The Thiol Reagent, Thimerosal, Evokes Ca2’ Spikes in HeLa Cells by Sensitizing the Inositol 1,4,5-Trisphosphate Receptor*

The thiol reagent, thimerosal, has been shown to cause an increase in intracellular CaZ+ concentration ([Ca2+Ii) in several cell types, and to cause Ca2+ spikes in unfertilized hamster eggs. Using single cell video-imaging we have shown that thimerosal evokes repet- itive Ca2+ spikes in intact Fura-2-loaded HeLa cells that were similar in shape to those stimulated by his- tamine. Both thimerosal- and histamine-stimulated Ca2+ spikes occurred in the absence of extracellular (Ca2+o), suggesting that they result from mobilization of Ca2+ from intracellular stores. Whereas histamine stimulated formation of inositol phosphates, thimerosal, at concentrations that caused sustained Ca2+ spik- ing, inhibited basal and histamine-stimulated formation of inositol phosphates. Thimerosal-evoked Ca2+ spikes are therefore not due to the stimulated produc- tion of inositol 1,4,5-trisphosphate (InsP3). The effects of thimerosal on Ca2+ spiking were probably due to alkylation of thiol groups on intracellular proteins because the spiking was reversed by the thiol-reducing compound dithiothreitol, and the latency between ad- dition of thimerosal

The thiol reagent, thimerosal, has been shown to cause an increase in intracellular CaZ+ concentration ([Ca2+Ii) in several cell types, and to cause Ca2+ spikes in unfertilized hamster eggs. Using single cell videoimaging we have shown that thimerosal evokes repetitive Ca2+ spikes in intact Fura-2-loaded HeLa cells that were similar in shape to those stimulated by histamine. Both thimerosal-and histamine-stimulated Ca2+ spikes occurred in the absence of extracellular (Ca2+o), suggesting that they result from mobilization of Ca2+ from intracellular stores. Whereas histamine stimulated formation of inositol phosphates, thimerosal, at concentrations that caused sustained Ca2+ spiking, inhibited basal and histamine-stimulated formation of inositol phosphates. Thimerosal-evoked Ca2+ spikes are therefore not due to the stimulated production of inositol 1,4,5-trisphosphate (InsP3). The effects of thimerosal on Ca2+ spiking were probably due to alkylation of thiol groups on intracellular proteins because the spiking was reversed by the thiol-reducing compound dithiothreitol, and the latency between addition of thimerosal and a rise in [Ca2+Ii was greatly shortened in cells where the intracellular reduced glutathione concentration had been decreased by preincubation with DL-buthionine (S,R)-sulfoximine. In permeabilized cells, thimerosal caused a concentration-dependent inhibition of Ca2+ accumulation, which was entirely due to inhibition of Ca2+ uptake into stores because thimerosal did not affect unidirectional 4aCa2+ efflux from stores preloaded with 46Ca2+. Thimerosal also caused a concentration-dependent sensitization of InsPs-induced Ca2+ mobilization: half-maximal mobilization of Ca2+ stores occurred with 161 f 20 nM InsPs in control cells and with 62 2 5 nM InsPB after treatment with 10 " thimerosal. We conclude that thimerosal can mimic the effects of histamine on intracellular Ca2+ spiking without stimulating the formation of InsP3 and, in light of our results with permeabilized cells, suggest that thimerosal stimulates spiking by sensitizing cells to basal InsP3 levels.
In many cell types, hormones that activate phosphoinosi- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Fax: 223-324387. 7 To whom correspondence should be addressed. Tel.: 223-336603; tidase C stimulate repetitive increases in [Ca2+],' (Woods et al., 1986;Jacob et al., 1988;Rooney et al., 1989). The frequency of these Ca2+ spikes is often dependent on the hormone concentration suggesting that they may provide a frequency encoded intracellular signaling system. Within a cell, each Ca2+ spike appears to initiate at the same site and to then propagate across it as a regenerative Ca2+ wave (Rooney et al., 1990). In Xenopus oocytes regenerative Ca2+ waves have been shown to annihilate when they collide, suggesting that the Ca2+ stores are briefly unresponsive after the passage of a wave, the resulting patterns of wave propagation are complex and unstable (Lechleiter and Clapham, 1992).
Variations of two basic models have been proposed to explain the generation of intracellular Ca2+ spikes (Berridge and Galione, 1988;Tsien and Tsien, 1990;Berridge, 1990;Jacob, 1990;Meyer and Stryer, 1991). The first suggests that Ca2+ spikes reflect the periodic formation and degradation of inositol 1,4,5-trisphosphate (InsP3), and mechanisms have been proposed that could lead to cyclical changes in intracellular InsP3 levels (Woods et al., 1987;Meyer and Stryer, 1988). In the second class of models, hormones are proposed to cause a sustained increase in the intracellular InsPB concentration and that then leads to cyclical mobilization of Ca2+ from either Imp3-insensitive or InsP3-sensitive stores (Rooney et al., 1989;Parker and Ivorra, 1990;Missiaen et al., 1991).
Although Ca2+ spiking is usually triggered by the increase in intracellular InsP3 concentration that follows receptor activation, several reports suggest that Ca2+ spikes can be evoked by agents that do not stimulate InsP, formation. Caffeine stimulates Ca*+ spikes in rat chromaffin cells even when receptor-stimulated inositol phosphate formation is inhibited by neomycin (Malgaroli et al., 1989). In parotid cells, the CaZ+-ATPase inhibitor, thapsigargin (Jackson et al., 1988), evokes Ca2+ spikes without stimulating inositol phosphates production (Foskett et al, 1991). Similar observations were obtained from hepatocytes treated with the thiol-oxidizing compound tert-butyl hydroperoxide (t-BHP) (Rooney et al., 1991). The spatial organization of the t-BHP-and receptor-stimulated Ca2+ signals in hepatocytes also appears to be similar: both are initiated at the same intracellular locus and both then propagate across the cell as a Ca2+ wave. These diverse stimuli that evoke Ca2+ spikes, caffeine, thapsigargin, t-BHP, and InsP3, share an ability to mobilize intracellular Ca2+ stores, but the way in which that triggers Ca2+ spiking is unknown.

Thimerosal Sensitizes the Imp, Receptor
Thimerosal (sodium ethylmercurithiosalicylate), a thiol alkylating agent, mobilizes Ca'+ from intracellular stores in leucocytes (Hatzelmann et al., 1990) and platelets (Hecker et al., 1989), and mimics the Ca'+ spikes that follow fertilization of hamster (Swam, 1991) and mouse' eggs. The mechanism underlying these actions of thimerosal is unknown. In the present study we have used human HeLa carcinoma cells to investigate the mechanism of thimerosal-induced Ca2+ spiking.

EXPERIMENTAL PROCEDURES
Cell Culture-HeLa cells were grown in minimal essential medium supplemented with 5% mixed serum (50% newborn calf, 50% fetal calf), 2 mM glutamine, 60 units/ml penicillin, and 50 pg/ml streptomycin. Cells were grown either on plastic dishes or, for single cell imaging studies, on glass coverslips (22-mm diameter, Chance Propper Ltd, Smethick, Warley, United Kingdom) in a humidified atmosphere (5% COP, 95% air) at 37 "C and fed daily.
To examine the effect of thimerosal on the mobilization of Ca2+ by InsP3, permeabilized HeLa cells were loaded to steady-state (5 min) with '%a2+ (2 pCi/ml) in the presence or absence of thimerosal and then added to InSPs. Incubations were terminated after 60 s by rapid filtration through Whatman GF/C filters (Bootman et al., 1992). Concentration-response curves were fitted to a logistic equation. The effect of thimerosal on unidirectional "Ca2+ efflux from permeabilized cells was examined by loading the cells to steady-state with "Ca2+, and then simultaneously adding thimerosal and removing the ATP by addition of glucose (10 mM) and hexokinase (25 units/ ml) (Taylor and Potter, 1990). The incubations were stopped at 10-5 intervals by rapid filtration. The results were fitted to a single exponential decay equation using the GraphPAD Inplot curve-fitting program (GraphPAD software, San Diego) by a nonlinear leastsquares procedure.
In all of these experiments, InsPo metabolism was negligible because even after a 5-min incubation of cells with trace amounts of [3H]Ins(1,4,5)P3 more than 95% was recovered unchanged.
Fura-2 Measurements of Intact Cells-For single cell measurements of [Caz+];, cells grown on glass coverslips were washed with extracellular medium (EM) containing (mM): NaCl, 121; KCl, 5.4; MgC1,0.8; CaC12, 1.8; NaHC03, 6; glucose, 5.5; HEPES, 25; pH 7.3. They were then loaded with Fura-2 by incubation with 1 PM Fura-2 acetoxyme-thy1 ester (Fura-2/AM) (Molecular Probes Inc.) for 30 min at room temperature (20 "C) and then washed in EM. A coverslip was mounted at room temperature on the stage of a Nikon diaphot inverted epifluorescence microscope. Fluorescent images were obtained by alternate excitation at 340 and 380 nm (40 ms at each wavelength) using either twin xenon arc lamps (Spex Industries Inc.) or a rotating filter wheel (Magical, Joyce Loebl). The emission signal was collected at 510 nm using an intensified charge-coupled device video camera 2T. R. Cheek, C. Vincent, M. H. Johnson, and M. J. Berridge, unpublished observations. (Photonic Science) and the digitized signals stored and processed as described previously (O'Sullivan et al., 1989). The fluorescence ratio was obtained at video rate and filtered with a time constant of 200 ms (Spex System) or at 3-5 intervals (Joyce Loebl System).
All experiments involving Fura-2 shown in the present study were performed at 20 "C because, although Fura-P/AM can be loaded into HeLa cells at 37 "C, the hydrolyzed dye is rapidly lost from the cells. This loss of Fura-2 at 37 "C is slowed, but not completely inhibited, by the anion transport inhibitor, sulfinpyrazone; by loading cells at 20 "C and then warming them to 37 "C in the presence of sulfinpyrazone (100 pM), it was possible to observe similar responses to both histamine and thimerosal at 37 "C.
[3H]Inositol and [3H]glycer~pho~phoin~sitol were eluted with 8 ml of 60 mM ammonium formate/5 mM NazBOr, InsPl with 8 ml of 0.2 M ammonium formate, 0.1 M formic acid, InsPZ with 8 ml of 0.5 M ammonium formate, 0.1 M formic acid, and InSP3 and InsPl were together eluted with 8 ml of 1.25 M ammonium formate, 0.1 M formic acid. The activity in each inositol phosphate fraction was expressed as a fraction of the total cell labeling.
Measurement of Cellular GSH-HeLa cells were removed from dishes and resuspended in EM at a density of 1 X lo6 cells/ml, and incubated at 37 "C in stirred glass cuvettes. The intracellular concentration of GSH was measured by following, after addition of 100 p M monochlorobimane, the rate of formation of the fluorescent GSHbimane adduct using excitation and emission wavelengths of 385 and 478 nm, respectively (Rice et al., 1986;Fernandez-Checa and Kaplowitz, 1990). Materials-Fura-2/AM was from Molecular Probes (Eugene, OR). Cell culture materials were from GIBCO. Histamine chloride, hexokinase, monochlorobimane, BSO, and saponin were from Sigma. ATP was from Boehringer. InsPo was from Dr. Robin Irvine (Babraham, United Kingdom). ' ' CaC12, [3H]inositol, and [3H]InsP3 were from Amersham.

Histamine-and Thimerosal-induced Ca'+ Spikes-Histamine and thimerosal induced Ca'+ spikes in single HeLa cells in the presence or absence of extracellular Ca'+ (Ca",) (Figs. 1 and 2 ) .
With histamine-stimulated cells, the first spike usually had a different shape and was larger than successive spikes, which had similar amplitudes and were more consistent in their shape. The frequency of spiking was lower at 1 p~ histamine (Fig. 1, c and d), compared to the response at 25 p~ (Fig. 1, a and b). When Ca'+,, was reduced below 1 pM, histamine (25 p~, Fig. le) evoked a few spikes only, but spiking was resumed when Ca'+o was elevated to 1.8 mM.
In contrast to the regular spiking behavior observed with histamine, the shape and pattern of the Ca'+ spikes induced by thimerosal were much more variable, and were dependent upon the duration of thimerosal application, its concentration, and the presence of Ca'+,, (Fig. 2 ) . For example, in the presence of Ca'+,, continuous perfusion of cells with 100 p~ thimerosal stimulated an increased [Ca'+li in all of 53 cells examined, but the Ca2+ spikes progressively broadened and the basal [Ca"], rose until individual spikes were no longer distinguishable (Fig. 2a). The same perfusion protocol in the absence of Ca'+o, however, evoked a response that was similar to that triggered by histamine, namely repetitive Ca2+ spikes of very similar shape and no increase in the basal [Ca2+]i (Fig. 2b). Another consistently successful protocol whereby the response to thimerosal closely resembled the effects of histamine on Ca2+ spiking was to perfuse cells with 100 pM thi- merosal until there was a detectable change in [Ca2+Ii, and to thereafter perfuse continuously with 1 PM thimerosal. Under these conditions, thimerosal evoked repetitive Ca2+ spikes in the presence of Ca2+, in 18 out of 28 cells. Successive spikes had similar shapes, and the spiking was sustained for more than 20 min in 16 out of the 18 cells without an elevation in the basal [Ca2+Ii (Fig. 3).
