The Role of 17B-Hydroxysteroid Dehydrogenases in Modulating the Activity of 2-Methoxyestradiol in Breast Cancer Cells

Research Article

Simon P. Newman,1 Christopher R. Ireson,1 Helena J. Tutill,1 Joanna M. Day,1 Michael F.C. Parsons,1 Mathew P. Leese,2 Barry V.L. Potter,2 Michael J. Reed,1 and Atul Purohit1

1Endocrinology and Metabolic Medicine and Sterix Ltd., Faculty of Medicine, Imperial College, St Mary’s Hospital, London, United Kingdom and 2Medicinal Chemistry and Sterix Ltd., Department of Pharmacy and Pharmacology, University of Bath, Bath, United Kingdom

Abstract cultures of cells derived from breast tumors have shown an h The bis-sulfamoylated derivative of 2-methoxyestradiol increased level of expression of 17 -HSD type 1 compared with (2-MeOE2), 2-methoxyestradiol-3,17-O,O-bis-sulfamate (2- that in cells derived from normal tissue (4). Furthermore, in a case- control study of patients diagnosed with breast cancer, increased MeOE2bisMATE), has shown potent antiproliferative and h antiangiogenic activity in vitro and inhibits tumor growth expression of 17 -HSD type 1 correlated with an increased risk of in vivo. 2-MeOE2bisMATE is bioavailable, in contrast to relapse (5). 2-MeOE2 that has poor bioavailability. In this study, we have 2-Methoxyestradiol (2-MeOE2; Fig. 1, 2) is generated in vivo by examinedtheroleof17B-hydroxysteroid dehydrogenase hydroxylation and subsequent methylation of E2. 2-MeOE2 has (17B-HSD) type 2 in the metabolism of 2-MeOE2. In MDA- been shown to be an antiangiogenic and antiproliferative agent in vitro (6, 7). 2-MeOE2 also inhibits the in vivo growth of MB-231 cells, which express high levels of 17B-HSD type 2, and in MCF-7 cells transfected with 17B-HSD type 2, high- xenografts derived from human breast MDA-MB-435 carcinoma performance liquid chromatography analysis showed that a cells, Meth A sarcomas, B16 melanomas, and the multiple myeloma significant proportion of 2-MeOE2 was metabolized to inactive cell line KAS-6/1 in immunodeficient mice. However, comparatively 2-methoxyestrone. Furthermore, MCF-7 cells transfected high p.o. (75 mg/kg/d) or i.p. (150 mg/kg/d) doses of 2-MeOE2 with 17B-HSD type 2 were protected from the cytotoxic effects are necessary to reduce the growth of breast or myeloma tumors, of 2-MeOE2. In contrast, no significant metabolism of respectively (8, 9). In a recent study, we administered 2-MeOE2 2-MeOE2bisMATE was detected in transfected cells and (10 mg/kg) either p.o. or i.v. to female rats to investigate the 17B-HSD type 2 transfection did not offer protection against metabolism of the agent in vivo (10). 2-MeOE2 was below the 2-MeOE2bisMATE cytotoxicity. This study may go some way to limit of detection in plasma following p.o. dosing and was rapidly explaining the poor bioavailability of 2-MeOE2, as the removed from the plasma following administration of a single gastrointestinal mucosa expresses high levels of 17B-HSD type i.v. dose, which suggested that the agent was undergoing rapid 2. In addition, this study shows the value of synthesizing biotransformation in vivo. In contrast, the antiangiogenic/antitumor agent 2-methoxyestradiol-3,17-O,O-bis-sulfamate (2-MeOE2- sulfamoylated derivatives of 2-MeOE2 with C17-position modifications as these compounds have improved bioavail- bisMATE; Fig. 1, 3; refs. 11–17) could still be detected in plasma ability and potency both in vitro and in vivo. (Cancer Res 2006; 24 hours after a single p.o. dose (10). 66(1): 324-30) The difference between the plasma levels of 2-MeOE2 and 2-MeOE2bisMATE in both human and rodent may be a result of the differing metabolic fate of these compounds. In the present Introduction study, we test the hypothesis that 2-MeOE2 is metabolized and Estrogens are thought to have a role in the promotion and inactivated by 17h-HSD enzymes in estrogen receptor (ER)–positive progression of hormone-dependent tumors. Levels of estrogens (MCF-7) and ER-negative (MDA-MB-231) breast cancer cells. Fur- within these tumors are higher than in the normal breast tissue and thermore, using both radiometric enzyme assays and analysis of the the peripheral circulation (1, 2), indicating that these hormones metabolism of unlabeled substrates by high-performance liquid may be synthesized in situ. These metabolic pathways are catalyzed chromatography (HPLC), we measured the interconversion of E1 by several enzymes, among which steroid sulfatase, which catalyzes and E2 in wild-type and transfected MCF-7 and MDA-MB-231 cells. the biotransformation of estrone sulfate to estrone (E1), and the Finally, we examined the metabolism of 2-MeOE2bisMATE, a 17h-hydroxysteroid dehydrogenases (17h-HSD), which catalyze the compound expected to be resistant to 17h-HSD metabolism due to interconversion of the weak estrogen E1 and the biologically active a sulfamate moiety at the C17 position. These studies will provide estrogen estradiol (E2), are important. 17h-HSD type 1, using the further information on the metabolic fate of 2-MeOE2 and identify + reduced form of the cofactor NADP , catalyzes the reduction of E1 new anticancer compounds, such as 2-MeOE2bisMATE, which are + to E2. Conversely, 17h-HSD type 2 requires the cofactor NAD and more potent in vitro and in vivo than 2-MeOE2. catalyzes the oxidation of E2 to E1 (3). Studies using primary Materials and Methods Drug synthesis. The syntheses of 2-methoxyestrone (2-MeOE1; Fig. 1, 1) Requests for reprints: Simon P. Newman, Endocrinology and Metabolic Medicine, and 2-MeOE2 (Fig. 1, 2) have been reported (18). 2-MeOE2bisMATE (Fig. 1, 3) Faculty of Medicine, Imperial College, St Mary’s Hospital, London W2 1NY, United was synthesized by reaction of 2-MeOE2 with excess sulfamoyl chloride in Kingdom. Phone: 44-207-886-1210; Fax: 44-207-886-1790; E-mail: simon.newman@ dimethyl acetamide. 2-Methoxyoestradiol-17-O-sulfamate (2-MeOE2- imperial.ac.uk. I2006 American Association for Cancer Research. 17MATE; Fig. 1, 4) was synthesized from 2-MeOE2 in three steps by first doi:10.1158/0008-5472.CAN-05-2391 benzylating the 3-hydroxyl group, then sulfamoylating the 17-hydroxyl

