Ecology and population genetic structure of strains of 
Teretrius nigrescens (Coleoptera: Histeridae), predator of 
Prostephanus truncatus (Coleoptera: Bostrichidae) 
B o n a v e n t u r e O m o n d i A m a n O d u o r 
Thesis submitted in fulfilment of the requirements for the award of the 
degree Doctor of Philosophy in Environmental Sciences 
at the North West University (Potchefstroom Campus) 
Supervisor: Prof. Johnnie van den Berg 
Co-supervisors: Dr. Fritz Schulthess 
Dr. Daniel Masiga 
May 2009 
DEDICATION 
To 
Anneke Leeuwenberg, 
Friend without borders 
Who, sadly, must now take a rest. 
iii 
ACKNOWLEDGEMENTS 
This work benefited from the support of many people to whom I am sincerely grateful. First 
and foremost, I appreciate the guidance, patience and encouragement by my supervisors Prof. 
Johnnie van den Berg, Drs. Fritz Schulthess and Daniel Masiga who, together with The Late 
Dr. Charles Omwega, taught me so much and nurtured my fledgling skills. I appreciate the 
immense logistical, moral and surrogate supervisory support of the Noctuid Stemborer 
Biodiversity team and Dr. Bruno Le Rii, who hosted this project in its final one year. I 
appreciate the statistical input of Dr. Nanqing Jiang (UNEP, China) for modelling work, Dr. 
Emese Meglecz (Universite de Provence, Marseille) in MlCROFAMILY analysis and Dr. 
Bruce Anani for sundry statistical support. 
Our collaborators' support and contribution to this work is greatly appreciated. Dr. Josephine 
Songa, Dr. Francis Nang'ayo and Mr. Paddy Likhayo started us up with colony 
establishment and management; while Messrs. Robert Mutweti, David Makumi and Patrick 
Lumumba helped with monitoring LGB fieldwork. The strains of the predator used were 
donated by Dr. David Bergvinson (CiMMYT, Mexico); Cyprian Atcha (IITA - Benin), 
Rhodrick Ndawala (NARO - Malawi) and Dr. Richard Hodges QSTRI, UK) sent us samples 
from UK and offered lots of advice. Partnership with Dr. Alfredo Rueda and Ms. Lucia 
Orantes (Zamorano University, Tegucigalpa) enabled us to prospect for three new strains of 
T. nigrescens from Honduras. Dr. Jeffrey Grubber (Russel Laboratories, Madison) sent us 
samples of sister group species. Drs. Pierpaolo Vienna and Riaan Stals identified the samples 
and together with the Histeridologists Network, advised on outgroup species. Thank you all 
for your kindness. 
My heartfelt appreciation to the staff of Biosciences East and Central Africa BecA hub, ILRI: 
Moses Naira, Delia Wasawo, Julius Osaso, Lucy Muthui, Bernard Waswa, Mercy Kitavi and 
Dr. Rob Skilton for guidance during my sojourn at ILRI Laboratories. 
Support from Drs. Pedro Fernandes (Gulbenkian Institute, Lisbon) Paul-Andre Calatayud 
and George Ongamo expanded the available information vastly. I also appreciate the input of 
all my colleagues: Drs. Kipkoech, Muli, Fenning-Okwae and, Obonyo, Migiro, Niassy and 
all the others made this work manageable. My laboratory colleagues especially Mr. Kabii, 
Dr. Eric Kouam, David Wainaina, Dr. Paul Mireji, Steve Ger, Eunice Machuka, Pamela 
Seda, Fathiya Khamis and Vincent Owino. 
IV 
My profound gratitude to Sarine Adhiambo, Charles Kamonjo and Andrew Wanyonyi, who 
between, them maintained the insect colonies and helped with the experiments. I am also 
grateful for technical assistance from Gerphas Okuku, Julius Obonyo, David Wainaina and 
all my student-interns. Administrative support by Ms. Caroh Akal and Beatrice Gikaria 
(Biocontrol Project) the Capacity Building Office and Molecular Biology and Biotechnology 
Unit made our work smooth. I am also grateful to my family: Mother, Paschalia Auma 
Oduor, sisters Beatrice Achieng' and Anne Akinyi and brothers Clement Argwings and 
Stephen Israel for urging on these efforts. To the many people who helped in diverse ways, 
but who I did not mention by name, thank you very much. Nothing is too small. The work of 
your helping hands lives in the submission of this thesis. 
Finally I am hugely indebted to the Kirkhouse Trust, UK for a generous sponsorship my PhD 
Fellowship and for providing research funds; the Swiss Government for funding the 
prospecting for samples in Honduras; and BecA-Hub, ILRI-Kenya for Research Fellowship 
for access to ILRI molecular Biology laboratories. 
Above all, I thank God, by who's Grace all was possible. 
V 
ABSTRACT 
The larger grain borer (LGB) Prostephanus truncatus (Horn) is the most important pest of 
farm stored maize and cassava in Africa. This alien invasive species was introduced into the 
continent from Mesoamerica in the late 1970s and by 2008 had spread to at least 18 
countries. In contrast to indigenous primary storage pests, LGB exists as on-farm and as wild 
populations, hence, sustainable control must target both environments. Biological control is 
especially attractive for wild populations to reduce early season grain store infestation, while 
cultural and chemical methods are useful to protect stored produce directly. Two populations 
of the predator Teretrius nigrescens Lewis were introduced into several African countries' as 
a biocontrol agent. It has shown long-term success and cost effective control in warm-humid 
areas. Control has however not been successful in cool and hot-dry zones. The aim of this 
study was to investigate the possible underlying genetic and ecological explanations for these 
observations and the possibility of joint use of molecular markers and ecological parameters 
in the development of sustainable control strategies. A 28-month baseline monitoring and 
recovery activity was done in from 2004 in five regions in Kenya along an east-westerly 
transect. Monitoring and live sample collection was also done in the original outbreak area in 
eastern Kenya. There was greater LGB flight activity in western Kenya (high potential maize 
production area) than the low potential areas. Very few T. nigrescens were recovered, solely 
in the eastern regions. LGB flight activity followed a seasonal pattern mostly related to 
changes in the relative humidity at 12:00, rainfall and dew point temperature but with a 3 - 4 
week lag. A linear predictive model based on these factors predicted 27 % of the observed 
flight activity. The survival and predation of five strains of T. nigrescens were compared at 
eight temperature levels between 15 °C and 36 °C at low and high humidity. All the strains 
of T. nigrescens exerted a significant reduction of LGB population build-up between 21 °C 
and 33 °C with generally better performance under humid conditions. There was no evidence 
vi 
of T. nigrescens development at 15 °C. At 18 °C, T. nigrescens oviposition and development 
was observed but the effect on LGB did not differ significantly from the control. The KARI 
population was the least effective in preventing grain damage at lower temperatures, but 
performed better than other strains above 30 °C at low humidity conditions. There was no 
control at 18 °C and 36 °C under both high and low humidity conditions. Since the extent of 
genetic differentiation in T. nigrescens was unclear from prior studies, several molecular 
marker techniques were progressively used. The RAPD-PCR did not reveal any genetic 
diversity between geographical populations. A lOOObp region of the mitochondrial mtCOI 
gene revealed two distinct clades differing consistently at 26 segregating sites. The two 
clades can be identified by simple PCR-RFLP procedure using single or double sequential 
restriction with EcoKl, Hindi, Rsal and Ddel digestion. However, the two lineages co-exist 
among the mid-altitude Central American populations. The internal transcribed spacer 
regions ITS1 and ITS2 with some neighbouring coding sequences of the ribosomal DNA 
were cloned and sequenced. The spacer regions were so variable in length and sequence 
between T. nigrescens and related Histeridae species that direct sequence alignment was not 
meaningful. Within T. nigrescens, there was intragenomic variability of the spacer regions 
mostly involving insertions and deletions of variable tandem repeat units predominantly 
within the ITS regions. The short flanking coding (18S, 5.8S and 21S) regions were 
conserved across populations and six other Histeridae species. There was no significant 
secondary structure variation of the ITS regions among populations of T. nigrescens. 
Twenty-four novel variable microsatellite markers were developed and tested on the 
Honduras populations. Alleles per locus ranged between two and twelve with observed 
heterozygosity between 0.048 and 0.646. Six loci deviated significantly from Hardy 
Weinberg Equilibrium and possibly had null alleles. The success of microsatellite 
amplification in outgroup species and variability of markers declined with an increase in the 
phylogenetic distance between the test species and T. nigrescens. Genotyping 432 individuals 
vii 
from 13 geographic populations revealed a comparatively higher genetic diversity in field 
populations. Partial isolation by distance and time was observed. Population bottlenecks were 
not detected, but recent expansion was evident in laboratory populations. Although five 
dominant genetic clusters were identified by Bayesian methods, meaningful hierarchical 
population structure was observed at between two and nine population groups (p < 0.01; 
10,000 iterations). Biological control of the larger grain borer using T. nigrescens seems an 
important aspect of the sustainable integrated control approach of the pest. Ecological 
adaptations, appropriate release strategies and genetic diversity are all essential 
considerations in these efforts and could be responsible for the variable success already 
observed. There is some genetic differentiation between populations of T. nigrescens but, 
further studies would be necessary to ascertain the contribution of such diversity to its 
predatory performance. The effect of laboratory culturing in aggravating genetic drift should 
be accommodated to avoid loss of diversity during sampling, quarantine, rearing and release 
of the predator. 
Key words: Teretrius nigrescens, Prostephanus truncatus, genetic differentiation, molecular 
markers, DNA sequence analysis, microsatellites, ecological suitability. 
VIII 
LUTTREKSEL 
Die groot graanboorder (GGB), Prostephanus truncatus (Horn), is die belangrikste plaag van 
plaasgestoorde mielies en cassava in Afrika. Hierdie eksotiese indringerspesie is per abuis 
gedurende die laat 1970s vanaf meso-Amerika na die die Afrika-kontinent gebring en het 
sederdien versprei na ten minste 18 lande. In teenstelling met inheemse plae van gestoorde 
graan bestaan GGB beide op-plaas en as wilde populasies wat tot gevolg het dat 'n 
volhoubare beheerstrategie beide omgewings sal moet teiken. Biologiese beheer is veral 
geskik om vroee-seisoen infestasies in graanstore te beperk, terwyl kulturele en chemiese 
beheermetodes nuttig is om die produk direk te beskerm. Twee populasies van Teretrius 
nigrescens Lewis is na verskillende Afrika-lande ingevoer en het langtermyn sukses en 
koste-effektiewe beheer van GGB in warmhumide areas tot gevolg gehad. Beheer was egter 
nie effektief in koel en droe areas nie. Die doel van hierdie studie was om moontlike 
onderliggende genetiese en ekologiese verklarings vir hierdie waarnemings te ondersoek 
asook om die moontlike ge'integreerde gebruik van molekulere merkers en ekologiese 
parameters in die ontwikkeling van volhoubare biologiese-beheerstrategiee te ondersoek. 
Feromoonvalle is gebruik om populasies van GGB en T. nigrescens te monitor oor 'nperiode 
van 28 maande langs 'n Oos - Wes gradient in vyf streke van Kenia. Monitering asook 
versameling van lewende materiaal is gedoen in die oorspronklike uitbraak-area in die ooste 
van Kenia. GGB is versamel in alle streke in die land: Daar was groter vlugte in die westelike 
hoe-potensiaal mielieproduskiegebiede as in lae-potensiaalgebiede. T. nigrescens is slegs in 
die oostelike streke opgespoor en in lae hoeveelhede. GGB vlugaktiwiteit volg :n seisoenale 
patroon wat grootliks verband hou met veranderinge in relatiewe humiditeit teen 12:00, 
reenval en doupunt-temperatuur maar met cn 3-4 week sloering in respons tot verandering in 
hierdie kondisies. :n Liniere voorspellingsmodel gebaseer op hierdie faktore het 27 % van 
die waargenome vlugaktiwiteit voorspel. Die oorlewing en predasie van vyf biotipes van T. 
ix 
nigrescens is vergelyk by agt verskillende temperature tussen 15 °C en 36 °C. Alle biotipes 
van T. nigrescens het 'n betekenisvolle afname in GGB populasie-opbou veroorsaak tussen 
21 °C en 33 °C. Geen ontwikkeling van T. nigrescens is by 15 °C waargeneem nie. By 18 °C 
is T. nigrescens eierlegging en ontwikkeling waargeneem maar dit het nie verskil van die 
kontrole nie. Die KAPJ-populasie was die minste effektief in die voorkoming van 
graanskade by laer temperature maar het beter gedoen as die ander biotipes by temperature 
bo 30 °C en by lae humiditeit. Geen beheer is waargeneem by 18 °C en 36 °C onder beide 
hoe en lae humiditeit nie. Aangesien die omvang van genetiese variasie in T. nigrescens nie 
duidelik was vanuit vorige getuienis nie, is verskeie molekulere tegnieke oor tyd gebruik om 
hierdie aspek te ondersoek. Die RAPD-PCR-tegniek het nie enige genetiese differensiasie of 
duidelike potensiele reeksgemerkte lokusse aangedui nie. 'n lOOObp streek van die 
mitokondriale SO 1-geen het twee duidelik onderskeibare klades van die predator aangetoon 
wat deurgaans op 26 segregerende lokusse verskil het. Twee klades kan identifiseer word 
deux eenvoudige PCR-RFLP prosedures deur gebruik te maak van enkel of dubbel-
 opvolgende restriksie met EcoKl, Hiincll, Rsal en Ddel. Twee lyne word egter gedeel tussen 
die mid-sentraal Amerikaanse populasies. Die interne getranskribeerde spasiestreke ITS1 en 
ITS2, tesame met die naburige koderende reekse van die ribosomale DNA is geamplifiseer, 
gekloon en die volgorde bepaal. Die spasiestreke tussen T. nigrescens en aanverwante 
Histeridae spesies het so varieer in lengte en volgorde dat direkte volgorde-sinkronisasie nie 
moontlik was nie. Selfs binne T. nigrescens is intergenoomvariasie van die spasiestreke, wat 
meestal veranderlike tandem herhaalde eenhede behels, waargeneem. Die meerderheid van 
die variasie was binne die ITS1 opvolgreeks. Die "kort kant" sykant koderings-streke (18S, 
5.8S and 21S) is bewaar oor populasies en ses ander Histeridae spesies. Daar was geen 
betekenisvolle sekondere struktuurvariasie van die ITS-streke tussen populasies van T. 
nigrescens nie. Vier-en-twintig nuwe veranderlike mikrosattelietmerkers is ontwikkel om 
enige onlangse demografiese gebeurtenisse in T. nigrescens populasies te raam. Hierdie 
X 
merkers is getoets op 'n veldpopulasie vanaf Honduras. Die aantal allele per lokus het varieer 
tussen twee en 12 met 'n waargenome heterosigositeit tussen 0.048 and 0.646. Ses loci het 
betekenisvol afgewyk vanaf die Hardy Weinberg Ekwilibrium en het aangedui dat daar geen 
amplifiserende allele voorkom nie. Die loci het die veranderlike allele in sewe ander 
Histeridae spesies geamplifiseer maar die sukses en veranderlikheid het afgeneem met 'n 
toename in die pohgenetiese afstand tussen die toetsspesie en T. nigrescens. 'n Groter 
genetiese diversiteit is by Genotipe 432 onder veldtoesande aangetoon. Gedeeltlike isolasie 
oor afstand en tyd is waargeneem. Geen bottelnekke is in populasies waargeneem nie maar 
onlangse uitbreiding is waargeneem in laboratoriumpopulasies. Alhoewel vyf dominante 
genetiese groepe geidentifiseer is met behulp van Bayesiaanse metodes, is betekenisvolle 
hierargiese populasiestrukture waargeneem tussen twee en nege populasiegroepe (p < 0.01; 
10,000 iterasies). Biologiese beheer van GGB met T. nigrescens is 'n belangrike komponent 
van 'n volhoubare geihtegreerde beheerstrategie. Ekologiese aanpassings, toepaslike 
vrylatingsstrategiee en genetiese diversiteit is noodsaaklike oorwegings in hierdie pogings en 
mag verantwoordelik wees vir die verskillende vlakke van sukses wat waargeneem is. 
Alhoewel daar wel genetiese diversiteit binne populasies van die predator waargeneem is, is 
verdere studies nodig om die bydrae van hierdie diversiteit tot die effektiwiteit van beheer 
onder veldtoestande te bepaal. Die effek van laboratorium-geteelde kolonies in die toename 
in genetiese drywing moet geakkomodeer word ten einde 'n verlies aan diversiteit en 
aanpassing by teelomstadighede te verhoed. 
Sleutelwoorde: Teretrius nigrescens, Prostephanus fruncatus, genetiese differensiasie, 
molekulere merkers, DNA volgorde-analises, mikrosatelliete, ekologiese volhoubaarheid. 
xi 
TABLE OF CONTENTS 
D E D I C A T I O N II 
ACKNOWLEDGEMENTS HI 
ABSTRACT '. V 
UITTREKSEL VTH 
TABLE OF CONTENTS XI 
1 C H A P T E R O N E 1 
GENERAL INTRODUCTION AND LITERATURE REVIEW 1 
1.1 INTRODUCTION.., 1 
1.2 GENERAL LITERATURE REVIEW 5 
1.2.1 Maize and cassava in Africa 5 
1.3 POST-HARVEST LOSSES 6 
1.3.1 Prostephanus truncatus in Africa 8 
1.4 MONITORING AND DETECTION OF STORED PRODUCT PESTS 8 
1.4.1 Monitoring P. truncatus 1 10 
1.5 CONTROL OF STORED PRODUCT PESTS 11 
1.5.1 Control of the larger grain borer 11 
1.5.2 Biological control. 12 
1.6 TERETRIUS NIGRESCENS 13 
1.7 MOLECULAR METHODS IN INSECT POPULATION GENETICS 16 
1.7.1 RAPD-PCR : 16 
1.7.2 Amplified fragment length polymorphism 17 
1.7.3 Mitochondrial DNA markers 18 
1.7.4 Nuclear DNA markers 19 
1.7.5 Microsatellite markers 20 
1.8 POPULATION GENETICS IN BIOLOGICAL CONTROL 21 
1.9 REFERENCES 23 
2 C H A P T E R T W O 37 
GENERAL MATERIALS AND METHODS 37 
2.1 STUDY MATERIALS 37 
2.1.1 Insects 37 
2.2 MONITORING AND RECOVERY OF P. TRUNCATUS AND T. NIGRESCENS 41 
2 .3 OUTGROUP SPECIES 4 1 
2.4 INSECT REARING 42 
2.4.1 Recognition of life stages 43 
2.5 GENERAL MOLECULAR PROCEDURES 46 
2.5.1 DNA extraction 46 
2.5.2 Molecular cloning 48 
2.5.3 DNA sequencing reactions 48 
2.5.4 Sequence Analysis 49 
2.6 REFERENCES ? 50 
xu 
3 C H A P T E R T H R E E 52 
THE FLIGHT ACTIVITY OF PROSTEPHANUS TRUNCATUS (HORN) AND TERETRIUS 
NIGRESCENS LEWIS IN KENYA 52 
3.1 ABSTRACT 52 
3.2 INTRODUCTION 52 
3.3 MATERIALS AND METHODS 54 
3.3.1 Sampling sites 54 
3.3.2 Insect trapping 55 
3.3.3 Meteorological data 58 
3.3.4 Data analysis 58 
3.4 RESULTS = 59 
3.4.1 Seasonal fluctuation in numbers ofLGB and I nigrescens 59 
3.4.2 The LGB flight model 65 
3.5 DISCUSSION 66 
3.6 REFERENCES 73 
4 C H A P T E R F O U R 78 
EFFECT OF TEMPERATURE AND HUMIDITY ON THE PREDATION OF TERETRIUS 
NIGRESCENS ON PROSTEPHANUS TRUNCATUS 78 
4.1 ABSTRACT 78 
4.2 INTRODUCTION 78 
4.3 MATERIALS AND METHODS 79 
4.3.1 Insects 79 
4.3.2 Experimental set up 80 
4.3.3 Damage and loss assessment 81 
4.3.4 Data Analysis 82 
4.4 RESULTS 83 
4.5 DISCUSSION 89 
4.6 REFERENCES 93 
5 C H A P T E R F I V E 97 
PHYLOGENETIC RELATIONSHIP BETWEEN TEN POPULATIONS OF TERETRIUS 
NIGRESCENS 97 
5.1 ABSTRACT 97 
5.2 INTRODUCTION 97 . 
5.3 MATERIALS AND METHODS 100 
5.3.1 DNA extraction 100 
5.3.2 Polymerase Chain Reaction 100 
5.3.3 PCR-RFLP analysis 102 
5.3.4 Data Analysis 103 
5.4 RESULTS 104 
5.4.1 MtCOI Sequence variation 106 
5.4.2 Genetic diversity 106 
5.4.3 PCR-RFLP identification of populations 107 
xiii 
5.4.4 Internal transcribed spacer regions 114 
5.5 DISCUSSION 114 
5.6 REFERENCES 118 
6 C H A P T E R S I X 125 
DEVELOPMENT OF MICRO SATELLITE MARKERS FOR TERETRIUS NIGKESCENS 125 
6.1 ABSTRACT 125 
6.2 INTRODUCTION 126 
6.3 MATERIALS AND METHODS 128 
6.3.1 Microsatellites-enriched library construction 128 
6.3.2 Identification of tandem repeats 130 
6.3.3 Primer design and optimisation 130 
6.3.4 Allele calling 132 
6.3.5 Cross species amplification 132 
6.3.6 Multiplexing Microsatellite primers 133 
6.3.7 Data Analysis 134 
6 A RESULTS 134 
6.4.1 Multiplex sets 135 
6.5 DISCUSSION 146 
6.6 REFERENCES 151 
7 C H A P T E R S E V E N '. 155 
POPULATION GENETIC STRUCTURE AND DEMOGRAPfflC fflSTORY OF TERETRIUS 
NIGRESCENS 155 
7.1 ABSTRACT 155 
7.2 INTRODUCTION 155 
7.3 MATERIALS AND METHODS 158 
7.3.1 Sampling and PCR 158 
7.3.2 Data Analysis 158 
7.4 RESULTS 163 
7.4.1 Genetic diversity 163 
7.4.2 Genetic differentiation 164 
7.4.3 Cluster analysis 164 
7.4.4 Demographic history 165 
7.5 DISCUSSION 174 
7.6 REFERENCES 180 
8 C H A P T E R E I G H T 188 
GENERAL DISCUSSION AND CONCLUSIONS 188 
8.1 REFERENCES :...195 
1 
C H A P T E R ONE 
General Introduction and Literature Review 
1.1 Introduction 
The larger grain borer (LGB) Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) is 
the worst pest of farm-stored maize and cassava in Africa today (Schneider et al., 2004). It 
was accidentally introduced into the continent in the 1970s and has since spread to at least 18 
countries (Schneider et al., 2004; Gueye et al., 2008). The LGB causes a mean weight loss of 
about 34 % within the first six months of grain storage, five times that caused by all the 
indigenous storage pest fauna (Hodges et al., 1983, Giles et al., 1996). This damage is 
particularly costly as it occurs when all the production costs have been incurred and when 
plant compensatory growth is not possible. Economic losses due to poor product 
marketability, pricing, seed viability and possible association with post-damage 
contamination can lead to total economic loss. 
In Kenya, annual losses were estimated at 730,000 tonnes, valued at KShs. 8.1 billion (US $ 
108 million at current exchange rates (www.xe.com, March, 14, 2009) (Anonymous, 2003). 
The larger grain borer spread into the country from Tanzania and initially established in the 
semi-arid grain deficient coastal and eastern region (Hodges et al., 1998). Strict containment 
efforts were imposed on the movement'of maize within Kenya, to prevent P. truncatus 
spreading to the main maize-surplus areas in the west of the country (Hodges et al., 1998). 
The pest has recently been reported in the cool high production highlands like the western 
Rift Valley in Kenya. This expansion poses the threats of both direct damage of bulk-stored 
grain and rapid dispersal from these maize-surplus regions during grain movement. It is also 
a new impediment in the grain industry, food security and the contribution of agriculture to 
the country's GDP. 
2 
The establishment of the LGB in Africa has also changed the post-harvest pest control 
paradigm. Storage of unshelled maize was hitherto effective against native post-harvest pests 
like the maize weevil, Sitophilus zeamais (Motchulsky) (Coleoptera: Curculionidae) and 
Angoumis grain moth Sitotroga cerealella Olivier (Lepidoptera: Gelechiidae), but facilitates 
infestation of maize by the LGB. Good store hygiene, cob segregation and sun-drying have 
been used in the traditional systems, but are not sufficient against the LGB (Giles et al., 
1996). Although shelling maize significantly reduces damage by the LGB (Cowley et al., 
1980), storage in sacks stabilises it enough to encourage attack. Treating shelled grain with 
binary pyrethroid-organophosphate insecticide dusts (such as Actellic Super) has been useful 
in pre-infestation protection. Recently, a spinosad-based insecticide has been introduced 
against storage insects (Fang et al., 2002). Chemical control is not technically sustainable in 
the subsistence storage systems due to the possibility of using inappropriate toxicant doses, 
counterfeit and diluted pesticides or development of resistance to pesticides. In high LGB 
risk areas, many farmers evade losses by disposing of their produce at low prices soon after 
harvest, forcing them to buy grain at higher prices later in the season (Meikle et al., 2000; 
Hodges, 2002). 
The presence of the LGB populations outside storage environments is a challenge to 
conventional control measures. Prostephanus truncatus is not exclusively a post-harvest 
agricultural pest but a woodborer, which has adapted to dried stored products. It can survive 
on several species of trees in Africa (Nang'ayo et al., 1993; Nansen et al., 2004). Cultural 
and chemical measures would be impractical or too expensive in the wild habitats. Adapted 
natural enemies that can locate the pest outside the agricultural ecosystem could help reduce 
the rate of re-infestation of disinfected stores and the early season pest pressure (Nansen and 
Meikle, 2002). 
3 
Teretrius (formerly Teretriosoma) nigrescens Lewis (Coleoptera: Histeridae), a predatory 
beetle, found to be consistently associated with P. truncatus in the native range, has been 
introduced into Africa in a classical biological control approach (Hoppe, 1986; Boye et al., 
1988; Borgemeister et al., 2001). This egg and larval predator has specific preference for P.-
 truncatus (Ayertey et al., 1999; Rees, 1991) and cues in on the pest's aggregation 
pheromone, as a kairomone, to locate its prey (Rees et al., 1990; Scholz et al., 1998). It is 
also able to disperse fast in the wild on its own (Borgemeister et al., 1997a). 
Biological control of the LGB using T. nigrescens has been carried out in several African 
countries with variable levels of success. The release of a Costa Rican population of the 
predator resulted in lower LGB flight activity and a reduction of grain damage trends in 
warm humid coastal areas of Togo, Benin and Ghana (Borgemeister et al., 1997a). An 80 % 
reduction in LGB flight activity and 70km dispersal was observed within three years of 
release of a Mexican population in Kenya (Hill et al., 2003). Yet, while control in coastal 
West Africa was sustained over 10 years, that in eastern Africa has largely collapsed. In 
Guinea Conakry, the Benin population failed to establish under cool highland conditions, 
comparable to the failure of the predator to pursue the LGB beyond the semi arid eastern 
region of Kenya. Climate and strain characteristics, therefore, determine the success of this 
approach. Cold and dry climates limit the performance of T. nigrescens more than that of P. 
truncatus (Tigar et al., 1994). 
Population differentiation and ecological and climatic limitations of the predator are not fully 
understood. Tigar et al. (1994) recovered it from highland areas of La Laguna and Texcoco 
in Mexico, while Rees (1990) observed the predator in very dry coastal areas near Yucatan 
Peninsula. Weather conditions of these areas are comparable to those in either cool humid 
Highlands or the dry Sahel, both of which have failed to sustain the establishment of its 
4 
populations. Although intraspecific variation has been reported in the LGB (Guntrip et al., 
1996; Marshed-Kharusy and Dawah, 1999) it does not significantly affect predation by T. 
nigrescens (Bergvinson, CiMMYT, Mexico, personal communication). It is therefore likely 
that predator population characteristics and their adaptation to environmental conditions play 
a key role in the levels of success observed in the biological control of the LGB in Africa. 
Six new populations of T. nigrescens recovered from various areas in Mexico and Honduras 
and two populations from Mexico and Costa Rica earlier released in Africa, are being 
studied under quarantine at the International Centre of Insect Physiology and Ecology 
(ICIPE), Nairobi, for possible use in LGB control programmes in eastern and southern 
Africa. Their release must be preceded by the development of reliable means of monitoring 
the strains and their biological interactions with each other, the pest and the environment. 
Characterisation of the strains of T. nigrescens is essential in quantifying, assessing and 
interpreting the performance of the strains of T. nigrescens used. Since biological differences 
have been noted without associated morphological differences between the putative biotypes, 
unequivocal differentiation using molecular markers is invaluable in this programme. 
This study, therefore, sought to answer the following questions: what is the level of genetic 
differentiation within T. nigrescens"! Do ecotypes of the predator exist? Do T. nigrescens 
geographic populations or putative strains have distinct preferences to measurable 
environmental conditions? Can the putative strains be discerned using simple molecular 
markers? How can diversity within T. nigrescens be harnessed for sustainable control of the 
LGB in Africa? 
Consequently, baseline monitoring of P. truncatus and T. nigrescens was done to assess the 
current distribution and flight activity of the two species in Kenya. Putative geographic 
5 
strains of the predator were compared to for their preference to temperature and humidity 
regimes. Mitochondrial and nuclear genetic markers were used to trace the phylogenetic 
history of the populations and to apportion its variation. Novel microsatellite markers were 
developed and utilised to assess the population genetic structure and demographic history of 
the predator. Finally, a simple molecular procedure has been proposed for differentiating 
four populations. 
1.2 General Literature Review 
1.2.1 Maize and cassava in Africa 
Maize and cassava are the most important staple food crops in Africa. The acreage of maize 
in the continent is about 26 million hectares, with an annual production of 42 million tonnes. 
In Kenya, 2.3 million tonnes are produced on some 1.5 million hectares (FAO, 2004). Over 
103 million tonnes of cassava is produced on the continent almost exclusively by subsistence 
farmers (FAO, 2004). Because of its drought tolerance, cassava is an important food security 
crop to about 20 million Africans (Legg, 1994). Significant failure of the two commodities 
therefore threatens food security and economic stability of the poor communities least able to 
bear the burden. 
Production of the two crops remains low, mainly due to climatic constraints and biotic 
factors. Field pests such as lepidopteran stem borers and diseases, for example, the maize 
streak virus, downy mildew and grey leaf spot, are major constraints to maize production. 
Similarly, the cassava mosaic geminiviruses seriously threaten yield stability of cassava 
(Legg et ah, 1994). Post-harvest losses are therefore especially painful since they occur after 
all production costs have been incurred. In developing countries, about 90 % of cereals 
produced are intended for human consumption, so, improved post-harvest protection 
measures could increase food supply by 30 - 40 % (Pantenius, 1988). 
6 
1.3 Post-harvest losses 
The LGB causes multi-dimensional and very costly damage to grains. Losses of quality, 
grain weight and those due to forgone income or extra expenditure in grain protection efforts 
can total to over 50 % of the loss in product value (Magrath et ah, 1996) (Figures 1.1, 1.2, 
1.3). Mean grain losses in Africa have been estimated at between 9 % and 13 % in the first 
six months (Golob, 1988; Pantenius, 1988). Insects account for 80-90 % of this loss 
(Pantenius, 1988). The lower yielding indigenous landraces of maize generally suffer lower 
post-harvest losses than the high yielding improved varieties (Prevett, 1990). 
The level of post-harvest grain losses in sub-Saharan Africa increased dramatically following 
the introduction and establishment of the LGB in the late 1970s and 1980s and its subsequent 
spread (Farrel and Schulten, 2002). Reported loss estimates include 8 % per month in Togo; 
30 - 50 % on unshelled maize in Burundi and up to 34 % in Kenya (Tantenius, 1988; 
Nang'ayo, 1996). The presence of the LGB also complicates food supply to grain deficit 
areas where containment approaches are effected. Yet farmers shell and sell maize early in 
the season in most affected areas to avoid rapid damage by the pest. Early disposal of surplus 
maize and repurchase during deficit seasons leads to double loss of selling at low prices and 
later buying the same produce (usually of a lower quality) at higher prices. 
7 
Figure 1.1: The Larger Grain Borer, Prostephanus truncatus (courtesy of G. Goergen, IITA) 
c • 
-11 - : i 
^^^^fJSSwi'^ W'. ?*! r. i t l^W'^JL >£*?» W^% 
Figure 1.2: LGB damage on different substrates: A- maize kernels, B- wood and C-cob of 
maize (pictures B and C: courtesy of C. Borgemeister, ICIPE). LGB frass was cleaned off 
substrates to expose damage. 
8 
1.3.1 Prostephanus truncatus in Africa 
The LGB is a native of Central America where it has long been recognised as an occasional 
pest of stored maize (Boxall, 2002). It was accidentally introduced into Africa in the 1970s 
(Dunstan and Magazini, 1981; Krall, 1984) and has now spread to at least 18 countries 
(Schneider et al., 2004; Gueye et al., 2008). The LGB causes rapid deterioration in the mass 
and quality of maize, dried cassava chips and even wood (Borgemeister et al., 2001; Boxall, 
2002) (Figure 1.1). Its introduction and spread has been attributed to movement of grain in 
trade and distribution of relief food. In fact, heavy infestations are often recorded near maize 
transit routes and grain markets (Tyler and Hodges, 2002). 
The LGB is a polyphagous insect, surviving and reproducing on agricultural and non-
 agricultural substrates. It mainly infests maize and cassava but also wheat, paddy pulses, 
groundnuts, cocoa and coffee beans and dry tubers of yam and sweet potato (Shires, 1977; 
Boxall, 2002). It can also infest above-ground wood, woody roots and seeds of several tree 
species (Nang'ayo et ah, 2002; Nansen et al., 2004). In Africa, high trap catches have been 
recorded in forest environments, away from any farmlands, grain storage or market areas, 
underlining the importance of the natural non-agricultural habitats in P. truncatus ecology 
(Hill et al., 2003). Woody tree branches girdled by cerambicid beetles are, so far, the only 
non-agricultural hosts in which the LGB adults and larvae have been found in the forest 
(Nansen et al., 2004). Perhaps, such hosts play an important role in maintaining populations 
in the off-season before other food substances become available (Nang'ayo et al., 1993; 
Ramirez-Martinez et al., 1994). 
1.4 Monitoring and detection of stored product pests 
Surveys of post-harvest pests generally focuses on detection and monitoring populations. 
Detection aims at finding the presence of the pest, especially to support early warning 
9 
systems or to aid to control decision making (Likhayo and Hodges, 2000; Hodges, 2002). 
Conversely, monitoring attempts to estimate changes of pest numbers over time. Early 
detection of storage pests is important since they uniquely infest the portion of the plant of 
the highest quality, and also at a time when compensatory growth is no longer possible. 
Losses cannot be made up for and are relatively economically more costly compared to 
early-season foliage damage. Many post harvest pests are cryptic and, therefore, detection of 
infestation is more critical than the estimation of population size since the latter is often 
difficult to determine (Hodges, 2002; Throne et al, 2003; Xing and Guyer, 2008). Since 
storage structures generally do not permit individual inspection of stored units, pest detection 
is invaluable. Data from pest monitoring are useful in detecting trends over time which 
would be associated with specific environmental conditions (Hodges, 2002). 
The methods of pest monitoring and detection vary from physical inspection of samples, 
transmittance spectroscopy and pheromone luring both inside and outside of the storage 
environments (Hodges, 2002; Xing and Guyer, 2008). For the LGB, traps are generally more 
effective than physical inspection. Probe traps are used for sampling insects moving through 
grain and can be baited to increase their attractiveness (reviewed by Hodge, 2002 and Throne 
et al., 2003). However, these traps require entering into the storage facility and are labour 
intensive. Also they are not useful for detecting insect stadia feeding and developing inside 
grains. Flight traps have been used for pest monitoring in the grain headspace and outside the 
store. Pheromone baited flight traps are in use for detecting Sitophilus zeamais (Dendy et al., 
1991; Likhayo and Hodges, 2000). The design of such traps depends on a clear 
understanding of the pest biology and ecology and an understanding of their working. 
( 
10 
1.4.1 Monitoring P. truncatus 
Probe and pheromone-baited crevice traps were initially used for LGB monitoring in grain 
stores (Dendy et ah, 1991). However these traps could detect the pest in only one third of the 
infestations in Mexico (Rees et al., 1990). This is perhaps because the aggregation 
pheromones are much more attractive to migrating insects looking for feeding substrates 
(Fadamiro and Wyatt, 1995). The LGB have also been caught in unbaited light traps, but 
these catch many species of insects and catch identification may be difficult. Similarly, they 
are not specifically attractive to the pest species. Flight traps are now the standard 
monitoring tool for LGB detection. The design of flight traps influences their attractiveness 
to P. truncatus. Delta or wing traps are not as attractive as funnel traps but much easier to 
deploy and not as prone to vandalism. Funnel traps are more protected from saturation, allow 
for addition of arrestants for trapped animals and are more consistent. They are therefore 
more useful for live recovery of samples (Hodges, 2002). 
The interpretation of flight trap data is critical. Flight traps generally catch insects in the 
dispersal phase and (for the LGB) are more attractive to young mated females (Hodges, 
2002). The LGB aggregation pheromone orientates flying insects towards the pheromone 
source but does not induce flight (Fadamiro and Wyatt, 1995). Ambient temperature during 
development, food quality and other environmental triggers of flight have all been shown to 
contribute to dispersal (Fadamiro and Wyatt, 1995; Scholz et al., 1997; Borgemeister et ah, 
1997b; Hodges et al., 2003). Instantaneous flight trap catches are therefore an indication of 
the propensity of the pest to disperse, but may only give an indication of the actual available 
population if long term trends are considered and accounted for (Hodges, 2002). 
11 
1.5 Control of stored product pests 
Cultural, physical and chemical means are used to manage post harvest pests in Africa. 
Maintenance of hygiene by complete emptying of and cleaning of stores before re-stocking 
reduces pest carry over between seasons. Sun drying of grains and treatment with inert dusts 
and botanical products are common in subsistence farming systems (Giles et al., 1996). 
These methods are effective against most post-harvest pests but do not ensure adequate 
control of the LGB. For instance, maize storage in husks or on cobs discourages infestation 
by most weevils and grain moths, but promotes infestation by the LGB (Giles et al., 1991; 
Ayertey et al., 1999). Shelling maize before storage and treating it with insecticides have 
been effective against indigenous post-harvest pests and the LGB in parts of East Africa, but 
not in West Africa (Ayertey et al., 1999). 
1.5.1 Control of the larger grain borer 
Methods of control of the LGB have evolved as more knowledge of the pest becomes 
available (Farrel and Schulten, 2002). Containment by exclusion and eradication was the first 
approach attempted, but was technically and logistically difficult to implement (Hodges, 
2002). Eradication of the pest succeeded in Israel and Arabia but not in Africa and integrated 
approaches were proposed (Farrell and Schulten, 2002). 
Both preventive and curative control measures have been used. Restriction of grain 
movement from and through infested regions has been attempted, but has been difficult to 
enforce, since cross-border smuggling and informal movement of grain is common (Golob, 
2002). Such activities, perhaps, introduced the pest into Burundi from Tanzania (Farrell and 
Schulten, 2002). Chemical treatment in farm stores and natural habitat has been attempted. 
Fumigation of stored produce using methyl bromide and aluminium phosphide (phostoxin) 
has been effective (Golob, 2002) although it is technically and logistically impractical at 
12 
small farm level. Contact and stomach insecticides, spread in inert powder or as liquids, 
protect the grain surface, and are therefore more potent on shelled maize or when used before 
infestation. Once cobs are attacked, adult LGB adults live and reproduce inside grain and 
might escape control (Golob, 2002). Use of pesticides is also limited by knowledge levels, 
hence the possibility of using fake and diluted products and incorrect doses. The 
effectiveness of a combination of organophosphates and pyrethroids is reducing as the pests 
become resistant (Golob, 2002). New pesticides based on Spinosad have been introduced in 
the market (Fang et al., 2002). Spinosyns are fungal secondary metabolites, so, the 
development of resistance is likely in the long run. An integrated management approach 
including the storage technology, hygiene, post-harvest treatment, biological control and 
supported by proper monitoring tools are presently favoured (Farrell and Schulten, 2002). 
1.5.2 Biological control 
Prostephanus truncatus is an introduced species and only of minor pest importance in its 
area of origin (Pantenius, 1988). Practical challenges to chemical control rendered it an ideal 
candidate for classical biological control (Schultz and Laborius, 1987, 1988). Two Protozoa, 
Mattesia and Nosema, were not effective under semi-field trials while two strains of bacteria 
caused significant mortality in the laboratory but were ineffectual in field studies (Poshko, 
1994). Isolates of the fungus Metarrhiziwn anisopiale (Metchinkoff) Sorokin were highly 
virulent but their use was not pursued due to technical and safety reasons (Schulz and Meikle 
etal, 2002). 
Of the natural enemies recovered, T. nigrescens was the most consistently associated with 
the pest and effective in causing significant mortality in the laboratory and in the field (Boye, 
1988). It was also significantly coadapted to predation on the LGB. The predator was 
therefore released in several areas in Africa between 1991 and 1996 for the control of the 
13 
LGB (Meikle et al, 2002). Several studies have reported reduction in flight activity, damage 
level and infestation rates (Borgemeister et al, 2001; Farrel and Schulten, 2002; Hill et al, 
2003). Other workers argue that the sustainable successful use of T. nigrescens is doubtful, 
and fluctuations in flight activity were mainly due to climatic conditions, hence clear 
understanding of the post-harvest system is a prerequisite to its use (Hoist and Meikle, 2002; 
Meikle et al, 2002). Integrated management approaches should include knowledge of the 
ecological limitations of the pest and predator and development of effective monitoring and 
warning systems (Boxall, 2002; Farrel and Schulten, 2002). 
1.6 Teretrius nigrescens 
Teretrius nigrescens Lewis belongs to the family Histeridae as do over 3700 other species 
(Mazur, 1997). They are tiny beetles, 2-3mm long with a compact body (Figure 1.3) 
(Poshko, 1994). Histerids are predators of necrophagous and wood boring insects and mites. 
Most histerids are either oligophagous or polyphagous living in carcasses and leaf litter 
(Degallier and Gomy, 1983; Stewart-Jones et al, 2004). 
Teretrius nigrescens shows a distinct preference for P. truncatus (Poshko, 1994). Although it 
can also prey upon Dinoderus spp., Rhizopertha dominica (Fabricius) (Coleoptera: 
Bostrichidae), Sitophilus spp. (Coleoptera: Curculionidae) and Triboliwn castaneum (Herbst) 
(Coleoptera: Tenebrionidae), which are common storage pests of cereals in Africa (Rees, 
1987; 1990; Poshko et al, 1992; Ayertey et al, 1999). Its performance on these species is 
however very poor. Studies showed that T. nigrescens does not pose a threat to the local 
beneficial insect fauna due to its prey finding strategy and distinct prey preference (Farrel 
and Schulten, 2002). In stores, it survives on grain without the prey for long periods, but 
reproduction would be suppressed under these conditions. 
14 
Figure 1.3: Adult Teretrius nigi-escens (courtesy: G. Goergen, IITA, Benin). 
The bar represents 1 mm. 
15 
Teretrius nigrescens is well adapted to the habitat and habits of its prey. The body size and 
fossorial adaptation of the fore tibia of the adults enable it to pursue the borer into the 
substrate. The larvae are voracious predators of the eggs and larvae of the LGB. 
Behaviourally, the predator exhibits strong attraction to the aggregation pheromones of P. 
truncatus (Scholz et al., 1998) and non-volatile compound(s) in the dust of the LGB (Rees, 
1990; Stewart-Jones et al., 2004). In Mexico and Costa Rica, it was only observed in 
association with the LGB, both in natural environments and maize stores (Boye, 1988; Rees, 
1990). 
The predatory capacity of T. nigrescens and its ability to control the LGB have been widely 
studied under laboratory and field conditions. Both larvae and adults prey on P. truncatus 
eggs and larvae (Rees, 1985; Poshko, 1994). A single larva consumes about 60 LGB larvae 
(about 5 daily) to develop to an imago, while an adult kills about 1.1 larvae a day (Boye, 
1988). An 80 % decline in the LGB population has been recorded in laboratory bioassays. In 
the field, the predator disperses faster than its prey P. truncatus (Borgemeister et al., 1997a). 
Field releases of T. nigrescens have resulted in a reduction in losses, maize damage, LGB 
flight activity and infestation levels (Giles et al., 1996; Borgemeister et al., 2001; Hill et al., 
2003; Schneider et al., 2004). The performance of T. nigrescens as a biological control agent 
depends on several environmental factors. Grain damage and weight loss are usually lower 
and suppression of LGB population growth better on shelled than cob-stored maize (Poshko, 
1994). There is little evidence however of control of the LGB by T. nigrescens on cassava, 
yam and other bulk substrates. 
