The T lymphocytes that infiltrate the exocrine glands in Sjögren’s syndrome (SS) play a key role in damaging glandular epithelial cells, but the mechanisms of this damage by T lymphocytes are not fully understood. To determine the cellular basis of this phenomenon, we focused our attention on the T lymphocytes around acinar epithelial cells in SS. We showed that CD8+ but not CD4+ T lymphocytes were located around the acinar epithelial cells and that a majority of these CD8+ T lymphocytes possess an unique integrin, αEβ7 (CD103). The acinar epithelial cell adherent with αEβ7 (CD103)+ CD8+ T lymphocytes was apoptotic. Both the perforin/granzyme B and Fas/Fas ligand pathways were implicated in the process of programmed cell death in lacrimal glands. These results suggested that αEβ7 integrin, by interacting with E-cadherin, mediates the adhesion between CD8+ T lymphocytes and acinar epithelial cells in SS and participates in inducing epithelial cell apoptosis, leading to secretory dysfunction of exocrine glands, a hallmark of SS.

Sjögren’s syndrome (SS)3 is an autoimmune disease characterized by severe lymphocytic infiltration in the lacrimal and salivary glands, leading to sicca syndrome which includes xerophthalmia and xerostomia (1). In general, the degree of lymphocytic infiltration is well correlated with the decrease in glandular secretion (1, 2, 3, 4), suggesting that the infiltrating lymphocytes are involved in the destruction of the lacrimal and salivary glands. The majority of the infiltrating lymphocytes are CD4+ T cells, but the CD8+ T cells, B cells, and plasma cells are also observed in the inflamed tissue (1, 3). These cells are expressing an activation marker, IL-2 receptor, and one of cytotoxic granules, granzyme A (5, 6), suggesting that the lymphocytes activated in situ may be involved in the destruction of salivary gland epithelial cells. Electron microscopic study has demonstrated that some T lymphocytes attach to the acinar epithelial cells (7). These results raised the possibility that such adherent T cells, directly communicating with the target cells, play a pathological role in the destruction of epithelial cells (7).

Recently, it has been shown that the acinar epithelial cells in SS were Fas and Fas ligand (FasL) positive, and these cells died by apoptosis (8). In contrast, most lymphocytes in the exocrine glands in SS express not only Fas but also bcl-2, resulting in minimal cell death, particularly in dense periductal foci. These results indicate that Fas-mediated programmed cell death in acini, but not in lymphocytes, may be an important mechanism leading to the glandular destruction found in SS.

In this regard, it has been proposed that the mechanism of tissue destruction in the nonobese diabetic strain of mice, a model of human insulin-dependent diabetes mellitus, is 1) recognition linked, in the sense that CD8+ T lymphocytes directly attack the target islet β cells, and 2) activation linked, meaning that activated CD4+ or CD8+T lymphocytes kill the bystander β cells (9, 10). Three distinct death effector systems are implicated in the destruction of islet β cells in nonobese diabetic mice, Fas/FasL, perforin, and TNF/TNFR (10). In a recognition model, CTL kill target cells by three distinct steps. First, T lymphocytes recognize an Ag in the context of MHC. Next, the interaction of adhesion molecules further confers tight adhesion between these cells in Ag recognition. Finally, the T cells release perforin and cytotoxic granules or cross-link Fas Ag on the target by FasL on the T lymphocytes and kill the target cells.

Although the activated T cells in the affected exocrine glands are most likely responsible for the pathogenesis of the glandular destruction in SS, the phenotype and adhesion molecules on the T cells, necessary to adhere to the glandular epithelial cells, and the precise mechanism by which these T cells attack the exocrine glands in SS are not fully understood.

In this study, we examined the surface structure of T lymphocytes located around the acinar epithelial cells and the expression of death mediators in the exocrine glands to explore the mechanism how infiltrating T lymphocytes damage exocrine glands in SS patients.

Xerophthalmia was diagnosed when compromised tear dynamics and ocular surface abnormalities were both present, as previously described (11). Patients with xerophthalmia were further categorized into two groups: those who satisfied the criteria for SS and those who did not (non-SS) (12). Six non-SS patients (all female; mean age, 50.8 yr; range, 37–58 yr) and seven SS patients (all female; mean age, 47.5 yr; age range, 31–56 yr) underwent lacrimal and salivary gland biopsies as described previously (3, 11) after obtaining their informed consent to perform the procedure.

Lacrimal secretion was evaluated by the Schirmer test and the nasal stimulation test (NST) as described previously (11). Salivary secretion was evaluated by the Saxon test as described (13). Normal values for individual test are follows: Schirmer test >5 mm/5 min; NST >10 mm; Saxon test >3.0 g/2 min.

Biopsy specimens were cut into two pieces, one for hematoxylin-eosin (HE) staining and one for immunostaining. For HE staining, the tissue sample was fixed with 4% paraformaldehyde and embedded in paraffin wax. All sections were cut 4 μm thick and stained with Mayer’s HE. For immunostaining, the biopsy tissue was snap-frozen, embedded in Tissue-Tek OCT compound (Lab-Tek Products, Elkhart, IN), and stored at −70°C until used. Frozen sections were cut 5 μm thick, blocked with PBS containing 10% normal goat serum for 20 min at room temperature, fixed in cold acetone for 10 min, and treated with 0.3% H2O2 in methanol for 30 min to inhibit endogenous peroxidase activity.

