|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 279, Issue 42, 43555-43559, October 15, 2004
A Genetic Screen for the Identification of Thiamin Metabolic Genes*
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
-hydroxyethyl)thiazole phosphate and 4-amino-2-methyl-5-hydroxymethylpyrimidine pyrophosphate (HMP-PP) (Fig. 1) (15). 5-Aminoimidazole ribonucleotide serves as the precursor to 4-amino-2-methyl-5-hydroxymethylpyrimidine phosphate (HMP-P) (6), and 4-methyl-5-(
-hydroxyethyl)thiazole phosphate is formed in a complex condensation from 1-deoxy-D-xylulose-5-phosphate, glycine, and cysteine (7). A phosphorylation catalyzed by ThiL provides the active cofactor TPP.
|
A common strategy for identifying antibiotic targets encoded in a plasmid library exploits the fact that overexpression of the target induces antibiotic resistance. For instance, recombinant cells with increased doses of murZ, the E. coli gene encoding UDP-GlcNAc enolpyruvate transferase, displayed phosphomycin resistance (8). Clones resistant to thiolactomycin were found to contain increased levels of fabB, which encodes
-ketoacyl-acyl carrier protein synthase I, the cellular target of thiolactomycin (9). Increased levels of prolipoprotein signal peptidase rendered cells resistant to globomycin, a specific inhibitor of that enzyme (10). Other selections have been used to identify proteins that could either detoxify an antibiotic or reverse the effects of the toxin (11, 12).
Our selection strategy utilizes the HMP analogs bacimethrin and4-amino-2-trifluoromethyl-5-hydroxymethylpyrimidine(CF3-HMP) (Fig. 1), both of which inhibit E. coli cell growth with a minimum inhibitory concentration (MIC) in the low micromolar range. A recent study demonstrated that the thiamin biosynthetic enzymes convert bacimethrin to 2'-methoxythiamin pyrophosphate (MeO-TPP), which inhibits thiamin-utilizing enzymes (13). In the presence of low levels of exogenous thiamin, cell growth is not inhibited by bacimethrin (14, 15). CF3-HMP is converted by HMP-P kinase (ThiD) to CF3-HMP pyrophosphate (CF3-HMP-PP), which inhibits thiamin phosphate synthase (ThiE) (16). Increased levels of biosynthesized thiamin should overcome growth inhibition caused by CF3-HMP. Enzymes that degrade TPP and its biosynthetic precursors will likely display similar activity toward MeO-TPP and CF3-HMP-PP and thereby detoxify the antibiotics. Thus, selection against bacimethrin and CF3-HMP toxicity should uncover genes involved in thiamin biosynthesis and degradation.
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
The B. subtilis genomic library was constructed by ligating fragments (
3 kb) from mechanically sheared DNA into the SacI site of pGEM-3Z. The resulting plasmid library was transformed into E. coli DH10B (Invitrogen). Selection of resistant clones and MIC determinations were carried out as described above.
Assays for Phosphatase ActivityAssay mixtures of 100 µl total volume contained 100 mM Tris-HCl, pH 8, 5 mM MgCl2, 0.4 mM HMP-PP or TPP, and 1 mg of protein/ml of cell-free extract from ymfB or cof overexpression strains (described in supplemental material). Mixtures were incubated at 37 °C for 30 min, and protein was removed by filtration through a Microcon 10 membrane (Millipore). 30 µl of the filtered reaction mixture was analyzed by HPLC on a Supelcosil LC-18-T analytical column (15 cm x 4.6 mm, 3 µm) using the following gradient: 0 min, 100% A (0.1 M K2HPO4 pH 6.6); 6 min, 100% A; 7 min, 90% A, 10% B (methanol); 20 min, 40% A, 30% B, 30% C (water). Elution was carried out at 1 ml/min and monitored at 254 nm. Under these conditions the following retention times were observed: HMP-PP (2.3 min), HMP-P (2.7 min), MeO-HMP-PP (2.8 min), MeO-HMP-P (4.0 min), CF3-HMP-PP (4.2 min), 4-amino-2-trifluoromethyl-5-hydroxymethylpyrimidine phosphate (6.6 min), TPP (9.3 min), thiamin monophosphate (11.7 min), HMP (12.7 min), thiamin (12.9 min), MeO-TPP (12.1 min), and 2'-methoxythiamin monophosphate (12.5 min). Preparation of the substrates is described in supplemental material. For competition experiments to determine relative rates of hydrolysis, assays were carried out as described above, except equimolar amounts (
0.4 mM each) of the two substrates were used in the assay. (V1/K1)/(V2/K2) was calculated as the ratio of products at 525% conversion divided by the ratio of substrates at 0% conversion.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
Functional Analysis of yajO, yjjX, and yveNWe suspected that YajO, YjjX, and YveN might also degrade TPP or its related metabolites. However, HPLC analysis of mixtures containing YajO, YjjX, and YveN incubated with TPP, thiamin monophosphate, thiamin, HMP-PP, HMP-P, or HMP indicated that no reaction had occurred. As the biosynthesis of HMP is not biochemically well defined, we considered that YajO, YjjX, and YveN may be involved in HMP formation. However, when these proteins were added to our recently developed in vitro HMP biosynthesis reaction mixture, no effect was observed (26).
| DISCUSSION |
|---|
|
|
|---|
We reasoned that increased levels of ymfB must confer resistance to bacimethrin and CF3-HMP because it is intact on plasmids 1, 2, and 3. thiC must be responsible for the antibiotic resistance observed in clones 4, 5, 6, 7, 8, and 9 because it is the only intact gene present on the plasmids carried by clones 4, 8, and 9. yjjX is likely the gene responsible for bacimethrin and CF3-HMP resistance in clones 10 and 11 because trpR (Trp regulator) and gpmB (phosphoglycerate mutase) are unlikely to be involved in thiamin biosynthesis or in the detoxification of these antibiotics. For clone 12, cof is most likely the gene responsible for resistance because cof is up-regulated in response to CF3-HMP.2 In clones 13 and 14, either dxs or yajO could be the gene responsible for the observed antibiotic resistance. thiE is the cellular target for CF3-HMP, which explains the behavior of clone 15. Finally, yveN is the only intact gene of clone 16, and thus must be responsible for the bacimethrin resistance of this clone. This analysis was confirmed by demonstrating the increased antibiotic resistance of the thiC, thiE, dxs, yjjX, yajO, cof, ymfB, and yveN overexpression strains.
We predicted that our screen would uncover genes involved in thiamin biosynthesis. The identification of the known thiamin biosynthetic genes thiC, thiE, and dxs on 9 of the isolated clones supported this hypothesis. An alternative possibility was that resistance could occur by detoxifying bacimethrin and CF3-HMP. Because ymfB and cof both contain hydrolase motifs, we considered the possibility that the detoxification of bacimethrin and CF3-HMP might occur by the hydrolysis of their phosphorylated metabolites. This hypothesis was confirmed by demonstrating that YmfB and Cof catalyzed the hydrolysis of MeO-HMP-PP and CF3-HMP-PP to give MeO-HMP-P and 4-amino-2-trifluoromethyl-5-hydroxymethylpyrimidine phosphate. This hydrolysis generates resistance to the antibiotics by reducing the formation of their toxic forms, MeO-TPP and CF3-HMP-PP (13, 16). Cof and YmfB selectively hydrolyze MeO-HMP-PP over HMP-PP with selectivities of 2.8 and 4.5, respectively, and Cof and YmfB hydrolyze CF3-HMP-PP over HMP-PP with selectivities of 2.2 and 2.3, respectively. It is notable that YmfB also catalyzes the hydrolysis of TPP and MeO-TPP. MeO-TPP hydrolysis is much slower than the hydrolysis of MeO-HMP-PP, and there is no selectivity for MeO-TPP hydrolysis over TPP hydrolysis.
The mechanisms by which yjjX, yajO, and yveN elicit resistance to bacimethrin and CF3-HMP are yet to be determined. YajO, YjjX, and YveN do not affect the biosynthesis of HMP. As the remaining steps of bacterial thiamin biosynthesis are biochemically well defined, it is unlikely that YajO, YjjX, or YveN are thiamin biosynthetic enzymes. In addition, these proteins have no apparent activity toward thiamin or its biosynthetic precursors. It is possible that these proteins are cellular targets of bacimethrin and CF3-HMP or that they promote efflux of the antibiotics (811).
The genetic screen described here is based on the resistance of clones isolated from E. coli and B. subtilis genomic libraries to antibiotics that inhibit thiamin biosynthesis and utilization. This is a powerful approach that resulted in the isolation of five genes involved in thiamin metabolism. Three of these genes (thiC, thiE, dxs) had been previously identified as thiamin biosynthetic genes (2). Two new genes (cof, ymfB) are involved in the hydrolysis of HMP-PP and TPP. Three additional genes (yajO, yjjX, yveN) that were isolated are still without functional assignment. Provided suitable inhibition can be identified, this approach should be generally useful for the identification of natural product biosynthetic genes (812).
Until now, it has been assumed that cellular TPP concentration is controlled by thiamin-mediated repression of the translation of key thiamin biosynthetic genes (2, 2729). However, our studies on Cof and YmfB demonstrate that E. coli contains enzymes able to catalyze the hydrolysis of TPP and HMP-PP. The physiological significance of these results is undetermined, and the possibility that Cof and YmfB can catalyze the hydrolysis of other metabolites cannot be excluded. Our results have allowed us to assign a possible function to two genes of previously unassigned function, and the results suggest that degradation of TPP and its biosynthetic precursors may play an important role in controlling intracellular concentrations of this cofactor.
| FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains additional Experimental Procedures. ![]()
¶ To whom correspondence should be addressed. Tel.: 607-255-7133; E-mail: tpb2{at}cornell.edu.
1 The abbreviations used are: TPP, thiamin pyrophosphate; HMP-PP, 4-amino-2-methyl 5-hydroxymethylpyrimidine pyrophosphate; HMP-P, 4-amino-2-methyl 5-hydroxymethylpyrimidine phosphate; HMP, 4-amino-2-methyl 5-hydroxymethylpyrimidine; CF3-HMP, 4-amino-2-trifluoromethyl 5-hydroxymethylpyrimidine; MeO-TPP, 2'-methoxythiamin pyrophosphate; CF3-HMP-PP, 4-amino-2-trifluoromethyl 5-hydroxymethylpyrimidine pyrophosphate; MeO-HMP-PP, 4-amino-2-methoxy-5-hydroxymethylpyrimidine pyrophosphate; HPLC, high pressure liquid chromatography; MIC, minimum inhibitory concentration. ![]()
2 J. Perkins and T. P. Begley, unpublished data. ![]()
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Ogawa, K. Yoshimura, H. Miyake, K. Ishikawa, D. Ito, N. Tanabe, and S. Shigeoka Molecular Characterization of Organelle-Type Nudix Hydrolases in Arabidopsis Plant Physiology, November 1, 2008; 148(3): 1412 - 1424. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. B. Thomaides, E. J. Davison, L. Burston, H. Johnson, D. R. Brown, A. C. Hunt, J. Errington, and L. Czaplewski Essential Bacterial Functions Encoded by Gene Pairs J. Bacteriol., January 15, 2007; 189(2): 591 - 602. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Karunakaran, K. Ebert, S. Harvey, M. E. Leonard, V. Ramachandran, and P. S. Poole Thiamine Is Synthesized by a Salvage Pathway in Rhizobium leguminosarum bv. viciae Strain 3841. J. Bacteriol., September 1, 2006; 188(18): 6661 - 6668. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Schyns, S. Potot, Y. Geng, T. M. Barbosa, A. Henriques, and J. B. Perkins Isolation and Characterization of New Thiamine-Deregulated Mutants of Bacillus subtilis J. Bacteriol., December 1, 2005; 187(23): 8127 - 8136. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |