Abstract

Abstract 2068

HbF interferes with deoxygenated HbS polymerization, and is a major genetic modifier of sickle cell anemia severity. In this study, we attempted to identify genetic factors responsible for HbF production in a small group of African American sickle cell anemia patients who have markedly elevated HbF levels. Initially, patients with HbF of more than 11% as determined by HPLC were selected. The following were excluded: age less than 5; MCV greater than 100; presence of HPFH 1 and 2 based on specific gap-PCR tests; other β-globin gene deletions as determined by multiplex ligation-dependent probe amplification (MLPA). We further excluded rare HBG1 and 2 promoter HPFH mutations by nucleotide sequencing. In the end, a unique group of 20 patients were identified for further studies. Their mean age was 16.3 ± 8.3 years; Hb 9.0 ± 1.3 g/dL; MCV 87.9 ± 7.3 fL; and Hb F 17.2 ± 4.8% (range 11–29%). A control group of 30 African American patients were chosen. They had similar age, Hb, and MCV, but their HbF was 5.0 ± 2.5% (range 0.5–8.8%). These patients were examined for the 3 known major HbF quantitative trait loci: the Xmn1 restriction site C/T polymorphism at NT -158 upstream of HBG2; the BCL11A polymorphism on Chr2p16; the HBS1L-MYB intergenic polymorphism on Chr6q23. These 3 HbF quantitative trait loci collectively account for 20–50% of HbF variance in different populations. We found a significant association between high HbF and BCL11A and HBS1L-MYB intergenic region QTLs in these patients, but these account for only 20% of HbF variance (Table). These results were further validated in 590 patients of similar age from the Cooperative Study of Sickle Cell Disease, 57 patients with HbF 20.6 ± 8.2% and 533 patients with HbF 3.1 ± 1.5% (Table). To further explore other possible causes of elevated HbF, we sequenced 8.6 kb DNA fragment between HBG1 and HBD in 15 high HbF and 15 control patients. This DNA fragment includes the 7.2 kb Corfu deletion that is associated with elevated HbF levels and also binding sites for BCL11A. Twenty SNPs were found. The minor allele frequencies were consistently higher in the high HbF group, but the difference was found to be statistically significant only in 4 SNPs, 3 SNPs between positions 49213 and 49994 and 1 SNP at position 54541 (P = 0.001 to 0.04), suggesting that polymorphisms in this region might contribute to HbF expression in African American sickle cell anemia patients. The G→A polymorphism at position 49876 creates a C/EBP binding site which is not present in the major allele. The G→A polymorphism at position 49994 eliminates an AP-1 and NF-E2 binding sites, which are present in the major alleles. All 3 factors are erythroid transcription factors. The possible functional roles of these minor alleles found in significantly higher frequencies in the high HbF patients need to be further investigated.

ChromosomeGeneSNPMinor allele frequency (MAF) according to HapMapStudy cohortCSSCD cohort
YorubansEuropeansMAF in high HbF patientsP*MAF in high HbF patientsP
2p16 BCL11A rs766432 0.25 0.12 0.45 0.05 0.47 6.4E-06 
6q23 HBS1L-MYB rs9399137 0.04 0.25 0.18 0.02 0.09 7.0E-04 
11p15 HBG2 rs7482144 — — 0.1 1.0 0.30 6.6E-06 
ChromosomeGeneSNPMinor allele frequency (MAF) according to HapMapStudy cohortCSSCD cohort
YorubansEuropeansMAF in high HbF patientsP*MAF in high HbF patientsP
2p16 BCL11A rs766432 0.25 0.12 0.45 0.05 0.47 6.4E-06 
6q23 HBS1L-MYB rs9399137 0.04 0.25 0.18 0.02 0.09 7.0E-04 
11p15 HBG2 rs7482144 — — 0.1 1.0 0.30 6.6E-06 
P*

value in the study cohort was calculated by Fisher exact analysis comparing the MAF in 20 high HbF patients with 30 control patients.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.