Differentiated markers in undifferentiated cells: Expression of smooth muscle contractile proteins in multipotent bone marrow mesenchymal stem cells

Authors


Author to whom all correspondence should be addressed.

Email: nieruqiong@126.com

Abstract

In studying the differentiation of stem cells along smooth muscle lineage, smooth muscle cell (SMC) contractile proteins serve as markers for the relative state of maturation. Yet, recent evidence suggests that some SMC markers are probably expressed in multipotent mesenchymal stem cells (MSCs). Such a paradox necessitates investigations to re-examine their role as differentiated markers in MSCs. We tried to detect the expression of four widely used SMC markers including α-smooth muscle actin (α-SMA), h1-calponin, desmin and smooth muscle myosin heavy chain (SM-MHC), as well as the other isoforms of calponin family in resting MSCs. Then we used three different conditions to initiate MSCs differentiation along SMC lineage, and examined the alternation of SMC markers expression at both the transcript level and protein level. Desmin and h1-calponin are expressed in MSCs, in the presence or absence of SMC induction conditions. Moreover, MSCs are shown to express all known isoforms of calponin. Double-staining reveals that h1-calponin +/α-SMA – cells constitute the majority of resting MSCs. Under differentiated conditions, expression of SM-MHC was initiated and expression of α-SMA was promoted. The expression of SM-MHC and upregulation of α-SMA are relatively reliable indications of a mature smooth muscle phenotype in MSCs. Given that the cells are particularly rich in calponins expression, we postulate possible roles of these proteins in regulating cellular function by taking part in actin cytoskeleton and signaling. These findings imply that an extensive study of the cell physiology of MSCs should focus on the functional roles for these proteins, rather than simply regard them as differentiated markers.

Introduction

A profound understanding of the development and differentiation of smooth muscle cells (SMCs) plays a vital role in defining the pathphysiologic process of vascular disease. It is demonstrated that in atherogenesis, not only the proliferation of resident SMCs, but also the circulating progenitor cells/stem cells from bone marrow, such as hematopoietic cells (HSCs) and mesenchymal stem cells (MSCs), contribute to intimal thickening (Saiura et al. 2001; Sata et al. 2002; Wagers et al. 2002). In the state of vascular injury, bone marrow derived cells migrate to facilitate vasculature and actively become part of neointima formation by differentiating into SMCs (Shimizu et al. 2001) (Hillebrands et al. 2001; Caplice et al. 2003). Among a variety of progenitor cells/stem cells and vascular cells, MSCs in particular seem to have a high potential for neointimal growth (Wang et al. 2008).

Mesenchymal stem cells (MSCs) are a group of multipotent progenitor cells characterized by their extensive consistent differentiation to multiple mesenchymal lineages (Pittenger et al. 1999). In vivo, MSCs are able to differentiate into pericytes or smooth muscle cells by direct injection into adult heart (Gojo et al. 2003) and contribute to the construction of tissue-engineered vascular autografts (Matsumura et al. 2003). In vitro, conditioned medium derived from smooth muscle, or, co-cultured with SMCs (Sata et al. 2002; Kanematsu et al. 2005), cultured in low concentration serum (Tamama et al. 2008), high concentration serum (Hegner et al. 2005), mechanical strain (Kurpinski et al. 2006) and biochemical factors treatment such as transforming growth factor-beta (TGF-β) (Kurpinski et al. 2010), and thromboxane A2 (Kim et al. 2009) proved to be able to reproducibly induce MSCs to differentiate into SMC-like cells.

In identifying a SMC-like population in the stem cell candidates, several smooth muscle isoforms of contractile apparatus serve as markers, such as α-smooth muscle actin (α-SMA), h1-calponin (bCaP), smooth muscle myosin heavy chain (SM-MHC) and desmin. α-SMA, the earliest known protein expressed in differentiation of the SMC during development, is highly selective for SMC or SMC-like cells in adult animals under normal circumstances, and it is also transiently expressed by a variety of mesodermally derived cells during development, tissue repair, and neoplastic growth. H1 (basic)-calponin, a relatively late marker during development in SMCs, appears to be restricted almost exclusively to smooth muscle, although it is now found in cardiac myocytes, myofibroblasts, and a variety of tumor cells. SM-MHC, also a late marker during development, shows the highest degree of cell specificity of any of the known markers of differentiated SMC (Owens 1995; Owens et al. 2004). Desmin, the principal intermediate filament protein in vascular SMCs of some blood vessels, was shown to be closely linked to the contractile function of muscle tissue (Capetanaki et al. 1997). In studying cellular differentiation of stem cells along smooth muscle lineage, the expressions of contractile proteins mentioned above logically serve as useful markers of the relative state of differentiation-maturation.

Yet, recent evidence posed threats to this approach to indentifying a differentiation-maturation state in MSCs. MicroSAGE analysis of gene expression of undifferentiated human MSCs revealed that the human MSCs colony simultaneously expressed transcripts characteristics of various mesenchymal cell lineages, including chondrocytes, osteoblasts, and myoblasts (Tremain et al. 2001). Primary cultured MSCs were demonstrated to express mRNA for most SMC markers at similar levels to SMCs, with an exception of SM-MHC (Ball et al. 2004; Tamama et al. 2008). Another study concerning the functional and proteomic changes as consequences of aging identified h1-Calponin was age-dependent expressed, which was expressed at a relatively high level in MSCs from young donors and significantly downregulated with age (Kasper et al. 2009). Smooth muscle contractile proteins serving as differentiated markers are detected in mesenchymal stem cells with multi-lineage potential.

However, the evidence mentioned above was obtained by high-throughput screening or assays at RNA level. Some researchers identified α-SMA/ h1-calponin positive cells as differentiated, smooth muscle like population in MSCs. In the present study, we conducted a straightforward investigation on the expression of some most widely used SMCs markers in MSCs at protein and mRNA level. We hypothesize that multipotent MSCs simultaneously express some of the SMC differentiated markers; consequently, they may fail to indicate a differentiation- maturation state of MSCs. Moreover, we hypothesize that MSCs are particularly rich in calponin expression, regardless of the differentiation state. We tried to detect the presence and intracellular distribution of SMC markers including α-SMA, h1-calponin, desmin and SM-MHC, as well as the other members in calponin families, in resting MSCs. Then we investigated the varieties of these SMC phenotypes under differentiation conditions, including (i) co-culture, a commonly used method to initiate differentiation of stem cells towards certain lineage (Sata et al. 2002; Kanematsu et al. 2005; Tian et al. 2010); (ii) low concentration of serum, a treatment reported to induce SMC gene markers expressions (Tamama et al. 2008; Treguer et al. 2009); and (iii) high concentration of serum, a condition that was believed to support optimal growth of stem cells and meanwhile, was probably a smooth muscle differentiation condition for MSCs (Hegner et al. 2005).

Materials and methods

Materials and antibodies

Cell culture complete medium for cells growth was Dulbecco's modified Eagle's medium-low glucose (DMEM, GIBCO, cat. no. 12800-116) supplemented with 10% heat inactivated fetal bovine serum (FBS; GIBCO, cat. no. 10270-106). Antibodies were purchased as follows: rabbit monoclonal [EP798Y] antibody to calponin1 (ab46794), mouse monoclonal [DE-U-10] to desmin (ab6322) and rabbit polyclonal to α smooth muscle actin (ab5694) were obtained from Abcam (Cambridge, UK). Rabbit polyclonal antibody to myosin smooth muscle heavy chain (sc-98705), goat polyclonal antibody to h2-Calponin [N-18] (sc-16607) and rabbit polyclonal antibody to h3-Calponin [H-55] (sc-28546) were purchased from Santa Cruz Biotechnology (CA, USA). Alexa Fluor 555 goat anti-rabbit antibody, Alexa Fluor 488 goat anti-mouse antibody and Alexa Fluor 488 donkey from Molecular Probes (OR, USA) were used as secondary antibody for immunofluorescence. Fluorescein isothiocyanate (FITC)-conjugated CD29, CD 45 and phycoerythrin (PE)-conjugated CD 34, CD 44 were from Becton Dickinson (CA, USA).

Cell culture

Sprague–Dawley rats, 6–8 weeks old, were killed by cervical dislocation. Mesenchymal stem cells were obtained from rat bone marrow by the Masoud Soleimani protocol (Soleimani & Nadri 2009) characterized by frequently changing medium during the initial phase of culturing. Briefly, rat's femurs and tibias were dissected and cleaned of adherent soft tissue on ice and bone marrow was harvested by flushing with complete medium, and then incubated at 37°C with 5% CO2 in a humidified chamber without any disturbing. After 3 h, the non-adherent cells that accumulated on the surface of the dish were removed by changing the medium. The medium was replaced after 8 h, and this step was repeated every 8 h for up to 72 h. When the cells reached 65–70% confluence, they were treated with 0.025% trypsin containing 0.025% ethylenediaminetetraacetic acid (EDTA) for 2 min at room temperature (25°C).

Cells at passage 2–5 were used for experiments. To investigate the smooth muscle markers in MSCs under differentiation conditions along SMC lineage, three different culture conditions were tested for 6 days: (i) co-cultured with aorta smooth muscle cells (to be elaborated later); (ii) cultured in DMEM with 0.5% FBS; or (iii) cultured in DMEM with 20% FBS. The first two treatments were known to induce differentiation along smooth muscle lineage by previous studies, while the effects of the last treatment were somehow obscure in that high concentration serum culture, of interest, suggested to be a differentiation condition for MSCs (Hegner et al. 2005).

The indirect co-culture was established by a transwell system with 0.4 μm pore polyester coated membrane (Corning, NY, USA). Smooth muscle cells from rat aorta were seeded on the upper compartment and MSCs were planted in the lower compartment.

Characteristic

Surface phenotypes analysis

Mesenchymal stem cells isolated from bone marrow were analyzed for surface marker expression using a FACS Calibur flow cytometry. Cultured cells were stained according to the manufacturer's recommendations with FITC-conjugated CD29, CD45 and PE-conjugated CD34, CD44. Background fluorescence was obtained from cells stained with FITC or PE alone. Quantitative analysis was done by a flow cytometer from Becton Dickinson (CA, USA) using CellQuestk software with 20 000 events recorded for each sample.

Adipogenic, osteogenic and chondrogenic differentiation assays

Like stem cells, cultured MSCs with multi-lineage potential must differentiate to osteoblasts, adipocytes and chondroblasts in vitro. The tri-lineage inductions were performed according to standard protocols as reported previously (Pittenger et al. 1999; Hegner et al. 2005). After the induction procedures, lipid vacuoles in adipogenesis were stained with 0.5% Oil red O, matrix mineralization in osteogenesis was stained with 5% Alizarin red-S and cartilage matrix in chodrogenesis was stained by alcian blue. Cultures were photographed using an inverted phase-contrast microscope and a CCD camera (Nikon Eclipse Ti-S).

Immunofluorescent staining

Depending on experimental design, MSCs cultured in 24-well chamber slides or 96-well chamber slides were fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100.No-specific binding was blocked with 5% normal goat serum in phosphate-buffered saline (PBS) for 1 h. Cells were stained, depending on experiment, with antibodies of h1-Calponin at 1:300 dilution, α-SMA at 1:400 dilution, desmin at 1:20 dilution, h2-calponin at 1:100 dilution, h3-calponin at 1:100 dilution, or SM-MHC at 1:50 dilution for 2 h at room temperature or 4°C overnight. Cells were then rinsed three times with PBS supplemented with 10% goat serum before being incubated with secondary antibodies. As secondary antibodies, AlexaFluor555 conjugated goat anti-rabbit IgG, AlexaFluor488 conjugated goat anti-mouse IgG or AlexaFlour488 conjugated were added at 1:400 dilutions. After three final washings, for nuclei staining and cells mounting, cells were covered with Prolong Gold anti-fade reagent with DAPI (Invitrogen, Oregon, USA) and were observed by a Nikon fluorescence microscope with a CCD video camera under identical excitation and exposure conditions (Nikon Eclipse Ti-S). For each experiment, negative controls were performed by omitting primary antibodies.

RNA extraction and reverse-transcriptase polymerase chain reaction

Total cellular RNA was extracted by Trizol (Invitrogen, USA) and reverse-transcribed into cDNA according to the manufacturer's instructions (TOYOBO, Japan). Real-time quantitative reverse transcription–polymerase chain reaction (RT–PCR) primers targeting rat α-SMA, h1-calponin (CNN1), h3-calponin (CNN3), SM heavy chain 1/2 and desmin, and mouse h2-calponin (CNN2) were designed by Primer express 2. 0 (sequences are listed in Table 1). All real-time RT–PCR assays were conducted using a MJ Opticon2 (Bio-Rad, USA) to determine the threshold cycle (Ct). The reactions were carried out with the standard reaction conditions. Determinations of the relative quantification of target gene expression levels (SMA, SMHC, desmin, CNN1, CNN2, CNN3) in comparison to a reference gene (GAPH) was made by the Ct deviation (ΔCt) method.

Table 1. Oligonucleotide primers for gene expression analysis by real time reverse transcription–polymerase chain reaction (RT–PCR)
GenePrimer sequences, 5′–3′Accession numberSize of amplified fragment, bp
  1. Fwd, forward; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Rev, reverse; SM-MHC, smooth muscle myosin heavy chain; α-SMA, α-smooth muscle actin. All of the primers were designed to target gene of rat except for primers of h2-calponin which targeted gene of mouse.

α-SMA

Fwd: GATAGAACACGGCATCATCACCA

Rev: GCAGGGACATTGAAGGTCTCAAA

NM_031004. 2183
h1-calponin

Fwd: CCTGTCTGCTGAGGTAAAGAACA

Rev: CCGGCTGGAGCTTGTTGATAAAT

NM_031747. 1170
h2-calponin

Fwd: ATGAGCTCCACGCAGTTCAACAA

Rev: CTTTCAGACCCTTCTGGAAGTCA

NM_007725. 2166
h3-calponin

Fwd: GGCTCTGTGAAGAAAGTCAACGA

Rev: GTAGTCTGAACCTGGGTCATGTT

NM_019359. 1160
desmin

Fwd: AGATGGCCTTGGATGTGGAGATT

Rev: CGGGTCTCAATGGTCTTGATCAT

NM_022531. 1178
SM-MHC

Fwd: CCAAGAACATGGACCCGCTAAAT

Rev: CGGAACATGCCCTTTTTGGTCTT

NM_001170600. 1175
GAPDH

Fwd: CCTCGTCTCATAGACAAGATGGT

Rev: GGGTAGAGTCATACTGGAACATG

NM_017008. 3169

Western blot essay

Western blots were performed as described previously (Hegner et al. 2005). Total proteins were extracted and the extraction was scraped into micro-tubes, centrifuged, and stored at −20°C until analysis.

