Zhiqiang Qin

Antibodies Against Nucleolin in Recipients of Organ Transplants

Background. Patients who reject allografts frequently make strong antibody responses against donor human leukocyte antigens and autoantigens such as vimentin, collagen V, or alpha-tubulin and it has been postulated that autoantibodies may play a role in allograft failure. Methods. We have used serum from patients who recently rejected an allograft as a source of antibodies in combination with lysates of human umbilical vein endothelial cells as a source of target antigens. Immunoprecipitation and protein identification was performed by mass spectrometry. Recombinant nucleolin was produced and sera were assayed for antibodies by enzyme-linked immunosorbent assay. Results. Immunoprecipitation with serum WW led to the recognition of the protein nucleolin as a target antigen. By enzyme-linked immunosorbent assay, with recombinant nucleolin (r-nucleolin), the frequency of antibodies to nucleolin were found to be 2.0% in normal subjects, 9.1% in patients waiting for a kidney transplant, 25.5% after irreversible rejection of a kidney allograft, 17.1% after a heart transplant, and 43.8% in heart transplant recipients developing transplant-related coronary artery disease. Antibodies against nucleolin from mice or from transplant patients inhibited endothelial cell proliferation and in vitro capillary-like tube formation and caused apoptosis of human umbilical vein endothelial cells. Conclusions. Antibodies against nucleolin seem to inhibit and produce apoptosis of proliferating endothelial cells. These antibodies were found in many transplant patients and seemed to be associated with rejection of kidney allografts and with coronary artery disease in heart transplant recipients.

Polymorphisms of MICA recognized by human alloantibodies

MICA antigens are polymorphic glycoproteins expressed on the surface of human endothelial cells and other cells. Antibodies against MICA have been found in transplant recipients and were found to be associated with decreased survival of kidney allografts. In the present work, we investigated the polymorphisms that are recognized by antibodies against MICA. Soluble MICA recombinant proteins representing 11 common alleles, two hybrid alleles, and two single amino acid mutated alleles were produced. Patterns of reactivity were determined with MICA bound to Luminex beads. In some studies, sera containing antibodies against MICA were absorbed by cell lines transfected with MICA*001, MICA*002, MICA*008, and MICA*009 or with untransfected cells, followed by testing of antibody reactivity against MICA proteins bound to beads. The monoclonal antibodies and sera used in this study were found to recognize up to 14 distinct MICA epitopes as demonstrated by their differential absorption/reactivity patterns. Among these, nine epitopes correlated with a single unique amino acid: one shared two signature amino acids, one shared three signature amino acids in close proximity, and three epitopes involved multiple amino acids in a nonlinear sequence. Two groups of public epitopes (MICA-G1 and MICA-G2) were characterized. MICA shared epitopes were determined by reactivity loss in single MICA antigen bead assays by absorption with MICA transfectants. Since these epitopes may be targets for antibody binding and possibly antibody-mediated allograft rejection, epitope identification may help understand the development of MICA antibodies and to identify suitable donors for sensitized transplant recipients.

The emerging issue of MICA antibodies: antibodies to MICA and other antigens of endothelial cells.

The major histocompatibility complex (MHC) encodes the HLA class I antigens expressed on the surface of most nucleated cells and the HLA class II antigens which are expressed mostly in B lymphocytes, monocytes and dendritic cells. Mismatched HLA antigens are the main source of the immune response that leads to the rejection of allografts. In some patients however, rejection may occur without a detectable response to donor HLA antigens. We have been interested in characterizing antibodies that develop in transplant recipients who do not appear to have antibodies against HLA. For this purpose, we focused our attention to antigens which are expressed on the surface of endothelial cells and are not found on peripheral blood lymphocytes. These include the MICA and MICB antigens, which are encoded by loci in the MHC; certain autoantigens expressed on the endothelium; and a family of polymorphic antigens expressed on endothelial cells which are distinct from HLA and elicit production of antibodies that appear also to be associated with graft failure. Antibodies against MICA have been associated with allograft rejection. MICB antibodies are only rarely found. The autoantibodies and the endothelial specific alloantibodies are being characterized in ongoing studies.

Identification of endothelial cell surface antigens encoded by genes other than HLA. A combined immunoprecipitation and proteomic approach for the identification of antigens recognized by antibodies against endothelial cells in transplant recipients

It has been known for some time that transplant recipients may have antibodies to endothelial cells which are not detected on lymphocytes. However, little progress has been made in the analysis of these endothelial antigens. In the present experiments we have attempted to characterize endothelial cell surface antigens to which antibodies were produced during graft rejection. We have used a panel of endothelial cells from umbilical cord veins and found that antibodies with a polymorphic pattern in the panel appeared to correlate with transplant failure of kidney allografts and with the development of transplant-related coronary artery disease (TCAD) in heart transplant recipients. Among 39 patients with kidney allografts, 21 were negative for antibodies to endothelial cells and did well and 18 were positive and had frequent transplant loss (p=0.001). In 18 patients with TCAD and 20 patients of a comparator group without TCAD, association of coronary disease with endothelial cell antibodies was observed (p<0.02). To characterize the endothelial antigens responsible for these serologic reactions we performed immunoprecipitation of reactive antibodies with the corresponding endothelial cell surface antigens, followed by protein identification of the target antigens. Nine proteins were identified in these experiments, 5 were non-polymorphic and appeared to represent autoantigens. Four of the isolated proteins appeared to be polymorphic. They were the Human Major Histocompatibility Complex class I chain-related gene A (MICA), already known to be associated with antibody production and graft failure, human keratin 1, a protein known to be polymorphic and expressed on the surface of endothelial cells, eukaryotic translation initiation factor (EIF) 2A and ErbB3-binding protein 1. The possible role of keratin 1 and the other antigens in allograft rejection requires further investigation.

Alleles of keratin 1 in families and populations

Keratin 1 is found in the upper layers of the epidermis, on the surface of endothelial cells and in the membrane of the neuroblastoma NMB7. It is important for the structural integrity of the skin, has been found to regulate the activity of kinases, such as protein kinase C (PKC) and SRC, to participate in complement activation by the lectin pathway and to be involved in fibrinolysis, angiogenesis and the response to oxidative stress. Studies of the polymorphisms of the Keratin 1 (KRT1) gene have been driven mostly by interest in its role in skin diseases. However, much of the KRT1 variation occurs in normal populations and is not associated with dermal pathology. In the present experiments, we have investigated the polymorphism of KRT1 genes by nucleotide sequencing in normal families and normal populations of European, African, Hispanic and Asian background. The frequencies of the KRT1 alleles were strikingly different in the four ethnic groups and most of the mutations resulted in amino acid substitutions, with only 3 out of 19 being synonymous. Analysis of selective neutrality by the Ewens-Watterson and Tajima D statistics showed that KRT1 allele homozygosity was decreased in three of the populations suggesting that KRT1 genes may be under the influence of balancing selection. It is possible that the role of KRT1 as a receptor, rather than its structural function in the epidermis, is what drives the selective forces that are apparent in the inheritance of this gene.