HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 30, 27232-27239, July 26, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/30/27232    most recent M203415200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Villard, S. Articles by Granier, C. Search for Related Content PubMed PubMed Citation Articles by Villard, S. Articles by Granier, C. Social Bookmarking                      What's this?

Low Molecular Weight Peptides Restore the Procoagulant Activity of Factor VIII in the Presence of the Potent Inhibitor Antibody ESH8^*

Sylvie Villard, Dominique Piquer, Sanjee Raut, Jean-Paul Léonetti, Jean-Marie Saint-Remy^§, and Claude Granier^¶

From the Unité Mixte de Recherche, CNRS 5094, Faculté de Pharmacie, BP 14491, 34093 Montpellier, France,  the Division of Haematology, National Institute for Biological Standards and Control, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, United Kingdom, and the ^§ Center for Molecular and Vascular Biology, University of Leuven, 3000 Leuven, Belgium

Received for publication, April 9, 2002, and in revised form, May 14, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Following repeated administration of factor VIII (FVIII), a significant number of hemophilia A patients develop antibodies (Abs), inhibiting the procoagulant activity of infused FVIII. We have designed an approach based on the blocking of the deleterious activity of these Abs by peptide decoys mimicking the anti-FVIII Ab epitopes. Here, the well characterized inhibitory monoclonal Ab ESH8 served as a model. Several phage peptide libraries were screened for specific binding to ESH8. Seven constrained dodecapeptide sequences were obtained. Six sequences carried the consensus motif, hydrophobic-(Y/F)GKTXL. This motif showed a certain similarity with the ²²³¹QVDFQKTMKV²²⁴⁰ sequence of the C₂ domain. In the seventh sequence, YCNPSIGDKNCR, the residues GDKN are similar to the sequence ²²⁶⁷DGHQ²²⁷⁰. Upon inspection of the C₂ domain crystallographic structure, the two stretches QVDFQKTMKV and DGHQ appeared close together in space and might constitute a discontinuous epitope. Corresponding synthetic peptides were able to inhibit the binding of ESH8 to FVIII in a specific and dose-dependent manner. Moreover, the ability of the selected peptides to neutralize the inhibitory activity of ESH8 was demonstrated in functional tests as well as in vivo in a murine model of hemophilia A. This study demonstrates the potential of this approach to neutralize the activity of potent inhibitory Abs.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Coagulation defects attributed to the absence or dysfunction of blood coagulation factor VIII (FVIII)¹ are observed in 0.01-0.02% of the male population (1). This recessive X-linked bleeding disorder is known as hemophilia A. The adequate treatment of this disease relies on infusions of human FVIII concentrates. However, a serious treatment complication is the development of a humoral immune response to FVIII in ~25% of patients with hemophilia A. These antibodies (Abs), also called inhibitors, block the FVIII procoagulant activity and complicate seriously the medical care of the patients by precluding further FVIII injections (2). This immune response is not only polyclonal but also heterogeneous in its specificity (3). Nevertheless, epitope mapping studies have shown that inhibitors recognize restricted binding sites, predominantly on the A₂, C₂, and A₃ domains on the FVIII molecule (4). The incidence of inhibitor occurrence in hemophilia A patients is difficult to estimate and can vary from 18 to 28% (5). This variation can be the result of a number of parameters, which needs to be considered, including quantitative criteria of inhibitor dosages, the elapsed time from the first apparition, the origin, the viral inactivation procedure applied, the purity of the administered product, and, of course, the patient.

A number of different therapeutic approaches have been developed to circumvent complications arising from inhibitors, such as the administration of porcine FVIII, which is less antigenic than its human counterpart (6), the administration of activated human FVII to bypass the intrinsic pathway (7), and the activation of the coagulation cascade by using prothrombin complex concentrates (8). Although these treatments have greatly improved the medical management of the disease, the ultimate goal for therapy, the specific neutralization of the immune response to FVIII, has not yet been achieved. As this complication of hemophilia A is largely antibody-mediated, we have applied a novel approach based on blocking the deleterious activity of these Abs by low molecular weight peptide decoys mimicking epitopes recognized by anti-FVIII Abs with a view to restore a normal procoagulant activity.

For the selection of peptide decoys, an approach based on phage peptide display was adopted (9). This technology has already been successfully applied in the area of thrombosis and hemostasis (10). In particular, a number of researchers have used phages displaying single chain variable fragment to study the molecular properties and functional characteristics of human anti-FVIII Abs (11). Nord et al. (12) used phages displaying ligands, which were able to bind to human FVIII to isolate them by affinity chromatography. However, to our knowledge, the phage peptide display technique has never been used to isolate peptides that are able to neutralize the inhibitory activity of anti-FVIII Abs. This methodology based on the display of random peptides fused to the minor (pIII) or major (pVIII) coat proteins of filamentous phages allows the identification of new molecules that influence protein-protein interactions as, for example, the antigen-antibody interaction. Such libraries allow to isolate mimotopes, i.e. peptides mimicking the binding properties of the natural antigen but with no apparent sequence homology with the antigen (13). The mimicking potential of the peptides is broad as mimotopes of linear, conformational, and even non-proteinaceous epitopes have been reported (14).

