Arch Toxicol DOI 10.1007/s00204-014-1352-1
MOLECULAR TOXICOLOGY
Novel role of hnRNP‑A2/B1 in modulating aryl hydrocarbon receptor ligand sensitivity See‑Wun Cho · Ken‑ichi Suzuki · Yoshiaki Miura · Tatsuhiko Miyazaki · Masato Nose · Hisato Iwata · Eun‑Young Kim
Received: 31 March 2014 / Accepted: 25 August 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract The aryl hydrocarbon receptor (AHR) is responsible for susceptibility to its ligand-dependent responses. However, the effect of non-AHR factors is less clear. To explore the non-AHR factors, we used two mouse strains with different AHR genetic variants, namely C3H/lpr and MRL/lpr strains with Ala and Val as the 375th amino acid residue, respectively. To assess the contribution of AHR alone, COS-7 cells transiently expressing AHR from each strain were treated with 6-formylindolo[3,2-b]
Electronic supplementary material The online version of this article (doi:10.1007/s00204-014-1352-1) contains supplementary material, which is available to authorized users. S.-W. Cho · E.-Y. Kim (*) Department of Life and Nanopharmaceutical Science, Kyung Hee University, Hoegi Dong, Dongdaemun‑Gu, Seoul 130‑701, Korea e-mail: [email protected] S.-W. Cho · E.-Y. Kim Department of Biology, Kyung Hee University, Hoegi Dong, Dongdaemun‑Gu, Seoul 130‑701, Korea K. Suzuki Graduate School of Science, Hiroshima University, Hiroshima, Japan Y. Miura · H. Iwata Center for Marine Environmental Studies (CMES), Ehime University, Matsuyama, Japan T. Miyazaki · M. Nose Department of Pathogenomics, Ehime University Graduate School of Medicine, Toon, Japan Present Address: M. Nose Department of Histopathology, Tohoku University Graduate School of Medicine, Sendai, Japan
carbazole (FICZ) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), and xenobiotic-responsive element (XRE)-driven reporter gene activities were measured. FICZ-EC50 values for the C3H/lpr and MRL/lpr AHR-mediated transactivation were 0.023 and 0.046 nM, respectively, indicating a similar susceptibility in both AHR genotypes. In contrast, C3H/lpr AHR was fourfold more sensitive to TCDD than MRL/lpr AHR. By a pull-down assay using a XRE-containing PCR product as bait and the hepatic nuclear extracts of both FICZ-treated mouse strains, we identified two interacting proteins as heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP-A2) and its splicing variant (hnRNPA2b). Immunoprecipitation assays demonstrated the AHR interaction with hnRNP-A2/B1. When hnRNP-A2 was coexpressed with the MRL/lpr or C3H/lpr AHR in COS-7, FICZ treatment decreased EC50 to about threefold in both AHR genotypes, compared with EC50 in AHR alone. Similarly, hnRNP-A2b co-expression also lowered the FICZEC50 values. In TCDD-treated COS-7, responses depended on the AHR genotype; while no change in TCDD-EC50 was observed for C3H/lpr AHR when hnRNP-A2 was co-expressed, the value was reduced to nearly tenfold for MRL/lpr AHR. Co-transfection with hnRNP-A2b attenuated the AHR sensitivity to TCDD. In conclusion, the hnRNP-A2/B1 interacting with AHR may be a modulator of the AHR ligand sensitivity. Keywords AHR · Sensitivity · FICZ · TCDD · C3H/lpr · MRL/lpr · hnRNP-A2/B1
Introduction The aryl hydrocarbon receptor (AHR), a member of the basic helix-loop-helix (bHLH) and Per-Arnt-Sim (PAS)
13
superfamily, is a ligand-dependent transcription factor. The PAS-B domain encompasses a ligand-binding site (Ge and Elferink 1998; Swanson and Bradfield 1993) and the C-terminal transactivation domain (TAD) interacts with several transcriptional cofactors (Rowlands et al. 1996). In the absence of ligands, AHR forms a complex with heat shock proteins (Hsp90s), a small protein (p23), and an immunophilin-like protein (XAP2) in the cytoplasm (Perdew 1988; Petrulis et al. 2000). Upon ligand binding, the AHR complex translocates into the nucleus, forms a heterodimer with AHR nuclear translocator (ARNT), and binds to a xenobiotic-responsive element (XRE; 5′-TNGCGGTG-3′) in the promoter region of the AHR gene battery which includes cytochrome P450 (CYP) 1A1 (Denison et al. 1988). Following the recruitment of a large number of co-activator/ co-repressor complexes, AHR consequently mediates various biochemical and toxic effects (Esser et al. 2009; Fernandez-Salguero et al. 1996; Mimura et al. 1997; Mandal 2005; Vorderstrasse et al. 2001). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and 6-formylindolo[3,2-b]carbazole (FICZ) are representative exogenous and endogenous AHR ligands, respectively, and are efficient inducers of the AHR gene battery (Safe 1990; Mukai and Tischkau 2007; Wincent et al. 2009). TCDD, one of the dioxin-like congeners including polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls, is the most potent AHR ligand (Safe 1990) and activates AHR to induce endocrine disruption, tumor promotion, and teratogenesis (Kerkvliet 2002; Taylor and Zhulin 1999). FICZ is formed from tryptophan in the UV-exposed skin and also binds AHR with high affinity (Fritsche et al. 2007; Rannug et al. 1987, 1995). AHR has recently emerged as a critical physiological regulator of innate and adaptive immune responses (Kerkvliet 2009; Veldhoen et al. 2008; Quintana et al. 2008; Kimura et al. 2008). AHR-mediated response to ligand exposure demonstrates strain- and species-specific differences. For example, the genetic polymorphism of AHRs in the C57BL/6 and DBA/2 mouse strains shows differential dioxin sensitivity. The 375th amino acid in the AHR ligand-binding site is Ala and Val in the C57BL/6 and DBA/2 strains, respectively. This difference results in lowered sensitivity to TCDD in DBA/2 compared to the C57BL/6 strain (Ema et al. 1994; Okey et al. 1989; Poland et al. 1994) and suggests that the 375th amino acid residue of the mouse AHR is one of the critical determinants for TCDD sensitivity. Nevertheless, factors other than AHR have also been implicated in this susceptibility (Kawakami et al. 2006; Moffat et al. 2010). Here, we hypothesize that the differences in the coactivators that interact with AHR may potentially modify ligand sensitivity, since the over-expression of some co-activators can enhance the transcriptional activity of
13
Arch Toxicol
several nuclear receptors in response to their respective ligands (Beischlag et al. 2002). Steroid receptor coactivator 1 (SRC1) and p300 as co-activators of AHR are known to play a critical role in the transcription of the AHR gene battery by forming a complex with general transcription factors (Hestermann and Brown 2003; Kollara and Brown 2006; Zhang et al. 2008). However, factors that affect the ligand sensitivity in AHR signaling pathways remain unexplored. Thus, identifying these factors is important for understanding the mechanisms underlying this susceptibility. In this study, we used two mouse models with different AHR genotypes, a C3H/lpr and an MRL/lpr strains with Ala and Val as the 375th amino acid residue in the AHR gene, respectively, to explore the non-AHR factors involved in AHR ligand sensitivity. Both strains have a single mutated gene, lymphocyte proliferation (lpr), that encodes the deletion mutant of Fas, resulting in the immunological disruption of Fas-mediated apoptosis (WatanabeFukunaga et al. 1992). Nevertheless, the MRL/lpr strain develops many symptoms of autoimmune diseases, while C3H/lpr strain shows less severe phenotypes (Nose et al. 1989), indicating that the onset of autoimmune diseases is controlled by a polygene network (Nose 2007). Initially, to assess the contribution of AHR alone, XRE-driven reporter gene activities were measured in COS-7 cells transiently expressing AHR from C3H/lpr or MRL/lpr strains treated with graded concentrations of FICZ or TCDD. Secondary, a pull-down assay was carried out using XRE as a bait for the MRL/lpr and C3H/lpr liver nuclear extracts. The highly expressed proteins in the liver nuclear extracts were identified as heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP-A2) and its splicing variant (hnRNP-A2b), which are well-known autoantigens involved in autoimmune diseases (Steiner et al. 1996; Cho et al. 2012). The hnRNP was transiently expressed with the MRL/lpr or C3H/lpr AHR into COS-7 treated with FICZ or TCDD, and the EC50 was compared with the EC50 value obtained from AHR expression alone to examine whether the hnRNP-A2/B1 is a critical factor that can regulate the sensitivity to TCDD and FICZ.
Materials and methods Animals C3H/lpr and MRL/lpr strains of mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). All mice used in this study were bred and housed in the animal facility of the Ehime University under a pathogen-free and climatecontrolled environment with 12-h light/dark cycles. All the animal studies were performed according to the guidelines
Arch Toxicol
set by the relevant animal regulatory committee of the Ehime University. Plasmid constructs The pcDNA™3.1/Zeo(+) expression vectors (Invitrogen) containing MRL/lpr AHR, C3H/lpr AHR, MRL/lpr ARNT, and C3H/lpr ARNT were constructed using the method described by Kim et al. (2011). The cDNA clone of hnRNP-A2/B1 (hnRNP-A2 and hnRNP-A2b) was amplified with primers containing a CACC site using KOD-Plus-Ver. 2 (Toyobo, Japan). The specific primers were designed for PCR using the National Center for Biotechnology Information (NCBI) genome database (forward, CACCATGGAG AGAGAAAAGGAA and reverse, ATATCTGCTCCTTCC ACCATA). PCR products were purified by QIAquick PCR Purification Kit (Qiagen) and ligated into the pcDNA™3.1 Directional TOPO® vector (Invitrogen). The reporter vector, pGudLuc 6.1, that contains the firefly luciferase reporter gene regulated by a 484-bp fragment with four XREs in the 5′-upstream region of mouse CYP1A1, was kindly provided by Dr. Michael S. Denison (University of California, Davis, USA). The XRE-containing fragment was cloned into the pGL4.10 firefly luciferase reporter vector (Promega). Luciferase reporter assay The constructed reporter plasmids were transfected