Interaction with anti-implantation and estrogen antagonistic activities of dl-ormeloxifene, a selective estrogen receptor modulator, by tetracycline in female Sprague-Dawley rats

Contraception 64 (2001) 261–269

Original research article

Interaction with anti-implantation and estrogen antagonistic activities of dl-ormeloxifene, a selective estrogen receptor modulator, by tetracycline in female Sprague-Dawley rats夞 R. Ghosh*, V. P. Kamboj, M. M. Singh Division of Endocrinology, Central Drug Research Institute, Lucknow-226 001, India

Abstract Among the 10 commonly used therapeutic agents investigated, concurrent oral administration of tetracycline (140 mg/kg) twice daily on Days 1–5 post-coitum (pc) interfered with the post-coital anti-implantation activity and almost completely abolished estrogen antagonistic activity of the single anti-implantation (1.5 mg/kg, orally) dose of dl-ormeloxifene administered on Day 1 pc, resulting in the occurrence of resorbed implantations in 50% of the females. However, no such interaction was evident when tetracycline was administered intramuscularly or when ormeloxifene was administered at twice its anti-implantation dose. There was no effect of ormeloxifene and/or tetracycline treatment on serum estradiol and progesterone levels, and all animals presented apparently normal corpora lutea. Ormeloxifene administered per se inhibited aminopyrine-N-demethylase (AD), glucose-6-phosphate dehydrogenase (G-6-PDH) and glutathione-Stransferase (GST) in the liver on the day of maximal endometrial receptivity, which was prevented by tetracycline co-administration. Aniline hydroxylase and AD were not detected in small intestine or uterus in vehicle control or any of the treatment groups. There was, however, no effect of ormeloxifene plus tetracycline treatment on serum total alkaline phosphatase activity. Findings suggest that interference with anti-implantation action of ormeloxifene by tetracycline might be due primarily to the almost complete abolition of its estrogen antagonistic activity at the uterine level, effected by decreased bioavailability of ormeloxifene and/or its active metabolite(s) by altered enterohepatic recirculation because of the effect on gut microflora. This might alternatively be related to an increased rate of its metabolism and elimination from the system via prevention of ormeloxifene-induced inhibition of hepatic AD, G-6-PDH, and GST, which, by effecting a decreased rate of metabolism, might be responsible for prolonged (⬃120 h) duration of estrogen antagonistic/anti-implantation action of ormeloxifene in this species. © 2001 Elsevier Science Inc. All rights reserved. Keywords: dl-Ormeloxifene; dl-Centchroman; Interaction; Anti-implantation; Estrogen agonistic; Antagonistic activities; Therapeutic agents; Tetracycline; Drug metabolic enzymes; Cytochrome P450; Aminopyrine-N-demethylase; Aniline hydroxylase; Glucose-6-phosphate dehydrogenase; Glutathione-Stransferase; Liver; Intestine; Uterus; Serum estradiol; Progesterone; Alkaline phosphatase

1. Introduction The combined oral contraceptives (COCs), though highly effective in controlled clinical trials (PI: 0.1), show up to a 5% failure rate during the first year in typical use [1]. This has been postulated to be the result primarily of interaction with certain concomitantly administered therapeutic agents.

夞 This research was supported by a grant of Emeritus Scientist scheme to Dr. V. P. Kamboj from the Extramural Research Division, Human Resource Development Group, Council of Scientific & Industrial Research, New Delhi, India. * Corresponding author. Tel.: ⫹91-522-213894; fax: ⫹91-522223405. E-mail address: [email protected] (R. Ghosh).

The first such clinical interaction in COC users receiving rifampicin and certain other antituberculous drugs was observed by Reimers and Jezek [2]. There have since been numerous case reports of COC users becoming pregnant while concomitantly receiving broad spectrum antibiotics, such as ampicillin, tetracycline, griseofulvin, chloramphenicol, sulfamides, and others [3–5]. Although failure of COCs in the presence of rifampicin is attributed primarily to the induction of hepatic mixed function oxidase system resulting in the rapid breakdown of the estrogen component of the COCs and its clearance from the system, broad spectrum antibiotics are believed to alter gut flora, preventing the release of active ethinyl estradiol (EE2) from its sulfuric and glucuronic acid conjugates, reducing reabsorption of EE2 into the portal circulation and hence its bioavailability [4].

0010-7824/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 0 1 0 - 7 8 2 4 ( 0 1 ) 0 0 2 5 7 - 8

262

R. Ghosh et al. / Contraception 64 (2001) 261–269

Ormeloxifene, a triphenylethylene selective estrogen receptor modulator (SERM), is known to provide good pregnancy protection in women in post-coital as well as weekly regimens [6]. Reports of its efficacy in the management of breast cancer in both men and women, dysfunctional uterine bleeding, post-menopausal osteoporosis, atherogenicity, dermatitis, restenosis, male infertility, and ovulation induction in amenorrhic women are available [6]. In view of the vast clinical application of SERMs [6,7] and the reported interaction of combined COCs with certain commonly used therapeutic agents (vide supra), evaluation of the effect of certain concomitantly administered therapeutic agents on anti-implantation/estrogen antagonistic activities of dlormeloxifene was needed. The present study reports the modulation of post-coital anti-implantation activity and estrogen agonistic and antagonistic activities of ormeloxifene and its effect on type I and II substrates of mixed function oxidase system on the day of maximal endometrial receptivity in the presence and absence of tetracycline in adult female Sprague-Dawley rats. Tetracycline, a broad spectrum antibiotic, finds vast use in pneumonia, peritonitis, chronic bronchitis, and urinary tract infections [8].

