Article pubs.acs.org/est
Rapid and Sensitive Screening of 17β-Estradiol Estrogenicity Using Fourier Transform Infrared Imaging Spectroscopy (FT-IRIS) Candice M. Johnson,† Nancy Pleshko,‡ Mohan Achary,§ and Rominder P. S. Suri*,† †
NSF Water & Environmental Technology (WET) Center, Department of Civil and Environmental Engineering, Temple University, Philadelphia, Pennsylvania 19122, United States ‡ Tissue Imaging and Spectroscopy Lab, Department of Bioengineering, Temple University, Philadelphia, Pennsylvania 19122, United States § Temple University, Department of Radiation Oncology, 3401 N. Broad Street, Philadelphia, Pennsylvania 19140, United States ABSTRACT: It is important to develop rapid and sensitive screening assays to assess the biological effects of emerging contaminants. In this contribution, the ability to determine the molecular level effects of 17β-estradiol on single MCF-7 cells using Fourier transform infrared imaging spectroscopy (FTIRIS) was investigated. The use of FT-IRIS enabled subcellular imaging of the cells and determination of a dose dependent response in mucin concentration at 24 and 48 h of incubation. The 48 h increase in mucin was comparable to increases in cellular proliferation (Pearson R = 0.978). The EC50 values for the E-screen and FT-IRIS assays were 2.29 and 2.56 ppt, respectively, indicating that the molecular changes, which are observed at the single cell level using FT-IRIS, are reflective of physiological changes that are observed as the cell population responds to 17ß-estradiol. The FT-IRIS method, when combined with principal component analysis, enabled differentiation and grouping of cells exposed to varying concentrations of 17ßestradiol. The FT-IRIS method shows potential to be used as a rapid and sensitive screening technique for the detection of biological responses to different emerging contaminants in relevant cells or tissues.
■
protein, and lipid content of fish liver tissue exposed to 17βestradiol. Similar changes were also observed for an estrogen mimic, nonylphenol.4 Holman et al.5 probed the single cell responses of HEPG2 liver cells to 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) using synchrotron-FTIR and observed dosedependent increases in the ratio of the symmetric to asymmetric phosphate modes (associated with nucleic acids), with increasing yet environmentally relevant concentrations of TCDD. Further, FT-IRIS has been used as a sensitive diagnostic tool to identify different tissue types and areas of malignancy.10,11 Motivated by these and other studies, the ability to detect dose-dependent changes in MCF-7 cells exposed to 17β-estradiol using Fourier transform infrared imaging spectroscopy (FT-IRIS), a modality that couples a conventional FTIR spectrometer with an imaging microscope, was investigated in the current study. The quality of images from FT-IRIS instruments with thermal sources have improved over the past decade mainly due to optimally designed optic systems, and the use of very sensitive detectors.12 While this allows for analysis of features up to ∼6 to 10 μm, including single cells, the images produced may yield less information than the brighter synchrotron
INTRODUCTION Emerging contaminants (ECs) of concern include chemicals, biological agents, or nanomaterials that may potentially threaten human or environmental health. There is limited information on regulatory guidelines for these contaminants, and standards related to this are still evolving. ECs include both natural and man-made chemicals such as hormones, pharmaceuticals, personal care products, perfluorinated compounds, and flame-retardants, among others. In order to gain further understanding of the risks posed by these chemicals, it is important to monitor the occurrence and biological activity of the ECs in different environmental media. Analytical instruments such as liquid (or gas) chromatography tandem mass spectrometry along with sophisticated extraction methods enable the quantification of the ECs; however, they give little indication into their biological activity.1 As such, bioanalytical tools, or bioassays, are relied on to estimate the total biological effects of both known and unknown mixtures of varying components.1,2 Fourier transform infrared (FTIR) analysis of environmental contaminants has been investigated both at the tissue and cellular level in varying matrices. The ability of FTIR to identify compositional changes and classify cells based on the altered biochemical fingerprint caused by exposure to contaminants has been recognized. Previous studies have highlighted the utility of FTIR spectroscopy in environmental toxicological assessments.3−9 Cakmak et al.4 observed changes in the glycogen, © 2014 American Chemical Society
Received: Revised: Accepted: Published: 4581
January 6, 2014 March 14, 2014 March 20, 2014 March 20, 2014 dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
sources.13 Nonetheless, FTIR imaging instruments with globar sources have been used in the analysis of single cells and provide pertinent information on the cell’s biochemistry.14,15 Recently, Llabjani et al.6 investigated the response of MCF-7 cells to 17β-estradiol using FTIR (without imaging capability) on bulk cells and have shown that the cell spectra could be segregated based on the concentrations of 17β-estradiol to which the cells were exposed. In the current study, the ability of benchtop FT-IRIS using conventional sources to detect subcellular responses to 17β-estradiol in single MCF-7 cells was investigated, and the findings were compared to the physiological changes observed on the cellular population level. In this capacity, infrared spectroscopy probed the molecular changes that occurred in response to exposure to an emerging contaminant or environmental stressor, and thus provided information on the genotype-envirotype relationship. In addition to gaining further insight and confirmation of response mechanisms, the molecular level analysis may provide support for development of a more rapid and sensitive detection method for bioactive chemicals.
with mid and near IR spectral range. Hyperspectral images (consisting of a full spectrum in the Z direction for every X, Y pixel coordinate) were collected in imaging mode with transflection geometry at 8 cm−1 resolution, with 128 coadded scans per pixels, and a 6.25 μm pixel size. A spectrum obtained from the low-E slide was used to correct for the atmospheric and instrumental background signal. The images were imported into Isys 5.0 software (Malvern Instruments, Columbia, MD) for data processing. In total, 10 randomly selected cells were imaged for each treatment concentration at the 24 and 48 h time points, and 100 spectra were extracted from the cytoplasm of each 10 cells. Absorbances that arise from the lipid, protein (amide I and II absorbance), carbohydrate (mucin, glycogen), and nucleic acid components (PO2 absorbances) of the cells are the primary contributors to the cell spectra, Figure 1. The
■
MATERIALS AND METHODS E-Screen Assay. MCF-7 cells were routinely maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 2% 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer, 2% glutamine, 1% penn/strep, and 10% fetal bovine serum (FBS). Experimental medium contained phenol-red free DMEM supplemented with HEPES, glutamine, and penn/strep as above in addition to 10% charcoal stripped FBS. A stock solution of 17β-estradiol was prepared in ethanol and diluted in experimental medium such that the final ethanol concentration did not exceed 0.1% v/v. Cells were seeded into 6-well plates at an initial density of 25 000 cells/mL and allowed to attach. After 24 h, medium containing the test compound was added to the cells and the plates incubated for 6 days. Following incubation, the cells were trypsinized and counted using a TC Biorad automatic cell counter. The estrogenic activity was quantified as the proliferative effect (PE).16 This represents the ratio of the highest cell yield obtained with the test substance to the nontreated control.
Figure 1. Example of a FT-IRIS acquired spectrum showing the positions of relevant macro-molecules.
spectra were normalized to the Amide I peak. Using a polynomial baseline correction method, the spectra were baseline corrected from 748 cm−1 to 4000 cm−1. The image was spatially interpolated to reduce the pixelation of the image. The comparable sizes of the nucleus and the wavelength of the incident light result in strong scattering artifacts by single cells, manifested as an undulating baseline, and a shift to lower wavenumbers in the expected position of the amide I peaks. A Resonant Mie Scatter Correction algorithm, developed by Bassan et al.,17−20 was used to remove these distortions in cell spectra, thereby allowing an accurate absorption spectra of the cells to be revealed. Infrared spectra follow Beer−Lambert’s law for absorbance, and accordingly, the integrated area under the peaks correlate to the concentration of the biomolecules that give rise to the absorption. The 1140−1180 cm−1 spectral region, containing absorbances related to carbohydrates such as mucin (∼1150 cm−1) and serine, tyrosine or threonine amino acids (∼1168 cm−1), was baseline corrected and the integrated area under the peak was determined for each of the time points investigated. The integrated areas were then normalized to the average of the area for the control group and plotted as “fold increase in response”. Similarly, the cell numbers in the E-Screen assay were normalized to that of the control.
