| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Oregon National Primate Research Center,3 Oregon Health & Science University, Beaverton, Oregon 97006
Departments of Obstetrics and Gynecology4 and Physiology and Pharmacology,5 Oregon Health & Science University, Portland, Oregon 97201
ABSTRACT
Embryonic stem cells (ESCs) may be able to cure or alleviate the symptoms of various degenerative diseases. However, unresolved issues regarding survival, functionality, and tumor formation mean a prudent approach should be adopted towards advancing ESCs into human clinical trials. The rhesus monkey provides an ideal model organism for developing strategies to prevent immune rejection and test the feasibility, safety, and efficacy of ESC-based medical treatments. Transcriptional profiling of rhesus monkey ESCs provides a foundation for pre-clinical ESC research in this species. In the present study, we used microarray technology, immunocytochemistry, reverse transcription polymerase chain reaction (RT-PCR) and quantitative real-time PCR (qPCR) to characterize and transcriptionally profile rhesus monkey ESCs. We identified 367 stemness gene candidates that were highly (>85%) conserved across five different ESC lines. Rhesus monkey ESC lines maintained a pluripotent undifferentiated state over a wide range of POU5F1 (also known as OCT4) expression levels, and comparisons between rhesus monkey, mouse, and human stemness genes revealed five mammalian stemness genes: CCNB1, GDF3, LEFTB, POU5F1, and NANOG. These five mammalian genes are strongly expressed in rhesus monkey, mouse, and human ESCs, albeit only in the undifferentiated state, and represent the core key mammalian stemness factors.
embryo, gene regulation, microarray, rhesus, stem cells
Embryonic stem cells (ESCs) can proliferate indefinitely, maintain an undifferentiated pluripotent state, and differentiate into any cell type [1]. Differentiation of ESCs into specific cell types may provide resources to cure or alleviate the symptoms of various degenerative diseases [2]. However, currently, we do not know whether ESC-derived differentiated cells will: 1) survive for extended periods of time following transfer to a patient, 2) maintain or lose their function in patients, 3) undergo uncontrolled proliferation and become tumors. With the possibility of these cells becoming cancerous and doing more harm than good, a prudent approach towards advancing ESCs into human clinical trials is required. While the vast majority of animal research is conducted in rodents, we believe that the unanswered questions regarding ESC in vivo function, apoptosis, and tumor formation would be best addressed using a non-human primate. Rhesus macaques (the standard non-human primate model organism) possess remarkable anatomical, physiological, and metabolic similarities to humans, and many human neurological diseases, such as Alzheimer and Parkinson diseases, can only be accurately modeled in the non-human primate. In addition to the biological similarities between rhesus macaques and humans, significant numbers of primate ESC lines are now available [3], and ESCs derived from the rhesus monkey or human blastocysts demonstrate extensive similarities that are not observed with murine ESCs [4–6]. This leads to the conclusion that rhesus macaques provide an accurate model for developing strategies to prevent immune rejection and to test the feasibility, safety, and efficacy of ESC-based medical treatments.
Despite the importance of rhesus macaques as a model for ESC-based cell replacement therapy, relatively little is known about the genetic programming of ESCs in this species. The microarray is a new and powerful technology that is capable of analyzing simultaneously the expression of thousands of genes. Many groups have used microarrays and related technologies to carry out transcriptional profiling of human and mouse ESC lines [5, 7–13]. However, to date, transcriptional profiling has not been described for any non-human primate ESC line. Performing transcriptional profiling of rhesus monkey ESC lines would provide important information on the rhesus monkey stemness genes needed to maintain rhesus monkey ESCs in a pluripotent state, demonstrate the similarities and differences between human and primate ESCs, and provide a foundation for future pre-clinical ESC research using non-human primates as the model organism. In the present study, we use microarray analysis immunocytochemistry, reverse transcription polymerase chain reaction (RT-PCR) and quantitative real-time PCR (qPCR), to characterize and transcriptionally profile rhesus monkey ESCs. We identify rhesus monkey stemness gene candidates and demonstrate their high levels of conservation (>85%) across five different rhesus monkey ESC lines. We also demonstrate that rhesus monkey ESC lines maintain a pluripotent undifferentiated state over a wide range of POU5F1 (also known as OCT4) expression levels, and by comparing rhesus monkey, human, and murine data, we identify key mammalian stemness genes.
The ORMES lines and somatic fibroblasts used in this study were produced in our laboratory [3], and the H1 line was provided courtesy of Dr. J. Thomson. The ORMES (Oregon Rhesus Macaque Embryonic Stem) cells were grown at 37°C in 5% CO2-balanced air atmosphere upon mitotically inactivated mouse embryonic fibroblast (MEF) feeder cells in Dulbecco modified Eagle medium (DMEM/F12) (Invitrogen, Grand Island, NY) that was supplemented with 15% fetal bovine serum (FBS) (Hyclone, Logan, UT), 0.1 mM ß-mercaptoethanol (Sigma, St. Louis, MO), 1% nonessential amino acids (Invitrogen), 2 mM L-glutamine (Invitrogen), 4 ng/ml FGF2 (Sigma), with the pH adjusted to 7.2 with NaOH. The human ESC line H1 was cultured under the same conditions as the ORMES lines, except that 20% KNOCKOUT serum replacement (Invitrogen) was used instead of 15% FBS in the culture medium. ESC colonies were passaged manually every 4 to 5 days. For embryoid body (EB) formation, entire ESC colonies were loosely detached from the feeder cells and transferred manually into feeder-free, 6-well, Ultra Low Adhesion plates (Costar; Corning Incorporated, Corning, NY) and cultured in suspension in ESC medium for 5 to 7 days. To induce spontaneous differentiation, the EBs were transferred into gelatin-coated, 60-mm culture dishes (Becton Dickinson, Bedford, MA) to allow attachment of the cells, and cultured for 1 mo in FGF2-free ESC culture media. Rhesus monkey somatic cells were cultured in DMEM/F12 that was supplemented with 10% FBS. In harvesting the ESCs for analysis, only flat colonies with a large nuclear to cytoplasmic ratio and a cobbled, tightly packed morphology were selected (Fig. 1A). The selected colonies were removed manually from the dish without using enzymes, to minimize MEF contamination. In harvesting differentiated EBs for analysis, RNA lysis buffer (Invitrogen) was added directly to the dish that contained the differentiated EBs, and the RNA samples from all the morphologically different EBs were pooled for analysis.
|
The MIAME (Minimum Information About a Microarray Experiment) guidelines for microarray research [14] were incorporated into the design and implementation of these studies. All the microarray information and individual cell intensity (CEL) files are available online at the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE4446). All the microarray comparison analysis data cited in the text have been deposited as Supplementary Tables 1–9 and are available at ftp://ftp.ncbi.nih.gov/pub/geo/DATA/supplementary/series/GSE4446/. Total RNA was isolated from cell colonies selected for the appropriate ESC morphology (flat monolayer colony with distinctive cobbled stem cell morphology and a high nucleo-cytoplasmic ratio) using the Invitrogen RNA purification kit with optional DNase treatment (Invitrogen). The RNA samples were quantified using the NanoDrop ND-1000 UV-Vis spectrophotometer (NanoDrop Technologies, Wilmington, DE) and the quality of the RNA was assessed using Lab-on-a-Chip RNA Pico Chips and the 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Samples with electropherograms that showed a size distribution pattern predictive of acceptable microarray assay performance were considered to be of good quality. Twenty nanograms of total RNA from each sample were amplified and labeled using a two-cycle cDNA synthesis and an in vitro transcription cRNA-RNA labeling system (GeneChip Two-Cycle Target Labeling and Control Reagents; Affymetrix, Inc., Santa Clara, CA). Following successful cRNA amplification, 10 µg of labeled target cRNA was hybridized to Rhesus Macaque Genome Arrays (Affymetrix) using standard protocols, as described in the GeneChip Expression Analysis manual (http://www.affymetrix.com/support/technical/manual/expression_manual.affx). The rhesus monkey array contains 52 865 probe sets, representing over 20 000 genes. The arrays were scanned using the GeneChip laser scanner (Affymetrix) and image processing, normalization, and expression analysis were performed with the Affymetrix GCOS ver. 1.2 software. MAS-5 statistical analysis was performed to calculate the signal log ratio (SLR) for each probe set comparison, and the gene expression fold-changes (FCs) between two samples were calculated from the SLR using the following formula: FC = (2SLR). The GCOS 1.2 MAS 5.0 software was used to calculate statistically significant upregulation/downregulation (P < 0.002) between the undifferentiated and differentiated rhesus monkey ESC samples. The following selection criteria were used to identify rhesus monkey putative stemness genes: 1) genes that were considered to be present (P < 0.05) in all three ORMES-6 biological replicates; 2) genes