Isolation and characterization of an age-related antigen present on senescent human red blood cells.

Autologous membrane-bound IgG was isolated from a subpopulation of human red blood cells (RBC) with specific density greater than 1.110, by affinity chromatography of purified RBC membrane glycoprotein preparations using immobilized wheat germ agglutinin and immobilized anti-human immunoglobulin (Ig) as immunoabsorbents. The Ig-containing population thus obtained, when further separated by chromatography on Sephadex G-200 in the presence of chaotropic agents, yielded four peaks (Ia, Ib, II, and III). Double immunodiffusion revealed the presence of Ig in the first three peaks (IgM in peak Ia, IgA in Ib, and IgG in II) but not in peak III. Peak III was precipitated by the Ig-containing peaks (Ia, Ib, and II) in immunodiffusion assays, suggesting that the antigenic membrane determinants responsible for the binding of autologous Ig to senescent human RBC were contained in this peak (III). Peaks Ia, Ib and II precipitate purified asialoglycophorin; peak III was reactive with purified autoantibodies directed against asialoglycophorin. These results suggest that an age-related antigenic determinant(s) present on senescent human RBC is exposed by desialylation of the major sialoglycoprotein component of the RBC membrane.

Binding of immunoglobulin classes to subpopulations of human red blood cells separated by density-gradient centrifugation.

Human erythrocytes (RBC) from whole blood were separated according to their specific densities by centrifugation on a polyvinyl-pyrrolidine-coated colloidal silica matrix (Percoll) into four major subpopulations. By indirect immunofluorescence assay, the most dense RBC subpopulation, with specific density greater than 1.110 g/ml (3%-5% of total RBC), was positive for membrane-bound immunoglobulin; the remaining, less dense subpopulations were negative. IgG was present on 85%-95%, IgM on 28%-32%, and IgA on 15%-20% of the RBC in the most dense population. When these immunoglobulins were eluted, radiolabeled, and used in binding studies with autologous RBC fractions subjected to thermal and/or enzymatic treatment, they reacted specifically with the less dense RBC subpopulations. These results suggest that previously cryptic antigens were revealed by the activity of neuraminidase on the plasma membranes of the treated RBC.

A comparison of the N-terminal amino acid sequences of early and late pools of anti-DNP and anti-DNP-p-aminobenzoylglutamate antibody light chains.

The N-terminal amino acid sequences of early and late pools of anti-DNP and anti-DNP-p-aminobenzoylglutamate (DNP-ABG) antibody light chains were quantitatively determined and compared. The amino acid composition at each locus of the N-terminal 20 amino acid residues of each light chain preparation was determined using automatic sequencing techniques coupled with high-pressure liquid chromatography and mass spectrometry. The sequence data obtained for the light chains corresponding to antibodies isolated early in the immune response (3-4 weeks) were essentially the same as those for light chains from antibodies isolated late in the response (12-14 weeks). In addition, it was observed that the sequence data obtained for the anti-DNP antibody light chain preparation were almost identical to those found for two anti-DNP-ABG antibody light "hain preparations. The sequence data obtained in the present study were compared with those obtained for normal rabbit light chains and with the composite sequence data published for other rabbit anti-hapten light chains.

Antigens of human trophoblasts: a working hypothesis for their role in normal and abnormal pregnancies.

This report describes the preparation and characterization of antisera to human trophoblast membranes. Rabbit antisera were raised to trophoblast microvilli prepared by differential ultracentrifugation. Antibodies to serum proteins were removed by solid-phase immunoabsorption with normal human serum, and indirect immunofluorescence experiments with cryostat sections of human placentas showed that the absorbed anti-trophoblast sera reacted with trophoblasts as well as with stromal cells and endothelium of chorionic villi. The antisera also produced membrane fluorescence when studied on viable lymphocytes and certain human cell lines. These anti-trophoblast sera were also lymphocytotoxic, and this reaction was abolished by prior absorption of the antisera with leukocytes. The leukocyte-absorbed anti-trophoblast sera retained their ability to react with trophoblasts and certain human cell lines, but no longer reacted with lymphocytes or placental stromal cells and endothelium. Two categories of trophoblast membrane antigens are thus defined: one present on trophoblasts and certain human cells lines (tentatively designated TA(1)), and the other on trophoblasts and lymphocytes, villous fibroblasts, and endothelium (tentatively designated TA(2)). A working hypothesis is proposed stating that normal pregnancy involves the generation of anti-TA(2) subsequent to blastocyst implantation and entrance of trophoblasts into the maternal circulation. This involves a mechanism similar to allogeneic cell stimulation and results in antibodies that block either the recognition or cytotoxicity of TA(1). Failure to mount this response allows TA(1) recognition and trophoblast immunopathology. Experimental and clinical studies in support of this working hypothesis, particularly involving abortion and toxemia, are cited from published reports.

Immunological mechanism for spontaneous abortion in systemic lupus erythematosus.

The incidence of lymphocytotoxic antibodies in patients with systemic lupus erythematosus (S.L.E.) was significantly lower during pregnancies ending in normal live births than in pregnancies ending in spontaneous abortions (P less than 0.005). It was possible to absorb the lymphocytotoxic antibodies from S.L.E. sera with purified trophoblast antigens. The presence of a trophoblast-reactive lymphocytotoxic antibody which fails to disappear during pregnancies which end in abortion suggests an immunological mechanism for spontaneous abortion in S.L.E.

N-bromosuccinimide oxidation of anti-DNP and anti-DNP-p-aminobenzoylglutamate antibodies in the presence and absence of protecting hapten.

The reagent N-bromosuccinimide (NBS) has been employed to investigate the role of tryptophan in hapten binding in anti-DNP (H-1) and anti-DNP-p-aminobenzoylglutamate (DNP-ABG) (I-13) antibodies. In 0-1 M acetate (pH 4-0) buffer fifteen and sixteen moles of tryptophan in the anti-DNP and anti-DNP-ABG antibodies respectively were reactive toward NBS. The hapten DNP-lysine protected 1 tryptophan in antibody H-1 and three tryptophans in antibody I-13 from NBS modification. DNP-ABG protected three tryptophans in antibody H-1 and five tryptophans in antibody I-13 from NBS oxidation. NBS treatment of the unprotected antibodies resulted in a significant, but not total inhibition of hapten binding, while in the hapten protected antibody preparation no significant loss of binding occurred due to NBS treatment. The binding activity of the anti-DNP antibody H-1 was more sensitive to NBS oxidation than was the anti-DNAP-ABG antibody I-13. This was apparently due to the larger number of oxidizable tryptophans in I-13 making it less sensitive to overall tryptophan modification. The results of these investigations are discussed in terms of the antibody combining site model proposed by Haselkorn et al. (Haselkorn, Friedman, Givol and Pecht, 1974) derived from kinetic mapping of the antibody-combining site by chemical relaxation spectroscopy.

A comparison of the isoelectrofocusing porperties of antibodies to DNP, DNP-glycylglycylglycine and DNP-p-aminobenzoylglutamate.

The charge of heterogeneity of antibodies to DNP-glycylglycylglycine (DNP-Gly-3) and DNP-p-aminobenzoylglutamate (DNP-pABG) has been investigated using preparative liquid isoelectric focusing techniques. The focusing profiles of the two antibody preparations are qualitatively similar, each containing four or five major peaks. The results are also similar to the profiles obtained earlier for DNP antibodies. The binding properties of refocused fractions from the electro-focusing separation of the two DNP-specific antibodies were investigated. Both the unfocused and refocused DNP-pABG antibody fractions were found to be functionally less heterogeneous than the corresponding DNP-Gly-3 fractions, suggesting that the site-filling capacity of the haptens contributed to the observed functional heterogeneity. A marked increase in K0 with increasing pI of the focused antibody fraction was observed for the DNP-pABG antibody preparation. A much smaller increase of K0 with pI was observed for DNP-Gly-3 antibodies and no observed change was found for DNP antibodies. These differences probably reflect the difference in the net charge on the three DNP hapten groups. The isoelectric focusing properties of the separated light and heavy chains from a DNP-antibody preparation and a DNP-pABG antibody preparation were also determined. The light chains in both instances focused as one or two major bands, while the profile of the heavy chain in both cases exhibited five to six major bands. These results are discussed in terms of the model proposed by Haselkorn, Friedman, Givol and Pecht (1974).

Purification of DNP specific antibodies using sepharose bound DNP-para-aminobenzoylglutamate.

An immunoabsorbent column employing a DNP derivative having restricted molecular freedom (dinitrophenylated-para-aminobenzoylglutamate--Sepharose 4B) was prepared in order to isolate and purify DNP specific antibodies in high yield. The antibodies were recovered from the immunoabsorbent column in yields from 80 to 90% using 3 M sodium thiocyanate, and these antibodies retained their immunologic activity. There appeared to be limited selection of antibodies with specific affinities as was noted in a parallel experiment comparing a second antibody purification procedure.