G. A. Robison

THE ROLE OF CYCLIC AMP IN THE INTERACTION OF GLUCAGON AND INSULIN IN THE CONTROL OF LIVER METABOLISM

It is now well established that insulin exerts direct effects on mammalian liver to inhibit the production of glucose and urea and to promote the uptake of potassium ions. It has been proposed that these effects of insulin may be partly due to a decrease in liver cyclic AMP.l The proposal is based on the following observations: ( 1) insulin produces a small but significant decrease in the level of cyclic AMP in the perfused rat liver; (2) depletion of insulin in vivo by treatment with insulin antiserum or alloxan results in a twofold increase in liver cyclic AMP; (3) exogenous cyclic AMP and hormones such as glucagon and epinephrine which raise the level of cyclic AMP produce effects on the liver which are opposite to those caused by insulin; (4) insulin antagonizes the actions of epinephrine, glucagon, or cyclic AMP in the perfused liver; and ( 5 ) insulin reduces the accumulation of liver cyclic AMP in the presence of glucagon. In this article we will present more recent observations on the interaction of glucagon, epinephrine, and insulin in the control of hepatic metabolism. The investigations have employed the isolated rat liver perfused by a modification of the technique of Mortimore.* The perfusion medium consisted of Krebs-Henseleit bicarbonate buffer containing 3 % bovine albumin and 20% bovine erythrocytes. Livers were from fed rats weighing 100-150 g and the perfusion flow rate was about 7 ml per minute. The perfusions were carried out in two ways. In most cases, livers were perfused for one hour with recirculating medium and hormones were infused into the portal vein at a constant rate. In these experiments, the hormone concentrations were calculated by dividing the quantity of hormone infused during the hour by the final volume of perfusate. Since this does not allow for degradation of hormone, the values are doubtlessly overestimated. In the second type of experiment livers were perfused initially for 20 minutes with recirculating media containing no additions in order to establish steady metabolic rates and levels of cyclic AMP. The perfusion system was then changed to one with nonrecirculating medium by diverting the perfusate leaving the liver into a beaker. After a six-minute control period, infusions of hormone were commenced and livers and effluent media were sampled at designated intervals. In these experi-

The Role of Cyclic AMP in the Control of Carbohydrate Metabolism

Cyclic AMP plays an important role in the regulation of metabolism generally. Emphasis in the present review has been placed on carbohydrate metabolism, but lipid metabolism has also been discussed to some extent. The chief role of cyclic AMP in several tissues seems to be to facilitate or promote the mobilization of glucose and fatty acid reserves. In the liver, glucagon and the catecholamines cause an increase in the intracellular level of cyclic AMPby stimulating adenyl cyclase. This increase in the level of .cyclic AMP leads to a net increase in hepatic glucose production by at least three mechanisms: stimulation of phosphorylase activation, suppression of glycogen synthetase activity, and stimulation of gluconeogenesis. The catecholamines also stimulate adenyl cyclase in muscle and adipose tissue. Among the principal effects of cyclic AMP in these tissues are glycogenolysis in muscle and lipolysis in adipose tissue. Another role of cyclic AMP is to enhance or promote the release of insulin from pancreatic beta cells. Insulin then travels to the liver and adipose tissue to suppress the accumulation of cyclic AMP, and may also antagonize the action of cyclic AMP in muscle. Cyclic AMP is thus seen to mediate the actions of several catabolic hormones as well as promote the release of an anabolic hormone which acts in part by opposing cyclic AMP. Since cyclic AMP is involved in the release as well as several of the actions of insulin, the possible role of cyclic AMP in diabetes has been discussed. Human diabetes mellitus is recognized as the result of a basic genetic defect, the nature of which is undefined. One line of evidence implicates basement membrane thickening as an early event in the patho genesis of diabetes. Further study of the formation and breakdown of the basement membrane may therefore lead to a better understanding of the genetic defect. Whether or not cyclic AMP plays a regulatory role in basement membrane synthesis is presently unknown. Another defect recognizable in prediabetics is faulty insulin release in response to glucose infusion. This could be secondary to basement membrane thickening, but there is also evidence that the cyclic AMP mechanism may be defective. At another level, the role of cyclic AMP is more obvious: insulin deficiency leaves unopposed the actions of hormones which stimulate the production of cyclic AMP, thereby contributing to the glucose plethora and ketosis so often seen in the later stages of the disease.

