Gap junctional communication modulates intercellular calcium signaling in many cell types. We have investigated whether gap junctional communication modulates calcium oscillatory behavior of cells responding to an agonist. Extracellular UTP induced calcium oscillations in 70% of HeLa cells cultured in monolayer, and neighboring cells oscillated independently of each other. In HeLa cell transfectants expressing connexin43 (HeLa/Cx43), extracellular UTP induced calcium transients, but calcium oscillations occurred in only 10% of cells. Inhibition of gap junctional communication with anandamide in HeLa/Cx43 transfectants substantially restored oscillations (55% of cells). In HeLa/Cx45 transfectants, UTP initiated calcium oscillations similar to those seen in HeLa cells (63% of cells), but HeLa/Cx46 transfectants demonstrated calcium oscillations that were dampened compared to those of the parental HeLa cells, and occurred in only 40% of cells. These experiments demonstrate that gap junctional communication modulates calcium oscillatory behavior in cell monolayers, presumably by allowing cells to share a small molecule such as inositol trisphosphate. These studies suggest that gap junctional communication may alter the nature of signals induced by calcium mobilizing agonists in a connexin-dependent fashion by modulating calcium oscillatory behavior.

Introduction

Gap junctional communication allows cells to share information in the form of aqueous solutes, and to respond to signals as a group rather than as individuals. The coordination of calcium signaling among cells has been studied by several investigators, but the role of gap junctional communication in calcium signaling remains unclear (Boitano et al., 1992; Charles et al., 1996; Finkbeiner, 1992; Guan et al., 1997; Newman and Zahs, 1997; Paemeleire et al., 2000; Sanderson et al., 1994; Sneyd et al., 1995; Tordjmann et al., 1997; Toyofuku et al., 1998). Thus while inositol (1,4,5)-trisphosphate (Ins(1,4,5)P3) can traverse at least some gap junction pores (Fry et al., 2001; Saez et al., 1989), it is unclear what conditions allow calcium signals to be propagated from cell to cell via this mechanism and which connexins mediate this kind of communication. In many instances, investigators have examined the propagation of intercellular calcium signals by stimulating single cells in monolayers and observing intercellular calcium waves that spread to neighboring cells. These mechanically induced intercellular calcium waves appear frequently to be propagated by diffusion of extracellular messengers rather than by gap junctional communication (Cotrina et al., 1995; Guan et al., 1997; Hassinger et al., 1996), particularly where Ins(1,4,5)P3-mediated release of intracellular calcium stores is required (Cao et al., 1997; Jørgensen et al., 1997).

Gap junctional communication could also modulate calcium signaling in conditions where all cells in a monolayer are subjected to the same calciotropic stimulus. For instance, variability in receptor expression levels or in the response to ligand binding could lead to a heterogeneous response by uncoupled cells in a monolayer. If this were true, then gap junctional communication among those cells could alter the response of individual cells by allowing some degree of signal averaging among cells. To test this hypothesis, we investigated whether gap junctional communication altered the calcium response to an agonist that activates an Ins(1,4,5)P3-mediated pathway in monolayers of HeLa cells and HeLa cells transfected with different gap junction proteins. Our results support the above concept, and demonstrate that different connexins differentially modulate agonist-stimulated calcium responses.

Materials and Methods

Cells and chemicals

HeLa cells were cultured in minimal essential medium (MEM; no. 11095-056; Gibco-BRL, Gaithersberg, MD) containing 10% heat-inactivated bovine calf serum (Hyclone, Logan, UT), 2 mM glutamine, 1 mM sodium pyruvate, 1% nonessential amino acids (Gibco-BRL), 5 U/ml penicillin, and 5 μg/ml streptomycin (MEM + BCS). Fura-2/AM, Lucifer Yellow, and DAPI were purchased from Molecular Probes (Eugene, OR). Other chemicals were from Sigma (St. Louis, MO).

Transfection of HeLa cells

HeLa/Cx46 (rat) transfectants have been described previously (Koval et al., 1997). HeLa/Cx43 (rat) and HeLa/Cx45 (mouse) transfectants were also generated using the calcium phosphate precipitation technique. Myc-tagged constructs were made by ligation of PCR-amplified connexin cDNAs into a pcDNA3 vector into which we had inserted the myc epitope sequence.

Calcium imaging

Fluorescence ratio imaging was performed with an IM-4000 system from Georgia Instruments (Roswell, GA). Cells plated on 25 mm diameter coverslips were loaded with 5 μM fura-2/AM for 30 minutes at 37°C, washed, and incubated in fresh medium for 15 minutes at 37°C to allow dye hydrolysis. Probenecid (2.5 mM) was used in media during loading and washing to inhibit dye efflux and compartmentalization (Di Virgilio et al., 1988). Coverslips were mounted in a PDMI-2 incubation chamber (Medical Systems, Greenvale, NY) on a Zeiss Axiovert 35, and ratio imaging was performed using two 150 Xenon light sources and monochromators set at excitation wavelengths of 340 nm and 380 nm. Images were recorded at intervals using a DAGE MTI CCD 72 camera and image intensifier. Ratio images were generated by the FL-4000 software package. Cytosolic calcium was estimated by generating a calibration curve with solutions of known calcium concentration (Molecular Probes, Eugene OR).

Dye transfer by microinjection

Chemical coupling via gap junctions was assessed by dye transfer between cells in a monolayer. Microinjection of Lucifer Yellow and DAPI were performed as previously described (Steinberg et al., 1994). Cells adherent to 25 mm coverslips were mounted in a PDMI-2 incubation chamber. Single cells were injected with Lucifer Yellow, DAPI or both, and the number of cells containing dye was monitored after 5 minutes.

Immunofluorescence microscopy and immunoblotting

Adherent cells were fixed in methanol/acetone (1:1) for 2 minutes at room temperature, and cells were prepared for immunofluorescence studies as previously described (Steinberg et al., 1994), using primary antibodies against Cx43 (Beyer and Steinberg, 1991), Cx45 (Lecanda et al., 1998), Cx46 (Koval et al., 1997) and Cy3 or rhodamine labeled goat anti-rabbit IgG. For protein immunoblots, cells were solubilized in SDS sample buffer, electrophoresed on 10% polyacrylamide gels, transferred to PVDF membranes, and incubated in anticonnexin antibodies as reported previously (Steinberg et al., 1994).

Results

UTP elicits calcium transients and asynchronous calcium oscillations in HeLa cell monolayers

Lack of intercellular transfer of microinjected dye in HeLa cells indicates that they are not coupled by gap junctions. HeLa transfectants that express different gap junction proteins, or connexins, have been used to assess the behavior of different connexins (Elfgang et al., 1995; Paemeleire et al., 2000). HeLa cells do express P2Y receptors that generate Ins(1,4,5)P3-mediated calcium transients upon addition of nucleotides to the extracellular medium (Cotrina et al., 1995; Paemeleire et al., 2000). Addition of 1 μM UTP to a monolayer of HeLa cells induced calcium transients in almost all cells in the monolayer (Fig. 1, top panel). By defining regions of interest in selected cells, we were able to determine that the majority of cells did not undergo a smooth decrement in [Ca2+]i after the initial rapid calcium spike, but instead exhibited a series of oscillations that died out over several minutes (Fig. 1, bottom panel). The oscillatory behavior in adjacent cells was not synchronized, and there did not appear to be any coordination of calcium behavior among cells in the monolayer.

Fig. 1.

UTP elicits asynchronous calcium oscillations in HeLa cells. (Top) Monolayers of HeLa cells were loaded with the calcium indicator fura-2 and 1 μM UTP was added to the medium while ratio imaging was performed. Time after UTP exposure is indicated on each panel. Scale bar: 40 μm. Oscillating cells (as detected in the bottom panel) are marked with an asterisk. (Bottom) Graph of cytosolic calcium concentration over time for single HeLa cells as shown in the top panel. An initial rise in cytosolic calcium was followed by random, asynchronous calcium oscillations in most cells.

Fig. 1.

UTP elicits asynchronous calcium oscillations in HeLa cells. (Top) Monolayers of HeLa cells were loaded with the calcium indicator fura-2 and 1 μM UTP was added to the medium while ratio imaging was performed. Time after UTP exposure is indicated on each panel. Scale bar: 40 μm. Oscillating cells (as detected in the bottom panel) are marked with an asterisk. (Bottom) Graph of cytosolic calcium concentration over time for single HeLa cells as shown in the top panel. An initial rise in cytosolic calcium was followed by random, asynchronous calcium oscillations in most cells.

HeLa cells expressing Cx43, Cx45, or Cx46 have gap junctions with differing molecular permeability

To investigate the effect of gap junctional communication mediated by different connexins on the asynchronous oscillatory responses observed above, we transfected HeLa cells with the gap junction proteins connexin43, connexin45 or connexin46. The connexin inserts were either full length or with a myc epitope tag added to the carboxyl terminus of the connexin. Results were similar for tagged or untagged connexins. These transfectants were analyzed by protein immunoblot for connexin expression along with the parental HeLa cell line (Fig. 2). The HeLa cells used for these transfections did not express endogenous Cx43 or Cx46, but a faint Cx45 band was detected on immunoblots of these cells. Thus the parental HeLa cell line did express small quantities of Cx45, although these cells were not functionally coupled, as assessed below. Each of the transfectants expressed a new immunoreactive band for the transfected protein that migrated with the appropriate molecular mass.

