Editorials and Perspectives |
1 Department of Medicine, Division of Hematology/Oncology, Vanderbilt University Medical Center, Nashville, TN, USA
2 Medical Service, VA Tennessee Valley Healthcare System, South Nashville, TN, USA. E-mail:stephen.brandt{at}vanderbilt.edu andmark.koury{at}vanderbilt.edu
Erythropoiesis, the process by which mature red blood cells differentiate from hematopoietic stem cells, provides about 2x1011 new erythrocytes daily to replace the 1% of old cells removed from the circulation. Red cell production increases several-fold after blood loss or hemolysis but does not overshoot because it is so tightly regulated. This regulation involves three basic processes in erythroid progenitor and precursor cell populations: proliferation, differentiation, and survival. Although they act in concert from the earliest cell committed to erythropoiesis, the burst-forming unit-ery-throid (BFU-E), through the last cell division of erythroblasts, each of these three processes can be regulated independently of another. After blood loss or hemolysis, stem cell factor (SCF/Kit-ligand) and glucocorticoids increase proliferation of cells in the BFU-E to colony-forming unit-erythroid (CFU-E) stages.1 However, SCF and glucocorticoids have little effect on survival or differentiation of these erythroid progenitor cells. Erythopoietin (EPO), the major physiologic regulator of erythropoiesis, is regulated by hypoxia at the level of its transcription.2 EPO promotes the survival of erythroid cells in the CFU-E through basophilic erythroblast stages without affecting their proliferation or differentiation.
Contact with macrophages in erythroblastic islands regulates proliferation of these EPO-dependent cells without affecting their survival or differentiation.3 Differentiation of late-stage erythroblasts is characterized by the synthesis and accumulation of hemoglobin, the most abundant and major functional protein of mature erythrocytes. Accumulation of heme and globin chains, which is finely regulated to assure erythroblast survival,4,5 has no effect on cellular proliferation, which has largely ceased by this stage of differentiation.
Proliferation, differentiation, and survival of erythroid progenitors are ultimately controlled through activation and repression of specific genetic programs. An important transcriptional regulator in erythropoiesis is LIM domain-only protein 2 (LMO2). Originally discovered from its involvement by chromosomal translocation in T-cell acute lymphoblastic leukemia, the LMO2 gene encodes a small nuclear protein with little beyond its two LIM domains. This LIM domain, while related structurally to the zinc finger DNA-binding domains of the GATA family of transcription factors, is involved in protein-protein rather than protein-DNA interactions. LMO2 and related LIM-only proteins appear to function as adaptor molecules and assist in the assembly of multimeric transcription factor complexes.
The first indication that LMO2 was important for erythroid differentiation came from gene targeting studies in embryonic stem (ES) cells. Homozygous inactivation of the Lmo2 gene in mice led to the complete absence of yolk sac erythropoiesis and early embryonic lethality.6 Further, both Lmo2–/– embryonic stem (ES) cells6 and wild-type ES cells transduced with an anti-LMO2 single-chain antibody7 were completely blocked in erythroid differentiation. In contrast, enforced Lmo2 expression was reported to inhibit the differentiation of erythroid progenitor cell lines.8 However, while LMO2 protein could have opposite actions in early- versus late-stage erythropoiesis, it more likely has dominant negative effects when overexpressed.9
Given its significance in erythropoiesis, the mechanism by which LMO2 is regulated in the cell is an important subject. It is noteworthy, therefore, that Felli et al.10 report in a paper in this issue of the journal that microRNA 223 (miR-223) reduces expression of LMO2 message and protein. miRNAs are small non-coding RNAs that affect gene expression at a post-transcriptional level by both promoting the degradation of specific mRNAs and inhibiting their translation. The authors found, first, that miR-223 expression declined during erythroid differentiation of cord blood CD34+ cells, while LMO2 was upregulated. They then showed that miR-223 interacted with the 3 untranslated region of LMO2 mRNA and that forced expression of miR-223 reduced LMO2 message and protein and inhibited erythroid differentiation. The paper suggests that down-modulation of miR-223 is required for erythroid differentiation and establishes LMO2 as one of its major targets (Figure 1A). Two other studies in press also found that miRNA-223 down-regulates LMO2 expression.11,12
![]() View larger version (26K): [in a new window] [Download PPT slide] |
Figure 1. Regulation of LMO2 in erythropoiesis. (A) Regulation of LMO2 mRNA by miRNA-223. Specific binding of this small non-coding RNA to the LMO2 3 untranslated region inhibits the translation of this mRNA and promotes its degradation, with consequences for erythroid differentiation. (B) Regulation of LMO2 protein by single-stranded DNA-binding proteins (SSBPs). These SSBPs inhibit LMO2 and LIM Domain-Binding Protein 1 (Ldb1) interaction with an E3 ubiquitin ligase, RING Finger LIM Domain-Binding Protein (RLIM), sparing them from proteasomal destruction and increasing their steady state concentrations in the cell RISC, RNA-induced silencing complex.
|
Finally, recent studies have documented recruitment of the E box-GATA-binding complex to the beta-globin locus control region and proximal promoter, with additional data suggesting it regulates globin gene transcription through an effect on DNA looping.15 Of considerable interest given its own regulation by an miRNA, LMO2 was recently implicated in down-regulating miRNA-142.16
In summary, recent studies, including the one reported by Felli et al.10, have shown that LMO2 is regulated at both a posttranscriptional and post-translational level in erythroid progenitors. This undoubtedly contributes to the precision with which erythrocyte production is controlled in vivo.
Footnotes
Drs. Stephen Brandt and Mark Koury are Professors of Medicine at Vanderbilt University.
Work described in this manuscript was supported by Merit Review Awards from the Department of Veterans Affairs (SJB and MJK) and by National Institutes of Health grant R01 HL49118 (SJB).
References
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||