GnRH cell size and number in a teleost fish with two male reproductive morphs: sexual maturation, final sexual status and body size allometry

Ontogeny of Inner Ear Saccular Development in the Plainfin Midshipman (Porichthys notatus)

The auditory system of the plainfin midshipman fish (Porichthys notatus) is an important sensory system used to detect and encode biologically relevant acoustic stimuli important for survival and reproduction including social acoustic signals used for intraspecific communication. Previous work showed that hair cell (HC) density in the midshipman saccule increased seasonally with reproductive state and was concurrent with enhanced auditory saccular sensitivity in both females and type I males. Although reproductive state-dependent changes in HC density have been well characterized in the adult midshipman saccule, less is known about how the saccule changes during ontogeny. Here, we examined the ontogenetic development of the saccule in four relative sizes of midshipman (larvae, small juveniles, large juveniles, and nonreproductive adults) to determine whether the density, total number, and orientation patterns of saccular HCs change during ontogeny. In addition, we also examined whether the total number of HCs in the saccule differ from that of the utricle and lagena in nonreproductive adults. We found that HC density varied across developmental stage. The ontogenetic reduction in HC density was concurrent with an ontogenetic increase in macula area. The orientation pattern of saccular HCs was similar to the standard pattern previously described in other teleost fishes, and this pattern of HC orientation was retained during ontogeny. Lastly, the estimated number of saccular HCs increased with developmental stage from the smallest larvae (2,336 HCs) to the largest nonreproductive adult (145,717 HCs), and in nonreproductive adults estimated HC numbers were highest in the saccule (mean ± SD = 28,479 ± 4,809 HCs), intermediate in the utricle (mean ± SD = 11,008 ± 1,619 HCs) and lowest in the lagena (mean ± SD = 4,560 ± 769 HCs).

Functional differences in the hypothalamic-pituitary-gonadal axis are associated with alternative reproductive tactics based on an inversion polymorphism

The evolution of social behavior depends on genetic changes, yet, how genomic variation manifests itself in behavioral diversity is still largely unresolved. Chromosomal inversions can play a pivotal role in producing distinct behavioral phenotypes, in particular, when inversion genes are functionally associated with hormone synthesis and signaling. Male ruffs exhibit alternative reproductive tactics (ARTs) with an autosomal inversion determining two alternative morphs with clear behavioral and hormonal differences from the ancestral morph. We investigated hormonal and transcriptomic differences in the pituitary and gonads. Using a GnRH challenge, we found that the ability to synthesize testosterone in inversion carriers is severely constrained, whereas the synthesis of androstenedione, a testosterone precursor, is not. Inversion morphs were able to produce a transient increase in androstenedione following the GnRH injection, supporting the view that pituitary sensitivity to GnRH is comparable to that of the ancestral morph. We then performed gene expression analyses in a second set of untreated birds and found no evidence of alterations to pituitary sensitivity, gonadotropin production or gonad sensitivity to luteinizing hormone or follicle-stimulating hormone across morphs. Inversion morphs also showed reduced progesterone receptor expression in the pituitary. Strikingly, in the gonads, inversion morphs over-expressed STAR, a gene that is located outside of the inversion and responsible for providing the cholesterol substrate required for the synthesis of sex hormones. In conclusion, our results suggest that the gonads determine morph-specific differences in hormonal regulation.

A conceptual framework for understanding sexual differentiation of the teleost brain

Vertebrate brains are sexually differentiated, giving rise to differences in various physiological and behavioral phenotypes between the sexes. In developing mammals and birds, the neural substrate underlying sex-dependent physiology and behavior undergoes an irreversible process of sexual differentiation due to the effects of perinatal gonadal steroids and sex chromosome complement. The differentiated neural substrate is then activated in the adult by the sex-specific steroid milieu to facilitate the expression of sex-typical phenotypes. However, this well-established concept does not hold for teleost fish, whose sexual phenotypes (behavioral or otherwise) are highly labile throughout life and can be reversed even in adulthood. Indeed, the available evidence suggests that, in teleosts, neither gonadal steroids early in development nor the sex chromosome complement contribute much to brain sexual differentiation; instead, steroids in adulthood serve to both differentiate the neural substrate and activate it to elicit sex-typical phenotypes in a transient and reversible manner. Evidence further suggests that marked sexual dimorphisms and adult steroid-dependent lability in the neural expression of sex steroid receptors constitute the primary molecular basis for sexual differentiation and lability of the teleost brain. The consequent sexually dimorphic but reversible steroid sensitivity in response to the adult steroid milieu may enable the teleost brain to maintain lifelong sexual lability and to undergo phenotypic sex reversal.

Characterization of the gonadotropin-releasing hormone system in the Neotropical teleost, Steindachneridion parahybae during the annual reproductive cycle in captivity

This study evaluated by immunohistochemical and Western blot methods, the distribution of two distinct gonadotropin-releasing hormones (GnRHs), corresponding to catfish GnRH (cfGnRH or GnRH1) and chicken-II GnRH (cGnRH-II or GnRH2), in Steindachneridion parahybae females in captivity, focusing these analyses on the reproductive cycle by semi-quantification of optical density (OD). Further, we found that the GnRH neuronal systems co-localized with their respective GnRH-associated peptides (GAPs). A group of neurons immunoreactive (ir) to GnRH1 were identified along the ventral region of the olfactory bulb (vOB) in the telencephalon (vTel) and in the main areas of the diencephalon (especially the medial basal hypothalamus, HBM), including fibers extending into the pituitary gland. In contrast, GnRH2 neurons were confined to the midbrain tegmentum, close to the ventricular surface, without projections to the pituitary gland. Moreover, a cfGAP (GnRH1)-specific band (9 kDa) was identified in the brain and pituitary gland, while a cGAP-II (GnRH2)-specific band (26 kDa) was observed only in the brain extract. During the reproductive cycle, GnRH1-ir presented greater OD values at the vitellogenic and regression stages than at the previtellogenic stage and after artificially induced to spawn. Larger GnRH2-ir neurons were observed during the reproductive cycle, but a higher OD was identified only in the regression stage compared with the other maturation stages. Finally, GnRH1 axons were found to be directed towards the pituitary, and this GnRH type, which is probably the hypophysiotropic form, can contribute to the reproductive dysfunction that occurs in S. parahybae females in captivity, whereas GnRH2 may act as a neuromodulator and/or neurotransmitter.

Expression of tachykinin3 and related reproductive markers in the brain of the African cichlid fish Astatotilapia burtoni

Neurokinin B, encoded by the tachykinin3 gene, plays a crucial role in regulating reproduction in mammals via KNDy neurons and interaction with GnRH. Previous work in teleost fishes has focused on hypothalamic tac3 expression for its role in reproduction, but detailed studies on extra-hypothalamic tac3 expression are limited. Here, we identified two tac3 genes in the social African cichlid fish Astatotilapia burtoni, only one of which produces a functional protein containing the signature tachykinin motif. In situ hybridization for tac3a mRNA identified cell populations throughout the brain. Numerous tac3a cells lie in several thalamic and hypothalamic nuclei, including periventricular nucleus of posterior tuberculum, lateral tuberal nucleus (NLT), and nucleus of the lateral recess (NRL). Scattered tac3-expressing cells are also present in telencephalic parts, such as ventral (Vv) and supracomissural (Vs) part of ventral telencephalon. In contrast to other teleosts, tac3 expression was absent from the pituitary. Using double-fluorescent staining, we localized tac3a-expressing cells in relation to GnRH and kisspeptin cells. Although no GnRH-tac3a colabeled cells were observed, dense GnRH fibers surround and potentially synapse with tac3a cells in the preoptic area. Only minimal (<5%) colabeling of tac3a was observed in kiss2 cells. Despite tac3a expression in many nodes of the mesolimbic reward system, it was absent from tyrosine hydroxylase (TH)-expressing cells, but tac3a cells were located in areas with dense TH fibers. The presence of tac3a-expressing cells throughout the brain, including in socially relevant brain regions, suggest more diverse functions beyond regulation of reproductive physiology that may be conserved across vertebrates.

Behavioural tactic predicts preoptic-hypothalamic gene expression more strongly than developmental morph in fish with alternative reproductive tactics

Reproductive success relies on the coordination of social behaviours, such as territory defence, courtship and mating. Species with extreme variation in reproductive tactics are useful models for identifying the neural mechanisms underlying social behaviour plasticity. The plainfin midshipman (Porichthys notatus) is a teleost fish with two male reproductive morphs that follow widely divergent developmental trajectories and display alternative reproductive tactics (ARTs). Type I males defend territories, court females and provide paternal care, but will resort to cuckoldry if they cannot maintain a territory. Type II males reproduce only through cuckoldry. We sought to disentangle gene expression patterns underlying behavioural tactic, in this case ARTs, from those solely reflective of developmental morph. Using RNA-sequencing, we investigated differential transcript expression in the preoptic area-anterior hypothalamus (POA-AH) of courting type I males, cuckolding type I males and cuckolding type II males. Unexpectedly, POA-AH differential expression was more strongly coupled to behavioural tactic than morph. This included a suite of transcripts implicated in hormonal regulation of vertebrate social behaviour. Our results reveal that divergent expression patterns in a conserved neuroendocrine centre known to regulate social-reproductive behaviours across vertebrate lineages may be uncoupled from developmental history to enable plasticity in the performance of reproductive tactics.

Is the kisspeptin system involved in responses to food restriction in order to preserve reproduction in pubertal male sea bass (Dicentrarchus labrax)?

Previous works on European sea bass have determined that long-term exposure to restrictive feeding diets alters the rhythms of some reproductive/metabolic hormones, delaying maturation and increasing apoptosis during gametogenesis. However, exactly how these diets affect key genes and hormones on the brain-pituitary-gonad (BPG) axis to trigger puberty is still largely unknown. We may hypothesize that all these signals could be integrated, at least in part, by the kisspeptin system. In order to capture a glimpse of these regulatory mechanisms, kiss1 and kiss2 mRNA expression levels and those of their kiss receptors (kiss1r, kiss2r) were analyzed in different areas of the brain and in the pituitary of pubertal male sea bass during gametogenesis. Furthermore, other reproductive hormones and factors as well as the percentage of males showing full spermiation were also analyzed. Treated fish fed maintenance diets provided evidence of overexpression of the kisspeptin system in the main hypophysiotropic regions of the brain throughout the entire sexual cycle. Conversely, Gnrh1 and gonadotropin pituitary content and plasma sexual steroid levels were downregulated, except for Fsh levels, which were shown to increase during spermiation. Treated fish exhibited lower rates of spermiation as compared to control group and a delay in its accomplishment. These results demonstrate how the kisspeptin system and plasma Fsh levels are differentially affected by maintenance diets, causing a retardation, but not a full blockage of the reproductive process in the teleost fish European sea bass. This suggests that a hormonal adaptive strategy may be operating in order to preserve reproductive function in this species.

Sexual Fate Reprogramming in the Steroid-Induced Bi-Directional Sex Change in the Protogynous Orange-Spotted Grouper, Epinephelus coioides

Androgen administration has been widely used for masculinization in fish. The mechanism of the sex change in sexual fate regulation is not clear. Oral administration or pellet implantation was applied. We orally applied an aromatase inhibitor (AI, to decrease estrogen levels) and 17α-methyltestosterone (MT, to increase androgen levels) to induce masculinization to clarify the mechanism of the sex change in the protogynous orange-spotted grouper. After 3 mo of AI/MT administration, male characteristics were observed in the female-to-male sex change fish. These male characteristics included increased plasma 11-ketotestosterone (11-KT), decreased estradiol (E2) levels, increased male-related gene (dmrt1, sox9, and cyp11b2) expression, and decreased female-related gene (figla, foxl2, and cyp19a1a) expression. However, the reduced male characteristics and male-to-female sex change occurred after AI/MT-termination in the AI- and MT-induced maleness. Furthermore, the MT-induced oocyte-depleted follicle cells (from MT-implantation) had increased proliferating activity, and the sexual fate in a portion of female gonadal soma cells was altered to male function during the female-to-male sex change. In contrast, the gonadal soma cells were not proliferative during the early process of the male-to-female sex change. Additionally, the male gonadal soma cells did not alter to female function during the male-to-female sex change in the AI/MT-terminated fish. After MT termination in the male-to-female sex-changed fish, the differentiated male germ cells showed increased proliferating activities together with dormancy and did not show characteristics of both sexes in the early germ cells. In conclusion, these findings indicate for the first time in a single species that the mechanism involved in the replacement of soma cells is different between the female-to-male and male-to-female sex change processes in grouper. These results also demonstrate that sexual fate determination (secondary sex determination) is regulated by endogenous sex steroid levels.

Sexual dimorphism in the hypophysiotropic tyrosine hydroxylase-positive neurons in the preoptic area of the teleost, Clarias batrachus

BACKGROUND

Dopamine (DA) neurons in the anteroventral periventricular nucleus (AVPV) in the preoptic area (POA) of mammals express estrogen receptors, regulate luteinizing hormone (LH) secretion, and show distinct sexual dimorphism. In teleosts, hypophysiotropic DA neurons of the nucleus preopticus periventricularis (NPP), located in the anteroventral POA, express estrogen receptors, innervate LH cells, and emerged as a neuroanatomical substrate for inhibiting LH cells. Interestingly, the NPP and AVPV seem to share several similarities. Whether DAergic neurons in the NPP show sexual dimorphism is, however, not known. Based on the proposed homology to AVPV and previous studies showing greater tyrosine hydroxylase (TH) mRNA and enzyme activity levels in the brain of female catfish, we hypothesize that females have greater number of DAergic neurons in the NPP and correspondingly more TH-immunoreactive fiber innervation of the pituitary.

METHODS

Adult, male and female Clarias batrachus collected during the prespawning phase of their reproductive cycle were used. Fish were anesthetized and perfused transcardially with phosphate-buffered saline (pH 7.4) and 4 % paraformaldehyde in phosphate buffer. Sections through the rostro-caudal extent of the POA and pituitary were processed for TH immunofluorescence. Using double immunofluorescence, the association between TH-immunoreactive fibers and LH cells in the pituitary was explored. Sections were analyzed using semiquantitative analysis.

RESULTS

NPP in POA of C. batrachus has two distinct subdivisions, viz, anterior (NPPa) and posterior (NPPp), and TH neurons were observed in both the subdivisions. Compared to that in the males, a significantly higher (P < 0.05) number of TH neurons was consistently observed in the NPPa of females. TH neurons in NPPp, however, showed no difference in the number or immunoreactivity. Since DA neurons in NPPa are hypophysiotropic, we compared TH-fiber innervation of the pituitary in both sexes. Compared to males, proximal pars distalis and LH cells in this region of the pituitary in females were densely innervated by TH fibers.