2.3. Post-coital contraceptive efficacy Adult female rats mated to males of proven fertility (Day 1: day of sperm-positive vaginal smear) were randomized into different groups and treated orally with the single anti-implantation (1.5 mg/kg) [10] dose of ormeloxifene or the vehicle (25% ethanol in distilled water) on Day 1 postcoitum (pc) along with certain commonly used therapeutic agents (Table 1). The dose of each therapeutic agent was calculated from the human dose on a surface area basis [11] and administered in a similar schedule. The first dose of each therapeutic agent was administered about 30 min after ormeloxifene treatment. At autopsy on Day 10 pc, number and status of corpora lutea and implantations in each animal were recorded (Table 1). Genital tracts of representative animals of each group were fixed in 10% formalin in phosphate-buffered saline for 24 h, preserved in 70% ethanol, and photographed. Because among the various therapeutic agents evaluated, interaction with post-coital contraceptive activity of ormeloxifene was observed only with concomitantly administered tetracycline, all further studies on the effect on estrogen agonistic and antagonistic activities and activity of hepatic, intestinal, and uterine drug metabolic enzymes were carried out with tetracycline.

2. Materials and methods 2.4. Estrogen agonistic and antagonistic activities 2.1. Chemicals and therapeutic agents Aminopyrine; aniline; nicotinamide adenine dinucleotide phosphate; nicotinamide; bovine serum albumin (BSA, fraction V); 17␣-ethinyl estradiol; and 1-chloro-2,4-dinitrobenzene were purchased from Sigma Chemical Co. (St. Louis, Missouri), and glutathione (reduced) and glucose-6phosphate from Sisco Research Laboratories (Mumbai). All other chemicals were of high purity and were purchased from Glaxo and E. Merck (India) Pure samples of cetrizine HCl, ciprofloxacin (Dr. Reddy’s Lab Ltd., Hyderabad), chlorpheniramine maleate (Cipla Ltd., Mumbai), and fluoxetine HCl (Torrent Research Center, Ahmedabad) were received as gifts. Digene, ibuprofen (Knoll Pharma Ltd., Mumbai), daonil (Hoechst Marrion Roussel Ltd., Mumbai), tetracycline (Terramycin, oxytetracycline HCl; Pfizer Ltd., Mumbai), resochin (chloroquine phosphate; Bayer India Ltd., Mumbai) and halopidol (Haloperidol; NR Jet Enterprises Ltd., Mumbai) were purchased from the local market. dl-Ormeloxifene (dl-centchroman) was synthesized [9] at the Central Drug Research Institute, Lucknow, India. 2.2. Animals Colony-bred immature (21 days old, 25–35 g) and adult (180 –220 g) Sprague-Dawley rats, maintained under standard conditions (22 ⫾ 1°C) with alternate 12 h light/dark periods and free access to pellet diet (Lipton India Ltd., Bangalore) and tap water were used.

Twenty-one day old, immature female rats were bilaterally ovariectomized under light ether anesthesia, and after post-operative rest for 7 days, were randomized into different treatment groups (Table 2). For estrogen agonistic activity, each rat received a single anti-implantation dose (1.5 mg/kg, orally) of ormeloxifene on Day 28 of age and/or tetracycline [140 mg/kg, orally/intramuscularly (im) twice daily] for 3 days beginning Day 28 of age. For estrogen antagonistic activity, each rat, in addition, received a 0.02 mg/kg dose of 17␣-ethinyl estradiol in 10% ethanol-distilled water once daily for 3 days on Days 28 –30 of age. Separate groups of animals receiving only the vehicle(s) for similar duration served as controls. At autopsy on Day 31 of age, a vaginal smear of each rat was taken, and the uterus was carefully excised, gently blotted, and weighed. 2.5. Drug metabolic enzymes Adult female rats treated orally with the single antiimplantation dose (1.5 mg/kg) of ormeloxifene on Day 1 pc and/or tetracycline (140 mg/kg, twice daily) on Days 1– 4 pc beginning 15 min after ormeloxifene treatment or the vehicle were decapitated and autopsied at 10:00 AM on Day 5 pc, i.e., day of maximal endometrial receptivity to blastocyst signal(s) in this species [10], and liver, small intestine (25 cm portion distal to pylorus), and uterus were quickly excised, washed with ice-cold physiological saline (0.85% NaCl in distilled water), and immediately processed for enzyme assays. About 2 mL blood samples were collected

R. Ghosh et al. / Contraception 64 (2001) 261–269

263

Table 1 Effect of certain common therapeutic agents on post-coital anti-implantation activity of ormeloxifene in Sprague-Dawley rats Treatment (oral) Vehicle Ormeloxifene Antacid Digenee Antiallergic Chlorpheniramine maleatee Cetrizine hydrochloridee Antidepressant Fluoxetine hydrochloridee Antidiabetic Glibenclamidee Anti-infective Tetracyclinee Ciprofloxacine Antimalarial Chloroquine phosphatee

Neuroleptic Haloperidole Nonsteroidal Anti-inflammatory Ibuprofene

Dosea (mg/kg)

