Histidine-proline-rich glycoprotein binding to platelets mediated by transition metals.

Histidine-proline-rich glycoprotein (HPRG) binds zinc, which in turn promotes HPRG binding to lymphocytes and monocytes. We examined the possibility that zinc and other transition metals also promote HPRG binding to platelets. Only non-specific, unsaturable association of HPRG with resting or activated platelets was observed in the absence of transition metals. However, nickel, cobalt, copper, cadmium, and zinc greatly increased HPRG association with the cells. In the presence of zinc, specific, saturable binding of HPRG to platelets was demonstrated. The cell binding capacity for HPRG could be increased by increasing the zinc saturation of HPRG from 10% to 30% as well as by activating the platelets with thrombin. Because platelets contain relatively high concentrations of secretable zinc, it is possible that significant amounts of HPRG bind to activated platelets at sites of blood clotting and that this has a physiologic function.

Comparison of the effect of histidine-rich glycoprotein and 6-aminohexanoic acid on plasmin production and fibrinolysis in vitro.

Because histidine-rich glycoprotein binds to the kringle 1-3 domain of plasminogen, it may affect fibrinolysis by reducing fibrin-dependent plasmin production, and in this way it could be mechanistically analogous to 6-aminohexanoic acid. We tested this hypothesis by comparing the effects of histidine-rich glycoprotein and 6-aminohexanoic acid in an in vitro assay of fibrin-dependent plasmin production mediated by tissue plasminogen activator. Whereas 1 mM of 6-aminohexanoic acid increased the K(m) of the reaction from approximately 0.22 microM to approximately 1.7 microM, 2 microM of histidine-rich glycoprotein had no discernible effect. Similar results were obtained in an assay based upon fibrin clot lysis. Therefore, we could not document an effect of histidine-rich glycoprotein on the rate of fibrin-dependent plasmin production.

Serologic evidence of heparin sensitization in cancer patients receiving heparin flushes of venous access devices.

Cancer patients with venous access devices (VADs) often receive daily flushes of heparin. Even this relatively small heparin exposure has been reported to induce immune-mediated thrombocytopenia. To estimate how frequently this occurs we tested for heparin-related antibodies in 49 patients receiving daily heparin flushes of their VADs. Although one-third of the patients showed evidence of heparin sensitization on at least one occasion during their surveillance, their antibody titers were generally low and typical of those found in other cohorts of patients who become sensitized to heparin but do not develop secondary thrombocytopenia. However, one patient developed a positive serotonin release assay indicative of a more significant sensitization, but without thrombocytopenia. Therefore, our observations suggest that heparin-induced thrombocytopenia (HIT) related to heparin flushes of VADs is uncommon but still an important diagnosis to entertain.

Fibrinolytic and coagulant responses to regional limb perfusions of tumor necrosis factor, interferon-gamma, and/or melphalan.

Regional limb perfusion with antineoplastic agents stresses the local vasculature in a variety of ways. However, by monitoring the perfusates from limbs treated with melphalan alone or with melphalan plus tumor necrosis factor (TNF) and interferon-gamma (IFN-gamma), we were able to distinguish the effect of the cytokines on the observed coagulant and fibrinolytic responses. We collected samples of effluent from a series of lower extremities that were perfused with the cytokines and/or melphalan as treatment for localized melanoma. Both regimens produced statistically significant evidence of coagulant and fibrinolytic activation. However, limbs receiving cytokines in addition to the melphalan responded with a sharper rise in tissue plasminogen activator (tPA) and plasmin (plasmin-antiplasmin complexes [PAP]) than limbs treated with melphalan alone. Evidence of thrombin formation (prothrombin fragment 1 + 2 [F1 + 2], thrombin-antithrombin complexes [TAT]) was also greater when the cytokines were included, although the response was delayed and less consistent than the fibrinolytic activation.

Mechanisms underlying the morning increase in platelet aggregation: a flow cytometry study.

OBJECTIVES: Mechanisms underlying the morning increase in platelet aggregation produced by arising and assuming the upright posture were studied by examining 1) the expression on the platelet surface of activation-dependent markers; 2) platelet aggregation in whole blood; and 3) hematologic factors likely to influence aggregation. BACKGROUND: The morning increase in thrombotic cardiovascular events has been attributed, in part, to the morning surge in platelet aggregability, but its mechanisms are poorly understood. METHODS: Expression of seven platelet surface antigens (including P-selectin, activated GPIIb,IIIa and GPIb-IX), whole-blood platelet aggregation, platelet count and hematocrit were measured before and after arising in 17 normal volunteers. The fibrinolytic variables, tissue-type plasminogen activator, plasminogen activator inhibitor 1 and catecholamine levels were also measured. RESULTS: On arising and standing, platelet aggregation increased by 71% (p < 0.01) and 27% (p < 0.03) in response to collagen and adenosine diphosphate, respectively. However, there was no change in any of the activation-dependent platelet surface markers. Whole-blood platelet count and hematocrit increased by 15% and 7% (both p < 0.0001), respectively. Norepinephrine and epinephrine levels increased by 189% (p < 0.0001) and 130% (p < 0.01), respectively. Tissue-type plasminogen activator antigen increased (31%, p < 0.01), but there was no significant increase in plasminogen activator inhibitor 1, suggesting an overall increase in fibrinolysis on standing. Prothrombin fragment 1.2 increased by 28% (p < 0.02), indicating a small increase in thrombin generation. The increases in hematocrit and platelet count that occurred on standing were carefully mimicked in vitro and resulted in a 115% (p < 0.05) increase in platelet aggregation in response to adenosine diphosphate. CONCLUSIONS: These data demonstrate that the morning increase in platelet aggregation is not accompanied by expression of activation-dependent platelet surface receptors and suggest that the increase in whole-blood aggregation may be primarily due to the increases in catecholamine levels, platelet count and hemoconcentration.

1-Desamino-8-arginine-vasopressin corrects the hemostatic defects in type 2B von Willebrand's disease.

DDAVP is effective treatment in most types of von Willebrand's disease; however, in type 2B von Willebrand's disease the use of DDAVP has been contraindicated due to DDAVP-induced thrombocytopenia. Several reports have confirmed the thrombocytopenic effects of DDAVP and the presence of circulating platelet aggregates in type 2B von Willebrand's disease. We have infused three type 2B patients with DDAVP. The three patients had different mutations of their vWf. All three patients had a missense mutation which resulted in a single amino acid substitution in the disulfide loop of the A1 domain. Administration of 20 micrograms of DDAVP resulted in significant elevations of factor VIII, vWf antigen, and ristocetin cofactor levels. In contrast to other studies, DDAVP did not induce or enhance thrombocytopenia in these three patients. When blood was obtained by fingerstick and diluted into sodium oxalate (Unopette) or EDTA (Microvette), the platelet counts did not change over 4 hr. In contrast, blood collected directly into evacuated tubes containing sodium citrate, lithium heparin, or EDTA consistently demonstrated varying degrees of thrombocytopenia and platelet clumping. We also observed a shortening of the pre-infusion bleeding time over the 4 hr period. All three patients have been studied twice and each has shown consistent results. DDAVP appears to be a useful form of treatment in type 2B vWd.

Activated platelets in paroxysmal nocturnal haemoglobinuria.

One of the major causes of morbidity and mortality in paroxysmal nocturnal haemoglobinuria (PNH) is venous thrombosis. We have studied fibrinolysis, coagulation and platelets in 11 patients with PNH in an attempt to identify the possible mechanism(s) of thrombosis in PNH. In this study we did not identify any fibrinolytic defects, evidence of coagulation activation, nor reduction in coagulation inhibitors. In contrast, in this cohort of 11 PNH patients we have identified varying degrees of platelet activation as defined by the surface expression of activation-dependent proteins and the binding of adhesive proteins to the platelet surface. The thrombotic events in PNH usually occur in the venous system. Our studies and previous experimental studies suggest that anti-platelet therapy may be efficacious in reducing the incidence and severity of venous thrombosis in PNH.

Reductions in tissue plasminogen activator and thrombomodulin in blood draining veins damaged by venous access devices.

A frequent complication of venous access devices (VADs) is axillary-subclavian venous thrombosis. To study this problem we have compared blood drawn through VADs with peripheral blood samples in a group of oncology patients with venographically demonstrated venous damage (N = 14) and a group with normal venograms (N = 21). The samples were assayed for a battery of proteins believed to be involved in thrombogenesis. After approximately six weeks of catheterization the venographically abnormal patients had significantly less thrombomodulin (P = 0.0055) and significantly higher PAI:tPA (P = 0.022) in catheter-drawn samples as compared with the venographically normal group. Although the data are inconclusive, it is hypothesized that these changes resulted from local endothelial injury.

False positive seroreactivity to Borrelia burgdorferi in systemic lupus erythematosus: the value of immunoblot analysis.