PE = cell yield with test chemical/cell yield control
■
FT-IRIS Cell Culture and Fixation. Cells were grown directly on low-E slides (Kevley Technologies, OH) at 1 × 105 cells/mL in experimental medium containing 17β-estradiol ranging from 0.27 ppt to 2.7 ppb or without any hormone (control). The control class was exposed to 0.1% v/v ethanol, similarly to the E-Screen method. After incubation for 24, 48, 72, 96, and 144 h the cells were rinsed in phosphate buffered saline (PBS) and then fixed in 4% formalin in PBS for 20 min. Once fixed, the slides were rinsed in PBS followed by a 3 s wash in water. Following a brief period of air-drying, the slides were stored in a desiccator until imaging. The 24 and 48 h time point represent the average of two independent experiments of five replicates each per treatment concentration. The other FT-IRIS time points represent triplicate measures. The E-Screen results are the average of eight replicates from four independent experiments. FT-IRIS Imaging. Cells were imaged using a Perkin-Elmer Spotlight 400 IR Imaging System (Perkin-Elmer, Shelton, CT) 4582
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
nucleus and a greater contribution of histone proteins.23 In the current study, these findings were used to segment the nucleus and cytoplasm. Figure 3 shows infrared spectral images of a
Principal Component Analysis. The infrared spectra of the cells contain vast amounts of information, including changes in peak intensities, shapes, and position, which may be too subtle to observe in the raw spectra and/or spectral derivatives.21 As such, multivariate data analysis methods were used to reduce the dimensionality and retain the variability of the FT-IRIS acquired spectra. This allowed further investigation into the spectral differences that result from exposing the cells to varying concentrations of 17ß-estradiol. The spectra were baseline corrected and normalized by peak height. Second derivative spectra were obtained from the cytoplasm of each cell and averaged prior to multivariate data analysis. A mean centered-PCA was performed in Unscrambler v10 (Camo, Norway).
■
RESULTS AND DISCUSSIONS Identification of Subcellular Regions. Cellular response mechanisms are very dynamic and vary between different parts of the cell. Once 17β-estradiol enters the cells, estrogen receptors become activated and dimerize. The hetero/homo dimers then bind with estrogen response elements and begin the transcription of estrogen related genes. The mRNA that is transcribed dictates the synthesis of proteins, which function to control the physiological cellular response.22 Among the physiological factors that are affected by 17β-estradiol are changes in metabolism and cell cycle regulation. The latter is used as a measure of the estrogenic response in MCF-7 cells and is the basis for the E-Screen assay which measures the increase in cell number as a function of estrogen concentration.16 Figure 2 shows a visual image of a single MCF-7 cell.
Figure 3. Images assembled according to the wavenumbers indicated below each panel. The regions with higher lipid intensity represent the cytoplasm.
single cell showing the distribution of the Amide I and lipid associated modes. When the images were created based on the lipid modes, (Figure 3 B, C, D) higher intensities were found closer to the periphery of the cell, and less intense regions occurred toward the cell center. However, the amide I mode was found to be most intense at the center of the cell, although it was also observed to be distributed throughout (Figure 3A). Some cells showed different patterns in distribution and were thought to be undergoing mitosis as indicated by the unique spectra and is described in more detail by Holman et al., 2000. 24,25 In these cases the lipids appeared to be homogenously distributed throughout the cells and were not included in the analysis. Other cells were thought to be in either S or G-phase. In all cells, regions that showed the most intense lipids were assigned as the cytoplasm. Spectra were then extracted from the cytoplasm and further analyzed. Biochemical Cellular Response to 17β-Estradiol. The MCF-7 response to 17ß-estradiol is complex, and there are several cascades of events that occur leading to the proliferative effect. This includes the stimulation and deactivation of genes and enzymes that are responsible for cell cycle regulation, proliferation and other metabolic processes. It was observed that the region with the most spectral differences between the cells treated with 17β-estradiol and nontreated cells was the 1300−1100 cm−1 range, Figure 4A. The major components that contribute in this area are phosphate (∼1240 cm−1), carbohydrates such as mucin (∼1150 cm−1) and serine, tyrosine, or threonine amino acids (∼1168 cm−1).26 The most consistent changes occurred in the carbohydrates, mucin, serine, tyrosine, and amino acid regions in the cytoplasm of the cells as shown in Figure 4B. Mucins also produce a peak at ∼1050 cm−1,27 however, other macromolecules (phosphodiesters of nucleic acids) also absorb in this region, making it difficult to interpret the changes that may occur. Although other carbohydrates such as glycogen may absorb in the 1150 cm−1 region, the assignment of mucin is more consistent with increased C−O and C−OH stretching modes, which are likely due to serine, threonine, and tyrosine in amino acids that occur at 1168 cm−1. Mucins are a family of highly glycosylated proteins which are known to be rich in serine and threonine and play a role in cell signaling as well as cell proliferation.28 Previous studies have shown that variants of mucin-related
Figure 2. Example of a visual image of a single cell. Visualization and delineation of the cytoplasm and nucleus is not immediately apparent.
