Eur. J. Biochem. 98, 77-82 (1979)

Hormonal Regulation of Glucokinase in Primary Cultures of Rat Hepatocytes Christian S C H U D T Fachbereich Biologie, Universitat Konstanz (Received August 11, 1978/April 20, 1979)

Glucokinase activity in rat hepatocyte cultures declined with a half-time, tip, of 32 h during 3 days under serum-free conditions. Addition of insulin and triamcinolone to the culture medium prevented this decay. Glucokinase levels in hepatocytes derived from fasted rats could be elevated from 7.4 to 16.4 mU/mg protein in the presence of insulin and triamcinolone. In 2-day-old cultures glucokinase was induced in the presence of both hormones with a half time, tip, of 5.1 h. In cultures treated for 2 days with triamcinolone, insulin induced a 80 % increase of glucokinase even in the absence of glucocorticoids. Insulin induction was dependent on protein synthesis but occurred in the absence of RNA synthesis. Glucocorticoid action, however, depended on RNA synthesis suggesting that glucocorticoids control transcription. Insulin evoked half-maximal effects at 3 nM and dexamethasone and triamcinolone at 0.1 and 1 nM respectively. Degradation of glucokinase was initiated in 2-day-old hepatocytes after removal of triamcinolone and insulin. Protein synthesis was essential for the onset of degradation and glucagon did not affect the rate of glucokinase degradation.

Glucokinase activity is subject to complex regulation [l]. Influences of diet [2,3], insulin [3-51, glucagon [6,7] and glucocorticoids [7 - 91 have been reported and also an interdependence of insulin and glucocorticoids has been discussed [lo]. The analysis of the underlying mechanisms in vivo however, was hampered by the influence of hormones and blood glucose on each other. Hepatocytes derived from adult rats can be cultivated in serum-free media and can be exposed to defined hormone concentrations for several days [ l l - 131. Therefore, this cell system appeared to be a suitable tool to reinvestigate the hormonal regulation of glucokinase in vitro. In the present paper the influences of insulin, glucagon, glucocorticoids and glucose concentrations on glucokinase activity in hepatocytes are studied during 3 days in culture. The control of glucokinase synthesis and degradation by glucagon and insulin are analyzed in 2-day-old cultures after pretreatment with and without steroids. By using the inhibitors cycloheximide and actinomycin D the different molec~

-~~ ~

~

Enqwws. Glucokinase (EC 2.7.1.2); hexokinase (EC 2.7.1.1); fructose-I ,6-bisphosphatase (EC 3.1.3.1 1).

ular mechanisms involved in insulin and glucocorticoid regulation of glucokinase activity are approached.

MATERIALS A N D METHODS Cell Culture Isolated hepatocytes were derived by liver perfusion according to Berry and Friend [14]. Male rats, 150-250 g, which had been fed ad libitum were used normally. Fasted rats had been deprived of food for 24 h. Perfusion was performed between 9.30 and 10.30 a.m. Isolated parenchymal cells were suspended in a cold (4- 6 "C) culture medium containing salts and supplements according to Dulbecco's modification of Minimum Essential Medium, 10 mM glucose, 5 m M lactate and 5 newborn calf serum. 10 ml of the cold suspension (2.0- 3.0 x lo5 cells/ml) were inoculated into 10-cm tissue-culture dishes (Greiner, Nurtingen, F.R.G.) and incubated at 37°C and 8 % C02/92';.;, air. 110 to 180 dishes were obtained per rat. Viable parenchymal cells attached to the dishes within 1 h. 3 h after seeding the medium was removed and ex-

Glucokinase in Hepatocyte Cultures

78

changed against a serum-free medium (37 'C) containing 0.1 mg albumin/ml and the indicated hormones.

A.ssaj.s

For enzyme assays dishes were washed twice with ice-cold saline and immediately frozen on a metal block cooled by liquid N2. Cells from 10-cm dishes were scraped into 0.5 ml buffer, sonicated twice for 5 s and centrifuged 20 min at 150000 x g. Glucokinase and hexokinase activity were determined according to Pilkis [I51 using 100 mM or 0.5 mM glucose in the assay mixture. Fructose-I ,6-bisphosphatase was determined as described by Pontremoli and Melloni [16]. Protein concentration was measured fluorimetrically [17].

