- Email: [email protected]
Protein synthesis in the uterus of the immature rat: The effects of short exposures to oestradiol-17β
J. swroid Biochem. Vol. 19, No. 4, pp. 1503 1508, 1983
0022-4731/8353.00+0.00 Copyright (c 1983 Pergamon Press Ltd
Printed in Great Britain. All rights reserved
PROTEIN SYNTHESIS IN THE UTERUS OF THE I M M A T U R E RAT: THE EFFECTS OF SHORT EXPOSURES TO OESTRADIOL-17fl JOANNE M. BEAUMONT* and JOHN T. KNOWLERt Department of Biochemistry, University of Glasgow, Glasgow G12 8QQ, U.K.
(Receired 7 December 1982) Summary--Three methods we,e used to study the effect of oestrogen on the incorporation of radioactive precursor into uterine protein. Intact tissue was incubated in vitro. Isolated uterine epithelial, stromal and myometrial cells were labelled in vitro. Isolated polysomes were translated in cell free protein synthesising systems. In all of these systems, minor qualitative changes in protein synthesis were detected where the uteri were derived from oestrogen-treated rats. These changes were most dramatic in isolated stromal cells and were accompanied by a marked overall increase in protein synthesis. The translation of m RNA on isolated uterine polysomes revealed a sequence of minor, but reproducible, oestradiol-induced changes. It was difficult, however, to relate these changes to those detected in incubated tissue or cells, possibly because the cell free translation products were not subject to normal posttranslational modification and processing.
Assay of total uterine protein synthesis
INTRODUCTION
We have shown that oestradiol-17fl induces profound qualitative and quantitative changes in the m R N A population of the immature rat uterus. O f the m R N A species isolated from the tissue 4 h after oestrogen administration, only 563o were present 2 h after treatment and only 11~o were detected in untreated animals[l]. These dramatic differences were unexpected as few oestrogen-induced proteins have been detected in the rat uterus (for a review see 2). Here we describe an attempt to clarify this ambiguity. We have examined the incorporation of precursor into uterine proteins by the fluorographic detection of radioactive polypeptides fractionated on twodimensional polyacrylamide gels and have analysed protein synthesis in the intact uterus, in isolated uterine ceils and in cell free protein synthesising systems. We confirm that the synthesis of those proteins which are abundant enough to be detected by these techniques, undergo only minor changes in response to oestrogen and we discuss how this relates to our previous findings. MATERIALS AND M E T H O D S
The source and maintenance of immature female rats, the administration of oestradiol-17fl and the in vitro incubation of excised uteri have been previously described. [3].
*Current address: Department of Pharmacy, University of Strathclyde, Royal College, Glasgow G1 1XW, U.K. /To whom all communications should be addressed. Abbreviations: Oestradiol- 17fl- 1,3,5,(10)-estratriene-3,17fldiol.
Total uterine protein was prepared from groups of 6 incubated uteri which were homogenised in 2 ml of a buffer containing 10raM Tris-HC1, pH 7.4, 10raM NaC1, 1.5 m M MgC12. This and subsequent steps were conducted at 0 ~ , C . Insoluble debris was removed by centrifugation at 10,000g for 5 rain in a Sorvall SS34 fixed-angle rotor and the resulting supernatant was precipitated with a 9 vol acetone containing 0.1 M HCI at - 2 0 ° C overnight. The precipitate was pelleted by centrifugation, washed with acetone, dried in a stream of nitrogen and dissolved in lysis buffer[4] for specific activity determination and two-dimensional gel electrophoresis.
Assa)' of protein ~Tnthesis by isolated cells Immature rat uteri were fractionated into epithelial, stromal and myometrial cells as described by M c C o r m a c k and Glasser[5] and were examined for purity and damage by phase contrast and transmission electron microscopy. The latter was performed on cell pellets fixed in glutaraldehyde-osmium tetroxide by I. Montgomery of the Department of Physiology, Glasgow University. Before analysis of protein synthesis, the separated cells were rested for 4 h in Eagles medium [6]. This was necessary because the freshly prepared cells were very inefficient in protein synthesis. The 4 h incubation was the minimum "rest" period required in order that the cells regained a rate of precursor incorporation which was comparable with that of the intact tissue. Rested cells were resuspended in 1 ml Eagles medium lacking methionine but containing 70 # Ci of L-[3H]methionine (800-1450Ci/mmol) and 5~o heat inactivated, charcoal stripped, foetal calf serum. The
1503
1504
JOANNE M. BEAUMONT and JOHN T. KNOWLER
uteri. A very complex pattern of labelled polypeptides, detected as radioactive spots on two dimensional gels [4] was obtained. When rats had received oestradiol at varying times before death, minor changes were observed in the in vitro incorporation of the radioactive amino acids but these were not very reproducible due to partial masking by the complex nature of the protein population (results not shown). Various methods of fractionation were employed in an attempt to reduce the complexity of the protein populations. Nuclear and cytoplasmic protein were isolated from the labelled uterine tissue but this did Assay of cell.li'ee polysomal protein synthesis little to reduce the problem. Cytoplasmic preparaUterine polysomes were prepared as described by tions were still very complex and resulted in patterns Merryweather and Knowler[7]. Cell free translation very similar to those since described in the mouse by mixes contained 8/~1 of micrococcal nuclease-treated Korach et al[ll]. The nuclear protein profile was rabbit reticulocyte lysate (Radiochemical Centre, distinct; however, only one reproducible difference Amersham), 2 ~1 of k-[3H]methionine (5/~Ci//il) and was detected in the proteins synthesised before and l/,l of a polysome preparation. Blanks contained after hormone treatment. A polypeptide of mol. wt sterile distilled water in place of polysomes. Assays 65,000 which focussed at a pH of about 5.2 disapwere incubated at 3 0 C for 90 rain and terminated by peared when rats had been treated with oestradiol for cooling at 0'C. Aliquots of the cell free transaction, varying times before death (results not shown). used to follow the time course and extent of protein McCormack and Glasser[5] have described a techsynthesis, were dried onto 2.5 cm Whatman 3 MM nique for the fractionation of different uterine cell filter discs (Fisons). The discs were then placed in ice types, which retain good cellular morphology and cold 10!I; (w/v) trichloroacetic acid for 10min to viability. We have applied their methods to an analprecipitate the protein, boiled in 10",, (w/v) tri- ysis of protein synthesis in uterine epithelial, stromal chloroacetic acid containing 10mM k-methionine and myometrial cells. Isolated cells were incubated, (Sigma) for 15 min and washed twice in cold 51~i(w/v) as described in the methods section, in 70 #Ci/ml of trichloroacetic acid. Discs were rinsed in ethanol, [3SS]methionine. The radioactive total cellular protein dried and counted in 10 ml 0.5,; w/v was then isolated and fractionated on two dimen2,5-diphenyloxazole in toluene. The remainder of sional polyacrylamide gels. Figure 1 illustrates the each assay was used to analyse the products of results obtained with epithelial and stromal cells translation by two-dimensional gel electrophoresis. isolated from the uteri of control rats and from the uteri of rats which had received oestradiol 4 h before Protein .fi'actionation and analysis death. The pattern of radioactive polypeptides detected The two dimensional fractionation of proteins was essentially as described by O'Farrell[4] except that the with epithelial cell preparations (Fig. la and b) was less complex than total uterine protein and was second dimension, sodium dodecyl sulphate polyacrylamide gels were prepared as described by dominated by a multi-component spot which coLeStourgeon and Beyer[8]. Proteins were assayed by migrated with rat and rabbit skeletal muscle actin. the method of Bramhall et a/[9] which avoided Figure la and lb show the labelled epithelial proteins interference by the ampholines present in lysis of cells derived from control animals and 4 h buffer [4]. Proteins, labelled by incubation in vitro, oestrogen-treated rats respectively. The only proteins were detected on the two-dimensional poly- which consistently appeared in the preparation as a acrylamide gels as described by Bonner and result of hormone treatment are indicated by arrows Laskey[10]. Rat leg muscle actin was the gift ofJ. Elce in Figure lb. Figure lc and ld shows labelled stromal proteins of Department of Biochemistry, Queens University, Kingston, Canada and rabbit skeletal muscle actin derived from control and 4 h oestrogen-treated rats. was prepared from an acetone powder which was the Hormone administration always resulted in a much gilt of J. Dow, Department of Biochemistry, Glasgow higher specific activity in the stromal proteins. Thus, University. Rat brain creatine kinase was the gener- in a typical experiment, the protein from control rats, ous gift of A. Kaye, Weizman Institute of Science, which received carrier only, incorporated radioactive methionine to 7.0 x 102cpm//tg while that of Rehovot, Israel. hormone-treated animals incorporated precursor to 2.23 x l03 cpm//ag. This difference in specific activity RESU LTS was reflected in the density of the spots on the Our initial experiments examined the in vitro incor- fluorograms and virtually all detected proteins were poration of both pSS]methionine and ~4C-mixed synthesised in greater quantities after oestrogen adamino acids into the total protein of immature rat ministration. The effect was more dramatic in some
cells were incubated for 90 min at 37 C under an atmosphere of 95°.,,, O2/5°i; CO> After the incubation, the cells were pelleted at 1000 rpm for 5 minutes in a Beckman T J-6 bench-top centrifuge and the medium discarded, the pellets were washed twice in 5ml phosphate buffered saline and repelleted. Pellets were resuspended in l ml isotonic saline, homogenised, and the solution precipitated with 9 vol of acetone, 0.1 M HCI overnight at - 2 0 C. Precipitates were washed with acetone, dried under nitrogen and dissolved in lysis buffer [4].
Effects of oestrogen on protein synthesis
1505 IEF
SDS
94
67 ,~W x10-3 43 3O
20
Fig. I. Polypeptides synthesised by preparations of uterine cells (a) polypeptides of epithelial cells from the uteri of untreated rats (100/~g--9.6 × 104cpm) (b) polypeptides of epithelial cells from the uteri of rats receiving oestrogen 4 h before death (100 t~g--l.2 × 105 zpm) polypeptides of stromal cells from the uteri of untreated rats (100#g--7 x 104cpm) (d) polypeptides of stromal cells from the uteri of rats receiving oestrogen 4 h before death (65 #g--l.4 x 105cpm).
proteins than others and was particularly striking in two polypeptides, indicated by open arrows in Figure ld, which were in the same positions as those identified as tubulins by Korach et al.[ll]. In addition, a number of newly labelled polypeptides appeared after oestradiol administration and nine, the appearance of which was reproducible, are labelled with solid arrows in Figure l d. Myometrial cell preparations incorporated radioactive amino acids into protein very inefficiently. Specific activities were such that only very few of the most abundant proteins could be detected on fluorograms and no realistic analysis of hormoneinduced changes could be made (results not shown), The efficient labelling of the separated uterine
cells was only possible after they had been allowed to recover from the trauma of isolation by resting in a complete, aerated medium for 4h. Thus, it was possible that the differences between the labelled protein patterns of isolated cell types and those of whole uterine protein stemmed from the labelling technique rather than the source of material. To test this possibility, uterine cell types were also isolated after the proteins had been labelled by incubating the intact tissue in vitro as described above. Analysis of the radioactive polypeptides in the isolated cells gave results very similar to those of Figure 1. Figure 2a-d shows the radioactive protein translation products of polysomes purified from the intact uteri of control, 2, 4 and 12 h treated rats respectively.
1506
JOANNE M. BEAUMONT and JOHN T. KNOWLI~R ...,r-----
IE F
5D5
l
1 9~
67 AAW
×10"3 43
30
20
a
b
d
Fig. 2. Translation products of polysomes purified from the intact uteri of (a) untreated rats (b) rats receiving oestrogen 2 h before death (c) rats receiving oestrogen 4 h before death (d) rats receiving oestrogen 12 h before death.
The most striking feature of this study was the high degree of correspondence between the controls (animals receiving carrier only) and all hormonal time points. Even the most minor spots could be detected consistently. However, hormone-induced changes were seen in the translation products of uterine polysomes from hormone treated rats. Those polypeptides which showed increased incorporation of precursor or which appeared for the first time by 2 h after oestrogen administration are indicated by arrows in Fig. 2b while three proteins which were synthesised in decreased quantities after hormone treatment are arrowed in the control (Fig. 2a). The
right hand of the two polypeptides arrowed in the centre of Fig. 2b probably represents the so called induced protein of Notides and Gorski[12] recently identified as an isoenzyme of creatine kinase by Reiss and Kaye[13]. Purified creatine kinase, kindly provided by A. Kaye, migrated to the same position on a parallel gel. This polypeptide was also detected in fractionations of total uterine protein labelled in t:itro. It was not however unambiguously located in epithelial or stromal cell preparations The changes in proteins synthesised by uterine polysomes from rats given oestradiol 2 h before death were maintained in those receiving 4h treatment
Effects of oestrogen on protein synthesis except that two polypeptides of tool. wt. 49,000 were no longer synthesised is increased quantities. Some further increases in the synthesis of specific polypeptides also occurred at this time and are indicated by arrows in Figure 2c. Proteins made by uterine polysomes isolated from rats receiving oestradiol 12 h before death showed little qualitative change compared with those from 4 h treated animals but, in most cases, maintained the changes occurring between 2 and 4 h (Fig. 2d). DISCUSSION
We have not found it possible to label rat uterine protein in t,h,o to a specific activity adequate for the analysis of individual components of the protein population. For this reason, a number of different approaches were employed by which the effects of hormone treatment in ~,ivo was monitored by assessing the incorporation of precursor into protein in vitro. There is, however, some precedent for believing that the incorporation of precursors in vitro may be abnormal. Thus, although oestrogen-induced hnRNA synthesis can be demonstrated in vitro [14], the stimulated synthesis of rRNA, which in ~:it~o is 10 12-fold control levels, is reduced to 2-3-fold in t~itro and can be explained by precursor uptake [15]. Nevertheless, oestrogen stimulated synthesis of IP [12] and of ribosomal protein (Muller and Knowler, unpublished observations) can be demonstrated in t,itro.
Studies with fractionated cells may solve some problems but create others. Problems of anoxia, which are likely to affect the inner cells of incubated tissue are reduced by cellular fractionation. Preferential uptake of radioactive precursor by the outer serosal cells of the tissue is also avoided. However, the cells are shocked by the trauma of isolation and require a 4 h rest before they will efficiently make protein. That the translation products of shocked cells can be markedly changed is shown by the heat shock response [16]. Response to hormones may also be lost during cellular fractionation. Higgins et al.[l 7] showed that castration markedly decreased the incorporation of methionine into proteins of seminal vesicle tissue but not of isolated, undamaged cells. They suggested that the cells in t'ivo were subject to vectorial constraints such as stromal-epithelial interactions. Notwithstanding the above constraints, the proteins synthesised in t~itro by incubated tissue and by fractionated cells are very similar to each other and between preparations. Qualitatively, there is considerable similarity in the proteins made by epithelial and stromal cells (Fig. 1) but the two cell types are strikingly different in their responses to the hormone. In both cell types there are minor qualitative changes in oestrogen.treated rats but only in the stroma is there a marked quantitative effect in which overall protein synthesis is considerably stimulated. This suggests a differential hypersR 1 9 4
1507
trophy in the stroma at this early stage of the oestrogenic response. If one can assume that mRNA molecules on polysomes are being actively transcribed, then the isolation of polysomes and their continued translation in eitro must most clearly represent protein synthesis in ~,i~'o. This assumes however that the translation in vitro produces full length polypeptides and even these may differ significantly in size and charge from proteins made in intact cells because they lack post-translational processing and modification. The spectrum of proteins made by isolated uterine polysomes show a number of similarities with those made by the tissue or isolated cells but there are also marked differences. There is little if any similarity in the oestrogen-induced proteins detected in the two systems. We are unable at present to identify many of the new polypeptides detected in either system. The two polypeptides arrowed in the centre of Fig. 2b are in the expected positions of the proteins collectively known as IP and including creatine kinase [13], an isoenzyme of enolase[18] and one other polypeptide [19]. The right hand component migrates in the same position as purified brain creatine kinase. We also detect this polypeptide in protein from total uterus and total uterine cytoplasm. It is not seen in nuclear preparations nor can it be definitely assigned in epithelial and stromal ceils. These observations confirm the previously published cytoplasmic locations of IP[20], and suggest that it may be a myometrial protein. However, two other reports[21,22] show that IP is synthesised in epithelial, stromal and myometrial cells. We are unable at the present time to explain these differences. One other oestrogen-induced component, that arrowed beneath actin in Fig. 2b, has been provisionally identified in the radioactive proteins isolated from in t'itro incubated nuclei. It is not, however, present in cytoplasmic proteins and may correspond to the 42,000 tool. wt non-histone protein described by King et a/.[23]. In all of the above systems, the qualitative effects of oestrogen on protein synthesis are minor. This supports the data derived from the two dimensional fractionations of total rat uterine proteins by Walker and Kaye[24] and of the mouse uterine proteins by Korach et al.[l 1]. The patterns of radioactive proteins described by both groups are very similar to our own. Nevertheless, the results are not those which might have been predicted from our previous studies of uterine mRNA populations[l] or from the very quiescent nature of the unstimulated immature rat uterus in which protein synthesis is at a very low level and most ribosomes are in the form of monosomes [7]. Estimates of the maximum resolving power of the two dimensional system range from 2500 5000 different components [4, 25] and O'Farrell[4] has demonstrated the ability of the system to detect more than 1000 bacterial proteins.
1508
JOANNE M. BEAUMONTand JOHN T. KNOWLER
However, in animal cells, relatively few proteins (50-60) may comprise greater than 50°~, o f the total protein mass [26]. This, together with the limitations on protein load and specific activity in the above experiments, mean that we are only likely to detect the m o s t a b u n d a n t o f the uterine proteins. Our previous findings that most oestrogen-induced changes occur in the m R N A species which are rare or o f intermediate a b u n d a n c e may therefore explain our inability to detect m o r e than a few oestrogen-induced changes in the synthesis o f specific proteins.
REFERENCES
1. Aziz S., Balmain A. and Knowler J. T.: Qualitative and quantitative changes in uterine mRNA populations in response to oestradiol treatment of rats. Eur. J. Biochem. 100 (1979) 95 100. 2. Beato M. (ed.): Steroid-Induced Uterine Proteins. Elsevier/North Holland Biomedical Press, Amsterdam (1980). 3. Knowler J. T. and Smellie R. M. S.: The synthesis of RNA in immature rat uterus responding to oestradiol17ft. Biochem. J. 125 (1971) 605-614. 4. O'Farrell P. H.: High resolution two-dimensional electrophoresis of proteins~ J. biol. Chem. 251 (1975) 4007-4021. 5. McCormack S. A. and Glasser S. R.: Differential response of individual uterine cell types from immature rats treated with oestradiol. Endocrinology 106 (1980) | 634-1649. 6. Busby D. W. G., House W. and Macdonald J. R.: In Virological Technique. Churchill, London (1964). 7. Merryweather M. J. and Knowler J. T.: The kinetics of the incorporation of newly synthesised RNA and protein into the ribosomes of the uterus of the oestrogenstimulated immature rat. Biochem. J. 196 (1980) 405-410. 8. LeStourgeon W. M. and Beyer A. L.: The rapid isolation, high-resolution electrophoretic characterisation and purification of nuclear proteins. Meth. cell. Biol. 16 (1977) 387-406. 9. Bramhall S., Noack N., Wu M. and Loewenberg J. R.: A simple colorimetric method for determination of protein. Analyt. Biochem. 31 (1969) 146 148. 10. Bonner W. M. and Laskey R. A.: A film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 46 (1974) 83 88. 11. Korach K. S., Harris S. E. and Carter D. B.: Uterine proteins influenced by estrogen exposure, analysis by two dimensional gel electrophoresis. Molec. cell. Endocr. 21 (1981) 243 254.
12. Notides A. C. and Gorski J.: Oestrogen-induced synthesis of a specific protein. Proc. natn. Acad. Sci., U.S.A. 56 (1966) 230 235. 13. Reiss N. A. and Kaye A. M.: Identification of the major component of the oestrogen-induced protein of rat uterus as the BB isoenzyme of creatine kinase. J. biol. ('hem. 256 (1981) 5741-5749. 14. Knowler J. T.: The incorporation of newly synthesized RNA into nuclear ribonucleoprotein particles after oestrogen administration to immature rats. Eur. J. Biochem. 64 (1976) 161 165. 15. Knowler J. T., Borthwick N. M. and Smellie R. M. S.: Early effects of oestrogen on the transcription of uterine. ribonucleic acid. Biochem. Soc. Trans. 3 (1975) 1177-1180. 16. Storti R. V., Scott M. P., Rich A. and Pardue M. L.: Translational control of protein synthesis in response to heat shock in D. melanogasta cells. Cell 22 (1980) 825 834. 17. Higgins S. J., Brooks D. E. and Fuller F. M.: Isolation of cells from rat seminal vesicles and epididymis and their use in studying androgen action. Molec. cell. Endocr. 23 (1981) 207-223. 18. Kaye A. M. and Reiss N.: The uterine estrogen-induced protein (IP) Purification, distribution and possible function. In Steroid Induced Proteins (Edited by M. Beato). Elsevier/North Holland Biomedical Press, Amsterdam (1980) pp. 3-20. 19. Skipper J. K., Eakle S. D. and Hamilton T. H.: Modulation by estrogen of the synthesis of specific uterine proteins. Cell 22 (1980) 69-78. 20. Pennequin P., Robel P. and Baulieu B.: Steroid-induced protein synthesis in rat uterus and prostate. Eur. J. Biochem. 60 (1975) 137-145. 21. Katzenellenbogen B. S. and Leake R. E.: Distribution of the oestrogen-induced protein and of total protein between endometrial and myometrial fractions of the immature and mature rat uterus. J. Endocr. 63 (1974) 439 449. 22. Dupont-Mairesse and Garland P.: Induction of the synthesis of a specific protein (IP) in the myometrium, the stroma and the luminal epithelium of the rat uterus. Endocrinology 96 (1975) 1587-1591. 23. King R. J. B., Somjen D., Kaye A. M. and Lindner H. R.: Stimulation by oestradiol-17fl of specific cytoplasmic and chromosomal protein synthesis in immature rat uterus. Molec. cell. Endocr. 1 (1974) 21 36. 24. Walker M. D. and Kaye A. M.: mRNA for the rat uterine oestrogen-induced protein. J. biol. Chem. 256 (1981) 23-26. 25. Chrambach A.: Electrophoresis and electrofocussing on polyacrylamide gel in the study of native macromolecules. Molec. cell. Bioehem. 29 (1980) 23-46. 26. Peterson J. L. and McConkey E. H.: Non histone chromosomal proteins from HeLa cells. J. biol. Chem. 251 (1976) 548-554.
