Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Hormone replacement therapy with estrogens can effectively treat symptoms, such as osteoporosis and climacteric complaints, that are related to the marked decline in plasma estradiol levels after menopause (Horsman et al., 1983). Ideally, the estrogenic effect of such a therapy should be evident in HRT¹ target tissues such as bone, the cardiovascular system, vagina, and the brain. An estrogenic, proliferative response, however, should be absent in the breast and in the uterus. In particular in the uterus, unopposed estrogen treatment leads to an increased risk of developing uterine cancers. Combined estrogen and progestogen treatment can prevent this risk increase. Therefore, HRT often consists of a combined estrogen and progestogen treatment, given either sequentially or continuously.

Norethisterone is a synthetic steroid that is used as a progestogen in contraception and in HRT in postmenopausal women. However, both progestogenic and antiprogestogenic effects of norethisterone have been described on uteroglobin and progesterone receptor expression in the uterus of progesterone-treated prepubertal rabbits (Pasapera et al., 1995). Moreover, estrogenic activity of norethisterone in rat uterus is mediated through 5-reduced metabolites of norethisterone (Mendoza-Rodriguez et al., 1999). Conversion of norethisterone in e.g., uterus into more estrogenic metabolites may affect the estrogen/progestogen balance that is the key feature of a combined HRT.

The aim of the present study is to investigate whether metabolism of norethisterone can take place in organs that are targets of HRT. In this study, the steroid metabolizing capacity of the uterus, vagina, and aorta was studied. To deduce which steroid-metabolizing enzymes are present, tissue fragments of uterus, vagina, and aorta from rats were incubated with ³H-labeled progesterone. Subsequently, incubations with ³H-labeled norethisterone were performed and its metabolites were analyzed. Furthermore, the effect on this metabolism of substitutions at the 7- and 11-position on the steroid skeleton was studied. The 7- and 11-position are known to maintain good estrogen receptor binding when they are substituted with small lipophilic groups such as a methyl or ethyl group (Zeelen and Bergink, 1980). The norethisterone derivatives studied are Org OM38, Org 4060, and Org 34694. All three steroids are similar to norethisterone, but have an additional 7-methyl- (Org OM38), an 11-ethyl- (Org 4060), or both a 7-methyl- and an 11-ethylidene substituent (Org 34694).

The radiolabeled metabolites were separated and isolated using HPLC and HPTLC. Identification of isolated metabolites was performed by cochromatography with unlabeled steroids in the different chromatographic systems.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. [1,2,6,7-³H]Progesterone (3589 GBq/mmol) was purchased from PerkinElmer Life Science Products (Hoofddorp, The Netherlands). The radiolabeled steroids [16-³H]norethisterone (1010 GBq/mmol), [16-³H]Org OM38 (1400 GBq/mmol), [11-³H]Org 4060 (560 GBq/mmol), and [16-³H]Org 34694 (925 GBq/mmol) were a gift from Organon (Oss, The Netherlands). The radiolabeled, 5-reduced metabolite of Org 4060 and the 3-hydroxylated metabolite of Org 4060 were synthesized from [11-³H]Org 4060 by Organon and yielded [11-³H]5-H-Org 4060 and [11-³H]3-OH-Org 4060 with specific activities of ca. 500 GBq/mmol. The radiolabeled, 5,3-reduced metabolite of Org 4060 and the 5,3-reduced metabolite were produced from [11-³H]5-Org 4060. Unlabeled norethisterone, Org 4060, Org OM38, and Org 34694 metabolites that were used as reference steroids were a gift from Organon. All other chemicals were obtained from local commercial sources and were of analytical grade.

Animals. Mature female Wistar strain Hsd/Cpd:Wu rats (Harlan, The Netherlands), with a weight between 225 and 250 g, were fed with standard pelleted diet (RMH-B; Hope Farms BV, Woerden, The Netherlands), and tap water ad libitum. The animals were subjected to a 14-h light/10-h dark daily cycle. The rats were randomly distributed in two groups. One group (n = 4) was ovariectomized 2 weeks before sacrifice. During 2 weeks, the other group (n = 5) was monitored daily for estrous cycle progression by vaginal smears. The animals were sacrificed on the morning of proestrus at 10.00 AM, and the uterus, vagina, and aorta were removed for the tissue fragment incubations. All animal procedures were approved by the Animal Ethics Committee.

Tissue Incubation. The dissected organs were minced and 50 mg of tissue was transferred to vials containing 3 ml of Leibowitz medium (L-15) (buffered with HEPES at pH 7.4 and supplemented with 100 U/ml penicillin and 100 µg/ml streptomycin) and 100 pmol of the ³H-labeled steroid dissolved in 50 µl of propyleneglycol. The final concentration of ³H-labeled steroid in the incubation medium is 33.3 nmol/l. The fragments were incubated at 37°C for 24 h in a shaking water bath. For each steroid, a 24-h control incubation without tissue fragments was done. For the time course incubation of norethisterone, the fragments were incubated for 1, 3, 6, and 24 h. Cooling on ice stopped the incubation and the samples were stored at 20°C until analysis.

Extraction. After separation from the medium, the tissue fragments were extracted twice with 3 ml of ethanol. The ethanol extract was dried under a stream of nitrogen, redissolved in 300 µl of methanol and combined with the medium. To this mixture, 2.5 µg of appropriate unlabeled carrier steroid was added. The mixture was then applied to an activated Sep-Pak C₁₈ column and eluted with 5 ml of water, hexane, and methanol consecutively. The radioactivity in each eluate was determined by liquid scintillation counting (Tri-Carb 1900 TR; Canberra Packard, Groningen, The Netherlands), and the methanol eluates were used for HPLC and HPTLC analysis. The extraction efficiency was calculated as the ratio of the radioactivity in the methanol eluate to the total radioactivity in the medium before incubation. The recovery of radioactivity in the methanol eluates was 101 ± 8.4, 83.6 ± 4.8, 86.8 ± 8.5, 94.8 ± 6.3, and 85.1 ± 12.3% for progesterone, norethisterone, Org 4060, Org OM38, and Org 34694, respectively.

HPLC Analysis of Metabolite Profiles. Separation of the steroids and their metabolites was performed using a Waters spherisorb S5 ODS-2 column (4.6 × 250 mm) (Phase Separations B.V., Emmen, The Netherlands) with a guard column (4.6 × 10 mm) and a gradient of water (solvent A) and acetonitrile (solvent B). The analytes were eluted with 40% of solvent B for 5 min, followed by a linear gradient of 40 to 75% solvent B in 35 min. After 5 min of elution with 75% solvent B, initial values were reestablished in 5 min. The flow rate was 1.0 ml/min. Alternatively, a second elution system with methanol as solvent B was used. In this system (system II), the analytes were eluted with 50% of solvent B for 5 min, followed by a linear gradient of 50 to 90% solvent B in 35 min. After 5 min of elution with 90% solvent B, initial values were reestablished in 5 min.

HPTLC Analysis. A 200-µl aliquot of the methanol Sep-Pak eluate was dried under a stream of nitrogen and redissolved in 20 µl of propanol. Together with appropriate standard steroids, the sample was applied to the sample concentration zone of an HPTLC plate (10 × 10 cm, Merck kieselgel 60 F254). The plates were first run in toluene/cyclohexane (50:50) for 10 min to concentrate the sample at the border of the concentration zone. After drying, the plates were developed in dichloromethane/diethyl ether (80:20) for 15 min at 4°C. The reference standards were visualized under UV light. Steroids with a 4-3-keto group were visible with exposure to 254 nm UV light, while the other steroids were detected at 360 nm UV light, after derivatization with primulin (Wright, 1971). For detection of the radioactive compounds on the HPTLC silica plate, the plates were developed on a tritium sensitive PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA) or on X-ray film after spraying with a scintillation reagent (EN³HANCE; PerkinElmer Life Science Products).

Isolation of Metabolites. After generation of metabolite profiles in HPLC system I, samples of progesterone, Org OM38, Org 4060, and Org 34694 incubations with uterus, vagina, and aorta tissue from four animals were pooled, extracted, and injected on HPLC system I. Using a switching valve between the UV- and on-line radioactivity detector, metabolites were isolated. The isolates were dried and redissolved in HPLC mobile phase. An aliquot of each isolated fraction was reinjected to verify that a single radioactive peak had been isolated.

Identification of Metabolites.

Progesterone The isolated progesterone metabolites were applied to a TLC plate (10 × 20 cm, Merck kieselgel 60 F254, with concentration zone), and developed with reference steroids in diisopropylether/chloroform/hexane (7:2:1). The radioactive metabolites were identified by cochromatography with unlabeled reference standards.

Org 4060. Metabolite profiles for Org 4060 were established in HPLC system I. This system with acetonitrile as the mobile phase was chosen so that the Org 4060, 3-OH-Org 4060, and 3-OH-Org 4060 internal standards could be detected at 215 nm. The isolated Org 4060 metabolites and the 5-H-Org 4060 samples were then applied to HPLC system II. The radioactive metabolites were identified by cochromatography with radioactive reference standards of 5-H-Org 4060, 3-OH-Org 4060, 3-OH,5-H-Org 4060, and 3-OH,5-H-Org 4060, which were all resolved in HPLC system II. The samples and the isolated Org 4060 metabolites were also applied to a HPTLC plate (10 × 10 cm, Merck kieselgel 60 F254, with concentration zone). The radioactive metabolites were identified by cochromatography with unlabeled reference standards (5-H-Org 4060, 3-OH-Org 4060, 3-OH-Org 4060, 3-OH,5-H-Org 4060, and 3-OH,5-H-Org 4060).

Org OM38. Metabolite profiles for Org OM38 were established in HPLC system I. This system with acetonitrile as the mobile phase was chosen so that the Org OM38, 3-OH-Org OM38, and 3-OH-Org OM38 internal standards could be detected at 215 nm. The isolated Org OM38 metabolites were applied to an HPTLC plate (10 × 10 cm Merck kieselgel 60 F254, with concentration zone). The radioactive metabolites were identified by cochromatography with unlabeled reference standards (5-H-Org OM38, 3-OH-Org OM38, 3-OH-Org OM38, 3-OH,5-H-Org OM38, and 3-OH,5-H-Org OM38).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Progesterone. Progesterone was readily metabolized in uterus and vagina and less in aorta. Figure 1 shows a representative HPLC radiochromatogram of vaginal tissue incubation with progesterone. Uterus tissue incubations with progesterone yielded similar profiles, and the amounts of metabolites are summarized in Table 1. There were no qualitative differences between uterus and vagina incubations; all peaks appeared in incubations of both tissues. In aorta, however, only peaks P25, P31, P36, and P39 were formed.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Representative HPLC (system I) chromatogram showing the metabolism of [3H]progesterone in vagina tissue fragments. The arrows indicate the formation of two unknown metabolites (P9 and P11) and identified metabolites (P25, P31, P33, P36, and P39). The radioactive peaks labeled P9-P39 were isolated and identified by cochromatography on TLC with unlabeled reference standards.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Autoradiogram of a TLC plate showing the identification of progesterone metabolites. The arrows show the comigration with unlabeled reference progesterone metabolites of the radioactivity in metabolite peaks (P31, P33, and P36), which were isolated using HPLC (system I). In lane 1 and 2 P33 and P36 can be identified as 5-pregnan-20-ol-3-one and 5-pregnan-3-ol-20-one, respectively. In lane 3 it is shown that progesterone and 5-pregnan-3,20-diol were not resolved in P31.

Norethisterone. The major metabolite of norethisterone in all three tissues was 3-OH,5-H-norethisterone, as is illustrated in Fig. 3. In incubations of norethisterone with uterus tissue and vaginal tissue there was also formation of 5-H-norethisterone and two unknown metabolites (N4 and N19). A time course incubation of uterus tissue with norethisterone is shown in Fig. 4, demonstrating the time-dependent conversion of norethisterone into 3-OH,5-H-norethisterone and 5-H-norethisterone The metabolism of norethisterone was not appreciably different between proestrus rats and ovariectomized rats (Student's t test, p < 0.05). Therefore, the data of the two groups were combined. Figure 5 summarizes the formation of the norethisterone metabolites in the three tissues. The metabolism of norethisterone (proestrus and ovariectomized animals, n = 8) was lower in uterus than in vagina (Table 2). Only 1.6% of total norethisterone was metabolized in aorta tissue. Clearly, 5-reduction is the main metabolic pathway for norethisterone in all tissues. The 5-H-norethisterone that is formed is then mainly metabolized by 3-HSD, since the major metabolite in all three tissues is 3-OH,5-H-norethisterone. Formation of 3-OH,5-H-norethisterone occurred in uterus and vagina, but was more pronounced in vagina tissue, demonstrating the presence of 3-hydroxysteroid dehydrogenase (3-HSD) activity.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Analysis of [3H]norethisterone (NET) metabolites in uterus, vagina, and aorta. A, representative HPLC (system I) radiochromatogram of a 24-h vagina tissue incubation with [3H]norethisterone. B, representative HPTLC autoradiogram of a control incubation of [3H]norethisterone without tissue (lane 1) and with uterus (lane 2), vagina (lane 3), and aorta tissue.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Time course of norethisterone (NET) metabolism in uterus tissue fragment incubations. Uterus tissue was incubated with [3H]norethisterone and analyzed by HPLC (system I) after 1, 3, 6, 12, and 24 h of incubation. There is a steady increase of the major metabolite 3-OH,5-H-norethisterone up to 24 h.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Norethisterone metabolism in rat. Metabolism of [3H]norethisterone (NET) in uterus, vagina, and aorta. [3H]Norethisterone and its metabolites after a 24-h incubation of uterus, vagina, and aorta tissue were analyzed using HPLC. Columns indicate average ± S.E.M. Major metabolites were 3-OH,5-H-norethisterone (3,5), 3-OH,5-H-norethisterone (3,5), and 5-H-norethisterone (5).

Org 4060. Contrary to norethisterone, its 11-ethyl derivative Org 4060 was poorly metabolized in uterus, vagina, and aorta tissue. A representative HPLC (system I) chromatogram is shown in Fig. 6A. Two radioactive peaks (F8 and F26), that were not present in the Org 4060 control incubation, appeared in uterus and vagina incubations. As is shown in Table 2, these metabolites represented each less than 3% of total radioactivity in any of the tissues examined. In the aorta incubation, a radioactive peak with a retention time of 25 min (F25) and F26 was present. The F26 peak cochromatographed in HPLC system I with the 3-OH-Org 4060 reference peak in the UV trace (data not shown). After isolation, peak F26 also cochromatographed with the tritiated 3-OH-Org 4060 standard in HPLC system II (Fig. 7A). The formation of 5-reduced forms of Org 4060 in the three tissues was excluded using HPTLC (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Metabolism of norethisterone derivatives. Representative HPLC (system I) radiochromatogram of a 24-h uterus tissue incubation with [3H]Org 4060 (A),[3H]Org OM38 (B), and [3H]Org 34694 (C). The upper trace in each panel represents the 24-h control incubation without tissue of the appropriate steroid. The arrows indicate the metabolites that were formed.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Analysis [3H]Org 4060 metabolites using HPLC system II. The upper trace represents a radiochromatogram of tritiated Org 4060 metabolites. A, identification of the F26 Org 4060 metabolite from uterus as 3-OH-Org-4060. B, conversion in aorta tissue of [3H]5-H-Org 4060 into 3-OH,5-H-Org 4060 and 3-OH,5-H-Org 4060.

Org OM38. Also Org OM38, the 7-methyl derivative of norethisterone, was less metabolized than norethisterone itself. In uterus, vagina, and aorta a single radioactive peak appeared (O21) that was not present in the control incubation (Fig. 6B). The O21 metabolite from all tissues had the same retention time in HPLC system I as the unlabeled 3-OH-Org OM38 standard. After isolation, this metabolite also cochromatographed with the same standard on HPTLC (data not shown). Org OM38 was metabolized to 3-OH-Org OM38 (O21) at 4.0 ± 0.7, 7.6, and 4.5% in uterus, vagina, and aorta, respectively (Table 2).

Org 34694. Similar to Org 4060, also Org 34694 was poorly metabolized in uterus, vagina, and aorta (Table 2). Three radioactive peaks (Fig. 6C) appear in the tissue incubations (T8, T9, and T26). The third radioactive peak (T26) had the same retention time as an unlabeled 3-OH-Org 34694 reference. Because of the low amount of radioactivity, attempts to isolate T26 were unsuccessful.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The incubations with progesterone in uterus tissue confirmed the presence of 5-reductase, 3-HSD, and 20-HSD, as was described earlier (Lisboa and Holtermann, 1976). The same progesterone-metabolizing enzymes also appeared to be present in vagina and, to a lesser extent, in aorta.

Next, it was investigated whether these enzymes also acted upon the 19-nor-progestagen norethisterone and norethisterone derivatives. In the three tissues, 5-reduction followed by 3-reduction was the major pathway of norethisterone metabolism. Norethisterone was metabolized into mainly 3-OH,5-H-norethisterone in uterus, vagina, and aorta of proestrus and ovariectomized rats. Conversion of norethisterone was higher in vagina than in uterus, while aorta showed only limited conversion. In uterus and vagina, also the formation of 3-OH,5-H-norethisterone was demonstrated. The conversion of all norethisterone derivatives (Org OM38, Org 4060, and Org 34694) was much lower than of norethisterone itself (Table 2). Furthermore, the formation of 5-reduced forms of Org OM38, Org 4060, or Org 34694 was not found. Only the formation of 3-hydroxylated forms of Org OM38, Org 4060, and Org 34694 in uterus, vagina, and aorta could be demonstrated.

There is limited information as to the physiological function of the progesterone-metabolizing enzymes in the female reproductive tract and the cardiovascular system. In mice uterus, there is a combined action of 5-reductase, 3-HSD, and 20-HSD to catabolize progesterone in uterus, and the impaired progesterone metabolism in the cervix of 5-reductase type 1 isoenzyme knockout mice leads to a parturition defect in these mice (Mahendroo et al., 1999).

The reduction of the 3-keto, 4 moiety of norethisterone and norethisterone derivatives may be important for their pharmacology in HRT target tissues, since this reduction changes its progestogenic character. While norethisterone acts as a progestogen in prepubertal rabbits, where it induces the uteroglobin content of uterine flushing, neither 5-H-norethisterone or 3-OH,5-H-norethisterone shows this effect (Cerbon et al., 1990). Moreover, 5-H-norethisterone and 3-OH,5-H-norethisterone exhibited strong in vivo estrogenic effects by inducing progesterone receptor mRNA in rabbit uterus (Pasapera et al., 1995) and both progesterone receptor and c-fos mRNA in rat uterus (Mendoza-Rodriguez et al., 1999). It was also shown that 3-OH,5-H-gestodene and 3-OH,5-H-gestodene posses a weak intrinsic estrogenic activity (Lemus et al., 2000). The mixed hormonal profile of norethisterone with both progestogenic and estrogenic properties may thus be mediated by metabolism of norethisterone. Since Org 4060, Org OM38, and Org 34694 were not 5-reduced in uterus, vagina, or aorta, estrogenic activity through formation of tetrahydro-reduced metabolites seems unlikely in these tissues. The pharmacological significance of the formation of minor amounts of 3(OH)-Org 4060, Org OM38, and Org 34694 needs further investigation.

