Protein Science ~2000!, 9:1791–1800. Cambridge University Press. Printed in the USA. Copyright © 2000 The Protein Society
Interaction of mammalian mitochondrial elongation factor EF-Tu with guanine nucleotides
YING-CHUN CAI,1 JAMES M. BULLARD,2 NANCY L. THOMPSON,1 and LINDA L. SPREMULLI 1 1
Department of Chemistry, University of North Carolina, Campus Box 3290, Chapel Hill, North Carolina 27599-3290
~Received May 9, 2000; Final Revision July 1, 2000; Accepted July 7, 2000!
Abstract Elongation factor Tu ~EF-Tu! promotes the binding of aminoacyl-tRNA ~aa-tRNA! to the acceptor site of the ribosome. During the elongation cycle, EF-Tu interacts with guanine nucleotides, aa-tRNA and its nucleotide exchange factor ~EF-Ts!. Quantitative determination of the equilibrium dissociation constants that govern the interactions of mammalian mitochondrial EF-Tu ~EF-Tumt ! with guanine nucleotides was the focus of the work reported here. Equilibrium dialysis with @ 3 H#GDP was used to measure the equilibrium dissociation constant of the EF-Tumt{GDP complex ~KGDP 5 1.0 6 0.1 mM!. Competition of GTP with a fluorescent derivative of GDP ~mantGDP! for binding to EF-Tumt was used to measure the dissociation constant of the EF-Tumt{GTP complex ~KGTP 5 18 6 9 mM!. The analysis of these data required information on the dissociation constant of the EF-Tumt{mantGDP complex ~KmGDP 5 2.0 6 0.5 mM!, which was measured by equilibrium dialysis. Both KGDP and KGTP for EF-Tumt are quite different ~about two orders of magnitude higher! than the dissociation constants of the corresponding complexes formed by Escherichia coli EF-Tu. The forward and reverse rate constants for the association and dissociation of the EF-Tumt{GDP complex were determined using the change in the fluorescence of mantGDP upon interaction with EF-Tumt . These values are in agreement with a simple equilibrium binding interaction between EF-Tumt and GDP. The results obtained are discussed in terms of the recently described crystal structure of the EF-Tumt{GDP complex. Keywords: elongation; elongation factor Tu; guanine nucleotides; mitochondria; protein synthesis; translation The binding of aminoacyl-tRNA ~aa-tRNA! to the acceptor site ~A-site! of the ribosome is mediated by elongation factor Tu ~EF-Tu!. EF-Tu is a member of the superfamily of the guanine nucleotide-binding proteins capable of binding either GTP or GDP. The EF-Tu{GTP complex binds aa-tRNA forming a ternary complex that positions the aa-tRNA in the A-site of the ribosome. Following codon:anticodon interaction, GTP is hydrolyzed and the inactive EF-Tu{GDP complex is released. Exchange of GDP for GTP is promoted by elongation factor Ts ~EF-Ts! that functions as a guanine nucleotide exchange factor ~Miller & Weissbach, 1977!. Mitochondrial elongation factor Tu ~EF-Tumt ! and elongation factor Ts ~EF-Tsmt ! have been purified from bovine liver as a tightly associated complex ~EF-Tu{Tsmt ! ~Schwartzbach & Spremulli, 1989!. EF-Tumt and EF-Tsmt of bovine and human origins have been cloned and sequenced ~Woriax et al., 1995; Xin et al., 1995!. Mature bovine EF-Tumt is 56 and 59% identical to the corresponding factors from Escherichia coli and Ther-
Reprint requests to: Linda L. Spremulli, Department of Chemistry, University of North Carolina, Campus Box 3290, Chapel Hill, North Carolina 27599; e-mail:
[email protected]. 2 Present address: University of Colorado Health Sciences Center, Biomedical Research Building, Room 311, 4200 E. 9th Avenue, Denver, Colorado 80262.
mus thermophilus, respectively. Like the T. thermophilus elongation factors, EF-Tumt and EF-Tsmt form a tight complex that cannot be dissociated even in the presence of high concentrations of guanine nucleotides ~Schwartzbach & Spremulli, 1989!. EF-Tumt is capable of forming a ternary complex with GTP and E. coli Phe-tRNA that binds to the A-site of the ribosome ~Schwartzbach & Spremulli, 1991; Woriax et al., 1996, 1997!. However, in the presence of EF-Tsmt , no significant amounts of intermediates equivalent to bacterial EF-Tu{GTP or EF-Tu{GDP can be detected in the mitochondrial system. These observations suggest that the equilibrium constants governing the interaction between EF-Tumt and its ligands may be considerably different from those observed in E. coli. EF-Tu has been subjected to extensive crystallographic studies, and the structures of several EF-Tu complexes have been determined. EF-Tu folds into three domains with guanine nucleotides binding exclusively to domain 1. The crystal structures of EF-Tu from Thermus aquaticus complexed with GDP and complexed with the nonhydrolyzable GTP analogue, guanosine-59-~ b, g-imido! triphosphate ~GDPNP! have been solved at 2.7 and 2.5 Å resolution, respectively ~Kjeldgaard et al., 1993; Polekhina et al., 1996!. The three-dimensional ~3D! structure of the bovine EF-Tumt{GDP complex has recently been determined ~Andersen et al., 2000!. These studies indicate that the overall structure of the mitochondrial factor is similar to that of prokaryotic EF-Tu. However, there
1791
1792
Y.-C. Cai et al.
are subtle structural differences between the bacterial and mitochondrial factors, and the orientation of domain 1 relative to domains 2 and 3 is different from that observed in E. coli EF-Tu ~Andersen et al., 2000!. In the current work, we have examined the binding constants that govern the interaction of EF-Tumt with GDP and GTP and have sought to explain the observations in terms of the 3D structures of EF-Tumt and E. coli EF-Tu. Results Equilibrium dissociation constant of the EF-Tumt{GDP complex Preliminary studies indicated that the binding of GDP to EF-Tumt was too weak to be measured by the traditional nitrocellulose filter binding assay or by gel filtration chromatography ~data not shown!. Thus, equilibrium dialysis was developed to assess the dissociation constant governing the formation of the EF-Tumt{GDP complex. Use of this approach was complicated by the difficulty of obtaining EF-Tumt in a nucleotide free form. Although several methods were used in an attempt to purify a nucleotide free form of EF-Tumt , none of these methods proved successful ~data not shown!. In general, EF-Tumt aggregated and precipitated out of solution very quickly following the removal of GDP and Mg 21 . Hence, EF-Tumt was prepared in the presence of low concentrations of GDP and the amount of GDP present in each sample was quantified. The total amount of GDP present in the equilibrium dialysis experiments was then corrected to account for the amount of GDP present in the EF-Tumt preparation ~see Materials and methods!. In general, it was necessary to have at least 0.2 mol of GDP present per mol of EF-Tumt to maintain the solubility of the factor. The calculation of the equilibrium dissociation constant ~KGDP ! for the EF-Tumt{GDP complex requires an evaluation of the percentage of the EF-Tumt that is capable of binding guanine nucleotides. This value was determined using different concentrations of GDP ~from 10 to 60 mM! in the dialysis buffer and 10 mM EFTumt containing a small amount of GDP present inside the bag. At 20 mM GDP, 91% of the EF-Tumt bound to GDP. At higher external concentrations of GDP ~40 and 60 mM!, the specific activity of the @ 3 H#GDP ~8.4 and 4.7 cpm0pmol! was too low to provide an accurate number. Hence, at least 91% of the EF-Tumt was capable of binding to GDP at subsaturating concentrations of GDP. Therefore, it was assumed that 100% of the soluble EF-Tumt was active in binding to GDP. To calculate KGDP , various concentrations of EF-Tumt ~containing small amounts of GDP! were dialyzed against buffer containing several concentrations of GDP including @ 3 H#GDP as a tracer. Following dialysis, the amounts of GDP inside and outside the dialysis bag were quantified from the amount of label present, and the total EF-Tumt concentration was assayed. These data were then used to calculate the amount of EF-Tumt{GDP, free GDP and free EF-Tumt present. By assuming the following mechanism, kd1
EF-Tumt 1 GDP
^
kd2
&
EF-Tumt{GDP,
~1!