The following mAbs were used for immunostaining: anti-CD4 (Leu-3a: IgG1), anti-CD8 (Leu-2a: IgG1), anti-ICAM-1 (LB-2: IgG2a), and HLA-DR (L243: IgG2a) (Becton Dickinson, San Jose, CA); anti-αEβ7 (2G5: IgG2a), anti-E-cadherin (67A4: IgG1), anti-HLA-A, -B, -C (B9.12.1: IgG2a), and anti-VCAM-1 (E1/6: IgG1) (Immunotech, Luminy, France); anti-APO2.7 (2.7A6A3: IgG1), anti-CD45 (KC56: IgG1), and anti-CD19 (J4.119: IgG1) (Coulter Immunology, Hialeah, FL); anti-Fas (UB2: IgG1) and anti-granzyme B (GrB-7: IgG2a) (MBL, Nagoya, Japan); and anti-FasL (Mike1, Mike2: rat IgG) (Alexis, Lexington, KY). Anti-VLA-4 (SM-27: IgG1) and anti-αEβ7 (SM-17: IgG1) were developed in our laboratory and described elsewhere (14, 15). Avidin-biotin-peroxidase staining with a Histofine kit (Nichirei, Tokyo, Japan) was used for immunohistochemical staining. For immunofluorescent staining, the primary Abs were detected with fluorescein-conjugated anti-mouse IgG Ab (Biosource International, Camarillo, CA). Stained tissue specimens were visualized with a laser image analyzer equipped with a confocal optical system (ACAS570, Meridian Instruments, Okemos, MI). Optical sections were made in the XY plane, and confocal images were acquired by single or dual methods. Gray scales were converted to 12-bit spectral color images, which were overlaid on the image obtained by merge scan. Background fluorescence was determined with isotype-matched control Ab (Coulter Immunology). Average fluorescent intensity was determined by scanning five randomly selected fields of the tissue section.

For counting of lymphocytes around acini, 10 acini in the randomly selected field are examined. The lymphocytes in contact with the epithelial cells composed of an acinus by merge scan on an ACAS 570 (×400 magnification) are judged to locate around acini and are counted. The number of lymphocytes around acini (a single acinus) is expressed as total counts of lymphocytes in contact with epithelial cells/10. For counting of lymphocytes in the tissue section, the sequential frozen section was stained by immunohistochemical method with biotin-labeled anti-CD45, anti-CD4, anti-CD8, and anti-CD19 and counted the positive cells in randomly selected field at ×400 magnification. The average counts in 10 given fields are expressed.

Formalin-fixed and paraffin-embedded tissue sections were deparaffinized and rehydrated. After extensive washing, the proteinization was conducted with proteinase K (Sigma, St. Louis, MO) at a concentration of 20 μg/ml for 15 min at room temperature. The slides were incubated at 37°C for 60 min with freshly prepared TdT mixture containing 0.4 U/μl TDT, 10 nM biotin-conjugated dUTP, and TdT buffer (Boehringer Mannheim, Indianapolis, IN), followed by staining with avidin-conjugated fluorescein at room temperature for 30 min. The slides without TdT served as negative controls.

PBLs were separated from heparinized venous blood by Ficoll-Hypaque density gradient centrifugation (Pharmacia, Uppsala, Sweden), as described previously (14), and cultured in RPMI 1640 supplemented with 10% FCS (Hazelton Biologics, Lenexa, KS) in the absence or presence of PHA (Difco, Detroit, MI) at a concentration of 10 μg/ml in a 25-cm2 flask in an upright position for 6 days. Flow cytometric analysis revealed that αEβ7 and LFA-1 were expressed on ∼30 and 90% of activated PBLs, respectively, whereas those were less than 5 and 90% in nonactivated PBL, respectively. T cell-to-epithelial cell adhesion assays were conducted as described previously (16). Cells from an epithelial cell line established from human submandibular glands (HSG) (a gift from Dr. Mitsunobu Sato, Tokushima University, Japan) were cultured to confluence in 12-well plates. The activated PBLs were preincubated with anti- αEβ7 and/or anti-LFA-1 mAb (1:50 dilution). After incubation for 30 min at 4°C, the nonactivated or activated PBL were added to confluent HSG cells that had been preincubated with anti-E-cadherin and/or anti-ICAM-1 mAb (1:50 dilution) for 5 min and incubated for 40 min at 37°C. Unbound PBLs were removed by gently washing with PBS, and the cells were fixed with acetone and stained with toluidine blue. The PBLs bound to HSG cells were counted under a microscope.

In this report, we examined five dry eye patients with SS and four patients with non-SS as a control. As shown in Fig. 1, the tear-secretory function, when assessed by Schirmer test, was impaired both in SS and non-SS patients, although the decrease in tear secretion was severer in SS than non-SS patients. Except for the Schirmer test, all of the other secretory functions were nearly normal in non-SS, but significantly depressed in SS patients. Histopathological analysis showed a significant increase in the lymphocytic infiltrates in SS, whereas only minimal infiltration of lymphocytes was detected in non-SS patients.

FIGURE 1.

Secretory function and pathological characteristics of the exocrine glands from patients with non-SS (□, n = 6) and SS (▪, n = 7). a, Tear secretion of the right eye (R) or left eye (L) was assessed by Schirmer test and NST as described in Materials and Methods. b, Saliva secretion was evaluated by Saxon’s test as described in Materials and Methods. c, Pathological grading for lacrimal glands or labial salivary glands were judged according to Greenspan’s scoring system (48 ). ∗, p < 0.05. Normal values are follows: Schirmer test >5 mm/5 min, NST >10 mm, Saxon test >3.0 g/2 min.

FIGURE 1.

Secretory function and pathological characteristics of the exocrine glands from patients with non-SS (□, n = 6) and SS (▪, n = 7). a, Tear secretion of the right eye (R) or left eye (L) was assessed by Schirmer test and NST as described in Materials and Methods. b, Saliva secretion was evaluated by Saxon’s test as described in Materials and Methods. c, Pathological grading for lacrimal glands or labial salivary glands were judged according to Greenspan’s scoring system (48 ). ∗, p < 0.05. Normal values are follows: Schirmer test >5 mm/5 min, NST >10 mm, Saxon test >3.0 g/2 min.

Close modal

HE-stained section of lacrimal gland samples revealed an enormous amount of mononuclear cells, observed as dense lymphocytic foci, while some lymphocytes were scattered around the acinar epithelial and glandular ductal cells, as seen in the representative SS case of Fig. 2,a. In contrast, only scattering of lymphocytes was detected in the lacrimal glands from non-SS patient (Fig. 2,a). To determine the phenotype of lymphocytes infiltrating into the lacrimal glands, immunohistochemical analysis with anti-CD4 or anti-CD8 Ab was conducted. As shown in Fig. 2,b, almost all of the CD4+ T lymphocytes were located in the periductal area as dense lymphocytic foci, whereas CD8+ T lymphocytes were predominantly observed around acinar cells. The bottom two sections in Fig. 2 b are at higher magnification (×400); arrows indicate CD8+ T lymphocytes around the acinar epithelial cells. A similar pattern of localization of the distinct T cell subsets was commonly observed in the four other SS patients and in the salivary gland samples from all SS patients.

FIGURE 2.

HE-stained sections of lacrimal glands from patients with non-SS (a, left) and SS (a, right). Immunohistochemical staining with anti-CD4 or anti-CD8 was conducted simultaneously for the representative SS patient (b). CD4+ T lymphocytes were detected as dense foci (upper panels) and the CD8+ T lymphocytes were observed predominantly around glandular epithelial cells (arrows). Original magnification, ×100; bottom panels, ×400.

FIGURE 2.

HE-stained sections of lacrimal glands from patients with non-SS (a, left) and SS (a, right). Immunohistochemical staining with anti-CD4 or anti-CD8 was conducted simultaneously for the representative SS patient (b). CD4+ T lymphocytes were detected as dense foci (upper panels) and the CD8+ T lymphocytes were observed predominantly around glandular epithelial cells (arrows). Original magnification, ×100; bottom panels, ×400.

Close modal

To confirm the above observation, we counted the number of lymphocytes located around the acinar cells. As shown in Fig. 3, left, the number of lymphocytes located around the acinar cells was significantly increased in SS as compared with non-SS patients (6.3 ± 1.5 vs 0.1). More importantly, >90% of these lymphocytes, located around the acinar cells, were CD8 positive. In contrast, the number of CD4+ T cells was three times greater than that of CD8+ T cells in the tissue section of SS patients (Fig. 3, right), indicating that the increased number of CD8+ T cells around acini was not merely reflected to the subpopulation of infiltrated lymphocytes in the exocrine glands. On the other hand, B lymphocytes as determined by anti-CD19 mAb were located predominantly in the interstitial area as lymphocytic foci, which were <10% of the aggregates (data not shown). Many reports have shown that a substantial number of B cells are detected in the lymphocytic infiltrates (1, 17). Because it has been reported that the number of B cells in the affected tissue is increasing when the duration of the illness is longer (18), one possible explanation would be that it was shorter in the patients under study than that in the previous reports.

FIGURE 3.

The number of lymphocytes located around the acinar epithelial cells (left) or total number of lymphocytes (right) in the biopsy specimens was counted. The average number of lymphocytes adhered per single acinus or that within a square millimeter is shown. ∗, p < 0.05.

FIGURE 3.

The number of lymphocytes located around the acinar epithelial cells (left) or total number of lymphocytes (right) in the biopsy specimens was counted. The average number of lymphocytes adhered per single acinus or that within a square millimeter is shown. ∗, p < 0.05.

Close modal

These results indicate that the majority of the lymphocytes infiltrating around the acinar cells in SS are CD8+ T lymphocytes, whereas CD4+ T lymphocytes are rarely observed in this area.

Because the ductal epithelial cells have been reported to be a primary target in the immune attack by the lymphocytes (1), one may wonder whether the lymphocytes next to the ductal epithelial cells have a role in damaging glands, as same as those around acinar cells. In the representative figure, we sometimes found the CD8+ T lymphocytes next to the ducts. However, it is hard to see the ducts consistently in a given field of the lacrimal gland section, which are much smaller than the minor salivary glands, and we could not count the number of the lymphocytes around the ducts.

Cell-to-cell adhesion is mediated by a particular set of adhesion molecules and their ligand (19). Accordingly, several pairs of adhesion molecules may mediate the adhesion of CD8+ T lymphocytes to the acinar cells of SS patients. We therefore examined the expression of the adhesion molecules on lacrimal gland samples (Fig. 4). A unique integrin adhesion molecule, αEβ7 (CD103) is known to be expressed on intestinal intraepithelial T lymphocytes, predominantly on the CD8+ phenotype. Moreover, the αEβ7 integrin is induced on peripheral blood T lymphocytes by in vitro activation with specific Ags, lectin, and MLR (20, 21, 22). Because the majority of the infiltrating lymphocytes were positive for CD8 and activated, the samples were stained for αEβ7 (CD103). None of the non-SS lymphocytes that adhered to acinar cells expressed αEβ7 (Fig. 4,c), whereas the lymphocytes infiltrating around SS acinar cells at sites of inflammation strongly expressed αEβ7 integrin (Fig. 4,d). Some αEβ7-positive signals were detected near the structures that appear to be ducts in HE-stained pictures of the sequential section (Fig. 4 b) in this representative figure.

FIGURE 4.

Immunofluorescent staining for surface structures in lacrimal gland tissues. Sections of lacrimal gland biopsy specimens from non-SS patients (left) and SS patients (right) were stained with HE (a and b), anti-αEβ7 (CD103) (c and d), anti-ICAM-1 (e and f), anti-HLA-DR (g and h), and anti-E-cadherin (i and j). Original magnification, ×400. Arrow, positively stained cells.

