β-thalassemia is a quantitative globin disorder that results from decreased levels of β-chain synthesis.2,3 The uncoupled α-chains form insoluble aggregates leading to ineffective erythropoiesis and shortened red cell survival. Iron overload from increased absorption and red cell transfusions contributes to end-organ damage. Globally, it is estimated that 1% to 5% of people are carriers, with certain geographical areas exhibiting a greater prevalence.2 The large clinical and hematologic variability can be partly accounted for by the milder phenotype seen in patients with a concomitant α-thalassemia mutation or with a compensatory increase in levels of the normally repressed γ-globin chain (hereditary persistence of fetal hemoglobin [HbF]), leading to clinically significant levels of HbF.3 Although some of the hereditary persistence of HbF mutations map to the globin locus, a significant focus of recent small- and large-scale genetic studies has been to identify nonlinked loci that achieve the same result.4
KLF1 (formerly known as EKLF) is an erythroid-enriched transcription factor that plays a critical global role in red cell gene regulation.5 Among its targets are genes within the β-like globin locus, where it directly (adult β-) and indirectly (fetal γ-) regulates the β-like globin switch during mammalian ontogeny.6 Although ablation of KLF1 is lethal, haploinsufficiency is benign but leads to altered gene expression of specific targets that are highly sensitive to KLF1 levels, such as the Lu antigen. Of relevance to the present study, another sensitive target is the Bcl11a gene, a repressor of γ-globin expression,7 whose levels are decreased such that HbF levels increase when KLF1 expression is lowered because of monoallelic disruption.8
The article focuses on the significant prevalence of KLF1 mutations and the evidence that these are linked to amelioration of the severity of β-thalassemia in those regions with high incidence; in the present case, the southern China provinces of Guangxi and Guangdong. Several aspects of this analysis are notable. Although there have been reports of KLF1 mutation occurring in patients with hemoglobinopathies, this is the first large population study to address how common this finding is. As a result, a major strength of the study is the large number of individuals surveyed (∼5000 total). This enabled significant conclusions to be made from a comparison of β-thalassemia endemic region (∼3800) vs non- thalassemia region (∼1200) samples, primarily that the KLF1 mutation prevalence was dramatically higher in the endemic samples. These numbers also allowed a comparison to be made of the median time to first transfusion between cohorts to show that patients with KLF1 mutations were significantly favorably affected. Perhaps surprisingly, patients with KLF1 mutations had a stronger ameliorative effect on severity than mutations within the β-locus, the HBS1L-MYB intergenic region, or in BCL11A. This resulted in patients who, although genetically β0-thalassemia homozygous (or compound β0-thalassemia heterozygotes) and therefore expected to have a thalassemia major phenotype, exhibited only a mild β-thalassemia intermedia phenotype and were largely transfusion-independent. Strikingly, 20% of nonthalassemic subjects with elevated HbA2 and HbF levels harbored KLF1 mutations.
Previously identified and novel KLF1 coding variants are described, and it is of interest that almost all mutations gave rise to truncation variants or were within the KLF1 zinc finger region, thus rendering 1 allele not expressed or functionally inactive.5
The present analyses strongly suggest that elevation of HbF and HbA2 levels, coupled with a decrease in CD44 expression, can be used as a basis to screen for KLF1 mutation. Identification of KLF1 mutations in individuals with β-thalassemia mutations can now be used along with other currently known predictors of disease severity to address prognosis and inform genetic counseling. Further, these types of analyses could well be directed at sickle cell disease patients, as a corollary to the present study is that monoallelic KLF1 mutations may also ameliorate the phenotypic severity in that population.9 In addition, including more of the KLF1 promoter region and introns in the analysis could also provide an additional source of mutation discovery relevant to alteration of expression.10 As with the present impressive study, it would be most optimal to characterize a large population of carefully characterized individuals.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
