Specific removal of anti-A and anti-B antibodies by using modified dialysis filters.

Removal of anti-A and anti-B blood group antibodies from human blood has been shown to facilitate cross-matched kidney transplantation by preventing hyperacute rejection. Patients in these studies had anti-A and anti-B antibodies removed by using plasmapheresis, followed by immunoadsorption onto packed bead columns. We conducted a study to assess the feasibility of selectively removing anti-A and anti-B antibodies directly from blood by using modified dialysis filters. An anti-A and anti-B specific antigen was covalently attached to the lumenal surfaces of hollow fibers within selected commercial dialysis modules. The filters were able to reduce the anti-A and anti-B titers of 300 ml of blood to 2 or below. A low molecular weight fraction of our antigen system was found to have no antibody binding capacity. The standard antigen was purified by removal of the low molecular weight fraction and a dialysis filter was modified by using the purified antigen. This filter displayed a six-fold higher capacity than a dialysis filter modified with the same mass of standard antigen. We conclude that selective blood group antibody removal by antigen modified dialysis filters is feasible and may be a simpler system than plasmapheresis followed by immunoadsorption.

Molecular mimicry in HLA-B27-related arthritis.

A unique feature of patients with ankylosing spondylitis and reactive arthritis is that almost all share the HLA type B27. The primary structures of the HLA-B27 antigens have been determined. At least six variants exist. However, disease predisposition does not appear to be restricted to a particular variant. One hypothesis about the pathogenesis of arthritis is that the bacteria that cause the arthritis carry components that are cross-reactive with HLA-B27 antigens. Several reactive bacterial components have indeed been identified using monoclonal anti-HLA-B27 antibodies. Even more striking is the identification, through a computerized search, of a Klebsiella protein. This protein carries a stretch of six amino acids identical to residues 72 to 77 of two of the HLA-B27 variants. A synthetic peptide carrying these six amino acids of HLA-B27 protein is reactive with serum antibodies in some patients with arthritis. With this knowledge, investigators will be able to formulate new approaches for examining the pathogenesis of HLA-B27-associated arthritis.

A monoclonal anti-HLA-B27 antibody which is reactive with a linear sequence of the HLA-B27 protein is useful for the study of molecular mimicry.

In the search for cross-reactivity between bacteria and HLA-B27, three groups of investigators have identified several bacterial envelope proteins which are reactive with the monoclonal anti-HLA-B27 antibodies B27.M1 and B27.M2. Since these two antibodies react poorly with HLA-B27-derived synthetic peptides, it is not possible to locate the reactive epitopes on the HLA-B27 using synthetic peptides. Here, we introduce Ye-2, a monoclonal anti-HLA-B27 antibody which, unlike B27.M1 and B27.M2, is reactive with a synthetic peptide derived from residues 63-84 of HLA-B27.1. Analysis with a cross-reactive peptide derived from residues 226-244 of bovine carbonic anhydrase suggests that only a few of the amino acid residues in the HLA-B27-derived peptide are responsible for the reactivity. This antibody should be a useful adjunct in a preliminary assessment of whether a bacterial protein mimics HLA-B27 in primary structure.

Chemotaxis under agarose: dependence upon polymorphonuclear leukocyte density.

Since the number of polymorphonuclear leukocytes (PMN) responding to a chemotactic stimulus in the Boyden chamber is influenced by the cell density, we studied whether this variable was important in determining chemotaxis under agarose. The chemotactic index was determined by summing the product of each cell that had migrated from the chemotactic well by its distance and correcting this sum for random migration. Cell density (number of PMN per mm2 surface area of the agarose plate well) influenced the number of cells responding to the chemotactic stimulus. Only when more than 1.8 x 10(2) PMN/mm2 were used was chemotaxis then detected. For cell densities greater than this number, there was a highly positive correlation between cell density and chemotactic index (P less than 0.001). These findings are consistent with previous reports and indicate that PMN density is a critical variable when using the agarose method. In addition, these studies provide further evidence for cellular cooperation in the initial phases of the chemotactic response.

Alteration of polymorphonuclear leukocyte surface charge by endogenous and exogenous chemotactic factors.

We investigated the effects of chemotaxins on the surface charge of isolated human PMN. Chemotaxis was ascertained using Boyden chambers. Surface charge was calculated using data derived from cell mobility as measured in a cytophorometer. The electrophoretic mobility of cells exposed to all chemotactic agents studied was altered. Endotoxin-activated serum containing C5a, PMN lysosomal extracts from "resting" and from crystal-stimulated cells, and Gly-His-Gly, a synthetic tripeptide, all significantly reduced the net negative surface charge of isolated neutrophils. Only chemotactically active substances effected this change; controls including heat-inactivated serum, other subcellular fractions, and an inert tripeptide, Gly-Gly-Gly, did not alter cell mobility in an electric field. These findings are consistent with studies by others on the effects of chemotactic factors on cation fluxes and cell aggregation and suggest a possible mechanism by which PMN directional migration is regulated.