Brief Report: Ectopic Expression of Nup98-HoxA10 Augments Erythroid Differentiation of Human Embryonic Stem Cells§


  • Disclosure of potential conflicts of interest is found at the end of this article.

  • Author contributions: J.J. and R.M.R.: conception and design, analysis and interpretation, manuscript writing; S.H.: collection of data; D.A., P.R., and K.H.: provision of study material; M.B.: conception and design, analysis and interpretation, manuscript writing. J.J. and R.M.R. contributed equally to this article.

  • §

    First published online in STEM CELLSEXPRESS February 15, 2011.


Hox genes encode highly conserved transcription factors that have been implicated in hematopoietic development and self-renewal of hematopoietic stem cells (HSCs) and hematopoietic development. The potency of NUP98-HOXA10hd (NA10) on adult murine bone marrow HSC self-renewal prompted us to examine its effect on specification and proliferation of hematopoietic cells derived from human embryonic stem cells (hESCs). Here, we demonstrate that expression of NA10 in hESCs influences the hematopoietic differentiation program. The specific effect of NA10 is dependent on the developmental stage of hematopoietic emergence from hESCs. Overexpression of NA10 in either undifferentiated hESCs or early hemogenic precursors augmented the frequency of CD45 GlycophorinA+ cells and erythroid progenitors (blast-forming unit-erythrocyte). In contrast, targeted NA10 expression in primitive CD34+ cells committed to the hematopoietic lineage had no effect on erythropoietic capacity but instead increased hematopoietic progenitor proliferation. Our study reveals a novel neomorphic effect of NA10 in early human erythroid development from pluripotent stem cells. STEM Cells 2011;29:736–741


Human embryonic stem cells (hESCs) are defined by their unique pluripotent properties and their ability to self-renew indefinitely in vitro [1]. These properties distinguish hESCs from all other tissue-specific stem cells [2]. The ability of hESCs to differentiate into virtually all possible cell types of the body endows them with a promising role in cell replacement therapy. However, to realize this potential, a fundamental understanding of the mechanisms that effectively control and direct differentiation of hESCs toward specific lineages must first be developed.

Control of stem cell fate is prevalently mediated by transcription factors [3]. Homeobox (Hox) genes represent highly conserved transcription factors [4] capable of regulating hematopoietic proliferation and lineage commitment [5]. Many Hox genes are expressed in both human and mouse primitive hematopoietic stem and progenitor cells and are downregulated during hematopoietic differentiation and maturation [1, 6, 7]. Multiple Hox genes have also been identified in human leukemia as fusion partners with a common partner, NUP98 [8]. Intriguingly, an engineered fusion of NUP98 and the homeodomain of HOXA10 (NA10) has been shown to be more potent than HOXB4 in stimulating in vitro expansion of murine hematopoietic stem cells (HSCs) without overt leukemogenic activity [6, 9, 10]. Given this potency, and previous studies implicating HOXA10 as a key regulator of primitive hematopoietic cell expansion and erythroid/megakaryocytic lineage choice [9, 11], we were interested in investigating the role of NA10 in regulating early hematopoiesis from hESCs. Here, we ectopically expressed NA10 in hESCs and defined cellular stages of hematopoietic development from hESCs. Our findings reveal a unique role of NA10 in regulating erythropoiesis and primitive hematopoietic progenitors at later stages of hESC hematopoiesis that provides a basis for applications using Hox genes and Hox fusions in hESC-hematopoietic development.


Culture of hESCs and Formation of Human Embryoid Bodies

Undifferentiated hESC lines H9 and H1 [12] were cultured on Matrigel (BD Biosciences, Bedford, MA, as previously described [13]. Formation and dissociation of human embryoid bodies (hEBs) were performed as previously reported by our group [13].

Isolation of Hemogenic Precursors and CD45+CD34+ Cells

The hemogenic precursor population from day 10 hEBs was purified based on the presence of CD31 and absence of CD45 [14]. CD45+CD34+ hematopoietic progenitors from day 15 hEBs were purified according to CD45 and CD34 expression. The hemogenic precursors and CD45+CD34+ cells were seeded on 96-well plates coated with fibronectin (BD Biosciences) and cultured for 7 or 4 days in serum-free hemogenic medium previously shown to sustain human HSCs [14].

Lentiviral Generation and Transduction

A 1.6-kb fragment encoding human NUP98 and HOXA10 homeodomain fusion gene from MIG FLAGNA10HD retroviral vector was subcloned into EF1a.GFP-PGK.PAC to replace GFP reporter gene (a gift from Dr. Benjamin E. Reubinoff; Jerusalem, Israel) and EF1a.GFP-EFla.Linker dual promoter lentiviral vectors, respectively. Lentiviral viruses were produced in 293FT cells (ATCC, Manassas, VA, as described [15]. Lentiviral transduction was performed following a previously reported protocol [16].

Flow Cytometry

Single cell suspensions were stained with anti-human CD45, CD31, CD34, and Glycophorin A antibodies (BD Biosciences). Samples were acquired using a FACSCalibur cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star Inc., Ashland, OR,

Colony-Forming Unit Assays

Cells indicated in the text were plated into methylcellulose H4230 (StemCell Technologies, Vancouver, Canada, supplemented with: 50 ng/ml stem cell factor, 3 U/ml erythropoietin, 10 ng/ml granulocyte–macrophage–colony-stimulating factor, and 10 ng/ml interleukin 3. Localized clusters of >50 cells showing morphological hematopoietic characteristics were counted as colonies after incubation for 10–14 days at 37°C and 5% CO2 in a humidified atmosphere.

Quantitative Polymerase Chain Reaction Analysis

mRNA was extracted using an mRNA extraction kit (Qiagen, Hilden, Germany, and was reverse-transcribed into cDNA using first-strand cDNA synthesis kit (Amersham Biosciences, Amersham, U.K.) according to the manufacturer's instructions. Expression of NA10 (Forward primers: 5′-GGGAACAACCAACCTAAGATTGGA-3′ and reverse primer: 5′-GCTTCTTCCGACCACTCTTGAATT-3′) was examined by regular polymerase chain reaction (PCR). Expression of GATA1 and β- and ε-hemoglobin [17] was quantified by quantitative-PCR. Signal intensities were normalized against glyceraldehyde 3-phosphate dehydrogenase [17] and the 2−ΔΔCt equation was used to calculate the relative gene expressions [18].

Statistical Analysis

Results were presented as mean ± SEM. Statistical significance was determined using an unpaired Student's t test and results were considered significant when p ≤ .05.


Ectopic Expression of NA10 Augments Erythroid Differentiation of hESCs

To investigate the effect of NA10 during hESC hematopoietic development, undifferentiated hESCs were first transduced with a lentiviral vector encoding NA10 fusion protein. After drug selection of transduced cells, EB-mediated hematopoietic differentiation was performed (Fig. 1A) [14, 17]. NA10 expression was sustained during EB formation, although at lower levels compared with undifferentiated stage (Supporting Information Fig. 2A). Both control and NA10-transduced hESCs gave rise to comparable frequencies and similar kinetic emergence of mature hematopoietic cells (CD45+ cells) during EB formation (Fig. 1B) [13]. However, a significant increase in erythropoiesis was detected in hESC-expressing NA10 by day 15 (Fig. 1C). Hematopoietic progenitor capacity, measured by the ability to form colony-forming units (CFUs), was comparable in NA10-transduced hESCs compared with control (Fig. 1D). Moreover, the relative frequency of each of the subtypes of CFUs was not affected by the overexpression of NA10 (Fig. 1D and Supporting Information Fig. 2C) indicating that clonal hematopoietic progenitor capacity to differentiate into the different blood cell types remained unaffected by NA10 expression.

Figure 1.

Overexpression of NA10 during EB formation leads to an increase in erythroid differentiation. (A): Undifferentiated hESCs were transduced with a control eGFP or NA10 vector, selected in puromycin, differentiated as EB, and analyzed. (B): Summary of CD45+ and (C) CD45-GlyA+ cell frequency during differentiation of control eGFP transduction (gray bars) and NA10 overexpressing (black bars) EBs. (D): Total number of CFUs derived from day 15 EBs. Abbreviations: BFU-E, blast-forming unit-erythrocyte; CFU, colony-forming unit; CFU-G, CFU-granulocyte; CFU-M, CFU-macrophage; eGFP, enhance green fluorescent protein; hESC, human embryonic stem cell.

Figure 2.

NA10 expression promotes the differentiation of red blood cells from hemogenic precursors. (A): Hemogenic precursors were isolated, transduced with the control or NA10 lentiviral vector, and the hemogenic potential was evaluated. (B): Representative flow profile of hematopoietic cells. (C): Frequency of RBCs (CD45-GlyA+) 7 days after transduction. (D): Total number of colonies derived from CD45-CD31+ cells transduced with control and NA10 construct. (E): Erythroid colonies. (F–K): Morphology of erythroid progenitors dissociated from individual CFU-erythrocytes. (L): Presence of NA10 transcripts in hemogenic precursors and individual CFUs. (M): GATA-1 and (N) ε-hemoglobin mRNA expression in hemogenic precursors and CFUs. Abbreviations: CFU, colony-forming unit; hESC, human embryonic stem cell; NTC, no template control; RBC, red blood cell.

NA10 Expression Promotes Generation of Erythroid Progenitors from Hemogenic Precursors but Not Committed Progenitors

Hemogenic precursors derived from hESCs represent a subpopulation of cells from human EBs, which lack CD45 but express PECAM, Flk-1, and VE-cadherin [14]. Identification of hESC-derived hemogenic precursors provides a model system for studying early events of human hematopoiesis from hemogenic precursors prior to hematopoietic commitment [14, 17] (Fig. 2A). Overexpression of NA10 in hemogenic precursors had no effect on the frequency of primitive hematopoietic-committed cells (CD34+CD45+) (Fig. 2B). However, NA10-transduced cells showed an increase in Glycophorin-A+-cells, indicative of augmented erythropoiesis (Fig. 2C). Moreover, overexpression of NA10 in hemogenic precursors enhanced their multilineage hematopoietic progenitor potential, as measured by the total number of hematopoietic progenitors (CFUs) (Fig. 2D). This overall increase in progenitor output was almost entirely due to augmented eythropoietic progenitors (Fig. 2D and Supporting Information Fig. 3D). In addition, cellularity of blast-forming unit-erythrocyte (BFU-E) colonies derived from NA10 cells was noticeably greater at a macrolevel compared with BFU-E transduced with control vector (Fig. 2E–2I). GATA-1, a key regulator in erythroid differentiation (Fig. 2M) [19], and ε- (Fig. 2N) and β-hemoglobin (Supporting Information Fig. 3E) were upregulated similarly to control. In addition, nuclei were detectable in differentiated erythroid cells (Fig. 2J and 2K), indicative of primitive erythropoiesis [20]. These results demonstrate that ectopic expression of NA10 in hESC-derived hemogenic precursors increased erythroid differentiation, overall erythroid progenitor number, and proliferative capacity of individual erythroid progenitors.

Figure 3.

NA10 expands hematopoietic progenitors and induces monocytic differentiation. (A): CD45+CD34+ hematopoietic progenitors were isolated, transduced with the control or NA10 lentiviral vector, and the hematopoietic potential was evaluated. (B): Cell expansion relative to cells originally seeded after 7 days in culture. (C): Frequency of hematopoietic progenitors (CD45+CD34+) and (D) erythrocytes (CD45-GlyA+) 7 days after transduction. (E): Total number and (F) subtypes of CFUs derived from hematopoietic progenitors transduced with NA10 vector and cultured during 7 days. Abbreviations: BFU-E, blast-forming unit-erythrocyte; CFU, colony-forming unit; CFU-G, CFU-granulocyte; CFU-M, CFU-macrophage; hESC, human embryonic stem cell.

NA10 Expands Hematopoietic Progenitors and Induces Monocytic Differentiation

As Hox gene expression patterns are tightly restricted to specific stages of in utero development [5], we analyzed the effect of NA10 on hematopoietic-committed hESCs. Hematopoietic CD34+CD45+ progenitors were isolated from hESC-derived EBs and transduced with NA10 (Fig. 3A). NA10 induced cell proliferation (Fig. 3B) but had little effect on the frequency of primitive hematopoietic progenitors (Fig. 3C), erythroid differentiation (CD45-Glycophorin-A+ cells) (Fig. 3D), or total CFU formation (Fig. 3E). However, a 32% increase in CFU-M and a 31% reduction in BFU-E relative to control (Fig. 3F) were observed on NA10 expression in CD34+CD45+ cells. Taken together, these results indicate that stable expression of NA10 in primitive cells committed to the hematopoietic lineage induces cell proliferation and favors monocyte differentiation while decreasing erythropoiesis.


Our study demonstrates that the effect of NA10 overexpression is dependent on the cellular target and its developmental stage during human hematopoietic specification from pluripotent state. Although ectopic expression of NA10 in hESC-derived hemogenic precursors augments erythroid progenitor output, no effect was observed on the generation of erythroid progenitors from committed hematopoietic CD45+CD34+ progenitors derived from hESCs. Our study indicates that NA10 acts on erythropoiesis initiated from early hemogenic precursors prior to hematopoietic commitment from pluripotent state of hESCs.

Mammalian erythropoiesis consists of two waves of development: (a) primitive erythropoiesis initiated in the yolk sac with the generation of large-nucleated erythroblasts and (b) definitive erythropoiesis arising from the fetal liver with the development of smaller enucleated erythrocytes. Primitive erythropoiesis is believed to originate from the hemangioblast, whereas hematopoietic-restricted progenitors give rise to definitive hematopoiesis, including erythrocytes [21]. Our results suggest that NA10 has a role in primitive, but not definitive, erythropoiesis as high levels of NA10 in undifferentiated hESCs increase erythropoiesis without affecting other hematopoietic lineages. When NA10 was expressed in hematopoietic precursor-like cells [14], the phenotype obtained was similar. However, NA10-expressing hematopoietic precursors, which give rise to definitive hematopoiesis, presented a higher self-renewal capacity and preferentially differentiated to myeloid lineages.

Our study reveals a previously unappreciated role of NA10 in promoting primitive erythropoiesis from hESCs. In addition, our results highlight the potential use of hemogenic precursors for generation of red blood cells in vitro instead of hematopoietic progenitors. Hemogenic precursors, in contrast to fully mature cells, maintain their proliferation capacity thus fewer cells may be required for a successful transfusion, similar to recent evidence where pancreatic beta cell progenitors derived from hESCs provided effects superior to transplanting mature beta cells [22]. However, additional work would be needed to fully characterize the potential clinical benefit of this approach as the erythroid cells maintained the nucleus and expressed embryonic globins and includes development of viral-free expression of NA10 and testing the in vivo potential. These avenues are currently being pursued in our laboratory.


Ectopic expression of NA10 augments erythroid differentiation of hESCs. NA10 expression promotes generation of erythroid progenitors from hemogenic precursors but not committed hematopoietic progenitors. NA10 expands resulting hematopoietic progenitors and favors monocytic differentiation.


This work was supported by grants from the Canadian Institute of Health Research (CIHR), National Centres of Excellence-StemCell Network, Canadian Cancer Society Research Institute (CCSRI) to M.B. M.B. is supported by the Canadian Chair Program and holds the Canada Research Chair in human stem cell biology. A fellowship from CCSRI supports R.M.R. Additional funding to R.K.H. was provided by the Terry Fox Foundation and the Canadian StemCell Network. We thank Bonan Zhong for his help in the experimental design and Clinton Campbell, Chelsea Maedler-Kron, Eleftherios Sachlos, and Eva Szabo for their invaluable insights during the preparation of the article.

Disclosure of Potential Conflicts of Interest

The authors indicate no potential conflicts of interest.