Cancer Res 2006; 66: (1). January 1, 2006 324 www.aacrjournals.org

unlabeled product (50 Ag) for visualization, and the diethyl ether phase was evaporated to dryness. The extracted steroids were resuspended in diethyl

ether and spotted onto TLC plates/silica gel 60 F254 (Merck, West Drayton, United Kingdom). The TLC plates were developed using dichloromethane/ ethyl acetate (4:1 v/v) as the solvent system. UV light (254 nm) was used to visualize the positions of the E1 and E2; the areas corresponding to the product were cut out, placed in scintillation vials, and the product was eluted in methanol (0.5 mL) before the addition of scintillation fluid (10 mL). Product and recovery radioactivities were counted using a liquid scintillation counter. In vitro metabolism of unlabeled steroids. Twenty-four hours after transfection, cells were incubated with either vehicle (THF) or 10 Amol/L of E2, E1, 2-MeOE2, 2-MeOE1, or 2-MeOE2bisMATE in medium for 48 hours. The compounds were also incubated with medium alone. After Figure 1. Structures: 1, 2-MeOE1; 2, 2-MeOE2; 3, 2-MeOE2bisMATE; 4, incubation, cells were removed from the flask by addition of trypsin. 2-MeOE2-17MATE. 17h-Epitestosterone was used as an internal standard. Medium (3 mL) and cells were combined and extracted twice with diethyl ether (4 mL) and group, and then hydrogenolytically deprotecting the 3-hydroxyl group. Full the lower aqueous layer frozen in a methanol/solid carbon dioxide mixture. details of these syntheses will be reported elsewhere. The upper organic phase from each extraction was decanted, combined, Cell culture. MCF-7 (ER positive) and MDA-MB-231 (ER negative) and evaporated to dryness under a stream of air at room temperature. The human breast cancer cells were obtained from the American Type Culture residues were stored at 20jC until they were analyzed by HPLC. Collection (LGC Promochem, Teddington, United Kingdom). MCF-7 cells To confirm that 2-MeOE1 had been generated from 2-MeOE2, the residue were cultured in MEM (Sigma, Poole, United Kingdom) supplemented with was reconstituted in 58% methanol and 0.02 mol/L ammonium sulfate 10% fetal bovine serum (FBS; Sigma) and 2 mmol/L glutamine (Sigma). (500 AL) and incubated with an excess of sodium borohydride for 1 hour at MDA-MB-231 cells were cultured in MEM (Sigma) supplemented with 5% room temperature. 2-MeOE2 and 2-MeOE1 were extracted from the reaction FBS (Sigma) and 2 mmol/L glutamine (Sigma). MCF-7 cells were used as mixture with diethyl ether (4 mL) and the organic phase was evaporated they have low 17h-HSD types 1 and 2 activities (19), whereas MDA-MB- to dryness. HPLC analysis was done as described by Ireson et al. (10). 231 cells were used as they have high oxidative 17h-HSD type 2 activity Briefly, 2-MeOE2bisMATE was separated from 2-MeOE2, 2-MeOE1, and (20, 21), reflecting the mainly oxidative direction of estrogen metabolism 2-methoxyestradiol-17-O-sulfamate (2-MeOE2-17MATE; Fig. 1, 4) using a in humans (22). reversed-phase HPLC method. Extracted samples were reconstituted in Cloning. The full-length 17h-HSD type 2 cDNA was a kind gift from Dr. mobile phase and injected on to a C3-phenyl column (250 5mm,5Amol/L) Luu-The (Laval University, Quebec, Quebec, Canada). The cDNA was cloned purchased from Phenomenex (Cheshire, United Kingdom). The agents were into the pAlter Max (Promega, Southampton, United Kingdom) expression separated from other medium components using an isocratic mobile phase vector using EcoRI restriction sites. To clone 17h-HSD type 1, a PCR-based consisting of 58% methanol in 0.02 mol/L ammonium sulfate containing method was used to incorporate the first 25 bp of the 17h-HSD type 1 1 mmol/L sodium azide. Analysis was done with a photodiode array sequence into an otherwise complete 17h-HSD type 1 cDNA, a kind gift detector set at 285 nm. from Dr. Luu-The. The full-length 17h-HSD type 1 sequence was subcloned Statistics. All experiments were done in triplicate and data presented are into the pAlter Max expression vector (Promega). All constructs were representative of one of three such experiments, all errors shown are the sequenced to confirm correct insertion of the respective cDNA into the mean F SD. Student’s t test was used to assess the significance of the vector. differences in cell proliferation. Transient transfections. On day 1, cells were seeded into T-25 tissue culture flasks at 20% confluency in growth medium. After 24 hours, cells Results were transfected with either 1 Ag per T-25 of pAlter 17h-HSD type 1, pAlter 17h-HSD type 2 or empty vector (mock transfected) combined with 3 AL 17B-HSD activity in MCF-7 and MDA-MB-231 cells. MCF-7 (ER Fugene 6 (Roche, Welwyn Garden City, United Kingdom) and 100 ALFBS- positive) breast cancer cells have relatively low endogenous 17h-HSD free MEM (Sigma) per flask. Compounds for metabolism studies were added reductive (type 1) and oxidative (type 2) activities (20, 21) and are, on day 3, and the cells and medium were extracted on day 5 for analysis therefore, a good model into which to transfect and study 17h-HSD h by HPLC. 17 -HSD enzyme activity was assayed on day 5, 72 hours after types 1 and 2 enzyme activities. The 17h-HSD type 1 activity was transfection. 12 fmol E2/h/106 cells and the corresponding 17h-HSD type 2 Proliferation assays. 2-MeOE2 (10 Amol/L) or 2-MeOE1 (10 Amol/L) activity was 10 fmol E1/h/106 cells following mock transfection. were added to cells in fresh growth medium 24 hours after transfection. Seventy-two hours after transfecting MCF-7 cells with the 17h-HSD Cells were grown for a further period of 72 hours. The cell monolayers were 6 then treated with Zaponin (Coulter, United Kingdom) and the number of type 1 cDNA, the type 1 activity was 2,817 fmol E2/h/10 cells, a 235- cell nuclei per flask was determined using a Coulter Counter. fold increase relative to mock-transfected cells. After transfection 17B-HSD enzyme assays. 17h-HSD activity in both the reductive with the cDNA for 17h-HSD type 2, the type 2 activity increased 51- (E1!E2), type 1, and oxidative (E2!E1), type 2, directions was measured fold relative to mock-transfected cells to 512 fmol E1/h/106 cells. using a method based on that described by Singh and Reed (20). Briefly, MDA-MB-231 breast cancer cells have a greater 17h-HSD intact cell monolayers were incubated with 5 pmol per T-25 flask of either oxidative activity than 17h-HSD reductive activity (20, 21) and tritiated [6,7-3H]E1 (50 Ci/mmol, NEN-Perkin-Elmer, Boston, MA) or 3 are, thus, a good model to study metabolism by endogenous [6,7- H]E2 (41.3Ci/mmol, NEN-Perkin-Elmer) in 2.5 mL of serum-free 17h-HSD type 2. Following mock transfection, MDA-MB-231 cells medium for 0.5 hours (reductive assay) or 3 hours (oxidative assay). Blank 17h-HSD enzyme activities were 11 and 54 fmol product/h/106 cells incubations were done in parallel by incubating labeled steroid in cell-free for types 1 and 2, respectively. The type 1 activity increased over flasks. After incubation, 2 mL of medium was removed and added to test 6 tubes containing 103 dpm of [4-14C]E2 (45 mCi/mmol, NEN-Perkin-Elmer) 11-fold to 128 fmol E2/h/10 cells 72 hours after transfection or [4-14C]E1 (54 mCi/mmol, NEN-Perkin-Elmer) to monitor procedural with the cDNA for 17h-HSD type 1. MDA-MB-231 cells were not losses. Diethyl ether (4 mL) was used to extract the steroids from the transfected with cDNA for 17h-HSD type 2 as they already express medium, the diethyl ether was transferred to a new glass tube containing high levels of this enzyme. www.aacrjournals.org 325 Cancer Res 2006; 66: (1). January 1, 2006