It can be observed that after cellular fixation it is difficult to identify the nucleus in the MCF-7 cells; therefore, the known distribution of biomolecules was used to approximate these regions in the FT-IRIS acquired hyperspectral images. Previous studies using FTIR imaging techniques have indicated that the lipid concentration is higher in the cytoplasm due to the higher amounts of phospholipids in membranous organelles.23 Further, the absorbance of the amide I modes (attributed to proteins) are higher in the nucleus. This may be due to the larger path length of the IR beam through the 4583
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
IRIS (∼3.3 fold). However, the 48 h FT-IRIS and the 144hr EScreen yielded comparable EC50 (effective concentration at 50% of the response) values when the linear part of the response was regressed and interpolated at 50% of the maximum response (Figure 6). From Figure 6, the values of
Figure 6. Linear dose ranges for FT-IRIS and E-Screen results used for determination of EC50 values.
2.29 and 2.56 ppt for the E-Screen assay and the FT-IRIS, respectively, are within the acceptable EC50 values of 1.904 −3.264 ppt previously described.32 This indicates that the FTIRIS method not only reduces the time of the assay, but also maintains the assay sensitivity. Further, since the FT-IRIS method interrogates changes on the single cell level, the comparable sensitivities indicate that the molecular changes that are observed in the single cells using FT-IRIS are reflective of physiological changes that are observed as the cell population responds to 17ß-estradiol. At 72 h the FT-IRIS-determined responses appeared to be increasing at lower concentration of 17ß-estradiol. At higher incubation times, no observable responses to FT-IRIS could be detected among the concentration range tested. While this should not be overinterpreted due to the small sample size used at these time points, it may indicate that the potency or sensitivity of the assay is leveling off at higher incubation times. At these time points, it would be interesting to see whether or not the cellular response is higher at concentrations lower than those tested, which would indicate hormesis. Biospectroscopy has the remarkable ability to detect responses to low and environmentally relevant concentrations of contaminants as well as the nonlinear or hormetic responses that may be associated with them.33 Such nonlinear responses were first reported by Llabjani et el., where the ability of biospectroscopy tools to detect nonlinear responses to environmentally relevant concentrations of polybrominated diphenyl ethers (PBDEs) was demonstrated.34 This is of significance to eco-toxicology as emerging contaminants are typically at very low concentrations in the environment. Another likely possibility for the decline in single cell response at higher times, is that these cells reflect the portion of the cell population that have a lessened response to 17ß-estradiol, as indicated by their lack of cellular division after prolonged exposure to 17ß-estradiol. Principal Component Analysis. Differentiation among the cells exposed to different concentrations of 17ß-estradiol can also be observed in the scores plot shown in Figure 7. The cells appear to be discriminated by PC1 and to a lesser extent about the origin of PC 2. The unsupervised nature of the PCA means that it does not differentiate between the variability within a particular data set (intragroup), or the variability
Figure 4. A. Full representative spectra of cells treated with and without 17β-estradiol at 48 h, B. Concentration dependent responses in the mucin/amino acid region in response to varying concentrations of 17β-estradiol at 48 h.
genes may be overexpressed in MCF-7 cells in response to 17βestradiol.29−31 A similar response was also stimulated by a xenoestrogen, nonyl-phenol in MCF-7 cells.30 Figure 5 shows the fold increase in the FT-IRIS responses compared to the cellular proliferation in the E-Screen as a
Figure 5. Comparison of E-Screen and FT-IRIS assays at 24, 48, 72, 96, and 144 h.
function of time. At 24 h, a concentration dependent response was observed. This response further increased at 48 h of exposure to 17ß-estradiol. The level of response for the 48 h treatment was comparable to the increase in proliferation as determined by the E-Screen assay although the responsiveness of the E-Screen assay (∼ 5 fold) was greater than that of FT4584
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
PCA. The second derivative spectra provide additional information to the original spectra and as such it highlights many differences between the different cell groups. However, the mucin peak appeared to increase in width as well as height. Since peak height in the second derivative is inversely proportional to the square of the original half width,38 the concurrent increases in peak height and width as a function of treatment may cancel each other in the second derivative spectra. This may have resulted in reductions in the sensitivity of the PCA toward the mucin contribution. However, concentration dependent shifts in the positions of the Amide I and II peaks could be observed in the second derivative spectra. These are suggestive of 17ß-estradiol’s effect on the cell cycle, although without initial cell synchronization it would be difficult to draw a definitive conclusion. At this point it is difficult to speculate on the extent to which biomolecules contribute to the intra- and intergroup variation. PCA in combination with supervised linear discriminant analysis (LDA) could be used to overcome this limitation; however, the number of spectra must be sufficiently large enough to prevent overfitting of the data.7,39 Platforms and MATLAB toolboxes such as those developed by Trevisan et al., could be useful for such analysis.40 Nonetheless, the ability of FT-IRIS combined with PCA to distinguish cells exposed to different amounts of 17ß-estradiol is evident. This study details the steps taken to investigate the ability of fourier transform infrared spectroscopy (FT-IRIS) to detect a dose dependent response to 17ß-estradiol. The cytoplasmic region of the cell was identified and spectra of cells treated with 17ß-estradiol showed dose dependent differences in the mucin concentrations. The FT-IRIS response may occur through the estrogen receptor, or they may be reflective of nongenomic responses due to the 17ß-estradiol, and further experiments are necessary to determine the specificity of the response. The FTIRIS response is comparable to, and correlates well, with results obtained by the E-Screen assay in one-third the time (48hrs versus 6 days) of the E-Screen assay, indicating that FT-IRIS of biological cells and tissues could be used to rapidly screen for the biological effects of environmental contaminants. Furthermore, the sensitivity of the assay is maintained using FTIRIS analysis. This has relevance to the analysis of environmental samples, which may be toxic or inhibitory upon enrichment. The sensitivity means that less sample enrichment will be necessary to detect a response and as such inhibitory effects may be reduced. The E-Screen assay also offers such sensitivity; however, the FT-IRIS has the added advantage of a shorter assay time.
Figure 7. Scores plot of the first two principal components on the FTIRIS acquired spectra of cells exposed to different concentrations of 17ß-estradiol. Each point represents an average of 100 spectra, extracted from the cytoplasm of cells.
between the groups (intergroup).35 In Figure 7, it can be seen that there is a fair amount of intragroup variation, although the intergroup variation is sufficiently large enough to result in discrimination among the groups. As such, the loadings for PC 1 includes information on the spectral variation among the different groups (or treatments), in addition to some spectral differences that are more consistent with the inherent variability of the analysis of single cells by FT-IRIS. The RMieS Correction algorithm successfully removed the reflection contribution and reduced the undulating baseline; in addition, the position of the amide I peak shifted to higher wave numbers. However, the degree of scattering varied from cell to cell and some cells tended to scatter more, possibly due to varying morphologies and geometries due to differences in the phase of the cell cycle. Cell synchronization could be used to reduce the intragroup variability as biochemical changes relating to cell cycle manifest in FT-IRIS analysis.24,36,37 The loading plot shown in Figure 8 indicates that the variability among the spectra could be attributed to a combination of contributions from the proteins (1650 cm−1, 1560 cm−1, 1460 cm−1, 1460 cm−1, 1172 cm−1), with some contribution from nucleic acids (∼1260 cm−1).26 It is important to point out that the mucin contribution (∼1150 cm−1) was not greatly observed in the
■
AUTHOR INFORMATION
Corresponding Author
*Phone: (215) 204-2376; fax: (215) 204-0622; e-mail: rsuri@ temple.edu. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was partly supported by the National Science Foundation (NSF) Water and Environmental Technology (WET) Center and Temple University. The opinions expressed in this article are those of the authors and do not necessarily reflect the views of the WET Center, NSF or Temple
Figure 8. Plot of loadings of PC1 that shows the spectral regions with the greatest variation among the cells. 4585
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
estrogensAn update on estrogenic environmental-pollutants. Environ. Health Perspect. 1995, 103, 113−122. (17) Bassan, P.; Byrne, H. J.; Lee, J.; Bonnier, F.; Clarke, C.; Dumas, P.; Gazi, E.; Brown, M. D.; Clarke, N. W.; Gardner, P. Reflection contributions to the dispersion artefact in FTIR spectra of single biological cells. Analyst 2009, 134 (6), 1171−1175. (18) Bassan, P.; Byrne, H. J.; Bonnier, F.; Lee, J.; Dumas, P.; Gardner, P. Resonant Mie scattering in infrared spectroscopy of biological materialsUnderstanding the ‘dispersion artefact’. Analyst 2009, 134 (8), 1586−1593. (19) Bassan, P.; Kohler, A.; Martens, H.; Lee, J.; Byrne, H. J.; Dumas, P.; Gazi, E.; Brown, M.; Clarke, N.; Gardner, P. Resonant Mie Scattering (RMieS) correction of infrared spectra from highly scattering biological samples. Analyst 2010, 135 (2), 268−277. (20) Bassan, P.; Kohler, A.; Martens, H.; Lee, J.; Jackson, E.; Lockyer, N.; Dumas, P.; Brown, M.; Clarke, N.; Gardner, P. RMieS-EMSC correction for infrared spectra of biological cells: Extension using full Mie theory and GPU computing. J. Biophotonics 2010, 3 (8−9), 609− 620. (21) Diem, M.; Papamarkakis, K.; Schubert, J.; Bird, B.; Romeo, M. J.; Miljkovic, M. The infrared spectral signatures of disease: Extracting the distinguishing spectral features between normal and diseased states. Appl. Spectrosc. 2009, 63 (11), 307A−318A. (22) Heldring, N.; Pike, A.; Andersson, S.; Matthews, J.; Cheng, G.; Hartman, J.; Tujague, M.; Strom, A.; Treuter, E.; Warner, M.; Gustafsson, J. A. Estrogen receptors: How do they signal and what are their targets. Physiol. Rev. 2007, 87 (3), 905−931. (23) Lasch, P.; Pacifico, A.; Diem, M. Spatially resolved IR microspectroscopy of single cells. Biopolymers 2002, 67 (4−5), 335− 338. (24) Holman, H. N.; Martin, M. C.; Blakely, E. A.; Bjornstad, K.; McKinney, W. R. IR spectroscopic characteristics of cell cycle and cell death probed by synchrotron radiation based fourier transform IR spectromicroscopy. Biopolymers 2000, 57 (6), 329−335. (25) Hammiche, A.; German, M. J.; Hewitt, R.; Pollock, H. M.; Martin, F. L. Monitoring cell cycle distributions in MCF-7 cells using near-field photothermal microspectroscopy. Biophys. J. 2005, 88 (5), 3699−3706. (26) Movasaghi, Z.; Rehman, S.; Rehman, I. U. Fourier transform infrared (FTIR) spectroscopy of biological tissues. Appl. Spectrosc. Rev. 2008, 43 (2), 134−179. (27) Lasch, P.; Boese, M.; Pacifico, A.; Diem, M. FT-IR spectroscopic investigations of single cells on the subcellular level. Vib. Spectrosc. 2002, 28 (1), 147−157. (28) Devine, P. L.; McKenzie, I. F. C. MucinsStructure, function, and associations with malignancy. Bioessays 1992, 14 (9), 619−625. (29) Porowska, H.; Paszkiewicz-Gadek, A.; Wolczynski, S.; Gindzienski, A. MUC1 expression in human breast cancer cells is altered by the factors affecting cell proliferation. Neoplasma 2002, 49 (2), 104−109. (30) Ren, L. F.; Marquardt, M. A.; Lech, J. J. Estrogenic effects of nonylphenol on pS2, ER and MUC1 gene expression in human breast cancer cells-MCF-7. Chem.-Biol. Interact. 1997, 104 (1), 55−64. (31) Zaretsky, J. Z.; Barnea, I.; Aylon, Y.; Gorivodsky, M.; Wreschner, D. H.; Keydar, I., MUCI gene overexpressed in breast cancer: Structure and transcriptional activity of the MUC1 promoter and role of estrogen receptor alpha (ER alpha) in regulation of the MUC1 gene expression. Mol. Cancer 2006, 5. (32) Soto, A. M.; Maffini, M. V.; Schaeberle, C. M.; Sonnenschein, C. Strengths and weaknesses of in vitro assays for estrogenic and androgenic activity. Best Pract. Res., Clin. Endocrinol. Metab. 2006, 20 (1), 15−33. (33) Ukpebor, J.; Llabjani, V.; Martin, F. L.; Halsall, C. J. Sublethal genotoxicity and cell alterations by organophosphorus pesticides in MCF-7 cells: Implications for environmentally relevant concentrations. Environ. Toxicol. Chem. 2011, 30 (3), 632−639. (34) Llabjani, V.; Trevisan, J.; Jones, K. C.; Shore, R. F.; Martin, F. L. Derivation by infrared spectroscopy with multivariate analysis of
University. The authors would like to thank Dr. Ana Soto for providing the MCF-7 cells.
■
REFERENCES
(1) Johnson, C. M.; Achary, M.; Suri, R. P. An interaction model for estimating in vitro estrogenic and androgenic activity of chemical mixtures. Environ. Sci. Technol. 2013, 47 (9), 4661−4669. (2) Streck, G. Chemical and biological analysis of estrogenic, progestagenic and androgenic steroids in the environment. Trac, Trends Anal. Chem. 2009, 28 (6), 635−652. (3) Cakmak, G.; Togan, I.; Uguz, C.; Severcan, F. FT-IR spectroscopic analysis of rainbow trout liver exposed to nonylphenol. Appl. Spectrosc. 2003, 57 (7), 835−841. (4) Cakmak, G.; Togan, I.; Severcan, F. 17 beta-estradiol induced compositional, structural and functional changes in rainbow trout liver, revealed by FT-IR spectroscopy: A comparative study with nonylphenol. Aquat. Toxicol. 2006, 77 (1), 53−63. (5) Holman, H. Y. N.; Goth-Goldstein, R.; Martin, M. C.; Russell, M. L.; McKinney, W. R. Low-dose responses to 2,3,7,8-tetrachlorodibenzo-p-dioxin in single living human cells measured by synchrotron infrared spectromicroscopy. Environ. Sci. Technol. 2000, 34 (12), 2513−2517. (6) Llabjani, V.; Jones, K. C.; Thomas, G. O.; Walker, L. A.; Shore, R. F.; Martin, F. L. Polybrominated diphenyl ether-associated alterations in cell biochemistry as determined by attenuated total reflection fourier-transform infrared spectroscopy: A comparison with DNAreactive and/or endocrine-disrupting agents. Environ. Sci. Technol. 2009, 43 (9), 3356−3364. (7) Llabjani, V.; Trevisan, J.; Martin, F. L. Application of infrared spectroscopy to biochemically signature the effects of chemical mixtures in mammalian cells. Mutagenesis 2010, 25 (6), 70. (8) Llabjani, V.; Crosse, J. D.; Ahmadzai, A. A.; Patel, I. I.; Pang, W. Y.; Trevisan, J.; Jones, K. C.; Shore, R. F.; Martin, F. L. Differential effects in mammalian cells induced by chemical mixtures in environmental biota as profiled using infrared spectroscopy. Environ. Sci. Technol. 2011, 45 (24), 10706−10712. (9) Bl, X. H.; Walsh, M. J.; Wei, X. M.; Sheng, G. Y.; Fu, J. M.; Martin-Hirsch, P. L.; Thomas, G. O.; Jones, K. C.; Martin, F. L. Infrared spectral analysis of MCF-7 cells treated with serum-lipid extracts segregates predominantly brominated flame retardant-exposed subjects from those with mainly organochlorine exposures. Environ. Sci. Technol. 2007, 41 (16), 5915−5922. (10) Kallenbach-Thieltges, A.; Grosseruschkamp, F.; Mosig, A.; Diem, M.; Tannapfel, A.; Gerwert, K. Immunohistochemistry, histopathology and infrared spectral histopathology of colon cancer tissue sections. J. Biophotonics 2013, 6 (1), 88−100. (11) Benard, A.; Desmedt, C.; Smolina, M.; Szternfeld, P.; Verdonck, M.; Rouas, G.; Kheddoumi, N.; Rothe, F.; Larsimont, D.; Sotiriou, C.; Goormaghtigh, E. Infrared imaging in breast cancer: Automated tissue component recognition and spectral characterization of breast cancer cells as well as the tumor microenvironment. Analyst 2014, 139 (5), 1044−1056. (12) Diem, M.; Romeo, M.; Matthaus, C.; Miljkovic, M.; Miller, L.; Lasch, P. Comparison of Fourier transform infrared (FTIR) spectra of individual cells acquired using synchrotron and conventional sources. Infrared Phys. Technol. 2004, 45 (5−6), 331−338. (13) Petibois, C.; Cestelli-Guidi, M.; Piccinini, M.; Moenner, M.; Marcelli, A. Synchrotron radiation FTIR imaging in minutes: A first step towards real-time cell imaging. Anal. Bioanal. Chem. 2010, 397 (6), 2123−2129. (14) Boydston-White, S.; Chernenko, T.; Regina, A.; Miljkovic, M.; Matthaus, C.; Diem, M. Microspectroscopy of single proliferating HeLa cells. Vib. Spectrosc. 2005, 38 (1−2), 169−177. (15) Marcsisin, E. J.; Uttero, C. M.; Miljkovic, M.; Diem, M. Infrared microspectroscopy of live cells in aqueous media. Analyst 2010, 135 (12), 3227−3232. (16) Soto, A. M.; Sonnenschein, C.; Chung, K. L.; Fernandez, M. F.; Olea, N.; Serrano, F. O. The E-Screen assay as a tool to identify 4586
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
Environmental Science & Technology
Article
bimodal contaminant-induced dose-response effects in MCF-7 cells. Environ. Sci. Technol. 2011, 45 (14), 6129−6135. (35) Bonnier, F.; Byrne, H. J. Understanding the molecular information contained in principal component analysis of vibrational spectra of biological systems. Analyst 2012, 137 (2), 322−332. (36) Boydston-White, S.; Romeo, M.; Chernenko, T.; Regina, A.; Miljkovic, M.; Diem, M. Cell-cycle-dependent variations in FTIR micro-spectra of single proliferating HeLa cells: Principal component and artificial neural network analysis. Biochim. Biophys. Acta 2006, 1758 (7), 908−914. (37) Flower, K. R.; Khalifa, I.; Bassan, P.; Demoulin, D.; Jackson, E.; Lockyer, N. P.; McGown, A. T.; Miles, P.; Vaccari, L.; Gardner, P. Synchrotron FTIR analysis of drug treated ovarian A2780 cells: An ability to differentiate cell response to different drugs? Analyst 2011, 136 (3), 498−507. (38) Susi, H.; Byler, D. M. Resolution-enhanced fourier-transform infrared-spectroscopy of enzymes. Methods Enzymol. 1986, 130, 290− 311. (39) Trevisan, J.; Angelov, P. P.; Carmichael, P. L.; Scott, A. D.; Martin, F. L. Extracting biological information with computational analysis of Fourier-transform infrared (FTIR) biospectroscopy datasets: Current practices to future perspectives. Analyst 2012, 137 (14), 3202−3215. (40) Trevisan, J.; Angelov, P. P.; Scott, A. D.; Carmichael, P. L.; Martin, F. L. IRootLab: A free and open-source MATLAB toolbox for vibrational biospectroscopy data analysis. Bioinformatics 2013, 29 (8), 1095−1097.
4587
dx.doi.org/10.1021/es5000676 | Environ. Sci. Technol. 2014, 48, 4581−4587
