Derivation by Infrared Spectroscopy with Multivariate Analysis of

ARTICLE pubs.acs.org/est

Derivation by Infrared Spectroscopy with Multivariate Analysis of Bimodal Contaminant-Induced Dose-Response Effects in MCF-7 Cells Valon Llabjani,†,‡ J
ulio Trevisan,† Kevin C. Jones,† Richard F. Shore,‡ and Francis L. Martin*,† † ‡

Centre for Biophotonics, Lancaster Environment Centre, Lancaster University, Bailrigg, Lancaster LA1 4YQ, U.K. NERC Centre for Ecology and Hydrology, Lancaster Environment Centre, Library Avenue, Bailrigg, Lancaster LA1 4AP, U.K.

bS Supporting Information ABSTRACT: Toxic responses to contaminants following exposure concentrations typically used in laboratory tests may not reflect how biological systems respond to lower environmental levels from which hormetic effect mechanisms have been suggested. We investigated the pattern of dose-response in mammalian cells to various environmental contaminants using a range of concentrations that span those that are environmentally relevant (10
12 M to 10
3 M). MCF-7 cell cultures were treated for 24 h with benzo[a]pyrene (B[a]P), lindane (γ-hexachlorocyclohexane), or polybrominated diphenyl ethers (PBDEs) congeners (47, 153, 183, and 209), then fixed in ethanol and interrogated using attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy. Mode of action was further studied by examining if test agents stimulated cell growth or altered CYP1A1 expression. Bimodal dose response curves were observed when MCF-7 cells were treated with PBDEs or lindane. The first peak distribution was associated with lower doses (10
12 M to 10
9 M), while the second occurred only after MCF-7 cells were exposed to concentrations >10
9 M. Cellular alterations associated with low-dose PBDEs were mainly due to lipid and secondary protein structural changes, whereas lindane induced DNA/RNA effects as well. In contrast, DNAreactive B[a]P gave rise to a monotonic linear dose-response relationship and induced mainly DNA/RNA cellular changes. This study shows that environmentally realistic exposures to chemical contaminants can induce nonmonotonic dose-responses in cellular systems.

’ INTRODUCTION Many chemical contaminants that are released into the environment are assimilated by biota and may induce adverse effects at high concentrations. Polybrominated diphenyl ethers (PBDEs) leach into the environment, are bioaccumulated, and have been found to induce a range of dose-related health effects in vitro or in vivo.1
3 The relationship between chemical dose and biological response traditionally falls into one of two categories. These are either linear dose-response caused by carcinogenic compounds where the biological effect increases with increasing concentration, or a threshold response for noncarcinogens where an agent needs to reach a certain concentration before it induces its effects [see Supporting Information (SI) Figure S1 A and B]. Laboratory tests are generally performed using high chemical concentrations and may not necessarily reflect how biological systems respond to lower environmental exposures.4 For instance, cellular responses at high dose may not be predictive of low-dose effects because there may be a “hormetic dose-response”,5 hormesis being low-dose stimulation followed by high-dose inhibition of response.6 The most common forms of chemical hormesis include β-curve or U-shaped curves (see SI Figure S1 C and D).7 The β-curve is characterized by low-dose stimulation and high-dose inhibition of response, whereas a U-shaped dose-response curve describes reduction and enhancement of adverse effect at low and r 2011 American Chemical Society

high doses, respectively.8 Organisms may counteract and neutralize chemical stressors through homeodynamic systems, by reaching an equilibrium between agent-induced toxicity and protective counteractive responses.9 In hormesis, stimulation of response may be associated with some form of overcorrection by the homeodynamic system in response to low-level exposure whereas an inhibitory response may occur when the toxicant concentration exceeds the maximum counter-response. Infrared (IR) spectroscopy is a powerful technique that allows one to generate a chemical signature based on structure and function in different biological systems.10 Attenuated total reflection Fourier-transform IR (ATR-FTIR) spectroscopy can distinguish single- and mixture-induced effects in cells resulting from contaminant exposures ranging from pM to μM concentration.11,12 IR spectroscopy may be applied in this fashion to interrogate the chemical structure of biomolecules because different chemical bonds absorb light at different wavelengths in the mid-IR region (λ = 2.5 μm
25 μm; 4000 cm
1
400 cm
1), and this generates a vibrational spectrum consisting of wavenumberabsorbance intensities often referred to as a “biochemical-cell Received: February 1, 2011 Accepted: May 31, 2011 Revised: May 27, 2011 Published: June 23, 2011 6129

dx.doi.org/10.1021/es200383a | Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology

ARTICLE

Figure 1. Dose-response effects in MCF-7 cells following treatment with brominated diphenyl ether (BDE) congeners (47, 153, 183, or 209). Left panels: LD1 scores plot following LDA of mixed-cell population (S- and G0/G1-phase) and fitted cubic splines using averages of all infrared (IR) spectra derived from triplicate experiments per cell cycle phase (n = 60; 10 IR spectra per slide  S- and G0/G1-phase cells  triplicate experiments). Right panels: Mean ( standard deviation IR absorbance ratio of protein (1400 cm
1) to lipid (1740 cm
1) of S-phase MCF-7 cells. ANOVA with Dunnett’s post hoc test was used to test significance between contaminant-exposed vs control cell populations (*, Pe0.05; **, Pe0.01). con, vehicle control; Conc, concentration.

fingerprint”.13 Toward handling the complex data sets that are generally obtained from this approach, IR spectra can thus be considered in a multidimensional space and the dimensionality is best reduced using multivariate analysis techniques [principal component analysis (PCA) and linear discriminant analysis (LDA)] that capture the most important variation. These techniques, either individually or combined, provide loadings vectors that can identify important wavenumbers, which can be related to chemical bonds of known vibrational frequency.

The aim of this study was to characterize in vitro doseresponses associated with exposure at realistic concentrations to a range of widely occurring environmental contaminants. We used estrogen-responsive MCF-7 cells as a model system and characterized responses to exposure using ATR-FTIR spectroscopy with multivariate analysis. We treated MCF-7 cells with benzo[a]pyrene (B[a]P), lindane (γ-hexachlorocyclohexane), or polybrominated diphenyl ethers (PBDEs)—specifically brominated diphenyl ether (BDE) congeners 47, 153, 183, and 209—at 6130

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology

ARTICLE

Figure 2. Dose-response effects in MCF-7 cells following treatment with lindane or B[a]P. Left panels: LD1 scores plot from LDA of mixed-cell population (S- and G0/G1-phase) and fitted cubic splines using averages of all infrared (IR) spectra derived from triplicate experiments per cell cycle phase (n = 60; 10 IR spectra per slide  S- and G0/G1-phase cells  triplicate experiments). Right panels: Mean ( standard deviation IR absorbance ratio of protein (1400 cm
1) to lipid (1740 cm
1) of S-phase MCF-7 cells. ANOVA with Dunnett’s post hoc test was used to test significance between contaminant-exposed vs control cell populations (**, Pe0.01). con, vehicle control; Conc, concentration.

doses ranging from 10
12 M to 10
3 M, and derived spectroscopy-based dose-response curves. We also discerned biochemical alterations associated with each chemical from IR absorbance spectra and examined if they stimulated cell growth by calculating protein-to-lipid ratio of treated cells compared to control cell populations. Finally, given that BDE47, the most abundant BDE congener in the environment,2,14 is thought to mediate dioxin-like action on AhR receptor,15 we examined whether it affects AhR activity in a dose-related manner by measuring the protein expression of CYP1A1.

’ EXPERIMENTAL SECTION Cell Culture. The human MCF-7 cell line was grown in Dulbecco’s modified essential medium supplemented with 10% heat-inactivated fetal calf serum (FCS), penicillin (100 U/mL) and streptomycin (100 μg/mL). Cells were grown in 5% CO2 in air at 37°C in a humidified atmosphere and were disaggregated with trypsin (0.05%)/EDTA (0.02%) solution before incorporation into experiments. Cell culture consumables were obtained from Invitrogen Life Technologies (Paisley, UK), unless otherwise stated. MCF-7 cells were used because they are metabolically active and express CYP isoforms;16 also, being hormone responsive they would likely be a representative model to determine target-cell susceptibility to endocrine-active contaminants. Test Agents. Stock solutions of PBDE standards [>99% pure, predissolved in nonane at 50 μg/mL (50 ppm; Cambridge Isotope Laboratories, Inc., Andover, MA)] and, B[a]P and lindane (Sigma Chemical Co., Poole, UK) were made up using dimethylsulfoxide (DMSO) as solvent. Serial dilutions in DMSO were added to cell incubates. The maximum DMSO concentration per incubate was 1% (v/v). Negative controls consisted of