UNLABELLED: The object of this study was to determine the incidence of seropositivity to B. burgdorferi by the commonly available enzyme-linked immunosorbent assay (ELISA) in patients with SLE and other rheumatic diseases and to evaluate immunoblot analysis as a tool to differentiate true from false positive ELISA. Sera were obtained from patients with SLE (n = 35), rheumatoid arthritis (n = 26), seronegative arthritis (n = 28) and Lyme disease (n = 18). Reactivity to B. burgdorferi antigens was analysed by two available diagnostic techniques: ELISA and immunoblot. Correlations were made between seroreactivity to B. burgdorferi and standard serological tests of autoimmunity: antibodies to nuclear antigens, dsDNA, cardiolipin, SSA and SSB. Seroreactivity to B. burgdorferi antigens by the ELISA system was detected in 40% of patients with SLE, 8% of patients with rheumatoid arthritis and 4% with seronegative arthritis. Among patients seropositive by ELISA, immunoblots were negative in all cases. However, eight of 14 patients with rheumatoid arthritis (57%) showed cross-reactivity to multiple borreli antigens. No significant correlations were found between Lyme seropositivity by ELISA and other autoantibodies except IgM rheumatoid factor (r = 0.61, P < 0.01) in patients with rheumatoid arthritis. IN CONCLUSION: a positive ELISA for Lyme disease was found in up to 40% of patients with established SLE and also in other rheumatic diseases. However, specific serum antibodies to Borrelia were not confirmed by the more specific immunoblot technique.(ABSTRACT TRUNCATED AT 250 WORDS)

Modulation of human T cell responses by nitric oxide and its derivative, S-nitrosoglutathione.

OBJECTIVE: To examine the effects of nitric oxide (NO) and its more stable derivative, S-nitrosoglutathione (SNO-GSH), on the response of activated T lymphocytes. METHODS: The effects of NO and SNO-GSH on DNA synthesis, interleukin-2 (IL-2) production, IL-2 receptor expression, and cGMP accumulation were determined in phytohemagglutinin-activated peripheral blood mononuclear cells (PBMC) and spleen T cells. RESULTS: Nitric oxide (half-life [T1/2] < 15 seconds) did not inhibit T cell proliferation. However, the derivative SNO-GSH (25 microM) (T1/2 > 2 hours) inhibited DNA synthesis by a mean +/- SD of 65 +/- 19.6% (P < 0.001) in PBMC and 75 +/- 15% (P < 0.001) in spleen cells. Macrophage depletion of PBMC did not abrogate the inhibition. SNO-GSH had no effect on IL-2 production or IL-2 receptor expression. NO (25 microM) increased the cGMP content of PBMC (0.65 +/- 0.15 pmoles/10(6) cells; P < 0.04), as did SNO-GSH (25 microM) in both PBMC (3.8 +/- 1; P < 0.001) and spleen T cells (5.2 +/- 1.2; P < 0.001). Methylene blue and hemoglobin, which are NO inhibitors, inhibited SNO-GSH-induced cGMP accumulation (P < 0.001). CONCLUSION: SNO-GSH inhibits T cell DNA synthesis independently of IL-2 production and in association with cGMP accumulation via a NO-dependent mechanism. We suggest that NO and its S-nitrosothiol derivatives may act as endogenous inhibitors of T cell-mediated inflammation.

Comparison of assay systems for detecting antibodies to nuclear ribonucleoproteins.

The specificity and sensitivity of two commercial enzyme linked immunosorbent assays (ELISAs), Diamedix (Miami, Florida) and Lipogen (Noxville, Tennessee), were assessed and compared with haemagglutination and immunodiffusion assays. Sera from 53 patients with various connective tissue disorders were examined for the presence of antibodies to nuclear antigens (ANA), double stranded DNA (dsDNA), Sm, RNP, SSA/Ro, and SSB/La. Of the 53 patients, 42 were ANA positive, 11 were ANA negative, and 22 had antibodies to dsDNA. Seven patients had antibodies to Sm by haemagglutination assay; these were also positive in both ELISA systems (only five of the seven patients were assayed by the Lipogen ELISA system). Two additional Sm positive values were obtained in each of the ELISA systems but only one of these was positive in both. Ten positive RNP results were obtained by haemagglutination and nine of these were also positive by the Diamedix ELISA. Only eight samples were tested by the Lipogen assay and seven of these were positive. Three additional RNP positive values were obtained by the Diamedix and six by the Lipogen ELISA assays. Of these, only two were positive in both. Antibodies to SSA/Ro were obtained in 11 patients by immunodiffusion and lines of partial identity were observed in nine. SSB/La antibodies were positive in six patients and two had lines of partial identity. All the SSA/Ro and SSB/La positive sera were also positive in both ELISA systems. Moreover, eight additional SSA/Ro positive values were obtained in each of the ELISAs, four of which had partial identity lines in the immunodiffusion assay. Furthermore, three additional SSB/La positive values were obtained by the Diamedix and four by the Lipogen assays. Of these, only two were positive in both ELISAs. This study shows that the above ELISAs are comparable in specificity and sensitivity with haemagglutination assay for detection of antibodies to Sm and RNP antigens and are more sensitive than immunodiffusion for the detection of SSA/Ro and SSB/La antigens.

