| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 50, 41178-41183, December 16, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Laboratoire de Physiologie Cellulaire Végétale UMR5168, CEA-CNRS-INRA-Université Joseph Fourier, 38054 Grenoble, France
Received for publication, August 24, 2005 , and in revised form, October 3, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To complete the analysis of Arabidopsis AK-HSDH enzymes the Arabidopsis thaliana AK-HSDH I isoform was cloned and overexpressed in E. coli and its kinetic and regulatory properties were determined. In the course of this work, a systematic survey of the effect of other amino acids led us to identify unexpected allosteric effectors of Arabidopsis Thr-sensitive AKs. The key point of this discovery was the measurement of AK activities in the presence of low (i.e. physiological) concentrations of the substrates ATP and Asp.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Protein Expression and Purification
Transformed E. coli BL21pLys(S) cells were grown at 37 °C in 800 ml of LB medium with 100 µg/ml carbenicillin and 34 µg/ml chloramphenicol. Isopropyl 1-thio-
-D-galactopyranoside was added to a final concentration of 0.4 mM when bacterial growth was equivalent to an A600 of 0.8. Cells were further grown for 15 h at 28 °C and collected by centrifugation. The preparation of the crude extract and the first two purification steps (anion exchange chromatography and gel filtration) of AK-HSDH I were as for AK-HSDH II (5). AK-HSDH I eluted from the DEAE column at a higher KCl concentration (300 mM) than AK-HSDH II (250 mM). After elution from the gel filtration column, the protein was concentrated on a Centricon device (50 K) and loaded on a Green Sepharose column equilibrated with 25 mM Hepes, pH 7.5, 10% (v/v) glycerol, and 1 mM Thr. AK-HSDH I protein was excluded. Pure protein was quickly frozen in N2 and stored at -80 °C for several months without any loss of activity. Recombinant AK-HSDH II protein was overexpressed in E. coli and purified as previously reported (5).
|
AK Assay—AK activity was measured with a coupled assay using aspartate semialdehyde dehydrogenase (ASADH) previously purified to homogeneity in the laboratory (6). The amount of ASADH was adjusted so that the flux through the enzymatic chain was independent on the ASADH concentration (typically 50 nM AK and 0.8 µM ASADH). Activity was measured in 50 mM Hepes, pH 8.0, 150 mM KCl, 20 mM MgCl2, 200 µM NADPH in the presence of fixed or variable amounts of ATP and Asp (see figure legends). Reaction was initiated by addition of the enzyme. In this coupled assay 2 mol of NADPH are consumed per mol of Asp phosphorylated (1 mol consumed by ASADH and 1 mol by HSDH). NADPH oxidation rate at steady-state was divided by two to obtain actual AK rates.
HSDH Assay—HSDH substrate, aspartate semialdehyde, was prepared as previously described (7). Activity measurements were carried out in 50 mM Hepes, pH 8.0, 150 mM KCl, 100 µM aspartate semialdehyde, and 200 µM NADPH. Reaction was initiated with the enzyme (final concentration 20 nM).
Free Amino Acid Analysis in Purified Arabidopsis Chloroplasts
A. thaliana (ecotype Columbia) plants were grown for 3 weeks on soil in a growth chamber (22 °C, 60% air humidity, light intensity of 150 µE m-2 s-1, 16 h light/8 h dark) before rosette leaves were harvested. Chloroplasts were purified on Percoll gradients as described in Ref. 8. Free amino acids were prepared from intact chloroplasts by three consecutive extractions in aqueous ethanol solutions buffered with Hepes-KOH, pH 7.5, as described in Ref. 9. Between the extraction steps, the samples were centrifuged for 10 min at 16,000 x g and the supernatants collected. The combined ethanol/water extracts were used for amino acid derivatization using o-phthaldialdehyde. The derivatized amino acids were separated by reverse phase high pressure liquid chromatography using a Hypersil C18 column (150 x 4.6 mm inner diameter; 3 µm) at 37 °C and 0.8 ml/min using a 65-min gradient of methanol in a sodium acetate buffer. Derivatized amino acids were detected using a SFM25 fluorimeter (Kontron) by excitation at 340 nm and emission at 455 nm. Free amino acid contents were normalized to chlorophyll concentration (10). Three extractions were performed for two independent purifications of chloroplasts to ensure reproducibility.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Comparison of AK-HSDH Isoform Kinetic Properties—AK I and AK II catalytic constants were similar (v/[E]0 = 5 ± 0.5 s-1 for AK I and v/[E]0 = 6.2 ± 0.5 s-1 for AK II). Compared with AK II, AK I displayed a higher Km for aspartate and a lower Km for ATP, see TABLE ONE.
