Inhibitory effects of growth differentiation factor 11 on autophagy deficiency-induced dedifferentiation of arterial smooth muscle cells

Abstract

Growth differentiation factor (GDF)11 has been reported to reverse age-related cardiac hypertrophy in mice and cause youthful regeneration of cardiomyocytes. The present study attempted to test a hypothesis that GDF11 counteracts the pathologic dedifferentiation of mouse carotid arterial smooth muscle cells (CASMCs) due to deficient autophagy. By real-time RT-PCR and Western blot analysis, exogenously administrated GDF11 was found to promote CASMC differentiation with increased expression of various differentiation markers (α-smooth muscle actin, myogenin, myogenic differentiation, and myosin heavy chain) as well as decreased expression of dedifferentiation markers (vimentin and proliferating cell nuclear antigen). Upregulation of the GDF11 gene by trichostatin A (TSA) or CRISPR-cas9 activating plasmids also stimulated the differentiation of CASMCs. Either GDF11 or TSA treatment blocked 7-ketocholesterol-induced CASMC dedifferentiation and autophagosome accumulation as well as lysosome inhibitor bafilomycin-induced dedifferentiation and autophagosome accumulation. Moreover, in CASMCs from mice lacking the CD38 gene, an autophagy deficiency model in CASMCs, GDF11 also inhibited its phenotypic transition to dedifferentiation status. Correspondingly, TSA treatment was shown to decrease GDF11 expression and reverse CASMC dedifferentiation in the partial ligated carotid artery of mice. The inhibitory effects of TSA on dedifferentiation of CASMCs were accompanied by reduced autophagosome accumulation in the arterial wall, which was accompanied by attenuated neointima formation in partial ligated carotid arteries. We concluded that GDF11 promotes CASMC differentiation and prevents the phenotypic transition of these cells induced by autophagosome accumulation during different pathological stimulations, such as Western diet, lysosome function deficiency, and inflammation.

NEW & NOTEWORTHY The present study demonstrates that growth differentiation factor (GDF)11 promotes autophagy and subsequent differentiation in carotid arterial smooth muscle cells. Upregulation of GDF11 counteracts dedifferentiation under different pathological conditions. These findings provide novel insights into the regulatory role of GDF11 in the counteracting of sclerotic arterial diseases and also suggest that activation or induction of GDF11 may be a new therapeutic strategy for the treatment or prevention of these diseases.

INTRODUCTION

Growth differentiation factor (GDF)11, a transforming growth factor-β superfamily member, is highly homologous to myostatin (30), which has been proposed as a “mythical fountain of youth” circulating factor because it can reverse age-related dysfunction in mouse cardiac (23), neural (30), and skeletal muscle (35) cells. Some clinical studies have revealed that higher GDF11 levels are associated with a lower risk of cardiovascular events and death, suggesting the GDF11 pathway as a potential new target to improve adverse cardiovascular outcomes (6, 31). Vascular smooth muscle cells (VSMCs) undergo a remarkable phenotypic transition during the development and progression of vascular diseases such as atherosclerosis (33). However, the role of GDF11 in atherosclerotic differentiation and phenotypic transition of VSMCs remains poorly understood.

Recent studies have indicated that autophagy is critical for cellular homeostatic processes in a variety of mammalian cells or tissues (15, 21, 38). Through autophagic processes, unnecessary or dysfunctional components within cells are degraded and/or recycled, maintaining their structural and functional integrity (33). It is well known that the autophagic process begins with the induction and formation of autophagosomes, which involves the sequestration of cytoplasmic components in double-membrane vesicles. Autophagosomes then fuse with lysosomes to form autophagolysosomes (APLs) and, in turn, digest the autophagic contents (51), which is referred to as autophagic flux. With respect to the regulation of the autophagic process by GDF11, Sinha et al. (35) reported that in uninjured skeletal muscle of aged mice treated with GDF11, the basal levels of autophagosome markers, as assessed by the ratio of autophagic intermediates light chain (LC)3B-II over LC3B-I, were increased, suggesting that GDF11 enhances autophagy and related mitochondrial biogenesis, regulating cellular remodeling of muscle fibers in the aging mouse model. However, there is also evidence that GDF11 may not have any antiaging effects in different animal models or cells (9, 37). It appears that the role of GDF11 in antiaging or tissue regeneration is far from clear. In particular, little is known as to whether or how GDF11 may be implicated in the regulation of vascular cell degeneration or regeneration, such as VSMCs, under different physiological and pathological conditions.

More recently, Li and coworkers (22, 42) demonstrated that CD38, a multifunctional enzyme as a cell differentiation marker in leukemic cells or other cells, mediates the production and metabolism of cADP-ribose and nicotinic acid adenine dinucleotide phosphate in many kinds of mammalian cells, including VSMCs, which play important roles in the regulation of vascular functions. It has also been reported that CD38 regulates lysosome trafficking and fusion to autophagosomes and thus determines the fate of the autophagosome controlling autophagic flux in VSMCs (44, 51). CD38 deficiency (CD38^−/−) led to autophagosome accumulation, which promoted a phenotype transition and proliferation of coronary VSMCs by accelerating cell cycle progression through the G₂/M phase (2). Other lysosome function inhibitors or disruptors were also found to cause autophagosome accumulation and phenotypic transition of coronary VSMCs to be more dedifferentiated or degenerative (3). In addition, autophagy, as an important regulator of cell differentiation and proliferation, has been well documented in the literature (27, 34). However, so far, it remains poorly understood whether the phenotypic transition induced by autophagy deficiency is regulated endogenously and whether there are any therapeutic interventions that can suppress this autophagy-associated degenerative process during inflammatory or atherogenic vascular diseases.

