Von Willebrand disease (VWD) is the most commonly inherited bleeding disorder with a symptomatic prevalence of 1:1000. Type 1 VWD comprises ∼80% of VWD cases and is defined by mild to moderate deficiencies of the large multimeric glycoprotein von Willebrand factor (VWF). Sequence variation in consensus splice sites is a common culprit in the induction of aberrant splicing, and occurs in approximately 10% of type 1 VWD cases. Investigation of the VWF mRNA is required in order to determine whether these sequence variants are causative mutations and the manner in which they affect splicing. In this study, aberrant VWF splicing from a consensus splice site mutation was investigated using mRNA from patient platelets and patient-derived blood outgrowth endothelial cells (BOEC). Both aberrant and alternative VWF splice forms were identified and their effects were further characterized through expression studies in HEK293T cells as well as further investigation of the patient BOEC.
Patient T152 presented with a type 1 VWD phenotype: VWF:Ag 0.49IU/mL, VWF:RCo 0.42IU/mL, FVIII:C 1.16IU/mL and an abnormal bleeding score of 8. Direct sequencing of the patient‘s VWF gene revealed c.5842+1G>C in intron 34. This variation was thought to cause the type 1 VWD phenotype through aberrant splicing. Reverse transcription of the VWF mRNA detected different splice variants within the platelet and BOEC RNA. Wildtype VWF mRNA as well as transcripts encoding in-frame double exon skipping of exons 33 and 34 were found in both RNA sources. An additional VWF transcript which skipped exon 33 and introduced a premature termination codon in exon 34 was present in the BOEC RNA. Sequencing of this transcript determined it was produced from the patient's normal allele and may be a product of alternative splicing. qRT-PCR quantification determined the patient VWF mRNA to be 51±5% wildtype, 28±4% exon33–34 skipped and 21±9% exon 33 skipped transcripts. The skipping of exon 33 was confirmed as an alternative transcript comprising 12±8% of VWF mRNA in 4 normal individuals.
VWF expression vectors were created through site-directed mutagenesis for all three of these transcripts. Calcium-phosphate mediated transfections of these mutant vectors into HEK293T cells with varying wildtype:mutant ratios (100:0, 90:10, 75:25, 50:50, 0:100) for both mutant transcripts, as well as in the biological ratio (51:28:21) determined as above from the patient qRT-PCR. The alternative exon 33 skipping transcript does not decrease VWF secretion compared to wildtype (p>0.5). The presence of even 10% of the double exon skipping mutant transcript decreased secretion of VWF to 62± 1%(p<0.05) of wildtype, demonstrating a dominant negative effect. Similar to the plasma VWF:Ag level of 0.49 IU/mL, VWF secretion decreased to 46± 8% (p<0.01) when transfected at biological mRNA ratios. When stimulated with PMA, the transfections mimicking the patient biological VWF mRNA ratio increased VWF secretion 22±10% which approximates the patient's clinical DDAVP response. A similar degree of increase was seen in PMA stimulated wildtype transfections. Multimer analysis of concentrated transfection media showed that the in-frame exon 33–34 is normally multimerized, which is consistent with the patient plasma multimers.
BOEC isolated from the patient enabled analysis of these VWF mutants ex vivo. Patient BOEC expressed 48±10% (p<0.01) less VWF relative to BOEC from normal individuals. Upon stimulus by thrombin and PMA for 24 hours, normal BOEC showed a 28±13% and 93±70% increase in VWF secretion respectively. Similar increases in secretion (35±11% and 120±32% respectively) were seen with T152's stimulated BOEC. Immunofluorescent staining of these BOEC showed abnormal morphology of Weibel-Palade bodies (WPBs). WPBs are characteristically cigar shaped; however, the patient's WPBs appear shortened and rounded relative to their normal counterparts.
We therefore conclude that the consensus splice site mutation c.5842+1G>C causes aberrant double exon skipping of exons 33 and 34 in this individual and is responsible for type 1 VWD through a dominant negative decrease in secretion. Additionally, the presence of VWF lacking exons 33 and 34 also affects the formation of WPBs. The alternative transcript which skips exon 33 was found to have no effect on VWF secretion and may be degraded by nonsense mediated decay.
James:CSL-Behring, Baxter, Bayer: Honoraria, Research Funding.
Asterisk with author names denotes non-ASH members.