It was also shown that, although Org 4060 is not 5-reduced in uterus, vagina, or aorta, 5-H-Org 4060 itself is readily metabolized in all three tissues into mainly 3-OH,5-H-Org 4060 (40.4, 35.8, and 33.6% for uterus, vagina, and aorta, respectively), and in 3-OH,5-H-Org 4060 (4.3%) in aorta. This, together with the fact that 3-OH-5-H-norethisterone and not 5-H-norethisterone is the main metabolite of norethisterone, demonstrates that 5-reduction is the rate-limiting step in the metabolism of norethisterone and norethisterone derivatives in the three HRT target tissues. Apparently, both the 11-ethyl substituent that is the only structural difference between norethisterone and Org 4060, and the 7-methyl substituent of Org OM38 renders both Org 4060 and Org OM38 poor substrates for 5-reduction, compared with norethisterone. The steric hindrance of the 7-methyl substituent for steroid 5-reduction is also known from 7-methyl-19-nortestosterone. In contrast to 19-nortestosterone, 7-methyl-19-nortestosterone is not 5-reduced (Sundaram et al., 1995). Substitutions at the 11-position of the steroid skeleton have not been reported before to interfere with 5-reductase activity. Interestingly, two 19-nor-progestogen steroids with an 11-methylene substituent, desogestrel and Org 30569, were extensively metabolized in rat liver microsomes into mainly 3-(OH),5-H-reduced derivatives (Verhoeven et al., 1998b). Apparently, the rigidity that is imposed on the steroid C ring by the double bond of the 11-methylene group in desogestrel and Org 30569 does not interfere unfavorably with the activity of 5-reductase (Verhoeven et al., 1998a). It is difficult to speculate on the nature of the interference of the 11-ethyl group of Org 4060 with the 5-reductase enzyme. The extreme insolubility of 5-reductase, a reflection of the membrane-bound nature of the enzyme, has made its purification and structural characterization very difficult. Still, photolabeling and mutagenesis studies have indicated an N-terminal steroid D-ring binding domain, an NADPH cofactor binding domain, and a putative C-terminal catalytic steroid A-ring region (Thigpen and Russell, 1992; Bhattacharyya et al., 1999; Wang et al., 1999). There seems to be considerable tolerance of 5-reductase for substitutions to the steroid D-ring. All known 5-reductases accept both testosterone and progesterone as a substrate. Also norethisterone, a 17-ethinylated-19-nortestosterone, is readily 5-reduced. Moreover, human 5-reductase type 1 or 2 can 5-reduce campesterol, a C-27 plant steroid, reduction in the plants (Li et al., 1997). Maybe the 11-ethyl substitution of Org 4060 does not interfere with the steroid D-ring binding domain of 5-reductase, but rather with the NADPH binding domain.

The fact that 3-OH,5-H-Org 4060 was formed from 5-H-Org 4060 in comparable amounts in uterus, vagina, and aorta shows the presence of 3-HSD in all three tissues. The 3-HSD enzyme is known to work in concert with 5-reductases in target tissues to regulate levels of active steroid (Penning, 1997). Only one isoform of 3-HSD has been described in rat, with high-level expression in the liver. It seems probable that this enzyme is also involved in the conversion of Org 4060 and Org OM38 in 3-OH-Org 4060 and 3-OH-Org OM38 in uterus and vagina.

In the rat, four isoforms of 3-HSD have been described, each with its own enzymatic characteristics and tissue expression pattern. Which of these isoforms is responsible for the conversion of 5-H-norethisterone into 3-OH,5-H-norethisterone remains unclear. Apart from their oxidative 3-hydroxysteroid dehydrogenase/⁵-⁴ isomerase activity, homogenates of HeLa cells expressing rat 3-HSD type 1 and type 2 are also able to reduce dihydrotestosterone into 5-androstan-3,17-diol (Zhao et al., 1991). Ribonuclease protection analysis revealed the presence of rat 3-HSD type 1 and type 2 mRNA in rat uterus (Simard et al., 1993). On the other hand, the rat 3-HSD type 3 displays no isomerase activity and is considered a pure 3-ketosteroid reductase. Its expression, however, has only been demonstrated in male rat liver (Labrie et al., 1992). So, either the 3-reduction of 5-norethisterone in vagina and the uterus is catalyzed by type 1 and 2 3-HSD or there may be another 3-HSD isoform present in these tissues with ketosteroid reductase activity.

Overall, it was shown in this article that norethisterone is metabolized to its estrogenic 5-reduced metabolites in uterus, vagina, and aorta. Substitutions at the 7- and 11-position as in Org 4060, Org OM38, and Org 34694 prevented the metabolism to 5-reduced forms. However, limited conversion of Org 4060, Org OM38, and Org 34694 into their 3-(OH)-reduced form does occur in uterus, vagina, and aorta. Furthermore, the conversion of 5-H-Org 4060 into 3OH,5-H-Org 4060 in all three tissues demonstrates that 5-reduction is the rate-limiting step in the formation of 3-OH-,5-H-reduced metabolites of norethisterone and norethisterone derivatives in HRT target tissues. It could thus be possible that circulating 5-reduced norethisterone derivatives are a source for generating 3-OH-,5-H-reduced metabolites in uterus, vagina, and aorta. It also indicates that in vivo effects of 5-reduced norethisterone and 5-reduced norethisterone derivatives in HRT target tissues should be interpreted with prudence, because of their conversion to 3-OH-,5-H-reduced forms, which display estrogenic activity.