Acquired hemophilia is a rare hemorrhagic disorder caused by the spontaneous appearance of inhibitory autoantibodies directed against endogenous coagulation factor VIII (FVIII). Inhibitory Abs also arise in patients with congenital hemophilia A as alloantibodies directed to therapeutic FVIII. Both autoimmune and alloimmune inhibitors neutralize FVIII by steric hindrance. We have described FVIII-hydrolyzing IgG in 50% of inhibitor-positive patients with severe hemophilia A that inactivate therapeutic FVIII. In this study, we investigated the presence of autoimmune FVIII-hydrolyzing IgG in patients with acquired hemophilia. Pooled IgG from healthy donors demonstrated moderate FVIII-hydrolyzing activity (56 ± 26 μmol/min/mol). Purified IgG from 21 of 45 patients with acquired hemophilia demonstrated FVIII hydrolysis rates (mean 219 ± 94 μmol/min/mol) significantly greater than that of control IgG. Three of four patients followed over the course of the disease had rates of FVIII hydrolysis that co-evolved with inhibitory titers in plasma, suggesting that IgG-mediated FVIII hydrolysis participates, in part, in FVIII inactivation. The present work extends the scope of the diseases associated with FVIII proteolysis and points toward the importance of FVIII as a key target substrate for hydrolytic immunoglobulins. Our data suggest that elevated levels of FVIII-hydrolyzing IgG in acquired hemophilia result from the exacerbation of a physiological catalytic immune response.

Acquired hemophilia is a rare but life-threatening hemorrhagic disorder caused by the spontaneous appearance of autoantibodies directed against the endogenous plasma coagulation factor VIII (FVIII)3 (1). Anti-FVIII autoantibodies are detected by their ability to inhibit FVIII procoagulant activity in plasma, and are referred to as FVIII inhibitors. Inhibitory anti-FVIII autoantibodies develop in about one per million individuals each year (2, 3). In up to 50% of the cases, they occur spontaneously without any underlying pathological condition. In the remaining cases, they develop in association with autoimmune disorders or malignancies, as reactions to drugs, or during the postpartum period. Acquired hemophilia has a reported mortality between 6.2 and 44.3% 1 year following diagnosis (3, 4, 5).

FVIII inhibitors also develop as anti-FVIII alloantibodies in 10–30% of patients with congenital hemophilia A, following replacement therapy with exogenous FVIII (6). Alloimmune FVIII inhibitors neutralize FVIII activity in a passive manner by steric hindrance; by binding to FVIII, they prevent its interaction with thrombin, activated FIX, FX, phospholipids, or von Willebrand factor (VWF), thus preventing the protective effect of VWF on FVIII half-life, blocking the activation of FVIII by thrombin, or impeding the formation of the tenase complex (7). Based on shared epitope specificities between allo- and autoimmune FVIII-specific Abs, the above mechanisms have also been attributed to autoimmune FVIII inhibitors.

We have previously demonstrated the presence of FVIII-hydrolyzing IgG in the plasma of patients with congenital severe hemophilia A (8). The rate of FVIII hydrolysis correlates with the inhibitory activities scored in the plasma of the patients (9). Furthermore, the kinetics of IgG-mediated FVIII degradation are compatible with a pathogenic role for hydrolyzing Abs in inactivation of the therapeutically administered FVIII (10). We hypothesized that autoimmune FVIII inhibitors may also neutralize the procoagulant activity of endogenous FVIII by proteolysis. In this report, we investigated the presence of autoimmune FVIII-hydrolyzing IgG in a French cohort of 45 patients with acquired hemophilia. Autoimmune FVIII-hydrolyzing IgG were found in 46.6% of the patients. Interestingly, the rates of IgG-mediated hydrolysis co-evolve with the FVIII inhibitory titers scored in patients’ plasma during the course of the disease, suggesting that FVIII hydrolysis mediated by anti-FVIII autoantibodies contributes to FVIII inactivation in acquired hemophilia.

Frozen plasma samples from 45 patients with acquired hemophilia were obtained from Centre Hospitalier Universitaire (CHU) de Rouen, CHU de Caen, Hôpital Cochin, Hôpital du Kremlin-Bicêtre, and Hôpital de Rennes and CHU de Compiègne, in accordance with the local ethical regulation. This study is ancillary to “Etude Sacha” on patients with acquired hemophilia in France (CHU de Rouen) that includes 82 patients. Plasma samples were obtained at the time of diagnosis of acquired hemophilia. Criteria for inclusion of the patients in the study were a residual FVIII activity below 30%, an inhibitory titer above or equal to 1 Bethesda units (BU)/ml, a prolonged activated partial thrombosplatin time (aPTT), and normal levels of other factors of the intrinsic pathway and of VWF. The age at the time of diagnosis was 68.3 ± 17.6 years (mean ± SD, ranging from 25 to 92). Patients included 28 males (mean age: 70.3 ± 16.4 years) and 13 females (mean age: 64.2 ± 20.2 years) (not documented in the case of four patients). The inhibitory titer and residual FVIII activity in plasma at the time of diagnosis were 84.1 ± 174.0 BU/ml and 5.6 ± 6.2%, respectively (Table I). The diseases underlying or existing before the development of anti-FVIII autoantibodies were documented in the case of 36 of 45 patients included in the study (Table I): allergic drug reactions (5/36, 13.9%), autoimmune disorders (2/36, 5.6%), malignancies (8/36, 22.2%), autoimmune disorders associated with malignancies (2/36, 5.6%), dermatologic disorders (3/36, 8.6%), diabetes (1/36, 2.8%), autoimmune diseases associated with diabetes (1/36, 2.8%), postpartum (2/36, 5.6%), and idiopathic (12/36, 33.3%). The survival 1 year after diagnosis, documented for 35 patients, was 60%. The clinical characteristics of the 45 patients included in the present study, 23 of whom are included in Etude Sacha, were similar to that of the 82 patients of Etude Sacha (unpublished data).

Table I.

