- Email: [email protected]
Inhibition of hepatic adenylate cyclase by NADH
Life Sciences, Vol. 49, pp. 915-923 Printed in the U.S.A.
Pergamon Press
INHIBITION OF HEPATIC A D E N Y L A T E C Y C L A S E BY NADH
Bipin G. Nair and Tarun B. Patel # Department of Pharmacology, The University of Tennessee, Memphis, The Health Science Center, 874 Union Avenue, Memphis, TN 38163, (901) 528-6006 (Received in final form July 17, 1991) Summarv
Adenylate cyclase activity in isolated rat liver plasma membranes was inhibited by NADH in a concentration-dependent manner. Half-maximal inhibition of adenylate cyclase was observed at 120 pM concentration of NADH. The effect of NADH was specific since adenylate cyclase activity was not altered by NAD+, NADP+, NADPH, and nicotinic acid. The ability of NADH to inhibit adenylate cyclase was not altered when the enzyme was stimulated by activating the Gs regulatory element with either glucagon or cholera toxin. Similarly, inhibition of Gi function by pertussis toxin treatment of membranes did not attenuate the ability of NADH to inhibit adenylate cyclase activity. Inhibition of adenylate cyclase activity to the same extent in the presence and absence of the Gpp(NH)p suggested that NADH directly affects the catalytic subunit. This notion was confirmed by the finding that NADH also inhibited solubilized adenylate cyclase in the absence of Gpp(NH)p. Kinetic analysis of the NADH-mediated inhibition suggested that NADH competes with ATP to inhibit adenylate cyclase; in the presence of NADH (1 mM) the Km for ATP was increased from 0.24 + 0.02 mM to 0.44 + 0.08 mM with no change in Vmax. This observation and the inability of high NADH concentrations to completely inhibit the enzyme suggest that NADH interacts at a site(s) on the enzyme to increase the Km for ATP by 2-fold and this inhibitory effect is overcome at high ATP concentrations. Previously, we demonstrated that, in the isolated perfused rat liver, the ability of physiological concentrations of glucagon to stimulate cAMP accumulation is determined by the cellular oxidation-reduction state (1). Hence, the maximal increase in cellular cAMP accumulation in response to 1 nM glucagon was observed when livers were perfused under conditions designed to hold the cellular redox potential at an intermediate level, i.e. perfusion with [lactate]/[pyruvate] ratio of 4 (1) or [3hydroxybutyrate/acetoacetate] ratio of 1 or 1.4 (2). Additionally, physiological concentrations of glucagon stimulated Ca 2+ efflux and increased the metabolic flux through the mitochondrial, Ca2+-sensitive (3-5), 2-oxoglutarate dehydrogenase complex in perfused livers only under conditions when glucagon-elicited cAMP accumulation was maximal (1,2). These findings, the observations that the cAMP analog 8-p-chlorophenylthio cAMP stimulated Ca 2+ efflux and metabolic flux through the 2-oxoglutarate dehydrogenase complex in livers irrespective of the cellular redox state (1,2) coupled with the ability of cAMP to increase cytosolic free-Ca 2+ # To whom correspondence should be addressed. 0024-3205/91
$3.00 +.00
Copyright (c) 1991 Pergamon Press plc
916
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
concentration in hepatocytes (6,7) suggested that redox potential regulates the accumulation of cAMP in response to glucagon and hepatic Ca 2+ fluxes and that activities of Ca2+-sensitive enzymes are determined by the extent to which cAMP accumulates. Regulation of cellular cAMP accumulation by oxidation-reduction state may involve alterations in the activities of adenylate cyclase and/or cAMP phosphodiesterase(s) either separately or in concert. To this end, studies of Clark and Jarret (8) have demonstrated that hepatic low Km cAMP phosphodiesterase activity is inhibited by NADH. Half-maximal inhibition of the cAMP phosphodiesterase activity was achieved at NADH concentrations of 20 I~M. Similarly, NADH has been demonstrated to inhibit (9) and NAD+ to stimulate (10) the adenylate cyclase activity of adipocyte plasma membranes and cardiac membranes, respectively. Although regulation of adenylate cyclase by NAD + and NADH has been reported (9,10) the mechanism(s) involved in this control remain to be elucidated. Additionally, although NAD+ stimulated cardiac adenylate cyclase, NADH did not alter the activity of this enzyme (10). In contrast, in adipocyte membranes NADH inhibited adenylate cyclase while NAD + did not exert any effects (8)° Hence, the effects of pyridine nucleotides on adenylate cyclase are tissue specific. Since, the effects of pyridine nucleotides on hepatic adenylate cyclase activity have not been documented, we have investigated the modulation of hepatic adenylate cyclase activity by pyridine nucleotides and examined the regulatory mechanism(s) involved. Materials and Methods
Male rats of the Sprague Dawley strain (175-200 g body weight) were used in these studies. Animals were allowed free access to food and water. Livers of rats anesthetized with pentobarbital sodium were excised and plasma membranes were isolated by the method of Neville (11). Protein was determined by the method of Bradford (12) employing bovine serum albumin as standard. Adenylate cyclase activity was assayed by the method of Salomon et al. (13), as described earlier (14) in a reaction mixture containing at the final concentrations: 50 mM Tris HCI, (pH 7.4), 5 mM MgCI 2, 12 mM phosphocreatine, 1mg/ml creatine kinase, l mM 3-isobutyl-l-methylxanthine (IBMX), and 0.1 mM ((~-32p)ATP (200 dpm/pmol). The inclusion of phosphocreatine/creatine phosphokinase ATP regenerating system maintained ATP concentrations at a constant level. Similarly, as monitored spectrophotometrically, NADH concentrations did not vary during the incubation period. Moreover, inclusion of lactate and lactate dehydrogenase (NADH regenerating system) in the incubation did not alter the effects of NADH indicating that the reduced pyridine nucleotide was not being metabolized at any significant rate. ADP-ribosylation of membrane proteins by pertussis toxin and cholera toxin was achieved employing the method of Lotersztajn et al. (15), as detailed in our earlier report (14). Adenylate cyclase activity was solubilized from membrane components by the method of Neer (16). Essentially, rat livers were homogenized in medium containing 0.1 M Tris-HCI (pH 7.6), 75 mM sucrose, 10 mM MgCI2, employing a loose fitting glass Dounce homogenizer. The homogenate was centrifuged at 15,000 X g for 20 minutes. The resulting pellet was resuspended in the aforementioned medium containing 2% (v/v) Triton X-100. This mixture was centrifuged at 100,000 X g for 2 hours. The resulting supernatant containing the solubilized adenylate cyclase activity was employed for further experiments.