FIGURE 4.

Immunofluorescent staining for surface structures in lacrimal gland tissues. Sections of lacrimal gland biopsy specimens from non-SS patients (left) and SS patients (right) were stained with HE (a and b), anti-αEβ7 (CD103) (c and d), anti-ICAM-1 (e and f), anti-HLA-DR (g and h), and anti-E-cadherin (i and j). Original magnification, ×400. Arrow, positively stained cells.

Close modal

Given the pivotal role of LFA-1 and ICAM-1 in the adhesion of CTL to target cells (19), the samples were stained with anti-ICAM Ab. Abundant ICAM-1 was detected in the aggregated lymphocytes in the region of the focus (Fig. 4,f), but only minimal expression was noted on the acinar cells, consistent with the previous results (23). This staining pattern was in sharp contrast to that of αEβ7 expression (Fig. 4,d). Similar expression of ICAM-1 was observed in non-SS samples (Fig. 4 e). These findings may indicate that the role of LFA-1/ICAM-1 in the adhesive interaction between CD8+ T lymphocytes and acinar cells is not of primary importance.

Because VCAM-1 was not detected on the acinar cells (data not shown), but intensely expressed on the venules (24), VLA-4/VCAM-1 interaction could not be responsible for the adhesion in SS patients. HLA-DR Ag was found to be expressed on the luminal surface of the acinar cells of the lacrimal glands of SS patients (Fig. 4,h), but it was not detected in non-SS dry eye patients (Fig. 4 g). The pattern of expression was similar to that of HLA-DP and -DQ molecules (data not shown).

Recently, Cepek et al. has reported that E-cadherin is a ligand for αEβ7 and that these molecules mediate adhesion between T lymphocytes and epithelial cells in intestinal epithelium (25). The expression of E-cadherin is restricted to certain types of cells, such as those of epithelial origin (26). The expression of E-cadherin in lacrimal glands from non-SS patients and SS patients is shown in Fig. 4, i and j, respectively. E-cadherin was expressed on the lateral and luminal surfaces of acinar cells, and there was no difference in the pattern of staining between SS and non-SS, whereas the intensity of the staining was much higher in SS than non-SS tissue. The ductal epithelial cells, as shown in the middle of the Fig. 4 j, were clearly stained by anti-E-cadherin. The same results were observed in the other sections from non-SS patients (data not shown), suggesting that not only acinar, but also ductal epithelial cells are expressing E-cadherin.

Considered as a whole, these findings indicate that αEβ7-E-cadherin interaction may play an important role in the adhesive interaction between the CD8+ T lymphocytes and the glandular epithelium, whereas the T lymphocytes aggregated as dense foci may rely on LFA-1/ICAM-1 adhesion molecules for homophilic adhesion. These conclusions are supported by the previous observation that in vitro adhesion of T lymphocytes to mammalian epithelial cell lines is primarily mediated by αEβ7/E-cadherin and that LFA-1/ICAM-1 is responsible for only 20% of the adhesion (16).

To confirm these observations, we set up an in vitro adhesion assay system, in which αEβ7-positive lymphocytes activated in vitro by PHA or nonactivated PBL as a control were added to HSG cells. Flow cytometric analysis showed that the HSG cells expressed the ICAM-1 and E-cadherin on their surface and that the mean fluorescence intensity of these Ags was comparable (data not shown). mAbs against αEβ7 and E-cadherin inhibited the adhesion of the αEβ7-positive lymphocytes to the HSG cells, whereas anti-ICAM-1 and anti-LFA-1 Ab could not inhibit adhesion (Fig. 5). These findings confirmed that αEβ7/E-cadherin plays an indispensable role in the adhesion.

FIGURE 5.

Inhibition of adhesion of PHA-activated PBLs to HSG epithelial cells by anti-αEβ7. Nonactivated or activated PBLs were added to HSG cells in the presence or absence of mAb to αEβ7, LFA-1, or ICAM-1. Results are expressed as the means + SEM of triplicate samples. ∗, p < 0.05.

FIGURE 5.

Inhibition of adhesion of PHA-activated PBLs to HSG epithelial cells by anti-αEβ7. Nonactivated or activated PBLs were added to HSG cells in the presence or absence of mAb to αEβ7, LFA-1, or ICAM-1. Results are expressed as the means + SEM of triplicate samples. ∗, p < 0.05.

Close modal

Here, one can raise a question of whether the acinar cells are killed by the CD8+ T lymphocytes. Given the observation that apoptosis of acinar cells is induced in SS patients (8), we investigated this possibility by using anti-APO2.7 mAb, which recognizes a novel mitochondrial membrane protein specifically induced in apoptotic cells (27). Recently, it was shown that this mAb is a useful and early marker for apoptosis, not only in hemopoietic cells but also in epithelial cells (28, 29, 30). As shown in a representative case, the acinar cells of a SS patient were stained strongly with anti-APO2.7 (Fig. 6,d), whereas those of a non-SS patient were rarely stained (Fig. 6,c). HE-stained overall histology of the sequential sections (Fig. 6, a and b) showed that the cell stained by APO2.7 in SS tissue appeared to be acinar epithelial cells. Furthermore, TUNEL staining showed that a significant but less intense staining was detected in the acinar epithelial cells in this representative SS patient (Fig. 6,f), whereas no positive signal was observed in the non-SS patient (Fig. 6,e). Because anti-APO2.7 Ab has been reported to detect the early phase of apoptosis (27), it allowed the identification of apoptotic acinar cells that appeared normal in conventional HE-stained sections or were not stained by TUNEL from SS patients. This may explain the normal histological appearance of acinar cells in lacrimal gland biopsy specimens from SS patients who failed to produce any reflex tears (11). Double staining with PE-labeled anti-CD8 and FITC-labeled anti-APO2.7 clearly showed that the acinar cells to which CD8+ T lymphocytes adhered by stained positively with anti-APO2.7 (Fig. 7,a). Around 50% of the acinar cells stained by anti-APO2.7 were in contact with CD8+ T cells (data not shown). One-half of the APO2.7-positive acinar cells were not adhered to by CD8+ T cells, suggesting that the other potential inducers can be involved in this process. In sharp contrast, red-labeled CD4+ T cells were located in the interstitial areas, whereas no CD4+ T cells were detected around the green-labeled acinar cells (Fig. 7 b). These results suggest that CD8+αEβ7+ T lymphocytes adhere to and kill acinar epithelial cells by inducing apoptosis.