CONCLUSIONS

Neurons of NPPa and their innervation to the pituitary seem to be a distinct sexually dimorphic DAergic system in C. batrachus. The DAergic system may serve as a component of the neural mechanisms controlling the sexually dimorphic LH surge in teleosts. Given the similarities shared by NPPa and AVPV, homology between these two nuclei is suggested.

Catecholaminergic Fiber Innervation of the Vocal Motor System Is Intrasexually Dimorphic in a Teleost with Alternative Reproductive Tactics

Catecholamines, which include the neurotransmitters dopamine and noradrenaline, are known modulators of sensorimotor function, reproduction, and sexually motivated behaviors across vertebrates, including vocal-acoustic communication. Recently, we demonstrated robust catecholaminergic (CA) innervation throughout the vocal motor system in the plainfin midshipman fish Porichthys notatus, a seasonal breeding marine teleost that produces vocal signals for social communication. There are 2 distinct male reproductive morphs in this species: type I males establish nests and court females with a long-duration advertisement call, while type II males sneak spawn to steal fertilizations from type I males. Like females, type II males can only produce brief, agonistic, grunt type vocalizations. Here, we tested the hypothesis that intrasexual differences in the number of CA neurons and their fiber innervation patterns throughout the vocal motor pathway may provide neural substrates underlying divergence in reproductive behavior between morphs. We employed immunofluorescence (-ir) histochemistry to measure tyrosine hydroxylase (TH; a rate-limiting enzyme in catecholamine synthesis) neuron numbers in several forebrain and hindbrain nuclei as well as TH-ir fiber innervation throughout the vocal pathway in type I and type II males collected from nests during the summer reproductive season. After controlling for differences in body size, only one group of CA neurons displayed an unequivocal difference between male morphs: the extraventricular vagal-associated TH-ir neurons, located just lateral to the dimorphic vocal motor nucleus (VMN), were significantly greater in number in type II males. In addition, type II males exhibited greater TH-ir fiber density within the VMN and greater numbers of TH-ir varicosities with putative contacts on vocal motor neurons. This strong inverse relationship between the predominant vocal morphotype and the CA innervation of vocal motor neurons suggests that catecholamines may function to inhibit vocal output in midshipman. These findings support catecholamines as direct modulators of vocal behavior, and differential CA input appears reflective of social and reproductive behavioral divergence between male midshipman morphs.

Sexually dimorphic distribution of calcium-binding protein, calretinin in the preoptic area of the freshwater catfish, Clarias batrachus (Linn.)

Preoptic area (POA) plays an important role in the hormonal regulation of the pituitary gland in vertebrates. In this study we report the sexually dimorphic distribution of calcium-binding proteins calretinin (CR) in the POA in the freshwater catfish, Clarias batrachus. Nissl staining highlighted the presence of the nucleus praeopticus periventricularis (NPP) and other subdivisions of the nucleus praeopticus (NPO), including supraoptic (NPOs), paraventricular (NPOp) and magnocellular (NPOm) divisions. In NPO, CR immunoreactivity was noted only in females but not in males. In both sexes, CR stained perikarya were found in the NPP. Sexually dimorphic localization of CR in the POA supports the notion that CR may play a gender-specific role and may be involved in hormonal regulation in fishes.

Morphometric changes associated with the reproductive cycle and behaviour of the intertidal-nesting, male plainfin midshipman Porichthys notatus

Morphometric changes in body condition, liver, sonic muscle and gonadal development associated with the annual reproductive cycle and behaviour of the intertidal-nesting male plainfin midshipman Porichthys notatus were investigated. Body condition of type I males rapidly increased during the pre-nesting (PN) period, peaked at the beginning of the summer nesting cycle and then gradually declined to lowest levels during the non-reproductive (NR) period. The gonado-somatic index of type I males peaked during PN and then declined during the summer nesting cycle to lowest levels at the end of the nest cycle and during NR. Indices of sonic muscle and liver of type I males were lowest during NR, gradually increased during PN and then peaked during the summer nesting cycle. Results indicate that body condition and fecundity of type I males were positively correlated with body mass at the end of the nest cycle. These findings as they relate to the annual reproductive cycle and behaviour of the type I male P. notatus are discussed.

Reproductive-tactic-specific variation in sperm swimming speeds in a shell-brooding cichlid

Theory predicts that males experiencing elevated levels of sperm competition will invest more in gonads and produce faster-swimming sperm. Although there is ample evidence in support of the first prediction, few studies have examined sperm swimming speed in relation to sperm competition. In this study, we tested these predictions from sperm competition theory by examining sperm characteristics in Telmatochromis vittatus, a small shell-brooding cichlid fish endemic to Lake Tanganyika. Males exhibit four different reproductive tactics: pirate, territorial, satellite, and sneaker. Pirate males temporarily displace all other competing males from a shell nest, whereas sneaker males always release sperm in the presence of territorial and satellite males. Due to the fact that sneakers spawn in the presence of another male, sneakers face the highest levels of sperm competition and pirates the lowest, whereas satellites and territorials experience intermediate levels. In accordance with predictions, sperm from sneakers swam faster than sperm from males adopting the other reproductive tactics, whereas sperm from pirates was slowest. Interestingly, we were unable to detect any variation in sperm tail length among these reproductive tactics. Thus, sperm competition appears to have influenced sperm energetics in this species without having any influence on sperm size.

Saccular potentials of the vocal plainfin midshipman fish, Porichthys notatus

The plainfin midshipman fish, Porichthys notatus, is a vocal species of teleost fish that generates acoustic signals for intraspecific communication during social and reproductive behaviors. All adult morphs (females and males) produce single short duration grunts important for agonistic encounters, but only nesting males produce trains of grunts and growls in agonistic contexts and long duration multiharmonic advertisement calls to attract gravid females for spawning. The midshipman fish uses the saccule as the main acoustic endorgan for hearing to detect and locate vocalizing conspecifics. Here, I examined the response properties of evoked potentials from the midshipman saccule to determine the frequency response and auditory threshold sensitivity of saccular hair cells to behaviorally-relevant single tone stimuli. Saccular potentials were recorded from the rostral, medial and caudal regions of the saccule while sound was presented by an underwater speaker. Saccular potentials of the midshipman, like other teleosts, were evoked greatest at a frequency that was twice the stimulus frequency. Results indicate that midshipman saccular hair cells of non-reproductive adults had a peak frequency sensitivity that ranged from 75 (lowest frequency tested) to 145 Hz and were best suited to detect the low frequency components (<or=105 Hz) of midshipman vocalizations.

Immunoneutralization of gonadotropin-releasing hormone type-III suppresses male reproductive behavior of cichlids

To investigate the roles of gonadotropin-releasing hormone (GnRH) types in reproductive behaviors, antisera against GnRH1, GnRH2 and GnRH3 were stereotaxically administered into the intracerebroventricular region to neutralize the three native GnRH types in the brain of male tilapia Oreochromis niloticus. Reproductive behaviors (nest-building and aggressive behaviors), and morphological changes of the three GnRH systems were investigated by immunocytochemistry. GnRH1, GnRH2 and GnRH3 immunoreactive fibers were significantly decreased following injections of GnRH antisera indicating successful neutralization of their respective endogenous GnRH peptides. GnRH1- and GnRH2-immunoneutralization did not inhibit reproductive behaviors but GnRH3-immunoneutralization significantly decreased nest-building ability (Saline: 26.5 +/- 3.7%/day versus GnRH3: 6.1 +/- 2.9%/day, P < 0.001), nest size (Saline: 0.67 +/- 0.09 points versus GnRH3: 0.10 +/- 0.05 points, P < 0.0002) and aggressive behavior (Saline: 2.34 +/- 0.19 points versus GnRH3 1.06 +/- 0.12 points, P < 0.0001). These observations provide evidence that GnRH3 is a potent neuromodulator of reproductive behaviors in male tilapia.

