Human embryonic stem cells (hESC) promise tremendous potential as a developmental and cell therapeutic tool. The combined effort of stimulatory and inhibitory signals regulating gene expression, which drives the tissue differentiation and morphogenetic processes during early embryogenesis, is still very poorly understood. With the scarcity of availability of human embryos for research, hESC can be used as an alternative source to study the early human embryogenesis. Hyaluronan (HA), a simple hydrating sugar, is present abundantly in the female reproductive tract during fertilization, embryo growth, and implantation and plays an important role in early development of the mammalian embryo. HA and its binding protein RHAMM regulate various cellular and hydrodynamic processes from cell migration, proliferation, and signaling to regulation of gene expression, cell differentiation, morphogenesis, and metastasis via both extracellular and intracellular pathways. In this study, we show for the first time that HA synthase gene HAS2 and its binding receptor RHAMM are differentially expressed during all stages of preimplantation human embryos and hESC. RHAMM expression is significantly downregulated during differentiation of hESC, in contrast to HAS2, which is significantly upregulated. Most importantly, RHAMM knockdown results in downregulation of several pluripotency markers in hESC, induction of early extraembryonic lineages, loss of cell viability, and changes in hESC cycle. These data therefore highlight an important role for RHAMM in maintenance of hESC pluripotency, viability, and cell cycle control. Interestingly, HAS2 knockdown results in suppression of hESC differentiation without affecting hESC pluripotency. This suggests an intrinsic role for HAS2 in hESC differentiation process. In accordance with this, addition of exogenous HA to the differentiation medium enhances hESC differentiation to mesodermal and cardiac lineages.
Disclosure of potential conflicts of interest is found at the end of this article.
Human embryonic stem cells (hESC) have promising diagnostic, clinical, and therapeutic applications, and this has prompted it to be in the forefront of medical research. It was almost two decades after the derivation of first mouse embryonic stem cell line (mESC)  that Thomson et al.  derived the first hESC line on mouse embryonic fibroblast (MEF) feeders. Since then, hESC have been isolated from the inner cell mass of 5–8-day-old blastocyst [2, –4], morula [5, 6], arrested early-stage embryos , and single blastomeres . Early development of preimplantation human embryos and the mechanisms that drive gene regulation during embryo development are not clearly understood. In view of this, hESC offer an important developmental model to understand the biology of early development.
Hyaluronan (HA) is a simple nonsulfated glycosaminoglycan (GAG) consisting of straight chains of alternating disaccharide units of β-1-glucuronic acid and N-acetyl-glucosamine that has long been known to be an important constituent of extracellular matrix . HA is synthesized as an unbranched polysaccharide by one of the three isoforms of hyaluronan synthase (HAS) gene, HAS1, HAS2, and HAS3 [9, –11]. The products of HAS genes are located at the inner surface of the cell membrane. Once synthesized, however, HA is extruded into the extracellular space as an unmodified polymer. HA was initially considered to be present only on the cell surface, but several reports have shown that HA is present intracellularly at various locations, such as cytoplasm, rough endoplasmic reticulum, nuclei and nucleoli of proliferating cells, and linked to centrosome and cytoskeletal compartments [12, –14]. HA mediates its diverse functions both extracellularly and intracellularly by interacting with many proteins termed hyaladherins.
The two main receptors known to bind HA are CD44 and RHAMM (receptor for hyaluronic acid-mediated motility). RHAMM is a unique B(X)7 B hyaladherin protein family member. The encoding gene is located on human chromosome 5q33.2 and has 18 exons [15, –17]. These receptors are now established to be the main signaling receptors for HA (reviewed in ), and along with HA, they regulate various cellular and dynamic processes, such as cell-to-cell adhesion, cell migration, morphogenesis, cell proliferation (including mitosis), cell signaling, regulation of gene expression, RNA splicing, cell differentiation, and metastasis [8, 19, 20]. HA is abundant in female reproductive tract fluids [21, 22] and has been shown to play a significant role in mammalian fertilization, implantation, extraembryonic cavitation, and embryo growth, as well as tissue morphogenesis, signifying its importance in embryo-maternal communication [23, , , –27]. Whereas HAS2 knockout in mice was detrimental to embryo survival and growth because of lethal cardiovascular anomalies , RHAMM knockout mice were viable with normal growth and appearance but had reduced fertility when mated with homozygous mice .
Although RHAMM is not essential for embryo viability, it has been found to play a profound role in several cellular events, such as mitosis, cell proliferation, and migration . RHAMM is highly expressed in the G2/M phase of the cell cycle, thus controlling mitosis; it is believed to increase cellular motility through direct interaction with the cytoskeleton and may have a role in the separation and migration of daughter cells following mitosis [15, 30, –32]. RHAMM has been noted to be highly expressed during tumorigenesis [29, 33] and also in some actively dividing normal tissues, such as thymus, spleen, testes, and placenta . It is evident that RHAMM regulates cell proliferation at low culture density but has no influence at higher cell confluence, suggesting that RHAMM protein promotes interaction of cell signaling with extracellular milieu [29, 35].
Today, there is increasing evidence linking HA to stem cell biology. It has recently been demonstrated that HA is synthesized and secreted by adult hematopoietic stem cells and mesenchymal progenitors and is involved in the regulation of their proliferation and differentiation in vitro, as well as their mobilization and trafficking [36, –38].
Taken together, these data suggest that HA and its two key genes, HAS2 and RHAMM, may play a significant role in human early embryogenesis and embryonic stem cell characterization. Notwithstanding this, the role of RHAMM and HAS2 in preimplantation human embryos and hESC has not been reported. Here, we report for the first time expression of HA-related genes RHAMM and HAS2 during preimplantation human embryo development up to the hESC characterization. We show that RHAMM is expressed in hESC and is significantly downregulated during the differentiation process. Inhibition of RHAMM in hESC causes loss of pluripotency and cell viability, as well as changes in cell cycle, thus highlighting an important role for RHAMM in maintenance of hESC pluripotency and proliferation. In addition, we highlight an important role for HAS2 in regulating HA synthesis and hESC differentiation. Most importantly, we report in vitro enhancement of cardiogenesis in differentiating hESC on addition of HA to differentiation medium.
Materials and Methods
Source of Human Preimplantation Embryos and Embryo Culture
This study had the approval of the relevant local ethics committee and the Human Fertilization and Embryology Authority. Surplus embryos were collected after patient consent  and cultured as previously described . In addition to fresh embryos, frozen embryos were also obtained for this research, donated by couples who no longer wanted them for their own treatments. The frozen embryos were at day 2 or 3 of fertilization and were thawed using the MediCult thawing kit (MediCult, Jyllinge, Denmark, http://www.medicult.com) per the manufacturer's instructions. After thawing, the embryos were transferred into G2.3 medium droplets (Vitrolife, Vitrolife, Kungsbacka, Sweden, http://www.vitrolife.com) for further culture.
Source of Human Embryonic Stem Cell (hESC) Lines and Cell Culture
Three hESC lines were used in this study: hES-NCL1  and two commercially available cell lines, H1 and H9 from WiCell Research Institute (Madison, WI, http://www.wicell.org). The hESC were cultured on MEF as previously described  or under feeder-free conditions with MEF-conditioned media . The data presented throughout this study represent an average of results from at least two of the cell lines.
In Vitro Differentiation of hESC by EB Formation
The hESC colonies were broken into small pieces of 5–20 cells and transferred to ultralow-attachment 24 well plates (Corning International, New York, www.corning.com) for the formation of embryoid bodies. The differentiation medium consisted of 78% knockout Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 20% fetal calf serum (FCS) (HyClone), 1% l-glutamine + penicillin/streptomycin (Invitrogen), and 1% nonessential amino acids (Invitrogen). Medium was changed every 3 days, and the EBs were collected at different time points for RNA extraction and subsequent reverse transcription polymerase chain reaction (RT-PCR) analysis.
Immunostaining of Human Embryos and hESC
Immunostaining of human embryos and hESC was carried out as described in one of our earlier publications . In brief, the zona pellucida of the embryos was removed using acid Tyrode's solution (MediCult) before fixation with 4% paraformaldehyde for 10 minutes and permeabilization with 0.2% Triton X-100 in phosphate-buffered saline (PBST) for 30 minutes. After blocking with 5% nonfat dried milk in PBST for 30 minutes, the embryos were incubated with the primary antibodies in 1% nonfat dried milk in PBST for 2 hours. The embryos were washed with 1% nonfat dried milk in PBST before addition of the secondary antibody. The primary antibodies used in this study were raised in rabbit. RHAMM polyclonal antibody was obtained as a kind gift from Dr V. Assman (Hamburg, Germany), who generated this antibody (1:50 final concentration), and HAS2 polyclonal antibody was obtained as a kind gift from Dr. P. Heldin (Ludwig Institute for Cancer Research, Uppsala, Sweden) (10 μg/ml). The embryos were washed with 1% nonfat dried milk in PBST before addition of the secondary antibody (fluorescein isothiocyanate [FITC]-conjugated anti-rabbit IgG antibody, 6 μg/ml). The embryos were washed again before fluorescence microscopy in 5% FCS in phosphate-buffered saline (PBS). The embryos or hESC incubated with secondary antibodies alone were used as negative controls.
RNA extraction and subsequent RT-PCR analysis was performed on human embryos and hESC as described earlier [6, 41]. The gene accession numbers and the primer sequences are shown in supplemental online Table 1. The annealing temperature for the PCR was set at 58°C unless otherwise specified.