FIGURE 6.

Immunofluorescent staining with anti-APO2.7. Lacrimal gland sections from non-SS patients (left) and SS patients (right) stained with HE in the upper panel (a and b), anti-APO2.7 in the middle panel (c and d), and with TUNEL in the lower panel (e and f). Original magnification: ×400.

FIGURE 6.

Immunofluorescent staining with anti-APO2.7. Lacrimal gland sections from non-SS patients (left) and SS patients (right) stained with HE in the upper panel (a and b), anti-APO2.7 in the middle panel (c and d), and with TUNEL in the lower panel (e and f). Original magnification: ×400.

Close modal
FIGURE 7.

Double staining of apoptotic cells with FITC-labeled anti-APO2.7 (green) and CD4+ or CD8+ T lymphocytes with PE-labeled anti-CD4 or CD8 (red) in the lacrimal gland of an SS patient. The apoptotic acinar cells are surrounded by adherent CD8+ T lymphocytes. Left, merge scan (×100); middle, double staining (×100); right, double staining (×400). Arrow, CD8+ T lymphocytes around acinar cells positively stained with anti-APO2.7.

FIGURE 7.

Double staining of apoptotic cells with FITC-labeled anti-APO2.7 (green) and CD4+ or CD8+ T lymphocytes with PE-labeled anti-CD4 or CD8 (red) in the lacrimal gland of an SS patient. The apoptotic acinar cells are surrounded by adherent CD8+ T lymphocytes. Left, merge scan (×100); middle, double staining (×100); right, double staining (×400). Arrow, CD8+ T lymphocytes around acinar cells positively stained with anti-APO2.7.

Close modal

Cytotoxic T cells kill target cells by two major pathways, the Fas/FasL pathway and the perforin/granzyme pathway (31, 32, 33). Fig. 8 shows the overall histology after HE staining (left) and the immunofluorescent images of the sequential section (right). As shown in Fig. 8, b and d, the Fas Ag was expressed on the acinar cells, not only of SS patients, but of non-SS patients, although the level of expression was higher in SS. FasL was exclusively expressed on the lymphocytic aggregates and detected, to a lesser degree, on the T lymphocytes around the acini in SS, whereas it was faintly expressed in the interstitial areas in non-SS. Perforin and granzyme B, however, were expressed around the acinar cells to which CD8+αEβ7+ T lymphocytes were in close contact, suggesting that the perforin/granzyme B pathway is involved in damaging the acinar cells by cell-cell contact. Average fluorescent intensity of these images showed that expression of all the mediators are significantly enhanced in SS, compared with non-SS patients (Fig. 9).

FIGURE 8.

Immunofluorescent staining for molecules participating in apoptosis of glandular epithelial cells. Sections of lacrimal gland tissue from non-SS patients (left two rows) and SS patients (right two rows) were stained with anti-Fas (b and d), anti-FasL (f and h), anti-granzyme B (f and j), or anti-perforin (n and p). HE-stained sequential sections were shown in the left side of the each ACAS pictures (a, c, e, g, i, k, m, and o). Original magnification, ×400.

FIGURE 8.

Immunofluorescent staining for molecules participating in apoptosis of glandular epithelial cells. Sections of lacrimal gland tissue from non-SS patients (left two rows) and SS patients (right two rows) were stained with anti-Fas (b and d), anti-FasL (f and h), anti-granzyme B (f and j), or anti-perforin (n and p). HE-stained sequential sections were shown in the left side of the each ACAS pictures (a, c, e, g, i, k, m, and o). Original magnification, ×400.

Close modal
FIGURE 9.

Average fluorescent intensity of signals from the death mediators expressed on the biopsy samples. An indirect immunofluorescent study using anti-granzyme B, anti-perforin, anti-Fas, and anti-FasL was conducted by confocal image analyzer (ACAS 570), and average fluorescent intensity was determined by scanning five randomly selected fields of the tissue section. ∗, p < 0.05.

FIGURE 9.

Average fluorescent intensity of signals from the death mediators expressed on the biopsy samples. An indirect immunofluorescent study using anti-granzyme B, anti-perforin, anti-Fas, and anti-FasL was conducted by confocal image analyzer (ACAS 570), and average fluorescent intensity was determined by scanning five randomly selected fields of the tissue section. ∗, p < 0.05.

Close modal

Among the growing complexity in the mechanism of autoimmune diseases, the potential role of CD8+ T lymphocytes to mediate disease is claimed (9, 10). In autoimmune inflammatory diseases of skeletal muscle, clonally expanded autoreactive CD8+ T cells have been demonstrated to contact and invade muscle fibers, by releasing perforin toward the contact areas (34). This suggests that the recognition-linked death effector system would be responsible for the mechanism of autoimmune attack in human disease. In SS, it has been shown that T cells with a common TCR clonotype accumulate in the lacrimal and labial salivary glands (35), suggesting that these clonally expanded T cells recognize putative autoantigen(s) on exocrine glands at sites of inflammation (36). Binding to TCR with Ag in the context of MHC or cross-linking of TCR has been demonstrated to result in polarized secretion of cytotoxic granules toward the target cells (37, 38). These observations are consistent with our detection of perforin and granzyme B predominantly around the acinar cells, but not lymphocytic foci.

