Effects of BMSCs interactions with adventitial fibroblasts in transdifferentiation and ultrastructure processes

Abstract

In this study an in vitro model of simulated blood vessel injury was used to study the effects of bone marrow-derived mesenchymal stem cells (BMSCs) morphology and to detect vascular smooth muscle actin (SM α-actin) expression in the presence of adventitial fibroblasts. BMSCs from rats with DAPI-labeled nuclei were co-cultured with adventitial fibroblasts for 7 days, while BMSCs cultured alone served as controls. Cell morphology of BMSCs was assessed by laser confocal microscopy and SM α-actin or calponin expression in BMSCs was detected by immunofluorescence staining. The expression of SM α-actin mRNA was identified using RT-PCR. Cell ultrastructure was assessed by electron microscopy. The results demonstrate that BMSCs with DAPI-labeled nuclei were smaller compared with fibroblasts, and their nuclei emitted a blue fluorescence. Most BMSCs displayed a polygonal shape changing from their original long fusiform shape. BMSCs with blue nuclei and red cytoplasm (SM α-actin positive or calponin positive) were observed, and a substantial number of filaments were present in the cytoplasm as observed under electron microscopy. The number of these cells increased as a function of culture duration. However, SM α-actin expression was weak and calponin expression was not detected in the control group. This study provides important new information on the characterization of artherosclerosis pathogenesis and vascular restenosis after blood vessel injury. Our findings demonstrate that direct interactions with adventitial fibroblasts can induce vascular smooth muscle-like cell differentiation in BMSCs.

Introduction

Cardiovascular diseases represent a leading cause of mortality. Vascular remodeling plays an important role in the pathology of many cardiovascular diseases, such as atherosclerosis and post-injury vascular restenosis. It has been suggested that the hyperplasia of vascular endothelial cells and mesolamella smooth muscle cells act as critical regulators involved with modulating the structure and function of the vasculature, while the action of vascular adventitia has long been neglected. Recent evidence has indicated that activation of the vascular adventitia, specifically the fibroblasts, contributes to the formation of intimal atherosclerotic lesion or vascular restenosis. Adventitial fibroblasts are activated early in the course of the injury and are involved in the formation of vascular restenosis. Additional factors associated with this process include SM α-actin expression, differentiation into myofibroblasts, migration into the intima and synthesis and release of cytokines and collagen, all of which appear to participate in the formation of vascular restenosis. Findings presented by Xu et al. from our laboratory [1,2] showed that early proliferative changes of adventitial fibroblasts are observed prior to the intima changes of atherosclerosis, and an expression of monocyte chemotactic protein-1 of vascular adventitia in the early stages of atherosclerotic inflammation has been detected. However, it has also been reported that some bone marrow-derived mesenchymal stem cells (BMSCs) participate in the adhesion and differentiation processes of vascular remodeling [3-5]. In addition, bioactive substances released after vascular injury along with BMSCs could enter the circulation at the site of vascular injury to differentiate into vascular smooth muscle cells and, in this way, participate in the development of hyperplasia [6,7]. In vascular injury, adventitial fibroblast secrete a large number of cytokines, chemokines, adhesion molecules in the early stages of injury, all of which can influence cell migration, proliferation, and differentiation [8-10]. Some of these factors may then play a role in stem cell mobilization.

Regulation of BMSCs differentiation is a complex process which is influenced by many factors. A prevailing consensus is that the surrounding microenvironment provides the main stage for this differentiation process. However, whether BMSCs differentiation requires contact with vascular adventitia has not been reported. In the current experiment, we simulated the microenvironment of vascular injury in vitro to assess the effect of direct contact with adventitial fibroblasts on transdifferentiation and ultrastructure of BMSCs. With this model it is possible to explore the mechanisms of BMSCs involvement in atherosclerosis and formation of vascular restenosis, as well as to evaluate the interaction between vascular adventitial fibroblasts and BMSCs. In addition, the findings generated from these experiments will provide the foundation for future work that can be applied for the study of pathogenesis of vascular restenosis and atherosclerosis after blood vessel injury.

Materials and methods

Animals

Male and female SD rats (90~100 g weight, N = 10) were provided by the experimental animal center of Binzhou Medical University. Rats were euthanized with an overdose of 3% pentobarbital anesthesia and then immersed and sterilized with 75% alcohol.

Isolation, cultivation and identification of BMSCs

Under aseptic conditions, bilateral femurs and tibias of rats were isolated and removed. Bone marrow was rinsed with DMED solution and centrifuged at 1,500 rpm for 5 min. The supernatant was discarded and the precipitate resuspended with PBS solution. Five ml of this solution was gradually applied onto 5 ml of Lymphocyte separation medium and centrifuged at 1,500 rpm for 5 min. The cells in the resultant middle white layer were carefully extracted, washed with PBS, and incubated in a humidified atmosphere of 5% CO₂ in air at 37°C in DMEM culture medium (containing 15% fetal bovine serum). Two days after the first exchange of medium, the cells of the suspension were discarded. The culture medium was changed every two days. The growth and morphology of cells were assessed daily with use of an inverted microscope. After reaching confluence, the cells were subcultured and the resultant ~2-3 generations of cells were used in the experiments. The cells were plated onto 25-mm cover slips for immunofluorescence. The cells were fixed in 4% paraformaldehyde, and then incubated with primary antibodies in blocking solution overnight at 4°C, washed, and incubated for 1h with secondary antibodies. BMSCs were identified by immunohistochemical staining using the primary antibodies CD105, CD34, CD44, SM α-actin (1:100) and calponin (a smooth muscle cell phenotypic marker protein of contraction used as markers for further differentiation of vascular smooth muscle cells). All antibodies were purchased from the Boster Company.

Isolation, cultivation and identification of vascular adventitial fibroblast

Under aseptic conditions, abdominal aortas of the rats were isolated, the adventitia was stripped and cut into small tissue blocks. Tissue blocks were incubated in a humidified atmosphere of 5% CO₂ at 37°C in DMEM culture medium (containing 15% fetal bovine serum). The medium was changed every two days until the cells covered the bottom of the bottle. The cells were subcultured and the resultant ~2-3 generations of cells, cultured on 25-mm cover slips, and used in the experiments. Vascular adventitial fibroblasts were identified by immunohistochemical staining using the primary antibodies Vimentin , Desmin and SM α-actin (1:100).

BMSCs co-cultured with vascular adventitial fibroblasts

After obtaining the ~2-3 generations of BMSCs, they were labeled with the nuclear dye DAPI (0.02 g/L fluorescent dye 4, 6 - acetyl 2-2 - phenyl indole, Santa Cruz Biotech) overnight, washed 10X with PBS solution, digested and resuspended. BMSCs and vascular adventitial fibroblasts were co-cultured by applying an identical density of 5 x 105/ml cells onto 6 orifice plates. An identical density of BMSCs (with no DAPI label) and vascular adventitial fibroblasts were mixed in culture bottles in preparation for electron microscopy observation. For comparison, untreated BMSCs were cultured with a density of 1 x 106/ml to serve as a control. Cells were cultured for 3, 5, or 7 days. Cells from each group were plated onto 25-mm cover slips for immunofluorescence staining using the primary antibody SM α-actin or calponin, and IgGCy3 (1:200 as fluorescence labeling sheep fight rabbit) as the second antibody. The cells were incubated with primary antibodies overnight at 4°C, washed 3 X in PBS and incubated with the secondary antibody in a 37°C incubator for 30 min. Cells of the control group were incubated with PBS solution in the absence of the primary antibody. Laser confocal microscopy was used for fluorescent analysis. The experiment was repeated three times. Ten different horizons of cells were randomly selected, and the rates of positive cell conversions were calculated.

SM-α-actin mRNA expression by reverse transcription PCR

Co-cultured cells were assessed at 3, 5 or 7 days post-culture. These cells were divided into three groups, and the expression of SM α-actin mRNA was measured by RT-PCR (n = 4). Extraction of total RNA was performed according to the Trizol reagent kit instructions and subsequently according to the two-step PCR detection kits manual of the TaKaRa Company. The sequences of primers used in this study were as follows:

SM α-actin forward: 5′-CGAGAAGCTGCTCCAGCTATGTG-3′, SM α-actin reverse: 5′-CTCTCTTGCTCTGCGCTTCGT-3′, β-actin forward: 5′-GGAGATTACTGCCCTGGCTCCTA-3′, β-actin reverse: 5′-GACTCATCGTACTCCTGCTTGCTG-3′.

The primers were synthesized by the Bao Biological Engineering Company. A reverse transcription PCR amplification system in a total of 12.5 μl were applied to the following PCR program: 5 min @ 94°C (pre denaturation), 30 s @ 94°C (initial denaturation), 30 s @ 58°C, 1min @ 72°C, repeated for 30 times, and 5 min @ 72°C (amplification). The PCR products were assessed on 1.5% agarose gels. The result of electrophoresis was scanned by a Gel imaging system for quantitative analysis and photographed.

Electron microscopy

Cells were digested with a pancreatic enzyme, centrifuged at 1,500 rpm for 10 min, and the supernatant discarded. The cells were then fixed with 2% glutaraldehyde at 4°C for 1.5 h; washed 3 X, fixed with 1% osmium tetroxide for 1 h, washed 3 X and dehydrated by gradient acetone. The cells were then embedded, ultra-thin sections sliced, dyed by osmic acid and photographed by JEM-2000-EX transmission electron microscopy.

Statistical analysis

Data was presented as the mean ± SEM. For determination of the significant differences, statistical analyses were performed using one-way ANOVA by SPSS software. Values of p < 0.05 will be regard as statistically significant.

Results

Morphology and immunofluorescence of BMSCs and fibroblasts

As assessed under inverted microscopy, the primary cultured BMSCs were found adherent to the culture bottle initially displaying short bar shapes (Figure 1A). Subsequently, BMSCs extended into a polygonal shape with diverse morphological characteristics. BMSCs were subcultured every 6-7 days. Eventually their shape gradually transformed into long spindles with an increasing nuclear size. After the ~2-3 generations, BMSCs grew robustly into a vortex shape. After ~2-3 d of culture, fibroblasts were transformed from the edges of adventitial tissue blocks with some showing adherent growth (Figure 1B). Primary cultured adventitial fibroblasts showed an irregular polygonal shape. Fibroblasts proliferated rapidly and showed a robust growth with a multilayer overlapping crest shape. The ~2-3 generations of BMSCs were cultured on 25-mm cover slips. Immunohistochemical assay with anti CD105 and CD44 revealed a rate of 90% BMSC staining (Figure 1C). With the use of anti SM α-actin, only weak staining of BMSCs was observed while anti CD34 and calponin were negative. BMSCs nuclei labeled with DAPI emitted a blue fluorescence at the rate of 100%. Staining of vascular adventitial fibroblasts with anti SM α-actin or calponin were both negative.

Figure 1.

Morphological observation and Immunohistochemical expression of BMSCs and fibroblasts. A. The primary cultured BMSCs (×100). B. Fibroblasts grown form blocks of vascular adventitia (×100). C. Immunohistochemical expression of CD44 in the 3rd generation of BMSCs (×100).

Co-cultured cells morphology and immunofluorescence

Co-cultured cells were observed under an inverted microscope. Gradually, two types of cells reached fusion showing a similar appearance, but were not easy to identify. After 7 days in culture, co-cultured and control cells achieved 70% fusion, but co-cultured cells lost their typical crest swirl appearance. Under confocal microscopy as assessed with different light sources, BMSCs appeared smaller, lost their typical long spindle shape and displayed an irregular polygonal shape. Their nucleus labeled with DAPI emitted a blue fluorescence and the cell cytoplasm of these BMSCs as treated with anti SM α-actin or calponin showed red fluorescence staining. The strongest positive response of the double-labeled cells treated with anti SM-α-actin was observed at 7 days with 83% demonstrating a red cytoplasmic and a blue nuclear stain (Figure 2A-C). This rate was only 31% in anti SM-calponin positive double-labeled cells (Figure 2D).

Figure 2.

Immunofluorescence expression of BMSCs. A-C. BMSC treated with anti SM α-actin showing a blue-stained nucleus and a red-stained cytoplasm after co-culture for 3, 5 or 7 d (×200). SM α-actin expressions of BMSCs positive rates were 33% at 3 d, 59% at 5 d, and 83% at 7 d. The number of cells increased as a function of culture duration. D. BMSCs treated with anti calponin showing a blue-stained nucleus and a red-stained cytoplasm after co-culture for 7 d (×200).

Analysis of SM α-actin mRNA expression

The PCR reaction was evaluated by each band of agarose gel absorbance and analysis of the grey PCR product by a gel imaging system (Figure 3A). The Grey values were calculated relative to β-actin. SM α-actin mRNA expression showed a statistically significant increase (P < 0.05) over the 7-day period. Values presented in Figure 3B represent mean + SD and statistical analyses reveal statistically significant differences among each of the three groups (P < 0.05) (Figure 3B).

Figure 3.

Expression of SM α-actin mRNA of co-cultured cells. A. Expressions of SM α-actin mRNA by RT-PCR (n = 4) at 3, 5 or 7 d of co-cultured cells. B. SM α-actin mRNA expression increased as a function of culture duration with statistically significant differences being obtain among each of the three groups (P < 0.05).

Observation of electron microscope

Ultrastructure of BMSCs was assessed under electron microscopy. In the control group, the nuclei of BMSCs were small, irregularly shaped and, morphologically diverse with ~1-2 nucleoli. A large amount of glycogen was present in the cytoplasm, but fewer organelles such as the ribosomes, rough endoplasmic reticulum and Golgi apparatus were observed and no myofilament structure was found in the cytoplasm of these BMSCs. When BMSCs were co-cultured with fibroblasts for 7 days, the cytoplasmic glycogen of BMSCs was substantially reduced. But cytoplasmic organelles were more developed (Figure 4A). In addition, some scattered myofilament structural formations were present in the cytoplasm (Figure 4B).

Figure 4.

Ultrastructure of BMSCs under electron microscopy. A. Cytoplasmic organelles of BMSC after co-culture were more developed. B. Presence of myofilaments in the cytoplasm of BMSC after co-culture for 7 d (×20000).

Discussion

Vascular remodeling after injury, in particular atherosclerosis and restenosis, represents a major issue associated with cardiovascular diseases [11]. Vascular smooth muscle and adventitia play an important role in the proliferation of vascular remodeling. In recent years, the significance of vascular adventitial fibroblasts in vascular restenosis has been recognized. Adventitial fibroblasts are activated in the early phase of vascular injury and can change their phenotype to myofibroblasts. These myofibroblasts show characteristics of smooth muscle with actin in plasma and migrate toward the lumen neointimal where they are involved in the proliferation of vascular remodeling [12,13]. However, observations resulting from stem cell research and extensive arterial interventional surgery have indicated that the proliferation of smooth muscle cells do not appear to originate completely from the middle layer and adventitia. Findings presented by Sata et al. [7] indicated that bone marrow-derived stem cells are involved in the formation of neointimal lesions after vascular injury, and these cells express SM α-actin. Hu et al. [14] reported that many progenitor cells contained stem cell markers such as sca-1 and c-kit as observed in arterial wall defects of apoE mice. Moreover, findings from Abedi’s laboratory showed that BMSCs existed in normal arterial walls and capillaries and these BMSCs that entered the circulation were involved with circulating progenitor cells and perfused other tissues [15]. Recently, Hoshino found that stem cells existed in vascular adventitia, and cultured human vascular adventitia had characteristics of stem cells [16]. Xu Q et al. [17,18] reported that progenitor cells existed in blood and blood vessels, and these cells were involved in the formation of atherosclerosis.

Recently, the relationship between stem/progenitor cells and vascular disease has received a considerable amount of attention as the source of these stem cells, their expression, differentiation and dysfunction, may be critically involved in the occurrence and development of cardiovascular diseases. BMSCs have the potential for multiple differentiations [19,20]. Under physiological conditions, the stem/progenitor cells exist in bone marrow and blood vessels in an undifferentiated state. With blood vessel injury, some cytokines and chemokines can mobilize these stem cells into the circulation for local perfusion of the injury site [21-23]. It has been reported that BMSCs have the capacity to differentiate into vascular endothelial and smooth muscle cells as demonstrated in vitro and in vivo [24,25]. Although the mechanisms through which stem cells differentiate into smooth muscle remain unclear, results from several studies have indicated that various signal transductions and signaling pathways are involved in driving this differentiation process [26-28]. It has also been demonstrated that signal transducers can function as enhancers or inhibitors of stem cells differentiation toward the smooth muscle cells lineage [29-31]. Therefore, further work on the mechanisms through which stem/progenitor cells differentiate to endothelial and smooth muscle cells and the cell signal transduction pathways involved will provide critical information required for understanding vascular remodeling and atherosclerosis and serve as the basis for new strategies in the treatment of vascular diseases.

In this experiment, the micro-environment of BMSCs capacity for direct contact with vascular adventitial fibroblasts was simulated in vitro. Under such conditions, the morphology and differentiation of BMSCs were observed to change into vascular component cells. An understanding of this relationship between adventitial hyperplasia and BMSCs differentiation is of considerable importance, because it will serve as a foundation for new protocols to study vascular restenosis and identify new targets to treat various cardiovascular diseases.

In this experiment, the goal of these co-cultures was to enable us to induce cell interactions among cytokines, paracrine, contact reactions and gap junctions and, in this way, observe their differentiation and phenotypic transformations. To accomplish this goal, we used purified BMSCs that were labeled with DAPI, mixed with vascular adventitial fibroblasts for co-culture and compared these with purified control BMSCs. Cytoplasmic immunofluorescence staining of these cells was then performed after 7 days in culture. In the control group a weak expression of SM α-actin and no expression of calponin was observed, but co-cultured BMSCs showed a strong positive expression of SM α-actin, and a positive expression of calponin at 7 days. In addition, SM α-actin mRNA was significantly up-regulated in these cells over this 7 day period. Ultrastructure of BMSCs as revealed by transmission electron microscopy (sem) showed that a substantial amount of glycogen was present in the cytoplasm of the BMSCs control groups. After co-culture, BMSCs in this phase of regulatory morphogenesis with plasticity and glycogen decreased significantly. Some scattered myofilament structures were found in BMSCs after co-culture. Such findings suggest that when BMSCs are in direct contact with vascular adventitial fibroblasts, they differentiate and promote the expression of smooth muscle-like properties in this microenvironment.

Based upon the results of the present experiment, it can be postulated that large amounts of bioactive substances are released by vessel injury and that these substances, which include cytokines or superoxides, may not only activate fibroblast proliferation and secretion, but also induce vascular smooth muscle-like cell differentiation in BMSCs. In this way, BMSCs and fibroblast may both participate in the proliferation of vascular remodeling through processes involving differentiation, migration and phenotype transformation as resulting from their interactions. At present, the exact role of BMSCs in the repairing process of vascular injury remains unclear. It has been reported that BMSCs accelerate endothelial repair and inhibit vascular smooth muscle hyperplasia [32,33]. However, it remains critical to elucidate further the mechanisms of repair and angiogenesis and to identify means of promoting beneficial angiogenesis while inhibiting smooth muscle hyperplasia. Such information will provide a critical step to prevent and treat the vascular adventitia remodeling of the arteries and further clarify BMSCs role in cardiovascular disease.

It should be noted that this experiment has some limitations resulting from the many physical and chemical factors in the environment which can influence cellular differentiation. Moreover, cellular differentiation needs to be assessed within different models to provide a comprehensive analysis of the many factors that may contribute to this effect.

Disclosure of conflict of interest

None.