Anti-Ig-induced calcium influx in rat B lymphocytes mediated by cGMP through a dihydropyridine-sensitive channel.

In contrast to excitable tissues where calcium channels are well characterized, the nature of the B lymphocyte calcium channel is unresolved. Here, we demonstrate by single cell analysis of freshly isolated rat B cells that the anti-immunoglobulin (Ig)-induced calcium influx takes place through a channel which shares pharmacologic and serologic properties with the L-type calcium channel found in excitable tissues. It is sensitive to the dihydropyridines nicardipine and Bay K 8644, to calciseptine, and to an anti-peptide antibody raised against the alpha1 subunit of the L-type calcium channel, but is voltage-insensitive. Anti-alpha1 and anti-alpha2 antibodies stain B but not T lymphocytes. Application of a cGMP agonist, measurement of cGMP levels in anti-Ig-stimulated B cells, and examining the effect of a guanylyl cyclase inhibitor on the anti-Ig response show that cGMP mediates the influx. This possibly involves a cGMP-dependent protein kinase. The anti-Ig-induced response is not abolished by prior treatment of B cells with a high dose of thapsigargin. These findings undermine the widely held belief of a categorical divide between excitable and non-excitable tissue calcium channels, demonstrate the limitations of the capacitative calcium influx theory, and point to a distinction between the calcium response mechanisms utilized by B and T lymphocytes.

In contrast to excitable tissues where calcium channels are well characterized, the nature of the B lymphocyte calcium channel is unresolved. Here, we demonstrate by single cell analysis of freshly isolated rat B cells that the anti-immunoglobulin (Ig)-induced calcium influx takes place through a channel which shares pharmacologic and serologic properties with the L-type calcium channel found in excitable tissues. It is sensitive to the dihydropyridines nicardipine and Bay K 8644, to calciseptine, and to an anti-peptide antibody raised against the ␣ 1 subunit of the L-type calcium channel, but is voltage-insensitive. Anti-␣ 1 and anti-␣ 2 antibodies stain B but not T lymphocytes. Application of a cGMP agonist, measurement of cGMP levels in anti-Ig-stimulated B cells, and examining the effect of a guanylyl cyclase inhibitor on the anti-Ig response show that cGMP mediates the influx. This possibly involves a cGMP-dependent protein kinase. The anti-Ig-induced response is not abolished by prior treatment of B cells with a high dose of thapsigargin. These findings undermine the widely held belief of a categorical divide between excitable and non-excitable tissue calcium channels, demonstrate the limitations of the capacitative calcium influx theory, and point to a distinction between the calcium response mechanisms utilized by B and T lymphocytes.
Cross-linking of the B cell antigen receptor can lead to the generation of intracellular signals. The earliest detected biochemical event is the tyrosine phosphorylation of intracellular proteins (1). This is due to the activation of a number of cytoplasmic tyrosine kinases (2). The activation of Ras, phosphatidylinositol 3-kinase, and phosphoinositide signaling pathways are among the most significant consequences of tyrosine phosphorylation (3), with the latter leading to an increase in cytosolic calcium levels (4). Calcium is the most common signal transduction element in cells. It affects the activity of various enzymes and helps to regulate universal processes such as cell growth and development (5). In order to sustain a high level of intracellular calcium during activation, calcium conductance through the B cell plasma membrane is transiently increased (6). However, unlike the case in excitable tissues, where fine structural and functional details of numerous channels have been unraveled (7), the mechanism of calcium influx in B cells has not been identified. Bearing in mind the conservation of calcium channel structure in excitable tissues, we hypothesized the potential presence of homologous structures in B lymphocytes. Single cell fluorescence imaging techniques, flow-cytometry, and radioimmunoassays were used to explore this possibility. Here, we present our findings about the nature of the B cell calcium channel and its mode of regulation.

EXPERIMENTAL PROCEDURES
B Cell Purification-Lymphocytes freshly isolated from the cervical and mesenteric lymph nodes of 8 -14-week-old non-immunized inbred PVG rats (8) were used in a double rosetting procedure by depletion of CD4 ϩ and CD8 ϩ cells (9)  2) (Sigma) was used both for loading and carrying out the calcium assays. Free cytosolic calcium concentration was determined as described previously (10). Images were collected every 12 s. All assays were conducted for a 12-20-min period (unless specified in the text). Basal calcium levels were in the order of 140 -180 nM.
cGMP Assay-The cGMP assay was performed using a radio-immunoassay kit (Amersham Corp.) according to Doshi et al. (16). Protein was measured by the method of Lowry et al. (17), as modified by Miller (18).