Days pc* of treatment

Pregnantb/ treated rats

Corpora luteab,c

Implantationsb,d

1 1

6/6 0/5

11.67 ⫾ 1.02 10.60 ⫾ 0.40

9.33 ⫾ 0.76 –

1–5

0/4

10.75 ⫾ 0.85

–

0.6h 0.7g

1–5 1–5

0/5 0/6

8.63 ⫾ 0.75 7.83 ⫾ 3.20

– –

2.8f

1–5

0/6

9.33 ⫾ 0.33

–

0.7f

1–5

0/7

12.17 ⫾ 0.61

–

140g 70g

1–5 1–5

3/6 0/6

12.17 ⫾ 0.83 10.33 ⫾ 0.88

12.67 ⫾ 0.33m –

140k decreased to 24.5

1–3

0/5

11.20 ⫾ 3.84

–

1–5

0/6

9.33 ⫾ 0.21

–

1–5

0/6

11.00 ⫾ 0.84

–

– 1.5f 224f

0.7g 60h

* pc: post-coitum. Dose of each therapeutic agent was calculated from human dose on surface area basis. b Day 10 pc. c Mean ⫾ SEM of all treated rats. d Mean ⫾ SEM of only the pregnant rats. e Animals in these groups received, in addition, single anti-implantation dose (1.5 mg/kg, orally) of ormeloxifene on Day 1 pc 30 min before initiation of treatment of the therapeutic agent administered fonce, gtwice, or hthrice daily. k First dose of 140 mg/kg on the morning of Day 1 pc, followed by 70, 24.5, and 24.5 mg/kg doses at 6, 24, and 48 h, respectively, thereafter. m All implantations were resorbing. a

from each rat, kept at room temperature for 30 min, and centrifuged at 1000 ⫻ g. Serum samples were stored at ⫺20°C until, assayed. Small intestine samples were longitudinally incised, and the mucosal layer was scrapped by using a sharp scalpel blade. Liver samples were finely minced by using sharp scissors. Liver, intestinal mucosa, and uterus samples from each rat were separately homogenized in 10 mM Na⫹/K⫹ isotonic (1.15%) KCl, pH 7.4. Liver [12], intestinal [13], and uterine [14] microsomes were prepared by differential centrifugation. Microsomal pellets were dispersed in 50 mM phosphate buffer containing 10 mM EDTA, pH 7.4 (liver), 50 mM Tris buffer, 20% (v/v) glycerol, pH 7.8 (intestinal mucosa), or 10 mM TrisHCl buffer containing 0.5 M CaCl2, pH 7.2 (uterus). Microsomal protein content was determined [15] by using BSA (fraction V) as standard. 2.6. Enzyme assays 2.6.1. Microsomal incubation Mixtures containing ⬃2 mg of the microsomal protein (10,000 ⫻ g supernatant fraction) and an NADPH-generating system (consisting of 0.65 mol NADP, 10 mol glucose-

Table 2 Effect of tetracycline co-administration on estrogen agonistic and antagonistic activities of ormeloxifene* in ovariectomized immature rats Treatment

Vehicle Ormeloxifene 17␣-Ethinyl estradiol Tetracycline Ormeloxifene ⫹ tetracycline 17␣-Ethinyl estradiol ⫹ tetracycline Ormeloxifene ⫹ 17␣-Ethinyl estradiol Ormeloxifene ⫹ 17␣-Ethinyl estradiol ⫹ tetracycline

Uterine weight (mg/10g body weight) Oral

Intramuscular

5.22 ⫾ 0.25 8.62 ⫾ 0.75a 19.37 ⫾ 0.49b 11.06 ⫾ 0.86b 10.35 ⫾ 0.64b 17.72 ⫾ 0.39b,g

11.69 ⫾ 0.34b,d 11.87 ⫾ 0.93b,c 19.87 ⫾ 1.58b,f

14.53 ⫾ 0.86d,e 16.84⫾1.09d,f

15.49 ⫾ 0.61h

* Ormeloxifene: 1.5 mg/kg, orally, on Day 28 of age; Tetracycline: 140 mg/kg, orally twice daily, on Days 28 –30 of age. a p ⬍ 0.01, b p ⬍ 0.001 vs. vehicle control group. c p ⬍ 0.05, d p ⬍ 0.01 vs. corresponding ormeloxifene per se-treated group. e p ⬍ 0.01 vs. 17␣-ethinyl estradiol per se-treated group. f p ⬍ 0.01, g p ⬍ 0.001 vs. corresponding tetracycline per se-treated group. h p ⬍ 0.05 vs. corresponding 17␣-ethinyl estradiol ⫹ tetracyclinetreated group. All other relevant comparisons were insignificant.

264

R. Ghosh et al. / Contraception 64 (2001) 261–269

6-phosphate, 50 mol nicotinamide, and 25 mol magnesium chloride in 0.05 M sodium phosphate buffer, pH 7.4) were used for all incubations. All reactions were initiated by the addition of a NADPH-generating system and were conducted at 37°C in an oscillating water bath. 2.6.2. Aminopyrine-N-demethylase The rate of aminopyrine-N-demethylation was estimated by the formation of formaldehyde. Aminopyrine (5 mol) and semicarbazide (45 mol) were added to the incubation mixture to a final volume of 1.5 mL. Reaction was terminated after 30 min by the addition of 0.5 mL 15% zinc sulfate and 0.5 mL saturated barium hydroxide solution. The precipitated protein was removed by centrifugation, and the concentration of formaldehyde was quantified by the method of Nash [16]. Commercial formaldehyde was used as standard. 2.6.3. Aniline hydroxylase Hydroxylation of aniline to paraminophenol (PAP) was determined as described previously [17]. Briefly, aniline (5 mol) was added to the incubation mixture to a final volume of 1 mL. Reaction was terminated after 30 min by the addition of 0.5 mL 20% trichloroacetic acid. The precipitated protein was removed by centrifugation, and the resultant supernatant was mixed with 0.25 mL 10% sodium carbonate and 0.5 mL of 2% phenol in 2 N sodium hydroxide for measurement of PAP.

Fig. 1. Genital tracts of (A) vehicle control, (B) ormeloxifene per se-, (C) tetracycline per se- and (D) ormeloxifene plus tetracycline-treated rats. Note the presence of all resorbed implantations in ormeloxifene plus tetracycline-treated rats and the normal implantation in rats receiving tetracycline per se or the vehicle. The absence of implantations in ormeloxifene per se-treated rats indicates its anti-implantation action. Scale bar represents 1 cm.

2.7. Statistical analysis The data were analyzed by the Student’s t-test.

3. Results 3.1. Post-coital contraceptive efficacy of ormeloxifene

2.6.4. Glucose-6-phosphate dehydrogenase The assay mixture contained 20 mM Tris-HCl buffer (pH 7.4), 10 mM magnesium chloride, 10 mM glucose-6-phosphate, 0.24 mM NADP, and an appropriate amount of cytosolic protein. The enzyme was assayed spectrophotometrically for 5 min at 340 nm [18]. 2.6.5. Glutathione-S-transferase Glutathione-S-transferase (GST) was assayed in cytosolic fraction spectrophotometrically at 340 nm. The assay mixture contained 1 mL 0.1 M potassium phosphate buffer (pH 6.5), 10 mM glutathione (reduced), an appropriate amount of cytosolic protein, and 2 mM 1-chloro-2,4-dinitrobenzene in a total volume of 3 mL [19]. 2.6.6. Serum total alkaline phosphatase Serum total alkaline phosphatase was assayed spectrophotometrically (405 nm) by using standard commercial kit (cat. no. 396494; Boehringer Mannheim, Germany). 2.6.7. Serum estradiol and progesterone Serum estradiol (cat. no. 1298470) and progesterone (cat. no. 1204475) were assayed by ELISA by using standard commercial kits (Boehringer Mannheim).

Of the 10 commonly used therapeutic agents evaluated in rats (Table 1), tetracycline exhibited interaction with the post-coital contraceptive action of ormeloxifene, as evidenced by the presence of resorbed implantations (Fig. 1) in 50% of the animals. There was no effect of ormeloxifene and/or tetracycline treatment on serum estradiol and progesterone levels (Table 3), and all animals presented apparently normal corpora lutea. However, no interaction was observed when either tetracycline was administered by im

Table 3 Effect of post-coital ormeloxifene and tetracycline treatment on serum levels of estradiol, progesterone, and total alkaline phosphatase activity on day of maximum endometrial receptivity in adult female SpragueDawley rats Treatment

Estradiol (pg/dL)

Progesterone (ng/dL)

Total alkaline phosphatase (U/L)

Vehicle Ormeloxifene ⫹ tetracycline

55.22 ⫾ 5.79 72.70 ⫾ 8.65

78.56 ⫾ 9.07 68.67 ⫾ 2.48

272.33 ⫾ 20.46 255.33 ⫾ 28.35

Ormeloxifene : 1.5 mg/kg, orally, Day 1 pc; Tetracycline : 140 mg/kg, orally, twice daily, Days 1–5 pc; All relevant comparisons were insignificant.

R. Ghosh et al. / Contraception 64 (2001) 261–269

265

Table 4 Effect of tetracycline co-administration on post-coital anti-implantation efficacy of ormeloxifene in rat Treatment

Vehicle Ormeloxifene Tetracycline Ormeloxifene ⫹ Tetracycline Ormeloxifene ⫹ Tetracycline Ormeloxifene ⫹ Tetracycline

Dose (mg/kg)

Route

– 1.5b 140c 1.5b 140c 1.5b 140c 3.0b 140c

Oral Oral Oral Oral Oral Oral im Oral Oral

Days pc of treatment

Pregnanta/ treated rats

Corpora luteaa

1–5 1 1–5 1 1–5 1 1–5 1 1–5

6/6 0/5 6/6 3/6

Implantationsa Total

Normal

Resorbed

70 53 72 73

56 0 52 38

56

0

52 0

0 38

0/4

51

0

0/5

60

0

* pc: post-coitum; im: intramuscular. Day 10 pc; administered bonce or ctwice daily.

a

route or ormeloxifene was administered at twice its antiimplantation dose (Table 4). 3.2. Estrogen agonistic and antagonistic activities Single oral administration of ormeloxifene at its antiimplantation dose (1.5 mg/kg, orally) to ovariectomized immature rats induced a weak uterotrophic effect (65%, p ⬍ 0.001, versus vehicle control group; Table 2; Fig. 2a), and only 4 – 6% of the cells in the vaginal smears were cornified. Tetracycline administered alone (orally/im) or in combination with ormeloxifene, too, caused significant uterine weight gain (112%, 124%, 98%, and 127%, p ⬍ 0.001, versus vehicle control group), and the extent of cornification of vaginal epithelium was comparable to ormeloxifene per se treatment group. The increase in uterine weight, when compared to the ormeloxifene per se treatment group, was significantly greater when tetracycline was administered im either alone (p ⬍ 0.01) or in combination with ormeloxifene (p ⬍ 0.05). The apparently lower uterotrophic response in rats receiving tetracycline by the oral (98%) rather than by the im (127%) route along with ormeloxifene was, however, insignificant. 17␣-Ethinyl estradiol (0.02 mg/kg) administered orally to ovariectomized immature rats once daily for 3 days produced about a four-fold increase in uterine weight (p ⬍ 0.001, versus vehicle control group) and an estrous type of vaginal smear picture. Although tetracycline (orally/im) coadministration produced no significant effect, ormeloxifene per se caused significant inhibition (25%, p ⬍ 0.01; Fig. 2b) in EE2-induced uterine weight gain. In rats receiving, in addition, tetracycline by the oral route, there was an almost complete abolition of the estrogen antagonistic action of ormeloxifene, and the inhibition in EE2-induced uterine weight gain was only of the order of about 5%. However, no such interaction was observed when tetracycline was administered by the im route, with about 22% inhibition in EE2-induced uterine weight gain.

3.3. Activity of drug metabolic enzymes 3.3.1. Liver Single anti-implantation dose (1.5 mg/kg, orally) of ormeloxifene administered on Day 1 pc inhibited (p ⬍ 0.001, versus vehicle control group, Table 5) activity of aminopyrine-N-demethylase, glucose-6-phosphate dehydrogenase (G-6-PDH), and GST in the liver, while aniline hydroxylase (AH) activity remained unaltered. In comparison, tetracycline, although increasing the activity of hepatic AH, GST (p ⬍ 0.001), and G-6-PDH (81%, insignificant), also inhibited (p ⬍ 0.05) aminopyrine-N-demethylase activity; the decrease, however, was of a lesser order than that after ormeloxifene per se treatment (p ⬍ 0.05). When coadministered with ormeloxifene, tetracycline prevented ormeloxifene-induced inhibition in the activity of G-6PDH, GST (p ⬍ 0.001, versus ormeloxifene per se treatment group), as well as aminopyrine-N-demethylase (58%, insignificant), and the levels were comparable (p ⬎ 0.05) to that after tetracycline per se treated group. The activity of AH, although significantly lower than in the tetracycline per se treated group (p ⬍ 0.001), was comparable to the vehicle or ormeloxifene per se treated groups. 3.3.2. Extrahepatic tissues Aminopyrine-N-demethylase as well as AH activity were not detected in either the small intestine or the uterus in vehicle control or any of the treatment groups (Tables 6 and 7). In the small intestine, ormeloxifene and/or tetracycline treatment produced a marginal (insignificant) decrease in G-6-PDH activity. Although ormeloxifene (36%) as well as tetracycline (p ⬍ 0.05, versus vehicle control group) per se treatment significantly decreased GST activity in the intestine, no change in its activity was evident when these were administered in combination (p ⬎ 0.05, versus vehicle control; p ⬍ 0.05 versus tetracycline per se treatment group, Table 6). In the uterus, like that in the liver, ormeloxifene per se treatment markedly inhibited the activity of G-6-PDH as

266

R. Ghosh et al. / Contraception 64 (2001) 261–269

Fig. 2. (A) Estrogen agonistic and (B) antagonistic activities of ormeloxifene (Orm) following tetracycline (T) co-administration in ovariectomized immature rats. Both ormeloxifene (1.5 mg/kg, single dose on Day 1) and 17␣-ethinyl estradiol (EE2; 0.2 mg/kg, once daily on Days 1–3) were administered by the oral route, whereas tetracycline (140 mg/kg, twice daily on Days 1–3) was administered by either the oral or the im route. Note weak estrogen agonistic activity of ormeloxifene as well as tetracycline (per se or in combination) in comparison to EE2, but complete abolition of estrogen antagonistic action of ormeloxifene (24.8% inhibition in EE2-induced increase in uterine weight) when tetracycline was coadministered by the oral route (5% inhibition), but not when administered parenterally (22% inhibition).

well as GST enzymes (p ⬍ 0.001 and p ⬍ 0.01, respectively, versus vehicle control group), which was prevented by tetracycline co-administration (p ⬍ 0.001 and p ⬍ 0.01, respectively, versus ormeloxifene per se-treated group), and the levels were comparable to that of the vehicle control group (Table 7). Tetracycline per se treatment had no significant effect on G-6-PDH or GST activity.

4. Discussion Results of this study provide evidence of the interaction of the orally administered broad spectrum antibiotic tetracycline, commonly used for pneumonia, peritonitis, chronic

bronchitis, and urinary tract infections [8], with the postcoital anti-implantation action of the selective estrogen receptor modulator dl-ormeloxifene in female Sprague-Dawley rats, resulting in the occurrence of resorbed implantations in 50% of the treated females. Ormeloxifene acts by inhibiting endometrial receptivity to blastocyst signal(s) by antagonism of the action of nidatory estrogen at the receptor level. It has no effect on hypothalamo-pituitary-ovarian axis, mating behavior, ovulation, fertilization, blastocyst formation, time or extent of secretion of nidatory estrogen and progesterone, or uterine progesterone receptor kinetics during the pre-implantation period at the contraceptive dose [10]. The precise reason for resorption of all implantations in ormeloxifene plus tetracycline-treated rats in the present study, in view of the unaltered plasma steroid levels, remains unexplained. The failure of anti-implantation action of ormeloxifene in the presence of orally administered tetracycline appears to be due primarily to almost complete abolition of its estrogen antagonistic activity at the uterine level, effected by decreased bioavailability of ormeloxifene and/or its active metabolites by altered enterohepatic recirculation caused by the effect on gut microflora. This might alternatively be related to an increased rate of its metabolism and elimination from the system via prevention of ormeloxifene-induced inhibition of hepatic aminopyrine-N-demethylase activity. This is supported by the observed significant decrease in activity of GST, a phase II enzyme responsible for detoxification of electrophilic alkylating agents including carcinogens and cytotoxic drugs [20] in the liver and uterus in ormeloxifene per se-treated rats, and the reversal in the presence of tetracycline might support the reduced bioavailability of ormeloxifene via increased excretion of its conjugates from the system. Pertinently, tetracycline co-administration has also recently been reported to markedly decrease Cmax1 and Cmax2 of ormeloxifene observed at 4 and 18 h, AUC0 –24h (area under the curve), absorption rate constant, AUC0-⬀ and absolute bioavailability of ormeloxifene and an increase in its mean absorption time and clearance rate. Interestingly, contraceptive efficacy of ormeloxifene was restored when tetracycline was administered along with Sporlac (Lactobacillus sporogens [21]). The significant decrease in G-6-PDH in the uterus and liver in ormeloxifene per se-treated rats and the prevention of this inhibition when it is co-administered with tetracycline might support the suggested role of free radicals in implantation and anti-implantation action of ormeloxifene in this species [10]. Absence of aminopyrine-N-demethylase and AH enzyme activities in the small intestine might be related to its reported lower protein content and catalytic activity [22]. Selective estrogen receptor modulators of the triphenylethylene series, such as ormeloxifene, are known to have a long duration of anti-implantation/estrogen antagonistic action [23]. This has been suggested to be related to their storage in body fat [24] caused by their high lipophilicity

