INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Thymidine (dT) auxotrophy in Salmonella typhimurium and Escherichia coli can result from either mutation of thyA, the gene for thymidylate synthase, or a lack of availability of dUMP, the substrate for the enzyme. As illustrated in Fig. 1, biosynthesis of dUMP occurs through two distinct pathways (18). The quantitatively more important pathway involves the deamination of dCTP to dUTP by dCTP deaminase, followed by the hydrolysis of dUTP by dUTP nucleotidohydrolase (dUTPase) to yield dUMP, with 75% of endogenous dUMP arising through this route. The second pathway generating the remaining 25% of dUMP consists of the reduction of UDP by ribonucleoside diphosphate reductase to dUDP, which is phosphorylated by nucleoside diphosphokinase to dUTP and subsequently hydrolyzed to dUMP by dUTPase.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   Pyrimidine nucleotide interconversions. The individual enzymes are identified by their gene symbols: dcd, dCTP deaminase (EC 3.5.4.13); dut, dUTPase (EC 3.6.1.23); ndk, nucleoside diphosphate kinase (EC 2.7.4.6); nrdAB, ribonucleoside diphosphate reductase (EC 1.17.4.1); pyrH, UMP kinase (EC 2.7.4.4); tmk, dTMP kinase (EC 2.7.4.9); thyA, thymidylate synthase (EC 2.1.1.45); tdk, thymidine kinase (EC 2.7.1.21); cdd, cytidine (deoxycytidine) deaminase (EC 3.5.4.5); and deoA, thymidine phosphorylase (EC 2.4.2.4). The dotted arrow represents an unknown pathway (see text) for the degradation of dCTP to dC.

Alternatively, dUMP may be produced by pyrimidine salvage through reaction of dU with dT kinase. The dU, in turn, may arise either from dC through deamination catalyzed by cytidine (deoxycytidine) deaminase or by the condensation of uracil and deoxyribose 1-phosphate mediated by thymidine phosphorylase, although this latter reaction is believed to act predominantly in the catabolic direction (18).

Consistent with there being two biosynthetic routes for dUMP production, it was shown that at least two distinct mutations were required to generate a growth requirement for dT (or dU) in S. typhimurium (3). One was identified as a mutation affecting dcd, the gene encoding dCTP deaminase, while the other pertained to a previously uncharacterized locus, termed dum (for dUMP), for which an approximate map position was established. The work described here was undertaken with the aim of characterizing dum, its role in relation to dT (dU) auxotrophy in S. typhimurium, and further characterizing aspects of dUMP and thymine nucleotide biosynthesis.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial strains and plasmid vectors. Bacterial strains used were derivatives of either S. typhimurium LT2 or E. coli K-12 and are listed, along with their genotypes and relevant properties, in Table 1. Plasmid vectors used as cloning vehicles are also presented in Table 1; those constructed in the study are described in the text.

Media and growth conditions. Lennox broth (7) was used as the complex medium. Bochner medium (6) was used for the positive selection of loss of tetracycline resistance (Tc^r). Minimal medium A, described previously (12), contained 0.2% glucose as the carbon source and, when included, Casamino Acids at 0.1%. Media were supplemented with the following (in micrograms per milliliter) as required: individual amino acids, 50; fluorouridine, 10; hypoxanthine, 20; thiamine, 2; dT, 10; uracil, 25; and 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal), 40. When used, antibiotics were added at the following concentrations (microgram per milliliter): ampicillin, 100; chloramphenicol, 20; gentamicin, 15; kanamycin, 60; and tetracycline, 15. Solid media were prepared by addition of agar to 1.5%. Unless otherwise noted, cultures were grown at 37°C. Liquid cultures were incubated on an air shaker operating at 250 rpm, and growth was monitored by measuring cell turbidity with a Klett-Summerson colorimeter.

