The Chromatin Remodeling Complex CHD1 Regulates the Primitive State of Mesenchymal Stromal Cells to Control Their Stem Cell Supporting Activity

Abstract

The primitive state (stemness) of mesenchymal stromal cells (MSCs) is responsible for supporting the function of tissue-specific stem cells to regenerate damaged tissues. However, molecular mechanisms regulating the stemness of MSCs remain unknown. In this study, we found that the primitive state of MSCs is hierarchically regulated by the expression levels of the chromatin remodeling complex, CHD1, with CHD1 expression levels higher in the undifferentiated state, and decreasing upon MSC differentiation. Consistently, CHD1 expression levels decrease during progressive loss of clonogenic progenitors (CFU-F) induced by passage cultures. Moreover, knockdown (KD) of CHD1 decreased CFU-F frequency, whereas CHD1 overexpression increased it. In addition, the expression of stem cell-specific genes was down- or upregulated upon KD or overexpression of CHD1, respectively, accompanied by associated changes in chromatin condensation. Importantly, altering CHD1 expression levels affected the ability of MSCs to support the self-renewing expansion of hematopoietic stem cells (HSCs). Furthermore, CHD1 levels were significantly decreased in MSCs from acute myeloid leukemia or aplastic anemia patients, where CFU-F and HSC-supporting activities are lost. Altogether, these findings show that chromatin remodeling by CHD1 is a molecular parameter that influences the primitive state of MSCs and their stem cell-supporting activity, which controls tissue regeneration.

Introduction

Mesenchymal stromal cells (MSCs) are a non-hematopoietic adherent cell population. MSCs are derived from bone marrow (BM), adipose tissue, or placental tissue. MSCs exhibit multilineage differentiation potential toward diverse types of tissues, including bone, cartilage, and adipose tissues [1 –3]. Accumulating studies have shown that the primary mode of action for MSCs is the paracrine support of tissue regeneration both by inhibiting apoptosis and fibrosis [4], and by stimulating the regeneration of endogenous stem cells such as hematopoietic stem cells (HSCs), neuronal stem cells, and other tissue-specific endogenous stem cells [5,6]. As such, MSCs have been widely utilized for many clinical cell therapy trials. So far, more than 425 clinical trials have been performed using ex-vivo expanded MSCs to facilitate the regeneration of injured tissues [7].

However, MSCs are highly heterogeneous, and only specific subsets of primitive MSCs have the ability to form colonies (CFU-F), or to support endogenous stem cells. For example, in BM, only primitive MSC subsets that retain colony-forming potential could contribute to the stem cell niches [8,9]. Subsequent studies revealed that only the subsets of MSCs expressing nestin [10], leptin-receptor [11], or prx-1 [12], which are enriched with CFU-F, serve as major HSC-supporting niche cells. Moreover, in ex-vivo cultured MSCs, there was a positive correlation between the frequency of CFU-F and their ability to support HSC self-renewal, despite the functional discrepancies between cultured MSCs and in-vivo isolated MSCs [13]. Accordingly, interests are highlighted on the mechanisms that maintains the primitive state of MSCs, but the key molecular regulators have yet to be identified.

Recently, studies have shown that maintenance of the undifferentiated state in stem cells is characterized by a dynamic epigenetic plasticity. Specifically, pluripotent embryonic stem cells (ESCs) contain less-condensed chromatin structures [14,15] and there is poised (primed but held-in-check) expression of lineage-associated regulatory genes through a bivalent mode of histone modifications by polycomb (PcG) group proteins [14,16]. Moreover, a hierarchy of stemness was correlated to the degree of open, “loose” chromatin with higher rates of histone modification exchange, which became “condensed” heterochromatin upon differentiation [17 –19], thus illustrating the dynamic nature of chromatin in the undifferentiated state of stem cells.

The dynamic nature of chromatin and its remodeling are mediated by multiple family members of chromatin helicase DNA-binding domain (CHD) proteins through binding to methylated histones [20]. The CHD family of chromatin remodeling enzymes comprises two chromo domains and two Sucrose NonFementable2 (SNF2)-like ATP-dependent helicase domains [21], while each subfamily harbors additional specific domains [20]. Accordingly, CHD1 was shown to be necessary for the maintenance of open chromatin and the pluripotency of ESCs, as well as for reprogramming fibroblasts into a pluripotent state [22]. However, little is known about whether these chromatin remodeling activities could be mechanisms for regulating the primitive state of MSCs to control their clonogenic potential and ability to support stem cell regeneration. In this study, we investigated the possibility that CHD1 controls the cell fate of MSCs to influence their stem cell niche function in supporting HSCs.

Materials and Methods

Human cells and culture

Human MSCs were separated from BMs of acute myeloid leukemia (AML) and aplastic anemia (AA) patients under informed consent. Human cord bloods and MSCs were obtained from normal donors under informed consent MC19TNSI0012. This study was approved by the Institutional Review Boards of St. Mary's Hospital and Catholic University of Korea. Human MSCs were cultured in low-glucose Dulbecco's modified Eagle's medium (DMEM; HyClone, Waltham, MA) supplemented with 10% fetal bovine serum (FBS; HyClone) as described [23].

Coculture

Human CD34⁺ cells from umbilical cord blood were isolated by immunomagnetic column using Dynabeads (Invitrogen, Carlsbad, CA) according to the instructions. For coculture assays, MSCs were irradiated with 15Gy of ¹³⁷Cs γ-rays (Gammacell 1000; MDS Nordion, Ottawa, ON, Canada) 24 h before coculture with human CD34⁺ cells for 4 days in DMEM supplemented with FBS in the presence of a cytokine mixture (20 ng/mL human SCF; 20 ng/mL human Flt3L (ProSpec-Tany TechnoGene Ltd.), 4 ng/mL human IL3, IL6 (R&D Systems, Minneapolis, MN), and 4 ng/mL human G-CSF (ProSpec) supplemented with 10⁻⁶M hydrocortisone (Sigma-Aldrich, St Louis, MO).

Flow cytometry and cell cycle analysis

Human CD34⁺ cells were stained with CD45-PE, CD34-APC, and CD90-FITC antibodies (BD PharMingen). Cell cycles of MSCs were analyzed by staining the MSCs with 10 μmol/L Hoechst33342 at 37°C for 45 min, and Pyronin Y was then added to give a final concentration of 2.5 μg/mL. Data were analyzed with Diva software using a LSRII.

Transduction and transfection of MSC

A negative scrambled control shRNA sequence and a CHD1/shRNA sequence were cloned into the pMirzip lentiviral vector (System Biosciences, Palo Alto, CA). For overexpression of CHD1, wild-type CHD1 cDNA was cloned into the pMIRNA1 lentiviral vector (SBI). MSCs were transduced with each lentiviral vector and sorted for transduced (GFP⁺) cells. For transfection of siRNA into MSCs, CHD1-specific siRNA and the control siRNA were purchased from GenePharma (Shanghai, China) and each siRNAs (50 nM) were transfected in the presence of oligofectamine (Invitrogen).

CFU-F assay and differentiation of MSCs

For CFU-F assays, MSCs were plated in 100 mm tissue culture dish. Fourteen days after plating, the cells were fixed with methanol and then stained with Crystal Violet (Sigma) for visualization. For osteogenic or adipogenic differentiation of MSCs, MSCs under growth medium were switched to differentiation-inducing medium for 14 days for osteogenic and 14 days for adipogenic differentiation. The extent of terminal differentiation was quantified by mineralization or lipid droplets in the differentiated cells using Alizarin Red S or Oil Red O staining, followed by spectrophotometry at 550 and 520 nm, respectively.

Real-time polymerase chain reaction

Total RNA was extracted from MSCs using TRIzol (Invitrogen). cDNA was synthesized from 1 μg of total RNA with reverse transcriptase (Invitrogen). Real-time quantitative polymerase chain reaction (RT-qPCR) was performed with SYBR Premix Ex Taq (TaKaRa, Japan). The threshold cycle (Ct) value for each gene was normalized to the Ct value of 18sRNA. The relative mRNA expression was calculated by using the formula; 2^−ΔΔCt, where ΔCt = Ct_sample− Ct_18sRNA and ΔΔCt = ΔCt_sample − ΔCt_reference group.

Western blotting

For western blot analysis, MSC lysates were subjected to gel electrophoresis [sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS-PAGE)], transferred to membranes, and incubated with primary antibodies against each specific antibody. The membranes were then incubated with horseradish peroxidase-conjugated secondary antibodies, and proteins were visualized using a chemiluminescence substrate (Thermo Scientific, Abbott Park, IL). Primary antibodies against CHD1 were purchased from Santa Cruz Biotechnology (Dallas, TX).

Senescence-associated β-galactosidase assay

Senescence-associated β-galactosidase (SA-β-gal) activity was examined by the Senescence β-Galactosidase Staining Kit (Cell Signaling Technology, Danvers, MA) according to the instruction. Stained cells were observed under a light microscope.

Chromatin immunoprecipitation asssay

Chromatin immunoprecipitation (ChIP) was performed with the EZ-ChIP™ Kit (Millipore/Upstate) according to the manufacturer's instruction. Briefly, MSCs were fixed in 1% formaldehyde, washed and lysed, then sonicated to shear DNA. Subsequently, anti-tri-MeH3K9 (Cell Signaling Technology) was added for immunoprecipitation along with protein A/G sepharose beads. After washes, protein/DNA complexes were reverse crosslinked for 4 h at 65°C, and analyzed by quantitative PCR for each gene promoter-specific sequences.

Chromatin accessibility

Chromatin accessibility of the Oct4 gene promoter in hMSCs was assessed by nuclease protection assay using the EpiQ Chromatin Analysis Kit (Bio-Rad, Munich, Germany). Briefly, cells were incubated in EpiQ chromatin buffer with or without nuclease at 37°C for 1 h, then genomic DNA was extracted. Chromatin accessibility was assessed by quantitative PCR and calculated using the EpiQ Chromatin Kit data analysis tool. The β hemoglobin (HBB) gene was used as the inaccessible reference for measuring relative chromatin accessibility at Oct4 gene locus. PCR primers: Oct4, 5′-CACTGCACTCCAGTCTGGGCAACAA and 5′-TGCCTAATGGTGGTGGCAATGGTGT; HBB, 5′-AGCCAGTGCCAGAAGAGCCAAGGA and 5′-CCCACAGGGCAGTAACGGCAGACTT.

Results

To determine whether CHD1 is involved in regulating the primitive state of MSCs, we first investigated whether CHD1 expression in MSCs changes with respect to the frequency of mesenchymal progenitors. For this, we examined the expression levels of CHD1 in MSCs during serial passage culture that causes a loss in the number of colony-forming unit fibroblasts (CFU-F). As shown, serial culture of MSCs caused a progressive loss of CFU-F with increased passage numbers (Fig. 1A). This decrease in the frequency of CFU-F was accompanied by a concomitant decrease of CHD1 expression levels in MSCs (Fig. 1B).

FIG. 1.

Changes in CHD1 expression levels in MSCs during passage cultures and differentiation. (A, B) Expression levels of CHD1 in MSCs during passage culture. MSCs obtained from normal donors were subjected to passage subculture. For each indicated passage (p), CHD1 expression levels were analyzed by RT-qPCR, and the frequency of clonogenic cells (colony-forming unit fibroblast; CFU-F) were analyzed by colony-forming assays followed by Crystal Violet staining for visualization of the colonies. (C) Expression levels of CHD1 in MSCs during terminal differentiation. MSCs were induced to terminally differentiate into either osteogenic or adipogenic lineages for 14 days. After terminal differentiation, CHD1 levels were compared relative to the undifferentiated MSCs (n = 6, *P < 0.05). CHD, chromatin helicase DNA-binding domain; MSC, mesenchymal stromal cell; RT-qPCR, real-time quantitative polymerase chain reaction; SEM, standard error of the mean.

