Vascular Smooth Muscle Cell Apoptosis Promotes Transplant Arteriosclerosis Through Inducing the Production of SDF-1α

Authors


Zifang Song, zsongmd@gmail.com

Abstract

Transplant arteriosclerosis is a leading cause of late allograft loss. Medial smooth muscle cell (SMC) apoptosis is considered to be an important event in transplant arteriosclerosis. However, the precise contribution of medial SMC apoptosis to transplant arteriosclerosis and the underlying mechanisms remain unclear. We transferred wild-type p53 to induce apoptosis of cultured SMCs. We found that apoptosis induces the production of SDF-1α from apoptotic and neighboring viable cells, resulting in increased SDF-1α in the culture media. Conditioned media from Ltv-p53-transferred SMCs activated PI3K/Akt/mTOR and MAPK/Erk signaling in a SDF-1α-dependent manner and thereby promoted mesenchymal stem cell (MSC) migration and proliferation. In a rat aorta transplantation model, lentivirus-mediated BclxL transfer selectively inhibits medial SMC apoptosis in aortic allografts, resulting in a remarkable decrease of SDF-1α both in allograft media and in blood plasma, associated with diminished recruitment of CD90+CD105+ double-positive cells and impaired neointimal formation. Systemic administration of rapamycin or PD98059 also attenuated MSC recruitment and neointimal formation in the aortic allografts. These results suggest that medial SMC apoptosis is critical for the development of transplant arteriosclerosis through inducing SDF-1α production and that MSC recruitment represents a major component of vascular remodeling, constituting a relevant target and mechanism for therapeutic interventions.

Abbreviations: 
BM

bone marrow

CM

conditioned medium

CXCR4

CXC chemokine receptor 4

DAPI

4′6-diamidino-2-phenylindole

ECs

endothelial cells

HSCs

hematopoietic stem cells

Ltv

lentivirus

MSCs

mesenchymal stem cells

mTOR

mammalian target of rapamycin

PDGF-BB

platelet-derived growth factor-BB

PI3K

phosphoinositide-3 alpha/kinase

SDF-1α

stromal cell-derived factor-1alpha

SMCs

smooth muscle cells

SM22α

smooth muscle protein 22α

TA

transplant arteriosclerosis

TGF-β

transforming growth factor-beta

Introduction

Solid-organ transplantation has become the gold standard therapy for patients suffering from end-stage organ failure (1). The short-term outcome of organ transplantation has been improved considerably in recent decades because of remarkable advances in immunosuppressive agents that largely overcome acute rejection. Nonetheless, the long-term expectations of clinical organ transplantation have not been achieved, mainly due to chronic allograft rejection, also referred to as transplant arteriosclerosis (TA) (2,3). TA is histopathologically characterized by diffuse neointimal formation in graft vasculature that primarily results from excessive accumulation of vascular smooth muscle cells (VSMCs) and massive deposition of extracellular matrix (4–6). Although both alloantigen-dependent and alloantigen-independent factors are identified as risk factors for TA, its pathogenesis is thought to be alloimmune-mediated injury to the graft vessels (4,6). The damage of vascular wall cells, mainly endothelial cells (ECs) and VSMCs, initiates a generalized repair process, leading to neointimal formation with lumen occlusion and eventually loss of graft function. This process was formerly considered to be primarily directed by the proliferation and migration of VSMCs within the vessel wall (7). However, recent evidence suggests that bone marrow (BM)-derived cells contribute to neointimal formation by giving rise to neointimal cells (8–10). It has been demonstrated that BM hematopoietic stem cells (HSCs) (11) and mesenchymal stem cells (MSCs) (8,12) migrate to the sites of injury and actively participate in the repair and remodeling of the injured vessel wall. Furthermore, stromal cell derived-factor 1α (SDF-1α) has been shown to be induced in injured vascular wall and contribute to the remodeling process through the recruitment of vascular progenitor cells (13,14).

In line with endothelial injury in the graft vasculature (15), allogeneic graft VSMCs constitute another target of alloimmune attack that initiates damage and apoptosis of vascular wall cells (16,17). Various alloimmune-mediated pathways have indeed been implicated in mediating medial SMC apoptosis, which is principally responsible for the massive loss of medial SMCs in the graft vasculature undergoing TA (18,19). Although it has been observed that apoptosis of medial SMCs precedes neointimal formation in allografts after transplantation in several animal models (20–22), its direct contribution to the vascular remodeling process and the underlying mechanisms remain unclear.

In this study, we generated a recombinant lentiviral vector carrying the anti-apoptotic gene BclxL, whose selective expression in VSMCs is modulated by a minimal SM22α promoter (23). Our results indicated that selective inhibition of medial SMC apoptosis diminishes the recruitment of MSCs within the vascular lesions and attenuates neointimal formation and lumen stenosis in aortic allografts, probably owing to reduced SDF-1α production in the vessel wall. In vitro studies confirmed that VSMC apoptosis induces SDF-1α production from both apoptotic and neighboring viable cells and that promotes the migration and proliferation of MSCs via activating PI3K/Akt/mTOR and MAPK/Erk signaling. These data define a critical role for medial SMC apoptosis in the pathogenesis of TA and implicate medial SMC apoptosis as a promising therapeutic target to modulate the vascular remodeling process following transplantation.

