The ring dove (Streptopeliafi sotia) has proved an excellent animal model for studies in reproductive endocrine-behaviour interactions for many years. It shows bi-parental care and has a stereotyped breeding cycle. Androgen receptor and progesterone receptor immunoreactivity (AR-ir and PR-ir) were localised in the brain of male and female ring doves. Nuclear AR-ir in courting birds was widespread throughout the telencephalong
diencephalon, and mesencephalon. Nuclear PR-ir was only localised in discrete regions of the preoptic-hypothalamus of both sexes. In the anterior and posterior hypothalamus, high density of AR-ir and PR-ir neurones was concentrated and co-localised in the nucleus preopticus anterior (POA), nucleus preopticus mediahs (POM), nucleus medialis, pars
medianis (POMm), nucleus preopticus paraventricularis magnocellularis (PPM), nucleus hypothalami laterahs posterioris (PLH), and tuberal. hypothalamus (Tu). Investigation of the effects of the breeding cycle on anterior pituitary gland and hypothalamic AR-ir and
PR-ir showed a significant (p<0.001) increase in AR-ir cells in courting birds when compared with birds at other stages of the breeding cycle. AR-ir cells in birds brooding
young was significantly reduced or absent altogether. In the anterior pituitary gland of both sexes, PR-ir cells increased during courtship and significantly decreased in brooding birds.
In the anterior hypothalamus of both sexes, no significant change in PR-ir density was observed. However, a significant increase (p<0.05) of PR-ir neurones was seen 'in the ventral and lateral Tu of incubating male and female doves, respectively, whilst a significant decrease in PR-ir neurones was seen in the ventral region of the Tu in birds brooding young (p<0.001). Although no sex difference was obtained in the number of AR-ir neurones, significantly higher staining intensity for AR-ir was measured in male than female doves, in all hypothalamic regions investigated. Similarly, increased PR-ir staining intensity was measured in courting birds than in birds at other breeding stages. In the anterior pituitary gland and all regions of the hypothalamus the percentage co locahsation between AR-ir and PR-ir changes With breeding cycle. The highest percentage of PR-ir and AR-ir
co-localisation (70-909/6w) as obtained in courting birds and the lowest (0.27-24%) in birds brooding young.
The association between steroid receptor expression and aromatase P450 enzyme(P450AROM)activity in the brain of the ring dove was also investigated. Sexually experienced
ring doves were injected intramuscularly with the aromatase inhibitor fadrozole (0.2 ml of 1 mg n-A-') and paired with saline injected controls in a cage containing a nest bowl and nesting material. The fadrozole/ saline vehicle was administered for 3 days at 12-hour intervals. Saline injected control males displayed aggressive and nest-orientated behaviours whereas fadrozole treated males showed none of these behaviours. Similarly, saline injected
control females displayed nest-onented behaviours and were sitting on the nest by day 2, whereas fadrozole injected females showed none of these behaviours. On day 4 of
courtship, birds were perfuse-fixed and the brains and anterior pituitary glands subjected to immunocytochernistry for androgen receptor (AR), progesterone receptor (PR) and
aromatase (P450ARONýIn* saline treated birds of both sexes, nuclear AR-ir and PR-ir were seen in the PON, PPM, POM, POMm, PLH and Tu. In contrast, fadrozole treatment in
both sexes resulted in the complete absence of detectable nuclear AR-ir expression, together with a significant decline in the expression of PR-ir in the POM and POMm, and
all regions of the Tu (p< 0.001). A sex difference in PR-ir expression was seen in the POA and PPM (p<0.001) of fadrozole treated birds. Fadrozole administration caused decrease in the percentage co-localisation of AR-ir and PR-ir in both sexes. A high density of AROMir neurones was observed in the POM and POMm of saline injected males, which was found to co-express PR-ir (99%).
The viability of maintaining in vitro brain slice preparations was studied with the aim to identify and confirm neural circuits involving the anterior and posterior hypothalamus in the initiation of sexual and parental behaviours in this species. Sections were pressure
injected with fluorogold (FG) or Dil (8 -15 nl) in the POA, POM, and the ventral, lateral and dorsal regions of the Tu. Results obtained confirm previous reports of neural
connections between the Tu and POA. Neural connections were also seen between the Tu and the nucleus medialis hypothalan-ii posterioris (PMH), and several other regions of the mesencephalon and telencephalon, including the nucleus intercollicularis JCo), nucleus ovoidalis (Ov), dorsolateralis posterior thalami (DLP), archistriatum, pars ventrale (Av) and archistriatum, pars dorsalis (Ad).
It is concluded that in the dove, central AR-ir and PR-ir expression and co-localisation are closely associated with the sexual stages of the reproductive cycle. Increases 'in AR-ir neurones in the preoptic-hypothalamus in courting birds are consistent with the behavioural role of androgen acting in this brain area during this time. Similarly, the high PR-ir neurones in the preoptic-hypothalamus are consistent with the progesterone-induced initiation of incubation in this species. The high percentage of AR-ir and PR-ir colocalisation in the preoptic-hypothalamus of courting doves supports previous reports involving progesterone acting in these brain regions to terminate the androgen-dependent aggressive courtship behaviour in male doves. The decrease in the number of co-localised
AR-ir and PR-ir neurones M the preoptic-hypothalamus following oestrogen synthesis inhibition suggests a role for oestrogen in the progesterone-dependent termination of
aggressive courtship display in males, and models proposing a mechanism involving oestrogen in the orchestration of central systems associated with courtship behaviour in the
ring dove are presented. Neural connections between the POA and Tu, and Tu with the PMH, ICo, Ov, DLP, Av and Ad may represent the neuroanatomical pathways involved in
the integration of courtship display, vocalisation (nest-coos and/or bow-coos), auditory, propriosceptive signals and gonadotropM secretion in the reproductive cycle of the ring dove. These results are consistent with the hypothesis that the POA, Tu and the midhypothalamus are involved in the orchestration of sexual and parental behaviour of the ring dove.