Genetic system for analyzing Escherichia coli thymidylate synthase.

Random in vitro mutagenesis of the thyA gene is being used to delineate its regulatory elements as well as the functional domains of its product, thymidylate synthase (EC 2.1.1.45). Streamlined procedures have been developed for the isolation and characterization of the mutants. Positive selection for synthase-deficient thyA Escherichia coli permitted the isolation of 400 mutants, which are being categorized by phenotypic and genetic criteria. An in situ 5-fluorodeoxyuridylate binding assay was devised to rapidly probe the substrate binding domain, whereas facile mapping procedures, based on pBR322- or M13-borne thyA deletion derivatives, were developed to localize mutations. The sequence changes of one amber mutation and another mutation that abolishes catalysis while maintaining substrate binding activity are presented. The orientation of the thyA gene on the E. coli chromosome was established.     

Properties of a defined mutant of Escherichia coli thymidylate synthase

A mutant of Escherichia coli thymidylate synthase (F3-TS), resulting from the replacement of a tyrosine for a cysteine 50 amino acids from the amino-terminal end, has been purified to homogeneity and found to contain less than 0.2% of the activity of the native enzyme (thyA-TS). Although this protein formed a ternary complex with 5-fluoro-2'-deoxyuridine 5'-monophosphate (FdUMP) and 5,10-methylenetetrahydrofolate, like the native enzyme, the extent of complex formation was significantly impaired as determined by equilibrium dialysis and circular dichroism. Thus, unlike the native enzyme, where 2 mol of FdUMP were present in each mole of ternary complex, F3-TS contained less than 1 mol of FdUMP/mol of ternary complex. Similarly, the binding of dUMP by F3-TS was greatly diminished relative to thyA-TS, but its binding as well as that of FdUMP could be improved by the presence of either the folate substrate or a tight binding folate analogue, 10-propargyl-5,8-dideazafolate (PDDF). However, despite the fact that PDDF enhanced the binding of FdUMP and dUMP to F3-TS, the binding of PDDF to the mutant enzyme was also greatly impaired. This contrasts with the native enzyme, which, under the same conditions, bound about 2 mol of PDDF/mol of enzyme in the presence or absence of either FdUMP or dUMP. Circular dichroism analyses with PDDF in the presence of dUMP or FdUMP yielded analogous results, but the effects were less dramatic than those obtained by equilibrium dialysis. Evidence in support of a structural difference between thyA-TS and F3-TS was obtained by demonstrating that the latter protein was 15-fold slower in forming a ternary complex with dUMP and PDDF than the former and that the mutant enzyme was less stable than the native enzyme.     

Expression of human thymidylate synthase in Escherichia coli

A cDNA clone encoding thymidylate synthase (TS) has been isolated from a human T-cell library and modified in the 5'-untranslated region to incorporate several unique cloning sites. The gene has been cloned as a cassette into several Escherichia coli expression vectors which did not provide detectable amounts of the enzyme. A successful approach used a constitutive E. coli expression vector developed for the enzyme from Lactobacillus casei. A 115-base pair 5'-untranslated region from the L. casei TS which contains a ribosomal binding site and other regulatory sequences has been fused to the coding region of the human TS gene to provide a construct that is expressed in E. coli. The level of expression was further enhanced by altering the nucleotide sequence of the first 90 base pairs to accommodate common codon use in E. coli. In our best expression system, catalytically active human TS is expressed to a level that represents about 1.6% of the total soluble protein. The recombinant human TS has been purified and characterized; except for the presence of an amino-terminal blocking group, the enzyme has physical and kinetic properties similar to the enzyme isolated from human cells.     