E1 and E2 metabolism. In addition to measuring enzyme Table 1. Metabolism of E2 and E1 in MCF-7 cells activities with labeled substrates, the metabolism of E1 and E2 by transfected with 17h-HSD type 1 or 2 MCF-7 cells was also examined using unlabeled substrates and subsequent HPLC analysis in wild-type and transfected cells. No Transfection E2 to E1 E1 to E2 discernable peaks were detected (Fig. 2A) in mock-transfected MCF-7 cells to which vehicle (THF) was added. In mock- Wild type 10 F 28F 3 transfected cells incubated with E2 (10 Amol/L), a small peak 17h-HSD type 1 8 F 393F 4 h F F was detected that eluted with the same Rt as the E1 standard (Fig. 17 -HSD type 2 98 331 2B, peak 2). In contrast, as shown in Fig. 2C, metabolism in cells transfected with the cDNA for 17h-HSD type 2 resulted in the F oxidation of E2 to E1. Similar experiments with the cDNA for 17h- NOTE: Values are expressed as a percentage of total steroid SD (n = 3). HSD type 1 revealed that E1 was almost completely reduced to E2. The results from this series of experiments are quantified in Table 1 and clearly show that whereas wild-type MCF-7 cells have low intrinsic reductive and oxidative activity, transfection with the incubated with 2-MeOE2 (10 Amol/L), which eluted with the same h cDNAs for 17 -HSD type 1 or 2 greatly enhances their ability to Rt as the 2-MeOE1 standard. Table 2 shows that 72% of the activate E1 to E2 or inactivate E2 by conversion to E1. 2-MeOE2 was oxidized to the biologically inactive 2-MeOE1. 2-MeOE1 and 2-MeOE2 metabolism. As 17h-HSD types 1 and 2 When these cells were transfected with the 17h-HSD type 1 cDNA, interconvert E1 and E2, the ability of these enzymes to reduce and 2-MeOE2 was protected from oxidation and only 44% was oxidized oxidize 2-MeOE1 and 2-MeOE2, respectively, was examined using to the inactive 2-MeOE1 (Table 2). unlabeled substrates and subsequent HPLC analysis. In mock- 2-MeOE2bisMATE metabolism. We have previously shown transfected MCF-7 cells incubated with 2-MeOE2 (10 Amol/L), a that bis-sulfamoylation of 2-MeOE2 to generate 2-MeOE2bisMATE small peak was detected (Fig. 3A, peak 2), which eluted with the offers considerable advantages by increasing both the potency and same Rt as the 2-MeOE1 standard. Following transfection with the bioavailability of this class of compound. Due to the presence of cDNA for 17h-HSD type 2, 2-MeOE2 was below the limit of the sulfamate group at the C17 position, it was postulated that this detection and oxidized to 2-MeOE1 (Fig. 3B, peak 2). Similar compound would not be a substrate for either 17h-HSD type 1 or 2. experiments with the cDNA for 17h-HSD type 1 revealed that To test this hypothesis, MCF-7 cells were transfected with either 2-MeOE1 was reduced to 2-MeOE2. The results from this series of 17h-HSD type 1 or 2 and incubated with 2-MeOE2bisMATE experiments are quantified in Table 2. (10 Amol/L) for 48 hours. Figure 4 shows that the metabolism of We have shown that the breast cancer cell line MDA-MB-231 has 2-MeOE2bisMATE is identical in mock-, 17h-HSD type 1–, and 17h- endogenously high levels of 17h-HSD type 2 enzyme activity HSD type 2–transfected cells, with a significant peak eluting with h relative to 17 -HSD type 1 enzyme activity and is, therefore, a good the same Rt as 2-MeOE2bisMATE, indicating that this compound model to look at the oxidation and inactivation of 2-MeOE2. No is resistant to metabolism in MCF-7 cells. A small second peak discernable peaks were detected in wild-type mock-transfected was also detected in extracts from all three transfections, MDA-MB-231 cells to which vehicle (THF) was added (Fig. 3C). A which elutes at the same Rt as 2-MeOE2-17MATE (Fig. 1, 4), significant peak was detected (Fig. 3D, peak 2) in wild-type cells a breakdown product of 2-MeOE2bisMATE. In the absence of cells, in medium-only control experiments the proportion of the breakdown product 2-MeOE2-17MATE doubles relative to the parent compound 2-MeOE2bisMATE (data not shown). Cell proliferation. Observation of mock-transfected MCF-7 cells incubated with 2-MeOE2 (10 Amol/L) for 48 hours indicated that cell proliferation was inhibited. A large number of apoptotic-like cells were also observed relative to vehicle-treated cells (Fig. 5A and B). However, MCF-7 cells transfected with 17h-HSD type 2, 24 hours before incubation with 2-MeOE2 (10 Amol/L), seemed to be resistant to the inhibition of cell proliferation and induction of cell death by 2-MeOE2 (Fig. 5C). In contrast, transfection with 17h-HSD type 2 offered no apparent protection from the inhibition of cell proliferation and induction of cell death by 2-MeOE2bisMATE (Fig. 5D and E). Cell proliferation was quantified to confirm these observations. The addition of 2-MeOE2 (10 Amol/L) to MCF-7 cells resulted in a 60% reduction in cell proliferation in both mock-transfected and 17h-HSD type 1–transfected MCF-7 cells (Fig. 6). However, in MCF-7 cells transfected with 17h-HSD type 2, there was only a Figure 2. High-performance chromatograms of extracted medium and cells 16% inhibition of proliferation relative to mock-transfected cells. showing the metabolism of E2 in MCF-7 cells mock transfected or transfected h with 17h-HSD type 2. A, mock-transfected MCF-7 cells were incubated with Treatment of mock-transfected and 17 -HSD type 2–transfected vehicle (THF) for 48 hours. B, mock-transfected MCF-7 cells were incubated with MCF-7 cells with 2-MeOE1 (10 Amol/L) had no significant effect on E2 (10 Amol/L) for 48 hours. C, MCF-7 cells transfected with 17h-HSD type 2 proliferation with respect to untreated cells. However, in 17h-HSD were incubated with E2 (10 Amol/L) for 48 hours. Peaks 1 and 2 were identified as E2 and E1, respectively, by cochromatography with authentic standards type 1–transfected cells, a 57% reduction in cell proliferation was (I.S, internal standard, 17h-epitestosterone). AU, arbitrary unit. measured after incubation with 2-MeOE1 (10 Amol/L; Fig. 6).

Cancer Res 2006; 66: (1). January 1, 2006 326 www.aacrjournals.org

Figure 3. High-performance chromatograms of extracted medium and cells showing the metabolism of 2-MeOE2 in MCF-7 and MDA-MB-231 cells. A, mock-transfected MCF-7 cells were incubated with 2-MeOE2 (10 Amol/L) for 48 hours. B, MCF-7 cells transfected with 17h-HSD type 2, 24 hours before incubation with 2-MeOE2 (10 Amol/L) for 48 hours. C, wild-type MDA-MB-231 cells were incubated with vehicle (THF) for 48 hours. D, wild-type MDA-MB-231 cells were incubated with 10 Amol/L 2-MeOE2 for 48 hours. Peaks 1 and 2 were identified as 2-MeOE2 and 2-MeOE1, respectively, by cochromatography with authentic standards (I.S, internal standard, 7h-epitestosterone).

Treatment of MCF-7 cells with 2-MeOE2bisMATE (10 Amol/L) based drugs. The natural metabolite of E2, 2-MeOE2, is generated in resulted in an 85% inhibition of cell proliferation. Transfection vivo from 2-hydroxyestradiol by catechol-O-methyl transferase, an with either 17h-HSD type 1 or 2 had no significant effect on the enzyme that is expressed in a wide range of mammalian tissues (23). inhibition of proliferation by 2-MeOE2bisMATE (10 Amol/L; Fig. 7). 2-MeOE2 has been shown to be an antiproliferative and antiangiogenic agent in vitro and to inhibit tumor growth in vivo and is Discussion currently in phase I/II trials for breast cancer (24, 25). In this study, we showed that 2-MeOE2 is oxidized to the inactive metabolite 2- In this study, we used two breast cancer cell lines to explore MeOE1 (26) by MCF-7 cells transfected with the 17h-HSD type the roles of the reductive and oxidative 17h-HSD enzymes in the 2 enzyme. After just 48 hours, 95% of the added 2-MeOE2 was interconversion of E1 and E2, and in the metabolic fate of the metabolized to 2-MeOE1. Wild-type MDA-MB-231 cells have a 5-fold anticancer agent 2-MeOE2. The ER-positive MCF-7 cells have both greater oxidative 17h-HSD activity than reductive activity and were low 17h-HSD types 1 and 2 activities and as such provide a suitable able to inactivate 72% of the 2-MeOE2 after 48 hours. This result model into which to transfect either 17h-HSD type 1 or 2. To look suggested that in hormone-independent tumors, in which 17h-HSD at endogenous 17h-HSD type 2 activity, we used the ER-negative type 2 is the prominent activity (27), the anticancer agent 2-MeOE2 MDA-MB-231 cells, which have higher oxidative than reductive would be rapidly inactivated. This hypothesis is supported by 17h-HSD activity (20, 21). In this study, we showed that MCF-7 cells only converted 8% of the added E1 to E2. This is not surprising, as when assayed both the 17h-HSD types 1 and 2 enzyme activities were low and approximately equal. This result suggests that in vivo ER-positive tumors, which lack the pathways to synthesize E2 themselves, may be reliant on the peripheral tissue to generate the E2 required for their growth. In addition to their role in E2 metabolism in tumors, the 17h-HSD enzymes have an important role to play in the metabolism of other steroids, and in addition may affect the bioavailability of steroid-

Table 2. Metabolism of 2-MeOE2 and 2-MeOE1 in MCF-7 and MDA-MB-231 cells transfected with 17h-HSD type 1 or 2

Transfection 2-MeOE2 2-MeOE1 to 2-MeOE1 to 2-MeOE2

MCF-7 wild type 5 F 15F 1 MCF-7 17h-HSD type 1 ND 95 F 2 MCF-7 17h-HSD type 2 95 F 29F 3 MDA-MB-231 wild type 72 F 416F 5 Figure 4. High-performance chromatograms of extracted medium and cells h MDA-MB-231 17h-HSD type 1 44 F 256F 0 showing the metabolism of 2-MeOE2bisMATE by 17 -HSD types 1 and 2. A, mock-transfected cells were incubated with 2-MeOE2bisMATE (10 Amol/L) for 48 hours. B, MCF-7 cells were transfected with 17h-HSD type 1, 24 hours before incubation with 2-MeOE2bisMATE (10 Amol/L) for 48 hours. C, MCF-7 NOTE: Values are expressed as a percentage of total drug F SD (n = 3). cells were transfected with 17h-HSD type 2, 24 hours before incubation with Abbreviation: ND, not detectable. 2-MeOE2bisMATE (10 Amol/L) for 48 hours. Peaks 3 and 4 were identified as 2-MeOE2bisMATE and 2-MeOE2-17MATE, respectively, by cochromatography with authentic standards (I.S, internal standard, 17h-epitestosterone). www.aacrjournals.org 327 Cancer Res 2006; 66: (1). January 1, 2006