|
Inhibition of AK Activities by Thr at Low Substrate Concentrations— The inhibition of AK II by Thr was previously measured in the presence of 50 mM Asp and 20 mM ATP (7). In the present work, to describe and compare the inhibition by Thr under more physiological conditions, we used low substrate concentrations, representative of the chloroplastic environment, i.e. 1 mM Asp and 2 mM ATP (11). Fig. 1A shows that AK I, which was predicted to be a Thr-sensitive AK, was indeed inhibited by Thr. The inhibition curve was sigmoidal. The K0.5 value for Thr was about twice as high as that of AK II (91 ± 3 and 49 ± 3 µM for AK I and AK II, respectively). Also the cooperativity of the inhibition was much higher for AK I (Hill number, nH = 3.7 ± 0.5) than for AK II (nH = 1.7 ± 0.15).
It is worth noting that the K0.5 value of AK II for Thr was 10-fold lower than previously reported for this enzyme when activity was measured in the presence of saturating concentrations of substrates (7). A lower K0.5 value for Thr in the presence of low substrate concentrations was expected as the inhibition of the AK activity by Thr is competitive with respect to the substrates (7).
Inhibition of HSDH Activities by Thr—Thr inhibited HSDH I activity in a hyperbolic manner (Fig. 1B). Inhibition was partial at saturation. In the presence of 100 µM aspartate semialdehyde, a K0.5 value for Thr of 407 ± 56 µM was calculated for HSDH I. In the same experimental conditions, HSDH II displayed a K0.5 value for Thr about 1 order of magnitude higher than HSDH I (K0.5 = 8500 ± 1800 µM). At saturation of Thr HSDH I retained 14% of activity, whereas HSDH II retained 25% of activity.
Identification of New Allosteric Effectors of AK I—As the present work was the first characterization of AK I we checked the effect of various other amino acids on its activity, in the absence of Thr and in the presence of physiological concentrations of substrates (1 mM Asp, 2 mM ATP). We initially tested amino acids produced in the Asp-derived amino acid pathway (Lys, Met, Leu, Val, Ile), or connected to this pathway (Cys, the precursor of Met). Interestingly, Cys, Ile, and Val stimulated the enzyme activity about 3-fold at 2.5 mM. On the other hand, Leu inhibited AK I. These results engaged us to test other amino acids external to the pathway but of similar size (Gly, Ser, and Ala). Surprisingly, Ser and Ala (2.5 mM) also stimulated AK I activity about 3-fold. Response curves are displayed in Fig. 2A. The activation was hyperbolic or slightly sigmoidal (for Cys). Cys was the strongest activator. Activation by cysteine did not reflect a general sulfhydryl effect, because dithiothreitol did not activate AK at 1 mM. Parameters of the activation by Ala, Cys, Ile, Ser, Val, and of the inhibition by Leu are displayed in TABLE TWO. To test for synergistic effects between different activators the different couples were tested at concentrations representing one-fourth the K0.5 value calculated from the response curves in Fig. 2A. For each couple tested, activation was additive, indicating the absence of synergy between two different activators. In the same manner no synergy could be detected between the inhibitors Leu and Thr (not shown).
|
|
|
Re-examination of AK-HSDH II Regulatory Properties—We checked the effect of the allosteric effectors identified for AK I on the AK II activity, in the presence of physiological concentrations of ATP and Asp.
Interestingly, the effect on the AK II activity was even more pronounced than for AK I and qualitatively different. A 13-fold stimulation of AK II activity was observed with Ala, Cys, Ile, Ser, and Val each at 2.5 mM. Leu, an inhibitor of AK I, was stimulatory with AK II although to a lower extent than observed for the other activators (4-fold activation at 2.5 mM). The response curves for the strongest activators of AK II displayed sigmoidal shapes (Fig. 2B). The saturation curve by Cys (Fig. 2B) had the lowest K0.5 value (80 ± 5 µM). K0.5 values for Val, Ile, Ala, and Ser were similar and about 4-fold higher (350 µM) than for Cys (see TABLE TWO). The saturation curve for Leu (Fig. 2B) displayed a higher K0.5 value (635 ± 100 µM) and was hyperbolic. As observed for AK I no synergy could be detected between different activators of AK II (not shown).
At 2.5 mM, the 6 activators of AK II were without any detectable effect on HSDH II activity in the presence or the absence of Thr, with the exception of Cys (not shown). Inhibition of HSDH II by Cys was hyperbolic and partial at saturation. The enzyme retained 15% of activity at saturation with Cys and a K0.5 value of 715 ± 171 µM was calculated. Thus, as observed for HSDH I, Cys was a more potent inhibitor of HSDH II than Thr. HSDH II K0.5 value for Cys was higher than that of HSDH I. No synergy between Thr and Cys for the inhibition of HSDH II could be observed (not shown).
Activation Results from a Decrease in the Km for the Substrates—AK I and AK II kinetic parameters in the absence and presence of a saturating concentration of Ala (10 mM) were compared. No catalytic effect could be detected (not shown). However, as indicated in TABLE ONE, in the presence of 10 mM Ala, AK I and AK II displayed much lower K0.5 values for ATP than in the absence of Ala (13- and 5-fold decrease, respectively). In the presence of Ala, AK II Km for Asp was reduced 4-fold, whereas AK I displayed a Km for Asp virtually unchanged. The minor effect of Ala on AK I apparent affinity for Asp explains why AK I is less efficiently activated than AK II at physiological concentrations of Asp and ATP (see Fig. 2). Both isoforms displayed similar kinetic parameters with respect to ATP and Asp in the presence of 10 mM Ala. The activation thus results from an increase in the apparent affinity for the substrates, and thus can only be detected at low substrate concentrations.
Enzyme Response in the Presence of Physiological Concentrations of Effectors—From a physiological point of view, the existence of several allosteric activators for a single enzyme, with some amino acids produced downstream in the pathway (Leu, Val, and Ile) and others external to the pathway (Cys, Ser, and Ala) was perplexing. To examine this further in a quantitative manner, the abundance of the different amino acids was measured in the stroma of chloroplasts isolated from A. thaliana leaves. Results presented in TABLE THREE indicated that Ala and Thr were from 5- to 20-fold more abundant than the other amino acids. Cys, the strongest activator of AK I and AK II and the best inhibitor of HSDH I and II was the lowest abundant of the 7 effectors. In the absence of observed synergy between different amino acids, these results suggested that the broad specificity for the control of AK activity observed when amino acids were tested one by one (Fig. 2) might be restricted to the two highly abundant amino acids Ala and Thr. We thus designed experiments in which AK I and AK II activities were measured in the presence of the 7 allosteric effectors at their chloroplastic concentrations. From a physiological point of view it was more pertinent to consider relative changes of concentration rather than absolute changes. Thus, the concentration of each of the allosteric effectors of the AK activities was varied in turn, from zero to several times their chloroplastic concentration, in the presence of the other effectors at their chloroplastic concentration. Fig. 3, A and B, show that, when the modification of amino acid abundance was expressed in terms of fold-changes, AK I and AK II activities were efficiently inhibited by Thr and activated by Ala only. Manyfold changes in Cys, Ile, Leu, Ser, or Val concentrations had only minor quantitative effects on AK I and AK II activities (Fig. 3). Results in Fig. 3, A and B, also show that Ala and Thr are able to displace each other. Moreover, the inhibition by Thr and the activation by Ala were observed for rather similar ranges of concentrations of effectors for AK I and AK II. Indeed, half-inhibition by Thr was observed at a concentration of Thr of 104 µM and 114 µM for AK I and AK II, respectively. Also, half-maximum activation was observed for 2250 and 3400 µM Ala for AK I and AK II, respectively.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The broad specificity observed for the control of the Arabidopsis AK-HSDH activities was surprising. We first checked whether some interactions could be synergistic (as observed for the Lys-sensitive plant AK that is synergistically inhibited by Lys and S-adenosylmethionine, (13)). As we could not detect any synergy between different effectors, one predicted that the relative abundance of the effectors determines their contributions to the control of the enzyme activity. We thus quantified the concentrations of the different allosteric effectors in Arabidopsis chloroplasts. The inhibitor Thr and the activator Ala proved to be much more abundant than the other allosteric effectors suggesting that Thr and Ala were the physiological effectors of the enzymes. In vitro kinetics in the simultaneous presence of the seven allosteric effectors confirmed this prediction. These experiments also showed that AK I and AK II were modulated by similar ranges of Thr or Ala concentrations. This was a priori non-obvious at the examination of the results in Fig. 2. Indeed, one may conclude from these results that AK I and AK II strongly differ with respect to their sensitivity to Ala. Indeed, AK I displayed much higher K0.5 values for the activators Ala than AK II. Also, the activation factor was lower than for AK II. However, results in Fig. 3 show that these quantitative differences observed when amino acids were tested in isolation were strongly attenuated when activities were measured in the simultaneous presence of the inhibitor Thr and the activator Ala. As AK I displayed a lower apparent affinity than AK II for both the activator and the inhibitor, the ratio of concentration rather than the absolute concentrations of Ala and Thr determined the output.
