Abstract
Patients with sickle cell disease have an increased risk of bacteremia. To study the genetic basis for this increased susceptibility we studied the association of single nucleotide polymorphisms (SNPs) in candidate genes that might affect the risk of infection and therefore bacteremia. The Cooperative Study of Sickle Cell Disease enrolled 4,082 patients who were observed for about 5 years. Blood from these patients was used for globin gene analysis and SNP genotyping. We limited the present studies to patients with sickle cell anemia, with or without coincident α thalassemia, yielding a pool of 1473 patients with genotype, demographic and clinical data for which DNA samples were available. A case was defined as a patient seen in a clinic, emergency department, or hospital for which a positive blood culture not associated with a known source of infection such as osteomyelitis, septic arthritis, pneumonia, or meningitis, was found. Control patients had no history of bacteremia and no incident bacteremic events. For genetic studies, the samples were genotyped for informative SNPs in candidate genes selected from public and proprietary databases using the Sequenom mass spectrometry SNP genotyping system. For quality control purposes about 3% of the DNA samples were regenotyped and Hardy-Weinberg equilibrium was assessed for each SNP among controls. Genotypic counts were compared between sickle cell patients with bacteremia and patients using multiple logistic regression adjusting for leukocyte count, penicillin prophylaxis, fetal hemoglobin and total hemoglobin levels. In our initial screen, we considered a SNP to have an association with a phenotype when the p-value was equal to or less than 0.01, or if this and other SNPs in the same gene were significant at the 0.05 level. If a SNP met these criteria, a second phase of genotyping was done to study additional haplotype tagging (ht) SNPs in the gene. Because of the importance of the TGF-β beta pathway in the immune response, additional genes in this pathway were also studied. The ABI SNPlex system was utilized for the second phase of genotyping. Among the 201 subjects with bacteremia and 1238 controls, there was no significant difference in age, sex, HbF concentration, distribution of β-globin gene cluster haplotypes or the presence of coincident α thalassemia. Patients with bacteremia had a slightly lower hemoglobin concentration and higher leukocyte count. Subjects who received antibiotic prophylaxis (p < 0.0001) or H. influenza vaccination (p <0.004) were more likely to develop bacteremia, but with a ‘non-covered’ organism. Four SNPs in BMP6 (rs270387, rs267188, rs267196 and rs366386; p values < 0.039), 2 SNPs in TGFBR3 (rs2148322 and rs2765888; p values < 0.033) and 2 SNPs in SMAD3 (rs10518707 and rs11631380; p values < 0.012) were associated with bacteremia. The TGF-β/BMP pathway modulates immunosuppression, cell migration, wound healing and angiogenesis, among other functions. In some studies, reduced levels of these proteins were associated with increased severity of bacterial pneumonia. SNPs that show an association with susceptibility to bacteremia may help target sickle cell anemia patients who require more prolonged antibiotic prophylaxsis.
Author notes
Corresponding author
