Androgen synthesis in a songbird: a study of cyp17 (17alpha-hydroxylase/C17,20-lyase) activity in the zebra finch.

Androgens and estrogens influence the maturation and function of numerous tissues in both male and female birds, especially the brains of the oscine songbirds. Although there exist a very large number of studies that have investigated circulating sex steroids in many species of wild and captive-held songbirds, there remain a significant number of questions about the sites of synthesis of the active steroids that act on the songbird brain. Estrogens are derived from androgen. Thus, the synthesis of androgen itself is critical for both androgen- and estrogen-dependent actions in both male and female songbirds. Therefore, we have undertaken studies of the enzyme 17alpha-hydroxylase/C17,20-lyase (Cyp17), the enzyme responsible for the synthesis of androgens from their progestin or pregnane precursors via their 17alpha-hydroxy intermediates. Here we have characterized optimal conditions for measuring Cyp17 in gonads of adult zebra finches via the conversion of tritiated [3H]progesterone into 17alpha-hydroxy P (17alpha-hydroxylase activity) and androstenedione and testosterone (C17,20-lyase) activity. Cyp17 activity is abundant in testis, with lesser amounts in ovary. Low levels of Cyp17 activity were also detected in male adrenals, but not in any other tissue, including brain. Testicular Cyp17 activity is readily inhibited in vitro by ketoconazole, a specific Cyp17 inhibitor. Ketoconazole works less well in vivo. In males castrated and/or treated with fadrozole, an inhibitor of aromatase, we detected no extragonadal sites of Cyp17 activity, although fadrozole appeared to increase circulating androgens in both castrated and intact males. Thus, we still do not know the site of androgen synthesis in these males. Further studies of Cyp17 will be useful in understanding more about the mechanisms of androgen delivery to neural circuits in adult and developing songbirds.

A putative 5 alpha-reductase inhibitor demasculinizes portions of the zebra finch song system.

One model of the sexual differentiation of the zebra finch song system holds that both major metabolites of testosterone, dihydrotestosterone (DHT) and estradiol (E2), act together to masculinize the song system. To test this model, we administered a putative inhibitor of 5 alpha-reductase (MK-434) to decrease the synthesis of DHT from testosterone (T) in hatchling zebra finches. We tested MK-434's inhibition of 5 alpha-reductase, 5 beta-reductase, and aromatase in vivo and in vitro. In vivo, MK-434 significantly inhibited 5 alpha-reductase activity but also reduced the activities of 5 beta-reductase and aromatase. In vitro, MK-434 was extremely effective in inhibiting 5 alpha-reductase in the rat prostate but only slightly inhibited 5 alpha-reductase in the zebra finch telencephalon, where it also reduced aromatase and 5 beta-reductase activities. These results suggest that MK-434 might differentially influence the availability of androgenic and estrogenic substrates, depending on the relative abundance of these enzymes in brain. MK-434 demasculinized (decreased) the number and decreased the density of RA neurons but did not significantly affect any other sexually dimorphic aspect of the song system, including the volumes of RA, HVC, and Area X; the size of neural somata in IMAN, HVC, and RA; and the number of neurons in HVC and IMAN. The differential influence of MK-434 on sexually dimorphic characteristics suggests that the various sexually dimorphic characteristics of the song system (1) are sensitive to different hormones, depending on the characteristic; or (2) have different sensitivities to hormone levels, some being easily affected by slightly reduced hormone levels whereas others are not; or (3) have markedly different critical periods depending on the characteristic. Regardless of the reason(s) for differential effects on the sexually dimorphic characteristics of the song system, the data clearly suggest that steroid hormones play a role in the normal masculine development of the song system.

Sexual differentiation of the brain in songbirds.

The brain regions that control song in zebra finches are much larger in males, who sing, than in females, who do not. Two major theories have been proposed to explain sexual differentiation of the neural song circuit. The 'mammalian' theory suggests that sex steroid secretions of the tests cause masculine development in males. The 'avian' theory suggests that ovarian secretions induce feminine patterns of development in females. Although experimental evidence provides some support for the mammalian theory, neither theory comfortably predicts the outcomes of experiments that bear on the mechanisms of sexual differentiation. In particular, it has been relatively difficult to block sex steroid synthesis and action in genetic males in a way that prevents masculine neural differentiation. Moreover, genetic females that possess large amounts of testicular tissue can have a feminine neural song circuit, suggesting that testicular secretions are not solely responsible for the masculine patterns of differentiation. The results indicate that new theories are needed to explain sexual differentiation of the song system.

Lack of a synergistic effect between estradiol and dihydrotestosterone in the masculinization of the zebra finch song system.

