Hemophilia A (OMIM +306700) is the most common inherited bleeding disorder, with an incidence of about 1:5000 male births. It is caused by mutations in the F8 gene, which comprises 26 exons and encodes the procoagulation factor (F) VIII – an essential cofactor in the intrinsic pathway of the coagulation cascade. The most prevalent mutation in the F8 gene is a genomic rearrangement, the intron-22-inversion, which is present in more than 40% of severe hemophilia A patients [1,2]. In addition, a wide spectrum of other causative mutations in the F8 gene has been published: missense and nonsense mutations, splice site mutations, small and large deletions, as well as insertions [3–8]. So far, only one duplication of a whole exon of the F8 gene, exon 13, has been described [9,10]. This may be an ascertainment bias because whole-exon duplications are difficult to detect by PCR-based methods, which are usually applied for mutation screening. Recently developed techniques for screening of quantitative genomic changes (copy number variations) significantly facilitate the detection of such rearrangements. Multiplex ligation-dependent probe amplification (MLPA; [11]) has rapidly found wide-spread application and has, for example, identified a large number of different duplications in the DMD gene causing Duchenne and Becker muscular dystrophies [12].

Over the past 15 years, a cohort of about 2000 patients suffering from mild, moderate or severe hemophilia A was referred to our laboratory for genetic diagnosis. DNA was isolated from peripheral blood samples using standard procedures. All patients were screened for mutations in the F8 gene by (i) long-range PCR for intron-22-inversions and intron-1-inversions, (ii) DGGE (denaturing gradient gel electrophoresis) for sequence alterations, and (iii) direct sequencing of all 26 exons and flanking intronic regions of the F8 gene. We could identify the causative mutations in all but 80 patients (about 96% detection rate). These 80 hemophiliacs comprise severe, moderate and mild cases and form the present study cohort.

DNA samples of the 80 hemophilia A patients were analyzed by MLPA (Kit P178, MRC Holland, Amsterdam, The Netherlands) for quantitative genomic changes in the F8 gene according to the manufacturer’s instructions. In brief, the F8 MLPA kit uses a mixture of 44 sequence-specific probes, which are hybridized to different target sequences within the F8 exons. Each MLPA probe consists of two adjacent oligonucleotides, which are ligated to each other after hybridization to their target sequence. Only ligated probes can be amplified in a single-tube PCR reaction using a universal primer pair that anneals to adaptors attached to the probe oligonucleotides. This results in the amplification of a set of exon-specific products between 130 and 476 bp in length. MLPA PCR products were separated on an ABI 310 automatic sequencer, preanalyzed by Genemapper V4.0 and evaluated by the Excel-based software Coffalyser V5.2, which normalizes the data in relation to control peaks from other genes and converts the peak areas into bar graphs. With a normal gene dosage, the ratio of control and test bars lies at 1.0 (variance: 0.8–1.2), in the case of a hemizygous duplication of an exon the corresponding peak is doubled, and in the case of a heterozygous duplication the peak should have a ratio of approx. 1.5.

Among the 80 patients screened by MLPA analysis for quantitative genomic changes, we could detect large duplications in nine patients (11%). Patients A and B showed duplications of single exons: exon 13 and exon 14, respectively. The other duplications affected more than one exon: exons 1 to 5 were duplicated in patient C, exons 5 to 25 in patient D, exons 23 to 25 in the unrelated patients E and F, exons 2 to 25 in patient G, exons 14 to 21 in patient H, while in patient I exons 7 to 11 were found to be duplicated (Fig. 1A). All duplications could be confirmed by quantitative DHPLC analysis (Fig. 1B). Therefore, a semi-quantitative multiplex PCR was performed by 25 cycles at touch-down PCR conditions (details on request), with each reaction comprising one or more exons expected to be duplicated, a human growth hormone (HGH) gene fragment as internal standard and one or more exons not involved in the duplication as PCR controls. Separation was done on an ion pair reverse phase C18 column at a semi-automated high throughput LC system (WAVE, Transgenomic Ltd, Omaha, NE, USA) (conditions on request). Navigator 1.5.3. software (Transgenomic Ltd) was used for data analysis. The chromatograms of patient samples were normalized to the HGH peak and superposed onto those of a normal control.


Figure 1.  MLPA (panel A) and DHPLC analysis (panel B) of the F8 gene of patients C, H and I. Solid black lines in DHPLC diagrams indicate wild-type graphs, dashed grey lines patient graphs. HGH, human growth hormone (control peak); c, control.

Download figure to PowerPoint

In the 80 hemophilia A patients studied, we could detect seven novel large duplications and one recently described single-exon duplication within the F8 gene.

Patient A suffers from a mild hemophilia A with 7% residual FVIII activity. His exon 13 duplication formally leads to an internal in frame duplication within the FVIII protein. This duplication has already been described in an Italian boy with mild hemophilia A, where it resulted from an unequal crossing-over between two misaligned normal chromosomes [10]. In a further study, this duplication could be shown to be present in 32% of northern Italian patients with mild hemophilia A [13]. MLPA analysis of the mother of patient A showed an exon 13 peak with a ratio of 1.5, which identified her as a carrier of hemophilia A.

Patients B to I all had severe hemophilia A, with less than 1% residual FVIII activities. Except for the duplication of exons 23–25, which was found in two unrelated hemophiliacs (patients E and F), all novel duplications detected are predicted to result in frameshifts of the F8 mRNA and protein sequence if they were re-arranged in a head-to-tail mode (tandem duplication). This is in accordance with the severe phenotype of the disease in these patients. Seven of the hemophiliacs with the severe subtype did not develop FVIII inhibitors, for one this is not known.

Not all large duplications are necessarily tandem duplications, depending on the mechanism of origin of the duplication. A duplication could in fact be a transposition of the duplicated part to a genomic position outside the affected gene, which would have no effect on the reading frame [12]. Such transpositions are not detectable by hybridization- and PCR-based methods like MLPA. However, because a transposition of a duplicated part should not compromise transcription and translation of the remaining intact F8 gene, it is unlikely to lead to a hemophilia phenotype.

Large duplications have been reported for a number of genes, for example Emery-Dreifuss muscular dystrophy (EMD), Charcot-Marie-Tooth neuropathy, and spinal muscular atrophy (SMA) (for a review see: [14]). They are thought to result from non-allelic homologous recombination (NAHR) between low-copy repeats (LCRs). Unequal crossing-over between Alu repeats has, amongst others, been described for the Duchenne/Becker muscular dystrophies (DMD/BMD) [15]. The intronic sequences of the F8 gene are very rich in Alu repeats, especially introns 22 and 25. Four of the nine duplications found in this study extend into intron 25 (duplication of exons 5–25, exons 2–25 and two times exons 23–25). Thus, Alu repeats may mediate the formation of these duplications. More complex mechanisms have also been reported; for example, in patients with Pelizaeus-Merzbacher disease tandem duplications are formed by a coupled homologous and non-homologous recombination mechanism [16]. Formally, a simple unequal crossing-over mechanism is expected to produce deletions and duplications at equal frequencies by a reciprocal recombination mechanism [17]. The hemophilia A database HAMSTeRS ( lists the deletions corresponding to all duplications reported here except for the duplications of exons 5–25 and exons 7–11. In the dystrophin gene, in contrast, large deletions are observed in about 65% of patients while duplications occur in only 10% of cases (Leiden database: The reasons for this divergence remain speculative at present.

In conclusion, nine of our 80 prescreened hemophilia A patients showed large duplications of one or more exons. This equals to approximately 0.5% of the 2000 hemophilia A patients studied by our group. Therefore, duplications of the F8 gene contribute to the recurrent low-frequency mutation types, like the inversion of intron 1 [18].

MLPA is a reliable, easy-to-use and fast method, which detects deletions and duplications in a quantitative way. As the whole F8 gene can be analyzed in one assay, we suggest applying MLPA prior to any screening method for point mutations in the F8 gene. Due to its quantitative nature, MLPA is the method of choice for the detection of heterozygous duplications and deletions in the F8 gene in potential female carriers of hemophilia A.

Disclosure of Conflicts of Interests

  1. Top of page
  2. Disclosure of Conflicts of Interests
  3. References

The authors state that they have no conflict of interest.


  1. Top of page
  2. Disclosure of Conflicts of Interests
  3. References