Structure and expression of the human θ_l globin gene

The recently identified θ-globin gene subfamily conists of the θ₁-globin gene located downstream from the α₁ -globin gene^1-5, and several other members including at least one truncated, processed pseudogene ψθ² (refs 1,6). Unlike the θ₁ -globin genes of the rabbit³ and galago ⁴, the structure of these genes in the orang-utan and baboon ^1,5 and their flanking regions show no apparent defects that would prevent their expression. Both θ₁ -globin genes are split into three exons with the potential to code for a polypeptide of length 141 amino acids. Besides differing by 26% in replace-ment-site substitutions, the θ₁ and α₁ -globin genes of the orang-utan and baboon also differ in their promoter structures, in the use of TGA versus TAA as the termination codon, and in the use of AGTAAA versus AATAAA as the polyadenylation signal^1,5. In contrast, the two θ₁,- globin genes from primates only differ by 1.7% in the replacement-site substitutions⁵. Here we present the complete DNA sequence of a cloned θ₁ -globin gene of humans, and show that it contains no apparent defects that would abolish its expression. Furthermore, by primer extension of single-stranded oligonucleotide probes, we show that the θ₁ -globin gene of humans is transcribed in an erythroleukemia cell line K562. Three messenger RNA species were detected, with 5′-ends mapping to ∼70 base pairs (bp) downstream from a TAT A promoter sequence, at 8 bp downstream from a GGGCGG promoter sequence and at 40 bp upstream from the ATG inititrion codon, respectively. Haemin treatment of the K562 cells slightly enhances the level of the longest θ₁-transcript. Our results provide strong evidence that the θ₁ -globin gene of humans is transcriptionally active in cells of erythroid orign, and suggests the presence of a functional θ₁-polypeptide in specific cells, possibly those of early erythroid tissue.