It is interesting to observe from results in Fig. 3 that the other activators do contribute altogether to the activation of the enzyme. For example, in the presence of a physiological concentration of Thr (300 µM), the six allosteric activators together stimulate 25-fold the AK II activity. Sixty percent of this activation are because of the abundant amino acid Ala. Nevertheless, the other activators contribute to the remaining 40%. Whether this non-negligible contribution constitutes a noise in the regulatory processes that cannot be avoided or whether it has important physiological function remains to be determined.
An interesting conclusion from this work is that the cellular context rather than the enzyme structure determines the specificity of the allosteric controls of Arabidopsis AK-HSDH enzyme activities. It is worth noting that the present conclusions are valid for Arabidopsis leaf chloroplasts. The conclusions might be different in a different tissue, under different environmental or developmental conditions leading to changes in the relative abundance of the seven allosteric effectors.
AK I, AK II, and HSDH I activities are inhibited by physiological concentrations of Thr. However, one can conclude from our results that HSDH activities are not stringent control points by Thr. Indeed, in a reaction media that mimics the metabolic composition of the leaf chloroplast, half-inhibition of HSDH I occurs for a concentration of Thr 6-fold higher than that required for half-inhibition of AK I or AK II (600 versus 100 µM, compare Fig. 3, A and C). Also, molecular analyses using a GUS reporter-gene strategy suggested that the promoters of AK-HSDH I and AK-HSDH II were simultaneously active in different tissues of the plant (15) and especially in the leaf. If AK-HSDH I and AK-HSDH II enzymes are indeed simultaneously expressed in the same compartment, and HSDH II is fully active at a physiological concentration of Thr, then HSDH activities are definitely not stringent control points by Thr. Note also that, although the inhibition of HSDH activities by Cys is more efficient than by Thr in vitro, the low abundance of this metabolite in vivo excludes a sensitive control of HSDH activities by Cys.
Functional redundancy is frequently invoked to justify the existence of isoforms. The present analysis shows that the two enzymes are not exactly similar. First, one observed that AK I was about 4-fold less active than AK II when activity was measured in the presence of the seven allosteric effectors at their physiological concentration (see Fig. 3, A and B). Also, as shown in these figures, the response curve for Ala is shifted to slightly higher concentrations of Ala for AK I compared with AK II. Finally, the two isoforms strongly differ with respect to the control of the HSDH activity by Thr. Whether these differences have important physiological implications remains to be determined. Other levels of controls, which would affect the two isoforms differentially are not excluded.
From a functional point of view, the existence of activators and inhibitors of an enzyme activity presents two advantages. First, it increases the sensitivity to small changes in the activator or the inhibitor concentrations. Second, as Ala not only displaces Thr but also activates the enzyme by itself, the existence of an activator and an inhibitor increases the range of enzyme activity for which control can be exercised. Additional function, if any, of the activation of plant AKs by Ala, remains obscure. In plants, after incorporation into glutamate, nitrogen is quickly distributed to other amino acids, much of it going directly to Ala and Asp via transamination of pyruvate and oxaloacetate, respectively (14). Ala is thus directly linked to the carbon and the nitrogen metabolisms of the cell. In this respect, Ala may be a feed-forward activator of the aspartate-derived amino acid pathway. Activation of AKs by Ala might couple the supply of carbon and nitrogen to the demand for Asp-derived amino acids synthesis. This is new information that might prove useful for understanding of the aspartate-derived amino acid metabolism in vivo and the improvement of crop quality.
|
FOOTNOTES |
|---|
1 To whom correspondence should be addressed. Tel.: 33-4-38-78-23-64; Fax: 33-4-38-78-50-91; E-mail: gcurien{at}cea.fr.
2 The abbreviations used are: AK, aspartate kinase; ASADH, aspartate semialdehyde dehydrogenase; HSDH, homoserine dehydrogenase.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Mas-Droux, G. Curien, M. Robert-Genthon, M. Laurencin, J.-L. Ferrer, and R. Dumas A Novel Organization of ACT Domains in Allosteric Enzymes Revealed by the Crystal Structure of Arabidopsis Aspartate Kinase PLANT CELL, July 1, 2006; 18(7): 1681 - 1692. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
), Cys (
), Ile (
), Leu (
), Ser (
), and Val (
). Curves are the best fit obtained by non-linear regression analysis of the experimental data to hyperbolic equations except for Leu where a Hill equation was used. Parameters of the activation are summarized in 