PBlackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382XBlackwell Publishing Ltd, 2005? 200557411131126Original ArticleCharacterization of a mycobacterial monomeromycolyl-diacylglycerolL. Kremer et al .
Molecular Microbiology (2005) 57(4), 1113–1126
doi:10.1111/j.1365-2958.2005.04717.x
Identification and structural characterization of an unusual mycobacterial monomeromycolyldiacylglycerol Laurent Kremer,1*† Chantal de Chastellier,2 Gary Dobson,3 Kevin J. C. Gibson,4 Pablo Bifani,1 Stéphanie Balor,2 Jean-Pierre Gorvel,2 Camille Locht,1 David E. Minnikin4 and Gurdyal S. Besra4 1 Laboratoire des Mécanismes Moléculaires de la Pathogénie Microbienne, INSERM U629, Institut Pasteur de Lille, 1, rue du Prof. Calmette, F-59019 Lille, France. 2 Centre d’Immunologie de Marseille-Luminy, INSERMCNRS-Université de la Méditerranée, 13288 Marseille Cedex 09, France. 3 Department of Chemistry, University of Newcastle upon Tyne, Newcastle upon Tyne, NE1 7RU, UK. 4 School of Biosciences, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK. Summary Systematic thin layer chromatographic (TLC) analysis of apolar lipids in Mycobacterium kansasii revealed the presence of a previously uncharacterized novel component. The product was ubiquitously found in a panel of M. kansasii clinical isolates, as well as other pathogenic and non-pathogenic mycobacterial species. TLC analysis of [14C]-acetate- or [14C]-glycerollabelled M. kansasii cultures tentatively assigned the novel product as an unusual triacylglycerol-related lipid. Subsequent purification, followed by structural determination using 1H-nuclear magnetic resonance (NMR) and electrospray mass spectrometry (ES/MS), led to the identification of this product as a monomeromycolyl-diacylglycerol (MMDAG). Treatment of M. kansasii with either isoniazid (INH), a well-known type II fatty acid synthase (FAS-II) and mycolic acid biosynthesis inhibitor, or tetrahydrolipstatin (THL), a drug approved for treating obesity, correlated with a reduced incorporation of [14C]-acetate into both mycolic acids and MMDAG. Addition of INH or THL to the cultures induced major morphological changes and, surprisingly, resulted in an increased number of lipid storage bodies, as determined by electron Accepted 3 May, 2005. *For correspondence. E-mail laurent. Fax
[email protected]; Tel. (+33) 4 67 14 33 81; (+33) 4 67 42 86. †Present address: Laboratoire de Dynamique Moléculaire des Interactions Membranaires, CNRS-UMR 5539, Université de Montpellier II, Case 107, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France.
© 2005 Blackwell Publishing Ltd
microscopy. The potent antimycobacterial activity of THL was confirmed against a variety of mycobacterial species, including INH-susceptible and -resistant Mycobacterium tuberculosis strains. Therefore, THL and other b-lactones may be promising drugs for the development of new antitubercular therapy. Introduction The mycobacterial cell envelope, also called the ‘waxy coat’, contains a remarkable array of lipids and glycolipids (Minnikin, 1982; Daffé and Draper, 1998; Minnikin et al., 2002; Kremer and Besra, 2004), and has been the subject of intensive research. Few organisms have evolved to produce such a diverse collection of lipophilic structures as in Mycobacterium tuberculosis (Minnikin et al., 2002). The mycobacterial cell wall is unique, consisting of three covalently linked polymers, peptidoglycan, arabinogalactan (AG) and mycolic acids (Brennan and Nikaido, 1995). Peptidoglycan is attached to AG via a phosphodiester bridge and about two-thirds of the non-reducing termini of AG are esterified with mycolic acids (McNeil et al., 1991). This covalently linked cell wall skeleton is often referred to as the mycolyl–arabinogalactan–peptidoglycan complex (mAGP) (Brennan and Nikaido, 1995). It is considered that the cell wall-bound mycolic acids form a monolayer lipid structure whose outer leaflet interacts with a range of unusual ‘extractable’ lipids to form a coherent lipid bilayer (Minnikin, 1982). These complex free lipids range in polarity from apolar triacylglycerols (TAG), phthiocerol dimycocerosates to polar glycopeptidolipids and lipooligosaccharides (Minnikin, 1982; Brennan and Nikaido, 1995). Many more potential lipid biosynthetic activities can be deduced from the M. tuberculosis genome sequence than there are known products in in vitro-grown tubercle bacilli, raising the possibility that many novel lipid species remain to be isolated and characterized. Triacylglycerols are essentially present in animals, plants and eukaryotic microorganisms, such as Plasmodium falciparum (Vielemeyer et al., 2004), as the major lipid storage molecules. Although rather uncommon in prokaryotes, recent work reported TAG accumulation in different bacteria, mainly in those belonging to the actinomycetes genera, including Nocardia (Alvarez and Steinbüchel, 2002), Rhodococcus (Alvarez et al., 2000),
1114 L. Kremer et al. teria, consisting of a novel TAG bearing a meromycolate substituent. This lipid was ubiquitously found in all mycobacterial species tested, including pathogenic species, emphasizing an important function in mycobacterial physiology. Its potential role as a source of meromycolates during mycolic acid biosynthesis is discussed.
Streptomyces (Olukoshi and Packter, 1994) and Mycobacterium (Garton et al., 2002; Daniel et al., 2004). Streptomyces have been observed to accumulate TAG as lipid bodies in mycelia at the stationary phase of growth (Packter and Olukoshi, 1995). It has also been suggested that TAG may act as a carbon source for the production of secondary metabolites, such as antibiotics (Olukoshi and Packter, 1994). Studies on the environmental isolate Rhodococcus opacus strain PD630 have demonstrated the accumulation and mobilization of TAGs into intracytoplasmic lipid inclusions (Alvarez et al., 1996; 2000). A recent mechanism of lipid body formation, different from the current models of lipid inclusion in eukaryotes, has been proposed in Acinetobacter calcoaceticus and R. opacus (Wältermann et al., 2005). Interestingly, previous investigations in the pathogenic M. tuberculosis strain indicated an important role of lipid accumulation in vitro and in human sputum (Garton et al., 2002). The proposed storage role of lipidic inclusions and TAGs may be advantageous for mycobacteria in vivo and be relevant to survival in a non-replicating state. This notion was recently supported with the evidence that diacylglycerol acyltransferases involved in TAG biosynthesis are upregulated in stationary-phase M. tuberculosis cultures and that TAG accumulates in dormant-like cultures (Daniel et al., 2004). Overall, these findings indicate that TAG biosynthesis and accumulation are important aspects of mycobacterial lipid metabolism, and suggest that TAG represents a potential target for treatment of mycobacterial diseases. The work presented in this report was undertaken to identify and characterize new mycobacterial lipids in order to decipher new sources of lipid structural diversity in mycobacteria. We describe here the identification and structural characterization of an unusual lipid in mycobac-
A
Results Identification of a new apolar lipid in Mycobacterium kansasii To identify novel mycobacterial lipids in the clinical strain Mycobacterium kansasii PHRI 901, the mycobacteria were grown in the presence of [1,2-14C]-acetate, and the total apolar and polar lipids were extracted and analysed by thin layer chromatography (TLC) spanning the whole range of lipid polarities (Besra, 1998). Figure 1A shows a typical two-dimensional TLC (2D-TLC) profile of total apolar lipids from M. kansasii PHRI 901, characterized by an intense spot (surrounded by a circle) known to correspond to TAG, according to Dobson et al. (1985). Figure 1B illustrates the purity of the material corresponding to this spot, following extraction with diethyl ether and resolution using petroleum ether/ethyl acetate (98/2) thrice in the first dimension and petroleum ether/acetone (98/2) in the second dimension. When another solvent system was used [petroleum ether/acetone (95/5) in the first dimension and toluene/acetone (99/1) in second dimension], the purified TAG shown in Fig. 1B gave rise to a second lipid component, referred to as ‘apolar lipid’ in addition to the major spot identified as TAG (Fig. 1C). This new product appeared to be less polar than TAG, suggesting that it may be a related more hydrophobic TAG.