For efficient killing of target cells, a pair/pairs of adhesion molecules are required to enhance contact between the T and target cells in addition to Ag recognition by TCR. In this regard, our results suggest that the interaction between αEβ7 and E-cadherin play a key role in inducing apoptosis in the acinar epithelial cells. Because E-cadherin is expressing on the ductal epithelial cells, we could speculate that accelerated apoptosis by the similar mechanism may be responsible for ductal injury in SS patients. However, in this study, we could not extensively examine the apoptosis of ductal epithelial cells to draw a definitive conclusion because of the limited sample size of lacrimal glands.

αEβ7, a novel integrin, is abundantly expressed on the intestinal intraepithelial cells, particularly on the CD8+ T lymphocytes (39). Although <5% of peripheral blood T cells express αEβ7, stimulation by lectin, cytokines such as TGF-β and IL-2, and MLR up-regulate the expression of αEβ7 on up to 30–50% of T cells (20, 21). Because our previous study has demonstrated that the expression of αEβ7 was not elevated on peripheral blood T cells from SS patients (15), those CD8+ T cells expressing αEβ7 in the exocrine glands appeared to be not simply due to the increase in the peripheral blood. Rather, it is tempting to speculate that the resident CD8+ T cells expressing αEβ7 or CD8+ T cells acquiring αEβ7 in situ are expanded in the exocrine glands. This notion is consistent with the above observations demonstrating the clonally expanded T cells in the glands.

It is now proposed that epithelial cells are the targets of the immunological injury in certain autoimmune disorders and that they be termed “autoimmune epithelitis” (40). They include Hashimoto’s thyroiditis, primary biliary cirrhosis, type 1 diabetes mellitus, and SS. Because the target cells in these diseases are of epithelial origin, the adhesive interactions between αEβ7 and E-cadherin may involve a common mechanism, giving rise to enhanced contact between CD8+ T lymphocytes and epithelial cells.

One-half of the APO2.7-positive cells are not in contact with CD8+αEβ7+ T lymphocytes, even though they express E-cadherin, a ligand for αEβ7. These results indicate that the case of induction of apoptosis by means unrelated to contact with CD8+ cells is at least as strong as the case for that induced by CD8+αEβ7+ T lymphocytes. Because the adhesive interaction between the T and the target cells would facilitate the killing by cytotoxic granules or Fas/FasL pathway, the other killing mechanism may be responsible for the apoptosis of epithelial cells. In this regard, it has been reported that hepatitis C virus-specific human CD8+ T lymphocytes kill bystander target cells as well as the Ag-bearing specific targets (41). Whereas the Ag-specific killing is mediated by Fas/Fas ligand, TNF-α, and cytotoxic granules, bystander killing is primarily mediated by Fas/FasL and TNF-α, but not by cytotoxic granules. Although the effective bystander killing requires a close contact between T and target cells, TNF-α released from the CD8+ T lymphocytes mediates lysis without close contact (41). These results suggest a possible role of TNF-α in the SS patients. Given the demonstration that the level of TNF-α in the saliva and salivary glands is increased (42), it should be determined that TNF-α is produced from CD8+ T lymphocytes contacted with acini, inducing the bystander killing near the contact sites.

Recently, destruction of target cells by a suicide mechanism in which both Fas and FasL are induced on the β cells or thyrocytes, leading to apoptosis of these cells, has been shown in NOD and Hashimoto’s thyroiditis (43, 44). Cytokines such as IL-1β or direct contact with CD8+ T lymphocytes are postulated to up-regulate the death-related structures on target cells. Kong et al. (8) demonstrated that Fas-FasL system on the acinar and ductal epithelial cells play a key role in inducing apoptosis in the epithelium. The present results support their observations. In addition, it is plausible that CD8+αEβ7+ T lymphocytes bearing FasL, by adhering to acinar cells, induce apoptosis of the acinar cells expressing Fas. Alternatively, perforin/granzyme may have a role in inducing apoptosis in the acinar epithelial cells. Given the recent demonstration that granzyme B can directly activate a key caspase, CPP32 (45), ultimately leading to DNA strand break, our results showing that the expression of perforin and granzyme B was up-regulated around the acinar cells support the notion. Taken together, both the Fas/FasL and perforin/granzyme B pathway may contribute to destruction of glandular epithelial cells in SS, as suggested by target cell lysis by CTL (46, 47).

We thank Ms. Kimie Tadano for technical assistance and Dr. Tsubura Suzuki and Tetsuo Shimoyama for obtaining labial samples.

1

This work was supported by a Grant-in-Aid for Scientific Research (C), a Ministry of Education, Science and Culture grant, a grant from the Ministry of Health and Welfare, Japan, and Maruki Memorial Research Foundation Grant 95001.

3

Abbreviations used in this paper: SS, Sjögren’s syndrome; FasL, Fas ligand; NOD, nonobese diabetic; NST, nasal stimulation test; HE, hematoxylin-eosin; HSG cells, cells from an epithelial cell line established from human submandibular glands.

