Chimeric anti-HLA antibody receptor engineered human regulatory T cells suppress alloantigen-specific B cells from pre-sensitized transplant recipients

Abstract

Organ transplantation is a lifesaving procedure, with 50,000 transplants happening every year in the United States. However, many patients harbor antibodies and B cells directed against allogeneic human leukocyte antigen (HLA) molecules, notably HLA-A2, greatly decreasing their likelihood of receiving a compatible organ. Moreover, antibody-mediated rejection is a significant contributor to chronic transplant rejection. Current strategies to desensitize patients non-specifically target circulating antibodies and B cells, resulting in poor efficacy and complications. Regulatory T cells (Tregs) are immune cells dedicated to suppressing specific immune responses by interacting with both innate and adaptive immune cells. Here, we genetically modified human Tregs with a chimeric anti-HLA antibody receptor (CHAR) consisting of an extracellular HLA-A2 protein fused to a CD28-CD3zeta intracellular signaling domain, driving Treg activation upon recognition of anti-HLA-A2 antibodies on the surface of alloreactive B cells. We find that HLA-A2 CHAR Tregs get activated specifically by anti-HLA-A2 antibody-producing cells. Of note, HLA-A2 CHAR activation does not negatively affect Treg stability, as measured by expression of the Treg lineage transcription factors FOXP3 and HELIOS. Interestingly, HLA-A2 CHAR Tregs are not cytotoxic towards anti-HLA-A2 antibody-producing cells, unlike HLA-A2 CHAR modified conventional CD4⁺ T cells. Importantly, HLA-A2 CHAR Tregs recognize and significantly suppress high affinity IgG antibody production by B cells from HLA-A2 sensitized patients. Altogether, our results provide proof-of-concept of a new strategy to specifically inhibit alloreactive B cells to desensitize transplant recipients.

Keywords: B cells; HLA sensitization; antibody production; engineered immune receptors; human immunology; regulatory T cells; transplant rejection; transplantation.

Figure 1

HLA-A2 CHAR Tregs are activated specifically by anti-HLA-A2 antibody producing cells. (A) Schematic representation of chimeric anti-human leukocyte antigen (HLA) antibody receptor (CHAR) featuring a CD28-CD3zeta signaling domain, expressed on the surface of a human regulatory T cell (Treg), binding an anti-HLA-A2 antibody on the surface of an allogeneic B cell from an HLA-A2 pre-sensitized patient. The B cell receptor (BCR) on the surface of the B cell comprises a surface-bound antibody and the signaling heterodimer CD79A and CD79B. Upon engagement, the HLA-A2 CHAR Treg suppresses anti-HLA-A2 expressing B cell function. (B) Cell surface expression of HLA-A2 CHAR construct in lentivirus transduced Tregs, as assessed by co-expression of HLA-A2 and a reporter gene, truncated nerve growth factor receptor (NGFRt), linked to the CHAR gene by a 2A peptide. UT, untransduced. (C) CHAR Treg activation upon 48-hour co-incubation with irradiated PA2.1 (anti-HLA-A2), but not alone or with IVA12 (anti-HLA-DR, DP, DQ) hybridoma cells, as assessed by surface expression of CD69. (D) Frequency of CD69-expressing cells among UT or CHAR Tregs alone or co-incubated with PA2.1 or IVA12 cells for 48h. (E) Expression levels (MFI, median fluorescence intensity) of CD71 on UT or CHAR Tregs alone or co-incubated with PA2.1 or IVA12 cells for 48h. (F) Expression levels (MFI) of CD25 on UT or CHAR Tregs alone or co-incubated with PA2.1 or IVA12 cells for 48h. (G) Enrichment of CHAR-expressing Tregs upon 9-day co-incubation with irradiated PA2.1 cells, but not alone or with IVA12 cells, as assessed by surface expression of NGFRt reporter. (H) Frequency of NGFRt-expressing Tregs (CHAR Tregs) alone or co-incubated with PA2.1 or IVA12 cells for 9 days. For (D–F, H), bars represent mean and standard deviation (n = 3 technical replicates, one blood donor representative of two). Data were analyzed by one-way ANOVA with multiple comparisons. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 2

HLA-A2 CHAR Tregs remain stable and are not cytotoxic upon activation. (A) Representative flow cytometry analysis of CHAR Treg FOXP3 and HELIOS expression after 9 days of co-culture with irradiated PA2.1 (anti-HLA-A2) and IVA12 (anti-HLA-DR, DP, DQ) hybridoma cells. Untransduced (UT) T effector (Teff) cells and CHAR Teff cells were used as negative controls. (B) Frequency of FOXP3⁺ CHAR Tregs alone or co-cultured with PA2.1 and IVA12 hybridoma cells for 9 days. (C) FOXP3 expression (mean fluorescence intensity - MFI) in CHAR Tregs alone or co-cultured with PA2.1 and IVA12 hybridoma cells for 9 days. (D) Frequency of FOXP3+HELIOS+ CHAR Tregs alone or co-cultured with irradiated PA2.1 and IVA12 hybridoma cells for 9 days. (E) Cytotoxicity of CHAR Tregs and CHAR Teff cells towards PA2.1, as measured by lactate dehydrogenase (LDH) release after 24-hour co-incubation at a 1:1 ratio. Bars in (B–E) represent mean and standard deviation (n = 3 technical replicates, one blood donor representative of two). Data were analyzed by one-way ANOVA with multiple comparisons. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 3

HLA-A2 CHAR Tregs suppress IgG specific production on highly pre sensitized patients. (A) HLA-A2 sensitized patient (SEN) and healthy control (HD) demographics and clinical characteristics. (B) Experimental design for assessing CHAR Treg function in the presence of HLA-A2 pre-sensitized patient cells. HLA-A2 sensitized donor-derived peripheral blood mononuclear cells (PBMCs) were co-incubated with HLA-A2-expressing K562 cells to induce expansion of anti-HLA-A2 B cells and anti-HLA-A2 IgG antibody production. If CHAR Tregs are added, a decrease in antibody production elicited by exposure to HLA-A2 is expected. (C) IgG antibody production 48 hours or 5 days after pre-sensitized patient PBMC co-incubation with HLA-A2-K562 in the presence or absence of CHAR T regs, as assessed by ELISA. n=3 sensitized patients (SEN) and n=1 healthy donor (HD) control). Data were analyzed by one-way ANOVA with multiple comparisons. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 4

HLA-A2 CHAR Tregs were activated and reduced the frequency of total B cells in pre-sensitized patient peripheral blood samples. (A) Flow cytometry gating strategy to identify HLA-A2⁺NGFR⁺CD19^- CHAR Tregs, CD19⁺CD20⁺ B cells, and B cell subsets in co-cultures of sensitized patient PBMCs with irradiated HLA-A2-K562 and CHAR Tregs. (B, D) Uniform manifold approximation and projection (UMAP) representation of sensitized patient PBMCs with irradiated HLA-A2-K562 (iA2K562) and CHAR Tregs depicting naive, marginal zone, memory and IgD^-CD27^- B cells, as well as plasmablasts, after 48 hours (B) and 5 days (D) of co-culture. (C, E) Frequency of CD27 expression and CD38 expression, as well as CD71 surface expression levels, in CHAR Tregs co-incubated with sensitized patient PBMCs for 48 hours (C) and 5 days (E). n=3 sensitized patients (SEN) and n=1 healthy donor (HD) control. Data were analyzed by one-way ANOVA with multiple comparisons. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.