Flow Cytometry Analysis of RHAMM Expression in hESC
The hESC were subjected to Accutase digestion for 15 minutes to obtain single-cell suspension and then spun in 15-ml centrifuge tube. The supernatant was discarded and the cells were then fixed by resuspending in 0.25% paraformaldehyde in PBS for 10 minutes at 37°C. The cells were spun again briefly, and the fixative removed. The cells were washed once with PBS and centrifuged, and the supernatant was discarded. Cell permeabilization was carried out by incubating for 20 minutes with ice-cold 90% methanol. After several washes in PBS, cells were incubated with primary antibody, RHAMM (1:50 dilution) or HAS2 (10 μg/ml) for 60 minutes at room temperature and washed twice, and then the FITC-conjugated anti-rabbit IgG antibody (6 μg/ml) was added and incubated for an additional 30 minutes at room temperature. The analysis was carried out using the FACSCalibur (BD Biosciences, San Diego, http://www.bdbiosciences.com) and the CellQuest software. At least 10,000 events were acquired in each case. RHAMM cell surface expression was carried out using similar techniques, with the exception of fixation and permeabilization steps, which were omitted.
HA Uptake Studies
BODIPY fluorescein-HA (FL-HA) conjugate (Molecular Probes Inc., Eugene, OR, http://www.probes.invitrogen.com) was used to study receptor-mediated uptake of HA by viable human embryos and hESC. Human embryos at various stages of development were incubated overnight at 37°C in G2.3 medium droplets containing 100 μg/ml concentration of FL-HA conjugate. The embryos were later washed three times in fresh G2.3 droplets before being subjected to fluorescence microscopy using an FITC filter. The images were acquired using a Zeiss microscope and Axiovision software (Carl Zeiss, Jena, Germany, http://www.zeiss.com). For HA uptake studies in live hESC, a similar protocol was used, but instead of G2.3 medium droplets, 300 μl of human embryonic stem medium with FL-HA conjugate was added to hESC colonies in each well of four-well plates. The cells incubated with fluorescein only were used as a negative control for this experiment.
Small Interfering RNAs and Transfections of hESC
Stealth RNA interference (RNAi) has been successfully used for RNAi in embryonic stem cells . Stealth RNAi sequences were obtained from Invitrogen. Three target Stealth RNAi sequences were ordered for RHAMM and HAS2 as follows.
RHAMM sequences (GenBank accession no. U29343):
R1, CCAGUAUCCUUUCAGAAAUCACAAA (163–187)
R2, GGCGUCUCCUCUAUGAAGAACUAUA (1,799–1,823)
R3, GGAGUCUCGAAGAGUCUCAAGGGAA (662–686).
HAS2 sequences (GenBank accession no. NM_005328):
H1, CCAGCCUCAUCUGUGGAGAUGGUAA (1,190–1,214)
H2, GCAGGCGGAAGAAGGGACAACAAUA (2,148–2,172)
H3, GGGCAGAGACAAAUCAGCCACUUAU (949–973).
Scrambled small interfering RNAs (siRNAs) with similar guanine cytosine (GC) content were purchased from Invitrogen and used as negative controls for transfections. Transfection of Stealth RNAi was carried out using Lipofectamine 2000 following the manufacturer's instructions. The transfection efficiency was assessed under the microscope by examining the intracellular uptake of BLOCK-iT Fluorescent Oligo (Invitrogen) as described earlier . The best transfection efficiency achieved in embryonic carcinoma (EC) cells was 70%–85%, whereas in hESC, more than 60% using 50 pmol of siRNA in each well of a four-well plate. The cells were incubated for 24 hours at 37°C in a humidified CO2 incubator, and for maximum transfection efficiency, this process was repeated after 24 hours. The cells were examined under a Zeiss microscope for morphological changes, and the samples were collected for various assays at different time points as described below.
LightCycler Real-Time PCR Analysis
LightCycler real-time PCR analysis was carried out as described previously . GAPDH was used as endogenous reference gene to calculate relative gene expression. The relative gene expression in Stealth RNAi-delivered hESC samples was normalized to that of the respective GC content-matched negative control samples, which was considered as 1 (100%). The data were analyzed by using LightCycler Relative quantification software, version 1.01 (Roche Diagnostics, Basel, Switzerland, http://www.roche-applied-science.com). Statistical significance was calculated by pairwise Student's t test, and p < .05 was considered statistically significant.
