Endothelin-1 Reduces Catabolic Activity of Human Mesenchymal Stem/Stromal Cells During Chondro- and Osteo-lineage Differentiation

Abstract

Human mesenchymal stem/stromal cells (hMSCs) reside in a vascularized microenvironment and experience a host of blood vessel secretions, including endothelin-1 (ET1). Previously, our group has demonstrated improved induction of osteogenesis and chondrogenesis in hMSCs through an ET1-induced increase in production of anabolic factors. The current study explores effects of ET1 on catabolic factors secreted by hMSCs during chondrogenesis and osteogenesis. Cell proliferation and extracellular matrix (ECM) deposition were also explored. Our results demonstrated that ET1 reduced mRNA transcript levels of MMP2, MMP13, ADAMTS4, and ADAMTS5 in chondrogenic hMSCs, and MMP13 and ADAMTS5 in osteogenic hMSCs. Furthermore, ET1-treated chondrogenic and osteogenic hMSCs showed more intense stains for Alcian blue and Alizarin red, respectively, than control cells. Immunocytochemical results demonstrated that the ET1-mediated reduction of MMP13 could be reversed through blocking ET1 induction. Overall, our findings indicate that hMSCs treated with ET1 during chondrogenic or osteogenic induction attenuate catabolic activities of the cell to reduce ECM degradation, suggesting that it may be beneficial to use ET1 to enhance hMSC differentiation and protect newly synthesized ECM from degradation.

1. INTRODUCTION

The use of human mesenchymal stem/stromal cells (hMSCs) in regenerative medicine and tissue engineering holds promise for therapeutic and research purposes. Paracrine and autocrine factors are critical in modulating activities including proliferation and differentiation of the cell; therefore, it is imperative to understand the importance of all the biomolecules within the hMSC cellular environment. Within the perivascular niche, hMSCs reside on the blood vessel wall and experience a host of blood vessel secretions, including endothelin-1 (ET1) [1–3]. Previously, our group has demonstrated an increase in induction of osteogenesis and chondrogenesis of hMSCs through the effect of endothelin-1 (ET1) and activation of the AKT signaling pathway [4, 5]. While the findings have elucidated the effect of ET1 on anabolic activities of hMSCs, effects of the molecule on their catabolic activities were not explored.

Induction of hMSCs for chondrogenesis and osteogenesis involves producing ECM components specific for differentiated cells. A balance between ECM production and degradation is maintained for tissue homeostasis [6–8]. In cartilage, collagen type 2 (COL2) and aggrecan (ACAN) comprise most of the dry ECM components [9–11]. Similarly, dry ECM components in bone include collagens, and glycoproteins, but most of its content is attributed to hydroxyapatite [12]. The collagen support structure of bone is comprised predominantly of collagen type 1 (COL1), which is mineralized by hydroxyapatite deposits. In addition to the anabolic products that build cartilage and bone, various protein enzymes degrade the ECM. Matrix metalloproteinases (MMPs) degrade collagens, and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTs) degrade proteoglycans. In cartilage, MMP2 and MMP13 degrade COL2, and their overexpression is associated with osteoarthritis [13, 14]. MMPs, particularly MMP13, are involved in bone formation and remodeling [15]. ADAMTS1, ADAMTS4, and ADAMTS5, are the primary aggrecanases that degrade aggrecans in cartilage and are similarly linked to osteoarthritis [16]. These aggrecanases are also known to participate in bone resorption and development [17]. The purpose of this study is to explore the effect of ET1 on the production of ECM molecules and degradation enzymes.

ET1 is a potent vasoconstrictor, but it has many physiological roles beyond regulating vascular activities. It has been shown that ET1 impacts cartilage health, bone synthesis and resorption, and immunomodulation [18]. For example, ET1 production is upregulated by transforming growth factor beta-1 (TGFB1) and inflammatory cytokines, potentially associating ET1 with inflammatory response [19, 20]. Additionally, in mice, ET1 knockout disrupts musculoskeletal development whereas overexpression induces skeletal and neuromuscular remodeling [21–23]. There is also a link between ET1 and pathological conditions of skeletal tissues. For example, ET1 concentration in the joint is correlated with severity of knee osteoarthritis [24]. Similarly, increased ET1 during osteoarthritis triggers nitric oxide and collagenase production by chondrocytes, promoting further cartilage ECM degradation [25–27]. In bone, ET1 is implicated in fibrosis and sclerosis, particularly during osteoarthritis [28]. Degradation of the ECM in response to ET1 in joint tissues is associated with stimulation of MMP1 and MMP13 production [28, 29]. To date, the understanding of the catabolic response of skeletal progenitor cells, hMSCs, to ET1 stimulation is still limited, particularly during cartilage and bone lineage differentiation; thus, the current investigation was conducted to determine the role of ET1 in regulating the catabolic activity of hMSCs undergoing chondrogenesis and osteogenesis.

2. MATERIALS AND METHODS

2.1. Cell culture

Procurement of human bone marrow was granted by the Institutional Review Board at the University of Wisconsin-Madison. Four bone marrow-derived hMSC lines were isolated from donors following a previous protocol [4]. Harvested cells were maintained in an incubator at 37°C in a humidified 5% CO2 atmosphere. Expansion medium was comprised of low-glucose DMEM, 10% fetal bovine serum, and 1% antibiotics. Cells were passaged at 70–80% confluence using 0.05% trypsin/EDTA (Gibco) and re-plated at a density of 1,000 cells/cm². Medium was replaced every 3 days. Cells between passages 2 and 5 were used for this study. Treated cells were exposed to 0.1 μM ET1 during medium changes. In the experiment of blocking ET1 activity, 10 μM bosentan was administered 30 minutes prior to ET1 treatment.

2.2. Proliferation analysis

The effect of ET1 on cell proliferation was determined by comparing DNA content of MSCs treated with different concentrations of the compound. Briefly, cells were seeded at 1,000 cells/cm² and expanded with 0.001 μM, 0.01 μM, and 0.1 μM for 11 days. DNA content of the cells was measured by the PicoGreen assay following manufacturer’s instructions.

2.3. Osteo- and chondro-genic differentiation

MSCs were expanded in culture with or without 0.1μM ET1 for two passages prior to differentiation. Control groups were cells expanded for two passages and induced for differentiation without ET1 whereas treated groups were administered 0.1 μM ET1. For chondrogenic induction, 250,000 cells were collected to make a pellet as previously described and then induced by medium comprised of high-glucose DMEM, 1% ITS, 50 μg/ml L-ascorbic acid-2-phosphate, 0.1μM dexamethasone, 40 μg/ml L-proline, 0.9 mM sodium pyruvate, and antibiotics [4]. ET1 and TGFB1 at 0.1 μM and 10 ng/ml, respectively, were administered fresh during medium changes. Differentiation medium was replaced every 3 days during induction.

For osteogenic differentiation, hMSCs seeded at the density of 5,000 cells/cm² were induced by medium comprised of low-glucose DMEM, 10% FBS, 10 mM b-glycerophosphate, 50 μg/ml L-ascorbic acid-2-phosphate, 0.1 μM dexamethasone, 0.01 μM 1α, 25-Dihydroxyvitamin D3, and antibiotics. ET1 at the concentration of 0.1 μM was administered fresh during medium changes. Fresh medium was replaced every 3 days.

2.4. Quantitative polymerase chain reaction (qPCR)

Total RNA was extracted from cells using Zymo-Spin columns (ZYMO Research) following the manufacturer’s instructions. RNA amounts and purity were measured using the NanoDrop 1000 spectrophotometer. Reverse transcription by the High-Capacity cDNA reverse Transcription Kit (Thermo Fisher Scientific) was performed according to the manufacturer’s instructions. Quantitative PCR analysis using SYBR Green Supermix was performed with primers listed in Table S1. Expression levels of mRNA transcripts of interest were normalized to those of housekeeping Ubiquitin C.

2.5. Histology

Chondrogenic pellets were fixed using 4% paraformaldehyde in two series of 1-hour incubations at room temperature. Fixed pellets were rinsed with distilled water prior to serial dehydration up to 100% ethanol. Pellets were infiltrated with xylene prior to paraffin embedding. Embedded pellets were cut into 8-μm sections with a microtome. Sections were deparaffinized and rehydrated prior to staining with Alcian blue and Hematoxylin & Eosin.

Osteogenic culture plates were gently rinsed with PBS and fixed with 60% isopropanol. Fixed culture plates were rinsed with distilled water 3 times before staining with 2% Alizarin red solution for localization of calcium deposits.

2.6. Glycosaminoglycan and calcium content

Chondrogenic pellets were digested in papain buffer (0.1M Sodium Acetate, 0.05 M EDTA, pH 5.53) with 5mM L-cysteine hydrochloride hydrate and 20 μg/ml papain. Glycosaminoglycan (GAG) production was quantified using dimethylmethylene blue (DMMB) assay with normalization to DNA amounts measured by PicoGreen assay. Calcium deposition from osteogenic hMSCs was extracted from culture plates using 0.5M hydrochloric acid after rinsing with PBS. Extracted calcium samples were shaken for 24 hours at 4°C and centrifuged for 2 mins at 500g. Calcium in the supernatant was measured using the Calcium (CPC) Liquicolor Kit.

2.7. Immunocytochemical analysis

Immunocytochemical staining was used to analyze MMP13 localization. Sectioned pellets were deparaffinized, incubated in 4% paraformaldehyde and washed 3 times with PBS. Specimens were then incubated for 30 mins at room temperature with blocking buffer comprised of 1% bovine serum albumin in PBS with 0.1% Tween 20. Mouse IgG primary antibody (R&D #MAB511-SP) detecting human MMP13 was prepared in blocking buffer and incubated overnight at 4°C with gentle shaking. Goat anti-mouse IgG secondary antibody (Alexa Fluor 488 Intvitrogen #A-21121) was prepared in the blocking buffer and incubated 1hr at room temperature in the dark. After washing 3 times with PBS, sections were counterstained for cell nuclei with 1mg/mL 4’,6-diamidino-2-phenylindole (DAPI) in VECTASHIELD mounting medium for 15 min at 37°C and observed under a confocal microscope using FITC and DAPI filters.

2.8. Statistical analysis

All quantitative assays were performed in triplicate with 4 biological replicates (n=4) and the data were presented as mean ± standard error of the mean. A Student’s t-test or one-way ANOVA with pairwise t-tests was used for statistical comparison. A p-value less than 0.05 was considered statistically significant.

3. RESULTS

3.1. ET1 elicits dose-dependent response of hMSCs

We determined if there was a dose-dependent response of hMSCs to ET1 by first treating the cell with various concentrations of the molecule to analyze the mitogenic effect. No significant difference was detected in the DNA quantification between control and treated groups over 11 days (Fig. 1A). In addition to the proliferation analysis, we investigated the dose-dependent effect of ET1 on the production of a representative catabolic factor of hMSCs. The mRNA expression of MMP13 was significantly reduced with increasing ET1 concentration (Fig. 1B). The final concentration of 0.1 μM ET1 was selected for the other assays because it is the most effective dose for stimulation.

Figure 1.

Dose-dependent effects of recombinant ET1 on pre-differentiated hMSCs. (A) Proliferation of hMSCs treated with or without different doses of ET1 analyzed by quantifying total DNA content. (B) Relative mRNA expression levels of MMP13 in hMSCs treated with or without different doses of ET1 after two passages of cell expansion.

3.2. ET1 reduces activities of ECM proteinases in hMSC-derived chondrocytes

To determine the effect of ET1 on catabolic and anabolic activities of chondrocytes derived from hMSCs, chondrogenic pellets after 21-day induction were analyzed. Treated hMSC pellets expressed significantly less MMP2, MMP13, ADAMTS4, and ADAMTS5 than control ones while the expression of MMP1, MMP3, ADAMTS1, and ADAMTS7 between the two groups was comparable (Fig. 2A). On the other hand, the production of GAG in treated pellets was significantly greater (p = 0.04) than that in control ones (Fig. 2B), also revealed by the result of Alcian blue showing more intense staining in the treated than control group (Fig. 2C).

Figure 2.

Analysis of hMSC chondrogenesis treated with or without ET1 for 2 passages of expansion and 21-day differentiation. (A) Relative mRNA expression of catabolic molecules in chondrogenic pellets of control and treated groups. (B) GAG content normalized to DNA content in chondrogenic pellets. (C) Hematoxylin & Eosin (left) and Alcian blue (right) staining of chondrogenic pellets. Scale bar = 200 μm.

3.3. ET1 reduces activities of ECM proteinases in hMSC-derived osteoblasts

The effect of ET1 on catabolic and anabolic activities of osteoblasts derived from hMSCs was investigated by analyzing the cell after 21 days of osteogenic induction. The result of mRNA expression showed that treated osteogenic hMSCs expressed significantly less MMP13 and ADAMTS5 than control cells while levels of MMP1, MMP2, MMP3, ADAMTS1, ADAMTS4, and ADAMTS7 were comparable between the two groups (Fig. 3A). In terms of the anabolic activity compared to control cells, treated osteogenic hMSCs showed a significant increase in calcium deposition (Fig. 3B) and a more intense staining of Alizarin red (Fig. 3C).

Figure 3.

Analysis of osteogenic induction of hMSCs treated with or without ET1 for 2 passages of expansion and 21-day differentiation. (A) Relative mRNA expression of catabolic molecules in osteogenic hMSCs of control and treated groups. (B) Calcium deposition normalized to DNA content in osteogenic culture. (C) Alizarin red staining in osteogenic culture of hMSCs. Scale bar = 200 μm.

3.4. Inhibition of ET1 signaling increases MMP13 production

To determine if ET1 induction resulted in a reduction in the production of MMP13, we inhibited ET1 signaling of cells’ chondrogenic pellets by treating them with Bosentan to block ETAR and ETBR. Immunofluorescent staining results showed that while MMP13 protein content in chondrogenic pellets was reduced by ET1 addition compared to that in control ones without ET1, the reduction of MMP13 was reversed in those treated with Bosentan (Fig. 4). This finding indicates that ET1 is involved in the regulation of MMP13 in hMSC-derived chondrocytes.

Figure 4.

Fluorescent immunocytochemical staining of chondrogenic pellets treated with or without ET1 and Bosentan. FITC staining indicates MMP13 localization, and DAPI represents the counterstain for cell nuclei. Scale bar = 200 μm.

4. DISCUSSION

This study explores the impact of ET1 on the catabolic activity of hMSCs during osteogenesis and chondrogenesis. Previously, our group has demonstrated that ET1 primes hMSCs in expansion culture for chondro- and osteo- lineage differentiation [4, 5]. In this study, we extended ET1 induction to the differentiation culture by treating hMSCs with ET1 for two passages as well as 21 days of differentiation. Consistent with our previous findings, ET1 increased ECM deposition of hMSCs during osteo- and chondro- lineage differentiation. ET1 also induced a stark reduction in the production of catabolic factors, MMPs and ADAMTS. Overall, ET1 appears to decrease the catabolic activity of hMSCs undergoing chondrogenesis and osteogenesis.

Studies have shown that MMP2 and MMP13 degrade COL2 in cartilage and MMP13 is involved in bone remodeling [13–15]. ADAMTS1, ADAMTS4, and ADAMTS5 are the primary aggrecanases in cartilage and contribute to bone resorption [16, 17, 30]. Notably, our data show that ET1 treatment leads to a significant reduction in the expression of MMP13 and ADAMTS5 during both osteo- and chondro-genesis, and additionally that of ADAMTS4 during chondrogenesis, suggesting that ET1 is capable of mitigating activities of MMPs and aggrecanases to protect newly synthesized matrix during chondrogenesis and osteogenesis of hMSCs.

Effects of ET1 on the catabolic activity of cartilage and bone cells have been previously investigated but there is limited information available about whether ET1 regulates that of hMSCs during chondrogenesis and osteogenesis. Studies have shown that ET1 in the joint space is associated with bone/cartilage resorption and osteoarthritis progression. For example, in one of these studies, results show that the concentration of ET1 in synovial fluid is correlated with severity of knee osteoarthritis [24]. In another study, ET1 is found to be associated with cartilaginous endplate degeneration and intervertebral disc degeneration in humans [31]. It is known that during pathological states of articular chondrocytes and synoviocytes, ET1 stimulates MMP1 and MMP13 production [28, 29]. ET1 in osteoarthritic chondrocytes induces nitric oxide production and increases collagenase production to cause joint degradation [25–27]. Moreover, a study has reported that ET1 is implicated in fibrosis and sclerosis of subchondral bone, particularly during osteoarthritis [28]. Contrary to these previous results, our data in this study show that ET1 inhibits catabolic activities of hMSCs by reducing the production of MMPs and aggrecanases. Our findings suggest that dependent on cell type, ET1 may stimulate cells to either increase or decrease ECM degradation, dependent on skeletal cell types.

Overall, our results indicate that hMSCs treated with ET1 during chondrogenic or osteogenic induction attenuate catabolic activities of the cell to reduce ECM degradation. The finding suggests that it may be beneficial to use ET1 to enhance hMSC differentiation and protect newly synthesized ECM from degradation. However, it remains unclear why the anabolic role is reversed as seen in diseased conditions of cartilage and bone such as osteoarthritis. More investigation may be warranted to determine the role of ET1 in the regulation of hMSCs and derived skeletal cells.

Supplementary Material

1

List of supplementary files

Table S1. Primer sequences for quantitative RT-PCR

HIGHLIGHTS.

• Endothelin-1 reduces MMP13 and ADAMTS5 in chondro- and osteo-genic hMSCs.

• Reduction in MMP13 of chondrogenic hMSCs is mediated by activities of endothelin-1 receptors.

• Endothelin-1 can be used to enhance hMSC chondrogenesis and osteogenesis.

• Endothelin-1 can protect newly synthesized ECM proteins from degradation.