Protein Engineering vol. 16 no. 12 pp. 1071-1079, 2003
© 2003 Oxford University Press
Structural and functional analysis of a truncated form of Saccharomyces cerevisiae ATP sulfurylase: C-terminal domain essential for oligomer formation but not for activity
1Department of Biochemistry, 2Medical Biophysics and Molecular & Medical Genetics, University of Toronto, Toronto, Ontario M5S 1A8, 3Molecular and Structural Biology, Ontario Cancer Institute/Princess Margaret Hospital, University Health Network, 610 University Avenue, Toronto, Ontario M5G 2M9, Canada and 4Department of Biochemistry, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461-1926, USA
5 To whom correspondence should be addressed at: Division of Molecular and Structural Biology, Ontario Cancer Institute/Princess Margaret Hospital, University Health Network, 610 University Avenue, Toronto, Ontario M5G 2M9, Canada. e-mail: pai{at}hera.med.utoronto.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
ATP sulfurylase catalyzes the first step in the activation of sulfate by transferring the adenylyl-moiety (AMP
) of ATP to sulfate to form adenosine 5'-phosphosulfate (APS) and pyrophosphate (PPi). Subsequently, APS kinase mediates transfer of the 
-phosphoryl group of ATP to APS to form 3'-phosphoadenosine 5'-phosphosulfate (PAPS) and ADP. The recently determined crystal structure of yeast ATP sulfurylase suggests that its C-terminal domain is structurally quite independent from the other domains, and not essential for catalytic activity. It seems, however, to dictate the oligomerization state of the protein. Here we show that truncation of this domain results in a monomeric enzyme with slightly enhanced catalytic efficiency. Structural alignment of the C-terminal domain indicated that it is extremely similar in its fold to APS kinase although not catalytically competent. While carrying out these structural and functional studies a surface groove was noted. Careful inspection and modeling revealed that the groove is sufficiently deep and wide, as well as properly positioned, to act as a substrate channel between the ATP sulfurylase and APS kinase-like domains of the enzyme.
Keywords: APS kinase/ATP sulfurylase/channeling/domain evolution/yeast
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Activated sulfate (adenosine 5'-phosphosulfate, or APS) is an ancient, high-energy metabolite whose cellular function has adapted over evolutionary time to serve a variety of metabolic roles. Dissimilatory sulfate reducing bacteria, present on the earth since prior to its oxygen atmosphere, activate sulfate to accomplish the eight-electron reduction of SO42– to sulfide that is used to drive oxidative phosphorylation. Chemolithotrophs and photolithotrophs use APS as an intermediate in the oxidation of reduced sulfur compounds, which provide electrons and energy for metabolic reactions. In assimilatory reducers, APS synthesis is essential for the synthesis of reduced sulfur metabolites (Segel, 1975
; Sanchez et al., 2001). Eukaryotes use APS, in the form of 3'-phosphoadenosine 5'-phosphosulfate (PAPS) for an entirely different purpose: sulfuryl-group transfer, a process used widely by the cell to regulate metabolism and essential for cellular processes such as blood clotting (Jevons, 1963; Morita and Jackson, 1986), growth factor/receptor interactions (Ishihara et al., 1993) and cell adhesion (Brauer et al., 1990).
The phosphoric-sulfuric acid anhydride bond present in both APS and PAPS is the chemical hallmark of activated sulfate. The energy of this bond is unusually large [
Go' (hydrolysis) –19 kcal/mol]. It is this energy that activates the otherwise nonreactive SO42– to participate in its subsequent metabolic biochemistry. In assimilatory and dissimilatory sulfate reducers, APS synthesis is catalyzed by the enzyme ATP sulfurylase (ATP:sulfate adenylyltransferase, EC 2.7.7.4) in a reaction in which the adenylyl-moiety (AMP) of ATP is transferred to sulfate, to produce APS and inorganic pyrophosphate (PPi) (Figure 1A) (Segel, 1975). APS is then phosphorylated by the enzyme APS kinase (ATP:adenylylsulfate 3'-phosphotransferase, EC 2.7.1.25) that transfers the -phosphoryl group from ATP to the 3'-hydroxyl of APS, to form PAPS and ADP (Figure 1B).