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
917
Highly purified (>99% pure) NAD+, NADH, glucagon, guanyl-5'yl imidodophosphate (Gpp(NH)p), cAMP, IBMX, cholera toxin, and creatine phosphate were obtained from Sigma Chemical Co. 3H-labeled cyclic AMP was obtained from New England Nuclear Corp. [¢-32p]ATP was purchased from ICN Radiochemicals, Irvine, CA. Pertussis toxin was purchased from List Biological Laboratories. NADP +, NADPH, and rabbit skeletal muscle creatine kinase were obtained from Boehringer Mannheim Inc. All other chemicals were of the highest purity commercially available. Results and
Discussion
Initially, the effect of NADH on adenylate cyclase activity of liver plasma membranes was investigated. The data presented in Figure 1 demonstrate that NADH, in a concentration dependent manner, inhibited the adenylate cyclase activity in plasma 120 110
NAD +
100 ¢q
9O 8O
q,v ~*.,
~v
70 6O 50 4O 0.0
0.2 0.4 0.6 0.8 Nucleotide Concentration (mM)
1.0
FIG. 1
The effect of varying NADH and NAD + concentrations on the activity of adenylate ¢yclase In rat liver plasma membranes. Adenylate cyclase activity was
assayed in the presenceof 10 IIM Gpp(NH)p and with or without various concentrationsof NADH or NAD+. Data are the mean + S.EM percent of control activity (47.3 +_ 6.8 pmol/min/mg protein),
membranes isolated from rat livers. Maximal inhibition of adenylate cyclase (50% of control) was observed at concentration of NADH > 1 mM, and half-maximal inhibition required 0.120 mM of this nucleotide (figure 1). These findings are to some extent, similar to the findings of Low and Werner (9) which demonstrate that NADH also inhibits adenylate cyclase activity of adipocyte plasma membranes by 50%. However, Low and Werner (9) observed half-maximal inhibition of adenylate cyclase at 3 mM concentration of NADH. This difference in the potency of NADH appears to be related to the assay conditions employed since under conditions similar to those of Low and Werner (9), i.e. in the presence of DTT (2 mM), concentrations of NADH required to inhibit the adenylate cyclase activity in plasma membranes of livers were similar to those reported for the enzyme in adipocyte membranes (9; data not shown). Although the adenylate cyclase activity was more sensitive to inhibition by NADH in our experiments, it should be noted that the sensitivity of this enzyme is at least 6-fold less than the sensitivity of the hepatic low Km cAMP phosphodiesterase (8). Since the cAMP phosphodiesterase is half-maximally inhibited at NADH concentrations of 20p.M (8), while similar inhibition of adenylate cyclase requires 120 I~M NADH, it is conceiveable that in intact hepatocytes or livers, the extent to which the adenylate
918
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
cyclase and cAMP phosphodiesterase are inhibited by a given concentration of NADH, determines the accumulation of cellular cAMP. The ability of NADH to modulate adenylate cyclase activity appeared to be specific since the enzyme was much less sensitive to modulation by NAD+ (figure 1). Similarly, NADP +, NADPH, and nicotinic acid also did not alter the hepatic adenylate cyclase activity (data not shown). Moreover, the inclusion of increasing NAD+ concentrations (0.1 to 1.0 mM) also did not alter the degree of inhibition of adenylate cyclase by 1 mM NADH (data not shown). Since the activity of adenylate cyclase is regulated by GTP binding regulatory proteins, Gs and Gi (17), and because the experiments depicted in figure 1 were performed in the presence of the GTP analog, Gpp(NH)p (10 t~M), the possibility that NADH inhibited adenylate cyclase by alterating the activity of either Gs or Gi was examined. The data in Table I demonstrate that, although glucagon (10 nM) stimulated hepatic adenylate cyclase by 2.3-fold, the ability of NADH to inhibit the enzyme was not attenuated, suggesting that alterations in Gs activity do not affect the responses of adenylate cyclase to NADH. These findings are further supported by our observations that when membranes were treated with cholera toxin to ADP-ribosylate and maximally activate Gs (18), the activity of adenylate cyclase was stimulated by 2-fold. However, the ability of NADH to inhibit the enzyme activity was not altered (Table II). Complete ADP-ribosylation of Gs by cholera toxin was indicated by the inability of glucagon to further stimulate adenylate cyclase activity (data not shown). Additionally, when membranes were TABLE I Effect of NADH on Glucagon-Stimulated Adenylate Cyclase Activity of Liver Membranes.
Incubation Condition
Adenylate Cyclase Activity (pmoVmin/mg)
Basal +Glucagon (10 nM) +Giucagon+ 0.1 mM NADH +Glucagon + 1.0 mM NADH
32.9 _+0.5 74.9 + 2.5*** 64.9 + 1.4" 49.5 _+2.9**
Liver plasma membraneswere assayedfor adenylatecyclase activity in the presence of 10 idvl Gpp(NH)p. NADH was added to the assays at the two concentrations indicated. Data are presented as mean + S.E.M. of 3 to 5 experiments. *p<0.05; **p<0.005;***p<0.005 (Students t-test) as compared with corresponding control. TABLE. Effect of NADH on Adenylate Cyclase Activity In Rat Liver Plasma Membranes Treated with either Cholera Toxin or Pertussis Toxin.
Incubation condition Basal +0.1 mM NADH + 1.0 mM NADH
Control Choleratoxin Pertussistoxin Membranes treatedmembranes treated membrane Adenylate Cyclase Activity (prnol/min/mgprotein) 30.1 _+1.4 61.6+_8.7** 30.8 + 3.9 N.D. 38.4_+2.4* N.D. 19.1 _+1.2* 27.6_+1.5" 13.8+ 1.7*
Membrane proteins were ADP-ribosylated as described in the Materials and Methods section. Control membranes were similarly treated in the absence of toxins. Adenylate cyclase was assayed in the presence of 10 ~M Gpp(NH)p. Data presented are mean + S.E.M. of three separate experiments. N.D.: Not Determined. *p<0.005; **p<0.001.
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
919
treated with pertussis toxin to ADP-ribosylate and inactivate Gi, the ability of NADH to inhibit adenylate cyclase activity was also not altered (Table II). As previously reported by us and others (14,19), pertussis toxin treatment of membranes did not alter the basal adenylate cyclase activity (Table II). Therefore, to determine the extent of pertussis toxin-mediated ADP-ribosylation and inactivation of Gi, the ability of the adenosine analog, (-)-N6-(R-phenylisopropyl)-adenosine (PIA) to inhibit forskolinstimulated adenylate cyclase activity was investigated (20,21). PIA (1 I~M) inhibited by 30% the forskolin (50 i~M)-stimulated adenylate cyclase activity of control membranes incubated with ADP-ribosylation mixture in the absence of the toxin. However, the forskolin-stimulated activity of pertussis toxin treated membranes was not altered by PIA (1 I~M) (data not shown). The experiments involving covalent modifications of Gs and Gi by cholera and pertussis toxins, respectively (Table II), suggest that NADH does not mediate its inhibitory actions on adenylate cyclase via alterations in the functional activity of its regulatory G proteins. To address this possibility that NADH altered the activity of the catalytic subunit directly, the ability of NADH to inhibit adenylate cyclase activity in liver plasma membranes in the absence of the GTP analog, Gpp(NH)p, was tested. The absence of any endogenous GTP in these experiments is indicated by the inability of glucagon to stimulate enzyme activity (Table III). As demonstrated by the data in Table III, in the absence of Gpp(NH)p, NADH inhibited the activity of adenylate cyclase in a concentration dependent manner to the same extent as that in the presence of the guanine nucleotide. Additionally, when the activity of adenylate cyclase was stimulated by forskolin which activates the catalytic subunit (21), the ability of NADH to inhibit the enzyme was not attenuated. It is noteworthy that forskolin stimulated adenylate cyclase to a greater extent in the absence of Gpp(NH)p than in the presence of the guanine nucleotide (Table III). This difference, which has previously been reported (22) is a reflection of activation of Gi in the presence of Gpp(NH)p. The data in Table III suggest that NADH inhibits adenylate cyclase in liver plasma membranes by directly altering the activity of the catalytic subunit. Despite this suggestion, one additional possibility which must not be excluded is that NADH alters the structural constraints exerted on the catalytic subunit by the membrane components, and therefore, changes the conformation of the catalytic subunit such that its activity is attenuated. To test this possibility, adenylate cyclase was solubilized by triton X-100 TABLE III NADH-Medlated Inhibition of Adenylate Cyclase of Liver Plasma Membranes in the presence and absence of Gpp(NH)p, and Forskolln.
Incubation Condition Basal + Glucagon (10 nM) + 0.1 mM NADH + 1.0 mM NADH + Forskolin (50 llM) + Forskolin + 0.1 mM NADH + Forskolin+ 1.0 mM NADH
+Gpp(NH)p
-Gpp(NH)p
Adenylate CyclaseActivity(pmol/min/mg protein) 47.3 ± 6.8 11.6 +_2.0t 83.9 ± 4.8** 13.6 ± 3.0 36.5 ± 1.7"* 7.3 ± 0.2* 23.1 ± 2.4** 5.8 ± 0.5** 101.7+ 0.7** 128.4± 2.3"*t 92.4 ± 0.2** 95.3 ± 1.1"* 79.5 ± 1.0"* 76.0 ± 1.4**
Adenylate cyclase assays were performed either in the absence or presence of the GTP analog, Gpp(NH)p (10 I.LM). Data presented are mean + S.E.M.of 3 to 5 separate experiments. *p<0.05; **p<0.001 as compared with corresponding control within the same group; tp<0.001 as compared with similarconditionwith Gpp(NH)p. treatment of plasma membranes as described under Materials and Methods section. As demonstrated in Table IV, NADH inhibited the solubilized enzyme assayed in the absence of Gpp(NH)p, in a manner similar to the membrane bound enzyme assayed
920
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
either in the presence or absence of Gpp(NH)p (figure 1, Table III). Similar to our findings with the membrane bound enzyme, the solubilized adenylate cyclase was TABLE IV The Effect of NADH on Solubilized Adenylate Cyclase Activity Derived from Rat Liver Plasma Membranes.
Incubation Conditions Basal + 0.1 mM NADH + 1.0 mM NADH + 5.0 mM NADH + 10 mM NADH + Forskolin(50 pM) + Forskolin+ 0.1 mM NADH + Forskolin+ 1.0 mM NADH + Forskolin+ 5.0 mM NADH + Forskolin+ 10 mM NADH
Adenylate Cyclase Activity (pmol/min/mgprotein) 44.0 + 0.9 31.9 + 1.1* 23.9 + 0.5* 24.1 + 0.4* 21.7 -+0.7* 321.3 + 4.5* 260.1 + 3.1" 205.0 + 1.8" 166.3 + 2.4* 144.4 + 2.8*
Adenylate cyclase was solubilized as described in Materials and Methods section and assayed in the absence of the GTP analog, Gpp(NH)p. Data presented are mean + S.E.M. of at least four determinations. *p<0.001 as compared with corresponding control. also not inhibited by NADP + (1 mM), NADPH (1 mM) or NAD + (0.5 mM) (data not shown). Additionally, as demonstrated for the membrane bound enzyme (Table III), NADH inhibited both the basal and forskolin-stimulated adenylate cyclase activity (Table IV). Moreover, irrespective of the starting activity, i.e. basal or forskolinstimulated, the maximum inhibition obtained in the presence of NADH was the same (approximately 50%) (Table IV). These data strongly indicate that NADH inhibits adenylate cyclase activity by directly interacting with the catalytic subunit. To further characterize the nature of the interaction between NADH and catalytic subunit, experiments were performed to examine the effects of NADH on solubilized adenylate cyclase activity in the presence of different ATP concentrations. In these studies the GTP analog, Gpp(NH)p was not included. Lineweaver-Burk plot representation of data from these experiments (figure 2) demonstrate that NADH inhibits adenylate cyclase by competing with the substrate ATP. NADH (1 mM) increased (p<0.001) the Km for ATP (control = 0.24 + 0.02 (n=9); +NADH = 0.44 + 0.08 mM, (n=9)) without altering the Vmax of the enzyme (figure 2). Similar results were also obtained both in the presence and absence of glucagon (10 nM) employing rat liver membrane preparations in place of the solubilized enzyme (data not shown). However, competitive inhibition of an enzyme dictates that at high concentrations of inhibitor near complete inhibition of activity should be observed. Clearly this is not the case with NADH-mediated inhibition of adenylate cyclase since the maximum inhibition obtained with NADH in either membranes (figure 1) or solubilized enzyme preparation (Table IV) was only 50%. Our observations could be explained if two different catalytic subunits of adenylate cyclase were present in the liver membrane or solubilized preparations and only one of these enzymes were inhibited by NADH. However, the existence of isozymes would predict that the Lineweaver-Burk plot would be non-linear. Since we have not observed a non-linear Lineweaver-Burk plot under any condition tested, this possibility is unlikely. However, stringent evaluations for the existence of adenylate cyclase isoforms in liver must await further investigations. The most plausible explanation of our data would be that NADH interacts at a site(s) on the catalytic subunit of adenylate cyclase that is(are) distinct from the catalytic site and doubles the Km for ATP. The resulting partial competitive inhibition (23), however, is overcome at concentrations of ATP that are greater than the Km in the presence of
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
921
°
P
0.03
~o
..= =
0.01
== E >
-5
0.01
0
!
i
i
!
z
5
10
15
20
25
I/ATP (raM "1)
FIG. 2
Effect of NADH on solubilized adenylate cyclase activity in the presence of varying concentrations of ATP. Enz;yme assays were performed with different ATP concentrations in the presence and absence of NADH. Representative experiment is shown. K m values for ATP estimated from several experiments were: Control = 0.24 + 0.02 mM (n=9); +NADH (1 mM) = 0.44 + 0.08 mM * (n=9). *p<0.001.