Few studies have conclusively reported the effects of climate on the development or 
performance of T. nigrescens in the control of the LGB. Anecdotal observations indicate that 
T. nigrescens is more limited by cool temperatures than its prey P. truncatus. Oussou et al. 
16 
(1998) did not establish any relationship between relative humidity conditions and larval 
survival, although 30 % RH appeared marginally more suitable than higher and lower 
humidity conditions. Leliveldt (1990) observed higher predatory ability of T. nigrescens at 
30 °C than at 26 °C on P. truncatus reared on maize. The recent reports of the pest dispersal 
to cool maize surplus areas especially in Kenya are, therefore, a major concern, as the 
ecological limitations of the T. nigrescens strains are not clearly understood. 
1.7 Molecular methods in insect population genetics 
Several molecular markers are available for the study of inter- and intra-specific variation 
and genetic structure of insect populations. These have been variously applied to delineate 
species, determine insect phylogenetic relationships (Caterino and Vogler, 2002) and 
elucidate intraspecific variation (Omondi et al, 2004; Berry et al., 2004). Caterino and 
Vogler (2002) used the 18S rDNA sequences in combination with morphological 
characteristics to resolve the phylogeny of the superfamily Histeroidea. However, each 
marker system comes with a set of assumptions, strengths and limitations restricting the 
ecological question it can be used to address (reviews by Loxdale and Lushai, 1998; Avise, 
2000; Estoup et al, 2002; Zhang and Hewitt, 2002). 
Within T. nigrescens, no reports of DNA based markers available. Molecular markers for 
such exploratory studies should be those that either require no prior sequence information for 
the species (e.g. RAPD-PCR, AFLP) or those for which probes may be obtained from other 
closely related studied species such as sequence analysis of a conserved gene. 
1.7.1 RAPD-PCR 
Randomly Amplified Polymorphic DNA - Polymerase Chain Reaction (RAPD-PCR) 
technique characterises DNA by amplifying fragments whose location is priorly unknown. 
17 
Short (8 — 10 bp) primers of arbitrary sequence, which anneal to several locations on the 
genome, are used (Welsh et al., 1990; Williams et al., 1990). The products are then separated 
by electrophoresis and a comparison based on the extent of band sharing between pairs of 
individuals. 
RAPD-PCR technique is likely to sample the genome more randomly, does not require prior 
sequence information on the subject species and utilises available random primers across 
species (Lynch and Milligan, 1994). It is also technically simple. The results however are 
usually not reproducible between laboratories, experimental protocols and time. Also 
comigration of equal sized fragments of different sources may exaggerate the similarity 
signal (Lynch and Milligan, 1994). Being a dominant marker it is limited on the potential 
number of alleles for the comparison of samples. Practical procedures are available for the 
improvement of the reliability of this marker (Gawell and Bartlett, 1993; Lima et al., 2002). 
RAPD-PCR has been useful in studies of population structure of insect species (Hadrys et 
al., 1992; Cenis et al., 1993; Lima et al., 2002; Omondi et al., 2004) identification of insect 
biotypes (De Barro and Driver, 1997; Moya et al., 2001) mating studies (de Barro and Hart, 
2000; Omondi et al., 2005). Sequencing unique RAPD bands has also led to the development 
of sequence characterised amplified regions (SCARs) markers (Rugienius et al., 2006). 
1.7.2 Amplified fragment length polymorphism 
Amplified (restriction) fragment length polymorphism (AFLP) technique combines the 
specificity of restriction enzyme digestion with the reliability of PCR to detect DNA 
polymorphism. It is based on the selective amplification of a fraction of the fragments 
obtained after DNA restriction, allowing high resolution of genetic differences. These 
markers are dominant as they display the presence or absence of restriction fragments and 
not their length differences. In entomological investigations, AFLP has been used to study 
18 
genetic variation among whiteflies (Cervera, 2000), linkage mapping in moths (Heckel et al., 
1998) and fruitfly taxonomy (Kibogo, 2005). It is therefore of potential utility in the design 
of group specific primers for molecular diagnosis, if any Sequence Tagged Sites (STSs) can 
be recovered and sequenced (Brugmans et al., 2003). It is a method of choice for its 
repeatability and robustness while using arbitrary primers. 
1.7.3 Mitochondrial DNA markers 
Mitochondrial genomes are a rich source of markers for studies of population genetics and 
evolution. In animals, the mitochondrial genome is generally circular (4-17kb) is maternally 
inherited and has a relatively simple genetic structure and high mutation rates (Adams and 
Palmer, 2003; Thao et al, 2004). The DNA molecule is made up of 37 genes coding for 22 
transfer RNAs, two ribosomal RNAs and 13 messenger RNAs and generally lacks introns. 
The rDNA on the other hand has large families of repetitive DNA, pseudogenes and large 
spacer sequences (Loxdale and Lushai, 1998). Within the class Insecta, the order of 
mitochondrial genes is highly conserved leading to the proposal of an ancestral gene array 
(Shao and Baker, 2003). 
Different gene regions of the mtDNA have been found to change at different rates (Lunt et 
al., 1996). The control region changes more rapidly both within and between species, 
whereas the ribosomal RNA genes (e.g. 12S and 16S) evolve more slowly (Moritz, 1987; 
McPheron and Han, 1997). Mitochondrial DNA (mtDNA) is therefore used for taxonomic 
and population genetic studies of insects (Loxdale and Lushai, 1998). Since there are more 
mitochondria compared to nuclei in a cell, there are more copies of mtDNA than those of 
nuclear DNA hence mtDNA can be detected in old and minute and physically degraded 
samples. The mtDNA is probably the most widely used source of DNA for molecular 
19 
investigations and a good starting point for preliminary investigations (Zhang and Hewitt, 
2003). 
Universal PCR primers have been described to amplify across different sections of the 
mtDNA (Zhang and Hewitt, 1997; Loxdale and Lushai, 1998). Divergence of nucleotide 
sequences of various mtDNA genes and subsequent sequence alignment to reveal nucleotide 
differences has been used to designate insect biotypes, notably the use of the mtCOI gene in 
Bemisia tabaci biotype designation (Berry et al., 2004). A 688 bp region of the 5D end 
mtCOI gene has been proposed as a possible marker evolving at the rate of speciation in 
animals and is the gene of choice for the genetic barcoding and identification of animal 
species (Hebert et al, 2003; Barrett and Hebert, 2005). 
1.7.4 Nuclear DNA markers 
The use of nuclear genetic markers in population genetics has been reviewed by Zhang and 
Hewitt (2002). Nuclear sequences have an advantage over mitochondrial genomes because 
they are inherited from both parents, hence can be transferred across species and population 
boundaries in hybrid zones (Zhang and Hewitt, 2003). However, recombination may distort 
the genetic content in nuclear makers leading to a false inference of genealogy (Zhang and 
Hewitt, 2003). For specific ecological studies, the choice of the marker to use is critical. 
Expressed (or coding) sequences are often more stable and useful for studying higher 
taxonomy differences since they have greater fitness costs to possible mutational changes. 
Conversely, introns may accumulate a greater extent of changes without a fitness cost and so 
find practical application in population level studies (e.g. Fritz et al., 1994; Gomez-Zurita et 
al, 2000). 
20 
Ribosomal genomes provide a uniquely powerful source of genomic information. The 
ribosomal genome consists of tandem repeats of exons, introns and internal transcribed 
spacers that are transcribed into RNA but not expressed into protein. These segments of 
DNA are widely used as data can be obtained from both sequence and secondary structure 
characteristics. Although inherited from both parents, concerted evolution events tend to fix 
changes within genomes and populations making them useful for studying population 
boundaries (Zhang and Hewitt, 2003). The structure and sequence of coding regions of 
rDNA are well conserved, while spacers are highly variable. Therefore, primers developed 
from the conserved region in one species can be used to amplify the variable regions for fine-
 scale population studies in an even remotely related species (Fritz et al., 1994). Similarly, 
sequences of the conserved regions are useful for studying deep phylogenetic relationships 
(e.g. Caterino and Vogler, 2002). 
1.7.5 Microsatellite markers 
Microsatellite sequences are short core (2-1 Obp) tandem repeats up to 500 bp long scattered 
throughout the genome (Weber and May, 1989). They mainly occur in non-coding DNA, 
closely associated with conserved loci (Ellegren, 2004). These markers are co-dominant and 
hyper-variable and their inheritance can be determined by simple mating studies. The level 
of genetic variability produced has made these markers, the most popular in ecological 
genetics studies. Microsatellites are useful Mendelian markers, with a fairly simple modes of 
evolution (Estoup et al., 1995; Ellegren, 2004) and are useful in population genetics, 
parentage studies and reconstruction of recent demographic events (Selkoe and Toonen, 
2006). 
An important set back in the use of microsatellites is the requirement ofde novo isolation of 
new markers if they are not yet available for a particular species or its congeneric relatives. 
Since each marker isolated comprises just a single data point, several markers must be 
21 
isolated and tested for use in population genetic studies (Zane et al, 2002; Selkoe and 
Toonen, 2006). However, once primers have been developed, they may be used in closely 
related taxa (Ellegren, 2004). Advances in sequencing technologies (Hudson, 2008) have 
significantly reduced the technical effort and cost of required in isolating new microsatellites 
markers (Zane et al, 2002; Selkoe and Toonen, 2006; Santana et al., 2009). 
1.8 Population genetics in biological control 
The availability of genetic tools for population genetics has increased the precision with 
which we can now study the interactions between crops, pests and their natural enemies in 
diverse environmental conditions. Roderick and Navajas (2003) have reviewed the utility of 
genetic information in biological control. Biological control would benefit from an accurate 
identification of the geographic source of an invasive species, tracking of genetic changes in 
populations since introduction and guided maintenance of genetic diversity of natural 
enemies. Molecular markers have been used to assess the genetics of population 
establishment, audit and improvement of approaches to biological introductions (Omwega 
and Overholt, 1996) and to study the basis of pest-natural enemy interactions (Gitau et al., 
2007). It is appropriate to focus on the source population of the invader to recover closely 
coadapted populations of natural enemies. Similarly, bioprospecting for natural enemies 
would be improved by ecological matching and genetic typing of ecological populations of 
the natural enemies to improve establishment in the areas of introduction. The chances of 
success may also be increased through breeding and genetic modification for better 
establishment and survival of the natural enemies, their effectiveness (especially pathogens) 
and ability to locate their hosts (Handler and Beeman, 2003). Advances in molecular 
biology, the development of multilocus markers and sequence data as well as analysis 
protocols has improved our ability to detect the genealogical origin, population history and 
underlying genetic interactions of populations and species. Similarly, the information of 
22 
insect species genome sequencing projects, including Drosophila melanogaster (Diptera; 
Drosophilidae), Apis mellifera (Hynemoptera: Apidae), Bombyx mori (Lepidoptera: 
Bombycidae), Anopheles gambiae (Diptera: Culicidae) and the post harvest pest Tribolium 
castaneum (Coleoptera: Tenebrionidae), have availed the available information for the 
genetic manipulation and monitoring of pests and their natural enemies. 
The effectiveness of biological depends on the establishment and adaptation of the 
introduced natural enemies to exploit the pest in the new environment (Roderick and 
Navajas, 2003). If adaptation of well characterised natural enemies is important, then it is 
essential to preserve a high level of genetic diversity in introduced natural enemies too 
(Roderick and Navajas, 2003). Establishment of an introduced population is often correlated 
with the propagule pressure (also called introduction effort) which is a product of the 
introduced numbers (propagule size) and the genetic diversity (Blackburn and Duncan, 2001; 
Lockwood et al., 2005; Bacigalupe, 2008). Rapid adaptation of natural enemies has been 
shown for microorganisms but less for insect predators and parasitoids (Holt and Hochberg, 
1997). Such adaptation depends on the availability underlying genetic diversity upon which 
natural selection in the new environment would act (Lockwood et al., 2007; Bacigalupe, 
2008). A geographical analysis of the genetic architecture of the populations of species 
would therefore provide important information in facilitating biological control efforts. 
23 
1.9 References 
Adams, K. L. and Palmer, J. D. (2003). Evolution of mitochondrial gene content: gene loss 
and transfer to the nucleus. Molecular Phylo genetics and Evolution 29: 380 - 395. 
Anonymous (2003). The status of the larger grain borer in Kenya. Report of the Larger 
Grain Borer Task Force, July, 2003. 
Ayertey, J. N., Meikle, W. G., Borgemiester, C , Camara, M. and Markham, R. H. 
(1999). Studies on predation of Prostephanus truncatus (Horn) (Col., Bostrichidae) 
and Sitophilus zeamais Mots. (Col., Curculionidae) at different densities on maize by 
Teretriosoma nigrescens Lewis (Col., Histeridae). Journal of Applied Entomology 
123 :265-271 . 
Bacigalupe, D. L. (2008). Biological invasions and phenotypic evolution: a quantitative 
genetic perspective. Biological Invasions doi 10.1007/sl0530-0089411-2. 
Barrett, R. D. EL, and P. D. N. Hebert, P. D. N. (2005). Identifying spiders through DNA 
barcodes. Canadian Journal of Zoology 83: 481- 491. 
Berry, S. D. Fondong, V. N., Christine, R., Rogan, D., Fauquet, C. M. and Brown, J. K. 
(2004). Molecular Evidence for five distinct Bemisia tabaci (Homoptera: 
Aleyrodidae) geographic haplotypes associated with cassava plants in sub-Saharan 
Blackburn, T. M. and Duncan, R. P. (2001). Determinants of establishment success in 
introduced birds. Nature 44: 195 — 197. 
Borgemeister, C , Djossou F., Adda, C , Schneider, H., Djomamou, B., Degbey, P., 
Azoma, YL and Markham, R. H. (1997a). Establishment, spread and impact of 
Teretriosoma nigrescens (Coleoptera: Histeridae), an exotic predator of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in south-western 
Benin. Environmental Entomology 26: 1405 — 1415. 
Borgemeister, C , Meikle, W. G., Scholz, D., Adda, C , Degbey, P. and Markham, R. H. 
(1997b). Seasonal and weather factors influencing the annual flight cycle of 
24 
Prostephanus truncatus (Coleoptera: Bostrichidae) and its predator Teretriosoma 
nigrescens (Coleoptera: Histeridae) in Benin. Bulletin of Entomological Research 87: 
239-246 . 
Borgemeister, C , Schneider, H., Affognon, H., Schulthess, F., Bell, A., Zweigert, M. E., 
Poehling, H. M. and Setamou, M. (2001). Impact Assessment of Teretrius 
nigrescens Lewis (Col.: Histeridae) in West Africa, A predator of the larger grain 
borer, Prostephanus truncatus (Horn) (Col.: Bostrichidae). In: 1st Symposium on 
Biological Control of Arthropods, 17 t h -21 s t September 200 lHonolulu, Hawaii. 
Boxall, R. A. (2002). Damage and loss caused by the larger grain borer, Prostephanus 
truncatus. Integrated Pest Management Reviews 7: 105 — 124. 
Boye, J. (1988). Autoecological investigations on the behaviour of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in Costa Rica. PhD 
Thesis, Institute for Phytopathology, Christian Albrechts University, Kiel. 
Boye, J. A., Durde, S., Kiel, H. Laborius, G. A. and Schulz, F. A. (1988). The possibilities 
of biological control of the larger grain borer (Prostephanus truncatus [Horn]) in 
Africa. In: G. G. M. Schulten and A. J. Toet [eds]: Technical Papers Presented at the 
Workshop on the Containment and Control of the Larger Grain Borer, Arusha, 
Tanzania, 16- 21, May 1988. Ministry of Agriculture and Livestock Development 
(Tanzania). 
Brugmans, B., van der Hulst, R. G. M., Viscer, R. G. F., Lindout, P. and Eck, H. J. 
(2003). A new and versatile method for the successful conversion of AFLP markers 
into simple locus markers. Nucleic Acids Research 31: e55. 
Caterino, M. S. and Vogler, A. P. (2002). The phylogeny of Histeroidea (Coleoptera: 
Staphyliniformia). Cladistics 18: 394 -415 . 
25 
CeniSi J. L., Perez, P. and Fereres, A. (1993) Identification of aphids (Homoptera: 
Aphididae) species and clones by Random Amplified Polymorphic DNA (RAPD-
 PCR). Annals of the Entomological Society of America 86: 545 — 550. 
Cervera, M. T., Cabezas, J. A., Simon, B., Martinez-Zapater, J. M., Beitia, F. and Cenis, 
J. L. (2000). Genetic relationships among biotypes of Bemisia tabaci (Hemiptera: 
Aleyrodidae) based on AFLP analysis. Bulletin of Entomological Research 90: 391 -
 396. 
Cowley, R. J., Howard, D. C. and Smith, R. H. (1980). The effect of grain stability on 
damage caused by Prostephanus truncatus (Horn) and three other beetle pests of 
stored maize. Journal of Stored Products Research 16: 75 — 78. 
De Barro, P. J. and Driver, F. (1997). Use of RAPD PCR to distinguish the B biotype from 
other biotypes of Bemisia tabaci (Gennadius) (Hemitera: Aleyrodidae). Australian 
Journal of Entomology 36: 149 — 152. 
De Barro, P. J. and Hart, P. J. (2000). Mating interactions between two biotypes of the 
whitefly Bemisia tabaci (Hemiptera: Aleyrodidae) in Australia. Bulletin of 
Entomological Research 90: 103 - 2 1 8 . 
Degallier N.' and Gomy Y. (1983). Caracteres. generaux et techniques de recolte des 
Coleopteres Histeridae. L 'Entomologiste 39: 9 - 17 
Dendy, X, Dobie, P., Saidi, J. A., Smith, J. and Urono, B. (1991). Trials to assess the 
effectiveness of new synthetic pheromone mixtures for trapping Prostephanus 
truncatus (Horn) (Coleoptera: Bostrichidae) in maize stores. Journal of Stored 
Products Research 27: 187-1-91. 
Dunstan, W. R. and Magazini, A. I. (1981). Outbreaks and new records, United Republic of 
Tanzania: the larger grain borer on stored products. FAO Plant Protection Bulletin 
29: 8 0 - 8 1 . 
26 
Ellegren, H. (2004). Microsatellites: Simple sequences with complex evolution Nature 
Reviews - Genetics 5: 435 - 445. 
Estoup, A., Jarne, P. and Cornuet, J. M. (2002). Homoplasy and mutation model at 
microsatellite loci and their consequences for population genetics analysis. Molecular 
Ecolology 11:1591 -1604 . 
Estoup, A., SchoII, A., Pouvreau, A. and Solignac, M. (1995). Monoandry and polyandry 
in bumble bees (Hymenoptera: Bombidae) as evidenced by highly variable 
microsatellites. Molecular Ecology 4: 89 — 93. 
Fadamiro, H. Y. and T. D. Wyatt. (1995). Flight initiation of Prostephanus truncatus in 
relation to time of day, temperature, relative humidity and starvation. Entomologia 
Experimental et Applicata 75: 273-277. 
Fang, L., Subramanyam, B. and Arthur, F. H. (2002). Effectiveness of Spinosad on four 
classes of wheat against five stored-product insects, Journal of Economic 
Entomology 95: 640 - 650. 
FAO (2004). FAO Statistics, FAOSTAT Agriculture Database, Agricultural production 
database website: httpf/faostat.org/faostat (accessed 7th April 2005). 
Farrel G. and Schulten G. G. M. (2002). The larger grain borer in Africa; a history of 
efforts to limit its impact. Integrated Pest Management Reviews 1:61 — 84. 
Fritz, G. N., Conn, J., Cockburn, A. and Seawright, J. (1994). Sequence analysis of 
ribosomal ITS2 from populations of Anopheles nuneztovari (Diptera: Culiciade) 
Molecular Biology and Evolution 11: 406 - 416. 
Gawell, N. J. and Bartlett, A. C. (1993). Characterisation of differences between whitefiies 
• using RAPD-PCR. Insect Molecular Biology 2: 3 3 - 3 8 . 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). 
Release and establishment of the predator Teretriosoma nigrescens Lewis for the 
27 
biological control of Prostephanus truncatus (Horn) in Kenya. African Crop Science 
Journal 4: 325 - 337. 
Gitau, C. W. Gundersen-Rindal, D., Pedroni, M., Mbugi, P. J. and Dupas, S. (2007). 
Differential expression of the CrVl haemocyte inactivation-associated polydnavirus 
gene in the African maize stem borer Busseola fusca (Fuller) parasitized by two 
biotypes of the endoparasitoid Cotesia sesamiae (Cameron). Journal of Insect 
Physiology 53: 676-684 
Golob, P. (1988). Current status of the larger grain borer Prostephanus truncatus (Horn) in 
Africa. Insect Science and its Application 9:737 — 745. 
Golob, P. (2002). Chemical, physical and cultural control of Prostephanus truncatus. 
Integrated Pest Management Reviews 7: 245 — 277. 
Gomez-Zurita, J., Juan;, C. and Petitpierre, E. (2000). Sequence, secondary structure and 
phylogenetic analyses of the ribosomal internal transcribed spacer 2 (ITS2) in the 
Timarcha leaf beetles (Coleoptera; Chrysomelidae). Insect Molecular Biology 9: 591 
- 6 0 4 . 
Gueye, M. T., Goergen, G., Badiane, D. and Hell, K. and Lamboni, L. (2008). First report 
on occurrence of the larger grain borer Prostephanus truncatus (Horn) (Coleoptera: 
Bostrychidae) in Senegal. African Entomology 16: 309-311. 
Hadrys, H., Malick, M. and Schierwater, B. (1992). Applications of random amplified 
polymorphic DNA (RAPD) in molecular ecology. Molecular Ecology 1: 55 - 64 
Handler, A. M. and Beeman, R. W. (2003). United States Department of Agriculture 
Agricultural Research Service: advances in the molecular genetic analysis of insects 
and their application to pest management. Pest Management Science 59: 728—735 
Hebert, P.D.N., Cywinska, A., Ball, S. L. and J. R. de Waard, J. R. (2003). Biological 
identifications through DNA barcodes. Proceedings of the Royal Society of London 
Serie B: Biological Sciences 270: 313 - 321. 
28 
Heckel, D. G., Gahan, L. J., Daly, J. C. and Trowell, S. (1998). A genomic approach to 
understanding Heliothis and Helicoverpa resistance to chemical and biological 
insecticides. Philosophical Transactions of the Royal Society of London - Serie B: 
Biological Sciences 353: 1713 - 1722. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
predatory beetle Teretrius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 - 306. 
Hodges, R. J. (2002). Detection and monitoring of the larger grain borer, Prostephanus 
truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest Management Reviews 7: 
223-243. 
Hodges, R. J., Addo, S. and Birkinshaw, L. (2003). Can observation of climatic variables 
be used to predict the flight dispersal rates of Prostephanus truncatus1? Agricultural 
and Forest Entomology 5: 123-135. 
Hodges, R. J., Dunstan, W. R., Magazini, I. and Golob, P. (1983). An outbreak of 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in East Africa. Protection 
Ecology 5: 183-194. 
Hodges, R. J., Hall, D. R., Mbugua, J. N. and Likhayo, P. W. (1998). The response of 
Prostephanus truncatus (Coleoptera: Bostrichidae) and Sitophilus zeamais 
(Coleoptera: Curculionidae) to pheromone and synthetic maize volatiles as lures in 
crevice traps. Bulletin of Entomological Research 88: 131 - 139. 
Hoist, N. and Meikle, W. (2002). Teretrius nigrescens against the larger grain borer 
Prostephanus truncatus in African maize stores: biological control at work? Journal 
of Applied Ecology 40: 307 - 319. 
Holt, R. D. and Hochberg, M. E. (1997). When is biological control evolutionarily stable 
(or is it)? Ecology 78: 1673 - 1683. 
29 
Hoppe, T. (1986). Storage insects of basic food grains in Honduras. Tropical Science 26: 25 
- 3 8 . 
Hudson, M. E. (2008). Sequencing breakthroughs for genomic ecology and evolutionary 
biology. Molecular Ecology Resources 8 : 3 - 1 7 . 
Kibogo, H. G. (2005).Identification of fruitfly species in the genera Ceratitis and 
Trirhitirum (Diptera: Tephritidae) using molecular markers. MSc Thesis, Jomo 
Kenyatta University of Agriculture and Technology, 13 0pp. 
Krall, S. (1984). A new threat to farm-level maize storage in West Africa: Prostephanus 
truncatus (Coleoptera: Bostrichidae). Tropical Stored Products Information 50: 26 -
 31. 
Legg, J. P. (1996). Host-associated strains within Ugandan populations of Bemisia tabaci 
(Gennadius) (Homoptera: Aleyrodidae). Journal of Applied Entomology 99: 13 - 23. 
Leliveldt, B. (1990). Natural enemy complex of Prostephanus truncatus (Horn) (Coleoptera 
Bostrichidae). PhD Thesis, Freie University, Berlin. 
Likhayo, P. W. and Hodges, R. J. (2000). Field monitoring Sitophilus zeamais and 
Sitophilus oryzae (Coleoptera: Curculionidae) using refuge and flight traps baited 
with synthetic pheromone and cracked wheat. Journal of Stored Products Research 
36:341-353. 
Lima, L. H. C , Campos, L., Moretzsohn, M. C , Navia, D. and de Oloviera, M. R. V. 
(2002). Gebetic diversity of Bemisia tabaci (Genn.) populations in Brazil revealed by 
RAPD markers. Genetics and Molecular Biology 25: 217 — 223. 
Lockwood, J. L., Cassey, P. and Blackburn, T. (2005). The role of propagule pressure in 
explaining species invasions. Trends in Ecology and Evolution 20: 223—228. 
Lockwood, J. L., Hoopes, M. F. and Marchetti, M. P. (2007). Invasion Ecology. Blackwell, 
Oxford 
30 
Loxdale, H. D. and Lushai, G. (1998). Molecular markers in entomology. Bulletin of 
Entomological Research 88: 557 — 600. 
Lunt, D. H., Zhang, D. X., Szymura, J. M. and Hewitt, G. M. (1996). The insect 
cytochrome oxidase 1 gene: evolutionary patterns and conserved primers for 
phylogenetic studies. Insect Molecular Biology 5: 153 - 165. 
Lynch, M. and Milligan, B. G. (1994). Analysis of population structure with RAPD 
markers. Molecular Ecology 3: 91 - 99. 
Magrath, P. A. Compton, J. A. F.; Motte, F. F. and Awuku, M. (1996). Coping with a 
new storage pest: The impact of the larger grain borer in Eastern Ghana. Cantham: 
Natural Resources Institute UK.. 
Marshed-Kharusy, M. N. and Dawah, H. A. (1999). Geographical variation in body size 
and the ability of Prostephanus truncatus (Horn) to damage stored maize. In: Zuxun, 
J., Quan, L., Yongsheng, L., Xiaanchang, T., Lianghua, G (eds) Stored Product 
Protection. Proceedings of the 7th International Working Conference on Stored 
Product Protection - Volume 2. 14-19 October 1998, Beijing, China. 
Mazur, S. (1997). A world catalogue of Histeridae (Coleoptera: Histeroidea). Pensoft, 
373pp. 
McPheron, B.. A. and Han, H. Y. (1997). Phylogenetic analysis of North American 
Rhagolettis (Diptera: Tephritidae) and related genera using mitochondrial DNA 
sequence data. Molecular Phylogenetics and Evolution 7: 1 — 16. 
Meikle, W. G., Hoist, N., Degbey, P. and Oussou, R. (2000). Evaluation of sequential 
sampling plans for the larger grain borer (Coleoptera: Bostrichidae) and the maize 
weevil (Coleoptera: Curculionidae) and of visual grain assessment in West Africa. 
Journal of Economic Entomology 93: 1822 — 1831. 
31 
Meikle, W. G., Rees, D. and Markham, R. H. (2002). Biological control of the larger grain 
borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest 
Management Reviews 7: 123 - 138. 
Moritz, C , Dowling, T. E. and Brown, W. M. (1987). Evolution of animal mitochondrial 
DNA: relevance for population biology and systematics. Annual Review of 
Ecological Systems 18: 269 - 292. 
Moya, A., Guirao, P., Cifuentes, D., Beitia, F. and Cenis, L. (2001) Genetic diversity of 
Iberian populations of Bemisia tahaci based on random amplified polymorphic DNA 
-polymerase chain reaction. Molecular Ecology 10: 891 - 897. 
Nang'ayo, F. L. O. (1996). Ecological Studies on the Larger Grain Borer in Savanna 
Woodlands of Kenya. PhD Dissertation, University of London, 179pp. 
Nang'ayo, F. L. O. Hill, M. G. and Wright, D. J. (2002). Potential hosts of Prostephanus 
truncatus (Horn) (Coleoptera; Bostrichidae) among native agroforestry trees in 
Kenya. Bulletin of Entomological Research 92: 499 — 506. 
Nang'ayo, F. L. O. Hill, M. G., Chandi, E. A., Chiro, C. T., Nzeve, D. N. and Obiero, J. 
(1993). The natural environment as a reservoir of the larger grain borer Prostephanus 
truncatus (Horn) (Coleoptera: Bostrichidae) in Kenya. African Crop Science Journal 
1 :39 -47 . 
Nansen, C. and Meikle, W. G. (2002). The biology of the larger grain borer, Prostephanus 
truncatus (Horn) (Col.: Bostrichidae). Integrated Pest Management Reviews 7: 91— 
104. 
Nansen, C , Meikle, W. G. and Korie, S. (2002). Spatial analysis of Prostephanus truncatus 
(Horn) (Coleoptera: Bostrichidae) flight activity near maize stores and different forest 
types in southern Benin. Environmental Entomology 30: 1135 — 1143. 
32 
Nansen, C , Meikle, W. G. Tigar, B., Harding, S. and Tchabi A. (2004). Non-agricultural 
hosts of Prostephanus truncatus (Coleoptera: Bostrichidae) in West African Forest. 
Annals of the Entomological Society of America 97: 481 — 491. 
Omondi, A. B., Obeng-Ofori, D., Danquah, E. Y. and Kyerematen, R. A. (2005). Mating 
interactions between okra and cassava biotypes of Bemisia tabaci (Homoptera: 
Aleyrodidae) on eggplant. International Journal of Tropical Insect Science 25: 159 — 
167. 
Omondi, A. B., Obeng-Ofori, D., Danquah, E. Y. and Offei, S. K. (2004). Genetic 
diversity and oviposition host preference of Bemisia tabaci associated with four crop 
hosts in Ghana. Journal of the Ghana Science Association 6: 105 — 116. 
Omwega, C. O. and Overholt W. A. (1996). Genetic changes occurring during laboratory 
rearing of Cotesia flavipes Cameron (Hymenoptera: Braconidae) an imported 
parasitoid for the control of graminaceous stem borers in Africa. African Entomology 
4 :231 -237 . 
Oussou, R. D., Meikle, W. G. and Markham, R. H. (1998). Factors affecting larval 
survivorship and development rate of Teretriosoma nigrescens Lewis. Insect Science 
and it Applications, 18: 53-58. 
Pantenius, C. U. (1988). Storage losses in traditional maize granaries in Togo. Insect 
Science and it Application 9: 725 — 735. 
Poshko, M. (1994). Research into the biology and host preference of the predator 
Teretriosoma nigrescens, a potential natural antagonist of the larger grain borer, 
Prostephanus truncatus. Eschborn: Deutsche Gesellschaft fur Technische 
Zusammenarbeit (GTZ) GmbH. 
Ramirez-Martinez, M., de Alba-Avila, A. and Ramirez-Zurbia, (1994). Discovery of the 
larger grain borer in- a tropical deciduous forest in Mexico. Journal of Applied 
Entomology 118: 354 - 360. 
33 
Rees, D. P. (1985). The life history of Teretriosoma nigrescens Lewis (Coleoptera: 
Histeridae) and its ability to suppress population of Prostephanus truncatus (Horn) 
(Coleoptera: Bostrichidae). Journal of Stored Products Research 21: 115-118. 
Rees, D. P. (1987). Laboratory studies on predation by Teretriosoma nigrescens Lewis (Col., 
Histeridae) on Prostephanus truncatus (Horn.) Col.: Bostrichidae) infesting maize 
cobs in the presence of other storage pests, Insect Science and its Application 23: 191 
- 1 9 5 . 
Rees, D. P. (1991). The effect of Teretriosoma nigrescens Lewis (Coleoptera: Histeridae) on 
three species of Bostrichidae infesting shelled maize. Journal of Stored Products 
Research 27: 83 - 86. 
Rees, D. P., Rivera, R. R. and Rodriguez, F. S. H. (1990). Observations on the ecology of 
Teretriosoma nigrescens (Lewis) (Col., Histeridae) and its prey Prostephanus 
truncatus (Horn) (Col., Bostrichidae) in the Yucatan peninsula, Mexico. Tropical 
Science 30: 153-165. 
Roderick, G.K. and Navajas, M. (2003). Genes in new environments: Genetics and 
evolution in biological control. Nature Reviews - Genetics 4: 889-899. 
Rugienius, R., Siksnianas, T., Stanys, V., Gelvonanskiene, D. and Bendokas, V. (2006). 
Use of RAPDs and SCAR markers for identification of Strawberry genotypes 
carrying red stele (Phytopthorafragariae) resistance gene RpfL. Agronomy Research 
4 :335 -339 . 
Santana, Q. C , Coetzee, M. P. A, Steenkamp, E. T., Mlonyeni, O. X., Hammond, G. N. 
A., Wingfield, M. J. and Wingfield, B. D. (2009). Micro satellite discovery by deep 
sequencing of enriched genomic libraries. Bio Techniques 46:217 - 223. 
Schneider, H., Borgemeister, C. Setamou, M., Affognon, H., Bell, A., Zweigert, M. E., 
Poehling, H. and Schulthess, F. (2004). Biological control of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) by its predator Teretrius 
34 
nigrescens (Lewis) (Coleoptera: Histeridae) in Togo and Benin. Biological Control 
30:241-255 . 
Scholz, D., Borgemeister, C. and Poehling, M. (1998). EAG responses of Prostephanus 
truncatus and its predator Teretriosoma nigrescens to the borer-produced aggregation 
Pheromone. Physiological Entomology 23: 265 - 273. 
Scholz, D., Borgemeister, C , Markham, R. H. and Poehling, M. (1997). Flight initiation 
in Prostpehanus truncatus: influence of population density and aggregation. 
Entomologia Experimental at Applicata 85: 237 — 245. 
Selkoe, K. A. andToonen, R. J. (2006). Microsatellites for ecologists: a practical guide to 
using and evaluating microsatellites markers. Ecology Letters 9: 615 - 629. 
Shao, R. and Baker, S. C. (2003). The highly rearranged mitochondrial genome of plague 
thrips, Thrips imagines (Insecta: Thysanoptera): convergence of two novel gene 
boundaries and an extraordinary arrangement or rKNA genes. Molecular Biology and 
Evolution2Q: 362-370 . 
Shires, S. W. (1977). Ability of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) 
to damage and breed on several food commodities. Journal of Stored Products 
Research 13: 205 - 2 0 8 . 
Stewart-Jones, A., Hodges, R. J., Farman, I. D. and Hall, D. R. (2006). Solvent extraction 
of cues in the dust and frass of Prostephanus truncatus and analysis of behavioural 
mechanisms leading to arrestment of the predator Teretrius nigrescens. Physiological 
Entomology 31: 63—72. 
Thao, M. L., Baumann, L. and Baumann, P. (2004). Organisation of the mitochondrial 
genomes of whiteflies, aphids and psyllids (Hemiptera, Stenorrhyncha). BioMed 
Central Evolutionary Biology 4: 25 - 30. 
35 
Throne, J. E., Hallman, G. J., Johnson, J. A. and Follett P. A. (2003). Post-harvest 
entomology research in the United States Department of Agriculture-Agricultural 
Research Service, Pest Management Science 59: 619 - 628. 
Tigar, B. J., Osborne, P. E., Key, G. E., Flores-S., M. E. and Vazquez-A., M. (1994). 
Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) by 
its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bulletin of 
Entomological Research 84: 555 — 565. 
Tyler, P. S. and Hodges, R. J. (2002). Phytosanitary measures against the Larger Grain 
Borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae), in international 
trade. Integrated Pest Management Reviews 7: 279 -289 . 
Weber, J. L. and May, P. E. (1989). Abundant class of human DNA polymorphism which 
can be typed using polymerase chain reaction. American Journal of Human Genetics 
44:388-396 . 
Welsh, J., Patterson, C. and McClelland, M. (1990). DNA polymorphisms amplified by 
arbitrary primers are useful as genetic markers. Nucleic Acids Research 19: 9213 — 
7218. 
Williams, J. G.K., Kubelik, A. R., Livak, K. J., Rafalski, J. A. and Tingey, S. V. (1990). 
DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. 
Nucleic Acids Research 18: 6531 — 6535. 
Xing, J. and Guyer, D. (2008). Detecting internal infestation in tart cherry using 
transmittance spectroscopy. Postharvest Bio logy andTechno logy 49: 411 - 4 1 6 . 
Zane, L., Bargelloni, L. and Patarnello, T. (2002). Strategies for Microsatellite isolation: a 
review. Molecular Ecology 1 1 : 1 - 1 6 . 
Zhang, D. -X. and Hewitt, G. M. (1997). Assessment of the universality of a set of 
conserved mitochondrial COI primers. Insect Molecular Biology 6: 143 150. 
36 
Zhang, D.X. and Hewitt, G. M. (2003). Nuclear DNA analysis in genetic studies of 
populations: practice, problems and prospects. Molecular Ecology 12: 563 — 584. 
37 
C H A P T E R T W O 
General Materials and Methods 
2.1 Study materials 
2.1.1 Insects 
Samples of Prostephanus truncatus collected from maize stores and the field in Kenya were 
used to establish cultures on maize based on the region of origin at the Animal Rearing and 
Quarantine Unit (ARQU) at ICEPE. Starter colonies of Teretrius nigrescens were obtained as 
outlined in Table 2.1. The KARI (Kenya Agricultural Research Institute) strain was 
subcultured from the remnant colony of the T. nigrescens originally coming from Mexico, 
released in Kenya in 1992 and now maintained at KARI, Kiboko Field Station (Giles et al., 
1996). The Benin strain was obtained from International Institute of Tropical Agriculture 
(IITA), Abomey-Calavi, Benin. This population had been established from adults recovered 
from granaries in Benin, three years after field releases in Togo (C. Atcha, WARD A, 
personal communication). The Togo releases had been recovered from Sardinal, Guanacaste, 
north-western Costa Rica (Boye, 1990). These populations were assumed to be descendants 
of insects that had been released in Kenya, Benin and Togo, respectively, in biological 
control efforts between 1990 and 2000. Three geographical strains from colonies established 
using insects recovered from granaries at El Batan (hereafter referred to as the Batan strain), 
Oaxaca and Tlatizaspan in Mexico were imported from CiMMYT, Mexico. Three other 
populations were collected from Teupasenti, Gualaso and Yoro in Honduras between 2006 -
 2007 and used to found laboratory cultures. The Kiboko colony was established using insects 
recovered in 2007 from the Kiboki Range, eastern Kenya within the original outbreak and 
biocontol area (Giles et al., 1996). The 'Store' colony was accidentally found in LGB-
 infested wooden pegs from a store in ICEPE. These pegs had reportedly been used for field 
sampling cereal stem borers and their parasitoids a year earlier. The origin of this population 
38 
was unknown. All T. nigrescens strains were reared on P. truncatus raised on maize at the 
Animal Rearing and Quarantine Unit (ARQU) at ICIPE, Duduville, Nairobi, Kenya. The 
LGB and T. nigrescens were reared using a modification of protocols described by Giles et 
al. (1996) and Atcha and Borgemeister (unpublished IITA manual). 
Three populations of alcohol-preserved samples were also used. The Ghana strain was 
obtained from a colony maintained at the Natural Resources Institute (NRI), UK, established 
with insects initially recovered from the field in Ghana in 1998 following releases in Togo, 
Benin and possibly Volta Region, Ghana (R. Hodges, NRI, personal communication). 
Malawi samples, originated from a colony maintained at the Crop Protection Agency, 
Ministry of Agriculture, Malawi. This colony was a subculture of the Benin colony 
(Anonymous, 1999). The Mombasa population comprised of five individuals caught on a 
pheromone-baited sticky trap at Mombasa. These traps were set after the discovery of the 
"Store" population, since the substrate material was it was discovered in was had been 
reportedly used in this region. 
39 
Table 2.1: Field sources and culturing history of the populations T. nigi'escens used in this study 
Name Field Origin Type Date 
sampled 
Rearing history Remarks 
KARI Neuvo Leone, 
Mexico 
Laboratory colony 1990 KARI colony, 16 years Earlier released for BC in Kenya 
Benin Guanacaste, 
Costa Rica 
Laboratory population 1989 IITA colony 18 years Established from field samples 
collected in Benin 
Ghana1 Ghana Laboratory colony 1989 NRI colony from field samples 
for 10 year culturing 
Possible progeny of Benin samples 
Teupasenti Honduras Directly recovered from the 
field 
2007 • Laboratory culturing for 1 year Originating from south western 
Honduras 
Gualaso Honduras Directly recovered from the 
field 
Directly recovered from the 
2007 Laboratory culturing for 1 year Mid altitude Honduras population 
Yoro Honduras 2007 Laboratory culturing for 1 year High altitude Central Honduras 
field population 
Oaxaca Mexico Young laboratory colony 2003 CiMMYT colony, 3 years Mid altitude population, southern 
Mexico 
ElBatan Mexico Young laboratory colony 2003 CiMMYT rearing for 3 years Low altitude population Central 
Mexico 
Tlaltizaspan Mexico Young laboratory colony 2003 CiMMYT rearing for 3 years High altitude population, upper central 
Mexico 
Malawi1 Costa Rica Laboratory colony 1990 From Benin colony, rearing 
for 9 years 
Sub-culture of Benin colony 
Store Unknown Recovered from LGB- 2006 1 year Antiquity unknown, Possible 
(Kenya) infested wood in Project 
Store 
Direct recovery from field 
descendants of KARI releases) 
Field Kenya 2007 Laboratory rearing 1 year Possible descendants of KARI releases 
Mombasa Kenya Field caught adults 2007 None Caught on sticky traps 
Ethanol-preserved samples, only used for molecular studies. The size of the original field-recovered population of the CiMMYT samples was 
about 200 beetles. The original size number of traps used and sampling duration for the recovery of KARI and Benin populations are not known. 
40 
Costa Rica 
Panama 
Figure 2.1: Map of Central America showing the field sources of Teretiiiis nigrescens used in this study. The Kr and samples from Costa Rica 
had been released in Africa in the 1990s founding (KARJ, Benin. Ghana and Malawi populations). Population abbreviations: (Kr - KARl, Ts -
 Tlaltizaspan, Ox - Oaxaca, Te -Teupasenti. 
41 
2.2 Monitoring and recovery of P. truncatus and T. nigrescens 
Pheromone-baited traps were used to monitor and recover P. truncatus and T. nigrescens 
samples in Kenya and Honduras. Disposable sticky traps (Pherocone III Delta traps, Trece, 
Inc.) were used for monitoring of the flight activity of P. truncatus. This trap consists of a 
folded card loaded, sticky on the inner surface. The traps were baited with Pherocon Lures, a 
moisture of the two components of the LGB aggregation pheromones Trun-calll (1 -
 Methylethyl (2E)-2-methyl-2-pentenoate and Trun-call 2 (1 methylethyl (2E,4E)-2,4-
 dimethyl-,4-heptadienoate). Delta traps were replaced every three weeks. 
For live recoveries of T. nigrescens and LGB in Kenya and Honduras, a modification of the 
Japanese beetle funnel trap was used. This traps consisted of a bucket receptacle, funnel and 
flat circular top. The roof had a holder into which the pheromone vial was inserted and 
secured. And the three parts fitted into the bucket. The trap bucket was loaded with 2 grams of 
LGB frass and 5 grams of maize grains. The frass and maize grains had been sterilised in an 
oven at 40 °C for four hours (Jembere et al., 1995). These traps were left in situ throughout the 
survey and recovery period and serviced every fortnight. Servicing funnel traps involved 
removal of bucket contents for laboratory analysis and replacement of both the pheromone 
lure and the frass-and-maize bait. 
