A polymorphism in the angiogenesis inhibitor, endostatin, in multiple myeloma

The past decade has seen an increasing interest in the dependence of the growth of multiple myeloma (MM) on angiogenesis [1]. Thus, a possible relationship between some angiogenic factors such as vascular endothelial growth factor, the fibroblastic growth factor, and the hepatocyte growth factor as well as cytokines such as interleukin-6, have also been reported [1]. They may account for the increased microvessel density that has been observed in the bone marrow of patients with MM and that correlate with disease progression and poor prognosis [2].

In addition to producing proangiogenic cytokines, solid tumours also play a role in the generation of endogenous antiangiogenic proteins, such as endostatin [3] but it is unknown whether this protein may be produced by neoplastic plasma cells. Endostatin is a 20kDa C-terminal fragment of collagen XVIII, the product of the COL18A1 gene. Higher serum levels of endostatin induced experimentally in mice caused regression of leukaemia and solid tumours [4]. In addition, Down’s syndrome patients have a decreased incidence of solid tumours possibly due to the high serum levels of the protein produced by their three copies of the COL18A1 gene [5]. Thus, different levels of endostatin seem to be associated with varying susceptibility to tumour development.

Furthermore, a COL18A1 gene polymorphism (D104N) located in the COOH-terminal globular domain, NC1, of collagen XVIII, the encoding region for endostatin was recently associated with increased risk for the prostatic adenocarcinoma, which was attributed to an impairment in the protein function [6].

Considering that it is unknown whether D104N polymorphism alters the risk for MM, this was the aim of this study. For this purpose, genomic DNA from the bone marrow samples of 58 MM patients (31 men, 27 women; mean age±S.D.: 57.5±11.4 years), seen at the University Hospital, and 300 controls (197 men, 103 women; mean age±S.D.: 53.9±2.9 years) were analysed using the polymerase chain reaction (PCR) followed by restriction endonuclease digestion with Mse I (Fig. 1).

View Large Image Download to PowerPoint

Fig. 1. PCR and restriction endonuclease digestion for detection of D104N polymorphism of the COL18A1 gene in multiple myeloma patients. Brometo-stained 12% polyacrylamide gel showing fragments of 169bp corresponding to the absence of D104N polymorphism and fragments of 101bp and 68bp corresponding to the presence of D104N polymorphism. Lanes 1 and 2 show the DNA size marker 100bp ladder and 25bp, respectively. Lane 3 shows the product of PCR amplification. Lanes 4–10 show the results from individuals with D104N polymorphism in heterozygosity. Lane 11 shows the result from individual with D104N polymorphism in homozygosity. Lanes 12–14 show the results from individuals without the gene polymorphism.

Both the patients’ and controls’ samples were in Hardy–Weinberg equilibrium (X2=0.61, P=0.43, X2=1.69, P=0.19, respectively). Similar frequencies of heterozygotes were observed for the D104N polymorphism in MM patients and controls (17.2 and 14.0%, respectively; P=0.66). Similar risks for the disease were also seen in individuals with D104N polymorphism in comparison with those without the gene polymorphism (OR=1.45, 95% CI: 0.60–3.51; P=0.41), after adjustment by age and sex.

When MM patients alone were considered, no differences in the frequencies of D104N polymorphism were found according to age (7 of 32 patients under 60 years versus 3 of 26 patients in older age; P=0.49), gender (5 of 31 males versus 5 of 27 females; P=1.0), ethnic origin (6 of 42 caucasians versus 4 of 15 blacks; P=0.43), and Durie and Salmon’s stage (3 of 18 patients at stages I+II versus 5 of 34 patients at stage III; P=1.0).

In conclusion, our results present preliminary evidence that the D104N polymorphism of the COL18A1 gene may be an unimportant determinant of the MM susceptibility.