Expression of Scl in mesoderm rescues hematopoiesis in the absence of Oct-4

Oct-4 is a POU transcription factor that is highly expressed in embryonic stem cells (ESCs), embryonic epiblasts, and primordial germline cells.^1-3  The precise level of Oct-4 tightly regulates the differentiation capacity of ESCs. Loss of Oct-4 expression in ESCs results in a loss of pluripotency and the induction of trophectoderm differentiation.^4,5  In contrast, doubling Oct-4 levels in ESCs induces differentiation toward the mesodermal and endodermal embryonic lineages.⁴  Oct-4 is normally expressed in early tissues, including oocyte, blastocysts, and embryonic ectoderm before gastrulation.^1,2  Oct-4 is absent in hematopoietic stem cells and is only detected at high levels in germ tissue of adults.^2,6  Maintaining expression of Oct-4 during ESC differentiation has been shown to have effects on neurogenesis and hematopoiesis.^7,8 

In this study, we have used an ESC line (ZHBTc4) in which both endogenous Oct-4 alleles have been deleted and a tetracycline-repressible Oct-4 transgene has been introduced.⁴  When Oct-4 expression was repressed at the beginning of differentiation (day 0), formation of embryoid bodies (EBs) was severely hindered and there was a decrease in the expression of hematopoietically associated genes. Similar results were observed when we stably transfected the cells with the basic helix-loop-helix transcription factor stem cell leukemia (Scl).^9,10  However, if Oct-4 was allowed to express for 48 hours after induction for differentiation, the expression of Scl was able to rescue the ESCs to form functional hematopoietic EBs. Our study demonstrates that Oct-4 is essential for at least 2 different roles in hematopoiesis. It must first specify brachyury-positive (bry⁺) mesoderm and then facilitate the proper hematopoietic development of uncommitted mesoderm.

ZHBTc4 ESCs were generated and cultured as previously described.⁴  Full-length Scl cDNA was subcloned into the pCAPP chicken β-actin promoter-containing vector. ZHBTc4 cells were electroporated with pCAPP parental or Scl/pCAPP vector using the Bio-Rad Gene Pulser (240 V, 500 μFD; Bio-Rad). Puromycin (2.5 μg/mL; Sigma-Aldrich) was added to culture medium 48 hours after electroporation for clonal selection. All ZHBTc4 and ZHBTc4-derived ESCs (pCAPP control and Scl clones) are Oct-4 knockout cells with tet-regulated Oct-4 transgene.

Total RNA was extracted from multiple clones of pCAPP control or Scl ZHBTc4 clones using RNeasy (QIAGEN). First-strand cDNA was synthesized using the Invitrogen cDNA synthesis kit according to the manufacturer's instructions (Invitrogen). Primers used for polymerase chain reaction (PCR) reactions are shown in supplemental Table 1 (available on the Blood website; see the Supplemental Materials link at the top of the online article).¹¹ 

Total protein was extracted from undifferentiated ESCs or differentiating EBs using 1× lysis buffer (50 mM Tris-Cl, pH 7.8, 150 mM NaCl, 1 mM ethylenediaminetetraacetic acid, 1 mM dithiothreitol, 0.5% NP-40 Alternative; Calbiochem; and 1× protease inhibitor cocktail, Roche Diagnostics). A total of 60 μg of each protein sample was used for sodium dodecyl sulfate–polyacrylamide gel electrophoresis and subsequently transferred to a polyvinylidene difluoride membrane for immunoblotting. Primary antibodies used were anti–Tal-1 polyclonal (Scl; Calbiochem), anti–Oct-4 monoclonal (Santa Cruz Biotechnology), antihemoglobin-α polyclonal (Santa Cruz Biotechnology), and anti–β-actin monoclonal (Sigma-Aldrich). Secondary antibodies were donkey anti–mouse IgG conjugated with horseradish peroxidase (HRP; GE Healthcare) and sheep anti–rabbit IgG HRP-conjugated (GE Healthcare).

Embryoid bodies were generated from ESCs either in liquid culture or in semisolid media. For liquid culture, ESCs were cultured in primary differentiation medium according to an established protocol.¹²  For differentiation in semisolid media, pCAPP control or Scl clones were cultured in conditions as previously described.¹³ 

Photomicrographs of EBs formed after 12 to 14 days cultured in methylcellulose medium were taken by Axioskop fluorescence microscopy (Carl Zeiss) via a 10×/0.3 NA air objective. Images taken via AxioCam camera (Carl Zeiss) were analyzed with Slidebook software (Olympus).

Day 14 EBs were collected and disaggregated in 0.25% collagenase at 37°C for 1 hour. Single-cell suspensions were collected, washed, and incubated with 100 μL antibody cocktail (CD45-phycoerythrin and CD11b-allophycocyanin, 1:100; BD Biosciences) at ambient temperature for 30 minutes. Cells were washed once and resuspended in phosphate-buffered saline for fluorescence-activated cell sorter analysis (FACSCalibur; BD Biosciences). Data collected were analyzed by FlowJo software (TreeStar).

Immunocytochemistry was performed as previously described.¹⁴ 

In pluripotent ESCs, any disruption of Oct-4 natural expression pattern results in an increase or decrease of their mesodermal commitment.^4,7  To investigate whether there is a possible role for Oct-4 during hematopoietic differentiation, we systematically switched off Oct-4 at different time points using a tet-regulated, Oct-4 knockin ESC line (ZHBTc4) and examined EB generation. Both EB formation and hematopoietic development were hampered when Oct-4 was turned off at day 0 (supplemental Figure 1A). Reverse-transcribed (RT) PCR analyses showed both mesodermal and hematopoietic markers (Scl, BMP4, Brachyury [bry], flk1, PU.1, EKLF, embryonic globin, and EpoR) were repressed (supplemental Figure 1B).

Because Scl is an early-acting factor required for hematopoietic stem cell development from mesoderm, we generated Scl-expressing ZHBTc4 ESCs (Scl clones) to determine whether Scl clones could bypass the requirement for Oct-4 in hematopoiesis (supplemental Figure 2).¹⁰  Scl failed to rescue hematopoiesis when Oct-4 was silenced at day 0 of differentiation (Figure 1A). Major mesodermal and hematopietic markers (bry, PU.1, and EKLF) were down-regulated in both tet-treated control and Scl clones (supplemental Figure 3). However, an increased expression in various globin genes (α-, ζ-, and βmaj-) was noted from the tet-treated Scl ESCs but not the tet-treated control (supplemental Figure 3).

To identify the time point when Oct-4 becomes dispensable for EB and hematopoietic development, we silenced Oct-4 in differentiating ESCs at day 2 or 4 and examined EB development. No hematopoietic EBs could be derived from control ESCs at any time point, although the size of day 4 tet-treated EBs was larger than day 0 and 2 tet-treated EBs (Figure 1A). In Scl ESCs, hemoglobinized red cells were seen in both day 2 and day 4 tet-treated EBs (Figure 1A), and expression of hemoglobin protein in these EBs was confirmed by immunoblot analyses (Figure 1B). The presence of myeloid progenitor population (CD45⁺CD11b⁺ cell population) in day 2 and day 4 tet-treated Scl EBs was demonstrated by flow cytometry (Figure 1C-D) and immunostaining (CD11b⁺; Figure 1E). RT-PCR analysis verified that expression of hematopoietic genes (PU.1, LMO2, Gata1, α-, ζ-, and βmaj-goblins) was restored in Scl EBs when Oct-4 was silenced at day 2 or day 4 of differentiation (Figure 2).

Our data support the notion that, without Oct-4, there is no generation of bry⁺ mesoderm and thus no mesoderm-derived hematopoietic progenitors. However, if we allow Oct-4 to be expressed for 2 or 4 days after differentiation induction, bry⁺ progenitors are formed and can be induced to hematopoietic tissue by Scl. In the absence of exogenous Scl, these bry⁺/Oct-4⁻ progenitors are not induced to become hematopoietic cells. However, silencing Oct-4 at day 6 in control EBs does not inhibit hematopoiesis (not shown). The requirement for Oct-4 correlates with the time (days 2.5-4.0 after differentiation induction) when the early hematopoietic progenitors (hemangioblasts/BL-CFCs) are generated in the EBs.¹⁵ 

In conclusion, we have revealed a novel role for Oct-4 in embryonic development beyond the initial specification of early bry⁺ mesoderm. By shutting off Oct-4 expression immediately after the formation of early mesoderm, we have shown that Oct-4 is necessary for hematopoietic development. However, this requirement can be bypassed by the expression of Scl in mesoderm.

This work was supported by the National Institutes of Health (Bethesda, MD; CA102283, HL075783; R.H.), the American Cancer Society (Atlanta, GA; RSG-06-170-01; R.D.), the American Society of Hematology (Washington, DC; Junior Faculty Scholar Award; R.D.), and the Leukemia & Lymphoma Society (White Plains, NY; SCOR 7388-06; R.H., R.D.).

National Institutes of Health

Contribution: K.Y.K. performed experiments, prepared data for publication, contributed conceptually, and prepared the manuscript; E.A.W., J.H.R., and T.T. performed experiments; and R.H. and R.D. contributed conceptually, supervised experiments, and edited the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Richard Dahl, Department of Internal Medicine and the Cancer Research and Treatment Center, University of New Mexico Health Sciences Center, 900 Camino de Salud, CRF 125A, MSC10 5550, Albuquerque, NM 87131; email: rdahl@salud.unm.edu.