- Email: [email protected]
Real-time electrical impedance detection of cellular activities of oral cancer cells
Biosensors and Bioelectronics 25 (2010) 2225–2231
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Real-time electrical impedance detection of cellular activities of oral cancer cells L. Renea Arias, Carla A. Perry, Liju Yang ∗ Biomanufacturing Research Institute and Technology Enterprises (BRITE), Department of Pharmaceutical Sciences, North Carolina Central University, Durham, NC 27707, USA
a r t i c l e
i n f o
Article history: Received 24 November 2009 Received in revised form 23 February 2010 Accepted 24 February 2010 Available online 3 March 2010 Keywords: Real-time Electrical impedance Cellular activity Cancer cell Apoptosis
a b s t r a c t In this study, the electric cell-substrate impedance sensing (ECIS) system was used to study the cellular activities of oral squamous cell carcinoma (OSCC) cells in a real-time and label-free manner. Various cellular activities, including cell adhesion, spreading, proliferation, and drug-induced apoptosis and inhibition of apoptosis, were monitored. A linear relationship was found between the impedance-based cell index and the cell number in the range of 3500 to 35,000 cells/well. Anti-cancer drug-cisplatin-induced OSCC cell apoptosis at the minimal concentration of 5 M after 20 h of treatment and followed a linear dose-dependent manner in the concentration range from 10 M to 25 M. The inhibition of cisplatininduced apoptosis by the carcinogen, nicotine, at concentrations from 0.1 M to 10 M was monitored. The most significant inhibitory effect of nicotine on cisplatin-induced apoptosis was observed at concentrations of 0.5–1 M. The results obtained with impedance method correlated well with microscopic imaging analysis of cellular morphology and cell viability analysis. This study demonstrated that the impedance-based method can provide real-time information about the cellular activity of viable cells and detect drug-induced cellular activities much earlier than commonly used cell-based image analysis. This impedance-based method has the potential to provide a useful analytical approach for cancer research. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Analysis of cellular properties and physiological functions of a certain type of cells in human body has been an effective approach to understand many biological and biomedical problems. For example, detection and analysis of cancer cells presented in body fluids can be a indicator of prognosis and diagnosis in oncology (Pantel et al., 1999). Viable tumor-derived epithelial cells or circulating tumor cells (CTCs) have been identified in peripheral blood from cancer patients, and have been found to correlate with clinical stage (Benez et al., 1999; Weihrauch et al., 2002), patient survival after therapy (Braun et al., 2000), and tumor metastasis (Beitsch and Clifford, 2000). In addition, cell-based methods are important for developing rapid and simple techniques to study cancerous cells and their interactions with drugs, especially in anti-cancer drug discovery and screening. Current typical approaches for cellular analysis include flow cytometry and microscopic imaging. While these methods can provide insights into the physiological function of any particular cell or of pathological changes that may have occurred; they usually require fluorescent, chemiluminescent, or radioactive labeling steps which often involve destruction of cells. The labeling processes may thus lead to the loss of very important biological information about viable cells.
∗ Corresponding author. Tel.: +1 919 530 6704; fax: +1 919 530 6600. E-mail address: [email protected] (L. Yang). 0956-5663/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2010.02.029
Label-free and non-invasive detection of cells and their functions is advantageous to provide real-time and/or kinetic cellular activities of viable cells, which could be of great use in many biomedical applications. A number of label-free technologies, including cell-substrate impedance (Giaever and Keese, 1986, 1993), quartz microbalance (Zhou et al., 2000; Marx et al., 2005), and optical waveguide lightmode spectroscopy (Ramsden et al., 1995; Fang, 2007), have been reported as a means for monitoring live cell status in a non-invasive and real-time manner. Amongst these, electrical/electrochemical impedance spectroscopy (EIS) has been recognized as a powerful electrochemical technique that can monitor live cell behavior in real-time. The distinct electrical properties associated with specific biological cells satisfy the impedance techniques for cellular analysis. By culturing biological cells on an electrode surface, EIS can directly sense detailed information about cellular activities occurring on an electrode or substrate’s surface by measuring the induced capacitance and/or resistance changes, eliminating multiple labeling and amplification steps typically used in many other cell-based methods, and allowing label-free and non-invasive study of cellular properties. In addition, impedance technique is amenable to miniaturized electronic systems to meet the growing needs of microdevices for point-of-care analysis. Impedance measurements using microelectrodes were first used to study the characteristics of anchorage dependent cultured cell lines by Giaever and Keese (1984). Since then, they have continuously improved and refined their method and named it electric cell-substrate impedance sensing (ECIS) (Giaever and
2226
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Keese, 1991, 1992, 1993). Other researchers have also been using similar impedance sensing to enumerate cells attached to a substrate (Ehret et al., 1997, 1998). Recently, the impedance-based sensing technology has gained a great deal of attention for studying cancer cells and monitoring drug-induced cellular activities for drug discovery (Solly et al., 2004; Linderholm et al., 2007; McGuinness, 2007; Klo et al., 2008; Chen et al., 2008; Liu et al., 2009). In this study, the ECIS system was used for label-free detection of oral cancer cells and studying the cellular activities induced by drugs. Oral cancer is one of the most common cancers worldwide, and oral squamous cell carcinoma (OSCC) occurs with the highest frequency. In the US, OSCC represents 2–4% of the annually diagnosed malignancies, accounting for 8000 deaths every year (CDC, 1998; USDHHS, 2000). Research has indicated that long-term tobacco usage is one of the important risk factors that predominately cause OSCC (Rodriguez et al., 2004), but limited information about the effect of nicotine on OSCC cell cellular activities was available. It has been suggested that nicotine, the major component of tobacco, plays several roles to alter cellular functions, including activation of mitogenic pathways, angiogenesis, and cell growth, in many cell types. It has also been reported that nicotine can inhibit the drug- and radiation-induced apoptosis of different cell lines (Wright et al., 1993; Xu et al., 2007). In this study, the impedancebased system was used not only to monitor the cell adhesion, spreading, and proliferation of the human oral cancer cell line, CAL27, but also to study the cellular activities of CAL27 cells induced by a well-known anti-cancer drug, cisplatin, and the inhibition of cisplatin-induced apoptosis by nicotine. 2. Materials and methods
based cell index was measured in the designed period of time with time intervals varying from 5 min to 30 min. 2.3. Drug treatments The well established anti-cancer drug, cisplatin, was obtained from Sigma–Aldrich Co. (St. Louis, MO). In the drug treatment tests, OSCC cells (5000 cells/well) were seeded in 16-well plates overnight (approximately 20 h), then aliquots of 15 l cisplatin solution were added to the wells at final concentrations of 5, 10, 15, 20, 25 M. Impedance-based cell indexes were recorded continuously over the entire experimental period. Nicotine (Sigma–Aldrich, St. Louis, MO), a carcinogen, was used to test its effect on inhibition of cisplatin-induced apoptosis of OSCC cells. Nicotine with final concentration of 0.1 M, 0.5 M, 1.0 M, 5.0 M and 10.0 M was used to treat the OSCC cells in the presence and absence of cisplatin. 2.4. Imaging The microscopic images of OSCC cells in the wells were taken on a Nikon ECLIPSE E600FN microscope (Japan) with a Coolsnap HQ camera (Roper Scientific, Inc., Photometric, Tucson, AZ) using reflective bright field mode. 2.5. Statistical analysis All experiments were repeated in triplicate. Data in the figures were expressed as either mean of three repeated results or mean ± standard deviation (SD). Statistical significance of the data was determined using the Student’s t-test. Differences were considered significant at the level of P < 0.05.
