Variation in branchial expression among insulin-like growth-factor binding proteins (igfbps) during Atlantic salmon smoltification and seawater exposure.

BACKGROUND: In preparation for migration from freshwater to marine habitats, Atlantic salmon (Salmo salar L.) undergo smoltification, a transformation that includes the acquisition of hyposmoregulatory capacity. The growth hormone (Gh)/insulin-like growth-factor (Igf) axis promotes the development of branchial ionoregulatory functions that underlie ion secretion. Igfs interact with a suite of Igf binding proteins (Igfbps) that modulate hormone activity. In Atlantic salmon smolts, igfbp4,-5a,-5b1,-5b2,-6b1 and-6b2 transcripts are highly expressed in gill. We measured mRNA levels of branchial and hepatic igfbps during smoltification (March, April, and May), desmoltification (July) and following seawater (SW) exposure in March and May. We also characterized parallel changes in a broad suite of osmoregulatory (branchial Na+/K+-ATPase (Nka) activity, Na + /K + /2Cl - cotransporter 1 (nkcc1) and cystic fibrosis transmembrane regulator 1 (cftr1) transcription) and endocrine (plasma Gh and Igf1) parameters. RESULTS: Indicative of smoltification, we observed increased branchial Nka activity, nkcc1 and cftr1 transcription in May. Branchial igfbp6b1 and -6b2 expression increased coincidentally with smoltification. Following a SW challenge in March, igfbp6b1 showed increased expression while igfbp6b2 exhibited diminished expression. igfbp5a,-5b1 and-5b2 mRNA levels did not change during smolting, but each had lower levels following a SW exposure in March. CONCLUSIONS: Salmonids express an especially large suite of igfbps. Our data suggest that dynamic expression of particular igfbps accompanies smoltification and SW challenges; thus, transcriptional control of igfbps may provide a mechanism for the local modulation of Igf activity in salmon gill.

Hepatic insulin-like growth-factor binding protein (igfbp) responses to food restriction in Atlantic salmon smolts.

The growth hormone (Gh)/insulin-like growth-factor (Igf) system plays a central role in the regulation of growth in fishes. However, the roles of Igf binding proteins (Igfbps) in coordinating responses to food availability are unresolved, especially in anadromous fishes preparing for seaward migration. We assayed plasma Gh, Igf1, thyroid hormones and cortisol along with igfbp mRNA levels in fasted and fed Atlantic salmon (Salmo salar). Fish were fasted for 3 or 10days near the peak of smoltification (late April to early May). Fasting reduced plasma glucose by 3days and condition factor by 10days. Plasma Gh, cortisol, and thyroxine (T4) were not altered in response to fasting, whereas Igf1 and 3,5,3'-triiodo-l-thyronine (T3) were slightly higher and lower than controls, respectively. Hepatic igfbp1b1, -1b2, -2a, -2b1 and -2b2 mRNA levels were not responsive to fasting, but there were marked increases in igfbp1a1 following 3 and 10days of fasting. Fasting did not alter hepatic igf1 or igf2; however, muscle igf1 was diminished by 10days of fasting. There were no signs that fasting compromised branchial ionoregulatory functions, as indicated by unchanged Na(+)/K(+)-ATPase activity and ion pump/transporter mRNA levels. We conclude that dynamic hepatic igfbp1a1 and muscle igf1 expression participate in the modulation of Gh/Igf signaling in smolts undergoing catabolism.

In vivo effects of 17α-ethinylestradiol, 17ß-estradiol and 4-nonylphenol on insulin-like growth-factor binding proteins (igfbps) in Atlantic salmon.

Feminizing endocrine disrupting compounds (EDCs) affect the growth and development of teleost fishes. The major regulator of growth performance, the growth hormone (Gh)/insulin-like growth-factor (Igf) system, is sensitive to estrogenic compounds and mediates certain physiological and potentially behavioral consequences of EDC exposure. Igf binding proteins (Igfbps) are key modulators of Igf activity, but their alteration by EDCs has not been examined. We investigated two life-stages (fry and smolts) of Atlantic salmon (Salmo salar), and characterized how the Gh/Igf/Igfbp system responded to waterborne 17α-ethinylestradiol (EE2), 17ß-estradiol (E2) and 4-nonylphenol (NP). Fry exposed to EE2 and NP for 21 days had increased hepatic vitellogenin (vtg) mRNA levels while hepatic estrogen receptor α (erα), gh receptor (ghr), igf1 and igf2 mRNA levels were decreased. NP-exposed fry had reduced body mass and total length compared to controls. EE2 and NP reduced hepatic igfbp1b1, -2a, -2b1, -4, -5b2 and -6b1, and stimulated igfbp5a. In smolts, hepatic vtg mRNA levels were induced following 4-day exposures to all three EDCs, while erα only responded to EE2 and E2. EDC exposures did not affect body mass or fork length; however, EE2 diminished plasma Gh and Igf1 levels in parallel with reductions in hepatic ghr and igf1. In smolts, EE2 and E2 diminished hepatic igfbp1b1, -4 and -6b1, and stimulated igfbp5a. There were no signs of compromised ionoregulation in smolts, as indicated by unchanged branchial ion pump/transporter mRNA levels. We conclude that hepatic igfbps respond (directly and/or indirectly) to environmental estrogens during two key life-stages of Atlantic salmon, and thus may modulate the growth and development of exposed individuals.

Alternative splice variants of the rainbow trout leptin receptor encode multiple circulating leptin-binding proteins.

In mammals, leptin (Lep) binding proteins (LepBPs) derived from Lep receptor (LepR) gene or protein bind most of the circulating Lep, but to date, information on LepBPs in nonmammalian vertebrate classes is lacking. This study details the characterization of multiple LepBPs in rainbow trout (Oncorhynchus mykiss), an early poikilothermic vertebrate, and presents the complete coding sequences for 3 of them. Size-exclusion chromatography and cross-linking assay identified plasma proteins bound to Lep ranging from 70 to 100 kDa. LepBPs were isolated from plasma by affinity chromatography, and their binding specificity was assessed by a competitive binding assay. A RIA for LepBPs indicates that plasma LepBP levels decline after fasting for 3 weeks. Immunoblotting of LepBPs using antibodies against different LepR epitopes shows that the LepBPs are indeed LepR isoforms. The alternatively spliced LepR transcripts (LepR(S1-3)) that include only the extracellular segment transcribe the 90-kDa LepBP1, the 80-kDa LepBP2, and the 70-kDa LepBP3, respectively. LepR(S1) generally has lower expression than the long-form LepR in most tissues. LepR(S2) is primarily expressed in adipose tissue, whereas LepR(S3) is expressed abundantly in brain and spleen, and moderately in liver and gills. The mRNA levels of hepatic LepR(S3) increase after 2 weeks of fasting. This study demonstrates a mechanism in fish for the generation of LepBPs that differs from that seen in mammals and indicates that the physiologic action of Lep in these poikilothermic vertebrates can be modulated, both centrally and peripherally, by the differentiated, tissue-specific expression of multiple LepBPs.

Plasma ghrelin levels in rainbow trout in response to fasting, feeding and food composition, and effects of ghrelin on voluntary food intake.

Ghrelin, a peptide hormone which stimulates growth hormone (GH) release, appetite and adiposity in mammals, was recently identified in fish. In this study, the roles of ghrelin in regulating food intake and the growth hormone (GH)-insulin-like growth factor I (IGF-I) system of rainbow trout (Oncorhynchus mykiss) were investigated in three experiments: 1) Pre- and postprandial plasma levels of ghrelin were measured in relation to dietary composition and food intake through dietary inclusion of radio-dense lead-glass beads, 2) the effect of a single intraperitoneal (i.p.) injection with rainbow trout ghrelin on short-term voluntary food intake was examined and 3) the effect of one to three weeks fasting on circulating ghrelin levels and the correlation with plasma GH and IGF-I levels, growth and lipid content in the liver and muscle was studied. There was no postprandial change in plasma ghrelin levels. Fish fed a normal-protein/high-lipid (31.4%) diet tended to have higher plasma ghrelin levels than those fed a high-protein/low-lipid (14.1%) diet. Plasma ghrelin levels decreased during fasting and correlated positively with specific growth rates, condition factor, liver and muscle lipid content, and negatively with plasma GH and IGF-I levels. An i.p. ghrelin injection did not affect food intake during 12-hours post-injection. It is concluded that ghrelin release in rainbow trout may be influenced by long-term energy status, and possibly by diet composition. Further, in rainbow trout, ghrelin seems to be linked to growth and metabolism, but does not seem to stimulate short-term appetite through a peripheral action.

