• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jacLink to Publisher's site
J Antimicrob Chemother. Mar 2009; 63(3): 451–457.
Published online Dec 24, 2008. doi:  10.1093/jac/dkn507
PMCID: PMC2721702

Morphological features and signature gene response elicited by inactivation of FtsI in Mycobacterium tuberculosis

Abstract

Objectives

Universally conserved events in cell division provide the opportunity for the development of novel chemotherapeutics against Mycobacterium tuberculosis. The aim of this study was to use the β-lactam antimicrobials cefalexin and piperacillin to inhibit FtsI and characterize the morphological changes and global transcriptional activities of genes to identify a signature response to FtsI inactivation.

Methods

Cefalexin and piperacillin were used to block cell division, and microscopy was used to evaluate the effects on bacterial morphology and ultrastructure. Global transcriptional analysis was performed to determine the impact of FtsI inhibition on cell cycle processes and to identify molecular markers.

Results

Inhibition of FtsI with cefalexin and piperacillin resulted in filamentous cells with multiple concentric rings and occasional branching as visualized by light and electron microscopy. Whole genome microarray-based transcriptional profiling and transcriptional mapping allowed the evaluation of cell cycle processes in response to inhibition of FtsI and characterization of transcriptional response and cell cycle processes.

Conclusions

This study substantiated that FtsZ-ring constriction and septal resolution require the transpeptidase activity of FtsI, making FtsI essential for cell division in M. tuberculosis. Therefore, FtsI is a target for drug discovery, and these studies provided a molecular signature of FtsI inactivation that can be applied to screening strategies for novel FtsI inhibitors.

Keywords: M. tuberculosis, cell division, microarray, cefalexin, piperacillin

Introduction

Despite the historical success of chemotherapy against Mycobacterium tuberculosis, this pathogen continues to cause high morbidity and mortality. The lengthy and complex multidrug regimens required to treat tuberculosis, the development of multidrug resistance to two or more of the front-line drugs and the increase in disease in endemic countries as a result of HIV/AIDS underscore the urgent public health need for new therapies.13 Many of the current tuberculosis-specific chemotherapeutics target components of the cell wall and have proved ineffective for treating established infections.46 However, analyses of the M. tuberculosis genome reveal a large number of essential conserved metabolic functions and numerous unexploited opportunities for drug discovery.

The cell division process, and in particular the penicillin-binding protein FtsI, represent a new target for M. tuberculosis drug discovery.1,6 At the time of septation, peptidoglycan synthesis and remodelling at the site of division is coordinated with constriction of the FtsZ ring and other components of the cell wall to produce two viable daughter cells.7 Studies in other bacteria have demonstrated that FtsI is required for peptide cross-linking of peptidoglycan at the septum and bacteria without functional FtsI proteins fail to divide, resulting in filamentous cells.8 In addition to filamentation, treatment with the FtsI inhibitors cefalexin and piperacillin leads to the formation of concentric rings along the bacterial filament and a branching morphology.811

The limited potency of β-lactam antibiotics against M. tuberculosis has been attributed primarily to the presence of β-lactamase activity, and secondarily to reduced binding affinity of β-lactam antibiotics for mycobacterial penicillin-binding proteins.1215 Therefore, much effort has been focused on inhibition of β-lactamase activity.16 Recently, the M. tuberculosis β-lactamase BlaC was crystallized and modelling has been undertaken to develop mycobacterial-specific β-lactamase inhibitors.12 Targeting BlaC to facilitate the use of β-lactam antibiotics is substantiated by mutant studies that confirmed that M. tuberculosis resistance to β-lactam antibiotics is mediated through BlaC.17,18

Alternatively, the development of novel FtsI inhibitors that are not susceptible to β-lactamase activity is another promising approach. The identification of inhibitors and advancement of lead compounds involve screening drug candidates for mode of action and off-target effects in bacteria, in addition to potency and inhibition of enzymatic activity.46 Accordingly, in this work, we inhibited FtsI activity and cell division with cefalexin and piperacillin, and report the corresponding alterations in morphology and response. Furthermore, characterization of these responses provides markers useful for developing appropriate drug screens to identify novel FtsI inhibitors.

