Biochimica el Biophysica Acta, 1095 ( lqgl ) 165- 168 ~ 1901 Elsevier Science Publishers B.V. All rights reserved I)lf'7-488t)/UI/5f)3.5{I ADONIS 1)16748899[(}0277A

161

BBAMCR 13037

Vasopressin stimulates pyruvate utilization through a Ca 2+-dependent mechanism and lactate formation by a prot:-in kinase C-dependent mechanism in isolated rat hepatocytes Kevin M. Conricode ~ and Raymond S. Ochs Dt7mrtment of Nutrition, ('ase Western [~t'sert~¢"Unicer~ilv Mr. ),'iraqiMedical ('emcr, ('lereland, 01t tU..*i.A.t (Received If, April ltJgl) (Revised manuscript received 15 .hdy Iraqi)

Key words: Vasnpressin: Epidermal grt~wlh factor: Protein kinasc ('; Fatty acid oxidation; (Rat hepatocyte)

Vasopressin stimulates lactate production by hepatocytes from fed rats, an effect which has been attributed exclusively to Ca 2+ activation of glycogenolysis. We provide evidence here for two further actions of vasopressin which affect lactate formation by rat hepatocytes, in the presence of 50 mM glucose, vasopressin inhibited lactate production by hepatocytes. The inhibition was relieved by the presence of a.cyano-.4-hydroxycinnamate (or-CHC), which blocks mitochondrial pyruvate transport. This suggests that vasopressin stimulates pyruvate utilization in the presence of a high concentration of glucose. Epidermal growth factor (EGF), which also increases lactate formation by hepatocytes, did not similarly decrease lactate accumulation in the presence of high glucose, suggesting no stimulation of lactate and pyruvate utilization by this hormone. In ceils depleted of Ca z*, vasopressin also stimulated lactate formation. Although vasopressin did not cause the apparent translocation of protein 10nase C between cell spaces, phospholipase C treatment of hepatocytes did duplicate vasopressin stimulation of lactate formation, provided fatty acid oxidation was suppressed by the simultaneous presence of the inhibitor palmixorate. We conclude that three actions of vasopressin affect lactate and pyruvate formation: the calcium-linked activations of glycogenolysis and mitoehondrial py~vate utilization, and a stimulation of glycolysis likely mediated by protein kinase C.

Introduction

Vasopressin and other Ca2+-mobilizing hormones, acting on isolated hepatoeytes, stimulate the hydrolysis of pbosphatidylinositol 4,5-bisphosphate (PIPz). This generates two intracellular messengers: inositol 1,4,5trisphosphate (IP3) and 1,2-diacylglycerol (DAG) [1]. IP3 releases Ca2+ from an internal cellular pool [2-4] and DAG is an activator of protein kinase C [5].

* Present address: Department of Physiology, Vanderbih University, Nashville, TN, U,S.A. Abbreviations: PIP2, phosphatidylinositol 4,5-bisphosphate; IP3, inositol 1,4,5-trisphosphatc; DAG, 1,2.diacylglycerol; TPA, 12-O-tetrade¢anoylphorbol-13-acetate; a-CHC, ot-cyano.4-hydroxycinnamate: EGF, epidermal growth factor; ATP, adenosine 5'qriphosphate; PMSF, phenylmethylsuifonylfluoride. Correspondence: R.5. Ochs, Depar:ment o[ Nutrilion. Mr. Sinai Medical Center, 1 Mr. Sinai Drive, Case Western Resewve University, Cleveland, OH 44106-4198, U.S.A.

Recent evidence s,~.gests that phosphatidyIcholine, rather than PIP,, i ~e major precursor of hormonegenerated DAG in ,tepatocytes [6,7]. Receptor activation thus triggers two separate signal transduction pathways. Hormonally induced increases in hepatocyte calcium have been linked with stimulation of glycogenolysisand glycogen phor.phorylase [8], a-ketoglutarate dehydrogendse [9], pyruvate dehydrogenase [10] and of fatty acid oxidation to CO_, [11]. Cellular effects which have been suggested to be mediated throagh protein kinase C include inactivation of glycogen synthase [12,13], activation of fatty acid synthesis and acetyI-CoA carboxylase [14] and inhibition of earnitine palmitoyltransferase [15]. Vasopressin is also known to stimulate giycolysis in hepatocytes isolated from fed rats [16,17]. This effect is believed to result from stimulation of glycogen breakdown [18], a CaZ+-mediated event. A possible role of protein kinase C in the vasopressin-induced stimulation of glycolysis has not been investigated.

162 Previous studies from our laboratory have demonstrated that calcium activation of a mitochondr[al enzyme, a-ketoglutarate dehydrogenase, is involved in calcium-dependent hormone stimulation of gluconeogenesis and ureogenesis by hepatocytes from fasted rats [9], later confirmed for perfused rat liver by others [19]. We provide evidence in the present report for a calcium dependent mitochondrial action on lactate formation by hepatocytes from fed rats. A recent study in our laboratory revealed that TPA, the activator of protein kinase C, stimulates lactate production in hepatocytes [20]. This suggests that protein kinase C may play a role in the stimulation of glycolysis by vasopressin. In the present study, we provide evidence that supports this conclusion. Materials and Methods

Materials [8-Arginine]vasopressin, TPA, a-cyano-4-hydroxycinnamic acid, phosphatidylserine, phospholipase C (type Xlll from B. cereus) and enzymes for metabolite measurements were from Sigma. EGF was from Boehringer-Mannheim. Palmoxirate (sodium 2-tetradecyl-oxiranecarboxylate dihydrate, MEN-3802)was a gift from McNeil Pharmaceutical. PS1 phosphocellulose paper was from Whatman. Histone H1 (Iysine-rieh fraction) was purified from calf thymus as described previously [21] with the exception that 0.2 mM phenylmethylsalfonylfluoride(PMSF) was included in the initial extraction. The cAMP-dependent protein kinase inhibitor was purified from rabbit muscle as described previousty [22] through the acid precipitation step.

Preparation of isolated hepatocytes Hepatocytes were prepared from ad lib fed rats by the method of Berry and Friend [23] as modified by Krebs et al. [24]. The perfusion medium was Ca2+-free Krebs-Henseleit bicarbonate buffer containing 25 mM glucose, 10 mM dihydroxyacetone and 10 mM glutamine. For experiments in which the concentration of extracellular Ca2+ was held constant (see figure legends), hepatocytes were washed and resuspended in Krebs-Henseleit bicarbonate buffer containing 1.2 mM CaCI2. When the extracetlular Ca2+ concentration was varied, hepatocytes were washed and resuspended in CaZ+-free buffer and Caz÷ was added to the incubations as indicated.

Incubation of hepatocytes Hepatocytes were incubated in 25 ml flasks at a concentration of 40-65 mg wet wt./ml in a total volume of 2 ml. The atmosphere was 95% 02/5% CO 2. Flasks were treated with dimethyldichlurosilane before use to minimize adhesion of the cells to the glass surface. Palmixorate was dissolved in 15% bovine serum

albumin (BSA) and diluted 100-fold into the incubations. Parallel incubations in the absence of the inhibitor contained an equal amount of BSA. Phospholipase C was separated from low molecular weight components before use with Sephadex G25 [25].

Metabolite determinations Incubations were terminated by the addition of 0.1 mi 70% perchloric acid. Lactate, glucose and ATP were measured in the neutralized extract by standard enzymatic endpoint procedures [26]. Glucose and lactate accumulation were linear with time up to 40 rain.

