Thrombocytopenia-absent radii syndrome (TAR) is a rare congenital disorder characterized by bilateral radius aplasia and thrombocytopenia due to virtual absence of bone marrow megakaryocytes (MKs). As other blood lineages are not affected, TAR is considered to be a bone marrow failure syndrome. At birth, platelet counts are often below 50/nL, but ameliorate in many patients during the first two years of life, however, without reaching the lower norm. Platelet reactivity in reponse to thrombopoietin is abrogated, mostly when patients are young while it is restored to normal in older patients. Interestingly, this patterning is not correlated with platelet counts. In 2007 we described a microdeletion on chromosome 1q21 in all patients analyzed as necessary, but not sufficient to cause TAR, as it was also present in unaffected family members. Using next generation sequencing, we recently identified the presence of one of two single nucleotide polymorphisms (SNPs) in non-coding, regulatory regions of the gene RBM8A whose compound inheritance together with the microdeletion causes TAR syndrome (Nature Genetics, 44, 435). All patients inherited one of the two SNPs from one of one parents and the microdeletion from the other. In 25% of cases the microdeletion was acquired de novo. This gene, which is located within the microdeleted region, encodes for Y14, a major subunit of the exon junction complex. In the German/Swiss/Polish cohort of 28 patients, 20 patients had the more frequent SNP in the 5'UTR, whereas 8 patients had a previously undescribed SNP in intron 1. The 5'UTR G->A introduces a novel binding site for transcriptional repressor Evi1. This factor is expressed in megakaryocytes and still present in platelets. Using electrophoretic mobility shift assays (EMSA) with lysates from platelets we found a complex exclusively present when a labeled oligonucleotide with the 5'UTR-SNP G->C was used and not with the wildtype sequence. This complex could be partially supershifted with an anti-Evi1 antibody, suggesting that an additional factor might bind to this complex. In contrast, the intronic SNP is predicted to abrogate a binding site for Mzf1. We used EMSA with oligonucleotides of both intronic SNP and wildtype sequence and indeed found a complex binding to the wildtype sequence as well as to a consensus sequence for Mzf1, but no or reduced binding to the intronic SNP. This complex could not be supershifted by adding an anti-Mzf1 antibody. However, when nuclear extracts from Meg01 cells were used, we found that one of three subcomplexes showed reduced binding in the presence of the antibody, even when the intronic SNP was used. As both SNPs showed different binding properties we carefully evaluated clinical data and found that platelet counts were more severely reduced in patients with the 5'UTR compared to patients harboring the intronic SNP. Although case numbers for the intronic SNP are low, in a direct comparison of age-matched patients, platelet counts were always higher in TAR patients with the intronic SNP. However, we did not find any obvious differences in skeletal features between the two groups. Further work is ongoing to better understand the causative role of either SNP for impaired megakaryopoiesis and thrombocytopenia in TAR syndrome.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.