Advances in clinical immunology.

Autoantibodies (antibodies to self-antigens) are a hallmark of rheumatic diseases. These antibodies are directed against nuclear or nonnuclear antigens and are regarded as marker antibodies for particular diseases.

A novel association of DQ alpha and DQ beta genes in the DRw10 haplotype. Determination of a DQw1 specificity by the DQ beta-chain.

The association of the class II genes of the DRw10 haplotype from a cell line, NASC, initiated from a member of a well characterized family, was analyzed by sequencing cDNA clones corresponding to DR beta I, DQ alpha, and DQ beta genes. An identical haplotype was also identified in the Raji cell line. In addition to typing as DRw10 and DQw1 with HLA typing sera both, the NASC and Raji cell lines were shown to react strongly with the monoclonal antibodies 109d6 (specific for DRw10 beta 1 and DRw53 beta 2 gene products) and Genox 3.5.3 (specific for DQw1) and exhibited the restriction fragment length polymorphism indicative of a DRw10, DQw1 haplotype. The DR beta 1 gene corresponding to the DRw10 specificity was found to have a first domain sequence different from all other DR beta I genes. Sequence analysis of the 3'-untranslated region of this DR beta-chain gene showed a significant divergence from the 3' untranslated region of the DRw53 family of haplotypes and a lesser divergence from that of the DRw52 and DR1/DR2 families. The sequence of the DQ beta genes corresponding to the DQw1 specificity in the DRw10 haplotype was found to be identical to the DQ beta gene from a DR1, DQw1 haplotype. Surprisingly, however, the DQ alpha gene did not resemble other DQw1-like DQ alpha genes, but was identical in sequence to the DQ alpha gene found in DR4 haplotypes. The novel association of DQ alpha and DQ beta genes in the DRw10 haplotype revealed in these studies may result from a double recombinational event. More consequentially, these studies strongly suggest that the DQw1 specificity recognized by Genox 3.5.3 is determined by the DQ beta chain and is not affected by the DQ alpha-chain.

Class II major histocompatibility complex gene sequences in rheumatoid arthritis. The third diversity regions of both DR beta 1 genes in two DR1, DRw10-positive individuals specify the same inferred amino acid sequence as the DR beta 1 and DR beta 2 genes of a DR4 (Dw14) haplotype.

The DR1 and DRw10 beta 1 chain genes were isolated from each of 2 individuals with rheumatoid arthritis who were heterozygous for these class II major histocompatibility complex specificities. The sequences of the DR1 beta 1 chains from both patients were identical, differing from previously reported DR beta 1 chains of individuals without RA by 2 amino acid substitutions, at positions 85 (Val-Ala) and 86 (Gly-Val), and by a silent mutation at the last nucleotide of codon 78 (C-T), resulting in the loss of a Pst I restriction endonuclease site. Identical DRw10 beta 1 chain genes were found in both patients. These were shown to encode the epitope recognized by monoclonal antibody 109d6. This antibody also recognizes an epitope on the DRw53 beta 2 chain of the DR4 haplotype. The third diversity regions of the DR1 beta (amino acids 67-74) and the DRw10 beta 1 chains (amino acids 67-73) were identical, respectively, with those of the DR4 (Dw14) beta 1 and beta 2 chains, raising the possibility that in these patients, the third diversity regions of the two DR beta 1 chain genes present in trans are conformationally equivalent to the cis-encoded third diversity regions of the DR4 (Dw14), DR beta 1, and beta 2 chains. The nucleotide sequences of the DQ beta complementary DNA clones were identical to that of the DQw1 beta chain, and no DR beta 2 complementary DNA clones were identified.

