SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Despite over a century of research, tuberculosis remains a leading cause of infectious death worldwide. Faced with increasing rates of drug resistance, the identification of genes that are required for the growth of this organism should provide new targets for the design of antimycobacterial agents. Here, we describe the use of transposon site hybridization (TraSH) to comprehensively identify the genes required by the causative agent, Mycobacterium tuberculosis, for optimal growth. These genes include those that can be assigned to essential pathways as well as many of unknown function. The genes important for the growth of M. tuberculosis are largely conserved in the degenerate genome of the leprosy bacillus, Mycobacterium leprae, indicating that non-essential functions have been selectively lost since this bacterium diverged from other mycobacteria. In contrast, a surprisingly high proportion of these genes lack identifiable orthologues in other bacteria, suggesting that the minimal gene set required for survival varies greatly between organisms with different evolutionary histories.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Tuberculosis remains a major public health concern despite the availability of effective chemotherapy. This is largely a result of the tenacity of the causative organism itself, as the treatment of a Mycobacterium tuberculosis infection requires many months of therapy with multiple drugs. This extensive course of treatment results in poor compliance and the emergence of multidrug-resistant strains, which represent an increasing fraction of TB cases in much of the world. Whereas several antibiotics are effective in treating mycobacterial infections, these drugs target a surprisingly small number of essential functions in the cell. Therefore, the identification of the pathways that are required for mycobacterial growth would provide many new targets for the rational design of more effective antimycobacterial agents that could be active against drug-resistant strains.

The genome sequence of an organism provides a list of all of the encoded genes, which includes all potential drug targets. However, a large proportion of the predicted genes, ∼ 40% in the case of M. tuberculosis (Cole et al., 1998), have no known function. Although many of these genes undoubtedly perform essential functions, the large-scale identification of genes that are required for growth is technically difficult and has been attempted for only a small number of organisms. Although the methods differed, in each case a library of bacterial mutants was generated and strains with growth defects were identified. In the case of Haemophillus influenzae, a library of transposon mutants was generated, and the location of insertions was determined by PCR-mapping (Akerley et al., 2002). Alternatively, antisense inhibition of gene expression was used in Staphylococcus aureus, allowing the identification of conditionally lethal phenotypes (Ji et al., 2001).

Although it is tempting to assume that essential pathways are universally conserved, it is not clear whether this information can be generalized between structurally and metabolically distinct organisms. We have therefore undertaken an exhaustive search for the genes that are required for the growth of mycobacteria. Many of these genes can be assigned to essential pathways, providing insight into the structural and metabolic requirements of mycobacteria. Interestingly, these requirements seem to be quite different from H. influenzae, demonstrating that the minimal set of genes required for life differs greatly between bacterial species. This represents both a challenge and an opportunity. Although it appears that these essential pathways will have to be defined individually for different pathogens, this information should be very useful in the development of more specific antimicrobial agents. In addition to previously characterized pathways, we also identified many genes with no known function. These genes presumably participate in undefined processes that are essential for cellular growth, and this information greatly expands the set of functions that can be targeted for drug design.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Construction of transposon mutant libraries

In order to identify genes that are required for the optimal growth of mycobacteria, we constructed large and diverse libraries of transposon insertion mutants in M. tuberculosis H37Rv and M. bovis BCG Pasteur, a closely related vaccine strain, using a Himar1-based transposon delivered by a transducing bacteriophage. After mutagenesis, bacteria were plated on defined media. Each surviving mutant strain contained one copy of the transposon integrated into the chromosome (Sassetti et al., 2001), and previous work suggests that transposition is relatively random aside from an absolute requirement for the dinucleotide TA (Rubin et al., 1999). Each mutant library contains ∼100 000 independent clones. As there are ∼ 4000 predicted open reading frames (orfs) we expect that mutagenesis is close to saturating, and that, without selective pressure, all genes with TA dinucleotides should contain insertions. However, insertions in genes that are required for growth or survival should be significantly under-represented in the population of mutants.

Use of TraSH to identify genes required in vitro

To identify these genes that are unable to sustain insertions, we used transposon site hybridization (TraSH). Transposon site hybridization can be used to analyse large pools of mutants by using a DNA microarray to detect genes that contain insertions (Sassetti et al., 2001). Spotted microarrays and therefore the TraSH method, are best suited for comparing two populations. As the characteristics of hybridization vary at each spot, depending on the sequence and the amount of probe, it is difficult to determine the absolute presence or absence of fluorescence signal corresponding to any given orf.

