Characterization and Biological Potency of Mono- to Tetra

Article pubs.acs.org/est

Characterization and Biological Potency of Mono- to TetraHalogenated Carbazoles Nicole Riddell,*,† Un-Ho Jin,*,‡ Stephen Safe,‡ Yating Cheng,‡ Brock Chittim,† Alex Konstantinov,† Robert Parette,§ Miren Pena-Abaurrea,∥,⊥ Eric J. Reiner,∥,⊥ David Poirier,∥ Tomislav Stefanac,† Alan J. McAlees,† and Robert McCrindle†,# †

Wellington Laboratories Inc., 345 Southgate Drive, Guelph, Ontario Canada N1G 3M5 Department of Veterinary Physiology & Pharmacology, Texas A&M University, College Station, Texas 77843-4466, United States § Matson & Associates, Inc., 331 East Foster Avenue, State College, Pennsylvania 16801, United States ∥ Ontario Ministry of the Environment, 125 Resources Road, Toronto, Ontario M9P 3 V6, Canada ⊥ Department of Chemistry, University of Toronto, Toronto, Ontario M5S 3H6, Canada # Department of Chemistry, University of Guelph, Guelph, Ontario N1G 2W1, Canada ‡

S Supporting Information *

ABSTRACT: This paper deals with the characterization and aryl hydrocarbon receptor (AhR) agonist activities of a series of chlorinated, brominated, and mixed bromo/chlorocarbazoles, some of which have been identified in various environmental samples. Attention is directed here to the possibility that halogenated carbazoles may currently be emitted into the environment as a result of the production of carbazole-containing polymers present in a wide variety of electronic devices. We have found that any carbazole that is not substituted in the 1,3,6,8 positions may be lost during cleanup of environmental extracts if a multilayer column is utilized, as is common practice for polychlorinated dibenzo-p-dioxin (dioxin) and related compounds. In the present study, 1H NMR spectral shift data for 11 relevant halogenated carbazoles are reported, along with their gas chromatographic separation and analysis by mass spectrometry. These characterization data allow for confident structural assignments and the derivation of possible correlations between structure and toxicity based on the halogenation patterns of the isomers investigated. Some halogenated carbazoles exhibit characteristics of persistent organic pollutants and their potential dioxin-like activity was further investigated. The structure-dependent induction of CYP1A1 and CYP1B1 gene expression in Ahresponsive MDA-MB-468 breast cancer cells by these carbazoles was similar to that observed for other dioxin-like compounds, and the magnitude of the fold induction responses for the most active halogenated carbazoles was similar to that observed for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). 2,3,6,7-Tetrachlorocarbazole was one of the most active halogenated carbazoles and, like TCDD, contains 4 lateral substituents; however, the estimated relative effect potency for this compound (compared to TCDD) was 0.0001 and 0.0032, based on induction of CYP1A1 and CYP1B1 mRNA, respectively.

■

ments may originate from industrial sources.4,7,9 Such a possibility draws support from the observation that extraction of a cyanobacterium growing on the shores of Ping Chau, Hong Kong yielded11 2,7-dibromocarbazole as well as 3,6-dibromoand 3,6-diiodo-carbazole. The authors commented on the extreme novelty of these compounds as natural products. It is perhaps more than coincidence that these three specific compounds have been used as intermediates for the production

INTRODUCTION

During the past 30 years, halogenated carbazoles have been detected in river deposits1,2 and lake sediments3,4 from the Great Lakes region of North America. Their occurrence in European soils,5−7 river deposits,8 and marine shore sediments7 confirms that these compounds are widely disseminated in the environment. Recently, it has been suggested that halogenated indigo dyes may be a likely source of 1,3,6,8-tetrabromocarbazole as well as some of the other halogenated carbazoles detected in environmental samples.9 However, the existence of 3,6-dichlorocarbazole in soils probably arises from other anthropogenic sources or has natural origins,9,10 while the dibromocarbazoles present in more recently deposited sedi© 2015 American Chemical Society

Received: Revised: Accepted: Published: 10658

June 4, 2015 July 24, 2015 July 30, 2015 July 30, 2015 DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology

Figure 1. Structures of 1H-indole, 9H-carbazole, and phenanthrene.