Materials and Methods

Expanded materials and methods are available in the supplemental material.

Animals

All of the animal protocols were approved by the Animal Care and Use Committee of Tongji Medical College. Inbred male Sprague-Dawley and Wistar rats and male BALB/c mice were purchased from the Center of Experimental Animals (Tongji Medical College, Huazhong University of Science and Technology).

Lentiviral vectors production

Recombinant lentiviral vectors carrying a specific SM22α promoter and the downstream rat BclxL (Ltv-BclxL) or p53 genes (Ltv-p53) fused to enhanced green fluorescence protein (EGFP) were generated as described previously (24,25).

Aortic transplantation, lentivirus infection and pharmaceutical treatment

Sprague-Dawley rats were used as donors and syngeneic recipients, and Wistar rats served as allogeneic recipients, as described by Ji et al. (26). Abdominal aortic transplantation was performed using a modified technique described initially by Mennander et al. (27). For lentivirus infection, Ltv-BclxL and Ltv-vector were used to infect aortic allografts before transplantation. For pharmaceutical treatment, rapamycin (1.5 mg/kg/day) and PD98059 (10 mg/kg/day) were administrated to allogeneic recipients via intraperitoneal injection, respectively. Control animals received the appropriate vehicle injections. Treatment started on day 11 after surgery and lasted for 3 weeks.

Histology and morphometry

Two and 8 weeks after transplantation, the aortic grafts were harvested. Serial cross-sections (5 μm thick) were cut. Six cross-sections from each animal were selected at 250-μm intervals for hematoxylin-eosin and Masson staining, followed by morphometric analysis.

Immunostaining

Immunohistochemistry and immunofluorescence were performed on cultured VSMCs and MSCs, as well as paraffin-embedded sections of arteries, using the avidin–biotin peroxidase complex method.

TUNEL assay

Apoptosis of VSMCs was detected using an In Situ Cell Death Detection Kit (Roche Diagnostics, Mannheim, Germany).

Cell culture

Primary VSMCs were isolated from the descending thoracic aorta of 4-week-old Sprague-Dawley rats, using an explant method (28). All experiments were conducted on cells between passages 3 and 8. Rat BM MSCs were obtained from femurs of Sprague-Dawley rats as previously described with some modifications (29).

Lentiviral infection

VSMCs were incubated with Opti-MEM I medium (Invitrogen) containing 200 multiplicity of infection (MOI) per cell of Ltv-vector or Ltv-p53 in the presence of polybrene (5 μg/ml).

Preparation of conditioned medium

Conditioned media (CM) from Ltv-p53-infected VSMCs (Ltv-p53 CM) or Ltv-vector-infected VSMCs (Ltv-vector CM) were prepared. Conditioned medium from uninfected VSMCs (uninfected CM) was prepared as a control.

MSC migration and proliferation assay

MSC migration was performed using transwell chambers containing 5-μm pore filters (Corning, Inc., New York, NY, USA). MSC proliferation was quantified using a BrdU Labeling and Detection Kit (Roche Applied Science, Penzgerg, Germany).

Quantitative RT-PCR

RNA was isolated from cultured SMCs or rat aortic grafts using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). RNA expression was analyzed by RT-PCR using iQ SYBR Green Supermix in an iCycler Real-Time PCR Detection System (Bio-Rad, Munchen, Germany).

Western blotting

Western blotting was used to detect the expression of p53, TGF-β, PDGF-BB and SDF-1α in cultured VSMCs, as well as the expression of signal molecules in cultured MSCs.

Statistics

All data were presented as mean ± SEM. After demonstration of homogeneity of variance with Bartlett test, one-way ANOVA, followed by Student–Newman–Keuls test where appropriate, was used to evaluate the statistical significance. Values of p < 0.05 were considered statistically significant.

Results

VSMC apoptosis induces the production of SDF-1α

To induce cell apoptosis in vitro, we transferred wild-type p53 gene into cultured VSMCs by lentivirus. Cell apoptosis was evaluated by DAPI staining 72 h after infection. The condensed or fragmented nuclei of apoptotic cells were noted in Ltv-p53-infected cells, but not Ltv-vector-infected cells (Figure 1A). In accordance with DAPI staining, flow cytometry analysis showed that p53 transfer induced a significant increase in both early and late apoptosis of VSMCs (Figure 1B).

Figure 1.

VSMC apoptosis induces the production of SDF-1α. Apoptosis of VSMCs was induced by wild-type p53 transfer mediated by lentivirus. (A) Apoptosis induction was determined by DAPI staining 72 h after infection. Arrowheads represent the condensed or fragmented nuclei of apoptotic cells. Scale bars: 50 μm. (B) Quantification of apoptotic cells was carried out by flow cytometry after Annexin V-APC/PI staining. Representative plots of Annexin V-APC/PI flow cytometry from three independent experiments are presented. The number represents the percentage of cells in each quadrant. p53 overexpression induced a significant increase in both early (Q4) and late (Q2) apoptosis. (C) Expression of mRNA for SDF-1α, TGF-β and PDGF-BB was determined by quantitative RT-PCR in infected and uninfected VSMCs. mRNA levels are normalized to GAPDH. (D) Western blot analysis demonstrated increased SDF-1α protein levels, unchanged levels of TGF-β and PDGF-BB in Ltv-p53 infected VSMCs, compared with Ltv-vector-infected VSMCs. Representative western blot (upper panel) of three independent experiments. GAPDH served as loading control. Densitometric analyses (lower panel) are presented as the relative ratio of each protein to GAPDH. (E) Concentration (pg/ml) of SDF-1α, TGF-β and PDGF-BB was measured by ELISA in conditioned medium from VSMCs with or without Lentivirus infection. All data presented are mean ± SEM (n = 3). ** p <0.01 versus Ltv-vector. (F) Double immunofluorescence staining for TUNEL (red) and SDF-1α (green) showing their colocalization in Ltv-p53 infected VSMCs. Scale bars: 50 μm. (G) Double immunofluorescence staining for TUNEL (red) and SDF-1α (green) showing their colocalization in aortic allografts 2 weeks after transplantation. Scale bars: 20 μm.

We then determined the influence of VSMC apoptosis on production of chemokines and cytokines such as SDF-1α, TGF-β and PDGF-BB, all of which are implicated in the development of TA (13,30,31). We found that their mRNA and protein levels were low in uninfected and Ltv-vector infected VSMCs. In Ltv-p53-infected VSMCs, the SDF-1α mRNA and protein levels were both significantly elevated, but there was no difference between Ltv-p53 and Ltv-vector-infected VSMCs in the expression levels of TGF-β and PDGF-BB (Figure 1C, D). In line with their intracellular expression, SDF-1α was higher in medium from Ltv-p53 transferred VSMCs than that of Ltv-vector transferred VSMCs. However, there were no media-related differences in the levels of TGF-β and PDGF-BB (Figure 1E).

To further determine the cellular source of SDF-1α, we conducted double immunofluorescence for TUNEL and SDF-1α in cultured VSMCs. We observed increased SDF-1α expression in apoptotic cells and neighboring viable VSMCs with Ltv-p53 (Figure 1F). It was also confirmed in vivo by double immunostaining of aortic grafts 2 weeks after transplantation, demonstrating enhanced immunoreactivity in both apoptotic and neighboring cells, mainly located in the media of allografts (Figure 1G). Taken together, these data suggest that VSMC apoptosis induces the production of SDF-1α from apoptotic and neighboring viable VSMCs.

VSMC apoptosis induces MSC migration and proliferation in a SDF-1α-dependent manner

SDF-1 is known to play a critical role in regulating stem cell mobilization, recruitment and proliferation through activating its receptor CXCR4 (13, 14), which is expressed on MSCs as indicated by immunofluorescence staining (Figure 2A). To explore the impact of VSMC apoptosis on the cellular biology of MSCs, we prepared CM from VSMCs and conducted cell migration and proliferation assays. The presence of Ltv-p53 CM in the lower chamber induced a robust migration of MSCs relative to MSCs exposed to Ltv-vector CM (Figure 2B). To determine whether SDF-1α is responsible for the increased migration of MSCs induced by Ltv-p53 CM, neutralizing anti-SDF-1α antibody was used together with CM to treat MSCs. Blockade of SDF-1α abolished Ltv-p53 CM-induced MSC migration (Figure 2B). We also evaluated the effect of CM on MSC survival and proliferation. Consistent with the effect on cell migration, exposure of serum-starved MSCs to Ltv-p53 CM caused a marked increase in DNA synthesis as measured by BrdU incorporation. This effect was abolished in the presence of neutralizing anti-SDF-1α antibody (Figure 2C). In addition, we noted that both Ltv-p53 CM and neutralizing anti-SDF-1α antibody had no effect on cell viability as assessed by trypan blue exclusion during the incubation. Thus, these findings suggest that VSMC apoptosis promotes MSC migration and proliferation in a SDF-1α-dependent manner.

Figure 2.

Vascular SMC apoptosis induces MSC migration and proliferation in a SDF-1α-dependent manner. (A) Immunofluorescence staining for CXCR4 (green) in MSCs. Cell nuclei were detected by DAPI (blue). Scale bars: 20 μm. (B) Transwell assay was performed to evaluate MSC migration towards CM derived from either Ltv-vector or Ltv-p53 infected VSMCs in the presence or absence of neutralizing anti-SDF-1α. Representative photomicrographs of transwell membranes with DAPI-stained MSCs are shown (upper panel). Scale bar: 100 μm. Quantification of migrated cells is shown (lower panel). Relative cell migration is expressed as fold increase compared with control (Ltv-vector CM with IgG). Values are mean ± SEM from triplicate determinations of three independent experiments. (C) MSC proliferation was assessed by BrdU incorporation in response to Ltv-vector or Ltv-p53 CM in the presence or absence of neutralizing anti-SDF-1α. The value is presented as mean fold increase ± SEM compared with control (Ltv-vector CM with IgG) of triplicates from three independent experiments. ** p < 0.01 versus Ltv-vector CM. ## p < 0.01 versus Ltv-p53 CM.

VSMC apoptosis via SDF-1α activates PI3K/Akt and MAPK/Erk signaling pathways in MSCs

PI3K/Akt and MAPK/Erk signaling pathways have been shown to mediate cell migration and proliferation induced by SDF-1α in different cell types (31,32). Thus, we detected whether VSMC apoptosis activates Akt and Erk in MSCs. Compared with Ltv-vector CM, Ltv-p53 CM induced a significant increase in phosphorylated Akt (Figure 3A) and Erk (Figure 3B). The addition of neutralizing anti-SDF-1α antibody to the Ltv-p53 CM largely inhibited the phosphorylation of Akt and Erk (Figure 3A, B). Moreover, Ltv-p53 CM-induced phosphorylation of Akt and Erk was also abolished by pretreatment with LY294002 (Figure 3A) and PD98059 (Figure 3B), respectively. These results demonstrate that VSMC apoptosis activates both PI3K/Akt and MAPK/Erk signaling pathways in MSCs and that is mediated by SDF-1α.

Figure 3.

VSMC apoptosis via SDF-1α activates PI3K/Akt and MAPK/Erk signaling pathways in MSCs. (A) Western blot showing phosphorylation of Akt in MSCs in response to Ltv-p53 CM in the presence or absence of neutralizing anti-SDF-1α or LY294002. Representative western blots (left panel) of three independent experiments. Densitometric analysis (right panel) is presented as the relative ratio of pAkt to total Akt. (B) Western blot showing phosphorylation of Erk1/2 in MSCs in response to Ltv-p53 CM in the presence or absence of neutralizing anti-SDF-1α or PD98059. Representative western blots (left panel) of three independent experiments. Densitometric analysis (right panel) is presented as the relative ratio of pErk to total Erk. (C) Western blots showing phosphorylation of mTOR, p70S6K and 4E-BP1 in MSCs in response to Ltv-p53 CM in the presence or absence of neutralizing anti-SDF-1α, LY294002 or rapamycin. Representative western blots (left panel) of three independent experiments. Densitometric analyses (right panel) are presented as the relative ratio of each phosphorylated protein to corresponding total protein. All values are mean ± SEM from triplicate determinations of three independent experiments. ** p < 0.01 versus Ltv-vector CM.

The mammalian target of rapamycin (mTOR) is an important downstream effector of PI3K/Akt signaling (33, 34). We then evaluated the effect of VSMC apoptosis on the activation of mTOR and its downstream targets, p70S6K and 4E-BP1. Western blotting showed that incubation of MSCs with Ltv-p53 CM led to strong phosphorylation of mTOR, as well as p70S6K and 4E-BP1 (Figure 3C). Either LY294002 or rapamycin totally blocked Ltv-p53 CM-induced phosphorylation of mTOR, p70S6K and 4E-BP1 (Figure 3C). These results show that PI3K/Akt acts upstream of mTOR in MSCs to mediate VSMC apoptosis-induced phosphorylation of mTOR, p70S6K and 4E-BP1.

PI3K/Akt/mTOR and MAPK/Erk signaling pathways mediate VSMC apoptosis-induced MSC migration and proliferation

To further explore whether PI3K/Akt/mTOR and MAPK/Erk signalings are involved in regulating MSC migration and proliferation induced by VSMC apoptosis, Transwell migration and BrdU incorporation assays were conducted in the presence of pharmacologic inhibitors. Preincubation of MSCs with LY294002, rapamycin, or PD98059, led to a dramatic decrease in cell migration towards Ltv-p53 CM (Figure 4A). In agreement with the effect on cell migration, all of these inhibitors significantly suppressed MSC proliferation (Figure 4B). These data suggest that both PI3K/Akt/mTOR and MAPK/Erk signaling pathways contribute to regulate VSMC apoptosis-induced MSC migration and proliferation.

Figure 4.

PI3K/Akt/mTOR and MAPK/Erk signaling pathways mediate VSMC apoptosis-induced MSC migration and proliferation. (A) Transwell assay was performed to evaluate MSC migration toward Ltv-vector or Ltv-p53 CM in the presence or absence of LY294002 (10 μM), rapamycin (10 nM), or PD98059 (100 μM). Relative cell migration is expressed as fold increase compared with control (Ltv-vector CM with DMSO). (B) MSC proliferation was assessed by BrdU incorporation in response to Ltv-vector or Ltv-p53 CM in the presence or absence of LY294002, rapamycin, or PD98059. All values are mean ± SEM from triplicate determinations of three independent experiments. ** p < 0.01 versus Ltv-vector CM. ## p < 0.01 versus Ltv-p53 CM.

Medial SMC apoptosis induces SDF-1α production in aortic allografts

Based on our in vitro observations, we explored the impact of VSMC apoptosis on the expression of SDF-1α in arterial wall in an established model of allograft arteriosclerosis. Selective inhibition of medial SMC apoptosis in aortic grafts was achieved by lentivirus-mediated gene transfer of BclxL. Selective expression of target genes in the media of aortic allografts 1 week after surgery was confirmed by fluorescence microscopy (Supplemental Figure IIIA). A marked increase in BclxL mRNA was also noted in the infected allografts (Supplemental Figure IIIB). TUNEL assays were conducted to detect apoptotic cells in the aortic grafts 2 weeks after transplantation, a time point at which vascular-cell apoptosis is apparent (21). Numerous apoptotic cells were detected predominantly in the medial layer of aortic allografts, but not the isografts (Figure 5A). However, the percentage of apoptotic medial cells was significantly decreased in Ltv-BclxL-infected allografts compared with Ltv-vector-infected allografts (Figure 5A). These findings demonstrate that medial SMC apoptosis occurs in the allogeneic grafts in the early phase after transplantation and that can be effectively suppressed by BclxL transfer.

Figure 5.

Medial SMC apoptosis induces the production of SDF-1α in aortic allografts. (A) Representative TUNEL-stained cross-sections from uninfected isografts, Ltv-vector or Ltv-BclxL allografts showing apoptotic cells in aorta 2 weeks after transplantation (upper panel). Arrowheads denote TUNEL positive cells. Scale bar: 20 μm. The bar graph (lower panel) shows the mean percentage TUNEL positive cells in aortic media (± SEM) from five vessels for each experimental group in four sections. (B) Representative cross-sections from uninfected isografts, Ltv-vector or Ltv-BclxL allografts immunostained for SDF-1α (brown) in aorta 2 weeks after transplantation (upper panel). Nuclei are stained blue. Scale bar: 20 μm. The bar graph (lower panel) shows the mean percentage of positively stained area to the measured area in aortic media (± SEM) from five vessels for each experimental group in four sections. ** p < 0.01 versus uninfected isograft. ## p < 0.01 versus Ltv-vector allograft. (C) mRNA expression of SDF-1α was determined by quantitative RT-PCR in aortic isografts and allografts 2 weeks after surgery. mRNA levels are normalized to GAPDH. (D) The level of SDF-1α in serum of rats 2 weeks after aortic transplantation was measured by ELISA. Values are shown as mean ± SEM, n = 6 rats per group. ** p < 0.01 versus uninfected isograft. # p < 0.05, ## p < 0.01 versus Ltv-vector allograft.

We next used immunohistochemistry to determine SDF-1α expression in the arterial wall at the same time point. SDF-1α level was minimal in isograft arteries. Nevertheless, SDF-1α expression, primarily located in the media, was strikingly increased in aortic allografts. Compared with Ltv-vector-infected allografts, Ltv-BclxL-infected allografts exhibited a significant decrease in local SDF-1α level (Figure 5B). Similarly, quantitative RT-PCR analysis demonstrated that the SDF-1α mRNA was much lower in Ltv-BclxL-infected allografts than that in Ltv-vector-infected allografts (Figure 5C). Moreover, we measured SDF-1α protein levels in plasma by ELISA at this time point. Consistent with results in the local arterial wall, plasma SDF-1α was notably elevated in rats that received allografts. However, compared with rats that received Ltv-vector-infected allografts, plasma SDF-1α was substantially decreased in rats that received Ltv-BclxL-infected allografts (Figure 5D). Collectively, these results imply that medial SMC apoptosis contributes to the induction of SDF-1α in aortic allografts.

Inhibition of medial SMC apoptosis diminishes MSC recruitment and prevents the development of allograft arteriosclerosis

As SDF-1α is known to be critical for neointimal formation and recruitment of vascular progenitor cells (13,14), we postulated that inhibition of medial SMC apoptosis would limit the development of allograft arteriosclerosis. We employed BclxL gene transfer to selectively inhibit medial SMC apoptosis of aortic allografts and examined the neointimal lesions 8 weeks after transplantation. As indicated by H&E and Masson staining, neointimal lesions were evident in allograft arteries but were minimal or absent in isograft arteries (Figure 6A). Compared with Ltv-vector-infected allografts, neointimal formation as determined by intimal area (Figure 6B) and intima/media ratio (Figure 6C) was significantly attenuated in Ltv-BclxL-infected allografts, resulting in a marked decrease of lumen stenosis (Figure 6D). We observed no differences in the media and total vessel areas among isografts, Ltv-vector-infected allografts and Ltv-BclxL-infected allografts (Figure 6E), excluding the possibility that the differences of the neointimal lesions and lumen stenosis resulted from altered aortic geometry.

Figure 6.

Inhibition of medial SMC apoptosis diminishes MSC recruitment and prevents the development of allograft arteriosclerosis. (A) Representative H&E-stained sections (upper panel) and Masson-stained sections (lower panel) from aortic grafts 8 weeks after transplantation in rat. Arrowheads denote the internal elastic lamina. Scale bars: 200 μm. Quantitative analysis of vascular lesions in aortic grafts, measured as intimal area (B), intima/media ratio (C), lumen stenosis (D), medial area and total vessel area (E). (F) Representative images illustrating the accumulation of CD90 (red) and CD105 (green) double-positive cells in aortic allografts 8 weeks after transplantation, with the dashed square indicating an area of higher magnification shown in the upper-right corner. Cell nuclei were detected by DAPI (blue). Dotted lines indicate the internal elastic lamina. Scale bars: 20 μm. Quantitative analysis shown in lower panel. Data are expressed as mean ± SEM (n = 5). ** p < 0.01 versus uninfected isograft. ## p < 0.01 versus Ltv-vector allograft.

We also explored the impact of medial SMC apoptosis on the recruitment of MSCs to the lesions using double immunostaining. We detected the existence of cells expressing MSC markers such as CD90 and CD105 in aortic allografts, mostly located in neointimal layer and partially in media (Figure 6F), supporting the concept that MSCs are implicated in the development of neointimal formation. Importantly, compared with Ltv-vector-infected allografts, Ltv-BclxL-infected allografts displayed a significant reduction in the number of CD90+CD105+ double-positive neointimal cells (Figure 6F), which is consistent with the impaired neointimal formation in the aortic allografts. Taken together, these results implicate that medial SMC apoptosis contributes to the recruitment of MSCs and the development of neointimal formation.

Blockade of PI3K/Akt/mTOR or MAPK/Erk signaling pathway limits the recruitment of MSCs and the development of allograft arteriosclerosis

Immunofluorescence staining detected robust activation of Akt, mTOR and Erk in the aortic allografts, mainly in the neointimal lesions (Figure 7A). Given that PI3K/Akt/mTOR and MAPK/Erk signalings mediate MSC migration and proliferation, we hypothesized that blocking PI3K/Akt/mTOR or MAPK/Erk signaling might impair the recruitment of MSCs and limit neointimal formation in the aortic allografts. Therefore, pharmacologic inhibitors rapamycin and PD98059 were systemic administrated to recipient animals from 10 days after surgery. As expected, the accumulation of CD90+CD105+ cells in the neointimal lesions was significantly reduced following either rapamycin or PD98059 treatment (Figure 7F). In parallel, treatment with either rapamycin or PD98059 led to a significant reduction in neointimal formation (Figure 7B–D) and lumen stenosis (Figure 7E) in the aortic allografts. These data validate our in vitro experiments and confirm that both PI3K/Akt/mTOR and MAPK/Erk signaling pathways are critical for the recruitment of MSCs and the development of neointimal formation after transplantation.

Figure 7.

Blockade of PI3K/Akt/mTOR or MAPK/Erk signaling pathway limits the recruitment of MSCs and the development of allograft arteriosclerosis. (A) Representative images showing the expression of phosphoralated Akt, mTOR and Erk1/2 (red) in aortic allografts 8 weeks after transplantation. Cell nuclei were detected by DAPI (blue). Scale bars: 200 μm. (B) Representative H&E-stained sections (upper panel) and Masson-stained sections (lower panel) from aortic allografts 8 weeks after transplantation. Arrowheads denote the internal elastic lamina. Scale bars: 200 μm. Quantitative analysis of vascular lesions in aortic allografts, measured as intimal area (C), intima/media ratio (D), and lumen stenosis (E). (F) Quantitative analysis of the percentage of CD90+CD105+ double-positive cells within the neointimal lesions in aortic allografts. Data are expressed as mean ± SEM (n = 5). * p < 0.05 versus Ltv-vector allograft.

Discussion

Vascular remodeling of graft arteries in response to alloreactivity is a complex process predominantly involving proliferation, migration and apoptosis of vascular SMCs. Numerous publications have described a critical role for the proliferation and migration of medial VSMCs, as well as the apoptosis of neointimal SMCs, in the progression of allograft arteriosclerosis. However, relatively little is known regarding the precise contribution of medial VSMC apoptosis to the pathogenesis of allograft arteriosclerosis. Here, we provide the first direct evidence that medial VSMC apoptosis plays a central role in the development of allograft arteriosclerosis. In vitro, VSMC apoptosis induces the production of SDF-1α that modulates MSC migration and proliferation via activating the PI3K/Akt/mTOR and MAPK/Erk signaling pathways. In vivo, inhibition of medial VSMC apoptosis results in a reduced level of SDF-1α, leading to a marked decrease in neointimal formation in the aortic allograft, accompanied by diminished infiltration of CD90+CD105+ cells within the lesions. Furthermore, systemic administration of pharmacologic inhibitors targeting PI3K/Akt/mTOR or MAPK/Erk pathway also reduces the accumulation of CD90+CD105+ cells in the lesions and attenuates neointimal formation and lumen stenosis in the aortic allografts.

