Eur. J. Biochem. 62, 109-115 (1976)
Effect of Triiodothyronine on Rat Liver Chromatin Protein Kinase Jacques KRUH and Lydie TICHONICKY Institut de Pathologie Moleculaire, Paris (Received June 1610ctober 23, 1975)
1. Injection of triiodothyronine to rats stimulates protein kinase activity in liver chromatin nonhistone proteins. A significant increase was found after two daily injections. A 4-fold increase was observed with the purified enzyme after eight daily injections of the hormone. No variations were observed in cytosol protein kinase activity. Electrophoretic pattern, effect of heat denaturation, effect ofp-hydroxymercuribenzoate seem to indicate that the enzyme present in treated rats is not identical to the enzyme in control animals, which suggests that thyroid hormone has induced nuclear protein kinase. Diiodothyronine, 3,3’,5’-triiodothyroninehave no effect on protein kinase. 2. Chromatin non-histone proteins isolated from rats injected with triiodothyronine incorporated more 32Pwhen incubated with [p3’P]ATP than the chromatin proteins from untreated rats. Thyroidectomy reduced the in vitro 32Pincorporation. It is suggested that some of the biological activity of thyroid hormone could be mediated through its effect on chromatin non-histone proteins.
The role of thyroid hormone in cell growth and differentiation is well-documented: triiodothyronine is able to induce several enzymes prematurely in liver and brain [l]. It is then very likely that this hormone acts partly at the gene level. Injection of thyroid hormone stimulates the rate of RNA synthesis [2,3], essentially ribosomal RNA [4]. It enhances the activity of polymerase I [ 5 ] . It has recently been shown that it also induces an early and transient activation of polymerase I1 [6]. Triiodothyronine stimulates protein synthesis, mainly at the level of ribosomes, it increases the amount of newly formed ribosomes and the incorporation of amino acids in vitro [3,7]. The hormone also increases the concentration of chromatin non-histone proteins [6]. It has been recently established that triiodothyronine is able to bind to nuclear proteins [8,9] and more precisely to chromatin non-histone proteins [lo]. These proteins are able to activate gene transcription [11 - 161. This gene activation may be related to the physiological effect of thyroid hormone. In previous works, we have purified from liver and heart chromatin proteins, protein kinases active with histone, nuclear phosphoproteins [12,13], casein and phosvitin [17,18]. We have recently observed that the injection of triiodothyronine to rats strongly increases the protein kinase activity and the phosphorylation in vitro of myocardial Chromatin non-histone proteins [19]. Since the hormone induces heart hypertrophy, it is difficult Enzyme. Protein kinase (EC 2.7.1.37).
to know whether phosphorylation is a direct effect of triiodothyronine or if it is an early step of cardiac hypertrophy. Since triiodothyronine has no effect on liver size in adult rat, it is interesting to see whether this hormone has any effect on chromatin protein kinase and phosphoproteins. A decrease of cytosol histone kinase activity was observed in thyroidectomized rats [20]. The data presented in this paper provide evidence for a specific stimulation by triiodothyronine of liver chromatin protein kinase activity without any effect on the cytosol kinases. The hormone also increases the ability of chromatin phosphoproteins to be phosphorylated in vitro in the presence of ATP. It is suggested that part of the effect of thyroid hormone at the gene level could be mediated through its effect on chromatin protein kinase activity and chromatin protein phosphorylation. MATERIALS AND METHODS Chemicals ATP, adenosine 3’ :5’-monophosphate, phosvitin and casein, triiodothyronine were obtained from Sigma Chemical (St Louis, Mo., U.S.A.), [y-32P]ATP with a specific activity in the range of 2- 3 Ci per mole was obtained from the Commissariat a I’Energie Atomique (Saclay, France). DEAE-cellulose (DE 11) was from Whatman. All other chemicals were obtained from Merck (Darmstadt, Germany).
110
Animals The experiments were performed with Wistar rats with a body weight range of 150-200 g. They were injected intraperitoneally daily with 15 pg of triiodothyronine/lOO g body weight.
Preparation of Chromatin Non-Histone Proteins Rats were killed by cervical fracture and bleeding. Liver nuclei were prepared by the technique of Chauveau et al. [21]. They were washed 4 times by centrifugation in 0.25 M sucrose containing 10 mM MgCl,. The complete procedure was performed at 4 "C. Chromatin non-histone proteins were prepared according to Takeda et al. [22] except that the last purification step was a precipitation by 40% ammonium sulfate. The precipitate contained the total protein kinase activity; it was dissolved in the following buffer: 50 mM Tris buffer (pH 8.0)/1 mM EDTA/ 1 mM 2-mercaptoethanol. The sample was dialyzed against several changes of the same buffer. Approximately 1 mg of protein was obtained from 1 g of liver. This preparation, which will be referred to as unfractionated chromatin non-histone proteins, was used as a source of protein kinase and also as a source of endogenous substrate for nuclear protein kinase. A 1 mg aliquot of this preparation is able to transfer 50 nmol of 32Pfrom [y-32P]ATPon phosvitin in 30min at 37 "C.
Fractionation of Chromatin Protein Kinase About 50 mg of chromatin non-histone proteins were loaded onto a DEAE-cellulose (DE 11) column (20 x 1 cm) equilibrated with the Tris/EDTA/mercaptoethanol buffer. The column was washed thoroughly with this buffer. After the removal of unretained proteins, a 0 - 0.35 M NaCl linear concentration gradient in the buffer was established. The volume of the eluting buffer of the reservoir and of the mixing chamber was 75ml each. The flow rate was 20ml per hour. Fractions of 2.5 ml each were collected. The protein kinase activities were measured on 0.1 ml aliquots in the presence of 50 pg of either casein or phosvitin. As previously described [17], 2 peaks of activity were obtained, kinase I active with casein, kinase I1 active with casein and phosvitin. The fractions were pooled, dialyzed against a 50 mM Tris buffer (pH 7.8) containing 0.4 M NaCl, and concentrated by dialysis in vacuum. A 1 mg aliquot of kinase I1 is able to transfer 0.25 pmol of 32P from [y-32P]ATPon phosvitin in 30 min at 37 "C.
Preparation of Cytosol Protein Kinases In previous studies [23,24], we observed that the pH 4.8 supernatant from liver and heart cytosol
~-3,.5,3'-Triiodothyronineand Rat Liver Chromatin Protein Kinase
contained kinases active on histone and protamine, and that the pH 4.8 precipitate from liver and heart cytosol contained a protein kinase active on casein and phosvitin. Parts of the livers taken for the preparations of nuclei were homogenized with a Potter homogenizer in 0.3 M sucrose containing 4 mM EDTA. The procedure was performed at 4°C. The homogenate was centrifuged at 30000 x g for 20 rnin and the supernatant centrifuged for 2 h at 105000 x g . The pH of the final supernatant was adjusted to pH 4.8 by dropwise addition of acetic acid. The supernatant was adjusted to pH 7.0 by addition of 1 M Tris buffer (pH 8.0). Solid ammonium sulfate was added up to 50%. The precipitate was dissolved in a 50 mM Tris buffer (pH 7.5) containing 10 mM MgCI, and dialyzed against the same buffer. The pH 4.8 precipitate was collected by a 30 rnin centrifugation at 30000 x g and solubilized in 0.1 M Tris buffer (pH 7.8). Solid ammonium sulfate was added up to 65%. The precipitate was dissolved in a 50 mM Tris buffer (pH 7.5) containing 10 mM MgCl, and dialyzed against the same buffer with several changes. Assay.s.for Protein Kinase Activities The standard incubation medium for the nuclear enzyme contained in a total volume of 0.25 ml: the enzyme preparation (containing either 60 pg of unfractionated chromatin non-histone proteins or 20 pg of protein kinase purified on a DEAE-cellulose column), 100 pg of either casein or phosvitin, 20 pmol of [ Y - ~ ~ P I A T containing P 2 x 10' counts/min, MgCl, (20 mM with casein, 5 mM with phosvitin), 0.10 M Tris buffer (pH 7.8). The incubations were performed for 30 rnin at 37 "C. The incubation medium for the cytosol histone kinase contained in a total volume of 0.25 ml: 100 pg of the cytosol fraction prepared from the p H 4 .8 supernatant, 1OOpg of histone F, from rabbit bone-marrow, 5 pM adenosine 3' :5'-monophosphate, 10 mM MgCl,, 25 pmol Tris buffer (pH 7.8), 2 x lo6 counts/min [ Y - ~ ~ P I A TThe P . incubations were performed for 15 rnin at 37 C. The incubation medium for phosvitin-casein kinase contained in a total volume of 0.25 ml: 50 pg of the cytosol fraction prepared from the pH 4.8 precipitate, 100 pg of casein or phosvitin, MgC1, (20 mM for casein, 5 mM for phosvitin), 0.10 M Tris buffer (pH 7.8 for casein, pH 7.2 for phosvitin), 2 x lo6 counts/min [ Y - ~ ~ P I A TThe P . incubations werc performed for 30 rnin at 37 "C. The reactions were stopped by the addition of 200 pg of ATP and 1 ml (for experiments using histone and casein as substrates) or 0.2 ml (when phosvitin was the substrate) of 31 "/, trichloroacetic acid. The precipitates were collected on Millipore HA 0.45 km
111
J. Kruh and L. Tichonicky
filters and washed with 5 ml of trichloroacetic acid (25 when histone or casein were the substrates, 10 ?< for experiments using phosvitin). The radioactivity was measured in a Nuclear Chicago liquid scintillation spectrometer by the Cerenkov effect. Variations of specific activities of 5 % were found in preparations from the same animal. Variations within 10% were found when the measures werc made in preparations from different rat livers. Polyacrylumide Gel Electrophoresis Electrophoreses were performed using the system 398 of Rodbard and Chrambach [25,26] modified by Gill and Garren [27] with 7 . 5 x acrylamide. Bromophenol blue was uscd as front indicator. After electrophoresis, 2 mA per tube, at 4 "C for 150 min, gels were sliced into 2-mm-wide sections. The gels were used for two purposes. For distribution of protein kinase activities : samples containing 200 pg of purified protein kinase wcre applied on each gel. After electrophoresis, each slice was extracted overnight with 0.4 ml of 0.10 M Tris buffer (pH 7.8) containing 10 mM MgCI,. Each sample was divided into 3 aliquots. Each aliquot was incubated with [y3'P]ATP containing 1 x lo6 counts/min and with 50 pg of either casein or phosvitin under the optimal conditions described above or in the absence of substrate. The endogenous activities were substracted from the activities obtained in the presence of substrate. The other use for the gels was distribution of bound 32Pafter incubation of chromatin non-histone proteins with [ Y - ~ ~ P I A:Tthe P incubation mixture consisted of 100 pg of unfractionated chromatin non-histone proteins, 10 mM MgCI,, 25 pmol Tris buffer (pH 7.0), [p3'P]ATP containing 2 x lo6 counts/ min, in a total volume of 0.1 ml. The incubations were performed for 60 min at 37 "C. The samples were applied on polyacrylamidc gel. After electrophoresis the radioactivity was measured in each 2-mm-wide slice by the Cerenkov effect. Other Techniques Histone F, was prepared by the method number 2 of Johns [28] from bone-marrow of rabbit treated with phenylhydrazine. Proteins were estimated by the spectrophotometric method of Kalckar [29]. RESULTS AND DISCUSSION
Ejject of Triiodothyronine in vivo on Chromatin Protein Kinase Activity in vitro Rats were injected daily with 15 pg/lOO g body weight of triiodothyronine for 1, 2, 6, 9 and 14 days.
2.25
i
If 1.0-
1
0
5
10
15
Injection number
Fig. 1, Ejrrct of irvrctions of ~r.iio(lut/i~r.orzii~e on clironiutiri protein kinasr specific aclivities. Rats were injected daily with 15 pg of triiodothyronine per 100 g of weight. Samples of 60 pg of unfractionated chromatin non-histone proteins were incubated with 100 pg of either casein ( 0 d )or phosvitin ( x -__ x ) and [;.-32P]ATP. Thc activity ratio is activity with triiodothyronine over control activity
Chromatin non-histone proteins extracted from liver nuclei werc incubated with either casein or phosvitin and [y-32P]ATP (Fig. 1). A small increase in the protein kinase specific activity was observed with both substrates after one injection; a much larger increase was found after two injections. The enzyme specific activity increased further, but to a lesser extent, up to the 9th day. After 9 injections the increase in specific activity was of approximately 100 %. Afterwards the enzyme activity decreased, although the hormone injections were continued. Since several kinase activities are present in rat liver chromatin [13,17], we fractionated chromatin non-histone proteins prepared from rats injected 8 times with triiodothyronine and from control rats on a DEAE-cellulose column, as described in the Materials and Methods section. The protein kinase I specific activity was found identical in the two samples of non-histone proteins, but the protein kinase I1 specific activity was much higher in the preparation from hormone-treated rats. The following investigations were performed with protein kinase I1 from liver chromatin non-histone proteins prepared from rats injected 8 times with triiodothyronine and from untreated rats. These preparations will be referred to as purified protein kinases. The specific activity of this enzyme was found 4-times higher with any concentration of substrates (Fig. 2). This increase in specific activity was observed in incubations performed from pH 6.4 to 9.0 (Fig. 3). The pH had little effect on casein phosphorylation, the optimal pH was 7.1 when phosvitin was the substrate, but with the enzyme prepared from treated rats, the specific activity decreased much more at higher pH values, with the exception of a shoulder at pH 8.2, than with the enzyme from control rats.
~-3,5,3’-Triiodothyronine and Rat Liver Chromatin Protein Kinase
112
P
a, , /’
I
6.4
7.1
I
.
7.5 7.88.0
8.5
9.0
PH Fig. 2. &J’?BC~ of suhstrate conccntrution on protein kinnse specifi’c uctivifies. Samples containing 20 pg of purified protein kinase from livers from triiodothyronine-injected (----) and control rats (-) wcrc incubated with varying amounts of either casein ( 0 ) or phosvitin ( x )
Fig. 3. Eflec.1 of’ p H on protein kinuse .spec$(. w f i v i / i e s . Samples containing 20 kg of purified protein kinase from livers from triiodothyronine-injected (----) and control rats (-) were incubated with 100 pg of either casein ( 0 ) or phosvitin ( x ) in the presence of0.10 M Tris buffer adjusted to various pH values
This increase in protein kinase specific activity in chromatin from triiodothyronine-treated rats could be explained in several ways : (a) a triiodothyroninemediated removal or inhibition of synthesis of a protein kinase inhibitor. In order to test this possibility, we have compared the specific activities of both enzyme preparations separately and after mixing. Under our experimental conditions, the 32Pincorporation into phosvitin was 132200 counts/min with protein kinase from control rats, 384600 counts/min with an cqual amount of the enzyme from injected rats, and 497800 counts/min when a mixture of both enzyme preparations was used. The incorporation was approximately additive, which argues against the presence of an inhibitor in the samples of protein kinase from untreated rats ; (b) a hormone-mediated translocation of the enzyme from cytosol into the nucleus. In a previous work, we found that the same enzyme, active on phosvitin and casein, is present in liver cytosol and in chromatin [17]. We measured the protein kinase specific activities in cytosol fractions from rats injected with triiodothyronine. We found no significant variations in the presence of histone, phosvitin or casein as substrates when compared to cytosol from control rats, which excludes the possibility that the increase in protein kinase activity results from the presence in chromatin of enzymes translocated from
cytosol; (c) a stimulation by the hormone of protein kinase activity. Triiodothyronine added to the incubation medium at concentrations from 10 pM to 1 mM had no effect on protcin kinasc activity. Nevcrthcless a stimulation could have occurred in vivo through an indirect mechanism; (d) hormone-induced synthesis of protein kinase. In this case the new synthesized enzyme could have properties different from those of the enzymes from untreated rats. This led us to compare purified protein kinase preparations from treated and from control rats. Fig. 4 shows the electrophoretic distribution of protein kinases. With the enzyme preparations from untreated animals, 3 peaks active with casein and 4 peaks active with phosvitin were found. With the enzyme preparations from injected rats, the electrophoretic pattern was different: not only the same peaks, with much higher activities, were found, but several additional peaks were also observed. We have compared the resistance to heat denaturation of both enzymes (Fig. 5). Purified protein kinases were preincubated at 31 “C for 15, 30 and 60 min, and the residual activity was measured after the addition of substrate and of radioactive ATP. As previously described [17], the preincubation for 60 min of protein kinase from control rats had no significant effect on its activity with casein and decreased to less than 40 ”/,
113
J. Kruh and L. Tichonicky
Phosvitin
I \
0
I--, i
0
i !
60
Fig. 5 . Effect ojprotein kinase prtJincubation at 37 C on i l s specific activity. Samples of 20 pg of purified protein kinase from triiodothyronine-injected (----) and control rats (-) were incubated at 37 “C for 15, 30 and 60 min. Casein (0)or phosvitin ( x ) and [y-32P]ATPwere added and the incubation continued for a further 30 min
/ i I
15 30 Preincubation time (min)
!
M
10
30
Slice number
Fig. 4. Electrophoretic pattern of chromatin protein kinases. Samples containing 200 pg of purified protein kinase from triiodothyroninetreated (----)and control rats (-) were submitted to gel eleclrophoreses. The gels were cut into 2-mm-wide slices. Each slice was eluted with 0.4 ml of 0.10 M Tris buffer (pH 7.8) containing 10 mM MgCI2. Samples of 0.1 ml were incubated with 50 Fg of either casein or phosvitin or in the absence of substrate. The endogenous activities of each slice were subtracted from the activities found in the prcscnce of substrate
the activity with phosvitin. The effect of the preincubation was different with protein kinase from treated rats; activities with both substrates were reduced to 60- 70 ”/, after 1 h of preincubation. Protein kinase from triiodothyronine-treated rats was less inhibitcd by p-hydroxymercuribenzoate than the enzyme from untreated rats (Fig. 6 ) , although with both enzymes, casein phosphorylation was more affected than phosvitin phosphorylation. These experiments seem to indicate that at least part of the enzymes present in chromatin from triiodothyronine-treated rats is not identical to the enzymes from untreated animals, which is in favor of an induction of protein kinase by thyroid hormone. In order to correlate the physiological action of the hormone with its effect on chromatin protein kinase, we injected at a daily rate of 15 pg/lOO g of body weight for 8 days, two compounds structurally
0 0
10 x) [ p -Hydroxymercuribenzoate] (pM)
30
Fig. 6 . Lflect of p-h~~ros~mercuriben--oate on prorein kinase specific activities. Samples of 20 pg of purified protein kinase from triiodothyronine-injected ( ---) and control rats (-) were incubated with various concentrations of p-hydroxymercuribenzoate for 15 min at 4 ’C. Casein ( 0 )or phosvitin ( x ) and [;‘-32P]ATPwere then added and the incubation continued for a further 30 min at 31 “C
related to the thyroid hormone, but with no biological activity : 3,5-diiodothyronine and 3,3’,5’-triiodothyronine. The injection did not significantly modify the chromatin protein kinase activities in the presence of casein and of phosvitin. Effect qf Triiodothyronine in vivo on the Phosphorylation in vitro of’ Chromatin Non-Histone Protein by A T P
Liver chromatin non-histone proteins from rats injected with triiodothyronine for 1, 2, 6, 9 and 14 days were incubated with [p3’P]ATP. Endogenously labelled phosphoproteins were separated on polyacrylamide gel (Fig. 7). No modifications were found after one or two triiodothyronine injections (not shown in the figure). After six injections there was a large increase in labelling, afterwards
1.-3.5.3'-Triiodothyronineand Rat Livcr Chromatin Protein Kinase
I
H
I
ll
10
20
30
Slice number
10
20
30
Slice number
Fig. 7. Elc,c.troi~horc,,ii, p u l t w n qf " P incorporulrd in vitro into ckromutin nun-Ristonc proteitis ,from r r i i ~ ~ ~ o t l l ~ ~ r o n i n e - t r crats. ~ated Chromatin non-histone proteins werc prepared from rats submitted x ), 9 (----) and 14 (. . . . . .) injections of triiodothyroat 6 ( x ninc and froin control rats (-). The samples containing 100 pg of proteins were incubated with [ Y - ~ ~ P I Afor T P 60 inin at 37 "C and submitted to gel clcctrophoresis. The gels wcre cut into 2-mniwidc slices. The radioactivity was measured in each slice
Fig. 8. L+c.trophori~tic putlrrti of P iticorpor.tr~i~c/ in vitro rn1o chromutin non-hislone proteins ,from ihjroidecromixd rats. Chromatin non-histone proteins were prcparcd from rats submitted to x ) before sacrithyroidcctomy 7 days ( ---) and 14 days ( x fice. and from rats submittcd to thyroidectomy and treated after 7 days with 7 injections of triiodothyronine (-----). The samples were incubated with [i'-32P]ATPfor 60 min a t 37 'C and submitted to gel electrophoresis. The gels were cut into 2-mm-widc slices. Thc radioactivity was measured in each slice
the phosphorylation decreased although the injections were continued. Chromatin non histone proteins were prepared from rats thyroidectomized 7 and 14 days before sacrifice and from rats thyroidectomized and injected 7 days later 7 times with triiodothyronine and each sample analyzed by gel electrophoresis (Fig. 8). The level of phosphorylation was much lower in chromatin prepared 14 days than in chromatin prepared 7 days after thyroidectomy. In chromatin from rats submitted to thyroidectomy 7 days before, a large peak was constantly found in slices 18- 19. The injection of triiodothyronine in thc thyroidectomized rats increased the level of phosphorylation in vitro, but the peak in slices 18- 19 partly disappeared. It may be concluded that thyroid hormone increases the ability of chromatin non-histone proteins to be phosphorylated in vitro. This could be explained in several ways: (a) the effect of the hormone on
chromatin protein phosphorylation rcsults from its effect on protein kinase. In this case we should haw expected an increase of phosphorylation in vivo, which would have reduced the number of sites available for the phosphorylation in vitro. Moreover, two injections of triiodothyronine strongly increased protein kinase activity but had no effect on chromatin protein phosphorylation; (b) triiodothyronine increased the turnover of protein-bound phosphate, which results in a loss of phosphate in vivo and in an increase of the number of sites available for phosphorylation in vitro ; (c) a hormone-induced change in phosphoprotein conformation which increases the number of phosphorylation sites; (d) a hormoneinduced synthesis of new phosphoprotein molecules, which would be consistent with the increase in chromatin protein concentration in triiodothyronine-treated rats [6]. The problem remains open at the present time.