Previous studies have suggested that both major active metabolites of testosterone, estradiol (E2) and dihydrotestosterone (DHT), are needed for complete masculinization of the brain regions that control song in passerine birds. However, DHT treatment of hatchling female zebra finches has only small masculinizing effects on the song system. To assess whether E2 and DHT have a synergistic effect on the masculinization of the zebra finch song system, female zebra finches were given Silastic implants of E2 on the day of hatching (day 1) either without any additional hormone treatment or in combination with DHT on days 1, 14, or 70. At 105 to 110 days of age, we measured the volumes of Area X, higher vocal center (HVC), robust nucleus of the archistriatum (RA), soma sizes in HVC, RA, and the lateral magnocellular nucleus of the neostriatum (IMAN), and neuron density and number in RA. E2 masculinized all of the measures in the song system with the exception of the number of neurons in RA. DHT did not synergize with E2 to produce any additional masculinization of the attributes measured. These data demonstrate that the combination of E2 and DHT did not result in the complete masculinization of the song control nuclei and argue against the importance of androgen in sexual differentiation of the song system.

A direct comparison of the masculinizing effects of testosterone, androstenedione, estrogen, and progesterone on the development of the zebra finch song system.

To assess which hormones are capable of masculinizing the neural song system of zebra finch hatchlings, we implanted female hatchlings with estrogen (estradiol [E2], 75 micrograms, n = 9), testosterone (T, 75-88 micrograms, n = 13), androstenedione (AE, 75 micrograms, n = 7), progesterone (P, 117 micrograms, n = 10), or nothing (Blanks, n = 10) and compared these to unimplanted males (n = 7). Implants, consisting of a hormone and Silastic mixture encased in polyethylene tubing, were placed under the skin of the breast on the day of hatching. Birds were killed when they were sub-adult (58 to 68 days old). We measured volumes of area X, the higher vocal center (HVC), and the robust nucleus of the archistriatum (RA); measured soma sizes in the lateral magnocellular nucleus of the neostriatum (lMAN), HVC, and RA; and counted RA neurons. E2 masculinized all measures in the song system and nearly sex-reversed the size of RA neurons. T masculinized volumes of nuclei and soma sizes but not the number or spacing of RA neurons. E2 was always at least as effective as T in masculinizing measures of the song system and was usually more effective. AE and P did not significantly masculinize any measure. These data suggest that E2 is more potent than aromatizable androgens or P in masculinizing the female song system in development and that the action of E2 alone may be sufficient to masculinize the volume of song control nuclei and the size and number of neurons.

Distribution of GABA-like immunoreactivity in the song system of the zebra finch.

GABA-like immunoreactivity (GABA-LIR) was mapped in the male and female zebra finch song system using a polyclonal antibody to GABA. GABA-LIR was found throughout the song system in neurons and neuropil of the robust nucleus of the archistriatum (RA), the higher vocal center (HVC), Area X, the magnocellular nucleus of the neostriatum (MAN), and the dorsomedial portion of the nucleus intercollicularis (DM of ICo). Puncta present in the lateral division of MAN (lMAN) may be local interneurons since the only known afferents of lMAN are from the dorsolateral nucleus of the anterior thalamus (DLM), which did not appear to have any cell bodies with GABA-LIR. Distinct and dense puncta with GABA-LIR were present in DLM, and may be projections from Area X/lobus parolfactorius (LPO). Dramatic sex differences in GABA-LIR distribution were found. Females did not appear to have any GABA-LIR above background in either RA or HVC. Females also did not appear to have a distinct Area X, although they did have many small, lightly staining cell bodies in the corresponding LPO. The distribution of GABA-LIR and sex differences in its distribution suggests that GABAergic neurons may play a role in the acquisition and/or production of song in the zebra finch.

Local intracerebral implants of estrogen masculinize some aspects of the zebra finch song system.

The song system of zebra finches is sexually dimorphic: the volumes of the song control nuclei and the neurons within these nuclei are larger in males. The song system of hatchling female zebra finches is masculinized by systemic treatment with estrogen. We investigated the locus of this estrogen action by using microimplants of estradiol benzoate (EB). We implanted female zebra finch nestlings 10-13 days old with Silastic pellets containing approximately 2 micrograms EB at one of several sites: near the higher vocal center (HVC), in the brain distant from HVC, or in the periphery either under the skin of the breast or in the peritoneal cavity. Controls were either unimplanted or implanted near HVC with Silastic pellets without hormone. The brains were fixed by perfusion at 60 days, and the volumes of the song control regions as well as the sizes of individual neurons were measured. Neurons in HVC were larger (more masculine) in the HVC-implanted group than in the other groups, which did not differ among themselves. The size of neurons in the robust nucleus of the archistriatum (RA) and the lateral magnocellular nucleus of the neostriatum (lMAN) were inversely correlated with the distance of the EB pellet to HVC; neurons in RA and lMAN were larger when the EB pellets were closer to HVC. This result suggests that implants near HVC were at or near a site of estrogen action. To our knowledge, this is the first demonstration that localized brain implants of estrogen cause morphological masculinization in any species.

Prenatal beta-endorphin can modulate some aspects of sexual differentiation in rats.

Sexually dimorphic traits were studied in offspring of rats injected with 33 micrograms rat beta-endorphin (beta-END) three times daily from Day 14 to Day 21 of pregnancy. beta-END males had shorter neonatal anogenital distances than did controls and were more likely to show the female lordosis pattern as adults, but they did not differ in male copulatory behavior. When given a choice between spending time with an estrous female or a male, beta-END males showed a lower preference for the female than did control males. The number and somal size of neurons in the bulbocavernosus and dorsolateral nucleus of the lumbar spinal cord were unaffected by drug exposure. Elevated beta-END during fetal ontogeny apparently alters the differentiation of some, but not all, sexually dimorphic traits. The data suggest that endogenous opioids may contribute to the etiology of the prenatal stress syndrome.

Prenatal flutamide alters sexually dimorphic nuclei in the spinal cord of male rats.

The effects of prenatal exposure to the antiandrogen flutamide on two sexually dimorphic nuclei of the lumbar spinal cord, the dorsolateral nucleus (DLN) and the spinal nucleus of the bulbocavernosus (SNB), were investigated. Rat dams were given daily injections of 5 mg flutamide or vehicle alone from day 11 through 21 of pregnancy. The spinal cords and perineal morphology of their male and female offspring were examined in adulthood. Flutamide reduced the number of SNB and DLN neurons, reduced the somal and nuclear area of SNB neurons, and reduced the weight of the perineal muscles in males. Flutamide produced no effect in females. No sexual dimorphism was found in the mean somal area of DLN neurons, but a sexual dimorphism was found in the distribution of somal areas in our samples; females had proportionately more large neurons than males. Flutamide-treated males also had proportionately more large neurons than control males but fewer than females. A sexual dimorphism was found in the nuclear areas of DLN neurons but flutamide did not influence this trait.

Prenatal stress alters sexually dimorphic nuclei in the spinal cord of male rats.

The spinal nucleus bulbocavernosus (SNB), the dorsolateral nucleus of the spinal cord (DLN), and the bulbocavernosus/levator ani (BC/LA) muscle complex were examined in prenatally stressed and control adult male rats, which had been screened for male copulatory behavior. There was a small but significant decrease in the number of DLN (5%) and SNB (3%) neurons in prenatally stressed males compared to controls. Prenatal stress had no effect on the somal or nuclear area of individual neurons within either nucleus, nor did it affect the weight of the BC/LA muscle complex. There were no differences in any of these measures between males that ejaculated and those did not in either the stressed or the control group. These data suggest that exposure of pregnant rats to transient environmental stressors may result in permanent alterations in androgen-sensitive CNS structures in their male offspring.

Effects of dorsal and medial cortex lesions on reversals in turtles.

Two experiments were performed to investigate the effect of cortical lesions on the acquisition and reversal of simultaneous discriminations in turtles. The first experiment examined the effect of cortical lesions on the acquisition and reversal of a spatial discrimination. The results of the first experiment revealed that lesions of the dorsal cortex produced a deficit in spatial learning. The results of the first experiment also revealed that when damage to the dorsal cortex was accompanied by substantial damage to the medial cortex, no deficit was manifest. The second experiment examined the effects of cortical lesions on the acquisition and reversal of a brightness discrimination. The results of the second experiment revealed that damage to neither the dorsal cortex nor the medial cortex produced a deficit. It was suggested that brightness is not represented in the thalamofugal visual pathway but is instead represented in the tectofugal visual pathway in reptiles. It was also suggested that the medial cortex, which is the evolutionary precursor to the mammalian hippocampal formation, functions differently from the mammalian hippocampus.

Function of the dorsal and medial cortex of turtles in learning.

The effects of damage to the dorsal and medial cortex of turtles were investigated in two experiments. In the first, damage to the dorsal cortex disrupted acquisition and reversal of a go-no-go discrimination but had no effect on retention of the discrimination if it had been learned preoperatively. Medial cortex damage had no effect. In the second experiment, dorsal cortex damage impaired acquisition, but not extinction or reacquisition, of a discrete-trial keypress. Again, medial cortex damage had no effect. The results suggest that the dorsal cortex is involved in learning in turtles.