Total protein content was determined with bicinchoninic acid (BCA) protein reagent. Equal amounts of protein were separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS–PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes. Then the PVDF membranes were incubated with a primary antibody overnight at 4°C. The secondary antibody (horseradish-peroxidase conjugated anti-rabbit/ mouse/goat IgG antibody) was added. After washing with PBS twice, electrochemiluminescence (ECL) system was used for the detection of protein bands. β-actin served as an internal control.

Statistical analysis

All experiments were carried out at least three times. Results of Western blots were quantified by densitometry using the software of Quantity One (Bio-Rad). Cell counting in immunofluorescence was done by Image-pro plus software or by manual counting. All values represented the mean ± SD. Comparison was performed using unpaired, double-sided t-test, t'-test or analysis of variance (anova), as appropriate. P < 0.05 was considered statistically significant. While in multiple groups' comparison, P-value was subjected to Bonferroni correction.

Results

Primary cultured cells exhibit characteristics of MSCs by morphology, surface markers and multilineage differentiation potential

The cultured cells showed spindle-shape single cell morphology in the first few days after plating. At 3–7 days, cell colonies formed and the size and number of colonies gradually increased. After the first passage, cells were propagated by conventional monolayer culture with homogeneous fibroblastoid morphology. Similarly, the cells maintained this morphology even at relatively high passages (passage 4) (Fig. 1A).

Figure 1.

Characteristics of primary cultured bone marrow mesenchymal stem cells (MSCs). (A) At day 6, colonies formed and cells appeared the spindle-shape morphology (left). Cells reached confluence at day 10 (middle), and they appeared conventional fibroblastoid morphology after passage 0 (right). (B) The immunophenotype of the cells were analyzed by fluorescence-activated cell sorting (FACS). The histogram showed positive expression for CD29, CD44, and negative expression of CD34 and CD45. The background signal of fluorescein isothiocyanate (FITC) and phycoerthrin (PE) was detected as negative control for the interested surface antigens. (C) Multilineage potential of the cells were verified when they differentiated into adipocytes stained with oil red (left), differentiated into mineralizing cells stained with alizarin red (middle), and differentiated into chondrocytic lineage cells stained with Alician blue (right).

By flow cytometric analysis, the cultured MSCs comprised a single phenotypic population. Consistent with prior studies, the cells were uniformly expressed CD29 (99. 93 ± 0.09% homogeneous), which is also named β1-integrin, and CD44 (99. 35 ± 0.97% homogeneous) as the receptor for hyaluronate and osteopontin. The cells were negative for markers of the hematopoietic lineage, including the lipopolysaccharide receptor CD34 (98. 65 ± 0.19% homogeneous) and the leukocyte common antigen CD45 (97. 75 ± 0.68% homogeneous) (Fig. 1B).

Like stem cells, MSCs are characterized by differentiation potential into tri-lineage of adipocytes, osteoblasts and chondrocytes. When the cells were cultured in adipogenic medium, lipid droplets were detectable after only 4 days, and reach a climax at day 14 (Fig. 1C left). In osteoinductive medium, cultured MSCs changed in cellular morphology from spindle-shape to cuboidal and meanwhile, accumulation of mineralized bone-like matrix was observed (Fig. 1C middle). The micromass culture technique was used to assess the chondrogenic differentiation. The development and accumulation of cartilage matrix was shown by staining the mucopolysaccharides (Fig. 1C right).

Resting BMMSCs express some of SMCs contractile proteins

Smooth muscle contractile proteins including α-SMA, h1-calponin, SM-MHC and desmin are specific to differentiated smooth muscle cells, consistent with a role in contractile function (Owens et al. 2004). To investigate the expression and intracellular localization of these markers in MSCs, immunofluorescent Staining was performed. As exhibited in Figure 2A, desmin protein resulting in fine reticular in cytoplasm were expressed in the majority of cells. Cells with h1-calponin diffusely expressing in cytoplasm could be widely detected, which constituted about 80% of the analyzed cells. α-SMA positive cells with well-organized thin SM α-actin filaments were detectable, accounted for around 15% of the analyzed cells. By contrast, SM-MHC expression was completely absent in resting MSCs. The proportions of positive cells for each protein were summarized in Figure 2B.

Figure 2.

Expression and intracellular localization of contractile proteins in resting mesenchymal stem cells (MSCs). (A) Immunofluorescent staining shows the expression of desmin, h1-calponin and α smooth muscle actin (α-SMA) in resting MSC, while expression of smooth muscle myosin heavy chain (SM-MHC) was absent. Desmin resulted in fine reticular was detected in the majority of cells. H1-calponin expression in cytoplasm was found in the bulk of cells. Cells with α-SMA expression along stress fibers could be detected, though at a low portion. Expression of SM-MHC was undetectable. (B) The proportions of immunofluorescence positive cells were analyzed to generally assess the expression level of each protein in the cells. (C) Expression level of contractile proteins in resting MSCs and SMCs from aorta was determined by western-blotting. *P < 0.05 and **P < 0.01 for comparison with control. bCaP, h1-calponin. Bar, 100 μm.

To corroborate the existence of the contractile markers expressed in resting bone marrow MSCs and to draw a comparison with SMCs from aorta, total protein of the cells were extracted and western-blotting was carried out. As exhibited in Figure 2C, desmin, h1-calponin and α-SMA were confirmed to be expressed in resting MSCs, whereas expression of SM-MHC was totally absent in MSC cells. A major issue of stem cell staining is the possible involvement of cell populations that are already differentiated along a certain lineage. To address this, we also examined the SMC markers in isolated MSCs at passage 0, at 3–5 days after initial culture when spindle-shape cells were first observed. The results mentioned above in immunofluorescent staining were proved to be reproducible (data not shown).

The expression of contractile proteins logically represents candidates in a relative mature state in the differentiation and development of SMCs. α-SMA is the earliest known marker of SMCs, while h1-calponin is regarded as a relatively late marker. Yet, our observation in MSCs was rather intriguing, for most of the resting MSCs expressed h1-calponin, while only a minority of cells expressed α-SMA. To further examine the difference in distribution of the two markers, double staining was used and the same analysis was carried out on SMCs from rat aorta serving as control. As depicted in Figure 3A, the dramatic distinction in expression levels between h1-calponin and α-SMA in resting MSCs was confirmed, for more than 50% of analyzed cells were h1-calponin +/ α-SMA −. Around 10% of cells were h1-calponin +/ α-SMA +. α-SMA expression was detected in h1-calponin positive cells as well as h1-calponin negative cells (as indicated by an arrow in Fig. 4A). By contrast, in SMCs from aorta, most cells co-expressed h1-Calponin and α-SMA. Roughly 10% of SMCs were α-SMA +/ h1-calponin −, while h1-calponin +/ α-SMA – SMCs were not detectable. The proportions of positive cells for α-SMA and h1-calponin in double-staining were summarized in Figure 3B.

Figure 3.

Double-staining of h1-calponin and α-smooth muscle actin (α-SMA) in resting mesenchymal stem cells (MSCs) and SMCs. (A) In resting MSCs, h1-calponin was expressed in most cells, while SMA positive cells were rarely found. SMA could be expressed in h1-calponin negative cells, as pointed out by the arrow, suggested that these two markers may indicate two distinct populations of resting MSCs. In comparison, SMCs from aorta identically co-express h1-calponin and α-SMA. (B) The proportions of positive cells for α-SMA and h1-calponin in double-staining. *P < 0.05 and **P < 0.01 for comparison with control. Bar, 100 μm.

Figure 4.

Expression and intracellular localization of h2-calponin and h3-calponin in resting mesenchymal stem cells (MSCs). (A) As members of the calponin family, both isoforms were widespread in the analyzed cells. H2-calponin was expressed in cytoplasm and seemed to accumulate around the nuclei, while h3-calponin was distributed along filaments in cytoplasm across the cells. (B) The proportions of positive cells are summed up. (C) Expression level of h2-calponin and h3-calponin in resting MSCs and smooth muscle actins (SMCs) from aorta was determined by western-blotting. Bar, 100 μm.

Besides h1 calponin, resting MSCs express h2 and h3 isoforms of calponin families

The relatively late markers H1 (basic) calponin was revealed to be expressed in bone marrow MSCs with multi-lineage potential by our study. To test whether MSCs were particularly rich in calponin expression, we investigated the expression of the remaining two known isoforms of calponin in MSCs, h2 (neutral)-calponin and h3 (acidic)-calponin. The expression and biological function of these two isoforms in muscle and non-muscle cells remains to be investigated. h2-calponin shows a broad tissue distribution pattern, including developing and remodeling smooth muscle (Hossain et al. 2003), endothelial cells (Tang et al. 2006), and culture human keratinocytes (Hossain et al. 2006). Expression of h3-calponin appears mostly restricted to neuronal tissues (Plantier et al. 1999), although it has also been detected in SMCs, myofibroblast and pancreatic cells. In our study, cellular immunofluorescence demonstrated that both h2-calponin and h3-calponin were expressed in resting bone marrow MSCs. Coinciding with previous observation, h2-calponin specifically located around the nuclei, a nuclei ring structure as previously described, and h3-calponin was expressed along filaments in a pattern similar with desmin (Fig. 5A). The proportions of fluorescence positive cells were summed up in Figure 5B. Immunoblotting showed that resting MSCs expressed h2-calponin and h3-calponin at similar levels in SMCs from aorta (Fig. 5C).

Figure 5.

Effects of high serum, low serum and co-culture with smooth muscle actins (SMCs) on smooth muscle myosin heavy chain (SM-MHC) expression in mesenchymal stem cells (MSCs). (A) By immunoflurescent staining, three induction conditions were proved to be able to give rise to SM-MHC expression, which was distributed along stress fibers in the cells. (B) The proportions of positive cells are summed up. (C) Expression level of SM-MHC in MSCs treated with three induction conditions was determined by western-blotting. Bar, 100 μm.

Under differentiation conditions, expression of SMA and SMMHC is promoted

Our results demonstrated that resting MSCs expressed desmin, h1-calponin and a low level of α-SMA. However, SM-MHC, which was recognized by a number of studies as the most restricted markers of differentiated SMC, was not detectable. We wondered if the expression level of these markers would change under conditions that induced stem cells to differentiate along smooth muscle lineage. Resting MSCs were cultured in low serum (medium with 1.5% FBS), high serum (medium with 20% FBS), or indirectly co-cultured with smooth muscle cells from aorta. Curiously, all three conditions proved to be able to initiate the expression of SM-MHC and enhance the expression of α-SMA. Yet, there was no significant difference on expression of desmin, h1-calponin, h2-calponin or h3-calponin between resting MSCs and induced MSCs (Fig. 6).

As conveyed in Figure 7, positive staining of specific SM-MHC antibody was detected in all of three induction conditions, distributing in a similar fashion to that in differentiated vascular SMCs. The increase in protein expression was widespread as immunofluorescent staining showed that roughly 75% of MSCs expressed SM-MHC after being treated with induction conditions for 6 days.

Figure 6.

Effects of three induction conditions on h1-calponin and α-smooth muscle actin (α-SMA) expression in mesenchymal stem cells (MSCs). (A, B) Immunoflurescent double-staining and statistical results revealed the enhancement of α-SMA expression in induced MSCs. The majority of induced cells co-expressed h1-calponin and α-SMA, a similar phenotype to that of SMCs in Figure 3A. (B) Expression level of α-SMA in MSCs treated with three induction conditions by western-blotting. Bar, 100 μm.

Figure 7.

Influence of induction stimulation on transcripts expression of primary cultured mesenchymal stem cells (MSCs). Three induction stimulations enhanced (α-smooth muscle actin (α-SMA) mRNA (left) and induced the expression of smooth muscle myosin heavy chain (SM-MHC) (right). Supplement. Expression of Calponin isoforms in embryonic stem cells (ESC), SMCs and MSCs cultured in different conditions.

Meanwhile, fluorescent double staining of h1-calponin and α-SMA was carried out to examine the modification of these markers. All three conditions brought about upregulation of α-SMA, but had no influence on expression of h1-calponin. As manifested in Figure 7, cells cultured in high serum, low serum and co-cultured with SMC were positively stained with α-SMA antibody and characteristically exhibited well-organized thin filaments across a broad area of the cells. Yet, in the high serum group, there was considerable variation in expression level of α-SMA among different experiments. When cultured in low serum and co-cultured with SMCs, MSCs exhibited some morphological changes, including turning flat (particularly apparent in low serum group) and expressing distinct thin filament distributed in cytoplasm. By comparison, cells cultured in high serum seemed to have smoother margins than low serum. In the control group, cells with well-organized α-SMA filaments made up only small portions and cells with diffuse non-filamentous α-SMA immunostaining in cytoplasm were detected. Immunofluorescence also showed expression of desmin in induced and non-induced cells, irrespective of their induction status. Another interesting finding was from analysis of h2-calponin and h3-calponin. Neither of these isoforms altered in expression level after being induced for 6 days, indicating that these proteins were not affected by SMCs lineage induction (data not shown).

Differentiated conditions enhance the transcripts expression of α-SMA and SM-MHC

In studying SMC phenotypes in response to cell culture conditions, the changes in gene expression patterns were proved to have occurred immediately, while alterations in actin and myosin content occurred relatively slowly (Chamley-Campbell et al. 1979). To further investigate the gene expression patterns of resting and induced MSCs, real-time quantitative RT–PCR was performed. The resting primary cultured MSCs expressed transcripts of desmin and α-SMA, as well as three isoforms of the calponin family. SM-MHC transcripts can be detected at the mRNA by RT–PCR. After being treated with the indicated differentiated condition for 6 days, α-SMA transcription was upregulated, especially in the co-culture group. SM-MHC, a SMC marker with the highest specificity, was enhanced in the treatment of induction condition at the transcript level. This effect on SM-MHC mRNA was again particularly remarkable in the co-culture group.

Discussion

It becomes clear that bone marrow MSCs are particularly prone to taking part in neointima growth by differentiating into vascular SMCs. In previous studies, the expressions of smooth muscle isoforms of contractile apparatus, such as α-SMA, h1-calponin, desmin and SM-MHC, logically served as evidence for identification of a differentiated SMC-like state. In our study, it was demonstrated that resting MSCs with multi-lineage potential express h1-calpnin, desmin, and a low level of α-SMA at both mRNA level and protein level, but SM-MHC is virtually absent at the protein level. Under differentiation conditions along SMC lineage, expression of SM-MHC is initiated and expression of α-SMA enhanced, while no significant difference is observed in expression of h1-calponin and desmin. Besides h1-calponin, the remaining known isoforms of calponin are verified to be expressed in MSCs, irrespective of their differentiation state.

Expression of SM-MHC has been extensively scrutinized by various studies from different laboratories and shows a high degree of specificity as an indicator for SMCs in both mature and developing organisms. By in situ hybridization analyses, SM-MHC expression was examined throughout development and maturation of whole mouse embryos and found no evidence for expression in cell types other than SMCs (Miano et al. 1994). The high degree of SMC specificity of expression of a −4.2 to 11.7 SM-MHC promoter enhancer reporter gene in transgenic mice was established using either direct measurement of a lacZ reporter transgene (Madsen et al. 1998) or a highly sensitive recombinase inducible system (Regan et al. 2000). SM-MHC may be the most rigorous marker for identification of differentiated SMC (Owens 1995). Confirming and extending the previous observations, our results demonstrated that smooth muscle markers including h1-calponin, desmin and α-SMA can be detected in multi-lineage MSCs, while the expression of SM-MHC was restricted to induced SMC-like MSCs.

Desmin is largely, but not exclusively, expressed in muscle cells, and its expression is developmentally regulated (Owens 1995). Desmin was shown to play an essential role in the maintenance of whole muscle tissue structural and functional integrity (Capetanaki et al. 1997). In regions of intimal thickening pointed to the presence of dedifferentiated smooth muscle cells, expression of desmin was remarkably downregulated, implying that its expression is associated with the contractile phenotype rather than with the synthetic phenotype (Yamada et al. 1997; Jimenez et al. 1999). Our present study revealed that MSCs with multi-lineage potential expressed desmin, and the expression levels didn't alter under induction conditions, suggesting that desmin may fail to distinguish a differentiated, SMC-like state of MSCs.

α-SMA is the most abundant of the actin isoforms in mature fully differentiated vascular smooth muscle. It is also the first known marker of differentiated SMC that is expressed during vasculogenesis (Owens 1995). However, α-SMA is not a definitive SMC marker for its wide variety in non-SMC cell types under certain circumstances (Owens et al. 2004). It is shown by our study that expression of α-SMA could be detected in about 10% of resting MSCs, and that the expression was upregulated in differentiated MSCs. As the most widely used SMC marker, the upregulation of α-SMA is, we believe, useful for representing the relative state of differentiation-maturation of MSCs.

h1-calponin expression appears to be restricted almost exclusively to smooth muscle, although it is also expressed in other cell types in the context of diverse pathological situations, such as epithelial cell sarcoma and mesangial glomerulonephritis (Owens 1995; Rozenblum & Gimona 2008). Of interest, our study demonstrated that h1-calponin was expressed in most of the resting MSCs with multilineage potential. Under induction conditions, the expression levels did not alter. Our observation is consistent with a recent study concerning the proteomic change on account of MSCs aging (Kasper et al. 2009). The effects of h1-calponin regulation on differentiation and cellular function in MSCs remain to be clearly elucidated, but it is evident that to simply treat it as a reliable marker for smooth muscle differentiation would be out of focus.

Double labeling of h1-calponin and α-SMA provided further evidence on the relation between smooth muscle contractile proteins and MSCs differentiation. In studying the development of SMCs during vasculogenesis, α-SMA is first detected in the presumptive SMCs and increases in its level of expression during development, making it the earliest known marker of differentiated SMCs (Barja et al. 1986). H1-calponin is, by comparison, a relatively late marker in development, which appears to be restricted almost exclusively to smooth muscle (Frid et al. 1992). Yet, it was observed in our study that in resting MSCs, most analyzed cells were h1-calponin +/ α-SMA −. Presumptive SMCs during development or synthetile SMCs are h1-calponin −/ α-SMA +, while mature SMCs are h1-calponin +/ α-SMA +. We figure that SMCs and MSCs may have different time courses of protein expression in maturation, therefore the developmental regulation of contractile protein expression in vascular SMCs may fail to adequately identify the state of differentiation in MSCs.

The observation that MSCs spontaneously expressed h1-calponin, desmin, as well as the rest isoforms in calponin family are thought-provoking and can prompt fresh thinking about the roles of these contractile-related proteins in MSCs. Basic to any understanding of SMC phenotypes is the remarkable plasticity of the cells in response to changes in local environmental cues. SMCs within adult blood vessels, regarded as contractile phenotype, proliferate at an extremely low rate, exhibit very low synthetic activity and express a unique repertoire of contractile proteins. These contractile proteins are specifically expressed in contractile SMCs, rather than their synthetic counterparts, thus are established as indicators for a relative state of maturation. But when it comes to other cell types, this criterion may, in our opinion, not be reasonably applied before it is extensively verified. The expression of h1-calponin is undisputedly reliable for the differentiated state in SMCs. Yet, pancreatic AR42J-B13 cells, which express all three calponin isoforms, downregulated h1-calponin expression and produced insulin in response to treatment of activin-A and hepatocyte growth factors, whereas overexpression of h1-calponin markedly suppressed differentiation (Morioka et al. 2003).

In our study, it is shown that all members of the calponin family were expressed in MSCs, irrespective of their differentiation state. As far as we are concerned, it is the first time for these isoforms to be reported in resting or induced MSCs. Calponin has been postulated to have both cytoskeletal and signaling function, although it awaits extensively experimental verification. H1-calponin has been investigated in detail and yet its functional role remains highly controversial. One theory is that h1-calponin may directly regulate contractility by inhibiting actomyosin ATPase activity of myosin heads cross-lined to actin (Takahashi et al. 1988; Winder & Walsh 1990). The other theory argues a qualitatively different functional mechanism of facilitating agonist-induced signal transduction and contractility by serving as a scaffold protein, which is based on the observation in vascular SMCs that h1-calponin acts as an adaptor protein to directly interact with the signaling proteins PKC (Leinweber et al. 2000) and ERK1/2 (Leinweber et al. 1999), then cotranslocates together with ERK1/2 and PKC to the cell cortex upon stimulation (Menice et al. 1997), in addition, is proved to strongly promote PKC activity (Leinweber et al. 2000). Mice lacking h1-calponin expression displayed enhanced ectopic bone formation in vivo, which is mediated by bone morphogenetic protein (BMP), suggesting that h1-calponin may serve as negative regulator of osteogenic program (Yoshikawa et al. 1998). Less than a handful of studies have investigated the properties and functional role of h2 (neutral)-calponin in smooth muscle and non-muscle cells. The significant expression of h2-calponin in growing smooth muscle and downregulation in quiescent smooth muscle cells indicate its involvement in the regulation of cell proliferation (Hossain et al. 2003). In a zebrafish model, h2-calponin plays a critical role in endothelial cell migration and wound healing and its expression is vital for proper vascular development (Tang et al. 2006). The function of h3 (acidic)-calponin has not been extensively investigated. Overexpression of h3-calponin in cultured hippocampal neurons results in an elongation of spines and an increase of density, indicating that h3-calponin regulates dendritic spine morphology and density, likely via regulation of the actin cytoskeleton reorganization and dynamic (Rami et al. 2006). Cell–cell fusion is an intriguing differentiation process, and h3-calponin is possibly a negative regulator of trophoblast fusion by regulating actin cytoskeleton rearrangement (Shibukawa et al. 2010). It is found that h3-calponin constitutively associates with both ERK1/2 and PKC α in vivo and cotranslocates to the cell cortex and podosome-like structures together with these two signaling molecules after stimulation with PDBu, and probably in this way regulates the migration in wound healing (Appel et al. 2010).

In this study, the expression of three isoforms of calponin was detected in rat bone marrow MSCs, in the presence or absence of SMC induction conditions. The spontaneous expression of calponin in resting MSCs made it fail to indicate smooth muscle differentiation of the cells, and the functional role of calponin in MSCs remains to be investigated. ERK signaling pathway has been implicated in mediating a diverse array of cellular functions including cell differentiation, proliferation, and inflammatory responses (Burkhard et al. 2009). Bone marrow MSCs differentiated into the smooth muscle lineage by blocking ERK signaling pathway, suggesting an anti-myogenic function of ERK pathway in cell differentiation of MSCs (Tamama et al. 2008). We speculate a possible character of calponin in regulating the cellular function of MSCs including cell differentiation, proliferation, and motility, on account of its role as a signal transducing and integrating molecule in actin cytoskeleton and ERK1/2 pathway, yet, it waits experimental verification.

In summary, this study demonstrates that in studying the differentiation of MSCs along smooth muscle lineage, the expression of SM-MHC and upregulation of α-SMA are relatively appropriate indicators of the state of differentiation-maturation. Desmin and three isoforms of calponin are shown to be expressed in resting MSCs and induced MSCs, with potential roles of regulating biological function of the cells, which are yet to be determined.

Acknowledgments

This work was supported by the National Science Foundation of China (NSFC) to Ruqiong Nie (30770899). We thank Yongqing Lin for providing mouse embryonic stem cells for control setting.

Ancillary