Von Willebrand disease (VWD) is a heterogeneous bleeding disorder characterized by quantitative (type 1 and 3) and qualitative (type 2) abnormalities of von Willebrand factor (VWF) (Sadler et al, 2006). VWD type 3 is the most severe form and has a prevalence of 0·55–3·2 per million in Western countries (Mannucci et al, 1984) to 6·0 per million in Iran (Lak et al, 2000). It is characterized by a virtually complete deficiency of VWF, and by severe haemorrhagic symptoms. Type 3 VWD inheritance is autosomal recessive. Patients with type 3 VWD are either homozygous or compound heterozygous for two defective VWF alleles. Heterozygous carriers are generally asymptomatic and occasionally may show a mild phenotype similar to mild type 1 VWD. Nevertheless, it has been confirmed that obligatory carriers for type 3 VWD represent a distinctive population from type 1 (Castaman et al, 2006). Different mutation types, such as frameshift, nonsense, splice site, and missense mutations have been detected in severe patients (http://www.vwf.group.shef.ac.uk/). Inhibitor development after replacement therapy with factor VIII/VWF concentrates is infrequent, and initial reports were associated with large gene deletions. The present study aimed to ascertain the genetic defects of two severe VWD patients, the propositus and the son of a first cousin, and to carry out a family study. DNA samples from the patients and other six relatives were obtained after informed consent. The patients showed undetectable VWF antigen (VWF:Ag), ristocetin cofactor (VWF:RCo), and VWF collagen-binding capacity (VWF:CB) levels, whereas their parents showed borderline normal values. The propositus developed alloantibodies with an inhibitor titre of 4·7, 5·3 and 5·7 Bethesda units/ml against VWF:Ag, VWF:RCo, and VWF:CB, respectively. The 52 exons and their intron/exon boundaries of the VWF gene were amplified by polymerase chain reaction (PCR). In both patients only the regions from exon 1 to exon 14, and from exon 44 to exon 52 could be amplified. The PCR technique failed to amplify from exon 15 to exon 43, indicating that these exons had undergone an homozygous deletion. However, all exons were successfully amplified in the parents, the sister of the propositus and a non-carrier relative. While the lack of amplification indicated a large deletion of the corresponding exons in the patients, the carrier diagnoses in this family were difficult because the presence of a normal allele hinders identification, by conventional amplification methods, of a complete or partial loss of one copy of the VWF gene used. Recently, methodologies based on gene dosage, such as multiple ligation-dependent probe amplification (MLPA), have been successfully used to detect carriers in a large number of disorders including haemophilia (Casaña et al, 2009). However, this is the first report of carrier detection in VWD.
Two salsa MLPA VWF kits, PO11-B1 and PO12-B1 (MRC-Holland, Amsterdam, the Netherlands), were employed following the manufacturer’s recommendations. A first normalization intra-sample was prepared by dividing the peak area of each probe by the total area of the control probes. A second normalization inter-sample was prepared by dividing the previous results by the average of the normalized results from three or more control individuals (Fig 1A). The patients’ data showed an almost total reduction of the area of the peaks corresponding to exons 16–43 (Fig 1B). The area of these same peaks was around 50% in 5 relatives (Fig 1C), including the patient’s parents and sister, showing their heterozygous state. These results apparently were in disagreement with the PCR results. While there was a lack of exon 15 amplification by PCR in the patients, MLPA produced normal data for probe 15 in all family members including both patients. This suggested that the breakpoint of the deletion was located behind the exon 15 probe and before the exon 15 reverse primer used for PCR. Keeping in mind that the maximum and minimum expected size of mutated allele should be from 391 to 4845 bp, the genomic region that includes exons 15–44 was amplified by long and accurate (LA) PCR technology in order to determine the deletion breakpoints. A PCR product of approximately 1636 bp was obtained in both the patients and the carriers. The sequencing of this product showed that the deletion covered from the 5′ end of intron 15–3266 bp within intron 43. These results could explain the discordant data between PCR and MLPA because intron 15 breakpoint was located just within the nucleotide sequence annealing the reverse primer. In summary, the patients have a large deletion of approximately 84 Kb and all carriers share the same deletion in the heterozygous state.
Following VWF gene cloning, the first genetic defects identified in VWD were large gene deletions detected by Southern Blot method in a minority of type 3 patients (Ngo et al, 1988). Large deletions in VWF can be detected in homozygous patients with VWD by an absence of PCR products of the deleted fragment. However, in relatives carrying VWF gene large deletions the heterozygous status cannot be recognized by conventional PCR methods due to the amplification of exons present in the normal chromosome. Segregation analysis may also suggest a possible deletion if it does not fit with a Mendelian inheritance pattern, showing, for example, an heterozygosity loss of several polymorphisms in genomic DNA (Eikenboom et al, 1998). We here report a type 3 VWD family whose patients are homozygous for a large deletion of exons 16–43 in the VWF gene. It is the first time that MLPA has been used to identify a large deletion in VWF, both in patients and heterozygous carriers. This method allowed us to offer more suitable genetic counselling to all members of this family. In our experience, MLPA is a useful tool for determining the dosage of the 52 exons of VWF, and may be very helpful for identifying possible insertions/deletions carriers when genomic DNA is analysed. Although complete or partial gene deletions are infrequent in VWD, the use of MLPA prior to direct sequencing of the 52 exons of VWF may identify insertions or deletions that are undetectable by standard methods.