2.3 Outgroup species 
Other Histeridae species were directly sampled from the field. Teretrius americanus LeConte 
(= T. latebricola Lewis) was collected from bark of fallen trees around Wisconsin by Jeffrey 
Grubber, Russell Labs, Madison (USA). Saprophilic histerid samples were collected around 
Nairobi from decomposing cadavers of a cat and bird between the third and seventh days post 
mortem as well as from pitfall traps baited with liver. Histerids were also sampled from the 
soil underneath herbivore dung patches, compost heaps and using a flight intercept trap set 
42 
near fallen tree trunks at the Kiboko Range (Figure 2.2). Outgroups species selected for this 
study were: T. americanus, Acriius nigricornis (Hoffmann), Chaetabraeus sp. (Mazureus), 
Saprinus bicoloroides Dahlgren, Saprinus splendens (Paykull), Atholus confinis (Erichson) 
and Carcinops pumilio (Erichson). These species were selected on the basis of their 
phylogenetic relationship with T. nigrescens (Caterino and Vogler, 2004) and sampled 
according to methods used by Degallier and Gomy (1983). All insect identities were 
determined by Dr. Peierpaolo Vienna, Italy. Insect samples for DNA extraction were 
transferred to absolute ethanol and stored at -20 °C till ready for use. 
2.4 Insect Rearing 
Insect cultures were set up in one-litre glass jars containing maize grains. Untreated maize 
grains (unidentified variety) was purchased locally, cleaned and winnowed before use. A 
sample of every batch of new maize was exposed to adult Sitophulis zeamais Motchulsky 
(Coleoptera: Curculionidae) overnight to infer the possibility of it having been treated with 
insecticidal compounds. The maize was then sterilised in an oven at 40 °C for 4 hours. Maize 
used for laboratory tests was treated at 40 °C for four hours, as both higher temperatures and 
longer treatment times had been found to influence the chemical and food quality properties of 
the maize and subsequent insect behaviour (Jembere et ah, 1995). Maize was allowed to cool 
on the bench for 24 hours at room conditions. The next day, 200 adult LGB were transferred 
from a healthy colony and added into the jar. This set up was either maintained at room 
conditions (about 27 - 30 °C) for seven weeks to multiply adult LGB for other experiments or 
used for T. nigrescens colony establishment. At every harvesting, T. nigrescens from several 
jars were mixed before being redistributed to fresh jars to increase genetic admixture. 
Jars used for T. nigrescens rearing were kept at room conditions for one week, for LGB brood 
establishment, after which 20 adult T. nigrescens individuals were transferred from established 
43 
colonies into each jar. This set up was either left until some T. nigrescens were needed or 
selected for experiments as desired. When a uniform age was essential, T. nigrescens adults 
were allowed 48 hours oviposition time after which they were sieved out. In all cases, the jars 
were kept in the laboratory (or incubator if for experiments) until ready for use. 
2.4.1 Recognition of life stages' 
The life stages of the larger grain borer juvenile stages of were identified as previously 
described (Farrel and Haines, 2002) (Figure 2.3). The larval stages of T. nigrescens were 
determined though observation of development and as described by Rees (1985) and the eggs 
as described by Stewart-Jones et al. (2006) (Figure 2.4). 
44 
Figure 2.2: Flight intercept trap used to sample Histeridae predating on bark-boring 
Coleoptera. Insects collected in soapy water in the yellow troughs were transferred into 
absolute ethanol every 12 hours. 
A: Adult LGB. (G. Goergen, IITA) 
D: LGB Pupa exposed from pupal case 
C: Third instar larva 
Figure 2.3: The life stages of Prostephanus truncatus. The pupa was carefully removed from a 
pupal case for photographing. The bar in B (egg) represents 0.5 mm, for the other 
photographs, the bar represents 1 mm. 
B: LGB egg 
45 
A: Freshly eclosed adult B: Egg 
D: Pupa C: First instarLarva 
Figure 2.4: The life stages of Teretrius nigrescens. For each photo, the bar included represents 
1 mm. 
46 
2.5 General molecular procedures 
2.5.1 DNA extraction 
Genomic DNA 
Genomic DNA was extracted from the head and thorax of individual T, nigrescens adults. 
Insect tissue was stored at -20 °C till ready for DNA extraction. For outgroup species, DNA 
was extracted from a single left hind leg and the rest of the specimen sent for species 
identification. Larger tissues of T. nigrescens were used to have sufficient DNA for all 
experimental procedures from the same samples. 
The DNA was extracted using the phenol-chloroform-isoamyl alcohol protocol (Sambrook et 
aL, 1989) with slight modifications. Briefly, insect tissue was homogenised in 100 u.1 of tissue-
 grinding buffer (10 mM Tris-Cl, 10 mM EDTA, 150 mM Sucrose, 60 mM NaCl, 0.5 % SDS, 
25 U/ml Proteinase; pH 7.5) and incubated at 55 °C for one hour. An equal volume of the lysis 
buffer (0.3 M Tris-Cl, 0.1 M EDTA, 0.15 M Sucrose, 60 mM NaCl, 0.75 % SDS, pH 7.5) was 
added to the homogenate, mixed gently and incubated on ice for 10 min. One volume of 
equilibrated phenol was added to the lysate, mixed by gentle inversion and centrifuged at 5000 
g for 10 min. An equal volume (200 ul) of chloroform isoamyi alcohol (24:1) was added to the 
supernatant in a fresh tube, mixed gently and centrifuged at 13000 g for 5 min. The DNA was 
precipitated by mixing the supernatant with a tenth volume of 5M sodium acetate (pH 5.2) 
followed by an equal volume of cold isopropanol and incubating it overnight at -20 °C. DNA 
was pelleted by centrifuging at 13000 g for 25 min and tilting off the alcohol gently. The 
pellet was washed twice with 70 % ethanol, each time centrifuging at 13000 g for 5 min 
before decanting off the liquid phase. It was dried in a flow hood and resuspended in 300 ul of 
TE Buffer (10 mM Tris pH 8.0, 1 mM EDTA). Extracted DNA was stored at -20 °C until 
ready for use. 
47 
Purification ofPCR products 
DNA was recovered from PCR products either from agarose gels or reacted PCR mix using 
Qiagen MinElute Gel Purification and PCR purification Kits respectively (QUIAGEN). Target 
DNA bands were excised from agarose gels, weighed and dissolved in to a 3 volumes of the 
gel solubilisation buffer in a water bath at 50 °C for 10 min. DNA was then precipitated with 
one volume of isopropanol and applied to a DNA-binding column. The column was 
centrifuged to bind to the DNA to the silica matrix while the solubilisation buffer was 
discarded. The column was washed once with the gel solubilisation buffer to remove traces of 
agarose, then washed once with the columns washing buffer (containing 70 % ethanol). DNA 
was eluted in 10 ul elution buffer (2 mM Tris, pH 8). 
PCR products were cleaned by mixing them with five parts binding buffer (containing 
isopropanol) and applying to the column as for PCR gel purification above. Columns were 
washed with the column washing buffer and DNA eluted as already described above. 
Plasm id DNA 
Plasmid DNA was extracted from overnight LB colonies using the Quiprep Spin Miniprep Kit 
(QUIAGEN). Briefly, 3 ml of the overnight culture was centrifuged at 5000 g to pellet the 
bacterial cell bodies. The pellet was resuspended in a resuspension buffer with 100 jxg/ml 
RNAse A for 5 minutes. A lysis buffer was added mixed gently and incubated at room 
temperature for 5 minutes. The denaturing buffer was then added to the lysis mix, quickly 
mixed thoroughly and centrifuged at 9000 g to separate cell debris from the lysate. The lysate 
was applied to a column and washed once with the binding buffer and column washing buffer 
respectively. Plasmid DNA was eluted in 55 fj.1 of 2 mM Tris-Cl buffer (pH 8.0). 
48 
2.5.2 Molecular cloning 
Cloning was done for eventual sequencing using pGEMT-Easy plasmid vector according to 
manufacturer's protocol. This plasmid vector subclones small direct PCR products generated 
by polymerases with A-tailing activity, without enzyme digestion. Ligation was done 
overnight at 4 °C. To transform competent cells, 2 ul of the ligation mix was mixed with 50 ul 
of freshly thawed DH5cc competent cells and left on ice for 30 min. The mix was then heat 
shocked at 42 °C in a water bath for 50 s and immediately placed back on ice for two minutes. 
This mixture was added to 950 p.1 of SOC Medium (Promega, 2005) and grown at 37 °C with 
vigorous shaking for two hours. The culture was then centrifuged at 3000 g and the bacterial 
pellet gently resuspended in 100 (il of SOC medium. This suspension was inoculated onto 
Luria-Bertani/Agar plates (containing 70 mM ampicillin, 80 fig/ml 5-Bromo-4-chloro-3-
 indolyl (3-D-galactopyranoside (X-Gal) and 0.5 mM isopropyl B-D-thiogalactopyranoside 
(IPTG)) overnight for blue-white colony screening. Positive colonies were screened by PCR 
using the same conditions for amplifying the insert (Chapters 5 and 6). 
Recombinant colonies that showed a single PCR product of definite size were grown singly in 
5 ml of Luria-Bertani medium with 50 u.g/ml ampicillin overnight at 37 °C with vigorous 
shaking to achieve high cell growth. Plasmid DNA was purified using the QIAprep Spin 
Miniprep Kit as described above. 
2.5.3 DNA sequencing reactions 
All DNA fragments were sequenced under the BigDye terminator cycle sequencing chemistry 
at a commercial facility (Macrogen Inc, Korea). Products of PCR were sequenced using the 
same primers as those used for the respective reactions, while inserts in pGEMT-E were 
sequenced using universal (Ml 3 pUC) forward and reverse primers. Ethanol-precipitated 
49 
sequencing PCR products were run on an ABI 3730x1 DNA Analyzer (Applied Biosystems). 
Where necessary, sequences were edited to remove vector sequences using VEC SCREEN 
Software (NCBI) and by manual alignment and editing. 
2.5.4 Sequence Analysis 
Apart from some very short micro satellite markers, most fragments were sequenced in both 
directions. Forward and reverse sequences were contigged using BIOEDIT. A local alignment 
of the forward sequence with the reverse complement of the reverse sequence was done 
around the overlapping termini (allowing two ends to slide). Once a perfect alignment was 
obtained, a consensus sequence was created. Ambiguous nucleotides were confirmed by 
comparison with the sequencing trace files. Sequence termini were then trimmed to include 
only portions with high chromatographic quality. 
50 
2.6 References 
Anonymous (1999). Integrated Management of Maize Pests and Diseases. Annual Report, 
Plant Health Management Division, 1999. International Institute of Tropical 
Agriculture, Cotonou, Benin. 
Boye, J. (1990). Ecological aspects of Prostephanus truncatus (Horn) (Coleoptera: Histeridae) 
in Costa Rica. Proceedings of the IITA/FAO Coordination Meeting, Cotonou, Benin 
pp. 73 - 86. 
Caterino, M. S. and Vogler, A. P. (2004). The phylogeny of Histeroidea (Coleoptera: 
Staphyliniformia). Cladistics 18: 394 -415 . 
Degallier, N. and Gomy, Y. (1983). Caracteres generaux et techniques de recolte des 
Coleopteres Histeridae. L 'Entomologiste 39: 9 — 17. 
Farrel, G., and Haines, C. P. (2002). The taxonomy, systematics and identification of 
Prostephanus truncatus (Horn). Integrated Pest Management Reviews 7: 85 — 90. 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). Release 
and establishment of the predator Teretriosoma nigrescens Lewis for the biological 
control of Prostephanus truncatus (Horn) in Kenya. African Crop Science Journal 4: 
325-337. 
Jembere, B., Obeng- Ofori, D. and Hassanali, A. (1995). Products derived from the leaves 
of Ocimum kilmandscharicum (Labiatae) as post harvest grain protectants against the 
infestation of three major stored insect product pests. Bulletin of Entomological 
Research 85: 361- 367. 
Promega (2005). Technical Manual for pGEM-T and pGEM-T Easy Vectors. Promega 
Corporation, Madison, USA. 
Rees, D. P. (1985). The life history of Teretriosoma nigrescens Lewis (Coleoptera: 
Histeridae) and its ability to suppress population of Prostephanus truncatus (Horn) 
(Coleoptera: Bostrichidae). Journal of Stored Products Research 21: 115-118 . 
51 
Sambrook, J. Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning, A Laboratory 
Manual 2nd edn. Cold Spring Harbour Laboratory Press, New York. 
Stewart-Jones, A., Hodges, R. J., Farman, I. D. and Hall, D. R. (2006). Solvent extraction 
of cues in the dust and frass of Prostephanus truncatus and analysis of behavioural 
mechanisms leading to arrestment of the predator Teretrius nigrescens. Physiological 
Entomology 31: 63—72. 
52 
CHAPTER THREE 
The Flight Activity of Prostephanus fruncatus (Horn) and Teretrius 
nigrescens Lewis in Kenya 
3.1 Abstract 
A baseline survey of the pest and its introduced natural enemy was carried out at five locations 
along an east-westerly transect across Kenya. The LGB was detected in all regions sampled, 
with higher flight activity in high potential maize growing regions in western Kenya. Its flight 
activity was auto correlated with the previous level and closely associated with relative 
humidity levels, temperature and vapour pressure deficit. There was a delayed response to 
meteorological variables. Teretrius nigrescens was only recovered after an intensive 
monitoring and recovery activity, in the original release region in eastern Kenya. A linear 
model explained 27 % of the flight activity observed and was well applicable to data collected 
14 years earlier. The effect of T, nigrescens in semi-arid areas seems seriously affected by 
environmental fragmentation. We suggest an ecologically and genetically aggressive effort to 
ensure the establishment of the predator in Western Kenya. 
3.2 Introduction 
The larger grain borer spread into Kenya from Tanzania and initially established in the drier 
coastal and eastern regions (Hodges et al., 1996). Containment efforts were imposed on the 
movement of maize within Kenya to prevent P. truncatus spreading to the main maize-surplus 
areas in the west of the country (Hodges et al., 1996). Although first reported in semi-arid 
maize deficit areas receiving food aid (Giles et al., 1996), the pest has spread to the high 
potential production areas like the western and Rift Valley provinces in Kenya (Anonymous, 
2003). This expansion poses the threats of both direct damage of bulk-stored grain and rapid 
S3 
dispersal from these maize-surplus regions during grain movement. At the farm-level, greater 
grain loss compromises the food and economic security situation in the region. The expanded 
range of distribution of LGB poses a new challenge to the grain industry in Kenya and the 
contribution of agriculture to the country's GDP. 
Biological control of the LGB using T. nigrescens has been only partially successful in Kenya. 
A Mexican population was released in Wundanyi and Makueni, resulting in an 80 % reduction 
in pest flight activity and predator dispersal of over 70km within the first three years of release 
(Nang'ayo, 1996; Hill et ah, 2003). However, the predator did not disperse beyond the semi-
 arid eastern areas of Kenya. Conversely, in West Africa, the release a Costa Rican population 
of the predator resulted in lower LGB flight activity and a reduction of damage trends was in 
warm humid coastal areas of Togo and Benin (Schneider et al., 2004). However, the predator 
was not efficient in the dry-hot areas. Likewise, the Costa Rican population failed to establish 
under the cool highland conditions of Guinea Conakry. 
The work of Nansen et al. (2001) and Hodges et al. (2003) demonstrated that climatic factors 
could cause strong year-to-year fluctuations in P. truncatus flight activity even in areas where 
T. nigrescens has been present for some time. Similar fluctuations were observed for the 
maize weevils Sitophilus zeamais Motschulsky, which should not be affected by T. nigrescens. 
Hodges et al. (2003) designed models of LGB flight activity, with main meteorological 
predictors, but incorporating specific modifiers coefficients under different environmental 
conditions. They suggested that their models could be used to separate the effects of T. 
nigrescens from those of climate, with differences in predicted and observed trap catches of P. 
truncatus in a locality after predator introduction providing a measure of the predator's 
impact. 
54 
This study formed a baseline survey of the prevalence and distribution LGB and T. nigrescens 
in Kenya. It is part of a pre-release study, forming a basis of the new effort to the use of 
potential geographically distinct populations of the predator in the control of the LGB in all 
outbreak regions in Kenya. The objectives of this study therefore were to monitor the flight 
activity of LGB in five regions in Kenya, to model possible ecological factors influencing 
their abundance and to establish whether T. nigrescens ever spread beyond the original release 
area in eastern Kenya. In an attempt to separate the effect of climate factors from those of the 
predator the model was validated with previous work done at the Kenya site where T. 
nigrescens was first released (Hill et ai, 2003). 
3.3 Materials and methods 
3.3.1 Sampling sites 
Twenty trapping sites were established four each at Kakamega, Kitale, Thika, Kitui and 
Mombasa, which represent different agro-ecological zones (Figure 3.1). At every sampling 
location, traps were located at least a kilometre away from each other, and placed at least 100 
m from any grain stores. Kitale and Kakamega are located in western Kenya and the Rift 
Valley and are characterized by cool humid conditions (Table 3.1). The latter are major maize 
production areas, with a tropical rainforest vegetation type. Thika is on the periphery of the 
semi-arid area, while Kitui and Kibwezi lie in the semi-arid lower mid-altitudes, within an 
early outbreak zone (Hodges et ah, 1996). Mombasa is in the coastal lowlands, with a warm 
sub-humid climate, also within the first outbreak zone of LGB. Another 3-month survey was 
carried out in 2007 in Kibwezi and Kiboko close to Makueni to confirm the recovery of T. 
nigrescens and to provide samples for a related molecular genetics study. Kibwezi is a semi-
 arid area with a bimodal rainfall distribution with long rains from March to May and short 
rains from November and January. The land use pattern consists of scattered small-holder 
farmlands maize, beans and pigeon pea as the main crops. Vegetation is of the savannah 
55 
woodland type characterised by diverse species of woody trees dominated by trees of the 
genera Commiphora (Burseraceae) and Acacia (Mimosaceae). 
3.3.2 Insect trapping 
Two types of traps were used. Pherocon III delta traps (Trece Inc., Aldair, OK, USA) were 
used during the 28 - month survey, when the objective was to monitor insect populations. 
Delta traps were composed of a folded card to create a flat-based triangular shaped chamber 
with sticky internal sides. They were baited with the synthetic LGB aggregation pheromone 
mixture trunc-call 1 and trunc-call 2 (Trece Inc.). As both T. nigrescens and P. truncatus are 
attracted to this pheromone, the same traps were used for both species. Traps were collected 
every three weeks, replaced with fresh traps and sent to' the laboratory for trap quantification 
and identification. In total, there were 38 collections during the 28-month monitoring period 
from July 2004 to September 2006. Japanese Beetle Funnel traps (Trece Inc.) were used in the 
three-month (February - May 2007) T. nigrescens survey in Kibwezi, with the aim of 
recovering any surviving predators alive. The traps were baited with a synthetic LGB 
aggregation pheromone mixture and also 50 g of sterilised maize grains and a spoonful of 
sterilised LGB frass (about 2 g) to arrest any arriving LGB and T. nigrescens. Frass and maize 
had been sterilised in an oven at 40 °C for four hours or 50 °C for two hours. Every fortnight, 
frass and maize from funnel traps were collected into labelled falcon tubes and moved to the 
laboratory for grain dissection and counts of adult LGB and T. nigrescens. This was done 
within a week of collection to avoid confounding results with any Fl-generation that could 
emerge from infested grains. The target species, P. truncatus and T. nigrescens, were 
identified under a stereomicroscope at xlO magnification and counted. 
56 
Table 3.1: The geographic and climatic conditions in the five monitoring regions in Kenya 
Region Altitude Rainfall TMax Tmin 
(m a.s.l.) (mm) (°C) (°C) 
Mombasa 0-500 1200 29.4 20.0 
Thika 500-1200 500-1000 28.6 16.4 
Kitale 1000-1500 800-2500 23.0 10.0 
Kakamega 1000-1500 1000-2000 23.3 13.4 
Kitui 200 - 400 600 32.0 22.0 
Kibwezi 100-500 600 31.0 20.0 
Tmin and TMax = average annual minimum and maximum temperatures respectively. 
57 
Kakam 
Humid 
Sub-Humid 
Semi-Humid 
Semi humid to semi arid 
Semi-Arid 
Arid 
Very arid 
Figure 3.1: Map of Kenya showing major agro-ecological zones and the locations of sampling stations monitored for LGB and T. nigrescens flight 
activity on Kenya between 2004 and 2007 (Source: USDA, 2004: http://www.fas.usda.gov/pecad/highlights/2004/12/Kenya/images) 
58 
3.3.3 Meteorological data 
Weather stations were present at all trapping sites except for Kitui and Kibwezi, for which 
data from the nearest site of Makindu was used. Monthly rainfall, minimum (Tmin) and 
maximum temperature (TMax), relative humidity (RH6, RH12) and dew point at 0600 and 
1200 hours (DP6, DP12, respectively) were obtained from the Kenya Meteorological 
Department. Relative humidity and dew point data were used to calculate monthly vapour 
pressure deficit (VPD). 
3.3.4 Data analysis 
Because LGB data was collected about every three weeks, the average monthly climatic data 
was converted to the same time scale using the interpolation function of MATLAB 
(MATLAB, 2000) to enable the assessment of the relationship between LGB trap catches and 
climatic variables. These data were then analysed by means of classical principal component 
analysis (PCA) on covariance matrix (Devillers and Karcher, 1991). Briefly, PCA replaces the 
original variables of a data set with a smaller number of uncorrelated variables called principal 
components (PCs). The method is linear in that the new variables are a linear combination of 
the original ones. The first principal component (PCI) accounts for the most part of the 
variance of the system, followed by PC2, PC3, and so on. PCA was performed on data 
converted into their loglO value. 
The study of the relationships between the LGB and the environmental variables were 
investigated by means of partial least squares (PLS) regression analysis (SAS, proc PLS). The 
climatic variables and number of LGB are centred by subtracting their means and scaled by 
dividing by their standard deviations. 
59 
Standardised x: x = — 
LGB data from each site at Mombasa, Kakamega, Kitale and Thika at time t were regressed 
using the partial least squares (PLS) model with climatic data as predictors and with or 
without LGB as auto-correlated factor. The order of the LGB autocorrelation model was 
determined by an autoregressive moving average (ARMA) models (proc ARTMA, SAS). The 
PLS procedure was used to fit models when the predictors were highly correlated, which is 
possible for climatic variables, with LGB as an independent factor. It sought to determine the 
minimum number of uncorrelated factors that significantly explain both LGB flight and 
climatic variations. The model was validated to the LGB data from a survey done at Kibwezi 
between 1991 andl997 (Hill et a/., 2003). 
3.4 Results 
3.4.1 Seasonal fluctuation in numbers of LGB and T. nigrescens 
The highest mean trap catches were recorded in Kakamega followed by Kitale. They were 
considerably lower and did not vary significantly in the other three locations (Figure 3.2). 
Average trap catches varied from 30 per trap in Mombasa to 550 in Kakamega, with a peak of 
over 2000 in July, 2005. LGB flight activity was bimodal in Thika and Kitui, with a major 
peak in November-January and a minor peak around March - May, just following the short 
rain season and at the beginning of long rain season. 
No T. nigrescens was trapped during the 28-month monitoring period in these five locations. 
During the intense survey in Kibwezi, only three adults of T. nigrescens were caught in the 
Kiboko range and none at the original release site in Kibwezi area. Four adults were caught in 
a single sentinel 2-week trapping period at one site in Mombasa. 
60 
Projection of the LGB flight into its principle components did not lead to a separation of the 
different sites and the models were therefore applied to all four regions. The first two principal 
components 1 and 2 contributed 58.02 % and 25.60 % to the total variance, the third, to the 
seventh components contributing 16.39 % (Table 3.2). Dew point, minimum temperature and 
rainfall contributed positively to both principle components. 
Table 3.2: Eigenvalues and weights for the first two principal components computed from 
logio transformed meteorological data. 
Weight 
Variable 
Proportion of covariance 
Eigenvalues 
Maximum Temperature 
Minimum temperature 
Relative humidity at 0600 hrs 
Relative humidity at 1200 hrs 
Dew point 0600 hrs 
Dew point 1200 hrs 
Rainfall 
PCI PC2 
0.5802 0.2560 
0.0412 0.0219 
-0.371 0.441 
0.049 0.669 
0.431 -0.259 
0.456 -0.157 
0.403 0.371 
0.382 0.350 
0.397 0.077 
Table 3.3: Percent variation accounted for by partial least squares factors 
Model effects Dependent variables 
Factors1 Current Total Current Total 
1 51.675 51.675 26.152 26.152 
2 24.706 76.381 8.037 34.189 
3 9.097 85.478 4.021 38.211 
4 3.058 88.536 1.091 39.302 
5 3.959 92.495 0.226 39.528 
Number of variables included into the variable building 
61 
3000 
2500 
Q- 2000 
c 1500 
CD 
(3 1000 -
 500 -
 8/04 
Kakamega 
Kitale 
Kitui 
Mombasa 
Thika 
12/04 4/05 8/05 12/05 
Date(Month/Year) 
4/06 8/06 
Figure 3.2: The flight activity of Prostephanus truncatus in five regions in Kenya (2004 
2007). 
Monitoring period between 1992 and 1997 
Figure 3.3: The temporal dynamics of standardized value of observed number of Larger Grain 
Borer (solid line), predicted LGB with LGB as auto-correlated factor (dash line), and 
predicted LGB without LGB as auto-correlated factor (star) in Makueni between 1991-1997. 
X-axis represents sampling times with the unit of time (month), and y-axis is the standardized 
value of LGB by the deviation with its mean value divided by the standard deviation. Arrow 
indicates time of first Teretrius nigrescens release in Makueni in 1992. 
a- (c> 
J.I 
<S 
IS " > 1 1.5 
A 1 rM Jte {#*% 
'g US £tf *\ \/fc e 
O -0-5 
i / [ ? I I ^ F Py*\ 1 '" ' '!> i ^ 
IS :25 JjT 3ft 3S 
(=3 v*> f *%% -1 - X 
-1.5 ~ Time (3-week) Time (3-weefc) 
Figure 3.4: The temporal dynamics of standardized value of observed number of larger grain borer (solid line), predicted LGB with LGB as auto-
 correlated factor (dash line), and predicted LGB without LGB as auto-correlated factor (star) at 4 sites: (a) Kakamega; (b) Kitale; (c) Mombasa; (d) 
Thika. X-axis represents 38 sampling times between 2004-2006 with the unit of time (3-week), and y-axis is the standardized value of LGB, which 
is the deviation with its mean value divided by the standard deviation. 
64 
Table 3.4: Parameter estimates for seven variables used to predict the flight activity of the 
LGB in four zones in Kenya. 
Source/Parameter Type I S S Type inss t value P > | t | 
F P F p 
Maximum Temperature 130.33 0.000 10.16 0.002 -3.19 0.002 
Minimum temperature 0.26 0.608 0.12 0.730 0.35 0.730 
Relative humidity at 0600 hrs 23.27 0.000 28.13 0.000 5.30 0.000 
Relative humidity at 1200 hrs 19.82 0.000 11.74 0.001 -3.43 0.001 
Dew point 0600 hrs 0.22 0.638 0.02 0.886 0.14 0.886 
Dew point 1200 hrs 0.15 0.700 0.16 0.693 -0.40 0.693 
Vapour Pressure Deficit 0.47 0.495 0.47 0.495 -0.68 0.495 
Table 3.5: Parameter estimates for data percentage contribution to model estimates. 
Centred Scaled 
Intercept 0.000 -0.014 
LGB(0 0.383 0.377 
Maximum Temperature -0.231 -0.224 
Minimum temperature 0.126 0.123 
Relative humidity at 0600 hrs 0.386 0.374 
Relative humidity at 1200 hrs -0.256 -0.247 
Dew point 0600 hrs 0.012 0.012 
Dew point 1200 hrs -0.083 -0.080 
Vapour Pressure Deficit -0.034 -0.591 
Rainfall -0.051 -0.050 
65 
3.4.2 The LGB flight model 
1) The model without LGB number as an auto-correlated variable 
From the CV test of cross-correlation analysis of the PLS model, the minimum number of 6 
variables, contributed 96 % of the variation in the predictors but only explained 22 % of the 
variation in responses (R2 = 0.281, F = 24.93, p < 0.0001) (Table 3.4). The following model 
was thus derived: 
LGB(t)=-0.97*Tavg+0.65*TMax+0.43*Tmin+0.37*RH6-0.32*RH12-0.32*VPD 
2) With LGB catches as an auto-correlated variable 
Time series analysis indicated that LGB dynamics at all sites followed an AR (1) model, in 
which LGB at time t is dependent on time at t - 1. Therefore, LGB sit-I was considered as 
one of the factors. From the coefficient of variation test of the cross-correlation analysis of the 
PLS model, the minimum number of three variables significantly contributed 87 % of the 
variation in the predictors and 37 % of the variation in responses (Table 3.5). To determine 
which factors to eliminate from the analysis, the regression coefficients of each factor in the 
PLS model, which determines the importance of each factor in the prediction of the response, 
were compared. If a predictor had a relatively small coefficient (in absolute value), then it was 
considered a prime candidate for deletion. 
The results from Kakamega, Kitale, Mombasa and Thika sites indicated that LGB at time t 
was affected by LGB at the previous sampling time (LGB[t-l]), RH6 and TMax. 
LGB(0 = -0.023+0.486*LGB(M)+0.122*RH6-0.049*TMax 
Figure 3.4 shows observed and predicted LGB trap catches. There was a one unit time delay 
due to the time series model. 
66 
In Kibwezi, LGB trap catches during the February - June 2007 period were low (< 300 beetles 
per fortnight) with most catches below 50 beetles per fortnight. However, traps located near 
markets or established grains stores yielded higher LGB numbers than those located further 
inland. Still, some LGB was detected in the Kiboko Range, 10km away from the nearest 
homestead of a maize farmer. 
3.5 Discussion 
In Africa, the spread of P. truncatus could be related to movements of contaminated maize 
consignments, usually from maize surplus to maize deficit areas (Bosque-Perez et al., 1991; 
Adda et al, 1996; Schneider et al., 2004). Thus, as also reported for West Africa, in Kenya 
trap catches were higher close to roads and markets than in the interior where homesteads 
were sparse. The dispersal of LGB beyond the maize deficit areas of eastern Kenya to the high 
production areas in the west took more than ten years after it was first discovered in the 
country. It is most likely that the infestations in western Kenya stemmed from maize 
importations from Tanzania during lean years into deficit areas close to Kitale and Kakamega 
(Anonymous, 2003) 
Though LGB reached western Kenya in recent years only, trap catches were considerably 
higher in these high production areas than in the maize deficit areas of eastern Kenya, where 
the pest was introduced in the country in the early 1990s (Hodges et al., 1996; Giles et al., 
1996). They were comparable to those in maize surplus areas of Ghana and Benin prior to the 
introduction of T. nigrescens (Schneider et al., 2004). This appears to be indicative of an 
introduction of the pest into the region without T. nigrescens or the inability of the predator to 
establish as a result of climatic limitations. However, according to the regression model of 
Tigar et al. (1994), who used average relative humidity, annual rainfall and annual mean 
temperature as independent variables, maximum catches in pheromone traps are expected at 
67 
temperatures of 23 - 25 °C and relative humidity of 50 - 52 %. This would suggest high trap 
catches in the cool areas of western East Africa corroborating the results of the present study. 
In Kitui, Kiboko and Mombasa, LGB trap catches were considerably lower than in western 
Kenya. Kitui and Kiboko lie in a maize deficit hot-dry zone with sparse shrub land but with 
several alternate tree hosts of LGB (Nang'ayo, 1996). The sustained trap catches of below 200 
adults per trap per month were significantly lower than those observed in the same region 
before the release of T. nigrescens and comparable to the flight activity recorded after the 
establishment of the predator in the area (Giles et al., 1996; Hill et al., 2003). However, they 
were in the range of those in the Guinea and Sudan Savannas in West Africa, where. T. 
nigrescens appeared not to exert control over LGB (Schneider et al, 2004). Not a single adult 
T. nigrescens was captured during the two year survey in the mid-altitudes, and just three 
specimens in the more intensive survey in the Kiboko range, where no single trap captured the 
predator more than once during three-month trapping period using 51 traps. In the coastal area 
around Mombasa, trap catches were even lower than in the semi-arid mid-altitudes. However, 
the traps were located in the south coast where both precipitation and maize production is low 
compared to the hot-humid coastal area of West Africa, where T. nigrescens appeared to 
control P. truncatus (Schneider et al., 2004). In the coastal area only four T. nigrescens 
specimens were captured. Hill et al. (2003) recovered an average of one T. nigrescens adult in 
every 20th trap in early 1995 and none during a three-year monitoring period thereafter. By 
contrast in Benin, Borgemeister et al. (1997a; 2001) recorded 100 - 400 T. nigrescens per trap 
per month, while Schneider et al. (2004) observed 20 - 110 adults per trap in the forest 
savannah region, and about two insects per trap in the drier northern Guinea savannah. 
Though the presence of predator in eastern Kenya confirms its establishment and spread over 
15 years since its release, the very low numbers recovered suggest that the predator is not 
responsible for the low LGB trap catches. 
68 
The introduction of a natural enemy results in an initial increase in the population of the 
natural enemy and a concomitant decrease in the number of the pest species, followed by a 
reduction in both species as they, ideally, level off into stable equilibrium (Hoddle, 2001). In 
Kenya, although this trend was initially observed during the first few years after releasing T. 
nigrescens in Kibwezi (Hill et al, 2003), the predator is now exceedingly rare, and in some 
areas LGB catches have gone back to almost the levels before the release, e.g. high catches of 
up to 200 individuals per trap were observed close to the Kiboko Range station during the 
second intensive survey. The question arises if the original decrease in pest densities were due 
to T. nigrescens and why the predator so scarce in the area. 
According to Zabel and Tscharntke (1998), the effect of habitat fragmentation depends on the 
food specialization of the species, whereby monophagous herbivores have a higher probability 
of being absent from small patches than polyphagous species. In addition, species richness of 
herbivores was shown to be positively correlated with habitat area while species richness of 
predators appeared to be negatively correlated with habitat isolation. They argued that due to 
instability of higher trophic level populations and limitations to disperse, predators were more 
affected by habitat isolation than herbivores. Nansen et al. (2002) and Hill et al. (2003) 
showed that the flight activity of P. truncatus was consistently higher near maize stores than 
inside forests, underlining the superior quality or attractiveness of maize compared to woody 
substrates, which was demonstrated by Hill et al. (2003). The original release area consists of 
scattered smallholder farmland, where crop fields are surrounded by savannah woodland, thus 
maize fields and stores are relatively scarce. Furthermore, because of insufficient rain, grain 
deficits are common in the area because maize harvests fail in 50% of the seasons 
(Anonymous, 1983). Thus, the relative scarcity of stored grain, when compared to the high-
 production zones in western Kenya, the low maize production potential and the frequent crop 
failure should lead to a highly fragmented habitat in time and space. In addition, LGB is an 
69 
oligophagous species mainly feeding on stored products that allow for much higher population 
growth rates than woody substrates (Hill et al., 2003), whereas T. nigrescens is highly 
monophagous (Rees et al, 1990; Scholz et al, 1998). Thus according to Pimm (1991), 
Lawton (1995), Holt (1996) and Zabel and Tscharntke (1998), the predator should be more 
susceptible and go extinct more readily than its phytophagous prey if habitat fragmentation 
increases. It is hypothesized therefore that in eastern Kenya, due to habitat isolation and lack 
of habitat connectivity as a result of the low number of release sites, T. nigrescens went 
locally extinct. It would also follow that in the semi-arid and humid coastal areas of Kenya, 
the role of wild, woody host plants in maintaining LGB and predator populations as suggested 
by Nang'ayo (1996), Borgemeister et al. (1998a,b) and Nansen et al. (2002) is limited. It is 
recommended to release it again, but with a much higher number of release sites and released 
populations to increase the interconnectedness between the different populations occupying 
cultivated and wild habitats. It is also suggested that the chances of permanent establishment 
of T. nigrescens are higher in the high production, maize surplus areas in western Kenya than 
the maize deficit areas in the eastern semi-dry mid-altitudes given the predator strain is 
adapted to the cooler climates. 
The effect of weather parameters on the LGB abundance and flight activity was not clear cut. 
Auto-regression of LGB appeared to be more important than weather factors in explaining 
seasonal variation in LGB trap catches. Another reason responsible for the lower significance 
of the weather variables is that at each site, the same meteorological parameters were applied 
to four different traps, which could not reflect the real micro-changes in climate. This study 
agrees with Hodges et al. (2003) who observed that flight activity is affected by weather 
parameters, mostly in their effects on the available population of LGB adults and their 
Influence on triggering flight. Earlier studies had Identified temperature, rainfall and humidity 
as the major abiotic determinants of LGB flight activity in Central America, West and East 
70 
Africa (Tigar et al, 1994; Nang'ayo, 1996; Nansen et al, 2001; Hodges et al, 2003). 
However, other studies suggest that sharp rises in flight activity at onset of rains is caused by 
the response to environmental cues rather than rapid reproduction leading to overcrowding 
and substrate degradation that may also trigger dispersal (Fadamiro and Wyatt, 1995; Hill et 
al, 2002). Borgemeister et al. (1997b), on the other hand, speculated that peak trap catches 
were related to storage patterns, coinciding with cleaning of grain stores for restocking. By 
contrast, Nang'ayo (1996) observed a significant correlation between the onset of flight with 
RH and temperature and found that flight peaks were not associated with maize storage. 
It is not clear whether the decrease in LGB trap catches observed in the 90s in the release area 
were due to the predator or due to a combination of climatic factors which also affected yields 
and therefore amount of stored maize. In Ghana, Hodges et al. (2003) could attribute seasonal 
and yearly fluctuations of LGB solely to climatic factors and there was no clear-cut decrease 
in trap catches after the release of the predator. Similar observations were made by Nansen et 
al. (2001) in southern and central Benin using climatic models contradicting results by 
Borgemeister et al. (1997b) and Schneider et al. (2004), who had however used a much higher 
number of pheromone traps over a wider area. Nansen et al. (2001) found that the same 
climatic parameters determined flight in various locations in southern Benin, whereas for 
central Benin new coefficients for the same environmental variables were needed to produce 
adequate predictions, while for northern Benin the model could not predict flight behaviour of 
LGB. In addition, different driving variable were used by Hodges et al. (2003) and Nansen et 
al. (2001), which indicate that other environmental variable not measured in the two studies 
are required to give reliable predictions of LGB. 
Due to the very low number of the predators caught and the erratic nature of their recovery, it 
is not possible to directly evaluate their effect in this survey. Such paucity of the predator 
71 
punctuated with long periods of no predator capture in flight traps had been observed by Hill 
et al. (2003), even when LGB flight activity remained low. Hill et al. (2003) recovered an 
average of a single T. nigrescens adult in every 20 trap in early 1995 and none during a five-
 year survey thereafter. In Benin, Borgemeister et al. (1997a; 2001) recorded a predator prey 
ratio of approximately 0.025 in the drier North to 0.15 in the maize surplus hot-humid regions 
in the South. Schneider et al. (2004) observed a lower sustained activity of about 20 adults per 
trap per month in the forest savannah region and about two insects per trap in the drier 
northern Guinea savannah. 
Flight traps sample individuals in the dispersal phase and may not be a reflection of the actual 
population sizes (Hodges, 2002). The use of prey pheromones may also not be optimum for 
the predator species, whose attraction depends on the nutritional and physiological state too 
(Larsson et al., 2009). However trap catches over a long period may still reflect the relative 
abundance of the pest (Hodges, 2002; Schneider, 2004). The predator-prey ratio in the drier 
parts of eastern and the coastal regions of Kenya were lower than those observed in the dry 
regions of northern Benin. In northern Benin, peak ratios of 0.01 were observed compared to 
about 0.03 in Kenya in 1995 (Nang'ayo, 1996; Nansen et al., 2001; Hill et al., 2003; Schneider 
et al., 2004). In Mombasa, Kenya, predator-prey ratio was as high as 0.02 per trap during the 
brief survey that took just a month. In Mexico, Tigar et al. (1994) observed ratios of 0.01 in 
the maize surplus area of La Laguna where P. truncatus was abundant. However, Rees et al. 
(1990) observed much higher predator-prey ratios of between 0.24 and 0.48 with higher ratios 
in maize production areas of Mexico using fewer traps and for a one-month survey. These 
sharp variations in the predator-prey ratios in Central America and Africa are interesting. 
The establishment and success of the predator in Kenya is lower than that in West Africa. 
Differences between the two release approaches exist. In West Africa, while about 130 000 
72 
were released over a six month period in Togo alone, Kenyan releases were about 4 900 
insects ( Giles et al., 1996). This single small release may have compromised the ability of the 
predator to find and establish in suitable habitats in suitable numbers. The strain released in 
West Africa was also sourced from the same latitudinal range as the release point, a factor in 
eventual success of the natural enemy (Phillips et al., 2008). This study concludes that 
although T. nigrescens may be playing a role in controlling LGB populations, it is unlikely to 
be a key source of LGB population regulation in Kenya at the moment. A more genetically 
and temporally aggressive release especially in western Kenya holds promise for the success 
of the biological control of the LGB. The rest of this thesis examines the ecological and 
genetic parameters of geographical populations of T. nigrescens that might be candidates to 
these new efforts. 
73 
3.6 References 
Adda, C , Borgemeister, C , Meikle, W. G., Markham, R. H., Olaleye, I., Abdou, K. S. 
and Zakari, M. O. (1996). First record of the larger grain borer, Prostephanus 
truncatus (horn) (Coleoptera: Bostrichidae), in the republic of Niger. Bulletin of 
Entomological Research 86: 83 - 85. 
Anonymous (1983). Farm Management Handbook of Kenya. Kenya Ministry of Agriculture 
and GTZ. 450pp 
Anonymous (2003). The status of the larger grain borer in Kenya. Report of the Larger Grain 
Borer Task Force, July, 2003. 
Borgemeister, C , Djossou F., Adda, C , Schneider, H., Djomamou, B., Degbey, P., -
 Azoma, K. and Markham, R. H. (1997a). Establishment, spread and impact of 
Teretriosoma nigrescens (Coleoptera; Histeridae), and exotic predator of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in south-western Benin. 
Environmental Entomology 26: 1405 - 1415. 
Borgemeister, C , Meikle, W.G., Scholz, D., Adda, C , Degbey, P., Markham, R.H., 
(1997b). Seasonal and meteorological factors influencing the annual flight cycle of 
Prostephanus truncatus (Coleoptera: Bostrichidae) and its predator Teretriosoma 
nigrescens (Coleoptera: Histeridae) in Benin. Bulletin Entomological Research 87: 
239-246. 
Borgemeister, C , Goergen, G., Tchabi, A., Awande, S.,Markham, R.H. and Scholz, D. 
(1998a). Exploitation of a woody host plant and Cerambycid-associated volatiles as 
host-finding cues by the larger grain borer (Col.: Bostrichidae). Annals of the 
Entomological Society of America 91: 741 — 747. 
Borgemeister, C , Tchabi, A. and Scholz, D. (1998b). Trees or stores? The origin of 
migrating Prostephanus truncatus collected in different ecological habitats in 
Southern Benin. Entomologia Experimentalis et Applicata 87: 285 - 294. 
74 
Borgemeister, C , Schneider, H., Affognon, H., Schulthess, F., Bell, A., Zweigert, M. E., 
Poehling, H. M. and Setamou, M. (2001). Impact Assessment of Teretrius nigrescens 
Lewis (Col.: Histeridae) in West Africa, A predator of the larger grain borer, 
Prostephanus truncatus (Horn) (Col.: Bostrichidae). In: 1st Symnposium on Biological 
Control of Arthropods, 17 t h - 21 s t September 2001Honolulu, Hawaii. 
Bosque-Perez, N. A., Markham, R. H. and Fajemisin, J. M. (1991). Occurence of the larger 
grain borer Prostephanus truncatus in Burkina Faso. FAO Plant Protection Bulletin 
39: 182-183 . 
Devillers, J. and W. Karcher, W. (eds) (1991). Applied Multivariate Analysis in SAR and 
Environmental Studies. Kluwer Academic Publishers. 
Fadamiro, H. Y. and T. D. Wyatt. (1995). Flight initiation of Prostephanus truncatus in 
relation to time of day, temperature, relative humidity and starvation. Entomologia 
Experimentalis et Applicata 75: 273-277. 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). Release 
and establishment of the predator Teretriosoma nigrescens Lewis for the biological 
control of Prostephanus truncatus (Horn) in Kenya. African Crop Science Journal 4: 
325-337. 
Hill, M. G., Borgemeister, C. and Nansen, C. (2002). Ecological studies on the larger grain 
borer, Prostephanus truncatus (Horn) (Col.: Bostrichidae) and their implications for 
integrated pest management. Integrated Pest Management Reviews 7: 201 - 221. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
predatory beetle Teretrius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 — 306. 
75 
Hoddle, M. S. (2001). Classical biological control of arthropods in the 21st century In: 1st 
Symposium on Biological Control of Arthropods, 17th - 21 s t September 2001, 
Honolulu, Hawaii. 
Hodges, R. J., Addo, S. and Birkinshaw, L. (2003). Can observation of climatic variables be 
used to predict the flight dispersal rates of Prostephanus truncatus! Agricultural and 
Forest Entomology 5: 123-135. 
