CD155/TIGIT signalling plays a vital role in the regulation of bone marrow mesenchymal stem cell-induced natural killer-cell exhaustion in multiple myeloma
Zhao-Yun Liu,Ling Deng,Yue Jia,Hui Liu,Kai Ding,Wei Wang,Hongkai Zhang,Rong Fu
DOI: https://doi.org/10.1002/ctm2.861
IF: 8.554
2022-01-01
Clinical and Translational Medicine
Abstract:Dear Editor, T-cell immunoreceptor with immunoglobulin and ITIM domain (TIGIT) is a potential immune checkpoint for natural killer (NK)-cell exhaustion in multiple myeloma (MM), and its ligand, CD155 on bone marrow mesenchymal stem cells (BMSCs), is highly expressed in newly diagnosed MM (NDMM) but expressed at extremely low levels on myeloma cells. CD155/TIGIT signalling is involved in the regulation of BMSC-induced NK-cell exhaustion. MM is an incurable disease, and most patients eventually relapse or develop drug-resistant disease. BMSCs, an important cellular component of the bone marrow microenvironment, are critical for cell-based immunotherapy because they modulate the functions of several immune cell types. BMSCs promote myeloma cell malignant proliferation in various ways. Meanwhile, myeloma cells can also reprogram BMSCs, which promotes the formation of a malignant regulatory loop based on BMSC–myeloma cell interactions.1-4 NK cells are lymphocytes that kill tumour cells without prior sensitisation, because the expression of MHC class I is low on myeloma cells and they are easily recognised by NK cells at the early disease stage.5, 6 With disease progression, the expression of inhibitory immune checkpoints on NK cells is increased, leading to NK-cell exhaustion and tumour cell immune escape. TIGIT is expressed on lymphocytes with an inhibitory role.7 Its immunoglobulin domain is similar to the amino terminal domain of CD226, both of which can bind their ligands CD155, CD112 and CD113.8 Competitive analysis showed that TIGIT has a higher affinity for CD155 than CD226, which can inhibit the interaction between CD155 and CD226 in a dose-dependent manner. Zavidij et al. demonstrated immune-state changes in MM during early-to-late-stage transition by single-cell RNA sequencing (10× Genomics). They found that NK cells were increased during MM early stages and correlated with the expression of chemokine receptors.9 Their study revealed an overall change of immune cells in MM, but they did not focus on the inhibitory receptors on NK cells. We re-analysed these sequencing results to explore the expression of TIGIT and CD96 on NK cells, which are both inhibitory receptors. TIGIT was highly expressed in monoclonal gammopathy of undetermined significance (n = 5), low-risk smouldering myeloma (SMM) (n = 3), high-risk SMM (n = 8) and NDMM (n = 7) compared to that in healthy controls (HCs) (n = 9), whereas CD96 expression was not significantly different (Figure 1). Next, we used clinical samples to verify the significant high expression of TIGIT in MM (Table S1). TIGIT expression was higher in NDMM (61.46% ± 12.51%) than in the CR (complete response) group (19.52% ± 13.36%) and in HCs (13.80% ± 5.075%) (both p < .01). Expression of the active receptor CD226 was lower in the NDMM group (50.66% ± 13.59%) than in the CR (84.99% ± 8.390%) and HC (86.43% ± 13.26%) groups (both p < .01; Figure 2B,E). Activated receptors (NKG2D and CD107a) and functional biomarkers (IFN-γ and perforin) on NK cells were decreased significantly in NDMM patients (Figure 2C,E). On NK cells, changes in levels of CD226, TIGIT, NKG2D, CD107a, the function marker perforin and IFN-γ in bone marrow were consistent with those in peripheral blood (Tables S2 and S3). The expression of activated and functional molecules on TIGIT+NK cells was lower than that on TIGIT−NK cells, whereas those on CD226+NK cells were elevated (Figure 2D), indicating that TIGIT expression was significantly increased and correlated with NK-cell modulation. As the interaction between the bone marrow environment and NK cells has been reported,10 we speculated that the expression of CD155, CD112 and CD113, ligands of TIGIT and CD226, might be expressed on BMSCs and even myeloma cells. Interestingly, the expression of CD155 was greatly upregulated on BMSCs from NDMM patients (95.07% ± 3.644%), but CD112 and CD113 were expressed at very low levels (Figure 2F). However, CD155, CD112 and CD113 levels were very low in myeloma cells (Figures 2G and S2C). Furthermore, levels of the NK-cell functional molecules perforin and IFN-γ were negatively associated with CD155 expression on BMSCs. The expression of CD155 was closely associated with MM clinical characteristics (Figure 2H). Considering that the TIGIT ligand CD155 was highly expressed on BMSCs, we speculated that its immunosuppressive functions in NK cells occur via the interaction with CD155 on BMSCs. Next, we revealed the mechanism through which BMSCs regulate NK cells via CD155/TIGIT using in vitro co-culture (Figure 3A). After NK cells were amplified with IL-2, expression of the activated markers NKG2D, NKp44, CD69 and NKp30 on NK cells were increased, but these levels were decreased after NK-cell co-culture with BMSCs for 6 days (Figure S3B). In co-culture systems, TIGIT mAb and TIGIT+CD226 mAb groups showed restored expression of these markers on NK cells and IFN-γ secretion was increased in NK cells (Figure 3B left, C). To verify the changes in NK-cell function, we added the U266 cell line to co-culture systems; apoptosis of U266 gradually increased after adding TIGIT mAbs (Figure 3B right). We further detected related protein changes in the CD155/TIGIT signalling pathway; here, SHIP-1 expression decreased and Erk phosphorylation increased with TIGIT blockade (Figure 3D). As TIGIT binds to CD155 with higher affinity and out-competes CD226 for CD155 binding, we purified the agonist CD226 mAb (WO 2020/023312 A1) to activate CD226 while suppressing TIGIT (Figure S3E). A blocking TIGIT mAb and agonist CD226 mAb were added to BMSC and NK-cell co-culture systems, and we found that the activation of NK cells increased with TIGIT combined with the agonist CD226 mAb and IFN-γ secretion, and U266 cell apoptosis was elevated (Figure S3C,D,F). In summary, we demonstrated that TIGIT was upregulated, but that CD226 was downregulated, on NK cells from NDMM patients. Regarding the TIGIT ligand, CD155, but not CD112 and CD113, was highly expressed on BMSCs but was expressed at low levels on myeloma cells. In vitro co-culture results demonstrated that CD115/TIGIT signalling plays a vital role SI in the interaction between BMSCs and NK cells. Blocking TIGIT could greatly restore NK-cell exhaustion, providing a potential avenue for antitumour immunotherapy for MM (Figure 4). This work was supported by the National Natural Science Foundation of China Youth Project (Grant no. 81900131), the Tianjin Municipal Natural Science Foundation (Grant no. 18JCQNJC80400), the Tianjin Education Commission Research Project (Grant no. 2018KJ043), the Tianjin Education Commission Research Project (Grant no. 2018KJ045) and the Tianjin Science and Technology Planning Project (Grant no. 20YFZCSY00060). Tianjin Municipal Health Commission Youth Project (Grant no. TJWJ2021QN001); Medjaden Academy & Research Foundation for Young Scientists (Grant no. MJR20221011); Tianjin Key Medical Discipline (Specialty) Construction Project. We thank the Biomedical Big Data Platform at Shanghai Institute for Advanced Immunochemical Studies (SIAIS), ShanghaiTech University for the analyses of single-cell RNA-data. All authors declare no conflict of interest. Figure S1 Natural killer (NK)-cell individual markers in the immune landscape of multiple myeloma (MM). Figure S2 Identification of bone marrow mesenchymal stem cells (BMSCs) and the ligand expression of CD155, CD112, and CD113 are extremely low on myeloma cells. Figure S3 Blockade of TIGIT and activating CD266 might better restore natural killer (NK)-cell function. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.