|
Despite the fact that the
–ß bond of ATP is cleaved in forming APS, the reaction is extremely unfavorable. The Keq for ATP sulfurylase catalyzed formation of APS is estimated at 1x10–8 under near-physiological conditions (Robbins and Lipmann, 1958). This energetic dilemma is resolved differently in different organisms. ATP sulfurylase isolated from Escherichia coli is composed of two subunits, called CysD and CysN, which catalyze APS synthesis and GTP hydrolysis, respectively (Leyh and Suo, 1992). The GTP hydrolysis reaction is kinetically and energetically linked to the synthesis of APS, and thus provides a –7.5 kcal/mol driving force for APS synthesis (Liu et al., 1994). The primary sequence of the E.coli ATP sulfurylase shares little, if any, similarity with those from eukaryotes—these enzymes are homomeric, and do not contain the GTPase subunit. In several animals, including the marine worm (Rosenthal and Leustek, 1995), fruit fly (Jullien et al., 1997), mouse (Li et al., 1995) and human (Venkatachalam et al., 1998), the genes for ATP sulfurylase and APS kinase are fused into a single bifunctional polypeptide, PAPS synthetase, which is known to channel APS between its active sites (Lyle et al., 1994). APS channeling provides an alternate means of overcoming the unfavorable energetics of APS synthesis and prevents the potent APS-inhibition of both ATP sulfurylase (Seubert et al., 1985) and APS kinase (Satishchandran and Markham, 1989; Renosto et al., 1991).
Recently, two crystal-structure analyses of yeast ATP sulfurylase have been performed independently by a group at the Max-Planck Institute in Munich (Ullrich and Huber, 2001; Ullrich et al., 2001) and in our Toronto laboratory (PDB code 1J70). Crystal structures are also known for the enzymes from Penicillium chrysogenum (MacRae et al., 2001, 2002) and the bacterial symbiont of the hydrothermal vent tubeworm Riftia pachyptila (Beynon et al., 2001). The yeast enzyme adopts a homohexameric arrangement. Each monomer separates into four domains: an N-terminal domain that loosely resembles pyruvate kinase, a catalytic domain that is structurally homologous to nucleotidyltransferases that connects via a linker domain to the C-terminal domain. This last domain is spatially quite separated from the catalytic domain of the ATP sulfurylase monomer and whereas sequence alignment provided little information regarding function, structural alignment revealed that this domain’s fold is extremely similar to that of APS kinase. This suggests that the yeast ATP sulfurylase is evolutionarily related to a bifunctional enzyme.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Materials
The pET expression systems were obtained from Novagen (Madison, WI). The pUBS520 and mgk rare-codon expression plasmids were kind gifts from Professor Ralf Mattes (Universität Stuttgart, Germany) and Dr Dinesh Christendat (University of Toronto), respectively. The pTYB1 plasmid and chitin affinity resin were from New England Biolabs (Beverly, MA). The BL21(DE3) cells were from Stratagene (La Jolla, CA). Complete EDTA-free protease inhibitor cocktail and DNase I were purchased from Roche (Mannheim, Germany). Superflow nickel agarose resin was from Qiagen GmbH (Hilden, Germany). The MWCO centrifugal filters were from Millipore (Bedford, MA) and the Tube-O-Dializer from Chemicon International Inc. (Temecula, CA). MonoQ HR 5/5 and Superdex 200 columns were from Amersham Biosciences Corp. (Piscataway, NJ). Crystal Screens I and II were from Hampton Research (Laguna Niguel, CA). PEG 6000 was from Fluka (St Louis, MO). Lactate dehydrogenase (rabbit muscle), pyruvate kinase (rabbit muscle), pyrophosphatase (yeast) and ß-NADH were from Roche Diagnostics Corp. ATP and phosphoenol pyruvate (PEP) were from Sigma-Aldrich Co. (St Louis, MO). Sodium sulfate was purchased from Fisher Scientific International Inc. (Pittsburgh, PA). APS kinase was purified according to previously published protocols (Wei et al., 2002).
Cloning and expression of yeast ATP sulfurylase
The ATP sulfurylase coding region (MET3, accession no. CAA60932) was cloned from a yeast cDNA library using the N- and C-terminal primers, 5'-AGTATAATTCATATGCCT GCTCCTCAC-3' and 5'-GAATGGATCCTTAAAATA CAAAAAAGCCATTGTC-3', respectively, and inserted into the pET-28-c(+) expression plasmid and sequenced. ATP sulfurylase was excised from this plasmid with the N-terminal His-tag coding sequence and ligated into pET-11d. The pET-11d and pUBS520 ArgU anti-codon expression plasmids were cotransformed into E.coli BL21(DE3) cells.
