Development of Responsiveness to Steroid Hormones: Advances in the Biosciences

Development of Responsiveness to Steroid Hormones is a collection of papers presented at the Bat-Sheva Seminar of the same name which was held in Rehovot, Israel, on October 18-26, 1978 and sponsored by the Bat-Sheva de Rothschild Foundation for the Advancement of Science in Israel in cooperation with The Weizmann Institute of Science. Contributors explore how the steroid receptor complex modulates transcription of RNA and cover topics ranging from the sequential acquisition of responsiveness to estrogen in the rat uterus to the ontogeny of steroid receptors in the guinea pig.

This book is comprised of 29 chapters and begins with a review of estrogen and estrogen effects in rat uterus and pituitary in culture. The embryonic chick Müllerian duct and several fetal guinea pig organs are then considered, followed by discussions on mammalian and chick liver, multi-hormonal control in rat liver; progesterone and decidualization; and rat mammary gland in culture. Glucocorticoids in developing pancreas, lung, and liver are examined, along with ecdysteroids in both locusts and Drosophila. Refractoriness is exemplified by gonadotropin action in the ovary. The remaining chapters deal with the role of steroid hormone receptors in brain development; the neural trigger for ovulation; aromatization and development of responsiveness of the brain to gonadal steroids; and neuroendocrine correlates of female-offspring interaction in maternal rats.

This monograph will be of value to physiologists, biologists, and biochemists.

• Language: English
• Publisher: Elsevier Science
• Release date: Oct 22, 2013
• ISBN: 9781483153087

Book preview

Committee

Sequential Acquisition of Responsiveness to Estrogen in the Rat Uterus

A.M. Kaye, N. Reiss and M.D. Walker,      Department of Hormone Research, The Weizmann Institute of Science, Rehovot, Israel

ABSTRACT

The postnatal development of responsiveness to estrogen in the rat uterus can be divided into three stages. In the first stage (lasting for a period of approximately 10 days after birth) a single dose of estrogen, administered to Wistar-derived rats, results in the stimulation of the synthesis of a limited number of uterine proteins. Those which are presently known are the ‘estrogen induced protein’ (IP) first described by Notides and Gorski, ornithine decarboxylase, the first and rate-limiting enzyme in the pathway of polyamine biosynthesis, and estrogen and progesterone receptors. During the second stage of responsiveness, seen at approximately two weeks after birth, both RNA and protein synthesis are stimulated by estrogen, while there is no effect on DNA synthesis. By three weeks after birth, all growth parameters, including DNA synthesis, are increased by estrogen treatment.We have concentrated our investigation on the early and late stages of responsiveness. IP has been characterized and identified as a constitutive component of uterus as well as of pituitary, hypothalamus and cerebral cortex. It has been purified from rat brain where it occurs in both males and females; antisera against it have been made in rabbits and a radioimmunoassay for IP is under development.The stimulation of DNA synthesis by estrogen, as measured by incorporation of tritiated thymidine, was found to be paralleled by an increase in the activity of DNA polymerase α, the putative replicative polymerase, with no increase in DNA polymerase β, possibly a repair enzyme.Throughout the entire period of postnatal development, the presence and, recently, the replenishment, of estrogen receptors have been demonstrated. We therefore use the working hypothesis that the acquisition of responsiveness to estrogen is a result of differentiation of specific chromosomal ‘acceptor sites’ for the estrogenreceptor complex.

Keywords

Estrogen

induced protein

DNA polymerase α

ornithine decarboxylase

receptors

uterus

brain

INTRODUCTION

A steroid hormone such as estrogen, which stimulates the growth and division of uterine cells, displays a wide variety of those actions which culminate in cell division. From both a basic viewpoint-elucidation of the mechanism of regulation of macromolecular synthesis-and an applied approach leading to the safer use of estrogens in contraception and therapy, it is crucial to know whether independent activation of any of estrogen’s actions is experimentally demonstrable. Such a functional dissection of estrogen’s action should provide powerful tools essential for the study of its molecular biology.

Fortunately, for investigation of this problem, the uterus of the rat shows a progressive and differential sensitivity to stimulation by estrogen during the period from birth to puberty. Price and Ortiz (1944) found that when a series of 6 daily injections of estradiol-benzoate (totalling 10 μg) was given to non-inbred rats aged 0 to 50 days, the wet weight increase and estrogen-associated histologic changes in uterus became more pronounced with age up to 26 days after birth. Thereafter the magnitude of these responses declined. A single injection of 0.5 μg estradiol–17β to Sprague-Dawley rats, ovariectomized at 21 days, showed an increase in their uterine weight response until puberty, followed by a decline in responsiveness in older animals (Liu, 1960).

More interestingly, the responsiveness to estrogen can be shown to be acquired in distinct developmental stages when individual components of the overall estrogen response are analyzed (Kaye and colleagues, 1972, 1974, 1975; Katzenellenbogen and Greger, 1974; Sömjen and colleagues, 1973a, b, 1974; Walker and colleagues, 1976, 1978; Peleg and colleagues, 1977, 1979; Peleg and Kaye, 1978). In perinatal life, although estrogen receptors are present in uterine cytoplasm and are capable of being transferred into the nucleus, there is a period of approximately two weeks after birth during which there is a stimulation of synthesis of a limited number of proteins following a single estrogen injection. Bulk protein and RNA synthesis can be stimulated by estrogen only after, the second week of life. DNA synthesis is accelerated by estrogen only in uteri of rats that are 20 days of age or older.

The emphasis in this review will be placed on the initial stage of the acquisition of estrogen responsiveness epitomized by the ‘estrogen induced protein’ first described by Notides and Gorski (1966) and on the culminating stage of stimulation of DNA synthesis.

THE PERINATAL STAGE OF RESPONSIVENESS: STIMULATION OF A SMALL GROUP OF PROTEINS

Steroid Receptors

Estrogen receptors have been demonstrated in rat fetal Mullerian duct (the precursor of the uterus) in fetuses one day before birth (Sömjen and colleagues, 1976). In order to detect the characteristic 8S receptor in the presence of the overwhelming fetal concentration of α-fetoprotein (see Nunez and colleagues, 1979; Raynaud and colleagues, 1979), which sediments at 4S, ³H-diethylstilbestrol was used as ligand instead of ³H-estradiol–17β. This synthetic estrogen was used (Fig. 1) because it shows only one thirteenth of the binding of ³H-estradiol to α-fetoproteinrich blood plasma from 11 day old rats. The developmental implication of circulating α-fetoprotein in newborn rats have been reviewed recently (c, f, Kaye, 1978; Raynaud and colleagues, 1979).

Fig. 1 ) of 8 rat fetuses taken on the 20th day of gestation were incubated in 1 nM (³H)diethylstilbestrol for 2 h at 0°C and centrifuged through 5–20% sucrose gradients. The arrows indicate the position of bovine serum albumin (4.3 S). Direction of sedimentation is from right to left. (From Sömjen and colleagues, 1976.)

Nuclear binding of estrogen was detected in uteri of one day old rats by sucrose gradient centrifugation (Sömjen and colleagues, 1973b), and was quantitated during post natal development using the exchange binding technique of Anderson and colleagues (1972).

A comparison of cytoplasmic and nuclear binding of estrogen during post natal development (Fig. 2) reveals the parallel curves predicted by the analysis of Williams and Gorski (1972). These authors showed that nuclear and cytoplasmic binding of estrogen maintained a constant ratio over all the concentrations of estrogen tested. Interestingly, a peak concentration of bound estrogen is found in 10 day old uteri, which corresponds to a peak in incorporation of amino acids into proteins at 10 days in uteri (Kaye and colleagues, 1974) and other organs of the rat (D. Kaye, unpublished data) including the brain.

Fig. 2 cytosol, average of 3 to 4 pooled samples (redrawn from , cytosol from uteri of rats with closed vaginas, or open vaginas Δ—-Δaverage of 5 to 7 pooled samples (drawn from data of Lee and Jacobson, 1971). Figure from Kaye (1978).

The replenishment of cytoplasmic receptors following estrogen administration is due, at least in part, to synthesis of receptor proteins (Sarff and Gorski, 1971; Cidlowski and Muldoon, 1978). The process of replenishment, which is evidence for the stimulation of receptor protein synthesis, takes place at 6 days to essentially the same extent and with the same time course shown at 10 and 20 days (Fig. 3). Furthermore, in 4-day-old rats, there is an approximately 6-fold induction of uterine progesterone receptors by estrogen (Raynaud and colleagues, 1979).

Fig. 3 ; 25-day-old rats. Each point represents the average value from pooled uteri in 2 independent experiments. (From Peleg and colleagues, 1979.)

Uterine Ornithine Decarboxylase

The protein whose induction has been demonstrated at the earliest post natal age is ornithine decarboxylase, the rate limiting enzyme in polyamine synthesis. At the age of two days, the specific activity of ornithine decarboxylase induced by estrogen injection is indistinguishible from the specific activity of ornithine decarboxylase attained after induction at 21 days (Kaye and colleagues, 1973, Fig. 4).

Fig. 4 Hormonal induction of ornithine decarboxylase as a function of age. Ovaries (at least five pairs per point) collected from rats injected 4 h previously with LH (30 μg/rat) were homogenized and portions of the 38, 000gmaxsupernatant fraction (0.04-0.6 mg of protein) were used for the assay. Uteri were collected from rats injected 4 h previously with 17β-estradiol (˜15 ng/g-wt, i.e)enzyme from uteri of dilute ethanol treated rats. The vertical bracket represents the 95% confidence limits of the mean value. (From Kaye and colleagues, 1973.)

The Estrogen-Induced Protein

Among the proteins specifically induced in the rat uterus by estrogen, the ‘estrogen-induced protein’ (IP) discovered and named by Notides and Gorski (1966), has two properties which make it a preferred protein marker for estrogen action. Stimulation of IP synthesis can be demonstrated within 40 minutes of estrogen administration (Barnea and Gorski, 1970) and IP synthesis can be induced by estradiol in surviving uteri in vitro (Katzenellenbogen and Gorski, 1972). This protein, which can be detected by double isotope labeling (v. review by Katzenellenbogen and Gorski, 1975) or by fluorography (Walker and colleagues, 1976) of ³⁵S-labeled proteins separated by electrophoresis on SDS–polyacrylamide gel slabs (Fig. 5), is induced as early as 5 days after birth in Wistar-derived rats. (Walker and colleagues, 1976, Fig. 6). It was previously reported by Katzenellenbogen and Greger (1974) that IP was induced by estrogen in 6-day-old Sprague-Dawley rats. During the period between 5 and 10 days after birth the inducibility of IP by estrogen increases (Fig. 6).

Fig. 5 Autoradiograms and corresponding densitometric tracings of sodium dodecyl sulfate 10–20% gradient polyacrylamide gels of uterine cytosol and partially purified IP preparations. Uteri from both untreated 20-day-old rats and from rats treated for 1 h with 5 μg estradiol–17β were separately incubated in the presence of either ³⁵S-methionine or ³H-methionine. The cytosol fraction was prepared from each of the 4 resulting groups of uteri. A mixture was made of ³⁵S-methionine labeled cytosol from uteri of untreated rats (³⁵S{C}) with ³H-methionine labeled cytosol from uteri of estrogen-treated rats (³H{E}). The complementary mixture of cytosol preparations was also made i.e., ³H {C} + ³⁵S{E}. Both mixtures were analyzed by polyacrylamide gel electrophoresis and autoradiography (left panel). Since the autoradiographic patterns are due almost entirely (> 95%) to the stronger ³⁵S radioactivity, the autoradiograms are designated according to the preparation labeled with ³⁵S-methionine e.g. the pattern designated C results from electrophoresis of cytosol mixture ³⁵S{C}+ ³H{E}.

The presence of the mixture of ³⁵S and ³H labeled cytosols, in both cases, permitted the isolation of the IP fractions from Cellogel electropherograms, using the characteristic elevation of the isotope ratio for detection of IP. The Cellogel purified IP fractions containing ³⁵S-methionine from uteri of untreated (Cp) and estrogen treated (Ep) rats were then re-analyzed by polyacrylamide gel electrophoresis (right panel). The arrows indicate the position of IP.

Fig. 6 Age dependence of IP synthesis. Uteri from 5-, 10-, 20- and 30-day-old rats (2–10 per group) injected with 1% ethanol vehicle (C) or estradiol 17β (1 μg/7 g body weight) were incubated in 1 ml PBS (Dulbecco and Vogt, 1954) in the presence of 50 μCi of ³⁵S-methionine. After 2 h incubation, uterine cytosol was prepared and subjected to electrophoresis on sodium dodecyl sulfate 10–20% polyacrylamide gels which were subsequently dried and autoradiographed. The arrows indicate the molecular weights (in order of increasing size) of actin, tubulin and serum albumin.

Because of the advantages of IP as a marker protein for estrogen action, we have selected it as the model protein to represent the small number of specific proteins synthesized in response to estrogen during the first few days of life by the rat uterus. In order to improve the resolution of IP from other proteins and to search for additional proteins which may be induced by estrogen, we subjected ³⁵s methionine-labeled uterine extracts to combined isoelectric focusing and SDS polyacryl amide slab gel electrophoresis, essentially according to O’ Farrell (1975) but using non-denaturing conditions in the first dimension. We observed (Fig. 7) a single spot in fluorograms of the gels which was increased by estrogen in every test. The molecular weight (46, 000–48, 000) and the isoelectric point (4.6–4.7) of this spot was consistent with the characteristics of IP (Iacobelli and colleagues, 1973; Sömjen and colleagues, 1973c; King and colleagues, 1974; Katzenellenbogen and Williams, 1974).

Fig. 7 Fluorogram of 2-dimensional polyacrylamide gel separation of uterine cytosol proteins using non-denaturing conditions in the first dimension. ³⁵S-labeled proteins derived from uterine cytosol of untreated (C) and estrogen-treated (E) rats were subjected to isoelectric focusing run under non-denaturing conditions, followed by electrophoresis in the presence of SDS. The origin of migration is at upper right of figure; the arrows indicate the position of IP.

When uterine cytosol was analyzed for IP following the original method of O’Farrell (1975), which employs urea as denaturing agent for isoelectric focusing (Fig. 8), the spot of Mr 46, 000-48, 000, which was increased by estrogen treatment, was found to focus at a position much closer to neutrality than when not denatured, and therefore to shift its position relative to other easily identifiable proteins on the fluorograms such as the most prominent spot, actin (Mr = 43, 000). The reason for this shift is currently under investigation.

Fig. 8 Fluorogram of 2-dimensional polyacrylamide gel separation of uterine cytosol proteins using denaturing conditions in the first dimension. ³⁵S-labeled proteins derived from cytosol of untreated (C) and estrogen-treated (E) uteri were subjected to isoelectric focusing run under denaturing conditions followed by electrophoresis in the presence of SDS. The origin of migration is at upper right of figure; the arrows indicate the position of IP.

Under both native and denaturing conditions no consistent change has been observed in any protein of the fluorograms other than IP.