C
B
apolar lipid TAG
TAG TAG
2
2 1
2 1
1
Fig. 1. Identification of a new apolar lipid in M. kansasii PHRI 901. Cells were labelled with [1,2-14C]-acetate for 16 h, and apolar lipids were extracted as described in Experimental procedures. (A) Characteristic apolar lipid profile obtained by loading approximately 200 000 cpm onto a silica gel plate and separating the labelled lipids by 2D-TLC using petroleum ether/ethyl acetate (98/2) in the first dimension (¥3) and petroleum ether/acetone (98/2) in the second dimension. Circled material was extracted from a preparative plate, solubilized in diethyl ether, and approximately 50 000 cpm of the purified lipids was subjected to 2D-TLC analysis using in the first dimension petroleum ether/ethyl acetate (98/2, ¥3) and in the second dimension petroleum ether/acetone (98/2) (B), or petroleum ether/acetone (95/5) in the first dimension and toluene/acetone (99/1) in the second dimension (C). © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1115 mycobacterial species by 2D-TLC/autoradiography. Figure 2B shows that the product is present in Mycobacterium gastri ATCC 15754, M. bovis BCG 1173P2, M. smegmatis mc2155 and M. thermoresistibile ATCC 19527. This lipid is thus not unique to M. kansasii but ubiquitous in mycobacterial species, suggesting that it might be involved in important physiological functions.
Lipid distribution in M. kansasii clinical isolates and other mycobacterial species To investigate whether the presence of this novel apolar lipid was restricted to M. kansasii PHRI 901, we analysed the distribution of this product in a collection of 31 M. kansasii clinical isolates by one-dimensional TLC (1DTLC) using toluene/acetone (99/1) after extraction of the apolar lipids from acetate-labelled cultures. As illustrated in Fig. 2A for a representative TLC of 12 independent isolates, the apolar lipid was found in all strains examined. We then analysed the distribution of this lipid in other
Structural analysis of the ‘apolar lipid’ Extraction of the apolar lipids from a culture of M. kansasii labelled with [UL-14C]-glycerol and analysis by TLC/auto-
A
apolar lipid TAG
origin PHRI K6 901
K12
K99
K269 K400 K402 K406 K423 K429 K430 K431
B apolar lipid
apolar lipid
TAG
TAG
apolar lipid TAG
2
2
2 1
1
M. kansasii
1
M. gastri
M. smegmatis
apolar lipid
apolar lipid
TAG
TAG
2
2 1
M. bovis BCG
1
M. thermoresistibile
© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Fig. 2. Distribution of the new apolar lipid in M. kansasii clinical isolates and other mycobacterial species. Mid-log phase cultures were labelled with [1,2-14C]-acetate for 24 h, and apolar lipids were extracted as described in Experimental procedures. A. Representative 1D-TLC/autoradiography of 12 clinical isolates of M. kansasii. Approximately 100 000 cpm were loaded in each lane. Total apolar lipids were developed in toluene/ acetone (99/1) and exposed overnight. B. 2D-TLC/autoradiography showing the apolar lipid profile in M. kansasii PHRI 901, M. gastri ATCC 15754, M. smegmatis mc2155, M. bovis BCG 1173P2 and M. thermoresistibile ATCC 19527, following development in petroleum ether/acetone (95/5) in the first dimension and toluene/acetone (99/1) in the second dimension. Approximately 100 000 counts was loaded for each TLC plate.
1116 L. Kremer et al. radiography revealed that both the TAG and the apolar lipid were labelled (data not shown), indicating that the latter also contained a glycerol moiety. This component was tentatively referred to as ‘apolar-TAG’. To determine the structure of apolar-TAG, 1H-nuclear magnetic resonance (1H-NMR) spectra of both apolarTAG and TAG purified from M. kansasii ATCC 12478 were determined and were found to be similar to that of a commercial standard glycerol-tritetradecanoate (trimyristin). For trimyristin, the following resonances were present: d0.85 (terminal CH3, 9H, triplet), 1.23 (CH2, broad), 1.58 (CH2·CH2·COO, 6H, distorted triplet), 2.28 (CH2·COO, 6H, two overlapping triplets), 4.09, 4.13, 4.25, 4.28 (CH2·OOC, 4 1H, four doublets), 5.23 (CH·OOC, 1H, multiplet). This indicated that the latter resulted from the CH2·COO group esterified at the 2-position of the glycerol, whereas the former resulted from these groups esterified at the 1- and 3-positions of glycerol. These signals were also present in the 1H-NMR spectra of the two M. kansasii TAG components, although their chemical shift were slightly more downfield, as shown in Fig. 3A and B for resonances a–m. The precise values for these signals were d0.87 (d), 1.23 (e), 1.59 (f), 2.30 (h), 4.11 (i), 4.15 (j), 4.26 (k), 4.30 (l), 5.25 (m) and 0.87 (d) 1.23 (e), 1.59 (f), 2.30 (h), 4.12 (i), 4.15 (j), 4.27 (k), 4.30 (l), 5.25 (m) for TAG (Fig. 3A) and apolar-TAG (Fig. 3B) respectively. Additionally, both spectra contained resonances indicative of double bonds at d2.0 (g) (CH2CH=CH2, 4H, multiplet) and d5.33 (n) (cis-CH=CH, 2H, distorted triplet). The integration of these signals suggested that monounsaturated fatty acids accounted for approximately one-third of the total fatty acids in both TAG forms, assuming that fatty acids with a greater degree of unsaturation were absent, which was supported by gas chromatography/mass spectrometry (GC/MS) analyses (see below). The spectrum (Fig. 3B) of apolar-TAG contained multiplets at d0.35 (a), 0.55 (b) and 0.63 (c), indicating the presence of cis-cyclopropane rings, which are suggestive of the mero-chain portion of mycolic acids. Analysis by electrospray mass spectrometry (ES/MS) revealed a series of sodiated molecular ions, m/z 797.7/ 799.8–909.9/911.8, differing by 28 amu for TAG (Fig. 4). Several attempts to obtain similar data for apolar-TAG were unsuccessful. Alkaline deacylation followed by methylation of both TAG forms released petroleum ether-extractable fatty acids, which co-chromatographed on TLC with a C12-C24 straight-chain fatty acid methyl ester (FAME) standard [Rf 0.46; petroleum ether/acetone (98/2, v/v)]. Apolar-TAG also gave rise to a second FAME, which was less polar (Rf 0.64) than the C12-C24 FAME standard. The results of the GC and GC/MS analysis of the FAMEs, migrating on TLC (Rf 0.46) with the standard, are shown in Table 1. In both TAG and apolar-TAG, C16 was a major component
Table 1. Fatty acid composition found in TAG and apolar-TAG from M. kansasii. FAME
TAG
Apolar-TAG
C14 C16:1 C16 C18:1 C18 TSA C20 C22 C24 C26
5.6 2.8 8.8 31.0 13.3 2.0 3.8 6.5 24.7 1.5
6.6 4.2 9.5 51.6 16.5 2.9 – 3.9 4.8 –
The percentage of each FAME was obtained after alkaline deacylation and methylation followed by GC/MS analysis. The involatile apolar FAMEs released from the apolar-TAG did not produce a GC and GC/MS profile. TSA, tuberculostearic acid.
with substantial amounts of C18 and C18:1. Only the FAMEs from TAG contained large amounts of C20-C26 FAMEs, with C24 particularly abundant. The involatile apolar FAMEs released from the apolar-TAG did not produce a GC and GC/MS profile. The 1H-NMR spectra of the apolar FAMEs released from apolar-TAG possessed characteristic resonances of a cis-cyclopropane ring-containing FAME, originally observed in the 1H-NMR spectrum of native apolar-TAG (Fig. 3B). Electron impact/mass spectrometry (EI/MS) analysis showed a series of peaks ranging from m/z 672– 840, and differing by 14 amu, corresponding to the molecular ions of dicyclopropyl FAMEs ranging from 45 to 57 carbons in the parent fatty acid. The major peak was at m/z 756, corresponding to a fatty acid of 51 carbons, and peaks corresponding to odd carbon-chain fatty acids also predominated. The 1H-NMR spectrum of native apolarTAG suggested one meromycolate chain per molecule of apolar-TAG. This implies that the apolar lipid was a monomeromycolyl diacylglycerol (MMDAG), which would account for its apolar characteristics relative to TAG (Fig. 5). Expression profile of TAG and MMDAG during growth of M. kansasii Work performed in M. smegmatis suggested that TAG accumulation was essentially dependent of the medium as well as the age of the culture (Garton et al., 2002). However, little is know about the time-course of these lipids in other mycobacteria. We therefore examined whether growth phase may affect TAG/MMDAG production in M. kansasii. Figure 6A illustrates the growth curve of M. kansasii PHRI 901, characterized by a dense culture after 2 weeks of incubation in Sauton medium. Samples were collected at various time points during growth, and © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1117
A
B
Fig. 3. 300 MHz NMR analysis of TAG (A) and apolar-TAG (B). The 1H-NMR spectra of purified TAG and apolar-TAG from M. kansasii PHRI 901 were obtained in CDCl3 at a concentration of 5 mg ml-1 on a Bruker 300 MHz instrument.
labelled with [14C]-acetate for 16 h. Apolar lipids were then extracted and resolved by 2D-TLC/autoradiography. Figure 6B shows that both the TAG and the MMDAG pools increased significantly during growth, especially during the late log phase. This is in agreement with findings suggesting that TAG accumulated in M. tuberculosis cells going into the non-replicative state (Daniel et al., 2004). These results suggest that production of MMDAG and TAG are growth state-dependent. © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Biosynthesis of the MMDAG is type I fatty acid synthase (FAS-I)-independent In mycobacteria, two types of fatty acid synthase (FAS) systems coexist, FAS-I and FAS-II (Kremer et al., 2000). Cell-free experiments demonstrated that, FAS-I provides the C12 acyl primers used by FAS-II for further elongation, yielding long-chain C14-C26 fatty acids up to C56 meromycolic acids, which are the precursors of mycolic acids.
1118 L. Kremer et al. Fig. 4. ES/MS analysis of TAG. The purified TAG from M. kansasii PHRI 901 were analysed using ES/MS in the positive ion mode on a triple quadropole LCT instrument (Micromass, Altrincham, UK) fitted with an atmospheric pressure electrospray source. The samples were directly injected using a Rheodyne injector with methanol as the mobile phase. The machine was run at a flow rate of 200 ml min-1 with the cone voltage at 35 V and the spraying needle voltage set at 3 kV. The scan rate was 1 s, and the mass range was from 200 to 2000.
Cerulenin (CER), an inhibitor of mycobacterial FAS-I (Parrish et al., 1999; Boshoff et al., 2002), was used to determine the role of FAS-I in TAG and MMDAG biosynthesis. When M. kansasii cultures were pre-treated with CER, the incorporation of label into the fatty acids of TAG was drastically diminished (Fig. 7A), supporting the notion that the acyl chains of TAG originate from FAS-I. In contrast, CER treatment resulted in the accumulation of MMDAG, suggesting that the glycerol-bound acyl chains of MMDAG, in contrast to TAG, were synthesized independently of FAS-I, presumably via FAS-II. Inhibition of MMDAG synthesis by FAS-II blocking drugs As MMDAG contains a meromycolate substituent bound
A
B
Fig. 5. Essential structure of apolar-TAG. A. The basic triacylglycerol structure, with R and X¢ or X¢¢ as longchain fatty acids (see Table 1). B. The dicyclopropyl meromycolate analogue, esterified at X¢ or X¢¢; x + y + z = 37–49, with main component of 43.
to a glycerol moiety, and as meromycolates are direct precursors of mycolic acids, we examined the effect of the mycolic acid biosynthesis inhibitor INH on MMDAG biosynthesis. M. kansasii PHRI 901, highly susceptible to killing by INH (Table 2), was treated with increasing concentrations of INH and labelled with [1,2-14C]-acetate. Mycolic acid methyl esters (MAMEs) and FAMEs were extracted and analysed by TLC. As expected, INH affects the synthesis of a, methoxy and ketomycolates (Fig. 7B). Synthesis of the MAMEs was completely abrogated at 5 mg ml-1 of INH, whereas FAMEs remained unaffected, as previously described for other mycobacterial species (Vilchèze et al., 2000; Kremer et al., 2003). Interestingly, a more apolar band which migrates ahead of the FAMEs (Fig. 7B) is also inhibited
Table 2. MICs of INH and THL against various mycobacterial species (in mg ml-1). Strain
INH
THL
M. kansasii PHRI 901 M. bovis BCG 1173P2 M. tuberculosis W (INH-resistant) M. tuberculosis W4 (INH-susceptible) M. marinum ATCC 927 M. smegmatis mc2155 M. thermoresistibile ATCC 19527
0.4–0.8 < 0.5 2 0.02–0.1 >5 5 10
30 30 < 30 < 30 40 > 250 50
© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1119
A
B
18
6
16
OD (600 nm)
1 MMDAG
5
14
2
TAG
MMDAG TAG
3
MMDAG TAG
12 2
2
10
4
8 6 4
1
2
4 MMDAG
3
1
4
8
12
16
6
MMDAG TAG
TAG
2
2
0
1
5 MMDAG
TAG
2
0
2 1
1
2 1
1
20
incubation time (days) Fig. 6. Influence of the growth state on TAG/MMDAG production in M. kansasii. A. Growth of M. kansasii PHRI 901 in Sauton medium (without shaking) was monitored by measuring the optical density at 600 nm. Samples were collected for analysis of the apolar lipid profile at various time points as indicated (from 1 to 6). B. Cultures were labelled with [1,2-14C]-acetate for 16 h at various time points as indicated by the numbers (from 1 to 6). Following extraction, apolar lipids were developed in petroleum ether/acetone (95/5) in the first dimension and toluene/acetone (99/1) in the second dimension. Approximately 50 000 counts were loaded for each TLC plate and exposed overnight.