Expression and purification of yeast ATP sulfurylase
Transformed E.coli [BL21(DE3)] were grown at 37°C in 2YT media. Expression was induced at OD600 = 0.6 by the addition of isopropyl ß-D-1-thiogalactopyranoside (IPTG) to 0.4 mM. The cells were harvested 2.5–3 h later and stored at –70°C. Pellets were resuspended in 30 ml/l E.coli of binding buffer (5 mM imidazole, 50 mM HEPES, 0.5 M NaCl, 5% glycerol, pH 7.5) containing one tablet of Complete EDTA-free protease inhibitor cocktail and 300–1000 U DNase I. The cells were lysed using a French Press, the debris was spun down and the 0.22 µm filtered supernatant was loaded onto an equilibrated 5 ml Ni–NTA Superflow nickel agarose column. The column was washed with wash buffer (30 mM imidazole, 50 mM HEPES, 0.5 M NaCl, 5% glycerol, pH 7.5) until the OD280 reached baseline and the bound protein was eluted with elution buffer (250 mM imidazole, 50 mM HEPES, 0.5 M NaCl, 5% glycerol, pH 7.5). The fractions containing the protein were pooled and concentrated using a 30K MWCO centrifugal filter. The concentrated sample was loaded onto a Superdex 200 size-exclusion column. The protein yielded two peaks. Most of the protein was present in a peak corresponding to a hexamer while the smaller peak corresponded to a dodecamer. The fractions containing the hexamer were pooled and concentrated. For crystal screens the protein was dialyzed against 5 mM HEPES, 1 mM NaN3, 0.5 mM EDTA, pH 7.5. The protein concentration was adjusted to 10–15 mg/ml.
Se-methionine derivative
Single colonies of BL21(DE3) cells containing the ATP sulfurylase expression system were grown to saturation at 37°C in 5 ml of 2YT media. They were gently pelleted and suspended twice in M9 minimal medium before inoculation into 1 l of fresh M9 media. The bacteria were grown at 37°C for 12–15 h at which point, 40 ml of M9 media containing 200 mg of each of lysine, threonine, phenylalanine, leucine, isoleucine, valine and 100 mg of selenomethionine were added. After 15 min, expression was induced by the addition of IPTG to a final concentration of 40 µM. The bacteria were harvested 8 h later.
Cloning, expression and purification of the C-terminal truncate of ATP sulfurylase
The DNA encoding the truncate (residues 1–393) was cloned from pET-11d containing the complete ATP sulfurylase gene using the N-terminal and C-terminal primers, 5'-AGT ATAATTCATATGCCTGCTCCTCAC-3' and 5'-AAAAGC AGGAAGAGCCTTTTGGTCTTGGTGGGTTGG-3', respectively. The product was inserted into the pTYB1 plasmid and transformed into E.coli containing the mgk rare codon plasmid. Transformed cells were grown at 37°C to an OD600 of 0.6, induced with 1 mM IPTG, grown for 4 h at 25°C, harvested and then frozen at –70°C. One liter of frozen cells was resuspended in 30 ml of lysis buffer [20 mM HEPES (pH 8.0) and 500 mM NaCl] containing one tablet of Complete EDTA-free protease inhibitor cocktail and 300–1000 U DNase I. After passing a French-press, debris was spun down and the 0.22 µm filtered supernatant was loaded onto an equilibrated 10 ml chitin affinity column. The column was washed at 4°C with lysis buffer until the OD280 reached baseline and then with 3 vols of lysis buffer containing 50 mM DTT. The column was incubated for 18 h at room temperature. The protein was eluted with lysis buffer. Fractions containing protein were pooled and concentrated using 30K MWCO centrifugal filters and were loaded onto a Superdex 200 gel filtration column. The truncate eluted in a single peak. Using a 30K MWCO centrifugal filter the protein was concentrated to 30–40 mg/ml for crystal screens.
Protein crystallization and diffraction data collection
Using the vapor-diffusion hanging-drop technique, initial crystallization conditions were tested with Crystal Screens I and II and promising conditions were refined. Wedge-shaped hexagonal rod ATP sulfurylase crystals formed within a few days in 4 µl droplets (3:1 protein:precipitant ratio) of 4% PEG 6000, 1.8 M LiCl, 50 mM HEPES (pH 7.0) and 20% glycerol. The crystals had dimensions of 0.4x0.1x0.05 mm3 and belonged to space group P321 with unit cell axes a = b = 230.85 Å, c = 69.71 Å. The Se-derivative crystallized under very similar conditions and formed crystals of the same space group and almost identical unit cell axes. Crystals were picked up in cryo-loops and mounted in a stream of nitrogen cooled to 100 K. A three-wavelength data set from a single Se-methionine labeled protein crystal was collected at the Advanced Photon Source beamline 14BMD (BioCARS) with a Quantum Q1 CCD detector with oscillations of 0.5°. The structure was solved using the MAD technique with selenium as the anomalous scatterer and refined to 2.5 Å. A native protein data set of 2.3 Å resolution was collected at the National Synchrotron Light Source (NSLS) beamline X8-C using a Quantum Q4 detector and an oscillation range of 0.2° (Table I).
|
Crystals of the ATP sulfurylase truncate formed in 4 µl droplets (1:1 protein:precipitant ratio) of 1.9 M (NH4)SO4, 0.1 M MES (pH 6.5), 0.01 M cobaltous chloride and 8% glycerol within a few days. They had dimensions of
0.3x0.3x0.3 mm3 and belonged to space group P212121 with unit cell axes a = 50.2 Å, b = 59.1 Å and c = 131.2 Å. Crystals were dipped into 1.9 M (NH4)SO4, 0.1 M MES (pH 6.5), 0.01 M cobaltous chloride and 16% glycerol before being mounted in a stream of nitrogen cooled to 100 K. A data set for the ATP sulfurylase truncate was collected at the Advanced Photon Source beamline 14BMC (BioCARS) with a Quantum Q4 CCD detector using an oscillation range of 1.0° (Table I). All data were processed with DENZO and reduced with SCALEPACK (Otwinowski and Minor, 1997). Phase determination, model building and refinement
For the crystals of the full-length ATP sulfurylase, the selenium sites were determined with the help of the program package SOLVE (Version 1.17, Terwilliger and Berendzen, 1999) and refined in SHARP (de la Fortelle and Bricogne, 1997). The resulting electron density map was improved by solvent-flattening and non-crystallographic symmetry averaging/masking techniques, in combination with phase extension using CCP4 (Version 2.0, Cowtan and Main, 1996). Model visualization and rebuilding were done with O (Version 6.2.1, Jones et al., 1991). Refinement of the model and picking of water molecules was accomplished with CNS (Brünger et al., 1998). Molecules A, B and C were individually checked with PROCHECK (Laskowski et al., 1993) and the secondary structures assigned with PROMOTIF (Hutchinson and Thorton, 1996). The structure of the truncate was determined at 1.9 Å using Patterson search techniques as implemented in CNS (Version 1.0) (Brünger et al., 1998) and then refined at 1.4 Å. All other procedures were as described for the full-length enzyme. Refinement statistics are found in Table II. The atomic coordinates for the full-length and truncated forms of yeast ATP:sulfate adenylyltransferase have been deposited in the Protein Data Bank under accession numbers 1J70 and 1R6X, respectively.
|
Kinetic analysis
APS formation was monitored at 339 nm by coupling the production of APS to the oxidation of NADH with the coupling enzymes APS kinase (0.5 U/ml), pyruvate kinase (15 U/ml) and lactate dehydrogenase (5.0 U/ml). Pyrophosphatase (0.5 U/ml) was added to remove the other primary product, pyrophosphate, and thus keep the reaction in the initial-velocity stage. The reaction conditions were as follows: PEP (1 mM), NADH (0.20 mM), and MgCl2 ([ATP] + 1.0 mM), 50 mM HEPES/K+, pH 8.0. The coupling enzymes were chosen so that the coupling reactions achieved 98% of their steady-state levels in <15 s. The concentration of native enzyme was 0.6 µM, and the concentration of the truncate was 0.4 µM. Velocities were measured at the 16 concentrations defined by a 4x4 matrix of substrate and activator concentrations. For the assays of the wild-type enzyme, the SO4 concentrations were 5.0, 0.56, 0.29 and 0.20 mM, and the ATP concentrations were 1.0, 0.11, 0.059 and 0.040 mM, respectively. In the assays of the truncated enzyme, the concentrations of both sulfate and ATP were 1.0, 0.11, 0.059 and 0.040 mM. The data were fit with the non-linear least-squares algorithm developed by Cleland using the model for a sequential binding mechanism (SEQUEN) (Cleland, 1979).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Structural and functional analysis of yeast ATP sulfurylase
The results of our crystallographic analysis of native yeast ATP sulfurylase are completely consistent with those reported by Ullrich et al. (2001). They revealed that the enzyme is a homohexamer and each monomer can be separated into four domains (Figure 2). Domain I (residues 1–167) displays weak but significant homology to the ß domain of pyruvate kinase (Larsen et al., 1994). Domain II (residues 168–327) forms the catalytic site and adopts a Rossmann-fold, typical for nucleotide-binding enzymes (Rossmann et al., 1974). A DALI search indicates that it is highly homologous to the nucleotidylyltransferase family of enzymes (Izard and Geerlof, 1999; Saridakis et al., 2001). Domain III (residues 328–393) links domain II to domain IV. Surprisingly, despite the absence of kinase activity, domain IV is structurally very similar to APS kinase, indicating a potential evolutionary relationship with a distinct bifunctional PAPS synthetase.
|
The three monomers of the ring-shaped trimer associate by ionic and hydrophobic interactions between the N-terminal residues 168–330 of one monomer and the C-terminal (residues 392–511) of its neighbor. Two of these rings attach upside down, rotated 60° relative to each other to form the hexamer. The two rings associate by hydrogen bonding between the N-terminal residues 1–167 and the C-terminus, while residues across the interface contribute hydrophobic interactions. The hexamer has a central pore that is surrounded by a densely packed protein ring. Six peripherally directed spikes emerge from this ring that reach out 150 Å from the center (Figure 2C). From this description it is obvious that the C-terminal domain is intimately involved in the association of three monomers into a trimer ring and also in the association of two of those rings into the mature hexamer.
Based on these structural observations, we hypothesized that removing the C-terminal domain might change the oligomerization state of the enzyme but keep the enzyme’s catalytic function intact. To test this hypothesis, a C-terminal truncate was constructed by deletion of residues 394–511. After the protein was expressed and purified, N-terminal sequencing showed that the N-terminal methionine residue had been cleaved. Size exclusion chromatography revealed that the ATP sulfurylase truncate adopted a monomeric form, thus establishing that the truncate could adopt a stable three-dimensional fold. Its molecular weight, as assessed by size exclusion chromatography, SDS–PAGE and mass spectrometry was, within error of measurement, equal to the predicted molecular weight of the truncate, 44.5 kDa. This result proved that the C-terminal domain is essential for oligomerization, consistent with the intricate interactions between the monomers mitigated by domain IV in the hexameric, full-length form of the protein.
To explore whether the change in oligomeric state was also reflected in an altered fold, the crystal structure of the truncate was determined. The truncate formed highly ordered crystals, distinctly superior to those of the full-length protein. Clear electron density was found for all residues of the truncate, except for amino acids 388–393 located at the C-terminus and for residues 345–351 located in a surface loop. Both of these regions are apparently mobile. The crystallographic refinement converged at an Rfactor of 19.7% and an Rfree of 21.7% at a resolution of 1.4 Å. A cobalt atom (B-factor = 24.3 Å2) was assigned to a high electron density peak coordinated between His 235 (distances = 2.19 Å, 2.98 Å), His 236 (distances = 2.15 Å, 3.03 Å, 3.07 Å), Asp 168 (distance = 2.74 Å) and two water molecules. Three other high-density peaks could be well fitted by sulfate ions. The first of these peaks (B-factor = 15.9 Å2) lies in the active site and represents a sulfate group tightly coordinated between Gln 195 (distance = 2.75 Å), Arg 197 (distances = 2.57 Å, 2.82 Å, 3.17 Å) and Ala 293 (distance = 2.84 Å) and by four water molecules, probable remnants of the original hydration shell. The second peak (B-factor = 31.4 Å2) is coordinated between His 166 (distance = 2.74 Å), Tyr 167 (distance = 2.97 Å), Asp 168 (distances = 3.09, 2.78) and Arg 173 (distance = 3.11 Å) and four water molecules. The third peak (B-factor = 27.9 Å2) is on the surface of the protein and is coordinated between Arg 205 (distances = 2.82, 2.96 Å), Arg 213 (distances = 2.92, 3.12 Å) and a single water molecule. Residues 2–387 of the truncate and the corresponding amino acids of one monomer of the hexameric form superimpose with a root mean square deviation (r.m.s.d.) of 1.05 Å for all common atoms, indicating that the truncate adopts an almost identical fold and even the side chain orientations are very similar. The superimposition of these two structures illustrates the high degree of structural similarity (Figure 3).
|
The catalytic consequences of truncating ATP sulfurylase were explored in a comparative study of the initial-rate behavior of native enzyme and truncate. Rather than erode turnover at the active site of ATP sulfurylase, removing the domain actually slightly increases the catalytic efficiency of the enzyme, kcat, is essentially unaffected, while the steady-state affinity of the active site for both of its substrates is enhanced [Km (SO4) and Km (ATP) decrease 3.5- and 1.5-fold, respectively] (Table III). This is consistent with the sulfate group being tightly coordinated in the active site. The ATP sulfurylase and APS kinase-like domains of the fungal (P.chrysogenum) enzyme communicate allosterically through structural changes driven by the binding of PAPS, which decrease the affinity for sulfate and potently inhibits APS synthesis (Renosto et al., 1990
; Foster et al., 1994; MacRae et al., 2001; MacRae et al., 2002). It is perhaps not surprising that truncating this domain, which is allosterically linked to the steady-state affinity of sulfate in a close structural homologue, would influence the steady-state affinity of the enzyme for sulfate. As hoped for, the enzyme’s activity proved stable toward change of its oligomeric state caused by the loss of its C-terminal domain.
|
Modeling of a substrate channel onto ATP sulfurylase
A structural alignment of the C-terminal domain of yeast ATP sulfurylase and P.chrysogenum APS kinase is presented in Figure 4. With the exception of the three APS kinase loops shown in red, residues 392–511 of ATP sulfurylase and residues 22–208 of APS kinase (the kinase residues 143–170 are flexible and have not been assigned) are structurally very similar. The r.m.s.d. for the structural alignment of the 116 equivalent C positions is 2.85 Å. Thus, the structural core of the APS kinase has been well conserved in the C-terminus of the ATP sulfurylase.
|
The presence of a kinase core is particularly interesting as animals express a bifunctional PAPS synthetase that contains both ATP sulfurylase and APS kinase activities on a single polypeptide (Li et al., 1995
; Rosenthal and Leustek, 1995; Venkatachalam, 1998). This bifunctionality facilitates the channeling of APS between the ATP sulfurylase and APS kinase active sites (Lyle et al., 1994) promoting an energetically unfavorable APS synthesis reaction while at the same time avoiding APS-mediated enzyme inhibition (Seubert et al., 1985; Satishchandran and Markham, 1989; Renosto et al., 1991).
The yeast ATP sulfurylase has no measurable APS kinase activity but the enzymes from P.chrysogenum and other moulds (Renosto et al., 1990), have retained the ability to bind PAPS, the product of the APS kinase reaction, suggesting a relationship with a bifunctional enzyme. The ATP sulfurylase domains of the yeast and Penicillium enzymes are highly conserved over the first 390 amino acids [65% identical plus 22% similar using the CLUSTALW algorithm (Thompson et al., 1994)]. In contrast, the APS kinase-like domains are 20% identical and 31% similar; in addition, the fungal kinase-like domain is 59 residues longer. Consequently, the alignment contains several gaps and loops. The largest gap (Figure 5, gap 2) corresponds to a loop (loop 2) that is highly disordered in several APS kinase structures. Biochemical and structural evidence have demonstrated that this loop is important in PAPS binding and/or allosteric regulation of the ATP sulfurylase domain of the Penicillium enzyme (Martin et al., 1989; Renosto et al., 1990; Foster et al., 1994). A DALI three-dimensional alignment search identified a third kinase homologue, thymidylate kinase, whose structure is notable because the loop that is disordered in the APS kinase is found well ordered in this enzyme as it closes over and interacts directly with its co-crystallized substrate (Lavie et al., 1997a,b, 1998). The thymidylate kinase loop-residues are well conserved in both the Penicillium ATP sulfurylase, and APS kinase sequences (Figure 6); and the most highly conserved loop residues are those that directly contact the substrate. Thus, this loop, or lid, provides what appears to be a good structural model for the homologous regions of the Penicillium ATP sulfurylase kinase-like domain and APS kinase in a substrate complex. The absence of this loop in the C-terminal domain of yeast ATP sulfurylase would explain this proteins inability to bind PAPS, abolishing PAPS regulation. The second, smaller gap in the sequence alignment between yeast and fungal ATP sulfurylase is 18 amino acids long (Figure 5, gap 1) and corresponds to a loop (loop 1) in the APS kinase structure that again is missing in the C-terminal domain of yeast ATP sulfurylase.
|
|
To assess whether the ATP sulfurylase and kinase-like domains in the yeast and fungal enzymes might be evolutionarily related to a bifunctional enzyme that channeled APS, a hybrid APS kinase-like domain was modeled by grafting the PAPS-binding lid (i.e., loop 2 from the thymidylate kinase structure) and loop 1 from the APS kinase structure, onto the catalytically inactive APS kinase-like domain of yeast ATP sulfurylase (Figure 7). The conformations of both loops were not changed. The structure of such a chimera reveals a deep channel that is
20 Å long, 12 Å deep and at its narrowest point (between the side chains of amino acids 124 and 135) 5.5 Å wide. It starts at the active site of the ATP sulfurylase, between the kinase-like and ATP sulfurylase domains of the same monomer, and leads directly into the postulated PAPS-binding pocket of that kinase-like domain which, in the hexamer, lies directly beneath the ATP sulfurylase domain from which the channeled substrate originated. The channel seems very well suited to guide substrate between the two sites. Loop 1, shown in green, would cover a section of the channel near the APS-synthesizing active site, forming a small tube, or pore, that would be large enough to allow easy passage of nucleotides.
|
The C-terminal domain of yeast ATP sulfurylase is structurally homologous to fungal APS kinase. Although it is enzymatically inactive, this domain has been maintained in spite of selection pressures, suggesting that it has a function. The ATP sulfurylase structure supports the hypothesis that the APS kinase-like domain is structurally independent of the catalytically active N-terminal domains and that now its primary function is to stabilize the oligomeric state of the enzyme. The results presented here confirm that hypothesis. Separating the domains produces a stable fully functional ATP sulfurylase that has a sulfate group tightly coordinated in the active site and consistent with this, binds sulfate with a higher steady-state affinity and thus is a slightly more efficient catalyst than its hexameric predecessor. Careful inspection of the structure reveals that the yeast ATP sulfurylase exhibits a surface groove that connects the active site of the adenylyltransferase and defunct active site of the APS kinase-like domain. It is tempting to speculate that this groove represents a substrate channel in transition either toward or away from full functionality. Certain yeast-like systems, notably the Aquifex aeolicus ATP sulfurylase, contain a fully functional APS kinase domain (Hanna et al., 2002
) offering the possibility of testing the channeling hypothesis. As the hexameric and monomeric forms of ATP sulfurylase possess equivalent enzymatic activity, why does ATP sulfurylase oligomerize? Consistent with the presence of the APS kinase-like domain and the surface groove, oligomerization may be a remnant of or an intermediate towards bifunctionality. As the surface channel mentioned above connects active sites located in separate monomers, oligomerization would be required for efficient transfer of APS from the ATP sulfurylase to the APS kinase active sites. In a more physiological sense, it will be interesting to assess whether the monomeric enzyme is as competent as the hexamer in supporting the metabolic requirements of Saccharomyces cerevisiae.
