Concentration-response relationships and temporal patterns in hepatic gene expression of Chinook salmon (Oncorhynchus tshawytscha) exposed to sewage.

Changes in liver gene expression were examined in juvenile Chinook salmon (Oncorhynchus tshawytscha) exposed in vivo for 8d to seawater (control) or one of 5 concentrations of sewage (environmentally-relevant dilutions of 0.05%, 0.1%, and 0.7%; 2%, 5% or 10%) and subsequently transferred to clean seawater for an 8-d recovery period. Livers were sampled on days 1, 4, 8 (sewage-exposed) and 16 (8d of sewage exposure plus 8d of recovery). A custom cDNA microarray using a universal DNA reference design was used to examine trends of altered gene expression across sewage concentrations, across timepoints, and at the end of the recovery period. Alterations in gene expression followed four distinct concentration-dependent patterns: (1) concentration response (e.g. estrogen receptor alpha), (2) inverse-concentration response (e.g. insulin receptor beta), U-shaped (e.g. mineralocorticoid receptor), (3) inverse U-shaped (e.g. benzodiazepine receptor), and (4) concentration-independent responses (e.g. ubiquitin). Temporal trends included: (1) peak gene expression at one of the sewage exposure timepoints with recovery to baseline levels after the depuration phase (e.g. vitelline envelope protein beta), (2) gene expression alterations that did not recover (e.g. glucose transporter 3), and (3) delayed gene expression alterations initiated only at the recovery timepoint (e.g. insulin-like growth factor 2). In summary, patterns in gene expression changes were found across sewage concentrations and exposure timepoints. This study is the first to show gene expression trends of this nature.

Growth hormone transgenesis influences carbohydrate, lipid and protein metabolism capacity for energy production in coho salmon (Oncorhynchus kisutch).

Growth hormone (GH) transgenesis results in increased growth, feed intake and consequent metabolic rates in fish, and alters the utilization of dietary and stored carbohydrates, lipid and protein. However, the manner in which GH transgenesis differentially alters these energy sources in fish has not been well explored. We examined the effects of GH transgenesis and dietary carbohydrate, lipid and protein levels on metabolic enzyme activity in coho salmon (Oncorhynchus kisutch). In white muscle, increased activities of glycolytic enzymes and decreased activities of lipolytic enzymes in transgenic fish indicate a sparing of lipids through the preferential use of carbohydrates for energy production. In liver, transgenic fish showed increased activity of lipid synthesis enzymes and a shift in amino acid metabolism from catabolic to synthetic roles, suggesting a larger emphasis on anabolic pathways in transgenic fish to support accelerated growth. Unlike nontransgenic fish, transgenic fish fed a diet high in carbohydrates maintained growth rates, had increased capacity for lipid synthesis, and increased potential for biosynthetic roles of amino acids. GH transgenesis influences metabolic reactions in coho salmon by emphasizing carbohydrate degradation for energy production and lipid synthesis, and increasing utilization of lipids and proteins for synthetic roles necessary to maintain accelerated growth.

Cell volume affects glycogen phosphorylase activity in fish hepatocytes.