proprietary methods. The purity (>99.5%) of all compounds subjected to testing for TCDD-like toxicological potential was determined by 1H NMR and GC/MS. 1H NMR spectra were recorded for dilute solutions (ca., 2 mg solute) in dichloromethane-d2 (CD2Cl2, 0.8 mL) or chloroform-d1 (CDCl3, 0.8 mL) on a 600 MHz Bruker spectrometer at 24 °C with a delay of 6 s (NMR Centre, University of Guelph, Guelph, ON). The chemical shifts (reported in ppm) of samples run in CDCl3 were referenced to trimethylsilane (TMS) at δ 0.00 ppm, all others were referenced to residual CHDCl2 at δ 5.32 ppm. Gas chromatographic separations and mass spectrometry analyses were conducted on an Agilent 7890A GC/Agilent 5975C MSD using an Agilent J&W 30 m DB-5 column (250 μm ID, 0.25 μm film thickness). The injections were done in splitless mode, with the injector temperature at 250 °C and a helium carrier gas flow of 1.0 mL/min. The following temperature program was used: initial oven temperature, 100 °C ; hold of 5 min; ramp at 10 °C/min to 325 °C; hold 20 min. Data were collected in full scan mode (50−1000 amu). Cell Culture, Quantitative Real-Time Polymerase Chain Reaction (PCR), and Western Blot Analysis. MDA-MB-468 human breast cancer cell lines were obtained from the American Type Culture Collection (Manassas, VA). MDA-MB-468 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) nutrient mixture supplemented with 0.22% sodium bicarbonate, 0.011% sodium pyruvate, and 10% fetal bovine serum (FBS). Cells were maintained at 37 °C in the presence of 5% CO2. Cells (3 × 105 per well in 6-well plate format) were treated with different concentrations of the carbazoles in DMSO in fresh DMEM containing 2.5% FBS. Complementary DNA (cDNA) was prepared from the total RNA of cells using High Capacity RNA-to-cDNA Kit (Applied Biosystems, Foster City, CA). Each PCR was carried out using SYBR Green Mastermix (Applied Biosystems) and the Bio-Rad iCycler (MyiQ2) real-time PCR System. Values for each gene were normalized to expression levels of the housekeeping gene TATA-binding protein (TBP). The sequences of the primers used for real-time PCR were as follows: CYP1A1 sense 5′-GAG GCC AGA AGA AAC TCC GT-3′, antisense 5′-CCC AGC TCA GCT CAG TAC CT-3′; CYP1B1 sense 5′-ACC TGA TCC AAT TCT GCC TG-3′, antisense 5′-TAT CAC TGA CAT CTT CGG CG-3′; and TBP sense 5′-TTC TGA ATA GGC TGT GGG GT-3′, and antisense 5′-GAT CAG AAC AAC AGC CTG CC-3′. Dose response effects of the compounds were analyzed using GraphPad Primer version 6.05 for Windows (GraphPad Software, La Jolla California U.S.A., www.graphpad.com). Cellular lysates for protein were prepared in a lysis buffer containing 50 mM Tris-HCl (pH 7.5), 2 mM ethylenediaminetetraacetic acid, 150 mM NaCl, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM NaF, 1 mM Na3VO4, 1 mM phenylmethyl sulfonyl fluoride, 1X Protease inhibitor cocktail, and 1% NP-40. The same amount of

of polymers present in a wide variety of electronic devices, including LED-displays.12,13 Indeed, the Pearl River Delta region of China is one of the major manufacturing locations in the world of electronic products.14 This has led us to speculate that their presence in the extract was a result of the assimilation of artifacts of human activity rather than products of secondary metabolism. Of course, many aquatic organisms, including cyanobacteria, are known to bioaccumulate compounds of this nature.15 Of the approximately 4000 naturally occurring halogenated compounds documented,16 many of which are reportedly produced by marine organisms, we know of only this one case where the isolation of halogenated carbazoles has been attributed to a natural source. Indeed, the fact that the only other location from which similar compounds have been isolated7 involves shore sediments close to an industrial site may indicate that, in this case also, discharge from a terrestrial source has occurred. There are currently no published reports on degradation pathways of these compounds in sedimentary environments. The similarity of these carbazoles in structure to halogenated dibenzofuran derivatives raises questions about their potential toxicity, especially because of their known persistence in the environment.3,4 Indeed, 3-chlorocarbazole, 3,6-dichlorocarbazole, and 3,6-dibromocarbazole were recently shown to have dioxin-like activity using ethoxyresorufin-O-deethylase (EROD) induction in rat hepatoma cells.17 In order to definitively derive a structure to toxicity relationship, comprehensive NMR and GC/MS characterizations were completed. We report here a comparison of the aryl hydrocarbon receptor (AhR)-mediated effects of several chloro-, bromo-, and bromo/chloro-carbazoles by examining their relative potencies compared to 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) as inducers of CYP1A1 and CYP1B1 gene expression in MDA-MB-468 human breast cancer cells. This cell line is particularly sensitive to the induction of CYP1A1 and CYP1B1 by TCDD and different structural classes of AhR ligands, and also highly expresses the human AhR.18 Our results show that some di- and the tri- and tetrahalocarbazoles exhibit dioxin-like activity in these in vitro assays. We also include a brief summary of the characterization of these compounds by 1H NMR spectroscopy and gas chromatography coupled to a low resolution mass spectrometer (GC/MS).

■

MATERIALS AND METHODS

Chemicals and Chemical Characterization. Carbazole (for numbering of the carbon atoms see Figure 1) was purchased from Alfa Aesar (Ward Hill, MA), 3-bromocarbazole from Matrix Scientific (Columbia, SC) and 2,7-dibromo-, 3,6dibromo-, and 3,6-dichloro-carbazole from TCI America (Montgomeryville, PA). 3′,4′-Dimethoxy-α-naphthoflavone was purchased from Sigma-Aldrich (St. Louis, MO). All other compounds were synthesized at Wellington Laboratories using 10659

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

RESULTS AND DISCUSSION H NMR Spectra. To allow for direct comparison of chemical shift values, the NMR spectra were obtained in a single solvent (CD2Cl2) which was selected due to the appreciable solubility of more highly substituted carbazoles in this solvent. Previous work has provided chemical shift data for four bromocarbazoles in acetone-d6 or DMSO-d6,19 and for 2,7and 3,6-dibromocarbazole in CDCl3.11 In addition, in an extensive study of substituent effects on the NMR spectra of carbazoles, chemical shift values and some spin−spin coupling constants have been reported for six chlorocarbazoles and five bromocarbazoles in CDCl3.20 The relevant 1H NMR data obtained in the present study for carbazole and the 11 halogenated carbazoles are summarized in Table 1. Since the spectra of these 12 closely related compounds were available to us, assignments of the chemical shifts were straightforward based largely on peak multiplicity. The low-field signal arising from the NH in these carbazoles is clearly not under exchange but is broadened, presumably mainly, by partially coalesced coupling to the quadrupolar 14N nucleus. A comparison of the chemical shift data obtained here for solutions of carbazoles in CD2Cl2 with those published earlier20 for the same compounds in CDCl3 revealed a number of, what appeared to be, substantial solvent-induced chemical shifts. Obvious differences between our values and those reported20 in the Bonesi study were noted for H-1 in carbazole and its 3chloro derivative (previously reported at 7.16 and 7.05 ppm, respectively); for H-6 in the 3-chloro- and 3-bromo-carbazole (previously reported at 7.60 and 7.61 ppm, respectively); and for H-8 in these two halogenated compounds (previously reported at 7.18 and 7.16 ppm, respectively). As a result of this observation, the 1H NMR spectra of these three compounds were rerun using CDCl3 as solvent. This did reveal very small solvent-induced upfield shifts (see Table 1) for H-1 and H-8 in all three compounds. These shifts presumably result from subtle modifications in the transient interactions21 between the local dipoles in the solvent and the solute, especially for the region around the NH, for the two different solvents. Nevertheless, these small upfield shifts for H-1 and H-8 in 3chloro-9H-carbazole when the solvent is changed from CD2Cl2 to CDCl3 cause the resonances of these protons to “move through” those of H-2 and H-7, respectively, and results in an interesting alteration in the appearance of this part of the

7.49 m

7.25 m 7.25 m 7.37 q

7.54 7.59 7.80 7.46 7.66

q q d m d

7.45 m 7.45 m

7.60 7.51 7.44 7.63 7.38 7.44

brs q q brd q q

8.20 8.05 8.26 8.25 8.06 8.29 8.59 8.29 8.27 8.08 8.27 8.31 8.48 8.47 8.46 8.44 m m m m m q 7.47 7.47 7.48 7.47 7.44 7.42 m m m m m q d 7.42 7.42 7.45 7.45 7.45 7.41 7.53 m m m m m 7.23 7.24 7.25 7.24 7.26

m m m m m m brd brs m m brd brd m brd brd q

H-5

■

8.08 8.08 8.04 8.04 8.03 8.00 7.94 8.10 8.04 8.03 7.91 8.16 8.15 8.11 8.00 7.96

H-6

H-7

H-8

NH

extracted proteins was electrophoresed on 10% SDS− polyacrylamide gel, transferred to a PVDF membrane (BioRad, Hercules, CA), and blotted with CYP1A1 (SC-20772), CYP1B1 (SC-374228), and AhR (SC-5579) antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) using GAPDH as a loading control. Statistics. The induction of CYP1A1 and CYP1B1 mRNAs by the halogenated carbazoles in MDA-MB-468 cells, and the inhibition studies were determined at least three times (3 biological replicates), and statistical significant differences between the solvent (DMSO) and compound-induced RNA levels were analyzed using the Student’s t test. The results are expressed as means with error bars representing 95% confidence intervals for three biological replicates for each compound unless otherwise indicated, and a P value of less than 0.05 was considered statistically significant. A SpearmanKarber or Probit analysis was conducted to determine halfmaximum effective concentrations (EC50) values with 95% confidence intervals for CYP1A1 and CYP1B1 inductions.

brs brs brs brs brs brs brs brs brs brs brs brs brs brs brs brs

Environmental Science & Technology

10660

7.37 q q d d d d 7.54 7.75 7.80 7.62 7.66

7.48 q 7.50 q

brs q q brd q 7.60 7.38 7.32 7.63 7.38

a

Referenced to TMS. All others referenced to residual CHDCl2 at δ 5.32 (note entries for 3-Chloro-9H-carbazole in CD2Cl2).

m m m m m m brd brs m m brd brd q brd q q 8.08 8.08 8.05 8.05 8.04 8.00 7.94 8.10 8.20 8.19 7.91 8.16 8.12 8.11 7.98 7.96 7.23 m 7.24 m m m q q q q d 7.42 7.42 7.38 7.38 7.37 7.41 7.53 m m q q q q 7.47 7.44 7.41 7.41 7.35 7.42

CD2Cl2 CDCl3a CD2Cl2 CD2Cl2a CDCl3a CD2Cl2 CD2Cl2 CD2Cl2 CD2Cl2 CDCl3a CD2Cl2 CD2Cl2 CD2Cl2 CD2Cl2 CD2Cl2 CD2Cl2 CZ CZ 3-CCZ 3-CCZ 3-CCZ 36-CCZ 1368-CCZ 2367-CCZ 3-BCZ 3-BCZ 27-BCZ 36-BCZ 136-BCZ 1368-BCZ 1-B-36-CCZ 18-B-36-CCZ 9-H-carbazole 9-H-carbazole 3-chloro-9H-carbazole 3-chloro-9H-carbazole 3-chloro-9H-carbazole 3,6-dichloro-9H-carbazole 1,3,6,8-tetrachloro-9H-carbazole 2,3,6,7-tetrachloro-9H-carbazole 3-bromo-9H-carbazole 3-bromo-9H-carbazole 2,7-dibromo-9H-carbazole 3,6-dibromo-9H-carbazole 1,3,6-tribromo-9H-carbazole 1,3,6,8-tetrabromo-9H-carbazole 1-bromo-3,6-dichloro-9H-carbazole 1,8-dibromo-3,6-dichloro-9H-carbazole

H-4 H-3 H-2 H-1 solvent acronym

Table 1. Detailed Summary of 1H NMR Chemical Shifts of Halogenated Carbazoles in CD2Cl2 and CDCl3

1

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology

Figure 2. Magnified region of the 1H NMR spectrum of 3-CCZ displaying the H-1, H-2, H-7, and H-8 protons in (A) CDCl3 and (B) CD2Cl2.

protons and, in addition, depending on the substitution pattern, further couplings. The presence of this 5J coupling (which is analogous to homobenzylic coupling) could not be verified by performing a COSY experiment on carbazole itself probably because of the breadth of the NH signal. However, in a homonuclear decoupling experiment, irradiation at the resonant frequency of the NH proton resulted in a marked simplification of the H-4/H-5 multiplet. Although this long-range coupling was not discussed in the publications listed above, analogous couplings have been noted for a number of related compounds, for example indole. A literature search on this topic commencing from a fairly recent publication23 led back 50 years to one that reported24 finding comparable long-range couplings between the NH proton and H-4 in indole and, also H-4/H-5 in carbazole itself. In the earlier reports listed above, the downfield shifts for H4/H-5 in these carbazoles (ca., 8.0 ppm) were attributed to their peri-position. More extreme examples are observed25 for the analogous crowded bay protons in some condensed aromatic hydrocarbons, such as phenanthrene in which these protons resonate at 8.70 ppm. For H-4/H-5 in carbazoles, the downfield shifts could result largely from the proton’s location in the deshielding zone along the perimeter of the adjacent aromatic rings. Indeed, it is possible to visualize a rudimentary physical image along the following lines. Starting from benzene, fusing in a pyrrole ring to produce indole would expose H-4 to an additional ring current that would be a major contributor to

spectrum (Figure 2). We have no explanation for the differences in our and the earlier assignments apart from noting that the current work is based on spectra obtained at much higher field strength, namely, 600 MHz vs 200 MHz. Using carbazole itself as the reference molecule, the substituent chemical shifts observed for these carbazoles generally conform with the expected values.22 Hydrogen atoms ortho to a bromine substituent are shifted downfield, while meta and para hydrogens experience shielding. Also, as expected, the introduction of a chlorine atom into a phenyl ring leads to shielding of all three hydrogen atoms. However, when two chlorine atoms are present in a single phenyl ring, as in 1,3,6,8-tetrachlorocarbazole, the H-2/H-7 protons suffer appreciable deshielding as do H-1/H-8 in its 2,3,6,7-tetrachloro analogue. This was commented on earlier for the former isomer and attributed20 to an enhancement of the inductive, over the resonance, effect of the chlorine atoms. Consideration should also be given to the possibility that this effect results from significant changes in local dipole moments with a concomitant alteration in the preferred modes of transient interactions with the solvent.21 The marked downfield chemical shifts and the spin−spin coupling constants of H-4 and H-5 in these carbazoles warrant separate discussion. In all 12 compounds, these protons resonate well downfield from all others, apart from the broad peaks attributable to the amino protons. These protons all exhibit 5J spin−spin coupling, of about 0.8 Hz, to the NH 10661

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology

Figure 3. (A) HRGC/LRMS chromatogram of a mixture of the chloro-, bromo-, and chloro/bromo-carbazoles investigated. (B) HRGC/LRMS chromatogram of the same mixture after treatment using a standard multilayered (acid/base silica) column.29

shifts result. This deshielding has been ascribed to the migration of electron density away from the interacting protons.28 In phenanthrene (Figure 1), the shielding effect experienced by the bay protons is greater since the larger (sixvs five-membered) central ring results in closer proximity of each proton to the adjacent proton and phenyl ring. Gas Chromatography/Mass Spectrometry. The chromatogram resulting from GC separation of a mixture of all 11 halogenated carbazoles is shown in Figure 3. It is pertinent to emphasize that the more common cleanup procedures used in

the downfield shift of this proton from 7.36 ppm in benzene to 7.64 ppm in indole. This proton suffers an additional downfield shift to 8.08 ppm when a phenyl ring is fused with the C-2, C-3 atoms of indole to give carbazole. In this case, deshielding contributions from the ring current of this phenyl ring and a steric interaction with H-5 may be the major factors. The closeness of approach of the hydrogens (and of the phenyl ring) is revealed by X-ray structural studies of carbazole.26,27 It is well-known that where molecular geometry forces compression between hydrogen atoms, appreciable downfield 10662

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology

Figure 4. Induction of CYP1A1 and CYP1B1 by TCDD and the halogenated carbazoles. (A) CYP1A1 and CYP1B1 mRNA. MDA-MB-468 cells were treated with the test compounds for 24 h and CYP1A1 and CYP1B1 mRNA levels relative to TBP mRNA was determined by real time PCR. (B) Inhibition of CYP1A1 mRNA induction. Cells were treated with the test compounds for 24 h and CYP1A1 mRNA levels were determined by real time PCR. (C) CYP1A1, CYP1B1, and AhR protein. MDA-MB-468 cells were treated with the test compounds for 24 h and whole cell lysates were analyzed by Western blots as outlined in the Materials and Methods. CYP1A1 is significantly (p < 0.05) induced at all concentrations (>5-fold induction); significant (p < 0.05) induction of CYP1B1 is indicated (*). Significant inhibition of CYP1A1 induction is also indicated in B (**).

