Effect of insulin on glycogen metabolism in isolated catfish hepatocytes

Effect of insulin on glycogen metabolism in isolated catfish hepatocytes

0300-9629/84$3.00+ 0.00 0 1984Pergamon Press Ltd Vol. 78A, No. 4, pp. 705-710, 1984 Camp. Biochem. Physiol. Printed in Great Britain EFFECT OF INS...

719KB Sizes 1 Downloads 17 Views

0300-9629/84$3.00+ 0.00 0 1984Pergamon Press Ltd

Vol. 78A, No. 4, pp. 705-710, 1984

Camp. Biochem. Physiol.

Printed in Great Britain

EFFECT OF INSULIN ON GLYCOGEN METABOLISM IN ISOLATED CATFISH HEPATOCYTES C. OTTOLENGHI*, A. C. F’UVIANI*, A. BARUFFALDI~ and L. BRIGHENTI* *Institute of General Physiology and TInstitute of Comparative Anatomy, University of Ferrara, 44100 Ferrara, Italy. Telephone: 0532-34221 (Received 28 November 1983)

Abstract-l. Insulin effect on carbohydrate metabolism in caffish hepatocytes consisted of a significant decrease of cell glycogen concentration both in the absence and in the-presence of glucose in the medium. 2. The hormone did not influence either the output of glucose from the cell or the intracellular glucose level. 3. Experiments with radioactive glucose showed a very low uptake of the sugar by the hepatocytes; correspondingly the incorporation of radioactivity into glycogen was very low and not influenced by insulin. 4. The glycogen content in catfish liver cells was influenced by the hormone in the opposite way to rat liver cells.

INTRODUCTION In mammals, a fiue homeostatic control for an adequate supply of glucose to the it depends on a balanced action of hypoglycemic hormones. In fish the hormonal

is essential tissues and and hyper-

regulation of carbohydrate metabolism still seems to be very uncertain and an accurate control appears less important (Larsen, 1976; Lin et al., 1978; Ablett et al., 1981). The physiological significance of insulin, the only hypoglycemic hormone known in vertebrates, has been studied in fish by many authors (Gray, 1929; Root et al., 1931; Leibson and Plisetskaya, 1968; Tashima and Cahill, 1968; Epple, 1969; DemailSuard and Garin, 1970; Patent, 1970; Inui et al., 1975; Akhtar et al., 1979; De Roos and De Roos, 1979). However, a direct effect of insulin on carbohydrate metabolism of liver, as well as of heart and musculature, has not been clearly demonstrated, since different and sometimes opposite effects of the hormone have been observed in the various fish species studied (Bentley and Follet, 1965; Matty and Falkmer, 1965; Leibson and Plisetskaya, 1968). In catfish, either fed or fasted, we found (Ottolenghi et al., 1982) that insulin in vivo always induces an evident decrease of blood glucose level accompanied by a prolonged (over 72 hr) reduction in liver glycogen content which starts 4 hr after injection. Moreover, an increase in the glycogen level of white and red muscles occurs. The aim of the present work was to verify a direct effect of insulin in catfish liver glycogen studied on isolated hepatocytes. MATERIALS AND METHODS

Hepatocytes preparation

Adult catfish, Zctalurus melas, (200-300 g weight) were purchased from a local dealer and transferred to the laboratory storehouse in tanks containing 2001. of aerated dechlorinated tap water. Fish were anaesthetized by immersion (15 min) in a chloretone (1 , 1,1,-trichlor-Zmethyl2-propanol) solution (60ml of saturated solution/l tap 705

water). An abdominal incision was made to expose the liver and glass cannulae were inserted into hepatic porta and cava veins. The liver was then removed, washed and immediately placed in perfusion apparatus, similar to that previously described for isolated rat liver perfusion (Ottolenghi and Cavagna, 1968). The liver was perfused (5 min) with Ca2+-free teleost Ringer solution (PH 7.4), aerated with a 95% 0,: 5% CO, gas mixture, in order to remove red blood cells; the perfusion was continued (5min) with the same saline solution containing 0.02% EDTA (ethylenediaminetetraacetic acid). The medium was then replaced with teleost Ringer saline to which 1 mM Ca2+ and 0.1% (w/v) collagenase (Seglen, 1972) were added; the perfusion continued for about 30 min at room temperature. At the end of perfusion the cells were dispersed, centrifuged (for 5 min at 50g) and washed twice with 15 ml of cold teleost Ringer saline (pH 7.4). The yield of hepatocytes was determined by cell counting, using a haemocytometer; about 95-97x viable cells were obtained, as determined by exclusion of trypan blue (Seglen, 1972). Isolated hepatocytes from rats (fed or fasted for 48 hr) were obtained according to the procedure of Seglen (1972). Hepatocytes incubation

After washing, hepatocytes (about 50mg wet wt, corresponding to approx. lo6 cells) were suspended in 2 ml of teleost Ringer solution (PH 7.4) in 25 ml Erlenmeyer flasks and incubated in shaking bath for 1 hr at 30°C. Glucose, [U-Wlglucose (3 11 mC/mmol) (Radiochemical Centre, Amersham, UK) and insulin (porcine insulin, Novo Industries, Copenhagen, Denmark) were added according to experimental protocol, to a final volume of 2.5 ml. Analytical procedures Glycogen. After incubation, cell samples were centrifuged (50g for 5min at 4°C). Pellets were rapidly washed with cold unlabelled 1OmM glucose and after further centrifugation (5 min), boiled (30 min) with 2 ml 30% KOH. Glycogen was precipitated from extracts and assayed spectrophotometrically according to Seifter et al. (1950) using the anthrone method. For radioactivity measurements, 0.2ml of glycogen suspension were placed on Whatman glass microfibre paper discs (type GF/B), dried, placed in disposable minivials containing 4 ml Bray scintillation solu-



tion (Bray, 1960) and counted in an “Intertechnique SL 30” liquid scintillation counter. Glucose. After incubation, samples were centrifuged as for glycogen; 0.2 ml 20% HClO, and 1 ml 6.6% HClO, were added to supernatant and pellets, respectively. Since glycogen present in isolated hepatocytes can interfere with glucose determination, the polysaccharide was precipitated by adding to 1 ml of the perchloric solution 50~1 of 11% Na,SO,, 50 ~1 of 30% NaOH and 2 ml of absolute ethanol. After glycogen removal by centrifugation, glucose content assayed in supernatants using the glucose was oxidase-peroxidase method (Bergmeyer and Bernt, 1963) with a standard kit (Boehringer, Mannheim, FRG). The radioactivity was estimated on 0.2 ml of perchloric solution placed in disposable minivials containing 4 ml Bray solution (Bray, 1960) and counted in the same scintillation counter as glycogen.


Analysis of data. Values of both experimental data and percent variations are expressed as the mean + SEM. Statistical analysis was performed by the paired Student’s t-test.


After incubation for 1 hr the glycogen concentration in hepatocytes decreased about 20% (P < 0.01) both in the presence or in the absence of glucose (A in Fig. 1). Insulin enhanced this effect (B and C in Fig. l), mainly in the presence of glucose (P < 0.01). The effect of insulin on glycogen decrease after 60 min is shown by the bars in the lower site of the graph. As shown in Fig. 2, after incubation for 1 hr in the



’ -INS


+INS 1





+INS 10 mU







Ii T





10 mM






Fig. 1. Glycogen content of isolated catfish hepatocytes after incubation with insulin (5 and 10 mU in the sample) and glucose (10 and 20 mM). Each bar represents the mean _+ SEM. The number of samples for each experimental condition is indicated into the bar. Levels of significance (paired Student’s t-test): 0, P i 0.01 with respect to value at zero time; x , P < 0.05; and x x , P < 0.01 with respect to the value of the corresponding sample in the absence of insulin (A).


Effect of insulin on catfish glycogen metabolism -





+INS 5mU















10 mU


Fig. 2. Glucose content of isolated catfish hepatocytes after incubation with insulin (5 and 10 mU in the sample) and glucose (10 and 20 mM). Each bar represents the mean f SEM. The number of samples for each experimental condition is indicated into the bar. Levels of significance (paired Student’s t-test): 0, P < 0.05; and l 0, P < 0.01 with respect to value at zero time. absence of both insulin and glucose, a small but significant (P < 0.05) increase of intracellular glucose occurred. This increase rose greatly (more than 100% P < 0.01) when glucose was present in the medium;

but the difference between samples containing 10 or 20 mM glucose was not statistically significant. Insulin, at the two concentrations tested, did not show any effect both in the presence and in the absence of glucose. The glucose release from hepatocytes during incubation in a glucose-free medium, is shown in Table 1: the small amount of extracellular glucose present at the beginning of incubation rose considerably (more than 200x, P < 0.01). The output of glucose was not statistically influenced by insulin. The uptake of glucose and the glycogen turnover were also evaluated by addition of [U-‘4C]glucose to the medium. The incorporation of labelled glucose into glycogen (Table 2) was small but it was proportional to the added glucose concentration.

When glycogen radioactivity is expressed as 1-18[ 14C] glycogen/g glycogen, the incorporation of labelled glucose shows a low significant (0.10 > P > 0.05) increase when insulin is present, whereas this does not occur if the radioactivity is expressed as pg [‘“Cl glycogen/g cells. Table 3 shows the values of radioactivity of cell acid soluble fraction, which actually includes glucose as well as some soluble metabolites, such as lactate and intermediate compounds of the citric acid cycle. It appears that radioactivity present in this fraction was proportional to the concentration of added glucose and that insulin did not show any effect. For comparison, some experiments were performed on isolated liver cells from rats, either fed or starued. Results are reported in Table 4. Glycogen level in fed rat hepatocytes decreased more than 60% following incubation. Insulin did not affect this decrease both in the absence and in the presence of glucose. In starved rats the glycogen level of liver cells


708 Table

1. Glucose release by isolated catfish hepatocytes incubation with insulin in a glucose-free medium

Time (min)

Insulin (total mU in in the sample)

0 60 60 60

0 0 5 10


Glucose release ‘AVariation* mg/g Cells 3.58 i 10.09 * 9.64 + 9.89 *

0.52 0.66 0.80 0.92

+251.8 + 59.9t +224.9 + 52.0t +229.8 * 51.7t

Mean values + SEM of ten experiments. *With respect to control samples at zero time. tLeve1 of significance (paired Student’s t-test), P i 0.01

was very low; the incubation decreased the polysaccharide concentration, but to a lesser extent (about 20%) than in fed rats. Insulin increased slightly (about 20%, 0.10 > P > 0.05) the glycogen level in the presence of 10 mM glucose. The glucose content of hepatocytes increased in the presence of glucose in the medium, but insulin did not affect this content in hepatocytes from both fed and starved rats. The glucose release, higher in hepatocytes of fed rats, was not influenced by insulin. DISCUSSION

Several studies have been performed in vivo on carbohydrate metabolism, and on the role of insulin in fish. However, there has been little research (Hayashi and Ooshiro, 1979; Renaud and Moon, 1980) on isolated fish hepatocytes and none of them concerned catfish. Therefore, it is not easy to compare our data with those of other authors. The high glycogen content of catfish hepatocytes agrees with data which are reported in viuo (Ottolenghi et al., 1981). In our experimental conditions, results indicate that a 1-hr incubation of catfish hepatocytes in a glucose-free medium caused a decrease of glycogen content of liver cells, and that the incubation with glucose did not affect the decrease of glycogen level, nor did it stimulate its synthesis (Fig. 1). When labelled glucose was added to the medium, the radioactivity of glycogen was very low (Table 2), but doubled with doubling of glucose addition. Since the level of glycogen remained unchanged in the presence of glucdse in the medium (Fig. l), the incorporation of labelled glucose could represent an expression of a turnover, that nevertheless seems to Table 2. Incorporation of [U-‘4C]glucose into glycogen of isolated catfish hepatocytes after incubation with labelled glucose and insulin Glucose (mM)

Insulin (total mU in the sample)


[ %]glycogen formed pg/g Glycogen* pg/g Cells

10 10 10 20 20 20

0 5 10 0 5 10

9 9 9 6 6 6

165 + 64 174+64 167 + 50 360 + 139 366 & 145 320+ 118

3.06 + 1.18 3.79 _+1.44t 3.99 * 1.47t 6.94 _t 2.79 8.64 & 3.44t 8.35 + 3.41.t

Mean values + SEM. N, number of experiments. About 150,000 or 300,000 cprn of [U-‘%]glucose were added to each sample with carrier glucose 10 or 20 mM, respectively (=6000cpm/~mol glucose added). *Calculated from specific radioactivity of added [U-14C]glucose. tLeve.1 of significance (paired Student’s f-test), 0.10 > P > 0.05 with respect to samples without insulin.

Table 3. Radioactivity in the acid soluble fraction of isolated catfish hepatocytes after incubation with labelled glucose and insulin Glucose (mM)

Insulin (total mU in the sample)


[ ?Z]glucose in the cells fig/g Glucose* fig/g Cells

10 10 10 20 20 20

0 5 10 0 5 10

8 8 8 6 6 6

829 + 83 914 + 161 773 * 70 1528 I144 1492+ 169 1590+ 139

125 k 23 221+ 38 187_t22 337 + 29 343 * 20 353 & 29

Mean values + SEM. N, number of experiments. About 150,000 or 300,OOOcpm p-?Jglucose were added to each sample with carrier glucose 10 or 20 mM, respectively (= 6000 cpm/pmol glucose added). *Calculated from specific radioactivity of added [U-14C]glucose.

be very low. Probably this is due to a limited activity of the enzymes of glycogen metabolism. A low glycogen synthetase activity might be indicated by the fact that in our conditions the glycogen content of isolated hepatocytes, which was already high, was not affected by adding glucose. At the same time glycogen phosphorylase activity does not appear very high either, since only 20% of glycogen was broken down in the absence of glucose. These results confirm our findings obtained in vivo, namely that liver glycogen level in catfish remained high even after long periods of starvation (Ottolenghi et al., 1981). The glycogen content of hepatocytes was changed by insulin: in fact, under the influence of the hormone, the cell glycogen level decreased, chiefly when glucose was added to the medium (Fig. 1). The low incorporation of labelled glucose into glycogen appeared slightly increased by insulin when values were expressed as pg [ 14C] glycogen/g glycogen, but not when expressed as /*g [ 14C]glycogen/g cells (Table 2). This seems to confirm the effect of insulin on total glycogen content in liver cells, for labelled glycogen was greater than in controls if it was expressed with respect to the level of total glycogen, that was actually decreased by insulin. With regard to the low incorporation of radioactive glucose found in hepatocyte glycogen, if we can compare in vivo with in vitro experiments, our results agree with those of Ablett et al. (1981). They reported that, when radioactive glucose and leucine were injected in vivo in rainbow trout, the glucose radioactivity present in aqueous fraction of liver (which contains all carbohydrates), was low 8 hr after radiolabel administration and that insulin produced a reduction of both glucose and leucine radioactivity present in such a fraction. Moreover, Inui et al. (1978) did not observe any significant insulin effect on the incorporation of labelled glycine or its metabolites into glycogen, into TCA-soluble fraction, or into protein in the cyclostome Eptatretus stouti. It has been demonstrated (Ottolenghi et al., 1982), that in vivo effects of insulin on carbohydrate metabolism of catfish consist of a decrease of plasma glucose accompanied by a liver glycogen decrease, which starts 8 or 24 hr after treatment, according to feeding conditions. Two hours after injection, in vivo insulin effect on glycogen content is very small (a decrease of 15%); the decrease is of a similar magnitude to that observed in isolated hepatocytes in these experiments.

Effect of insulin on cattish glycogen



Table 4. Glycogen, intracellular glucose and glucose released by rat isolated liver ceils after incubation with glucose and insulin Time tmin)

Glucose (mM)

Insulin (total mU in the sample)

Glucose Glycogen


Released .-

1.8O_tO.11 1.81 If 0.25 2.03 i 0.22 8.61 + 208t 8.94 f 2.05t

0.163 *rt 0.041 0.564 0.116? 0.552 & 0.114t

Fed rats (9) $; (C) (D) (E)

600 ;;

(A) (B) (C) (D) (E)



60 60 60

0O 0 10 10

0 5 0 5

27.22 9.51 ?_ * 2.81 1.71t 9.17+ 1.35t 11.55 + 1.42t 11.75 * 1.591

0 0 1::

0 0 5 0 5

Starved rats (4) 0.427 & 0.078 0.340 +0x%5* 0.321 i 0.041+ 0.335 * 0.046 0.409 + 0.053$


0.612 0.383 0.352 4.547 4.783

rt 0.051 f 0.068 It: 0.068* k 0.589P i: 0.341t

0.010 * 0.003 0.039 * 0.012* 0.040 + 0.012*

Mean values & SEM, expressed as mg/g cells. Number of animals in parentheses. Levels of significance (paired Student’s r-test) in each group (fed or starved): *P < 0.05 and tP < 0.01 with respect to (A); and $P < 0.05 with respect to (D).

The incubation of hepatocytes in a glucose-free medium caused a small increase of intracellular glucose (Fig. 2) and a very significant rise of the sugar released in the medium (Table 1). When glucose was added, there was an important increase of sugar level in hepatocytes (Fig. 2). This is probably due chiefly to i~bition of the release, owing to the mass action effect of external glucose. In fact, the very low specific radioactivity of intracellular glucose (Table 3), indicates a small input of the sugar which is diluted with the cold glucose coming from glycogen breakdown or other cellular processes. Since the maximal level of intracellular glucose was already reached at IO mM concentra~on in the medium and since it was observed that total glycogen level did not increase, it is likely that some of the assumed glucose follows a metabolic pathway different from glycogen synthesis, perhaps as lipid or protein storage and/or it is metabolized by oxidative processes. Insulin did not affect the glucose content or the glucose release of isolated hepatocytes, both in the absence and in the presence of glucose in the medium. In rainbow trout, after insulin treatment, hypoglycemia occurs, accompanied by measurable glycogenolysis (Ablett et al., 1981); the authors think that in this fish the glucoregulatory role of insulin is to direct carbohydrate metabolites towards the oxidative pathway and/or towards possible storage compartments other than glycogen. Since insulin induced a decrease of total glycogen level (Fig. 1) in catfish liver cell too, it is possible to assume an analogous effect for this fish. In view of very low glucose oxidation rates in fish as compared to mammals {Stimpson, 1965; Bever et al., 1977; A~strong et al., 1979), a greater role of the hormone in control of protein and lipid metabolism can be supposed in the former. Our findings of glycogen level in fed rat hepatocytes agree with those found by Seglen (1974). The comparative study of insulin effect on isolated hepatocytes showed that there are many differences between the effects of the hormone in rat and in catfish. The glycogen level in fed rat hepatocytes was lower and the polysaccharide was less stable than in catfish hepatocytes, in fact the decrease of glycogen level after 1 hr of incubation was about 65% in rat and

20% in catfish liver cells. In rats starved for 48 hr the glycogen level was drastically reduced, so the incubation in a glucose free medium caused a decrease of about 20% only. Insulin was effective only on glycogen of starved rats hepatocytes and in the presence of glucose, with an increase of polysacharide of roughly 20%. This result is similar to that obtained by Ottolenghi et al. (1973) in the isolated perfused rat liver and agrees with data reported by Beynen and Geelen (1981). These authors, studying the control of glycogen metabolism by insulin in isolated hepatocytes prepared from fed rats and incubated with 10 mM glucose, found that insulin does not increase giycogen deposition as measured chemically; and they think that during incubation for 1 hr the increased synthesis and the depressed breakdown of glycogen, due to insulin, modify only to a small extent the total cellular glycogen level. The same authors found a significant increase of about 30% of the glycogen level only in hepatocytes prepared from rats starved for 24 hr, an increase similar to that observed by us. In conclusion, the results of the present study, together with our findings in vivo (Ottolenghi et al., 1982), indicate that insulin effect on carbohydrate metabolism in catfish hepatocytes consists chiefly in a decrease of cell glycogen concentration.

REFERENCES Ablett R. F., Russel O., Sinnhuber R. 0. and Selivonchick D. P. (1981) The effect of bovine insulin on [ ‘T]glucose and [3H]leucine incorporation in fed and fasted rainbow trout (Salmo guirdseri). Gera.camp. Endocr. 44,418-427. Akhtar S., Butt M. A. and Ahmad M. M. (1979) Effect of insulin and glucose loading on the blood sugar level of the freshwater teleost Channa punctatus. Pak. J. Zool. 11, 179-181. Armstrong M. K., Romsos D. R. and Reveille G. A. (1979) Glucose turnover in fasted Cinomolgus monkeys (Macaca fmciadaris) as measured by (2-3H) and (6-3H) and (U-‘“C) glucose. Comp. ~ioc~em. Physiol. 62A, 101l-1016. Bentley P. J. and Follet B. K. (1965) The effects of hormones on the carbohydrate metabolism of the lamprey (LampetraJuviatilis). J. Endow. 31, 127-137. Bergmeyer H. U. and Bernt E. (1963) D-glucose deter-



mination with glucose oxidase and peroxidase. In Melhads of Enzymatic Analysis (Edited by Bergmeyer H. U.), pp. 123-130. Academic Press, New York. Bever K., Chenoweth M. and Dunn A. (1977) Glucose turnover in kelp bass (Parahzbrux spp.). In uivo studied with (6-3H, 6-‘4C)-glucose. Am. J. Physiol. 232, R6&72. Beynen A. C. and Geelen M. J. H. (1981) Control of glycogen metabolism by insulin in isolated hepatocytes. Horm. Metub. Res. 13, 376378. Bray G. A. (1960) A simple efficient liquid scintillator for counting aqueous solutions in a liquid scintillation counter. Anaiyt. Biochem. 1, 279-285. Demael-Suard A. and Garin D. (1970) Mise en evidence du role de l’insuline dans la regulation du metabolism glucidique de Tinca tincu. C.R. Sot. Biol. Paris 164,831-835. De Roos R. and De Roos C. C. (1979) Severe insulin induced hypoglycemia in the spiny dogfish shark (Squalus ucunthias). Gen. camp. Endocr. 37, 186-191. Epple A. (1969) The endocrine pancreas. In Fish Physiology ‘(Edited‘by Hoar W. S. and-Randall D. J.), Vol. II, pp. 275-319. Academic Press. New York. Gray I. E. (1929) The effect of insulin on fishes. Sci. Mon. 28, 27 l-274. Hayashi S. and Ooshiro Z. (1979) Gluconeogenesis in isolated liver cells of the eel (Anguillu juponica). J. camp. Physiol. 132, 343-359. Inui Y., Arai S. and Yokote M. (1975) Gluconeogenesis in the eel-VI. Effects of hepatectomy, alloxan and mammalian insulin on the behaviour of plasma aminoacids. Bull. Japan. Sot. Sci. Fish. 41, 1105-l 111. Inui Y., Yu J. Y. L. and Gorbman A. (1978) Effect ofbovine insulin on the incorporation of [ t4C]glycine into protein and carbohydrate in liver and muscle of hagfish (Eptutretus stouti). Gen. camp. Endocr. 36, 133-141. Larsen L. 0. (1976) Blood glucose level in intact and hypophysectomized river lamprey (Lumpetrafluviatilis L.) treated with insulin stresses or glucose, before and during the period of sexual maturation. Gen. camp. Endocr. 29, l-13. Leibson L. and Plisetskaya E. M. (1968) Effect of insulin on blood sugar level and glycogen content in organs of some cyclostomes and fish. Gen. camp. Endocr. 11, 381-392. Lin II., Romsos D. R., Tack P. I. and Leveille G. A. (1978) Determination of glucose utilization in Coho salmon (C. kisutch) with (6-rH) and (U-‘4C)-glucose. Comp. Biochem. Physiol. 59A, 189-19 1. Matty A. J. and Falkmer S. (1965) Hormonal control of carbohydrate metabolism in Myxine glutinosu. Gen. camp. Endocr. 5. 701.