The activity of glycogen phosphorylase (GPase) in the active a-form (GPase a) is dependent on the hydration state of hepatocytes. We establish that GPase a catalysis in catfish (Ameiurus nebulosus) hepatocytes is a function of medium osmolarity and that a linear relationship exists between GPase a activity and osmolarity between 254 mosmol l(-1) and 478 mosmol l(-1). Exposure of isolated hepatocytes to hyperosmotic media increases enzyme activity up to 7-fold, indicative of covalent phosphorylation. GPase activation associated with cell shrinkage peaks within 10 min of exposure. The average degree of activation (2.7-fold-increase of GPase a) is only slightly less than in hepatocytes exposed to glucagon (3.1-fold-increase) under isosmotic conditions; with glucagon, the maximum is reached within 2 min. Phosphorylation status remains elevated during the entire 40 min experimental period; cells do not undergo regulatory volume increase (RVI) during this period and do not regain pre-exposure volume. We interpret the increased GPase a activity as an inherent response to hyperosmotic stress, likely brought about by molecular crowding. Activation of the enzyme results in increased glucose production from endogenous glycogen. Glucose is not retained in the liver cells, but may act as an oxidative substrate in extrahepatic tissues for the increased metabolic demand of ion regulation. Protein kinase A or intracellular Ca(2+) make apparently small contributions to the activation of GPase, leaving us to speculate on alternate routes of enzyme activation. Conversely, hepatocyte swelling in hyposmotic medium leads to significant decreases in GPase a activity and curtailed glucose output. A minimum is attained in 10 min, and pre-insult rates are re-established within 40 min, somewhat lagging behind readjustment in cell volume by regulatory volume decrease (RVD). We conclude that cell swelling and subsequent RVD do not signify stress to the cells and metabolic demand may be decreased under cell swelling conditions. Alteration of GPase phosphorylation with extracellular osmolarity appears to be a general phenomenon, since we also find it in hepatocytes of another freshwater catfish (Clarias batrachus) and a marine scorpaenid (Sebastes caurinus).

Metabolic zonation in teleost gastrointestinal tract. Effects of fasting and cortisol in tilapia.

Activities of several metabolic enzymes show distinct patterns of zonation along the intestinal tract of tilapia (Oreochromis niloticus), rainbow trout (Oncorhynchus mykiss) and copper rockfish (Sebastes caurinus). Zonation is species and enzyme specific, with different metabolic activities concentrated in specific areas, and few generalizations can be made. The rockfish show the smallest degree of zonation, with highest activities in the third quarter of the intestine, and shallow gradients to either side, and a general upswing in activity towards the distal end. In the trout, mitochondrial enzyme activities (citrate synthase, glutamate dehydrogenase, malate dehydrogenase) are highest in the pyloric caeca and decrease along the length of the small intestine. This pattern is accentuated for malic enzyme and glucose 6-phosphate dehydrogenase. These enzymes drop precipitously in activity after the first few sections of the small intestine, while other NADP-linked dehydrogenases (isocitrate dehydrogenase, and 6-phosphogluconate dehydrogenase) show moderate activity in pyloric caeca and peak toward the distal section of the small intestine. In tilapia, glutamate dehydrogenase shows a similar decrease as in trout, but citrate synthase peaks towards the distal sections. NADP-dependent dehydrogenases reveal distinct patterns, peaking in different sections of the intestine-malic enzyme in the proximal midsection, glucose 6-phosphate dehydrogenase in the distal mid-section, and isocitrate dehydrogenase in the anal section. Enzyme activities in the stomach of trout and tilapia also show zonation, with the midsection generally displaying the highest activities. A 5-day treatment of tilapia with an intraperitoneal cortisol deposit (25 mg kg(-1) wet mass) drastically alters metabolic performance along the gut in enzyme specific patterns, generally increasing enzyme activities in site-specific arrangements. Cortisol treatment also leads to the expected increases in activities of phosphoenolpyruvate carboxykinase, pyruvate kinase and aspartate aminotransferase in liver, but not in kidney. Aspartate aminotransferase is the only enzyme in brain significantly increased by cortisol treatment. Short-term food deprivation changes enzyme patterns, often resembling those observed after cortisol administration. We conclude that brain, liver and intestinal amino acid metabolism is an important target for cortisol action in fish and that metabolic zonation is a key factor to be reckoned with when analyzing physiological phenomena in the fish intestine.

Glutamine synthetase in tilapia gastrointestinal tract: zonation, cDNA and induction by cortisol.

