Development and validation of an animal model of treatment-resistant depression with psychotic features

Abstract

Major depression (MD) affects up to 17% of the global population. While effective treatments for MD exist, an underlying interaction between genetic propensity and chronic and/or early life adversity may render these drugs ineffective in managing the disorder. Treatment resistant depression (TRD) is a growing problem with various risk factors. Although its underlying biology remains elusive, comorbid psychosis is known to contribute to its development. MD with psychotic features, also known as psychotic depression (MDpsy) is one such disorder wherein the patient presents with co-occurring depressive and psychotic symptoms. MDpsy is characterised by depressive symptoms that are more severe than those occurring in MD, but similar to SCZ and bipolar disorder (BD), while presenting psychosis is generally less severe than in SCZ. This disorder is often misdiagnosed as MD or schizophrenia, resulting in patients not receiving appropriate treatment which further contributes to the development of treatment resistance. While antidepressant or antipsychotic monotherapy is ineffective, a combination of fluoxetine (FLX) and olanzapine (OLZ), originally approved by the United States Food and Drug Administration (FDA) for the treatment of TRD, has proven useful in treating MDpsy. A lack of animal models of MDpsy has hindered research and development into new and improved treatments. Thus we sought to develop such a model based on a gene-X-environment approach to give insight into the pathophysiology of MDpsy as well as to provide a platform for the development of more effective treatment strategies. The Flinders Sensitive Line (FSL) is a genetic rodent model of MD, while post-weaning social isolation rearing (SIR) is a neurodevelopmental model of schizophrenia. Both models reflect the bio-behavioural features of their respective human disorder. We hypothesised that combining the two models would produce an animal that presents with the bio-behavioural features of MDpsy later in life. The current investigation was thus conducted to achieve the following three aims: 1) establish the depressive-like profile of the FSL-SIR rat and resistance to FLX; 2) establish the psychotic-like profile of the FSL-SIR rat and its resistance to or partial improvement with OLZ; 3) demonstrate therapeutic benefits across mood and psychotic manifestations in FSL-SIR rats following OLZ-FLX combination (OFC) treatment.

To establish the SIR-exposed FSL (FSL-SIR) rat as a model of TRD, male FSL rats were reared socially or in isolation for 8 weeks from post-natal day (PND) 21. These rats received either saline (1 mg/kg, SAL) or FLX (10 mg/kg) subcutaneously (s.c) for 14 days from PND 63. Coping and depressive-like behaviour, locomotor activity, recognition memory and social interactive behaviours were assessed in the forced swim test (FST), open field test (OFT), novel object recognition test (nORT), and social interaction test (SIT), respectively. To establish the FSL-SIR as a model of MDpsy, male SD and FSL rats were reared in isolation for 8 weeks from PND 21. These rats received either saline (1 mg/kg, SAL) or OLZ (5 mg/kg) subcutaneously (s.c) for 14 days from PND 63. To assess the therapeutic effects of OFC in the FSL-SIR rats, another group of FSL-SIR rats received a combination of FLX and OLZ (OFC), at doses of 10 mg/kg and5 mg/kg, respectively. The effects of OLZ and OFC in the FST were also assessed in FSL-SIR rats. Sprague-Dawley (SD) rats receiving only SAL treatment were used in both studies as the healthy control. Post-mortem cortico-hippocampal monoamine levels were subsequently determined by high performance liquid chromatography (HPLC) while plasma dopamine-β-hydroxylase (DBH), corticosterone (CORT), interleukin-6 (IL-6), tumour necrosis factor (TNF-α), and brain-derived neurotrophic factor (BDNF) levels were determined by enzyme-linked immunosorbent assay (ELISA). FSL displayed significant manifestations of depressive-like, behaviours, including, a trend to psychomotor retardation in the OFT, significantly elevated immobility (despair) and reduced swimming (survival behaviour) in the FST, and significant social withdrawal and social anxiety-like behaviour in the SIT versus SD-SAL rats. Significant reversals in these deficits were observed following FLX treatment while a trend toward reduced social anxiety-like behaviour was observed. Climbing behaviour in the FST was unaffected by condition or treatment in FSL rats and not altered by FLX. Downward trends in novelty discrimination were observed in the FSL rats and left unchanged by FLX. FSL rats presented with significantly reduced cortico-hippocampal NE and hippocampal 5-HT and frontocortical 5-HT and cortico-hippocampal DA levels similar to SD controls. FLX did not reverse these deficits but rather significantly reduced hippocampal 5-HT. Plasma DBH, CORT, IL-6, and TNF-α levels were unchanged in FSL versus SD rats. All except for CORT were unaltered by FLX. FLX treatment significantly increased plasma CORT in FSL rats. The downward trend in plasma BDNF noted in these rats was continued by FLX treatment. Locomotor activity was increased by a large effect size in SAL-treated FSL-SIR rats versus FSL-SAL rats while a trend toward reduced activity followed FLX treatment. Like socially-housed FSL rats, FSL-SIR rats

displayed significant immobility versus SD animals, although less than FSL rats by a large effect size. Swimming was similar in FSL-SIR rats and SD controls while there was an upward trend in climbing. FLX significantly increased immobility and decreased swimming versus SD rats. OLZ and OFC did not improve immobility but rather significantly reduced swimming in these rats. The upward trend in climbing was continued by FLX but reversed by OLZ and OFC. FSL-SIR did not display cognitive impairment in the nORT, however, FLX showed a trend to improve novelty discrimination. FSL-SIR rats exhibited significant disordered social behaviour and asocial (anxious) behaviour versus SD rats. Neither was reversed by FLX. Significant cortico-hippocampal NE and 5-HT were observed while downward trends in cortico-hippocampal DA were observed in FSL-SIR rats. FLX significantly reversed all these deficiencies. DBH was significantly reduced in FSL-SIR versus SD rats, with a very large effect size reduction versus SAL-treated FSL rats. There were not reversed by FLX. CORT was not significantly elevated in FSL-SIR versus FSL controls and remained unaltered by FLX. FSL-SIR rats did not present with significant elevations in IL-6 and TNF-α although a downward trend in TNF-α followed FLX treatment. Although presenting similarly to SD rats, BDNF in FSL-SIR rats followed a downward trend after treatment with FLX. SD-SIR exhibited psychotic-like behaviour manifested as significantly decreased %PPI which was not improved by OLZ treatment. SD-SIR rats did not exhibit hyperlocomotion but did trend toward thigmotactic behaviour in the OFT versus SD controls which was not improved by OLZ treatment. SD-SIR rats did not manifest social withdrawal, anxiety or aggression in the SIT. SD-SIR rats presented with trends toward reduced cortico-hippocampal NE and raised cortico-hippocampal 5-HT levels versus SD rats were observed. These aberrations were normalised by OLZ treatment. Cortico-hippocampal DA was similar in SD-SIR and SD rats. Plasma DBH was significantly reduced in SD-SIR rats and trended toward a further reduction following OLZ treatment. An upward trend in SD-SIR plasma CORT levels versus SD controls was observed. This was reversed by OLZ treatment. Elevated plasma IL-6 and TNF-α levels were not observed in SD-SIR rats and both remained unaltered by OLZ treatment. FSL-SIR rats did not exhibit hyperlocomotion but did display significant thigmotactic behaviour with downward trends and significant reductions in both parameters following OLZ and OFC, respectively. While FSL-SIR did not exhibit cognitive impairment in the nORT, a trend to increase novelty discrimination was noted following treatment with OLZ. OFC had no effect on novelty discrimination. Significant social withdrawal and social anxiety-like behaviour were observed in the FSL-SIR versus SD controls; these were not improved

by OLZ or OFC treatment. There was a trend to aggression by the FSL-SIR rats which was attenuated by OLZ and normalised by OFC. Significantly reduced %PPI in the FSL-SIR rat was not improved by OLZ but a trend to increase %PPI followed OFC treatment. Significant cortico-hippocampal NE and 5-HT were observed while downward trends in cortico-hippocampal DA were observed in the FSL-SIR rat. OLZ and OFC caused a significant rise in cortico-hippocampal NE and 5-HT in FSL-SIR rats. Upward trends in cortico-hippocampal DA followed OLZ and OFC treatment. OLZ and OFC did not alter the trend toward raised CORT in the FSL-SIR versus SD rats. Neither OLZ nor OFC altered the downward trend in plasma DBH levels noted in the FSL-SIR rat. Although not significantly different from SD controls, OLZ and OFC showed a trend to reduce both plasma IL-6 and TNF-α. BDNF was not altered in the FSL-SIR rat; however, an upward trend in this neurotrophin followed OLZ treatment. OFC had no effect on BDNF in these rats. These results confirm that exposing FSL rats to post-weaning SIR produces animals that present with both depressive- and psychotic-like behaviours akin to those observed in MDpsy, including immobility in the FST, decreased %PPI, psychomotor agitation, social impairment, anxiety, and aggression and accompanied by increased plasma CORT and reduced DBH. These depressive- and psychotic-like symptoms as well as CORT and DBH demonstrate resistance to FLX and OLZ