0022-4731/8353.00+0.00 Copyright (c 1983 Pergamon Press Ltd
Printed in Great Britain. All rights reserved
PROTEIN SYNTHESIS IN THE UTERUS OF THE I M M A T U R E RAT: THE EFFECTS OF SHORT EXPOSURES TO OESTRADIOL-17fl JOANNE M. BEAUMONT* and JOHN T. KNOWLERt Department of Biochemistry, University of Glasgow, Glasgow G12 8QQ, U.K.
(Receired 7 December 1982) Summary--Three methods we,e used to study the effect of oestrogen on the incorporation of radioactive precursor into uterine protein. Intact tissue was incubated in vitro. Isolated uterine epithelial, stromal and myometrial cells were labelled in vitro. Isolated polysomes were translated in cell free protein synthesising systems. In all of these systems, minor qualitative changes in protein synthesis were detected where the uteri were derived from oestrogen-treated rats. These changes were most dramatic in isolated stromal cells and were accompanied by a marked overall increase in protein synthesis. The translation of m RNA on isolated uterine polysomes revealed a sequence of minor, but reproducible, oestradiol-induced changes. It was difficult, however, to relate these changes to those detected in incubated tissue or cells, possibly because the cell free translation products were not subject to normal posttranslational modification and processing.
Assay of total uterine protein synthesis
INTRODUCTION
We have shown that oestradiol-17fl induces profound qualitative and quantitative changes in the m R N A population of the immature rat uterus. O f the m R N A species isolated from the tissue 4 h after oestrogen administration, only 563o were present 2 h after treatment and only 11~o were detected in untreated animals[l]. These dramatic differences were unexpected as few oestrogen-induced proteins have been detected in the rat uterus (for a review see 2). Here we describe an attempt to clarify this ambiguity. We have examined the incorporation of precursor into uterine proteins by the fluorographic detection of radioactive polypeptides fractionated on twodimensional polyacrylamide gels and have analysed protein synthesis in the intact uterus, in isolated uterine ceils and in cell free protein synthesising systems. We confirm that the synthesis of those proteins which are abundant enough to be detected by these techniques, undergo only minor changes in response to oestrogen and we discuss how this relates to our previous findings. MATERIALS AND M E T H O D S
The source and maintenance of immature female rats, the administration of oestradiol-17fl and the in vitro incubation of excised uteri have been previously described. [3].
*Current address: Department of Pharmacy, University of Strathclyde, Royal College, Glasgow G1 1XW, U.K. /To whom all communications should be addressed. Abbreviations: Oestradiol- 17fl- 1,3,5,(10)-estratriene-3,17fldiol.
Total uterine protein was prepared from groups of 6 incubated uteri which were homogenised in 2 ml of a buffer containing 10raM Tris-HC1, pH 7.4, 10raM NaC1, 1.5 m M MgC12. This and subsequent steps were conducted at 0 ~ , C . Insoluble debris was removed by centrifugation at 10,000g for 5 rain in a Sorvall SS34 fixed-angle rotor and the resulting supernatant was precipitated with a 9 vol acetone containing 0.1 M HCI at - 2 0 ° C overnight. The precipitate was pelleted by centrifugation, washed with acetone, dried in a stream of nitrogen and dissolved in lysis buffer[4] for specific activity determination and two-dimensional gel electrophoresis.
Assa)' of protein ~Tnthesis by isolated cells Immature rat uteri were fractionated into epithelial, stromal and myometrial cells as described by M c C o r m a c k and Glasser[5] and were examined for purity and damage by phase contrast and transmission electron microscopy. The latter was performed on cell pellets fixed in glutaraldehyde-osmium tetroxide by I. Montgomery of the Department of Physiology, Glasgow University. Before analysis of protein synthesis, the separated cells were rested for 4 h in Eagles medium [6]. This was necessary because the freshly prepared cells were very inefficient in protein synthesis. The 4 h incubation was the minimum "rest" period required in order that the cells regained a rate of precursor incorporation which was comparable with that of the intact tissue. Rested cells were resuspended in 1 ml Eagles medium lacking methionine but containing 70 # Ci of L-[3H]methionine (800-1450Ci/mmol) and 5~o heat inactivated, charcoal stripped, foetal calf serum. The
1503
1504
JOANNE M. BEAUMONT and JOHN T. KNOWLER
uteri. A very complex pattern of labelled polypeptides, detected as radioactive spots on two dimensional gels [4] was obtained. When rats had received oestradiol at varying times before death, minor changes were observed in the in vitro incorporation of the radioactive amino acids but these were not very reproducible due to partial masking by the complex nature of the protein population (results not shown). Various methods of fractionation were employed in an attempt to reduce the complexity of the protein populations. Nuclear and cytoplasmic protein were isolated from the labelled uterine tissue but this did Assay of cell.li'ee polysomal protein synthesis little to reduce the problem. Cytoplasmic preparaUterine polysomes were prepared as described by tions were still very complex and resulted in patterns Merryweather and Knowler[7]. Cell free translation very similar to those since described in the mouse by mixes contained 8/~1 of micrococcal nuclease-treated Korach et al[ll]. The nuclear protein profile was rabbit reticulocyte lysate (Radiochemical Centre, distinct; however, only one reproducible difference Amersham), 2 ~1 of k-[3H]methionine (5/~Ci//il) and was detected in the proteins synthesised before and l/,l of a polysome preparation. Blanks contained after hormone treatment. A polypeptide of mol. wt sterile distilled water in place of polysomes. Assays 65,000 which focussed at a pH of about 5.2 disapwere incubated at 3 0 C for 90 rain and terminated by peared when rats had been treated with oestradiol for cooling at 0'C. Aliquots of the cell free transaction, varying times before death (results not shown). used to follow the time course and extent of protein McCormack and Glasser[5] have described a techsynthesis, were dried onto 2.5 cm Whatman 3 MM nique for the fractionation of different uterine cell filter discs (Fisons). The discs were then placed in ice types, which retain good cellular morphology and cold 10!I; (w/v) trichloroacetic acid for 10min to viability. We have applied their methods to an analprecipitate the protein, boiled in 10",, (w/v) tri- ysis of protein synthesis in uterine epithelial, stromal chloroacetic acid containing 10mM k-methionine and myometrial cells. Isolated cells were incubated, (Sigma) for 15 min and washed twice in cold 51~i(w/v) as described in the methods section, in 70 #Ci/ml of trichloroacetic acid. Discs were rinsed in ethanol, [3SS]methionine. The radioactive total cellular protein dried and counted in 10 ml 0.5,; w/v was then isolated and fractionated on two dimen2,5-diphenyloxazole in toluene. The remainder of sional polyacrylamide gels. Figure 1 illustrates the each assay was used to analyse the products of results obtained with epithelial and stromal cells translation by two-dimensional gel electrophoresis. isolated from the uteri of control rats and from the uteri of rats which had received oestradiol 4 h before Protein .fi'actionation and analysis death. The pattern of radioactive polypeptides detected The two dimensional fractionation of proteins was essentially as described by O'Farrell[4] except that the with epithelial cell preparations (Fig. la and b) was less complex than total uterine protein and was second dimension, sodium dodecyl sulphate polyacrylamide gels were prepared as described by dominated by a multi-component spot which coLeStourgeon and Beyer[8]. Proteins were assayed by migrated with rat and rabbit skeletal muscle actin. the method of Bramhall et a/[9] which avoided Figure la and lb show the labelled epithelial proteins interference by the ampholines present in lysis of cells derived from control animals and 4 h buffer [4]. Proteins, labelled by incubation in vitro, oestrogen-treated rats respectively. The only proteins were detected on the two-dimensional poly- which consistently appeared in the preparation as a acrylamide gels as described by Bonner and result of hormone treatment are indicated by arrows Laskey[10]. Rat leg muscle actin was the gift ofJ. Elce in Figure lb. Figure lc and ld shows labelled stromal proteins of Department of Biochemistry, Queens University, Kingston, Canada and rabbit skeletal muscle actin derived from control and 4 h oestrogen-treated rats. was prepared from an acetone powder which was the Hormone administration always resulted in a much gilt of J. Dow, Department of Biochemistry, Glasgow higher specific activity in the stromal proteins. Thus, University. Rat brain creatine kinase was the gener- in a typical experiment, the protein from control rats, ous gift of A. Kaye, Weizman Institute of Science, which received carrier only, incorporated radioactive methionine to 7.0 x 102cpm//tg while that of Rehovot, Israel. hormone-treated animals incorporated precursor to 2.23 x l03 cpm//ag. This difference in specific activity RESU LTS was reflected in the density of the spots on the Our initial experiments examined the in vitro incor- fluorograms and virtually all detected proteins were poration of both pSS]methionine and ~4C-mixed synthesised in greater quantities after oestrogen adamino acids into the total protein of immature rat ministration. The effect was more dramatic in some
cells were incubated for 90 min at 37 C under an atmosphere of 95°.,,, O2/5°i; CO> After the incubation, the cells were pelleted at 1000 rpm for 5 minutes in a Beckman T J-6 bench-top centrifuge and the medium discarded, the pellets were washed twice in 5ml phosphate buffered saline and repelleted. Pellets were resuspended in l ml isotonic saline, homogenised, and the solution precipitated with 9 vol of acetone, 0.1 M HCI overnight at - 2 0 C. Precipitates were washed with acetone, dried under nitrogen and dissolved in lysis buffer [4].
Effects of oestrogen on protein synthesis
1505 IEF
SDS
94
67 ,~W x10-3 43 3O
20
Fig. I. Polypeptides synthesised by preparations of uterine cells (a) polypeptides of epithelial cells from the uteri of untreated rats (100/~g--9.6 × 104cpm) (b) polypeptides of epithelial cells from the uteri of rats receiving oestrogen 4 h before death (100 t~g--l.2 × 105 zpm) polypeptides of stromal cells from the uteri of untreated rats (100#g--7 x 104cpm) (d) polypeptides of stromal cells from the uteri of rats receiving oestrogen 4 h before death (65 #g--l.4 x 105cpm).
proteins than others and was particularly striking in two polypeptides, indicated by open arrows in Figure ld, which were in the same positions as those identified as tubulins by Korach et al.[ll]. In addition, a number of newly labelled polypeptides appeared after oestradiol administration and nine, the appearance of which was reproducible, are labelled with solid arrows in Figure l d. Myometrial cell preparations incorporated radioactive amino acids into protein very inefficiently. Specific activities were such that only very few of the most abundant proteins could be detected on fluorograms and no realistic analysis of hormoneinduced changes could be made (results not shown), The efficient labelling of the separated uterine
cells was only possible after they had been allowed to recover from the trauma of isolation by resting in a complete, aerated medium for 4h. Thus, it was possible that the differences between the labelled protein patterns of isolated cell types and those of whole uterine protein stemmed from the labelling technique rather than the source of material. To test this possibility, uterine cell types were also isolated after the proteins had been labelled by incubating the intact tissue in vitro as described above. Analysis of the radioactive polypeptides in the isolated cells gave results very similar to those of Figure 1. Figure 2a-d shows the radioactive protein translation products of polysomes purified from the intact uteri of control, 2, 4 and 12 h treated rats respectively.
1506
JOANNE M. BEAUMONT and JOHN T. KNOWLI~R ...,r-----
IE F
5D5
l
1 9~
67 AAW
×10"3 43
30
20
a
b
d
Fig. 2. Translation products of polysomes purified from the intact uteri of (a) untreated rats (b) rats receiving oestrogen 2 h before death (c) rats receiving oestrogen 4 h before death (d) rats receiving oestrogen 12 h before death.
The most striking feature of this study was the high degree of correspondence between the controls (animals receiving carrier only) and all hormonal time points. Even the most minor spots could be detected consistently. However, hormone-induced changes were seen in the translation products of uterine polysomes from hormone treated rats. Those polypeptides which showed increased incorporation of precursor or which appeared for the first time by 2 h after oestrogen administration are indicated by arrows in Fig. 2b while three proteins which were synthesised in decreased quantities after hormone treatment are arrowed in the control (Fig. 2a). The
right hand of the two polypeptides arrowed in the centre of Fig. 2b probably represents the so called induced protein of Notides and Gorski[12] recently identified as an isoenzyme of creatine kinase by Reiss and Kaye[13]. Purified creatine kinase, kindly provided by A. Kaye, migrated to the same position on a parallel gel. This polypeptide was also detected in fractionations of total uterine protein labelled in t:itro. It was not however unambiguously located in epithelial or stromal cell preparations The changes in proteins synthesised by uterine polysomes from rats given oestradiol 2 h before death were maintained in those receiving 4h treatment
Effects of oestrogen on protein synthesis except that two polypeptides of tool. wt. 49,000 were no longer synthesised is increased quantities. Some further increases in the synthesis of specific polypeptides also occurred at this time and are indicated by arrows in Figure 2c. Proteins made by uterine polysomes isolated from rats receiving oestradiol 12 h before death showed little qualitative change compared with those from 4 h treated animals but, in most cases, maintained the changes occurring between 2 and 4 h (Fig. 2d). DISCUSSION
We have not found it possible to label rat uterine protein in t,h,o to a specific activity adequate for the analysis of individual components of the protein population. For this reason, a number of different approaches were employed by which the effects of hormone treatment in ~,ivo was monitored by assessing the incorporation of precursor into protein in vitro. There is, however, some precedent for believing that the incorporation of precursors in vitro may be abnormal. Thus, although oestrogen-induced hnRNA synthesis can be demonstrated in vitro [14], the stimulated synthesis of rRNA, which in ~:it~o is 10 12-fold control levels, is reduced to 2-3-fold in t~itro and can be explained by precursor uptake [15]. Nevertheless, oestrogen stimulated synthesis of IP [12] and of ribosomal protein (Muller and Knowler, unpublished observations) can be demonstrated in t,itro.
Studies with fractionated cells may solve some problems but create others. Problems of anoxia, which are likely to affect the inner cells of incubated tissue are reduced by cellular fractionation. Preferential uptake of radioactive precursor by the outer serosal cells of the tissue is also avoided. However, the cells are shocked by the trauma of isolation and require a 4 h rest before they will efficiently make protein. That the translation products of shocked cells can be markedly changed is shown by the heat shock response [16]. Response to hormones may also be lost during cellular fractionation. Higgins et al.[l 7] showed that castration markedly decreased the incorporation of methionine into proteins of seminal vesicle tissue but not of isolated, undamaged cells. They suggested that the cells in t'ivo were subject to vectorial constraints such as stromal-epithelial interactions. Notwithstanding the above constraints, the proteins synthesised in t~itro by incubated tissue and by fractionated cells are very similar to each other and between preparations. Qualitatively, there is considerable similarity in the proteins made by epithelial and stromal cells (Fig. 1) but the two cell types are strikingly different in their responses to the hormone. In both cell types there are minor qualitative changes in oestrogen.treated rats but only in the stroma is there a marked quantitative effect in which overall protein synthesis is considerably stimulated. This suggests a differential hypersR 1 9 4
1507
trophy in the stroma at this early stage of the oestrogenic response. If one can assume that mRNA molecules on polysomes are being actively transcribed, then the isolation of polysomes and their continued translation in eitro must most clearly represent protein synthesis in ~,i~'o. This assumes however that the translation in vitro produces full length polypeptides and even these may differ significantly in size and charge from proteins made in intact cells because they lack post-translational processing and modification. The spectrum of proteins made by isolated uterine polysomes show a number of similarities with those made by the tissue or isolated cells but there are also marked differences. There is little if any similarity in the oestrogen-induced proteins detected in the two systems. We are unable at present to identify many of the new polypeptides detected in either system. The two polypeptides arrowed in the centre of Fig. 2b are in the expected positions of the proteins collectively known as IP and including creatine kinase [13], an isoenzyme of enolase[18] and one other polypeptide [19]. The right hand component migrates in the same position as purified brain creatine kinase. We also detect this polypeptide in protein from total uterus and total uterine cytoplasm. It is not seen in nuclear preparations nor can it be definitely assigned in epithelial and stromal ceils. These observations confirm the previously published cytoplasmic locations of IP[20], and suggest that it may be a myometrial protein. However, two other reports[21,22] show that IP is synthesised in epithelial, stromal and myometrial cells. We are unable at the present time to explain these differences. One other oestrogen-induced component, that arrowed beneath actin in Fig. 2b, has been provisionally identified in the radioactive proteins isolated from in t'itro incubated nuclei. It is not, however, present in cytoplasmic proteins and may correspond to the 42,000 tool. wt non-histone protein described by King et a/.[23]. In all of the above systems, the qualitative effects of oestrogen on protein synthesis are minor. This supports the data derived from the two dimensional fractionations of total rat uterine proteins by Walker and Kaye[24] and of the mouse uterine proteins by Korach et al.[l 1]. The patterns of radioactive proteins described by both groups are very similar to our own. Nevertheless, the results are not those which might have been predicted from our previous studies of uterine mRNA populations[l] or from the very quiescent nature of the unstimulated immature rat uterus in which protein synthesis is at a very low level and most ribosomes are in the form of monosomes [7]. Estimates of the maximum resolving power of the two dimensional system range from 2500 5000 different components [4, 25] and O'Farrell[4] has demonstrated the ability of the system to detect more than 1000 bacterial proteins.
1508
JOANNE M. BEAUMONTand JOHN T. KNOWLER
However, in animal cells, relatively few proteins (50-60) may comprise greater than 50°~, o f the total protein mass [26]. This, together with the limitations on protein load and specific activity in the above experiments, mean that we are only likely to detect the m o s t a b u n d a n t o f the uterine proteins. Our previous findings that most oestrogen-induced changes occur in the m R N A species which are rare or o f intermediate a b u n d a n c e may therefore explain our inability to detect m o r e than a few oestrogen-induced changes in the synthesis o f specific proteins.
REFERENCES
1. Aziz S., Balmain A. and Knowler J. T.: Qualitative and quantitative changes in uterine mRNA populations in response to oestradiol treatment of rats. Eur. J. Biochem. 100 (1979) 95 100. 2. Beato M. (ed.): Steroid-Induced Uterine Proteins. Elsevier/North Holland Biomedical Press, Amsterdam (1980). 3. Knowler J. T. and Smellie R. M. S.: The synthesis of RNA in immature rat uterus responding to oestradiol17ft. Biochem. J. 125 (1971) 605-614. 4. O'Farrell P. H.: High resolution two-dimensional electrophoresis of proteins~ J. biol. Chem. 251 (1975) 4007-4021. 5. McCormack S. A. and Glasser S. R.: Differential response of individual uterine cell types from immature rats treated with oestradiol. Endocrinology 106 (1980) | 634-1649. 6. Busby D. W. G., House W. and Macdonald J. R.: In Virological Technique. Churchill, London (1964). 7. Merryweather M. J. and Knowler J. T.: The kinetics of the incorporation of newly synthesised RNA and protein into the ribosomes of the uterus of the oestrogenstimulated immature rat. Biochem. J. 196 (1980) 405-410. 8. LeStourgeon W. M. and Beyer A. L.: The rapid isolation, high-resolution electrophoretic characterisation and purification of nuclear proteins. Meth. cell. Biol. 16 (1977) 387-406. 9. Bramhall S., Noack N., Wu M. and Loewenberg J. R.: A simple colorimetric method for determination of protein. Analyt. Biochem. 31 (1969) 146 148. 10. Bonner W. M. and Laskey R. A.: A film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 46 (1974) 83 88. 11. Korach K. S., Harris S. E. and Carter D. B.: Uterine proteins influenced by estrogen exposure, analysis by two dimensional gel electrophoresis. Molec. cell. Endocr. 21 (1981) 243 254.
12. Notides A. C. and Gorski J.: Oestrogen-induced synthesis of a specific protein. Proc. natn. Acad. Sci., U.S.A. 56 (1966) 230 235. 13. Reiss N. A. and Kaye A. M.: Identification of the major component of the oestrogen-induced protein of rat uterus as the BB isoenzyme of creatine kinase. J. biol. ('hem. 256 (1981) 5741-5749. 14. Knowler J. T.: The incorporation of newly synthesized RNA into nuclear ribonucleoprotein particles after oestrogen administration to immature rats. Eur. J. Biochem. 64 (1976) 161 165. 15. Knowler J. T., Borthwick N. M. and Smellie R. M. S.: Early effects of oestrogen on the transcription of uterine. ribonucleic acid. Biochem. Soc. Trans. 3 (1975) 1177-1180. 16. Storti R. V., Scott M. P., Rich A. and Pardue M. L.: Translational control of protein synthesis in response to heat shock in D. melanogasta cells. Cell 22 (1980) 825 834. 17. Higgins S. J., Brooks D. E. and Fuller F. M.: Isolation of cells from rat seminal vesicles and epididymis and their use in studying androgen action. Molec. cell. Endocr. 23 (1981) 207-223. 18. Kaye A. M. and Reiss N.: The uterine estrogen-induced protein (IP) Purification, distribution and possible function. In Steroid Induced Proteins (Edited by M. Beato). Elsevier/North Holland Biomedical Press, Amsterdam (1980) pp. 3-20. 19. Skipper J. K., Eakle S. D. and Hamilton T. H.: Modulation by estrogen of the synthesis of specific uterine proteins. Cell 22 (1980) 69-78. 20. Pennequin P., Robel P. and Baulieu B.: Steroid-induced protein synthesis in rat uterus and prostate. Eur. J. Biochem. 60 (1975) 137-145. 21. Katzenellenbogen B. S. and Leake R. E.: Distribution of the oestrogen-induced protein and of total protein between endometrial and myometrial fractions of the immature and mature rat uterus. J. Endocr. 63 (1974) 439 449. 22. Dupont-Mairesse and Garland P.: Induction of the synthesis of a specific protein (IP) in the myometrium, the stroma and the luminal epithelium of the rat uterus. Endocrinology 96 (1975) 1587-1591. 23. King R. J. B., Somjen D., Kaye A. M. and Lindner H. R.: Stimulation by oestradiol-17fl of specific cytoplasmic and chromosomal protein synthesis in immature rat uterus. Molec. cell. Endocr. 1 (1974) 21 36. 24. Walker M. D. and Kaye A. M.: mRNA for the rat uterine oestrogen-induced protein. J. biol. Chem. 256 (1981) 23-26. 25. Chrambach A.: Electrophoresis and electrofocussing on polyacrylamide gel in the study of native macromolecules. Molec. cell. Bioehem. 29 (1980) 23-46. 26. Peterson J. L. and McConkey E. H.: Non histone chromosomal proteins from HeLa cells. J. biol. Chem. 251 (1976) 548-554.