Stefaan Keppens

On the role of calcium as second messenger in liver for the hormonally induced activation of glycogen phosphorylase

We have studied the mode of action of three hormones (angiotensin, vasopressin and phenylephrine, an alpha-adrenergic agent) which promote liver glycogenolysis in a cyclic AMP-independent way, in comparison with that of glucagon, which is known to act essentially via cyclic AMP. The following observations were made using isolated rat hepatocytes: (a) In the normal Krebs-Henseleit bicarbonate medium, the hormones activated glycogen phosphorylase (EC 2.4.1.1) to about the same degree. In contrast to glucagon, the cyclic AMP-independent hormones did not activate either protein kinase (EC 2.7.1.37) or phosphorylase b kinase (EC 2.7.1.38). (b) The absence of Ca2+ from the incubation medium prevented the activation of glycogen phosphorylase by the cyclic AMP-independent agents and slowed down that induced by glucagon. (c) The ionophore A 23187 produced the same degree of activation of glycogen phosphorylase, provided that Ca2+ was present in the incubation medium. (d) Glucagon, cyclic AMP and three cyclic AMP-dependent hormones caused an enhanced uptake of 45Ca; it was verified that concentrations of angiotensin and of vasopressin known to occur in haemorrhagic conditions were able to produce phosphorylase activation and stimulate 45Ca uptake. (e) Appropriate antagonists (i.e. phentolamine against phenylephrine and an angiotensin analogue against angiotensin) prevented both the enhanced 45Ca uptake and the phosphorylase activation. We interpret our data in favour of a role of calcium (1) as the second messenger in liver for the three cyclic AMP-independent glycogenolytic hormones and (2) as an additional messenger for glucagon which, via cyclic AMP, will make calcium available to the cytoplasm either from extracellular or from intracellular pools. The target enzyme for Ca2+ is most probably phosphorylase b kinase.

(3H)-vasopressin binding to isolated rat hepatocytes and liver membranes: regulation by GTP and relation to glycogen phosphorylase activation.

Specific vasopressin binding to rat hepatocytes and rat liver membranes was measured using biologically active (3H)-Tyr2-Lys8-vasopressin (8.5 Ci/mM). In both systems, vasopressin binding was found to be time-dependent, reversible, and saturable. The kinetic parameters for vasopressin binding were: apparent dissociation constants (Kd): 4.9 nM and 15 nM; maximal binding capacities: 0.83 pmoles/mg protein and 2.10(5) sites/Cell for purified membranes and intact cells respectively. The relative affinities of 19 vasopressin structural analogues were deduced from competition experiments and compared to the previously determined glycogenolytic (or antiglycogenolytic) potencies of these analogues. For both agonists and antagonists, a highly significant correlation was demonstrated between pKd and pKa (or pKi) values, suggesting that the detected binding sites are the physiological receptors involved in the glycogenolytic action of vasopressin on the rat liver. The affinity of antagonists for binding to these receptors is the same for both membranes and cells. In contrast, agonists which bind to vasopressin receptor sites have a higher affinity for purified membranes than for intact cells (Kd membranes/Kd cells = 8 +/- 1). GTP (0.1mM) reduced the affinity of agonists but not of antagonists for binding to membranes and abolished the differences between Kd values for binding to hepatocytes and membranes.

The nature of the hepatic receptors involved in vasopressin-induced glycogenolysis.

We have found a close correlation between the known vasopressor potency of arginine vasopressin and fourteen structural analogs, and the ability of these peptides to activate glycogen phosphorylase in isolated rat hepatocytes; there was no relation with the known antidiuretic activity of the analogs. We have also found that the pA2 values characterizing the known antivasopressor capacity of five analogs against vasopressin were close to those obtained for their inhibition of the vasopressin-induced activation of hepatic glycogen phosphorylase. We propose therefore that the hepatic receptors responsible for the glycogenolytic activity of vasopressin share characteristics with and appear therefore related to those responsible for pressor activity in vivo.

The activation of liver glycogen phosphorylase by vasopressin

It has been shown [ 1 ] that vasopressin stimulates glycogenolysis in the rat liver at concentrations which can occur in vivo, especially during haemorrhagic shock. We report here that the glycogenolytic effect of vasopressin is due to the activation of glycogen phosphorylase, the rate limiting enzyme in glycogen degradation. The mechanism involved appears different from that of glucagon or adrenaline: it is apparently not mediated by cyclic AMP, since in contrast with the two latter hormones no increase in protein kinase activity is observed with vasopressin. A preliminary communication describes part of these results PI.

Extracellular ATP and UTP exert similar effects on rat isolated hepatocytes

1 Extracellular UTP and ATP show obvious similarities in their control of several metabolic functions of rat isolated hepatocytes. 2 They have a similar time‐course and concentration‐dependency for the activation of glycogen phosphorylase, the generation of inositol trisphosphate (IP3), the inhibition of glycogen synthase and the lowering of adenosine 3′: 5′‐cyclic monophosphate (cyclic AMP) levels. 3 There is a similar synergism of the nucleotides with glucagon in activating phosphorylase. 4 They undergo a similar inhibition by phorbol myristic acid of their glycogenolytic effect. 5 The ATP and UTP effect on IP3 levels are not additive. 6 It is tentatively concluded that UTP and ATP use a common receptor.

The activation of liver phosphorylase b kinase by glucagon

Although cyclic AMP was discovered in studies of the mode of action of glucagon in promoting liver glycogenolysis (see [l] ), the entire sequence of events initiated by the cyclic nucleotide is less well known in liver than in muscle. Indeed, in this tissue, as has been worked out very well by Krebs and co-workers (see [2] ) cyclic AMP brings about a sequential activation of protein kinase, phosphorylase kinase and glycogen phosphorylase, leading ultimately to an enhanced rate of glycogen breakdown. In liver, the activation of glycogen phosphorylase by glucagon also involves a rise in the cyclic AMP content (see [ 11) and an activation of protein kinase [3,4] but up till now, there has only been one report describing an enhanced activity of phosphorylase kinase [5]. In the present report, we show that glucagon does indeed cause a substantial increase in the liver phosphorylase kinase activity; we show furthermore that this activation is caused by the action of the cyclic AMP dependent protein kinase.

The complex interaction of ATP and UTP with isolated hepatocytes. How many receptors

1. ATP exerts multiple receptor-mediated effects on isolated hepatocytes: glycogenolysis through the activation of glycogen phosphorylase (cAMP-independent, IP3/calcium-mediated), inactivation of glycogen synthase, inhibition of the glucagon effect on cAMP, activation of phospholipase D. The fact that some of these effects can be selectively altered and that they are not, or differently, reproduced by some other analogues of ATP, suggests the presence of more than one receptor. (i) Pertussis toxin abolishes the anti-glucagon effect of ATP without affecting its glycogenolytic effect. (ii) Single cell calcium measurements reveal major differences between ATP and ADP, (iii) 2MeSATP and ADP beta S, in clear contrast to ATP, barely increase the levels of IP3 and their glycogenolytic effects is completely blocked by phorbol ester treatment of hepatocytes. (iv) 2MeSATP differs from ADP beta S since it has no anti-glucagon effect. 2. Effects of UTP on isolated hepatocytes so far do not show any difference with effects of ATP, suggesting interaction with the same receptor(s). 3. It is proposed that liver plasma membranes contain (at least) three different receptors mediating (a) the activation of phospholipase C, (b) the activation of phospholipase D and (c) the inhibition of adenylate cyclase.

Characterization of purinoceptors present on human liver plasma membranes.

Using ATPα35S as radioligand, we have detected the presence of specific purinoceptors on human liver plasma membranes. They are characterized by a K d value of 0.19 μM and a B max of 24 pmol/mg membrane protein. These purinoceptors belong to the P2Y subclass as demonstrated by the high degree of similarity with rat liver purinoceptors, previously shown to be P2Y[(1986) Biochem. J. 240, 367–371] and known to be involved in the control of liver glycogenolysis.

Characterization of the biological effects of 2‐methylthio‐ATP on rat hepatocytes: clear‐cut differences with ATP

1 In several tissues, 2‐methylthio adenosine triphosphate (2MeSATP) is a very potent P2Y‐purine agonist. In rat hepatocytes, 2MeSATP half‐maximally activated glycogen phosphorylase at 20 nm and was therefore about 25 times more effective than ATP (Ka 0.5–0.8 μm). This strong glycogenolytic potency of 2MeSATP suggests on its own the presence of P2Y‐purinoceptors in liver. 2 Displacement of the radioligand ATPα[35S] from its receptor however occurred at much higher concentrations of 2MeSATP than was anticipated on the basis of its glycogenolytic potency. 3 The interaction of 2MeSATP with the receptor, characterized with ATPα[35S] as radioligand, cannot be considered as a pure competitive interaction. 4 2MeSATP did not share the ability of ATP to counteract the effect of glucagon on the adenosine 3′:5′‐cyclic monophosphate levels. 5 2MeSATP barely increased the levels of inositol trisphosphate (IP3). 6 The glycogenolytic effect of 2MeSATP was completely abolished by pretreatment of the hepatocytes with phorbol myristic acetate. 7 It is tentatively concluded that 2MeSATP and ATP are interacting with different P2 purinoceptors.

The somatostatin analogue TT-232 induces apoptosis in A431 cells Sustained activation of stress-activated kinases and inhibition of signalling to extracellular signal-regulated kinases

TT-232 is a somatostatin analogue containing a five-residue ring structure. The present report describes TT-232-induced signalling events in A431 cells, where a 4-h preincubation with the peptide irreversibly induced a cell death program, which involves DNA-laddering and the appearance of shrunken nuclei, but is unrelated to somatostatin signalling. Early intracellular signals of TT-232 include a transient two-fold activation of the extracellular signal-regulated kinase (ERK2) and a strong and sustained activation of the stress-activated protein kinases c-Jun NH(2)-terminal kinase (JNK)/SAPK and p38MAPK. Blocking the signalling to ERK or p38MAPK activation had no effect on the TT-232-induced cell killing. At the commitment time for inducing cell death, TT-232 decreased EGFR-tyrosine phosphorylation and prevented epidermial growth factor (EGF)-induced events like cRaf-1 and ERK2 activation. Signalling to ERK activation by FCS, phorbol 12-myristate 13-acetate (PMA) and platelet-derived growth factor (PDGF) was similarly blocked. Our data suggest that TT-232 triggers an apoptotic type of cell death, concomitant with a strong activation of JNK and a blockade of cellular ERK2 activation pathways.