Fig. 2.

Protein immunoblot analyses for Cx43, Cx45 and Cx46 in HeLa transfectants. Samples from HeLa and HeLa transfectants were resolved by SDS-PAGE, electrophoretically transferred to PVDF membranes, and then probed with antisera against Cx43, Cx45 or Cx46. Migration of molecular mass standards is indicated in the figures.

Fig. 2.

Protein immunoblot analyses for Cx43, Cx45 and Cx46 in HeLa transfectants. Samples from HeLa and HeLa transfectants were resolved by SDS-PAGE, electrophoretically transferred to PVDF membranes, and then probed with antisera against Cx43, Cx45 or Cx46. Migration of molecular mass standards is indicated in the figures.

We next examined the localization of transfected protein by indirect immunofluorescence microscopy (Fig. 3). The parent HeLa cells did not stain for Cx43 or Cx46, but did display a small amount of punctate intercellular staining for Cx45 consistent with the immunoblot data above. HeLa/Cx43 and HeLa/Cx45 transfectants demonstrated linear punctate staining for the transfected connexin at the interface between adjacent cells in a characteristic gap junction staining pattern. HeLa/Cx46 transfectants also demonstrated a large amount of staining with Cx46 antiserum, but this staining was in a more diffuse linear pattern at the plasma membrane.

Fig. 3.

Immunofluorescence localization of connexins in HeLa transfectants. HeLa (a-c), HeLa/Cx43 (d), HeLa/Cx45 (e), or HeLa/Cx46 (f) cells were cultured on glass coverslips, fixed, permeabilized, and then incubated with rabbit antiserum to either Cx43 (a,d), Cx45 (b,e), or Cx46 (c,f) PBS for 1 hour at room temperature. The cells were then washed, labeled with Cy3-conjugated anti-rabbit IgG, and photographed. HeLa cells do not express detectable Cx43 (a) or Cx46 (c), and minimal Cx45 (b). In Cx43 and Cx45 transfected cells, the transfected connexins appear largely in a linear punctate pattern at cell-cell boundaries. In contrast, Cx46 appeared as diffuse linear intercellular staining in HeLa/Cx46 cells. Scale bar: 40 μm.

Fig. 3.

Immunofluorescence localization of connexins in HeLa transfectants. HeLa (a-c), HeLa/Cx43 (d), HeLa/Cx45 (e), or HeLa/Cx46 (f) cells were cultured on glass coverslips, fixed, permeabilized, and then incubated with rabbit antiserum to either Cx43 (a,d), Cx45 (b,e), or Cx46 (c,f) PBS for 1 hour at room temperature. The cells were then washed, labeled with Cy3-conjugated anti-rabbit IgG, and photographed. HeLa cells do not express detectable Cx43 (a) or Cx46 (c), and minimal Cx45 (b). In Cx43 and Cx45 transfected cells, the transfected connexins appear largely in a linear punctate pattern at cell-cell boundaries. In contrast, Cx46 appeared as diffuse linear intercellular staining in HeLa/Cx46 cells. Scale bar: 40 μm.

The function of transfected connexins was assessed by measuring the intercellular spread of dye after microinjection of the negatively charged dye Lucifer Yellow or the positively charged dye DAPI into single cells within monolayers. The results of these experiments are summarized in Table 1. The parent HeLa cells demonstrated no gap junctional communication as assessed in these experiments. All of the transfected cell lines demonstrated some degree of dye transfer, but differed in the extent of dye transfer and in the relative degree of dye transfer of the positively and negatively charged species. HeLa/Cx43 transfectants allowed the greatest degree of diffusion of the negatively charged Lucifer Yellow, but in our hands did not allow passage of the positively charged DAPI dye. In contrast HeLa/Cx45 transfectants allowed much less Lucifer Yellow transfer than did the HeLa/Cx43 transfectants, but also allowed intercellular transfer of DAPI. HeLa/Cx46 transfectants also allowed passage of both negatively charged and positively charged species, but with both tracers allowed less intercellular diffusion than did the Cx45 transfectants.

Table 1.

Gap junction molecular permeability in HeLa transfectants.

Lucifer Yellow n DAPI n
HeLa   0.5±0.7   13   0   15  
HeLa/Cx43   9.9±4.8   34   0   8  
HeLa/Cx45   2.9±2.8   47   2.1±2.0   16  
HeLa/Cx46   0.2±0.4   8   0.5±0.5   13  
Lucifer Yellow n DAPI n
HeLa   0.5±0.7   13   0   15  
HeLa/Cx43   9.9±4.8   34   0   8  
HeLa/Cx45   2.9±2.8   47   2.1±2.0   16  
HeLa/Cx46   0.2±0.4   8   0.5±0.5   13  

Adherent HeLa, HeLa/Cx43, HeLa/Cx45, and HeLa/Cx46 cells were microinjected with either 10 mM Lucifer Yellow or 15 mM DAPI. Results are expressed as the average number of neighboring cells that contained dye after 5 minutes.

To provide a more direct comparison of the ability of the transfectants to permit passage of positively charged and negatively charged dyes, we co-injected Lucifer Yellow (LY) and DAPI and assessed transfer of both dyes simultaneously (Fig. 4). For these experiments we used two independent clones each of HeLa/Cx43 and HeLa/Cx45 transfectants. Both Cx43-transfected clones transferred the negatively charged LY among cells well, averaging 13-15 cells per microinjected cell, but neither clone transferred the positively charged dye DAPI among cells. In contrast, cells of both HeLa/Cx45 clones transferred much less LY than did the Cx43 clones, and transferred DAPI more efficiently than LY. These data confirm that Cx43 and Cx45 form gap junctions that have different molecular permeabilities, and that the relative gap junctional permeability to LY compared to DAPI is greater for Cx43 channels than for Cx45 channels.

Fig. 4.

Dye transfer in HeLa/Cx43 and HeLa/Cx45 transfectants. Single cells were co-injected with Lucifer Yellow (LY) and DAPI, and dye transfer was assessed as the number of cells receiving dye per injected cell at 4 minutes. Both Cx43 transfected cell lines transferred LY well but did not pass DAPI. Conversely, both Cx45 transfected cell lines transfer DAPI better than LY.

Fig. 4.

Dye transfer in HeLa/Cx43 and HeLa/Cx45 transfectants. Single cells were co-injected with Lucifer Yellow (LY) and DAPI, and dye transfer was assessed as the number of cells receiving dye per injected cell at 4 minutes. Both Cx43 transfected cell lines transferred LY well but did not pass DAPI. Conversely, both Cx45 transfected cell lines transfer DAPI better than LY.

Mechanical stimulation of single HeLa cells elicits intracellular calcium waves that require activation of P2Y receptors but not gap junctional communication

Mechanical stimulation of single cells elicits calcium transients that in many cells propagate radially to other cells via gap junctional communication or activation of P2Y receptors by released nucleotides. We determined whether HeLa cells and the transfectants we had generated could transmit mechanically induced calcium waves, and if so, what mechanism was responsible for these waves. Monolayers of HeLa cells or transfectants were loaded with fura-2, and single cells were mechanically stimulated with a glass micropipette. In the parent HeLa cells and in all the HeLa/Cx43, HeLa/Cx45, and HeLa/Cx46 transfectants, mechanical stimulation of a single cell resulted in the propagation of calcium transients to adjacent cells, although the number of cells to which the calcium transient spread varied somewhat among the different transfectants (Fig. 5).

Fig. 5.

Mechanically induced calcium waves in HeLa and Cx-transfected HeLa cells are inhibited by desensitization of ATP receptors. Cell monolayers were loaded with fura-2 and single cells were mechanically stimulated before or 5 minutes after desensitization of P2 receptors by addition of 10 μM ATP. Numbers represent the number of cells to which the subsequent calcium wave propagated.

Fig. 5.

Mechanically induced calcium waves in HeLa and Cx-transfected HeLa cells are inhibited by desensitization of ATP receptors. Cell monolayers were loaded with fura-2 and single cells were mechanically stimulated before or 5 minutes after desensitization of P2 receptors by addition of 10 μM ATP. Numbers represent the number of cells to which the subsequent calcium wave propagated.

We next sought to determine whether the mechanism of these intercellular calcium waves required gap junctional communication or activation of P2Y receptors. Addition of the gap junction inhibitor anandamide to these cells did not block the propagation of mechanically induced calcium waves (data not shown). In contrast, desensitization of P2Y receptors by prior addition of ATP prevented calcium wave propagation in all of the cell types tested (Fig. 5). Thus, in HeLa cells expressing Cx43, Cx45 and Cx46, gap junctional communication was not able to allow propagation of calcium signals among cells when the stimulus was mechanical and of a single cell. These experiments show that in HeLa cells, as in many other cell types, propagation of mechanically induced intercellular calcium waves requires activation of P2Y receptors, presumably by released nucleotides.

Gap junctional communication mediated by Cx43 prevents UTP-induced calcium oscillations

The above experiments demonstrated that the HeLa cell transfectants expressed connexins and were coupled as assessed by dye transfer studies. They also showed that a mechanically elicited calcium response in a single cell did not propagate to surrounding cells in any of the transfectants. We next asked whether gap junctional communication altered the pattern of asynchronous calcium oscillations that occurred after addition of extracellular nucleotides to HeLa cells as demonstrated in Fig. 1. Monolayers of HeLa /Cx43 transfectants were loaded with fura-2, and 1 μM UTP was added while the cells were imaged as described previously. As before, the cells in the monolayer responded to the nucleotide by raising their intracellular calcium levels. However, the HeLa/Cx43 transfectants did not demonstrate calcium oscillations after the initial calcium rise as did the parent HeLa cells, but instead showed a smooth decay in the cytosolic calcium concentration as shown in single cell calcium traces in Fig. 6. In contrast, HeLa/Cx45 transfectants demonstrated oscillatory responses similar to those seen in the parental HeLa cells, and HeLa/Cx46 cells demonstrated oscillatory calcium behavior intermediate between that of the HeLa and HeLa/Cx43 cells.

Fig. 6.

UTP-mediated calcium responses in connexin-transfected HeLa cells. Cells were loaded with fura-2, exposed to 1 μM UTP, and ratio imaging was performed. Single cell calcium traces were constructed as above. HeLa/Cx45 transfectants exhibited asynchronous calcium oscillations similar to the parent HeLa cells. In contrast HeLa/Cx43 transfectants displayed a calcium rise that was followed by a gradual decline to baseline without much oscillatory behavior. HeLa/Cx46 transfectants displayed dampened oscillatory behavior.

Fig. 6.

UTP-mediated calcium responses in connexin-transfected HeLa cells. Cells were loaded with fura-2, exposed to 1 μM UTP, and ratio imaging was performed. Single cell calcium traces were constructed as above. HeLa/Cx45 transfectants exhibited asynchronous calcium oscillations similar to the parent HeLa cells. In contrast HeLa/Cx43 transfectants displayed a calcium rise that was followed by a gradual decline to baseline without much oscillatory behavior. HeLa/Cx46 transfectants displayed dampened oscillatory behavior.

We assessed differences in the calcium responses among the different cell lines primarily by determining the percentage of cells that demonstrated calcium oscillations, and we also measured the height of the initial calcium response in each cell and the number of calcium oscillations that occurred in any cell that demonstrated oscillatory behavior. The quantitate results shown in Table 2 confirm that the percentage of cells demonstrating oscillatory behavior was similar in HeLa and in HeLa./Cx45, markedly reduced in HeLa/Cx43 transfectants, and intermediate in HeLa/Cx46 transfectants.

Table 2.

Gap junctional communication modulates UTP-induced calcium oscillations in HeLa transfectants

1 μM UTP in MEM HeLa n=73 HeLa/Cx43 n=48 HeLa/Cx45 n=46 HeLa/Cx46 n=81
Peak [Ca2+]i (nM±s.d.)   470±280   250±90   470±200   190±80  
Number of oscillations (oscillations/5min±s.d.)   3.0±2.2   1.6±1.1   3.7±2.2   2.0±1.5  
% Cells showing oscillations   67   8   63   40  
1 μM UTP in MEM HeLa n=73 HeLa/Cx43 n=48 HeLa/Cx45 n=46 HeLa/Cx46 n=81
Peak [Ca2+]i (nM±s.d.)   470±280   250±90   470±200   190±80  
Number of oscillations (oscillations/5min±s.d.)   3.0±2.2   1.6±1.1   3.7±2.2   2.0±1.5  
% Cells showing oscillations   67   8   63   40  

Cell monolayers were loaded with fura-2 and ratio imaging was performed after addition of 1 μM UTP to the medium.

To confirm that the dampening of calcium oscillatory behavior seen in HeLa/Cx43 cells depended on gap junctional communication, we assessed UTP-induced oscillatory behavior in HeLa and HeLa/Cx43 in the presence or absence of the gap junction inhibitor anandamide. Anandamide preincubation (50 μM) almost completely inhibited transfer of LY in HeLa/Cx43 transfectants (Fig. 7). The addition of the gap junction inhibitor had little effect on the calcium responses seen in untransfected HeLa cells (Table 3). In contrast, addition of anandamide substantially restored calcium oscillatory behavior in HeLa/Cx43 cells. These experiments are therefore consistent with the hypothesis that Cx43-mediated gap junctional communication is responsible for the absence of calcium oscillations seen after addition of UTP to HeLa/Cx43 cells.

Fig. 7.

Anandamide inhibits Cx43-mediated dye transfer. HeLa/Cx43 cells were microinjected with Lucifer Yellow in the presence or absence of anandamide, and dye transfer was assessed. Scale bar: 40 μm.

Fig. 7.

Anandamide inhibits Cx43-mediated dye transfer. HeLa/Cx43 cells were microinjected with Lucifer Yellow in the presence or absence of anandamide, and dye transfer was assessed. Scale bar: 40 μm.

Table 3.

The gap junction inhibitor anandamide restores calcium oscillations in HeLa/Cx43 cells

HeLa n=27 HeLa+ anandamide n=38 HeLa/Cx43 n=27 HeLa/Cx43+ anandamide n=51
Peak [Ca2+]i (nM±s.d.)   400±150   370±150   180±60   280±100  
Number of oscillations (oscillations/5 min±s.d.)   4.5±1.9   4.4±1.0   2.2±0.8   1.4±0.6  
% Cells showing oscillations   70   67   10   55  
HeLa n=27 HeLa+ anandamide n=38 HeLa/Cx43 n=27 HeLa/Cx43+ anandamide n=51
Peak [Ca2+]i (nM±s.d.)   400±150   370±150   180±60   280±100  
Number of oscillations (oscillations/5 min±s.d.)   4.5±1.9   4.4±1.0   2.2±0.8   1.4±0.6  
% Cells showing oscillations   70   67   10   55  

Cells were loaded with fura-2 and response to addition of 1 μM UTP in the presence or absence of 50 μM anandamide was assessed by ratio imaging. Anandamide had no effect on the calcium responses in the parental HeLa cells, but substantially restored oscillatory behavior in HeLa/Cx43 cells.

Modulation of agonist response by gap junctional communication varies by agonist concentration

Temporal variability in calcium responses may signal differences in agonist concentration, and may therefore be a way of sensing differences in agonist concentration. We therefore assessed calcium responses in HeLa transfectants over a range of agonist concentrations (Fig. 8). For these experiments, fura-2-loaded cells were exposed to different concentrations of UTP while ratio imaging was performed. All cells in the field were analyzed and scored as no response, partial response (submaximal transient, no oscillations), maximal response (maximal transient, no oscillations) or oscillation (maximal transient with subsequent oscillations). The untransfected HeLa cells and the HeLa/Cx43 transfectants responded similarly to the submaximal dose of 0.1 μM UTP: in both cell types, about 20% of cells displayed oscillatory behavior, and about 30% of cells had a maximal rise without oscillations. The only difference between the two cells was that all of the HeLa/Cx43 transfectants had at least some response to 0.1 μM UTP whereas 15% of HeLa cells did not respond at all.

Fig. 8.

Gap junction-mediated effects on calcium oscillations are dependent on the dose of agonist applied. Monolayers of HeLa cells and transfectants were loaded with fura-2 and calcium response to different concentrations of UTP was assessed by video imaging. Single cell traces were categorized as no response, submaximal rise, maximal response without oscillations, or response with oscillations, as described in the text.

Fig. 8.

Gap junction-mediated effects on calcium oscillations are dependent on the dose of agonist applied. Monolayers of HeLa cells and transfectants were loaded with fura-2 and calcium response to different concentrations of UTP was assessed by video imaging. Single cell traces were categorized as no response, submaximal rise, maximal response without oscillations, or response with oscillations, as described in the text.

All cells responded to 1 μM UTP, but as described above, a high percentage of HeLa cells demonstrated oscillatory behavior, whereas few of HeLa/Cx43 transfectants did. As the dose of UTP was increased to 10 and 100 μM, the percentage of HeLa cells that oscillated decreased to 30%, and in the HeLa/Cx43 transfectants the percentage of oscillating cells decreased further to zero. Thus the ratio of oscillating to maximal responding cells varied in an agonist-dependent fashion in HeLa cells over a broad range of UTP concentrations, but HeLa/Cx43 cells demonstrated significant oscillatory behavior only at submaximal UTP concentrations. We also performed this analysis on HeLa/Cx45 and HeLa/Cx46 transfectants. As anticipated from the above results, the HeLa/Cx45 transfectants responded in a manner similar to that of HeLa cells with the percentage of oscillating cells declining to about 20% with 100 μM UTP. The HeLa/Cx46 cells appeared to be less affected by agonist concentration, with about 50% of the oscillating cells remaining with concentrations of 1-100 μM UTP. These results demonstrate that gap junctional communication mediated by different connexins alters the calcium responses in HeLa cells over a broad range of agonist concentrations. Put another way, transfection with different gap junction proteins altered agonist concentration sensing in HeLa cells in dramatically different fashion.

Discussion

These studies demonstrate that gap junctional communication mediated by different connexins alters the nature of calcium signaling in response to a calcium mobilizing agonist. HeLa cells, which expressed a barely detectable amount of Cx45 and were not coupled as assessed by our dye transfer assays, demonstrated asynchronous calcium oscillations that occurred in most cells at near maximal agonist concentration and occurred less frequently at higher or lower agonist concentrations. In contrast, HeLa cells transfected to express Cx43 did not demonstrate significant oscillatory behavior in response to UTP except at low agonist concentrations, and even then it was only seen in a minority of the cells. Inhibition of gap junctional communication restored asynchronous oscillatory behavior in these cells, confirming the role of Cx43-mediated gap junctional communication in the inhibition of calcium oscillatory behavior. HeLa/Cx45 transfectants showed a minimal decrease in oscillatory behavior, and HeLa/Cx46 transfectants demonstrated moderate dampening of calcium oscillations after stimulation with UTP.

Many of the previous studies that have examined the role of gap junctional communication in intercellular calcium signaling have focused on the role of gap junctions in mediating the propagation of calcium signals induced in a single cell to its surrounding cells. Frequently mechanical stimulation of a single cell has been used to induce a calcium response in the first cell. In the current studies, mechanically stimulated HeLa cells transfected with Cx43, Cx45 or Cx46 were not able to propagate calcium transients to adjacent unstimulated cells after desensitization of purinergic receptors. However, gap junctional communication in connexin transfectants was able to change the pattern of the response to a calciotropic stimulus when this stimulus was applied to all cells simultaneously. The primary effect of gap junctional communication was to inhibit the generation of asynchronous oscillations that occurred in most of the HeLa cells after addition of agonist. Calcium oscillatory behavior is understood to be important for the initiation of down stream activities that follow calcium mobilization (Berridge, 1997).

It is probable that the differences in calcium response to the agonist UTP seen in HeLa cells and HeLa/Cx43 cells are accounted for by the ability of Cx43 gap junctions to allow the diffusion of a small signaling molecule among cells, thereby lowering its concentration in the cell in which it was originally generated. Although the dye transfer studies presented here show that the molecular permeability of gap junctional channels formed by Cx43, Cx45 and Cx46 are different for LY and DAPI as has been previously shown by others (Elfgang et al., 1995), it is difficult to extrapolate from these findings to the nature of this signaling molecule.

Because of its role in calcium signaling, inositol trisphosphate (Ins(1,4,5)P3) is a good candidate for the signaling molecule in question. Ins(1,4,5)P3 has been demonstrated to traverse gap junctions as demonstrated by the presence of Ins(1,4,5)P3-mediated calcium transients in neighboring cells after microinjection of Ins(1,4,5)P3 or caged Ins(1,4,5)P3 (Fry et al., 2001; Niessen et al., 2000; Saez et al., 1989). Ins(1,4,5)P3 permeability is likely to be more similar to LY permeability than DAPI permeability. Niessen and co-workers studied the diffusion of Ins(1,4,5)P3 through channels formed of Cx26, Cx23, Cx43 and found that these gap junctions differ in their ability to pass Ins(1,4,5)P3 (Niessen et al., 2000). The idea that inositol trisphosphate, rather than Ca2+, influences intercellular calcium signaling is supported by studies in hepatocytes (Clair et al., 2001).

Although connexins are present in most cells and gap junctional communication is widespread, the biological function of this type of communication among cells is frequently not well understood. These studies demonstrate a way in which gap junctional communication modifies the calcium response to an agonist to which all cells in a monolayer are capable of responding independently. Because the spatial and temporal patterns of calcium response can largely determine the downstream effects of calcium mobilization (Berridge, 1997), the connexin profile of a tissue may be an important determinant of how that tissue will respond to a calciotropic agonist.

Acknowledgements

This work was supported by NIH grants HD37799, DK46686 and GM54660.

References

Berridge, M. J. (
1997
). The AM and FM of calcium signalling.
Nature
386
,
759
-760.
Beyer, E. C. and Steinberg, T. H. (
1991
). Evidence that the gap junction protein connexin-43 is the ATP-induced pore of mouse macrophages.
J. Biol. Chem.
266
,
7971
-7974.
Boitano, S., Dirksen, E. R. and Sanderson, M. J. (
1992
). Intercellular propagation of calcium waves mediated by inositol trisphosphate.
Science
258
,
292
-295.
Cao, D., Lin, G., Westphale, E. M., Beyer, E. C. and Steinberg, T. H. (
1997
). Mechanisms for the coordination of intercellular calcium signaling in insulin-secreting cells.
J. Cell Sci.
110
,
497
-504.
Charles, A. C., Kodali, S. K. and Tyndale, R. F. (
1996
). Intercellular calcium waves in neurons.
Molec. Cell. Neurosci.
7
,
337
-353.
Clair, C., Chalumeau, C., Tordjmann, T., Poggioli, J., Erneux, C., Dupont, G. and Combettes, L. (
2001
). Investigation of the roles of Ca(2+) and InsP(3) diffusion in the coordination of Ca(2+) signals between connected hepatocytes.
J. Cell Sci.
114
,
1999
-2007.
Cotrina, M. L., Lin, J. H., Alves-Rodrigues, A., Liu, S., Li, J., Azmi-Ghadimi, H., Kang, J., Naus, C. C. and Nedergaard, M. (
1995
). Connexins regulate calcium signaling by controlling ATP release.
Proc. Natl. Acad. Sci. USA
95
,
15735
-15740.
Di Virgilio, F., Steinberg, T. H., Swanson, J. A. and Silverstein, S. C. (
1988
). Fura-2 secretion and sequestration in macrophages. A blocker of organic anion transport reveals that these processes occur via a membrane transport system for organic anions.
J. Immunol.
140
,
915
-920.
Elfgang, C., Eckert, R., Lichtenberg-Frate, H., Butterweck, A., Traub, O., Klein, R. A., Hulser, D. F. and Willecke, K. (
1995
). Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells.
J. Cell Biol.
129
,
805
-817.
Finkbeiner, S. (
1992
). Calcium waves in astrocytes–filling in the gaps.
Neuron
8
,
1101
-1108.
Fry, T., Evans, J. H. and Sanderson, M. J. (
2001
). Propagation of intercellular calcium waves in C6 glioma cells transfected with connexins 43 or 32.
Microsc. Res. Tech.
52
,
289
-300.
Guan, X., Cravatt, B. F., Ehring, G. R., Hall, J. E., Boger, D. L., Lerner, R. A. and Gilula, N. B. (
1997
). The sleep-inducing lipid oleamide deconvolutes gap junction communication and calcium wave transmission in glial cells.
J. Cell Biol.
139
,
1785
-1792.
Hassinger, T. D., Guthrie, P. B., Atkinson, P. B., Bennett, M. V. L. and Kater, S. B. (
1996
). An extracellular signaling component in propagation of astrocytic calcium waves.
Proc. Natl. Acad. Sci. USA
93
,
13268
-13273.
Jørgensen, N. R., Geist, S. T., Civitelli, R. and Steinberg, T. H. (
1997
). ATP-dependent and gap junction-dependent intercellular calcium signaling in osteoblastic cells.
J. Cell Biol.
139
,
497
-506.
Koval, M., Harley, J. E., Hick, E. and Steinberg, T. H. (
1997
). Connexin46 is retained as monomers in a transGolgi compartment of osteoblastic cells.
J. Cell Biol.
137
,
847
-857.
Lecanda, F., Towler, D. A., Ziambaras, K., Cheng, S.-L., Koval, M., Steinberg, T. H. and Civitelli, R. (
1998
). Gap junctional communication modulates gene expression in osteoblastic cells.
Molecular Biol. Cell
9
,
2249
-2258.
Newman, E. A. and Zahs, K. R. (
1997
). Calcium waves in retinal glial cells.
Science
275
,
844
-847.
Niessen, H., Harz, H., Bedner, P., Kramer, K. and Willecke, K. (
2000
). Selective permeability of different connexin channels to the second messenger inositol 1,4,5-trisphosphate.
J. Cell Sci.
113
,
1365
-1372.
Paemeleire, K., Martin, P. E., Coleman, S. L., Fogarty, K. E., Carrington, W. A., Leybaert, L., Tuft, R. A., Evans, W. H. and Sanderson, M. J. (
2000
). Intercellular calcium waves in HeLa cells expressing GFP-labeled connexin 43, 32, or 26.
Mol. Biol. Cell
11
,
1815
-1827.
Saez, J. C., Connor, J. A., Spray, D. C. and Bennett, M. V. L. (
1989
). Hepatocyte gap junctions are permeable to the second messenger, inositol 1,4,5-trisphosphate, and to calcium ions.
Proc. Natl. Acad. Sci. USA
86
,
2708
-2712.
Sanderson, M. J., Charles, A. C., Boitano, S. and Dirksen, E. R. (
1994
). Mechanisms and function of intercellular calcium signaling.
Molec. Cell. Endrocrinol.
98
,
173
-187.
Sneyd, J., Wetton, B. T., Charles, A. C. and Sanderson, M. J. (
1995
). Intercellular calcium waves mediated by diffusion of inositol trisphosphate: a two-dimensional model.
Am. J. Physiol.
268
,
C1537
-C1545.
Steinberg, T. H., Civitelli, R., Geist, S. T., Robertson, A. J., Veenstra, R. D., Wang, H.-Z., Warlow, P. M., Hick, E., Laing, J. G. and Beyer, E. C. (
1994
). Connexin43 and connexin45 form gap junctions with different molecular permeabilities in osteoblastic cells.
EMBO J.
13
,
744
-750.
Tordjmann, T., Berthon, B., Claret, M. and Combettes, L. (
1997
). Coordinated intercellular calcium waves induced by noradrenaline in rat hepatocytes: dual control by gap junction permeability and agonist.
EMBO J.
16
,
5398
-5407.
Toyofuku, T., Yabuki, M., Otsu, K., Kuzuya, T., Hori, M. and Tada, M. (
1998
). Intercellular calcium signaling via gap junction in connexin-43-transfected cells.
J. Biol. Chem.
273
,
1519
-1528.