by INH (and THL see below), which are possibly very long-chain FAMEs, presumably meromycolates. The biosynthesis of MMDAG was also inhibited upon INH treatment, while synthesis of TAG remained relatively unaffected (Fig. 7C). As INH inhibits FAS-II activity, the C45-C57 meromycolate substituent of MMDAG are most likely synthesized via FAS-II, consistent with the fact that MMDAG synthesis is insensitive to the FAS-I inhibitor CER. Tetrahydrolipstatin (THL) belongs to a well-characterized family of esterase inhibitors, which have been shown to be selective and irreversible inhibitors of pancreatic lipases (Hadváry et al., 1991). The common structural feature of THL is the reactive b-lactone ring (Fig. 7D). Given the structural analogy between THL and mycolic acids, we examined whether THL may inhibit the biosynthesis of mycolic acids and MMDAG. As shown in Fig. 7B, THL partially inhibited the incorporation of acetate in mycolic acids. Therefore, compared with INH for which a complete cessation of mycolic acid production was observed, THL inhibited only partially mycolic acid biosynthesis. A plateau of inhibition was achieved at around the minimal inhibitory concentration (MIC) value (40 mg ml-1) and no further inhibition could be observed even at higher THL doses (up to 250 mg ml-1, data not shown). As observed for INH, THL did not appear to inhibit TAG biosynthesis. However, as shown in Fig. 7C, THL also inhibited the incorporation of acetate into MMDAG in a dosedependent manner. © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Antimycobacterial activity of THL The inhibitory properties of THL on the synthesis of mycolic acids and on meromycolate analogues prompted us to investigate whether THL exhibits antimycobacterial activity. The MIC of THL for various mycobacteria, including pathogenic and non-pathogenic species (Table 2), was determined. The MIC value for M. kansasii and M. bovis BCG was approximately 30 mg ml-1. Interestingly, the saprophytic strain M. smegmatis was found to be resistant to THL treatment with an MIC value >250 mg ml-1. We also examined the effect of THL on M. tuberculosis strains and found that the drug inhibited growth of both INH-sensitive and INH-resistant M. tuberculosis, with MIC values of about 30 mg ml-1 in both cases (Table 2). Thus, THL and related b-lactones may potentially be used as new antitubercular drugs effective even against INHresistant strains.
Effect of INH and THL on the ultrastructural appearance of M. kansasii It was previously reported that the treatment of susceptible strains of M. aurum and M. tuberculosis with INH results in different morphological changes that can be visualized by electron microscopy (EM) (Bardou et al., 1996). To determine whether INH or THL treatment had any incidence on the morphological appearance of M. kansasii PHRI 901, exponentially growing bacteria were
1120 L. Kremer et al.
CER 0 0.2 1 10
A
INH
B
0 0.1 0.5 5
THL 0 20 40 100
FAMEs
MMDAG MAMEs TAG
a M K
origin
C
INH 0 0.1 0.5 5
origin
D
THL 0 20 40 100
MMDAG NHC HO
TAG O
O
O
O
origin Fig. 7. Effect of drug treatment on the biosynthesis of TAG and MMDAG and mycolic acids. A. M. kansasii PHRI 901 cultures were treated with the indicated concentrations of CER (in mg ml-1) for 8 h and then incubated with [1,2-14C]acetate for another 16 h. Apolar lipids were then extracted as described in Experimental procedures and equal counts (50 000 cpm) were loaded in each lane. The TLC plate was developed in toluene/acetone (99/1). B. The inhibitory effect of either INH or THL on the incorporation of [1,2-14C]-acetate into mycolic acids was assayed by labelling M. kansaii PHRI 901 in the presence of the indicated drug concentrations (in mg ml-1). The FAMEs and MAMEs (a, alpha; M, methoxy; K, keto) were extracted, and equal counts (200 000 cpm) were subjected to TLC. C. The effect of INH and THL treatment was also analysed with respect to TAG and MMDAG production. Cultures were treated with INH or THL and labelled with [1,2-14C]-acetate as described above. Total apolar lipids were extracted, and equal counts (100 000 cpm) were subjected to TLC. D. Chemical structure of THL.
either left untreated or treated for 24 h with either drug before being fixed and processed for EM. Whole bacteria were first observed either by transmission electron microscopy (TEM) after negative staining or by scanning electron microscopy (SEM). Bacteria kept their rod shape and no major differences in cell width or length were observed between untreated and INH- or THL-treated bacteria. For all three cases, the size of the bacteria was heterogeneous in the sense that about 10% of the bacteria were either more rounded or much longer than the bulk of the population. In contrast, drug treatments induced surface modifications, i.e. the appearance of small blebs at the surface of INH-treated M. kansasii, consistent with
previous observations for M. smegmatis- or M. tuberculosis-treated cultures (Takayama et al., 1973; Vilchèze et al., 2000), and small depressions in THL-treated ones (Fig. 8). In addition, observation of thin sections by TEM showed major differences in terms of degradation. Only about 10% of the population of untreated bacteria were degraded, as defined by the integrity of the cell wall, cytoplasmic membrane or cytoplasm (Fréhel et al., 1997), whereas up to 40% of the treated bacteria were in the process of being degraded. Given the inhibitory effect of both drugs on mycolic acid and MMDAG biosynthesis, we next focused our attention on the appearance of the cell wall and intracellular lipid bodies, which have been pro© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1121 transversal bacterial profile (Fig. 10). The latter were usually large and/or distorted. Addition of Malachite green to the fixatives in order to stabilize and stain lipids confirmed that the inclusions contained lipids (Fig. 9D). In some instances, the entire cytoplasm was filled with small lipid bodies, suggesting that the bacteria stored increasingly larger amounts of lipids following drug treatment. Discussion The present study describes the isolation of an unusual TAG-related molecule, characterized by the presence of a meromycolate substituent esterified to glycerol (Fig. 5), which we named MMDAG. Although very minor as compared with the overall pool of TAG, MMDAG was detected in a panel of 31 independent clinical isolates of M. kan-
A ILI ILLBI
ILLBI
A
B
ILI
L ILBI
B
C
Fig. 8. Morphological appearance of whole bacteria under the scanning electron microscope after treatment with INH or THL. Bacteria were either left untreated (A) or treated with 40 mg ml-1 THL (B) or 5 mg ml-1 INH (C) for 24 h before fixation and processing for SEM. The surface of untreated bacteria (A) is smooth, whereas that of THLtreated bacteria (B) displays small depressions (arrows) and that of INH-treated bacteria (C) displays small blebs (arrows). Bar = 1 mm.
posed to serve as fatty acid storage compartments under certain stress conditions (Garton et al., 2002). A striking difference was observed in the appearance (Fig. 9) and the number (Fig. 10) of lipid bodies between the untreated and treated groups. In untreated bacteria, 90% of the transversal profiles displayed between 0 and 2 lipid bodies that were usually rounded and small (Fig. 9A). Treatment with either INH (Fig. 9B) or THL (Fig. 9C and D) induced a clear shift towards higher values of lipid bodies per © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
ILLIB
ILI
C
D
Fig. 9. Morphological changes of INH- and THL-treated M. kansasii as observed by transmission electon microscopy. Bacteria were either left untreated (A) or treated with 5 mg ml-1 INH (B) or 40 mg ml-1 THL (C and D) for 24 h before fixation and processing for TEM. A. Thin sections of untreated bacteria display between 0 and 2 electron translucent lipid bodies. B and C. (B) INH-treated bacteria and (C) THL-treated bacteria: most of the bacteria display several irregularly shaped or large lipid bodies. D. Malachite green was added during the fixation steps of THL-treated bacteria: the bodies appear denser than in (C), thereby confirming the presence of lipids in the inclusions. LB, lipid body. Bar = 0.5 mm.
1122 L. Kremer et al.
Profiles (%)
70
I: 0 II: 1-2 III: 3-4 IV: 5-6 V: 7-8 VI: ≥9
control
60 50 40 30 20 10 0
I1
2 II
3 III
IV
V VI
Profiles (%)
70
INH
60 50 40 30 20 10 0
I1
II2
III IV 3 4
V
VI
5 V
6 VI
Profiles (%)
70
THL
60 50 40 30 20 10 0
I1
II2
3 III
4 IV
Class of intensity (lipid bodies/bacterial profile) Fig. 10. Histogram distribution of the number of lipid bodies per bacterial cell profile. The number of lipid bodies was scored in 100–150 different transversal sections of untreated (top), INH-treated (middle) or THL-treated (bottom) M. kansasii. Data are expressed as the percentage of cell profiles containing a given number of lipid bodies as indicated in the classes of intensity (from I to VI). Only morphologically intact, and therefore viable bacteria were considered, and care was taken to avoid serial sections of the same bacterium.
sasii, as well as in several other mycobacterial strains, including pathogenic and non-pathogenic species, suggesting that this molecule may play an important physiological role. Actinomycetes, including Mycobacterium, accumulate large amounts of TAG, providing a carbon and
energy source for maintaining viability during nutrient starvation (Alvarez et al., 2000; Alvarez and Steinbüchel, 2002). TAG represent a source of fatty acids for the biosynthesis of phospholipids, as well as a carrier of unusual fatty acids involved in regulating fatty acid composition of membrane lipids. Therefore, in addition to serving as an inert lipid deposit, TAG plays an important role in the metabolism of lipid-accumulating bacteria. It has also been suggested that the occurrence of significant amounts of TAG in mycobacteria could be one of the factors that determine the wide distribution of these microorganisms in nature and their ability to cope with adverse environmental conditions, providing them with an evolutionary advantage over other bacteria (Alvarez and Steinbüchel, 2002). TAG is also the major component of mycobacterial lipid bodies (Garton et al., 2002). Given the structural relatedness between MMDAG and TAG, it is very likely that MMDAG is also stored in lipid bodies, although this remains to be clearly demonstrated. The latter have been shown to sequester and store fatty acids, especially when conditions become unfavourable for mycobacterial growth (Garton et al., 2002). Our results show that M. kansasii accumulates both TAG and MMDAG during exponential growth, especially during late log phase. This growth phase dependency is consistent with the observation from Daniel et al. (2004) showing that TAG synthesis is upregulated in stationary-phase M. tuberculosis. Mycolic acid biosynthesis occurs intracellularly, but the final products are mainly localized within the cell wall. Takayama and Armstrong (1976; 1977) noted an early labelling of 6-mycolyl-6¢-acetyl trehalose (MAT) in M. tuberculosis, and proposed that MAT is an intermediate carrier for newly synthesized mycolates, which are then transferred to trehalose dimycolate (TDM) and AG. In addition, a 6-mycolyl-b-D-mannopyranosyl-phosphopolyprenol has also been isolated from M. smegmatis and proposed to serve as a mycolyl carrier that transfers mycolic acid residues to trehalose monomycolate (TMM), TDM and cell wall AG (Besra et al., 1994). By analogy with these molecules, MMDAG may represent another intermediate ‘carrier’ of newly synthesized meromycolates. However, the main meromycolate component of the dicyclopropyl a-mycolates from M. kansasii has x = 17, y = 14, z = 17 (Fig. 5) (Watanabe et al., 2002), corresponding to an even-numbered meromycolate with 56 carbons. There appears to be no direct relationship therefore between the main odd-numbered 51 carbon component of the MMDAG (Fig. 5). Further detailed experiments are required to elucidate the precise relationship between MMDAG and the other mycolate-containing entities present in mycobacteria. As MMDAG appear being produced in exponentially growing bacteria, and as replicating mycobacteria are believed to actively synthesize mycolic acids, our results suggest that MMDAG may rep© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1123 resent a ‘carrier’ of mycolic acid precursors in growing cells. The control of tuberculosis is seriously hampered by the emergence of multidrug-resistant strains of M. tuberculosis. Therefore, new approaches to the treatment of tuberculosis are needed (Kremer and Besra, 2002). TAG and/ or MMDAG may constitute attractive targets for novel drug development, as alterations of these compounds is expected to affect lipid metabolism and/or mycolic acid synthesis or transfer of mycolic acids to the cell wall and thus the mycobacteria’s ability to survive in a non-replicative state of dormancy. We provide evidence that THL exhibits antimycobacterial activity against various pathogenic strains, including INH-sensitive and INH-resistant M. tuberculosis. THL is known to inhibit digestive lipases, including pancreatic lipases from several species and human gastric lipases (Borgström, 1988; Hadváry et al., 1991). It is currently used for the treatment of obesity, and has been associated with significantly greater weight loss than that seen with dieting alone (Thearle and Aronne, 2003). Our results suggest that treatment with THL was correlated with partial inhibition of mycolic acid biosynthesis, although to a lesser extent than INH, which completely abrogated mycolic acid biosynthesis. In addition, both the amount and the size of lipid bodies increased after treatment with either INH or THL, suggesting that precursors of mycolic acids accumulate within these bodies, possibly as a consequence of mycolic acid biosynthesis inhibition. Vilchèze et al. (2000) demonstrated that treatment of M. smegmatis with INH is sufficient to induce the accumulation of FAS-I end-products. The absence of TAG accumulation following drug treatment suggests that the FAS-I end-products may be either esterified to molecules different from TAGs and subsequently stored within lipid bodies or eventually found to be present as free lipids as proposed by Garton et al. (2002). This hypothesis is in line with early work of Gale and McLain (1963) demonstrating that treatment of M. smegmatis with ethambutol was accompanied by an increase in the size of lipidic inclusions, presumably due to the accumulation of mycolic acid precursors, as ethambutol inhibits mycolic acid transfer into the cell wall (Takayama et al., 1979). Alternatively, these inclusions might contain wax esters (consisting of long-chain fatty alcohols and long-chain fatty acids), representing another class of neutral lipids observed in M. tuberculosis (Wang et al., 1972) and M. smegmatis (Kalscheuer et al., 2003). Minimal inhibitory concentration (MIC) values of M. bovis BCG strains overexpressing different enzymes involved in mycolic acid biosynthesis (the b-ketoacyl ACP synthase KasA, the b-ketoacyl ACP reductase MabA and the enoyl ACP reductase InhA; Kremer et al., 2000) were found to be similar to the wild-type strain (data not shown). This indicates that overexpression of these FAS-II © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
enzymes did not induce THL resistance, thus ruling out the possibility of these enzymes being in vivo THL targets. The fact that mycolic acid biosynthesis was only partially inhibited by THL, even at concentrations at least eightfold higher than the MIC value, suggests that this drug may have additional targets. Interestingly, the use of microarray experiments to determine the gene expression profile of M. tuberculosis in response to THL has revealed that different lipase-encoding genes were upregulated, along with genes encoding putative transporters and transcriptional regulators (Waddell et al., 2004). Of particular interest is the induction of Rv3855 (ethR) encoding the transcriptional repressor of the monooxygenase gene ethA, involved in the activation of ethionamide, a drug inhibiting mycolic acid synthesis (Baulard et al., 2000; DeBarber et al., 2000; Vannelli et al., 2002). Kalscheuer and Steinbüchel (2003) recently identified a new family of bifunctional enzymes expressing wax ester synthase (WS)/acyl-CoA:diacylglycerol acyltransferase (DGAT) activities, which are involved in the biosynthesis of wax ester and TAG. This family is widely distributed within all members of the Mycobacterium genus and M. tuberculosis H37Rv possesses 13 genes coding WS/DGAT-related proteins. These genes were individually expressed in Escherichia coli, and it was found that Rv3130c shows the highest DGAT activity (Daniel et al., 2004). We therefore investigated whether THL may inhibit the activity of Rv3130c. The corresponding gene was cloned into a pET28b vector, and the recombinant protein was expressed in E. coli, purified by affinity chromatography, and tested for DGAT activity in the presence of 1,2-dioleoyl-sn-glycerol, as described by Daniel et al. (2004). No inhibition of the DGAT activity was observed in the presence of THL, suggesting that DGAT are not THL targets (data not shown). Recent studies have reported a potent antitumour activity of THL by inhibiting the thioesterase domain of eukaryotic FAS-I, an enzyme strongly linked to tumour progression (Knowles et al., 2004; Kridel et al., 2004). The thioesterase domain allows for the liberation of palmitate from FAS-I after completion of fatty acid biosynthesis. We demonstrate here that treatment of M. kansasii cultures with THL was not accompanied by a reduction of fatty acids originating from the FAS-I system, thus excluding FAS-I as a primary target of THL in mycobacteria. Careful inspection by SEM of drug-treated mycobacteria revealed different morphological changes, characterized by the appearance of small blebs at the surface in the presence of INH-treated bacteria and small depressions in THLtreated cells. This features are consistent with the fact that INH and THL do not share the same target(s). The possibility that the antimycobacterial activity of THL results from the combined action of several different targets is particularly interesting, as the presence of multiple targets
1124 L. Kremer et al. would reduce the frequency of appearance of THL resistance in M. tuberculosis. Future work may lead to the development of more potent analogues of THL for the treatment of mycobacterial infections. Experimental procedures Strains and culture conditions All mycobacterial strains were grown on Middlebrook 7H11 agar plates supplemented with 10% oleic acid-albumin-dextrose-catalase (OADC; Difco) or under shaking in Sauton medium at 37∞C. M. kansasii PHRI 901 was isolated from an HIV-positive patient. Thirty-one M. kansasii clinical strains were isolated primarily from immunocompromised patients in eight New York City hospitals between March 1999 and April 2000. M. kansasii was cultured from induced or natural sputum, broncho-alveolage, lymphatic fluid and stool.
Drug susceptibility testing The MIC of INH and THL was determined by serial 10-fold dilutions of the mycobacterial cultures on solid Middlebrook 7H11 medium containing 10% OADC. After incubation at 37∞C the number of colony-forming units (cfu) were determined, and the MIC was defined as the lowest concentration of the drug resulting in a 99% reduction of cfu.
either INH, THL or CER were added at various concentrations followed by incubation at 37∞C for 8 h. 1 mCi ml-1 [1,214 C]-acetate (50–62 mCi mmol-1, Amersham) was then added to the cultures followed by further incubation at 37∞C for 16 h. The 14C-labelled apolar lipids were extracted and analysed as described above. Alternatively, to extract fatty acids and mycolic acids, the 14C-labelled cells were harvested by centrifugation at 2000 g and washed with phosphate-buffered saline (PBS). The 14C-labelled FAMEs and MAMEs were then extracted as described previously (Kremer et al., 2002). Equal counts were subjected to TLC, developed in petroleum ether/acetone (95/5; v/v) and exposed overnight to Kodak X-Omat film.
Purification of TAG and newly identified lipid The apolar lipids were fractionated by flash column chromatography at medium pressure using Fisher (UK) silica gel 60 (35–70 microns) equilibrated in petroleum ether. Both the TAG and the MMDAG from M. kansasii were isolated using 2.5% ethyl acetate in petroleum ether and further separated using a preparative TLC (plastic-backed silica gel 60 F254, Merck, Darmstadt, Germany). The plates were run several times in toluene/acetone (99/1; v/v), monitored with ethanolic rhodamine 6G as a stain and visualized under long wave (366 nm) UV light. The compounds were then scraped from plates and extracted using diethyl ether and finally evaporated to dryness.
Extraction and analysis of mycobacterial apolar lipids Saturated mycobacterial pre-cultures were used to inoculate fresh Sauton medium (1/50 dilution) and all cultures were grown to mid-log phase before the addition of 1 mCi ml-1 [1,214 C]-acetate (50–62 mCi mmol-1, Amersham) or 1 mCi ml-1 [UL-14C]-glycerol (153.5 mCi mmol-1, Sigma) for 16–24 h. The non-polar fraction was separated from the polar fraction according to Dobson et al. (1985). The [1,2-14C]-acetatelabelled apolar lipids were extracted by adding 2 ml of CH3OH/ 0.3% NaCl (100/10; v/v) and 2 ml of petroleum ether to the cell pellet followed by stirring for 30 min. After centrifugation, the upper petroleum ether layer was removed, and 2 ml of petroleum ether was added to the lower phase, and the process was repeated once. The combined petroleum ether extracts were then evaporated under nitrogen to yield apolar lipids that were resuspended in CH2Cl2 and analysed by 2DTLC using silica gel plates (5735 silica gel 60F254; Merck, Darmstadt, Germany), as reported earlier (Besra, 1998). For the first dimension, three developments of petroleum ether/ ethyl acetate (98/2; v/v) were used and for the second dimension, petroleum ether/acetone (98/2; v/v) was used. TAG, as well as the newly identified apolar lipid were separated by either 1D-TLC using toluene/acetone (99/1; v/v) or 2D-TLC using petroleum ether/acetone (95/5; v/v) in the first dimension and toluene/acetone (99/1; v/v) in the second dimension. TLCs were exposed to a Kodak X-Omat film for 24 h.
