Ca2+ Signaling in Identified T-lymphocytes from Human Intestinal Mucosa

Ca2+ entry across the plasma membrane is necessary for the activation and proliferation of T-lymphocytes. Human intestinal lamina propria lymphocytes physiologically exhibit minimal proliferation in response to antigen receptor stimulation when compared with peripheral blood T-lymphocytes. This hyporeactivity is partially abolished in inflammatory bowel disease. We hypothesized that differences in Ca2+ signaling could be related to the disease. To test this possibility, we measured Ca2+ signals in identified lymphocytes from human blood and human intestinal mucosa. Ca2+ signals in lamina propria T-lymphocytes from non-inflamed tissue were drastically reduced when compared with Ca2+ signals of blood T-lymphocytes from the same persons. However, Ca2+ signals in T-lymphocytes from inflamed intestinal mucosa were much higher than the ones from non-inflamed mucosa and almost reached levels of Ca2+ signals in peripheral blood T-cells. Furthermore, Ca2+ influx was closely linked to cell proliferation in both peripheral blood T-lymphocytes and lamina propria lymphocytes cells. We conclude that differences in Ca2+ signaling can explain the differences of T-lymphocyte reactivity in blood versus lamina propria and, importantly, also between T-lymphocytes from inflamed and non-inflamed intestinal mucosa. Ca2+ channels in the plasma membrane of T-lymphocytes might thus prove an excellent target to screen for immunosuppressiva to potentially treat the symptoms of inflammatory bowel disease.

Ca 2ϩ influx across the plasma membrane following stimulation of Jurkat T-cells or peripheral blood T-lymphocytes is a necessary signal for T-cell activation (1). Following cross-linking of the TCR, 1 a number of signaling cascades are activated, one of which results in Ins-1,4,5-P 3 production. Ins-1,4,5-P 3 initiates Ca 2ϩ signaling in T-cells by two coupled processes: Ca 2ϩ release from the endoplasmic reticulum through the Ins-1,4,5-P 3 receptor channel and subsequent activation of plasma membrane Ca 2ϩ channels. Activation of the Ca 2ϩ channels is dependent on the Ca 2ϩ filling state of the endoplasmic reticulum. A low Ca 2ϩ concentration in the endoplasmic reticulum provides the signal to activate the plasma membrane Ca 2ϩ channels, whereas a high Ca 2ϩ concentration in the endoplasmic reticulum terminates the response (2). These channels are therefore referred to as store-operated Ca 2ϩ channels. In Tlymphocytes, the store-operated Ca 2ϩ channels are highly Ca 2ϩ selective (3) and are also termed CRAC channels (1, 4 -6). CRAC channels provide the Ca 2ϩ influx necessary for T-cell activation (7), and their absence correlates with greatly reduced T-cell functionality resulting in severe combined immunodeficiencies in humans (8 -10). These findings as well as numerous other observations (summarized in Ref. 1) stress the conclusion that CRAC channel activity determines T-cell reactivity. The molecular structure of CRAC channels is still unknown.
T-lymphocytes from peripheral blood and also Jurkat T-cells activate very well following stimulation through the TCR (1,7,10). In contrast, T-lymphocytes from the intestinal mucosa do not respond very well to TCR stimulation and are therefore referred to as hyporeactive (11). In patients with IBD, such as Crohn's disease or ulcerative colitis, pathologically enhanced LP T-cell proliferation is believed to be one of the reasons causing inflammation and it is concluded that T-cells play a central role for the pathogenesis of IBD (12)(13)(14)(15)(16)(17). T-lymphocytes from inflamed tissue are hyperreactive and reach proliferation rates comparable to blood T-lymphocytes following TCR stimulation (12). The reason for the differences in reactivity of PB and LP T-cells is not known.
Because Ca 2ϩ signals are closely linked with activation and proliferation of T-cells (1,7), we hypothesized that differences in Ca 2ϩ signaling in LP T-cells could explain their hyporeactivity (compared with PB T-cells) and also their hyperreactivity in inflamed tissue. We developed protocols that allowed us to measure Ca 2ϩ signals in identified T-cells from the intestinal mucosa and peripheral blood of healthy individuals and patients with active IBD. We found that store-operated Ca 2ϩ influx correlated well with the reactivity of the cells and that the amplitude of Ca 2ϩ signals due to Ca 2ϩ influx was linked to the proliferative response of the cells. Our results explain the differences of reactivity found in intestinal T-lymphocytes from inflamed and non-inflamed tissues.

