1-Methylguanosine-Deficient tRNA of Salmonella enterica Serovar Typhimurium Affects Thiamine Metabolism
Department of Molecular Biology, Umeå University, S-90 187 Umeå, Swedeno|@q&, 百拇医药
Received 2 July 2002/ Accepted 1 November 2002o|@q&, 百拇医药
ABSTRACTo|@q&, 百拇医药
In Salmonella enterica serovar Typhimurium a mutation in the purF gene encoding the first enzyme in the purine pathway blocks, besides the synthesis of purine, the synthesis of thiamine when glucose is used as the carbon source. On carbon sources other than glucose, a purF mutant does not require thiamine, since the alternative pyrimidine biosynthetic (APB) pathway is activated. This pathway feeds into the purine pathway just after the PurF biosynthetic step and upstream of the intermediate 4-aminoimidazolribotide, which is the common intermediate in purine and thiamine synthesis. The activity of this pathway is also influenced by externally added pantothenate. tRNAs from S. enterica specific for leucine, proline, and arginine contain 1-methylguanosine (m1G37) adjacent to and 3' of the anticodon (position 37). The formation of m1G37 is catalyzed by the enzyme tRNA(m1G37)methyltransferase, which is encoded by the trmD gene. Mutations in this gene, which result in an m1G37 deficiency in the tRNA, in a purF mutant mediate PurF-independent thiamine synthesis. This phenotype is specifically dependent on the m1G37 deficiency, since several other mutations which also affect translation fidelity and induce slow growth did not cause PurF-independent thiamine synthesis. Some antibiotics that are known to reduce the efficiency of translation also induce PurF-independent thiamine synthesis. We suggest that a slow decoding event at a codon(s) read by a tRNA(s) normally containing m1G37 is responsible for the PurF-independent thiamine synthesis and that this event causes a changed flux in the APB pathway.
INTRODUCTIONj3z0, http://www.100md.com
Cellular metabolism consists of high- and low-carbon-flux pathways to ensure that there is efficient use and synthesis of various metabolites. The pathways for synthesis of vitamins, such as thiamine, are examples of low-carbon-flux pathways. These pathways may therefore be suitable for studying small regulatory changes, which may induce an easily scored phenotype. Proteins expressed at low levels, such as the proteins in low-carbon-flow pathways and regulatory proteins, are encoded on mRNAs with a codon usage distinct from that of mRNAs that encode highly expressed proteins (44). mRNA which encodes proteins that are expressed at low levels may be more sensitive to small aberrations in the translation apparatus, such as deficiencies in tRNA modification or small changes in the environment of the ribosome. Indeed, the mRNA that is expressed at a low level and encodes the regulatory protein VirF in the virulence cascade of Shigella is dependent on the presence of specific modified nucleosides in tRNA, although efficient translation of many other mRNAs in the cell is not dependent on the same modifications (17-19). In this study we examined whether the synthesis of enzymes involved in low-carbon-flow pathways is translationally regulated, and we found that the metabolism of thiamine is regulated in this way.
In Salmonella enterica serovar Typhimurium synthesis of thiamine, which consists of a thiazole and a pyrimidine moiety, occurs by two independent pathways . Thiazole monophosphate and 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate, which are the end products of these pathways, are then condensed, resulting in thiamine monophosphate. Following phosphorylation, thiamine pyrophosphate (TPP), which is the active cofactor in bacteria, is formed. Synthesis of the pyrimidine moiety of thiamine starts with 4-aminoimidazole ribotide (AIR), which is also an intermediate in the synthesis of purine (37, 48) (Fig. 1). For growth on glucose as the carbon source, cells having a mutation in the purF gene require thiamine or pantothenate in addition to a purine (16). However, under certain conditions, such as anaerobic conditions or growth on certain carbon sources other than glucose and citric acid cycle intermediates, purF mutants are able to grow without addition of thiamine or pantothenate (14, 16, 41). This thiamine-independent growth under low-flux conditions requires activation of the alternative pyrimidine biosynthesis (APB) pathway, which requires an active apbA gene product (15). The apbA gene was also identified as the panE gene (27), which encodes ketopantoate reductase, a product required for synthesis of pantothenate. Although the involvement of the panE gene product in the APB pathway explains the link(s) between pantothenate metabolism and thiamine metabolism, the molecular mechanism has not been elucidated (22). The APB pathway also requires active purD, purG, and purI gene products, demonstrating that the APB pathway feeds into the purine pathway at the step producing phosphoribosylamine (PRA), which is the product of the first dedicated step in the purine biosynthesis catalyzed by the purF gene product. The APB pathway requires an active oxidative pentose phosphate pathway (21), showing that ribose 5-phosphate is important for the formation of PRA. Thus, the APB pathway for the synthesis of thiamine is linked not only to the high-carbon-flux purine pathway but also to the oxidative pentose pathway and in some unknown manner to the pantothenate biosynthetic pathway
fig.ommittedwp-&8, 百拇医药
Synthesis of purines, pantothenate, and thiamine. PurF is the first gene in the purine biosynthetic pathway whose end products are adenosine and guanosine. AIR is an intermediate in the synthesis of purine and thiamine. The APB pathway for synthesizing AIR feeds into the purine pathway at the formation of PRA. The APB pathway requires an active oxidative pentose phosphate pathway (gnd and zwf gene products), the PurD, PurG, and PurI proteins, and the panE gene product. The panE gene was previously called the apbA gene. A mutation in this gene blocks the APB pathway and thereby blocks the synthesis of AIR and the synthesis of thiamine. Also, addition of pantothenate stimulates the APB pathway in an unknown manner (question mark). Since growth stimulation by pantothenate requires an active pur pathway between PRA and AIR (data not shown), the pantothenate may feed into the purine pathway at the PRA step, although there may be other explanations. HMP-PP, 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate; THZ-P, thiazole monophosphate; TMP, thiamine monophosphate.wp-&8, 百拇医药
tRNAs from all organisms contain modified nucleosides, which are derivatives of the nucleosides adenosine (A), guanosine (G), uridine (U), and cytidine (C). Although modified nucleosides are present in various positions of the tRNAs, two positions are frequently modified, and numerous different modified nucleosides can be found there. One of these two positions is position 34, which harbors the wobble nucleoside, and the other is position 37, which is the nucleoside adjacent to and 3' of the anticodon. The modified nucleosides found in the latter position are correlated with the coding ability of the tRNA. tRNA reading codons starting with U and A have a hydrophobic modified nucleoside, such as an isopentenyl derivative of A, and a threonylated derivative of A, respectively. The tRNA reading codons CUN (Leu), CCN (Pro), and CGG (Arg) in all organisms contain at position 37 1-methylguanosine (m1G37) (1, 5). Indeed, m1G37 was most likely present in the tRNA of the progenitor which preceded the emergence of organisms into the domains Archaea, Bacteria, and Eucarya, since the tRNA(m1G37)methyltransferases from members of the three phylogenetic domains show sequence similarities (6). This modified nucleoside has profound effects on the function of tRNA; it improves the overall efficiency (34), it increases the rate with which the aminoacyl-tRNA in the ternary complex enters the A site in a tRNA-dependent way (36), and it prevents +1 frameshifting (8, 30). There are reasons to believe (7) that tRNA is not fully modified under all physiological conditions. Therefore, it has been suggested that the degree of tRNA modification may be a regulatory device, since modified nucleosides have such a strong impact on the activity of tRNA (7). In this paper we report results which support such a suggestion, and we show that a lack of m1G37 in tRNA influences the metabolism of thiamine, whereas a lack of some other modified nucleosides does not. We also show that small amounts of antibiotics known to inhibit translation influence the metabolism of thiamine and pantothenate. We conclude that the metabolism of thiamine is controlled on the translational level as well as on the well-established transcriptional level.
MATERIALS AND METHODS4i, 百拇医药
Bacterial strains and plasmids. The bacterial strains used were derivatives of S. enterica strain LT2 (Table 1). Plasmid pBY03 contains the tetracistronic trmD operon, the ffh gene (previously called the 48K gene), which is upstream of the trmD operon, and a gene encoding a 16-kDa protein from Escherichia coli (10). Plasmid pBY12 lacks the last gene of the operon, rplS, and plasmid pMW123 (47) has a Tn5 insertion in the trmD gene, which is the third gene of the operon. Plasmids pUMV4 and pUSC1 harbor the trmD orthologues from Methanococcus vannielii and Saccharomyces cerevisiae, respectively (6).4i, 百拇医药
fig.ommitted4i, 百拇医药
Strains of S. enterica4i, 百拇医药
Media and growth conditions. NCE medium, which was used as the basal salt medium, is medium E (46) lacking citrate and supplemented with 1 mM MgSO4. The carbon sources used were each added at a final concentration of 0.4%. All supplements were added at concentrations suggested previously (13). TYS agar plates (10 g of Trypticase peptone broth per liter, 5 g of yeast extract per liter, 5 g of NaCl per liter, 15 g of Bacto Agar [Difco] per liter) was used as a rich solid medium. The rich liquid medium used was NAA, which contained nutrient broth (0.8%; Difco Laboratories, Detroit, Mich.) supplemented with 0.5% NaCl and adenine, tryptophan, tyrosine, phenylalanine, p-hydroxybenzoate, 2,3-dihydroxybenzoate, and p-aminobenzoate at concentrations recommended by Davis et al. (13). Growth on various plates was determined by resuspending a colony from a TYS agar plate in 0.9% NaCl and then making single-cell streaks on basal salt plates containing supplements. The plates were incubated at the appropriate temperature and scored as described below. The error in determining the colony size was 10%.
Determination of TPP. Strain GT4500 (purF2085 trmD+) was grown in NCE medium containing 0.4% glucose, adenine (50 µg/ml), and thiamine (0.001 µg/ml) to an optical density at 600 nm (OD600) of about 0.6, and then the cells were collected by centrifugation, washed once in 0.9% NaCl, and resuspended in the same growth medium lacking thiamine. The cells were incubated overnight at 37°C. During this time the OD600 did not increase. Strain GT5357 (purF2085 trmD27) was grown in NCE medium containing 0.4% glucose and adenine (50 µg/ml) to an OD600 of 0.6, and then the cells were collected by centrifugation. The TPP was extracted as described previously (23), and derivatization was performed as described by Kawasaki (32). Briefly, the washed cells from the cultures were resuspended in 1 ml of H2O. One-half of each of the suspensions was made 0.1 M HCl and centrifuged, and the supernatants were passed through a 0.45-µm-pore-size filter (Eppendorf centrifugal filter tubes). To 100 µl of each extract 12.5 µl of 0.3 M CNBr was added, and this was followed by neutralization with 12.5 µl of 1 M NaOH. Ten microliters of each extract was subjected to high-performance liquid chromatography (HPLC) by using a Waters 474 fluorescence detector to monitor thiamine, thiamine monophosphate, and TPP contents. The other half of each cell suspension was used to determine the dry weight of the cells.
Determination of CoA derivatives. A method which measures all coenzyme A (CoA) derivatives was obtained from D. Downs, University of Wisconsin, Madison. Cells were grown in 200 ml of MOPS glucose medium at 37°C to an OD600 of about 0.4. The culture was centrifuged, and the cells were resuspended in 1.8 ml of water. One-half of the suspension was pelleted and dried at 65°C for 24 h for dry weight determination. A cell extract was prepared from the other half of the suspension by adding 0.1 ml of 1 M HCOOH. Following centrifugation for 10 min at 10,000 x g, the supernatant was neutralized with 1 M NH4OH. To 1 ml of cell extract (or a standard solution of CoA) 0.4 ml of 2% dithiothreitol was added. The cell extract (or standard solution of CoA) was added to a mixture containing sodium malate, acetylphosphate, NAD, malate dehydrogenase, and citrate synthase. Addition of phosphotransacetylase started the enzymatic reaction, and the rate of NADH synthesis was monitored with a spectrophotometer at 340 nm. The rate of NADH synthesis was converted to the amount of CoA in the extract by using a standard curve generated from assays performed with samples containing known amounts of CoA.
Determination of m1G content of tRNA. Cells were grown in NCE medium containing glucose at 37°C to a density of about 100 Klett units (about 5 x 108 cells/ml). The cells from 25 ml of the culture were harvested and resuspended in 0.4 ml of 25 mM Tris-HCl (pH 7.4)-60 mM KCl-10 mM MgCl2. Total RNA was extracted as described previously (20). The method used included lysozyme treatment without sucrose, addition of a detergent mixture, and finally phenol extraction. Total RNA dissolved in 2 ml of R200 buffer (100 mM Tris-phosphate, 15% ethanol, 200 mM KCl; pH 6.3) was applied to a Nucleobond column equilibrated with the same buffer. The column was washed with 6 ml of R200 buffer. Following washing with 2.5 ml of R650 buffer (same as R200 buffer except that the KCl concentration was 650 mM), the tRNA was eluted in 7 ml of R650 buffer. This tRNA preparation was free of rRNA and DNA. The tRNA was precipitated with isopropanol and digested with nuclease P1 to nucleosides as described previously (29), and the hydrolysate was analyzed by HPLC (28).
RESULTSa9sw!q., http://www.100md.com
Initial observations. Synthesis of purine and synthesis of thiamine are linked through a common intermediate (5'-phosphoribosyl-5-aminoimidazole or AIR) . Upon addition of adenine, which is one of the end products of the purine pathway, repression of the purine biosynthesis occurs and a shortage of AIR is induced, resulting in a thiamine deficiency (38). Therefore, the growth of wild-type S. enterica is inhibited by addition of adenine to the growth medium, and this growth inhibition is relieved by addition of thiamine . This adenine sensitivity is observed only in glucose minimal medium and not when the carbon source is, for example, glycerol . However, strain GT875 (trmD3) is resistant to the addition of adenine, implying that the internal pool of thiamine in this strain is sufficient to overcome the adenine repression.a9sw!q., http://www.100md.com
fig.ommitteda9sw!q., http://www.100md.com
. Growth of trmD+ and trmD3 strains on media containing various carbon sources with or without adeninea
The trmD3 mutation mediates thiamine-independent growth of a strain with a deletion of the purF gene. Deletion of the purF gene (purF2085) results in a growth requirement for thiamine in addition to adenine (37) (Table 3). However, introduction of the trmD3 mutation results in thiamine-independent growth on plates containing adenine . This suggests that the trmD3 mutation increases the pool of thiamine, which is consistent with the adenine-resistant phenotype observed for strain GT875 .pr:ot), 百拇医药
fig.ommittedpr:ot), 百拇医药
Observed thiamine independence is caused by the trmD3 mutation and not by any mutation outside the trmD genepr:ot), 百拇医药
Thiamine independence of a purF mutant is caused by the trmD3 mutation and not by any other mutation close to the trmD gene. The trmD3 mutation was isolated after localized mutagenesis by hydroxylamine (8). Therefore, the unexpected thiamine independence of a trmD3 purF2085 mutant may not be caused by the trmD3 mutation but by a mutation in another gene closely linked to the trmD gene. The trmD gene is part of an operon consisting of rpsP (encoding ribosomal protein S16), rimM (encoding a 21-kDa protein involved in ribosome maturation) (9), trmD [encoding tRNA(m1G37)methyltransferase (TrmD peptide)], and rplS (encoding ribosomal protein L19), which are transcribed in that order. We therefore introduced different plasmids into strains GT875 (trmD3) and GT4501 (purF2085 trmD3). Plasmid pBY03 carries the entire trmD operon, whereas plasmid pBY12 contains all genes except the last gene, rplS. Plasmid pMW123 contains the whole trmD operon, but there is a Tn5 transposon in the trmD gene. The thiamine independence of strain GT4501 (trmD3 purF2085) was inhibited by all plasmids except plasmid pMW123, which has an inactive trmD gene. Similarly, the adenine resistance phenotype of the trmD3 mutant was inhibited by all plasmids except the vector and plasmid pMW123 . Thus, the observed thiamine independence was caused by the trmD3 mutation and not by any other mutation linked to the trmD gene. The trmD3 mutation is a missense mutation causing a proline-to-leucine change at position 184 (35) and therefore is not likely to have a polar effect on the synthesis of L19. All these results are consistent with the suggestion that the unexpected thiamine independence of a trmD3 purF2085 mutant and the adenine resistance phenotype of the trmD3 mutant are indeed due to the trmD3 mutation, which causes m1G37 deficiency in the tRNA.
The thiamine independence of a trmD3 purF2085 mutant is caused by a lack of m1G37 in tRNA and not by a loss of a possible second function of the TrmD peptide. The tRNA (m1G37)methyltransferase (TrmD peptide) methylates a subset of tRNAs in the cell. Therefore, the assumption is that an m1G37-deficient tRNA(s) influences the synthesis of or requirement for thiamine by causing aberrant translation of a key mRNA in the metabolism of thiamine. However, another tRNA methyltransferase, the tRNA(m5U54)methyltransferase (TrmA peptide), has two functions; one function is to methylate the tRNA, and the other essential function is unknown (40). Therefore, the TrmD peptide might also have a second function that influences the synthesis of thiamine. If so, it may be possible to isolate mutations in the trmD gene that do not influence the capacity to methylate the tRNA but still are able to cause thiamine independence in the presence of a purF mutation. We therefore grew phage P22 on a strain (GT3534 [zff-2521::Tn10dCm]) containing a transposon closely linked to the trmD gene and treated the phage stock with hydroxylamine. Using strain GT3670 (purF2085) as the recipient, we selected Cmr transductants on plates containing adenine, thiamine (4 x 10-4 µg/ml), and chloramphenicol, and from these transductants we selected clones that grew without added thiamine at 37°C. All 10 such thiamine-independent transductants had an m1G37-deficient tRNA Thus, no thiamine-independent mutants of strain GT3670 (purF2085) that were still able to fully methylate the tRNA were isolated. Moreover, we isolated several mutants with mutations in the trmD gene that do not influence the methylation of tRNA (35), and these mutations did not result in thiamine independence of a purF mutant . Thus, so far, all mutations in the trmD gene that result in an m1G37 deficiency also result in thiamine independence of a purF2085 mutant, whereas mutations in the trmD gene that do not result in an m1G37 deficiency do not mediate thiamine independence. These results suggest that the thiamine independence of a purF mutant is caused by m1G-deficient tRNA and not by a possible alternative function of the TrmD peptide.
fig.ommittedhm:m, 百拇医药
m1G37 deficiency induces thiamine independence in a purF2085 genetic background. The gray bars indicate the m1G/pseudouridine ratio. The level of m1G was determined by HPLC by using bulk tRNA prepared from cells grown in NCE medium containing glucose at 37°C and was expressed as the absorbance at 254 nm of m1G relative to the absorbance of pseudouridine. The open bars indicate the growth of the various mutants as determined by single-cell streaks on NCE medium plates containing glucose, adenine, and thiamine. The diameters of the colonies were measured after 48 h of incubation at 37°C. The colony size was determined by determining the average size of about five colonies in relation to the size of wild-type colonies. Thus, 0.1 indicates that the growth of the mutant was 10% of that of the wild-type strain. The solid bars indicate the colony size on glucose adenine medium as determined after 48 h of incubation at 37°C. The sizes of the colonies of the various mutants were determined relative to the size of wild-type colonies on plates containing glucose, adenine, and thiamine.
Introduction of plasmid pUMV4, which harbors the trmD orthologue from the archaeon M. vannielii, into strain GT5357 (purF2085 trmD27) suppressed the thiamine independence of this strain and concomitantly increased the level of m1G in the tRNA . Similar results were also obtained with plasmid pUSC1, which harbors the trmD orthologue from S. cerevisiae (designated TRM5 in the yeast [6]). Thus, the thiamine independence was correlated with the lack of m1G in the tRNA. Moreover, since it is very unlikely that the orthologues from an archaeon and a eukaryote also have a possible second function similar to that of the TrmD peptide of S. enterica, we concluded that it is the undermethylated tRNA and not an alternative function of the TrmD peptide that causes the thiamine-independent growth in a purF2085 genetic background.)c@5{;, http://www.100md.com
fig.ommitted)c@5{;, http://www.100md.com
trmD orthologues from M. vannielii and S. cerevisiae suppress the thiamine independence of the purF2085 trmD27 mutant, although they only partially restore the level of m1G37 in tRNAa
Evidently, there is no correlation between the degree of thiamine independence of a purF mutant and the degree of m1G37 deficiency, since even though several of the mutants isolated had no detectable m1G, they had different abilities to grow on medium lacking thiamine (compare trmD28 and trmD30 in Fig. 2). Some mutants had a measurable level of m1G (8 to 40% of the wild-type level) in the tRNA, but they were still able to grow without thiamine as efficiently as some of the mutants that had no detectable level of m1G (for example, the trmD30 mutant with no detectable m1G and the trmD3 mutant with 30% of the wild-type level of m1G grew similarly without thiamine). Moreover, although plasmids pUMV4 (trmD+ from M. vannielii) and pUSC1 (TRM5) completely inhibited the thiamine independence of a purF mutant, they only partially restored the m1G level (Table 4).6;3vp, 百拇医药
Other mutations that result in tRNA modification deficiency, reduce polypeptide growth rates, or induce +1 frameshifting do not mediate thiamine independence of a purF mutant. It has been shown previously that a lack of ms2i6A37 in the tRNA of E. coli results in slow growth and a derepressed level of the Gnd enzyme, whose activity is pivotal in the oxidative pentose pathway (31). Since a lack of m1G also induces slow growth (34), slow growth as such may result in derepressed synthesis of thiamine or an unbalanced metabolism in which thiamine is involved. However, the miaA1 mutation, which results in a lack of ms2io6A37 in tRNA, slow growth, and a reduced polypeptide elongation rate (24), did not induce thiamine independence of a purF mutant (data not shown). Since the miaA mutation induces even slower growth than the trmD3 mutation (24, 34), slow growth as such does not induce the ability to grow without thiamine. A mutation in the rpsL gene, which also reduces the growth rate, did not confer thiamine independence (data not shown). Thus, neither a tRNA modification deficiency nor slow growth as such resulted in thiamine-independent growth of a purF mutant. Apparently, an m1G37 deficiency in tRNA is unique in mediating thiamine independence of a purF mutant.
It has been shown previously that a lack of m1G37 in tRNA mediates the ability to suppress +1 frameshifting at sites that consist of runs of Cs, which are also suppressed by the sufA6 and sufB2 frameshift suppressors (8). The observed thiamine-independent growth in the presence of a purF mutation may be caused by extensive frameshifting in some mRNA(s) critical to obtain this phenotype. Since the specificity of the trmD3-mediated frameshifting and the specificity induced by the sufA6 and sufB2 frameshift suppressors are similar, we tested whether these two frameshift suppressors conferred thiamine independence in a purF2085 background. This was not the case (data not shown), suggesting that the mechanism by which the trmD3 mutation makes a purF2085 mutant independent of thiamine does not involve frameshifting at sites similar to those at which the sufA6 and sufB2 tRNAs frameshift.17h4^, 百拇医药
The trmD3-mediated thiamine independence of a purF mutant requires an active APB pathway. The PurF protein catalyzes the first step in purine synthesis, resulting in the formation of PRA, which is converted to AIR by the products of the purD, purG, and purI genes (Fig. 1). Under low-flux conditions, such as growth on gluconate, an active APB pathway is required. This pathway is blocked by mutations in the panE gene (27) (previously designated the apbA gene [15]) and by mutations in the purD, purG, or purI gene (49). The APB pathway is also blocked by mutations in the gnd and zwf genes, whose products block the oxidative pentose phosphate pathway (21). The coaA gene encodes the enzyme pantothenate kinase, which catalyzes the first dedicated step in the synthesis of CoA by forming 4'-phosphopantothenate. Mutations in this gene also block the APB pathway (26). Since mutations in the purD, purG, purI, panE, and coaA genes blocked the trmD-mediated ability of a purF mutant to grow on glucose without thiamine (data not shown), we suggest that an m1G37-deficient tRNA affects the metabolism of thiamine at some step(s) in the APB pathway.
