*4.3. The Detection of HTLV-1 Genes and Proteins in SGs in SS*

Similar to the involvement of EBV in SS, the infection of salivary glands by HTLV-1 has also been debated. In 1992, it was reported that the expression of HTLV-1 p19 (a member of the family of gag antigens) was observed in SGECs from 31% of SS patients, 24% of RA patients with SS, and 21% of patients with sicca [105], suggesting the possibility of the presence of an endogenous retrovirus other than the human endogenous retroviral sequence (HRES-1). Mariette et al. [106] reported that *tax* gene (but not *gag*, *pol*, *env*) was detected by both ISH and PCR in labial salivary glands of two of nine patients with SS. They also observed the localization of *tax* in the nuclei of both SGECs and lymphocytes. Sumida et al. [107] later described similar findings with the positive expression of *tax* gene in labial salivary glands from Japanese patients with SS, and they reported that the nucleotide

sequence of the *pX* region in labial salivary glands was completely identical to that of the MT-2 cell line, suggesting that the findings of these two studies [106,107] support the concept that at least *tax* gene might be involved in the pathogenesis of SS.

In an aforementioned study conducted in Nagasaki [113], the serum titers of anti-HTLV-1 antibody in patients with HTLV-1-seropositive SS was as high as that in HAM patients, and IgM-class antibodies were frequently detected in the former group. Interestingly, salivary IgA antibodies to HTLV-1 were observed in HTLV-1-seropositive SS patients although a low frequency of these antibodies was observed in the HAM patients and healthy carriers. Four responses to that study [113] arose from France and Austria [108], in which there were similar observations regarding the detection of these antibodies with additional observations in HTLV-1-seropositive SS patients. In these responses, the involvement of *tax* gene and the local synthesis of IgA class anti-HTLV-1 antibody in saliva was discussed. In contrast, a study of salivary glands from 49 patients with SS in Great Britain demonstrated no *tax* expression unlike that of ERV-3, an endogenous human retrovirus [110], suggesting differences in regional characteristics between endemic and non-endemic areas as well as the possibility of PCR contamination.

With respect to HTLV-1-seropositive patients with SS, an investigation using an in situ PCR technique detected HTLV-1 proviral DNA in infiltrating MNCs but not salivary gland epithelial cells of these patients [121]. In contrast, we detected *tax* gene in 18% of the salivary glands of HTLV-1-seronegative SS patients by nested PCR [122], suggesting that the involvement of HTLV-1 in HTLV-1-seronegative patients might be limited because the viral load in salivary glands from HTLV-1-seronegative SS patients was much lower than that from HTLV-1-seropositive patients. A study conducted in France and using ISH detected HTLV-1 RNA in both salivary gland epithelial cells and lymphoid cells of HTLV-1-seronegative SS patients, although *tax* gene was expressed in LSGs from these patients [123]. Sasaki et al. later described the presence of TCR Vbeta gene in infiltrating lymphocytes of salivary glands from HTLV-1-seropositive SS patients, in which specific T cells with Vbeta5.2, 6, and 7 were used, indicating the involvement of an HTLV-1-driven T-cell activation in SS [124].

As for other inflammatory conditions, Mariette et al. examined the expression of *tax* gene in LSGs from SS patients and patients with chronic inflammatory diseases [125]. They detected *tax* gene in 30% of the SS patients and 28% of the non-SS patients. These results suggested that HTLV-1 may contribute to the development of chronic inflammation regardless of the presence/absence of SS. Although the number of publications concerning HTLV-1-related genes in the salivary glands of individuals with SS has decreased since 2000, a 2012 report from Korea noted that *tax* gene was detected in 3.8% of a series of SS patients, and HTLV-1 p19 or Tax protein was revealed by IHC in LSGs from >40% of the SS patients [126]. With regard to HTLV-1 genes other than the previously reported genes, we examined the expressions of HTLV-1 bZIP factor (*HBZ*) and *tax* gene by ISH, which showed that both genes expressed in infiltrating MNCs and SGECs from HAM-SS patients and a patient with adult T-cell leukemia (ATL), although the expression of *tax* gene was dominant in MNCs of the HAM-SS patients [127] (Figure 1A,B).

*HBZ* gene, which is encoded by a minus strand of the HTLV-1 provirus, is known to induce chronic inflammation in ATL through the expression of Foxp3, and it was suggested that the expression of *HBZ* has the potential to induce Foxp3 gene transcription in CD4<sup>+</sup> T cells from *HBZ* transgenic mice [111,128]. In accord with these experimental observations, we also observed a high expression of Foxp3 in salivary gland tissue from a patient with ATL and patients with HAM-SS. The involvement of *HBZ* gene in HTLV-1-seropositive SS patients was newly considered after the observed outbreak of chronic inflammation in this population.

**Figure 1.** The expression of *tax*/*HBZ* and HTLV-1 virions in salivary glands (SGs) of patients with Sjögren's syndrome (SS). (**A**) Massive lymphocytic infiltration was by hematoxylin-eosin staining in a labial salivary gland from a patient with sicca symptoms and adult T-cell leukemia (ATL). The expression of *tax*/HTLV-1 bZIP factor (*HBZ*) in SGs from a patient with ATL (**B**) and patients with HTLV-1-associated myelopathy complicated with SS (**C**)**,** examined by in situ hybridization. A dominant expression of *HBZ* (*green*) was observed in the ATL SGs in both infiltrating mononuclear cells (MNCs) and ducts (**B**). In contrast, a dominant expression of *tax* (*red*) was observed in MNCs of salivary glands from patients with HAM-SS (**C**). Electron microscopy (**D**) revealed the existence of HTLV-1 virions (*arrowheads*) at the contact face between HCT-5 cells (an HTLV-1-infected cell line) and salivary gland epithelial cells (SGECs).
