|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||
Background Factor VIII inhibitors are IgG alloantibodies that arise during replacement therapy in 25 to 50 percent of patients with severe hemophilia A. The hydrolysis of factor VIII by anti–factor VIII antibodies has been proposed as a mechanism of inactivation of factor VIII.
Methods We purified IgG from patients with severe hemophilia A. The proteolytic activity of the antibodies was assessed by incubating the IgG with biotinylated human factor VIII and analyzing patterns of factor VIII cleavage by sodium dodecyl sulfate–polyacrylamide-gel electrophoresis and immunoblotting. The controls were normal human IgG and IgG purified from plasma of patients with hemophilia who did not have inhibitory antibodies.
Results Significant proteolytic activity was detected in IgG from 13 of 24 inhibitor-positive patients. No hydrolytic activity was detected in control antibodies of IgG from patients without inhibitors. The rate of hydrolysis of factor VIII by purified IgG correlated positively with the factor VIII–neutralizing activity of IgG in plasma (r2=0.67, P=0.029). Principal-component analysis of migration profiles of digestion fragments demonstrated the heterogeneity of the catalytic potential of factor VIII inhibitors among patients.
Conclusions Proteolysis is a mechanism by which IgG antibodies against factor VIII can inactivate factor VIII.
Catalytic antibodies are immunoglobulins with a capacity to hydrolyze the antigen for which they are specific. The proteolysis of factor VIII by anti–factor VIII alloantibodies has recently been reported as an additional mechanism by which factor VIII inhibitors can neutralize the procoagulant function of factor VIII.9 In the present study, we investigated the prevalence of catalytic IgG antibodies against factor VIII among patients with severe hemophilia A and inhibitors.
Methods
Patients
Frozen plasma samples from 24 patients with factor VIII inhibitors and severe hemophilia A (plasma factor VIII levels less than 1.0 percent of normal) were obtained from the Hôpital Cochin (Paris), Hôpital du Kremlin-Bicêtre (Paris), Hôpital de Saint-Etienne (St. Etienne, France), Gasthuisberg Leuven (Leuven, Belgium), and Christian Medical College Hospital (Vellore, India). Most of the patients had received a diagnosis of severe hemophilia within the first two years of life. The mean (±SD) age of the patients was 40.6±18.4 years (range, 10 to 70). Three patients had a low inhibitory titer (less than 5 Bethesda units [BU] per milliliter), and 21 had an inhibitory titer of at least 5 BU per milliliter. Plasma samples were also obtained from four patients with severe hemophilia A and no detectable inhibitor activity in the plasma. Inhibitor-positive patients who had mild hemophilia, or who had severe hemophilia and were undergoing induction of immune tolerance, were not included in the study.
Purification of IgG
IgG was isolated from plasma by ammonium sulfate precipitation followed by chromatography on protein G Sepharose. A therapeutic preparation of pooled normal human immune globulin (Sandoglobulin, ZLB Bioplasma, Bern, Switzerland) was used as a source of normal IgG. To exclude contaminating proteases, size-exclusion chromatography of IgG and immune globulin was performed on a superose-12 column equilibrated with 50 mM TRIS, 8 M urea, and 0.02 percent sodium azide (pH 7.7) at a flow rate of 0.25 ml per minute. The IgG-containing fractions were then pooled and dialyzed against 50 mM TRIS, 100 mM glycine, and 0.02 percent sodium azide (pH 7.7) for two days at 4°C. The purity of IgG preparations was assessed by sodium dodecyl sulfate–polyacrylamide-gel electrophoresis (SDS-PAGE) and immunoblotting under nonreducing conditions. IgG was quantified by enzyme-linked immunosorbent assay. In the case of Patient 22, anti–factor VIII IgG antibodies were purified from total IgG by affinity chromatography on a human factor VIII matrix, as previously described.10
Evaluation of Factor VIII–Inhibitory Activity
Factor VIII–inhibitory activity was measured in plasma and in purified IgG fractions according to the method of Kasper et al.11 The detection limit of the Bethesda assay was 0.3 BU.