Clinical characteristics of patients with acquired hemophilia

Characteristics (number of patients documented)aNumber (percent or range) of patients reported
Male/Female (n = 41) 28/13 
Age (n = 34) 68.3 ± 17.6 years (25–92 years) 
aPTT (n = 22)b 2.35 ± 0.66 (1.6–4.2) 
FVIII:c in plasma (n = 30)c 5.6 ± 6.2 (<1–30%) 
Inhibitory titer (n = 45)d 84.1 ± 174.0 BU/ml (1–1050 BU/ml) 
Prior or preexisting conditions (n = 36)  
 Allergic drug reactions 5 (13.9) 
 Autoimmune disorders (AiD) 2 (5.6) 
 Malignancies 8 (22.2) 
 AiD associated with malignancies 2 (5.6) 
 Dermatologic disorders 3 (8.6) 
 Diabetes 1 (2.8) 
 AiD associated with diabetes 1 (2.8) 
 Postpartum 2 (5.6) 
 Idiopathic 12 (33.3) 
Survival (n = 35)e 21 (60%) 
Characteristics (number of patients documented)aNumber (percent or range) of patients reported
Male/Female (n = 41) 28/13 
Age (n = 34) 68.3 ± 17.6 years (25–92 years) 
aPTT (n = 22)b 2.35 ± 0.66 (1.6–4.2) 
FVIII:c in plasma (n = 30)c 5.6 ± 6.2 (<1–30%) 
Inhibitory titer (n = 45)d 84.1 ± 174.0 BU/ml (1–1050 BU/ml) 
Prior or preexisting conditions (n = 36)  
 Allergic drug reactions 5 (13.9) 
 Autoimmune disorders (AiD) 2 (5.6) 
 Malignancies 8 (22.2) 
 AiD associated with malignancies 2 (5.6) 
 Dermatologic disorders 3 (8.6) 
 Diabetes 1 (2.8) 
 AiD associated with diabetes 1 (2.8) 
 Postpartum 2 (5.6) 
 Idiopathic 12 (33.3) 
Survival (n = 35)e 21 (60%) 
a

Depending on the clinical characteristic, data were received in the case of 22–45 of the 45 patients included in the study.

b

aPTT, Activated partial thromboplastin time, expressed as the ratio of the measured value over that of the physiological value.

c

FVIII:c, Factor VIII activity in plasma.

d

Inhibitory titers were assessed using the modified Bethesda assay.

e

Survival scored within the first year after diagnosis.

In the case of four patients, blood samples were obtained sequentially over periods of 6–140 days (4–7 samples per patient). For these four patients, the underlying treatments included corticotherapy, therapeutic FVIII, recombinant activated FVII, and/or intravenous Ig (IVIg).

FVIII-inhibitory activity was measured in plasma using the modified method of Kasper et al., (11) and expressed in BU/ml. Plasma was heated 1 h at 56°C before testing. Heated plasma was incubated with an equal volume of pooled citrated human plasma (Dade-Behring) for 2 h at 37°C. Residual FVIII activity was measured in a 1-stage clotting assay, as described. The detection limit of the assay was 0.3 BU.

IgG was isolated from plasma by affinity-chromatography on protein G Sepharose (Amersham Biosciences). A therapeutic preparation of pooled normal human Ig (IVIg; Sandoglobulin) was used as a source of normal IgG. To exclude potentially contaminating proteases, size-exclusion chromatography of patients’ IgG and IVIg was performed on a superose-12 column (Amersham Biosciences), equilibrated with 50 mM Tris, 8 M urea, and 0.02% NaN3 (pH 7.7), at a flow rate of 250 μl/min. IgG-containing fractions were pooled and dialyzed against PBS-0.01% NaN3 for 48 h at 4°C, followed by dialysis against 50 mM Tris (pH 7.7), 100 mM glycine, 0.02% NaN3, and 5 mM CaCl2 (catalytic buffer) for 24 h at 4°C. We have previously demonstrated that urea-treated purified IgG retain the inhibitory activity toward FVIII (9). The purity of IgG preparations was assessed by (i) SDS-PAGE and immunoblotting under non-reducing conditions and by (ii) matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) analysis of trypsin digests of the IgG preparations. IgG was quantified by OD measurements at 280 nm.

Commercially available recombinant human FVIII (Kogenate FS; Bayer HealthCare), 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. FVIII was reconstituted in distilled water to a final concentration of 600 μg/ml, desalted by dialyzing against 100 mM borate (pH 7.0), 150 mM NaCl, and 5 mM CaCl2 (Borate buffer). Sulfo-NHS-LC-biotin (250 μg) was allowed to react with FVIII (600 μg) for 2 h at 4°C. Biotinylated FVIII was dialyzed against catalytic buffer for 3 h at 4°C, aliquoted, and stored at −20°C until use. The protocol was essentially identical for the biotinylation of human serum albumin (HSA; Laboratoire Français du Fractionnement et des Biotechnologies), activated FVII (FVIIa; NovoSeven), and prothrombin (Kordia). Biotinylated FVIII, HSA, FVIIa, and prothrombin (185 nM) were incubated in catalytic buffer with IgG (10 μg/ml, 67 nM) for 24 h at 37°C. Samples were mixed with Laemmli’s buffer without 2-ME (1:1 v/v), and 20 μl of each sample was subjected to 10% SDS-PAGE. Protein fragments were then transferred onto nitrocellulose membrane (Schleicher & Schuell Microscience). Following overnight blocking in TBS, 0.2% Tween 20 at 4°C, membranes were incubated with streptavidin-coupled alkaline phosphatase (Kirkegaard & Perry Laboratories) diluted 1/4000 in blocking buffer, for 60 min at room temperature. After washing in TBS containing 0.1% Tween 20, labeled proteins were revealed using the BCIP/NBT kit.

High-resolution images were acquired by scanning the immunoblots using a SnapScan 600 scanner (Agfa). Black and white images were converted to negatives using the Adobe Photoshop CS2 (Version 9.0.2) software. A macro was written using the National Institutes of Health image 1.62b7 software (OD macro; Heudes D & Nicoletti A) to calculate mean image densities. Briefly, the negative images were imported into the National Institutes of Health image 1.62b7 software using the OD macro. The negative images were converted back to the positive mode by applying an arithmetic logarithmic (log) process. The “log process” does not affect the image pixels. For calculating the rates of hydrolysis of FVIII, we measured, for each sample, the mean density i) of the total area of the lane of the FVIII migration profile and ii) of the area of the protein bands with molecular weights below 75 kDa (i.e., the L chain of unprocessed FVIII). The percentage of FVIII hydrolyzed was calculated as the ratio of the mean density of the hydrolyzed area over the mean density of the total area of the lane. This approach yielded a linear correlation between the calculated mean OD and the amount of FVIII loaded on the gels for a range of 23–741 nM of biotinylated FVIII (Spearman rank correlation; p = 0.02, Rho = 1, R2 = 0.97).