Since the CoaA enzyme catalyzes an early step in the synthesis of CoA (Fig. 1) and since it is required for an active APB pathway (26), we also measured the levels of CoA derivatives in strains GT4219 (trmD+), GT4218 (trmD3), and GT5454 (trmD27) grown in MOPS glucose medium at 37°C. The levels of CoA derivatives were the same in trmD+ and in trmD3 strains, whereas the levels of CoA derivatives in the trmD27 mutant were somewhat reduced (76% of the wild-type level) (data not shown). Thus, the trmD mutations do not drastically influence the levels of CoA derivatives.{nv$ft, 百拇医药
We concluded that the trmD-mediated thiamine independence of a purF mutant requires an active APB pathway, including an active CoaA enzyme, suggesting that the metabolic flow downstream of the CoaA-catalyzed step and ultimately the synthesis of coenzyme A are pivotal for the trmD-mediated thiamine independence in a purF2085 genetic background.{nv$ft, 百拇医药
The trmD3-mediated thiamine independence of a purF mutant is not caused by poor translation of the purR or purE mRNAs. A mutation in the purR gene, which encodes the purine repressor, derepresses the synthesis of the purine pathway and thus the synthesis of AIR. A purE mutation blocks the conversion of AIR to purine and thereby allocates AIR to the synthesis of thiamine. Therefore, a purF purR purE triple mutant grew on glucose adenine medium without added thiamine (data not shown). One possible explanation for the observed thiamine independence of the trmD3 purF2085 mutant is poor translation of the purR and purE mRNAs, which should result in a phenotype similar to that of the triple purF purR purE mutant. However, we believe that a translational defect in the purR mRNA is unlikely, since several purR-regulated genes were not derepressed in the trmD3 mutant (data not shown). Furthermore, the ß-galactosidase activities of a PurE-LacZ protein fusion were comparable in trmD+ and trmD3 cells (data not shown). These results suggest that poor translation of the purE mRNA in conjunction with poor translation of the purR mRNA is not the cause of the trmD3-mediated thiamine independence in a purF2085 genetic background.
The trmD3-induced Thi + phenotype in a purF mutant is caused by increased synthesis of thiamine and not by a sparing effect of thiamine. The thiamine independence in a purF2085 genetic background mediated by the trmD3 mutation may result from decreased utilization of thiamine (e.g., by blocking of a reaction that requires thiamine, which saves thiamine for other thiamine-dependent reactions). However, the trmD3 mutant can sustain restreaking on plates lacking thiamine and repeated dilutions in a liquid medium, suggesting that it is synthesis of thiamine and not saving of thiamine that is influenced by the trmD3 mutation. Indeed, the level of TPP was at least threefold higher in the purF2085 trmD27 mutant than in the purF2085 trmD+ strain when they were incubated in glucose adenine medium . We concluded that trmD mutations caused increased synthesis of thiamine and not sparing of thiamine..)evn4, 百拇医药
fig.ommitted.)evn4, 百拇医药
Levels of TPP in the wild-type strain and in the trmD27 mutant. Strain GT4500 (purF2085 trmD+) was grown in MOPS glucose medium containing adenine and thiamine to an OD600 of 0.6 and then transferred to the same medium lacking thiamine. Following overnight incubation, during which the cell density did not increase, the level of TPP was determined (average of three determinations of three independent extracts). Strain GT5357 (purF2085 trmD27) was grown in MOPS glucose medium containing adenine to an OD600 of 0.6 and then harvested and examined to determine the level of TPP (average of three determinations of two independent extracts).
Some antibiotics known to reduce the polypeptide chain growth rate (CGR) also induce PurF-independent synthesis of thiamine. A codon early in a gene that is decoded slowly results in a queuing of ribosomes. Such queuing may eventually interfere with translation initiation and thereby decrease the efficiency of translation of an mRNA. If CUN (Leu), CCN (Pro), or CCG (Arg) codons were present close to the 5' end of a gene critical for PurF-independent synthesis of thiamine, the efficiency of decoding of these codons would be sensitive to the level of m1G37 in the corresponding tRNAs. If queuing of ribosomes in the beginning of some mRNA can explain the trmD3-mediated PurF-independent synthesis of thiamine, antibiotics that are known to affect CGR but are not codon specific may induce a similar phenotype. Therefore, we tested the abilities of various antibiotics which are known to affect the CGR to mediate PurF-independent synthesis of thiamine. Chloramphenicol, tetracycline, or fusidic acid but not kanamycin, streptomycin, or erythromycin allowed the purF2085 mutant to grow without thiamine (Table 5). We conclude that antibiotics known to inhibit the CGR also mediated PurF-independent synthesis of thiamine similar to the synthesis mediated by an m1G37 deficiency. Like trmD27-mediated thiamine independence, antibiotic-mediated thiamine independence required an active CoaA enzyme (data not shown), suggesting that the antibiotics affected translation of mRNAs that are involved in the same metabolism as the trmD-induced phenotype.
fig.ommitted'pp{6, 百拇医药
Abilities of different antibiotics to mediate thiamine independencea'pp{6, 百拇医药
DISCUSSION'pp{6, 百拇医药
We show here that an m1G37 deficiency in tRNA influences the metabolism of thiamine and the metabolism of pantothenate. The trmD3-mediated effect on the PurF-independent synthesis of thiamine is through the low-carbon-flux APB pathway that requires an active oxidative pentose phosphate pathway and an active pantothenate kinase (coaA gene product), which catalyzes the first dedicated step in the synthesis of CoA. The trmD-mediated PurF-independent synthesis of thiamine results from a translational effect of m1G37-deficient tRNA and not from a general aberration in translation. We concluded that some step(s) in thiamine metabolism is sensitive to the efficiency of translation of some specific mRNA(s).'pp{6, 百拇医药
There is no apparent correlation between the degree of hypomodification of bulk tRNA and the degree of PurF-independent synthesis of thiamine . In S. enterica there are seven tRNA species that contain m1G37, and any m1G37-deficient tRNA species may cause the PurF-independent thiamine synthesis observed. The tRNA(m1G37)methyltransferase can be altered in such a way that it recognizes two forms of the same tRNA species differently (35). Therefore, the degrees of hypomodification of various tRNA species may not be the same in the different mutants. A residual activity of an altered TrmD enzyme may methylate differently the critical tRNA species that mediate PurF-independent thiamine synthesis, although the level of m1G in bulk tRNA is about the same.
It has been suggested that the ribosome is the sensor for heat and cold shock responses (45). It has also been suggested that antibiotics which result in an empty A site induce a heat shock response, whereas antibiotics that fill the A site generate a cold shock response. Although all six antibiotics tested in this study reduced the growth rate and the average CGR, only three of them (chloramphenicol, tetracycline, and fusidic acid) mediated PurF-independent synthesis of thiamine . These three antibiotics result in a filled A site and induce a cold shock response. However, erythromycin, which also results in a filled A site and induces a cold shock response, did not mediate PurF-independent synthesis of thiamine . Moreover, whereas chloramphenicol, tetracycline, fusidic acid, and erythromycin reduce the level of ppGpp (25), erythromycin did not induce PurF-independent synthesis of thiamine. Accordingly, the trmD27 mutation did not alter the ppGpp level in cells grown in MOPS glucose medium (data not shown). Thus, the PurF-independent synthesis of thiamine is not correlated with some general signals from the ribosome, like the signal that induces cold shock or reduces the level of ppGpp. Interestingly, changes in the synthesis of some metabolic enzymes are triggered by various antibiotics to different degrees (25). Similarly, our results also indicate that there is specificity in the responses of antibiotics, since not all antibiotics tested induced PurF-independent synthesis of thiamine although they all reduced the growth rate and the average CGR.
The efficiency with which a tRNA species reads a certain codon is influenced by its concentration in the cell, the degree of modification, and the fraction of the tRNA population that is aminoacylated. The modified nucleosides play a pivotal role by improving the decoding capacity of the tRNA and also by decreasing the codon sensitivity, by influencing the codon choice, and by maintaining the reading frame. Moreover, the impact of a specific modified nucleoside is tRNA species dependent (for reviews see references 2,, 3, 4, and 12). Unbalanced metabolism caused by, for example, various environmental changes may induce unbalanced synthesis of tRNA, resulting in hypomethylated tRNA; e.g., limitation of leucine induces hypomodification of tRNAPhe (33; for a review see reference 4). Such a hypomethylated tRNA may in turn affect the translation of some specific mRNAs. Thus, the degree of modification of a tRNA influences the efficiency of translation and may thereby regulate various parts of the metabolism (for a review see reference 7). Thus, some environmental changes may induce m1G37 deficiency, and the hypomethylated tRNA may induce an altered flow of some specific metabolic pathway, as illustrated in this study by derepressed synthesis of thiamine. There are several ways that such an m1G37-deficient tRNA may induce altered gene expression, and two of them are discussed below.
One way that such a translational regulatory event may occur is by a low level of frameshifting caused by m1G37 deficiency, resulting in a low level of an enzyme or a regulatory protein in thiamine or pantothenate metabolism. Such frameshifting should occur at CUN, CCN, or CGG codons, since these codons are read by m1G37-containing tRNA. However, the frameshift suppressor mutations sufA6 and sufB2, which alter the m1G37-containing and , respectively, so that they frameshift at a CCC codon (42), did not induce PurF-independent synthesis of thiamine like the mutations in the trmD gene did. If frameshifting were caused by an m1G37-deficient tRNAPro at such a site, we would expect that the sufA6 and the sufB2 frameshift suppressors should mediate similar PurF-independent synthesis of thiamine. Furthermore, some antibiotics which are known to influence the amino acid step time on the ribosome caused PurF-independent synthesis of thiamine. However, they did not suppress several +1 frameshift mutations in the his operon, which are known to be suppressed by a lack of m1G37 (8) (data not shown). Therefore, we do not favor the hypothesis that a frameshifting event by an m1G37-deficient tRNAPro is responsible for the PurF-independent synthesis of thiamine. However, frameshifting by an m1G37-deficient or tRNALeu cannot be ruled out.
Another way that an m1G37-deficient tRNA may alter the efficiency of translation of an mRNA is by inducing stalling of ribosomes. If a ribosome is retarded early on an mRNA, a queuing of ribosomes occurs that eventually may inhibit translation initiation. Within the first 25 codons rare codons are preferentially used (11), and two of them, CCC (Pro) and CUA (Leu), are decoded by tRNAs having m1G37. A deficiency of m1G37 reduces the rates with which Leu tRNAs, Pro tRNAs, and Arg tRNAs decode CUU/A/G (Leu), CCN (Pro), and CGG (Arg) codons, respectively (34, 36). Thus, slow decoding of these codons early in an mRNA(s) encoding an enzyme or a regulatory protein involved in thiamine metabolism may result in reduced translation initiation and thus less expression of the protein. Even a small change in the synthesis of such an enzyme or regulatory protein may give a distinct phenotype. Indeed, there is precedence for such a suggestion: a lack of queuosine (Q34), which is present in the wobble position of a subset of tRNAs, reduces the synthesis of the positive regulatory protein VirF, and this results in reduced synthesis of proteins required for the virulence of Shigella flexneri (17, 18). This observation was somewhat surprising, since a mutant (tgt) lacking this modified nucleoside in E. coli does not show any evident phenotypic difference compared to the wild type (39). One interpretation of these observations is that there is a sequence(s) in the virF mRNA that is not present in most other mRNAs and is especially sensitive to being decoded by the Q34-deficient tRNA. Also, a lack of ms2io6A37 in S. enterica increases synthesis of the Gnd enzyme but not synthesis of the Zwf enzyme; these two enzymes are both involved in the oxidative pentose phosphate pathway (31). Thus, our results showing that a lack of m1G37 influences the metabolism of thiamine are similar to the observation that translation of the virF and gnd mRNAs may be more dependent on the status of tRNA modification than translation of other mRNAs is. A low-abundance mRNA, such as an mRNA encoding a regulatory protein or an enzyme in a low-carbon-flux pathway, may have some special requirement for proper translation and therefore may be more sensitive than other mRNAs to small but significant changes in translation efficiency or fidelity.
We hypothesize that there is a target mRNA(s) with codons read by m1G37-containing tRNAs close to the translation initiation site and that in such a codon context the mRNA(s) would be sensitive to the level of m1G37 in tRNA. The target mRNA(s) encodes an enzyme catalyzing a reaction involved in thiamine or pantothenate metabolism. A low level of m1G37 would result in a low level of this target enzyme and thereby cause a low flux in the corresponding pathways. Such a restriction in a biosynthetic pathway should result in a buildup of a metabolite(s) upstream of the restriction point, and thus the metabolite(s) may be channeled into an alternative pathway, leading to increased flux in, for example, the APB pathway. Alternatively, the target mRNA encodes a regulatory protein. A reduced level of this regulatory protein in turn induces an altered flux in the APB pathway. Interestingly, increased production of pantothenate by overexpression of the PanB enzyme slightly increases the CoA level and concomitantly mediates PurF-independent synthesis of thiamine (43). Reduced flux through the purine pathway, like the flux in a purF mutant, results in an increased requirement for CoA to synthesize thiamine (26). It has been suggested that some CoA thioesters may be involved in modifying (e.g., succenylating) an enzyme involved in the conversion of AIR to 4-amino-5-hydroxymethyl-2-methylpyrimidine. Although we did not observe any altered CoA levels in the trmD mutants, increased flux in the pantothenate-CoA pathway in the trmD mutant may still be the reason for the observed trmD-mediated thiamine independence. We suggest that the m1G37 deficiency mediates its translational effect on thiamine metabolism by acting on some step(s) following the production of CoA derivatives (e.g., in a modifying enzyme that uses a CoA thioester as a cofactor). Such an increase in flux would result in increased synthesis of thiamine.
ACKNOWLEDGMENTS0'', 百拇医药
This work was supported by the Science Research Council (project BU-2930) and by the Cancer Foundation (project 680).0'', 百拇医药
We thank Diana Downs, University of Wisconsin, Madison, for generous gifts of strains and for introducing us to thiamine metabolism. We thank Kerstin Jacobsson for excellent technical assistance with the HPLC analysis and for the use of plasmids pUMV4 and pUSC1. Gunilla Jäger is acknowledged for providing strains having the transposon zff-2521::Tn10dCm. Critical reading of the manuscript by Olof Persson and Mikael Wikström is gratefully acknowledged.0'', 百拇医药
REFERENCES0'', 百拇医药
Björk, G. R. 1986. Transfer RNA modification in different organisms. Chem. Scr. 26B:91-95.0'', 百拇医药
Björk, G. R. 1995. Genetic dissection of synthesis and function of modified nucleosides in bacterial transfer RNA. Prog. Nucleic Acid Res. Mol. Biol. 50:263-338.0'', 百拇医药
Björk, G. R. 1995. Biosynthesis and function of modified nucleosides in tRNA, p. 165-205. In D. Söll and U. L. RajBhandary (ed.), tRNA: structure, biosynthesis, and function. ASM Press, Washington, D.C.
Björk, G. R. 1996. Stable RNA modification, p. 861-886. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington.\[i|+, http://www.100md.com
Björk, G. R. 1998. Modified nucleosides in positions 34 and 37 of tRNAs and their predicted coding capacities, p. 577-581. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.\[i|+, http://www.100md.com
Björk, G. R., K. Jacobsson, K. Nilsson, M. J. O. Johansson, A. S. Byström, and O. P. Persson. 2001. A primordial tRNA modification required for the evolution of life? EMBO J. 20:231-239.\[i|+, http://www.100md.com
Björk, G. R., and T. Rasmuson. 1998. Links between tRNA modification and metabolism and modified nucleosides as tumor marker, p. 471-491. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.\[i|+, http://www.100md.com
Björk, G. R., P. M. Wikström, and A. S. Byström. 1989. Prevention of translational frameshifting by the modified nucleoside 1-methylguanosine. Science 244:986-989.
Bylund, G. O., L. C. Wipemo, L. A. C. Lundberg, and P. M. Wikström. 1998. RimM and RbfA are essential for efficient processing of 16S rRNA in Escherichia coli. J. Bacteriol. 180:73-82.|h\qo2/, http://www.100md.com
Byström, A. S., K. J. Hjalmarsson, P. M. Wikström, and G. R. Björk. 1983. The nucleotide sequence of an Escherichia coli operon containing genes for the tRNA(m1G)methyltransferase, the ribosomal proteins S16 and L19 and a 21-K polypeptide. EMBO J. 2:899-905.|h\qo2/, http://www.100md.com
Chen, G. F., and M. Inouye. 1990. Suppression of the negative effect of minor arginine codons on gene expression; preferential usage of minor codons within the first 25 codons of the Escherichia coli genes. Nucleic Acids Res. 18:1465-1473.|h\qo2/, http://www.100md.com
Curran, J. 1998. Modified nucleosides in translation, p. 493-519. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.|h\qo2/, http://www.100md.com
Davis, W., D. Botstein, and J. R. Roth. 1980. A manual for genetic engineering: advanced bacterial genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Downs, D. M. 1992. Evidence for a new, oxygen-regulated biosynthetic pathway for the pyrimidine moiety of thiamine in Salmonella typhimurium. J. Bacteriol. 174:1515-1521.7j][((, http://www.100md.com
Downs, D. M., and L. Petersen. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.7j][((, http://www.100md.com
Downs, D. M., and J. R. Roth. 1991. Synthesis of thiamine in Salmonella typhimurium independent of the purF function. J. Bacteriol. 173:6597-6604.7j][((, http://www.100md.com
Durand, J. M., B. Dagberg, B. E. Uhlin, and G. R. Björk. 2000. Transfer RNA modification, temperature and DNA superhelicity have a common target in the regulatory network of the virulence of Shigella flexneri: the expression of the virF gene. Mol. Microbiol. 35:924-935.7j][((, http://www.100md.com
Durand, J. M., N. Okada, T. Tobe, M. Watarai, I. Fukuda, T. Suzuki, N. Nakata, K. Komatsu, M. Yoshikawa, and C. Sasakawa. 1994. vacC, a virulence-associated chromosomal locus of Shigella flexneri, is homologous to tgt, a gene encoding tRNA-guanine transglycosylase (Tgt) of Escherichia coli K-12. J. Bacteriol. 176:4627-4634.
Durand, J. M. B., G. R. Björk, A. Kuwae, M. Yoshikawa, and C. Sasakawa. 1997. The modified nucleoside 2-methylthio-N-6-isopentenyladenosine in tRNA of Shigella flexneri is required for expression of virulence genes. J. Bacteriol. 179:5777-5782.dzc$], http://www.100md.com
Emilsson, V., and C. G. Kurland. 1990. Growth rate dependence of transfer RNA abundance in Escherichia coli. EMBO J. 9:4359-4366.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1996. Involvement of the oxidative pentose phosphate pathway in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 178:1476-1479.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1997. Mutations in sdh (succinate dehydrogenase genes) alter the thiamine requirement of Salmonella typhimurium. J. Bacteriol. 179:3989-3996.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1999. Biosynthesis of the pyrimidine moiety of thiamine independent of the PurF enzyme (phosphoribosylpyrophosphate amidotransferase) in Salmonella typhimurium: incorporation of stable isotope-labeled glycine and formate. J. Bacteriol. 181:841-848.
Ericson, J. U., and G. R. Björk. 1986. Pleiotropic effects induced by modification deficiency next to the anticodon of tRNA from Salmonella typhimurium LT2. J. Bacteriol. 166:1013-1021.\a7yfsy, 百拇医药
Evers, S., K. Di Padova, M. Meyer, H. Langen, M. Fountoulakis, W. Keck, and C. P. Gray. 2001. Mechanism-related changes in the gene transcription and protein synthesis patterns of Haemophilus influenzae after treatment with transcriptional and translational inhibitors. Proteomics 1:522-544.\a7yfsy, 百拇医药
Frodyma, M., A. Rubio, and D. M. Downs. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:236-240.\a7yfsy, 百拇医药
Frodyma, M. E., and D. Downs. 1998. The panE gene, encoding ketopantoate reductase, maps at 10 minutes and is allelic to apbA in Salmonella typhimurium. J. Bacteriol. 180:4757-4759.\a7yfsy, 百拇医药
Gehrke, C. W., and K. C. Kuo. 1990. Ribonucleoside analysis by reversed-phase high performance liquid chromatography. J. Chromatogr. Libr. 45A:A3-A71.
Gehrke, C. W., K. C. Kuo, R. A. McCune, K. O. Gerhardt, and P. F. Agris. 1982. Quantitative enzymatic hydrolysis of tRNAs: reversed-phase high-performance liquid chromatography of tRNA nucleosides. J. Chromatogr. 230:297-308.3, 百拇医药
Hagervall, T. G., T. M. Tuohy, J. F. Atkins, and G. R. Björk. 1993. Deficiency of 1-methylguanosine in tRNA from Salmonella typhimurium induces frameshifting by quadruplet translocation. J. Mol. Biol. 232:756-765.3, 百拇医药
Jones, W. R., G. J. Barcak, and R. E. Wolf, Jr. 1990. Altered growth-rate-dependent regulation of 6-phosphogluconate dehydrogenase level in hisT mutants of Salmonella typhimurium and Escherichia coli. J. Bacteriol. 172:1197-1205.3, 百拇医药
Kawasaki, T. 1986. Determination of thiamin and its phosphate esters by high-performance liquid chromatography. Methods Enzymol. 122:15-29.3, 百拇医药
Kitchingman, G. R., and M. J. Fournier. 1977. Modification-deficient transfer ribonucleic acids from relaxed control Escherichia coli: structures of the major undermodified phenylalanine and leucine transfer RNAs produced during leucine starvation. Biochemistry 16:2213-2220.
Li, J. N., and G. R. Björk. 1995. 1-Methylguanosine deficiency of tRNA influences cognate codon interaction and metabolism in Salmonella typhimurium. J. Bacteriol. 177:6593-6600.8-l$h, 百拇医药
Li, J. N., and G. R. Björk. 1999. Structural alterations of the tRNA(m1G37)methyltransferase from Salmonella typhimurium affect tRNA substrate specificity. RNA 5:395-408.8-l$h, 百拇医药
Li, J. N., B. Esberg, J. F. Curran, and G. R. Björk. 1997. Three modified nucleosides present in the anticodon stem and loop influence the in vivo aa-tRNA selection in a tRNA-dependent manner. J. Mol. Biol. 271:209-221.8-l$h, 百拇医药
Newell, P. C., and R. G. Tucker. 1967. New pyrimidine pathway involved in the biosynthesis of the pyrimidine of thiamine. Nature 215:1384-1385.8-l$h, 百拇医药
Newell, P. C., and R. G. Tucker. 1968. Precursors of the pyrimidine moiety of thiamine. A new route of pyrimidine biosynthesis involving purine intermediates. Biochem. J. 106:279-287.8-l$h, 百拇医药
Noguchi, S., Y. Nishimura, Y. Hirota, and S. Nishimura. 1982. Isolation and characterization of an Escherichia coli mutant lacking tRNA-guanine transglycosylase. Function and biosynthesis of queuosine in tRNA. J. Biol. Chem. 257:6544-6550.
Persson, B. C., C. Gustafsson, D. E. Berg, and G. R. Björk. 1992. The gene for a tRNA modifying enzyme, m5U54-methyltransferase, is essential for viability in Escherichia coli. Proc. Natl. Acad. Sci. USA 89:3995-3998.*kefdj1, 百拇医药
Petersen, L., J. Enosberlage, and D. M. Downs. 1996. Genetic analysis of metabolic crosstalk and its impact on thiamine synthesis in Salmonella typhimurium. Genetics 143:37-44.*kefdj1, 百拇医药
Qian, Q., J. N. Li, H. Zhao, T. G. Hagervall, P. J. Farabaugh, and G. R. Björk. 1998. A new model for phenotypic suppression of frameshift mutations by mutant tRNAs. Mol. Cell 1:471-482.*kefdj1, 百拇医药
Rubio, A., and D. M. Downs. 2002. Elevated levels of ketopantoate hydroxymethyltransferase (PanB) lead to a physiologically significant coenzyme A elevation in Salmonella enterica serovar Typhimurium. J. Bacteriol. 184:2827-2832.*kefdj1, 百拇医药
Sharp, P. M., and W. H. Li. 1986. Codon usage in regulatory genes in Escherichia coli does not reflect selection for 'rare' codons. Nucleic Acids Res. 14:7737-7749.
