398 Biochimica et Biopl~vsica Acta 846 (1985) 398 404 Elsevier

BBA 11545

Effects of insulin on glucose metabolism in isolated heart myocytes from adult rats

Victor Chen *, Kathleen H. McDonough and John J. Spitzer **

Department of Physiology, Louisiana State Unit~ersitv Medical ('enter. 1901 Perdtdo Street. New Orleans, LA 70112 tU.S.A.)

(Received December 21st, 1984)

Key words: Glucose metabolism: Insulin: (Rat myocyte)

The study examined the effect of insulin on glucose metabolism in freshly isolated calcium-tolerant heart myocytes from adult rats. The uptake of 2-deoxyglucose demonstrated an initial lag in response to insulin and the maximal insulin effect was not attained until after 3 min preincubation with the hormone. A dose-response study of 14CO2 production from [14C]glucose revealed that the maximum insulin stimulation of glucose utilization occurred with 5 m U / m i . Both the uptake and the oxidation of glucose proceeded at a linear rate in the absence and presence of insulin. However, insulin exerted a greater effect on the uptake (42-54%) than on the oxidation (17-22%) of exogenous glucose. Incorporation of glucose into glycogen was markedly increased by insulin and resulted in the myocyte glycogen concentration returning to in vivo levels. In the absence of insulin, glucose incorporation plateaued within 10 min of incubation and the glycogen concentration was not altered. Our findings also indicate that at equilibrium, insulin-treated cells exhibited a

higher glycogen turnover rate. It thus appears that insulin exerts a differential effect on the different pathways in glucose metabolism in the isolated cardiac cells. This may be related in part to their quiescent state and lower energy demand.

i n t r o d u c t i o n

The stimulatory effect of insulin on myocardial glucose metabolism has been demonstrated in iso- lated perfused hearts [1-3] and more recently in freshly isolated heart cells from adult rats [4-7]. Dose-response studies [6,7] have shown that the half-maximal and maximal insulin effects on glu- cose uptake by the isolated myocytes are compara- ble to those observed in other cell preparations such as the isolated adipocytes [8]. The data con- cerning maximal stimulation of glucose uptake

and utilization by insulin are, however, not uni- form [4-7]. The discrepancies may be due in part to differences in the physical and metabolic state of the isolated heart myocytes from different pre- parations. They may also imply a differential ef- fect of insulin on glucose metabolism. This study was therefore undertaken to examine how insulin affects the uptake of exogenous glucose and its flux through the glycolytic pathway into either the tricarboxylic acid cycle or into the production of lactate as well as glucose incorporation into glyco- gen stores.

* Present address: Department of Pathology, Yale University School of Medicine, 310 Cedar Street, New Haven, ( 'T 06510 (U.S.A.)

** To whom correspondence should be addressed.

Materials and Methods

Isolation of heart cells. Male Sprague-Dawley rats weighing 176 200 g were used in all experi-

0167-4889/85/$03.30 ' 1985 Elsevier Science Publishers B.V. (Biomedical Division)

ments. Calcium-tolerant myocytes were isolated by initial perfusion of the heart and subsequent mechanical dispersion of diced ventricular tissue with 0.1% collagenase in Joklik medium supple- mented with 0.1% bovine serum albumin, 1 mM MgSO 4 and 8 mM glutamate as previously de- scribed [9]. The isolated cells were washed thrice with Joklik solution, containing 1% bovine serum albumin, 0.5 mM CaC12 (1 and 1.5 mM in subse- quent washes) plus 5 mM glucose and were used immediately for the experiments. Cell viability was estimated by elongated morphology and exclusion of Trypan blue stain. Protein concentration of the cells was determined [10] with bovine serum al- bumin as standard.