Efficient synthesis of mouse thymidylate synthase in Escherichia coli

The coding region of the mouse thymidylate synthase (TS)-encoding cDNA (ts) was inserted downstream from the phage T7 promoter and translation initiation signals of the expression vector, pET-3a, and transformed into Escherichia coli BL21(DE3)[pLysS]. When the wild-type (wt) cDNA sequence was used, mouse TS was synthesized in the bacterial cells in response to induction, but the level of expression was low. When the second codon (Leu) was changed from CUG, found in the normal mRNA, to CUU, the level of expression increased 17-fold and TS represented 5-10% of total cell protein. The recombinant enzyme was purified to homogeneity by affinity chromatography. The recombinant TS had the same Mr as the enzyme from cultured mouse fibroblasts. Kinetic studies with the recombinant enzyme showed that the apparent Km values for deoxyuridylate and 5,10-methylenetetrahydrofolate were 10.5 and 22 microM, respectively, which were similar to the values for TS from mouse cell extracts. The mouse ts expression vector will be useful for the large-scale production of the wt enzyme and for the creation and analysis of mutant enzymes by protein engineering techniques.     

Catalytic role of histidine 147 in Escherichia coli thymidylate synthase

Nine mutant thymidylate synthases were isolated that only differed in sequence at position 147. The wild-type enzyme (which had a histidine residue at 147) and mutant enzymes were purified to near homogeneity and their kinetic properties were compared. Although the kcat values for the mutant enzymes were 10-10,000-fold lower than for the wild-type enzyme, the Km values for both 2'-deoxyuridylate and 5,10-methylenetetrahydrofolate were nearly identical for all the enzymes indicating that His-147 is not significantly involved in initial substrate binding. By comparing the wild-type (His-147) to the glycine (Gly-147) enzyme, the side chain of His-147 was estimated to lower the activation energy of the catalytic step by 1.6-2.9 kcal mol-1. In contrast to the wild-type enzyme, the activity of the Gly-147 enzyme decreased when the pH was raised above 7.5. The activity loss coincided with the deprotonation of a residue that had a pKa of 9.46 +/- 0.2 and an enthalpy of ionization (delta Hion) of 12.1 +/- 0.9. These values are consistent with the involvement of a lysine or an arginine residue in the catalytic process. An inspection of the rates of ternary complex formation among enzyme, 5-fluoro-2'-deoxyuridylate, and 5,10-methylenetetrahydrofolate for the mutant enzymes indicated that His-147 is not needed for the proton removal from C-5 of 2'-deoxyuridylate but rather participates in an initial catalytic step and alters the pKa value of a catalytically important lysine or arginine residue.     

Demonstration of a specific Escherichia coli SecY-signal peptide interaction

Protein translocation in Escherichia coli is initiated by the interaction of a preprotein with the membrane translocase composed of a motor protein, SecA ATPase, and a membrane-embedded channel, the SecYEG complex. The extent to which the signal peptide region of the preprotein plays a role in SecYEG interactions is unclear, in part because studies in this area typically employ the entire preprotein. Using a synthetic signal peptide harboring a photoaffinity label in its hydrophobic core, we examined this interaction with SecYEG in a detergent micellar environment. The signal peptide was found to specifically bind SecY in a saturable manner and at levels comparable to those that stimulate SecA ATPase activity. Chemical and proteolytic cleavage of cross-linked SecY and analysis of the signal peptide adducts indicate that the binding was primarily to regions of the protein containing transmembrane domains seven and two. The signal peptide-SecY interaction was affected by the presence of SecA and nucleotides in a manner consistent with the transfer of signal peptide to SecY upon nucleotide hydrolysis at SecA.     

Characterization of a cis-acting regulatory element in the protein coding region of thymidylate synthase mRNA

Thymidylate synthase (TS) functions as an RNA-binding protein by interacting with two different sequences on its own mRNA. One site is located in the 5′-upstream region of human TS mRNA while the second site is located within the protein coding region corresponding to nt 434–634. In this paper, a 70 nt RNA sequence, corresponding to nt 480–550, was identified that binds TS protein with an affinity similar to that of full-length TS mRNA and TS434–634 RNA. In vitro translation studies confirmed that this sequence is critical for the translational autoregulatory effects of TS. To document in vivo biological significance, TS sequences contained within this region were cloned onto the 5′-end of a luciferase reporter plasmid and transient transfection experiments were performed using H630 human colon cancer cells. In cells transfected with p644/TS434–634 or p644/TS480–550, luciferase activity was decreased 2.5-fold when compared to cells transfected with p644 plasmid alone. Luciferase mRNA levels were identical for each of these conditions as determined by RNase protection and RT–PCR analysis. Immunoprecipitation of TS ribonucleoprotein complexes revealed a direct interaction between TS protein and TS480–550 RNA in transfected H630 cells. Treatment with 5-fluorouridine resulted in a nearly 2-fold increase in luciferase activity only in cells transfected with p644/TS434–634 and p644/TS480–550. This study identifies a 70 nt TS response element in the protein coding region of TS mRNA with in vitro and in vivo translational regulatory activity.     

Efficient and seamless DNA recombineering using a thymidylate synthase A selection system in Escherichia coli

λ-Red system-based recombinogenic engineering is a powerful new method to engineer DNA without the need for restriction enzymes or ligases. Here, we report the use of a single selectable marker to enhance the usefulness of this approach. The strategy is to utilize the thymidylate synthase A (thyA) gene, which encodes an enzyme involved in the synthesis of thymidine 5′-triphosphate, for both positive and negative selection. With this approach, we successfully created point mutations in plasmid and bacterial artificial chromosome (BAC) DNA containing the mouse Col10a1 gene. The results showed that the thyA selection system is highly efficient and accurate, giving an average of >90% selection efficiency. This selection system produces DNA that is free from permanent integration of unwanted sequences, thus allowing unlimited rounds of modifications if required.     

Vibrationally enhanced hydrogen tunneling in the Escherichia coli thymidylate synthase catalyzed reaction

The enzyme thymidylate synthase (TS) catalyzes a complex reaction that involves forming and breaking at least six covalent bonds. The physical nature of the hydride transfer step in this complex reaction cascade has been studied by means of isotope effects and their temperature dependence. Competitive kinetic isotope effects (KIEs) on the second-order rate constant (V/K) were measured over a temperature range of 5-45 degrees C. The observed H/T ((T)V/K(H)) and D/T ((T)V/K(D)) KIEs were used to calculate the intrinsic KIEs throughout the temperature range. The Swain-Schaad relationships between the H/T and D/T V/K KIEs revealed that the hydride transfer step is the rate-determining step at the physiological temperature of Escherichia coli (20-30 degrees C) but is only partly rate-determining at elevated and reduced temperatures. H/D KIE on the first-order rate constant k(cat) ((D)k = 3.72) has been previously reported [Spencer et al. (1997) Biochemistry 36, 4212-4222]. Additionally, the Swain-Schaad relationships between that (D)k and the V/K KIEs reported here suggested that at 20 degrees C the hydride transfer step is the rate-determining step for both rate constants. Intrinsic KIEs were calculated here and were found to be virtually temperature independent (DeltaE(a) = 0 within experimental error). The isotope effects on the preexponential Arrhenius factors for the intrinsic KIEs were A(H)/A(T) = 6.8 +/- 2.8 and A(D)/A(T) = 1.9 +/- 0.25. Both effects are significantly above the semiclassical (no-tunneling) predicted values and indicate a contribution of quantum mechanical tunneling to this hydride transfer reaction. Tunneling correction to transition state theory would predict that these isotope effects on activation parameters result from no energy of activation for all isotopes. Yet, initial velocity measurements over the same temperature range indicate cofactor inhibition and result in significant activation energy on k(cat) (4.0 +/- 0.1 kcal/mol). Taken together, the temperature-independent KIEs, the large isotope effects on the preexponential Arrhenius factors, and a significant energy of activation all suggest vibrationally enhanced hydride tunneling in the TS-catalyzed reaction.     

Characterization of the interaction between DNA gyrase inhibitor and DNA gyrase of Escherichia coli.