VanBogelen, R. A., and F. C. Neidhardt. 1990. Ribosomes as sensors of heat and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA 87:5589-5593.7\9eh, http://www.100md.com
Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218:97-106.7\9eh, http://www.100md.com
Wikström, P. M., A. S. Byström, and G. R. Björk. 1988. Non-autogenous control of ribosomal protein synthesis from the trmD operon in Escherichia coli. J. Mol. Biol. 203:141-152.7\9eh, http://www.100md.com
Yura, T. 1956. Evidence of nonidentical alleles in purine requiring mutants of Salmonella typhimurium. Publ. Carnegie Inst. 612:63-75.7\9eh, http://www.100md.com
Zilles, J. L., and D. M. Downs. 1996. A novel involvement of the PurG and PurI proteins in thiamine synthesis via the alternative pyrimidine biosynthetic (APB) pathway in Salmonella typhimurium. Genetics 144:883-892.(Glenn R. Björk and Kristina Nilsson)
Received 2 July 2002/ Accepted 1 November 2002o|@q&, 百拇医药
ABSTRACTo|@q&, 百拇医药
In Salmonella enterica serovar Typhimurium a mutation in the purF gene encoding the first enzyme in the purine pathway blocks, besides the synthesis of purine, the synthesis of thiamine when glucose is used as the carbon source. On carbon sources other than glucose, a purF mutant does not require thiamine, since the alternative pyrimidine biosynthetic (APB) pathway is activated. This pathway feeds into the purine pathway just after the PurF biosynthetic step and upstream of the intermediate 4-aminoimidazolribotide, which is the common intermediate in purine and thiamine synthesis. The activity of this pathway is also influenced by externally added pantothenate. tRNAs from S. enterica specific for leucine, proline, and arginine contain 1-methylguanosine (m1G37) adjacent to and 3' of the anticodon (position 37). The formation of m1G37 is catalyzed by the enzyme tRNA(m1G37)methyltransferase, which is encoded by the trmD gene. Mutations in this gene, which result in an m1G37 deficiency in the tRNA, in a purF mutant mediate PurF-independent thiamine synthesis. This phenotype is specifically dependent on the m1G37 deficiency, since several other mutations which also affect translation fidelity and induce slow growth did not cause PurF-independent thiamine synthesis. Some antibiotics that are known to reduce the efficiency of translation also induce PurF-independent thiamine synthesis. We suggest that a slow decoding event at a codon(s) read by a tRNA(s) normally containing m1G37 is responsible for the PurF-independent thiamine synthesis and that this event causes a changed flux in the APB pathway.
INTRODUCTIONj3z0, http://www.100md.com
Cellular metabolism consists of high- and low-carbon-flux pathways to ensure that there is efficient use and synthesis of various metabolites. The pathways for synthesis of vitamins, such as thiamine, are examples of low-carbon-flux pathways. These pathways may therefore be suitable for studying small regulatory changes, which may induce an easily scored phenotype. Proteins expressed at low levels, such as the proteins in low-carbon-flow pathways and regulatory proteins, are encoded on mRNAs with a codon usage distinct from that of mRNAs that encode highly expressed proteins (44). mRNA which encodes proteins that are expressed at low levels may be more sensitive to small aberrations in the translation apparatus, such as deficiencies in tRNA modification or small changes in the environment of the ribosome. Indeed, the mRNA that is expressed at a low level and encodes the regulatory protein VirF in the virulence cascade of Shigella is dependent on the presence of specific modified nucleosides in tRNA, although efficient translation of many other mRNAs in the cell is not dependent on the same modifications (17-19). In this study we examined whether the synthesis of enzymes involved in low-carbon-flow pathways is translationally regulated, and we found that the metabolism of thiamine is regulated in this way.
In Salmonella enterica serovar Typhimurium synthesis of thiamine, which consists of a thiazole and a pyrimidine moiety, occurs by two independent pathways . Thiazole monophosphate and 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate, which are the end products of these pathways, are then condensed, resulting in thiamine monophosphate. Following phosphorylation, thiamine pyrophosphate (TPP), which is the active cofactor in bacteria, is formed. Synthesis of the pyrimidine moiety of thiamine starts with 4-aminoimidazole ribotide (AIR), which is also an intermediate in the synthesis of purine (37, 48) (Fig. 1). For growth on glucose as the carbon source, cells having a mutation in the purF gene require thiamine or pantothenate in addition to a purine (16). However, under certain conditions, such as anaerobic conditions or growth on certain carbon sources other than glucose and citric acid cycle intermediates, purF mutants are able to grow without addition of thiamine or pantothenate (14, 16, 41). This thiamine-independent growth under low-flux conditions requires activation of the alternative pyrimidine biosynthesis (APB) pathway, which requires an active apbA gene product (15). The apbA gene was also identified as the panE gene (27), which encodes ketopantoate reductase, a product required for synthesis of pantothenate. Although the involvement of the panE gene product in the APB pathway explains the link(s) between pantothenate metabolism and thiamine metabolism, the molecular mechanism has not been elucidated (22). The APB pathway also requires active purD, purG, and purI gene products, demonstrating that the APB pathway feeds into the purine pathway at the step producing phosphoribosylamine (PRA), which is the product of the first dedicated step in the purine biosynthesis catalyzed by the purF gene product. The APB pathway requires an active oxidative pentose phosphate pathway (21), showing that ribose 5-phosphate is important for the formation of PRA. Thus, the APB pathway for the synthesis of thiamine is linked not only to the high-carbon-flux purine pathway but also to the oxidative pentose pathway and in some unknown manner to the pantothenate biosynthetic pathway
fig.ommittedwp-&8, 百拇医药
Synthesis of purines, pantothenate, and thiamine. PurF is the first gene in the purine biosynthetic pathway whose end products are adenosine and guanosine. AIR is an intermediate in the synthesis of purine and thiamine. The APB pathway for synthesizing AIR feeds into the purine pathway at the formation of PRA. The APB pathway requires an active oxidative pentose phosphate pathway (gnd and zwf gene products), the PurD, PurG, and PurI proteins, and the panE gene product. The panE gene was previously called the apbA gene. A mutation in this gene blocks the APB pathway and thereby blocks the synthesis of AIR and the synthesis of thiamine. Also, addition of pantothenate stimulates the APB pathway in an unknown manner (question mark). Since growth stimulation by pantothenate requires an active pur pathway between PRA and AIR (data not shown), the pantothenate may feed into the purine pathway at the PRA step, although there may be other explanations. HMP-PP, 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate; THZ-P, thiazole monophosphate; TMP, thiamine monophosphate.wp-&8, 百拇医药
tRNAs from all organisms contain modified nucleosides, which are derivatives of the nucleosides adenosine (A), guanosine (G), uridine (U), and cytidine (C). Although modified nucleosides are present in various positions of the tRNAs, two positions are frequently modified, and numerous different modified nucleosides can be found there. One of these two positions is position 34, which harbors the wobble nucleoside, and the other is position 37, which is the nucleoside adjacent to and 3' of the anticodon. The modified nucleosides found in the latter position are correlated with the coding ability of the tRNA. tRNA reading codons starting with U and A have a hydrophobic modified nucleoside, such as an isopentenyl derivative of A, and a threonylated derivative of A, respectively. The tRNA reading codons CUN (Leu), CCN (Pro), and CGG (Arg) in all organisms contain at position 37 1-methylguanosine (m1G37) (1, 5). Indeed, m1G37 was most likely present in the tRNA of the progenitor which preceded the emergence of organisms into the domains Archaea, Bacteria, and Eucarya, since the tRNA(m1G37)methyltransferases from members of the three phylogenetic domains show sequence similarities (6). This modified nucleoside has profound effects on the function of tRNA; it improves the overall efficiency (34), it increases the rate with which the aminoacyl-tRNA in the ternary complex enters the A site in a tRNA-dependent way (36), and it prevents +1 frameshifting (8, 30). There are reasons to believe (7) that tRNA is not fully modified under all physiological conditions. Therefore, it has been suggested that the degree of tRNA modification may be a regulatory device, since modified nucleosides have such a strong impact on the activity of tRNA (7). In this paper we report results which support such a suggestion, and we show that a lack of m1G37 in tRNA influences the metabolism of thiamine, whereas a lack of some other modified nucleosides does not. We also show that small amounts of antibiotics known to inhibit translation influence the metabolism of thiamine and pantothenate. We conclude that the metabolism of thiamine is controlled on the translational level as well as on the well-established transcriptional level.
MATERIALS AND METHODS4i, 百拇医药
Bacterial strains and plasmids. The bacterial strains used were derivatives of S. enterica strain LT2 (Table 1). Plasmid pBY03 contains the tetracistronic trmD operon, the ffh gene (previously called the 48K gene), which is upstream of the trmD operon, and a gene encoding a 16-kDa protein from Escherichia coli (10). Plasmid pBY12 lacks the last gene of the operon, rplS, and plasmid pMW123 (47) has a Tn5 insertion in the trmD gene, which is the third gene of the operon. Plasmids pUMV4 and pUSC1 harbor the trmD orthologues from Methanococcus vannielii and Saccharomyces cerevisiae, respectively (6).4i, 百拇医药
fig.ommitted4i, 百拇医药
Strains of S. enterica4i, 百拇医药
Media and growth conditions. NCE medium, which was used as the basal salt medium, is medium E (46) lacking citrate and supplemented with 1 mM MgSO4. The carbon sources used were each added at a final concentration of 0.4%. All supplements were added at concentrations suggested previously (13). TYS agar plates (10 g of Trypticase peptone broth per liter, 5 g of yeast extract per liter, 5 g of NaCl per liter, 15 g of Bacto Agar [Difco] per liter) was used as a rich solid medium. The rich liquid medium used was NAA, which contained nutrient broth (0.8%; Difco Laboratories, Detroit, Mich.) supplemented with 0.5% NaCl and adenine, tryptophan, tyrosine, phenylalanine, p-hydroxybenzoate, 2,3-dihydroxybenzoate, and p-aminobenzoate at concentrations recommended by Davis et al. (13). Growth on various plates was determined by resuspending a colony from a TYS agar plate in 0.9% NaCl and then making single-cell streaks on basal salt plates containing supplements. The plates were incubated at the appropriate temperature and scored as described below. The error in determining the colony size was 10%.
Determination of TPP. Strain GT4500 (purF2085 trmD+) was grown in NCE medium containing 0.4% glucose, adenine (50 µg/ml), and thiamine (0.001 µg/ml) to an optical density at 600 nm (OD600) of about 0.6, and then the cells were collected by centrifugation, washed once in 0.9% NaCl, and resuspended in the same growth medium lacking thiamine. The cells were incubated overnight at 37°C. During this time the OD600 did not increase. Strain GT5357 (purF2085 trmD27) was grown in NCE medium containing 0.4% glucose and adenine (50 µg/ml) to an OD600 of 0.6, and then the cells were collected by centrifugation. The TPP was extracted as described previously (23), and derivatization was performed as described by Kawasaki (32). Briefly, the washed cells from the cultures were resuspended in 1 ml of H2O. One-half of each of the suspensions was made 0.1 M HCl and centrifuged, and the supernatants were passed through a 0.45-µm-pore-size filter (Eppendorf centrifugal filter tubes). To 100 µl of each extract 12.5 µl of 0.3 M CNBr was added, and this was followed by neutralization with 12.5 µl of 1 M NaOH. Ten microliters of each extract was subjected to high-performance liquid chromatography (HPLC) by using a Waters 474 fluorescence detector to monitor thiamine, thiamine monophosphate, and TPP contents. The other half of each cell suspension was used to determine the dry weight of the cells.
Determination of CoA derivatives. A method which measures all coenzyme A (CoA) derivatives was obtained from D. Downs, University of Wisconsin, Madison. Cells were grown in 200 ml of MOPS glucose medium at 37°C to an OD600 of about 0.4. The culture was centrifuged, and the cells were resuspended in 1.8 ml of water. One-half of the suspension was pelleted and dried at 65°C for 24 h for dry weight determination. A cell extract was prepared from the other half of the suspension by adding 0.1 ml of 1 M HCOOH. Following centrifugation for 10 min at 10,000 x g, the supernatant was neutralized with 1 M NH4OH. To 1 ml of cell extract (or a standard solution of CoA) 0.4 ml of 2% dithiothreitol was added. The cell extract (or standard solution of CoA) was added to a mixture containing sodium malate, acetylphosphate, NAD, malate dehydrogenase, and citrate synthase. Addition of phosphotransacetylase started the enzymatic reaction, and the rate of NADH synthesis was monitored with a spectrophotometer at 340 nm. The rate of NADH synthesis was converted to the amount of CoA in the extract by using a standard curve generated from assays performed with samples containing known amounts of CoA.
Determination of m1G content of tRNA. Cells were grown in NCE medium containing glucose at 37°C to a density of about 100 Klett units (about 5 x 108 cells/ml). The cells from 25 ml of the culture were harvested and resuspended in 0.4 ml of 25 mM Tris-HCl (pH 7.4)-60 mM KCl-10 mM MgCl2. Total RNA was extracted as described previously (20). The method used included lysozyme treatment without sucrose, addition of a detergent mixture, and finally phenol extraction. Total RNA dissolved in 2 ml of R200 buffer (100 mM Tris-phosphate, 15% ethanol, 200 mM KCl; pH 6.3) was applied to a Nucleobond column equilibrated with the same buffer. The column was washed with 6 ml of R200 buffer. Following washing with 2.5 ml of R650 buffer (same as R200 buffer except that the KCl concentration was 650 mM), the tRNA was eluted in 7 ml of R650 buffer. This tRNA preparation was free of rRNA and DNA. The tRNA was precipitated with isopropanol and digested with nuclease P1 to nucleosides as described previously (29), and the hydrolysate was analyzed by HPLC (28).