Genetic techniques. Bacteriophages P22HT105/int-201 (9) and P1CMclr100 (17) were used for transductions with S. typhimurium. In some instances, 10 mM EGTA was added to plating medium to limit lysogeny of transductants. Methods for transposon technology with Tn10 or Tn10dTc were as previously reported (9, 13). Rapid transduction mapping (4) was performed with Kit-22 from the Salmonella Genetic Stock Centre (SGSC). Pulsed-field gel electrophoresis (PFGE) was carried out by S.-L. Liu at the SGSC. Mutants harboring a cdd mutation were selected by resistance to 5-fluorodeoxcytidine (18). Transduction to cdd⁺ was selected by plating to medium containing cytidine (100 µg/ml) as the sole nitrogen source.

DNA techniques. The methods used were primarily adapted from the manual of Sambrook et al. (21). Procedures drawn from this manual include chromosomal and plasmid DNA isolations; restriction endonuclease digestion, modification, and ligation of DNA; transformation; and agarose gel electrophoresis. Fragments from digested plasmid DNA were isolated and purified by using Geneclean III (Bio 101 Inc.) or by electroelution. Transformation of plasmids between E. coli and S. typhimurium was carried out as previously described (1).

Cloning of dcd⁺ and construction of a dcd::Km null mutant. Plasmid pNJK2, containing the dcd gene on an 8.1-kbp fragment, was obtained by transducing a pBR328 library of S. typhimurium DNA (1) to KP1361 and selecting for dU (dT) independence and chloramphenicol resistance (Cm^r). Subcloning of a 1.8-kbp EcoRV-HindIII into pUC19 yielded the dcd⁺ construct pNJK14, which mediated an increase in dCTP deaminase activity of approximately 100-fold relative to the control (Fig. 2). The unique BssHII site of pNJK14 was modified by S1 nuclease digestion to remove the four-nucleotide 5' overhang, producing pNJK17, a construct imparting no increase in dCTP deaminase activity (Fig. 2), thus delimiting the BssHII site to within dcd. (The foregoing manipulations and assays for dCTP deaminase activity were done with E. coli MC1061 as the host because it has a greater efficiency of transformation than any of the relevant S. typhimurium strains.) Since KP1361 was unable to tolerate the presence of a dcd⁺ high-copy-number-plasmid, in vivo-complementation testing was done with low-copy-number plasmids obtained by transferring the EcoRV-HindIII fragments to pWSK29.

View larger version (8K): [in this window] [in a new window]   FIG. 2.   Structures and properties of specific plasmids. The thick solid line for pNJK14 pertains to the cloned S. typhimurium DNA present in the plasmid; the positions of pertinent restriction sites are indicated. For pNJK17, the vertical arrow above the line indicates that the BssHII site was modified; for pNJK23, the diagram illustrates the insertion of the Kmr cassette at the BssHII site. The ability of each of the cloned or modified DNAs to complement the dcd mutation of KP1361 is shown at the right.

Enzyme assays. Cell extracts for enzyme assays were prepared by sonic disruption of cells resuspended in 40 mM potassium phosphate buffer (pH 7.0). Aspartate transcarbamoylase (ATCase) (12) and dCTP deaminase (3) assays were carried out according to published procedures. UMP kinase assay was performed by a modification of the procedure described by Blondin et al. (5). The reaction mixture contained 50 mM Tris-HCl (pH 7.5), 50 mM KCl, 5 mM MgCl₂, 0.5 mM ATP, 1 mM phosphoenolpyruvate, 0.13 mM NADH, 0.5 mM UMP, pyruvate kinase (2 U/ml), lactate dehydrogenase (10 U/ml), and cell extract. For the reference reaction, UMP was omitted, and the two reactions were run in parallel at 37°C. UMP kinase activity was calculated after correction for the background NADH oxidase activity observed in the reference reaction. Protein content was determined by the method of Lowry et al. (16), using bovine serum albumin as a standard. Specific activity was expressed as nanomoles of substrate transformed per minute per milligram of cell extract protein.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Genetic characterization of dum. The dum locus was previously mapped by conjugation to be near purE, located at 12 centisomes (Cs) on the current linkage map of S. typhimurium (22), and was not further characterized. In this study, a Tn10dTc insertion linked to dum was constructed, and rapid transduction mapping (4) based on selection for loss of Tc^r served to approximate the location of dum between 3 and 7 Cs (data not shown). The pan locus is located at 4.5 Cs. P22 and P1 lysates prepared on KRM12 were used to transduce KR1318 to pan⁺ and then score for Tc^r; 2 and 48% cotransduction, respectively, of the two markers was observed (500 transductants from each cross were tested). In parallel, PFGE analysis of KR1638 chromosomal DNA defined the location of the linked Tn10 insertion to be about 375 kbp counterclockwise from XbaI cleavage site 1 (15), placing it clockwise of pan at approximately 5 Cs.