Hydrolysis of Biotinylated Factor VIII
Commercially available recombinant human factor VIII (Kogenate II, Bayer, Pittsburgh), which contains imidazole, glycine, sucrose, chloride, sodium, calcium, histidine, and polysorbate 80 as stabilizing agents and is free of other proteins, was used in all experiments. After desalting on a PD-10 column (Amersham Pharmacia Biotech, Uppsala, Sweden), factor VIII (3 ml at a concentration of 198 µg per milliliter) in borate buffer (100 mM boric acid [pH 7.0], 150 mM sodium chloride, and 5 mM calcium chloride) was conjugated with biotin (440 µl at a concentration of 25 µg per milliliter) in the dark for two hours at 4°C, as previously described.12 Biotinylated factor VIII (100 µg per milliliter, 385 mM) was incubated in 40 µl of 50 mM TRIS–hydrochloric acid (pH 7.7), 100 mM glycine, 0.025 percent Tween 20, and 0.02 percent sodium nitrate buffer, alone or in the presence of the IgG to be tested (25 µg per milliliter, 167 nM) for 24 hours at 37°C. Samples were mixed with Laemmli's buffer without mercaptoethanol (1:1 vol/vol). Twenty microliters of each sample was then subjected to 7.5 percent or 10 percent SDS-PAGE. The proteins were transferred onto nitrocellulose (Schleicher and Schuell, Dassel, Germany). After overnight blocking in phosphate-buffered saline, 1 percent bovine serum albumin, and 0.1 percent Tween 20 at 4°C, the membranes were incubated for 30 minutes at room temperature with streptavidin-coupled horseradish peroxidase (Amersham Pharmacia Biotech) diluted 1:3000 in blocking buffer. After being washed in phosphate-buffered saline and 0.1 percent Tween 20, the labeled proteins were revealed with the ECL kit (Amersham Pharmacia Biotech) and BioMax ML films (Kodak, Chalon-sur-Saône, France). The films were scanned with a SnapScan 600 scanner (Agfa, Mortsel, Belgium), and the images were processed with Color It software (MicroFrontier, Des Moines, Iowa). To calculate the rates of factor VIII hydrolysis, the densitometry of the protein bands was computed with National Institutes of Health image 1.62b7 software. Spontaneous hydrolysis occurring on incubation of factor VIII in the presence of buffer alone was considered to represent the background level and was subtracted from each analysis. The significance of the increase in the hydrolytic activity of patient IgG as compared with that of immune globulin (used as a negative control) was assessed by an analysis of variance post hoc test. The reported P values are two-sided.
Results
Inhibitory Activity against Factor VIII
The mean inhibitory activity against factor VIII was 2.7±1.5 BU per milliliter (range, 1 to 4) in the plasma of 3 patients with low titers and 54.6±62.5 BU per milliliter (range, 6.7 to 280) in the plasma of 21 patients with high titers. Inhibitory activity against factor VIII was also measured in purified IgG from the plasma of 21 of the 24 inhibitor-positive patients; in the case of Patients 2, 19, and 20, the quantity of IgG available was insufficient for all the assays. The inhibitory activity of purified IgG ranged between undetectable levels and 10.8 BU per milligram of IgG (Table 1). On average it was 1/10 that found in whole plasma, a result in agreement with an average IgG concentration in plasma of 10 mg per milliliter. There was a significant correlation between inhibitor activity in unfractionated plasma and that in the purified IgG fraction from plasma (r2=0.72, P<0.001 by Spearman rank correlation).
|
Figure 1 depicts the electrophoretic pattern of migration of biotinylated factor VIII after incubation with buffer or antibodies from different sources. Biotinylated factor VIII gave a characteristic pattern of several bands corresponding to molecular masses of 250 to 337 kD and an additional band of 82±2 kD. Incubation of biotinylated factor VIII with IgG purified from the plasma of inhibitor-positive patients caused a detectable cleavage of factor VIII in 14 of the 24 samples. Hydrolysis of factor VIII was most complete in Patients 2, 3, 6, 11, and 20 (Figure 1). The migration profiles obtained with antibodies from these five patients showed nine cleavage products of 57±1, 50±1, 47±1, 44±2, 38±1, 35±1, 33±1, 30±1, and 27±1 kD. Intermediate cleavage fragments were also observed, with molecular masses greater than 82 kD (Figure 2A). N-terminal sequencing of the cleaved fragments confirmed that all hydrolysis products were derived from factor VIII (data not shown). The migration profiles of factor VIII that had been incubated with normal human IgG that had no detectable inhibitory activity against factor VIII (Figure 1), or with IgG from the plasma of patients with hemophilia A who had no inhibitor (data not shown), were similar to that of factor VIII incubated in buffer alone. Four samples of purified IgG that hydrolyzed recombinant factor VIII did not hydrolyze recombinant factor IX or albumin under the same experimental conditions.
|
|
|
Discussion
Several mechanisms have been proposed to account for inactivation of factor VIII in patients with hemophilia A and factor VIII inhibitors.3,4,5,6,7,8,15 We recently reported that anti–factor VIII IgG antibodies of two patients with a factor VIII inhibitor could hydrolyze factor VIII.9 On the basis of the derived catalytic efficiency of these antibodies and the half-life of circulating IgG, we proposed that the hydrolysis of factor VIII may account, at least in part, for the rapid elimination of factor VIII from the plasma after the infusion of factor VIII into patients with factor VIII inhibitors.9 We found that 13 of 24 unselected patients with factor VIII inhibitors also had catalytic antibodies against factor VIII in their plasma. The correlation between the factor VIII–hydrolyzing activity of IgG and the factor VIII–neutralizing activity in plasma was statistically significant but not definitively established.
The prevalence of proteolytic antibodies found in our study may be an underestimate, since an isolated population of anti–factor VIII antibodies from the plasma of a patient with no detectable hydrolytic activity in the total IgG fraction (Patient 22) had hydrolytic activity against factor VIII. We have previously documented that the proteolytic activity of anti–factor VIII IgG is mediated by F(ab')2 fragments, that the purified IgG is free from contaminating proteases, and that the hydrolytic activity of the antibodies is not inhibited by a broad range of generic protease inhibitors.9,16,17 In the present study, the purification of IgG on protein G Sepharose, followed immediately by size-exclusion chromatography in 8 M urea, made it unlikely that the hydrolysis of factor VIII was caused by contaminating proteases.
Hydrolytic antibodies against vasoactive intestinal peptide, DNA, thyroglobulin, and prothrombin have been described in asthma,18,19 systemic lupus erythematosus,20,21 autoimmune thyroiditis,16 and multiple myeloma.17 The clinical relevance of such antibodies is, however, unclear. In hemophilia, by contrast, the association between the hydrolytic activity of antibodies against factor VIII and clinical manifestations (resistance to replacement therapy) is plausible.
Patients with hemophilia A constitute a heterogeneous group. In patients with mild-to-moderate or severe hemophilia A, factor VIII inhibitors can develop during treatment with human factor VIII. The probability of developing factor VIII inhibitors is greater in patients with large deletions in the factor VIII gene.22 We restricted our analysis to patients with severe hemophilia A and factor VIII–neutralizing activity in plasma. The relation between the hydrolytic activity of IgG and factor VIII–neutralizing activity was not consistent. Antibodies from some patients caused hydrolysis of factor VIII at low rates, but the plasma had strong inhibitory activity (Patients 9 and 16); in other cases, the IgG caused hydrolysis of factor VIII at high rates, but the plasma had weak inhibitory activity (Patients 3 and 14); in three samples, there were high rates of both hydrolysis and inhibitory activity (Patients 2, 11, and 20). This inconsistency may be a reflection of multiple mechanisms that participate simultaneously in inactivating factor VIII. Principal-component analysis of the results of factor VIII digestion, which takes into consideration the amount of factor VIII hydrolyzed per unit of time and the size of the proteolytic products, distinguished two groups of patients among the 13 patients, further emphasizing the heterogeneity of inhibitors in severe hemophilia A.
Sites on factor VIII that may be common targets of proteolytic anti–factor VIII antibodies are currently under investigation. Such studies may provide a rationale for the design of novel approaches to the treatment of patients with factor VIII inhibitors with hydrolytic activity.
Supported by the Institut National de la Santé et de la Recherche Médicale, the Centre National de la Recherche Scientifique, Bayer Corporation (Pittsburgh), and ZLB Bioplasma (Bern, Switzerland). Part of the work was supported by a grant from INSERM–Indian Council of Medical Research. Dr. Horn is the recipient of a fellowship from the Swiss National Foundation, and Dr. Bayry is the recipient of a fellowship from the Ministère de l'Education, France.
We are indebted to Dr. Alok Srivastava of the Department of Haematology of the Christian Medical College Hospital of Vellore (India) and Dr. Jacqueline Reynaud of the Département d'Hématologie des Hôpitaux de Saint-Etienne (France) for providing us with plasma samples; to Dr. Robert Calvora (Baxter, France) for providing us with recombinant human factor IX; to Stéphanie Rose and Cécile Artaud for technical assistance; and to the Bayer Corporation for providing factor VIII.
Source Information
From INSERM Unité 430 and Université Pierre et Marie Curie, Hôpital Broussais, Paris (S.L.-D., J.B., N.M., M.P.H., A.P., M.D.K., S.V.K.); Unité Mixte de Recherche 5094, Centre National de la Recherche Scientifique, Faculté de Pharmacie, Montpellier, France (S.V.); Centre des Hémophiles, Hôpital Cochin et Bicêtre, Paris (N.S., R.O.); the Center for Molecular and Vascular Biology, Katholieke Universiteit Leuven, Leuven, Belgium (J.-M.S.-R.); and Unité Propre de Recherche 9021, Centre National de la Recherche Scientifique, Institut de Biologie Moleculaire et Cellulaire, Strasbourg, France (J.H.).
Other authors were Joseph Reinbolt, Ph.D. (Unité Propre de Recherche 9002, Centre National de la Recherche Scientifique, IBMC, Strasbourg, France), and Dipika Mohanty, M.D., Ph.D. (Institute of Immunohaematology, Mumbai, India).
Address reprint requests to Dr. Lacroix-Desmazes at INSERM Unité 430, Hôpital Broussais, 96 rue Didot, Paris 75014, France, or at sebastien.lacroix{at}brs.ap-hop-paris.fr.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