Variability of GnRH secretion in two goby species with socially controlled alternative male mating tactics

Male reproductive phenotypic plasticity related to environmental-social conditions is common among teleost fish. In several species, males adopt different mating tactics depending on their size, monopolizing mates when larger, while parasitizing dominant male spawns when smaller. Males performing alternative mating tactics are often characterized by a strong dimorphism in both primary and secondary reproductive traits. According to studies on sex-changing species and on species where only one male morph is reproductively active, male alternative phenotypes are expected to vary also in gonadotropin-releasing hormone (GnRH) neurons in forebrain preoptic area (POA). Here, we compared the intra- and inter-sexual variations in number and size of GnRH neurons, along with gonads and male accessory structure investment, in two goby species, the grass goby, Zosterisessor ophiocephalus, and the black goby, Gobius niger, characterized by male alternative mating phenotypes. In both species, older and larger males defend nests, court and perform parental care, while younger and smaller ones try to sneak territorial male spawning. We found that grass goby and black goby have different patterns of GnRH expression. Grass goby presents a clear intra-sexual dimorphism in GnRH expression, related to the occurrence of alternative mating tactics, while in the black goby, only inter-sexual differences are observed. The inter- and intra-specific variability in the GnRH neurons in these two goby species is discussed in light of the differences in migratory behavior, nest type, and mating system.

Sonic and electric fish: at the crossroads of neuroethology and behavioral neuroendocrinology

Field and laboratory studies of weakly electric and sound-producing teleost fishes demonstrate how steroidal and non-steroidal hormones mediate the translation of neural events into behavior. The development of this research program has depended upon an interdisciplinary neuroethological approach that has characterized the neurophysiological properties of the motor and sensory pathways that lead to the production and detection of easily quantified highly stereotyped behaviors, namely, electric organ discharges (EODs) and vocalizations. Neuroethological studies of these teleosts have now integrated a behavioral neuroendocrinology approach that has provided several examples of how hormone-sensitive neurobiological traits contribute to adaptive behavioral plasticity in natural habitats. As such, these studies provide guideposts for comparable studies in other groups of teleosts and vertebrates in general.

Ontogenetic changes in the response properties of individual, primary auditory afferents in the vocal plainfin midshipman fish Porichthys notatus Girard

The auditory system of adult midshipman fish Porichthys notatusGirard is an important sensory receiver system used during intraspecific social communication to encode conspecific vocalizations, but the response properties and function of this system in the pre-adult stages are unknown. Midshipman fish, like other teleosts, use the saccule as the main acoustic end organ of the inner ear. In this study, we examined the discharge properties and the frequency response dynamics of auditory saccular afferent neurons in pre-adult midshipman (∼4–12 months of age) to determine whether encoding of auditory information, inclusive of conspecific vocalizations,changes across life history stages. Extracellular single unit recordings were made from saccular afferents while sound was presented via an underwater speaker. Comparisons with adult data show that the resting discharge rate and auditory threshold sensitivity increased with age/size,while temporal encoding of frequency did not show any significant shifts. The results indicate that the saccular afferents of juveniles, like those of non-reproductive adults, are best adapted to temporally encode the low frequency components (≤100 Hz) of midshipman vocalizations. This report represents the first in vivo investigation of age-related changes in the encoding properties of individual auditory neurons for any fish species.

Seasonal variation of steroid hormone levels in an intertidal-nesting fish, the vocal plainfin midshipman

This study characterized the seasonal variation of the steroid hormones testosterone (T), 11-ketotestosterone (11-KT), 17beta-estradiol (E2), and cortisol (F) as they relate to the gonadal development and reproductive behavior of the plainfin midshipman fish, Porichthys notatus. The plainfin midshipman is a deep-water teleost that seasonally migrates into the shallow intertidal zone where type I, or "singing," males build nests, acoustically court and spawn with females. The gonadosomatic index and plasma steroid levels were measured from adult type I males and females collected over four time periods (non-reproductive, pre-nesting, nesting, and post-nesting) that corresponded to seasonal fluctuations in midshipman reproductive biology and behavior. Among type I males, plasma levels of T and 11-KT were low during the winter non-reproductive period, gradually increased during seasonal recrudescence of the testes in the spring pre-nesting period, and then peaked at the beginning of the summer nesting period. In the latter half of the nesting period and during the fall post-nesting period, plasma levels of T and 11-KT were low or non-detectable. Low, detectable levels of E2 were also found in the plasma of 50% or more type I males during every seasonal period except the winter non-reproductive period. Among females, plasma levels of T and E2 were low throughout the year but briefly peaked in April during the spring pre-nesting period when ovaries underwent seasonal recrudescence. Plasma F levels were correlated with collection depth and were lower in males than females when fish were collected deeper than 120 m. The sex-specific peaks of steroid hormone levels in male and female midshipman may serve differential functions related to the physiology, reproductive behavior, and vocal communication of this species.

Development of galanin-like immunoreactivity in the brain of the brown trout (Salmo trutta fario), with some observations on sexual dimorphism

The development of galanin-like immunoreactive (GAL-ir) cells and fibers was investigated in the brain of brown trout embryos, alevins, juveniles, and adults (some spontaneously releasing their gametes). The earliest GAL-ir neurons appeared in the preoptic region and the primordial hypothalamic lobe of 12-mm embryos. After hatching, new GAL-ir neurons appeared in the lateral, anterior, and posterior tuberal nuclei, and in late alevins, GAL-ir neurons appeared in the area postrema. In juveniles, further GAL-ir populations appeared in the nucleus subglomerulosus and magnocellular preoptic nucleus. The GAL-ir neuronal groups present in juveniles were also observed in sexually mature adults, although the area postrema of males lacked immunoreactive neurons. Moreover, spawning males exhibited GAL-ir somata in the olfactory bulb and habenula, which were never observed in adult females or in developing stages. In adults, numerous GAL-ir fibers were observed in the ventral telencephalon, preoptic area, hypothalamus, neurohypophysis, mesencephalic tegmentum, ventral rhombencephalon, and area postrema. Moderate to low GAL-ir innervation was seen in the olfactory bulbs, dorsomedial telencephalon, epithalamus, medial thalamus, optic tectum, cerebellum, and rhombencephalic alar plate. There were large differences among regions in the GAL-ir innervation establishment time. In embryos, GAL-ir fibers appeared in the preoptic area and hypothalamus, indicating early expression of galanin in hypophysiotrophic centers. The presence of galanin immunoreactivity in the olfactory, reproductive, visual, and sensory-motor centers of the brain suggest that galanin is involved in many other brain functions. Furthermore, the distribution of GAL-ir elements observed throughout trout development indicates that galaninergic system maturation continues until sexual maturity.

Neurochemical correlates of male polymorphism and alternative reproductive tactics in the Azorean rock-pool blenny, Parablennius parvicornis

In the common Azorean rock-pool blenny, Parablennius parvicornis, males exhibit alternative reproductive morphologies: (1) larger males defend nest sites, provide parental care, have anal glands (involved in pheromone release), testicular glands, and low gonad:body weight ratio (GSI) and (2) smaller, younger, males do not defend nests, have reduced glands and high GSI. These smaller non-nesting males behave as satellites (associated with nests) or sneakers (moving among nests), attempting to achieve parasitic fertilizations via sperm competition. In non-mammals, arginine vasotocin (AVT) is a key hypothalamic peptide involved in the control of reproductive behavior and physiology, and several fish species that exhibit alternative male reproductive morphs show polymorphism in AVT brain chemistry. We conducted an immunocytochemical study to generate comparative data on this intertidal blenny. Our analysis showed no difference in AVT-immunoreactive cell number or size between the male morphs, which is consistent with studies on other fish, including blennies. The number of AVT cells was positively correlated to fish body mass, while cell size showed no such relation. If corrected for body mass, the smaller non-nesting males have significantly more cells than the large nesting males. Our data suggest that the size and number of forebrain AVT cells develops initially to allow for reproduction in the young non-nesting males and this pattern does not appear to change when males take on the nesting morphotype later in life. This result appears to be consistent in many fishes with alternative male morphotypes.

Neural mechanisms and behaviors for acoustic communication in teleost fish

Sound communication is not unique to humans but rather is a trait shared with most non-mammalian vertebrates. A practical way to address questions of vocal signal encoding has been to identify mechanisms in non-mammalian model systems that use acoustic communication signals in their social behavior. Teleost fishes, the largest group of living vertebrates, include both vocal and non-vocal species that exploit a wide range of acoustic niches. Here, we focus on those vocal species where combined behavioral and neurobiological studies have recently begun to elucidate a suite of adaptations for both the production and the perception of acoustic signals essential to their reproductive success and survival. Studies of these model systems show that teleost fish have the vocal-acoustic behaviors and neural systems both necessary and sufficient to solve acoustic problems common to all vertebrates. In particular, behavioral studies demonstrate that temporal features within a call, including pulse duration, rate and number, can all be important to a call's communicative value. Neurobiological studies have begun to show how these features are produced by a vocal motor system extending from forebrain to hindbrain levels and are encoded by peripheral and central auditory neurons. The abundance and variety of vocal fish present unique opportunities for parallel investigations of neural encoding, perception, and communication across a diversity of natural, acoustic habitats. As such, investigations in teleosts contribute to our delineating the evolution of the vocal and auditory systems of both non-mammalian and mammalian species, including humans.

Vocal-acoustic circuitry and descending vocal pathways in teleost fish: convergence with terrestrial vertebrates reveals conserved traits

Vocal behavior is multifaceted and requires that vocal-motor patterning be integrated at multiple brain levels with auditory, neuroendocrine, and other social behavior processes (e.g., courtship and aggression). We now provide anatomical evidence for an extensive vocal network in teleost fishes (Batrachoididae: Porichthys notatus; Opsanus beta) that is strongly integrated with neuroendocrine and auditory pathways and that exhibits striking similarities to the vocal-acoustic circuitry known for mammals. Biotin compound injections into neurophysiologically identified vocal regions of the forebrain (preoptic area and anterior hypothalamus) and of the midbrain (periaqueductal gray and paralemniscal tegmentum) reveal extensive connectivity within and between these regions, as well as reciprocal relationships with the auditory thalamus and/or auditory midbrain (torus semicircularis). Thus, specific components of the basal forebrain and midbrain are here designated as the forebrain vocal-acoustic complex (fVAC) and midbrain vocal-acoustic complex (mVAC), respectively. Biotin injections into the mVAC and a previously identified hindbrain vocal pattern generator likewise provide anatomical evidence for a distributed network of descending projections to the vocal pacemaker-motoneuron circuitry. Together, the present experiments establish a vocal-auditory-neuroendocrine network in teleost fish that links the forebrain and midbrain to the hindbrain vocal pattern generator (i.e., fVAC --> mVAC --> pattern generator) and provides an anatomical framework for the previously identified neuropeptide modulation of vocal activity elicited from the forebrain and midbrain, which contributes to the expression of sex- and male morph-specific behavior. We conclude with a broad comparison of these findings with those for other vertebrate taxa and suggest that the present findings provide novel insights into the structure of conserved behavioral regulatory circuits that have led to evolutionary convergence in vocal-acoustic systems.

Variation in reproductive behaviour within a sex: neural systems and endocrine activation

Intrasexual variation in reproductive behaviour, morphology and physiology is taxonomically widespread in vertebrates, and is as biologically and ecologically significant as the differences between the sexes. In this review, we examine the diverse patterns of intrasexual variation in reproductive behaviours within vertebrates. By illustrating the genetic, cellular, hormonal and/or neural mechanisms underlying behavioural variation in a number of species, another level of complexity is added to studies of brain organization and function. Such information increases our understanding of the unique and conserved mechanisms underlying sex and individual differences in behaviour in vertebrates as a whole. Here, we show that intrasexual variation in behaviour may be discrete or continuous in nature. Moreover, this variation may be due to polymorphism at a single genetic locus or many loci, or may even be the result of phenotypic plasticity. Phenotypic plasticity simply refers to cases where a single genotype (or individual) can produce (or display) different phenotypes. Defined in this way, plasticity subsumes many different types of behavioural variation. For example, some behavioural phenotypes are established by environmental factors during early ontogeny, others are the result of developmental transitions from one phenotype early in life to another later in life, and still other strategies are facultative with different behaviours displayed in different social contexts.

Gonadotropin releasing hormone (GnRH) neurones of triploid catfish, Heteropneustes fossilis (Bloch): an immunocytochemical study

Gonadotropin-releasing hormone (GnRH), a regulator of gonadal maturation in vertebrates, is primarily secreted by neurosecretory cells of the pre-optic area (POA) in the forebrain of teleosts. GnRH-immunoreactive (GnRH-ir) cells of this area demonstrate positive correlation in number and size of soma with gonadal maturity and directly innervate the pituitary in most teleosts. Gonadal development in triploid fish remains impaired due to genetic sterility. The gonadal immaturity in triploid fish may be due to low levels of gonadotropin and sex steroids during the vitellogenic phase of reproductive cycle. However, the nature of GnRH-ir cells in triploid fish is not yet known. Triploid catfish (H. fossilis) showed significant decrease (P<0.001) in size and number of immunoreactive-GnRH cells of POA and low immunoreactivity in pituitary in comparison to their diploid full-sibs during the late pre-spawning phase of ovarian cycle. This study suggests that low activity of GnRH-cells in triploid may be due to lack of positive feedback stimulation by sex steroids and/or reduced responsiveness of sensory cells to environmental cues required for gonadal maturation in teleosts.

Sex differences in the brain of goldfish: gonadotropin-releasing hormone and vasotocinergic neurons

The differences between male and female behaviors are reflected in sexual dimorphism of brain structures and are found throughout the nervous system in a variety of vertebrates. The present study examined neurons immunolabeled for gonadotropin-releasing hormone and arginine vasotocin in the brain of the goldfish Carassius auratus to determine if these neurons are sexually dimorphic. There was no sex difference or influence of sex steroids on the neuronal volume and optical density of staining of arginine vasotocin neurons. Similarly, gonadotropin-releasing hormone neurons of the terminal nerve and midbrain tegmentum did not differ between sexually mature males, females and maturing females replaced with sex steroids with respect to distribution, numbers, optical density of staining, or gross morphology. In maturing females, testosterone specifically recruited additional preoptic gonadotropin-releasing hormone neurons to equal those in sexually mature individuals. Since estrogen had no effect, the influence of testosterone on gonadotropin-releasing hormone neuronal numbers appears to be independent of aromatization. Specifically, the preoptic gonadotropin-releasing hormone neuronal size was significantly larger in sexually mature males than females. 11-Ketotestosterone-replacement to ovariectomized maturing females induced male-typical secondary characters and male-type courtship behavior but did not masculinize the preoptic gonadotropin-releasing hormone neuronal size. Our results show that the sexually dimorphic preoptic gonadotropin-releasing hormone neuronal size is determined by factors (genetic) other than gonadal steroids. Further, we propose the hypothesis that phenotypic and behavioral sex differences need not be accompanied by structural differences in gonadotropin-releasing hormone and arginine vasotocin in the brain.

Gonadotropin-releasing hormone-immunoreactive neurons and associated nicotinamide adenine dinucleotide phosphate-diaphorase-positive neurons in the brain of a teleost, Rhodeus amarus

Using combined nicotinamide adenine dinucleotide phosphate-diaphorase (NADPHd) histochemistry and salmon gonadotropin-releasing hormone (sGnRH) immunocytochemistry, it is reported for the first time that possible potential contacts occur between the nitric oxide (NO)- and the GnRH-containing neurons in the brain of a freshwater teleost, Rhodeus amarus. GnRH-immunoreactive (ir) neurons were observed in the olfactory nerve (OLN), olfactory bulb (OB), medial olfactory tract (MOT), ventral telencephalon (VT), nucleus preopticus periventricularis (NPP), nucleus lateralis tuberis (NLT), and midbrain tegmentum (MT). Although NADPHd neurons were widely distributed in the brain, only those having an association with GnRH-ir neurons are described. Based on the nature of the association between the GnRH and the NADPHd neurons, the former were classified into three types. The Type I GnRH neurons were characterized by the presence of NADPHd-positive granules in the perikarya and processes and occurred in the OLN, OB, MOT, and VT. The Type II GnRH neurons, having soma-soma or soma-process contacts with the NADPHd neurons, were restricted to the MT; the long processes of NADPHd cells crossed over either the perikarya or the thick processes of GnRH cells. However, the Type III GnRH neurons, found in the NPP and NLT, did not show direct contact, but a few NADPHd fibers were present in the vicinity. The terminal-soma contacts in the olfactory system and the VT and the soma-soma contacts in the MT represent the sites of possible potential contacts indicating a direct NO involvement in GnRH function, although NO action by diffusion remains possible. NO may influence the NPP and NLT GnRH cells by diffusion only, since a direct contact was not observed.

Vasotocin innervation and modulation of vocal-acoustic circuitry in the teleost Porichthys notatus

Arginine vasotocin (AVT) and its mammalian homologue arginine vasopressin (AVP) modulate reproduction-related and other social behaviors in a broad range of vertebrate species. These functions of AVT/AVP may be in part achieved through the modulation of sensorimotor integration, although experimental evidence supporting this hypothesis remains limited. In the present experiments, we demonstrate (1) AVT innervation of candidate vocal-acoustic brain regions, and (2) AVT modulation of vocal-motor physiology in the plainfin midshipman fish (Porichthys notatus), which uses vocalizations in both mate attraction and agonistic contexts. AVT distribution was compared with known vocally active brain regions and to central auditory and vocal pathways. AVT-immunoreactive fibers and putative terminals descend almost exclusively from the preoptic area and are found in two primary candidate sites for vocal-acoustic integration - the anterior tuberal hypothalamus and paralemniscal midbrain tegmentum. AVT immunoreactivity is also located in several other vocally active regions, including the ventral tuberal nucleus, periaqueductal gray, and paraventricular regions of the isthmus and rostral hindbrain. The parvocellular preoptic area itself is also vocally active, although thresholds are substantially higher than for other regions. The functional significance of AVT input to vocal-acoustic regions was demonstrated in the paralemniscal midbrain where local delivery of AVT modulated electrically evoked, rhythmic vocal-motor output, which precisely mimicked natural vocalizations. AVT produced dose-dependent inhibitions of parameters associated with call initiation (burst latency and number of vocal-motor bursts elicited) but not of vocal-motor patterning (fundamental frequency and burst duration). Together, these findings provide support for the proposal that AVT modulates sensorimotor processes underlying social/acoustic communication.

Differential distribution of gonadotropin-releasing hormone-immunoreactive neurons in the stingray brain: functional and evolutionary considerations

Gonadotropin-releasing hormone (GnRH) is a neuropeptide that occurs in multiple structural forms among vertebrate species. Bony fishes, amphibians, reptiles, birds, and mammals express different forms of GnRH in the forebrain and endocrine regions of the hypothalamus which regulate the release of reproductive gonadotropins from the pituitary. In contrast, previous studies on bony fishes and tetrapods have localized the chicken GnRH-II (cGnRH-II) nucleus in the midbrain tegmentum and, combined with cladistic analyses, indicate that cGnRH-II is the most conserved form throughout vertebrate evolution. However, in elasmobranch fishes, the neuroanatomical distribution of cGnRH-II and dogfish GnRH (dfGnRH) cells and their relative projections in the brain are unknown. We used high-performance liquid chromatography and radioimmunoassay to test for differential distributions of various GnRH forms in tissues from the terminal nerve (TN) ganglia, preoptic area, and midbrain of the Atlantic stingray, Dasyatis sabina. These experiments identified major peaks that coelute with cGnRH-II and dfGnRH, minor peaks that coelute with lamprey GnRH-III (lGnRH-III), and unknown forms. Immunocytochemistry experiments on brain sections show that dfGnRH-immunoreactive (-ir) cell bodies are localized in the TN ganglia, the caudal ventral telencephalon, and the preoptic area. Axons of these cells project to regions of the hypothalamus and pituitary, diencephalic centers of sensory and behavioral integration, and the midbrain. A large, discrete, bilateral column of cGnRH-II-ir neurons in the midbrain tegmentum has sparse axonal projections to the hypothalamus and regions of the pituitary but numerous projections to sensory processing centers in the, midbrain and hindbrain. Immunocytochemical and chromatographic data are consistent with the presence of lGnRH-III and other GnRH forms in the TN that differ from dfGnRH and cGnRH-II. This is the first study that shows differential distribution of cGnRH-II and dfGnRH in the elasmobranch brain and supports the hypothesis of divergent function of GnRH variants related to gonadotropin control and neuromodulation of sensory function.

Differences in mGnRH and cGnRH-II contents in pituitaries and discrete brain areas of Rana rugulosa W. according to age and stage of maturity

(1) In tadpoles, chicken-II gonadotropin-releasing hormone (cGnRH-II) could be measured in the brains before metamorphosis, but mammalian gonadotropin-releasing hormone (mGnRH) did not appear until the stage of metamorphosis, i.e. cGnRH-II appeared earlier than mGnRH during ontogenesis. (2) During the metamorphic climax, mGnRH content increased more rapidly than cGnRH-II; the content of mGnRH was about two times of that of cGnRH-II. (3) In juveniles and adults, the content of mGnRH and cGnRH-II, and the distribution pattern of mGnRH (but not cGnRH-II) in the brains and pituitaries changed with age and stages of gonadal development. mGnRH mainly distributed in the rostral brain areas, whereas cGnRH-II had a widespread distribution in the brain. (4) Both mGnRH and cGnRH-II were present in the pituitaries at each stage of maturity. The gonadotropin-releasing hormone (GnRH) content at sexually maturity was significantly higher than that at other stages of gonadal development, and the content of mGnRH was about 15-18 times of that of cGnRH-II. (5) These results suggest that both mGnRH and cGnRH-II are potentially involved in the direct regulation of pituitary gonadotropes, and mGnRH may be the major active form, cGnRH-II may also serve as a neurotransmitter or neuromodulator in the brain.

Preoptic GnRH and AVT: axes for sexual plasticity in teleost fish

Alternative reproductive tactics within one sex, adult sex or role change, and reproductive suppression are all forms of reproductive plasticity commonly exhibited among teleost fishes. The two neuropeptides that have been most extensively studied with regard to such behavioral plasticity are gonadotropin releasing hormone (GnRH) and arginine vasotocin (AVT). Here, we review intra- and intersexual variation in the number and size of GnRH and AVT neurons along with gonadal phenotype in those species of teleosts showing intraspecific plasticity in reproductive behavior. In several species, male dimorphisms in the number and/or size of GnRH neurons in the forebrain's preoptic area parallel a divergence in relative gonad size and reproductive tactics. The available studies of AVT-containing neurons in the preoptic area also indicate intrasexual dimorphisms among males, although a proximate link to other reproductive traits and behavioral outcomes is more difficult to recognize. For both GnRH and AVT, there are also species-typical patterns in the coupling between structural (e.g., neuronal and gonadal) traits and reproductive tactic expressed, which likely reflect distinct patterns of adaptation to particular ecological environments. As discussed, neurophysiological, biochemical, and receptor density studies are now essential to establish the functional significance of the diverse organizational patterns of GnRH and AVT neurons in teleosts. Similar studies also need to be carried out in species of other vertebrate groups that show comparable behavioral plasticity.

Adaptation, Exaptation, and Constraint: A Hormonal Perspective

We approach conceptual issues in evolutionary biology from an endocrinological perspective, noting that single hormones typically act on several target tissues and thereby mediate suites of correlated phenotypic traits. When several components of such a suite are beneficial, an important evolutionary question is whether all are adaptations or some are exaptations. The answer may depend on whether the traits arose in response to selection on variation in systemic levels of the hormone on variation in responsiveness of target tissues to invariant levels of the hormone. If the former, selection probably acted directly on fewer than all traits; beneficial traits arising indirectly would be exaptations. In contrast, multiple beneficial traits that arose out of independent changes in target-tissue sensitivity to invariant hormone levels could all be adaptations. Knowledge of specific hormonal mechanisms as well as of historical selective regimes will be necessary to draw such distinctions. Endocrine constraints on evolution can be studied experimentally by applying hormones systemically and measuring interdependent responses of beneficial and detrimental traits to selection (phenotypic engineering with hormones). Supposing that alteration of one trait in isolation would enhance fitness, cases in which the net effect of endocrine alteration of multiple traits is to depress fitness provide evidence for constraints. We briefly report results of recent studies employing hormonal manipulations, stressing our own work on the dark-eyed junco (Junco hyemalis: Emberizidae).

Sexual differentiation of the vertebrate brain: principles and mechanisms

A wide variety of sexual dimorphisms, structural differences between the sexes, have been described in the brains of many vertebrate species, including humans. In animal models of neural sexual dimorphism, gonadal steroid hormones, specifically androgens, play a crucial role in engendering these differences by masculinizing the nervous system of males. Usually, the androgen must act early in life, often during the fetal period to masculinize the nervous system and behavior. However, there are a few examples of androgen, in adulthood, masculinizing both the structure of the nervous system and behavior. In the modal pattern, androgens are required both during development and adulthood to fully masculinize brain structure and behavior. In rodent models of neural sexual dimorphism, it is often the aromatized metabolites of androgen, i.e., estrogens, which interact with estrogen receptors to masculinize the brain, but there is little evidence that aromatized metabolites of androgen play this role in primates, including humans. There are other animal models where androgens themselves masculinize the nervous system through interaction with androgen receptors. In the course of masculinizing the nervous system, steroids can affect a wide variety of cellular mechanisms, including neurogenesis, cell death, cell migration, synapse formation, synapse elimination, and cell differentiation. In animal models, there are no known examples where only a single neural center displays sexual dimorphism. Rather, each case of sexual dimorphism seems to be part of a distributed network of sexually dimorphic neuronal populations which normally interact with each other. Finally, there is ample evidence of sexual dimorphism in the human brain, as sex differences in behavior would require, but there has not yet been any definitive proof that steroids acting early in development directly masculinize the human brain.

Preoptic AVT immunoreactive neurons of a teleost fish with alternative reproductive tactics

Recent reports have implicated an important role for arginine vasotocin (AVT) in the socially mediated sexual differentiation of fishes. This study focuses on the plainfin midshipman (Porichthys notatus) which exhibits two male morphs, type I and type II, differing in a suite of behavioral, neurobiological, and endocrine traits. Immunocytochemical techniques were used to detect neurons containing AVT-like peptide in the forebrain of juveniles, adult females, and type I and type II males. AVT immunoreactive (ir) somata were localized to three regions: the terminal nerve ganglion, the preoptic area (POA), and the pineal stalk. The profile area, or size, of AVT-ir POA neurons differed across the four classes of midshipman and was strongly correlated to differences in body size among the groups. By contrast, the number of AVT-ir cells in the POA exhibited no difference across the classes of midshipman. The number of POA cells containing AVT is therefore likely to be set early in development and not to change with the growth of the animal. An analysis of AVT-ir cell number normalized by body mass revealed that the larger morphs, type I males and females, have fewer cells per gram body mass than type II males and juveniles. Therefore, type II males have a juvenile-like AVT POA phenotype with smaller cells and more numerous cells per unit body mass than type I males. Type II males also exhibit more variability in the number of AVT-ir cells found in the POA compared to type I males.

Fish as models for the neuroendocrine regulation of reproduction and growth

Models are essential for the full understanding of neuroendocrine control processes. In this regard fish offer a rich source of biological material. They have diverse growth and reproductive strategies, inhabiting most of the Earth's aquatic ecological niches. Fish possess many of the common vertebrate features but also offer several unique aspects to allow the biologist easy access to the study of hypothalamic and pituitary function. Several key examples of how teleosts, or the bony fish, can offer insight into fundamental mechanisms of vertebrate sex differentiation, growth and reproduction are reviewed.

Sex differences in the nervous system of reptiles

1. The study of sex differences in the brain and behavior of reptiles presents an excellent opportunity both to discern general principles of sexual differentiation in the nervous system and to explore the evolutionary history of this process in amniote vertebrates. 2. Findings in several reptiles suggest that some sex differences found in mammals and birds are conserved while others are not. Conserved features include areas in the limbic forebrain involved in the regulation of social and sexual behaviors. As in mammals and birds, it is rare to find differences in the distribution of sex steroid concentrating neurons in reptiles but common to find differences in the distribution of the various steroid hormone receptors and in their regulation. 3. This research has revealed that differences in social and sexual behavior are reflected better by the activity, not by the size, of hormone-sensitive limbic areas. 4. Finally, species differences in plasma levels of sex hormones are paralleled by differences in behavioral sensitivity to these hormones as well as by differences in the regulation of genes coding for steroid hormone receptors.

