Abstract
LIN28 proteins bind to RNA and regulate developmental timing events in multicellular organisms, in part, by reducing cellular levels of the let-7 family of microRNAs. High-level LIN28 expression in stem cells promotes their self-renewal. Over-expression of the LIN28 proteins causes suppression of let-7 in hematopoietic stem and early progenitor cell populations (CD34+) from adult donors and manifests a more fetal-like phenotype in the erythroid lineage. Here we explore LIN28expression that is restricted to erythroid cells, rather than stem or multi-potential progenitor cells.
For this purpose, lentiviral transduction vectors were produced with LIN28A expression driven by erythroid-specific gene promoter regions of the human KLF1 or SPTA1 genes, as well as an internal ribosomal entry site for puromycin selection (vectors: KLF1-LIN28A-OE and SPTA1-LIN28A-OE). Viral supernatants from these constructs were compared with empty-vector controls in matched transductions of CD34+ cells from three adult human volunteers. The cells were transduced and cultured using a three-phase, serum-free model for ex vivo erythropoiesis. Erythroblast proliferation and differentiation were comparable between control and LIN28-transduced cells assessed by cell counting and flow cytometry with staining for CD71, glycophorin A and thiazole orange. To validate restricted expression of LIN28 in the erythroid lineage, colony formation assays were performed in semisolid methylcellulose containing 1.0 ug/ml puromycin. BFU-E, CFU-GM, CFU-G, CFU-M and GEMM colonies were enumerated 14 days after plating. Puromycin addition to KLF1-LIN28A-OE and SPTA1-LIN28A-OE transductions resulted in selection of the erythroid colonies (BFU-E as a percentage of total colonies: Control: 44.6 ± 6.1%; KLF1-LIN28A-OE: 98.4 ± 0.7%, p=0.003; SPTA1-LIN28A-OE: 95.2 ± 1.1%, p=0.005).
LIN28A over-expression was confirmed by RT-QPCR (KLF1-LIN28A-OE: 2.1E+05 ± 7.0E+04 copies/ng; SPTA1-LIN28A-OE: 2.2E+05 ± 8.3E+04 copies/ng; Controls: below detection limits) and Western analyses after transduction. Suppression of all let-7 miRNA family members to less than 30% control levels were detected for both vectors resulting in a reduction in total let-7 miRNA (RT-QPCR: Control: 2.0E+07 ± 9.7E+05 copies/ng; KLF1-LIN28A-OE: 5.6E+06 ± 5.6E+05 copies/ng, p=0.003; SPTA1-LIN28A-OE: 4.6E+06 ± 6.2E+05 copies/ng, p=0.003). BCL11A expression levels were also measured by RT-QPCR and Western analyses. While BCL11A showed no significant change at the mRNA level (Control: 1.2E+03 ± 4.5E+02 copies/ng; KLF1-LIN28A-OE: 2.9E+02 ± 7.4E+01 copies/ng, p=0.07; SPTA1-LIN28A-OE: 4.2E+02 ± 3.3E+02 copies/ng, p=0.07), protein analyses of nuclear BCL11A showed moderately reduced levels after KLF1-LIN28A-OE and SPTA1-LIN28A-OE transductions.
Globin mRNA and protein levels were investigated and compared with controls. Gamma-globin mRNA was significantly increased in LIN28A-OE samples (Control: 3.6E+06 ± 8.2E+05 copies/ng; KLF1-LIN28A-OE: 1.9E+07 ± 1.7E+06 copies/ng, p=0.007; SPTA1-LIN28A-OE: 1.7E+07 ± 8.9E+05 copies/ng, p=0.003). Fetal hemoglobin (HbF) production was measured at the end of the culture period using High Performance Liquid Chromatography, and was increased in the KLF1-LIN28A-OE and SPTA1-LIN28A-OE samples compared to the control (Control: 7.0 ± 1.4%; KLF1-LIN28A-OE: 31.9 ± 2.7%, p=0.004; SPTA1-LIN28A-OE: 43.0 ± 6.2%, p=0.004). Flow cytometry analyses demonstrated a pan-cellular HbF distribution.
In contrast to promoting self-renewal in stem cells, these data suggest that adult erythroblast-restricted LIN28 functions to partially reverse the fetal-to-adult developmental transition in hemoglobin expression.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
