Vascular Cell Apoptosis
Cell Type–Specific Modulation by Transforming Growth Factor-β1 in Endothelial Cells Versus Smooth Muscle Cells
Background—It is postulated that vascular lesion formation and remodeling involves a balance between vascular cell death and cell proliferation. Transforming growth factor-β1 (TGF-β1) is a pleiotropic factor expressed within vascular cells that regulates cell growth in a tissue-specific manner. This study tested the hypothesis that the control of vascular cell apoptosis involves cell type–specific regulation by TGF-β1.
Methods and Results—In response to serum withdrawal, cultured endothelial cells and vascular smooth muscle cells exhibited apoptosis as evidenced by DNA laddering and quantitated by analysis of nuclear chromatin morphology. Addition of TGF-β1 to endothelial cells in serum-free media further potentiated the induction of apoptosis in a dose-dependent fashion. Moreover, TGF-β1 promoted endothelial cell death despite the presence of 10% serum. However, endothelial cells plated on collagen I were resistant to TGF-β1–induced apoptosis. This antiapoptotic influence of the matrix was mimicked by integrin activation with anti-β1 antibodies and associated with increased expression of the antiapoptotic factor bcl-2. In accord with the hypothesis that the modulation of antiapoptotic gene expression may mediate the effects of TGF-β1 and β1 integrins on cell fate, we observed that endothelial cells with constitutive upregulation of bcl-2 remained viable despite exposure to TGF-β1 in serum-free conditions. In contrast to the proapoptotic effect of TGF-β1 in endothelial cells, addition of TGF-β1 to vascular smooth muscle cells in serum-free media inhibited apoptosis.
Conclusions—These findings suggest that the effect of cytokines such as TGF-β1 on cell fate is contextual and is modulated by cell-matrix interactions in a cell type–specific manner.
The pathogenesis of vascular diseases such as atherosclerosis and restenosis after angioplasty is characterized by the accumulation of cells within the intima. Although studies of the pathogenesis of vascular disease have focused on the role of cell proliferation as a determinant of vessel cellularity, a growing body of evidence suggests that vascular structure and lesion formation is determined by a balance between cell growth and cell death by apoptosis.
The process of apoptosis appears to determine tissue structure in health and disease in a variety of contexts.1 Seminal studies by Cho et al2 have documented that programmed cell death may play a critical role in the ontogeny of the vasculature. Moreover, recent studies of human vascular disease specimens have provided evidence that apoptosis may occur within the context of atherogenesis and restenosis after angioplasty.3 4 5 However, the factors that determine the balance between cell death and cell survival within the vessel wall remain to be characterized.
Studies of human atherosclerotic plaques as well as restenosis lesions after angioplasty have documented transforming growth factor-β1 (TGF-β1) expression by vascular cells within lesions.6 Similarly, TGF-β1 expression within the vasculature is upregulated in the context of diabetes and hypertension in animal models of vascular disease.7 8 TGF-β1 is postulated to play a mediator role in promoting alterations in vessel structure on the basis of its pleiotropic properties as a modulator of cell growth, differentiation, migration, and matrix production.9 Furthermore, a recent study suggests that alterations in the local abundance of TGF-β1 in the arterial wall promotes vascular cell transdifferentiation, vascular wall remodeling, arterial lesion growth, and lesion regression associated with apoptosis.10 The biological effect of TGF-β1 is cell specific and contextual. Given the pleiotropic effects of TGF-β1 on vascular cell function, the present study tested the hypothesis that TGF-β1 also modulates the process of vascular cell death by apoptosis. Our findings indicate that TGF-β1 is a bifunctional modulator of cell death that exerts a proapoptotic influence on endothelial cells and is antiapoptotic for vascular smooth muscle cells (VSMC). These observations have important implications for understanding the role of cytokine growth factors such as TGF-β1 in the pathogenesis of vascular disease.
Endothelial Cell Experimental Protocols
Bovine aortic endothelial cells (BAEC) were isolated and characterized as previously described.11 BAEC (passages 6 to 12) were maintained in Dulbecco’s modified Eagle’s medium (DME) (Gibco Laboratories) supplemented with 10% heat-inactivated calf serum (CS) (Gibco), penicillin (100 U/mL), and streptomycin (100 μg/mL) and incubated in a humidified atmosphere of 95% air/5% CO2. The medium was changed every 2 to 3 days, and cells were routinely passaged by trypsin/EDTA with a split ratio of 1:6. Human umbilical vein endothelial cells (HUVEC) were obtained as primary cultures from Clonetics Corp (San Diego, Calif). HUVEC (passage 2 to 5) were maintained in endothelial growth medium (EGM) containing 10 ng/mL human recombinant epidermal growth factor, 1 μg/mL hydrocortisone, 2% fetal bovine serum (FBS), 50 μg/mL gentamicin, 50 ng/mL amphotericin-B, 12 μg/mL bovine brain extract, and 10 μg/mL heparin. Bovine pulmonary artery endothelial cells (CPAE) were obtained from American Type Culture Collection (Rockville, Md) and maintained (passages 18 to 22) in DME medium 20% FBS (Gibco). For growth and apoptosis studies, HUVEC, BAEC, and CPAE were grown to a near-confluent state on Labtek glass chamber slides (Nunc, Inc).
BAEC DNA Synthesis
BAEC were grown as described above and incubated in serum-free DME with or without supplementation with human recombinant basic fibroblast growth factor (bFGF) (50 ng/mL) (Sigma Chemical Co, St Louis, Mo) in the presence or absence of human recombinant TGF-β1 (1 ng/mL) (R&D Systems, Minneapolis, Minn). 5-bromo-2′-deoxyuridine (BrdU) (Sigma Chemical Co) (5 μg/mL) was added to each well for a total incubation of 48 hours before fixation in 70% ethanol. Nuclei incorporating BrdU were detected with a streptavidin-biotin detection system (Zymed) and the percent positive nuclei determined by counting 200 total nuclei.
BAEC Population Dynamics
BAEC were grown as described above and incubated for 48 hours in serum-free DME with or without supplementation with FGF (50 ng/mL) in the presence or absence of TGF-β1 (1 ng/mL). Adherent cells were harvested by trypsinization, and mean cell numbers were determined with a Coulter counter at baseline and after 48 hours.
Kinetics of BAEC Apoptosis
BAEC were grown as described above and incubated for 8, 18, 24, or 48 hours in serum-free DME, DME supplemented by FGF (50 ng/mL), or TGF-β1 (1 ng/mL). Cells were stained with fluorescent DNA-binding dyes, harvested, and analyzed for apoptosis as described below. To define the dose-response relations, cells were exposed to increasing concentrations of TGF-β1 (0, 0.1 0.25, 0.5, 1, and 5 ng/mL) for 48 hours in 10% CS or serum-free DME with or without supplementation with FGF (50 ng/mL). After 48 hours, the adherent and nonadherent cells were harvested for in situ end-labeling, extraction of chromosomal DNA, or stained with fluorescent DNA-binding dyes and analyzed for apoptosis as described below.
TGF-β1 Induction of Apoptosis in HUVEC
HUVEC were grown to a near-confluent state in EGM before incubation with 1 ng/mL of TGF-β1. After 24 hours, the adherent and nonadherent cells were harvested and stained with fluorescent DNA-binding dyes and analyzed for apoptosis as described below. For collagen coating experiments, rat tail collagen type I (Collaborative Biomedical Products, Bedford, Mass) was used according to the manufacturer’s specifications to apply a thin coating (50 μg/mL) on the surface of the glass chamber slides. The slides were incubated at room temperature for 1 hour before rinsing with PBS. For antibody coating experiments, glass chamber slides were incubated for 2 hours with 20 μg/mL mouse anti-human β1 integrin monoclonal antibody (clone TS2/16; Endogen, Cambridge, Mass) or control purified mouse IgG (Sigma) before rinsing with PBS. For collagen and antibody coating experiments, HUVEC were plated at 1.5×105 cells/well in fully supplemented EGM in the presence of 1 ng/mL TGF-β1. After 24 hours, the adherent and nonadherent cells were harvested and stained with fluorescent DNA-binding dyes and analyzed for apoptosis as described below.