Escherichia coli DNA gyrase is comprised of two subunits, GyrA and GyrB. Previous studies have shown that GyrI, a regulatory factor of DNA gyrase activity, inhibits the supercoiling activity of DNA gyrase and that both overexpression and antisense expression of the gyrI gene suppress cell proliferation. Here we have analyzed the interaction of GyrI with DNA gyrase using two approaches. First, immunoprecipitation experiments revealed that GyrI interacts preferentially with the holoenzyme in an ATP-independent manner, although a weak interaction was also detected between GyrI and the individual GyrA and GyrB subunits. Second, surface plasmon resonance experiments indicated that GyrI binds to the gyrase holoenzyme with higher affinity than to either the GyrA or GyrB subunit alone. Unlike quinolone antibiotics, GyrI was not effective in stabilizing the cleavable complex consisting of gyrase and DNA. Further, we identified an 8-residue synthetic peptide, corresponding to amino acids (89)ITGGQYAV(96) of GyrI, which inhibits gyrase activity in an in vitro supercoiling assay. Surface plasmon resonance analysis of the ITGGQYAV-containing peptide-gyrase interaction indicated a high association constant for this interaction. These results suggest that amino acids 89--96 of GyrI are essential for its interaction with, and inhibition of, DNA gyrase.     

Characterization of Escherichia coli B glycogen synthase enzymatic reactions and products.

Previous reports implicate UDPglucose as an active glucosyl donor for the unprimed reaction and “glucoprotein” formation in glycogen biosynthesis in Escherichia coli. Results presented here indicate that UDPglucose and GDPglucose are glucosyl donors in the primed and unprimed reactions catalyzed by purified E. coli B glycogen synthase at less than 5% the rate observed when ADPglucose is the donor. The unprimed reaction is stimulated by 0.25 m citrate and a high molecular weight product is formed similar to that produced when ADPglucose is the glucosyl donor. Physiological amounts of branching enzyme and high concentrations of glycogen inhibit transfer from UDPglucose and GDPglucose. In addition, transfer from UDPglucose is inhibited by ADPglucose. These results strongly suggest that ADPglucose is the physiological donor in both the primed and unprimed reactions. Furthermore, these and previously reported results suggest that one enzyme is involved in the catalysis of the primed, unprimed, and TCA-insoluble product formation reactions. Antiserum prepared against purified E. coli B glycogen synthase inactivates transfer of glucose from either ADPglucose or UDPglucose in the above reactions catalyzed by E. coli B crude extracts. Purified E. coli B glycogen synthase preparations contain significant amounts of α-glucan primer. Evidence shows that this glucan is not covalently attached to the enzyme. Results presented show that formation of material insoluble in TCA and previously considered to be due to “glucoprotein” formation, is in fact due to the generation of long chain length glucan molecules intrinsically acid insoluble. The data suggest that previous results purported to be de novo synthesis of glycogen are due to glucan associated with the glycogen synthase and not to formation of a “glucoprotein” intermediate which then acts as primer for further oligosaccharide synthesis.     

Characterization of a galactose specific adhesin of enteroaggregative Escherichia coli

A fimbrial adhesin was identified from an enteroaggregative Escherichia coli strain. The adhesin was purified to 740-fold by sequential chromatography on an affinity matrix and gel filtration column in the FPLC system. The homogeneity of the purified protein was established by analytical isoelectrofocussing (pI 7.25). The native adhesin appeared as a high-molecular-weight aggregative protein as revealed by gel filtration chromatography on Superose 12HR10/30 column. However, in sodium dodecyl sulfate-polyacrylamide gel electrophoresis the molecular weight of the adhesin was found to be 18 kDa and this was further confirmed by gel filtration chromatography on Superose 6HR 10/30 column presence of 6 M guanidine hydrochloride. The N-terminal 15-amino-acid sequence of the adhesin did not show homology with any of the previously reported fimbrial adhesins. The purified adhesin showed adhesion to human erythrocytes in the presence of Ca(2+) (5 mM). The optimum temperature and pH for the hemadhesion activity was found to be 25 degrees C and 6.5, respectively. The inhibition study clearly suggested that the binding site of the adhesin could recognize galactose as the specific sugar. The fluorescence of 4-methylumbelliferyl-alpha-D-galactopyranoside was quenched on binding to the adhesin and maximum reversal of fluorescence quenching was observed by competitive substitution titration with raffinose. The adhesin was found to contain one binding site per monomer for its specific sugar residue. The association constant and the free energy of binding were obtained as 3.98 x 10(5) M(-1) and -31.97 kJ/mol, respectively. The adherence of the bacteria to HEp-2 monolayer was inhibited in presence of galactose and this was further supported by a significant reduction in the bacterial adherence to the HEp-2 cells, pretreated with beta-D-galactosidase.     

Characterization of Escherichia coli ATP synthase beta-subunit mutations using a chromosomal deletion strain.

(1) We constructed Escherichia coli strain JP17 with a deletion in the ATP synthase beta-subunit gene. JP17 is completely deficient in ATP synthase activity and expresses no beta-subunit. Expression of normal beta-subunit from a plasmid restores haploid levels of ATP synthase in membranes. JP17 was shown to be efficacious for studies of beta-subunit mutations. Site-directed mutants were studied directly in JP17. Randomly generated chromosomal mutants were identified by PCR and DNA sequencing, cloned, and expressed in JP17. (2) Eight novel mutations occurring within the putative catalytic nucleotide-binding domain were characterized with respect to their effects on catalysis and structure. The mutations beta C137S, beta G152D, beta G152R, beta E161Q, beta E161R, and beta G251D each impaired catalysis without affecting enzyme assembly or oligomeric structure and are of interest for future studies of catalytic mechanism. The mutations beta D301V and beta D302V, involving strongly conserved carboxyl residues, caused oligomeric instability of F1. However, growth characteristics of these mutants suggested that neither carboxyl side chain is critical for catalysis. (3) The mutations beta R398C and beta R398W rendered ATP synthase resistant to aurovertin, giving strong support to the view that beta R398 is a key residue in the aurovertin-binding site. Neither beta R398C or beta R398W impaired catalysis significantly.     

Analysis of mRNA recognition by human thymidylate synthase

Expression of hTS (human thymidylate synthase), a key enzyme in thymidine biosynthesis, is regulated on the translational level through a feedback mechanism that is rarely found in eukaryotes. At low substrate concentrations, the ligand-free enzyme binds to its own mRNA and stabilizes a hairpin structure that sequesters the start codon. When in complex with dUMP (2′-deoxyuridine-5′-monophosphate) and a THF (tetrahydrofolate) cofactor, the enzyme adopts a conformation that is unable to bind and repress expression of mRNA. Here, we have used a combination of X-ray crystallography, RNA mutagenesis and site-specific cross-linking studies to investigate the molecular recognition of TS mRNA by the hTS enzyme. The interacting mRNA region was narrowed to the start codon and immediately flanking sequences. In the hTS enzyme, a helix–loop–helix domain on the protein surface was identified as the putative RNA-binding site.     

Characterization of the cofactor composition of Escherichia coli biotin synthase

The cofactor content of in vivo, as-isolated, and reconstituted forms of recombinant Escherichia coli biotin synthase (BioB) has been investigated using the combination of UV-visible absorption, resonance Raman, and M?ssbauer spectroscopies along with parallel analytical and activity assays. In contrast to the recent report that E. coli BioB is a pyridoxal phosphate (PLP)-dependent enzyme with intrinsic cysteine desulfurase activity (Ollagnier-deChoudens, S., Mulliez, E., Hewitson, K. S., and Fontecave, M. (2002) Biochemistry 41, 9145-9152), no evidence for PLP binding or for PLP-induced cysteine desulfurase or biotin synthase activity was observed with any of the forms of BioB investigated in this work. The results confirm that BioB contains two distinct Fe-S cluster binding sites. One site accommodates a [2Fe-2S](2+) cluster with partial noncysteinyl ligation that can only be reconstituted in vitro in the presence of O(2). The other site accommodates a [4Fe-4S](2+,+) cluster that binds S-adenosylmethionine (SAM) at a unique Fe site of the [4Fe-4S](2+) cluster and undergoes O(2)-induced degradation via a distinct type of [2Fe-2S](2+) cluster intermediate. In vivo M?ssbauer studies show that recombinant BioB in anaerobically grown cells is expressed exclusively in an inactive form containing only the as-isolated [2Fe-2S](2+) cluster and demonstrate that the [2Fe-2S](2+) cluster is not a consequence of overexpressing the recombinant enzyme under aerobic growth conditions. Overall the results clarify the confusion in the literature concerning the Fe-S cluster composition and the in vitro reconstitution and O(2)-induced cluster transformations that are possible for recombinant BioB. In addition, they provide a firm foundation for assessing cluster transformations that occur during turnover and the catalytic competence of the [2Fe-2S](2+) cluster as the immediate S-donor for biotin biosynthesis.     

The identification of thymidylate synthase peptide domains located in the interface region that bind thymidylate synthase mRNA

Thymidylate synthase (TS) is a critical chemotherapeutic target and intracellular levels of TS are an important determinant of sensitivity to TS inhibitors. Translational autoregulation represents one cellular mechanism for controlling the level of expression of TS. This mechanism involves the binding of TS protein to its own messenger RNA (mRNA), thus, repressing translational efficiency. The presence of excess substrate or inhibitors of TS leads to derepression of protein binding to mRNA, resulting in increased translational efficiency and ultimately increased levels of TS protein. TS protein has been shown to bind to two distinct areas on its mRNA. The goal of the present work is to define the TS domains responsible for this interaction. Using a separate series of overlapping 17-mer peptides spanning the length of both the human and Escherichia coli TS sequences, we have identified six potential domains located in the interface region of the TS protein that bind TS mRNA. The identified domains that bind TS mRNA include three concordant regions in both the human and E. coli peptide series. Five of the six binding peptides contain at least one invariant arginine residue, which has been shown to be critical in other well-defined protein-RNA interactions. These data suggest that the identified highly conserved protein domains, which occur at the homodimeric interface of TS, represent potential participating sites for binding of TS protein to its mRNA.     

Purification and characterization of selenomethionyl thymidylate synthase from Escherichia coli: comparison with the wild-type enzyme

Replacement of methionine (Met) residues by selenomethionine (SeMet) was recently shown to facilitate the crystallographic analysis of protein structure through the application of multi-wavelength anomalous diffraction techniques [Yang et al. (1990) Science (Washington, D.C.) 249, 1398-1405]. The availability of SeMet-containing proteins provides an excellent opportunity to evaluate the effects of the complete replacement of Met by SeMet. We chose to compare the properties of selenomethionyl thymidylate synthase isolated from Escherichia coli DL41 (a methionine auxotroph) and wild-type (wt) enzyme obtained from E. coli Rue10. An improved purification procedure for thymidylate synthase was developed which permitted the isolation of 25 mg of pure protein from 2 g of E. coli in 90% yield in no more than 8 h. The pure wt and SeMet enzymes exhibited specific activities 40% higher than published values. Thermal stability studies at 30 degrees C in degassed buffer showed that the SeMet enzyme (t1/2 67 h) was 8-fold less stable than wt enzyme (t1/2 557 h). The half-lives for the latter enzymes in nondegassed buffers at 30 degrees C were decreased by 2-fold, thus indicating the sensitivity of the enzyme to dissolved oxygen. Both enzymes exhibited essentially the same kinetic and binding properties, including Km(dUMP) (1.2 x 10(-6) M), specificity constant (1.6 x 10(6) s-1 M-1), and Kd for 5-fluorodeoxyuridylate binding (1.2 nM) in covalent inhibitory ternary complexes. In addition, X-ray crystallographic analysis by difference Fourier synthesis showed there was no significant difference in conformation between the SeMet enzyme and the wt enzyme.