Spontaneous hydrolysis occurring upon incubation of FVIII in the presence of buffer alone was considered to represent the background level and was subtracted from each analysis. The data was expressed as μmol of FVIII protein hydrolyzed per minute per mol of IgG. Significant differences between the rates of FVIII hydrolysis of patients’ IgG and that of IVIg (used as a control), were assessed using an ANOVA post-hoc test (Fisher’s protected least significant difference). The reported p values are one-sided.

Plasma from 45 patients was collected at the time of diagnosis of acquired hemophilia. The mean age of the patients at the time of inclusion was 68.3 ± 17.6 years (ranging from 25 to 92 years, Table I). The mean inhibitory activity against FVIII in plasma was 84.1 ± 174.0 BU/ml (range, 1–1050 BU/ml, Table I). All patients had anti-FVIII IgG, as assessed by enzyme-linked immunosorbant assay (data not shown). Sixty percent of the patients were still alive 1 year after diagnosis. Reminiscent of previous reports (3, 4, 5), we found no significant correlation between the residual FVIII activity, aPTT, and inhibitory titers, measured in plasma at the time of diagnosis (p ≥ 0.08 in all cases). In addition, none of these clinical parameters exhibited significant correlations with the survival of the patients 1 year following diagnosis (p = 0.06, 0.70, and 0.82, respectively).

IgG was purified from the plasma of the 45 patients. We investigated the capacity of purified IgG to hydrolyze human recombinant FVIII. The hydrolytic patterns of biotinylated FVIII incubated in buffer alone or with the purified IgG are depicted in Fig. 1,A, in the case of IgG from 10 patients. FVIII exhibited a characteristic electrophoretic pattern, with major protein bands migrating at molecular masses between 90 and 337 kDa and an additional band at 82 kDa. Incubation of FVIII with purified IgG (67 nM) from several patients (e.g., patients 1, 8, 15, and 17) resulted in the hydrolysis of FVIII. In contrast, the migration profile of FVIII remained unchanged when it was incubated in buffer alone (Ctl). The migration profiles of hydrolyzed FVIII were subjected to densitometric analysis so as to compute the specific rates of FVIII hydrolysis in the case of IgG from the 45 patients and of IVIg (Fig. 2 and Table II). IVIg exhibited a hydrolytic activity of 55.6 ± 25.9 μmol/min/mol toward FVIII, a value statistically indistinguishable from the mean rate of FVIII hydrolysis measured in the case of IgG from nine healthy blood donors (54.9 ± 22.5 μmol/min/mol, data not shown). Purified IgG from 21 of the 45 patients (46.6%) included in the study (i.e., patients 1, 2, 5, 8, 10, 13, 14, 15, 16, 17, 21, 22, 23, 29, 32, 35, 36, 37, 41, 43, and 45) exhibited a FVIII-hydrolyzing activity that was significantly higher than that of IVIg (Table II, p < 0.05). The mean hydrolytic activity of IgG of these 21 patients was 218.8 ± 93.9 μmol/min/mol (coefficient of variation: 0.29 ± 0.20).

FIGURE 1.

Hydrolysis of FVIII by IgG purified from the plasma of patients with acquired hemophilia. A, Biotinylated recombinant human FVIII (185 nM, in the case of 10 patients), (B) biotinylated HSA (185 nM, in the case of 3 patients), (C) biotinylated recombinant human activated FVII (FVIIa, 185 nM, in the case of 3 patients), and (D) biotinylated human prothrombin (185 nM, in the case of 3 patients) were incubated alone (Ctl) or in the presence of IgG (67 nM) for 24 h at 37°C. IVIg was used as a source of normal IgG and as a negative control. Samples were subjected to 10% SDS-PAGE and transferred onto a nitrocellulose membrane, before revelation of biotinylated fragments.

FIGURE 1.

Hydrolysis of FVIII by IgG purified from the plasma of patients with acquired hemophilia. A, Biotinylated recombinant human FVIII (185 nM, in the case of 10 patients), (B) biotinylated HSA (185 nM, in the case of 3 patients), (C) biotinylated recombinant human activated FVII (FVIIa, 185 nM, in the case of 3 patients), and (D) biotinylated human prothrombin (185 nM, in the case of 3 patients) were incubated alone (Ctl) or in the presence of IgG (67 nM) for 24 h at 37°C. IVIg was used as a source of normal IgG and as a negative control. Samples were subjected to 10% SDS-PAGE and transferred onto a nitrocellulose membrane, before revelation of biotinylated fragments.

Close modal
FIGURE 2.

Rates of IgG-mediated hydrolysis of FVIII. Electrophoretic profiles of digestion of FVIII by purified IgG from patients with acquired hemophilia were subjected to densitometric analysis. Spontaneous hydrolysis occurring upon incubation of FVIII in the presence of buffer alone was considered to represent the background level and was subtracted from each analysis. The figure depicts the mean hydrolytic rates (±SD) from four independent measurements. Significant differences between the rates of FVIII hydrolysis of patients’ IgG and that of IVIg (used as a control) were assessed using an ANOVA post-hoc test. The reported p values are one-sided (*, p < 0.05).

FIGURE 2.

Rates of IgG-mediated hydrolysis of FVIII. Electrophoretic profiles of digestion of FVIII by purified IgG from patients with acquired hemophilia were subjected to densitometric analysis. Spontaneous hydrolysis occurring upon incubation of FVIII in the presence of buffer alone was considered to represent the background level and was subtracted from each analysis. The figure depicts the mean hydrolytic rates (±SD) from four independent measurements. Significant differences between the rates of FVIII hydrolysis of patients’ IgG and that of IVIg (used as a control) were assessed using an ANOVA post-hoc test. The reported p values are one-sided (*, p < 0.05).

Close modal
Table II.