VSMC apoptosis occurs in a variety of vascular pathologies, including allograft arteriosclerosis (20, 21), restenosis after vascular injury (35), atherosclerosis (36) and aneurysm formation (37). Under these pathological conditions, VSMC apoptosis can be triggered by various stimuli, such as inflammatory responses, mechanical injuries, wall shear stress, hypoxia and oxidized lipoproteins (38). In TA, however, the alloreactive response guided by T cells appears to be responsible for cell apoptosis in the graft vasculature (16–18). Our observation that medial SMC apoptosis occurred when the aortic grafts were allogeneic but not syngeneic in rats supports the concept that allogeneic graft SMCs represent a target of alloimmune attack that initiates medial cell apoptosis and gross loss.

Apoptosis is a genetically regulated and finely orchestrated process of selective cell deletion. Both p53 and Bcl-2 family members are shown to be crucial regulators of the death pathway (39). The p53 gene acts as an antioncogene associated with up-regulation of apoptosis. Normal vascular cells express little p53, but increased expression of p53 has been reported to exist in medial SMCs undergoing apoptosis in human aneurysm wall (37). Accordingly, in this study, VSMC apoptosis was successfully achieved by lentivirus-mediated gene transfer of p53. In contrast, BclxL, a member of the Bcl-2 family with an antiapoptotic effect, endows the medial VSMCs with a resistance to alloimmune-mediated apoptosis. This is in accordance with the fact that BclxL expression is upregulated in neointimal cells that display an apoptosis-resistant phenotype (40).

VSMC apoptosis has been proposed to be implicated in the pathogenesis of vascular proliferative diseases. In atherosclerosis, recent studies in animal models have demonstrated that VSMC apoptosis accelerates atherosclerotic plaque growth, induces an unstable plaque phenotype, prevents expansive remodeling, resulting in more severe stenosis (41, 42). In restenosis after vascular injury, however, medial VSMC apoptosis directly contributes to the development of neointimal formation and vascular remodeling (43). Likewise, local treatment with caspase inhibitor ZVAD-fmk following vascular injury in rabbits results in reduced apoptosis of medial SMCs and decreased neointimal lesions (44). Contrary to the effects of medial SMC apoptosis on neointimal formation, apoptosis of neointimal SMCs protects against lesion growth and lumen stenosis after vascular injury (45). Although it has been observed that apoptosis of medial SMCs precedes neointimal formation in allografts after transplantation in several animal models (20, 21), the direct role of medial SMC apoptosis in the development of TA is unclear. In this study, selective inhibition of VSMC apoptosis was achieved by lentivirus-mediated gene transfer of BclxL in a rat aortic transplant model. This experimental model makes it possible to define the direct role of graft medial SMC apoptosis in vascular remodeling. In this model, BclxL transfer is sufficient to inhibit the apoptosis of medial SMCs, leading to reduced neointimal lesions and alleviated lumen stenosis in the aortic allografts. Our results suggest that medial SMC apoptosis plays an active role in the development of TA and, accordingly, medial SMC apoptosis represents a potential target for therapeutic intervention to prevent neointimal formation.

VSMCs are the main cellular component of the arterial wall and fundamental for the maintenance of normal vascular function and structure. Following transplantation, massive apoptosis of medial VSMCs leads to loss of medial cells and impaired vascular function. Thus, it was recently proposed that TA is a host healing process initiated by damage of vascular wall to restore graft vascular structure and function, consisting of regeneration of ECs and repopulation of VSMCs (4). During the repair process, apart from the proliferation and migration of VSMCs from the media, the recruitment of recipient BM-derived progenitor cells has been described to contribute to lesion progression (8–13). MSCs mainly reside in the BM and can also be mobilized to blood circulation in response to injuries. MSCs are multipotent stem cells and have the capacity to differentiate into multiple cell types, including VSMCs (46), ECs and cardiomyocytes (47). Using the defined mesenchymal markers CD90 and CD105 (48), we observed the accumulation of MSCs within the neointimal lesions in aortic allografts. Although the final destination of CD90+CD105+ cells was undefined in this study, our results are in agreement with previous studies showing that MSCs are recruited to vascular lesion and participate in the process of vascular remodeling after vascular injury (8, 12). More importantly, the present study also indicates a direct correlation between medial VSMC apoptosis and the number of CD90+CD105+ cells as well as neointimal lesion size, implying that increased MSC recruitment within vascular lesions constitutes a mechanism accounting for accelerated neointimal formation in the aortic allografts. However, apart from linking medial SMC apoptosis to BM-derived cells, recently VSMC apoptosis has also been implicated in migration and proliferation of local VSMCs, by which medial SMC apoptosis contributes to vascular remodeling after vascular injury (43). In our study, we could not rule out the possibility that medial SMC apoptosis contributes to neointimal formation through modulating the cellular activities of residual VSMCs during TA.

In addition to increased medial cell apoptosis 2 weeks after transplantation, the aortic allografts, predominantly the media, display elevated expression of SDF-1α, accompanied by the enhanced serum level of SDF-1α, and that is concomitant with the apparent apoptosis of medial cells. In vitro experiments additionally provided evidence that both apoptotic and adjacent viable VSMCs produce SDF-1α. In support of this finding, apoptotic SMCs has been shown to release cytokines IL-1α and β, which may stimulate the surrounding viable SMCs to produce cytokines and chemokines (49). It is now well established that chemokine SDF-1α and its receptor CXCR4 are crucial for the recruitment of BM-derived progenitor cells to the sites of injury (13, 14, 50). Considering CXCR4 abundantly expressed on MSCs, our data imply that SDF-1α is responsible for the recruitment of MSCs to the graft vasculature induced by VSMC apoptosis. This was further evident by our in vitro observations that CM from apoptotic VSMCs induces the migration and proliferation of MSCs and that, conversely, blockade SDF-1α signaling with its neutralizing antibody abolishes these effects. Furthermore, emerging evidence suggests that the SDF-1/CXCR4 axis contributes to the regulation of MSC survival and migration (51, 52). Therefore, these data suggest that VSMC apoptosis modulates the recruitment of MSCs to the graft vasculature by the SDF-1α/CXCR4 axis.

Another important finding of this study is that PI3K/Akt signaling pathway is pivotal for MSC migration and proliferation induced by VSMC apoptosis. This is based on the observation that both SDF-1α neutralizing antibody and PI3K-specific inhibitor LY294002 blocked Akt activation and inhibited MSC proliferation and migration induced by VSMC apoptosis. We also observed phosphoralation of the downstream components of Akt, including mTOR, p70S6K and 4E-BP1 (33, 34), while mTOR inhibitor rapamycin exerts inhibitory effects on MSC proliferation and migration both in vitro and in vivo. PI3K/Akt is known to mediate signals for G protein-coupled receptors in multiple cell types and play an important role in cell survival, proliferation, migration and differentiation (53). mTOR is a downstream target of Akt. The assembly of mTOR into the mTOR complex1 (mTORC1) and mTOR complex2 (mTORC2) controls cell growth and proliferation by regulating the initiation of translation and cell-cycle progression via activation of p70S6K and inhibition of 4E-BP1 (33,34,54). Recent data have shown that mTOR also plays a critical role in the regulation of cell migration and chemotaxis. mTORC1 regulates cell motility via both S6K1 and 4E-BP1 pathways, while mTORC2 has distinct effects on the actin cytoskeleton that participates in the regulation of cell migration (34,55). These together with our findings confirm that VSMC apoptosis-induced MSC migration and proliferation are dependent on PI3K/Akt/mTOR signaling pathway. This concept is also confirmed by our in vivo results, indicating that systemic administration of rapamycin reduces the accumulation of MSCs in the vascular lesions and the neointimal lesion size.

The present study indicated that VSMC apoptosis also activates the MAPK/Erk signaling pathway in MSCs via SDF-1α/CXCR4 axis. MAPK/Erk signaling pathway has been shown to modulate a wide variety of cellular processes including cell migration, proliferation and survival (56), all of which are involved in neointimal formation. Studies on SDF-1α have suggested that MAPK/Erk signaling mediates the SDF-1α-induced migration in different cell types; however, it is cell type dependent. In PC-3 carcinoma cells, Erk signaling is required for SDF-1-induced chemotaxis (32), whereas PI3K/Akt activation, but not Erk activation, is required for SDF-1α-induced cell migration in T lymphocytes and hematopoietic progenitor cells (57). Our in vitro studies indicated that MAPK/Erk signaling pathway contributes to mediate VSMC apoptosis-induced MSC migration and proliferation, which constitutes a mechanism accounting for impaired neointimal formation in rats by treating with PD98059. These data suggest that both PI3K/Akt/mTOR and MAPK/Erk signaling pathways are required for VSMC apoptosis-induced MSC migration and proliferation and subsequent neointimal formation in the allografts.

Vascular wound repair and remodeling is a complex process involving vascular wall cells, inflammatory cells and vascular progenitor cells. Although previous work and our present study indicate that the recruitment of MSCs participates in the repair process of the injured vasculature, the precise role of MSCs in neointimal formation remains largely undefined. MSCs may directly repopulate the neointimal lesions by giving rise to neointimal cells (8,12), while MSCs have recently been shown to possess paracrine functions and immunomodulatory features (58–60), by which MSCs may play an indirect or even negative role in the development of neointimal formation and TA. Thus, additional studies are needed to clarify the role of MSCs in the development of TA. Another limitation of this study is that it is unclear about the molecular mechanism by which apoptotic cells induce the neighboring viable VSMCs to produce SDF-1α. Future studies are also required to address this issue.

In conclusion, our findings implicate a critical role of VSMC apoptosis in the development of TA through inducing SDF-1α production and further indicate that the recruitment of MSCs constitutes an important component of pathological vascular remodeling. The targeted manipulation of cell signaling for medial VSMC apoptosis and MSC recruitment may benefit the treatment of vascular proliferative diseases.

Acknowledgments

Funding sources: This work was supported by the grants from the National Natural Science Foundation of China to Z.S. (No.30700398, No.81170441) and Q.Z. (No.30600786). The work was also supported by the grant from the Specialized Research Fund for the Doctoral Program of Higher Education of Ministry of Education of China to Z.S. (No.20070487158).

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary