Role of protein kinase C in the regulation of cytosolic Ca2+ in A431 cells: separation of growth factor and bradykinin pathways

TGF-alpha-induced breakdown of stress fibers and degradation of tropomyosin in NRK cells is blocked by a proteasome inhibitor

Treatment of NRK cells with TGF-alpha in the presence of serum initiates disassembly of cytoskeletal stress fibers and suppresses the synthesis of tropomyosin isoforms (TMs) 1, 2, and 3 but not TMs 4 and 5 (Cooper et al., Cancer Res. 47, 4493-4500, 1987). In order to determine how the loss of tropomyosin is induced and what role it plays in cytoskeletal disruption, the turnover of tropomyosin was studied in the presence of the transforming growth factor and protease inhibitors. Cells were pulse-labeled with [35S]methionine and chased in the absence or the presence of the growth factor. It was found that TMs 1, 2, and 3 are degraded at about twice the rate of TMs 4 and 5 in control cells and that the rate of degradation of TMs 1-3 is accelerated by the growth factors. Degradation of TMs in control and growth factor-treated cells is blocked by a membrane-permeable inhibitor of cysteine proteases (LLnL) that acts upon calpains and proteasomes, and the cells maintain a flattened shape with a normal complement of stress fibers. Application of inhibitors that block calpains but not proteasomes does not block TM degradation. Treatments (suspension culture or cytochalasin B) that disrupt stress fibers without application of the growth factors also accelerate TM degradation, suggesting that acceleration of TM degradation is a consequence of its release from stress fibers during their breakdown. The normally more rapid turnover of the TM isoforms 1-3 that are lost in the phenotypically transformed cells could serve to facilitate the cytoskeletal reorganization that follows the activation of signal transduction pathways by the transforming growth factors observed in this study or during other rearrangements of the cytoskeleton such as occur during cell migration or mitosis.

A minimal model for calcium signal generated by tyrosine kinase and G protein linked receptors; a stochastic computer simulation with CALSIM

A software was designed to simulate the calcium signal following hormone or growth factor stimulation in epithelial cells. The software written in C runs on a PC under Windows environment. It is based on a Markov process where the dynamic of the system is characterised by phenomenological transition probabilities. Moreover a minimal model is proposed to analyse the role of plasma channels and IP3 receptors, together with the opposite action of the CaATPase pumps, in the cytosolic and endoplasmic reticulum (ER) calcium signal control. The simulation is applied on the calcium response following stimulation by carbacol (protein G coupled receptors) or epidermal growth factor (tyrosine kinase type receptors) in A431 epithelial cells. The experimental calcium signals can be grouped in three classes; a spike and a return to the basal level (signal A), a spike and a decrease to a plateau level (signal B) or a slow increase to a plateau (signal C). Epidermal growth factor induces signal A and B while carbacol gives signal B and C. When a 'pseudo' steady state is reached oscillations occur. Computer simulations show that signal A can result from the activation of IP3 receptors while signal C would result from the activation of the plasma channels; signal B appears as the additive contribution of both channels, while oscillations are compatible with a calcium induced calcium release mechanism. Simulations suggest that the calcium dynamic in the ER is a mirror of cytosolic calcium but that a simple way to produce similar calcium elevation in these two compartments is to activate plasma channels. Implications of such a mechanism is discussed.

Corticotropin-releasing factor-like peptides increase cytosolic [Ca2+] in human epidermoid A-431 cells

This study investigated whether sauvagine and urotensin I change [Ca2+]i in human epidermoid A-431 cells and whether these changes are correlated with their anti-edema properties in vivo. A-431 cells were used because they possess the corticotropin-releasing factor (CRF) receptor 2. Treatment with either sauvagine or urotensin I led to an immediate increase in [Ca2+]i, the magnitude of which depended on the concentration of the drug. Sauvagine was more effective than urotensin I, with a median effective concentration (EC50) of 1.4 +/- 0.2 fM, compared to an EC50 of 66 +/- 7 fM for urotensin I. Both were more effective at stimulating increases in [Ca2+]i than CRF (EC50 of 6.8 +/- 0.1 pM). There was a correlation between the EC50 for increasing [Ca2+]i and the median effective dose (ED50) for inhibiting edema induced by heating rat paw (r = 0.99). Removal of extracellular Ca2+ or incubation with La3+ eliminated the increase in [Ca2+]i stimulated by either peptide. Pretreatment with a CRF receptor antagonist reduced the increase in [Ca2+]i by these peptides. This occurred in an antagonist concentration-dependent manner, with median inhibitory concentrations (IC50) of 1.99 +/- 0.04 nM and 0.85 +/- 0.04 nM, respectively. Both pertussis toxin (an inhibitor of G proteins) and U-73122 (an inhibitor for inositol trisphosphate (InsP3) production) partially inhibited the increases. InsP3 was measured to determine whether these peptides mobilized Ca2+ from an InsP3-sensitive store. Both sauvagine and urotensin I increased InsP3. The InsP3 increases were inhibited by U-73 122 and CRF antagonist, but not by removal of external Ca2+. Both peptides elevated protein tyrosine phosphorylation. In summary, these peptides increase [Ca2+]i as a result of Ca2+ influx via CRF receptor-operated Ca2+ channels coupled to pertussis toxin-sensitive G proteins and a Ca2+ mobilization from InsP3-sensitive Ca2+ pools. Their in vivo effectiveness at inhibiting edema is related to their respective capacities to stimulate elevations of [Ca2+]i, supporting a role for intracellular Ca2+ in this process.

The Ras-related protein Cdc42Hs and bradykinin promote formation of peripheral actin microspikes and filopodia in Swiss 3T3 fibroblasts

The Ras-related protein Cdc42 plays a role in yeast cell budding and polarity. Two related proteins, Rac1 and RhoA, promote formation in mammalian cells of membrane ruffles and stress fibers, respectively, which contain actin microfilaments. We now show that microinjection of the related human Cdc42Hs into Swiss 3T3 fibroblasts induced the formation of peripheral actin microspikes, determined by staining with phalloidin. A proportion of these microspikes was found to be components of filopodia, as analyzed by time-lapse phase-contrast microscopy. The formation of filopodia was also found to be promoted by Cdc42Hs microinjection. This was followed by activation of Rac1-mediated membrane ruffling. Treatment with bradykinin also promoted formation of microspikes and filopodia as well as subsequent effects similar to that seen upon Cdc42Hs microinjection. These effects of bradykinin were specifically inhibited by prior microinjection of dominant negative Cdc42HsT17N, suggesting that bradykinin acts by activating cellular Cdc42Hs. Since filopodia have been ascribed an important sensory function in fibroblasts and are required for guidance of neuronal growth cones, these results indicate that Cdc42Hs plays an important role in determining mammalian cell morphology.

Evidence for involvement of a GTP-binding protein in activation of Ca2+ influx by epidermal growth factor in A431 cells: effects of fluoride and bacterial toxins

Aluminium fluoride (AlF4-), a G protein activator, was used to study a possible role of G protein in the control of the pathways for Ca2+ influx through plasma membrane of human carcinoma A431 cells. Fluorimetric measurements with the Ca2+ indicator Indo-1 have shown that addition of fluoride induces an increase in concentration of cytosolic free calcium ([Ca2+]in) due to both release of Ca2+ from intracellular stores and Ca2+ influx from the extracellular medium. The cells stimulated by fluoride became unresponsive to subsequent addition of epidermal growth factor (EGF), histamine and bradykinin. The Ca2+ signal induced by fluoride as well as one induced by EGF was inhibited by the pretreatment of cells with protein kinase C activator, phorbol myristate acetate (PMA). The pretreatment of the cells with pertussis toxin produced no effect on EGF-induced calcium response. In contrast, the pretreatment with cholera toxin (CTX) increased the basal level of [Ca2+]in and abolished the effect of EGF. The effects of CTX could not be reproduced by treating the cells with forskolin or IBMX, agents known to elevate cAMP content in the cell. Patch clamp experiments have shown that fluoride increases the activity of Ca(2+)-permeable channels identical to those activated by EGF from the extracellular side of the membrane [Mozhayeva et al. (1991) J. Membr. Biol. 124, 113-126]. The results obtained suggest the involvement of GTP-binding protein in signal transduction from the EGF receptor to Ca(2+)-permeable channel of plasma membrane in A431 cells.

Effect of heat shock, [Ca2+]i, and cAMP on inositol trisphosphate in human epidermoid A-431 cells