|
Acknowledgements |
|---|
We thank Professor H.Penefsky, Dr J.Nachmann and Dr B.Eger for technical advice. We are grateful to Professor T.Hofmann and Dr B.Eger for reading and commenting on this manuscript. For their generous time commitment and support, we would like to thank the staff at Brookhaven National Laboratory beam line X8C (supported in part by a grant from the Canadian Insititutes for Health Research and the National Science and Engineering Council of Canada) and at BioCARS sector beamlines at the Advanced Photon Source, Argonne National Laboratories. This research was supported by the National Institutes of Health Grant GM54469 (T.S.L.) and by the Canada Research Chairs Program and the National Sciences and Engineering Research Council (E.F.P.). Use of the Advanced Photon Source was supported by the United States Department of Energy, Basic Energy Sciences, Office of Science, under Contract W-31-109-Eng-38. Use of BioCARS Sector 14 was supported by the National Institutes of Health, National Center for Research Resources, under Grant Number RR07707. Research carried out (in whole or in part) at the National Synchrotron Light Source, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Division of Materials Sciences and Division of Chemical Sciences, under Contract No. DE-AC02-98CH10886.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Beynon,J.D., MacRae,I.J., Huston,S.L., Nelson,D.C., Segel,I.H. and Fisher,A.J. (2001) Biochemistry, 40, 14509–14517.[CrossRef][Medline]
Brauer,P.R., Keller,K.M. and Keller,J.M. (1990) Development, 110, 805–813.
Brünger,A.T. et al. (1998) Acta Crystallogr. D, 54, 905–921.[CrossRef][Medline]
Christopher,J.A. (1998) SPOCK The Structural Properties Observation and Calculation Kit. The Center for Macromolecular Design, Texas A & M University, College Station, TX.
Cleland,W.W. (1979) Methods Enzymol., 63, 103–138.[Medline]
Cowtan,K. and Main,P. (1996) Acta Crystallogr. D, 52, 43–48.[CrossRef][Medline]
de la Fortelle,E. and Bricogne,G. (1997) Methods Enzymol., 276, 472–494.[CrossRef][Web of Science]
Foster,B.A., Thomas,S.M., Mahr,J.A., Renosto,F., Patel,H.C. and Segel,I.H. (1994) J. Biol. Chem., 269, 19777–19786.
Hanna,E., MacRae,I.J., Medina,D.C., Fisher,A.J. and Segel,I.H. (2002) Arch. Biochem. Biophys., 406, 275–288.[CrossRef][Web of Science][Medline]
Hutchinson,E.G. and Thornton,J.M. (1996) Protein Sci., 5, 212–220.[Web of Science][Medline]
Ishihara,M., Tyrrell,D.J., Stauber,G.B., Brown,S., Cousens,L.S. and Stack,R.J. (1993) J. Biol. Chem., 268, 4675–4683.
Izard,T. and Gerlof,A. (1999) EMBO J., 18, 2021–2030.[CrossRef][Web of Science][Medline]
Jevons,F.R. (1963) Biochem J., 89, 621–624.[Web of Science][Medline]
Jones,T.A., Zou,J.Y., Cowan,S.W. and Kjeldgaard,M. (1991) Acta Crystallogr. A, 47, 110–119.[CrossRef]
Jullien,D., Crozatier,M. and Kas,E. (1997) Mech. Dev., 68, 179–186.[CrossRef][Web of Science][Medline]
Kraulis,P.J. (1991) J. Appl. Crystallogr., 24, 946–950.[CrossRef][Web of Science]
Larsen,T.M., Laughlin,L.T., Holden,H.M., Rayment,I. and Reed,G.H. (1994) Biochemistry, 33, 6301–6309.[CrossRef][Medline]
Laskowski,R.A., MacArthur,M.W., Moss,D.S. and Thornton,J.M. (1993) J. Appl. Crystallogr., 26, 283–291.[CrossRef][Web of Science]
Lavie,A., Vetter,I.R., Konrad,M., Goody,R.S., Reinstein,J. and Schlichting,I. (1997a) Nat. Struct. Biol., 4, 601–604.[CrossRef][Web of Science][Medline]
Lavie,A., Schlichting,I., Vetter,I.R., Konrad,M., Reinstein,J. and Goody,R.S. (1997b) Nat. Med., 3, 922–924.[CrossRef][Web of Science][Medline]
Lavie,A., Ostermann,N., Brundiers,R., Goody,R.S., Reinstein,J., Konrad,M. and Schlichting,I. (1998) Proc. Natl Acad. Sci. USA, 95, 14045–14050.
