• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 6, 2004; 101(1): 314–319.
Published online Dec 26, 2003. doi:  10.1073/pnas.0305439101
PMCID: PMC314182

A polyketide synthase catalyzes the last condensation step of mycolic acid biosynthesis in mycobacteria and related organisms


Mycolic acids are major and specific constituents of the cell envelope of Corynebacterineae, a suborder of bacterial species including several important human pathogens such as Mycobacterium tuberculosis, Mycobacterium leprae, or Corynebacterium diphtheriae. These long-chain fatty acids are involved in the unusual architecture and impermeability of the cell envelope of these bacteria. The condensase, the enzyme responsible for the final condensation step in mycolic acid biosynthesis, has remained an enigma for decades. By in silico analysis of various mycobacterial genomes, we identified a candidate enzyme, Pks13, that contains the four catalytic domains required for the condensation reaction. Orthologs of this enzyme were found in other Corynebacterineae species. A Corynebacterium glutamicum strain with a deletion in the pks13 gene was shown to be deficient in mycolic acid production whereas it was able to produce the fatty acids precursors. This mutant strain displayed an altered cell envelope structure. We showed that the pks13 gene was essential for the survival of Mycobacterium smegmatis. A conditional M. smegmatis mutant carrying its only copy of pks13 on a thermosensitive plasmid exhibited mycolic acid biosynthesis defect if grown at nonpermissive temperature. These results indicate that Pks13 is the condensase, a promising target for the development of new antimicrobial drugs against Corynebacterineae.

Many clinical isolates of Mycobacterium tuberculosis, the causative agent of tuberculosis, have become resistant to one or several chemotherapeutic agents, and these resistant strains may account for up to 41% of M. tuberculosis clinical isolates in some countries according to the World Health Organization (1), emphasizing the need to develop new and efficient therapeutic agents. Many of the drugs commonly used to treat tuberculosis, such as isoniazid, ethionamide, and ethambutol, target cell envelope biogenesis in mycobacteria (2). The envelope is typified by the occurrence of high molecular weight 2-alkyl and 3-hydroxy fatty acids, called mycolic acids (Fig. 1). These fatty acids are present in a limited range of microorganisms, which are grouped in the Corynebacterineae suborder. They are present either as esters of trehalose or as ester of the terminal pentaarabinofuranosyl units of arabinogalactan, the polysaccharide that, together with peptidoglycan, forms the cell wall skeleton (3). Both types of mycolate-containing components have been shown to play a crucial role in the structure and function of the cell envelope. Mycolic acids attached to the cell wall are organized with other lipids to form a permeability barrier that contributes to the very low permeability of the envelope of Corynebacterineae and the natural resistance of these microorganisms to various antibiotics (46). Trehalose mycolates have been implicated in numerous biological functions, notably in mycobacterial virulence where the structure of the mycolates has been found to be important for initial replication and persistence in vivo (7, 8).

Fig. 1.
Structure of mycolic acids and proposed terminal steps in their biosynthesis. (A) Structure of mycolic acids from M. tuberculosis (α type) and C. glutamicum. These 2-alkyl 3-hydroxyl fatty acids are elaborated by members of the Corynebacterineae ...

The structure and biosynthesis of mycolic acids have been the subject of intense research efforts for a number of years, primarily because the enzymes involved in their metabolism offer attractive and selective targets for the development of new antimycobacterial drugs. The R1 and R2 chains of mycolic acids vary in length and structure according to the Corynebacterineae genus and the species considered, but all display a common structural feature: the mycolic motif (Fig. 1). This feature suggests that enzymatic steps involved in the formation of this motif are common to all members of this group of bacteria. Thus, the enzyme that catalyzes the condensation of two fatty acids to yield the mycolic motif, hereafter referred to as condensase, represents a good potential target for the development of new and specific drugs against Corynebacterineae. The condensase has remained a mystery despite the considerable efforts made by several laboratories during the last 40 yr. Various mechanisms have been suggested for this condensation reaction, and the possibility of a Claisen-type reaction has attracted much attention (9). This reaction would involve a condensation of a carboxylated acyl-coenzymeA and a second activated acyl chain to yield a 3-oxo intermediate, which would then be reduced to form mycolic acid (Fig. 1). As pointed out by Gastambide-Odier and Lederer (9), this reaction is very similar to the condensation of acyl-CoA with methylmalonyl-CoA to form methyl-branched fatty acid, a reaction recently shown to be catalyzed in mycobacteria by a family of enzymes called type I polyketide synthases (Pks) (10, 11).

In this paper, we identified a candidate type I Pks for being the condensase, the enzyme responsible for the final condensation step in mycolic acid biosynthesis. We demonstrated the role of the candidate enzyme in mycolic acid biosynthesis and envelope biogenesis in Corynebacteria and Mycobacteria. We also showed that this enzyme is essential for the viability of mycobacteria.

Materials and Methods

Strains and Culture Conditions. Corynebacterium glutamicum (ATCC13032 RES) (12) was cultured on BHI medium (Difco). Rhodococcus rhodochrous (ATCC13808) and Mycobacterium smegmatis mc2155 were cultured on Luria broth (LB) medium (Difco) supplemented with 0.05% Tween 80 for M. smegmatis,to prevent aggregation. Kanamycin, hygromycin, chloramphenicol, and sucrose were added when required at final concentrations of 40 μg/ml or 25 μg/ml (for M. smegmatis and C. glutamicum respectively), 50 μg/ml, 15 μg/ml, and 10% (wt/vol), respectively.

Computer Analysis. M. tuberculosis strain H37Rv and Mycobacterium leprae DNA sequences were obtained from the Pasteur Institute Website (www.pasteur.fr). Research of Pks13 orthologs on Mycobacterium avium, Mycobacterium marinum, M. smegmatis, Corynebacterium diphtheriae, C. glutamicum, and Corynebacterium efficiens genomes were performed at the National Center for Biotechnology Information web site (www.ncbi.nlm.nih.gov) by using the blast program. The sequences of the various Pks13, FadD32, and AccD4 proteins were compared by using the needlemanwunsh program on the Pasteur Institute web site.

PCR Amplification of the pks13 Locus from Rhodococcus. Total bacterial DNA was extracted from 5 ml of saturated liquid cultures as described (13). DNA pellets were resuspended in 100 μl of 10 mM Tris (pH 8). Degenerate primers were designed from a multiple alignment of DNA sequences from M. tuberculosis, M. leprae, M. smegmatis, and C. glutamicum (Table 2, which is published as supporting information on the PNAS web site). DNA fragments internal to the fadD32, pks13, or accD4 genes were first amplified by PCR by using genomic DNA from R. rhodochrous and the degenerate primers corresponding to the various genes. PCR was performed in a final volume of 50 μl containing 2.5 units of TaqDNA polymerase (Roche Molecular Biochemicals), 10% Me2SO, and each primer at a concentration of 4 μM. The PCR products were analyzed and inserted into the pGEM-T easy vector according to the manufacturer's instructions (Promega). Inserts were sequenced with the T7 and SP6 primers. We used the resulting sequences to design specific primers for R. rhodochrous (Table 2), which were used to amplify a region covering almost the entire fadD32-pks13-accD4 8.7-kb locus (≈600 nt of the 5′ region of fadD32 were missing). The various fragments were cloned into pGEM-T easy vectors and sequenced (Genome Express, Meylan, France).

Construction of the C. glutamicum Δpks13::km Mutant and Complementation Plasmid. Two DNA fragments (0.9 kb and 0.7 kb in length) overlapping the pks13 gene at its 5′ and 3′ extremities were amplified by PCR from C. glutamicum total DNA by using primers pkdel5 + pkdel2 and pkdel3 + pkdel4, respectively (Table 2). These fragments were inserted into plasmid pMCS5 (MoBiTec, Göttingen, Germany). A kanamycin resistance cassette was inserted between these two PCR fragments to give pCMS5::pks. This plasmid was transferred into C. glutamicum by electroporation, and transformants were selected on plates containing kanamycin. Transformants in which allelic replacement had occurred between the WT chromosomal pks13 gene and the mutated plasmid-borne allele were identified on the basis of their rough colony morphology and characterized by PCR by using primers fa2, ac2, K10, K7, pk1, and pk2. One strain, C. glutamicum Δpks13::km, was selected for further studies.