The basal levels of inositol monophosphate, inositol bisphosphate, and inositol trisphosphate (InsP3) in A-431 cells incubated in Na(+)-Hanks' solution were, respectively, 1.23 +/- 0.18, 0.17 +/- 0.03, and 0.69 +/- 0.07% of the total radioactivity in the cell. When cells were heated, InsP3 increased in a temperature-dependent manner related to the duration of heating. The active form of InsP3, inositol 1,4,5-trisphosphate, increased 237 +/- 17% after heating (45 degrees C, 20 min) then returned to baseline within 15 min after the return to 37 degrees C. The heat-induced increase in InsP3 was not observed in the absence of extracellular Ca2+ or with amiloride treatment. Treatment with the nonhydrolyzable GTP analogue 5'-guanylylimidodiphosphate stimulated that component of the InsP3 increase due to G proteins. U-73122, an inhibitor of phospholipase C-mediated processes, blocked the increase in InsP3 resulting from heat exposure. Both pertussis toxin (30 ng/ml, 24 h), an inhibitor of G inhibitory protein, and cholera toxin (1 microgram/ml, 1 h), a stimulator of G stimulatory protein, increased InsP3 in unheated cells, and heating failed to induce a further increase, suggesting that heat activates G proteins. Likewise, 8-bromoadenosine 3',5'-cyclic monophosphate (8-BrcAMP), 3-isobutyl-1-methylxanthine, Ro 20-1724, or forskolin increased InsP3 in unheated cells, and heat did not cause an additional increase. The InsP3 increase induced by 8-BrcAMP was inhibited by removal of extracellular Ca2+ or treatment with verapamil, suggesting that an influx of extracellular Ca2+ stimulates InsP3 production.(ABSTRACT TRUNCATED AT 250 WORDS)

Spontaneous calcium oscillations and mechanically and chemically induced calcium responses in mammary epithelial cells

Changes of intracellular calcium activity (Ca2+i) in mouse mammary epithelial cells in primary culture (normal) and in an established cell line (MMT060562, cancerous) were investigated by microfluorometry and image analysis of fura-2 fluorescence. In both types of cells, some populations exhibited occasional Ca2+i oscillations with a period of 50-160 s. Slight mechanical stimulation of a cell with a fine glass pipette induced a Ca2+i increase, which spread from the stimulated cell to the surrounding cells with a speed of 7-12 microns/s. ATP (> 1 mumol/l) and ADP, but not AMP induced a Ca2+i increase in both cell types. Bradykinin was highly effective (> 10 nmol/l) only in the cancerous mammary epithelial cells. In Ca(2+)-free solution, all these Ca2+i responses remained unchanged at the first application, and decreased abruptly at the second trial. La3+ (> 0.5 mmol/l) suppressed the response to ATP but not the response to bradykinin. Addition of extracellular Mn2+ rapidly quenched the fura-2 fluorescence in the cell even in a non-stimulated state. Influx of Mn2+ did not increase during Ca2+i responses. These results indicate that the sources of Ca2+i responses in mammary epithelial cells are intracellular stores, which exchange Ca2+ with the extracellular medium.

Manipulation of intracellular calcium in NCB-20 cells

A number of lines of evidence indicate that the Ca2+ and cyclic AMP signalling systems interact in NCB-20 cells. However, to date, the regulation of [Ca2+]i homeostasis has not been studied in this cell line. The present study aimed to clarify our understanding of [Ca2+]i homeostasis in these cells and to evaluate tools that manipulate [Ca2+]i, independently of protein kinase C effects. Bradykinin, by a B2-receptor, elevated [Ca2+]i by a pertussis-toxin-insensitive mechanism. The BK-stimulated [Ca2+]i rise originated from intracellular sources, without a contribution from Ca2+ entry mechanisms. The effect of BK was precluded by pretreatment with thapsigargin and ionomycin--compounds that elevated [Ca2+]i independent of phospholipase C activation. Both compounds, however, exerted effects in addition to stimulating release of Ca2+ from BK-sensitive stores; the BK-sensitive Ca2+ pool was a subset of the thapsigargin-sensitive pool; ionomycin strongly stimulates Ca2+ entry. Activation of protein kinases A and C attenuated the duration of the BK-induced rise in [Ca2+]i, without affecting the peak [Ca2+]i, suggesting interference with the BK response at a step downstream of the activation of phospholipase C. Application of these approaches should enhance the delineation of the consequences of Ca2+ mobilization on cyclic AMP accumulation.

ERK1 and ERK2, two microtubule-associated protein 2 kinases, mediate the phosphorylation of tyrosine hydroxylase at serine-31 in situ

