Tobacco smoke reduces viability in human lung fibroblasts: protective effect of glutathioneS-transferase P1

Takeo Ishii, Takeshi Matsuse, Hiroko Igarashi, Michiaki Masuda, Shinji Teramoto, Yasuyoshi Ouchi


Cigarette smoking is thought to be a major risk factor in various lung diseases including lung cancer and emphysema. However, the direct effect of cigarette smoke on the viability of lung-derived cells has not been fully elucidated. In this study, we investigated the viability of human lung fibroblast-derived (HFL1) cells to different concentrations of cigarette smoke extract (CSE). CSE induced apoptosis at lower concentrations (10–25%) and necrosis at higher concentrations (50–100%). We also examined the effects of glutathioneS-transferase P1 (GSTP1), one of the xenobiotic metabolizing and antioxidant enzymes in the lung, against the cytotoxicity of CSE. Our results indicated that the level of HFL1 cell death was decreased by transfection with a GSTP1 expression vector and was increased by GSTP1 antisense vector transfection. Therefore, transient overexpression and underexpression of GSTP1 appeared to inhibit and enhance the cytotoxic effects of CSE on HFL1 cells, suggesting that GSTP1 may have protective effects against cigarette smoke in the airway cells.

  • human lung-derived cell
  • apoptosis
  • necrosis
  • xenobiotic enzyme

tobacco smoke (TS) is thought to be associated with various diseases, especially with lung diseases such as lung cancer and emphysema (22). TS contains about 4,000 various constituents, including numerous chemicals that result in the production of reactive oxygen species. In addition, TS contains several major carcinogens (12) as well as other chemicals reported to cause apoptosis in several kinds of cells (16, 25, 35). Recently, TS was reported to cause apoptosis and necrosis in mammalian cell lines (33), and the mechanism of apoptosis induction by TS was thought to occur in association with mitochondrial depolarization (2). TS-mediated depletion of lung glutathione is thought to lead to increased lipid peroxidation, neutrophil sequestration, and transcription of proinflammatory cytokine genes. Although these changes are thought to be involved in the pathogenesis of chronic obstructive pulmonary disease (COPD) (24), TS-induced apoptosis and necrosis of human lung-derived cells have not yet been studied extensively.

It is possible that xenobiotic metabolizing enzymes and antioxidants could have major roles in protection against TS-related lung diseases (3). Several investigators (5, 11, 14, 27) have reported that the genetic polymorphisms of xenobiotic metabolizing enzymes are associated with COPD, a major TS-related disease. Recently, Ishii et al. (14) also reported that glutathioneS-transferase (GST) P1 (GSTP1) gene polymorphism and susceptibility to COPD are correlated. GSTP1 is a member of the GST gene family, which consists of major detoxification enzymes, and has been identified in alveoli, alveolar macrophages, and respiratory bronchioles in the peripheral lungs (6). GSTs consist of a superfamily of dimeric phase II metabolic enzymes that catalyze the conjugation of reduced glutathione with various electrophilic compounds (21). Thus GSTP1 may play an important role in cellular defense by detoxifying a variety of toxic substrates in TS.

In this study, we first determined the viability of human embryo lung fibroblast-derived (HFL1) cells to cigarette smoke extract (CSE) at different concentrations using annexin V-phycoerythrin (PE) and 7-amino-actinomycin D (7-AAD). Then we carried out transfection of the HFL1 cells with GSTP1 sense and antisense vectors to examine the protective effects of GSTP1 against the cytotoxicity of CSE.


Cell culture.

HFL1 cells were obtained from Riken Cell Bank (Tsukuba City, Japan) and maintained in Dulbecco's modified Eagle's medium (DMEM; GIBCO BRL, Life Technologies) containing 10% fetal bovine serum (FBS; Interon), 100 U/ml of penicillin, and 100 μg/ml of streptomycin. The cells were passaged twice a week and used for experiments between passages 14 and 18.

Preparation of CSE solutions and exposure.

CSE was prepared as previously described (29), with a slight modification. Ten pieces of commercial cigarettes (Mild Seven; 84 mm long, with a diameter of 8 mm; purchased from Japan Tobacco, Tokyo, Japan) were smoked continuously by a pump-smoke machine named Hamburg II (Leybold-Heraes, Hamburg, Germany), and this smoke was used to generate 10 ml of TS-bubbled DMEM without FBS as shown in Fig.1. This TS-bubbled DMEM was passed through a 0.22-μm-pore filter (Millipore Japan, Tokyo, Japan) to remove large particulates that would not reach the lower airways (31). This medium was defined as 100% CSE, and this CSE was diluted with DMEM without FBS in the following experiments. Nicotine, one of the major components of TS, in 100% CSE was measured by gas chromatography-mass spectrometry as previously described (9) to analyze the stability of the components of the CSE. The concentration of nicotine was 10,157.4 ± 741.3 ng/ml (n = 5 experiments).

Fig. 1.

Preparation of cigarette smoke extract (CSE) solution. Ten pieces of commercial cigarettes were smoked continuously by a peristaltic pump-smoke machine, and this smoke was drawn and was used to generate 10 ml of tobacco smoke-bubbled DMEM without fetal bovine serum (FBS).

Construction of GSTP1 sense and antisense vectors.

Plasmid vector pIRES2-EGFP carrying the enhanced green fluorescent protein (EGFP) gene expressed in an internal ribosomal entry site (IRES)-dependent manner was purchased from Clontech Laboratories (Palo Alto, CA). The cDNA of GSTP1 inserted into pUC19 was a gift from Prof. M. Muramatsu (Department of Biochemistry, Saitama Medical School, Saitama, Japan). Restriction enzymes were purchased from New England Biolabs. The GSTP1 cDNA fragment was prepared by EcoRI digestion of the plasmid and inserted into the EcoRI site of pBluescript II SK(−) (Stratagene) with a ligation kit (Takara Shuzo) to construct pBS-GSTP1. Then, the SacI-SalI fragment of pBS-GSTP1 containing the cDNA was inserted between theSacI and SalI sites of pIRES2-EGFP. The resulting plasmid had the GSTP1 coding sequence in the direction that transcription by the cytomegalovirus promoter was expected to generate the sense RNA. Therefore, this plasmid was defined as the sense vector (Fig. 2). The antisense vector, which generates the GSTP1 antisense RNA, was constructed by inserting the XhoI-PstI GSTP1 fragment prepared from pBS-GSTP1 between the XhoI and PstI sites of pIRES2-EGFP (Fig. 2).

Fig. 2.

Simplified construct of the sense and antisense vector. The cDNA of glutathione S-transferase P1 (GSTP1) was integrated into the multiple cloning site of pIRES2-EGFP, a eukaryotic expression vector that coexpresses enhanced green fluorescent protein (EGFP) in the transfected cells in the sense (A) and antisense (B) directions. Arrows, direction of the insert. The digestive regions with the restriction enzymes used for the construction of the plasmids are also shown. PCMV, promoter of cytomegalovirus; IRES, internal ribosome entry site.

DNA transfection.

The cells were seeded in a six-well culture plate, and 24 h later, DNA transfection was carried out with the FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Basel, Switzerland) according to the manufacturer's protocol. Forty-eight hours after transfection, the cells were analyzed.

Preparation of the cells for the experiments.

For analyzing cellular viability, 60–80% confluent cells were trypsinized and plated in six-well culture plates at a density of 1 × 105 cells/well in DMEM with 10% FBS. After 24 h, the CSE solution diluted with DMEM without FBS was added to each well. For analysis of GSTP1-transfected cells, the CSE solution was added 48 h after transfection.

Flow cytometry.

GSTP1 expression in the cells transfected with the pIRES2-EGFP-based vectors was detected as previously described (19), with a slight modification. In brief, the cells trypsinized and harvested 48 h after transfection were fixed with 1% paraformaldehyde and treated with a fluorescence-activated cell sorter (FACS) permeabilizing solution (Becton Dickinson). The membrane-permeabilized cells were reacted with mouse anti-human GSTπ monoclonal antibody (DAKO), followed by R-PE-conjugated rat anti-mouse Ig, κ light chain monoclonal antibody (PharMingen). The cell emissions were then analyzed by flow cytometry with FACScaliber (Becton Dickinson), with fluorescence channel (FL) 1 to detect EGFP (excitation wavelength 488 nm; emission wavelength 507 nm) expression for determining the transfection efficiency and FL2 to detect PE (emission wavelength 570 nm).

Apoptosis and necrosis of the cells were analyzed as previously described (2), with a slight modification. Surface exposure of phosphatidylserine, thought to be the early signal of apoptosis, was monitored with the high-affinity binding properties of annexin V for phosphatidylserine. Membrane disruption indicating necrosis was assessed with 7-AAD uptake. The cells were washed twice with phosphate-buffered saline (PBS) and incubated for 15 min with 5 μl of annexin V-PE (PharMingen) and 5 μl of 7-AAD solution (Via-Probe, PharMingen) in 1× binding buffer (10 mM HEPES, pH 7.4 with NaOH, 140 mM NaCl, and 2.5 mM CaCl2) at a final concentration of 1 × 105 cells/100 μl solution. Then, the cells were analyzed by flow cytometry for fluorescence from PE-conjugated annexin V (emission wavelength 570 nm) and 7-AAD (emission wavelength 660 nm), which was analyzed with FL3. Dual labeling of the cells with a combination of annexin V and 7-AAD allowed the detection and differentiation of annexin-negative and 7-AAD-negative viable cells, annexin-positive and 7-AAD-negative apoptotic cells, and 7-AAD-positive necrotic cells. The cell percentages for apoptosis and necrosis were calculated concretely as followsapoptosis(%)no.of annexinV­positive and7­AAD­negative cellsno.of7­AAD­negative cells×100 necrosis(%)no.of7­AAD­positive cellsno.of all cells×100 For the vector-transfected cells, the gate for FL1 was set up so that only EGFP-positive cells were analyzed. Under these conditions, induction of apoptosis and necrosis was examined by measuring annexin V binding and 7-AAD uptake, respectively (Fig.3).

Fig. 3.

Fluorescence-activated cell sorter (FACS) analysis of apoptosis and necrosis of the transfected cells by triple labeling. Transfected cells were labeled with EGFP, analyzed in fluorescence channel (FL) 1, and selected by gating. For investigation of cell viability, these cells were labeled simultaneously with phycoerythrin (PE) and 7-amino-actinomycin D (7-AAD) as described inmaterials and methods and analyzed in FL2 and FL3.

Statistical methods.

Results are presented as means ± SD. Comparisons were made by Student's t-test or analysis of variance when appropriate. These analyses were performed with the StatView J-4.5 application program (SAS Institute). P values < 0.05 were considered to indicate significance.


CSE induced apoptosis at lower concentrations and necrosis at higher concentrations in human lung fibroblasts.

To characterize the effects of CSE on the viability of human lung fibroblasts, HFL1 cells were treated with various concentrations of CSE for different periods of time (Fig. 4). Apoptosis was induced in the presence of 5–25% CSE rather than other concentrations, and the proportion of apoptotic cells was highest after 20–24 h of CSE exposure. There was a statistical difference between the proportion of apoptotic cells in 0% CSE and that in 25% CSE (P = 0.0366). Under these conditions, necrotic cells increased in a time-dependent manner, especially after 20–24 h. In contrast, in the presence of a 50% or higher concentration of CSE, the cells became necrotic at earlier time points and apoptotic cells were rarely detected. Based on these results, treatment with 25% CSE was used in the following experiments to study the effects of GSTP1 on the viability of CSE-exposed HFL1 cells.

Fig. 4.

Apoptosis (A) and necrosis (B) of HFL1 cells after indicated times of exposure to CSE of indicated concentrations. Values are means ± SD;n = 3 experiments. * Significantly different from 0% CSE, P < 0.05.

Levels of GSTP1 expression in the vector-transfected cells.

To analyze GSTP1 expression in the cells transfected with different vectors, successful DNA transfection was monitored by GFP expression, and the level of GSTP1 expression was measured by immunofluorescent flow cytometry. As shown in Fig. 5, HFL1 cells transfected with the sense vector produced a higher level of GSTP1 than the cells transfected with the control vector. In contrast, the level of GSTP1 expression in the HFL1 cells transfected with the antisense vector was markedly lower than that of the control cells.

Fig. 5.

Intracellular GSTP1 level of the vector-transfected cells was analyzed by FACScaliber. The results are from 1 representative experiment of 3. 1st Ab (−), primary antibody negative control.

To characterize these cell populations in a more quantitative manner, each population was divided into three groups, “low,” “moderate,” and “high,” based on the level of GSTP1 expression. The border between the low and moderate groups was defined as the fluorescence intensity below which 90% of the negative control cells, which were not incubated with the anti-GSTP1 antibody but only with the secondary antibody, were included. Similarly, the border between the moderate and high groups was defined as the fluorescence intensity below which 90% of the cells transfected with the control vector, which were incubated with both the anti-GSTP1 and secondary antibodies, were included. According to this categorization, the majority of HFL1 cells transfected with the control vector were in the moderate group, and ∼8 and 13% were the high and low groups, respectively (Table 1). In contrast, nearly 40% of HFL1 cells transfected with the sense vector were in the high group, whereas >70% of the antisense vector-transfected cells were in the low group (Table 1). The proportion of the high group of cells transfected with the sense vector was significantly larger than that of the control vector-transfected cells (P = 0.0062). Similarly, the proportion of the low group of cells transfected with the antisense vector was significantly larger that that of the control vector-transfected cells (P = 0.0087). These results strongly suggest that transfection of the GSTP1 sense and antisense vectors led to overexpression and repression of GSTP1, respectively.

View this table:
Table 1.

Intracellular GSTP1 level of the transfectants of the vectors

Effects of GSTP1 sense and antisense vector transfection on induction of apoptosis and necrosis by CSE.

To examine the effects of GSTP1 on CSE-mediated cellular damage, HFL1 cells transfected with the GSTP1 sense or antisense vector were exposed to 25% CSE. As shown in Fig.6 A, no significant difference was observed between the cell populations transfected with the different vectors with regard to the proportion of apoptotic cells.

Fig. 6.

Kinetics of 25% CSE-induced apoptosis (A) and necrosis (B) of the transfected cells. Values are means ± SD; n = 3 experiments.

In contrast, the level of necrosis was clearly affected by vector transfection (Fig. 6 B). The cells transfected with the GSTP1 antisense vector were significantly more susceptible to CSE-induced necrosis than the control vector-transfected cells (P< 0.0001). Although the cells transfected with the GSTP1 sense vector and the control vector showed similar levels of necrosis until 16 h after CSE exposure, the sense vector-transfected cells appeared to be more resistant to CSE-induced necrosis at later time points (P = 0.0014).


In this study, we demonstrated the ability of CSE to induce apoptosis and necrosis in human lung fibroblasts (HFL1 cells) in a dose-dependent manner. Aqueous extracts of cigarette tar have been used to model the types of species that lung cells are exposed to in smokers' lungs (4, 13). Although the CSE with our modified method is thought to include the gas-phase smoke components, these components are supposed to vaporize in the early phase and water-soluble components remain during the whole experimental period, which was thought to be similar to the situation in the “smoker's lungs.” Additionally, tar microdeposits, which the lung fluid is thought to wash out continuously, were removed with a 0.22-μm-pore filter, also as previously described (4, 13). Although the redox-sensitive tar components are soluble in water with our method, collection of the extract in methanol or acetone would have provided better extraction of the components in the gas phase (23). The concentration of nicotine was 10,157.4 ± 741.3 ng/ml in 100% CSE in our study. Because it was reported that the concentrations of nicotine in the serum and urine were 14.7 and 1,749.9 ng/ml, respectively (15), and that the half-life of the concentration of nicotine in the serum was ∼30 min, we considered that 100% CSE and the dilution were appropriate to use in our study.

Cytotoxic effects of cigarette smoke on mammalian cells have previously been reported for the A549 alveolar epithelial cell line (20) and U937 premonocytic cells (2, 33). Our present study is the first to show the induction by CSE of apoptosis and necrosis in human lung-derived fibroblasts. The proportion of apoptosis was <10% in most cases, and we considered that this was very small. Indeed, the proportion of apoptotic lung fibroblasts was reported to have been 20–40% with thiol depletion or lovastatin stimulation (1,30). In those studies, apoptosis was analyzed by terminal deoxynucleotidyltransferase dUTP nick end labeling or morphological methods, and with these methods, detected apoptotic cells included not only the early apoptotic cells but also the secondary necrotic cells (postapoptotic cells). The latter part of the “apoptotic cells” was not detected as apoptosis with our method with annexin V, which might lead the proportion of apoptosis in our experiments to be smaller than in the other reports.

We also analyzed the effects of GSTP1 sense and antisense vector transfection on the susceptibility of the cells to CSE-induced apoptosis and necrosis. In the presence of 25% CSE, transfection with the GSTP1 sense and antisense vectors led to a significant decrease and increase in necrotic cells, respectively. Although the cells transfected with the sense vector and the control vector showed similar levels of necrosis until 16 h, the sense vector-transfected cells appeared to be more resistant to CSE-induced necrosis at later time points. The sense-transfected cells are thought to have a more metabolizing capacity than the control cells, and the xenobiotics are supposed to be less accumulated in the sense-transfected cells after 16 h of CSE exposure, which might lead to resistance in the sense vector-transfected cells. In contrast, transfection with the sense and antisense vectors did not show significant effects on the cellular susceptibility to CSE-induced apoptosis.

Cigarette smoke is thought to be one of the most important causes of various lung diseases (22). Although many studies on the association between smoking and lung diseases were done and much evidence on the association has already been accumulated, the actual role of cigarette smoke in the pathogenesis of these diseases has not yet been fully elucidated (22). Cigarette smoke contains ∼4,000 species of substances including oxidants, carcinogenic agents, and other xenobiotics (12). Therefore, it is possible that some proteins, such as antioxidant and xenobiotic metabolizing enzymes, play critical roles in the protection of the lung against cigarette smoke (24). Supporting this possibility are previous studies that showed that glutathione protected HFL1 cells from hydrogen peroxide-induced apoptosis and necrosis (32) and that thiol depletion induced apoptosis in human lung fibroblasts (1). According to these reports and the fact that other important enzymes including glutathione peroxidase and other GSTs also use glutathione as substrate and contribute to the detoxification of peroxides such as hydrogen peroxide, lipid peroxides, and various xenobiotics (24), we have not modified intracellular GSH concentration, which is also thought to have a crucial association with the activity of GSTP1.

In the present study, we examined the effects of GSTP1, a member of the GST family, because GSTs are known as antioxidative xenobiotic enzymes. GSTP1 is expressed in alveoli, alveolar macrophages, and respiratory bronchioles in the peripheral lung (6). It has been suggested that overexpression of GST increases resistance to oxidative stress (36) and to various drugs including carcinogenic (10) and chemotherapeutic agents (8). Because GSTP1 has catalytic activity against base propenals (17), it may also have a major role in the protection of DNA and contribute to cellular viability. In this study, we showed that GSTP1 had a protective effect against CSE, leading to a reduction in necrotic cells. Conversely, a decrease in GSTP1 expression level appeared to enhance cellular susceptibility to CSE-induced necrosis. Intriguingly, transfection efficiency of the GSTP1 antisense vector (2.12 ± 0.38%) was generally lower than that of the sense and control vectors (4.28 ± 0.69 and 4.25 ± 0.82%, respectively). This may indicate that inhibition of GSTP1 expression by the antisense vector by itself may also affect cellular viability.

Emphysema is one of the most important lung diseases associated with cigarette smoking. Although there are two major hypotheses on its pathogenesis, namely the “protease-antiprotease” hypothesis and the “oxidant-antioxidant” hypothesis (18, 28), the actual mechanism of disease induction is still unknown. The results of this study suggest that fibroblasts exposed to cigarette smoke may be killed through necrosis or apoptosis and that the level of extracellular fibrous proteins, such as collagen and elastin, normally produced by fibroblasts, may be decreased. This resembles the collagenase-overexpressed situation in which the lungs of transgenic mice were reported to become emphysematous (7). Although lung fibroblasts are not thought to be a primary target for the direct exposure to TS, we considered that they are one of the targets, especially for water-soluble components of TS that are thought to pass the basement membrane, and that they have an important role as the source of the extracellular matrix, which is what led us to use lung fibroblasts for this study. Alternatively, but not mutually exclusively, cigarette smoke may enhance local inflammation of the lung tissue because cellular necrosis leads to inflammatory responses (34). Activation of proteases in the emphysematous lung (26) might be a result, rather than a cause, of the inflammatory responses.

In conclusion, we report that cigarette smoke induced apoptosis and necrosis of human lung fibroblasts and that this cytotoxic effect was reduced with GSTP1, one of the important xenobiotic metabolizing enzymes, by decreasing necrotic cells. This suggests that GSTP1 may have a protective role in human lungs against cigarette smoke and contribute to preventing them from developing cigarette smoke-related lung diseases including emphysema.


We thank Prof. Masami Muramatsu (Department of Biochemistry, Saitama Medical School, Saitama, Japan) for the cDNA of glutathioneS-transferase P1 (GSTP1). We also thank Dr. J. Nakajima (Department of Thoracic Surgery, Tokyo University Hospital, Tokyo, Japan) for permission to use the FACScaliber. We are indebted to Prof. J. Patrick Barron (Medical Communications Center, Tokyo Medical University, Tokyo, Japan) for review of this manuscript.


  • This study was partly supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan; Health Sciences Research Grants from the Ministry of Health and Welfare of Japan; and the Smoking Research Foundation of Japan.

  • Address for reprint requests and other correspondence: T. Matsuse, Dept. of Pulmonary Medicine, Yokohama City Univ. Medical Center, 4-57, Urahune-cho, Minami-ku, Yokohama City 232-0024, Japan.

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract