Ontogeny of corticotropin-releasing factor and of hypothalamic-pituitary-interrenal axis responsiveness to stress in tilapia (Oreochromis mossambicus; Teleostei).

The ontogeny of the corticotropin-releasing factor (CRF) system and of the ability of the hypothalamic-pituitary-interrenal (HPI) axis to respond to stressors (capture or confinement), or to cortisol treatment was investigated in tilapia (Oreochromis mossambicus). In 2 days post hatching (dph) larvae, the first developmental stage used for immunohistochemistry, CRF-immunoreactivity (ir) was observed in the nucleus preopticus (npo), and in two hypothalamic nuclei (nlt and nrl). In this stage, CRF- and AVT-ir was found in the neural part of the pituitary, and endocrine cells in the pars distalis and pars intermedia contained POMC-derived peptides. In the ventral telencephalon, CRF-ir cells were first observed 5 dph, whereas projections from these cells into the anterior part of the latero-dorsal telencephalon (Dla) from 7 dph onwards. CRF, ACTH, alpha-MSH, and cortisol were quantified by radioimmunoassays in homogenates of the anterior-cranial region of the larvae containing brain, pituitary, and headkidneys. CRF contents increased from 43 +/- 3 to 1070 +/- 70 pg/larvae between 5 and 110 dph. Larvae of age 5, 12, 24, and 42 dph were captured sequentially from a group. All life stages were able to rapidly increase their cortisol content in response to this stressor (ANOVA: P < 0.001). Overall, the developmental stage affected cortisol content (ANOVA: P < 0.001), but developmental stage did not influence the cortisol reaction to stress (ANOVA: P > 0.162). Whole brain CRF content did not change during the 20 min stress period and the relationship between CRF-producing neurons and the initial HPI stress response in early life stages remains to be established. Cortisol feeding of 18 and 29 dph larvae for periods ranging from 2 to 24 days resulted in elevations of the CRF content (P < 0.003) in comparison to controls. In 18 dph larvae cortisol feeding abolished the cortisol response to capture stress as observed in control fed larvae (P < 0.008). We propose that cortisol induced upregulation of CRF takes place in the telencephalon and is restricted to a time period during larval development, characterised by the absence of glucocortoid receptor (GR) expression in the telencephalic Dm region in these larvae. Finally, the stress response to 24 h confinement was compared between saltwater adapted and freshwater adapted juveniles (age 77 dph). Confinement stress (24 h) affected cortisol and CRF content (ANOVA: P < 0.001, P < 0.008, respectively), but not ACTH content. Interactions were observed between salinity and confinement regarding cortisol and alpha-MSH contents (ANOVA: P < 0.02), but not regarding CRF and ACTH contents. The increase in cortisol levels induced by confinement was remarkably high in freshwater adapted larvae (five times higher than in saltwater adapted larvae). Regarding the cortisol response it is concluded that during and after the period of mouth breeding tilapia larvae respond to capture stress in a similar fashion (onset and height) as adults. Previously, we reported that the initial plasma cortisol response to capture stress in adult tilapia occurred independently from changes in plasma ACTH levels. The current finding that also brain CRF contents do not alter during the initial cortisol response in larvae further indicates that the initial cortisol response in this species may be regulated independently from CRF and ACTH.

Distribution and quantification of corticotropin-releasing hormone (CRH) in the brain of the teleost fish Oreochromis mossambicus (tilapia).

The recent characterization of the corticotropin-releasing hormone (CRH) prehormone of the fish tilapia (Oreochromis mossambicus) showed that more variation exists between vertebrate CRH amino acid sequences than recognized before. The present study investigates whether the deviating composition of tilapia CRH coincides with an atypical distribution of CRH in the brain. For this purpose we applied immunohistochemistry, as well as radioimmunoassay (RIA) quantification in brain slices. The results are plotted in a new atlas and reconstruction of the tilapia brain. The largest population of CRH-immunoreactive (ir) neurons is present in the lateral part of the ventral telencephalon (Vl). Approximately tenfold less CRH-ir neurons are observed in the preoptic and tuberal region. The CRH-ir neurons observed in the preoptic region are parvocellular and do not, or hardly, display arginine-vasotocin (AVT) immunoreactivity. CRH-ir neurons are also present in the glomerular layer of the olfactory bulb, in the periventricular layer of the optic tectum, and caudal to the glomerular nucleus. A very dense plexus of CRH-ir terminals is located in the most rostral part of the dorsal telencephalon. This region has not been described in other teleosts and is in the present study subdivided into the anterior part of the dorsal telencephalon (Da) and the anterior part of the laterodorsal telencephalon (Dla). High densities of CRH-ir terminals were observed in and around Vl, in the tuberal region, around the rostral part of the lateral recess, and in the caudal part of the vagal lobe. In the pituitary, CRH-ir terminals are concentrated in the neuro-intermediate lobe. Overall, the immunohistochemical and quantitative data correlated well, as the RIA CRH profile in serial 160-microm slices revealed four peaks, which corresponded with major ir-cell groups and terminal fields. Our results strongly suggest that the CRH-ir cells of Vl project to the rostro-dorsal telencephalon. Consequently, they may not be primarily involved in regulation of pituitary cell types but may subserve other functions. The presence of a CRH-containing Vl-Da/Dla projection seems to be restricted to the most modern group of teleosts, i.e., the Acanthopterygians. Further anatomic indications for non-pituitary-related functions of CRH are found in the vagal lobe and the optic tectum of tilapia. Although the low CRH content of the preoptic region reported here for tilapia may be typical for unstressed fish, the fact remains that remarkably few CRH-ir neurons are involved in regulating the pituitary. Overall, the CRH distribution in the brain of tilapia is more widespread than previously reported for other teleosts.