The complementation plasmid, pCGL2308, was produced by inserting the C. glutamicum pks13 gene plus 417 bp of the upstream region, obtained in a 5.3-kb PCR fragment with primers pk3 and pk4, into vector pCGL482 (14).

Construction of the M. smegmatis Conditional Mutant Strain. Two DNA fragments, each ≈1 kb in length, overlapping the pks13 gene at its 5′ and 3′ extremities, were amplified by PCR from M. smegmatis total DNA by using primers 13F + 13G and 13H + 13I, respectively (Table 2). These fragments were inserted into plasmid pJQ200 (15). A hygromycin resistance cassette was inserted between the two PCR fragments to give pDP28. This plasmid was transferred into M. smegmatis by electroporation, and transformants were selected on plates containing hygromycin. Transformants in which pDP28 had been integrated by single crossover between the WT and mutated copies of pks13 were characterized by PCR, by using primers 13J, 13K, H1, and H2. One strain, named PMM47, was selected for further studies.

To produce the complementation plasmid pDP32, the full-length pks13 gene was amplified by PCR from M. smegmatis total DNA by using primers 13R and 13P (Table 2). The pks13 gene was inserted into a thermosensitive mycobacterial plasmid derived from pCG63 (16) and containing a mycobacterial expression cassette derived from pMIP12 (17). The resulting plasmid, pDP32, contains the entire pks13 gene under the control of the pBlaF* promotor. Plasmid pDP32 was transferred by electroporation into PMM47, and transformants were selected on plates containing kanamycin and hygromycin. The second crossover event at the chromosomal pks13 locus was selected by plating a liquid culture of PMM47:pDP32 cultured at 30°C on plates containing kanamycin, hygromycin, and sucrose, which were then incubated at 30°C. Colonies were screened by PCR by using primers 13J, 13K, H1, and H2; and a strain, named PMM48:pDP32, in which the WT chromosomal copy of pks13 was replaced by the mutated pks13::hyg allele, was selected for phenotypic analysis. In these conditions, 8% of the selected HygR, KmR, SucR colonies resulted from allelic exchange, the other clones resulting from mutation in the sacB gene.

Biochemical Analysis of C. glutamicum and M. smegmatis Strains. Cultures of C. glutamicum were grown to exponential growth phase and labeled by incubation with 0.5 μCi/ml (1 Ci = 37 GBq) [14C]acetate (specific activity of 54 mCi/mmol; ICN, Orsay, FR) for 3 hr. For radiolabeling of the M. smegmatis conditional mutant at nonpermissive temperature, PMM48:pDP32 and the WT mc2155 were first cultured at 30°C. These cultures were then diluted in fresh media to an OD600 nm of 0.005 and incubated at 42°C until the OD600 nm reached 0.3. Cells were then labeled by incubation for 3 h with 0.5 μCi/ml [14C]acetate. Fatty acids were prepared from the labeled cells and separated by analytical TLC on Durasil 25 by using dichloromethane or petroleum ether/diethyl ether (9:1) as the eluent as described (18). Labeled compounds were quantified on a PhosphorImager (Amersham Pharmacia Biosciences).

For GC and GC-MS analyses, trimethylsilyl derivatives of fatty acids were obtained as described (19). GC analyses were performed by using a Girdel Series 30 instrument equipped with an OV1 capillary column (0.30 mm × 25 m) by using helium gas (0.7 bar) with the flame ionization detector at 310°C. The temperature program was from 60–310°C, at 5°C/min. GC-MS analyses were performed on a Hewlett-Packard 5889 X mass spectrometer (electron energy, 70 eV) working in electron impact (EI) by using NH3 as the reagent gas (CI/NH3), coupled with a Hewlett-Packard 5890 Series II gas chromatograph fitted with a similar OV1 column (0.30 mm × 12 m).

Freeze-Fracture Preparations for Electron Microscopy. Bacterial suspensions were harvested by centrifugation. The pellet was placed between a thin copper holder and a thin copper plate before quenching in liquid propane (20). The frozen sample was fractured at -125°C in a vacuum of ≈10-7 torr by removing the upper plate with a knife that had been cooled in liquid propane in a Balzers 301 freeze-etching unit. The fractured sample was replicated with a 1.5-nm-thick platinum-carbon deposit and backed with chromic acid before being washed with distilled water and was observed with a Hitachi 7000 electron microscope operating at 100 kV.