Hodges, R. J., Farrell, G. and Golob, P. (1996). Review of the larger grain borer outbreak in 
East Africa - rate of spread and pest status. In G. Farrell, A.H. Greathead, M. G. Hill 
and G.N Kibata (eds) Management of Farm Storage Pests in East and Central Africa. 
Proceedings of East and Central Africa Storage Workshop, 14 - 19, April 1996, 
Naivasha, Kenya. Ascot: International Institute of Biological Control. 
Holt, R. D. (1996). Food webs in space: an island biogeographical perspective. In: Polis GA, 
Winemiller KO (eds) Food Webs. Chapman & Hall, New York, pp 313 - 323. 
Larsson, M. C. and Svensson, G. P. (2009). Pheromone monitoring of rare and threatened 
insects: exploiting a pheromone-kairomone system to estimate prey and predator 
Abundance. Conservation Biology. DOI: 10.1111/j.l523-1739.2009.01263.x 
Lawton, J. H. (1995). Population dynamic principles. In: Lawton JH, May RM (eds) 
Extinction Rates. Oxford University Press, Oxford, pp 147-163 . 
Nang'ayo, F. L. O. (1996). Ecological Studies on The Larger Grain Borer in Savanna 
Woodlands of Kenya. PhD Dissertation, University of London, 179pp. 
Nansen, C , Meikle, W. G. and Korie, S. (2002). Spatial analysis of Prostephanus truncatus 
(Horn) (Coleoptera: Bostrichidae) flight activity near maize stores and different forest 
types in southern Benin, West Africa. Annals of the Entomological Society of America 
95: 66 - 1A. 
76 
Nansen, C , Korie, S., Meikle, W. G. and Hoist, N. (2001). Sensitivity of Prostephanus 
truncatus (Coleoptera: Bostrichidae) flight activity to environmental variables in 
Benin, West Africa. Environmental Entomology 30: 1135-1143 
Phillips, C. B., Baird, D. B., Iline, 1.1., McNeill, M. R., Proffitt, J. R. Goldson, S. L. and J. 
M. Kean, J. M. (2008). East meets west: adaptive evolution of an insect introduced for 
biological control. Journal of Applied Ecology 45: 948 — 956. 
Rees, D. P., Rivera, R. R. and Rodriguez, F. S. H. (1990). Observations on the ecology of 
Teretriosoma nigrescens (Lewis) (Col., Histeridae) and its prey Prostephanus 
truncatus (Horn) (Col., Bostrichidae) in the Yucatan peninsula, Mexico. Tropical 
Science 30: 153-165 . 
Schneider, H. (1999). Impact assessment of Teretriosoma nigrescens Lewis (Coleoptera 
Histeridae), a predator of the larger grain borer Prostephanus truncatus (Horn) 
(Coleoptera: Bostrichidae). PhD Thesis, University of Hannover 148pp. 
Schneider, H., Borgemeister, C. Setamou, M., Affognon, H., Bell, A., Zweigert, M. E., 
Poehling, H. and Schulthess, F. (2004). Biological control of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) by its predator Teretrius 
nigrescens (Lewis) (Coleoptera: Histeridae) in Togo and Benin. Biological Control 30: 
241 -255 . 
Scholz, D., Borgemeister, C. and Poehling, M. (1998). EAG responses of Prostephanus 
truncatus and its predator Teretriosoma nigrescens to the borer-produced aggregation 
pheromone. Physiological Entomology'23: 265 - 2 7 3 . 
Tigar, B. J., Osborne, P. E., Key, G. E., Flores-S., M. E. and Vazquez-Arista, M. (1994). 
Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) by 
its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bulletin of 
Entomological Research 84: 555 — 565. 
77 
Zabel, J. and Tscharntke, T. (1998). Does fragmentation of Urtica habitats affect 
phytophagous and predatory insects differentially? Oecologia 116: 4 1 9 - 4 2 5 . 
78 
C H A P T E R F O U R 
Effect of temperature and humidity on the predation of Teretrius nigrescens 
on Prostephanus truncatus 
4.1 Abstract 
The variable efficacy of two geographic populations of Teretrius nigrescens has led to the 
hypothesis that ecological strains of the predator exist. Thus sustainable biological control of 
the larger grain borer in Africa could benefit from the exploitation of this strain diversity. To 
test this hypothesis, five putative strains of the predator were tested for their ability to 
multiply, control LGB, and to prevent maize kernel damage at 18, 21, 24, 27, 30, 33 and 36 °C 
and under low (40 - 55 %) and high (70 - 85 %) relative humidity. Both the LGB and T. 
nigrescens performed better under humid than dry conditions. Between 21 and 33 °C, T. 
nigrescens reduced LGB population build up by about 80 %. Although the predator could not 
exert as much control under extreme conditions, adult survival was high at both the lower and 
upper extremes 
4.2 Introduction 
Teretrius nigrescens was introduced into Africa, after studies showed that it was an effective 
predator closely coadpted with its prey's biology and habitat. T. nigrescens locates its prey by 
cuing in on its aggregation pheromones (Rees et at, 1985,1990; Boye, 1988; Boye et at, 
1988). This enhances natural dispersal and efficacy of the predator in the wild (Nang'ayo, 
1996; Schneider et at, 2004). However, its establishment has been population and climate-
 dependent. Although a Costa Rica population established well and efficiently contributed to 
the control of LGB in the warm humid regions of southern Benin and Togo, it was not 
efficient in the dry, hot savannas (Borgemeister et al., 2001; Schneider et at, 2004). In Kenya, 
a Mexican population established in the warm semi arid regions of eastern Kenya, but flight 
79 
activity has remained very low (Hill et al., 2003). Both populations did not establish in the 
cooler highland regions of Guinea Conakry and Kenya (Giles et al., 1996; Anonymous, 1999). 
Both LGB and T. nigrescens have been recovered from both sub-humid and semi-arid zones 
across Central America (Rees et ah, 1990; Tigar et al., 1994). It is therefore possible that the 
failure of the predator to establish in parts of Kenya could have been a result of suitability of 
local ecological conditions around the release sites for the strains of the predator released. The 
larger grain borer has now spread over most ecological environments in sub-Sahara Africa 
between southern Senegal and the Venda region of South Africa. Sustainable development of 
biological control components, therefore, involves predator populations adapted to the diverse 
climatic conditions in Africa. Four new populations of the predator and two previously 
released ones are currently being considered for release in eastern Africa. Based on the 
preliminary observations of genetic diversity and ecological preference of T. nigrescens, the 
suitability of each population for specific regions should be determined before release. 
This study investigates the effects of fourteen constant temperature and humidity 
combinations on the predation of the LGB by five putative strains of the T. nigrescens on 
shelled maize. The response of T. nigrescens is measured in terms of population increase, 
while predatory performance are measured in terms of control of LGB population growth, 
LGB mortality, grain damage and grain weight loss. 
4.3 Materials and methods 
4.3.1 Insects 
The colony of P. truncatus used in this experiment originated from a laboratory culture 
originally recovered from Kitale (Kenya) in 2004 and maintained at room temperature on 
maize kernels in the Animal Rearing and Quarantine Unit (ARQU), ICIPE (Chapter 2). All 
80 
maize used in the experiments was purchased locally. Five putative strains of T. nigrescens 
were used in this study. Three colonies recovered from Mexico by CIMMYT - Mexico from 
Batan, Tlaltizaspan and Oaxaca. The Kenyan population was obtained from the Kenya 
Agricultural Research Institute (KARI) Kenya Kiboko Field Station. This was the remnant of 
a colony originating from around Nuevo Leone, Mexico, initially released for the control of 
the LGB in Kenya in 1992 (Nang'ayo, 1996; P. Likhayo, KARI, Kenya, R. J. Hodges, NRI, 
UK, personal communication). The Benin population was donated by the Centre for 
Biological Control, International Institute of Tropical Agriculture (IITA) in Benin. This 
population had previously been released in Benin and Togo in 1993. The colony was 
maintained in the laboratory on LGB on shelled maize with regular introgression with field 
collected samples, the last being in 1996 (C. Atcha, WARD A, Ivory Coast, personal 
Communication). In our laboratories T. nigrescens were reared on maize grains and LGB 
collected from Kitale. All insects were reared under similar conditions for at least six months 
before use in experiments to minimise the effects of past rearing conditions and hence have a 
common starting point. 
4.3.2 Experimental set up 
The effect of temperature and humidity conditions on the ability of T. nigrescens to suppress 
LGB populations was studied under eight temperature levels i.e. 15, 18, 21, 24 27, 30, 33 and 
36 °C and two relative humidity regimes (low: 45 — 55 and high: 75 - 85 %) in environmental 
chambers (Sanyo, Japan). One hundred unsexed P. truncatus adults were introduced into 250 
ml glass jars containing 100 g of hand-selected, undamaged maize kernels, sterilised at 50 °C 
for four hours in an oven (Jembere et al., 1995). The number of grains per jar, recorded for 
eventual damage determination, averaged 228.4 ±13.2. To each jar, 100 unsexed LGB adults 
were added and left at room temperature for a week to allow for oviposition. From each strain, 
twenty unsexed, adult T. nigrescens, less than 3 months old, were then added to each jar. The 
81 
jars were closed and kept in an incubator at the respective constant temperature and humidity 
conditions for 70 days. Two controls, one with P. truncatus alone and the other with 
uninfested maize were also set up. Each treatment was replicated six times. 
To assess grain damage, the substrate mixture was sieved with a 3.5 mm and 0.4 mm mesh 
sieve (Helbig, 1999). Particles passing through the 3.5 mm mesh sieve were classified as loss 
and those passing through the 0.4 mm mesh as insect frass. The fraction retained by 3.5 mm 
mesh was sorted by hand into damaged and undamaged grain and counted. The number, 
weight and moisture content of damaged and undamaged grains and weight and moisture 
content of LGB frass were determined. Damaged grains were dissected and the insects 
counted. Insect counts were restricted to adults, to limit the data to the first generation 
individuals and to avoid error from possible decomposition of dead larvae due to their soft 
body tissues. The moisture content of the grains in the uninfested control was used to account 
for grain moisture content m. determining grain damage. 
4.3.3 Damage and loss assessment 
In this study, the terms 'grain damage' and 'grain loss' were used sensu Boxall (2002). 
Damage refers to the proportion of the grain found to bear signs of LGB attack while loss was 
restricted to a decline in grain weight. As LGB caused near total grain destruction, grain 
damage was determined by posterior subtraction of uninfested grain in each set up at the end 
of the experiment from the initial number of grains, i.e. 
Grain damage % = 100 x [(a-b)ld\ 
where a was the initial number of grain and b the number of undamaged grain at the end of the 
experiment. 
82 
Weight loss was determined as 
Weight Loss % = 100 x {[Wa-(Wb+Wc)]IWa } 
where Wa was the initial grain weight grain, Wh the weight of undamaged grains and Wc the 
weight of damaged grain at the end of the experiment. 
LGB and T. nigrescens population build-up was determined as the final number of individuals 
counted (adults only for LGB) minus the initial number of individuals added at set-up of the 
experiment. Insect mortality was determined as the number of dead adults of the respective 
species counted as a fraction of the total number of adult insects recovered at the end of the 
experiment. Juvenile stages were excluded to avoid error due to the possible decomposition of 
their soft body tissues. 
The index of grain loss prevention by 71 nigrescens (Figures 4.4 and 4.5) was determined at 
each temperature level as: 
Grain loss prevention index = 100 x (1- (Dc - Dj)/Dc)) 
Where D,- is the mean loss for each strain and Dc mean loss for untreated control. 
4.3.4 Data Analysis 
Two-way analysis of variance (ANOVA) was performed in SAS (SAS Institute, 2000) using 
the general linear model (proc glm) with temperature, humidity and 7. nigrescens strains as 
the main effects. Data on insect counts (LGB and 71 nigrescens) were logio(x) and logio(x + 1) 
transformed before ANOVA procedure in SAS. Damage level data (as a proportion of the 
original grain numbers and weight) were arcsine square root transformed before ANOVA. 
83 
When ANOVA showed significant differences, mean separation was done using the Student-
 Newman-Keul (SNK) test (in SAS). Untransformed data are presented. 
4.4 Results 
Teretrius nigrescens survived the ten weeks of the experiments under all temperature and 
humidity regimes but did not reproduce at 15 °C, thus this temperature was removed from 
further studies. At 18 °C, under both humidity regimes, both T. nigrescens and P. truncatus 
did not develop to maturity within the 10 weeks though juvenile stages of both species were 
observed. Teretrius nigrescens population growth was highest at 30 °C for all populations 
tested (Figure 4.1 and 4.2) apart from the Benin population that performed best at 27 °C. 
The effect of T. nigrescens on LGB population growth is summarised in figures 4.3 and 4.4. 
The LGB populations increased under all temperature conditions. The effect of T. nigrescens 
in reducing LGB population growth was over 80 % between 21 - 33 °C. Above 33 °C the 
effect of T. nigrescens appeared higher than at the other temperature levels. This is however a 
possible combination of the effect of this temperature on both species and the exaggerated 
effect of minimal predation by surviving T. nigrescens adults. 
Slight differences between the T. nigrescens strains' ability to control the population build up 
and confer grain protection against the LGB were observed (Figures 4.5 and 4.6). The KARI 
population appeared less efficacious than the other populations at low temperature levels, but 
similar under high temperature conditions. For instance, grain weight loss was significantly 
higher with the KARI population at 21 °C (F = 20.9, p < 0.0001) and 24 °C (F = 38.8, p < 
0.0001) while it was not significantly different from the other populations at higher 
temperature levels (Table 4.1). Similar trends were observed for survivorship and LGB 
84 
population increase. The other populations did not differ in their ability to control grain 
damage. 
Grain damage trends were generally similar to the weight loss trends but they were more 
consistent (Table 4.1). The effect of humidity was generally more discernable above 24 °C 
where higher humidity resulted in higher grain damage. At low temperature levels, the effect 
of humidity was minimal. There were no discernable trends in mortality levels of T. 
nigrescens and P. truncatus between treatments. 
85 
E 
0) 
w 
+1 c re o 
E. 
c o 
3 a. o 
Q. 
10 
c 
0) o 10 
£ 
a> 
TO c re .c O 
25 
20 
15 
10 
5 -
 -5 
? Batan 
-Benin 
? KARI 
-Oaxaca 
-Tlaltiza 
21 24 27 
Temperature °C 
30 33 36 
Figure 4.1: The effect of temperature on the population increase of Teretrius nigrescens under 
40 - 55 % relative humidity regime. 
60 
3 Q. O Q. 
10 
c 
<D 
O 
<0 
£ 
.O) 
c 
a 
D) c re si O 
50 
E 
Q) 
CO 
+1 c re o> 
£ 40 c o 
30 
20 
10 
-10 
Temperature °C 
Figure 4.2: The effect of temperature on the population increase of Teretrius nigrescens under 
40 - 55 % relative humidity regime. 
86 
10 
£ 
to 
+i 
c re 
i H 
o 
p 
c o 
^ 2 H 
re 
Q. O Q. 
m O 
-2 J 
i fk 
i 
r^i 
33 
Temperature °C 
D Control 
K B atari 
? Benin 
BKARi 
n Oaxaca 
0 Tlaltizaspan 
- r J = L 
Figure 4.3: Effect of Teretrius nigrescens on LGB population growth at six temperature 
conditions at 40 - 55 % relative humidity. 
E 
<B </) +1 
C 
re 
E 
o o 
o 
c o 
3 a. o a. 
m o 
12 -i 
10 
? Control 
H Batan 
H Benin 
aKARI 
D Oaxaca 
0 Tlaltizaspan 
Temperature °C 
Figure 4.4: Effect of Teretrius nigrescens on LGB population growth at six temperature levels 
at 70 - 85 % relative humidity conditions. 
87 
24 27 30 
Temperature (°C) 
Figure 4.5: Effect of temperature on grain weight loss prevention by five strains of Teretrius 
nigrescens at 40 - 55 % relative humidity (control = 0 % protection). 
120 -i 
16 21 24 27 
Temperature (°C) 
Figure 4.6: Relative effect of temperature on gram weignt loss prevention by five strains of 
Teretrius nigrescens at 70 - 85 % relative humidity conditions (control = 0 % protection). 
88 
Table 4.1: Effect of temperature, humidity and Teretrius nigrescens strain on indicators of 
grain loss (mean ± sem) caused by LGB on maize grains after 70 weeks. 
Grain Weight Loss (%) Grain Damage f%1 LGB Frass produced fgl 
RH=70-85 RBM0-55 RH=70-85 RH=40-55 RH=70-85 RH=40-55 
Population n Mean ± sem Mean ± sem Mean ± sem Mean ± sem Mean ± sem Mean ± sem 
18 "C 
Batan 6 9.35 ± 1.85a 7.57 ±0.65" 52.01 ±1.00°b 40.01±2.13,b 11.96±1.05" 8.08 ± 0.50b 
Benin 8 9.56 ± 1.09a 9.62*0.47* 52.86 ±1.47°b 42.28 ±1.23*b 12.10 ±0.96" 9.48 ± 0.37b 
KARI 6 17.85 ±4.02*b 9.41 ±0.98' 55.14 ±1.49b° 41.48 ± 2.1 Tb 21.83 ±5.35" 9.12±0.76b 
Oaxaca 6 11.11 ±4.25 ,b 9.99 ± 0.62* 53.67 ±2.05" 44.13 ± 1.2 l'b 11.67 ±1.09" 9.73 ± 0.40b 
Tlaltizaspan 6 9.06 ±0.89" 8.23 ± 0.58" 53.01 ±0.85"b 38.25 ±1.56" 13.21 ± 1.73" 8.64±0.51b 
Control 6 23.68 ±2.59b 13.44 ±0.34b 55.39 ±1.44c 46.55 ± 0.63b 20.86 ±1.51* 12.17 ±0.26' 
21 "C 
Batan 6 15.41 ± 1.25" 32.22 ±1.98° 72.74 ±3.48'b 68.13 ±1.88" 16.07 ±0.95" 22.43 ± 0.78b 
Benin 8 16.73 ± 0.82" 29.95 ±2.19* 72.76 ±1.44* 67.34 ±2.80" 16.10 ±0.58" 22.06 ±1.44b 
KARI 6 27.49 ± 5.04b 29.30 ±1.90° 78.83 ± 0.63b 66.64 ±2.53" 22.37 ±4.27" 21.88±l . l lb 
Oaxaca 6 13.99 ± 0.92* 30.04 ±1.25* 68.09 ±2.05"b 69.15 ±1.23" 14.61 ±0.94" 22.49 ± 0.93b 
Tlaltizaspan 6 16.43 ± 1.07" 27.33 ±2.24* 71.66 ±2.08*b 66.87 ±1.37" 15.57 ±1.12" 22.43 ± 0.78b 
Control 6 41.17 ±2.14° 55.28 ± 0.98b 81.90 ± 1.49= 82.82 ± 1.3 9b 21.77 ±3.13° 40.22 ±1.32* 
24 "C 
Batan 6 23.30 ±1.20* 23.33 ±0.83* 79.50 ±1.99" 64.13 ±1.60° 20.43 ± 0.71sb 16.83 ±0.55° 
Benin 8 20.42 ±0.80* 27.23 ± 1.61b 74.94 ±1.86"b 67.14±1.84* 18.62 ±0.75b 19.51 ± 1.08^ 
KARI 5 41.09 ±5.21b 28.47 ±2.06b 85.54 ±0.88b 68.13 ±3.65* 22.90 ± 1.17** 20.60 ±1.45b 
Oaxaca 6 22.92 ±1.66* 24.50 ± 0.67°b 81.00 ±2.59* 64.14 ±0.74° 20.57 ±1.49b 17.68 ± 0.51^ 
Tlaltizaspan 6 20.98 ±1.61* 25.19 ±1.27b 78.43 ± 1.35* 66.13 ±1.51* 18.90±l.llb 17.87 ±0.82bc 
Control 6 63.79 ±3.65° 71.31 ±0.61= 94.70 ±0.48° 87.58 ± 0.58b 28.90 ±4.83* 44.12 ±0.31* 
27 *C 
Batan 6 25.58 ±2.21* 18.99 ±0.59* 77.45 ±1.82* 54.42 ±1.39*. 26.09 ±2.90b 13.50 ± 0.46° 
Benin 8 22.67 ±2.10* 23.40 ±1.50b 79.73 ±1.61"b 59.96±2.11* 20.06 ±1.59b 16.08 ±0.94bo 
KARI 6 34.57 ±2.96* 25.69 ± 2. l l b 84.03 ±1.14b 64.46 ±2.19' 25.31 ± 2.12b I8.07±1.24b 
Oaxaca 6 20.68 ±2.21* 21.09 ±0.90°b 77.73 ± 1.05* 58.02 ±1.04°b 20.34 ±1.15b 14.92 ±0.61° 
Tlaltizaspan 6 22.40 ± 3.40° 24.50 ± 0.66°b 77.43 ±1.90°b 61.29 ±0.90° 21.79 ±1.52b 16.13 ±0.49b° 
Control 6 69.60 ±4.88b 54.63 ±1.16° 95.86 ±0.85° 80.37 ±1.26° 37.56 ±6.47* 33.88 ±0.60* 
30 "C 
Batan 6 16.97 ±2.59* 29.25 ±1.24b 76.94 ±4.44* 65.16 ±1.51*b 13.12 ±1.70b 19.88 ±0.82b 
Benin 14 31.76 ±3.69b 24.84 ± 0.47b 82.13 ±0.99"b 61.77 ±1.05° 21.40 ±1.70' 17.10 ±0.33° 
KARI 11 30.05 ±1.99b 31.03 ±1.75b 86.68 ±1.28b 68.31 ±1.41b 21.21 ±1.19* 21.06 ±1.06b 
Oaxaca 6 17.73 ± 1.60° 24.36 ±0.81* 78.00 ± 3.26* 62.81 ± 0.88" 13.53 ±1.21b 16.81 ±0.57° 
Tlaltizaspan 11 20.96 ±1.59* 32.02 ±1.89*b 79.21 ± 1.67°b 66.45 ±1.87° 19.00 ±1.47*b 21.10 ±0.97b 
Control 6 78.39 ±1.62c 79.11 ±1.20° 97.78 ± 0.48° 86.52 ±0.44° 16.79 ±2.06"b 43.31 ±0.45* 
33 'C 
Batan 6 20.62 ±0.92* 15.83 ± 0.27° 66.83 ±2.54* 34.93 ± 0.62° 16.25 ±1.28° 8.06±0.14b 
Benin 8 20.79 ± 0.93* 15.39 ±0.46° 70.34 ±2.26° 33.90 ±1.93" 16.44 ±0.75* 7.74 ± 0.27b 
KARI 6 36.57 ± 7.12b 15.92 ±1.47* 74.81 ±3.34" 38.58 ±1.26b 16.40 ±6.61* 9.36 ± 0.76b 
Oaxaca ? 6 20.49 ±1.33* 14.84 ±0.56* 69.64 ±3.88" 32.71 ± 1.24" 16.28 ±1.10° 7.59 ± 0.32b 
Tlaltizaspan 6 19.78 ± 1.07* 15.69 ±0.21b 67.72 ± 2.30° 35.60 ± 0.52" 15.89 ±0.98" 8.12 ± 0.1 lb 
Control 6 71.74 ±2.95° 29.94 ±1.47° 90.85 ± 0.76b 48.18 ±1.44b 25.43 ± 5.50" 14.36 ± 0.76" 
36 "C 
Batan 5 10.99 ± 0.20' 17.60 ± 0.371* 44.50 ± 2.82*b 37.18 ±0.82b 9.00 ± 0.26° 7.94±0.18b 
Benin 5 11.30 ±0.45* 14.41 ± 0.44* 47.26 ± 2.83°b 29.63 ± 0.82" 9.84 ±0.40" 6.69 ± 0.23b 
KARI 5 12.68 ± 2.26* 13.87 ±0.15* 42.55 ±1.58b 29.47 ±1.40" 8.77 ±0.46" 6.47±0.12b 
Oaxaca 6 12.54 ± 2.02* 16.76 ±0.25b 50.97 ±6.13"b 35.64 ±0.84b 10.79 ±1.76" 7.49 ± 0.19b 
Tlaltizaspan 5 12.09 ± 0.32* 17.61 ±0.18bo 47.12±1.17*b 36.53 ± 0.76b 10.00 ±0.30" 7.63±0.15b 
Control 6 26.47 ±9.39b 21.76 ±1.96° 59.11 ±5.41° 36.60 ±1.35b 17.56 ±4.64" 9.37 ±0.63° 
Means within sub columns followed by the same letter are not significantly different (SNK test, after A N O V A ; / < O.OS). 
89 
4.5 Discussion 
Temperature influences population trends of poikilotherms through its effect on physiological 
activities of the organism which can be detected in its development time, fecundity and 
survival (Hagstrum and Milliken, 1988). The importance of relative humidity in the 
development of storage pests has been demonstrated by Markham (1981), Bell and Watters 
(1982) and Throne et al. (1994) and of T. nigrescens by Oussou et at. (1998). Predicting the 
rate of growth from ambient temperature and humidity levels is important in biological control 
both in their effect on the natural enemy and on the pest species. Other than rainfall, these two 
parameters also influence the global distribution, dispersal and establishment of new 
populations of pests and their natural enemies and the level of damage or control they exert 
(Thomson et al., 2009). Substrate-boring storage insects are generally adapted to minimal 
fluctuations in humidity and temperature conditions. Thus, estimates based on constant 
temperature and humidity conditions are likely to be more applicable to this category of 
insects. In this experiment response variables were likely to be influenced by the effect of 
these conditions on both species and their interaction with each other. Therefore most 
responses (e.g. grain damage, grain protection, frass and LGB population growth) are only 
compared between strains within the other two fixed variable levels or within the same 
temperature and different humidity levels. 
Teretrius nigrescens survived all temperature levels for ten weeks, with limited population 
growth at extreme temperature regimes. Predation on LGB and resultant control of damage 
ranged between 20 %? and 80 %. The apparent high grain protection index at 36 °C was most 
likely due to the direct negative effects of temperature on LGB survival and population growth 
and thus positive effect on grain damage. Under such conditions, minimal predation by the 
surviving T. nigrescens adults is likely to result in a disproportionate decline in comparative 
damage levels relative to the control. 
90 
Earlier workers found 26 - 30 °C to be optimum for T. nigrescens (Rees et al., 1985), 
corroborating the findings of the present experiments. Also, dry conditions seem to limit T. 
nigrescens more than its predator resulting in a lower index of damage control supporting the 
findings of Oussou et al. (1998). In fact, under dry conditions, there was a significant decrease 
in P. truncatus damage control exerted by T. nigrescens at the optimal temperature of 27 °C 
for almost all populations. Due to vigorous multiplication in the predator population, the 
shortage of prey resulting in reduced survival of the predator larvae, or a delay in maturity 
could have caused this decline. Disproportionately high predation resulting from rapidly rising 
T. nigrescens populations and their effect on T. nigrescens survivorship have been reported by 
Oussou et al. (1998) and Rees et al. (1990). Egg and larval cannibalism has been suspected to 
occur under high density rearing conditions in the absence of suitable prey (Atcha and 
Borgemeister, IITA rearing manual). Rapid early population growth, resulting into crowding 
could have similar effects if cannibalism occurs, although its effects would be greater in the 
second generation larvae. 
At 24 °C, the Batan population was marginally superior to other populations under humid 
conditions, while the Oaxaca population exerted better LGB population reduction at lower 
than higher humidity levels. These reductions are not corroborated by other measures of 
survival. Again, two scenarios are possible. A voracious predator population possibly exerted 
greater control of the prey, eventually limiting its own population build-up. This results in the 
eventual observation of low numbers of both species. This scenario is suspected for the Benin 
and Oaxaca population at 30 °C. A second possibility, also associated with inconsistent trends 
observed under high humidity, is the sex ratio of the predator in each set up. Teretrius 
nigrescens cannot be sexed using observable secondary features (Boye, 1988). Consequently, 
the 20 insects starting culture could have had varying proportions of females, with a resultant 
variation in the contribution of each starting individual to the population counted at the end of 
91 
the experiment. This random event is however more likely to increase the variation of 
observations within treatments across the all populations than to be more pronounced at a 
specific temperature level. In both cases, however, it would be inconclusive to compare the 
predator populations from these data without conducting single predation and development 
experiments, which would include post experimental sex determination. 
Extreme damage levels were observed in this experiment. Between 24 °C and 30 °C, there 
was near total destruction of the grains with most damaged kernels barely discernible. Such 
levels of damage are unlikely in the storage environment in such a short time and were the 
result of a high initial infestation level. The extreme damage of maize grain could also have 
limited LGB development and hence T. nigrescens. Lower LGB infestation rates would 
aggravate the effects of sex ratio differences between treatments and replicates, while higher 
maize quantities would significantly increase the time required to dissect and count insects 
from the grains, which poses logistical limitations. 
The hypothesis that biotypes of P. truncatus and T. nigrescens exist has been invoked to 
explain the variation in pest status and performance of the predator under different agro-
 ecological zones (Nansen and Meikle, 2002). Geographical populations of the pest have 
shown significant variation in fitness and phenotypic traits including development time, 
longevity, oviposition rate fertility, enzyme activity and diversity of gut flora (Guntrip et ah, 
1996; Vazquez-Arista et al., 1999; Nansen and Meikle, 2002). Such variation has not been 
clear in T. nigrescens. Anecdotal evidence shows that a comparison of mid-altitude Mexican 
T. nigrescens and the Costa Rican population originally released in West Africa did not reveal 
significant variation in their response to temperature variation (Anonymous, 1999). However, 
the hypothesis that there are different populations of T. nigrescens adapted to different 
climatic conditions, which would affect their efficiency, could not be conclusively tested by 
92 
this study. Complete life table studies, or at least studies of the effect of temperature and 
humidity conditions on the developmental, fitness phenotypic characteristics would be 
necessary. 
93 
4.6 References 
Anonymous (1999). Integrated Management of Maize Pests and Diseases. Annual Report, 
Plant Health Management Division, 1999. International Institute of Tropical 
Agriculture, Cotonou, Benin. 
Bell, R. J. and Watters, F. L. (1982). Environmental factors influencing the development and 
rate of increase of Prostephanus truncatus (Horn) Coleoptera: Bostrichidae) on stored 
maize. Journal of Stored Products Research 18: 131 — 142. 
Borgemeister, C , Schneider, H., Affognon, H., Schulthess, F., Bell, A., Zweigert, M. E., 
Poehling, H. M. and Setamou, M. (2001). Impact assessment of Teretrius nigrescens 
Lewis (Col.: Histeridae) in West Africa, A predator of the larger grain borer, 
Prostephanus truncatus (Horn) (CoL: Bostrichidae). In: 1st Symposium on Biological 
Control of Arthropods, 17 t h -21 s t September 2001, Honolulu, Hawaii. 
Boxall, R. A. (2002). Damage and loss caused by the larger grain borer, Prostephanus 
truncatus. Integrated Pest Management Reviews 7: 105 — 124 
Boye, J. (1988). Auto ecological Investigations on the Behaviour of the Larger Grain Borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in Costa Rica. PhD Thesis, 
Institute for Phytopathology, Christian Albrechts University, Kiel. 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). 
Release and establishment of the predator Teretriosoma nigrescens Lewis for the 
biological control of Prostephanus truncatus (Horn) in Kenya. African Crop Science 
Journal 4: 325-337. 
Guntrip, J., Sibly, R. M. and Smith R. H. (1996). A phenotypic and genetic comparison of 
egg to adult life-history traits between and within two strains of the larger grain borer, 
Prostephanus truncatus Horn (Coleoptera: Bostrichidae). Journal of Stored Products 
Research 32: 213 - 2 2 3 . 
94 
Hangstrum, D. W. and Milliken, G. A. (1988). Quantitative analysis of temperature, 
moisture and diet factors affecting insect development. Annals of the Entomological 
Society of America 81: 539 — 546. 
Helbig, J. (1999). The Ecology of Prostephanus truncatus in Togo with particular Emphasis 
on Interaction with the Predator Teretriosoma nigrescens. Eschborn: GTZ. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
predatory beetle Tereti'ius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 - 306. 
Jembere, B., Obeng- Ofori, D. and Hassanali, A. (1995). Products derived from the leaves 
of Ocimum kilmandscharicum (Labiatae) as post harvest grain protectants against the 
infestation of three major stored insect product pests. Bulletin of Entomological 
Research 85: 361-367. 
Markham, R. H. (1981). The Ecology of Insect Populations in Maize Storage Cribs in 
Nigeria. PhD Thesis, University of London. 
Meikle, W. G., Rees, D. and Markham, R. H. (2002). Biological control of the larger grain 
borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest 
Management Reviews 7: 123 — 138. 
Nang'ayo, F. L. O. (1996). Ecological Studies on the Larger Grain Borer in Savanna 
Woodlands of Kenya. PhD Dissertation, University of London, 179pp. 
Nansen, C. and Meikle, W. G. (2002). The biology of the larger grain borer, Prostephanus 
truncatus (Horn) (Col.: Bostrichidae). Integrated Pest Management Reviews 7: 91— 
104. 
Nansen, C., Meikle, W. G. and Korie, S. (2002). Spatial analysis of Prostephanus truncatus 
(Horn) (Coleoptera: Bostrichidae) flight activity near maize stores and different forest 
types in southern Benin. Environmental Entomology 30: 1135-1143. 
95 
Oussou, R. D., Meikle, W. G. and Markham, R. H. (1998). Factors affecting larval 
survivorship and development rate of Teretriosoma nigrescens Lewis. Insect Science 
and it Application 18: 53-58. 
Rees, D. P., Rivera, R. R. and Rodriguez, F. S. H. (1990). Observations on the ecology of 
Teretriosoma nigrescens (Lewis) (Col., Histeridae) and its prey Prostephanus 
truncatus (Horn) (Col., Bostrichidae) in the Yucatan peninsula, Mexico. Tropical 
Science 30: 153-165 . 
Rees, D.P. (1985). Life history of Teretriosoma nigrescens (Lewis) (Coleoptera: Histeridae) 
and its ability to suppress populations of Prostephanus truncatus (Horn) (Coleoptera: 
Bostrichidae). Journal of Stored Products Research 21: 115-118. 
SAS Institute (2000). SAS/STAT Software: changes and enhancements through Release 7. 
Cary, North Carolina. 
Schneider, H., Borgemeister, C. Setamou, M., Affognon, H., Bell, A., Zweigert, M. E., 
Poehling, H. and Schulthess, F. (2004). Biological control of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) by its predator Teretrius 
nigrescens (Lewis) (Coleoptera: Histeridae) in Togo and Benin. Biological Control 30: 
2 4 1 - 2 5 5 . 
Thomson, L. J., Macfadyen, S. and Hoffmann, A. A. (2009). Predicting the effects of 
climate change on natural enemies of agricultural pests, Biocontrol 
doi:10.1016/i.biocontrol.2009.01.022 
Throne, J. E. (1994). A life history of immature maize weevils (Coleoptera: Curculionidae) 
on corn stored at constant temperatures and relative humidities in the laboratory. 
Environmental Entomology 23: 1459 — 1471. 
Tigar, B. J., Osborne, P. E., Key, G. E., Flores-S., M. E. and Vazquez-Arista, M. (1994). 
Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) by 
96 
its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bulletin of 
Entomological Research 84: 555 — 565. 
Vasquez-Arista, M., Smith, R. H., Martinez-Gallardo, N. A. and Blanco-Labra, A. (1999). 
Enzymatic differences in the digestive system of the adult and larva of Prostephanus 
truncatus (Coleoptera: Bostrichidae). Journal of Stored Products Research 35: 167 — 
174. 
97 
C H A P T E R F I V E 
Phylogenetic relationship between ten populations of Terefrius nigrescens 
5.1 Abstract 
Molecular markers are a powerful tool for studying population genetics and phylogenetics. 
They may reveal details about population sources, migration history and evolutionary events. 
Data from Teretrius nigrescens monitoring studies and results and physiological tests, suggest 
that ecological or geographical strains of do exist. This study sought to identify phylogentic 
relationships between geographical populations of T. nigrescens and to design molecular 
markers for the identification any possible trains. Ten populations of T. nigrescens were 
studied using RAPD-PCR, sequence analysis of ribosomal ITS1, 5.8S and ITS2 and mtCOI 
from 10 populations. The mtCOI variation revealed two clades associated with geographical 
regions in Central America. It also indicated considerable reduction in the genetic diversity in 
laboratory cultures. RAPD-PCR did not reveal any potential SCAR markers. The ITS 
variation mainly involved insertion and deletion (indels) of simple sequence repeats even 
within individuals. This study concludes that there exist two parapatric mitochondrial lineages 
of the predator associated with the geographical origin of populations. However, the study 
suggests the inclusion of more variable markers to clarify the observations in demographic 
time scale. 
5.2 Introduction 
The success of biological control of the larger grain borer (LGB) by T. nigrescens Lewis in 
Africa has been largely dependent on the strain used and the ecological conditions in the area 
of release (Meikle et aL, 2002; Schneider et al., 2004). This led to the hypothesis that the 
strains released would only succeed where environmental conditions were suitable. The low 
pest status of the LGB and presence of the predator in a wide range of environmental 
98 
conditions in Africa led to the hypothesis that ecotypes exist. However, a review of the 
literature does not reveal distinct discontinuity, among geographical populations making it 
difficult to estimate the level of differentiation that would be expected among and between the 
populations. Knowledge of the genetic structure of T. nigrescens populations has potential use 
in guiding the exploration for new populations, monitoring released populations and their 
interactions with the environment, other populations and the pest. 
Molecular markers have become invaluable in ecological and evolutionary studies. In 
entomology particularly, DNA markers provide a means of identification of organisms where 
morphological differences are not obvious or where taxonomically important stages or 
characters are not clear (Loxdale and Lushai, 1998). Despite these attributes, the choice of the 
marker to apply in a specific situation is sometimes subjective, as a result of the variation in 
evolution rates of specific loci (Caterino et ah, 2000). The challenge is to find a gene that is 
variable enough to reveal significant differences, yet conserved enough to be meaningfully 
analysable. 
Mitochondrial DNA (mtDNA) is now commonly used in phylogenetic and population genetic 
studies. It is matrilineally inherited, expressed, nearly conserved in function and arrangement 
across many taxa and evolve much faster than the nuclear genes (Hwang and Kim, 1999). 
Therefore, gene arrangement, sequence variation and protein sequences can all be used at 
different levels of taxonomic studies. The existence of universal primers for mitochondrial 
genes enables the amplification of loci without detailed prior sequence information (Simon et 
a!., 1994; Zhang and Hewitt, 1997). The subunit 1 of the mitochondrial cytochrome oxidase 
gene (mtCOI) is widely used due to variation in the rate of evolution in different regions (Lunt 
et aL, 1996). Coding genes, however, have a greater restriction to applicability in 
differentiating between populations. 
99 
The internal transcribed spacer (ITS) regions provide a powerful tool in population and 
phylogenetic studies since both hypervariable nucleotide sequence and less variable folding 
pattern of the initial RNA transcript of these sequences can be studied (Muller, 2007; 
Coleman, 2009). The nuclear ribosomal DNA (rDNA) consists of bundles of tandem arrays of 
genes of high copy number, containing both coding and non-coding regions (Zhang and 
Hewitt, 2003). Nuclear rDNA spacer regions evolve much faster than coding regions since 
mutations are not limited by fitness constraints. Nuclear DNA undergoes genetic 
recombination resulting in possible nucleotide and length variation within individuals and 
populations. The process of concerted evolution somewhat corrects these difference making 
certain sequences more or less fixed within breeding populations (Hwang and Kim 1999; 
Zhang and Hewitt, 2003). This makes rDNA useful in studying the mating interactions 
between populations (Fritz et ah, 1994; Zhang and Hewitt, 2003). 
The separate introduction of two populations of T. nigrescens into Africa, isolated over 800 
kilometres apart, raises the possibility that distinct strains were involved. Both strains however 
were not effective in controlling the larger grain borer in warm dry climates (Giles et ah, 
1996; Hill et ah, 2003; Hodges et ah, 2003; Schneider et ah, 2004). After extensive surveys 
using pheromone traps, Tigar et ah (1994) modelled the conditions responsible for the survival 
of both species in Mexico and inferred that they could exist in a wide range of regions. Rees et 
ah (1990) recovered both LGB and T. nigrescens under very warm, dry conditions in the 
Yucatan Peninsula. It is thus possible that the limitations of the T. nigrescens populations 
introduced into Africa are partially due to their inherent genetic limitations. To address this 
problem, four populations of the predator from Mexico and Honduras have been imported into 
the quarantine facilities of ICIPE, Nairobi, for eventual release. Pertinent questions still to be 
answered are: What is the extent of genetic variation in T. nigrescens!. Do geographical 
populations of T. nigrescens represent distinct ecotypes? Can molecular markers be used to 
100 
distinguish between these ecotypes? How can this genetic diversity be incorporated into a 
biological control program? To investigate this questions, the sequence variation of mtCOI 
gene and both ITS1 and ITS2 spacer regions among the populations of the predator was 
studied (Table 2.1). 
5.3 Materials and Methods 
5.3.1 DNA extraction 
Total genomic DNA was isolated from the heads and thoraxes of individual adult T. 
nigrescens using the Phenol-Chloroform protocol (Sambrook et al., 1989). DNA was stored in 
TE buffer (10 mM TrisHCl pH 8, 1 mM EDTA) at -20 °C until ready for use (Chapter 2). For 
out-group species, single legs were used and the rest of the insect submitted for identification. 
Insect identification to species level was done by Prof. Pierpaolo Vienna, Venezia Lido, Italy. 
5.3.2 Polymerase Chain Reaction 
RAPD-PCR trials 
Random amplified polymorphic DNA Polymerase Chain Reaction (RAPD-PCR) (Welsh et 
al., 1990; Williams et al., 1990) was used to screen the wide diversity within and among 
geographic populations. The KARI, Oaxaca, Tlaltizaspan, El Batan and Benin geographic 
populations were used in this study representing the widest geographical range of the 
populations available. Twenty five decamer random primers previously used in insect genetic 
diversity studies (Moya et al., 2001; Omondi et al, 2004) were screened for polymorphism, 
and reproducibility. Thirteen of these were selected and used to amplify DNA from samples 
across populations. 
101 
The PCR reaction was done in a 25 ul mix containing lx Genscript Taq polymerase buffer 
with 3.0 mM MgCl2, 400 uM dNTPs, 300 nM single RAPD Primer, 1U Taq polymerase and ~ 
20 ng template DNA. The amplification was done in an ABI 9800 thermal cycler programmed 
as follows: one cycle of 94 °C for 3 min, followed by 35 cycles of a minute each of 94 °C 
denaturation, annealing at 35 °C and extension at 72 °C, and a final extension step of 72 °C for 
10 min. Products were held at 4 °C till ready for use. The amplicons were separated on 2 % 
agarose (in TAE) gels stained with ethidium bromide under 2.5 volts/cm potential gradient for 
2 hours. Gels were visualised under UV light. Since there were no bands for potential use as to 
develop sequence characterised amplified region (SCAR) markers, RADP PCR was not 
pursued further. 
Cytochrome Oxidase 1 
A 1200 bp region of the mtCOI gene, was amplified by PCR using the universal insect mtCOI: 
C1J-17363 (5' TATAGCATTCCCACTAATAAATAA 3') forward and TL2-N-3014 (5' 
TCCAATGACTAATCTGCCATATTA 3') reverse primers previously described by Zhang 
and Hewitt (1997). The PCR amplifications were carried out in 30 ul reaction volumes 
containing lx Genscript Taq polymerase buffer with 1.5 mM MgCl2, 400 uM dNTPs, 150 nM 
each of the forward and reverse primers, 1U Taq polymerase and ~ 20 ng genomic DNA. The 
thermal cycling programme was 94 °C 3 min, followed by 35 cycles of denaturation at 94 °C 1 
min, annealing at 52 °C for 40 s and extension at 72 °C for 90 s, and then a final extension 
step of 72 °C for 10 min. The PCR product was separated on a 1.2 % agarose gel. The 
amplification product of 1200 bp size was excised from the gel and purified using Quiagen 
Min Elute Gel purification kit (Quiagen) according to manufacturer's instructions (Chapter 2). 
The eluted product was sequenced in both directions using the PCR primers on an ABI chain 
termination sequencing technology at a commercial facility (Macrogen Inc, Korea). (Chapter 
2) 
102 
Internal transcribed spacer regions (ITS) 
Amplification of the two ITS regions and 5.8S rRNA coding site was done as above using 
standard forward primer ITS 5 (5' GGAAGTAAAAGTCGTAACAAGG 3') and reverse 
primer ITS4 (5' TCCTCCGCTTATTGATATGC 3') (White et al., 1990). The thermocycling 
conditions were similar to those of mtCOI (above) but with an annealing temperature of 54 °C 
and running length of 25 cycles. Fewer cycles were used to increase the integrity of the 
product for cloning. Preliminary direct sequencing of ITS genes produced equivocal results; 
hence cloning was done to reveal possible intra-genomic variation. 
The gel purified fragment was cloned into pGEMT-E vector (Promega) and used to transform 
DH5a chemically competent cells as already described (Chapter 2). Recombinant colonies 
were screened by PCR to confirm the presence of the insert. Five positive recombinant 
colonies per insect sample were selected for bacterial propagation and plasmid purification. 
Plasmid purification was done using Quiagen Miniprep Kit according to manufacturer's 
protocol. Plasmid and DNA samples were sequenced by the ABI terminator system at 
Macrogen Inc. Korea using M13FpUC and M13RpUC universal primers (Promega) at 
Macrogen Inc. 
5.3.3 PCR-RFLP analysis 
Sequence alignment revealed parsimony informative segregating sites, associated with 
populations, so, NEB CUTTER version 2 (Vincze et aL, 2003) software was used to predict the 
diagnostic utility of the restriction endonucleases cleaving parsimony informative sites. The 
selected restriction enzymes were then used to digest PCR products of the mtCCT from the 
five populations of T. nigrescens (Ghana, Oaxaca, Benin, KARI and Tlaltizaspan) 
representing the main sub clusters in sequence analysis. 
103 
5.3.4 Data Analysis 
DNA sequences were edited manually to confirm the positions of ambiguous nucleotide calls 
using BIOEDIT (Hall, 1999). The termini of the sequences were trimmed to remove regions of 
unreliable sequencing quality. The reverse complement of the reverse sequence was contiged 
with the forward sequence after pair wise alignment of the termini produced at least 100 bp 
complete consensus overlap region. All complete sequences were aligned together and 
segregating sites reconfirmed by comparison with the trace diagrams and translation using the 
Invertebrate Mitochondrial codon table. Since mtCOI is an expressed gene, only fragments 
translating into a complete reading frame protein after sequence confirmation were 
considered. This would avoid error from possible nuclear mitochondrial pseudogenes 
(Bensasson et al., 2000). Haplotype identification and frequencies were done manually by 
sequences alignment and grouping of sequences according to similarity in BloEDIT. Haplotype 
network was determined in TCS version 2.1 (Clement et al., 2000). Haplotypes were grouped 
into series of clades from the smallest group of individual haplotypes. Due to the distinct 
separation of three clades (at least 22 steps), and rearing history of laboratory colonies, 
population inferences were based on the haplotype network of trees rather than a statistical 
nested clade analysis. 
Multiple sequence alignment was performed using Clustal W version 1.8 as implemented in 
BloEDIT. The numbers and types of segregating sites, Kimura 2-parameter distances (K2P) 
between individuals, populations and groups of populations were determined using MEGA 4.0 
(Kumar et al., 2001). In this approach, the mean number of base substitutions per site from all 
59 sequences and 1084 bases per sequence was conducted using the Maximum Composite 
Likelihood method in MEGA4 (Tamura et al., 2004; 2007). Average within and between' 
geographic population K2P distances and their standard deviation were calculated using MEGA 
4.0. Due sequence similarity among samples from Honduras were combined into one 
104 
population. In determining the genetic diversity between groups of populations, the 'Store' 
Population was analysed separately since its antiquity could not be ascertained. 
The phylogenetic relationships between the samples were inferred using a neighbour-joining 
tree on MEGA 4 (Saitou and Nei, 1987). Sequence alignment was done in Clustal X to create 
an alignment file for MEGA. The evolutionary relationships were inferred with Neighbour-
 Joining tree method, with 1000 replicates. The evolutionary distances used to compute tree-
 branch length were determined. Similar length mtCOI sequence from Teretrius americanus 
LeConte (= T. latebricola Lewis) was used as an out group species. 
The ITS1 and ITS2 sequences were aligned manually on BroEDIT, since, due to numerous 
indels, automatic alignment was not satisfactory. The repeat regions were identified manually 
from the BIOEDIT alignment and using SSR Finder (Benson, 1999). The boundaries of the 
18S, 5.8S and 21S rRNA genes were delimited based comparisons with sequences of 
Timorcha and Tetranynchus sequences (Navajas et al., 1998; Gomez Zurita et al., 2000). 
5.4 Results 
RAPD-PCR profiles obtained using all the 13 primers were reproducible with numerous 
polymorphic bands (Figure 5.1). However the polymorphism of profiles was not associated 
with any geographical or demographic historical characteristics of the populations. 
105 
CD C30C& CS C33 030002 So© 0 3 (33 OS (»M<Iffi®StifeO tJfeg'ieS ^JQ CD 
^*^<- .V' i - :**i$3& £ 
Figure 5.1: RAPD PCR profiles of four populations of Teretrius nigrescens from primer OPB 
11 run on 2% agarose (in TAE) gel. Population used are: K - Kari, Bn - Benin; Bt - Batan; Ox 
- Oaxaca and Ts - Tlaltizaspan. M represents 1 kb DNA ladder. 
106 
5.4.1 MtCOI Sequence variation 
A 1200 bp region was amplified with the mtCOI primers. After editing, a 1084 bp fragment 
was used in the analyses whose results are presented below. The mtCOI sequence was 63.4 % 
A/T rich. The nucleotide frequencies were 0.285 (A), 0.349 (T), 0.203 (C), and 0.163 (G), 
comparable to mitochondrial sequences in other Coleoptera (Cognato et al., 2003; Mynhardt 
et al., 2007). There were 102 polymorphic sites, of which 72 were parsimony informative 
(Figure 5.2). This variation occurred most at third (83.3 %), first (13.7 %) and second (9 %) 
codon positions. The rate and direction of nucleotide substitutions is summarised in Table 5.1. 
No insertions or deletions were observed among T. nigrescens samples or species of Histeriae 
used as out groups. 
5.4.2 Genetic diversity 
Pair-wise divergence between populations based on K2P parameter are shown in Tables 6.2 
and 6.3. There was greater evolutionary divergence among samples from the Honduras/Costa 
Rica than there was from the samples from Mexico. However, the Mexican Tlaltizaspan 
population was closest to individuals from Ghana and Malawi both originating from the Costa 
Rica population reared and released in Benin. 
The evolutionary relationships between the 57 T. nigrescens samples revealed two major 
clades supported by > 98% bootstrap values (Figure 5.3). One cluster contained most samples 
originally recovered from Mexico, while the other grouped samples from Honduras, Costa 
Rica and Tlaltizaspan (Mexico). Other minor clusters were highly supported too (e.g. all Store 
samples). These grouped samples from highly monomorphic populations or single haplotype. 
Twenty-two haplotypes were observed of which 15 were unique (observed in only one 
individual each) (Figure 5.2). Haplotype diversity was highest among the Honduras 
populations (7 haplotypes) and lowest among the KARI, Store and Malawi populations (2 
107 
haplotypes each) (Figure 5.4). Haplotype HI9 was both the most diverse and the most 
frequent, occurring in five of the populations and 23 % of the individuals (Figure 5.5). 
5.4.3 PCR-RFLP identification of populations 
In silico restriction analysis showed that the restriction endonucleases AcfL, BstYl, Ddel, Fold 
and Rsal, had their recognition sites flanking parsimony-informative sites and so could resolve 
four populations into two major clades and (Table 5.4). The results of these digestions are 
shown in Figure 5.6. 
108 
Variable positions marked from the beginning of the 1083 by fragment 
H8 
H10 
H9 
H12 
H15 
H13 
H14 
H3 
H4 
HI 
H2 
Hll 
H7 
H5 
H6 
H16 
H18 
H19 
H20 
H21 
H22 
H17 
1111 
245669112 3 
7213931235 
GAATATTGTA 
1111122222 
4677745566 
4247865614 
AGATTCACGT 
.G C. 
. G. . . . TC . 
.AG.C.GT. 
.AG.C.GT. 
.AG...GT. 
2 2 2 2 2 3 3 3 3 3 
7 7 8 9 9 0 1 1 1 2 
0 4 2 4 7 6 15 8 7 
ACTCGGCTCA 
G 
G 
...TT 
...TT 
333 3344444 
4 4 6 8 9 0 0 1 2 3 
2 83 1 6 2 8 4 0 0 
TTCGTCGGCA 
44444445 55 
4457789035 
147 5842112 
CTCTCTGCCA 
5 5 5 5 6 6 6 6 6 6 
7 8 8 8 0 0 1 1 1 2 
63 583 92 587 
ACTAATATAT 
6 6 6 6 7 7 7 7 7 7 6 7 7 78 8 8 I 8 8 8 7 
5 8 8 9 0 1 2 4 56 5 7 8 9 1 2 2 2 4 4 7 7 
5 1 4 6 3 1 6 4 4 5 5 4 3 2 9 2 4 6 3944 
TAGTTTTTGT AACGTCCTTT 
8 8 8 9 9 9 9 9 9 9 8 
7 9 9 U 0 0 2 J4 4 7 
9 17 3 6 6 7 3 2 5 9 
1 1 1 1 1 1 11 
9999000000 00 
4699024556 66 
8906507366 78 
AGCCCTGGTT ATTCACTAAA GT 
G 
.A.TAA. 
.A.TAA. 
.A.TA.. 
T..T.. 
T..T.. 
.T..TT. 
C. 
C. 
.c. 
. .A 
C....C.C. 
C....C.C. 
. .G. 
A.G. 
A.G, 
A.G. 
A.G. 
. .G. 
A.G. 
.A. 
.A. 
.A. 
.A. 
.A. 
.A. 
A.G.-A. 
A.G.-A. 
.GGCCAC 
.GGCC.C 
.GGCC.C 
.GGCC 
.GGCC 
.GGCC 
.GGCC 
C. 
.C 
.C 
.c 
.c 
,c 
.T 
.C 
c 
,c 
.c 
.c 
. .c 
, .c 
, .c 
. .c 
,AG 
GA. 
GA. 
GA. 
GA. 
GA. 
GA. 
GA. 
GA. 
.A 
.A 
.A 
.A 
.A 
• GT. . A. TA. 
.A.C 
A.C 
.AC 
.AC 
.AC 
.AC 
.AC 
.AC 
. .C 
. .C 
.A. 
.A. 
.A. 
.A. 
.A. 
.A. 
.A. 
G. . . 
G. . . 
G. . . 
G. . . 
G. . . 
G. . . 
G. . . 
G. . . 
GAC. 
.C. 
.C. 
.C. 
.C. 
.c. 
A. 
A. 
A. 
A. 
T. 
, TG 
T. 
.T..TT. 
.TCA... 
.T.A... 
. T. 
.T. 
. X. 
,C..T.A. 
C..T.A. 
C..T.A. 
. .T.T. 
.AT.TG 
.AT.TG 
.AT.TG 
..T.TG 
..T.TG 
.AT.TG 
ACT 
. .T 
. .T 
. .T 
. .T 
. .T 
. .T 
.T. 
.T. 
.T. 
.T. 
.T. 
.T. 
.T. 
.T. 
.T. 
..T.A. 
CT. A. 
. CT . A. 
. .T. 
T.T. 
. . C. . . . 
...GCT... 
...GCT... 
...GCT... 
...GCT... 
...GCT... 
,..GCT.G. 
,CGGCT... 
.CGGCT... 
...GCT.G. 
.TG 
.TG 
.TG 
.TG 
.TG 
.TG 
TG. 
TG. 
TG. 
TG, 
TG. 
TG. 
C 
.C 
.c. 
.c. 
.c 
c 
C.C 
C.C 
C.C 
C.C 
C.C 
C.C 
CG...C.C 
GS...C.C 
C.ACCC.C 
.ACCC 
.ACCC 
.ACCC 
.ACCC 
.A.CC 
.A.... 
.A.... 
.A..T. 
A.CC 
A.CC 
CC 
C.A 
C.A 
. -A 
C.A 
CCCCA 
CCCC. 
C.CC. 
CCCC. 
CCCC. 
CCCC. 
.G. 
.G. 
.G. 
.G. 
.G, 
.AT. . . 
.A. . . . 
.AT. . . 
.AT... 
.AT... 
.G.ACAT... 
.G.ACATC.. 
.ACAT... 
.ACAT.,. 
C. 
C . C . 
... T... AC. 
...T... AC. 
TA. . . .A. . . 
TA. . . .A. . . 
TA. . . .A.C. 
TA....A... 
TA....A... 
TA. . . .A. . . 
C. 
C. 
.... AC 
... AC 
... AC 
... AC 
... AC 
i. . . AC 
.... AC 
.... AC 
... AC 
.... A. 
T AAC AC 
.G. 
.G. 
G.TACT. 
G.TACT. 
G.TACT. 
G.TACT. 
G.TACT. 
G.TACT. 
CC 
CC 
CC 
. CC 
. CC 
.CC 
T. . 
TA. 
TAT, 
TAT 
TAT 
TAT 
TAT 
TAT 
..A.C. 
..A... 
T.AACC 
AACC 
AACC 
AACC 
AACC 
AACC 
C 
c. 
c 
,c 
,c 
.c. 
. . .G 
. . .G 
. . .G 
. . .G 
. .TG 
G G 
AC 
A. 
.C 
.C 
.C 
.C 
.C 
.G AC 
Figure 5.2: Nucleotide variation among 22 mtCOl haplotypes (HI to H22) on 1084 bases of the mtCOI T. nigrescens used in this study. Numbers at 
the top of each nucleotide position represent the segregating site position relative to the first nucleotide of this fragment. Haplotypes are arranged in 
order of relative similarity. Dots represent similarity with haplotype H8 while the nucleotide in each segregating site is labeled vertically (such that 
the first is 27 bp and the last is 1068 bp from the origin). 
109 
Table 5.1: Maximum composite likelihood estimate of the pattern of nucleotide substitution. 
Nucleotide 
A T C G 
A - 2.33 1.36 12.89 
T 1.9 - 18.87 1.09 
C 1.9 32.39 - 1.09 
G 22.5 2.33 1.36 -
 Each entry shows the probability of substitution from one base (row) to another base (column). 
Table 5.2 Estimates of evolutionary divergence over sequence pairs between populations  
Kibok Oaxac Hondura Malaw 
Population KAPJ Benin o a s Batan Store i Ghana 
0.025 
Benin 4 
0.021 0.028 
Kiboko 6 7 
0.008 0.028 
Oaxaca 7 1 0.0248 
0.041 0.036 
Honduras 7 6 0.0361 0.0379 
0.002 0.027 
Batan 2 7 0.0239 0.0109 0.0441 
0.043 0.038 0.045 
Store 2 6 0.0344 0.0387 0.0226 6 
0.047 0.031 0.049 0.023 
Malawi 3 7 0.0354 0.0425 0.0216 7 2 
0.052 0.040 0.055 0.029 
Ghana 7 5 0.0428 0.0483 0.0264 2 9 0.0261. 
Tlaltizaspa 0.052 0.032 0.054 0.035 0.021 
n 3 1 0.0399 0.0490 0.0302 8 6 0.0219 3 
110 
Table 5.3: K2P distances between pairs of major population clades of Teretrius nigrescens. 
"Store" is treated separately as the exact source is unresolved. 
Mexico Benin 
Mexico 0.0000 
Benin 0.0418 ± 0.0049 0.0000 
Store 0.0398 ±0.0052 0.0303 ±0.0036 
I l l 
34 . 
100 
MaIawiL5 
Teupasent l l 
Teupasent i2 
Teupasent l3 
Y o r o F I 
MalawIMS 
Yorom 1 
S t A n t o n i o 
— O a x a c a F 3 
— O a x a c a M I 
r- Store7 
Store 1 
Store6 
S t o r e 5 
S t o r e 2 
S t o r e 3 
Store4 
p Ghana3 
3y I G h a n a l 
95 I Ghana2 
T la l t lzaspanFI 
Ben inMI 
BeninM3 
MalawiL7 
MalawiS 
MalawM 0 
TeupasentI-4 
St Anton io2 
Ghanan3 
Ghanan4 
G h a n a n l 
T lat izaspanF2 
Ghanan2 
721 OaxacaMT 
OaxacaMS 
OaxacaMg 
BatCA11Ed 
B atari A11 
K ibokoNI 
KIbokoN2 
BatanM3 
OaxacaE13 
OaxacaD13 
OaxacaE14 
BenJnM4 
KariF3 
K ibokoM2 
OaxacaM6 
KARIM1 
S t A n t M l 
OaxaoaMS 
O a x a c a n l 
KARIM2 
KAR1F2 
KARln4 
Ben inM2 
KAR1M3 / 
Oaxacan2 -* 
Teretrius amer icanus 
Southern 
Geographical 
populations 
(Samples from 
West of Ridge) 
'Nothern' 
Geographical 
populations 
(Samples 
from East of 
Ridae) 
-i 
Figure 5.3: Evolutionary relationships of 57 samples of Teretrius nigrescens from seven 
populations analysed using UPGMA method. The tree is rooted using a homologous sequence 
from Teretrius americanus. This bootstrap consensus tree inferred from 5000 replicates is 
shown, with the bootstrap values next to the branches 
112 
Btn Bnn Ghn Kar Str Kbk Tit Oxc Hnd Miw 
Geographic Population 
Figure 5.4: Haplotype frequency of different populations of T. nigrescent. Btn- Batan; Bnn- Benin; 
Ghn - Ghana, Kar — KARI, Tit -Tlatizaspan; Oxc — Oaxaca; Hnd- Honduras; Mlw — Malawi 
• " # " • 
\ 1 1-
 - / / ? I — / / — - 1 r- t A — | 1 - » H 1 - I — ' I -
 -L) 
i—i—•—i-
 O 
Figure 5.5: Network showing the most parsimonious relationship between 22 mtCOI 
haplotypes of T. nigreescens from 10 geographical populations. Circles represent haplotypes, 
black dots signify intermediate (unsampled) haplotypes and lines between haplotypes and 
denote single nucleotide change. Major colour shades viz: Blue - Mexican populations; 
Red/Orange (Costa Rican Populations); Green; Honduras populations and grey ("Field" or 
"Store" populations with unidentified source). 
113 
Table 5.4: Predicted products of the cleavage of 1200 bp mtCOI PCR product using four 
potentially diagnostic endonucleases1. 
Enzyme Fragments Produced Remarks 
Mexico Costa Rica 
JRsal 39,432,701, 432, 740 Good 
Fokl 123,510,598 124,213,443,451 Fair 
BstYI 59, 475, 697 475,755 Good 
Acil 59, 469, 703 199,200,832 Good 
Ddel 123,212,869 124, 267, 270, 570 Fair 
fragments in bold, for each enzyme, are not distinguishable on the agarose gel 
A: 1200 bp PCR product B: Digestion with^tccl 
C: Digestion with Rsal 
M K t M Bti Bt* Bnt B1126162 Gxi 0x2 Tfi Tiz -ve 
D: Digestion with Ddel 
Figure 5.6: The -1200 bp amplicon and products of its single digestion with three restriction 
enzymes. Sample abbreviations are: K - KARI; Bt: Batan; Ox: Oaxaca; G - Ghana, Tl -
 Tlatizapan; '-ve': negative control set up without DNA. 
114 
5.4.4 Internal transcribed spacer regions 
Cloning enabled the sequencing of the whole of the ITS amplicons with flanking coding 
sequences. This enabled comparison between samples and outgroup species. The length of the 
ITS sequences ranged between 1250 and 1350 bp. Intragenomic variation of up to 10 bases 
were observed. Most of this variation involved insertion and deletion of tandem sequence 
repeat units (Appendix A). Differences were therefore in multiples of repeat lengths. Because 
of little variation in the non repeat sequences and high polymorphism in the repeat variable 
regions, the ITS was not found useful for the phylogenetic analysis in this study, without 
major editing to remove variable parts. This gene was therefore not considered further in the 
phylogenetics and identification of the populations of T. nigi'escens. 
5.5 Discussion 
Of the three methods used in this study, the mtCOI sequences were the most useful in the 
detection of population genetic variations in T. nigrescens. The RAPD-PCR technique used in 
preliminary analysis could not detect variations between populations. The RAPD-PCR 
technique amplifies fragments whose location in the genome is unknown (Loxdale and Lushai 
et al., 1998). Although the technique suffers technical reproducibility limitations (Lynch and 
Milligan, 1994) these can be minimized by optimizing reaction conditions and careful 
selection of primers to use. The RAPD-PCR technique has been used to differentiate between 
biotypes and detect interbreeding between populations of insects (Moya et al., 2001; De Barro 
and Hart, 2000; Omondi et al., 2005). This technique has also been used for the development 
of diagnostic sequence characterized amplified region sequence (SCAR) markers from by 
sequencing unique fragments (Rugienius et al, 2006). Such studies depend on prior 
identification of fixed differences in RAPD profiles in these populations (Omondi et al., 
2004). Although amplifications were repeatable, there were no population-specific markers. 
This was attributed to three possible causes: the insects might have been too closely related 
115 
genetically to have significant differences in their DNA detectable by this procedure; the 
primers used were simply not amplifying variable loci genetic populations were possibly 
sympatric at sampling locations and could not be defined by geographical sampling approach 
of this study. In the first two cases, greater primer screening would have revealed useful 
markers amplifying sites variable between populations. 
The ITS 1 and ITS 2 sequences showed comparable sizes and coding region similarity with 
the previously studied beetle species (Gomez-Zurita et al, 2000). Since the variation was 
predominantly due to 2 or 3 base pair tandem repeats it might be interesting to investigate how 
the repeat-length fit with microsatellite models in population genetic analysis. However, one 
must first resolve the apparent polyploid genotype pattern created by the intragenomic 
variation of these markers. 
The mtCOI gene sequence analysis revealed two major clades of T. nigrescens associated with 
defined geographical ranges. Apart from the "Store" samples whose origin was not clearly 
known, all insect samples from Africa clustered with their original source populations 
confirming suitability of this marker for phylogenetic analysis (Otranto et al., 2003). The 1200 
bp region sequenced in this study covered two highly variable regions and a much more 
conserved portion of this gene (Lunt et al., 1996; Zhang and Hewitt, 1997). This strategy 
promised greater accuracy since the phylogenetic differentiation of this species was initially 
unknown. Several regions of the sequence with a greater density of parsimony informative 
variable sites have been identified in this study. Shorter fragments spanning these desirable 
regions would be more practical, to reduce the potential analysis challenges associated with a 
large number of single sequence haplotypes and potential error in sequencing large fragments. 
Although very few parsimony-informative segregating sites are within 5 bases of each other, 
116 
careful primer design might reveal a single step multiplex PCR protocol for identifying these 
populations (cf. Lindell and Murphy, 2008). 
The 22 haplotypes separated into two clusters, suggesting the existence of two distinct 
maternal lineages, suggesting historical geographical and hence genetic fragmentation. This 
could have impeded interaction between populations causing independent evolution of this 
gene in the two populations. This study identified the region between Oaxaca and Tlaltizaspan 
as the possible contact zone where both subtypes exist. Lindell and Murphy (2008) have 
identified the Baja peninsula in north-western Mexico as a possible hybrid zone mitochondrial 
forms of Lizard Uta stansburiana, Baird & Girard. This hybrid zone was explained to reflect a 
complex of geological history events since the Miocene era (Lindell et al., 2006). Although all 
samples of T. nigrescens were obtained south of the Baja peninsula, the extent of the effects 
historical geological disturbances would vary for individual species depending on their 
original distribution and the effect of geological events on their eventual dispersal. Thus, a 
study of the geological and ecological history of the natural range of T. nigrescens would 
clarify this possibility. This was beyond the scope of the study, however. Alternatively, the T. 
nigrescens could have spread north and southwards from the zone around Oaxaca, founding 
the two different clades. This hypothesis is supported by the haplotype differentiation. 
There is a possibility of population fragmentation as suggested by a high frequency of unique 
haplotypes only shared among a few individuals within specific populations. The effect of 
local geographical features may also be responsible for the variation and lineage divisions 
observed in this study. It is recognized that the adaptability of an invader, rather than sheer 
genetic diversity or numbers, are responsible for the establishment of a species in the new 
range (Lloyd et al., 2005; Zayed et al., 2007; Bacigalupe, 2008). However, adaptability is also 
a factor of the diversity of the gene pool introduced upon which selection occurs to produce 
117 
resulting in an adaptable population (Roderick and Navajas, 2004). The chances of 
establishment and range expansion may therefore be higher with an increased genetic base 
(Kolar and Lodge, 2001; Allendorf and Lunquist, 2001; Phillips et al, 2008). Adequate 
sampling for a greater genetic pool is important in recovery of natural enemies (Omwega and 
Overholt, 1996; Phillips et at., 2008) and should guide future sampling plans for a more 
genetically inclusive approach to the biological control of the LGB. 
Genetic diversity in laboratory samples was generally lower than that of field samples. 
Similarly, the Malawi population showed a much lower level of mitochondrial genetic 
diversity than its parental Benin population. This suggests the disproportionate contribution of 
a few related females to the genetic pool of these populations, population expansion from a 
few haplotypes, or problems with genetic bottlenecks during recovery, quarantine or rearing 
(Lloyd et ah, 2005; Zayed et al., 2007). The demographic history of the laboratory populations 
used in this study is not clear on a few important events. For instance, insects isolated from 
Mexico were studied in the Benin laboratory but their possible contribution to the Benin 
populations is not clear (W. Meikle, personal Communication). In Kenya, mite infestation and 
possible mycoplasma infections were reported leading to culling of a part of the colony, 
though the eventual founding populations are not documented or the extent of this possible 
bottleneck known (R. Hodges, P. Likhayo, G. Kibata personal Communication). Recent 
population genetic shifts and admixture events could have contributed significantly to the 
population genetics of the populations we used and eventual performance of populations 
released for biological control. These hypotheses may be tested more efficiently in recent 
population time scales using hypervariable markers such as microsatellites. The next section 
of the study therefore focuses on the development testing and use of microsatellite makers to 
test these hypotheses. 
118 
5.6 References 
Allendorf, F. W. and Lundquist, L. L. (2003). Introduction: Population biology, evolution, 
and control of invasive species. Conservation Biology 17: 24—30. 
Bacigalupe, D. L. (2008). Biological invasions and phenotypic evolution: a quantitative 
genetic perspective. Biological Invasions doi 10.1007/sl0530-0089411-2. 
Bensasson, D., Zhang, D. -X. and Hewitt, G. M. (2000). Frequent Assimilation of 
Mitochondrial DNA by Grasshopper Nuclear Genomes. Molecular Biology and 
Evolution 17: 406-415. 
Benson, G. (1999). Tandem Repeats Finder: a program to analyze DNA sequences. Nucleic 
Acids Research 27: 573 - 580. 
Caterino, M. S., Cho, S. and Sperling, F. A. H. (2000). The current state of insect 
molecular systematics: a thriving Tower of Babel, Annual Review of Entomology 45: 
1-54. 
Clement, M., Posada, D. and Crandall, K. (2000). TCS: a computer program to estimate 
gene genealogies. Molecular Ecology 9: 1657 — 1660. 
Cognato, A. I., Harlin, A. D. and Fisher, M. L. (2003). Genetic structure among pinyon 
pine beetle populations (Scolytinae: Ips confusus). Environmental Entomology 32: 
1262-1270. 
Coleman, A. W. (2009). Is there a molecular key to the level of "biological species" in 
eukaryotes? A DNA guide. Molecular Phylogenetics and Evolution 50:197-203. 
De Barro, P. J. and Hart, P. J. (2000). Mating interactions between two biotypes of the 
whitefly Bemisia tabaci (Hemiptera: Aleyrodidae) in Australia. Bulletin of 
Entomological Research 90: 103 —218. 
Felsenstein, J. (1985). Confidence limits on phylogenies: An approach using the bootstrap. 
Evolution 39: 783 -791 . 
119 
Fritz, G. N., Conn, J., Cockburn, A. and Seawright, J. (1994). Sequence analysis of 
ribosomal ITS2 from populations of Anopheles nuneztovari (Diptera: Culiciade) 
Molecular Biology and Evolution 11: 406 - 416. 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). 
Release and establishment of the predator Teretriosoma nigrescens Lewis for the 
biological control of Prostephanus truncatus (Horn) in Kenya. African Crop Science 
Journal 4: 325 — 337. 
Gomez-Zurita, J., Juan, C. and Petitpierre, E. (2000). Sequence, secondary structure and 
phylogenetic analyses of the ribosomal internal transcribed spacer 2 (ITS2) in the 
Timarcha leaf beetles (Coleoptera; Chrysomelidae). Insect Molecular Biology 9: 591 
- 6 0 4 . 
Hall, T.A. (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis 
program for Windows 95/98A^T. Nucleic Acids Symposium Series 4: 95-98. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
predatory beetle Teretrius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 — 306. 
Hodges, R. J., Addo, S. and Birkinshaw, L. (2003). Can observation of climatic variables 
be used to predict the flight dispersal rates of Prostephanus truncatusl Agricultural 
and Forest Entomology 5: 123 - 135. 
Hwang, U.-W. and Kim, W. (1999). General properties and phylogenetic utilities of nuclear 
ribosomal DNA and mitochondrial DNA commonly used in molecular systematics. 
The Korean Journal ofParasitology 37: 215—228. 
120 
Kimura, M. (1980). A simple method for estimating evolutionary rate of base substitutions 
through comparative studies of nucleotide sequences. Journal of Molecular Evolution 
16:111-120. 
Kolar, C. S. and Lodge, D. M. (2001). Progress in invasion biology: predicting invaders. 
Trends in Ecology and Evolution 16: 199-204. 
Lindell, J. and Murphy, R. T. W. (2008). Simple identification of mitochondrial lineages in 
contact zones based on lineage-selective primers. Molecular Ecology Resources 8: 
66-73. 
Lindell, J., Ngo, A. and Murphy, R. T. W. (2006). Deep genealogies and the mid-
 peninsular seaway of Baja California. Journal of Bio geography 33: 1327-1331. 
Lloyd, C. J., Hufbauer, R. A., Jackson, A., Nissen, S. J.and Norton, P. N. (2005). Pre-
 and post-introduction patterns in neutral genetic diversity in the leafy spurge gall 
midge, Spurgia capitigena (Bremi) (Diptera: Cecidomyiidae). Biological Control 33: 
153-164. 
Loxdale, H. D. and Lushai, G. (1998). Molecular markers in entomology. Bulletin of 
Entomological Research 88: 557 — 600. 
Lunt, D. H., Zhang, D. X., Szymura, J. M. and Hewitt, G. M. (1996). The insect 
cytochrome oxidase 1 gene: evolutionary patterns and conserved primers for 
phylogenetic studies. Insect Molecular Biology 5: 153 — 165. 
Lynch, M. and Milligan, B. G. (1994). Analysis of population structure with RAPD 
markers. Molecular Ecology 3: 91 - 99. 
Meikle, W. G., Rees, D. and Markham, R. H. (2002). Biological control of the larger grain 
borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest 
Management Reviews 7: 123-138 . 
121 
Moya, A., Guirao, P., Cifuentes, D., Beitia, F. and Cenis, L. (2001). Genetic diversity of 
Iberian populations of Bemisia tabaci based on random amplified polymorphic DNA 
-polymerase chain reaction. Molecular Ecology 10: 891 - 897. 
Miiller, T., Philippi, N., Dandekar, T., Schultz, J. and Wolf, M. (2007). Distinguishing 
species. SNA 13: 1469-1472. 
Mynhardt, G., Harris, M. K. and Cognato, A. I. (2007). Population genetics of the pecan 
weevil (Coleoptera: Curculionidae) Inferred from mitochondrial nucleotide data. 
Annals of the Entomological Society of America 100: 582 — 590. 
Navajas, M., Lagnel, J., Gutierrez, J. and Boursot, P. (1998). Specieswide homogeneity 
of nuclear ribosomal ITS2 sequences in the spider mite Tetranychus urticae contrasts 
with extensive mitochondrial COI polymorphism. Heredity 80: 742-752. 
Omondi, A. B., Obeng-Ofori, D., Danquah, E. Y. and Kyerematen, R. A. (2005). Mating 
interactions between okra and cassava biotypes of Bemisia tabaci (Homoptera: 
Aleyrodidae) on eggplant. International Journal of Tropical Insect Science 25: 159 — 
167. 
Omondi, A. B., Obeng-Ofori, D., Danquah, E. Y. and Offei, S. K. (2004). Genetic 
diversity and oviposition host preference of Bemisia tabaci associated with four crop 
hosts in Ghana. Journal of the Ghana Science Association 6: 105 - 116. 
Omwega, C. O. and Overholt W. A. (1996). Genetic changes occurring during laboratory 
rearing of Cotesia flavipes Cameron (Hymenoptera: Braconidae) an imported 
parasitoid for the control of graminaceous stem borers in Africa. African Entomology 
4 : 2 3 1 - 2 3 7 . 
Otranto, D., Traversa, D. Guida, B., Tarasitano, E., Fiorente, P. and Stevens, J. R. 
(2003). Molecular characterisation of the mitochondrial cytochrome oxidase 1 gene 
122 
of Oestridae species causing myiasis. Medical and Veterinary Entomology 17: 307 — 
315. 
Phillips, C. B., Baird, D. B., Iline, I. L, McNeill, M. R., Proffitt, J. R. Goldson, S. L. and 
J. M. Kean, J. M. (2008). East meets west: adaptive evolution of an insect 
introduced for biological control. Journal of Applied Ecology 45: 948 - 956. 
Rees, D. P., Rivera, R. R. and Rodriguez, F. S. H. (1990). Observations on the ecology of 
Teretriosoma nigrescens (Lewis) (Col., Histeridae) and its prey Prostephanus 
truncatus (Horn) (Col., Bostrichidae) in the Yucatan peninsula, Mexico. Tropical 
Science 30: 153-165 . 
Roderick, G. K. and Navajas, M. (2003). Genes in new environments: Genetics and 
evolution in biological control. Nature Reviews - Genetics 4: 889-899. 
Rugienius, R., Siksnianas, T., Stanys, V., Gelvonanskiene, D. and Bendokas, V. (2006). 
Use of RAPDs and SCAR markers for identification of Strawberry genotypes 
carrying red stele (Phytopthora jragariae) resistance gene Rpfl. Agronomy Research 
4 : 3 3 5 - 3 3 9 . 
Saitou, N. and Nei, M. (1987). The neighbor-joining method: A new method for 
reconstructing phylogenetic trees. Molecular Biology and Evolution 4:406-425. 
Sambrook, J. Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning, A Laboratory 
Manual. 2ndedn. Cold Spring Harbour Laboratory Press, New York. 
Schneider, H., Borgemeister, C. Setamou, M., Affognon, H., Bell, A., Zweigert, M. E., 
Poehling, H. and Schulthess, F. (2004). Biological control of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) by its predator Teretrius 
nigrescens (Lewis) (Coleoptera: Histeridae) in Togo and Benin. Biological Control 
30 :241-255 . 
123 
Simon, C , Frati, F. Beckenbach, A., Crespi, B., Liu, H. and Flook, P. (1994). Evolution, 
weighting and phylogenetic utility of mitochondrial gene sequences and a 
compilation of conserved polymerase chain reaction primers. Annals of the 
Entomological Society of America 87: 651—701. 
Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007). MEGA4: Molecular Evolutionary 
Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution 
24:1596-1599. 
Tamura, K., Nei, M. and Kumar, S. (2004). Prospects for inferring very large phylogenies 
by using the neighbor-joining method. Proceedings of the National Academy of 
Sciences (USA) 101:11030-11035. 
Tigar, B. J., Osborne, P. E., Key, G. E., Flores-S., M. E. and Vazquez-Arista, M. (1994). 
Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) by 
its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bulletin of 
Entomological Research 84: 555 — 565. 
Vincze, T., Posfai, J. and Roberts, R.J. (2003). NEBCUTTER: a program to cleave DNA 
with restriction enzymes Nucleic Acids Research 31: 3688 — 3691. 
Welsh, J., Patterson, C. and McClelland, M. (1990). DNA polymorphisms amplified by 
arbitrary primers are useful as genetic markers. Nucleic Acids Research 19: 9213 -
 7218. 
White, T. J., Bruns, T., Lee, S. and Taylor, J. (1990). Amplification and direct sequencing 
of fungal ribosomal RNA genes for phylogenetics. In: Innes, M., Gelfand, D., 
Sninsky, J. and White, T. (eds) PCR Protocols: A Guide to Methods and 
Applications, pp. 315-322 Academic Press, San Diego, CA. 
124 
Williams, J. G.K., Kubelik, A. R., Livak, K. J., Rafalski, J. A. and Tingey, S. V. (1990). 
DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. 
Nucleic Acids Research 18: 6531 - 6535. 
Zayed, A., Constantin, S. A. and Packer, L. (2007). Successful biological invasion despite 
a severe genetic load. PLoS ONE2: e868. doi:10.1371/journal.pone.0000868. 
Zhang, D. -X. and Hewitt, G. M. (1997). Assessment of the universality of a set of 
conserved mitochondrial COI primers. Insect Molecular Biology 6: 143 150. 
Zhang, D.X. and Hewitt, G. M. (2003). Nuclear DNA analysis in genetic studies of 
populations: practice, problems and prospects. Molecular Ecology 12: 563 — 584. 
125 
C H A P T E R S I X 
Development of Microsatellite Markers for Teretrius nigrescens 
6.1 Abstract 
Microsatellites are the most commonly used neutral multilocus molecular markers in 
ecological genetics today. Several attributes make them useful in fine-scale variation studies: 
they are hypervariable, co dominant, reproducible and amenable to automation. 
Microsatellites were the best markers we could use in addressing the questions raised in the 
previous chapters which gene sequence data could not answer. However, since none had 
been developed for Histeridae, fresh markers had to be developed for this study. Twenty four 
novel microsatellite markers were isolated and characterised. Their population parameters 
were tested on 50 individuals from a field population from Honduras. Alleles per locus 
ranged between 2 and 12, and observed heterozygosity between 0.037 and 0.652. Six loci 
deviated significantly from Hardy-Weinberg equilibrium and showed evidence of null 
alleles. These markers will be useful for studies of the predator's population structure and 
characterizing populations for control of LGB. 
Published as: Omondi, A. B., Orantes, L. C, van den Berg, X Masiga, D. and Schulthess, F. (2009). 
Isolation and Characterization of microsatellite markers from Teretrius nigrescens Lewis (Coleoptera: 
Histeridae), predator of the storage pest Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) Molecular 
Ecology Resources 9: 1236 - 1239 
126 
6.2 Introduction 
Molecular markers are useful in determining the level of genetic differentiation, genetic and 
demographic history of animal populations. A combination of the molecular markers of 
appropriate discriminative power and ecological factors may be used to clarify varying levels 
and possible causes of population fragmentation, differentiation and specialization events. 
Microsatellites are among the most variable and popular marker systems in genetics. As 
opposed to unique DNA, microsatellite variation derives mainly from length variability 
rather than primary sequence variation, and the genetic variation is characterized by high 
heterozygosity and the presence of multiple alleles hence a high genetic information content. 
They are used in genetic research for diversity studies, genetic map development, linkage 
analyses, marker-assisted selection and fingerprinting studies (Ellegren, 2004). There are 
generally two types of microsatellites those linked to expressed sequences (type 1) and the 
anonymous ones (type II) associated with non-coding DNA. The EST linked markers are 
usually more conserved and are useful in selection, breeding and gene mapping studies. 
Anonymous microsatellites which, are often highly variable, are useful in fine scale 
population studies of recent genetic and demographic history of populations (Ellegren, 2004; 
Ng et ah, 2005; Vasemagi et al, 2005; Maneeruttanarungroj et al., 2006). 
The wide usage of microsatellite markers is based on their co-dominant nature, abundance in 
eukaryotic genomes, robustness, hypervariability, high information content, requirement for 
simple sample preparation and amenability to automation (Zane et al., 2002; Selkoe and 
Toonen, 2006). However, they also have some technical limitations such as the cost of 
isolation and de novo characterization and the number of markers required since one assay 
forms only one datum point per genotype (Masi et al, 2003). Practical problems also arise in 
127 
their analysis as many theoretical models fail to accurately explain allele frequency 
distributions in natural populations (Dakin and Avise, 2004; Ellegren, 2004). 
There are several approaches to isolating microsatellite markers depending on the 
availability of prior information the material and resources available. Several authors have 
reviewed these procedures, the preliminary evaluation and inclusion criteria for the use of 
microsatellites in ecological studies (Zane et al, 2002; Selkoe and Toonen, 2006). For 
widely sequenced species, sequences in genomic databases can be searched for microsatellite 
repeats and primers designed for these loci. Microsatellites from a closely related species 
may be used tested and used in a related species with considerable level of success in 
confamilial organisms (Selkoe and Toonen, 2006). However, the success of the above 
methods depends on the level of scientific interest in the related taxa. De novo isolation may 
be carried out by developing microsatellite libraries especially for species with little 
available sequence information. 
At the beginning of this work, only 924 Expressed Sequence Tagged (EST) sequences were 
available from the family Histeridae (Caterino and Vogler, 2002). Microsatellites were 
therefore developed de novo for the T. nigrescens to complement information obtained by 
use of RAPD-PCR, ITS and mtCOI sequence analysis. The objective of this study, therefore, 
was to develop a panel of novel ecologically useful microsatellite markers for from T. 
nigrescens to test their population genetic characteristics; to test their multiplexing 
possibilities and assess their potential usefulness in related species. 
128 
6.3 Materials and Methods 
6.3.1 Microsatellites-enriched library construction 
Three microsatellite libraries were developed following the enrichment capture protocol 
Glenn and Schable (2005). Genomic DNA was extracted from the heads and thoraxes of a 
pool of ten freshly killed T. nigrescens adults of both sexes from the Benin Laboratory 
population using the Phenol-Chloroform-isolated procedure (Sambrook et ah, 1989). About 
2-4 ug of total genomic DNA was digested to completion with Rsal, BstU or EcoRV and 
XmnI in (New England Biolabs) in 20 ul volumes. Double stranded linkers (SNX 24 
Forward: 5D GTTTAAGGCCTAGCTAGCAGAATC; SuperSNX24+4P Reverse: 5D 
GATTCTGCTAGCTAGGCCTTAAACAAAA) (Glenn and Schable, 2005) were ligated to 
both ends of DNA fragments in 30ul reaction volumes containing 7.0 ul dsSuperSNX-24 
linkers, 1.0 X ligase buffer and 2.0ul DNA ligase and 20 u.1 of the digested product. Linker-
 ligated DNA was amplified in 50 ul reaction volumes containing IX PCR Buffer (with 
1.5mM MgCl2); 200 uM dNTPs and 2 Units of Taq DNA polymerase (GenScript), 
approximately lOng ligated DNA and 200 nM of the Super SNX primer. Thermal cycling 
conditions were 4 min at 94 C, followed by 30 cycles of 94 °C for 25 s, 55 °C for 1 min, 72 
°C for 3 min and a final extension of 72 °C for 15 min. The Super SNX24 forward primer 
was used as the sole primer. 
The PCR products were hybridised to a combination of dinucleotide, trinucleotide and 
tetranucleotide probes in 50 ul mixtures containing IX hybridisation solution, 0.2 |_iM of 
each of the oligonucleotide probes, 10 JJ.1 linker ligated DNA from pervious step. The 
hybridisation programme denatured the DNA at 95 °C for 5 min then cooled to 70 °C - 50 
°C at 0.2 °C every 5 s (99 cycles); 50 °C for 10 m; 50 - 40 °C every at 0.5 °C every 5 s and a 
final hold at 10 °C. This complex was added to washed and resuspended streptavidin-coated 
129 
magnetic beads (Dynabead M-28, Invitrogen) and mixed by gentle rotation for 35 min. The 
mixture was washed twice each with 2X SSC, 0.1 % SDS solution and IX SSC, 0.1 % SDS 
at a maximum of 50 °C, each time, capturing the beads with a magnetic particle concentrator 
(Invitrogen). The enriched fragments were then eluted by denaturing 95 °C for five minutes 
in TE buffer (10 mM Tris pH 8, 0.2 mM EDTA) and precipitating in 2 volumes absolute 
ethanol and 0.1 volume NaOAc/EDTA at -20 °C overnight. The DNA was pelleted and 
washed with 70 % ethanol. 
The recovered DNA was double-enriched for microsatellite-containing fragments by nested 
PCR, in three 25 |LX1 reactions each containing IX PCR buffer (GenScript), 1.5 mM MgCl2 
200 uM of each dNTP 0.5 uM SuperSNX24 forward primer, 1U Taq polymerase 
(GenScript) and 2 ul of enriched genomic DNA fragments. Thermal cycling was done in 
Palmer-Cetus (TTC 100) thermocycler at 95 °C for 2 min, 25 cycles of 95 °C for 20s, 60s for 
20 s and 72 °C for 90 s and a final elongation at 72 °C for 30 min. Products from the three 
replicates were bulked for use as templates for the nested PCR. The subsequent PCR 
fragments were cleaned using MinElute PCR Purification Kit (QUIAGEN) for coning. 
