We read with great interest the article by Ogawa et al (1) about the involvement of interferon-γ–induced T cell–attracting chemokines in salivary gland lesions of patients with Sjögren's syndrome (SS). The authors demonstrated that IP-10 and Mig proteins were expressed in the ductal epithelium adjacent to lymphoid infiltrates of SS salivary glands, but not in normal glands. Furthermore, they found that most CD3+ infiltrating lymphocytes in dense periductal foci expressed CXC receptor 3 (CXCR3).
Since the expression of chemokine receptors by salivary epithelial cells was not mentioned, we thought it might be of interest to report our experience. We studied minor salivary gland (MSG) biopsy specimens from 18 patients with primary SS (all female; mean ± SD age 49.0 ± 15.5 years [range 29–76]) fulfilling the new American-European Consensus Group diagnostic criteria for SS (2) and from 12 subjects with sicca symptoms (all female; mean ± SD age 51.0 ± 20.5 years [range 20–75]) but lacking any other features of SS.
Glandular tissues were characterized as containing a significant mononuclear cell infiltration or as normal, by the Chisholm and Mason diagnostic criteria (3). The expression of IP-10 and CXCR3 was determined by immunohistochemistry on frozen sections of gland tissues, using an avidin–biotin–immunoperoxidase system as previously described by us (4).
As shown in Table 1, our findings are not entirely consistent with those of Ogawa et al. Specifically, in our study, the frequency of IP-10–positive cells was similar between SS patients and control subjects. Expression of the CXCR3 chemokine receptor by T lymphocytes was evident in all SS MSGs examined and in MSGs from 4 of 12 controls. Furthermore, CXCR3 was also expressed in the ductal epithelium in all the salivary glands from all of the SS patients and 11 of the 12 controls.
|Intensity of staining†||Lymphocytes, CXCR3||Ductal epithelium|
The positive IP-10 staining found in controls might have been explained by the possible presence of early, preclinical SS or subclinical infection. To rule out the first hypothesis, we performed oral, ocular, and serologic diagnostic examinations. Test results were normal in all control subjects, although 2 of 12 MSG biopsy samples showed a focus score of >1. This finding is not surprising in that the frequency of positive focus scores in our control subjects (17%) is similar to that recently observed in healthy volunteers by Radfar et al (5).
With regard to the possibility of a subclinical bacterial or viral infection, periodontitis was excluded in all of the subjects by a comprehensive oral evaluation. Furthermore, polymerase chain reaction and in situ hybridization were used to analyze MSG biopsies from patients and controls, using primers and probes to specific sequences of cytomegalovirus (CMV), Epstein-Barr virus, and herpes simplex virus. CMV genome was detected in 1 patient with SS and in 4 of the control subjects. The virus was detected not only as genome but also as messenger RNA transcripts (early and late gene), indicative of active viral replication.
Since the selection of controls differed between our case series and that of Ogawa et al, we cannot exclude the possibility that the inclusion of subjects with sicca symptoms may have introduced bias to our study, despite the absence of any other features of SS. However, the expression of CXCR3 in ductal epithelial cells in all of the SS patients and in 11 of 12 controls is intriguing. Given that ductal epithelial cells produce IP-10, it is notable that these cells coexpress CXCR3 on their surface. There are at least two ways to interpret these results. First, IP-10 may act on ductal epithelial cells as an autocrine factor. Second, the chemoattracting signal of IP-10 may be down-regulated by means of the CXCR3 expressed on ductal epithelial cells.
In conclusion, further studies are needed to determine the role of IP-10 and its receptor, CXCR3, in the pathogenesis of SS. However, should findings of this preliminary study be confirmed, they would indicate that CXCR3 expression in ductal epithelial cells might play a role in physiologic settings.