Results and Discussion

Identification of a Gene Putatively Encoding the Condensase. The proposed Claisen-type reaction involved in the condensation of two fatty acid chains to form mycolic acid is very similar to the one catalyzed by type I Pks during the formation of methyl-branched fatty acids in M. tuberculosis (9). We therefore hypothesized that a type I Pks with an unusual substrate specificity might be responsible for catalyzing the condensation reaction that generates mycolic acid. This hypothesis leads to several predictions: (i) the candidate Pks must contain the catalytic domains required for the condensation reaction: an acyl transferase domain (AT), a ketosynthase domain (KS), an acyl carrier protein domain (ACP), and a thioesterase domain (TE) (Fig. 1); (ii) this enzyme should be present in the various species of bacteria that produce mycolic acid.

We investigated the available genome sequences of various mycobacterial species. We first compared the type I Pks found in M. tuberculosis and M. leprae. Both these species are known to produce various types of mycolic acid despite the massive gene decay that was shown to have occurred in the genome of M. leprae (21). A systematic search for the 16 type I Pks of M. tuberculosis (22) in M. leprae revealed that only nine of these putative enzymes were present in M. leprae. Seven of these nine putative enzymes have been already shown to be involved in the biosynthesis of another group of lipids in M. tuberculosis: the phthiocerol dimycocerosates and structurally related phenolglycolipids (19, 23). One of the two remaining candidates, ML1229, was found to display a common domain organization and strong sequence similarities with the subfamily of M. tuberculosis type I Pks involved in polymethyl-branched fatty acid biosynthesis. The members of this subfamily of Pks have keto reductase, dehydratase, and enoyl reductase catalytic domains that would not be expected to be present in the condensase. The remaining candidate, named ML0101 in M. leprae, is 83% identical over the full length (1733 aa) to the Pks13 of M. tuberculosis and the two proteins contain the required ketosynthase, acyl transferase, acyl carrier protein, and thioesterase catalylic domains. In addition, the genes encoding these proteins are located downstream from the fbp genes, the products of which have been shown to be involved in the transfer of mycolic acid onto arabinogalactan in M. tuberculosis (24). As genes encoding proteins involved in a given biosynthesis pathway are often clustered on the chromosome, this organization of genes suggests that Pks13 may be involved in the metabolism of mycolic acids. Moreover, two genes predicted to encode an acyl-CoA synthase (fadD32) and a subunit of an acyl-CoA carboxylase (accD4), two enzymes that may be required for activation of the condensase substrates, flank pks13 in both M. tuberculosis and M. leprae (Fig. 1). As expected, pks13, fadD32, and accD4 orthologs were found to be organized in the same way also on the chromosomes of other mycobacterial species (M. smegmatis, M. marinum, and M. avium) for which partial genome sequences are available; the sequence of the three encoded proteins displayed >70% similarity to their M. tuberculosis orthologs (Table 1).

Table 1.
Similarities over the entire length of the protein among the Pks13, FadD32, and AccD4 sequences of various Corynebacterineae species

We investigated whether Pks13 was present in other mycolic-acid-producing bacterial genera by searching for Pks13 orthologs in three corynebacterial species (C. glutamicum, C. efficiens, and C. diphtheriae) for which the complete genome sequences are available. Pks13, but also FadD32 and AccD4 orthologs, were found in these three species (Table 1), and the genetic organization of the region surrounding pks13 was again highly conserved. Importantly, pks13 was the only type I pks gene found in corynebacteria. The fadD32-pk13-accD4 locus was also amplified from a reference strain of Rhodococcus by PCR with degenerate and specific primers. The sequence of the fragments amplified revealed that the encoded FadD32, Pks13, and AccD4 proteins were very similar to those from M. tuberculosis (Table 1). Thus, Pks13 is found in all of the Corynebacterineae analyzed and is the only type I Pks meeting the criteria for identification as a candidate condensase. We therefore identified Pks13 as a good potential candidate for the condensase.

Deletion of pks13 in C. glutamicum Induces Major Changes of Microbiological Phenotypes. We investigated whether Pks13 was the condensase by deleting the WT copy of pks13 in C. glutamicum ATCC13032. We chose to study this species because it has been demonstrated that Corynebacteria spp. can survive without producing mycolates, as shown by the existence of. C. amycolatum (25), whereas isoniazid, an inhibitor of mycolic acid synthesis, has been shown to be lethal to Mycobacteria spp (26). A kanamycin resistance cassette flanked by arms, 900 bp and 700 bp identical to the 5′ and 3′ parts of pks13, was cloned into a vector unable to replicate in Corynebacteria. This construct was transferred into C. glutamicum. Several kmR transformants were analyzed by PCR by using various combination of primers, and one clone gave the amplification pattern consistent with allelic replacement of the WT copy of pks13 by the mutated copy with the 4.3-kb internal deletion into which the kanamycin cassette was inserted (Fig. 2). In comparison with the WT strain, the Δpks13::km mutant exhibited significant changes of microbiological phenotypes. The smooth shiny colonies of the WT strain were replaced by rough colonies in the mutant. In addition, the mutant strain aggregated strongly in liquid culture, and no such aggregation was observed with WT cells. More importantly, the mutant grew considerably more slowly (doubling time, 160 min) than the WT (doubling time, 58 min), and it was unable to grow at temperatures above 30°C (whereas its WT counterpart produced colonies on plates incubated at temperature of up to 37°C). All of the unusual phenotypic features of the mutant strain displayed a partial reversion to WT when the mutant strain was transformed with a plasmid, pCGL2308, carrying a 5.3-kb fragment containing the entire pks13 from C. glutamicum plus 417 bp of the upstream region. The colony morphology of C. glutamicum Δpks13::km: pCGL2308 was similar to that of the WT strain, and the complemented strain exhibited an intermediate level of aggregation in liquid culture and grew at 34°C. These complementation results demonstrated that the observed unusual phenotypes were due to the deletion of pks13. Therefore, although a functional pks13 gene was not required for the viability of C. glutamicum, deletion of the pks13 gene resulted in striking phenotypes suggestive of cell envelope modification.

Fig. 2.
Construction of a C. glutamicum Δpks13::km mutant. (A) Schematic representation of the genetic structure of the pks13 locus in WT C. glutamicum and in the Δpks13::km mutant. The boxes indicate the various genes of the pks13 locus. The ...

The Δpks13::km C. glutamicum Mutant Does Not Produce Mycolic Acids and Exhibited an Altered Cell Envelope. We investigated the origin of these phenotypic changes by comparing the fatty acids produced by the WT and mutant strains. Cultures of C. glutamicum WT, the Δpks13::km mutant, and the complemented strain were grown to exponential growth phase and labeled with [14C]acetate. Fatty acids were then released from the bacteria by saponification of whole cells. TLC and GC analysis of the fatty acid methyl esters revealed a complete absence for the C. glutamicum Δpks13::km mutant of both mycolates and palmitone, a degradation product of the 3-oxo intermediate resulting from the condensation reaction (Fig. 1B) that usually appears on alkaline hydrolysis (Fig. 3A and data not shown). However, the Δpks13::km mutant strain produced large amounts of C16-C18 fatty acids, the precursor of corynomycolates (Fig. 3A and data not shown). Complementation restored the production of mycolic acids, but the amount of mycolates was much lower than in the WT strain: the ratio mycolate/C16-C18 fatty acids was 12 times lower (Fig. 3A). These observation were confirmed by GC-MS analysis (Fig. 6, which is published as supporting information on the PNAS web site). Several reasons may explain the observed partial reversion: (i) either the expression of pks13 from the plasmid in the complemented strain was not as strong as that in the WT, (ii) or the chromosomal insertion of the kanamycin cassette exerted a polar effect on the expression of accD4 that may be important for the observed phenotypes, (iii) or both. Nevertheless, these results clearly demonstrated that the Δpks13::km mutant is no longer producing mycolic acids because of the mutation of pks13.

Fig. 3.
Biochemical and ultrastructural analysis of the envelope of C. glutamicum Δpks13::km mutant. (A) Fatty acid contents of WT C. glutamicum, the Δpks13::km mutant strain, and the Δpks13::km mutant strain, complemented with a plasmid ...

In Corynebacterineae, mycolic acids are known to be part of the outer lipid bilayer, which corresponds functionally to the outer membrane of Gram-negative bacteria (5, 6). The presence of this additional membrane is convincingly demonstrated by freeze-fracture electron microscopy in which a fracture plane is seen that propagates exclusively between the two layers of this pseudo-outer membrane in WT cells (27, 28). Comparative analysis of the WT and Δpks13::km mutant cells by this technique showed that the fracture plane propagated in the plasma membrane of the mutant whereas it clearly occurred in the cell wall of the WT cells (Fig. 3B). Thus, the outer lipid bilayer, which consists mostly of cell wall-linked and trehalose mycolates, was not present in the mutant of C. glutamicum, a finding consistent with the lack of production of corynomycolates by the Δpks13::km mutant cells.

These results indicate that C. glutamicum (Δpks13::km) lacks an enzyme essential for the Claisen-type condensation reaction by which mycolic acids are generated from two activated fatty acid molecules, and therefore that Pks13 is involved in this reaction. Three reactions are expected between the two fatty acid molecules and the 3-oxo intermediate resulting from the condensation reaction: (i) activation of the fatty acid substrates by the formation of acyl-CoA, (ii) carboxylation of one acyl-CoA molecule to form alkylmalonyl-CoA, and (iii) condensation of one acyl-CoA molecule with one alkylmalonyl-CoA molecule to form the 3-oxo intermediate (9). The first two enzymatic reactions are catalyzed by well known enzymes, the acyl-CoA synthase and acyl-CoA carboxylase, respectively. Therefore, Pks13, which contains the required enzymatic domains, probably catalyzes the third reaction, the condensation itself. Interestingly, two genes predicted to encode an acyl-CoA synthase (fadD32) and a subunit of an acyl-CoA carboxylase (accD4) flank pks13 in all of the Corynebacterineae analyzed, suggesting that these two genes may encode enzymes involved in the substrate activation reactions. If Pks13 is involved in the condensation reaction itself, one would expect accumulation of a carboxylated intermediate in the Δpks13::km mutant. However, the pks13 and accD4 genes are separated by only 20 bp, suggesting that they form an operon. If so, the km insertion in pks13 may decrease the expression of accD4, preventing the formation of large amounts of carboxylated intermediates, a hypothesis consistent with the partial reversion observed with C. glutamicum Δpks13::km:pCGL2308. Alternatively, the formation of this compound may be strictly regulated.

Pks13 Is Essential for the Viability of Mycobacteria and Is also Involved in Mycolic Acid Biosynthesis in These Bacteria. Mycolic acid biosynthesis is known to be essential for mycobacterial growth (26, 29). We investigated the role of Pks13 in the biosynthesis of mycolic acid in mycobacteria by first considering whether this enzyme was essential. We used a genetic approach in the model strain M. smegmatis mc2155 to address this issue. A nonreplicative vector containing the counterselectable marker sacB (30) and a mutated copy of pks13 was inserted into the chromosome by single crossover between the WT chromosomal allele of pks13 and the mutated allele, to give the strain PMM47 (Fig. 4A). Plating a culture of this strain at various temperatures (25°C, 30°C, or 37°C) on medium containing 10% sucrose and hygromycin generated clones with mutations in the sacB gene but failed to select the second recombination event that would have produced a strain carrying only the mutated pks13::hyg allele. This result suggests that this gene is essential for mycobacterial growth. To demonstrate conclusively that this gene was essential, we transferred a second WT copy of pks13 in a thermosensitive mycobacterial vector into PMM47. In this genetic context, the selection of clones resistant to sucrose and hygromycin gave mutants (≈8% of the SucR, HygR colonies) in which a second recombination event had occurred in PMM47 between the two chromosomal alleles of pks13 leaving only the nonfunctional copy of pks13 on the chromosome (Fig. 4). This recombinant strain, PMM48:pDP32, contains a deletion and insertion in pks13 on the chromosome and a functional pks13 gene on a thermosensitive plasmid. Streaking this recombinant strain on hygromycin plates at 30°C or 42°C revealed that it was unable to form colonies at high temperature. In liquid culture, this strain grew as rapidly as the WT at 30°C, a permissive temperature for plasmid replication. However, if the culture was shifted to 42°C, a nonpermissive temperature for plasmid replication, the number of viable bacteria increased during the first 12–24 h, during which time the temperature-sensitive plasmid was cured and then remained stable during the next 24 h before declining, the only viable bacteria being those retaining a copy of the complementing plasmid (Fig. 5A). These experiments demonstrated that pks13 is essential for the survival of M. smegmatis. We then investigated mycolic acid synthesis in cultures growing at 30°Cor shifted to 42°C. Cultures of WT and conditional mutants were labeled for 3 h, and total lipids were extracted and analyzed by TLC after saponification. This analysis revealed that the level of mycolate production was markedly lower (60% lower) at high temperature in the conditional mutant PMM48:pDP32 than in the WT strain (Fig. 5B). As expected, mycolic acid synthesis was not entirely abolished in the culture as the remaining bacterial population still harboring the complementing plasmid was able to produce mycolic acids, even though the plasmid could not replicate. Nevertheless, these results demonstrated that Pks13 was involved in the biosynthesis of mycolates and essential for the physiology of mycobacteria.

Fig. 4.
Construction of a conditional mutant of M. smegmatis.(A) Schematic representation of the genetic structure of the pks13 genomic locus obtained during construction of the conditional M. smegmatis mutant. The boxes indicate the various genes of the pks13 ...
Fig. 5.
Growth characteristics and mycolic acid contents of the conditional M. smegmatis mutant incubated at 30°C and 42°C. (A) Colony-forming units (cfu) counted during growth of the WT strain or PMM48:pDP32 at 30°C or 42°C. The ...


Pks13 is an enzyme found in all mycolic acid-producing bacteria tested. It is involved in the condensation reaction in corynebacteria and contains the catalytic domains required for the condensation reaction itself and the release of the substrate from the synthase. It is also involved in the biosynthesis of mycolates in mycobacteria. Therefore, taken together, our data reveal that pks13 encodes the condensase, the enzyme catalyzing the condensation of two fatty acids to form mycolic acids. Moreover, we propose that the substrates of the Pks13 condensase are activated by the acyl-CoA synthase, FadD32, and an acyl-CoA carboxylase containing the accD4 protein. The Pks13 enzyme is essential for the survival of mycobacteria and possibly also other Corynebacterineae. It may therefore be seen as a specific target for the development of drugs for the treatment of human infections caused by mycobacteria and possibly other Corynebacterineae pathogens.

Supplementary Material

Supporting Information:


We thank Drs. M.-A. Lanéelle and H. Montrozier for valuable help in the biochemical analysis of the various strains. We are grateful to Dr. A. Engel for his support and enthusiastic discussions about the cryomicroscopy analysis of the various strains. We also thank Dr. A. Quémard for providing the Rhodococcus strain. D.P. holds a Fellowship from the Ministère Délégué à la Recherche et aux Nouvelles Technologies (MDRNT). This work was supported by the Centre National de la Recherche Scientifique (CNRS, France) and the MDRNT [Actions Concertées Incitatives (ACI) “Molécules et cibles thérapeutiques”].


This paper was submitted directly (Track II) to the PNAS office.

Abbreviation: Pks, polyketide synthase.

Data deposition: The sequence of the pks13 locus from Rhodococcus rhodochrous reported in this paper has been deposited in the GenBank database (accession no. AY439008).


1. Anonymous (2000) The WHO/IUATLD Global Project on Antituberculosis Drug Resistance Surveillance (World Health Organization, Geneva).
2. Webb, V. & Davies, J. (1999) in Mycobacteria: Molecular Biology and Virulence, eds. Ratledge, C. & Dale, J. (Blackwell Science, Oxford, U.K.), Vol. 1, pp. 287-307.
3. Daffé, M. & Draper, P. (1998) Adv. Microbiol. Physiol. 39, 131-203. [PubMed]
4. Jarlier, V. & Nikaido, H. (1990) J. Bacteriol. 172, 1418-1423. [PMC free article] [PubMed]
5. Brennan, P. J. & Nikaido, H. (1995) Annu. Rev. Biochem. 64, 29-63. [PubMed]
6. Draper, P. (1998) Frontiers Biosci. 3, d1253-d1261. [PubMed]
7. Dubnau, E., Chan, J., Raynaud, C., Mohan, V. P., Lanéelle, M.-A., Yu, K., Quémard, A., Smith, I. & Daffé, M. (2000) Mol. Microbiol. 36, 630-637. [PubMed]
8. Glickman, M. S., Cox, J. S. & Jacobs, W. R., Jr. (2000) Mol. Cell 5, 717-727. [PubMed]
9. Gastambide-Odier, M. & Lederer, E. (1960) Biochem. Z. 333, 285-295. [PubMed]
10. Mathur, M. & Kolattukudy, P. E. (1992) J. Biol. Chem. 267, 19388-19395. [PubMed]
11. Dubey, V. S., Sirakova, T. D. & Kolattukudy, P. E. (2002) Mol. Microbiol. 45, 1451-1459. [PubMed]
12. Dusch, N., Puhler, A. & Kalinowski, J. (1999) Appl. Environ. Microbiol. 65, 1530-1539. [PMC free article] [PubMed]
13. Belisle, J. T. & Sonnenberg, M. G. (1998) in Mycobacteria Protocols, eds. Parish, T. & Stoker, N. G. (Humana, Totowa, NJ), Vol. 101, pp. 31-44.
14. Peyret, J.-L., Bayan, N., Joliff, G., Gulik-Krzywicki, T., Mathieu, L., Shechter, E. & Leblon, G. (1993) Mol. Microbiol. 9, 97-109. [PubMed]
15. Quandt, J. & Hynes, M. F. (1993) Gene 127, 15-21. [PubMed]
16. Guilhot, C., Gicquel, B. & Martin, C. (1992) FEMS Microbiol. Lett. 98, 181-186. [PubMed]
17. Le Dantec, C., Winter, N., Gicquel, B., Vincent, V. & Picardeau, M. (2001) J. Bacteriol. 183, 2157-2164. [PMC free article] [PubMed]
18. Laval, F., Lanéelle, M.-A., Deon, C., Montsarrat, B. & Daffé, M. (2001) Anal. Chem. 73, 4537-4544. [PubMed]
19. Constant, P., Perez, E., Malaga, W., Lanéelle, M.-A., Saurel, O., Daffé, M. & Guilhot, C. (2002) J. Biol. Chem. 277, 38148-38158. [PubMed]
20. Aggerbeck, L. P. & Gulik-Krzywicki, T. (1986) Methods Enzymol. 128, 457-472. [PubMed]
21. Cole, S. T., Eiglmeier, K., Parkhill, J., James, K. D., Thomson, N. R., Wheeler, P. R., Honore, N., Garnier, T., Churcher, C., Harris, D., et al. (2001) Nature 409, 1007-1011. [PubMed]
22. Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S. V., Eiglmeier, K., Gas, S., Barry, C. E., III, et al. (1998) Nature 393, 537-544. [PubMed]
23. Azad, A. K., Sirakova, T. D., Rogers, L. M. & Kolattukudy, P. E. (1996) Proc. Natl. Acad. Sci. USA 93, 4787-4792. [PMC free article] [PubMed]
24. Puech, V., Guilhot, C., Perez, E., Tropis, M., Armitige, L. Y., Gicquel, B. & Daffé, M. (2002) Mol. Microbiol. 44, 1109-1122. [PubMed]
25. Collins, M. D., Burton, R. A. & Jones, D. (1988) FEMS Microbiol. Lett. 49, 349-352.
26. Vilchèze, C., Morbidoni, H., Weisbrod, T. R., Iwamoto, H., Kuo, M., Sacchettini, J. C. & Jacobs, W. R., Jr. (2000) J. Bacteriol. 182, 4059-4067. [PMC free article] [PubMed]
27. Chami, M., Bayan, N., Dedieu, J., Leblon, G., Shechter, E. & Gulik-Krzywicki, T. (1995) Biol. Cell 83, 219-229. [PubMed]
28. Puech, V., Chami, M., Lemassu, A., Lanéelle, M.-A., Schiffler, B., Gounon, P., Bayan, N., Benz, R. & Daffé, M. (2001) Microbiology 147, 1365-1382. [PubMed]
29. Takayama, K., Wang, L. & David, H. L. (1972) Antimicrob. Agents Chemother. 2, 29-35. [PMC free article] [PubMed]
30. Pelicic, V., Reyrat, J.-M. & Gicquel, B. (1996) Mol. Microbiol. 20, 919-925. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene links
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...