To evaluate the possibility of disrupting the interaction between the FVIII molecule and its inhibitors by peptide decoys to restore the procoagulant activity of FVIII, a model system was investigated, namely, the well characterized murine inhibitor mAb ESH8 (15). We used this mAb to screen four different libraries displaying foreign peptides at the surface of the major coat protein, pVIII. We present here evidence both in vitro and in vivo that mimotopes of the C₂ domain can efficiently neutralize the inhibitory activity of mAb ESH8 and as such could represent an efficient method for the treatment of hemophilia A patients with inhibitory antibodies.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Screening the Phage Library with mAb ESH8-- The different M13 phage libraries used expressing 15-mer (X₁₅), 30-mer (X₃₀), 17-mer including a fixed cysteine residue (XCX₁₅), and 12-mer peptides including two fixed cysteine residues (XCX₈CX,) were all described by Bonnycastle et al. (16) and were obtained from Dr. J. Scott (Simon Fraser University, Burnaby, British Columbia, Canada). MAb ESH8 was obtained from American Diagnostica Inc. (Greenwich, CT) and has been characterized previously (15). Pannings were performed as described by Smith (9). mAb ESH8 was coated onto a polystyrene Petri dish (10 × 1.5 cm, Falcon 1029) overnight at 4 °C at a concentration of 5 µg/ml for the first two pannings and a concentration of 0.5 µg/ml for the last panning in 100 mM NaHCO₃, pH 8.6, on a shaking platform. For the first panning, 5 × 10¹² transducting units of each library were incubated with the adsorbed mAb. After incubation, unbound phages were washed, and bound phages were removed by acidic elution. Eluted phages were then used to infect Escherichia coli K91 cells. After three rounds of enrichment, individual phage clones were isolated and further analyzed (17).

Phage Binding Analysis by ELISA-- ELISA microtiter plates were coated with 0.5 µg/ml ESH8 or control mAbs in 100 mM NaHCO₃, pH 8.6, overnight at 4 °C. Plates were washed with PBS, 0.1% Tween 20 (v/v) and then blocked with PBS, 0.1% Tween 20, 2% nonfat dried milk (w/v) for 1 h at 37 °C. A mixture of 50 µl of a phage dilution and 50 µl of blocking buffer then was added to each well. Phage particles were incubated for 2 h at 37 °C. Binding was detected using a peroxidase-conjugated anti-M13 antibody (Roche Molecular Biochemicals) diluted 1:3000 in blocking buffer. After 1 h at 37 °C and washing, color was developed by adding the peroxidase substrate. The resulting absorbance was measured at 450 nm with an automated microtiter plate reader (Dynatech MR 5000).

Competitive ELISA with Phages-- The buffers used were the same as those described above. ELISA plates were coated with 0.5 µg/ml ESH8 overnight at 4 °C and then washed and blocked for 1 h at 37 °C. Phages at 10¹² transducting units/ml and FVIII at increasing concentrations were incubated together for 2 h at 37 °C. For the control, the phages were incubated with an irrelevant protein, bovine serum albumin. The binding of the phages was detected by a peroxidase-conjugated anti-M13 antibody incubated 1 h at 37 °C. The resulting absorbance was measured at 450 nm as described above.

Sequence Identification-- For each phage, ~9 µg of single-stranded DNA was purified using the QIA prep Spin M13 protocol (Qiagen). Sequencing reactions were done by the dideoxy chain termination method (18) using the Abi Prism kit (PE Applied Biosystems) with ABI PRISM 377 (PerkinElmer Life Sciences). The primer 5'-TCGGCAAGC TCTTTTAGG-3' was used for annealing.

Synthesis and Immunoassay of Cellulose-bound Peptides-- The different phage sequences as well as overlapping peptides of 15 and 25 amino acids frameshifted by one residue representing the C₂ domain of FVIII were prepared by the Spot technique. The general protocol has been described previously (19). The membranes were obtained from Intavis (Bergish Gladbach, Germany). Fmoc amino acids and N-hydroxybenzotriazole were obtained from Novabiochem. An ASP222 robot (Abimed) was used for the coupling steps. All of the peptides were acetylated at their N terminus. After the peptide sequences had been assembled, the side-chain protecting groups were removed by trifluoroacetic acid treatment. The set of membrane-bound peptides was probed by incubation with ESH8 (5 µg/ml). Its binding to peptides was detected by using an alkaline phosphatase-conjugated anti-mouse antibody (Sigma) diluted 1:1000 with its substrate, BCIP-MTT (5-bromo-4-chloro-3-indolylphosphate, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, Sigma), which gives a blue precipitate on peptide spots recognized by ESH8. A plot of spot intensities was obtained with the Scion Image software after scanning the membrane. The membrane was further treated to remove precipitated dye and bound antibodies and was reused (19).