that demonstrated statistically significant upregulation in all three biological replicates of ORMES-6 following GCOS comparisons with MAS-5 statistical analysis, and 3) genes that demonstrated on average a 3-fold or higher level of expression in undifferentiated ESCs compared to the pooled differentiated EBs. Since some microarray probe sets can represent the same gene, we removed any duplicate probe sets. All of the normalized microarray data sets generated from these studies can be found in the Supplemental information on the GEO website, as noted above. Supplemental Table 1 contains the comparison analysis for all 52 865 probe sets, while Supplemental Table 2 contains the rhesus monkey genes that were significantly upregulated (FC>3) in the ORMES-6 biological replicates, and Supplemental Table 3 contains the rhesus monkey genes that were significantly upregulated (FC>3) in the pooled differentiated EBs, and Supplemental Tables 4–8 represent genes that were significantly upregulated in ORMES 6A, 7, 9, 10 and 13, respectively.
|
|
|
|
Total RNA was extracted from ESCs, pooled differentiated EBs, and somatic cells using the PureLink Micro-to-Midi Total RNA purification kit (Invitrogen). Total RNA was treated with DNase I, to remove genomic contamination, before cDNA preparation using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen) according to the manufacturers instructions. The first-strand cDNA was further amplified by PCR using individual primer pairs for specific genes. The RT-PCR primers sequences, annealing temperatures, and PCR product sizes for POU5F1, SOX2, GAPDH and NANOG are listed in Supplemental Table 9. The NANOG and SOX2 primer sequences were based on a consensus of the human sequences acquired from GenBank. The POUF51 and GAPDH primer sequences were based on rhesus macaque sequences determined in our laboratory (GAPDH, GenBank AY624140; POU5F1, unpublished sequence). The PCR conditions were as follows (denaturation/annealing/extension): for NANOG and GAPDH, 40 cycles of 94/60/68°C for 30/60/60 s; for POUF51 and SOX2, 30 cycles of 94/60/68°C for 30/60/60 sec. Amplicons were electrophoresed through 1.6% 0.5x TAE agarose gels, stained with ethidium bromide, and visualized on a UV transilluminator.
Quantitative real-time PCR (qPCR) was performed on total RNA samples from each rhesus monkey ESC line, i.e., ORMES 6A, 6B, 6C, 7, 9, 10, and 13, and the pooled differentiated EBs. RNA from each ESC line was treated with RNase-free DNase (Invitrogen) to remove genomic DNA. The RNA concentrations were determined using an Agilent Labchip Bioanalyzer (Agilent Technologies). The cDNAs were synthesized from 2 µg of each total RNA sample with SuperScript III reverse transcriptase (200 U/µl) (Invitrogen) using oligo(dT) primers. TaqMan probes were designed using ABI Primer Express ver. 2.0.0 and synthesized by Applied Biosystems (Foster City, CA). FAM and VIC are reporter dyes, and MGBNFQ is the Molecular-Groove Binding Non-fluorescence Quencher (see Supplemental Table 9). The real-time PCR primers were designed using the IDT website (http://scitools.idtdna.com/Primerquest/) and synthesized by IDT (Integrated DNA Technologies, Coralville, IA) (Supplemental Table 9). qPCR was performed on an ABI 7900HT Fast Real-time PCR System with the SDS 2.2.2 program and using the ABI TaqMan universal PCR master mix (Applied Biosystems). All reactions were carried out in triplicate. The final concentration of the real-time primers was 300 nM, and the final concentration of the real-time probes was 250 nM. Initially, we examined the ORMES POU5F1 qPCR results obtained for a 5-fold dilution series of a control sample of ESC cDNAs, to determine the optimum primer dilution for future reactions. The control sample consisted of equal amounts of each ORMES cell line. With each subsequent run, we included triplicate 5-fold dilutions of control POU5F1 and GAPDH as standard curves. The cycling profile for each run was 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s followed by 60°C for 1 min, using the default ramp rate. POU5F1 gene expression for each sample was normalized relative to the endogenous housekeeping gene GAPDH. POU5F1 expression in each undifferentiated ESC sample was compared to POU5F1 expression in the pooled differentiated EBs. In each FC calculation, the average cycle-threshold (Ct) value of the POU5F1 expression level in the undifferentiated ESC line (normalized against the GAPDH expression level in the undifferentiated ESC sample) was compared to the average Ct value of the POU5F1 expression level in the pooled differentiated EBs (normalized against the GAPDH expression level in the pooled differentiated EB sample).
Undifferentiated ESCs were allowed to attach and grow in MEF-coated Lab-Tek II chamber slides (Fisher Scientific, Santa Clara, CA) for 2 days and were then fixed in 4% formaldehyde for 15 min. Whole differentiated EBs were fixed overnight before sectioning. Half of the sectioned EBs were subjected to Nissl staining according to standard protocols, to investigate the morphologically different cell types that comprised the differentiated EBs. The ESC samples and the remainder of the sectioned EBs were permeabilized for 40 min with 0.2% Triton X-100 and 0.1% Tween-20, and non-specific reactions were blocked with 10% normal goat serum (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). The cells were then incubated for 40 min in primary antibodies, washed three times, and exposed to secondary antibodies that were conjugated to fluorochromes (Jackson ImmunoResearch), before costaining with 2 µg/ml 4',6-diamidino-2-phenylindole (DAPI) for 10 min, whole-mounting onto slides, and examination by epifluorescence microscopy. All the primary antibodies (against OCT4, SSEA1, SSEA4, AFP, c-peptide, and nestin) were from Santa Cruz Biotechnology (Santa Cruz, CA).
Identification of Rhesus Monkey Stemness Gene Candicates
In order to identify genes that are implicated in maintaining rhesus monkey ESCs in an undifferentiated state (putative rhesus monkey stemness genes), the transcriptomes of undifferentiated and differentiated rhesus monkey ESCs were compared, together with the identification of genes that demonstrated a significant change in gene expression. ORMES-6 was used because it has the appropriate cellular morphology (flat monolayer colony with distinctive cobbled stem cell morphology and a high nucleo-cytoplasmic ratio) (Fig. 1A), stable strong proliferation rate, and normal karyotype [3]. After 1 mo of spontaneous differentiation, ORMES-6 EBs differentiated into several unique cell types (Fig. 1B). We detected fibroblasts and what appeared to be mature neurons by direct in vitro observation (Fig. 1B). Approximately 5% of the EBs differentiated into neurospheres or small tightly packed balls of neuronal cells, while another 5% of the EBs started to contract spontaneously, indicating the presence of cardiomyocytes. Sectioning and Nissl staining demonstrated that the interiors of the differentiated EBs contained a wide spectrum of morphologically distinct cell types (Fig. 1B), and immunocytochemistry demonstrated the presence of 
-fetoprotein (AFP)-, c-peptide-, and nestin-positive cells (Fig. 1B). AFP is a marker of early liver development, c-peptide is a marker of de novo insulin production, and nestin is a marker of neural progenitor cells. Histological analysis demonstrated that after 1 mo of differentiation, ORMES-6 ESCs formed a variety of different cell types, including representatives of all three germ layers, i.e., the endoderm (AFP), mesoderm (cardiomyocytes), and ectoderm (neurons). As significant cell type variability was observed among the EBs, the differentiated EBs were pooled for the microarray analysis.
To investigate the variation introduced by the microarray procedure and the possibility of superficial intrasample variability, two undifferentiated samples of ORMES-6 from the same passage were compared as biological replicates to the pooled differentiated EBs. Scatter plot analysis of the two biological replicates demonstrated a correlation coefficient of 0.99 (Supplemental Table 1, available online at GEO), which reflects acceptable levels of reproducibility and accuracy (Fig. 2A). As expected, many statistically significant differences were observed when undifferentiated ESCs were compared to pooled differentiated EBs (correlation coefficient of 0.75) (Fig. 2A, Supplemental Table 1). Subsequently, three biological replicates of ORMES-6 from two different passages were used in a comparison analysis with the pooled differentiated EBs and the results were filtered (see Materials and Methods section) to obtain a list of putative rhesus monkey stemness genes. Briefly, genes were selected that were present (P < 0.05) in all three biological replicates and significantly upregulated (>3-fold difference) in the ORMES-6 cells relative to the pooled differentiated EBs. We identified 367 rhesus monkey stemness gene candidates (Supplemental Table 2). In contrast, 3200 genes were significantly upregulated in the pooled differentiated EBs relative to the undifferentiated ORMES-6 cells (Supplemental Table 3), reflecting the presence of a large number of divergent cell types. The ontology of the stemness gene candidates was determined using both Affymetrix Netaffx and Amigo (Fig. 2B). The largest identified category corresponded to transcription factors/DNA-binding proteins, which perhaps reflects the wide array of potential differentiation pathways available to ESCs.
|
Conservation of Expression of Stemness Gene Candidates among Different Rhesus Monkey ESC Lines
The conservation of rhesus monkey stemness gene expression was investigated in five lines, ORMES 6, 7, 9, 10 and 13, which were chosen to reflect lines that were passaged initially by different techniques and that had slightly different growth characteristics. However, all five lines were karyotypically normal and all demonstrated standard ESC morphology and marker expression [3]. The characteristics of each of the five ORMES lines used in this study are summarized in Table 1. A level of conservation was assigned based on the number of ORMES lines that demonstrated significantly increased expression of a stemness gene (MAS 5.0; P < 0.002) compared to the pooled differentiated EBs (Table 2). A very high level of conservation meant the stemness gene was significantly upregulated in all five rhesus monkey ESC lines (5/5), and no conservation (None) meant that the stemness gene was significantly upregulated only in ORMES-6 (1/5). Over 85% of the rhesus monkey stemness genes candidates demonstrated high (4/5) or very high (5/5) conservation of expression across the five different rhesus monkey ESC lines (Supplemental Table 2).
POU5F1 (OCT4) Expression in Rhesus Monkey ESC Lines
As POU5F1 expression levels are tightly regulated in mouse ESCs [15], the level of POU5F1 expression in rhesus monkey ESC lines was determined using both a microarray and real-time qPCR (Table 3). As shown in Table 3, the fold-change calculations produced by the microarray analysis were, in most cases, significantly more conservative than those calculated by real-time qPCR. It should be noted that we observed approximately 20% variation in the qPCR calculated POU5F1 expression level between biological replicates of ORMES-6A and ORMES-6B, which reflects either biological differences in POU5F1 expression between the replicates or artificial variation introduced during the qPCR analysis. Despite the 20% variation between the replicates, we observed more than 350% variability in POU5F1 expression between the ORMES lines (Table 3). This is almost double the level of variability in POU5F1 expression that mouse ESCs can tolerate and still maintain a pluripotent state [15]. Therefore, we decided to investigate whether our five rhesus monkey ESC lines expressed the appropriate stem cell markers and whether they possessed the capacities to form EBs and differentiate into multiple cell types in vitro. RT-PCR analysis of the five undifferentiated rhesus monkey ESC lines confirmed the microarray results, demonstrating that all five rhesus monkey ESC lines expressed both POU5F1 and the stem cell marker NANOG (Fig. 3B). No POU5F1 or NANOG expression was detected in the pooled differentiated EBs or in the adult skin fibroblasts used as a negative somatic cell control (Fig. 3B). Immunocytochemical analysis demonstrated that all five rhesus monkey ESC lines expressed the POU5F1 protein (OCT4) and the human stem cell marker SSEA4, but not the mouse stem cell marker SSEA1 (Fig. 3A). It is interesting to note that although both the human ESC line (H1) and the rhesus monkey ESC lines (ORMES 6, 7, 9, 10, and 13) all expressed SSEA4, there were differences in the expression morphology, which probably reflect species-specific differences (Fig. 3A). When we allowed each ESC line to form EBs and spontaneously differentiate in vitro, all five lines demonstrated the ability to form differentiated EBs that contained multiple cell types. In addition to this spontaneous differentiation, it has been demonstrated that ORMES 6, 7, and 13 can undergo directed differentiation into neural, cardiac, and retinal cells [3].
|
Comparison of Rhesus Monkey, Human, and Murine ESC Gene Expression Profiles
Bhattacharya and coworkers have previously used comparisons of pooled RNA samples from multiple tissues to identify 92 genes that are expressed in five human ESC lines (BG01, BG02, GE01, GE09, and TE06) and in a pooled ESC sample [10]. Seventy-seven of the 92 genes were upregulated in undifferentiated human ESCs when compared to pooled differentiated EBs [11]. Of these 77 genes, 70 were represented on our microarray, and of these 70 genes, 27 were significantly upregulated in all three biological replicates of ORMES-6, 14 showed increased expression in some but not all of the ORMES-6 replicates, and 29 demonstrated no significant increase in expression in any of the ORMES-6 replicates in a comparison analysis with the pooled differentiated EBs (Table 4, Supplemental Table 1). Thus, 59% (41/70) of human stemness gene candidates were expressed in at least one rhesus monkey ESC biological replicate. We did not observe significant upregulation of REXO1, FOXD3 or TERT in the undifferentiated rhesus monkey ESC lines, while RT-PCR analysis indicated that SOX2 was expressed in both undifferentiated and differentiated rhesus monkey ESCs (Fig. 3A). Our results indicate that rhesus monkey ESCs primarily utilize the FGF signaling pathway (Table 5), similar to human ESCs [13]. The TGF-ß and Wnt signaling pathways undergo upregulation upon differentiation (Table 5). In order to identify mammalian stemness genes, we compared the upregulated expression profiles of rhesus monkey stemness genes (Supplemental Table 2) with the publicly available databases for human [10, 11] and mouse stemness genes [7], as defined by a 3-fold or greater change in expression. Five genes were strongly expressed (FC>3) in human, rhesus monkey, and murine ESCs (Table 6).
|
|
This report, to our knowledge, represents the first genome-wide expression analysis of rhesus monkey undifferentiated and differentiated ESCs. Initially, we compared two ORMES-6 biological replicates, to demonstrate an acceptable level of reproducibility and accuracy; we observed a correlation coefficient of 0.99. We then identified 367 putative stemness genes that were significantly upregulated in three biological undifferentiated replicates of ORMES-6 in a comparison analysis with pooled differentiated EBs. In all, 3200 genes were significantly upregulated, i.e., showed a >3-fold or greater change in expression, in the pooled differentiated EBs compared to the undifferentiated ESCs, which suggests a significantly higher level of complexity in the pooled differentiated EBs. It is likely that the large number of cell types into which ESCs can differentiate results in more genes being expressed in the pooled differentiated EBs than in the undifferentiated ESCs. However, ESCs that demonstrate greater transcriptional complexity in the undifferentiated state rather than the differentiated state have been described [13]. The differences between the results obtained in the present and previous studies may reflect the amount of time allowed for differentiation. In the study conducted by Wei et al. [13], ESCs were differentiated for 12 days, in contrast to the 1 mo used in the present study. We chose the 1-mo differentiation period, as certain differentiated cell types, such as neurons and cardiomyocytes, do not always appear until the third week of differentiation. Thus, the increased differentiation time may have significantly increased the complexity of the pooled differentiated EBs used in the comparison analysis with the undifferentiated ESCs. The five ORMES lines analyzed in this study (ORMES 6, 7, 9, 10, and 13) were obtained from different passage numbers (Table 1) and a proportion of the gene expression differences between the ORMES lines may be attributable to differences in the numbers of cell passages. However, when the expression levels of 367 putative rhesus stemness genes were analyzed across the five ORMES lines, despite the fact that the ORMES lines were analyzed at different passage numbers, over 85% of the putative rhesus stemness genes demonstrated high (4/5) or very high (5/5) levels of conservation of gene expression (Supplemental Table 2), which supports the candidacies of these 367 genes and suggests that ESC passage number may not be a significant factor in determining the expression of stemness genes in undifferentiated rhesus ESC lines.
We examined the expression of POU5F1 in rhesus monkey ESC lines. The POU5F1 gene encodes a POU transcription factor that induces and controls the expression of a wide variety of early embryonic genes. In mice, POU5F1 is expressed in ES cells, early embryonic cells, and in the germ line [16, 17]. POU5F1 expression is required to maintain pluripotency, whether in vivo in the embryo [18, 19] or in vitro in cultured ES cells [15], and it has been suggested that POU5F1 may be required not only to maintain pluripotency, but to establish it [16]. At the murine blastocyst stage, POU5F1 expression becomes restricted to the inner cell mass (ICM) cells, with downregulation in the trophectoderm cells [20]. In humans, POU5F1 is a quantitative pluripotency/stem cell marker that is expressed at high levels in the ICM [21] and in human ES cells [22], and expressed at very low levels in trophectoderm cells [21] and a variety of non-pluripotent adult tissues [23]. Similarly, in rhesus macaques, POU5F1 is expressed both in the ICM of rhesus monkey blastocysts and in rhesus monkey ESCs [24]. Previous research has revealed that when the expression of POU5F1 in mouse embryonic cells falls below 50% of the normal endogenous level, these cells differentiate into trophectoderm. In contrast, when the expression of POU5F1 is more than 50% higher the normal endogenous level, differentiation into parietal endoderm occurs [15]. We analyzed the expression of POU5F1 in five ORMES lines using both microarray analysis and qPCR. Our results demonstrate that the fold-change calculations produced by the microarray analysis are, in most cases, significantly more conservative than the fold-changes calculated by qPCR (Table 3). This supports the contention of Yuen and coworkers [25] that Affymetrix microarray chip analyses are accurate and reliable but tend to underestimate differences in gene expression [25]. Analysis by qPCR analysis also demonstrated a wide variability in POU5F1 expression, with a >3.5-fold variation from the lowest to highest observed level in the ORMES lines (Table 3). Despite this being almost double the variation in POU5F1 expression that mouse ESCs can tolerate and still maintain a pluripotent state [15], all five ORMES lines expressed the appropriate stem cell markers (as analyzed by RT-PCR and immunocytochemistry) and retained pluripotency (as analyzed by EB formation and spontaneous differentiation into multiple cell types). It is possible that rhesus monkey ESCs can tolerate a higher degree of POU5F1 gene expression variability than murine ESCs. This theory is supported by recent research that has revealed significant (3.3-fold) variability in POU5F1 expression between pluripotent human ESC lines [13]. However, it should be noted that the POU5F1-controlled differentiation observed by Niwa and coworkers occurred through an alteration in POU5F1 expression in a single murine ESC line [15]. The previously mentioned variability in POU5F1 expression levels in rhesus monkey and human ESC lines has been observed between different lines. Therefore, it remains a possibility that different primate ESC lines are able to maintain a pluripotent state over widely varying POU5F1 expression levels but undergo differentiation if the intraline POU5F1 expression level is altered by an increase or decrease of 50%. This is an interesting area for future study.
Several groups have analyzed the human ESC transcriptome and noted differences between human ESC lines [9–11, 26–31]. In the present study, we incorporated data from a group that identified putative human stemness genes based on the characterization of five different human ESC lines [10, 11]. It should be noted that the human-rhesus comparative results reported here may be unique to the human ESC study employed. By comparative analysis of human and rhesus monkey stemness candidate genes, we have demonstrated that 59% (41/70) of the human stemness gene candidates are expressed in at least one rhesus monkey ESC biological replicate, which indicates a significant correlation between the genetic programs of rhesus monkey and human ESCs. We did not observe significant upregulation of REXO1, FOXD3 or TERT in the undifferentiated rhesus ESC lines when compared to the pooled differentiated EBs, which mirrors the findings obtained for human ESCs [10]. We were surprised not to observe significant SOX2 upregulation in the undifferentiated rhesus monkey ESCs, since SOX2 is usually considered to be a major marker of stemness [10]. Bhattachrya and coworkers observed SOX2 upregulation, albeit only in some of the human ESC lines examined [10]. Due to the uncertainty surrounding the status of SOX2, we performed RT-PCR analysis and confirmed SOX2 expression in all of the undifferentiated rhesus monkey ESC lines (Fig. 3B). However, we also observed SOX2 expression in the pooled differentiated EBs (Fig. 3B), which suggests that SOX2 expression does not necessarily correlate with the undifferentiated ESC state. Our preliminary analysis of signaling pathway genes suggests that rhesus monkey ESCs primarily utilize the FGF signaling pathway (Table 5), similar to human ESCs [13]. This is not surprising, as both human and rhesus monkey ESCs require FGF2 to maintain the undifferentiated state, while murine ESCs require LIF [32].
A comparison of rhesus monkey, human, and murine stemness gene candidates led to the identification of the following five genes that are strongly expressed in undifferentiated ESCs: CCNB1, GDF3, LEFTB, POU5F1, and NANOG (Table 6). The cell cycle factor cyclin B1 (CCNB1) has previously been demonstrated to be extremely abundant in ESCs [33], reflecting the rapid rate of cell proliferation. GDF3 is a member of the TGF-ß superfamily that is expressed in both human and mouse ESCs [34] and blocks BMP signaling, thereby maintaining ESCs in an undifferentiated state [34]. LEFTY-B (LEFTB) is an inhibitor of nodal signaling that is downregulated very early upon differentiation [35]. In addition, POU5F1 and NANOG control a transcriptional network that is intricately connected to the maintenance of pluripotency, self-renewal, genome surveillance, and cell fate determination in ESCs [36]. Thus, the CCNB1, GDF3, LEFTB, POU5F1, and NANOG genes are significantly expressed in human, murine, and rhesus monkey ESCs and may represent the most basic core mammalian stemness genes.
Microarray analysis, immunocytochemistry, RT-PCR, and qPCR were used to characterize and transcriptionally profile rhesus monkey ESCs. We identified 367 rhesus monkey putative stemness genes, with a high level (>85%) of conservation across five different rhesus monkey embryonic stem cell lines. We have also demonstrated that rhesus monkey ESC lines maintain a pluripotent undifferentiated state over a wide range of POU5F1 expression levels, and we have compared rhesus monkey, human, and murine stemness gene candidates to identify five key mammalian stemness genes, CCNB1, GDF3, LEFTB, POU5F1, and NANOG. This transcriptional profiling of rhesus monkey ESCs provides a foundation for future research using rhesus monkey ESCs as an accurate model to develop strategies to prevent immune rejection and test the feasibility, safety, and efficacy of ESC-based medical treatments.
ACKNOWLEDGMENTS
We thank the Oregon Health Sciences University West Campus Affymetrix Microarray Core for assistance in performing the microarray analysis, the OHSU Assisted Reproductive Technology and Embryonic Stem Cell core for help in culturing ESCs, Dr. Terry Morgan for performing histological identification of the teratomas, and Dr. Jon Hennebold for his helpful review of this manuscript.
FOOTNOTES
1Supported by Oregon National Primate Research Center pilot grant GPRC46174 (to J.A.B.) and National Institute of Health grant RR00163 (to P.K.). 
Correspondence: 2 FAX: 503 690 5563; e-mail: wolfd{at}ohsu.edu
Received: 13 May 2006.
First decision: 9 June 2006.
Accepted: 26 July 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  S. P. Gong, H. Kim, E. J. Lee, S. T. Lee, S. Moon, H.-J. Lee, and J. M. Lim Change in gene expression of mouse embryonic stem cells derived from parthenogenetic activation Hum. Reprod., December 23, 2008; (2008) den388v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Dighe, L. Clepper, D. Pedersen, J. Byrne, B. Ferguson, S. Gokhale, M. C. T. Penedo, D. Wolf, and S. Mitalipov Heterozygous Embryonic Stem Cell Lines Derived from Nonhuman Primate Parthenotes Stem Cells, March 1, 2008; 26(3): 756 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Navara, J. D. Mich-Basso, C. J. Redinger, A. Ben-Yehudah, E. Jacoby, E. Kovkarova-Naumovski, M. Sukhwani, K. Orwig, N. Kaminski, C. A. Castro, et al. Pedigreed Primate Embryonic Stem Cells Express Homogeneous Familial Gene Profiles Stem Cells, November 1, 2007; 25(11): 2695 - 2704. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