CYCLIC AMP AND THE FUNCTION OF EUKARYOTIC CELLS: AN INTRODUCTION

The chemical structure of adenosine 3’,5’-monophosphate, otherwise known as cyclic AMP, is shown in FIGURE 1. It was discovered as the mediator of the hepatic glycogenolytic effect of epinephrine and glucagon,’ and is now recognized as a versatile regulatory agent mediating a host of hormonal effects.2 Research in this area is expanding at such a rate that it is very difficult, if not impossible, for a single group to keep track of it all. Our hope for this present monograph was that it would update and supplement an earlier one,2 which we hope will continue to be useful as a source of background material. Both of these hopes will probably be realized to some extent, but the incredible pace of research in this area cannot be overemphasized. We are writing these words in October, 1970, with the conference less than a month away, and we fully expect that they will have a faintly old-fashioned ring to them by the time they are in print. Perhaps by extrapolating from what is said here and in the rest of this monograph, the serious reader will gain at least some idea of where the subject is likely to lead in the future. The structure of adenyl cyclase is still poorly understood. Information about this enzyme was summarized previously,”.’ and an important recent development is discussed in this monograph by Lefkowitz et Still more recently, Rodbell and his colleagues 6 * have made some useful contributions. There is no longer any question that the receptors for some hormones are very closely related to the adenyl cyclase system as a whole, but most of the details remain to be worked out. We still do not know how the hormone-receptor interaction leads to a change in the catalytic activity of the enzyme. To whatever extent our earlier model (FIGURE 2) represents an aspect of reality, it would appear that the catalytic and regulatory subunits do not necessarily develop at the same time. A more operationally correct way of summarizing the available data R , would be to say that the catalytic activity of adenyl cyclase and its ability to be stimulated by hormones do not necessarily develop at the same time. There is now evidence that these two components of the system can be separated,*O and further work along these lines will be watched with interest. The suggestion that GTP or GDP may play an important role in the effects of hormones on adenyl cyclase has also been of interest, but whether future research will substantiate or invalidate this hypothesis remains to be seen. The detrimental effects of cell breakage on adenyl cyclase, and especially on its sensitivity to hormonal stimulation, were emphasized in an earlier review.J However, we can now see that this may be even more variable than it was then thought to be. Platelet adenyl cyclase, for example, seems to be just as sensitive to stimulation by prostaglandins in broken cell preparations as it is in intact cells, as noted by Krishna and his colleagues.ll Brain cyclase stands at the other end of the scale, with most other cells and tissues falling some-

EFFECTS OF PROSTAGLANDINS ON FUNCTION AND CYCLIC AMP LEVELS OF HUMAN BLOOD PLATELETS

platelet adenyl cyclase. This was noted several years ago by R.E.Scott and his colleagues(Butcher et a1,1967;Scott, 1970),and has been confirmed since by at least three other groups of investigators(Wo1fe and ShulmanIl969;Marquis et al,l969;Moskowitz et a1,1970). The order of potency of different prostaglandins,to the extent that this has been studied,is the same as their order of potency as inhibitors of platelet aggregation. With the exception of the fluoride ion,to be mentioned laterlagents which do not inhibit aggregation are not known to be capable of stimulating platelet adenyl cyclase. The second line of evidence derives from studies with intact platelets. As expected from their effect on adenyl cyclase,the prostaglandins increase the steady state level of cyclic AMP in these cells(Fig.l)(Robison et a1,1969). The greater potency of PGEl over PGE2 is in line with their known potencies as inhibitors of aggregation(Weeks et all 1969;Robison et a1,1969). Other investigators(Moskowitz et all 1970;Vigdahl et all 1969)have used 3H-adenine or 14Cadenosine to label endogenous ATP,and have obtained essentially similar results: the prostaglandins were found to stimulate the incorporation of radioactivity into cyclic AMP,reflecting an increased rate of synthesis of the cyclic nucleotide in the intact cells. Conversely,several agents which stimulate platelet aggregation have been shown to produce a fall in the level of cyclic AMP in these cells. These are epinephrine (Moskowitz et alIl970;Robison et alIl969;Salzman & Neri, 1969;Marquis et a1,1970)and adenosine diphosphate(ADP), (Salzman & Neri,l969;Cole et alIl970a),both of which are thought to play an important role in regulating platelet function in vivo(Mustard ti Packham,l970). The effect of epinephrine is mediated by adrenergic a-receptors, First,the prostaglandins are potent stimulants of