RESULTSa9sw!q., http://www.100md.com
Initial observations. Synthesis of purine and synthesis of thiamine are linked through a common intermediate (5'-phosphoribosyl-5-aminoimidazole or AIR) . Upon addition of adenine, which is one of the end products of the purine pathway, repression of the purine biosynthesis occurs and a shortage of AIR is induced, resulting in a thiamine deficiency (38). Therefore, the growth of wild-type S. enterica is inhibited by addition of adenine to the growth medium, and this growth inhibition is relieved by addition of thiamine . This adenine sensitivity is observed only in glucose minimal medium and not when the carbon source is, for example, glycerol . However, strain GT875 (trmD3) is resistant to the addition of adenine, implying that the internal pool of thiamine in this strain is sufficient to overcome the adenine repression.a9sw!q., http://www.100md.com
fig.ommitteda9sw!q., http://www.100md.com
. Growth of trmD+ and trmD3 strains on media containing various carbon sources with or without adeninea
The trmD3 mutation mediates thiamine-independent growth of a strain with a deletion of the purF gene. Deletion of the purF gene (purF2085) results in a growth requirement for thiamine in addition to adenine (37) (Table 3). However, introduction of the trmD3 mutation results in thiamine-independent growth on plates containing adenine . This suggests that the trmD3 mutation increases the pool of thiamine, which is consistent with the adenine-resistant phenotype observed for strain GT875 .pr:ot), 百拇医药
fig.ommittedpr:ot), 百拇医药
Observed thiamine independence is caused by the trmD3 mutation and not by any mutation outside the trmD genepr:ot), 百拇医药
Thiamine independence of a purF mutant is caused by the trmD3 mutation and not by any other mutation close to the trmD gene. The trmD3 mutation was isolated after localized mutagenesis by hydroxylamine (8). Therefore, the unexpected thiamine independence of a trmD3 purF2085 mutant may not be caused by the trmD3 mutation but by a mutation in another gene closely linked to the trmD gene. The trmD gene is part of an operon consisting of rpsP (encoding ribosomal protein S16), rimM (encoding a 21-kDa protein involved in ribosome maturation) (9), trmD [encoding tRNA(m1G37)methyltransferase (TrmD peptide)], and rplS (encoding ribosomal protein L19), which are transcribed in that order. We therefore introduced different plasmids into strains GT875 (trmD3) and GT4501 (purF2085 trmD3). Plasmid pBY03 carries the entire trmD operon, whereas plasmid pBY12 contains all genes except the last gene, rplS. Plasmid pMW123 contains the whole trmD operon, but there is a Tn5 transposon in the trmD gene. The thiamine independence of strain GT4501 (trmD3 purF2085) was inhibited by all plasmids except plasmid pMW123, which has an inactive trmD gene. Similarly, the adenine resistance phenotype of the trmD3 mutant was inhibited by all plasmids except the vector and plasmid pMW123 . Thus, the observed thiamine independence was caused by the trmD3 mutation and not by any other mutation linked to the trmD gene. The trmD3 mutation is a missense mutation causing a proline-to-leucine change at position 184 (35) and therefore is not likely to have a polar effect on the synthesis of L19. All these results are consistent with the suggestion that the unexpected thiamine independence of a trmD3 purF2085 mutant and the adenine resistance phenotype of the trmD3 mutant are indeed due to the trmD3 mutation, which causes m1G37 deficiency in the tRNA.
The thiamine independence of a trmD3 purF2085 mutant is caused by a lack of m1G37 in tRNA and not by a loss of a possible second function of the TrmD peptide. The tRNA (m1G37)methyltransferase (TrmD peptide) methylates a subset of tRNAs in the cell. Therefore, the assumption is that an m1G37-deficient tRNA(s) influences the synthesis of or requirement for thiamine by causing aberrant translation of a key mRNA in the metabolism of thiamine. However, another tRNA methyltransferase, the tRNA(m5U54)methyltransferase (TrmA peptide), has two functions; one function is to methylate the tRNA, and the other essential function is unknown (40). Therefore, the TrmD peptide might also have a second function that influences the synthesis of thiamine. If so, it may be possible to isolate mutations in the trmD gene that do not influence the capacity to methylate the tRNA but still are able to cause thiamine independence in the presence of a purF mutation. We therefore grew phage P22 on a strain (GT3534 [zff-2521::Tn10dCm]) containing a transposon closely linked to the trmD gene and treated the phage stock with hydroxylamine. Using strain GT3670 (purF2085) as the recipient, we selected Cmr transductants on plates containing adenine, thiamine (4 x 10-4 µg/ml), and chloramphenicol, and from these transductants we selected clones that grew without added thiamine at 37°C. All 10 such thiamine-independent transductants had an m1G37-deficient tRNA Thus, no thiamine-independent mutants of strain GT3670 (purF2085) that were still able to fully methylate the tRNA were isolated. Moreover, we isolated several mutants with mutations in the trmD gene that do not influence the methylation of tRNA (35), and these mutations did not result in thiamine independence of a purF mutant . Thus, so far, all mutations in the trmD gene that result in an m1G37 deficiency also result in thiamine independence of a purF2085 mutant, whereas mutations in the trmD gene that do not result in an m1G37 deficiency do not mediate thiamine independence. These results suggest that the thiamine independence of a purF mutant is caused by m1G-deficient tRNA and not by a possible alternative function of the TrmD peptide.
fig.ommittedhm:m, 百拇医药
m1G37 deficiency induces thiamine independence in a purF2085 genetic background. The gray bars indicate the m1G/pseudouridine ratio. The level of m1G was determined by HPLC by using bulk tRNA prepared from cells grown in NCE medium containing glucose at 37°C and was expressed as the absorbance at 254 nm of m1G relative to the absorbance of pseudouridine. The open bars indicate the growth of the various mutants as determined by single-cell streaks on NCE medium plates containing glucose, adenine, and thiamine. The diameters of the colonies were measured after 48 h of incubation at 37°C. The colony size was determined by determining the average size of about five colonies in relation to the size of wild-type colonies. Thus, 0.1 indicates that the growth of the mutant was 10% of that of the wild-type strain. The solid bars indicate the colony size on glucose adenine medium as determined after 48 h of incubation at 37°C. The sizes of the colonies of the various mutants were determined relative to the size of wild-type colonies on plates containing glucose, adenine, and thiamine.
Introduction of plasmid pUMV4, which harbors the trmD orthologue from the archaeon M. vannielii, into strain GT5357 (purF2085 trmD27) suppressed the thiamine independence of this strain and concomitantly increased the level of m1G in the tRNA . Similar results were also obtained with plasmid pUSC1, which harbors the trmD orthologue from S. cerevisiae (designated TRM5 in the yeast [6]). Thus, the thiamine independence was correlated with the lack of m1G in the tRNA. Moreover, since it is very unlikely that the orthologues from an archaeon and a eukaryote also have a possible second function similar to that of the TrmD peptide of S. enterica, we concluded that it is the undermethylated tRNA and not an alternative function of the TrmD peptide that causes the thiamine-independent growth in a purF2085 genetic background.)c@5{;, http://www.100md.com
fig.ommitted)c@5{;, http://www.100md.com
trmD orthologues from M. vannielii and S. cerevisiae suppress the thiamine independence of the purF2085 trmD27 mutant, although they only partially restore the level of m1G37 in tRNAa
Evidently, there is no correlation between the degree of thiamine independence of a purF mutant and the degree of m1G37 deficiency, since even though several of the mutants isolated had no detectable m1G, they had different abilities to grow on medium lacking thiamine (compare trmD28 and trmD30 in Fig. 2). Some mutants had a measurable level of m1G (8 to 40% of the wild-type level) in the tRNA, but they were still able to grow without thiamine as efficiently as some of the mutants that had no detectable level of m1G (for example, the trmD30 mutant with no detectable m1G and the trmD3 mutant with 30% of the wild-type level of m1G grew similarly without thiamine). Moreover, although plasmids pUMV4 (trmD+ from M. vannielii) and pUSC1 (TRM5) completely inhibited the thiamine independence of a purF mutant, they only partially restored the m1G level (Table 4).6;3vp, 百拇医药
Other mutations that result in tRNA modification deficiency, reduce polypeptide growth rates, or induce +1 frameshifting do not mediate thiamine independence of a purF mutant. It has been shown previously that a lack of ms2i6A37 in the tRNA of E. coli results in slow growth and a derepressed level of the Gnd enzyme, whose activity is pivotal in the oxidative pentose pathway (31). Since a lack of m1G also induces slow growth (34), slow growth as such may result in derepressed synthesis of thiamine or an unbalanced metabolism in which thiamine is involved. However, the miaA1 mutation, which results in a lack of ms2io6A37 in tRNA, slow growth, and a reduced polypeptide elongation rate (24), did not induce thiamine independence of a purF mutant (data not shown). Since the miaA mutation induces even slower growth than the trmD3 mutation (24, 34), slow growth as such does not induce the ability to grow without thiamine. A mutation in the rpsL gene, which also reduces the growth rate, did not confer thiamine independence (data not shown). Thus, neither a tRNA modification deficiency nor slow growth as such resulted in thiamine-independent growth of a purF mutant. Apparently, an m1G37 deficiency in tRNA is unique in mediating thiamine independence of a purF mutant.
It has been shown previously that a lack of m1G37 in tRNA mediates the ability to suppress +1 frameshifting at sites that consist of runs of Cs, which are also suppressed by the sufA6 and sufB2 frameshift suppressors (8). The observed thiamine-independent growth in the presence of a purF mutation may be caused by extensive frameshifting in some mRNA(s) critical to obtain this phenotype. Since the specificity of the trmD3-mediated frameshifting and the specificity induced by the sufA6 and sufB2 frameshift suppressors are similar, we tested whether these two frameshift suppressors conferred thiamine independence in a purF2085 background. This was not the case (data not shown), suggesting that the mechanism by which the trmD3 mutation makes a purF2085 mutant independent of thiamine does not involve frameshifting at sites similar to those at which the sufA6 and sufB2 tRNAs frameshift.17h4^, 百拇医药
The trmD3-mediated thiamine independence of a purF mutant requires an active APB pathway. The PurF protein catalyzes the first step in purine synthesis, resulting in the formation of PRA, which is converted to AIR by the products of the purD, purG, and purI genes (Fig. 1). Under low-flux conditions, such as growth on gluconate, an active APB pathway is required. This pathway is blocked by mutations in the panE gene (27) (previously designated the apbA gene [15]) and by mutations in the purD, purG, or purI gene (49). The APB pathway is also blocked by mutations in the gnd and zwf genes, whose products block the oxidative pentose phosphate pathway (21). The coaA gene encodes the enzyme pantothenate kinase, which catalyzes the first dedicated step in the synthesis of CoA by forming 4'-phosphopantothenate. Mutations in this gene also block the APB pathway (26). Since mutations in the purD, purG, purI, panE, and coaA genes blocked the trmD-mediated ability of a purF mutant to grow on glucose without thiamine (data not shown), we suggest that an m1G37-deficient tRNA affects the metabolism of thiamine at some step(s) in the APB pathway.
Since the CoaA enzyme catalyzes an early step in the synthesis of CoA (Fig. 1) and since it is required for an active APB pathway (26), we also measured the levels of CoA derivatives in strains GT4219 (trmD+), GT4218 (trmD3), and GT5454 (trmD27) grown in MOPS glucose medium at 37°C. The levels of CoA derivatives were the same in trmD+ and in trmD3 strains, whereas the levels of CoA derivatives in the trmD27 mutant were somewhat reduced (76% of the wild-type level) (data not shown). Thus, the trmD mutations do not drastically influence the levels of CoA derivatives.{nv$ft, 百拇医药
We concluded that the trmD-mediated thiamine independence of a purF mutant requires an active APB pathway, including an active CoaA enzyme, suggesting that the metabolic flow downstream of the CoaA-catalyzed step and ultimately the synthesis of coenzyme A are pivotal for the trmD-mediated thiamine independence in a purF2085 genetic background.{nv$ft, 百拇医药
The trmD3-mediated thiamine independence of a purF mutant is not caused by poor translation of the purR or purE mRNAs. A mutation in the purR gene, which encodes the purine repressor, derepresses the synthesis of the purine pathway and thus the synthesis of AIR. A purE mutation blocks the conversion of AIR to purine and thereby allocates AIR to the synthesis of thiamine. Therefore, a purF purR purE triple mutant grew on glucose adenine medium without added thiamine (data not shown). One possible explanation for the observed thiamine independence of the trmD3 purF2085 mutant is poor translation of the purR and purE mRNAs, which should result in a phenotype similar to that of the triple purF purR purE mutant. However, we believe that a translational defect in the purR mRNA is unlikely, since several purR-regulated genes were not derepressed in the trmD3 mutant (data not shown). Furthermore, the ß-galactosidase activities of a PurE-LacZ protein fusion were comparable in trmD+ and trmD3 cells (data not shown). These results suggest that poor translation of the purE mRNA in conjunction with poor translation of the purR mRNA is not the cause of the trmD3-mediated thiamine independence in a purF2085 genetic background.
The trmD3-induced Thi + phenotype in a purF mutant is caused by increased synthesis of thiamine and not by a sparing effect of thiamine. The thiamine independence in a purF2085 genetic background mediated by the trmD3 mutation may result from decreased utilization of thiamine (e.g., by blocking of a reaction that requires thiamine, which saves thiamine for other thiamine-dependent reactions). However, the trmD3 mutant can sustain restreaking on plates lacking thiamine and repeated dilutions in a liquid medium, suggesting that it is synthesis of thiamine and not saving of thiamine that is influenced by the trmD3 mutation. Indeed, the level of TPP was at least threefold higher in the purF2085 trmD27 mutant than in the purF2085 trmD+ strain when they were incubated in glucose adenine medium . We concluded that trmD mutations caused increased synthesis of thiamine and not sparing of thiamine..)evn4, 百拇医药
fig.ommitted.)evn4, 百拇医药
Levels of TPP in the wild-type strain and in the trmD27 mutant. Strain GT4500 (purF2085 trmD+) was grown in MOPS glucose medium containing adenine and thiamine to an OD600 of 0.6 and then transferred to the same medium lacking thiamine. Following overnight incubation, during which the cell density did not increase, the level of TPP was determined (average of three determinations of three independent extracts). Strain GT5357 (purF2085 trmD27) was grown in MOPS glucose medium containing adenine to an OD600 of 0.6 and then harvested and examined to determine the level of TPP (average of three determinations of two independent extracts).
Some antibiotics known to reduce the polypeptide chain growth rate (CGR) also induce PurF-independent synthesis of thiamine. A codon early in a gene that is decoded slowly results in a queuing of ribosomes. Such queuing may eventually interfere with translation initiation and thereby decrease the efficiency of translation of an mRNA. If CUN (Leu), CCN (Pro), or CCG (Arg) codons were present close to the 5' end of a gene critical for PurF-independent synthesis of thiamine, the efficiency of decoding of these codons would be sensitive to the level of m1G37 in the corresponding tRNAs. If queuing of ribosomes in the beginning of some mRNA can explain the trmD3-mediated PurF-independent synthesis of thiamine, antibiotics that are known to affect CGR but are not codon specific may induce a similar phenotype. Therefore, we tested the abilities of various antibiotics which are known to affect the CGR to mediate PurF-independent synthesis of thiamine. Chloramphenicol, tetracycline, or fusidic acid but not kanamycin, streptomycin, or erythromycin allowed the purF2085 mutant to grow without thiamine (Table 5). We conclude that antibiotics known to inhibit the CGR also mediated PurF-independent synthesis of thiamine similar to the synthesis mediated by an m1G37 deficiency. Like trmD27-mediated thiamine independence, antibiotic-mediated thiamine independence required an active CoaA enzyme (data not shown), suggesting that the antibiotics affected translation of mRNAs that are involved in the same metabolism as the trmD-induced phenotype.
fig.ommitted'pp{6, 百拇医药
Abilities of different antibiotics to mediate thiamine independencea'pp{6, 百拇医药
DISCUSSION'pp{6, 百拇医药
We show here that an m1G37 deficiency in tRNA influences the metabolism of thiamine and the metabolism of pantothenate. The trmD3-mediated effect on the PurF-independent synthesis of thiamine is through the low-carbon-flux APB pathway that requires an active oxidative pentose phosphate pathway and an active pantothenate kinase (coaA gene product), which catalyzes the first dedicated step in the synthesis of CoA. The trmD-mediated PurF-independent synthesis of thiamine results from a translational effect of m1G37-deficient tRNA and not from a general aberration in translation. We concluded that some step(s) in thiamine metabolism is sensitive to the efficiency of translation of some specific mRNA(s).'pp{6, 百拇医药
There is no apparent correlation between the degree of hypomodification of bulk tRNA and the degree of PurF-independent synthesis of thiamine . In S. enterica there are seven tRNA species that contain m1G37, and any m1G37-deficient tRNA species may cause the PurF-independent thiamine synthesis observed. The tRNA(m1G37)methyltransferase can be altered in such a way that it recognizes two forms of the same tRNA species differently (35). Therefore, the degrees of hypomodification of various tRNA species may not be the same in the different mutants. A residual activity of an altered TrmD enzyme may methylate differently the critical tRNA species that mediate PurF-independent thiamine synthesis, although the level of m1G in bulk tRNA is about the same.
It has been suggested that the ribosome is the sensor for heat and cold shock responses (45). It has also been suggested that antibiotics which result in an empty A site induce a heat shock response, whereas antibiotics that fill the A site generate a cold shock response. Although all six antibiotics tested in this study reduced the growth rate and the average CGR, only three of them (chloramphenicol, tetracycline, and fusidic acid) mediated PurF-independent synthesis of thiamine . These three antibiotics result in a filled A site and induce a cold shock response. However, erythromycin, which also results in a filled A site and induces a cold shock response, did not mediate PurF-independent synthesis of thiamine . Moreover, whereas chloramphenicol, tetracycline, fusidic acid, and erythromycin reduce the level of ppGpp (25), erythromycin did not induce PurF-independent synthesis of thiamine. Accordingly, the trmD27 mutation did not alter the ppGpp level in cells grown in MOPS glucose medium (data not shown). Thus, the PurF-independent synthesis of thiamine is not correlated with some general signals from the ribosome, like the signal that induces cold shock or reduces the level of ppGpp. Interestingly, changes in the synthesis of some metabolic enzymes are triggered by various antibiotics to different degrees (25). Similarly, our results also indicate that there is specificity in the responses of antibiotics, since not all antibiotics tested induced PurF-independent synthesis of thiamine although they all reduced the growth rate and the average CGR.
The efficiency with which a tRNA species reads a certain codon is influenced by its concentration in the cell, the degree of modification, and the fraction of the tRNA population that is aminoacylated. The modified nucleosides play a pivotal role by improving the decoding capacity of the tRNA and also by decreasing the codon sensitivity, by influencing the codon choice, and by maintaining the reading frame. Moreover, the impact of a specific modified nucleoside is tRNA species dependent (for reviews see references 2,, 3, 4, and 12). Unbalanced metabolism caused by, for example, various environmental changes may induce unbalanced synthesis of tRNA, resulting in hypomethylated tRNA; e.g., limitation of leucine induces hypomodification of tRNAPhe (33; for a review see reference 4). Such a hypomethylated tRNA may in turn affect the translation of some specific mRNAs. Thus, the degree of modification of a tRNA influences the efficiency of translation and may thereby regulate various parts of the metabolism (for a review see reference 7). Thus, some environmental changes may induce m1G37 deficiency, and the hypomethylated tRNA may induce an altered flow of some specific metabolic pathway, as illustrated in this study by derepressed synthesis of thiamine. There are several ways that such an m1G37-deficient tRNA may induce altered gene expression, and two of them are discussed below.
One way that such a translational regulatory event may occur is by a low level of frameshifting caused by m1G37 deficiency, resulting in a low level of an enzyme or a regulatory protein in thiamine or pantothenate metabolism. Such frameshifting should occur at CUN, CCN, or CGG codons, since these codons are read by m1G37-containing tRNA. However, the frameshift suppressor mutations sufA6 and sufB2, which alter the m1G37-containing and , respectively, so that they frameshift at a CCC codon (42), did not induce PurF-independent synthesis of thiamine like the mutations in the trmD gene did. If frameshifting were caused by an m1G37-deficient tRNAPro at such a site, we would expect that the sufA6 and the sufB2 frameshift suppressors should mediate similar PurF-independent synthesis of thiamine. Furthermore, some antibiotics which are known to influence the amino acid step time on the ribosome caused PurF-independent synthesis of thiamine. However, they did not suppress several +1 frameshift mutations in the his operon, which are known to be suppressed by a lack of m1G37 (8) (data not shown). Therefore, we do not favor the hypothesis that a frameshifting event by an m1G37-deficient tRNAPro is responsible for the PurF-independent synthesis of thiamine. However, frameshifting by an m1G37-deficient or tRNALeu cannot be ruled out.
Another way that an m1G37-deficient tRNA may alter the efficiency of translation of an mRNA is by inducing stalling of ribosomes. If a ribosome is retarded early on an mRNA, a queuing of ribosomes occurs that eventually may inhibit translation initiation. Within the first 25 codons rare codons are preferentially used (11), and two of them, CCC (Pro) and CUA (Leu), are decoded by tRNAs having m1G37. A deficiency of m1G37 reduces the rates with which Leu tRNAs, Pro tRNAs, and Arg tRNAs decode CUU/A/G (Leu), CCN (Pro), and CGG (Arg) codons, respectively (34, 36). Thus, slow decoding of these codons early in an mRNA(s) encoding an enzyme or a regulatory protein involved in thiamine metabolism may result in reduced translation initiation and thus less expression of the protein. Even a small change in the synthesis of such an enzyme or regulatory protein may give a distinct phenotype. Indeed, there is precedence for such a suggestion: a lack of queuosine (Q34), which is present in the wobble position of a subset of tRNAs, reduces the synthesis of the positive regulatory protein VirF, and this results in reduced synthesis of proteins required for the virulence of Shigella flexneri (17, 18). This observation was somewhat surprising, since a mutant (tgt) lacking this modified nucleoside in E. coli does not show any evident phenotypic difference compared to the wild type (39). One interpretation of these observations is that there is a sequence(s) in the virF mRNA that is not present in most other mRNAs and is especially sensitive to being decoded by the Q34-deficient tRNA. Also, a lack of ms2io6A37 in S. enterica increases synthesis of the Gnd enzyme but not synthesis of the Zwf enzyme; these two enzymes are both involved in the oxidative pentose phosphate pathway (31). Thus, our results showing that a lack of m1G37 influences the metabolism of thiamine are similar to the observation that translation of the virF and gnd mRNAs may be more dependent on the status of tRNA modification than translation of other mRNAs is. A low-abundance mRNA, such as an mRNA encoding a regulatory protein or an enzyme in a low-carbon-flux pathway, may have some special requirement for proper translation and therefore may be more sensitive than other mRNAs to small but significant changes in translation efficiency or fidelity.
We hypothesize that there is a target mRNA(s) with codons read by m1G37-containing tRNAs close to the translation initiation site and that in such a codon context the mRNA(s) would be sensitive to the level of m1G37 in tRNA. The target mRNA(s) encodes an enzyme catalyzing a reaction involved in thiamine or pantothenate metabolism. A low level of m1G37 would result in a low level of this target enzyme and thereby cause a low flux in the corresponding pathways. Such a restriction in a biosynthetic pathway should result in a buildup of a metabolite(s) upstream of the restriction point, and thus the metabolite(s) may be channeled into an alternative pathway, leading to increased flux in, for example, the APB pathway. Alternatively, the target mRNA encodes a regulatory protein. A reduced level of this regulatory protein in turn induces an altered flux in the APB pathway. Interestingly, increased production of pantothenate by overexpression of the PanB enzyme slightly increases the CoA level and concomitantly mediates PurF-independent synthesis of thiamine (43). Reduced flux through the purine pathway, like the flux in a purF mutant, results in an increased requirement for CoA to synthesize thiamine (26). It has been suggested that some CoA thioesters may be involved in modifying (e.g., succenylating) an enzyme involved in the conversion of AIR to 4-amino-5-hydroxymethyl-2-methylpyrimidine. Although we did not observe any altered CoA levels in the trmD mutants, increased flux in the pantothenate-CoA pathway in the trmD mutant may still be the reason for the observed trmD-mediated thiamine independence. We suggest that the m1G37 deficiency mediates its translational effect on thiamine metabolism by acting on some step(s) following the production of CoA derivatives (e.g., in a modifying enzyme that uses a CoA thioester as a cofactor). Such an increase in flux would result in increased synthesis of thiamine.
