GENERAL

AND

COMPARATIVE

Hormonal

ENDOCRINOLOGY

87, 44-53 (1992)

Effects on Glycogen Metabolism in Isolated Hepatocytes of a Freeze-Tolerant Frog THOMAS P. MOMMSEN” AND KENNETH B. STOREY?

*Department of Biochemistry and Microbiology, University of Victoria, Victoria, British Columbia, Canada V8W 3P6; and )“lnstitute of Biochemistry, Departments of Biology and Chemistry, Carleton Universiw, Ottawa, Ontario, Canada KlS 5B6 Accepted November 13, 1991 To determine whether specific hormonal responses were involved in the production of cryoprotectant (glucose) by liver of the freeze tolerant wood frog, Rana sylvatica, metabolically active hepatocytes were isolated in reasonable yields (mean 20.1 * 1.30% SEM, n = 29) by in situ liver perfusion with collagenase. Freshly isolated cells from autumncollected frogs contained large amounts of glycogen (650 pmol glucosyl units/g packed cells) and produced glucose from this endogenous reserve at a rate of 10 pmol g-’ hr-’ at 0”. Glucose output from cells was highly responsive to the addition of hormones; rates of glucose release increased 2.1-, 1.7-, and 1.7-fold with the addition of lo-’ M bovine glucagon, lo-’ M epinephrine, and 5 x 1O-6 M dibutyryl-cyclic AMP, respectively. Norepinephrine, 5-hydroxytryptamine, and bovine insulin were without effect at 0.1 FM/~. Hormone stimulation of glucose release was correlated with an increase in both the total activity and the percentage a of glycogen phosphorylase in hepatocytes. However, none of the hormones tested affected the kinetic properties of hepatocyte pyruvate kinase, suggesting the absence of covalent modification control of the enzyme. The data indicate that the freezingstimulated production of large quantities of glucose as a cryoprotectant by R. sylvatica liver does not involve qualitative differences in the hormonal control of liver glycogenolysis, compared with other lower vertebrates. However, quantitative differences were seen, such as the much greater phosphorylase activity, 4.38 + 0.33 pmol min-’ g-’ packed cells, in freshly isolated R. sylvarica hepatocytes compared with 0.36 2 0.06 pmol min-’ g-’ in Rana pipiens hepatocytes. 0 1992 Academic Press, Inc.

Several species of terrestrially-hibernating frogs have developed the capacity to tolerate freezing of extracellular body fluids as a strategy for winter survival (Storey, 1990). Among the adaptations supporting freezing survival is the accumulation of low molecular weight carbohydrate cryoprotectants. These act in a colligative manner to regulate cell volume reduction during freezing and also have protective functions in stabilizing proteins and macromolecular structures (Storey and Storey, 1988). The wood frog, Rana sylvatica, uses glucose as its cryoprotectant, which reaches amounts of up to 250-300 kmol/g wet wt in blood and organs of frozen frogs, compared with control values in unfrozen frogs of 1 to 5

pmol/g (Storey, 1987a; Storey and Storey, 1986). Glucose is produced from massive reserves of liver glycogen (up to 180 mg/g wet wt) that are accumulated for this purpose prior to hibernation. As soon as freezing starts, at peripheral sites in the skin, liver glycogenolysis is rapidly activated and glucose is released and pumped out into the blood to be quickly distributed to all other organs before the inward movement of freezing front halts blood circulation (Storey, 1987a; Storey and Storey, 1986). Glucose output from liver is stimulated within 5 min of the beginning of ice formation in the body extremities, resulting from an activation of liver glycogen phosphorylase and 44

0016-6480/92 $4.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

HORMONAL

EFFECTS

ON

leading to rates of glucose accumulation exceeding 20 p,mol g-’ hr-’ at -2.5” (Storey and Storey, 1988). The link between the stimulus of ice nucleation and the response of glucose output by liver must be hormonal or nervous. Indeed, in a recent study, we have found that intraperitoneal injections of the beta-adrenergic antagonist propranolol sharply reduce cryoprotectant output from liver during freezing (K. B. Storey and J. M. Storey, unpublished resuits) . The present study details for the first time the successful isolation of metabolically active hepatocytes from two amphibians, the wood frog Rana sylvatica and the leopard frog Rana pipiens. Subsequently, it examines the hormonal control of glycogenolysis in R. sylvatica cells at low temperature to determine whether cryoprotectant synthesis may involve unique responses to hormones by liver of a freezetolerant animal. MATERIALS