~
115
J. Kruh and L. Tichonicky
I n these experiments. the amount of injected triiodothyronine was highcr than the physiological amounts of the hormone. In a set of experiments, we injected 8 timcs 1.5 pg/lOO g of body weight of triiodothyronine. Under these conditions we observed an increase in the chromatin protein phosphorylation in vi/ro. but we werc unablc to obscrve any increase in protein kinasc activity. Nevertheless we cannot excludc that a small increase in thc enzyme activity would not have been detected by, the current methods used for thc measurement of protein kinase activity. The modifications in protein kinase activity and in chromatin protein phosphorylation seems to result not from a high level of thyroid hormone but from a rapid increase in the hormone level, since the modifications induccd by the hormone are transient and disappear progressively with time even if the hormone is further administercd. It has bcen shown that the binding of triiodothyronine to chromatin occurs within 30 min after the injection of the hormone [8]. Intravenous injection of small amounts of the hormone stimulates RNA polymerase I1 synthesis within 20 min [6]. The effect of the injection of approximately 15 pg of triiodothyronine on polymerase I was detected after 10 h. The incorporation of labclled orotic acid into nuclear R N A increased 3-4 h after the injection of the hormone. The accumulation of newly formed ribosomes and the increase in the amino acid incorporation activity of ribosomes were detected approximately 30 h after triiodothyronine injection [2]. We found a significant increase in protein kinase activity only after 2 daily injcctions of the hormone. Although it is not possible to compare measurements made in different laboratories, the fact that a single injection of triiodothyronine failed to produce a significant effect on protein kinase activity when it would have had a dramatic effect on transcription, casts doubt on the relationship between the two processes. Nevertheless the demonstration that the nuclear receptor of triiodothyronine is a chromatin nonhistone protein, the demonstration that these proteins and that Chromatin protein phosphorylation play a fundamental role in the regulation of gene activity [11 - 161, the absence of effect of compoundschemically related to triiodothyronine but devoid from any biological activity makes it very likely that the effcct of the hormone on chromatin protein kinase activity is relevant to its physiological action. The authorsare much indebted to Drs N. Defer and B. Dastugue for helpful discussions, and to Dr N. Etling for analyzing purity of trrrodolhyroninc and providing 3,3',5'-triiodothyronine and 3.5-diiodothyroninc.
This work was supported by Grants from thc lnstiiur :Vur,oitcil cle lu Ri~cltmhcMcYicule, thc ('iviire h'urioirul dc lu Rucherelie Scio,ri/irluc and thc Uc;ltgurinn Gitt+rdi>u lr Rivltcrc.N.he
(k.lu Suirii ri
Scienr:fiquv er Technique.
REFERENCES 1 . Greengard. 0. (1970)in Biocli~miccrl Acrinns n/' Hornmnes (Litwack, Ci., cd.) vol. 1. pp. 53-87. Acadcmic P r w . Ncw York and London. 2. Tala, J. R. & Widnell, C. C. (1966)Bicchem. J . 98,604-620. 3. Tata. J. R. (1966)Pmg. Nuclvic Acid Res. Mol. Uinl. 5, 191
-
250. 4. Wyatt, G.R . & Tata, J. R . (1968)BincAm. J . IOY, 253-258. 5. Smuckler. E. A. & Tata, J. R. (1971) Nuiure 1 Lond.) 234. 37- 39. 6. Jothy, S.. Bilodeau, J. L., Champsaur. H. B Simpkins, H. ( 1975)Biocltt*ni.J . 150, 133 - 135. 7. Tata, J. R. & Williams-Ashman, H. (3. (1967)Eur. J . Binc~hrm. 2.366- 374. 8. Surks. M.I., Kacrncr. D., Dillman, W. & Oppcnhcimer, J. H. (1973).I. B i d . Chem. 2448.7066-7072. 9. Samucls, H. H. & Tsai. J. S. (1973) Proc. Nurl Acurl. Sci. U.S.A. 70.3488- 3492. 10. Thomopoulos, P., Dastuguc, B. & Defer. N. (1974)Biocltmi. Biophys. Rcs. Cornmun~58,499- 506. 11. Kamiyama. M. & Wang.T. Y . (1971)Bicichim. Biop1t.w. Actu. 225,563- 576. 12. Kamiyama, M., Dastuguc, B. B Kruh, J. (1971) Biochem. Biopltw. Res. Contmun. 44.29- 36. 13. Kamiyama, M., Dastugue, B., Defer, N. & Kruh, J. (1972) Binchint. Rinpltys. Acta, 277.576 - 583. 14. Stein, G . S.,Spelsberg, T.C. & Kleinsmith. L. J. (1974)Sciiwec~ ( Wu.vh. D.C.) 183.8 17- 824. 15. Kruh, J. Dastugue. B., Defer, N., Kamiyama, M. & Tichonicky, I,. (1975)Biochimie, 56,995- 1001. 16. Kruh, J., Courtois, Y., Dastugue, B.. Defer, N., Gibson, K., Kamiyama, M. & Tichonicky, L. (1975)in Regularion I$ Growdi and D@wvttiated Function in Eukuryctic Cel1.s.
Raven Press, New York, p. 139- 154. 17. Dastugue. B., Tichonicky, L. B Kruh, J. (1974) Biochimie, 56,491- 500. 18. Gibson, K., Tichonicky, L. & Kruh. J . (1974)Binchimic. 56. 1417- 1423. 19. Gibson, K.. Tichonicky. L. & Kruh, J. (1975)Mnl. Cell. Bioclieni. Y, 79- 83. 20. Correze, C., Pincll, P. & Nunez, J. (1972)FEES Lerr. 23.87- 91. 21. Chauveau, J., Mouk, Y. & Rouiller. C. (1956)E.rp. Cell Rus. 11. 317- 321. 22. Takcda. M., Yamamura, H. & Ohga. Y. (1971)Biocliem. Bioplrys. Rvs. Commun. 42, 103- 110. 23. Dastugue, B.. Tichonicky, L. & Kruh, J. (1973)Biochimic, 55. 1021- 1030. 24. Gibson. K., Tichonicky, L. & Kruh. J. (1974)Biochimii., 56, 1409- 1416. 25. Rodbard, D. & Chrambach, A. (1971) Anal. Binchetn. 40, 95-134. 26. Chrambach, A. & Rodbard, D. (1971)Science ( Wmh. D.C.) 172,440-451. 27. Gill, G . N. & Garren, L. D. (1971) Proc. Nor1 Acud. Sci. U.S.A. 68,786-790. 28. Johns, E. W. (1964)Biochem. J . 92,55-59. 29. Kakkar, H. M. (1947)J. Biol. Chem. 167,461-469.
J . Kruh and L. Tichonicky, Institut de Palhologie Mokukdire, 24,rue du Faubourg St-Jacques, F-75014Paris, France
Effect of Triiodothyronine on Rat Liver Chromatin Protein Kinase Jacques KRUH and Lydie TICHONICKY Institut de Pathologie Moleculaire, Paris (Received June 1610ctober 23, 1975)
1. Injection of triiodothyronine to rats stimulates protein kinase activity in liver chromatin nonhistone proteins. A significant increase was found after two daily injections. A 4-fold increase was observed with the purified enzyme after eight daily injections of the hormone. No variations were observed in cytosol protein kinase activity. Electrophoretic pattern, effect of heat denaturation, effect ofp-hydroxymercuribenzoate seem to indicate that the enzyme present in treated rats is not identical to the enzyme in control animals, which suggests that thyroid hormone has induced nuclear protein kinase. Diiodothyronine, 3,3’,5’-triiodothyroninehave no effect on protein kinase. 2. Chromatin non-histone proteins isolated from rats injected with triiodothyronine incorporated more 32Pwhen incubated with [p3’P]ATP than the chromatin proteins from untreated rats. Thyroidectomy reduced the in vitro 32Pincorporation. It is suggested that some of the biological activity of thyroid hormone could be mediated through its effect on chromatin non-histone proteins.
The role of thyroid hormone in cell growth and differentiation is well-documented: triiodothyronine is able to induce several enzymes prematurely in liver and brain [l]. It is then very likely that this hormone acts partly at the gene level. Injection of thyroid hormone stimulates the rate of RNA synthesis [2,3], essentially ribosomal RNA [4]. It enhances the activity of polymerase I [ 5 ] . It has recently been shown that it also induces an early and transient activation of polymerase I1 [6]. Triiodothyronine stimulates protein synthesis, mainly at the level of ribosomes, it increases the amount of newly formed ribosomes and the incorporation of amino acids in vitro [3,7]. The hormone also increases the concentration of chromatin non-histone proteins [6]. It has been recently established that triiodothyronine is able to bind to nuclear proteins [8,9] and more precisely to chromatin non-histone proteins [lo]. These proteins are able to activate gene transcription [11 - 161. This gene activation may be related to the physiological effect of thyroid hormone. In previous works, we have purified from liver and heart chromatin proteins, protein kinases active with histone, nuclear phosphoproteins [12,13], casein and phosvitin [17,18]. We have recently observed that the injection of triiodothyronine to rats strongly increases the protein kinase activity and the phosphorylation in vitro of myocardial Chromatin non-histone proteins [19]. Since the hormone induces heart hypertrophy, it is difficult Enzyme. Protein kinase (EC 2.7.1.37).
to know whether phosphorylation is a direct effect of triiodothyronine or if it is an early step of cardiac hypertrophy. Since triiodothyronine has no effect on liver size in adult rat, it is interesting to see whether this hormone has any effect on chromatin protein kinase and phosphoproteins. A decrease of cytosol histone kinase activity was observed in thyroidectomized rats [20]. The data presented in this paper provide evidence for a specific stimulation by triiodothyronine of liver chromatin protein kinase activity without any effect on the cytosol kinases. The hormone also increases the ability of chromatin phosphoproteins to be phosphorylated in vitro in the presence of ATP. It is suggested that part of the effect of thyroid hormone at the gene level could be mediated through its effect on chromatin protein kinase activity and chromatin protein phosphorylation. MATERIALS AND METHODS Chemicals ATP, adenosine 3’ :5’-monophosphate, phosvitin and casein, triiodothyronine were obtained from Sigma Chemical (St Louis, Mo., U.S.A.), [y-32P]ATP with a specific activity in the range of 2- 3 Ci per mole was obtained from the Commissariat a I’Energie Atomique (Saclay, France). DEAE-cellulose (DE 11) was from Whatman. All other chemicals were obtained from Merck (Darmstadt, Germany).
110
Animals The experiments were performed with Wistar rats with a body weight range of 150-200 g. They were injected intraperitoneally daily with 15 pg of triiodothyronine/lOO g body weight.
Preparation of Chromatin Non-Histone Proteins Rats were killed by cervical fracture and bleeding. Liver nuclei were prepared by the technique of Chauveau et al. [21]. They were washed 4 times by centrifugation in 0.25 M sucrose containing 10 mM MgCl,. The complete procedure was performed at 4 "C. Chromatin non-histone proteins were prepared according to Takeda et al. [22] except that the last purification step was a precipitation by 40% ammonium sulfate. The precipitate contained the total protein kinase activity; it was dissolved in the following buffer: 50 mM Tris buffer (pH 8.0)/1 mM EDTA/ 1 mM 2-mercaptoethanol. The sample was dialyzed against several changes of the same buffer. Approximately 1 mg of protein was obtained from 1 g of liver. This preparation, which will be referred to as unfractionated chromatin non-histone proteins, was used as a source of protein kinase and also as a source of endogenous substrate for nuclear protein kinase. A 1 mg aliquot of this preparation is able to transfer 50 nmol of 32Pfrom [y-32P]ATPon phosvitin in 30min at 37 "C.
Fractionation of Chromatin Protein Kinase About 50 mg of chromatin non-histone proteins were loaded onto a DEAE-cellulose (DE 11) column (20 x 1 cm) equilibrated with the Tris/EDTA/mercaptoethanol buffer. The column was washed thoroughly with this buffer. After the removal of unretained proteins, a 0 - 0.35 M NaCl linear concentration gradient in the buffer was established. The volume of the eluting buffer of the reservoir and of the mixing chamber was 75ml each. The flow rate was 20ml per hour. Fractions of 2.5 ml each were collected. The protein kinase activities were measured on 0.1 ml aliquots in the presence of 50 pg of either casein or phosvitin. As previously described [17], 2 peaks of activity were obtained, kinase I active with casein, kinase I1 active with casein and phosvitin. The fractions were pooled, dialyzed against a 50 mM Tris buffer (pH 7.8) containing 0.4 M NaCl, and concentrated by dialysis in vacuum. A 1 mg aliquot of kinase I1 is able to transfer 0.25 pmol of 32P from [y-32P]ATPon phosvitin in 30 min at 37 "C.
Preparation of Cytosol Protein Kinases In previous studies [23,24], we observed that the pH 4.8 supernatant from liver and heart cytosol
~-3,.5,3'-Triiodothyronineand Rat Liver Chromatin Protein Kinase
contained kinases active on histone and protamine, and that the pH 4.8 precipitate from liver and heart cytosol contained a protein kinase active on casein and phosvitin. Parts of the livers taken for the preparations of nuclei were homogenized with a Potter homogenizer in 0.3 M sucrose containing 4 mM EDTA. The procedure was performed at 4°C. The homogenate was centrifuged at 30000 x g for 20 rnin and the supernatant centrifuged for 2 h at 105000 x g . The pH of the final supernatant was adjusted to pH 4.8 by dropwise addition of acetic acid. The supernatant was adjusted to pH 7.0 by addition of 1 M Tris buffer (pH 8.0). Solid ammonium sulfate was added up to 50%. The precipitate was dissolved in a 50 mM Tris buffer (pH 7.5) containing 10 mM MgCI, and dialyzed against the same buffer. The pH 4.8 precipitate was collected by a 30 rnin centrifugation at 30000 x g and solubilized in 0.1 M Tris buffer (pH 7.8). Solid ammonium sulfate was added up to 65%. The precipitate was dissolved in a 50 mM Tris buffer (pH 7.5) containing 10 mM MgCl, and dialyzed against the same buffer with several changes. Assay.s.for Protein Kinase Activities The standard incubation medium for the nuclear enzyme contained in a total volume of 0.25 ml: the enzyme preparation (containing either 60 pg of unfractionated chromatin non-histone proteins or 20 pg of protein kinase purified on a DEAE-cellulose column), 100 pg of either casein or phosvitin, 20 pmol of [ Y - ~ ~ P I A T containing P 2 x 10' counts/min, MgCl, (20 mM with casein, 5 mM with phosvitin), 0.10 M Tris buffer (pH 7.8). The incubations were performed for 30 rnin at 37 "C. The incubation medium for the cytosol histone kinase contained in a total volume of 0.25 ml: 100 pg of the cytosol fraction prepared from the p H 4 .8 supernatant, 1OOpg of histone F, from rabbit bone-marrow, 5 pM adenosine 3' :5'-monophosphate, 10 mM MgCl,, 25 pmol Tris buffer (pH 7.8), 2 x lo6 counts/min [ Y - ~ ~ P I A TThe P . incubations were performed for 15 rnin at 37 C. The incubation medium for phosvitin-casein kinase contained in a total volume of 0.25 ml: 50 pg of the cytosol fraction prepared from the pH 4.8 precipitate, 100 pg of casein or phosvitin, MgC1, (20 mM for casein, 5 mM for phosvitin), 0.10 M Tris buffer (pH 7.8 for casein, pH 7.2 for phosvitin), 2 x lo6 counts/min [ Y - ~ ~ P I A TThe P . incubations werc performed for 30 rnin at 37 "C. The reactions were stopped by the addition of 200 pg of ATP and 1 ml (for experiments using histone and casein as substrates) or 0.2 ml (when phosvitin was the substrate) of 31 "/, trichloroacetic acid. The precipitates were collected on Millipore HA 0.45 km
111
J. Kruh and L. Tichonicky
filters and washed with 5 ml of trichloroacetic acid (25 when histone or casein were the substrates, 10 ?< for experiments using phosvitin). The radioactivity was measured in a Nuclear Chicago liquid scintillation spectrometer by the Cerenkov effect. Variations of specific activities of 5 % were found in preparations from the same animal. Variations within 10% were found when the measures werc made in preparations from different rat livers. Polyacrylumide Gel Electrophoresis Electrophoreses were performed using the system 398 of Rodbard and Chrambach [25,26] modified by Gill and Garren [27] with 7 . 5 x acrylamide. Bromophenol blue was uscd as front indicator. After electrophoresis, 2 mA per tube, at 4 "C for 150 min, gels were sliced into 2-mm-wide sections. The gels were used for two purposes. For distribution of protein kinase activities : samples containing 200 pg of purified protein kinase wcre applied on each gel. After electrophoresis, each slice was extracted overnight with 0.4 ml of 0.10 M Tris buffer (pH 7.8) containing 10 mM MgCI,. Each sample was divided into 3 aliquots. Each aliquot was incubated with [y3'P]ATP containing 1 x lo6 counts/min and with 50 pg of either casein or phosvitin under the optimal conditions described above or in the absence of substrate. The endogenous activities were substracted from the activities obtained in the presence of substrate. The other use for the gels was distribution of bound 32Pafter incubation of chromatin non-histone proteins with [ Y - ~ ~ P I A:Tthe P incubation mixture consisted of 100 pg of unfractionated chromatin non-histone proteins, 10 mM MgCI,, 25 pmol Tris buffer (pH 7.0), [p3'P]ATP containing 2 x lo6 counts/ min, in a total volume of 0.1 ml. The incubations were performed for 60 min at 37 "C. The samples were applied on polyacrylamidc gel. After electrophoresis the radioactivity was measured in each 2-mm-wide slice by the Cerenkov effect. Other Techniques Histone F, was prepared by the method number 2 of Johns [28] from bone-marrow of rabbit treated with phenylhydrazine. Proteins were estimated by the spectrophotometric method of Kalckar [29]. RESULTS AND DISCUSSION
Ejject of Triiodothyronine in vivo on Chromatin Protein Kinase Activity in vitro Rats were injected daily with 15 pg/lOO g body weight of triiodothyronine for 1, 2, 6, 9 and 14 days.
2.25
i
If 1.0-
1
0
5
10
15
Injection number
Fig. 1, Ejrrct of irvrctions of ~r.iio(lut/i~r.orzii~e on clironiutiri protein kinasr specific aclivities. Rats were injected daily with 15 pg of triiodothyronine per 100 g of weight. Samples of 60 pg of unfractionated chromatin non-histone proteins were incubated with 100 pg of either casein ( 0 d )or phosvitin ( x -__ x ) and [;.-32P]ATP. Thc activity ratio is activity with triiodothyronine over control activity
Chromatin non-histone proteins extracted from liver nuclei werc incubated with either casein or phosvitin and [y-32P]ATP (Fig. 1). A small increase in the protein kinase specific activity was observed with both substrates after one injection; a much larger increase was found after two injections. The enzyme specific activity increased further, but to a lesser extent, up to the 9th day. After 9 injections the increase in specific activity was of approximately 100 %. Afterwards the enzyme activity decreased, although the hormone injections were continued. Since several kinase activities are present in rat liver chromatin [13,17], we fractionated chromatin non-histone proteins prepared from rats injected 8 times with triiodothyronine and from control rats on a DEAE-cellulose column, as described in the Materials and Methods section. The protein kinase I specific activity was found identical in the two samples of non-histone proteins, but the protein kinase I1 specific activity was much higher in the preparation from hormone-treated rats. The following investigations were performed with protein kinase I1 from liver chromatin non-histone proteins prepared from rats injected 8 times with triiodothyronine and from untreated rats. These preparations will be referred to as purified protein kinases. The specific activity of this enzyme was found 4-times higher with any concentration of substrates (Fig. 2). This increase in specific activity was observed in incubations performed from pH 6.4 to 9.0 (Fig. 3). The pH had little effect on casein phosphorylation, the optimal pH was 7.1 when phosvitin was the substrate, but with the enzyme prepared from treated rats, the specific activity decreased much more at higher pH values, with the exception of a shoulder at pH 8.2, than with the enzyme from control rats.
~-3,5,3’-Triiodothyronine and Rat Liver Chromatin Protein Kinase
112
P
a, , /’
I
6.4
7.1
I
.
