Schizophrenia onset peaks around puberty and occurs more frequently and earlier in males than females, suggesting a mediating role of sex hormones. Brain-derived neurotrophic factor (BDNF) supports neurotransmitter systems implicated in schizophrenia and its expression is reduced in patients. We have previously shown in mice that forebrain BDNF expression correlates with pubertal testosterone elevation. However, it is unclear how testosterone and BDNF interact during adolescence. We therefore examined the effects of pre-pubescent hormone manipulation in a BDNF heterozygous (het) mouse model of psychotropic-induced hyperlocomotion and pre-pulse inhibition (PPI) – endophenotypes of schizophrenia. Pre-pubescent male wild-type (WT) and het mice were gonadectomised and implanted with placebo, testosterone or dehydrotestosterone (DHT). Upon adulthood, amphetamine and MK801-induced hyperactivity and PPI were examined.

For locomotor, neither genotype nor gonadectomy affected response to MK801. However, testosterone caused a reduction in MK801-induced locomotion only in WT mice. DHT had no effect, suggesting that the testosterone effect is mediated through conversion to estrogen and that this is compromised in het mice. Amphetamine responses in shams were comparable between genotypes. Gonadectomy did not alter amphetamine-induced hyperactivity. Both testosterone and DHT reduced amphetamine responses in both genotypes. For PPI, het mice displayed increased startle compared to WT. Gonadectomy decreased startle responses whereas testosterone and DHT implants reasserted it in both genotypes. No genotype differences were seen in PPI across treatments in sham animals. Gonadectomy reduced PPI in response to amphetamine in WT, which was restored by testosterone but not DHT implants, suggesting estrogen effects. In het mice, sex hormone manipulation did not have significant effects in PPI responses. 

These findings demonstrate that reduced BDNF levels alter the effects of marked pubescent hormonal influences on psychotropic-induced hyperlocomotion, startle levels and PPI, suggesting a complex interaction between BDNF, testosterone and both NMDAR and dopamine signalling.