Cell cycle regulation of the G0/G1 transition in 5-fluorouracil-sensitive and -resistant human colon cancer cell lines.

PURPOSE: Resistance to 5-fluorouracil (5-FU) has been associated with thymidylate synthase (TS) gene amplification and increased TS protein levels. Increased TS protein expression has also been found to be a significant independent prognostic factor for disease-free survival and overall survival in patients treated with adjuvant 5-FU-based chemotherapy. In these studies and in our prior preclinical studies, TS has been considered a marker of proliferative capacity. The purpose of the current study was to further evaluate the association between TS levels and cell cycle regulation, by investigating cell cycle kinetics in a 5-FU-resistant cell line with constitutive overexpression of TS. The influence of increased TS levels on cell cycle progression may provide insight into methods to overcome 5-FU resistance. MATERIALS: 5-FU-sensitive NCI H630(WT) and 5-FU-resistant NCI H630(R1) (with 15- to 20-fold higher TS protein levels) were utilized in this investigation to determine the influence of constitutive overexpression of TS on cell cycle kinetics. RESULTS: There was no apparent influence of increased TS levels on cell cycle distribution during asynchronous growth, and both cell lines reach plateau growth phase in 120 hours, arresting in G0/G1 as determined by flow cytometry. In the H630(WT) cells, this G0/ G1 arrest was associated with a 14- to 17-fold reduction in TS activity and protein levels (using the TS-106 monoclonal antibody), whereas in the H630(R1) cells, only a two- to fivefold reduction was noted. Flow cytometry analysis utilizing Ki-67 indicated that there was no evidence of a G0 population in the confluent H630(R1), whereas 26% +/- 7% of confluent H630(WT) cells were Ki-67 negative (G0) and the remainder had low Ki-67 signal intensity. Analysis of pRb phosphorylation and p16 and p21 expression suggested that the arrest point for both cell lines was before the point at which Rb phosphorylation takes place, yet the confluent H630(R1) cells had threefold higher p21 than confluent H630(WT) cells. DISCUSSION: These data suggest that the 5-FU-resistant H630(R1) cell lines arrest at a later point in G0/G1 and have a potentially greater capacity for proliferation.

In vivo modulation of iododeoxyuridine metabolism and incorporation into cellular DNA by 5'-amino-5'-deoxythymidine in normal mouse tissues and two human colon cancer xenografts.

This in vivo study examines the ability of 5'-amino-5'-deoxythymidine (5'-AdThd) to modulate 5-iododeoxyuridine (IdUrd) cellular metabolism in two human colon cancer xenografts (HT 29 and HCT-116), two actively proliferating normal mouse tissues (bone marrow and intestine), and a quiescent normal mouse tissue (liver). 5'-AdThd is a thymidine analogue that at low concentrations (<30 micrometer) can increase thymidine kinase activity, which is the rate-limiting enzyme for activation of IdUrd. We reported recently that the in vitro incubation of HT 29 and HCT-116 cells in 5'-AdThd + IdUrd resulted in an enhancement of 5-iodo-2'-dUTP pools, IdUrd DNA incorporation, and subsequent radiosensitization compared with incubation with IdUrd alone (Clin. Cancer Res., 1: 407-416, 1995). These in vitro effects were more significant in the radioresistant cell line HT 29. Using a 6-day continuous infusion of IdUrd (50 or 100 mg/kg/day) and/or 5'-AdThd (200 mg/kg/day), no increase in systemic toxicity (percentage of body weight loss) was observed in athymic nude mice with 5'-AdThd alone or when combined with IdUrd. There was significant dose-dependent, systemic toxicity with IdUrd, which was reversible within 3 days of completing the lower-dose IdUrd infusion. However, a comparison of plasma levels during the 6-day continuous infusion of IdUrd +/- 5'-AdThd showed a significant interaction of IdUrd and 5'-AdThd, resulting in higher plasma levels by day 6 of both compounds and the principal metabolites, iodouracil and deoxyuridine, which is consistent with nonlinear saturating effects on dihydrouracil dehydrogenase. Coadministration of IdUrd and 5'-AdThd resulted in an increase in the percentage of IdUrd DNA incorporation in the two proliferating normal tissues, which was significant only with the lower IdUrd dose. No effect on IdUrd DNA incorporation was found in normal liver at either IdUrd dose +/- 5'-AdThd. Similar to our in vitro data, the continuous infusion of IdUrd and 5'-AdThd showed a significant effect by increasing the percentage of IdUrd DNA incorporation in HT-29 xenografts at both IdUrd doses, whereas coadministration of 5'-AdThd had no such effect in HCT-116 xenografts.

Increased thymidylate synthase protein levels are principally associated with proliferation but not cell cycle phase in asynchronous human cancer cells.

We have analysed cell cycle variations in thymidylate synthase (TS) protein in asynchronously growing NCl H630 and HT 29 colon cancer and MCF-7 breast cancer cell lines. Western immunoblot analysis using the TS 106 monoclonal antibody revealed a 14- to 24-fold variation in TS levels between the peak exponential and confluent growth phase in the three cell lines. Similar variations in TS levels and TS activity were detected using the 5-fluorodeoxyuridine monophosphate and deoxyuridine monophosphate biochemical assays. The percentage of cells in S-phase, which paralleled changes in TS levels, reached a maximum of 38-60% in asynchronous exponentially growing cells compared with 5-10% in confluent cells. In asynchronous exponential cells, analysis of TS levels in each cell cycle phase using two-parameter flow cytometric analysis revealed that TS protein levels were 1.3- to 1.5-fold higher in S than in G0/G1 phase cells, and 1.5- to 1.8-fold higher in G2/M than G0/G1 cells. Similar differences of 1.1- to 1.5-fold between G0/G1 and S-phase and 1.6- to 1.9-fold between G0/G1 and G2/M-phase were detected by Western immunoblot and biochemical assays. TS protein was not detectable by Western blot analysis, flow cytometry or biochemical analysis in the G0/G1 population of confluent cells. Twenty-six per cent of cells in this population were G0 cells compared with 2% in exponentially growing cells. In contrast to TS, a 4-fold difference in thymidine kinase (TK) was detected between G0/G1 and S-phase cells in exponentially growing MCF-7 cells. The level of TS enzyme is associated with cellular proliferation and the percentage of cells in S-phase; however, TS protein is not exclusively associated with S-phase in asynchronously growing cells. The variation in TS levels between exponentially growing and confluent cell population appears to be due to differences in TS levels between G0 and G1 cells.

Turnover of inositol pentakisphosphates, inositol hexakisphosphate and diphosphoinositol polyphosphates in primary cultured hepatocytes.

We have used a non-transformed cell model, the primary cultured hepatocyte, to explore the turnover of inositol hexakisphosphate, multiple isomers of inositol pentakisphosphate and two novel diphosphoinositol polyphosphates. All of these compounds gradually accumulated radioactivity throughout a 70 h period of labelling with [3H]inositol. However, a rapid metabolic rate was revealed upon inhibition of diphosphoinositol polyphosphate biphosphatase(s) with 1 mM fluoride for 40 min: this treatment elevated levels of [3H]diphosphoinositol polyphosphates up to 10-fold, indicating that their cellular pools were normally turning over at least 10 times every 40 min. This was accompanied by a turnover of about 10% of the pool of inositol hexakisphosphate. Control experiments established that 200 nM vasopressin brought about a typical activation of phospholipase C in hepatocytes after 62 h of primary culture. This agonist treatment did not affect steady-state levels of [3H]inositol pentakisphosphates, [3H]inositol hexakisphosphate or [3H]diphosphoinositol polyphosphates. However, prolonged treatment of hepatocytes with 2 microM thapsigargin reduced steady-state levels of [3H]diphosphoinositol polyphosphates by 50-70%. This effect of thapsigargin was also observed in the presence of fluoride, indicating that thapsigargin inhibited the rate of synthesis of diphosphoinositol polyphosphates.

In situ imaging of agonist-sensitive calcium pools in AR4-2J pancreatoma cells. Evidence for an agonist- and inositol 1,4,5-trisphosphate-sensitive calcium pool in or closely associated with the nuclear envelope.

The activation of phospholipase C by hormones and neurotransmitters activates a complex combination of Ca2+ release and accumulation by intracellular organelles. Previously, we demonstrated that, in some cell types, the fluorescent Ca2+ indicator, fura-2, can be loaded into intracellular, agonist-sensitive Ca2+ pools (Glennon, M. C., Bird, G. St. J., Kwan, C.-Y., and Putney, J. W., Jr. (1992) J. Biol. Chem. 267, 8230-8233). In the current study, we have attempted to exploit this phenomenon by employing digital fluorescence imaging of compartmentalized fura-2 to investigate the localization and function of the major intracellular sites of Ca2+ regulation in AR4-2J pancreatoma cells. By judicious use of a surface receptor agonist together with the Ca(2+)-ATPase inhibitor, thapsigargin, cellular regions were identified whose behavior indicates that they contain the sites of agonist- and inositol 1,4,5-trisphosphate-mediated intracellular Ca2+ release. These regions were located throughout the cell and may include the nuclear envelope. They were distinct in locus and behavior from two other regions, which counterstained with fluorescent markers for nuclei and mitochondria. Fura-2 in mitochondrial regions reported low resting levels of [Ca2+], and revealed that organelles in these regions accumulate and retain Ca2+ after agonist activation. These findings demonstrate that fluorescent Ca2+ indicators can be employed to directly monitor changes in [Ca2+] in the major Ca(2+)-regulating organelles, and provide the first in situ visualization and localization of the major sites of Ca2+ regulation in cells.

Actions of vasopressin and the Ca(2+)-ATPase inhibitor, thapsigargin, on Ca2+ signaling in hepatocytes.

When hepatocytes were loaded with fura-2 by incubation with the acetoxymethyl ester (fura-2/AM), addition of Mn2+ resulted in a rapid quench of a fraction of cellular fura-2 fluorescence. Addition of vasopressin caused a second, rapid quench of cellular fura-2, whereas the addition of thapsigargin had no effect. When hepatocytes were loaded by microinjection of fura-2 acid, addition of Mn2+ caused a slower, sustained rate of quench, and both vasopressin and thapsigargin increased this rate of quench. When Mn2+ was removed from the medium of fura-2/AM-loaded cells after preincubation with Mn2+, vasopressin still caused quench of cellular fura-2. In contrast, neither vasopressin nor thapsigargin increased fura-2 quench when Mn2+ was removed from fura-2-injected cells. When fura-2/AM-loaded cells were permeabilized with saponin, only a fraction of the cell-associated fura-2 was quenched by addition of Mn2+. A second fraction was then quenched by addition of inositol 1,4,5-trisphosphate. These results indicate that in hepatocytes loaded with the acetoxymethyl ester of fura-2, the increased quench of cellular fura-2 seen with phospholipase C-linked agonists is not due to effects of the agonist on Mn2+ entry across the plasma membrane, but rather is due to agonist activation of Mn2+ penetration into an intracellular organelle, presumably through inositol 1,4,5-trisphosphate-regulated channels. Thus, it appears that compartmentalization of fura-2 accounts for previously reported anomalies in Ca2+ signaling in hepatocytes, such as the apparent failure of Ca(2+)-ATPase inhibition to increase divalent cation entry, as well as the apparent ability of phospholipase C-linked agonists to stimulate efflux of Ca2+.

Malonyl-CoA and long chain acyl-CoA esters as metabolic coupling factors in nutrient-induced insulin secretion.

Several approaches were used to test the hypothesis proposing a role for acyl-CoA esters in nutrient-induced insulin release (Prentki, M., and Matschinsky, F. M. (1987) Physiol. Rev. 67, 1185-1248; Corkey, B. E., Glennon, M. C., Chen, K. S., Deeney, J. T., Matschinsky, F. M., and Prentki, M. (1989) J. Biol. Chem. 264, 21608-21612). Exogenous saturated long chain fatty acids markedly potentiated glucose-induced insulin release and elevated long chain acyl-CoA esters in the clonal beta-cell line (HIT). The secretory action depended on the fatty acid chain length, occurred in the range 3-20 microM (free concentration of palmitate), and was reversible and inhibitable by the neuromodulator somatostatin. 2-Bromopalmitate, an inhibitor of carnitine palmitoyl transferase I, suppressed the oxidation of endogenous fatty acids and promoted release of insulin. Only the nutrients or the combination of nutrients that caused secretion elevated malonyl-CoA. The short-chain acyl-CoA profile of HIT cells stimulated by various nutrients was determined in the presence of the nonstimulatory fuel glutamine. Glucose and leucine each provoked similar changes in acyl-CoA compounds. Both secretagogues elevated malonyl-CoA 3-6-fold, whereas succinyl-CoA, free CoASH, acetyl-CoA, and the free CoASH to acetyl-CoA ratio remained unaltered. Furthermore, only when inhibition of fatty acid oxidation was associated with a rise in malonyl-CoA did the total (mitochondrial plus cytoplasmic) content of long chain acyl-CoA esters correlate inversely with insulin release promoted by various nutrients. The results are consistent with the concept that fuel stimuli cause a rise in malonyl-CoA which by inhibiting fatty acid oxidation increase cytosolic long chain acyl-CoA esters. These data provide further support for a model in which malonyl-CoA and long chain acyl-CoAs esters serve as metabolic coupling factors when pancreatic beta-cells are stimulated with glucose and other nutrient secretagogues.

Purification of an inositol (1,3,4,5)-tetrakisphosphate 3-phosphatase activity from rat liver and the evaluation of its substrate specificity.

Hepatic inositol (1,3,4,5)-tetrakisphosphate 3-phosphatase activity was detected in a 100,000 x g soluble fraction and a detergent-solubilized particulate fraction. Activity in both fractions increased up to 40-fold after anion-exchange chromatography due to removal of endogenous inhibitors (Hodgson, M.E., and Shears, S.B. (1990) Biochem. J. 267, 831-834); at this stage the detergent-solubilized particulate activity comprised over 90% of total activity. The particulate phosphatase was further purified by affinity chromatography using heparin-agarose and red-agarose. The latter column resolved two peaks of enzyme activity (designated 1 and 2 by their order of elution from the column). Their proportions varied between experiments, but peak 2 generally predominated and so this was further purified by hydroxylapatite chromatography. The final preparation was typically 38,000-fold purified with a 7% yield. The apparent molecular mass of this enzyme was 66 kDa, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and gel filtration. The enzyme had little or no affinity for the following: inositol (1,3,4,6)-tetrakisphosphate, inositol (1,3,4)-trisphosphate, inositol (1,3)-bisphosphate, inositol (3,4)-bisphosphate, and para-nitrophenylphosphate. At pH 7.4 the Km for inositol (1,3,4,5)-tetrakisphosphate was 130 nM and the Vmax was 4250 nmol/mg protein/min. The purified enzyme also dephosphorylated inositol (1,3,4,5,6)-pentakisphosphate to inositol (1,4,5,6)-tetrakisphosphate (Km = 40 nM, Vmax = 211 nmol/mg protein/min), and inositol hexakisphosphate to at least five isomers of inositol pentakisphosphate (Km = 0.3 nM, Vmax = 12 nmol/mg protein/min). The latter affinity is the highest yet defined for an enzyme involved in inositol phosphate metabolism. Determinations of IC50 values, and Dixon plots, revealed that with the (1,3,4,5)-tetrakisphosphate as substrate, the pentakis- and hexakisphosphates were potent competitive inhibitors; the Ki values (25 and 0.5 nM, respectively) were similar to their substrate Km values. The kinetic properties of this enzyme, as well as estimates of the cellular levels of its potential substrates, indicate that inositol pentakisphosphate and inositol hexakisphosphate are likely to be the preferred substrates in vivo.

A role for malonyl-CoA in glucose-stimulated insulin secretion from clonal pancreatic beta-cells.