Isolation of Lamina Propria Lymphocytes (LPLs) and Peripheral
Blood Lymphocytes (PBLs)-LPLs were obtained from 6 -8 mucosa biopsies taken from colonoscopic examinations. Biopsies were washed three times in HBSS (PAA Laboratories GmbH, catalog number H15-009). The mucosa was cut into little pieces and shaken for 3 h at 37°C in collagenase medium of the following composition: 500 ml of RPMI 1640 medium (Invitrogen, catalog number 21875-034); 10% fetal bovine serum (Invitrogen), catalog number 10270-106); 2.5 ml of gentamycin (Biochrom, catalog number A2710); 1 ml of amphotericin B (500x, Roche Applied Science, catalog number 14977600); 12.5 ml of HEPES (1 M); 0.5 ml of mercaptoethanol (0.5 M), 5 ml of collagenase (10 mg/ml, Biochrom, collagenase-type CLS III (141 units/mg), catalog number C3-22); 5 ml of trypsin inhibitor (10 mg/ml, Sigma, trypsin inhibitor from Glyxine max (soybean), type I-S, product number T6522); 5 ml of DNase I (10 mg/ml, Roche Applied Science, DNase I from bovine pancreas (100 mg), catalog number 1284932); 50 units/ml penicillin; and 50 g/ml streptomycin (Invitrogen, catalog number 15140-122). Following digestion, the suspension was carefully sucked through a syringe to disrupt the remaining tissue. Cells were separated from larger mucosa remnants by filtration through a 70 m cell sieve. The suspension was then centrifuged for 10 min at 360 ϫ g at room temperature. The pellet was resuspended in 30% Percoll (Amersham Biosciences, catalog number 17-0891-01, diluted with 0.9% NaCl) and layered over a 70% Percoll solution. The following density centrifugation for 20 min at 1230 ϫ g yielded a population of mononuclear cells at the interface. Cells were carefully recovered and washed twice in medium by centrifugation at 360 ϫ g for 10 min at room temperature. Before use, cells were kept for 1-8 h at 37°C in culture medium at a concentration of 1 million/ml. PBLs were obtained from fresh blood anti-coagulated with heparin. 10 ml of blood was diluted with 10 ml of phosphate-buffered saline (Invitrogen, catalog number 14190-094) and layered over Ficoll (Ficoll-Paque Plus, Amersham Biosciences, catalog number 17-1440-03). Density centrifugation for 20 min at 1230 ϫ g yielded a population of mononuclear cells at the interface. The following isolation steps were equal to those described for LPL isolation.
For proliferation experiments, PBLs were purified from leukocyte reduction filters from the local blood bank. Cells were collected by back-flushing the filter with 120 ml of HBSS (PAA, catalog number 15-009). Peripheral blood mononuclear cells were isolated by a density gradient centrifugation at 450 ϫ g for 30 min at room temperature (Ficoll-Paque Plus, Amersham Biosciences, catalog number 17144002) in 50 ml of Leucosep tubes (Greiner, catalog number 227290). The peripheral blood mononuclear cell layer was washed in HBSS. Remaining red blood cells were removed by the addition of 3 ml of lysis buffer (155 mM NH 4 Cl, 10 mM KHCO 3 , 0.1 mM EDTA, pH 7.3) for 3 min. After lysis, peripheral blood mononuclear cells were washed with HBSS (200 ϫ g for 10 min at room temperature). Viability of the cells was checked by staining with trypan blue. Cells were further purified by adhering to plastic for 24 h in RPMI 1640 complete medium at 37°C (1.5 ϫ 10 6 cells/ml). Non-adherent cells, mostly PBLs, were collected and used for proliferation assays.
Ca 2ϩ Imaging-Cells were loaded at 22-23°C for 20 min with 1 M Fura-2/AM (Molecular Probes) in medium with 10 mM HEPES added, washed with fresh medium, stored at room temperature for 10 min, and immediately used. Cells were allowed to adhere to poly-L-ornithinecoated (0.1 mg/ml, Sigma) glass coverslip chambers on the stage of an Olympus IX 70 microscope equipped with a ϫ40 Uplan/Apo (numerical aperture 1.0) objective. Cells were alternately illuminated at 340 and 380 nm with the Polychrome IV monochromator (TILL Photonics). The fluorescence emissions at Ͼ 440 nm were captured with a CCD camera (TILL Imago), digitized, and analyzed using TILL Vision software (TILL Photonics). Ratio images were recorded at intervals of 5 s. [Ca 2ϩ ] i was estimated from the relation [Ca 2ϩ ] i ϭ K * (R Ϫ R min )/(R max Ϫ R) where the values of K * , R min , and R max were determined from an in situ calibration of Fura-2/AM in Jurkat T-cells as described previously (18). Ca 2ϩ Ringer's solution contained (in mM): 155 NaCl, 4.5 KCl, 1 CaCl 2 , 1 MgCl 2 , 10 D-glucose, and 5 HEPES (pH 7.4 with NaOH). CaCl 2 was replaced by MgCl 2 in the Ca 2ϩ -free Ringer's solution with 1 mM EGTA added. Thapsigargin (1 M, stock 1 mM in Me 2 SO, Molecular Probes) and anti-CD3 (OKT-3) monoclonal antibody (10 g/ml, stock at 4 mg/ml, obtained from the hybridoma OKT-3, ATCC) were used to stimulate the cells. A sandwiched self-made chamber was used for all of the measurements, which allowed for a complete solution exchange Ͻ1 s.
Identification of T-lymphocytes-To allow identification of different subtypes of lymphocytes, cells were stained with anti-CD3-PE, anti- CD4-Cy5, and anti-CD8-fluorescein isothiocyanate antibodies (diluted 1:12 in 0.9% NaCl containing 3% fetal bovine serum, Dako) on the stage of the microscope following the Ca 2ϩ -imaging experiment. After 5-10 min, the antibody solution was washed away with Ringer's solution. Standard HQ filter sets (AHF Analysentechnik) were used for identification of cells, which were excited with the Polychrome IV monochromator at 480 nm (fluorescein isothiocyanate), 540 nm (PE), or 620 nm (Cy5).
Proliferation Assays-Proliferation experiments using the EZ4U assay were carried out in 96-well cell culture plates (BD Biosciences, catalog number 353072, flat bottom), and data points were measured as triplicates. 50,000 PBLs were cultured in a total volume of 200 l in each well. Cells were stimulated using 5 nM phorbol 12-myristate 13acetate (PMA, Sigma, catalog number P1585) and 0.5 M ionomycin (Sigma, catalog number I0634). Plates were incubated for 72 h at 37°C, 5% CO 2 , and 95% humidity. After incubation time, the number of living cells was determined by the reduction of the tetrazolium salt EZ4U to formazan derivatives (Biozol, catalog number BI-5000). 20 l of EZ4U reagent was added to each well, and plates were incubated for another 4 h. Optical density was measured in a EL X 800 UV universal microplate reader (BIO-TEK Instruments) at wavelength settings of 465-630 nm.
For proliferation experiments using [ 3 H]thymidine incorporation, 5 ϫ 10 4 purified cells were added in 0.2 ml of RPMI 1640 supplemented with 10% fetal bovine serum, 2% L-glutamine, and antibiotics. After 96-h incubation at 37°C, 1 Ci of [ 3 H]thymidine was incorporated over a 6-h period before harvesting the cells (Inotec, Wohlen, Switzerland). [ 3 H]Thymidine incorporation was measured in a liquid scintillation spectrometer (Beckmann, Munich, Germany). All of the assays were performed in triplets, and the results differed by Ͻ15%. To analyze the impact of [Ca 2ϩ ] i on cell proliferation, the Ca 2ϩ concentration in the medium was modified by adding extra Ca 2ϩ or EGTA.
Data Analysis-Data were analyzed using TILL Vision, Igor Pro (Wavemetrics), and Microsoft Excel. Averages are presented as the mean Ϯ S.E. For statistical analysis, an unpaired two-sided Student's t test was used. Data were considered significantly different if p was smaller than 0.01.