Modification of gonadotropin releasing hormone (GnRH) mRNA expression in the retinal-recipient Thalamus

Although the environmental cues that trigger reproductive behaviors are known for many species, the mechanisms through which these signals influence the neurochemistry of the brain to produce behavior have been elusive. In this study, we describe a retinally modulated system of gonadotropin releasing hormone (GnRH) producing neurons in the thalamus of the plainfin midshipman fish, Porichthys notatus. Previously, we cloned and sequenced the cDNA for prepro-GnRH in midshipman. Here, using in situ hybridization, we localized prepro-GnRH mRNA to the ventrolateral nucleus of the thalamus, three divisions of the preoptic area, the ganglion of the terminal nerve, and the olfactory bulb. Since the thalamus, terminal nerve ganglion, and preoptic area have been associated with visual functions, we investigated the retinal connections in midshipman. In particular, biocytin tract tracing delineated a reciprocal connection between the ventrolateral nucleus of the thalamus and the retina. Retinofugal projections are exclusively contralateral. Experimental manipulation of this retinalthalamic loop through complete optic nerve transection shows that GnRH mRNA expression in the contralateral ventrolateral nucleus may be influenced by the retina. We hypothesize that a reciprocal retinothalamic GnRH circuit is important in modulating the expression of seasonal reproductive behaviors.

Independent effects of incubation temperature and gonadal sex on the volume and metabolic capacity of brain nuclei in the leopard gecko (Eublepharis macularius), a lizard with temperature-dependent sex determination

The extent to which variation within and between the sexes can be assigned to genes vs. environment is problematic, because, in most vertebrates, males and females differ genetically. However, factors other than sex chromosomes and the consequent sex-typical gonadal hormone secretions may play important roles in the differentiation of the neural mechanisms underlying individual and sex differences in aggressive and sexual behavior. The leopard gecko, like many oviparous reptiles, lacks sex chromosomes. Instead, gonadal sex is determined by temperature during embryogenesis, with low and high incubation temperatures producing females and intermediate temperatures producing mixed sex ratios. In essence, this allows for the study of individual and sex differences without the confounding variable of genetically determined gender. Experiments have shown that the temperature experienced during incubation plays a critical role in establishing the adult morphological, endocrinological, and behavioral phenotype. In this experiment, the independent effects of incubation temperature and gonadal sex on the morphology and metabolic capacity of specific brain nuclei were determined. Both individual and sex differences in the volume of the preoptic area and ventromedial nucleus of the hypothalamus are determined primarily by incubation temperature, not by gonadal sex. However, incubation temperature and gonadal sex are both important in determining the metabolic capacity in the anterior hypothalamus, external amygdala, dorsal lateral nucleus of the hypothalamus, dorsal lateral nucleus of the thalamus, dorsal ventricular ridge, habenula, lateral hypothalamus, nucleus rotundus, nucleus sphericus, periventricular nucleus of the hypothalamus, preoptic area, periventricular nucleus of the preoptic area, septum, striatum, torus semicircularis, and ventromedial nucleus of the hypothalamus. This is the first demonstration in a vertebrate that factors other than gonadal sex hormones, which arise from the individual's genetic constitution, can affect the sexual differentiation of the brain.

In situ hybridization for two differentially expressed GnRH genes following estrogen and triiodothyronine treatment in the brains of juvenile tilapia (cichlid)

Using in situ hybridization, messenger RNAs for gonadotropin-releasing hormone (GnRH) were seen distributed differently in the brains of 40-day-old juvenile tilapia Oreochromis mossambicus. While salmon-GnRH mRNA was localized in the nucleus olfactoretinalis (terminal nerve ganglia), chicken II-GnRH mRNA mRNA was observed in the midbrain nucleus. Various concentrations (0.1-10 mg) of estradiol benzoate and triiodothyronine, given over a 24 h period, had no effects on mRNA levels of salmon- and chicken II-GnRH. Analysis of variance indicated significantly higher levels of salmon- but not chicken II-GnRH mRNA in larger (> 1.5 mm) versus smaller (1.3 mm) brains, among juveniles of the same age and same genetic brood. Thus, salmon-GnRH neurons in the nucleus olfactoretinalis display greater variance depending on the body mass. Since reproductively active tilapia differ with respect to body size at sexual maturation, therefore, besides the age and treatment effects, body size should be taken into consideration.

Gonadotropin-releasing hormone gene expression in teleosts

Expression of multiple molecular forms of gonadotropin-releasing hormone (GnRH) mRNAs and GnRH peptides were examined in the brains of tilapia (Oreochromis mossambicus) and sockeye salmon (Oncorhynchus nerka), using in situ hybridization histochemistry and immunohistochemical techniques. After otherwise identical conditions, lesser background and stronger GnRH hybridization signals were observed on cryostat vs. paraffin sections. In both fresh and Bouin's-fixed paraffin-embedded tissues, there was a good correlation between the distribution of GnRH mRNA and GnRH peptide-containing cells. Although the brains of tilapia and the sockeye were immunoreactive to three forms of the GnRH molecule (salmon, mammal, chicken-II), GnRH mRNA expression was site-specific and species-specific. In the tilapia, ganglionic cells of the nucleus olfactoretinalis, basal telencephalon and the anteroventral preoptic area were immunoreactive to salmon-, and mammalian-GnRH peptide. Neurons of the nucleus olfactoretinalis expressed cichlid-GnRH I mRNA. The preoptic neurons, despite the immunoreactivity, expressed no hybridization signals. Midbrain neurons were immunoreactive to salmon-GnRH but expressed cichlid-GnRH II beta (= chicken-GnRH II) mRNA hybridization signals. In the sockeye, ganglionic cells along the extracerebral course of the nervus terminalis were immunoreactive to mammalian-, chicken-II and salmon-GnRH. These neurons expressed only salmon-GnRH mRNA hybridization signals. Intracerebral GnRH expression in the sockeye was delayed till smoltification. The basal telencephalon and midbrain neurons immunoreactive to salmon-GnRH, formed no hybridization signals with GnRH antisense probes. Oligonucleotide probes complementary to chicken-GnRH I and mammalian-GnRH revealed no hybridization signals in the tilapia and in the sockeye brain. Fibers, immunoreactive to salmon-, mammalian-, and chicken II-GnRH were seen in close association with growth hormone cells. Chicken-GnRH II-immunoreactive fibers were also seen in close proximity to somatolactin cells in the sockeye salmon.

Nonsequential developmental trajectories lead to dimorphic vocal circuitry for males with alternative reproductive tactics

Midshipman fish, Porichthys notatus, have two male reproductive morphs: type 1 males generate long duration advertisement calls ("hums") to attract females to a nest; type II males sneak-spawn and, like females, do not produce mate calls but generate short duration agonistic calls. A vocal pacemaker circuit includes: motoneurons in the caudal brain stem and rostral spinal cord that innervate vocal/sonic muscles; pacemaker neurons that are located ventrolateral to motoneurons and establish their fundamental discharge frequency; and a ventral medullary nucleus that couples the motoneuron-pacemaker circuit bilaterally. Transneuronal biocytin transport identified morph-specific developmental trajectories for the vocal circuit. Among nonreproductive, juvenile type I males, motoneuron soma size and motor nucleus volume increase most during a stage prior to sexual maturation. An additional increase in motoneuron size and nucleus volume is coupled to the greatest increase in pacemaker soma size at a stage coincident with the onset of sexual maturity; ventral medullary neurons show similar growth increments during both stages. Type II males (and females) mature with no or little change in cell size or motor nucleus volume. The results indicate that alternative mating tactics are paralleled by alternative developmental trajectories for the neurons that determine tactic-specific behaviors, in this case vocalizations. Together with aging data based on otolith growth, the results support the hypothesis that alternative male morphs in midshipman fish adopt nonsequential, mutually exclusive life history tactics.