Cells undergoing apoptosis can be enumerated using the Annexin V-FITC apoptosis detection kit (BD Biosciences). The protocol was carried out in accordance with the manufacturer's instructions and briefly comprises the following. Cells were harvested using Accutase, washed twice with ice-cold phosphate-buffered saline, and counted. Cells (1 × 105) were suspended in 100 μl of 1× binding buffer (supplied), and then 5 μl of Annexin V-FITC and 5 μl of 7-aminoactinomycin D (7-AAD) solution were added. The mixture was vortexed gently and incubated for 15 minutes at room temperature in the dark. Four hundred microliters of 1× binding buffer was added, and the cells were analyzed by flow cytometry (FACSCalibur; Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com).
5-Bromo-2′-Deoxyuridine Cell Proliferation Assay
Human ESC proliferation was measured by incorporation of 5-bromo-2′-deoxyuridine using the BD Pharmingen BrdU Flow kit (San Diego, http://www.bdbiosciences.com/index_us.shtml) following the manufacturer's instructions. Cells were stained with FITC anti-5-bromo-2′-deoxyuridine (anti-BrdU) and 7-aminoactinomycin. Cells from the same population that were not BrdU-labeled were used as negative control. Flow cytometry analysis was carried out using a FACSCalibur (Becton Dickinson) and the CellQuest software.
Cell Cycle Analysis
Cell cycle analysis was performed using the CycleTest Plus DNA reagent kit (Becton Dickinson). Human ESC were harvested by Accutase treatment and counted (using a hemocytometer). Five hundred thousand cells were fixed, permeabilized, and stained in accordance with manufacturer's instructions, and the sample was analyzed by flow cytometry (FACSVantage; Becton Dickinson) measuring FL2 area versus total counts. The data were analyzed using ModFit (Verity Software House, Topsham, ME, http://www.vsh.com) and FlowJo to generate percentages of cells in G1, S, and G2/M phases.
Alkaline Phosphatase Assay
The alkaline phosphatase (AP) staining was carried out using the Alkaline Phosphatase Detection Kit following the manufacturer's instructions (Chemicon, Temecula, CA, http://www.chemicon.com) as previously described .
HA Quantification in Cell Culture Supernatant in HAS2 Knockdown Experiment
To establish the effect of HAS2 knockdown on HA synthesis, HA in cell culture medium supernatant was assessed by enzyme-linked binding protein assay using Corgenix HA-enzyme-linked immunosorbent assay (ELISA) kit (Corgenix, Broomfield, CO, http://www.corgenixonline.com) according to the manufacturer's instructions.
Exogenous HA-Supplemented Ex Vivo hESC Differentiation Experiment
To assess the short-term effect of HA-induced hESC differentiation, the test group differentiation medium was supplemented with 0.75 mg/ml high molecular weight HA (Fluka Biochemica, Buchs, Switzerland, http://goliath.ecnext.com). In vitro culture of cells in regular differentiation media, as described earlier , was taken as the control group for comparison. To assess the specificity of HA-induced differentiation compared with the property of a GAG in general, another GAG, chondroitin sulfate (0.75 mg/ml) was supplemented to differentiation media as the second comparable group in the experiment. The cells were grown as EBs in their respective culture media for up to 7 days and then were plated onto gelatin-coated tissue culture plates for further differentiation for another 10 days at a density of 3–5 EBs per well (2 cm2 surface area).
Morphological Assessment of Contractile Areas and Real-Time RT-PCR Analysis
Daily microscopic monitoring was carried out to assess the number of contracting EBs. The percentage of EBs displaying contractile activity and the distribution of the timing of onset of spontaneous beating were evaluated and plotted on a graph. The number of contractions per minute was calculated to assess the beating rate. The cells were cultured as Ebs for 7 days. The Ebs were plated on gelatin-coated dishes for a further 10 days, postulating in a total 17 days in culture, and the samples were then collected for real-time RT-PCR analysis as described above for gene expression of cardiac markers. We used previously published primer sequences for the expression of early cardiac markers as well as β-III tubulin . The annealing temperature for these primers was set at 62°C.
Immunocytochemistry of hESC-Derived Cardiomyocytes
The cells were fixed with 90% methanol for the immunostaining procedure. Cells were incubated with the primary antibody staining for 2 hours with α-actinin sarcomeric (10 μg/ml) or tropomyosin sarcomeric (10 μg/ml) from Sigma-Aldrich (St. Louis, http://www.sigmaaldrich.com) and then washed three times with PBS and 2% fetal bovine serum. This was followed by 1 hour of incubation with FITC-conjugated anti-mouse IgG (6 μg/ml) as the secondary antibody. 4,6-Diamidino-2-phenylindole nuclear staining was performed for a further 5 minutes and finally, after an additional three washes, the cells were viewed under fluorescent microscopy. The negative control group contained cells stained only with secondary antibody.
Two-tailed pairwise Student's t test was used to analyze results obtained from two samples with one time point. Analysis of variance (single factor or two factors with replication) was used to compare multiple samples (at one time or several time points). The results were considered significant if p < .05.