Glutamine synthetase, an enzyme generally associated with ammonia detoxication in the vertebrate brain and with hepatic nitrogen turnover in mammals, shows substantial activities in the gastrointestinal tract of teleostean fishes. Enzyme activity is highest in the central area of the stomach and reveals a distinct distribution pattern in stomach and along the intestine of tilapia (Oreochromis niloticus), rainbow trout (Oncorhynchus mykiss) and copper rockfish (Sebastes caurinus). In all three species, intestinal activity peaks in the distal region of the intestine. The brain contains the highest titre of the enzyme (46 U g(-1) in tilapia brain versus 15 U g(-1) in tilapia stomach), but because of the relative mass of the stomach, the largest glutamine synthetase pool in tilapia body appears to be localized in the stomach. Activities in white and red muscle are very modest at 0.1% of the brain. Independent of distribution, peak activities of glutamine synthetase in selected areas of tilapia stomach and intestine are significantly (two- to fourfold) increased after a 5-day treatment with an intraperitoneal cortisol deposit. Cortisol also increases glutamine synthetase activity in tilapia liver, white and red muscle, while activities in brain remain unaffected. We cloned and sequenced the predominant transcript of tilapia stomach glutamine synthetase (about 1.9 kb), encoding a 371-amino acid peptide. The open reading frame shows considerable identity with glutamine synthetase in toadfish (92% at peptide level, 87% at nucleotide level), but possesses a longer 3'-untranslated region than the toadfish. The tilapia glutamine synthetase mRNA contains a remnant of a putative mitochondrial leader sequence, but without a conserved second site for initiation of translation. We also find evidence for additional transcripts of glutamine synthetase in tilapia, suggesting multiple genes. Finally, we present evidence for similar abundance of glutamine synthetase transcripts in all regions of rockfish intestine. The physiological significance of the presence of glutamine synthetase in teleostean intestine is discussed.

Novel role for prostaglandin E2 in fish hepatocytes: regulation of glucose metabolism.

Prostaglandin E(2) (PGE(2)) potently activated glycogenolysis and gluconeogenesis in isolated rockfish (Sebastes caurinus) hepatocytes. The average degree of activation for glycogenolysis was 6.4+/-0.67-fold (mean+/-S.E.M.; n=37), and could be as much as 19-fold. Analysis of dose-concentration relationships between glycogenolytic actions and PGE(2) concentrations yielded an EC(50) around 120 nM in hepatocyte suspensions and 2 nM for hepatocytes immobilized on perifusion columns. For the activation of gluconeogenesis (1.74+/-0.14-fold; n=10), the EC(50) for suspensions was 60 nM. Intracellular targets for PGE(2) actions are adenylyl cyclase, protein kinase A and glycogen phosphorylase. Concentrations of cAMP increased with increasing concentrations of PGE(2), and peaked within 2 min of hormone application. In the presence of the phosphodiesterase inhibitor, isobutyl-3-methylxanthine, peak height was increased and peak duration extended. The protein kinase A inhibitor, Rp-cAMPS, counteracted the activation of glycogenolysis by PGE(2), implying that the adenylyl cyclase/protein kinase A pathway is the most important, if not exclusive, route of message transduction. PGE(2) activated plasma membrane adenylyl cyclase and hepatocyte glycogen phosphorylase in a dose-dependent manner. The effects were specific for PGE(2); smaller degrees of activation of glycogenolysis were noted for PGE(1), 11-deoxy PGE(1), 19-R-hydroxy-PGE(2), and prostaglandins of the A, B and Falpha-series. The selective EP(2)-receptor agonist, butaprost, was as effective as PGE(2), suggesting that rockfish liver contains prostaglandin receptors pharmacologically related to the EP(2) receptors of non-hepatic tissues of mammals. Rockfish hepatocytes quickly degraded added PGE(2) (t((1/2))=17-26 min). A similar ability to degrade PGE(2) has been noted in catfish (Ameiurus nebulosus) hepatocytes, but no glycogenolytic or gluconeogenic actions of the hormone are noted for this species. We conclude that PGE(2) is an important metabolic hormone in fish liver, with cAMP-mediated actions on glycogen and glucose metabolism, and probably other pathways regulated by cAMP and protein kinase A. The constant presence of EP(2)-like receptors is a unique feature of the fish liver, with interesting implications for function and evolution of prostaglandin receptors in vertebrates.

