Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 771–775
Expression of 25 hydroxyvitamin D3-1␣-hydroxylase in human endometrial tissue Steffi Becker ∗ , Tim Cordes, Dagmar Diesing, Klaus Diedrich, Michael Friedrich Department of Gynaecology and Obstetrics, University of Schleswig-Holstein, Campus L¨ubeck, Ratzeburger Allee 160, 23562 L¨ubeck, Germany
Abstract 1,25(OH)2 D3 (calcitriol) has been shown to play an important role in cell proliferation, differentiation and immune responsiveness. The enzyme responsible for calcitriol synthesis 25 hydroxyvitamin D3 -1␣-hydroxylase (1␣-OHase) has been reported in many human tissues. The aim of this study was to investigate the expression of 1␣-OHase in gynaecological tissues. Using a highly specific nested touchdown PCR we examined the expression of 1␣-OHase in normal and malignant endometrial tissue and in human endometrial Ishikawa cells. In addition, we analyzed the protein expression of 1␣-OHase by Western blot. The expression of 1␣-OHase in normal and malignant endometrial tissue and Ishikawa cells was detected and splice variants of the enzyme in Ishikawa cells were identified. These data suggest an alternative splicing of 1␣-OHase in malignant endometrial tissue and cells. We postulate that the expression of 1␣-OHase gene variants may contribute to the antiproliferative effects of calcitriol. In conclusion, the modulation of the 1␣-OHase opens up a new target for vitamin D3 related therapies in endometrial cancer. © 2006 Elsevier Ltd. All rights reserved. Keywords: Extrarenal 1,25-dihydroxyvitamin D3 ; 1␣-OHase; Alternative splicing; Endometrial cancer
1. Introduction The active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2 D3 , calcitriol] plays an integral role in maintaining calcium and phosphate homeostasis, immune responsiveness, cell growth and differentiation [1–3]. In general, the effects of calcitriol on target cells are mediated via binding to the nuclear vitamin D receptor, a ligand-dependent transcription factor but there is an increasing evidence for the involvement of other pathways mediated by a proposed membrane receptor [4,5]. The synthesis of calcitriol from the major circulating form of vitamin D, 25 hydroxyvitamin D3 is catalyzed by 25 hydroxyvitamin D3 -1␣-hydroxylase (1␣-OHase), a mitochondrial cytochrome P450 enzyme in the kidney [6]. However, extrarenal activity of 1␣-OHase expression has been described in a variety of tissues [7] and several extrarenal cells, including keratinocytes, osteoblasts and microglia produce calcitriol from its precursor in vitro ∗
Corresponding author. Tel.: +49 451 500 3604; fax: +49 451 5004760. E-mail address:
[email protected] (S. Becker).
0960-0760/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2006.12.075
[8]. In addition, the expression of 1␣-OHase has been documented in cancer cells (prostate, breast, glioblastoma). As reported for numerous normal and cancer cells, 1,25(OH)2 D3 is an antiproliferative and prodifferentiating hormone, which induces apoptosis and inhibits migration [9–11]. The ability of calcitriol in regulating cancer cell growth makes the hormone a potential target for tumor management. Therefore, the development of new vitamin D analogues with minor calcaemic side effects is an important subject for the pharmaceutical industry [12]. Endometrium, a part of the female reproductive system, is a highly proliferative tissue during the first part of the menstrual cycle. Although the proliferation of endometrial cells is stimulated by high serum levels of estrogens, hyperestrogenism is one of the major risk factors for the development of endometrial hyperplasia and finally, endometrial cancer [13]. Endometrial tumors are one of the most aggressive malignancies in women. Every year half a million incidences are registered world wide. Human endometrial tissues and endometrial cancer cell lines have been shown to express the vitamin D receptor and the
772
S. Becker et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 771–775
effects of calcitriol to promote differentiation of these cells have been reported [14,15]. Thus, endometrium might be a suitable target of antiproliferative effects of calcitriol. Several studies documented that alternative splicing occurs frequently in human cancer cells (e.g. in brain and skin cancer) [16]. Recently, the alternative splicing of cytochrome P450 genes has been reported [17,18]. This might have a biological function in regulating the activity and level of enzymes and play an important role for local vitamin D metabolism in endometrial cancer. In this study, we describe the expression of full length 1␣-OHase in human normal and malignant endometrial tissue and report new splice variants of the enzyme in human endometrial Ishikawa cells.