RESULTS
To allow Ca 2ϩ measurements in identified T-lymphocytes from intestinal mucosa and peripheral blood without potential pre-activation of the cells, separation steps of the cells with antibodies through magnetic beads or FACS were avoided. A protocol was developed that allowed parallel Ca 2ϩ imaging and identification of the same cells with up to three different antibodies after the experiment. Fig. 1 shows a typical example of such a LPL experiment with an infrared picture (A), [Ca 2ϩ ] i measurements of cells at one time point (B), and resulting [Ca 2ϩ ] i traces for two of the cells over time (C). To enable identification of T-cell subtypes, cells were stained on the stage of the microscope with anti-CD3, anti-CD4, and anti-CD8, respectively (D-F).
To activate store-operated Ca 2ϩ channels, cells were stimulated with 1 M TG in Ca 2ϩ -free Ringer's solution. TG fully inhibits the SERCA Ca 2ϩ -ATPases of the endoplasmic reticulum, causing a small transient rise in [Ca 2ϩ ] i (Fig. 1C) due to the unopposed leakage of Ca 2ϩ from stores followed by extrusion across the plasma membrane. Depletion of Ca 2ϩ stores by this method maximally activates store-operated Ca 2ϩ channels in the plasma membrane, also referred to as CRAC channels in T-lymphocytes (4,5). No Ca 2ϩ entry through the channels is observed during TG stimulation when the extracellular solution does not contain any Ca 2ϩ . Changing the extracellular Ca 2ϩ concentration to 1 mM following complete store depletion, then allows Ca 2ϩ influx through store-operated Ca 2ϩ channels, giving rise to long-lasting Ca 2ϩ signals in T-cells (Fig. 1C).
To compare LP T-cells, which are hyporeactive, with "normal" reactive PB T-cells, we isolated PBLs from the same patients. PBLs were analyzed using exactly the same protocol applied for the LPLs in Fig. 1. Fig. 2 compares the Ca 2ϩ signals of LPLs and PBLs grouped into CD4 ϩ and CD8 ϩ T-cells. Fig. 2, A and B, shows representative experiments, and the statistical analysis of all of the experiments is presented in Fig. 2C. PB T-cells display larger Ca 2ϩ signals following activation of storeoperated Ca 2ϩ entry through CRAC Ca 2ϩ channels than LP T-cells (p Ͻ 0.0001). This is true for CD4 ϩ and CD8 ϩ cells, whereas in general, CD4 ϩ cells respond slightly but significantly better than CD8 ϩ cells (p Ͻ 0.01). The larger Ca 2ϩ signals in PB T-cells correlate well with their reactivity and proliferation following activation of the cells, whereas the smaller Ca 2ϩ signals in LP T-cells correlate with their previously described hyporeactivity (11). We did not observe differences in resting Ca 2ϩ levels between LP and PB T-cells, which have been reported by another group (19). The reason for this discrepancy was not further investigated. It is noteworthy, however, that we observed higher LPL resting Ca 2ϩ signals when the cell preparation was not optimal, which was evident by a very low yield of cells and many debris and "degranulated" cells.
Whereas LP T-cells of healthy individuals are hyporeactive, this is completely changed in T-cells from inflamed tissue of patients with active IBD. In the latter case, the T-cells proliferate quite well and almost reach rates comparable with "normal" reactive PB T-cells. Considering the importance of intracellular Ca 2ϩ signals for T-cell proliferation, we hypothesized that Ca 2ϩ signals in T-cells from inflamed tissue could be increased, thus explaining the higher proliferation rates. clear difference of Ca 2ϩ signals between LP T-cells of noninflamed and inflamed mucosa, the latter of which reached Ca 2ϩ concentrations almost as high as the respective PB CD8 ϩ T-cells. The same was found for the CD4 ϩ cells of the patient (data not shown). In Fig. 3B, the CD3 ϩ T-cells from all of the patients are compared. Ca 2ϩ signals due to Ca 2ϩ influx are clearly enhanced in LP T-cells from inflamed tissue versus non-inflamed tissue and reach levels of Ca 2ϩ signals in PB T-cells. Ca 2ϩ signals in PB T-cells did not differ between cells from IBD and non-IBD individuals. Fig. 3, C and D, depicts the statistical analysis of the Ca 2ϩ peak and plateau for the different T-cell populations analyzed. In CD4 ϩ and CD8 ϩ T-cells as well as in CD3 ϩ CD4 Ϫ CD8 Ϫ T-cells, we found the same pattern. Whereas Ca 2ϩ peak and plateau in all of the LP T-cell subpopulations from non-inflamed tissue are significantly lower than the signals from the respective PB T-cell subpopulations (p Ͻ 0.0001), Ca 2ϩ signals in LP T-cells from inflamed tissue are significantly increased compared with the ones from non-inflamed tissue (p Ͻ 0.01 for Ca 2ϩ peak and plateau) and almost reach levels of Ca 2ϩ signals in PB T-cells. These findings correlate well with the hyperreactivity of LP T-cells from inflamed intestine of individuals with active IBD.