Leyh,T.S. and Suo,Y. (1992) J. Biol. Chem., 267, 542–545.
Li,H., Deyrup,A., Mensch,J.R.,Jr, Domowicz,M., Konstantinidis,A.K. and Schwartz,N.B. (1995) J. Biol. Chem., 270, 29453–29459.
Liu,C., Suo,Y. and Leyh,T.S. (1994) Biochemistry, 33, 7309–7314. [CrossRef][Medline]
Lyle,S., Ozeran,J.D., Stanczak,J., Westley,J. and Schwartz,N.B. (1994) Biochemistry, 33, 6822–6827.[CrossRef][Medline]
MacRae,I.J., Segel,I.H. and Fisher,A.J. (2000) Biochemistry, 39, 1613–1621.[CrossRef][Medline]
MacRae,I.J., Segel,I.H. and Fisher,A.J. (2001) Biochemistry, 40, 6795–804.[CrossRef][Medline]
MacRae,I.J., Segel,I.H. and Fisher,A.J. (2002) Nat. Struct. Biol., 9, 945–949.[CrossRef][Web of Science][Medline]
Martin,R.L., Daley,L.A., Lovric,Z., Wailes,L.M., Renosto,F. and Segel,I.H. (1989) J. Biol. Chem., 264, 11768–11775.
Merrit,E.A. and Murphy,M.E.P. (1991) Acta Crystallogr. D, 50, 869–873.
Morita,T. and Jackson,C.M. (1986) J. Biol. Chem., 261, 4008–4014.
Otwinowski,Z. and Minor,W. (1997) Methods Enzymol., A276, 307–326.
Rossmann,M.G., Moras,D. and Olsen,K.W. (1974) Nature, 250, 194–199.[CrossRef][Medline]
Renosto,F., Martin,R.L. and Segel,I.H. (1989) J. Biol. Chem., 264, 9433–9437.
Renosto,F., Martin,R.L., Wailes,L.M., Daley,L.A. and Segel,I.H. (1990) J. Biol. Chem., 265, 10300–10308.
Renosto,F., Martin,R.L. and Segel,I.H. (1991) Arch. Biochem. Biophys., 284, 30–34.[CrossRef][Web of Science][Medline]
Robbins,P.W. and Lipmann,F. (1958) J. Biol. Chem., 233, 686–690.
Rosenthal,E. and Leustek,T. (1995) Gene, 165, 243–248.[CrossRef][Web of Science][Medline]
Sanchez,O., Ferrera,I., Dahl,C. and Mas,J. (2001) Arch. Microbiol., 176, 301–305.[CrossRef][Web of Science][Medline]
Saridakis,V., Christendat,D., Kimber,M.S., Dharamsi,A., Edwards,A.M. and Pai,E.F. (2001) J. Biol. Chem., 276, 7225–7232.
Satishchandran,C. and Markham,G.D. (1989) J. Biol. Chem., 264, 15012–15021.
Siegel,L.M. (1975) In Greenberg,D.M. (ed.), Metabolic Pathways, Vol. VII, 3rd edn. Academic Press, New York, NY, pp. 217–286.
Seubert,P.A., Renosto,F., Knudson,P. and Segel,I.H. (1985) Arch. Biochem. Biophys., 240, 509–523.[CrossRef][Web of Science][Medline]
Terwilliger,T.C. and Berendzen,J. (1999) Acta Crystallogr. D, 55, 849–861.[CrossRef][Medline]
Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) Nucleic Acids Res., 22, 4673–4680.
Ullrich,T.C. and Huber,R. (2001) J. Mol. Biol., 313, 1117–1125.[CrossRef][Web of Science][Medline]
Ullrich,T.C., Blaesse,M. and Huber,R. (2001) EMBO J., 20, 316–329.[CrossRef][Web of Science][Medline]
Venkatachalam,K.V., Akita,H. and Strott,C.A. (1998) J. Biol. Chem., 273, 19311–19320.
Wei,J., Tang,Q.X., Varlamova,O., Roche,C., Lee,R. and Leyh,T.S. (2002) Biochemistry, 41, 8493–8498.[CrossRef][Medline]
Received July 17, 2003; revised October 27, 2003; accepted October 28, 2003
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||