The recovered SSR-DNA fragments was ligated into a pGEMT-Easy (Promega) cloning 
vector according to suppliers' recommendations and used to transform chemically competent 
E. coli (DH5a strain) cells. Transformed white colonies (n = 500) were selected visually 
from Luria-Bertani/Agar/ampicillin plates with X-Gal and IPTG and screened for the insert 
by PCR using Super SNX forward primer alone as forward and reverse primer as described 
in Chapter 2. Positive colonies were therefore those that showed a single band. Two hundred 
cloned fragments of over 400 bp were sequenced under BigDye terminator cycle sequencing 
chemistry using M13F-pUC(-40) forward and reverse primers (Macrogen Inc, Korea) 
130 
(Chapter 2). Sequences were cleaned to remove vector sequences using Vec Screen and by 
manual alignment and editing. 
6.3.2 Identification of tandem repeats 
Microsattellite repeat regions were identified using Tandem Repeats Finder Server (Benson, 
1999). This programme allows the finding of SSR repeats without a priori specification of 
the repeat motif or size. It uses statistically-based recognition criteria to identify 
microsatellites, modelling them according to the percentage identity and frequency of indels. 
MlCROFAMtLY software (Meglecz et al, 2007) was used to screen sequences for flanking 
sequence similarity and to eliminate redundant and repeated members of microsatellite 
families (based on the degree of flanking sequence similarity). 
6.3.3 Primer design and optimisation 
Flanking sequences were manually demarcated for primer design using an online version of 
PRIMER 3 software Rozen and Skaletsky (1998). For each locus, forward and reverse 
primers were designed to anneal at least 20 bp away from the microsatellite core, have at 
least 30 % GC content and an annealing temperature of about 50 °C. The difference in the 
annealing temperatures of the forward and reverse primers was set to be within 2 °C of each 
other. Amplification product sizes were targeted for between 130 and 450 bp to enable allele 
sizing using a single internal size standard. 
Up to six primer sets were selected depending on the characteristics of the flanking 
sequence, yielding a total of 121 primer combinations (MWG Biotech, Germany). All 
possible combinations forward and reverse primer pairs were tested for each locus. The 
initial optimisation reactions contained lx Genscript Taq polymerase buffer with 1.5 mM 
131 
MgCl2, 400 uM dNTPs, 150 nM Primers, 1U Taq polymerase and ~ 20ng template DNA. 
The annealing temperature was initially set at 5 °C below the lower melting temperature 
(Tm) of the primer pair. Where there was no amplification, the annealing temperature was 
lowered to a maximum of 10 °C below the Primer melting temperature. The highest possible 
annealing temperature (Ta) at which a definite reproducible amplification product of around 
the correct predicted size was adopted as the working Ta for that primer set. Primers were 
screened on two insects from both sexes per population to identify those producing clear 
reproducible polymorphic amplification product. PCR products were separated in 2 % 
agarose (in TAE) gels at 2.6 V/cm for 2 hours. 
Twenty-seven primer sets that gave amplifications in at least one individual from each 
population during the pre-screening were selected. For each locus, only one primer set was 
chosen. Fluorescent dye-labelled primer pairs were designed for these loci for automated 
multiple genotyping using ABI sequencer (ABI 3130) (Missiaggia and Grattapaglia, 2006). 
For each primer set, the forward primer was labelled using one fluorescent dye (PET, NED, 
6-FAM or VIC) while the reverse primer was not labelled. The PCR products from each co-
 loading set were diluted and mixed according to the optimum predetermined signal strength. 
Each analysis well was loaded with 10 ul of sample. Each sample was a mixture of 9 ul of 
Hi-Di Formamide, (with 0.15 ul of LIZ 500 internal size standard) and 1 \A of the mix of the 
diluted PCR products. Each co loading set comprised of four PCR products each labelled 
with a different fluorescent dye. AUele sizing was done in an ABI 3130 or 3730 instrument 
(Applied BioSystems). Polymorphism and population genetics characteristics were assessed 
on 50 individuals, randomly selected from wild caught adults, sampled using funnel traps 
baited with LGB pheromone mimic (Pherocon Technologies) and heat sterilized LGB frass 
and maize grains. Samples had been collected from Honduras between January and July 
132 
2008 and sent to our laboratory alive. This geographic population represented the most 
recent field recovery. 
6.3.4 Allele calling 
The sizes of alleles were estimated from electrophoregrams using GENEMAPPER 4.0 software 
(ABI Biosystems), using LIZ 500 internal standard profile to estimate the allele sizes. The 
allele detection threshold was set at 200 and 100 for homozygotes and heterozygotes 
respectively. Peaks were manually edited to confirm correct allele calls. All products that 
showed equivocal, missing or more than two alleles per sample were assumed unsuccessful 
and repeated (Figure 6.1). Reactions that did not amplify after two attempts were reamplified 
with the other sets of earlier primers designed for the locus and the mtCOI primers. Those 
that amplified were considered null or non-amplifying alleles. A subset of successfully 
genotyped individuals was re-analysed to test the genotyping error rate of each marker and as 
positive controls in repeat reactions. 
6.3.5 Cross species amplification 
Cross species amplification was tested on six histerid species from varying 18S sequence-
 based relative phylogenetic distance from T. nigrescens (Caterino and Vogler, 2002) 
(Chapter 2). One to six individuals from the species T. latebricola (= T. americanus), 
Chatabraeus spp. and Acritus nigricornis (Abraeinae); Carcinops pumilio (Dendrophilinae); 
Saprinus bicoloroides (Saprininae) and Hister zulu (Histerinae). The PCR was carried out 
under the same conditions as those for T. nigrescens with two positive controls. PCR 
products- were analysed by direct allele sizing and scored as described for T. nigrescens 
above. 
133 
6.3.6 Multiplexing Microsatellite primers 
Multiplexing is the amplification of different fragments on a single template, each with its 
own set of primers, in the same reaction. Products are then separated by size or other 
attribute on a suitable matrix. Multiplexing microsatellite fragment analysis based on 
different fluorescent labels reduces the cost and time of the process. However, it involves a 
potentially erroneous step of coloading set mixing and dilution before analysis. PCR 
multiplexing reduces the cost of consumables, time and number of reactions necessary and 
eliminates the need for mixing step. Theoretically, the main considerations of the 
multiplexing process are primer annealing temperature and dye label. The practical 
challenges include possibilities of product competition, higher risk of large allele drop out 
(van Oosterhout et al., 2004), or other causes of failure of generation of specific products. 
These can lead to problems from false null alleles to inaccurate genotyping and PCR failure. 
Therefore PCR multiplexing must be optimised for their ability at correct genotyping 
abilities before use. 
Seven possible multiplex sets each comprising three or four differently labeled primers of 
nearly similar annealing temperatures were tested. Primers quantities were initially set 
depending on the annealing temperature relative to the others, primer relative amplification 
and genotyping strength and colour dye (table 3). The PCR assays were carried out as above 
in 20 ul reaction mixes with 400 uM dNTPs, 2 U Taq polymerase and 2 ul template DNA. 
Amplification with standard Taq polymerase (Genscript) and Hot Start (Amplitaq Gold) 
polymerases were compared. Touchdown PCR was eventually with Annealing Temperature 
(Ta) decreasing at 0.5 °C per cycle (from highest Ta + 1 degree to lowest Ta from earlier 
optimization reactions). Products were loaded undiluted (0.5 ul + 10 ul) of Hi-Di formamide 
containing 0.12 ul of LIZ 500 internal size standard. 
134 
6.3.7 Data Analysis 
Allele frequencies and deviations from HWE were calculated with Bonferroni corrections 
using CERVUS 3.0 (Kalinowski et at, 2007). The occurrence of null alleles for each marker 
was estimated with MlCROCHECKER at 95 % level of significance with 1000 iterations (van 
Oosterhout et at, 2004). Linkage disequilibrium was tested using GENEPOP version 3.4 
software (Raymond and Rousset, 1995; 2004), under both the Fischer's and Markov Chain 
method (10000 dememorization steps, 100 batches and 1000 iterations per batch). For this 
test, only samples collected from Tegucigalpa, Honduras, were included to avoid the 
possible effect of the genetic structure and covariation of alleles between populations. 
Marker reliability was assessed by re-amplification of 10 randomly selected individuals 
twice and comparing the results. 
6.4 Results 
Of the 200 fragments sequenced, 64 fragments with micro satellite repeats were screened for 
flanking region similarity using MlCROFAMILY (Meglecz et at, 2007) (Table 6.1). About 
three primer pairs per locus were designed using PRIMER 3 (Rozen and Skaletsky, 1998) for 
the resulting 39 unique fragments. Twenty-seven primer pairs consistently amplifying loci 
from both sexes and six test populations were selected for polymorphism assessment. The 
loci TnM58, TnN6 and TriN13 gave inconsistent amplification while TnM50 
([GGTAGAA]?) produced uncorrectable stutter peaks on while the markers was clearly 
polymorphic on agarose gel analysis (Figure 6.2). General amplification failure was 
observed with TnM541 and TnN12. The number of alleles per locus ranged between two and 
12 (average = 5.75) (Table 6.4). Seven loci deviated significantly from Hardy Weinberg 
Equilibrium (Table 6.4). There was also evidence of presence of alleles in all these loci and 
135 
in TnM541 and TnN21 (Table 6.4). Three groups of markers showed significant linkage 
disequilibrium: TnM36-TnM70-TnM53; TnM79-TnM85 and TnM41-Tnm71. A BLAST 
query of available sequences in the public, genetic database did not reveal any significantly 
similar sequences. 
Cross amplification was successful for most samples and most loci (Figure 6.3, Table 6.5). 
About 80 % of all the primers selected for T. nigrescens amplified the other Abraeinae 
specimens and 70 % were polymorphic. Histerinae (H. zulu) and Dendrophilinae (C. 
pumilio) gave the poorest amplification and polymorphism under the conditions used for T. 
nigrescens. 
6.4.1 Multiplex sets 
Multiplexing sets B, C, D, E and F produced correct genotyping reactions for the four 
multiplexing test samples with both standard (GenScript) and Hot Start polymerases 
(Amplitaq Gold; Applied Biosystems), after a single trial (Figure 6.4). The reactions that 
failed in these multiplexing sets had also failed in the simplex reactions. Multiplex set 'A' 
was the least successful, with only half the reactions being unequivocally scorable. 
136 
Table 6.1: Marker attrition rate in the development of T. nigrescens microsatellites 
Parameter Number %of last % of original 
Number of positive clones 
Clones with >400bp (sequenced) 
Sequences with SSRs 
Non-redundant SSR sequences 
Amplified primer sets (each population) 
Amplified loci 
Primers selected for genotyping 
Polymorphic loci 
324 - 100.00 
200 61.73 61.73 
64 32.00 19.75 
39 60.94 12.04 
66 51.97* -
 33 50 20.37 
27 81.82 8.33 
24 88.89 7.41 
* Percentage of the 127 primer sets developed. For each locus, at between two and six primer 
sets targeting regions of different sizes were designed and tested. 
Table 6.2: Frequency of various types of repeats in sequenced motifs 
Repeat Type Isolated Amplifying Remarks 
Mono- 11 -
 Di- 18 9 
Tri- 48 17 
Tetra- 3 1 
Penta- 1 0 
Hexa- 4 0 
Other 1 (septa-) 1 
Compound 26 6 
Associated with other repeats 
Most very short or not amplifying 
Good candidates 
Long (CATA)n candidate not 
amplifying 
Not amplified 
Some resolved to trinucleotide repeats 
Variable on gel, many stutter peaks 
One and >1 repeats together . 
137 
ISO KM 
A,A2 
2 M ?ac 
B2B? 
1» I« 
190 100 210 
A1A3 B1B3 
JL 
ISO 20c 220 230 2 » 
A2A2 B1B2 
k !\ / I 
t» WC 
Figure 6.1: Alieles scored by their size in base pairs using GENEMAPPER. The expected 
lengths are marked in for automatic correction of sizes of all alieles falling within this bin 
range. Three individuals are scored on two loci and three different alieles. 'x ' represents pull 
up or background noises visible during genotyping. 
138 
Table 6.3: PCR conditions for multiplex sets of primers. Ta refers to the conditions used in 
the final reactions 
Set Name Dye Ta Primer Qty Reaction Ta Success rate 
A TnNl 6-FAM 48 250 1/4; 2/4 
TnM70 
TnM47 
VIC 
NED 
49 
50 
250 
350 
53-- 4 8 inconsistent 
Non specific 
TnM48 PET 51 350 4/4; 4/4 
B TnM50 VIC 50 375 Non specific 
TnM36 
TnM58 
6-FAM 
PET 
51 
52 
375 
500 
55 -50 4/4; 4/4 
No amplification 
TnM71 NED 53 500 2/4; 3/4 
C TnM40 VIC 52 375 4/4; 4/4 
TnM66 
TnN28 
6-FAM 
NED 
51 
53 
375 
500 
55 -50 4/4; 4/4 4/4; 4/4 
TnN2a PET 54 500 4/4; 4/4 
D TnM54 PET 53 500 4/4; 4/4 
TnM85 
TnN12 
VIC 
NED 
53 
54 
500 
500 
56 -52 3/4; 3/4 
Non specific 
TnM79 6-FAM 53 250 4/4; 3/4 
E TnN21 VIC 53 350 3/4; 3/4 
TnNl 3 
TnM23 
PET 
NED 
53 
54 
500 
500 
56--52 Non specific 
3/4; 3/4 
TnM55 6-FAM 53 350 3/4; 3/4 
F TnM541 NED 55 500 2/4; 2/4 
TnM72 
TnM53 
PET 
VIC 
55 
56 
500 
400 
5 8 - 52.5 
4/4; 4/4 
3/4; 2/4 
TnM51 6-FAM 53 350 2/4; not specific 
G TnM41 NED 56 250 2/2; 2/2 
TnN6 PET 54 350 58--51 No amplification 
TnN14 6-FAM 51 500 2/2; 1/2 
For each amplification set, the fraction represents success rate (for production of the correct 
genotype as simplex reactions) when attempted on the four samples of known genotypes. 
The first fraction represents GenScript Taq polymerase and the second Amplitaq Gold 
master mix. 
139 
Table 6.4: Characteristics of 24 microsatellite loci and primers for T. nigrescent tested on 50 LGB pheromone trap-sampled adults from Honduras. 
Locus/ Repeat motif 
Accession No. Primer sequence 5D - 3 D 
Ta 
(°C) 
Size range 
(bp) 
N % Ho HE 
140 -169 38 5 0.053 0.366f* 
211-220 42 5 0.286 0.546f* 
172-184 39 6 0.118 0.168 
239 - 252 39 6 0.419 0.400 
208-211 47 5 0.511 0.628 
167-217 36 5 0.417 0.686f* 
183-193 47 4 0.128 0.160 
193-239 46 10 0.652 0.683 
166-181 50 4 0.600 0.568 
163-166 45 3 0.133 0.205 
206 - 217 39 2 0.065 0.104 
194-214 45 8 0.622 0.652 
TnM47 
EU826153 
(CAT)7 F: R: 
TnM48 
EU826156 
(ATT)3(ATG)6GGG(ATG),0 F: R: 
TnM55 
EU826152 
(CCG)5-(ATG)6 F: R: 
TnM70 
EU826157 
CTGA)16(AAC)10 F: R: 
TnM71 
EU826158 
(ACT)17 F: R: 
TnM53 
EU826151 
(CATTj4..CTG)17 F: R: 
TnM79 
EU826160 
(TGA)17 F: 
R: 
TnM85 
EU826161 
(TACA)9 F: R: 
ToM36 
EU826154 
(GAT)20 F: R: 
TnM51 
EU826155 
(GCT)7 F: R: 
TnM54 
EU826149 
(TGA)6 F: R: 
TnM72 
EU826159 
(CAA)22(CAG)2(CAA)29 F: R: 
NED-AAATCGCTTTTCTTCTTGCAT 
ATGCACGTGATTTTGCATTC 
PET- CCCTTTGACGGCATTGTTTA 
TGCGTTCTTGAAATGCAAAC 
6-FAM-GGCCCCAACTTTCCATTATT 
CTTTACAGGTCCCCGCTTTC 
VIC- GAAGAATTCTACTAATTTATATGTG 
NED-CACGCTGGGTGGTGAATAG 
TTCTTTCCCGAAAGCTCAAC 
VIC- GCCGAGCGCATTAGATACAT 
ACAGCTGATGTTCTTCCAACG 
6-FAM-CTCGATTGCTCATGTGTTCG 
TTAGAATGTGGCGGGATTGT 
VIC- GTTTTGCGGAAAGACCGTAG 
TGAGCCGACCAACTTAATGA 
6-FAM-ACATGTTAAGGATCCGAAAACC 
CGATGCTTTTTCATTTGAGC 
6-FAM-GCAACAAGATTCCGAACAATC 
GTTTCCGTTTGAGTGCGTTT 
PET- CCAGCTGAAATAGGCACAAA 
CGGCACAATATATTCCAAAGAA 
PET- AGGTCCGACCGTTGGTAAC 
GAACGCACCACCGATACTTT 
50 A 
51A 
53E 
49 A 
53B 
56 F 
53 D 
53 D 
51B 
53F 
53 D 
54 F 
Table 6.4 continued 
140 
Locus/ 
Accession No. 
Repeat motif Piimet sequence 5D - 3 D 
Ta 
(°C) 
Size range 
(bp) 
N nA Ho H„ 
TnN21 
EU826167 
CTCA)40 
TnN28 
EU826168 
(GAT), 
TnM541 
EU826162 
(CA)6(GAT)6 
TnN12 
EU826165 
(TGA)17 
TnN14 
EU826166 
(cn)u 
TnN2 
EU826164 
(TTG)60 
TnM23 
FJ531687 
(CA)16(CT)12 
TnNl 
FJ531692 
CirG)34 
TnM40 
FJS31688 
(CA)13 
TnM41 
FJ531693 
(GAT)10 
TnM58 
FJ531689 
(GT)„ 
TnM66 
FJ531690 
(GA)14 
F: VIC- CACAAAGTCAATTTCGACGTG 
R: AACCACCATTTGACCTGGAA 
F: NED-TAGCGCAGAAGGACAAATGA 
R: CCCTGGGTACATGCAAAAAT 
F: NED-GATTTCCATCGTCCTGGCTA 
R: CTGACCTGTGTTCGTTTTGC 
NED-GGAGAAAAACTGCGAAGTTACG 
R: TCCAATTAGAAAGTGGCTGGA 
F: 6-FAM-AAGACGCCGTTTTCTTGAAT 
R: TAAATTTCCGGTTGCCTTCC 
F: PET-CTAGGCTTTTGTTGTTGTTGC 
R: AGGAACACTCAAGAACATTCG 
F: NED-CGGTCTCTTAAGGACAGAAGTTG 
R: GAGGGAAAATGAATTGACAAGG 
F: 6FAM-AATTCTAGAGCAGTAGGC 
R: TTCCAACAGACGTTCCAACA 
F: YIC-CTGCTACGAAACTGGTGCTG 
R: GCCAAAGACGAGCCATAAAT 
F: NED-AAGTAACGAGGGCGAGAAGAC 
R: GTAGAATCCGCACCCAGAAG 
F: PET-CGCAATGGTGTTCCTGTTAG 
R: AAAAACCCGAGAGCCAATTC 
F: 6FAM- GGAAGGTTCAAGCTGTCCAA 
R: ATGCCTTCCCGTACGATTT 
53h 209--228 50 5 0.320 0.397f 
53c 175--183 37 3 0.060 0.169f 
54" 163--186 27 2 0.037 0.037f 
54D 229--237 19 4 0.316 0.585-f* 
51G 133--146 48 5 0.146 0.196 
53c 216--234 48 4 0.375 0.434 
54E 140--166 48 12 0.500 0.811-1* 
48A 253--272 48 4 0.396 0.484 
52c 145--156 50 6 0.5.80 0.686 
55F 116 -131 48 6 0.646 0.630 
52B 116--131 34 5 0.529 0.652 
51c 141--179 46 11 0.196 0.491f* 
Interrupted repeat motifs are denoted by (...). 
All forward primers were direcuy labelled on the 5D end, reverse primers were not labelled. 
-f- markers showing existence of null alleles, * loci showing significant deviation from Hardy Weinberg Equilibrium. 
Ta: annealing temperature; multiplex sets tested shown byA'B,c'D,EiF. 
N: number of individuals successfully genotyped, nA: number of alleles detected. 
141 
Table 6.5: Cross-species amplification of 19 microsatellite primers isolated from Teretrius nigi'escens on seven species in four subfamilies within 
the family Histeridae. 
Locus Teretrius Acritus Abraeus sp Chatabraeus sp Saprinus Carcinops Hister zulu 
americanus nigiicomis bicoloroides pumilio 
TnM23 + 
148-159 
+ 
144- 161 
+ 
161-175 ' ' 
TnM40 + 
141 -178 
+ 
144 
+ 
144 
- + 
154 
- -
 TnNl + 
263 - 272 
+ 
263 
+ 
235-263 
- - - -
 TnM66 + 
146-150 
+ 
143 
+ 
243 
+ 
262 
- - -
 TnM36 + + + + + + + 
175-181 174-?178 178-181 178- 181 .175 -178 178-181 175 -178 
TnM47 + + + - + + + 
162-226 162--226 162-226 162 162-234 162 
TnM48 + + + + + + + 
211-220 211--214 214-220 220 211-220 211-214 214-220 
TnM55 + + + + + - -
 175 -179 184 184 184 184 
TnM70 + 
249 
+ 
254 
+ 
238-254 
- + 
249 - 254 
- -
 TnM71 + 
195-208 
+ 
208 
+ 
208 
+ 
208 
- - -
142 
Locus Teretrius Acritus- Abraeus sp Chatabraeiis sp Saprimis Carcinops Ulster zulu 
americamis nigiicomis bicoloroides pwnilio 
TnM51 + + + + - + -
 160-163 163 163 163 163 
TiiM53 + + + + + + -
 202 - 205 202-?205 202 202 202 202 
TnM54 + + + + - + -
 195-216 203-•216 203--216 214-217 216 
TnM541 + 
187 
+ 
187 
+ 
187 
+ 
164-217 
- - -
 TnM79 + + + + - - -
 137-146 176-?192 176--198 176-198 
TnM85 + 
232 - 237 
+ 
232 
+ 
232 
+ 
232 
- - -
 TnN12 + 
232 - 237 
+ 
233 
+ 
233 
+ 
233 
- - -
 TnN2 - + 
226 
+ 
222--226 
+ 
226 
- - -
 TnN21 + + + + - - + 
222 - 238 222 221 222 212 
TnN28 + + + + + - -
 170 170 • -222 170--177 170-177 177 
Successful amplification of at least one sample (+), non-specific oi" failed amplification (-); size tange of the amplified products (from automated generic 
analyzer) is given. 
143 
Figure 6.2: PCR products of TnM50 marker on 2 % agarose gel populations. Accessions: M 
(100 bp DNA size standard); 1-2 (KARI); 3,4 (Benin); 5-6 (Batan); 7-8 (Oaxaca); 9-10 (NRI) 
11-12 (Tlaltizaspan) and 13 (-ve control -water). 
Onthophilinse 
Anapleini 
Saprininae 
Niponiinae 
Paromalini 
Bacaniini 
Abraeinae 
?? Chlamydopsinae 
^ Dendrophilini 
Tribaltnae 
' Histerinae 
Dsndrophilinae 
Saprininae 
/"braeinae 
Teretrius 
T.nigrescens 
Phylogenetic relationships of 
Histerid subfamilies 
20 40 60 80 
Amplification success 
Figure 6.3: Comparison between the phylogenetic relationship of Histeridae subfamilies sensu 
Caterino and Vogler (2000) (left) and percentage cross amplification success of 27 unique 
microsatellites isolated (right). 
144 
(DO 111 120 !« 1JO ISO 170 iso 200 m m 
. L ~ ' ? - * ' - ? ? 
jOI.MtQXin (fr<_f BiTSuidAT 
110 120 130 140 150 ISO 170 1 190 200 210 220 230 2 « 250 250 270 280 
_gxi * 
100 110 
Epi.li 
120 130 
tOlmi a 
ISO 
1 !? 
170 180 180 200 210 220 230 240 250 260 270 140 150 280 
vex 
20000 
IKW 
120O0 
BOOO 
4000 
ft JL A t . 1A . I 1 1 
Figure 6.4: Allele sizing profiles from multiplex analysis of amplicons from genotyping 
reactions. Profiles show a: size standard (orange colour peaks) and products labelled with the 
four fluorescent dyes. Relative signal strength is shown on the height (x — axis) while y-axis 
shows the fragment size. 
145 
41133 
sz 133.07 
4IH8 
sz 148.40 
41156 
sz 155.72 
41165 
st 164.11 «r? ET ti? sz204.86Eill.6!»a:218.68 
41133 all40 all48 41156 ill62 41170 
sz 132,94 a 14036 | g 148.24 Isz 155,55^z 162.54 g 169.81 
41133 41140 41148 all56 U.162 41170 
sz 133.08 si 140.44 si 14835 si 155 JS)z 16238 sz 169,65 
Figure 6.5: Stutter bands associated with TnM 50 from a random sample of three genotypes. A 
and C are perhaps heterozygotes while B is most likely a homozygote as it displays a clear 
reduction in stutter peak height away from the main peak. '+ ' denotes the probable right allele. 
A is also a potential result of amplification of two alleles from the same family from different 
locations in the genome (Meglecz et al., 2004). 
146 
6.5 Discussion 
Dinucleotide repeats are the most abundant and polymorphic followed by mono and 
tetranucleotide repeats, while trinucleotide repeats are the least dominant (if a high minimum 
size threshold is considered) (Fornage et al, 1992; Ellegren, 2004). In fact several studies 
using a mixture of probes have ended up either solely or mostly with dinucleotide repeats 
(Theissinger et al., 2008). Since they are also the most variable it is also common that 
microsatellite isolation studies focus solely on these motif types (Khamis et al., 2008; Vinson 
et al., 2008). In this study, trinucleotide repeats were the most abundant accounting for (62 %) 
of all SSRs isolated and (71 %) of eventual SSRs genotyped. It is not clear why this occurred. 
For a given repeat length, dinucleotide repeats would be 33 % shorter than the trinucleotide 
repeats, hence would elute at a lower temperature during the washing steps of hybridisation 
capture. A second library construction using dinucleotide probes alone yielded only an 
additional 3 microsatellites in the final genotyping. As expected, dinucleotide and 
tetranucleotide repeats were generally more variable (average 9.7 alleles per locus) compared 
to trinucleotide repeats (4.94 alleles per locus). The marker attrition rate was higher among 
dinucleotide repeats where 66 % of those isolated did not eventually meet the utility criteria 
(Selkoe and Toonen, 2006). Although 87 % of them were unique, most failed to give 
reproducible amplifications, were not in HWE or showed significant evidence of null alleles. 
All these observations were possibly as a result of hypervariability of both the core repeat 
motifs and flanking sequences of these markers, which made universal primer annealing sites 
difficult to find. 
Accurate microsatellite scoring is often complicated by the presence of stutter bands. These 
bands introduce possible error in the genotyping of populations if not accommodated (Dakin 
and Avise, 2004). Stutters are inaccurate peaks caused by the introduction of errors during 
147 
PCR, due to polymerase slippage and the amplification of the incorrect copies amplifying 
products that are multiples of repeat motif longer or shorter than the template allele (Lai and 
Sun, 2004). Their occurrence in earlier cycles leads to accumulation that masks the correct 
allele sizes, making it difficult especially to differentiate homozygotes from heterozygotes. 
Such errors are more likely in mononucleotide and dinucleotide repeats, but are also found in 
other peaks (Walsh et al, 1996). Clarke et al. (2001) found high error rates in TA and CA 
repeats longer than 18 repeat units, with both Pfu and Taq polymerases. Several microsatellite 
development assays drop markers showing significant levels of stutter repeats (Selkoe and 
Toonen, 2006). However, stutter repeats can be corrected for clearly scorable peaks by allele 
scanning programmes if known, this problem is perhaps not quite serious. In this study, one 
marker (TnM50 [GGTAGAA]7) showed great variability on gel but abundance of stutters that 
differed by 7 bp (Figure 6.5). It was impossible to score this marker meaningfully, partly due 
to the high stutter ratio and absence of a provision for a heptanucleotide repeat in the 
microsatellite repeat scoring and scanning programmes available. 
Multiplexing was done at the allele sizing level (using colour dyes). Multiplexing enables the 
simultaneous analysis of several markers in a single reaction hence reduces the cost of 
analysis and increases genotyping efficiency. This also reduces the ardour of PCR analysis, 
promoting automation and reducing potential error and contamination. PCR multiplexing, 
amplifying several markers in a single reaction further enables simultaneous analysis of 
several markers. However, PCR multiplexing requires careful optimisation to determine which 
samples can be run together to ensure comparable ranges of signal strength and reduce allele 
drop out or amplification failure due to preferential amplification of certain fragments (Masi et 
al, 2003). Some loci earlier initially failed to amplify multiplex sets when primers were used 
in equimolar concentrations, especially with rather difficult loci. Adjusting the relative primer 
148 
concentrations relative to their signal strength in simplex reactions and observed peak 
dominance (Table 6.5).The multiplex analysis used here can be expanded to include primers 
with similar colour labels and non-overlapping size ranges once they can be convincingly 
established in various populations. While multiplexing provides a cost effective option to 
microsatellite assays, its use with these markers must be preceded by careful optimisation for 
template quality, polymerase, primers and laboratory conditions, since it is not clear they 
might be easily repeatable across laboratories. 
Eight loci deviated significantly from Hardy Weinberg Equilibrium in the wild Honduras 
population used for maker characterization (Table 6.4). This could have been caused by the 
Wahlund effect or existence of genetic substructure given that the Honduras population was 
sampled three times across six months period and a geographical range of about 400 km and > 
1500 altitude difference. There was also evidence of null alleles in all these loci and in 
TnM541 and TnN21. Other than mutations in the microsatellite annealing sites, the markers 
showing null alleles could also have been associated with coding DNA (within or physically 
linked to exons) which if under selection favouring one variant, could lead to the 
disproportionate survival of individuals bearing this. This and the distribution of null alleles 
may explain why some loci alone were in significant deviation from HWE. 
Three groups of markers showed significant gametic disequilibrium (p < 0.01): TnM36-
 TnM70-TnM53; TnM49-TnM85 and TnM41-TnM71. In the MlCROFAMTLY analysis, that 
compared by sequence alignment each set of flanking sequence to a database containing all 
the microsatellite sequences identified, all these primers were listed as unique. Co-variation 
between these loci could have been caused by the physical proximity of these loci on the 
149 
chromosome allowing joint inheritance of the alleles. Only a single member of each group 
would thus be useful for population analysis if neutral markers are needed. 
Cross species amplification is possible because of the highly conserved nature of 
microsatellite flanking sequences enabling their use in studies of related species, often to the 
family level (Ottowell et al, 2005; Selkoe and Toonen, 2006). However, the level of 
polymorphism and success of amplification declines with an increase in the phylogenetic 
distance between the source and test species (Primmer et al, 1996; Neff and Gross, 2001; 
Wright, 2004). Most of the primers in this study isolated were able to amplify loci of other 
species in the family Histeridae confirming the observations in plants and vertebrate taxa 
(Peakall et al., 1998). The success rate declined by the 28S r-DNA sequence based 
phylogenetic distance between the test species and T. nigrescens, based on the phylogenetic 
distance between histerid subfamilies (Caterino and Vogler, 2002). Curiously, TnN2 failed to 
amplify in the congeneric T. americanus, yet amplified in the other Abraeinae and distantly 
related Histerini. Although some decline in allelic diversity was observed in this study, 
statistical ascertainment of this bias was not done owing to the dearth of individuals of related 
species. Such a study would need genotyping at least 20 well-sampled individuals of at least 
one wild population of these species, which was beyond the scope of this study. The success 
rate observed with these loci however shows that these primers good candidates in the similar 
analyses and microsatellites marker development in the family Histeridae, especially 
Abraeinae, Saprininae and related subfamilies. The success rate and reliability of these 
primers may be increased by further optimization, relaxing annealing stringency and 
increasing DNA quantity in the PCR mix, which we did not do. Being the first set of primers 
developed in this family of beetles and due to the paucity of histerid sequences in genetic 
databases, these markers will be useful in genetic studies in this family. 
150 
The criteria for identifying and testing micro satellites to be used in ecological studies (Selkoe 
and Toonen, 2006) and techniques of overcoming some practical problems with microsatellite 
analyses (Dakin and Avise, 2004) improve the accuracy of data generated from this technique. 
Hardy Weinberg Equilibrium, Mendelian inheritance, repeatability and scorability have 
already been tested in the 24 loci isolated in this study. These will be used for further testing 
and evaluation of these loci for use in the population studies of T. nigrescens. 
This study successfully developed and characterisaed the first set of microsatellite markers for 
Teretrius nigrescens. To our knowledge, this is also the first set of SSR markers developed in 
the family Histeridae. The markers cover a range of repeat types, and levels of variability, and 
should find utility in ecological and genetic studies of this and related species. The next 
chapter describes the use of these tools in the studies of genetic structure and demographic 
history of 13 geographical populations of T. nigrescens. 
151 
6.6 References 
Benson, G. (1999). Tandem Repeats Finder: a program to analyze DNA sequences. Nucleic 
Acids Research 27: 573-580. 
Caterino, M. S. and Vogler, A. P. (2002). The phylogeny of Histeroidea (Coleoptera: 
Staphyliniformia). Cladistics 18: 394-415 . 
Clarke, L. A, Rebelo, C. S., Goncalves X, Boavida, M. G. and Xordan, P. (2001). PCR 
amplification introduces errors into mononucleotide and dinucleotide repeat 
consequences. Journal of Clinical Pathology: Molecular Pathology 54: 351 - 353. 
Dakin, E. E. and Avise, X C. (2004). Micro satellite null alleles in parentage analysis. 
Heredity 93: 504 - 509. 
Ellegren, H. (2004). Microsatellites: Simple sequences with complex evolution Nature 
Reviews — Genetics 5: 435 - 445. 
Fornage, M., Chan, L., Siest, G. and Boerwinkle, E. (1992). AUele frequency distribution of 
the (TG)n and (AG)n microsatellite in the apolipoprotein C-II gene. Genomics 12: 63 -
 68. 
Glenn, T. C. and Schable, N. A. (2005). Isolating microsatellite DNA loci. Methods in 
Enzymology 395: 202 - 222. 
Kalinowski, S. T., Taper, M. L-, Marshall, T. C. (2007). Revising how the computer 
program CERVUS accommodates genotyping error increases success in paternity 
assignment. Molecular Ecology 16: 1099 - 1006. 
Khamis, F., Karam, N. C , Guglielmino, R., Ekesi, S-, Masiga, D., de Meyer M., Kenya, 
E. U. and Malacrida, A. R. (2008). Isolation and characterization of microsatellite 
markers in the newly discovered invasive fruit fly pest in Africa, Bactrocera invadens 
(Diptera: Tephritidae). Molecular Ecology Resources 8: 1509-1511. 
152 
Lai, Y. and Sun, F. (2004). Sampling distribution for microsatellites amplified by PCR: mean 
field approximation and its applications to genotyping Journal of Theoretical Biology 
228: 185-194. 
Maneeruttanarungroj, C , Pongsomboon, S., Wuthisuthimethavee, S., Klinbunga S. 
Wilson, K. J., Swan, X, Li, Y., Whan, V., Chu, K-H., Li, C. P., Tong, X, Glenn, K., 
Rothschild, M., Jerry, D. and Tassanakajon, A. (2006). Development of 
polymorphic expressed sequence tag-derived microsatellites for the extension of the 
genetic linkage map of the black tiger shrimp (Penaeus monodori). Animal Genetics 
37: 363-8. 
Masi, P., Zeuli, P. L. S. and Donini, P. (2003). Development and analysis of multiplex 
microsatellite markers sets in common bean (Phaseolus vulgaris L.) Molecular 
Breeding 11: 303 - 3 1 3 . 
Meglecz, E. (2007). MlCROFAMTLY (version 1): a computer program for detecting fianking-
 region similarities among different microsatellite loci. Molecular Ecology Notes 7: 18-
 20. 
Meglecz, E., Petenian, F., Dan chin, E., D'Acier, A. C , Rasplus X -Y. and Faure, E. 
' (2004). High similarity between flanking regions of different microsatellites detected 
within each of two species of Lepidoptera: Parnassius apollo and Euphydryas auriana. 
Molecular Ecology 13: 1693 - 1700. 
Missiaggia, A. and Grattapaglia, D. (2006). Plant microsatellite genotyping with 4-color 
fluorescent detection using multiple-tailed primers. Genetics and Molecular Research 
5: 72-78. 
Neff, B. D. and Gross, M. R. (2001). Microsatellite evolution in vertebrates: inference from 
AC dinucleotide repeats. Evolution 55: 1717 — 1733. 
153 
Ng, S. H. S., Chang, A., Brown, G.D., Koop, B. F. and Davidson, W. S. (2005). Type I 
microsatellite markers from Atlantic salmon (Salmo salar) expressed sequence tags. 
Molecular Ecology Notes 5: 762—6. 
Ottowell, K. M., Donnellan, S. C , Moran, G. F. and Paton, D. C. (2005). Multiplexed 
microsatellite markers for the genetic analysis of Eucalyptus leucoxylon (Myrtaceae) 
and their utility for ecological and breeding studies in other Eucalyptus species. 
Journal of Heredity 96: 445 - 451. 
Peakall, R., Gilmore, S., Keys, W., Morgante, M. and Rafalski, A. (1998). Cross-Species 
Amplification of Soybean (Glycine max) Simple Sequence Repeats (SSRs) Within the 
Genus and Other Legume Genera: Implications for the Transferability of SSRs in 
Plants. Molecular Biology and Evolution 15: 1275 — 1287. 
Primmer, C. R., Moller, A. P. and Ellegren, H (1996). A wide range survey of cross species 
microsatellite amplification in birds. Molecular Ecology 5: 365 — 378. 
Raymond, M. and Rousset, F. (1995). GENEPOP (version 1.2): population genetics software 
for exact tests and ecumenicism. Journal of Heredity 86: 248 — 249. 
Raymond, M. and Rousset, F. (2004). GENEPOP on the Web. 
http://wbiomed.curtin.edu.au/genepop/index.html. 
Rozen, S. and Skaletsky, H. J. (1998). PRIMER3. Code available at http://www-
 genome.wi.mit.edu/genome_software/other/primer3.html. AccessedSep 2007. 
Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning, a laboratory 
manual. 2nd edn. Cold Spring Harbour Laboratory Press, New York. 
Selkoe, K. A. and Toonen, R. J. (2006). Microsatellites for ecologists: a practical guide to 
using and evaluating microsatellites markers. Ecology Letters 9: 615 — 629. 
Theissinger, K., Feldheim, K. A., Taubmann, J., Seitz, A. and Pauls, S. U. (2008). Isolation 
and characterization of 10 highly polymorphic di- and trinucleotide microsatellite 
154 
markers in the mayfly Ameletus inopinatus (Ephemeroptera: Siphlonuridae). 
Molecular Ecology Resources 8: 1285-1287. 
van Oosterhout, C , Hutchinson, W. F., Willis D. P. M. and Shipley, P. (2004). Micro-
 Checker: Software for identifying and correcting genotyping errors in microsatellite 
data. Molecular Ecology Notes 4: 535 - 538. 
Vasemagi, A., Nilsson, J. and Primmer, C. R. (2005). Expressed sequence tag-linked 
microsatellites as a source of gene-associated polymorphisms for detecting signatures 
of divergent selection in Atlantic salmon (Salmo salar L.). Molecular Biology and 
Evolution 22: 1067-76. 
Vinson, C. C , Sampaio, I. and Ciampi, A. Y. (2008). Eight variable microsatellite loci for a 
Neotropical tree, Jacaranda copaia (Aubl.) D. Don {Bignoniaceae). Molecular 
ecology Resources 8: 1288 - 1290. 
Walsh, P. S., Fildes, N." J. and Reynolds, R. (1996). Sequence analysis and characterization 
of stutter products at the tetranucleotide repeat locus vWA. Nucleic Acids Research 24: 
2807-2812. 
Wright, T. F., Johns, P. M., Walters, J. R., Lerner, A. P., Swallow, J. G. and Wilkinson, 
G. S. (2004). Microsatellite variation among divergent populations of stalk-eyed flies, 
genus Cyrtodiopsis. Genetics Research 84: 27 — 40. 
Zane, L., Bargelloni, L. and Patarnello, T. (2002). Strategies for microsatellite isolation: a 
review. Molecular Ecology 11: 1 — 16. 
155 
C H A P T E R S E V E N 
Population Genetic Structure and Demographic History of Teretrius 
nigrescens 
7.1 Abstract 
Molecular markers have so far, in this work shown the existence of two distinct maternal 
genealogies of T. nigrescens. It is possible that genetic and ecological interactions accompany 
this differentiation, but this has not been clear. This study uses the hypervariable markers 
developed earlier in the work to study the genetic population structure, eco-genetic 
differentiation and demographic history 13 sample populations of T. nigrescens. A total of 432 
individuals from the putative populations were genotyped at 21 polymorphic microsatellite 
loci. The data were analysed with standard statistical methods and Bayesian approaches. The 
loci were polymorphic for most populations. Genetic diversity was higher in field than 
laboratory populations. Population bottlenecks were not detected, but recent expansion was 
observed in laboratory populations. The samples separated into five informative groups, 
bringing together populations from the same geographical zones. However, population 
structuring was hierarchical to a maximum of nine populations. While geographical distance 
had contributed to population genetic identities, the introduction of the predator into Africa 
resulted in significant genetic drift and allelic equilibrium changes. Thus, a genetic quality 
control approach to biological control is suggested. 
7.2 Introduction 
Hypervariable molecular markers are potentially essential tools in biological control. They can 
be used for the detection of significant historical demographic events, genetic variation and 
monitoring of established populations (Lowe et al., 2004; Lloyd et al., 2005). Classical 
156 
biological control is the introduction of a co-evolved natural enemy meant to control an 
inadvertently introduced invasive pest. Both introductions are often followed by rapid 
population expansion in the new geographical range usually leading to founder events 
depending on the size of the founding population (Grevstad, 1999; Sakai et al., 2001; 
Memrnott et al., 2005). If the introduced populations are small, a significant reduction in 
allelic diversity through.genetic drift,and inbreeding occurs (Nei et al., 1975). For natural 
enemies, this effect can be reduced by large releases, multiple isolation from the field over the 
expanse of the genetic range (Nei et al., 1975; Grevstad, 1999; Roderick and Navajas, 2003; 
Lloyd et al., 2005). However, initial population sizes may not be informative without 
knowledge of the underlying natural diversity, which could be exploited to increase diversity 
(Lockwood et al., 2005; Bacigalupe, 2008). Also, the initial diversity can easily be lost in 
laboratory culturing if strong selection forces occur to enable the organism to survive under 
such conditions (Bigler, 1992). Such may result in lower vigour of the natural enemy in the 
wild (Bellows et al, 1999). 
Nuclear micro satellite markers are highly variable and informative in population genetics. 
They are distributed throughout the eukaryotic genome and are amenable to genetic 
recombination through mating, hence useful in the studies of population structure, 
demographic dynamics, population movements and dispersal over variable time scales 
(Beaumont et al, 2001, Liu et al, 2009, Zhang and Hewitt, 2003; Ellegren, 2004; Bray et al, 
2009). In classical biological control, they can be used to investigate the source and area of 
origin of an introduced pest population, population-specific monitoring of changes in genetic 
diversity and the relationship between ecological suitability and genetic diversity. For 
instance, Lloyd et al. (2005) used microsatellite and mitochondrial markers to examine the 
effects of collecting, quarantining and establishment of the spurge gall midge Spurgia 
157 
capitigena (Bremi) (Diptera: Cecidomyiidae), a natural enemy of Euphorbia esula (L.) on its 
genetic diversity. Nuclear micro satellite markers have also been used in the study of invasion 
genetics (Liu et al., 2009), conservation biology (Beaumont et al., 2001) and even in human 
population genetics (Bonhomme et al., 2008). 
Teretrius nigrescens was introduced into Africa principally as two geographic populations -
 one from Costa Rica and one from Mexico- released separately (Giles et al., 1996; Meikle et 
al., 2002; Hill et al., 2003). A phylogeographic study of the mtCOI sequences of 10 local 
populations has shown the existence of two distinct mitochondrial lineages of T. nigrescens 
(Chapter 6) with possible coexistence of the mitochondrial halpotypes in mid-altitude 
populations of southern Mexico. Mitochondrial genes are generally clonally inherited, hence 
invaluable in determining population lineages and species boundaries (Adams and Palmer, 
2003; Thao et al., 2004). However, they may not reveal genetic recombination or recent 
population genetic events (Loxdale and Lushai, 1998). 
Genetic variation possibly had a role in the variability of T. nigrescens success in Africa 
(Anonymous, 1999), although its contribution has not been studied directly. Also, despite the 
importance of genetic diversity in biological control, little effort is usually given to 
maintaining diversity in the colonies and the genetic diversity of the released specimens 
relative to the native populations is virtually unknown. This study sought to provide baseline 
information on their diversity and ecological suitability is necessary for eventual informed 
management and eventual monitoring of their establishment in the field. To address these 
issues, neutral micro satellite markers were used to investigate the fine-scale population 
structure and demographic histories of the T. nigrescens populations from a north-southerly 
158 
transect between central Mexico and north eastern Costa Rica and of samples from the region 
already introduced into Africa (Meikle et al., 2002). 
7.3 Materials and Methods 
7.3.1 Sampling and PCR 
A total of 432 samples of adult T.. nigrescens were genotyped. Samples were variously 
obtained from laboratory and field colonies as described in Chapter 2. Sample preparation and 
DNA extraction were done using the standard phenol-chloroform protocol (Sambrook et al., 
1989). 
Genotyping was done in simplex PCR reactions as described in Chapter 2 using 21 variable 
labelled micro satellite primers developed by Omondi et al. (2009). Allele sizing was done on 
an ABI sequencer, with LIZ 500 internal standard. Genotypes were scored using GENEMAPPER 
Software Version 7 (ABI Biosciences) and exported into MS Excel worksheet through text 
editing applications for collation and preparation of input files. Any samples that failed to 
amplify were reanalyzed three times in simple PCR. Individuals that consistently failed to 
amplify with micro satellite markers were amplified with a 1300 bp fragment of the mtCOI 
gene and the ones that failed were excluded. Most of these individuals had a traceable 
handling history pointing to poor DNA quality. 
7.3.2 Data Analysis 
Null alleles occur when some samples fail to amplify with certain markers perhaps due to a 
mutation within the primer annealing site. Such mutations prevent annealing hence PCR does 
not occur under optimized conditions. Such results may be discerned from a general lack of 
amplification due to inadequate DNA quality; as such individuals amplify well at other loci. 
All non-amplifying samples were therefore re-amplified twice for each failed locus to 
159 
differentiate null alleles from amplification failure. MlCROCHECKER software (van Oosterhout, 
2004) was used to evaluate the existence of null alleles for each locus. 
Genetic Diversity 
The genetic diversity was tested with various applications. CERVUS version 3.0 (Kalinowski et 
al., 2007) was' used to calculate the number of alleles (IIA), allele frequency and Hardy 
Weinberg Equilibrium (ETWE) expectations of heterozygosity (HE and Ho). GENEPOP version 
3.4 (Raymond and Rouset, 1995) was used to test for population departure from HWE, and 
linkage disequilibrium using Markov Chain Monte Carlo (MCMC) methods with 10,000 
replications and 10,000 dememorisation steps. The allelic richness (TXR) was determined using 
FSTAT version 2.9.3.2 (Goudet, 2002). Fixation indices (Fis) were calculated according to 
Weir and Cockerham (1984) using GENETIX version 4.04 software package (Belkhir et al., 
2002). 
Demographic history 
Recent demographic events such as migration, population bottlenecks and expansions leave a 
genetic signature in the genotypes and allele frequencies of present day populations (Cornuet 
and Luikart, 1998; Luikart and Cornuet, 1998). Population contractions often lead to the loss 
of rare alleles and allelic diversity. However, the allelic loss in homozygotes occurs faster than 
the loss in allelic diversity especially for alleles of low frequency (e.g. 0.01) leading to 
heterozygote excess (Luikart et al., 1998). 
Each population (except Mombasa) was analysed for heterozygote deficiency or excess using 
BOTTLENECK version 1.2 (Cornuet and Luikart, 1998; Piry et ah, 1999) based on 10,000 
replications. Few microsatellites fit strictly into the stepwise mutation model, yet assumptions 
160 
of an Infinite Allele Model may lead to occurrence of a bottleneck without the attendant 
heterozygosity excess (Piry et al, 1998; Garza and Williamson, 2001). Also, one third of the 
microsatellites developed for this work were compound markers involving two different 
motifs, enabling allele differences that are not products of the motif length (Table 2.1) (Bull et 
al, 1999). The Two Phase Model (TPM) was therefore used assuming 50 % Infinite Allele 
mutation model (IAM) model was therefore used, with 10,000 replications. Demographic 
decisions were based on Wilcoxon's Test for the probability of heterozygosity excess and 
deficiency and two tailed tests of both events. 
Population structure analysis 
Pair-wise genetic distances based onNei's (1973) minimum genetic distance between pairs of 
populations was done based on all 21 alleles and restricted to alleles that were in HWE in the 
reference Teupasenti population alone. Population clustering was analysed based on 21 loci 
meeting the HWE threshold. We used Bayesian methods implemented in the program 
STRUCTURE version 2.2 (Pritchard et al, 2000; Falush et al, 2003, 2007) to test for the 
existence of a genetic structure between and within the 13 geographic populations of T. 
nigrescens genotyped. This method uses a parametric genetic mixture modelling to infer the 
number of populations and likelihood of assignment of individuals to each population based 
on allele frequency. It assumes that there are K populations contributing to the gene pool of 
the genotyped population. Each population has an unknown but characteristic allele frequency 
at each locus. Assuming that the loci are in Hardy Weinberg Equilibrium and are unlinked, the 
individuals are assigned probabilistically to one of the assumed populations depending on 
their genotype, such that the equilibrium is not violated. The number of populations modelled 
to change between known bounds and the proportional probability of membership of each 
161 
individual is determined. The number of clusters of populations (the most probable K) is then 
determined based on the assignment of individuals and estimates of the allele frequencies. 
Three modelling approaches were used to infer the log likelihoods of the number of clusters, 
two ad hoc (Evanno et al., 2005; Pritchard et al., 2007) and one learning approach (Beaumont 
et al., 2001). It was assumed that each group composed of unknown subgroups and that 
putative population identity of the insects were genetically uninformative. Any individual was 
equally likely to belong to any subgroup, but the number of subgroups was unknown and gene 
frequencies were correlated. The structure was run in four replicates per assumed K (between 
1 - 20). An admixture model was assumed, with default parameters advised in the program 
manual (migration prior of 0.05; initial value for alpha inference = 1) (Pritchard et al., 2007). 
A different lambda was inferred for each population from the genotype data. Burn-in period 
was fixed at 10,000 and running length to 100,000 under admixture ancestry model. This 
simulation was run on the original data (all polymorphic loci, some populations not in HWE) 
with null alleles coded as recessive alleles (Falush et al., 2007). 
Most individuals in this run were strongly assigned to one population (for K = 2 to K = 9), 
showing the existence of some population structure hence the need to estimate the true K 
(Pritchard et al., 2007). The probability of K [Ln P(D)] was plotted against values of assumed 
K between 1- 20, predicting the true value of K (Figure 7.1) as the point where the curve 
enters the plateaus phase (Pritchard et al., 2007). 
Finally, the approach of Evanno (2005) was used on the results of the first method to 
determine the most likely K, and to compare with the result obtained according to the user's 
STRUCTURE manual (Pritchard et al., 2007). Briefly, we analysed second order rate of change 
162 
of L(K), denoted as 'Ln P(D)' in the simulation results, depicting the true K as the point where 
there was the sharpest change in the slope of the distribution of L(K). This was determined by 
plotting AK= against assumed K values from the simulation (Figure 7.2): 
AK = m(\L"K\)/s[L(K)] 
Where AK is an ad hoc quantity based on the second order rate of change of the likelihood 
function with respect to K. It is the mean absolute value of L"K between the runs divided by 
the standard deviation of L(K) for each value of K {s[L(K)]}.thus: 
L"K = L'(K+l)-L'(K) 
The mean differences between the successive values of likelihood of K \L' (X)] are computed 
as: 
L'K= L(K) - L{K-Y) 
In the third approach to determine K, the reference population method (PopFlag = 1 option) of 
Beaumont et at. (2001) was used. Since in the first run, the Yoro geographical population 
clustered neatly together across all putative vales of K, they were used as a learning sample 
assumed to be correctly assigned to their genetic population. Modelling of other populations 
was therefore based on the a priori assumption that these samples were correctly classified. 
This approach improves the accuracy of classification and inference of genetic clusters, but 
would be less reliable if clusters are not very distinct genetically (Pritchard et al., 2007). 
163 
7.4 Results 
7.4.1 Genetic diversity 
For the gene diversity tests, only the 16 loci that were polymorphic and in HWE in the large 
populations are shown (Table 7.1). Most microsatellite loci were polymorphic, with a mean of 
3.8 alleles per locus. All samples considered together showed significant deviation from 
Hardy Weinberg equilibrium at all loci. Specific population parameters were variable (Table 
7.2). Although most geographic populations did not deviate significantly from the HWE 
expectations for most alleles, a few populations either deviated significantly or were 
monomorphic for alleles that were polymorphic and in equilibrium for other populations. This 
was most common in small populations from which of less than 10 individuals were sampled. 
The unbiased estimates of genetic differentiation are summarised on Table 7.2. Most samples 
from Mombasa failed to amplify at most loci after three attempts and hence the population 
was excluded from bottleneck and genetic diversity analyses as the general failure may have 
been caused by poor DNA quality. 
The analysis using MlCROCHECKER detected significant existence of null alleles in five loci for 
the test population. The mean frequency of null alleles was 0.25 ± 0.14 including individuals 
that did not amplify across most alleles, while the mean number of completely genotyped 
individuals per population and locus was 13.5 ± 10.1. The Benin population had the highest 
proportion of completely genotyped individuals, while the small populations (Kiboko and 
Yoro) did not have a single sample completely genotyped at all loci. 
164 
7.4.2 Genetic differentiation 
Pair-wise genetic distances between pairs of geographic populations were higher between the 
southern (Honduras/Costa Rica populations) and Mexico populations than within these 
general groupings (Table 7.3). However, these distances varied significantly within these 
population groups, especially among populations obtained from laboratory colonies or with 
recent such ancestry. 
7.4.3 Cluster analysis 
Without a priori assumptions on population origin and by coding null alleles as recessive, the 
analysis of the potential substructure in the pooled population based on 432 individuals and all 
21 polymorphic loci resulted in a maximum of nine genetic clusters, with individuals tending 
to be assigned to one grouping. The curve of the posterior probabilities of if against assumed 
K, used as an informal ad hoc indicator of the number of genetic populations (Pritchard et at., 
2007) rose till K= 5 and then levelled off (Figure 7.1). Thus the values of real K could be 
visually discerned at approximately K = 5. Using the second order change in the rate of the 
rise in the posterior probabilities, the most likely number of clusters was estimated at 5 (Figure 
7.2). The a priori set reference population flag approach of Beaumont et al. (2001) did not 
reveal a significant structure as all individuals were allocated to the possible clusters in equal 
proportions between K = 2 to K= 9. 
The results of the cluster analysis were in agreement with the proportion of membership of 
populations to each of the clusters (Figure 7. 3). The five inferred clusters generally reflected 
the geographical origin of the sampled populations. Cluster 1 included Batan, Tlaltizaspan 
and Mombasa samples (Mexican origin), cluster 2 Benin samples and cluster 3 Kiboko, KAPJ 
and "Store" populations. Cluster 4 grouped together Malawi and Ghana samples, while 
165 
Cluster 5 had all the populations from Honduras. Oaxaca geographic population was allocated 
partially to clusters 1 to 3, and 1, 3 and 4, respectively. The lowest proportion of membership 
of a given cluster was 0.00 for Kiboko to clusters 1 and 2 and Gualaso to clusters 1 and 3 and 
the highest was 0.996 in cluster 3 for the Kiboko population. The highest assignments were 
generally from small populations and colonies with a distinct laboratory culture history. 
7.4.4 Demographic history 
Analysis of possible genetic signatures of population bottlenecks and expansion revealed a 
significant mode-shift in the Gualaso population only. This is however not informative, since 
this was a very small population (n = 5). The rest of the populations showed allele frequency 
distributions that were typically hyperbolic, indicating no loss of rare alleles. Most populations 
showed significant heterozygote deficiency at more loci than those that were excess, leaving a 
signature of recent population expansion, but without a significant mode shift. 
v o 
v o 
a 
O 
4-1 
w a o 
2 o< o a, o 
o 
<D too 
to K 
C5J 
o 
t o 
•cs 
> 
o 
o 
CD 
?s 
CO 
O 
o 
a 
CD 
crj 
r* 
O 
+-» crj 
S-l CS 
> a ns 
a 
faO as 
CM CO 
O O 
>^ O 
0 0 
J 3 
aj H 
O 
o" o 
*>< 
s 
c3 3 
X 
w 
o 
<u o 5; 
W 
o 
o a. 
H 
P3 
N ^ O OO 
^-f v o CD - M 
v q - 3 ; i n p 
<N d o d 
3 o o O o c 9 T3 rt o CN o ^ t -a 13 CO CN o CO Tt- i—i O g <N d d o" ffi 
'9 8 v o CA ? M i—i CO CO -=d- o 
a. 
2 
CO cO VO CN 
i n 
d i n d o d 
O CA o r— 
O VO IT) i - l 
O \ f I N l O 00 ^ <N d d d 
o o o 
o o o 
O O O 
CO 
CO VO CA CO 
^ t - o \ C N i n 
C N ^ t ; i n o 
co d d d 
v o 
c o 
i- l O O 
m o c o i n 
o C A i n c o 
N ^ ^ ^1 ^ 
T-< d d d 
C A r-- c o o 
c o C N cN o 
CN i~ ! O O O 
i - i d d d 
oo ? M c o C A i - i CN T-i 
i n i n o 
CD CD CD , „ O O O * 5 v - i o O O CO • 
i - i o o CD CD CD 
r— vo 
o 
v o 
o 
o CN 
x t - p i n v q T-I 
-^i d d d 
o 
VO 
^ - O O CN o o in o o 
CN d d 
f- i n -4- as 
^ O i—< C \ i—I 
t ^ ; -3 - c q - ^ C N c o 
T-5 d d CD 
r ^ 1/1 Tf- IT) 
T—< CN C A i n 
c o cN T-^ i-H 
C N d d d 
o o o 
o o o 
o o o 
i-< d O 
t- CD CD 
CN o o 
^ ° *-? 
i - i d d 
c o c o i n C A 
_ ^ t - C A c o i n 
g ) cN i n p p v q 
?< CD CD CD C N 
i n 
o o 
o CD o 
O CD CD 
CD CD CD 
i - i d d 
O CD CD 
CD CD CD 
CD CD CD 
C N CN CN O i n ^H -^» o 
OJ CD CD CD CD 
v - i d d O 
O 
O 
i n i n 
r— o 
c o d 
o o o h ~ 
o o o v o 
m P I o m o i c 
C N d d d 
i n 
i n c o c o c o 
l O C M C ) 4 
v o c o i r i c o i j -
 CN d d d 
C N i n v o c o 
. _ x i - c o c o r -
 ^ n i n v o i - i o c o 
i - i d d d 
r-~ v o c-~ i n 
_ v o i n i n i - i 
?-n C N <N o o p 
-i CD CD CD i 
CO 
o i n 
o CN 
O i - i 
c o 
CO 
^ - i 
i n 
i n p 8 ^ °i 
i—i 
r-
 c o o 
^-i 
o 
i n 
CO 
T-l i n i-5 d d d d d d 
CN i n i n t— 
CA O f - VO 
c o CN ^ CO CO T-f 
-^i d d d 
. o ^ J - o o 
v _ v o i n o o i ° CN vN p p O 
i - i o o 
VO VO CA CN 
. _ v o r - c o C N 
^ C N ^ - 1 i - ; <N 
I -H o o o 
i n C N C A 
C N i n r— ? 
c o ^ o 1 * - i-H v o 
C N d d d 
t— CA CD CO ^i- co i n co 
o o c o c o " n r~~ c o 
T-< d d d 
CA CO ^cj- 0\ i n H ^ t " CO 
^ VO CA CA CN ^ o i n C A -sf t ; <N <N -^ CN • * r^ CN co C N cN i - i CN O d d CN CD CD CD 
CA ^ H CD CN 
CO CO CD CA 
c n ^ c o i t H 
i - i d d d 
-4 - ?<*? C N ' 
v o -3 - 1-H ? 
^ t - ^f- i n -^ - C N 
CN d d d 
i n 
c o C N 
CO -^f i - i vo co r— co r>] H 
CA 
o 35
  
1.6
 42
  
0.1
 35
  5 CO p o o o o o o o 
-^i d d d 3
 5 1.6
 42
  
0.1
 35
  
d d d d 
i n 
CO 
c o c o r~-
 ? * c i i n 
t - . v q v q 
c o d d 
v-t CA O CA 
C A c o v o i n 
v o o o i n i n p 
C N d d d 
CN i - i o t— 
,_ , m i n C N i - i 
^ i * r - T-i i - i CN 
-i d d d i n 
c o 
t - - CO "5f CO 
N CO r H CO 
c o i n m H 
CN d d d 
J5 O M a 
co CN co r -vo in in o 
c o C N o o o 
r— i n o C A 
c o cN c o i - i 
t^~ [ > ; VO -i-J l > ; 
c o d d d 
r"— -*J- i - i CM 
CD c o p-- o o i n i n i n c s o o 
c o d d d 
o i n 
i - l CA O ' 
CN IT)" ICJ" CO 
C A p v q -^h c o 
^t CD CD CD 
VO H O N 
rn ~-* CO CN CD CD 
^ ^ m v o i n ( N 
c o d o o 
i n 
i n 
is o co a W Q ^ 3 g t r l K p Q H Z 5 S S M £ H 2 rt q K f f i ' K f t " H ^ 33 £ o 
CO 
a o 
O 0 
3 
3 5 
pp 
o o 
•3 J o o "IT* jD 
1 ^ 
W 
o 
FQ 
^ j - i n i n i— 
NO o r— NO 
C \ t o CO H 
c o C N ^_; o d d 
O CD O CO 
o CM o t o 
LO cN O - ^ NO cO 
t N d d d 
o^ th N co 
o w i - i -
 CN o o o 
<N N N G\ 
_, N O C N r— t - i 
CN o o o 
N CA Cv f -
 N O ^ t o r--
 i - i o o o 
CD i - l CN ^ j -_ O * N O 
J j i f o t o t o o 
i-5 d o d 
ON CD O 
UO o o 
O N o o 
v 5 6 d 
t N O CO O c o NO i n i n 
r~~~ C N i—i c o c o o 
c o d d d 
c o c o r— r--
 „ C N o NO t - -
 C D ^ - j C N v o m o 
C N d d d 
CN ON ON ON 
c o r— <N i - t 
LO O -s t " 3 ; i - i 
c o d d d 
ON o to o 
CN CO CN o 
o \ -si- N O -3 -
 ° o C N ^_; 0 - 0 - o 
O CD O CO 
o cN o -^ f 
IT) [ N 9 ^ ^ ^ 
CN d d d 
i n oo 
c o CO C N c o 
\ t \ o i n v < 
t o G ; v o ? + t o 
C N d d d 
C N c \ c \ r— 
, _ N O i - i c o p -
 S M t~- M ^ t ^ 
CO * • * * 
i - i o o o 
VO ?<??<- NO 
CN CN C N -sj-
 ^-O cr, C N CN I - I i - i 
i-5 d d d CO 
LO o N O CO 
C N o N O r~ 
i j - v o u o ^ q 
i - i d d d 
p - t— m o 
C N i - i CN CD 
CO CN t~-; N-^ -rH O 
i-5 d d d 
c-~ CN O N—i 
^-1 -^H LO i—1 
<N t ^ l CN CN i-< 
CN d d d 
CO CO O i—l 
OO t-~ O N t 
CN o o o 
i—i C N i—i CD 
. NO CN t~~ -3 " 
' N T NO CO NO ^t* CO CO . . . . 
CO O O O 
o 
^ co ^ 
o 
r— 
xr co 
i-i oo t— o 
m 
r-N C O CJ NO CO 
NO 
NO 
cN 
NO 
CN 
ON 
o d d d CO d d d
 CO t— CN O N 
r^ l"-- - ^ CO CO 
™ CO ^ - i CO NO CN 
t o d d d 
W O u> 
-3" CN CN! icf-
 »^. \ t > 0 t o N t 
^ > o t o i n \ t ( S 
c o d d d 
-N|- "tf- LO o o 
i - i c o t—- o o 
CN N t t o CN! 
0 0 " * t o d d d 
o o o © 
o ? * o o 
i ^ ^ o i n o o 
c o d d d 
CD CO C N CO 
O CN N O i - i 
[ ^ O LO 1 ^ 1-1 
t o d d d 
N O 
c o 
t N N t r > m 
i n r - i - i c o 
[ f ! l O N+ nr)- rt 
cN d d d 
f-~ o C N r -
 i - i t o c o o 
IT) I H CM vH t 
oi d d d 
LO C N CD O N 
_ CN P - O CN 
© OO N O LO NO O 
i - i d d d 
r - r— L O o 
i t " i - i C N o 
CX) CN] c o rt H q 
-^i d d d 
i * i - i -st- c o 
N f i n c \ N t 
CD CN CN V H 
^ d d d 
i n \ t CO C \ 
NO CO CO OO 
t~_ LO f - ; LT) CN 
c o d d d 
O N r— c o C N 
, _ i - l ON ? * OO 
^ H ^ t ^ q 
LO CD O O 
CO O N O C N 
t O CO N t O 
t - - C N r— c-; o 
? ^ d d d 
OO CD o o NO 
- n CO L O LO c o 
™ c o r > N N O V H 
i f d d d 
i n C N o o 
r - i - i CD CD 
c o CN! q q 
i - i d d i - i 
O CD CD CD 
O OO CD CD 
LO CN CD i - i CN O 
CN d d d 
N O N O - ^ i - l 
m N t o t o t t 
c o ^ - C N N O cN v q 
CN d d d 
o CN NO LO .^ r- co o in 
2J" icj- r-7 L O C N N O 
CN d d d 
o o N O 1-1 t~-
 _ , c o N O t o I - I 
L N c o c o CO CD CN 
CN; d d d CO 
CD LO ON CN 
O O CO i - l 
^° t o o c o o r-; 
i - i d d d i n 
r— o o c-~- o o 
OO i—l ON o CD 
i-5 d d 
NO NO CD CD 
I N \ t o o 
ON CN CO c o o o 
C N d d i-5 
r~- -vt- o o 
^ LO CN CD CD 
^ j t o * - • > n o o 
C N d d i-< 
C N o i n i n 
_^ CN CO NO CO 
^ t o e s t o O C O 
C N d d o 
v- t NO CD OO 
_ CN CO CO NO 
^ 1 LO C~-; CO i-H NO 
C N C D d d 
NO 
CO 
t— r— t-- ON i n ON NO oo 
NO I^- CO i - l LO 
oi d d d 
! > . i - l CD O N 
i - i NO LO CD 
NO "3- CN LO 00 ^ CN d d d 
O O O CO 
O CO CD t-~ m CN! o N t N t N 
oi d d d 
CD LO CN NO 
. CN! N f u- i h -
 J 5 OO ^ LO ^ t " H 
co" d d d 
LO 
CO 
c o o N O r~-
 C N c o c o 
C\ LO -5f 
C N d d C D 
CO C N N O CN 
r » c o L O C N 
* " N f NO ' t j - i n vH 
C N d d d t o 
i d " O N NO c o 
_ i - i c o -*t- r -
 ° ^ CO CN NO f - - o 
t o ' d d d m 
c o x h i n c o 
LO c o CN CN 
CO 1)- C-; ^ h NO CN 
CN d d d 
CD CO c o ? 
NO t~- CO CN 
c o c o ^ i n 
co" d d d 
i f i n c \ i n 
i n i - i t~- CO 
° ? NO t - ; NO NO O 
t o d d d CN 
ON 1—1 i—l t— 
, _ t~- i - l t - - CO 
* n O N L O i n c q CN 
t o d d d t o 
N O o CD cN 
0 N h - I N C M t o g CN C N c o c - _ , 
LO CD CD CD 
c o CN 
CO i—i 
ff ^ ffi' ffi [£ H ^ 
CN 
^ ^ w o co ?5 
3 s f f i K f i H 
CN 
a o M O w fl 
O N O CD ^ f 
LO t~~ ON CD 
1 - ; r-; t - ; o 
i n d d d 
. BJ Q g K - K f ^ H ^ 3 I f K W ^ M Z 3 ^ f f i f f i f ^ 
{S 
u 
o p 
1 
o. 
0 
3 
c3 o 
.a 
a 
CU 
pq 
o 
o 
2 
o 
'£ 
2 3 
PQ 
o o o o o o o o o 
o o o o o o o o o 
M N N in 
CO CO rt VO 
CO T-H O O N O 
T-i d d d 
r— ON ON o * N M O N v^ q q q 
T-i d o o 
C N as o C N 
oo c\ i/i M in to N -t 
00 ^ c4 o o o 
o o o o o o o o o 
o o o 
_, o o o 
s - °. °. °. 
T-I o o 
l O T^ CO ^ 
N CO CO CN 
T-< o o o 
ON ON o r--m T—< in t— 
CO CN CN <N CN o 
CO O d d 
O PI o\ Oi 
__ N O -ret N O in cn i^- r— NO t~- i-i 
oi CD CD CD 
N G i CO lO 
_ CN ^t" CO "St 
C J co co m m o 
T-i o o d 
CXI CO CO CN 
?*? £?— NO Tf \o h >o in q 
CO d d o 
NO 
Nt 
in co 
Pi IO C» CO o ON co m CO CO vo t>; o 
Nt d d d 
NO O N CO NO 
ID U l -t CO N O in r— r— o 
Nt d o o 
H O O 
\f CO C\ N 
NO Nt CO CO ^ io NO T-I q co 
i-i d o d 
c\ o co i-,„ CN o co in 
S p l t M O l O r H 
T-I O O O 
co in co r^ -,„ m ?+ > o 
^ N t <N c o CN] ( N 
CN d d © 
m i n c o C N 
CN O N CO c o 
© t o N O CO c o O 
T-i d o o NO 
? * M O O 
ON T- i CN O 
0 0 CS CN] r H r H O 
T- i d d © 
in i>- ON o _ o in in o 
f " CNl N t O O O 
T-H o o o 
o « o o 
-=cf NO O © 
CN) o © c q © 
CN) d d T-3 
c o o t— T—i 
, „ i n i—i i n C N 
<£> ^j. i n T-H o N t 
T - I © o o 
C N c o o m 
_ N O c o o r -
 [ ^ m CN! % t N t q 
CN d d d 
r H C \ \ t CM 
o - T—I CO CO N t 
£ j NO H CN T - I O ) 
CN d d d 
o o o o o o cq o o 
N-< d d 
© o o o 
O CO CD CD m IN q ^ N q 
CN © d © 
in Nt cN co 
_ , (?-?_ Nt co in 
C N co oo CN CN CD 
v4 d d d 
NO CN) C N C N 
co co co Nt 
N= m CO T-H T-I CN 
T-i © © O 
t- ON Nt C N 
'T-I NO O N CN \t ^ M H X 
C N ON in in 
CO T-I CO Nt Nt in CN CD co 
oi d d d 
O CO CD T—I 
f- CN © C N co co ^  co in N—i 
CN © d © 
CO rf Nt NO 
CO NO C N C N 
^ Nt C-- CN CN) © 
CN © © O 
???-I N O Nt r~-Nt in ON CO 
NO C N Nt T—i in 
T-I d d d 
t-- co N O Nt 
,„ Nt to CO NO 
^ NO O . O ) CO o 
CN) O O 
T-I T-I CD CO -S h- -t Nt TH grj CN CN CN CN CD 
CN) CD CD CD 
C N NO T—i CO Nt C N T—I N O in C A H oi q 
T-i d d d 
CD O O 
CD CD CD 
CD CD CD 
CD CD CD CD 
O <N o © 
in CN) CD CO CD CD 
CN d d T-i 
r- in in co _ CN (N CN) © 
C N ^ T-I © o cq 
T-I o d o 
CN © r— o 
,__ T-I co T-J in J° CN1 ON CO Nt CN! 
T-I d d d 
in co CN NO 
, O T—I T-I T—I 
CN o o o 
ON NO r— f— 
CO T—I T—i GO g n i o i n > o r f CN o d o 
O CO O T—I 
O C N O O N co "n cq in T-I 
oi d d d 
^- in o C N 
T-I co o NO 
co N—; in NO o 
co d d CD 
CD co N O r— 
_, CO in CO C N 
CN) co t-; Nq in T--I CN O O CD 
O N C~- T—I [— 
_. CO O f- CO ^r t^- CN in ^ f cq 
CN d CD d CO 
C N 
CN) NO CN T-I 
f- O T-i O 
CO T—I NO NO O 
co d d d 
CM TJ- o r-
 O T—i N O C N co T-I T | TJ- q 
oi d 6 6 
CD O O 
O O CD 
CD CD CD 
CD CD CD CD 
CD ^t CD CD in CN) O CO CN] in 
co d d d 
IN CN) co r-
 O NO T-l ON f^ _ co co o o t— 
T-i d d d 
C N 
Nt C A CN CN1 
._ NO CO r-- O 
N3 CN C N CO ^f CN) 
T-i d d d CO 
CD CN T^J- T-i 
_. N O O CN CO J in O M O xt H 
<N d d d 
CN r- <N o 
_ CO CN) Tj- CN CN -(- in in in cq 
oi CD CD CD 
NO CO O T-H 
r-- ON O O N 
co co "^ co in T-I 
c4 CD CD CD 
CD NO CN T-I _, -ret- r- NO r-J2 co C A "* 'NI; O 
CN d d d 
NO r~ NO T-^  
_ P - "Nt" CO T-H 
C N ^ cq NO in T-I 
CN d d d CN 
CN co 
co in Nd- T-H 
"*f CO -NT f~-
 tj- Nt in co co 
CN d d d 
NO CO T—l co 
N O T-I r~- r-^ G \ co N O in q 
co d d d 
co 
in CN co ^ ct-
 co CN in -t H CO ^; ^t O 
CN) d d d 
%* 
& 
m o to .n Q Sffiffi £ H Z 3 § K X f£B Z 
CN 
c? g$£ C^H Z . . pj O w 3 d ffi.ffi EC 
T3 ci 
g -a 
O p 
J-< 
.2! 
o 
U 
d 
O 
NO 
CO 
£ 
a '3 o 
NO w 
o 
LO CN 
f- O oo NO 
in oo cN tN 
NO o 
CN O ON CO ON Tf" CN O NO -3-
 OO 
CO 
oi o d o » Ol ^ cJ o d o 
CO 
o o o o O <N o o c^  O CO o o 
pi d d r! 
tN 
m T-H CO O O T-< - t^* m cN oo \ o co x i -
 CN d d d 
[— ON G \ CO 
v-f CO OO O 
u~) O N N O CO ^ 
(N d d d 
in r-- r— <N 
CN -*-i N O i n -4- NO NO - ^ CN 
^ d d d 
oo r— m o 
NO v-1 CN O oo co co r^ ^ q 
co d d d 
H ON NO O N oo o o CN 
C*-) O NO t~~; N - ; 
co d d d 
o o o T- I o o o ON o m NO o 
oi d d d 
m 
CO 
co -sj- oo -sf 
ON -^ H CN T - 1 ^ \ t m m o 
CN d d d 
i n 
CO 
CO 
co -^H r - x f 
CN IT) U ) O ^ CNI r - N t \ t 
-Jr d d d 
^f- NO o cN 
N \ t co m 
O N i n r— i n CN 
•^r d d d 
O N f CO H co -=f r - o co co r-^  co i n 
-3- d d d 
oo 
CO 
\ t >o o i n 
CO ON NO LO 
NO t - ; t - ; f - O 
^f d d d 
N t rf \ t VO 
OO CO OO CO 
NO NO t NO ^"< 
?^r d d d 
o 
W O ?/> 
O in 
CN cN 
NO NO 
cN o o 
i-i co CN 
g m ^ m in 
CN d d 
CN r-o 
NO 
CN 
d d d d d 
<? 
OO ON NO cN 1 — I 
vH O CO f-CS Tf ^f -^< o cN d d d 
to 
NO CO 
cN 
ON NO 
o 
CO 
t-; 
CO O 
o 
ON 
ON CD 
t-< d d d d 
o 
NO 
CN O N O ("— I D 
O N T - I i n r~- i>-oo CN C N CN CN o i - ; 
*-< d d d d 
i n o co r— co 
CN co co [— o N \ t N t i n q T-M 
CN d d d d 
cN NO CO CO co o in 
CN in in -3-J^- C N 
O NO OO NO 
m in ^- co 
£ -^ C-; ^  in r-l 
CO o o o 
ON 
N—* 
co d d d d 
O v-H [ CO 
t~- NO t- O J \o ^ t; « NO q 
co d d d 
co 
CO CN 
d 
GO 
£ 
CO 
O 
O 
O 
ft 
o 
o 
ft 
1—1 
<N 
CO 
I 
"o o 
0 ft o ft 
ca u 
<*> 
. 6 b 
- C : cu 
< 4 - I 
O 
> '-3 
CD 
<u 
CO 
to 
o 
CO 
CD 
?s 
m 
CN 
<D 
o 
c3 
CO 
CS 
N 
•J3 
CO 
o 
C/5 
C« 
O 
cs 
o 
a 
o 
1 
cs 
§ 
o 
o 
3 
0 es 
CS 
t - l O ON CO o o o\ o LO T-i in o 
Tf; O C-; i n t— o 
o o o o o o 
t~~ O VO VO 0 \ o 
o \ o r-~ i n en ^ 
m r-- vo T-I vo vo 
s 
en en 
1—1 
CN 
en 
vo 
xr 
t--
 CM 
in 
o 
in 
CN 
o\ o 
in 
O 
CD 
o 
OO 
en 
o o o 
o o o 
in 
CN 
o o o o o o o o o o o 
CO 
VO 
in 
CO VD VO 
in O 
o oo 
o 
t-l 3 co CN r-o VO in 
o o o o o o o CD o o CD CD CD CD 
OO vo O ^ t 
7X CO CO t - l 
i n t f ) ^ O j 
o O O O O O O O O O O O O 
o en vo *o 
o -3- J^- CO 
O VO -vf CN 
o o o o 
T-l ? * -!f i H 
CN VO CD ON 
«3 l> CO -^ 
o o o o 
CN CN ?n en T H vo in CN vo CN oo CN o d^- cn r-- t^- t-~ en co VO o vo VO r-~ in VO in o T-l 
O CO VO CD VO xf CN 
h ^ n \ o ^ t o i n CN CD vq CD vo i n en 
o o o o o o o o 
o en i n o 
o ^ - Ol o 
o vq H CD 
CD CD CD CD 
c N T - < e n o m i n o G N < N i n 
c o c N v o o c N C N O C D c n C N 
x f e n i n c D T - j T - i c D i n c N T - i 
O C D C D C D C D C D C D C D C D C D ' 
CD CD o o in CN t~-
 <N CN CD CN cN co vo 
xh xf in c^h i-< <N vo 
CD CD CD CD CD CD CD 
c D e n i n r - - < N o c o r ~ ~ c n c 7 \ c N T - i r ~ - c N c o ^ o e n v 3 r - ~ T - i e n C 7 \ i - < C N c n ^ O - ^ c X 3 T - i o O \ c N o O T - i e n T - < c ~ ^ - 3 - v o i r - - e N c N i n c N T - ; c N C D V 3 - ^ i > - o q e n i n i n T ^ c N i n o o i n r - - ; V 3 T - < 
C D C D C D O C D O O O O O O C D O O O O O O C D C D O 
O O [~- <d CD O VO -^ 
o in 
CD O 
CD in VO ^ CD t--
 CD CD CD 
>l >l <ri f- <c! vo u o <d <d <d o 
CD T-i O CD 
JP ^ ^ <! <! ]z, o\ ]Zt !z; 
en en o\ co 
O en O CN 
in CD in CD 
CD CD CD CD 
vo -^ - CN CO 
-^ - T-I C^ 0\ 
<N en CD CD 
CD CD CD CD CD CD 
vq 
CO en 
1—1 
in o en in 
en 
CN 
CN 
<N 
oo 
-vi-
 C7N 
in 
o 
in 
CA 
CN 
CD 
V3 o 
in 
in oo 
T-; 
en 
en 
en 
en en C7i 
en 
xf 
CN 1—1 Xl-
 CN 
CN 
VO 
CD CD CD o CD CD CD CD CD o CD CD CD CD CD o CD 
co 
in 
CN m en CN VO 
xf 
T-l 
in 
oo 
in 
T-I 
in 
i—i 
CO 
in 
CN 
o in 
in r-- oo un C7N 
in 
co en o 
CD 
in en 
in 
VO 
CD 
CD 
VO 
CD 
O 
CD 
CD 
CD 
CA 
o o o 
O o 
VO 
CD 
in 
o 
in 
en 
o 
CD vo 
CD 
CD 
CN 
CD 
in 
in 
CD CD CD* CD CD CD CD CD CD CD CD CD 
co 
T-l 
in 
co 
CN 
CD S ^ O O CD CN CA in VO in CD en CN CN en T-l CN en VO CN CN CN en r-en in CA CD CD CD CD CD CD CD CD CD CD CD CD 
C D O C D C D C D O O O O 
C D O C D O O O O i n O 
c D o c N - ^ c N ^ J - ^ i n i n 
C D O C D C D C D C D C D C D C D 
C D O T - i C O e n c O O C N O O 
C D u - > c o i - i c o T - i r - - c N i n 
C D e n c N - ^ i n i n v q i n i n 
O C D C D C D C D C D C D C D C D 
c N i n T - i C D O i n i > - T - i i n i > - c A 
e n o r - C D T - i T - i L n r - c N i n c o 
? v f v o c D C D o - i n c N e n i N C N e n 
o o o o o o o o o o o 
<N en C\ 
ON vo in 
en vo o 
o d d 
^h o\ co 
CN en vo in en i—i 
CD CD CD CD CD 
en en in in CN oo oo CO 
T-I VO CD "Sf 
1-1 VO IT - CD 
vo <N co CN 
en in in en 
CD CD CD CD CD 
in CD vo 
ON CO CN - CD ' 
CD 
VO 
CD 
u n c o i n c D T - i o \ o \ r ^ c N i — i v o 
O v ^ J - i n i n v o o o o o - ^ - v o c D O 
i n v o v o c D v o i n v o o o i n o o ' v ^ 
d c D C D C D C D C D C D C D C D C D C D 
CO ?<* OO 
T - I r o " * i n o T - I i n m m m i > r~- CN CA 
H H H H H H c - i H H H H 
1—1 .^  t~-C\ c-~ CN r^  r-; en CD CD CD 
in ? * T-l oo TH JS z z ^ g rt H H H 
oo T - I en 
vo vo i n 
CN i n t~~ 
CD CD CD 
CN CA CN 
i—i i—l CO 
CN i n r— 
CD CD CD 
O en CO t~~ ^ " C \ CN xf" co 
c~- i n g \ CN 
?<*? v o 
en o \ i n en i n vo 
oo o \ 
O CD 
o \ •^^ 
vo i n 
vo ->d-
 ON VO 
r— i n 
oo CN t-o 
CN 'T' CN o 
H 
^ 
E5 ^ 
ft 
S3 cej 
erf O 
cej 
CD 
?a 
CD 
ca 
171 
Table 7.3: Multilocus Nei's minimum distance between pairs of populations based on all 21 polymorphic loci (above diagonal) and on 16 neutral 
markers (in HWE in Teupasenti reference population) including null alleles (below diagonal) 
Batan Benin Kiboko Ghana Gualaso KARI Malawi Oaxaca Store Teupasenti Tlaltizaspan Yoro 
Batan 0.143 0.177 0.226 0.173 0.105 0.252 0.065 0.276 0.244 0.005 0.235 
Benin 0.084 0.270 0.061 0.097 0.197 0.095 0.080 0.329 0.118 0.097 0.109 
Kiboko 0.121 0.161 0.351 0.416 0.181 0.370 0.212 0.384 0.380 0.177 0.441 
Ghana 0,135 0.030 0.190 0.171 0.316 0.058 0.207 0.465 0.147 0.177 0.163 
GuaJaso 0.147 0.069 0.282 0.087 0.227 0.173 0.150 0.446 0.042 0.135 0.069 
KARI 0.076 0.121 0.117 0.198 0.177 0.321 0.096 0.162 0.274 0.108 0.296 
Malawi 0.160 0.046 0.220 0.026 0.096 0.211 0.243 0.457 0.150 0.195 0.180 
Oaxaca 0.045 0.044 0.131 0.108 0.111 0.064 0.138 0.234 0.187 0.055 0.200 
Store 0.213 0.211 0.217 0.275 0.229 0.104 0.292 0.149 0.348 0.248 0.509 
Teupasenti 0.143 0.056 0.238 0.066 0.041 0.165 0.069 0.097 0.209 0.201 0.061 
Tlaltizaspan 0.012 0.059 0.132 0.110 0.126 0.083 0.126 0.040 0.206 0.118 0.187 
Yoro 0.180 0.066 0.290 0.075 0.053 0.221 0.091 0.130 0.294 0.040 0.152 
172 
-22000 
Figure 7.1: Estimation of the population structure of T. nigrescens based on ad hoc measure of 
the gradient of K against L(K) (Pritchard et al., 2007). 
100 
90 
80 
70 
60 
50 
40 ? 
30 
20 -
 10 
0 
AK=m(\L"K\)/s[L(K)] 
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 
Assumed number of populations 
Figure 7.2: Detection of the number of clusters of individuals based on the ad hoc measure 
(AK) (Evanno et ah, 2005). The asterisk marks the inferred K. sharp changes in gradient in after 
K = 5 suggests the existence of genetic substructure. 
t -
 o © ^ © 
© cc 
£ o 
« 
a 
.3 c 
u 
CO 
M 
3 
U 
- 3 - -<d- r - T -
 c o © L O O 
O O O ON 
o © © © ' 
N O CO ^ f ^ 3 
V". " * CN T -^- © o © 
©> o ©' © 
11 en H 
3 
CO 
m 
o 
ON co 
© 
© 
© 
CN 
CN 
© 
ON 
0 0 
© © c O o 
CO 
O 
© 
^ 1 -
 © 
LO 
O 
© 
o 
© o © © o © 
I N 
CO 
L O 
CN 
© 
ON 
© 
© 
© © © o © 
© 
CO 
© 
© 
LO CO CO © 
0 0 r-< CO m 
© CN r - . *- . 
© o © ©' 
(N LO O N \ C 
M O ?>*? r-H © © ON © 
CO ? = * CO CO r~-
 I O L O NO T—1 <??) o © CO © o 
© T-i © © 
© © © O 
© ©> ©> <N 3 O © © o © ' © 
LO 
© 
CN 
CN 
© 
© 
© 
CN 
ON 
© 
© 
© 
0 © © © O © 
0 
o © © © 
© 
© 
© 
NO 
ON 
ON 
CN 
© 
© 
© 
© 
r-< O l ? * r i CN 
CN L O © CN © 
© O N © © © 
© © > © ' © ' © 
CC 
© 
© - ^ r - i 
CO © © 
© © © 
© © © ' 
-a 
CD 
ex & n ,—, c c 
l > 
oJD "o" 
en CD o 
— 
o o . 
> 
c 3 O O ^ 3 
O 
33 
_03 
03 
3 
CD CJ O U 
O a! 
CU 03 C 
J C OJ 
-4-J n K 
-4-1 u 03 CX U 
CD C (1) 
CD 
u . 
CX 
CD 
u 
L - a, 
3 
* ^ o ca c n 03 
"5 o 
'fi 
CX c^ 03 
03 3 
T 3 
s 
1 
L J 
(11 
CD > o3 _ 3 
O 
03 
u CQ 
03 
CD 
O o j —^' -a 
O 
CX 
3 c 03 CX (i . r~ r U H 
01) O L* 
s 
Z) 
O 
o 
a j 2 
o 
CD 
X J 
s c -
 u E 
3 i 1 cr| 
r - T 3 
ID o 55 
3 
-a 
C r. > o in r^  -a CD r ? ( ) 
r. 3 \> L t -si w ^ n 
CD O D H c b -4—i 3 n ^o <IJ L- ( — 1 ca c O CJ " c r t CD a , 
CD 
03 
tOJ 
T 3 2 o L-
 D . CQ '£ i C 03 (D 
OX) 
c CD 
CD 
_ 3 
CQ 1) f -
 C/3 
3 C t o c^ 
co (D O O 
[33 
L t -
 o CO 
C u n , n 03 c o o n ex OJ o X ) -*-i o OCJ 3 _3 
O 
L ( _ 
O 
a, 1c 
C^l 
ci) 
C^3 
03 
c 
03 
r— 
en 
? a 
a) 
J D CJ si 
3 
C/] 
^ 5 5 
CD r^ 03 n a i or) CD 
ro a l S _CJ a) 
r-~ 3 1 ~^ i n T 3 5 
3 > 
'-5 >-
 -4—i 
CD 
H 
174 
7.5 Discussion 
Most samples in this study were unambiguously genotyped at most loci. Null alleles were 
relatively rare, as seen in the HWE expectations, and, were not corrected for in the measures 
of genetic distance. Other studies have shown little difference between genetic analyses 
correcting for null alleles and those that do not (Liu et al., 2009; Williams et al., 2005). A 
deviation from HWE expectations was observed in a number of allele-population 
combinations. All analyses combining more than one putative population as defined by 
geographical or sampling source resulted in a decreased adherence to HWE expectations. 