ACKNOWLEDGMENTS0'', 百拇医药
This work was supported by the Science Research Council (project BU-2930) and by the Cancer Foundation (project 680).0'', 百拇医药
We thank Diana Downs, University of Wisconsin, Madison, for generous gifts of strains and for introducing us to thiamine metabolism. We thank Kerstin Jacobsson for excellent technical assistance with the HPLC analysis and for the use of plasmids pUMV4 and pUSC1. Gunilla Jäger is acknowledged for providing strains having the transposon zff-2521::Tn10dCm. Critical reading of the manuscript by Olof Persson and Mikael Wikström is gratefully acknowledged.0'', 百拇医药
REFERENCES0'', 百拇医药
Björk, G. R. 1986. Transfer RNA modification in different organisms. Chem. Scr. 26B:91-95.0'', 百拇医药
Björk, G. R. 1995. Genetic dissection of synthesis and function of modified nucleosides in bacterial transfer RNA. Prog. Nucleic Acid Res. Mol. Biol. 50:263-338.0'', 百拇医药
Björk, G. R. 1995. Biosynthesis and function of modified nucleosides in tRNA, p. 165-205. In D. Söll and U. L. RajBhandary (ed.), tRNA: structure, biosynthesis, and function. ASM Press, Washington, D.C.
Björk, G. R. 1996. Stable RNA modification, p. 861-886. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington.\[i|+, http://www.100md.com
Björk, G. R. 1998. Modified nucleosides in positions 34 and 37 of tRNAs and their predicted coding capacities, p. 577-581. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.\[i|+, http://www.100md.com
Björk, G. R., K. Jacobsson, K. Nilsson, M. J. O. Johansson, A. S. Byström, and O. P. Persson. 2001. A primordial tRNA modification required for the evolution of life? EMBO J. 20:231-239.\[i|+, http://www.100md.com
Björk, G. R., and T. Rasmuson. 1998. Links between tRNA modification and metabolism and modified nucleosides as tumor marker, p. 471-491. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.\[i|+, http://www.100md.com
Björk, G. R., P. M. Wikström, and A. S. Byström. 1989. Prevention of translational frameshifting by the modified nucleoside 1-methylguanosine. Science 244:986-989.
Bylund, G. O., L. C. Wipemo, L. A. C. Lundberg, and P. M. Wikström. 1998. RimM and RbfA are essential for efficient processing of 16S rRNA in Escherichia coli. J. Bacteriol. 180:73-82.|h\qo2/, http://www.100md.com
Byström, A. S., K. J. Hjalmarsson, P. M. Wikström, and G. R. Björk. 1983. The nucleotide sequence of an Escherichia coli operon containing genes for the tRNA(m1G)methyltransferase, the ribosomal proteins S16 and L19 and a 21-K polypeptide. EMBO J. 2:899-905.|h\qo2/, http://www.100md.com
Chen, G. F., and M. Inouye. 1990. Suppression of the negative effect of minor arginine codons on gene expression; preferential usage of minor codons within the first 25 codons of the Escherichia coli genes. Nucleic Acids Res. 18:1465-1473.|h\qo2/, http://www.100md.com
Curran, J. 1998. Modified nucleosides in translation, p. 493-519. In H. Grosjean and B. Benne (ed.), Modification and editing of RNA. American Society for Microbiology, Washington, D.C.|h\qo2/, http://www.100md.com
Davis, W., D. Botstein, and J. R. Roth. 1980. A manual for genetic engineering: advanced bacterial genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Downs, D. M. 1992. Evidence for a new, oxygen-regulated biosynthetic pathway for the pyrimidine moiety of thiamine in Salmonella typhimurium. J. Bacteriol. 174:1515-1521.7j][((, http://www.100md.com
Downs, D. M., and L. Petersen. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.7j][((, http://www.100md.com
Downs, D. M., and J. R. Roth. 1991. Synthesis of thiamine in Salmonella typhimurium independent of the purF function. J. Bacteriol. 173:6597-6604.7j][((, http://www.100md.com
Durand, J. M., B. Dagberg, B. E. Uhlin, and G. R. Björk. 2000. Transfer RNA modification, temperature and DNA superhelicity have a common target in the regulatory network of the virulence of Shigella flexneri: the expression of the virF gene. Mol. Microbiol. 35:924-935.7j][((, http://www.100md.com
Durand, J. M., N. Okada, T. Tobe, M. Watarai, I. Fukuda, T. Suzuki, N. Nakata, K. Komatsu, M. Yoshikawa, and C. Sasakawa. 1994. vacC, a virulence-associated chromosomal locus of Shigella flexneri, is homologous to tgt, a gene encoding tRNA-guanine transglycosylase (Tgt) of Escherichia coli K-12. J. Bacteriol. 176:4627-4634.
Durand, J. M. B., G. R. Björk, A. Kuwae, M. Yoshikawa, and C. Sasakawa. 1997. The modified nucleoside 2-methylthio-N-6-isopentenyladenosine in tRNA of Shigella flexneri is required for expression of virulence genes. J. Bacteriol. 179:5777-5782.dzc$], http://www.100md.com
Emilsson, V., and C. G. Kurland. 1990. Growth rate dependence of transfer RNA abundance in Escherichia coli. EMBO J. 9:4359-4366.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1996. Involvement of the oxidative pentose phosphate pathway in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 178:1476-1479.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1997. Mutations in sdh (succinate dehydrogenase genes) alter the thiamine requirement of Salmonella typhimurium. J. Bacteriol. 179:3989-3996.dzc$], http://www.100md.com
Enos-Berlage, J. L., and D. M. Downs. 1999. Biosynthesis of the pyrimidine moiety of thiamine independent of the PurF enzyme (phosphoribosylpyrophosphate amidotransferase) in Salmonella typhimurium: incorporation of stable isotope-labeled glycine and formate. J. Bacteriol. 181:841-848.
Ericson, J. U., and G. R. Björk. 1986. Pleiotropic effects induced by modification deficiency next to the anticodon of tRNA from Salmonella typhimurium LT2. J. Bacteriol. 166:1013-1021.\a7yfsy, 百拇医药
Evers, S., K. Di Padova, M. Meyer, H. Langen, M. Fountoulakis, W. Keck, and C. P. Gray. 2001. Mechanism-related changes in the gene transcription and protein synthesis patterns of Haemophilus influenzae after treatment with transcriptional and translational inhibitors. Proteomics 1:522-544.\a7yfsy, 百拇医药
Frodyma, M., A. Rubio, and D. M. Downs. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:236-240.\a7yfsy, 百拇医药
Frodyma, M. E., and D. Downs. 1998. The panE gene, encoding ketopantoate reductase, maps at 10 minutes and is allelic to apbA in Salmonella typhimurium. J. Bacteriol. 180:4757-4759.\a7yfsy, 百拇医药
Gehrke, C. W., and K. C. Kuo. 1990. Ribonucleoside analysis by reversed-phase high performance liquid chromatography. J. Chromatogr. Libr. 45A:A3-A71.
Gehrke, C. W., K. C. Kuo, R. A. McCune, K. O. Gerhardt, and P. F. Agris. 1982. Quantitative enzymatic hydrolysis of tRNAs: reversed-phase high-performance liquid chromatography of tRNA nucleosides. J. Chromatogr. 230:297-308.3, 百拇医药
Hagervall, T. G., T. M. Tuohy, J. F. Atkins, and G. R. Björk. 1993. Deficiency of 1-methylguanosine in tRNA from Salmonella typhimurium induces frameshifting by quadruplet translocation. J. Mol. Biol. 232:756-765.3, 百拇医药
Jones, W. R., G. J. Barcak, and R. E. Wolf, Jr. 1990. Altered growth-rate-dependent regulation of 6-phosphogluconate dehydrogenase level in hisT mutants of Salmonella typhimurium and Escherichia coli. J. Bacteriol. 172:1197-1205.3, 百拇医药
Kawasaki, T. 1986. Determination of thiamin and its phosphate esters by high-performance liquid chromatography. Methods Enzymol. 122:15-29.3, 百拇医药
Kitchingman, G. R., and M. J. Fournier. 1977. Modification-deficient transfer ribonucleic acids from relaxed control Escherichia coli: structures of the major undermodified phenylalanine and leucine transfer RNAs produced during leucine starvation. Biochemistry 16:2213-2220.
Li, J. N., and G. R. Björk. 1995. 1-Methylguanosine deficiency of tRNA influences cognate codon interaction and metabolism in Salmonella typhimurium. J. Bacteriol. 177:6593-6600.8-l$h, 百拇医药
Li, J. N., and G. R. Björk. 1999. Structural alterations of the tRNA(m1G37)methyltransferase from Salmonella typhimurium affect tRNA substrate specificity. RNA 5:395-408.8-l$h, 百拇医药
Li, J. N., B. Esberg, J. F. Curran, and G. R. Björk. 1997. Three modified nucleosides present in the anticodon stem and loop influence the in vivo aa-tRNA selection in a tRNA-dependent manner. J. Mol. Biol. 271:209-221.8-l$h, 百拇医药
Newell, P. C., and R. G. Tucker. 1967. New pyrimidine pathway involved in the biosynthesis of the pyrimidine of thiamine. Nature 215:1384-1385.8-l$h, 百拇医药
Newell, P. C., and R. G. Tucker. 1968. Precursors of the pyrimidine moiety of thiamine. A new route of pyrimidine biosynthesis involving purine intermediates. Biochem. J. 106:279-287.8-l$h, 百拇医药
Noguchi, S., Y. Nishimura, Y. Hirota, and S. Nishimura. 1982. Isolation and characterization of an Escherichia coli mutant lacking tRNA-guanine transglycosylase. Function and biosynthesis of queuosine in tRNA. J. Biol. Chem. 257:6544-6550.
Persson, B. C., C. Gustafsson, D. E. Berg, and G. R. Björk. 1992. The gene for a tRNA modifying enzyme, m5U54-methyltransferase, is essential for viability in Escherichia coli. Proc. Natl. Acad. Sci. USA 89:3995-3998.*kefdj1, 百拇医药
Petersen, L., J. Enosberlage, and D. M. Downs. 1996. Genetic analysis of metabolic crosstalk and its impact on thiamine synthesis in Salmonella typhimurium. Genetics 143:37-44.*kefdj1, 百拇医药
Qian, Q., J. N. Li, H. Zhao, T. G. Hagervall, P. J. Farabaugh, and G. R. Björk. 1998. A new model for phenotypic suppression of frameshift mutations by mutant tRNAs. Mol. Cell 1:471-482.*kefdj1, 百拇医药
Rubio, A., and D. M. Downs. 2002. Elevated levels of ketopantoate hydroxymethyltransferase (PanB) lead to a physiologically significant coenzyme A elevation in Salmonella enterica serovar Typhimurium. J. Bacteriol. 184:2827-2832.*kefdj1, 百拇医药
Sharp, P. M., and W. H. Li. 1986. Codon usage in regulatory genes in Escherichia coli does not reflect selection for 'rare' codons. Nucleic Acids Res. 14:7737-7749.
VanBogelen, R. A., and F. C. Neidhardt. 1990. Ribosomes as sensors of heat and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA 87:5589-5593.7\9eh, http://www.100md.com
Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218:97-106.7\9eh, http://www.100md.com
Wikström, P. M., A. S. Byström, and G. R. Björk. 1988. Non-autogenous control of ribosomal protein synthesis from the trmD operon in Escherichia coli. J. Mol. Biol. 203:141-152.7\9eh, http://www.100md.com
Yura, T. 1956. Evidence of nonidentical alleles in purine requiring mutants of Salmonella typhimurium. Publ. Carnegie Inst. 612:63-75.7\9eh, http://www.100md.com
Zilles, J. L., and D. M. Downs. 1996. A novel involvement of the PurG and PurI proteins in thiamine synthesis via the alternative pyrimidine biosynthetic (APB) pathway in Salmonella typhimurium. Genetics 144:883-892.(Glenn R. Björk and Kristina Nilsson)