The etiology of SLE has not been fully elucidated but is related to many factors including genetic, environmental, infection, and hormones [22]. The production of BCs and autoantibodies is the key driving factor in SLE pathogenesis but without the proper support of TCs, BCs are ineffective and TCs provide BCs with the necessary costimulatory signals [23]. In the RA patient study, the ratio of CD19+PD-L1+BCs among untreated patients was significantly lower than that in HC patients, whereas after treatment, there was an increased frequency of PD-L1+BCs among patients with a positive response to treatment [20]. However, in patients with melanoma, the expression of B cell PD-L1 increased significantly and was positively correlated with the tumor stage [24]. SLE patients have been reported to have increased PD-1, Ki-67, or PD-L1 expression on CD11c+BCs [25]. Additionally, the ratio of CD4+PD-1+cells in psoriatic patients was much lower than compared to the control group [26]. In summary, PD-1 / PD-L1 expression is disease-specific on immune cells. In our study, there was a higher expression frequency of PD-L1 and PD-1 on CD4+T/ CD19+BCs from the PB of SLE patients compared to HCs, indicating the aberrant expressed pathway of PD-1/PD-L1 during the interaction of CD4+TCs with CD19+BCs.

Ki-67 protein, a common marker of proliferating cells, encodes a 359 KD non-histone nuclear protein [14] and was originally identified as an antigen in the nucleus of Hodgkin’s lymphoma, which is highly expressed in circulating cells but strongly downregulated in quiescent G0 cells [27]. Some studies have found that in early SLE, several expanded lymphocyte populations commonly express Ki-67 [28]. Ki-67+ cell populations, peripheral helper T (Tph) cells as well as age-related B cells (ABCs) remained elevated during the first year, and similar increases were observed in diagnosed SLE, suggesting these pathways to be activated early in SLE and characteristic of SLE pathologic immune responses [28]. Ki-67+NK cells are related to the increased production of autoantibodies, decreased levels of complement, and nephritis in SLE patients [29]. In vitro experiments have proved the influence of IL-15 on upregulating Ki-67 expression in NK cells [29]. Compared to HCs, Ki-67 expression in CD4+TCs and CD19+BCs of PB of patients with SLE was relatively higher. Also, the proportion of CD4+PD-L1-Ki-67+TCs, CD4+PD-1+/-Ki-67+TCs, CD19+PD-L1-Ki-67+, as well as CD19+PD-1-Ki-67+BCs in SLE patients, was higher compared to HCs. This suggests that the proliferative activity of CD4+TCs, CD19+BCs, CD4+PD-L1-TCs, CD4+PD-1+/-TCs, CD19+PD-L1- and CD19+PD-1-BCs was abnormal. In the research of tumors, it is generally believed that Ki67 is considered to be related to the prognosis of patients [30,31,32,33,34], while there are few relevant researches in SLE patients.

Some studies demonstrated among new patients with SLE, an upregulated serum level of PD-1 antibody, which is linked to the disease activity, was observed; besides, PD-1 autoantibody promoted the proliferation of CD4+TCs [12]. As reported by studies on patients with RA, CD19+BCs has a potentical in inhibiting CD8+TCs proliferation as well as cytokine production in a manner depending on PD-L1 [20]. In addition, PD-L1 immature BCs inhibit TCs depending on the level of PD-L1 among patients with melanoma (stage III and IV) [24]. In our study, activated CD19+B in HCs/SLE patients promoted the proliferation and PD-L1 expression in CD4+TCs. Thus, it was speculated that after activation, the negative signal increased and the proliferation decreased because of the increased expression of PD-L1. Therefore, PD-L1 blockers were added to investigate the impact of the axis of PD-1/PD-L1 on the interaction of CD4+TCs with CD19+BCs, showing that CD4+TC proliferation was restored by PD-1/PD-L1 pathway among HCs and SLE patients. In summary, we hypothesized that PD-1/PD-L1 signaling axis exerts a crucial impact on activating pathogenic cell responses of TCs and BCs in SLE patients; Besides, BCs have the potential to regulate the T cell immune response by expressing regulatory molecules including PD-L1.

PD-L1hiB cells can regulate the expansion of TFH cells [35]. PD-1 limits the upregulation of CXCR3 on Tfh cells, thus concentrating cells in the germinal center, where the PD-L1/PD-1 interaction of a single Tfh cell with BCs optimizes the competition as well as affinity maturation of BCs [36]. Also, PD-L1 solely has a potential in inducing the immature CD4+TCs to differentiate into Foxp3-induced regulatory T (iTreg) cells [37]. The upregulation of PD-1 expression among SLE patients with the PTPN22 Trp620 missense allele is related to the decrease of Treg inhibition ability and the increase of T cell proliferation ability [38]. In addition, CTLA-4 expressed by Treg depleted CD80/CD86 through macrophages and released free PD-L1 on antigen-presenting cells [39]. In our study, anti-PD-L1 restored the expression of Tfh but decreased that of Treg among HCs. Therefore, we believe that the inhibition of Tfh and the recovery of Treg expression by PD-1/PD-L1 signaling axis when TCs interacts with BCs is key for TC inhibition along with the maintenance of peripheral tolerance. However, we speculate that based on individual differences or the impairment of PD-1/PD-L1 signaling pathway among SLE patients, the role of the PD-1/PD-L1 signaling pathway in the expression frequency of Tfh/Treg is not statistically significant.

As reported by experimental studies, PD-1+/-TCs are beneficial for the increase of IFN-γ/IL-17 production [39]. In studies of melanoma patients, B cell subsets could suppress IFN-γ expression by TCs in a manner depending on PD-L1 [24]. Several researchers have also suggested that the IFN-γ gene may occur early in patients with SLE [40], possibly playing a critical role in LN [41]. IL-10 can specifically block B cell responses in SLE mice [42]. It is commonly believed that IL-10 is paradoxical pathogenic in SLE. Our results proved that PD-L1 inhibition inhibited IL-10 / IFN-γ secretion, therefore, roles of IFN-γ and IL-10 for SLE cannot be ignored.

The roles of PD-L1 are various in immune regulation. Besides, PD-1/PD-L1 signaling axis has complex pathogenicity in SLE. In addition to ITIM (inhibitory motif), the intracellular segment of PD-1 also has a structure called immune receptor complex tyrosine conversion motif (ITFM) which can bind tyrosine kinase or tyrosine phosphatase (such as SHP-2) according to the existence of a binding protein called SAP, to initiate negative regulation. The classical PD-1 pathway is related to the trans-interaction of PD-1 (on TCs) with PD-L1 (on APC or tumor cells), negatively regulating the immune response after the activation of TCs, thereby maintaining immune tolerance [12]. Recently, the existence of cis-interactions of PD-1 and APCs has been revealed on the same tumor cells or PD-L1 [43]. In addition, blocking [44] or activating [45] PD-1 can improve the survival of lupus mice. In addition, the combination of PD-1 agonist and low-dose IL-2 may have better therapeutic efficacy in SLE [46]. The susceptibility to SLE is also related to PD1.3/PD1.5 polymorphism, while the PD1.6 polymorphism is possibly protective for SLE [47].

In addition, this study has some limitations. (1) TCs and BCs were isolated for co-culture in vitro, while there are countless other cell types potentially expressing PD-L1, including dendritic cells, macrophages as well as bone marrow suppressor cells in the in vivo microenvironment. (2) In future studies, each cell type should be blocked by PD-L1 to distinguish effects of PD-L1 blockers on BCs or TCs. (3) However, its mechanism of action and its relationship with PD-1/PD-L1 signaling axis need to be further studied. Reseach has shown that PD-L1-Ig can inhibit the formation of Th17 cells in several organs such as spleen and the kidney, clear the abnormal production of cytokines (IFN-γ, IL-17, and IL-10) and anti-dsDNA autoantibodies in the serum, inhibit the deposition of IgG in the glomeruli with decreased proteinuria, and activate TCs in the urine [45]. In addition, PD-L1 binding to PD-1 on the surface of CD8+TCs, weakens the function of cytotoxic CD8+TCs and promotes immune escape in nasopharyngeal carcinoma [48]. In conclusion, both dysfunction of the cytokine IL-17 and regulatory CD8+TCs are involved in the pathogenesis of SLE. However, its mechanism of action and its relationship with PD-1/PD-L1 signaling axis need to be further investigation.