This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Qu, Q. Articles by Härkönen, P. Search for Related Content PubMed PubMed Citation Articles by Qu, Q. Articles by Härkönen, P.

Selective Estrogenic Effects of a Novel Triphenylethylene Compound, FC1271a, on Bone, Cholesterol Level, and Reproductive Tissues in Intact and Ovariectomized Rats¹

Institute of Biomedicine, Department of Anatomy, and MediCity Research Laboratory, University of Turku (Q.Q., H.Z., J.D., N.C., Z.P., K.V., P.H.); Orion Corp. (A.L.); and Hormos Medical Ltd. (L.K.), 20520 Turku, Finland; and the Center for Nutrition and Toxicology, Karolinska Institute (M.K., K.H.), S-14157 Huddinge, Sweden

Address all correspondence and requests for reprints to: Dr. Pirkko Härkönen, Institute of Biomedicine, Department of Anatomy, and Medicity Research Laboratory, University of Turku, Kiinamyllynkatu 10, Turku 20520, Finland. E-mail: pirkko.harkonen{at}utu.fi

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
FC1271a is a novel triphenylethylene compound with a tissue-selective profile of estrogen agonistic and weak antagonistic effects. It specifically binds to the estrogen receptor and ß with affinity closely similar to that of toremifene and tamoxifen. To study the in vivo effects of the compound, 4-month-old rats were sham operated (sham) or ovariectomized (OVX) and treated daily for 4 weeks with various doses of FC1271a or vehicle (orally). FC1271a was able to oppose OVX-induced bone loss by maintaining the trabecular bone volume of the distal femur. Accordingly, the OVX-induced loss of bone strength was prevented at doses of 1 and 10 mg/kg. FC1271a also prevented the OVX-induced increase in serum cholesterol in a dose-dependent manner. No significant changes in uterine wet weight or morphology were observed in the OVX-rats treated with 0.1 or 1 mg/kg FC1271a, but at a dose of 10 mg/kg it had a slightly estrogenic effect. In immature rats the effect of FC1271a on uterine wet weight was less stimulatory than that of toremifene or tamoxifen, but more stimulatory than that of raloxifene or droloxifene. The appearance of the dimethylbenzanthracene (DMBA)-induced mammary tumors was inhibited by treatment of DMBA-treated rats with FC1271a in a dose-dependent manner. In human MCF-7 breast cancer cell tumors raised in nude mice in the presence of estrogen, the growth and expression of pS2 marker gene could not be maintained after estrogen withdrawal by treatment with FC1271a. No formation of DNA adducts was observed in the liver of the FC1271a-treated rats. In conclusion, the bone-sparing, antitumor, and cholesterol-lowering effects of FC1271a combined with a low uterotropic activity and lack of liver toxicity indicate that FC1271a could be an important alternative in planning antiosteoporosis therapy for estrogen deficiency.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
DECREASING production of estrogen in aging women has been associated with degenerative changes in various organ systems, including the skeleton, cardiovascular system, and nervous system (1, 2, 3). In many women, decreasing levels of estrogen are also associated with menopausal symptoms. Estrogen replacement therapy effectively protects against degenerative changes as well as menopausal symptoms (4). However, long term estrogen therapy, even in combination with progestins, may also cause undesirable side-effects, including an increased risk of breast and uterine cancer (5, 6). Therefore, several attempts have recently been taken to develop new hormone replacement therapies with selective estrogen-like effects in bone, cardiovascular system, and/or central nervous system without stimulation or tumor promotion in reproductive or other organs (7).

New compounds presenting selective estrogen agonist and/or antagonist properties [selective estrogen receptor (ER) modulators (SERMs)] have recently been introduced. In addition, the antiestrogens tamoxifen (TAM), toremifene (TOR), raloxifene (RAL), and droloxifene (DRO), which were originally developed for breast cancer therapy, have been found to present several SERM-type effects. They all act as partial estrogen agonists in bone and cardiovascular system in postmenopausal women (8, 9). However, none of above therapies fulfils all of the requirements of optimal hormone replacement therapy in postmenopausal women. Concerns of endometrial cancer and deep vein thrombosis as well as induction of hepatocarcinogenesis in rodents (10, 11) prevent the use of TAM for hormone replacement therapy (12). TOR seems to lack the genotoxicity of TAM but there are no experiments or trials concerning the treatment of menopausal symptoms. The next generation of SERM RAL has estrogen agonistic effects in bone and cardiovascular system without causing significant uterine stimulation, but recent reports on clinical trials indicate that it is not effective against or may even increase climacteric symptoms (13).

We have recently identified a novel triphenylethylene compound FC1271a [chemical name: Z-2-(4-(4-chloro-1,2-diphenyl-but-1-enyl)phenoxy)ethanol; Fig. 1] by screening an estrogen/antiestrogen molecule library with various in vitro test systems. This compound also presented tissue-specific agonistic and antagonistic effects in vivo (14). In the in vitro analysis FC1271a was found to exert estrogen-like effects in bone marrow cultures by enhancing osteoblastic differentiation with a mechanism that differed from that of RAL (15). The effects of FC1271 on osteoblastic differentiation could be inhibited by the pure antiestrogen ICI 182,780, suggesting an ER-mediated mechanism. In osteoclast cultures the effects of FC1271a were also similar to those of estrogen, but clearly differed from those of TAM and RAL (16).

View larger version (10K): [in this window] [in a new window]   Figure 1. Chemical structure of FC1271a [Z-2-(4-(4-chloro-1,2-diphenyl-but-1-enyl)phenoxy)ethanol].

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Compounds
FC1271a (Z-isomer), TAM (Z-isomer), TOR (Z-isomer), RAL, and DRO were synthesized and purified in the Chemical Research Laboratory of Orion Corp. (Oulu, Finland). The purity of the compounds was more than 99%.

Animal protocols for in vivo studies
Adult female Sprague Dawley rats, 2.5 months old and weighing approximately 300 g at the beginning of the experiments, were obtained from Mollegaard Breeding Center (Skensved, Denmark). For immature uterus tests, 18-day-old female rats (Mollegaard) were used. The rats were housed at 22 ± 2 C with 50 ± 20% humidity. The temperature and a 12-h light, 12-h dark cycle were automatically controlled. Commercial laboratory rat food (Special Diet Services, Essex, UK) and water were available ad libitum. The rats were kept without food overnight before the blood samples were collected for cholesterol measurement.

Rats were anesthetized with 2 mg/kg diazepam (Diapam, Orion Corp.) ip, followed by a 0.2-ml ip injection of fentanyl-fluanisone anesthetic (Hypnorm, Janssen Pharmaceuticals Ltd., Grove, UK). Bilateral ovariectomy (OVX) or corresponding sham operation (sham) was performed using a dorsal approach. Upon recovery from anesthesia, animals were randomly grouped. In the first experiment (Exp I), FC1271a was administered at doses of 1 and 10 mg/kg to intact or ovariectomized (OVX) rats with or without simultaneous estrogen treatment. In the second experiment (Exp II), five different doses of FC1271a and one dose of RAL (3 mg/kg) or DRO (10 mg/kg) were given to OVX animals. In the third experiment (Exp III), 17-EE₂ and 17ß-E₂ were given to OVX animals orally or sc, respectively, to compare the effects of SERMs with those of estrogens. In all experiments each experimental group contained an average of 10 rats. 17-EE₂ and the SERMs were dissolved in polyethylene glycol 300 (PEG300, Orion Corp.) and administered daily by oral gavage in a volume of 5 ml/kg. 17ß-E₂ was given sc in sesame oil. All animal procedures and care were approved by the ethics committee of University of Turku.

Tissue collection and processing
After the treatments, the rats were weighed and killed by CO₂ asphyxia. Blood was collected by cardiac puncture. Aliquots of serum were assayed for cholesterol, and the rest was frozen and kept at -20 C for LH, FSH, and collagen type I cross-linked C-telopeptides (ICTP) assays. The liver and half of the uterus were rapidly excised and frozen in liquid nitrogen for storage at -70 C until use, and the other half of the uterus was fixed in 70% formalin. The lumbar vertebrae and both tibia and femurs were dissected. The proximal part of the right femur, left tibia, and lumbar spine (L1–L5) were frozen and stored at -20 C until analysis. The right tibiae were burned for 24 h at 600 C to obtain ash weights. The distal part of the right femur was fixed in 40% ethanol and stored at 4 C. Urine was collected in a metabolic cage for 24 h.

Serum (fasted) cholesterol was assayed using an enzymatic method (17). Serum LH and FSH levels were measured by a RIA kit (NIDDK, Baltimore, MD). Urinary deoxypyridinoline (DPD) was analyzed using an enzyme-linked immunosorbent assay (Pyrilinks, Metra Biosystems, Inc., Mountain View, CA) and normalized, with the content of creatinine measured enzymatically (Sigma, St. Louis, MO). ICTP was measured using a RIA kit (Orion Diagnostica, Oulunsalo, Finland).

Bone biomechanical measurements
The bone specimens were thawed before biomechanical tests and were kept moist during all handling and test procedures. The proximal part of the femur was used to measure failure load of the femoral neck as described by Peng et al. (18). In brief, the proximal femur was fixed perpendicularly to a polymethyl methacrylate plate with suitable holes. A plastic support was then placed on the compression stage of the testing machine (SEY 10, Magnetic Elektromotoren AG, Wadenswil, Switzerland). The concave compressing head loaded the femoral head at a constant vertical velocity of 0.155 mm/s. The femoral head-neck complex was pressed until failure by loading the head with a force parallel to the shaft. The maximal compression load (N) was recorded by a plotter (Perkin-Elmer Corp., model 165, Hitachi Ltd., Tokyo, Japan). Lumbar vertebrae 3 (L3) was separated and trimmed to exclude the spinous, transverse, and articular processes. The central part of the vertebral body was loaded along the longitudinal axis. Maximum load (N) was obtained directly from the load deformation curves.

Bone histomorphometry
The distal one third of the left femur was fixed in 40% ethanol, dehydrated in graded concentrations of ethanol, defatted in acetone, then embedded in methyl methacrylate. Undecalcified frontal sections of the distal femoral metaphases were cut at the same site at 4- and 10-µm thickness. The 4-µm sections were stained using the Masson-Goldner-Trichrome method, whereas the 10-µm sections remained unstained.

An M2 image analyzer (Imaging Research, Inc., Brock University, Ontario, Canada) was applied for histomorphometric measurements of the secondary spongiosa of the distal femoral metaphases between 1 and 2 mm distal to the growth plate-epiphyseal junction and extended to the endocortical surface in the lateral dimension (18). The region within the first 1 mm to the growth plate was omitted to restrict measurements to the secondary spongiosa. Total tissue area, trabecular bone area, trabecular bone perimeter, osteoclast number, and osteoclast perimeter were measured in the 4-µm sections. The percent trabecular bone volume (TBV) and osteoclast number per mm bone surface were calculated accordingly. The mineral apposition rate (MAR) was calculated on the basis of interlabeled width between the calcein and tetracycline double labeling on 10-µm sections.

Determination of DNA adducts
The formation of DNA adducts in the liver of rats treated either with FC1271a or TAM was studied using the ³²P postlabeling method of DNA and HPLC analysis (19). Thirty 90-day-old female Sprague Dawley rats were randomly divided into three groups. They were given a daily oral dose of either polyethylene glycol (control group) or 45 mg/kg·day FC1271a or TAM, respectively (treatment groups). After 2 weeks the animals were killed, and their livers were weighed and stored at -70 C until DNA isolation and adduct analysis.

DMBA tumors
Mammary carcinoma was induced by treating 50-day-old female Sprague Dawley rats with a single oral dose of DMBA (20 mg/rat) in sesame oil as described by Kangas et al. (20). The tumors were allowed to develop for 7–8 weeks. The rats were then randomly divided into experimental groups and treated daily with FC1271a suspended in polyethylene glycol or vehicle only by oral gavage for 4 weeks. After treatment, the tumors were followed for another 6 weeks. Tumor occurrence and growth were monitored by weekly palpation and measurements. Tumor volume was calculated according to a formula: V(m³) = (w² x l)/12, where w and l are perpendicular diameters of the tumor (l being the longest diameter). On the basis of these measurements, the tumors were classified into three categories at the end of the experiment: 1) growing tumors (tumor volume increased >4-fold during the treatment), 2) stabilized tumors, and 3) regressing tumors (tumor volume decreased 75%). In addition, 4) the number of the tumors that disappeared during treatment and 5) the number of the tumors that appeared during treatment were recorded.

MCF-7 cell cultures
MCF-7 cells were grown in phenol red-free RPMI 1640 medium supplemented with 5% dextran-charcoal-stripped FCS as described by Wärri et al. (21, 22). On the day after plating, the medium was changed, and 17ß-E₂ or FC1271a was added in dimethylsulfoxide (final concentration, 0.1%). Vehicle was added to the control wells. For growth curves, MCF-7 cells were seeded in a 24-well plates (Nunc, Rosklide, Denmark) at a density of 25,000 cells/well. Cells were trypsinized on the days indicated, and the cell number was counted using a Coulter counter (Coulter, Harpenden, UK). For determining the rate of DNA synthesis, the cells were seeded in 96-well plates (5000 cells/well) and labeled with [methyl-³H]thymidine (0.2 µC/well; Amersham Pharmacia Biotech, Aylesbury, UK) for 2 h. The labeled cells were detached with trypsin, harvested on a glass-fiber filter (Wallac, Inc.), and counted in a flat bed scintillation counter Microbeta LSCTM (Wallac, Inc.).

Nude mouse tumors
Female athymic nude mice (8–12 weeks old; nu/nu-BALB/cABom, Bomholtgaard, Rye, Denmark) were OVX and supplemented with sc estrogen pellets (17ß-E₂, 0.72 mg/pellet·mouse, 60-day release; Innovative Research of America, Toledo, OH) before injecting MCF-7 cells (10⁷ cells in 0.2 ml saline) sc into the shoulder region under light ether anesthesia. Tumor growth was monitored twice a week, and tumor size was measured in two perpendicular dimensions. Tumor volumes were calculated according to the formula: V (mm³) = /6(d1 x d2)^3/2, where d1 and d2 are perpendicular diameters of the tumor. Tumors were allowed to grow for 6 weeks (up to 200 mm³). The control group was kept with the estrogen pellet, and the tumors were allowed to grow for 30 days. In the other groups estrogen pellets were removed, and the mice were divided into experimental groups that were treated orally with vehicle or FC1271 at different doses for 48 days. At the end of the experiment the tumors were removed and frozen in liquid nitrogen for storage at -70 C or were fixed in neutral formaldehyde (37%).

Northern blot analysis of RNA
Total RNA was purified from MCF-7 cells and MCF-7 nude mouse tumors in guanidine-isothiocyanate as described by Ruohola et al. (23), size-fractionated on 1% agarose-formaldehyde gels, blotted on nylon membranes, and hybridized for quantitation of pS2 signal intensities (22).

Ligand binding assay for ERs
The ability of FC1271a to compete with 17ß-[³H]E₂ for binding to ER and -ß was evaluated and compared with that of 17ß-E₂. Purified, recombinant human ER and -ß proteins [PanVera Corp., Madison, WI; 2.4 pmol of each in 20 mM HEPES (pH 7.4), 1.5 mM EDTA, 0.5 mM dithiothreitol, and 10% (wt/vol) glycerol] were incubated in the presence of serial dilutions of unlabeled 17ß-E₂ (Sigma) or FC1271a with 4 nM [2,4,6,7-³H]E₂ (72 Ci/mmol; Amersham International) for 20–22 h at 22 C. Bound and free radioligands were separated on Sephadex G-25 PD-10 columns by washing with 0.05 M Tris-HCl (pH 7.4–7.5). The amount of [³H]E₂-bound receptor protein was determined by liquid scintillation counting (OptiPhase, HiSafe 3 and 1409, Wallac, Inc.). The relative counts per min were plotted against the concentration of the ligand. The numeral values of binding constants were obtained by fitting the data to the Hill’s equation (24) and computed by the International Mathematical and Statistical Libraries routine RNLIN.F90. The representative plots as well as the statistical analysis were performed using the Microsoft Corp. (Redmond, WA) Excel 7.0 package.

Statistical analysis
Data were expressed as the mean ± SEM of each group. Statistical differences in different groups in the in vivo and in vitro experiments were evaluated by one-way ANOVA, followed by a two-tailed Student’s t test if significance was found. P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Interaction of FC1271a with ERs
The interaction of FC1271a with ERs was evaluated by studying the ability of the compound to compete with 17ß-[³H]E₂ for binding to purified recombinant human ER and ERß proteins and by comparing that to unlabeled estradiol. FC1271a displaced estradiol in a concentration-dependent manner, and IC₅₀ values of 827 and 1633 nM were obtained for ER and -ß, respectively (Fig. 2, A and B). The IC₅₀ values for 17ß-E₂ under the corresponding conditions were 6.8 and. 9.1 nM. The relative binding affinities of FC1271a calculated from these values were 0.8% and 0.6% for ER and ERß, respectively.

View larger version (18K): [in this window] [in a new window]   Figure 2. Competition by 17ß-E2 (A) and FC1271a (B) for [3H]E2 binding to recombinant human ER and ERß protein. The molar concentrations of the compounds required to obtain 50% inhibition of [3H]E2 binding (IC50).

Body weight. OVX increased body weight gain markedly compared with that of sham-operated rats, in accordance with the report by Yamazaki and Yamaguchi (27). 17-EE₂, 17ß-E₂, FC1271a, DRO, and RAL all normalized the OVX-induced increase in body weight gain (Fig. 3). 17ß-E₂ was the most potent, with an ED₅₀ of 2.4 µg/kg·day, and 17-EE₂ was 3- and 10-fold more potent than FC1271a and DRO, respectively. FC1271a at 1 mg/kg was similarly efficacious as RAL at 3 mg/kg.

View larger version (17K): [in this window] [in a new window]   Figure 3. Effects of FC1271a, 17-EE2, and 17ß-E2 at various doses on body weight gain in OVX rats. The effects of 3 mg/kg RAL and 10 mg/kg DRO are shown as a comparison. Results are from Exp II and III. Rats (n = 9–11 rats/group) were OVX and treated daily orally with vehicle, 17-EE2, FC1271a, RAL, and DRO or sc with 17ß-E2 for 4 weeks. Body weight gain is the difference between body weight before and after treatment. Results are the mean ± SD compared with the OVX level (=100%) in each experiment. The OVX levels in Exp I–III were 45.1 ± 9.0, 57.9 ± 4.9, and 48.8 ± 8.9 g, respectively. Sham (16.6 ± 4.5 g) represents that in Exp III.

View larger version (64K): [in this window] [in a new window]   Figure 4. Goldner-Massen stained undecalcified sections from distal femur of sham (A), OVX (B), and 1 mg/kg (C) and 10 mg/kg (D) FC1271a-treated animals. OVX caused a marked loss of secondary spongiosa, and FC1271a effectively prevented this loss. The rectangle in B shows the area where TBV was measured from sections of each experimental group.

Two biochemical indicators of collagen degradation, urinary deoxypyridinoline (U-DPD) and serum ICTP (S-ICTP), were measured to assess the extent of bone resorption. Both U-DPD and S-ICTP increased significantly after OVX, indicating that excess bone resorption occurred (Fig. 5). FC1271a at 1–10 mg/kg prevented the increase in S-ICTP by 82–67% compared with OVX (P < 0.05), equivalent to the effects of RAL and DRO. 17ß-E₂ at 1–10 µg/kg and 17-EE₂ at 1–10 mg/kg also decreased S-ICTP levels to the same extent, but the efficacy was higher (1000- and 100-fold, respectively) than that of FC1271a, RAL, or DRO (Fig. 5A). U-DPD was also decreased in a dose-dependent manner by FC1271a. RAL at 3 mg/kg was somewhat more potent in reducing U-DPD levels than FC1271a or DRO (Fig. 5B).

View larger version (21K): [in this window] [in a new window]   Figure 5. Serum ICTP (A) and urinary DPD (B) after a 4-week treatment of OVX rats. All SERMs used reduced both values, although not as effectively as 17ß-E2. Results are the mean ± SD (n = 9–11 rats/group) from Exp II and III compared with the OVX level (=100%) in each experiment. The levels of ICTP and DPD in OVX rats were 19.9 ± 2.5 µg/liter and 90.4 ± 14.4 nM in Exp II, and 18.6 ± 2.2 µg/liter and 568.3 ± 182.5 nM in Exp III, respectively. The levels of ICTP and DPD in sham animals were 14.1 ± 2.3 µg/liter and 57.4 ± 17.2 nM, respectively, representing those in Exp III.

View larger version (17K): [in this window] [in a new window]   Figure 6. Effects of various doses of 17-EE2, 17ß-E2, and FC1271a on serum cholesterol levels in OVX rats. The effects of 3 mg/kg RAL and 10 mg/kg DRO were studied as a comparison. Rats were sham operated or OVX, and the OVX rats were treated with vehicle or the compounds studied. Rats were fasted overnight before death. Results are the mean ± SD (n = 9–11 rats/group) from Exp II and III compared with the OVX level (=100%) in each experiment (2.99 ± 0.15 mM in Exp III and 2.69 ± 0.13 mM in Exp II). Sham (3.43 ± 0.11 mM) represents that in Exp III.

View larger version (53K): [in this window] [in a new window]   Figure 7. Serum LH, FSH, and LH/FSH ratio in sham-operated rats, OVX rats, and OVX rats treated with different doses of FC1271a or with 3 mg/kg RAL and 10 mg/kg DRO. Results are the mean ± SD (n = 10–11 rats/group) relative to the OVX value (100%) from Exp II. OVX values for LH, FSH, and LH/FSH were 41.3 ± 14.6, 119.3 ± 12.6, and 0.31 ± 0.12 ng/ml, respectively. *, P < 0.05; **, P < 0.01 (vs. OVX).

View larger version (75K): [in this window] [in a new window]   Figure 8. Effect of 17-EE2, 17ß-E2, and FC1271a on uterus weight. Rats were sham operated or OVX, and the OVX rats were treated with various doses of 17-EE2, 17ß-E2, and FC1271a or vehicle (polyethylene glycol) for 4 weeks. Relative uterine wet weight (uterine weight/body weight) was determined (A). Results are the mean ± SD (n = 9–11 rats/group) from Exp I–III relative to the results in OVX rats (=100%) in each experiment. The weights of OVX rats in Exp I–III was 0.09 ± 0.01, 0.10 ± 0.02, and 0.08 ± 0.01 g, respectively. Sham (0.32 ± 0.10 g) represents that in Exp III. Histological sections (B) were prepared from the same uteri as in A. A, Sham operated; B, OVX; C, OVX, treated with 50 µg/kg 17ß-E2/day, sc, for 4 weeks; D, OVX, treated daily with 1 mg/kg FC1271a for 4 weeks; E, OVX, treated daily with 10 mg/kg FC1271a for 4 weeks. A representative section per group is shown. Hematoxylin-eosin stain; magnification, x250.

Immature uterus tests
The uterine effects were also studied by treating prepubertal 18-day-old female rats orally for 3 days with increasing doses of FC1271a and by comparing the effects with those of 17-EE₂ as well as the SERMs, TAM, TOR, RAL, and DRO. Figure 9A shows that FC1271a had a slight intrinsic agonistic activity similar to that of RAL and DRO up to 1 mg/kg. At higher doses, it was somewhat more estrogen agonistic, but at 10 mg/kg the effect of FC1271a was still clearly less than that of 10 µg/kg 17-EE₂ or 50 µg/kg 17ß-E₂. The stimulation of FC1271a at 50 and 100 mg/kg was at the level obtained with 1 mg/kg TAM.

View larger version (32K): [in this window] [in a new window]   Figure 9. Effects of various SERM molecules, 17-EE2, and 17ß-E2 on uterine weight in immature rats. Eighteen-day-old rats were treated for 3 days orally with the compounds or vehicle (polyethylene glycol) or sc in the case of 17ß-E2 or vehicle (sesame oil). Uterine weight was related to the body weight of each animal. Results are the mean ± SD (n = 5). The experiments with different compounds were repeated one to three times with corresponding results.

Effects of FC1271a in intact adult rats
The effects of FC1271a in intact adult rats were studied by treating them with the compound at the doses of 1 and 10 mg/kg for 4 weeks (Exp I). Uterine weights declined by 15% and 26%, respectively. Serum ICTP was slightly increased in the group treated with the higher dose of FC1271a, being 12.7 ± 1.6 µg/liter in the 10 mg/kg treatment group compared with 10.4 ± 1.8 µg/liter in the control group. Neither the ash weight of the tibia nor the maximum load strength of the femoral neck changed after treatment (Table 1). FC1271a at 10 mg/kg caused a slight increase in the maximum load of L3. In contrast, TBV was decreased, and the number of osteoclasts and MAR were significantly increased with 10 mg/kg FC1271a (Table 1), indicating that FC1271a at a higher dose had a slightly antiestrogenic effect in the ovary-intact animals.

Effect of FC1271a on breast cancer cells in vitro and in vivo
The estrogen-dependent MCF-7 human breast cancer cells (21, 22) were used as a model for studies on the effects of FC1271a on breast cancer cells. The addition of the compound at concentrations of 0.1 nM to 10 µM did not cause a significant increase in MCF-7 cell growth in vitro when studied by measuring ATP (Fig. 10A) or 3-[4,5-dimethylthiazol-2-yl]2,5-diphenyltetrazolium bromide levels, cell numbers, and rate of [³H]thymidine incorporation (data not shown) during a 7-day culture period. On the other hand, the compound did not inhibit the growth stimulation caused by 1 nM estradiol, except at a concentration 10 mM by only 30%. Similar results were obtained with ZR 75–1 cells (data not shown), another estrogen-dependent human breast cancer cell line (21). The cytotoxicity of FC1271a at high concentrations was therefore markedly lower than that for TAM, TOR (21, 28), or RAL (29).

View larger version (37K): [in this window] [in a new window]   Figure 10. Effect of FC1271a on the growth and pS2 gene expression of MCF-7 human breast cancer cells in vitro and in vivo. A, MCF7 cells were grown for 4 days in vitro in the presence of 0.1 nM to 10 µM FC1271a with or without 1 nM 17ß-E2. The MCF-7 cell growth rate in cultures was followed by measuring the relative mass of viable cells using an ATP method. The results are the mean ± SD of 6 parallel wells of a 96-well plate. The results are presented relative to those for the cells grown with or without 17ß-E2 and without FC1271a (100%). The ATP level in the cells grown with 17ß-E2 was 1.65-fold higher than that in cells grown without it. B, MCF-7 cells grown in OVX nude mice for 8 weeks in the presence of 17ß-E2 pellets. After the tumors were grown, the pellets were removed (except from 17ß-E2 controls, day 0, arrow), and the mice were treated from then on with 1, 10, or 50 mg/kg FC1271a or vehicle (polyethylene glycol) for the time periods indicated. Tumor sizes were measured and compared with the size of each tumor on day 1 (100%). The growth of the tumors in mice that retained the 17ß-E2 pellet as controls is shown as a comparison (inset) to the tumors withdrawn from 17ß-E2 pellet with or without FC1271a treatment. The results are the mean ± SD. The numbers of tumors were nine for vehicle-treated control group, eight for each group treated with FC1271a, and five for the group retaining the E2 pellet. The experiment was repeated once with similar results. C, Northern blot analysis of the pS2 gene in MCF7 cells grown in vitro for 6 days with or without (=control) 1 nM 17ß-E2 or in the presence of 1 nM or 1 µM FC1271a. D, Northern blot analysis of the pS2 gene in MCF7 cell tumors grown in nude mice treated with or without (=control) 1 nM 17ß-E2 pellet or with 1, 10, or 50 mg/kg FC1271.

The expression of the pS2 gene was studied as a model for an estrogen-dependent gene (30) expressed by both MCF-7 and ZR 75–1 cells. The expression of pS2 in MCF-7 cells grown in the presence of FC1271a at various concentrations (1 nM to 1 µM) was not observed except at a faint level in the cells grown with 10 or 100 nM FC1271a (Fig. 10C). Similar results were obtained with ZR 75–1 cells (data not shown). In nude mouse tumors a low level of pS2 messenger RNA was observed in the tumors from the mice treated with 50 mg/kg FC1271a, but very little was seen in the mice treated with 1 or 10 mg/kg FC1271a (Fig. 10D).

Effect of FC1271a on appearance and growth of DMBA tumors
The number of breast tumors in the DMBA-induced rats treated with FC1271a at the dose of 10 or 50 mg/kg for 4 weeks was significantly smaller than that in control rats (31% and 5%, respectively; Fig. 11, A and B). At follow-up at 6 weeks after the cessation of treatment, the number of the tumors in the rats treated with 50 mg/kg FC1271a was still only 35% of that in the nontreated control group. The effect of FC1271a treatment on the growth status of the tumors in a representative experiment is presented in Table 2. It shows that in the nontreated control group, 100% of tumors were continuously growing, whereas only 88%, 41%, and 21% of the tumors grew in the groups treated with 1, 10, and 50 mg/kg FC1271a, respectively. In addition, the number of tumors in a stationary or regressing state increased in a dose-dependent manner. Some tumors (4%) even disappeared in the group treated with the highest dose of 50 mg/kg.

View larger version (40K): [in this window] [in a new window]   Figure 11. Effect of FC1271a on the number of DMBA-induced tumors per rat. The mice were treated (orally) for 4 weeks with vehicle (polyethylene glycol), 1 or 10 mg FC1271a (Exp I), or 0.1 or 50 mg FC1271a (Exp II; open bar) and then followed for 6 weeks without any treatment (hatched bar). The results are the mean ± SD number of tumors per rat. The number of rats was 8–12/group. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (vs. control).

View larger version (21K): [in this window] [in a new window]   Figure 12. HPLC analysis of the 32P-postlabeled DNA from the livers of rats treated for 2 weeks with 45 mg/kg tamoxifen (A) or 45 mg/kg FC1271a (B).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

FC1271a prevented bone loss in the ovariectomized rat, which is an applicable model for early skeletal changes in postmenopausal osteoporosis. As in postmenopausal women, estrogen deficiency in ovariectomized rats produces high turnover osteoporosis, in which trabecular bone is preferentially lost due to markedly enhanced resorption. Although bone formation is also increased, the net loss of bone mass is a hallmark of estrogen deficiency-induced osteoporosis (32). In our experiments, OVX resulted in a significant osteopenic response in 4 weeks at multiple skeletal sites, as determined by bone histomorphometry, biomechanical test, and ash weight measurement. As expected, 17-EE₂ and 17ß-E₂ prevented OVX-induced bone loss, which has also been demonstrated in several previous reports (32, 33).

FC1271a prevented bone loss by suppressing the increase in bone turnover that occurs with estrogen withdrawal. The OVX-induced increase in MAR, for instance, was completely inhibited by 0.3 mg/kg FC1271a. It was also evident that administration of FC1271a reduced the number of osteoclasts and led to a decrease in other resorption parameters. The preservation of TBV of the distal femur and the ash weight of the tibial epiphyses confirmed the net bone-sparing effect of FC1271a. Moreover, serum ICTP and urinary DPD, which are biochemical markers of bone turnover, were decreased by FC1271a in a dose-dependent manner. In conclusion, the results from bone histomorphometry confirmed that FC1271a acts as an estrogen agonist in bone. The doses of 3 and 10 mg/kg were equally effective in preventing osteoporosis and comparable to those achieved by RAL at 3 mg/kg and DRO at 10 mg/kg, respectively.

The histomorphometric observations were associated with parallel results in the mechanical strength measurements. The OVX-induced decrease in the mechanical strength of the femoral neck, the lumbar vertebra, and the tibia could be prevented by FC1271a treatment. It has also been shown elsewhere that there is a clear positive correlation between the maximal load of the femoral neck and trabecular bone volume in the distal femur as well as femoral ash weight in the rat model (18).

FC1271a lowered serum cholesterol level in a dose-dependent manner in OVX rats. Previous work in OVX rats and intact rabbits showed that oral administration of 17-EE₂ produced a marked reduction in total serum cholesterol (34, 35). The mechanism involves the up-regulation of hepatic LDL receptors by estrogen, resulting in an enhanced clearance of circulating LDL (35, 36). Therefore, the rat model predicts the reduction of LDL cholesterol for agents producing estrogen agonist effects in liver. It is thus a sensitive model for monitoring pharmacological effects of estrogen on total cholesterol in vivo. Clinical observations have also shown that both estrogen and the SERM molecules, such as TAM, TOR, and RAL, were able to lower LDL cholesterol in postmenopausal women (37, 38, 39, 40).

In tests in immature rat uterus, FC1271a was 200- to 1000-fold less potent as an agonist than estrogen. TAM and TOR were clearly more potent estrogen agonists than FC1271a. The effects of RAL and DRO were comparable to those of FC1271a up to 1 mg/kg, but at higher doses these two were less agonistic. In OVX rats, FC1271a induced a 1.5-fold increase in uterine weight at the maximum. This was approximately 30% of that in 17-EE₂- or 17ß-E₂-treated rats and comparable to that in DRO (at 10 mg/kg) and slightly higher than that in RAL (3 mg/kg)-treated rats. At a higher dose (10 mg/kg), the weight increase was associated with an increase in the height of luminal epithelium, which was not observed at a dose of 1 mg/kg FC1271a. This suggests that at lower doses the uterine weight increase in OVX rats was preferentially caused by stromal components.

Immature rat tests as well as the experiments with intact rats also demonstrated that FC1271a has a weak antagonistic activity in uterus, as demonstrated by inhibition of estrogen-induced weight increase (50% inhibition at maximum). It was also notable that in OVX rats FC1271a given together with 17ß-E₂ was in most cases able to prevent the formation of stratified epithelial morphology that is typically and constantly induced by prolonged treatment with estrogen (data on file). These effects together suggest that treatment with FC1271a at the doses effective in bone is associated with a reduced risk of development of endometrial cancer compared with a chronic administration of estrogen. On the other hand, the lack of strongly antagonistic and atrophic effects of FC1271a on the reproductive tract would actually suggest a preservation of a rather normal endometrial, and possibly vaginal, epithelium and mucosa.

In the DMBA rat mammary carcinoma model, FC1271a showed a clear antitumor effect that seemed to be caused primarily by a decrease in the appearance of new tumors but also by a retardation of tumor progression. The DMBA tumor model is widely used in studying antitumor activity of the hormonally acting compounds. Approximately 90% of the DMBA tumors respond to estrogen withdrawal upon ovariectomy (41). It is not clear, however, by which mechanism FC1271a exerts its antitumor effect. It may not be caused by a direct antiestrogen effect, because, in contrast to TAM and TOR (21, 22), FC1271a had very little effect on estrogen-dependent proliferation of two human breast cancer cell lines MCF-7 and ZR 75–1 in vivo or in vitro. One possibility is that the antitumor effect of FC1271a was mediated indirectly by hypothalamic and/or hypophyseal effects. FC1271a acted as a weak agonist in decreasing the OVX-induced elevation of LH. It is also possible that FC1271a decreases the level of circulating PRL, which is known to stimulate DMBA-induced tumor growth (42). The serum PRL level has been found to decreased in TAM- or TOR-treated patients (43, 44). Unfortunately, the effects of FC1271a on PRL synthesis and/or release are not presently known.

The mechanisms of the multiple effects of FC1271a in different targets are not yet known. It is probable, however, that many of the effects are ER mediated, as suggested, for example, by the ability of the pure antiestrogen ICI 182,780 to inhibit FC1271a stimulation of osteoblastic differentiation in bone marrow cultures (15). The competition binding assay showed that FC1271a binds specifically to both ER and ERß. The binding affinity was low, but it was roughly in the range of that for TAM and TOR. The binding affinity for ER seemed to be higher than that for ERß, although the relative binding affinities of FC1271a for ER and ERß (0.8% and 0.6%, respectively) were rather close to each other. The binding affinity does not, however, necessarily correlate with the biological activity, as shown in vitro with MCF-7 cell proliferation assays (45) and with all the in vivo and in vitro effects of TAM and TOR. Accordingly, FC1271a has been shown to promote binding of the ER to the estrogen response element as well as to increase the ER- and ERß-mediated transcription of a reporter gene in trans-activation assays (data on file and Tasanen, M., K. Väänänen, and P. Härkönen, in preparation). It remains to be shown, however, by which molecular and cellular mechanisms various tissue-specific responses of FC1271a are induced (46, 47, 48, 49).

In conclusion, FC1271a showed beneficial effects, comparable to those of RAL and DRO, on bone metabolism in the OVX rat model. It acted as an estrogen agonist in bone and lipid metabolism, but had little estrogenic activity in uterus at doses that efficiently prevented bone loss. Furthermore, FC1271a is characterized by a weak antiestrogenic activity. This suggests that FC1271a is a truly selective estrogen that does not cause clinically unwanted antiestrogenic effects, such as menopausal symptoms. Taken together, these data strongly suggest that the characteristic profile of FC1271a makes it a novel alternative for the prevention of postmenopausal osteoporosis.

	Acknowledgments

 
We thank Timo Leino and Pirkko Rauhamäki for their excellent technical assistance, and Eeva Valve for help in biochemical assays.

	Footnotes

 
¹ This work was supported by grants from the Technology Development Center and the Academy of Finland (to P.H. and K.V.).

² Both authors should be considered as first authors of this manuscript.

Received September 15, 1999.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Kanis JA 1996 Estrogens, the menopause, and osteoporosis. Bone 19:S185–S190
2. Grodstein F, Stampfer MJ 1998 Estrogen for women at varying risk of coronary disease. Maturitas 30:19–26[CrossRef][Medline]
3. Halbreich U 1997 Role of estrogen in postmenopausal depression. Neurology [Suppl 7] 48:S16–S19
4. Marsburn PB, Carr BR 1992 Hormone replacement therapy: protection against the consequences of menopause. Postgrad Med J 92:145–159
5. Lobo RA 1995 Benefits and risks of estrogen replacement therapy. Am J Obstet Gynecol 173:982–990[CrossRef][Medline]
6. Colditz GA, Hankinson SE, Hunter DJ, Willett WC, Manson JE 1995 The use of estrogen and progestins and the risk of breast cancer in postmenopausal women. N Engl J Med 332:1589–1593[Abstract/Free Full Text]
7. Tonetti DA, Jordan C 1996 Targeted anti-estrogens to treat and prevent diseases in women. Mol Med Today 5:218–223
8. Love RR, Wiebe DA, Newcomb PA, Cameron L, Leventhal H 1991 Effects of tamoxifen on cardiovascular risk factors in post-menopausal women. Ann Intern Med 115:860–864
9. Black LJ, Sato M, Rowley ER, Magee DE, Bekele A 1994 Raloxifene (LY139481 HCI) prevents bone loss and reduces serum cholesterol without causing uterine hypertrophy in ovariectomized rats. J Clin Invest 93:63–69
10. Greaves P, Goonetilleke R, Nunn G, Topham J, Orton T 1993 Two-year carcinogenicity study of tamoxifen in alderley park Wistar-derived rats. Cancer Res 53:3919–3924[Abstract/Free Full Text]
11. Hirsimäki P, Hirsimäki Y, Nieminen L, Payne BJ 1993 Tamoxifen induces hepatocellular carcinoma in rat liver: a 1-year study with two antiestrogens. Arch Toxicol 67:49–54[CrossRef][Medline]
12. Seachrist L 1994 Restating the risks of tamoxifen. Science 263:910–911[Free Full Text]
13. Draper MW, Flowers DE, Huster WJ, Neild JA, Harper KD, Arnaud C 1996 A controlled trial of raloxifene (LY 139481) HCl: impact of bone turnover and serum lipid profile in healthy postmenopausal women. J Bone Miner Res 11:835–842[Medline]
14. Härkönen PL, Qu Q, Dahllund JC, Laine A, Hemminki K, Kangas L, Väänänen HK FC1271a is a novel tissue-specific antiestrogen with bone-sparing effect. 10th International Congress of Endocrinology, San Francisco, CA, 1996, vol 1:P1.256
15. Qu Q, Härkönen PL, Väänänen HK 1999 Comparative effects of estrogen and antiestrogens on differentiation of osteoblasts in mouse bone marrow culture. J Cell Biochem 73:500–507[CrossRef][Medline]
16. Parikka V, Lehenkari PP, Härkönen PL, Väänänen HK 1998 The effect of two selective estrogens raloxifene and FC1271a on osteoclast survival and resorption in vitro. J Bone Miner Res [Suppl 5] 23:S552
17. Kostner GM 1976 Enzymatic determination of cholesterol in high-density lipoprotein fractions prepared by polyanion precipitation. Clin Chem 22:695
18. Peng ZQ, Tuukkanen J, Zhang HX, Jämsä T, Väänänen HK 1994 The mechanical strength of bone in different rat models of experimental osteoporosis. Bone 15:523–532[Medline]
19. Hemminki K, Widlak P, Hou SM 1995 DNA adducts caused by tamoxifen and toremifene in human microsomal system and lymphocytes in vitro. Carcinogenesis 16:1661–1664[Abstract/Free Full Text]
20. Kangas L, Nieminen A-L, Blanco G, Grönroos M, Kallio S 1986 A new triphenylethylene compound, Fc-1157a. II. Antitumor effects. Cancer Chemother Pharmacol 17:109–113[CrossRef][Medline]
21. Wärri AM, Laine AM, Majasuo KE, Alitalo KK 1991 Estrogen suppression of erbB2 expression is associated with increased growth rate of ZR 75–1 human breast cancer cells in vitro and in nude mice. Int J Cancer 49:616–623[Medline]
22. Wärri AM, Huovinen RL, Laine AM, Martikainen PM Härkönen PL 1993 Apoptosis in toremifene-induced growth inhibition of human breast cancer cells in vitro and in vivo. J Natl Cancer Inst 85:1412–1418[Abstract/Free Full Text]
23. Ruohola JK, Valve EM, Vainikka S, Alitalo KK, Härkönen PL 1995 Androgen and FGF regulation of FGF receptors in S115 mouse mammary tumor cells. Endocrinology 136:2176–2188
24. Salomonsson M, Carlsson B, Häggblad J 1994 Equilibrium hormone binding to human estrogen receptors in highly diluted cell extracts is non-cooperative and has a Kd of approximately 10 pM. J Steroid Biochem Mol Biol 50:313–318[CrossRef][Medline]
25. Turner CH, Sato M, Bryant HU 1994 Raloxifene preserves bone strength and bone mass in ovariectomized rats. Endocrinology 135:2001–2005[Abstract]
26. Ke HZ, Simmons HA, Pirie CM, Crowford DT, Thompson DD 1995 Droloxifene, a new estrogen antagonist/agonist, prevents bone loss in ovariectomized rats. Endocrinology 136:2435–2440[Abstract]
27. Yamazaki I, Yamaguchi H 1989 Characteristics of an ovariectomized osteopenic rat model. J Bone Miner Res 4:13–22[Medline]
28. Kangas L 1990 Biochemical and pharmacological effects of toremifene metabolites. Cancer Chemother Pharmacol 27:8–12[CrossRef][Medline]
29. Black LJ, Jones CD, Falcome JF 1983 Antagonism of estrogen action with a new benzothiophene derived antiestrogen. Life Sci 32:1031–1036[CrossRef][Medline]
30. Kida N, Yoshimura T, Mori K, Hayashi K 1989 Hormonal regulation of synthesis and secretion of pS2 protein relevant to growth of human breast cancer cells (MCF-7). Cancer Res 49:3494–3498[Abstract/Free Full Text]
31. Pathak DN, Bodell WJ 1994 DNA adduct formation by tamoxifen with rat and human liver microsomal activation system. Carcinogenesis 15:529–532[Abstract/Free Full Text]
32. Kalu DN, Salerno E, Liu CC, Echon R, Ray M, Garza-Zapata M 1991 A comparative study of the actions of tamoxifen, estrogen, and progesterone in the ovariectomized rat. Bone Miner 15:109–123[CrossRef][Medline]
33. Wronski TJ, Cintron M, Doherty AL, Dann LM 1988 Estrogen treatment prevents osteopenia and depresses bone turnover in ovariectomized rats. Endocrinology 123:681–686[Abstract/Free Full Text]
34. Chao Y, Windler EE, Chen GC, Havel RJ 1979 Hepatic catabolism of rat and human lipoproteins in rats treated with 17--ethinyl estradiol. J Biol Chem 254:11360–11366[Free Full Text]
35. Ma PTS, Yamamoto T, Goldstein L, Brown MS 1986 Increased mRNA for low density lipoprotein receptor in livers of rabbis treated with 17--ehinyl estradiol. Proc Natl Acad Sci USA 83:792–796[Abstract/Free Full Text]
36. Brown MS, Goldstein JL 1980 The estradiol-stimulated lipoprotein receptor of rat liver. J Biol Chem 255:10464–10471[Abstract/Free Full Text]
37. Bush TL, Barrett-Connon E, Cowan LD, Criqui MH, Wallace RB 1987 Cardiovascular mortality and noncontraceptive use of oestrogen in women: results from the Lipid Research Clinics Program follow-up study. Circulation 75:1102–1109[Abstract/Free Full Text]
38. Love RR, Mazess RB, Barden HS, Epstein S, Newcomb PA 1992 Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. N Engl J Med 326:852–856[Abstract]
39. Saarto T, Blomqvist C, Ehnholm C, Taskinen MR, Elomaa I 1996 Antiatherogenic effects of adjuvant antiestrogens: a randomized trial comparing the effects of tamoxifen and toremifene on plasma lipid levels in postmenopausal women with node-positive breast cancer. J Clin Oncol 14:429–433[Abstract/Free Full Text]
40. Lufkin EG, Whitaker MD, Nickelsen T, Argueta R, Caplan RH 1998 Treatment of established postmenopausal osteoporosis with raloxifene: a randomized trial. J Bone Miner Res 13:1747–1754[CrossRef][Medline]
41. Huovinen R, Collan Y 1994 Cell loss in dimethylbenz(a)anthracene-induced rat mammary carcinoma treated with toremifene and ovariectomy. Tumour Biol 15:345–353[Medline]
42. Jordan VC, Murphy CS 1990 Endocrine pharmacology of antiestrogens as antitumor agents. Endocr Rev 11:578–610[Abstract/Free Full Text]
43. Groom CV, GriffithsK 1976 Effect of the antiestrogen tamoxifen on plasma levels of luteinizing hormone, follicle stimulating hormone prolactin, oestradiol and progesterone in hormal premenopausal women. J Endocrinol 70:421[Abstract/Free Full Text]
44. Szamel I, Hindy I, Vincze B, Eckhardt S, Kangas L Hajba A 1994 Influence of toremifene on the endocrine regulation in breast cancer patients. Eur J Cancer 30A:154–158
45. Ruenitz PC, Bourne CS, Sullivan KJ Moore SA 1996 Estrogenic triarylethylene acetic acids: effect of structural variation on estrogen receptor affinity and estrogenic potency and efficacy in MCF-7 cells. J Med Chem 39:4853–4859[CrossRef][Medline]
46. Webb P, Lopez GN, Uht LR Kushner PJ 1995 Tamoxifen activation of the estrogen receptor/AP-1 pathway: potential origin for the cell-specific estrogen-like effects of antiestrogens. Mol Endocrinol 9:443–456[Abstract/Free Full Text]
47. Yang NN, Venugopalan M, Hardikar S, Glasebrook A 1996 Identification of an estrogen response element activated by metabolites of 17ß-estradiol and raloxifene. Science 273:1222–1225[Abstract]
48. Willson TM, Norris JD, Wagner BL, Asplin I, Baer P 1997 Dissection of the molecular mechanism of action of GW5638, a novel estrogen receptor ligand, provides insights into the role of estrogen receptor in bone. Endocrinology 138:3901–3911[Abstract/Free Full Text]
49. Barkhem T, Carlsson B, Nilsson Y, Enmark E, Gustafsson J, Nilsson S 1998 Differential response of estrogen receptor and estrogen receptor ß to partial estrogen agonists/antagonists. Mol Pharmacol 54:105–112[Abstract/Free Full Text]

This article has been cited by other articles:

				J. M. Fitts, R. M. Klein, and C. A. Powers Estrogen and Tamoxifen Interplay with T3 in Male Rats: Pharmacologically Distinct Classes of Estrogen Responses Affecting Growth, Bone, and Lipid Metabolism, and Their Relation to Serum GH and IGF-I Endocrinology, October 1, 2001; 142(10): 4223 - 4235. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Qu, Q. Articles by Härkönen, P. Search for Related Content PubMed PubMed Citation Articles by Qu, Q. Articles by Härkönen, P.