CTP Synthase from Sulfolobus solfataricus

Master Thesis in Biochemistry Iben Havskov Lauritsen

June 2010 University of Copenhagen, Department of Biology Supervisor: Kaj Frank Jensen

PREFACE

This work represents my master thesis in Biochemistry at the University of Copenhagen. Most of the work was carried out at the Department of Biological Chemistry, Institute of Molecular Biology, University of Copenhagen under supervision by Kaj Frank Jensen. Crystallization was carried out at Centre for Crystallographic Studies, Department of Chemistry, University of Copenhagen under supervision by Eva Johansson. The preliminary work I did on solving the structure was done at Department of Chemistry, Technical University of Denmark under supervision by Pernille Harris. She later finished solving the structure, and a paper on the work is about to be submitted.

I thank Eva Johansson and Pernille Harris for teaching me how to make protein crystals and how to solve the structure of them. That has been a very exciting part of the project for me. I also thank Lise Schack for endless help and good company in the laboratory. Finally I thank Kaj Frank Jensen for encouraging supervision.

______Iben Havskov Lauritsen June 2010, Copenhagen

ABSTRACT

CTP synthase from the extreme thermoacidophilic archaeon Sulfolobus solfataricus has been investigated in several ways in this study. CTP synthase is responsible for de novo synthesis of CTP from UTP. The first part of the reaction is the deamination of glutamine to generate ammonia for the second part of the reaction, the CTP synthesis. This work is mostly focused on the kinetics of the first part of the reaction. GTP was found to activate this glutaminase reaction, by both lowering Km and increasing kcat. Furthermore the activating effect of GTP was found to be increased 20-fold in the presence of the substrates ATP and UTP. ATP and UTP also increased the reaction rate of the glutaminase reaction without GTP. The activation by ATP and UTP was investigated in more detail, and it turned out that neither ATP nor UTP alone displayed much activation. But when ATP was present, increasing concentrations of UTP gave increasing activity rates. Both ATP and UTP displayed positive cooperativity. GTP had a different effect on the production of CTP than it had on the activation of the glutaminase reaction; it was an activator of the production of CTP up to 50-100 M GTP, but inhibited the reaction at higher concentrations. This showed that the two reactions of CTP synthase from S. solfataricus are only coupled up to this concentration of GTP. Size analysis by sucrose gradient sedimentation showed that CTP synthase was a dimer that tetramerized in the presence of nucleotides. The protein was crystallized and the structure solved by X-ray diffraction. Crystal structures of CTP synthases from other organisms had already been published, but unlike these, which had crystallized as tetramers, CTP synthase from Sulfolobus solfataricus had crystallized as a dimer. The enzyme was found to be protected from trypsin-proteolysis by the nucleotides ATP, UTP and CTP. It was also found that CTP could be produced with ammonium chloride as ammonia source instead of glutamine.

INTRODUCTION

Sulfolobus solfataricus Sulfolobus solfataricus is an aerobic, hyperthermophilic, acidophilic crenarchaeon (Archaea) that grows optimally at 80 °C and pH 2-4. It metabolizes sulphur and is able to grow on a variety of carbon sources (2-4). Sulfolobus strains are widely studied, and are model organisms for studying the Crenarchaea branch of the Archaea kingdom, in part because they are relatively easy to culture. Strains of the genus Sulfolobus have been isolated all over the world from acidic solfataric fields, and are basically found everywhere where there is volcanic activity. The strain used in this work, S. solfataricus strain P2, was isolated from a solfataric field near Naples, Pisciarelly (3), and its genome has been sequenced (4). The genes for pyrimidine synthesis are highly conserved and are concentrated in two operon-like structures within 5.5 kb. Only genes encoding carbamoylphosphate synthetase are found elsewhere, clustered with arginine biosynthesis genes (4).

Pyrimidine metabolism Nucleotides are essential for all organisms. Besides their role in energy metabolism, they are used for the synthesis of DNA, RNA, phospholipids and several co-enzymes. The purines and the pyrimidines are synthesized via two different pathways. Since the need for nucleotides is universal for all living (growing) organisms, these de novo pathways are essentially identical throughout the biological world. All sequenced bacterial genomes, except some intracellular parasites, encode the enzymes required for de novo biosynthesis of pyrimidine nucleotides (5). Most organisms also have salvage pathways, where nucleotides are synthesized from nucleosides or bases taken up by the cell or from enzymatic breakdown of nucleic acids (6). These salvage pathways differ between organisms.

1 - 2 ATP + HCO pyrA pyrB pyrC 3 Carbamyl phosphate Carbamyl aspartate Dihydroorotate + glutamine

pyrD

pyrG ndk pyrH CTP UTP UDP UMP pyrF OMP pyrE Orotate

ndk udk upp

CDP Uridine udp Uracil uraA Uracil from outside the cell

cmk cdd codA

CMP udk Cytidine rihA Cytosine codB Cytosine from outside the cell

Figure 1. Schematic representation of pyrimidine nucleotide de novo and salvage pathways for E. coli. At the top of the figure is shown the unbranched de novo pathway leading ultimately to CTP and below that are shown the salvage reaction pathways. The enzymes encoded by the genes in the de novo pathway are: Carbamoyl phosphate synthase/CP synthase (pyrA); aspartate transcarbamoylase/ATCase ( pyrB); dihydroorotase/DHOase (pyrC); dihydroorotate dehydrogenase/DHODase (pyrD); orotate phosphoribosyltransferase/OPRTase (pyrE); orotidylate decarboxylase/OMP decarboxylase (pyrF);UMP kinase (pyrH); nucleoside diphosphate kinase/NDP kinase (ndk); CTP synthase (pyrG). Enzymes of the salvage pathways are: CMP kinase (cmk); uridine kinase (udk); cytidine deaminase (cdd); uridine phosphorylase (udp); ribonucleoside hydrolase A (rihA); Cytosine deaminase (codA); UPRTase (upp). The cytoplasmic membrane proteins are cytosine permease (codB) and uracil permease (uraA).

CTP synthase The pyrimidine de novo pathway is unbranched, and the last step is the amination of UTP to produce CTP (7). This process is catalyzed by CTP synthase/synthetase (CTPS) which is encoded by pyrG in bacteria, ctrA in some gram positive bacteria, URA7 and URA8 in Saccharomyces cerevisiae and CTPS1 and CTPS2 in humans. Most eukaryotes express two isoforms (8). The enzyme has been studied intensively for many years. Especially the Escherichia coli enzyme is well characterized, but a lot of work has also been done on CTPS from S. cerevisiae, Lactococcus lactis and recently also on human CTPS (9). In 2004 the structure of both E. coli and Thermus thermophilus CTPS was solved (8;10), and the structure of the synthase domain of human CTPS has also been published (11). CTP synthase is a Mg2+-demanding, glutamine amidotranferase, weighing approximately 60 kDa. It consists of a single polypeptide chain, and is a homotetramer in its fully active form. The overall reaction of CTP synthase is the deamination of glutamine to glutamate, and the amination of UTP to

2 CTP at the expense of an ATP to ADP dephosphorylation (12). CTPS from L. lactis and S. cerevisiae are also able to catalyze the conversion of dUTP to dCTP. For L. lactis CTPS the reaction is probably not physiologically relevant (13), but it might play a role in the nucleotide metabolism of S. cerevisiae (14). The action of CTPS consists of two separate reactions, taking place in different domains of the enzyme.

Figure 2. The reaction catalyzed by CTPS. CTPS catalyzes both hydrolysis of glutamine to glutamate and ammonia and ATP dependent synthesis of CTP from UTP. Picture is reproduced from (10).

Glutamine amide transfer domain In the C-terminal end of the enzyme is a class I glutamine amidotransferase (GATase) domain, with the catalytic triad Cys-His-Glu. Here ammonia is generated by glutamine hydrolysis via the formation of a covalent cysteinyl-glutamyl intermediate (8;10;12;15;16)

Amidoligase/synthase domain In the N-terminal end of the enzyme is the synthase domain (amidoligase, ALase). This domain structurally resembles dethiobiotin synthase (8;10). Here UTP is activated by ATP-dependent phosphorylation at the 4-position and then reacts with ammonia (12;17-19). The enzyme can use external ammonia instead of nascent ammonia from glutamine hydrolysis (20;21), but this is not thought to be of physiological importance (22).

Ammonia tunnel The ammonia from the glutaminase domain is delivered to the synthase domain via a tunnel in the enzyme (8;23-25). Such a tunnel is seen in several other enzymes with GATase domains (26). In most of these enzymes with molecular tunnels, the tunnel is preformed, and

3 exists also in the absence of ligands bound to the active site, but there is at least one example where the tunnel not formed before substrates have bound to their respective active sites (27). In the crystal structures of CTPS from T. thermophilus (10), no ammonia tunnel is seen, but it is suggested that binding of ATP and UTP makes the glutaminase domain approach the synthase domain to make a connecting channel between the two active sites, thereby coupling the two half-reactions. In the structure of E. coli CTPS (8), a tunnel is seen between the two active sites. The tunnel is accessible to solvent close to the GATase site at the proposed GTP binding site, and this is probably where exogenous ammonia enters the active site (8).

Regulation of the reaction Perhaps not surprisingly, the reaction is highly regulated. It is a key step of controlling the level of CTP in the cell, and the activity of CTPS is regulated by interaction with all four ribonucleotide triphosphates. It is directly inhibited by its own product, CTP, which act as a negative feedback control. The glutaminase half-reaction is activated by GTP. Both the substrates ATP and UTP control the reaction rate upon binding, by associating the enzyme to its active form as a tetramer. Moreover, ATP and UTP act as allosteric activators of the glutaminase half-reaction. In general, CTPS show positive cooperativity for ATP and UTP and negative cooperativity for glutamine and GTP (28).

Regulation by ATP and UTP ATP and UTP show strong positive cooperativity in E. coli CTPS, with Hill-coefficients of 3.8 and 3.2 respectively (21;22;28). Positive cooperativity is also seen in S. cerevisiae (29;30) and L. lactis (13), but not in T. thermophilus (10) or in mammalian CTPS (31). Besides being substrates of the synthase reaction, ATP and UTP also act as allosteric activators of the glutaminase half-reaction, which can take place in any oligomeric state of the enzyme (32). Studies on L. lactis CTPS suggest that formation of the 4-phosphorylated UTP intermediate also stimulate glutamine hydrolysis, acting as a co-activator with GTP (33).

Tetramerization CTPS is a dimer-of-dimers homotetramer in its active state. Without ligands and at low concentrations, it exists as dimers or monomers, but ATP and UTP binding induce tetramerization in a synergistic way (32;32;34-39). Both ATP and UTP alone can induce

4 tetramerization in E. coli CTPS at high enough concentrations (32). In S. cerevisiae CTPS, UTP alone can cause the enzyme to tetramerize, but ATP alone cannot. Data indicate that ATP facilitates the tetramerization of CTP synthase through its role in the phosphorylation of UTP (37). L. lactis CTPS is always on a tetramer form; the tetramer interface being stabilized by ionic interactions. If these interactions are destroyed, the enzyme forms tetramers in response to ATP and UTP as other CTP synthases. It shows cooperativity with ATP and UTP even though it is always found as a tetramer. The ATP and UTP binding sites are located at the interface of the dimer that contact the other dimer (8). This suggests that at least some of the positive cooperativity observed for ATP and UTP is caused by the substrate dependent association-dissociation mechanism of the enzyme (35). But allosteric effects of ATP and UTP binding probably also contribute to the positive cooperativity. At least this seems to be the case in the L. lactis enzyme, which is always a tetramer, but still shows positive cooperativity with ATP and UTP (13;40). For S. cerevisiae CTPS there is found to be correlation between the kinetics of activity and of tetramerization, and it is suggested that the cooperative effects seen with ATP and UTP, is a reflection of the tetramerization of the enzyme (37). The fact that ATP and UTP make the enzyme tetramerize makes it more complicated to describe the mechanism of this observed cooperativity.

Regulation by CTP

CTP inhibits the enzymatic activity by acting as a competitive inhibitor of UTP, with a Ki value that is close to the Km value for UTP (21;41). CTPS from S. cerevisiae, containing a mutation that makes it defective in product inhibition, shows 16-fold to 20-fold increased levels of CTP in the cell (42). CTP binds to the active site at a position that is overlapping the one proposed for UTP, in such a way that the same triphosphate-binding site is used (8) (See Figure 3). At low enzyme concentrations, CTP stimulates activity instead of inhibiting it, and induces tetramerization of the enzyme (37). Furthermore, CTP has been shown to induce positive cooperativity of UTP in CTPS from mammals (31;43) and to increase the positive cooperativity of UTP in CTPS from S. cerevisiae (29).

5 Figure 3. Proposed binding sites of UTP and CTP. UTP and CTP are proposed to share the same binding site for their 5’triphosphate moieties in the active site. Picture reproduced from (1).

Regulation by GTP The GTP binding site is located between the two active sites, at the dimer interface. It binds where the ammonia tunnel opens to the solvent. In this way it can close the tunnel and optimize channelling of ammonia from the GATase domain to the ALase domain (44). In this way GTP promotes coordination of the phosphorylation of UTP and hydrolysis of glutamine, so the two half-reactions become coupled (33). GTP acts allosterically to activate glutamine hydrolysis by stabilizing the enzyme conformation that binds the tetrahedral intermediates formed during glutamine hydrolysis (45), thereby enhancing the rate of formation of the intermediate (15;21). GTP both raises Kcat and lowers Km of the glutaminase half-reaction. This allosteric activation by GTP is unique for CTPS among the group of enzymes with a GATase domain (10;44). In a recent report (46), GTP is shown to inhibit glutamine-dependent CTP formation in E. coli above concentrations of 0.15 mM, while still stimulating the glutamine half- reaction. CTP formation with exogenous ammonia is inhibited at any concentration of GTP. CTPS from E. coli has been reported to show negative cooperativity for GTP when the enzyme is on its tetrameric form (15;28). Some newer reports have not been able to confirm this (47). Negative cooperativity of GTP and glutamine has also been reported for the URA7 encoded isoform of CTPS from S. cerevisiae (29), whereas no cooperativity for GTP and glutamine is seen with CTPS from the URA8 encoded isoform (30). For mammalian (48), and L. lactis CTPS (13), no negative cooperativity for GTP is seen.

6 Regulation by phosphorylation Studies of URA7 encoded S. cerevisiae CTPS have shown that this enzyme is phosphorylated by Protein Kinase A (PKA) and Protein Kinase C (PKC), and that this phosphorylation controls both activity and tetramerization (37;49-51). Human CTPS encoded by CTPS1 and expressed in S. cerevisiae has also been shown to be phosphorylated by these two protein kinases (52;53), but studies on the same enzyme expressed in human cells have failed to reproduce this result (54). However, the enzyme was found to be phosphorylated by a different protein kinase, GSK3, and the activity of the enzyme is regulated by this phosphorylation.

