INTERACTIONS OF OESTRADIOL-17\g=b\ AND TAMOXIFEN IN THE UTERUS OF THE PREGNANT RAT J. S. MAJOR, B. GREEN
Department
P. J. HEALD of Biochemistry, University of Strathclyde, Biosciences 31 Taylor Street, Glasgow, G4 ONR AND
Building,
(Received 22 January 1976) SUMMARY
uptake
and retention of
a radioactive post-coital antifertility agent have shown that the ovary retained more radioactivity than did any other reproductive organ. Studies have also been made of the uptake and distribution of [3H]tamoxifen and [3H]oestradiol-17\g=b\ in the uterus of the pregnant rat on days 2\p=n-\6post coitum. Twenty-four hours after administration of tamoxifen, either i.v. or orally, 40\p=n-\50% of the radioactivity was in the high speed pellet, 10\p=n-\20% in the nuclear fraction, and 15\p=n-\30% in the cytosol. An equivalent dose of [3H]oestradiol-17\g=b\ yielded distributions of 5 %, 5 % and 82 % respectively. Fractionation of uteri from animals given 0\m=.\2mg tamoxifen/kg on Day 2 of pregnancy followed by [3H]oestradiol 60 min before death showed little difference in total uptake of oestradiol or distribution in the subcellular fraction on Days 4, 5 and 6. Although uptake of oestradiol by uterine nuclei was reduced on Day 3 by previous administration of tamoxifen on Day 2, appreciable quantities were still bound to the nuclear receptors. Treatment of ovariectomized animals with tamoxifen at doses up to 40 \g=m\g/rat (i.e. 0\m=.\2mg/kg) led to the accumulation of oestrogen\p=n-\receptorcomplex in the nucleus. It is concluded that the antifertility properties of tamoxifen (under the conditions of these experiments) cannot be ascribed to the suppression of uptake and binding of oestradiol by the uterus.
Measurement of the
tamoxifen, by reproductive tissues of the
rat
INTRODUCTION
post-coital antifertility agent tamoxifen [ICI 46,474: trans-l-(p-/?-dimethylaminoethoxyphenyl)-l,2-diphenylbut-l-ene], have shown that in appropriate doses it can delay implantation by up to 25 h in the pregnant rat (Major & Heald, 1974). Examination of this phenomenon suggested that tamoxifen delays or eliminates the rise in nidatory plasma oestradiol primarily by inhibiting its synthesis by the ovary (Watson, Anderson, Alam, O'Grady & Heald, 1975; Watson & Alam, 1976). Tamoxifen is antioestrogenic in the rat (Harper & Walpole, 1967) and since it combines with uterine cytoplasmic oestrogen receptors in vitro (Skidmore, Walpole & Woodburn, 1972), it has been suggested that one alternative means by which it could exert its antifertility effect is by suppression of the uterine uptake of nidatory oestrogen. To examine this situation further the time course of uptake and retention of tamoxifen by pregnant rat uterus was measured and also the uptake of oestradiol-17/? in vivo by uteri of rats receiving tamoxifen on Day 2 Studies
on
the mechanism of action of the
of pregnancy. *
Present address: ICI Pharmaceuticals
Ltd, Alderley Park, Alderley Edge, Cheshire.
The subcellular distribution of tamoxifen was also examined and compared with that of oestradiol-17/?. Evidence for intra-nuclear binding of tamoxifen was obtained by utilizing the [3H]oestradiol exchange technique developed by Anderson, Clark & Peck (1972). MATERIALS AND METHODS
The rats used, conditions of mating and administration of tamoxifen previously (Major & Heald, 1974; Watson et al. 1975).
were as
described
Chromatographie identification of tamoxifen in uterine and liver extracts All glassware was treated with 5 % (v/v) dimethyldichlorosilane (BDH Chemicals Ltd, Poole, Dorset) in toluene to reduce adsorption losses. Animals were dosed orally on Day 2 with 0-4 mg [3H]tamoxifen/kg and killed 6-72 h later. The perfused liver was homogenized in 20 ml methanol containing non-radioactive tamoxifen as carrier (50 /¿g/ml), and after centrifuging and twice re-extracting the residue with 10 vol. methanol/carrier solution the combined extracts were evaporated to dryness under nitrogen at 40 °C. A similar procedure was used for uterine tissue except that 20 vol. methanol were employed in the re-extractions. The dried extracts were taken up in the smallest volume of methanol required to produce a clear solution and aliquots were separated on Kieselgel HF254 plates using benzene-triethylamine (9:1, v/v) (Fromson, Pearson & Bramah, 1973). After development tamoxifen was detected by u.v. light, as little as 250 ng producing a recognizable fluorescence. Areas of the chromatograms corresponding to a tamoxifen marker were scraped into counting vials con¬ taining 1-0 ml H20, 10 ml butyl PBD Triton X scintillator were added, and the samples counted.
[3H]Tamoxifen [sp. act. 240 /¿Ci/mg (89-2 mCi/mmol)] was supplied by Dr A. L. Walpole was at least 98 % pure as determined by t.l.c. [2,4,6,7-3H]Oestradiol-17/ff (45 Ci/mmol) was obtained from the Radiochemical Centre, Amersham.
(ICI Pharmaceuticals Ltd) and
of [3H]tamoxifen and [3H]oestradiol-17ß To study the retention of tamoxifen in the uterus and other tissues, rats in groups of three were dosed orally between 09.00 and 10.00 h on Day 2 of pregnancy with 0-2 mg [3H]tamoxifen/kg body wt and were killed 3, 6, 24, 48, 72 and 120 h after treatment. In another series, animals were dosed orally on the morning of Day 2 with 0-05,0-1, 0-2, or 0-4 mg [3H]tamoxifen/kg and killed 24 h later. Blood was collected either by heart puncture under ether anaesthesia or immediately after decapitation and was centrifuged to obtain plasma. Before removing any tissue the animal was perfused in situ through the inferior vena cava with about 30 ml 0-9 % (w/v) NaCl. Duplicate samples of tissues or single samples of whole organs such as the ovary, oviduct and pituitary gland were rapidly excised, rinsed in 0-9 % (w/v) NaCl, gently blotted on filter paper and weighed before dissolving in NCS tissue solubilizer for counting. To examine subcellular distribution of tamoxifen and oestradiol, rats in groups of three were injected i.v. on the morning of Day 2 of pregnancy with 0-2 mg [3H]tamoxifen/kg body Retention and subcellular distribution
or with 0-1 µg 360 min later.
wt
[2,4,6,7-3H]oestradiol-17/?/rat and
were
killed at intervals between 5 and
Fractionation of uteri free from fat and trimmed were mesentery, slit longitudinally, and minced finely with was pooled and homogenized in 10 vol. (w/v) icefrom three animals the tissue scissors, cold 0-32M-STKMC buffer (0-32 M-sucrose; 0-05 M-Tris; 0-025 M-KC1; 0-005 M-MgCl2;
Uteri
0-0033 M-CaCl2), pH 7-5, using an Ultra Turrax TP 18/2 disintegrator for 1-5 min at 60 V. After removing samples for the determination of total radioactivity, the remainder was filtered through 100 mesh nylon bolting cloth. The residue was washed with 10 vol. 0-32 mSTKMC buffer, rehomogenized in fresh buffer and again filtered. The combined filtrates were centrifuged at 600-700 g for 10 min to obtain the crude nuclear pellet. This was resuspended in 0-25 m-STKMC buffer, underlayered with 10 ml 0-32 m-STKMC buffer and centrifuged as above. The washed nuclear pellet was taken up in 2-4 m-STM buffer (2-4 msucrose; 0-05 M-Tris; 0-001 M-MgCl2), pH 7-5, and centrifuged at 78000 g for 1 h to yield a purified nuclear pellet. In some experiments, the crude nuclear pellet was suspended in 1-0 M-STM buffer, pH 7-5, containing 0-25 % (w/v) Triton X-100 and centrifuged at 2200 g for 10 min. The pellet was then purified as above by resuspending in 2-4 M-STM buffer
followed by centrifugation. The initial supernatant was centrifuged at 105000 g for 45 min to produce mitochondrial and microsomal pellet and a particle-free cytosol fraction. Nuclear
a
combined
[3H]oestradiol exchange
Female rats on Day 2 of pregnancy were bilaterally ovariectomized under tribromoethanol (Avertin) anaesthesia (Bayer Products Company, Surbiton-Upon-Thames) and were used 7 days later. The animals were treated by subcutaneous injection with either oestradiol or tamoxifen at 5, 100 or 200/ig/kg body weight, or received vehicle alone (0-5 % (w/v) Tween 80 in 0-9 % (w/v) saline containing 1 % (w/v) absolute ethanol). Three hours later, the animals were decapitated and uterine nuclei were prepared essentially as described in the previous section. Portions of each nuclear suspension in 0-25 m-SHM buffer (0-25 M-sucrose; 10 mMHEPES; 1 mM-MgCl2), pH 7-5, were dispensed into two series of tubes, A and B, such that each series comprised an identical range of nuclear suspensions. The nuclei were sedimented by centrifugation at 600-700 g for 10 min and the supernatants were discarded. To each drained pellet in series A was added 1-0 ml 0-25 M-SHM containing 10-8 m-[2,4,6,7-3H]oestradiol, and each pellet in series received 1-0 ml of the same buffer containing 10~8M-[2,4,6,7-3H]oestradiol and 10_e M-diethylstilboestrol. The pellets were resuspended and were incubated for 60 min at 37 °C. The nuclei were then sedimented by centrifugation and washed four times with 2-0 ml portions of ice-cold 0-25 M-SHM buffer. Nuclear-bound [3H]oestradiol was extracted with ethanol and the DNA content of residues determined by the procedure described by Ceriotti (1952). Tubes in series A were used to determine the total amount of [3H]oestradiol exchanged. Series contained the same concentrations of [3H]oestradiol as in A plus a 100-fold excess of diethylstilboestrol and was used to determine the specific [3H]oestradiol exchange. The quantity of [3H]oestradiol bound to nuclei in the presence of the 100-fold excess of stilboestrol was subtracted from the total nuclear binding with [3H]oestradiol alone to yield the quantity of specifically bound [3H]oestradiol.
Uptake of[3H]oestradiol into
the rat uterus following treatment with tamoxifen Mated rats were given 0-2 or 0-1 mg tamoxifen/kg body weight orally between 09.00 and 10.00 h on Day 2, the controls receiving vehicle alone. Groups of three animals were injected i.v. 24, 48,72 or 96 h later with 5 ng [2,4,6,7-3H]oestradiol-17^ (100 Ci/mmol)/100 g body wt. One hour later, the animals in each group were decapitated and the pooled uterine tissue was fractionated essentially as described in the preceding section. In this experiment, how¬ ever, the nuclei were purified by layering the crude suspension in 0-25 m-STM buffer con¬ taining 0-5 % (w/v) Triton X-100 over a gradient prepared 15-18 h earlier by successively
portions of 2-2 m-STM, 2-0 m-STM, and 1 -6 m-STM buffer. The purified pelleted by centrifugation at 32000 g for 45 min at 4 °C. While this procedure provided nuclei of satisfactory purity, Ihe method did not produce any obvious improve¬ ment in either the quality or the yield of the nuclei when compared with the method used previously. Analysis of nuclear oestradiol receptors Nuclear fractions, prepared from three or four animals, were suspended in 3-0 ml 0-8 mKC1 in 10 mM-Tris/HCl buffer, pH 8-5, containing 1-5 mM-EDTA and after 60 min on ice the suspension was centrifuged to yield the KC1 supernatant. Portions (300 µ\) of the extracts were analysed on 5-20 % linear sucrose gradients (Martin & Ames, 1961) prepared in 0-8 mKCl/Tris/EDTA buffer. After centrifuging for 18 h at 100000 g the gradients were frac¬ tionated by upward displacement with dense sucrose solution, fractions diluted to 1 -0 ml layering nuclei
10 ml
were
with water, mixed with 10 ml butyl PBD-Triton X scintillation fluid and counted. Separate gradients were loaded with ovalbumin (3-6 S) and y-globulin (7-0 S) [Sigma Chemical Co.] and after centrifugation and fractionation, protein was determined by the method of Lowry, Rosebrough, Farr & Randall (1951). Measurement
of radioactivity Samples of tissue homogenate and plasma samples were diluted to 0-8 ml with water and 0-2 ml of 72 % (w/v) perchloric acid was added to each. The mixtures were incubated over¬ night at 40 °C, mixed thoroughly, cooled, and taken up in 10 ml butyl PBD-Triton X scintillation fluid (12 g butyl-PBD; 1000 ml toluene; 500 ml Triton X-100). All other tissue suspensions and cytosol fractions were counted directly in butyl PBD-Triton X solution or were first digested at 40 °C with 1 -0 ml NCS tissue solubilizer (Amersham/Searle Corpora¬ tion) mixed with 0-03 ml glacial acetic acid, 2-0 ml toluene and 12-0 ml PPO toluene scintillator (7-5 g 2-5 diphenyl oxazole/litre toluene). Ethanol extracts were counted in 12 ml butyl PBD-toluene scintillation fluid (240 g naphthalene; 24 g butyl PBD; 1500 ml toluene; 900 ml 2-methoxyethanol). Radioactivity was determined in a Packard Tri-Carb 3320 liquid scintillation spectrometer and quench correction was applied by internal standardization using [3H]hexadecane. RESULTS
The time course of retention of radioactivity in various tissues following the oral administra¬ tion of 0-2 mg [3H]tamoxifen/kg body wt on Day 2 of pregnancy is presented in Fig. 1. In the reproductive tissues significant radioactivity was detected within 3 h, and rose to a maximum in 6 h remaining significantly (P < 0-001) higher for at least 48 h when compared with diaphragm and cerebral cortex. Seventy-two hours after tamoxifen administration there was no difference between reproductive and other organs. Determination of the proportion of radioactivity present as tamoxifen in the uterus at different times after oral administration of 0-4 mg/kg showed that 45-50% was present unchanged after 6 h, 25-29 % after 24 h, 15-20 % after 48 h and less than 15 % after 72 h. These percentages are equivalent to 140, 80, 26 and 10 fmol tamoxifen/mg wet tissue at 6, 24, 48 and 72 h respectively. Subcellular distribution
injection [3H]tamoxifen (0-2 mg/kg body weight) was followed by rapid in¬ into the uterus to reach a maximum wilhin 30 min (Fig. 2d). Some 40-45 % of corporation the radioactivity was in the high speed pellet, 10-15 % was in the nuclear fraction and a further 15-20 % was in the cytosol fraction (Fig. 2b). In a second experiment in which ani¬ mals were killed 24 h after receiving an oral dose of 0-2 mg tamoxifen/kg, the distribution of Intravenous
of
radioactivity was approximately 50 % in the high speed pellet, 20 % in the nuclear fraction and 30 % in the cytosol, indicating that the distribution was not simply dependent upon the
of administration. After i.v. injection of 0-1 µg [3H]oestradiol-17/?, uterine tissue levels (Fig. 3a) reached a maximum in 60 min but the distribution of ladioactivity was quite different from that of tamoxifen (Fig. 3b). Thus 41-57 % was associated with the nuclear fraction, 10-16 % with the high speed pellet and 37-43 % with the cytosol fraction. No previous data appear to exist on the distribution of oestradiol in the adult pregnant rat but the figures above com¬ pare well with the distribution of labelled oestradiol in subcellular fractions from the uteri of immature rats (Gorski, Toft, Shyamala, Smith & Notides, 1968; Jensen & DeSombre, route
1972).