the equilibrium dissociation constant for the EF-Tumt{GDP complex was calculated as KGDP 5
kd 2 k d1
5
@EF-Tumt # @GDP# @EF-Tumt{GDP#
.
~2!
Based on seventeen independent assays the value of KGDP was determined to be 1.0 6 0.1 mM. Equilibrium dissociation constant of the EF-Tumt{mantGDP complex To determine the equilibrium dissociation constant of the EF-Tumt{ GTP complex using competition assays with 29-~or 39-!-O-~Nmethylanthraniloyl! guanosine 59-diphosphate ~mantGDP! ~see below!, we first needed to determine the dissociation constant of the EF-Tumt{mantGDP complex ~KmGDP !. This value was again determined using equilibrium dialysis. Following dialysis, the amounts of mantGDP inside and outside the dialysis bag were determined by fluorescence spectroscopy and the total EF-Tumt concentration inside the bag was also determined. These data were used to calculate the concentrations of EF-Tumt{mantGDP, free EF-Tumt , and free mantGDP inside the bag. By assuming the following mechanism km1
ET-Tumt 1 mantGDP
^
km2
&
EF-Tumt{mantGDP,
~3!
the equilibrium dissociation constant for the EF-Tumt{mantGDP complex was calculated as Km GDP 5
k m2 k m1
5
@EF-Tumt # @mantGDP# @EF-Tumt{mantGDP#
.
~4!
Based on four independent measurements, the equilibrium dissociation constant of the EF-Tumt{mantGDP complex was found to be KmGDP 5 2.0 6 0.5 mM. Thus the dissociation constant is approximately twice that seen for the EF-Tumt{GDP complex. Effect of EF-Tumt on the fluorescence properties of mantGDP Preliminary analysis indicated that the binding of GTP to EF-Tumt is significantly weaker than the binding of GDP. Hence, it was not possible to use equilibrium dialysis to determine the equilibrium dissociation constant for the EF-Tumt{GTP complex. To obtain a value for this constant, a competition assay between mantGDP and GTP was developed. This approach required an initial investigation into the interaction of mantGDP with EF-Tumt . Mant nucleotide derivatives have been used in investigations of nucleotide binding to a number of proteins, including Ras ~John et al., 1990, 1993; Hazlett et al., 1993; Moore et al., 1993; Nixon et al., 1995!, Cdc42H ~Leonard et al., 1994; Nomanbhoy & Cerione, 1996!, F1 ATPase ~Divita et al., 1993!, MaxA ~Richter et al., 1995!, and EF-Tu ~Giovane et al., 1995; Wagner et al., 1995; Watson et al., 1995!. In all cases, interactions of the fluorescent derivatives with the proteins led to substantial increases in fluorescence quantum yields. It seems likely that the fluorescence enhancement seen when mant-conjugated nucleotides bind to proteins is due to alleviation of quenching interactions between the fluorescent group and solvent molecules. The crystal structures of EF-Tu bound to guanine nucleotides show that there is little direct interaction of residues from EF-Tu with either the 29 or 39 ribose hydroxyl groups of the guanine nucleotide ~Kjeldgaard et al., 1993; Sprinzl, 1994; Abel et al., 1996; Polekhina et al., 1996; Song et al., 1999!. Both the 29 and 39 hydroxyl groups of GDP are oriented toward the surface of the protein. Furthermore, previous results have indicated that either dGTP or ddGTP can promote the EF-Tumt directed binding of aminoacyl-tRNA to ribosomes ~Woriax et al., 1996!.
1793
Mitochondrial elongation factor Tu These results indicate that neither of these ribose hydroxyl groups plays a critical role in the interaction of EF-Tu with guanine nucleotides. Based on these observations, it was reasonable to use GDP with mant conjugated to the 29 or 39 hydroxyl to measure the binding of GTP to EF-Tumt . The change in the fluorescence emission of mantGDP upon binding to EF-Tumt was first investigated by adding EF-Tumt{GDP to solutions containing mantGDP. With excitation at 350 nm, mantGDP appeared to have an emission peak around 440 to 450 nm. The effect of EF-Tumt on the fluorescence emission spectrum was then examined. Since EF-Tumt could not be purified in a nucleotide free form, EF-Tumt{GDP preparations were used. During incubation, free EF-Tumt in this preparation could bind mantGDP. Furthermore, exchange of GDP in the EF-Tumt{GDP complex with mantGDP could occur allowing an examination of the effect of EF-Tumt on the fluorescence properties of mantGDP. Upon binding to EF-Tumt , the fluorescence emission increased although the shape of the emission spectrum did not change significantly ~data not shown!. To ensure that the change in fluorescence intensity observed in the presence EF-Tumt was dependent on the concentration of EF-Tumt , the fluorescence emission of mantGDP was measured as a function of the concentration of EF-Tumt at a constant concentration of mantGDP. Under these conditions, an increase in the fluorescence emission was observed in response to increasing concentrations of EF-Tumt . The increase in fluorescence emission was not due to the EF-Tumt alone since the emission of EF-Tumt in the absence of mantGDP was very low ~data not shown!. These experiments demonstrate that the fluorescence emission intensity of mantGDP is higher when it is bound to EFTumt than when it is free in solution. Relative fluorescence of mantGDP when free and when bound to EF-Tumt To determine the dissociation constant for EF-Tumt{GTP, we needed to determine the relative fluorescence of mantGDP in the free and bound states. To obtain this value, competition assays between GDP and mantGDP for binding to EF-Tumt were carried out in the fluorescence assay. In this approach, the fluorescence intensity of an equilibrium mixture of EF-Tumt and mantGDP was obtained. Increasing concentrations of GDP were then added and the fluorescence signal was monitored after equilibrium had been reestablished. As indicated in Figure 1, increasing concentrations of GDP resulted in a decrease in the fluorescence observed as the mantGDP was replaced by GDP in the complex with EF-Tumt . The data obtained using this approach can be analyzed by considering the mechanisms shown in Equations 1 and 3 when they are coupled. In these experiments, the values of KGDP and KmGDP determined by equilibrium dialysis, along with the known total concentrations of EF-Tumt , GDP, and mantGDP, were used to calculate the concentration of mantGDP bound to EF-Tumt . These calculations were carried out using the Mathematica software package to algebraically solve Equations 2, 4, and 5 for @EF-Tumt{mantGDP#, where @EF-Tumt # tot 5 @EF-Tumt # 1 @EF-Tumt{GDP# 1 @EF-Tumt{mantGDP# @GDP# tot 5 @GDP# 1 @EF-Tumt{GDP# @mantGDP# tot 5 @mantGDP# 1 @EF-Tumt{mantGDP#.
~5A! ~5B! ~5C!
Fig. 1. Competition between GDP and mantGDP for binding to EF-Tumt . A representative data set is shown. Different concentrations of GDP were incubated with 1 mM EF-Tumt and 0.36 mM mantGDP in 2 mL samples at 4 8C. Fluorescence intensities were monitored after equilibrium was reached. Points represent the experimental data while the line was from the predicted fluorescence calculated as described in the text.
In these equations, the subscript “tot” signifies the total concentration. The values of @mantGDP# tot and @EF-Tumt{mantGDP# were used to determine the fluorescence enhancement occurring when mantGDP binds to EF-Tumt . The total fluorescence intensity F of mantGDP in the presence of EF-Tumt can be expressed as ~Cantor & Schimmel, 1980; Lakowicz, 1999! F 5 A$@mantGDP# tot 1 ~B 2 1!@EF-Tumt{mantGDP#%
~6!
where F is the total fluorescence intensity, A is an instrumental constant, and B is the relative fluorescence of EF-Tumt{mantGDP to free mantGDP. Plots of F as a function of @EF-Tumt{mantGDP# were fit to lines in which the intercepts were A@mantGDP# tot and the slopes were A~B 2 1! ~data not shown!. The emission intensity increase of mantGDP upon binding to EF-Tumt was calculated from the ratios of the slopes to the intercepts. The value of B ~1.75 6 0.06! was determined in three independent experiments. The values of A and B were used to predict the fluorescence signal at each concentration of @GDP# tot . The predicted values agreed well with the experimentally obtained data ~Fig. 1!. The circles in this figure represent the experimental data while the line was calculated from the values of B, @EF-Tumt # tot , @GDP# tot , @mantGDP# tot , KGDP , and KmGDP . The calculated line fits the experimental data quite well. This result indicates that the data obtained by the fluorescence competition assay are consistent with the KmGDP determined by equilibrium dialysis if the fluorescence intensity of mantGDP is enhanced 75% upon binding to EF-Tumt .
1794
Y.-C. Cai et al.
Time course of association and dissociation of EF-Tumt with mantGDP
R~t ! 5
The association and dissociation kinetics of EF-Tumt with mantGDP were investigated to obtain information on the rate constants governing the formation and dissociation of the EF-Tumt{mantGDP complex. These measurements were carried out using fluorescence spectroscopy. In general, the rate of change of the fluorescence with time in these measurements is given by the rate of change of the concentration of the EF-Tumt{mantGDP complex ~Equation 6!. Association kinetics were monitored by treating solutions of mantGDP with preparations containing EF-Tumt and GDP. In these measurements, the time-dependent fluorescence displayed biphasic kinetics. This observation can be explained by the occurrence of two separate reactions during the incubation period. In the first ~rapid! phase, the free EF-Tumt in the preparation binds to mantGDP ~km1 , Equation 3!. The slower phase represents the dissociation of the EF-Tumt{GDP complex ~kd2 , Equation 1!. The slow dissociation step is followed by binding of mantGDP to the EFTumt that has released GDP. As shown in Figure 2A, a rapid increase in the fluorescence emission was observed after the addition of EF-Tumt ~with GDP! to mantGDP. This initial increase in the fluorescence signal reflects the association of mantGDP with the free EF-Tumt in the preparation and is a measure of the time-dependent concentration of the EF-Tumt{mantGDP complex. At these early times, d dt
@EF-Tumt{mantGDP# ' km1 @EF-Tumt # @mantGDP#
~7!
where km1 is the association rate constant for the formation of the complex ~Equation 3!. The use of Equation 7 assumes that there are no hidden intermediates present in the reaction. The integrated form of Equation 7 for unequal concentrations of reactants is ~Capellos & Bielski, 1980! ln@R~t !# @mantGDP# 0 2 @EF-Tumt # 0
5 k m1 t
~8A!
@EF 2 Tumt # 0 ~ @mantGDP# 0 2 @EF-Tumt{mantGDP# ! @mantGDP# 0 ~ @EF-Tumt # 0 2 @EF-Tumt{mantGDP# ! ~8B!
where @mantGDP# 0 and @EF-Tumt # 0 are the initial concentrations of free mantGDP and free EF-Tumt , respectively. The initial concentration of free mantGDP was known from the total amount of mantGDP used in a given measurement. The value of @EF-Tumt # 0 was calculated from the measured value of KGDP and the total amounts of EF-Tumt and GDP that were initially added to the free mantGDP. The time-dependent value of @EFTumt{mantGDP# was found from the time-dependent fluorescence by using Equation 6. In these calculations, the instrumental constant A was determined by the initial fluorescence of mantGDP in the absence of EF-Tumt and the value of the constant B was fixed at the previously determined value. Plots of the left side of Equation 8A vs. t, for early times, were linear ~Fig. 2B!. The slopes of these data, for three independent assays, gave a value for the association rate constant for EF-Tumt and mantGDP of km1 5 ~3 6 1! 3 10 3 M 21 s 21 . The value obtained for the association rate constant is significantly slower than for a diffusion controlled process. The slower rate obtained presumably reflects the number of productive collisions between EF-Tumt and GDP. Since the nucleotide binding site on this factor represents only a small fraction of its surface area, most of the collisions occurring would not be expected to lead to a productive interaction. The dissociation of mantGDP from the EF-Tumt{mantGDP complex was also examined. In these measurements, the time course for the dissociation of mantGDP from EF-Tumt in the presence of GDP was measured by monitoring the fluorescence intensity of the mantGDP following the addition of a large amount of GDP ~500 mM or 1 mM! to samples containing a mixture of EF-Tumt and mantGDP. Since the concentration of GDP was in great excess over the concentration of mantGDP, the EF-Tumt{mantGDP complex dissociated completely with time. Changes in the fluorescence resulting from the forward reaction governed by km1 ~Equation 3! were
Fig. 2. Time course of the association of mantGDP with EF-Tumt . MantGDP ~1 mM! was incubated with EF-Tumt ~2 mM! and GDP ~1.8 mM! in buffer V for the indicated time at 4 8C. Fluorescence was excited at 350 nm and monitored at 446 nm. A: The fluorescence increased with time after the addition of EF-Tumt and GDP, and exhibited biphasic behavior. B: The second order rate constant for the interaction of EF-Tumt with mantGDP was found by plotting the left side of Equation 8A as a function of time for early times ~see text!. The line shows a least-squares linear fit.
