Gates, Katherine V.

American Heart Association estimates the prevalence of valvular heart disease to be 2.5% in the United States. Biological heart valve replacements, such as glutaraldehyde‐fixed bovine pericardium (BP), are considered the current standard of care. The NHLBI xenotransplantation working group identified xenoantigenicity as the major barrier to expanding use of such xenogeneic tissues. Previous research has utilized an immunoproteomic approach based on two‐dimensional western blots to identify a small number of BP xenoantigens, which is time consuming and technically challenging. These limitations have resulted in an ever increasing use of shotgun proteomic approaches due to their superior throughput and reproducibility. However, traditional Liquid Chromatography‐Mass Spectrometry (LC‐MS) fails to separate proteins based on their antigenicity, and therefore it is impossible to determine which identified proteins are antigenic versus not. We hypothesized that an affinity chromatography method utilizing poly‐polyclonal serum would bridge this gap and represent an innovative shotgun proteomic antigen identification approach. Using affinity chromatography rabbit IgG antibodies were isolated from whole serum with Protein A HP SpinTrap (GE Healthcare). IgG capture was confirmed by serum protein electrophoresis and ELISA. Two types of antibody affinity columns were generated, with naive serum (D0) or serum from rabbits immunized with bovine pericardium (D84). Bovine pericardial protein extracts were run through both affinity columns, allowing for the specific capture of antigens, which were later eluted and collected. To confirm specificity of antigen capture, eluants were subjected to western blot, and probed with D84 serum. Affinity enriched antigenic proteins were identified by LC‐MS/MS, with statistical comparisons made between D0 and D84. This approach identified 133 putative antigenic proteins that were enriched on D84 compared to D0 columns. Ten of these antigens corresponded to previously published BP xenoantigens. Of the remaining 123 previously unidentified antigens, 30 have been indicated to be immunogenic in other fields such as cancer and autoimmunity. This method was then applied to three humans who had been implanted with bovine pericardial heart valves. Surprisingly, 300 to 500 proteins were identified from patient columns. Between 57 to 79% of the identified proteins were shared among the three individuals, as well as almost all of the proteins identified in the rabbit model were also present in the human. This work therefore represents the first high throughput method of antigen identification that can now be used to monitor currently implanted patients as well as inform the development of BP‐based biomaterials in the future. Support or Funding Information Research supported by grant number R01HL115205 from the National Heart Lung and Blood Institute (NHLBI) at the National Institutes of Health (NIH) and grant number 5T32OD010931‐10 from the NIH.

Abstract Valve replacement is an important therapeutic approach for management of valvular heart disease, accounting for over 200,000 surgeries annually worldwide. Deficiencies of current heart valve prostheses led the NHLBI cardiac surgery working group to highlight the need for improved heart valve leaflet biomaterials. Recently, our group has engineered a novel method to eliminate antigens from bovine pericardium (BP) extracellular matrix (ECM) scaffolds, while retaining biomaterial structure-function properties and regenerative capacity. However, despite reducing BP-ECM scaffold antigen burden by >92%, graft-specific humoral IgG response persisted at ~30% of the level associated with native BP in a rabbit model. In this study, we sought to determine the identities of residual antigens present in antigen removed BP-ECM scaffolds. New Zealand White Rabbits were subpannicularly implanted with either native, glutaraldehyde fixed, or antigen-removed BP-ECM scaffolds (n = 6 per group). Rabbit serum, collected pre- and at 56 d post-implantation, was used to generate IgG affinity chromatography columns. Native BP protein extracts were loaded onto columns, and captured antigens were identified by LC-MS/MS. A total of 259 antigens were identified in native BP rabbits. The glutaraldehyde fixed and antigen-removed rabbits responded to ~4% and ~12% of these same antigens, respectively. A small subset of antigens were uniquely identified in each of the treatment groups, not identified in native BP. These results inform ongoing efforts to eliminate antigens from xenogeneic tissues in development of ECM scaffolds, and may provide insight into the relative immunogenicity of individual xenoantigens in the context of tissue engineering.

Abstract Although immunosuppressive drugs improve short term survival, rejection incidence and long term survival remain largely unchanged in heart transplant patients. While Human Leukocyte Antigen (HLA) is a common initiator of graft-specific immune responses, non-HLA antibody mediated rejection (AMR) also commonly occurs. Furthermore, non-HLA antibodies have been implicated in cell mediated rejection and cardiac allograft vasculopathy. Therefore, there is a need to identify the role of non-HLA antigens in heart transplant rejection. Serum from heart transplant patients lacking donor specific anti-HLA antibodies was collected longitudinally; pre-transplantation, acute rejection, and post-rejection for patients with AMR and matched controls. Serum IgG was used to form anti-human heart affinity columns and antigenic proteins from donor heart extracts captured. Antigens were eluted from the columns and subjected to LC-MS/MS. Longitudinal graft-specific changes in immune response were calculated. Response toward non-HLA antigens increased longitudinally, with 41 high prevalence antigens found in ≥50% of the cohort. Co-culturing cardiomyocyte and endothelial cells with patient IgG indicated cell function changes correlated with specific antigens (e.g., Prohibitin). Immunofluorescence staining of patient biopsies revealed that longitudinal increase in antigen expression correlated with non-HLA antibody response. The non-HLA antigens identified in this study elicit acute and chronic graft-specific adaptive immune responses in heart transplant recipients. Antibodies against these non-HLA antigens induced cellular dysfunction by changing expression of their target protein in the donor heart.

Background. The International Society for Heart and Lung Transplant consensus panel notes that too little data exist regarding the role of non-HLA in allograft rejection. We developed a novel shotgun immunoproteomic approach to determine the identities and potential roles non-HLA play in antibody-mediated rejection (AMR) in heart transplant recipients. Methods. Serum was collected longitudinally from heart transplant recipients experiencing AMR in the absence of donor-specific anti-HLA antibodies (n = 6) and matched no rejection controls (n = 7). Antidonor heart affinity chromatography columns were formed by recipient immunoglobulin G immobilization at transplantation, acute rejection, and chronic postrejection time points. Affinity chromatography columns were used to capture antigens from individual patient’s donor heart biopsies collected at transplantation. Captured proteins were subjected to quantitative proteomic analysis and the longitudinal response was calculated. Results. Overlap in antigen-specific response between AMR and non-AMR patients was only 8.3%. In AMR patients, a total of 155 non-HLAs were identified, with responses toward 43 high prevalence antigens found in ≥50% of patients. Immunofluorescence staining for representative high prevalence antigens demonstrated that their abundance increased at acute rejection, correlating with their respective non-HLA antibody response. Physiological changes in cardiomyocyte and endothelial cell function, following in vitro culture with patient immunoglobulin G, correlated with response toward several high prevalence antigens. Conclusions. This work demonstrates a novel high-throughput strategy to identify clinically relevant non-HLA from donor endomyocardial biopsy. Such a technique has the potential to improve understanding of longitudinal timing of antigen-specific responses and their cause and effect relationship in graft rejection.

PurposeThis case‐control retrospective discovery study is to identify antigenic bovine pericardium (BP) proteins that stimulate graft‐specific humoral immune response in patients implanted with glutaraldehyde fixed bovine pericardial (GFBP) heart valves.Experimental designBanked serum is collected from age‐ and sex‐matched patients who received either a GFBP or mechanical heart valve replacement. Serum IgG is isolated and used to generate poly‐polyclonal antibody affinity chromatography columns from each patient. Native and deglycosylated BP protein extracts are separately added to individual patient affinity chromatography columns, with unbound proteins washed through the column. Proteins captured in the affinity chromatography columns are submitted for proteomic identification. Differences between GFBP and mechanical heart valve replacement recipients are analyzed with Gaussian linearized modeling.ResultsCarbohydrate antigens overwhelm protein capture in the column, requiring BP protein deglycosylation prior to affinity chromatography. Nineteen BP protein antigens, which stimulated graft‐specific IgG production, are identified in patients who received GFBP valve replacements. Identified antigens are significantly over‐represented for calcium‐binding proteins.Conclusions and clinical relevancePatients implanted with GFBP valves develop a graft‐specific humoral immune response toward BP protein antigens, with 19 specific antigens identified in this work. The molecular functions of over‐represented antigens, specifically calcium‐binding proteins, may aid in understanding the underlying factors that contribute to structural valve deterioration.