AND METHODS

Animals. Wood frogs (R. syivatica) of both sexes were caught in eastern Ontario during September. Animals were held in the laboratory at 3” without food for up to 2 months before use. Both juvenile frogs, several months postmetamorphosis (mean weight 1.46 + 0.05 g SEM, n = 18) and adult frogs, second year or older (mean weight 6.36 -+ 0.40 g SEM, IZ = 15), were used. R. pipiens (mean weight 23.8 + 2.70 g SEM, n = 7) were purchased from a local supplier and kept in a large holding tank at about 18” without feeding for 2-3 weeks before use. Preparation of hepatocytes. Frogs were doublepithed and then cannulated through the abdominal vein using drawn-out PE 50 tubing. Livers were flushed of blood with a balanced frog saline solution containing 114 mM NaCl, 1.0 mM MgSO,, 2.25 nnI4 KCl, 0.44 mM KH,PO,, 0.33 mM Na,HPO,, 13 mM NaHCO,, and 2% defatted BSA buffered with 10 mM Hepes to pH 7.63 at 22”. Hepatocytes were obtained by in situ perfusion with collagenase (0.03%) in frog saline for 30 mm at a rate of 2 ml g - r liver min - r. Cells were then isolated by differential centrifugation as detailed by French et al. (1981) (for fish systems) with the substitution of frog saline containing 2 mM CaCl,. For the last centrifugation step and the final suspension, hepatocytes were taken up in frog saline containing 4.5% BSA and 2 mM CaCl,. Hydrogen ion and

FROG

HEPATOCYTES

45

bicarbonate concentrations of the suspension were chosen to match in vivo conditions (K. Storey and D. G. McDonald, unpublished data). Protein concentration of R. sylvatica blood was approximately 45 mg/ ml (T. Mommsen and K. Storey, unpublished data). Hepatocytes from RR. pipiens were isolated by the same procedure, the only difference being the use of drawn-out PE 90 tubing for canmdation because of the larger size of these frogs. Hepatocytes yields were not above 20% of the initial liver weight. Experimental procedure. Aliquots of fresh hepatocytes were removed for gravimetric determination of cell weight and for measurement of initial (time zero) values for metabolites and enzymes. Remaining cells were divided into batches of 1 to 3 mg and incubated with or without hormones in 500~pl polypropylene microcentrifirge tubes in a total volume of 120 J. After a 90-min incubation at 0”, cells were sampled. An incubation temperature of 0” was chosen since cryoprotectant is only produced at near zero-degree temperatures. An aliquot for metabolite determinations was acidified with perchloric acid to a final concentration of 0.6 N. Remaining cells were used for enzyme determinations. These were sedimented by ceutrifugation in a microcentrifuge. Supernatants were discarded and pellets were taken up in 200 ul of ice-cold 50 mM imidazole-HCl buffer (pH 7.4 at 22”) containing 100 mM NaF, 5 m&I EDTA, 5 mM EGTA, and 25 mM 2-mercaptoethanol; the presence of protein kinase (EDTA and EGTA) and protein phosphatase (NaF) inhibitors preserved the phosphorylation state of the enzyme throughout the isolation and assay procedures. Cells were sonicated twice for 15 set each and then centrifuged for 30 set at I2,OOOg. Enzyme activities were measured in the supematant fraction. Rates of enzymes and glucose production were linear with time. Assay conditions. Conditions for glycogen phosphorylase a were: 50 mM potassium phosphate buffer, pH 7.0; 2 mg/ml glycogen (previously dialyzed); 0.4 mM NADP; 10 t&f glucose-1,6-bisphosphate; 15 mIt4 MgCl,; 0.25 mM EDTA; and excess phosphoglucomutase and glucosed-phosphate dehydrogenase with the addition of 1.6 m&I AMP for the determination of total (a + b) phosphorylase. Assay conditions for pyruvate kinase were: 50 rnn imidazole-HCl buffer; pH 7.0; 5 mM ADP; 0.12 mM NADH; 100 m&I KCl; 10 m&I MgCl,; excess lactate dehydrogenase and phosphoenolpyruvate at saturating (5.0 m&4) or subsaturating (0.05 mM) levels. All enzyme assays were conducted at 22” using a Pye Unicam SP S-100 recording spectrophotometer. One unit is defined as the amount of enzyme utilizing substrate at 1 Fmol per min. Glycogen (determined as glucose after hydrolysis with amyloglucosidase) and glucose were determined enzymatically (Keppler and Decker, 1974). Hormones. Hormone solutions were made fresh for each experiment and dissolved in frog sahne containing 4.5% BSA and 2 mM CaCl*. Bovine glucagon was

46

MOMMSEN

AND STOREY

purchased from CalBiochem Co. (San Diego, CA); bovine insulin, N6,2’-O-dibutyryl-3’:5’-cyclic AMP (dbCAMP), and other hormones were from Sigma Chemical Co. (St. Louis, MO). Biochemicals were from Sigma Chemical Co. or Boehringer Mannheim Corp. (Dorval, Quebec, Canada). Statistics. Statistical analysis was by paired t test or Student-Newman-Keuls test (Sokal and Rohlf, 1981).

Analysis of hepatocyte metabolism was carried out at 0” because our interest was in the control of cryoprotectant production and glucose is produced only when frogs freeze, i.e., at body temperatures of -0.5 to -2”. The rate of endogenous glucose production was not, however, dependent on the initial amount of glycogen present in individual preparations of cells (Fig. 1). Hepatocytes from R. pipiens contained much less glycogen and glucose and produced glucose at a rate of only 8% of that of the R. sylvatica cells (Table 1). The rate difference may largely reflect the difference in total glycogen phosphorylase activity in freshly isolated hepatocytes of the two species: 4.38 + 0.33 units/g packed cells for R. sylvatica versus 0.36 + 0.06 units/g packed cells for R. pipiens. Isolated hepatocytes of R. sylvatica were highly responsive to the addition of hormones implicated in the breakdown of glycogen in other vertebrates, including amphibians. As shown in Table 2, glucagon and epinephrine significantly increased the rate of production of glucose from endogenous glycogen by hepatocytes, with glucagon more than doubling the rate of glucose output into the surrounding medium. The second messenger analogue, db-CAMP, had the same effect. Norepinephrine, 5-hydroxytryptamine, and insulin, at the doses tested, did not alter the endogenous rate of glucose production. Interestingly, the glucagon-dependent rate of glucose production was negatively correlated with the basal rate (Fig. 2), established in the absence of hormones.

RESULTS Isolated hepatocytes from R. sylvatica excluded trypan blue (>95%), contained few red blood cells (95%) were obtained from leopard frogs but yields were lower (a maximum of 20%) than for wood frogs, and contamination with red blood cells was as high as 5%. Freshly isolated liver cells from R. sylvatica contained a high amount of endogenous glycogen (Table 1). As is common to freshly isolated hepatocytes from other sources (cf. Moon et al., 1985), frog hepatocytes were in negative glycogen balance during the first few hours following isolation. In wood frog hepatocytes incubated at O”, glucose was produced at an appreciable rate, the amount of glucose present in the cells doubling in less than 1 hr (Table 1).

TABLE 1 METABOLITE CONTENTS OF FRESHLY ISOLATED HEPATOCYTES FROM THE WOOD FROG (Rana LEOPARD FROG (Ranapipiens) Glucose

Glycogen pm01 g wet wt-’ Rana Rana

sylvatica pipiens

658 c 39.1 (29) 118 -t- 19.8 (7)

7.89 + 0.88 (28) 1.35 f 0.28 (6)

sylvatica)

AND

Rate of glucose production atO”u,molg-‘h-’ 9.97 -c 1.10 (28) 0.78 k 0.26 (6)

Note. Values are given as means 2 SEM with number of independent observations in parentheses. Glycogen is given in glucose units.

HORMONAL

EFFECTS

ON FROG HEPATOCYTES

I 10

I 20

0 ; 0 GLYCOGENOLYSIS

(pmoles

glucose

g-’

47

h-‘)

1. Independence of the rate of glycogenolysis and the initial glycogen content of isolated hepatocytes of Rana sylvatica. Glycogen content was determined at the start of each incubation. Glucose production was measured over 90 min at 0”. Each point represents an individual hepatocyte preparation. FIG.

When analyzing the characteristics of glycogen phosphorylase, the key enzyme involved in glycogenolysis, two strategies of enzyme regulation were apparent in R.

sylvatica hepatoqytes

(Table 3). First, the total activity of the enzyme assayable in hepatocytes increased following treatment with glucagon, epinephrine, or db-CAMP.

TABLE 2 EFFECTSOF HORMONESONGLUCOSEPRODUCTION FROMENDOGENOUSGLYCOGEN INISOLATED Rana sylvatica HEPATOCYTES Hormone Control Insulin (lo-’ M) Glucagon (lo-’ M) Dibutyryl-CAMP (5 x 10W6 M) Epinephrine (lo-’ M) Norepinephrine (low7 kf) 5Hydroxytryptamine (10e5 M)

Rate of glucose production 104 214 171 169 112 104

100 r 7.1 _t 16.6 t 26.0 i: 20.6 t 8.4 k 4.8

n 22 8 22 8 9 9 8

Significance NS P < 0.01 F < 0.01 P < 0.01 NS NS -

Note. Hepatocytes were incubated with or without hormones for 90 min at 0”. Values are relative rates (means -I- SEM) compared to the control value which is set at 100; n is the number of independent observations on different preparations of hepatocytes. The absolute rate of glucose production for control hepatocytes was 10.0 k 1.39 km01 glucose g wet wt-’ h-i at 0” (n = 22). P values show significant differences compared with control; NS, not significantly different from the control value.

48

MOMMSEN

%

INCREASE

AND STOREY

WITH

GLUCAGON

FIG. 2. Dependence of glucagon-mediated activation of glycogenolysis on the control rate of glycogenolysis, measured in the absence of added hormones, in isolated hepatocytes of Rana sylvatica. The percentage increase in glycogenolysis following glucagon treatment is plotted against the control rate in freshly isolated cells. Each point represents an individual preparation of hepatocytes. Regression line: y = 20.91 - 0.0513x; correlation coefficient: 0.63 (n = 223.

Second, treatment with these hormones resulted in a higher proportion of the enzyme converted to the active (a) form. In combination, these two effects resulted in a twofold increase in the activity of active phosphorylase a in cells exposed to glucagon or dibutyryl-CAMP and a 67% increase in epinephrine treated cells. Other hormone treatments had no statistically significant effect on phosphorylase activity although a trend was discernible for insulin. Following exposure to insulin, cells showed the lowest value for total phosphorylase activity as well as the lowest percentage of enzyme in the a form. Glucagon effects on phosphorylase were also tested on R. pipiens hepatocytes. Glucagon treatment did not significantly alter either the total activity of phosphorylase or the percentage of a (Table 3). The effects of hormone treatments on pyruvate kinase (PK) activity in R. sylvatica hepatocytes were also assessed. In mammals and fish, hormone action modulates the phosphorylation state and kinetic properties of the liver enzyme. However, as

shown in Table 4, the activity ratio (enzyme velocity at subsaturating versus saturating phosphoenolpyruvate), an indicator of PK substrate affinity, was not affected by hormone treatment. The observed ratios, about 0.4 in all cases for activity with 0.05 versus 5.0 mM phosphoenolpyruvate, are in line with measured Km values for phosphoenolpyruvate of about 0.08 mM for R. sylvatica liver pyruvate kinase (Storey, 1987b), suggesting that the properties of this enzyme were not altered during the experimental manipulation of hepatocytes. DISCUSSION

Our study clearly shows that the isolated hepatocyte system is well suited for studies of hormonal effects on hepatic intermediary metabolism in anurans. Hepatocytes represent an experimentally uniform system, an advantage over the cultured pieces of liver that have been used in other studies of amphibian liver metabolism (Brown et al., 1975; Janssens et al., 1983; Janssens and Grigg, 1984). Furthermore, since sufficient

HORMONAL

EFFECTS

ON

FROG

TABLE HORMONAL

EFFECTS

ON GLYCOGEN HEPATOCYTES

Treatment Rana

3

PHOSPHORYLASE ACTIVITY AND PERCENTAGE OF Rana sylvatica AND Rana pipkns

Total activity (units/g)

a IN ISOLATED

% Phosphorylase a

sylvatica

Control 0 min 90 mm Insulin (0.1 p&f) Glucagon (0.1 ~.uW) db-CAMP (10 CtM) Epinephrine (0.1 @f) Norepinephrine (0.1 l&) 5-Hydroxytryptamine (0. I JLIM) Rana

49

HEPATOCYTES

4.38 3.61 2.99 6.90 7.36 5.45 3.84 4.42

* i f + -e + 2 r

0.33 0.29 0.54 1.08* 0.63* 0.77* 0.52 0.34

(28) (27) (7) (18) (9) (8) (8) (9)

48.2 49.3 43.8 61.0 62.4 64.2 51.8 55.8

+ 2.5 2 2.3 + 5.1 -t 2.7* -+ 3.4* 2 3.9* +- 2.1 + 4.0

(28) (27) (7) (18) (9) (8) (8) (9)

pipiens

Control 0 min 90 min Glucagon (0.1 pm

0.36 + 0.06 (6) 0.33 -+ 0.06 (6) 0.50 + 0.08 (6)

42.3 a 8.8 (6) 56.5 2 3.8 (6) 49.7 k 5.4 (6)

Note. Cells were incubated for 90 min at O”, collected by centrifugation, and then extracted in buffer containing protein kinase and phosphatase inhibitors to prevent interconversion of a and b forms of glycogen phosphorylase. Enzyme activity was then assayed at 22”. One unit of activity is the amount of enzyme catalyzing the formation of 1 pmol of glucose-l-phosphate per minute at 22”. Data are units/g packed cells, means 2 SEM with n values for the number of separate determinations given in brackets. * Significantly different from time zero control values (also from 90-min control values) at P < Q.05 by the Student-Newman-Keuls test.

active cells for numerous parallel treatments could be obtained from frogs of even 1 g body wt, it was possible to use a randomized block design for the experiments and a paired analysis for the treatment of results. Glucose output by R. sylvatica hepatocytes was undoubtedly the result of glycogen breakdown, glycogen levels being extremely high (up to 180 mg/g wet wt, the equivalent of 1000 pmol/g as glucose) in autumn-collected wood frogs (Storey, 1987a; Storey and Storey, 1986). Hepatocytes were not supplied with exogenous precursors suitable as substrates for gluconeogenesis and it is unlikely that endogenous precursors, other than glycogen, could support glucose synthesis at the rates seen. In addition, the rates of hepatic gluconeogenesis from precursors such as lactate or alanine are very low in wood frogs (T. Mommsen and K. Storey, unpublished data). It is

noteworthy that, in contrast to piscine systems (Mommsen, 1986), the rate of glycogen breakdown in control hepatocytes of R. sylvatica was not dependent on the actual amount of glycogen present (Fig. 1). Furthermore, since this rate was negative’ly correlated with the glucagon-dependent rate of glycogen breakdown, it can be proposed that both rates are governed by an upper limit on the activity of glycogen phosphorylase. In support of this, the rate of glycogen breakdown in the absence of added hormones and the activity of glycogen phosphorylase were each 12- to 13-fold higher, respectively, in R. sylvatica, compared with R. pip&s, hepatocytes. Glucose output from R . sylvatica hepatocytes and the activity of gly,cogen phosphorylase responded to hormone stimulation by glucagon and epinebhrine. The same effects occurred in response to the addition of the cyclic AMP analogue, dibu-

50

MOMMSEN TABLE

4

EFFECT

OF HORMONES ON PYRUVATE KINASE ACTIVITY AT Low VERSUS HIGH PHOSPHOENOLPYRUVATE CONCENTRATIONS IN ISOLATED HEPATOCYTES OF Rana sylvatica

Hormone Control Glucagon Insulin Epinephrine 5-Hydroxytryptamine Norepinephrine Dibutyryl-CAMP

Activity ratio (O.OSi5.0 mM PEP) 0.42 0.39 0.40 0.41 0.42 0.46 0.39

+- 0.02 f 0.02 +- 0.03 + 0.02 k 0.01 i 0.02 -+ 0.02

(24) (15) (7) (8) (7) (6) (9)

Note. Isolated hepatocytes were incubated with or without hormones (hormone concentrations as in Table 2) for 90 min at 0” and then collected by centrifugation. Pyruvate kinase activity was measured at low (0.05 mM, subsaturating) and high (5.0 mM, saturating) phosphoenolpyruvate with all other substrates/ cofactors at saturating levels. Values are means t SEM with the number of independent observations in parentheses.

tyryl-CAMP. Similar responses to these hormones occur throughout the vertebrates including mammals, fish, reptiles, and other amphibians (Harri and Lindgren, 1972; Brown et al., 1975; Farrar and Frye, 1979; Janssens et al., 1983; Moon et al., 1985; Janssens and Giuliano, 1989). The effectiveness of db-CAMP in eliciting the same metabolic responses as occurred under glucagon or epinephrine stimulation is consistent with studies of the hormonal regulation of glycogenolysis in other amphibian species. Thus, glucagon-stimulated glycogenolysis in Ambystoma mexicanum liver by increasing CAMP concentrations (Janssens and Maher, 1986) and catecholamine stimulation of glycogenolysis in amphibians was mediated via p-adrenergic, CAMP-dependent mechanisms (Janssens et al., 1983; Janssens and Grigg, 1984). Alphaadrenergic, Ca2+ -mediated mechanisms do not appear to be important in lower vertebrates (Janssens et al., 1983; Janssens and Grigg, 1984, 1987; Janssens and Giuliano, 1989).

AND STOREY

The doses of catecholamines used in the present study were chosen to be within the measured range of blood levels for these compounds in R. sylvatica (K. Storey and S. Perry, unpublished data). At these doses, neither norepinephrine nor 5-hydroxytryptamine affected hepatocyte metabolism in R. sylvatica and are unlikely, therefore, to be involved in the stimulation of cryoprotectant synthesis in vivo. Differential effects of epinephrine versus norepinephrine on amphibian glycogenolysis have also been noted by Janssens and Grigg (1984) who found that the half-maximal dose for eliciting glucose release from Xenopus laevis liver pieces in vitro was 1.4 X 1O-8 M for epinephrine but 23-fold higher for norepinephrine. By contrast, Harri and Lindgren (1972) found that the doseresponse curves for the two hormones were nearly identical for the stimulation of hyperglycemia in vivo in Rana temporaria. In general, then, our present results support the conclusions of previous studies on the hormonal control of glycogenolysis in amphibians. In this respect, we found no unique hormonal effects such as might be adaptive for the rapid biosynthesis of cryoprotectant in the freeze-tolerant frog. However, glycogen metabolism in hepatocytes of R. sylvatica differed in two significant ways from the freeze intolerant species R. pipiens: (1) the total activity of glycogen phosphorylase in R. sylvatica hepatocytes was 1Zfold greater than in R. pipiens hepatocytes, and (2) glycogen content of R. sylvatica hepatocytes was six times greater than that of R. pipiens cells. This indicates that one strategy of adaptation for enhancing the capacity for rapid glucose synthesis as a cryoprotectant is a quantitative, not qualitative, change in both the enzyme and the substrate of liver glycogen metabolism. The accumulation of large amounts of glycogen in liver as a preparation for hibernation has been reported for a number of anuran species from northern latitudes (Schlaghecke and Blum, 1978; Farrar and

HORMONAL

EFFECTS

ON

Dupre, 1983). Freeze-tolerant frogs have increased this glycogen content to very high levels (and increased the proportional size of liver) to accommodate the quantity of cryoprotectant that must be produced (organ contents of glucose can range to 300 pmol/g wet wt in frozen R. sylvatica) (Storey, 1987a; Storey and Storey, 1986). Furthermore, Harri and Lindgren (1972) observed that epinephrine readily induced hyperglycemia in cold-acclimated (5”) but not warm-acclimated (25”) R. temporaria and suggested that catecholamine regulation of blood glucose levels was important to anuran survival at cold temperatures. This response may have been adapted and exaggerated by freeze-tolerant frogs and tied to the trigger of peripheral ice nucleation to produce the massive build-up of cryoprotectant glucose that is initiated by freezing. At the enzymatic level, biochemical adaptations have also been made that pace the rapid output af glucose from liver during the early minutes/hours of freezing. For example, blood glucose can rise by 3.3-fold within 5 min of the initiation of ice nucleation (Storey and Storey, 1988). This is permitted by the maintenance of high activities of glycogen phosphorylase in liver and by an uncommon feature of the phosphorylase activation process in R. sylvarica liver. Thus, freezing stimulates both an increase in the percentage of phosphorylase in the a form and an increase in the total enzyme activity expressed. These responses have different time frames, the percentage of a increasing within 2 min whereas an increase in total phosphorylase activity does not appear until about 30 min after the start of freezing at -2.5“ (Storey and Storey, 1988). The present results show that both of these responses can be triggered by glucagon or epinephrine in isolated liver cells, with the comparable effects of db-CAMP indicating that’ these are probably ‘mediated in vivo by CAMP as the second messenger. Glucagon treatment, for example, resulted in a net twofold increase in the activity of