1795
Mitochondrial elongation factor Tu not observed, rather EF-Tumt which had released mantGDP, bound to the excess GDP. In this case, d dt
@EF-Tumt{mantGDP# ' 2k m2 @EF-Tumt{mantGDP#.
~9!
As shown in Figure 3A, the fluorescence decreased with time after addition of GDP. Assuming that the dissociation of the EF-Tumt{mantGDP complex is a first-order reaction, the integrated form of Equations 6 and 9 is ln
F
F 2 F` F0 2 F`
G
5 2k m2 t
~10!
where F is the fluorescence at time t, F0 is the initial fluorescence, and F` is the fluorescence at long times. The dissociation rate constant of the EF-Tumt{mantGDP complex was determined by plotting the left side of Equation 10 vs. t ~Fig. 3B!. Linear regression of five independent assays gave a value for the dissociation rate constant equal to km2 5 ~4.0 6 0.1! 3 10 23 s 21 . The dissociation rate constant of the E. coli EF-Tu{GDP complex is 0.38 3 10 23 s 21 ~Laurberg et al., 1998!. Therefore, the dissociation rate constant of EF-Tumt with mantGDP is 10 times higher than that of the comparable E. coli complex. The weaker binding of mantGDP or GDP to EF-Tumt is thus partially due to the higher dissociation rate constant. The equilibrium dissociation constant KmGDP calculated from the dissociation and association rate constants ~Equation 4! is 1.3 6 0.4 mM. This value is equivalent within experimental uncertainty to the value obtained from equilibrium dialysis. Thus, no hidden intermediates appear to occur in the interaction of EF-Tumt with mantGDP. Equilibrium dissociation constant of the EF-Tumt{GTP complex Most of the EF-Tu proteins from different species have a higher affinity for GDP than for GTP. Preliminary results indicated that the binding of bovine EF-Tumt to GTP is also weaker than its
binding to GDP ~K. Leanza & L. Spremulli, unpubl. data!. The weak binding of GTP to EF-Tumt made it difficult to obtain a dissociation constant for this interaction directly. The method used to obtain the binding parameters for GTP was based on the fact that GTP will compete with GDP and mantGDP for binding to EFTumt . This competition can be monitored by following the decrease in the fluorescence emission signal from mantGDP at increasing concentrations of GTP. The use of the competition assay was complicated by the fact that, in any GTP sample, there was a certain amount of GDP produced by the hydrolysis of GTP. Since GDP binds to EF-Tumt more tightly than GTP, it was important to obtain an estimate of the percentage of GDP in the GTP preparations. This assessment was carried out by analysis of the sample using high-performance liquid chromatography ~HPLC!, which clearly separates GDP and GTP ~data not shown!. After every assay, the same GTP sample used in measuring the GTP dissociation constant was assayed for purity using HPLC to check the percentage of GDP in the sample. In most of the samples, GDP accounted for 11% of the total nucleotide. The equilibrium dissociation constant of the EF-Tumt{GTP complex ~KGTP ! was obtained by competition of GTP with mantGDP for binding to EF-Tumt . Different concentrations of GTP were incubated with solutions of EF-Tumt and mantGDP. Fluorescence intensities were measured after equilibrium was reached. The fluorescence change observed as a function of the concentration of GTP added is shown in Figure 4A. This change is caused by the competition of GTP and the contaminating GDP with mantGDP for binding to EF-Tumt . The data are plotted as a function of the total guanine nucleotide concentration added ~GTP and GDP!. The amount of EF-Tumt bound to GTP was calculated as described below and is plotted as a function of the total concentration of GTP in Figure 4B. In the assay, EF-Tumt interacts with GDP ~Equation 1!, mantGDP ~Equation 3!, and GTP: kt 1 EF-Tumt 1 GTP
^
kt 2
&
EF-Tumt{GTP.
Fig. 3. Kinetics of the dissociation of the EF-Tumt{mantGDP complex. An EF-Tumt{mantGDP preparation containing 2.0 mM EF-Tumt was adjusted to a final concentration of 1.0 mM mantGDP in 2 mL to form the EF-Tumt{mantGDP complex. The dissociation of the EF–Tumt{mantGDP complex was monitored at 4 8C as a function of time following the addition of 1 mM GDP. Fluorescence was excited at 350 nm and was monitored at 446 nm. A: Intensity of fluorescence emission vs. time. B: Plot of the left side of Equation 10 vs. time. The line shows a least-squares linear fit.
~11!
1796
Y.-C. Cai et al.
Fig. 4. Ability of GTP to compete with mantGDP for binding to EF-Tumt . Different concentrations of GTP, including the contaminating GDP, were added to an EF-Tumt{mantGDP preparation ~1 mM EF-Tumt and 0.18 mM mantGDP!. A: The fluorescence change caused by the addition of the GTP preparation was monitored and plotted as a function of the total guanine nucleotide added ~GTP and GDP!. B: The amount of the EF-Tumt{GTP complex present at each concentration of GTP was calculated as a function of the concentration of GTP added ~see text!. The total concentration of GTP was corrected for contaminating GDP.
The equilibrium dissociation constant for the EF-Tumt{GTP complex ~KGTP ! is KGTP 5
k t2 k t1
5
@EF-Tumt # @GTP# @EF-Tumt{GTP#
.
~12!
To determine KGTP , it is necessary to know the concentration of free EF-Tumt , the concentration of free GTP and the concentration of the EF-Tumt{GTP complex. These values were affected by the binding of GDP and mantGDP in the sample to EF-Tumt . The three coupled equilibria are described by Equations 2, 4, 5B, 5C, and 12 along with @EF-Tumt # tot 5 @EF-Tumt # 1 @EF-Tumt{GDP#
Discussion
1 @EF-Tumt{mantGDP# 1 @EF-Tumt{GTP# ~13A! @GTP# tot 5 @GTP# 1 @EF-Tumt{GTP#
Finally, the calculation of KGTP required the concentration of free GTP in addition to the concentrations of free EF-Tumt and EF-Tumt complexed to GTP. The total concentration of GTP was the total concentration of guanine nucleotides minus the GDP contaminant present. The concentration of free GTP was calculated from the total concentration of GTP and the concentration of EF-Tumt{GTP by using Equation 13B. KGTP was then found at each point during the titration ~Fig. 4! from Equation 12. Based on three independent assays, the dissociation constant for EF-Tumt with GTP was found to equal 18 6 9 mM. This value is approximately 20-fold higher than the dissociation constant of the EF-Tumt{GDP complex.
~13B!
where the subscript tot refers to the total concentration of a given species. The concentration of @EF-Tumt{mantGDP# was determined from the fluorescence signal by using Equation 6. In this analysis, @mantGDP# tot was known from the total amount of fluorophore added in each measurement. The value of the instrumental constant A ~Equation 6! was determined by adding a saturating amount of GDP ~0.5 mM!, and the value of B was known from previous measurements ~see above!. The concentration of free mantGDP at each point was calculated from the known total concentration of mantGDP and Equation 5C. The concentration of free EF-Tumt was calculated by using the known value of KmGDP and Equation 4. The concentration of the EF-Tumt{GDP complex was calculated by using the known concentration of free EF-Tumt with Equations 2 and 5B. This calculation was possible since the total concentration of GDP was known from HPLC analysis of the GTP preparation. The concentration of the EF-Tumt{GTP complex was then found from Equation 13A and the known total concentration of EF-Tumt .
The equilibrium dissociation constants governing the binding of guanine nucleotides to EF-Tu and its cytoplasmic counterpart EF1a have been reported for a number of systems ~Table 1!. The dissociation constant for the EF-Tumt{GDP complex is more than two orders of magnitude higher than the values observed with either E. coli or T. thermophilus EF-Tu. The equilibrium dissociation constant for the EF-Tumt{GTP complex is also significantly higher ~about 60-fold! than the value observed for prokaryotic EF-Tu. The affinities of EF-Tumt for guanine nucleotides are closer to the values obtained for EF1a, chloroplast EF-Tu, and yeast EF-Tumt than to the values observed with bacterial EF-Tu ~Table 1!. EFTumt resembles the prokaryotic factors in having a significantly higher dissociation constant ~about 20-fold! for the GTP complex than for the GDP complex. The ratio of the equilibrium binding constants obtained for the binding of GDP and GTP to EF-Tumt is close to the ratio observed with E. coli EF-Tu although the values themselves are quite different. Alignment of the amino acid sequences of bovine EF-Tumt with those of the prokaryotic factors indicates that EF-Tumt has a high degree of similarity to them especially in domain 1 where guanine nucleotides are bound. All of the residues in EF-Tu observed in the crystal structures of E. coli and T. thermophilus EF-Tu making contact with GDP, GDPNP, or Mg 21 ~directly or via water molecules! are well conserved in the sequence of EF-Tumt .
1797
Mitochondrial elongation factor Tu Table 1. Equilibrium dissociation constants of EF-Tu from different species with guanine nucleotides
EF-Tu E. coli EF-Tu T. thermophilus EF1a Chloroplast EF-Tu EF-Tumt ~yeast! EF-Tumt ~bovine!
KGDP ~mM!
KmGDP ~mM!
KGTP ~mM!
References listed below a
0.0077 0.0013 0.01–3 0.3 25 1.0 6 0.1
~0.0011! b 0.0031
0.3 0.016 0.6 2.1 5 18 6 9
A B C D E This work
2.0 6 0.5
a A: Miller and Weissbach ~1970!, Louie and Jurnak ~1985!, Giovane et al. ~1995!; B: Wagner et al. ~1995!; C: Slobin and Moller ~1976!, Nagata et al. ~1977!, Roobol and Moller ~1978!, Carvalho et al. ~1984!, Crechet and Parmeggiani ~1986!, Saha and Chakraburtty ~1986!, Kinzy and Merrick ~1991!, van Damme et al. ~1992!, Edmonds et al. ~1998!; D: Sreedharan et al. ~1985!; E: Rosenthal and Bodley ~1987!. b This value was obtained using 39-O-anthraniloyl-GDP at 30 8C.
The crystal structure of bovine EF-Tumt complexed to GDP has recently been determined and is quite similar to the structure of the E. coli EF-Tu{GDP complex ~Fig. 5! ~Abel et al., 1996; Andersen et al., 2000!. Analysis of this structure indicates that the difference in KGDP between E. coli and mitochondrial EF-Tu cannot arise directly from the residues that are in contact with the GDP or the
Mg 21 ion. This observation is in agreement with the sequence conservation mentioned above. The distances from the GDP and Mg 21 to the residues in EF-Tumt that interact with them are close to those observed in prokaryotic EF-Tu. However, the GDP binding site in EF-Tumt appears to be more flexible than the nucleotide binding pocket in E. coli EF-Tu. This conformational flexibility may account at least for the higher dissociation rate constant observed for the EF-Tumt{GDP complex than for the corresponding complex formed with E. coli EF-Tu. The increased flexibility probably arises from several residues that differ between prokaryotic EF-Tu and EF-Tumt . For example, there are two glycine residues leading into the Switch I region in EF-Tumt ~residues 84–85!, which may give this region more structural flexibility compared to the corresponding residues Thr38 and Tyr39 in E. coli EF-Tu ~Fig. 5!. Another possible basis for increased structural flexibility in EF-Tumt is the absence of a salt bridge ~Arg171 to Glu150 in E. coli EF-Tu! that anchors part of the structure near the nucleotide binding site. Further, Trp184 near the region where the guanine base is located in E. coli EF-Tu is replaced by Gly in EF-Tumt . Thus, increased thermal motion is expected in this region of EF-Tumt . A second feature of EF-Tumt that may be affecting the strength of guanine nucleotide binding is the interaction of domain 1 with domain 2 and especially domain 3. The coordinates of the GDP moiety in EF-Tumt were superimposed on those of the GDP in E. coli EF-Tu. When examined in this orientation, it is clear that the overall orientation of domain 1 to domain 3 in EF-Tumt{GDP is shifted compared to the orientation observed in the E. coli structure ~arrow in Fig. 5!. As a result, domain 1 of EF-Tumt is farther
Fig. 5. Comparison of the structures of EF-Tumt{GDP and E. coli EF-Tu{GDP. The crystal structures of EF-Tumt{GDP ~A: PDB number 1D2E obtained at 1.94 Å resolution ~Andersen et al., 2000!, and the structure of E. coli EF-Tu{GDP ~B: PDB number 1EFC obtained at 2.5 Å resolution ~Abel et al., 1996!! were aligned so that the GDP molecules were superimposed using the RasMol model ~Sayle & Milner-White, 1995!. The Mg 21 is shown in red. Domain 1 is in blue with Switch I in yellow. The helical region leading into the linker between domains 1 and 2 ~residues 226–238 in EF-Tumt and the corresponding residues 181–191 in E. coli EF-Tu! are shown in white. Residues G84G85 in EF-Tumt and the corresponding residues T38Y39 in E. coli EF-Tu are shown as ball and stick models in cyan. Numbering for EF-Tumt begins with the first residue of the precursor sequence. The mature form of EF-Tumt begins at residue 44. Domain 2 is shown in purple and domain 3 in green. EF-Tumt has a C-terminal extension seen as the helical region extending from domain 3 ~pink!.
1798
Y.-C. Cai et al.
away from domain 3 and is rotated relative to the orientation of these domains observed in the prokaryotic factors. There are few contacts between domains 1 and 2 in the GDP-bound form of EF-Tu. But the helical region leading into the linker joining domains 1 and 2 is in a slightly different orientation in EF-Tumt than in E. coli EF-Tu ~white helical region in Fig. 5!. A number of studies have suggested that domains 2 and 3 interact with domain 1 to affect the equilibrium binding constant for the EF-Tu{GDP complex. When domain 1 of E. coli EF-Tu is expressed alone, the dissociation constant of the domain 1{GDP complex is three orders of magnitude higher than that obtained with intact EF-Tu ~Parmeggiani et al., 1987!. Furthermore, E. coli EF-Tu lacking either domain 2 or domain 3 alone has a decreased affinity for GDP and GTP, similar to that seen with domain 1 alone ~Cetin et al., 1998!. A chimeric form of EF-Tu in which domain 2 of E. coli EF-Tu has been replaced by the corresponding domain of EF-Tumt binds to GDP ;25% as well as that observed with E. coli EF-Tu ~Bullard et al., 1999!. Replacing domain 3 of E. coli EF-Tu with domain 3 of EF-Tumt results in a reduction of GDP binding to ;10% of that seen with the wild-type bacterial factor ~Bullard et al., 1999!. These observations suggest that domain 2 and particularly domain 3 affect guanine nucleotide binding, probably by influencing the precise conformation of domain 1.
a 50% slurry of Ni-NTA resin equilibrated in buffer I and incubated for 1 h while rocking at 4 8C. The sample was passed through a Qiagen polypropylene fritted column that retained the Ni-NTA resin. The resin was washed with buffer III ~100 mL! and EF-Tumt was eluted using buffer IV ~3 mL!. After isolation from the NiNTA column, EF-Tumt was dialyzed against buffer V ~1 L! containing 10 mM GDP for 3 h. This step was important to remove imidazole that interfered with the measurement of the absorbance spectrum of GDP. Protein concentrations were determined by using the micro-Bradford method ~Bio-Rad, Hercules, California! according to the manufacturer’s directions. Approximately 10 mg of EF-Tumt were obtained from 2 L of cell culture. All EF-Tumt preparations were greater than 95% pure as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis ~SDS-PAGE!. These preparations were free of nucleotidase activity ~data not shown!.
Materials and methods
Preparation of EF-Tumt{mantGDP
Materials Guanosine-59-triphosphate ~GTP! was purchased from Sigma Chemical Co. ~St. Louis, Missouri!. Guanosine-59-diphosphate ~GDP! was from ICN Biomedicals Inc. ~Costa Mesa, California!. @ 3H#GDP was obtained from Dupont-New England Nuclear ~Boston, Massachusetts!. 29-~or 39-!-O-~N-methylanthraniloyl! guanosine 59diphosphate ~mantGDP! was obtained from Molecular Probes ~Eugene, Oregon!. Ni-NTA fritted columns were purchased from Qiagen ~Hilden, Germany!. Nitrocellulose membrane filters HAWG ~0.45 mm pore size! were from Millipore Corporation ~Bedford, Massachusetts!. Spectrum dialysis bags were from Fisher ~Fair Lawn, New Jersey!. Buffers Buffer I contained 50 mM Tris-HCl, pH 7.6, 60 mM KCl, 7 mM MgCl2 . Buffer II was 50 mM Tris-HCl, pH 7.6, 60 mM KCl, 7 mM MgCl2 , 7 mM b-mercaptoethanol ~BME!, 0.1 mM phenylmethylsulfonylchloride ~PMSF!, and 10% glycerol. Buffer III consisted of 50 mM Tris-HCl, pH 7.6, 1 M KCl, 7 mM BME, 10 mM imidazole, and 10% glycerol. Buffer IV included 50 mM Tris-HCl, pH 7.6, 40 mM KCl, 0.15 M imidazole, 7 mM BME, and 10% glycerol. Buffer V was composed of 20 mM Hepes-KOH, pH 7.0, 40 mM KCl, 1 mM MgCl2 , 0.1 mM @ethylenedinitrilo#-tetraacetic acid ~EDTA!, and 10% glycerol. Buffer VI contained 25 mM Hepes-KOH, pH 7.6, 10% glycerol, 0.1 mM EDTA, and 50 mM NH4Cl. Buffer VII had 25 mM Hepes-KOH, pH 7.6, 10% glycerol, 0.1 mM EDTA, and 900 mM NH4Cl. Expression and purification of EF-Tumt The His-tagged form of mature bovine EF-Tumt was expressed in E. coli and purified as described previously ~Zhang et al., 1996!. The extract obtained from 6 g of cells was mixed with 0.6 mL of
Fluorescence spectroscopy All fluorescence intensity measurements were carried out on a spectrofluorimeter ~SLM 8000C! with the excitation wavelength at 350 nm. Unless otherwise noted, the emission wavelength was 446 nm and the temperature was maintained at 4 8C.
The fluorescent derivative of GDP, mantGDP, was used to replace the GDP in normal EF-Tumt preparations. Initial steps in the purification of the mantGDP-bound form of EF-Tumt were carried out as described ~Woriax et al., 1995! The EF-Tumt bound to the Ni-NTA resin was washed with 90 mL of buffer II. The EF-Tumt still retained on the resin was incubated twice with 10 mL of buffer II containing 5 mM inosine diphosphate ~IDP! for 1 h each time at 4 8C. This buffer was removed and the resin was further washed with an additional 100 mL of buffer II. The resin was incubated twice with 1 mL of buffer II containing mantGDP ~20 or 100 mM! for 1 h at 4 8C. The resin was washed with 100 mL of buffer III and the EF-Tumt containing mantGDP was eluted with 3 mL of buffer IV. Determination of the concentrations of GDP in EF-Tumt preparations The concentration of GDP in each EF-Tumt preparation was determined by absorbance spectroscopy. EF-Tumt preparations ~250 mL containing at least 50 mM EF-Tumt ! and controls ~buffer V, 250 mL! were denatured at 80 8C for 10 min. The denatured samples were centrifuged in an Eppendorf centrifuge for 30 min at 4 8C to remove the precipitated protein. The absorbance spectra of the supernatants were then measured using the buffer V control as the background. The concentrations of GDP were measured by the absorbance at 256 nm using an extinction coefficient of 13,804 M 21 cm 21 ~Grasselli & Ritchey, 1975!. Determination of the concentrations of mantGDP in EF-Tumt{mantGDP preparations Samples ~;100 mM EF-Tumt ! from EF-Tumt{mantGDP preparations were denatured at 80 8C for 10 min, centrifuged in an Eppendorf centrifuge for 30 min at 4 8C to remove precipitated protein, and diluted more than 10-fold with water. To determine the mant-
1799
Mitochondrial elongation factor Tu GDP concentration, the fluorescence intensity was compared to a standard curve developed using known concentrations of mantGDP ~from 0.125 to 4 mM! under the same conditions. Heat ~80 8C for 10 min! does not affect the emission spectrum of mantGDP ~data not shown!. Determination of the equilibrium dissociation constant of the EF-Tumt{GDP complex using equilibrium dialysis Equilibrium dialysis was carried out at 4 8C in 15 mL tubes containing 5 mm diameter dialysis bags. Dialysis was carried out at 4 8C due to the instability of EF-Tumt at higher temperatures. The dialysis bags held 150 mL samples with different concentrations of EF-Tumt varying from 2.5 to 30 mM. Dialysis was carried out against 14 mL of buffer V containing 13.8 nmol GDP ~0.986 mM! and 2 mL of @ 3 H#GDP ~7,200 cpm0pmol, specific activity 3,000 Ci0mmol, 0.014 mM!. The total radioactivity in the buffer outside the dialysis bag was measured by counting 20 mL of the starting dialysis buffer in a scintillation counter before dialysis. The amount of GDP present in each EF-Tumt sample was determined as described above. The specific activity ~as cpm0pmol! of the @ 3 H#GDP was corrected for this value. Overnight dialysis was used to ensure that equilibrium had been reached. Following dialysis, duplicate aliquots ~20 mL! from both outside and inside the dialysis bag were removed. The amount of @ 3 H#GDP was determined by counting the samples in 10 mL Scinti Safe cocktail. Any precipitated protein formed during dialysis was removed by centrifugation in an Eppendorf centrifuge. The concentration of the remaining soluble EF-Tumt was measured by the Bio-Rad method and this information was used to correct for any EF-Tumt inactivated during dialysis. The total soluble protein remaining after dialysis was considered to be the total concentration of EF-Tumt . To estimate the percentage of EF-Tumt capable of binding GDP, high concentrations of GDP ~10 to 60 mM! outside the dialysis bag were used to equilibrate the sample of EF-Tumt ~10 mM! and treated as described above. Association and dissociation kinetics of EF-Tumt with mantGDP The time course for the association of mantGDP with EF-Tumt was monitored by fluorescence spectroscopy for 15 min by adding EF-Tumt ~1, 2, or 3 mM! containing a known concentration of GDP to 1 mM mantGDP in buffer V. The time course for the dissociation of EF-Tumt{mantGDP was monitored ~for 40 min! by adding large amounts of GDP ~0.5 or 1 mM! to samples containing 0.4 mM mantGDP and 1.0 mM EF-Tumt . The fluorescence was constant in the absence of EF-Tumt ~for the association rate curve! and in the absence of GDP ~for the dissociation rate curve!. Determination of the equilibrium dissociation constant of the EF-Tumt{mantGDP complex by equilibrium dialysis and fluorescence spectroscopy Different concentrations of an EF-Tumt{mantGDP preparation ~about 7 mM EF-Tumt and 2.5 mM mantGDP or 3.5 mM EF-Tumt and 1.3 mM mantGDP in 150 mL! were dialyzed against various concentrations of mantGDP ~1, 2, and 5 mM in 14 mL buffer V! overnight at 4 8C. Protein concentrations inside the bag were measured by the Bio-Rad assay after dialysis. Samples ~30 mL! from outside and inside the bags were heated at 90 8C for 10 min and precipitated
protein was removed by Eppendorf centrifugation. The amount of mantGDP in each sample was then measured by fluorescence spectroscopy as described above following a 10-fold dilution with water. The dissociation constant for the EF-Tumt{mantGDP complex was confirmed by monitoring the fluorescence intensities of samples containing EF-Tumt{mantGDP ~1 mM EF-Tumt and 0.36 mM mantGDP! and different concentrations of GDP ~0.2 to 100 mM! in buffer V. Intensities were monitored after the samples reached equilibrium ~;40 min!. Free GDP did not interfere with the fluorescence signal from mantGDP ~data not shown!. Fluorescence measurements were carried out in the linear response range, and there were no problems with the inner filter effect ~data not shown!. Determination of the equilibrium dissociation constant of EF-Tumt{GTP by competition of GTP with mantGDP Competition between GTP and mantGDP for binding to EF-Tumt was used to measure the equilibrium dissociation constant of the EF-Tumt{GTP complex ~KGTP !. In these measurements, samples containing 1 mM EF-Tumt and 0.18 mM mantGDP in buffer V were titrated with 2.5 mM to 100 mM GTP. The emission of mantGDP was measured after equilibrium had been reached ~40 min!. Experimental uncertainties Experimental uncertainties in KGDP ~10%!, KmGDP ~25%!, km1 ~33%!, km2 ~3%!, and B ~8%! are reported as standard errors in the means obtained from repetitive trials. The standard error in the mean for KGTP was 17%. However, determination of KGTP in each individual trial depended on the values of KGDP and KmGDP ~see text!. By using standard methods of error propagation in these calculations, it was determined that the errors in KGDP and KmGDP resulted in ;50% uncertainty in the calculated value of KGTP . Therefore, this larger source of experimental error in the determination of KGTP is reported as its uncertainty. Acknowledgments This work was supported by funds from the National Institutes of Health ~Grant GM32734! to LLS and from the National Science Foundation ~Grant MCB9728116! to NLT.
References Abel K, Yoder MD, Hilgenfeld R, Jurnak F. 1996. An alpha to beta conformational switch in EF-Tu. Structure 4:1153–1159. Andersen GR, Thirup S, Spremulli LL, Nyborg J. 2000. High resolution crystal structure of bovine mitochondrial EF-Tu in complex with GDP. J Mol Biol 297:421– 436. Bullard JM, Cai Y-C, Zhang YL, Spremulli LL. 1999. Effects of domain exchanges between Escherichia coli and mammalian mitochondrial EF-Tu on interactions with guanine nucleotides, aminoacyl-tRNA and ribosomes. Biochim Biophys Acta 1446:102–114. Cantor CR, Schimmel PR. 1980. Biophysical chemistry: Techniques for the study of biological structure and function. San Francisco: W.H. Freeman and Company. Capellos C, Bielski B. 1980. Kinetic systems: Mathematical description of chemical kinetics in solution. Huntington, NY: Robert E. Krieger Publishing Co. Carvalho MDD, Carvalho JF, Merrick WC. 1984. Biological characterization of various forms of elongation factor 1 from rabbit reticulocytes. Arch Biochem Biophys 234:603– 611. Cetin R, Anborgh PH, Cool RH, Parmeggiani A. 1998. Functional role of the noncatalytic domains of elongation factor Tu in the interactions with ligands. Biochemistry 37:486– 495.