36
12
48 Time after
24
72
injection (h) course of distribution of radioactivity after a single oral dose of [3H]tamoxifen to Fig. 1. Time pregnant rats on Day 2 of pregnancy. Animals received 0-2 mg/kg body wt and tissues were processed as described in the Materials and Methods section. , Plasma; T, liver; ·, ovary; pituitary; A, uterus; V, cerebral cortex and diaphragm. .
The differences found between the distribution of [3H]tamoxifen and [3H]oestradiol are simply due to the much larger doses of tamoxifen administered. Thus, in a group of rats injected with 0-2 mg [3H]oestradiol/kg body weight (adjusted with carrier oestradiol to a specific activity of 88 mCi/mmol similar to that of tamoxifen) and killed 1 h later, 82 % of the radioactivity was present in the cylosol fraction, 5 % in the nuclear fraction and a further 5 % was found in the high speed pellet. Although this distribution indicated that high doses of either tamoxifen or oestradiol could be saturating the nuclear oestrogen receptors, the low amount of oestradiol associated with the high-speed pellet was still distinctly different from the distribution obtained with tamoxifen. not
-
Nuclear [3H]oestradiol exchange It has been shown (Clark, Anderson & Peck, 1972) that [3H]oestradiol will exchange with oestrogenic material bound in target tissue nuclei in vitro. Furthermore, the exchange tech¬ nique has been used to investigate nuclear accumulation of antioestrogen-receptor complex
& Peck, 1973). In the present studies, subcutaneous injection of 1, 20 or 40 //g tamoxifen into ovariectomized rats resulted in the appearance of intranuclear material which could undergo exchange with [3H]oestradiol, thus indicating accumulation of antioestrogen-receptor complex in the nuclear fraction (Fig. 4). An analogous pattern of accumulation was observed after injection of a similar dose range of oestradiol.
(Clark, Anderson
(ß) 10 _,
9
-
-
o
g* S. . * G ^ «
>>
=>
£ 1
' '
'
0-5
1
(b) c
? Öß
li
O o ta J3
O
_
«
o
50 40
30 20
10 05
J_ 4
6 5 Time after injection (h) Fig. 2. Uptake and distribution of [3H]tamoxifen in rat uterus after an i.v. injection of 0-2 mg/kg body weight on the morning of Day 2 of pregnancy, (a) ·, d.p.m./100 µ% DNA; A, d.p.m./10 mg wet tissue; , d.p.m./ µ plasma, (b) , high speed pellet; C*, cytosol fraction; T, nuclear 1
2
3
fraction.
Oestradiol uptake after treatment with tamoxifen Fractionation of the uteri of animals treated with tamoxifen on Day 2 and given [3H]oestradiol 60 min before death on Days 4, 5 or 6 showed little major difference in the total uptake of radioactivity. Thus mean values for two groups of animals (3 rats/group) for controls and animals receiving 0-1 mg and 0-2 mg tamoxifen/kg body wt respectively were: Day 4, 200, 160 and 180 d.p.m./mg wet tissue; Day 5, 240, 195 and 205 d.p.m./mg wet tissue; Day 6, 180, 200 and 160 d.p.m./mg wet tissue. There was also little, if any, difference in the proportions of total radioactivity found in the subcellular fractions (Fig. 5). In animals similarly treated with 0-2 mg tamoxifen/kg body wt and killed on Day 3, oestradiol uptake was depressed below that of the untreated controls (controls
10
9
h
8 7
6 5 4
h- 1
o
3 2 1
h 0-5
1
(A) 70 60 —
—
50
.2 •5
40
30 20 10
i_L 0-5
_L 1
JL
2 3 4 5 6 Time after injection (h) Fig. 3. Uptake and distribution of [3H]oestradiol in rat uterus after an i.v. injection of 01 /tg on the morning of Day 2 of pregnancy, (a) ·, d.p.m./10/íg DNA; , d.p.m./mg wet tissue, , d.p.m./ 10/il plasma, (b) , high speed pellet; O, cytosol fraction; , nuclear fraction.
0
IS 3
o
o e
fë o
S
0
2
4
6
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Oestradiol (/¿g) or tamoxifen (/¿g)
Fig. 4. Uterine nuclear [3H]oestradiol exchange in vitro after treatment of ovariectomized rats for 3 h in vivo with oestradiol or tamoxifen. ·, Rats injected s.c. with tamoxifen; O, rats injected s.c. with oestradiol.
60
p
50 «d
0
ai
G
40
-
30 O
i S
20
-
I
10
oöö
y
i
— -
—
~ ~
Day 4 Dose
Day 5 of tamoxifen (mg/kg body wt)
Day 6
Fig. 5. The subcellular distribution of radioactivity in rat uterus 60 min after i.v. administration of [3H]oestradiol (5 ng/100 g body wt) to animals receiving various oral doses of tamoxifen on Day 2 of pregnancy. , Nuclear fraction; 0, cytosol; HI, high speed pellet.
(b) 3-6 S
70 S
35 Fraction number
Fig. 6. Sedimentation patterns of 0-8 M-KC1 extracts of uterine nuclei from rats on Day 3 of pregnancy 60 min after receiving 5 ng [3H]oestradioI/100 g body wt i.v. Comparisons are between extracts from identical quantities of nuclei as determined by DNA content, (a) Control animals; (b) animals treated with 0-2 mg tamoxifen/kg body wt on Day 2 of pregnancy (09.00-10.00 h). The 3-6S, and 70S markers were ovalbumin and bovine-y-globulin respectively. O, Protein (Lowry et al. 1951); ·, radioactivity.
222+ 10 (s.E.M.) d.p.m./mg wet tissue (3 animals); tamoxifen treated, 145+ 12 d.p.m./mg wet tissue (3 animals)). The decrease in oestradiol uptake was examined further in the nuclear fraction. Nuclei were extracted with 0-8 m-KCI and the extract was subjected to sucrose density gradient analysis (Fig. 6). It will be seen that although the total level of [3H]oestradiol taken up by the nuclei was decreased slightly after treatment of the animals with tamoxifen, the pattern of distribution of radioactivity was identical with that of the un¬ treated controls, suggesting that tamoxifen was not eliminating the binding of oestradiol to the nuclear oestrogen receptors. DISCUSSION
Following the oral administration of [3H]tamoxifen to pregnant rats at a level of 0-4 mg/kg, radioactivity was found to be retained by reproductive tissues with the ovary showing the greatest retention. Analysis showed that a declining proportion of radioactivity in the uterus was present as unchanged tamoxifen. Thus within 6 h only 50 % of the total radioactivity was present as tamoxifen, declining to 25-29 % and 15-20 % in 24 and 48 h respectively. These proportions correspond to 140, 80 and 26 fmol tamoxifen/mg wet tissue. These quantities may be compared with values of 16, 35, 51 and 60 fmol oestradiol/mg wet wt reported by Mester, Martel, Psychoyos & Baulieu (1974) for the oestradiol binding capacity of rat uterine cytosol on Days 2-5 of pregnancy respectively. Tamoxifen can inhibit implantation completely if given to rats of our colony at 0-2 mg/ kg body wt on Day 2 of pregnancy only, and it has been previously considered to exert this action by inhibiting the uptake of oestradiol necessary for implantation or, if given at this early stage of pregnancy, by accelerating tubai transport of the fertilized eggs followed by their expulsion from the uterus (Harper & Walpole, 1967). In our colony, however in¬ creased tubai transport of fertilized eggs does not appear to be involved in the antifertility properties of tamoxifen when administered at the physiologically effective dose of 0-2 mg/kg on Day 2 of pregnancy (Major & Heald, 1974). In the present work, except for the period up to 24 h after administration, no convincing evidence was obtained that the uptake of oestradiol was suppressed either in the whole uterus, or in the subcellular fractions. On the other hand tamoxifen promoted the transfer of uterine cytosol oestradiol receptors to the nucleus to the same extent as an equal dose of oestradiol. At first sight this is a surprising finding for a compound which, in vitro, has 1/1000 the affinity of oestradiol for uterine cytosol receptors (Skidmore et al. 1972). How¬ ever it is now apparent that such measurements in vitro are by themselves insufficient to explain the mechanisms of antioestrogen action in vivo. Thus it has been shown that a single injection of nafoxidine in the immature rat causes protracted depletion of oestradiol recep¬ tors within the uterine cytosol, associated with their prolonged retention within the nuclei (Clark et al. 1973). Extension of this work by Katzenellenbogen & Ferguson (1975) revealed that in the immature rat, following treatment with nafoxidine or CI-628, oestradiol adminis¬ tered in vivo was still taken up and bound specifically in the uterine nuclei to the oestrogen receptors but was in these circumstances unable to promote uterine growth or the synthesis of induced protein. It is therefore possible that in the adult rat, oestradiol which binds within the uterine nuclei on Day 3 of pregnancy, 24 h after tamoxifen treatment, is also to a large extent ineffec¬ tive in promoting the changes essential to implantation. These considerations do not make it possible to state unequivocally that the antifertility action of tamoxifen does not include an action involving a suppression of the oestrogenic response of the uterus. On the other hand it is our belief that the balance of evidence indicates that the primary site of antifertility action, when given on Day 2 of pregnancy, is the ovary (Watson et al. 1975).
A CASE award to J.S.M. is gratefully acknowledged, from Dr A. L. Walpole, ICI Pharmaceuticals Ltd.
together with gifts of [3H]tamoxifen
REFERENCES
Anderson, J., Clark, J. H. & Peck, E. J., Jr. (1972). Oestrogen and nuclear binding sites. Determination of specific sites by [3H]oestradiol exchange. Biochemical Journal 126, 561-567. Ceriotti, G. (1952). Microchemical determination of desoxyribonucleic acid. Journal of Biological Chemistry 198, 297-303. Clark, J. H., Anderson, J. & Peck, E. J., Jr. (1972). Receptor-estrogen complex in the nuclear fraction of rat uterine cells during the estrous cycle. Science 176, 528-530. Clark, J. H., Anderson, J. N. & Peck, E. J. Jr. (1973). Estrogen receptor-antiestrogen complex: Atypical binding by uterine nuclei and effects on uterine growth. Steroids 22, 707-718. Fromson, J. M., Pearson, S. & Bramah, S. (1973). The metabolism of tamoxifen (I.C.I. 46,474). Part I. In laboratory animals. Xenobiotica 3, 623-709. Gorski, J., Toft, D., Shyamala, G., Smith, D. & Notides, A. (1968). Hormone receptors: Studies on the interaction of estrogen with the uterus. Recent Progress in Hormone Research 24, 45-80. Harper, M. J. K. & Walpole, A. L. (1967). Mode of action of I.C.I. 46,474 in preventing implantation in rats. Journal of Endocrinology 37, 83-92. Jensen, E. V. & DeSombre, E. R. (1972). Mechanism of action of the female sex hormones. Annual Review of Biochemistry 41, 203-230. Katzenellenbogen, B. S. & Ferguson, E. R. (1975). Antiestrogen action in the uterus: biological ineffective¬ ness of nuclear bound estradiol after antiestrogen. Endocrinology 97, 1-21. Lowry, O. H., Rosebrough, O. H., Farr, A. L. & Randall, Rose J. (1951). Protein measurement with the folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Major, J. S. & Heald, P. J. (1974). The effect of I.C.I. 46,474 on ovum transport and implantation in the rat. Journal of Reproduction and Fertility 36, 117-124. Martin, R. G. & Ames, B. N. (1961). A method for determining the sedimentation behaviour of enzymes: Application to protein mixtures. Journal of Biological Chemistry 236, 1372-1379. Mester, L, Martel, D., Psychoyos, A. & Baulieu, E. E. (1974). Hormonal control of oestrogen receptor in uterus and receptivity for ova implantation in the rat. Nature 250, 776-778. Skidmore, J., Walpole, A. L. & Woodburn, J. (1972). Effect of some triphenylethylenes on oestradiol binding in vitro to macromolecules from uterus and anterior pituitary. Journal of Endocrinology 52, 289-298. Watson, J. & Alam, M. (1976). Oestrogen synthesis during delayed implantation in the rat. Contraception 13, 101-107. Watson, J., Anderson, F. B., Alam, M., O'Grady, J. E. & Heald, P. J. (1975). Plasma hormones and pituitary luteinizing hormone in the rat during the early stages of pregnancy and after post-coital treatment with Tamoxifen (ICI 46,474). Journal of Endocrinology 65, 7-17.
Department
P. J. HEALD of Biochemistry, University of Strathclyde, Biosciences 31 Taylor Street, Glasgow, G4 ONR AND
Building,
(Received 22 January 1976) SUMMARY
uptake
and retention of
a radioactive post-coital antifertility agent have shown that the ovary retained more radioactivity than did any other reproductive organ. Studies have also been made of the uptake and distribution of [3H]tamoxifen and [3H]oestradiol-17\g=b\ in the uterus of the pregnant rat on days 2\p=n-\6post coitum. Twenty-four hours after administration of tamoxifen, either i.v. or orally, 40\p=n-\50% of the radioactivity was in the high speed pellet, 10\p=n-\20% in the nuclear fraction, and 15\p=n-\30% in the cytosol. An equivalent dose of [3H]oestradiol-17\g=b\ yielded distributions of 5 %, 5 % and 82 % respectively. Fractionation of uteri from animals given 0\m=.\2mg tamoxifen/kg on Day 2 of pregnancy followed by [3H]oestradiol 60 min before death showed little difference in total uptake of oestradiol or distribution in the subcellular fraction on Days 4, 5 and 6. Although uptake of oestradiol by uterine nuclei was reduced on Day 3 by previous administration of tamoxifen on Day 2, appreciable quantities were still bound to the nuclear receptors. Treatment of ovariectomized animals with tamoxifen at doses up to 40 \g=m\g/rat (i.e. 0\m=.\2mg/kg) led to the accumulation of oestrogen\p=n-\receptorcomplex in the nucleus. It is concluded that the antifertility properties of tamoxifen (under the conditions of these experiments) cannot be ascribed to the suppression of uptake and binding of oestradiol by the uterus.
Measurement of the
tamoxifen, by reproductive tissues of the
rat
INTRODUCTION
post-coital antifertility agent tamoxifen [ICI 46,474: trans-l-(p-/?-dimethylaminoethoxyphenyl)-l,2-diphenylbut-l-ene], have shown that in appropriate doses it can delay implantation by up to 25 h in the pregnant rat (Major & Heald, 1974). Examination of this phenomenon suggested that tamoxifen delays or eliminates the rise in nidatory plasma oestradiol primarily by inhibiting its synthesis by the ovary (Watson, Anderson, Alam, O'Grady & Heald, 1975; Watson & Alam, 1976). Tamoxifen is antioestrogenic in the rat (Harper & Walpole, 1967) and since it combines with uterine cytoplasmic oestrogen receptors in vitro (Skidmore, Walpole & Woodburn, 1972), it has been suggested that one alternative means by which it could exert its antifertility effect is by suppression of the uterine uptake of nidatory oestrogen. To examine this situation further the time course of uptake and retention of tamoxifen by pregnant rat uterus was measured and also the uptake of oestradiol-17/? in vivo by uteri of rats receiving tamoxifen on Day 2 Studies
on
the mechanism of action of the
of pregnancy. *
Present address: ICI Pharmaceuticals
Ltd, Alderley Park, Alderley Edge, Cheshire.
The subcellular distribution of tamoxifen was also examined and compared with that of oestradiol-17/?. Evidence for intra-nuclear binding of tamoxifen was obtained by utilizing the [3H]oestradiol exchange technique developed by Anderson, Clark & Peck (1972). MATERIALS AND METHODS
The rats used, conditions of mating and administration of tamoxifen previously (Major & Heald, 1974; Watson et al. 1975).
were as
described
Chromatographie identification of tamoxifen in uterine and liver extracts All glassware was treated with 5 % (v/v) dimethyldichlorosilane (BDH Chemicals Ltd, Poole, Dorset) in toluene to reduce adsorption losses. Animals were dosed orally on Day 2 with 0-4 mg [3H]tamoxifen/kg and killed 6-72 h later. The perfused liver was homogenized in 20 ml methanol containing non-radioactive tamoxifen as carrier (50 /¿g/ml), and after centrifuging and twice re-extracting the residue with 10 vol. methanol/carrier solution the combined extracts were evaporated to dryness under nitrogen at 40 °C. A similar procedure was used for uterine tissue except that 20 vol. methanol were employed in the re-extractions. The dried extracts were taken up in the smallest volume of methanol required to produce a clear solution and aliquots were separated on Kieselgel HF254 plates using benzene-triethylamine (9:1, v/v) (Fromson, Pearson & Bramah, 1973). After development tamoxifen was detected by u.v. light, as little as 250 ng producing a recognizable fluorescence. Areas of the chromatograms corresponding to a tamoxifen marker were scraped into counting vials con¬ taining 1-0 ml H20, 10 ml butyl PBD Triton X scintillator were added, and the samples counted.
[3H]Tamoxifen [sp. act. 240 /¿Ci/mg (89-2 mCi/mmol)] was supplied by Dr A. L. Walpole was at least 98 % pure as determined by t.l.c. [2,4,6,7-3H]Oestradiol-17/ff (45 Ci/mmol) was obtained from the Radiochemical Centre, Amersham.
(ICI Pharmaceuticals Ltd) and
of [3H]tamoxifen and [3H]oestradiol-17ß To study the retention of tamoxifen in the uterus and other tissues, rats in groups of three were dosed orally between 09.00 and 10.00 h on Day 2 of pregnancy with 0-2 mg [3H]tamoxifen/kg body wt and were killed 3, 6, 24, 48, 72 and 120 h after treatment. In another series, animals were dosed orally on the morning of Day 2 with 0-05,0-1, 0-2, or 0-4 mg [3H]tamoxifen/kg and killed 24 h later. Blood was collected either by heart puncture under ether anaesthesia or immediately after decapitation and was centrifuged to obtain plasma. Before removing any tissue the animal was perfused in situ through the inferior vena cava with about 30 ml 0-9 % (w/v) NaCl. Duplicate samples of tissues or single samples of whole organs such as the ovary, oviduct and pituitary gland were rapidly excised, rinsed in 0-9 % (w/v) NaCl, gently blotted on filter paper and weighed before dissolving in NCS tissue solubilizer for counting. To examine subcellular distribution of tamoxifen and oestradiol, rats in groups of three were injected i.v. on the morning of Day 2 of pregnancy with 0-2 mg [3H]tamoxifen/kg body Retention and subcellular distribution
or with 0-1 µg 360 min later.
