Haemophilia A, factor VIII intron 22 inversion screening using subcycling-PCR.

Thromb Haemost 2006; 95: 746–7 Dear Sir, Haemophilia A is an X-linked recessive disorder caused by mutations in the Factor VIII (FVIII) gene and leads to a deficiency of blood coagulation. One in 5000 males is affected by haemophiliaA, and approximately half of all severe cases are caused by a gene rearrangement involving intra-chromosomal homologous recombination between a 9.5 kb region of intron 22 (Int22h1) and two extragenic homologues (either Int22h2 or Int22h3) located near the Xq telomere (1–4). Liu, et al. (5) designed a multiplex PCR in which primers A+B amplified int22h-2 or int22h-3 (10 kb) and P+Q amplified int22h-1 (12 kb) in normal individuals. Intron 22 inversions in affected or carrier individuals were indicated by the presence of an 11 kb band following amplification with primer combinations A+Q or P+B. While this assay is attractive in its design, consistent and reliable amplification of bands within this region still proves difficult with high background smears, no amplification or preferential amplification of lower molecular weight bands within the multiplex, and occasionally the appearance of a non-specific 11 kb band (6, 7). Detection of intron 22 inversions by inverse PCR (8) or indirect polymorphism analysis (9) have also been described but require greater amounts of DNA, and appear more labour intensive than the assay described. Inverse PCR requires a four hour BclI restriction digest, an overnight self-ligation of restriction fragments and purification of products before multiplex PCR analysis (8), and indirect polymorphism analysis uses single primer pairs but involves restriction enzyme digestion in addition to nested PCR (9). Modifications to FVIII Intron 22 inversion testing using subcycling-PCR (S-PCR) (6) with individual primer pairs (A+B, P+Q, A+Q and P+B) as opposed to one reaction with four primers, as recommended by Polakova, et al. (7), provide a more reliable and reproducible assay for the detection of intron 22 inversion mutations. For diagnostic purposes, however, only two of the four original sets of primer pairs (P+Q and P+B) were found sufficient to give a conclusive result (Fig. 1). Primers E5S2 and E9AS, which amplify a 6.8 kb region of the MLL gene (10), were also included to control for the presence of high molecular weight DNA within each sample and substantially reduced the time required for band resolution during electrophoresis. Although the MLL control gene is outside of the region encoding FVIII and the telomeric intron 22 homologues, fewer primers involved in the amplification of the GC-rich and highly homologous Int22h regions (6) may be advantageous and decrease the occurrence of non-specific amplification. PCR reactions were set up using two reaction mixes. Mix I contained 0.2 μM forward and reverse inversion 22 primers, 0.02 μM MLL primers, 200 μM 7-deaza-dGTP (Roche, Mannheim, Germany), 310 μM dGTP and 500 μM of each of the other dNTPs (Roche, Mannheim, Germany), and 7.5% dimethylsulphoxide (DMSO) in a total volume of 6.7 μl. Mix II com-