Nucleotide sequence of a DRw10 beta chain cDNA clone. Identity of the third D region with that of the DRw53 allele of the beta 2 locus and as the probable site encoding a polymorphic MHC class II epitope.

The nucleotide and inferred amino acid sequence of a DRw10 beta chain was obtained from cDNA clones isolated from a DR1, DRw10 heterozygous cell line. The sequence of this beta chain gene was distinctive, differing from those of all other defined DR types. The DRw10 beta chain gene was shown by transfection experiments to encode a polymorphic epitope recognized by mAb 109d6 that is also encoded by the DRw53 beta 2 chain gene. Comparison of the nucleotide sequence of both genes revealed that their third D regions (amino acids 67 to 73) were identical. This suggested first that the 109d6 epitope could be encoded by residues of this region, and second, that a putative gene conversion event transferred this sequence along with the information encoding the 109d6 epitope from a donor gene such as DRw53 beta 2. The sequence of the DRw10 beta chain gene was observed to be identical to that of clone pII beta 4 derived from the non-DR3 haplotype in the Raji cell line, which was also demonstrated to express the determinant recognized by antibody 109d6, suggesting that the typing of this cell line is HLA-DR3/DRw10. No evidence was found for the existence of a DR beta 2 chain gene product encoded by the DRw10 haplotype. The DRw10 haplotype was of particular interest because it was present along with a DR1 haplotype in the propositus who had rheumatoid arthritis, and was shared by the DR4-positive son of the propositus, who also had rheumatoid arthritis. This raised the possibility that the DRw10 haplotype, and most probably one or more specific conformations encoded by the DR beta chain, are involved in the definition of the disease susceptibility phenotype.

Hypotheses on the molecular basis of susceptibility to rheumatoid arthritis.

An interpretive summary of recent immunologic and molecular biologic data concerning the molecular basis of susceptibility of rheumatoid arthritis will be presented. The central point of view is taken that the MHC class II molecules encoding disease susceptibility function in a specific immune recognition event. This could involve an antigen "X" that currently eludes characterization or be directed to polymorphic determinants on the MHC molecule itself. The problem of understanding the meaning of the association of susceptibility to rheumatoid arthritis with diverse MHC alleles such as DR4 (Dw4 and Dw14) and DR1 is approached by detailed biochemical analysis that led to the identification of common stretches of amino acid sequence, presumably encoding conformationally equivalent structures. The sequence shared by the otherwise unrelated DR1 and DR4 haplotypes from residue 67 in the DR a chain that appears to confer susceptibility is Leu-X-X-Gln-Arg/Lys. Non-classic MHC polymorphisms related to disease susceptibility but not associated with particular alleles such as identified by Ab109d6 prove especially valuable in suggesting new directions for attempting to understand the significance of these associations. Consideration is given to the possibility that a family of either slightly different or identical conformations encoded in either cis or trans cumulatively confer the liability to develop rheumatoid arthritis. This implies a highly non-classic mode of inheritance. It seems reasonable to consider the pathogenesis of rheumatoid arthritis as evolving from a typical immune response based on a simple immune recognition event directed to a single antigen.

Molecular diversity of HLA-DR4 haplotypes.

Complementary DNA (cDNA) clones encoding beta chains of the DR and DQ regions and alpha chains of the DQ region were isolated and sequenced from four homozygous DR4 cell lines of different HLA-D types: GM3103(Dw4), FS(Dw10), BIN(Dw14), and KT3(Dw15). When compared with each other and with a previously published sequence from a DR4 (Dw13 cell line), the variability of DR beta 1 gene products is generally restricted to the region around amino acid position 70, with an additional polymorphism at position 86. Many of these differences, including an unusual amino acid substitution at position 57 in the Japanese cell line KT3(Dw15), may be due to gene conversion events from the DR beta 2 or DX beta genes. In contrast, DR beta 2 molecules are identical in Dw15, Dw10, and Dw4 cell lines. DQ beta chains isolated from GM3103(Dw4), FS(Dw10), and BIN40(Dw14) are also identical. However, the DQ beta sequence from cell line KT3(Dw15) differs substantially from all other previously reported DQ beta alleles, consistent with its serological designation, DQ "blank." The first domain sequences of DQ alpha chains were identical in all four cell lines. The data suggest that relatively circumscribed amino acid changes in the DR beta 1 molecule are responsible for the HLA-D typing differences between some haplotypes.