The averaged shapes of thimerosal-and histamine-stimulated Ca2+ spikes are shown in Fig. 4. These results show that thimerosal-evoked Ca2+ spikes had the same amplitudes as those evoked by histamine, with no change in the resting [Ca2+]i. However, both the rising and recovery phases of thimerosal-evoked Ca2+ spikes were slower than for histamine-evoked Ca2+ spikes. Correspondingly, the thimerosalevoked Ca2+ spikes were broader than those evoked by histamine, with half-height widths of 20 and 10 s, respectively. T h e different kinetics are particularly evident when the rates of [Ca2+Ii change are plotted either against time (Fig. 4, aii and bii) or against [Ca2+Ii (Fig. 4, aiii and biii). The phase diagrams (Fig. 4, aiii and biii), clearly indicate that for both histamine-and thimerosal-stimulated spikes, the rate of [Ca2+Ii rise is faster than the rate of recovery. A brief incubation (50 s) with 100 FM thimerosal evoked Ca2+ spikes in 12 out of 29 cells; in 50% of these cells the spiking ceased within 6 min of the removal of thimerosal (not shown). Longer perfusions with 100 PM thimerosal, which caused the sustained increase in [Ca2+Ji (Fig. 2a), were not reversed after removal of thimerosal (results not shown). Both the spiking behavior initiated by brief incubation with thimerosal and the otherwise irreversible sustained increase in [Ca2+Ii that followed prolonged incubation with 100 IM thimerosal were rapidly reversed by addition of dithiothreitol (DTT) for 2 (Fig. 5a) or 5.5 (Fig. 5b) min; the effect of D T T was itself reversible (Fig. 5, a and b). These results, and the observation that histamine was able to evoke further Ca2+ mobilization in cells which had been incubated with thimerosal ( Fig. 2b) suggest that the effects of thimerosal are not the result of irreversible damage to intracellular Ca2+ stores. Subsequent experiments were designed to address the mechanism underlying the ability of thimerosal to reversibly trigger the episodic discharge of intracellular Ca2+ stores.
The interval between the application of thimerosal and the first detectable change in [Ca2+]i varied greatly between cells (e.g. Figs. 2a and 5) and could be as long as 15 min. Both this latency and its variability between cells were significantly decreased by preincubating cells for 24 h with the y-glutamyl cysteine synthetase inhibitor, DL-buthionine (S,R)-sulfoximine (BSO, 1 mM). The interval between application of thimerosal (100 p~) and the peak of the first Ca2+ spike was 4.5 f 0.2 min ( n = 104) in control cells and 2.4 f 0.1 min ( n = 95) in cells pretreated with BSO (Fig. 6). The BSO pretreatment, which had no effect on responses to histamine, caused a 72 k 1% ( n = 4) decrease in the intracellular concentration of reduced glutathione (GSH), which is similar to its effects on other cells (Griffith and Meister, 1979;Dethmers and Meister, 1981;Shrieve et al., 1988).
Effect of Thimerosal on Permeabilized Cells-A characteristic feature of the action of thimerosal, particularly at high concentrations, is the gradual elevation of the basal [Ca2+Ii (Fig. 2a). This is consistent with earlier observations suggesting that thiol reagents inhibit Ca2+-ATPases (Jones et al., 1983;Bellomo et al., 1983;Guillemette and Segui, 1988). The effects of thimerosal on Ca2+ pumping were examined in  . 5. DTT inhibits thimerosal-induced Ca2+ spikes. Cells were perfused with Ca2+-containing EM supplemented with 100 p~ thimerosal for the periods shown. The 3 cells shown in a were briefly exposed (2 min) to 1 mM DTT for the periods shown; DTT rapidly reversed the otherwise irreverible effects of thimerosal. In b, 3 cells were exposed to DTT for a longer period (5.5 rnin), which was much longer than the normal interspike interval, and again there was complete reversal of the effects of thimerosal. From both panels, the effects of DTT itself are clearly reversible. permeabilized cells, in order to study both the uptake into, and efflux from, internal Ca2+ stores. Thimerosal caused a decrease in the steady-state Ca2+ content of permeabilized cells from 5.6 & 0.2 nmol/mg protein in control cells ( n = 3) to 3.5 f 0.2 nmol/mg protein ( n = 3) in cells treated with 10 ~L M thimerosal (Fig. 7). The unidirectional 45Ca2+ efflux from permeabilized HeLa cells was unaffected by thimerosal (10 p~; Fig. 8), suggesting that its effects on 45Ca2+ accumulation were the sole consequence of thimerosal inhibiting Ca2+ uptake. It is noteworthy that 1 p~ thimerosal, which sustained Ca2+ spiking (Fig. 3) and gave a modest sensitization of InsPsinduced Ca2+ release (below), had no effect on the steadystate Ca2+ content (5.8 f 0.2 nmol/mg protein, n = 3).
Preincubation of permeabilized cells with thimerosal (5 min) caused a concentration-dependent increase in the sensitivity of the intracellular Ca2+ stores to InsPs (2.6-fold with for the period shown. In cells depleted (by 72%) of intracellular GSH, the Caz+ rise triggered by thimerosal was both more rapid and sustained than in untreated cells (compare with Figs. 2a and 5 ) .  10 p~ thimerosal), and a modest decrease in the fraction of the stores released by a maximal InsP3 concentration ( Fig. 9; Table I). This increased sensitivity to InsP3 occurred despite the decrease in Ca2+ content of the intracellular stores, which earlier work suggests may decrease the sensitivity of the InsP3 receptor to InsPB (Nunn and Taylor, 1992). The increased sensitivity of thimerosal-treated cells to InsP3 was not a consequence of the inhibition of Ca2+ uptake because in unidirectional 45Ca2' efflux experiments, a submaximal InsP3 concentration (100 nM) stimulated a greater 45Ca2+ release after thimerosal (10 p~) treatment (Table 11).
A lower histamine concentration (1 p~) , which evoked lesser Ca2+ signals than 100 K M thimerosal, caused a %fold increase (Fig. 10). Thimerosal, at a concentration (100 p~) sufficient  to evoke substantial increases in [Ca2+Ii, produced an inhibition of both the basal and histamine-stimulated rates of inositol phosphates formation (Fig. 10). These results suggest Cells were loaded to steady-state (5 min) with or without 10 W M thimerosal. ATP was then rapidly removed by addition of glucose and hexokinase simultaneous with addition of a maximal (4 WM) or submaximal (100 nM) concentration of Inspa and the incubations were continued for 10 s. The results (mean k S.E., n = 5 for control and n = 6 for thimerosal-treated cells) show that thimerosal pretreatment increases the sensitivity of the cells to the submaximal InsP3 concentration.
'5Ca2+ release (  that the effects of thimerosal on intracellular Ca2+ cannot be a consequence of increased InsPs formation.

DISCUSSION
Histamine, acting at HI receptors (Sauvi et al., 1987;Bootman et al., 1992), evokes repetitive Ca2+ spikes in Fura-2loaded HeLa cells. This Ca2+ spiking occurs in the presence or absence of Ca2+o, suggesting that it results from the periodic release of Ca2+ from intracellular stores, but the spiking is sustained only in the presence of Ca2+o (Fig. 1). Similar changes in [Ca2+]i have been observed in many cell types after activation of receptors that lead to production of InsP3. Under appropriate conditions, the thiol reagent, thimerosal, evoked repetitive Ca2+ spikes in HeLa cells (Fig. 3). The Ca" spikes evoked by both stimuli were of a similar amplitude and shape. However, the rates of [Ca2+Ii rise and recovery of the thimerosal-evoked spikes were slower than for histamine-evoked spikes (Fig. 4). The slowing of the rate of recovery may have been due to a slight inhibition of Ca2+-ATPases, as thimerosal reduced the steady-state Ca2+ content of permeabilized HeLa cells ( Fig. 7; Table I) without affecting unidirectional Ca2+ efflux from preloaded stores (Fig. 8). We conclude that thimerosal caused a concentration-dependent inhibition of Ca2+ uptake, and since CaZ+ uptake into these stores can also be fully inhibited by the Ca2+-ATPase inhibitor thapsigargin (not shown), it seems likely that thimerosal also inhibits a Ca2+-ATPase. A similar slowing in the rate of recovery leading to the same degree of spike broadening was observed with t-BHP-evoked spikes in hepatocytes (Rooney et al., 1991). The reason why the thimerosal-evoked spikes had a slower rising phase is unclear; such a response could arise from a less synchronous release of intracellular Ca2+ stores during thimerosal-evoked Ca2+ spiking.
Thimerosal did not mimic the effect of histamine on the accumulation of inositol phosphates, indeed it substantially inhibited both the basal and histamine-stimulated rates of accumulation (Fig. 10). It seems unlikely, therefore, that an increase in InsPs production can account for the ability of thimerosal to stimulate Ca2+ spikes. The stimulation of spiking by thimerosal is likely to result from the alkylation of critical intracellular sulfhydryl groups, because it can be reversed by D T T (Fig. 5 ) , and is more rapid in onset in cells depleted of GSH by prior incubation with BSO (Fig. 6). The potentiation of thimerosal action by BSO presumably reflected the loss of the internal reducing system based on GSH that is responsible for regenerating free sulfhydryl groups.
In HeLa cells, the inhibitory effect of thimerosal on Ca2+ uptake into intracellular stores and its stimulatory effect on InsP3-induced Ca2+ mobilization could both conceivably contribute to the Ca2+ spiking observed in intact HeLa cells. Thimerosal-triggered Caz+ spikes are unlikely to result solely from its inhibitory effects on Ca2+ uptake into intracellular stores, because the Ca2+-ATPase inhibitor, thapsigargin (Jackson et al., 1988;Thastrup et al., 1990), does not evoke repetitive Ca2+ spikes in HeLa cells (results not shown). During prolonged incubations with a high thimerosal concentration in the presence of Ca2+o, successive spikes progressively broaden and eventually fuse to give a sustained rise in [Ca2+Ii (Fig. 2a). This possibly results from progressive inhibition of Ca2+-ATPases. However, a low concentration of thimerosal (1 PM) in either the presence or absence of Ca2+o, and a high concentration (100 PM) in the absence of Ca2+o, evoked spikes without a change in the basal [Ca2+Ii. This suggests that thimerosal can evoke spikes under conditions where the cells are able to regulate [Ca2+],.
It seems unlikely, that pump inhibition can account for Ca2+ spike generation, and the ability of thimerosal to evoke repetitive spikes may therefore result from sensitization of InsPs-induced Ca2+ release. Thimerosal increased the sensitivity of intracellular Ca2+ stores to InsPs by up to 2.6-fold ( Fig. 9; Table I). This sensitization was also observed in the absence of Ca2+-ATPase activity (Table 11) indicating that it was a direct effect on the Ca2+ release process and not a consequence of inhibiting the opposing action of the Ca2+ uptake pathway. A similar sensitization to InsP3 by thimerosal and GSSG has been observed in permeabilized hepatocytes Thimerosal Sensitizes the Imp3 Receptor 25119 (Missiaen et al., 1991;Renard et al., 1992). In mouse oocytes, both InsP3 and thimerosal trigger Ca2+ spiking, but only the former is inhibited by microinjection of heparin (Carroll and Swann, 1992). Our results suggesting that thimerosal increases the sensitivity of the InsP3 receptor to InsP3 provide a possible explanation for this observation because if the affinity of the receptor for InsP3 is increased by thimerosal then the effectiveness of heparin, a competitive antagonist, would be reduced. Alternatively, both InsPs and ryanodine receptors may contribute to Ca2+ spiking in mouse oocytes and thimerosal may have effects on both receptors (Swann, 1991).
The effects of thimerosal on Ca2+ spiking in HeLa cells may be a consequence of its ability to increase the sensitivity of the InsP3 receptor to InsP3 and to thereby promote Ca2+ mobilization at resting levels of intracellular InsPs. Although the degree of sensitization is relatively modest (up to 2.6-fold, Table 11), similar changes in sensitivity in hepatocytes in response to either cyclic AMP (Burgess et al., 1991;Capiod et al., 1991) or t-BHP (Rooney et al., 1991) generate Ca2+ spikes similar to those evoked by activation of receptors that stimulate InsPs formation. We conclude that thimerosal evokes repetitive Ca2+ spikes in HeLa cells that are similar to those evoked by histamine, it does so without stimulating InsP3 formation, and may do so by sensitizing the InsPs receptor to resting levels of InsP3.