wt
[2,4,6,7-3H]oestradiol-17/?/rat and
were
killed at intervals between 5 and
Fractionation of uteri free from fat and trimmed were mesentery, slit longitudinally, and minced finely with was pooled and homogenized in 10 vol. (w/v) icefrom three animals the tissue scissors, cold 0-32M-STKMC buffer (0-32 M-sucrose; 0-05 M-Tris; 0-025 M-KC1; 0-005 M-MgCl2;
Uteri
0-0033 M-CaCl2), pH 7-5, using an Ultra Turrax TP 18/2 disintegrator for 1-5 min at 60 V. After removing samples for the determination of total radioactivity, the remainder was filtered through 100 mesh nylon bolting cloth. The residue was washed with 10 vol. 0-32 mSTKMC buffer, rehomogenized in fresh buffer and again filtered. The combined filtrates were centrifuged at 600-700 g for 10 min to obtain the crude nuclear pellet. This was resuspended in 0-25 m-STKMC buffer, underlayered with 10 ml 0-32 m-STKMC buffer and centrifuged as above. The washed nuclear pellet was taken up in 2-4 m-STM buffer (2-4 msucrose; 0-05 M-Tris; 0-001 M-MgCl2), pH 7-5, and centrifuged at 78000 g for 1 h to yield a purified nuclear pellet. In some experiments, the crude nuclear pellet was suspended in 1-0 M-STM buffer, pH 7-5, containing 0-25 % (w/v) Triton X-100 and centrifuged at 2200 g for 10 min. The pellet was then purified as above by resuspending in 2-4 M-STM buffer
followed by centrifugation. The initial supernatant was centrifuged at 105000 g for 45 min to produce mitochondrial and microsomal pellet and a particle-free cytosol fraction. Nuclear
a
combined
[3H]oestradiol exchange
Female rats on Day 2 of pregnancy were bilaterally ovariectomized under tribromoethanol (Avertin) anaesthesia (Bayer Products Company, Surbiton-Upon-Thames) and were used 7 days later. The animals were treated by subcutaneous injection with either oestradiol or tamoxifen at 5, 100 or 200/ig/kg body weight, or received vehicle alone (0-5 % (w/v) Tween 80 in 0-9 % (w/v) saline containing 1 % (w/v) absolute ethanol). Three hours later, the animals were decapitated and uterine nuclei were prepared essentially as described in the previous section. Portions of each nuclear suspension in 0-25 m-SHM buffer (0-25 M-sucrose; 10 mMHEPES; 1 mM-MgCl2), pH 7-5, were dispensed into two series of tubes, A and B, such that each series comprised an identical range of nuclear suspensions. The nuclei were sedimented by centrifugation at 600-700 g for 10 min and the supernatants were discarded. To each drained pellet in series A was added 1-0 ml 0-25 M-SHM containing 10-8 m-[2,4,6,7-3H]oestradiol, and each pellet in series received 1-0 ml of the same buffer containing 10~8M-[2,4,6,7-3H]oestradiol and 10_e M-diethylstilboestrol. The pellets were resuspended and were incubated for 60 min at 37 °C. The nuclei were then sedimented by centrifugation and washed four times with 2-0 ml portions of ice-cold 0-25 M-SHM buffer. Nuclear-bound [3H]oestradiol was extracted with ethanol and the DNA content of residues determined by the procedure described by Ceriotti (1952). Tubes in series A were used to determine the total amount of [3H]oestradiol exchanged. Series contained the same concentrations of [3H]oestradiol as in A plus a 100-fold excess of diethylstilboestrol and was used to determine the specific [3H]oestradiol exchange. The quantity of [3H]oestradiol bound to nuclei in the presence of the 100-fold excess of stilboestrol was subtracted from the total nuclear binding with [3H]oestradiol alone to yield the quantity of specifically bound [3H]oestradiol.
Uptake of[3H]oestradiol into
the rat uterus following treatment with tamoxifen Mated rats were given 0-2 or 0-1 mg tamoxifen/kg body weight orally between 09.00 and 10.00 h on Day 2, the controls receiving vehicle alone. Groups of three animals were injected i.v. 24, 48,72 or 96 h later with 5 ng [2,4,6,7-3H]oestradiol-17^ (100 Ci/mmol)/100 g body wt. One hour later, the animals in each group were decapitated and the pooled uterine tissue was fractionated essentially as described in the preceding section. In this experiment, how¬ ever, the nuclei were purified by layering the crude suspension in 0-25 m-STM buffer con¬ taining 0-5 % (w/v) Triton X-100 over a gradient prepared 15-18 h earlier by successively
portions of 2-2 m-STM, 2-0 m-STM, and 1 -6 m-STM buffer. The purified pelleted by centrifugation at 32000 g for 45 min at 4 °C. While this procedure provided nuclei of satisfactory purity, Ihe method did not produce any obvious improve¬ ment in either the quality or the yield of the nuclei when compared with the method used previously. Analysis of nuclear oestradiol receptors Nuclear fractions, prepared from three or four animals, were suspended in 3-0 ml 0-8 mKC1 in 10 mM-Tris/HCl buffer, pH 8-5, containing 1-5 mM-EDTA and after 60 min on ice the suspension was centrifuged to yield the KC1 supernatant. Portions (300 µ\) of the extracts were analysed on 5-20 % linear sucrose gradients (Martin & Ames, 1961) prepared in 0-8 mKCl/Tris/EDTA buffer. After centrifuging for 18 h at 100000 g the gradients were frac¬ tionated by upward displacement with dense sucrose solution, fractions diluted to 1 -0 ml layering nuclei
10 ml
were
with water, mixed with 10 ml butyl PBD-Triton X scintillation fluid and counted. Separate gradients were loaded with ovalbumin (3-6 S) and y-globulin (7-0 S) [Sigma Chemical Co.] and after centrifugation and fractionation, protein was determined by the method of Lowry, Rosebrough, Farr & Randall (1951). Measurement
of radioactivity Samples of tissue homogenate and plasma samples were diluted to 0-8 ml with water and 0-2 ml of 72 % (w/v) perchloric acid was added to each. The mixtures were incubated over¬ night at 40 °C, mixed thoroughly, cooled, and taken up in 10 ml butyl PBD-Triton X scintillation fluid (12 g butyl-PBD; 1000 ml toluene; 500 ml Triton X-100). All other tissue suspensions and cytosol fractions were counted directly in butyl PBD-Triton X solution or were first digested at 40 °C with 1 -0 ml NCS tissue solubilizer (Amersham/Searle Corpora¬ tion) mixed with 0-03 ml glacial acetic acid, 2-0 ml toluene and 12-0 ml PPO toluene scintillator (7-5 g 2-5 diphenyl oxazole/litre toluene). Ethanol extracts were counted in 12 ml butyl PBD-toluene scintillation fluid (240 g naphthalene; 24 g butyl PBD; 1500 ml toluene; 900 ml 2-methoxyethanol). Radioactivity was determined in a Packard Tri-Carb 3320 liquid scintillation spectrometer and quench correction was applied by internal standardization using [3H]hexadecane. RESULTS
The time course of retention of radioactivity in various tissues following the oral administra¬ tion of 0-2 mg [3H]tamoxifen/kg body wt on Day 2 of pregnancy is presented in Fig. 1. In the reproductive tissues significant radioactivity was detected within 3 h, and rose to a maximum in 6 h remaining significantly (P < 0-001) higher for at least 48 h when compared with diaphragm and cerebral cortex. Seventy-two hours after tamoxifen administration there was no difference between reproductive and other organs. Determination of the proportion of radioactivity present as tamoxifen in the uterus at different times after oral administration of 0-4 mg/kg showed that 45-50% was present unchanged after 6 h, 25-29 % after 24 h, 15-20 % after 48 h and less than 15 % after 72 h. These percentages are equivalent to 140, 80, 26 and 10 fmol tamoxifen/mg wet tissue at 6, 24, 48 and 72 h respectively. Subcellular distribution
injection [3H]tamoxifen (0-2 mg/kg body weight) was followed by rapid in¬ into the uterus to reach a maximum wilhin 30 min (Fig. 2d). Some 40-45 % of corporation the radioactivity was in the high speed pellet, 10-15 % was in the nuclear fraction and a further 15-20 % was in the cytosol fraction (Fig. 2b). In a second experiment in which ani¬ mals were killed 24 h after receiving an oral dose of 0-2 mg tamoxifen/kg, the distribution of Intravenous
of
radioactivity was approximately 50 % in the high speed pellet, 20 % in the nuclear fraction and 30 % in the cytosol, indicating that the distribution was not simply dependent upon the
of administration. After i.v. injection of 0-1 µg [3H]oestradiol-17/?, uterine tissue levels (Fig. 3a) reached a maximum in 60 min but the distribution of ladioactivity was quite different from that of tamoxifen (Fig. 3b). Thus 41-57 % was associated with the nuclear fraction, 10-16 % with the high speed pellet and 37-43 % with the cytosol fraction. No previous data appear to exist on the distribution of oestradiol in the adult pregnant rat but the figures above com¬ pare well with the distribution of labelled oestradiol in subcellular fractions from the uteri of immature rats (Gorski, Toft, Shyamala, Smith & Notides, 1968; Jensen & DeSombre, route
1972).
36
12
48 Time after
24
72
injection (h) course of distribution of radioactivity after a single oral dose of [3H]tamoxifen to Fig. 1. Time pregnant rats on Day 2 of pregnancy. Animals received 0-2 mg/kg body wt and tissues were processed as described in the Materials and Methods section. , Plasma; T, liver; ·, ovary; pituitary; A, uterus; V, cerebral cortex and diaphragm. .
The differences found between the distribution of [3H]tamoxifen and [3H]oestradiol are simply due to the much larger doses of tamoxifen administered. Thus, in a group of rats injected with 0-2 mg [3H]oestradiol/kg body weight (adjusted with carrier oestradiol to a specific activity of 88 mCi/mmol similar to that of tamoxifen) and killed 1 h later, 82 % of the radioactivity was present in the cylosol fraction, 5 % in the nuclear fraction and a further 5 % was found in the high speed pellet. Although this distribution indicated that high doses of either tamoxifen or oestradiol could be saturating the nuclear oestrogen receptors, the low amount of oestradiol associated with the high-speed pellet was still distinctly different from the distribution obtained with tamoxifen. not
-
Nuclear [3H]oestradiol exchange It has been shown (Clark, Anderson & Peck, 1972) that [3H]oestradiol will exchange with oestrogenic material bound in target tissue nuclei in vitro. Furthermore, the exchange tech¬ nique has been used to investigate nuclear accumulation of antioestrogen-receptor complex
& Peck, 1973). In the present studies, subcutaneous injection of 1, 20 or 40 //g tamoxifen into ovariectomized rats resulted in the appearance of intranuclear material which could undergo exchange with [3H]oestradiol, thus indicating accumulation of antioestrogen-receptor complex in the nuclear fraction (Fig. 4). An analogous pattern of accumulation was observed after injection of a similar dose range of oestradiol.
(Clark, Anderson
(ß) 10 _,
9
-
-
o
g* S. . * G ^ «
>>
=>
£ 1
' '
'
0-5
1
(b) c
? Öß
li
O o ta J3
O
_
«
o
50 40
30 20
10 05
J_ 4
6 5 Time after injection (h) Fig. 2. Uptake and distribution of [3H]tamoxifen in rat uterus after an i.v. injection of 0-2 mg/kg body weight on the morning of Day 2 of pregnancy, (a) ·, d.p.m./100 µ% DNA; A, d.p.m./10 mg wet tissue; , d.p.m./ µ plasma, (b) , high speed pellet; C*, cytosol fraction; T, nuclear 1
2
3
fraction.
Oestradiol uptake after treatment with tamoxifen Fractionation of the uteri of animals treated with tamoxifen on Day 2 and given [3H]oestradiol 60 min before death on Days 4, 5 or 6 showed little major difference in the total uptake of radioactivity. Thus mean values for two groups of animals (3 rats/group) for controls and animals receiving 0-1 mg and 0-2 mg tamoxifen/kg body wt respectively were: Day 4, 200, 160 and 180 d.p.m./mg wet tissue; Day 5, 240, 195 and 205 d.p.m./mg wet tissue; Day 6, 180, 200 and 160 d.p.m./mg wet tissue. There was also little, if any, difference in the proportions of total radioactivity found in the subcellular fractions (Fig. 5). In animals similarly treated with 0-2 mg tamoxifen/kg body wt and killed on Day 3, oestradiol uptake was depressed below that of the untreated controls (controls
10
9
h
8 7
6 5 4
h- 1
o
3 2 1
h 0-5
1
(A) 70 60 —
—
50
.2 •5
40
30 20 10
i_L 0-5
_L 1
JL
2 3 4 5 6 Time after injection (h) Fig. 3. Uptake and distribution of [3H]oestradiol in rat uterus after an i.v. injection of 01 /tg on the morning of Day 2 of pregnancy, (a) ·, d.p.m./10/íg DNA; , d.p.m./mg wet tissue, , d.p.m./ 10/il plasma, (b) , high speed pellet; O, cytosol fraction; , nuclear fraction.
0
IS 3
o
o e
fë o
S
0
2
4
6
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Oestradiol (/¿g) or tamoxifen (/¿g)
Fig. 4. Uterine nuclear [3H]oestradiol exchange in vitro after treatment of ovariectomized rats for 3 h in vivo with oestradiol or tamoxifen. ·, Rats injected s.c. with tamoxifen; O, rats injected s.c. with oestradiol.
60
p
50 «d
0
ai
G
40
-
30 O
i S
20
-
I
10
oöö
y
i
— -
—
~ ~
Day 4 Dose
Day 5 of tamoxifen (mg/kg body wt)
Day 6
Fig. 5. The subcellular distribution of radioactivity in rat uterus 60 min after i.v. administration of [3H]oestradiol (5 ng/100 g body wt) to animals receiving various oral doses of tamoxifen on Day 2 of pregnancy. , Nuclear fraction; 0, cytosol; HI, high speed pellet.
(b) 3-6 S
70 S
35 Fraction number
Fig. 6. Sedimentation patterns of 0-8 M-KC1 extracts of uterine nuclei from rats on Day 3 of pregnancy 60 min after receiving 5 ng [3H]oestradioI/100 g body wt i.v. Comparisons are between extracts from identical quantities of nuclei as determined by DNA content, (a) Control animals; (b) animals treated with 0-2 mg tamoxifen/kg body wt on Day 2 of pregnancy (09.00-10.00 h). The 3-6S, and 70S markers were ovalbumin and bovine-y-globulin respectively. O, Protein (Lowry et al. 1951); ·, radioactivity.
222+ 10 (s.E.M.) d.p.m./mg wet tissue (3 animals); tamoxifen treated, 145+ 12 d.p.m./mg wet tissue (3 animals)). The decrease in oestradiol uptake was examined further in the nuclear fraction. Nuclei were extracted with 0-8 m-KCI and the extract was subjected to sucrose density gradient analysis (Fig. 6). It will be seen that although the total level of [3H]oestradiol taken up by the nuclei was decreased slightly after treatment of the animals with tamoxifen, the pattern of distribution of radioactivity was identical with that of the un¬ treated controls, suggesting that tamoxifen was not eliminating the binding of oestradiol to the nuclear oestrogen receptors. DISCUSSION
Following the oral administration of [3H]tamoxifen to pregnant rats at a level of 0-4 mg/kg, radioactivity was found to be retained by reproductive tissues with the ovary showing the greatest retention. Analysis showed that a declining proportion of radioactivity in the uterus was present as unchanged tamoxifen. Thus within 6 h only 50 % of the total radioactivity was present as tamoxifen, declining to 25-29 % and 15-20 % in 24 and 48 h respectively. These proportions correspond to 140, 80 and 26 fmol tamoxifen/mg wet tissue. These quantities may be compared with values of 16, 35, 51 and 60 fmol oestradiol/mg wet wt reported by Mester, Martel, Psychoyos & Baulieu (1974) for the oestradiol binding capacity of rat uterine cytosol on Days 2-5 of pregnancy respectively. Tamoxifen can inhibit implantation completely if given to rats of our colony at 0-2 mg/ kg body wt on Day 2 of pregnancy only, and it has been previously considered to exert this action by inhibiting the uptake of oestradiol necessary for implantation or, if given at this early stage of pregnancy, by accelerating tubai transport of the fertilized eggs followed by their expulsion from the uterus (Harper & Walpole, 1967). In our colony, however in¬ creased tubai transport of fertilized eggs does not appear to be involved in the antifertility properties of tamoxifen when administered at the physiologically effective dose of 0-2 mg/kg on Day 2 of pregnancy (Major & Heald, 1974). In the present work, except for the period up to 24 h after administration, no convincing evidence was obtained that the uptake of oestradiol was suppressed either in the whole uterus, or in the subcellular fractions. On the other hand tamoxifen promoted the transfer of uterine cytosol oestradiol receptors to the nucleus to the same extent as an equal dose of oestradiol. At first sight this is a surprising finding for a compound which, in vitro, has 1/1000 the affinity of oestradiol for uterine cytosol receptors (Skidmore et al. 1972). How¬ ever it is now apparent that such measurements in vitro are by themselves insufficient to explain the mechanisms of antioestrogen action in vivo. Thus it has been shown that a single injection of nafoxidine in the immature rat causes protracted depletion of oestradiol recep¬ tors within the uterine cytosol, associated with their prolonged retention within the nuclei (Clark et al. 1973). Extension of this work by Katzenellenbogen & Ferguson (1975) revealed that in the immature rat, following treatment with nafoxidine or CI-628, oestradiol adminis¬ tered in vivo was still taken up and bound specifically in the uterine nuclei to the oestrogen receptors but was in these circumstances unable to promote uterine growth or the synthesis of induced protein. It is therefore possible that in the adult rat, oestradiol which binds within the uterine nuclei on Day 3 of pregnancy, 24 h after tamoxifen treatment, is also to a large extent ineffec¬ tive in promoting the changes essential to implantation. These considerations do not make it possible to state unequivocally that the antifertility action of tamoxifen does not include an action involving a suppression of the oestrogenic response of the uterus. On the other hand it is our belief that the balance of evidence indicates that the primary site of antifertility action, when given on Day 2 of pregnancy, is the ovary (Watson et al. 1975).
A CASE award to J.S.M. is gratefully acknowledged, from Dr A. L. Walpole, ICI Pharmaceuticals Ltd.
together with gifts of [3H]tamoxifen
REFERENCES
Anderson, J., Clark, J. H. & Peck, E. J., Jr. (1972). Oestrogen and nuclear binding sites. Determination of specific sites by [3H]oestradiol exchange. Biochemical Journal 126, 561-567. Ceriotti, G. (1952). Microchemical determination of desoxyribonucleic acid. Journal of Biological Chemistry 198, 297-303. Clark, J. H., Anderson, J. & Peck, E. J., Jr. (1972). Receptor-estrogen complex in the nuclear fraction of rat uterine cells during the estrous cycle. Science 176, 528-530. Clark, J. H., Anderson, J. N. & Peck, E. J. Jr. (1973). Estrogen receptor-antiestrogen complex: Atypical binding by uterine nuclei and effects on uterine growth. Steroids 22, 707-718. Fromson, J. M., Pearson, S. & Bramah, S. (1973). The metabolism of tamoxifen (I.C.I. 46,474). Part I. In laboratory animals. Xenobiotica 3, 623-709. Gorski, J., Toft, D., Shyamala, G., Smith, D. & Notides, A. (1968). Hormone receptors: Studies on the interaction of estrogen with the uterus. Recent Progress in Hormone Research 24, 45-80. Harper, M. J. K. & Walpole, A. L. (1967). Mode of action of I.C.I. 46,474 in preventing implantation in rats. Journal of Endocrinology 37, 83-92. Jensen, E. V. & DeSombre, E. R. (1972). Mechanism of action of the female sex hormones. Annual Review of Biochemistry 41, 203-230. Katzenellenbogen, B. S. & Ferguson, E. R. (1975). Antiestrogen action in the uterus: biological ineffective¬ ness of nuclear bound estradiol after antiestrogen. Endocrinology 97, 1-21. Lowry, O. H., Rosebrough, O. H., Farr, A. L. & Randall, Rose J. (1951). Protein measurement with the folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Major, J. S. & Heald, P. J. (1974). The effect of I.C.I. 46,474 on ovum transport and implantation in the rat. Journal of Reproduction and Fertility 36, 117-124. Martin, R. G. & Ames, B. N. (1961). A method for determining the sedimentation behaviour of enzymes: Application to protein mixtures. Journal of Biological Chemistry 236, 1372-1379. Mester, L, Martel, D., Psychoyos, A. & Baulieu, E. E. (1974). Hormonal control of oestrogen receptor in uterus and receptivity for ova implantation in the rat. Nature 250, 776-778. Skidmore, J., Walpole, A. L. & Woodburn, J. (1972). Effect of some triphenylethylenes on oestradiol binding in vitro to macromolecules from uterus and anterior pituitary. Journal of Endocrinology 52, 289-298. Watson, J. & Alam, M. (1976). Oestrogen synthesis during delayed implantation in the rat. Contraception 13, 101-107. Watson, J., Anderson, F. B., Alam, M., O'Grady, J. E. & Heald, P. J. (1975). Plasma hormones and pituitary luteinizing hormone in the rat during the early stages of pregnancy and after post-coital treatment with Tamoxifen (ICI 46,474). Journal of Endocrinology 65, 7-17.
