Abstract
purpose. Cytochrome P450 (CYP) enzymes metabolize endogenous compounds such as steroid hormones, fatty acids, and xenobiotics, including drugs and carcinogens. Expression of CYP enzymes in ocular tissues is poorly known. However, mutations in the CYP1B1 gene have been linked to congenital glaucoma. The aim of the present study was to investigate the expression and regulation of cytochrome P450 enzymes in a human nonpigmented ciliary epithelial cell line.
methods. Expression of mRNAs for major xenobiotic metabolizing CYPs in families 1–3 and regulatory factors involved in the induction of CYPs was studied using reverse transcriptase–polymerase chain reaction. For induction studies, the cells were treated with dexamethasone or 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) for 24 hours. RNA and immunoblotting analysis were used to study CYP induction. Transcriptional regulation of CYP1B1 gene was studied by transient transfection of reporter gene constructs.
results. mRNAs of CYP1A1, CYP1B1, and CYP2D6 and of the regulatory factors aryl hydrocarbon receptor (AHR), aryl hydrocarbon receptor nuclear translocator, and glucocorticoid receptor were expressed in the human nonpigmented ciliary epithelial cell line. CYP1B1 mRNA was strongly and dose dependently induced by TCDD. CYP1B1 protein was detected only after TCDD treatment of the human nonpigmented ciliary epithelial cells. CYP1B1 promoter was activated by TCDD. The major drug-metabolizing enzymes CYP1A2, CYP2Cs, and CYP3As were not detected in these cells, and dexamethasone treatment had no effect on CYP expression.
conclusions. TCDD potently induces CYP1B1 mRNA in human nonpigmented ciliary epithelial cells, suggesting the involvement of an AHR-mediated pathway in the regulation of ciliary CYP1B1 expression.
Cytochrome P450 (CYP) enzymes are a superfamily of heme-containing proteins important in oxidative metabolism.
1 2 CYP enzymes are responsible for the metabolism of endogenous substances such as steroid hormones and fatty acids and of xenobiotics including drugs and carcinogens.
1 Fifty-seven cytochrome P450 genes have been identified in the human genome, whereas the CYP1, CYP2, and CYP3 families contribute primarily to the metabolism of drugs.
2 CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4 are the most important CYPs responsible for the metabolism of most drugs.
1
CYP enzymes are most highly expressed in the liver. However, they are known to be present in many other tissues, such as in small intestine, lung, and kidney.
3 The expression and regulation of CYP enzymes in ocular tissues is poorly known. High expression of CYP1B1 mRNA has been reported in the human eye in the nonpigmented epithelium of the ciliary body and in the iris.
4 Very low levels of other CYP expression have recently been detected in human ocular tissues.
5
Several antiglaucoma medicines and many other ophthalmic drugs used topically are substrates, inhibitors, or inducers of CYP enzymes. For example, the anti-inflammatory and analgesic agent diclofenac is a substrate of CYP2C9.
6 Dexamethasone (DEX) is a CYP3A4 substrate and a potent inducer of CYP3A4 activity.
7 8 9 One widely used glaucoma medicine, the β-adrenergic blocking agent timolol, is a CYP2D6 substrate, and the antibiotic chloramphenicol is an inhibitor of CYP3A4.
10 11 However, scant information is available regarding whether the major drug-metabolizing CYPs are expressed in ocular tissues. The presence and activity of CYP enzymes may significantly affect the efficacy and safety of ophthalmic drugs.
Among the xenobiotic-metabolizing CYP enzymes in families CYP1 to CYP3,
CYP1B1 is the only member directly linked to a disease. CYP1B1 has been found to be a causative disease gene for primary congenital glaucoma and a modifier gene, or rarely a causative gene, in primary open angle glaucoma.
12 Furthermore,
CYP1B1 null mice exhibit abnormalities in their ocular drainage structure and trabecular meshwork, which are similar to those in patients with primary congenital glaucoma.
12
An important characteristic of many xenobiotic-metabolizing CYP enzymes is adjustment to chemical environment by induction. This induction is mediated primarily by intracellular xenobiotic-sensing receptors, including pregnane X receptor (PXR), constitutive androstane receptor (CAR), and aryl hydrocarbon receptor (AHR). AHR functions as a heterodimer with aryl hydrocarbon receptor nuclear translocator (ARNT). Furthermore, glucocorticoid receptor (GR) may also be involved in some induction phenomena.
13 DEX induces the expression of CYP3A6 enzyme in the rabbit lacrimal gland.
14 Almost nothing is known regarding the regulation of CYP enzymes in the human eye.
The aim of the present study was to investigate the expression and regulation of CYP enzymes in a human nonpigmented ciliary epithelial cell line. We show that few CYP forms are expressed in this cell line. More important, we show for the first time that CYP1B1 is inducible in a ciliary cell line by an AHR receptor ligand, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD).
The nonpigmented human ciliary cell line, developed from primary cultures of nonpigmented human ciliary epithelium,
15 16 was a generous gift from Miguel Coca-Prados (Yale University, New Haven, CT). This cell line has many characteristics of normal nonpigmented human ciliary epithelium. The nonpigmented human ciliary cells (passages 5–10) were cultured in Dulbecco modified Eagle medium (Sigma-Aldrich, Steinheim, Germany) supplemented with 10% (vol/vol) fetal bovine serum (Invitrogen, Paisley, Scotland, UK) and 1% penicillin-streptomycin. Cells were cultured in standard conditions at 37°C, 5% CO
2, and saturated humidity.
Nearly confluent cells were treated with DEX (Sigma-Aldrich) and TCDD (National Cancer Institute Chemical Carcinogen Repository, Bethesda, MD) for 24 hours in serum-free Dulbecco modified Eagle medium. Inducers were dissolved in dimethyl sulfoxide (DMSO). Concentrations used in screening studies (100 nM for DEX, 10 nM for TCDD) were selected based on our previous studies showing that these concentrations are sufficient to cause strong GR- and AHR-mediated inductions.
17 18 Control cells were incubated with an equal volume of DMSO for 24 hours. After 24-hour exposure, the cells were collected and used for RNA or protein extraction.
Nearly confluent cells were washed twice with phosphate-buffered saline and were collected into a small volume of lysis buffer for RNA extraction. Total RNA was extracted from cultured ciliary epithelial cells with reagent (TRI reagent; Sigma-Aldrich) according to the manufacturer’s protocol. The purity and concentration of the isolated RNA was measured spectrophotometrically, and mRNA was stored at −70°C until cDNA synthesis.
Complementary DNA (cDNA) was synthesized with a cDNA synthesis kit (First-Strand; Amersham Biosciences, Buckinghamshire, UK). One microgram RNA was used in each synthesis. Negative control was prepared by making a cDNA synthesis reaction without RNA. The cDNAs and the negative control were stored at −20°C until PCR amplifications were performed.
Each PCR reaction contained 1 μL cDNA (of a total 15 μL), 2.5 μL of 10× PCR reaction buffer (Finnzymes, Espoo, Finland), 0.5 μL dNTP (10 mM) reaction mixture (Finnzymes), 1 μL sense and antisense primer (10 μM), 0.5 μL DNA polymerase (DyNAzyme II, 2 U/mL; Finnzymes), and water to a final volume of 25 μL. The primers (Sigma-Genosys, Cambridge, UK) used in the study are presented in
Table 1 . The CYP2C8–19 primers detect all CYP2C8, CYP2C9, CYP2C18, and CYP2C19.
PCR cycles were conducted as follows: 1 minute at 94°C, then 35 cycles for 10 seconds at 94°C, 10 seconds at annealing temperature, 30 seconds at 72°C, and a final extension of 5 minutes at 72°C. The annealing temperature used for each primer pair is shown in
Table 1 . Every series of PCR reactions contained a negative control to exclude contamination and a human liver cDNA sample as positive control. After PCR amplification, 8 μL reaction mixture was electrophoresed into 0.5% agarose gel and stained with ethidium bromide. All PCR reactions were repeated at least twice.
Agarose gel was prepared with low melting point agar (Agarose Prep; Amersham Biosciences AB, Uppsala, Sweden), and the CYP2D6 RT-PCR products were run on the gel. The three most prominent bands were extracted from the gel with an extraction kit (Qiaquick Gel Extraction Kit; Qiagen GmbH, Hilden, Germany) according to the manufacturer’s protocol. Purified DNA products were then subjected to direct sequencing (Biocenter Oulu DNA Sequencing Core Facility, Oulu, Finland).
Five micrograms of the total RNA was electrophoretically resolved and transferred onto a nylon membrane (Hybond-N
+; Amersham Biosciences, Little Chalfont, UK). RNA was fixed by ultraviolet light cross-linking, and the membrane was hybridized with [α
32P] dCTP-labeled probes. CYP1A1 and CYP1B1 cDNA probes were prepared as previously described
19 ; the 18S probe was kindly provided by Heikki Ruskoaho (University of Oulu, Finland).
Cultured cells were washed and collected into phosphate-buffered saline. The cells were broken by sonication and centrifuged. Sixty micrograms of the 10,000
g supernatant protein was subjected to SDS-PAGE (7.5% % polyacrylamide). CYP1B1 supersomes (0.1 μg; BD Biosciences, San Jose, CA) were used as recombinant control. Proteins were transferred electrophoretically to a polyvinylidene difluoride transfer membrane (Immobilon; Millipore, Bedford, MA) in a semidry transfer cell (Trans-Blot SD; Bio-Rad Laboratories Inc., Hercules, CA) at 15 V for 30 minutes. The membrane was blocked in 5% milk in Tris-buffered saline/Tween 20 buffer overnight. The sheet was then incubated for 1 hour with an antibody raised against a human CYP1B1 pentapeptide
20 (1:1000 dilution) and for 1 hour with a secondary peroxidase-conjugated goat anti–rabbit antibody (Sigma-Aldrich, St. Louis, MO) (1:25,000 dilution). After washing, the immunoreactive bands were visualized with chemiluminescent peroxidase substrate (Sigma-Aldrich).
Preparation and site-directed mutagenesis of the
CYP1B1 5′-luciferase reporter constructs have been described previously.
21 pGL3-basic plasmids contained the human
CYP1B1 gene 5′-flanking region from −2299 to +25, from −910 to +25, and from −852 to +25. Plasmids −910/XRE3 mt and −910/XRE2 mt have mutations in AHR-binding sites. Ciliary epithelial cells were seeded into a 24-well plate and cultured for 24 hours before transient transfection using reagent (Tfx-20; Promega, Madison, WI) according to the manufacturer’s protocol. Cells in each well were transfected with 0.5 μg
CYP1B1/luc reporter gene plasmid and 0.1 μg control reporter plasmid pRL-TK (Promega) in 200 μL medium (Opti-MEM I; Invitrogen). Twenty-four hours after transfection, the cells were treated with 100 nM TCDD of vehicle (DMSO) only. After 24-hour treatment, luciferase activities were measured by the Dual-Luciferase Reporter Assay System (Promega).
Detection of CYP mRNA Expression in Human Nonpigmented Ciliary Epithelial Cell Line
Expression of several CYP genes in families CYP1–3 was studied by an RT-PCR method. The presence of a band of the correct size in agarose gel was regarded as evidence for gene expression. mRNAs of CYP1A1 and CYP1B1 were detected in the human nonpigmented ciliary epithelial cell line and the human liver
(Fig. 1) . mRNAs of CYP1A2, CYP2A6, CYP2B6, CYP2Cs, CYP2E1, CYP3A4, CYP3A5, and CYP3A7 were not detected in the ciliary epithelial cell samples, but all were present in the control liver sample.
RT-PCR for CYP2D6 mRNA produced a single band of correct size for the liver sample, as expected. In contrast, in the ciliary epithelial cell sample, four separate bands were detected
(Fig. 2A) . Of these, band number four was of expected size, but the three other products were larger. To establish the identity of these PCR products, the three most prominent bands (bands 1, 2, 4) were extracted from the gel and sequenced. CYP2D6 sense and antisense primers were designed to target sequences at exons 7 and 9, respectively. Band 4, of similar size as the liver band, represented correctly sliced CYP2D6 mRNA. Band 1 contained the introns between exons 7 to 8 and 8 to 9, and band 2 contained the intron between exons 7 to 8 but not between exons 8 to 9. Band 3 was not sequenced, but its size suggested that it might have contained a product with the intron between exons 8 to 9 but not between exons 7 to 8
(Fig. 2B) .
Detection of CYP-Regulating Receptor mRNA Expression in Human Nonpigmented Ciliary Epithelial Cell Line
Expression of mRNAs for major regulators of CYP induction, including the nuclear receptors CAR, PXR, and GR, and for basic helix-loop-helix factors AHR and ARNT was studied by an RT-PCR method. All these mRNAs could be detected in the liver, as expected. In the ciliary epithelial cell sample GR, AHR, and ARNT were readily detected, whereas CAR and PXR were not present
(Fig. 3) . These findings suggested that the GR- and AHR-mediated regulatory pathways could be functional in the current human nonpigmented ciliary epithelial cell line, whereas the CAR and PXR pathways were not.
Given that the RT-PCR results suggested that GR- and AHR-mediated regulatory pathways might have been functional in the cell line studied, we next treated the cells with DEX and TCDD, which are prototypic ligands for GR and AHR, respectively. Expression of all studied CYP forms was then screened with conventional RT-PCR. Although the qualitative RT-PCR method used was not designed to produce true quantitative data, it could be used to detect substantial differences in mRNA amounts or for the detection of mRNAs detectable only after treatment with an inducing agent. Expression of CYP1A1 and CYP1B1 mRNAs was strongly induced by 24-hour treatment with 10 nM TCDD but not by a similar exposure to 100 nM DEX
(Fig. 4) . Neither TCDD nor DEX induced expression of any other CYP forms studied.
To further establish the induction of CYP1 forms by TCDD in human nonpigmented ciliary epithelial cells by an accurate quantitative method, we carried out a dose-response experiment and detected the mRNAs with Northern blot analysis. The level of CYP1B1 mRNA was increased statistically significantly and dose-dependently by 1 to 100 nM TCDD
(Fig. 5) . The CYP1A1 mRNA could not be detected by Northern blot analysis in any of the samples, indicating that the expression level of CYP1A1 was much lower than that of CYP1B1 in this cell line.
Human nonpigmented ciliary epithelial cells were treated with 10nM TCDD or vehicle (DMSO) only for 24 hours. The presence of CYP1B1 protein was analyzed by immunoblotting with human anti–CYP1B1 antibody. A band similar in size to the recombinant CYP1B1 was detected in the TCDD-treated sample, but no clear protein expression was detected in the untreated cells
(Fig. 6) .
To further study regulatory mechanisms of
CYP1B1 induction by TCDD in human ciliary cells, we transfected
CYP1B1 5′-luciferase reporter constructs into the human ciliary epithelial cell line. TCDD induced
CYP1B1 −2299/+25 promoter fragment–regulated luciferase activity 5.7-fold, indicating transcriptional regulation of the
CYP1B1 gene by TCDD
(Fig. 7) . The current promoter fragment contained eight putative AHR-binding elements; several have been shown to bind AHR in vitro.
21 Previous studies in several cell lines suggested that the region from −910 to −852 containing two AHR-binding elements (XRE2 and XRE3) plays an important role in the AHR-mediated regulation of
CYP1B1.
21 However, in our present study, 5′ deletion of the
CYP1B1 promoter to −910 strongly reduced TCDD induction, indicating that in the human ciliary epithelial cell line, the more upstream AHR sites play the major role in TCDD induction. The 5′ deletions of the promoter or mutations of the two XRE sites had no major effect on the constitutive activity of the CYP1B1 promoter in the human ciliary epithelial cell line studied.
Few reports have been published on the expression pattern of xenobiotic metabolizing CYP enzymes in ocular tissues or in different ocular cell types. Furthermore, almost nothing is known of the regulation of CYP genes in the human eye. In the present study, we characterized the expression profile of the major CYP forms in families CYP1–3 in a nonpigmented ciliary epithelial cell line. Additionally, we elucidated the presence and functionality of the main regulatory pathways involved in the induction of CYP enzymes by chemical compounds.
CYP1A1 and CYP1B1 were found to be expressed in the nonpigmented ciliary epithelial cell line. Both are well known for their ability to metabolize a number of carcinogens, such as polycyclic aromatic hydrocarbons.
22 CYP1A1 and CYP1B1 are only weakly expressed in the human liver but, in contrast, can be found in a number of other tissues (e.g., breast).
22 23 24 25 In ocular tissues, CYP1B1 has previously been detected at mRNA level in the iris, ciliary body, and nonpigmented ciliary epithelial line and at lower levels in the cornea, retinal-pigment epithelium and retina.
26 27 Doshi et al.,
27 using an immunolocalization method, also suggested that CYP1B1 protein is present in the fetal and adult nonpigmented ciliary epithelium. CYP1B1 may have many functions in human ocular tissues, and it has been linked especially to various forms of glaucoma. Congenital glaucoma has been reported to be associated with chromosomal abnormalities in at least 17 different autosomes.
28 In particular, homozygous mutations in the
CYP1B1 gene, which codes CYP1B1 and is located on chromosome 2p22-p21, have been linked to congenital glaucoma.
29
In agreement with previous studies, we could detect a relatively high level of CYP1B1 mRNA expression in the nonpigmented ciliary epithelial line. More important, we showed for the first time the strong induction of CYP1B1 mRNA and protein (and, to a lesser extent, CYP1A1 mRNA) by TCDD in an ocular cell type. TCDD induction of CYP1B1 is mediated by AHR, and AHR may also play a role in constitutive expression of CYP1B1.
30 31 Previous studies in MCF-7 (breast carcinoma), HepG2 (hepatocellular carcinoma), LS-180 (colon carcinoma), and OMC-3 (ovarian carcinoma) cell lines suggested that the region from −910 to −852 containing two AHR binding elements (XRE2 and XRE3) plays an important role in AHR-mediated regulation of
CYP1B1 in these cell lines.
21 In contrast, our current findings indicate that in the nonpigmented ciliary epithelial cells further upstream, AHR binding sites play the predominant role. This suggests cell type specific regulation of
CYP1B1 promoter.
The role of CYP1B1 in the developing eye and in congenital glaucoma is not fully understood.
32 However, it has been suggested that CYP1B1 participates in the development of the iridocorneal angle.
33 In addition,
CYP1B1 has been identified as a modifier gene in primary open-angle glaucoma and rarely as a causative gene in this disorder.
12 Substrates of CYP1B1 include endogenous compounds such as steroids, retinoic acids, and melatonin.
12 Interindividual variability in the metabolism of these or other as yet unidentified substrates may contribute to the pathophysiology of various types of glaucoma. Recently, CYP1B1 was reported to decrease oxidative stress and to promote angiogenesis in retinal endothelial cells.
34 Whether this mechanism could be relevant to iridocorneal angle development remains to be shown. An ability to increase the CYP1B1 expression level could potentially represent an interesting novel treatment strategy. However, use of AHR activation, even locally, is probably limited by toxic effects. The AHR receptor may play an important physiological role during development,
35 and a number of endogenous ligands have been identified for AHR.
36 Whether the AHR-mediated regulation of CYP1B1 plays a significant role during fetal development and contributes to iridocorneal angle development remains an important subject for future studies. Furthermore, exposure to exogenous AHR ligands such as environmental contaminants may disturb physiological regulation of
CYP1B1 in the human eye.
Most CYP forms involved in the metabolism of pharmaceuticals, such as CYP1A2, CYP2Cs, and CYP3A4, were not expressed in the nonpigmented ciliary epithelial cells. This is in line with a recent study in which low levels of CYP mRNAs were observed in the human iris-ciliary body.
5 CYP2D6 was an exception, and some mRNA expression could be detected for this enzyme. CYP2D6 has been reported to be involved in the metabolism of approximately 20% to 30% of all commonly used drugs.
37 38 Recently timolol, which is used to lower intraocular pressure, has been shown to be metabolized mainly by CYP2D6.
11 However, only very low concentrations of timolol metabolites were detected in the human aqueous humor in a few patients after a single dose of ophthalmic timolol.
39 It seems evident that a negligible amount of CYP2D6 protein is expressed in ocular tissues. In the present study, we detected several different splice variants of CYP2D6 in the nonpigmented ciliary epithelial cell line, but not in the liver sample. Previously, Huang et al.
40 41 reported extensive alternative slicing of CYP2D6 in lung and breast tissues. The very low amount of correctly spliced CYP2D6 mRNA may explain the lack of significant CYP2D6 enzyme expression in the eye.
Previously, CYP3A expression has been shown be inducible in the rabbit lacrimal gland by topical treatment with DEX.
14 In the liver, DEX is known to induce CYP3A4 mainly through PXR, which in turn is regulated by GR.
42 However, in fetal hepatocytes and in the lung, DEX appears to regulate members of the CYP3A subfamily directly.
17 43 In the present study, GR, but not PXR, was found to be expressed in the nonpigmented ciliary epithelial cell line. However, DEX did not induce the expression of CYP3A enzymes, possibly because of the lack of PXR or other transcription factors necessary for CYP3A induction or in consequence of repression of
CYP3A genes in this cell type by epigenetic mechanisms.
In conclusion, most drug-metabolizing CYPs were not expressed in the human nonpigmented ciliary epithelial cells. CYP1A1, CYP1B1, and CYP2D6 mRNAs as well as mRNAs for regulatory factors GR, AHR, and ARNT were detected. TCDD induced CYP1B1 expression by transcriptional mechanism, suggesting that an AHR-mediated pathway may play an important role in the regulation of ciliary CYP1B1 expression. AHR activation may thus affect developmental functions of CYP1B1 in the human eye. The human nonpigmented ciliary epithelial cell line used here will offer a useful model to study the regulation of CYP1B1 in the ciliary cell type.
The authors thank Ritva Tauriainen and Päivi Tyni for excellent technical assistance, Miguel Coca-Prados for the nonpigmented human ciliary cell line, and Robert J. Edwards (Section on Experimental Medicine and Toxicology, Imperial College London, London, UK) for the anti–CYP1B1 antibody.