Tyrosine hydroxylase (TH) is phosphorylated at four sites in situ and in vivo, and the protein kinases that phosphorylate three of these sites (Ser8,Ser19,Ser40) have been identified. In intact cells, the phosphorylation of the fourth site (Ser31) is increased in response to phorbol esters or nerve growth factor (NGF). Here, we show that Ser31 is phosphorylated by ERK1 and ERK2, two myelin basic protein and microtubule-associated protein kinases. Extracts of NGF- or bradykinin-treated PC12 rat pheochromocytoma cells were fractionated on Mono Q columns. Protein kinase activity toward Ser31 in TH was present in two peaks corresponding to myelin basic protein kinase activities previously identified as ERK1 and ERK2. Phosphorylation of purified TH in vitro by both kinases was selective for Ser31 up to at least 0.6 mol of phosphate per mol of TH subunit. Treatment of intact PC12 cells with bradykinin or NGF increased both the phosphorylation of TH-Ser31 in situ and the catalytic activity of ERKs (measured subsequently in vitro with myelin basic protein as substrate). Pretreatment of the cells with genistein (a protein-tyrosine kinase inhibitor) decreased the bradykinin- but not the NGF-induced changes in both TH-Ser31 phosphorylation and ERK activity. Genistein also inhibited the increases in Ser31 phosphorylation produced by phorbol dibutyrate, muscarine, and Ba2+. The data indicate that ERK activity is responsible for phosphorylating TH at Ser31 in intact cells and suggest that TH-Ser31 phosphorylation may be regulated by multiple signaling pathways that converge at or prior to the activation of the ERKs.

Epidermal growth factor inhibits both cholecystokinin octapeptide-induced inositol 1,4,5-trisphosphate production and [CA2+]i increase in rat pancreatic acinar cells

We have studied the effects of epidermal growth factor (EGF) on both cholecystokinin octapeptide (CCK-OP)-induced inositol-1,4,5 trisphosphate (IP3) production and on cytosolic free calcium concentrations [Ca2+]i by fluorescence measurements in fura-2-loaded pancreatic acini. Our data show that EGF inhibits CCK-OP induced IP3 production by 40 +/- 9% and decreases CCK-OP induced rise in cytosolic Ca2+ by 41 +/- 9%. These data indicate that activation of EGF receptors leads to inhibition of CCK-OP induced stimulation of phospholipase C (PLC).

Effect of intracellular pH on cytosolic free [Ca2+] in human epidermoid A-431 cells

This study characterizes the correlation between intracellular pH (pHi) and the cytosolic free Ca2+ concentration ([Ca2+]i) in suspended and adherent human epidermoid A-431 cells. Using the fluorescent dyes 2,7-bis(carboxyethyl)carboxyfluorescein acetoxymethyl ester (BCECF) and fura-2, the resting pHi and [Ca2+]i in suspended cells were 7.23 +/- 0.03 and 209 +/- 30 nM; those in adherent cells were 7.28 +/- 0.02 and 87 +/- 5 nM. Removal of external Ca2+ did not change the resting pHi but reduced the resting [Ca2+]i, indicating the resting level of [Ca2+]i is in part maintained by an influx of Ca2+ from the external medium. When both suspended and adherent cells were acidified or alkalinized, resting [Ca2+]i was altered. An intracellular acidification induced a fall in [Ca2+]i, and a rise in pHi induced a rise in [Ca2+]i. These changes in [Ca2+]i were correlated with an uptake of 45Ca2+ from the external medium, whereas no Ca2+ efflux occurred. The alteration in [Ca2+]i induced by modification of pHi was abolished in the absence of external Ca2+ or by adding 2 mM CoCl2, LaCl3, and attenuated by the addition of 2 mM MnCl2 to the bathing medium. It was insensitive to the voltage-gated Ca2+ channel blockers nifedipine or verapamil (1 mM). CoCl2, LaCl3, and MnCl2 each induced changes in pHi and [Ca2+]i but verapamil and nifedipine did not. Because CoCl2, LaCl3, and MnCl2 are also known to block Na+/Ca2+ exchange, intracellular Na+ ([Na+]i) was measured by flame photometry in acidified or alkalinized cells. In either case no change in [Na+]i was observed. Furthermore, treatment with amiloride (100 microM), a blocker of the Na+/Ca2+ exchanger, did not inhibit the pH-induced changes in [Ca2+]i. 1,2-bis(o-Aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) (100 microM), a Ca2+ chelator, induced a decrease in pHi as well as a reduction of [Ca2+]i, also supporting the direct relation between pHi and [Ca2+]i. 3,4,5-Trimethoxybenzoic acid 8-(diethylamino)ocytl ester HCl (TMB-8) (100 microM), a known blocker of intracellular Ca2+ mobilization, did not change the resting pHi and [Ca2+]i in normal cells or cells acidified or alkalinized. This observation, taken together with data from cells incubated in the absence of external Ca2+, suggests intracellular Ca2+ pools are not involved in changes in [Ca2+]i that result from a modification of pHi. Resting pHi and [Ca2+]i in cells treated with either 8-bromo-dibutyryl cAMP (1 mM) or forskolin (150 microM) are not changed.(ABSTRACT TRUNCATED AT 400 WORDS)