Donor-specific B Cell Memory in Alloimmunized Kidney Transplant Recipients: First Clinical Application of a Novel Method.

BACKGROUND: HLA-specific memory B cells may contribute to the serum HLA antibody pool upon antigen reexposure. The aim of this pilot study was to investigate the presence of concurrent donor-specific memory B cell-derived HLA antibodies (DSA-M) in renal allograft recipients with pretransplant donor-specific HLA antibodies (DSA) and its association with occurrence of antibody-mediated rejection (AMR) using a recently developed method. METHODS: Twenty patients with Luminex single antigen bead (SAB) assay-defined DSA but negative complement-dependent cytotoxicity crossmatches were enrolled. Plasma samples and peripheral blood mononuclear cells were collected at 3 timepoints (pretransplant, mo 6, mo 12). We analyzed IgG-purified and concentrated culture supernatants from polyclonally activated peripheral blood mononuclear cells using SAB assays and compared HLA antibody profiles with same day plasma results. RESULTS: Plasma SAB analysis revealed 35 DSA in 20 patients pretransplant. DSA-M were detected in 9 of 20 (45%) patients and for 10 of 35 specificities (29%). While median mean fluorescence intensity values of DSA with concurrent DSA-M (5877) were higher than those of DSA without DSA-M (1476), 3 of 6 patients with AMR and low mean fluorescence intensity DSA (<3000) had DSA-M. Overall, pretransplant DSA/DSA-Mpos allograft recipients showed a higher incidence of biopsy-proven (sub)clinical AMR (P = 0.032) and a higher extent (g≥1 + ptc≥1) of microvascular inflammation (67% vs 9%, P = 0.02). In 17 patients (28 DSA) with posttransplant analyses, persisting DSA posttransplant had more often DSA-M (6/12; 50%) than nonpersisting DSA (2/16; 13%). CONCLUSIONS: Assessment of DSA-M might be a novel tool to supplement serum HLA antibody analysis for pretransplant risk stratification in patients with DSA.

An Easy and Sensitive Method to Profile the Antibody Specificities of HLA-specific Memory B Cells.

BACKGROUND: Pretransplant immunological risk assessment is currently based on donor-specific HLA antibodies in serum. Despite being an excellent source for antibodies produced by bone marrow-residing plasma cells, serum analysis does not provide information on the memory B-cell compartment. Although B-cell culture supernatants can be used to detect memory B cell-derived HLA antibodies, low IgG concentrations can preclude detectability of HLA antibodies in luminex single-antigen bead (SAB) assays. METHODS: Culture supernatants of polyclonally activated B cells from alloantigen exposed (n = 13) or nonexposed (n = 10) individuals were either concentrated 10-fold, or IgG was isolated by using a protein G affinity purification method to increase the IgG concentration. These processed culture supernatants, as well as paired serum samples were tested for the presence of HLA antibodies using luminex SAB analysis. RESULTS: In immunized individuals, 64% were found to have HLA-specific B-cell memory in concentrated supernatants, whereas 82% showed HLA-specific B-cell memory when IgG isolated supernatants were used for HLA antibody detection. IgG-isolated supernatants showed higher mean fluorescence intensity values compared with concentrated supernatants without increased background. In some individuals, HLA-specific B-cell memory was detected in the absence of accompanying serum antibody specificities. CONCLUSIONS: We developed a novel, highly sensitive method to assess the HLA-specific memory B-cell compartment using luminex SAB technology. This assay allows direct comparison to the serum compartment and may therefore provide a more complete picture of the humoral alloimmune response in patients with a history of alloantigen exposure.

No Evidence for Cross-reactivity of Virus-specific Antibodies With HLA Alloantigens.

BACKGROUND: Antibodies directed against HLA can develop through pregnancy, blood transfusions, or organ transplants. Anecdotal evidence suggests that virus-specific antibodies may have the capacity to cross-react with HLA, a phenomenon called heterologous immunity, which is well described for T-cell alloreactivity. METHODS: To determine whether antibody cross-reactivity between viral antigens and HLA is common, we tested 51 virus-specific human monoclonal antibodies (mAbs) specific for human immunodeficiency virus, varicella zoster virus, cytomegalovirus, and parvovirus, for reactivity against HLA class I and class II in single-antigen bead assays. In addition, we tested the reactivity of 41 HLA-specific human mAbs against common viral antigens of cytomegalovirus, varicella zoster virus, human immunodeficiency virus, Epstein-Barr virus, and BK polyomavirus. RESULTS: No cross-reactivity of any of the virus-specific mAbs with either HLA class I or class II molecules, as well as no cross-reactivity of any of the HLA-specific mAbs with any of the viral antigens was observed. CONCLUSIONS: These findings indicate that the frequency of cross-reactivity on the antibody level between viral antigens and HLA, if present at all, is low. The emergence of HLA antibodies upon viral infection or vaccination is therefore probably due to bystander activation of dormant HLA-specific memory B cells.

Kinetics of the pregnancy-induced humoral and cellular immune response against the paternal HLA class I antigens of the child.

Pregnancy can prime the maternal humoral immune response against paternal human leukocyte antigens (HLA) of the child. Previous studies have reported that formation of antibodies against inherited paternal HLA is associated with the presence of primed cytotoxic T lymphocytes (CTLs) specific for these antigens. Recently, we reported that primed CTLs can persist for more than 10 years after pregnancy even if the antibodies have disappeared. In the present study we studied the kinetics of the pregnancy induced immune response of the T-cell and B-cell compartment. In 12 women, who had specific antibodies against the paternal HLA antigens of the child (child mismatch) at the time of delivery, we analyzed the CTLp frequencies against the paternal HLA antigens from the time of delivery up to 2 years after. The contribution of primed CTLs to these CTLp frequencies was tested by limiting dilution analysis in the absence and presence of monoclonal antibodies specific for CD8. In contrast to naïve CTLs, primed CTLs are resistant to CD8 antibodies. Disappearance of the antibodies was not associated with a decrease of the number of CTLp directed against the paternal antigens, towards which the antibodies were originally directed. However, in women where the antibodies disappeared, a decrease of primed child mismatch specific CTLs was found, whereas in women where the antibodies persist, the population of primed CTLs remained stable up to 2 years after delivery. Our data suggest a functional correlation between the T-cell and B-cell allorepertoire. Although the kinetics do not run completely in parallel, disappearance of the anti-HLA antibodies in the first 2 years after delivery is related with a decrease of primed child mismatch specific CTLs. These data may be relevant for transplantation of female recipients with historical, pregnancy-induced HLA alloantibodies.