Differential identification of PT‐VWD from type 2B VWD and GP1BA nomenclature issues

Abstract

I read with interest the recent letter by Favaloro et al (2007), reporting that a simplified ristocetin-induced-platelet-agglutination (RIPA) mixing assay correctly identified three platelet type von Willebrand disease (PT-VWD) individuals in which the phenotype was later confirmed by genetic analysis. Although we should await responses from other hemostasis laboratories following the implementation of the assay in their laboratory, this assay has superiority over the cryoprecipitate challenge and standard RIPA because it provides simultaneous answers regarding the two disorders. Undoubtedly, the genetic approach supported by us and others (Whalley & Perry, 2007; Favaloro, 2008) remains the gold standard for identifying the gene mutation in VWF vs. platelet GP1BA genes. Furthermore, the potential misdiagnosis of PT-VWD as 2B VWD can only be revealed with large molecular studies. We recently proposed an international PT-VWD project and also called for a PT-VWD online database/disease registry (Othman, 2007; Othman & Lillicrap, 2007) to help address questions related to this diagnostic dilemma; the latter is now available at http://www.pt-vwd.org. Favaloro et al (2007) referred to the previously reported Gly233 Ser as Gly246Ser. This raises a critical point about the nucleotide and amino-acid (aa) numbering of GP1BA as well as mutation description. The human gene nomenclature committee (HGNC) (http://www.genenames.org/) that stems from the human genome organisation (HUGO) and the affiliated human genome variation society (HGVS) (http://www.hgvs.org/) recommend that nucleotide numbering should start at the ATG translation codon and all nucleotides upstream of this A should be numbered as −1 and the aa numbering should start from the first Met of the polypeptide chain. The GP1BA cloned in 1987 (Lopez et al, 1987) consists of two exons with 42 nucleotides of 5′ non-coding sequence. The entire coding sequence is contained within exon 2, the first 48 nucleotides of exon 2 codes for a 16 aa signal peptide and the mature protein starts at aa His. Accordingly, the PT-VWD mutation historically known as 788 G>A; Gly 233 Ser should ideally be referred to as the 746 G>A; Gly 249 Ser, as described by Favaloro et al (2007). Likewise, other GP1BA mutations/polymorphisms will have different nomenclature, as shown in Table I. Regarding the name of the platelet gene/protein, we support the name GP1BA (1 arabic numeral instead of Latin) as listed for this gene in the HGNC website (ID:4439), italicised when referring to the gene and non italicised when referring to the protein. Similarly, in the VWF gene, different Investigators have used different mutation nomenclature based on whether numbering starts from the first Met or the first aa of the mature protein (Ser 763) which have sometimes created confusion. Examples are: Cys 1149 Arg (Cys 386 Arg) and Cys 1130 phe (Cys 367 Phe), Type 1 Malmo/New york (Weiss & Sussman, 1986) Pro1266 Leu (Pro 504 Leu). It is worth noting that two recently reported type 1 VWD studies (Goodeve et al, 2007; James et al, 2007) have followed the standard nomenclature and that the VWD database (http://www.vwf.group.shef.ac.uk/) has been updated to accommodate the new nomenclature. Following a standard nomenclature is critically important for consistent description of gene sequence variations in order to reduce the confusion among clinicians and researchers regarding the diagnosis of genetic disorders. However, we realize it takes effort and great deal of awareness to consistently implement this nomenclature universally.