To circumvent this problem we used randomly labelled genomic DNA to control for hybridization of each orf. Transposon site hybridization probes, which hybridize to the region adjacent to each transposon insertion, were constructed from the pool of in vitro grown cells and labelled with a fluorophore (insertion probe). A second probe population was prepared from genomic DNA and labelled with another fluorophore (genomic probe). These were mixed and hybridized to a DNA microarray consisting of PCR products representing fragments of each of the predicted M. tuberculosis orfs (see Fig. 1). As expected, the absolute intensities of fluorescent signals produced by hybridizing the genomic probe were normally distributed (not shown). However, the ratios produced by dividing the signal from the insertion pool by that of the genomic pool for each gene showed a skewed distribution (Fig. 2), resulting from the presence of a population of orfs containing a smaller than expected number of transposon insertions, i.e. mutations that produce in vitro growth defects. This population should be under-represented among randomly chosen clones. Indeed, sequencing of the transposon insertion site of 101 random mutants revealed that the disrupted genes were represented in the larger normally distributed peak in the TraSH data (Fig. 2), demonstrating that TraSH provides an accurate reflection of the composition of the pool.

image

Figure 1. Protocol used to identify genes required for optimal growth. Immediately after mutagenesis, each mutant contains a single transposon insertion (triangles), and the library contains mutations in each gene (lettered boxes) in the genome. After a growth phase, mutants harbouring insertions in genes that are required for survival (grey boxes) are lost from the library. TraSH probe is generated from the selected library, and this ‘insertion probe’ consists only of sequences complementary to genes that contain insertions in the selected library. Randomly labelled chromosomal DNA (genomic probe) will hybridize to every gene represented on the array. Spots that hybridize to the genomic probe, but not the insertion probe represent genes that are required for growth.

Download figure to PowerPoint

image

Figure 2. Distribution of transposon insertions detected by transposon site hybridization (TraSH). Histogram plot of the ratio of insertion probe to genomic probe for each gene of M. tuberculosis (black line). Data is the average of M. tuberculosis and BCG libraries, each analysed twice. Red line indicates the data for genes that were disrupted in randomly isolated mutants for which insertion sites were determined by sequencing. 0.2-fold cutoff is indicated with a dashed line.

Download figure to PowerPoint

In order to define the genes that impair growth when disrupted, we chose 12 genes that have essential functions in almost all other bacteria (annotated on Table S1). All that had interpretable results had ratios of insertion/genomic probe signals that placed them below the normally distributed genes (not shown). We found that all had ratios below 0.2. As a ratio of 0.2 also excluded the variability due to differences in transposition frequency (see below), we chose this value as a cutoff for growth attenuation. For genes to be considered essential for optimal growth, we required the corresponding ratios to be below 0.2 and to be reproducible in both M. tuberculosis and M. bovis BCG libraries (each library was analysed twice, and mutants were considered to be reproducibly attenuated if the average of the resulting ratios was significantly different from 1.0 with P < 0.05 by t-test). Genes that lacked TA dinucleotides (five genes) and those that are known to be deleted in M. bovis BCG (110 genes) were excluded from analysis. A total of 614 genes met these criteria (Table S1).

Table 1. . Genes that are required for growth can sustain insertions on cloned DNA. Ratios (insertion pool/genomic pool) for genes that are predicted to be required for growth and are present on the mutagenized cosmid clones.
Gene nameMutant library ratioCosmid library ratio
purE 0.144.87
pcnA 0.173.56
ispA 0.136.99
accA3 0.081.53
Rv2611c 0.071.13
ppiB 0.042.20
hisS 0.100.60
ribF 0.181.09

TraSH analysis of insertions in cloned DNA

The relative lack of transposon insertions in the identified set of genes could be either due to transposon specificity or because these genes are required for optimal growth. To exclude the former possibility, we randomly chose seven cosmid clones from a M. tuberculosis genomic DNA library and used the mycobacterial transposon to perform in vitro mutagenesis (Lampe et al., 1996) Pools from the cosmids, containing insertions at the same density as the larger libraries, were used to prepare insertion probes. These were mixed with a genomic probe and analysed on the DNA microarray. For the 232 genes encoded on the cosmids, the insertion/genomic probe intensity ratios were normally distributed (Fig. 3). Among the genes on these cosmids were eight that appear to produce growth attenuation when mutated in both M. tuberculosis and M. bovis BCG (Table 1). On cosmids, these genes sustained insertions at a frequency indistinguishable from other genes. Applying the 0.2-fold cut-off used above to the cosmid data results in the exclusion of> 99% of the genes in this experiment. This provides a rough estimate of ∼ 1% as the rate at which genes are falsely determined to be required for growth using these criteria suggesting that ∼ 40 genes may be incorrectly classified.

image

Figure 3. Histogram plot of ratios derived from cloned M. tuberculosis chromosomal DNA fragments after in vitro mutagenesis. Data is the average of replicate experiments. Blue boxes indicate genes that that fall below the 0.2-fold cut-off in Fig. 2 and are predicted to be required for growth.

Download figure to PowerPoint

Organization of genes that are required for growth

The genes that we have identified as important for growth are distributed throughout the chromosome. In fact, the largest segment of the chromosome that does not contain one of these genes is only ∼115 kilobases. A closer inspection of the distribution reveals that many of these genes are clustered in operons and presumably co-transcribed. It is known that a transposon insertion near the beginning of an operon can attenuate the expression of downstream genes (Berg et al., 1980) a phenomenon known as polarity. Therefore, these polar effects could cause the phenotypes of downstream genes to be attributed to those encoded before them. This could lead to a higher proportion of genes in operons appearing to be required for growth, particularly near the 5′ end of the operon. To address the effect of polarity on our results, 21 operons that contain genes predicted to be required for growth were analysed. The number of genes identified by TraSH in either the 5′ or 3′ halves of these operons were determined. This analysis did not detect any bias toward the 5′ end of operons (in fact, a slight bias toward the 3′ end was observed) indicating that polarity has a negligible effect on our results. This is probably caused by the presence of a highly active promoter in the transposon (Rubin et al., 1999) that is able to drive the expression of downstream genes at a sufficient level to prevent polar effects.

Metabolic roles of genes required for optimal growth

The genes required for optimal growth include many predicted to play central roles in metabolism. Using pathway predictions we identified several central biosynthetic and metabolic pathways in which the majority of steps were mediated by genes that we predict to be required for growth (Table 2). These pathways include those required for synthesis of amino acids, nucleic acids and co-factors. The vast majority of genes that play roles in other central cellular processes, including replication, transcription, protein synthesis and cell division also produced growth attenuation when mutated. In addition to these genes in well-defined pathways, a large number of the identified genes have no clearly defined function. For example, a total of 211 genes lack systematic names and have only Rv number designations.

Table 2. . Essential pathways in M. tuberculosis and M. bovis BCG. Essential steps are those that are performed by genes identified as being required for optimal growth.
PathwayEssential steps/total steps
Alanine biosynthesis 2/2
Arginine biosynthesis 9/9
Asparagine biosynthesis 1/1
Aspartate biosynthesis 2/2
Chorismate biosynthesis 5/6
Cysteine biosynthesis 2/2
dTDP-rhamnose biosynthesis 3/4
Folate biosynthesis 5/9
Glutamate degradation 2/2
Glycine degradation 3/3
Haeme biosynthesis 7/10
Histidine biosynthesis 6/6
Homoserine biosynthesis 3/3
Homoserine/Methionine biosynthesis 2/3
Isoleucine biosynthesis 3/4
Leucine biosynthesis 3/4
Lysine and diaminopimelate synthesis 6/7
Mannose and GDP-mannose metabolism 2/3
Non-oxidative pentose phosphate pathway 3/4
Panthothenate and coenzyme A biosynthesis 3/5
Peptidoglycan synthesis 7/10
Proline biosynthesis 2/3
Proline utilization 2/2
PRPP biosynthesis 3/3
Purine biosynthesis10/13
Pyridoxal 5′phosphate biosynthesis 2/2
Pyrimidine biosynthesis 3/6
Riboflavin, FMN and FAD biosynthesis 3/3
Serine biosynthesis 2/2
Sulphur-containing amino acid metabolism 3/4
Thiamine biosynthesis 4/4
Threonine biosynthesis from homoserine 2/2
Trehalose anabolism 3/3
Tryptophan biosynthesis 4/4
Valine biosynthesis 2/3

Intermediate growth rate mutants

Many mutants may display mild growth attenuation that is not severe enough to be identified in the experiments described above. To identify these genes, we subjected the M. tuberculosis mutant library to a second round of growth. Mutants that grow relatively slowly should be under-represented in the second growth pool as opposed to the initial pool. Indeed, we found 42 genes with detectable insertions in the initial library that, when regrown, produced signals that met our criteria for growth attenuation (Fig. 4, Table S2). Included in these genes is the mbtB gene encoding part of the non-ribosomal peptide synthase responsible for the production of the siderophore, mycobactin (Quadri et al., 1998). A closer analysis of the mycobactin gene cluster revealed that the ratios for six of the eight mbt genes decrease by 2.9- to 13.5-fold after the second growth step. Thus, although previous work has shown that mbtB mutants grow normally in the short term under iron-replete conditions (De Voss et al., 2000), our work indicates that, over a longer period of growth, a minor defect can be detected.

image

Figure 4. Mutants with a slow growth phenotype are defined as those with an average ratio of> 0.4 (all measurements> 0.2) after one plating, and after an additional growth phase, meet the criteria for growth attenuation (an average ratio of < 0.2 and P-value < 0.05 by t-test). Pools from each growth phase were analysed twice and the average of these measurements is shown.

Download figure to PowerPoint

Evolutionary conservation of genes required for in vitro growth

The genome of the leprosy bacillus, Mycobacterium leprae, has undergone dramatic reductive evolution since diverging from other mycobacterial species. As a result, only 40% of the genes of M. tuberculosis have functional orthologues in M. leprae (Cole et al., 2001). We found that the majority of the genes that we predict to be required for the optimal growth of M. tuberculosis (78%) have been retained in the M. leprae genome. Genes that lack systematic names were almost as likely to have orthologues in M. leprae as those with annotated functions. In contrast, only 27% of genes that are not required for in vitro growth have M. leprae orthologues. This set of non-essential genes was defined as those that produced an insertion/genomic probe ratio of> 0.2 in all experiments (Table S3). Thus, M. leprae appears to have selectively conserved the majority of genes that are necessary for optimal growth.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Transposon mutagenesis has been used extensively to determine the importance of bacterial genes in various biological processes. Here we have used transposons to screen for genes important for growth by finding those genes that cannot sustain transposon insertions. We find that approximately one-sixth of all predicted orfs meet stringent criteria for being required for optimal growth. Whereas many are known and are predicted to encode products vital to cellular metabolism, many have no known function.

Transposon site hybridization is simply a screening tool, and several technical considerations limit our ability to define the growth requirement for any individual gene. First, insertions in non-essential regions either in or adjacent to essential genes might produce TraSH probes that are still able to hybridize, a particular problem for small genes. Second, transposition might have more site specificity in vivo (as in the mycobacterial mutant libraries) than in vitro (as in the cosmid experiment), and this could lead to an overestimate of the true number of genes that are required for growth. Conversely, our use of stringent criteria to avoid false positives likely excluded some mutants with actual growth defects. Indeed, several genes that have been reported previously to be essential for growth did not meet our criteria for growth attenuation. Some genes, such as ideR (Rodriguez et al., 2002), rmlD (Ma et al., 2002), and whiB2 (Gomez and Bishai, 2000) did not meet the ratio cutoff. Others, such as pimA (Kordulákováet al., 2002) and mtrA (Zahrt and Deretic, 2000), were excluded despite ratios below the cutoff because the data was not sufficiently reproducible between experiments (or libraries). Finally, our results are limited to growth on a defined medium. Further experiments using varied growth conditions should be useful for identifying genes that are required for growth under other conditions, and this in turn should provide functional clues for many uncharacterized mycobacterial genes.

Lesions in many different biochemical pathways are known to affect growth rate. Many of these pathways, such as protein and cell wall biosynthesis, are clearly vital to cellular survival. While we predicted that the majority of the genes in these pathways would be required for growth, many appeared to be dispensable. In some cases this is due to technical limitations (as discussed above), but in several others these observations reflect physiological complexity. In purine biosynthesis, for example, the products of the purT and purN genes provide alternative mechanisms for the formylation of 5-phosphoribosylglycineamide (Zalkin and Nygaard, 1996). As either of these genes is sufficient to perform this function, neither is apparently required. Similarly, both the serB and serB2 genes are predicted to perform the same function in serine, glycine and cysteine biosynthesis (Karp and Romero, 2002; http:www.biocyc.org). However, in this case we found that mutations in serB2, but not serB impeded growth, suggesting that the serB2 gene encodes an essential phosphoserine phosphatase, and the serB gene cannot compensate for the absence of this activity.

Why should genes required in vitro be of interest? After all, M. tuberculosis is an obligate pathogen that appears to have no significant environmental niche. However, the vast majority of genes required in vitro are probably required during infection. In addition, these genes make much more practical targets than those specifically required for infection, as screening for antibacterial compounds and testing for resistance is much simpler in vitro. Largely as a result of this, genes required for in vitro growth represent almost all of the targets for current antibiotics.

Mycobacterium leprae is a close relative of M. tuberculosis, with few unique genes of its own. Mycobacterium leprae grows remarkably slowly in vivo and cannot be cultured in vitro at all. It is likely that the loss of several pathways that we have identified as important for the growth of M. tuberculosis is responsible for the growth characteristics of M. leprae. For example, analysis of the M. leprae genome found defects in sulphur acquisition and reduction, and the synthesis of mycobactin, and cobalamin (Cole et al., 2001). We have identified these pathways, as well as proline degradation, and molybdopterin synthesis, as important for the growth of M. tuberculosis but absent from the M. leprae genome.

The Gram-negative bacterium Haemophilus influenzae appears to have very different growth requirements than mycobacteria, as it only encodes orthologues of 47% of genes required for mycobacterial growth. For example, we find that genes encoding enzymes required for rhamnose and lipid metabolism, critical for the mycobacterial cell wall (Brennan and Vissa, 2001), are required by mycobacteria but lack orthologues in H. influenzae. In addition, many amino acid biosynthetic genes are required by mycobacteria regardless of culture condition (Sassetti et al., 2001), as these bacteria lack the ability to efficiently transport some exogenous compounds from the medium (Pavelka and Jacobs, 1999). As H. influenzae is able to acquire these metabolites from the environment, several amino acids biosynthetic pathways, as well as those involved in thiamine, haeme, and NAD synthesis, are dispensable (Klein and Luginbuhl, 1979; Akerley et al., 2002).

Recently, a large number of the genes that are required for growth in several different organisms have been defined (Arigoni et al., 1998; Hutchison et al., 1999; Judson and Mekalanos, 2000; Chalker et al., 2001; Ji et al., 2001; Akerley et al., 2002), and the concept of a universal ‘minimal genome’ that is required for life has developed. Our analysis suggests that this concept needs to be reconsidered. With the gain and loss of functions through evolution, different bacteria have acquired a variety of metabolic and structural requirements. Thus, although genes that are essential in one organism are likely to be conserved in closely related species, they are much less likely to be required by more distant organisms. This has important practical consequences. For example, all antimicrobials target functions required for bacterial growth, and only a few of the potential targets are exploited by current antibiotics. Comparing genes required for growth in different bacteria allows the identification of targets that would provide relatively narrow therapeutic spectrums. Such focused antibiotics could be advantageous as they decrease the likelihood of selection for broad antibiotic resistance.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Identification of genes required for optimal growth

Library generation, bacterial culture, and TraSH analysis were performed essentially as described (Sassetti et al., 2001). Briefly, transposon libraries were generated using the transducing phage, phiMycoMarT7, and ∼100 000 mutants were grown on 7H10 agar (Difco) containing AD (0.5% bovine serum albumin, 0.2% glucose, 0.085% NaCl), for BCG-pasteur, or OADC (Becton Dickinson), for M. tuberculosis H37Rv. To identify mutants with minor growth defects, the M. tuberculosis library was collected by scraping plates, frozen at − 80°C in 15% glycerol, titred, and replated. In each case, chromosomal DNA was isolated from the surviving clones and TraSH probe (insertion probe) was generated. Labelling with aminoallyl dUTP was performed by reverse transcription from the adapter primer for the insertion probe, or with Klenow using random primers for M. tuberculosis chromosomal DNA (genomic probe), and NHS-esters of Cy3 or Cy5 were conjugated as described at http:cmgm.stanford.edupbrownprotocolsindex.html. The resulting probes were hybridized to the M. tuberculosis microarray. Data was collected using a GenePix4000 scanner (Axon) and analysed using GeneSpring software (Silicon Genetics). Each experiment was performed twice, and each microarray was printed in duplicate. Therefore, the data presented represents the average of four data points for each gene.

Mutagenesis of cosmids

A cosmid library of Sau3AI digested chromosomal DNA from M. tuberculosis strain H37Rv was made by standard cosmid cloning techniques. Plasmid was isolated from seven random clones, and mutagenized in vitro (Lampe et al., 1996; 1999) with a derivative of the MycoMarT7 transposon that carries a chloramphenicol resistance marker. Mutagenized cosmids were recovered by transformation into ElectroTen Blue cells (Stratagene) and selection for chloramphenicol resistance. Plasmid was pooled from each transposition reaction (∼2000 independent clones each), digested with PacI (to excise inserts from the vector), and TraSH probe was generated as above.

Genome comparisons

Functional pathways were identified using the BioCyc server (http:www.biocyc.org;Karp and Romero, 2002). Orthologous genes were identified using the MBGD server (http://mbgd.genome.ad.jp).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

We thank David Lampe for providing purified Himar1 transposase, the Motorola Academic Platform Program for slides, and Jyothi Rengarajan and members of the Rubin lab for helpful advice. This work was supported in part by NIH grants AI48704 and AI51929. C.M.S. is supported by the Cancer Research Fund of the Damon Runyon-Walter Winchell Foundation, DRG-1647.

Supplementary material

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Table S1. Genes required for optimal growth. Data from replicate experiments were analyzed using GeneSpring software. Ratios (insertion probe/genomic probe) and fluorescence intensities (‘insertion’ and ‘genomic’) are provided along with minimum and maximum values.

Table S2. Genes that produce slow growth when mutated. Data are the average of replicate experiments. Columns are labelled as in Table S1.

Table S3. Non-essential genes. Data are the average of replicate experiments. Columns are labelled as in Table S1.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information
  • Akerley, B.J., Rubin, E.J., Novick, V.L., Amaya, K., Judson, N., and Mekalanos, J.J. (2002) A genome-scale analysis for identification of genes required for growth or survival of Haemophilus influenzae. Proc Natl Acad Sci USA 99: 966971.
  • Arigoni, F., Talabot, F., Peitsch, M., Edgerton, M.D., Meldrum, E., Allet, E., et al. (1998) A genome-based approach for the identification of essential bacterial genes. Nat Biotechnol 16: 851856.
  • Berg, D.E., Weiss, A., and Crossland, L. (1980) Polarity of Tn5 insertion mutations in Escherichia coli. J Bacteriol 142: 439446.
  • Brennan, P.J., and Vissa, V.D. (2001) Genomic evidence for the retention of the essential mycobacterial cell wall in the otherwise defective Mycobacterium leprae. Lepr Rev 72: 415428.
  • Chalker, A.F., Minehart, H.W., Hughes, N.J., Koretke, K.K., Lonetto, M.A., Brinkman, K.K., et al. (2001) Systematic identification of selective essential genes in Helicobacter pylori by genome prioritization and allelic replacement mutagenesis. J Bacteriol 183: 12591268.
  • Cole, S.T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., et al. (1998) Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393: 537544.
  • Cole, S.T., Eiglmeier, K., Parkhill, J., James, K.D., Thomson, N.R., Wheeler, P.R., et al. (2001) Massive gene decay in the leprosy bacillus. Nature 409: 10071011.
  • De Voss, J.J., Rutter, K., Schroeder, B.G., Su, H., Zhu, Y., Barry, C.E., 3rd (2000) The salicylate-derived mycobactin siderophores of Mycobacterium tuberculosis are essential for growth in macrophages. Proc Natl Acad Sci USA 97: 12521257.
  • Gomez, J.E., and Bishai, W.R. (2000) whmD is an essential mycobacterial gene required for proper septation and cell division. Proc Nat Acad Sci USA 97: 85548559.
  • Hutchison, C.A., Peterson, S.N., Gill, S.R., Cline, R.T., White, O., Fraser, C.M., et al. (1999) Global transposon mutagenesis and a minimal Mycoplasma genome. Science 286: 21652169.
  • Korduláková, J.M.G., Mikusová K., Puzo G., Brennan, P.J., Gicquel, B., and Jackson, M. (2002) Definition of the first mannosylation step in phosphatidylinositol mannoside synthesis. PimA is essential for growth of mycobacteria. J Biol Chem 277: 3133531344.
  • Ji, Y., Van Zhang, B.S.F., Horn, W.P., Woodnutt, G., Burnham, M.K., and Rosenberg, M. (2001) Identification of critical Staphylococcal genes using conditional phenotypes generated by antisense RNA. Science 293: 22662269.
  • Judson, N., and Mekalanos, J.J. (2000) TnAraOut, a transposon-based approach to identify and characterize essential bacterial genes. Nat Biotechnol 18: 740745.
  • Karp, S.P., and Romero, P. (2002) The pathway tools software. Bioinformatics 18(Suppl 1): S225S232.
  • Klein, R.D., and Luginbuhl, G.H. (1979) Simplified media for the growth of Haemophilus influenzae from clinical and normal flora sources. J Gen Microbiol 113: 409411.
  • Lampe, D.J., Akerley, B.J., Rubin, E.J., Mekalanos, J.J., and Robertson, H.M. (1999) Hyperactive transposase mutants of the Himar1 mariner transposon. Proc Natl Acad Sci USA 96: 1142811433.
  • Lampe, D.J., Churchill, M.E., and Robertson, H.M. (1996) A purified mariner transposase is sufficient to mediate transposition in vitro. EMBO J 15: 54705479.
  • Ma, Y., Pan, F., and McNeil, M. (2002) Formation of dTDP-rhamnose is essential for growth of mycobacteria. J Bacteriol 184: 33923395.
  • Pavelka, M.S. Jr and Jacobs, W.R. Jr (1999) Comparison of the construction of unmarked deletion mutations in Mycobacterium smegmatis, Mycobacterium bovis bacillus Calmette-Guerin, and Mycobacterium tuberculosis H37Rv by allelic exchange. J Bacteriol 181: 47804789.
  • Quadri, L.E., Sello, J., Keating, T.A., Weinreb, P.H., and Walsh, C.T. (1998) Identification of a Mycobacterium tuberculosis gene cluster encoding the biosynthetic enzymes for assembly of the virulence-conferring siderophore mycobactin. Chem Biol 5: 631645.
  • Rubin, E.J., Akerley, B.J., Novik, V.N., Lampe, D.J., Husson, R.N., and Mekalanos, J.J. (1999) In vivo transposition of mariner-based elements in enteric bacteria and mycobacteria. Proc Natl Acad Sci USA 96: 16451650.
  • Rodriguez, G.M., Voskuil, M.I., Gold, B., Schoolnik, G.K., and Smith, I. (2002) ideR, an essential gene in Mycobacterium tuberculosis: role of IdeR in iron-dependent gene expression, iron metabolism, and oxidative stress response. Infect Immun 70: 33713381.
  • Sassetti, C.M., Boyd, D.H., and Rubin, E.J. (2001) Comprehensive identification of conditionally essential genes in mycobacteria. Proc Natl Acad Sci USA 98: 1271212717.
  • Zahrt, T.C., and Deretic, V. (2000) An essential two-component signal transduction system in Mycobacterium tuberculosis. J Bacteriol 182: 38323838.
  • Zalkin, H., and Nygaard, P. (1996) Biosynthesis of purine nucleotides. In Escherichia Coli and Salmonella: Cellular and Molecular Biololgy, Vol. 1. Neidhardt, F.C. (ed.). Washington, DC: American Society of Microbiology.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Table S1. Genes required for optimal growth. Data from replicate experiments were analyzed using GeneSpring software. Ratios (insertion probe/genomic probe) and fluorescence intensities ('insertion' and 'genomic') are provided along with minimum and maximum values. Table S2. Genes that produce slow growth when mutated. Data are the average of replicate experiments. Columns are labelled as in Table S1. Table S3. Non-essential genes. Data are the average of replicate experiments. Columns are labelled as in Table S1.

FilenameFormatSizeDescription
MMI_3425_sm_TableS1.xls172KSupporting info item
MMI_3425_sm_TableS2.xls30KSupporting info item
MMI_3425_sm_TableS3.xls593KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.