TG induces a complete depletion of Ca 2ϩ stores and thus allows us to study the maximal Ca 2ϩ entry through storeoperated Ca 2ϩ channels independent of the activation of TCRdependent signaling cascades. Activation of the TCR through OKT-3, which binds to CD3, also induces Ca 2ϩ signaling in T-lymphocytes but only partially depletes Ca 2ϩ stores because the released Ca 2ϩ can be pumped back into the Ca 2ϩ stores by the SERCA ATPases. TCR stimulation by 10 g/ml OKT-3 in Ca 2ϩ -free Ringer's solution induces Ca 2ϩ release transients because of the generation of Ins-1,4,5-P 3 , which releases Ca 2ϩ from endoplasmic reticulum stores and subsequent re-uptake of Ca 2ϩ into the stores or clearance of Ca 2ϩ across the plasma membrane. Fig. 4A shows examples of LP and PB T-cells to illustrate typical response patterns. In cells that do not show release transients in response to OKT-3 (one of the LP T-cells, thin solid line), only very little influx is observed. The influx is clearly enhanced when OKT-3 stimulation induces a Ca 2ϩ release spike, thereby partially depleting Ca 2ϩ stores that then results in activation of store-operated Ca 2ϩ entry. The release transients due to OKT-3 stimulation can occur at different times (Fig. 4A) and are therefore not visible in the Ca 2ϩ traces when averaging cells (Fig. 4B). We compared LP and PB T-cells from the same individuals. Approximately 44% PB T-cells from healthy individuals or patients with IBD showed clear Ca 2ϩ

FIG. 3. Comparison of Ca 2؉ signals in PB and LP T-lymphocytes from patients with active IBD and healthy individuals.
Experiments were carried out as described in Fig. 1 release transients during the OKT-3 application and were counted as responders and used for the average in Fig. 4B. In contrast, only 16% of the LP T-cells from healthy individuals or patients with IBD responded to OKT-3 stimulation. Comparing the Ca 2ϩ influx signals of PB and LP responders, we again observed that the Ca 2ϩ peak and plateau following Ca 2ϩ influx in responding LP T-cells was lower than that in the responding PB T-cells (p Ͻ 0.01, Fig. 4B, statistics on the calcium plateau in 4C). Similar to the TG stimulation protocol, Ca 2ϩ peak and plateau following influx were enhanced in LP T-cells from patients with active IBD compared with LP T-cells from non-IBD individuals (p Ͻ 0.01). Surprisingly, we also found an influx increase in PB T-cells from patients with active IBD when compared with PB T-cells from healthy individuals (p Ͻ 0.01), which was not observed following TG stimulation. The reason for this difference is currently unknown but could result from the T-cells of the IBD patients being already activated.
To link the observed differences in Ca 2ϩ signaling between LP T-lymphocytes from persons with and without active IBD to T-cell function, we analyzed the importance of Ca 2ϩ entry for cell proliferation. Proliferation of PBL and LPL cells were analyzed while varying the net Ca 2ϩ influx rate through changes in the extracellular Ca 2ϩ concentration (details are described in the figure legend). Fig. 5A illustrates the strong dependence of PBL proliferation rates on the external Ca 2ϩ concentration. Free [Ca 2ϩ ] ext below 100 M is not sufficient to induce the full proliferative response of PBLs, even when a strong stimulus (PMA, ionomycin) was used. Using TG or OKT-3 as a stimulus, cell proliferation was also completely blocked by low micromolar [Ca 2ϩ ] ext (Fig. 5B), whereas it could be slightly enhanced by adding 1 mM Ca 2ϩ to the medium, indicating that proliferation in PBL depends on the external Ca 2ϩ concentration and therefore on net Ca 2ϩ entry. The same pattern was found in LPL cells using TG as stimulus (Fig. 5C), the overall proliferation rates, however, being much smaller. The proliferation rates in LPL cells from inflamed tissue were found to be higher than the ones of LPL cells from non-inflamed tissue. These results link Ca 2ϩ signals in PBL cells, LPL cells, and LPL cells from inflamed mucosa with cell proliferation and the immune response. It can be speculated that reduced Ca 2ϩ signals are causally related to cell hyporeactivity LPL cells from healthy individuals, whereas the enhanced Ca 2ϩ signals might well account for the pathological hyperreactivity of LPLs from patients with active IBD.

DISCUSSION
Several groups have shown that T-lymphocytes of the intestinal mucosa are hyporeactive and respond with low proliferative response and low transcriptional activity upon antigen presentation (11,17,20,21). In contrast, T-lymphocytes from peripheral blood are able to react with strong proliferative activity upon antigen stimulation. The hyporeactivity of intestinal T-cells is one mechanism likely to avoid pathologic inflammatory responses in the intestine, which is continuously exposed to antigen. The molecular mechanism for the different reactivity of LP and PB T-cells is unclear. Because elevations of [Ca 2ϩ ] i have been closely linked to activity and proliferation of T-lymphocytes (1, 7, 10), we compared Ca 2ϩ signals in resting and activated LP and PB T-cells. We found resting [Ca 2ϩ ] i levels to be similar in LP and PB T-cells following stimulation; however, PB T-cells reached much higher [Ca 2ϩ ] i levels than LP T-cells, which correlate well with their proliferative responses. Ca 2ϩ signals in LP T-cells from inflamed tissue were drastically increased, which could explain the enhanced T-cell

FIG. 4. Comparison of Ca 2؉ signals in PB and LP T-lymphocytes from patients with active IBD and healthy individuals following TCR stimulation.
Experiments were carried out as described in Fig. 1, only that OKT-3 was used to activate the TCR instead of TG depleting internal Ca 2ϩ stores. A, Ca 2ϩ signals of individual cells following OKT-3 stimulation are shown. Activation of the TCR results in transient Ca 2ϩ signals that concomitantly activates the Ca 2ϩ influx observed following Ca 2ϩ re-addition. Only very little Ca 2ϩ influx was observed in case cells did not respond to OKT-3 stimulation as depicted for the one LP T-cell. B, averages of all of the cells, which did respond to OKT-3 stimulation. Ca 2ϩ transients are not visible because they average out when many cells are pooled. C, analysis of Ca 2ϩ peaks and plateaus of responding cells as described in the other figures. Because of the fact that only few LPL cells respond to OKT-3, cell numbers were small (ranging from 8 to 110 cells/2 to 5 patients). Error bars represent mean Ϯ S.E. proliferation described in those cells. The increased responsiveness of LP T-cells from inflamed tissue could be attributed to infiltration of the lamina propria by PB T-cells or to increased Ca 2ϩ influx of the resident LP T-cells themselves. It has been shown that lymphocytes, which are primed and differentiated in Peyer's patches and in mesenteric lymph nodes, migrate into the blood stream and preferentially return to Peyer's patches or the lamina propria of the gut (22,23). Tissue-specific accumulation and proliferation of T-cells and/or proliferation of resident T-cells may both modulate chronic inflammation such as IBD. At present, we cannot distinguish whether the increased conclusion. Whereas Ca 2ϩ release from stores was on average comparable in PB and LP T-cells following TG stimulation, only very few LP T-cells (16% compared with 44% PB T-cells) responded to TCR stimulation with clear Ca 2ϩ release transients during the 500-s OKT-3 exposure. Ca 2ϩ entry following activation of plasma Ca 2ϩ channels by OKT-3 or thapsigargin was present in all of the LP T-cells, but it was clearly reduced when compared with PB T-cells. Because activity of the plasma membrane Ca 2ϩ channels is closely correlated with T-cell activation and proliferation (7), the reduction of Ca 2ϩ entry could explain the hyporeactivity of LP T-cells upon antigen stimulation.
Recent publications highlight the importance of Ca 2ϩ signals for the induction of peripheral tolerance in T-cells. Macian et al. (25) reported that TCR stimulation without co-stimulation strongly favors Ca 2ϩ -dependent signal transduction via nuclear factor of activated T-cells over other TCR-dependent signal transduction pathways, thereby inducing T-cell anergy, which is a tolerance mechanism in which T-cells are functionally inactivated (26). On the other hand, oral tolerance can also be associated with an impairment of Ca 2ϩ -dependent signal transduction (27). At first sight, our results would better fit in with the latter observation. However, one can easily imagine that both mechanisms (25,27) could even co-exist in the same cell because Ca 2ϩ can have very different specific cellular functions depending on kinetics, amplitude, and localization of the [Ca 2ϩ ] i elevation (1,28,29). How the molecular mechanism of mucosal T-cell hyporeactivity exactly works is still unknown, but it has been speculated that the microenvironment of the intestinal mucosa is important for the suppression of T-cell responses (30,31). A continuous suboptimal stimulation by mucosal factors and/or antigen would be likely to induce the tolerance of mucosal T-lymphocytes. One could speculate that such a suboptimal stimulation leads to a reduction of Ca 2ϩ influx through CRAC channels that in turn would suppress activation and proliferation of the cells following optimal TCR stimulation.
Abnormal activity and proliferation of T-lymphocytes from the human intestinal lamina propria is thought to play an important role in IBD such as Crohn's disease or ulcerative colitis (12)(13)(14)(15)(16)(17). In contrast to cells from non-inflamed tissue, T-cells from inflamed tissue have been found to be hyperreactive. Following TCR stimulation, the cells show abnormally high proliferation rates that are thought to be one reason for the constant intestinal inflammation. If our hypothesis that Ca 2ϩ entry is closely linked to proliferation in LP T-cells were correct, we would expect to find increased Ca 2ϩ entry in LP T-cells from inflamed tissue. Indeed, this was the case. Whereas the number of responding cells following TCR stimulation was not increased in T-cells from inflamed intestine, there was a clear increase of Ca 2ϩ signals because of Ca 2ϩ influx across the plasma membrane following TCR or thapsigargin stimulation when compared with LP T-cells from noninflamed tissue. This increase of [Ca 2ϩ ] i following stimulation could explain the hyperreactivity of LP T-cells from inflamed intestine of patients with active IBD. Because enhanced proliferation of activated T-cells is believed to be one of the problems during IBD, decreasing Ca 2ϩ influx could potentially also decrease inflammation in the intestine.
An important question is how the higher Ca 2ϩ signals in LP T-cells from patients with active IBD are generated. There are two principal possibilities to account for the differences. One is an increase of net influx, the other is a decrease of net efflux of Ca 2ϩ from the cytosol. A change in the net efflux is not very likely because Ca 2ϩ clearance rates were not different between PB and LP T-cells from healthy individuals or patients with active IBD. 2 This leaves increased Ca 2ϩ entry as the best explanation, which could be achieved by higher expression of Ca 2ϩ channels or expression of a different type of Ca 2ϩ channels with higher Ca 2ϩ permeability in the plasma membrane. Alternatively, an increase in the net Ca 2ϩ influx rate could also be mediated by a more negative membrane potential of the cells. In any case, influx of Ca 2ϩ through store-operated CRAC Ca 2ϩ channels is increased in LP T-cells from inflamed tissue and reaches levels of PB T-cells, thus abolishing the hyporeactivity of the LP T-cells. Inhibition of store-operated Ca 2ϩ entry in LP T-cells of patients with active IBD could decrease activation and proliferation of the cells leading to a reduction of intestinal inflammation. Their localization in the plasma membrane makes the CRAC Ca 2ϩ channels in T-cells a good target for future immunosuppressiva to potentially treat the symptoms of IBD.