1
Fox, R. I..
1994
. Epidemiology, pathogenesis, animal models, and treatment of Sjögren’s syndrome.
Curr. Opin. Rheumatol.
6
:
501
2
Hikichi, T., A. Yoshida, K. Tsubota.
1993
. Lymphocytic infiltration of the conjunctiva and the salivary gland in Sjögren’s syndrome.
Arch. Ophthalmol.
111
:
21
3
Xu, K. P., S. Katagiri, T. Takeuchi, K. Tsubota.
1996
. Biopsy of labial salivary glands and lacrimal glands in the diagnosis of Sjögren’s syndrome.
J. Rheumatol.
23
:
76
4
McCartney-Francis, N. L., D. E. Mizel, R. S. Redman, M. Frazier-Jessen, R. B. Panek, A. B. Kulkarni, J. M. Ward, J. B. McCarthy, S. M. Wahl.
1996
. Autoimmune Sjögren’s-like lesions in salivary glands of TGF-β1-deficient mice are inhibited by adhesion-blocking peptides.
J. Immunol.
157
:
1306
5
Alpert, S., H. I. Kang, I. Weissman, R. I. Fox.
1994
. Expression of granzyme A in salivary gland biopsies from patients with primary Sjögren’s syndrome.
Arthritis Rheum.
37
:
1046
6
Tsubota, K., I. Saito, N. Miyasaka.
1994
. Granzyme A and perforin expressed in the lacrimal glands of patients with Sjögren’s syndrome.
Am. J. Ophthalmol.
117
:
120
7
Fox, R. I., F. V. Howell, R. C. Bone, P. Michelson.
1984
. Primary Sjögren’s syndrome: clinical and immunopathologic features.
Semin. Arthritis Rheum.
14
:
77
8
Kong, L., N. Ogawa, T. Nakabayashi, G. T. Liu, E. D’Souza, H. S. McGuff, D. Guerrero, N. Talal, H. Dang.
1997
. Fas and Fas ligand expression in the salivary glands of patients with primary Sjögren’s syndrome.
Arthritis Rheum.
40
:
87
9
Tisch, R., H. McDevitt.
1996
. Insulin-dependent diabetes mellitus.
Cell
85
:
291
10
Benoist, C., D. Mathis.
1997
. Cell death mediators in autoimmune diabetes: no shortage of suspects.
Cell
89
:
1
11
Tsubota, K., K. P. Xu, T. Fujihara, S. Katagiri, T. Takeuchi.
1996
. Decreased reflex tearing is associated with lymphocytic infiltration in lacrimal glands.
J. Rheumatol.
23
:
313
12
Vitali, C., S. Bombardieri, H. M. Moutsopoulos, G. Balestrieri, W. Bencivelli, R. M. Bernstein, K. B. Bjerrum, S. Braga, J. Coll, S. de Vita, et al
1993
. Preliminary criteria for the classification of Sjögren’s syndrome: results of a prospective concerted action supported by the European Community.
Arthritis Rheum.
36
:
340
13
Kohler, P. F., M. E. Winter.
1985
. A quantitative test for xerostomia: the Saxon test, an oral equivalent of the Schirmer test.
Arthritis Rheum.
28
:
1128
14
Takeuchi, T., K. Amano, H. Sekine, J. Koide, T. Abe.
1993
. Upregulated expression and function of integrin adhesive receptors in systemic lupus erythematosus patients with vasculitis.
J. Clin. Invest.
92
:
3008
15
Pang, M., T. Abe, T. Fujihara, S. Mori, K. Tsuzaka, K. Amano, J. Koide, T. Takeuchi.
1998
. Upregulation of αEβ7, a novel integrin adhesion molecule, on T cells from systemic lupus erythematosus patients with specific epithelial involvement.
Arthritis Rheum.
41
:
1456
16
Cepek, K. L., C. M. Parker, J. L. Madara, M. B. Brenner.
1993
. Integrin αEβ7 mediates adhesion of T lymphocytes to epithelial cells.
J. Immunol.
150
:
3459
17
Fox, R. I., T. E. Hugli, L. L. Lanier, E. L. Morgan, F. Howell.
1985
. Salivary gland lymphocytes in primary Sjögren’s syndrome lack lymphocyte subsets defined by Leu 7 and Leu 11 antigens.
J. Immunol.
135
:
207
18
Pepose, J. S., R. F. Akata, S. C. Pflugfelder, W. Voigt.
1990
. Mononuclear cell phenotypes and immunoglobulin gene rearrangements in lacrimal gland biopsies from patients with Sjögren’s syndrome.
Ophthalmology
97
:
1599
19
Springer, T. A..
1990
. Adhesion receptors of the immune system.
Nature
346
:
425
20
Schieferdecker, H. L., R. Ullrich, A. N. Weiss Breckwoldt, R. Schwarting, H. Stein, E. O. Riecken, M. Zeitz.
1990
. The HML-1 antigen of intestinal lymphocytes is an activation antigen.
J. Immunol.
144
:
2541
21
Parker, C. M., K. L. Cepek, G. J. Russell, S. K. Shaw, D. N. Posnett, R. Schwarting, M. B. Brenner.
1992
. A family of β7 integrins on human mucosal lymphocytes.
Proc. Natl. Acad. Sci. USA
89
:
1924
22
Cerwenka, A., D. Bevec, O. Majdic, W. Knapp, W. Holter.
1994
. TGF-β1 is a potent inducer of human effector T cells.
J. Immunol.
153
:
4367
23
St. Clair, E. W., J. C. Angellilo, K. H. Singer.
1992
. Expression of cell-adhesion molecules in the salivary gland microenvironment of Sjögren’s syndrome.
Arthritis Rheum.
35
:
62
24
Saito, I., K. Terauchi, M. Shimuta, S. Nishiimura, K. Yoshino, T. Takeuchi, K. Tsubota, N. Miyasaka.
1993
. Expression of cell adhesion molecules in the salivary and lacrimal glands of Sjögren’s syndrome.
J. Clin. Lab. Anal.
7
:
180
25
Cepek, K. L., S. K. Shaw, C. M. Parker, G. J. Russell, J. S. Morrow, D. L. Rimm, M. B. Brenner.