Failure to clearly genotype some samples or loci may be attributed to null alleles (i.e., a 
mutation within the primer annealing region) or failure of PCR due to the quality of DNA. 
Amplification failure due to null alleles are usually occurs at few loci only. Poor template 
DNA quality causes poor amplification across most loci in specific samples (Dakin and 
Avise, 2004). Null alleles may cause a shift from HWE, often leading to heterozygosity 
deficiency as many heterozygotes are coded as homozygotes (Liu et al., 2009). 
The genetic diversity, uncorrected for population size, did not show any trends between 
laboratory and field populations. The Mexican populations however had greater diversity 
within populations than the Honduras and Costa Rica populations. Genetic diversity was 
however higher in parent laboratory than in possible daughter sample populations (e.g. 
Benin>Malawi or Ghana; KARI > Kiboko, Store, Mombasa). Such trends indicate genetic 
drift, sampling effects or loss of some alleles during colony subdivision when establishing 
new colonies (Lloyd et al., 2005). 
A further analysis of possible bottlenecks in the population suggests that most populations 
were expanding, but without a genetic bottleneck signature. Such population disturbances are 
common in biological invasions, range expansion and insect rearing (Bigler, 1992; Memmott 
175 
et al., 2005). The ability to detect a population bottleneck signature based on the 
heterozygosity excess and deficiency alone has been questioned (Garza and Williamson, 
2001; Cournet et al., 1998). This analysis is dependent on prior assumptions of microsatellite 
mutation models, which are often not studied directly. While there is no general consensus 
for the correct mutation model to use and most scientists adopt the middle ground 
assumption of the TPM model, the different models gave contrasting results. Within two 
phase model (TPM), one must make a quasi arbitrary decision about the proportion of loci in 
the single step and infinite allele mutation models. The choice of one or the other mutation 
model determines the result and decision obtained. The correct model is only correctly 
discernible through mating studies and sequencing of alleles (Eliegren, 2004). In this study, 
one third of the alleles used were composite, some composed of different repeat core sizes. 
Though several studies have identified such markers, they are clearly not accounted for in 
the three mutation models available. The inability of the He/Ho ratio to detect bottlenecks 
has been reported in populations with known recent bottleneck events and within the 
'window for testing' for such events (Lawler, 2008). In this study, the Kiboko population, 
raised from just three adults, showed signs of expansion but not a bottleneck within one year 
of culturing, possibly resulting from the masking of the signal during subsequent population 
expansion. 
In biological control agents, the most probable population constrictions are likely to occur 
during isolation from the field, laboratory rearing, subdivision and lag phase after release. 
Under artificial rearing, these events are often followed by leading to a population expansion 
during rearing. Generally, population bottlenecks result in either population extirpation or if 
there is recovery, an expansion. Consequently heterozygosity deficiency is observed 
especially in a rapidly growing population (Bonhomme et al., 2008). In fact, in this study, 
176 
the program could not detect any bottleneck signal well within the bottleneck detection 
window (Lawler, 2008). Sub culturing is known to have been done for further releases 
(Meikle et aL, 2002). Insects released in Ghana, Malawi, Tanzania and Nigeria originated 
from the Benin population (Anonymous, 1999). Similarly, Benin and Ghana samples had a 
recent history of introgression with wild individuals caught in Africa, but virtually none from 
Central America, the predator's native range (Meikle et al., 2002). This lack of secondary 
isolation and reintroductions from Mexico was possibly partly to maintain colony purity and 
avoid contamination of laboratory colonies with potentially diseased field samples. The 
observation of greater heterozygosity deficiency in most laboratory populations compared to 
the Teupasenti sample - a recent field isolation - thus indicates a recent population 
expansion that could mask the effect of population bottlenecks. Further, T. nigrescens are 
reared in jars, often with a starting population of 20 adults of around 10 females. Rearing 
insects in sealed rearing jars creates an effective population fragmentation that might lead to 
allelic drifts, especially if colonies from different jars are not mixed after every few 
generations between jars. This would thus create a series of slight bottlenecks fragmentation 
during rearing and a significant founder effect at every sub-culturing. Consequently, loss of 
rare alleles and increased homozygosity may lead to the over-expression of deleterious 
genes, depletion of rare but important fitness linked alleles and increased sensitivity to 
environmental fluctuations or pathogens, reducing the potential establishment and success of 
eventual released individuals. Indeed, our data shows.a general reduction in numbers and 
allelic richness among laboratory populations compared to the field populations. 
The genetic mixture analysis revealed a clear genetic structure of T. nigrescens following the 
geographical distribution. Two methods of estimating the number of populations gave 
consistent results, with an indication of existence of a hierarchical structure. Five distinct 
177 
populations were estimated with the Oaxaca population showing a mixed ancestry at this 
level. Although the estimation of K in such a case is informative, it is more important to 
figure out what level ofK is most biologically informative (Evanno et al., 2005; Pritchard et 
al., 2007). For example, although structure revealed hidden genetic substructure and 
demographic history events in cat and insect populations (Beaumont et al.; 2001, Liu et al., 
2008), it could not account for certain known demographic events among Dexter cattle 
breeds (Bray et al., 2009). In this study, the assumption of two clusters (K = 2) grouped 
together all populations from Mexico with those of Mexican origin (KARI, Kiboko and 
"Store" and Mombasa), while those originating from all other locations formed the other 
cluster. At K = 3, the southern cluster was subdivided into two sub clusters: Honduras 
(Gualaso, Teupasenti and Yoro) and Costa Rica (Benin, Ghana and Malawi). This reflected 
the correct countries of initial origin of these populations. At K = 4, Mexican populations 
were further divided into two clusters: the first cluster contained samples from Batan, 
Tlaltizaspan and Oaxaca (recently sampled from the field), while the second included 
samples of Mexican origin released in Kenya in the 1990s. The distinction between the Costa 
Rican populations released in Ghana, Zambia and Malawi from the mother population 
maintained in Benin with little evidence frequent field introgressions was evident at K = 5. 
Beyond K = 5, subdivisions occurred but the Honduras populations always ended up in the 
same cluster. 
This hierarchical mode of clustering is informative. First, one is able to detect the origin of 
samples used, corroborating the reconstruction of the history of biological control in Africa. 
Also, there was evidence of allelic diversity shifts in culture after separation for just a few 
years (Hufbauer et al, 2004; Lloyd et al, 2005). Changes in allelic diversity and 
composition have been observed in laboratory colonies of Cotesia flavipes (Omwega and 
178 
Overholt, 1996) and Aphidius ervi (Hufbauer et aL, 2002; 2004). However, the association 
between these neutral changes and fitness-related traits is debatable (van Tienderen et aL, 
2002; Bekessy et aL, 2003). Due to logistical stringencies and because of their sheer rarity in 
the field, natural enemies are often collected as small subpopulations ill-representing the full 
genetic diversity in the source population (Lloyd et aL, 2Q05). Most of the populations 
released in Africa have been separated from the mother populations for about 20 years now 
(Giles et aL, 1996; Meikle et aL, 2003). This period corresponds to less than 300 generations 
of T. nigrescens, which is comparatively short in species evolution. The effect of genetic 
drift and founder effects may be more in play considering the rearing protocol of fragmented 
populations. Other than a small study population from mid-altitude Mexico introduced into 
the laboratories of IITA in 1990s (W. M. Meikle, USDA, Weslaco, USA, personal 
communication), there is little evidence of introgression of the gene pool using samples 
collected over time or from different geographical locations. This could be the main 
contributing factor to the hierarchical structure and maintenance of genetic distinctiveness of 
the populations recovered. 
The practical role of release sizes in the success of biological control establishments is 
controversial (Hopper and Roush, 1993; Gfrevstad, 1999; Memmott et aL, 2005). The genetic 
diversity of natural enemies is a key parameter in biological control and a possible 
determinant of their eventual establishment, ecological fitness and effectiveness. Rapid 
adaptation to local environmental conditions and eventual effectiveness of may not only 
depend on the suitability of the species introduced, but also the existence of well adapted 
geographical biotypes (Diehl and Bush, 1984). In fact, it has been recognised that variability 
of natural enemies is invaluable in biological control in providing pools of genes on which 
selection occurs (Roush, 1990; Roderick and Navajas, 2003). The process of collection, 
179 
quarantine rearing, importation and release may impose series of bottlenecks leading to loss 
of genetic variation (Hopper and Roush, 1993; Hopper et al., 1993; Lloyd et al., 2005; 
Fauvergue et al., 2007). Often, one starts with a small population losing on potentially 
important rare alleles (Nei et al., 1975; Bigler, 1992). Rearing protocols aiming at producing 
pure populations, long time laboratory culturing and subdivision may reduce the internal 
diversity of the natural enemies. Where information was available, new colonies and releases 
were shown to be founded by between 200 and 5000 individuals respectively (Anonymous, 
1999; P. Likhayo, KARI; H.Z. Irmgard, IITA, personal communication). On the face value, 
these numbers meet the minimum viable population size criterion for diploid organisms, but 
such diversity must take into account underlying genetic variability (Rai, 2003). However, 
our data show induced systematic bottleneck/expansion or drift effects perhaps during initial 
isolation from the field or by rearing in fragmented non-contiguous subpopulations. 
180 
7.6 References 
Adams, K. L. and Palmer, J. D. (2003). Evolution of mitochondrial gene content: gene loss 
and transfer to the nucleus. Molecular Phylo genetics and Evolution 29: 380 - 395. 
Anonymous (1999). Management of Maize Pests and Diseases, Annual Report, Plant Health 
Management Division, 1999. International Institute of Tropical Agriculture, Cotonou, 
Benin. 
Bacigalupe, D. L. (2008). Biological invasions and phenotypic evolution: a quantitative 
genetic perspective. Biological Invasions doi 10.1007/sl0530-0089411-2. 
Beaumont, M., Barratt, E. M., Gottelli, D., Kitchener, A. C , Daniels, M. J., Pritchard, 
J. K. and Brudford, M. W. (2001). Genetic diversity and introgression in the 
Scottish wildcat, Molecular Ecology 10: 319 — 336. 
Bekessy, S. A., Ennos, R. A., Burgman, M. A., Newton, A. C. and Ades, P. K. (2003). 
Neutral DNA markers fail to detect genetic divergence in an ecologically important 
trait. Biological Conservation 110: 267—275. 
Belkhir, K., Borsa, P., Chikhi, L., Raufaste, N. and Bonhomme, F. (2002). GENETIX, 
Version 4.04 logiciel sous Windows™ pour la Genetique des populations. 
Laboratoire Genome, Populations, Interactions CNRS UMR 5000. Universite 
Montpellier II, Montpellier, France. 
Bellows, T. S., Fisher, T. W. and Cattairone, L. E. (1999). Handbook of Biological 
Control: Principles and Application of Biological Control. Academic Press, 1046pp. 
Bigler, F. (1992). Quality Control in insect rearing systems, in: Ochieng-Odero, J. P. R. (ed) 
Techniques of Insect Rearing for the Development of Integrated Pest and Vector 
Management Strategies (pp 189 - 210). Proceedings of the International Group 
Training Course on Techniques of Insect Rearing for the Development of Integrated 
Pest and Vector Management, ICIPE, Nairobi. 
181 
Bonhomme, M., Blancher, A., Cuartero, S., Chikhi, L. and Crouau-Roy., B. (2008). 
Origin and number of founders in an introduced insular primate: estimation from 
nuclear genetic data. Molecular Ecology 17: 1009 - 1019. 
Bray, T. C , Chikhi, L., Sheppy, A. J. and Bruford, M. W. (2009). The population genetic 
effects of ancestry and admixture in a subdivided cattle breed. Animal Genetics 40: 
393-400 . 
Bull, L. N., Pabon-Pena, R. C. and Freimer, N. B. (1999) Compound Microsatellite 
Repeats: Practical and Theoretical Features. Genome Research 9:830 — 838. 
Cornuet, J. M. and Luikart, G. (1996). Description and power analysis of two tests for 
detecting recent population bottlenecks from allele frequency data. Genetics 144: 
2001-2014. 
Cornuet, X, Piry, S., Luikart, G., Estoup, A., and Solignac, M. (1999). New methods 
employing multilocus genotypes to select or exclude populationsas origins of 
individuals. Genetics 153: 1989-2000. 
Dakin, E. E. and Avise, J. C. (2004). Microsatellite null alleles in parentage analysis. 
Heredity 93: 5QA-5Q9. 
Diehl, S. R. and Bush, G. L. (1984). An evolutionary and applied perspective of insect 
biotypes. Annual Review of Entomology 29: 471 — 504. 
Ellegren, H. (2004). Micro satellites: Simple sequences with complex evolution Nature 
Reviews — Genetics 5: 435 - 445. 
Evanno, G., Regnaut, S. and Goudet, J. (2005). Detecting the number of clusters of 
individuals using the software STRUCTURE: a simulation study. Molecular Ecology 
14:2611-2620. 
182 
Falush, D., Stephens, M. and Pritchard, J. K. (2003). Inference of the population structure: 
Extensions to linked loci and correlated allele frequencies. Genetics 164: 1567 -
 1587. 
Falush, D., Stephens, M. and Pritchard, J. K. (2007). Inference of the population structure 
using multilocus genotype data: dominant markers and null alleles. Molecular 
Ecology Notes 1:51 A- 578. 
Fauvergue, X., Malausa, J. -C., Giuge, L. and Courchamp, F. (2007). Invading 
parasitoids suffer no Alee effect: a manipulative field experiment Ecology 88: 2392 -
 2403. 
Garza, J. C. and Williamson, E. G. (2001). Detection of reduction in population size using 
data from microsatellite loci. Molecular Ecology 10: 305 - 318. 
Giles, P. H., Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). 
Release and establishment of the predator Teretriosoma nigrescens Lewis for the 
biological control of Prostephanus truncatus (Horn) in Kenya. African Crop Science 
Journal,, 4: 325 — 337. 
Goudet, J. (2002). FSTAT, a program to estimate and test gene diversities and fixation 
indices (version 2.9.3). Available from http://www.unil.ch/izea/softwares/fstat.html. 
Grevstad, F. S. (1999). Experimental invasions using biological control introductions: The 
influence of release size on the chance of population establishment, Biological 
Invasions 1:313—323. 
Handler, A. M. and Beeman, R. W. (2003). United States Department of Agriculture 
Agricultural Research Service: advances in the molecular genetic analysis of insects 
and their application to pest management Pest Management Science 59: 728-735. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
183 
predatory beetle Teretrius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 — 306. 
Holt, R. D. and Hochberg, M. E. (1997). When is biological control evolutionarily stable 
(or is it)? Ecology 78: 1673 - 1683. 
Hopper, K. R. and Roush, R. T. (1993). Mate finding, dispersal, number released, and the 
success of biological control introductions. Ecological Entomology 18: 321-331. 
Hopper, K. R., Roush, R. T. and Powell, W. (1993). Management of genetics of biological-
 control introductions. Annual Review of Entomology 38: 27—51. 
Hufbauer, R. A. (2002). Evidence for non adaptive evolution in parasitoid virulence 
following a biological control introduction. Ecological Applications 12: 66 —78. 
Hufbauer, R. A., Bogdanowicz, S. M. and Harrison, R. G. (2004). The population genetics 
of a biological control introduction: mtDNA and microsatellite variation in native and 
introduced populations of Aphidius ervi, a parasitoid wasp. Molecular Ecology 13: 
337-348. 
Kalinowski, S. T., Taper, M. L. and Marshall, T. C. (2007). Revising how the computer 
program CERVUS accommodates genotyping error increases success in paternity 
assignment. Molecular Ecology 16: 1099-1006. 
Lawler, R. R. (2008). Testing for a historical population bottleneck in Wild Verreaux's 
? Sifaka (Propithecus verreauxi verreauxi) Using Microsatellite Data. American 
Journal of Primatology 70: 990 - 994. 
Liu, Y.D., Wu, K. M. and Guo, Y. Y. (2009). Population structure and introduction history 
of the pink bollworm, Pectinophora gossypiella, in China. Entomologia 
Experimentalis et Applicata 130: 160 — 172. 
Lloyd, C. J., Hufbauer, R. A., Jackson, A., Nissen, S. J.and Norton, P. N. (2005). Pre-
 and post-introduction patterns in neutral genetic diversity in the leafy spurge gall 
184 
midge, Spurgia capitigena (Bremi) (Diptera: Cecidomyiidae). Biological Control 33: 
153-164. 
Lockwood, J. L., Cassey, P. and Blackburn, T. (2005). The role of propagule pressure in 
explaining species invasions. Trends in Ecology and Evolution 20: 223 - 228.Lowe, 
A., Harris, S. and Ashton, P. (2006). Ecological Genetics: Design, Analysis and 
Application. Blackwell Publishing, 326pp. 
Loxdale, H. D. and Lushai, G. (1998). Molecular markers in entomology. Bulletin of 
Entomological Research 88: 557 - 600. 
Luikart, G. and Cornuet, J. M. (1998). Empirical evaluation of a test identifying recently 
bottlenecked populations from allele frequency data. Conservation Biology 12: 228-
 237. 
Luikart, G., AUendorf, F. W., Cornuet, J. M. and Sherwin, W. B. (1998). Distortion of 
allele frequency distributions provides a test for recent population bottlenecks. 
Journal of Heredity 89: 238-247. 
Meikle, W. G., Rees, D. and Markham, R. H. (2003). Biological control of the larger grain 
borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest 
Management Reviews 7: 123 - 13 8. 
Memmott, J., Craze, P. G., Harman, H. M., Syrett, P. and Fowler, S. V. (2005). The 
effect of propagule size on the invasion of an alien insect. Journal of Animal Ecology 
. 74:50-62. 
Nei, M. (1973). Analysis of gene diversity in subdivided populations. Proceedings of the 
National Academy of Sciences of the USA 70: 3321 — 3323. 
Nei, M. (1978). Estimation of average heterozygosity and genetic distance from small a 
number of individuals. Genetics 89: 583 —590. 
185 
Nei, M., Maruyama, T. and Chakraborty, R. (1975). The bottleneck effects and genetic 
variability in populations. Evolution 29: 1-10. 
Omondi, A. B., Orantes, L. C , van den Berg, J., Masiga, D. and Schulthess, F. (2009). 
Isolation and characterization of microsatellite markers from Teretrius nigrescens 
Lewis (Coleoptera: Histeridae), predator of the storage pest Prostephanus truncatus 
(Horn) (Coleoptera: Bostrichidae). Molecular Ecology Resources 9: 1236 - 1239. 
Omwega, C. O. and Overholt, W. A. (1996). Genetic changes occurring during laboratory 
rearing of Cotesia flavipes Cameron (Hymenoptera: Braconidae) an imported 
parasitoid for the control of graminaceous stem borers in Africa. African Entomology 
4 :231 -237 . 
Piry, S., Luikart, G. and Cornuet, J. M. (1998). BOTTLENECK: A program for detecting 
recent effective population size reductions from allele frequency data, available from: 
http://www. ensam. inra.fr/URLB  
Piry, S., Luikart, G. and Cornuet, J. M. (1999). BOTTLENECK: A computer program for 
detecting recent effective population size reductions from allele frequency data, 
available from: Journal of Heredity 90: 502 - 503. 
Pritchard, J. K., Stephens, M. and Donnely, P. (2000). Inference of population structure 
from genotype data. Genetics 155: 945 - 959. 
Pritchard, J. K., Wen, X. and Falush, D. (2007). Documentation for structure software: 
version 2.2 http//pritch.bsd.uchicago.edu/software. 
Rai, U. K. (2003). Minimum sizes for viable population and conservation biology. Our 
Nature 1: 3-9. 
Raymond M. and Rouset F. (1995). GENEPOP (Version 3.4): population genetics software 
for exact tests.and ecumenism. Journal of Heredity 86: 248 - 249. 
186 
Richard R. L. (2008). Testing for a historical population bottleneck in wild Verreaux's 
Sifaka (Propithecus verreauxi verreavxi) using microsatellite data. American Journal 
of Primatology 70: 990-994. 
Roderick, G. K. and Navajas, M. (2003). Genes in new environments: Genetics and 
evolution in biological control. Nature Reviews - Genetics 4: 889-899. 
Roush, R.T. (1990). Genetic considerations in the propagation of entomophagous species. 
In: Baker, R.A., Dunn, P.E. (Eds.), New Directions in Biological Control, 
Alternatives for Suppressing Agricultural Pests and Diseases. Alan R. Liss, New 
York, pp. 373-387 . 
Sakai, A. K., Allendorf, F. W., Holt, J. S., M. Lodge, D. M., Molofsky, J. With, K A., 
Baughman, S. Cabin, R. J. Cohen, J. E., Ellstrand, N. C , McCauley, D. E., 
O'Neil, P., Parker, I. M., Thompson, J. N. and Weller, S. G. (2001). The 
population biology of invasive species. Annual Review of Ecology and Systematics, 
32:305-32. 
Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning, a laboratory 
manual. 2ndedn. Cold Spring Harbour Laboratory Press, New York. 
Thao, M. L., Baumann, L. and Baumann, P. (2004). Organisation of the mitochondrial 
genomes of whiteflies, aphids and psyllids (Hemiptera, Stenorrhyncha). BioMed 
Centi'dl Evolutionary Biology 4: 25 - 30. 
Thao, M. L., Baumann, L. and Baumann, P. (2004). Organisation of the mitochondrial 
genomes of whiteflies, aphids and psyllids (Hemiptera, Stenorrhyncha). BioMed 
Central Evolutionary Biology 4: 2 5 - 3 0 . 
van Oosterhout, C, Hutchinson, W. F, Willis D. P. M. and Shipley, P. (2004). Micro-
 Checker: Software for identifying and correcting genotyping errors in microsatellite 
data. Molecular Ecology Notes 4: 535 - 538. 
187 
van Tienderen, P. H., de Haan, A. A., van der Linden, C. G. and Vosman, B. (2002). 
Biodiversity assessment using markers for ecologically important traits. Trends in 
Ecology and Evolution 17: 577-582. 
Weir, B. S. and Cockerham, C. C. (1984). Estimating F-statistics fort the analysis of 
population structure. Evolution 38: 1358-1370. 
Williams, D. A., Overholt, W. A., Cuda, J. P. and Hughes, C. R. (2005). Chloroplast and 
micro satellite DNA diversities reveal the introduction history of Brazilian peppertree 
(Schinus terrebinthifolus) in Florida. Molecular Ecology 14: 3643 — 3656. 
Zayed, A., Constantin, S. A. and Packer, L. (2007). Successful biological invasion despite 
a severe genetic load. PLoS ONE 2: e868. doi:10.1371/journal.pone.0000868. 
Zhang, D.X. and Hewitt, G. M. (2003). Nuclear DNA analysis in genetic studies of 
populations: practice, problems and prospects. Molecular Ecology 12: 563 — 584. 
188 
C H A P T E R E I G H T 
General Discussion and Conclusions 
The sustainable management of the invasive LGB in Africa depends on finding a suite of 
control measures and formulation of decision-making tools applicable in diverse 
environments. These should result in significant reduction in pest populations, yet be 
compatible with other post-harvest pest management measures. The correlation between 
weather conditions and the survival, abundance and flight activity of the larger grain borer 
has been studied before (Fadamiro and Wyatt, 1995; Hill et al, 2003; Hodges et al, 2003). 
Models of the flight activity generally identify warm humid conditions as being both suitable 
for development and initiation of flight activity of the pest (Rees et al., 1990; Tigar et al, 
1994; Hodges et al, 2003). Other than Tigar et al (1994) little modelling and prediction of 
the flight activity of the predator has been done. It has been documented that weather 
conditions may trigger flight in P. truncatus either directly or through their contribution to 
rapid population build-up leading to crowding (Fadamiro and Wyatt, 1995; Hodges et al, 
2003). The findings of this study generally concurred with previous workers on the seasonal 
patterns of the LGB flight. The differences in exact model parameters under different 
ecological conditions suggest that local adaptation may distort the predictive ability of 
models previously developed. Continuous surveillance of the pest through the national 
agricultural research systems is necessary. 
Although T. nigrescens established in Kenya (around the coast and semi-arid lower eastern 
region), its contribution to the control of the LGB in most of the areas sampled is doubtful. 
Giles et al (1996), Nang'ayo (1996) and Hill et al (2003) observed a marked reduction of 
the pest's flight activity following release of the predator into a largely maize deficit area. 
However, Hill et al. (2003) recorded very low subsequent T. nigrescens flight activity. 
189 
Fluctuation in the LGB population may be more a result of meteorological parameters than 
the action of the predator. It is unlikely that a predator that is barely detectable would be an 
important control agent of a ubiquitous pest. Population fragmentation, post release 
extirpation and the dearth of safe establishment habitats could have contributed to these 
observations (Grevstad, 1999; Shea, and Possingham, 2000) and should be accommodated in 
further releases. A more numerically and genetically aggressive release strategy should also 
be adopted. In Kenya, about 4900 beetles in total were released per location, compared to 
135,000 in Togo and 100,000 in Zambia. To increase the chances of success of the predator, 
a wider genetic diversity from across the predator's native range and ecological modelling 
and mapping studies should be incorporated. 
The existence of wild populations of T, nigrescens in Kenya (e.g. Mombasa, Store and Field 
populations) is fascinating. The virtual absence of the predator in western Kenya 
undetectable is informative. It would be interesting to know the selective barriers to 
westward dispersal of the predator, since the LGB has subsequently established in these. 
Climatic and topographical factors may be responsible (Giles et al., 1996). Possibly, the 
introduction of the LGB into the western part of the country was through independent 
anthropogenic causes such as commerce. Studying the population differentiation of the LGB, 
using molecular markers might elucidate the origin of the infesting populations countrywide. 
The role of micro evolution in the establishment of the LGB in Africa should be investigated. 
The levels of grain damage and environment of the pest in Africa are quite different from 
that observed in the native range (Rees et al, 1990; Farrel and Schulten, 2002). The 
successful establishment of a pest often depends on its ability to adapt to new ecological 
conditions (Hufbauer and Roderick, 2005). Rapid changes in ecology and genetic diversity 
190 
are recognised as the key to success of an invasion (Lee, 2002). Rapid selection of a subtype 
of the invader occurs, leading to the survival of better adapted individuals following the lag 
phase of the invasion. Such changes probably favoured adaptation to the farm-store 
environment unlike the semi wild founder populations. In this study, pest recoveries were 
higher nearer roads, markets and human dwellings. One issue still unresolved is the main 
hosts of the LGB in the wild (Nang'ayo et al, 1993; Nansen et ah, 2001). The role of local 
fauna in the perennation of the pest in Africa should be studied to incorporate sound 
environmental management in its control. 
The hypothesis that biologically discontinuous biotypes of the predator exist cannot be 
upheld in this study, but could be verified with other studies suggested below. Guntrip et al. 
(1996), reported significant biological differences between LGB populations from Mexico 
and Costa Rica, but this cannot be said of T. nigrescens from the two countries. One cannot 
however, rule out the possible existence of strain of T. nigrescens along the entire 
geographical expanse of the pest. Although the observations on the pest's temperature and 
humidity preference were not conclusive, difference in fitness especially under marginal 
temperature levels (21 °C and 30 - 33 °C) was observed. It would be important to study the 
effects of these factors on the developmental and life table characteristics of the pest. 
Constant temperature studies provide useful means of modelling populations and predicting 
the thermal requirements in their development. Such studies have provided useful data for 
pest management decisions such as pest predictions and timing of control interventions (e.g. 
Jaramillo et al., 2009). These studies are especially useful when variations are distinct, but 
adjustment to these conditions may mask inters train differences. Also, nature does not 
provide such constant conditions. Thus the most important effects of the environment would 
191 
be those extremes that elicit migration or just cause a significant increase in mortality, 
epizootics, etc. Stimulus mediated gene expression studies would provide a more precise 
means of understanding the reaction of such extremes on the pest and natural enemy 
populations (e.g. Fadamiro and Wyatt, 1995; Hodges et aL, 2003). Heat shock proteins and 
heat shock factors are molecular chaperons known to mediate such responses. The 
quantitative expression of this family of genes open a molecular estimate of the possible 
stress tolerance of living organisms (Malmendal et aL, 2005) and could provide a new 
frontier towards understanding possible differences in the adaptability of T. nigrescens 
populations. 
This study established the existence of two mitochondrial lineages of the predator associated 
with geographical zones. The variations associated with such ancestry may not be apparent 
now, but suggest ancient (and possibly present) genetic differentiation mediated by strong 
ecological fragmentation. The utilization of this variability across the natural range of the 
pest may lead to the eventual recovery of better adapted populations. Only two populations 
of T. nigrescens from limited geographical and temporal isolation efforts were released 
separately in Africa (Meikle et aL, 2003; Hill et aL, 2003; Schneider et aL, 2004). Eventual 
culturing, quarantine and release possibly reduced the variability and adaptability further. 
There is need to exploit the potential genetic diversity of T. nigrescens and other associated 
natural enemies for the control of the LGB. A potentially rich source of such diversity would 
be the contact hybrid zone from central Mexico to northern Honduras. Due to topographical 
fragmentation, it is possible that several locally adapted subpopulations exist in this zone. 
Latitudinal similarities with most of Africa also make populations from this area potentially 
better adapted to local conditions than more subtropical populations from Mexico. In fact the 
192 
Costa Rica population that established in West Africa shared similar latitudinal origin with 
the area of release. 
Significant genetic changes in T. nigrescens populations were observed in this study 
representing a period of only 300 generations of laboratory culturing. In nature, this period is 
too short to engender significant evolution (Lee, 2002). However, laboratory rearing exposes 
cultures to stochastic and random effects of genetic drift, bottlenecks, modified adaptive 
selection pressures and fragmentation of populations. Some of these changes may result in 
loss of fitness similar to the classical screw worm rearing case (Bellows et al., 1999). Diploid 
organisms are a lot more sensitive to such changes (Roderick and Navajas, 2003). High 
mortality following laboratory release, ecological fragmentation and difficulty to find mates 
after release pose further stringent bottlenecks in the field. The microsatellite markers 
developed in this study will be useful for monitoring the rearing, establishment, dispersal and 
population changes in the new released T. nigrescens populations. Consequently, timely 
intervention and informed decision-making in biological control through the management of 
genetic diversity would be possible. 
Clinal variation along the geographical range exists and may be associated with distinct 
fitness and predatory characteristics. Genetic recombination would useful in harnessing this 
diversity for sustainable biological control. Strain interbreeding or joint releases and 
molecular monitoring may support this approach. Teretrius nigrescens is a well adapted 
predator of the LGB able to effect better control in the wild or when released early. Although 
T. nigrescens reduces grain damage and limits LGB population growth, the damage observed 
on maize is still much higher than economically tolerable levels. Integrated approaches to 
LGB control with incorporating management of grain harvesting and processing, storage, 
193 
monitoring and forecasting, biological, physical and chemical control provide a more 
pragmatic approach. Tetetrius nigrescens would be used in controlling populations off­
 season and off farm. Collaborative activities to achieve cost effectiveness in fresh recoveries 
of more diverse populations would enable the utilisation of the full genetic potential of the 
predator. 
Conclusion and recommendations 
The initial assumption of this study visualised the success of T. nigrescens in Africa as a 
direct product of ecological suitability of the strains of the predator released. The variation in 
the species was thought to be distinct, perhaps comparable to well characterised biotypes. 
This study has shown that while genetic diversity exists between different geographic 
populations of this species, that variation may be graded and closely tied to local ecological 
conditions including barriers to migration and population interactions. Also, the methods 
used in field isolation, rearing and releases of the predator had a role to play in the success of 
biological control. The predator established in both releases, but effective control required 
that it locates most breeding populations of the pest. This might have been more difficult in 
fragmented landscapes, marginal regions and areas where release sizes were small (like in 
Kenya) compared to west Africa, where propagule pressure was much higher. Published 
results from West Africa reveal that T. nigrescens plays a role in the integrated control of the 
LGB. This study therefore recommends steps to increase the utility of the full genetic 
diversity of T. nigrescens in the integrated sustainable control of the LGB in other parts of 
Africa. 
194 
Recommendations 
• Greater sampling and characterisation of the populations of the larger grain borer 
across its natural range (Southern USA to northern Columbia/Peru) should be done 
and compared with the invasive populations in Africa, to locate the critical sources of 
the African invasion. 
• Characterisation of T. nigrescens populations from more sites in its native range may 
give a clearer picture of the extent of genetic diversity there, and perhaps, reveal 
more efficacious strains. 
• Genetic studies and ecological modelling above could be used to predict the regional 
ecological suitability of these species in Africa and to guide new releases and 
monitoring. 
• A direct study of the expression of stress-related genes and proteins such as heat 
shock factors, heat shock protein genes may directly reveal the adaptability of the 
strains to changing climatic conditions. This data could also guide laboratory 
breeding of laboratory strains for release. 
195 
8.1 References 
Bellows, T. S., Fisher, T. W. and Cattairone, L. E. (1999) Handbook of Biological 
Control: Principles and Application of Biological Control. Academic Press, 1046pp. 
Fadamiro, H. Y. and T. D. Wyatt. (1995). Flight initiation of Prostephanus truncatus in 
relation to time of day, temperature, relative humidity and starvation. Entomologia 
Experimentalis et Applicata 75: 273-277. 
Farrel G. and Schulten G. G. M. (2002). The larger grain borer in Africa; a history of 
efforts to limit its impact. Integrated Pest Management Reviews 7: 61 — 84. 
Giles, P. EL, Hill, M. G., Nang'ayo, F. L. O., Farrell, G. and Kibata, G. N. (1996). 
Release and establishment of the predator Teretriosoma nigrescens Lewis for the 
biological control of Prostephanus truncatus (Horn) in Kenya. African Crop Science 
Journal 4: 325 — 337. 
Grevstad, G. (1999). Experimental invasions using biological control introductions: the 
influence of release size on the chance of population establishment, Biological 
Invasions 1: 313 — 323 
Guntrip, J., Sibly, R. M. and Smith R. H. (1996). A phenotypic and genetic comparison of 
egg to adult life-history traits between and within two strains of the larger grain 
borer, Prostephanus truncatus Horn (Coleoptera: Bostrichidae). Journal of Stored 
Products Research 32: 213 — 223. 
Hill, M. G., Nang'ayo F. L. O. and Wright, D. J. (2003). Biological control of the larger 
grain borer Prostephanus truncatus (Coleoptera: Bostrichidae) in Kenya using a 
predatory beetle Teretrius nigrescens (Coleoptera: Histeridae). Bulletin of 
Entomological Research 93: 299 — 306. 
196 
Hodges, R. J Addo, S. and Birkinshaw, L. (2003). Can observation of climatic variables be 
used to predict the flight dispersal rates of Prostephanus truncatusl Agricultural and 
Forest Entomology 5: 123 — 135. 
Hufbauer, R. A. and Roderick, G. K. (2005). Microevolution in biological control: 
mechanisms, patterns, and processes. Biological Control 35: 227—239. 
Jaramillo, J., Chabi-Olaye, A., Kamonjo, C , Jaramillo. A., Vega, F. E., Poehling, H.-
 M.and Borgemeister, C. (2009). Thermal tolerance of the coffee berry borer 
Hypothenemus hampei: predictions of climate change impact on a tropical insect 
Pest. PLoS ONE 4: e6487. doi:10.1371/journal.pone.0006487. 
Lee, C. E. (2002). Evolutionary genetics of invasive species. Trends in Ecology and 
Evolution 17r: 386 - 3 9 1 . 
Malmendal, A., Overgaard, J., Bundy, J. G. S0rensen, J. G., Niels Chr. Nielsen, N. C , 
Volker Loeschcke, V. and Holmstrup, M. (2005). Metabolomic profiling of heat 
stress: hardening and recovery of homeostasis in Drosophila. American Journal of 
Physiology - Regulatory, Integrative and Comparative Physiology 291: 4205-^-212 
Meikle, W. G., Rees, D. and Markham, R. H. (2003). Biological control of the larger grain 
borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Integrated Pest 
Management Reviews 7: 123-138 . 
Nang'ayo, F. L. O. (1996). Ecological Studies on The Larger Grain Borer in Savanna 
Woodlands of Kenya. PhD Dissertation, University of London, 179pp. 
Nang'ayo, F. L. O. Hill, M. G., Chandi, E. A., Chiro, C. T., Nzeve, D. N. and Obiero, J. 
(1993). The natural environment as a reservoir of the larger grain borer Prostephanus 
truncatus (Horn) (Coleoptera: Bostrichidae) in Kenya. African Crop Science Journal 
1 : 3 9 - 4 7 . 
197 
Nansen, C , Korie, S., Meikle, W. G. and Hoist, N. (2001). Sensitivity of Prostephanus 
truncatus (Coleoptera: Bostrichidae) flight activity to environmental variables in 
Benin, West Africa. Environmental Entomology 30: 1135-1143 
Rees, D. P., Rivera, R. R. and Rodriguez, F. S. H. (1990). Observations on the ecology of 
Teretriosoma nigrescens (Lewis) (Col., Histeridae) and its prey Prostephanus 
truncatus (Horn) (Col., Bostrichidae) in the Yucatan peninsula, Mexico. Tropical 
Science 30: 153-165 . 
Roderick, G.K. and Navajas, M. (2003). Genes in new environments: Genetics and 
evolution in biological control. Nature Reviews - Genetics 4: 889-899. 
Schneider, H., Borgemeister, C. Setamou, M., Affognon, H., Bell, A., Zweigert, M. E., 
Poehling, H. and Schulthess, F. (2004). Biological control of the larger grain borer 
Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) by its predator Teretrius 
nigrescens (Lewis) (Coleoptera: Histeridae) in Togo and Benin. Biological Control 
30 :241-255 . 
Shea, K. and Possingham, H. (2000). Optimal release strategies for biological control 
agents: An application of stochastic dynamic programming to population 
management. Journal of Applied Ecology 37: 77 — 86. 
Tigar, B. X, Osborne, P. E., Key, G. E., Flores-S., M. E. and Vazquez-Arista, M. (1994). 
Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) by 
its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bulletin of 
Entomological Research 84: 555 — 565. 
198 
APPENDIX A: Variation in cloned sequences of Teretrius nigrescens internal transcribed spacer sequences (1TS1, ITS2) with flanking and intervening 
ribosomal RNA gene sequences. 
18S (SSU) ITS1 
hA 
hB 
hC 
hD 
hE 
hF 
hG 
hH 
hi 
hJ 
hK 
hL 
GGAAGTAAAA GTCGTAACAA GGTTT^CGTA liUTCjAAUCTG CGGAAGGATC ATTAACGTTT TTCTCCGCCG CCCTGCTGGT GCGCGCGCGA GAAACAACGC CAACTTATCG TCACGGGACG 120 
A 12 0 
120 
C A 120 
C 120 
120 
120 
120 
120 
C T 120 
120 
A 120 
hA 
hB 
hC 
hD 
hE 
hF 
hG 
hH 
hi 
hJ 
hK 
hL 
ITS1 
CCGTGCTCAG GTTTACCAGC GAGTGCCGCG AAGATCGGCA CACGCGGTCC CCGGCGGCCC GAAGACACCT TCTCTCGAAC GCACGACGCC AGCTTGGCGT GGTGGCTCGA GCGGCAGCAG 240 
240 
240 
240 
A 240 
240 
G T 240 
240 
240 
A 240 
G T 240 
240 
rrsi 
hA CGCGAGAACG CGAACAGAAG AAAGACCCCC GTCGACCGTG CTTCGCGTGA CACGCCGAGC GCGTG 
hB 
hC 
hD 
hE 
hF 
hG 
hH 
hi TG.C. .C GA 
h J 
hK 
hL 
GTC CGCCCGGACG GAGTGATTCG TCTTTCTCGT TCGTCTCGTT CTGCTCCTTC 358 
358 
358 
358 
358 
358 
358 
358 
360 
358 
358 
358 
(GT)5 (low variability) (ATTCGTCTT)5 (non variable) 
199 
ITS1 
hA GCGTCATACG CCTCGCACAG GTACACGCAC ACTCTCGAAC GCAGTGCACG TGCGCCAGCT CGAGGACACC CCGCCGTCGC CCGAGATATT TCGCCAGACG CGCGCACCCT TTCCCGCCGC 4 78 
hB 
hC 
hE 
hF 
hH 
hi 
hJ 
hK 
hB 
478 
478 
hD C A .T 47£ 
47£ 
47£ 
hG T 4 7 8 
478 
480 
478 
478 
hL C A .T 478 
1TS1 
hA ATGTGTTTCT TTCGCAGCGG CAGCGGATCG GCGTGTCGTC GCTGTCGGCC CGGACGATGT GCGTGCGTGT TTTGTCGTCC TCGCCGGCGC ACACGTGCGT GCGCTCCGAC GTGTCCGCGT 598 
598 
hD 
hE 
hF 
hG 
hH 
hi 
hJ 
hK 
hL 
hD 
hE 
hF 
hG 
hH 
hi 
hC T 598 
598 
598 
598 
598 
598 
600 
598 
598 
598 
ITS1 
?4 
hA TTCGCTTGCG AAAAAACGAG ACGCGTCGGT CGCGCGCCCG TGCAATTTTT TTTTT AATT GACACAAGTA CAGAACTCGC TTCCGATGCG TTGCGAAAAA CTGTCGTTGA AAAGATTACC 717 
hB 
hC T 
717 
716 
717 
716 
716 
717 
716 
718 
h J T 718 
hK T 718 
hL 717 
(T)n variable 
200 5.8S 
hA CTGAACGGTG GATCACTTGG CTCGTGGGTC GATGAAGAAC GCAGCTAATT GCGCGTCTAC TTGTGAACTG CAGGACACAT GAACATCGAC ATTTCGAACG CACATTGCGG TCCTCGGACG 8 37 
hB 837 
hC 836 
hD 837 
hE 836 
hF 8 36 
hG 837 
hH 836 
hi A 838 
hJ A 838 
hK 838 
hL 837 
5.8S ITS 2 
? 
hA TCTCGTTCCT GGACCACGTC TGTCTGAGGG TCGTCCTCGT ATCAAAATAG CTCCGTGTCT CGCGCACGGG GATTCTTGGG GTCTCGAAGG TCCGTCCGAC CGACGTGCCC TTAAAACACG 957 
hB G 957 
hC 956 
hD 957 
hE 956 
hF C 956 
hG 957 
hH 956 
hi C 958 
hJ . .-, 958 
hK 958 
hL 957 
ITS 2 
hA CGCGCAACGC CGACAGACGT TGCGTGCGTC CTCGCGACGG AACGATAGCC TGCGTCGTTC GTTCTCGTTC GATCGCTAGG CCGCCGTCGC GTGTCGATAC GCGCCCCCCG CGAGCGCGCC 107 7 
hB G 1077 
hC G 1076 
hD G 1077 
hE G A 1076 
hF G 1076 
hG G 1077 
hH G A 1076 
hi G A 1078 
hJ 1078 
hK C 1078 
hL G 1077 
(TCGTTCGATC)n not variable 
ITS 2 
201 
hA GACGCTCTCA CGTGTCCGTG TTGCCGTGCG GCCTTTACGG TCCGGCCCTG CGACGTGAAG CGATCGGTGT CTGAGCGTTC GCGTGCGAGT GCCGTCGTAT CGTCAGCGTT ACGTACAACA 1197 
hB 1197 
hC 1196 
hD 1197 
hE 1196 
hF A G. . . 1196 
hG A A 1197 
hH 1196 
hi 1198 
hJ 1198 
hK 1198 
hL 1197 
ITS 2 28S (LSU) 
hA 
hB 
hC 
hD 
hE 
hF 
hG 
hH 
hi 
hJ 
hK 
hL 
ACAA TAT TATTATTATT ATTATTTAAC AACAATA 
CAA T 
C M . G . 
CAA. 
TTAT TTTATTGTAC GCATAAAACG CGACCTCAGA TCAGGCGAGA TCACCCGCTG AATTTAAGCA TATCAATAAG 130 5 
GTA ATA C.G. ,.G T 1299 
ATG ATA C G 1295 
ATA ATA C.G. ,.G 12 99 
ATA -- 1298 
ATA A 1295 
ATA 1296 
ATA ----- 12 98 
ATA -- 1300 
ATA 1300 
ATA ATAATACC . . ..G 1300 
ATA ATA C.G. ..G 1299 
(TAT)n'(CAA)n' (TAT)n very variable compound repeat 
Sequenced ~ 1300 bp region of the ITS 1 and ITS 2 from T. nigrescens with intervening 5.8S and flanking 18S and 28S rRNA sequences. The coding 
regions of the first sequence (hA) are underlined and labeled with the names based on sequences from Oslrinia sp. and Tetranynchus sp.; repeat regions are 
underlined. Below each batch of sequences, repeat motif indicated and the level of variability observed from 221 cloned sequences from T. nigrescens.