Extraction of protein kinase C After incubation with the hormones, 0.5 ml aliquots of hepatocytes were removed and the cells were pelleted at 300 × g for 40 s. The medium was removed and the cells were dispersed in 1 ml of a buffer containing 20 mM Tris (pH 7.5), 3 mM EDTA, 0.5 mM D'IT, 0.25 M sucrose, 5 #g/ml leupeptin, 0.2 raM PMSF and 0.25 mg/ml saponin by vortexing. After 2 min, the extract was centrifuged at 14000 rpm for 40 s in an Eppendorf microfuge. The supernatant was the eytusolie extract. The pellet was dissolved 0.5 ml of a buffer containing 20 mM Tris (pH 7.5), 3 mM EDTA, 0.5 mM D'IT, 5 #g/ml leupeptin, 0.2 mM PMSF and 2% Triton. This was centrifuged for 20 rain at 14000 rpm at 4°C and the supernatant was the membrane extract.

Protein kinase C assay Cytosolic and membrane extracts (0.4 and 0.2 ml, respectively) were loaded onto 0.5 ml columns of DEAE-Sephacel equilibrated with 20 mM Tris (pH 7.5), 1 mM EDTA, 0.5 mM DTI" at 4 *C. The columns were washed with 1.5 ml of the same buffer and protein kinase C was eluted with the buffer containing 0.32 M NaCI in a volume of 1 ml. The eluate was diluted three-fold, as high salt concentrations inhibited protein kinase C activity. The assay contained 20 mM Tris (pH 7.5), 5 mM MgCI:, 2 mg/ml histone HI, 0.1 mg/ml protein kinase A inhibitor, 50 p.M [y-32P~TP (5.105 CPM/nmol) and 0.02 ml of the diluted eluate in a total volume of 0.'. ml. Protein kinase C activity was taken as the difi,:rence between the activity in presence and absence of 0.2 mg/ml phosphatidylserine: 4 #g/ml TPA. Since TPA rather than DAG was used as the activator, calcium did not affect the assay, as expected {27]. This is useful in excluding potential interference by other CaZ+-dependent kinases. Incubations were run at 30 °C for 20-40 min; the assay was linear with time for at least 40 rain with the dilutions used in these experiments. Aliquots (30 #i) of the reaction mixture were removed and spotte,~ onto P-81 phosphocellulose paper and immersed in 0.85%

163 phosphoric acid. The filter papers were washed with five changes of phosphoric acid and radioactivity was measured in a scintillation counter. Our values for protein kinase C activity agree well with those of nther investigators [28]. As a further control, we prepared brain extracts (a rich source of protein kinase C) which also gave activities that agreed with literature values [28].

0.8

E E ~L

0.6

°

.2 *6

0.4

o

0,2

Results 0.0

Effect of glucose on the vasopressin stimulation of lactate production As found by other investigators [16,17], vasopressin stimulated lactate production in the absence of added glucose in hepatocytes from fed rats. Exogenous glucose attenuated vasopressin stimulation, and at 50 mM glucose, vasopressin actually inhibited lactate formation (Fig. 1). By contrast, EGF stimulated lactate production at all glucose concentrations. The attenuation of the vasopressin effect was not surprising, since vasopressin and glucose have been shown to stimulate glycolysis by the same mechanism [18]. The inhibition observed with high glucose, however, suggested that vasopressin exerted an additional effect which was unmasked when the stimulation of glycolysis was reduced. To determine if va~pressin was influencing the utilization of lactate and pyruvate, hepatocytes were treated with a-cyano-4-hydroxycinnamate(a-CHC), an inhibitor of mitoehondriat pyruvate transport [29]. Fig. 2 shows that a-CHC reversed the inhibition of lactate production by vasopressin in the presence of 50 mM

¢~ 0.6

o

E 0.4 :a.

0,2~

~ . - - t , vosopress[n

,/0

_o 0,0 O

i

i

i

L

t0

20

50

40

50

glucose, mM

Fig. 1. Glucoseeffect on the vasopressinand EGF stimulationsof production.Hepatocyteswere incubatedfor 40 rain in the presence of various glucose concentrations.The incubationcon. rained 1.2 mM CaCIz and 64.2 mg wet wt./ml of hepatocytes. Concentrationsof the effectorswere lid nm vasopressinand 25 nM EGF. Similarresults were obtainedwith a separate preparationof hepalocytes; reproducibilityof the inhibitionobservedat high glucose concentrations is indicated by statistical analysis displayed in Fig. Z

lactate

control t'x'a vosopre~sin

"li ni

no Qddlt[on:;

glucose

0t-CHe

g[laCOtie + ~-CHC

Fig. 2. Reversal oi: t,e vasopressin-induced inhibition of laclate production in the presence of high glucose by a-CHC. Hepatocytcs were incubated for 40 rain with 50 mM glucose and/or 0.5 mM a-CHC. Incubalions contained 1.2 mM CaCI 2. The vasopressin concentration was lid aM. Results represent mcans_+S.E, of three separate hepatocyle preparations. * P < 0.05.

glucose. A small but significant vasopressin stimulation was observed in the presence of a-CHC.

Effects in Ca" +.depleted ceils Vasopressin stimulation of glucose output by hepatocytes from fed rats results largely from the Ca2+-dependent activation of glycogen phosphorylase, and requires medium Ca 2÷ [8]. We confirmed that vasopressin stimulates glucose output by hepatocytes from fed rats in the presence, but not in the absence, of medium Ca 2÷ (data not shown). We have previously found that the basal rate of lactate formation varies greatly between individual hepatocyte preparations from ad lib fed rats, and that the percent stimulation of this process by hormones was greater when the basal rate was relatively low [20]. To ensure low rates of lactate formation (and consequently large hormone effects), we added non-saturating concentrations of BtzcAMP to incubations for the studies shown in Fig. 3. Fig. 3A shows that vasopressin stimulated lactate production in the presence or absence of extracellular Ca2+ or in the presence of EGTA. The relatively small vasopressin stimulation of glucose release with calcium present (Fig. 3B) was due to the presence of the added Bt2cAMP; in the absence of the latter, we observed a 60% stimulation of glucose release by vasopressin (data not shown). Glucose release was decreased by vasopressin in the Ca2+-depleted cells (Fig. 3B). EGF also stimulated lactate production without increasing glycogenolysis. This hormone also has no effect on glucose release under conditions in which the vasopressin effect is large (i.e., in the absence of Bt2cAMP), suggesting it is not a member of the family of Ca-'+-mobilizing hormones [201.

t64

"~ 0.3 ._=

*

~=o~trol IEGF

• ~

AI

~

control

~

TeA

Ill

r'~ . . . . pressln

.~ 8

~

EGF

A1/ /

B

] i~"

17~ vasopressin

/

O 0~

0.2

~ 6

•- E 0.1

c- c

o.o

=

,

o_~

1.5

o Fig. 4. Failure of vasopressin and EGF to translocate protein kinase C, Incubations contained 1,2 mM CaC[ 2. Hepatocyles were preineubated for 20 rain and incubated for 8 rain with 1130nM vasopressln. 25 nM EGF or 0.5 #M TPA. Cells were fractionated and protein kinase C was assayed as described in Materials and Methods. Results represent means~:S.E, of three separate hepatocyte preparations. * P < 0.05.

;.0 'd o

0.5

~,, 0.0 t .2 mM Ca2+

no added Ca2+

I mM EGTA

Fig. 3. Stimulation of lactate production by vasopressin and EGF in CaZ".depleted hepatocytes. Hepatocytes were prepared in the absence of Ca t* and preincubated for 10 rain with 1,2 mM CaCI~, no added Ca t+ or 1 mM EGTA. All irtcubalions contained 300 nM Bt2cAMP. Incubations were started by addition of vasopressin (100 aM) or EGF (25 riM) and were conducted for 30 rain. Results represent rneans+S,E, of four separate hepatocyre preparations. * P