SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. References

MicroRNAs (miRNAs) are small noncoding RNAs that regulate gene expression by pairing with their target mRNAs, thereby inducing protein translation inhibition or/and mRNA degradation. There is now strong evidence that miRNAs play a crucial role in the regulation of hematopoiesis. Several groups have shown that miRNA expression change dynamically during hematopoietic differentiation and functional studies demonstrated that miRNAs control not only differentiation but also activity of hematopoietic cells by targeting transcription factors, growth factor receptors, and specific transcripts involved in the modulation of cellular responses to external stimuli. In this review, we will summarize the current knowledge of miRNA expression and function during hematopoiesis and discuss controversies and future directions. Am. J. Hematol., 2010. © 2010 Wiley-Liss, Inc.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. References

Hematopoiesis is a highly regulated process controlled by complex molecular events that simultaneously regulate commitment, differentiation, proliferation, and apoptosis of hematopoietic stem cells (HSC). Critical to this process is a network of transcription factors which activation or inhibition is required to initiate the commitment of HSC to different lineages precursors [1]. Chromatin modifications including histone acetylation and DNA methylation has been recently recognized as regulatory mechanisms for transcription factor expression regulation during hematopoiesis [2]. There is now strong evidence that a novel class of noncoding RNAs, called microRNAs, modulate hematopoietic differentiation, proliferation, and activity of hematopoietic cells by targeting the expression of transcription factors and genes involved in the regulation of cell cycling and proliferation.

MicroRNAs (miRNAs) are small (18-24 nucleotides) evolutionary conserved noncoding RNAs that bind to the 3′untranslated region (UTR) of target mRNAs resulting in translation repression or mRNA degradation [3]. MiRNAs are transcribed by polymerase II from unique miRNA genes or from “host” genes (within intrones of known protein-coding or noncoding genes) into initial precursor of variable length called pri-miRNA [3]. This initial precursor is processed by the RNase Drosha to a 70- to 120-nucleotides hairpin structure precursor called pre-miRNA that is subsequently exported to the cytoplasm by exportin 5 [4]. Once in the cytoplasm, this pre-miRNA is further processed by another RNAase; Dicer, resulting a 18- to 24-nucleotide duplex, that after losing the complementary strand or passenger strand, is incorporated into the RNA interfering silencing complex (RISC) [4]. The RISC is comprised by Dicer, argonautes and Di George critical region protein 8 proteins and is the effector pathway for miRNAs canonical activity [3, 4]. The mature miRNA strand will direct the RISC complex to the target mRNA based on the complementarity between the miRNA and its target 3′UTR, resulting in mRNA cleavage when the complementarity is perfect or in protein translation inhibition when the complementarity is imperfect [3, 4]. A single miRNA can control the levels of hundreds of different target genes and multiple miRNAs can regulate a single mRNA [3, 4]. Currently, there are 940 miRNAs registered in Sanger miRNA registry (miRBase version 15). Strong evidence indicates that miRNAs play important roles in cellular processes such as proliferation, development, differentiation, and apoptosis [5–7]. Not surprisingly, we and others have shown that miRNAs are dynamically expressed during hematopoiesis and regulate differentiation and activity of hematopoietic cells (Fig. 1).

thumbnail image

Figure 1. Expression levels and regulatory networks of miRNAs during hematopoiesis. CLP, common lymphoid progenitor; CMP, common myeloid progenitor; DN, double negative T-cell precursors; DP, double positive (CD4+CD8+ T cells); EMP, erythrocyte-megakaryocyte precursor; EP, erythrocyte precursor; GMP, granulocyte-monocyte precursor; GP, granulocyte precursor; HSC, hematopoietic stem cells; MP, megakaryocyte precursor. Dashed green arrows means positive regulation; Red dashed lines means negative regulation; Black Thick arrows follows hematopoiesis development and differentiation hierarchy; Small black arrows (up or down) means miRNA expression during a specific hematopoietic stage. MiRNA targets or transcription factors involved in the regulation of miRNAs are shown using blue letters. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Chen et al. [5] firstly demonstrated that miRNAs are differentially expressed in mice hematopoietic tissues. The authors found that miR-181 was expressed at high levels in the murine thymus and differentiated B lymphocytes, whereas it was detectable at lower levels in undifferentiated progenitor cells. MiR-223 expression was confined to myeloid cells and miR-142 expression was highest in B-lymphoid tissues. Ectopic expression of miR-181 in hematopoietic progenitor cells resulted in B lymphocyte proliferation upon hematopoetic reconstitution. However, ectopic expression of miR-142 or miR-223 did not induce any obvious phenotype [5].

Following this pioneer report, many groups have investigated the expression and function of miRNAs during hematopoiesis. In this review, we will describe miRNA expression patterns during hematopoiesis and the functional implications of these expression changes. In addition, we will discuss controversies and future directions in the field. The role of miRNAs in the initiation and progression of hematological malignancies has been extensively reviewed and will not be discussed in this review [8–10].

MicroRNAs are required in hematopoiesis

Dicer is an RNase-III enzyme that is critical for miRNA biogenesis. As expected, Dicer-deficient mice exhibit premature death at embryonic day 7.5 with a lack of detectable multipotent stem cells [11]. Conditional deletion of Dicer in murine embryonic stem cells renders these cells unable to differentiate [12]. Ablation of Dicer in early B-cell progenitors results in a developmental block at the pro- to pre-B-cell transition and the antibody repertoire is completely disturbed [13]. Within T cells, conditional Dicer deletion results in impaired T-cell development and aberrant T helper cell differentiation and cytokine production [14]. A severe block in peripheral CD8+ T cells in the thymus and a reduced number of CD4+ T cells are also observed; thereby indicating that miRNAs seems to be required for CD4/CD8 lineage commitment [15]. Altogether, data from germline and conditional Dicer deletion indicate that miRNAs play a critical role in hematopoiesis.

MicroRNAs in HSC

Georgantas et al. [16] measured miRNA expression in human CD34+ selected HSC obtained from pooling mobilized peripheral blood stem cell harvests (PBSCH) or BM samples from healthy donors. Using a microarray miRNA platform, the authors identified 33 miRNAs expressed (above background) in human CD34+ cells from both peripheral blood HSC (PBHSC) and BM. These most abundant miRNAs in human BM CD34+ cells were; miR-191, miR-181, miR-223, miR-25, miR-26, miR-221, and miR-222. The integration of in silico miRNA target predictions with the microarray miRNA data, suggested that miRNAs control hematopoietic differentiation through the translational control of mRNAs critical to hematopoiesis.

Liao et al. [17] isolated CD34+ CD38- HSC and CD34+ HSC from umbilical cord blood (UCB) and performed miRNA microarray analysis. The miRNA expression profile obtained from CD34+ HSC was in high concordance with the paper of Georgantas et al. [16]. The authors further compared CD34+ vs. CD34+ CD38- HSC, arguing that the latter subpopulation display stem cell properties compared to a more heterogenous and committed cell population identified only by positivity of CD34. The authors found 9 miRNA over-expressed and 22 down-regulated in CD34+ CD38- HSCs compared with CD34+CD38+ cells. Among the most up-regulated miRNAs in CD34+ CD38- HSC, miR-520h was predicted in silico to target ATP-binding cassette, subfamily G (ABCG2), which is a gene involved in stem cell maintenance. Transduction of miR-520h into CD34+ cells increased the numbers of multiple progenitor colonies (CFU-E, BFU-E, and CFU-GM) and the number of CD34+ cells as well [17]. Therefore, the authors reasoned that miR-520h may promote differentiation of HSC into committed progenitors by inhibiting ABCG2 expression. It is intriguing that higher miR-520h expression levels are observed in CD34+ CD38- HSC with respect to CD34+ cells. One would expect that miR-520h will increase with differentiation commitment. Further studies will be needed to establish the role of miR-520h in hematopoiesis.

Using Taqman low density arrays, Merkerova et al., [18] attained miRNA expression profiles of UCB CD34+ cells. There was high concordance in the miRNA expression of UCB CD34+ cells with respect to the other two previous studies [16, 17]. In this report, miR-520h was found highly expressed in CD34+ cells and deeply down-regulated in T lymphocytes, monocytes, and granulocytes. Furthermore, the authors found significant up-regulation of miRNAs that were previously confirmed to target HOX genes, which are involved in self-renewal of HSC/early progenitors. Consistent with previous reports indicating gene expression differences between UCB CD34+ cells and BM CD34+ cells [19], likely due that UCB contains a larger population of immature and pluripotent CD34+ CD38- cells, the authors found that the expression of 13 miRNAs was significantly different between these cell types [18]. Interestingly, the expression of the miRNA cluster comprised by miR-517c, miR-518a, miR-519d, and miR-520h, was detected only in UCB CD34+ cells. By integrating transcriptome and miRNA expression in these samples, the authors found negative correlations among in silico predicted miR-520 targets (51 genes) and miR-520h expression. Although ABCG2 mRNA expression was not correlated, ID1 and ID3 genes, both involved in terminal hematopoietic differentiation, were found to inversely correlate with miR-520h [18].

Erythropoiesis

Felli and colleagues firstly reported that two miRNAs (miR-221 and miR-222) were down-regulated during in vitro differentiation of UCB CD34+ cells to erythroid precursors (Table I and Fig. 1) [20]. The authors further demonstrated that the down-regulation of both miR-221 and miR-222 was necessary to promote early erythroid proliferation of UCB CD34+ cells. Notably, both miRNAs target the tyrosine kinase receptor c-KIT, which is critical for early erythroblasts expansion [21]. The down-regulation of both miR-221 and miR-222, unblocks KIT expression and allows erythroblast expansion [20]. Forced expression of miR-221 and miR-222 in UCB CD34+ HSC inhibits KIT protein expression and erythroid growth. While critical for early erythropoiesis, KIT expression is silenced in late erythropoiesis by multiple mechanisms, including the transcription factor GATA-1 [21]. Recently, Gabbianelli et al., [22] demonstrated that human perinatal hemoglobin switching is under control of the KIT receptor and partially by miR-221/-222. Over expression of both miRNAs in UCB progenitors caused a decrease of erythroblast proliferation and of fetal hemoglobin content.

Table I. MicroRNAs with Important Role in Hematopoiesis
MiRNAExpressionTargetsFunctionRegulated byReferences
  1. MK, megakaryopoiesis; Mo, monopoiesis; GC, germinal center; DN3, double negative stage 3; DN4, double negative stage 4; CMP, common myeloid progenitors; GMP, granulocyte-monocyte progenitors; Prog, progenitors; Tregs, T regulatory cells.

Erythropoiesis
 miR-221 miR-222Downc-KITInhibits early erythroid proliferation Controls perinatal HB switchingUnknown20,23-25
 miR-451 miR-144UpFOXO3 GATA-2 KlfdRegulates oxidative stress, positive regulation of terminal erythroid differentiation and in zebrafish hemoglobin synthesis (miR-144)GATA-123, 25-30
 miR-24Bi-phasichALK4Modulates negatively erythropoiesisUnknown32
 miR-155DownPU-1/ETS-1 CEBPb/SHIP1Inhibits erythropoiesisAP-1, JNK23,26, 34-35
 miR-223DownLMO2Inhibits erythropoiesisCEBPa, NF1A, E2F133
Megakaryopoiesis
 miR-130aDownMAFBUnknownUnknown37
 miR-10aDownHOXA1UnknownUnknown37
 miR-150Upc-MybDrives MK differentiationUnknown38,40
 miR-155DownEts-1/Meis1Inhibits MK differentiation and proliferationAP-1, JNK12, 34-35, 44
 miR-146aDownCXCR4Inhibits MK differentiation and proliferationPLZF41
 miR-125b-2DownDICER1/St18Increase MK proliferationUnknown43
Myelopoiesis
 miR-223Up GNFI-A/E2F1 MEF2CInhibits granulocytic proliferation and activityCEBPα, NFI-A, E2F145-48
 miR-21 miR-196aUp G/GMUnknownRegulation between CMP-GMP transitionGFI-152
 miR-424Up MoNFI-AInduces Mo differentiation and proliferationPU.153-54
 miR-17-92Down MoAML1Inhibits Monopoiesis in vitro but no effects in vivoAML155-56
Lymphopoiesis
 miR-150Down prog.c-MybInhibits the transition from pro-B to pre-B cellsUnknown64-65
Up mature B/T cellsBlock DN3 to DN4 T-cell transition
 miR-181aUp prog. and in DP T cells Down in mature B/T cellsBCL2/CD69 TCR/DUSP5 DUSP6/SHP2 PTPN22Increases CD19+ cells/Modulate late thymic T-cell development, TCR sensitivities and strenghtUnknown67-68
 miR-17-92Up in B/T precursors Down in mature B/T cellsBim PTENModulates positively the transition from pro-B to pre-B cellsc-Myc, E2F156,69
 miR-155Up in Tregs and in activated T/B cellsSOCS1 SHIP-1 CEBPb AIDEnhanced Tregs proliferation Maintain competitive Treg fitness Regulates GC-B cells responses, innate imunity, T-cell dependent antibody responses, negative regulator of somatic hypermutationFoxp335, 73-78

A systematic descriptive analysis of miRNA expression in erythrocyte precursors obtained from peripheral blood mononuclear cells and cultured in a three phase liquid system, revealed a progressive down-regulation of miR-150, miR-155, miR-221, and miR-222, up-regulation of miR-16 and miR-451 at late stages of differentiation and a biphasic regulation of miR-339 and miR-378 (Table I and Fig. 1) [23].

Others groups investigated miRNA expression using leukemic cell lines that exhibit erythroid differentiation on stimulation with chemicals (e.g., hemin), such as the erythroleukemia cell line K562 (blastic phase Chronic myeloid leukemia) [24] or murine erythroleukemia line (MEL) ,which are blocked at the pronormoblast stage of differentiation [25]. Although there were similarities in miRNA expression profiles attained using primary UCB samples with respect to the differentiated cell lines (e.g., up-regulation of miR-451, miR-24, and miR-16 and down-regulation of miR-221-222 and miR-155 during erythropoiesis), notable differences were observed probably due to the inability of cell lines to fully differentiate into mature erythrocytes and their leukemic origins.

Among all erythroid miRNA profiling studies, miR-451 is the most consistent and significantly upregulated miRNA after erythroid differentiation [23, 25–28] (Table I and Fig. 1). Furthermore, its expression is specific to erythroid tissues. MiR-451 is transcribed in cluster with miR-144 from chromosome 17 by the transcription factor GATA-1 [27]. Ectopic transfection of miR-451 promotes erythroid differentiation and hemoglobin content of murine erythroleukemia cells while the antagomir has opposite effects [25]. Using a zebrafish mutant (meunier: mnr) that expresses GATA-1 but not miR-144 and -451, Pase et al. [28] recently reported that erythropoiesis was initiated but the maturation was retarded in this model. MiR-451 over-expression partially rescued erythroid maturation.

The in vivo effects of miR-451 were explored by two groups using morpholino technology in zebrafish models. Dore et al. elegantly described that zebrafish embryos depleted of miR-451 but not miR-144 exhibited severe anemia, despite forming erythroid precursors [27]. Pase et al. [28] subsequently reported that the depletion of miR-451 but not miR-144 resulted in delayed erythroid maturation. However, the authors did not observe severe anemia as previously reported [27]. A possible explanation was recently provided by a recent review by Zhao et al., [29] arguing that the use of 1-phenyl-2-thiourea (PTU) in Dore et al. studies to improve visualization of internal embryo structures, may have caused excessive PTU-induced anemia due to its oxidative effects. It has been preliminary reported that the miR-144/451 locus ablation in mice causes mild hemolytic anemia at baseline and increased erythroid susceptibility to oxidative stress through the direct targeting of the transcription factor FoxO3, which activates numerous erythroid anti-oxidant genes [29]. Rasmussen et al., recently reported that mice deficient for the miR-144/451 cluster exhibits late erythroblast maturation defects, resulting in erythroid hyperplasia, splenomegaly, and a mild anemia. Again, it seems that the phenotype is mainly driven by the loss of miR-451 expression, because mice deficient for miR-451 has a phenotype indistinguishable from the miR-451/144 mice [30]. The data also suggest an importance for miR-451/144 in providing robustness to erythropoiesis under situation of oxidative stress. In addition, to FOX03, GATA-2 is a bona fide target of miR-451, at least in fish [27]. There is some evidence that miR-144 may be involved in the regulation of zebrafish hemoglobin synthesis by direct targeting of the Kruppel-like transcription factor (Klfd) that activates α-E1 gene transcription, a embryonic form of alpha globin. Klfd binds to the promoters of α-globin and miR-144/451 to activate their transcription constituting a negative feedback circuitry [31].

Among other miRNAs differentially expressed during erythropoiesis, miR-24 which is highly expressed in CD34+ HSC and down-regulated during erythropoiesis, modulates negatively erythropoiesis [32]. These effects are mediated by direct targeting of Activin type I receptor, which promotes erythropoiesis in cooperation with erythropoietin. Like miR-24, miR-223 is down modulated in unilineage erythroid culture of UCB CD34+ cells. Enforced expression of miR-223 inhibits erythroid maturation of human erythroleukemia cells and UCB CD34+ progenitors [33]. This effect is partially mediated by targeting the LIM-only protein 2 (LMO2) by miR-223 [33]. LMO2 is a transcription factor that along with GATA-1, SCL/TAL1 and LDB1 constitutes a multiprotein complex that promotes erythropoiesis (Table I and Fig. 1).

Many reports indicate that miR-155 is down-regulated during erythropoiesis induction in primary cells or cell lines [23, 26]. Consistent with this data, a decreased numbers of erythroid precursors and circulating red cells were observed in transplanted animals with HSC over-expressing miR-155 [34]. In contrast, there was no obvious erythroid phenotype changes observed in the miR-155 knock out mice (Table I and Fig. 1) [35].

Last, Zhao et al. showed a negative autoregulative feedback loop between miR-15a and c-Myb, whereas c-Myb binds to miR-15a promoter and regulates its expression. In turn, miR-15a binds to the 3′ UTR of c-Myb and blocks its translation. The expression of c-Myb and miR-15a are inversely correlated in CD34+ cells undergoing erythroid differentiation. Enforced expression of miR-15a in normal marrow mononuclear cells blocked erythroid and myeloid colony formation in vitro [36].

Megakaryopoiesis

Our group first reported miRNA expression profiles during in vitro megakaryocytic differentiation of BM CD34+ progenitors [37]. We identified 19 miRNAs down regulated during megakaryocytic differentiation of CD34+ progenitors, including miR-10a, -10b, -30c, -106, -126, -130a, -32, and -143 among others (Table I and Fig. 1). Three of them (miR-223, miR-15a, and miR-16-1) exhibited a biphasic expression pattern; early down-regulation and late up-regulation, suggesting a stage-specific function. Many of the downregulated miRNAs are predicted to target megakaryocytc specific genes, suggesting that the loss of these miRNAs, unblocks their targets resulting in megakaryocytic differentiation. Indeed, we found that miR-130 targets MAFB, which induces GPIIb in synergy with GATA1, SP1, and ETS-1 (Table I) [37].

Other groups have reported functional studies involving individual miRNAs in megakaryocytic differentiation. Lu et al. [38] showed that miR-150, which is moderately expressed in megakaryocyte/erythrocyte precursors (MEP), exhibits increased expression in cells undergoing megakaryocytic differentiation while it is down-regulated in cells differentiating to the erythroid lineage (Table I and Fig. 1). Using gain- and loss-of-function experiments, the authors demonstrated that miR-150 drives MEP differentiation towards megakaryocytes at the expense of erythroid cells in vitro and in vivo by targeting the transcription factor c-Myb. Mice bearing lower c-Myb expression as a result from targeted mutations, exhibits marked anemia and increased thrombocytosis with respect to controls, suggesting that c-Myb is a critical player in the cell fate decision between megakaryocyte/erythrocyte differentiations. This report challenges the dogma that miRNA are merely fine differentiation tuners or activity regulators. Although the data are provocative, experiments using available germline miR-150 knock out models may be warranted to strongly establish the role of miR-150 in MEP cell fate decisions. To this end, the miR-150 knock out mice were reported to be viable, fertile and morphologically normal, exhibiting only B-cell and Tcell alterations, as will be further described [39]. Supporting a role for miR-150 in megakaryocytic differentiation, Barroga et al. [40] reported that thrompoietin (TPO), the primary humoral regulator of platelet production, suppressed c-Myb expression by increasing miR-150 expression in megakaryocyte precursors and TPO-dependent cell lines. Another miRNA, miR-34a has been shown to represses c-Myb expression and cyclin dependent kinases during megakaryopoiesis in leukemic cell lines and primary CD34+ progenitors treated with TPO [41]. However, the down-regulation of c-Myb preceded the up-regulation of miR-34, indicating other mechanisms involved in c-Myb regulation during the early days of culture.

There has been a report implicating miR-146 in megakaryopoiesis. Labbaye et al. [42] found that miR-146a is suppressed during human megakaryopoiesis and that ectopic expression of this miRNA suppresses megakaryocyte differentiation and proliferation during in vitro culture of CD34+ cells. The authors found that miR-146 targets the chemokine receptor 4 (CXCR4), which integrity is indispensable for megakaryopoiesis (Table I and Fig. 1) [42].

Enforced expression of miR-125b-2 in human CD34+ cells increases the number and the size of CFU-MKs without blocking megakaryocytic differentiation, suggesting that this miRNA is controlling proliferation of MEP [43]. MiR-125b-2 seems to target ST18 (suppression of tumorigenicity 18) and Dicer (critical enzyme involved in miRNAs biogenesis). Indeed, the authors found a regulatory negative feedback loop between Dicer and miR-125b-2. Production of miR-125b by Dicer resulted in suppression of Dicer protein expression levels and, consequently, impaired overall miRNA processing (Table I).

Finally, the oncogenic miR-155 is highly expressed in CD34+ cells, but it is sharply down-regulated during in vitro UCB CD34+ cells differentiation to megakaryocytes [44]. The enforced expression of miR-155 in human CD34+ cells impaired the proliferation and differentiation of megakaryocytes in vitro and in vivo, likely through direct targeting of the pro-megakaryocyte transcription factors ETS-1 and Meis-1 (Table I and Fig. 1) [12].

Myelopoiesis

One of the first studies to suggest a role for miRNAs during myelopoiesis was reported by Fazi and colleagues. The authors found that miR-223 expression was strongly induced in acute promyelocytic leukemia (APL) cell lines by retinoic acid (RA) [45]. On RA treatment, CEBPa (a master myeloid transcription factor that promotes granulocytic-monocytic differentiation) binds to the miR-223 promoter, displaces the negative granulopoietic transcription factor NFIA, and activates miR-223 transcription. Ectopic expression of miR-223 in APL cell lines induces granulocytic differentiation, whereas the knockdown of miR-223 exhibited opposite effects. High expression of miR-223 was further confirmed in human peripheral blood granulocytes, whereas its expression was down-modulated during monopoiesis [45]. These data lead to the authors to reason that miR-223 is a positive regulator of granulopoiesis. However, mice deficient for miR-223 expression exhibit an increase in peripheral blood and bone marrow neutrophils [46]. The mutant neutrophils displayed unusual morphology, aberrant pattern of lineage specific markers expression and increased reactivity to activating stimuli, including evidence of spontaneous inflammatory lung pathology. Furthermore, the authors showed that MEF2c, a transcription factor that promotes myeloid progenitor differentiation, is a bona fide target of miR-223 and the phenotype of the miR-223 deficient mice was partially corrected when the MEF2c gene was genetically ablated (Table I and Fig. 1).

A recent study using acute myeloid leukemia (AML) cell lines reported that the CEBPa induced miR-223 targets the cell cycle regulator E2F1, thereby blocking cell cycle progression and myeloid proliferation [47]. Interestingly, E2F1 binds to the miR-223 promoter suppressing its transcription. Thus, CEBPa-miR-223 and E2F1 comprise a negative autoregulatory loop in AML [47]. The implication of this finding in normal hematopoiesis is unknown. These results also provide with a complementary explanation for the increased numbers of granulocytes observed in the miR-223 deficient mice, since these effects could also be explained by shorter cell cycle and increase proliferation due to higher levels of E2F1 in the miR-223 deficient mice with respect to the wild type controls. In conclusion, miR-223 decreases myeloid proliferation and dampens granulocytic activation. These data also highlight the importance of performing studies using knock in or knock out animal models, representing better models for establishing definitive functional roles of miRNAs during hematopoiesis.