Aliquots of cell suspensions were incubated in the absence or presence of insulin at 10 m U / m l (with the exception of dosage studies) for 10 min. Preliminary study has shown that the maximal response to insulin is attained in the isolated cardiac cells after 10 rain preincubation with the hormone. The incubation medium for all experi- ments was composed of Joklik medium supple- mented with 1 mM MgSO 4, 1.5 mM CaC12, 1 mM L-carnitine, 1% bovine serum albumin and 5 mM nonlabeled glucose. At 5 mM glucose, saturation kinetics were achieved in both insulin-treated and untreated myocytes. All experiments were per- formed in a shaker bath maintained at 37°C and 100 cycles/rain. An aliquot of the incubation medium was counted in the liquid scintillation counter and used for determination of specific activity of the substrate pool (cpm/nmol) .

The uptake of 2-deoxvglucose. This sugar is phosphorylated, but not further metabolized, and is therefore indicative of glucose uptake by the cells [11]. The reaction was initiated by adding deoxy[2-3H]glucose and [14C]inulin to cell suspen- sions preincubated with or without insulin. [~aC]Inulin was added to estimate extracellular trapping of 2-deoxyglucose. Approx. 5 s before the end of the incubation, an aliquot of the cell sus- pension was transferred to a microfuge tube con- taining bromododecane and was immediately centrifuged in a Beckman microfuge for 30 s. In an earlier study [6], cells were centrifuged with per- chloric acid in addition to bromododecane. We found this step to be unnecessary and it was therefore omitted in this study. The tip of the tube

399

was cut off just above the cell pellet and trans- ferred to a scintillation vial. The pellet was solubi- lized and the radioactivities were counted in a Beckman liquid scintillation counter equipped with a microprocessor for dual label counting. In each sample, extracellular trapping of 2-deoxyglucose was less than 5% of the total radioactivity. Both nonspecific glucose uptake, i.e., by simple diffu- sion as indicated by L-glucose uptake, and 2-de- oxyglucose efflux were negligible during the course of the experiment (data not shown).

Glycolytic flux of glucose. The rate of glycolysis, i.e., glucose hydrolysis to pyruvate, in the cardiac cells was determined by the production of ~H~O from [2-3H]glucose, as tritium in the 2-position of glucose is lost into water in glycolytic reactions [12]. The production of 3H20 from [2-~H]glucose by the heart cells was determined, according to Clark [13], after 40 rain of incubation with or without different concentrations of insulin. The reaction was stopped by adding 10% trichloro- acetic acid. After centrifugation, the acid super- hate was applied to a Dowex column to separate 3H20 from [2-3H]glucose. We found that 99 103% of the tritiated glucose was retained by the col- umn, whereas 97-101% of the 3H20 remained in the effluent.

Glucose oxidation. Dose-response curve of in- sulin versus glucose oxidation was determined by the production of 14CO~ from [U-UC]glucose dur- ing 40-rain of incubation without or with insulin at various concentrations. The 14CO: was trapped by hyamine hydroxide and counted. The time-course of glucose oxidation was also determined and in some experiments oxygen consumption by the heart cells was measured with a Clark electrode [14].

Glucose incorporation into glycogen. The reaction was initiated by adding [U-14C]glucose to myocytes preincubated with 5 mM non-labeled glucose in the presence and absence of insulin. At the end of incubation, the cell suspensions were immediately centrifuged, the supernate discarded, and 30% hot KOH added to hydrolyze the cells. Glycogen was then precipitated with 95% ethanol and washed several times with 60% ethanol. Preliminary stud- ies showed that 93-98% of the glycogen was re- covered by this procedure, and that only a negligi- ble amount of [U-14C]glucose remained trapped in

400

the glycogen pellet. For the estimation of glycogen content, glucose was omitted during preincubation with insulin. After 40 min of incubation with 5 mM non-labeled glucose, glycogen was extracted from the myocytes and its concentration de- termined [15].

Statistical analysis. Data from insulin-treated and control groups were analyzed by dependent Sludent's t-test. Significance was set at P < 0.05.

Chemicals. Joklik-modified minimum essential medium was purchased from KC Biological (Kansas City, KS), and collagenase from Worthington Biochemicals (Freehold, N J). Bio- chemicals, including bovine insulin and bovine serum albumin (essentially fatty-acid free) were purchased from Sigma (St. Louis, MO). Radioiso- topes were obtained from New England Nuclear (Boston, MA) and Amersham (Arlington Heights, 1L). Other reagents were commercial products of the highest available purity.

Results

The viability of the isolated myocytes was 78 + 1% ( N = 33). In experiments in which the myocytes were incubated for 40 rain, viability decreased slightly to 72 _+ 1% (N = 17) after incubation.

In order to determine the time-course of maxi- mum insulin stimulation, the heart cells were pre- incubated with insulin for varying time periods, and 2-deoxyglucose uptake estimated in a 30 s period. It is shown in Fig. 1 that insulin effect was

o) 4 -

-&.$

3 ~ 3 - ~ a

~ E 2 - (3~e1

X - - 0 0 t e E

L~ 0 l u

i

1

T

,i J

f I • [

Preincubat ion time (rain)

Fig. I. Time-course of insulin action on 2-deoxygiucose uptake by the isolaled myocytes. The cells were incubated with insulin (10 m U / m l ) for indicated time periods before 2-deoxyglucose uptake in a 15 s time period was determined. Values are means _4-S.E. of four experiments.

.c 4

o

E

2 g

o 1 ̧ 3:

CO

~, 0 co

r T T

10 2 0 3 0

i n s u l i n ( m u / m l )

- 4 £

o T-3o

3 5 '

"1 3

o

0 ~"

40 ~g

Fig. 2. Dose-response curves for effect of insulin on glucose utilization by isolated myocytes: 3H20 production from [3H]glucose (©) and 14C02 production from [U-~4C]glucose (O). Values are means_+ S.E. of five experiments. Glucose utili- zation rates were significantly (P < 0.05) increased when the cells were incubated with insulin at all dosage levels studied.

not evident until the myocytes were preincubated with insulin for at least 30 s, and maximum re- sponse was not achieved until after 3-4 min of incubation with insulin.

Fig. 2 shows the dose-response relationship be- tween insulin and glucose utilization in the iso- lated heart cells. Both the production of 3H20 from 5 mM D-[2-3H]glucose and 14CO 2 from 5 mM D-[U-14C]glucose responded to insulin in a

A C

212 . . .10

o 8

v

= 4 == o

~ 0 o

"o I cq

30 60 90 incubation time (sec)

120

Fig. 3. Uptake of 2-deoxyglucose by isolated myocytes in- cubated in the absence (D D) or presence (ll B) of insulin. Values are means_+S.E, of six experiments. Glucose uptake is significantly (P <0.001) higher in insulin-treated cells than in untreated cells.

.E 120

o

~100

~ 80

~ 60

~ 40

~ 20 o E c

o ,o 2'0 3'o 4'0 incubation time (rain)

Fig. 4. Oxidation of 5 mM glucose by isolated myocytes in the absence (iS D) or presence (111 m) of insulin. Values are means-+S.E, of four experiments. Glucose oxidation is significantly (P < 0.01) increased i n insulin-treated cells as compared to untreated cells.

similar pattern. Half-maximal insulin effect was achieved at approx. 1 m U / m l ( 4 . 5 0 - 9 M ) and maximum ( + 23% and + 17%, respectively) stimu- lation at 5 m U / m l (2.2.10 -8 M).

The rate of 2-deoxyglucose uptake by cells in- cubated with insulin was 42-54% greater than that in cells incubated without insulin (Fig. 3). The rate of uptake remained constant during the course of

2 8- o

E "~ 6 "

0 O.

4 " 0 o

o 2 o

o~

2 4 6 8 10 incubation time (min)

Fig. 5. Incorporation of exogenous glucose (5 mM) into glyco- gen in isolated myocytes in the absence (El E]) or pres- ence (11 II) of insulin. Values are means+S.E , of five experiments. Glucose incorporation rate is significantly (P < 0.01) higher in insulin-treated cells than in untreated cells.

401

TABLE 1

[~4C]GLUCOSE INCORPORATION INTO G L Y C O G E N IN ISOLATED MYOCYTES AFTER PREINCU BATION WITH NON-LABELED GLUCOSE

Values are means_+ S.E. of four experiments. Glucose incorpo- ration (nmol /10 m i n / m g protein) is significantly ( P < 0.001) higher in insulin-treated cells than in untreated cells.

