The expression of the prolactin receptor (PRL-R) gene has been investigated in various tissues of tilapia (Oreochromis niloticus) reared in fresh or brackish water. Using a cDNA probe spanning the extracellular domain of the tilapia PRL-R and Northern blot analysis, the presence of tilapia PRL-R mRNA has been confirmed in the osmoregulatory organs and has been detected in other tissues, including the skin, the brain, the reproductive organs, and the two major hematopoietic organs (spleen and head kidney), as well as circulating lymphocytes. These findings suggest a conservation of the physiological processes regulated by prolactin throughout the vertebrates, including immunity and central nervous activity. A non-radioactive in situ hybridization procedure has allowed us to detect the expression of the tilapia PRL-R in the branchial chloride cells and the intestinal mucosal layer of fresh water animals, confirming the direct control exerted by prolactin on the water and ionic exchanges in tilapia. In all the tissues examined one unique PRL-R transcript has been detected with a similar size (3·2 kb) whatever the salinity conditions. Thus, the transcriptional expression of the tilapia PRL-R strongly differs from the complex RNA pattern reported for the higher vertebrates PRL-R and provides an additional argument for the existence of a single PRL-R for both prolactin isoforms in this fish species.

In order to identify the specificity and functionality of salmon prolactin (sPRL) promoter, transgenic rainbow trout carrying a construct comprising the 2.4 kb fragment of the 5' flanking region of Atlantic Chinook sPRL gene fused either to the reporter genes cat (sPRL-cat) or lacZ (sPRL-lacZ) were produced. sPRL-cat in transgenic F0 fish expressed strongly CAT only in the pituitary gland. Transgenic in F1-F4 lines harbouring sPRL-lacZ expressed beta-galactosidase (beta-gal) only in the follicular PRL-producing cells of the adenohypophysis. We observed heterocellular, mosaic distribution of beta-gal within PRL cell population and enormous variation of lacZ expression level between the littermates in the same transgenic line. Regardless of the transgene copy number, age or sex of transgenic fish, beta-gal expression was lactotroph-specific but variegated in all the nine F2 hemizygous lines analysed. One line harbouring a multicopy integration was followed up to F4 generation: the transgene was transmitted without modifications. Analysis of genomic DNA from pituitaries showed that lacZ sequences were highly methylated. LacZ expression was low and its transcripts, analysed by in situ hybridisation, showed a mosaic distribution within the pituitary gland. These data suggest that variegated expression of lacZ can occur at the transcription level owing to the silencing effect of lacZ gene. After proving the tissue-specific expression of reporter genes driven by the sPRL promoter, we tried to obtain the genetic ablation of PRL-producing cells,by transferring the same construct comprising diphtheria toxin DT-A gene (tox). However, the high mortality rate of sPRL-tox transformed embryos has embedded this study and no transgenic fish expressing tox were produced. The appropriateness of using transgenic strategies to analyse gene function in Salmonids is discussed, especially the implications of the multicopy integration patterns and of the variegated transgene expression.