The soya isoflavone content of rat diet can increase anxiety and stress hormone release in the male rat

Rationale: Most commercial rodent diets are formulated with soya protein and therefore contain soya isoflavones. Isoflavones form one of the main classes of phytoestrogens and have been found to exert both oestrogenic and anti-oestrogenic effects on the central nervous system. The effects have not been limited to reproductive behaviour, but include effects on learning and anxiety and actions on the hypothalamo-pituitary axis. It is therefore possible that the soya content of diet could have significant effects on brain and behaviour and be an important source of between-laboratory variability. Objectives: To determine whether behaviour in two animal tests of anxiety, and stress hormone production, would differ between rats that were fed a diet which was free of soya isoflavones and other phytoestrogens (isofree) and those. that were fed a diet which contained 150 mug/g of the isoflavones genistein and daidzein (iso-150). This controlled diet has an isoflavone concentration similar to that in the maintenance diet routinely used in our institution. Methods: Male rats were randomly allocated to the iso-free and iso-150 diets and their body weights and food and water consumption were recorded for 14 days. They were then maintained on the same diets, but housed singly for 4 days, before testing in the social interaction and elevated plus-maze tests of anxiety. Corticosterone concentrations in both dietary groups were determined under basal conditions and after the stress of the two tests of anxiety. Vasopressin and oxytocin concentrations were determined after brief handling stress. Results The groups did not differ in food or water intake, body weight or oxytocin concentrations. Compared with the rats fed the iso-free diet, the rats fed the iso-150 diet spent significantly less time in active social interaction and made a significantly lower percentage of entries onto the open arms of the plus-maze, indicating anxiogenic effects in both animal tests. The groups did not differ in their basal corticosterone concentrations, but the iso-150 group had significantly elevated stress-induced corticosterone concentrations. Stress-induced plasma vasopressin concentrations were also significantly elevated in the iso-150 diet group compared with the isofree rats. Conclusions: Major changes in behavioural measures of anxiety and in stress hormones can result from the soya isoflavone content of rat diet. These changes are as striking as those seen following drug administration and could form an important source of variation between laboratories.