the same amount of DMSO as in treatment regimens, with PBDE negative controls also containing nonane (purity >95%; Promochem, Welwyn Garden City, Hertfordshire, UK) that matched the concentration present in the BDE test agent. Solutions were analytically tested throughout the experimental period; no fluctuation or degradation in test-agent concentration was observed. Cell Treatments and ATR-FTIR Spectroscopy. Cultures of MCF-7 cells were disaggregated, resuspended in complete medium, and then seeded in T25 flasks, whereupon they were either concentrated in S-phase (after 24-h incubation) or G0/G1phase (after 96 h) prior to treatment with or without test agent for a further 24 h. The range of test-agent concentrations was 10
12 M to 10
7 M for BDE congeners, 10
12 M to 10
3 M for lindane, and 0.1 μM to 10.0 μM for B[a]P. Following treatment, cells were disaggregated and the cell suspensions were immediately fixed with 70% ethanol and stored at 4°C until analyzed. Cellular material was then applied to IR-reflective “Low E” glass slides (Kevley Technologies, Chesterland, OH), allowed to airdry, and then desiccated. IR spectra were obtained using a Bruker Vector 27 FTIR spectrometer with Helios ATR attachment containing a diamond crystal (Bruker Optics Ltd., Coventry, UK). For each experimental condition (i.e., each slide), 10 spectra were acquired. The ATR crystal was cleaned with sodium dodecyl sulfate (SDS; Sigma Chemical Co.) prior to each new sample being analyzed, and a new background spectrum was collected prior to analysis of a new sample. Triplicate experiments were conducted for each chemical at three different times over a 9-month period. Spectral Processing. Raw IR spectra obtained from interrogated samples were preprocessed prior to computational analysis. Spectra were cut at the “biochemical-cell fingerprint” region (1800 cm
1 to 900 cm
1), baseline corrected using OPUS rubber band correction software and normalized to 1650 cm
1. 6131

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology Spectra obtained from S- or G0/G1-phase treated cells were then combined to form a mixed-cell population per test agent. Computational analysis was performed on a data set that contained 60 IR spectra per treatment (10 spectra per slide  S- and G0/G1phase cells  triplicate experiments). Agent effects on cell stimulation (protein-to-lipid ratio) were examined only using data from S-phase cells. Linear Discriminant Analysis (LDA). LDA is a supervised linear transformation that projects the wavenumbers into a variable-reduced space which is optimal for discrimination between treatment classes. LDA seeks directions in the original data set (wavenumber variables) where the ratio of the betweenclass (treatments) to within-class (replicates) variance is maximal. The variables generated through LDA (factors) are linear combinations of the wavenumber-absorbance intensity values.17,18 Each factor has an associated loadings vector. Each loadings vector contains the weights of the wavenumbers forming its corresponding factor. Loadings vectors provide an interpretation of which wavenumbers are responsible for discrimination between treatment concentrations. LDA was carried out using in-house software (http://biophotonics.lancs.ac.uk/software/) developed in MATLAB (The MathWorks, Inc., USA). LDA scores and loadings plots were derived for the biochemical-cell fingerprint region (1800 cm
1 to 900 cm
1). After LDA, the numbers of variables was reduced from 235 (3.85 cm
1 wavenumber spacing) to a few (3
7 LDs depending on number of treatments). For this study, the first LDA factor (LD1) was used to visualize the cellular responses to the test agents in 1-dimensional (D) scores plots that represented the main chemical alterations. In these scores plots, nearness of a treatment point to the spectral points of control cells implies chemical structure similarity, and distance suggests dissimilarity. We calculated the mean for each concentration (including control) in the LD1 space, and then derived doseresponse curves where the x-axis was formed by the concentrations and the y-axis represented the distance between each concentration mean and the control mean in the LD1 space. This is a 1-D version of a previous modified “cluster vector approach”,17 applied to examine binary-mixture effects12. LD1 was selected because it is the maximally discriminating LDA factor. For visualization purposes (Figures 1 and 2), a line of best fit was interpolated through the mean distances from the control cluster in the LD1 space. Interpolation was carried out using cubic splines. Test Agent Effect of Cell Growth Stimulation. Processed IR spectra (baseline corrected and normalized) resulting from treated MCF-7 cells were used to calculate cell proliferation rate as a marker for cell stimulation. Cell proliferation was calculated by measuring the protein-to-lipid ratio using the intensity absorbance at two vibration modes. The intensity absorbance at 1400 cm
1 was used as a protein marker as it corresponds to CdO symmetric stretching of amino acids and 1740 cm
1 is associated with CdO stretching vibrations of lipids. Repeated-measures one-way analysis of variance (ANOVA) with Dunnett’s post hoc tests were used to examine whether the contaminant induced effects observed in LD1, and if protein/lipid absorbance intensity ratios differed significantly between treated and control cell populations; there was no need to transform data to meet the underlying assumptions of homogeneity of variance between groups and normality of residuals. When there were only two groups to compare, differences between the groups were tested by a Student’s t-test. Western Blotting. Following 24-h treatment of MCF-7 cells with BDE47 concentrations (10
12 M to 10
7 M) or positive