7.5 7.88.0
8.5
9.0
PH Fig. 2. &J’?BC~ of suhstrate conccntrution on protein kinnse specifi’c uctivifies. Samples containing 20 pg of purified protein kinase from livers from triiodothyronine-injected (----) and control rats (-) wcrc incubated with varying amounts of either casein ( 0 ) or phosvitin ( x )
Fig. 3. Eflec.1 of’ p H on protein kinuse .spec$(. w f i v i / i e s . Samples containing 20 kg of purified protein kinase from livers from triiodothyronine-injected (----) and control rats (-) were incubated with 100 pg of either casein ( 0 ) or phosvitin ( x ) in the presence of0.10 M Tris buffer adjusted to various pH values
This increase in protein kinase specific activity in chromatin from triiodothyronine-treated rats could be explained in several ways : (a) a triiodothyroninemediated removal or inhibition of synthesis of a protein kinase inhibitor. In order to test this possibility, we have compared the specific activities of both enzyme preparations separately and after mixing. Under our experimental conditions, the 32Pincorporation into phosvitin was 132200 counts/min with protein kinase from control rats, 384600 counts/min with an cqual amount of the enzyme from injected rats, and 497800 counts/min when a mixture of both enzyme preparations was used. The incorporation was approximately additive, which argues against the presence of an inhibitor in the samples of protein kinase from untreated rats ; (b) a hormone-mediated translocation of the enzyme from cytosol into the nucleus. In a previous work, we found that the same enzyme, active on phosvitin and casein, is present in liver cytosol and in chromatin [17]. We measured the protein kinase specific activities in cytosol fractions from rats injected with triiodothyronine. We found no significant variations in the presence of histone, phosvitin or casein as substrates when compared to cytosol from control rats, which excludes the possibility that the increase in protein kinase activity results from the presence in chromatin of enzymes translocated from
cytosol; (c) a stimulation by the hormone of protein kinase activity. Triiodothyronine added to the incubation medium at concentrations from 10 pM to 1 mM had no effect on protcin kinasc activity. Nevcrthcless a stimulation could have occurred in vivo through an indirect mechanism; (d) hormone-induced synthesis of protein kinase. In this case the new synthesized enzyme could have properties different from those of the enzymes from untreated rats. This led us to compare purified protein kinase preparations from treated and from control rats. Fig. 4 shows the electrophoretic distribution of protein kinases. With the enzyme preparations from untreated animals, 3 peaks active with casein and 4 peaks active with phosvitin were found. With the enzyme preparations from injected rats, the electrophoretic pattern was different: not only the same peaks, with much higher activities, were found, but several additional peaks were also observed. We have compared the resistance to heat denaturation of both enzymes (Fig. 5). Purified protein kinases were preincubated at 31 “C for 15, 30 and 60 min, and the residual activity was measured after the addition of substrate and of radioactive ATP. As previously described [17], the preincubation for 60 min of protein kinase from control rats had no significant effect on its activity with casein and decreased to less than 40 ”/,
113
J. Kruh and L. Tichonicky
Phosvitin
I \
0
I--, i
0
i !
60
Fig. 5 . Effect ojprotein kinase prtJincubation at 37 C on i l s specific activity. Samples of 20 pg of purified protein kinase from triiodothyronine-injected (----) and control rats (-) were incubated at 37 “C for 15, 30 and 60 min. Casein (0)or phosvitin ( x ) and [y-32P]ATPwere added and the incubation continued for a further 30 min
/ i I
15 30 Preincubation time (min)
!
M
10
30
Slice number
Fig. 4. Electrophoretic pattern of chromatin protein kinases. Samples containing 200 pg of purified protein kinase from triiodothyroninetreated (----)and control rats (-) were submitted to gel eleclrophoreses. The gels were cut into 2-mm-wide slices. Each slice was eluted with 0.4 ml of 0.10 M Tris buffer (pH 7.8) containing 10 mM MgCI2. Samples of 0.1 ml were incubated with 50 Fg of either casein or phosvitin or in the absence of substrate. The endogenous activities of each slice were subtracted from the activities found in the prcscnce of substrate
the activity with phosvitin. The effect of the preincubation was different with protein kinase from treated rats; activities with both substrates were reduced to 60- 70 ”/, after 1 h of preincubation. Protein kinase from triiodothyronine-treated rats was less inhibitcd by p-hydroxymercuribenzoate than the enzyme from untreated rats (Fig. 6 ) , although with both enzymes, casein phosphorylation was more affected than phosvitin phosphorylation. These experiments seem to indicate that at least part of the enzymes present in chromatin from triiodothyronine-treated rats is not identical to the enzymes from untreated animals, which is in favor of an induction of protein kinase by thyroid hormone. In order to correlate the physiological action of the hormone with its effect on chromatin protein kinase, we injected at a daily rate of 15 pg/lOO g of body weight for 8 days, two compounds structurally
0 0
10 x) [ p -Hydroxymercuribenzoate] (pM)
30
Fig. 6 . Lflect of p-h~~ros~mercuriben--oate on prorein kinase specific activities. Samples of 20 pg of purified protein kinase from triiodothyronine-injected ( ---) and control rats (-) were incubated with various concentrations of p-hydroxymercuribenzoate for 15 min at 4 ’C. Casein ( 0 )or phosvitin ( x ) and [;‘-32P]ATPwere then added and the incubation continued for a further 30 min at 31 “C
related to the thyroid hormone, but with no biological activity : 3,5-diiodothyronine and 3,3’,5’-triiodothyronine. The injection did not significantly modify the chromatin protein kinase activities in the presence of casein and of phosvitin. Effect qf Triiodothyronine in vivo on the Phosphorylation in vitro of’ Chromatin Non-Histone Protein by A T P
Liver chromatin non-histone proteins from rats injected with triiodothyronine for 1, 2, 6, 9 and 14 days were incubated with [p3’P]ATP. Endogenously labelled phosphoproteins were separated on polyacrylamide gel (Fig. 7). No modifications were found after one or two triiodothyronine injections (not shown in the figure). After six injections there was a large increase in labelling, afterwards
1.-3.5.3'-Triiodothyronineand Rat Livcr Chromatin Protein Kinase
I
H
I
ll
10
20
30
Slice number
10
20
30
Slice number
Fig. 7. Elc,c.troi~horc,,ii, p u l t w n qf " P incorporulrd in vitro into ckromutin nun-Ristonc proteitis ,from r r i i ~ ~ ~ o t l l ~ ~ r o n i n e - t r crats. ~ated Chromatin non-histone proteins werc prepared from rats submitted x ), 9 (----) and 14 (. . . . . .) injections of triiodothyroat 6 ( x ninc and froin control rats (-). The samples containing 100 pg of proteins were incubated with [ Y - ~ ~ P I Afor T P 60 inin at 37 "C and submitted to gel clcctrophoresis. The gels wcre cut into 2-mniwidc slices. The radioactivity was measured in each slice
Fig. 8. L+c.trophori~tic putlrrti of P iticorpor.tr~i~c/ in vitro rn1o chromutin non-hislone proteins ,from ihjroidecromixd rats. Chromatin non-histone proteins were prcparcd from rats submitted to x ) before sacrithyroidcctomy 7 days ( ---) and 14 days ( x fice. and from rats submittcd to thyroidectomy and treated after 7 days with 7 injections of triiodothyronine (-----). The samples were incubated with [i'-32P]ATPfor 60 min a t 37 'C and submitted to gel electrophoresis. The gels were cut into 2-mm-widc slices. Thc radioactivity was measured in each slice
the phosphorylation decreased although the injections were continued. Chromatin non histone proteins were prepared from rats thyroidectomized 7 and 14 days before sacrifice and from rats thyroidectomized and injected 7 days later 7 times with triiodothyronine and each sample analyzed by gel electrophoresis (Fig. 8). The level of phosphorylation was much lower in chromatin prepared 14 days than in chromatin prepared 7 days after thyroidectomy. In chromatin from rats submitted to thyroidectomy 7 days before, a large peak was constantly found in slices 18- 19. The injection of triiodothyronine in thc thyroidectomized rats increased the level of phosphorylation in vitro, but the peak in slices 18- 19 partly disappeared. It may be concluded that thyroid hormone increases the ability of chromatin non-histone proteins to be phosphorylated in vitro. This could be explained in several ways: (a) the effect of the hormone on
chromatin protein phosphorylation rcsults from its effect on protein kinase. In this case we should haw expected an increase of phosphorylation in vivo, which would have reduced the number of sites available for the phosphorylation in vitro. Moreover, two injections of triiodothyronine strongly increased protein kinase activity but had no effect on chromatin protein phosphorylation; (b) triiodothyronine increased the turnover of protein-bound phosphate, which results in a loss of phosphate in vivo and in an increase of the number of sites available for phosphorylation in vitro ; (c) a hormone-induced change in phosphoprotein conformation which increases the number of phosphorylation sites; (d) a hormoneinduced synthesis of new phosphoprotein molecules, which would be consistent with the increase in chromatin protein concentration in triiodothyronine-treated rats [6]. The problem remains open at the present time.
~
115
J. Kruh and L. Tichonicky
I n these experiments. the amount of injected triiodothyronine was highcr than the physiological amounts of the hormone. In a set of experiments, we injected 8 timcs 1.5 pg/lOO g of body weight of triiodothyronine. Under these conditions we observed an increase in the chromatin protein phosphorylation in vi/ro. but we werc unablc to obscrve any increase in protein kinasc activity. Nevertheless we cannot excludc that a small increase in thc enzyme activity would not have been detected by, the current methods used for thc measurement of protein kinase activity. The modifications in protein kinase activity and in chromatin protein phosphorylation seems to result not from a high level of thyroid hormone but from a rapid increase in the hormone level, since the modifications induccd by the hormone are transient and disappear progressively with time even if the hormone is further administercd. It has bcen shown that the binding of triiodothyronine to chromatin occurs within 30 min after the injection of the hormone [8]. Intravenous injection of small amounts of the hormone stimulates RNA polymerase I1 synthesis within 20 min [6]. The effect of the injection of approximately 15 pg of triiodothyronine on polymerase I was detected after 10 h. The incorporation of labclled orotic acid into nuclear R N A increased 3-4 h after the injection of the hormone. The accumulation of newly formed ribosomes and the increase in the amino acid incorporation activity of ribosomes were detected approximately 30 h after triiodothyronine injection [2]. We found a significant increase in protein kinase activity only after 2 daily injcctions of the hormone. Although it is not possible to compare measurements made in different laboratories, the fact that a single injection of triiodothyronine failed to produce a significant effect on protein kinase activity when it would have had a dramatic effect on transcription, casts doubt on the relationship between the two processes. Nevertheless the demonstration that the nuclear receptor of triiodothyronine is a chromatin nonhistone protein, the demonstration that these proteins and that Chromatin protein phosphorylation play a fundamental role in the regulation of gene activity [11 - 161, the absence of effect of compoundschemically related to triiodothyronine but devoid from any biological activity makes it very likely that the effcct of the hormone on chromatin protein kinase activity is relevant to its physiological action. The authorsare much indebted to Drs N. Defer and B. Dastugue for helpful discussions, and to Dr N. Etling for analyzing purity of trrrodolhyroninc and providing 3,3',5'-triiodothyronine and 3.5-diiodothyroninc.
This work was supported by Grants from thc lnstiiur :Vur,oitcil cle lu Ri~cltmhcMcYicule, thc ('iviire h'urioirul dc lu Rucherelie Scio,ri/irluc and thc Uc;ltgurinn Gitt+rdi>u lr Rivltcrc.N.he
(k.lu Suirii ri
Scienr:fiquv er Technique.
REFERENCES 1 . Greengard. 0. (1970)in Biocli~miccrl Acrinns n/' Hornmnes (Litwack, Ci., cd.) vol. 1. pp. 53-87. Acadcmic P r w . Ncw York and London. 2. Tala, J. R. & Widnell, C. C. (1966)Bicchem. J . 98,604-620. 3. Tata. J. R. (1966)Pmg. Nuclvic Acid Res. Mol. Uinl. 5, 191
-
250. 4. Wyatt, G.R . & Tata, J. R . (1968)BincAm. J . IOY, 253-258. 5. Smuckler. E. A. & Tata, J. R. (1971) Nuiure 1 Lond.) 234. 37- 39. 6. Jothy, S.. Bilodeau, J. L., Champsaur. H. B Simpkins, H. ( 1975)Biocltt*ni.J . 150, 133 - 135. 7. Tata, J. R. & Williams-Ashman, H. (3. (1967)Eur. J . Binc~hrm. 2.366- 374. 8. Surks. M.I., Kacrncr. D., Dillman, W. & Oppcnhcimer, J. H. (1973).I. B i d . Chem. 2448.7066-7072. 9. Samucls, H. H. & Tsai. J. S. (1973) Proc. Nurl Acurl. Sci. U.S.A. 70.3488- 3492. 10. Thomopoulos, P., Dastuguc, B. & Defer. N. (1974)Biocltmi. Biophys. Rcs. Cornmun~58,499- 506. 11. Kamiyama. M. & Wang.T. Y . (1971)Bicichim. Biop1t.w. Actu. 225,563- 576. 12. Kamiyama, M., Dastuguc, B. B Kruh, J. (1971) Biochem. Biopltw. Res. Contmun. 44.29- 36. 13. Kamiyama, M., Dastugue, B., Defer, N. & Kruh, J. (1972) Binchint. Rinpltys. Acta, 277.576 - 583. 14. Stein, G . S.,Spelsberg, T.C. & Kleinsmith. L. J. (1974)Sciiwec~ ( Wu.vh. D.C.) 183.8 17- 824. 15. Kruh, J. Dastugue. B., Defer, N., Kamiyama, M. & Tichonicky, I,. (1975)Biochimie, 56,995- 1001. 16. Kruh, J., Courtois, Y., Dastugue, B.. Defer, N., Gibson, K., Kamiyama, M. & Tichonicky, L. (1975)in Regularion I$ Growdi and D@wvttiated Function in Eukuryctic Cel1.s.
Raven Press, New York, p. 139- 154. 17. Dastugue. B., Tichonicky, L. B Kruh, J. (1974) Biochimie, 56,491- 500. 18. Gibson, K., Tichonicky, L. & Kruh. J . (1974)Binchimic. 56. 1417- 1423. 19. Gibson, K.. Tichonicky. L. & Kruh, J. (1975)Mnl. Cell. Bioclieni. Y, 79- 83. 20. Correze, C., Pincll, P. & Nunez, J. (1972)FEES Lerr. 23.87- 91. 21. Chauveau, J., Mouk, Y. & Rouiller. C. (1956)E.rp. Cell Rus. 11. 317- 321. 22. Takcda. M., Yamamura, H. & Ohga. Y. (1971)Biocliem. Bioplrys. Rvs. Commun. 42, 103- 110. 23. Dastugue, B.. Tichonicky, L. & Kruh, J. (1973)Biochimic, 55. 1021- 1030. 24. Gibson. K., Tichonicky, L. & Kruh. J. (1974)Biochimii., 56, 1409- 1416. 25. Rodbard, D. & Chrambach, A. (1971) Anal. Binchetn. 40, 95-134. 26. Chrambach, A. & Rodbard, D. (1971)Science ( Wmh. D.C.) 172,440-451. 27. Gill, G . N. & Garren, L. D. (1971) Proc. Nor1 Acud. Sci. U.S.A. 68,786-790. 28. Johns, E. W. (1964)Biochem. J . 92,55-59. 29. Kakkar, H. M. (1947)J. Biol. Chem. 167,461-469.
J . Kruh and L. Tichonicky, Institut de Palhologie Mokukdire, 24,rue du Faubourg St-Jacques, F-75014Paris, France