Regulation of pyrG expression Genetic control of the level of PyrG expression has been shown in both Bacillus subtilis and L. lactis (5). For L. lactis it has been shown that the pyrG expression is directly correlated with to the intracellular concentration of CTP (55), and regulation of pyrG expression in B. subtilis has also been shown to be CTP sensitive. In B. subtilis, pyrG is regulated by transcription attenuation, and the conserved sequences that cause this regulation of pyrG expression in B. subtilis are found in many other gram-positive bacteria (5).

Prospective in medical treatment of diseases Conversion of UTP to CTP is the rate-limiting step in the formation of cytosine nucleotides, and it is an important point of regulation in pyrimidine biosynthesis (56). The activity of the de novo pyrimidine pathway is low in resting or fully differentiated cells, as the salvage pathway can supply the cells with pyrimidines. In proliferating cells, for example neoplastic cells and tumor cells, the de novo pyrimidine pathway is highly active (56). CTPS activity is increased in several human cancers (57;58). Viruses and parasites also have an extra need for nucleotides for RNA and DNA synthesis, and all of these are sensitive to CTP depletion caused by CTPS inhibition (59). CTPS is therefore the target of several drugs, and a lot of research is being done (1;1;60). There are two categories of CTPS inhibitors. One is the glutamine antagonists, for example 6-diazo-5-oxo-L-norleucine (DON) and acivicin, which have both been used successfully to control the development of African sleeping sickness caused by Trypanosoma brucei, in vitro and in infected mice (61;62). The structure of the glutaminase domain of T.

7 Brucei CTPS with bound Acivicin has been solved recently (PDB Code 2W7T). Acivicin also suppress HIV activity in infected cells and has anti-tumour activity (63). The second category of CTPS inhibitors is the nucleotide/nucleoside analogues, for example cyclopentenylcytosine (CPEC) and 3-deazauridine. These two pyrimidine analogues are effective CTP inhibitors that arrest cancer cell proliferation (41). In the parasite Plasmodium falciparum, which is responsible for malaria in humans, activity of CTPS is the only source of cytidine nucleotides, and CTPS can therefore be a target for the development of anti-malarial drugs (59). Several other conditions are also potentially treatable using anti-CTPS drugs, for example giardiosis (64), chlamydia (65;66) and hemorrhagic fever (67). Cytidine analogues are less attractive as anti-CTPS drugs; resistance to these drugs can be acquired without inhibition of CTP synthesis activity, because of the way the UTP and CTP binding sites overlap (8).

8 MATERIALS AND METHODS

General instrumentation Buch & Holm Bandelin Sonopuls sonicator HO2000 with sonication probe KE76 Sorvall RC 5B centrifuge with rotors: SS-34 (8·50 ml), GSA (6·250 ml), GS-3 (6·500 ml) Beckman ultracentrifuge with SW41 swinging bucket rotor Eppendorf table centrifuge Canberra-Packard instant imager CycloneTM Storage Phosphor Screen (Packard) phosphor imager Packard tri-carb 2200 CA liquid scintillation analyzer Eppendorf photometer Beckman DU 650 spectrophotometer Zeiss Specord S10 spectrophotometer BioRad BioLogic Workstation FPLC system

Materials Dyematrex Gel Red A was from Millipore Corporation. Diethylaminoethyl cellulose (DE52, preswollen) was from Whatman International Ltd. Poly(ethyleneimine)-impregnated cellulose thin-layer plates (PEI-plates) were made according to a described procedure (68) or were from MERCK. Cellulose thin-layer plates were from MERCK. L-[U-14C] glutamine was from Amersham Biosciences. [2-14C]uridine 5’-triphosphate (UTP) was from Moravek Biochemicals. Yeast alcohol dehydrogenase and bovine liver catalase was from Boehringer- Mannheim. Trypsin (type XIII) from bovine pancreas was from Sigma. Nap10 and Nap5-Pre- packed Sephadex G-25 columns were from Amersham Biosciences. Centriprep and centrifugal filters and Ultrafree MC centrifugal filter Biomax-30 (30,000 NMWL) were from Millipore Corporation. Other chemichals were purchased from various sources.

Construction of an expression vector DNA from S. solfataricus was used as template in a PCR reaction with Vent polymerase. Two primers were created, the one with a Shine Dalgarno sequence at the 5’ end for optimum alignment of the mRNA to the ribosome. The plasmid pUHE 23-3 (69), which contains a strong promoter, suitable cloningsite and a Lac operon was digested with Sal 1 (5’ overhang

9 G↓TCGAC) and BamHI (G↓GATC*C) and treated with alkaline phosphatase to prevent self- ligation. The PCR product was also digested with these digestion enzymes, and was ligated to the plasmid with a T4 DNA ligase. The plasmids were transformed into competent (Ca2+ treated) NF1830 cells (E. coli K-12, recA1/F’lacIq1 lacZ::Tn5) (70;71). This strain carries an F’lacIq1 episome and overproduces the lac-repressor so that transcription of the cloned gene is repressed until induction by isopropyl thio-β-D-galactoside (IPTG). The transformed cells were plated onto LB agar plates supplemented with ampicillin (100 g/ml) so only the transformed cells would grow. Only four colonies showed up, and of these only one, KDS4, contained plasmid with the insert. Sequencing confirmed the right insert, except for an extra BamHI site that had appeared.

Cell growth Cells were grown in LB medium (72) made with minimal AB medium (73) instead of water, and enriched with glucose, to give as many cells as possible. For five litres of final growth medium, 4.5 litre of LB were made, with B medium instead of water, using either NZ-amine or Tryptone, and with no salt added, since this is included in the AB medium. pH was set to 7.0 - 7.2 with solid NaOH. The solution was distributed in three big flasks and autoclaved at 120 °C. When cooled down, 500 ml of sterile A medium, 500 mg ampicillin (to 100 mg/l in solution) and 100 g glucose (to 2 % w/v in solution) was added. The glucose was first dissolved in a small volume of GDW sterilized by boiling. Five litres of medium was inoculated with an overnight culture of NF1830, containing the plasmid pKDS4, to an optical density OD436 of approximately 0.05. Cells were grown with vigorous aeration at 37 °C. OD436 was followed using the Eppendorf photometer, and at an OD436 of 0.5, expression of the pyrG gene on the plasmid was induced with 0.5 mM IPTG and growth was continued into stationary phase overnight. Cells were harvested by centrifugation with the GS-3 rotor at 5,000 rpm for 15 minutes. Pellets were re-suspended in cold 0.9% NaCl and spun down at 12,000 rpm with the SS-34 rotor for 15 minutes. This procedure gave 28-31 gram of cells that were stored at -20°C for later purification of CTP synthase.

10 Purification Approximately 30 g of cells were suspended in 100 ml sonication buffer and disrupted by ultrasonic treatment for 30 minutes at intervals, at power of approximately 35 %. After sonication, β-mercaptoethanole (β-ME) was added to a concentration of 2.0 mM. Cell debris was removed by centrifugation at 9,000 rpm for 10 minutes with the SS-34 rotor. The solution was heated to 70 °C in a water bath for 10 minutes under constant stirring, to denature non-heat stable E. coli proteins. The extract was cleared by centrifugation at 10,000 rpm for 15 minutes in the SS-34 rotor. A lot of precipitation was discarded.

Ammonium sulfat ((NH4)2SO4) was added to the solution at a concentration of 30%; the solution was stirred for half an hour and any precipitate was removed by centrifugation at 13,000 rpm for 15 minutes in rotor SS-34. The concentration of ammonium sulphate was then brought up to 60%, the solution was again stirred for half an hour and the precipitate was collected by centrifugation as before. The pellet was resuspended in 5.0 ml buffer A and dialysed against this same buffer overnight to remove the ammonium sulphate. The fraction was applied to a 60 ml DE-52 (DEAE-cellulose) column. The column was first run with 50 ml of buffer A, next with a 100 ml linear NaCl gradient of buffer A
buffer B and was finally washed with 50 ml buffer B. Forty fractions of 5.0 ml were collected. CTPS was found in the first peak of the salt gradient, at conductivity approximately 10 mS/cm to 28 mS/cm. The relevant fractions were pooled and dialysed for one hour against with buffer A. The dialysate was applied to a 30 ml Dyematrex gel red A column. The column was run with the same program as the DE-52 column, but with buffer C instead of buffer B. CTPS was found in the only peak of the salt gradient, at conductivity approximately 20 mS/cm to 40 mS/cm. The relevant fractions were pooled and concentrated using Centriprep according to the manufacturer’s instructions. Volume was reduced from 50 ml to approximately 11 ml. Finally the concentrated protein was dialysed overnight against buffer A with 50% glycerol, after which the volume was approximately 3 ml and the concentration of CTP synthase was approximately 7 mg/ml. The enzyme was then stored at -20 °C.

Determination of protein concentration: The extinction coefficient of CTPS was calculated from amino acid data. Theoretical absorption of UV light is estimated from the content of absorbing amino acids in the protein,

11 namely tryptophan, tyrosine and cysteine (74). The calculated extinction coefficient at A280 was 0.68517464, and A280 was measured on a Zeiss spectrophotometer to determine the protein concentration.

Purification buffer list Sonication buffer: 100 mM Tris/Cl (pH 7.6), 2 mM EDTA. Buffer A: 25 mM tris/Cl (pH 7.6), 0.1 mM EDTA, 2.0 mM β-ME. Buffer B: 25 mM tris/Cl (pH 7.6), 0.1 mM EDTA, 2.0 mM β-ME, 333 mM NaCl. Buffer C: 25 mM tris/Cl (pH 7.6), 0.1 mM EDTA, 2.0 mM β-ME, 1.0 M NaCl.

Crystallization Crystallization setup The crystallization experiments were made by the hanging drop vapour diffusion technique. This method works by concentrating the protein solution by evaporation. A small droplet of the concentrated protein together with a precipitating agent in a buffer is put in a sealed chamber containing a reservoir with a much larger volume of the same buffer and a precipitating agent at a higher concentration. In this closed environment, the droplet will equilibrate with the reservoir solution and slowly the concentrations of both protein and precipitant will increase by evaporation from the droplet. Hopefully crystals will start to form in the supersaturated solution. Glycerol was removed and buffer of CTPS changed using NAP-5 or NAP-10 columns. The enzyme was eluted from the column with 25 mM tris (pH 7.6) with 2 mM β- ME and 0.1 mM EDTA, and was concentrated using Ultrafree centrifugal filter to a concentration of approximately 10 mg/ml. The crystallization was set up in VDX plates from Hampton Research. Each well was filled with 0.5 ml of reservoir solution. Two l of enzyme (10 mg/ml CTPS, 25 mM tris (pH 7.6), 2 mM β-ME, 0.1 mM EDTA) was put on a siliconized cover slip; 2 l of the solution from the reservoir was added to this drop, and the cover slip was turned upside down and was sealed to the well with vacuum grease.

12 Solubility footprint screen (75) The screen consists of six different precipitants with three different pH values at four different concentrations. In general, the concentration of the protein to be crystallized is suitable when the first row, with the lowest precipitant concentration, does not precipitate the protein at all, and the last row, with the highest precipitant concentration, precipitates the protein almost immediately. The screen was set up with a protein solution of approximately 10 mg/ml, giving a concentration of 5 mg/ml in the drop. This protein concentration was suitable, since half of the conditions showed precipitate.

Solubility Precipitant/pH 2- footprint PEG400 PEG 4000 PEG 0000 (NH4)2SO4 PO3 Citrate screen 5.5 7.0 8.5 5.5 7.0 8.5 Precipitant 15% (cl) 10% (p) 7.5% (p) 0.75M (cl) 0.8M (cl) 0.75M (cl) concentration 24% (cl) 15% (p) 12.5% (p) 1.0M (cl) 1.32M (cl) 1.0M (cl) 33% (p) 20% (p) 17.5% (p) 1.5M (cl) 1.6M (cl) 1.2M (p) 42% (p) 25% (p) 22.5% (p) 2.0M (cl) 2.0M (cl) 1.5M (p) Table 1. Solubility footprint screen results. The PEG solutions are buffered to the right pH by 0.1 M Imidazole/malic acid buffer. (cl) = clear drop, no precipitation; (p) = precipitation.

Hampton Crystal Screen 1 & 2 (76) Polycrystalline aggregates or microcrystalline precipitate showed up with several of the reagents. The most promising results were seen with reagents 18 (0.1 M sodium cacodylate (pH 6.5), 0.2 M magnesium acetate, 20% PEG 8000) and 46 (0.1 M sodium cacodylate (pH 6.5), 0.2 M calcium acetate, 18% PEG 8000) from Crystal Screen 1, which gave star-shaped, needle-thin polycrystals.

Optimization Reagent 18 and reagent 46 are very much alike, and the following 4x6 optimization screen was set up; 5, 10, 15 and 20% PEG 8000 against 0.05, 0.1 and 0.2 M magnesium acetate and 0.05, 0.1 and 0.2 M calcium acetate, all in 0.1 M sodium cacodylate buffer (pH 6.5). This screen gave crystals at all conditions, except the four wells with only 5% PEG 8000. The best crystals were seen with 10% PEG 8000 and 0.1 M magnesium acetate. A 4x8 optimization screen was set up in order to optimize the crystals; 7, 8, 9 and 10% PEG 8000 against eight different buffers (Sodium Cacodylate (pH 6.5), Sodium Acetate (pH 5.5), MES (pH 6.0), MES (pH 6.5), Hepes (pH 7.0), Hepes (pH 7.5), Tris (pH 8.0), Tris

13 (pH 8.5)). All buffers were at 0.1 M and the magnesium acetate concentration was kept constant at 0.1 M. The best crystals from this screen were seen with 0.1 M sodium acetate, pH 5.5 and 7% PEG 8000. This is both the lowest pH value and the lowest PEG concentration in the screen, so another screen was set up to see if better crystals would be produced with even lower values. Four different PEG 8000 concentrations, 4, 5, 6 and 7%, were screened against sodium acetate at three different pH values; 4.5, 5.0 and 5.5. The magnesium acetate was kept constant at 0.1 M in all solutions. The best crystals showed up with pH 4.5 and 7% PEG 8000. A new portion of protein from the same batch was concentrated to approximately 10 mg/ml, making it necessary to screen the PEG concentration once again. A screen was then set up with 3, 4, 5, 6, 7 and 8% PEG 8000 against 0.1 M sodium acetate at pH 3.5, pH 4.0, pH 4.5 and pH 4.5 with 5.0 mM UTP (UTP present only in the protein solution; not in the reservoir), keeping the magnesium acetate concentration constant at 0.1 M. Here good crystals showed up at pH 4.5 at several concentrations of PEG 8000. Four crystals from the last tray were selected for data collection and mounted on cryoloops.

Diffraction data collection and statistics Diffraction data were collected with crystal no. 519 at MAX-lab in Lund, Sweden, at beam line I911-5 (with a fixed wavelength of 0.9059 Å). The crystal-to-detector distance was 200 mm. 338 images were collected with an oscillation of 0.50 degrees (covering a total of 165 degrees). The data were processed with the HKL2000 suite (77), which contains the program Denzo, used for autoindexing (deducing a crystal lattice), refinement of the crystal and detector parameters and integration of the dataset (measurement of spot intensities), and the program Scalepack, used for scaling (finding the relative scale factors between measurements), global refinement/post refinement and statistical analysis.

Calculations The volume of the unit cell, V, was calculated using equation 1 (78)

2 2 2 V = a ⋅ b ⋅ c ⋅ 1− cos ⋅α − cos ⋅ β − cos ⋅γ + 2 ⋅ cosα ⋅ cos β ⋅ cosγ (Eq. 1) where a, b and c are the cell dimensions in Ångstrøm (Å) and α, β and γ are cell dimensions in degrees, of the unit cell.

14 The Matthews volume, VM, was calculated using equation 2 (79)

VUC VM = (Eq. 2) M W ⋅ N AU ⋅ NUC where VUC is the volume of the unit cell, MW is the molecular weight of the protein monomer,

NAU is the number of monomers in the asymmetric unit and NUC is the number of asymmetric units in the unit cell. In the case of space group P1, NUC is 1.

The water content of the crystal, Vsolvent, was calculated using equation 3 (79) .1 23 Vsolvent = 1− (Eq. 3) VM

Size analysis by sucrose gradient sedimentation Sucrose gradient sedimentation of proteins can be used as a crude (but straightforward) way to estimate the molecular weight. The protein to be studied is run in the gradient with one or more marker enzymes of known molecular weight and/or sedimentation constant. The distance that a protein will travel in a linear gradient is directly proportional to the sedimentation constant and the molecular weight. Four 5-20% sucrose gradients of 12.0 ml were made in 50 mM tris-buffer (pH 8.0) containing 10.0 mM MgCl2 and 2.0 mM dithiothreitol (DTT). Two of the gradients had additionally 1.0 mM of UTP, ATP and GTP. The gradients were made with a gradient maker, which consists of two chambers, connected at the bottom. The exit chamber was filled with the 20% sucrose solution, and the other with the 5 % solution. This way the gradients would have a 20 % sucrose solution at the bottom and a 5 % solution at the top, with a linear gradient in between. The gradients were place at 4 °C overnight. The next day, 100 µl solution containing CTPS and the marker enzymes bovine liver catalase and yeast alcohol dehydrogenase was carefully placed on top of the gradients; the gradients were centrifuged for 23 hours at 39,000 rpm at 4 °C in a Beckman SW41 swinging bucket rotor. The gradients were tapped from the bottom with the aid of a peristaltic pump, with 10 drops in each fraction (approximately 46 fractions per gradient), and these fractions were assayed for CTPS, catalase and alcohol dehydrogenase.

15 Calculations The molecular weights of CTPS were calculated using Equation 4, which describes the ratio between the distance travelled by two proteins through the gradient (80)

2   3 dist1 s 1 MW1 (Eq. 4) R = = =   dist 2 s2  MW2 

↓

3  dist  2  1  MW1 =   ⋅ MW2  dist 2  where R is the ratio between two proteins, dist is the distance travelled by a protein through the gradient, s is the sedimentation constant and MW is the molecular weight. This method only gives approximate results, since it is assumed that the proteins have equal partial specific volumes (0.725 cm3 per g), and that they are spherical in shape. In the calculations of the apparent molecular weight of CTPS, the molecular weights of the marker enzymes were taken from (80), and these are for catalase 250 kDa (s20,w =11.4 S) and for alcohol dehydrogenase 150 kDa (s20,w =7.6 S). The distances travelled by the proteins were found by subtracting the peak position from the total amount of drops in the gradient. The approximate molecular weight of CTPS was then calculated from the position of both marker enzymes, and an average result was calculated. The final calculated weights represent an average of two gradients.

Trypsin limited proteolysis Trypsin is a proteolytic digestive enzyme, an endopeptidase. It has an optimal operating pH of 8 and optimal operating temperature of 37 °C. It cleaves proteins at the C-terminal side of lysine and arginine, except where these are followed by proline. Trypsin can activate its inactive precursor trypsinogen by autolysis. Proteolysis was conducted in 35 l assays in eppendorf tubes at 37 °C in a heating block. The reaction was initiated by addition of trypsin (160 g/ml) to the preheated reaction mixture (70.0 mM Hepes (pH 8.0), 0.5 mM EDTA, 10.0 mM MgCl2, 0.8 mg/ml CTPS and 5.0 mM of nucleotides, when these were present). Aliquots of 10 l were removed from the

16 reaction mixture at 10, 30 and 60 minutes and directly transferred to SDS-PAGE loading buffer to stop the reaction. The fractions were loaded onto and run on a 15% SDS-PAGE gel.

Assays Catalase assay

The assay follows the decrease in absorbance at A240. One ml assay mix containing 50 mM

NaCl, 1.0 mM EDTA and 0.06% H2O2 in 50 mM Tris/Cl (pH 7.5) was started with 20 l enzyme fraction and the reaction was measured in the Beckman spectrophotometer.

Alcohol dehydrogenase assay

The assay follows the increase in absorbance at A340. One ml assay mix containing 2.0 mM EDTA, 3% ethanol and 0.2 mM NAD in 50 mM Tris/Cl (pH 7.8) was started with 50 l enzyme fraction and the reaction was measured in the Beckman spectrophotometer.

CTPS assays – CTP synthesis All assays with CTPS were run at 60 °C, and the standard assay contained 2 mM DTT, 20 mM MgCl2, 10 mM glutamine, 1 mM ATP, 1 mM UTP and 1 mM GTP in 50 mM Hepes (pH 8.0). When different concentrations were used, specific details are given in legends to figures or in text. CTPS was taken from the 50% glycerol stock stored at -20 °C. The nucleotide-free sucrose gradients were assayed by a continuous -1 -1 spectrophotometric assay following the increase in A291 (ε1338 M ·cm ) (21). The assay was run in the Zeiss spectrophotometer, and was initiated by adding 50 l preheated enzyme fraction to 150 l preheated assay mixture without GTP. The sucrose gradients with nucleotides could not be assayed spectrophotometrically because GTP absorbs strongly at 291 nm. Therefore a radioactive TLC assay was made on PEI-plates using [14C]-UTP. A fixed-time assay with a reaction time of one hour was performed with 15 l of preheated assay mixture in eppendorf tubes. The reaction was started with 10 l preheated enzyme fraction. A start line was drawn with a soft pencil on the PEI plates 1.5 cm from the bottom, and start spots were drawn on this line 1.5 cm apart. Ten l of nucleotides in formic acid (2.0 M HCOOH, 1.0 mM ATP, 1.0 mM UTP, 1.0 mM GTP and 1.0 mM CTP) was applied to the start spots prior to application of 10 l of the reacted assay mixture. After the plates were dry, they were run up to the start-spot line in methanol, and

17 then developed in 0.85 M KPO4 (pH 3.4) (81), to separate CTP from UTP. When dry, the plates were counted in an instant imager or the spots were cut out and counted by liquid scintillation. This assay was also used to investigate the effect of GTP on the reaction rate and to measure reaction with NH4Cl as a substrate instead of glutamine. In these assays, the assay volume was 60 l, enzyme concentration was 0.020 mg/ml and 10 l aliquots were removed after 2, 5, 10 and 15 minutes and spotted onto the PEI plates. The plates were developed in 2M HCOOH, 2 M LiCl (1:1), and were counted in the instant imager when dry.

Initially, PEI plates were developed in 0.85 M KPO4 (pH 3.4), but the UTP and CTP separation was poor in this system, so after trying different solvents, 2M HCOOH, 2 M LiCl (1:1) (82) was chosen since separation of UTP and CTP was much better in this system.

CTPS assays - glutaminase The formation of glutamate from glutamine was followed with [U-14C] glutamine by a TLC assay on cellulose plates. The standard assay is the same as that for CTP synthese by TLC on PEI plates, and start line and start spots were made in the same way. Enzyme and nucleotides were pre-incubated together for 3-5 minutes at 60°C and assays were started by addition of glutamine. The assay volume was 50 l, and 10 l aliquots were removed at 5, 10 and 15 minutes and spotted onto cellulose plates. Alternatively only one sample was taken, if initial velocities were certain to be linear within the duration of the assay. In this case assay volumes were 15 l and aliquots of 6 l were spotted on the plates. The CTPS concentration was not constant in all assays; the enzyme concentration was varied so that 5-15% of the glutamine would be converted. This was done because a lesser degree of conversion gave unreliable results due to a high background and poor separation. At a higher degree of conversion, ATP or UTP would in most cases be used up. Using longer assay times also gave errors, due to evaporation. The plates were either half (10 cm) or whole (20 cm), and were developed in isopropanol : formic acid : GDW (80:2:20) (83) for approximately 7 hours. When plates were dry, they were either counted in the instant imager for approximately one hour or put on multipurpose screens overnight and counted in the phosphor imager. Because of problems with poor separation in this system, a decision was made to use whole plates instead of halves. One problem with the separation was of a different nature,

18 though, and it was the fact that the cellulose plates, though seeming symmetrical, had a direction in which they separated nicely and one in which they separated badly. The tin foil that the cellulose material was attached to roughned by small ridges for better adherence of the material. Only when these ridges were parallel to the solvent front did the spots separate as expected.

Quality of the glutaminase assay Some assay tests were performed. One test showed that glutamate did not inhibit the reaction to a great extent. Another test showed that the reaction rate was the same whether or not the enzyme had been incubated with nucleotides before performing the assay, and that the initial rate was linear for up to at least 15 % conversion of glutamine. A third test showed that the specific activity of the enzyme did not drop with CTPS concentrations as low as 0.01875 mg/ml. When the enzyme concentration was lower than that, the specific activity did change. Because of this, kinetic results at low substrate concentrations are missing in the report. These might have shed light on some details of the reaction kinetics. Not all initial rates started at the origin, and this was especially the case at low enzyme concentrations. This was a problem that has not been resolved in this work.

Datahandling and Equations Analysis of initial velocity data. All graphs were plotted with GraphPad Prism version 4 for Windows, and kinetic parameters were determined by fitting initial rate kinetic data to the appropriate equations by nonlinear regression analysis, using the same program. Standard deviations are the ones calculated by the program. A segment was fit for velocity at ligand-concentration zero except when glutamine was the varied substance, in which case the reaction rate was zero without substrate. Whenever kcat values are reported, the segment is included in this value. A fit was always attempted with both equations for hyperbolic kinetics (Eq.5) and for sigmoid kinetics (Eq.6). The reaction rates were divided by total concentration of enzyme subunits, so the reaction rates have the dimension of s-1.

19 Enzyme kinetics equations Equation for hyperbolic kinetics (Michaelis-Menten equation):

Vmax ⋅[L] v = + v0 (Eq. 5) K0.5 + []L

Equation for sigmoid kinetics (Hill equation):

n Vmax ⋅[L] v = n n + v0 (Eq. 6) K0.5 + []L

Equation for uncompetitive inhibition

Vmax ⋅[L] v = + v0   (Eq. 7)  []L  []L ⋅1+  + Ki  KA 

[L] is the concentration of ligand, whether substrate or effector v is the reaction rate Vmax is the limiting rate K0.5 is the concentration giving half limiting rate Ki is an inhibition constant KA is an activation constant v0 is reaction velocity at [L] = 0 n is the Hill coefficient

20 RESULTS

Cloning, expression and purification A bachelor project about CTPS from Sulfolobus solfataricus had been made in the laboratory by Karen Duus Sørensen (84) prior to my work. So an expression vector had already been constructed and this had been inserted into a suitable E. coli strain, and a procedure for the purification of the protein had been established. The purification procedure is outlined below. (More detail and information about construction of the expression vector can be found in the Materials and Methods section)

1 2 3 4 5 Sonication Heating to (NH4)2SO4 DE-52 Red of cells 70 °C precipitation column column

Figure 4. Outline of purification procedure, as established by Karen Duus Sørensen (84)

Figure 5 shows the purification of CTPS. The over expressed CTPS protein is visible in all lanes of the purification gel, also in lane 1, where only the cell debris has been removed. The purified protein is seen in lane 5.

M 1 2 3 4 5 M

97.4 66.2

45.0 Figure 5. SDS-PAGE showing purification of CTPS. M, marker lane; 1, supernatant 31.0 after sonication; 2, supernatant after heat treatment; 3, dialysate after fractionated 21.5 ammonium sulphate precipitation; 4, pooled 14.4 fractions after DE-52 column; 5, pooled fractions after red column; M, marker lane.

21 Crystallization First, a solubility footprint screen was set up to find a useful protein concentration for the crystallization attempts. This showed that a protein concentration of 10 mg/ml was suitable. Next, two commercial screens, Hampton Crystal Screen 1 and 2 were set up. These showed polycrystalline aggregates or microcrystalline precipitate with several of the reagents, and the most conditions that showed the most promising precipitate were used as a starting point for optimization. When good crystals were produced, four crystals were briefly exposed to a cryoprotectant to help prevent ice formation, picked and mounted on cryo-loops and flash cooled in liquid N2 for future data collection.

Identification no. Conditions for crystallization Cryoprotectant 582 4% PEG 8000, no UTP glycerol 612 6% PEG 8000, no UTP glycerol 519 7% PEG 8000, 5 mM UTP glycerol 610 7% PEG 8000, 5 mM UTP ethylene glycol Table 2. Conditions for crystallization. All crystallization conditions also contained 0.1 M sodium acetate at pH 4.5 and 0.1 M magnesium acetate. Cryoprotectant contained 0.1 M sodium acetate (pH 4.5), 5 % PEG 8000, 0.1 M magnesium acetate and 22% glycerol or ethylene glycol.

Diffraction data and statistics for crystal no. 519 Radiation source MAX-lab I911-5 (in Lund, Sweden) Temperature (K) 100 Space group P1 (triclinic) Cell dimensions a, b, c (Å) 43.4, 74.8, 98.2 Cell dimensions α, β, γ (°) 75.8, 84.6, 75.3 No. of measured reflections 54660 No. of unique reflections 46226 Redundancy 1.2 Resolution limits (Å) 20 – 2.4 (2.49-2.4) b a Rlinear (%) 6.7 (27.4) Completeness (%) 86.2 (44.5)a I/σ(I)c 12.3 (2.5)a Table 3. Diffraction data and statistics. aThe values in parentheses are for the highest resolution shell. b Rlinear = ∑ I − I ∑ I , where the sums are over all symmetry related reflections of intensity I, I = observed intensity and I = average intensity for multiple measurements. In all sums single measurements are excluded. cI/σ(I) = the average intensity of all reflections / the average error of all reflections.

A dataset was recorded with on of the crystals, and the crystal was found to belong to space group P1 (triclinic). This means that the unit cell consists of only one asymmetric unit, since

22 there is no crystallographic symmetry. The volume of the unit cell was found to be 295·103Å3 by equation 1. The number of monomers in the unit cell can be estimated by calculating the

Matthews volume, VM, and the approximate water content of the crystal, Vsolvent. This was done for 1, 2, 3 and 4 as possible number of subunits per asymmetric unit, using equation 2 and 3. 3 3 On average proteins have a VM of 2.4 Å /Da, ranging from about 1.9 to 4.2 Å /Da and a water content of approximately 43%, ranging from about 27% to 65% (79). Assuming 3 two subunits in the asymmetric unit, VM is 2.48 Å /Da and the water content is 50%, which are both very typical numbers for protein crystals. If the asymmetric unit contained three subunits, the values would be just on the lower limit of normal and one subunit would give values slightly above the upper limit of normal. It is not possible that the asymmetric unit contains four subunits, so the enzyme has not crystallized as a tetramer. The crystal structure has later been solved (see Appendix), and the crystal did indeed contain two monomers in the unit cell.

Figure 6. Crystals of CTPS. Crystals appeared overnight and grew up to a size of 400 m.

Size analysis by sucrose gradient sedimentation The aggregation state of CTPS in the presence and absence of nucleotides was analyzed by sucrose gradient sedimentation. Four gradients were made, two without nucleotides and two with 1.0 mM of ATP, GTP and UTP. In the nucleotide-free gradients, the CTPS came out as one peak, as can be seen in figure 7A, with a calculated molecular weight of approximately 100 kDa. Figure 7B shows a gradient with nucleotides, and here CTPS came out as two

23 peaks, or a peak with a shoulder. These two peaks represent molecular weights of approximately 105 kDa and 205 kDa.

0.5 0.5 A B 0.4 0.4

0.3 0.3

0.2 0.2

0.1 0.1 activity (arbitrary units) 0.0 activity (arbitrary units) 0.0 0 10 20 30 40 0 10 20 30 40 fraction number fraction number

Figure 7. Sedimentation of CTPS and marker enzymes in sucrose gradients. (○) is catalase, () is alcohol dehydrogenase and (▼) is CTPS. (A) Sucrose gradient without nucleotides. (B) Sucrose gradient with 1 mM ATP, 1 mM UTP and 1 mM GTP.

Trypsin limited proteolysis CTPS was digested with trypsin to see how different ligands would protect it from being cut, and with the hope to also be able to determine at what positions the enzyme was protected by these ligands. A total proteolysis would give quite a lot of fragments, but not all of the potential cleavage sites are immediately accessible because of the tertiary/quaternary structure of CTPS.

MPNKYIVVTG GVLSSVGKGT LVASIGMLLK RRGYNVTAVK IDPYINVDAG TMNPYMHGEV FVTEDGAETD LDLGHYERFM DVNMTKYNNI TAGKVYFEVI KKEREGKYLG QTVQIIPHVT DQIKDMIRYA SKINNAEITL VEIGGTVGDI ESLPFLEAVR QLKLEEGEDN VIFVHIALVE YLSVTGELKT KPLQHSVQEL RRIGIQPDFI VGRATLPLDD ETRRKIALFT NVKVDHIVSS YDVETSYEVP IILESQKLVS KILSRLKLED RQVDLTDWIS FVNNIKGINS KKTINIALVG KYTKLKDSYI SIKEAIYHAS AYIGVRPKLI WIESTDLESD TKNLNEILGN VNGIIVLPGF GSRGAEGKIK AIKYAREHNI PFLGICFGFQ LSIVEFARDV LGLSEANSTE INPNTKDPVI TLLDEQKNVT QLGGTMRLGA QKIILKEGTI AYQLYGKKVV YERHRHRYEV NPKYVDILED AGLVVSGISE NGLVEIIELP SNKFFVATQA HPEFKSRPTN PSPIYLGFIR AVASL

Figure 8. Potential cleavage sites by trypsin on CTPS. Trypsin cleaves proteins at the C-terminal side of lysine and arginine, except where these are followed by proline.

MgCl2 does not seem to help protect the enzyme from proteolysis This is seen in figure 8 in the lanes with CTP both with and without MgCl2. The level of protection seemed to be as follows; ATP ≥ CTP > UTP ≥ no ligand. Other trypsin digestion experiments, not shown here,

24 suggested that GTP did not offer much protection (less than UTP), and that ATP and UTP together offered the best protection of all. It is seen that nucleotides did protect the enzyme from enzymatic cleavage.

UTP ATP Tr. CTPS No ligand CTP CTP –MgCl2 M 10’ 30’ 60’ 10’ 30’ 60’ M 10’ 30’ 60’ 10’ 30’ 60’ 10’ 30’ 60’

97.4 97.4 66.2 66.2 45.0 45.0

31.0 31.0

21.5 21.5

14.4 14.4

Figure 9. Nucleotides protecting CTPS from enzymatic cleavage by trypsin. CTPS was cut by trypsin for one hour, and samples taken at 10, 30 and 60 minutes.

Glutaminase half-reaction kinetics Effect of GTP, ATP and UTP on glutaminase activity To see how the nucleotides ATP, UTP and GTP affect the reaction rate of the glutaminase half-reaction, glutamine concentration was varied both in the presence and absence of saturating concentrations of these, with ATP and UTP being present or absent together. The results are fit to equation 5 (hyperbolic kinetics). As can be seen in figure 10, the reaction could take place without the nucleotides, but the reaction rate was very low. Maximal reaction rate depended on all three nucleotides being present.

When only GTP was present, Km was lowered 4-fold and kcat was increased 9-fold, compared to the reaction rate without nucleotides. When only ATP and UTP were present Km was unchanged but kcat was increased 14-fold. When GTP, ATP and UTP were all present, kcat was increased almost 40-fold, but Km was not lowered.

25 5 A 0.125 B

4 0.100 -1 -1

3 0.075

2 0.050 Initial rate, s 1 Initial rate, s 0.025

0 0.000 0 2 4 6 8 10 0 2 4 6 8 10 Glutamine, mM Glutamine, mM

Figure 10. Effect of GTP, ATP and UTP on the glutaminase reaction. All concentrations of ATP, UTP and GTP are 10.0 mM or nothing. ○ = GTP, ATP and UTP; □ = GTP; ◊ = ATP and UTP; = no GTP, ATP or UTP. Panel B shows a magnified view of the reaction without nucleotides.

-1 Nucleotides present Kcat (s ) Km (mM) ATP + UTP, GTP 4.72 ± 0.22 1.17 ± 0.18 ATP + UTP 1.77 ± 0.01 1.50 ± 0.02 GTP 1.11 ± 0.27 0.37 ± 0.04 No nucleotides. 0.127 ± 0.002 1.39 ± 0.08 Table 4. Kinetic data for the glutaminae reaction with GTP, ATP and UTP as effectors. Nucleotides are present at 10 mM. The graphs are shown in figure 10.

Effect of varying concentrations of ATP and UTP on glutaminase activity In E. coli and other organisms, CTPS shows cooperativity with ATP and UTP, especially when one of them is unsaturated. In these experiments, both ATP and UTP displayed cooperativity, and all the curves are fit to equation 6 (sigmoid kinetics). When no ATP was present, UTP displayed a sigmoid curve, as can be seen in figure -1 11A, with a Hill coefficient of 3.26 ± 0.30, a kcat of 1.47 ± 0.06 s and a K0.5 value of 5.75 ± 0.21 mM. In the presence of ATP, the curves looked almost identical whether containing 0.2 -1 mM, 1.0 mM or 10.0 mM ATP. They all had the same kcat value, between 2.64 and 2.68 s , but there were too few data points at low UTP concentration to determine K0.5 values, and all the standard deviations were very high. All the curves seemed to be sigmoid, and none of them would fit to a hyperbolic equation. The one containing 10.0 mM ATP could only be fit to equation 6 when the Hill coefficient was fixed. Figure 11B shows that when the concentration of ATP was varied, the maximal velocity depended on the concentration of UTP, and the K0.5 values of ATP were very low

(less than 0.2 mM). Only the kcat values are useful, since there are too few points at low ATP concentrations to determine Hill coefficients and K0.5 accurately.

26 The four kcat values at different UTP concentrations were re-plotted and fit to equation 5 (hyperbolic kinetics) in figure 11C. This gave a Km value for UTP of 0.9 mM.

3.0 3.0 A B 2.5 2.5 -1 -1 2.0 2.0

1.5 1.5

1.0 1.0 Initial rate, s Initial rate, s 0.5 0.5

0.0 0.0 0 2 4 6 8 10 12 0 2 4 6 8 10 UTP, mM ATP, mM

3.0 C Fig. 11. Plots at different concentrations of ATP 2.5 and UTP. Assays are made as single point assays -1 2.0 at 10 minutes reaction time. (A) Varying UTP at 4 different concentrations of ATP. (○) is 0 mM ATP, 1.5 (x) is 0.2 mM ATP, () is 1.0 mM ATP and (□) is 1.0 10.0 mM ATP. (B) Varying ATP at 4 different concentrations of UTP. (○) is 0 mM UTP, (x) is Initial rate, s 0.5 0.2 mM UTP, () is 1.0 mM UTP and (□) is 10.0 mM UTP. (C) kcat values from figure 11B re- 0.0 0 2 4 6 8 10 plotted. kcat = 2.53 ± 0.26, Km = 0.91 ± 0.36. UTP, mM

GTP activation of the glutaminase reaction The activation by GTP of the glutaminase half-reaction was investigated with different concentrations of glutamine, and with or without 1.0 mM ATP and 1.0 mM UTP. Glutamine and glutamate did not separate properly in these assays, so some of the graphs show up to five separate experiments in one fit. All the results are fit to equation 5 (hyperbolic kinetics). Without ATP and UTP, GTP continued to activate the glutaminase reaction in an almost linear fashion, with a K0.5 of 15 mM. That is a 20-fold increase in K0.5 compared to the reaction also containing ATP and UTP. A re-plot of the kcat values with different glutamine concentrations was made to give a hint about the Km value of glutamine, even though it only contained three data points plus the origin. It was fit to equation 5 (hyperbolic kinetics) and gave a Km value of 2.8 ± 0.8.

An activation constant, K0.5, of 0.7 mM for GTP (of 0.7 mM) was also estimated from figure 12C.

27

0.8 2.5 0.7 A B 2.0 -1 0.6 -1

0.5 1.5 0.4 0.3 1.0 Initial rate, s

0.2 Initial rate, s 0.5 0.1 0.0 0.0 0 1 2 3 4 5 0 2 4 6 8 10 GTP, mM GTP, mM

4 1.4 C 1.2 D -1 -1 3 1.0 0.8 2 0.6 0.4 Initial rate, s Initial rate, s 1 0.2 0 0.0 0 1 2 3 4 5 0 2 4 6 8 10 GTP, mM GTP, mM

Fig. 12. Activation by GTP. (A) 0.5 mM glutamine, 1.0 mM ATP and 1.0 mM UTP. K0.5 for GTP is 0.22 mM ± -1 -1 0.07 mM and kcat is 0.76 s ± 0.08 s (B) 1.0 mM glutamine, 1.0 mM ATP and 1.0 mM UTP. K0.5 for GTP is -1 0.66 mM ± 0.19 mM and kcat is 1.9 s ± 0.20 (C) 10.0 mM glutamine, 1.0 mM ATP and 1.0 mM UTP. K0.5 for -1 -1 GTP is 0.7 mM ± 0.2 mM and kcat is 4.13 s ± 0.50 s . (D) 10.0 mM glutamine, no ATP, no UTP. K0.5 for GTP -1 -1 is 15.0 mM ± 6.3 mM and kcat is 2.7 s ± 0.75 s .

5

-1 4

3

2 Initial rate, s 1 Figure 13. Replot of kcat values from figure -1 12A, B and C. kcat is 6.58 ± 0.64 s and Km is 0 2.8± 0.8 mM. 0 2 4 6 8 10 Glutamine, mM

28 CTP synthesis kinetics GTP inhibition of the synthase reaction When the concentration of GTP was varied, it had a different effect on the production of CTP than it had on the glutaminase half-reaction, as can be seen in figure 14. The data were fit to equation 7 (uncompetitive inhibition). GTP increased the reaction rate up to a GTP concentration of 50-100 M. At higher concentrations the reaction rate was inhibited. The maximal reaction rate, at 50-100 M GTP, is 0.9 s-1. The maximum reaction rate is only 3- fold above the rate seen without GTP. If figure 14 is compared with figure 12C, which shows the rate of glutamine production in an assay performed at the same conditions, it is seen that the glutaminase reaction is not coupled to the synthesis of CTP above 50-100 M GTP, since the glutaminase reaction continues to be stimulated by GTP, whereas the CTP synthese activity is initially activated by GTP but then inhibited at higher concentrations.

0.9 0.8

-1 0.7 0.6 0.5 0.4 0.3 Initial rate, s 0.2 0.1 0.0 0 1 2 3 4 5 GTP, mM

Figure 14. Activation and inhibition by GTP of the rate of CTP production. The data are fit to -1 uncompetitive inhibition (equation 7). vmax is 0.864 ± 0.095 s ; Ki is 2.5 ± 1.0 mM and KA is 3.057 ± 1.296 M.

NH4Cl as ammonia source instead of glutamine

CTP synthesis was also measured with NH4Cl as ammonium source instead of glutamine. CTP and UTP did not separate well because of the high ionic strength, which may also have disturbed the reaction itself. Lower concentrations of NH4Cl may have given more information about the reaction rate. Figure 15 shows that the enzyme is able to use exogenous ammonia for CTP synthesis.

29

0.20

-1 0.15

0.10

Figure 15. Reaction with NH4Cl as substrate

Initial rate, s 0.05 instead of glutamine. NH4Cl concentration was from 12.5 – 50 mM. Assay conditions were 0.00 without GTP and glutamine. 0 10 20 30 40 50 NH4Cl, mM

30 DISCUSSION

The results found for CTPS from S. solfataricus were generally found to resemble the ones found for CTPS from E. coli and from other organisms. This should not be surprising, as CTP synthases show a high degree of homology, due to their vital function in the cell. There are some finer details though, and these I will now try to cast some light on.

Crystallization I did some preliminary structure solving of S. solfataricus CTPS, using the T. thermophilus PDB structure 1VCM for molecular replacement to solve the phase problem. The enzyme had crystallized as a dimer, unlike both E. coli and T. thermophilus CTPS, which had crystallized as tetramers. Pernille Harris has recorded a new dataset with the crystals and finished the work with the crystal structure. (See Appendix for manuscript to be submitted.)

Size analysis by sucrose gradient sedimentation In the sucrose gradient experiments, CTPS came out at a molecular weight of approximately 100 kDa in the gradient without nucleotides and weights of 105 kDa and 205 kDa when ATP, UTP and GTP were also present. These results confirm previous findings for CTPS from S. solfataricus (84), where a molecular weight of 100 kDa was found without nucleotides, and a weight of 205 kDa when ATP, UTP and GTP were present. The shoulder/second peak is not reported, but this could be due to a different concentration of CTPS. Similar results have been found with E. coli CTPS (32;34). The E. coli CTPS was found to be a dimer of molecular weight 105 that associated to a tetramer of molecular weight 210 in the presence of ATP and UTP. These weights were confirmed both by sucrose gradient sedimentation and by gel filtration and were found before the actual weight of 60 kDa for the monomer was determined. (12). The straightforward interpretation of the sucrose gradient results is that for some reason CTPS appears lighter in the sucrose gradient than it actually is and, like the E. coli enzyme, CTPS from S. solfataricus is in a dimer state without nucleotides, and that tetramerization is dependent on nucleotide concentration, since with nucleotides added in the sucrose gradient it existed both in a dimer and a tetramer state.

31 This is also the behaviour of S. cerevisiae CTPS (37); it exists in a dimer state without nucleotides, and as an increasing fraction of tetramer at increasing concentrations of ATP and UTP. Calf liver CTPS is also a dimer in the absence of nucleotides, and a tetramer in the presence of nucleotides, and both ATP and UTP are required for complete polymerization (39). Rat CTPS is found in both monomer, dimer and tetramer form. Without nucleotides, it is primarily a dimer, and when UTP and ATP are present, a mixture of dimers and tetramers are seen (43). The E. coli enzyme has been shown to reversibly dissociate to monomers in very dilute solutions and at cold conditions (35). It has also been shown that the enxyme does not solely exist as a dimer in the absence of ATP and UTP; it actually exists in an equilibrium of monomers, dimers and tetramers (36). At enzyme concentrations above 5 M, 40% of the enzyme exists as tetramers. The same is seen with CTPS from T. thermophilus, that is in a protein concentration-dependent equilibrium between monomers, dimers and tetramers in the absence of UTP and ATP, with the tetramer as the major component when the protein solution is 5 mg/ml (10). With this in mind, it could also be imagined that the 205 kDa peak represents a diffusion state between dimer and tetramer, and the 100 kDa peak represents a diffusion state between monomer and dimer.

Trypsin limited proteolysis The trypsin limited proteolysis experiments unfortunately showed more fragments than I had hoped for, and I was not able to determine the points of cleavage. These experiments were inspired by experiments done on the E. coli enzyme (25). Here the authors find four different cleavage fragments when no nucleotides protect the enzyme and only two fragments when ATP, UTP or CTP are present. Glutamine or GTP does not protect the enzyme. Studies on inactivation of CTPS by thiourea dioxide (85) have shown that both ATP, UTP and GTP will protect CTPS from E. coli from being inactivated, with half-maximal protection concentrations between 5-7 mM. When ATP and UTP are present together, the half-maximal protection concentration is less than 1 mM. Since ATP, UTP and GTP have separate binding sites, these results suggest that the mechanism of protection is not just physical protection, and since the nucleotides also show sigmoidal kinetics of protection, it is likely that protection is caused by conformational changes in the protein upon nucleotide binding.

32 In the trypsin digestion experiments I also saw that ATP and UTP together protect better than either of them alone and therefore I assume that most of the protective effect seen with nucleotides was not caused by physical protection but by tetramerization and the following conformational changes in the protein upon nucleotide binding.

Effect of GTP, ATP and UTP on glutaminase activity Table 4 and figure 10 show that without ATP, UTP or GTP the reaction rate of the glutaminase half-reaction was very low. ATP and UTP had the biggest effect on Kcat, but GTP both raised Kcat, and also lowered Km, and the highest reaction rate was seen when all three nucleotides were present. GTP activated the glutaminase reaction whether or not ATP and UTP were present, but it was most obvious when they were not; then GTP increased efficiency (Kcat/Km) 36-fold. When ATP and UTP were also present, there was only a 4-fold increase, which was entirely

Kcat. It is possible that the effect of GTP was somehow masked by the massive amounts of ATP and UTP that were present in the assay. There were not enough Mg2+ ions in the assay when ATP, UTP and GTP were present at the same time, since only 20 mM MgCl2 was present in the assay, but ATP, UTP and GTP were each at a concentration of 10 mM. CTPS needs Mg2+ for the triphosphates, but additionally it must have free Mg2+ ions for full activity (47), and this need was not met in all conditions of this assay. ATP and UTP are not substrates of the glutaminase reaction, but their presence activated the reaction rate. When GTP was not present, ATP and UTP increased kcat, 14-fold with no effect on Km. When GTP was also present, ATP and UTP increased the efficiency 12- fold, but Kcat was 4-fold higher while Km was 3-fold lower. In an investigation of the effect of GTP on the E. coli CTPS (15), Levitzki and Koshland investigate the glutaminase half-reaction by using the non-hydrolysable ATP analogue ADPNP instead of ATP. They find that GTP, without ADPNP and UTP being present, kcat of the reaction is increased 12-fold. When ADPNP and UTP are also present,

GTP increases kcat 10-fold. So the increase in reaction rate caused by GTP is almost the same, whether or not the substrate nucleotides are present. ADPNP and UTP give a 6-fold activation of kcat when GTP is not present and a 5-fold activation in the presence of GTP. When the production of CTP is measured in the presence of ATP and UTP, kcat is increased 7-fold and

33 Km is lowered 6-fold. GTP is found to have no effect on the reaction rate when ammonium instead of glutamine is the substrate. For L. lactis CTPS, the reaction rate of the glutamine-dependent CTP production is increased between 25-fold to 50-fold by GTP, increasing with increasing ATP and UTP concentrations (13). The coupling between the two half-reactions has been studied in more detail for the L. lactis enzyme (33). Using the non-hydrolysable ATP analogue ATP-γS to study the glutaminase half-reaction, Km is found to be the same at all conditions. GTP stimulates kcat 3-fold without UTP and ATP-γS, and 7-fold when they are present. In contrast to the E. coli results (and my results) there is no activation by UTP and ATP-γS alone, but a small activation is seen when GTP is also present. The GTP-activated glutaminase reaction, studied using the ATP analogue, never reaches the same level as the production of CTP, but is about 10 times slower; it is therefore suggested that it is the 4-phosphorylated UTP intermediate that activates the glutaminase reaction together with GTP. Since UTP and ATP- γS do stimulate a little when GTP is also present, some allosteric effects are probably also taking place. The increases of activation rates on the glutaminase half-reaction in S. solfataricus CTPS caused by GTP, ATP and UTP resemble the ones found for E. coli CTPS. I saw an activation of the reaction rate both by GTP and by ATP and UTP, and found that GTP also lowered Km for glutamine.

Effect of varying concentrations of ATP and UTP on glutaminase activity The effect on the glutaminase reaction rate of varying concentrations of ATP together with varying concentrations of UTP was investigated. Figure 10 showed that together they stimulated the reaction rate. From Figure 11 shows that it is the concentration of UTP that determines the level of maximum stimulation, as ATP alone showed almost no activation, consistent with the idea that the ATP-activated UTP is the activator of the reaction. Or possibly, that with CTPS from S. solfataricus, ATP alone cannot induce tetramerization of the enzyme, as is the case with CTPS from S. cerevisiae. The binding of ATP and UTP has been shown to be cooperative in several organisms. I have been unsuccessful in obtaining results from binding experiments. The results seen in Figure 11 are obscured by the fact that ATP and UTP cause the enzyme to tetramerize. ATP and UTP stimulate the activity of the glutaminase reaction, of which they

34 can be seen as allosteric activators, and they are substrates of the CTP production. In these assays 1 mM of GTP was used, which was shown (Figure 14) to result in a production rate of CTP that was slower than the rate of glutamine hydrolyzation. Therefore the increase in reaction rates seen with increasing concentrations of ATP and especially UTP (Figure 11) are not coupled to the production of CTP. ATP and UTP increase the reaction rate of the glutaminase activity, either by tetramerization or more subtle allosteric mechanisms or by an effect of the phosphorylated UTP. CTPS from E. coli show positive cooperativity with ATP and UTP (21). When the other substrates are at near saturation, all kinetics are hyperbolic. When either ATP or UTP are present at non-saturating concentrations, the saturation curve of the other is sigmoid, indicating cooperative substrate binding or interacting substrate sites. Hill coefficients of the reaction are 3.8 for ATP and 3.2 for UTP (22;28). Binding studies of ATP and UTP in the absence of the other show Hill coefficients for ATP and UTP of 1.6 and 2.0, respectivelly (32). Investigating the effect of enzyme concentration on reaction rates of CTP synthesis (35), CTPS was found to display hysteresis, cold lability, variations in specific activity with enzyme concentration, and alteration of these properties by ATP and UTP; these effects are more pronounced when ATP and UTP are non-saturating. This suggests that the association- dissociation behaviour of CTP synthase play a significant role in generating the positive cooperativity seen with ATP and UTP. But the fact that L. lactis CTPS show cooperativity with ATP and UTP even though it does not display this association-dissociation behaviour suggests that allosteric effects are also responsible for the cooperative effects

GTP activation of the glutaminase reaction GTP is an activator of the glutaminase half-reaction at all concentrations. When ATP and

UTP are not present, the reaction rate does not reach the limiting rate, and the K0.5 for GTP is 20 times lower than when ATP and UTP are also present. It seems that ATP and UTP somehow affect the binding of GTP, possibly by tetramerization of the enzyme; residues involved in the activation and coupling by GTP of the enzyme has been identified in both the glutaminase domain (25;86) and in the synthase domain (23) of CTPS. Analysis of the T. thermophilus structure suggest that the GTP binding site is formed when ATP and UTP binding bring these two consensus GTP binding sequences from different subunits of the

35 dimer closer together (10), see figure 16. Dynamic light-scattering experiments indicate that even when in a tetrameric form, caused by high protein concentration, the tetramer changes conformation when ATP and UTP bind, and becomes more compact (10). In the non- compacted structure of T. thermophilus CTPS, the two consensus binding sequences are 15- 25 Å apart, which mean that GTP cannot coordinate with both of them at the same time. Figure 17 shows the S. solfataricus CTPS dimer, with one pair of the consensus sequences highlighted as spheres. The shortest distance between the two sequences is approximately 13

Å. So the high K0.5 value seen in figure 12D, where no ATP or UTP is present, could reflect GTP binding to only the part of the GTP binding site that is situated on the glutaminase domain. A conformational change induced by ATP and UTP binding, similar to the one hypothesized for T. thermophilus CTPS may then reflect the much lower K0.5 value seen for GTP when ATP and UTP are also present.

The Km value for glutamine of 2.8 mM seen in Figure 13 is only a loose estimate, but it seems to be higher than the Km values reported from E. coli; Km values for glutamine of 0.32 mM (45) and 0.24 mM (25) has been reported.

Figure 16. Putative conformational change upon binding of ATP and UTP. A computer model of how the glutaminase domain (shown in brown) is rotated toward the synthase domain (shown in grey) when ATP and UTP are bound. The two consensus sequences for GTP binding are shown in green. Picture is reproduced from (10).

36 Figure 17. Dimer of CTPS from S. solfataricus with two GTP binding consensus sequences shown as spheres. The minimum distance between the two halves of the binding site was measured to be approximately 13 Ångstrøm using SwissPDBViewer. This distance is too great for GTP to coordinate with both sequences. The binding sequence in the glutaminase domain is shown on the blue subunit and the binding sequence in the synthase domain is shown on the green subunit. (Picture created with PyMOL)

GTP inhibition of the synthase reaction. GTP was found to inhibit the production of CTP in S. solfataricus CTPS above a certain -1 concentration. The data was fit to uncompetitive inhibition, and vmax was 0.86 s , KA was 2.5

M and Ki was 3.1 mM. Maximal conversion rate vmax is reached already at about 50-100 M GTP. This is a very different result than the one seen for the glutaminase reaction, investigated under similar conditions (Figure 12C), where GTP continues to activate the reaction rate. There is some inconsistency in the specific activity of the enzyme throughout this work, probably because the enzyme loses activity over time, but generally maximum kcat values of 3-6 s-1 was seen. In figure 14 however, the maximum rate of CTP synthesis is less than 1 s-1. This assay was done in parallel with an identical assay to measure the glutamine conversion, with enzyme taken from the same batch; in this assay the maximal conversion rate was more than 4 s-1. So the low rate of CTP production appears not to be random. This inhibition by GTP of the CTP synthase reaction is also seen in E. coli CTPS (46). Up to a concentration of 0.15 mM, GTP is an activator of the reaction, but at concentrations above that GTP behaves as a negative allosteric effector of CTPS, inhibiting glutamine-dependent CTP formation. GTP does not inhibit the glutaminase activity.

37 These results resemble the ones for S. solfataricus CTPS shown in Figure 14. However, the decoupling I saw seems stronger, as the reaction rate of CTP formation never even came close to the limiting rate of glutamine conversion. Glutamine conversion and production of CTP is originally reported to be coupled in E. coli CTPS, with the same rate of CTP production as the glutamine conversion. It is also found that the when ammonia is used as a substrate instead of glutamine, a higher maximal turnover for CTP formation is seen. This is taken as an indication that it is the glutamine half- reaction that is the rate-limiting step (16). But this is shown not to be the case above a GTP concentration of 0.15 mM for E. coli CTPS, and it seems that it is also not the case for S. solfataricus CTPS above 0.1 mM GTP.

38 REFERENCES References

1. Taylor, S. D., Lunn, F. A., and Bearne, S. L. (2008) Ground state, intermediate, and multivalent nucleotide analogue inhibitors of cytidine 5'-triphosphate synthase, ChemMedChem. 3, 1853-1857.

2. Brock, T. D., Brock, K. M., Belly, R. T., and Weiss, R. L. (1972) Sulfolobus: a new genus of sulfur-oxidizing bacteria living at low pH and high temperature, Arch. Mikrobiol. 84, 54-68.

3. Wolfram Zillig, Karl O.Stetter, Simon Wunderl, Wolfgang Schulz, Harro Priess, and Ingrid Scholz (1980) The Sulfolobus-"Caldariella" group: taxonomy on the basis of the structure of DNA-dependent RNA polymerases., Arch. Microbiol. 125, 259-269.

4. She, Q., Singh, R. K., Confalonieri, F., Zivanovic, Y., Allard, G., Awayez, M. J., C.C., Clausen, I. G., Curtis, B. A., De Moors, A., Erauso, G., Fletcher, C., Gordon, P. M., Heikamp-de Jong, I., Jeffries, A. C., Kozera, C. J., Medina, N., Peng, X., H.P., Redder, P., Schenk, M. E., Theriault, C., Tolstrup, N., Charlebois, R. L., Doolittle, W. F., Duguet, M., Gaasterland, T., Garrett, R. A., Ragan, M. A., Sensen, C. W., and van der, O. J. (2001) The complete genome of the crenarchaeon Sulfolobus solfataricus P2, Proc Natl Acad Sci U S A 98, 7835-7840.

5. Turnbough, C. L., Jr. and Switzer, R. L. (2008) Regulation of pyrimidine biosynthetic gene expression in bacteria: repression without repressors, Microbiol. Mol. Biol. Rev. 72, 266-300, table.

6. Christopher K.Mathews, K.E.van Holde, and Kevin G.Ahern (2000) Biochemistry Benjamin/Cunnings, San Fransisco, CA.

7. LIEBERMAN, I. (1956) Enzymatic amination of uridine triphosphate to cytidine triphosphate, J Biol Chem JID - 2985121R 222, 765-775.

8. Endrizzi, J. A., Kim, H., Anderson, P. M., and Baldwin, E. P. (2004) Crystal structure of Escherichia coli cytidine triphosphate synthetase, a nucleotide-regulated glutamine amidotransferase/ATP-dependent amidoligase fusion protein and homologue of anticancer and antiparasitic drug targets, Biochemistry JID - 0370623 43, 6447-6463.

9. van Kuilenburg, A. B., Meinsma, R., Vreken, P., Waterham, H. R., and van Gennip, A. H. (2000) Identification of a cDNA encoding an isoform of human CTP synthetase, Biochim Biophys Acta 1492, 548-552.

10. Goto, M., Omi, R., Nakagawa, N., Miyahara, I., and Hirotsu, K. (2004) Crystal structures of CTP synthetase reveal ATP, UTP, and glutamine binding sites, Structure (Camb) JID - 101087697 12, 1413-1423.

11. Kursula, P., Flodin, S., Ehn, M., Hammarstrom, M., Schuler, H., Nordlund, P., and Stenmark, P. (2006) Structure of the synthetase domain of human CTP synthetase, a

39 target for anticancer therapy, Acta Crystallogr. Sect. F. Struct. Biol. Cryst. Commun. 62, 613-617.

12. Weng, M., Makaroff, C. A., and Zalkin, H. (1986) Nucleotide sequence of Escherichia coli pyrG encoding CTP synthetase, J Biol Chem JID - 2985121R 261, 5568-5574.

13. Wadskov-Hansen, S. L., Willemoes, M., Martinussen, J., Hammer, K., Neuhard, J., and Larsen, S. (2001) Cloning and verification of the Lactococcus lactis pyrG gene and characterization of the gene product, CTP synthase, J Biol Chem JID - 2985121R 276, 38002-38009.

14. Pappas, A., Park, T. S., and Carman, G. M. (1999) Characterization of a novel dUTP- dependent activity of CTP synthetase from Saccharomyces cerevisiae, Biochemistry JID - 0370623 38, 16671-16677.

15. Levitzki, A. and Koshland, D. E. J. (1972) Role of an allosteric effector. Guanosine triphosphate activation in cytosine triphosphate synthetase, Biochemistry JID - 0370623 11, 241-246.

16. Levitzki, A. and Koshland, D. E. J. (1971) Cytidine triphosphate synthetase. Covalent intermediates and mechanisms of action, Biochemistry JID - 0370623 10, 3365-3371.

17. Lewis, D. A. and Villafranca, J. J. (1989) Investigation of the mechanism of CTP synthetase using rapid quench and isotope partitioning methods, Biochemistry JID - 0370623 28, 8454-8459.

18. von der, S., Anderson, P. M., and Villafranca, J. J. (1985) Mechanistic investigations of Escherichia coli cytidine-5'-triphosphate synthetase. Detection of an intermediate by positional isotope exchange experiments, J Biol Chem JID - 2985121R 260, 14993- 14997.

19. Weng, M. L. and Zalkin, H. (1987) Structural role for a conserved region in the CTP synthetase glutamine amide transfer domain, J Bacteriol JID - 2985120R 169, 3023- 3028.

20. Chrakraborty, K. P. and HURLBERT, R. B. (1961) Role of glutamine in the biosynthesis of cytidine nucleotides in Escherichia coli, Biochim Biophys Acta 47, 607-609.

21. Long, C. W. and Pardee, A. B. (1967) Cytidine triphosphate synthetase of Escherichia coli B. I. Purification and kinetics, J Biol Chem JID - 2985121R 242, 4715-4721.

22. D.E.Koshland, JR. and A.Levitzki (1974) CTP Synthetase and Related Enzymes, in The Enzymes 10 (Boyer, P. D., Ed.) pp 539-558, Academic Press, New York.

23. Iyengar, A. and Bearne, S. L. (2003) Aspartate-107 and leucine-109 facilitate efficient coupling of glutamine hydrolysis to CTP synthesis by Escherichia coli CTP synthase, Biochem J JID - 2984726R 369, 497-507.

40 24. Lunn, F. A. and Bearne, S. L. (2004) Alternative substrates for wild-type and L109A E. coli CTP synthases: kinetic evidence for a constricted ammonia tunnel, Eur J Biochem JID - 0107600 271, 4204-4212.

25. Simard, D., Hewitt, K. A., Lunn, F., Iyengar, A., and Bearne, S. L. (2003) Limited proteolysis of Escherichia coli cytidine 5'-triphosphate synthase. Identification of residues required for CTP formation and GTP-dependent activation of glutamine hydrolysis, Eur J Biochem JID - 0107600 270, 2195-2206.

26. Raushel, F. M., Thoden, J. B., and Holden, H. M. (1999) The amidotransferase family of enzymes: molecular machines for the production and delivery of ammonia, Biochemistry JID - 0370623 38, 7891-7899.

27. Raushel, F. M., Thoden, J. B., and Holden, H. M. (2003) Enzymes with molecular tunnels, Acc Chem Res JID - 0157313 36, 539-548.

28. Levitzki, A. and Koshland, D. E. J. (1969) Negative cooperativity in regulatory enzymes, Proc Natl Acad Sci U S A JID - 7505876 62, 1121-1128.

29. Yang, W. L., McDonough, V. M., Ozier-Kalogeropoulos, O., Adeline, M. T., Flocco, M. T., and Carman, G. M. (1994) Purification and characterization of CTP synthetase, the product of the URA7 gene in Saccharomyces cerevisiae, Biochemistry JID - 0370623 33, 10785-10793.

30. Nadkarni, A. K., McDonough, V. M., Yang, W. L., Stukey, J. E., Ozier- Kalogeropoulos, O., and Carman, G. M. (1995) Differential biochemical regulation of the URA7- and URA8-encoded CTP synthetases from Saccharomyces cerevisiae, J Biol Chem JID - 2985121R 270, 24982-24988.

31. McPartland, R. P. and Weinfeld, H. (1979) Cooperative effects of CTP on calf liver CTP synthetase, J Biol Chem JID - 2985121R 254, 11394-11398.

32. Levitzki, A. and Koshland, D. E. J. (1972) Ligand-induced dimer-to-tetramer transformation in cytosine triphosphate synthetase, Biochemistry JID - 0370623 11, 247-253.

33. Willemoes, M. and Sigurskjold, B. W. (2002) Steady-state kinetics of the glutaminase reaction of CTP synthase from Lactococcus lactis. The role of the allosteric activator GTP incoupling between glutamine hydrolysis and CTP synthesis, Eur J Biochem JID - 0107600 269, 4772-4779.

34. Long, C. W., Levitzki, A., and Koshland, D. E. J. (1970) The subunit structure and subunit interactions of cytidine triphosphate synthetase, J Biol Chem JID - 2985121R 245, 80-87.

35. Anderson, P. M. (1983) CTP synthetase from Escherichia coli: an improved purification procedure and characterization of hysteretic and enzyme concentration effects on kinetic properties, Biochemistry JID - 0370623 22, 3285-3292.

41 36. Robertson, J. G. (1995) Determination of subunit dissociation constants in native and inactivated CTP synthetase by sedimentation equilibrium, Biochemistry JID - 0370623 34, 7533-7541.

37. Pappas, A., Yang, W. L., Park, T. S., and Carman, G. M. (1998) Nucleotide-dependent tetramerization of CTP synthetase from Saccharomyces cerevisiae, J Biol Chem JID - 2985121R 273, 15954-15960.

38. Lunn, F. A., MacDonnell, J. E., and Bearne, S. L. (2008) Structural requirements for the activation of Escherichia coli CTP synthase by the allosteric effector GTP are stringent, but requirements for inhibition are lax, J. Biol. Chem. 283, 2010-2020.

39. McPartland, R. P. and Weinfeld, H. (1976) Cytidine 5'-triphosphate synthetase of calf liver. Size, polymerization, and reaction stoichiometry, J Biol Chem JID - 2985121R 251, 4372-4378.

40. Willemoes, M. and Larsen, S. (2003) Substrate inhibition of Lactococcus lactis cytidine 5'-triphosphate synthase by ammonium chloride is enhanced by salt- dependent tetramer dissociation, Arch Biochem Biophys JID - 0372430 413, 17-22.

41. Endrizzi, J. A., Kim, H., Anderson, P. M., and Baldwin, E. P. (2005) Mechanisms of Product Feedback Regulation and Drug Resistance in Cytidine Triphosphate Synthetases from the Structure of a CTP-Inhibited Complex(,), Biochemistry JID - 0370623 EDAT- 2005/10/12 09:00 MHDA- 2005/10/12 09:00 AID - 10. 1021/bi051282o [doi] PST - ppublish 44, 13491-13499.

42. Ostrander, D. B., O'Brien, D. J., Gorman, J. A., and Carman, G. M. (1998) Effect of CTP synthetase regulation by CTP on phospholipid synthesis in Saccharomyces cerevisiae, J Biol Chem 273, 18992-19001.

43. Thomas, P. E., Lamb, B. J., and Chu, E. H. (1988) Purification of cytidine- triphosphate synthetase from rat liver, and demonstration of monomer, dimer and tetramer, Biochim. Biophys. Acta. 953, 334-344.

44. Willemoes, M., Molgaard, A., Johansson, E., and Martinussen, J. (2005) Lid L11 of the glutamine amidotransferase domain of CTP synthase mediates allosteric GTP activation of glutaminase activity, FEBS J JID - 101229646 272, 856-864.

45. Bearne, S. L., Hekmat, O., and MacDonnell, J. E. (2001) Inhibition of Escherichia coli CTP synthase by glutamate gamma-semialdehyde and the role of the allosteric effector GTP in glutamine hydrolysis, Biochem J JID - 2984726R 356, 223-232.

46. MacDonnell, J. E., Lunn, F. A., and Bearne, S. L. (2004) Inhibition of E. coli CTP synthase by the "positive" allosteric effector GTP, Biochim Biophys Acta JID - 0217513 1699, 213-220.

47. Robertson, J. G. and Villafranca, J. J. (1993) Characterization of metal ion activation and inhibition of CTP synthetase, Biochemistry JID - 0370623 32, 3769-3777.

42 48. Savage, C. R. and Weinfeld, H. (1970) Purification and properties of mammalian liver cytidine triphosphate synthetase, J Biol Chem JID - 2985121R 245, 2529-2535.

49. Park, T. S., Ostrander, D. B., Pappas, A., and Carman, G. M. (1999) Identification of Ser424 as the protein kinase A phosphorylation site in CTP synthetase from Saccharomyces cerevisiae, Biochemistry. 38, 8839-8848.

50. Park, T. S., O'Brien, D. J., and Carman, G. M. (2003) Phosphorylation of CTP synthetase on Ser36, Ser330, Ser354, and Ser454 regulates the levels of CTP and phosphatidylcholine synthesis in Saccharomyces cerevisiae, J. Biol. Chem. 278, 20785-20794.

51. Carman, G. M. and Kersting, M. C. (2004) Phospholipid synthesis in yeast: regulation by phosphorylation, Biochem. Cell Biol. 82, 62-70.

52. Chang, Y. F., Martin, S. S., Baldwin, E. P., and Carman, G. M. (2007) Phosphorylation of human CTP synthetase 1 by protein kinase C: identification of Ser(462) and Thr(455) as major sites of phosphorylation, J. Biol. Chem. 282, 17613- 17622.

53. Choi, M. G. and Carman, G. M. (2007) Phosphorylation of human CTP synthetase 1 by protein kinase A: identification of Thr455 as a major site of phosphorylation, J. Biol. Chem. 282, 5367-5377.

54. Higgins, M. J., Graves, P. R., and Graves, L. M. (2007) Regulation of human cytidine triphosphate synthetase 1 by glycogen synthase kinase 3, J. Biol. Chem. 282, 29493- 29503.

55. Jorgensen, C. M., Hammer, K., and Martinussen, J. (2003) CTP limitation increases expression of CTP synthase in Lactococcus lactis, J Bacteriol JID - 2985120R 185, 6562-6574.

56. Evans, D. R. and Guy, H. I. (2004) Mammalian pyrimidine biosynthesis: fresh insights into an ancient pathway, J Biol Chem JID - 2985121R 279, 33035-33038.

57. Williams, J. C., Kizaki, H., Weber, G., and Morris, H. P. (1978) Increased CTP synthetase activity in cancer cells, Nature. 271, 71-73.

58. Ellims, P. H., Gan, T. E., and Medley, G. (1983) Cytidine triphosphate synthetase activity in lymphoproliferative disorders, Cancer Res. 43, 1432-1435.

59. Yuan, P., Hendriks, E. F., Fernandez, H. R., O'Sullivan, W. J., and Stewart, T. S. (2005) Functional expression of the gene encoding cytidine triphosphate synthetase from Plasmodium falciparum which contains two novel sequences that are potential antimalarial targets, Mol Biochem Parasitol 143, 200-208.

60. Roy, A. C., Lunn, F. A., and Bearne, S. L. (2010) Inhibition of CTP synthase from Escherichia coli by xanthines and uric acids, Bioorg. Med. Chem. Lett. 20, 141-144.

43 61. Hofer, A., Steverding, D., Chabes, A., Brun, R., and Thelander, L. (2001) Trypanosoma brucei CTP synthetase: a target for the treatment of African sleeping sickness, Proc Natl Acad Sci U S A JID - 7505876 98, 6412-6416.

62. Fijolek, A., Hofer, A., and Thelander, L. (2007) Expression, purification, characterization, and in vivo targeting of trypanosome CTP synthetase for treatment of African sleeping sickness, J. Biol. Chem. %20;282, 11858-11865.

63. Dereuddre-Bosquet, N., Roy, B., Routledge, K., Clayette, P., Foucault, G., and Lepoivre, M. (2004) Inhibitors of CTP biosynthesis potentiate the anti-human immunodeficiency virus type 1 activity of 3TC in activated peripheral blood mononuclear cells, Antiviral Res JID - 8109699 61, 67-70.

64. Jimenez, B. M. and O'Sullivan, W. J. (1994) CTP synthetase and enzymes of pyrimidine ribonucleotide metabolism in Giardia intestinalis, Int. J. Parasitol. 24, 713-718.

65. McClarty, G. and Qin, B. (1993) Pyrimidine metabolism by intracellular Chlamydia psittaci, J Bacteriol JID - 2985120R 175, 4652-4661.

66. Tipples, G. and McClarty, G. (1995) Cloning and expression of the Chlamydia trachomatis gene for CTP synthetase, J. Biol. Chem. 270, 7908-7914.

67. Andrei, G. and De, C. E. (1993) Molecular approaches for the treatment of hemorrhagic fever virus infections, Antiviral Res. 22, 45-75.

68. Randerath, K. and Randerath, E. (1967) [40] Thin-layer separation methods for nucleic acid derivatives, in Methods in Enzymology Part A: Nucleic Acids (Lawrence Grossman and Kivie Moldave, Ed.) Volume 12, Part 1 ed., pp 323-347, Academic Press.

69. Deuschle, U., Kammerer, W., Gentz, R., and Bujard, H. (1986) Promoters of Escherichia coli: a hierarchy of in vivo strength indicates alternate structures, EMBO J. 5, 2987-2994.

70. Bonekamp, F., Clemmesen, K., Karlstrom, O., and Jensen, K. F. (1984) Mechanism of UTP-modulated attenuation at the pyrE gene of Escherichia coli: an example of operon polarity control through the coupling of translation to transcription, EMBO J. 3, 2857-2861.

71. Andersen, J. T., Poulsen, P., and Jensen, K. F. (1992) Attenuation in the rph-pyrE operon of Escherichia coli and processing of the dicistronic mRNA, Eur. J. Biochem. 206, 381-390.

72. BERTANI, G. (1951) Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli, J. Bacteriol. 62, 293-300.

73. Clark, D. J. and Maaløe, O. (1967) DNA replication and the division cycle in Escherichia coli, Journal of Molecular Biology 23, 99-112.

44 74. Gill, S. C. and von Hippel, P. H. (1989) Calculation of protein extinction coefficients from amino acid sequence data, Anal Biochem JID - 0370535 182, 319-326.

75. Stura, E. A., Nemerow, G. R., and Wilson, I. A. (1992) Strategies in the crystallization of glycoproteins and protein complexes, Journal of Crystal Growth 122, 273-285.

76. JANCARIK, J. and SUNG-HOU, K. I. (1991) Sparse matrix sampling : a screening method for crystallization of proteins, J. Appl. crystallogr. 24, 409-411.

77. Otwinowski, Z. and Minor, W. (1997) [20] Processing of X-ray diffraction data collected in oscillation mode, in Methods in Enzymology Macromolecular Crystallography Part A (Charles, W. C., Ed.) Volume 276 ed., pp 307-326, Academic Press.

78. Kensal E.van Holde, W.Curtis Johnson, and P.Shing Ho. (1998) Principles of physical biochemistry.

79. Matthews, B. W. (1968) Solvent content of protein crystals, J Mol Biol 33, 491-497.

80. MARTIN, R. G. and AMES, B. N. (1961) A method for determining the sedimentation behavior of enzymes: application to protein mixtures, J Biol Chem JID - 2985121R 236, 1372-1379.

81. Cashel, M., Lazzarini, R. A., and Kalbacher, B. (1969) An improved method for thin- layer chromatography of nucleotide mixtures containing 32P-labelled orthophosphate, J. Chromatogr. 40, 103-109.

82. RANDERATH, K. and RANDERATH, E. (1964) Ion-exchange chromatography of nucleotides on poly-(ethyleneimine)-cellulose thin layers, J Chromatogr JID - 0427043 16, 111-125.

83. Houlberg, U. and Jensen, K. F. (1983) Role of hypoxanthine and guanine in regulation of Salmonella typhimurium pur gene expression, J Bacteriol JID - 2985120R 153, 837-845.

84. Karen Duus Sørensen (2003) Kloning af CTP-syntasegenet fra Sulfolobus solfataricus og forsøg på oprensning og karakterisering af Enzymet., Københavns Universitet, Molekylærbiologisk institut.

85. Robertson, J. G., Sparvero, L. J., and Villafranca, J. J. (1992) Inactivation and covalent modification of CTP synthetase by thiourea dioxide, Protein Sci JID - 9211750 1, 1298-1307.

86. Willemoes, M. (2003) Thr-431 and Arg-433 are part of a conserved sequence motif of the glutamine amidotransferase domain of CTP synthases and are involved in GTP activation of the Lactococcus lactis enzyme, J Biol Chem JID - 2985121R 278, 9407- 9411.

45

46

ABBREVIATIONS

β-ME – β-mercaptoethanol CTP – cytidine 5’-triphosphate CTPS – CTP synthase DON – 6-diazo-5-oxo-norleucine DTT – dithiothreitol GDW – glass distilled water GSK3 – Glycogen Synthase Kinase 3 Hepes – N-2-hydroxyethylpiperaxzine-N’-2-ehanesulfonic acid IPTG – isopropyl-thio-β-D-galactoside PCR – polymerase chain reaction PEG - polyethylene glycol PEI – polyethyleneimine PKA – Protein Kinase A PKC – Protein Kinase C RPM – rotations per minute SDS-PAGE – sodium dodecyl sulfate polyacrylamide gel electrophoresis TLC – thin layer chromatography

47

48 APPENDIX – Manuscript to be submitted

Dimer form of CTP synthase from Sulfolobus solfataricus

Iben Lauritsena, Martin Willemoës,a Kaj Frank Jensena and Pernille Harrisb* a Department of Biology, University of Copenhagen, 2200 Copenhagen N, Denmark, and bDepartment of Chemistry, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark. E- mail: [email protected]

Synopsis Crystal structure of the unliganded dimeric form of CTP synthase from Sulfolobus solfataricus unambiguously establishes the dimer. A comparison with the tetrameric form of the Escherichia coli enzyme suggests rather large intramolecular movements upon tetramerisation.

Abstract CTP synthase is catalyzing the last committed step in de novo pyrimidine nucleotide biosynthesis: Glutamine is hydrolysed in the class I glutamine amidotransferase domain and the nascent ammonia is channelled through the interior of the enzyme to the synthase domain, where it reacts with the intermediate 4-phosphoryl UTP obtained by ATP dependent phosphorylation of UTP. Active CTP synthase is a tetrameric enzyme composed of two dimers. In the presence of the substrate nucleotides, ATP and UTP, the tetramer is favoured and when saturated with nucleotide the tetramer completely dominates as the state of oligomeric state of the enzyme. Furthermore, phosphorylation has been shown to influence the oligomeric state of the enzymes from yeast and human. The crystal structure of the dimeric form of the CTP synthase from S. solfataricus enzyme has been determined to 2.5 Å resolution. A comparison of the dimeric interface with the intermolecular interfaces in the tetrameric structures of T. thermophilus CTP synthase and E. coli CTP synthase shows that the dimeric interfaces are almost identical in the three systems. Residues, which from a structural alignment with the E. coli enzyme must be involved in the tetramerization of S. solfataricus CTP synthase all have large thermal parameters in the dimeric form. Furthermore, it is seen that they undergo substantial movements upon tetramerization. A sequence alignment with CTP synthase from yeast offers no obvious explanation for the importance of these phosphorylation sites in the dimer- tetramer equilibrium.

Keywords: CTP synthase, dimer-tetramer equilibrium, nucleotide metabolism

1. Introduction

CTP synthase is catalyzing the last commited step in de novo pyrimidine nucleotide biosynthesis:

49 ATP + UTP + Glutamine → ADP + Pi + CTP + Glutamate

Glutamine is hydrolysed in the class I glutamine amidotransferase domain and the nascent ammonia is channelled through the interior of the enzyme to the synthase domain (Levitzki & Koshland, 1971,1972; Weeks et al., 2006; Willemoes, 2004; Iyengar & Bearne, 2003; Lunn & Bearne, 2003; Endrizzi et al., 2004). Here ammonia reacts with the intermediate 4-phosphoryl UTP obtained by ATP dependent phosphorylation of UTP (von der Saal et al., 1985; Lewis & Villafranca, 1989; Willemoes & Sigurskold, 2002). In this reaction the enzyme activity is regulated by GTP, an allosteric activator that greatly stimulates the hydrolysis of glutamine (Levitzki & Koshland, 1971; Willemoes et al. 2005; Bearne et al. 2001; Macdonnell et al. 2004; Lunn et al. 2008; Willemoes 2003). Alternatively, the reaction can take place with ammonia obtained from the solution in place of glutamine hydrolysis in which case the reaction proceeds at a similar rate to that of the glutamine dependent reaction in the presence of GTP (Willemoes 2004; Bearne et al. 2001). The product CTP also serves as an allosteric inhibitor; the triphosphate binding site overlaps with that of UTP but the nucleoside moiety of CTP binds in an alternative pocket opposite the binding site for UTP (Endrizzi et al. 2005). Active CTP synthase is a tetrameric enzyme. The enzyme tetramer is composed of two dimers that again have been shown to dissociate into monomers at dilute enzyme concentrations (Anderson 1983: Robertson 1995). In the presence of the substrate nucleotides, ATP and UTP, the tetramer is favoured and when saturated with nucleotide the tetramer completely dominates as the state of oligomeric state of the enzyme (Anderson 1983; Pappas et al. 1998; Levtzki & Koshland, 1972). One exception to this observation of an unstable tetramer in the absence of nucleotides is the L. lactis enzyme that remains a tetramer even at dilute enzyme concentrations (Wadskov-Hansen et al.2001) . Whereas E. coli CTP synthase is stabilised in the tetrameric state by increasing ionic strength (Anderson 1983: Robertson 1995) the opposite effect is seen with the L. lactis enzyme (Willemoes & Larsen, 2003). Also, the enzymes from yeast and Human have been shown to be regulated by phosphorylation by protein kinases A and C (Chang & Carman, 2008; Chang et al. 2007). Phosphorylation also influences the oligomeric state of the enzyme, so that treatment with alkaline phosphatase fully dissociates the yeast URA7 tetramer to dimers, even in the presence of ATP and UTP (Pappas et al. 1998). Mutations that affect the oligomeric state and prevents tetramer formation have also been described (Lunn et al. 2008).The glutamine amidotransferase domain of the L. lactis enzyme is fully active in the dimeric form of the enzyme (Levitzki & Koshland, 1971; Willemoes & Larsen, 2003) and from the structure of the E. coli enzyme it is also evident that the tetramerisation mainly affects the synthase domain and the composite formation of active sites with the contribution from three subunits (Endrizzi et al. 2004, 2005).

50 2. Materials and methods

2.1. Protein synthesis and purification

The reading frame of the S. solfataricus pyrG was obtained by PCR with chromosomal DNA from S. solfataricus P2 as a template (a gift from Dr Q. She, Department of Biology, University of Copenhagen) and using the oligonucleotides sspyrg1: CCGGATCCAGGAGAGAACATAATGCCAAACAAGTACATAGTCGTTACAGG and sspyrg2: CGACGTCGACTTAAAGACTAGCAACAGCTCTAATGAACCC as primers and containing a synthetic start codon and Shine-Dalgarno sequence indicated by underlining or bold lettering, respectively. By use of the restriction endonuclease sites Bam HI and Sal I incorporated into the primers as indicated in italics in the above sequences, an expression vector was constructed by cloning this pyrG fragment into the E. coli DNA vector pUHE23-2 (Deuschle et al. 1986) resulting in the plasmid pKDS4. CTP synthase was synthesised by growing E. coli strain NF1830 (Andersen et al.1992) transformed with pKDS4 in NYZ medium with apicillin (100 µg/ml) with vigorous shaking at 37°C to an OD436 of 0.5 when IPTG (0.5 mM) was added. Growth of the cell culture was continued over night and subsequently cells were harvested by centrifugation. Approximately 30 gram of cells were resuspended in 100 mL of 100 mM Tris-HCl, pH 7.6 and, 2 mM EDTA and subjected to sonication. Then mercaptoethanol was added to a final concentration of 2.0 mM and the cell debris was removed by centrifugation. The cell extract was heated in a water bath to 70°C for 10 minutes under constant stirring and then cleared by centrifugation. Next, (NH4)2SO4 was added to 30% saturation and the protein precipitate was removed by centrifugation. The solution was then made 60% saturated and the precipitate collected by centrifugation. The pellet was resuspended in Buffer A (25 mM Tris-HCl, pH7.6, 0.1 mM EDTA and 2 mM mercaptoethanol) and dialysed overnight against the same buffer. The dialysed protein was applied to a 60 mL DE-52 column (DEAE-cellulose, Whatman) and washed with 50 mL Buffer A. Elution of protein from the column was performed by a linear gradient (100 mL) obtained by mixing with Buffer B (25 mM Tris-HCl, pH7.6, 0.1 mM EDTA, 2 mM mercaptoethanol and 0.33 M NaCl). S. solfataricus CTP synthase eluted early from the column and the relevant fractions were pooled and dialysed for one hour against Buffer A. Finally, the dialysed protein was applied to a 30 mL Dyematrex Gel Red A column (Millipore) and eluted by a 100 mL gradient starting from Buffer A and ending with Buffer C (25 mM Tris-HCl, pH7.6, 0.1 mM EDTA, 2 mM mercaptoethanol and 1 M NaCl). The protein eluted as a single peak containing S. solfataricus CTP synthase. The relevant fractions were pooled and concentrated on a Centriprep (Millipore) to a final volume of 11 mL. Finally, the protein was dialysed

51 overnight against 50% glycerol in Buffer A. The enzyme was stored at -20°C at a concentration of approximately 7 mg/mL.

2.2. Analysis of the oligomeric state of S. solfataricus CTP synthase

Size determination of S. solfataricus CTP synthase by sedimentation in ultracentrifuge experiments was performed as previously described (Jensen & Mygind, 1996). Briefly, 5-20% sucrose gradients were made in 50 mM Tris-HCl, pH 8.0 containing 10.0 mM MgCl2 and 2.0 mM DTT. When indicated the nucleotides UTP, ATP and GTP were present at a concentration of 1.0 mM each. The S. solfataricus CTP synthase and the marker enzymes bovine liver catalase (250 KDa) and yeast alcohol dehydrogenase (150 KDa) were identified in the fractions by enzymatic activity.

2.3. Crystallization

The crystallization experiments were made with a protein solution containing approximately 10 mg/mL protein in 25 mM tris pH 7.6, 2 mM mercaptoethanol and 0.1 mM EDTA. A solubility footprint screen (Stura, Nemerow & Wilson, 1992) and Hampton Research crystal screens I and II (Jancarik & Kim, 1991) were setup using hanging-drop vapour diffusion in VDX plates. 2+2 µL drops were equilibrated against 500 µL reservoir solution. Conditions 18 and 46 from Hampton Research crystal screen I gave bunch of needles. These were optimized to give the largest crystals from drops made up from 2 µL 6% PEG8000, 0.1 M magnesium acetate and 0.1 M sodium acetate buffer pH 4.5 and 2 µL protein solution at a concentration of 10 mg/mL.

2.4. Data collection and processing

After having tested several cryo protectants, it turned out that we got the best data set from a single crystal, which was mounted using a litho loop and flash-cooled directly in the nitrogen stream. Diffraction data were collected to be as complete as possible (360◦). Integration and scaling of the data were performed using XDS and XSCALE (Kabsch, 1993). Data collection statistics are presented in Table 1

2.5. Structure solution and refinement

Molecular replacement was performed using MOLREP (Vagin & Teplyakov, 1997; Vaguine et al. 1999). The search unit was 1 subunit from Thermus thermophilus CTP synthase, which shares 54% 2- sequence identity with S. solfataricus CTP synthase, where the structure with SO4 bound gave the best solution (pdb-entry 1VCN). Two molecules were found in the asymmetric unit giving an R-factor

52 of 0.511 and a score of 0.599. Rigid body refinement was performed using Phenix.refine (Afonine et al., 2005). Changes in the sequence were performed automatically using Coot (Emsley & Cowtan, 2004) and checked manually. Structure refinement in Phenix.refine with successive rounds of model building in Coot was performed. The final R factor for the structure consisting of a dimer with 8384 protein atoms and 147 water molecules was 0.229 (Rfree=0.295). Validation was performed using the JCSG structure validation server. The Ramachandran plot gave 85.5% in the core areas, 12.7% in allowed areas, 1.3% in the generally allowed areas and 0.2% in the disallowed areas.

3. Results

Like the other characterised CTP synthases mentioned above the oligomeric state of the S. solfataricus enzyme is also a tetramer in the presence of ATP and UTP (Fig. 1(a)). Similarly, the enzyme dissociates to dimers in the absence of substrate nucleotides (Fig. 1(b)). Unlike the E. coli (Endrizzi et al. 2004) and the T. thermophilus (Goto et al. 2004) enzymes that both crystallised as tetramers even in the absence of ATP and UTP, the S. solfataricus enzyme crystallises as a dimer and the crystal structure unambiguously establishes the dimer of CTP synthase shown in Fig. 2. Otherwise, the overall fold of the protein is similar to the fold found in the CTP synthases from E. coli (pdb-entries: 1S1M (Endrizzi et al., 2004) and 2AD5 (Endrizzi et al., 2005)) and T. Thermophilus (pdb-entries 1VCN, 1VCM and 1VCO (Goto et al., 2004)). An analysis of the amino acid residues responsible for the dimer interactions was performed using the web-interface PISA at European Bioinformatics Institute (http://www.ebi.ac.uk/msd- srv/prot_int/pistart.html) (Krissinel & Henrick, 2007). A comparison with the intermolecular interfaces in the tetrameric structures of T. thermophilus CTP synthase (pdb-entry 1vcn) and E. coli CTP synthase (pdb-entry 1s1m) shows that residues which are responsible for the dimerisation are almost identical in the three systems - all are in the N-terminal domain and in highly conserved areas of the sequence (see Fig. 3 for sequence alignment and indication of dimer interface). The only outlier is an extra salt bridge between Glu199 and Arg202 in S. solfataricus structure. The areas in the sequences responsible for the dimer formation are highlighted with a black box in the sequence alignment shown in Fig. 3.

The 2Fobs-Fcalc electron density allowed for the building of the complete backbone structure of S. solfataricus CTP synthase. Some parts of the peptide chain are, however, build with very large thermal parameters, indicating a large flexibility in the chain and some uncertainty in the coordinates. Among these the loop that runs from residues 424 through 435, which has large thermal parameters. It is situated close to the glutamine binding site in pdb-entry 1VCO and the flexibility is probably coupled to the ability to bind to glutamine. The large flexibility could, however, also be due to the lack of crystal contacts in the area.

53 Also the residues from Val11 to Val16, from Val180 to Leu195 have very large thermal parameters. These residues constitute areas that in the E.coli and the T. thermophilus structures are responsible for the dimer-dimer interactions that form the tetramer. We have superposed the tetramer formed by E. coli CTP synthase (pdb-entry 1S1M) with two copies of the S. solfataricus CTP synthase model to form a pseudo-tetramer of the S. solfataricus structure. It may easily be seen in Fig. 4 that near the tetramer interface large differences in backbone coordinates indeed are found between the E. coli enzyme and the S. solfataricus enzyme. In the S. solfataricus pseudo-tetramer the helix from 190-200 is clashing with the similar helix from the other dimer; the loop from 180-190 is clashing with the loop from 11-16 from the other dimer. In the E. coli structure, which indeed is a true tetramer, the loops and helices have moved apart and as a consequence also the backbone residues from 215 to 230 move substantially (as is also indicated in Fig. 4). A sequence aligment (not shown) with CTP synthase from yeast allowed us to identify the approximate locations of the phosphorylation sites that have been identified as important for tetramerization in the yeast enzyme (Park et al., 1999, 2003; Choi et al., 2003). In Fig. 3 these phosphorylation sites are indicated by blue rectangles. As seen, the residues at positions corresponding to the sites of phophorylation are not even close to the tetramerization interface and the alignment in Fig. 3 offers no obvious explanation for the importance of these phosphorylation sites in the dimer- tetramer equilibrium.

4. Conclusion

Unlike the E. coli (Endrizzi et al. 2004) enzyme and the T. thermophilus (Goto et al. 2004) that both crystallised as tetramers even in the absence of ATP and UTP, the S. solfataricus enzyme crystallised as a dimer. This allowed for an analysis of the possible structural changes that take place upon tetramer formation of CTP synthase. It is evident from comparing the structures of S. solfataricus CTP synthase dimer with those of the E. coli tetramer whether unliganded (Endrizzi et al. 2004) or in comlex with CTP and ADP (Endrizzi et al. 2005), that where the two structures of the E. coli tetramers are very similar at the dimer-dimer interface, loop movements and a more "open" structure around the helices that constitute the dimer- dimer interface are seen in the S. solfataricus structure (fig. Y). The latter observation is in agreement with the effect on the oligomeric structure of binding the substrate nucleotides, ATP and UTP, which in this context serve to stabilize the tetramer once the "compact" interface has formed from the more flexible structure of this area in the dimer.

54 Figure 1 Oligomeric structure of the S. solfataricus CTP synthase. The influence of nucleotides on the oligomeric state of CTP synthase. The molecular weight calculated for the dimer and tetramer is 120 KDa and 240 KDa, respectively. Experiments were conducted as described in Materials and Methods. A) sedimentation of CTP synthase and marker enzymes in the presence of nucleotides: (diamonds) catalase activity (250 KDa), (squares) alcohol dehydrogenase activity (150 KDa) and (triangles) CTP synthase activity. B) sedimentation of CTP synthase and marker enzymes in the absence of nucleotides. Symbols are the same as above.

0.5 A B 0.4

0.3

0.2

0.1

units) (arbitrary Activity 0.0 0 10 20 30 40

Fraction number

0.5 B A 0.4

0.3

0.2

0.1

activity(arbitrary units) 0.0 0 10 20 30 40

fraction number

55 Figure 2 Dimeric structure of S. solfataricus CTP synthase. The A molecule is diplayed in green and the B molecule is displayed in blue. The dimeric contact area of the A-molecule is in red. The figure was prepared in Pymol (DeLano, 2002).

56 Figure 3 A sequence alignment of CTP synthases from S. solfataricus, E. coli and T. thermophilus. The alignment rendering was done with ESPript (Gouet et al. (1999)). Black bars indicate dimer interfaces, orange bars indicate dimer-dimer (tetramer) interfaces and blue bars indicate residues that align with residues which in the yeast enzyme are established as phosphorylation sites important for tetramerization (Park et al. 2003). Identical residues in the alignment are marked with red boxes and white characters. Similar residues are in red characters, while blue frame indicates similarity across groups.

57 Figure 4 The tetramer interface as it would appear between two copies of S. solfataricus CTP synthase (red and blue respectively). The structure of E.coli apo-protein (1s1m) is in grey. Clashes are seen between the neighbouring helices running from 190-200 (the helix from the blue subunit is in lime) and the loop running from residue 180 to 190 (emphasized in lime) and the loop from residue 11-16 from the neighbouring subunit (green). Furthermore the helix running from residue 215-230 is emphasized in green. The figure was prepared in Pymol (DeLano, 2002).

58 Table 1 X-ray data and refinement statistics. Values in parentheses are for the outermost resolution shell.

Data collection Beam line I9-11 2 MAXLab Sweden Detector MARResearch CCD Wavelength (Å) 1.0419 Temperature (K) 100 Space group P1 Unit cell parameters (Å, ◦) a = 43.45(4), b = 76.78(5), c = 98.87(7), α = 100.993(9), β = 95.36(3), γ = 108.42(1) Resolution range (Å) 20.00-2.50 (2.60-2.50) No. of observed reflections 152504(15232) No. of unique reflections 39001(3903) Mosaicity (◦) 0.25-0.65 Redundancy 3.9(3.9) Completeness 95.9(86.4) * Rmerge (%) 4.7(28.5) I/σ(I) 19.93(4.65) Refinement Resolution range 20.00-2.50(2.57-2.50) R factor 0.229(0.270)

Rfree 0.295(0.348) No. of subunits in ASU 2 No. of protein non-H atoms 8384 No. of water molecules 147 Average B factors (Å2) Main chain 51.9 Side chain 54.4 solvent 45.5 R.m.s.d. bond lengths (Å) 0.008 R.m.s.d. bond angles (◦) 1.078

* R = I (hkl) − I(hkl /) I (hkl) where I (hkl) is the mean intensity of a set of equivalent sym ∑hkl∑ i i ∑hkl∑ i i reflections.

59 Acknowledgements

We are grateful for the beamtime provided at MAX-Lab (Lund, Sweden) and for the Danish Natural Science Research Council contribution to DANSYNC. We also acknowledge the support by the European Community – Research Infrastructure Action under the FP6 programme “Structuring the European Research Area”.

References Afonine, P.V., Grosse-Kunstleve, R.W. & Adams, P.D. (2005). CCP4 Newsl. 42, contribution 8. Andersen, J. T., Poulsen, P. & Jensen, K. F. (1992). Eur J Biochem 206, 381-390. Anderson, P. M. (1983). Biochemistry 22, 3285-3292. Bearne, S. L., Hekmat, O. & Macdonnell, J. E. (2001). Biochem J 356, 223-232. Chang, Y. F. & Carman, G. M. (2008). Prog Lipid Res 47, 333-339. Chang, Y. F., Martin, S. S., Baldwin, E. P. & Carman, G. M. (2007). J Biol Chem 282, 17613-17622. Choi, M.-G., Park, T. S., Carman, G. M. (2003). J. Biol. Chem. 278, 23610-23616. DeLano, W.L. The PyMOL Molecular Graphics System (2002) DeLano Scientific, San Carlos, CA, USA. http://www.pymol.org. Deuschle, U., Kammerer, W., Gentz, R. & Bujard, H. (1986). Embo J 5, 2987-2994. Gouet, P., Courcelle, E., Stuart, D. I. & metoz, F. (1999). Bioinformatics 15 305-8 Emsley, P. & Cowtan, K. Acta Crystallogr. D60 (2004), p2126-2132. Endrizzi, J. A., Kim, H., Anderson, P. M. & Baldwin, E. P. (2004). Biochemistry 43, 6447-6463. Endrizzi, J. A., Kim, H., Anderson, P. M. & Baldwin, E. P. (2005). Biochemistry 44. 13491-13499. Goto, M., Omi, R., Nakagawa, N., Miyahara, I. & Hirotsu, K. (2004).Structure 12, 1413-1423. Iyengar, A. & Bearne, S. L. (2003). Biochem J 369, 497-507. Jancarik, J. & Kim, S.-H. (1991) J. Appl. Crystallogr. 24 409-411. Jensen, K. F. & Mygind, B. (1996). Eur J Biochem 240, 637-645. Kabsch, W. (1993). J. Appl. Cryst. 26, 795-800. Krissinel, E. & Henrick, K. (2007). J. Mol. Biol. 372, 774-797. Levitzki, A. & Koshland, Jr., D. E. (1971). Biochemistry 10, 3365-3371. Levitzki, A. & Koshland, Jr., D. E. (1972). Biochemistry 11, 241-246. Levitzki, A. & Koshland, Jr., D. E. (1972). Biochemistry 11, 247-253. Lewis, D. A. & Villafranca, J. J. (1989). Biochemistry 28, 8454-8459. Lunn, F.A. & Bearne, S. L. (2004). Eur J Biochem 271, 4204-4212. Lunn, F. A., Macdonnell, J. E. & Bearne, S. L. (2008). J Biol Chem 283, 2010-2020. Lunn, F. A., Macleod, T. J. & Bearne, S. L. (2008). Biochem J 412, 113-121. Macdonnell, J. E., Lunn, F. A. & Bearne, S. L. (2004). Biochim Biophys Acta 1699, 213-220. Pappas, A., Yang, W. L., Park, T. S. & Carman, G. M. (1998). J Biol Chem 273, 15954-15960. Park, T.-S., Ostrander, D. B., Pappas, A. & Carman, G. M. (1999). Biochemistry 38, 8839-8848.

60 Park, T.-S., O'Brien, D.J. & Carman, G.M. (2003). J. Biol. Chem. 278, 20785-20794. Robertson, J. G., (1995). Biochemistry 34, 7533-7541. von der Saal, W., Anderson, P. M., & Villafranca, J. J. (1985), J Biol Chem 260, 14993-14997. Stura, E.A., Nemerow, G.R., Wilson, I.A. (1992). Journal of Crystal Growth 122, 273-285. Vagin, A.A. & Teplyakov, A. (1997). J. Appl. Cryst. 30 1022 Vaguine, A.A., Richelle, J. & Wodak, S.J. (1999) Acta Cryst. D55 191-205. Wadskov-Hansen, S. L., Willemoes, M., Martinussen, J., Hammer, K., Neuhard, J. & Larsen, S. (2001). J. Biol. Chem. 276, 38002-38009 Weeks, A., Lund, L. & Raushel, F. M. (2006). Curr Opin Chem Biol 10 465-472. Willemoes, M. (2003). J Biol Chem 278, 9407-9411. Willemoes, M. (2004). Arch Biochem Biophys 424, 105-111. Willemoes, M. &Larsen, S. (2003). Arch Biochem Biophys 413, 17-22. Willemoes, M., Molgaard, A., Johansson, E. & Martinussen, J. (2005). Febs J 272, 856-864. Willemoes, M. & Sigurskjold, B. W. (2002). Eur J Biochem 269, 4772-4779. [Click here to enter a reference (Style: IUCr references)] [Click here to enter a reference (Style: IUCr references)]

61