Comparative gene expression in hematopoietic progenitor cells derived from embryonic stem cells

Experimental Hematology 30 (2002) 58–66

Comparative gene expression in hematopoietic progenitor cells derived from embryonic stem cells Shi-Jiang Lu, Fei Li, Loyda Vida, and George R. Honig Department of Pediatrics, College of Medicine, University of Illinois at Chicago, Chicago, Ill., USA (Received 29 June 2001; revised 21 August 2001; accepted 30 August 2001)

Objective. The aim of this study was to characterize at the molecular level the hematopoietic progenitor cells derived from rhesus monkey embryonic stem (ES) cell differentiation. Materials and Methods. We purified CD34 and CD34CD38 cells from rhesus monkey ES cell cultures and examined the expression of a variety of genes associated with hematopoietic development, by semiquantitative polymerase chain reaction analysis. For comparison, we examined cell preparations from fresh or cultured rhesus monkey bone marrow (BM) and from mouse ES cells and BM. Results. We observed a high degree of similarity in the expression patterns of these genes, with only a few exceptions. Most notably, the message of the flt3 gene was undetectable in rhesus monkey ES cell-derived CD34 and CD34CD38 cells, whereas substantial flt3 expression was observed in the corresponding cells from fresh BM and in CD34 cells from cultured BM. The integrin L and interleukin-6 (IL-6) receptor genes also were expressed in CD34CD38 cells from BM, but there was little or no expression of these genes in CD34CD38 cells derived from ES cells. Parallel analyses, using CD34Lin cells derived from murine ES cell cultures, showed no apparent expression of flt3, integrin L, or IL-6 receptor, whereas corresponding cell preparations isolated from mouse BM expressed high levels of all of these genes. Conclusions. ES cell-derived hematopoietic progenitors, both from the rhesus monkey and from the mouse, exhibited the same alterations in gene expression compared with BM-derived cells from these animals. These observations could reflect the presence of different subpopulations in the cell fractions that were compared, or they may represent altered biologic properties of ES cell-derived hematopoietic stem cells. © 2002 International Society for Experimental Hematology. Published by Elsevier Science Inc.

Long-term cultures of pluripotent embryonic stem (ES) cells have been established from a variety of mammalian species, including nonhuman primates and man [1–4]. Under appropriate culture conditions, ES cells undergo differentiation in vitro to form hematopoietic precursors [5–7]. From immunofluorescence analyses of rhesus monkey ES cell-derived hematopoietic progenitor cells, we have determined that a substantial percentage of these cells are CD34, and that they have morphologic features of undifferentiated blast cells. The hematopoietic character of these precursors is supported further by the demonstration that

Offprint requests to: George R. Honig, M.D., Ph.D., Department of Pediatrics, M/C 856, University of Illinois at Chicago, 840 South Wood Street, Chicago, IL 60612; E-mail: [email protected]

they express genes associated with early hematopoietic differentiation, and that they exhibit morphologic findings of erythroid and myeloid lineages among their progeny cells [7]. These results are entirely comparable to those from observations of murine ES cell differentiation in vitro [5,6]. Among the distinctive characteristics of hematopoietic stem cells (HSC) is their ability to repopulate bone marrow (BM)-ablated animals. HSC homing and engraftment now are understood to be complex processes in which a number of genes play important roles [8,9]. Studies of the hematopoietic differentiation of mouse ES cells in culture, as well as of gene knockout models, also have identified a variety of genes that are linked to the ontogeny of HSCs. These observations have defined genes whose functions are critical for the initiation, commitment, and progression of hematopoiesis [10–14].

0301-472X/02 $–see front matter. Copyright © 2002 International Society for Experimental Hematology. Published by Elsevier Science Inc. PII S0301-472X(01)0 0 7 6 7 - 6

S.-J. Lu et al./Experimental Hematology 30 (2002) 58–66

In an effort to characterize further the hematopoietic progenitor cells derived from rhesus monkey ES cell differentiation, we initiated experiments to analyze the expression patterns of multiple genes in CD34 and CD34CD38 cells derived from ES cells, in relation to those from comparable cell populations isolated from BM. The genes we examined included those associated with hematopoietic differentiation, HSC homing, and engraftment. Most HSCs in BM are believed to be in the G0 resting state. In contrast, ES cell-derived hematopoietic precursors, which have been exposed to hematopoietic growth factors included in the culture media, are in active growth, which could confound comparisons of gene expression between them. To circumvent this potential limitation, we also examined gene expression of CD34 and CD34CD38 cells purified from BM that had been maintained under the same culture conditions as those used for ES cell hematopoietic differentiation.

Materials and methods Purification of rhesus monkey CD34 and CD34CD38 cells The maintenance and induction of hematopoietic differentiation of R366.4 rhesus monkey ES cells were as previously described [7]. CD34 cells were isolated from differentiated ES colonies with biotinylated anti-human CD34 antibody (clone 12.8; Baxter) using a Coulter flow cytometer. For CD34CD38 cell preparation, CD38 cells first were depleted using anti-human CD38 antibody (clone OKT10; American Type Culture Collection [ATCC], Manassas, VA, USA), then CD34CD38 cells were sorted by two color parameters with a Coulter flow cytometer. BM from rhesus monkeys (age 2 to 8 years) was obtained through the animal tissue distribution network from the Wisconsin Regional Primate Center and from the Biologic Resources Laboratory of the University of Illinois at Chicago. Purified mononuclear cells were resuspended in Dulbecco’s modified Eagle medium with 10% fetal bovine serum and divided into two parts. One part was used for isolation of CD34 and CD34CD38 cells as described earlier. The other aliquot was incubated in ES differentiation media for 3 days and sorted for the purification of CD34 and CD34CD38 cells. Figure 1 shows the FACS profiles that define the gates for purification of the CD34CD38 cells. Purification of mouse CD34Lin cells The mouse D3 ES line was obtained from the ATCC. The D3 ES cells (3000/well) were cultured in Iscove’s modified Dulbecco medium supplemented with 15% fetal bovine serum, and the combination of recombinant murine stem cell factor (40 ng/mL), interleukin-3 (IL-3; 40 ng/mL), IL-6 (20 ng/mL), vascular endothelial growth factor (VEGF; 40 ng/mL), granulocyte colony-stimulating factor (G-CSF; 40 ng/mL), flt3 ligand (FL) (20 ng/mL), and erythropoietin (Epo; 2 U/mL) (R & D Systems). At day 13 of differentiation, hematopoietic-like clusters were collected and washed three times with phosphate-buffered saline supplemented with 2% bovine albumin. Cells were stained with fluorescein isothiocyanateconjugated anti-mouse CD34 (clone RAM34; B-D PharMingen), and biotinylated anti-mouse lineage panel (containing anti-CD3e, CD11b, CD45R/B220, Ly6G, and TER-119; Baxter) coupled with