RESULTS
Calcium Response to Anti-Ig Antibodies-B cells loaded with fura-2 AM were stimulated with an anti-IgM (10 g/ml) or anti-IgD (10 g/ml) antibody without further cross-linking. The antibodies used have been shown to be mitogenic. The stimu-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ To whom correspondence should be addressed.
lation led to a substantial increase in calcium levels ( Fig. 1, a and b). The response was transient, and the amplitude of the response in different cells varied. When monitored for a longer period of time (90 min), anti-Ig stimulation was shown to cause an oscillatory response in individual cells (data not shown).
Repeating the experiments under low extracellular calcium conditions practically eliminated the response (Fig. 1c), demonstrating an essential role for extracellular calcium.
The Effect of Calcium Channel Modulators-Next, the effect of calcium channel blockers on the calcium response was examined. The presence of nicardipine (10 M) (19) completely abolished the expected anti-Ig-induced calcium response (Fig.  1d). The experiment was then repeated using calciseptine (20), a peptide derived from black mamba venom, which is considered a specific blocker of L-type calcium channels. At 10 M, calciseptine inhibited the response as effectively as nicardipine (Fig. 1e). In another assay, the agonist Bay K 8644 (10 M) (21) was utilized. This dihydropyridine has been shown to bind the open state of L-type calcium channels and prolong their open time. Adding Bay K 8644 to the B cells after initiation of the anti-Ig-induced response led to maintenance of the calcium level at the peak (Fig. 1f). The calcium channel blocker -conotoxin (10 M) (22) had no effect on the anti-Ig-induced calcium response (data not shown). These results underscore the dihydropyridine sensitivity of the B cell calcium channel.
Surface Staining of Lymphocytes with Anti-channel Antibodies-To determine the possible presence of structures homologous to the L-type calcium channels on B cells, rat lymphocytes were labeled with polyclonal anti-peptide antibodies raised against sequences from the dihydropyridine-sensitive ␣ 1D subunit and the calcium channel ␣ 2 subunit (Fig. 2, a and  b). A biotinylated antilight chain antibody was employed to mark B cells within the whole lymphocyte population. Twocolor flow-cytometric analysis revealed surface membrane staining of B lymphocytes by both antibodies. The fluorescence intensity was compatible with the normally low density (roughly 1 m Ϫ2 ) (23) of ion channel expression. In contrast, neither of the two antibodies detectably stained T lymphocytes (Fig. 2,  a and b), pointing to a fundamental difference in the molecules involved in calcium entry in the two lymphocyte subsets.
Functional Effect of the Anti-␣ 1D Antibody-It has been shown previously that anti-␣ 1 antibodies can inhibit calcium currents in myocytes (24). On this basis, we examined the possible effect of the anti-␣ 1D antibody on the anti-Ig-induced calcium response. The presence of the anti-␣ 1D antibody completely blocked the response (Fig. 2c). Presence of the preimmune serum at the time of anti-Ig stimulation had no such effect. This emphasizes the direct involvement of an ␣ 1 -like structure in regulating anti-Ig-stimulated calcium influx in B cells.
The Effect of Depolarization-L-type calcium channels are voltage-sensitive. Upon depolarization, the ␣ 1 subunit S 4 segments are thought to sense the change in voltage. This leads to a conformational change in the channel, which results in calcium influx (7). To investigate the voltage-sensitivity of the B cell calcium response, we exposed these cells to KCl (50 mM) or gramicidin A (1 M). As in a previous report (25), and unlike excitable cells, depolarization of B lymphocytes did not lead to a calcium response (data not shown).
The Role of cGMP-We then turned to investigating the potential involvement of a second messenger in inducing the calcium response. Our point of departure was the chance finding that soluble NO can induce a transient B cell calcium

B Lymphocyte Calcium Channel 7298
response similar to that induced by the anti-immunoglobulins (Fig. 3a). NO is known to induce an increase in cGMP by activating the soluble forms of guanylyl cyclase (26). We therefore examined the effect of the analogue 8-bromo-cGMP on B cells. Stimulation of B cells with the analogue (10 M) led to a calcium response (Fig. 3b). The response was transient, with cells showing notable heterogeneity in the onset times of their responses. The analogue-induced response required extracellular calcium (Fig. 3c) and was blocked by nicardipine (10 M) (Fig. 3d).
Two separate approaches were used to examine the possible mediation of the anti-Ig-induced response by cGMP. First, we investigated the effect of anti-Ig stimulation on the B cell cGMP levels. B lymphocytes were stimulated with anti-IgD (10 g/ml) for 2 or 5 min. cGMP levels were then estimated by radioimmunoassay and compared to those of unstimulated samples. The stimulation led to an approximate 2.5-fold increase in B lymphocyte cGMP levels (Fig. 3e). In another approach, the effect of manipulating guanylyl cyclase on the anti-Ig-induced response was examined. B lymphocytes were stimulated with anti-IgD (10 g/ml) in the presence of LY83583 (10 M) (27). Treatment of B cells with this guanylyl cyclase inhibitor nearly abolished the anti-Ig-induced calcium influx (Fig. 3f). cGMP is therefore implicated in the anti-Ig-induced calcium response.
To define the target of cGMP in this pathway, we examined the effect of a nonspecific inhibitor of cGMP-dependent protein kinase I␣ on the calcium response (28). The presence of Rp-8-pCPT-cGMPS (100 M) led to a delayed and diminished calcium response to anti-Ig stimulation (Fig. 3g), indicating the possible involvement of a cGMP-dependent protein kinase in the B cell calcium influx.
The endogenous production of NO in human B cell lines has been documented (29). However, neither the addition of NO synthase inhibitors N G -methyl L-arginine (30) or N G -nitro-Larginine-methyl ester (31) to B cells a few minutes prior to anti-Ig stimulation nor the overnight incubation of the B cells with these inhibitors (500 M) had any effect on the anti-Iginduced calcium response (data not shown). This would suggest that endogenous NO is not involved in the coupling of surface immunoglobulin to guanylyl cyclase.
The Effect of Thapsigargin-Capacitative calcium entry has been shown to occur upon treatment of cells, not only with receptor agonists but also with endoplasmic reticulum Ca 2ϩ -ATPase blockers such as thapsigargin (32).
To investigate the possible role of this mechanism in the B lymphocyte calcium response, the cells were treated with thapsigargin. As shown in Fig. 4a, treatment of B lymphocytes with thapsigargin (1 M) leads to a substantial increase in cytosolic calcium levels. The profile of the response is different from that induced by anti-Ig stimulation, with calcium levels maintained at a raised plateau throughout the period of experimentation. The elevated plateau reflects an increased calcium influx through the cell membrane. We studied the effect of LY83583 on this response. B lymphocytes were stimulated with thapsigargin (1 M) in the presence of LY83583 (10 M). Under such circumstances, the thapsigargin-induced response remained unaffected (Fig. 4b). This indicates the lack of cGMP involvement in the thapsigargin-induced calcium response. Pretreatment of most cell types with high concentrations of thapsigargin abolishes Ca 2ϩ release by their receptor agonists (33). B cells were stimulated with anti-IgD (10 g/ml) after treatment with 10 M of thapsigargin. In spite of the high concentration of thapsigargin used in the experiment, the addition of anti-IgD could still induce a significant additional increase in the B lymphocyte calcium levels (Fig. 4c). DISCUSSION The sensitivity of the anti-Ig response to dihydropyridines and the anti-␣ 1D antibody and the serological evidence for the existence of ␣ 1 and ␣ 2 -like subunits in B cells indicate the existence of functional and structural homology between the B lymphocyte channel and the L-type calcium channel found in excitable tissues. The results obtained with the calcium channel blockers can be used to argue the existence of a structure similar to L-type calcium channels as the B cell calcium channel.