1800 Crechet J-B, Parmeggiani A. 1986. Characterization of the elongation factors from calf brain. 2. Functional properties of EF-1a, the action of physiological ligands and kirromycin. Eur J Biochem 161:647– 653. Divita G, Goody RS, Gautheron DC, Di Pietro A. 1993. Structural mapping of catalytic site with respect to alpha-subunit and noncatalytic site in yeast mitochondrial F1-ATPase using fluorescence resonance energy transfer. J Biol Chem 268:13178–13186. Edmonds BT, Bell A, Wyckoff J, Condeelis J, Leyh TS. 1998. The effect of F-actin on the binding and hydrolysis of guanine nucleotide by Dictyostelium elongation factor 1A. J Biol Chem 273:10288–10295. Giovane A, Balestrieri C, Balestrieri ML, Servillo L. 1995. Interaction studies between elongation factor Tu and anthraniloyl-fluorescent analogues of guanyl nucleotides. Eur J Biochem 227:428– 432. Grasselli JG, Ritchey WM. 1975. Spectral data and physical constants for organic compounds. In: CRC Atlas of spectral data and physical constants for organic compounds. CRC Press. pp 387–390. Hazlett TL, Moore KJM, Lowe PN, Jameson DM, Eccleston JF. 1993. Solution dynamics of p21ras proteins bound with fluorescent nucleotides: A timeresolved fluorescence study. Biochemistry 32:13575–13583. John J, Rensland H, Schlichting I, Vetter I, Borasio GD, Goody RS, Wittinghofer A. 1993. Kinetic and structural analysis of the Mg 21 -binding site of the guanine nucleotide-binding protein p21H-ras. J Biol Chem 268:923–929. John J, Sohmen R, Feuerstein J, Linke R, Wittinghofer A, Goody RS. 1990. Kinetics of interaction of nucleotides with nucleotide-free H-ras p21. Biochemistry 29:6058– 6065. Kinzy T, Merrick WC. 1991. Characterization of a limited trypsin digestion form of eukaryotic elongation factor 1a. J Biol Chem 266:4099– 4105. Kjeldgaard M, Nissen P, Thirup S, Nyborg J. 1993. The crystal structure of elongation factor EF-Tu from Thermus aquaticus in the GTP conformation. Structure 1:35–50. Lakowicz JR. 1999. Principles of fluorescence spectroscopy. Dordrecht, The Netherlands: Kluwer Academic0Plenum Press. Laurberg M, Mansilla F, Clark BFC, Knudsen CR. 1998. Investigation of functional aspects of the N-terminal region of elongation factor Tu from Escherichia coli using a protein engineering approach. J Biol Chem 273:4387– 4391. Leonard DA, Evans T, Hart M, Cerione RA, Manor D. 1994. Investigation of the GTP-binding0GTPase cycle of Cdc42Hs using fluorescence spectroscopy. Biochemistry 33:12323–12328. Louie A, Jurnak F. 1985. Kinetic studies of Escherichia coli elongation factor Tu-guanosine 59-triphosphate-aminoacyl-tRNA complexes. Biochemistry 24:6433– 6439. Miller D, Weissbach H. 1970. Studies on the purification and properties of factor Tu from E. coli. Arch Biochem Biophys 141:26–37. Miller D, Weissbach H. 1977. Factors involved in the transfer of aminoacyltRNA to the ribosome. In: Weissbach H, Pestka S, eds. Molecular mechanisms of protein biosynthesis. New York: Academic Press. pp 323–373. Moore KJM, Webb MR, Eccleston JF. 1993. Mechanism of GTP hydrolysis by p21N-ras catalyzed by GAP: Studies with a fluorescent GTP analog. Biochemistry 32:7451–7459. Nagata S, Iwasaki K, Kaziro Y. 1977. Purification and properties of polypeptide chain elongation factor EF-1a from pig liver. J Biochem 82:1633–1646. Nixon AE, Brune M, Lowe PN, Webb MR. 1995. Kinetics of inorganic phosphate release during the interaction of p21ras with the GTPase-activating proteins, p120-GAP and neurofibromin. Biochemistry 34:15592–15598. Nomanbhoy TK, Cerione RA. 1996. Characterization of the interaction between RhoGDI and Cdc42Hs using fluorescence spectroscopy. J Biol Chem 271:10004–10009. Parmeggiani A, Swart G, Mortensen K, Jensen M, Clark B, Dente L, Cortese R. 1987. Properties of a genetically engineered G domain of elongation factor Tu. Proc Natl Acad Sci USA 84:3141–3145.
Y.-C. Cai et al. Polekhina G, Thirup S, Kjeldgaard M, Nissen P, Lippmann C, Nyborg J. 1996. Helix unwinding in the effector region of elongation factor EF-Tu-GDP. Structure 4:1141–1151. Richter MF, Schwemmle M, Herrmann C, Wittinghofer A, Staeheli P. 1995. Interferon-induced MxA protein. GTP binding and GTP hydrolysis properties. J Biol Chem 270:13512–13517. Roobol K, Moller W. 1978. The role of guanine nucleotides in the interaction between aminoacyl-tRNA and elongation factor 1 of Artemia salina. Eur J Biochem 90:471– 477. Rosenthal LP, Bodley JW. 1987. Purification and characterization of Saccharomyces cerevisiae mitochondrial elongation factor Tu. J Biol Chem 262:10955– 10959. Saha SK, Chakraburtty K. 1986. Protein synthesis in yeast. Isolation of variant forms of elongation factor 1 from the yeast Saccharomyces cerevisiae. J Biol Chem 261:12599–12603. Sayle RA, Milner-White EJ. 1995. RASMOL: Biomolecular graphics for all. Trends Biochem Sci 20:374–376. Schwartzbach CJ, Spremulli LL. 1989. Bovine mitochondrial protein synthesis elongation factors: Identification and initial characterization of an elongation factor Tu-elongation factor Ts complex. J Biol Chem 264:19125–19131. Schwartzbach CJ, Spremulli LL. 1991. Interaction of animal mitochondrial EF-Tu:EF-Ts with aminoacyl-tRNA, guanine nucleotides and ribosomes. J Biol Chem 266:16324–16330. Slobin L, Moller W. 1976. Characterization of developmentally regulated forms of elongation factor 1 from Artemia salina. Eur J Biochem 69:367– 375. Song HW, Parsons MR, Rowsell S, Leonard G, Phillips SE. 1999. Crystal structure of intact elongation factor EF-Tu from Escherichia coli in GDP conformation at 2.05 Å resolution. J Mol Biol 285:1245–1256. Sprinzl M. 1994. Elongation factor Tu: A regulatory GTPase with an integrated effector. Trends Biochem Sci 19:245–250. Sreedharan SP, Beck CM, Spremulli LL. 1985. Euglena gracilis chloroplast elongation factor Tu. Purification and initial characterization. J Biol Chem 260:3126–3131. van Damme H, Amons R, Moller W. 1992. Identification of the sites in the eukaryotic elongation factor 1a involved in the binding of elongation factor 1b and aminoacyl-tRNA. Eur J Biochem 207:1025–1034. Wagner A, Simon I, Sprinzl M, Goody RS. 1995. Interaction of guanosine nucleotides and their analogs with elongation factor Tu from Thermus thermophilus. Biochemistry 34:12535–12542. Watson BS, Hazlett TL, Eccleston JF, Davis C, Jameson DM, Johnson AE. 1995. Macromolecular arrangement in the aminoacyl-tRNA:elongation factor Tu:GTP ternary complex. A fluorescence energy transfer study. Biochemistry 34:7904–7912. Woriax VL, Bullard JM, Ma L, Yokogawa T, Spremulli LL. 1997. Mechanistic studies of the translational elongation cycle in mammalian mitochondria. Biochim Biophys Acta 1352:91–101. Woriax VL, Burkhart W, Spremulli LL. 1995. Cloning, sequence analysis and expression of mammalian mitochondrial protein synthesis elongation factor Tu. Biochim Biophys Acta 1264:347–356. Woriax VL, Spremulli GH, Spremulli LL. 1996. Nucleotide and aminoacyltRNA specificity of the mammalian mitochondrial elongation factor EFTu:Ts complex. Biochim Biophys Acta 1307:66–72. Xin H, Woriax VL, Burkhart W, Spremulli LL. 1995. Cloning and expression of mitochondrial translational elongation factor Ts from bovine and human liver. J Biol Chem 270:17243–17249. Zhang YL, Li X, Spremulli LL. 1996. Role of the conserved aspartate and phenylalanine residues in prokaryotic and mitochondrial elongation factor Ts in guanine nucleotide exchange. FEBS Lett 391:330–332.