ARTICLE

Figure 3. (A) Mean ( standard deviation (SD) of the infrared (IR) absorbance ratio of protein (1400 cm
1) to lipid (1740 cm
1) in S-phase vs G0/G1-phase MCF-7 cells. Student’s t-test was used to test if ratios in S-phase were significantly different compared to G0/G1-phase cells (***, Pe0.001). (B) Western blot analysis of CYP1A1 expression in MCF-7 cells treated with BDE47. β-Actin was analyzed to confirm sample integrity and gel loading. B[a]P, benzo[a]pyrene; Con, control.

control B[a]P (5 μM), cells were detached from the culture flask, washed with PBS, and resuspended in 100 μL of lysis buffer [50 mM Tris/HCl (pH 8), 0.1 mM EDTA, 0.5% Triton X-100 and 12.5 mM 2-mercaptoethanol] (Sigma Chemical Co.). Cells were then centrifuged at 10 000 rpm for 15 min and the pellet was removed. Sample buffer [final concentration 6% glycerol, 0.83% β-mercaptoethanol, 1.71% Tris-HCl (pH 6.8), and 0.002% bromophenol blue] was added to the supernatant. SDS-PAGE was performed using 12% bis-acrylamide gel at room temperature and for each well 20 μg of total protein was loaded. Samples were resolved by electrophoresis at 125 V for 1 h and then transferred onto nitrocellulose membrane (Schleicher & Schuell Bioscience, Keene, NH). The blot was blocked with 5% (w/v) milk powder in TBST (Tris-buffered saline Tween-20) for 1 h and washed in TBST. The membrane was then incubated for 1 h with anti-CYP1A1 (Abcam, Cambridge, UK; AB79819), and incubated for 1 h with appropriate secondary antibodies. Protein bands were detected using ECL Detection Reagent (Amersham BioScience) before film exposure.

’ RESULTS AND DISCUSSION When MCF-7 cells were treated with BDE congeners, their response was nonlinear and varied depending on the concentration tested (Figure 1). PBDEs gave rise to distinct biphasic dose responses as distinguished by marked cell alterations compared to control treatments, and ANOVA showed that these were significant (Pe0.01) for most of the treatments. The first occurred at low doses (typically 10
12 M to 10
10 M), and the second occurred at higher concentrations (typically >10
9 M to 10
8 M). Thus, the mode of action for PBDEs on the cell cultures appeared to change at approximately 10
9 M to 10
8 M exposure. This dose-response pattern was also apparent and significant (Pe0.01) at most exposures when MCF-7 cells were treated with lindane except following 10
5 M treatment, although the transition between different modes occurred at 6132

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology

ARTICLE

Figure 4. LD1 loadings plots following LDA indicating wavenumber-associated molecular alterations responsible for segregation following treatment of MCF-7 cells with contaminant, as indicated. Represents all treatments combined compared to corresponding control.

higher (10
6 M) concentrations (Figure 2). In contrast, treatment of MCF-7 cells with DNA-reactive B[a]P elicited a response where the number of cellular alterations relative to the control increased in a dose-related manner (Pe0.01 at all concentrations) in a more classic linear dose-response relationship, up to a concentration of 5 μM (Figure 2). Increasing the concentration of B[a]P further reduced the number of cell alterations detected in the absorbance spectrum; this may have been due to cytotoxicity mechanisms setting in. An elevated protein-to-lipid ratio has been previously observed in cells undergoing active proliferation.19 Exponentially growing MCF-7 cells (S-phase) exhibited a significantly higher ratio of protein (1400 cm
1) to lipid (1740 cm
1) compared to G0/G1-phase cells (Pe0.001; t = 6.7; df = 26) (Figure 3A). The protein-to-lipid ratio was also significantly elevated in S-phase cell cultures treated with low-dose BDE47 or BDE183 (10
12 M or 10
11 M), indicating a stimulation of cell proliferation (Figure 1). The protein-to-lipid ratio in cells treated with all concentrations of BDE153 was also elevated compared with controls, although only significantly at concentrations of 10
12 M and 10
9 M (Figure 1). In contrast, cell proliferation seemed to decrease with increasing concentrations of BDE209, although the protein-to-lipid ratio was only significantly lower than in control cultures at 10
8 M (Figure 1). There appeared to be no

significant effect of lindane or B[a]P on cell proliferation with any of the concentrations tested (Figure 2). Treatment of MCF7 cells with BDE47 resulted in a concentration-related increase in CYP1A1 protein expression (Figure 3B). Low-dose BDE47 (i.e., 10
12 M) produced a small increase in CYP1A1 expression whereas high-dose BDE47 treatment (10
7 M) resulted in 2- to 3-fold higher levels of protein compared to vehicle control (Figure 3B). Examination of the loadings plots following LDA (Figure 4) indicated that the main biochemical alterations induced by BDE47 were on lipids (1705 cm
1 to 1750 cm
1), proteins [Amide III (∼1250 cm
1)], and carbohydrates (∼1150 cm
1). A similar response was also evident in BDE153-treated cells with alterations largely in the regions associated with lipids, proteins [Amide I (∼1650 cm
1), Amide II (∼1550 cm
1), and at ∼1400 cm
1 (CdO stretching of amino acids)], and to a lesser extent at carbohydrate (∼1140 cm
1) and DNA vibrations (∼1080 cm
1). BDE183 and BDE209 also induced changes in the protein region (Amide I, Amide III), glycogen (∼1057 cm
1), and carbohydrate (∼1155 cm
1). The mode of action of B[a]P and lindane contrasted with those of the BDE congeners. Lindane induced alterations in the lipid and protein regions but the most pronounced effects were on the DNA/RNA region. B[a]P also mainly affected the DNA/RNA region (∼1225 cm
1, 6133

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology ∼1080 cm
1) with relatively minor changes in the protein (Amide I; ∼1600 cm
1) region. Our findings show that PBDEs and lindane are capable of inducing biphasic responses at different dose-ranges in MCF-7 cells. Analysis of cell proliferation suggests that the first-phase response observed with the tetra- to hepta-BDE congeners (BDEs 47, 153, 183) at low-dose may be a result of cell stimulation induced by low-dose treatments (Figure 1). In contrast, the second phase response was not a result of cell stimulation; rather, this appeared to be associated with agent-induced toxic effects in MCF-7 cells, as when cells were treated with higher concentrations of B[a]P (Figure 2). Our results are consistent with those of other studies that have shown that exposure to BDE47 induced a nonlinear dose response, 20,21 and a natural mixture of persistent organic pollutants induced a nonmonotonic dose response in cell cultures.22 Cells have adaptive mechanisms that respond to a range of contaminants (or inhibitors) to counteract their action, and over-reaction can lead to a biphasic hormetic dose response.23 Overcorrected responses may occur following low-dose exposure and persist until there is a balance between the inhibitor action and cell response.9,24 Measuring chemical contaminant-induced effects at levels one might describe as low dose (i.e., environmental concentrations) has remained an elusive challenge to date. This study suggests that IR spectroscopy with computational analysis can be applied as a novel approach to investigate chemical dose-response relationships in mammalian cells as well as to obtain biomarkers that may be related with different test agents.25 Chemical contaminants may produce varied responses in the same cellular system, one activated at lower levels and the second response at higher concentrations. It is highly likely that the mechanisms activated by low vs high concentrations are different and might even be opposing (i.e., stimulatory vs. inhibitory). Future challenges point toward applying this approach to signature effects in target tissues of ecosystem biota that might have been variously exposed to such chemical contaminants. IR spectroscopy is a novel approach to derive chemical dose-response relationships in biological systems. In the future, the creation of a larger database containing an environmentally representative range of differentacting contaminants, together with an information system and computational methods in corresponding scale, may allow for the identification of chemical mode-of-action in an automated highthroughput and robust manner.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information that shows schematic representations of classic dose-response curves. This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel: þ44 (0)1524 510206; fax: þ44 (0)1524 510217; e-mail: [email protected].

’ ACKNOWLEDGMENT V.L. is a NERC-CEH algorithm student (NE/F008643/1).

ARTICLE

’ REFERENCES (1) Barber, J. L.; Walsh, M. J.; Hewitt, R.; Jones, K. C.; Martin, F. L. Low-dose treatment with polybrominated diphenyl ethers (PBDEs) induce altered characteristics in MCF-7 cells. Mutagenesis 2006, 21, 351– 360. (2) Darnerud, P. O.; Eriksen, G. S.; J
ohannesson, T.; Larsen, P. B.; Viluksela, M. Polybrominated diphenyl ethers: occurrence, dietary exposure, and toxicology. Environ. Health Perspect. 2001, 109 (Suppl 1), 49–68. (3) Hallgren, S.; Darnerud, P. O. Polybrominated diphenyl ethers (PBDEs), polychlorinated biphenyls (PCBs) and chlorinated paraffins (CPs) in rats - testing interactions and mechanisms for thyroid hormone effects. Toxicology 2002, 177, 227–243. (4) Martin, F. L. Complex mixtures that may contain mutagenic and/or genotoxic components: a need to assess in vivo target-site effect(s) associated with in vitro-positive(s). Chemosphere 2007, 69, 841–848. (5) Calabrese, E. J; Baldwin, L. A. Hormesis: A generalizable and unifying hypothesis. Crit. Rev Toxicol. 2001, 31, 353–424. (6) Stebbing, A. Hormesis - the stimulation of growth by low levels of inhibitors. Sci. Total Environ. 1982, 22, 213–234. (7) Calabrese, E. J.; Baldwin, L. A. Reevaluation of the fundamental dose-response relationship - A new database suggests that the U-shaped, rather than the sigmoidal, curve predominates. Bioscience 1999, 49, 725–732. (8) Calabrese, E. J; Baldwin, L. A Hormesis: U-shaped dose responses and their centrality in toxicology. Trends Pharmacol. Sci. 2001, 22, 285–291. (9) Stebbing, A. R. D. Interpreting 'dose-response' curves using homeodynamic data: with an improved explanation for hormesis. Dose Response 2009, 7, 221–233. (10) Martin, F. L.; Kelly, J. G.; Llabjani, V.; Martin-Hirsch, P. L.; Patel, I. I.; Trevisan, J.; Fullwood, N. J.; Walsh, M. J. Distinguishing cell types or populations based on the computational analysis of their infrared spectra. Nat. Protoc. 2010, 5, 1748–1760. (11) 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, 3356–3364. (12) Llabjani, V.; Trevisan, J.; Jones, K. C.; Shore, R. F.; Martin, F. L. Binary mixture effects by PBDE congeners (47, 153, 183, or 209) and PCB congeners (126 or 153) in MCF-7 cells: biochemical alterations assessed by IR spectroscopy and multivariate analysis. Environ. Sci. Technol. 2010, 44, 3992–3998. (13) Walsh, M. J.; German, M. J.; Singh, M. N.; Pollock, H. M.; Hammiche, A.; Kyrgiou, M.; Stringfellow, H. F.; Paraskevaidis, E.; MartinHirsch, P. L.; Martin, F. L. IR microspectroscopy: potential applications in cervical cancer screening. Cancer Lett. 2007, 246, 1–11. (14) Kalantzi, O. I.; Martin, F. L.; Thomas, G. O.; Alcock, R. E.; Tang, H. R.; Drury, S. C.; Carmichael, P. L.; Nicholson, J. K.; Jones, K. C. Different levels of polybrominated diphenyl ethers (PBDEs) and chlorinated compounds in breast milk from two UK regions. Environ. Health Perspect. 2004, 112, 1085–1091. (15) Wahl, M.; Lahni, B.; Guenther, R.; Kuch, B.; Yang, L.; Straehle, U.; Strack, S.; Weiss, C. A technical mixture of 2,2 0 ,4,4 0 -tetrabromo diphenyl ether (BDE47) and brominated furans triggers aryl hydrocarbon receptor (AhR) mediated gene expression and toxicity. Chemosphere 2008, 73, 209–215. (16) Jiao, H.; Allinson, S. L.; Walsh, M. J.; Hewitt, R.; Cole, K. J.; Phillips, D. H.; Martin, F. L. Growth kinetics in MCF-7 cells modulate benzo [a] pyrene-induced CYP1A1 up-regulation. Mutagenesis 2007, 22, 111–116. (17) Martin, F. L.; German, M. J.; Wit, E.; Fearn, T.; Ragavan, N.; Pollock, H. M. Identifying variables responsible for clustering in discriminant analysis of data from infrared microspectroscopy of a biological sample. J. Comput. Biol. 2007, 14, 1176–1184. (18) Walsh, M. J.; Singh, M. N.; Stringfellow, H. F.; Pollock, H. M.; Hammiche, A.; Grude, O.; Fullwood, N. J.; Pitt, M. A.; Martin-Hirsch, P. L.; 6134