Biochemical characterization of dum. The mapping data led to speculation that dum may be identical to pyrH, the gene encoding UMP kinase and known to be located around 4 to 5 Cs. In pyrH mutants, conversion of UMP to UDP is compromised and the accompanying starvation for pyrimidine nucleotides leads to derepression of all pyr genes of the de novo biosynthetic pathway (12, 23). Accordingly, we determined the levels of activity of ATCase (encoded by pyrBI) and UMP kinase in the original dum-1 mutant (KP1361) and a dum⁺ counterpart (KP1389) (Table 2). ATCase activity in KP1361 was at least 20-fold above that of KP1389, and UMP kinase activity was less than one-third of that of KP1389. Further, when a plasmid (pJES11) harboring the pyrH⁺ gene from E. coli (24) was transferred to KP1361, the resulting transformant no longer required dT or dU for growth (hereafter described as dT prototrophy or dT auxotrophy) and had a level of ATCase activity comparable to that of the KP1389 control (Table 2). To exclude the possibility of multicopy suppression (29), the pyrH⁺ DNA from pJES11 was subcloned as an EcoRI-HindIII fragment to pFZY1. KP1361 transformants harboring this pyrH⁺ low-copy-number derivative were also dT prototrophs but grew more slowly than transformants containing pJES11.

Effect of dcd::Km and on the production of thymine nucleotides. We did not know the nature of the dcd-1 mutation affecting dCTP deaminase activity in the primary isolate, KP1361, nor did we know whether the mutagenesis procedure had introduced additional mutations in the region. Therefore, we constructed a strain harboring a dcd::Km null mutation in order to critically assess the contribution of dCTP deaminase in dUMP biosynthesis.

Contribution of the cdd and dcd::Km mutations. JL1269dcd::Km contained only the triad of mutations dcd::Km, pyrH, and cdd, and its dT requirement was complemented by a plasmid containing S. typhimurium dcd⁺ (Table 3). JL1269dcd::Km was transduced cdd⁺ in the presence of dT, and the resulting transductants were dT prototrophs, albeit slow growing, and addition of dT in the medium stimulated growth. Similarly, the cdd⁺ transductants of KP1725dcd::Km and KP1361dcd::Km were also dT independent and exhibited increased growth in the presence of dT.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Previous work (3, 8) had shown that to establish dT auxotrophy (defined as an obligate requirement of a source for thymine nucleotide synthesis to prevent thymineless death) by mutations other than thyA, at least two distinct mutations were required to sufficiently limit synthesis of dUMP, the immediate precursor of dTMP. Such mutants can have their auxotrophy met by either dU or dT. In the pioneering work of Fuchs and Neuhard (8), a dcd cdd parental E. coli strain was used in the selections and the mutant obtained harbored a mutation affecting ribonucleoside diphosphate reductase (nrdAB), thus impairing the synthesis of dUMP through the combined effect of blocking the conversion of dCTP to dUTP (dcd) and UDP to dUDP (nrd). Applying a similar rationale, Beck et al. (3) isolated mutants of S. typhimurium defective in dUMP synthesis; as predicted, one of the mutations pertained to dcd, but the other, dum-1, remained an enigma, although it was shown that it was not an nrd mutation.

The accumulated evidence presented here supports the conclusion that the gene originally referred to as dum is pyrH. Identification of dum as pyrH was catalyzed by the realization that dum had been imprecisely mapped in the original study, as both rapid transduction mapping and PFGE confirmed the location of dum to be between 4 and 5 Cs on the current linkage map, not at 12 Cs as previously defined (3). Reconstruction experiments involving the transduction of pyrH⁺ and known pyrH alleles were fully consistent with the interpretation that dum-1 was a pyrH allele.

UMP is the precursor for the synthesis of all pyrimidine nucleotides, and UMP kinase (pyrH) is an essential enzyme catalyzing the conversion of UMP to UDP. Consequently, pyrH mutants have decreased pyrimidine ribonucleoside di- and triphosphate pools, resulting in elevated levels of the enzymes of the pyrimidine biosynthetic pathway (12, 23). KP1361 (dum-1) exhibited the characteristic increased level of ATCase activity observed for pyrH mutants, whereas its dum⁺ counterpart, KP1389, did not (Table 2), showing that this elevation was due to the dum-1 allele. Further, the introduction of plasmid-borne pyrH⁺ (pJES11) to KP1361 complemented the dum-1 mutation. Thus, in a dcd genetic background, where the major route for dUMP synthesis is blocked, the presence of a pyrH mutation has the effect of lowering the UDP pool, thereby compromising the second pathway for dUMP synthesis.

Mutants of S. typhimurium (3) and E. coli (19) defective in dCTP deaminase were previously shown to contain lowered dTTP pools and 10- to 20-fold-elevated dCTP pools which might, in turn, provide a source of dC. In the presence of dC deaminase, this could lead to production of dU, possibly precluding the isolation of second pathway dT auxotrophs. As mentioned earlier, this aspect had been taken into consideration in past studies, but these investigations did not address whether cdd⁺ versus the cdd mutant state was indeed of significance. Here, we showed that in a dcd::Km genetic background, the presence of a cdd mutation further compromised dUMP synthesis, as illustrated by the growth rate data presented in Results and the fact that a cdd pyrH dcd mutant is auxotrophic for dT, whereas its isogenic cdd⁺ counterpart is not. Thus, in dcd::Km pyrH double mutants, some level of dC must arise for conversion to dU. These results parallel the observations of Weiss and Wang (28) in their study on the effects of a dcd null mutation in E. coli; the mutation created a condition of dT bradytrophy, and the condition was more severe in cdd mutants than in cdd⁺ strains.

Based on the available data, we conclude that in thyA⁺ strains, the presence of three mutations, dcd, cdd, and pyrH, creates a condition of dT auxotrophy due to an inadequate level of dUMP biosynthesis. In wild-type cells, the two pathways, dCTP to dUTP and UDP to dUDP, jointly supply dUTP for production of dUMP (Fig. 1). Inactivation of dcd results in a marked increase in the dCTP pool, a stress for thymine nucleotides, and derepression of nrd (encoding ribonucleoside diphosphate reductase [25]) in an attempt to compensate. The elevated pool of dCTP can be metabolized to dC (by an unknown route), and the dC, in turn, can provide dU to be reutilized for the synthesis of dUMP, thereby providing a third pathway for dUMP biosynthesis, a pathway operating only in conditions of expanded dCTP pools. Thus, in dcd cdd backgrounds, any situation interfering with the production of intracellular dUDP will lead to a diminished supply of dUMP. The observation that impairment of UMP kinase or ribonucleoside diphosphate reductase activity in such backgrounds leads to dT auxotrophy is fully consistent with this scenario and raises the possibility that simple interference of de novo pyrimidine biosynthesis may have the same effect.