Synthesis of Soluble Peptides-- All of the soluble peptides were synthesized on an Abimed AMS422 synthesizer by Fmoc chemistry (20). Peptides were deprotected and released from the resin by trifluoroacetic acid treatment in the presence of appropriate scavengers. Peptides were lyophilized, and their purity assessed by reverse phase high pressure liquid chromatography and mass spectrometry. When necessary, peptides were purified by high pressure liquid chromatography to reach a purity greater than 90%.

Competitive ELISA Assay with Synthetic Peptides Derived from Phages-- ESH8 (0.05 µg/ml) was preincubated in PBS-0.1% Tween 20, 2% nonfat dried milk (w/v) with synthetic peptides at increasing concentrations overnight at 4 °C. ELISA plates were coated with 1 µg/ml recombinant FVIII overnight at 4 °C in PBS. Plates were washed and blocked for 1 h at 37 °C. The mixture of ESH8 and synthetic peptides was incubated 2 h at 37 °C. mAb binding was revealed by adding a peroxidase-conjugated anti-mouse IgG antibody (Sigma) diluted 1:3000 for 1 h at 37 °C. The resulting absorbance was measured at 490 nm after the addition of 50 µl of 4 N H₂SO₄. The reference absorbance was obtained with the incubation of ESH8 alone.

Modified Bethesda Assay-- This assay was based on the Bethesda assay (21) with the Nijmegen modification. ESH8 at a fixed concentration giving 50% FVIII activity inhibition was incubated overnight at 4 °C with the different peptides at various concentrations diluted with 0.05 M imidazole buffer, pH 7.3. Equal volumes (250 µl) of this sample and normal plasma buffered with 0.1 M imidazole, pH 7.4, were mixed together and incubated for 2 h at 37 °C. One-stage FVIII potency assays were then carried out. The inhibitory activity was read in Bethesda unit/ml from a semilogarithmic plot representing the correlation between residual FVIII activity (logarithmic) and inhibitor activity (linear). A minimum of three dilutions was made for each condition.

Thrombin Generation Test-- This assay is a modification of that carried out by Pitney and Dacie (22). Normal platelet-poor plasma was used as substrate. 375 µl of this normal platelet-poor plasma substrate was added to a 100-µl preincubated solution of ESH8 (175 nM) mixed with either Tris borate saline buffer (control) or with a 1000 molar excess of the different peptides. This mixture was incubated at 37 °C for 20 min prior to the addition of 80 µl of FIXa (14 nM) and further incubated for 1.5 min at 37 °C. To this mixture, 400 µl of phospholipid (3.1 µg/ml) and 400 µl of CaCl₂ (7.8 mM) were then added to start the reaction. At different time intervals, 50-µl aliquots were then removed and placed in 200 µl of fibrinogen in cups on a Deca coagulometer (Diagnostic Grifols S.A., Barcelona, Spain). The clotting times were then converted into thrombin units by use of a -thrombin standard curve. The peak thrombin level and the time taken to reach half-maximal (t_1/2max) coagulation were deduced from the thrombin generation curves.

Peptide Stability-- The peptides were incubated for 0, 1, 2, 3, 4, 5, and 24 h at 37 °C at 300 µM in cell culture medium containing 10% fetal calf serum. Following incubation, the samples (150 µl) were centrifuged for 8 min at 11,000 rpm through an ultrafiltration membrane (10 kDa, Amicon Microcon) to remove serum proteins. 100 µl aliquots of the filtrate were then analyzed by reverse phase-high pressure liquid chromatography and quantified.

In Vivo Experiments-- The FVIII knock-out mice used here were as described previously (23-25). All of the mice were used in groups of three individuals. 0.5 IU of human FVIII (Kogenate, Bayer Inc., Berkeley, CA) diluted to 100 µl in physiological medium was injected in the tail vein. A blood sample obtained by cardiac puncture after 15 min was taken, and the concentration of FVIII was evaluated in a chromogenic assay (Dade Behring). A group of mice was injected in the tail with 0.25 µg of ESH8 followed 30 min later by an injection of 0.5 IU of FVIII. Preliminary experiments had demonstrated that this amount of ESH8 was sufficient to inhibit 80% FVIII activity. Lastly, ESH8 was preincubated overnight at 4 °C with 1 mg of peptide 46 or an irrelevant peptide before injection in the tail vein, and the residual FVIII activity was determined as described above.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Selection of Phages Displaying Peptides Recognized by mAb ESH8-- To identify peptide binding to mAb ESH8, we screened four different phage libraries, two of which expressed linear peptides of either 15 (X₁₅) or 30 amino acids (X₃₀), whereas the other two phage libraries displayed peptides including either one or two fixed cysteines and whose sizes were 17 (XCX₁₅) or 12 amino acids (XCX₈CX). These random peptides are fused to the pVIII coat protein of filamentous phage. A significant enrichment for phage binding to the target Ab was obtained after three rounds of panning on the immobilized mAb ESH8. Three hundred phage clones were randomly picked from the third round of selection and were checked by ELISA for their ability to bind mAb ESH8. The specificity of the signal was tested by evaluating their reactivity on mAb ESH4 (15), another anti-C₂ mAb, or an anti-Troponin I mAb of the same isotype, mAb 11E12 (26). Thirty positive clones giving ELISA signals at least four times higher than background were further characterized. All of them bound to mAb ESH8 in a dose-dependent manner and in a specific way, because none of them cross-reacted with each of the control mAbs. The results obtained with the most reactive phage clone are shown in Fig. 1A. This phage was further used in a competition experiment with FVIII (Fig. 1B). FVIII competed with the phage for mAb ESH8 binding in a dose-dependent manner with a IC₅₀ of 4 µg/ml at 10¹¹ phages/ml. Reciprocally, the clone was able to inhibit FVIII binding to mAb ESH8. However, because of limitation in the available quantity of phages, no complete inhibition curve could be obtained (data not shown). Taken together, these results show that phage peptides specifically recognizing mAb ESH8 have been obtained.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Binding and specificity of the most reactive phage clone selected on mAb ESH8. A, the target mAb ESH8 and two control mAbs, ESH4, another anti-C2 domain mAb, and 11E12, an irrelevant Ab of the same isotype, were coated onto microtitration plates. The binding of the phage clone on ESH8 (), on ESH4 (), and on 11E12 () was detected by an anti-M13 antibody. B, mAb ESH8 was coated onto microtitration plates. The binding of the most reactive phage to mAb ESH8 was inhibited by increasing concentrations of FVIII () or an irrelevant protein, bovine serum albumin (BSA) (). Phage binding was detected as described above.

Sequence Determination and Comparison-- The phage DNA coding for foreign peptide was sequenced. The deduced peptide sequences are shown in Table I. A sequence analysis revealed striking features. Despite exhaustive screening of four phage peptide libraries, all of the selected peptides originated only from the dodecapeptide cysteine-constrained library (XCX₈CX), suggesting that both the primary sequence and the conformational context in which the sequence is presented are critical for binding. From the thirty selected clones, only seven distinct sequences were obtained with six peptides sharing the consensus motif GKTXL. These four invariant amino acids suggested that they might be important for binding either by providing residues contacting mAb ESH8 or by participating in the appropriate peptide folding. A closer inspection of the six related sequences revealed that the second position in the loop was exclusively occupied by a hydrophobic residue, a valine (3 times), a threonine (2 times), or a leucine (1 time). The third position in the loop corresponded to two aromatic residues, tyrosine or phenylalanine. Consequently, the final consensus motif could be the following: XCX(V/T/L)(Y/F)GKTXLCX with X being any amino acids. The seventh sequence (YCNPSIGDKNCR, peptide 73) is totally distinct from the others. To assess whether the selected peptides could still be recognized by mAb ESH8 out of phage context, they were synthesized on a cellulose membrane by the Spot technique. All of the sequences were recognized by mAb ESH8, suggesting that the binding activity was an intrinsic property of the peptide (Fig. 2A). In this format, the unique sequence, YCNPSIGDKNCR (peptide 73), as well as the sequences ECIVYGKTALCT (peptide 46) and QCQTFGKTMLCT (peptide 47) were the most reactive.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Reactivity of ESH8 with the selected peptides presented out of the phage. A, peptides synthesized by the Spot method were probed for reactivity with ESH8. The spot numbers correspond to the peptide sequences mentioned in Table I. B, substitution study of each residue of the peptide 46 was also evaluated. Each residue was changed into each one of the 19 amino acids, and these peptides were tested for reactivity with ESH8. Black boxes, 80-100% of the parent peptide reactivity; gray boxes, 50-79% of the parent peptide reactivity; white boxes, <50% of the parent peptide reactivity; ND, not determined.

Localization of Mimotopes within the C₂ Domain Sequence-- To identify peptide residues critical for binding to mAb ESH8, a mutational analysis based on the Spot technique was performed. A set of peptides was generated in which each residue of the lead sequence was replaced in turn by each one of the other nineteen naturally occurring amino acids. Fig. 2B shows the results on the analysis conducted on peptide 46, ECIVYGKTALCT. The four conserved residues, Gly, Lys, Thr, and Leu, of the consensus motif appeared important functionally and/or structurally as any substitution totally abrogated the binding. Similarly, the substitution of the two cysteines by two alanines or serines resulted in a complete loss of reactivity, underlying the possible importance of the disulfide bridge for binding. With regards to the second position, only the amino acid with hydrophobic properties (Ile, Leu, Val, Met, Trp, Tyr, and Thr) was tolerated, and a phenyl functionality at the third position was crucial because no substitution even by tryptophan was allowed. The same reactivity profile was obtained with the five other related sequences (data not shown). However, the same study conducted on the sequence YCNPSIGDKNCR (peptide 73) showed that the critical amino acids were isoleucine, glycine, aspartic acid, lysine, and the last asparagine residue as well as the two cysteines (data not shown). The sequence alignment between the C₂ domain of FVIII and the consensus motif indicated a putative three residue homology (²²³⁴FKT²²³⁷) in the sequence QVDFQKTMKV (residues 2231-2240) of the C₂ domain (Table I). By inspecting the three-dimensional structure of the C₂ domain (27), these three residues appeared to belong to an exposed loop, which faces residues ²²⁶⁷DGHQ²²⁷⁰ (Fig. 3A). On the other hand, although peptide 73 presented no apparent homology with the FVIII sequence, we observed possible similarities between its sequence, YCNPSIGDKNCR, and the sequence ²²⁶⁷DGHQ²²⁷⁰ of the C₂ domain. The aspartic acid and the glycine residues were present in both of them. His²²⁶⁹ can be replaced by conservative substitution with lysine and Gln²²⁷⁰ with asparagine. Therefore, it is possible that peptide 73 functionally mimicked the region ²²⁶⁷DGHQ²²⁷⁰ of FVIII, a region that is close to residues ²²³⁴FKT²²³⁷ in the three-dimensional structure of the C₂ domain (Fig. 3B). We postulate that the mAb ESH8 epitope is discontinuous and made up of two antigenic determinants represented by the two selected sequences (Fig. 3B). Another observation strengthens this hypothesis. Epitope mapping studies using overlapping 15-mer peptides covering the whole sequence of the C₂ domain gave no positive result, suggesting that mAb ESH8 recognized a conformational epitope (data not shown). However, when 25-mer peptides were used instead, an epitope with two components was uncovered, i.e. EFLISSSQDGHQWTLFFQNGK (residues 2259-2279) and KEWLQVDFQKTMKVT (residues 2227-2241) (Fig. 3A). Each one of these components contained the regions (underlined) we postulated as constituting the mAb ESH8 epitope.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Localization of the mAb ESH8 epitope on the C2 domain. Representation of the 2259EFLISSSQDGHQWTLFFQNGK2279 sequence, the epitope identified by Scandella et al. (15) (yellow), and the two component epitopes deduced from phage display, 2267DGHQ2270 and 2231QVDFQKTMKV2240 (red).

Inhibition of the Interaction between mAb ESH8 and FVIII in Vitro and in Functional Assays-- The seven sequences were chemically synthesized in a soluble form and tested for their ability to inhibit the interaction of mAb ESH8 with FVIII in ELISA. To this end, the binding of mAb ESH8 to FVIII was examined in the presence of increasing amounts of specific or unrelated peptides. As shown in Fig. 4A, the different peptides derived from the phage display selection efficiently inhibited the binding of mAb ESH8 to FVIII in a dose-dependent manner, which was not the case for an irrelevant peptide. At a concentration of ~10 µM, all of the six peptides were able to inhibit 50% mAb ESH8 binding, and complete inhibition was obtained at a concentration higher than 100 µM. Therefore, these results suggest that the different peptides derived from the phage display selection mimicked the C₂ domain for its binding properties to mAb ESH8.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Capacity of the peptides to neutralize the inhibitory activity of ESH8. A, inhibition of the binding of ESH8 to FVIII by increasing concentrations of peptides in ELISA. FVIII was coated onto microtitration plates, and the binding of ESH8 was inhibited by increasing concentrations of either specific peptides (45, 46, 47, 48, 72, and 73) or an irrelevant one. B, neutralization of the inhibitory activity of ESH8 by increasing concentrations of peptides in Bethesda test. The concentration of ESH8 was that giving 50% FVIII activity inhibition.

The capacity of peptides to neutralize the inhibitory activity of mAb ESH8 was measured in two different functional tests. First, in a modified Bethesda assay (21) in which mAb ESH8 was used at a concentration resulting in 50% inhibition of FVIII cofactor activity, the peptides were all able to neutralize mAb ESH8 inhibitory activity in a dose-dependent manner with complete neutralization achieved by peptides 46 and 47 when used in a 10⁶ molar excess, corresponding to a final concentration of 200 µM, over mAb ESH8 (Fig. 4B). The neutralization was specific, because an irrelevant peptide had no measurable activity in the same range of concentrations (Fig. 4B). The peptides by themselves had a minimal effect in the assay (data not shown). Interestingly, when the test was repeated with a concentration of mAb ESH8 giving 70% inhibition of FVIII activity, complete neutralization was still reached by using a 10⁶ molar excess of peptides 46 and 47.

These peptide properties were confirmed in a second functional assay based on thrombin generation (TG). Fig. 5 shows the TG profiles of normal platelet-poor plasma in the absence and presence of mAb ESH8. A lag phase was observed in the presence of mAb ESH8, although the peak thrombin produced remained very similar to the profile obtained in absence of mAb ESH8. The inhibition of FVIII by mAb ESH8 gave a t_1/2max value of 220 s compared with the t_1/2max value of 116 s for normal plasma alone (Fig. 5). When this test was repeated with mAb ESH8 that had been preincubated with the various peptides, the inhibition of the TG profile was corrected by all of the peptides with the exception of the control peptide, which showed only a slight correction of the lag phase. This finding suggests that neutralization of the inhibitory activity of mAb ESH8 by peptides 45, 46, 47, and 48 is indeed of physiological relevance.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Efficiency of the peptides in a thrombin generation assay. A, comparison of thrombin generation profiles of normal platelet-poor plasma (NPPP) in the presence of buffer (), ESH8 alone (), ESH8 pre-incubated with peptide 47 (), or with a control peptide (). B, summary of the neutralizing effects of the different peptides on ESH8 inhibition of FVIII expressed as TG parameters, t1/2max (seconds), and peak thrombin (IU/ml).

Efficiency of the Peptide in Vivo-- Based on these in vitro experiments, a potential application of such peptides for neutralizing FVIII inhibitor Abs in vivo was considered. We first assessed the stability of peptides 46 and 73 upon incubation in 10% fetal calf serum at 37 °C for various lengths of time using high pressure liquid chromatography. No degradation was observed for either peptide after 2 h of incubation, in contrast with the fate of a linear peptide that was already 50% degraded (data not shown). Over a longer period of time (24 h), peptide 73 was shown to be particularly resistant with only 50% degradation, whereas 84% peptide 46 was degraded. Under the same experimental conditions, the linear peptide was fully degraded.

These results encouraged us to examine whether or not such peptides could neutralize the FVIII inhibitory activity of mAb ESH8 in an in vivo model. We used FVIII-deficient mice obtained by targeted disruption of exon 16 of the FVIII gene (23, 24). An injection of 0.5 IU of human FVIII in the tail vein resulted in a FVIII level of 0.3 IU/ml after 10 min as measured by a chromogenic assay. 2 groups of 3 mice were pretreated with either 0.25 µg of mAb ESH8 or a mixture of 0.25 µg of mAb ESH8 with 1 mg of peptide 46 or control peptide preincubated overnight at 4 °C (this represents a 2 × 10⁵ molar excess of the peptide). Thirty minutes later, all of the mice were reconstituted with 0.5 IU of human FVIII. From Fig. 6, it can be seen that pretreatment with mAb ESH8 alone reduces FVIII activity by 88%, whereas mice treated with the mAb ESH8-peptide mixture had only a 58% reduction, indicating that peptide 46 exerted some neutralizing activity on mAb ESH8-mediated FVIII inhibition. Such neutralization was specific for peptide 46, because a control peptide was unable to restore FVIII activity.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Capacity of the peptide 46 to neutralize the inhibitory activity of ESH8 in a murine model of hemophilia A. Human FVIII was injected in the tail vein of mice to obtain a detectable level by chromogenic assay. This circulating level was evaluated in the absence or in the presence of ESH8, administered alone, or preincubated with peptide 46 or control peptide.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

A significant proportion of hemophilia A patients develops inhibitor Abs following repeated FVIII administration (28). Such Abs inhibit the procoagulant activity of FVIII and present a serious clinical complication in that they preclude FVIII replacement therapy. Several treatments have been attempted to overcome this problem, but these are therapeutically or economically not suitable, emphasizing the need to evaluate new approaches. One such attempt could be to prevent FVIII binding to the inhibitor Abs using epitope-mimicking peptides. By combining to the variable parts of these Abs, such peptides would neutralize the inhibitory capacity of anti-FVIII Abs. In an attempt to evaluate the feasibility of this approach, we chose an anti-C₂ antibody, mAb ESH8, which was used to identify peptides that were able to divert mAb ESH8 from FVIII. The choice for mAb ESH8 was motivated by several criteria. First, this mAb is well characterized (15, 29, 30) and is used as reference inhibitor Ab by a number of research teams. Second, its epitope is mapped to residues 2248-2285 on the C₂ domain of FVIII, a domain frequently recognized by human anti-FVIII Abs (31). Third, it is an effective high titer inhibitor Ab (6300 Bethesda unit/mg) (15). Finally, as monoclonal and human Abs that map to C₂ epitope 2218-2307, its inhibition mechanism is based on the slow release of thrombin-cleaved FVIII from von Willebrand factor, which reduces FVIII binding to phospholipids (29). The peptides binding to mAb ESH8 were selected by the phage display technology, a method allowing the identification of peptides with significant mimicry potential for a broad spectrum of applications (14). We have identified from a mixture of several library peptides, two that are able to functionally mimic mAb ESH8 epitope. By combining the information obtained from the consensus motifs of the two sequences with that derived from the three-dimensional structure of the C₂ domain (27), we were able to identify two distinct regions of the C₂ domain likely to be involved in the interaction with mAb ESH8. The two ligand peptides are postulated to mimic non-consecutive solvent-exposed stretches brought together in the three-dimensional structure of the C₂ domain and constituting a discontinuous epitope for mAb ESH8. One of the two components (²²⁶⁷DGHQ²²⁷⁰) belongs to a region that had already been described by Scandella et al. (15). They performed an epitope mapping using truncated forms of recombinant C₂ domain and identified a region encompassing residues 2248-2285 as the putative mAb ESH8 epitope. The second epitopic component (residues belonging to the fragment 2227-2241) was identified in this study. mAb ESH8 was found to have some affinity for each individual peptide sequence. The two unrelated and separated peptides were each able to block in an efficient and specific manner the in vitro binding of mAb ESH8 to FVIII. The simultaneous addition of the two peptides improved the inhibition (data not shown). An inspection of the crystallographic structure of the C₂ domain (27) indicates that the two parts of the epitope are in fact spatially close together (7.3 Å) and may form a coherent epitope. Our data indicate that the two contributing parts are equally involved in mAb ESH8 recognition. Scandella et al. (15) carried out the analysis using the N-terminal truncated C₂ domain in which the second antigenic part could not be identified. Therefore, our results do not stand in contradiction with this previous report. Consistent with these epitope-mimicking properties, the peptides identified here also exert functional properties in vitro. In the Bethesda test, the peptides completely neutralized the inhibitory activity of mAb ESH8, thereby restoring full FVIII cofactor activity. This neutralizing effect was also observed in the inhibition of the thrombin generation test in which peptides representative of the two epitopes were able to neutralize the mAb ESH8-dependent inhibition of thrombin generation. In both cases, 200 µM of peptides were required to completely neutralize the inhibitory activity of mAb ESH8. This is probably related to the fact that each peptide represents only a part of the epitope to which mAb ESH8 binds. This is in line with the findings of Reineke et al. (32) in a study on the reconstruction of a discontinuous binding site on interleukin 10. These authors (32) demonstrate that each of the peptides mimicking either one of the two parts of the binding site was only slightly efficient when used alone. However, combining these peptides representative of the two binding sites by a linker molecule resulted in a biological activity within the nanomolar range. The capacity to prevent the binding of antibodies to FVIII by specific epitope-like peptides has also been shown with another anti-FVIII mAb, mAb F7b4.² In such a case, full neutralization of the FVIII inhibitory activity was obtained with only a 50-fold molar excess (14 µM). Whether or not the combination of peptides would be required for improved neutralization, our data already demonstrate the feasibility of the approach, because peptides can efficiently neutralize in vitro the inhibitory activity of an anti-FVIII even when they mimic only part of the epitope.

An in vivo experiment using single peptides was carried out to establish the basis of a possible therapeutic application. Because mAb ESH8 does not inhibit the cofactor activity of mouse FVIII,³ we used the model of hemophilia A mice reconstituted by human recombinant FVIII (25). The inhibition of FVIII as measured in a chromogenic assay was compared using either ESH8 alone or in a combination with one of the peptide. A significant yet limited neutralization of the inhibitory capacity of ESH8 was observed. The efficacy of a peptide in vivo depends on a number of parameters and, in particular, on its stability. For example, it is commonly accepted that L-peptides are rapidly degraded in serum (33). Preliminary data had shown that peptide 46 exhibited an unusual resistance to protease degradation upon incubation in serum, which was deemed to be dependent on the presence of a disulfide bridge between Cys-2 and Cys-11, and to the derivatization of both peptide ends, acetylation of the N-terminal end, and amidation of the C-terminal end (34). These characteristics prompted us to use the peptide as such with no attempt to further increase its stability. However, this could be achieved by the replacement of L-amino acids by D-derivatives or by the addition of unnatural ones and/or modification of the peptide bond itself. In addition, the dosing regimens used for the in vivo bioassays were based on an extrapolation from in vitro results, which might not be optimal for animal administration.

The efficiency of a molecular decoy approach similar to the one described in this paper has already been studied in two models of human pathology, diabetes mellitus (35) and myasthenia gravis (36). In both cases, short RNA sequences were selected from a large random RNA library on the basis of their ability to specifically bind to a mAb representative of human autoantibodies. The authors (35, 36) demonstrated that the selected decoys were able to inhibit the binding of a few human sera to their target antigen, insulin, and acetylcholine receptor, respectively. However, the efficacy of such RNA decoys had not yet been validated in an animal model. Provided that technical improvements would increase peptide stability and capacity to neutralize inhibitory Abs, a number of clinical applications can already be envisioned. Moderate and mild hemophilia A patients with inhibitor could be more prone to benefit from peptide-based therapy insofar, because the anti-FVIII immune response of such patients appears to involve only a limited number of B cell clones, the specificity of which is directed essentially against the epitope corresponding to the region containing the FVIII mutation. In severe hemophilia A cases, although the immune response is usually more heterogeneous, the B cell epitopes cluster to only a limited number of regions, i.e. residues Arg⁴⁸⁴-Ile⁵⁰⁸ on the A₂ domain (37), residues Glu²¹⁸¹-Val²²⁴³ and His²³¹⁵-Asp²³³² on the C₂ domain (30, 38), and finally residues Gln¹⁷⁷⁸-Met¹⁸²³ (39). The present approach will now be repeated using Abs of human origin of both monoclonal and polyclonal origins. The optimal combination of peptides will be determined to obtain a "universal" decoy or a mixture of decoys that is able to inhibit a majority of anti-FVIII Abs. If full neutralization of inhibitor Abs cannot be obtained, we can expect that it will significantly help reduce significantly the antibody titers of the patients.

	ACKNOWLEDGEMENTS

We thank Dr. Sharon Lynn Salhi for the editorial revision of this paper. We also thank Sabrina Grailly for carrying out the in vivo experiments.

	FOOTNOTES

^* This work was partly supported by a grant from Wyeth.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: UMR CNRS 5094, Faculté de Pharmacie, BP 14491, 15 Ave. Charles Flahault, 34093 Montpellier Cedex 5, France. Tel.: 33-4-67-54-86-02; Fax: 33-4-67-54-86-10; E-mail: claude.granier@ibph.pharma.univ-montp1.fr.

Published, JBC Papers in Press, May 14, 2002, DOI 10.1074/jbc.M203415200

² S. Raut, S. Villard, S. Grailly, J.-G. Grilles, C. Granier, J. M. Saint-Remy, and T. Barrowdi, submitted for publication.

³ J. M. Saint-Remy, unpublished data.

	ABBREVIATIONS

The abbreviations used are: FVIII, factor VIII; Abs, antibodies; mAb, monoclonal antibody; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline; Fmoc, N-(9-fluorenyl)methoxycarbonyl; TG, thrombin generation.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				S. Planque, M. A. Escobar, K. C. Smith, H. Taguchi, Y. Nishiyama, E. Donnachie, K. P. Pratt, and S. Paul Covalent Inactivation of Factor VIII Antibodies from Hemophilia A Patients by an Electrophilic FVIII Analog J. Biol. Chem., May 2, 2008; 283(18): 11876 - 11886. [Abstract] [Full Text] [PDF]

				S. Lacroix-Desmazes, B. Wootla, S. Dasgupta, S. Delignat, J. Bayry, J. Reinbolt, J. Hoebeke, E. Saenko, M. D. Kazatchkine, A. Friboulet, et al. Catalytic IgG from Patients with Hemophilia A Inactivate Therapeutic Factor VIII J. Immunol., July 15, 2006; 177(2): 1355 - 1363. [Abstract] [Full Text] [PDF]

				T. C. Lei and D. W. Scott Induction of tolerance to factor VIII inhibitors by gene therapy with immunodominant A2 and C2 domains presented by B cells as Ig fusion proteins Blood, June 15, 2005; 105(12): 4865 - 4870. [Abstract] [Full Text] [PDF]

				K. Vanhoorelbeke, H. Depraetere, R. A. P. Romijn, E. G. Huizinga, M. De Maeyer, and H. Deckmyn A Consensus Tetrapeptide Selected by Phage Display Adopts the Conformation of a Dominant Discontinuous Epitope of a Monoclonal Anti-VWF Antibody That Inhibits the von Willebrand Factor-Collagen Interaction J. Biol. Chem., September 26, 2003; 278(39): 37815 - 37821. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/30/27232    most recent M203415200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Villard, S. Articles by Granier, C. Search for Related Content PubMed PubMed Citation Articles by Villard, S. Articles by Granier, C. Social Bookmarking                      What's this?