Specific rates of hydrolysis of FVIII by IgG purified from the plasma of 45 patients with acquired hemophilia

Patient No.Inhibitory Titer (BU/ml)aHydrolysis of FVIIIb (μmol/min/mol)Associated ConditioncSurvivald
40 449.7 ± 211.6e Malignancy 
63 154.3 ± 41.0e Idiopathic 
128 144.7 ± 85.5 Malignancy 
114 127.9 ± 50.6 Idiopathic 
380 168.3 ± 19.7e nd nd 
32 123.8 ± 54.4 AiD, Malignancy 
49.4 82.0 ± 43.7 nd nd 
3.1 381.6 ± 192.0e Allergic drug reaction 
42 132.3 ± 18.6 Malignancy 
10 172.1 ± 30.3e Idiopathic 
11 14 75.8 ± 13.6 Idiopathic 
12 119.8 ± 76.9 AiD 
13 40 158.2 ± 67.5e Postpartum 
14 222.6 ± 10.6e Allergic drug reaction 
15 80 178.3 ± 16.2e Idiopathic 
16 52 137.0 ± 27.2e Malignancy 
17 100 356.5 ± 69.2e AiD, Malignancy 
18 10 92.6 ± 25.9 Dermatologic disorder 
19 1.4 96.4 ± 40.5 Idiopathic 
20 18 82.0 ± 18.7 Malignancy 
21 1050 147.4 ± 99.4e nd nd 
22 330 136.2 ± 73.3e Idiopathic 
23 18 302.3 ± 61.5e Idiopathic 
24 1.3 48.0 ± 19.5 Idiopathic 
25 60 62.0 ± 13.2 Malignancy 
26 57.7 ± 34.2 AiD, Diabetes 
27 97.4 ± 58.6 Idiopathic 
28 56 68.3 ± 9.1 Idiopathic nd 
29 4.5 134.5 ± 26.0e Dermatologic disorder 
30 1.5 75.6 ± 18.4 Diabetes 
31 107.7 ± 98.6 Allergic drug reaction 
32 194.7 ± 20.8e Allergic drug reaction 
33 362.7 82.8 ± 35.5 AiD 
34 249.2 94.2 ± 68.7 nd nd 
35 62 317.6 ± 38.8e Allergic drug reaction 
36 54 212.5 ± 56.8e Malignancy 
37 128.7 ± 63.5e nd nd 
38 105.4 104.5 ± 18.8 nd nd 
39 92.5 91.0 ± 49.1 nd nd 
40 26.6 96.2 ± 70.2 Postpartum 
41 10 168.1 ± 16.8e Malignancy 
42 29.9 104.6 ± 37.9 nd nd 
43 4.9 303.8 ± 211.3e Idiopathic 
44 13 101.5 ± 47.8 Dermatologic disorder 
45 58.2 169.7 ± 31.3e nd nd 
Ig 55.6 ± 25.9 
Patient No.Inhibitory Titer (BU/ml)aHydrolysis of FVIIIb (μmol/min/mol)Associated ConditioncSurvivald
40 449.7 ± 211.6e Malignancy 
63 154.3 ± 41.0e Idiopathic 
128 144.7 ± 85.5 Malignancy 
114 127.9 ± 50.6 Idiopathic 
380 168.3 ± 19.7e nd nd 
32 123.8 ± 54.4 AiD, Malignancy 
49.4 82.0 ± 43.7 nd nd 
3.1 381.6 ± 192.0e Allergic drug reaction 
42 132.3 ± 18.6 Malignancy 
10 172.1 ± 30.3e Idiopathic 
11 14 75.8 ± 13.6 Idiopathic 
12 119.8 ± 76.9 AiD 
13 40 158.2 ± 67.5e Postpartum 
14 222.6 ± 10.6e Allergic drug reaction 
15 80 178.3 ± 16.2e Idiopathic 
16 52 137.0 ± 27.2e Malignancy 
17 100 356.5 ± 69.2e AiD, Malignancy 
18 10 92.6 ± 25.9 Dermatologic disorder 
19 1.4 96.4 ± 40.5 Idiopathic 
20 18 82.0 ± 18.7 Malignancy 
21 1050 147.4 ± 99.4e nd nd 
22 330 136.2 ± 73.3e Idiopathic 
23 18 302.3 ± 61.5e Idiopathic 
24 1.3 48.0 ± 19.5 Idiopathic 
25 60 62.0 ± 13.2 Malignancy 
26 57.7 ± 34.2 AiD, Diabetes 
27 97.4 ± 58.6 Idiopathic 
28 56 68.3 ± 9.1 Idiopathic nd 
29 4.5 134.5 ± 26.0e Dermatologic disorder 
30 1.5 75.6 ± 18.4 Diabetes 
31 107.7 ± 98.6 Allergic drug reaction 
32 194.7 ± 20.8e Allergic drug reaction 
33 362.7 82.8 ± 35.5 AiD 
34 249.2 94.2 ± 68.7 nd nd 
35 62 317.6 ± 38.8e Allergic drug reaction 
36 54 212.5 ± 56.8e Malignancy 
37 128.7 ± 63.5e nd nd 
38 105.4 104.5 ± 18.8 nd nd 
39 92.5 91.0 ± 49.1 nd nd 
40 26.6 96.2 ± 70.2 Postpartum 
41 10 168.1 ± 16.8e Malignancy 
42 29.9 104.6 ± 37.9 nd nd 
43 4.9 303.8 ± 211.3e Idiopathic 
44 13 101.5 ± 47.8 Dermatologic disorder 
45 58.2 169.7 ± 31.3e nd nd 
Ig 55.6 ± 25.9 
a

Inhibitory titers were measured in the plasma using the Bethesda assay.

b

The results are mean ± SD of four independent measurements. Rates of FVIII hydrolysis were calculated by densitometric analysis of Western blots. Spontaneous hydrolysis that occurred upon incubation of FVIII in the presence of buffer alone was subtracted from each analysis. The mean coefficient of variation was 0.37 ± 0.21 (range: 0.05–0.92).

c

Depicts for each patient the disorder associated with the development of FVIII inhibitors. AiD, Autoimmune disorder; and nd, not documented.

d

Survival status within the first year after diagnosis. A, Alive; D, deceased; and nd, not documented.

e

p < 0.05 for the comparison with IVIg, using an analysis of variance post hoc test.

We have previously documented that FVIII-hydrolyzing IgG behave as serine proteases in the absence of contamination by adventitious proteases (8). In this study, the absence of contaminating proteases was confirmed by MALDI-TOF analysis of trypsin digests of the IgG preparations (data not shown). rHSA, which presents with a single electrophoretic band of 65 kDa (Fig. 1,B), was not hydrolyzed when incubated with IgG of patients 14 and 15 (elevated rates of hydrolysis of FVIII) and with IgG of patient 20 (marginal hydrolysis of FVIII). Similarly, recombinant human FVIIa (50 kDa, Fig. 1,C) and human prothrombin (72 kDa, Fig. 1 D) were not hydrolyzed by the IgG of patients 1, 8, and 17 (elevated rates of hydrolysis of FVIII), thus indicating the specificity of hydrolysis toward FVIII and the absence of adventitious contaminating non-IgG proteases.

No significant correlation was found between the levels of FVIII-hydrolyzing IgG with either the age of the patients, their survival 1 year after diagnosis, or the FVIII inhibitory activity in plasma (data not shown). Patients with varying underlying disorders presented with different levels of FVIII-hydrolyzing IgG (Table II): 186.9 ± 123.2 μmol/min/mol (mean ± SEM) in the case of IgG from 10 patients with malignancies, 148.1 ± 119.6 for 5 patients with autoimmune diseases, 127.1 ± 43.8 for 2 postpartum patients, 244.8 ± 107.0 for 5 drug-related cases, 109.5 ± 22.1 for 3 patients with dermatologic diseases, and 146.7 ± 83.6 for 12 idiopathic patients. Thus, levels of FVIII-hydrolyzing IgG were 2-fold lower in patients with dermatologic diseases than in patients who had developed FVIII inhibitors in the context of malignancies or as a reaction to drugs (statistically nonsignificant and p = 0.052, respectively).

In the case of four patients (patients 10, 11, 32, and 35), plasma samples were collected at the time of diagnosis and during the course of the disease for periods of time ranging from 6–140 days. At the time of diagnosis, inhibitory titers in plasma were 6, 12, 7, and 24 BU/ml, respectively (Table II). At the time of the last plasma sampling, all patients had detectable inhibitory titers of 0.7, 3.5, 1.6, and 19 BU/ml, respectively (Fig. 3). In the case of 3 patients, inhibitory titers had decreased by 3.4- to 8.6-folds over the 24, 26, and 6 days of follow-up, respectively. In the case of patient 35, who was followed for a period of time of 140 days, the inhibitory titer varied constantly between a minimum of 19 BU/ml and a maximum of 62 BU/ml, and demonstrated an overall 1.3-fold reduction over the period.

FIGURE 3.

Follow-up of the inhibitory and hydrolytic activities of patients’ IgG. Plasma samples were collected from four patients with acquired hemophilia during periods of time comprised between 6 and 140 days. A total of 4–7 samples were collected for each patient. For each sample, the inhibitory titer toward FVIII was measured in plasma (BU/ml, •). IgG was purified from each plasma sample and assessed for FVIII hydrolysis (μmol/min/mol, ○). The graphs depict for each of the four patients the evolution with time of the inhibitory titer (left y-axis) and of the rates of IgG-mediated FVIII hydrolysis (right y-axis). The data for FVIII hydrolysis represent the means and SDs of four separate experiments.

FIGURE 3.

Follow-up of the inhibitory and hydrolytic activities of patients’ IgG. Plasma samples were collected from four patients with acquired hemophilia during periods of time comprised between 6 and 140 days. A total of 4–7 samples were collected for each patient. For each sample, the inhibitory titer toward FVIII was measured in plasma (BU/ml, •). IgG was purified from each plasma sample and assessed for FVIII hydrolysis (μmol/min/mol, ○). The graphs depict for each of the four patients the evolution with time of the inhibitory titer (left y-axis) and of the rates of IgG-mediated FVIII hydrolysis (right y-axis). The data for FVIII hydrolysis represent the means and SDs of four separate experiments.

Close modal

IgG was purified from each of the samples and assessed for FVIII hydrolyzing activity. In the case of patients 10, 11, and 32, the hydrolytic activity of IgG had decreased by 1.4 ± 0.2-folds, between the time of diagnosis and the time of the last plasma sampling (Table II and Fig. 3). IgG from patient 35 demonstrated a significant overall decrease of 2.1-folds of the rate of IgG-mediated FVIII hydrolysis (p < 0.05). However, rates of FVIII hydrolysis by IgG from patient 35 evolved inversely to the inhibitory titers measured in plasma (Rho = −0.5; and R = 0.48). In contrast, FVIII hydrolysis rates were found to co-evolve with the inhibitory titers in the case of IgG from patients 10, 11, and 32 (Rho values equal to 1.0, 0.8, and 0.2; and R = 0.97, 0.65, and 0.36, respectively, as calculated using the Spearman rank correlation test, Fig. 3).

Human Ab catalysts have been reported in a large number of pathologies, including multiple myeloma, asthma, Hashimoto’s thyroiditis, systemic lupus erythematosus, scleroderma, rheumatoid arthritis and multiple sclerosis, and HIV-related immune thrombocytopenia (12, 13, 14, 15, 16, 17, 18, 19, 20). Target Ags for disease-associated hydrolytic Abs include prothrombin, vasoactive intestinal peptide, thyroglobulin, DNA, RNA, and the anti-platelet integrin GPIIIa(β3). We have previously described the presence of FVIII-hydrolyzing IgG in two instances: in the course of iatrogenic alloimmunization against therapeutic human FVIII in patients with severe hemophilia A, and during acute inflammation in some patients with severe sepsis (9, 21). The present description of FVIII-hydrolyzing IgG in patients with acquired hemophilia extends the scope of the diseases associated with FVIII proteolysis. Acquired hemophilia may be idiopathic or associated with a plethora of underlying inflammatory, neoplastic, or autoimmune conditions (5, 22, 23). Indeed, in our study cohort, while development of FVIII inhibitors was idiopathic or associated with pregnancy in roughly 40% of the cases, underlying diseases in the other cases were allergic drug reactions, malignancies, dermatologic disorders, diabetes, autoimmune disorders, and autoimmune disorders associated with malignancies or diabetes. Probably owing to the restricted number of patients in each of the different disease groups included in our study, the occurrence of FVIII-hydrolyzing IgG among patients with acquired hemophilia could not be attributed to a particular underlying pathology. Nevertheless, our results point toward the importance of FVIII as a key target substrate for hydrolytic immunoglobulins, and raise the question of the physiopathological relevance of FVIII-hydrolyzing IgG. By limiting the extent of coagulation, FVIII-hydrolyzing IgG may be beneficial in cases of disseminated vascular thrombosis (21). They may, however, be detrimental in situations where normal hemostasis is hampered by a congenital or acquired deficit of coagulation factors.

Abs to FVIII emerge as alloantibodies in patients with hemophilia A following treatment with exogenous FVIII, or develop spontaneously as autoantibodies against endogenous FVIII in patients with acquired hemophilia (6). Autoantibodies and alloantibodies against FVIII share similar properties with respect to their isotype distribution, epitope specificities, and mechanisms of inhibition of FVIII activity. They may, however, differ in their kinetics of FVIII inactivation (24, 25, 26, 27). We have documented earlier that anti-FVIII alloantibodies, from ∼50% of inhibitor-positive patients with severe hemophilia A, hydrolyze FVIII. The observed correlation between the rate of FVIII hydrolysis and the inhibitory activity measured in plasma (9), as well as the kinetics of IgG-mediated FVIII degradation (10), were compatible with a deleterious role of FVIII-hydrolyzing IgG in the patients. In this study, we demonstrate the presence of FVIII-hydrolyzing IgG autoantibodies in the plasma of 47% of patients with acquired hemophilia, thus demonstrating an elevated prevalence of FVIII-hydrolyzing IgG in acquired hemophilia, similar to that found in patients with congenital hemophilia A. The longitudinal study performed in the case of four patients, for the rates of FVIII hydrolysis and Bethesda titers in plasma, suggested that the catalytic activity evolves with the inhibitory titer following similar trends in the case of three patients. Together, the data suggest that high levels of FVIII-hydrolyzing IgG occur as a result of the disease and accompany the evolution of the pathological anti-FVIII immune response. In contrast to our findings in congenital hemophilia A (9), no correlation was observed between the scored rates of hydrolysis of FVIII (catalytic activity) at the time of diagnosis and the Bethesda titers (inhibitory activity) measured in plasma. The discrepancy between these two sets of data may be due to differences in the relative importance of the various mechanisms of FVIII inhibition that coexist in different patients (28, 29, 30, 31, 32, 33), i.e., non-proteolytic FVIII inactivation by steric hindrance vs Ab catalyst-mediated FVIII hydrolysis. Alternatively, the difference between the two studies may relate to the different immunological contexts that characterize the development of the anti-FVIII immune response. Indeed, while the development of FVIII inhibitors in patients with congenital hemophilia A is seen as an alloimmune response against an exogenous therapeutic Ag, the occurrence of anti-FVIII autoantibodies is the result of a break of tolerance toward an endogenous protein.

The origin of Abs with catalytic activity has been the centre of intense speculation. Seminal works in the field of Ab catalysis have shown that Abs with germline configuration are endowed with catalytic functions as diverse as oxy-cope rearrangement (34), Diels-Alder reaction (35, 36), or peptide hydrolysis (37, 38), thus demonstrating that the catalytic potential of immunoglobulins is encoded in the genome. The observation of amidase activity performed by circulating IgG, IgM, and IgA from healthy donors (39, 40), and our finding of baseline levels of FVIII hydrolysis by normal immunoglobulins, indicates that the repertoire of immunoglobulins endowed with catalytic functions is expressed under physiological conditions. A role for catalytic Abs under physiological conditions in the removal of metabolic wastes has been proposed (41). Accordingly, IgG-mediated hydrolytic inactivation of FVIII may be a mechanism of control of the circulating levels of FVIII under physiological conditions.

The expression of the repertoire of catalytic Abs is controlled under physiological conditions. Indeed, MRL/lpr mice immunized with transition-state analogues of an esterase reaction have been shown to make higher numbers of clones expressing catalytic Abs as compared with control BALB/c or wild type MRL/++ mice (42). MRL/lpr mice are deficient in Fas-mediated apoptosis and do not delete self-reactive B cell clones as efficiently as wild-type animals. It remains to be investigated whether, in patients with acquired hemophilia, the exacerbation of the catalytic anti-FVIII autoimmune response results from a breakdown in physiological regulatory mechanisms, followed by the recruitment of natural FVIII-hydrolyzing IgG-producing B cell clones, or from the de novo selection of catalytic IgG-producing B cells.

We have previously documented that the proteolytic activity of anti-FVIII IgG is mediated by the F(ab′)2 of the Abs, that purified IgG is free from contaminating proteases, and that the hydrolytic activity of Abs is not inhibited by a broad range of generic protease inhibitors. In the present study, the purification of IgG on protein G-Sepharose immediately followed by size-exclusion chromatography in 8 M urea further excluded the possibility that FVIII hydrolysis was mediated by contaminating proteases. Accordingly, the absence of contaminating proteases within IgG preparations was confirmed by MALDI-TOF analysis, and by the specificity of IgG-mediated hydrolysis for FVIII and not for albumin, prothrombin, and FVIIa.

The clinical characteristics of the patients included in our study are essentially similar to that described in the case of a recently published English study encompassing 172 patients (3). The survival rate in our cohort (60%) within the first year following diagnosis was similar to that of the English cohort (55.7%). Because of the heterogeneous nature of the underlying conditions that are associated with acquired hemophilia, no single diagnostic parameter for acquired hemophilia has been identified as yet. Indeed, diagnosis of acquired hemophilia still relies on low levels of circulating FVIII, on the presence of inhibitory anti-FVIII Abs, and on a prolonged aPTT, together with normal levels of other factors of the intrinsic pathway and of VWF. More importantly, no prognostic parameter for the outcome of acquired hemophilia has ever been identified; at the time of diagnosis, neither the inhibitory titer, the aPTT, the FVIII coagulant activity in plasma, nor the underlying disease are associated with the survival or mortality of the patients (3, 5, 22, 23). Accordingly, in our study, there was no association between either one of the latter clinical parameters or the presence of FVIII-hydrolyzing IgG measured at the time of diagnosis, and the survival/death of the patients 1 year following inclusion. These observations raise the question of the relevance of FVIII-related clinical parameters as pertinent markers for disease outcome.

We are indebted to Dr. Roseline d’Oiron (Centre des hémophiles, Hôpital Kremlin-Bicêtre, Bicêtre, France), Dr. Valérie Gay (Centre Hospitalier général service hémophilie, Chambéry, France), Dr. Luc Darnige (Service d’hémotologie, Hôpital Européen Georges-Pompidou, Paris), and Drs. Natalie Stieltjes and Valérie Roussel-Robert (Centre des hémophiles, Hôpital Cochin, Paris, France) for providing us with plasma samples, and to Prof. Sudhir Paul (University of Texas, Houston Medical School, Houston, TX) for providing us with biotinylated covalently reactive analog. We are grateful to Dr Zéra Tellier (LFB) for constant support and to Dr. Françoise Truong (Novonordisk) for assistance with collection of patients’ data. Human recombinant FVIII and plasma-derived HSA were gifts from Bayer Healthcare and LFB, respectively.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Institut National de la Santé et de la Recherche Médicale, by Centre National de la Recherche Scientifique, by Indo-French Center for Promotion of Advanced Research, by Novonordisk (La Defense, France), and by a Grant from Agence Nationale de la Recherche (ANR-05-MRAR-012). B.W., S.D., and J.D.D. are the recipients of fellowships from Laboratoire Français de Fractionnement et des Biotechnologies (LFB, Les Ulis, France), from Fédération de la Recherche Médicale, and from Institut National de la Santé et de la Recherche Médicale postes-verts, respectively.

