Investigation of the microbial diversity and functionality of soil in fragmented South African grasslands along an urbanization gradient
Van Rensburg, Jacobus Petrus Jansen
MetadataShow full item record
The diversity of microorganisms and the influence of their enzymatic activities in soil are critical to the maintenance of good soil health. Changes in these parameters may be the earliest predictors of soil quality changes, potentially indicating anthropogenic influences. The goal of this study was to investigate the soil microbial diversity and function of grasslands along an urbanization gradient. Soil samples were collected in the Potchefstroom municipal area, South Africa, at specific sites. Sampling sites were described as urban, suburban and rural - according to the V-I-S (Vegetation-Impervious surface-Soil) model of Ridd (1995). Soil samples were collected over a warmer, wet season (May) and a colder, dry season (August) over two years (2007 and 2008). Collected soil samples were characterised using certain physical and chemical parameters. Plant species composition and abundance were determined at each site, along with basic site data (soil compaction, percentage ground cover, percentage bare ground, percentage organic material present). The Shannon-Weaver diversity index was used to calculate biodiversity values for all the investigated sites regarding collected plant species composition. The microbial component of the soil was quantified and characterized using culture-dependent and culture-independent techniques. Culture-dependent techniques included the investigation of the aerobic heterotrophic bacteria and fungi. Organisms were plated out on different media, and the bacterial component was broadly grouped using morphology. Dominant organisms were identified by sequencing of PCR amplified 16S ribosomal DNA fragments. Shannon-Weaver index for bacterial diversity was determined for each of the sites. Denaturing gradient gel electrophoresis (DGGE) profiling of selected bacterial communities were also conducted. Microbial community function was determined using enzyme assays of five major groups of enzymes, namely (i) dehydrogenase; (ii) β-glucosidase; (iii) acid phosphatase, (iv) alkaline phosphatase and (v) urease. Plant species results were then brought into context with microbiological diversity and functionality results using multivariate statistics. Physical and chemical parameters of the collected soil samples revealed patterns present along the urbanization gradient. The pH values were mostly higher in the sub-urban and urban sites than in the rural sites. Electrical conductivity values were generally highest in the sub-urban sites. Plant species composition revealed trends along the urbanization gradient. Ordinations clearly grouped the plant species into rural, sub-urban and urban groups regarding plant species composition. Rural sites had the highest number of plant species. Shannon-Weaver values regarding the plant diversity supported the plant species composition data indicating higher plant diversity in the rural areas, followed by the sub-urban and the urban areas. Plant structural data indicated that forbs were most numerous in the rural sites, and less so in the urban sites. Higher average aerobic heterotrophic bacterial levels were present in the urban soil samples. The bacterial levels were lower in the sub-urban and rural soil samples. Subsequent identification of the dominant bacteria in the soil samples revealed organisms of the genus Bacillus dominated the aerobic heterotrophic bacterial communities in the soil samples. Bacillus species dominated the soil samples along the urbanization gradient. Shannon-Weaver indices based on culture-dependent methods indicated that urban sites had the highest biodiversity. These results could have been exaggerated, because of an overestimation of the number of bacterial morphotypes present in samples. Fungal levels were higher in the soil from samples collected at the rural samples sites. The culture-independent method (DGGE) was not optimized and inconclusive results were obtained. Enzyme assays revealed that potential dehydrogenase, β-glucosidase and urease activity followed a trend along the urbanization gradient, with urban samples registering the highest values and rural sites the lowest. Enzymes involved in carbohydrate catabolism (β-glucosidase and dehydrogenase) registered significantly higher potential activity in urban sites than the sub-urban and rural sites. The results could indicate that urban sites have the potential to lose carbon at higher rates than the rural sites. This aspect may need further investigation. Higher potential urease activity could indicate higher N-cycling in the urban soil environment. Ordination results for soil-, plant- and microbial diversity as well as microbial functionality indicated certain trends along the urbanization gradient. Plant species composition and structure data indicated that urbanization has a definite effect on the plant communities in the urban ecosystem. Results regarding aerobic heterotrophic bacteria populations and potential enzyme activity of the dehydrogenase, β-glucosidase (both active in the carbon cycle) and urease (active in the nitrogen cycle) illustrated clear trends along the urbanization gradient. In conclusion, results indicated that urbanization has an effect on plant species composition, and the population and function of aerobic heterotrophic bacteria and the fungal population. Furthermore, this study demonstrated the potential of using microbial diversity and activity as tools to investigate carbon utilization and storage along an urban-rural gradient.