We sought to verify this by an independent approach. The results obtained with the anti-channel antibodies adequately serve this purpose. Of these, the data obtained with the anti-␣ 1D antibody is of paramount significance. This antibody is targeted against a sequence residing between the S 5 and S 6 segments of the ␣ 1 subunit domain IV, an area thought to form part of the transmembrane pore and the dihydropyridine binding site. While the staining of B cells with the antibody signifies

B Lymphocyte Calcium Channel 7299
the serologic similarity of the B cell structure with the L-type calcium channel, its ability to block the anti-Ig-induced calcium response indicates its functional similarity to that structure. However, the calcium response in B cells is not voltagesensitive. As already mentioned, treatment of these cells with KCl (50 mM) or gramicidin A (1 M) gave no calcium response. Moreover, surface staining of excitable cells with anti-␣ 1D , or reduction of their calcium conductance by this antibody, is dependent on prior depolarization with KCl, suggesting that in these cells the antibody's target epitope is only exposed following a conformational change in the channel as a result of depolarization (34,35). We have found this to be unnecessary for B cells. The B cell calcium channel could therefore be a dihydropyridine-sensitive complex devoid of a membrane voltage sensor. Two models in the literature serve as precedents to this hypothesis (36,37). (i) The Drosophila melanogaster trp gene product has a sequence similarity to the ␣ 1 subunit of L-type calcium channels but lacks the charged residues in the S 4 segment (36); (ii) a murine erythroleukemia cell line expresses a truncated form of the ␣ 1 subunit, in which the first four transmembrane segments are absent (37). This would enable it to interact with dihydropyridines without having a voltage-dependent gating pattern.
Our results demonstrate a role for cGMP in B cell calcium influx. In theory, this might be by (i) direct cGMP gating of the channel or (ii) its regulation through a cGMP-dependent protein kinase.
Cyclic nucleotide gated channels are structurally related to voltage-gated cation channels (38,39). The S 5 -S 6 linkers of the two groups show significant resemblance. Bearing in mind that the anti-␣ 1D antibody is raised against this region, the B cell channel can be potentially seen as a cyclic nucleotide gated entity. However, direct channel gating by cGMP in rod photoreceptors is apparently carried out by comparatively low affinity interactions (40). If we were to extrapolate this condition to B lymphocytes, the low cGMP levels found in these cells (in comparison to rod photoreceptors) would make a direct gating mechanism implausible.
A better case can be made for the involvement of a cGMPdependent protein kinase. The ability of Rp-8-pCPT-cGMPS to diminish the calcium response favors the role of a kinase as the immediate target of cGMP. The slow escape from inhibition with this compound may be due to the dissociation of the inhibitor and recovery of the cGMP-dependent protein kinase. The 8-substituted analogues of cGMP (such as 8-bromo-cGMP) are more potent activators of cGMP kinase I␣ than cGMP itself (41). This normally accounts for the biological activity of these compounds in intact cells. In fact, a reduction in cGMP-dependent protein kinase expression would make cells less responsive to 8-bromo-cGMP, while the restoration of its expression would restore the 8-bromo-cGMP response (42). Therefore, the B cell calcium response to 8-bromo-cGMP can be seen as another indicator of cGMP-dependent protein kinase involvement. The 8-bromo-cGMP result also argues against the involvement of cGMP regulated phosphodiesterases since it cannot interact with the allosteric binding site of these molecules (41).
Prior treatment of B cells with a high dose of thapsigargin did not prevent an additional increase in intracellular calcium upon anti-Ig stimulation, nor did the guanylyl cyclase inhibitor LY83583 affect the thapsigargin-induced calcium response. This indicates that the anti-Ig-induced increase in cytosolic calcium cannot be explained by the depletion of calcium from a thapsigargin-sensitive pool. This is in contrast to the reports on T cells where engagement of the T cell receptor after maximal thapsigargin treatment fails to evoke a further calcium response (43,44). One can suggest that T and B cell receptor-mediated increases in calcium are mechanistically dissimilar, a notion that concurs with the B cell specificity of anti-channel staining among lymphocytes.
Our findings introduce a novel calcium influx channel in non-excitable cells, provide ample evidence for the role of cGMP as a second messenger in B cell calcium influx, indicate a distinction between B and T cell calcium channels, and demonstrate the limitations of the capacitative calcium theory in explaining calcium influx in non-excitable tissues.