Abstract 1540

Poster Board I-563

Tumor necrosis factor alpha (TNFα) is a pro-inflammatory cytokine that stimulates phagocytosis, neutrophil recruitment, and expression of adhesion molecule VCAM-1. Plasma levels of TNFα have been found to be increased in sickle cell disease (SCD), and in vitro studies show that TNFα causes increased adherence of sickle red blood cells to human umbilical vein endothelial cells. A polymorphism in the promoter region of the TNFα gene has previously been associated with stroke in children with SCD (Hoppe et al., 2007). The current study was designed to identify associations of additional TNFα single nucleotide polymorphisms (SNPs) with SCD clinical complications. We analyzed five SNPs in the TNFα gene in 509 DNA samples of SCD patients from Duke University, University of North Carolina at Chapel Hill, and Emory University. In our data set, cerebrovascular events (CVEs), including overt stroke, seizures, and transient ischemic attacks, occurred in 133 out of 509 SCD patients (26.1%). SNP genotyping was performed using Taqman genotyping assays from Applied Biosystems. Due to low minor allele frequencies (<0.05) for all the SNPs examined, genetic associations with SCD clinical complications were examined by using allele tests. After controlling for age, gender, and use of hydroxyurea, two of the five TNFα SNPs, rs2228088 and rs3093665, were significantly associated with CVEs (p=0.013 and 0.029, respectively). The odds that SCD patients with a G allele at rs2228088 suffered from CVEs were 0.485 times that for patients with the T allele, suggesting that the G allele had a protective effect. The odds of having the A allele at rs3093665 and suffering from CVEs was also reduced, at 0.45 compared to the C allele. Neither SNP was found to be in linkage disequilibrium (LD) with any of the other SNPs analyzed (r2≤0.002). There was also strong association of SNP rs2228088 with acute chest syndrome (ACS; p=0.003), occurring in 382 out of 509 SCD patients (75%). However, in this analysis, the G allele was associated with increased risk for ACS (OR=2.313). In addition to the association with CVEs, the SNP rs3093665 was also significantly associated with priapism (p=0.03), reported by 86 of 223 male SCD patients (38.6%). In this analysis, the A allele was protective, as had been observed for CVE (OR=0.188). Additionally, we found no difference in steady state plasma TNFα levels between genotypes for the two SNPs. The functional significance of these SNPs is presently unknown. SNP rs2228088 is a synonymous SNP located in the coding region, and rs3093665 is located in the 3' untranslated region of the TNFα gene. While the G to T change at SNP rs2228088 does not translate to a change in amino acid sequence, the A to C change at SNP rs3093665 may affect mRNA stability due to its location. It is also possible that one or both of these SNPs is in LD with another functionally relevant SNP. Our findings thus support previous data implicating TNFα polymorphisms in risk for central nervous system events. Interestingly, ACS has been previously associated with seizures, stroke and altered mental status in adults and children with SCD (Vinchinsky et al., 2000) and with silent cerebral infarcts and reversible posterior leukoencephalopathy syndrome in children with SCD (Henderson et al., 2003). However, in our dataset, ACS and the occurrence of CVEs were not associated (p=0.24). Further studies are required to elucidate these and other factors that potentially correlate with SCD clinical complications.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution