0013-7227/06/$15.00/0 Printed in U.S.A.
Endocrinology 147(11):5333–5339 Copyright © 2006 by The Endocrine Society doi: 10.1210/en.2006-0778
Human Hydroxysteroid (17-) Dehydrogenase 1 Expression Enhances Estrogen Sensitivity of MCF-7 Breast Cancer Cell Xenografts Bettina Husen,* Kaisa Huhtinen,* Taija Saloniemi, Josef Messinger, Hubert H. Thole, and Matti Poutanen Solvay Pharmaceuticals Research Laboratories (B.H., J.M., H.H.T.), 30173 Hannover, Germany; and Department of Physiology (K.H., T.S., M.P.), Institute of Biomedicine, University of Turku, 20520 Turku, Finland Hydroxysteroid (17-) dehydrogenase 1 (HSD17B1) catalyzes the conversion between estrone (E1) and estradiol (E2). The reaction is reversible in vitro, but the data in cultured cells suggest that E2 production is the predominant reaction in physiological conditions. However, the hypothesis has not been verified in vivo. In the present study, estrogen-dependent MCF-7 human breast cancer cells were stably transfected with an expression plasmid for human HSD17B1. The enzyme efficiently converted E1 to E2 and enhanced the estrogendependent growth of cultured MCF-7 cells in the presence of hormonally less active E1. The HSD17B1-expressing cells also formed estrogen-dependent tumors in immunodeficient nude mice. After treating the mice with an appropriate dose of the substrate (E1, 0.1 mol/kg䡠d), a marked difference in tumor growth was observed between nontransfected and HSD17B1-
H
YDROXYSTEROID (17-  ) DEHYDROGENASES (HSD17Bs) catalyze the interconversions between the hormonally less active 17-keto steroids and the highly active 17-hydroxy steroids, thereby controlling the last step in the formation of the highly potent androgens and estrogens. A number of HSD17B enzymes having differential expression profiles, substrate specificities, and regulatory mechanisms have been identified, and currently more than 10 different HSD17B enzymes with steroidmetabolizing activity have been characterized (1), of which HSD17B type 1 (HSD17B1) catalyzes the production of estradiol (E2) and type 2 (HSD17B2) is responsible for E2 inactivation. It is hypothesized that HSD17B enzymes play a significant role in regulating the availability of the highly active ligands for receptor binding at estrogen target cells (2). Furthermore, a number of estrogendependent diseases have been shown to be associated with aberrant expression of HSD17B1 in the target tissues. Studies have indicated that, especially in breast cancer tissue, the intracellular E2 concentration is often higher First Published Online August 17, 2006 * These authors have contributed equally to the work. Abbreviations: E1, Estrone; E2, estradiol; ESR1, estrogen receptor ␣; HSD17B, hydroxysteroid (17-) dehydrogenase; HSD17B1, HSD17B type 1; HSD17B2, HSD17B type 2; qRT-PCR, quantitative RT-PCR; STS, estrone sulfatase. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.
transfected MCF-7 cells, mean tumor weights at the end of E1 treatment being 23.2 and 130.4 mg, respectively. Furthermore, estrogen-dependent growth of the HSD17B1-expressing xenografts in the presence of E1 was markedly inhibited by administering 5 mol/kg䡠d of a specific HSD17B1 inhibitor. After a 4-wk treatment, the tumor size was reduced by 59.8% as compared with the nontreated tumors, whereas the uterine growth of the mice was not affected by the HSD17B1 inhibitor used. This was in line with the induction of apoptosis of the tumors. The results evidently show that estrogenic response for E1 is enhanced by the local action of HSD17B1 in vivo, and thus, the enzyme is a potential target for pharmacological inhibition of estrogen action. (Endocrinology 147: 5333–5339, 2006)
than that found in normal breast (3, 4). The presence of HSD17B1 in up to 50 – 60% of the breast tumors is in line with these studies (5, 6). Therefore, HSD17B1 is considered to be a promising therapeutic target (7), i.e. to inhibit estrogen-dependent growth of breast cancer in postmenopausal women (8, 9). In addition to breast, the enzyme is also expressed in other sex steroid target tissues, such as endometrium (10) and endometriotic tissue (11). The reaction catalyzed in vitro is reversible, and thus, there is ongoing debate whether the enzyme is able to form a concentration gradient of E2 between plasma and the cells expressing the enzyme. Studies in cultured cells suggest that HSD17B1 prefers the production of E2, whereas less estrone (E1) is produced (10), but models in vivo are essential to support this hypothesis. Proliferation of the human breast carcinoma cell line MCF-7 is estrogen dependent, and the cells can be grown as tumors in immunodeficient mice (12–14). In previous studies of different authors, the estrogen-dependent growth of MCF-7 cells has been studied, i.e. after stable transfection with human E2-synthesizing enzymes, such as aromatase (CYP19A1, cytochrome P450, family 19, subfamily A, polypeptide 1) (15, 16) and estrone sulfatase [STS, steroid sulfatase (microsomal), arylsulfatase C, isozyme S] (17). These studies have shown that the estrogen concentration promoting the tumor growth can be supplied by locally expressed steroid metabolizing enzymes. In the present study, human MCF-7 breast cancer cells were stably transfected with human HSD17B1, and the role of the enzyme in
5333
5334
Endocrinology, November 2006, 147(11):5333–5339
modifying estrogen-dependent growth of the cells in the presence of E1 was evaluated in tumor xenografts. Materials and Methods Reagents E1 and E2 were purchased from Sigma Aldrich (Taufkirchen, Germany). E2 pellets (total dose of 0.72 mg for 60 d release; Innovative Research of America, Sarasota, FL) and E1 injections of 8 g/kg䡠d sc in 1% EtOH in sesame oil were used in estrogen sensitivity study in vivo. For further studies, E2 was dissolved in 10% hydroxypropyl--cyclodextrin for sc injection. The specific HSD17B1 inhibitor B10721325 [3-hydroxy15-(4-morpholin-4-yl-4-oxo-butyl)-estra-1,3,5(10)-trien-17-one; see Ref. 18, example 1] had been proven to inhibit conversion of E1 to E2 by recombinant human HSD17B1 (IC50 of ⬃1 m). This compound was synthesized and provided by Solvay Pharmaceuticals Chemical Design and Synthesis unit (Hannover, Germany). E1 and B10721325 were dissolved in 50% dimethylsulfoxide/50% propanediol to be applied via ALZET osmotic minipumps (DURECT Corp., Cupertino, CA; distributed by Charles River, Sulzbach, Germany). ICI 182,780 (Faslodex) was obtained from Astra Zeneca (Cheshire, UK).
Husen et al. • HSD17B1 in Vivo Model
Analyzing estrogen-dependent tumor growth and its inhibition in vivo On d 1, each mouse was injected with 0.1 ml of the cell suspension (MCF-7-HSD17B1 clone 12, 2.5 ⫻ 106 or 4 ⫻ 106 cells) sc in both flanks. On d 1–7, the mice were daily injected with 1 mol/kg E2 sc. On d 8, osmotic minipumps were implanted sc for continuous delivery of different treatments over a time period of 14 d (ALZET microosmotic pump, model 1002). Pumps released vehicle only (50% dimethylsulfoxide/50% propanediol), 0.1 mol/kg䡠d E1, or 0.1 mol/kg䡠d E1 ⫹ 5 mol/kg䡠d HSD17B1 inhibitor B10721325. After 14 d the pumps were replaced by new ones, extending the treatments by an additional 2 wk. ICI 182,780 was administered in its commercial formulation at 179 mg/kg sc once per week starting on d 8. Growth of the xenografts was determined once a week by measuring the largest and smallest tumor diameter (d1, d2) with calipers. Tumor area was calculated according to the formula: tumor area ⫽ 䡠 d1/2 䡠 d2/2. At the end of the experiment, the mice were killed by CO2 asphyxiation, and wet weights of tumors and uteri were determined. For histological examination, tumors were immediately fixed in 0.4% formaldehyde.
Immunohistochemistry Establishing MCF-7 cells expressing human HSD17B1 MCF-7 cells were stably transfected with a pRC-CMV plasmid (Invitrogen, Carlsbad, CA) expressing human HSD17B1 (accession no. NM 000413) using FuGene 6 transfection reagent (Roche Applied Science, Mannheim, Germany). After clonal selection using G-418 sulfate, 22 cell lines were retrieved. Both reductive and oxidative HSD17B activities in cultured cells were assessed in triplicate as previously described (19). The estrogen-dependent growth of the cell clones in vitro was measured in five replicates using a MTT Cell Proliferation Kit I (Roche Applied Science) as previously described (19). For production of the xenografts in nude mice, we selected two cell lines with different levels of enzyme activity and high proliferation response to E1 (MCF-7-HSD17B1, clones 12 and 17). For inoculation into immunodeficient mice, subconfluent MCF-7 or MCF-7-HSD17B1 cells were harvested and resuspended in Matrigel basement membrane matrix, Phenol-Red-Free (BD Biosciences, Heidelberg, Germany).
Formaldehyde-fixed tumors were embedded in paraffin. Five-micrometer-thick sections were deparaffinized, hydrated, and stained according to the Dako-LSAB AP protocol using new fuchsin as chromogen (Dako Diagnostika, Hamburg, Germany). As primary antibody we used mouse antihuman Ki67 proliferation antigen (Ki67, clone MIB-1; Dako; 1:50 dilution). Apoptosis was detected using ApopTag peroxidase in situ oligo ligation apoptosis detection kit (Chemicon International, Temecula, CA) according to the manufacturer’s instructions. Diaminobenzidine was used as the substrate for visualization of the enzymatic reactions. In negative controls the primary antibody was replaced by nonimmune mouse IgG. The negative control for apoptosis was prepared without a DNA ligase. All sections were counterstained by hematoxylin. For determination of labeling indices for Ki-67 and apoptosis, a single observer counted 1000 cells from three different tumors of each treatment group using representative photomicrographs. Subsequently the percentage of labeled cells was calculated.
Animals
Quantitative RT-PCR
Adult female BALB/cABom- or Crl:NU/NU mice at 10 –11 wk of age were obtained from Taconic Europe (Ejby, Denmark) or Charles River (Sulzfeld, Germany), respectively. The animals were housed under controlled conditions of light and humidity and received food and water ad libitum. The study protocol was approved by the local animal care committee (University of Turku, Finland, and Bezirksregierung Hannover, Germany).
Expression of human ESR1, HSD17B1, and HSD17B2 genes were determined in MCF-7 cells, MCF-7-HSD17B1 clones 12 and 17, and tumors derived from MCF-7 and MCF-7-HSD17B1 clone 12 cells. RNA was isolated with Trizol reagent (Invitrogen) and further purified with NucleoSpin RNA II-RNA isolation kit (Macherey-Nagel, Du¨ren, Germany). Quantitative RT-PCR (qRT-PCR) was performed using DyNAmo HS qRT-PCR kit for two-step SYBR Green qRT-PCR (Finnzymes Oy, Espoo, Finland). Expression of all genes analyzed was normalized for human hypoxanthine phosphoribosyltransferase gene expression. The following forward and reverse primers were used, respectively: human ESR1 5⬘-TGG AGA TCT TCG ACA TGC TG-3⬘ and 5⬘-GCC ATC AGG TGG ATC AAA GT-3⬘; human HSD17B1 5⬘-ACA CCT TCC ACC GCT TCT AC-3⬘ and 5⬘-GAA CGT CGC CGA ACA CTT-3⬘; human HSD17B2 5⬘-AAC TGA TGG GGA GCT TCT TCT TAT-3⬘ and 5⬘-CCT CCT CCC ATG CTG CTG ACA-3⬘; and human hypoxanthine phosphoribosyltransferase (HPRT) 5⬘-TGC TCG AGA TGT GAT GAA GG-3⬘ and 5⬘-TCC CCT GTT GAC TGG TCA TT-3⬘.
Estrogen sensitivity of MCF7-HSD17B1 clones 12 and 17 in vivo The mice were ovariectomized under anesthesia according to standard procedures, and the E2 pellets (0.72 mg for 60 d release, Innovative Research of America) were implanted sc 1–2 d before cell inoculation to induce maximal tumor growth. On experimental d 1, each mouse received in both flanks a sc injection of 2.5 ⫻ 106 cells per 0.1 ml MCF-7 or MCF7-HSD17B1 cells (clones 12 or 17). The size of the tumors formed was determined by measuring three diameters (d1, d2, d3) with calipers twice a week throughout the experiment. Tumor volumes were calculated according to the formula V ⫽ 4/3 䡠 d1/2 䡠 d2/2 䡠 d3/2. On d 33, the E2 pellets were removed and the size of xenografts was followed for 2 more weeks. On d 47–58, the mice were daily injected with 8 g/kg E1 sc, and the tumor size was followed up as before. At the end of the experiment, the mice were killed by CO2 asphyxiation. The tumor growth rates in the presence of E2 (d 19 –31) and E1 (d 47–58) were then calculated (⌬tumor size/⌬treatment time) to demonstrate the estrogen responsiveness of the cell clones presenting differential growth properties. According to these results, MCF-7-HSD17B1 clone 12 was selected for further studies.
Statistical analysis Mean ⫾ sd was calculated for enzyme activities, cell proliferation, tumor areas, tumor wet weights, and uterine wet weights for each treatment group. One-way-ANOVA and Tukey’s post test were performed using Sigma Stat 3.1 (SPSS Inc., Chicago, IL) for in vitro studies and GraphPad Prism 4.0 (GraphPad Software, Inc., San Diego, CA) for in vivo studies. For qRT-PCRs, mean ⫾ sem was calculated and analyzed by Student’s t test or Mann-Whitney rank sum test, one-way-ANOVA, or Kruskal-Wallis one-way-ANOVA of ranks using SigmaStat 3.1 (SPSS).
Husen et al. • HSD17B1 in Vivo Model
Endocrinology, November 2006, 147(11):5333–5339
TABLE 1. Conversion of E1 to E2 (percent of the substrate converted) in cultured MCF-7 parental cells and seven different MCF-7-HSD17B1 cell clones Cell clone no.
MCF-7 7 10 12 15 17 31 35
Time (h) 0.5
1
2
3
4
24
0 17 18 27 20 3 20 4
0 33 32 44 37 5 34 9
0 60 54 72 69 6 62 10
0 73 75 80 75 16 75 19
0 86 83 89 90 23 89 28
0 99 100 100 100 100 100 100
Cultures of 200,000 cells/well were incubated with 0.4 nmol substrate in six-well plates.
Results HSD17B1 expression enhances E1 to E2 conversion in cultured cells in vitro
MCF-7 cells were stably transfected with a plasmid expressing human HSD17B1 under CMV-promoter, and 22 clonal cell lines were generated. In the culture conditions used, a marked reductive in HSD17B activity (conversion of E1 to E2) was detected in seven of them (Table 1), whereas in the parental MCF-7 cells, no activity was detected. Furthermore, no significant oxidative activity was found in either the nontransfected or HSD17B1-transfected MCF-7 cells. The clones were then tested for their estrogen responsiveness in culture by measuring the estrogen-dependent proliferation. All the clones were responsive to E2 to a very similar degree as the parental MCF-7 cells, in which a 5-pm concentration of E2 was already significantly increasing the growth rate. Compared with the nontransfected cells, the HSD17B1expressing clones were significantly more sensitive to E1. In the parental cell line, 50 pm of E1 induced growth, whereas in HSD17B1-expressing cell clones 12 and 17, already a 5-pm E1 concentration increased the growth rate significantly more than in nontransfected MCF-7 cells (Fig. 1). Furthermore, in HSD17B1-expressing cells, a similar growth rate was detected with E1 and E2 with doses higher than 5 pm. The basal growth rates of these clones were variable, and in the absence of estrogens, clone 12 showed an identical growth rate to the MCF7 cells, whereas clone 17 was growing faster in vitro (data not shown). FIG. 1. Stable transfection of MCF-7 cells with human HSD17B1 enhances their estrogen sensitivity in vitro. Cell number (mean ⫾ SD) of parental nontransfected MCF-7 cells (MCF-7) and MCF-7 cells stably expressing HSD17B1 (MCF-7-HSD17B1 no. 12 and no. 17) were determined after 7-d culture in the presence of absence of different doses of E1 (white bars) and E2 (black bars). In the presence of identical E1 concentration, the proliferation of the HSD17B1-expressing cells is increased significantly, compared with the parental MCF-7 cells, as denoted by the asterisks. **, P ⫽ 0.004; ***, P ⱕ 0.001.
5335
HSD17B1 expression enhances E1 action in MCF-7 cells in vivo
Based on HSD17B1 expression, estrogen responsiveness and basal growth rate in vitro, we selected MCF-7-HSD17B1 clones 12 and 17 to be evaluated in vivo. The formation of tumors in ovariectomized nude mice and the tumor growth rates in the presence of E2 and E1 were evaluated to find out whether the model could be useful for testing of HSD17B1 inhibitors in vivo. Clone 12 expressing a high amount of HSD17B1 showed a similar growth rate in vivo in the presence of both about 480 g/kg䡠d E2 and even lower doses of 8 g/kg䡠d E1, indicating an effective conversion of E1 to E2 in the tumors (Fig. 2). However, growth rates of parental MCF-7 cells and the clone 17 with lower level of HSD17B1 expression were remarkably reduced with E1. Expression of HSD17B1 and ESR1 was analyzed and compared in cultured cells and tumor xenografts using qRT-PCR (Fig. 3) and immunohistochemistry (data not shown). As expected by the activity measurements in cultured cells, HSD17B1 expression was detected only in the MCF-7HSD17B1 cells and tumor xenografts derived from them. ESR1 was present in all types of cells and xenografts. qRTPCR was also used to confirm the absence of HSD17B2 (data not shown) in all types of cells and tumors. Based on these results, we used the MCF-7-HSD17B1 clone 12 cells for further studies in vivo. To assess inhibition of HSD17B1 in terms of E1-dependent tumor growth, it is essential to choose a dose of E1 that has only negligible proliferative effect on tumor growth on its own. Tumor growth should be fully dependent on conversion of E1 to E2. The tumor formation of HSD17B1-expressing and parental MCF-7 cells was first supported by E2 during the first week after inoculation, after which estrogen-dependent growth of the cells was analyzed in the presence of E1. Studies were conducted treating the mice with various doses of E1 (0.008, 0.04, 0.2, or 1 mol/kg䡠d sc) via osmotic minipumps implanted, and an optimized dose of 0.1 mol/kg䡠d of E1 was identified, which did not cause significant estrogenic effect on nontransfected MCF-7 cells but a marked reduction in the tumor area (Fig. 4A) and a low tumor weight at autopsy (Fig. 4B, top). In contrast, the HSD17B1-expressing MCF-7 cells showed a clear estrogen response with this E1 dose because the tumor area was maintained during the E1 treatment (Fig.
5336
Endocrinology, November 2006, 147(11):5333–5339
FIG. 2. The growth rates of the nontransfected parental MCF-7 cells (open circle), MCF-7-HSD17B1 no. 12 (filled square) and MCF-7HSD17B1 no. 17 cells (filled triangle) as tumor xenografts were determined in the presence of E2 (0.72 mg per 60 d release) and E1 (8 g/kg䡠d). In the presence of E1, the growth rate of MCF-7 and MCF7-HSD17B1 no. 17 (low level of HSD17B1 expression) was markedly reduced, compared with that observed with E2. In contrast, the MCF7-HSD17B1 no. 12 cells expressing high amounts of HSD17B1 showed a similar growth in the presence of both E2 and E1. This indicated an effective conversion of E1 to E2 in the tumors. Asterisks denote significant difference in tumor growth rate in the presence of E1 vs. E2. *, P ⫽ 0.03; **, P ⫽ 0.004; n ⫽ 16 tumors for MCF-7 and n ⫽ 7 tumors each for clones 12 and 17.
Husen et al. • HSD17B1 in Vivo Model
In contrast to the tumor weight, the uterine weight was similar in the mice treated with E1 or E1 plus the HSD17B1 inhibitor (Fig. 5B). This indicates that the inhibitor did not alter the systemic E1 action and that the compound was devoid of estrogenic or antiestrogenic properties. Our results evidently show that local HSD17B1 expression enhances estrogen response in the presence of E1 and that the nude mouse model is suitable for testing the efficacy of HSD17B1 inhibitors in vivo. According to histopathology the tumors of vehicle-treated mice and all ICI 182,780-treated animals were highly necrotic (Fig. 6). In addition, necrosis was found in tumors of mice treated with the HSD17B1 inhibitor, but it was less obvious in those treated with E1 only. The labeling for the proliferation marker Ki67 was reduced in tumors of vehicle-treated and E1 ⫹ ICI 182,780-treated tumors, compared with those treated with E1 only or E1 ⫹ HSD17B1 inhibitor. The reduced tumor weight in HSD17B1 inhibitor-treated mice was associated with induced apoptosis (Fig. 7). This finding suggests a less aggressive effect on tissue integrity for HSD17B1 inhibitors, compared with pure antiestrogens. Discussion
4, A and B, top). In contrast to the different effect of E1 on tumor weights between the MCF-7 and MCF-7-HSD17B1 cells, a similar increase in the uterine weights was observed for both treatments (Fig. 4B, bottom), indicating that the systemic estrogen exposure for the animals was identical. E1 response in HSD17B1-expressing MCF-7 cells is reduced by a specific HSD17B1 inhibitor in vivo
We next studied whether a HSD17B1 inhibitor could reduce the estrogen-dependent growth induced by E1 in MCF7-HSD17B1 cells (clone 12) grown in the presence of 0.1 mol/kg䡠d of E1. In line with our hypothesis, the data indicated that treating the mice with 5 mol/kg䡠d of a specific HSD17B1 inhibitor, a significant reduction (of up to 60%) was observed in the mean tumor weight of HSD17B1-expressing MCF-7 tumors (Fig. 5). However, using the pure antiestrogen ICI 182,780, an additional reduction in the tumor growth was obtained. This difference is in line with the growth-promoting effect of the systemic E1 provided by the osmotic pumps.
FIG. 3. Expression of HSD17B1 and ESR1 mRNA was analyzed by quantitative RTPCR. As expected, in line with the activity measurements, HSD17B1 expression was detected only in the cultures of stably transfected cells (no. 12, no. 17; A) and tumors formed from clone 12 (MCF-7-HSD17B1; B). ESR1 expression was found in all cells and tumors. Asterisks denote statistical significance level vs. parental MCF-7 cell tumors. **, P ⱕ 0.010; ***, P ⱕ 0.001. HPRT, Hypoxanthine phosphoribosyltransferase.