3

Abbreviations used in this paper: FVIII, factor VIII; IVIg, intravenous immunoglobulin; HSA, human plasma-derived albumin; VWF, von Willebrand factor; aPTT, activated partial thrombosplatin time; BU, Bethesda units; MALDI-TOF, matrix-assisted laser desorption/ionization-time-of-flight; CHU, Centre Hospitalier Universitaire.

1
Kessler, C. M..
2000
. Acquired factor VIII autoantibody inhibitors: current concepts and potential therapeutic strategies for the future.
Haematologica
85
:
57
-61.
2
Franchini, M., G. Gandini, T. Di Paolantonio, G. Mariani.
2005
. Acquired hemophilia A: a concise review.
Am. J. Hematol.
80
:
55
-63.
3
Collins, P. W., S. Hirsch, T. P. Baglin, G. Dolan, J. Hanley, M. Makris, D. M. Keeling, R. Liesner, S. A. Brown, C. R. Hay.
2007
. Acquired hemophilia A in the United Kingdom: a 2-year national surveillance study by the United Kingdom Haemophilia Centre Doctors’ Organisation.
Blood
109
:
1870
-1877.
4
Green, D..
1991
. Cytotoxic suppression of acquired factor VIII: C inhibitors.
Am. J. Med.
91
:
14S
-19S.
5
Morrison, A. E., C. A. Ludlam, C. Kessler.
1993
. Use of porcine factor VIII in the treatment of patients with acquired hemophilia.
Blood
81
:
1513
-1520.
6
Mannucci, P. M., E. G. Tuddenham.
2001
. The hemophilias: from royal genes to gene therapy.
N. Engl. J. Med.
344
:
1773
-1779.
7
Lavigne-Lissalde, G., J. F. Schved, C. Granier, S. Villard.
2005
. Anti-factor VIII antibodies: a 2005 update.
Thromb. Haemostasis
94
:
760
-769.
8
Lacroix-Desmazes, S., A. Moreau Sooryanarayana, C. Bonnemain, N. Stieltjes, A. Pashov, Y. Sultan, J. Hoebeke, M. D. Kazatchkine, S. V. Kaveri.
1999
. Catalytic activity of antibodies against factor VIII in patients with hemophilia A.
Nat. Med.
5
:
1044
-1047.
9
Lacroix-Desmazes, S., J. Bayry, N. Misra, M. P. Horn, S. Villard, A. Pashov, N. Stieltjes, R. d'Oiron, J. M. Saint-Remy, J. Hoebeke, et al
2002
. The prevalence of proteolytic antibodies against factor VIII in hemophilia A.
N. Engl. J. Med.
346
:
662
-667.
10
Lacroix-Desmazes, S., B. Wootla, S. Dasgupta, S. Delignat, J. Bayry, J. Reinbolt, J. Hoebeke, E. Saenko, M. D. Kazatchkine, A. Friboulet, et al
2006
. Catalytic IgG from patients with hemophilia A inactivate therapeutic factor VIII.
J. Immunol.
177
:
1355
-1363.
11
Kasper, C. K., L. M. Aledort, R. B. Counts.
1975
. A more uniform measurement of factor VIII inhibitors.
Thromb. Diath. Haemorrh.
34
:
869
-872.
12
Paul, S., L. Li, R. Kalaga, P. Wilkins-Stevens, F. J. Stevens, A. Solomon.
1995
. Natural catalytic antibodies: peptide-hydrolyzing activities of Bence Jones proteins and VL fragments.
J. Biol. Chem.
270
:
15257
-15261.
13
Li, L., S. Paul, S. Tyutyulkova, M. D. Kazatchkine, S. Kaveri.
1995
. Catalytic activity of anti-thyroglobulin antibodies.
J. Immunol.
154
:
3328
-3332.
14
Paul, S., L. Lan, R. Kalaga, J. O'Dell, R. E. Dannenbring, Jr, S. Swindells, S. Hinrichs, P. Caturegli, N. R. Rose.
1997
. Characterization of thyroglobulin-directed and polyreactive catalytic antibodies in autoimmune disease.
J. Immunol.
159
:
1530
-1536.
15
Kozyr, A. V., A. V. Kolesnikov, N. A. Zelenova, L. P. Sashchenko, S. V. Mikhalap, M. E. Bulina, A. N. Ignatova, P. V. Favorov, A. G. Gabibov.
2000
. Autoantibodies to nuclear antigens: correlation between cytotoxicity and DNA-hydrolyzing activity.
Appl. Biochem. Biotechnol.
83
:
255
-268.
16
Baranovskii, A. G., N. A. Ershova, V. N. Buneva, T. G. Kanyshkova, A. S. Mogelnitskii, B. M. Doronin, A. N. Boiko, E. I. Gusev, O. O. Favorova, G. A. Nevinsky.
2001
. Catalytic heterogeneity of polyclonal DNA-hydrolyzing antibodies from the sera of patients with multiple sclerosis.
Immunol. Lett.
76
:
163
-167.
17
Saveliev, A. N., D. R. Ivanen, A. A. Kulminskaya, N. A. Ershova, T. G. Kanyshkova, V. N. Buneva, A. S. Mogelnitskii, B. M. Doronin, O. O. Favorova, G. A. Nevinsky, K. N. Neustroev.
2003
. Amylolytic activity of IgM and IgG antibodies from patients with multiple sclerosis.
Immunol. Lett.
86
:
291
-297.
18
Ponomarenko, N. A., O. M. Durova, I. I. Vorobiev, E. S. Aleksandrova, G. B. Telegin, O. G. Chamborant, L. L. Sidorik, S. V. Suchkov, Z. S. Alekberova, N. V. Gnuchev, A. G. Gabibov.
2002
. Catalytic antibodies in clinical and experimental pathology: human and mouse models.
J. Immunol. Methods
269
:
197
-211.
19
Ponomarenko, N. A., O. M. Durova, I. I. Vorobiev, A. A. Belogurov, Jr, I. N. Kurkova, A. G. Petrenko, G. B. Telegin, S. V. Suchkov, S. L. Kiselev, M. A. Lagarkova, et al
2006
. Autoantibodies to myelin basic protein catalyze site-specific degradation of their antigen.
Proc. Natl. Acad. Sci. USA
103
:
281
-286.
20
Nardi, M., S. Tomlinson, M. A. Greco, S. Karpatkin.
2001
. Complement-independent, peroxide-induced antibody lysis of platelets in HIV-1-related immune thrombocytopenia.
Cell
106
:
551
-561.
21
