Abstract

Characterization of the regulatory elements that control α-spectrin (ASp) gene expression is important for understanding the pathogenesis of ASp-linked hemolytic anemia. Our previous studies demonstrated that the ASp promoter directs low levels of expression, and, addition of a downstream region of noncoding exon 1 and intron 1 containing 2 GATA-1 sites conferred a 10-fold increase in activity in transient transfection assays. Transgenic (TG) mouse lines with the ASp promoter, the Asp promoter-exon 1-intron 1, or ASp promoter-exon 1-intron 1 with mutations of both splice sites linked to the human Aγ-globin gene as reporter were created. In reticultocytes, no expression was detected in any of the 8 lines transmitting the ASp promoter-Aγ-globin transgene. TG mice with the ASp promoter-exon1-intron 1 demonstrated significant levels of Aγ-globin gene expression in reticulocytes, with levels of Aγ-globin mRNA of ~0.4% of mouse α-globin mRNA/transgene copy #. This expression was nearly position independent, as 22/24 lines expressed the transgene. Using a FACS-based assay, γ-globin protein was present in 100% of erythrocytes. Expression levels comparable to the Asp promoter-exon 1-intron 1 TG were detected in 9/9 lines with the mutated splice sites, indicating splicing did not contribute to changes in expression. DNaseI hypersensitive site (HS) mapping identified a broad, erythroid-specific HS across exon 1 and intron 1. The presence of a DNaseI HS site suggested the presence of a positive regulatory element or a chromatin modification such as a boundary element. Analysis of a positive regulatory element in vivo was sought by stably transfecting the following luciferase (luc) plasmids into K562 cells: ASp promoter, ASp promoter-exon 1-intron 1, ASp promoter-exon 1, ASp promoter-intron 1, and ASp promoter-exon 1-intron 1 with both GATA-1 sites mutated. Clones with copy # ≤5 were analyzed; ≥9 independent clones/line were analyzed. Normalized luc activity of the ASp promoter-exon 1-intron 1 was significantly higher than the ASp promoter in stably transfected cells, 86±15 v 28±3 (p<0.001). Mutation of both GATA-1 sites in the exon 1-intron 1 plasmid reduced activity to background. Normalized luc activity from the promoter-exon was 46±6; from the promoter-intron 101±31, suggesting the intron functions as a positive regulatory element. A barrier assay was performed by flanking a β-globin promoter-EGFP gene using wild type (WT) exon 1, exon 1 with the GATA site abolished, or WT intron 1, and stably transfecting the plasmids into K562 cells. The WT exon 1 and mutant exon 1 expressed GFP in 10/12 and 7/8 lines, respectively, indicating a barrier function for exon 1 independent of GATA-1 activity. Only 1/9 lines expressed EGFP when the cassette was flanked by the ASp intron and 0/8 expressed EGFP when there were no sequences flanking the β-globin promoter. TG mouse lines with the Asp promoter-exon 1 or the Asp promoter-intron 1 linked to the Aγ-globin gene were created. 1/5 TG lines with ASp promoter-exon 1 expressed at low levels and 3/7 TG lines with ASp promoter-intron 1 expressed at levels comparable to the ASp promoter-exon 1-intron 1. These results demonstrate that 2 elements downstream of the ASp promoter are required for high-level, erythroid-specific expression. Exon 1 has barrier activity and intron 1 functions as a positive regulatory element. This is an excellent candidate region for mutations associated with ASp-linked inherited hemolytic anemia.

Author notes

Corresponding author