Discussion
The critical role of the type I IFN pathway in the pathogenesis of pSS
has been widely addressed [30, 31]. Acting as the premier producer
of IFN-α, the role of pDC in the pathogenesis of pSS still remains
inclusive. In the present study, we have deciphered the redundant role
of pDC in the type I signature and disease development in pSS.
Consistent with previous studies, we have detected a peripheral
reduction in pSS [20, 26]. We have demonstrated a negative feedback
loop, in which, pDC promoted antibody production and in turn, high
immunoglobulin could induce pDC apoptosis. It has been reported that
IgG-complexed adenoviruses induce the apoptosis of human pDCs [32].
Hyperactivated B cells, characteristic autoantibodies (SSA, SSB), and
hypergammaglobulinemia are hallmarks of pSS [33]. Previous studies
have found that pDCs could efficiently promote the differentiation of B
cells into plasmablasts and plasma cells through type I IFNs and IL-6
[34, 35]. Moreover, pDCs could enhance the autoreactivity of B cells
via type I IFNs in a T cell-independent manner [36]. Although our
data showed that, pSS-derived pDC did not display a stronger capacity
for promoting plasma cell differentiation and antibody production, we
gave a good explanation that the excess immunoglobulin (IgG and IgA)
produced by hyperactivated B cells might result in the apoptosis and
reduction of pDC in peripheral blood of pSS. In addition,
Hydroxychloroquine (HCQ) effectively inhibits B cell activation [37]
and a recent systematic review revealed that serum IgA in patients with
pSS decreased significantly after using HCQ [38], which is
consistent with our results that pDC frequency rebounded after HCQ
therapy.
Nonetheless, pDC is noted for its unique ability in producing type I
IFNs, our data suggest that it might not be the culprit of hyperactive
type I IFN signaling in pSS. Recent studies in pSS have emphasized the
critical role of dysregulated epithelial cells [29, 39], especially
the active IFN signaling in salivary gland epithelial cells (SEGCs)
[29]. SEGCs from pSS were found to be sensitive to the stimulation
of TLR agonists and produced type I IFNs as a response to the
stimulation [27, 28, 40]. Type I IFNs released by SEGCs could
further promote the secretion of BAFF in an NF-κB dependent way [12,
27]. These results imply that gland epithelial cells are not purely
innocent victims, but can also be the trigger of type I IFN signaling in
pSS.
In SSc, pDCs directionally migrated to target organs and secreted IFN-α
and CXCL4, thereby accelerated tissue fibrosis [15, 41]. In another
autoimmune disease with fibrosis signature, IgG4-related disease, active
pDCs and relevant IFN-α signaling were also observed in related
pancreas, pDC also enhanced the production of IgG4 by B cells [42].
Dual role of pDCs on IFN-α production and promoting activation of
pathogenic T cells in psoriasis has been revealed in psoriasis [43].
Study in RA mouse model suggested that pDC aggravated joint inflammation
and bone erosion via TLR7 dependent type I IFNs [17]. Reduction of
pDC in the peripheral blood as well as the concomitant infiltration and
activation in related tissue were reported in SLE [44, 45]. A recent
randomized controlled trial in SLE clarified that Litifilimab
(anti-BDCA2 antibody, BDCA2 is an exclusive marker of pDC) is effective
in disease remission [46], which indicated the feasibility of pDC
purge strategy in SLE treatment. Whereas, newly published research
revealed that SLE-derived pDC had senescence and inert phenotype, active
keratinocytes should be responsible for the tupe I IFN signaling rather
than pDC [26]. In general, pDCs contributed to the pathogenesis of
autoimmune diseases through type I IFNs or interplay with other immune
cells [47]. However, further studies are needed to elucidate the
potential mechanism about pDC’s contribution to SLE and pSS.
Our available evidence does not support an overactive phenotype (resting
and activating status) of pDCs from pSS patients, particularly in the
production of IFN-α. However, a previous transcriptional study reported
that pDCs from patients with pSS secreted more type I IFNs after TLR7
stimulation [21]. Antonios et al. ’s work that pDCs from patients
with SLE or pSS did not show stronger secretion ability for inflammatory
cytokines, especially IFNα [26], which is consistent with our
findings. These divergent results may be caused by the difference of
detection methods, further study will help to elucidate underlying
mechanism.
In conclusion, we demonstrated the decreased percentage of pDC in pSS
patients, which might result from the excess immunoglobulin (IgG and
IgA) induced apoptosis. Moreover, we showed that pDC might not be the
major contributor to the hyperactivation of type I interferon signaling
in pSS patients. Our research provides a good addition to pSS
pathogenesis and gives implications that targeting pDC might not be a
good strategy for clinical pSS treatment.