FROG

HEPATOCYTES

51

active phosphorylase a in hepatocytes (a combination of the increase in total phosphorylase activity from 4.38 to 6.90 units/g packed cells and increased percentage of phosphorylase a from 48.2 to 61.0%) (Tab1.e 3) whereas in vivo the combination of these two effects resulted in a rise in phosphorylase a activity of 7- to l-)-fold within 2-3 h of freezing exposure at - 2.5” (Storey and Storey, 1988). Thus, the hormonally-induced mechanisms of phosphorylase activation are the same in isolated hepatocytes in vitro as those seen in vivo in response to whole animal freezing. It therefore appears likely that hormones may mediate the freezinginduced production of glucose in vivo. The molecular mechanisms involved in phosphorylase activation in R. sylvatica have recently been explored and appear to suit the particular needs of R. sylvatica liver: (1) to hold phosphorylase activity within limits in the unfrozen frog, (2) to support extremely high glucose output over the early minutes/hours of freezing despite a subzero body temperature, and (3) to sustain high phosphorylase activity even as levels of glucose, a potent enzyme inhibitor, rise to several hundred millimolar (Crerar et al., 1988; Risman et al., 1991). In both mammalian and piscine systems, glucagon actions on liver include an inactivation of pyruvate kinase via CAMPmediated protein phosphorylation (EngStrom et ‘al., 1987; Mommsen and Moon, 1990). This promotes gluconeogenesis. Glucagon-mediated phosphorylation of pyruvate kinase produces an enzyme form that has a greatly reduced affinity for phosphoenolpyruvate, as well as altered responses to inhibitors and activators. However, the present results, using an activity ratio (0.05/ 5.0 mM phosphoenolpyruvate) to assless changes in substrate affinity of pyruvate kinase, gave no indication af stimulated modification of this enzyme in R. sylvatica hepatocytes. Although CQSItrary to the situation in mammals and fish, these results agree with other recent studies

52

MOMMSEN

of frog liver pyruvate kinase. Both R. sylvatica and R. pipiens liver pyruvate kinase showed low K, values for phosphoenolpyruvate and hyperbohc substrate saturation kinetics (Fournier and Guderley, 1986; Storey, 1987b). Freezing exposure modified the properties of glycogen phosphorylase and phosphofructokinase (via enzyme phosphorylation) in R. sylvatica liver but did not alter the properties of pyruvate kinase (Storey, 1987b). In addition, R. pipiens liver pyruvate kinase showed no alteration of kinetic or regulatory properties under conditions that should promote enzyme phosphorylation by CAMP-dependent mechanisms (Fournier and Guderley, 1986). Overall, then, it appears that anuran liver pyruvate kinase is not subject to regulation via CAMP-dependent, hormonestimulated reversible phosphorylation. ACKNOWLEDGEMENTS We thank Dr. F. Schueler for the collection of wood frogs, Ms. S. Walters for excellent technical assistance, and J. Storey for critical reading of the manuscript. Supported by operating Grants from the National Institutes of Health (GM43796) to K.B.S. and from the Natural Sciences and Engineering Research Council to T.P.M.

REFERENCES Brown, D., Fleming, N., and Balls, M. (1975). Hormonal control of glucose production by Amphiuma means liver in organ culture. Gen. Cornp. Endocrinol. 27, 380-388. Crerar, M. M., David, E. S., and Storey, K. B. (1988). Electrophoretic analysis of liver glycogen phosphoryiase activation in the freeze-tolerant wood frog. Biochim. Biophys. Acta 971, 72-84. Engstrom, L., Ekman, P., Humble, E., and Zetterqvist, 0. (1987). Pyruvate kinase. In “The Enzymes” (P. D. Boyer and E. G. Krebs, Eds.), Vol. 18, pp. 47-75. Academic Press, New York. Farrar, E. S., and Frye, B. E. (1979). A comparison of adrenalin and glucagon effects on carbohydrate levels of larval and adult Rana pipiens. Gen. Comp. Endocrinol. 39, 372-380. Farrar, E. S., and Dupre, R. K. (1983). The role of diet in glycogen storage by juvenile bullfrogs prior to overwintering. Comp. Biochem. Physiol. A 75, 255-260.

AND STOREY Fournier, P., and Guderley, H. (1986). Evolution of the functional properties of pyruvate kinase isozymes: Pyruvate kinase L from Rana pipiens. J. Comp. Physiol. 156, 691-699. French, C. J., Hachachka, P. W., and Mommsen, T. P. (1981). Utilization of amino acids in isolated hepatocytes from rainbow trout. Eur. .I. Biochem. 113, 311-317. Harri, M. E., and Lindgren, E. (1972). Adrenergic control of carbohydrate metabolism in the frog, Rana temporaria. Comp. Gen. Pharmacol. 3, 226-234. Janssens, P. A., Caine, A. G., and Dixon, J. E. (1983). Hormonal control of glycogenolysis and the mechanism of action of adrenaline in amphibian liver in vitro. Gen. Comp. Endocrinol. 49, 477-484. Janssens, P. A., and Grigg, J. A. (1984). Adrenergic regulation of glycogenolysis in liver of Xenopus laevis in vitro. Comp. Biochem. Physiol. C 77, 403-408. Janssens, P. A., and Grigg, J. A. (1987). Hormonal regulation of hepatic glycogenolysis in the toad Xenopus laevis is mediated by cyclic AMP and not Cazf. Gen. Comp. Endocrinol. 67, 227-233. Janssens, P. A., and Giuliano, M. (1989). Hormonal regulation of hepatic glycogenolysis in Amphibolurus nuchalis, the eastern netted dragon: An in vitro study. J. Comp. Physiol. B 159, 323-331. Janssens, P. A., and Maher, F. (1986). Glucagon and insulin regulate in vitro hepatic glycogenolysis in the axolotl Ambystoma mexicanum via changes in tissue cyclic AMP concentration. Gen. Comp. Endocrinol. 61, 64-70. Keppler, D., and Decker, K. (1974). Glycogen: Determination with amyloglucosidase. In “Methods of Enzymatic Analysis” (H.-U. Bergmeyer, Ed.), pp. 1127-113 1, Verlag Chemie, Weinheim. Mommsen, T. P. (1986). Comparative gluconeogenesis in salmonid hepatocytes. Can. J. 2001. 64, 1110-1117. Mommsen, T. P., and Moon, T. W. (1990). Metabolic actions of glucagon-family hormones in teleost hepatocytes. J. Endocrinol. 126, 109-118. Moon, T. W., Walsh, P. J., and Mommsen, T. P. (1985). Fish hepatocytes: A model metabolic system. Can. J. Fish. Aquaf. Sci. 42, 1772-1782. Risman, C. A., David, E. S., Storey, K. B., and Crerar, M. M. (1991). Glucose and caffeine regulation of liver glycogen phosphorylase activity in the freeze-tolerant wood frog, Rana sylvatica. Biothem. Cell Biol. 69, 251-255. Schlaghecke, R., and Blum, V. (1978). Seasonal variations in liver metabolism of the green frog, Rana esculenta (L.). Experientia 34, 45-57.

HORMONAL

EFFECTS

Sokal, R. R., and Rohlf, F. J. (1981). Biomefry. 2nd ed. Freeman, San Francisco. Storey, K. B. (1987a). Organ-specific metabolism during freezing and thawing in a freeze-tolerant frog. Am. J. Physiol. 253, R292-R297. Storey, K. B. (1987b). Glycolysis and the regulation of cryoprotectant synthesis in liver of the freeze tolerant wood frog. J. Comp. Physiol. B 157, 373380.

ON

FROG

HEPATOCYTES

53

Storey, K. B. (1990). Life in a frozen state: Adaptive strategies for natural freeze tolerance in amphibians and reptiles. Am. J. Physiol. 258, R559-R568. Storey, K. B., and Storey, J. M. (1986). Freeze tolerant frogs: Cryoprotectants and tissue metabolism during freeze/thaw cycles. Can. J. Zool. 64, 4956. Storey, K. B., and Storey, J. M. (1988). Freeze tolerance in animals. Physiol. Rev. 68, 27-84.

Hormonal effects on glycogen metabolism in isolated hepatocytes of a freeze-tolerant frog.

To determine whether specific hormonal responses were involved in the production of cryoprotectant (glucose) by liver of the freeze tolerant wood frog...
917KB Sizes 0 Downloads 0 Views