R. Ghosh et al. / Contraception 64 (2001) 261–269

267

Table 5 Effect of post-coital ormeloxifene and/or tetracycline treatment on certain drug metabolic enzymes in the liver on day of maximum endometrial receptivity in rat* Treatment

AminopyrineN-demethylasea

Aniline hydroxylaseb

Glucose-6phosphate dehydrogenasec

Glutathione-Stransferasec

Vehicle Ormeloxifene Tetracycline Ormeloxifene ⫹ tetracycline

1.90 ⫾ 0.31 0.59 ⫾ 0.11e 0.96 ⫾ 0.02d,f 0.93 ⫾ 0.16d

0.53 ⫾ 0.07 0.44 ⫾ 0.02 1.20 ⫾ 0.01e,g 0.43 ⫾ 0.03h

10.12 ⫾ 0.92 5.43 ⫾ 0.04e 18.35 ⫾ 9.65 15.60 ⫾ 0.17e,g

58.00 ⫾ 2.52 26.00 ⫾ 5.90e 164.00 ⫾ 0.40e,g 170.00 ⫾ 4.00e,g

* Values are mean ⫾ SEM of 2–3 independent assays; Ormeloxifene: 1.5 mg/kg, orally, Day 1 pc; Tetracycline: 140 mg/kg, orally, twice daily, Days 1–5 pc. a ␮g HCHO formed/mg protein. b ␮g p-aminophenol formed/mg protein. c D/␮g/sec, ⫻10⫺6. d p ⬍ 0.05, e p ⬍ 0.001 vs. corresponding vehicle control group. f p ⬍ 0.05, g p ⬍ 0.001 vs. corresponding ormeloxifene per se-treated group. h p ⬍ 0.001 vs. corresponding tetracycline per se-treated group. All other relevant comparisons were insignificant.

(dl-ormeloxifene: log p ⫽ 7.41 [25]) or binding to certain antiestrogen binding sites (dl-ormeloxifene: RBA-AEBS: 19⫹5; RBA-ER: 5.42⫹1.45; percent of estradiol-17␤ [26, 27]), and their gradual and consistent release into the system from these storage sites. This is suggested to result in prolonged availability of such antiestrogens and/or their active metabolites, which by causing marked depletion in the available estrogen receptor pool appear to render uterus refractory to action of nidatory estrogen secreted late on Day 4 pc in this species [23]. The present study, in addition, provides evidence of a delicate mechanism whereby selective estrogen receptor modulators such as dl-ormeloxifene might regulate their own metabolism and elimination from the system, effected by the inhibition in the activity of certain crucial drug metabolic enzymes in the liver, which is prevented by co-administered tetracycline. This is evidenced by the significant inhibition in the activity of hepatic aminopyrine-N-demethylase, known to have broad sub-

strate specificity for multiple P450 isoforms, including CYP1A1, 1A2, 2B1, 2B2, 2C11, and 2C12 [28], by ormeloxifene, and reversal of this inhibition in rats treated conjointly with ormeloxifene and tetracycline. According to Meltzer et al. [29], this inhibition in hepatic cytochrome P450-dependent mixed function oxidation in rats pretreated with tamoxifen as well as its N-desmethyl and 4-hydroxy metabolites was due to their occupying the Type-I binding sites of cytochrome P450, limiting accessibility of other substrates to active sites of the enzyme. Pertinently, although ormeloxifene is actively metabolized by the rat liver homogenate in vitro to biologically active as well as inactive metabolites, with active metabolites contributing to estrogen agonistic and anti-implantation activities and the inactive metabolites accounting for its gradual metabolic disposition [30], all three biologically active metabolites constitute its demethylated (viz., 7-desmethyl-, 2-desmethyl- and 2-monomethyl-) products [30]. Of these, 7-des-

Table 6 Effect of post-coital ormeloxifene and/or tetracycline treatment on certain drug metabolic enzymes in the small intestine on day of maximum endometrial receptivity in rat Treatment

AminopyrineN-demethylasea

Aniline hydroxylaseb

Glucose-6phosphate dehydrogenasec

Glutathione-Stransferasec

Vehicle Ormeloxifene Tetracycline Ormeloxifene ⫹ tetracycline

ND ND ND ND

ND ND ND ND

6.35 ⫾ 2.10 5.02 ⫾ 1.76 4.88 ⫾ 2.42 4.27 ⫾ 1.97

29.10 ⫾ 4.12 18.50 ⫾ 6.03 10.60 ⫾ 5.26d 30.80 ⫾ 4.68e

␮g HCHO formed/mg protein. ␮g p-aminophenol formed/mg protein. c D/␮g/sec, ⫻10⫺6. d p ⬍ 0.05 vs. corresponding vehicle control group. e p ⬍ 0.05 vs. corresponding tetracycline per se-treated group. Other legends are the same as in Table 5. ND: Not detected. All other relevant comparisons were insignificant. a

b

268

R. Ghosh et al. / Contraception 64 (2001) 261–269

Table 7 Effect of post-coital ormeloxifene and/or tetracycline treatment on certain drug metabolic enzymes in the uterus on day of maximum endometrial receptivity in rat Treatment

AminopyrineN-demethylasea

Aniline hydroxylaseb

Glucose-6phosphate dehydrogenasec

Glutathione-Stransferasec

Vehicle Ormeloxifene Tetracycline Ormeloxifene ⫹ tetracycline

ND ND ND ND

ND ND ND ND

18.50 ⫾ 0.92 1.00 ⫾ 0.20e 13.30 ⫾ 4.10f 13.20 ⫾ 0.63d,h

41.00 ⫾ 7.06 6.85 ⫾ 2.35d 35.40 ⫾ 11.60f 28.80 ⫾ 3.10g

␮g HCHO formed/mg protein. ␮g p-aminophenol formed/mg protein. c D/␮g/sec, ⫻10⫺6. d p ⬍ 0.01, e p ⬍ 0.001 vs. corresponding vehicle control group. f p ⬍ 0.05, g p ⬍ 0.01, hp ⬍ 0.001 vs. corresponding ormeloxifene per se-treated group. Other legends are the same as in Table 5. ND: Not detected. All other relevant comparisions were insignificant. a

b

methyl ormeloxifene has earlier been considered as the possible active metabolite of ormeloxifene in vivo [31]. Marked increase in the activity of hepatic microsomal AH, aminopyrine N-demethylase, cytochrome P450, and cytochrome-b5, indicating rapid disposal of the compound from the body, has also been observed in adult female rhesus monkeys treated with a 25 mg/kg dose of ormeloxifene 8 h before autopsy [32]. Reports of similar interactions between certain anti-tubercular drugs [2], anticonvulsants [3,33], antibiotics [4], etc., and COCs resulting in occurrence of unplanned pregnancies [1] are available. Broad spectrum antibiotics, such as chloramphenicol, ampicillin, and sulphamides [5], have also been reported to reduce bioavailability of estrogen component of the COCs by altering gut microflora and preventing release of the active EE2 component of the COCs from its sulfuric and glucuronic acid conjugates and reducing its reabsorption into the portal circulation [4]. In the present study, however, tetracycline (140 mg/kg) administered twice daily for 3 consecutive days by oral or im route did not appear to significantly alter the estrogenic profile of 17␣-ethinyl estradiol at the uterine level in this species. The lack of effect on EE2 bioavailability (vide supra) in the present study might be due to comparatively much shorter duration of treatment in this study. To the best of our knowledge, this is the first report of the interaction of a selective estrogen receptor modulator with the broad spectrum antibiotic tetracycline at the uterine level. Information is also not available on the effects of such an interaction on the other estrogen target tissues. However, in view of the wide clinical application of the multifunctional molecules such as ormeloxifene [6,7,34] and the consequence of similar drug-drug interactions on the occurrence of unplanned pregnancies and in the management of hormone-related clinical disorders, a cautious approach on use of additional/alternate method of contraception and treatment modalities during the period of such broad spec-

trum antibiotic therapy appears highly warranted to safeguard the few unidentifiable women at risk.

Acknowledgments The authors thank Mr. Jagdish Prasad and Mr. Bishambhar Prasad Misra for the handling of animals, Mr. Manoj Khurana for help in treatments, and Mr. Mahesh Tewari and Mrs. Sudha Vaishnavi for the handling of instruments.

References [1] Weaver K, Glasier A. Interaction between broad-spectrum antibiotics and the combined oral contraceptive pill. Contraception 1999;59: 71– 8. [2] Reimers D, Jezek A. Rifampicin und andere antituberkulostatika bei gleichzeitige oraler kontrazeption. Prax Pneumol 1971;25:255– 62. [3] Fernandez N, Sierra M, Diez MJ, Teran T, Pereda P, Garcia JJ. Study of the pharmacokinetic interaction between ethinylestradiol and amoxicillin in rabbits. Contraception 1997;55:47–52. [4] Back DJ, Bates M, Bowden A. The interaction of phenobarbital and other anticonvulsants with oral contraceptive steroid therapy. Contraception 1980;22:495–503. [5] Szoka PR, Edgren RA. Drugs interaction with oral contraceptives: compilation and analysis of an adverse experience report database. Fertil Steril 1998;49:31– 8. [6] Singh MM. Centchroman, a selective estrogen receptor modulator, as a contraceptive and for the management of hormone related clinical disorders. Med Res Rev 2001;21:302– 47. [7] Macgregor JI, Jordan VC. Basic guide to the mechanism of antiestrogen action. Pharma Rev 1998;50:151–96. [8] Murakami K, Tateda K, Matsumoto T, Mujizaki S, Yamaguchi K. Efficacy of a novel tetracycline derivative glycycline against penicillin resistant streptococcus pneumonia in a mouse models of pneumonia. J Antimicrob Chem 2000;46:629 –32. [9] Ray S, Grover PK, Kamboj VP, Setty BS, Kar AB, Anand N. Antifertility agents. 12. Structure-activity relationship of 3,4-diphenyl chromenes and chromans. J Med Chem 1976;19:276 –9. [10] Singh MM, Trivedi RN, Chauhan SC, Srivastava VML, Kamboj VP. Uterine estradiol and progesterone receptor concentration, activities

R. Ghosh et al. / Contraception 64 (2001) 261–269

[11]

[12]

[13]

[14] [15] [16] [17]

[18]

[19]

[20] [21]

[22]

of certain antioxidant enzymes and dehydrogenases and histoarchitecture in relation to time of secretion of nidatory estrogen and high endometrial sensitivity in rat. J Steroid Biochem Mol Biol 1996;59: 215–24. Freireich EJ, Gehan EA, Rall DP, Schmidt LH, Skipper HE. Quantitative comparison of toxicity of anti-cancer agents in mouse, rat, dog, monkey and man. Cancer Chemother Repts 1966;50:219 – 44. Kremers P, Beaune P, Cresteil T, et al. Cytochrome P-450 monooxygenase activity in human and rat liver microsomes. Eur J Biochem 1981;118:599 – 606. Lindeskog P, Haapranta T, Norgard M, Glaumann H, Hansson T, Gustaffson J. Isolation of rat intestinal microsomes: Partial characterization of mucosal cytochrome P-450. Arch Biochem Biophys 1986;244:492–501. Lyttle RC, DeSombre ER. Uterine peroxidase as a marker for estrogen action. Proc Nat Acad Sci, USA 1977;74:3162– 6. Lowry OH, Rosenbrough NJ, Farr AR, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–75. Nash T. The colorimetric estimation of formaldehyde by means of the Hantzsch reaction. Biochem J 1953;55:412–22. Schenkman JB, Remmer H, Estabrook RW. Spectral studies of drug interaction with hepatic microsomal cytochrome. Mol Pharmacol 1967;3:113–23. Shonk CE, Boxer GE. Enzyme patterns in human tissues I. Methods for the determination of glycolytic enzymes. Cancer Res 1964;24: 709 –20. Habig WH, Pabst UJ, Jakoby WB. Glutathione-S-transferase A. The first step in mercapturic acid formation. J Biol Chem 1974;249: 7130 –9. Rushmore TH, Pickett CB. Glutathione-S-transferase: structure, regulation, and therapeutic implications. J Biol Chem 1993;268:11475– 8. Gupta S, Mukhopadhyay A, Ray S, Giri AK. Comparative antimutagenic effects of d- and l-centchroman and their comparison with tamoxifen in salmonella assay. Mutation Res 1999;50:219 – 44. Lin JH, Chiba M, Baille TA. Is the role of the small intestine in first-pass metabolism overemphasized? Pharmacol Rev 1999;51:135– 53.

269

[23] Singh MM, Sreenivasulu S, Kamboj VP. Duration of anti-implantation action of the triphenylethylene antiestrogen centchroman in adult female rats. J Reprod Fertil 1994;100:371– 4. [24] Trivedi RN, Chauhan SC, Dwivedi A, Kamboj VP, Singh MM. Time-related effects of triphenylethylene antiestrogen-induced changes in uterine weight, estrogen receptors, and endometrial sensitivity in rats. Contraception 1995;51:367–79. [25] Saha A, De AU, Sengupta C. Centchroman: effect on blood lipids. Indian J Pharm Sci 1999;61:268 –70. [26] Sudo K, Monsma EJ Jr, Katzenellenbogen BS. Antiestrogen binding sites distinct from the estrogen receptors. Endocrinology 1983;112: 425–34. [27] Saeed A, Durani N, Durani S, Ray S, Kapil RS. Cis-isomer of centchroman—a selective ligand for the microsomal antiestrogen binding site. Biochem Biophys Res Commun 1984;125:346 –52. [28] Wei X, Loi C, Schmucker DL, Vestal RE. Characterization of the independent and combined effects of two inhibitors on oxidative drug metabolism in rat liver microsomes. Biochem Pharmacol 1995;49: 1657– 63. [29] Meltzer NM, Stang P, Sternson LA. Influence of tamoxifen and its N-desmethyl and 4-hydroxy metabolites on rat liver microsomal enzymes. Biochem Pharmacol 1984;33:115–23. [30] Ratna S, Roy SK, Ray S, Kole PL, Salman M, Madhusudan K, et al. Centchroman: in vitro metabolism by rat liver homogenate. Drug Dev Res 1986;7:173– 8. [31] Paliwal JK, Gupta RC. Tissue distribution and pharmacokinetics of centchroman a new nonsteroidal post-coital contraceptive agent and its 7-desmethyl metabolite in female rats after a single oral dose. Drug Metab Dispos 1996;24:148 –55. [32] Mishra NC, Ray S, Roy SK. Influence of centchroman on some hepatic microsomal enzymes in female rhesus monkeys. Med Sci Res 1990;18:881–2. [33] Burrough KE, Chambliss ML. Antibiotics and oral contraceptive failure. Arch Fam Med 2000;9:81–2. [34] Sato M, Grese TA, Dodge JA, Bryant HU, Turner CH. Emerging therapies for the prevention or treatment of post-menopausal osteoporosis. J Med Chem 1999;42:1–24.