The present study was designed to address the effects of exogenous GDF11 on VSMC dedifferentiation in cultured cells and a mouse model with autophagy deficiency and test the hypothesis that GDF11 counteracts the pathological dedifferentiation of mouse carotid arterial smooth muscle cells (CASMCs) due to deficient autophagy. We first performed cell experiments to examine the effects of GDF11 administration and induction on VSMC dedifferentiation and autophagic flux upon stimulations of atherosclerotic stimuli, such as 7-ketocholesterol (7-Ket) and the CD38^−/− gene. We then addressed the effects of trichostatin A (TSA), a GDF11 inducer, on phenotypic transition and neointima formation in the carotid artery using partial ligated carotid arteries (PLCAs) from CD38^−/− mice fed a Western diet (WD). Our results demonstrated that increased GDF11 promotes CASMC differentiation and prevents dedifferentiation by a reduction of autophagosome accumulation induced by atherosclerotic stimuli.

MATERIALS AND METHODS

Isolation and culture of mouse CASMCs.

CASMC isolation from mice has been previously described (1). Briefly, 2% isoflurane was used to anesthetize the mice. The carotid arteries were then removed and put into PBS on ice. The adventitia was removed from the artery using angled forceps under a microscope. The tissue was washed three times with PBS and cut into pieces that were ~1–2 mm using microdissecting scissors in a cell culture hood. The dissected tissues were washed two to three times with cell culture medium and then added into a cell culture dish without medium for 2 h. Fresh medium was then added into the culture dish, and the tissues were incubated in a humidified 37°C, 5% CO₂ incubator. FBS (10%) and DMEM supplemented with 2% antibiotics were used to culture the tissues. CASMCs were isolated when they grew out from the dissected tissue. After 5–10 days, CASMCs were cloned by the selection of those cells from cell-growing islands in the dish. The identification and purity of CASMCs in culture were confirmed as previously described (45). Cells were used at passages 4–10 for in vitro experiments.

Cell transfection.

Cells (1.5 × 10⁵) were prepared in a six-well plate for cell transfection. Transfection of GDF11 CRISPR-cas9 activation plasmid (sc-420524-ACT, Santa Cruz Biotechnology, Dallas, TX) was performed with the UltraCruz transfection reagent (sc-395739, Santa Cruz Biotechnology) according to the manufacturer’s instructions. Cells were collected at different times to extract RNA or protein.

Western blot analysis.

Western blot analysis was used to analyze specific protein, as previously described (43). Briefly, treated cells were placed on ice and washed with ice-cold PBS. PBS was aspirated, and 60 μl lysis buffer with protein inhibitor was added to the dish. Cells were scraped off and transferred into a cold microcentrifuge tube. The suspension was maintained in ice for 30 min and then centrifuged at 25,000 g for 15 min at 4°C. The supernatant was collected, and the concentration was measured. Protein (20 μg) was loaded into the wells of a 12% SDS-PAGE gel at a voltage of 100 V for 2–3 h. Protein was transferred from the gel into the membrane at a voltage of 100 V for 1 h in the cold room. Nonfat milk (5%) in Tris-buffered saline-Tween 20 (TBST) buffer was used to block the membrane for 1 h at room temperature. The blot was incubated with primary antibodies against vimentin (1:5,000, Abcam, Cambridge, MA), smooth muscle 22α (SM22α; 1:5,000, Abcam), proliferating cell nuclear antigen (PCNA; 1:2,000, Abcam), p62 (1:2,000, Abcam), or GDF11 (1:500, BD Biosciences, San Jose, CA) overnight at 4°C. The blot was rinsed three times for 10 min with TBST and then incubated with a second antibody labeled with horseradish peroxidase for 1 h at room temperature. The blot was rinsed three times for 10 min with TBST and developed with Odyssey FC Imaging (LI-COR, Lincoln, NE). Anti-β-actin antibody (1:20,000, Santa Cruz Biotechnology) was used to probe this housekeeping gene expression as a control. ImageJ 6.0 [National Institutes of Health (NIH), Bethesda, MD] was used to quantify the intensity of a specific protein.

mRNA expression analysis.

mRNA expression was detected by real-time RT-PCR as previously described (46). Briefly, TRIzol RNA isolation reagents (ThermoFisher Scientific, Waltham, MA) were used to extract total RNA according to the manufacturer’s protocol. RNA (1 μg) was synthesized to cDNA with iScript cDNA synthesis kits (Bio-Rad, Hercules, CA). PCR was conducted on a Bio-Rad 96 Real-Time PCR System in a 20-μl final reaction mixture using optimized SYBR Green I quantitative PCR solution for monitoring the products. The reaction conditions were as follows: preincubation at 95°C for 2 min and then 45 cycles of 15 s at 95°C and 1 min at 60°C. β-Actin was used to normalize the expression of mRNA. Fold differences were calculated as follows: 2^{−ΔΔ threshold cycle}. The primers used in this study were synthesized by Operon, and the sequences were as follows: β-actin, sense 5′-TCGCTGCGCTGGTCGTC-3′ and antisense 5′-GGCCTCGTCACCCACATAGGA-3′; GDF11, sense 5′-GTCGCCTAGAGGCATCAAGT-3′ and antisense 5′-CCCAGTTAGGGGTTTCAGRCGT-3′; vimentin, sense 5′-CGGAAAGTGGAATCCTTGCA-3′ and antisense 5′-CACATCGATCTGGACATGCTGT-3′; myosin heavy chain (MyHC), sense 5′-AAGCGAAGAGTAAGGCTGTC-3′ and antisense 5′-CTTGCAAAGGAACTTGGGCTT-3′; myogenic differentiation (MyoD), sense 5′-GTGGCAGCGAGCACTACAGT-3′ and antisense 5′-ACACAGCCGCACTCTTCCCT-3′; myogenin, sense 5′-GCACTGGAGTTCGGTCCCAA-3′ and antisense 5′-TATCCTCCACCGTGATGCTG-3′; SM22α, sense 5′-TCCAGTCCACAAACGACCAAGC-3′ and antisense 5′-GAATTGAGCCACCTGTTCCATCTG-3′; and PCNA, sense 5′-CAGAGCTCTTCCCTTACGCA-3′ and antisense 5′-GTCCTTGAGTGCCTCCAACA-3′.

Mice.

CD38 wild-type (CD38^+/+) and CD38 knockout (CD38^−/−) C57BL/6J male mice (8–12 wk old) were used in the present study. All mice were maintained in an environmentally controlled room (25°C and 40~50% humidity) with a 12:12-h light-dark cycle. Both CD38^+/+ and CD38^−/− mice were randomly separated into four different experimental groups and fed with a normal diet (ND) or WD for 3 wk. PLCA surgery was performed in the fourth week, and mice on the ND or WD were injected intraperitoneally with TSA (58880-19-6, Cayman Chemical, Ann Arbor, MI; 0.5 mg/kg ip every day) for another 3 wk. All procedures were carried out in accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of Virginia Commonwealth University.

PLCA model for neointima.

The surgery of PLCA as a neointima formation model was carried out as in previous reports in our laboratory (18, 40) and others (20). Briefly, 2% inhaled isoflurane was used to anesthetize mice for 5 min, and the anesthesia was continued through a nose cone. The neck of the mouse was shaved, and the area was disinfected with betadine solution followed by a wipe with a 70% alcohol swab. A sterile drape was used to maintain the area clean. A midline incision (1–2 cm) was made, and the left carotid artery was exposed. The carotid arteries (external, internal, and occipital) were tightly ligated with 6.0 silk suture but the superior thyroid artery was left to serve as the sole source for blood circulation. The nonligated right carotid artery was used as an internal control. The incision was closed with 5.0 silk suture and disinfected with betadine solution. Mice were then temporarily placed on a heating pad until they recovered from anesthesia. After 3 wk of PLCA, mice were euthanized and perfused with 4% paraformaldehyde for the collection of carotid arteries. Part of the artery was frozen in liquid nitrogen for dual-fluorescence staining and confocal microscopic analysis. Part of the artery was stocked in 10% formalin to prepare wax slides for immunohistochemistry.

Immunofluorescence staining.

Cells cultured on cover slides or the carotid artery on frozen slides were washed two to three times for 5 min with PBS and then fixed in 4% paraformaldehyde in PBS for 10–15 min on ice. Samples were rinsed two to three times in PBS for 10 min and permeabilized with 0.1% Triton X-100 in PBS for another 10 min. After being washed two to three times for 10 min with PBS, samples were incubated with primary antibody for 2 h or overnight at 4°C. Samples were then rinsed three times for 10 min in PBS and incubated with secondary antibody, labeled with either Alexa 488 or Alexa 555 for 1 h at room temperature in the dark room. Finally, mounting medium with DAPI and nail polish was used to mount and seal the slides after they had been washed with PBS. Pictures were taken by a confocal laser scanning microscope (Fluoview FV1000, Olympus, Tokyo, Japan). The staining intensity of the cell or tissue was measured and analyzed with ImageJ 6.0 (NIH). The colocalization of a lysosomal transmembrane marker, lysosome-associated membrane protein 1 (Lamp-1; 1:200, Abcam) with light chain (LC)3B (1:200, Abcam) or p62 (1:200, Abcam), was detected with double staining and measured with Image-Pro Plus software (version 6.0, Media Cybernetics, Bethesda, MD). Pearson correlation coefficients were used to summarize the colocalization, as previously described (41).

Immunohistochemistry.

Paraffin sections of carotid arteries were heated for 30 min at 65°C. Deparaffinization was performed in xylene for 10 min, and hydration was carried out in graded ethanol (100%, 95%, and 75%). After antigen retrieval in retrieval buffer (pH 6) at 98°C for 15 min, 3% H₂O₂ in methanol was used to quench endogenous peroxidase activity. Sections were blocked with 2.5% horse serum for 1 h at room temperature. Incubation with primary antibodies [SM22α (1:1500, Abcam), vimentin (1:1500, Abcam), and GDF11 (1:100, BD Biosciences)] was performed overnight at 4°C. After being rinsed three times with PBS, sections were incubated with biotinylated secondary antibodies and streptavidin-peroxidase complex for 20 min at room temperature. Sections were then sequentially developed with diaminobenzidine solution for 5 min. Finally, sections were counterstained in hematoxylin for 5 min, dehydrated in graded ethanol (75%, 95%, and 100%), and mounted with a mixture of distyrene, plasticizer, and xylene (4).

Morphological examination and medial thickening analysis.

Hematoxylin and eosin staining was performed as previously described (47) to study the morphological changes. Briefly, carotid arteries were collected and immersed in 10% neutral buffered formalin for over 48 h. The formalin-fixed carotid artery was embedded in paraffin and then cut into 5- to 7-μm serial sections for histopathological evaluation. For hematoxylin and eosin staining, sections were deparaffinized using dimethylbenzene and dehydrated with 100%, 95%, and 75% ethanol to water. Sections were then immersed in hematoxylin and hydrochloride alcohol. As soon as the color turned blue, sections were stained with eosin. After that, sections were rinsed with running water and dehydrated with different grades of ethanol. Finally, the slides were mounted with a mixture of distyrene, plasticizer, and xylene, and pictures were taken under the microscope. Intimal thickening was examined using Image-Pro Plus 6.0 software (Media Cybernetics).

Statistics.

Data are presented as means ± SD. Values were analyzed for significant differences between and within multiple groups using ANOVA for repeated measures followed by Duncan’s multiple-range test. Significant differences between the two groups of experiments were examined using Student’s t-test. Statistical analysis was performed with Sigma Plot 12.5 software (Systat Software, San Jose, CA). Statistical significance was defined at P < 0.05.

RESULTS

Exogenously administrated GDF11 promoted CASMC differentiation and phenotypic transition.

Since the phenotypic transition of VSMCs is a hallmark of vascular inflammation and atherosclerosis, which may be initiated by arterial endothelial dysfunction and injury, we first analyzed the gene expression associated with the differentiation and dedifferentiation of CASMCs isolated from wild-type (CD38^+/+) mice to determine the basic effect of exogenously administrated GDF11 on their differentiation. By real-time RT-PCR, we found that GDF11 treatment markedly increased RNA expression of CASMC differentiation marker genes such as SM22α, MyHC, myogenin, and MyoD but decreased the expression of dedifferentiation marker genes, including vimentin and PCNA (Fig. 1A). With the use of Western blot analysis, we observed that GDF11 treatment significantly increased the protein expression of CASMC differentiation markers SM22α and calponin but decreased the expression of dedifferentiation markers, including vimentin and PCNA (Fig. 1, B–F). These results suggest that GDF11 stimulates a CASMC phenotypic transition to a more differentiated, mature, and contractile status.

Fig. 1.

Exogenous growth differentiation factor (GDF)11 promoted carotid arterial smooth muscle cell (CASMC) differentiation and phenotypic transition. In vitro, primary CASMCs isolated from the carotid artery in normal mice were used. A: mRNA expression of differentiation markers [smooth muscle (SM)22α, myosin heavy chain (MyHC), myogenin, and myogenic differentiation (MyoD)] and dedifferentiation markers [vimentin and proliferating cell nuclear antigen (PCNA)] after GDF11 treatment (100 ng/ml, 24 h) detected by RT-PCR. B: representative Western blot analysis showing the effects of GDF11 on the protein expression of differentiation markers (SM22α and calponin) and dedifferentiation markers (vimentin and PCNA) in CASMCs treated with different doses of GDF11 (0–100 ng/ml, 24 h). C–F: summarized data showing the effects of GDF11 on the protein expression of differentiation markers (SM22α and calponin) and dedifferentiation markers (vimentin and PCNA) in CASMCs treated with different doses of GDF11. Data are expressed as means ± SD; n = 5. *P < 0.05 vs. the control (Ctrl) group.

GDF11 gene activation and induction promoted CASMC differentiation and phenotypic transition.

We also examined the effect of GDF11 gene induction or activation on CASMC phenotypic transition by overexpression of the GDF11 gene using GDF11-specific CRISPR-cas9 activating plasmids or treated cells with a known GDF11 inducer, TSA (5, 11, 14, 48, 53). We that GDF11-specific CRISPR-cas9 activating plasmids stimulated GDF11 expression in CASMCs, as detected by real-time PCR and Western blot analysis (Fig. 2, A and B). This upregulation of the GDF11 gene, via gene editing, induced the differentiation of CASMCs, as shown by the remarkable decrease in vimentin but increase in SM22α (Fig. 2, C and D). We also observed that the GDF11 gene inducer TSA induced GDF11 overexpression in CASMCs, as detected by real-time PCR and Western blot analysis (Fig. 3, A, C, and D). This GDF11 overexpression induced the differentiation of CASMCs with a significant decrease in the ratio of vimentin to SM22α, as shown by the results from real-time RT-PCR (Fig. 3B). We also found that TSA induced a decrease in vimentin but induced an increase of SM22α by Western blot analysis (Fig. 3, C, E, and F). It is clear that induction or activation of the GDF11 gene in CASMCs also enhances CASMC differentiation.

Fig. 2.

Endogenous growth differentiation factor (GDF)11 induced carotid arterial smooth muscle cell (CASMC) differentiation and phenotypic transition. A: effect of GDF11-specific CRISPR-cas9 activating plasmids on GDF11 RNA expression by RT-PCR (n = 5). B: effects of GDF11-specific CRISPR-cas9 activating plasmids on GDF11 protein expression by Western blot analysis (n = 3). C and D: time-dependent changes of vimentin and smooth muscle (SM)22α RNA expression induced by GDF11-specific CRISPR-cas9 activating plasmids (n = 5). Data are expressed as means ± SD. *P < 0.05 vs. the CRISPR control (Ctrl) group.

Fig. 3.

Endogenously produced growth differentiation factor (GDF)11 (GDF11) by trichostatin A (TSA)-induced carotid arterial smooth muscle cell (CASMC) differentiation and phenotypic transition. A: dose-dependent effects of TSA on GDF11 mRNA expression by RT-PCR (n = 5). B: dose-dependent effects of TSA on the ratio of vimentin to smooth muscle (SM)22α by RT-PCR (n = 5). C: representative Western blot analysis showing the dose-dependent effects of TSA on GDF11, vimentin, and SM22α protein expression. D: summarized data showing the dose-dependent effects of TSA on GDF11 protein expression (n = 5). E: summarized results showing the dose-dependent effects of TSA on expression of vimentin. F: summarized data showing the dose-dependent effects on the expression of SM22α (n = 5). Data are expressed as means ± SD. *P < 0.05 vs. the control (Ctrl) group.

GDF11 and TSA reversed CASMC dedifferentiation and autophagosome accumulation induced by the lysosome function inhibitor.

Because VSMCs possess remarkable phenotypic transition and autophagosome accumulation during atherosclerosis, we tested the effect of GDF11 on CASMC dedifferentiation and autophagosome accumulation induced by a proatherogenic factor (7-Ket) or by suppression of lysosome function with bafilomycin (Baf), which has been reported to inhibit lysosome function and disrupt autophagic flux. By Western blot analysis, 7-Ket treatment was found to remarkably increase the expression of vimentin but decrease the expression of SM22α (Fig. 4, A–C, and Fig. 5, A–C), producing a CASMC phenotypic transition into dedifferentiation status. 7-Ket treatment also significantly increased autophagosome accumulation, as shown by the large increase in p62 protein (Fig. 4, A and D, and Fig. 5, A and D). However, these increases were almost completely attenuated by GDF11 (Fig. 4, A–D) or TSA treatment (Fig. 5, A–D). Similarly, both GDF11 and TSA markedly attenuated the Baf-induced phenotypic transition and autophagosome accumulation in CASMCs (Fig. 4, A–D, and Fig. 5, A–D).

Fig. 4.

Growth differentiation factor (GDF)11 reversed carotid arterial smooth muscle cell (CASMC) dedifferentiation and autophagosome accumulation induced by autophagy inducer or lysosome function inhibitor. GDF11 (100 ng/ml), 7-ketocholesterol (7-Ket; 5 μg/ml), and bafilomycin (Baf; 5 nM) were used to treat CASMCs for 24 h. A: representative Western blot analysis showing the expression of vimentin, smooth muscle (SM)22α, and p62. B: summarized data showing the expression of vimentin. C: summarized data showing the expression of SM22α. D: summarized data showing the expression of p62. Data are expressed as means ± SD; n = 5. *P < 0.05 vs. the vehicle (Vehl)-control (Ctrl)-DMSO group; #P < 0.05 vs. the Ctrl-DMSO group; &P < 0.05 vs. the Ctrl-Baf group.

Fig. 5.

Trichostatin A (TSA) reversed carotid arterial smooth muscle cell (CASMC) dedifferentiation and autophagosome accumulation induced by autophagy inducer or lysosome function inhibitor. TSA (20 ng/ml) was used to treat CASMCs for 24 h. A: representative Western blot analysis showing the expression of vimentin, smooth muscle (SM)22α, and p62. B: summarized data showing the expression of vimentin. C: summarized data showing the expression of SM22α. D: summarized data showing the expression of p62. Data are expressed as means ± SD; n = 5. *P < 0.05 vs. the vehicle (Vehl)-control (Ctrl)-DMSO group; #P < 0.05 vs. the Ctrl-DMSO group; &P < 0.05 vs. the Ctrl-bafilomycin (Baf) group. 7-Ket, 7-ketocholesterol.

GDF11 reversed dedifferentiation during the suppression of autophagic flux induced by CD38^−/−.

CD38 gene deletion causes defective autophagosome trafficking, which, in turn, impairs autophagic flux in VSMCs (51). We further confirmed the effects of GDF11 on CASMC dedifferentiation during stimulation of autophagosome generation or suppression of lysosome function by the CD38^−/− gene. We found that deletion of the CD38 gene increased the expression of vimentin but decreased the expression of SM22α compared with CD38^+/+ cells, even under the normal conditions without proatherogenic stimulations, indicating that CD38^−/− changes the CASMC phenotype to a more dedifferentiated status. In CASMCs from CD38^−/− mice, 7-Ket stimulation enhanced the dedifferentiated status, whereas GDF11 substantially blocked the occurrence of this dedifferentiated status (Fig. 6A). Figure 6, B and C, shows the summarized data, clearly demonstrating that GDF11 reversed the CASMC phenotypic transition into dedifferentiated status induced by CD38^−/− or 7-Ket stimulation. These results further suggest that the inhibitory effect of GDF11 on the phenotypic transition of CASMCs is associated with lysosome function and autophagic flux.

Fig. 6.

Effects of growth differentiation factor (GDF)11 on dedifferentiation in carotid arterial smooth muscle cells (CASMCs) induced by CD38 gene deficiency (CD38^−/−) . A: representative Western blot analysis showing the expression of vimentin and smooth muscle (SM)22α as CASMC markers. B: summarized data showing the expression of vimentin. C: summarized data showing the expression of SM22α. Data are expressed as means ± SD; n = 5. *P < 0.05 vs. the vehicle (Vehl)-control (Ctrl)-CD38 wild-type (CD38^+/+) group; #P < 0.05 vs. the Ctrl-CD38^+/+ group; &P < 0.05 vs. the Ctrl-CD38^−/− group. 7-Ket, 7-ketocholesterol.

Inhibitory effects of TSA on the neointima formation in the PLCA model.

The findings from our cell experiments presented above strongly link the effects of GDF11 on the CASMC phenotypic transition to its action to improve autophagic flux or lysosome function. To confirm these findings further and study their translational relevance, we tested the role of GDF11 induction in vascular injury or VSMC phenotypic transition in a mouse model with PLCA and WD. It was observed that intimal thickening was much more in the carotid artery wall of CD38^−/− compared with CD38^+/+ mice, even when they were on the ND. The WD markedly increased neointima formation in both CD38^−/− and CD38^+/+ mice. It seems that CD38-mediated mechanisms participate in WD-induced neointima formation. Interestingly, we found that TSA significantly inhibited neointima formation in CD38^+/+ or CD38^−/− mice. In particular, it was clear that TSA prevented the enhancement of neointima formation induced by CD38 gene deletion and WD (Fig. 7A). Quantification of the ratio of the intima versus media is shown in Fig. 7B, clearly demonstrating that TSA decreased the ratio of the intima versus media under normal condition or with WD.

Fig. 7.

Effects of trichostatin A (TSA) on neointima formation and growth differentiation factor (GDF)11 expression in the partial ligated carotid artery (PLCA) mouse model. A: representative hematoxylin and eosin staining showing neointima formation in the PLCA wall. Black arrowheads represent the medial area; white arrowheads represent the intima area. B: quantification analysis of the ratio between intima and media of the arteries in PLCA. C: representative immunohistochemistry staining showing the expression of GDF11 in the carotid artery of PLCA CD38 wild-type (CD38^+/+) and CD38 gene-deficient (CD38^−/−) mice treated with a normal diet (ND) or Western diet (WD). D: summarized data showing the quantification of GDF11 relative expression by the calculation of the percentage of positive staining in the artery wall. L, lumen side. Data are expressed as means ± SD; n = 5. *P < 0.05 vs. the ND-control (Ctrl)-CD38^+/+ group; #P < 0.05 vs. the Ctrl-CD38^+/+ group; &P < 0.05 vs. Ctrl-CD38^−/− group.

Effects of TSA treatment on the expression of GDF11 in PLCA mice.

To confirm the effects of TSA on GDF11 expression, the expression of GDF11 in the ligated carotid artery wall (disease model) was detected in both CD38^+/+ and CD38^−/− mice. As shown in Fig. 7, C and D, in CD38^+/+ PLCA mice, the WD significantly decreased the expression of GDF11 in the carotid arterial wall, whereas TSA increased the expression of GDF11 under the ND condition and attenuated the effects of the WD on GDF11 expression. Especially, we found that CD38 gene deletion also decreased the expression of GDF11 under the ND or WD and that TSA could recover it in PLCA CD38^−/− mice. These results suggest that GDF11 expression in the arterial wall was reduced by the WD, and, in particular, the decrease in GDF11 expression even occurs in CD38^−/− mice without WD challenge. Based on these results, we believe that the CD38 gene and its product importantly control GDF11 expression and that stimuli or challenges such as the WD or CD38 gene deletion may cause a reduction in GDF11, leading to related pathological changes. Therapeutic interventions, such as TSA, to increase GDF11 expression may have a counteracting effect on pathological changes, such as neointimal hyperplasia, as described above.

Inhibitory effects of TSA on the CASMC phenotypic transition in PLCA mice.

Corresponding to the effects of TSA on neointima formation, we found that in carotid arteries with partial ligation from CD38^+/+ mice, the WD markedly increased the expression of vimentin but decreased the expression of SM22α. TSA significantly attenuated the increase in vimentin expression in carotid arteries with partial ligation from CD38^+/+ mice on the ND or WD but blocked the decrease in SM22α with the WD (Fig. 8, A and C). In CD38^−/− mice, increased vimentin but decreased SM22α expression in ligated carotid arteries was observed compared with CD38^+/+ mice, even though the mice were on the ND. Interestingly, TSA was found to significantly attenuate this increase in vimentin and decrease in SM22α expression induced by CD38 gene deletion (Fig. 8, A–D). These in vivo data further confirmed that TSA inhibits the CASMC phenotypic transition associated with disturbed autophagic flux in the medial layer of the arterial wall.

Fig. 8.

Inhibitory effects of trichostatin A (TSA) on the phenotypic transition of carotid arterial smooth muscle cells (CASMCs) in the partial ligated carotid artery (PLCA) model of CD38 wild-type (CD38^+/+) and CD38 gene-deficient (CD38^−/−) mice on a normal diet (ND) or Western diet (WD). A: representative immunohistochemistry staining for vimentin. B: percentage of positive vimentin staining in the artery wall in PLCA. C: typical representative immunohistochemistry staining for smooth muscle (SM)22α. D: quantitative analysis of SM22α staining in PLCA represented by the calculation of the percentage of positive staining in the artery wall. Data are expressed as means ± SD. *P < 0.05 vs. the ND-control (Ctrl)-CD38^+/+ group; #P < 0.05 vs. the Ctrl-CD38^+/+ group; &P < 0.05 vs. the Ctrl-CD38^−/− group.

Inhibitory effects of TSA on APL accumulation in the carotid arteries with PLCA.

With the use of the confocal microscope, we also examined the effects of TSA on APL aggregation or accumulation in carotid arteries with PLCA from CD38^+/+ and CD38^−/− mice. In CD38^+/+ mice, TSA was found to significantly decrease the colocalization of Lamp-1 versus LC3B-II or Lamp-1 versus p62 in the carotid arterial medial layer with partial ligation, even though the mice were on the ND (Fig. 9, A–D). In CD38^−/− mice, however, the colocalization of Lamp-1 versus LC3B-II or Lamp-1 versus p62 was much lower in the carotid arterial wall with PLCA; this may be due to the CD38^−/− gene and, consequently, decrease in the fusion of autophagosomes with lysosomes. It is possible that GDF11 promotes CASMC differentiation and prevents phenotypic transition of these cells by improvement of lysosome function, increasing autophagic flux and reducing autophagosome accumulation in these cells.

Fig. 9.

Inhibitory effects of trichostatin A (TSA) on autophagosome accumulation of carotid arterial smooth muscle cells (CASMCs) in the partial ligated carotid artery model of CD38 wild-type (CD38^+/+) and CD38 gene-deficient (CD38^−/−) mice on a normal diet (ND) or Western diet (WD). A and C: representative fluorescent confocal microscopic images showing the colocalization of lysosome-associated membrane protein 1 (Lamp-1) with p62 or light chain 3B (LC3B). B and D: summarized data showing the fold changes in the Pearson correlation coefficient (PCC) for the colocalization of Lamp-1 with p62 or LC3B-II (n = 5). Data are expressed as means ± SD. *P < 0.05 vs. the ND-control (Ctrl)-CD38^+/+ group; #P < 0.05 vs. the Ctrl-CD38^+/+ group.

DISCUSSION

The present study demonstrates that both exogenously administrated GDF11 and induction of GDF11 genes promoted CASMC differentiation and reversed their phenotypic transition both in vitro and in vivo under pathological conditions. In cultured CASMCs, exogenously administrated GDF11 increased the expression of differentiation markers such as SM22α, MyHC, myogenin, and MyoD and decreased the expression of dedifferentiation markers such as vimentin and PCNA. In addition, GDF11-specific CRISPR-cas9 activating plasmids and the GDF11 gene inducer TSA increased GDF11 production and induced differentiation of CASMCs, as shown by the remarkable decreases in mRNA expression of vimentin but increases in expression of SM22α. These results indicate that both exogenously administrated and endogenously induced GDF11 enhance CASMC differentiation. Moreover, 7-Ket and Baf induced dedifferentiation of CASMCs, as shown by the increased expression of vimentin associated with increased autophagosome accumulation, as shown by the increase in p62 levels, which were blocked by GDF11 or TSA. In CASMCs from CD38^−/− mice, the dedifferentiation status, with or without 7-Ket stimulation, was reversed by GDF11. We also demonstrated in animal experiments that deletion of the CD38 gene, along with the WD, resulted in decreased GDF11 expression and phenotypic transition in CASMCs in the carotid arterial wall from the mouse PLCA model, which was suppressed by GDF11 induction with TSA. All of these in vitro and in vivo experiments strongly suggest that GDF11 may mainly promote CASMC differentiation and prevent a phenotypic switch of these cells induced by various pathological stimuli.

GDF11 is a myostatin (GDF8) homologous protein from the transforming growth factor-β superfamily, which plays a role in the inhibition of myogenesis, muscle cell growth, and differentiation (30). Although VSMCs undergo remarkable phenotypic switching during the development and progression of vascular diseases such as atherosclerosis, the effects of GDF11 on the VSMC phenotypic switch are still poorly understood. In the present study, GDF11 was found to promote CASMC differentiation and decrease its pathological phenotypic transition induced by proatherogenic stimulation or lysosome function inhibitors, suggesting that GDF11 importantly controls vascular plasticity under physiological and pathological conditions. Although there are few reports about its role in the regulation of vascular function or plasticity, GDF11 has been shown to have a potent action on cell differentiation from different organs or tissues. In this regard, Lu et al. (25) reported that in bone marrow mesenchymal stem cells under in vitro conditions, GDF11 inhibited osteoblastic differentiation via Smad2/3 signaling, and it also inhibited bone formation and decreased bone mass in mice. Zhang et al. (49) reported that GDF11 treatment increased osteoblastogenesis but inhibited adipogenesis from bone marrow mesenchymal stem cells in vitro by inhibiting the activity of peroxisome proliferator-activated receptor-γ. In some studies on cardiac plasticity, restoration of more youthful levels of systemic GDF11 have been reported to reverse age-related cardiac hypertrophy (23). It has also been demonstrated that intraperitoneal injection of recombinant GDF11 increased satellite cell frequency and function in muscles of GDF11-treated mice. These satellite cells act as precursors to skeletal muscle cells and give rise to differentiated skeletal muscle cells (35). In contrast, many other studies have indicated that GDF11 may have no effect or even negative effects on aging-related cardiovascular diseases and muscle dysfunction (50). The debate on the role of GDF11 in aging and cardiovascular diseases remains unclear, and more studies are needed.

In some vascular studies, reports about the effects of GDF11 on endothelial cells have also been controversial. In a study by Katsimpardi et al. (17), GDF11 was found to increase the proliferation of primary brain capillary endothelial cells by 22.9% in the presence of VEGF and EGF, enhance cell migration in culture media without FBS (13), and strengthen the viability of human umbilical endothelial cells (52). However, several studies have shown that GDF11 had no significant effects on these endothelial functions or behaviors (13, 26, 52). So far, the effects of GDF11 on VSMCs remain unclear. The present study performed a systematic analysis of various SMC differentiation markers, such as α-smooth muscle actin, myogenin, MyoD, and MyHC, as well as their dedifferentiation markers, vimentin and PCNA. It was found that both exogenous administration of GDF11 and upregulation of the GDF11 gene by TSA or specific CRISPR-cas9 activating plasmids remarkably stimulated the differentiation of CASMCs and substantially blocked 7-Ket-, Baf-, and CD38 gene deletion-induced CASMC dedifferentiation. Interestingly, this action of GDF11 was found to antagonize the CASMC dedifferentiation induced by autophagosome accumulation and lysosome dysfunction under these pathological conditions. It is clear that GDF11 serves as an important factor in the promotion of SMC differentiation and prevention of their pathological phenotypic transition, which is associated with its counteracting action on blunted autophagic flux or autophagosome accumulation. To our knowledge, these findings, for the first time, elucidate the role of GDF11 in the regulation of CASMC phenotype properties in connection with autophagy and, in particular, autophagic flux.

We also carried out animal experiments to validate our findings using a PLCA mouse model. We found that TSA induced GDF11 and decreased vimentin expression accompanied by increased SM22α in the carotid artery of PLCA in CD38^−/− mice. Moreover, TSA also decreased neointima formation in both WD-fed CD38^−/− and CD38^+/+ mice. These results demonstrated that induction of the GDF11 gene by TSA treatment importantly contributed to enhanced CASMC differentiation that prevents neointima formation in the PLCA mouse model treated with WD, which is consistent with previous reports showing that adenovirus vector-transfected GDF11 (16) and recombinant GDF11 reduced atherosclerosis in apolipoprotein E^−/− mice (26). These WD-induced pathological changes in the arterial wall are associated with hyperlipidemia and locally produced cytokines. With respect to the effects of TSA, Findeisen et al. (12) also demonstrated that it reduced SMC proliferation via repression of cyclin D1 and PCNA and suppressed neointima formation in response to vascular injury in vivo (32). Several studies in other tissues have shown that histone deacetylase inhibitors, including TSA, induced growth arrest and the associated reduction in tumor growth in various species. All of these results support that TSA promotes SMC differentiation and reduces neointima formation by limiting SMC dedifferentiation. Although previous studies have proposed that the effects of TSA are potentially attributed to epigenetic alterations in SMC phenotypic transition or neointimal lesions (7, 39), the mechanisms that mediate the TSA action remain controversial. The present study confirmed that TSA upregulated GDF11 expression, which may contribute to its role in the regulation of CASMC plasticity. Despite no evidence in the vascular system, this TSA action, through GDF11, has also been reported in many studies on other tissues or cells under different conditions (5, 11, 14, 48, 53). It should be noted that due to the high cost for long-time chronic experiments, the present study did not attempt to test the role of direct administration of recombinant GDF11. Locally transfected GDF11 in the carotid arterial wall to observe its action on SMC dedifferentiation will be done shortly.

One of the more important findings in the present study is that the actions of GDF11 in the control of CASMC differentiation and counteraction on phenotypic transition upon pathological stimulations are associated with its regulation of autophagy in these cells, in particular, the autophagic flux process. It is well known that autophagy is a cellular homeostatic process that degrades unwanted or dysfunctional components of cells for recycling or removal (2, 10). Defective autophagy promotes the proliferation and dedifferentiation of VSMCs, contributing to the generation and deterioration of atherosclerosis (8, 19). In this regard, Sinha et al. (35) demonstrated that uninjured muscles in aged mice showed strikingly improved myofibrillar and mitochondrial morphology upon GDF11 treatment, which was associated with increased basal levels of autophagosome (macroautophagy) markers, assessed as the ratio of autophagic intermediates LC3B-II over LC3B-I. Recently, our laboratory reported that autophagy impairment upon CD38 gene deletion or lysosomal dysfunction led to p62 accumulation, which, in turn, resulted in dedifferentiation and proliferation of CASMCs. Accumulation of p62 caused compromised ubiquitinylation and degradation of cyclin-dependent kinase 1 and enhanced G₂/M cell phase progress during the cell cycle (2). These results showed that dysfunctional autophagy might result in a phenotype transition of VSMCs in response to pathogenic stimulations. In this context, the present study demonstrated that TSA, a GDF11 inducer, or GDF11 gene activation reversed the CASMC phenotypic transition from synthetic to contractile status by the decrease of autophagosome accumulation in both in vivo and in vitro experiments. This action of GDF11 on CASMCs, through the enhanced autophagic process, is similar to those reported in the myocardium and other tissues or cells (23, 35, 54).

The present study mainly focused on the effects of GDF11 on autophagy deficiency and associated dedifferentiation but did not yet attempt to define the mechanisms by which autophagosome accumulation affects cell dedifferentiation. In the literature, however, the role of autophagy has been reported, and several mechanisms have been proposed to regulate cell differentiation. One of the major mechanisms is the role of accumulated p62 in autophagosomes, which may serve as a signaling hub regulating cell proliferation, differentiation, and many other activities. Lysosome dysfunction or deficient autophagic efflux may lead to the p62 increase in cells, which activates different regulatory pathways, such as NF-κB, nuclear factor (erythroid-derived 2)-like 2, and cyclin-dependent kinase-mediated signaling or regulatory mechanisms, thereby resulting in changes in cell differentiation (28, 29).

In conclusion, the present study demonstrated that GDF11 promoted CASMC differentiation and prevented phenotypic transition of these cells by the reduction of autophagosome accumulation induced by different pathological stimuli. Our results may define a novel regulatory mechanism mediating VSMC differentiation and plasticity associated with autophagy.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-122937, HL-057244, and HL-075316.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.-L.L. conceived and designed research; X.Y. and H.L. performed experiments; X.Y. analyzed data; X.Y. interpreted results of experiments; X.Y. prepared figures; X.Y. and O.W.B. drafted manuscript; X.Y., O.W.B., H.L., N.L., Y.Z., and P.-L.L. edited and revised manuscript; P.-L.L. approved final version of manuscript.