Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Jul 2.
Published in final edited form as: Cancer Res. 2008 Oct 15;68(20):8473–8481. doi: 10.1158/0008-5472.CAN-08-0131

Nicotine Promotes Mammary Tumor Migration via A Signaling Cascade Involving PKC and cdc42

Jinjin Guo 1, Soichiro Ibaragi 2, Tongbo Zhu 1, Ling-Yu Luo 1, Petra S Huppi 3, Chang Yan Chen 1
PMCID: PMC3698481  NIHMSID: NIHMS479643  PMID: 18922921

Abstract

Nicotine, one of major components in tobacco, has been shown to be at high concentrations in the blood stream of cigarette smokers. However, the mechanisms of how nicotine affects tumor development and whether nicotine is a potential carcinogen for malignancies induced by second-hand smoking are not fully understood yet. Here, we investigate the signaling pathways by which nicotine potentiates tumorigenesis in human mammary epithelial-like MCF10A or cancerous MCF7 cells. We demonstrate that human MCF10A and MCF7 cells both express four subunits of nAChR. The treatment of these cells with nicotine enhances the activity of protein kinase C (PKC) α without altering the expression level of this kinase. Nicotine also stimulates [3H]thymidine incorporation into the genome of these cells as well as forces serum-starved cells to enter S phase of the cell cycle, resulting of growth promotion. Importantly, upon nicotine treatment, the mobility of MCF10A and MCF7 cells are enhanced, which can be blocked by the addition of nAChR or PKC inhibitor. Experiments using siRNA knockdown or ectopic expression of cdc42 demonstrated that cdc42 functions as a downstream effector of PKC and is crucial in the regulation of nicotine-mediated migratory activity in the cells. Together, our findings suggest that nicotine, through interacting with its receptor, initiates a signaling cascade that involves PKC and cdc42, and consequently promotes migration in mammary epithelial or tumor cells.

Introduction

Tobacco contains various components, of which many are known to be carcinogenic and/or mutagenic (1, 2). Although studies have linked cigarette smoking with various onsets of human malignancies, little is known about how second-hand smoking promotes tumor development or causes the onset of cancer. Nicotine has been identified as one of the important constituents of tobacco (3). Through interacting with nicotine acetylcholine receptor (nAChR), nicotine functions on either the motor endplate of muscle or at the central nervous system responsible for tobacco addiction (4–6). Recently, it has been recognized that nicotine is able to activate various intracellular signaling pathways in non-neuronal cells, indicating that nicotine may possess a function for tumor promotion (7–11). After engaging its receptor, nicotine rapidly activates mitogenic-related, intracellular signaling pathways in endothelial cells or keratinocytes (11, 12). Emerging evidence demonstrated that nicotine potently induces secretion of different types of calpain from lung cancer cells, which then promotes cleavage of various substrates in the extracellular matrix to facilitate metastasis and tumor progression (10).

nAChR is composed of nine α-subunits (α 2–10) and two β-subunits (α2 and 4) (13, 14). The receptor has been demonstrated to be expressed on the surface of various non-neuronal cells, such as lung epithelial cells, keratinocytes, vascular smooth muscle cells (7–14). On the surface of these cells, the α 3, α 5, α 7, α 9, β 2 and β 4 subunits are expressed. Among the subunits, α 3, α 5, β 2 and β 4, in different combinations, form heteromeric channels (11, 15–18). The homomeric channel often is composed by several α 7 or α 9 subunits. These hetero- or homomeric channels are highly Ca++ permeable, which allow releases of Ca++ from intracellular stores to the cytosol of cells (19, 20). In human epidermal cells, the interaction of nicotine and its receptor have been shown to activate calmodulin-dependent protein kinase II, PKC, phosphodylinositol-3-kinase (PI3K)/Akt and Rac/cdc42 that are often involved in the regulation of cell adhesion, migration and invasion (11, 21, 22). The activation of nicotine receptor is able to activate Ras/Raf/MEK/ERK signaling (11). Recently, it has also been reported recently that upon the ligation of nAChR by nicotine, the tyrosine kinase JAK-2 and transcription factor STAT-3 in macrophages are activated (22).

PKC consists of more than 12 different isoforms of protein-serine/threonine kinases that are important components of phospholipase-coupled growth factor receptor signaling pathways (23–26). The enzymatic activities of PKC isoforms are regulated by diacylglycerol (DAG) or Ca++, which divide these kinases into three subclasses: DAG-dependent (PKC δ, ε, θ, and μ), DAG/Ca++-dependent (PKC α, β I and β II) and atypical (PKC ζ, λ and τ) isoforms. These isoforms have different tissue distributions and are involved in different biological processes, including cell growth, migration, apoptosis and differentiation (23–26). In the blood stream of smokers, nicotine is found to be at pharmacological concentrations of 90–1000 nM, which has been shown to be able to activate PKC in cultured human or murine lung cancer cells (27, 28). The activation of PKC has been shown to be responsible for Bcl-2 phosphorylation, which, in turn, antagonizes drug-induced apoptosis in lung cancer cells (28). It has been reported that nicotine can elicit the activity of Ras or Raf-1 (9, 11, 29). In addition, studies have demonstrated that Akt was phosphorylated in the lungs of nicotine-treated mice and in human lung cancer tissues derived from smokers (30). All these observations suggest that nicotine is able to promote pro-survival activity in cells and is important for tumorigenesis.

Cell migration or tumor metastasis is a crucial process in tumor development. Numerous factors play a role in the regulation of this process, which includes growth factors, kinases, phosphatases as well as extracellular matrix components. Growth-related receptors, when activated by corresponding ligands, contribute to cell proliferation and migratory or invasive capacity of cells. After recruiting downstream effectors, growth-related receptors often exert their functions by organizing their donwstream effectors, such as PKC, PI3K or cdc42 to initiate signaling cascades, which affect various biological processes, including cell migration and cancer cell invasion (31–36). Nicotine has been demonstrated to induce the phosphorylation of different subtypes of calpains, resulting in enhanced cell migration or, in more specifically, lung cancer metastasis (10, 34).

Although numerous studies indicated the role of nicotine exposure in tumor promotion, little is known about the effect of nicotine on breast tumor development, especially on the metastatic process of breast cancer. Here, we demonstrated that 4 different subunits of nAChRs were expressed in MCF10A and MCF7 cells, and the expression of these subunits was not affected by nicotine exposure. However, the treatment with nicotine augmented the activity of PKC in these cells. Although nicotine stimulated DNA synthesis as well as the S phase entry of these cells under serum-starvation conditions, the suppression of PKC activity did not significantly block the growth promoting effects of nicotine. Furthermore, nicotine exposure induced both “normal” mammary epithelial or mammary cancer cells to migrate under serum-starvation conditions, which could be suppressed by either nAChR or PKC inhibitor. Acting as a downstream effector of PKC, cdc42 was identified as a major effector in this nicotine-mediated action to promote migratory activity. Together, our investigation suggests that nicotine mobilizes PKC and cdc42 signaling in both “normal” mammary epithelial and mammary cancer cells to promote cell migratory activity.

Methods and Materials

Materials

Nicotine and nAChR inhibitor mecamylamine hydrochloride (MCA) were purchased from Sigma-Aldrich, Inc. (St. Louis, MO, USA). The PKC inhibitor 12-(2-cyanoethyl)-6, 7, 12, 13-tetrahydro-13-methyl-5-oxo-5H-indolo (2, 3-α’ pyrolo (3, 4-c)-carbazole (GO6976) was obtained from EMD Chemicals Inc. (San Diego, CA, USA). The anti-PKC antibody was purchased from BD Parmingen (Lo Jolla, CA, USA). Wortmannin and UO16 were obtained form Cell Signaling Tech. (Calbiochem, San Diego, USA).

Cell Culture and Transfection

Human breast cancer MCF10A and MCF7 cells were purchased from ATCC (Manassas, VA, USA). MCF10A cells were cultured in DMEM/F12 medium supplemented with 5% donor horse serum, 20 ng/ml epidermal growth factor (EGF), 10 µg/ml insulin, 0.5 µg/ml hydrocortisone, 100 ng/ml cholera toxin and antibiotics. MCF7 cells were maintained in Dulbecco’s modified Eagle’s (DME) medium with 10% fetal calf serum, 4 mM L-glutamine and antibiotics.

Cdc42 was inserted into MSCV retroviral vector and subsequently transiently infected into the cells. Small interference RNAs were chemically synthesized and inserted into a lentiviral vector. The vector carrying targeted siRNA sequence was introduced into the cells. The antisense strand siRNA was targeted against GTPase using 21-nucleotide sequences (5’-AAGAAGTCAAGCATTTCTGTC-3’) for RhoA, (5’-AAGATAACTCACCACTGTCCA-3’) for cdc42, and (5’-AAGTTCTTAATTTGCTTTTCC-3’) for Rac1 (35).

PCR and Real-Time PCR

Total RNAs were extracted from cultured cells with or without treatments using RNease Mini Kit (Qiagen, Valencia, CA, USA) following the protocol provided by the manufacturer. Primers for the genes encoding the subunits of nAChR were designed with the assistance of the Primer Express software (Applied Biosystems, Foster City, CA, USA). For Real-Time PCR, the gene expression was normalized using GAPDH as the control.

PKC Enzymatic Assay

After various treatments, cells were lysed in the lysis buffer (25 mM Tris-HCl, Ph 7.5, 1% NP-40, 20 mM MgCl2, 150 mM NaCl). The lysates were normalized for protein concentration. The equal amount of total proteins was immunoprecipitated with anti-PKC α antibody and the immunoprecipitates were analyzed for PKC α activity using a PKC enzymatic kit (Millipore, Bilerica, MA, USA). Briefly, the immunoprecipitates were mixed with the substrate cocktail, the inhibitor cocktail, and the lipid activator in microcentrifuge tubes. After adding Mg++/ATP cocktail containing [γ-32P] ATP, the samples were incubated at 30°C for 10 min. and then spotted on P81 phosphocellulose papers. The radioactivity of the 32P-incorporating substrate was measured by a scintillation counter.

[3H]thymidine incorporation

MCF10A or MCF7 cells were grown in Petri dishes until 60–70% confluence. Controls and treated cells were done in triplicate. The cells were cultured in the medium containing 0.5% serum for 24 h. Subsequently, the cells were grown in the medium containing 0.5% or 10% serum plus 4 µCi/ml of [3H]thymidine (Perkin Elmer Life Sciences, Waltham, MA, USA) with or without nicotine treatment. The cells were labeled for 4 h at 37° C. After precipitation with cold 10% trichloroacetic acid, the cells were dissolved in 0.5 ml of 0.1 M NaOH overnight at 4° C. The amount of radioactivity in each sample was counted using a scintillation machine.

Cell Proliferation Assay

The cells (2 × 105) were plated in 12-well plates and cultured in the medium containing 0.5% serum, which is designated as day 1. Subsequently, the cells with or without treated with nicotine or re-fed with 10% serum were grown for another 4 days. The numbers of viable cells were determined by trypan blue staining and counted daily using a hemocytometer.

Cell Cycle Analysis

Cell cycle distribution of DNA content was measured using a flow cytometer as described (36). The cells, after the treatments, were fixed in 65% DME medium and 35% ethanol for 2–4 h at 4° C and stained with propidium iodide. Subsequently, the DNA profiles were analyzed using Cell Quest software.

Cell Migration and Invasion Assay

Following various treatments, cell migration was analyzed using a Boyden cell migration assay plates (Neuroprobe, Inc., Gaithersburg, MD, USA) according to the manufacturer’s instructions. Untreated or treated cells were labeled with Dil-labeled acetylated low-density lipoprotein (10 µg/ml) for 4 h. Cells were trypsinized and resuspended in phenol red-free medium at 10,000-cells/25 µl. Each experiment was performed in triplicate and three separate experiments were performed in each experimental group.

Monolayer Wound Healing Assay

Cell were seeded and allowed to grow to 90% confluent. A cell scraper was used to wipe away the cell monolayer on one side of the start line that had been drawn on the bottom of the plate. Wounded monolayer was washed 4 times with the medium to remove cell debris. Cells were then treated. Twenty-four hours later, the wounded areas were photographed at various time points with the assistance of the landmarks drawn on the undersurface of the plate.

Experimental Lung Metastasis Assay

MCF7 cells (1 × 106) were injected into the tail veins of nude mice. A group of 15 mice were peritoneally injected with nicotine (30 µg/mouse) every 2 days (37). Another group of the same numbers of mice was left untreated. Five mice from each group were sacrificed at 3 and 4 weeks after tumor cell injection. The lung tissues were isolated and slides were prepared. After stained with Heomatoxylin Eosin (HE), the approximate number of foci of blue-stained metastatic tumor cells present in the lung was estimated with an Olympus dissecting microscope. The scoring system was used to score the degree of metastasis, based on the estimated counts: −, no blue cells; +, about 1–50 blue cells; ++, about 50–100 blue cells; +++, about 100–200 blue cells (37).

Immunoblotting

Following treatments, cell lysates were prepared, and proteins were separated by SDS-PAGE gels. Membranes were incubated with the designated primary antibody (1:1000 for all antibodies) overnight in a cold-room at 4° C. Bound primary antibodies were reacted with corresponding second antibodies for 2 h and detected by chemiluminescence.

Statistical Analysis

Three to five independent repeats were conducted in all experiments. Error bars represent these repeats. A Student’s T test was used and a p value of <0.05 was considered significant.

Results

Activation of PKC by nicotine treatment in breast epithelial and cancer cell lines

It is known that nAChRs are expressed not only on the surface of neuronal cells, but also lung epithelia and vascular endothelial cells, through which nicotine affects various biological or patho-biological processes including angiogenesis, ischemia, or growth promotion (7–12). To explore whether nicotine could influence breast cancer development, the gene expression of nAChRs in normal breast epithelial-like MCF10A cells and breast cancer MCF7 cells were examined using PCR technique. Total mRNAs from these cells were isolated and subsequently the gene expression of each nAChR subunit was analyzed (Fig. 1a). The results from PCR revealed that nAChR α 5, 7, 9 and β 4 were consistently expressed in both cells. The expressions of the genes of other nAChR subunits were undetectable in these cells (data not shown). To determine whether nicotine could stimulate nAChR expression, the cells were treated with nicotine and the levels of nAChR α 5, α 7, α 9 and β 4 expression were measured by real-time PCR (Fig. 1b). The amount of the gene expression of each nAChR subunit in untreated cells was similar as that in treated cells.

Figure 1.

Figure 1

Open in a new tab

The expression of the subunits of nAChR in MCF10A and MCF7 cells. a. The total RNA from each cell line was isolated. Equal amounts of RNA were reverse-transcribed, and the expressions of nAChR α and α subunits were identified by PCR. b. With or without nicotine treatment (0.5 µM) for 24 h, the total RNAs were obtained. After reverse-transcription, real-time PCR was performed. The abundance of nAChRs was normalized to actin. Error bars represent the standard deviation over 3 independent experiments.

Nicotine exposure has been shown to rapidly increase extracellular signal-regulated kinase (ERK) activation aw well as serine phosphorylation of signal transducer and activator of transcription (STAT)-1 and 3. Studies also demonstrated that nicotine is able to stimulate PI3K/Akt and Ras/Raf pathways (21, 22). We previously showed that nicotine treatment upregulates PKC activity in mouse lung epithelial cells (9, 29). It has been reported that nicotine receptor engagement correlates well with the migration promotion of human lung cancer cell, in which PKC plays a crucial role (10). Since the phorbol ester and calcium-dependent isoforms of PKC are predominantly expressed in many types of cells and their activity has been shown to be involved in growth promotion (23–26), the expression of PKC α in MCF7 or MCF10A cells in regards to nicotine treatment was first examined by immunoblotting (Fig. 2a). A similar level of PKC α expression was recognized by the antibody in both breast cancer cell lines with or without the treatment, indicating that nicotine plays no role in the expression of this kinase. To determine whether nicotine treatment affects PKC α activity, the kinase activity was measured, using a PKC-specific kinase activity assay (Fig. 2b). After growing the cells in the medium containing 0.5% serum for 24 h, a basal activity of this kinase was detected in MCF7 and MCF10A cells (Fig. 2b, lanes 1 and 6). In contrast, the kinase activity was dramatically increased after nicotine exposure under serum-starvation conditions (about 4–5 folds) (Fig. 2b, lanes 2 and 7) or re-fed with 10% serum (about 5–6 folds) (Fig. 2b, lanes 4 and 9). The addition of MCA (a nAChR inhibitor) abrogated the induction of PKC α activity mediated by nicotine in serum-starved cells (Fig. 2b, lanes 3 and 8), but had no negative effects on the cells re-fed with serum (Fig. 2b, lanes 5 and 10). The expression and activity of other PKC isoforms in nicotine-treated cells did not changed in comparison with that in untreated cells (data not shown). Therefore, the data suggest that nicotine, through its receptor, activates PKC α in these breast cells.

Figure 2.

Figure 2

Open in a new tab

PKC activation in nicotine-treated MCF cells. a. The cells were treated with nicotine (0.5 µM) for 30 min, cell lysates were extracted for immunoblotting analysis. β-Actin was used to determine equal loading of total protein per lane. b. After nicotine treatment in the presence or absence of MCA (50 nM), the cells were harvested and lysates were prepared for PKC α activity assay. Error bars represent the standard deviation over 3 independent experiments.

Effect of nicotine engagement on breast cancer growth

Emerging evidence demonstrates the potential growth promoting activity rendered by nicotine (21, 22). Increase in cell proliferation is a crucial step in tumorigenesis. Therefore, we used [3H]thymidine incorporation assay to test if nicotine treatment has an effect on DNA uptake in MCF7 (Fig. 3a) and MCF10A cells (Fig. 3b). After growing in the medium containing 0.5% serum for 48 h, MCF7 and MCF10A cells were treated with nicotine under the serum-starvation condition or after re-feeding with 10% serum in the presence of [3H]thymidine. Subsequently, rates of DNA synthesis were measured. Under serum depletion condition, little [3H]thymidine incorporation was observed in either MCF7 or MCF10A cells (Figs. 3a and b, lane 1). A moderate amount of [3H]thymidine (about 2–3 folds) was incorporated in nicotine-treated cells under serum-starvation conditions (Figs. 3a and b, lane 2). The higher rate of [3H]thymidine intake (4–5 folds) was seen in the cells re-fed with 10% serum (Figs. 3a and b, lane 3). However, the addition of nicotine and 10% serum into starved cells caused an even greater increase of [3H]thymidine incorporation into their genomes (about 5–6 folds) (Figs. 3a and b, lane 4). We further determined the role of nAChR or PKC in nicotine-mediated DNA synthesis promotion. The addition of MCA or GO6976 blocked [3H]thymidine incorporation in nicotine treated MCF7 and MCF 10A cells (Figs. 3a and b, lanes 5 and 6). The effect of nicotine on cell proliferation was confirmed by the assay for cell growth. After 24 h of serum starvation, MCF7 or MCF10A cells were grown in the medium containing 0.5%, 10% serum or with nicotine or for consecutive 4 days (Figs. 3c and d). The cells exposed to nicotine, like those cultured in the growth medium, were consistently dividing in the medium lacking serum over the testing period.

Figure 3.

Figure 3

Figure 3

Open in a new tab

Effect of nicotine exposure on the proliferation and cell cycle progression in MCF10A and MCF7 cells. a and b. Cells were grown in the medium containing 0.5% serum for 24 h, and then subjected to various treatments. Subsequently, the rate of [3H] incorporation was measured by a scintillation counter. Error bars represent the standard deviation over 5 independent experiments (n = 5, *, p< 0.05) c and d. After culturing in the medium containing 0.5% serum, cell proliferation of MCF7 and MCF10A with or without nicotine treatment was measured. Error bars represent the standard deviation over 5 independent experiments. e. After cultured in the medium containing 0.5% serum, cells were treated with nicotine or re-fed with 10% serum for 4 h, and DNA profiles of the cells were analyzed by a flow cytometer. f. The plot represents the percentages of the cells in the S phase of the cell cycle. Error bars represent the standard deviation over 5 independent experiments.

Next, we tested if nicotine was able to promote cell cycle progression in the cells. After serum depletion for 24 h, MCF10A and MCF7 cells were stimulated with nicotine for 4 h and cell cycle analysis was performed to measure the percentage of the cells in the S phase (Fig. 3e and f). Under the serum depletion, there was in general a very low percentage of MCF10A and MCF7 cells in the cell cycle beyond S phase (less than 4%) (Fig. 3f, lanes 1 and 4). In contrast, nicotine exposure caused a more than 2 fold increase in the S phase population of both MCF10A and MCF7 cells in comparison with untreated controls (Fig. 3f, lanes 2 and 5). After re-fed with 10% serum, the percentage of both cells in the S phase was increased to about 3 folds, as expected (Fig. 3f, lanes 3 and 6). Together, the data indicate that nicotine partially mimics mitogenic stimulation and promotes cell cycle progression or proliferation in MCF10A or MCF7 cells.

Nicotine treatment promotes the migratory activity of MCF10A and MCF7 cells

Growth factors, such as EGF or IGF, not only stimulate cell proliferation but also migration. Nicotine has been shown to be able to activate various intracellular signaling pathways that promote cell growth or survival (21, 22). We further examined whether nicotine exposure affects migration of MCF breast cells, using the Boyden chamber and cell wound-healing assays. The Boyden chamber assay revealed that the migration phenomenon in MCF10A or MCF7 cells was undetectable when growing in the medium containing 0.5% serum (Fig. 4a, lanes 1 and 7). The addition of nicotine, under the condition of serum depletion, increased the number of the migratory cells (more than 2 folds) (Fig. 4a, lanes 2 and 8), the magnitudes of which are similar as those re-fed with 10% serum (Fig. 4a, lanes 3 and 9). The addition of GO6976 dramatically suppressed the nicotine-induced migratory activity in serum-starved MCF10A and 7 cells (Fig. 4a, lanes 4 and 10), but the inhibition of PI3K/Akt by wortmannin (Fig. 4a, lanes 5 and 11) or MAP/ERK signaling by UO16 (Fig. 4a, lanes 6 and 12) had no effect on the promotion of migration mediated by nicotine. Collectively, the data indicate that nicotine, through mobilizing PKC signaling pathway, promotes migration in the breast cancer cells.

Figure 4.

Figure 4

Open in a new tab

Acceleration of cell migration and wound healing by nicotine exposure. a. After cultured in the medium containing 0.5% serum for 24 h, cells were untreated (lanes 1 and 7), or treated with nicotine (lanes 2 and 8); 10% serum (lanes 3 and 9); nicotine plus GO6976 (lanes 4 and 10); nicotine plus wartmannin (lanes 5 and 11) and nicotine plus U016 (lanes 6 and 12). Cells then were labeled with 10 µg/ml Dil-labeled acetylated low density lipoprotein for 4 h at 37°C. Subsequently, cells were loaded in Boyden chamber and incubated overnight. The migratory cells at the bottom wells were measured by fluorescence. Error bars represent the standard deviation over 3 independent experiments. b. Cells were partially wiped away and then subjected to various treatments. Photos are cells migrating to the wiping-off area at 48 h. The line represents the edge of the cell monolayer immediately after wiping away the cells. c. Four weeks after the injection of MCF7 cells through tail veins of the mice, the lung tissues from mice without nicotine treatment (left) or with nicotine exposure (right) were isolated, coated on slides, and stained with HE. Arrows point to the area with metastatic tumor cells.

We then tested the effect of nicotine on MCF 7 cell migration using a wound-healing assay (Fig. 4b). MCF 7 cells were unable to migrate under serum-starvation condition. Following nicotine treatment, the starved cells displayed a high capacity to migrate to the wounded area. However, in the presence of MCA or GO6976, the migration mediated by nicotine was completely suppressed. The addition of UO16 had no effect on nicotine-induced MCF7 cell migration. Consistently, the treatment with wortmannin to block PI3K/Akt signaling did not affect nicotine-induced wound-healing (data not shown). Together, the data strongly suggest that nicotine, like growth factors, functions through its receptor and governs multiple downstream pathways. Among these signaling pathways, PKC appears to be involved in the regulation of the activity of cell migration.

To further test whether nicotine is able to promote metastasis, experimental lung metastasis assay was performed, which has been used frequently to detect micrometastases (37–39). MCF7 cells that are know to be poorly metastatic or invasive in nude mice (37–39) were employed. After injected MCF7 cells into the tail veins of nude mice, a group of the mice was peritoneally injected with nicotine every two days for 28 days and another group with the same amount of mice was left untreated. Subsequently, a metastatic phenotype was examined in the lungs harvesting from the mice treated with nicotine for 21 and 28 days or from untreated mice (Table 1). Four weeks after tumor cell injection, there were abundant blue-stained tumor cells deposited around vessels in the lung from nicotine treated mouse (Fig. 4c). In comparison, there was no micrometastasis in the lungs of the mice without the administration of nicotine. Thus, the in vivo data further support the notion that nicotine facilitates the process of tumor metastasis.

Table 1.

Nicotine Exposure Promotes i.v. Injected MCF7 Cells to Metastasize to the Lungs of Nude Mice

Group Cell line Time after
Injection		 Lung metastases
Incidence Extent
Control MCF7 21 days 0/5a −
Control MCF7 28 days 0/5 −
Exposure to MCF7 21 days 0/5 −
Exposure to MCF7 28 days 1/5 +++
Open in a new tab

one million MCF7 cells were injected into the tail veins of nude mice. One group of the mice was injected with nicotine and another group was left untreated. Mice were sacrificed at the indicated times, lungs were harvested and stained, and metastases were quantified. Score system: −, no blue cells; +, about 1–50 blue cells; ++, about 50–100 blue cells; +++, about 100–200 blue cells. Extent: Extent of metastases, according to scoring system;

a

number of mice with metastases/number of mice injected.

Cdc4 2 functions as a downstream effector of PKC to promote the migratory activity mediated by nicotine

GTPases have been reported to act downstream of PKC to regulate cell adhesion or migration (35). Therefore, it led us to explore the involvement of GTPases in nicotine-induced migratory promotion. In order to evaluate the individual role of each GTPase, the expression of RhoA, Rac1 and cdc42 and the inhibition of their expression by the siRNAs were first examined by immunoblotting (Figs. 5a and b). MCF7 cells express RhoA, Rac1 and cdc42. After introducing the corresponding siRNAs into MCF7 cells, the levels of protein expressions of these GTPases were dramatically suppressed (6–8 folds). The expression of cdc42 in MCF10A cells, after the introduction of siRNA-cdc42, was also examined and the amount of the protein was reduced (about 7 folds) in comparison with the control.

Figure 5.

Figure 5

Open in a new tab

Involvement of cdc42 in nicotine-mediated cell migration. a. After transfected the siRNAs into MCF7 cells, the expression of RhoA, Rac1 and cdc42 were examined by immunoblotting analysis. b. Expression of cdc42 in MCF10A cells after introducing siRNA-cdc42. c. Expression of cdc42 in MCF7 or MCF10A cells after transient introduction of cdc42. d. After ectopically expressing cdc42 or introducing the siRNAs, MCF7 (left panel) or MCF10A (right panel) cells were cultured in the medium containing 0.5% serum for 24 h prior to nicotine treatment. Subsequently, the Boyden chamber assay was performed. Error bars represent the standard deviation over 3 independent experiments (n = 3, *, p < 0.01).

Before testing the effect of these GTPases on nicotine-induced migratory activity, cdc42 was ectopically expressed in MCF 7 or MCF10A cells (Fig. 5c). A significant increased amount of exogenous cdc42 (about 4 folds) was detected by the antibody in both cells. Subsequently, Boyden chamber assays were then conducted (Fig. 5d). The suppression of cdc42 by the siRNA dramatically blocked the nicotine-induced migration in serum-starved MCF7 cells (Fig. 5d, left penal, lane 4). The introduction of the siRNA into the cells to suppress either RhoA or Rac1 did not affect nicotine-mediated migratory activity (Fig. 5b, left penal, lanes 5 and 6). Transient overexpression of cdc42 in MCF7 cells further augmented the magnitude of nicotine-mediated migratory activity (Fig. 5b, left penal, lane 7). Since the inhibitory effect of cdc42 siRNA on the migration was similar as that occurred after treated with PKC inhibitor GO6976 (comparing figure 3a, lanes 4 and 8), it indicates that both molecules are involved in this nicotine-mediated action. To determine the possible linear relationship between PKC and cdc42, serum-starved MCF7 cells with transient overexpression of cdc42 were treated with nicotine in the presence of GO6976 and then subjected to the Boyden chamber assay. The suppression of PKC dramatically suppressed the nicotine-mediated migratory effect the cells, regardless of the augmented expression of cdc42 (Fig. 5b, left penal, lane 8). The similar results were obtained from MCF10A cells (Fig. 5b, right panel). Overall, the data suggest that cdc42 participates in nicotine-mediated migration and its activation depends upon the existence or activity of PKC.

Discussion

Tobacco smoke has been implicated in the promotion of the onset of not only lung cancer but also malignancies in various other organs (1–3). Increasing incidences of human cancers caused by second-hand smoking has drawn a lot of attention for the needs to understand the etiology of the malignancies induced by such environmental smoking pollution. Nicotine is originally thought to be mainly responsible for tobacco addiction. However, many studies now reveal that this tobacco component is able to modulate various key biological activities in non-neuronal tissues (7–12). In particular, nicotine has been shown to promote survival of many cell types, including keratinocytes, and head or neck tumor cells (11, 40). Through associating with nicotine receptor, nicotine has been demonstrated to modulate the expressions of a diverse set of genes, the products of which are involved in the regulation of gene transcription, RNA processing and protein modification (21). In lung cancer, studies indicate that nicotine exposure facilitates the spread of the cancer in the body (10, 34). In the process of tumor promotion, nicotine has been shown to stimulate phosphorylation of μ- and m-calpains to enhance proteolytic activity and further to accelerate cell invasion (10, 34). Therefore, nicotine exposure appears to be able to promote not only cancer cell survival, but also tumor metastasis.

Here, our study demonstrates that human breast cancer MCF10A and MCF7 cells, like lung epithelial cells or keratinocytes, express at least 4 subunits of nAChR. Based on the similarity of the expression patterns of nAChR subunits in human breast cancer cells and cells from other types of tumors, we have examined the effect of nicotine in regulation of cell migration as well as on growth. We showed in this study that nicotine exposure is moderately mitogenic in serum-starved MCF10A and MCF7 cells. Furthermore, nicotine treatment, via stimulating PKC and cdc42 signaling cascade, promotes cell migration and invasion as measured by the Boyden chamber and wound-healing assays. This nicotine-mediated migratory activity requires the engagement of nicotine receptor. Therefore, from the experiments using human breast cancer cell lines, we conclude that one way for nicotine to promote breast tumor development may be through potentiating metastasis.

Cell surface growth factors and adhesion molecules are often involved in regulation of mitogenesis and metastasis that are critical steps during tumor development. It has been shown that the addition of nicotine mobilizes Ras and its downstream signaling pathways in mouse lung epithelial cells, resulting in the upregulation of cyclin D1 and subsequent entry of cell cycle following G1 stimulation (9). Furthermore, upon nicotine exposure, the lung cells were unable to arrest in G1 phase of the cell cycle, which renders the cells a genetic background for the establishment of genetic instability (29). Our current study using breast epithelial-like or cancer cell lines again demonstrates that nicotine is able to mobilize multiple intracellular signaling pathways to modulate growth- or survival-related activities. Through binding to the receptor, nicotine disrupts the G1 restriction point induced by serum starvation and, at the same time, stimulates cell migration or invasion. It is possible that nicotine, after binding to its receptor, recruits angiogenic or migration-related factors to the surface of cells and creates a high concentration of ligands near the receptor for the potentiation of cell migration. In such a cross talk between nicotine receptor and angiogenic receptors, nicotine promotes interactions of intracellular angiogenic factors and their receptors. Our study also showed that nicotine has a profound effect on cell proliferation after the cells were re-fed with the growth medium containing 10% serum (figure 2a). It is also conceivable that nicotine synergizes with growth factor receptor-induced pathways, which often occurs in the cooperation between VEGF and other growth factors (41–43).

PKC has been reported to regulate the expression or structural change of cytoskeleton-associated proteins, which further affects cytoskeleton reorganization and cell invasion (35, 44, 45). Upon mitogenic stimulation, PKC, acting as a second messenger, associates with the plasma membrane and then transmits signals to downstream effectors, including GTPases (35, 44, 45). In human endothelial cells, cdc42 has been shown to function downstream of PKC for re-organization of stress fibers after exposing to a stress environment (35). In this study, we showed that the suppression of PKC or cdc42 blocks nicotine-mediated migratory activity in breast cancer cells. The overexpression of cdc42 is able to further enhance the migration. In contrast, although cdc42 was overexpressed, there were a few migratory cells after PKC activation is inhibited. Given the fact that PKC activation occurs at the early stage and the plasma membrane level upon various stimulations and GTPases often function as intracellular effectors, our data suggest that PKC and cdc42 work in a linear signaling relationship to transduce the signals initiated by nicotine to accelerate cell migration.

Many growth factors and their downstream effectors are aberrantly activated or overexpressed, which contribute to growth deregulation in cancer. It has been reported that the addition of nicotine increases EGFR expression in colon cells (14). Therefore, it is possible that nicotine could elevate the sensitivity of breast epithelial or cancer cells to growth factors, and subsequently promote cell proliferation and metastasis. Since it is known that src is responsible for the phosphorylation of both EGFR and nAChR, it would be interesting to determine if src acts as a mediator to transmit the signals elicited by nicotine to EGFR pathway.

MCF10A is a spontaneously immortalized, but nontransformed human mammary epithelial cell line derived from the breast tissue of a patient with fibrocystic changes (46). MCF10A cells have been considered as normal breast epithelial cells, because these cells can not grow in soft agar, and are dependent on growth factors and hormones in the cultured medium for growth and survival (46). Our investigations demonstrate that the mammary epithelial MCF10A cells, like MCF7 breast carcinoma cells, express the same subunits of nAChR, and nicotine treatment does not alter the expression pattern of each subunit. The magnitudes of growth and migratory promotion by nicotine in these two cell lines are very similar. Moreover, both PKC and cdc42 have been shown to be involved in nicotine-induced cell migration. Thus, the data suggest that the same machinery in these two different cell lines is employed by nicotine to enhance the migratory or proliferation activity.

In summary, we have demonstrated that nicotine treatment can stimulate human breast epithelial-like MCF10A as well as cancer MCF7 cells to proliferate, migrate, and wound-heal. Metastasis plays a critical role in the process of tumorigenesis. The results also provided an insight into how nicotine potentiates cell migration in normal breast epithelial or cancer cells. Overall, our study provides an evidence to suggest that nicotine is a possible component for the initiation of breast cancer induced by second-hand smoking. The present investigation also warrants a caution for the clinic use of nicotine to relieve chronic pain or aid in the cessation of cigarette smoking.

Acknowledgments

This work was supported by Flight Attendant Medical Research Institute and by National Institutes of Health Grant RO1 CA124490 (to CY. C).

References

  • 1.Conti-Fine BM, Navaneetham D, Lei S, Maus AD. Neuronal nicotinic receptors in non-neuronal cells: new mediators of tobacco toxicity? Eur. J. Phamacol. 2000;393:279–274. doi: 10.1016/s0014-2999(00)00036-4. [DOI] [PubMed] [Google Scholar]
  • 2.Friedl P, Wolf K. Tumor-cell invasion and migration: diversity and escape mechanisms. Nature rev. Cancer. 2003;3:362–374. doi: 10.1038/nrc1075. [DOI] [PubMed] [Google Scholar]
  • 3.Schuller HM, Orloff M. Tobacco-specific carcinogenic nitrosamines. Logands for nicotinic acetylcholine receptors in human lung cancer cells. Bkochem. Pharmacol. 1998;55:1377–1384. doi: 10.1016/s0006-2952(97)00651-5. [DOI] [PubMed] [Google Scholar]
  • 4.Lindstrom J. Nicotinic acetylcholine receptors in health and disease. Mole. Neurobiol. 1997;15:193–222. doi: 10.1007/BF02740634. [DOI] [PubMed] [Google Scholar]
  • 5.Lukas RJ, Changeux JP, Le Novere N. Cruurent status of the nomenclature for nicotinic acetylcholine receptors and their subunits. Phamacol. Rev. 1999;51:397–401. [PubMed] [Google Scholar]
  • 6.Grando SA, Kawashima K, Wessler I. The non-neuronal cholinergic system in human. Life Sci. 2003;72:2009–2012. doi: 10.1016/s0024-3205(03)00063-8. [DOI] [PubMed] [Google Scholar]
  • 7.Heusch WL, Maneckjee R. Signalling pathways involved in nicotine regulation of apoptosis of human lung cancer cells. Carcinogenesis. 1998;19:551–556. doi: 10.1093/carcin/19.4.551. [DOI] [PubMed] [Google Scholar]
  • 8.Minna JD. Nicotine exposure and bronchial epithelial cell nicotinic acetylcholine receptor expression in the pathogenesis of lung cancer. J Clin Invest. 2003;111:31–33. doi: 10.1172/JCI17492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Guo J, Chu M, Abbeyquaye T, Chen C-Y. Persisten nicotine Treatment potentiates DHFR amplification in rat lung epithelial cells as a consequence of Ras activation. J. Biol. Chem. 2005;280:30422–30431. doi: 10.1074/jbc.M504688200. [DOI] [PubMed] [Google Scholar]
  • 10.Xu L, Deng X. Protein Kinase Cτ promotes nicotine-induced migration and invasion of cancer cells via phosphorylation of μ- and m-calpains. J. Biol. Chem. 2006;281:4457–4466. doi: 10.1074/jbc.M510721200. [DOI] [PubMed] [Google Scholar]
  • 11.Arredondo J, Chernyavsky AI, Jolkovsky DL, Pinkerton KE, Grando SA. Receptor-mediated tobacco toxicity: cooperation of the Ras/Raf-1/MEK1/ERK and JAK-2/STAT-3 pathways downstream of α7 nicotinic receptor in oral keratinocytes. The FASEB J. 2006;20:2093–2101. doi: 10.1096/fj.06-6191com. [DOI] [PubMed] [Google Scholar]
  • 12.Zhu BQ, Heeschen C, Slievers RE, Karliner JS, Parmley WW, Glantz SA, Cooke JP. Second hand smoke stimulates tumor angiogenesis and growth. Cancer Res. 2003;4:191–196. doi: 10.1016/s1535-6108(03)00219-8. [DOI] [PubMed] [Google Scholar]
  • 13.Li S, Zhao T, Xin H, Ye L-H, Zhang X, Tanaka H, Nakamura A, Kohama K. Nictotinic acetylcholine receptor α7 subunit mediates migration of vascular smooth muscle cells toward nicotine. J. Pharmacol. Sci. 2004;94:334–338. doi: 10.1254/jphs.94.334. [DOI] [PubMed] [Google Scholar]
  • 14.Ye YN, Liu ESL, Shin VY, Wu WKK, Luo JC, Cho CH. Nicotine promoted colon cancer growth via epidermal growth factor receptor, c-src, and 5-lipoxygenase-mediated signal pathway. J. of Pharmacol. And Exp. Therapeutics. 2004;308:66–72. doi: 10.1124/jpet.103.058321. [DOI] [PubMed] [Google Scholar]
  • 15.Nguyen VT, Hall LL, Gallacher G, Ndoye A, Jolkovsky DL, Webber RJ, Buchli R, Grando SA. Choline acetyltransferase, acetylcholinesterase, and nicotinic acetylcholine receptors of human gingival and esophageal epithelia. J. Dent. Res. 2000;79:939–949. doi: 10.1177/00220345000790040901. [DOI] [PubMed] [Google Scholar]
  • 16.Nguyen VT, Ndoye A, Grando SA. Novel human α9 acetylcholine receptor regulating keratinocyte adhesion is targeted by pemphigus vulgaris autoimmunity. Am. J. Pathol. 2000;157:1377–1391. doi: 10.1016/s0002-9440(10)64651-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Arredondo J, Nguyen VT, Chernyavsky AI, Jolkovsky DI, Pinkerton KE, Grando SA. A receptor-mediated mechanism of nicotine toxicity in oral keratinocytes. Lab. Invest. 2001;81:1653–1668. doi: 10.1038/labinvest.3780379. [DOI] [PubMed] [Google Scholar]
  • 18.Arredondo J, Chernyavsky AI, Marubio LM, Beauder AI, Jolkovsky DI, Pinkerton KF, Grando SA. Receptor-mediated tobacco toxicity: Regulation of gene expression through α3β2 nicotinic receptor in oral epithelial cells. Am. J. Pathol. 2005;166:597–613. doi: 10.1016/s0002-9440(10)62281-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Seguela P, Wadiche J, Dineley-Miller K, Dani JA, Patrick JW. Molecular cloning, functional properties, and distribution of rat brain α7: a nicotinic cation channel highly permeable to calcium. J. Neurosci. 1993;13:596–604. doi: 10.1523/JNEUROSCI.13-02-00596.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Castro NG, Albuquerque EX. α-Bungarotoxin-sensitive hippocampal nicotinic receptor channel has a high calcium permeability. Biophys. J. 1995;68:516–524. doi: 10.1016/S0006-3495(95)80213-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Dunckley T, Lukas RJ. Nicotine modulates the expression of a diverse set of genes in the neuronal SH-SY5Y cell line. J. Biol. Chem. 2003;278:15633–15640. doi: 10.1074/jbc.M210389200. [DOI] [PubMed] [Google Scholar]
  • 22.Li JM, Cui TX, Shiuchi T, Liu HW, Min LJ, Okumura M, Jinno T, Wu L, Iwai M, Horiuchi M. Nicotine enhances angiotensin II-induced mitogenic response in vascular smooth muscle cells and fibroblasts. Arterioscler. Thomb. Vase. Biol. 2004;24:80–84. doi: 10.1161/01.ATV.0000104007.17365.1c. [DOI] [PubMed] [Google Scholar]
  • 23.Dumont JA, Bitonti AJ. Biochem. Biophys. Modulation of human melanoma cell metastasis and adhesion may involve integrin phosphorylation mediated through protein kinase C. Res. Commun. 1994;204:264–272. doi: 10.1006/bbrc.1994.2454. [DOI] [PubMed] [Google Scholar]
  • 24.Newton AC. Protein kinase C: structure, function, and regulation. J. Biol. Chem. 1995;270:28495–28498. doi: 10.1074/jbc.270.48.28495. [DOI] [PubMed] [Google Scholar]
  • 25.Mellor H, Parker PJ. The extended protein kinase C superfamily. Biochem. J. 1998;332:281–292. doi: 10.1042/bj3320281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Gomez DE, Skilton G, Alonso DF, Kazanietz M. The role of protein kinase C and novel phorbol ester receptors in tumor cell invasion and metastasis. Oncol. Rep. 1999;6:1363–1370. doi: 10.3892/or.6.6.1363. [DOI] [PubMed] [Google Scholar]
  • 27.Heusch WL, Maneckjee R. Signalling pathways involved in nicotine regulation of apoptosis of human lung cancer cells. Carcinogenesis. 1998;19:551–556. doi: 10.1093/carcin/19.4.551. [DOI] [PubMed] [Google Scholar]
  • 28.Mai H, May WS, Gao F, Jin Z, Deng X. A functional role for nicotine in Bcl-2 phosphorylation and suppression of apoptosis. J. Biol. Chem. 2003;278:1886–1891. doi: 10.1074/jbc.M209044200. [DOI] [PubMed] [Google Scholar]
  • 29.Chu M, Guo J, Chen C-Y. Long-term exposure to nicotine, via Ras pathway, induces cyclin D1 to stimulate G1 cell cycle transition. J. Biol. Chem. 2005;280:6369–6379. doi: 10.1074/jbc.M408947200. [DOI] [PubMed] [Google Scholar]
  • 30.West KA, Brognard J, Clark AS, et al. Rapid Akt activation by nicotine and a tobacco carcinogen modulates the phenotype of normal human airway epithelial cells. J. Clin. Invest. 2003;111:81–90. doi: 10.1172/JCI16147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Horwitz AR, Parsons JT. Cell migration-moving’ on. Science. 1999;286:1102–1103. doi: 10.1126/science.286.5442.1102. [DOI] [PubMed] [Google Scholar]
  • 32.Mamoune A, Luo JH, Lauffenburger DA, Wells A. Calpain-2 as a target for limiting prostate cancer invasion. Cancer Res. 2003;63:4632–4640. [PubMed] [Google Scholar]
  • 33.Sahal E. Mechanisms of cancer cell invasion. Curr. Opin. Genet. Dev. 2005;15:87–96. doi: 10.1016/j.gde.2004.12.002. [DOI] [PubMed] [Google Scholar]
  • 34.Xu L, Deng X. Suppression of cancer cell migration and invasion by protein phosphatase 2A through dephosphorylation of μ and m-calpains. J. Biol. Chem. 2006;281:35567–35575. doi: 10.1074/jbc.M607702200. [DOI] [PubMed] [Google Scholar]
  • 35.Tatin F, Varon C, Genot E, Moreau V. A signaling cascade involving PKC, Src and cdc42 regulates podosome assembly in cultured endothelial cells in response to phorbol ester. J. of Cell Sci. 2005;119:769–781. doi: 10.1242/jcs.02787. [DOI] [PubMed] [Google Scholar]
  • 36.Deeds L, Teodorescu S, Chu M, et al. A p-53-independent G1 cell cycle checkpoint induced by the suppression of protein kinase C α and θ isoforms. J. Biol. Chem. 2003;278:39782–39793. doi: 10.1074/jbc.M306854200. [DOI] [PubMed] [Google Scholar]
  • 37.Zhang L, Kharbanda S, McLeskey SW, et al. Overexpression of fibroblast growth factor 1 in MCF7 breast cancer cells facilitates tumor cell dissemination but does not support the development of macrometastases in the lungs or lymph nodes. Cancer Res. 1999;59:5023–5029. [PubMed] [Google Scholar]
  • 38.Kurebayashi J, McLeskey SW, Johnson MD, et al. Quantitative demonstration of spontaneous metastasis by MCF7 human breast cancer cells cotransfected with fibroblast growth factor 4 and lacZ. Cancer Res. 1993;53:2178–2187. [PubMed] [Google Scholar]
  • 39.Zhang L, Kharbanda S, Chen D, et al. MCF7 breast carcinoma cells overexpressing FGF-1 form vascularized, metastatic tumors in ovariectomized or tamosifen-treated nude mice. Oncogene. 1997;15:2093–2108. doi: 10.1038/sj.onc.1201386. [DOI] [PubMed] [Google Scholar]
  • 40.Onoda N, Nelmi A, Weiner D. Nicotine affects the signaling of the death pathway, reducing the response of head and neck cancer cell line to DNA damaging agents. Head Neck. 2001;23:860–870. doi: 10.1002/hed.1125. [DOI] [PubMed] [Google Scholar]
  • 41.Wang SL, Feng J, Correa A. Effects of in vivo treatments of nicotine and benxo[a]pyrene on the epidermal growth factor receptor in hamster buccal pouch. Toxicology. 1996;107:31–38. doi: 10.1016/0300-483x(95)03198-o. [DOI] [PubMed] [Google Scholar]
  • 42.Mathur RS, Mathur SP, Young RC. Up-regulation of epidermal growth factor-receptors (EGF-R) by nicotine in cervical cancer cell lines: this effect may be mediated by EGF. Am. J. Reprod. Immunol. 2000;44:114–120. doi: 10.1111/j.8755-8920.2000.440207.x. [DOI] [PubMed] [Google Scholar]
  • 43.Nakayama H, Numakawa T, Ikeuchi T. Nicotine-induced phosphorylation of Akt through epidermal growth factor receptor and src in PC12 cells. J. Neurochem. 2002;83:1372–1379. doi: 10.1046/j.1471-4159.2002.01248.x. [DOI] [PubMed] [Google Scholar]
  • 44.Hai C, Hahne P, Harrington E, Gimona M. Conventional protein kinase C mediates phorbol-dibutyrate-induced cytoskeletal remodeling in a7r5 smooth muscle cells. Exp. Cell Res. 2002;280:64–74. doi: 10.1006/excr.2002.5592. [DOI] [PubMed] [Google Scholar]
  • 45.Moreau V, Tatin F, Varon C, Genot F. Actin can reorganize into podosomes in aortic endothelial cells, a procell controled by cdc42 and RhoA. Mole. Cell. Biol. 2003;23:6809–6822. doi: 10.1128/MCB.23.19.6809-6822.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kiosses W, Hahn K, Giannelli G, Quaranta V. Characterization of morphological and cytoskeletal changes in MCF10A breast epithelial cells plated on laminin-5: comparison with breast cancer cell line MCF7. Cell Coomun. Adhes. 2001;8:29–44. doi: 10.3109/15419060109080705. [DOI] [PubMed] [Google Scholar]

RESOURCES