2. Materials and methods 2.1. Immunohistochemistry Samples of normal and malignant endometrial tissues were obtained from two patients and immediately fixed with 4% formaldehyde. Paraffin embedded tissues sections were dewaxed and rehydrated. Endogenous peroxidase activity was blocked by incubation with 3% H2 O2 /PBS and tissue sections were incubated with rabbit serum (1:80) in PBS to prevent non specific binding of the first antibody for 1 h. Thereafter, primary polyclonal 1␣-OHase antibody (Biologo, Kiel, Germany) was applied in a dilution of 1:500. Primary antibody binding was detected by using a two-step biotin/streptavidin-based antibody detection system employing peroxidase-mediated DAB staining (DAKO, Germany). Finally, tissue sections were counterstained with haematoxylin and mounted in permanent mounting media. The evaluation of immunohistochemical staining was performed semi-quantitatively comparing visual differences of staining intensities in normal and cancer tissues compared to negative controls stained with the secondary antibody. 2.2. Cell culture The human endometrial cancer cell line Ishikawa was purchased from the European Collection of Cell Culture (ECACC no. 99040201, Wiltshire, UK). Cells were maintained in RPMI1640 medium (GIBCO-BRL, Germany) supplemented with 25 mM HEPES, 1% l-Glutamin and 10% foetal bovine serum. 2.3. RNA and poly (A)-RNA isolation Total RNA and mRNA from Ishikawa cells were extracted from several passages of cell culture with TRIZOL (Invitrogen, Germany) and Oligotex mRNA Kit (Qiagen, Germany) according to manufacturer’s instructions. The RNA was quantified spectrophotometrically and its integrity was controlled by 1% agarose gel electrophoresis in MOPS buffer.
2.4. Nested touchdown PCR Prior utilization, RNA was DNaseI (Invitrogen, Germany) treated. First strand cDNA was synthesized with Onmiscript reverse transcriptase (Quiagen, Germany) and oligo-d(T)15 primer (Invitrogen, Germany). The first PCR was performed using primers Sp1aFOR1 (5 -GGAGAAGCGCTTTCTTTCG-3 ) and Sp1aRev3 [5 -TGGGGCAAACCCACTTAATA-3 ) with 10 cycles (2 min at 98 ◦ C, 15 s at 94 ◦ C, 20 s at 68 ◦ C, 4 min at 68 ◦ C, 15 min at 68 ◦ C). The PCR product was purified (Machery-Nagel, Germany) and 5 l template were used for the second PCR using primers HE1 (5 -CAGACCCTCAAGTACGCC-3 ) and Sp1aRev2 (5 -AAACCAGGCTAGGGCAGATT-3 ). This PCR consisted of 12 cycles (30 s at 96 ◦ C, 10 s at 94 ◦ C, 20 s with a touchdown from 68 ◦ C to 62 ◦ C in 0.5 ◦ C intervals, 4 min at 68 ◦ C) followed by 18 cycles (10 s at 64 ◦ C, 20 s at 62 ◦ C, 4 min at 68 ◦ C). The PCR reactions were performed using 2,5 Units RedACCu TaqTM LA DNA polymerase (Sigma, Germany). The obtained PCR products were separated on a 1% agarose gel. 2.5. Cloning and plasmid purification Cloning of 1␣-OHase PCR products in pCR® 4-TOPO vector was carried out using TOPO-TA Cloning Kit for sequencing (Invitrogen, Germany) according to manufacturer’s instructions. To screen Escherichia coli transformants, plasmid DNA from bacterial cultures were isolated using alkaline lysis and column purification (Plasmid Mini Kit, Qiagen, Germany). 2.6. Sequence analysis Sequencing was performed according to manufacturer’s instructions. Plasmid inserts were sequenced with an automated sequencer (MWG Biotech, Germany). The homology search of the obtained sequences was done with Bioedit Sequence Alignment Editor (Isis Pharmaceuticals, USA). 2.7. Western blot analysis Cells from five passages of cell culture were lysed in sample buffer (125 mM Tris, 30% Glycerine, 8% SDS, pH 6.8). Fifteen micrograms of proteins were subjected to 12% SDS–polyacrylamide gel electrophoresis (PAGE) under reducing conditions and transferred to a nitrocellulose membrane (Schleicher Schuell, Germany). After blocking with TBST containing 5% non fat powdered milk, the membranes were incubated with the primary antibody against human 1␣-OHase (Biologo, Germany) in a dilution of 1:2000. The secondary antibody (rabbit anti sheep IgG) conjugated to horseradish peroxidase (Amersham Biosciences, Germany) was added in a dilution of 1:4000. Bands were visualized using the enhanced chemiluminescence
S. Becker et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 771–775
(ECL) detection system (Amersham Biosciences, Germany). The obtained signals were compared to -actin as internal standard.
773
3.2. Expression of 1α-OHase mRNA and protein in Ishikawa cells
To examine the expression of 1␣-OHase in endometrium, we carried out immunohistochemistry with an antibody raised against the human 1␣-OHase. As well in healthy proliferating as in malignant endometrial tissue sections we detected specific signals of the 1␣-OHase expression compared to negative controls (Fig. 1). However, we observed no visible differences of 1␣-OHase expression in healthy and malignant endometrial tissues.
To adequately describe the expression of 1␣-OHase mRNA in human endometrial cancer cells, we used a highly specific approach that combined nested and touchdown PCR. Using specific primers of exon 1 and 9, we detected a variety of PCR products (1.8–4 kb). We confirmed the expression of the normal enzyme (2.4 kb) in Ishikawa cells of five different cell culture passages (Fig. 2A) and additionally splice variants at 1.8 kb, 2.5 kb and 3 kb. In Western blot analysis, using a sheep polyclonal antiserum against the human 1␣-OHase, we detected the 56 kDa band corresponding to the predicted 1␣-OHase protein. Furthermore, we identified single bands at 23 kDa, 37 kDa and between 50 kDa and 56 kDa, suggesting the presence of different 1␣-OHase variants (Fig. 2B). All of the examined passages appear to have a complex, similar expression pattern of the 1␣-OHase gene and protein.
Fig. 1. Expression of 1␣-OHase in normal (A) and malignant (B) human endometrial tissue. Tissue sections of healthy and malignant endometrium were paraffin embedded and incubated with human 1␣-OHase antibody. Biotin/streptavidin-based antibody detection system employing peroxidasemediated DAB staining was used to visualize the binding of the primary antibody. The cells were counterstained with heamatoxylin.
Fig. 2. Expression of 1␣-OHase in human endometrial Ishikawa cells. RNA and proteins were isolated from five different passages of cell culture and analyzed by nested-touchdown PCR (A) and Western blot (B). The 2.4 kb and 56 kDa bands correspond to the full length enzyme.
3. Results 3.1. Detection of 1α-OHase in normal and malignant endometrial tissue
774
S. Becker et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 771–775
Fig. 3. Organization of the normal and alternative spliced 1␣-OHase mRNA in Ishikawa cells. The PCR products were cloned into pCR® 4-TOPO vector and after sequencing aligned to the full length sequence of the enzyme.
These approaches did not allow quantitative measurements due to the high product variety. Additional control experiments were done with antibodies preabsorbed with 50× immunizing peptide to specify of the obtained bands. 3.3. Cloning and identification of 1α-OHase splice variants in Ishikawa cells In order to analyze the detected variants of 1␣-OHase, we subcloned the obtained cDNA of the nested-touchdown PCR in pCR4-TOPO Cloning vector. Restriction analysis of the resulting transformants revealed three different inserts (data not shown). Sequence analysis of the subcloned cDNA and the alignment to the full length enzyme indicated two splice variants of 1␣-OHase and the full length gene. The first variant we identified showed an insertion of intron1 resulting in short protein of 9 kDa. This is due to a termination codon in intron1 (Fig. 3). A further variant lacks exon 3–5 due to a premature stop codon in exon 6 and resulted in a 17 kDa protein. This variant is corresponding to the 1.8 kb PCR product we detected before subcloning (3.2.). These proteins contain neither the ferredoxin in exon 6 nor the haem binding side in exon 8 and therefore result in inactive forms of the enzymes.
4. Discussion In this study, we describe the expression of the 25 hydroxyvitamin D3 converting enzyme 1␣-OHase in healthy and malignant endometrial tissue and in endometrial Ishikawa cells. In addition, our current study is the first that reports splice variants of 1␣-OHase in Ishikawa cells. The expression of the enzyme has been previously shown in many extrarenal tissues and in vitro a variety of nonrenal cells synthesized calcitriol, the active form of vitamin D. Furthermore, the activity of 1␣-OHase was also detected in various cancer cells [7,8]. Alternative splicing frequently occurs in human cancer cells and causes tissue specific variations. As described in skin and malignant glioma cells, many of the detected variants might be inactive [19]. These results are consistent with our data. Cloning of PCR products showed not only the full length 1␣-OHase but also splice variants containing intron 1 or dele-
tion of exon 3, 4 and 5. This causes early stop codons and the mRNA of these variants will not be translated in a functional protein. But the functional significance of the resulting cDNAs and proteins is still unclear. It has been reported that the expression and activity of 1␣-OHase in extrarenal cells differs from that observed in kidney. Recent studies have shown that the addition of exogenous calcitriol inhibits the renal 1␣-OHase in contrast to that observed in macrophages [20]. Expression study presented here confirmed that the 1␣-OHase activity of human endometrial Ishikawa cells is due to a single gene product. Alternative splicing might have a biological function in regulating the enzyme level and causes the differences between renal and extrarenal tissues. Although we detected 1␣-OHase in normal and malignant endometrial tissues by immunohistochemistry, the proof of the presence of splice variants of the enzyme and data about the effect of calcitriol on 1␣-OHase expression in endometrium in correlation to existing gene variants are still missing and under investigation. Thus, the differential regulation in vitro and in vivo remains unclear. The expression of 1␣-OHase has been previously shown in human decidua during early pregnancy and preliminary data postulated the production of calcitriol in vitro by uterine cells in absence of decidual cells [21]. It seems likely that calcitriol plays not only immunmodulatory but also autocrine functions in endometrium. The protective effects of calcitriol have been shown in common cancers. It promotes active cell death in lung and breast cancer cells [22,23]. But there are contrary data about the activity of 1␣-OHase in cancer cells. It has been proposed that a reduced activity of the enzyme in cancer cells resulted in a diminished activity of the enzyme and local synthesis of calcitriol [24,25]. In contradiction to this, some reports documented an up regulation of the activity of 1␣OHase in breast tumors. The abrogation of the effect of calcitriol was due to abnormal activity of the calcitriol catabolizing enzyme, vitamin D 24-hydroxylase [26]. However, these investigations were not performed in correlation to the expression of 1␣-OHase splice variants and no exact information about primer settings or protein expression of the full length enzyme and its variants was demonstrated in these studies. Therefore, we assume the experimental settings of
S. Becker et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 771–775
these reports include functional and non functional 1␣-OHase gene variants and resulted in divergent data. Further analysis of regulation, function and activity of 1␣-OHase gene variants and other vitamin D metabolizing enzymes may clarify these questions and open up a new paradigm in understanding the role of calcitriol in modulating cell growth in gynaecological malignoma.
5. Conclusion In summary, our data demonstrate the expression of 1␣OHase in human endometrium and in endometrial Ishikawa cells. The detection of alternative splicing of 1␣-OHase in Ishikawa cells may contribute to the antiproliferative effects of calcitriol. These findings support an important autocrine role for calcitriol in endometrium and may therefore provide important target for the prevention and treatment of human endometrial cancer.
Acknowledgments We thank C. Siggel and B. Jablonski for the technical support. References [1] C.M. Hansen, L. Binderup, K.J. Hamber, C. Carlberg, Vitamin D and cancer: effects of 1,25 (OH)2 D3 and its analogues on growth control and tumorigenesis, Front Biosci. 6 (2001) 820–848. [2] M.T. Cantorna, B.D. Mahon, D-hormone and immune system, J. Rheumatol. Suppl. 76 (2005) 11–20. [3] Y.R. Lou, S. Qiao, R. Talonpoika, H. Syvala, P. Tuohima, The role of vitamin D3 metabolism in prostate cancer, J. Steroid Biochem. Mol. Biol. 92 (2004) 317–325. [4] S. Christakos, M. Raval-Pandya, R.P. Wernyi, W. Yang, Genomic mechanisms involved in pleiotropic actions of 1,25-dihydroxyvitamin D3, Biochem. J. 316 (1996) 361–371. [5] D.T. Baran, Nongenomic actions of the steroid hormone 1␣-25dihydroxyvitamin D3, J. Cell. Biochem. 56 (1994) 303–306. [6] R. Bland, E.A. Walker, S.V. Hughes, P.M. Stewart, M. Hewison, Constitutive expression of 25-hydroxyvitamin D3-1␣-hydroxylase in transformed human proximal tubule cell line: evidence for direct regulation of vitamin D metabolism by calcium, Endocrinology 140 (1999) 2027–2034. [7] P. Zehnder, R. Bland, M.C. Williams, et al., Extrarenal expression of 25-hydroxyvitamin D3-1␣-hydroxylase, J. Clin. Endocrinol. Metab. 86 (2001) 888–894. [8] D. Zehnder, R. Bland, S.V. Hughes, et al., Analysis of the tissue distribution of 1␣-hydroxylase identifies novel extra-renal sites for the synthesis of 1,25–dihydroxyvitamin D3, in: A.W. Norman, R. Bouillon, M. Thomasset (Eds.), Vitamin D Endocrine System, University of California, Riverside, 2000, pp. 159–162. [9] P. Naveilhan, C. Baudet, F. Berger, A.L. Benabid, P. Brachet, D. Wion, Induction of glioma cell death by 1,25-dihydroxyvitamin D3, in: A.W. Norman, R. Bouillon, M. Thomasset (Eds.), Vitamin D: A Pluripotent Steroid Hormone: Structural Studies, Molecular Endocrinology and Clinical Applications, Walter de Gruyter, New York, 1994, pp. 644–645.
775
[10] C.M. Hansen, T.L. Frandsen, N. Brunner, L. Binderup, 1 alpha, 25Dihydroxyvitamin D3 inhibits the invasive potential of human breast cancer cells in vitro, Clin. Exp. Metastasis 12 (3) (1994) 195–202. [11] J.P.T.M. Van Leeuwn, H.A.P. Pols, Vitamin D: anti cancer and differentiation, in: D. Feldman, F.H. Glorieux, J.W. Pike (Eds.), Vitamin D, Academic Press, London, 1997, pp. 1089–1099. [12] S.Y. James, A.G. Mackay, K.W. Colsten, et al., Effects of 1,25 dihydroxyvitamin D3 and its analogues on induction of apoptosis in breast cancer cells, J. Steroid Biochem. Mol. Biol. 58 (1996) 395–401. [13] R. Kaaks, A. Lukanova, M.S. Kurzer, Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review, Cancer Epidemiol. Biomarkers Prev. 11 (12) (2002) 1531–1543. [14] A. Vienonen, S. Miettinen, M. Blauer, P.M. Martikainen, E. Tomas, P.K. Heinonen, T. Ylikomi, Expression of nuclear receptors and cofactors in human endometrium and myometrium, J. Soc. Gynecol. Investig. 11 (2004) 104–112. [15] H. Yabushita, M. Hirata, M. Noguchi, M. Nakanishi, Vitamin D receptor in endometrial carcinoma and the differentiation-inducing effect of 1,25-dihydroxyvitamin D3 on endometrial carcinoma cell lines, J. Obstet. Gynaecol. Res. 22 (1996) 529–539. [16] B. Diesel, M. Seifert, J. Radermacher, U. Fischer, W. Tilgen, J. Reichrath, E. Meese, Towards a complete picture of splice variants of the gene for 25-hydroxyvitamin D3 1alpha-hydroxylase in brain and skin cancer, J. Steroid Biochem. Mol. Biol. 89-90 (2004) 527–532. [17] J.N. Flanagan, L. Wang, V. Tangpricha, J. Reichrath, T.C. Chen, M.F. Holick, Regulation of the 25-hydroxyvitamin D-1alpha-hydroxylase gene and its splice variant, Recent Results Cancer Res. 164 (2003) 157–167. [18] S. Ren, L. Nguyen, S. Wu, C. Encinas, J.S. Adams, M. Hewison, Alternative splicing of vitamin D-24-hydroxylase: a novel mechanism for the regulation of extrarenal 1,25-dihydroxyvitamin D synthesis, J. Biol. Chem. 280 (21) (2005) 20604–20611. [19] B. Diesel, J. Rademacher, M. Bureik, R. Bernhardt, M. Seifert, J. Reichrath, U. Fischer, E. Meese, Vitamin D3 metabolism in human glioblastoma multiforme: functionality of CYP27B1 splice variants, metabolism of calcidiol and effect of calcitriol, Clin. Cancer Res. 11 (15) (2005) 5370–5380. [20] H. Reichel, H.P. Koeffler, R. Barbers, A.W. Norman, Regulation of 1,25-dihydroxyvitamin D3 production by cultured alveolar macrophages from normal human donors and from patients with pulmonary sarcoidosis, J. Clin. Endocrinol. Metab. 65 (6) (1987) 1201–1209. [21] M. Kachkache, C. Rebut–Bonneton, J. Demignon, E. Cynober, M. Garabedian, Uterine cells other than stromal decidual cells are required for 1,25-dihydroxyvitamin D3 production during early human pregnancy, FEBS Lett. 333 (1993) 83–88. [22] G. Jones, H. Ramshaw, A. Zhang, et al., Expression and activity of vitamin D-metabolizing cytochrome P450s (Cyp1␣ and Cyp24) in human non small lung carcinomas, Endocrinology 140 (1999) 3303–3310. [23] K. Townsend, K.N. Evans, M.J. Campbell, K.W. Colston, J.S. Adams, M. Hewison, Biological actions of extra-renal 25-hydroxyvitamin D-1alpha-hydroxylase and implications for chemoprevention and treatment, J. Steroid Biochem. Mol. Biol. 97 (1–2) (2005) 103–109. [24] U. Segersten, P. Holm, P. Bj¨orklund, O. Hessman, H. Nordgern, L. Binderup, G. Akerson, P. Hellman, G. Westin, 25-Hydroxyvitamin D3 1␣ hydroxylase expression in breast cancer and use of non 1␣hydroxylated vitamin D analogue, Breast Cancer Res. 7 (6) (2005) 980–986. [25] J.Y. Hsy, D. Feldmann, J.E. Mc Neal, D.M. Peel, Reduced 1␣hydroxylase activity in human prostate cancer cells correlates with decreased susceptibility to 25-hydroxyvitamin D3-induced growth inhibition, Cancer Res. 61 (2001) 2852–2856. [26] K. Townsend, C.M. Vanwell, M. Guy, K.W. Colston, J.L. Mansi, P.M. Stewart, M.J. Campbell, M. Hewison, Autocrine metabolism of vitamin D in normal and malignant breast tissues, Clin. Cancer Res. 11 (2005) 3579–3586.