Insulin Preincubation time (min):

( m U / m l ) 10 40 60

2.10_+0.31 1.90_+0.21 2.38_+0.35 10 5.82_+0.70 3.23_+0.35 3.72_+0.53

incubation regardless of the absence or presence of insulin. The regression lines were y = 1.10 + 2.56 x and y = 1.77 + 3.53 x, respectively.

A time-course study of the effect of insulin on glucose oxidation is shown in Fig. 4. Glucose oxidation in the insulin-treated cells was 17-22% greater than that in the non-treated cells. In ad- dition, the oxidation rate in the presence as well as the absence of insulin remained linear during the 40 min incubation. The respective regression lines were y = 0.13 + 2.23 x and y = 1.15 + 2.57 x. De- spite the increased glucose oxidation in the pres- ence of insulin, oxygen consumption by the iso- lated myocytes remained unaltered at 16.6 + 6.9 and 15.8_+ 4.5 n m o l / m g protein per min when incubated with and without insulin.

Incorporation of [U-14C]glucose into glycogen proceeded at a much faster rate when the isolated myocytes were incubated with insulin (Fig. 5). Within 5 min, insulin-treated cells had accu- mulated 55% more labeled glucose, whereas after 10 min incubation the increase was 153%. The incorporation gradually increased over 20 min be- fore reaching equilibrium (data not shown). The increase in glucose incorporation was accompa- nied by an increase in glycogen concentration (from 61 + 8 to 86 + 8 n m o l / m g protein after 40 min incubation with insulin). Without insulin, glu- cose incorporation plateaued within 10 min and glycogen content remained unchanged at 64_+ 7 n m o l / m g protein. Glycogen synthesis was also estimated under steady-state conditions by [U- ~4C]glucose incorporation into glycogen after the cells had been incubated with non-labeled glucose

402

for 40 and 60 min (Table I). In the absence of insulin, the amount of radioactivity accumulated during a 10 min incubation, following 40 or 60 rain incubation with non-labeled glucose, was comparable to that in cells preincubated with glu- cose for only 10 min, suggesting a constant glucose turnover in glycogen. However, insulin stimulation of glucose incorporation into glycogen was still demonstrated when glucose turnover had already attained equilibrium.

Discussion

The results of this study suggest a differential stimulatory effect of insulin on glucose metabo- lism in the freshly isolated heart myocytes. Insulin has a greater influence on glucose uptake and incorporation into glycogen than on glucose hy- drolysis. It appears that the effect of insulin on glucose hydrolysis to pyruvate is greater than on glucose flux all the way through to oxidation. This may be related to an already near-maximal rate of glucose oxidation in the quiescent myocyte, even in the absence of insulin. In the absence of insulin, 87% of the phosphorylated glucose was eventually oxidized (Figs. 3 and 4). By converting CO 2 pro- duction from glucose to oxygen equivalents, it can be seen that the oxidation of glucose accounted for most (83%) of the oxygen consumed by the cells, in agreement with earlier studies [14]. With insulin in the medium, virtually all (97%) of the oxygen consumed could be accounted for by glucose oxidation. This is in agreement with the conclusion that in isolated myocytes, exogenous substrate oxidation can account for essentially the total oxygen utilization [14,16].

The greater utilization of exogenous glucose by the isolated heart myocytes than by the isolated perfused hearts [1,2] may be explained by dif- ferences in their contractile activities and, there- fore, their energy demands. The limited use of exogenous glucose by the isolated hearts is attribu- table to the fact that glucose transport and phos- phorylation are rate-limiting [1], resulting in greater dependence on endogenous substrates to maintain metabolic stability while performing mechanical work. In contrast, the cardiac cells are quiescent and require much less energy, since their main expenditure is to maintain basal metabolic activi-

ties. Thus, even when glucose uptake in the ab- sence of insulin has become rate-limiting at 2.5 nmol /min per mg protein (unpublished data), the sugar is already furnishing most of the fuel required by the myocytes.

The delayed onset of insulin action on 2-de- oxyglucose uptake noted in the isolated myocytes (Fig. 1) has also been observed in the isolated adipocytes [17]. Both the length of lag phase and the time required to achieve a maximum insulin action were comparable between the two cell pre- parations. Eckel et al. [7] reported a lag in insulin action on glucose transport across the cell mem- brane as indicated by 3-O-methylglucose uptake. The response to insulin, however, was compara- tively faster than that of 2-deoxyglucose uptake. The time lag in insulin action has been attributed to a delay between insulin binding to the receptors and onset of glucose uptake involving a chemical or physical signal [7,17]. Thus, the difference in the time lag for 3-O-methylglucose and 2-deoxyglu- cose uptakes may be indicative of a graded trans- mission of the insulin signal through different cellular compartments.

The 42 54% increase in 2-deoxyglucose uptake (Fig. 3) by insulin-treated cells is comparable to the maximal stimulation of 3-O-methylglucose transport by insulin [7]. The insulin effect, how- ever, is substantially less than that reported by Haworth et al. [6]. Furthermore, 2-deoxyglucose uptake was previously determined in the presence of 5 mM acetate, which is known to be readily utilized by the myocardium and is a potent inhibi- tor of glucose uptake, even in the presence of insulin [18,19]. Stimulation of glucose utilization by insulin in these experiments (Fig. 2) is also lower than the 2-fold increase observed by Kao et al. [5], and is similar to the rate observed by Bihler et al. [20]. In a previous study [5], the rate of 3H_,O production from [2-3H]glucose in the absence of insulin (0.54- 1.31 nmol /min per mg protein) was considerably lower than in our studies (2.43 nmol /min per mg protein). The reason for this discrepancy is not clear.

The greater increase in glucose uptake than glucose hydrolysis in insulin-treated cells suggests that the excess hexose phosphate is diverted into alternate metabolic pathways, such as glycogen synthesis. This is illustrated in Table I1, which

403

TABLE II

EFFECT OF INSULIN ON GLUCOSE METABOLISM IN ISOLATED MYOCYTES

All values are expressed in nmol/min per mg protein. In parentheses are percentages of the glucose uptake value. The '% change' values are not identical with the values found in Results because they were obtained from the calculated regression lines.

Glucose Glycolytic Glucose Incorporation uptake flux oxidation into glycogen

Control 2.56 2.43 (95%) 2.23 (87%) 0.21 (8%) Insulin (10 mU/ml) 3.53 2.99 (85%) 2.57 (73%) 0.60 (25%)

% change + 38% + 23% + 15% + 186%

summarizes the effects of insulin on glucose metabolism in isolated myocytes. The combined rates of glucose incorporation into glycogen (0.60 nmol /min per mg protein, Fig. 5) and glucose flux through glycolysis (2.99 nmol /min per mg protein, Fig. 2) account for all the glucose uptake (3.53 nmol /min per mg protein, Fig. 3) by the myocytes.

The greater increase in glucose uptake than glucose hydrolysis in insulin-treated cells suggests that the excess hexose phosphate is diverted into alternate metabolic pathways, such as glycogen synthesis. This is illustrated in Table II, which summarizes the effects of insulin on glucose metabolism in isolated myocytes. The combined rates of glucose incorporation into glycogen (0.60 nmol /min per mg protein, Fig. 5) and glucose flux through glycolysis (2.99 nmol /min per mg protein, Fig. 2) account for all the glucose uptake (3.53 nmol /min per mg protein, Fig. 3) by the myocytes. This further suggests that an increase in glycogen synthesis may be the direct consequence of in- creased substrate availability. Earlier studies [2] on isolated rat-hearts have also showed that stimula- tion of glycogen synthesis by insulin did not in- volve the conversion of the enzyme glycogen syn- thetase from the D to I (inactive to active) form. Myocardial UDPglucose level was also elevated by insulin.

In the absence of insulin, there was only minimal glycogen turnover in the isolated cardial cells as indicated by the low glucose incorporation rate of 0.21 nmol /min per mg protein (Fig. 5, Table II) together with the unaltered glycogen content of 61 nmol /mg protein. These findings are also in accordance with the hypothesis that, in the iso- lated myocytes, energy generated from endoge- nous substrates is not derived from glycogen, but

primarily from the triacylglycerol pool. After a 40 min incubation without insulin, the

glycogen concentration of the heart myocytes did not change. Glycogen concentration also remained at a similar (58-65 nmol /mg protein) level throughout most of the cell isolation (unpublished observations). During and before the perfusion with collagenase, approximately one-third of the myocardial glycogen store was hydrolyzed (the in vivo glycogen concentration was 90 nmol /mg pro- tein). The lower glycogen content in the cells, however, was rapidly restored by insulin. It is important to note that despite the smaller effect of insulin on glucose utilization in the isolated heart cells than in the isolated rat hearts, the insulin effect on glycogen turnover in these cells is com- parable to that found in Langendorff perfused hearts [21]. When glycogen turnover in the isolated myocytes had attained equilibrium, the rate of glucose incorporation into glycogen remained higher with insulin present in the incubation medium (Table I). This reiterates the notion that the increased glycogen synthesis is due to in- creased substrate availability. Moreover, it is con- ceivable that the simultaneous glycogenolysis in the heart cells is also the result of accelerated substrate flow rather than a change in phosphory- lase a/b activities.

Acknowledgements

This study was supported by grants from Na- tional Institutes of Health (HL-23157) and 'American Heart Association, Louisiana, Inc. Dr. V. Chen was a post-doctoral trainee supported by National Institutes of Health National Research Service Award (HL-07098).

404

References

1 Morgan, H.E., Henderson, M.J., Regen, D.M. and Park, C.R. (1961) J. Biol. Chem. 236, 253 261

2 Chain, E.B., Mansford, K.R.L. and Opie, L.H. (1969) Bio- chem. J. 155:537 546

3 Cheung, J.Y., Conover, C., Regen, D.M., Whitfield, C.F. and Morgan, H.E. (1978) Am. J. Physiol. 234, E70 E78

4 Grosso, D.A., Frangakis, C.J., Carlson, E.C. and Bressler, R. (1977) Prep. Biochem. 7, 383-401

5 Kao, R.L., Christman, E.W., Luh, S.L., Krauhs, J.M., Tyers, G.F.O. and Williams, E.H. (1980) Arch. Biochem. Biophys. 203, 587 599

6 Haworth, R.A., Hunter, D.R. and Berkoff, H.A. (1982) Febs Lett. 141, 37 40

7 EckeL J., Pandalis, G. and Reinauer, H. (1983) Biochem. J. 212, 385-392

8 Siegel, J. and Olefsky, J.M. (1980) Biochemistry 19, 2183-2190

9 Montini, J., Bagby, G.J., Burns, A.H. and Spitzer, J.J. (1981) Am. J. Physiol. 240, H659 H663

10 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275

11 Wick, A.N., Drury, P.R., Hakeda, H.I. and Wolfe, J.B. (1957) J. Biol. Chem. 224, 963-969

12 Rovetto, M.J., Lamberton, W.F. and Neely, J.R. (1975) Circ. Res. 37, 742-751

13 Clark, M.G. (1978) Int. J. Biochem. 9, 17-18 14 Montini, J., Bagby, GJ . and Spitzer, J.J. (1981) J. Mol. ('cll.

Cardiol. 3, 903-911 15 Seifters, S., Dayton, S., Novic, B. and Muntwyler, E. (1950)

Arch. Biochem. 144~ 83 89 16 Chen, V., Baby, G.J. and Spitzer, J.J. (1983) Am. J. Physiol.

245. C46-C51 17 Ciaraldi, T.P. and O[efsky, J.M. (1979) Arch. Biochem.

Biophys. 193, 221 231 18 Randle, P.J., Newsholme, E.A. and Garland. P.B. (1970)

Biochem. J. 93, 652-665 19 Randle, P.J., England, PJ. and Denton, R.M. (1970) Bio-

chem. J. 117, 677-695 20 Bihler, I., Ho, T.K. and Sawh, P.C. (1983) Can. J. Physiol.

Pharmacol. 62, 581 588 21 Das. 1. (1973) Am. J. Physiol. 224, 7-12