the analysis for polychlorinated dioxins (PCDDs) and dibenzofurans (PCDFs) as well as polybrominated diphenyl ethers (BDEs) may not be suited for halogenated carbazoles. The use of a typical multilayer column29 containing both sulfuric acid-treated and base treated-silica will likely remove any carbazole which is not substituted in the 1,3,6,8 positions. This may be because the four sites that are the most active toward electrophilic aromatic substitution30 (e.g., sulfonation) are not available for reaction or, alternatively, access to the N atom by a reactive species is blocked by the neighboring halogen atoms. It is interesting to note that, with the present mixture, only the compounds that have the 1,3,6,8-substitution pattern survived such treatment (Figure 3), although their apparent recoveries were still much lower than 100% (1,3,6,8-

tetrachloro-, 1,3,6,8-tetrabromo-, 1,8-dibromo-3,6-dichloro-carbazole, 18%, 29%, and 52% respectively). From Figure 3, it seems clear that the most important physicochemical factors in retention times are molecular size and ease of interaction of the NH with the stationary phase. A variety of comparisons of the data support this conclusion, but perhaps the most compelling are the relative retention times of the two tetrachlorocarbazoles and 1,3,6,8-tetrabromocarbazole. 1,3,6,8-Tetrachlorocarbazole with a retention time of 21.94 min is much less retained than its brominated analogue (25.47 min), presumably largely because of its smaller size. However, 2,3,6,7-tetrachlorocarbazole has a longer retention time (25.61 min) than the tetrabromobromocarbazole investigated, possibly to some extent because of a higher 10663

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology Table 2. Estimated REPs for Halogenated Carbazoles EC50 (nM) compound 2,3,7,8-tetrachlorodibenzodioxin 3-bromo-9H-carbazole 2,7-dibromo-9-H-carbazole 3,6-dibromo-9-H-carbazole 1,3,6-tribromo-9H-carbazole 1,3,6,8-tetrabromo-9H-carbazole 3-chloro-9H-carbazole 3,6-dichloro-9H-carbazole 1,3,6,8-tetrachloro-9H-carbazole 2,3,6,7-tetrachloro-9H-carbazole 1-bromo-3,6-dichloro-9H-carbazole 1,8-dibromo-3,6-dichloro-9H-carbazole a

CYP1A1

range of REPs CYP1B1

a

0.046 2500 3000 2700 490a 150a 1700 410 70 450 740 140

0.29 1100 2300 2000 340a 30a n/a n/a 50a 90a 870 30a

CYP1A1 1.8 1.3 1.7 9.0 3.1 2.7 1.1 6.6 1.0 6.0 3.2

CYP1B1

1.0 × 10−5 × 10−5 × 10−5 × 10−5 × 10−4 × 10−5 × 10−4 × 10−4 × 10−4 × 10−5 × 10−4

2.6 1.3 1.5 8.5 9.7

1.0 × 10−4 × 10−4 × 10−4 × 10−4 × 10−3

5.8 3.2 3.3 9.7

× × × ×

10−3 10−3 10−4 10−3

Values determined using Probit Analysis (all other EC50 values were calculated using the Spearman-Karber method).

charge separation31 but, more importantly, because of the interaction of the NH with the phenyl groups in the DB-5 stationary phase. In the two compounds where halogen atoms are attached to C-1 and C-8, NH···π hydrogen bonding32 will be greatly obstructed. Indeed, the retention times of 2,3,7,8tetrachlorodibenzofuran (21.38 min) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (21.62 min) under identical conditions are very similar to that of 1,3,6,8-tetrachlorocarbazole. Consistent with the behavior of halogenated dibenzofurans,33 the EI mass spectra of all 11 compounds have the parent ion as the base peak (Figure S1). Each loses all of its halogen atoms consecutively yielding a fragment which is representative of the molecular skeleton. The marked stability of the carbazole structure to electron impact is evidenced by the intensity of the molecular ion in the mass spectrum of carbazole itself.34 Not unexpectedly, loss of bromine from the molecular ion is easier than loss of chlorine. For example, based on the intensity of the parent ion as 100%, [MCl]+ from 3-chlorocarbazole has a 35.7% relative intensity (RI) while [MBr] + from 3bromocarbazole has a 59.4% RI. (Table S1). Similarly, loss of one of the two chlorines from 1-bromo-3,6-dichlorocarbazole to give [MCl]+ has an 8.8% RI, while loss of the only bromine atom to give [MBr]+ has an 18.9% RI. Ah-Responsive Activity and Relative Congener Potencies. TCDD and related toxic halogenated aromatic hydrocarbons (HAHs) bind the AhR and induce a common pattern of biochemical and toxic responses. The induction of CYP1A1 and CYP1A1-dependent activities (e.g., EROD activity) are among the most sensitive indicators of TCDDlike activity.35,36 Since these compounds act through the AhR, there is a structure-dependent correlation between the potencies of the individual congeners. This relationship has been used to determine relative effect potencies (REPs) of HAHs compared to TCDD, which is assigned an REP of 1.0.37−39 This approach has been extensively used by regulatory agencies to determine the total “dioxin” toxic equivalencies (TEQs) of a mixture of HAHs; which is equal to their individual concentrations or amounts times the REP for each individual HAH.38 Indeed, Figure 4A shows that TCDD significantly induced CYP1A1 mRNA at concentrations as low as 0.01 nM. We observed a structure-dependent induction of CYP1A1 and CYP1B1 gene expression by the halogenated carbazoles in MDA-MB-468 cells; the most active compounds were the 1,3,6,8- and 2,3,6,7-chloro/bromocarbazoles which were more potent than their lower halogenated analogs. In