-

Time ( d a y s )

Fig. 1. G'luc~ohinii.sc~ oc~tjvifiesin pritizcir): c.uliuws of' liipcrtoc~~tc~s. Hepatocytes were cultivated either in hormone-free medium (control, 0- o),o r in the presence of 20 nM insulin and 0.1 pM At the first and the second day control triamcinolone (.--.) cultures werc changed to the medium containing insulin and triamcinolone (e).At the time points indicated three dishes of each condition were uashed twice with cold saline and stored at - 30 T for enzyme determination. Data are taken from five experiments S.D. and are given as mean ~

~

Cliemiculs Insulin, triamcinolone, dexamethasone and hydrocortisone were obtained from Serva (Heidelberg, F.R.G.). Albumin (Cohn fraction V) was obtained from Sigma (Munchen, F.R.G.). Glucagon was a kind gift from Eli Lilly (Bad Homburg, F.R.G.). Glucocorticoids were solubilized at 10 mM in 100 ethanol immediately before addition to the culture medium.

RESULTS Ckcrnges oj Glucok inuse A ct ivit ies iii Hepatocytcs under Various Culture Conditions Primary hepatocyte cultures contained about 20 mU/mg protein of glucokinase activity 3 h after inoculation (Fig. 1). During 3 days culture in a serumfree, synthetic medium glucokinase decreased with a half-time, t l / 2 , of 51 h (specific activity) or t l , 2 of 32 h (activity/dish) (Fig. 1). This enzyme decay could be prevented and the initial specific activity was maintained if the culture medium contained insulin and triamcinolone (Fig. 1). Addition of insulin of triamcinolone separately did not prevent the glucokinase decline (Table 1). In hepatocyte cultures derived from fasted rats, 7.4 2.4 mU/mg protein glucokinase were found. After 2 days of culture this low value was enhanced to 16.4 2.6 mU/mg protein providing both hormones had been added (Table 1). Additional supplementation of the medium with glucagon antagonized the influences of insulin and glucocorticoids on glucokinase in cultures from fed or fasted rats (Table 1). A direct influence of glucose

concentrations on glucokinase activity was not found. Table 3 shows, that 60 mM glucose did not change activity of hepatocytes of either source. Changes of fructose-l,6-bisphosphataseand hexokinase activity levels were determined under identical conditions. While hexokinase activity remained stable under all culture conditions, fructose-I ,6-bisphosphatase was found to decline to less than 50% of initial activity within 2 days culture. A significant reduction of this decay was observed only in the simultaneous presence of glucagon, insulin and triamcinolone (Table 1).

Elevation and Degradation

of' Glucokinase

Addition of insulin and triamcinolone to 2-dayold cultures induced a 2-fold net increase of glucokinase activity within 24 h (Table 2). Enzyme activity rose immediately after hormone addition if the cells had been pretreated by 0.1 pM glucocorticoid, whereas a significant delay of about 4 h was obvious in control cultures (data not shown). The final activities established in hepatocytes of either pretreatment were similar (Table 2). Kinetic analysis of the increase of excess glucokinase revealled a half-time, t ~ , ' of ~, 5.1 11. Separate addition of insulin or triamcinolone to control cultures did not significantly change glucokinase (Table 2). However, insulin alone did induce a rise of glucokinase in glucocorticoid pretreated cultures (Table 2). Glucagon alone did not change the basal activity levels, whereas the activity elevation observed with insulin and triaincinolone was re-

79

C. Schudt

Table 1. Activities of glucokinuse, ,fructo,se-I,6-bisphosphata.re und hexokinuse in heppatocyte cultures derived from f e d or 24-liJbstcd rots Enzyme activities were measured in hepatocytes 3 h and 48 h after onset of culture. Defined hormone concentrations were present in the medium from 3 h to 48 h culture. Insulin and glucagon were added at 20 nM and triamcinolone 0.1 pM. Data were taken from four experiments and are given as mean If: S.D. n gives the number of dishcs measured separately. n.d., not determined

Cells

Enzyme activity

Additions

~

~~

-

__

-

._ ~

glucokinase -~

~~~

fasted n=8

___

fed n=8

fructose-I ,6-bi\phosphatase fed n=8

liexokinnse fed n=6

40.0 16.5 18.2 15.0 16.9 22.4 26.4 n.d.

1.5 f 0.3 1.3 f 0.7 1.4 k 0.2 1.2 f 0.3 1.3 f 0.4 1.8 f 0.3 1.9 f 0.3 n.d.

mU/mg protein ~~

3h 2 days 2 days 2 days 2 days 2 days 2 days 2 days

5 serum control insulin glucagon triamcinolone insulin + triamcinolone insulin + glucagon + triamcinolone 60 m M glucose

7.4 6.3 6.3 6.7 6.4 16.4 6.6 7.0

Table 2. Induuc,tion of glucokinase in 2-day-old hepatocyte cultures Hepatocytes were cultivated in the absence or presence of 0.1 pM triaincinolone. After 2 days a glucokinase activity of 28 k 4 m u / dish was measured under both conditions. Cultures were washed once with 10 ml hormone-free cultur medium for 30 min in order to reduce the concentration of triamcinolone below 10 pM. Then fresh medium containing additions as indicated was added. Cells were harvested 20 h later by two washings with cold saline and were stored at - 30 C until enzyme activity determination. Data are taken froin five experiments ( n = 14) and are given as mean & S.D. Additions during 20 h incubation

Glucokinase activity arter preculture with ~~

~

control

~~~

& 2.4 & 1.4 f 0.9 1.3 f 0.7 f 2.6 & 1.2 1.4

.

-~

....

20.8 & 2.5 12.6 2.1 11.9 k 1.6 13.1 f 0.8 12.2 k 0.9 20.9 & 2.3 11.0 f 1.6 12.2 k 1.9

*

f 6.9 f 2.3 f 2.4 k 1.7 f 2.1 f 2.1 k 2.1

Table 3. Degradation of’glucokinase in 2-daj.-old hepparocj.ro culturcs Hepatocytes were cultivated for 2 days in the presence of in3ulin (20 nM) and triamcinolone (0.1 pM), Glucokinase activity was 40 mU/dish. The cultures were incubated with 10 ml hormone-free medium for 30 min in order to remove hormones and then frcsh medium cobtaining the additions as indicated was added ( t = 0). After the indicated time periods, four culture dishes of each condition were washed and stored for enzyme activity determination. Data are taken from three experiments and are given as mean & S.D. Additions to the medium

Glucokinase activity after incubation for ~~

-

~~

12 I1

8h

~~

triamcinolone

mUidish -~

+

mU/dish Control Insulin 20 nM Triamcinolonc 0.1 pM Glucagon 20 nM Insulin triamcinolone Insulin triamcinolone + glucagon Cyclohexiniide 1 pg,’ml Insulin triamcinolone + cycloheximide

+ + +

24.0 24.8 23.4 25.1 44.6 28.3 20.4

f 3.5 & 2.7 k 3.1 f 4.1 f 4.1 f 3.1 f 3.4

22.4 f 2.8

-

38.1 22.8 23.6 47.8 31.2 22.1

k 4.4 f 2.9 & 3.3 f 3.8 & 4.1 & 3.1

25.2 k 4.1

inarkably suppressed by glucagon (Table 2). If protein synthesis was blocked by addition of 10 pg cyclohexiinide/ml there was no influence on the basal glucokinase activity but total inhibition of the hormoneinduced rise was observed (Table 2). Glucokinase degradation was studied in hepatocytes which had been incubated for 2 days in the combined presence of insulin and triamcinolone. After removal of the hormones, glucokinase activity de-

24 I1

Insulin triamcinolone Control Glucagon 20 nM Cycloheximide 1 pg/ml Insulin + triamcinolone cycloheximide

+

*

~

~.

41.2 4.5 33.6 f 3.0 34.2 If: 4.1 40.2 k 3.4

39.6 k 2.6 30.0 f 2.1 31.4 2.9 39.3 f 2.4

39.8 I 2.1 24.8 j 1.9 23.9 i 3.6 38.4 I 3 . 2

40.8 f 3.9

41.2 _+ 4.1

40.4

t 3.6

creased (Table 3 ) . Addition of glucagon did not change the rate of this spontaneous degradation. However, if protein synthesis was blocked either in the absence or presence of insulin and triamcinolone, no decline of glucokinase activity was observed (Table 3).

Dose-Response Relationship between Horn~onrs and Increase qf‘ Glucokinase Activitj* In 2-day-old hepatocyte cultures incubated for 24 h with varying glucocorticoid concentrations and with 20 nM insulin, half-maximal effects of dexamethasone and triamcinolone were found at 0.1 and 1 nM re-

Glucokinase in Hepatocyte Cultures

80 N

60

."r. -E 40 ._ .-

40

U

3

-

5 ..

I

20

>

c

n

m

3

0

-E

m

C

..-

14 20

2.

+

1

U 2

I

u o

0

c ..x

40

-12

0 0

-11 -0 -9 -8 log [Steroid] / M

-7

-6

a 0 0

0 U 3

3 2c

C

4

1

0

' -12

I

' -10

'

I

-9 -8 log [ ~ n s u ~ i/nM ]

-11

' -7

Fig. 2. Changes of glucokinase activities in 2-day-old hepatocytes after 20 h incubation in the pre.c.ence of various concentrations of glucocorticoids and insulin. Hepatocytes were cultivated for 2 days at hormone-free conditions and contained 24 f 3.6 mU/dish ( t = 0). Then various concentrations of glucocorticoids in the presence of constant insulin concentration (A) or various insulin concentrations in the absence of presence of 0.1 pM triamcinolone (B) were added. After 20 h dishes were harvested and glucokinase activities were determined. Data are taken from three experiments and are given as mean t- S.D. (A) (I) Dexamethasone + 20 nM insulin; (0)triamcinolone + 20nM insulin; (A) hydrocortisone 20nM insulin. (B) (+) Insulin + 0.1 KM triamcinolone; (0)insulin

104 -Actinomycin

10-8

10'

D (g/mI)

Fig. 3 . Inhibition qfglucokinase inciuction and ('Hluridine incorporation by actinomycin D. Hepatocytes were cultivated for 2 days in the absence of presence of triamcinolone. Glucokinase activity was 20.2 2.5 mU/dish after both pretreatments (0).Dishes were washed with 10 ml of hormone-free medium for 30 min in order to remove triamcinolone. Then fresh medium containing actinomycin D and hormones was added ( t = 0). (0) Insulin and triamcinolone to triamcinolone-pretreated cultures; (0)insulin to triamcinolone-pretrealed culcur-es; (0) insulin and triamcinolone to control cultures. 20 h later dishes were harvested and glucokinase activity was determined. Data are taken from three experiments and are given as mean k S.D. Incorporation of [3H]uridine (4 pCi/ml) was measured for 30min in hepatocytes after 12 h incubation at the respective actinomycin D concentration. Relative inhibition was similar on control and in triamcinolone-pretreated cells (A)

+

spectively (Fig. 2 A). The maximum of glucokinase activity measured 24 h after hormone addition was identical with both substances. A significant effect of hydrocortisone was obtained at concentrations above 0.1 pM (Fig. 2A). The insulin concentration necessary for half-maximal glucose enhancement in the simultaneous presence of 1 pM triamcinolone was in the range of 3 nM. None of the applied insulin concentrations could increase glucokinase activity significantly in the absence of glucocorticoids (Fig. 2B).

served. Glucocorticoid-pretreated hepatocytes appeared to be less sensitive to actinomycin D since no change of cellular morphology was noticed up to 300 ng/ml. In these cells actinomycin D also reduced glucokinase elevation. However, even in the presence of 100 ng actinomycin D/ml, insulin alone increased glucokinase 40 % (Fig. 3). At this concentration of RNA synthesis inhibitor, [3H]uridine incorporation was blocked by 98 % (Fig. 3). Actinomycin D interfered also with the rate of protein synthesis. After 12 h preincubation of hepatocytes with 10 ng actinomycin D/ml, incorporation of ['4C]leucine into hepatocyte protein during 30min was reduced to 60'x (data not shown).

Injluence of Actinomycin D on Hormone-Induced Glucokinase Elevation

DISCUSSION

When actinomycin D at various concentrations was added together with insulin and triamcinolone to 2-day-old hepatocyte cultures, 3 ng actinomycin D/ ml were sufficient to abolish the entire activity increase of glucokinase observed in cells precultured under control conditions (Fig. 3). At actinomycin D concentrations above 10 ng/ml, glucokinase activity was decreased further but cell destruction was ob-

The regulation of glucokinase by insulin is a well known phenomenon with regard to the decrease of this enzyme in diabetes [l]. The present results show that liver cells in culture specifically synthesize glucokinase in response to insulin addition if these cells are simultaneously treated with glucocorticoids (Table 2). Specificity was demonstrated in vitro by comparison with hexokinase and fructore-1,6-bis-

C. Schudt

phosphatase, in which only the activity of the latter was hormone dependent and was synergistically increased in the combined presence of insulin, glucagon and triamcinolone (Table 1). The insulin-induced activity increase of glucokinase was apparently due to de novo synthesis. This was evident from the suppression of activity induction when protein synthesis was blocked by cyclohexiniide (Table 2) and supports earlier findings in vivo [3,8, 18,191. In contrast to studies in vivo [3,8], RNA synthesis was not found to be essential for the insulin induction of glucokinase synthesis (Fig. 3). Therefore, it appears that insulin controls cellular glucokinase activities by specific interference with protein synthesis. Glucagon inhibited glucokinase synthesis (Table 2), whereas it did not significantly affect degradation (Table 3). This suggests that glucagon antagonizes the insulin indwtion of enzyme synthesis. Half-maximal effect of insulin was observed at about 3 nM. This concentration is in the range of insulin levels in the blood [20] and is in accordance with other dose-response relationships between insulin and hepatocyte metabolism [11,13,21]. In the absence of any hormones, glucokinase activity of hepatocytes declined spontaneously (Fig. 1, Table 3). This shows that glucokinase synthesis is arrested after withdrawal of insulin and triamcinolone and only degradation proceeds. Furthermore, active degradation depended on protein synthesis (Table 3). Therefore, it must be assumed that either proteases or their regulators have to be synthesized, and that their activity was reduced in the presence of insulin and triamcinolone. Degradation of glucokinase in vivo was reported to be much faster [22]. This indicates, that other hormones may control enzyme degradation. Glucokinase synthesis was stimulated by insulin in hepatocyte cultures only in the presence of, or after previous treatment with, glucocorticoids (Table 2). Similar multihormonal effects on glucokinase regulation have been observed in vivo [lo]. In other experiments with adrenalectomized rats glucokinase levels were found to be low and they varied in a small range following fasting or feeding. Normal glucokinase levels, however, were established following injection of hydrocortisone [6,8]. The present results with actinomycin D show that glucocorticoids are involved specifically in R NA synthesis. N o enzyme synthesis could be induced in control cells if RNA synthesis was inhibited (Fig. 3). In glucocorticoid-pretreated cells, however, glucokinase activity could be enhanced by insulin in the absence of RNA synthesis. This suggests, that glucocorticoids provide the mRNA for glucokinase. This function of glucocorticoids explains the delay of glucokinase synthesis in control cells (see Results) and the insulin-induced synthesis in the absence of glucocorticoids in glucocorticoid pretreated cells (Table 2). The parallel inhibition of protein syn-

81

thesis by actinomycin D may explain the reduced rates of glucokinase synthesis in the presence of insulin (Fig. 3). Glucocorticoid function in glucokinase induction might be otherwise explained as an induction of transcription for other proteins which mediate the insulin signal such as the insulin receptor or ion transport systems. However, previous studies of glycogen metabolism showed that hepatocytes cultivated under control conditions maintained sensitivity towards insulin [13] whereas in the present study no insulin action was observed in the absence of glucocorticoids (Fig. 2 B). Hepatocytes are apparently more sensitive to synthetic than to physiological glucocorticoids (Fig. 2 A). Degradation of hydrocortisone may explain its low efficiency. The changes of actual activities of glucokinase in hepatocyte cultures may now be explained by the different influences of hormones : Insulin and glucagon regulate the rate of glucokinase synthesis provided mRNA is available. Glucocorticoids exert a permissive effect most likely by control of mRNA synthesis. Degradation is reduced in the presence of insulin and triamcinolone and is again induced by de novo synthesis after removal of these hormones. High glucose concentrations do not influence glucokinase activities in v i m (Table 1). This supports earlier assumptions [1,7], that the influence of blood glucose levels on glucokinase activity is mediated by changes of hormone concentrations. This work was supported by Deut.sche Forse~~un~.~jiemi~iri.srl~u/t. Sonderforschungsbereicli 1 38 'Biologisi,hc Gren:fIi'chet~ und .'j~ii~z(/;r&'. The author thanks Prof. Pette for helpful discussions and Dr Patti Nemeth for reading the manuscript. The excellent technical assistance of Mrs K. Tilch, Mr P. Bley and Mr A. Toledo is gratcfully acknowledged.

REFERENCES 1. Weinhouse, S. (1976) Curr. Top. Cell. Re&. 10, 1-50. 2. DiPietro, D. L. & Weinhouse, S. (1960) J . Bid. C'hcm. 235. 2542 - 2545. 3. Sharma, C., Manjeshwar, R. & Weinhouse, S. (1963) J . B i d . C'hem. 238, 3840-3845. 4. Niemeyer, H., Perez, N. & Codoceo, R. (1967) J . Biol. C'hem. 242, 860-864. 5. Ruderman, N. B., Lauris, V. & Herrerd, M. G . (1967) An7. .J. Physiol. 212, 1169-1173. . 215, 461 -476. 6. Pilkis, S. J. (1970) Biochem. B i o p h j ~ Acra, 7. Ureta, T., Radojkovic, J. & Niemeyer. H. (1970) J . Biol. Chem. 24.5, 4819-4824. 8. Sharma, C., Manjeswhar, R. & Weinhouse, S. (1964) A h . Enzyme Regul. 2, 189-200. 9. Walker, G. & Rao, S. (1964) Biochcm. J . 90, 360-368. 10. Seitz, H. J., Luth, W. & Tarnowski, W. (1978) Abstr. 12th Meet. Fed. Eur. Biochem. Soc. 2701. 11. Kletzien, R. F., Pariza, M . W., Becker, J. E., Potter, V. R. & Butcher, F. R. (1976) J . Biol. Chem. 2.5/, 3014-3020. 12. Pariza, M. W., Butcher, F. R., Kletzien. R. F., Becker, J . E. & Potter, V. R. (1976) Proc. Nut1 Acad. Sri. U . S . A . 73, 4511 -4515

82

C . Schudt : Glucokinase in Hepatocytc Cultures

Schudt, C. (1979) Eur. J . Biorhivn. in the press. Berry, M. N. Sr Friend, D. S. (1 969) J . CeN B i d . 43, 506 520. Pilkis, S.J. (1975) Mefhods Eitzyniol. 42, 31-39. Pontrenioli, S. Sr Melloni, E. (1975) Mrrliocls E ~ y n z o l .42, 354- 359. 17. Fairbanks, G., Steck, T. L. Sr Wallach, D. F. H . (1971) Bioche//ii.rtrJ;,10, 2606- 261 7. 18. Salas, M., Vinuela, E. Sr Sols, A. (1963) J . B i d . C'hem. 238, 3535-3538.

13. 14. 15. 16.

-

C. Schudt, Fachbereich Biologie, Universitit Konstanz. Poitfach 7 733, 1)-7750 Konstanz, Federal Republic of Germany

19. Niemeyer, 11.. Perez, N., Clark-Turri, L. Sr Rabajille, E. (1965) Arch. Bioclieni. Biophj..s. 109, 634- 645. 20. Dawson, C. M. Sr Hales. C. N. (1969) Bioclirri?. Biopliq..s. ACIN. 184, 287 - 298. 21. Van de Werve, G., Huc. L. Sr Hers, H.-G. (1977) Biocliiw. J . 162. 135-142. 22. Adelnian, R. C . Sr Freeman. C. (1972) E~idocri/io/ogj~,90, 1551 -1560.

Hormonal regulation of glucokinase in primary cultures of rat hepatocytes.

Eur. J. Biochem. 98, 77-82 (1979) Hormonal Regulation of Glucokinase in Primary Cultures of Rat Hepatocytes Christian S C H U D T Fachbereich Biologi...
480KB Sizes 0 Downloads 0 Views