Mechanisms Regulating Increased Embryonic βh1 Globin Expression in Adult Nan anemic Mice

Sickle Cell Disease affects 90-100,000 in the US including 1/500 African-Americans born each year. Elevation of fetal hemoglobin (HbF) by co-inheritance of positive genetic modifiers of HbF expression or hydroxyurea (HU) treatment ameliorates disease severity. Because HU can have significant side effects, novel therapies aimed at elevating postnatal HbF expression are actively being sought. Three major loci modify HbF expression. Together, they account for ~50% of the variation in HbF expression, indicating that additional modifiers exist. Previously, we described the semi-dominant inbred mouse model Nan (neonatal anemia) that carries a missense mutation (E339D) in the second zinc finger of Krüppel-like factor 1 (KLF1/formerly EKLF) causing severe anemia accompanied by a striking failure of hemoglobin switching in Nan/+ mice (homozygotes die in utero). Embryonic βh1 globin expression is upregulated in Nan E14.5 fetal liver and in adult spleen where, remarkably, it accounts for nearly 100% of β-like globin gene expression. To extend these studies, we examined potential mechanisms regulating βh1 expression in adult Nan. Nan expression of Bcl11a, a downstream target of KLF1 that plays a major, conserved role in β-like globin gene switching, is 60-80% that of both untreated and phenylhydrazine treated (PHZ) wild type (WT) controls in the spleen. In peripheral blood, Nan BCL11A protein level is >50% of WT by western blotting. Importantly, prior studies by other investigators showed that newborn Bcl11a heterozygous knockout mice expressing Bcl11a at 50% of WT levels are haplosufficient, showing no differences in β-like globin gene expression. These data indicate that upregulated βh1 expression in Nan is BCL11A independent. We next examined erythropoiesis by flow cytometry using CD44, Ter119, and forward scatter (FSC) as markers. Nucleated erythroid precursors were strikingly decreased in Nan vs. PHZ-treated and phlebotomized (PHB) spleen, unusual in an anemic mouse model. Similar results were obtained using CD71, Ter119 and FSC gating. Despite this, βh1 expression normalized to saline-injected non-anemic controls was dramatically higher in Nan (93.0 ± 13.7, AU, X ± SEM) than either PHZ (3.6 ± 0.7) or PHB (7.4 ± 0.7) mice (p < 0.0001). Thus, increased βh1 in Nan is not simply due to stress erythropoiesis with concomitantly increased erythroid precursors. To analyze βh1 expression genetically, we constructed two Nan congenic lines on two different inbred genetic backgrounds, BALB/cBy and 129/SvImJ. Marked variation in adult spleen βh1 expression levels is seen among the three Nan strains. Similarly, we analyzed βh1 expression in an outbred high resolution mapping population derived from eight inbred strains, the diversity outcross (DO). Substantial variation in expression was seen among DO individuals. These data firmly establish the existence of modifying genes exerting profound influences on βh1 expression. We established an F2 intercross between 129S1/SvImJ-Nan/+ and C57BL/6J mice to take advantage of both Nan and DO mice to identify quantitative trait loci (QTL) modifying βh1 expression. Preliminary statistical analysis of 173 phenotyped (βh1 expression by RT-PCR) and SNP-genotyped F2 Nan/+ mice using R/qtl software identified a highly significant QTL for βh1 on Chr 7 encompassing the β-globin cluster and 3 suggestive QTL (Chr 4, 5 and 14). Analysis of 261 DO mice using QTL/ReL software identified 2 significant QTL (Chr 6, 7) and 6 suggestive QTL (Chr 2, 4[2], 6, 10, 14), with three in common with the QTL identified in F2 mice. Our analyses to date identify QTL overlapping three loci known to influence β-like globin gene switching (the β globin locus, Chr 7; LSD1, Chr 4; Mi-2β, Chr 6), providing proof of principle for our strategy. More importantly, additional loci identified contain no known modifiers, indicating the influence of novel genes. In summary, elevated βh1 expression in adult Nan spleen (1) occurs independently of Bcl11a; (2) is not mediated solely by stress erythropoiesis; (3) is highly influenced by genetic background; and (4) is influenced by novel genetic regulators of β-like globin switching.