Lacroix-Desmazes, S., J. Bayry, S. V. Kaveri, D. Hayon-Sonsino, N. Thorenoor, J. Charpentier, C. E. Luyt, J. P. Mira, V. Nagaraja, M. D. Kazatchkine, J. F. Dhainaut, V. O. Mallet.
2005
. High levels of catalytic antibodies correlate with favorable outcome in sepsis.
Proc. Natl. Acad. Sci. USA
102
:
4109
-4113.
22
Green, D., K. Lechner.
1981
. A survey of 215 non-hemophilic patients with inhibitors to factor VIII.
Thromb. Haemostasis
45
:
200
-203.
23
Sohngen, D., C. Specker, D. Bach, B. M. Kuntz, M. Burk, C. Aul, G. Kobbe, A. Heyll, K. A. Hollmig, W. Schneider.
1997
. Acquired factor VIII inhibitors in nonhemophilic patients.
Ann. Hematol.
74
:
89
-93.
24
Shapiro, S..
1967
. The immunologic character of acquired inhibitors of antihemophiliac globulin (Factor VIII) and the kinetics of their interaction with FVIII.
J. Clin. Invest.
46
:
147
-156.
25
Fulcher, C. A., S. de Graaf Mahoney, T. S. Zimmerman.
1987
. Factor VIII inhibitor IgG subclass and FVIII polypeptide specificity determined by immunoblotting.
Blood
69
:
1475
-1480.
26
Fulcher, C. A., K. Lechner, S. de Graaf Mahoney.
1988
. Immunoblot analysis shows changes in FVIII inhibitor chain specificity in FVIII inhibitor patients over time.
Blood
72
:
1348
-1356.
27
Scandella, D., H. Nakai, M. Felch, W. Mondorf, I. Scharrer, L. Hoyer, E. Saenko.
2001
. In hemophilia a and autoantibody inhibitor patients: the factor viii a2 domain and light chain are most immunogenic.
Thromb. Res.
101
:
377
-385.
28
Arai, M., D. Scandella, L. W. Hoyer.
1989
. Molecular basis of factor VIII inhibition by human antibodies: antibodies that bind to the factor VIII light chain prevent the interaction of factor VIII with phospholipid.
J. Clin. Invest.
83
:
1978
-1984.
29
Saenko, E. L., M. Shima, G. E. Gilbert, D. Scandella.
1996
. Slowed release of thrombin-cleaved factor VIII from von Willebrand factor by a monoclonal and a human antibody is a novel mechanism for FVIII inhibition.
J. Biol. Chem.
271
:
27424
-27431.
30
Shima, M., D. Scandella, A. Yoshioka, H. Nakai, I. Tanaka, S. Kamisue, S. Terada, H. Fukui.
1993
. A factor VIII neutralizing monoclonal antibody and a human inhibitor alloantibody recognizing epitopes in the C2 domain inhibit factor VIII binding to von Willebrand factor and to phosphatidylserine.
Thromb. Haemostasis
69
:
240
-246.
31
Zhong, D., E. L. Saenko, M. Shima, M. Felch, D. Scandella.
1998
. Some human inhibitor antibodies interfere with factor VIII binding to Factor IX.
Blood
92
:
136
-142.
32
Foster, P. A., C. A. Fulcher, R. A. Huoghten, S. de Graaf Mahoney, T. S. Zimmerman.
1988
. Localization of the binding regions of a murine monoclonal anti-FVIII antibody and a human anti-factor VIII alloantibody, both of which inhibit factor VIII procoagulant activity, to amino acid residues threonine351-serine365 of the factor VIII heavy chain.
J. Clin. Invest.
82
:
123
-128.
33
Lubahn, B. C., J. Ware, D. W. Stafford, H. M. Reisner.
1989
. Identification of a FVIII epitope recognized by a human hemophilic inhibitor.
Blood
73
:
497
-499.
34
Ulrich, H., E. Mundorff, B. Santarsiero, E. Driggers, R. Stevens, P. Schultz.
1997
. The interplay between binding energy and catalysis in the evolution of a catalytic antibody.
Nature
389
:
271
-275.
35
Romesberg, F. E., B. Spiller, P. G. Schultz, R. C. Stevens.
1998
. Immunological origins of binding and catalysis in a Diels-Alderase antibody.
Science
279
:
1929
-1933.
36
Xu, J., Q. Deng, J. Chen, K. N. Houk, J. Bartek, D. Hilvert, I. A. Wilson.
1999
. Evolution of shape complementarity and catalytic efficiency from a primordial antibody template.
Science
286
:
2345
-2348.
37
Gao, C., C. H. Lin, C. H. Lo, S. Mao, P. Wirsching, R. A. Lerner, K. D. Janda.
1997
. Making chemistry selectable by linking it to infectivity.
Proc. Natl. Acad. Sci. USA
94
:
11777
-11782.
38
Gololobov, G., M. Sun, S. Paul.
1999
. Innate antibody catalysis.
Mol. Immunol.
36
:
1215
-1222.
39
Planque, S., H. Taguchi, G. Burr, G. Bhatia, S. Karle, Y. X. Zhou, Y. Nishiyama, S. Paul.
2003
. Broadly distributed chemical reactivity of natural antibodies expressed in coordination with specific antigen binding activity.
J. Biol. Chem.
278
:
20436
-20443.
40
Planque, S., Y. Bangale, X. T. Song, S. Karle, H. Taguchi, B. Poindexter, R. Bick, A. Edmundson, Y. Nishiyama, S. Paul.
2004
. Ontogeny of proteolytic immunity: IgM serine proteases.
J. Biol. Chem.
279
:
14024
-14032.
41
Friboulet, A., B. Avalle, H. Debat, D. Thomas.
1999
. A possible role of catalytic antibodies in metabolism.
Immunol. Today
20
:
474
-475.
42
Tawfik, D. S., R. Chap, B. S. Green, M. Sela, Z. Eshhar.
1995
. Unexpectedly high occurence of catalytic antibodies in MLR/lpr and SLJ mice immunized with a transition-state analog: is there a linkage to autoimmunity.
Proc. Natl. Acad. Sci. USA
92
:
2145
-2149.