a single p.o. dose of 2-MeOE2 (10 mg/kg) was administered to rats, 2-MeOE2 could not be detected in the plasma 1 hour after ad ministration. After i.v. administration, however, 2-MeOE2 could be detected in the plasma, although its half-life was only 14 minutes. In a recent phase I trial, a daily p.o. dose of 1,000 mg 2-MeOE2 was given to 24 patients with advanced metastatic breast cancer (24, 25). Metabolism studies showed that 80% to 95% of the 2-MeOE2 was oxidized to 2-MeOE1 and furthermore 80% to 90% of both 2- MeOE2 and 2-MeOE1 were conjugated to glucuronides/sulfates. 17h-HSD type 2 activity but not that of type 1, has been detected in the human gastrointestinal tract. Immunohistochemistry has localized this expression to the absorptive epithelium of the stomach, duodenum, ileum, colon, and rectum (29). It is now proposed that 17h-HSD type 2 expression in the gastrointestinal tract has both an important role in regulating exposure to p.o. administered sex steroids and protecting the human body against exposure to environmental and/or bacterially synthesized sex steroids (29–31). Evidence for this role is provided by the observation that p.o. administered E2 and testosterone are inactivated rapidly and do not enter the circulation in significant amounts (32, 33). Furthermore, studies have shown that 17h-HSD type 2 can metabolize several p.o. administered steroidal compounds, includ- ing those used as oral contraceptives and hormone replacement therapy (34). The metabolism of 2-MeOE2 to 2-MeOE1 by 17h-HSD type 2, in conjunction with high E1 sulfotransferase immunoreac- Figure 5. Representative micrographs showing mock-transfected MCF-7 cells tivity that has been detected in epithelial cells of the gastrointes- (A) incubated with vehicle alone (THF); mock-transfected MCF-7 cells (B) tinal tract (35), may explain both the high levels of 2-MeOE1 and incubated with 2-MeOE2 (10 Amol/L); MCF-7 cells transfected with 17h-HSD type 2 (C) incubated with 2-MeOE2 (10 Amol/L); mock-transfected MCF-7 cells sulfate conjugates detected in the recent phase I study (25). (D) incubated with 2-MeOE2bisMATE (10 Amol/L); and MCF-7 cells transfected The metabolic inactivation of 2-MeOE2 has led to the search for with 17h-HSD type 2 (E) incubated with 2-MeOE2bisMATE (10 Amol/L). 2-MeOE2 derivatives, which are not so rapidly inactivated. One of MCF-7 cells were seeded in T-25 flasks and transfected 24 hours later with the relevant cDNA. Following a further 24 hours, the medium was replaced with fresh these derivatives, 2-MeOE2bisMATE, has shown promise as an growth medium containing the compounds and the cells were incubated for antitumor, antiangiogenic agent and has good bioavailability another 48 hours. (10–17). In this current study, no metabolism of 2-MeOE2bisMATE was detected when cells overexpressing either 17h-HSD type 1 or research from our group (12), which showed that although C17- 2 were treated with this compound. In all experiments where position modified 2-MeOE2 derivatives were potent inhibitors of 2-MeOE2bisMATE was incubated with cells, f20% of the MDA-MB-231 proliferation (IC50 < 0.40 Amol/L), 2-MeOE2 was a relatively weak inhibitor with an IC50 in excess of 4.0 Amol/L. The inactivation of 2-MeOE2 in MDA-MB-231 cells was partially reversed by overexpressing the 17h-HSD type 1 enzyme. This increased the reductive metabolism within the cell. MCF-7 cells transfected with 17h-HSD type 1 were able to convert 95% of added 2-MeOE1 to the active 2-MeOE2. These two results indicated that in tumors with a greater reductive 17h-HSD activity, 2-MeOE2 may not be metabolized to 2-MeOE1 and any circulating unconjugated 2-MeOE1 could be reduced to the active 2-MeOE2. Further evidence showing the vital balance between oxidative and reductive 17h-HSD activities, in relation with the efficacy of 2-MeOE2, was the observation that MCF-7 cells overexpressing 17h-HSD type 2 were significantly protected from the cytotoxic effects of 2-MeOE2. Conversely, proliferation of MCF-7 cells overexpressing type 1 was inhibited by treatment with the inactive metabolite 2-MeOE1 (26). In both breast and myeloma xenograft models in vivo, 2-MeOE2 has successfully inhibited tumor growth, although in both cases relatively high doses of 2-MeOE2 were required, 75 and 150 mg/kg/d, Figure 6. Effects of 2-MeOE2 and 2-MeOE1 on cell proliferation in transiently respectively (8, 9). In phase I trials, a clinical benefit was shown in transfected MCF-7 cells. MCF-7 cells were seeded in T-25 flasks and transfected 24 hours later with the relevant cDNA. Following a further 24 hours, the only two patients who were receiving 1,600 to 3,200 mg/d of medium was replaced with fresh growth medium containing the compounds and 2-MeOE2; further dose escalations of up to 6,000 mg/d are planned the cells were incubated for a further 72 hours. Cell numbers were counted (28). The ability of some tumor cell lines to inactivate 2-MeOE2 using a Coulter Counter. wt, mock-transfected MCF-7s; 2, MCF-7 cells transfected with 17h-HSD type 2; 1, MCF-7 cells transfected with 17h-HSD does not fully explain the high doses of 2-MeOE2 required to show type 1. Representative of one of three such experiments (n =3).Columns, efficacy in these trials. Previously, our group (10) showed that when mean; bars, SD. a, P < 0.001 versus mock-transfected untreated cells.

Cancer Res 2006; 66: (1). January 1, 2006 328 www.aacrjournals.org

the compound detected. This observation implies that 2-MeOE2- bisMATE may be stabilized in cells, possibly by binding to tubulin, as this class of compound is known to disrupt tubulin dynamics (36). Alternatively, the acidification of the medium by the cells may make the loss of the 3-sulfamate moiety less probable. Further evidence for the stability of 2-MeOE2bisMATE was that over- expression of type 2 offered no protection from the cytoxicity of 2-MeOE2bisMATE in cell proliferation experiments in contrast to 2-MeOE2. Modification of the C17 position not only imparts improved resistance to metabolism but can also increase the potency of 2-MeOE2-like compounds (37). Furthermore, Utsumi et al. (13) showed that 2-MeOE2bisMATE and another C17 position– modified compound, 2-methoxyestradiol-17h-cyanomethyl-3-O- sulfamate, inhibited both in vitro cell proliferation (MCF-7 and MDA-MB-231 cells) and caused significant tumor regression (20 mg/kg p.o.) in a MCF-7 xenograft model. Figure 7. Effects of 2-MeOE2bisMATE on cell proliferation in transiently In conclusion, we have shown that the anticancer agent 2- transfected MCF-7 cells. MCF-7 cells were seeded in T-25 flasks and transfected 24 hours later with the relevant cDNA; following a further 24 hours, the medium MeOE2 can be oxidized by the 17h-HSD type 2 enzyme to an was replaced with fresh growth medium containing the compounds and the cells inactive metabolite, 2-MeOE1, and this may go some way to were incubated for a further 72 hours. Cell numbers were counted using a Coulter Counter. wt, mock-transfected MCF-7s; 2, MCF-7 cells transfected explaining the poor bioavailability of this compound observed in with 17h-HSD type 2; 1, MCF-7 cells transfected with 17h-HSD type 1. clinical trials. In contrast, 2-MeOE2bisMATE, the promising Representative of one of three such experiments (n = 3). Columns, mean; bars, antitumor/antiangiogenic agent, is resistant to metabolism by SD. a, P < 0.001 versus mock-transfected untreated cells. both the 17h-HSD type 1 and type 2 enzymes. compound was detected as the inactive breakdown product 3 2-MeOE2-17MATE. In contrast, when incubated with medium Acknowledgments alone, this percentage increased significantly (P < 0.001) to 33% of Received 7/8/2005; revised 9/14/2005; accepted 10/18/2005. Grant support: Sterix Ltd., a member of the Ipsen group. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance 3 Unpublished observation. with 18 U.S.C. Section 1734 solely to indicate this fact.

References the microtubule inhibitors 2-methoxyestradiol and 18. Leese MP, Hejaz HAM, Mahon MF, et al. A-ring Taxol. Cancer Res 1997;57:81–6. substituted estrogen-3-O-sulfamates: potent multi-tar- 1. Geisler J. Breast cancer tissue estrogens and their 10. Ireson CR, Chander SK, Purohit A, et al. Pharmaco- geted anti-cancer agents. J Med Chem 2005;48:5243–56. manipulation with aromatase inhibitors and inactiva- kinetics and efficacy of 2-methoxyestradiol and 2- 19. Miettinen MM, Poutanen MH, Vihko RK. Character- tors. J Steroid Biochem Mol Biol 2003;86:245–53. methoxyestradiol-bis-sulfamate in vivo in rodents. Br J ization of estrogen-dependent growth of cultured MCF-7 2. Bonney RC, Reed MJ, Davidson K, Beranek PA, James Cancer 2004;90:932–7. human breast-cancer cells expressing 17h-hydroxysteroid VHT. The relationship between 17h-hydroxysteroid 11. Newman SP, Leese MP, Purohit A, et al. Inhibition of dehydrogenase Type 1. Int J Cancer 1996;68:600–4. dehydrogenase activity and estrogen concentrations in in vitro angiogenesis by 2-methoxy- and 2-ethyl- 20. Singh A, Reed MJ. Insulin-like growth factor Type I and human breast tumors and in normal breast tissue. Clin estrogen sulfamates. Int J Cancer 2004;109:533–40. insulin-like growth factor Type II stimulate estradiol-17h Endocrinol 1983;19:727–39. 12. Raobaikady B, Reed MJ, Leese MP, Potter BVL, Purohit hydroxysteroid dehydrogenase (reductive) activity in 3. Labrie F, Luu-The V, Lin S-X, et al. The key role of 17h- A. Inhibition of MDA-MB-231 cell cycle progression and breast cancer cells. J Endocrinol 1991;129:R5–8. hydroxysteroid dehydrogenases in sex steroid biology. cell proliferation by C-2-substituted estradiol mono- and 21. DuncanLJ,ReedMJ.Theroleandproposed Steroids 1997;62:148–58. bis-3-O-sulfamates. Int J Cancer 2005;117:150–9. mechanism by which estradiol 17h-hydroxysteroid 4. Speirs V, Green AR, Atkin SL. Activity and gene 13. Utsumi T, Leese MP, Chander SK, et al. The effects of dehydrogenase regulates breast tumor estrogen concen- expression of 17h-hydroxysteroid dehydrogenase Type I 2-methoxyestrogen sulfamates on the in vitro and trations. J Steroid Biochem Mol Biol 1995;55:565–72. in primary cultures of epithelial and stromal cells in vivo proliferation of breast cancer cells. J Steroid 22. Reed MJ, Beranek PA, Ghilchik MW, James VHT. derived from normal and tumorous human breast Biochem Mol Biol 2005;94:219–27. Conversion of estrone to estradiol and estradiol to tissue: the role of IL-8. J Steroid Biochem Mol Biol 14. Day JM, Newman SP, Comninos A, et al. The effects of estrone in postmenopausal women. Obstet Gynecol 1998;67:267–74. 2-substituted estrogen sulfamates on the growth of 1985;66:361–5. 5. Gunnarsson C, Olsson BM, Stal O. Abnormal expres- prostate and ovarian cancer cells. J Steroid Biochem Mol 23. Mannisto PT, Kaakkola S. Catechol-O-methyltrans- sion of 17h-hydroxysteroid dehydrogenases in breast Biol 2003;84:317–25. ferase (COMT): biochemistry, molecular biology, phar- cancer predicts late recurrence. Cancer Res 2001;61: 15. Suzuki RN, Newman SP, Purohit A, Leese MP, Potter macology, and clinical efficacy of the new selective 8448–51. BVL, Reed MJ. Growth inhibition of multi-drug-resistant COMT inhibitors. Pharmacol Rev 1999;51:593–628. 6. Fotsis T, Zhang Y, Pepper MS, et al. The endogenous breast cancer cells by 2-methoxyoestradiol-bis-sulpha- 24. Lakhani NJ, Sarkar MA, Venitz J, Figg WD. 2- estrogen metabolite 2-methoxyestradiol inhibits angio- mate and 2-ethyloestradiol-bis-sulphamate. J Steroid Methoxyestradiol, a promising anticancer agent. Phar- genesis and suppresses tumor growth. Nature 1994;368: Biochem Mol Biol 2003;84:269–78. macotherapy 2003;23:165–72. 237–9. 16. Ho YT, Newman SP, Purohit A, Leese MP, Potter BVL, 25. Sledge GW, Miller KD, Hanye LG, et al. A phase I 7. Zhu BT, Conney AH. Is 2-methoxyestradiol an Reed MJ. The effects of 2-methoxy estrogens and their study of 2-methoxyestradiol in patients with refractory endogenous estrogen metabolite that inhibits mammary sulfamoylated derivatives in conjunction with TNF-a on metastatic breast cancer. Proc Am Soc Clin Oncol 2002; carcinogenesis? Cancer Res 1998;58:2269–77. endothelial and fibroblast cell growth, morphology and Abstract no. 441. 8. Dingli D, Timm M, Russell SJ, Witzig TE, Rajkumar apoptosis. J Steroid Biochem Mol Biol 2003;86:189–96. 26. Purohit A, Hejaz HA, Walden L, et al. The effect of 2- SV. Promising preclinical activity of 2-methoxyestra- 17. Wood L, Leese MP, Mouzakiti A, et al. 2-MeOE2bis- methoxyestrone-3-O-sulfamate on the growth of breast diol in multiple myeloma. Clin Cancer Res 2002;12: MATE induces caspase-dependent apoptosis in CAL51 cancer cells and induced mammary tumors. Int J Cancer 3948–54. breast cancer cells and overcomes resistance to TRAIL 2000;85:584–9. 9. Klauber N, Parangi S, Flynn E, Hamel E, D’Amato RJ. via cooperative activation of caspases. Apoptosis 2004;9: 27. Abul-Hajj YJ. Relationship between estrogen recep- Inhibition of angiogenesis and breast cancer in mice by 323–32. tors, 17h-hydroxysteroid dehydrogenase and estrogen www.aacrjournals.org 329 Cancer Res 2006; 66: (1). January 1, 2006

content in human breast cancer. Steroids 1979;34: 31. Mustonen MV, Poutanen MH, Kellokumpu S, of the wild-type enzyme in the endoplasmic reticulum. 217–25. et al. Mouse 17h-hydroxysteroid dehydrogenase Type 2 Endocrinology 1999;140:3334–41. 28. Dahut WL, Lakhani NJ, Kohn EC, et al. A phase I study mRNA is predominantly expressed in hepatocytes 35. Miki Y, Nakata T, Suzuki T, et al. Systemic of 2-methoxyestradiol (2ME2) in patients with solid and in surface epithelial cells of the gastrointesti- distribution of steroid sulfatase and estrogen sulfo- tumors. Proc Am Soc Clin Oncol 2003;Abstract no. 833. nal and urinary tracts. J Mol Endocrinol 1998;20: transferase in human adult and fetal tissues. J Clin 29. Sano T, Hirasawa G, Takeyama J, et al. 17h- 67–74. Endocrinol Metab 2002;87:5760–8. Hydroxysteroid dehydrogenase Type 2 expression and 32. Lobo RA, Cassidenti DL. Pharmacokinetics of oral 36. MacCarthy-Morrogh L, Townsend PA, Purohit A, enzyme activity in the human gastrointestinal tract. Clin 17h-estradiol. J Reprod Med 1992;37:77–84. et al. Differential effects of estrone and estrone-3-O- Sci (Lond) 2001;101:485–91. 33. Farthing MJ, Vinson GP, Edwards CR, Dawson AM. sulfamate derivatives on mitotic arrest, apoptosis, and 30. English MA, Hughes SV, Kane KF, Langman MJ, Testosterone metabolism by the rat gastrointestinal microtubule assembly in human breast cancer cells. Stewart PM, Hewison M. Estrogen inactivation in the tract, in vitro and in vivo. Gut 1982;23:226–34. Cancer Res 2000;60:5441–50. colon: analysis of the expression and regulation of 34. Puranen TJ, Kurkela RM, Lakkakorpi JT, et al. 37. Leese MP, Leblond B, Newman SP, Purohit A, Reed 17h-hydroxysteroid dehydrogenase isozymes in nor- Characterization of molecular and catalytic properties MJ, Potter BVL. Anti-cancer activities of novel D-ring mal colon and colonic cancer. Br J Cancer 2000;83: of intact and truncated human 17h-hydroxysteroid modified 2-substituted estrogen-3-O-sulfamates. J Ste- 550–8. dehydrogenase type 2 enzymes: intracellular localization roid Biochem Mol Biol 2005;94:239–51.

Cancer Res 2006; 66: (1). January 1, 2006 330 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2006 American Association for Cancer Research. The Role of 17β-Hydroxysteroid Dehydrogenases in Modulating the Activity of 2-Methoxyestradiol in Breast Cancer Cells

Simon P. Newman, Christopher R. Ireson, Helena J. Tutill, et al.

Cancer Res 2006;66:324-330.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/66/1/324

Cited articles This article cites 35 articles, 8 of which you can access for free at: http://cancerres.aacrjournals.org/content/66/1/324.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/66/1/324.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/66/1/324. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.