We also compared the expression of CHD1 during osteogenic or adipogenic differentiation of cultured MSCs. The expression level was highest in the undifferentiated state of MSCs, but decreased during their terminal differentiation into osteogenic or adipogenic lineages (Fig. 1C and Supplementary Fig. S1). These results indicate that CHD1 expression levels of MSCs are regulated along with changes in cell fate such that higher levels of expression of CHD1 are associated with the primitive, undifferentiated state of MSCs.

To further examine the functional significance of CHD1 expression levels in MSCs, we next examined the effect of CHD1 knockdown (KD) on the clonogenic potential and multilineage differentiation of MSCs. Lentiviral vector expressing shRNA against CHD1 was transduced into MSCs, which caused a significant decrease in CHD1 expression levels (Fig. 2A–C and Supplementary Fig. S2).

FIG. 2.

Effects of CHD1-KD on proliferation and frequency of CFU-F. (A) Schematic illustration of the vector for shRNA against CHD1. copGFP; Copepod green fluorescent protein, T2A; translational cleavage site, CMV; cytomegalovirus promoter, LTR. (B, C) KD of CHD1 in MSCs. Decrease of CHD1 expression levels in MSCs were confirmed by RT-qPCR (B) and western blot (C) (n = 3, *P < 0.05). (D) Changes in proliferative activity of MSCs by CHD1-KD. MSCs transduced with control vector or CHD1-KD vector were analyzed for proliferative activity. Shown are the increase in cell number after 3 days of culture from plating 2 × 10⁴ cells. (E) Cell cycle analysis of CHD1-KD MSCs. MSCs were transduced with control or CHD1-shRNA vector. Cell cycling status of transduced (GFP⁺) cells were analyzed by Hoechst33342/Pyronin Y staining. Shown are the representative flow cytometry profiles (left) and quantification of cells in each cycling stage (right) (mean ± SEM, n = 3, *P < 0.05). (F) Effects of CHD1-KD on the frequency of CFU-F in MSCs. Relative frequency of CFU-F in CHD1-KD MSCs are shown in comparison to the frequencies of CFU-F in control MSCs (n = 6, *P < 0.05). (G) Comparisons of senescence in MSCs. MSCs transduced by each vector were compared for SA-β-gal activity. Shown are the mean ± SEM (n = 15). KD, knockdown; LTR, long terminal repeat; SA-β-gal, senescence-associated β-galactosidase.

The KD of CHD1 caused a moderate decrease in cell proliferation, with an increase in the number of cells in the G0 phase of the cell cycle (Fig. 2D, E). Strikingly, MSCs with CHD1-KD exhibited a profound decrease in the number of CFU-F compared with control MSCs (Fig. 2F and Supplementary Fig. S3A). This decrease of CFU-F was not associated with an increase in cell senescence, as evidenced by comparable levels of SA-β-gal activity (Fig. 2G and Supplementary Fig. S3B). These results show that high expression levels of CHD1 are critical for the maintenance of mesenchymal progenitor cells, and decreased expression of CHD1 causes the loss of clonogenic potential.

To determine whether the effects of CHD1-KD on the clonogenic potential of MSCs is related to changes in stemness secondary to its known chromatin remodeling activity, we examined the expression changes of pluripotency-related genes. As shown, the expression levels of OCT4, NANOG, and SOX2 were significantly decreased in CHD1-KD MSCs compared with the control MSCs (Fig. 3A). The decrease in expression levels of these pluripotent genes were similarly reproduced in MSCs with siRNA-induced KD of CHD1 (Supplementary Fig. S4). The decrease in pluripotent gene expression was associated with condensation of chromatin, as evidenced by an increase in the repressive histone mark, H3K9-me3 [24], in the OCT4 promoter (Fig. 3B) as well as a decrease in chromatin accessibility to nuclease at the promoter (Fig. 3C, D). Of note, the ChIP assay for MSCs with CHD1-KD exhibited increase of H3K9-me3 in the promoters of genes for primitive state of MSCs, including SOX2, KLF4, and NESTIN [10], but not tissue-specific genes such as RUNX2 or OSTEOCALCIN (Supplementary Fig. S5), suggestive of preferential silencing of genes involved in the primitive state of MSCs. These results indicate that CHD1 regulates the stemness of MSCs through its chromatin remodeling activities.

FIG. 3.

Effects of CHD1-KD on the transcription of pluripotent genes. (A) Effects of CHD1-KD on the expression of pluripotent genes. Shown are the relative expression levels in CHD1-KD MSCs relative to the control MSCs (mean ± SEM, n = 3, *P < 0.05). (B) Chromatin immunoprecipitation assay for the OCT4 promoter. MSCs with control or CHD1-KD were immunoprecipitated from chromatin with a H3K9me-3-specific antibody, and the coprecipitation of DNA sequences specific for the OCT4 promoter were quantitatively measured by RT-qPCR. Shown are the levels of OCT4 promoter sequence relative to the control (mean ± SEM, n = 3). (C, D) Chromatin accessibility test for the OCT4 promoter. Chromatin from control or CHD1-KD MSCs were subjected to nuclease treatment followed by DNA extraction. The OCT4 promoter sequence was amplified by region-specific PCR primers for quantitative measurement of OCT4 sequences protected from nuclease digestion. Shown are the representative profiles of RT-qPCR in the presence or absence of nuclease in each group (C), and quantitative measurement of chromatin accessibility as determined by the difference in Cq value between the untreated and nuclease-treated samples (D) (mean ± SEM, n = 3, *P < 0.05).

We next examined the effect of CHD1-KD on the multilineage differentiation potential of MSCs. When subjected to osteogenic or adipogenic differentiation, MSCs with CHD1-KD exhibited a significant decrease in their terminal differentiation into both lineages (Fig. 4A, B). These results show that CHD1, while contributing to the maintenance of the undifferentiated state, also plays a role in guidance of multilineage differentiation process.

FIG. 4.

Effects of CHD1-KD on multilineage differentiation of MSCs. CHD1-KD or control MSCs were subjected to osteogenic or adipogenic differentiation, and stained with Alizarin Red S or Oil Red, respectively. Shown are the representative profiles for osteogenic and adipogenic differentiation (A) and quantification as determined by absorbance at 550 nm (osteogenic) and 520 nm (adipogenic) (B) (mean ± SEM, n = 3, *P < 0.05).

Based on the effects of CHD1-KD in MSCs, we next questioned whether increasing CHD1 expression would cause the opposite effects on the characteristics of MSCs. Thus, MSCs overexpressing CHD1 were generated by transduction with lentiviral vector encoding wild-type CHD1 (Fig. 5A–C). As shown, the CHD1-overexpressing MSCs exhibited an increase in the frequency of CFU-F and increased expression of pluripotent genes without changes in SA-β-gal activity (Fig. 5D–F). The increased expressions of pluripotent genes were associated with decondensation of chromatin as evidenced by increased accessibility to nuclease (Fig. 5G, H). Similarly, MSCs overexpressing CHD1 exhibited an increase in osteogenic differentiation potential while maintaining adipogenic differentiation potential (Fig. 5I). Altogether, these results show that the expression levels of CHD1 is a factor controlling the stemness and multilineage differentiation potential of MSCs.

FIG. 5.

Effect of CHD1 overexpression in MSCs. (A) Schematic illustration of lentiviral vectors, EF1; Elongation factor 1α promoter, copGFP; Copepod green fluorescent protein. (B, C) Overexpression of CHD1 in MSCs. Expression levels of CHD1 were compared between control and CHD1-transduced MSCs by RT-qPCR (B) and western blot (C). Shown are the mean ± SEM of expression levels relative to the control transduced MSC (n = 4, *P < 0.05), and representative profile for western blot, respectively. (D) Effects of CHD1 overexpression on the frequency of CFU-F. Fold changes in the frequency of CFU-F after cell plating (1,000 cells) are shown relative to the frequencies of CFU-F in control MSCs (mean ± SEM, n = 6, *P < 0.05). (E) Effects of CHD1 overexpression on the expression of pluripotent genes. Expression levels of each indicated gene were measured by RT-qPCR in the control and CHD1-overexpressing MSCs. Shown are the expression levels of CHD1-overexpressing (OE) cells relative to the control MSCs. (F) Effects on senescence of MSCs. Shown are the % of cells exhibiting positive staining for SA-β-gal activity (mean ± SEM, n = 15, *P < 0.05). (G, H) Changes in the chromatin accessibility of OCT4 promoter in MSCs by overexpression of CHD1. Chromatin from control or CHD1-OE MSCs were subjected to nuclease treatment followed by DNA extraction. The OCT4 promoter sequence was amplified by region-specific PCR primers for quantitative measurement of OCT4 sequences protected from nuclease digestion. Shown are the representative profiles of RT-qPCR in the presence or absence of nuclease in each group (G), and quantitative measurement of chromatin accessibility as determined by the difference in Cq value between the untreated and nuclease-treated samples (H). (I) Effects of CHD1 overexpression on multilineage differentiation of MSCs. Control and CHD1-overexpressing MSCs were subjected to osteogenic or adipogenic differentiation. Shown are the representative profile (left) and quantitative measurement of differentiation as determined by absorbance at 550 and 520 nm for osteogenic and adipogenic cells, respectively (right) (mean ± SEM, n = 5, *P < 0.05).

Since the primitive stemness of MSCs is related to their niche function supporting endogenous stem cells [10,23,25 –27], we next questioned whether CHD1 in MSCs was also involved in the control of niche activity to support HSCs.

The HSC-supporting activity of MSCs is reflected by a shift toward expressing epithelial/mesenchymal transition (EMT) genes [28,29]. Thus, we first examined the influence of CHD1 expression levels on expression changes of EMT genes. MSCs with CHD1-KD exhibited significantly lower expression levels of EMT-regulating genes, including ZEB1, ZEB2, TWIST1, SNAI1, and SLUG (Fig. 6A). Moreover, KD of CHD1 caused a significant decrease in NESTIN and CD146 levels, which are markers for mesenchymal niche cells displaying HSC-supporting activities [9,10,25,30] (Fig. 6B). In contrast, overexpression of CHD1 in MSCs upregulated the EMT-related genes (Fig. 6C) and the ratio of E-cadherin to N-cadherin reflecting EMT was concomitantly changed with KD or overexpression of CHD1 (Fig. 6D), indicating that EMT genes are controlled by CHD1 expression levels.

FIG. 6.

Effects of CHD1 expression changes on the HSC-supporting activity of MSCs. MSCs were transfected with control or siRNA against CHD1. (A) Effects of CHD1-KD on EMT-related gene expression in MSCs. Expression levels of EMT-related genes in CHD1-KD MSCs analyzed by RT-qPCR are shown relative to the control MSCs (n = 3, mean ± SEM, *P < 0.05). (B) Effects of CHD1-KD on the expression of pericyte markers. Expression levels of each indicated pericyte marker were compared for control and CHD1-KD MSCs by RT-qPCR. Shown are the expression levels in CHD1-KD MSCs relative to the control MSCs (n = 3, mean ± SEM, *P < 0.05). (C) Effects of CHD1 overexpression on the expression of EMT-related genes in MSCs. Expression levels of each indicated EMT gene in CHD1-overexpressing MSCs determined by RT-qPCR are shown in reference to the levels in control MSCs (n = 3, mean ± SEM, *P < 0.05). (D) Comparisons of E-cadherin/N-cadherin expression levels in MSCs with KD or overexpression of CHD1. MSCs with KD or overexpression of CHD1 were analyzed for relative expression levels of E-cadherin and N-cadherin in reference to the control-transduced MSCs. Shown are the changes in the ratio of expression levels for E-cadherin to N-cadherin in each group MSCs in comparison to the control MSCs as determined by RT-qPCR. (n = 3, mean ± SEM, *P < 0.05). (E) Effects of CHD1 expression levels in MSCs on HSC-supporting activity. MSCs overexpressing CHD1 or CHD1-KD were cocultured with umbilical cord blood-derived CD34⁺ cells. Four days after coculture, the expansion of primitive hematopoietic subpopulations (CD34⁺90⁺) were examined. Shown are the relative expansion fold of CD34⁺90⁺ hematopoietic cells after coculture with each indicated MSC group in comparison to the coculture with control MSCs. (n = 3 for OE, n = 9 for siRNA, mean ± SEM, *P < 0.05). EMT, epithelial/mesenchymal transition; HSC, hematopoietic stem cell.

Consistent with the effects on MSCs, umbilical cord blood-derived CD34⁺ cells cocultured with CHD1-KD MSCs also exhibited a significant decrease in the self-renewing expansion of primitive hematopoietic subsets (CD34⁺CD90⁺), the subset previously characterized as long-term repopulating HSCs [31]. In contrast, coculture of CD34⁺ hematopoietic cells with CHD1-overexpressing MSCs significantly enhanced the self-renewing expansion of CD34⁺90⁺ hematopoietic cells (Fig. 6E). Altogether, these results show that the changes in expression levels of CHD1 in MSCs is a molecular parameter that can cause a dynamic change in the HSC-supporting activity of MSCs.

To further explore the physiological significance of CHD1 expression changes for controlling HSC-supporting activity of MSCs, we examined CHD1expression levels under pathological conditions of altered niche activity. For this, we first examined MSCs derived from AML patients. The mesenchymal niche cells from AML patients exhibited loss of CFU-F and multilineage differentiation potential (Fig. 7A, B) and accordingly, loss of HSC-supporting activities (Fig. 7C), consistent with previous findings [30,32,33]. When CHD1 expression levels were examined, we found significantly lower levels in AML-derived MSCs compared with MSCs derived from normal donors (Fig. 7D). Thus, functional changes of MSCs induced by CHD1-KD, including loss of CFU-F, multilineage differentiation and HSC-supporting activity, were similarly reproduced in MSCs of AML patients, where CHD1 expression level is decreased.

FIG. 7.

Changes of CHD1 expression in MSCs from hematological disorders with degeneration of the stem cell niche. (A–D) Changes of MSCs under leukemia condition. MSCs obtained from normal donor and AML patients were compared for the frequency of CFU-F, multilineage differentiation, HSC-supporting activities, and expression levels of CHD1. (A) Total numbers of CFU-F in bone marrow samples were measured for normal and AML patients (n = 5 for normal, n = 9 for AML patients, mean ± SEM, *P < 0.05). (B) Normal or AML MSCs were subjected to osteogenic or adipogenic differentiation, and CFU-F, respectively. Shown are the representative profiles for osteogenic and adipogenic differentiation and CFU-F assays. (C) MSCs from each group were cocultured with CD34⁺ cells for 4 days. Expansion of CD34⁺ cells after coculture with the two different types of MSCs are shown (n = 5, mean ± SEM, *P < 0.05). (D) Changes in the expression levels of CHD1 in AML-MSCs. The expression levels of CHD1 in AML patients were examined in comparison to the expression in normal MSCs. (n = 7 for normal, n = 9 for AML patients, mean ± SEM, *P < 0.05). (E–H) Changes of MSCs under AA. MSCs from AA patients were compared with MSCs from healthy donors. Frequency of CFU-F (E), the representative profiles for osteogenic and adipogenic differentiation and CFU-F (F), HSC-supporting activities (G), and changes in CHD1 expression levels (H) are shown (n = 4 for normal, n = 5 for AA, mean ± SEM, *P < 0.05). AA, aplastic anemia; AML, acute myeloid leukemia.

We also examined MSCs from patients with AA, as an independent model for loss of stemness in MSCs. MSCs from AA similarly exhibited significantly decreased levels of CFU-F and multilineage differentiation potential (Fig. 7E, F), decrease of HSC-supporting activities (Fig. 7G). Similar to the MSCs from AML, CHD1 levels in AA-derived MSCs were significantly lower than the levels in normal MSCs (Fig. 7H), similarly mirroring the MSCs with CHD1-KD.

Altogether, these results show that the changes of the CHD1 expression levels in MSCs are related to the molecular mechanism regulating the niche activity of MSCs in the context of diseases characterized by degenerative niche conditions and could comprise another pathogenic mechanism for alteration of HSC-supporting activities in patients with hematological insufficiency.

Discussion

Despite the widespread use of MSCs in clinical cell therapy trials, significant functional heterogeneity exists in their ability to stimulate tissue-specific stem cells, where primitive subset of MSCs with clonogenic potential play the major role as niche cells in-vivo [8 –12,34] and in-vitro [23]. However, factors that maintain the stemness of MSCs have not been well characterized.

In the current study, we found that the primitive state of MSCs is maintained by high expression levels of the chromatin remodeling complex, CHD1, which functions in the dynamic plasticity of chromatin structure.

The CHD1 expression levels are hierarchically regulated in a manner that higher expression levels are associated with the primitive and undifferentiated state of MSCs and KD or overexpression of CHD1 in MSCs could concomitantly influence their primitive state. Thus, dynamic remodeling of chromatin mediated by CHD1 may play a major role in the maintenance of the primitive state of MSCs, reminiscent of their role in maintenance of the pluripotent state and reprogramming into the stem cell state [19,20,22,35].

Supporting the possibility, KD of CHD1 caused selective downregulation in the transcription of pluripotent genes, but not tissue-specific genes. However, taking that the principal role of CHD1 has been implicated in the chromatin remodeling and hypertranscription phenomenon [35], it is not clear yet how the selectivity of target genes are determined in the transcription process.

Interestingly, the role of CHD1 in MSCs was not confined to the maintenance of primitive state, but was also involved in the guidance of multilineage differentiation process, that is, KD of CHD1 in MSCs inhibited their osteogenic or adipogenic differentiation.

Studies have shown that the stem cell state with multilineage differentiation potential is associated with higher epigenetic plasticity with open chromatin structure to permit the cell fate changes into various types of differentiated cells [36 –39]. In contrast, differentiated cells exhibit less epigenetic plasticity and epigenetic alterations leads to the apoptosis of the cells, stably maintaining the given cell identity [17].

Supporting the possibility, it was shown that CHD1 was also necessary for induction of osteoblast-specific genes [40] or for multilineage differentiation of ESCs [41]. Thus, dynamic remodeling of chromatin mediated by CHD1 is necessary for guidance of a differentiation program as well as for maintenance of primitive state for multiple spectrums of stem cells with different levels of stemness.

However, it is noteworthy that multiple members of the chromatin helicase domain (CHD) family have been shown to mediate chromatin remodeling [20] in a manner that is cell type specific. For example, chromatin remodeling by chd2 is required for determination of myogenic cell fates and induction of muscle-specific genes, including MyoD [42]. Similarly, chromatin remodeling by chd4 establishes a lymphopoiesis-specific gene expression profile [43], and chd9 is involved in the expression of osteocalcin for bone development [44,45]. In this regard, it is possible that CHD1, among multiple CHD family members, is primarily involved in the establishment of stemness signatures in multiple types of stem cells.

Of note, the primitive state of mesenchymal cells was directly related to their niche cell function to promote regeneration of tissue-specific stem cells, HSCs, as inferred from the influence of CHD1 expression on the self-renewing expansion of cocultured HSCs.

Previous studies have shown that subsets of MSCs that can serve as niche for HSCs in BM are characterized by subsets exhibiting higher frequencies of colony-forming cells and expression of specific markers such as Prx1, Nestin, Leptin, or CD51/PDGFR [11,12,46 –48]. Moreover, increase of colony-forming cells were correlated to higher HSC-supporting activities during in-vitro culture model [23], suggestive of the relationship between primitive state of MSCs and their niche function.

Moreover, given that HSC-supporting activities of MSCs were either upregulated or downregulated with respect to CHD1 expression levels, it is possible that CHD1 expression is a molecular parameter that can dictate the dynamic changes of MSC niche activities during their adaptive response to various extrinsic signals for regulating hematopoiesis [49]. Supporting this possibility, CHD1 expression levels were significantly downregulated in MSCs from patients with AML or AA, where MSCs exhibit a loss of stemness and decreased HSC-supporting activities [30,32,33]. The functional defect of AML is caused by extensive complexity of cellular and molecular alterations, but exhibit degenerative changes of MSCs with loss of HSC-supporting activities [30,50 –54]. Similarly, the cellular changes in the MSCs of AA is caused by multiple idiopathic etiologies, exhibiting similar degenerative changes [55 –57].

Taking that the degenerative changes of MSCs caused by CHD1-KD, including loss of CFU-F, multilineage differentiation or loss of HSC-supporting activity, were similarly reproduced in MSCs from AML or AA patients with a decrease of CHD1 expression, it is possible that changes in CHD1 expression comprise another target of molecule leading to the degenerative changes of MSCs in those patients exhibiting decrease of HSC-supporting activities.

At present, the upstream regulators that can modulate the expression changes of CHD1 remain unclear. Therefore, further studies are warranted to dissect the mechanism controlling dynamic changes of CHD1 expression levels in MSCs as well as signals that can control their expression under various physiological conditions. Nevertheless, our findings show another physiological role of chromatin remodeling factor, CHD1, as a new regulatory mechanism controlling the microenvironment of stem cells, in addition to role of the chromatin remodeling for maintenance of stem cell state, self-renewal or differentiation of normal and cancer stem cells [36 –39,58].

Conclusion

Our findings reveal that the chromatin remodeling activity mediated by CHD1 is a molecular parameter that can influence the primitive state of MSCs and their stem cell niche activity, pointing it as potential molecular target to control tissue regeneration.

Footnotes

Acknowledgment

The human MSCs from healthy donors (MASTER Cells) were supplied from the Catholic Institute of Cell Therapy (CIC, Seoul, Korea).

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study is supported by the NRF of Korea and funded by the Ministry of Science, ICT, & Future Planning (2017M3A9B3061947).

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Figure S4

Supplementary Figure S5

References

Keating

. (2006). Mesenchymal stromal cells. Curr Opin Hematol, 13:419–425.

Pittenger

, Mackay

, Beck

, Jaiswal

, Douglas

, Mosca

, Moorman

, Simonetti

, Craig

and Marshak

. (1999). Multilineage potential of adult human mesenchymal stem cells. Science, 284:143–147.

Prockop DJ. (1997). Marrow stromal cells as stem cells for nonhematopoietic tissues. Science, 276:71–74.

Caplan

and Correa

. (2011). The MSC: an injury drugstore. Cell Stem Cell, 9:11–15.

Murphy

, Moncivais

and Caplan

. (2013). Mesenchymal stem cells: environmentally responsive therapeutics for regenerative medicine. Exp Mol Med, 45:e54.

Frenette

, Pinho

, Lucas

and Scheiermann

. (2013). Mesenchymal stem cell: keystone of the hematopoietic stem cell niche and a stepping-stone for regenerative medicine. Annu Rev Immunol, 31:285–316.

Squillaro

, Peluso

and Galderisi

. (2016). Clinical trials with mesenchymal stem cells: an update. Cell Transplant, 25:829–848.

Chan

, Chen

, Luppen

, Kim

, DeBoer

, Wei

, Helms

, Kuo

, Kraft

and Weissman

. (2009). Endochondral ossification is required for haematopoietic stem-cell niche formation. Nature, 457:490–494.

Sacchetti

, Funari

, Michienzi

, Di Cesare

, Piersanti

, Saggio

, Tagliafico

, Ferrari

, Robey

, Riminucci

and Bianco

. (2007). Self-renewing osteoprogenitors in bone marrow sinusoids can organize a hematopoietic microenvironment. Cell, 131:324–336.

10.

Mendez-Ferrer

, Michurina

, Ferraro

, Mazloom

, Macarthur

, Lira

, Scadden

, Ma'ayan

, Enikolopov

and Frenette

. (2010). Mesenchymal and haematopoietic stem cells form a unique bone marrow niche. Nature, 466:829–834.

11.

Ding

, Saunders

, Enikolopov

and Morrison

. (2012). Endothelial and perivascular cells maintain haematopoietic stem cells. Nature, 481:457–462.

12.

Greenbaum

, Hsu

, Day

, Schuettpelz

, Christopher

, Borgerding

, Nagasawa

and Link

. (2013). CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature, 495:227–230.

13.

Mitchell

, McIntosh

, Zvonic

, Garrett

, Floyd

, Kloster

, Di Halvorsen

, Storms

, Goh

, et al. (2006). Immunophenotype of human adipose-derived cells: temporal changes in stromal-associated and stem cell-associated markers. Stem Cells, 24:376–385.

14.

Azuara

, Perry

, Sauer

, Spivakov

, Jorgensen

, John

, Gouti

, Casanova

, Warnes

, Merkenschlager

and Fisher

. (2006). Chromatin signatures of pluripotent cell lines. Nat Cell Biol, 8:532–538.

15.

Spivakov

and Fisher

. (2007). Epigenetic signatures of stem-cell identity. Nat Rev Genet, 8:263–271.

16.

Bernstein

, Mikkelsen

, Xie

, Kamal

, Huebert

, Cuff

, Fry

, Meissner

, Wernig

, et al. (2006). A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell, 125:315–326.

17.

Chung

, Kim

, Hong

, Kwon

, An

, Park

, Lee

and Oh

. (2009). Undifferentiated hematopoietic cells are characterized by a genome-wide undermethylation dip around the transcription start site and a hierarchical epigenetic plasticity. Blood, 114:4968–4978.

18.

Efroni

, Duttagupta

, Cheng

, Dehghani

, Hoeppner

, Dash

, Bazett-Jones

, Le Grice

, McKay

, et al. (2008). Global transcription in pluripotent embryonic stem cells. Cell Stem Cell, 2:437–447.

19.

Meshorer

, Yellajoshula

, George

, Scambler

, Brown

and Misteli

. (2006). Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev Cell, 10:105–116.

20.

Micucci

, Sperry

and Martin

. (2015). Chromodomain helicase DNA-binding proteins in stem cells and human developmental diseases. Stem Cells Dev, 24:917–926.

21.

Hall

and Georgel

. (2007). CHD proteins: a diverse family with strong ties. Biochem Cell Biol, 85:463–476.

22.

Gaspar-Maia

, Alajem

, Polesso

, Sridharan

, Mason

, Heidersbach

, Ramalho-Santos

, McManus

, Plath

, Meshorer

and Ramalho-Santos

. (2009). CHD1 regulates open chromatin and pluripotency of embryonic stem cells. Nature, 460:863–868.

23.

Kim

, Lee

, Choi

, Kim

, Chu

, Leem

and Oh

. (2016). Heterogeneous niche activity of ex-vivo expanded MSCs as factor for variable outcomes in hematopoietic recovery. PLoS One, 11:e0168036.

24.

Ninova

, Fejes Tóth

and Aravin

. (2019). The control of gene expression and cell identity by H3K9 trimethylation. Development, 146:dev181180.

25.

Qian

, Le Blanc

and Sigvardsson

. (2012). Primary mesenchymal stem and progenitor cells from bone marrow lack expression of CD44 protein. J Biol Chem, 287:25795–25807.

26.

Morikawa

, Mabuchi

, Kubota

, Nagai

, Niibe

, Hiratsu

, Suzuki

, Miyauchi-Hara

, Nagoshi

, et al. (2009). Prospective identification, isolation, and systemic transplantation of multipotent mesenchymal stem cells in murine bone marrow. J Exp Med, 206:2483–2496.

27.

Pinho

, Lacombe

, Hanoun

, Mizoguchi

, Bruns

, Kunisaki

and Frenette

. (2013). PDGFRalpha and CD51 mark human nestin+ sphere-forming mesenchymal stem cells capable of hematopoietic progenitor cell expansion. J Exp Med, 210:1351–1367.

28.

Jeon

, Lee

, Park

, Kim

, Shin

, Lee

and Oh

. (2017). Shift of EMT gradient in 3D spheroid MSCs for activation of mesenchymal niche function. Sci Rep, 7:6859.

29.

Wei

, Nakahara

, Asada

, Zhang

, Gao

, Xu

, Alfieri

, Brodin

, Zimmerman

, et al. (2020). Snai2 maintains bone marrow niche cells by repressing osteopontin expression. Dev Cell, 53:503–513.e5.

30.

Kim

, Shim

, Lee

, Yim

, Kim

, Leem

, Lee

, Min

and Oh

. (2015). Microenvironmental remodeling as a parameter and prognostic factor of heterogeneous leukemogenesis in acute myelogenous leukemia. Cancer Res, 75:2222–2231.

31.

Peault

, Weissman

, Buckle

, Tsukamoto

and Baum

. (1993). Thy-1-expressing CD34+ human cells express multiple hematopoietic potentialities in vitro and in SCID-hu mice. Nouv Rev Fr Hematol, 35:91–93.

32.

Schepers

, Pietras

, Reynaud

, Flach

, Binnewies

, Garg

, Wagers

, Hsiao

and Passegue

. (2013). Myeloproliferative neoplasia remodels the endosteal bone marrow niche into a self-reinforcing leukemic niche. Cell Stem Cell, 13:285–299.

33.

