5-Iodotubercidin and proglycosyn: a comparison of two glycogenic compounds in hepatocytes from fasted rats

ELSEVIER

Bi,x'himicaet Biophy~icaAcla 1311 (;9q6) 77-84

5-Iodotubercidin and proglycosyn: a comparison of two glycogenic compounds in hepatocytes from fasted rats Rosina

Fliickiger-lsler

~ Edelgard

Kux. Paul Walter

Departme,~t q! Bit,,-hemi,~tr~. Ve~tllianum. Ve~ul,,'tt.~t, 1. (H-4051 Basel. Swit:erland

Received28 August 1995. rcvi~ed5 December 1995;accepted 5 December 1995

Abstract 5-1odotubercidin (Itu) and proglycosyn (Pro) have similar glycogenic properties. To compare their mechanisms of action, we tested them in hepatocytes from fasted rats. We show that both compounds are similar in that they stimulated glycogen synthesis, increased the concentration of synthase a, decreased that of phospho~'lase u and Io~ered the concentration of F-2,6-P~ in the presence of glucose, lactate-pyruvate and amino acids. However. when amino acids were absent. Pro was the better stimulator of glycogenesis than Itu and in combination they elevated glycogen and synthase a concentrations ~ynergi~tieall~. Further they differ in that (I) Itu enhanced the levels of cyclic AMP whereas Pro did not; (2) Pro depressed glucose production from gluconeogenic substrates, whereas Itu stimulated this process; (3) the inhibition of F-2•6-P_, formation and glycolysis by Pro became much weaker than that by ltu when glucose concentrations were raised from I0 to 20 mM. Inhibition of glycolysis but not that of glycogen synthesis was partly due to a phosphorylated metabolite of ltu. The present study indicates that despite their simi!ar glycogenic eflects, Itu and Pro do not share a common mechanism of action. Further, the inhibition of glycolysis and :-2,6-P, formation by hu cannot be explained if it acts solely as a general inhibitor of protein kinases. Keywords:

Glycogen synthesis: Gl~.colysis:Fructose-2.6-bisphosphatc:(Rat hepattx:~te)

1. Introduction In hepatocytes from fasted rats, the adenosine kinase inhibitor 5-iodotubercidin strongly stimulates glycogen synthesis from various substrates [I]. The mechanism of this stimulation differs from the one of amino acids in that the combined action of Itu and amino acids on glycogen depostion is more than additive [I] and further, in contrast to certain amino acids [2,3], Itu increases neither the cell volume nor the concentrations of aspartate and glutamate [I]. The promotion of glycogenesis by Itu is accompanied by an increase in the concentration of glycogen synthase a and a decrease in that of active phosphorylase [1]. Very similar effects have been described for the phenacylimidazolium compound proglycosyn [4-6]. In hepatocytes from normal and streptozotocin diabetic rats, Pro and glutamine s.,nergistically stimulate the synthesis of glycogen [6]. Pro

Abbreviations:Itu, 5-iodotubercidin;Pro. proglycosyn:F-2.6-P_,. fructose-2.6-bisphosphate. • Correspondendingauthor. Fax: + 41 61 2673566. 0167-4889/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0167-4889(95 )00196-4

activates hepatic glycogen synthase and inactivates glycogen phosphorylase without causing cell swelling [5]. The apparently identical glycogenic properties of two chemically different compounds might suggest that Itu and Pro share a common mechanism of action. The only difference in metabolic effects between ltu and Pro observed so far is that Itu enhances the levels of cyclic AMP [I] whereas Pro has no significant effect [5]. As Itu has been shown to be a general inhibitor of protein kinases [7], its glycogenic action could possibly result from an inhibition of phosphorylase kinase and various synthase kinases. Although Pro. or its demethylated, glucuronidated derivative [8]. has been proposed to stimulate glycogen synthesis by an activation of protein phosphatase(s), its inhibitory effects on protein kinases were not investigated and therefore cannot be excluded [5]. In addition, the glycogenic action of both compounds cannot be explained by an effect on cyclic AMP or calcium [1,5,7]. Pro has been reported to have several glucagon-like effects on hepatic metabolism in that it decreases the concentration of F-2,6-P:, inhibits glycolysis. inhibits fatty acid synthesis and stimulates fatty acid oxidation [5]. Furthermore. Pro inhibits net glucose pro-

R. Fliickiger-lsleret al. / Biochimicaet Bioph.vsicaActa 1311 ( 19961 77-,~4

duction from gluconeogenic precursors [4,6] by diverting the gluconeogenic flux from glucose to glycogen [6]. However. the observed effects of Pro on F-2,6-~ concentrations and on glycolysis cannot be explained by the suggestion that it activates protein phosphatase(s). Until now. Pro and Itu have been investigated separately by differcnt laboratories and under different incubation conditions. To get a direct comparison, we tested the two compounds in the same system and compared their effects on glycogen metabolism, on cyclic AMP and F-2.6-P 2 levels, on glucose production and on glycolysis.