Materials and methods

Bacterial growth conditions and recombinant strains

For all experiments, M. tuberculosis H37Rv was cultivated at 37°C in Middlebrook 7H9 liquid medium containing 0.2% glycerol, ADC (albumin, dextrose and catalase enrichment) and 0.05% Tween 80 or on Middlebrook 7H11 agar containing OADC (oleic acid, albumin, dextrose and catalase enrichment). For determination of MICs, M. tuberculosis was grown to an OD600 of ~0.5 and diluted 1:10. Cefalexin and piperacillin were added to final concentrations of 500–0.5 µM in a total volume of 0.1 mL, and tested in triplicate. The MIC was defined as the lowest concentration of drug that prevented bacterial outgrowth as monitored by OD600 after 7 days of incubation. For viability testing, drugs were added to 30 mL cultures. Each day, dilutions were plated on Middlebrook 7H11 agar, and viability was determined by enumeration of cfu. For microarray experiments, M. tuberculosis cultures (30 mL) were grown to an OD600 of 0.3, each drug was added at its respective MIC (20 µM cefalexin or 40 µM piperacillin) or untreated for a control, and the cultures incubated at 37°C for 5 or 24 h.

The FtsI open reading frame was amplified from M. tuberculosis H37Rv genomic DNA (TB Vaccine Testing and Research Material Contract HHSN266200400091c) using Accuprime pfx DNA polymerase with ftsI-5′: atg agc cgc gcc gcc and ftsI-3′: cta ggt ggc ctg caa gac c including engineered asymmetric NdeI and HindIII restriction sites for insertion into the mycobacterial-inducible shuttle vector pVV16ap. The pVV16ap vector contains an acetamidase promoter region that provides enhanced expression in the presence of 2% acetamide.

Ultrastructural analysis by scanning electron microscopy (SEM)

Bacteria were collected by centrifugation and washed three times in PBS, pH 7.4, and fixed with 2.5% gluteraldehyde in buffer A [0.1 M potassium phosphate (pH 7.4), 1 mM CaCl2 and 1 mM MgCl2] at 4°C for 48 h. The fixed bacterial cells were collected by centrifugation, washed three times in buffer A and treated with 1% OsO4 in buffer A for 30 min at 4°C. Again cells were washed three times with buffer A, and prepared for SEM with a graded series of ethanol treatments (20% to 100%). Ultrastructural examination was performed using a JOEL JEM-100CX electron microscope.

Microarray processing and data analysis

The M. tuberculosis microarrays were obtained through the TB Vaccine Testing and Research Materials Contract (HHSN266200400091c) at Colorado State University. Treated and control bacterial cells were suspended in TRIzol and physically disrupted with 0.1 mm zirconium beads.1 Total RNA was purified using an RNeasy Kit (Qiagen). Approximately, 8 µg of total RNA from each treatment was converted into cDNA in the presence of either Cy5- or Cy3-labelled nucleotides as previously described.1 Hybridization was performed at 42°C for 12 h. Slides were scanned using a VersArray Chipreader Pro. Data reduction and global normalization were performed on the raw fluorescent intensities. The normalized intensity values of treated and control cultures were used to generate ratio and log2 expression values for each gene. The final microarray dataset used for transcriptional mapping resulted from combining the two biologically independent replicates of 24 h cefalexin and piperacillin treatments and four controls. For defining and evaluating the molecular markers of FtsI inhibition, independent biological replicates of 5 h cefalexin and piperacillin treatments and four 24 h cefalexin and piperacillin treatments were used.

Quantitative real-time PCR

Quantitative real-time PCR was performed on selected genes to verify differential gene expression observed through microarray data analysis. Quantitative real-time PCR was performed using SYBR-green (Invitrogen). PCR amplification was performed with a thermocycling programme of 55°C for 5 min then 95°C for 2 min and 45 cycles of 95°C for 15 s, 60°C for 30 s and 72°C for 45 s. The relative number of transcripts for each gene was determined based on linear regression analysis of 100, 10 and 1 ng of M. tuberculosis genomic DNA. The total number of targets (n) was calculated by the equation n= a+ b log(x), where a is the intercept, b the slope of the standard curve and x the threshold cycle obtained by amplifying n targets. All reactions were performed in triplicate on two independent RNA preparations from M. tuberculosis treated with cefalexin or piperacillin using primers as described.1

Results

Inhibition of FtsI results in bacterial filaments and branching and budding

The susceptibility and viability of M. tuberculosis to cefalexin and piperacillin were determined by microbroth dilution assay and plating.1,19 The MICs of cefalexin and piperacillin were found to be 20 and 40 µM, respectively, for wild-type bacteria. Treatment with cefalexin or piperacillin at the MIC for 3 days reduced bacterial viability by 4 and 5 logs, respectively (Figure 1), and there was no bacterial growth after exposure to either drug for 5 days. Overexpression of ftsI resulted in a 6-fold increase in the MIC values of each drug (125 and 250 µM for cefalexin and piperacillin, respectively), which corresponds to the relative increase in the level of ftsI expression in the ftsI merodiploid strain as determined by quantitative real-time PCR.

Figure 1
Bacterial growth in the presence of FtsI inhibitors. Bacterial cfu after treatment with 20 µM cefalexin or 40 µM piperacillin at designated timepoints.

When visualized by acid-fast staining, piperacillin- and cefalexin-treated bacteria possessed dark staining regions at equal intervals (Figure 2a and b). SEM allowed for length determinations of piperacillin- and cefalexin-treated cells to be performed. Treatment for 3 days with piperacillin or cefalexin resulted in a mean length of 7.1 ± 1.6 and 7.2 ± 1.3 µm, respectively, which is in sharp contrast to the 3.3 ± 1.0 µm length of untreated cells (P < 0.05). Closer examination using electron microscopy revealed the presence of concentric rings in addition to filamentation (Figure 2c and d). The observation of clearly visible concentric rings indicative of septa was consistent with our previous observations and those from other organisms treated with FtsI inhibitors.1,2022 Bacteria with prolonged exposure to cefalexin also developed clearly visible budding and branching; this, however, was not observed as frequently for piperacillin treatment (Figure 2e and f). Together, these data confirm that FtsI is essential for cell division and overall bacteria viability.

Figure 2
Bacterial morphology of M. tuberculosis treated with the cell division inhibitors. Bacteria treated with piperacillin or cefalexin were visualized by acid-fast staining (a and b) and SEM (c–f). Inhibition of FtsI with 40 µM piperacillin ...

Transcriptional mapping and identification of a signature profile for FtsI inactivation

To evaluate the overall effect of FtsI inactivation on cellular processes, such as DNA replication, cell division and macromolecular synthesis, the global transcriptional response of M. tuberculosis to cefalexin and piperacillin treatment was assessed through DNA microarray analysis. When the overall expression profiles of cells treated for 24 h with cefalexin and piperacillin were compared [Table S1, available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/)], there was a 78% concordance in the transcriptional response between these treatments, indicating that FtsI inhibition was similar in the mode of action for both these β-lactam drugs. The transcriptionally active open reading frames were subjected to self-organizing mapping (SOM) analyses. Sets of discriminant genes representing the major cell cycle processes of DNA initiation, replication and chromosome segregation; cell division; peptidoglycan synthesis; lipoarabinomannan synthesis; arabinogalactan; and mycolic acid synthesis were selected to provide a more concise picture of how genes involved in cell division are grouped relative to other cell cycle processes (Figure 3). Overall, cell cycle processes were distributed in an overlapping fashion consistent with their execution in cellular replication. Inhibition of cell division via FtsI resulted in repression of DNA initiation, replication and segregation of discriminant genes. Genes encoding cell division components were distributed in a bimodal fashion. Specifically, genes involved in septum resolution were induced and genes involved in formation of the septum were repressed. Discriminant genes encoding for pathways for cell wall synthesis were largely distributed between those involved in cell division and DNA initiation. Gene expression levels observed by microarray analysis were confirmed by quantitative real-time PCR (Figure 4). The arrangement of cell cycle processes is consistent with inactivation of FtsI and inhibition of the final step in cell division. The results were also similar to reports on other bacteria of temporal regulation of genes encoding septum formation or septum resolution.23,24

Figure 3
Transcriptional map of cell cycle process in M. tuberculosis. (a) Mean expression activity of each SOM group. (b) Coordinates of the transcriptional map used to map cell cycle processes as determined by SOM analysis. (c) Bacterial cell cycle processes ...
Figure 4
Quantitative real-time PCR analysis of discriminant cell cycle genes. Data are means (±SD) of log2 expression from at least two biological independent samples. Ratios were calculated using the total number of gene targets from treated bacteria ...