addition, the AhR antagonist 3′,4′-dimethoxy-α-naphthoflavone40 also inhibited TCDD- and 2,3,6,7-tetrachlorocarbazoleinduced CYP1A1 mRNA (Figure 4B), and similar inhibition was observed for 3′-methoxy-4′-nitroflavone (data not shown). Interestingly, both AhR antagonist were partial agonists in MDA-MB-468 cells. The major structure−activity difference between the halogenated carbazoles and the corresponding halogenated dibenzo-p-dioxins and dibenzofurans was the higher activity of the 1,3,6,8- versus the 2,3,6,7-tetrahalocarbazoles (Table 2). The 2,3,6,7 substitution pattern contains four lateral halogen substituents comparable to that present in 2,3,7,8-TCDD; however, our results indicate that, for the chlorinated carbazoles, only two lateral substituents (3 and 6) are necessary for high CYP1A1/CYP1B1 induction. For instance, both the 1-bromo-3,6-dichloro- and 3,6-dichlorocarbazole significantly induced CYP1A1 mRNA levels at a concentration of 1 μM and were more potent than 3bromocarbazole, 3-chlorocarbazole, 3,6-dibromocarbazole, and 2,7-dibromocarbazole. Differences in enzyme induction potency between the bromo- versus chloro-carbazoles were observed with the chlorinated carbazoles displaying a higher degree of potency. This is especially evident when comparing the CYP1A1 EC50 values of 3,6-dichlorocarbazole (410 nM) and 3,6-dibromocarbazole (2700 nM). In general, the REP values determined for the halogenated carbazoles investigated (based on comparisons between the induction of CYP1A1 and CYP1B1 by TCDD) are consistent with the suggested structure−activity relationships. The effects of TCDD (1 nM) and the halogenated carbazoles (1 μM) on expression of CYP1A1, CYP1B1, and AhR proteins in MDA-MB-468 cells were also determined (Figure 4C). At the concentrations used in this experiment, the induction of CYP proteins was observed only for the most active halogenated carbazoles; however, the carbazoles appeared to be more effective than TCDD as inducers of CYP1B1 vs CYP1A1 proteins. TCDD and, to a lesser extent, 1,3,6,8- and 2,3,6,7-tetrachloro-carbazole decrease the cytosolic AhR protein; this is routinely observed for TCDD in most cancer cell lines.18 Surprisingly, some of the halogenated carbazoles, such as 3-bromocarbazole, also decreased AhR expression and the reason for this response is not understood. REPs for the 1,3,6,8-tetrahalocarbazoles ranged from 0.00031 to 0.00066 (CYP1A1 induction) to 0.0097 to 0.0058 (CYP1B1 induction) which overlap with the WHO-TEFs for higher chlorinated dibenzofurans, dibenzo-p-dioxins, and some coplanar 10664

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology PCBs.37,38 It is clear from our results and previous reports17 that halogenated carbazoles are members of the HAH family of environmental contaminants that act through the AhR and their REPs for CYP induction are relatively low in MDA-MB-468 cells. However, a direct comparison of 2,3,6,7-tetrachlorocarbazole (REP range of 0.0001 to 0.0032) with the isosteric 2,3,7,8tetrachlorodibenzofuran (REP ≈ 0.1) indicates that replacement of the oxygen bridging atom with NH reduces dioxin-like activity in the in vitro assays. The relative in vivo potencies of these compounds in animal models will be required to derive a REP that can be used for hazard assessment of these compounds and their potential adverse environmental and human health effects. Moreover, like other persistent organic pollutants, halogenated carbazoles may exhibit AhR-independent effects which should also be investigated. The lower AhR agonist activities of halogenated carbazoles should also be evaluated in terms of potential exposure given the levels measured in sediments. For example, the concentration of 1,3,6,8-tetrachlorocarbazole found1 at a site in the Buffalo River was quantified at 25 ng/g while similarly high levels (up to 65 ng/g) were measured3,4 for 1,3,6,8tetrabromocarbazole in several sediment cores from Lake Michigan. Concerns regarding the potential adverse impacts of halogenated carbazoles on environmental species and humans will require more extensive studies that include environmental sampling, additional quantitative in vivo/in vitro toxicity studies, potential for bioaccumulation, risk to wildlife, as well as an evaluation of possible synergistic effects. However, the results of this study demonstrate that halogenated carbazoles represent yet another class of persistent organic pollutants with dioxin-like activity.

■

(2) Pena-Abaurrea, M.; Jobst, K. J.; Ruffolo, R.; Shen, L.; McCrindle, R.; Helm, P. A.; Reiner, E. J. Identification of potential novel bioaccumulative and persistent chemicals in sediments from Ontario (Canada) using scripting approaches with GCxGC-TOF MS analysis. Environ. Sci. Technol. 2014, 48 (16), 9591−9599. (3) Zhu, L. Y.; Hites, R. A. Identification of brominated carbazoles in sediment cores from Lake Michigan. Environ. Sci. Technol. 2005, 39 (24), 9446−9451. (4) Guo, J. H.; Chen, D.; Potter, D.; Rockne, K. J.; Sturchio, N. C.; Giesy, J. P.; Li, A. Polyhalogenated carbazoles in sediments of Lake Michigan: a new discovery. Environ. Sci. Technol. 2014, 48 (21), 12807−12815. (5) Reischl, A.; Joneck, M.; Dumler-Gradl, R. Chlorocarbazoles in soils. Umweltwiss. Schadst.-Forsch. 2005, 17 (4), 197−200. (6) Trobs, L.; Henkelmann, B.; Lenoir, D.; Reischl, A.; Schramm, K. W. Degradative fate of 3-chlorocarbazole and 3,6-dichlorocarbazole in soil. Environ. Sci. Pollut. Res. 2011, 18 (4), 547−555. (7) Grigoriadou, A.; Schwarzbauer, J. Non-target screening of organic contaminants in sediments from the industrial coastal area of Kavala City (NE Greece). Water, Air, Soil Pollut. 2011, 214 (1−4), 623−643. (8) Kronimus, A.; Schwarzbauer, J.; Dsikowitzky, L.; Heim, S.; Littke, R. Anthropogenic organic contaminants in sediments of the Lippe river, Germany. Water Res. 2004, 38 (16), 3473−3484. (9) Parette, R.; McCrindle, R.; McMahon, K. S.; Pena-Abaurrea, M.; Reiner, E. J.; Chittim, B.; Riddell, N.; Voss, G.; Dorman, F. L.; Pearson, W. N. Halogenated indigo dyes: A likely source of 1,3,6,8tetrabromocarbazole and some other halogenated carbazoles in the environment. Chemosphere 2015, 127 (0), 18−26. (10) Mumbo, J.; Lenoir, D.; Henkelmann, B.; Schramm, K. W. Enzymatic synthesis of bromo- and chlorocarbazoles and elucidation of their structures by molecular modeling. Environ. Sci. Pollut. Res. 2013, 20 (12), 8996−9005. (11) Lee, S. C.; Williams, G. A.; Brown, G. D.; Maculalactone, L. and three halogenated carbazole alkaloids from Kyrtuthrix maculans. Phytochemistry 1999, 52 (3), 537−540. (12) Morin, J. F.; Leclerc, M.; Ades, D.; Siove, A. Polycarbazoles: 25 years of progress. Macromol. Rapid Commun. 2005, 26 (10), 761−778. (13) Karon, K.; Lapkowski, M.; Juozas, G. Electrochemical and UVVis/ESR spectroelectrochemical properties of polymers obtained from isomeric 2,7-and 3,6-linked carbazole trimers; influence of the linking topology on polymers properties. Electrochim. Acta 2014, 123, 176− 182. (14) Zhang, K.; Wei, Y.-L.; Zeng, E. Y. A review of environmental and human exposure to persistent organic pollutants in the Pearl River Delta, South China. Sci. Total Environ. 2013, 463, 1093−1110. (15) Katagi, T. Bioconcentration, bioaccumulation, and metabolism of pesticides in aquatic organisms. In Reviews of Environmental Contamination and Toxicology; Whitacre, D. M., Ed.; Springer-Verlag: New York, 2010; Vol. 204; pp 1−132. (16) Gribble, G. W. The diversity of naturally produced organohalogens. Chemosphere 2003, 52 (2), 289−297. (17) Mumbo, J.; Henkelmann, B.; Abdelaziz, A.; Pfister, G.; Nguyen, N.; Schroll, R.; Munch, J.; Schramm, K.-W. Persistence and dioxin-like toxicity of carbazole and chlorocarbazoles in soil. Environ. Sci. Pollut. Res. 2015, 22 (2), 1344−1356. (18) Jin, U.-H.; Lee, S.-O.; Safe, S. Aryl hydrocarbon receptor (AHR)-active pharmaceuticals are selective AHR modulators in MDAMB-468 and BT474 breast cancer cells. J. Pharmacol. Exp. Ther. 2012, 343 (2), 333−341. (19) Smith, K.; James, D. M.; Mistry, A. G.; Bye, M. R.; Faulkner, D. J. A new method for the bromination of carbazoles, beta-carbolines and iminodibenzyls by use of N-bromosuccinimide and silica-gel. Tetrahedron 1992, 48 (36), 7479−7488. (20) Bonesi, S. M.; Ponce, M. A.; Erra-Balsells, R. A study of substituent effect on H-1 and C-13 NMR spectra of mono, di and poly substituted carbazoles. J. Heterocycl. Chem. 2005, 42 (5), 867−875. (21) Engler, E. M.; Laszlo, P. New description of nuclear magnetic resonance solvent shifts for polar solvents in weakly associating aromatic solvents. J. Am. Chem. Soc. 1971, 93 (6), 1317−1327.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b02751. Figure S1: HRGC/LRMS spectra associated with the halogenated carbazoles presented in Figure 2. Table S1: Summary of the relative intensities of the prominent mass spectral signals obtained during the HRGC/LRMS analysis. (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*Phone: 1-519-822-2436; fax: 1-519-822-2849; e-mail: nicole@ well-labs.com (N.R.). *Phone: 1-979- 845-9181; fax: 1-979-862-4929; e-mail: [email protected] (U.-H.J.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The financial support of the Syd Kyle Chair, Texas A&M AgriLife Research, and the National Institutes of Health (P30 ES023512) is gratefully acknowledged.

■

REFERENCES

(1) Kuehl, D. W.; Durhan, E.; Butterworth, B. C.; Linn, D. Tetrachloro-9H-carbazole, a previously unrecognized contaminant in sediments of the Buffalo River. J. Great Lakes Res. 1984, 10 (2), 210− 214. 10665

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666

Article

Environmental Science & Technology (22) Pretsch, E.; Clerc, T.; Seibl, J.; Simon, W. Tables of Spectral Data for Structure Determination of Organic Compounds; Springer-Verlag: Berlin, Heidelberg, New York, Tokyo, 1983. (23) Akhmedov, N. G.; Dacko, C. A.; Guven, A.; Soderberg, B. C. G. Complete analysis of the H-1 and C-13 NMR spectra of diastereomeric mixtures of (R,S- and S,S-)-3,6-dimethoxy-2,5dihydropyrazine-substituted indoles and their conformational preference in solution. Magn. Reson. Chem. 2010, 48 (2), 134−150. (24) Black, P. J.; Heffernan, M. L. Analysis of Proton Magnetic Resonance Spectra of Heteroaromatic systems.4. Benzofuran, Indole, and related compounds. Aust. J. Chem. 1965, 18 (3), 353−361. (25) Abraham, R. J.; Canton, M.; Reid, M.; Griffiths, L. Proton chemical shifts in NMR. Part 14. Proton chemical shifts, ring currents and pi electron effects in condensed aromatic hydrocarbons and substituted benzenes. Journal of the Chemical Society-Perkin Transactions 2 2000, 4, 803−812. (26) Gajda, K.; Zarychta, B.; Kopka, K.; Daszkiewicz, Z.; Ejsmont, K. Substituent effects in nitro derivatives of carbazoles investigated by comparison of low-temperature crystallographic studies with density functional theory (DFT) calculations. Acta Crystallogr., Sect. C: Struct. Chem. 2014, 70, 987−U248. (27) Kurahash, M.; Fukuyo, M.; Shimada, A.; Furusaki, A.; Nitta, I. Crystal and molecular structure of carbazole. Bull. Chem. Soc. Jpn. 1969, 42 (8), 2174−2179. (28) Cheney, B. V. Magnetic deshielding of protons due to intramolecular steric interactions with proximate hydrogens. J. Am. Chem. Soc. 1968, 90 (20), 5386−5390. (29) Lamparski, L. L.; Nestrick, T. J.; Stehl, R. H. Determination of parts-per-trillion concentrations of 2,3,7,8-tetrachlordibenzo-paradioxin in fish. Anal. Chem. 1979, 51 (9), 1453−1458. (30) Bonesi, S. M.; ErraBalsells, R. On the synthesis and isolation of chlorocarbazoles obtained by chlorination of carbazoles. J. Heterocycl. Chem. 1997, 34 (3), 877−889. (31) Do, L.; Geladi, P.; Haglund, P. Multivariate data analysis to characterize gas chromatography columns for dioxin analysis. Journal of Chromatography A 2014, 1347, 137−145. (32) Pfaffen, C.; Infanger, D.; Ottiger, P.; Frey, H.-M.; Leutwyler, S. N-H center dot center dot center dot pi hydrogen-bonding and largeamplitude tipping vibrations in jet-cooled pyrrole-benzene. Phys. Chem. Chem. Phys. 2011, 13 (31), 14110−14118. (33) Donnelly, J. R.; Sovocool, G. W. Mass-spectral characteristics of bromochlorinated dibenzo-para-dioxins and dibenzofurans. Chemosphere 1990, 20 (3−4), 295−300. (34) Hallberg, A.; Martin, A. R. Mass-spectral fragmentation patterns of heterocycles.6. Carbazole and 1,8-dideuteriocarbazole. J. Heterocycl. Chem. 1984, 21 (3), 837−840. (35) Poland, A.; Knutson, J. C. 2,3,7,8-Tetrachlorodibenzo-paradioxin and related halogenated aromatic hydrocarbons. Examination of the mechanism of toxicity. Annu. Rev. Pharmacol. Toxicol. 1982, 22, 517−554. (36) Safe, S. Polychlorinated- biphenyls (PCBS), dibenzo-paradioxins (PCDDS), dibenzofurans (PCDFS), and related compounds. Environmental and mechanistic considerations which support the development of toxic equivalency factors (TEFS). Crit. Rev. Toxicol. 1990, 21 (1), 51−88. (37) Van den Berg, M.; Denison, M. S.; Birnbaum, L. S.; DeVito, M. J.; Fiedler, H.; Falandysz, J.; Rose, M.; Schrenk, D.; Safe, S.; Tohyama, C.; Tritscher, A.; Tysklind, M.; Peterson, R. E. Polybrominated dibenzo-p-dioxins, dibenzofurans, and biphenyls: inclusion in the toxicity equivalency factor concept for dioxin-like compounds. Toxicol. Sci. 2013, 133 (2), 197−208. (38) Van den Berg, M.; Birnbaum, L. S.; Denison, M.; De Vito, M.; Farland, W.; Feeley, M.; Fiedler, H.; Hakansson, H.; Hanberg, A.; Haws, L.; Rose, M.; Safe, S.; Schrenk, D.; Tohyama, C.; Tritscher, A.; Tuomisto, J.; Tysklind, M.; Walker, N.; Peterson, R. E. The 2005 World Health Organization reevaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol. Sci. 2006, 93 (2), 223−241.

(39) Van den Berg, M.; Birnbaum, L.; Bosveld, A. T. C.; Brunstrom, B.; Cook, P.; Feeley, M.; Giesy, J. P.; Hanberg, A.; Hasegawa, R.; Kennedy, S. W.; Kubiak, T.; Larsen, J. C.; van Leeuwen, F. X. R.; Liem, A. K. D.; Nolt, C.; Peterson, R. E.; Poellinger, L.; Safe, S.; Schrenk, D.; Tillitt, D.; Tysklind, M.; Younes, M.; Waern, F.; Zacharewski, T. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 1998, 106 (12), 775−792. (40) Murray, I. A.; Flaveny, C. A.; Chiaro, C. R.; Sharma, A. K.; Tanos, R. S.; Schroeder, J. C.; Amin, S. G.; Bisson, W. H.; Kolluri, S. K.; Perdew, G. H. Suppression of cytokine-mediated complement factor gene expression through selective activation of the Ah receptor with 3′,4′-dimethoxy-alpha-naphthoflavone. Mol. Pharmacol. 2011, 79 (3), 508−19.

10666

DOI: 10.1021/acs.est.5b02751 Environ. Sci. Technol. 2015, 49, 10658−10666