59

streptavidin-phycoerythrin. CD34Lin cells were sorted with a Coulter flow cytometer. BM cells were obtained by flushing the tibias and femurs of strain 129/sv mice (The Jackson Laboratory). CD34Lin cells were purified from mononuclear cells as described earlier. Cytoplasmic RNA isolation and cDNA pool construction A total of 3  106 CD34 cells (purity 97.5%) from fresh rhesus monkey BM, 5  105 cells (purity 98%) from cultured BM, and 2.8  105 cells (purity 98%) from rhesus monkey ES cell cultures were prepared. Cytoplasmic RNA was isolated from each preparation of purified CD34 cells, using an RNeasy Mini Kit (Qiagen), following the procedure recommended by the supplier. The RNA was resuspended in a minimum volume of RNase-free water (30 L). RNA, 3 L, was used for first-strand cDNA synthesis with SMART II and CDS primers (Clontech), using Superscript II reverse transcriptase (BRL). cDNA pools were constructed using the SMART cDNA synthesis kit (Clontech). cDNA pools generated by the SMART procedure have been shown to preserve the relative abundance relationship in original mRNA populations [15], and this procedure has been used successfully in the construction of cDNA pools using less than 1000 cells [16]. A total of 1.1  105 CD34CD38 cells (purity 95.4%) from fresh rhesus monkey BM, 1.5  105 cells (purity 97.1%) from cultured BM, and 8  103 cells (purity 96%) from rhesus monkey ES cultures were obtained. Cytoplasmic RNAs were purified from the CD34CD38 cells, and cDNA pools were constructed as described earlier. A total of 1.7  103 CD34Lin cells (purity 95.1%) from fresh mouse BM and 9  103 cells (purity 95.7%) from D3 ES cell cultures were obtained. Cytoplasmic RNA isolation and cDNA pool construction were performed as described earlier. Five microliters of each of the cDNA pools was diluted 100fold with TE buffer, and various quantities (1, 2, 4, 8, 16, and 32 L) of the diluted cDNA pools were used for polymerase chain reaction (PCR) amplification, using primers of the hypoxanthine phosphoribosyltransferase (HPRT) and -tubulin genes, in a total volume of 50 L. Ten microliters of the PCR products were separated on 1.5% agarose gels and stained with ethidium bromide. The signal intensity of the corresponding bands was estimated visually under an ultraviolet transilluminator. Based on the signal intensities of the HPRT and -tubulin bands, the DNA template quantities in the cDNA pools were equalized by dilution with TE buffer. Gene expression quantification by semiquantitative PCR Inasmuch as many of the genes we undertook to study have not been cloned from the rhesus monkey, whenever rhesus monkeyspecific sequences were unavailable, we used the sequences of the corresponding human genes to design our PCR primers, assuming that the generally close homology between human and rhesus monkey would compensate for minor differences in gene sequences in PCR amplification [17]. The sense and anti-sense primer sequences and the corresponding cDNA PCR product sizes are shown in the Supplemental Table (see web posting at www.elsevier.com/locate/exphemonline). Cytoplasmic RNA (5 g) from fresh rhesus BM was reverse transcribed to singlestranded cDNA and used to optimize PCR conditions for most of the genes, except for the VE-cadherin (VE-CA) and von Willebrand factor (vWF) genes, for which total human placental RNA (Clontech) was used. In general, the PCR reaction was performed

60

S.-J. Lu et al./Experimental Hematology 30 (2002 ) 58–66

Figure 1. Immunofluorescence flow cytometric gate profiles for CD34CD38 cells derived from fresh (I) and cultured (II) rhesus monkey bone marrow, and ES cell differentiation (III). Cells in gate R were defined as CD34CD38 cells.

in a volume of 50 L containing 20 mM TRIS-HCl (pH 8.0), 50 mM KCl, 0.2 mM dNTP, 0.25 g of each primer, and 2.5 U of Taq DNA polymerase (BRL) with different concentrations of MgCl2 for each specific gene, as shown in the Supplemental Table. Various quantities (1, 2, 4, 8, 16, and 32 L) of the equalized cDNA pools at two dilutions (10 and 100) were used in the PCR analyses. From 35 to 40 cycles of PCR amplification were carried out, with 1 minute for denaturation at 94C, 1.5 minute for annealing at optimal temperatures for the various genes that were analyzed, and 2 minutes for polymerization at 72C. Ten microliters of PCR products was separated on 1.5% agarose gel and visualized by ethidium bromide staining. The relative expression levels in cDNA from ES cells and cultured BM cells were estimated visually, according to the relative band intensities compared to those of cells directly isolated from fresh BM. The expression of genes in mouse CD34Lin cells was analyzed similarly. To confirm the fidelity of reverse transcriptase (RT)-PCR, flt3, -globin, and -globin PCR products were confirmed by DNA sequence analysis.

Results Hematopoietic receptors We examined the expression of a number of cytokine and chemokine receptor genes that have been shown to be important in normal hematopoiesis. Results are summarized in Table 1. IL3 receptor A was shown to be expressed at very low levels in rhesus monkey CD34 cells of all cell origins, with no evidence of its expression in ES cellderived CD34CD38 cells. No differences were found for IL-6 receptor expression in any of the CD34 cell preparations, but a dramatically reduced level was seen in CD34 CD38 cells derived from ES cells and from BM following cytokine stimulation (Fig. 2). G-CSF receptor and fms (macrophage-CSF receptor) were expressed at relatively high levels in both CD34 and CD34CD38 progenitors, but no expression of fms was detected in CD34CD38 cells isolated from cultured BM. In studies that examined mpl, the receptor of the cytokine thrombopoietin (TPO), no significant difference in expression was observed between progenitor cells

isolated from ES cells and from fresh BM. However, the expression of mpl was reduced to an undetectable level in both CD34 and CD34CD38 progenitor cells isolated from BM after cytokine stimulation. Epo-receptor expression could not be detected in any of the purified progenitor cells, although a clear positive signal was obtained from the whole BM control. Expression of the structurally related flt3 and c-kit genes from rhesus monkey cells demonstrated quite different patterns. The flt3 was expressed at relatively high levels in CD34 cells from fresh BM and cultured BM, but showed no detectable message in ES cell-derived CD34 cells. In CD34CD38 cells, expression of flt3 remained at a substantial level in cells from fresh BM, but there was no detectable expression in cells derived from ES cells or from cultured BM (Fig. 2). Concordant with these observations, we observed that CD34 cells derived from rhesus monkey ES cell cultures showed very similar cobblestone areaforming activity with or without added flt3 ligand in secondary plating experiments (data not shown). On the other hand, a very low level of flt3 expression was present in undifferentiated ES cells and in CD38 cells of ES cell origin (data not shown). In contrast, the c-kit gene was expressed at a substantial level in both CD34 and CD34CD38 cells of all origins. In CD34CD38 cells derived from cultured BM, c-kit expression was reduced to approximately 25% compared to that of fresh BM. The chemokine receptor CXCR4 was expressed at very high levels in both CD34 and CD34CD38 cells from rhesus monkey ES cells, fresh BM, and cultured BM. However, its expression was about five-fold less in ES cellderived CD34 and CD34CD38 cells compared to BM, with or without cytokine stimulation (Fig. 2). Expression of CXCR4 also could be demonstrated in undifferentiated ES cells (data not shown). No difference was apparent in the expression of P-selectin glycoprotein ligand-1 (PSGL-1), a cell surface mucin-like glycoprotein that serves as ligand for P-, L-, and E-selectins, among CD34 or CD34CD38 cells from ES cells, fresh BM, and cultured BM.

S.-J. Lu et al./Experimental Hematology 30 (2002) 58–66 Table 1. Gene expression in CD34 and CD34CD38 cells derived from rhesus monkey embryonic stem cells and bone marrow with and without added cytokines CD34

Receptors CXCR4 KDR/flk-1 c-kit flt3 fms mpl IL-3 receptor A IL-6 receptor Epo receptor G-CSF receptor Transcription factors scl PU.1 GATA-2 GATA-1 P45 NF-E2 c-myb Integrins 1 2 7 1 2 3 4 5 6 L M Lineage-specific genes CD34 MPO LZ CD14 -Globin -Globin -Globin RAG-1 pax5

CD34CD38

ES

BM

BMC

ES

BM

BMC

↓ ↑    ↑    

         

↓         ↓

↓ ↑   ↑   ↓  ↑

         

↓  ↓    ↓ ↓  

 ↓ ↓   

     

 ↓ ↓ ↓  

↑    ↑ 

     

     

↑   ↑ ↑    ↑  

          

↑    ↓   ↓ ↓  

↑       ↑ ↑  

          

          

   ↓ ↓ ↓ ↑  

        

↓   ↓ ↓ ↓ ↑ ↓ 

     ↓ ↑  

        

↓     ↓   

↑ increased expression compared to BM; ↓ decreased expression compared to BM;  undetectable;  similar expression level compared to BM; BM  C BM with added cytokines.

KDR/flk-1 is a receptor kinase for VEGF, which is expressed on endothelial cells and on the earliest identified hematopoietic and endothelial progenitor cells [18–20]. As shown in Figure 2, KDR/flk-1 was expressed at very high levels in both CD34 and CD34CD38 cells of ES cell origin, compared to the corresponding cell populations isolated from fresh rhesus monkey adult BM. In cell samples prepared from fresh BM, expression of KDR/flk-1 in CD34CD38 cells was much higher than in CD34 cells from the same BM (Fig. 2 and data not shown), suggesting that purging of CD38 cells substantially enriched the

61

KDR/flk-1 cell population. This observation is in accordance with findings from human BM, which showed that 0.1 to 0.2% of CD34 cells were KDR/flk-1, whereas 1 to 2% of CD34CD38 cells showed positive staining for the KDR/flk-1 receptor [21]. Unexpectedly, we observed that culture (expansion) of BM in ES cell culture medium for 3 days resulted in total loss of KDR/flk-1 expression, both in CD34 and CD34CD38 cells (Fig. 2 and Table 1). To determine if the KDR/flk-1 message might have come from mature endothelial cells present in the samples of fresh BM, we examined the expression of two endothelial markers, vWF and VE-CA, in both the ES cell- and BM-derived hematopoietic precursors. As shown in Figure 3, both vWF and VE-CA were expressed in CD34 cells from all three preparations, albeit at different levels, but no expression of either marker could be detected from cells with the CD34CD38 phenotype. These findings suggest that the detected signals of KDR/flk-1, at least in the CD34CD38 cells, likely did not result from endothelial cell contamination. Hematopoietic transcription factors Expression of various transcription factors that have been shown to play an important role in normal hematopoiesis was examined in rhesus monkey progenitor cells from the different sources, and the results are summarized in Table 1. PU.1 is a member of the ets family, and disruption of its function results in abnormal myeloid and lymphoid development and impaired stem cell homing and engraftment in mice [22,23]. We observed only a moderate level of PU.1 expression in CD34 cells isolated from fresh rhesus monkey BM, with very low (5%) levels in the same subset of cells from both ES cell progeny and cultured BM. Furthermore, no detectable PU.1 message could be detected in CD34CD38 cells from all sources, although the PU.1specific PCR product was clearly present in unfractionated BM samples analyzed under the same conditions. The transcription factor scl initially was identified as an oncogene associated with some forms of leukemia. Later it was shown to be an important regulator of the early development of endothelial and hematopoietic lineages [24,25]. We observed no difference in its expression among CD34 cells of all three sources. However, we observed an approximate eight-fold increase in scl expression in ES cellderived CD34CD38 cells, whereas no detectable mRNA was seen in the counterpart derived from cultured BM (Table 1). GATA-2 has been shown to be important in early hematopoiesis [26]. We observed a high level of expression of GATA-2 in CD34 cells from fresh BM, but relatively low expression in CD34 cells from ES cells and cultured BM. Interestingly, there was no detectable expression of GATA-2 in CD34CD38 cells from all sources, suggesting that GATA-2 may not function at very early stages of hematopoiesis, as previously suggested [26]. In contrast, GATA-1, a critical regulator of erythroid lineage differentiation [27], showed little or no variation in its mRNA level among the

62

S.-J. Lu et al./Experimental Hematology 30 (2002 ) 58–66

Figure 2. Representative analyses of gene expression in CD34 and CD34CD38 cells derived from rhesus monkey ES cell differentiation (ES), fresh BM (BM), and BM cultured with cytokines for 3 days (BMC). Cytoplasmic RNA from CD34 and CD34CD38 cells was used to construct cDNA pools, and the expression of genes was examined by semiquantitative PCR. The number at the top of each lane indicates the amount (in microliters) of cDNA used in the 50-L PCR reaction. M 1 kb DNA marker (BRL); WBM unfractionated rhesus monkey adult BM.

isolated CD34 cells. However, in the population of CD34 CD38 cells, GATA-1 expression was only observed in ES cell-derived progenitors, and none could be detected in cells from fresh BM or cultured BM. A high level of expression of both the p45NF-E2 and myb genes was observed in both

Figure 3. Analyses of vWF and VE-CA expression in CD34 and CD34CD38 cells derived from rhesus monkey ES cell differentiation (ES), fresh BM (BM), and BM cultured with cytokines for 3 days (BMC). Cytoplasmic RNA from CD34 and CD34CD38 cells was used to construct cDNA for the analyses of vWF and VE-CA expression by semiquantitative PCR. PL-RT human placental total RNA (Clontech).

CD34 and CD34CD38 cells from all three sources, and no significant difference in expression was found among them (Table 1). Integrins Integrins are adhesive molecules that are expressed in most types of leukocytes and in HSC [28,29]. Some subtypes of integrins form heterodimers, which play a vital role in homing and engraftment of HSC to BM [8]. In particular, the VLA-4 (41) and VLA-5 (51) antigens have been shown to be critical for HSC lodging in the BM [8,30,31]. We therefore examined integrin gene expression in CD34 and CD34CD38 cells from ES cells and BM. RT-PCR analyses showed that most integrins were expressed in rhesus monkey ES-derived and BM (fresh and cultured) CD34 cells (Table 1) with few exceptions. Integrin 2 was only expressed in ES cell-derived CD34 cells, whereas the 7 message was detected exclusively in CD34 cells from fresh BM. In comparison to CD34 cells from BM, integrins 1, 1, 2, and 6 showed substantially elevated expression levels in CD34 cells derived from ES cells. The relatively elevated expression of these genes presumably is not attributable to cytokine stimulation, because the same integrins did not exhibit significantly increased levels of expression in CD34 cells from cytokinestimulated BM compared with fresh BM. No notable differences were found in the expression of integrins 4 and 5 in CD34 and CD34CD38 cells from the three different sources. On the other hand, expression of integrins 2, 7, 1, 2, 3, and M was undetectable in CD34CD38 cells of all origins, except that very low levels of 1 message were detected in the corresponding populations from ES cells (Table 1). In the case of integrin L, this gene was expressed at a rela-

S.-J. Lu et al./Experimental Hematology 30 (2002) 58–66

tively high level in CD34 cells of all origins; however, L was not detected in CD34CD38 cells derived from ES cells, whereas a low but clearly detectable level of expression was observed in both cytokine-stimulated and nonstimulated cell populations from BM (Fig. 2). Lineage-specific genes All of the progenitor cells examined in this study were isolated using the CD34 antigen as a positive marker. We initially analyzed the expression of the CD34 gene in both CD34 and CD34CD38 cells (Table 1). A moderate level of CD34 mRNA was found in all progenitor cells, with a moderate decrease observed in both CD34 and CD34CD38 cells isolated from cultured BM. No expression of CD34 was found in undifferentiated ES cells [7]. Very high mRNA levels of the MPO and LZ genes, both markers of myeloid cells, were observed in both CD34 and CD34CD38 cells from ES, fresh BM, and cultured BM. For erythoid-specific gene expression, -globin and -globin mRNAs were assessed in both CD34 and CD34CD38 cells. Substantial levels of both genes were expressed in cell preparations from all three sources. However, a much higher level of -globin message was detected in CD34CD38 cells derived from ES cells compared to corresponding cells from BM. On the other hand, a relatively low level of -globin gene expression was found in CD34 cells of all origins, with the highest level observed in cells of fresh BM origin. No detectable -globin gene expression was observed in CD34CD38 cells derived from ES cells, fresh BM, or cultured BM. Expression of CD14, a monocyte surface marker, could be detected in CD34 cells from ES cells, as well as from fresh BM and cultured BM, but negligible levels of its mRNA were observed in CD34CD38 cells from any origin. We observed high mRNA levels of the lymphoidspecific gene RAG-1 in fresh BM-derived CD34 cells, but its expression was decreased in CD34 cells from BM cultured in vitro, and no expression of this gene could be detected in ES cell-derived CD34 cells. Moreover, the RAG-1 message was only detectable in CD34CD38 cells derived from fresh BM, and it was absent from cells derived from ES cell differentiation or cultured BM. We also examined the expression of the pax5 gene, which normally is expressed only in pre-B cells. Correspondingly, there was no detectable level of pax5 mRNA in any of the progenitor cells, except for CD34 cells isolated directly from fresh BM, in which a moderate level of pax5 message was observed (Table 1). These results are consistent with previous observations that expression of multilineage genes precedes the onset of specific lineage differentiation in the hematopoietic system [14,32,33]. Gene expression in mouse CD34Lin cells Parallel studies to those of rhesus monkey HSC gene expression were performed to compare their expression in murine hematopoietic precursors. For these studies, we purified hematopoietic CD34Lin cells from strain 129/sv mouse BM and from differentiation cultures of mouse D3 ES cells.

63

These experiments showed no apparent expression of the flt3, IL-6 receptor, or integrin L genes in CD34Lin cells derived from the ES cell cultures, whereas substantial levels of mRNA messages were observed in corresponding cell populations isolated from BM (Table 2 and Fig. 4), thus substantiating our findings from the rhesus monkey ES cells. On the other hand, we did not observe a decrease in CXCR4 gene expression in CD34Lin cells derived from murine ES cell cultures. With the mouse system, there was an increased level of expression of this gene compared with CD34Lin cells purified from BM (Fig. 4). Expression of the integrin 1 and c-kit genes was examined, and again no difference could be detected for the c-kit gene between CD34Lin cells from murine ES cells and those harvested from BM. A somewhat increased level of integin 1 expression was observed in the corresponding cells from murine ES cell differentiation. Discussion We previously demonstrated that rhesus monkey ES cells can differentiate in vitro to form hematopoietic progenitors that morphologically resemble undifferentiated blast cells, and that are replatable under stroma-dependent culture conditions [7]. The present study was directed toward further characterizing these cells at the molecular level by examining the expression of genes associated with hematopoietic cell development. Our findings demonstrate overall a remarkable degree of similarity in the expression patterns of these genes in both CD34 and CD34CD38 cells from all sources, which is consistent with recently reported studies in which gene expression profiles were compared between fresh and cytokine-stimulated hematopoietic progenitor cells [34].

Figure 4. Representative analyses of gene expression in CD34Lin cells derived from mouse D3 ES cell differentiation (ES) and BM. Cytoplasmic RNA from CD34Lincells was used to construct cDNA pools, and the expression of genes was examined by semiquantitative PCR. The number at the top of each lane indicates the amount (in microliters) of cDNA pool used in the 50-L PCR reaction. M 1 kb DNA marker (BRL); WBM

unfractionated strain 129/sv mouse BM.

64

S.-J. Lu et al./Experimental Hematology 30 (2002 ) 58–66

Table 2. Gene expression in CD34Lin cells derived from mouse embryonic stem cells and bone marrow Gene

ES

BM

flt3 c-kit CXCR4 IL-6 receptor Integrin L Integrin 1

  ↑   ↑

     

↑ increased expressin compared to BM;  undetectable;  similar expression level compared to BM.

We observed, however, that the flt3 gene, which was well expressed in CD34 and CD34CD38 cells from rhesus monkey BM, was undetectable in the corresponding populations derived from rhesus monkey ES cells. Moreover, in parallel with these observations, flt3 displayed robust expression in CD34Lin cells from murine BM, but had no apparent expression in CD34Lin cells from murine ES cells. The expression of both the integrin L and IL-6 receptor genes also was reduced to very low levels in CD34CD38 cells derived from rhesus monkey ES cell cultures and was absent in CD34Lin cells from mouse ES cells. The flt3 gene encodes a tyrosine kinase receptor closely related to the c-kit and fms gene products [35,36]. Targeted disruption of the flt3 gene and that of its ligand FL in mice resulted in decreased numbers of myeloid and lymphoid progenitor cells [37,38]. Transplantation experiments also demonstrated impaired reconstitution of the hematopoietic system of irradiated recipients using HSCs from flt3/ mice [37]. Gene expression analyses revealed the presence of flt3 in multipotential myeloid and lymphoid precursors, with decreased expression associated with differentiation to erythroid and mast cell lineages [35,36,39,40]. The expression status of flt3 in hematopoietic cell progenitors remains controversial. In the mouse, flt3 mRNA was detected in early AA4LinSca1 cells, but no significant difference was found in the reconstitution potential between flt3 and flt3 subtypes of AA4Sca, AA4CD34, or Lin Sca populations [41]. Progenitor cells with low levels of flt3 expression had a greater percentage of their population in the G0 stage (27.6%) compared with those having high levels of flt3 (3.3%) [41]. About 45% of cells with high levels of flt3 were found to be at the G1 stage of the cell cycle, and only 5% of cells with low levels of flt3 were in G1 [41]. Moreover, Billia et al. [42] recently showed that the flt3 transcript was only detectable in purified reconstituting cells from mouse BM, but not in later cells. Similarly, more than 40% of CD34flt3low cells from human BM were found to be in G0, whereas only 1.5% of the CD34flt3high population were in the G0 phase. Moreover, more than 75% of CD34flt3high cells were in the G1 phase,

but only 36.8% of CD34flt3low cells were in G1 [43]. Glimm et al. [44] recently demonstrated that transplantable stem cells from cultured cord blood were restricted to the G1 stage cell fraction. Taken together, these findings suggest that the CD34flt3 cell fraction described earlier may represent the most primitive hematopoietic progenitors, with repopulating potential. However, whether the CD34 antigen is a reliable HSC marker in the mouse remains debatable [45–48]. Götze et al. [43] and Rappold et al. [40] also showed that almost equal percentages of CD34CD38 and CD34 CD38 cells from BM expressed flt3, and that the highest expression of flt3 was localized to the CD34CD38 population. Hematopoietic colony-forming cell assays demonstrated that CD34flt3 cells formed mainly granulomonocytic colonies, whereas CD34flt3 cells contained mainly erythroid colonies. Xiao et al. [49] reported that greater than 90% of freshly isolated CD34CD38 cells from mobilized peripheral blood expressed flt3, whereas only few CD34 CD38 cells presented flt3. Furthermore, CD34CD38 precursors lost flt3 expression as the cells matured into CD34CD38 progenitors in culture, which is clearly at variance with the findings of Götze et al. [43] and Rappold et al. [40]. Our analyses of flt3 expression revealed no detectable message of flt3 in both rhesus monkey CD34 and CD34 CD38 cells, as well as in murine CD34Lin cells, derived from ES cells, although a low level of flt3 message could be detected in CD38 populations from rhesus monkey ES cells (data not shown). In contrast, we detected a high level of flt3 message in CD34 cells from both fresh BM and cultured BM. However, flt3 transcript was only detectable in CD34CD38 cells derived from fresh but not cultured BM or ES cells. Our results clearly demonstrate that only flt3 progenitors are generated from both mouse and rhesus monkey ES cells that have undergone differentiation in vitro. IL-6 receptor exists as a heterodimer: its IL-6 binding subunit (IL-6R) is a 80-kDa protein; its other subunit (gp130) is a component of various cytokine receptors [50]. IL-6R is expressed in a number of cell types, including hematopoietic lineages [50]. Its ligand IL-6 is a multifunctional cytokine involved in a number of physiologic responses [50]. Previous studies demonstrated that IL-6 also acts on early hematopoietic progenitors in combination with other cytokines, and it is believed to induce stem cells to enter the cell cycle [51,52]. The IL-6R gene is constitutively expressed in ES cells, and its level of expression is increased during in vitro differentiation in both murine and rhesus monkey ES cells [7,13]. Our observations indicate no difference in IL-6R expression in CD34 cells derived from rhesus monkey ES cells, fresh BM, or cultured BM. However, the expression of IL-6R was substantially reduced, to about 5%, in the more primitive hematopoietic CD34CD38 cells derived from rhesus monkey ES cell

S.-J. Lu et al./Experimental Hematology 30 (2002) 58–66

differentiation compared to the same cell population from fresh BM. Similarly, no IL-6R message was detected in CD34Lin cells isolated from murine ES cell differentiation. These observations could have resulted from cytokine stimulation, inasmuch as similarly decreased expression levels were observed in CD34CD38 cells derived from cultured rhesus monkey BM (Fig. 2). Integrin L, which when associated with integrin 2 constitutes the leukocyte-function associated (LFA-1) antigen, is expressed on most myeloid and lymphoid cell types at various stages of maturity and on a subset of more immature cells, but disappears from cells differentiating to the erythroid lineage [53–55]. LFA-1 mediates cell–cell interactions by binding to cell surface ligands, the intercellular adhesion molecules-1, -2 and -3, and plays a crucial role in regulation of inflammatory and immune responses mediated by mature granulocytes and lymphocytes [29]. Long-term BM culture studies have demonstrated that the CD34LFA-1 cells derived from human BM generated more colony-forming cells than CD34 LFA cells, suggesting that LFA-1 cells are more primitive than LFA-1 cells [54]. Pruijt et al. [56] showed that the LFA-1 cells isolated from murine BM and from cytokine-mobilized blood possess more colony-forming cells than do the LFA-1 fractions. Furthermore, the LFA-1 fractions were found to contain the majority of cells with radioprotective capacity in BM transplantation experiments [56], which is in agreement with the findings in man [54]. In this study, we demonstrated that integrin 2 is expressed at a substantial level in CD34 cells isolated from ES cell differentiation. No message was observed in CD34 cells from fresh BM or cultured BM, or in CD34CD38 cells from any cell origin. The other subunit of LFA-1, integrin L, was expressed at relatively high levels in all CD34 cells. However, the expression of L was undetectable in the more primitive hematopoietic precursors CD34CD38 and CD34Lin cells derived from ES cells (Figs. 2 and 4). These findings suggest that relatively more primitive hematopoietic progenitor cells may have been generated from the ES cell cultures in vitro. The findings from this study suggest that the expression of genes in hematopoietic progenitor cells purified from BM and from ES cell differentiation are remarkably comparable, with the apparent exception of flt3, IL-6 receptor, and integrin L. Either of two broad hypotheses might account for the observed differences. The CD34 and/or CD34CD38 cells of rhesus monkey and CD34Lin cells of mouse purified from BM and ES cell differentiation could contain different subpopulations of hematopoietic precursors, possibly at varying stages of differentiation. Simultaneous characterization for the content of functionally distinct precursors and analysis of gene expression in the purified cell populations [42] might permit verification of this hypothesis. Alternatively, ES cell differentiation under culture conditions might result in altered regulation of certain genes in hematopoietic progenitor/stem cells. These changes, regard-

65

less of the underlying mechanisms for their development, could provide important clues regarding the altered behavior of progenitor cells from murine ES cell cultures, which lack long-term reconstitution potential when transplanted into adult recipient mice [57,58].

Acknowledgments We thank Dr. James A. Thomson (Wisconsin Regional Primate Research Center, University of Wisconsin) for providing the rhesus monkey ES cells for this study; the Wisconsin Regional Primate Center and the Biologic Resources Laboratory of the University of Illinois at Chicago for providing rhesus monkey BM; Dr. Jianxun Li and Ximing Zhou (College of Dentistry, University of Illinois at Chicago) for assistance with graphic photo imaging; and Dr. Karen Hagen (Research Resources Center, University of Illinois at Chicago) for flow cytometric analyses.

References 1. Evans MJ, Kaufman MH (1981) Establishment in culture of pluripotential cells from mouse embryos. Nature 292:154 2. Thomson JA, Kalishman J, Golos TG, et al. (1995) Isolation of a primate embryonic stem cell line. Proc Natl Acad Sci U S A 92:7844 3. Thomson JA, Itskovitz-Eldor J, Shapiro SS, et al. (1998) Embryonic stem cell lines derived from human blastocysts. Science 282:1145 4. Reubinoff BE, Pera MF, Fong CY, Trounson A, Bongso A (2000) Embryonic stem cell lines from human blastocysts: somatic differentiation in vitro. Nat Biotechnol 18:399 5. Keller GM (1995) In vitro differentiation of embryonic stem cells. Curr Opin Cell Biol 7:862 6. Weiss MJ, Orkin SH (1996) In vitro differentiation of murine embryonic stem cells. New approaches to old problems. J Clin Invest 97:591 7. Li F, Lu S-J, Vida L, Thomson JA, Honig GR (2001) Bone morphogenetic protein-4 induces efficient hematopoietic differentiation of rhesus monkey embryonic stem cells in vitro. Blood 98:335 8. Quesenberry PJ, Becker PS (1998) Stem cell homing: rolling, crawling, and nesting. Proc Natl Acad Sci U S A 95:15155 9. Whetton AD, Graham GJ (1999) Homing and mobilization in the stem cell niche. Trends Cell Biol 9:233 10. Keller G, Lacaud G, Robertson S (1999) Development of the hematopoietic system in the mouse. Exp Hematol 27:777 11. Keller G, Kennedy M, Papayannopoulou T, Wiles MV (1993) Hematopoietic commitment during embryonic stem cell differentiation in culture. Mol Cell Biol 13:473 12. Schmitt RM, Bruyns E, Snodgrass HR (1991) Hematopoietic development of embryonic stem cells in vitro: cytokine and receptor gene expression. Genes Dev 5:728 13. McClanahan T, Dalrymple S, Barkett M, Lee F (1993) Hematopoietic growth factor receptor genes as markers of lineage commitment during in vitro development of hematopoietic cells. Blood 81:2903 14. Hu M, Krause D, Greaves M, et al. (1997) Multilineage gene expression precedes commitment in the hemopoietic system. Genes Dev 11:774 15. Zhumabayeva B, Diatchenko L, Chenchik A, Siebert PD (2001) Use of SMART-generated cDNA for gene expression studies in multiple human tumors. Biotechniques 30:158 16. Zheng J, Shen W, He DZ, Long KB, Madison LD, Dallos P (2000) Prestin is the motor protein of cochlear outer hair cells. Nature 405:149 17. Gomes I, Sharma TT, Mahmud N, et al. (2001) Highly abundant genes in the transcriptosome of human and baboon CD34 antigen-positive bone marrow cells. Blood 98:93 18. Kabrun N, Buhring HJ, Choi K, Ullrich A, Risau W, Keller G (1997)

66

19.

20.

21. 22. 23.

24.

25.

26. 27.

28.

29. 30.

31.

32. 33.

34.

35. 36.

37.

38.

S.-J. Lu et al./Experimental Hematology 30 (2002 ) 58–66 Flk-1 expression defines a population of early embryonic hematopoietic precursors. Development 124:2039 Yamaguchi TP, Dumont DJ, Conlon RA, Breitman ML, Rossant J (1993) flk-1, an flt-related receptor tyrosine kinase is an early marker for endothelial cell precursors. Development 118:489 Yamashita J, Itoh H, Hirashima M, et al. (2000) Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors. Nature 408:92 Ziegler BL, Valtieri M, Porada GA, et al. (1999) KDR receptor: a key marker defining hematopoietic stem cells. Science 285:1553 Fisher RC, Scott EW (1998) Role of PU.1 in hematopoiesis. Stem Cells 16:25 Fisher RC, Lovelock JD, Scott EW (1999) A critical role for PU.1 in homing and long-term engraftment by hematopoietic stem cells in the bone marrow. Blood 94:1283 Porcher C, Swat W, Rockwell K, Fujiwara Y, Alt FW, Orkin SH (1996) The T cell leukemia oncoprotein SCL/tal-1 is essential for development of all hematopoietic lineages. Cell 86:47 Kallianpur AR, Jordan JE, Brandt SJ (1994) The SCL/TAL-1 gene is expressed in progenitors of both the hematopoietic and vascular systems during embryogenesis. Blood 83:1200 Tsai FY, Keller G, Kuo FC, et al. (1994) An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371:221 Pevny L, Simon MC, Robertson E, et al. (1991) Erythroid differentiation in chimaeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1. Nature 349:257 Coulombel L, Auffray I, Gaugler MH, Rosemblatt M (1997) Expression and function of integrins on hematopoietic progenitor cells. Acta Haematol 97:13 Hynes RO (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69:11 Frenette PS, Subbarao S, Mazo IB, von Andrian UH, Wagner DD (1998) Endothelial selectins and vascular cell adhesion molecule-1 promote hematopoietic progenitor homing to bone marrow. Proc Natl Acad Sci U S A 95:14423 Zanjani ED, Flake AW, Almeida-Porada G, Tran N, Papayannopoulou T (1999) Homing of human cells in the fetal sheep model: modulation by antibodies activating or inhibiting very late activation antigen-4– dependent function. Blood 94:2515 Strobl H, Takimoto M, Majdic O, et al. (1993) Myeloperoxidase expression in CD34 normal human hematopoietic cells. Blood 82:2069 Mao M, Fu G, Wu JS, et al. (1998) Identification of genes expressed in human CD34() hematopoietic stem/progenitor cells by expressed sequence tags and efficient full-length cDNA cloning. Proc Natl Acad Sci U S A 95:8175 Oh IH, Lau A, Eaves CJ (2000) During ontogeny primitive (CD34() CD38()) hematopoietic cells show altered expression of a subset of genes associated with early cytokine and differentiation responses of their adult counterparts. Blood 96:4160 Rosnet O, Schiff C, Pebusque MJ, et al. (1993) Human FLT3/FLK2 gene: cDNA cloning and expression in hematopoietic cells. Blood 82:1110 Matthews W, Jordan CT, Wiegand GW, Pardoll D, Lemischka IR (1991) A receptor tyrosine kinase specific to hematopoietic stem and progenitor cell-enriched populations. Cell 65:1143 Mackarehtschian K, Hardin JD, Moore KA, Boast S, Goff SP, Lemischka IR (1995) Targeted disruption of the flk2/flt3 gene leads to deficiencies in primitive hematopoietic progenitors. Immunity 3:147 McKenna HJ, Stocking KL, Miller RE, et al. (2000) Mice lacking flt3 ligand have deficient hematopoiesis affecting hematopoietic progenitor cells, dendritic cells, and natural killer cells. Blood 95:3489

39. Pharr PN, Hofbauer A (1997) Loss of flk-2/flt3 expression during commitment of multipotent mouse hematopoietic progenitor cells to the mast cell lineage. Exp Hematol 25:620 40. Rappold I, Ziegler BL, Kohler I, et al. (1997) Functional and phenotypic characterization of cord blood and bone marrow subsets expressing FLT3 (CD135) receptor tyrosine kinase. Blood 90:111 41. Zeigler FC, Bennett BD, Jordan CT, et al. (1994) Cellular and molecular characterization of the role of the flk-2/flt-3 receptor tyrosine kinase in hematopoietic stem cells. Blood 84:2422 42. Billia F, Barbara M, McEwen J, Trevisan M, Iscove NN (2001) Resolution of pluripotential intermediates in murine hematopoietic differentiation by global complementary DNA amplification from single cells: confirmation of assignments by expression profiling of cytokine receptor transcripts. Blood 97:2257 43. Götze KS, Ramirez M, Tabor K, Small D, Matthews W, Civin CI (1998) Flt3high and Flt3low CD34 progenitor cells isolated from human bone marrow are functionally distinct. Blood 91:1947 44. Glimm H, Oh IH, Eaves CJ (2000) Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G(2)/M transit and do not reenter G(0). Blood 96:4185 45. Osawa M, Hanada K, Hamada H, Nakauchi H (1996) Long-term lymphohematopoietic reconstitution by a single CD34 low/negative hematopoietic stem cell. Science 273:242 46. Donnelly DS, Zelterman D, Sharkis S, Krause DS (1999) Functional activity of murine CD34 and CD34 hematopoietic stem cell populations. Exp Hematol 27:788 47. Tajima F, Sato T, Laver JH, Ogawa M (2000) CD34 expression by murine hematopoietic stem cells mobilized by granulocyte colonystimulating factor. Blood 96:1989 48. Krause DS, Theise ND, Collector MI, et al. (2001) Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 105:369 49. Xiao M, Oppenlander BK, Plunkett JM, Dooley DC (1999) Expression of Flt3 and c-kit during growth and maturation of human CD34 CD38 cells. Exp Hematol 27:916 50. Keller ET, Wanagat J, Ershler WB (1996) Molecular and cellular biology of interleukin-6 and its receptor. Front Biosci 1:d340 51. Leary AG, Ikebuchi K, Hirai Y, et al. (1988) Synergism between interleukin-6 and interleukin-3 in supporting proliferation of human hematopoietic stem cells: comparison with interleukin-1 alpha. Blood 71:1759 52. Ikebuchi K, Wong GG, Clark SC, Ihle JN, Hirai Y, Ogawa M (1987) Interleukin 6 enhancement of interleukin 3-dependent proliferation of multipotential hemopoietic progenitors. Proc Natl Acad Sci U S A 84:9035 53. Kansas GS, Muirhead MJ, Dailey MO (1990) Expression of the CD11/ CD18, leukocyte adhesion molecule 1, and CD44 adhesion molecules during normal myeloid and erythroid differentiation in humans. Blood 76:2483 54. Gunji Y, Nakamura M, Hagiwara T, et al. (1992) Expression and function of adhesion molecules on human hematopoietic stem cells: CD34 LFA-1 cells are more primitive than CD34LFA-1 cells. Blood 80:429 55. Papayannopoulou T, Brice M (1992) Integrin expression profiles during erythroid differentiation. Blood 79:1686 56. Pruijt JF, van Kooyk Y, Figdor CG, Willemze R, Fibbe WE (1999) Murine hematopoietic progenitor cells with colony-forming or radioprotective capacity lack expression of the beta 2-integrin LFA-1. Blood 93:107 57. Hole N, Graham GJ, Menzel U, Ansell JD (1996) A limited temporal window for the derivation of multilineage repopulating hematopoietic progenitors during embryonal stem cell differentiation in vitro. Blood 88:1266 58. Palacios R, Golunski E, Samaridis J (1995) In vitro generation of hematopoietic stem cells from an embryonic stem cell line. Proc Natl Acad Sci U S A 92:7530