Regulation of sex steroid receptor-mediated signaling by ligand metabolism is an attractive therapeutic option. Accordingly, we and others have hypothesized that HSD17Bs are able to modulate sex steroid action via catalyzing the reaction between the low and the highly active sex steroids locally at the target cells. HSD17B1 is highly expressed in about 50 – 60% of postmenopausal breast cancer specimens (5, 6) and thus could have a role in estrogen-dependent growth of these tumors. The enzyme is also expressed in normal endometrium (10) and endometriotic tissues (11). In postmenopausal women the circulating androgens are likely to be the major precursor for local E2 formation, and HSD17B1 activity together with that of CYP19A1 is expected to be essential for E2 production locally in peripheral tissues. However, there is a difference in the intratissular localization of the enzymes: CYP19A1 is typically in the stromal compartment, whereas HSD17B1 is expressed in the epithelial cells (5, 6). Recent studies have also shown that HSD17B1 expression is associated with poor prognosis of breast cancer
Husen et al. • HSD17B1 in Vivo Model
FIG. 4. Development of tumors and uteri in mice with and without 0.1 mol/kg䡠d E1 supplementation after inoculation with 2.5 ⫻ 106 cells/ flank of either nontransfected MCF-7 cells (MCF-7) or MCF-7 cells expressing HSD17B1 clone 12 (MCF-7-HSD17B1). Mean ⫾ SEM, n ⫽ 5 animals/treatment. A, Growth curves of tumors show that only E1-supplemented tumors from MCF-7-HSD17B1 cells maintain their initial size, whereas all other tumors are regressing. B, Tumor weights and uterine weights after 6 wk of E1 treatment. At the chosen dose of E1, MCF-7-HSD17B1 tumors have retained their size, although uterine growth is comparable in animals bearing MCF-7 tumors and animals bearing MCF-7-HSD17B1 tumors. Asterisks denote significance level vs. E1 treatment at the same time point. *, P ⱕ 0.05; **, P ⱕ 0.01.
patients (8, 9), further indicating a putative role for the enzyme in the growth and progression of breast cancer. In the present study, we have shown that HSD17B1 expression in the breast cancer cells is able to enhance the estrogen dependency of the cells in vivo in the presence of the weakly active E1. Whereas the enzyme catalyzes a reversible reaction in vitro, the present study provides for the first time direct
Endocrinology, November 2006, 147(11):5333–5339
5337
FIG. 5. Development of tumors and uteri in mice treated with the pure antiestrogen ICI 182,780 (ICI) and HSD17B1 inhibitor B10721325 (HSD-Inh.) with and without E1 supplementation after inoculation with 4 ⫻ 106 of MCF-7-HSD17B1 cells (clone 12) per flank (mean ⫾ SEM, n ⫽ 6 animals/treatment). A, Growth curves of tumors show that those mice treated with only E1 maintain their initial tumor sizes, whereas all other tumors are regressing to different extents. B, Tumor weights and uterine weights after 4 wk of treatment. ICI completely abolishes systemic effects of endogenous and exogenous estrogens regardless of any additional treatment. Tumor weights but not uterine weights are reduced to the same level as observed with vehicle only where systemic endogenous estrogens are available. In contrast, the specific HSD17B1 inhibitor acts locally, affecting only the HSD17B1-expressing tumor tissue but not uterine weights. Asterisks denote significance level vs. treatment with only E1 at the same time point. Frames summarize groups of the same significance level. *, P ⱕ 0.05; **, P ⱕ 0.01; ***, P ⱕ 0.001.
evidence for the role of local HSD17B1 as enhancing the E1 action in vivo. The growth of the estrogen-dependent dimethyl-benz(a)anthracene-induced mammary gland tumors in the rat has also been suggested to involve local action of HSD17B1 (20). However, the contribution of the HSD17B1 to
5338
Endocrinology, November 2006, 147(11):5333–5339
Husen et al. • HSD17B1 in Vivo Model
FIG. 6. Immunohistochemical staining for Ki67 proliferation antigen and labeling of apoptotic cells in tumors originating from MCF-7 cells expressing HSD17B1 after administration of vehicle, E1, E1 with the HSD17B1 inhibitor B10721325, and E1 with the pure antiestrogen ICI 182,780 (ICI). Tumors of mice treated with vehicle or ICI appear highly necrotic.
the estrogen receptor-mediated growth could not be adequately determined. The estrogen-dependent human breast cancer cell line
FIG. 7. The labeling index for the proliferation marker Ki67 (white bars) was reduced in all tumors treated with ICI 182,780 (ICI) and vehicle, compared with tumors treated with E1 and E1 ⫹ the HSD17B1 inhibitor (HSD-Inh.). The labeling index for apoptosis (black bars) was increased only in tumors treated with E1 ⫹ HSD17B1 inhibitor. Three tumors from each treatment group were evaluated. In the case of apoptotic cells, asterisks denote significance level vs. treatment with E1 ⫹ HSD-Inh. In the case of Ki67-positive cells, asterisks denote significance level vs. treatment with E1 as well as vs. treatment with E1 ⫹ HSD-Inh. *, P ⱕ 0.05, **, P ⱕ 0.01; ***, P ⱕ 0.001.
MCF-7 has been widely used as a tumor model in vivo (12) and for studying factors affecting the estrogen responsiveness in vivo. The cells are characterized by only a low level of endogenous HSD17B activity (10, 19), offering the opportunity to study the role of various transfected HSD17Bs, i.e. on estrogen-dependent growth. In the present study, HSD17B1-mediated tumor growth could be assessed and differentiated from the HSD17B1-independent growth. Our studies also showed for the first time that such a model is suitable for evaluating the efficacy of HSD17B1 inhibitors in vivo. Analogous models have been used to evaluate CYP19A1 or STS inhibitors (15–17). But for the evaluation of CYP19A1 and STS action, the respective substrates (androstenedione, estrone sulfate) show no, or only very low, estrogenicity and thus could be administered in more or less unlimited amounts. However, this is not possible with the weakly active estrogen E1 used as a substrate for HSD17B1 as in the present study. Comparing the tumor and uterine weights, we were able to distinguish the local intratumoral effects of human HSD17B1 action from the systemic estrogenic effects of the substrate provided. In the mouse uterus, HSD17B1 has been detected (21), but our data suggest that the mouse enzyme does not contribute to the estrogenic response in the uterus.
Husen et al. • HSD17B1 in Vivo Model
Endocrinology, November 2006, 147(11):5333–5339
With the study design used, we were also able to show that the inhibitor tested acts on only the human HSD17B1 but not murine ovarian HSD17B1, being the main source of systemic E2 supply in adult female mice. In our study, the uteri of all mouse groups were growing similarly due to the estrogenic effect of the E1 treatment. The inhibitor used affected only the human HSD17B1 expressed in the MCF-7 cell tumors. Evidently the local intracrine action of human HSD17B1 enhanced the estrogen sensitivity of MCF-7 cells in vivo. Estrogenic or antiestrogenic action of the inhibitor could also be ruled out with the similar uterus weight of the nontreated and inhibitor-treated mice. The growth inhibition of the tumors by the HSD17B1 inhibitor was weaker than was observed in the ICI 182,780-treated mice, which is in line with the less necrotic appearance of the tumors in the HSD17B1 inhibitor-treated mice. Thus, the data strongly support the concept that targeted inhibition of HSD17B1 is a promising therapeutic approach to modulate estrogen response in the target tissues with less severe side effects as compared with antiestrogens.
12.
Acknowledgments
13.
We are grateful to Sabine Brandes, Birgit Ku¨hlein, and Szilvia Eiterer for excellent technical assistance, and we are indebted to Lauri Kangas and colleagues from Hormos Medical (Turku, Finland) for helpful advice. Harry Kujari (Department of Pathology, University of Turku, Turku, Finland) is acknowledged for histopathological evaluation of the tumors. Received June 9, 2006. Accepted August 10, 2006. Address all correspondence and requests for reprints to: Bettina Husen, Solvay Pharmaceuticals Research Laboratories, Hans-Bo¨cklerAllee 20, 30173 Hannover, Germany. E-mail:
[email protected]. Disclosure summary: K.H., T.S., and M.P. have nothing to declare. B.H., J.M., and H.H.T. are employed by Solvay Pharmaceuticals. B.H., J.M., and H.H.T. are inventors on patent WO2005047303.
References 1. Mindnich R, Mo¨ller G, Adamski J 2004 The role of 17 -hydroxysteroid dehydrogenases. Mol Cell Endocrinol 218:7–20 2. Penning TM 2003 Hydroxysteroid dehydrogenases and pre-receptor regulation of steroid hormone action. Hum Reprod Update 9:193–205 3. Bonney RC, Reed MJ, Davidson K, Beranek PA, James VH 1983 The relationship between 17-hydroxysteroid dehydrogenase activity and oestrogen concentrations in human breast tumours and in normal breast tissue. Clin Endocrinol (Oxf) 19:727–739 4. McNeill JM, Reed MJ, Beranek PA, Bonney RC, Ghilchik MW, Robinson DJ,
5.
6. 7. 8. 9. 10.
11.
14. 15. 16. 17.
18. 19. 20.
21.
5339
James VH 1986 A comparison of the in vivo uptake and metabolism of 3Hoestrone and 3H-oestradiol by normal breast and breast tumour tissues in postmenopausal women. Int J Cancer 38:193–196 Sasano H, Frost AR, Saitoh R, Harada N, Poutanen M, Vihko R, Bulun SE, Silverberg SG, Nagura H 1996 Aromatase and 17-hydroxysteroid dehydrogenase type 1 in human breast carcinoma. J Clin Endocrinol Metab. 81:4042– 4046 Poutanen M, Isomaa V, Lehto VP, Vihko R 1992 Immunological analysis of 17-hydroxysteroid dehydrogenase in benign and malignant human breast tissue. Int J Cancer 50:386 –390 Poirier D 2003 Inhibitors of 17-hydroxysteroid dehydrogenases. Current Med Chem 10:453– 477 Oduwole OO, Li Y, Isomaa VV, Mantyniemi A, Pulkka AE, Soini Y, Vihko PT 2004 17-hydroxysteroid dehydrogenase type 1 is an independent prognostic marker in breast cancer. Cancer Res 64:7604 –76049 Gunnarsson C, Hellqvist E, Stal O 2005 17-Hydroxysteroid dehydrogenases involved in local oestrogen synthesis have prognostic significance in breast cancer. Br J Cancer 92:547–552 Miettinen MM, Mustonen VJ, Poutanen MH, Isomaa VV, Vihko RK 1996 Human 17-hydroxysteroid dehydrogenase type 1 and type 2 isoenzymes have opposite activities in cultured cells and characteristic cell- and tissuespecific expression. Biochem J 314:839 – 845 Zeitoun K, Takayama K, Sasano H, Suzuki T, Moghrabi N, Andersson S, Johns A, Meng L, Putman M, Carr B, Bulun SE 1998 Deficient 17-hydroxysteroid dehydrogenase type 2 expression in endometriosis: failure to metabolise 17-estradiol. J Clin Endocrinol Metab 83:4474 – 4480 Shafie SM, Grantham FH 1981 Role of hormones in the growth and regression of human breast cancer cells (MCF-7) transplanted into athymic nude mice. J Nat Cancer Inst 67:51–56 Noel A, Simon N, Raus J, Foidart JM 1992 Basement membrane components (Matrigel) promote the tumorigenicity of human breast adenocarcinoma MCF7 cells and provide an in vivo model to assess the responsiveness of cells to estrogen. Biochem Pharmacol 43:1263–1267 Clarke R 1996 Human breast cancer cell line xenografts as models of breast cancer: the immunobiologies of recipient mice and the characteristics of several tumorigenic cell lines. Breast Cancer Res Treatment 39:69 – 86 Yue W, Zhou D, Chen S, Brodie A 1994 A new nude mouse model for postmenopausal breast cancer using MCF-7 cells transfected with the human aromatase gene. Cancer Res 54:5092–5095 Yue W, Wang JP, Hamilton CJ, Demers LM, Santen RJ 1998 In situ aromatization enhances breast tumor estradiol levels and cellular proliferation. Cancer Res 58:927–932 Nakata T, Takashima S, Shiotsu Y, Murakata C, Ishida H, Akinaga S, Li PK, Sasano H, Suzuki T, Saeki T 2003 Role of steroid sulfatase in local formation of estrogen in post-menopausal breast cancer patients. J Steroid Biochem Mol Biol 86:455– 460 Messinger J, Thole HH, Husen B, Van Steen B, Schneider G, Hulshoff JBE, Koskimies P, Johansson N, Adamski J 2005 Novel 17-hydroxysteroid dehydrogenase inhibitors, Patent WO 2005047303, pp 1–189 Miettinen MM, Poutanen MH, Vihko RK 1996 Characterization of estrogen dependent growth of cultured MCF-7 human breast-cancer cells expressing 17-hydroxysteroid dehydrogenase type I. Int J Cancer 68:600 – 604 Li S, Martel C, Dauvois S, Belanger A, Labrie F 1994 Effect of estrone on the growth of 7,12-dimethylbenz(a)anthracene-induced mammary carcinoma in the rat: a model of postmenopausal breast cancer. Endocrinology 134:1352– 1357 Nokelainen P, Puranen T, Peltoketo H, Orava M, Vihko P, Vihko R 1996 Molecular cloning of mouse 17-hydroxysteroid dehydrogenase type 1 and characterization of enzyme activity. Eur J Biochem 236:482– 490
Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.
