Lack of association of the heparanase gene single-nucleotide polymorphism Arg307Lys with acute lymphoblastic leukaemia in patients from Northern Ireland

Heparan sulphate is a key structural component of the extracellular matrix (ECM) that contributes to the self-assembly and integrity of the matrix by binding structural proteins including collagen, laminin and fibronectin. The endoglycosidase enzyme, heparanase, participates in the degradation and remodelling of the ECM by cleaving heparan sulphate molecules to yield bioactive fragments that modulate growth factor activity. Heparanase activity also induces an angiogenic response by releasing angiogenic factors sequestered by heparan sulphate in the ECM. In contrast to the various enzymes that degrade protein components of the ECM, a single mammalian heparanase appears to be responsible for degrading heparan sulphate. As such, expression of heparanase activity is associated with a range of normal and pathological processes that involve degradation and remodelling of the ECM including wound healing, inflammation, neovascularization and tumour metastasis.1 Expression of heparanase in normal tissues is restricted to a few cell types including endothelial cells, keratinocytes and placental trophoblasts. In the haematopoietic system, heparanase is expressed in platelets, neutrophils, macrophages and activated lymphocytes, where it mediates extravasation during inflammatory and immune responses.2 In contrast to normal tissues, heparanase is widely expressed in primary human tumours where high levels of expression are correlated with increased local and distant metastasis, increased tumour vascularization and reduced post-operative survival in cancer patients.1, 3 In animal models, overexpression of heparanase confers accelerated tumour growth and an invasive phenotype.1 Conversely, heparanase gene silencing is associated with a marked inhibition of tumour progression.4 Expression of heparanase has been detected in blasts of human leukaemias, including acute myeloid leukaemia and acute lymphoblastic leukaemia (ALL); however, its involvement in haematological malignancies has not been systematically assessed.5, 6 Ostrovsky et al.7 have recently shown that a single-nucleotide polymorphism (SNP) of the heparanase gene (rs11099592), which causes the substitution of arginine with lysine at position 307 (Arg307Lys) of the heparanase polypeptide, was significantly associated with ALL. They reported that genotype and allele frequencies of the SNP in a group of 43 Israeli ALL patients were significantly different from a group of 103 healthy control individuals ( 22 = 6.384, P=0.041; for genotype comparison and 12 = 4.96, P=0.026 for allele comparison). In addition, levels of heparanase expression in bone marrow or blood of 18 ALL patients were significantly lower compared to 31 control individuals (RQ values: 3.281.53 vs 8.812.23; P<0.0001), and a haplotype of three heparanase SNPs (rs4693608; rs11099592; rs4364254) was identified that was associated with reduced heparanase expression in control individuals. The authors concluded that heparanase gene SNPs may influence basal expression of the gene and that the Arg307Lys SNP is an important determinant in the pathogenesis of ALL.7 We sought to confirm the association of the Arg307Lys SNP with ALL reported by Ostrovsky et al.7 by comparing the genotype and allele frequencies of the SNP in a population of Northern Irish ALL patients with a group of healthy control individuals. We speculated that it would be of interest to determine if the findings reported for the Israeli population extended to other, less heterogeneous ethnic groups such as the Northern Irish. Genotype and allele frequencies for the Arg307Lys SNP were compared in a group of 58 Northern Irish ALL patients (32 male patients, 26 female patients; age: 2–70 years) and 45 healthy control individuals (23 male subjects, 22 female subjects; age: 2–65 years). A 273 bp fragment of the heparanase gene spanning the site of the rs11099592 SNP was amplified by PCR. Digestion of the PCR product with the BstNI restriction enzyme yielded diagnostic bands of 192 bp indicating the Arg allele and 257 bp indicating the Lys allele.7 In contrast to the findings of Ostrovsky et al.,7 a comparison of genotype and allele frequencies for the Arg307Lys SNP showed no significant differences between the Northern Irish ALL patients and the healthy control group ( 22 = 2.866, P=0.260 for genotype comparison and 12 = 2.629, P=0.105 for allele comparison). Furthermore, a comparison of our data with those reported by Ostrovsky et al.7 revealed substantial differences in the genotype and allele frequencies for the Northern Irish and Israeli ALL and control populations (Table 1). The lack of concordance between genotype and allele frequencies in the Northern Irish and Israeli populations may suggest ethnic variation in the frequency of Arg307Lys SNP. The Israeli population studied by Ostrovsky et al.7 included individuals from a number of ethnic groups including Ashkenazi, North African, Mediterranean and Near Eastern Jewish populations. In a previous study by the same group, significant differences between these populations in the genotype and allele frequencies of several heparanase SNPs were identified, although these did not include the Arg307Lys SNP.8 It is of interest to note, however, that the allele frequencies for the Arg307Lys SNP reported by Ostrovsky et al.8 in this earlier study are strikingly different from those reported in their subsequent study7 and bear a close similarity to our data on the Northern Irish population (Table 2). All of the patients and control individuals included in our study were of Northern Irish origin. Due to its geographical isolation and historically low levels of inward migration, the Irish population may be considered to be genetically more homogeneous relative to other populations, including the Israeli population studied by Ostrovsky et al.7 As such, the influence of genetic factors predisposing to complex conditions such as ALL may be more apparent in our population. In conclusion, our findings on Northern Irish ALL patients have failed to confirm the association between the heparanase Arg307Lys SNP and ALL previously reported for the Israeli population. This may be related to the heterogeneous nature of the Israeli population studied by Ostrovsky et al.7 Further studies may provide clarification on the precise prevalence of the Arg307Lys SNP in the Israeli Jewish population.