Expression Analyses of HA-Related Genes HAS2 and RHAMM in Early Preimplantation Human Embryos
The expression of HA-related genes HAS2 and RHAMM in early human embryos was studied by immunofluorescence, and relative expression was investigated by real-time RT-PCR analyses. Immunocytochemistry analysis in human embryos indicated that both genes are expressed in all stages of preimplantation human embryos up to the blastocyst stage (Fig. 1A). RHAMM expression was restricted to the cell surface in very early-stage embryos (2–3-cell stage; Fig. 1A); however, both cell surface and intracellular staining were observed during later stages of preimplantation development (Fig. 1A). HAS2 showed similar intense staining in all stages of human preimplantation embryos (Fig. 1A). Staining of a six-cell-stage embryo and a blastocyst with the secondary antibody only did not show a distinct fluorescence pattern, thus confirming the validity of the antibody staining for both RHAMM and HAS2 (Fig. 1A).
To confirm the results obtained by fluorescence microscopy, real-time RT-PCR analysis was carried out. This analysis showed that the relative expression of RHAMM increased transiently from 4-cell to 8–12-cell cleavage-stage embryos and then remained more or less static at the morula and blastocyst stages before a further dramatic increase noted during the blastocyst expansion stage (p < .05; Fig. 1B). HAS2 expression remained consistent until the early blastocyst stage and then rapidly increased to a significant fold (p < .05) at the expanded blastocyst stage (Fig. 1C).
RHAMM and HAS2 Expression in hESC and EBs
A combination of real-time RT-PCR, flow cytometry, and immunocytochemistry analysis was used to assess the expression of RHAMM and HAS2 in hESC. Immunocytochemistry revealed intense RHAMM reactivity in undifferentiated hESC (Fig. 2A). RHAMM expression was noted both on the cell membrane and intracellularly using immunocytochemistry and flow cytometry (Fig. 2A, 2C). Immunocytochemistry and flow cytometry analysis showed that intracellular HAS2 staining was noted in 97% of hESC (Fig. 2B, 2C).
Relative quantification of these genes by real-time RT-PCR demonstrated that RHAMM is downregulated almost twofold on differentiation of hESC (p < .01) (Fig. 2D), similar to the DNA Affymetrix data analyses for RHAMM expression in both H1 and hES-NCL1 hESC (data not shown). Although HAS2 mRNA levels were high in undifferentiated hESC, the expression pattern showed a maximum peak after 10 days of differentiation as embryoid bodies (p < .01; Fig. 2E).
HA Uptake by Early Human Embryos and hESC
We assessed the uptake of exogenous fluorescein-labeled HA by viable preimplantation embryos and undifferentiated hESC, and our findings show that exogenous HA is taken up by both early human embryos and hESC, thus suggesting the presence of HA-binding receptors (Fig. 3). Our experiments showed that the uptake of HA was significantly correlated to the dose of HA added to the hESC media (supplemental online Fig. 1A). In addition, the uptake of HA seemed to increase over time until it reached a plateau at 30 hours. Longer incubation with HA did not increase the uptake any further (supplemental online Fig. 1B). The specificity of HA uptake was confirmed in competition experiments where HA did interfere with the uptake of FL-labeled HA (supplemental online Fig. 1C).
Stealth RNAi-Mediated Gene Knockdown in hESC
The Lipofectamine 2000 transfection method has been used efficiently in mouse embryonic stem cells for introduction of siRNAs . To assess the transfection efficiency in human EC cells and hESC, we used BLOCK-iT Fluorescent Oligo (Invitrogen). As shown in supplemental online Figure 2, we found transfection efficiencies of approximately 70%–85% for human EC cells and 60%–75% for hESC. We used three short-strand Stealth RNAi sequences each for RHAMM and HAS2 in human EC cells to determine the best specific sequence for RNAi-mediated knockdown. Universal GC content-matched (low GC for RHAMM and medium GC for HAS2) were used as negative controls for all RNAi experiments, as described in one of our earlier publications . The maximum knockdown efficiency of the intended target gene was observed at 48 hours post-transfection in EC cells (Fig. 4A, 4B), with 85% (±7.02%) knockdown for RHAMM sequence 1 (R1) and 91% (±1%) for HAS2 sequence 1 (H1). These sequences achieved 65% and 86% knockdown in the expression of RHAMM and HAS2, respectively, in hESC compared with their respective negative controls (Fig. 4C, 4D). The knockdown efficiency in hESC was similar to previously reported papers [41, 42]. The mRNA levels for the respective targeted genes remained lower compared with the control siRNA-treated cells (p < .05) for up to 6 days post-transfection (Fig. 4C, 4D). The effects of siRNA transfection on gene regulation are transient, and this is apparent in the elevated levels of expression of the targeted gene from days 2–4 and 6 (Fig. 4C, 4D). We have reported similar effects during the downregulation of NANOG in hESC by RNAi . An additional factor could be increased cell death or reduced proliferation/survival as result of knockdown of genes crucial to human ESC growth (such as NANOG and RHAMM), which results in faster growth of wild-type cells (that have not taken the siRNA) compared with transfected cells. To minimize these effects and increase the efficiency of RNAi targeting, repeated rounds of transfections, as reported in here and our previous publication , need to be carried out.
RHAMM Knockdown Influences Cell Viability and Cell Proliferation
The effect of RHAMM knockdown on cell apoptosis and proliferation was assessed 48 and 96 hours post-transfection by flow cytometry, as shown in Figure 5. RHAMM downregulation resulted in a significant increase in number of late apoptotic cells (Annexin V+/7-AAD+) compared with negative control (p < .05; Fig. 5A). Cell proliferation analysis by flow cytometry demonstrated that RHAMM knockdown did not prevent DNA synthesis but permitted very few cells to undergo mitosis (Fig. 5B). This finding was also confirmed by cell cycle analysis using propidium iodide (data not shown), which showed increased percentages of cells in the G1 and S phases of the cell cycle and decreased percentages in G2/M.
Downregulation of RHAMM Results in Loss of hESC Pluripotency and Differentiation Toward Extraembryonic Lineages
Downregulation of RHAMM during hESC differentiation suggests that RHAMM is likely to be involved in maintenance of hESC pluripotency. To investigate this further, we performed RHAMM knockdown studies using RNA interference as described in Materials and Methods. Alkaline phosphatase staining studies performed 48 hours after RNAi procedure indicated that there was more often a central differentiation noted in hESC colonies transfected with RHAMM siRNA compared with the ones transfected with control siRNA, resulting in a significant reduction in the number of AP+ colonies after RHAMM knockdown (Fig. 5C). Real-time RT-PCR analysis also suggested that RHAMM knockdown results in downregulation of pluripotent markers such as OCT4, NANOG, and REX1 (Fig. 5D). In addition, real-time RT-PCR analysis for some of the early differentiation genes showed upregulation of the trophectodermal marker CDX2, as well as primitive endoderm markers GATA4 and IHH at 6 days post-siRNA transfection in hESC (Fig. 5E). We did not find significant differences in expression of primitive ectodermal (FGF5) and mesodermal (BRACHYURY, VIMENTIN) markers between control and RHAMM siRNA-transfected hESC (data not shown).
Downregulation of HAS2 Alters HA Synthesis and hESC Differentiation
Real-time RT-PCR analysis after HAS2 siRNA transfection suggested no changes in expression of pluripotent markers (data not shown), indicating that HAS2 is unlikely to play a role in maintenance of hESC pluripotency. ELISAs indicated that downregulation of HAS2 lead to a reduction in HA synthesis by hESC as expected (Fig. 6A). Human ESC transfected with siRNA for HAS2 or scrambled control were plated at low density on Matrigel (to encourage spontaneous differentiation) for 6 days and then analyzed by real-time RT-PCR for the expression of lineage markers. This analysis indicated that HAS2 knockdown suppresses differentiation to primitive endoderm, primitive ectoderm, and mesoderm, as well as ectodermal and trophectodermal lineages (Fig. 6B, 6C). Since HAS2 has been reported to be crucial for cardiac and vascular development in mice , we investigated the expression of several hematoendothelial markers (CD31, KDR, and CD34), as well as cardiac markers (cTnT and MLC2A; Fig. 6D) by real-time RT-PCR. This indicated that development of all lineages was suppressed upon HAS2 knockdown. It is currently unclear whether this phenomenon is due to suppression of mesodermal differentiation or to an additional effect of HAS2 in early hematoendothelial and cardiac development.
Exogenous HA Enhances Cardiogenic Differentiation of hESC
The role of the most prominent HA synthase gene HAS2 in cardiac morphogenesis is well reported to date [28, 45, 46]. Notwithstanding this, the role of HAS2 in early human cardiac development is as yet unknown. Early embryonic lethality in Has2 knockout mice due to severe cardiac and vascular abrogation and its rescue observed in explants on addition of HA  suggest that exogenous HA may have a potential role in cardiogenesis. This hypothesis is further supported by another study that has shown a positive effect for a hyaluronan ester in induction of cardiogenesis in murine embryonic stem cells . This prompted us to explore the role of exogenous HA on cardiogenesis using hESC as a developmental model. We used the EB differentiation system described previously  and added either HA or chondroitin sulfate (as a GAG control) to the differentiation media (both at 0.75 mg/ml) to assess any effects on the number of EBs showing spontaneous contractions in culture. The number of contracting EBs observed by microscopy was significantly enhanced by HA compared with standard differentiation media and a GAG control (Fig. 7A). The specificity of contractile areas to be cardiac cells was confirmed by cardiac-specific immunostaining (Fig. 7B). Addition of different amounts of exogenous HA up to 0.75 mg/ml (0.25, 0.5 and 0.75 mg/ml) to the differentiation media did not result in significant changes in the total numbers of cells obtained from the respective EBs after 17 days of culture (supplemental online Fig. 3). Significant differences in cell number (p < .05) were found between control EBs (0 mg/ml HA) and the ones that were cultured in the presence of 1 mg/ml HA (supplemental online Fig. 3). Although significant differences in contractile activity were found between the control group (0 mg/ml) and EBs cultured in the presence of 0.75 mg/ml HA (Fig. 7A), different amounts of HA (0.25, 0.5 and 1 mg/ml) did not show significant changes in this activity (data not shown). Addition of different amounts of exogenous HA to hESC showed significant increase (p < .005) in cell number with increasing concentrations of HA (supplemental online Fig. 4).
Real-time RT-PCR analyses showed a significant upregulation of BRACHYURY as well as a number of cardiac markers (NKX2.5, MLC2A, cTnT, ANP, and GATA4) in the HA-treated group compared with the control (Fig. 7C). These data suggest that addition of HA is likely to enhance commitment of hESC to mesodermal lineages. Most interestingly, addition of HA seems to enhance hESC commitment to cardiac lineages; however, at present it is unclear whether this is related to the effects of HA on mesodermal commitment or is independent from those effects.
In this study, we have assessed the role of hyaluronan, its synthase gene HAS2, and its binding receptor RHAMM in early human embryogenesis and in hESC pluripotency and differentiation. Here, we show that HAS2 and RHAMM are expressed in all stages of early preimplantation human embryos as well as in hESC and are regulated upon differentiation of hESC. Furthermore, we show that HA is taken up and internalized by early human embryos and hESC and ex vivo differentiation of hESC in the presence of HA enhances differentiation toward mesodermal lineages. Knockdown experiments of RHAMM and HAS2 presented in this study show that RHAMM is important for the maintenance of hESC pluripotency, viability, and cell proliferation and cell cycle, whereas HAS2 is crucial for differentiation capacity of hESC.
The significance of HA during oocyte maturation and embryo development in other species is well acknowledged [21, 23, 25, –27, 48, –50]. Notwithstanding this progress, there is a paucity of data regarding the role of HA in early human development that prompted us to undertake this study.
Studies carried out in bovine embryos have shown that RHAMM expression was not visible until the eight-cell stage . Its expression seemed to decrease from the 8-cell to the 16-cell stage and then increased in morula and blastocyst, with the highest expression seen in expanded blastocysts . Our study indicates that RHAMM expression is found in all stages of human preimplantation development, with the very early embryos (less than four-cell stage) showing only pericellular staining and later embryos showing both cell surface and intracellular staining. Real-time RT-PCR analysis indicated variations in RHAMM expression in different stages of preimplantation development; however, the highest expression was observed in expanded blastocysts.
Investigations of HA synthases in mammalian embryos have shown that HAS2 is the major contributor of HA synthesis and is the most crucial of the three HAS enzymes [14, 28, 51, , –54]. The results presented in this study show that HAS2 is expressed in all stages of human preimplantation development; however, a dramatic increase is observed at the expanded blastocyst stage, which is concomitant with the fact that HA synthesis increases during the implantation phase of blastocysts . Thus, the differential regulation of human HAS2 expression in early embryo development may implicate its possible role in tissue remodeling.
Our study reveals for the first time that both RHAMM and HAS2 are expressed in hESC and that these cells are capable of exogenous fluorescein-labeled HA uptake. Previous studies have shown that intracellular HA and RHAMM have a close spatial interaction and that by blocking RHAMM, the intracellular binding of fluorescein-HA is affected in hematopoietic cells [38, 55]. Moreover, the size of HA may influence its uptake pattern by cells, with most intracellular HA being of low molecular weight and localized in vesicles [56, 57]. Fragmented fluorescein-HA localized near the nuclei, whereas high molecular weight fluorescein-HA was mostly located in large endosomes . Our fluorescein-HA was of high molecular weight (HMW) and was taken up by cells to localize into the cytoplasm, suggesting that hESC is capable of even HMW HA uptake. The expression of RHAMM and the ability of hESC to internalize exogenous HA cumulatively emphasize the possible interaction of HA with its binding proteins in hESC to exhibit its cellular functions.
The molecular basis of RNAi is to use homologous short RNA transcripts, causing the deprivation of target mRNA, which in turn leads to post-transcriptional silencing. Following the successful application of siRNA duplexes for gene silencing in cultured mammalian cells [58, 59], RNAi has now been increasingly used to study the gene importance in embryonic stem cell characterization. Recent reports show that the effective downregulation of OCT4 and NANOG in embryonic stem cells by RNAi leads to differentiation toward extraembryonic lineages and establishes them as regulators of pluripotency [41, 42, 60]. Similarly, downregulation of RHAMM by RNAi results in downregulation of key pluripotency markers (OCT4, NANOG, and REX1) and shows predilection toward extraembryonic cell fate, suggesting an important role for RHAMM in maintenance of hESC pluripotency that is consistent with its observed downregulation during hESC differentiation.
RHAMM is also known to be important for spindle stability and mitotic progression through the G2/M phase of the cell cycle, and its deletion results in cell cycle arrest in G2/M phase but does not impede the cells from progressing through S phase [13, 30, –32]. In accordance with this, we found that RHAMM knockdown in hESC did not alter progression through S phase, as shown by BrdU incorporation, but significantly, very few cells entered G2/M phase. RHAMM has been shown to be involved in motility and invasiveness of tumorigenesis and metastasis by influencing cell proliferation, and its blockade can be a potential anticancer therapeutic target [29, 61]. Apoptosis dependent on tumor suppressor gene p53 can be related to upregulation of certain apoptotic targets, but evidence now reports that it can also be explained by transcriptional repression of genes causing growth arrest, and RHAMM was found to be one of the transcriptionally downregulated genes, with up to 3.85-fold reduction in its expression [62, 63]. Increased apoptosis has also been observed when a 15-fold reduction in RHAMM expression was noted in malignant cells by anticancer treatment . This corroborates with our results showing a dramatic increase in apoptosis when RHAMM was knocked down in hESC. The induced apoptosis can be explained by the effect of RHAMM suppression on mitosis [13, 30]. It is likely that cell cycle arrest in G2/M phase precedes the apoptosis seen after RHAMM knockdown.
This study has shown convincingly that HAS2 expression is significantly increased during hESC differentiation. Using RNAi, we were able to downregulate HAS2 expression, and this resulted in a significant downregulation of all markers studied, confirming that differentiation to extraembryonic lineages (trophectoderm, primitive ectoderm, and primitive endoderm), as well as three germ lineages (mesoderm, endoderm, and ectoderm), was suppressed upon HAS2 downregulation. These data suggest an important role for HAS2 in hESC differentiation and merit additional studies to unravel the mechanisms and signaling pathways that are likely to be affected by HAS2. Most importantly, suppression of differentiation can be exploited for large-scale culture of hESC by adding specific HAS2 inhibitors to the culture media.
HA has been suggested to have an important role in cardiogenesis during embryo development [28, 45, 64, , –67] and during differentiation of mESC , which prompted us to investigate the role of exogenous HA in hESC differentiation to cardiomyocytes. It is known that glycosaminoglycan in itself, because of its physicochemical properties, can influence cell proliferation and differentiation [68, 69], and hence to study HA-specific role in differentiation of hESC, we deemed it essential to include another GAG as a second control apart from the control hESC grown in normal differentiating media. We used chondroitin sulfate (CS) as the GAG control with the same final concentration in the normal differentiating media as the HA (0.75 mg/ml) to compare the effects on differentiation of hESC toward cardiac lineage. Assessing the cumulative percentage of contractile EBs over a period of days after EB plating is a documented method to assess the function of hESC to differentiate into cardiomyocytes . We found that there was a significant increase in the number of contractile EBs grown in HA media compared with the control and, surprisingly, noted that adding CS in culture media markedly reduced the contractile ability of EBs. These results were also confirmed by real-time RT-PCR analysis that indicated upregulation of the mesodermal marker BRACHYURY as well as cardiac markers (NKX2.5, ANP, GATA4, cTnT, and MLC2A). In view of these results, we concluded that HA enhances cardiogenic potential of hESC. Future work, however, is necessary to understand whether this is an effect of HA on mesodermal differentiation or is a more general mechanism. Whatever mechanism is involved, it should still be possible to exploit the data derived from this work to enhance cardiomyocyte differentiation from hESC.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.
We are grateful to Sun Yung and Dennis Kirk for technical assistance and also to the patients at Newcastle Fertility Centre at Life who donated their embryos for the stem cell research. We are thankful to Dr. Volker Assmann, London, U.K., for donation of RHAMM antibody and to Dr. Paraskevi Heldin, Uppsala, Sweden, for HAS2 antibody. This work was supported by funding from Medical Research Council (Grant G0301182), One NorthEast Regional Development Agency, and Newcastle upon Tyne Hospitals NHS Trust. P.S. is currently affiliated with the Centro de Investigación Príncipe Felipe, Valencia, Spain.