Ottolenghi C. and Cavagna R. (1968) Effect of hypophyseal growth hormone on the activity of tyrosine transaminase of the isolated perfused rat liver. Endocrinology 83, 924926. Ottolenghi C., Caniato A., Manservigi R. and Barnabei 0. (1973) Carbamylcholine and insulin control of glycogenesis in the isolated perfused rat liver. In Isolated Liver Perfusion and its Applications (Edited by Bartosek I., Guaitani A. and Miller L. L.), pp. 179-185. Raven Press, New York. Ottolenghi C., Puviani A. C. and Brighenti L. (1981) Glycogen in liver and other organs of catfish (Zctahaw melus): seasonal changes and fasting effects. Comp. Biothem. Physiol. 68A, 313-321. Ottolenghi C., Puviani A. C., Baruffaldi A. and Brighenti L. (1982) In uivo effects of insulin on carbohydrate metabolism of catfish (Zctalurus melus). Comp. Biochem. Physiol. 72A, 35-41. Patent G. J. (1970) Comparison of some hormonal effects on carbohydrate metabolism in an elasmobranch (Sqaalus acanthius) and a holocephalan (Hydrolagus colliei). Gen. camp. Endocr. 14, 215-242. Renaud J. M. and Moon T. W. (1980) Starvation and the metabolism of hepatocytes isolated from the American eel (Anguillu rostrata). J. camp. Physiol. B135, 127-137. Root R. W., Hall F. G. and Gray I. E. (1931) The influence of insulin on glycogen distribution in marine fishes. J. biol. Chem. 91, 27-35. Seglen P. 0. (1972) The effect of perfusate pH on respiration and glycolysis in the isolated rat liver perfused with an erythrocyteand protein-free medium. Biochim. biophys. Acta 264, 3988410. Seglen P. 0. (1974) Autoregulation of glycolysis, respiration, gluconeogenesis and glycogen synthesis in isolated parenchymal rat liver cells-under aerobic and anaerobic conditions. Biochim. bioohvs. Actu 338, 317-336. Seifter S., Seymour S., ‘Nbvic B. and Muntwyler E. (1950) Quoted by Hassid W. Z. and Abraham S. (1957) Chemical procedures for analysis of polysaccharides-I. Determination of glycogen and starch. In Methods in Enzymology (Edited by Colowick S. P. and Kaplan N. O.), Vol. III, pp. 34-40. Academic Press, New York. Stimpson J. H. (1965) Comparative aspects of the control of glycogen utilization in vertebrate liver. Comp. Biochem. Physiol. 15, 178-197. Tashima L. and Cahill G. F. Jr (1968) Effects of insulin in the toad fish (Opsanus tau). Gen. camp. Endocr. 11, 262-271.