2. Materials and methods 2.1. Materials

The chemicals used are listed with their sources: enzymes and coenzymes (S'~ma. St. Louis, USA and Boehringer-Mannheim GmbH, Germany). lactate (Serva. Heidelberg. Germany). alanine, glutamine, dihydroxyacetone (Fluka. Bachs, CH), glucose, fructose, g!ycine (Merck, Basel, CH) pyruvate (Calbiochem. La Jolla, USA). 5iodombercidin (Research Biochemicals, MA. USA), albumine bovine fraction V (Miles. Zurich. CH). UDP-D-[U14C]glucose, cyclic AMP [3HI assay system (Amersham Rahn AG. Zurich, CH). Proglycosyn was a gift from Lilly Research Laboratories, Eli Lilly. Indianapolis, IN 46285. USA. All other chemicals were of analytical grade.

mined spectrophotometrically in neutralized HCIOa-extracts of the cell suspension. Glycogen was measured as glucose after acid hydrolysis [15]. Glycogen synthase a activity was measured at 25°C in the presence of 5 mM UDP-[U-laC]glncose. 1% glycogen, 4 mM EDTA. 20 mM KF, 50 mM Hepes (pH 7.8) and 10 mM Na,SO 4 [16]. Phosphorylase a activity ~as measured at 25°C in the presence of 50 mM glucose-l-phosphate. I% glycogen. 150 mM NaF and 0.5 mM caffeine [17]. For the determination of cyclic AMP. the cells were freeze-thawed in 50 mM Tri~ buffer containing 4 mM EDTA (Titriplex lll) in order to prevent enzymatic degradation of cyclic AMP [18]. and then heated to 90°C for several minutes to coagulate protein. After centrifugation, cyclic AMP was determined in the supernatant using the cyclic AMP [3H] assay system from Amersham. 2,4. Expression t~l results

A value of 1.25 × 10s cells per g liver wet weight was used for calculations [19,20]. The results are given as means + S.E. for the indicated numbers ol cell preparations. Statistical differences were tested by Student's t-test [21]. When more than two groups were compared (Fig. 2). statistical differences were determined by one-way ANOVA followed by Fisher's protected least significant difference method [21 ].

3. Results 2.2. Preparation and incubation o f hepatocytes

Hepatocytes from 18-24 h fasted male Wistar rats were prepared as described previously [9]. The viability of the cells was determined by trypan blue exclusion and was 85-95%. ( 8 - 1 5 ) × 106 cells were incubated for I h under a O~/CO2 atmosphere (95/5%) at 37°C in Krebs bicarbonate buffer containing 0.5% ( w / v ) albumin, defatted and dialyzed according to [I0]. Before addition of substrates, the cells were preincubated for 20 min in the presence or absence of Itu, Pro or both together. The inhibitors used in Fig. 5 were also preincubated with the cells. 2.3. Analytical methods

Fructose-2,6-bisphosphate was determined using pyrophosphate-dependent 6-phosphofructo-I-kinase from potato tubers as described in [11]. The enzyme was partially purified except that the last purification step of the enzyme on DEAE-cellulose was omitted. The concentrations of fructose-6-phosphate used in the assays was 20 mM for the determination of Vm~~ and I mM in the assays with varying concentrations of F-2,6-P 2. Glucose [12], lactate [13] and pyruvate [14] were deter-

3.1. Dose-response curces o f glycogen, synthase a and phosphorylase a ill the presence o f Itu and Pro

The optimal concentrations for stimulation of glycogen synthesis are reported to be 10-50/zM for Itu [I] and 100 ixM for Pro [6]. Glycogen, synthase a and phosphorylase a concentrations were measured in cells preincubated with Itu and Pro and then incubated for 60 min with 10 mM glucose, 5 mM lactate-pyruvate (9:1) and an amino acid mixture containing 5 mM of each glutamine, alanine and glycine (Fig. I). Both effectors led to a marked stimulation of glycogen deposition. Maximal stimulation by Pro was achieved between 50 and 100 ArM (32 p, mol/g), whereas higher concentrations caused a lower stimulation. With ltu. the stimulation curve levelled out between I0 and 100/zM (30-33/zmol/g, Fig. la). Both effeetors stimulated glycogen deposition to the same maximal extent but the potency of Pro was ten times less. In agreement with the proposed mechanisms of action of Itu (protein kinase inhibition [7]) and Pro (protein phosphatase stimulation [5]), the stimulation curves of glycogen synthesis correlated well with the corresponding stimulation curves of synthase a activity (Fig. lb, r2=0.92 and 0.89) and the inhibition of phosphorylase a activity (Fig. Ic, r-" = 0.97 and 0.79).