Several groups have also systematically profiled the expression of miRNAs during RA-induced granulocytic differentiation of APL cells [48–50]. Our group found 10 up-regulated miRNAs (let-7a-3, miR-16-1, −223, −15a, −15b, let-7c, let-7d, −342, 107, and −147), whereas miR-181b was down regulated [48]. We further confirmed that the RA-induced down-regulation of the oncogenes BCL-2 and RAS is partially mediated by miR-15a/-16-1 and several let-7 family members. The expression profiles obtained by our group in NB4 cells treated with ATRA were validated independently by an independent group in primary APL samples [50].

The growth factor independence gene (GFI-1) encodes a transcriptional repressor which is required for normal granulopoiesis [1, 51]. Mice and humans deficient in GFI-1 expression exhibit an arrest in myeloid differentiation and have increased granulocytic precursors [1, 51]. Using cell lines and GFI-1 deficient mice, Velu et al. [52] found that GFI-1 binds to the promoter region of miR-21 and miR-196b and suppress their transcription. GFI-1 up-regulation in granulocyte-monocyte progenitors (GMPs) from wild-type mice was associated with a dramatic reduction of both miR-21 and miR-196b. These results were validated in human leukemic cell lines models of granulocytic and monocytic differentiation. Over-expression of miR-21 in Lin- murine BM cells resulted in a significant increase of monocytic colonies, whereas the use of anatgomiR-21 had opposite effects [52]. Ectopic expression of miR-196b resulted in a significant loss of granulocytic colonies. However, co-expression of both miR-21 and miR-196b completely blocks G-CSF induced granulopoiesis and lead to the accumulation of cells that exhibited morphological and immunophenotyping features of immature granulocyte and monocyte precursors. Thus, the data indicate that these two miRNAs are critical players in the GFI-1 activity controlling the transition between GMPs to granulocyte precursors (Table I and Fig. 1) [52].

In monocyte development, it was reported that the transcription factor PU.1 activates miR-424, which in turn induces monocytic/macrophage differentiation in AML cell lines and human CD34+ cells [53]. These effects are mediated at least in part by activating specific genes important for monocyte/macrophage differentiation such as M-CSF receptor, through the direct repression of the negative regulator of monopoiesis: NFIA. The up-regulation of miR-424 was confirmed by other group [54] using cell lines monocytic differentiation models (Table I and Fig. 1).

Fontana et al. [55] reported that the miR-17-5p, miR-20a, and miR-106a cluster are down regulated during monocytic differentiation and maturation of human CD34+ cells. Consequentially, the transcription factor AML-1, which is a direct target for the three miRNAs, is unblocked and up-regulated, thereby promoting monocytic-macrophage maturation and differentiation, while suppressing further the transcription of these three miRNAs (Table I and Fig. 1). The ectopic expression of the three miRNAs delays terminal differentiation of monocytes, whereas their inhibition accelerates differentiation. Notably, mouse monopoiesis occurring in the absence of miR-17-92 cluster expression is normal [56]. Whether this is due to differences between mouse and human monopoiesis or this is inherent to the in vitro models used by Fontana et al., further clarification is needed. This is another example on how mouse models contradict sharply to data obtained using cell lines or primary cells in vitro differentiation models.

As we discussed previously, miR-125b has an important role in increasing proliferation and self-renewal of human and mouse megakaryocytic progenitors and MEPs [43]. In contrast, overexpression of miR-125b in human and mouse HSC, results in increased proliferation of inmature granulocytes (Table I and Fig. 1). Bousquet et al. [57] reported a strong up-regulation of miR-125b in MDS and AML with t(2;11)(p21;q23). In vitro experiments revealed that miR-125b was able to interfere with primary human CD34+ cell differentiation, and also inhibited terminal (monocytic and granulocytic) differentiation in HL60 and NB4 leukemic cell lines. While these results may be relevant for leukemia, in particular megakaryoblastic leukemia in the context of Down syndrome and AML/MDS with t(2 ;11), the relevance of these findings for normal hematopoiesis is unclear.

The role of miR-155 during normal human hematopoiesis was established using knock out animal models. Notably, there were no myeloid abnormalities detected in the mice deficient for miR-155 expression [35]. Although miR-155 is not required for normal hematopoiesis, it may need to be repressed during this process. Supporting this hypothesis, sustained miR-155 expression in murine HSC causes myeloproliferation with extramedullary hematopoiesis, splenomegaly and dysplastic changes [34]. There was also a decreased in the erythroid and megakaryocyte colonies [12]. Certainly, these findings are important for leukemogenesis, since aberrant expression of miR-155 has been described in distinct subsets of AML [58, 59].

Jurkin et al., [60] reported that miR-146 expression is detected at low levels in monocytes and it was absent in neutrophils. The authors further identified miR-146a as a regulator of monocyte and dendritic cell (DC) activation but not myeloid/DC subset differentiation. Ectopic miR-146a in monocytes and intDCs interfered with TLR2 downstream signaling and cytokine production, without affecting phenotypic DC maturation.

Lymphopoiesis

Over the past few years, several groups have reported strong data supporting a critical role for subsets of miRNAs in lymphoid cells development and immune function. Numerous in depth reviews has been recently published [61–63]. Here, we will highlight the most important miRNAs involved in the modulation of this lineage (Fig. 1).

miR-150

MiR-150 is mainly expressed in lymphoid tissues, including lymph nodes and spleen. The expression is restricted to mature T and B lymphocytes but not in their progenitors. Two groups independently reported that ectopic expression of miR-150 in murine HSC greatly impairs the formation of mature B cells and inhibits the transition from pro-B to the pre-B-cell stage partially by increasing apoptosis [64, 65]. In one study, a less pronounced block was apparent in the T-cell lineage, at the double-negative 3(DN3) to double-negative 4 (DN4) transition. Mice deficient for miR-150 expression were viable and fertile, but exhibited phenotype changes restricted to the lymphoid tissues, consisting of expansion of B1 cells in the spleen and in the peritoneal cavity, higher levels of serum immunoglobulins and enhanced T-cell dependent immune responses [64]. The authors further showed that these changes were caused at least in part by up-regulation of c-Myb, a confirmed miR-150 target. C-Myb is an essential transcription factor for early lymphoid development and its targeted loss in B cells leads to a block in B-cell differentiation from pro-B to pre-B-cell stage, identical to the phenotype observed on ectopic over-expression of miR-150 [66] (Fig. 1).

miR-181

MiR-181 is comprised of three clusters, miR-181a-1/miR-181-b-1 located in chromosome 1, miR-181a-2/miR-181b-2 located in chromosome 9, and miR-181c/miR-181d on chromosome 19. MiR-181a is expressed highly in the thymus and at lower levels in the lymph nodes and BM [5]. Specifically, miR-181a expression is high in the early B-cell differentiation stages and decreases from the pro-B-cell to the pre-B-cell stage. The ectopic expression of mir-181a in murine HSC resulted in an increase in the percentage of CD19+ B cells and decrease of CD8+ T cells without affecting other hematopoietic lineages in hematopoetic reconstitution assays in vivo [5].

Further studies are required to establish the role of miR-181 in B-cell development. With respect to T cells, miR-181a is dynamically up-regulated at the double positive (CD4+ CD8+) stage of thymocyte development and correlated inversely with the expression of three predicted miR-181 targets; BCL-2, CD69 and the T-cell receptor (TCR) alpha, all of them involved in positive T-cell selection [67]. Further experiments confirmed that these proteins are bona fide targets of miR-181. A different group confirmed experimentally that the ectopic expression of miR-181a in thymic progenitor cells can promote CD4 and CD8 double-positive T-cell development, supporting a role for miR-181 in late T-cell thymic development [68]. In addition, to development functions, miR-181 also modulates TCR signaling strength and sensitivities, thereby influencing T-cell sensitivity to antigens. These effects are mediated by down-regulated expression of several protein tyrosine phosphatases, including SHP-2, PTPN22, DUSP5 and DUSP6 which in turn results in activation of TCR signaling molecules Lck and Erk [68] (Fig. 1).

miR-17-92 cluster

The miR-17-92 cluster, which comprises six miRNAs (miR-17, miR-18a, miR-19a, miR-20a, miR-19b-1, and miR-92-1), is highly expressed in B and T precursor cells but its expression decreases after maturation. Mice with targeted deletion of miR-17-92 cluster die rapidly after birth. This was likely due to severely hypoplastic lungs and ventricular septal defects [56]. In hematopoiesis, the absence of this cluster inhibits B-cell development at the pro-B to pre-B transition by increasing the levels of the proapoptotic protein Bim, a miR-17-92 target [56]. These experiments suggest that the miR-17-92 cluster acts specifically during the transition from pre-B to pro-B lymphocyte development, enhancing the survival of the B cells at this stage by targeting Bim. Consistent with this data, mice with targeted over-expression of the miR-17-92 cluster in the lymphocyte compartment develop severe lymphoproliferative disorder and autoimmunity [69]. These mice have decreased levels of Bim but also of the pro-apoptotic PTEN, which is another confirmed miR-17-92 target with critical functions in lymphopoiesis (Fig. 1).

Other miRNAs

It has been reported that miR-146a expression might be involved in cell fate determination in mouse lymphocytes, because its level is substantially increased in T helper 1 cells and decreased in T helper 2 cells relative to its expression in naïve T cell [67, 70] (Fig. 1). In addition systematic profiling of miRNAs was also performed in naïve, memory and effector T cells, revealing that very few miRNAs are dynamically up-regulated on antigen specific T-cell differentiation [71].

Immune function

miR-155.

MiR-155 is processed from a primary transcript called B-cell integration cluster encoded by a gene originally isolated near a common retroviral integration site in avian leucosis virus-induced lymphomas [72]. Although miR-155 is dispensable for lymphocyte development, it has crucial roles in the modulation of T- and B-cell responses in vivo. Mice deficient for B-cell integration cluster/miR-155 are viable, fertile, exhibits early lung inflammation, and have a dampened innate immune response [35]. Impaired B-cell responses were evidenced by significantly reduced amount of immunoglobulin M (IgM) and switched antigen-specific antibodies. Furthermore, miR-155 deficient mice were less responsive to immunization [35]. Using both, knock out and knock in animal models, Thai et al. [73] showed that miR-155 plays a critical role in the formation and activity of germinal centers in the lymph nodes and in T-cell dependent antibody responses. T cells from miR-155 deficient mice failed to produce significant levels of interleukin-2 and interferon-γ and there was a bias toward Th2 differentiation, suggesting that miR-155 promots T helper type 1 responses. Overall, these broad miR-155 immune effects are mediated in part by controlling cytokine production. However, other miR-155 targets are likely to be important, such as the transcription factor PU.1, which regulates immunoglobulin switching or c-Maf (T-cell responses) [74]. A recent study also indicates that miR-155 may be involved in the regulation and function of T regulatory cells in concert with the transcription factor FOXP3 [75, 76].

Two recent articles reported that miR-155 negatively regulates somatic hypermutation and Class-switch recombination of the immunoglobulin locus by targeting the activation-induced cytidine deaminase (AID) protein in B cells [77, 78]. AID plays a critical role in the regulation of these processes, assuring the generation of a diverse antibody repertoire. Disruption of the interaction between miR-155 and AID, resulted in increased class switch recombination, defective affinity maturation and genomic instability as evidenced by the association with Myc-related translocations [77, 78]. This stunning finding is in concordance with a previous report indicating that AID was required for MYC translocations that occurred in the context of IL6 transgene expression [79]. Although, within this context miR-155 seems to protect from Myc-related translocations, mice deficient for miR-155 do not exhibit any B-cell malignancy suggesting that this mechanism by itself may not be sufficient to induce malignancy [35].

Innate immunity miRNAs

MiRNAs have also involved in the regulation of mechanisms that defend the host from infection by other organism, in a nonspecific manner (known as innate immunity processes).

Critical to these processes is the recognition of bacteria, viruses, and other pathogens by the Toll family of extracellular receptors and the initiation of intracellular signaling events, in particular NF-Kb activation that is ultimately critical to establish a defensive response. Three miRNAs, miR-155, miR-146, and miR-132, are dramatically up-regulated on treatment of monocytes with endotoxin lipopolysacharide (LPS) that mimics signaling by bacterial infection [80]. Interestingly, miR-146a and b are NF-Kb induced miRNAs, that target TNF receptor-associated factor 6 (TRAF6) and IL-1 receptor-associated kinase 1 (IRAK1) genes, both genes encoded for two key adapter molecules downstream of Toll-like and cytokine receptors [80]. This constitute a negative feedback system whereby bacterial components induce NF-κB, resulting in up-regulation of the miR-146 genes, which, on processing, down-regulate levels of IRAK1 and TRAF6 proteins, reducing the activity of the pathway. This creates a negative feedback loop to limit TLR signaling after exposure to extracellular ligand.

The up-regulation of miR-155 in response to LPS may play a critical role in inducing myeloproliferation following infection. O'Connell et al. [34] found a strong but transient induction of miR-155 in mouse bone marrow after injection of LPS that associated with granulocyte-macrophage expansion. Similar results were observed upon sustained expression of miR-155 in murine HSC hematopoietic reconstitution models [34].

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. References

A wealth of reports establish that miRNAs are part of complex regulatory networks involving transcriptional factors and cytokines that co-operate to govern the expression of the genetic program underlying the differentiation, proliferation and activity of hematopoietic cells. Although most of the studies were carried out analyzing a single miRNA or a cluster, it is likely that miRNAs may cooperate with other miRNAs in the regulation of these processes. To that end, the contribution of each miRNA to a certain phenotype would need to be systematically investigated by performing loss or gain of function in vivo studies.

Notable, a single miRNAs can act on several hematopoietic lineages and exert completely different functions stressing the notion that miRNA effects are dependent on the cell context and the expression of the target genes within that cell.

Over the next years, research should focus on developing loss or gain of function animal models to provide definitive information about miRNA functions during hematopoiesis. Studies based only on cell lines or in vitro cultures of primary cells have proven to be inconsistent or insufficient for obtaining definitive answers. However, it will be important to recognize that these discrepancies may have a biological meaning. There are numerous examples of therapeutic modification of miRNA in adult mice by administering antagomirs, miRNAs mimics or via retroviral approaches that elicits different effects than germline manipulation of miRNA genes. These differences may be due to experimental artifacts such as off target effects, but they also could be due to differences in how organisms handle miRNAs at different times in development, such as compensatory mechanisms. It may be important to elucidate and understand such differences, because manipulating miRNAs in adult organisms is a potential therapeutic possibility.

The advent of high throughput small RNA sequencing will allow the discovery of novel miRNAs and other non coding RNAs expressed during hematopoiesis that will require further functional characterization. The integration of miRNA expression with other datasets, including whole mRNA expression, genome-wide DNA sequencing, and epigenome studies may uncover functional relationships and improved overall the understanding of non coding RNAs in the regulation of hematopoiesis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. References