Estradiol impairs hyposmoregulatory capacity in the euryhaline tilapia, Oreochromis mossambicus.

Freshwater (FW)-adapted tilapia (Oreochromis mossambicus) were treated with estradiol (E(2)) for 4 days to stimulate protein synthesis and sampled at 0, 4, and 24 h after exposure to 50% seawater (SW). E(2) increased circulating vitellogenin (VTG) levels in large amounts, indicative of unusually high rates of hepatic protein synthesis. E(2) treatment prevented the recovery of plasma osmolality in 50% SW that was evident in the sham group. Plasma sodium concentration was significantly elevated with E(2) in FW, but the levels did not change in 50% SW. Gill Na(+)-K(+)-ATPase activity was significantly lower in the E(2) group compared with sham-injected tilapia in 50% SW. No significant differences were noted in plasma cortisol, thyroxine, triiodothyronine, or glucose concentration with E(2) in 50% SW. E(2) significantly lowered several key liver enzyme activities and also decreased gill lactate dehydrogenase and malate dehydrogenase activities over a 24-h period. Together, our results suggest that E(2) impairs ion regulation in tilapia, partially mediated by a decreased metabolic capacity in liver and gill. The decreased tissue metabolic capacity is likely due to E(2)-induced energy repartitioning processes that are geared toward VTG synthesis at the expense of other energy-demanding pathways.

Effects of arginine on pancreatic hormones and hepatic metabolism in rainbow trout.

Arginine (Arg), injected intraperitoneally into rainbow trout (Oncorhynchus mykiss), increases plasma concentrations of glucagon, glucagon-like peptide-1 (GLP-1), and insulin by three- to 10-fold. Resulting ratios of glucagon and GLP-1 over insulin are unchanged in 20-d food-deprived fish (saline, 1.28 vs. Arg, 0.93; not significant) while slightly increased in feeding trout (saline, 0.70 vs. Arg, 0.92; P<0.05). In food-deprived juveniles, Arg injection leads to significant decreases in plasma fatty acids (saline, 1.65 mM L(-1) vs. Arg, 1.09 mM L(-1); P<0.05) and increases in glycogen phosphorylase total activity (saline, 3.7 units g(-1) vs. Arg, 4.6 units g(-1); P<0.05) and degree of phosphorylation (saline, 1.7 units g(-1) vs. Arg, 2.33 units g(-1); P<0.05). Plasma and liver glucose and liver enzymes (glucose-6-phosphate dehydrogenase, 6-phosphogluconate dehydrogenase, pyruvate kinase, phosphoenolpyruvate carboxykinase, lactate dehydrogenase, and malic enzyme) are unaffected. Otherwise, fish show the changes in plasma metabolites expected with food deprivation. Arg injection into feeding fish results in decreases in plasma fatty acids, liver glycogen, and glucose, while liver glucose 6-phosphate concentrations increase. Hepatocytes isolated from feeding fish injected with Arg 2 h previously show significantly lower rates of lactate oxidation than controls (85% of control), while rates of gluconeogenesis and hormonal responses to mammalian glucagon and GLP-1 remain unchanged. Rates of lactate oxidation and gluconeogenesis are significantly decreased by 5%-10% on treatment with porcine insulin. Complete immunoneutralization of insulin with rabbit antisalmon insulin serum decreases hepatic glucose 6-phosphate concentrations and abolishes the Arg-dependent effects on glycogen phosphorylase. It appears that short-term increases in pancreatic hormones cause only minor metabolic readjustments in the relatively short time frame covered in these experiments. Surprisingly, complete removal of insulin does not have immediate altering or detrimental effects on key metabolites and metabolic pathways, even if glucagon and GLP-1 concentrations are concurrently several-fold higher than usual. Our data clearly show the dual role of Arg in fish metabolism.

Amphibian glucagon family peptides: potent metabolic regulators in fish hepatocytes.

Peptides analogous to glucagon-like peptide-1 (GLP-1) have been isolated from amphibian pancreas and intestine, and their amino acid sequences and cDNA structures elucidated. Just like their mammalian counterpart, these peptides are potent insulinotropins in mammalian pancreatic cells. We show here that these peptides also exert strong glycogenolytic actions when applied to dispersed fish hepatocytes. We compared the potencies of three synthetic GLP-1s from Xenopus laevis and two native GLP-1s from Bufo marinus in the activation of glycogenolysis in the hepatocytes of a marine rockfish (Sebastes caurinus) and two freshwater catfish (Ameiurus nebulosus and A. melas), and demonstrated their effectiveness in increasing the degree of phosphorylation of glycogen phosphorylase. We also compared the glycogenolytic potency of the peptides with those of human GLP-1 and glucagons from human and B. marinus. Sensitivity to these peptides is species-specific, with the rockfish responding at lower concentrations to GLP-1s and the two catfish reacting better to glucagons. However, the relative potency of the amphibian GLP-1s and glucagons is similar in the three species. Xenopus GLP-1C (xGLP-1C) is consistently more potent than xGLP-1B, while xGLP-1A displays the smallest activation of glycogenolysis. Similarly, Bufo GLP-1(32)-the peptide with the highest amino acid sequence identity to xGLP-1C-always shows a higher potency than Bufo GLP-1(37), which is closely related to xGLP-1B. The relative hierarchy of these glycogenolytic GLP-1s differs from their ranking as insulinotropins in mammalian beta-cells. In the rockfish system, Bufo glucagon-36, a C-terminally extended glucagon, is more potent than the shorter bovine glucagon and Bufo glucagon-29 in the activation of glycogenolysis; when tested in A. nebulosus hepatocytes, bovine and amphibian glucagons are equipotent. Amphibian GLP-1s and glucagons activate glycogenolysis in fish hepatocytes through increased phosphorylation of glycogen phosphorylase, implying involvement of the adenylyl cyclase/protein kinase A system in signal transduction. We conclude that the broad physiological effectiveness of GLP-1 has been retained throughout vertebrate evolution, and that both insulinotropic activity and glycogenolytic actions belong to the repertoire of GLP-1.

Paradigms of growth in fish.

Most fish are indeterminate growers with white muscle making up the majority of the acquired bulk. Within the muscle, the myofibrillar fraction accounts for almost two-thirds of the protein synthetic activity, implying that it is accretion of myofibrillar proteins that makes the single most important contribution to fish growth. Fish muscle growth itself is not linear and occurs through a combination of hyperplasia and hypertrophy in post-juvenile stages. Superimposed on periodicity of growth in length and mass can be other phases governed by lunar, reproductive or circannual cycles. Data on fish growth are discussed in the framework of site-specific muscle abundance, metabolic and functional zonation of muscle, proliferation and differentiation of satellite cells and the contribution of myofibrillar proteins. Hormonal control of muscle growth is described against the backdrop of plasma availability of myogens (insulin, IGF-I, growth hormone), distribution and dynamics of their respective receptors, and their interactions. Important contributions of the 'supply side' are discussed with hormones regulating amino acid resorption from the intestine, intestinal growth, liver processing and amino acid uptake by the muscle. Data are also interpreted from metabolic angles, to explain lipolytic and nitrogen-sparing effects of growth hormones, and lipogenic effects of insulin and high protein diets. Finally, special attention is devoted to the multifaceted roles of arginine in fish growth, as precursor, intermediate and hormone secretagogue.

Glucagon-like peptide-1 activates the adenylyl cyclase system in rockfish enterocytes and brain membranes.

Glucagon-like peptide (GLP) exerts important physiological functions in fish liver, but extrahepatic sites of action and physiological roles have been largely ignored. We show here that GLP activates adenylyl cyclase in isolated brain and enterocyte membranes and increases cellular cyclic adenosine monophosphate (cAMP) levels in isolated enterocytes of rockfish (Sebastes caurinus). Following exposure to synthetic zebrafish GLP (zf-GLP) (1 nM-1 microM), a concentration-dependent increase in enterocyte cAMP is noted. The maximum increase in cAMP levels is observed at 1 microM zf-GLP, and represents a 30% increase above control values. Exendin-4, a GLP receptor agonist in mammals, elicits a similar concentration-dependent increase in enterocyte cAMP. In contrast, norepinephrine or prostaglandin E2 (at 1 microM) increased cAMP levels by 2 and 4-fold, respectively. Brain membrane adenylyl cyclase is activated 20-40% by zf-GLP, and to a smaller extent by zf-glucagon, while exendin-4 is as effective as zf-GLP at a dose of 100 nM. These results suggest potential physiological roles of GLP in brain and intestine in piscine systems analogous to GLP-1 functions in these tissues described for mammals.

Molecular characterization of glycosomal NAD(+)-dependent glycerol 3-phosphate dehydrogenase from Trypanosoma brucei rhodesiense.

The primary structure of a 38-kDa protein isolated from membrane preparations of African trypanosomes was determined by protein and DNA sequencing. Searching of the protein database with the trypanosome translated amino acid sequence identified glycerol 3-phosphate dehydrogenase (EC 1.1.1.8) from various prokaryotic and eukaryotic organisms as the optimal scoring protein. Surprisingly, the eukaryotic trypanosome enzyme showed the highest degree of sequence identity with the corresponding enzyme from the prokaryote Escherichia coli. The trypanosome molecule was expressed in Escherichia coli and found to be enzymatically active, thus confirming the identity of the molecule as an NAD(+)-dependent glycerol 3-phosphate dehydrogenase. A monoclonal antibody specific for the 38-kDa protein was used to localize the enzyme to glycosomes. Immunoblotting showed that the monoclonal antibody bound to a 38-kDa protein in African trypanosomes but not in T. cruzi, Leishmania or Crithidia. The enzyme has a pI of 9.1, a net charge of +17 and contains the peroxisome-like targeting tripeptide SKM at its C-terminus, all characteristic of glycosomal enzymes. Amino acids predicted to be involved in the NAD(+)-dependent glycerol 3-phosphate dehydrogenase active site have diverged from those of the mammalian enzyme. Kinetic analyses of the trypanosome GPD and GPD from rabbit muscle showed that the Km values of the two enzymes are different. The data suggest that the trypanosome protein may be a candidate target for rational drug design.

Glucagon and glucagon-like peptides in fishes.

Glucagon and glucagon-like peptides (GLPs) are coencoded in the vertebrate proglucagon gene. Large differences exist between fishes and other vertebrates in gene structure, peptide expression, peptide chemistry, and function of the hormones produced. Here we review selected aspects of glucagon and glucagon-like peptides in vertebrates with special focus on the contributions made by analysis of piscine systems. Our topics range from the history of discovery to gene structure and expression, through primary structures and regulation of plasma concentrations to physiological effects and message transduction. In fishes, the pancreas synthesizes glucagon and GLP-1, while the intestine may contribute oxyntomodulin, glucagon, GLP-1, and GLP-2. The pancreatic gene is short and lacks the sequence for GLP-2. GLP-1, which is produced exclusively in its biologically active form, is a potent metabolic hormone involved in regulation of liver glycogenolysis and gluconeogenesis. The responsiveness of isolated hepatocytes to glucagon is limited to high concentrations, while physiological concentrations of GLP-1 effectively regulate hepatic metabolism. Plasma concentrations of GLP-1 are higher than those of glucagon, and liver is identified as the major site of removal of both hormones from fish plasma. Ultimately, GLP-1 and glucagon exert effects on glucose metabolism that directly and indirectly oppose several key actions of insulin. Both glucagon and GLP-1 show very weak insulinotropic activity, if any, when tested on fish pancreas. Intracellular message transduction for glucagon, especially at slightly supraphysiological concentrations, involves cAMP and protein kinase A, while pathways for GLP are largely unknown and may involve a multitude of messengers, including cAMP. In spite of fundamental differences in GLP-1 function between fishes and mammals, fish GLP-1 is as powerful an insulinotropin for mammalian B-cells as mammalian GLP-1 is a metabolic hormone if tested on piscine liver.