2.1. Cell culture 3. Results and discussion The OSCC cell line, CAL27, derived from the tongue (ATCC # CRL-2095), was cultured in Dulbecco’s modified eagle’s medium (DMEM) (Sigma) supplemented with 10% of fetal bovine serum and antibiotics (100 IU/ml penicillin and 100 g/ml streptomycin) (Lonza, Walkersville, MD). Cells were cultured in 75 cm2 flask and were incubated at 37 ◦ C in an atmosphere of 5% CO2 in air. The medium was renewed every 2–3 days. When confluent, cells were detached from the flask using 0.25% trypsin with 0.53 mM EDTA solution. Cells were centrifuged to remove trypsin. Cell pellet was resuspended with fresh DMEM medium. Cell number in the suspension was determined using the Vi-cell XR cell counting system (Beckman Coulter, Miami, FL). Desired cell concentrations were obtained by diluting the cell suspension with fresh DMEM for further experiments. Cell viability analysis was also performed using the Vi-cell system, which utilizes the trypan blue dye exclusion method. The dead and non-viable cells are permeable and can uptake the trypan blue dye, thus become darker than the viable cells. The Vi-cell system analyzes up to 50 images for the determination of cellular viability which is defined as the percentage of viable cells in the total cell number. 2.2. Real-time impedance measurement of cellular activities The real-time impedance-based measurement of OSCC cellular activities was performed using the RT-CES system (ACEA Biosciences, Inc., San Diego, CA). Detailed information about the components and the principle of the system is in the Supplementary Materials S1 and S2. For OSCC cell measurements, 50 l of DMEM was added to the wells of the 16-well E-plate for taking background readings, then 100 l of OSCC cells in DMEM with the desired concentration was added to each well. Impedance-
3.1. Real-time monitoring of the attachment, spreading and proliferation of OSCC cells In cancer biology, cell adhesion to the extracellular matrix is important to tumor metastasis, thus cell morphology is another most important characteristic in study of cancer cells. By culturing OSCC cells on the microelectrode surface, impedance signals from the microelectrodes can reflect both the cell adhesion to the electrodes and the cell morphology. Fig. 1A and B shows two representative cellular activity curves of OSCC cells at (A) high (3.5 × 104 cells/well) and (B) low (3.5 × 103 cells/well) cell concentrations cultured on the microelectrodes for 20 h. The cell index curves show the average values of three measurements with the error bar showing the standard deviation of the three measurements at each data point. For high cell concentration, the cell index curve (Fig. 1A) presents three distinct regions with respect to the rate of cell index change, which most likely reflects the initial cell adhesion, followed by cell spreading, and cell proliferation stages. In the first 1–2 h, the cell index increased slowly with an increase rate of 0.085/h. This period of time corresponded to the initial cell adhesion stage, in which the cells started to settle down toward the electrode surface and became adherent to the electrodes. In the next 2–4 h, the cell index increased rapidly at a rate of 0.42/h. Most likely, in this period, the adhered cells spread over the electrode surface. After spreading, cells started to proliferate, and the rate of increase of the cell index again slowed to 0.084/h. At low cell concentration (3.5 × 103 cells/well), only two distinct regions on the cell index curve corresponding to the cell spreading and cell proliferation stages are visible-no distinct region can be observed between the cell adhesion and cell spreading stages. A possible explanation could be that because of the low cell number, the time required for the cells
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
2227
Fig. 1. (A) and (B) Two representative cell index curves of OSCC cells at high (3.5 × 104 cells/well) and low (3.5 × 103 cells/well) cell concentrations cultured on the microelectrodes for 20 h. (C)–(F) The microscopic images of the OSCC cells (3.5 × 104 cells/well) after they were seeded in the well for 30 min, 2 h, 5 h, and 22 h.
to settle and adhere to the electrodes was minimal, and the cells quickly start to spread over the electrode surface. The cell proliferation stage started approximately 4 h after the cells were seeded in the well. Fig. 1C–F shows the microscopic images of the OSCC cells (3.5 × 104 cells/well) after they were seeded in the well for 30 min, 2 h, 5 h and 22 h. It can be seen that at 30 min, the cells started to settle down onto the electrode surface, and the round shape of the
cells indicated that they were not spread yet. Over the first 2 h, cells continued to settle down onto the electrode surface from the suspension. The cells are still round, indicating that significant cell spreading had not yet occurred. At 5 h, cells started to spread over the electrode surface. After spreading, the cells started to proliferate. At 22 h, the cells fully covered on the electrode surface. These images provide strong evidence to interpret the pattern of the cell index curve in Fig. 1A. Correlating to the cell index change in Fig. 1A,
Fig. 2. (A) A group of cell index curves with growth time for the samples with different cell numbers ranging from 1.0 × 103 to 3.5 × 104 cells/well. (B) The linear relationship between the cell number and the cell index measured at 15 h after the cell were seeded in the wells. (C)–(F) The microscopic images of cell coverage on the microelectrodes after the cells fully spread when 1000, 5000, 10,000, and 20,000 cells were seeded in the wells. Cell index curves in (A) shows averaged values from three measurements, and error bars in (A) and (B) are the standard deviations of three measurements.
2228
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Fig. 3. (A) A group of cell index curves reflecting the apoptosis of OSCC cells induced by different concentrations of cisplatin (5 M, 10 M, 15 M, 20 M and 25 M). Initial number of OSCC cells: 5000 cells/well. (B) The dose-dependent cell index measured at 3 h, 5 h, 10 h, 15 h, and 20 h after cisplatin treatment, derived from three repeated tests as in (A). The error bar on each data point is the standard deviation of the three measurements,.
we can see that the cell index increased slowly due to the increasing number of cells settled down on the electrode surfaces in the first 1–2 h. While the cell spreading caused the most significant and rapid increase in the cell index, the cell proliferation generated slower increases in the cell index. The rate of cell index increase due to cell spreading is about 5 times the rate measured during initial cell adhesion or cell proliferation. These results suggest that the impedance-based cell index can be used as reliable measure to study the kinetics of cell spreading and proliferation. 3.2. Detection of OSCC cells Fig. 2A shows the cell index curves with growth time for the samples with different cell numbers ranging from 1.0 × 103 to 3.5 × 104 cells/well. For each sample, the cell index increases with the time which corresponds the cellular activities including cell attachment, spreading, and proliferation over time. Between the samples with different cell numbers, at each time point, the higher cell number has higher cell index. As the cell size, morphology, the extent of cell spreading, and the rate of proliferation are relatively constant in the same type of cells, the impedance-based cell index should be indicative of the cell number. Derived from Fig. 2A, Fig. 2B shows the relationship between the cell number and the cell index measured at 15 h during the cell proliferation stage. It shows that the cell index is proportional to the cell number in the range of cell number between 3500 and 35,000 cells/well, with the linear regression equation of I (cell index) = 8 × 10−5 N (cells/well) − 0.1124. The sample with 1000 cells/well did not generate a cell index that is significantly different from the blank, indicating that such low number of cells is unable to generated detectable cell index. As the cell index is related to the cell coverage on electrode surface, we examined the cell coverage of samples with different cell numbers. Fig. 2C–F shows the representative images of cell coverage on the electrode generated by different number of cells after these cells fully spread on the electrode surfaces. Estimated from several images taken at different locations in each well, the average cell coverage generated by 1000, 5000, 10,000, and 20,000 cells/well are approximately 2.4%, 12.0%, 38.0%, 72.0%, respectively. Linked to the cell index measurements, the results indicated that a ∼10% cell coverage on the electrode surface is required to generate a detectable impedance-based cell index. The results here demonstrated that this impedance-based method could offer a label-free and non-invasive quantitative method to detect OSCC cells. This can be useful in clinic diagnostics, such as detection of the presence of cancer cells in body fluids. Detection of CTCs can be a potential indicator of prognosis and diagnosis in oncology (Pantel et al., 1999), since the presence of cancer cells in the bloodstream or bone marrow is correlated with clinical stage (Benez et al., 1999; Weihrauch et al., 2002), patient survival
after therapy (Braun et al., 2000), and tumor metastasis (Beitsch and Clifford, 2000). In the mouth, the surface layer of cells is replaced about every 2–4 h; and the turnover time of the oral epithelium is about 4.5 days. If a person is developing an oral cancer, cancer cells can be shed into saliva at very early stage of the cancer; and the number of cancer cells in saliva can be a measure of the cancer stage. Mauk et al. (2007) reported that more than 1000 cells/ml and 9000 cells/ml of OSCC cells were separated from tumor stage 1 and 4 patient saliva, respectively. This makes saliva an ideal sample for early screening and detection of oral cancers. The impedance-based method has the potential to be a sensitive non-invasive screening method for detecting early stage cancer by detecting cancer cells in saliva, and an approach to obtain quantitative information about cancer stage or to monitor the progress of cancer treatment. 3.3. Real-time monitoring of apoptosis of OSCC induced by cisplatin Fig. 3A shows a group of representative cell index curves reflecting the effects of different concentrations of cisplatin on CAL27 cells cultured on the microelectrodes at an initial cell number of 5000 cells/well. The increased cell index in the control sample indicated cell growth with time, whereas the decreased cell index in other samples treated with increasing concentrations of cisplatin (5 M, 10 M, 15 M, 20 M and 25 M) reflected the increasing extents of cell death resulting from cisplatin-induced apoptosis. Each cell index curve also provided detailed kinetic information of cell responses to the particular cisplatin concentration. For instance, compared to the control, the cell index of OSCC cells treated with 5 M cisplatin starts to decrease approximately 20 h after treatment, whereas the one treated with 10 M cisplatin starts to decrease at approximately 10 h after treatment. With treatments of 15 M, 20 M and 25 M cisplatin, the cell indexes start to decline within 1–3 h after treatments. The results indicated that low concentrations of cisplatin required longer time to induce OSCC cell apoptosis, while high concentrations of cisplatin-induced OSCC cell apoptosis quickly. And this impedance-based method can monitor cisplatin-induced cell apoptosis in a real-time and label-free manner, and provide dynamic information about cellular response to drug treatment over the entire experimental period. Derived from three repeated tests as in Fig. 3A, Fig. 3B shows the dose-dependent cell index curves of OSCC cells in response to different concentrations of cisplatin at 3 h, 5 h, 10 h, 15 h and 20 h after treatment. The reproducibility of the measurement is indicated by the error bars which are the standard deviation of the three measurements. At each measurement time, the change in cell index induced by cisplatin treatment increased with the increasing concentration of cisplatin. For all these measurement times, the linear dose response range can be observed in the cisplatin concentration
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
2229
Fig. 4. The images of OSCC cells showing the cell morphology at 5 h and 22 h after treatment with 20 M of cisplatin, along with the images of the untreated control cells. Initial cell number: 5000 cells/well and 35,000 cells/well.
range from 10 M to 25 M. Obviously, the cell indexes monitored at longer treatment times (15 h, 20 h, or more) show clearer apoptosis effects. The time required to induce significant apoptosis of OSCC cells is also related the concentration of cisplatin. The apoptosis induced by 10 M of cisplatin can be detected at 20 h after treatment, while the apoptosis induced by 20 M of cisplatin can be detected as early as 5 h after treatment. Fig. 4 shows the images of OSCC cells (5000 cells/well and 35,000 cells/well) at 5 h and 22 h after treatment with 20 M of cisplatin, along with the images of the untreated control cells. The images of OSCC cells at the two cell number levels show similar patterns: at 5 h after treatment, no significant morphology change can be seen, and at 22 h after treatment, cells showed significant morphology change which indicated cell apoptosis. Linking to the cell index change in Fig. 3A, the declining of cell index was clearly observed as early as 5 h after treatment, indicating that the impedance-based cell index is more sensitive to the changes in cellular activities induced by cisplatin treatment at earlier time. The effects of drug treatment can be obtained before any morphology change is observed in imaging analysis. At 22 h after cisplatin treatment, the cells are not tightly attached to the electrode surface, the apoptotic cells became round in shape. Correspondingly, the cell index at 22 h after treatment decreased significantly (Fig. 3A), indicting the impedance results are correlated with the significant morphology changes observed in imaging analysis. The morphology changes in the two different levels of cell numbers confirmed that cell density in the wells (confluent or not) did not affect the cisplatin-induced apoptosis on OSCC cells. Cell viability after cisplatin treatment was analyzed by the Vi-cell system. Upon the treatment of 20 M cisplatin, the cell viability decreased from 97.0 ± 2.1% (control) to 92.3 ± 2.8% at 5 h after treatment, and to 82.8 ± 4.0% at 22 h after treatment (Fig. S3(A) in Supplementary Materials). Statistical analysis showed that the cell viability at 22 h after treatment was significantly different from the control, while that at 5 h after treatment was not significantly different. Upon the treatment of cispatin at 10 M, 20 M and 30 M, the cell viabilities measured at 22 h after treat-
ment were 92.7 ± 0.9%, 82.8 ± 4.0% and 59.5 ± 1.3%, respectively (Fig. S3(B) in Supplementary Materials). Statistical analysis showed significant decrease in cell viability at 22 h upon the treatment of 20 M and 30 M cisplatin. The decreases in cell viability with treatment time and with the increasing cisplatin concentration were correlated with the trend of impedance-based cell index decrease observed in Fig. 3A. 3.4. Real-time monitoring of nicotine inhibiting the apoptosis induced by cisplatin Nicotine, as a major component in tobacco, has long been recognized as a carcinogen. It has been implicated as a risk factor for a number of tobacco-related diseases, such as coronary artery disease, cardiovascular disease, atherosclerosis formation (Cucina et al., 2007; Ambrose and Barua, 2004), and several types of cancers (Catassi et al., 2008). Nicotine has been shown to affect various cellular systems, including lung cancer (Zhang et al., 2006; Catassi et al., 2008), pancreas cancer (Galitovsky et al., 2004), and oral cancer (Xu et al., 2007). Several studies have focused on the correlation between nicotine and apoptosis in a variety of cell types, but the results are contradictory. Some studies show that nicotine seems to act as an inhibitor of apoptosis, while others show that it is an inductor of apoptosis. For example, Zhang et al. (2006) reported that nicotine prevented the apoptosis induced by menadione in human lung cancer cells. Galitovsky et al. (2004) reported nicotine-induced apoptosis in pancreatic cancer cells, and the most significant apoptotic effect was observed at nicotine concentration of 100 M to 1 mM. In smooth muscle cells (SMC), Cucina et al. (2007) found that nicotine not only inhibited the SMC apoptosis but also increased the SMC proliferation. Carlisle et al. (2007) reported that nicotine may play a role in regulating survival of non-small cell lung cancer (NSCLC) cells, by activating cell-signaling pathways through muscle-type and neuronal nicotintic acetylcholine receptors in NSCLC cells. Catassi et al. (2008) reviewed the multiple roles of nicotine on cell proliferation and inhibition of apoptosis in human lung cancers. Different research groups have studied the
2230
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Fig. 5. (A) The representative cell index curves of the control, the cells treated with 10 M nicotine only, 20 M cisplatin only, and the cells treated with 0.1 M, 1.0 M and 5.0 M in the presence of 20 M of cisplatin. (B) The cell indexes measured at 18 h after they were treated with 0.1 M, 0.5 M, 1.0 M, 5.0 M and 10 M of nicotine in the presence of 20 M cisplatin, along the cell indexes of the cells treated with 20 M cisplatin only and the control. The error bars are the standard deviations of three repeated test results. Data labeled with an asterisk was significantly different from the result of cisplatin-treated sample, P < 0.05.
effect of nicotine on head and neck cancer cells (Xu et al., 2007; Arredondo et al., 2006; Onoda et al., 2001). For example, Xu et al. (2007) reported that nicotine inhibited cisplatin-induced apoptosis in human oral cancer cell line Tca8113. Others reported that nicotine affected the signaling of the death pathway and resulted in a decreased cytotoxicity of various anticancer agents such as cisplatin and gamma-radiation in head and neck cancer cell lines UMSCC 10b and UMSCC5 (Onoda et al., 2001). In this study, we used the real-time impedance system to monitor the cellular responses of OSCC cell line cells, CAL 27, when they were treated with nicotine in the presence and absence of cisplatin. Fig. 5A shows a group of representative cell index curves reflecting the effect of different concentrations of nicotine on the OSCC cells in the presence and absence of cisplatin. It shows that the pattern of the cell index curve of OSCC cells with 10 M nicotine treatment is similar to that of the control, indicating that nicotine only did not affect the OSCC cells’ growth. The cells treated with 20 M cisplatin shows reduced cell index compared with the control, indicating the cisplatin-induced apoptosis. The cells co-treated with 1 M nicotine and 20 M cisplatin presents higher cell indexes than that with 20 M cisplatin treatment, indicating that 1 M nicotine has an inhibitory effect on the cisplatin-induced cell apoptosis. We also tested the effect of different concentrations of nicotine in the presence of 20 M cisplatin, and found that 0.1 M nicotine did not cause significant inhibition on the cisplatin-induced apoptosis. The cell index curve shows that treatment with 0.5 M nicotine exhibited significant inhibition on the cisplatin-induced apoptosis. However, when the nicotine concentration increased to 5 M, the cell indexes are lower than those with 1 M and 0.5 M nicotine treatments, but still higher than those of cisplatin treatment only, which suggest that the 5 M nicotine inhibitory effect on the cisplatin-induced apoptosis is not as strong as that by 1 M and 0.5 M nicotine. Derived from three repeated tests as in Fig. 5A, Fig. 5B shows the cell indexes read at 40 h of OSCC cell growth and approximately 18 h after treatments, when the cells were treated with 20 M cisplatin only, and 20 M cisplatin with different concentrations of nicotine (0.1 M to 10 M). The results clearly show that low concentration of nicotine did not exhibit inhibitory effect on 20 M cisplatin-induced apoptosis; Nicotine at concentrations of 0.5 M to 1 M exhibited the most significant inhibitory effect on 20 M cisplatin-induced apoptosis; and Nicotine inhibitory function decreased when its concentration increased from 1 M to 10 M. With 0.5 M and 1 M nicotine treatments, the cell indexes increased by 68.2% and 73.9%, respectively, compared with the cell index of cisplatin treatment, whereas 5 M and 10 M nico-
tine treatment increased the cell indexes by 52.1% and 17.3%, respectively. Considering the multiple roles of nicotine on cell proliferation and inhibition of apoptosis (Catassi et al., 2008), it is not surprising that we observed the reduced nicotine inhibitory effect on cisplatin-induced apoptosis at nicotine concentration > 1 M. However, the detailed mechanisms of nicotine effects are limited and are beyond the scope of this study. The observation of nicotine inhibitory effect on cisplatininduced apoptosis in OSCC cells at nicotine concentration of 1 M is consistent with the results reported by Xu et al. (2007), in which they found that 1 M nicotine could suppress apoptosis induced by 20 M cisplatin in human oral cancer cell line quiescent Tca8113 cells. According to their terminal transferase dUTP nick end labeling (TUNEL) assay of the apoptosis, 1 M nicotine caused a ∼50% decrease in the percentage of TUNEL positive cells analyzed at 35 h after treatment, compared with that of cisplatin treatment. While conventional methods can only analyze the cellular activity at one time point (for example, 36 h after treatment), in this study, the real-time impedance system allowed us to monitor the cellular responses in the entire experimental period, and enabled us to observe the nicotine inhibitory effect as early as 8 h after treatment. In addition, the real-time impedance method allowed us to test the effect of different concentration of nicotine on cell apoptosis in a high throughput manner. The study demonstrated that the real-time impedance system has advantages over conventional cellular analysis methods. In general, conventional cell-based assays require a labeled ligand or enzyme substrate or a tracer molecule. In this respect, despite their considerable success in drug screening and cellular analysis, they are still limited in several aspects, including cost and time of labels, and increased assay complexity due to wash steps. Moreover, the labeling steps often involve destruction of cells, which lead to the loss of very important information about viable cells. In addition, the conventional method for detection of cellular responses to a drug is usually an end-point analysis, in which the real-time measurement of viable cell activities are not possible. For example, using the TUNEL assay kit, the cells must be processed, fixed, stained and observed under the microscope, the effects of apoptosis can only be analyzed at one time point (Xu et al., 2007). This real-time impedance analysis overcomes the shortcomings of conventional cell-based assays. Its advantages include: (i) it can provide valuable real-time cellular activities of viable cells in the entire history of treatment; (ii) it is a label-free method, requiring no labeling steps which not only save cost and time but also avoid the alternation of cellular information during labeling steps; (iii) this impedancebased method is more sensitive in monitoring early cell responses
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
compared with traditional image analysis; and (iv) the impedancebased method can provide quantitative information about the cell responses in a dose-dependent manner. 4. Conclusions In this study, we have demonstrated the use of impedance-based method for label-free detection of oral cancer cells and real-time monitoring of various cellular activities including cell adhesion, cell spreading, cell proliferation on microelectrodes, and drug-induced apoptosis and inhibition of cell apoptosis. The impedance-based cell index was indicative of the cell number seeded in the detection well and presented a linear relationship with the cell number in the range of 3500 to 35,000 cells/well. This impedance-based method successfully monitored OSCC cell apoptosis induced by the well-known anti-cancer drug-cisplatin. The results showed that cisplatin induced OSCC cell apoptosis at the minimal concentration of 5 M after 20 h of treatment and followed a linear dose-dependent manner in the concentration range from 10 M to 25 M. Microscopic imaging analysis of cell morphology and cell viability analysis confirmed cisplatin-induced apoptosis and correlated well with the impedance-based results. Importantly, the impedance-based method not only provided real-time information about the cellular activity of viable cells, but also detected drug-induced cellular activities much earlier than imaging analysis. The inhibition of cisplatin-induced apoptosis by the carcinogen, nicotine, at concentrations from 0.1 M to 10 M was monitored. The most significant inhibitory effect was observed at concentrations of 0.5–1 M, whereas, high concentrations of nicotine (5 M and 10 M) exhibited less inhibitory effect and low concentrations of nicotine (0.1 M) showed no inhibitory effect on 20 M cisplatin-induced apoptosis. The results in this study have shown the potential of the impedance-based method as a useful analytical approach for cancer research. Acknowledgements The authors acknowledge the funding support from National Science Foundation (CBET-0916138), the Golden LEAF Foundation, and the NCBIOIMPACT initiative of the state of North Carolina through the Biomanufacturing Research Institute and Technology Enterprise (BRITE) at North Carolina Central University (NCCU). We also thank Dr. John Bang at the Department of Environmental Sciences, NCCU for providing the ECIS instrument. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bios.2010.02.029.
2231
References Ambrose, J.A., Barua, R.S., 2004. J. Am. Coll. Cardiol. 43, 1731–1737. Arredondo, J., Chernyavsky, A.I., Grando, S.A., 2006. Cancer Biol. Ther. 5, 511–517. Beitsch, P., Clifford, E., 2000. Am. J. Surg. 180, 446–449. Benez, A., Geiselhart, A., Handgretinger, R., Schiebel, U., Fierlbeck, G., 1999. J. Clin. Lab. Anal. 13, 229–233. Braun, S., Pantel, K., Muller, P., Janni, W., Hepp, F., Kentenich, C., Gastroph, S., Wischnik, A., Dimpfl, T., Kindermann, G., Riethmuller, G., Schlimok, G., 2000. N. Engl. J. Med. 342, 525–533. Carlisle, D.L., Liu, X., Hopkins, T.M., Swick, M.C., Dhir, R., Siegfried, J.M., 2007. Pulm. Pharmacol. Ther. 20, 629–641. Catassi, A., Servent, D., Paleari, L., Cesario, A., Russo, P., 2008. Mutat. Res. 659, 221–231. Centers for Disease Control and Prevention (CDC), 1998. MMWR, 47 (RR-14), 1–12 (http://www.cdc.gov/mmwr/preview/mmwrhtml/00054567.htm). Chen, Y., Zhang, J., Wang, Y., Zhang, L., Julien, R., Tang, K., Balasubramanian, N., 2008. Biosens. Bioelectron. 23, 1390–1396. Cucina, A., Fuso, A., Coluccia, P., Cavallaro, A., 2007. J. Surg. Res. 150, 227–235. Ehret, R., Baumann, W., Brischwein, M., Schwinde, A., Stegbauer, K., Wolf, B., 1997. Biosens. Bioelectron. 12 (1), 29–41. Ehret, R., Baumann, W., Brischwein, M., Schwinde, A., Wolf, B., 1998. Med. Biol. Eng. Comput. 36, 365–370. Fang, Y., 2007. Sensors 7, 2317–2329. Galitovsky, V., Chowdhury, P., Zharov, V.P., 2004. Life Sci. 75, 2677–2687. Giaever, I., Keese, C.R., 1992. Chemtech. Feb, 116–125. Giaever, I., Keese, C.R., 1991. Proc. Natl. Acad. Sci. U. S. A. 88, 7896–7900. Giaever, I., Keese, C.R., 1984. Proc. Natl. Acad. Sci. U. S. A. 81, 3761–3764. Giaever, I., Keese, C.R., 1986. IEEE Trans Biomed. Eng. 33, 242–247. Giaever, I., Keese, C.R., 1993. Nature 366, 591–1591. Klo, B.D., Kurz, R., Jahnke, H.G., Fischer, M., Rothermel, S., Anderegg, U., Simon, J.C., Robitzki, A.A., 2008. Biosens. Bioelectron. 23, 1473–1480. Linderholm, P., Vannod, J., Barrandon, Y., Renaud, P., 2007. Biosens. Bioelectron. 22, 789–796. Liu, Q., Yu, J., Xiao, L., Tang, J.C.O., Zhang, Y., Wang, P., Yang, M., 2009. Biosens. Bioelectron. 24, 1305–1310. Marx, K.A., Zhou, T., Montrone, A., McIntosh, D., Braunhut, S.J., 2005. Anal. Biochem. 343, 23–34. Mauk, M.G., Ziober, B.L., Chen, Z., Thompson, J., Bau, H., 2007. Ann. N. Y. Acad. Sci. 1098, 467–475. McGuinness, R., 2007. Curr. Opin. Pharmacol. 7, 535–540. Onoda, N., Nehmi, A., Weiner, D., Mujumdar, S., Christen, R., Los, G., 2001. Head Neck 23, 860–870. Pantel, K., Cote, R.J., Fodstat, O., 1999. J. Natl. Cancer Inst. 91 (3), 1113–1124. Ramsden, J.J., Li, S.Y., Heinzle, E., Prenosil, J.E., 1995. Cytometry 19, 97–102. Rodriguez, T., Altieri, A., Chatenoud, L., Gallus, S., Bosetti, C., Negri, E., Franceschi, S., Levi, F., Talamini, R., La Vecchia, C., 2004. Oral Oncol. 40, 207–213. Solly, K., Wang, X., Xu, X., Strulovici, B., Zheng, W., 2004. Assay Drug Dev. Technol. 2, 363–372. U.S. Department of Health and Human Services (USDHHS), 2000. Oral health in America: a report of the surgeon general—executive summary. US Department of Health and Human Service, National Institute of Dental and Craniofacial Research, National Institutes of Health, Rockville, MD, pp. 1–13. Weihrauch, M.R., Skibowski, E., Koslowski, T.C., Voiss, W., Re, D., Kuhn-Regnier, F., Bannwarth, C., Siedek, M., Diehl, V., Bohlen, H., 2002. J. Clin Oncol. 20 (21), 4338–4343. Wright, S.C., Zhong, J., Zheng, H., Larrick, J.W., 1993. FASEB J. 7, 1045–1051. Xu, J., Huang, H., Pan, C., Zhang, B., Liu, X., Zhang, L., 2007. Int. J. Oral Maxillofac. Surg. 36, 739–744. Zhang, T., Lu, H., Shang, X., Tian, Y., Zheng, C., Wang, S., Cheng, H., Zhou, R., 2006. Biochem. Biophys. Res. Commun. 342, 928–934. Zhou, T., Marx, K.A., Warren, M., Schulz, H., Braunhut, S.J., 2000. Biotechnol. Prog. 16, 268–277.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Real-time electrical impedance detection of cellular activities of oral cancer cells L. Renea Arias, Carla A. Perry, Liju Yang ∗ Biomanufacturing Research Institute and Technology Enterprises (BRITE), Department of Pharmaceutical Sciences, North Carolina Central University, Durham, NC 27707, USA
a r t i c l e
i n f o
Article history: Received 24 November 2009 Received in revised form 23 February 2010 Accepted 24 February 2010 Available online 3 March 2010 Keywords: Real-time Electrical impedance Cellular activity Cancer cell Apoptosis
a b s t r a c t In this study, the electric cell-substrate impedance sensing (ECIS) system was used to study the cellular activities of oral squamous cell carcinoma (OSCC) cells in a real-time and label-free manner. Various cellular activities, including cell adhesion, spreading, proliferation, and drug-induced apoptosis and inhibition of apoptosis, were monitored. A linear relationship was found between the impedance-based cell index and the cell number in the range of 3500 to 35,000 cells/well. Anti-cancer drug-cisplatin-induced OSCC cell apoptosis at the minimal concentration of 5 M after 20 h of treatment and followed a linear dose-dependent manner in the concentration range from 10 M to 25 M. The inhibition of cisplatininduced apoptosis by the carcinogen, nicotine, at concentrations from 0.1 M to 10 M was monitored. The most significant inhibitory effect of nicotine on cisplatin-induced apoptosis was observed at concentrations of 0.5–1 M. The results obtained with impedance method correlated well with microscopic imaging analysis of cellular morphology and cell viability analysis. This study demonstrated that the impedance-based method can provide real-time information about the cellular activity of viable cells and detect drug-induced cellular activities much earlier than commonly used cell-based image analysis. This impedance-based method has the potential to provide a useful analytical approach for cancer research. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Analysis of cellular properties and physiological functions of a certain type of cells in human body has been an effective approach to understand many biological and biomedical problems. For example, detection and analysis of cancer cells presented in body fluids can be a indicator of prognosis and diagnosis in oncology (Pantel et al., 1999). Viable tumor-derived epithelial cells or circulating tumor cells (CTCs) have been identified in peripheral blood from cancer patients, and have been found to correlate with clinical stage (Benez et al., 1999; Weihrauch et al., 2002), patient survival after therapy (Braun et al., 2000), and tumor metastasis (Beitsch and Clifford, 2000). In addition, cell-based methods are important for developing rapid and simple techniques to study cancerous cells and their interactions with drugs, especially in anti-cancer drug discovery and screening. Current typical approaches for cellular analysis include flow cytometry and microscopic imaging. While these methods can provide insights into the physiological function of any particular cell or of pathological changes that may have occurred; they usually require fluorescent, chemiluminescent, or radioactive labeling steps which often involve destruction of cells. The labeling processes may thus lead to the loss of very important biological information about viable cells.
∗ Corresponding author. Tel.: +1 919 530 6704; fax: +1 919 530 6600. E-mail address: [email protected] (L. Yang). 0956-5663/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2010.02.029
Label-free and non-invasive detection of cells and their functions is advantageous to provide real-time and/or kinetic cellular activities of viable cells, which could be of great use in many biomedical applications. A number of label-free technologies, including cell-substrate impedance (Giaever and Keese, 1986, 1993), quartz microbalance (Zhou et al., 2000; Marx et al., 2005), and optical waveguide lightmode spectroscopy (Ramsden et al., 1995; Fang, 2007), have been reported as a means for monitoring live cell status in a non-invasive and real-time manner. Amongst these, electrical/electrochemical impedance spectroscopy (EIS) has been recognized as a powerful electrochemical technique that can monitor live cell behavior in real-time. The distinct electrical properties associated with specific biological cells satisfy the impedance techniques for cellular analysis. By culturing biological cells on an electrode surface, EIS can directly sense detailed information about cellular activities occurring on an electrode or substrate’s surface by measuring the induced capacitance and/or resistance changes, eliminating multiple labeling and amplification steps typically used in many other cell-based methods, and allowing label-free and non-invasive study of cellular properties. In addition, impedance technique is amenable to miniaturized electronic systems to meet the growing needs of microdevices for point-of-care analysis. Impedance measurements using microelectrodes were first used to study the characteristics of anchorage dependent cultured cell lines by Giaever and Keese (1984). Since then, they have continuously improved and refined their method and named it electric cell-substrate impedance sensing (ECIS) (Giaever and
2226
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Keese, 1991, 1992, 1993). Other researchers have also been using similar impedance sensing to enumerate cells attached to a substrate (Ehret et al., 1997, 1998). Recently, the impedance-based sensing technology has gained a great deal of attention for studying cancer cells and monitoring drug-induced cellular activities for drug discovery (Solly et al., 2004; Linderholm et al., 2007; McGuinness, 2007; Klo et al., 2008; Chen et al., 2008; Liu et al., 2009). In this study, the ECIS system was used for label-free detection of oral cancer cells and studying the cellular activities induced by drugs. Oral cancer is one of the most common cancers worldwide, and oral squamous cell carcinoma (OSCC) occurs with the highest frequency. In the US, OSCC represents 2–4% of the annually diagnosed malignancies, accounting for 8000 deaths every year (CDC, 1998; USDHHS, 2000). Research has indicated that long-term tobacco usage is one of the important risk factors that predominately cause OSCC (Rodriguez et al., 2004), but limited information about the effect of nicotine on OSCC cell cellular activities was available. It has been suggested that nicotine, the major component of tobacco, plays several roles to alter cellular functions, including activation of mitogenic pathways, angiogenesis, and cell growth, in many cell types. It has also been reported that nicotine can inhibit the drug- and radiation-induced apoptosis of different cell lines (Wright et al., 1993; Xu et al., 2007). In this study, the impedancebased system was used not only to monitor the cell adhesion, spreading, and proliferation of the human oral cancer cell line, CAL27, but also to study the cellular activities of CAL27 cells induced by a well-known anti-cancer drug, cisplatin, and the inhibition of cisplatin-induced apoptosis by nicotine. 2. Materials and methods
based cell index was measured in the designed period of time with time intervals varying from 5 min to 30 min. 2.3. Drug treatments The well established anti-cancer drug, cisplatin, was obtained from Sigma–Aldrich Co. (St. Louis, MO). In the drug treatment tests, OSCC cells (5000 cells/well) were seeded in 16-well plates overnight (approximately 20 h), then aliquots of 15 l cisplatin solution were added to the wells at final concentrations of 5, 10, 15, 20, 25 M. Impedance-based cell indexes were recorded continuously over the entire experimental period. Nicotine (Sigma–Aldrich, St. Louis, MO), a carcinogen, was used to test its effect on inhibition of cisplatin-induced apoptosis of OSCC cells. Nicotine with final concentration of 0.1 M, 0.5 M, 1.0 M, 5.0 M and 10.0 M was used to treat the OSCC cells in the presence and absence of cisplatin. 2.4. Imaging The microscopic images of OSCC cells in the wells were taken on a Nikon ECLIPSE E600FN microscope (Japan) with a Coolsnap HQ camera (Roper Scientific, Inc., Photometric, Tucson, AZ) using reflective bright field mode. 2.5. Statistical analysis All experiments were repeated in triplicate. Data in the figures were expressed as either mean of three repeated results or mean ± standard deviation (SD). Statistical significance of the data was determined using the Student’s t-test. Differences were considered significant at the level of P < 0.05.
2.1. Cell culture 3. Results and discussion The OSCC cell line, CAL27, derived from the tongue (ATCC # CRL-2095), was cultured in Dulbecco’s modified eagle’s medium (DMEM) (Sigma) supplemented with 10% of fetal bovine serum and antibiotics (100 IU/ml penicillin and 100 g/ml streptomycin) (Lonza, Walkersville, MD). Cells were cultured in 75 cm2 flask and were incubated at 37 ◦ C in an atmosphere of 5% CO2 in air. The medium was renewed every 2–3 days. When confluent, cells were detached from the flask using 0.25% trypsin with 0.53 mM EDTA solution. Cells were centrifuged to remove trypsin. Cell pellet was resuspended with fresh DMEM medium. Cell number in the suspension was determined using the Vi-cell XR cell counting system (Beckman Coulter, Miami, FL). Desired cell concentrations were obtained by diluting the cell suspension with fresh DMEM for further experiments. Cell viability analysis was also performed using the Vi-cell system, which utilizes the trypan blue dye exclusion method. The dead and non-viable cells are permeable and can uptake the trypan blue dye, thus become darker than the viable cells. The Vi-cell system analyzes up to 50 images for the determination of cellular viability which is defined as the percentage of viable cells in the total cell number. 2.2. Real-time impedance measurement of cellular activities The real-time impedance-based measurement of OSCC cellular activities was performed using the RT-CES system (ACEA Biosciences, Inc., San Diego, CA). Detailed information about the components and the principle of the system is in the Supplementary Materials S1 and S2. For OSCC cell measurements, 50 l of DMEM was added to the wells of the 16-well E-plate for taking background readings, then 100 l of OSCC cells in DMEM with the desired concentration was added to each well. Impedance-
3.1. Real-time monitoring of the attachment, spreading and proliferation of OSCC cells In cancer biology, cell adhesion to the extracellular matrix is important to tumor metastasis, thus cell morphology is another most important characteristic in study of cancer cells. By culturing OSCC cells on the microelectrode surface, impedance signals from the microelectrodes can reflect both the cell adhesion to the electrodes and the cell morphology. Fig. 1A and B shows two representative cellular activity curves of OSCC cells at (A) high (3.5 × 104 cells/well) and (B) low (3.5 × 103 cells/well) cell concentrations cultured on the microelectrodes for 20 h. The cell index curves show the average values of three measurements with the error bar showing the standard deviation of the three measurements at each data point. For high cell concentration, the cell index curve (Fig. 1A) presents three distinct regions with respect to the rate of cell index change, which most likely reflects the initial cell adhesion, followed by cell spreading, and cell proliferation stages. In the first 1–2 h, the cell index increased slowly with an increase rate of 0.085/h. This period of time corresponded to the initial cell adhesion stage, in which the cells started to settle down toward the electrode surface and became adherent to the electrodes. In the next 2–4 h, the cell index increased rapidly at a rate of 0.42/h. Most likely, in this period, the adhered cells spread over the electrode surface. After spreading, cells started to proliferate, and the rate of increase of the cell index again slowed to 0.084/h. At low cell concentration (3.5 × 103 cells/well), only two distinct regions on the cell index curve corresponding to the cell spreading and cell proliferation stages are visible-no distinct region can be observed between the cell adhesion and cell spreading stages. A possible explanation could be that because of the low cell number, the time required for the cells
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
2227
Fig. 1. (A) and (B) Two representative cell index curves of OSCC cells at high (3.5 × 104 cells/well) and low (3.5 × 103 cells/well) cell concentrations cultured on the microelectrodes for 20 h. (C)–(F) The microscopic images of the OSCC cells (3.5 × 104 cells/well) after they were seeded in the well for 30 min, 2 h, 5 h, and 22 h.
to settle and adhere to the electrodes was minimal, and the cells quickly start to spread over the electrode surface. The cell proliferation stage started approximately 4 h after the cells were seeded in the well. Fig. 1C–F shows the microscopic images of the OSCC cells (3.5 × 104 cells/well) after they were seeded in the well for 30 min, 2 h, 5 h and 22 h. It can be seen that at 30 min, the cells started to settle down onto the electrode surface, and the round shape of the
cells indicated that they were not spread yet. Over the first 2 h, cells continued to settle down onto the electrode surface from the suspension. The cells are still round, indicating that significant cell spreading had not yet occurred. At 5 h, cells started to spread over the electrode surface. After spreading, the cells started to proliferate. At 22 h, the cells fully covered on the electrode surface. These images provide strong evidence to interpret the pattern of the cell index curve in Fig. 1A. Correlating to the cell index change in Fig. 1A,
Fig. 2. (A) A group of cell index curves with growth time for the samples with different cell numbers ranging from 1.0 × 103 to 3.5 × 104 cells/well. (B) The linear relationship between the cell number and the cell index measured at 15 h after the cell were seeded in the wells. (C)–(F) The microscopic images of cell coverage on the microelectrodes after the cells fully spread when 1000, 5000, 10,000, and 20,000 cells were seeded in the wells. Cell index curves in (A) shows averaged values from three measurements, and error bars in (A) and (B) are the standard deviations of three measurements.
2228
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Fig. 3. (A) A group of cell index curves reflecting the apoptosis of OSCC cells induced by different concentrations of cisplatin (5 M, 10 M, 15 M, 20 M and 25 M). Initial number of OSCC cells: 5000 cells/well. (B) The dose-dependent cell index measured at 3 h, 5 h, 10 h, 15 h, and 20 h after cisplatin treatment, derived from three repeated tests as in (A). The error bar on each data point is the standard deviation of the three measurements,.
we can see that the cell index increased slowly due to the increasing number of cells settled down on the electrode surfaces in the first 1–2 h. While the cell spreading caused the most significant and rapid increase in the cell index, the cell proliferation generated slower increases in the cell index. The rate of cell index increase due to cell spreading is about 5 times the rate measured during initial cell adhesion or cell proliferation. These results suggest that the impedance-based cell index can be used as reliable measure to study the kinetics of cell spreading and proliferation. 3.2. Detection of OSCC cells Fig. 2A shows the cell index curves with growth time for the samples with different cell numbers ranging from 1.0 × 103 to 3.5 × 104 cells/well. For each sample, the cell index increases with the time which corresponds the cellular activities including cell attachment, spreading, and proliferation over time. Between the samples with different cell numbers, at each time point, the higher cell number has higher cell index. As the cell size, morphology, the extent of cell spreading, and the rate of proliferation are relatively constant in the same type of cells, the impedance-based cell index should be indicative of the cell number. Derived from Fig. 2A, Fig. 2B shows the relationship between the cell number and the cell index measured at 15 h during the cell proliferation stage. It shows that the cell index is proportional to the cell number in the range of cell number between 3500 and 35,000 cells/well, with the linear regression equation of I (cell index) = 8 × 10−5 N (cells/well) − 0.1124. The sample with 1000 cells/well did not generate a cell index that is significantly different from the blank, indicating that such low number of cells is unable to generated detectable cell index. As the cell index is related to the cell coverage on electrode surface, we examined the cell coverage of samples with different cell numbers. Fig. 2C–F shows the representative images of cell coverage on the electrode generated by different number of cells after these cells fully spread on the electrode surfaces. Estimated from several images taken at different locations in each well, the average cell coverage generated by 1000, 5000, 10,000, and 20,000 cells/well are approximately 2.4%, 12.0%, 38.0%, 72.0%, respectively. Linked to the cell index measurements, the results indicated that a ∼10% cell coverage on the electrode surface is required to generate a detectable impedance-based cell index. The results here demonstrated that this impedance-based method could offer a label-free and non-invasive quantitative method to detect OSCC cells. This can be useful in clinic diagnostics, such as detection of the presence of cancer cells in body fluids. Detection of CTCs can be a potential indicator of prognosis and diagnosis in oncology (Pantel et al., 1999), since the presence of cancer cells in the bloodstream or bone marrow is correlated with clinical stage (Benez et al., 1999; Weihrauch et al., 2002), patient survival
after therapy (Braun et al., 2000), and tumor metastasis (Beitsch and Clifford, 2000). In the mouth, the surface layer of cells is replaced about every 2–4 h; and the turnover time of the oral epithelium is about 4.5 days. If a person is developing an oral cancer, cancer cells can be shed into saliva at very early stage of the cancer; and the number of cancer cells in saliva can be a measure of the cancer stage. Mauk et al. (2007) reported that more than 1000 cells/ml and 9000 cells/ml of OSCC cells were separated from tumor stage 1 and 4 patient saliva, respectively. This makes saliva an ideal sample for early screening and detection of oral cancers. The impedance-based method has the potential to be a sensitive non-invasive screening method for detecting early stage cancer by detecting cancer cells in saliva, and an approach to obtain quantitative information about cancer stage or to monitor the progress of cancer treatment. 3.3. Real-time monitoring of apoptosis of OSCC induced by cisplatin Fig. 3A shows a group of representative cell index curves reflecting the effects of different concentrations of cisplatin on CAL27 cells cultured on the microelectrodes at an initial cell number of 5000 cells/well. The increased cell index in the control sample indicated cell growth with time, whereas the decreased cell index in other samples treated with increasing concentrations of cisplatin (5 M, 10 M, 15 M, 20 M and 25 M) reflected the increasing extents of cell death resulting from cisplatin-induced apoptosis. Each cell index curve also provided detailed kinetic information of cell responses to the particular cisplatin concentration. For instance, compared to the control, the cell index of OSCC cells treated with 5 M cisplatin starts to decrease approximately 20 h after treatment, whereas the one treated with 10 M cisplatin starts to decrease at approximately 10 h after treatment. With treatments of 15 M, 20 M and 25 M cisplatin, the cell indexes start to decline within 1–3 h after treatments. The results indicated that low concentrations of cisplatin required longer time to induce OSCC cell apoptosis, while high concentrations of cisplatin-induced OSCC cell apoptosis quickly. And this impedance-based method can monitor cisplatin-induced cell apoptosis in a real-time and label-free manner, and provide dynamic information about cellular response to drug treatment over the entire experimental period. Derived from three repeated tests as in Fig. 3A, Fig. 3B shows the dose-dependent cell index curves of OSCC cells in response to different concentrations of cisplatin at 3 h, 5 h, 10 h, 15 h and 20 h after treatment. The reproducibility of the measurement is indicated by the error bars which are the standard deviation of the three measurements. At each measurement time, the change in cell index induced by cisplatin treatment increased with the increasing concentration of cisplatin. For all these measurement times, the linear dose response range can be observed in the cisplatin concentration
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
2229
Fig. 4. The images of OSCC cells showing the cell morphology at 5 h and 22 h after treatment with 20 M of cisplatin, along with the images of the untreated control cells. Initial cell number: 5000 cells/well and 35,000 cells/well.
range from 10 M to 25 M. Obviously, the cell indexes monitored at longer treatment times (15 h, 20 h, or more) show clearer apoptosis effects. The time required to induce significant apoptosis of OSCC cells is also related the concentration of cisplatin. The apoptosis induced by 10 M of cisplatin can be detected at 20 h after treatment, while the apoptosis induced by 20 M of cisplatin can be detected as early as 5 h after treatment. Fig. 4 shows the images of OSCC cells (5000 cells/well and 35,000 cells/well) at 5 h and 22 h after treatment with 20 M of cisplatin, along with the images of the untreated control cells. The images of OSCC cells at the two cell number levels show similar patterns: at 5 h after treatment, no significant morphology change can be seen, and at 22 h after treatment, cells showed significant morphology change which indicated cell apoptosis. Linking to the cell index change in Fig. 3A, the declining of cell index was clearly observed as early as 5 h after treatment, indicating that the impedance-based cell index is more sensitive to the changes in cellular activities induced by cisplatin treatment at earlier time. The effects of drug treatment can be obtained before any morphology change is observed in imaging analysis. At 22 h after cisplatin treatment, the cells are not tightly attached to the electrode surface, the apoptotic cells became round in shape. Correspondingly, the cell index at 22 h after treatment decreased significantly (Fig. 3A), indicting the impedance results are correlated with the significant morphology changes observed in imaging analysis. The morphology changes in the two different levels of cell numbers confirmed that cell density in the wells (confluent or not) did not affect the cisplatin-induced apoptosis on OSCC cells. Cell viability after cisplatin treatment was analyzed by the Vi-cell system. Upon the treatment of 20 M cisplatin, the cell viability decreased from 97.0 ± 2.1% (control) to 92.3 ± 2.8% at 5 h after treatment, and to 82.8 ± 4.0% at 22 h after treatment (Fig. S3(A) in Supplementary Materials). Statistical analysis showed that the cell viability at 22 h after treatment was significantly different from the control, while that at 5 h after treatment was not significantly different. Upon the treatment of cispatin at 10 M, 20 M and 30 M, the cell viabilities measured at 22 h after treat-
ment were 92.7 ± 0.9%, 82.8 ± 4.0% and 59.5 ± 1.3%, respectively (Fig. S3(B) in Supplementary Materials). Statistical analysis showed significant decrease in cell viability at 22 h upon the treatment of 20 M and 30 M cisplatin. The decreases in cell viability with treatment time and with the increasing cisplatin concentration were correlated with the trend of impedance-based cell index decrease observed in Fig. 3A. 3.4. Real-time monitoring of nicotine inhibiting the apoptosis induced by cisplatin Nicotine, as a major component in tobacco, has long been recognized as a carcinogen. It has been implicated as a risk factor for a number of tobacco-related diseases, such as coronary artery disease, cardiovascular disease, atherosclerosis formation (Cucina et al., 2007; Ambrose and Barua, 2004), and several types of cancers (Catassi et al., 2008). Nicotine has been shown to affect various cellular systems, including lung cancer (Zhang et al., 2006; Catassi et al., 2008), pancreas cancer (Galitovsky et al., 2004), and oral cancer (Xu et al., 2007). Several studies have focused on the correlation between nicotine and apoptosis in a variety of cell types, but the results are contradictory. Some studies show that nicotine seems to act as an inhibitor of apoptosis, while others show that it is an inductor of apoptosis. For example, Zhang et al. (2006) reported that nicotine prevented the apoptosis induced by menadione in human lung cancer cells. Galitovsky et al. (2004) reported nicotine-induced apoptosis in pancreatic cancer cells, and the most significant apoptotic effect was observed at nicotine concentration of 100 M to 1 mM. In smooth muscle cells (SMC), Cucina et al. (2007) found that nicotine not only inhibited the SMC apoptosis but also increased the SMC proliferation. Carlisle et al. (2007) reported that nicotine may play a role in regulating survival of non-small cell lung cancer (NSCLC) cells, by activating cell-signaling pathways through muscle-type and neuronal nicotintic acetylcholine receptors in NSCLC cells. Catassi et al. (2008) reviewed the multiple roles of nicotine on cell proliferation and inhibition of apoptosis in human lung cancers. Different research groups have studied the
2230
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
Fig. 5. (A) The representative cell index curves of the control, the cells treated with 10 M nicotine only, 20 M cisplatin only, and the cells treated with 0.1 M, 1.0 M and 5.0 M in the presence of 20 M of cisplatin. (B) The cell indexes measured at 18 h after they were treated with 0.1 M, 0.5 M, 1.0 M, 5.0 M and 10 M of nicotine in the presence of 20 M cisplatin, along the cell indexes of the cells treated with 20 M cisplatin only and the control. The error bars are the standard deviations of three repeated test results. Data labeled with an asterisk was significantly different from the result of cisplatin-treated sample, P < 0.05.
effect of nicotine on head and neck cancer cells (Xu et al., 2007; Arredondo et al., 2006; Onoda et al., 2001). For example, Xu et al. (2007) reported that nicotine inhibited cisplatin-induced apoptosis in human oral cancer cell line Tca8113. Others reported that nicotine affected the signaling of the death pathway and resulted in a decreased cytotoxicity of various anticancer agents such as cisplatin and gamma-radiation in head and neck cancer cell lines UMSCC 10b and UMSCC5 (Onoda et al., 2001). In this study, we used the real-time impedance system to monitor the cellular responses of OSCC cell line cells, CAL 27, when they were treated with nicotine in the presence and absence of cisplatin. Fig. 5A shows a group of representative cell index curves reflecting the effect of different concentrations of nicotine on the OSCC cells in the presence and absence of cisplatin. It shows that the pattern of the cell index curve of OSCC cells with 10 M nicotine treatment is similar to that of the control, indicating that nicotine only did not affect the OSCC cells’ growth. The cells treated with 20 M cisplatin shows reduced cell index compared with the control, indicating the cisplatin-induced apoptosis. The cells co-treated with 1 M nicotine and 20 M cisplatin presents higher cell indexes than that with 20 M cisplatin treatment, indicating that 1 M nicotine has an inhibitory effect on the cisplatin-induced cell apoptosis. We also tested the effect of different concentrations of nicotine in the presence of 20 M cisplatin, and found that 0.1 M nicotine did not cause significant inhibition on the cisplatin-induced apoptosis. The cell index curve shows that treatment with 0.5 M nicotine exhibited significant inhibition on the cisplatin-induced apoptosis. However, when the nicotine concentration increased to 5 M, the cell indexes are lower than those with 1 M and 0.5 M nicotine treatments, but still higher than those of cisplatin treatment only, which suggest that the 5 M nicotine inhibitory effect on the cisplatin-induced apoptosis is not as strong as that by 1 M and 0.5 M nicotine. Derived from three repeated tests as in Fig. 5A, Fig. 5B shows the cell indexes read at 40 h of OSCC cell growth and approximately 18 h after treatments, when the cells were treated with 20 M cisplatin only, and 20 M cisplatin with different concentrations of nicotine (0.1 M to 10 M). The results clearly show that low concentration of nicotine did not exhibit inhibitory effect on 20 M cisplatin-induced apoptosis; Nicotine at concentrations of 0.5 M to 1 M exhibited the most significant inhibitory effect on 20 M cisplatin-induced apoptosis; and Nicotine inhibitory function decreased when its concentration increased from 1 M to 10 M. With 0.5 M and 1 M nicotine treatments, the cell indexes increased by 68.2% and 73.9%, respectively, compared with the cell index of cisplatin treatment, whereas 5 M and 10 M nico-
tine treatment increased the cell indexes by 52.1% and 17.3%, respectively. Considering the multiple roles of nicotine on cell proliferation and inhibition of apoptosis (Catassi et al., 2008), it is not surprising that we observed the reduced nicotine inhibitory effect on cisplatin-induced apoptosis at nicotine concentration > 1 M. However, the detailed mechanisms of nicotine effects are limited and are beyond the scope of this study. The observation of nicotine inhibitory effect on cisplatininduced apoptosis in OSCC cells at nicotine concentration of 1 M is consistent with the results reported by Xu et al. (2007), in which they found that 1 M nicotine could suppress apoptosis induced by 20 M cisplatin in human oral cancer cell line quiescent Tca8113 cells. According to their terminal transferase dUTP nick end labeling (TUNEL) assay of the apoptosis, 1 M nicotine caused a ∼50% decrease in the percentage of TUNEL positive cells analyzed at 35 h after treatment, compared with that of cisplatin treatment. While conventional methods can only analyze the cellular activity at one time point (for example, 36 h after treatment), in this study, the real-time impedance system allowed us to monitor the cellular responses in the entire experimental period, and enabled us to observe the nicotine inhibitory effect as early as 8 h after treatment. In addition, the real-time impedance method allowed us to test the effect of different concentration of nicotine on cell apoptosis in a high throughput manner. The study demonstrated that the real-time impedance system has advantages over conventional cellular analysis methods. In general, conventional cell-based assays require a labeled ligand or enzyme substrate or a tracer molecule. In this respect, despite their considerable success in drug screening and cellular analysis, they are still limited in several aspects, including cost and time of labels, and increased assay complexity due to wash steps. Moreover, the labeling steps often involve destruction of cells, which lead to the loss of very important information about viable cells. In addition, the conventional method for detection of cellular responses to a drug is usually an end-point analysis, in which the real-time measurement of viable cell activities are not possible. For example, using the TUNEL assay kit, the cells must be processed, fixed, stained and observed under the microscope, the effects of apoptosis can only be analyzed at one time point (Xu et al., 2007). This real-time impedance analysis overcomes the shortcomings of conventional cell-based assays. Its advantages include: (i) it can provide valuable real-time cellular activities of viable cells in the entire history of treatment; (ii) it is a label-free method, requiring no labeling steps which not only save cost and time but also avoid the alternation of cellular information during labeling steps; (iii) this impedancebased method is more sensitive in monitoring early cell responses
L.R. Arias et al. / Biosensors and Bioelectronics 25 (2010) 2225–2231
compared with traditional image analysis; and (iv) the impedancebased method can provide quantitative information about the cell responses in a dose-dependent manner. 4. Conclusions In this study, we have demonstrated the use of impedance-based method for label-free detection of oral cancer cells and real-time monitoring of various cellular activities including cell adhesion, cell spreading, cell proliferation on microelectrodes, and drug-induced apoptosis and inhibition of cell apoptosis. The impedance-based cell index was indicative of the cell number seeded in the detection well and presented a linear relationship with the cell number in the range of 3500 to 35,000 cells/well. This impedance-based method successfully monitored OSCC cell apoptosis induced by the well-known anti-cancer drug-cisplatin. The results showed that cisplatin induced OSCC cell apoptosis at the minimal concentration of 5 M after 20 h of treatment and followed a linear dose-dependent manner in the concentration range from 10 M to 25 M. Microscopic imaging analysis of cell morphology and cell viability analysis confirmed cisplatin-induced apoptosis and correlated well with the impedance-based results. Importantly, the impedance-based method not only provided real-time information about the cellular activity of viable cells, but also detected drug-induced cellular activities much earlier than imaging analysis. The inhibition of cisplatin-induced apoptosis by the carcinogen, nicotine, at concentrations from 0.1 M to 10 M was monitored. The most significant inhibitory effect was observed at concentrations of 0.5–1 M, whereas, high concentrations of nicotine (5 M and 10 M) exhibited less inhibitory effect and low concentrations of nicotine (0.1 M) showed no inhibitory effect on 20 M cisplatin-induced apoptosis. The results in this study have shown the potential of the impedance-based method as a useful analytical approach for cancer research. Acknowledgements The authors acknowledge the funding support from National Science Foundation (CBET-0916138), the Golden LEAF Foundation, and the NCBIOIMPACT initiative of the state of North Carolina through the Biomanufacturing Research Institute and Technology Enterprise (BRITE) at North Carolina Central University (NCCU). We also thank Dr. John Bang at the Department of Environmental Sciences, NCCU for providing the ECIS instrument. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bios.2010.02.029.
2231
References Ambrose, J.A., Barua, R.S., 2004. J. Am. Coll. Cardiol. 43, 1731–1737. Arredondo, J., Chernyavsky, A.I., Grando, S.A., 2006. Cancer Biol. Ther. 5, 511–517. Beitsch, P., Clifford, E., 2000. Am. J. Surg. 180, 446–449. Benez, A., Geiselhart, A., Handgretinger, R., Schiebel, U., Fierlbeck, G., 1999. J. Clin. Lab. Anal. 13, 229–233. Braun, S., Pantel, K., Muller, P., Janni, W., Hepp, F., Kentenich, C., Gastroph, S., Wischnik, A., Dimpfl, T., Kindermann, G., Riethmuller, G., Schlimok, G., 2000. N. Engl. J. Med. 342, 525–533. Carlisle, D.L., Liu, X., Hopkins, T.M., Swick, M.C., Dhir, R., Siegfried, J.M., 2007. Pulm. Pharmacol. Ther. 20, 629–641. Catassi, A., Servent, D., Paleari, L., Cesario, A., Russo, P., 2008. Mutat. Res. 659, 221–231. Centers for Disease Control and Prevention (CDC), 1998. MMWR, 47 (RR-14), 1–12 (http://www.cdc.gov/mmwr/preview/mmwrhtml/00054567.htm). Chen, Y., Zhang, J., Wang, Y., Zhang, L., Julien, R., Tang, K., Balasubramanian, N., 2008. Biosens. Bioelectron. 23, 1390–1396. Cucina, A., Fuso, A., Coluccia, P., Cavallaro, A., 2007. J. Surg. Res. 150, 227–235. Ehret, R., Baumann, W., Brischwein, M., Schwinde, A., Stegbauer, K., Wolf, B., 1997. Biosens. Bioelectron. 12 (1), 29–41. Ehret, R., Baumann, W., Brischwein, M., Schwinde, A., Wolf, B., 1998. Med. Biol. Eng. Comput. 36, 365–370. Fang, Y., 2007. Sensors 7, 2317–2329. Galitovsky, V., Chowdhury, P., Zharov, V.P., 2004. Life Sci. 75, 2677–2687. Giaever, I., Keese, C.R., 1992. Chemtech. Feb, 116–125. Giaever, I., Keese, C.R., 1991. Proc. Natl. Acad. Sci. U. S. A. 88, 7896–7900. Giaever, I., Keese, C.R., 1984. Proc. Natl. Acad. Sci. U. S. A. 81, 3761–3764. Giaever, I., Keese, C.R., 1986. IEEE Trans Biomed. Eng. 33, 242–247. Giaever, I., Keese, C.R., 1993. Nature 366, 591–1591. Klo, B.D., Kurz, R., Jahnke, H.G., Fischer, M., Rothermel, S., Anderegg, U., Simon, J.C., Robitzki, A.A., 2008. Biosens. Bioelectron. 23, 1473–1480. Linderholm, P., Vannod, J., Barrandon, Y., Renaud, P., 2007. Biosens. Bioelectron. 22, 789–796. Liu, Q., Yu, J., Xiao, L., Tang, J.C.O., Zhang, Y., Wang, P., Yang, M., 2009. Biosens. Bioelectron. 24, 1305–1310. Marx, K.A., Zhou, T., Montrone, A., McIntosh, D., Braunhut, S.J., 2005. Anal. Biochem. 343, 23–34. Mauk, M.G., Ziober, B.L., Chen, Z., Thompson, J., Bau, H., 2007. Ann. N. Y. Acad. Sci. 1098, 467–475. McGuinness, R., 2007. Curr. Opin. Pharmacol. 7, 535–540. Onoda, N., Nehmi, A., Weiner, D., Mujumdar, S., Christen, R., Los, G., 2001. Head Neck 23, 860–870. Pantel, K., Cote, R.J., Fodstat, O., 1999. J. Natl. Cancer Inst. 91 (3), 1113–1124. Ramsden, J.J., Li, S.Y., Heinzle, E., Prenosil, J.E., 1995. Cytometry 19, 97–102. Rodriguez, T., Altieri, A., Chatenoud, L., Gallus, S., Bosetti, C., Negri, E., Franceschi, S., Levi, F., Talamini, R., La Vecchia, C., 2004. Oral Oncol. 40, 207–213. Solly, K., Wang, X., Xu, X., Strulovici, B., Zheng, W., 2004. Assay Drug Dev. Technol. 2, 363–372. U.S. Department of Health and Human Services (USDHHS), 2000. Oral health in America: a report of the surgeon general—executive summary. US Department of Health and Human Service, National Institute of Dental and Craniofacial Research, National Institutes of Health, Rockville, MD, pp. 1–13. Weihrauch, M.R., Skibowski, E., Koslowski, T.C., Voiss, W., Re, D., Kuhn-Regnier, F., Bannwarth, C., Siedek, M., Diehl, V., Bohlen, H., 2002. J. Clin Oncol. 20 (21), 4338–4343. Wright, S.C., Zhong, J., Zheng, H., Larrick, J.W., 1993. FASEB J. 7, 1045–1051. Xu, J., Huang, H., Pan, C., Zhang, B., Liu, X., Zhang, L., 2007. Int. J. Oral Maxillofac. Surg. 36, 739–744. Zhang, T., Lu, H., Shang, X., Tian, Y., Zheng, C., Wang, S., Cheng, H., Zhou, R., 2006. Biochem. Biophys. Res. Commun. 342, 928–934. Zhou, T., Marx, K.A., Warren, M., Schulz, H., Braunhut, S.J., 2000. Biotechnol. Prog. 16, 268–277.