Pandey, G. et al (2013) Polymorphisms in the F8 gene and MHC-II variants as risk factors in the development of inhibitory anti-factor VIII antibodies during the treatment of haemophilia A: A computational assessment. DOI:10.1371/journal.pcbi.1003066 (PubMed ID 23696725)
The bleeding disorder Haemophilia A is due to a deficiency in (or faulty) Factor VIII protein, an integral part of the blood clotting cascade. The condition is treated by infusion with either plasma-derived or more commonly, recombinant Factor VIII (rFVIII). However 25-30% of patients treated in this way develop neutralising anti-rFVIII antibodies (commonly called ‘inhibitors’). Inhibitors are twice as prevalent in African American Haemophilia A patients compared with Caucasian patients. Common non-synonymous single nucleotide polymorphisms (ns-SNPs) in the F8 gene exist as six haplotypes (designated H1 to H6). The endogenous FVIII is much more polymorphic in individuals of black African descent compared with Caucasians. Haplotypes H3 and H4 which have been found exclusively so far in Americans of black African descent have been associated with a higher prevalence of inhibitors suggesting that such polymorphisms in FVIII are risk factors in inhibitor development.
Development of FVIII inhibitors require fragments of the FVIII to be presented complexed with an MHC class II molecule to CD4+ T cells and it is important that this peptide-MHC II complex is stable. Pandey and colleagues employed ProImmune’s REVEAL® binding assay to measure stability of complexes of common MHC class II haplotypes with FVIII peptide fragments from the polymorphic regions of the F8 gene and used this to predict sequences that are likely to be immunogenic.
ProImmune’s REVEAL® binding and stability scores can be used to identify potential T cell epitopes
ProImmune’s REVEAL® binding assay is a rapid in vitro assay that measures peptide-MHC II complex stability and can therefore be used to identify potential epitopes. The bound complex is recognised by a fluorescently-labelled monoclonal antibody that only binds to the MHC class II molecule if there is bound peptide. The assay is available for 26 HLA-DR, 11-DQ and 19-DP alleles as well as mouse class II alleles H-2IAb and H-2 IAd making it useful for work with mouse haemophilia models. ProImmune REVEAL® is a single-step assay that determines the HLA restriction of an immune response without consuming precious patient samples. By comparing the binding results with negative and positive controls, peptide sequences most likely to be immunogenic can be identified. ProImmune REVEAL® Class II Complete Rate Assay can then be used to measure on and off rates for each complex. A slow off rate for a peptide with an intermediate score from the REVEAL® binding assay indicates a potentially significant epitope.
In the first experiment performed by Sauna’s team (shown in figure 1) MHC II-peptide stability was compared between well characterized T cell epitopes from FVIII that have been associated previously with inhibitor formation (positive controls) and negative controls where there is no inhibitor formation. 6 common MHC class II alleles were used. It was found that the median binding for the positive controls was significantly higher than for the negative controls thus demonstrating that complex stability as determined by the REVEAL® assay could be used as a marker for inhibitor formation. These results included data on three specific missense mutations Y205C, R2150H and W2229C that have been associated with inhibitor formation in 30-40% of patients (A).
Figure 1. A Shows the prevalence of inhibitors in patients with three particular missense mutations Y2105C, R2150H and W2229C (blue). This is significantly greater than the overall prevalence of inhibitors averaged across all HA patients with missense mutations (grey). ProImmune REVEAL® Class II HLA-peptide binding assays (Fig B) and REVEAL® Class II Complete Rate Assay measuring on/off rates (Fig C) showed that there was a correlation between prevalence of inhibitors and the formation of a more stable FVIII peptide-MHC II complex. 6 common MHC II alleles were used (DRB1*03:01, DRB1*04:01, DRB1*07:01, DRB1*11:01, DRB1*15:01 or DRB1*15:03)PC – positive control (known T cell epitopes), NC – negative controls.
Potential FVIII epitopes associated with the polymorphic regions of F8 gene have been predicted using ProImmune REVEAL® Class II HLA-peptide binding and stability assays and may explain the the prevalence of inhibitors in patients of black African descent
Sauna’s team went on to use the ProImmune REVEAL® Class II HLA-peptide binding assay to measure the binding between 30 FVIII peptides (covering the locations of the common polymorphisms – sites of mismatch between patient endogenous FVIII and the rFVIII) and the 6 common MHC-II alleles mentioned above. The study revealed significant binding of at least one peptide from each of the polymorphic regions of FVIII to several MHC class II variants with many of these peptides having affinities greater than the positive controls (Figure 2). Some peptides bound up to 83% of MHC-II alleles.
Figure 2 – (A) Binding scores for FVIII 14-amino acid peptides spanning the ns-SNP region. Each row represents a peptide and each column represents an MHC-II allele. Peptides are grouped based on position in the FVIII gene (haplotypes H2-H6). Positive-control peptides (PC) are known T-cell epitopes of FVIII and negative control (NC) peptides are from regions of the FVIII protein where no T-cell epitopes have been identified. Dark colours represent peptides that bind with high affinity to the MHC-II allele. (B) shows peptides with binding scores ≥ 15% – these are considered to be potential T-cell epitopes for the MHC-II variants shown. (C) shows the binding promiscuity score for each peptide shown in Fig B and indicates the fraction of MHC-II variants that each of these peptides binds to.
ProImmune REVEAL® Class II REVEAL® Class Complete Rate Assays were used to measure on and off rates for each complex because stability of complexes has been shown to be an important factor for predicting immune response. More stable complexes were generated with test-peptides compared to the negative controls (Figure 3)
FIG 3. ProImmune REVEAL stability scores. (A) Normalized half-life times for peptides with each peptide-MHC-II were determined. Each row represents a peptide and each column an MHC-II allele. (B) Stable (stability score≥6) FVIII peptide- MHC-II complexes. (C) Binding promiscuity scores based on stability (the fraction of MHC-II alleles each peptide binds to with a score ≥6 is shown).
This study underlines the importance of testing candidate FVIII therapeutic peptides on as wide a selection of MHC Class II haplotypes and F8 gene variants as possible to minimize the risk of inhibitor formation in Haemophilia A patients. This is particularly significant since current rFVIII therapeutics have tended to be based on F8 gene variants that predominate in the Caucasian population (probably giving rise to a higher incidence of inhibitors in patients of black African descent). ProImmune REVEAL® Class II assays are particularly useful at addressing this need because they (i) enable rapid high-throughput screening of putative epitopes, (ii) determine HLA-restriction in one step and (iii) the results are more reliable than in silico prediction methods (in general computational tools to predict peptide-MHC binding are less accurate for Class II than Class I). Use of the REVEAL® assay for the in vitro screening described allow just the most promising candidate drugs to be taken forward into functional assays such as T cell proliferation.