In vivo effect of drug treatment on mycolic acid and apolar lipid biosynthesis in M. kansasii Mycobacterium kansasii was grown to mid-log phase, and
Structural determination 1
H-NMR spectra for the purified samples dissolved in CDCl3 were obtained on a Bruker 300 MHz instrument. The intact TAG and the novel lipid were analysed using ES/MS. The measurements were carried out in the positive ion mode on a triple quadropole LCT instrument (Micromass, Altrincham, UK) fitted with an atmospheric pressure electrospray source. The samples were directly injected using a Rheodyne injector (Rheodyne Europe GmbH, Bensheim, Germany) with methanol as the mobile phase. The machine was run at a flow rate of 200 ml min-1 with the cone voltage at 35 V and the spraying needle voltage set at 3 kV. The scan rate was 1 s, and the mass range was from 200 to 2000. Fatty acids were released from the TAG and the MMDAG as FAMEs. Briefly, 2 mg of sample were dried under reduced pressure, then resuspended in anhydrous toluene and again dried under reduced pressure. The samples were then resuspended in 3 ml of anhydrous dichloromethane/sodium methoxide (10/1; v/v) and left overnight at room temperature. The resulting solution was mixed for 30 min with 1 g of a cation exchange resin (Amberlite CG-50, Sigma), filtered, dried under reduced pressure and resuspended in CHCl 3/H2O (1/ 1; v/v). The FAMEs were recovered in the CHCl3 extract following centrifugation, dried and analysed by GC and GC/ MS. Gas chromatography (GC) analysis was performed using a Thermoquest Trace GC 2000 equipped with a flame ionization detector. The samples were separated using a temperature programme as follows. The injector temperature was set at 50∞C, held for 1 min and then increased to 110∞C at © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
Characterization of a mycobacterial monomeromycolyl-diacylglycerol 1125 20∞C min-1. The oven was held at 110∞C, then ramped to 290∞C at 8∞C min-1 and held for 5 min to ensure that all products had eluted from the column. GC/MS analysis was performed on a Thermoquest Trace GC coupled to a Finnigan Polaris mass spectroscopy unit (energy, 70 eV) working in the positive electron impact mode, and the samples were separated as described above. All data were collected and analysed using the Xcaliber (v. 1.2) software.
Processing of bacteria for EM After growth of M. kansasii PHRI 901 in Sauton medium to mid-log phase, INH (5 mg ml-1) or THL (40 mg ml-1) was added, and the culture continued for 24 h. The bacteria were then fixed for 2 h at room temperature with 2.5% glutaraldehyde (Sigma) in 0.1 M sodium cacodylate (pH 7.2), 5 mM CaCl2 and 5 mM MgCl2, washed twice with cacodylate buffer and post-fixed overnight at room temperature with 1% OsO4 in the same buffer. Following concentration in 2% agar in the same buffer and treatment for 1 h at room temperature with 1% uranyl acetate in Veronal buffer, the bacteria were then dehydrated in a graded series of acetone and embedded in Spurr resin. Thin sections were stained with Reynold’s lead citrate. To stabilize and better visualize the lipids within the inclusions, 0.1% Malachite green was added to some of the samples during the aldehyde fixation and osmium post-fixation steps according to Pourcho et al. (1978).
Acknowledgements Tetrahydrolipstatin (THL) was kindly provided by P. Hadvary and Hoffman-La Roche, Basel, Switzerland. The authors thank Jean Paul Chauvin (Electron Microscopy Unit of the Institut de Biologie du Développement de Marseille, Marseille, France) and Chantal Brézac (Station Marine d’Endoume, UMR 6570, Université de la Méditerranée et Centre d’Océanographie de Marseille, Marseille, France) for expert technical assistance in TEM and SEM respectively. This work received financial support from the Centre National de la Recherche Scientifique (Microbiology Contract No. 2G5027 to C.C. and L.K.) and from INSERM. G.S.B. acknowledges support as a Lister Institute-Jenner Research Fellow and the Medical Research Council (UK).
References Alvarez, H.M., and Steinbüchel, A. (2002) Triacylglycerols in prokaryotic microorganisms. Appl Microbiol Biotechnol 60: 367–376. Alvarez, H.M., Mayer, F., Fabritius, D., and Steinbüchel, A. (1996) Formation of intracytoplasmic lipid inclusions by Rhodococcus opacus strain PD630. Arch Microbiol 165: 377–386. Alvarez, H.M., Kalscheuer, R., and Steinbüchel, A. (2000) Accumulation and mobilization of storage lipids by Rhodococcus opacus PD630 and Rhodococcus ruber NCIMB 40126. Appl Microbiol Biotechnol 54: 218–223. Bardou, F., Quemard, A., Dupont, M.A., Horn, C., Marchal, G., and Daffe, M. (1996) Effects of isoniazid on ultrastructure of Mycobacterium aurum and Mycobacterium tuber© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
culosis and on production of secreted proteins. Antimicrob Agents Chemother 40: 2459–2467. Baulard, A.R., Betts, J.C., Engohang-Ndong, J., Quan, S., McAdam, R.A., Brennan, P.J., et al. (2000) Activation of the pro-drug ethionamide is regulated in mycobacteria. J Biol Chem 275: 28326–28331. Besra, G.S. (1998) Preparation of cell-wall fractions from mycobacteria. In Methods in Molecular Biology, Vol. 101. Parish, T., and Stoker, N.G. (eds). Totowa, NJ: Humana Press, pp. 91–107. Besra, G.S., Sievert, T., Lee, R.E., Slayden, R.A., Brennan, P.J., and Takayama, K. (1994) Identification of the apparent carrier in mycolic acid synthesis. Proc Natl Acad Sci USA 91: 12735–12739. Borgström, B. (1988) Mode of action of tetrahydrolipstatin: a derivative of the naturally occurring lipase inhibitor lipstatin. Biochim Biophys Acta 962: 308–316. Boshoff, H.I., Mizrahi, V., and Barry, C.E., III (2002) Effects of pyrazinamide on fatty acid synthesis by whole mycobacterial cells and purified fatty acid synthase I. J Bacteriol 184: 2167–2172. Brennan, P.J., and Nikaido, H. (1995) The envelope of mycobacteria. Annu Rev Biochem 64: 29–63. Daffé, M., and Draper, P. (1998) The envelope layers of mycobacteria with reference to their pathogenicity. Adv Microbiol Physiol 39: 131–203. Daniel, J., Deb, C., Dubey, V.S., Sirakova, T.D., Abomoelak, B., Morbidoni, H.R., and Kolattukudy, P.E. (2004) Induction of a novel class of diacylglycerol acyltransferases and triacylglycerol accumulation in Mycobacterium tuberculosis as it goes into a dormancy-like state in culture. J Bacteriol 186: 5017–5030. DeBarber, A.E., Mdluli, K., Bekker, L.G., Barry, C.E., 3rd (2000) Ethionamide activation and sensitivity in multidrugresistant Mycobacterium tuberculosis. Proc Natl Acad Sci USA 97: 9677–9682. Dobson, G., Minnikin, D.E., Minnikin, S.M., Parlett, J.H., Goodfellow, M., et al. (1985) Systematic analysis of complex mycobacterial lipids. In Chemical Methods in Bacterial Systematics. Goodfellow, M., and Minnikin, D.E. (eds). London: Academic Press, pp. 237–265. Fréhel, C., Offredo, C., and de Chastellier, C. (1997) The phagosomal environment protects virulent Mycobacterium avium from killing and destruction by clarithromycin. Infect Immun 65: 2792–2802. Gale, G.R., and McLain, H.H. (1963) Effect of ethambutol on cytology on Mycobacterium smegmatis. J Bacteriol 86: 749–756. Garton, N.J., Christensen, H., Minnikin, D.E., Adegbola, R.A., and Barer, M.R. (2002) Intracellular lipophilic inclusions of mycobacteria in vitro and in sputum. Microbiology 148: 2951–2958. Hadváry, P., Sidler, W., Meister, W., Vetter, W., and Wolfer, H. (1991) The lipase inhibitor tetrahydrolipstatin binds covalently to the putative active site serine of pancreatic lipase. J Biol Chem 266: 2021–2027. Kalscheuer, R., and Steinbüchel, A. (2003) A novel bifunctional wax ester synthase/acyl-CoA: diacylglycerol acyltransferase mediates wax ester and triacylglycerol biosynthesis in Acinetobacter calcoaceticus ADP1. J Biol Chem 278: 8075–8082.