Divergence of insulin-like growth factors I and II in the elasmobranch, Squalus acanthias.

Recent studies have shown that vertebrates, including teleostean fishes, amphibians, birds and mammals, contain two distinct insulin-like growth factor (IGF) genes. In contrast agnathans, represented by hagfish, apparently have only one IGF that has features characteristic of both IGF-I and IGF-II. Between these groups the elasmobranchs occupy a critical position in terms of the phylogeny of IGFs. We sought to determine if gene duplication and divergence of IGF-I and IGF-II occurred before or after divergence of elasmobranchs from other vertebrates by cloning IGF-like molecules from Squalus acanthias. Our analysis shows that Squalus liver produces two distinct IGF-like molecules. One has greater sequence identity to, and conserved features characteristic of, known IGF-I molecules, while the other is more IGF-II like. These results suggest that the prototypical IGF molecule duplicated and diverged in an ancestor of the extant gnathostomes.

Characterization of insulins and proglucagon-derived peptides from a phylogenetically ancient fish, the paddlefish (Polyodon spathula).

The North American paddlefish, Polyodon spathula (Order Acipenseriformes) is an extant representative of a group of primitive Actinopterygian (ray-finned) fish that probably shared a common ancestor with present-day teleosts. Two molecular forms of insulin which differ by a single amino acid substitution, His or Asp at position 15 of the A chain, were isolated from the pancreas of the paddlefish. Paddlefish insulins show greatest structural similarity to insulin from the garfish (order Lepisosteiformes) and resemble mammalian insulins more strongly than do insulins from teleost fish. The primary structures of several proglucagon-derived peptides, two molecular forms of glucagon which differ by the single amino acid substitution Arg18-->Ser, and glucagon-like peptide, have been less well conserved during evolution. The paddlefish glucagons contain 31 amino acid residues, rather than the usual 29, and show several structural features, such as Met5, Glu24 and Gly29, not previously observed in glucagons from other species. In spite of considerable differences in structure between paddlefish and mammalian glucagons (10 or 11 amino acid substitutions), both paddlefish glucagons are equally as effective as bovine glucagon in stimulating glycogenolysis in dispersed hepatocytes from the teleost fish Sebastes caurinus (rockfish). However, the substitution Arg18-->Ser in the paddlefish glucagon results in a 6-fold decrease in potency in this system.

Effects of temperature and freezing on hepatocytes isolated from a freeze-tolerant frog.

Metabolically active hepatocytes prepared from freeze-tolerant wood frogs, Rana sylvatica, were used to examine the direct effects of temperature and freezing on cryoprotectant synthesis and to assess the effectiveness of the natural cryoprotectant glucose in the freezing preservation of the isolated cells. Freshly isolated hepatocytes showed slow leakage of lactate dehydrogenase, readily synthesized urea, and oxidized a variety of 14C-labeled substrates. Effects of temperature on glucose production by isolated hepatocytes showed a normal Arrhenius relationship. However, compared with 0 degrees C control cells, either incubation at higher temperatures or freezing at -3 degrees C reduced the activity of glycogen phosphorylase alpha. These data suggest that the freezing-induced cryoprotectant production that occurs in vivo is not due to direct action of either low temperature or freezing on liver cell metabolism. The natural cryoprotectant glucose was also an excellent cryoprotectant of hepatocytes in vitro. In the absence of glucose, freezing caused a substantial leakage of lactate dehydrogenase from isolated hepatocytes, the rate of leakage increasing as freezing temperature decreased. Addition of 200-600 mM glucose to the incubation medium (similar to natural levels) fully protected cells against damage during freezing at -4 or -8 degrees C, normal freezing temperatures experienced by these frogs. Glucose also greatly improved freezing survival of isolated frog hepatocytes at ultralow temperatures (-80 or -196 degrees C).