Molecular markers of FtsI inhibition

Based on transcriptional activity, cell cycle-associated genes segregate into two defined groups, those induced and those repressed in response to FtsI inhibition (Figure 5a). The transcriptional response was further evaluated using independent experiments of cefalexin and piperacillin treatments at 5 and 24 h to identify a subset of genes that can serve as molecular markers for FtsI inhibition. Hierarchal clustering analysis of the transcriptional response of cell cycle discriminant genes assigned the independent treatment experiments into two groups that correspond to either 5 or 24 h exposures (Figure 5b) and revealed five separate groups of genes with high correlations of transcriptional activities (Figure 5b). Based on this analysis and the overall trend in transcriptional activity upon inhibition of FtsI, genes from Groups Ia and VIa were selected for further evaluation as to whether they were highly predictive of FtsI inhibition for high throughput screening (Figure 5c and d). Upon analysis of the experimental expression profiles, the transcriptional response of signature genes had a 96% correlation at both the 5 and 24 h exposure times, demonstrating that this set of genes could discern inactivation of FtsI independent of exposure time. The mean log2 expression of each FtsI inactivation signature gene was consistently expressed over six experimental treatments, thus providing a quantifiable set of molecular markers. To assess whether the consensus response profile for FtsI inactivation was similar to the transcriptional response to treatment with drugs with other modes of action, they were compared with previously reported expression profiles of drug treatment and with inhibition of FtsZ.1,6,25 This inspection did not identify overlaps in transcriptional responses of FtsI inhibition with other treatments, thus adding confidence that the observed response of the consensus genes was due to inhibition of FtsI.

Figure 5
Transcriptional response of cell cycle discriminant genes and signature transcriptional pattern of FtsI inhibition. (a) Transcriptional response of cell cycle discriminant genes determined by whole genome microarray analysis. (b) Hierarchal clustering ...

Discussion

One of the most challenging tasks in developing chemotherapeutics against a specific target is the identification of robust whole cell molecular markers that are informative of the mode of action and can discern off-target activity.46,25 This is largely because of a lack of tools to assess the suitability of a specific cellular process as a target for drug development. A requirement of such screening markers is the ability to differentiate with confidence between compounds with either an unrelated or closely related mode of action and those with the desired mode of action.5 Accordingly, morphological analyses and global transcriptional profiling of M. tuberculosis treated with the β-lactams cefalexin and piperacillin were undertaken to identify morphological characteristics and define molecular markers of FtsI inhibition that offer the potential for the development of facile whole cell screens.

Visualization of M. tuberculosis treated with cefalexin and piperacillin revealed that similar to the prevention of septum formation, inhibition of FtsI resulted in filamentation.9 However, an important distinction between the filaments resulting from cefalexin and piperacillin treatment and those observed when M. tuberculosis was treated with FtsZ inhibitors was that FtsI inhibition leads to the development of concentric rings at even intervals along the filament.1 Concentric rings are indicative of septa, and their presence and persistence indicate that FtsI activity is essential for completion of cell division in M. tuberculosis. It also indicates that the initiation of FtsZ assembly at other sites is not negatively regulated with respect to the completion of the previous round of cell division. This observation agrees with previous conclusions that FtsI activity is not required for assembly of FtsZ rings at future sites,9 and substantiates the notion that regulation of cell division in M. tuberculosis occurs at the level of FtsZ polymerization, despite the general lack of annotated FtsZ regulatory elements. Importantly, initiation of cell division is uncoupled to completion of the previous round of cell division in mycobacterium. This is further supported by the presence of occasional budding and branching as a result of FtsI inhibition caused by an imbalance in the regulation and activity of septal proteins such that asymmetric cell division events occur.11,26

An area of interest regarding the development of new chemotherapeutics against novel or underexploited molecular targets is the ability to discern the mode of action during the screening process. The identification of a signature transcriptional pattern of genes allows for the development of whole cell-based mode of action screens. Transcriptional mapping organized genes into groups consistent with cell cycle processes and it was found that genes encoding proteins involved with cell cycle processes that precede the completion of cell division were repressed and those involved in resolution of the septum and cell wall synthesis were induced. This general expression profile was in contrast to the transcriptional response to inhibition of septum formation previously reported.1,6 Notably, these trends in the transcriptional response highlight the temporal nature of cell cycle events and the dynamic regulation between early and late cell cycle processes. Together, these studies demonstrated that morphological and transcriptional profiling can be used to identify features predictive of FtsI inhibition and substantiated that these genes can be used as a tool to discern the mode of action in the evaluation of new drug candidates.

The availability of a transcriptional response-based screening tool capable of discerning the mode of action for high throughput screening of compound libraries provides an additional level of information about cellular processes. Such gene responses reflect metabolic reactions to a specific deficit and have been successfully translated to report a specific pathway, thus revealing the mode of action.1,6,25,27 Further analysis of the transcriptional response of cell cycle discriminant genes led to the identification of molecular markers suitable for high throughput screening of anti-FtsI candidates. When these molecular markers were evaluated using independent treatments of FtsI inhibitors, it was found that in each case they were capable of reporting the mode of action with a 96% correlation. Identification of these molecular markers based on transcriptional mapping established that a limited number of genes could be used to discern FtsI inhibition based solely on non-subjective criteria, and inclusion of the distinct morphological features resulting from FtsI inhibition provides additional evidence of a compound’s mode of action.

Importantly, with regard to the development of novel chemotherapeutics that target cell division, the identification of high content molecular features capable of reporting the inhibition of a molecular target with high predictability affords the opportunity to screen compounds for FtsI mode of action. Not only can a single feature be chosen to design a recombinant strain for high throughput screening, an approach that has proved useful for the identification of novel compounds with a specific mode of action,28 multiple features can be used together as a platform for determining the mode of action and potentially identifying off-target effects. Notably, the signature pattern of FtsI inhibition and the set of molecular markers identified in this study were capable of categorizing known FtsI inhibitors, and therefore can be used to predict the mode of action of FtsI inhibition. This provides a molecular tool for the development of reporter strains and high content screening platforms capable of identifying potential FtsI inhibitors and discerning potential off-target effects.

Funding

This work was supported by RO1 AI055298 (R. A. S.).

Transparency declarations

None to declare.

Supplementary Material

[Supplementary Data]

Acknowledgements

We gratefully acknowledge Dr Philip Chapman for assistance with the statistical analysis and Ms Laurel Respicio and Ms Melissa Boyne for technical assistance. We acknowledge the support provided by the TB Vaccine Testing and Research Materials Contract HHSN266200400091c.

References

1. Slayden RA, Knudson DL, Belisle JT. Identification of cell cycle regulators in Mycobacterium tuberculosis by inhibition of septum formation and global transcriptional analysis. Microbiology. 2006;152:1789–97. [PubMed]
2. Zhang Y, Post-Martens K, Denkin S. New drug candidates and therapeutic targets for tuberculosis therapy. Drug Discov Today. 2006;11:21–7. [PubMed]
3. Zhang Y, Amzel LM. Tuberculosis drug targets. Curr Drug Targets. 2002;3:131–54. [PubMed]
4. Barry CE, 3rd, Slayden RA, Sampson AE, et al. Use of genomics and combinatorial chemistry in the development of new antimycobacterial drugs. Biochem Pharmacol. 2000;59:221–31. [PubMed]
5. Slayden RA, Crick D, Neil MM, et al. The Role of Genomics in Antibacterial Drug Discovery. New York: Marcel Dekker; 2003. Genomics in tuberculosis drug discovery; pp. 111–34.
6. Respicio L, Nair PA, Huang Q, et al. Characterizing septum inhibition in Mycobacterium tuberculosis for novel drug discovery. Tuberculosis (Edinb) 2008;88:420–9. [PubMed]
7. Errington J, Daniel RA, Scheffers DJ. Cytokinesis in bacteria. Microbiol Mol Biol Rev. 2003;67:52–65. [PMC free article] [PubMed]
8. Eberhardt C, Kuerschner L, Weiss DS. Probing the catalytic activity of a cell division-specific transpeptidase in vivo with β-lactams. J Bacteriol. 2003;185:3726–34. [PMC free article] [PubMed]
9. Pogliano J, Pogliano K, Weiss DS, et al. Inactivation of FtsI inhibits constriction of the FtsZ cytokinetic ring and delays the assembly of FtsZ rings at potential division sites. Proc Natl Acad Sci USA. 1997;94:559–64. [PMC free article] [PubMed]
10. Maki N, Gestwicki JE, Lake EM, et al. Motility and chemotaxis of filamentous cells of Escherichia coli. J Bacteriol. 2000;182:4337–42. [PMC free article] [PubMed]
11. Gullbrand B, Akerlund T, Nordstrom K. On the origin of branches in Escherichia coli. J Bacteriol. 1999;181:6607–14. [PMC free article] [PubMed]
12. Wang F, Cassidy C, Sacchettini JC. Crystal structure and activity studies of the Mycobacterium tuberculosis β-lactamase reveal its critical role in resistance to β-lactam antibiotics. Antimicrob Agents Chemother. 2006;50:2762–71. [PMC free article] [PubMed]
13. Chambers HF, Moreau D, Yajko D, et al. Can penicillins and other β-lactam antibiotics be used to treat tuberculosis? Antimicrob Agents Chemother. 1995;39:2620–4. [PMC free article] [PubMed]
14. Jarlier V, Nikaido H. Mycobacterial cell wall: structure and role in natural resistance to antibiotics. FEMS Microbiol Lett. 1994;123:11–8. [PubMed]
15. Zhang Y, Steingrube VA, Wallace RJ., Jr β-Lactamase inhibitors and the inducibility of the β-lactamase of Mycobacterium tuberculosis. Am Rev Respir Dis. 1992;145:657–60. [PubMed]
16. Abate G, Miorner H. Susceptibility of multidrug-resistant strains of Mycobacterium tuberculosis to amoxycillin in combination with clavulanic acid and ethambutol. J Antimicrob Chemother. 1998;42:735–40. [PubMed]
17. Flores AR, Parsons LM, Pavelka MS., Jr Characterization of novel Mycobacterium tuberculosis and Mycobacterium smegmatis mutants hypersusceptible to β-lactam antibiotics. J Bacteriol. 2005;187:1892–900. [PMC free article] [PubMed]
18. Flores AR, Parsons LM, Pavelka MS., Jr Genetic analysis of the β-lactamases of Mycobacterium tuberculosis and Mycobacterium smegmatis and susceptibility to β-lactam antibiotics. Microbiology. 2005;151:521–32. [PubMed]
19. Huang Q, Kirikae F, Kirikae T, et al. Targeting FtsZ for antituberculosis drug discovery: noncytotoxic taxanes as novel antituberculosis agents. J Med Chem. 2006;49:463–6. [PMC free article] [PubMed]
20. Romberg L, Levin PA. Assembly dynamics of the bacterial cell division protein FTSZ: poised at the edge of stability. Annu Rev Microbiol. 2003;57:125–54. [PubMed]
21. Wang L, Lutkenhaus J. FtsK is an essential cell division protein that is localized to the septum and induced as part of the SOS response. Mol Microbiol. 1998;29:731–40. [PubMed]
22. Weiss DS, Chen JC, Ghigo JM, et al. Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J Bacteriol. 1999;181:508–20. [PMC free article] [PubMed]
23. Goehring NW, Gueiros-Filho F, Beckwith J. Premature targeting of a cell division protein to midcell allows dissection of divisome assembly in Escherichia coli. Genes Dev. 2005;19:127–37. [PMC free article] [PubMed]
24. Katis VL, Harry EJ, Wake RG. The Bacillus subtilis division protein DivIC is a highly abundant membrane-bound protein that localizes to the division site. Mol Microbiol. 1997;26:1047–55. [PubMed]
25. Boshoff HI, Myers TG, Copp BR, et al. The transcriptional responses of Mycobacterium tuberculosis to inhibitors of metabolism: novel insights into drug mechanisms of action. J Biol Chem. 2004;279:40174–84. [PubMed]
26. Gullbrand B, Nordstrom K. FtsZ ring formation without subsequent cell division after replication runout in Escherichia coli. Mol Microbiol. 2000;36:1349–59. [PubMed]
27. Wilson M, DeRisi J, Kristensen H-K, et al. Exploring drug-induced alterations in gene expression in Mycobacterium tuberculosis by microarray hybridization. Proc Natl Acad Sci USA. 1999;96:12833–8. [PMC free article] [PubMed]
28. Lee RE, Protopopova M, Crooks E, et al. Combinatorial lead optimization of [1,2]-diamines based on ethambutol as potential antituberculosis preclinical candidates. J Comb Chem. 2003;5:172–87. [PubMed]

Articles from Journal of Antimicrobial Chemotherapy are provided here courtesy of Oxford University Press
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Compound
    Compound
    PubChem Compound links
  • Gene
    Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • MedGen
    MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links
  • Taxonomy
    Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree