166
Biochimica et Biophysica ,4 eta, t t ~97 ( 1991 ) 16¢~. 17!i
~e~ 1991 Elsevier Science Publishers B.V, All rights reserved 0925-4439/91/$113.511 A D O N I S 092544399100115E BBADIS 61070
Glucocorticoid-mediated muscle atrophy: alterations in transcriptional activity of skeletal muscle nuclei G.A.J. G o o d l a d and C.M. Clark Department o f Biochemistry and Microbiology, University of St. Andrews, St. Andrews, Fife ( U.K.)
(Received 7 February 1991)
Key words: Skeletal muscle nucleus; Dexamethasone; RNA polymerase
Glucocorticoid-induced muscle atrophy is associated with a decrease in the level of protein synthesis and a loss of RNA. This paper reports the behaviour of RNA polymerase I- and RNA polymerase II-directed transcription (EC 2.7.7.6) in nuclei isolated from skeletal muscles of rats given a catabolic dose of dexamethasone acetate (5 mg per Kg body weight) over a period of 4 days. Both activities were altered by the dexamethasone treatment. In the case of RNA polymerase 1-mediated transcription there was a loss of template-engaged enzymes indicating the existence of an inhibition of initiation of transcription while the rate of elongation of bound enzymes was unaltered. The number of RNA polymerase ll-chromatin bound enzymes was increased, but the mean polynucleotide elongation rate was reduced. The possibility that giucocorticoids may impair the elongation stage of transcription in skeletal muscle by increasing the frequency of premature termination of transcripts is discussed. No evidence was obtained for any increase in ribonuclease activity in muscle nuclei of dexamethasone-treated animals.
Introduction
Glucocorticoids when administered in excessive amounts or when overproduced by the adrenal cortex have a net anabolic effect on liver while causing atrophy of other tissues, e.g. lymphoid tissue or skeletal muscle. This group of steroids also has a positive action on the regulation of the expression of certain specific genes and a negative action on the expression of others [1]. Indeed the same gene may be affected in opposing directions when present in different tissues [2]. Both positive and negative gene regulation by glucocorticoids involve the initiation stage of transcription where a complex of the steroid and its receptor interact with a regulatory sequence of the gene [3]. However, the number of genes which have been shown to be directly responsive to glucocorticoids is relatively small and thus the activity of more than just those genes may be affected in tissues showing a response to these steroids. There is a fall in the total RNA content of muscles of glucocorticoid-treated rats [4-6] and studies on RNA turnover in skeletal muscle in the whole animal have
Correspondence: G.A.J. Goodlad, Department of Biochemistry and Microbiology, Irvine Building, University of St. Andrews, St. Andrews, Fife, KY16 9AL, Scotland, U.K.
shown that in the case of ribosomal RNA this is due to a decrease in the rate of synthesis and not an increase in degradation [6]. Cytoplasmic levels of ribosomal and mRNA depend on both transcriptional and post-transcriptional processes. An assessment of the state of in vivo transcriptional activity independent of post-transcriptional processing can however be obtained by studying RNA formation in isolated nuclei [7] and the present paper describes an investigation of transcriptional activities directed by RNA polymerases I and II (EC 2.7.7.6) in skeletal muscle nuclei from rats given a catabolic dose of dexamethasone acetate over a 4-day period. Methods
Treatment of animals. Virgin female Wistar rats of mean fasting body weight 200 + 5 g were fed individually for 5 days 14 g Breeding diet (Special Diet Services, Stepfield, Essex, U.K.). After this time animals were divided into two groups and received daily for 4 days either a single subcutaneous injection of 5 mg per Kg body weight of dexamethasone acetate (Sigma, Poole) finely suspended in 0.85% (w/v) NaCI or a subcutaneous injection of 0.85% (w/v) NaCI alone. All rats were housed individually and maintained on a 12-hour light-dark cycle (lights on 07.00 h) at a temper-
167 ature of 22 _+ 2 ° C. Rats continued to receive a 14 g diet daily. Food was given as a single meal at 16.00 h and was consumed completely throughout the experimental period. Animals were killed 24 h after the final injection. Estimation o f protein and nucleic acid content of muscle. Left gastrocnemius muscles were dissected out, weighed and non-collagenous protein, R N A and D N A content estimated as described previously [8]. Assay o f R N A polymerase acticity; The method was based on that of Ljungquist and Astrom [9]. Nuclei from mixed hind-leg muscles were isolated as described [10]. Nuclei equivalent to 70-100 ~ g D N A were incubated in a total vol. of 0.5 ml containing 25 m M Tris-HC1 (pH 7.8 at 25°C), 2 mM MgCI2, 2 m M MnCI 2, 5 m M dithiothreitol, 4% ( v / v ) glycerol, 3 mM phosphocreatine (Sigma, Poole), 0.25 U creatine phosphokinase (BCL, Lewes), 0.6 m M ATP, G T P and CTP and 0.12 m M [ u - I n C ] U T P (0.1 ~Ci per assay) (Amersham International). (NH4)2SO 4 was present at 70 mM. a-Amanitin (BCL, Lewes) was added to differentiate activities associated with the different R N A polymerases present. With the preparations of muscle nuclei used, no increase in inhibition was observed when the concentration of the inhibitor was raised from 2 to 200 /~g per ml indicating, as had been observed previously [10], the absence of any R N A polymerase III activity. 2 p.g per ml a-amanitin was therefore employed to differentiate R N A polymerase I and II activities. In some experiments heparin (Sigma, Poole) was present at a concentration of 1 mg per ml. Free R N A polymerase activities were measured in the presence of 5 0 / z g per ml poly [d(A-T)] (BCL, Lewes) [10]. Reactions were carried out at 37 ° C for periods from 2 - 1 2 min and terminated by the addition of 0.5 ml 20% ( w / v ) trichloracetic acid containing 0.06 M Na4P20 7. The precipitates were collected on glass fibre discs ( G F / A 2.5 cm diam.) and washed 6 times with 5 ml 10% ( w / v ) trichloracetic acid containing 0.01 M N a a P 2 0 7 and 3 times with 6 ml ethanol. Discs were dried at 100 ° C for 1 h and radioactivity estimated by liquid scintillation counting.