1126 L. Kremer et al. Knowles, L.M., Axelrod, F., Browne, C.D., and Smith, J.W. (2004) A fatty acid synthase blockade induces tumor cellcycle arrest by down-regulating Skp2. J Biol Chem 279: 30540–30545. Kremer, L., and Besra, G.S. (2002) Current status and future development of antitubercular chemotherapy. Expert Opin Invest Drugs 11: 1033–1049. Kremer, L., and Besra, G.S. (2004) A waxy tale, by Mycobacterium tuberculosis. In Tuberculosis and the Tubercle Bacillus. Cole, S.T., Eisenach, K.D., McMurray, D.N., and Jacobs, W.R., Jr (eds). Washington, DC: American Society for Microbiology Press, pp. 287–308. Kremer, L., Baulard, A.R., and Besra, G.S. (2000) Genetics of mycolic acid biosynthesis. In Molecular Genetics of Mycobacteria. Hatfull, G.F., and Jacobs, W.R., Jr (eds). Washington, DC: American Society for Microbiology Press, pp. 173–190. Kremer, L., Dover, L.G., Carrère, S., Nampoothiri, K.M., Lesjean, S., Brown, A., et al. (2002) Mycolic acid biosynthesis and enzymic characterization of the b-ketoacyl-ACP synthase A-condensing enzyme from Mycobacterium tuberculosis. Biochem J 364: 423–430. Kremer, L., Dover, L.G., Morbidoni, H.R., Vilchèze, C., Maughan, W.N., Baulard, A., et al. (2003) Inhibition of InhA activity, but not KasA activity, induces formation of a KasAcontaining complex in mycobacteria. J Biol Chem 278: 20547–20554. Kridel, S.J., Axelrod, F., Rozenkrantz, N., and Smith, J.W. (2004) Orlistat is a novel inhibitor of fatty acid synthase with antitumor activity. Cancer Res 64: 2070– 2075. McNeil, M., Daffé, M., and Brennan, P.J. (1991) Location of the mycolyl ester substituents in the cell walls of mycobacteria. J Biol Chem 266: 13217–13223. Minnikin, D.E. (1982) Lipids: complex lipids, their chemistry, biosynthesis and roles. In The Biology of Mycobacteria. Ratledge, C., and Stanford, J. (eds). London: Academic Press, pp. 95–184. Minnikin, D.E., Kremer, L., Dover, L.G., and Besra, G.S. (2002) The methyl-branched fortifications of Mycobacterium tuberculosis. Chem Biol 9: 545–553. Olukoshi, E.R., and Packter, N.M. (1994) Importance of stored triacylglycerols in Streptomyces: possible carbon source for antibiotics. Microbiology 140: 931–943. Packter, N.M., and Olukoshi, E.R. (1995) Ultrastructural studies of neutral lipid localisation in Streptomyces. Arch Microbiol 164: 420–427. Parrish, N.M., Kuhajda, F.P., Heine, H.S., Bishai, W.R., and Dick, J.D. (1999) Antimycobacterial activity of cerulenin and its effects on lipid biosynthesis. J Antimicrob Chemother 43: 219–226. Pourcho, R.G., Bernstein, M.H., and Gould, S.F. (1978) Mal-
achite green: applications in electron microscopy. Stain Technol 53: 29–35. Takayama, K., and Armstrong, E.L. (1976) Isolation, characterization, and function of 6-mycolyl-6¢-acetyltrehalose in the H37Ra strain of Myocobacterium tuberculosis. Biochemistry 15: 441–447. Takayama, K., and Armstrong, E.L. (1977) Metabolic role of free mycolic acids in Mycobacterium tuberculosis. J Bacteriol 130: 569–570. Takayama, K., Wang, L., and Merkal, R.S. (1973) Scanning electron microscopy of the H37Ra strain of Mycobacterium tuberculosis exposed to isoniazid. Antimicrob Agents Chemother 4: 62–65. Takayama, K., Armstrong, E.L., Kunugi, K.A., and Kilburn, J.O. (1979) Inhibition by ethambutol of mycolic acid transfer into the cell wall of Mycobacterium smegmatis. Antimicrob Agents Chemother 16: 240–242. Thearle, M., and Aronne, L.J. (2003) Obesity and pharmacologic therapy. Endocrinol Metab Clin North Am 32: 1005– 1024. Vannelli, T.A., Dykman, A., and Ortitz De Montellano, P.R. (2002) The antituberculosis drug ethionamide is activated by a flavoprotein monooxygenase. J Biol Chem 277: 12824–12829. Vielemeyer, O., McIntosh, M.T., Joiner, K.A., and Coppens, I. (2004) Neutral lipid synthesis and storage in the intraerythrocytic stages of Plasmodium falciparum. Mol Biochem Parasitol 135: 197–209. Vilchèze, C., Morbidoni, H.R., Weisbrod, T.R., Iwamoto, H., Kuo, M., Sacchettini, J.C., and Jacobs, W.R., Jr (2000) Inactivation of the inhA-encoded fatty acid synthase II (FASII) enoyl-acyl carrier protein reductase induces accumulation of the FASI end products and cell lysis of Mycobacterium smegmatis. J Bacteriol 182: 4059–4067. Waddell, S.J., Stabler, R.A., Laing, K., Kremer, L., Reynolds, R.C., and Besra, G.S. (2004) The use of microarray analysis to determine the gene expression profiles of Mycobacterium tuberculosis in response to anti-bacterial compounds. Tuberculosis 84: 263–274. Wältermann, M., Hinz, A., Robenek, H., Troyer, D., Reichelt, R., Malkus, U., et al. (2005) Mechanism of lipid-body formation in prokaryotes: how bacteria fatten up. Mol Microbiol 55: 750–763. Wang, L., Takayama, K., Goldman, D.S., and Schnoes, H.K. (1972) Synthesis of alcohol and wax ester by a cell-free system in Mycobacterium tuberculosis. Biochim Biophys Acta 260: 41–48. Watanabe, M., Aoyagi, Y., Mitome, H., Fujita, T., Naoki, H., Ridell, M., and Minnikin, D.E. (2002) Location of functional groups in mycobacterial meromycolate chains; the recognition of new structural principles in mycolic acids. Microbiology 148: 1881–1902.
© 2005 Blackwell Publishing Ltd, Molecular Microbiology, 57, 1113–1126