To gain insight into the relationship between acyl coenzyme A (CoA) esters and glucose-induced insulin release, acyl-CoA profiles were determined in clonal pancreatic beta-cells (HIT). A high sensitivity high performance liquid chromatography method was used to measure malonyl, succinyl, beta-hydroxy beta-methylglutaryl and acetyl-CoA esters and free CoASH. Malonyl-CoA content increased more than 3-fold following exposure of HIT cells to 10 mM glucose. The rise in malonyl-CoA, which preceded insulin secretion, was evident 2 min after exposure to glucose and was sustained for at least 30 min. The increase in malonyl-CoA was associated with inhibition of fatty acid oxidation, increased de novo lipid synthesis and a rise in diacylglycerol content. Succinyl-CoA levels, which may reflect anaplerotic influx into the citric acid cycle, were elevated in the presence of glucose. The concentration of acetyl-CoA and the ratio of free CoASH to acetyl-CoA was unchanged. The data are consistent with a metabolic model in which malonyl-CoA mediates the switch from fatty acid catabolism to lipid synthesis during glucose stimulation of beta-cells. We suggest that these changes in lipid metabolism, by leading to increased diacylglycerol synthesis or protein acylation could play a pivotal role in the regulation of the sustained phase of insulin secretion.

Selective activation of Ca2+ influx by extracellular ATP in a pancreatic beta-cell line (HIT).

The action of exogenous ATP on cytoplasmic free Ca2+ ([Ca2+]i) was studied in insulin secreting cells using fura-2. Stimulation of clonal pancreatic beta-cells (HIT) with ATP (range 2-20 microM) evoked a sustained elevation in [Ca2+]i. ATP selectively promoted Ca2+ influx and not Ca2+ mobilization since (1) the effect required external Ca1+ and (2) was observed in cells in which internal stores were depleted with ionomycin (3) the rate of Mn2+ influx, measured as the quenching of the fura-2 signal, was accelerated by ATP. The action of ATP was unaffected by the voltage-sensitive Ca2+ channel blockers nifedipine and verapamil as well as by a depolarizing concentration of K+. The effect on [Ca2+]i was highly specific for ATP since AMP, ADP, adenosine 5'-[gamma-thio]triphosphate, adenosine 5'-[beta, gamma-methylene]triphosphate, GTP and adenosine were ineffective. In normal pancreatic islet cells, both exogenous ATP (range 0.2-2 microM) and ADP induced a transient Ca2+ elevation that did not require external Ca2+. The nucleotide specificity of the effect on [Ca2+]i suggests that ATP activates P2 gamma purinergic receptors in normal beta-cells. Thus, ATP evokes a Ca2+ signal in clonal HIT cells and normal islet cells by different transducing systems involving distinct purinoreceptors. A novel mechanism for increasing [Ca2+]i by extracellular ATP is reported in HIT cells, since the nucleotide specificity and the selective activation of Ca2+ influx without mobilization of internal Ca2+ stores cannot be explained by mechanisms already described in other cell systems.

Relationship between unusual hepatic acyl coenzyme A profiles and the pathogenesis of Reye syndrome.

This study examines the relationship between impaired fatty acid oxidation and the pathogenesis of Reye syndrome. We present a hypothesis proposing that many clinical signs of this childhood disease are caused by accumulation of unusual acyl CoA esters, precursors to deacylated metabolites found in the patients' blood and urine. A new method was developed to measure acyl CoA compounds in small human liver biopsy samples, offering several advantages over previous techniques. A major finding was an accumulation in Reye syndrome patients of short- and medium-chain acyl CoA intermediates of fatty acid and branched-chain amino acid oxidation. These metabolites included octanoyl, isovaleryl, butyryl, isobutyryl, propionyl, and methylmalonyl CoA esters. The findings were explained in a model of hepatic fatty acid oxidation involving three interrelated pathways: mitochondrial beta-oxidation, peroxisomal beta-oxidation, and omega-oxidation in the endoplasmic reticulum. The results suggest that pathogenesis in Reye syndrome stems from generalized mitochondrial damage resulting in accumulation of acyl CoA esters. High levels of these compounds lead to inhibition of mitochondrial pathways for ureogenesis, gluconeogenesis, and fatty acid oxidation. The inhibited pathways, in turn, could cause the hyperammonemia, hypoglycemia, and hypoketonemia observed in patients. The model also explains underlying biochemical differences between patients with Reye syndrome and medium-chain acyl CoA dehydrogenase deficiency, another disorder of fatty acid metabolism. Acetyl CoA levels, in the latter disease, were dramatically decreased, compared with both human controls and Reye syndrome patients.

Cell-specific patterns of oscillating free Ca2+ in carbamylcholine-stimulated insulinoma cells.

The effect of the muscarinic agonist carbamylcholine on cytoplasmic Ca2+ concentration ([Ca2+]i was examined at the single cell level in clonal pancreatic beta-cells (HIT). Cells were loaded with the indicator dye fura 2, and [Ca2+]i was measured by microfluorimetry. Carbamylcholine induced changes in Ca2+ that differed from cell to cell and provoked in some cells oscillatory Ca2+ fluctuations. During a transient, free Ca2+ rose to a peak within 1-3 s. The frequency of the oscillations increased with agonist concentration. Oscillations in [Ca2+]i occurred in the absence of external Ca2+. When cells were perifused for a sufficient period of time without carbamylcholine, near identical Ca2+ responses were elicited in each cell by successive applications of the agonist. Thus, individual cells displayed characteristic and reproducible Ca2+ responses with respect to amplitude, frequency, and shape of the transients as well as latency in onset of the initial Ca2+ rise. We propose that the biological response to a Ca2+ agonist in a given cell is not only determined by the frequency and amplitude of Ca2+ oscillations but is governed by the unique pattern of the Ca2+ signal of each cell, which may be termed "Ca2+ fingerprint."

Regulation of steady-state free Ca2+ levels by the ATP/ADP ratio and orthophosphate in permeabilized RINm5F insulinoma cells.

Stimulation of insulin secretion in the pancreatic beta-cell by a fuel such as glucose requires the metabolism of the fuel and is accompanied by increases in oxygen consumption and intracellular free Ca2+. A very early signal for these events could be a decrease in the cytosolic ATP/ADP ratio due to fuel phosphorylation. To test this hypothesis the regulation of free Ca2+ was evaluated in permeabilized RINm5F insulinoma cells that sequester Ca2+ and maintain a low medium free Ca2+ concentration (set point), between 100 and 200 nM, in the presence of Mg2+ and ATP. ATP, creatine, creatine phosphate, and creatine phosphokinase were added to the media to achieve various constant ratios of ATP/ADP. Free Ca2 was monitored using fura-2. The results demonstrated that the steady-state free Ca2+ concentration varied inversely with the ATP/ADP ratio and orthophosphate (Pi) levels. In contrast, no correlation between free Ca2+ and the phosphorylation potential (ATP/ADP.Pi) was found. Regulation of the Ca2+ set point by the ATP/ADP ratio was observed at ratios between 5 and 50 and at Pi concentrations between 1 and 7 mM, irrespective of whether mitochondria were participating in Ca2+ sequestration or were inhibited. Increasing the ATP/ADP ratio stimulated Ca2+ uptake by the nonmitochondrial pool but did not modify Ca2+ efflux. Glucose 6-phosphate (1 mM) had no effect on the Ca2+ set point. The data suggest that variations in the cytosolic ATP/ADP ratio induced by fuel stimuli may regulate Ca2+ cycling across nonmitochondrial compartments and the plasma membrane by modulating the activity of Ca2+ -ATPases. A mechanism linking fuel metabolism and cytosolic ATP/ADP ratio to activation of the Ca2+ messenger system in pancreatic beta-cells is proposed.

Linked oscillations of free Ca2+ and the ATP/ADP ratio in permeabilized RINm5F insulinoma cells supplemented with a glycolyzing cell-free muscle extract.

We show in the accompanying paper that the steady-state level of free Ca2+ maintained by the organelles of permeabilized RINm5F insulinoma cells varies inversely with the ATP/ADP ratio when this ratio is set by addition of creatine phosphokinase and fixed ratios of creatine to creatine phosphate. We, therefore, asked whether acute cyclic alterations in the cytosolic ATP/ADP ratio in the range known to modulate O2 consumption might be involved in regulating the physiological activity of Ca2+ -ATPases and the cytosolic free Ca2+ level. To explore this hypothesis we combined two experimental systems: 1) permeabilized RINm5F insulinoma cells that can maintain a low medium Ca2+ concentration and 2) a cell-free extract of rat skeletal muscle that spontaneously exhibits oscillatory behavior of glycolysis and linked oscillations in the ATP/ADP ratio, when provided with glucose. The free Ca2+ level maintained by the permeabilized cells oscillated in phase with the glycolytic oscillations and correlated closely with the ATP/ADP ratio but not with glucose 6-phosphate, fructose 6-phosphate, orthophosphate, or pH. When glucokinase replaced hexokinase as the glucose phosphorylating enzyme, Ca2+ oscillations were induced by increasing the glucose concentration from 2 to 8 mM. The results demonstrate a link between metabolite changes and free Ca2+ levels in a reconstituted physiological system. They support a model in which oscillations in glycolysis and the ATP/ADP ratio may cause oscillations in cytosolic free Ca2+, beta-cell electrical activity, and insulin release.

Lowering pH in blood platelets dissociates myosin phosphorylation from shape change and myosin association with the cytoskeleton.

Platelet shape change induced by ADP is relatively independent of external pH over the range 6-7. If the chloride ion in the buffer is replaced by weak acids, however, shape change is rapidly and reversibly inhibited as a function of lowered pH (92% at pH 6.0). This inhibition is correlated with lowered internal pH caused by the weak acids, as measured by the 5,5-dimethyloxazolidine 2,4-dione technique. Shape change was 50% inhibited at internal pH 6.4 when 50 mM NaCl was replaced by propionate (PR). When platelets were stimulated with ADP 10-20 s after addition of PR to a final pH of 6 (PR6), both myosin light chain (MLC) phosphorylation and myosin and actin association with the cytoskeleton were reduced in correlation with the inhibition of shape change. But when ADP was added 30 s after PR6, the MLC phosphorylation was essentially the same in PR or in chloride, although shape change and myosin and actin association with the cytoskeleton remained inhibited. This was shown to be due mainly to endogenous phosphorylation of MLC. On return to neutral pH, platelets in PR immediately changed shape and myosin and actin became associated with the cytoskeleton. Two-dimensional tryptic peptides of MLC showed two major spots after PR6 treatment, indicating that both the MLC kinase site and the protein kinase C sites were phosphorylated. The results show that increased internal pH is not required for shape change, although it may affect the rate. In PR6, as after phorbol esters, MLC phosphorylation can be uncoupled from shape change. The association of myosin and actin with the cytoskeleton is closely correlated with shape change, suggesting that shape change requires the active interaction of these contractile proteins.

Cyclic AMP raises cytosolic Ca2+ and promotes Ca2+ influx in a clonal pancreatic beta-cell line (HIT T-15).

The effect on cytosolic Ca2+ concentration ([Ca2+]i) of cAMP analogues and the adenylate cyclase-stimulating agents forskolin, isoproterenol and glucagon has been examined in an insulin-secreting beta-cell line (HIT T-15) using fura 2. All these manipulations of the cAMP messenger system promoted a rise in [Ca2+]i which was blocked by the Ca2+ channel antagonists verapamil and nifedipine or by removal of extracellular Ca2+. The action of the adenylate cyclase activator forskolin was glucose-dependent. The results suggest that cAMP elevates [Ca2+]i in HIT cells by promoting Ca2+ entry through voltage-sensitive Ca2+ channels, not through mobilization of stored Ca2+. Activation of Ca2+ influx may be an important component of the mechanisms by which cAMP potentiates fuel-induced insulin release.