into an African green monkey kidney cell line, COS7. The cells were seeded at a concentration of 6 × 104 cells/well in 24-well plates and were maintained in Roswell Park Memorial Institute (RPMI) medium-1640 (Hyclone, USA) supplemented with 10 % fetal bovine serum (Hyclone) at 37 °C under 5 % CO2. Lipofectamine LTX (Invitrogen) was used for transfections according to the manufacturer’s instructions. Twenty nanograms of the mouse CYP1A1 5′-flanking region cloned into pGL4.10 firefly luciferase reporter vector was co-transfected with 5 ng of pcDNA™3.1/Zeo(+)_AHR, 50 ng of pcDNA™3.1/Zeo(+)_ARNT expression vector, and 5 ng of pGL4.74 [hRluc/TK] Renilla luciferase transfection control vector (Promega). In addition, 20 ng of the expression vector of hnRNP-A2/B1 (hnRNP-A2 or hnRNP-A2b) was also transfected to examine the effect of this protein on the AHR ligand sensitivity. After 5 h of the transfection, the cells were treated with graded concentrations of FICZ or TCDD dissolved in dimethyl sulfoxide (DMSO; Sigma). The test solutions of individual chemicals were prepared by serially diluting the stock solutions. Following ligand treatment for 18 h, luciferase activity as a measure of the response of C3H/lpr and MRL/lpr mouse AHRs to the tested chemicals was measured with the
Dual-Luciferase Assay Kit (Promega) and a Multi-Mode Microplate Reader (BioTek Synergy2). The final luminescence values were expressed as the ratio of firefly luciferase units to Renilla luciferase units. The reporter assay for each tested chemical was carried out in triplicates by setting up four-replicate wells per chemical concentration in each independent experiment. The transcriptional activation potential of AHR by each chemical was evaluated from the respective dose–response curve. EC50 values were also calculated from the dose–response curves using Prism 4 (GraphPad, USA). FICZ treatment of C3H/lpr and MRL/lpr mice FICZ was purchased from Sigma-Aldrich (USA) and dissolved in DMSO. Ten-week old C3H/lpr and MRL/lpr mice were intraperitoneally injected with 100 μg/kg body weight of FICZ solution. After 2 h of exposure, FICZtreated mice were killed using ether anesthesia and the livers were immediately collected. Protein extraction and pull‑down assay Nuclear extracts of mouse livers were prepared using a Nuclear Extraction Kit (Active Motif, USA), according to the supplier’s protocol. The XRE cluster site (620 bp) of mouse CYP1A1 5′-flanking region was amplified by KOD-Plus-Ver.2 using the biotinylated primers: forward, GTTAGTTAGGAACAGGTTGA; reverse, ATGGTG GAGGAAAGGGTGGA. Following the purification of biotinylated PCR products by QIAquick PCR Purification Kit, they were incubated for 12 h at 4 °C with streptavidinagarose beads contained in Biotinylated Protein Interaction Pull-Down Kit (Pierce, USA). Nuclear extracts were dialyzed in Tris-buffered saline (TBS) containing Complete EDTA-free Protease Inhibitor Cocktail (Roche, Switzerland) and 1 mM dithiothreitol (DTT) for 12 h at 4 °C and were further diluted with TBS to 1 mg/ml of protein concentration. The extracts were incubated with the PCR product-linked streptavidin beads for 12 h at 4 °C. After washing with TBS, the beads were eluted by LDS sample buffer containing 10 mM DTT (Invitrogen, USA). The eluates were then separated by 4–12 % gradient NuPAGE (Invitrogen) and visualized by Coomassie Brilliant Blue R-250 (CBB). Protein identification using MALDI‑TOF/TOF CBB-stained bands were cut and then digested using InGel Tryptic Digestion Kit (Pierce, USA) and Sequence Grade Modified Trypsin (Promega, USA), according to the supplier’s protocol. The tryptic peptides were purified by ZipTip C18 (Millipore, USA) prior to mass spectrometry
13
analysis. The eluates were mixed with saturated α-cyano4-hydroxycinnamic acid (CHCA) and then spotted on 384-well plates. Mass spectra of peptides were acquired in a positive reflector ion mode on matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF/TOF) 5800 (AB SCIEX, USA). For protein identification, the acquired MS/MS spectra were subjected to Mascot MS/MS Ion Search (Matrix Science, UK). Immunoprecipitation assay To investigate the interaction between hnRNP-A2/B1 and AHR in the MRL/lpr and C3H/lpr mouse strains, nuclear fractions of FICZ-treated mouse livers were prepared using a Nuclear Extraction Kit (Active Motif, USA), according to the supplier’s protocol. Immunoprecipitation was performed using ImmunoCruz™ IP/WB Optima F System (sc45043; Santa Cruz Biotechnology, USA), as well as antiAHR (BML-SA210; Enzo Life Sciences, Germany) and anti-hnRNP-A2/B1 (ab31645; Abcam, USA) antibodies. For the preparation of precleared nuclear fraction, 1 ml of nuclear extracts was incubated with 60 μl of the suspended (25 % v/v) preclearing matrix in a 1.5-ml microcentrifuge tube at 4 °C on a rotator for 30 min. After centrifugation at a maximum speed for 30 s at 4 °C, only a supernatant was collected. For the formation of the IP antibody-IP matrix complex, 60 μl of suspended (25 % v/v) IP matrix (beads), 5 μg of IP antibody (anti-AHR or anti-hnRNP-A2/B1 antibody), and 500 μl of TBS were added in a microcentrifuge tube. The mixture was incubated at 4 °C on a rotator for 4 h. After incubation of the IP antibody with the IP matrix, a pellet matrix was obtained via microcentrifugation at maximum speed for 30 s at 4 °C. The pellet matrix was then washed two times with 500 μl of TBS. After the final wash of the IP antibody-IP matrix complex, the precleared supernatant of nuclear fraction was transferred to the pellet matrix, and the mixture was then incubated at 4 °C on a rotator for overnight. After incubation, the mixture was centrifuged at a maximum speed for 30 s at 4 °C to pellet the IP matrix complex. The pellet was washed 4 times with TBS. After final wash, the pellet was resuspended in 50 μl of 2× LDS sample buffer containing 10 mM DTT (Invitrogen). The suspended samples were boiled for 10 min at 70 °C and then subjected to the following Western blot analysis. Western blot analysis Cell extracts were prepared with RIPA buffer (150 mM NaCl, 1 % NP-40, 0.1 % sodium dodecyl sulfate (SDS), 2 mM EDTA, 6 mM Na2HPO4, 4 mM NaH2PO4, 50 mM NaF, 0.2 mM Na3VO4, and 1 × protease inhibitor). Protein concentrations were normalized using the Bradford assay
13
Arch Toxicol
(Bio-Rad Laboratories, USA), and 50 µg of protein was separated by 12 % SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to an activated PVDF membrane. The membranes were incubated with antibodies against AHR (N-19) (sc-8088; Santa Cruz Biotechnology, USA) and V5-tag [E10] (ab53418; Abcam, USA) for the detection of hnRNPA2/B1, as well as with anti-actin (I-19) (sc1616; Santa Cruz Biotechnology, USA). Primary antibody binding to each antigen was detected using the secondary antibodies: donkey anti-goat IgG-HRP (sc-2020; Santa Cruz Biotechnology, USA), goat anti-mouse IgG-HRP (sc-2005; Santa Cruz Biotechnology, USA), or donkey anti-rabbit IgG-HRP (sc-2313; Santa Cruz Biotechnology, USA). Bands on the membrane were visualized using ECL Prime Western Blotting Detection System (Amersham Biosciences).
Results Comparison of amino acid sequences of AHR and ARNT between MRL/lpr and C3H/lpr strains We sequenced full-length AHR cDNAs from C3H/lpr and MRL/lpr mice and found that the AHR cDNAs from both strains have an open reading frame of 849 amino acid residues with a predicted molecular mass of 96 kDa. Comparison of the deduced AHR amino acid sequences indicated that a total of 3 amino acid residues (at positions 348, 375, and 758) were different between the two strains (Fig. S1). The ARNT cDNAs from both strains have an open reading frame of 792 amino acid residues with a predicted molecular mass of 87 kDa and a single amino acid mutation at position 497 (Fig. S2). Comparison of ligand‑dependent transactivation potency of AHRs from MRL/lpr and C3H/lpr strains In vitro luciferase activities mediated by C3H/lpr and MRL/lpr mouse AHRs were measured to evaluate the transactivation potency by the treatment with FICZ and TCDD. We found a dose-dependent induction of the luciferase activity by FICZ and TCDD (Fig. 1). EC50 values of FICZ were estimated to be 0.023 and 0.046 nM for C3H/lpr AHR and MRL/lpr AHR, respectively (Fig. 1a; Table 1), while the TCDD-EC50 values were 0.054 nM for C3H/lpr AHR and 0.23 nM for MRL/lpr AHR (Fig. 1b; Table 1). Comparison of EC50 values indicates that C3H/lpr AHR is more sensitive to TCDD than MRL/lpr AHR, although EC50 values of FICZ are almost similar between the two strains. In addition, the EC50 values of 2,3,4,7,8-pentachlorodibenzo-p-dioxin (PeCDD), 2,3,7,8-tetrachlorodibenzofuran (TCDF), and 2,3,4,7,8-pentachlorodibenzofuran (PeCDF)
Arch Toxicol Fig. 1 Dose–response curves of FICZ and TCDD for the induction of luciferase activity from a firefly luciferase reporter gene, mediated by C3H/lpr or MRL/lpr AHR. Relative luciferase activity levels (firefly/Renilla luciferase activity ratio) are plotted against the concentration of FICZ or TCDD. The bars represent mean ± SEM of the responses from three independent experiments (replicates/experiment, n = 4)
Table 1 EC50 values (mean ± standard deviation) of FICZ and TCDD for in vitro AHR-mediated transactivation in the presence or absence of hnRNP-A2/B1 Chemical
FICZ TCDD
Strain
EC50 (nM) No hnRNP-A2/B1
hnRNP-A2
hnRNP-A2b
C3H/lpr
0.023 (±0.005)
0.007 (±0.0007)*
0.009 (±0.001)*
MRL/lpr
0.046 (±0.003)
0.017 (±0.0004)*
0.021 (±0.001)
C3H/lpr
0.054 (±0.009)
0.076 (±0.0007)
0.15 (±0.014)
MRL/lpr
0.23 (±0.014)
0.026 (±0.001)*
1.25 (±0.007)*
Relative luciferase activity (firefly/Renilla luciferase activity ratio) is plotted against each concentration. Data are from three independent experiments (replicates/experiment, n = 4). EC50 values of FICZ and TCDD were calculated from the respective dose–response curves using Prism 4 (GraphPad, USA). Student’s t-test was carried out to detect the statistical difference in EC50 values. * Statistical difference (p
MOLECULAR TOXICOLOGY
Novel role of hnRNP‑A2/B1 in modulating aryl hydrocarbon receptor ligand sensitivity See‑Wun Cho · Ken‑ichi Suzuki · Yoshiaki Miura · Tatsuhiko Miyazaki · Masato Nose · Hisato Iwata · Eun‑Young Kim
Received: 31 March 2014 / Accepted: 25 August 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract The aryl hydrocarbon receptor (AHR) is responsible for susceptibility to its ligand-dependent responses. However, the effect of non-AHR factors is less clear. To explore the non-AHR factors, we used two mouse strains with different AHR genetic variants, namely C3H/lpr and MRL/lpr strains with Ala and Val as the 375th amino acid residue, respectively. To assess the contribution of AHR alone, COS-7 cells transiently expressing AHR from each strain were treated with 6-formylindolo[3,2-b]
Electronic supplementary material The online version of this article (doi:10.1007/s00204-014-1352-1) contains supplementary material, which is available to authorized users. S.-W. Cho · E.-Y. Kim (*) Department of Life and Nanopharmaceutical Science, Kyung Hee University, Hoegi Dong, Dongdaemun‑Gu, Seoul 130‑701, Korea e-mail: [email protected] S.-W. Cho · E.-Y. Kim Department of Biology, Kyung Hee University, Hoegi Dong, Dongdaemun‑Gu, Seoul 130‑701, Korea K. Suzuki Graduate School of Science, Hiroshima University, Hiroshima, Japan Y. Miura · H. Iwata Center for Marine Environmental Studies (CMES), Ehime University, Matsuyama, Japan T. Miyazaki · M. Nose Department of Pathogenomics, Ehime University Graduate School of Medicine, Toon, Japan Present Address: M. Nose Department of Histopathology, Tohoku University Graduate School of Medicine, Sendai, Japan
carbazole (FICZ) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), and xenobiotic-responsive element (XRE)-driven reporter gene activities were measured. FICZ-EC50 values for the C3H/lpr and MRL/lpr AHR-mediated transactivation were 0.023 and 0.046 nM, respectively, indicating a similar susceptibility in both AHR genotypes. In contrast, C3H/lpr AHR was fourfold more sensitive to TCDD than MRL/lpr AHR. By a pull-down assay using a XRE-containing PCR product as bait and the hepatic nuclear extracts of both FICZ-treated mouse strains, we identified two interacting proteins as heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP-A2) and its splicing variant (hnRNPA2b). Immunoprecipitation assays demonstrated the AHR interaction with hnRNP-A2/B1. When hnRNP-A2 was coexpressed with the MRL/lpr or C3H/lpr AHR in COS-7, FICZ treatment decreased EC50 to about threefold in both AHR genotypes, compared with EC50 in AHR alone. Similarly, hnRNP-A2b co-expression also lowered the FICZEC50 values. In TCDD-treated COS-7, responses depended on the AHR genotype; while no change in TCDD-EC50 was observed for C3H/lpr AHR when hnRNP-A2 was co-expressed, the value was reduced to nearly tenfold for MRL/lpr AHR. Co-transfection with hnRNP-A2b attenuated the AHR sensitivity to TCDD. In conclusion, the hnRNP-A2/B1 interacting with AHR may be a modulator of the AHR ligand sensitivity. Keywords AHR · Sensitivity · FICZ · TCDD · C3H/lpr · MRL/lpr · hnRNP-A2/B1
Introduction The aryl hydrocarbon receptor (AHR), a member of the basic helix-loop-helix (bHLH) and Per-Arnt-Sim (PAS)
13
superfamily, is a ligand-dependent transcription factor. The PAS-B domain encompasses a ligand-binding site (Ge and Elferink 1998; Swanson and Bradfield 1993) and the C-terminal transactivation domain (TAD) interacts with several transcriptional cofactors (Rowlands et al. 1996). In the absence of ligands, AHR forms a complex with heat shock proteins (Hsp90s), a small protein (p23), and an immunophilin-like protein (XAP2) in the cytoplasm (Perdew 1988; Petrulis et al. 2000). Upon ligand binding, the AHR complex translocates into the nucleus, forms a heterodimer with AHR nuclear translocator (ARNT), and binds to a xenobiotic-responsive element (XRE; 5′-TNGCGGTG-3′) in the promoter region of the AHR gene battery which includes cytochrome P450 (CYP) 1A1 (Denison et al. 1988). Following the recruitment of a large number of co-activator/ co-repressor complexes, AHR consequently mediates various biochemical and toxic effects (Esser et al. 2009; Fernandez-Salguero et al. 1996; Mimura et al. 1997; Mandal 2005; Vorderstrasse et al. 2001). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and 6-formylindolo[3,2-b]carbazole (FICZ) are representative exogenous and endogenous AHR ligands, respectively, and are efficient inducers of the AHR gene battery (Safe 1990; Mukai and Tischkau 2007; Wincent et al. 2009). TCDD, one of the dioxin-like congeners including polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls, is the most potent AHR ligand (Safe 1990) and activates AHR to induce endocrine disruption, tumor promotion, and teratogenesis (Kerkvliet 2002; Taylor and Zhulin 1999). FICZ is formed from tryptophan in the UV-exposed skin and also binds AHR with high affinity (Fritsche et al. 2007; Rannug et al. 1987, 1995). AHR has recently emerged as a critical physiological regulator of innate and adaptive immune responses (Kerkvliet 2009; Veldhoen et al. 2008; Quintana et al. 2008; Kimura et al. 2008). AHR-mediated response to ligand exposure demonstrates strain- and species-specific differences. For example, the genetic polymorphism of AHRs in the C57BL/6 and DBA/2 mouse strains shows differential dioxin sensitivity. The 375th amino acid in the AHR ligand-binding site is Ala and Val in the C57BL/6 and DBA/2 strains, respectively. This difference results in lowered sensitivity to TCDD in DBA/2 compared to the C57BL/6 strain (Ema et al. 1994; Okey et al. 1989; Poland et al. 1994) and suggests that the 375th amino acid residue of the mouse AHR is one of the critical determinants for TCDD sensitivity. Nevertheless, factors other than AHR have also been implicated in this susceptibility (Kawakami et al. 2006; Moffat et al. 2010). Here, we hypothesize that the differences in the coactivators that interact with AHR may potentially modify ligand sensitivity, since the over-expression of some co-activators can enhance the transcriptional activity of
13
Arch Toxicol
several nuclear receptors in response to their respective ligands (Beischlag et al. 2002). Steroid receptor coactivator 1 (SRC1) and p300 as co-activators of AHR are known to play a critical role in the transcription of the AHR gene battery by forming a complex with general transcription factors (Hestermann and Brown 2003; Kollara and Brown 2006; Zhang et al. 2008). However, factors that affect the ligand sensitivity in AHR signaling pathways remain unexplored. Thus, identifying these factors is important for understanding the mechanisms underlying this susceptibility. In this study, we used two mouse models with different AHR genotypes, a C3H/lpr and an MRL/lpr strains with Ala and Val as the 375th amino acid residue in the AHR gene, respectively, to explore the non-AHR factors involved in AHR ligand sensitivity. Both strains have a single mutated gene, lymphocyte proliferation (lpr), that encodes the deletion mutant of Fas, resulting in the immunological disruption of Fas-mediated apoptosis (WatanabeFukunaga et al. 1992). Nevertheless, the MRL/lpr strain develops many symptoms of autoimmune diseases, while C3H/lpr strain shows less severe phenotypes (Nose et al. 1989), indicating that the onset of autoimmune diseases is controlled by a polygene network (Nose 2007). Initially, to assess the contribution of AHR alone, XRE-driven reporter gene activities were measured in COS-7 cells transiently expressing AHR from C3H/lpr or MRL/lpr strains treated with graded concentrations of FICZ or TCDD. Secondary, a pull-down assay was carried out using XRE as a bait for the MRL/lpr and C3H/lpr liver nuclear extracts. The highly expressed proteins in the liver nuclear extracts were identified as heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP-A2) and its splicing variant (hnRNP-A2b), which are well-known autoantigens involved in autoimmune diseases (Steiner et al. 1996; Cho et al. 2012). The hnRNP was transiently expressed with the MRL/lpr or C3H/lpr AHR into COS-7 treated with FICZ or TCDD, and the EC50 was compared with the EC50 value obtained from AHR expression alone to examine whether the hnRNP-A2/B1 is a critical factor that can regulate the sensitivity to TCDD and FICZ.
Materials and methods Animals C3H/lpr and MRL/lpr strains of mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). All mice used in this study were bred and housed in the animal facility of the Ehime University under a pathogen-free and climatecontrolled environment with 12-h light/dark cycles. All the animal studies were performed according to the guidelines
Arch Toxicol
set by the relevant animal regulatory committee of the Ehime University. Plasmid constructs The pcDNA™3.1/Zeo(+) expression vectors (Invitrogen) containing MRL/lpr AHR, C3H/lpr AHR, MRL/lpr ARNT, and C3H/lpr ARNT were constructed using the method described by Kim et al. (2011). The cDNA clone of hnRNP-A2/B1 (hnRNP-A2 and hnRNP-A2b) was amplified with primers containing a CACC site using KOD-Plus-Ver. 2 (Toyobo, Japan). The specific primers were designed for PCR using the National Center for Biotechnology Information (NCBI) genome database (forward, CACCATGGAG AGAGAAAAGGAA and reverse, ATATCTGCTCCTTCC ACCATA). PCR products were purified by QIAquick PCR Purification Kit (Qiagen) and ligated into the pcDNA™3.1 Directional TOPO® vector (Invitrogen). The reporter vector, pGudLuc 6.1, that contains the firefly luciferase reporter gene regulated by a 484-bp fragment with four XREs in the 5′-upstream region of mouse CYP1A1, was kindly provided by Dr. Michael S. Denison (University of California, Davis, USA). The XRE-containing fragment was cloned into the pGL4.10 firefly luciferase reporter vector (Promega). Luciferase reporter assay The constructed reporter plasmids were transfected into an African green monkey kidney cell line, COS7. The cells were seeded at a concentration of 6 × 104 cells/well in 24-well plates and were maintained in Roswell Park Memorial Institute (RPMI) medium-1640 (Hyclone, USA) supplemented with 10 % fetal bovine serum (Hyclone) at 37 °C under 5 % CO2. Lipofectamine LTX (Invitrogen) was used for transfections according to the manufacturer’s instructions. Twenty nanograms of the mouse CYP1A1 5′-flanking region cloned into pGL4.10 firefly luciferase reporter vector was co-transfected with 5 ng of pcDNA™3.1/Zeo(+)_AHR, 50 ng of pcDNA™3.1/Zeo(+)_ARNT expression vector, and 5 ng of pGL4.74 [hRluc/TK] Renilla luciferase transfection control vector (Promega). In addition, 20 ng of the expression vector of hnRNP-A2/B1 (hnRNP-A2 or hnRNP-A2b) was also transfected to examine the effect of this protein on the AHR ligand sensitivity. After 5 h of the transfection, the cells were treated with graded concentrations of FICZ or TCDD dissolved in dimethyl sulfoxide (DMSO; Sigma). The test solutions of individual chemicals were prepared by serially diluting the stock solutions. Following ligand treatment for 18 h, luciferase activity as a measure of the response of C3H/lpr and MRL/lpr mouse AHRs to the tested chemicals was measured with the
Dual-Luciferase Assay Kit (Promega) and a Multi-Mode Microplate Reader (BioTek Synergy2). The final luminescence values were expressed as the ratio of firefly luciferase units to Renilla luciferase units. The reporter assay for each tested chemical was carried out in triplicates by setting up four-replicate wells per chemical concentration in each independent experiment. The transcriptional activation potential of AHR by each chemical was evaluated from the respective dose–response curve. EC50 values were also calculated from the dose–response curves using Prism 4 (GraphPad, USA). FICZ treatment of C3H/lpr and MRL/lpr mice FICZ was purchased from Sigma-Aldrich (USA) and dissolved in DMSO. Ten-week old C3H/lpr and MRL/lpr mice were intraperitoneally injected with 100 μg/kg body weight of FICZ solution. After 2 h of exposure, FICZtreated mice were killed using ether anesthesia and the livers were immediately collected. Protein extraction and pull‑down assay Nuclear extracts of mouse livers were prepared using a Nuclear Extraction Kit (Active Motif, USA), according to the supplier’s protocol. The XRE cluster site (620 bp) of mouse CYP1A1 5′-flanking region was amplified by KOD-Plus-Ver.2 using the biotinylated primers: forward, GTTAGTTAGGAACAGGTTGA; reverse, ATGGTG GAGGAAAGGGTGGA. Following the purification of biotinylated PCR products by QIAquick PCR Purification Kit, they were incubated for 12 h at 4 °C with streptavidinagarose beads contained in Biotinylated Protein Interaction Pull-Down Kit (Pierce, USA). Nuclear extracts were dialyzed in Tris-buffered saline (TBS) containing Complete EDTA-free Protease Inhibitor Cocktail (Roche, Switzerland) and 1 mM dithiothreitol (DTT) for 12 h at 4 °C and were further diluted with TBS to 1 mg/ml of protein concentration. The extracts were incubated with the PCR product-linked streptavidin beads for 12 h at 4 °C. After washing with TBS, the beads were eluted by LDS sample buffer containing 10 mM DTT (Invitrogen, USA). The eluates were then separated by 4–12 % gradient NuPAGE (Invitrogen) and visualized by Coomassie Brilliant Blue R-250 (CBB). Protein identification using MALDI‑TOF/TOF CBB-stained bands were cut and then digested using InGel Tryptic Digestion Kit (Pierce, USA) and Sequence Grade Modified Trypsin (Promega, USA), according to the supplier’s protocol. The tryptic peptides were purified by ZipTip C18 (Millipore, USA) prior to mass spectrometry
13
analysis. The eluates were mixed with saturated α-cyano4-hydroxycinnamic acid (CHCA) and then spotted on 384-well plates. Mass spectra of peptides were acquired in a positive reflector ion mode on matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF/TOF) 5800 (AB SCIEX, USA). For protein identification, the acquired MS/MS spectra were subjected to Mascot MS/MS Ion Search (Matrix Science, UK). Immunoprecipitation assay To investigate the interaction between hnRNP-A2/B1 and AHR in the MRL/lpr and C3H/lpr mouse strains, nuclear fractions of FICZ-treated mouse livers were prepared using a Nuclear Extraction Kit (Active Motif, USA), according to the supplier’s protocol. Immunoprecipitation was performed using ImmunoCruz™ IP/WB Optima F System (sc45043; Santa Cruz Biotechnology, USA), as well as antiAHR (BML-SA210; Enzo Life Sciences, Germany) and anti-hnRNP-A2/B1 (ab31645; Abcam, USA) antibodies. For the preparation of precleared nuclear fraction, 1 ml of nuclear extracts was incubated with 60 μl of the suspended (25 % v/v) preclearing matrix in a 1.5-ml microcentrifuge tube at 4 °C on a rotator for 30 min. After centrifugation at a maximum speed for 30 s at 4 °C, only a supernatant was collected. For the formation of the IP antibody-IP matrix complex, 60 μl of suspended (25 % v/v) IP matrix (beads), 5 μg of IP antibody (anti-AHR or anti-hnRNP-A2/B1 antibody), and 500 μl of TBS were added in a microcentrifuge tube. The mixture was incubated at 4 °C on a rotator for 4 h. After incubation of the IP antibody with the IP matrix, a pellet matrix was obtained via microcentrifugation at maximum speed for 30 s at 4 °C. The pellet matrix was then washed two times with 500 μl of TBS. After the final wash of the IP antibody-IP matrix complex, the precleared supernatant of nuclear fraction was transferred to the pellet matrix, and the mixture was then incubated at 4 °C on a rotator for overnight. After incubation, the mixture was centrifuged at a maximum speed for 30 s at 4 °C to pellet the IP matrix complex. The pellet was washed 4 times with TBS. After final wash, the pellet was resuspended in 50 μl of 2× LDS sample buffer containing 10 mM DTT (Invitrogen). The suspended samples were boiled for 10 min at 70 °C and then subjected to the following Western blot analysis. Western blot analysis Cell extracts were prepared with RIPA buffer (150 mM NaCl, 1 % NP-40, 0.1 % sodium dodecyl sulfate (SDS), 2 mM EDTA, 6 mM Na2HPO4, 4 mM NaH2PO4, 50 mM NaF, 0.2 mM Na3VO4, and 1 × protease inhibitor). Protein concentrations were normalized using the Bradford assay
13
Arch Toxicol
(Bio-Rad Laboratories, USA), and 50 µg of protein was separated by 12 % SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to an activated PVDF membrane. The membranes were incubated with antibodies against AHR (N-19) (sc-8088; Santa Cruz Biotechnology, USA) and V5-tag [E10] (ab53418; Abcam, USA) for the detection of hnRNPA2/B1, as well as with anti-actin (I-19) (sc1616; Santa Cruz Biotechnology, USA). Primary antibody binding to each antigen was detected using the secondary antibodies: donkey anti-goat IgG-HRP (sc-2020; Santa Cruz Biotechnology, USA), goat anti-mouse IgG-HRP (sc-2005; Santa Cruz Biotechnology, USA), or donkey anti-rabbit IgG-HRP (sc-2313; Santa Cruz Biotechnology, USA). Bands on the membrane were visualized using ECL Prime Western Blotting Detection System (Amersham Biosciences).
Results Comparison of amino acid sequences of AHR and ARNT between MRL/lpr and C3H/lpr strains We sequenced full-length AHR cDNAs from C3H/lpr and MRL/lpr mice and found that the AHR cDNAs from both strains have an open reading frame of 849 amino acid residues with a predicted molecular mass of 96 kDa. Comparison of the deduced AHR amino acid sequences indicated that a total of 3 amino acid residues (at positions 348, 375, and 758) were different between the two strains (Fig. S1). The ARNT cDNAs from both strains have an open reading frame of 792 amino acid residues with a predicted molecular mass of 87 kDa and a single amino acid mutation at position 497 (Fig. S2). Comparison of ligand‑dependent transactivation potency of AHRs from MRL/lpr and C3H/lpr strains In vitro luciferase activities mediated by C3H/lpr and MRL/lpr mouse AHRs were measured to evaluate the transactivation potency by the treatment with FICZ and TCDD. We found a dose-dependent induction of the luciferase activity by FICZ and TCDD (Fig. 1). EC50 values of FICZ were estimated to be 0.023 and 0.046 nM for C3H/lpr AHR and MRL/lpr AHR, respectively (Fig. 1a; Table 1), while the TCDD-EC50 values were 0.054 nM for C3H/lpr AHR and 0.23 nM for MRL/lpr AHR (Fig. 1b; Table 1). Comparison of EC50 values indicates that C3H/lpr AHR is more sensitive to TCDD than MRL/lpr AHR, although EC50 values of FICZ are almost similar between the two strains. In addition, the EC50 values of 2,3,4,7,8-pentachlorodibenzo-p-dioxin (PeCDD), 2,3,7,8-tetrachlorodibenzofuran (TCDF), and 2,3,4,7,8-pentachlorodibenzofuran (PeCDF)
Arch Toxicol Fig. 1 Dose–response curves of FICZ and TCDD for the induction of luciferase activity from a firefly luciferase reporter gene, mediated by C3H/lpr or MRL/lpr AHR. Relative luciferase activity levels (firefly/Renilla luciferase activity ratio) are plotted against the concentration of FICZ or TCDD. The bars represent mean ± SEM of the responses from three independent experiments (replicates/experiment, n = 4)
Table 1 EC50 values (mean ± standard deviation) of FICZ and TCDD for in vitro AHR-mediated transactivation in the presence or absence of hnRNP-A2/B1 Chemical
FICZ TCDD
Strain
EC50 (nM) No hnRNP-A2/B1
hnRNP-A2
hnRNP-A2b
C3H/lpr
0.023 (±0.005)
0.007 (±0.0007)*
0.009 (±0.001)*
MRL/lpr
0.046 (±0.003)
0.017 (±0.0004)*
0.021 (±0.001)
C3H/lpr
0.054 (±0.009)
0.076 (±0.0007)
0.15 (±0.014)
MRL/lpr
0.23 (±0.014)
0.026 (±0.001)*
1.25 (±0.007)*
Relative luciferase activity (firefly/Renilla luciferase activity ratio) is plotted against each concentration. Data are from three independent experiments (replicates/experiment, n = 4). EC50 values of FICZ and TCDD were calculated from the respective dose–response curves using Prism 4 (GraphPad, USA). Student’s t-test was carried out to detect the statistical difference in EC50 values. * Statistical difference (p