R. Fliickiger-I~h,r el ld. / Bio('lltmi,'ll el Bi,wh~ ~ica ,.~(t¢t 1311 t 1996/ 77-84 3.2. Separate a n d c o m b i n e d effec'ts a n d (3"clio A M P leuuls

tm

glycogen metabolism

The effects of I0 /zM Itu, I00 g M Pro or their combination were studied in the absence (Fig. 2a.c.e) and in the presence of amino acids (Fig. 2b.d,f). When amino acids were omitted, glycogen synthesis from gluconeogenic substrates was significantly enhanced by ltu but clearly more so by Pro (Fig. 2a). A combined addition of Itu and Pro resulted in a very pronounced synergistic effect. The stimulation of glycogen production was significantly higher upon a combined addition compared to the calculated sums of glycogen after a separate addition of h a and Pro (36.4 versus 27.0, glucose and lactate-pyruvate and 44.7 versus 32.4. dihydroxyacetone: P < 0.05, horizontal lines). A similar pattern was observed for glycogen synthase a (Fig. 2c). Although the stimulation of Pro was not larger than that of ltu, their combination led again to a more than additive effect. The activity of phosphorylase a (Fig. 2e) was significantly inhibited by h u but stronger by Pro and further decreased by their combination. Addition of amino acids to the gluconeogenic substrates led to a significant increase of the values of glycogen (Fig. 2b) and synthase a (Fig. 2d) in the absence and presence of either ltu or Pro. The effect of Pro, however, was equal to that of Itu and a combined addition no longer resulted in a synergistic stimulation. Further, the concentrations of glycogen and synthase a attained with Ira plus Pro in the presence and absence of amino acids ,,,,'ere the same, indicating that maximal concentrations of glycogen and synthase a within I h were achieved. The increase of phosphorylase a activity (Fig. 2f) induced by amino acids was clearly suppressed by both agents. Despite their similar effects there was a difference between Itu and Pro: in apparent contradiction of its glycogenic action, Itu significantly raised the levels of cyclic A M P from 0.25 +_ 0.01 to 0.34 +_ 0.03 n m o l / g (glucose and lactate-pyruvate, n = 4) and from 0.29 +_ 0.03 to 0.44 +_ 0.03 n m o l / g (glucose. lactate-pyruvate and amino acids, n = 4), whereas Pro did not change them. After their combined addition, the cyclic A M P concentrations remained as high as observed with Itu alone.

40 I

(a~

o 10

20

30

40

5°

[odotubo mld~n

Effector (~M)

~o

o 100

2o0

300

400

500

Prog~ycosyn

lo

20

30

40

50

lodOtubercldin

400 40

500 50

Eftector (~M)

ld(~ T (C)

20

100 10

200 20

300 3o

Proglycosyn Iodotuberc~aln

Effector (~M)

3.3. D o s e - d e p e n d e n t pilate concentrations

inhibition

o f ./'ructo.~e-2.6-bi.wlu~s-

The dose-dependent effects of hu and Pro on the concentrations of F-2.6-P, are shown in Fig. 3. In the presence of 10 mM glucose, lactate, pyruvate and amino acids, increasing concentrations of Itu and Pro resulted in a dramatic decrease of F-2.6-P, formation, h u was again more than 10 times more potent than Pro but at the highest effector concentrations used (50 and 500 /,tM, respectively), the decrease in the concentration F-2,6-P, was equally strong with Itu and Pro. Such a steep drop could indicate loss of cell viability. However. concomitant with

Fig. I. Con,.:entration-dependenteffects of hu and Pro on glycogen s,,nthesis, s',ntha~,e a and phosphorytase ,~t.Hepatocytes ',,.ere incubated for i h v.ith 10 mM glucose. 5 mM lactate-pyruvate(9:i) and an amino acid mixture containing 5 mM of each glutamine, alanine and glycine: ( © ). in the ~hsence of effector: preincubated with hu (O) or Pro ( • ) for 20 rain. Values are the means+S.E~ for 4-8 different experiments. ' " P < 0.01 '.er,.us control vaiues. ~ ~ P < O.OI for ;'alues in the presence of ltu '.or,u, values in the presence of corresponding concentrations of Prt~ a~, determined by unpaired Student's t-test.

the strong decrease in the concentrations of F-2.6-P_,. there was -fficient glycogen synthesis (Fig. la). Within the whole concentration range itu and with Pro up to 10(3 ,u,M

R. I"liicki~c,l-I';h'r et al. / Biochimic~l el Biophysica Actd 1311 ( 19961 77-84

8O

substrate (Table 1). G l y c o l y s i s . measured by the formation

there was a good correlation between I h e decreases o f F-2.6-P, levels and the corresponding increases o f glycogen ( r "~= 0.98 and 0.90. respectively).

3.4. Efh'c'ts ~[ liu a n d Pro

~,ll

o f lactate plus pyruvate, was inhibited with I0 /~M ltu by 85% and this was paralelled by a 71% dect'ease o f F-2.6-P_, concentrations. A higher concemration o f Itu further decreased g l y c o l y s i s but not F-2,6-P,, formation, The inhibition o f F-2.6-P 2 formation by 100 or 500 /zM Pro in the presence o f glucose was only 30% or 50%, respectively and thus much weaker than the one observed with the substrate mixture shown in Fig. 3 (84% and 93% inhibi-

glycolysis

Since F-2.6-P, is an important positive effector o f the glycolytic enzyme 6-phosphofructo- I-kinase [21]. Itu and Pro were tested on g l y c o l y s i s using 20 m M glucose as

l [] Control

D

G,LP

B

(~HA

1200

G.L.P

C

3oo

c

...._L_

A

U

G.L.P

"

~

"

G,L,P

(e)

1o

DHA

C

c

~

"

,,

"

(f

A

8

10

B

8 s

DHA Substrate

"

•

n ~.

G.L,P

~

'

0

n

B

G,L,P

DHA Substrate

Fig. 2. Effects of I0 /zM hu and 100 #M Pro in the presence or absence of amino acids. Hepatocytes were incubated for I h with 10 mM glucose and 5 mM lactate-pyruvate (9:1) (G.I,.P. n = 7) or with I0 mM dihydroxyacetone (DHA, n = 5) either in the absence (a.c,e) or in the presence of an amino acid mixture containing 5 mM of each glutamine, alanine and glycine (b.d.f). Concentrations at the end of preincubation were 0.5 + 0.1 # m o l / g (glycogen), 173 + 18 mU/g (synthase) and 10.3 _+0.9 U / g (phosphorylase). The horizontal lines indicate the calculated sums of the concentrations of glycogen (a.b) and synthase a (c.d) after a separate addition of ltu and Pro. Statistical analysis: Columns with different index letters differed significantly (P < 0.05) within a substrate group as determined by least significant difference test; " P < 0.05 and " " P < 0.01 for values with amino acids versus corresponding values without amino acids as determined by paired Student's t-test.

R. Fliilkiger.l~ler et al. / Bio~himi~a et Biophx ~il,t Acnl 1311 I I996t 77-84

,06 80

so

,so 4

°.

i ~o o 0

100 10

200 20

300 30

400 40

500 50

proglycosyn IodotubercLdlr

Effector (pM)

Fig. 3. Dose-dependent efl~.cts of hu and Pro on F-2.b-P, formation. Substrates were 10 mM glucose, 5 mM lacuue-pyruvatc (9:1) and amino acids (5 mM of each glutamine, alanine and glycme). Incubation time was I h. (O). v,ithout efl~ctor: (Q), preincubated ~ith I0 .aM ha: t • ). preincubated with 100 /.tM Pro; 1•). preincubated ~ith hu plus Fro. Values are the means_+S.E, for 4--6 cell preparations. Unpaired t-test results: " P < 0.05 and " " P < 0.01 ~ersus control value~ wilhout efl~zctor.

tion). As a result, g l y c o l y s i s was inhibited only by 2 5 - 2 9 c k . A c o m b i n e d addition o f Itu and Pro led to an additional s i g n i f i c a n t inhibition o f g l y c o l y s i s and F-2,6-P, formation. The weak inhibition o f g l y c o l y s i s by Pro, however, was o b s e r v e d only at 20 m M or higher glucose concentrations (Fig. 4). In the presence o f I 0 m M glucose. Pro inhibited lactate plus pyruvate formation by 79% and ltu by 87%. 3.5. Effect o f I m a n d Pro on g h w o s e production Since Itu strongly decreased the concentration o f F-2,6P,, we i n v e s t i g a t e d whether the c o m p o u n d had an effect on g l u c o n e o g e n e s i s (Table 2). G l u c o s e production from lactate-pyruvate, d i h y d r o x y a c e t o n e or fructose was stirnu-

- -

10

20 30 Glucose (raM)

40

Fig. 4 Gluct~c concentration curve in the presence of Itu and Pro.

Hepatoc~te, ~'.ere incubated for I h v,,ithdifferent glucose concentrations. IO). control: I l l . preincubated with 10 p.M hu: ( • I. preincubate.J 'xith ItXI ,u.M Prt~. Values are the means _+S.E. for 5-7 L:ell preparations. Unpaired t-te,t results: " P < 0.05 and " " P < 0.OI versus control values '.~ithout effector: ~ " P < 0.01 for values in the presence of hu versus ~alue~ in the presence of corresponding concentrations of Pro.

lated by I 0 /xM Itu as reflected by the positive differences in glucose between control and hu-treated cells. 100 /.tM Pro. on the other hand. i n h i b i t e d net glucose production from all g l u c o n e o g e n i c substrates tested. A comparison o f the difference between the concentrations o f glucose in control and effector-treated cells to that o f g l y c o g e n revealed that ltu stimulated both glucose and g l y c o g e n production, whereas Pro stimulated only g l y c o g e n synthesis. although more efficiently than Itu. A g o o d correlation between the decreases in glucose and increases in glycogen concentrations due to Pro ( r -~= 0.93) confirms the o b s e r v a t i o n o f G u o et al. [6] that Pro diverts the gluconeogenic flux into g l y c o g e n . G l u c o s e production in the presence o f ltu p l u s Pro was e v e n further decreased than with Pro alone, T h i s further decrease in glucose production due to a c o m b i n a t i o n o f ltu and Pro seemed to account for the additional increase in g l y c o g e n concentrations, since

Table I Effect of hu and Pro on glycolysis and the concentration of fructose-2,6-bisphospate Additions Glycolytic substrate (mM)

Effizctor (.u,M)

Lactate plus pyruvate ( .u,mol/g)

Fructose-2.6-bis phosphate (nmol/g)

Glucose (20)

Itu (10) ltu (50) Pro t IOn1 Pro (500) Itu + Pro(lO+ IO0) ltu + Pro 150 + 500)

54.1 + 1.9 (8) 7.9 ~. !. 1 (8) 5.5 _ 0.7 (8I 40.5 _+ 1.9 (8) 38.3 _.+.2.6 (8) 3.4+0.4(8) 2.3 _+0.3 (8)

20.5 "t"2.0 (5) 6.0 _+0.4 (5) 6.0 + 0.4 (5) 14.3 _+ 1.6 (5) 10.2 + 1.2 (5) 5.2_+0.3(5) 4.7 _+0.3 (5)

"" "" ' " "" " ": " " ~#

"" "" "' " " .zz " " ~#

Cells were preincubated for 20 rain with the effectors before substrates were added. Incubation time was 60 rain. Endogenous lactate plus pyruvate formation ',,.,as 1.08+0.11 in the absence and 1.11 +0.16 and 1.12+O.43 # m o l / g in the presence of Itu and Pro. respectively. Endogenous concentrations of fructose-2.6-biphosphate were 0.46 ___0.08 in the absence and 0.19 _+0.03 and 0.12 + 0.05 nmol/g in the presence of Itu and Pro. respectivel~ The numbers of experiments are indicated in brackets. Unpaired t-test results: ' P < 0.05 and ' " P < 0.01 versus control values without effeetor; ° P<0.01 for values with 10.u.M ltu plus 100/zM versus values with 10.uMltu alone; ## P
R. Flilckiger-I,~h,rel al. / Biochimica el Biophy.~icaActa 1311 ~19961 77-c'¢4

82

the total amount of net gluct)sc and glycogen production by Itu plus Pro was the same as c o m p a r e d to ltu alone, Interestingly. Itu alone inhibited glucose produt, tion and stimulated glycogen synthesis to the same extent as Pro when the gluconeogenic snbstrates were supplemented with a m i n o acids, resulting in a correlation coefficient equal to that of Pro (not shown).

and the inhibition was almost reversed in the presence o f A D A (Fig. 5b). "[he stimulation o f glycogenesis by Itu. on ihe other hand. was not reduced, indicating that glycogen m e t a b o l i s m m a y be less affected by a phosphorylatea ltu product than glycolysis. A D A did not influence the effect of Pro on glycogen synthesis.

3.6. A phosphot3"lated metabolite is partly respot:sible fiJr the ~ffects o f Itu

4. Discussion

It has been postulated that the effects o f Pro m a y be mediated by a metabolite [5] and Van Shaftingen and De H o f f m a n n proposed glucuronidated derivative(s) to be the active molecules o f Pro [8]. ltu, on the other hand, is not only an inhibitor but also a substrate for adenosine kinase [22] and we tested whether a phosphorylated product could be responsible for it,, effects on g l y c o g e n synthesis and glycolysis. With this aim in view. two inhibitors. Y-amino5'deoxyadenosine ( A D A ) , another inhibitor o f adenosine kinase. [23,24] and 6 - m e t h y l m e r c a p t o p u r i n e riboside ( M e M P R ) , were used. M e M P R . an inhibitor o f purine [25.26] and g l y c o g e n synthesis [27]. was chosen for c o m parison because it exerts its effects only after phosphorylation by adenosine kinase [25,26]. Fig. 5a shows that M e M P R inhibited glycolysis from 20 m M glucose to the s a m e extent as Itu. This inhibition could be cancelled by blocking the phosphorylation o f M e M P R with A D A . T h e inhibition o f lactate plus pyruvate formation by Itu was also reduced from 80% to 52-~ in the presence o f A D A . indicating that Itu acts partly via a phosphorylated product. The decrease o f glyco~ysis in the pesence o f Pro and A D A c o m p a r e d to Pro alone could be ascribed to the slight inihibiting effect o f A D A under control conditions. G l y c o gen production from glucose, lactate, pyruvate and a m i n o acids was strongly depressed by the addition o f M e M P R

It was considered earlier that the effects of Itu and Pro on g l y c o g e n m e t a b o l i s m m a y be based on a c o m m o n m e c h a n i s m o f action [I]. The two c o m p o u n d s were s h o w n by different laboratories to stimulate glycogen synthesis in a m a n n e r that is different from the one o f a m i n o acids [I .4-6]. T h e i r direct c o m p a r i s o n in this study r e v e a l e d that Pro. with a 10 times lower potency (Fig. 1), is the better stimulator of glycogenesis from glucose and gluconeognic substrates in the absence (Fig. 2. Table 2) but not in the presence o f a m i n o acids (Fig. 2). Pro also leads to a stronger decrease of phosphorylase a concentration than Itu. provided that a m i n o acids are absent (Fig. 2). W h e n Itu and Pro are added together, their stimulatory effect on g l y c o g e n production and synthase a concentration is m o r e than additive in the absence o f a m i n o acids. This additive stimulation is concealed w h e n a m i n o acids are present p r e s u m a b l y because o f an additional s y n e r g i s m o f a m i n o acids and the combination o f itu plus Pro: the concentrations o f g l y c o g e n and synthase a attained with either Itu or Pro are further increased by the presence o f a m i n o acids. But w h e n itu and Pro are added together, g l y c o g e n and synthase a concentrations are the s a m e in the absence and presence of a m i n o acids, indicating that m a x i m a l levels are attained. Cyclic A M P levels are increased by Itu and Itu plus Pro but not by Pro alone. Since g l y c o g e n production

Table 2 Glucose and glycogen production different gluconegenic sub,;trates in the presence and absence of Itu and Pro Additions Glucoeneogenic (mM)

Effector (,uM)

Glucose (p_mol/g)

Lactate-pyru;'ate (4.5-0.5)

Itu (10) Pro (100) Itu + Pro Itu (10) Pro (1011) Itu + Pro Itu 1101 Pro (100) hu + Pro

49.7 + 2.1 0.9 ± 0.1 58.4±2.5 2.9+0.4 40.9 + 1.0 12.2± 1.0 33.3 5:2.2 26.7 ± 3.4 102.3 ± 6.3 3.5 ± 0.7 111.7±6.7 13.0+2.2 87.8+3.8 21.05:1.1 82.0 5:3.6 43.1 5:3.4 10/,.0 ± 3.2 5.8 ± 0.5 110.8±2.9 16.2+ 1.2 92.2 5:2.5 17,1 _+ 1.2 89.0 ± 2.2 38.3 ± 3.8

Dihydroxyacetone (10)

Fructose 15)

Gbcogen (/J.mol of glucose/g)

...1Gluco~'~ (p.mol/g)

...IGlycogen ( gmol of glucose/g)

'. Glucose + glycogen I ,u,mol of glucose/g)

+ 8.7 + 1.4 " " - 8 . 8 + 1.6 " " -16.4:t:2.7 " . 2 c

+ 2.0 + 0.4 " " +11.3+0.9 " " +25.8__.4.3 ' - o o

10.7 + 1.6 " " +2.5:t: 1.8 +9.45:5.8 " "

+9.4+2.2 " -14.5__.2.88 "" -20.3 5:3.2 ' .';C

+9.5+ 1.9 *" +17.5+1.0 "" +39.6-t-3.4 " .OC

+18.9+3.1 "" +3.0±2.7 +19.35:3.6 " . 0 0

+10.8± 1.7 " " -7.8 + U.6 " " -10.1±1.0 ..C

+10.4±2,4 " " +11.35:1.0 " " +32.5+3.6 .OC

+21.2±2.0 " " +3.55:1.2 +22.4±3.8 -.CO

Hepat(x:yles were preincubated for 20 min with the effectors before substrates were added. Incubation time ',,.'as 60 min. Glucose and glycogen concentrations after preincubation were 5.3.4- 0.5 ar.d 0.4 _4:.0.1 /zmol/g, respecti,.,ely. Values are the uneans + S.E. of 7 experiments. Paired t-test result:.;: • P < 0,05 and " " P < 0.01 versus control values in the absence of ellector; o p < 0.05 and o ~ p < 0.01 for values with Itu + Pro versus values with Pro,

R. l"liickit, e r - l d , ' r ,'t al. / Bi,,,'himi, ,~ ,.t I~tt,l,ht ~l, u 4, t,, 1311 I 19961 77-,'¢4

.

.

.

no

. FT

Etfector

so b)

no

~.

'

.

~

.

Enector

Fig. 5. Effects of inhibitionof the phosph(~r.,,lation~1 hu and 6-mcth'.lmercaptopurine riboside by 5'-amino-5'-deox3aden~,sine.Ct,ntr~,l xzlluc~ and the effects of 20 #M 6-methyh'nercaptopurincriho,ide (Me~',IPR). It) /,tM ltu and 100 p.M Pro on glycolysir,la) and g',lcL-gcn,,.,.nthcsi~(hi arc shown in the presence iv,hire bars) and absence (~re.~ har~) of the adenosine kinase inhibitor 5'-arnino-5'deox~adconsinc(ADA). The ~ubstrafes were 20 mM glucose (glycolysis) and 10 mM g[ucose. 4..5 mM lactate. 0.5 mM pyruvate and 5 mM of each glutamine, alanine and glycine (glycogen synthesis).Incubationtime ~a~ 1 h. The ~alues are the means_+S.E, for 6-7 experiments. Statistical anal~,is: column~ with different index letters differed significantly(P < 0.()5) a,, determined b) least significant difference test: " P < 0.05 and " " P < t).OI for xaluc,, with ADA versus corresponding valuc~,with~,ut ADA as determined b~ paired Student's t-test.

is accelerated by a combination of ltu and Pro, the glycogenie action of both effectors seems to be independent of changes in cyclic AMP concentrations. In the presence of I0 mM glucose, lactate-pyruvate and amino acids, ltu artd Pro decrease the concentration of F-2,6-P~ equally strong (Fig. 3). whereas with 20 mM glucose as substrate (Table I). Pro inhibits F-2.6-P 2 formation to a much lesser extent than Itu. Since a drop in F-2,6-P:, a positive allosteric effector of 6-phosphofmctokinase !, results in an inhibition of glycolysis [28], the subsequent decrease of lactate plus pyruvate accumulation in the presence of Pro is correspondingly smaller than in the presence of Itu (Table I). When Itu and Pro are added together, inhibition of glycolysis and of F-2,6-P, l'ormatiot~ is significantly more pronounced than with Itu alone. The 30% inhibition of glycolysis by 100 /zM Pro in the

83

presettce ttf 2 0 mM glucose is markedly weaker than the ,.)ne ~b,,c~xcd x~ith 10 mM (79%. Fig. 3) or reported by Guo ctal. [6] (83% inhibition of lactate formation). Obviously. ,qtl~_t~c can o',ercome the inhibiting effect of Pro on glycol.~,is better than that of hu. The experiments with 5'-amino-5'-deoxyadenosine (ADA). un~thcr inhibitor of adenosine kinase [23.24] show that a phosphorylated product of hu is. at least in part. responsible for the inhibition of glyeolysis by ltu (Fig. 5a). The lack of sensitivity of glycogen metabolism to an inhibition of the phosphorylation of Itu (Fig. 5b) is in agreement with the results of Massilon et al. [7] which showed that hu directly inhibits protein kinases involved in the inactixation of phospborylase and activation of synthase. Unlike Pro ~hich inhibits net glucose synthesis by directing the gluconeogenic flux away from glucose and into glycogen (Table 2). [4-6], Itu stimulates net glucose production from gluconeogenic precursors and thus increases the o~erall rate of glucose (free glucose plus glycogen) lbrmation (Table 2). One may therefore conelude that ltu affects, besides glycogen synthase, also the glucose-6-phosphate-glucokinase cycle, ltu has been reported to inhibit glucokinase of liver cytosol by 37% [7]. When llu is added together with Pro it not only supports thc stintuhlting effect of Pro on glycogenesis but also the inhibiting effect on net glucose production (Table 2). These rcsutts indicate that Pro redirects the enhanced production of glucose caused by ltu further into glycogen and therefore increases the glycogenic effect of hu. This study shows that hu and Pro have many similarities although they have no chemical analogy. They have almost identical effects on glycogen metabolism. However, since their combined effect on glycogen production is synergistic, it is unlikely that a common mechanism is involved. Further, Itu. but not Pro. raises t.he levels of cyclic AMP and stimulates net glucose production, whereas Pro apparently inhibits this process. On the other hand, they both reduce the concentration of F-2.6-P_, and inhibit glycolysis. The strong inhibition of F-2,6-P2 formation by ltu and Pro (Figs. 3 and 4) with a simultaneous stimulation of glycogenesis (Fig. In) seems to be inconsistent with the mechanisms of action proposed for the two compounds: Itu was identified as a general inhibitor of protein kinases. including cyclic AMP-dependent protein kinase [7] and Pro was found to stimulate protein phosphatase 2A [5]. Such properties should result in an increase rather than the observed decrease in the concentration of F-2.6-P, since the only protein kinase known to catalyze phosphorylation of hepatic 6-phosphofructose-2-kinase/F-2.6-bisphosphatase (PFK.,/F-2,6P_,ase) is cyclic AMP-dependent protein kinase and dephosphorylation of the enzyme is catalyzed primarily by protein phosphatase 2A [28]. One may theretbre suggest that the two compounds regulate the concentration of F-2.6-P 2 by a different mechanism than that of glycogen. "[his could apply to Itu which stimulates

R. FI i 'k ~,,,r-lsler el aL / B m c h i m i c a et Biophvsica Acta I "~1 ' ~1996 7 7 - 8 4

g l y c o g e n production directly whereas it inhibits glycolysis. at least in part, via a phosphorylated metabolite (Fig. 5). A further possibility would be that the inhibition o f F-2,6-P, formation by Itu and Pro occurs via direct inhibition o f P F K , . A n d again, in the case o f Itu with its k n o w n inhibitory effect on kinases other than protein k i n a s e s (adenosine kinase [23.29.30]. g l u c o k i n a s e [7]). an inhibition o f P F K , would be z: reasonable explanation. Exactly h o w Itu and Pro or their raetabolites which, in apparent contradiction to their g l y c o g e n i c action, inhibit F-2,6-P 2 formation r e m a i n s to be established. An additional c o m p a r ison o f Itu and Pro with purified e n z y m e s , e.g.. P F K : , protein kinases and, in v i e w o f the effects o f Pro on lipid m e t a b o l i s m [5]. a c e t y I - C o A carboxylase, w o u l d be worthwile for further studies o f the interrelations i n v o l v e d in liver cell functions.

Acknowledgements W e thank Irene M e y h a c k for the preparation o f potato tuber p y r o p h o s p h a t e - d e p e n d e n t 6 - p h o s p h o f r u c t o - l - k i n a s e and the m e a s u r e m e n t s o f F-2,6-P,.

References [I] FliJckiger-lsler. R.E. and Walter. p. (1993) Biochem. J. 292.85-912. [2] Baquet A.. Hue. L.. Van Woerkom. G.M. and Plomp, J.A.M. (1990) J. Biol. Chem. 265. 955-959, [3] Plomp. P.J.A.M., Boon. L., Caro, L.H.P.. Van Woerkom, G.M. and Meijer. A.J. (1990)J. Biol. Chem. 191. 237-243. [4] Harris. R.A.. Yamanouchi. K.. Roach. P.J.. Dominianni. S.J. and Stephens. T.W (1989)J. Biol. Chem. 264, 14674-14680. [5] Yamanouchi. K.. Stephens, T.W.. Chikada. K., Dominianni. S.J., Behforouz. H.. Scislowski. P.. DePaoli-Roach, A., AIImann. D.W. and Harris. R,A. (1992) Arch. Biochem. Biophys. 294. 609-615. [6] Guo, Z.K.. Wals. P.A. and Katz, J (1991} J. Biol. Chem. 266, 22323--22327. [7] Massilon. D.. Stalmans. W.. Van de Werve, G. and Bollen, M. (1994) Biochem. J. 299, 123-128.

[8] Van Shaftingen. E. and De Hoffmann. E. (1993) Eur. J. ~3iochem. 218. 745-751. [9] Nyfeler. F.. Fasel, P. and Walter. P. (1981) Biochim. Biophys. Acta 675. 17-23. [lO1 Chen, R.F. (1967) J. Biol. Chem. 242. 173-181. [I IJ Van Schaftingen. E.. Lederer. B., Bartrons. R. and Hers. H.G. ((1982} Eur. J. Biocehm. 129, 191-195. [12] Kunst. A. Draeger. B. and Ziegenhorn. J. (1985} in Methods of Enzymatic Analysis. 3rd edn. (Bergmeyer, H.U., ed.). pp. 163-172. Verlag Chemie, Weinheim. [13] Gutmann, 1. and Wahlefeld. A.W. (1974} in Methoden der enzymatischen Analyse, 3rd edn. (Bergmeyer, H.U., ed.}, pp. 1511-1514. Verlag Chemie. Weinheim. [14] Lamprecht. W. and Heinz. F. (1985) in Methods of Enzymatic Analysis. 3rd edn. (Bergmeyer. H.U.. ed.}. pp. 570-577. Verlag Chemie. Weinheim. [15] Walaas. O. and Walaas. E. (1950) J. Biol. Chem. 187. 769-776. [16] Doper~, F.. Vanstapel. F. and Stalmans. W. (1981}} Eur. J. Biochem. 104, 137-146. [17] Hue, L.. Bontemps, F. and Hers. H.G. (1975) Biochem. J. 152. 105-114. [18] Cheung, W.Y. (1970J in Role of Cyclic AMP in Cell Function (Greengard. P. and Costa. E., eds.), pp. 52-65. Raven Press. New York. [19] Weibel. E.R.. Stfiubli. W.. Gnfigi, H.R. and Hess. F. (1969) J. Cell Biol. 42.68-92. [20] Whealley. D.N. (1972) Exp. Cell Res. 74. 455-362. [21] Snedecor. G.W. and Cochran. W.G. (1980} in Statistical Methods, 7th edn, pp. 215-237. Iowa University Press. Ames, IA. [22] Harley. M.C.. Lin. B. and Fox. I.H. (1985) J. Biol. Chem. 260. 15765-1568 I. [23] Bontemps. F.. Van den Berghe. G. and Hers. H.G. (1983) Proc. Natl. Acad. Sci. USA 80, 2829-2833. [24] Newby, A.C.. Holmquist. C.A., Illingworth. J. and Pearson, J.D. (1983) Biochem. J. 214. 317-323. [25] Crabtree. G.. Scholar. E.. Chu. S.-H. and Parks. R.E. (1973) Biochem. Pharmacol. 22. 155-162. [26] Stet. E.H.. De Abreu. A., BiSkkerink. J.P.M., Blom, H.J., Lambooy. H.J., Vogels-Menmink. T.M.. Graaf-Hess. A.C., Van Raay-Selten, B. and Trijbels, F.J. (1994} Biochem. J. 304. 163-168. [27] Isler, R. and Walter, P. (1985) Experientia 41. D783. [28] Pilkis, S.J. and Claus. T.H. (1991) in Annu. Rev. Nutr. II (Olson, R.E.. ed.). pp. 465-515, Annual Rev., Palo Alto. CA. [29] Newby, A.C., Holmquist. C.A.. lllingworlh, J.D. and Pearson. J.D. (1983) Biochem. J. 214, 317-323. [30] Davies. L.P., Baird-Lambert, J. and Marwood. J.F. (1986) Biochem. Pharmacol. 35. 3021-3029.