Zhang

, Ho

, Huang

, Maeda

, Lin

, Lee

, Hair

, Holyoake

, Huettner

and Bhatia

. (2012). Altered microenvironmental regulation of leukemic and normal stem cells in chronic myelogenous leukemia. Cancer Cell, 21:577–592.

34.

Wang

, Zhang

, Liu

and Zou

. (2013). Hierarchical organization and regulation of the hematopoietic stem cell osteoblastic niche. Crit Rev Oncol Hematol, 85:1–8.

35.

Percharde

, Bulut-Karslioglu

and Ramalho-Santos

. (2017). Hypertranscription in development, stem cells, and regeneration. Dev Cell, 40:9–21.

36.

Therizols

, Illingworth

, Courilleau

, Boyle

, Wood

and Bickmore

. (2014). Chromatin decondensation is sufficient to alter nuclear organization in embryonic stem cells. Science, 346:1238–1242.

37.

and Xie

. (2005). Stem cell self-renewal controlled by chromatin remodeling factors. Science, 310:1487–1489.

38.

Yadav

, Quivy

and Almouzni

. (2018). Chromatin plasticity: a versatile landscape that underlies cell fate and identity. Science, 361:1332–1336.

39.

Dixon

, Jung

, Selvaraj

, Shen

, Antosiewicz-Bourget

, Lee

, Ye

, Kim

, Rajagopal

, et al. (2015). Chromatin architecture reorganization during stem cell differentiation. Nature, 518:331–336.

40.

Baumgart

, Najafova

, Hossan

, Xie

, Nagarajan

, Kari

, Ditzel

, Kassem

and Johnsen

. (2017). CHD1 regulates cell fate determination by activation of differentiation-induced genes. Nucleic Acids Res, 45:7722–7735.

41.

Piatti

, Lim

, Nat

, Villunger

, Geley

, Shue

, Soratroi

, Moser

and Lusser

. (2015). Embryonic stem cell differentiation requires full length CHD1. Sci Rep, 5:8007.

42.

Harada

, Okada

, Konno

, Odawara

, Yoshimi

, Yoshimura

, Kumamaru

, Saiwai

, Tsubota

, et al. (2012). Chd2 interacts with H3.3 to determine myogenic cell fate. EMBO J, 31:2994–3007.

43.

Arends

, Dege

, Bortnick

, Danhorn

, Knapp

, Jia

, Harmacek

, Fleenor

, Straign

, et al. (2019). CHD4 is essential for transcriptional repression and lineage progression in B lymphopoiesis. Proc Natl Acad Sci U S A, 116:10927–10936.

44.

Shur

and Benayahu

. (2005). Characterization and functional analysis of CReMM, a novel chromodomain helicase DNA-binding protein. J Mol Biol, 352:646–655.

45.

Shur

, Socher

and Benayahu

. (2006). In vivo association of CReMM/CHD9 with promoters in osteogenic cells. J Cell Physiol, 207:374–378.

46.

Chitteti

, Cheng

, Streicher

, Rodriguez-Rodriguez

, Carlesso

, Srour

and Kacena

. (2010). Osteoblast lineage cells expressing high levels of Runx2 enhance hematopoietic progenitor cell proliferation and function. J Cell Biochem, 111:284–294.

47.

Méndez-Ferrer

, Michurina

, Ferraro

, Mazloom

, Macarthur

, Lira

, Scadden

, Ma'ayan

, Enikolopov

and Frenette

. (2010). Mesenchymal and haematopoietic stem cells form a unique bone marrow niche. Nature, 466:829–834.

48.

Pinho

, Lacombe

, Hanoun

, Mizoguchi

, Bruns

, Kunisaki

and Frenette

. (2013). PDGFRα and CD51 mark human nestin+ sphere-forming mesenchymal stem cells capable of hematopoietic progenitor cell expansion. J Exp Med, 210:1351–1367.

49.

Jeong

, Kim

and Oh

. (2018). The adaptive remodeling of stem cell niche in stimulated bone marrow counteracts the leukemic niche. Stem Cells, 36:1617–1629.

50.

Behrmann

, Wellbrock

and Fiedler

. (2018). Acute myeloid leukemia and the bone marrow niche-take a closer look. Front Oncol, 8:444.

51.

, Jeong

and Kim

. (2019). Normal and leukemic stem cell niche interactions. Curr Opin Hematol, 26:249–257.

52.

Shafat

, Gnaneswaran

, Bowles

and Rushworth

. (2017). The bone marrow microenvironment—home of the leukemic blasts. Blood Rev, 31:277–286.

53.

Wang

and Zhong

. (2018). Roles of the bone marrow niche in hematopoiesis, leukemogenesis, and chemotherapy resistance in acute myeloid leukemia. Hematology, 23:729–739.

54.

Yamashita

, Dellorusso

, Olson

and Passegué

. (2020). Dysregulated haematopoietic stem cell behaviour in myeloid leukaemogenesis. Nat Rev Cancer, 20:365–382.

55.

Fattizzo

, Giannotta

and Barcellini

. (2020). Mesenchymal stem cells in aplastic anemia and myelodysplastic syndromes: the “Seed and Soil” crosstalk. Int J Mol Sci, 21:5438.

56.

Medinger

, Drexler

, Lengerke

and Passweg

. (2018). Pathogenesis of acquired aplastic anemia and the role of the bone marrow microenvironment. Front Oncol, 8:587.

57.

Schoettler

and Nathan

. (2018). The Pathophysiology of acquired aplastic anemia: current concepts revisited. Hematol Oncol Clin North Am, 32:581–594.

58.

Liau

, Sievers

, Donohue

, Gillespie

, Flavahan

, Miller

, Venteicher

, Hebert

, Carey

, et al. (2017). Adaptive chromatin remodeling drives glioblastoma stem cell plasticity and drug tolerance. Cell Stem Cell, 20:233–246.e7.

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

0.06 MB

0.32 MB

0.06 MB

0.08 MB