IGF2BP1 Expression in Human Mesenchymal Stem Cells Significantly Affects Their Proliferation and Is Under the Epigenetic Control of TET1/2 Demethylases

Abstract

Mesenchymal stem cells (MSCs) are a population of cells harboring in many tissues with the ability to differentiate toward many different lineages. Unraveling the molecular profile of MSCs is of great importance due to the fact that these cells are very often used in preclinical and clinical studies. We have previously reported the expression of insulin-like growth factor 2 mRNA binding protein 1 (IGF2BP1) an oncofetal mRNA-binding protein—in different stem cell types such as bone marrow (BM)-MSC and umbilical cord blood (UCB)-hematopoietic stem cells. Here, we demonstrate that MSCs of adipose tissue, BM, and UC origin have a differential pattern of IGF2BP1 and ten-eleven-translocate 1/2 (TET1/2) expression that could correlate with their proliferation potential. Upon IGF2BP1 interference, a significant reduction of cell proliferation is observed, accompanied by reduced expression of c-MYC and GLI1 and increased p21. We also present, for the first time, evidence that IGF2BP1 is epigenetically regulated by TET1 and TET2 demethylases. Specifically, we show that TET1 directly binds to the promoter of IGF2BP1 gene and affects the hydroxymethylation status of its promoter. These results indicate that IGF2BP1 and TET1/2 contribute to the stemness of MSCs, at least regarding their proliferative potential.

Introduction

Mesenchymal stem cells (MSCs) were initially discovered by Friedenstein in the 1970s in the bone marrow (BM) as cells with clonogenic potential that could differentiate toward osteocytes under the appropriate stimuli [1]. Since then, great progress has been made and MSCs have been found in almost all tissues, including adipose tissue (AT), umbilical cord (UC), and umbilical cord blood (UCB) [2] while there is also evidence that they differentiate toward various lineages, such as neuroectoderm and epithelium [3]. Unraveling the molecular profile and understanding the physiology of these cells is essential, as MSCs are very often used in cell therapy and the selection of the most appropriate population for each application is pivotal.

We have previously reported the expression of the insulin-like growth factor 2 mRNA-binding protein 1 (IGF2BP1) in different stem cell populations, specifically BM-MSC and UCB-hematopoietic stem cells (UCB-HSC) [4,5]. IGF2BP1 is a member of a highly conserved gene family of RNA-binding proteins. It was first discovered by its ability to specifically bind to c-MYC mRNA and protect it from endonucleolytic cleavage, thus increasing its half-life [6]. Multiple studies have established an important role for IGF2BP1 in embryo development. Knockdown of the gene caused high levels of perinatal death (60%), while the animals that were finally born, were significantly smaller with hypoplastic tissues in almost every organ [7]. It was shown that IGF2BP1 and its homologues in chicken or Xenopus, bind to various mRNAs affecting their half-life, their translatability, and their subcellular localization. Among these targets are c-MYC [8], IGFII-leader3′ [9], H19 [10], tau, β-actin [11], and GLI1 [12]. All these genes have a key role in embryo development, in tissue-specific differentiation or even in cellular proliferation, differentiation, and motility. We have previously shown that this gene is under epigenetic control in BM-CD34⁺ isolated progenitors [4].

Methylation is a major mechanism through which the spatiotemporal pattern of expression of genes involved in cell development and differentiation is controlled. Defects in methylation patterns change the molecular profile and produce aberrant phenotypes in embryonic stem cells (ESCs) and embryos by causing loss of lineage restriction and transdifferentiation toward the extraembryonic trophoblast lineage [13]. It is noteworthy that the degree of methylation at the promoter loci of these “embryonic” genes in MSCs is greater than that in ESCs but less in fibroblasts, which suggests that the biology of MSCs as a group may be epigenetically restricted [14].

Recently, the ten-eleven-translocate (TET) family of proteins was identified as a new family of enzymes that alter the methylation status of DNA [15,16] and, therefore, may have an impact on the expression of genes crucial to pluripotency or cellular differentiation. As demonstrated by Ito et al., mouse ESC (mESC) express only Tet1 and Tet2 and knocking down of Tet1, resulted in significant reduction of Nanog expression in preimplantation embryos, diminished ESC proliferation, and overt ESC differentiation [17]. Another report by Koh et al. states that Tet1 depletion in mESC correlated with decreased expression of the Nodal antagonists Lefty1 and Lefty2, increased expression of the mesoderm/endoderm transcription factors Brachyury and Foxa2, and upregulated expression of the trophoectoderm transcription factors Cdx2, Eomes, and Elf5, which all underlie the role of TET proteins as key regulators of early embryonic development and differentiation that control lineage determination at critical decision points [18].

On the other hand, various studies reported that TET2 loss leads to increased replicating capacity in vitro and enhanced HSC function in vivo, suggesting that TET2 restrains aberrant self-renewal and expansion of HSC [19,20]. Thus, the TET family of proteins could influence a plethora of genes and play different roles in many cell populations at different time points.

In this study, we wanted to gain an insight into the role of IGF2BP1 and TET demethylases in human MSCs. Our data demonstrate coordinated regulation of IGF2BP1 by TET1 and TET2 suggesting a possible role for all these genes in the physiology of MSCs.

Materials and Methods

MSC isolation and culture

Isolation of MSCs from any source was performed within 24 h and under sterile conditions. All donors were informed and had signed a consent form. All procedures had prior approval from the Institutional Review Board of Saint Savas Cancer Hospital and from the Research Ethics Committee of the University of Athens. Furthermore, at the time of sample collection, all donors were disease free.

