Jun 10, 2016 - by EF-Ts, it follows that the dissociation of GDP in the presence of .... to the vial and the radioactivity retained by the filter was determined in a liquid .... The exchange reaction can be represented by the following reaction .... GDP)T, is much smaller than ... Using this value, kc, Kb, and K& are estimated to be.
T H E .]OURNAI. O F BlOLOGlCAL CHEMISTRY
Vol. 156. No. I I . Issue of .June 10, pp. 5591-5596.1981 Pr'rmted m U.S A.
Kinetic Studies on the Interactionsof Escherichia coli K12 Elongation Factor Tu with GDP and Elongation FactorTs* (Received for publication, August 20, 1980, and in revised form, February 5, 1981)
Vincent Chaul, Guillermo Romerog,and Rodney L. Biltonen From the Departmentsof Biochemistry and Phermacology, University of Virginia School of Medicine, Charlottesuille, Virginia 22908
The kinetic parameters describing the dissociation of GDP from EF-Tu by forming anEF-TueEF-Ts complex GDP from the elongation factor Tu (EF-Tu) GDP com- which has a stability comparable to that of the EF-TU- GDP plex in the absence and presence of elongation factor complex (2,3).A mechanism for the regeneration of the active Ts (EF-Ts) have been characterized using an equilib- EF-TU. GTPcomplex based upon this finding has been prorium isotope exchange technique. The rate constant for posed (8): dissociation of GDP from EF-Tu was found to be 1.7 X EF-TU GDP + EF-TS EF-Tu. EF-TS + GDP s". Since this dissociation rate is greatly enhanced by EF-Ts, it followsthat the dissociation of GDP in the EF-Tu. EF-TS+ GTP G? EF-Tu. GTP+ EF-TS presence of EF-Tsproceeds via the formation of a ternary EF-TuoGDP-EF-Ts complex as represented be- However, the exchange of the bound GDP from the EF-Tu. GDP complex with guanine nucleotide free in solution is low : greatly accelerated by EF-Ts. This result strongly suggested EF-TU GDP + EF-TS e EF-Tu* GDP EF-TS @ that a ternary complex of EF-Tu, GDP, andEF-Ts exists, and EF-TU. EF-TS + GDP that the rapid observed exchange of GDP occurred via this complex. In order to establish whether this is the case and Analysis of the exchange kinetics according to this define clearly the situation, a detailed kinetic study of this reaction scheme yields a rate constant for the dissocia- dissociation reaction in the presence and absence of EF-Ts tion of GDP from the ternary complex of 21270 s-'. The equilibrium association constants for GDP and EF-Ts using equilibrium isotope exchange techniques was initiated. with a mechanism in to form the ternary complex were found to be 6.4 X lo4 The resultsobtainedareconsistent which the rapid dissociation of GDP in the presence of EF-Ts M" and 1.8 X lo5 M-', respectively. These results demonstrate that the dissociation of GDP from EF-Tuin the is indeed due to the transientformation of a ternary EF-Tu a detailed analysis of the presence of EF-Ts is not the rate-limiting process in GDP .EF-Ts complex. In this report, EF-Ts-catalyzed GDP exchange reaction is presented. protein synthesis.
-
EXPERIMENTALPROCEDURES
Materials-E. coli K12 frozen cell paste (% log phase) was purThe elongation factors Tu and Ts participate in the poly- chased from Grain Processing Co. (Muscatine, Iowa) and stored at "C until used. ["HIGDP disodium salt (9 mCi/pmol) in 50% peptidechain elongation cycle during proteinsynthesis in -20 ethanol was obtained from New England Nuclear and used without prokaryotes (1).These proteins have been purified to apparent further purification. Nitrocellulose fiiter discs (Schleicher and Schuell homogeneity from Escherichia coli. (2, 3). EF-Tu' in the type BA 85, 25-mm diameter) were used in all the filtration experipresence of GDP promotes the binding of aminoacyl-tRNA to ments described. GDP from Sigma was used without further purifiribosomes (4, 5 ) . Subsequently, GTP is hydrolyzed to GDP cation. All other chemicals were of analytical grade. Methods-Elongation factors were purified by a minor modificaand P, (6, 7). GDP and EF-Tu are then released from the ribosome in the form of an EF-Tu. GDP complex. This se- tion of the procedures of Arai et el. (3).After separation of the EF-Tu. GDP and EF-Tu.EF-Ts complexes by DEAE cellulose chromatogquence of reactions can be schematically represented as fol- raphy, both complexes were chromatographed on two separate hylows: droxyapatite columns (2 X 40 cm). Bothcolumns were equilibrated in 1 mM potassium phosphate buffer (pH 6.8) containing 1 mM dithioEF-Tu. GTP+ aa-tRNA G? EF-Tu. GTP. aa-tRNA threitol. After loading the protein solution, the hydroxyapatite columns were washed successively with 200 ml of 0.01 M phosphate and + ribosome G? EF-Tu. GTP. aa-tRNA 200 ml of 0.02 M phosphate. The EF-Tu complexes were eluted with ribosome.aa-tRNA + EF-Tu.GDP + P, a linear potassium phosphate gradient (1 liter each of 0.02 M potasphosphateand 0.2 M potassium phosphate). All phosphate It has been observed that EF-Ts can quantitatively displace sium solutions were equilibrated at pH6.8 and contained 1 mM dithiothreitol. This modification, successfully employed by Wittinghofer and * This work was supported by United States Public Health Service Leberman (9) in the purification of Bacillusstearotherrnophilus Grants GM20637 and GM26894. The costs of publication of this elongation factors, yielded electrophoretically homogeneous EF-Tu. article were defrayed in part by the payment of page charges. This GDP and EF-Tu. EF-Ts complexes. article must therefore be hereby marked "advertisernerd" in accordThe EF-Tu. EF-Ts complex was concentrated and dialyzed against ance with 18 U.S.C. Section 1734 solely to indicate this fact. 0.02 M potassium phosphate buffer containing 0.1 M NaC1,10 mM Present address, National Institutes of Health, Bethesda, Md. MgCL, and 1 mM dithiothreitol (buffer A). EF-Tu.GDP was dialyzed 20205. against buffer A containing 10 p~ GDP. Both protein solutions were 9 Present address, Universdad Cayetano Heredia, Departmento di stored in 50% glycerol at -20 "C and were found to be stable for Biologia, Apartado 5045, Lima, Peru. several months. ' The abbreviations used are:EF-Tu, polypeptide elongation factor EF-Tu Assay-The concentrations of EF-Tu.GDP and EF-Tu. Tu; EF-Ts, polypeptide elongation factor Ts; aa-tRNA, aminoacyl- EF-Ts were measured by theGDP binding and fiitrationassay transfer ribonucleic acid GDP, guanosine 5"diphosphate; GTP, gua- described by Miller and Weissbach (2). Theprotein was incubated at nosine 5"triphosphate; P,, inorganic phosphate. 20 "C with 2-10 PM [:'H]GDP (1 mCi/pmol) in buffer A at a final
+
5591
E .Interactions coli Factor Elongation
5592
volume of 100 pl. At the end of the incubation period (30 min for EFTu-GDP, 5 min for EF-Tu.EF-TS), the reaction was stopped by addition of 2 ml of ice-cold buffer A and fdtered rapidly through a nitrocellulose filter disc. The filter was washed twice with 2-ml aliquots of buffer A and dissolved in 1 ml of 2-methoxyethanol in a 7ml counting vial. Five ml of a standard toluene/2,5-diphenyloxazole/ 1,4-bis[2-(5-phenyloxazolyl)]benzenescintillation mixture were added to thevial and the radioactivity retained by the filter was determined in a liquid scintillation counter. Preparation of EF-Tu.f 3H]GDP--The EF-TuFHlGDP complex was prepared by incubation of the purified EF-Tu-GDP complex with 0.1 p~ [JH]GDP (8 mCi/pmol) at 20 "C in a final volume of 15 ml. The formation of the radioactive EF-Tu-GDP complex was monitored by the fdtration assay described above. EquilibriumIsotope Exchange-All exchange experiments described herein were carried out in buffer A at 21 "C. In the absence of EF-Ts, the reaction was initiated by addition of 5 ml of a 50 p~ solution of unlabeled GDP to an equal volume of EF-Tu-r'HlGDP complex prepared as described above. Atspecified times, 0.5-ml aliquots of the reaction mixture were withdrawn, the reaction was stopped by rapid fitration of the aliquot through a nitrocellulose filter disc, and the amount of radioactive complex was measured as described. Because the EF-Tu.["H]GDP complex was prepared by incubating the protein with radioactive GDP at a concentration about 30-fold higher than the GDP dissociation constant, the additional unlabeled GDP added to initiate the exchange reaction had no appreciable effect on the state of dissociation of the complex. Therefore, the total concentration of complex will remain unchanged during the time course of the reaction, ie. the exchange reaction occurs under equilibrium conditions. The exchange reaction in the presence of EF-Ts was carried out in a rapid mixing filtration apparatus (Fig. 1 ) described in detail else-
mixinq unit
where (IO). Briefly, the apparatus consists of separate mixing and fdtration units connected to one another. The mixing unit consists of two thermostated syringes connected to a tangential mixer. The filtration unit is composed of a thermostated filter holder connected to a vacuum pump. The two reaction solutions are loaded separately into the thermostated syringes of the mixing unit. The reaction is initiated by simultaneously driving both solutions through the mixing chamber, delivering the reaction mixture onto the filter disc. The reaction mixture is allowed to incubate on the filter, and after apreset time that is controlled by a variable time switch, the reaction is stopped by rapid filtration of the solution. The filter is then washed rapidly, and the amount of radioactive EF-Tu. GDP complex remaining on the filter is measured as described. The mixing time of the mixing chamber used for the construction of the apparatus is a few milliseconds.However, the time resolution of the apparatus is a function of the total volume of the reaction mixture. The rate of filtration of the solution through the nitrocellulose filter disc is about 5 ml/s, and the typical volume of reaction mixtures for each experiment was 1 ml. Thus, the time resolution of the apparatus under the current experimental conditions is about 200 ms. Analysis of the Kinetic Data-The data obtained from the equilibrium exchange experiments were fit to a single exponential function by a nonlinear least squares fitting procedure (11, 12). The results were interpreted in terms of a generalized reaction scheme described in the Appendix. RESULTS
Exchange in the Absence of EF-Ts-EF-Tu isolated from
E. coli contains a single, tightly bound GDP molecule ( 2 ) . This bound GDP is exchangeable with GDP free in solution. Fig. 2A shows the time course of the exchange of r3H]GDP bound to EF-Tu with unlabeled GDP free in solution. The amount of EF-Tu- ["HIGDPcomplex at any given time was assayed by rapidly passing an aliquot of the reaction mixture through a nitrocellulose filter disc. Since only GDP bound to EF-Tu is retained after the fiter disc is washed with buffer, the amount of radioactivity on the filter disc measures the amount of the EF-Tu.["HIGDP complex. The exchange reaction can be represented by the following reaction scheme: k-1 EF-Tu.r3H]GDP c3 EF-Tu kl EF-TU + GDP
k, " +
+ [3H]GDP (1)
EF-Tu.GDP
X 1
When the amount of unlabeled GDP is in large excess, the reverse reaction is negligible, and the rateequation is
;'!K
-d(EF-Tu. ['H]GDP)/dt = k-l(EF-Tu. [JHIGDP)
(2a)
which, upon integration and application of the appropriate boundary conditions, yields (EF-Tu.["H]GDP)
filiratlm unit
=
(EF-Tu. rH]GDPh exp (-k-,t)
Ob)
.0
reservoir
FIG. 1. Schematic diagramof the rapidfiltration apparatus. The mixing unit consists of two thermostated syringes mounted on an aluminum block, a mixing chamber connected to theoutlets of the reagent syringes, and a mechanically actuated plunger that controls the volume delivered by the reagent syringes. The fitration unit consists of a thermostated filter holder connected to a vacuum pump via a vacuum flask that serves at thesame time as a waste container. The vacuum pump is connected to the fiter holder via a solenoid valve controlled by an electronic timer. The movement of the plunger initiates the reaction by driving an equal volume of each reagent syringe onto the filter holder. Simultaneously, a microswitch activates the electronic timer. At the end of the incubation period, the solenoid valve is opened by the timer, exposing the fiter to thevacuum line. The reaction is stopped by the rapid filtration of the reaction mixture. The mixing chamber was carefully cleaned after each experiment.
B
20
8 10
8 -
0
0
SECONDS
200 400 600 BOO I000 1200
SECONDS
FIG. 2. The kinetics of the GDP exchange reaction in the absence of EF-Ts. The reaction mixture contained initially 10 nM EF-Tu. [3H]GDP and 1 p~ GDP. The dissociation of the EF-Tu. [?H]GDp complex was monitored as described under "Materials and Metho&." A , time course of the reaction; B, semilogarithmic plot of the data. The curues were obtained by fitting the data to Equation 2b. The first order rate constant obtained from the fitting procedure was 1.7 (k0.1) X 10"' sK1.
Interactions Factor Elongation E. coli
5593
that in all these exchange experiments, the initial concentration of EF-Tu. ["HIGDPis at least 50-fold greater than the total concentration of EF-Ts. Theseresults demonstrate that EF-Ts enhances the rateof GDP dissociation from EF-TUby a catalytic process. The dependence of kohson the initial concentration of EFTu.["H]GDP was studied at 0.5 nM EF-TS and 50 pM GDP. Values of kohsat various concentrations of EF-Tu (from 20 nM to 6.8 p ~ are ) given in Table I. 0 IO 20 30 Mechanism of the EF-Ts-mediated Exchange-The enSECONDS SECONDS hanced exchange rate observed in the presence of EF-Ts can FIG. 3. The [3H]GDP exchange reaction in the presence of best be explained by a reaction mechanism in which EF-Ts EF-Ts. The reaction mixture contained initially 50 nM EF-Tu. interacts with EF-Tu. GDP to form a ternary complex be['HIGDP, 50 p~ GDP, and 0.5 nM EF-Ts. The time course of the which is in rapidequilibreaction is shown in A . The semilogarithmic plot of the data is shown tween EF-Tu, EF-Ts, and GDP, and in B. The solid lines were obtained by fitting the data to a single rium with GDP and the EF-Tu. GDP and EF-Tu. EF-Ts exponential function. The first order rate constant of the reaction, complexes. The simplest mechanism involving such a ternary k,,, was calculated to be 0.113 f 0.003 s-I. complex is: where (EF-Tu. [:'H]GDP) and (EF-Tu. [:IH]GDP)o are the concentrations of EF-Tu. ["HIGDP at any time t and time zero, respectively. Fig. 2B is a semilogarithmic plot of the data in Fig. 2 4 . As shown, the exchange reaction can be described accurately by a single exponential function. An average value of 1.7 (kO.1) X lo-,'' s-' was calculated for k - , from three separate experiments at 21°C. This value is in reasonably good agreement with results reportedpreviously (2,13).Since unlabeled GDP wasin large excess, huhs, as expected, was found to be independentof the initial concentration of EF-Tu. [?H]GDP (20 nM to 5 p ~ in) the reaction mixture. Exchange in the Presence of EF-Ts-It has been shown previously that the exchange of the ["HIGDP bound to EFTu is accelerated greatly in the presence of EF-TS (2, 3 ) . Fig. 3 shows the effect of EF-Ts in the kinetics of theGDP exchange reaction. The time course of the reaction was followedin the rapid filtration apparatus as described above. The reaction was initiated by mixing 0.5 ml of 0.1 p~ EF-Tu['HIGDP with an equal volume of a solution containing 100 GDP and 1 nM EF-Tu. EF-Ts. Thesemilogarithmic representation of the data (Fig. 3B) shows that the exchange in the presence of EF-Ts is also described by a single exponential function. hob.;, the observed first order rate constant of the exchange, was calculated to be 0.113 & 0.003 s-I. In order to obtain additional information on the EF-Tsmediated exchange, the reaction was carried out at various concentrations of EF-Tu.["H]GDP andEF-Ts. Under all conditions studied,the exchange reaction was found to follow first order kinetics. Fig. 4 shows that kc,hsis linearly dependent on the totalconcentrations of EF-Ts. Itshould be emphasized
EF-TU.['HIGDP
kz
+ EF-TS 2 EF-TU.["HIGDP k-z
k-4
.EF-TS 2 EF-Tu. EF-TS+ ['HIGDP k4
(3)
k4
EF-Tu-EF-Ts + GDP S EF-TU. GDP k-4
k-a -EF-Ts Z EF-TU-GDP+ EF-TS ha
In this mechanism, EF-Ts interacts with EF-Tu-["HIGDP to form the ternary complex, EF-Tu.["H]GDP.EF-Ts. The labeled GDP in the ternarycomplex then exchanges with the unlabeled GDP free in solution. The faster exchange rate observed in the presence of EF-Ts can only be rationalized by assuming that the dissociation rate of [3H]GDP from the ternary complex is faster than the dissociation of VHIGDP from EF-Tu .['H]GDP. Since one EF-Ts facilitates the exchange of several ["HIGDPs, the cycle shown in the above reaction scheme is repeated until complete exchange occurs. Therefore, it is also necessary that the dissociation of EF-Ts from the ternary complex is at least as fast as the observed exchange rate. With this mechanism, the exchange rate is dependent on the concentrations of EF-Tu .["HIGDP, GDP, and EF-Ts.kohs,the first order rate constant of the exchange, as derived in the Appendix, is given by k, .(EF-Ts)T
k h
(GDP)T+ (EF-Tu.GDPh) (GDP),I.(EF-Tu.GDP).I.
where (EF-Ts)~, (EF-Tu.GDP)T, and (GDP)T are the total TABLEI The dependence of k,,,,*on the concentration ofthe EF-Tu.GDP complex EF-Tu .GDP
k,>t,.
WM
In
n
0.113
0
1
0.113 0.098
[EF-Ts] (nM)
FIG.4. The dependence of kohs on the concentration of EFTS.The reaction mixture contained initially 0.1 PM EF-Tu. [ 'HIGDP, 50 PM G D P , and EF-Ts at the concentrations indicated.
0.065
~
~
0.02 0.05 0.075 0.10 0.37 0.76 1.52 2.53 3.78 6.08 _6.83_
~~-
S-'
~
~~
& 0.003 0.114 f 0.003 -C 0.004 0.110 f 0.003 0.106 i 0.002 i 0.002 0.087 i 0.002 0.072 i 0.003 f 0.005 0.053 i 0.001 0.050 f 0.001 "
E. Interactions coli Factor Elongation
5594
concentrations of EF-Ts, EF-Tu. GDP, and GDP, respectively, in the reaction mixture. KGs and K b are association constants defined by the equilibrium scheme below: K EF-TU-GDP+ TS c ' EF-Tu-GDP li\
(5)
K'