The extent of hydrolysis of U T P by muscle nuclei under conditions used for the R N A polymerase assays was determined by separation of acid-soluble uridine and uridine nucleotides on poly(ethyleneimine)-cellulose sheets (Polygram Cell 300 PE1) [11] and estimating their radioactivity by liquid scintillation counting. Estimation o f relative number o f template-engaged R N A polymerase molecules and rate o f polynucleotide chain elongation. The number of bound transcribing enzyme molecules and the rate of polynucleotide chain elongation were estimated essentially as described by Coupar and Chesterton [12]. The amount of nuclei per assay was increased to 350-400 /zg DNA. [3H]UTP (Amersham International) was present at a concentration of 10/~M and a specific activity of 6/~Ci per nmol to ensure measureable radioactivity in uridine obtained from hydrolysis of R N A formed. Incubation temperature was reduced to 21 ° C and the period of incubation was from 30 s to 2 min. The synthesised R N A was digested for 16 h at 3 7 ° C in 0.3 M K O H and the products of digestion separated on poly(ethyleneimine)-cellulose sheets [11] and radioactivity present in uridine and U M P spots measured by scintillation counting. Results Table I shows that administration of 5 mg dexamethasone acetate per Kg body weight daily to rats for 4 days caused marked loss of body and gastrocnemius weight. The steroid treatment had no effect on the total amount of D N A per muscle but there was a fall in both muscle protein and R N A content when these were expressed per mg DNA. The effects of dexamethasone treatment on R N A polymerase transcription in nuclei from mixed hind-leg muscles are shown in Table II. The activities of both R N A polymerases I and II showed decreases of about 33% below control levels in preparations from steroidtreated rats. Heparin has been shown to activate R N A polymerase II transcription in nuclei of several cell types including skeletal muscle by causing an increase
TABLE I The effect of daily administration of dexamethasone acetate at a dose of 5 mg per Kg for 4 days on body and left gastrocnemius weights and on gastrocnemius DATA, RNA and non-collagenous protein content Results are expressed as mean± S.E. of at least six experiments. Differences between means significant at the 0.1% (a) or 0.5% (b) level. (Student's t-test).
Control Dexamethasone-treated
Body weight change (g)
g gastrocnemius per 100 g initial body weight
mg DNA per gastrocnemius per 100 g initial body weight
mg RNA mg DNA
mg protein mg DNA
+ 4.0 + 0.5 - 31.5 + 1.2 a
0.59 ± 0.01 0.39 ± 0.01 a
0.38 + 0.02 0.34 ± 0.01
3.88 ± 0.17 2.70 ± 0.12 a
244 ± 14 184 ± 6 b
168 TABLE II The effect of daily administration to rats of dexamethasone acetate (5 mg / Kg) for 4 days on template-bound RNA polymerase activities in isolated nuclei from skeletal muscle
Nuclei were incubated for 10 min at 37°C in the absence or presence of heparin (1 mg/ml) as described in Methods. Results are expressed as mean_+S.E. Numbers in parenthesis represent the number of animals in each group. Differences between means of preparations from control and dexamethasone-treated rats significant at 2% (a), 0.5% (b), levels. (Students' t-test). Heparin
RNA polymerase activity (pmol UMP incorporated per 10 min per 100 ~zg DNA) I
-
(6)
+
(4)
II
Control
Dexameth- Control Dexamethasone asone
41_+2 48-+1
28+_4 a 34-+2 h
26-+1 133_+3
17_+2 b
123_+6
in the elongation rate of the transcript [10,12,13]. When nuclei from muscles of control and dexamethasonetreated animals were assayed in the presence of the proteoglycan both preparations showed several-fold increases in R N A polymerase II activity and the difference between the two was no longer apparent. Heparin had no effect on the activity of R N A polymerase I with either set of nuclei. Although glucocorticoid administration has been reported to decrease ribonuclease activity in skeletal muscle [14], the possibility that the observed falls in R N A polymerase activities might be due to increased hydrolysis of the nascent transcripts by endogenous ribonuclease in the nuclei was examined. Nuclei from control and dexamethasone-treated animals were incubated under standard conditions for 5 min except that a-amanitin was omitted. Actinomycin D and a-amanitin were then added to some of the incubations to inhibit any further R N A synthesis and all reactions allowed to continue for a further 7 min. Fig. 1 shows that in the presence of inhibitors no difference in the loss of radioactivity from newly synthesised R N A in the two groups of nuclei occurred (Fig. 1). The amounts of [3H]UTP remaining in the incubation medium after 10 min were not significantly different in the two sets of preparations (results not shown) indicating that decreased incorporation of nucleotide into R N A following dexamethasone administration could not be attributed to an increased hydrolysis of the nucleotide. The existence of a pool of R N A polymerase molecules which are not bound to the chromatin and therefore not actively engaged in transcription has been demonstrated in several tissues including skeletal muscle [10,15,16] and alteration in activity of this pool in response to various stimuli has been reported [16]. In
the present study however dexamethasone administration had no effect on the levels of activity of free R N A polymerase I or II when poly[d(A-T)] was used as template and transcription of chromatin was blocked by actinomycin D (results not shown). The alterations observed in transcriptional activity of nuclei from corticosteroid-treated rats could be due to changes in either the initiation or elongation stages of the transcription process. An estimate of the relative number of enzyme molecules which were bound to their template at the time of isolation can be obtained by hydrolysing the newly synthesised R N A formed during incubation with alkali and estimating the radioactivity found in uridine residues. Each transcript should yield one nucleoside residue from its 3'terminus. To ensure that there was no additional production of uridine by the action of endogenous nuclease activity the temperature of the incubation was reduced to 21° C. At this temperature the amounts of [3H]uridine recovered from alkaline digestion of transcripts were found to be independent of the incubation time over the period 30 s to 2 rain. All subsequent incubations were therefore of 1 min duration. Table III shows that [3H]uridine resulting from alkaline hydrolysis of transcripts in nuclei from dexamethasone-treated rats was lower than that obtained from transcripts in nuclei from control animals when incubations were carried out in the presence of a-amanitin, indicating that the steroid treatment had caused a reduction in number of actively transcribing R N A polymerase I molecules. On the otherhand, the number of transcribing R N A poly-
80
60
©
........ 40
.
.
.
.
.
.
.
9__.,. .
.
.
- - °
.
20
.
.
,
|
l
, l 5
|
l
l
.
h . 10
*
l
15
Time min
Fig. 1. Time-course of RNA synthesis and degradation of nascent RNA by skeletal muscle nuclei. Nuclei (equivalent to 70 /~g DNA) from control (o) or dexamethasone-treated (o) rats were incubated at 37°C in the absence of a-amanatin. At 5 min 2 /zg per ml a-amanitin and 50/zg per ml actinomycin D were added to some of the incubations (dashed lines) and all incubations were continued for a further 7 min. Results are the mean of two experiments.
169 T A B L E III
Incorporation of radioactivity from [3H]UTP into internal residues and 3'-termini of RNA by skeletal muscle nuclei from control and dexamethasone-treated rats Nuclei (350 p.g D N A ) incubated for 1 rain at 21 ° C. Radioactivity from [3H]UTP incorporated into internal U M P and terminal uridine residues estimated as described in Methods. Results are the M e a n + S.E. of four experiments. Differences between m e a n s significant at the 5% (a), 2% (b) and I% (c) level. pmol incorporated per mg D N A [ 3H ] U M P
Control Dexamethasone-treated
[ 3H]uridine
[ 3H ] U M P / [ 3H]uridine
I
II
I
II
1
II
8.02 +0.45 5.78 c +0.36
11.41 +0.33 13.15 a -+0.56
0.259 +0,013 0.203 b -+0,009
0.097 _+0.006 0.134 c _+0.006
31.02+1.53 28.43--+ 1.09
118.70 -+4.17 97.91 c _+0.56
merase II molecules showed an increase in nuclei of rats given dexamethasone. The mean polynucleotide chain elongation rate as measured by the ratio of radioactivity incorporated into UMP of internal residues to the radioactivity found in uridine was not altered by dexamethasone treatment in the case of R N A polymerase 1-mediated transcription. R N A polymerase II-mediated transcription measured at 21°C showed a small rise in UMP incorporation, however when this was related to the number of chains initiated there was a reduction in the elongation rate (Table III). Discussion
Previous work on the effect of glucocorticoid administration on the turnover of pre-labelled r R N A in skeletal muscle has shown that the loss in R N A due to these steroids can be attributed to a decrease in the rate of synthesis; the rate of degradation showing an actual decrease compared with that in normal animals [6]. The present studies show that the decrease in rRNA synthesis is associated with a decrease in R N A polymerase I activity. Inhibition of ribosomal R N A synthesis at the transcriptional level has been observed in certain other tissues in response to glucocorticoid action. When thymocytes are incubated in the presence of dexamethasone there is a loss of R N A which has been attributed to an impairment of the elongation stage of the process, the number of transcribing enzyme molecules remaining unaffected [17]. In lymphosarcoma cells growing in the presence of glucocorticoids R N A polymerase I activity is reduced due to a defect at the initiation stage, in this case the polynucleotide elongation rate is not affected [18]. The response of R N A polymerase I in skeletal muscle to glucocorticoids therefore appears similar to that of lymphosarcoma cells in that it can be attributed to a defect at initiation which causes a decrease in the number of actively transcribing enzyme molecules and no change in the mean polynucleotide chain elongation rate (Table III). The decrease in initiation observed in
lymphosarcoma cells exposed to dexamethasone has been shown to be due to a loss in activity of the RNA polymerase I initiation factor TFIC [18]. Whether this is the reason for the effect observed in skeletal muscle cannot be determined from the present work, but the absence of a fall in free R N A polymerase I activity suggests that the decreased initiation is not due to a lack of enzyme molecules. The data obtained in the present work show that glucocorticoid treatment causes both an increase in the number of RNA polymerase II molecules bound to the chromatin template and a decrease in the mean polynucleotide chain elongation rate in skeletal muscle (Table III). The net result of these two effects is a small increase in the observed incorporation of UMP into RNA. However when the temperature is raised to 37 o C and the duration of incubation is increased to 10 min a 35% decrease in R N A polymerase II-mediated R N A synthesis compared to control level is observed in nuclei of the treated animals (Table II). No such temperature dependence of the effect of dexamethasone on RNA polymerase I-mediated R N A synthesis is apparent; similar relative decreases in this activity are observed at both incubation temperatures in nuclei from the steroid-treated animals (Tables II and III). From studies with several systems it is becoming evident that gene expression in eukaryotic cells can be regulated at the elongation stage as well as at the initiation phase of transcription. Attenuation sites have been demonstrated at promoter proximal regions of the chromatin where premature termination may occur [13,19,20,21]. It is therefore postulated that the effect of dexamethasone on elongation by R N A polymerase II in muscle can be explained by alterations in chromatin structure which bring about an increase in premature termination. When nuclei are incubated at 37 ° C for 10 rain more polymerase molecules would be able to progress along the template to attenuation sites and any effect of the steroid of this nature would be more pronounced. This possibility is also supported by the fact that when heparin is present in the incubation medium RNA polymerase II-directed RNA synthesis
170 increases several-fold and there is no significant difference between activities in nuclei from control and dexamethasone-treated rats (Table II). Heparin which removes almost all proteins from chromatin while leaving bound RNA polymerase II in a transcriptionally active state [22] has been shown to overcome the attenuation of elongation which has been demonstrated in cultured cells and also the premature termination caused by the exposure of cells to the adenosine analogue, 5,6-dichloro- 1-/3-D-ribofuranosylbenzimidazole [13]. We have previously observed that transcription by R N A polymerase II is impaired in nuclei isolated from skeletal muscle in the tumour-bearing animal where there is also a loss of protein and R N A [10]. Here too the defect occurs at the elongation stage and is not apparent when heparin is added to the incubation system. These findings together with those in the present paper suggest that there is a mechanism in skeletal muscle which modulates transcription independently of initiation and which is activated in situations associated with muscle protein depletion. Acknowledgements This work was supported by grants from the Smith Kline Foundation and The University of St. Andrews Research Committee. We are grateful to Mr. Andrew Oliver for the maintenance of the animals used in this work.
References 1 Rousseau, G.G. (1984) Biochem. J. 224, 1-12. 2 Nechushtan, H., Benvenisty, N., Brandeis, R. and ResheL L. (1987) Nuc. Acids Res. 15, 6405-6417. 3 Burnstein, K.L. and Cidlowski, J.A. (1989) Annu. Rev. Physiol. 51,683-689. 4 Rannels, S.R. and Jefferson, L.S. (1980) Am. J. Physiol. 238 (Endocrinol. Metab. 1), E564-E572. 5 Odedra, B.R. and Millward, D.J. (1982) Biochem. J. 204, 663-672. 6 Goodlad, G.A.J. and Onyezili, F.N. (1981) Biochem. Med. 25, 34-47. 7 Greenberg, M.E. and Ziff, E.B. (1984) Nature 311,433-438. 8 Clark, C.M. and G ooodlad, G.A.J. (1971) Eur. J. Cancer 7, 3-9. 9 Ljungquist, I. and Astrom, S. (1984) Int. J. Biochem. 16, 69-74. 10 Goodlad, G.A.J. and Clark, C.M. (1988) Biochim. Biophys. Acta 950, 296-347. 11 Randerath, K. and Randerath, E. (1969) Methods Enzymol. 12A, 323-302. 12 Coupar, B.E.H. and Chesterton, C.J. (1977) Eur. J. Biochem. 79, 525-533. 13 Tamm, I. and Kikuchi, T. (1979) Proc. Natl. Acad. Sci. 76, 5750-5754. 14 Kreshner, R.M. and Meyer, W.L. (1976) Biochem. Biophys. Res. Commun. 70, 513-518. 15 Yu, F.L. (1975) Biochem. Biophys. Res. Commun. 64, 1109-1115. 16 Sentenac, A. (1985) C.R.C. Crit. Rev. Biochem. 18, 31-90. 17 Dembinski, T.C. (1984) FEBS Lett. 173, 129-133. 18 Cavanaugh, A.J. and Thompson, E.A., Jr. (1985) Nuc. Acids Res. 13, 3357-3369. 19 Nepveu, A,, Marcu, K.B., Skoultchi, A.I. and Lachman, H.M. (1987) Genes and Development 1,938-945. 20 Watson, R.J. (1988) Oncogene 2, 267-272. 21 Chodosh, L.A., Fire, A., Samuels, M. and Sharp, P.A. (1989) J. Biol. Chem. 264, 2250-2257. 22 Choder, M. and Aloni, Y. (1988)J. Biol. Chem. 263, 12994-13002.
Biochimica et Biophysica ,4 eta, t t ~97 ( 1991 ) 16¢~. 17!i
~e~ 1991 Elsevier Science Publishers B.V, All rights reserved 0925-4439/91/$113.511 A D O N I S 092544399100115E BBADIS 61070
Glucocorticoid-mediated muscle atrophy: alterations in transcriptional activity of skeletal muscle nuclei G.A.J. G o o d l a d and C.M. Clark Department o f Biochemistry and Microbiology, University of St. Andrews, St. Andrews, Fife ( U.K.)
(Received 7 February 1991)
Key words: Skeletal muscle nucleus; Dexamethasone; RNA polymerase
Glucocorticoid-induced muscle atrophy is associated with a decrease in the level of protein synthesis and a loss of RNA. This paper reports the behaviour of RNA polymerase I- and RNA polymerase II-directed transcription (EC 2.7.7.6) in nuclei isolated from skeletal muscles of rats given a catabolic dose of dexamethasone acetate (5 mg per Kg body weight) over a period of 4 days. Both activities were altered by the dexamethasone treatment. In the case of RNA polymerase 1-mediated transcription there was a loss of template-engaged enzymes indicating the existence of an inhibition of initiation of transcription while the rate of elongation of bound enzymes was unaltered. The number of RNA polymerase ll-chromatin bound enzymes was increased, but the mean polynucleotide elongation rate was reduced. The possibility that giucocorticoids may impair the elongation stage of transcription in skeletal muscle by increasing the frequency of premature termination of transcripts is discussed. No evidence was obtained for any increase in ribonuclease activity in muscle nuclei of dexamethasone-treated animals.
Introduction
Glucocorticoids when administered in excessive amounts or when overproduced by the adrenal cortex have a net anabolic effect on liver while causing atrophy of other tissues, e.g. lymphoid tissue or skeletal muscle. This group of steroids also has a positive action on the regulation of the expression of certain specific genes and a negative action on the expression of others [1]. Indeed the same gene may be affected in opposing directions when present in different tissues [2]. Both positive and negative gene regulation by glucocorticoids involve the initiation stage of transcription where a complex of the steroid and its receptor interact with a regulatory sequence of the gene [3]. However, the number of genes which have been shown to be directly responsive to glucocorticoids is relatively small and thus the activity of more than just those genes may be affected in tissues showing a response to these steroids. There is a fall in the total RNA content of muscles of glucocorticoid-treated rats [4-6] and studies on RNA turnover in skeletal muscle in the whole animal have
Correspondence: G.A.J. Goodlad, Department of Biochemistry and Microbiology, Irvine Building, University of St. Andrews, St. Andrews, Fife, KY16 9AL, Scotland, U.K.
shown that in the case of ribosomal RNA this is due to a decrease in the rate of synthesis and not an increase in degradation [6]. Cytoplasmic levels of ribosomal and mRNA depend on both transcriptional and post-transcriptional processes. An assessment of the state of in vivo transcriptional activity independent of post-transcriptional processing can however be obtained by studying RNA formation in isolated nuclei [7] and the present paper describes an investigation of transcriptional activities directed by RNA polymerases I and II (EC 2.7.7.6) in skeletal muscle nuclei from rats given a catabolic dose of dexamethasone acetate over a 4-day period. Methods
Treatment of animals. Virgin female Wistar rats of mean fasting body weight 200 + 5 g were fed individually for 5 days 14 g Breeding diet (Special Diet Services, Stepfield, Essex, U.K.). After this time animals were divided into two groups and received daily for 4 days either a single subcutaneous injection of 5 mg per Kg body weight of dexamethasone acetate (Sigma, Poole) finely suspended in 0.85% (w/v) NaCI or a subcutaneous injection of 0.85% (w/v) NaCI alone. All rats were housed individually and maintained on a 12-hour light-dark cycle (lights on 07.00 h) at a temper-
167 ature of 22 _+ 2 ° C. Rats continued to receive a 14 g diet daily. Food was given as a single meal at 16.00 h and was consumed completely throughout the experimental period. Animals were killed 24 h after the final injection. Estimation o f protein and nucleic acid content of muscle. Left gastrocnemius muscles were dissected out, weighed and non-collagenous protein, R N A and D N A content estimated as described previously [8]. Assay o f R N A polymerase acticity; The method was based on that of Ljungquist and Astrom [9]. Nuclei from mixed hind-leg muscles were isolated as described [10]. Nuclei equivalent to 70-100 ~ g D N A were incubated in a total vol. of 0.5 ml containing 25 m M Tris-HC1 (pH 7.8 at 25°C), 2 mM MgCI2, 2 m M MnCI 2, 5 m M dithiothreitol, 4% ( v / v ) glycerol, 3 mM phosphocreatine (Sigma, Poole), 0.25 U creatine phosphokinase (BCL, Lewes), 0.6 m M ATP, G T P and CTP and 0.12 m M [ u - I n C ] U T P (0.1 ~Ci per assay) (Amersham International). (NH4)2SO 4 was present at 70 mM. a-Amanitin (BCL, Lewes) was added to differentiate activities associated with the different R N A polymerases present. With the preparations of muscle nuclei used, no increase in inhibition was observed when the concentration of the inhibitor was raised from 2 to 200 /~g per ml indicating, as had been observed previously [10], the absence of any R N A polymerase III activity. 2 p.g per ml a-amanitin was therefore employed to differentiate R N A polymerase I and II activities. In some experiments heparin (Sigma, Poole) was present at a concentration of 1 mg per ml. Free R N A polymerase activities were measured in the presence of 5 0 / z g per ml poly [d(A-T)] (BCL, Lewes) [10]. Reactions were carried out at 37 ° C for periods from 2 - 1 2 min and terminated by the addition of 0.5 ml 20% ( w / v ) trichloracetic acid containing 0.06 M Na4P20 7. The precipitates were collected on glass fibre discs ( G F / A 2.5 cm diam.) and washed 6 times with 5 ml 10% ( w / v ) trichloracetic acid containing 0.01 M N a a P 2 0 7 and 3 times with 6 ml ethanol. Discs were dried at 100 ° C for 1 h and radioactivity estimated by liquid scintillation counting.
The extent of hydrolysis of U T P by muscle nuclei under conditions used for the R N A polymerase assays was determined by separation of acid-soluble uridine and uridine nucleotides on poly(ethyleneimine)-cellulose sheets (Polygram Cell 300 PE1) [11] and estimating their radioactivity by liquid scintillation counting. Estimation o f relative number o f template-engaged R N A polymerase molecules and rate o f polynucleotide chain elongation. The number of bound transcribing enzyme molecules and the rate of polynucleotide chain elongation were estimated essentially as described by Coupar and Chesterton [12]. The amount of nuclei per assay was increased to 350-400 /zg DNA. [3H]UTP (Amersham International) was present at a concentration of 10/~M and a specific activity of 6/~Ci per nmol to ensure measureable radioactivity in uridine obtained from hydrolysis of R N A formed. Incubation temperature was reduced to 21 ° C and the period of incubation was from 30 s to 2 min. The synthesised R N A was digested for 16 h at 3 7 ° C in 0.3 M K O H and the products of digestion separated on poly(ethyleneimine)-cellulose sheets [11] and radioactivity present in uridine and U M P spots measured by scintillation counting. Results Table I shows that administration of 5 mg dexamethasone acetate per Kg body weight daily to rats for 4 days caused marked loss of body and gastrocnemius weight. The steroid treatment had no effect on the total amount of D N A per muscle but there was a fall in both muscle protein and R N A content when these were expressed per mg DNA. The effects of dexamethasone treatment on R N A polymerase transcription in nuclei from mixed hind-leg muscles are shown in Table II. The activities of both R N A polymerases I and II showed decreases of about 33% below control levels in preparations from steroidtreated rats. Heparin has been shown to activate R N A polymerase II transcription in nuclei of several cell types including skeletal muscle by causing an increase
TABLE I The effect of daily administration of dexamethasone acetate at a dose of 5 mg per Kg for 4 days on body and left gastrocnemius weights and on gastrocnemius DATA, RNA and non-collagenous protein content Results are expressed as mean± S.E. of at least six experiments. Differences between means significant at the 0.1% (a) or 0.5% (b) level. (Student's t-test).
Control Dexamethasone-treated
Body weight change (g)
g gastrocnemius per 100 g initial body weight
mg DNA per gastrocnemius per 100 g initial body weight
mg RNA mg DNA
mg protein mg DNA
+ 4.0 + 0.5 - 31.5 + 1.2 a
0.59 ± 0.01 0.39 ± 0.01 a
0.38 + 0.02 0.34 ± 0.01
3.88 ± 0.17 2.70 ± 0.12 a
244 ± 14 184 ± 6 b
168 TABLE II The effect of daily administration to rats of dexamethasone acetate (5 mg / Kg) for 4 days on template-bound RNA polymerase activities in isolated nuclei from skeletal muscle
Nuclei were incubated for 10 min at 37°C in the absence or presence of heparin (1 mg/ml) as described in Methods. Results are expressed as mean_+S.E. Numbers in parenthesis represent the number of animals in each group. Differences between means of preparations from control and dexamethasone-treated rats significant at 2% (a), 0.5% (b), levels. (Students' t-test). Heparin
RNA polymerase activity (pmol UMP incorporated per 10 min per 100 ~zg DNA) I
-
(6)
+
(4)
II
Control
Dexameth- Control Dexamethasone asone
41_+2 48-+1
28+_4 a 34-+2 h
26-+1 133_+3
17_+2 b
123_+6
in the elongation rate of the transcript [10,12,13]. When nuclei from muscles of control and dexamethasonetreated animals were assayed in the presence of the proteoglycan both preparations showed several-fold increases in R N A polymerase II activity and the difference between the two was no longer apparent. Heparin had no effect on the activity of R N A polymerase I with either set of nuclei. Although glucocorticoid administration has been reported to decrease ribonuclease activity in skeletal muscle [14], the possibility that the observed falls in R N A polymerase activities might be due to increased hydrolysis of the nascent transcripts by endogenous ribonuclease in the nuclei was examined. Nuclei from control and dexamethasone-treated animals were incubated under standard conditions for 5 min except that a-amanitin was omitted. Actinomycin D and a-amanitin were then added to some of the incubations to inhibit any further R N A synthesis and all reactions allowed to continue for a further 7 min. Fig. 1 shows that in the presence of inhibitors no difference in the loss of radioactivity from newly synthesised R N A in the two groups of nuclei occurred (Fig. 1). The amounts of [3H]UTP remaining in the incubation medium after 10 min were not significantly different in the two sets of preparations (results not shown) indicating that decreased incorporation of nucleotide into R N A following dexamethasone administration could not be attributed to an increased hydrolysis of the nucleotide. The existence of a pool of R N A polymerase molecules which are not bound to the chromatin and therefore not actively engaged in transcription has been demonstrated in several tissues including skeletal muscle [10,15,16] and alteration in activity of this pool in response to various stimuli has been reported [16]. In
the present study however dexamethasone administration had no effect on the levels of activity of free R N A polymerase I or II when poly[d(A-T)] was used as template and transcription of chromatin was blocked by actinomycin D (results not shown). The alterations observed in transcriptional activity of nuclei from corticosteroid-treated rats could be due to changes in either the initiation or elongation stages of the transcription process. An estimate of the relative number of enzyme molecules which were bound to their template at the time of isolation can be obtained by hydrolysing the newly synthesised R N A formed during incubation with alkali and estimating the radioactivity found in uridine residues. Each transcript should yield one nucleoside residue from its 3'terminus. To ensure that there was no additional production of uridine by the action of endogenous nuclease activity the temperature of the incubation was reduced to 21° C. At this temperature the amounts of [3H]uridine recovered from alkaline digestion of transcripts were found to be independent of the incubation time over the period 30 s to 2 rain. All subsequent incubations were therefore of 1 min duration. Table III shows that [3H]uridine resulting from alkaline hydrolysis of transcripts in nuclei from dexamethasone-treated rats was lower than that obtained from transcripts in nuclei from control animals when incubations were carried out in the presence of a-amanitin, indicating that the steroid treatment had caused a reduction in number of actively transcribing R N A polymerase I molecules. On the otherhand, the number of transcribing R N A poly-
80
60
©
........ 40
.
.
.
.
.
.
.
9__.,. .
.
.
- - °
.
20
.
.
,
|
l
, l 5
|
l
l
.
h . 10
*
l
15
Time min
Fig. 1. Time-course of RNA synthesis and degradation of nascent RNA by skeletal muscle nuclei. Nuclei (equivalent to 70 /~g DNA) from control (o) or dexamethasone-treated (o) rats were incubated at 37°C in the absence of a-amanatin. At 5 min 2 /zg per ml a-amanitin and 50/zg per ml actinomycin D were added to some of the incubations (dashed lines) and all incubations were continued for a further 7 min. Results are the mean of two experiments.
169 T A B L E III
Incorporation of radioactivity from [3H]UTP into internal residues and 3'-termini of RNA by skeletal muscle nuclei from control and dexamethasone-treated rats Nuclei (350 p.g D N A ) incubated for 1 rain at 21 ° C. Radioactivity from [3H]UTP incorporated into internal U M P and terminal uridine residues estimated as described in Methods. Results are the M e a n + S.E. of four experiments. Differences between m e a n s significant at the 5% (a), 2% (b) and I% (c) level. pmol incorporated per mg D N A [ 3H ] U M P
Control Dexamethasone-treated
[ 3H]uridine
[ 3H ] U M P / [ 3H]uridine
I
II
I
II
1
II
8.02 +0.45 5.78 c +0.36
11.41 +0.33 13.15 a -+0.56
0.259 +0,013 0.203 b -+0,009
0.097 _+0.006 0.134 c _+0.006
31.02+1.53 28.43--+ 1.09
118.70 -+4.17 97.91 c _+0.56
merase II molecules showed an increase in nuclei of rats given dexamethasone. The mean polynucleotide chain elongation rate as measured by the ratio of radioactivity incorporated into UMP of internal residues to the radioactivity found in uridine was not altered by dexamethasone treatment in the case of R N A polymerase 1-mediated transcription. R N A polymerase II-mediated transcription measured at 21°C showed a small rise in UMP incorporation, however when this was related to the number of chains initiated there was a reduction in the elongation rate (Table III). Discussion
Previous work on the effect of glucocorticoid administration on the turnover of pre-labelled r R N A in skeletal muscle has shown that the loss in R N A due to these steroids can be attributed to a decrease in the rate of synthesis; the rate of degradation showing an actual decrease compared with that in normal animals [6]. The present studies show that the decrease in rRNA synthesis is associated with a decrease in R N A polymerase I activity. Inhibition of ribosomal R N A synthesis at the transcriptional level has been observed in certain other tissues in response to glucocorticoid action. When thymocytes are incubated in the presence of dexamethasone there is a loss of R N A which has been attributed to an impairment of the elongation stage of the process, the number of transcribing enzyme molecules remaining unaffected [17]. In lymphosarcoma cells growing in the presence of glucocorticoids R N A polymerase I activity is reduced due to a defect at the initiation stage, in this case the polynucleotide elongation rate is not affected [18]. The response of R N A polymerase I in skeletal muscle to glucocorticoids therefore appears similar to that of lymphosarcoma cells in that it can be attributed to a defect at initiation which causes a decrease in the number of actively transcribing enzyme molecules and no change in the mean polynucleotide chain elongation rate (Table III). The decrease in initiation observed in
lymphosarcoma cells exposed to dexamethasone has been shown to be due to a loss in activity of the RNA polymerase I initiation factor TFIC [18]. Whether this is the reason for the effect observed in skeletal muscle cannot be determined from the present work, but the absence of a fall in free R N A polymerase I activity suggests that the decreased initiation is not due to a lack of enzyme molecules. The data obtained in the present work show that glucocorticoid treatment causes both an increase in the number of RNA polymerase II molecules bound to the chromatin template and a decrease in the mean polynucleotide chain elongation rate in skeletal muscle (Table III). The net result of these two effects is a small increase in the observed incorporation of UMP into RNA. However when the temperature is raised to 37 o C and the duration of incubation is increased to 10 min a 35% decrease in R N A polymerase II-mediated R N A synthesis compared to control level is observed in nuclei of the treated animals (Table II). No such temperature dependence of the effect of dexamethasone on RNA polymerase I-mediated R N A synthesis is apparent; similar relative decreases in this activity are observed at both incubation temperatures in nuclei from the steroid-treated animals (Tables II and III). From studies with several systems it is becoming evident that gene expression in eukaryotic cells can be regulated at the elongation stage as well as at the initiation phase of transcription. Attenuation sites have been demonstrated at promoter proximal regions of the chromatin where premature termination may occur [13,19,20,21]. It is therefore postulated that the effect of dexamethasone on elongation by R N A polymerase II in muscle can be explained by alterations in chromatin structure which bring about an increase in premature termination. When nuclei are incubated at 37 ° C for 10 rain more polymerase molecules would be able to progress along the template to attenuation sites and any effect of the steroid of this nature would be more pronounced. This possibility is also supported by the fact that when heparin is present in the incubation medium RNA polymerase II-directed RNA synthesis
170 increases several-fold and there is no significant difference between activities in nuclei from control and dexamethasone-treated rats (Table II). Heparin which removes almost all proteins from chromatin while leaving bound RNA polymerase II in a transcriptionally active state [22] has been shown to overcome the attenuation of elongation which has been demonstrated in cultured cells and also the premature termination caused by the exposure of cells to the adenosine analogue, 5,6-dichloro- 1-/3-D-ribofuranosylbenzimidazole [13]. We have previously observed that transcription by R N A polymerase II is impaired in nuclei isolated from skeletal muscle in the tumour-bearing animal where there is also a loss of protein and R N A [10]. Here too the defect occurs at the elongation stage and is not apparent when heparin is added to the incubation system. These findings together with those in the present paper suggest that there is a mechanism in skeletal muscle which modulates transcription independently of initiation and which is activated in situations associated with muscle protein depletion. Acknowledgements This work was supported by grants from the Smith Kline Foundation and The University of St. Andrews Research Committee. We are grateful to Mr. Andrew Oliver for the maintenance of the animals used in this work.
References 1 Rousseau, G.G. (1984) Biochem. J. 224, 1-12. 2 Nechushtan, H., Benvenisty, N., Brandeis, R. and ResheL L. (1987) Nuc. Acids Res. 15, 6405-6417. 3 Burnstein, K.L. and Cidlowski, J.A. (1989) Annu. Rev. Physiol. 51,683-689. 4 Rannels, S.R. and Jefferson, L.S. (1980) Am. J. Physiol. 238 (Endocrinol. Metab. 1), E564-E572. 5 Odedra, B.R. and Millward, D.J. (1982) Biochem. J. 204, 663-672. 6 Goodlad, G.A.J. and Onyezili, F.N. (1981) Biochem. Med. 25, 34-47. 7 Greenberg, M.E. and Ziff, E.B. (1984) Nature 311,433-438. 8 Clark, C.M. and G ooodlad, G.A.J. (1971) Eur. J. Cancer 7, 3-9. 9 Ljungquist, I. and Astrom, S. (1984) Int. J. Biochem. 16, 69-74. 10 Goodlad, G.A.J. and Clark, C.M. (1988) Biochim. Biophys. Acta 950, 296-347. 11 Randerath, K. and Randerath, E. (1969) Methods Enzymol. 12A, 323-302. 12 Coupar, B.E.H. and Chesterton, C.J. (1977) Eur. J. Biochem. 79, 525-533. 13 Tamm, I. and Kikuchi, T. (1979) Proc. Natl. Acad. Sci. 76, 5750-5754. 14 Kreshner, R.M. and Meyer, W.L. (1976) Biochem. Biophys. Res. Commun. 70, 513-518. 15 Yu, F.L. (1975) Biochem. Biophys. Res. Commun. 64, 1109-1115. 16 Sentenac, A. (1985) C.R.C. Crit. Rev. Biochem. 18, 31-90. 17 Dembinski, T.C. (1984) FEBS Lett. 173, 129-133. 18 Cavanaugh, A.J. and Thompson, E.A., Jr. (1985) Nuc. Acids Res. 13, 3357-3369. 19 Nepveu, A,, Marcu, K.B., Skoultchi, A.I. and Lachman, H.M. (1987) Genes and Development 1,938-945. 20 Watson, R.J. (1988) Oncogene 2, 267-272. 21 Chodosh, L.A., Fire, A., Samuels, M. and Sharp, P.A. (1989) J. Biol. Chem. 264, 2250-2257. 22 Choder, M. and Aloni, Y. (1988)J. Biol. Chem. 263, 12994-13002.