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135

Environmental Science & Technology

ARTICLE

Martin, F. L. FTIR microspectroscopy coupled with two-class discrimination segregates markers responsible for inter- and intra-category variance in exfoliative cervical cytology. Biomark. Insights 2007, 3, 179–189. (19) Mourant, J. R.; Yamada, Y. R.; Carpenter, S.; Dominique, L. R.; Freyer, J. P. FTIR spectroscopy demonstrates biochemical differences in mammalian cell cultures at different growth stages. Biophys. J. 2003, 85, 1938–1947. (20) Suvorov, A.; Girard, S.; Lachapelle, S.; Abdelouahab, N.; Sebire, G.; Takser, L. Perinatal exposure to low-dose BDE-47, an emergent environmental contaminant, causes hyperactivity in rat offspring. Neonatology 2009, 95, 203–209. (21) Talsness, C. E.; Kuriyama, S. N.; Sterner-Kock, A.; Schnitker, P.; Grande, S. W.; Shakibaei, M.; Andrade, A.; Grote, K.; Chahoud, I. In utero and lactational exposures to low doses of polybrominated diphenyl ether-47 alter the reproductive system and thyroid gland of female rat offspring. Environ. Health Perspect. 2008, 116, 308–314. (22) Gregoraszczuk, E. L.; Milczarek, K.; W
ojtowicz, A. K.; Berg, V.; Skaare, J. U.; Ropstad, E. Steroid secretion following exposure of ovarian follicular cells to three different natural mixtures of persistent organic pollutants (POPs). Reprod. Toxicol. 2008, 25, 58–66. (23) Calabrese, E. J. Evidence that hormesis represents an “overcompensation” response to a disruption in homeostasis. Ecotoxicol. Environ. Saf. 1999, 42, 135–137. (24) J€arv, J. A model of nonexclusive binding of agonist and antagonist on g-protein coupled receptors. J. Theor. Biol. 1995, 175, 577–582. (25) Martin, F. L. Shining a new light into molecular workings. Nat. Methods 2011, 8, 385–387.

’ NOTE ADDED AFTER ASAP PUBLICATION There were minor text errors in the version published ASAP on June 23, 2011. The corrected version was published on June 27, 2011.

6135

dx.doi.org/10.1021/es200383a |Environ. Sci. Technol. 2011, 45, 6129–6135