Purified HLA antigens to probe human alloantibody specificity

Susceptibility of liver allografts to high or low concentrations of preformed antibodies as measured by flow cytometry

Liver grafts are more resistant to damage by HLA antibodies than other organ allografts, but it is not clear if the antibodies are associated with graft rejection or graft loss, or if different antibody concentrations have different effects. To explore potential associations between antibody concentrations and outcome, preformed IgG antibodies against donor cells were quantified by flow cytometry in 465 consecutive liver transplant recipients. Antibody-positive patients were classified according to whether they had high or low antibody concentrations and analyzed for possible correlation with graft rejection or graft loss. The results showed that the incidence of rejection was not significantly different between antibody-positive and negative patients. However, patients with high antibody concentrations had a higher incidence of steroid-resistant rejections (31% at 1 year) than patients with low antibody (4%) or no antibody (8%, p < 0.0004). These effects were mainly due to T-cell (HLA class 1) antibodies. The overall incidence of rejection at 1 year was 69% for high antibody patients, 51% for patients with low antibodies and 53% for patients with no antibodies (p not significant). In an apparent paradox, antibody-positive patients underwent fewer early graft losses. Thus, the associations of preformed antibodies and outcome depend, on the one hand, on antibody concentrations, and on the other hand on whether the outcome measured is steroid-sensitive rejection, steroid-resistant rejection or graft survival. These complex interactions may explain the controversial results observed in previous studies.

Detection of alloantibodies by flow cytometry: relevance to clinical transplantation

Before an organ transplant is performed, donor-recipient compatibility must be established by a crossmatch in much the same way as it is done for blood transfusions. The target antigens in organ transplantation, however, are HLA rather than blood group molecules, and the target cells are lymphocytes instead of red cells. If antidonor antibodies are detected, it is important to know whether they are IgG or IgM, whether they recognize T and/or B cells, and whether the antibody reactivity is weak or strong. These test requirements are better met by flow cytometry than by the standard cytotoxicity technique. A growing body of evidence now indicates that flow cytometry can provide more sensitive and timely crossmatch information than cytotoxicity assays to decide whether or not a transplant should be done. Flow cytometry crossmatch (FCXM) is a new and evolving technique that has already been found to be extremely useful in the clinical transplantation setting, even though significant questions yet remain about the precision and reliability of using flow cytometry to quantify alloantibodies and about the limits of normal reactivity in the assay. This article reviews important technical details of the FCXM and its interpretation and clinical application in transplantation medicine.

Quantitation of soluble HLA class I antigen in human albumin and immunoglobulin preparations for intravenous use by solid-phase immunoassay

Soluble HLA class I antigens in human plasma preparations possibly play a role in HLA sensitization and modulation of the immune response. We therefore have determined their concentration in albumin and immunoglobulin preparations from several commercial sources and compared these values to the concentration in normal human sera. For this purpose we used a newly developed solid-phase enzyme immunoassay employing rabbit anti-mouse antibody, monomorphic HLA class I monoclonal antibody and a polyclonal enzyme-linked beta 2-microglobulin-specific antiserum. Soluble HLA antigen concentration in 14 albumin batches from 6 manufacturers and in 16 immunoglobulin batches from 11 manufacturers ranged from 0 to 9.6 and from 0 to 20.9 ng/ml. The concentration in normal human sera 1,328 +/- 954 ng/ml (n = 54). We conclude that the concentration of soluble HLA concentration in albumin and immunoglobulin preparations is more than 50 times lower than in normal human serum, but considerable differences exist between products of various manufacturers.

Binding of IgG to B cell via HLA molecules

Binding of immunoglobulins to major histocompatibility complex (MHC) molecules was demonstrated by two different assays: the binding of IgG to B cells by flow cytometry, and purified MHC antigens with an Elisa assay. Fc fragment from immune-complex binds to the Fc receptor on B lymphocytes. Here, Fab was also shown to bind to B cells. This binding was inhibited by specific human allo anti-HLA Class I and II sera directed at the polymorphic sites. Thus, in addition to the Fc receptor, MHC can also serve as a binding site for IgG. In an Elisa assay using purified antigens, IgG was shown to bind to HLA Class I and II molecules. Other proteins such as transferrin, human serum albumin, gelatin, etc., did not bind to the MHC proteins. Immunoglobulins bound to MHC molecules by sites on the Fab fragment independent of the hypervariable region. This was demonstrated by the retention of antibody activity even after binding of antibody (anti-lactoferrin) to MHC. The relative avidity between Fab and HLA Class I and II was 4-8 x 10(5) M-1.

An enzyme-linked immunosorbent assay for the detection of platelet antibodies using detergent-solubilized platelets immobilized on nitrocellulose discs

A method is detailed for the solubilization of human platelets using a dialyzable detergent, decanoyl-N-methylglucamide (Mega-10). At a detergent/protein ratio of 1:12, the efficiency of solubilization was 27%. This platelet lysate (PLy) was then bound to nitrocellulose (NC) discs to assay retention of the native immunological functions of the platelet membrane antigens. Using alkaline phosphatase-coupled anti-IgG, the major platelet membrane glycoproteins GPIb, GPIIb, and GPIIb/IIIa were detectable with as little as 20 ng of monoclonal antibody. Antisera to the class I histocompatibility antigens HLA-A1, B7, B8, the PlA1 allodeterminant, and serum from multiply transfused, alloimmunized patients were reactive even after 100 days storage of the discs at 4 degrees C, and with as little as 1.0 micrograms of NC-bound PLy. The binding of the same antisera to intact, immobilized platelets as well as specific complement-mediated lymphocytotoxicity was also inhibited by PLy. PLy from HLA-A3- or B44-positive donors, however, did not inhibit cytotoxicity of lymphocytes expressing either antigen using several different antisera. Our results indicate that Mega-10 is an excellent solubilizing agent for the immunological study of platelet membranes. The fact that clinically relevant platelet membrane antigens are preserved, immunologically reactive, and stable over long periods of storage, makes this assay amenable to a routine crossmatching procedure for platelet transfusions.

HLA-A2 epitopes recognized by alloantibodies from broadly sensitized patients

Alloantibodies in broadly sensitized patients are difficult to characterize because they comprise mixtures with different specificities. By studying the ability of patients' sera to inhibit or enhance the binding of two anti-HLA-A2 monoclonal antibodies, BB7.2 and MA2.1, reactivity with A2 epitopes could be operationally defined. One group of patients had antibodies that inhibited the binding of both monoclonal antibodies. The inhibitory activity was removed by absorption with A2 positive cells but not with A28 or B17 positive cells. These alloantibodies apparently recognized the A2-Bw69 epitope also defined by the BB7.2 monoclonal antibody. In other patients, the inhibitory activity was removed by A2 and B17 positive cells, suggesting the involvement of the same epitope recognized by the MA2.1 monoclonal antibody. A third type of reactivity was detected in other patients. These alloantibodies enhanced the binding of BB7.2, inhibited MA2.1, and were absorbed by A2 and B17 positive cells indicating that the A2-B17 epitope was involved in these cases too. These studies underscore the potential value of monoclonal antibodies in dissecting the specificities of complex alloantibody mixtures.