NADH. As described by Segel (23), when (a) the substrate (S) and the inhibitor (I) bind to the enzyme (E) at different sites to yield ES, El and ESl complexes, (b) the substrate binds to the free enzyme with a greater affinity than to the El complex, and (c) the ES and ESI complexes both yield product with equal facility, then the inhibition observed can be termed partial competetive inhibition. This model ideally fits the partial (incomplete) yet competitive inhibition of adenylate cyclase by NADH. Since the effects of NADH on hepatic adenylate cyclase can be overcome at high ATP concentrations, it may be argued that NADH-mediated regulation of this enzyme may have no physiological significance. However, it should be noted that most of the cellular adenine nucleotides are in the bound form and unavailable for enzymatic reactions (24,25). Moreover, because intracellular ATP may not be uniformly distributed (26), it is possible that ATP available for the membrane bound adenylate cyclase may be low enough for NADH to inhibit its activity. This coupled with the presence of NADH in livers at concentrations sufficient to significantly inhibit adenylate cyclase, i.e. between 0.1 and 0.2 mM (27), indicate that NADH may regulate adenylate cyclase activity. In this regard, inhibition of adenylate cyclase by NADH in the liver may explain a number of experimental findings from various laboratories. For instance, since the cAMP phosphodiesterase is known to be inhibited by NADH (8), it may be expected that in livers perfused under highly reducing conditions, e.g. lactate alone, glucagon-elicited cAMP accumulation would be very high. A similar argument would also apply to hepatocytes incubated with lactate alone. However, in our previous studies (1), when rat livers were perfused with lactate, a substrate which increases cellular NADH/NAD+ ratio, glucagon (1 nM) elevated cAMP levels to a lesser extent as compared to the cAMP accumulation observed in livers perfused with lactate plus pyruvate such that the ratio of [lactate]/[pyruvate] was held constant at 4. Similarly, in hepatocytes incubated with lactate alone the glucagon-mediated elevation of cellular cAMP content was also lower than that observed in hepatocytes incubated with lactate plus pyruvate (8). Since lactate and pyruvate do not alter the activities of adenylate cyclase (data not shown) or cAMP phosphodiesterase (8), the findings described above would suggest that an elevation in cellular NADH content
922
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
inhibits the adenylate cyclase activity in hepatocytes. This contention is further supported by the findings of Zederman et al. (28) that in hepatocytes incubated with the cAMP phosphodiesterase inhibitor, IBMX, and lactate, glucagon increased cAMP levels to a lesser extent than in hepatocytes incubated with IBMX alone. In summary, our data demonstrate that NADH, selectively inhibits the hepatic adenylate cyclase activity. The experimental evidence provided also indicates that NADH directly inhibits the catalytic subunit of hepatic adenylate cyclase without altering the activities of Gi or Gs. Moreover, the data suggest that NADH by interacting at a site(s) distinct from the catalytic site increases the Km for ATP, and in the presence of excess ATP the inhibition due to increased Km is overcome. Overall, these findings can explain the modulation by cellular redox potential of agonist stimulated increases in cellular cAMP content in perfused livers and hepatocytes.
A~knowledaements Supported by grants DK 35713 and DK 01742 from the NIH. T.B.P. is a recipient of NIH Research Career Development Award (DK 01742).
RefQrences 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.
H.M RASHED and T.B. PATEL, J. Biol. Chem. 262 15953-15958 (1987). H.M. RASHED, F.M. WALLER and T.B.PATEL, J. Biol. Chem. 263 5700-5706 (1988). J.G. MCCORMACK, Biochem. J. 231 581-595 (1985). J.G. MCCORMACK, Biochem. J. 231 597-608 (1985). J.G. MCCORMACK and R.M. DENTON, Biochem. J. 180 533-544 (1979). F.D. SlSTARE, R.A. PICKING and R.C. HAYNES, Jr, J. Biol. Chem. 21~0 1274412747 (1985). P.F. BLACKMORE and J.H. EXTON, J. Biol. Chem. :~(~1 11056-11063 (1986). M.G. CLARK and I.G. JARRETF, Biochem. J. 17i~ 805-816 (1978). H. LOW, and S. WERNER, FEBS Lett. 65 96-98 (1976). A.M. FELDMAN, M.A. LEVINE, K.L. BAUGHMAN and C. VAN DOP, Biochem. Biophys. Res. Commun. 142 631-637 (1987). D.M. NEVILLE, Jr., Biochem. Biophys. Acta. 154 540-552 (1968). M.M. BRADFORD, Anal. Biochem. 72248-254 (1976). Y. SALOMON, C. LONDOS and M. RODBELL, Anal. Biochem. 58 541-548 (1974). B.G. NAIR., H. M. RASHED., and T. B. PATEL, Biochem. J. 264 563-571 (1989) S. LOTERSZTAJN, C. PAVOINE, A. MALLAT, D. STENGEL, P.A. INSEL and F. PECKER, J. Biol. Chem. 262 3114-3117 (1987). E.J. NEER, J. Biol. Chem. 254 2089-2096 (1979). A.G. GILMAN, Ann. Rev. Biochem. 56 615-649 (1987). J. MOSS, and M. VAUGHN, Ann. Rev. Biochem. 48 581-600 (1979). M.B. ANAND-SRIVASTAVA, A.K. SRIVASTAVA and M. CANTIN, J. Biol. Chem. 262 4931-4937 (1987). J. WOLFF, C. LONDOS and D. M. F. COOPER, Adv. Cyclic Nucleotide Res.14 199-214 (1981). K. SEAMON and J.W. DALY, J. Biol. Chem. 256 9799-9801 (1981). T. KATADA, G.M. BOKOCH, J.K. NORTHUP, M. UI and A. G. GILMAN, J. Biol. Chem. 259 3568-3577 (1984). I. H. SEGEL, Raoid eouilibrium oartial and mixed tvoe inhibition. In "Enzvme Kinetics: Behaviour and Analv$is of Raoid Eauilibrium and Steadv State Enz;cme Systems." John Wiley & Sons, pp. 161-226, New York; Chichester; Brisbane; Toronto (1975).
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
923
24. R.L. VEECH, J.W.R. LAWSON, N.W. CORNELL, and H.A. KREBS, J. Biol. Chem. 254 6538-6547 (1979). 25. T.P.M. AKERBOOM, H. BOOKELMAN, P.F. ZUURENDONK, R. VAN DER MEER and J.M. TAGER, Eur. J. Biochem. 84 413-420 (1978). 26. D.S. MILLER and S.B. Horowitz, J. Biol. Chem. 261 13911-13915 (1986). 27. M. KLINGENBERG, Nicotinamide-adenine dinucleotides and dinucleotide ohosohates ~NAD. NADP. NADH NADPH~. In "Methods of Enzymatic Analysi,~," ill. U. Bergmeyer ed.) Vol. VII, pp 251-271, VCH Verlagsgesellschaft mbH, Weinheim, Germany (1985). 28. R. ZEDERMAN, H. LOW, and K. HALL, FEBS Lett. 75 291-294 (1977).
Pergamon Press
INHIBITION OF HEPATIC A D E N Y L A T E C Y C L A S E BY NADH
Bipin G. Nair and Tarun B. Patel # Department of Pharmacology, The University of Tennessee, Memphis, The Health Science Center, 874 Union Avenue, Memphis, TN 38163, (901) 528-6006 (Received in final form July 17, 1991) Summarv
Adenylate cyclase activity in isolated rat liver plasma membranes was inhibited by NADH in a concentration-dependent manner. Half-maximal inhibition of adenylate cyclase was observed at 120 pM concentration of NADH. The effect of NADH was specific since adenylate cyclase activity was not altered by NAD+, NADP+, NADPH, and nicotinic acid. The ability of NADH to inhibit adenylate cyclase was not altered when the enzyme was stimulated by activating the Gs regulatory element with either glucagon or cholera toxin. Similarly, inhibition of Gi function by pertussis toxin treatment of membranes did not attenuate the ability of NADH to inhibit adenylate cyclase activity. Inhibition of adenylate cyclase activity to the same extent in the presence and absence of the Gpp(NH)p suggested that NADH directly affects the catalytic subunit. This notion was confirmed by the finding that NADH also inhibited solubilized adenylate cyclase in the absence of Gpp(NH)p. Kinetic analysis of the NADH-mediated inhibition suggested that NADH competes with ATP to inhibit adenylate cyclase; in the presence of NADH (1 mM) the Km for ATP was increased from 0.24 + 0.02 mM to 0.44 + 0.08 mM with no change in Vmax. This observation and the inability of high NADH concentrations to completely inhibit the enzyme suggest that NADH interacts at a site(s) on the enzyme to increase the Km for ATP by 2-fold and this inhibitory effect is overcome at high ATP concentrations. Previously, we demonstrated that, in the isolated perfused rat liver, the ability of physiological concentrations of glucagon to stimulate cAMP accumulation is determined by the cellular oxidation-reduction state (1). Hence, the maximal increase in cellular cAMP accumulation in response to 1 nM glucagon was observed when livers were perfused under conditions designed to hold the cellular redox potential at an intermediate level, i.e. perfusion with [lactate]/[pyruvate] ratio of 4 (1) or [3hydroxybutyrate/acetoacetate] ratio of 1 or 1.4 (2). Additionally, physiological concentrations of glucagon stimulated Ca 2+ efflux and increased the metabolic flux through the mitochondrial, Ca2+-sensitive (3-5), 2-oxoglutarate dehydrogenase complex in perfused livers only under conditions when glucagon-elicited cAMP accumulation was maximal (1,2). These findings, the observations that the cAMP analog 8-p-chlorophenylthio cAMP stimulated Ca 2+ efflux and metabolic flux through the 2-oxoglutarate dehydrogenase complex in livers irrespective of the cellular redox state (1,2) coupled with the ability of cAMP to increase cytosolic free-Ca 2+ # To whom correspondence should be addressed. 0024-3205/91
$3.00 +.00
Copyright (c) 1991 Pergamon Press plc
916
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
concentration in hepatocytes (6,7) suggested that redox potential regulates the accumulation of cAMP in response to glucagon and hepatic Ca 2+ fluxes and that activities of Ca2+-sensitive enzymes are determined by the extent to which cAMP accumulates. Regulation of cellular cAMP accumulation by oxidation-reduction state may involve alterations in the activities of adenylate cyclase and/or cAMP phosphodiesterase(s) either separately or in concert. To this end, studies of Clark and Jarret (8) have demonstrated that hepatic low Km cAMP phosphodiesterase activity is inhibited by NADH. Half-maximal inhibition of the cAMP phosphodiesterase activity was achieved at NADH concentrations of 20 I~M. Similarly, NADH has been demonstrated to inhibit (9) and NAD+ to stimulate (10) the adenylate cyclase activity of adipocyte plasma membranes and cardiac membranes, respectively. Although regulation of adenylate cyclase by NAD + and NADH has been reported (9,10) the mechanism(s) involved in this control remain to be elucidated. Additionally, although NAD+ stimulated cardiac adenylate cyclase, NADH did not alter the activity of this enzyme (10). In contrast, in adipocyte membranes NADH inhibited adenylate cyclase while NAD + did not exert any effects (8)° Hence, the effects of pyridine nucleotides on adenylate cyclase are tissue specific. Since, the effects of pyridine nucleotides on hepatic adenylate cyclase activity have not been documented, we have investigated the modulation of hepatic adenylate cyclase activity by pyridine nucleotides and examined the regulatory mechanism(s) involved. Materials and Methods
Male rats of the Sprague Dawley strain (175-200 g body weight) were used in these studies. Animals were allowed free access to food and water. Livers of rats anesthetized with pentobarbital sodium were excised and plasma membranes were isolated by the method of Neville (11). Protein was determined by the method of Bradford (12) employing bovine serum albumin as standard. Adenylate cyclase activity was assayed by the method of Salomon et al. (13), as described earlier (14) in a reaction mixture containing at the final concentrations: 50 mM Tris HCI, (pH 7.4), 5 mM MgCI 2, 12 mM phosphocreatine, 1mg/ml creatine kinase, l mM 3-isobutyl-l-methylxanthine (IBMX), and 0.1 mM ((~-32p)ATP (200 dpm/pmol). The inclusion of phosphocreatine/creatine phosphokinase ATP regenerating system maintained ATP concentrations at a constant level. Similarly, as monitored spectrophotometrically, NADH concentrations did not vary during the incubation period. Moreover, inclusion of lactate and lactate dehydrogenase (NADH regenerating system) in the incubation did not alter the effects of NADH indicating that the reduced pyridine nucleotide was not being metabolized at any significant rate. ADP-ribosylation of membrane proteins by pertussis toxin and cholera toxin was achieved employing the method of Lotersztajn et al. (15), as detailed in our earlier report (14). Adenylate cyclase activity was solubilized from membrane components by the method of Neer (16). Essentially, rat livers were homogenized in medium containing 0.1 M Tris-HCI (pH 7.6), 75 mM sucrose, 10 mM MgCI2, employing a loose fitting glass Dounce homogenizer. The homogenate was centrifuged at 15,000 X g for 20 minutes. The resulting pellet was resuspended in the aforementioned medium containing 2% (v/v) Triton X-100. This mixture was centrifuged at 100,000 X g for 2 hours. The resulting supernatant containing the solubilized adenylate cyclase activity was employed for further experiments.
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
917
Highly purified (>99% pure) NAD+, NADH, glucagon, guanyl-5'yl imidodophosphate (Gpp(NH)p), cAMP, IBMX, cholera toxin, and creatine phosphate were obtained from Sigma Chemical Co. 3H-labeled cyclic AMP was obtained from New England Nuclear Corp. [¢-32p]ATP was purchased from ICN Radiochemicals, Irvine, CA. Pertussis toxin was purchased from List Biological Laboratories. NADP +, NADPH, and rabbit skeletal muscle creatine kinase were obtained from Boehringer Mannheim Inc. All other chemicals were of the highest purity commercially available. Results and
Discussion
Initially, the effect of NADH on adenylate cyclase activity of liver plasma membranes was investigated. The data presented in Figure 1 demonstrate that NADH, in a concentration dependent manner, inhibited the adenylate cyclase activity in plasma 120 110
NAD +
100 ¢q
9O 8O
q,v ~*.,
~v
70 6O 50 4O 0.0
0.2 0.4 0.6 0.8 Nucleotide Concentration (mM)
1.0
FIG. 1
The effect of varying NADH and NAD + concentrations on the activity of adenylate ¢yclase In rat liver plasma membranes. Adenylate cyclase activity was
assayed in the presenceof 10 IIM Gpp(NH)p and with or without various concentrationsof NADH or NAD+. Data are the mean + S.EM percent of control activity (47.3 +_ 6.8 pmol/min/mg protein),
membranes isolated from rat livers. Maximal inhibition of adenylate cyclase (50% of control) was observed at concentration of NADH > 1 mM, and half-maximal inhibition required 0.120 mM of this nucleotide (figure 1). These findings are to some extent, similar to the findings of Low and Werner (9) which demonstrate that NADH also inhibits adenylate cyclase activity of adipocyte plasma membranes by 50%. However, Low and Werner (9) observed half-maximal inhibition of adenylate cyclase at 3 mM concentration of NADH. This difference in the potency of NADH appears to be related to the assay conditions employed since under conditions similar to those of Low and Werner (9), i.e. in the presence of DTT (2 mM), concentrations of NADH required to inhibit the adenylate cyclase activity in plasma membranes of livers were similar to those reported for the enzyme in adipocyte membranes (9; data not shown). Although the adenylate cyclase activity was more sensitive to inhibition by NADH in our experiments, it should be noted that the sensitivity of this enzyme is at least 6-fold less than the sensitivity of the hepatic low Km cAMP phosphodiesterase (8). Since the cAMP phosphodiesterase is half-maximally inhibited at NADH concentrations of 20p.M (8), while similar inhibition of adenylate cyclase requires 120 I~M NADH, it is conceiveable that in intact hepatocytes or livers, the extent to which the adenylate
918
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
cyclase and cAMP phosphodiesterase are inhibited by a given concentration of NADH, determines the accumulation of cellular cAMP. The ability of NADH to modulate adenylate cyclase activity appeared to be specific since the enzyme was much less sensitive to modulation by NAD+ (figure 1). Similarly, NADP +, NADPH, and nicotinic acid also did not alter the hepatic adenylate cyclase activity (data not shown). Moreover, the inclusion of increasing NAD+ concentrations (0.1 to 1.0 mM) also did not alter the degree of inhibition of adenylate cyclase by 1 mM NADH (data not shown). Since the activity of adenylate cyclase is regulated by GTP binding regulatory proteins, Gs and Gi (17), and because the experiments depicted in figure 1 were performed in the presence of the GTP analog, Gpp(NH)p (10 t~M), the possibility that NADH inhibited adenylate cyclase by alterating the activity of either Gs or Gi was examined. The data in Table I demonstrate that, although glucagon (10 nM) stimulated hepatic adenylate cyclase by 2.3-fold, the ability of NADH to inhibit the enzyme was not attenuated, suggesting that alterations in Gs activity do not affect the responses of adenylate cyclase to NADH. These findings are further supported by our observations that when membranes were treated with cholera toxin to ADP-ribosylate and maximally activate Gs (18), the activity of adenylate cyclase was stimulated by 2-fold. However, the ability of NADH to inhibit the enzyme activity was not altered (Table II). Complete ADP-ribosylation of Gs by cholera toxin was indicated by the inability of glucagon to further stimulate adenylate cyclase activity (data not shown). Additionally, when membranes were TABLE I Effect of NADH on Glucagon-Stimulated Adenylate Cyclase Activity of Liver Membranes.
Incubation Condition
Adenylate Cyclase Activity (pmoVmin/mg)
Basal +Glucagon (10 nM) +Giucagon+ 0.1 mM NADH +Glucagon + 1.0 mM NADH
32.9 _+0.5 74.9 + 2.5*** 64.9 + 1.4" 49.5 _+2.9**
Liver plasma membraneswere assayedfor adenylatecyclase activity in the presence of 10 idvl Gpp(NH)p. NADH was added to the assays at the two concentrations indicated. Data are presented as mean + S.E.M. of 3 to 5 experiments. *p<0.05; **p<0.005;***p<0.005 (Students t-test) as compared with corresponding control. TABLE. Effect of NADH on Adenylate Cyclase Activity In Rat Liver Plasma Membranes Treated with either Cholera Toxin or Pertussis Toxin.
Incubation condition Basal +0.1 mM NADH + 1.0 mM NADH
Control Choleratoxin Pertussistoxin Membranes treatedmembranes treated membrane Adenylate Cyclase Activity (prnol/min/mgprotein) 30.1 _+1.4 61.6+_8.7** 30.8 + 3.9 N.D. 38.4_+2.4* N.D. 19.1 _+1.2* 27.6_+1.5" 13.8+ 1.7*
Membrane proteins were ADP-ribosylated as described in the Materials and Methods section. Control membranes were similarly treated in the absence of toxins. Adenylate cyclase was assayed in the presence of 10 ~M Gpp(NH)p. Data presented are mean + S.E.M. of three separate experiments. N.D.: Not Determined. *p<0.005; **p<0.001.
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
919
treated with pertussis toxin to ADP-ribosylate and inactivate Gi, the ability of NADH to inhibit adenylate cyclase activity was also not altered (Table II). As previously reported by us and others (14,19), pertussis toxin treatment of membranes did not alter the basal adenylate cyclase activity (Table II). Therefore, to determine the extent of pertussis toxin-mediated ADP-ribosylation and inactivation of Gi, the ability of the adenosine analog, (-)-N6-(R-phenylisopropyl)-adenosine (PIA) to inhibit forskolinstimulated adenylate cyclase activity was investigated (20,21). PIA (1 I~M) inhibited by 30% the forskolin (50 i~M)-stimulated adenylate cyclase activity of control membranes incubated with ADP-ribosylation mixture in the absence of the toxin. However, the forskolin-stimulated activity of pertussis toxin treated membranes was not altered by PIA (1 I~M) (data not shown). The experiments involving covalent modifications of Gs and Gi by cholera and pertussis toxins, respectively (Table II), suggest that NADH does not mediate its inhibitory actions on adenylate cyclase via alterations in the functional activity of its regulatory G proteins. To address this possibility that NADH altered the activity of the catalytic subunit directly, the ability of NADH to inhibit adenylate cyclase activity in liver plasma membranes in the absence of the GTP analog, Gpp(NH)p, was tested. The absence of any endogenous GTP in these experiments is indicated by the inability of glucagon to stimulate enzyme activity (Table III). As demonstrated by the data in Table III, in the absence of Gpp(NH)p, NADH inhibited the activity of adenylate cyclase in a concentration dependent manner to the same extent as that in the presence of the guanine nucleotide. Additionally, when the activity of adenylate cyclase was stimulated by forskolin which activates the catalytic subunit (21), the ability of NADH to inhibit the enzyme was not attenuated. It is noteworthy that forskolin stimulated adenylate cyclase to a greater extent in the absence of Gpp(NH)p than in the presence of the guanine nucleotide (Table III). This difference, which has previously been reported (22) is a reflection of activation of Gi in the presence of Gpp(NH)p. The data in Table III suggest that NADH inhibits adenylate cyclase in liver plasma membranes by directly altering the activity of the catalytic subunit. Despite this suggestion, one additional possibility which must not be excluded is that NADH alters the structural constraints exerted on the catalytic subunit by the membrane components, and therefore, changes the conformation of the catalytic subunit such that its activity is attenuated. To test this possibility, adenylate cyclase was solubilized by triton X-100 TABLE III NADH-Medlated Inhibition of Adenylate Cyclase of Liver Plasma Membranes in the presence and absence of Gpp(NH)p, and Forskolln.
Incubation Condition Basal + Glucagon (10 nM) + 0.1 mM NADH + 1.0 mM NADH + Forskolin (50 llM) + Forskolin + 0.1 mM NADH + Forskolin+ 1.0 mM NADH
+Gpp(NH)p
-Gpp(NH)p
Adenylate CyclaseActivity(pmol/min/mg protein) 47.3 ± 6.8 11.6 +_2.0t 83.9 ± 4.8** 13.6 ± 3.0 36.5 ± 1.7"* 7.3 ± 0.2* 23.1 ± 2.4** 5.8 ± 0.5** 101.7+ 0.7** 128.4± 2.3"*t 92.4 ± 0.2** 95.3 ± 1.1"* 79.5 ± 1.0"* 76.0 ± 1.4**
Adenylate cyclase assays were performed either in the absence or presence of the GTP analog, Gpp(NH)p (10 I.LM). Data presented are mean + S.E.M.of 3 to 5 separate experiments. *p<0.05; **p<0.001 as compared with corresponding control within the same group; tp<0.001 as compared with similarconditionwith Gpp(NH)p. treatment of plasma membranes as described under Materials and Methods section. As demonstrated in Table IV, NADH inhibited the solubilized enzyme assayed in the absence of Gpp(NH)p, in a manner similar to the membrane bound enzyme assayed
920
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
either in the presence or absence of Gpp(NH)p (figure 1, Table III). Similar to our findings with the membrane bound enzyme, the solubilized adenylate cyclase was TABLE IV The Effect of NADH on Solubilized Adenylate Cyclase Activity Derived from Rat Liver Plasma Membranes.
Incubation Conditions Basal + 0.1 mM NADH + 1.0 mM NADH + 5.0 mM NADH + 10 mM NADH + Forskolin(50 pM) + Forskolin+ 0.1 mM NADH + Forskolin+ 1.0 mM NADH + Forskolin+ 5.0 mM NADH + Forskolin+ 10 mM NADH
Adenylate Cyclase Activity (pmol/min/mgprotein) 44.0 + 0.9 31.9 + 1.1* 23.9 + 0.5* 24.1 + 0.4* 21.7 -+0.7* 321.3 + 4.5* 260.1 + 3.1" 205.0 + 1.8" 166.3 + 2.4* 144.4 + 2.8*
Adenylate cyclase was solubilized as described in Materials and Methods section and assayed in the absence of the GTP analog, Gpp(NH)p. Data presented are mean + S.E.M. of at least four determinations. *p<0.001 as compared with corresponding control. also not inhibited by NADP + (1 mM), NADPH (1 mM) or NAD + (0.5 mM) (data not shown). Additionally, as demonstrated for the membrane bound enzyme (Table III), NADH inhibited both the basal and forskolin-stimulated adenylate cyclase activity (Table IV). Moreover, irrespective of the starting activity, i.e. basal or forskolinstimulated, the maximum inhibition obtained in the presence of NADH was the same (approximately 50%) (Table IV). These data strongly indicate that NADH inhibits adenylate cyclase activity by directly interacting with the catalytic subunit. To further characterize the nature of the interaction between NADH and catalytic subunit, experiments were performed to examine the effects of NADH on solubilized adenylate cyclase activity in the presence of different ATP concentrations. In these studies the GTP analog, Gpp(NH)p was not included. Lineweaver-Burk plot representation of data from these experiments (figure 2) demonstrate that NADH inhibits adenylate cyclase by competing with the substrate ATP. NADH (1 mM) increased (p<0.001) the Km for ATP (control = 0.24 + 0.02 (n=9); +NADH = 0.44 + 0.08 mM, (n=9)) without altering the Vmax of the enzyme (figure 2). Similar results were also obtained both in the presence and absence of glucagon (10 nM) employing rat liver membrane preparations in place of the solubilized enzyme (data not shown). However, competitive inhibition of an enzyme dictates that at high concentrations of inhibitor near complete inhibition of activity should be observed. Clearly this is not the case with NADH-mediated inhibition of adenylate cyclase since the maximum inhibition obtained with NADH in either membranes (figure 1) or solubilized enzyme preparation (Table IV) was only 50%. Our observations could be explained if two different catalytic subunits of adenylate cyclase were present in the liver membrane or solubilized preparations and only one of these enzymes were inhibited by NADH. However, the existence of isozymes would predict that the Lineweaver-Burk plot would be non-linear. Since we have not observed a non-linear Lineweaver-Burk plot under any condition tested, this possibility is unlikely. However, stringent evaluations for the existence of adenylate cyclase isoforms in liver must await further investigations. The most plausible explanation of our data would be that NADH interacts at a site(s) on the catalytic subunit of adenylate cyclase that is(are) distinct from the catalytic site and doubles the Km for ATP. The resulting partial competitive inhibition (23), however, is overcome at concentrations of ATP that are greater than the Km in the presence of
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
921
°
P
0.03
~o
..= =
0.01
== E >
-5
0.01
0
!
i
i
!
z
5
10
15
20
25
I/ATP (raM "1)
FIG. 2
Effect of NADH on solubilized adenylate cyclase activity in the presence of varying concentrations of ATP. Enz;yme assays were performed with different ATP concentrations in the presence and absence of NADH. Representative experiment is shown. K m values for ATP estimated from several experiments were: Control = 0.24 + 0.02 mM (n=9); +NADH (1 mM) = 0.44 + 0.08 mM * (n=9). *p<0.001.
NADH. As described by Segel (23), when (a) the substrate (S) and the inhibitor (I) bind to the enzyme (E) at different sites to yield ES, El and ESl complexes, (b) the substrate binds to the free enzyme with a greater affinity than to the El complex, and (c) the ES and ESI complexes both yield product with equal facility, then the inhibition observed can be termed partial competetive inhibition. This model ideally fits the partial (incomplete) yet competitive inhibition of adenylate cyclase by NADH. Since the effects of NADH on hepatic adenylate cyclase can be overcome at high ATP concentrations, it may be argued that NADH-mediated regulation of this enzyme may have no physiological significance. However, it should be noted that most of the cellular adenine nucleotides are in the bound form and unavailable for enzymatic reactions (24,25). Moreover, because intracellular ATP may not be uniformly distributed (26), it is possible that ATP available for the membrane bound adenylate cyclase may be low enough for NADH to inhibit its activity. This coupled with the presence of NADH in livers at concentrations sufficient to significantly inhibit adenylate cyclase, i.e. between 0.1 and 0.2 mM (27), indicate that NADH may regulate adenylate cyclase activity. In this regard, inhibition of adenylate cyclase by NADH in the liver may explain a number of experimental findings from various laboratories. For instance, since the cAMP phosphodiesterase is known to be inhibited by NADH (8), it may be expected that in livers perfused under highly reducing conditions, e.g. lactate alone, glucagon-elicited cAMP accumulation would be very high. A similar argument would also apply to hepatocytes incubated with lactate alone. However, in our previous studies (1), when rat livers were perfused with lactate, a substrate which increases cellular NADH/NAD+ ratio, glucagon (1 nM) elevated cAMP levels to a lesser extent as compared to the cAMP accumulation observed in livers perfused with lactate plus pyruvate such that the ratio of [lactate]/[pyruvate] was held constant at 4. Similarly, in hepatocytes incubated with lactate alone the glucagon-mediated elevation of cellular cAMP content was also lower than that observed in hepatocytes incubated with lactate plus pyruvate (8). Since lactate and pyruvate do not alter the activities of adenylate cyclase (data not shown) or cAMP phosphodiesterase (8), the findings described above would suggest that an elevation in cellular NADH content
922
NADH Regulation of Hepatic Adenylate Cyclase
Vol. 49, No. 12, 1991
inhibits the adenylate cyclase activity in hepatocytes. This contention is further supported by the findings of Zederman et al. (28) that in hepatocytes incubated with the cAMP phosphodiesterase inhibitor, IBMX, and lactate, glucagon increased cAMP levels to a lesser extent than in hepatocytes incubated with IBMX alone. In summary, our data demonstrate that NADH, selectively inhibits the hepatic adenylate cyclase activity. The experimental evidence provided also indicates that NADH directly inhibits the catalytic subunit of hepatic adenylate cyclase without altering the activities of Gi or Gs. Moreover, the data suggest that NADH by interacting at a site(s) distinct from the catalytic site increases the Km for ATP, and in the presence of excess ATP the inhibition due to increased Km is overcome. Overall, these findings can explain the modulation by cellular redox potential of agonist stimulated increases in cellular cAMP content in perfused livers and hepatocytes.
A~knowledaements Supported by grants DK 35713 and DK 01742 from the NIH. T.B.P. is a recipient of NIH Research Career Development Award (DK 01742).
RefQrences 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.
H.M RASHED and T.B. PATEL, J. Biol. Chem. 262 15953-15958 (1987). H.M. RASHED, F.M. WALLER and T.B.PATEL, J. Biol. Chem. 263 5700-5706 (1988). J.G. MCCORMACK, Biochem. J. 231 581-595 (1985). J.G. MCCORMACK, Biochem. J. 231 597-608 (1985). J.G. MCCORMACK and R.M. DENTON, Biochem. J. 180 533-544 (1979). F.D. SlSTARE, R.A. PICKING and R.C. HAYNES, Jr, J. Biol. Chem. 21~0 1274412747 (1985). P.F. BLACKMORE and J.H. EXTON, J. Biol. Chem. :~(~1 11056-11063 (1986). M.G. CLARK and I.G. JARRETF, Biochem. J. 17i~ 805-816 (1978). H. LOW, and S. WERNER, FEBS Lett. 65 96-98 (1976). A.M. FELDMAN, M.A. LEVINE, K.L. BAUGHMAN and C. VAN DOP, Biochem. Biophys. Res. Commun. 142 631-637 (1987). D.M. NEVILLE, Jr., Biochem. Biophys. Acta. 154 540-552 (1968). M.M. BRADFORD, Anal. Biochem. 72248-254 (1976). Y. SALOMON, C. LONDOS and M. RODBELL, Anal. Biochem. 58 541-548 (1974). B.G. NAIR., H. M. RASHED., and T. B. PATEL, Biochem. J. 264 563-571 (1989) S. LOTERSZTAJN, C. PAVOINE, A. MALLAT, D. STENGEL, P.A. INSEL and F. PECKER, J. Biol. Chem. 262 3114-3117 (1987). E.J. NEER, J. Biol. Chem. 254 2089-2096 (1979). A.G. GILMAN, Ann. Rev. Biochem. 56 615-649 (1987). J. MOSS, and M. VAUGHN, Ann. Rev. Biochem. 48 581-600 (1979). M.B. ANAND-SRIVASTAVA, A.K. SRIVASTAVA and M. CANTIN, J. Biol. Chem. 262 4931-4937 (1987). J. WOLFF, C. LONDOS and D. M. F. COOPER, Adv. Cyclic Nucleotide Res.14 199-214 (1981). K. SEAMON and J.W. DALY, J. Biol. Chem. 256 9799-9801 (1981). T. KATADA, G.M. BOKOCH, J.K. NORTHUP, M. UI and A. G. GILMAN, J. Biol. Chem. 259 3568-3577 (1984). I. H. SEGEL, Raoid eouilibrium oartial and mixed tvoe inhibition. In "Enzvme Kinetics: Behaviour and Analv$is of Raoid Eauilibrium and Steadv State Enz;cme Systems." John Wiley & Sons, pp. 161-226, New York; Chichester; Brisbane; Toronto (1975).
Vol. 49, No. 12, 1991
NADH Regulation of Hepatic Adenylate Cyclase
923
24. R.L. VEECH, J.W.R. LAWSON, N.W. CORNELL, and H.A. KREBS, J. Biol. Chem. 254 6538-6547 (1979). 25. T.P.M. AKERBOOM, H. BOOKELMAN, P.F. ZUURENDONK, R. VAN DER MEER and J.M. TAGER, Eur. J. Biochem. 84 413-420 (1978). 26. D.S. MILLER and S.B. Horowitz, J. Biol. Chem. 261 13911-13915 (1986). 27. M. KLINGENBERG, Nicotinamide-adenine dinucleotides and dinucleotide ohosohates ~NAD. NADP. NADH NADPH~. In "Methods of Enzymatic Analysi,~," ill. U. Bergmeyer ed.) Vol. VII, pp 251-271, VCH Verlagsgesellschaft mbH, Weinheim, Germany (1985). 28. R. ZEDERMAN, H. LOW, and K. HALL, FEBS Lett. 75 291-294 (1977).