BM-MSCs were isolated as previously described [21]. AT-MSCs were isolated according to Zuk et al. [22]. Briefly, 5–10 mL of processed lipoaspirate were washed thrice using phosphate-buffered saline (PBS)-Dulbecco (Biochrom AG, Berlin, Germany, www.biochrom.de/), followed by incubation in 1% collagenase type IV (0.075% w/v) (Sigma-Aldrich Co. LLC., St. Louis, MO, www.sigmaaldrich.com) for 60 min at 37°C, under shaking. Upon centrifugation at 1,200 g for 10 min, the pelleted stromal vascular fraction was resuspended in complete medium-10% fetal bovine serum (Biosera Ltd, East Sussex, United Kingdom, www.biosera.com/), α-MEM with GlutaMAX™ (Gibco, by Life Technologies Invitrogen, Paisley, United Kingdom, www.invitrogen.com/site/us/en/home/brands/Gibco.html), and 1×antibiotic-antimycotic (Gibco)—and plated in a T25 culture flask (Greiner Bio-One GmbH, Frickenhausen, Germany, www.greinerbioone.com/en/start/) for 24 h after which nonadherent cells were removed.

UC-MSC isolation was as follows: received UC was rinsed in PBS-Dulbecco (Biochrom AG, Berlin, Germany) with 1× antibiotic-antimycotic (Gibco). A hematoma-free piece of 5 cm was dissected and the blood vessels were removed. The remaining tissue was cut into small pieces and plated in a T175 culture flask (Greiner Bio-One GmbH) in complete medium as explants.

MSC culture conditions were similar, independent of tissue origin. Two to 3 weeks upon initial plating (depending upon the colonies observed), cells were trypsinised using 0.25% trypsin-ethylenediaminetetraacetic acid (EDTA) (1×; Gibco) and re-plated in complete medium with a cell density of 250/cm². Medium was replenished twice a week and further passage was performed when cells reached 80% confluence.

Phenotypic and functional characterization

Calculation of doubling time (T _D)

The population doubling time was calculated as previously described [5] using the following equation: T _D=ln2×Dt/(lnN _t−lnN ₀), where Dt is the time for which the culture was allowed to grow and N ₀ is the inoculum cell number while N _t is the cell harvest number.

Calculation of population doublings

The population doubling (PD) was calculated by the following equation: PD=(logN _t−logN ₀)/log2, where N ₀ is the inoculum cell number and N _t is the cell harvest number.

Calculation of growth index and cumulative GI

The growth index (GI) was calculated using the following equation: GI=N _t/N ₀, where N ₀ is the inoculum cell number and N _t is the cell harvest number. The cumulative GI (cGI) was calculated using the following equation: cGI=GIn×GI(n−1), where n is the passage number.

Surface staining

Phenotypic characterization of MSCs took place in every other passage. MSCs were stained, fixed, and analyzed, as previously described [5] on a FACSCalibur Flow Cytometer (BD Biosciences, San Jose, CA, www.bdbiosciences.com/) using CellQuest software or FlowJo 7.8. All anti-human antibodies used were purchased from BD Pharmingen (San Jose, CA): CD73 PE, CD45 PerCP, CD106 FITC, CD44 PE, CD29 APC, CD14 PE, HLA-class I FITC, and HLA-DR PE; with the exception of CD34-FITC (Immunotech, Marseille, France, www.immunotech.cz/en/company_en.htm), CD105-FITC, C-MYC-PE (AbD Serotec, Oxford, United Kingdom, www.abdserotec.com/) and CD90 FITC (eBioscience, San Diego, CA, www.ebioscience.com/).

In vitro differentiation

Adipogenic and osteogenic lineage differentiation assays were performed on passages 2 and 4, as previously described [21]. Differentiation was assessed by OilRedO staining and Alizarin Red S staining, as previously described [5], for adipogenic and osteogenic differentiation respectively. All chemicals used in both stains were purchased from Sigma-Aldrich Co.

RNA isolation and quantitative RT-PCR analysis

Total RNA was extracted from MSCs using either the Nucleospin RNA II kit (Macherey-Nagel GmbH&Co, Dueren, Germany, www.mn-net.com/) or RNAqueous^®-Micro Kit (Ambion, Inc., Austin, TX, www.invitrogen.com/site/us/en/home/brands/ambion.html), depending on cell number, according to the manufacturer's instructions. rDNase treatment was performed to remove any residual genomic DNA. Total RNA (1 μg) was reverse transcribed using ImProm-II Reverse Transcriptase (Promega, Madison, WI, www.promega.com/) and random hexamer oligonucleotides (Invitrogen).

Quantitative RT-PCR (qPCR) was performed on a Rotor-Gene 3000 (Corbett Research, Sydney Australia, www.corbettlifescience.com/). Two hundred nanograms of cDNA solution, KAPA SybrFast qPCR kit (KapaBiosystems, Boston, MA, www.kapabiosystems.com/), and 200 nM of primers were used in a 25 μL reaction. PCR conditions were 95°C for 10 min, with 40 cycles of 95°C for 20 s, 59°C–65°C for 20 s depending on primer pair, and 72°C for 20 s. To confirm amplification specificity, a melting curve analysis was performed at the end of each cycle. Primer pairs used are listed in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/scd).

Furthermore, the ΔΔC_t method was used for data analysis of gene expression levels. GAPDH mRNA levels were used for internal normalization. NANOG levels were normalized to MCF-7, a breast cell line overexpressing embryonic markers. In addition, IGF2BP1, its targets, cell cycle genes, TET1 and TET2 levels were normalized to K562, a leukemic cell line.

RNA interference

Small RNA interference (siRNA) transfection experiments were performed using siIMPORTER™ (Upstate, Milton Keynes, United Kingdom, Upstate, www.millipore.com/company/cp1/redirect-ab) according to the manufacturer's instructions. Two different siRNA molecules were designed and tested against each target molecule and the most effective was chosen. The chemically synthesized 21 nucleotide-long double-stranded RNAs, directed against IGF2BP1, TET1, and TET2 mRNAs used in this study are listed in Supplementary Table S2.

Control-Alexa Fluor 488-labeled siRNA was used to monitor the cellular uptake of each experiment (Qiagen Ltd, West Sussex, United Kingdom, www.qiagen.com/default.aspx), by fluorescence-activated cell sorting (FACS) analysis, 24 h post-transfection. Negative control siRNA (Ambion, Inc.) was also included to assess any off-target effects. Cells were harvested at various time points after siRNA transfection depending on the assay performed. All si experiments were performed in MSC cultures at passages 2 or 3.

Western blotting

Whole-cell lysates were prepared from cells transfected with control or IGF2BP1 siRNAs for 24, 48, and 72 h. Cells were lysed in ice-cold RIPA buffer containing protease inhibitors. Protein concentration of each lysate was determined using a DC protein assay (Bio-Rad, Hercules, CA). Ten micrograms of total protein of each sample was analyzed in a NuPage 4%–12% Bis-Tris acrylamide gel and transferred to a PVDF membrane. Western blotting was performed using the following primary antibodies: IGF2BP1 (sc-166344) and β-TUBULIN (sc-5274) from Santa Cruz Biotechnology, Inc. (Dallas, TX, www.scbt.com/). The secondary antibody used was anti-mouse IgG-HRP (P0447) from Dako (Glostrup, Denmark, www.dako.com/). Super Signal West Pico Chemiluminescent Substrate (Catalog No. 34077; Thermo, Inc., Fisher Scientific, Rockford, IL, www.thermofisher.com) was used to develop blot signals.

Immunocytochemistry

MSCs were cultured in Lab-Tek slides (Nalge NUNc International, Rochester, NY, www.nuncbrand.com), fixed with acetone solution, air dried and hydrated in a series of graded alcohol to water. Automated immunohistochemistry validation was performed using the automated Leica Bond system. Antigen retrieval was performed using Bond Epitope Retrieval Solution 2. Slides were then incubated with TET1 Ab (GeneTex, GTX627420) and TET2 (GeneTex, GTX124205), diluted to a concentration of 1:100). Slides were counterstained and mounted for light microscopy analysis.

Proliferation assays

Thymidine incorporation

MSCs were seeded at a density of 6,000 cells/cm², in 96-well plates (Greiner Bio-One GmbH) 24 h prior to siRNA transfection. After 48 h of incubation, cells were pulsed with 1 μCi/well [methyl-³H] thymidine (specific activity: 53.0 Ci/mmol, 1 mCi/mL; Amersham, GE Healthcare, Amersham, United Kingdom, www3.gehealthcare.com/en/Global_Gateway) for an additional 18 h. Cells were harvested, and counted on a liquid scintillation counter (Wallace Trilux, No. 1450 Microbeta; PerkinElmer Life Sciences, Waltham, MA, www.labx.com/v2/adsearch/detail3.cfm?adnumb=480708).

Colony forming units-fibroblast assay

To evaluate the self-renewing capacity of MSCs, 150 cells/cm² were seeded at passage 2 onto a six-well plate (Greiner Bio-One GmbH) in complete medium. At day 14, the medium was removed; cells were washed twice with PBS and stained with 0.5% crystal violet (Merck &Co, Inc., Whitehouse Station, NJ, www.merck.com/index.html) in methanol. Colonies larger than 50 cells were counted under a stereoscopic microscope. All experiments were performed in triplicates.

Hydroxymethylated DNA immunoprecipitation

Initially the promoter of the IGF2BP1 gene was retrieved using the Ensembl Genome Browser, www.ensembl.org/index.html, and analyzed for its CpG content using http://genome.ucsc.edu/cgi-bin/hgBlat?command=start. Two regions of interest around 1 kb upstream of the transcription start site (TSS) were identified (Fig. 6A). Primers specific to these regions were designed (Supplementary Table S3).

The hMeDIP kit (Cat No. AF-110-0016) and the MeDIP kit (Cat No. mc-green-001) from Diagenode (Seraing, Belgium, www.diagenode.com/en/index.php) were used to perform immunoprecipitation of hydroxymethylated and methylated DNA respectively, according to the manufacturer's instruction. Briefly, genomic DNA was extracted using QIAamp DNA Mini Kit from Qiagen Ltd (Catalog No. 51304). The DNA was then disrupted into random fragments using a bench-top sonicator (Ultrasonic, BRANSON200). One microgram of fragmented DNA was denatured at 95°C for 10 min, followed by overnight incubation at 4°C, with the specific antibody and beads. The next day beads were washed and resuspended in the corresponding buffer. The immunoprecipitated (IP) DNA was eluted and analyzed by qPCR.

Chromatin immunoprecipitation

For the chromatin immunoprecipitation (ChIP) assay, the SimpleChIP^® Enzymatic Chromatin IP Kit (#9003; Cell Signalling, Danvers, MA) was used, according to the manufacturer's instructions. Briefly, 5×10⁶ UC-MSCs were fixed with formaldehyde, to “preserve” the protein-DNA interactions. Cells were then lysed and chromatin was harvested and fragmented using a Dounce homogenizer and a bench-top sonicator (Ultrasonic, BRANSON200). The chromatin was then subjected to immunoprecipitation using TET1-specific Ab (GeneTex, GTX627420). DNA sequences associated with the protein of interest co-precipitated as part of the cross-linked chromatin complex and the relative amount of that DNA sequence was enriched by the immunoselection process. After immunoprecipitation, the protein-DNA cross-links were reversed and the DNA was purified. qPCR was used to detect the enrichment of the sequence of interest. The Histone H3 Rabbit mAb serves as a positive control IP for almost any locus studied and Normal Rabbit IgG serves as a negative control IP. In addition, ribosomal protein L30 (RPL30) primers were used as positive control.

Statistics

The Student's t-test with a 95% confidence interval was performed using GraphPad Prism version 5.00 for Windows, GraphPad 5 Software (www.graphpad.com). Paired analyses were performed in all knockdown experiments. The level of significance was set at 0.05 (*P<0.05, **P<0.01, and ***P<0.001).

Results

IGF2BP1 and TET1/2 expression in MSCs of different origin

IGF2BP1 expression has been previously detected by our group in HSC and BM-MSCs, with indications of epigenetic regulation [4,5]. However, the normal role of this protein in stem cells has not yet been described. Recent publications assign to TET1 and TET2 DNA demethylases an important role in ESC physiology [17,18,23] by demethylating the Nanog gene promoter [17].

To explore the role of IGF2BP1 and TET1/2 in MSCs, we initially tested populations of fetal and adult origin for their expression pattern. All cell preparations met the phenotypic and functional criteria for MSC definition (Supplementary Table S4). As shown in Figure 1, IGF2BP1, TET1, and TET2 (A, B, and C respectively) mRNA levels of expression were considerably higher in UC-MSCs, as compared to both adult, BM, and AT. These differences in mRNA were also reflected in the corresponding protein levels (Supplementary Fig. S1).

FIG. 1.

IGF2BP1 (A), TET1 (B), and TET2 (C) levels of expression in MSC of adult and fetal origin. mRNA levels were estimated by qPCR in MSC populations of different origin, harvested at passages 2–3. UC-MSCs exhibit the highest levels for all the three genes tested. (○) Represent individual values from different samples of each source. BM, bone marrow n=4; AT, adipose tissue n=5; UC, umbilical cord; IGF2BP1, insulin-like growth factor 2 mRNA binding protein 1; TET1/2, ten-eleven-translocate 1/2 (*P<0.05, **P<0.01, and ***P<0.001). MSC, mesenchymal stem cell; qPCR, quantitative RT-PCR.

In line with other reports, we found that adult MSCs have a lower in vitro proliferative capacity compared with UC-derived MSCs [24] (Fig. 2A). It is also known that in vitro changes in MSC physiology, including their proliferative potential, can be attributed to changes in their gene expression pattern [25]. In this point we should underline that the different conditions used in cell isolation of each source, do not allow direct comparisons of the in vitro proliferative history spanning from seeding to first passage. Therefore, we decided to investigate the effect of the number of population doublings (PD)—as a measure of the proliferation rate—on gene expression within the same source, that is, the UC-derived MSCs that possess the highest proliferative capacity. As seen in Figure 2B, initially UC-MSCs proliferate fast [passages 1–3, with a median doubling time (T _D) of 1.46 days] whereas after passage 6 they start to slowdown (median T _D=5.11 days). To this end, we compared gene expression of UC-MSCs at passages 1–2 (PD estimated from P0=5–15) versus passages >7 (PD>30) (Fig. 2B). Indeed, the number of PD significantly affected IGF2BP1 mRNA levels, although never reaching the levels observed in adult MSCs (Fig. 2C). TET1 expression was significantly decreased to levels comparable to adult MSCs (Fig. 2D), while no significant decrease was observed in TET2 expression in relation to UC-MSC PDL (Fig. 2E). Altogether, these results are an indication that IGF2BP1 might be implicated in the regulation of UC-MSC proliferation.

FIG. 2.

Proliferative capacity of human MSCs and expression of IGF2BP1, TET1, and TET2. (A) The avg T _D of adult MSCs is longer than that of UC-MSCs. Avg T _D is depicted as median values; (○) represent values from different samples of each source. BM n=8; AT n=5; (B) cGI of UC-MSCs during their culture, n=6. Data represent mean values±SEM at different passages. Squares in the graph indicate the populations analyzed in graphs (C–E), in relation to their PD (5–15 vs. >30 respectively). IGF2BP1 (C), TET1 (D), and TET2 (E) mRNA levels of expression significantly vary between UC-MSCs that have undergone different number of PD. Bars represent mean±SEM, Avg T _D, average doubling time; PD, population doublings (**P<0.01, and ***P<0.001). cGI, cumulative growth index.

Inhibition of IGF2BP1 expression in UC-MSCs

To further investigate whether IGF2BP1 is involved in the regulation of UC-MSC proliferation, we applied RNA interference. Initially, we determined the efficiency of the assay at different time points. As depicted in Figure 3A, IGF2BP1 mRNA levels significantly reduced compared with control cultures for 48 h post-transfection and then gradually increased, but still remained at lower levels compared with control cultures 72 h post-transfection. This reduction in the mRNA levels was further confirmed at the protein level by western blot analysis (Fig. 3B).

FIG. 3.

siRNA mediated reduction of IGF2BP1 levels of expression in human UC-MSCs. (A) IGF2BP1 gene knockdown in UC-MSC cultures at different time points. (○) Represent values from individual samples. The % reduction at different time points is depicted as median values. (B) Western blot analysis 48 and 72 h upon RNA interference. β-Tubulin protein levels were used as loading control. (C) Effect of IGF2BP1 knockdown in UC-MSCs on established mRNA targets, 48–72 h post-transfection. Bars represent mean±SEM (D) Significant decrease of c-MYC protein levels upon IGF2BP1, knockdown. (Di, ii) Are representative plots of flow cytometry analysis of c-MYC expression upon siControl or siIGF2BP1 treatment of UC-MSCs, respectively. (Diii) Bars represent fold difference compared to isotype control values (mean±SEM) from three independent experiments. MFI, mean fluorescence intensity (*P<0.05, **P<0.01, and ***P<0.001).

Then, we examined the expression of certain genes previously reported to be targets of IGF2BP1 in other cell types, to verify whether the reduction of IGF2BP1 could result in a biological effect [12,26]. As shown in Figure 3C, c-MYC, β-ACTIN, and GLI1 mRNA levels were reduced and IGFII-leader3 mRNA levels were increased. The effect of IGF2BP1 knockdown was further controlled at the protein level of c-MYC, as its mRNA and protein molecules have a short half-life. Indeed, a significant reduction of c-MYC mean fluorescent intensity values—which represent the antigen intensity per cell—was observed (Fig. 3D).

Having established that RNA interference has the potential to elicit a biological effect, we further tested the outcome of IGF2BP1 interference on MSC proliferative capacity by ³H-TdR incorporation and colony forming units-fibroblast formation. Indeed, decreased IGF2BP1 expression caused a significant reduction in ³H-thymidine incorporation in cultures of UC-MSCs (Fig. 4A). In addition, a decrease in colony formation was observed (Fig. 4B and Supplementary Fig. S2).

FIG. 4.

Effect of IGF2BP1 inhibition on UC-MSC proliferation. (A) The proliferative capacity of MSCs upon iRNA was reduced as illustrated by ³H-TdR incorporation. Each experiment was performed in triplicates and (○) represent the average of a triplicate (B) IGF2BP1 gene knockdown affected UC-MSCs in vitro self-renewal as demonstrated by reduced colony number in a CFU-f assay. Each experiment was performed in triplicates and (○) represent the average of a triplicate. (C) Cell cycle arrest was accompanied by elevation of p21 mRNA levels. (○) Represent individual values of different samples treated with either control or IGF2BP1 siRNA. iRNA, RNA interference; CFU-f, colony forming unit-fibroblasts (**P<0.01). siRNA, small RNA interference.

We have previously shown that inhibition of proliferation in MCF-7 breast cancer cells after IGF2BP1 knockdown was related to reduction of c-MYC and elevation of cyclin-dependent kinase inhibitor 1A (p21) [4]. Bearing this in mind, siRNA-treated UC-MSC cultures were also tested for p21 expression. As expected, the mRNA levels of this cell cycle regulator were increased between 48 and 72 h in all samples tested (Fig. 4C).

Similar results after IGF2BP1 knocking down were also obtained in adult MSCs (Supplementary Fig. S3A). Furthermore, UC-MSCs at passage >7, exhibiting lower IGF2BP1 levels compared with their counterparts at passages 1–2, also expressed decreased GLI1 and p21 levels (Supplementary Fig. S3B).

To ascertain that these effects on proliferation were not due to apoptosis, we tested MSC cultures with Annexin-V/7AAD staining. No difference was observed in the percent of positively staining cells between control and siRNA-treated cultures (data not shown).

Taken altogether, these findings could point to the possible involvement of IGF2BP1 in MSC proliferative capacity.

TET1 and TET2 downregulation in UC-MSCs

To test our hypothesis that IGF2BP1 may be a downstream target of TET1 or TET2 demethylases, we knocked down each one of the two demethylases and evaluated any changes on IGF2BP1 mRNA levels by qPCR. As shown in Supplementary Figure S4A and D, TET1 and TET2 expression was significantly reduced, 48 h post-tranfection. Gene expression changes were also accompanied by decrease in protein expression levels (Supplementary Fig. S4B, E).

Prior to analyzing the relation of IGF2BP1 with TET proteins, we tested whether NANOG gene expression is affected by TET1 downregulation in UC-MSCs, as initially reported by Ito et al. in mESC [17]. Indeed, MSCs treated with siTET1 exhibited a significant reduction in NANOG expression (Supplementary Fig. S4C), while siTET2 resulted in elevation or no alteration of NANOG expression (Supplementary Fig. S4F).

To analyze the effect of TET1 or TET2 interference on IGF2BP1 and on MSC proliferation, we performed a series of experiments in UC-MSCs. As depicted in Figure 5B and C respectively, there was a significant reduction in IGF2BP1 mRNA levels upon exposure to siTET1, while a milder reduction was observed upon TET2 siRNA treatment. These findings indicate for the first time, to our knowledge, that IGF2BP1 gene could be a direct target of TET1.

FIG. 5.

IGF2BP1, TET1, or TET2 interference and proliferation of UC-MSCs. Effect of IGF2BP1 (A), TET1 (B), or TET2 (C) interference on proliferation and the c-Myc/p21 axis. Effect of TET1 (D) and TET2 (E) interference on each other and on cell cycle regulators CDK4 and 6. Bars represent mean percent change±SEM of mRNA levels of either targeted demethylase gene (no fill), IGF2BP1 (gray filled), cell cycle genes (pattern filled) or proliferation rate (black filled) of specific-si treated cells in comparison to control-si treated cells. si, small interference RNA (*P<0.05, **P<0.01, and ***P<0.001). CDK, cyclin-dependent kinase.

We next investigated how decreased TET1 or TET2 gene expression influences UC-MSC proliferation, compared to IGF2BP1 siRNA treatment. As shown in Figure 5, there was a significant reduction in ³H-thymidine incorporation by the siRNA treated UC-MSC cultures for TET1 (Fig. 5B) and for TET2 (Fig. 5C), similar to siIGF2BP1 treatment (Fig. 5A). Subsequently, we wanted to test whether IGF2BP1 reduced levels of expression and the reduced proliferation caused by TET1 or TET2 gene interference correlate with the same molecular events.

A common pattern of events is observed between IGF2BP1 and TET1 interference (Fig. 5A, B respectively). Specifically, TET1 inhibition reduces IGF2BP1 mRNA to similar levels with those after treatment with siIGF2BP1. Levels of c-MYC are downregulated, while p21 is upregulated by both siIGF2BP1 and siTET1. On the other hand, TET2 reduction resulted in slightly increased c-MYC expression and decreased p21 expression (Fig. 5C).

Since inhibiting both, TET1 or TET2, resulted in reduced proliferation, although at varying levels, we proceeded to explore whether other molecules involved in cell cycle were differentially affected during this process. We observed an interaction between the two demethylases, since, siRNA against TET1 diminished TET2 mRNA levels and vice versa. Further exploring the molecular mechanisms for the reduced proliferation upon TET1 or TET2 interference, we found that cyclin-dependent kinases 4 and 6 (CDK4/6) mRNA levels significantly declined (Fig. 5D, E respectively).

Epigenetic status of IGF2BP1 gene in UC-MSCs

A number of studies report that only TET1 protein includes a CXXC domain, which is responsible for DNA binding, while TET2 molecule lacks this and the ability of direct binding [27].

In this context, to evaluate the epigenetic status of IGF2BP1 gene, relative levels of hydroxy-methylated IP-DNA were calculated in the genomic DNA of UC-MSCs treated with siTET1 versus UC-MSCs treated with Control si. As shown in Figure 6B, TET1 gene interference caused a significant reduction of hydroxymethylation levels in the tested CpG regions of the promoter (Fig. 6A), 1,000 bp upstream of the gene TSS. This reduction was mirrored by an increase in the methylation levels of the IGF2BP1 promoter in at least two of the rich CpG regions tested (Supplementary Fig. S5).

FIG. 6.

TET1 directly binds to IGF2BP1 and inhibits its methylation. (A) Promoter region of the IGF2BP1 gene. Arrows and numbers show the location of primers used in these analyses. CpG islands are shown in white. (B) Relative levels of hydroxymethylation in the IP-DNA were evaluated by qPCR, using the corresponding primer pairs indicated above, following the manufacturer's instructions. Bars represent control si (no fill) and siTET1 (gray filled). (C) Signal of TET1-IP DNA relative to the total amount of input DNA, analyzed by qPCR using the specific primer pairs indicated in (A). RPL30 specific primers were used as a positive control of the PCR reaction for the IP-DNA. Bars represent the results from the positive control Histone3-specific antibody (no fill) and TET1 specific antibody (gray filled). IP, immunoprecipitated; RPL30, ribosomal protein L30; TSS, transcription start site.

To further establish that TET1 controls IGF2BP1 methylation status by direct binding to its gene promoter, we performed ChIP analysis using a TET1-specific antibody (Fig. 6C). We detected significant enrichment of the IGF2BP1 promoter DNA sequences of all loci examined above for their hydroxymethylation status.

Discussion

ESC have been extensively investigated and are always used as a reference population for the deciphering of pathways involved in development and differentiation. Their characteristics comprise what is referred to as “stem” signature. A lot of scientific interest has focused on determining differences and similarities among stem cell populations, with MSCs being one of the most studied. They can be found in a multitude of tissues, other than the originally defined source of the BM [28]. In this study, we demonstrate the differential expression of important embryonic genes: IGF2BP1 and the TET1 and TET2 demethylases in human MSCs from different sources. We further look into the role of these genes and their possible interaction in regulating cellular proliferation.

IGF2BP1 is an mRNA-binding protein with a fetal pattern of expression [26]. We have previously reported that IGF2BP1 is expressed in stem cells with high renewal capacity like human UCB-HSC and mouse BM-MSCs [4,5]. Furthermore, we [4] and others [29] have assigned a role to IGF2BP1 in the control of cell proliferation in malignant cells.

In this study, we show that MSCs of fetal origin, UC-derived, which proliferate in vitro faster than their adult counterparts from BM and AT, also express significantly higher levels of IGF2BP1. Furthermore, UC-MSCs, which have undergone a limited number of PDL in vitro, also express higher levels of IGF2BP1 compared with their in vitro exhausted counterparts. This observation was further substantiated when we knocked down this gene and monitored MSC proliferation rate. Moreover, the significant reduction in IGF2BP1 expression resulted in a considerable reduction in the levels of c-MYC, GLI1, and p21. It is well established that c-MYC is a key regulator of cell cycle proliferation in almost all cell types [30]. Furthermore, a number of studies have reported c-MYC to be one of the major regulators of stem physiology [31], with prominent role in the production of induced pluripotent stem (iPS) cells, where the quantity and the exact time point of c-MYC transient expression are detrimental for the number of the iPS cells produced [32].

The GLI transcription factors—effectors of Hedgehog (Hh) signaling—are involved in cell fate determination, proliferation, and patterning in many cell types and most organs during embryo development, and in regulating stem cell maintenance and regeneration [33]. Moreover, GLI1 has been reported to be involved in the regulation of HSC proliferation [34]. Noubissi et al. [12] identified a novel mechanism, in colorectal tumorigenesis, by which Wnt signaling regulates the transcriptional outcome of Hh signaling pathway, via GLI1 mRNA stabilization by the RNA-binding protein IGF2BP1. Our data suggest that this mechanism might also be active in normal MSCs, partly to regulate their proliferation.

We have previously reported that knocking down of IGF2BP1 in MCF-7 breast cancer cells resulted in growth inhibition through c-MYC reduction and induction of p21 [4], a negative regulator of cell cycle, similarly to our current observations with MSCs.

In our previous work on IGF2BP1 expression in HSC, we showed that treatment of BM-CD34⁺ cells with 5-azacytidine resulted in de novo activation of the gene [4]. Additionally, it was recently reported that Tet1 and 2 demethylases play an important role in epigenetic control in mESC physiology [17,18]. These observations prompted us to investigate the possibility that these TET family members might also regulate IGF2BP1 expression through its methylation status in UC-MSCs.

Since no data exist regarding the pattern of expression of TET in MSC populations, we initially tested MSCs from all sources for TET1/2 expression. Thus, we report here for the first time, to our knowledge, that both these DNA demethylases are expressed in all samples tested, irrespective of their tissue of origin. However, it is worth mentioning that TET1 and TET2 are significantly higher in UC-MSCs, similar to IGF2BP1.

To further delineate the role of TET1/2 proteins, we studied UC-MSCs upon gene knockdown. The data presented herein show that TET1 and TET2 knockdown significantly affected the proliferative capacity of UC-MSCs, though to a different degree. This observation is in accordance with the studies by Koh et al. [18] who report that the effects of Tet1 depletion in mESCs were dominant over an often milder effect of Tet2 depletion. TET1 knocking down led to significantly reduced proliferation by increased p21 levels and a concomitant reduction in c-MYC and GLI1 levels, similar to IGF2BP1 knockdown. A weaker, though significant, reduction in cell proliferation was also observed upon TET2 silencing despite a reduction in p21 levels and an elevation in c-MYC levels. On the other hand, TET1 and TET2 depletion caused a significant reduction in the levels of CDK4/6, key regulators of cell cycle. Thus, TET1 affects both, c-MYC/p21 and CDK4/6 pathways, explaining the more robust inhibition of cell proliferation upon its silencing, whereas the milder effects observed with TET2 targeting may be explained by its ability to downregulate only CDK4/6.

Moreover, TET1 or TET2 inhibition resulted in a significant reduction in the mRNA of IGF2BP1. However, TET1 knockdown brought about a nearly 2-fold higher reduction in IGF2BP1 levels than did TET2. These findings could suggest that IGF2BP1 may be directly targeted only by TET1; a hypothesis further substantiated by the direct binding of TET1 on the IGF2BP1 promoter and the analysis of its hydroxymethylation status. Furthermore, it was recently reported that TET1 and TET2 have a physical association with NANOG. However, only TET1 and NANOG co-occupy genomic loci of pluripotency and lineage commitment genes in ESC, possibly assigning a co-factor role for TET2. Among these genes, IGF2BP1 is at the third place of the top-ranked target loci [35]. Thus, our results come in accordance with the notion that TET1 affects gene expression more dramatically while TET2 acts a co-factor whose loss can be compensated, and therefore the effect of its inhibition is milder.

The data presented in this study denote that human MSCs share with embryonic tissues many common molecular regulatory pathways, including the TET1/2 DNA demethylation machinery and the post-transcriptional regulatory protein IGF2BP1 (Fig. 7). It appears that these factors are part of a highly complex molecular network comprising the specific “stem” profile of the MSC population and their crosstalk needs to be further clarified. Whether IGF2BP1 is only related to the proliferative potential of MSC, an essential feature of stem cells, or is more fundamentally implicated in the stem physiology, remains to be investigated.

FIG. 7.

Schematic summary of the results obtained in this study. (A) IGF2BP1 inhibition in MSCs results in decreased proliferation through decreased c-MYC and GLI1 expression and upregulation of p21. (B) TET1 interferance results in decreased IGF2BP1 expression leading to the downstream effects observed with direct IGF2BP1 knocking down. TET1 interferance induces inhibition of proliferation also through the downregulation of CDK4/6. TET2 seems to exert its role through interaction with TET1. (C) TET1 epigenetically controls IGF2BP1 expression through direct binding to its promoter. TET2 may act as a cofactor along with TET1.

Footnotes

Acknowledgments

We would like to thank E. Anastasopoulou for her assistance in flow cytometry and N. Cacoullos for her help in editing this article. We acknowledge a donation from OPAP SA to M.P.

Author Disclosure Statement

The authors indicate no potential conflicts of interest.

References

Friedenstein

, Gorskaja

and Kulagina

. (1976). Fibroblast precursors in normal and irradiated mouse hematopoietic organs. Exp Hematol, 4:267–274.

Kern

, Eichler

, Stoeve

, Kluter

and Bieback

. (2006). Comparative analysis of mesenchymal stem cells from bone marrow, umbilical cord blood, or adipose tissue. Stem Cells, 24:1294–1301.

Phinney

and Prockop

. (2007). Concise review: mesenchymal stem/multipotent stromal cells: the state of transdifferentiation and modes of tissue repair—current views. Stem Cells, 25:2896–2902.

Ioannidis

, Mahaira

, Perez

, Gritzapis

, Sotiropoulou

, Kavalakis

, Antsaklis

, Baxevanis

and Papamichail

. (2005). CRD-BP/IMP1 expression characterizes cord blood CD34+ stem cells and affects c-myc and IGF-II expression in MCF-7 cancer cells. J Biol Chem, 280:20086–20093.

Katsara

, Mahaira

, Iliopoulou

, Moustaki

, Antsaklis

, Loutradis

, Stefanidis

, Baxevanis

, Papamichail

and Perez

. (2011). Effects of donor age, gender, and in vitro cellular aging on the phenotypic, functional, and molecular characteristics of mouse bone marrow-derived mesenchymal stem cells. Stem Cells Dev, 20:1549–1561.

Bernstein

, Herrick

, Prokipcak

and Ross

. (1992). Control of c-myc mRNA half-life in vitro by a protein capable of binding to a coding region stability determinant. Genes Dev, 6:642–654.

Hansen

, Hammer

, Nielsen

, Madsen

, Dalbaeck

, Wewer

, Christiansen

and Nielsen

. (2004). Dwarfism and impaired gut development in insulin-like growth factor II mRNA-binding protein 1-deficient mice. Mol Cell Biol, 24:4448–4464.

Lemm

and Ross

. (2002). Regulation of c-myc mRNA decay by translational pausing in a coding region instability determinant. Mol Cell Biol, 22:3959–3969.

Nielsen

, Christiansen

, Lykke-Andersen

, Johnsen

, Wewer

and Nielsen

. (1999). A family of insulin-like growth factor II mRNA-binding proteins represses translation in late development. Mol Cell Biol, 19:1262–1270.

10.

Runge

, Nielsen

, Lykke-Andersen

, Wewer

and Christiansen

. (2000). H19 RNA binds four molecules of insulin-like growth factor II mRNA-binding protein. J Biol Chem, 275:29562–29569.

11.

Chao

, Patskovsky

, Patel

, Levy

, Almo

and Singer

. (2010). ZBP1 recognition of beta-actin zipcode induces RNA looping. Genes Dev, 24:148–158.

12.

Noubissi

, Goswami

, Sanek

, Kawakami

, Minamoto

, Moser

, Grinblat

and Spiegelman

. (2009). Wnt signaling stimulates transcriptional outcome of the Hedgehog pathway by stabilizing GLI1 mRNA. Cancer Res, 69:8572–8578.

13.

Feng

, Jacobsen

and Reik

. (2010). Epigenetic reprogramming in plant and animal development. Science, 330:622–627.

14.

Yannarelli

, Pacienza

, Cuniberti

, Medin

, Davies

and Keating

. (2013). The potential role of epigenetics on multipotent cell differentiation capacity of mesenchymal stromal cells. Stem Cells, 31:215–220.

15.

Iyer

, Tahiliani

, Rao

and Aravind

. (2009). Prediction of novel families of enzymes involved in oxidative and other complex modifications of bases in nucleic acids. Cell Cycle, 8:1698–1710.

16.

Tahiliani

, Koh

, Shen

, Pastor

, Bandukwala

, Brudno

, Agarwal

, Iyer

, Liu

, Aravind

and Rao

. (2009). Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science, 324:930–935.

17.

Ito

, D'Alessio

, Taranova

, Hong

, Sowers

and Zhang

. (2010). Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification. Nature, 466:1129–1133.

18.

Koh

, Yabuuchi

, Rao

, Huang

, Cunniff

, Nardone

, Laiho

, Tahiliani

, Sommer

, et al. (2011). Tet1 and Tet2 regulate 5-hydroxymethylcytosine production and cell lineage specification in mouse embryonic stem cells. Cell Stem Cell, 8:200–213.

19.

Moran-Crusio

, Reavie

, Shih

, Abdel-Wahab

, Ndiaye-Lobry

, Lobry

, Figueroa

, Vasanthakumar

, Patel

, et al. (2011). Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell, 20:11–24.

20.

, Huang

, Jankowska

, Pape

, Tahiliani

, Bandukwala

, An

, Lamperti

, Koh

, et al. (2010). Impaired hydroxylation of 5-methylcytosine in myeloid cancers with mutant TET2. Nature, 468:839–843.

21.

Sotiropoulou

, Perez

, Salagianni

, Baxevanis

and Papamichail

. (2006). Characterization of the optimal culture conditions for clinical scale production of human mesenchymal stem cells. Stem Cells, 24:462–471.

22.

Zuk

, Zhu

, Mizuno

, Huang

, Futrell

, Katz

, Benhaim

, Lorenz

and Hedrick

. (2001). Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng, 7:211–228.

23.

Freudenberg

, Ghosh

, Lackford

, Yellaboina

, Zheng

, Li

, Cuddapah

, Wade

, Hu

and Jothi

. (2012). Acute depletion of Tet1-dependent 5-hydroxymethylcytosine levels impairs LIF/Stat3 signaling and results in loss of embryonic stem cell identity. Nucleic Acids Res, 40:3364–3377.

24.

Choudhery

, Badowski

, Muise

and Harris

. (2013). Comparison of human mesenchymal stem cells derived from adipose and cord tissue. Cytotherapy, 15:330–343.

25.

Delorme

, Ringe

, Pontikoglou

, Gaillard

, Langonne

, Sensebe

, Noel

, Jorgensen

, Haupl

and Charbord

. (2009). Specific lineage-priming of bone marrow mesenchymal stem cells provides the molecular framework for their plasticity. Stem Cells, 27:1142–1151.

26.

Ioannidis

, Mahaira

, Papadopoulou

, Teixeira

, Heim

, Andersen

, Evangelou

, Dafni

, Pandis

and Trangas

. (2003). CRD-BP: a c-Myc mRNA stabilizing protein with an oncofetal pattern of expression. Anticancer Res, 23:2179–2183.

27.

Williams

, Christensen

and Helin

. (2012). DNA methylation: TET proteins-guardians of CpG islands?. EMBO Rep, 13:28–35.

28.

Orbay

, Tobita

and Mizuno

. (2012). Mesenchymal stem cells isolated from adipose and other tissues: basic biological properties and clinical applications. Stem Cells Int, 2012:461718

29.

Mongroo

, Noubissi

, Cuatrecasas

, Kalabis

, King

, Johnstone

, Bowser

, Castells

, Spiegelman

and Rustgi

. (2011). IMP-1 displays cross-talk with K-Ras and modulates colon cancer cell survival through the novel proapoptotic protein CYFIP2. Cancer Res, 71:2172–2182.

30.

Singh

and Dalton

. (2009). The cell cycle and Myc intersect with mechanisms that regulate pluripotency and reprogramming. Cell Stem Cell, 5:141–149.

31.

Satoh

, Matsumura

, Tanaka

, Ezoe

, Sugahara

, Mizuki

, Shibayama

, Ishiko

, Nakajima

and Kanakura

. (2004). Roles for c-Myc in self-renewal of hematopoietic stem cells. J Biol Chem, 279:24986–24993.

32.

Takahashi

, Tanabe

, Ohnuki

, Narita

, Ichisaka

, Tomoda

and Yamanaka

. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131:861–872.

33.

Robbins

, Fei

and Riobo

. (2012). The Hedgehog signal transduction network. Sci Signal, 5:re6

34.

Merchant

, Joseph

, Wang

, Brennan

and Matsui

. (2010). Gli1 regulates the proliferation and differentiation of HSCs and myeloid progenitors. Blood, 115:2391–2396.

35.

Costa

, Ding

, Theunissen

, Faiola

, Hore

, Shliaha

, Fidalgo

, Saunders

, Lawrence

, et al. (2013). NANOG-dependent function of TET1 and TET2 in establishment of pluripotency. Nature, 495:370–374.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.05 MB

0.08 MB

0.13 MB

0.05 MB

0.02 MB

0.12 MB

IGF2BP1 Expression in Human Mesenchymal Stem Cells Significantly Affects Their Proliferation and Is Under the Epigenetic Control of TET1/2 Demethylases

Abstract

Introduction

Materials and Methods

MSC isolation and culture

Phenotypic and functional characterization

Calculation of doubling time (T D)

Calculation of population doublings

Calculation of growth index and cumulative GI

Surface staining

In vitro differentiation

RNA isolation and quantitative RT-PCR analysis

RNA interference

Western blotting

Immunocytochemistry

Proliferation assays

Thymidine incorporation

Colony forming units-fibroblast assay

Hydroxymethylated DNA immunoprecipitation

Chromatin immunoprecipitation

Statistics

Results

IGF2BP1 and TET1/2 expression in MSCs of different origin

Inhibition of IGF2BP1 expression in UC-MSCs

TET1 and TET2 downregulation in UC-MSCs

Epigenetic status of IGF2BP1 gene in UC-MSCs

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

Supplementary Material

Calculation of doubling time (T _D)