Generation of hematopoietic repopulating cells from human embryonic stem cells independent of ectopic HOXB4 expression

Abstract

Despite the need for alternative sources of human hematopoietic stem cells (HSCs), the functional capacity of hematopoietic cells generated from human embryonic stem cells (hESCs) has yet to be evaluated and compared with adult sources. Here, we report that somatic and hESC-derived hematopoietic cells have similar phenotype and in vitro clonogenic progenitor activity. However, in contrast with somatic cells, hESC-derived hematopoietic cells failed to reconstitute intravenously transplanted recipient mice because of cellular aggregation causing fatal emboli formation. Direct femoral injection allowed recipient survival and resulted in multilineage hematopoietic repopulation, providing direct evidence of HSC function. However, hESC-derived HSCs had limited proliferative and migratory capacity compared with somatic HSCs that correlated with a distinct gene expression pattern of hESC-derived hematopoietic cells that included homeobox (HOX) A and B gene clusters. Ectopic expression of HOXB4 had no effect on repopulating capacity of hESC-derived cells. We suggest that limitations in the ability of hESC-derived HSCs to activate a molecular program similar to somatic HSCs may contribute to their atypical in vivo behavior. Our study demonstrates that HSCs can be derived from hESCs and provides an in vivo system and molecular foundation to evaluate strategies for the generation of clinically transplantable HSC from hESC lines.

Figure 1.

Generation of hESC-derived hematopoietic cells from hemogenic precursors and comparison with somatic hematopoietic cells using in vitro and in vivo assays. (A) Derivation of a cell population isolated from d 10 hEBs that lacks CD45 but expresses PECAM-1, Flk1, and VE-cadherin (termed CD45^negPFV) (3). After 7 d of culture in serum-free medium containing hematopoietic cytokines (SCF, FLT3L, IL-3, IL-6, and G-CSF), these hemogenic precursors give rise to hematopoietic cells (CD45⁺) and a substantial proportion of primitive hematopoietic cells (CD45⁺/CD34⁺). (B) Purified CD45⁺/CD34⁺/CD38⁻/Lin⁻ hematopoietic cells sorted from somatic CB (n = 8) and hESC-derived hematopoietic cells (n = 6) demonstrate similar CFU capacity and (C) CFU type. (D) 8-wk survival of NOD/SCID mice receiving i.v. injection of 5 × 10⁵–1.6 × 10⁷ somatic hematopoietic cells over. (E) Level of human chimerism in the BM of recipient NOD/SCID mice surviving to 8 wk transplanted with somatic or hESC-derived hematopoietic cells.

Figure 2.

In vivo pulmonary emboli and aggregation of hESC-derived hematopoietic cells in response to mouse serum. Hematoxylin and eosin–stained lung cross-sections of NOD/SCID recipient mice receiving (A) somatic hematopoietic cells and (B) hESC-derived hematopoietic cells, 24 h after i.v. transplantation. Normal lung structures and pulmonary capillary (dashed lines) were observed in mice receiving somatic hematopoietic cells, but mice receiving a similar dose of hESC-derived hematopoietic cells had numerous emboli in the serial lung cross sections. Bar, 10 μm. A representative embolus (B, inset) lodged in the pulmonary capillary (dashed line) causes blood vessel obstruction (B and C). The human origin of pulmonary emboli was confirmed by i.v. transplantation of GFP-expressing hESC-derived hematopoietic cells (C, inset; bar, 10 μm; green, GFP-expressing hESC-derived hematopoietic cells; blue, nucleus stained by DAPI [arrows]). (D) Size and complexity of somatic and hESC-derived hematopoietic cells were evaluated by flow cytometric measurement and by direct comparison of physical size by microscopy. No differences in these properties were detected between somatic and hESC-derived hematopoietic cells. Bar, 50 μm. (E) Microscopic examination for cellular aggregation of hESC-derived hematopoietic cells and somatic hematopoietic cells after 1–2 h in vitro treatment with or without serum from different mouse and human sources. Addition of adult mouse serum did not trigger aggregation of somatic hematopoietic cells but did cause rapid aggregation of hESC-derived hematopoietic cells. Bar, 50 μm. (F) Quantitative analysis of cellular aggregation indicated that up to 80% of hESC-derived hematopoietic cells aggregated in response to addition of adult mouse and rat serum (n = 7); 32% aggregated in response to human adult serum (n = 3); 21% aggregated in response to human neonatal CB serum (n = 3); 12% aggregated in response to human fetal blood serum (n = 3); and 10% aggregated in response to FBS (n = 3). All sera tested were used at 20% by volume.

Figure 3.

Femoral IBMT of hESC-derived hematopoietic cells in NOD/SCID mice. (A) 8-week survival of NOD/SCID recipients receiving somatic or hESC-derived hematopoietic cells by IBMT. (B) Southern blot analysis of recipient mouse BM DNA using the human-specific α-satellite sequences to demonstrate human chimerism from independent experiments as indicated. (C) Representative example of multilineage human hematopoietic (CD45⁺) reconstitution of BM cells from the injected femur of NOD/SCID recipients receiving hESC-derived hematopoietic cells that include lymphoid, myeloid, and erythroid lineages. (D) PCR analysis of recipient mouse DNA extracted from BM cells using the human-specific α-satellite sequence to demonstrate human chimerism in the injected femur, contralateral femur, and other long bones. (E) Summary of level of human chimerism in individual mice transplanted with somatic or hESC-derived hematopoietic cells from hemogenic CD45^negPFV precursors. (F) Comparative analysis of average level of human chimerism in injected femur, contralateral femur, and other long bones.

Figure 4.

Comparative analysis of the molecular profile of candidate hESC-derived versus somatic hematopoietic stem cells. Hematopoietic cells derived from both hESC-hemogenic precursors and human UCB were isolated based on the expression of CD34 and absence of CD38 (CD34⁺/CD38⁻/CD45⁺). Total RNA from both hESC-derived and somatic stem cell populations was extracted, and amplified RNA was generated to hybridize human HG-U133AB Affymetrix chips. Expression levels of hESC and somatic CD34⁺/CD38⁻/CD45⁺ cells were calculated relative to undifferentiated hESC cells. Differentially regulated genes were defined as those being up-regulated more than twofold and being statistically significant (P < 0.01). Genes differentially expressed in primitive hematopoietic cells derived from hESC and UCB were categorized as those involved in (A) cell–cell contact and migration, (B) cell replication, (C) transcriptional regulators, and (D) HOX genes. A complete dataset of the differential microarray profile is provided in Table S1.

Figure 5.

Ectopic expression of HOXB4 induces proliferation but does not confer engraftment potential on hESC-derived hematopoietic cells. (A) Control vector (GFP) and HOXB4 bicistronic retroviral construct. HOXB4 cDNA was subcloned into the control vector backbone upstream of an IRES. Enhanced GFP (eGFP) cDNA downstream of IRES sequence acts as a reporter for selection of stable cell lines and tracking of transduced cells. (B) Western blot analysis of PG13 packaging cell line transduced with vector or HOXB4 retrovirus showing specific HOXB4 expression from HOXB4-containing retrovirus. (C) Quantitative RT-PCR confirmed that HOXB4-transduced cells had up to 300-fold increase in HOXB4 expression as compared with controls shown. (D) HOXB4-overexpressing CB Lin⁻/CD34⁺ cells show a threefold in vitro proliferative advantage after 4 d culture as compared with their GFP-expressing counterparts (n = 4). (E) Vector and HOXB4-transduced CB Lin⁻/CD34⁺ were seeded in methylcellulose assays, and CFU potential was evaluated 14 d after plating. (F) A representative example of human engraftment in NOD/SCID mice repopulated with either vector or HOXB4-transduced SRC from CB (n = 6). CD45⁺ human cells were gated, and the extent to which vector and HOXB4-transduced SRC contributed to the engraftment was analyzed. Vector (top) and HOXB4-transduced human cells (GFP⁺CD45⁺) (bottom) were gated and further analyzed for immature (CD34⁺), myeloid (CD33⁺/CD13⁺) and B-lymphoid (CD19⁺) composition. (G) hESC-derived hematopoietic cells were transduced with vector versus HOXB4-expressing retrovirus, and in vitro proliferation was assessed 4 d after viral exposure. Similar to CB cells, HOXB4-overexpressing hESC-derived hematopoietic cells displayed a 2.5-fold higher expansion than vector-expressing cells but retained equivalent survival measured by the percentage that was 7-AAD–negative (n = 4). (H) Vector or HOXB4-transduced hESC-derived hematopoietic cells were plated in methylcellulose assays to test CFU potential. (I) Hematopoietic progeny differentiated from HOXB4-transduced CD45^negPFV precursors maintain their hematopoietic phenotype. Gene transfer efficiency into hESC-derived hematopoietic cells was 22% on average and was 10.5% for GFP and HOXB4, respectively. A representative experiment of the phenotypic analysis of vector- (top) and HOXB4-transduced (bottom) hematopoietic progeny differentiated from hESCs is shown for cell surface CD45 and CD34 expression. H and I show that HOXB4 overexpression does not alter the in vitro developmental capacity of CD45^negPFV precursors. (J) Human chimerism in the marrow flushed from the different mice bones was not detectable. DNA from one mouse transplanted with equivalent number of CB Lin⁻/CD34⁺ cells was used as a positive control as shown. (K) Summary of the number of cells transplanted and the frequency of human chimerism detected in the BM of recipient mice from vector- or HOXB4-transduced hematopoietic cells.