1994
. Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the αEβ7 integrin.
Nature
372
:
190
26
Takeichi, M..
1991
. Cadherin cell adhesion receptors as a morphogenetic regulator.
Science
251
:
1451
27
Zhang, C., Z. Ao, A. Seth, S. F. Schlossman.
1996
. A mitochondrial membrane protein defined by a novel monoclonal antibody is preferentially detected in apoptotic cells.
J. Immunol.
157
:
3980
28
Koester, S. K., P. Roth, W. R. Mikulka, S. F. Schlossman, C. Zhang, W. E. Bolton.
1997
. Monitoring early cellular responses in apoptosis is aided by the mitochondrial membrane protein-specific monoclonal antibody APO2.7.
Cytometry
29
:
306
29
Yaguchi, M., K. Miyazawa, M. Otawa, T. Katagiri, J. Nishimaki, Y. Uchida, O. Iwase, A. Gotoh, Y. Kawanishi, K. Toyama.
1998
. Vitamin K2 selectively induces apoptosis of blastic cells in myelodysplastic syndrome: flow cytometric detection of apoptotic cells using APO2.7 monoclonal antibody.
Leukemia
12
:
1392
30
De Saint Jean, M., F. Bringnole, G. Feldmann, C. Baudouin.
1998
. Apoptosis and expression of inflammation-related proteins following IFN γ treatment of epithelial conjunctival cells in vitro.
Invest. Ophthalmol. Visual Sci.
39
:
548
31
Doherty, P. C..
1993
. Cell-mediated cytotoxicity.
Cell
75
:
607
32
Berke, G..
1994
. The binding and lysis of target cells by cytotoxic lymphocytes: molecular and cellular aspects.
Annu. Rev. Immunol.
12
:
735
33
Liu, C. C., P. M. Persechini, J. D. Young.
1995
. Perforin and lymphocyte-mediated cytolysis.
Immunol. Rev.
146
:
145
34
Goebels, N., D. Michaelis, M. Engelhardt, S. Huber, A. Bender, D. Pongratz, M. A. Johnson, H. Wekerle, J. Tschopp, D. Jenne, R. Hohlfeld.
1996
. Differential expression of perforin in muscle-infiltrating T cells in polymyositis and dermatomyositis.
J. Clin. Invest.
97
:
2905
35
Matsumoto, I., K. Tsubota, Y. Satake, Y. Kita, R. Matsumura, H. Murata, T. Namekawa, K. Nishioka, I. Iwamoto, Y. Saitoh, T. Sumida.
1996
. Common T cell receptor clonotype in lacrimal glands and labial salivary glands from patients with Sjögren’s syndrome.
J. Clin. Invest.
97
:
1969
36
Haneji, N., T. Nakamura, K. Takio, K. Yanagi, H. Higashiyama, I. Saito, S. Noji, H. Sugino, Y. Hayashi.
1997
. Identification of α-fodrin as a candidate autoantigen in primary Sjögren’s syndrome.
Science
276
:
604
37
Liu, C. C., C. M. Walsh, J. D. Young.
1995
. Perforin: structure and function.
Immunol. Today
16
:
194
38
Henkart, P. A., M. S. Williams, H. Nakajima.
1995
. Degranulating cytotoxic lymphocytes inflict multiple damage pathways on target cells.
Curr. Top. Microbiol. Immunol.
198
:
75
39
Cerf Bensussan, N., A. Jarry, N. Brousse, B. Lisowska Grospierre, D. Guy Grand, C. Griscelli.
1987
. A monoclonal antibody (HML-1) defining a novel membrane molecule present on human intestinal lymphocytes.
Eur. J. Immunol.
17
:
1279
40
Moutsopoulos, H. M..
1994
. Sjögren’s syndrome: autoimmune epithelitis.
Clin. Immunol. Immunopathol.
72
:
162
41
Ando, K., K. Hiroishi, T. Kaneko, T. Moriyama, Y. Muto, N. Kayagaki, H. Yagita, K. Okumura, M. Imawari.
1997
. Perforin, Fas/Fas ligand, and TNF-α pathways as specific and bystander killing mechanisms of hepatitis C virus-specific human CTL.
J. Immunol.
158
:
5283
42
Fox, R. I., H. I. Kang, D. Ando, J. Abrams, E. Pisa.
1994
. Cytokine mRNA expression in salivary gland biopsies of Sjögren’s syndrome.
J. Immunol.
152
:
5532
43
Giordano, C., G. Stassi, R. De Maria, M. Todaro, P. Richiusa, G. Papoff, G. Ruberti, M. Bagnasco, R. Testi, A. Galluzzo.
1997
. Potential involvement of Fas and its ligand in the pathogenesis of Hashimoto’s thyroiditis.
Science
275
:
960
44
Chervonsky, A. V., Y. Wang, F. S. Wong, I. Visintin, J. C. A. F. R. A., J. C. A. Janeway, L. A. Matis.
1997
. The role of Fas in autoimmune diabetes.
Cell
89
:
17
45
Darmon, A. J., D. W. Nicholson, R. C. Bleackley.
1995
. Activation of the apoptotic protease CPP32 by cytotoxic T-cell-derived granzyme B.
Nature
377
:
446
46
Kagi, D., F. Vignaux, B. Ledermann, K. Burki, V. Depraetere, S. Nagata, H. Hengartner, P. Golstein.
1994
. Fas and perforin pathways as major mechanisms of T cell-mediated cytotoxicity.
Science
265
:
528
47
Kojima, H., N. Shinohara, S. Hanaoka, Y. Someya Shirota, Y. Takagaki, H. Ohno, T. Saito, T. Katayama, H. Yagita, K. Okumura, et al
1994
. Two distinct pathways of specific killing revealed by perforin mutant cytotoxic T lymphocytes.
Immunity
1
:
357
48
Greenspan, J..
1974
. The histopathology of Sjögren’s syndrome in labial salivary gland biopsies.
Oral Surg. Oral Med. Pathol.
37
:
217