Circulating levels of cholecystokinin and gastrin-releasing peptide in rainbow trout fed different diets.

Cholecystokinin (CCK) and gastrin-releasing peptide (GRP) are gastrointestinal peptides thought to be important regulators of intake and digestion of food in vertebrates. In this study, pre- and postprandial plasma levels of CCK and GRP were measured in rainbow trout (Oncorhynchus mykiss) by the establishment of homologous radioimmunoassays, and the hormonal levels assessed in relation to dietary lipid:protein ratio and food intake. Fish were acclimated to either a high protein/low lipid diet (HP/LL diet; 14.1% lipids) or a normal protein/high lipid diet (NP/HL diet; 31.4% lipids). On three consecutive sampling days, radio-dense lead-glass beads were included in the diets for assessment of feed intake. Fish were terminally sampled for blood and stomach contents prior to feeding at time 0, and at 0.3, 1, 2, 4, 6, and 24 h after feeding. There was a postprandial elevation of plasma CCK levels, which was most evident after 4 and 6 h. Fish fed the NP/HL diet had higher plasma CCK levels compared with those fed the HP/LL diet. Plasma CCK levels were not affected by the amount of food ingested. GRP levels in plasma were not influenced by sampling time, diet, or feed intake. The results indicate that the endocrine release of gastrointestinal CCK is increased during feeding and may be further influenced by the dietary lipid:protein ratio in rainbow trout. Plasma GRP levels, on the other hand, appear not to be influenced by feeding or diet composition.

Purification, characterization, and biological activity of insulins from the spotted dogfish, Scyliorhinus canicula, and the hammerhead shark, Sphyrna lewini.

Insulin was purified from pancreatic extracts of two elasmobranch species belonging to different families in the order Carcharhiniformes, the European spotted dogfish, Scyliorhinus canicula (Scyliorhinidae), and the hammerhead shark, Sphyrna lewini (Carcharhinidae). The amino acid sequence of dogfish insulin was established as A-chain GIVDHCCRNT(10)CSLYDLEGYC(20)NQ and B-chain LPSQHLCGSH(10)LVETLYFVCG(20)QKGFYYVPKV(30). The primary structure of hammerhead shark insulin was similar to that of dogfish insulin with only 2 amino acid substitutions at A8 (R --> H) and B30 (V --> I). The elasmobranch insulins were markedly different from human insulin (17 amino acid substitutions) but all the residues in human insulin that are believed to be important in determining the receptor binding conformation (B6, B8, B11, B13, B23, B24, B25, A2, A3, and A19) have been conserved in the elasmobranch insulins with the exception of the conservative substitution Phe --> Tyr at B25. Consistent with this, dogfish and human insulin showed almost identical binding affinity to the recombinant solubilized human insulin receptor (K(D) values of 14.0 and 18.6 pM, respectively; relative potency 133%). Previous studies have shown that bovine insulin produces severe and sustained hypoglycemia in elasmobranchs but the effect is of slow onset. Bolus arterial injections of dogfish insulin (10 nmol x kg(-1)) into unanesthetized, fasting dogfish (n = 9) produced no changes in blood glucose, 3-hydroxybutyrate, and acetoacetate concentrations over a 4-h period. In a second series of experiments (n = 7), dogfish insulin (10 nmol x kg(-1)) produced a significant (P < 0.05) fall in blood glucose after 12 h that persisted for at least 48 h, but no change in ketone body concentrations. The data indicate that the metabolic actions of an endogenous elasmobranch insulin in an elasmobranch are similar to those previously described for mammalian insulin.