For more rapid surveys than are possible using conventional two dimensional gel analysis, we took advantage of the properties of Cellogel, gelatinized cellulose acetate, as a support medium for electrophoresis of IP (Sömjen and colleagues, 1973b). Cellogel, a sponge-like matrix, has no molecular sieving effect so that electrophoretic migration in Cellogel is solely due to the net charge of the molecule. Additionally, proteins can simply be squeezed out of Cellogel using a syringe, either by hand or by centrifugation. Uterine cytosol extracts were separated by electrophoresis on Cellogel blocks into 8 fractions which were subsequently subjected to SDS polyacrylamide slab gel electrophoresis, a separation dependent on molecular weight.

Fraction 7 from the Cellogel block (Fig. 9) contained most of the IP, permitting the use of this fraction for the comparison of extracts from organs of control and estrogen-treated rats (compare Fig. 5). This technique confirmed the constitutive presence of IP in uteri from immature rats (Walker and colleagues, 1976). It also facilitated the search for IP in other organs in which IP exists constitutively but, in contrast to the uterus, is not induced by estrogen.

Fig. 9 Fluorogram of SDS polyacrylamide gel electrophoresis of unfractionated and Cellogel-fractionated cytosol. Samples of ³⁵S-labeled uterine cytosol proteins from control (C) and estrogen-treated (E) rats were subjected to Cellogel electrophoresis and separated into 8 fractions according to mobility relative to bovine serum albumin (BSA). Fraction 1, −0.1 to −0.1 times mobility of BSA; 2, 0.1 to 0.3; 3, 0.3 to 0.5; 4, 0.5 to 0.7; 5, 0.7 to 0.9; 6, 0.9 to 1.05; 7, 1.05 to 1.2; 8, 1.2 to 1.4. Unfractionated cytosol and fractions 2–8 were then subjected to electrophoresis on a 10–20% polyacrylamide gel. The exposure time of the total cytosol samples was less than that of the fractions, in order to avoid over-exposure. From Walker and colleagues, 1979.

IP was found in the pituitary gland and in the hypothalamus and cerebral cortex of female rats as well as in the cerebral cortex of male rats (Walker and colleagues, 1979, Fig. 10). Very much smaller concentrations of IP were found in liver and in muscle. In none of these organs has induction of IP by estrogen yet been demonstrated. Thus, while induction of IP by estrogen may be a uterine specific response, the presence of IP is more general and the protein is found in both male and female rats.

Fig. 10 ), liver (Li) and muscle (Mu). The Cellogel fraction shown is that corresponding to the mobility of IP (fraction 7 as shown in Fig. 9. Lanes 1 and 16 show molecular weight markers (BSA, 67, 000; ovalbumin, 45, 000; hemoglobin, 16, 000). IP migrates slightly slower than ovalbumin. (From Walker and colleagues, 1979)

To obtain further evidence for the close relationship or identity of IP from pituitary, brain and uterus, IP samples were cut from SDS polyacrylamide gels and placed, along with S. aureus protease V8 in the slots of a second SDS polyacrylamide gel (Cleveland and colleagues, 1977). Digestion was allowed to proceed in the stacking gel. The resultant fluorogram of the slab gel electropherogram (Fig. 11) showed a pattern of bands in which IP samples from different organs were indistinguishable, but were grossly different from actin and from tubulin, proteins which have similar electrophoretic migration to IP on SDS gels (see Fig. 5).

Fig. 11 s. aureus V8 protease partial digestion pattern of IP-like proteins from uterus (Ut), pituitary (Pi), hypothalamus (Hy) and cerebral cortex (Co) of rats. Purified samples of IP-like proteins from the above organs were cut from SDS polyacrylamide slab gels and subjected to electrophoresis on SDS 15–20% polyacrylamide gels in the presence or absence of s. aureus protease. Lanes 1 and 8 contain molecular weight markers (BSA, ovalbumin, α-chymotrypsin-ogen, myoglobin and cytochrome C). Lanes 2-7: IP-like proteins in the absence of protease; Lanes 9–14: digestion pattern resulting from addition of protease. Organ abbreviations are those used in the legend to Fig. 10. The letters under the organ abbreviations indicate preparations from control (C) and estrogen-treated (E) rats. Lanes 15 and 16 show rat tubulin (T) and rabbit muscle actin (A) respectively, in the presence of protease. The patterns for uterus and pituitary represent fluorograms of the gel after electrophoresis, whereas those for the remainder are Coomassie brilliant blue staining patterns.

The finding that IP, a protein originally considered uterine-specific, is found in other organs of both males and females, is paralleled by the case of uteroglobin in rabbit uterus. Uteroglobin, also called blastokinin, was originally thought to be specifically induced by progesterone in rabbit uterus. Recently, however, it has been found in the male reproductive tract and in the respiratory tract (Noske and Feigelson, 1976; Feigelson and colleagues, 1977; Torkkeli and colleagues, 1977).

Since brain appeared to be the richest source of IP available, both in terms of concentration in the organ and the weight of the organ per rat, brain was chosen as the source for IP purification. Successive steps of ammonium sulfate precipitation, DEAE cellulose chromatography and preparative gel electrophoresis yielded a product that was >85% pure (Fig. 12), This IP preparation was injected into male and female rabbits to raise anti-IP sera. Immunoprecipitation, using inactivated S. aureus cells, instead of second antibody, revealed antibodies against IP in all rabbits injected (Fig. 13). The S. aureus cell membranes contain protein A, which has the property of binding the Fc fragment of IgG and also often binds actin, as in this system (Fig. 13).

Fig. 12 Purification of IP from rat brain. 1, 40, 000 × g supernatant fraction; 2, ammonium sulfate 40–63 fraction; 3, DEAE cellulose fraction; 4, preparative SDS-polyacrylamide gel electrophoresis fraction, 1 μg; 5, 2 μg (as number 4); 6, 3 μg (as number 4).

Fig. 13 Immune precipitation of IP in uterine cytosol. The immune complex was precipitated by formaldehyde-treated S. aureus. Lanes 1–4, preimmune serum from 4 rabbits; 5–8, rabbit serum after 2 injections of IP; 9–12, rabbit serum after 3 injections of IP; 13, S. aureus + cytosol alone (with no serum addition). The lower band which appears in all lanes is actin.

The fact that an antibody prepared against a brain IP preparation specifically precipitates uterine IP is additional evidence for the close similarity (or identity) of IP derived from brain and uterus. The complementary observation has been made by S. Iacobelli (personal communication) who finds the greatest reaction with anti-uterine IP serum in brain cytosol.

The availability of reasonable quantities of IP from brain will permit a much more efficient search for its function. Preliminary evidence indicates that during the first month after birth the amount of IP in brain increases both relatively and absolutely (as shown by radioimmunoassay) (I. Gozes, unpublished observations; N. Reiss, unpublished observations).

THE INTERMEDIATE STAGE OF RESPONSE: STIMULATION OF ALL MACROMPLECULAR SYNTHESES EXCEPT DNA

At 5 or 10 days after birth, estradiol–17β does not cause a significant increase in the weight, protein, RNA or DNA content of Wistar rat uteri (Fig. 14) when measured 24 h after administration, the time of maximal effect in older rats. However at the age of 15 days there is a significant increase in wet weight, including increases in both protein and RNA content with no change in the content of DNA (Fig. 14).

Fig. 14 , R2858. (from Sömjen and colleagues, 1973a)

In addition to the change in responsiveness to estrogen shown by IP synthesis during the period between 5 and 10 days after birth, discussed in the previous section, Katzenellenbogen and Greger (1974) found that estradiol causes a very small increase in 2-deoxyglucose phosphorylation at the age of 9 days. This response is capable of maximal stimulation at 19 days. In the same series of experiments, an interesting change in responsiveness measured by wet weight increase, was found to occur between days 12 and 13 after birth. By 12 days after birth, the uteri of the Sprague-Dawley rats respond to estrogen by an 80% weight increase. This value is the same whether a single injection is given, or a series of three daily injections is begun on day 12. When injections are begun on day 13, three daily injections lead to a 230% increase.

An intriguing age dependent morphologic response to estrogen has been observed recently in rat uterine epithelial cells by LeGoascogne and Baulieu (1977). ‘Nuclear bodies’ (round nuclear organelles, between 0.2 and 1.0 μm in diameter in the un-condensed chromatin region during interphase) appear on day 8 after birth and remain constant in number between day 10 and 20. If diethylstilbestrol (0.5 μg) is injected on days 8 and 9, the number of nuclear bodies on day 10 is triple that in untreated controls.

THE STAGE OF COMPLETE RESPONSIVENESS: DNA SYNTHESIS

Thymidine Incorporation Studies

Estrogen administration at the age of 15 days does not result, in our strain of Wistar derived rats, in stimulation of DNA synthesis as measured by thymidine incorporation. However, by the age of 20 days a two-fold stimulation is observed (Fig. 15). The use of a ‘priming’ injection of estradiol at 13 days before a second estradiol injection at 15 days, does not lead to any increase in the rate of DNA synthesis measured on day 16.

Fig. 15 rat given 5 μg estradiol. Vertical lines indicate 95% confidence intervals. Inserted histogram 01, proestrus; Co estrus; L1 metestrus and L2 diestrus. Open bars, control rats; lined bars, rats given 5 μg estradiol. The cycling rats were 40 days old, the pregnant rat was a multiparous adult. (From Kaye and colleagues, 1972)

Mitotic index counts (Kaye and colleagues, 1972) established that the observed increases in DNA synthetic rate do indeed reflect increased cell division. These experiments also showed that all the cell types of the 20-day-old rat uterus are stimulated to divide by a single injection of as little as 5 ng of estradiol–17β per rat. Although the luminal epithelium showed, as expected, the highest response, even the myometrium was highly stimulated to reach half the mitotic index shown by luminal epithelium. The fact that this extent of cell division in 20-day old rat uterus is not reflected by an increase in DNA content when measured 24 h after estrogen treatment (Fig. 14) was noted as early as 1958 by Mueller and colleagues. However in 25-day-old rats, an estrogen induced increase in DNA content is seen, which reaches a highly significant value in uteri of 30-day-old rats. Luck and colleagues (1974) confirmed the absence of an estrogen induced increase in uterine DNA content at 20 days, but reported such an increase at 7 days in the unspecified strain of rats they employed.

Stimulation of DNA Polymerase α

The presumptive replicative polymerase, DNA polymerase a (Weissbach, 1977) shows an increase in activity in immature rat uterus following estrogen administration (Harris and Gorski, 1976, 1978; Walker and colleagues, 1978). The time course of stimulation of DNA polymerase a parallels that in the thymidine incorporation studies discussed in the previous section. Moreover, although uterine DNA polymerase αactivity can be stimulated by doses of estrogen as low as 0.6 ng/g body weight, in 20- and 25-day-old rats, doses as high as 170 ng/g body weight failed to cause any increase in DNA polymerase a activity in uteri of 10 or 15 day old rats (Fig. 16). Thus, both the ability to respond to estrogen by an increase in thymidine incorporation into DNA and by an increase in DNA polymerase a activity develop in parallel between 15 and 20 days after birth in Wistar-derived rats, raising the possibility that a limiting factor for estrogen-stimulated DNA synthesis may be DNA polymerase αactivity.

Fig. 16 Age dependence of the stimulation of uterine DNA polymerase α activity by estradiol–17β at 24 h after administration. Activity was measured under low salt (open bars, measuring both αand βpolymerases) and high salt conditions (hatched bars, measuring only polymerase β). C = untreated; E = estrogen-treated rats. Values shown are the mean ± SEM of 3–4 independent determinations on groups of 4 animals per age group.

FUTURE PROSPECTS

A collection of mutant estrogen-responsive cells in culture, would be a most powerful tool for investigation of the mechanism of estrogen action. The closest approach at present to such a genetic system may be the investigation of post natal development of responsiveness to estrogen described in this review. The search for factors which regulate the sequential acquisition of responsiveness to estrogen will probably involve investigation of the non-histone proteins of uterine chromatin, for as in most developmental systems, the most likely candidates for the regulatory molecules arising by differentiation are these as yet scarcely investigated proteins. Moreover, among these non-histone proteins in chromatin are found components of the ‘acceptor sites’ for several steroid hormone receptor complexes (King and Mainwaring, 1974).

An immediate aim is to find a function of the ‘induced protein’ in order to provide some insight into its role in the uterus and perhaps to simplify it determination.

The longer range aim of the research reported from our laboratory is the construction of a cell-free system in which estrogen action can be reconstituted in vitro, in order to study the mechanism of estrogen action at the molecular level.

ACKNOWLEDGEMENTS

We thank all our collaborators, whose names are included in the references to our joint papers, for their contribution to this project. Work in our laboratory has been supported in part by grants (to H. R. Lindner) from the Ford Foundation and Population Council, N.Y., USA.

REFERENCES

Anderson, J.M., Clark, J.H., Peck, E.J. Biochem. J. 1972; 126:561.

Barnea, A., Gorski, J. Biochemistry. 1970; 9:1899.

Cidlowski, J.A., Muldoon, T.G. Biol. Reprod. 1978; 18:234.

Clark, J.H., Gorski, J. Science. 1970; 169:76.

Cleveland, D.W., Fischer, S.G., Kirschner, M.W., Laemmli, U.K. J. Biol. Chem. 1977; 252:1102.

Dulbecco, R., Vogt, N. J. Expt. Med. 1954; 99:183.

Feigelson, M., Noske, I.G., Goswami, A.K., Kay, E. Ann. N.Y. Acad. Sci. 1977; 286:273.

Harris, J.M., Gorski, J. Fed. Proc. 1976; 35:1528.

Harris, J.N., Gorski, J. Mol. Cell. Endocr. 1978; 10:293.

Iacobelli, S., Paparatti, L., Bompiani, A. FEBS Lett. 1973; 32:199.

Kaye, A.M., Sheratzky, D., Lindner, H.R. Biochim. Biophys. Acta. 1972; 261:475.

Kaye, A.M., Icekson, I., Lamprecht, S.A., Gruss, R., Tsafriri, A., Lindner, H.R. Biochemistry. 1973; 12:3072.

Kaye, A.M., Sömjen, D., Sömjen, G., Walker, M., Icekson, I., Lindner, H.R. Biochem. Soc. Trans. 1975; 3:1151.

Kaye, A.M., Sömjen, D., King, R.J.B., Sömjen, G., Icekson, I., Lindner, H.R. Adv. Exp. Med. Biol. 1974; 44:383.

Kaye, A.M.Litwack, G., eds. Biochemical Actions of Hormones; 5. Academic Press, N.Y., 1978:149.

King, R.J.B., Mainwaring, W.I.P.Steroid Cell Interactions. London: Butterworth, 1974.

Katzenellenbogen, B.S., Gorski, J. J. Biol. Chem. 1972; 247:2399.

Katzenellenbogen, B., Gorski, J. Methods in Enz. 1975; 36:444.

Katzenellenbogen, B.S., Greger, N.G. Mol. Cell. Endocr. 1974; 2:31.

Katzenellenbogen, B.S., Williams, L.B. Proc. Nat. Acad. Sci. USA. 1974; 71:1281.

King, R.J.B., Sömjen, D., Kaye, A.M., Lindner, H.R. Mol. Cell. Endocr. 1974; 1:21.

Lee, C., Jacobson, H.I. Endocrinology. 1971; 88:596.

Le Goascogne, C., Baulieu, E-E. Revue Biologie Cellulaire. 1977; 30:195.

Liu, F.T.Y. Am. J. Physiol. 1960;