Collagen type I was used to coat tissue culture flasks as described above. CPAE were plated at 3×106 cells/flask in serum-free medium in the presence or absence of collagen I. After 24 hours, the adherent and nonadherent cells were harvested and homogenized in lysis buffer (10 mmol/L Tris, 1% SDS solution). Protein concentrations were determined by a BCA assay (Pierce Chemical). Cell lysates (50 μg) were loaded on a 12% SDS-polyacrylamide gel, electrophoretically transferred to a nitrocellulose membrane (Hybond ECL, Amersham), and the membrane stained with india ink to verify equal loading and transfer efficiency. The membrane was blocked in PBS, 0.1% Tween, 5% nonfat dry milk and probed with a polyclonal antibody directed against bcl-2 (#492, Santa Cruz Biotechnology, Santa Cruz, Calif) (1:100 dilution) followed by a streptavidin-biotinylated-horseradish peroxidase–linked secondary antibody (1:100 dilution) (Dako Corp) and detected with electrochemical luminescence detection reagents (Amersham) by autoradiography.
Retroviral Constructs and Infection of Endothelial Cells
Retroviral constructs containing the human bcl-2 gene under the control of the retroviral long-terminal repeat promoter and the neomycin resistance gene were generated, titered, and characterized as previously described.12 For infections of BAEC, 1×106 cells were treated with viral supernatants added at 12-hour intervals for 3 days. Stable transfectants were selected by incubation with G418 (1 mg/mL). Medium was changed twice weekly for 2 weeks, until G418-resistant colonies appeared. Cultures were subsequently trypsinized and maintained as described above. Pools of cells were used for subsequent experiments to minimize differences that might have resulted from viral integration. Expression of the human bcl-2 protein was confirmed by immunohistochemical staining with a monoclonal antibody directed against human bcl-2 (Dako, Carpenteria, Calif) as previously described.12 The majority of BAEC transfected with the human bcl-2 vector exhibited positive immunostaining for bcl-2 (>90%), whereas BAEC transfected with the control vector containing the neomycin resistance cassette alone exhibited no detectable staining (data not shown).
VSMC Experimental Protocols
Rabbit smooth muscle cells (VSMC) were isolated and characterized as previously described.13 Human umbilical artery smooth muscle cells were purchased from Clonetics Corp (San Diego, Calif). VSMC (passages 5 to 9) were maintained in DME medium and Ham’s F12 (DME/F12) (Gibco) supplemented with either 10% (rabbit VSMC) or 20% (human VSMC) heat-inactivated FBS (Gibco), penicillin (100 U/mL), and streptomycin (100 μg/mL) and 25 mmol/L HEPES buffer. For each of the following experiments, VSMC were grown in 10% FBS to a near-confluent state.
VSMC DNA Synthesis
VSMC were grown as described above and incubated in 10% FBS or serum-free DME in the presence or absence of TGF-β1(1 ng/mL). BrdU (5 μg/mL) was added to each well for a 24-hour incubation before fixation and analysis as described above.
VSMC Population Dynamics
VSMC were grown as described above and incubated for an additional 24 hours in 10% FBS or serum-free DME in the presence or absence of TGF-β1(1 ng/mL). Mean cell numbers were determined with a Coulter counter.
TGF-β1 Inhibition of Apoptosis in VSMC
VSMC were grown as described above and incubated in serum-free DME in the presence or absence of TGF-β1 (0.25, 0.50, or 1 ng/mL) for 24 hours. After 24 hours, the adherent and nonadherent cells were harvested for extraction of chromosomal DNA or stained with fluorescent DNA-binding dyes and analyzed for apoptosis as described below.
Microscale Analysis of DNA Fragmentation
In Situ 3′-End-Labeling of DNA
Identification of fragmented DNA in situ was performed by collecting adherent and nonadherent cells through trypsinization and centrifugation and fixing in 4% buffered formaldehyde. In situ labeling of 3′-OH ends was performed by an adaptation of a previously described procedure.16 Specimens in which the terminal deoxynucleotide transferase was omitted served as a negative control, and a DNase-treated specimen served as a positive control.
Quantitative Analysis of Apoptosis by Fluorescent Microscopy
As an additional assay of cell death by apoptosis, we used fluorescent DNA-binding dyes to define nuclear chromatin morphology as a quantitative index as previously described by our laboratory.14 Counts are expressed as apoptotic nuclei/total nuclei to obtain percent apoptotic nuclei.
The apoptotic index obtained is a cross-sectional and not a cumulative assessment at the time of cell harvest and may be an underestimate of the total apoptotic cell population. We have observed a high concordance between rates of apoptosis observed with this method compared with assays of subdiploid apoptotic populations with fluorescence-activated cell sorter analysis and phase-contrast time-lapse video microscopy.
Experiments were performed in triplicate except as indicated and repeated a minimum of 3 times. Data are presented as mean±SD. Statistical analyses were performed by ANOVA or unpaired 2-tailed Student’s t test.
Endothelial Cell Growth Kinetics: Response to bFGF Versus TGF-β1
In accordance with previous reports,17 the initial set of experiments characterized the effects of serum deprivation, bFGF, and TGF-β1 on cell growth and kinetics. Under serum-free conditions, we observed a net decrease in cell number over a 48-hour period. The addition of bFGF prevented the net cell loss induced by serum withdrawal and effectively maintained cellularity at baseline levels (Figure 1⇓). However, the administration of TGF-β1 potentiated the net cell loss induced by serum withdrawal. Moreover, TGF-β1 induced a net cell loss compared with baseline even in the presence of bFGF (Figure 1⇓). We hypothesized that both serum withdrawal and TGF-β1 administration result in a net decrease in cell number by inducing cell death in addition to the well-described effects on inhibiting cell cycle progression.
Endothelial Cell Apoptosis: Induction by Serum Withdrawal and TGF-β1
To test the hypothesis that TGF-β1 induces endothelial cell loss by apoptosis, we used 3 complementary assays. In DNA extracts obtained from endothelial cells maintained in medium containing 10% serum, there was little evidence of apoptosis by DNA laddering. In contrast, DNA extracted from cells exposed to serum-free conditions for 48 hours exhibited extensive cleavage of DNA in the characteristic internucleosomal pattern. The addition of bFGF inhibited apoptosis, whereas the administration of TGF-β1 further enhanced this process of cell death (Figure 2⇓). The finding that the net cell loss after exposure to TGF-β1 or serum withdrawal was due to apoptotic death was further confirmed by analysis of nuclear chromatin morphology with fluorescence microscopy with the use of a DNA-binding dye (Hoechst 33342) as well as by the in situ end-labeling assay (data not shown). All 3 techniques were concordant in documenting apoptosis in response to serum withdrawal and TGF-β1 administration.
The process of apoptosis was quantitated by the nuclear chromatin morphology assay. Cells maintained in 10% serum have a low basal apoptotic index of ≈2%. The induction of apoptosis in response to serum withdrawal was detectable above baseline within 18 hours, achieved a plateau within 24 hours, and was sustained for up to 48 hours (Figure 3⇓). The administration of TGF-β1 (1 ng/mL) caused a significant increase in the percentage of apoptotic cells noted at these time points. Indeed, picomolar concentrations of TGF-β1 induced apoptosis in a dose-dependent manner under serum-free conditions (Figure 4⇓).
On the basis of the observation that bFGF effectively reverses the induction of apoptosis in response to serum withdrawal, we assessed the effect of bFGF supplementation on TGF-β1–induced apoptosis. Although bFGF administration substantially reduced the baseline level of apoptosis, cells treated with TGF-β1 continued to undergo cell death by apoptosis in a dose-dependent manner despite coincubation with bFGF. To further confirm the proapoptotic effect of TGF-β1, we assessed the response in the presence of 10% serum. Although multiple mitogens and putative survival factors were present under these conditions, exposure to TGF-β1 induced endothelial cell death in a dose-dependent manner (Figure 4⇑). To confirm the broad applicability of this observation, we also assessed the effect of TGF-β1 on human umbilical vein endothelial cells and confirmed that responses observed in bovine aortic cells could be replicated in endothelial cells from another species and vascular bed in the presence of fully supplemented growth medium (11%±0.7% [1 ng/mL TGF-β1], n=6, vs 4%±0.6% [vehicle control], n=6; P<0.001).
Modulation of TGF-β1-Induced Apoptosis: Cell-Matrix Interactions
We were intrigued by the observation that survival factors present within serum failed to abolish TGF-β1–induced apoptosis. Given that the response to pleiotropic cytokines is often contextual, we tested the hypothesis that the proapoptotic influence of TGF-β1 is modulated by factors within the microenvironment such as matrix proteins. Indeed, we observed that plating HUVEC on collagen I markedly attenuated the induction of apoptosis in response to TGF-β1 (1 ng/mL) as compared with cells plated on glass (19%±3.5%[ glass] vs 2%±0.6% [collagen], n=6; P<0.001). As an additional set of control experiments, we confirmed previous observations that plating on collagen failed to modulate other TGF-β1–induced cellular responses such as the inhibition of DNA replication (data not shown).
To further characterize this countervailing antiapoptotic influence of the cell substratum, we tested the hypothesis that the enhanced cell survival on collagen I was mediated by β1 integrins that bind to this matrix protein. We observed that HUVEC plated on immobilized control IgG exhibited a similar induction of TGF-β1–induced apoptotic death when compared with cells plated on glass (19%±3.5% [glass] vs 17%±2% [IgG], n=6; P>0.05). In contrast, HUVEC plated on immobilized anti-β1 integrin antibodies18 proved to be resistant to the induction of apoptosis in response to TGF-β1 administration (1 ng/mL) (17%±2% [IgG] vs 7%±2.2% [anti-β1 integrin], n=6; P<0.001) (Figure 5⇓). Thus activation of β1 integrins by collagen I within the microenvironment activates a survival signal that can prevent TGF-β1–induced apoptotic death.
It is intriguing that we observed a more potent inhibition of TGF-β1–induced apoptosis by cell-matrix interactions compared with peptide growth factors. We hypothesized that the survival-promoting effects of the collagen matrix was mediated by the upregulation of the antiapoptotic gene bcl-2. In contrast to the relatively low levels of bcl-2 expression observed while plated on tissue culture plastic, plating on collagen I markedly increased the expression of bcl-2 (Figure 6⇓). To confirm that the upregulation of bcl-2 observed in response to matrix-integrin interactions is sufficient to inhibit TGF-β1–induced death, we examined the response of BAEC stably transfected with the bcl-2 gene.
Although neomycin-selected BAEC control cells remained susceptible to TGF-β1–induced cell death, BAEC with constitutive upregulation maintained cell viability despite expose to TGF-β1 in serum-free conditions (2%±1.3% [serum free] vs 4%±1.2% [TGF-β1]; P>0.05) (Figure 7⇓). Taken together, these findings suggest that the proapoptotic effect of TGF-β1 is modulated by matrix-integrin interactions involving the upregulation of bcl-2.
Given that TGF-β1 is a cytokine factor produced by both endothelial cells and VSMC, we examined whether TGF-β1 had similar effects on VSMC apoptosis regulation.
VSMC Growth Kinetics: Response to TGF-β1
We initially confirmed the effect of serum deprivation and TGF-β1 administration on cell growth kinetics as previously defined.19 The administration of TGF-β1 (1 ng/mL) significantly inhibited DNA replication in VSMC grown in the presence of 10% serum (29%±1.2% [10% serum] vs 21%±1.0% [TGF-β1]; P<0.001) and potentiated the decrease in cell cycle progression induced by serum withdrawal over a 24-hour period (18%±0.4% [serum withdrawal] vs 13%±0.3% [TGF-β1]; P<0.001).
To further define the growth kinetics, changes in cell number were assessed in parallel with changes in DNA replication. In the presence of 10% serum, as compared with baseline cell counts, the near-confluent cells continued to accumulate over a 24-hour period, whereas cells exposed to serum withdrawal exhibited a significant net decrease in cell number. In response to TGF-β1 administration, we observed an apparent paradox. Although TGF-β1 inhibited cell cycle progression in VSMC, the administration of TGF-β1 actually increased cellularity by preventing the net cell loss that accompanied serum withdrawal (Figure 8⇓). Given this discordance between the inhibitory effect of TGF-β1 on DNA replication and the net increase in cell number, we hypothesized that TGF-β1 preserved cellularity by preventing VSMC death.
VSMC Apoptosis: Effect of TGF-β1
VSMC grown to near confluence maintained in 10% serum exhibited little evidence of apoptosis, as determined by the gel electrophoresis assay for DNA fragmentation (data not shown). In contrast, DNA extracted from cells exposed to serum withdrawal for 24 hours had substantial DNA fragmentation that was markedly attenuated by the administration of TGF-β1 (Figure 9⇓). These results assayed by gel electrophoresis were verified and quantitated by examining nuclear chromatin morphology with fluorescent microscopy. In contrast to the proapoptotic effect observed with endothelial cells, TGF-β1 at picomolar concentrations prevented the death of VSMC triggered by serum withdrawal (Figure 10⇓). To confirm the broad applicability of this observation, the effect of TGF-β1 on human VSMC was also determined. In response to serum withdrawal, we observed an increase in apoptotic frequency from 2%±0.4% (n=9) to 23%±2.1% (n=6) as compared with serum controls (P<0.001). Exposure to TGF-β1 (1 ng/mL) significantly inhibited apoptosis in human VSMC (11%±0.8% [serum withdrawal+TGF-β1], n=6, vs 23%±2.1% [serum withdrawal], n=6; P<0.001). These findings suggest that although TGF-β1 exerts a similar antiproliferative effect on both endothelial cells and VSMC, there are distinct cell type–specific regulatory mechanisms that govern programmed cell death in these vascular cells.
It has been postulated that vascular structure and lesion formation is determined in part by a balance between cell proliferation and cell death by apoptosis. Indeed, recent studies of human vascular specimens and animal models suggest that apoptosis is involved in the pathogenesis of vascular disease.3 4 5 20 The cytokine factors that regulate apoptosis within the vessel wall remain to be further characterized. The present study provides the first evidence that the pleiotropic growth factor TGF-β1 is a bifunctional modulator of vascular cell apoptosis. TGF-β1 is an inducer of endothelial cell death, whereas it prevents VSMC death. In addition, we are intrigued by the observation that the proapoptotic effect of TGF-β1 can be abrogated by countervailing signals induced by cell-matrix interactions. These in vitro observations suggest that the regulation of apoptosis within an intact vessel by cytokine growth factors such as TGF-β1 may be dependent on the cell type as well as the local cellular context.
The specificity of the effect of TGF-β1 on cell death is substantiated by several lines of evidence including DNA laddering, nuclear chromatin morphology, and in situ end labeling. The specificity of the response was further evidenced by the fact that the effect of TGF-β1 was dose dependent and observed within the picomolar range. These picomolar concentrations of TGF-β1 are similar to stimulated levels in response to shear stress or angiotensin II.11 13 In addition, the observation that TGF-β1 had opposing effects on different cell types indicates that the administration of TGF-β1 did not have a nonspecific toxic effect. Finally, additional studies in human vascular cells documented that the bifunctional response was not limited to the bovine or rabbit cell isolates that were the primary focus of this study. The effects of TGF-β1 on vascular cell apoptosis were therefore cell type specific and not species specific. Overall, our findings in vascular cells are consistent with previous reports in hepatic cells that have shown that TGF-β1 has the capacity to induce apoptosis.21 During the preparation of this report, 2 reports confirmed the observation that TGF-β1 induces endothelial cell apoptosis.22 23 We have further extended these findings by documenting that the proapoptotic effect is contextual and that engagement of β1 integrins by matrix proteins such as collagen I can abrogate the induction of apoptosis in response to TGF-β1. Furthermore, our findings in an in vitro model appear to be supported by data from a recent in vivo study in which transient overexpression of TGF-β1 induces vascular lesion growth followed by vascular cell apoptosis during TGF-β1 withdrawal and lesion regression.
It is intriguing that our findings on cell type specificity in the regulation of cell death are similar to previous observations that have documented cell type–specific effects of TGF-β1 in the regulation of cell growth.17 24 25 26 In addition, the observation that the response to TGF-β1 is contextual indicates that the endothelial cell fate is dependent on a variety of proapoptotic and antiapoptotic mediators present within the microenvironment. Further studies are necessary to characterize the actual role of apoptosis in vascular remodeling and lesion formation in vivo.
The regulation of apoptosis in vascular cells remains poorly characterized. Studies in other cell systems suggest that a balance of proapoptotic versus antiapoptotic factors such as bax and bcl-2 may be involved.27 28 Given our observation that the regulation of apoptosis in vascular cells is cell type specific, it is clear that pathways defined in other cell systems may not be directly applicable to elucidating the regulation of apoptosis in vascular cells. It is noteworthy that we observed a dissociation between the regulation of cell growth and cell death by TGF-β1. Although TGF-β1 inhibited cell cycle progression in both endothelial and vascular smooth muscle cells, we documented diametric effects on the regulation of apoptosis. This observation suggests that these cellular signaling pathways regulating growth and death may be divergent in vascular cells.
The pathway that links TGF-β1 receptor activation with cell death remains to be clarified. It was recently reported that TGF-β1 stimulation results in decreased bcl-2 protein levels in endothelial cells.22 We demonstrated that microenvironmental cues such as the cell substratum can modulate the proapoptotic influence of TGF-β1 on endothelial cell fate. We observed that engagement of β1 integrins promotes an antiapoptotic signal that can override the proapoptotic effect of TGF-β1. Furthermore, plating on collagen I results in increased expression of bcl-2 in association with an antiapoptotic effect. Moreover, forced expression of bcl-2 prevents both endothelial cell death in response to serum withdrawal and TGF-β1. Taken together, these findings suggest that endothelial cell fate is determined in part by the modulation of bcl-2 expression by the countervailing influence of proapoptotic factors such as TGF-β1 and antiapoptotic mediators such as β1 integrins. Further studies are necessary to clarify the role of other antiapoptotic members of the bcl-2 family in endothelial cells29 as well as the receptor-mediated signaling pathways that determine the expression of apoptosis regulatory genes. Similarly, further investigation is necessary to define the cellular mediators involved in the regulation of VSMC fate.
In summary, the present study has documented that TGF-β1 is an inducer of endothelial cell apoptosis, whereas it prevents VSMC death. The effect of TGF-β1 on cell fate is contextual and is modulated by cell-matrix interactions mediated by β1 integrins. This observation further extends the mechanisms by which this pleiotropic factor can modify vascular structure. The cell type specificity of the effect of TGF-β1 on cell viability has implications for understanding the role of this cytokine factor in the process of vascular remodeling and intimal lesion formation. Further investigation is necessary to elucidate the molecular basis of the bifunctional effects of TGF-β1 on vascular cell apoptosis and characterize the actual role of apoptosis as a determinant of vascular structure in vivo.
This work was supported by the Baxter Foundation, the American Heart Association, and National Institutes of Health grant HL-48638. Dr Gibbons is a Pew Biomedical Research Scholar. Dr Pollman is a recipient of a California Affiliate American Heart Association Postdoctoral Fellowship Award. Dr Naumovski is an Amgen/American Society of Hematology Scholar.
- Received August 1, 1998.
- Revision received November 17, 1998.
- Accepted December 7, 1998.
- Copyright © 1999 by American Heart Association
Cho A, Courtman DW, Langille BL. Apoptosis (programmed cell death) in arteries of the neonatal lamb. Circ Res. 1995;76:168–175.
Isner JM, Kearney M, Bortman S, Passeri J. Apoptosis in human atherosclerosis and restenosis. Circulation. 1995;91:2703–2711.
Nikol S, Isner JM, Pickering JG, Kearney M, Leclerc G, Weir L. Expression of transforming growth factor-β1 is increased in human vascular restenosis lesions. J Clin Invest. 1992;90:1582–1592.
Park IS, Kiyomoto H, Abboud SL, Abboud HE. Expression of transforming growth factor beta and type IV collagen in early streptozotocin-induced diabetes. Diabetes. 1997;46:473–480.
Kim S, Ohta K, Hamaguchi A, Omura T, Yukimura T, Miura K, Inada AY, Ishimura Y, Chatani F, Iwao H. Angiotensin II type I receptor antagonist inhibits the gene expression of transforming growth factor-β1 and extracellular matrix in cardiac and vascular tissues of hypertensive rats. J Pharmacol Exp Ther. 1995;273:509–515.
Border WA, Ruoslahti E. Transforming growth factor-beta in disease: the dark side of tissue repair. J Clin Invest. 1992;90:1–7.
Schulick AH, Taylor AJ, Zuo W, Qiu CB, Dong G, Woodward RN, Agah R, Roberts AB, Virmani R, Dichek DA. Overexpression of transforming growth factor β1 in arterial endothelium causes hyperplasia, apoptosis, and cartilaginous metaplasia. Proc Natl Acad Sci U S A. 1998;95:6983–6988.
Ohno M, Cooke JP, Dzau VJ, Gibbons GH. Fluid shear stress induces endothelial transforming growth factor beta-1 transcription and production. J Clin Invest. 1995;95:1363–1369.
Naumovski L, Cleary ML. Bcl2 inhibits apoptosis associated with terminal differentiation of HL-60 myeloid leukemia cells. Blood. 1994;83:2261–2267.
Gibbons GH, Pratt RE, Dzau VJ. Vascular smooth muscle cell hypertrophy vs hyperplasia. autocrine transforming growth factor-β1 expression determines growth response to angiotensin II. J Clin Invest. 1992;90:456–461.
Pollman MJ, Yamada T, Horiuchi M, Gibbons GH. Vasoactive substances regulate vascular smooth muscle cell apoptosis: countervailing influences of nitric oxide and angiotensin II. Circ Res. 1996;79:748–756.
Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol. 1992;119:493–501.
Heimark RL, Twardzik DR, Schwartz SM. Inhibition of endothelial regeneration by type-beta transforming growth factor from platelets. Science. 1986;233:1078–1080.
Meredith JE Jr, Fazeli B, Schwartz MA. The extracellular matrix as a cell survival factor. Mol Biol Cell. 1993;4:953–961.
Owens GK, Geisterfer AAT, Yang YW-H, Komoriya A. Transforming growth factor-β-induced growth inhibition and cellular hypertrophy in cultured vascular smooth muscle cells. J Cell Biol. 1988;107:771–780.
Oberhammer FA, Pavelka M, Sharma S, Tiefenbacher R, Purchio AF, Bursch W, Shulte-Hermann R. Induction of apoptosis in cultured hepatocytes and in regressing liver by transforming growth factor β1. Proc Natl Acad Sci U S A. 1992;89:5408–5412.
Tsukada T, Eguchi K, Migita K, Kawabe Y, Kawakami A, Matsuoka N, Takashima H, Mizokami A, Nagataki S. Transforming growth factor beta 1 induces apoptotic cell death in cultured human umbilical vein endothelial cells with down-regulated expression of bcl-2. Biochem Biophys Res Commun. 1995;210:1076–1082.
Choi ME, Ballermann BJ. Inhibition of capillary morphogenesis and associated apoptosis by dominant negative mutant transforming growth factor-beta receptors. J Biol Chem. 1995;270:21144–21150.
Assoian RK, Sporn MB. Type β transforming growth factor in human platelets: release during platelet degranulation and action on vascular smooth muscle cells. J Cell Biol. 1986;102:1217–1223.
Agrotis A, Saltis J, Bobik A. Transforming growth factor-β1 gene activation and growth of smooth muscle from hypertensive rats. Hypertension. 1994;23:593–599.
Steller H. Mechanisms and genes of cellular suicide. Science. 1995;267:1445–1449.
Reed JC. Mini-review: cellular mechanisms of disease series: Bcl-2 and the regulation of programmed cell death. J Cell Biol. 1994;124:1–6.