Accurate Molecular Mass Determination of Mycolic Acids by MALDI

Synthetic trehalose esters of cis-alkene and diene α′-mycolic acids of Mycobacteria. Salam G. Taher , Maged Muzael , Juma'a R. Al Dulayymi , Mark S...
0 downloads 0 Views 175KB Size
Anal. Chem. 2001, 73, 4537-4544

Accurate Molecular Mass Determination of Mycolic Acids by MALDI-TOF Mass Spectrometry Franc¸ oise Laval, Marie-Antoinette Lane´elle, Catherine De´on, Bernard Monsarrat, and Mamadou Daffe´*

Institut de Pharmacologie et Biologie Structurale du Centre National de la Recherche Scientifique et Universite´ Paul Sabatier (UMR 5089), 205 route de Narbonne, 31077 Toulouse, Cedex 04 (France)

Mycolic acids, major and specific long-chain fatty (C70C90) acid components of the mycobacterial cell envelope, were analyzed for the first time using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry operating in a reflectron mode. The various types of purified mycolates from representative mycobacterial species were analyzed using 2,5-DHB as matrix, because less than 10 pmol of mycolates was sufficient to obtain well-resolved mass spectra composed exclusively of pseudomolecular [M + Na]+ ions consistent with the structures deduced from the chemical analytical techniques applied to these molecules. Examination of the MALDI mass spectra demonstrated that the chain lengths of the various mycolates correlated with the growth rate of mycobacterial strains. Although slow growers, such as Mycobacterium tuberculosis and Mycobacterium ulcerans, produced a series of odd carbon numbers (C74C82) of r-mycolic acids, rapid growers synthesized both odd and even carbon numbers. In addition, the main chain of oxygenated mycolic acids from slow growers were four to six carbon atoms longer than the corresponding r-mycolic acids, whereas rapid growers elaborated oxygenated homologues possessing the same chain lengths as their r-mycolic acids. Furthermore, a comparative analysis of the crude fatty acid mixtures from a wild-type strain of M. tuberculosis and its isogenic mutant effected in the synthesis of oxygenated mycolates by MALDI mass spectrometry revealed structural differences between the r-mycolates from the two strains. Thus, this technique appeared to be a rapid and highly sensitive technique for the analysis of mycolic acids, not only by providing accurate molecular masses and new structural information, but also by both reducing sample consumption and saving time. Mycolic acids, very long chain (up to C90) R-branched β-hydroxylated fatty acids, are the hallmark of the Mycobacterium genus that comprises several human pathogens such as Mycobacterium tuberculosis and Mycobacterium leprae, the causative agents of tuberculosis and leprosy, respectively. These molecules represent major cell envelope components (40-60% of the cell * Corresponding author: Phone: (+33) 561 175 569. Fax: (+33) 561 175 994. E-mail: [email protected]. 10.1021/ac0105181 CCC: $20.00 Published on Web 08/16/2001

© 2001 American Chemical Society

dry weight) and are found covalently linked to the cell wall arabinogalactan or esterifying trehalose and glycerol; both types of mycolic acid-containing components are believed to play a crucial role in the structure and function of the mycobacterial cell envelope.1,2 Mycolic acids attached to the cell wall arabinogalactan are organized with other lipids to form an outer permeability barrier that confers an extremely low fluidity and an exceptional low permeability to mycobacteria and may also explain their intrinsic resistance to many antibiotics.3 Trehalose mycolates have been implicated in numerous biological functions related both to the physiology and virulence of M. tuberculosis.2 The occurrence of various chemical groups on the main chain of these acids, called meromycolic chain, allows the definition of mycolate subclasses; thus, the R-mycolic acid type, which is present in all the mycobacterial species described so far, is composed of C70-C76 fatty acids and contains two unsaturations (either cis cyclopropane rings or one or both of cis and trans double bonds) and no additional oxygenated groups. The slightly more polar and of lower molecular mass (C60-C68) mycolic acids, designed as R′, are restricted to some mycobacterial species, devoid of additional oxygenated groups, and contain only a cis double bond. In the other types of mycolic acids, the meromycolic chain carries an oxygenated group in the distal position (closest to the ω end of the meromycolic chain) consisting of a keto, methoxy, epoxy, or ester group. Representative structures of the seven known subclasses of mycolic acids are shown in Figure 1. Most mycobacterial species elaborate a combination of different types of mycolic acids (Figure 1), and the mycolate composition of mycobacteria has been extensively used for taxonomic purposes.4,5 Interestingly, subtle variations in the mycolic acid structure have profound effects on the physiology and virulence of the tubercle bacillus.6-9 For instance, the replacement of a (1) Barry, C. E., III; Lee, R. E.; Mdluli, K.; Sampson, A. E.; Schroeder, B. G.; Slayden, R. A.; Yuan, Y. Prog. Lipid Res. 1998, 37, 143-179. (2) Daffe´, M.; Draper, P. Adv. Microb. Physiol. 1998, 39, 131-203. (3) Jarlier, V.; Nikaido, H. J. Bacteriol. 1990, 172, 1418-1423. (4) Daffe´, M.; Lane´elle, M. A.; Asselineau, C.; Le´vy-Frebault, V.; David, H. Ann. Microbiol. 1983, 134B, 241-256. (5) Minnikin, D. E.; Minnikin, S. M.; Parlett, J. H.; Goodfellow, M.; Magnusson, M. Arch. Microbiol. 1984, 139, 225-231. (6) Liu, J.; Barry, C. E., III; Besra, G. S.; Nikaido, H. J. Biol. Chem. 1996, 271, 29545-29551. (7) Yuan, Y.; Zhu, Y.; Crane, D. D.; Barry, C. E., III. Mol. Microbiol. 1998, 29, 1449-1458. (8) Glickman, M. S.; Cox, J. S.; Jacobs, W. R., Jr. Mol. Cell 2000, 5, 717-727. (9) Dubnau, E.; Chan, J.; Raynaud, C.; Mohan, V. P.; Lane´elle, M.-A.; Yu, K.; Que´mard, A.; Smith, I.; Daffe´, M. Mol. Microbiol. 2000, 36, 630-637.

Analytical Chemistry, Vol. 73, No. 18, September 15, 2001 4537

Figure 1. Structures of the major mycolic acids described in mycobacteria.

The main values of n1 and n3 are 15, 17, and 19; those of n2 depend on the nature of A and B. When no methyl branch is present in A and B, or when a methyl branch occurs in both A and B, n2 is 12, 14, and 16. In contrast, when a methyl branch is present in A or B, n2 is 13, 15, or 17. n4 is always 19, 21, or 23.

cyclopropane ring by a double bond in the least polar mycolic acid (R-mycolate) esterifying trehalose totally abolishes the formation of cords typifying virulent tubercle bacilli and profoundly affects the virulence of the mutant strain.8 Similarly, the lack of production of oxygenated mycolates (keto- and methoxymycolates) in M. tuberculosis, results in both a change of the permeability and an attenuation of the mutant strain.9 These observations, in addition to the fact that the biosynthesis pathway leading to mycolic acids is the target of the most effective antituberculous drug, isoniazid, explain the renewed interest in studies related to these compounds. In the past decades, electron impact (EI) mass spectrometry has been exhaustively used for the structural elucidation of mycolic acids.10-14 Although this technique has provided important structural information through the analysis of fragments that resulted both from pyrolysis of the R-branched β-hydroxylated fatty acids and cleavages of the main chain at positions that correspond to the location of chemical functions, the fragmentation patterns are complicated because of the existence of series of homologues and the loss of water and methanol molecules from

the native fragments and compounds. Furthermore, thanks to the development of molecular biology of mycobacteria and the availability of the M. tuberculosis genome sequence,15 many attempts have been made to rapidly analyze the consequences of inactivation of genes putatively involved in the biosynthesis of mycolic acids. Consequently, more sensitive analytical methods are needed to comparatively evaluate the mycolate content of the wild-type and isogenic mutants at a small-scale culture level. Because of its high sensitivity, matrix-assisted laser desorption/ ionization time-of-flight (MALDI-TOF) mass spectrometry represents an attractive approach for a rapid structural analysis of mycolic acids. In the present study, we first applied MALDI-TOF mass spectrometry in a reflectron mode to the study of the various types of mycolic acids and showed that this technique provides highly resolved mass spectra. We then compared the crude fatty acid mixtures of a M. tuberculosis mutant defective in the biosynthesis of oxygenated mycolic acids to that of its parent strain. Analysis of the mass spectra from the mutant strain indicated the accumulation of putative intermediates of the biosynthesis of oxygenated mycolic acids of M. tuberculosis.

(10) Toubiana, R.; Berlan, J.; Sato, H.; Strain, M. J. Bacteriol. 1979, 139, 205211. (11) Minnikin, D. E. In The Biology of the Mycobacteria; Ratledge, C., Stanford, J., Eds.; Academic Press Inc.: London, 1982; Vol. 1, pp 95-184. (12) Dubnau, E.; Lane´elle, M. A.; Soares, S.; Benichou, A.; Vaz, T.; Prome´, D.; Prome´, J. C.; Daffe´, M.; Que´mard, A. Mol. Microbiol. 1997, 23, 313-322. (13) Davidson, L. A.; Draper, P.; Minnikin, D. E. J. Gen. Microbiol. 1982, 128, 823-828. (14) Que´mard, A.; Lane´elle, M. A.; Marrakchi, H.; Prome´, D.; Dubnau, E.; Daffe´, M. Eur. J. Biochem. 1997, 250, 758-763.

EXPERIMENTAL SECTION Bacterial Strains. Strains used in this study as sources of the different types of mycolates were M. tuberculosis H37Rv (ATCC

4538

Analytical Chemistry, Vol. 73, No. 18, September 15, 2001

(15) Cole, S. T.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher, C.; Harris, D.; Gordon, S. V.; Eiglmeier, K.; Gas, S.; Barry, C. E.; Tekaia, F.; Badcock, K.; Basham, D.; Brown, D.; Chillingworth, T.; Connor, R.; Davies, R.; Devlin, K.; Feltwell, T.; Gentles, S.; Hamlin, N.; Holroyd, S.; Hornsby, T.; Jagels, K.; Barrell, B. G. Nature 1998, 393, 537-544.

27294), Mycobacterium ulcerans ATCC 19423, Mycobacterium smegmatis ATCC 607, Mycobacterium phlei ATCC 11758, and Mycobacterium alvei CIPT 103464. The isogenic mutant strain of M. tuberculosis H37Rv referred as hma::hyg strain, was obtained by disruption in the hma gene, as previously described.9 Growth Conditions. Bacteria were grown on Sauton medium as surface pellicles at 37 °C for various periods of time according to their growth rate. M. smegmatis, M. alvei, and M. phlei were grown for 3, 5, and 7 days, respectively, whereas M. tuberculosis strains were grown for 8 weeks. Cells were collected by pouring off the medium. Isolation and Purification of Mycolates. Although several methods have been described to release mycolic acids from mycobacterial cells, it is important to preserve the native structural features of the molecules, because mycolic acids may contain acidlabile functions such as epoxy ring. This explains why alkaline conditions are preconized for their isolation.16 In addition, the method should avoid a lyophilization step that represents a serious hazard when analyzing live pathogens; alternatively, dead lyophilized bacteria can be used. Consequently, we have developed a simple method suitable for systematic analyses of both pathogenic and nonpathogenic mycobacteria that is based on saponification of wet bacteria.4 Briefly, under a biosafety cabinet type II, a loopful of mycobacterial wet cells, consisting of roughly 5-10 mg, was transferred in a screw-capped tube containing 1.4 mL of methoxyethanol. A volume of 0.2 mL of 40% aqueous KOH was added, and the tightly closed tube was heated in a oven at 100 °C for 3 h. After cooling and acidification by 0.5 mL of 20% aqueous H2SO4, fatty acids were extracted three times with diethyl ether. The pooled ethereal phases were throroughly washed with distilled water three times to eliminate traces of H2SO4 and methoxyethanol and were then dried under vacuum. Methylation of extracted fatty acids was performed under a well-ventilated hood with an ethereal solution of diazomethane.17 The mycolate methyl ester pattern of the strains was determined by analytical TLC on Silica Gel 60 using either petroleum ether/diethyl ether (9/1,v/v, five runs) or dichloromethane as eluents.4 Revelation of lipid spots was performed by spraying the plates with molybdophosphoric acid (10% in ethanol), followed by charring. Purification of mycolates was achieved by preparative TLC (Silica Gel 60, Macherey-Nagel) run in the same eluents; mycolate bands were first visualized by spraying the plates with rhodamine B (0.01% in 0.25M NaH2PO4), scraped off the plates and then eluted from the gel with diethyl ether. Then rhodamine B was separated from the mycolates by chromatography on a Florisil column (Pasteur pipet) irrigated with a diethyl ether elution. The purity of the various types of mycolates was checked by analytical TLC, as described above. Mycolates and other hydroxylated lipid compounds were converted into trimethylsilyl derivatives, as described.18 An authentic synthetic corynomycolate, kindly provided by Dr J. F. Tocanne (IPBS, Toulouse), was used as a reference compound. MALDI-TOF Mass Spectrometry. Matrix Solutions. Because both the nature of the matrix and the method of sample (16) Minnikin, D. E.; Parlett, J. H.; Magnusson, M.; Ridell, M.; Lind, A. J. Gen. Microbiol. 1984, 130, 2733-2736. (17) Kates, M. In Laboratory Techniques in Biochemistry and Molecular Biology; Burdon, R. H., van Knippenberg, P. H., Eds.; Elsevier: Amsterdam, 1986; Vol. 3, pp 326-425. (18) Sweeley C. C.; Bentley, R.; Makita M.; Wells W. W. J. Am. Chem. Soc. 1963, 85, 2497-2507.

preparation are critical for obtaining good signals, we first addressed the question of the best matrix to be used for mycolic acids analysis. The choice needed to take into account both the crystal quality and the obtaining of maximum sensitivity.19,20 All of the solid matrixes used in these studies were purchased from Aldrich (France). Matrix solutions were prepared as follows: •R-Cyano-4-hydroxycinnamic acid (R-CHCA): 10 mg/mL in acetonitrile/water (1:1,v/v) containing 0.1% trifluoroacetic acid (TFA); •Sinapinic acid: 10 mg/mL in acetonitrile/water (1/1,v/v) containing 0.1% TFA; •2,5-Dihydroxybenzoic acid (2,5-DHB): 10 mg/mL in methanol/ chloroform (1:1,v/v); •5-Chloro-2 mercaptobenzothiazole (CMBT): 10 mg/mL in methanol/chloroform (1:1,v/v); •2-(4-Hydroxyphenylazo) benzoic acid (HABA): 10 mg/mL in methanol/chloroform (1:1,v/v); •1-Hydroxyisoquinoline (HIQ)/2,5-DHB in the molar ratio of 1:3 in methanol/chloroform (1:1,v/v);21 and •2-Hydroxy-5-methoxybenzoic acid /2,5-DHB: a mixture (1:9, v/v) of the stock solutions at 10 mg/mL in methanol/chloroform (1:1,v/v), referred to as ,super 2,5-DHB..21 From the seven different matrix solutions tested, 2,5-DHB, gave the best results in terms of signal-to-noise ratio and homogeneity of detection. Moreover, this matrix is readily soluble in the organic solvents tested in which mycolic acids are soluble, providing a homogeneous target surface and giving both reproducible results and highly resolved mass spectra. Sample Preparation. The stock solutions of mycolates were prepared in chloroform at a concentration of 1 mM and were directly applied onto the sample plate as 1-µL droplets, followed by the addition of 0.5 µL of matrix solution. The samples were then allowed to crystallize at room temperature. Premixed sample/ matrix mixtures deposited onto the target did not improve crystallization properties. To investigate the sensitivity, aliquots from the stock solutions were diluted with the required volume of chloroform to obtain final concentrations between 100 and 10 µM, and the various dilutions were applied onto the sample plate as above. Mass Spectrometry Analysis. MALDI-TOF mass spectra (in the positive mode) were acquired on a Voyager-DE STR mass spectrometer (PerSeptive Biosystems, Framingham, MA) equipped with a pulsed nitrogen laser emitting at 337 nm. Samples were analyzed in the Reflectron mode using an extraction delay time set at 100 ns and an accelerating voltage operating in positive ion mode of 20 kV. To improve the signal-to-noise ratio, 150 single shots were averaged for each mass spectrum, and typically, four individual spectra were accumulated to generate a summed spectrum. An external mass spectrum calibration was performed using the calibration mixture 1 of Sequazyme Peptide Mass Standards Kit (Perseptive Biosystems), including known peptide standards in a mass range from 900 to 1600 Da. Internal mass calibration was performed using either sodiated synthetic corynomycolic acid ([C32H64O3+Na]+; MW, 519.4748) and its methyl (19) Schiller, J.; Arnhold, J.; Benard, S.; Muller, M.; Reichl, S.; Arnold, K. Anal. Biochem. 1999, 267, 46-56. (20) Benard, S.; Arnhold, J.; Lehnert, M.; Schiller, J.; Arnold, K. Chem. Phys. Lipids 1999, 100, 115-125. (21) Harvey, D. J. Mass Spectrom. Rev. 1999, 18, 349-450.

Analytical Chemistry, Vol. 73, No. 18, September 15, 2001

4539

Figure 2. TLC of purified methyl mycolates under study: I, R-mycolates; II, R′-mycolates; III, methoxymycolates; IV, ketomycolates; V, epoxymycolates; VI, ω-carboxymycolates; X, ω-1-methoxymycolates; lane t, mycolates from M. tuberculosis (types I, III, IV); lane p, mycolates from M. phlei (types I, IV, VI); lane a, mycolates from M. alvei (types I, X); adsorbant, silicic acid; solvent, petroleum ether/diethyl ether (90/10,v/v), five runs; visualization, molybdophosphoric acid spray, followed by heating. The arrow indicates the solvent front.

ester derivative ([C33H66O3+Na]+; MW, 533.4904) or with the sodiated C76 mycolic acid methyl ester previously described ([M + Na]+ ) 1146.1477).10 RESULTS AND DISCUSSION Structural Characteristics of Mycolic Acids. The mycolate profiles of the various species have been a valuable tool in the classification of mycobacteria,4,5,22 the different subclasses of mycolic acids being readily separated by TLC, thanks to the presence of various chemical groups on the meromycolic chain. This observation led us to isolate the various methyl mycolates under study by this technique as pure substances (Figure 2). The characteristic mycolate pattern of M. tuberculosis (lane t), which consists of R-, keto-, and methoxymycolates, is shared by several slow-growing pathogenic species, including the pathogenic species Mycobacterium bovis, M. leprae, Mycobacterium microti, M. ulcerans, or Mycobacterium kansasii, and the nonpathogenic species Mycobacterium gastri and Mycobacterium gordonae. M. smegmatis contains R-, R′-, and epoxymycolic acids (lanes I, II, and V, respectively), found mainly in some rapidly growing non photochromogenic species.4,5 M. phlei (lane p), a rapidly growing photochromogenic species, as well as members of the opportunistic pathogens Mycobacterium avium-intracellulare com(22) Levy-Frebault, V. V.; Portaels, F. Int. J. Syst. Bacteriol. 1992, 42, 315-323.

4540 Analytical Chemistry, Vol. 73, No. 18, September 15, 2001

Figure 3. Reflectron MALDI-TOF mass spectra of the synthetic C32H64O3 corynomycolic acid: a, free acid; b, the methyl ester derivative with, in insert, an expanded view of the pseudomolecular [M + Na]+ ion. Samples were dissolved in chloroform at a final concentration of 1 mM and applied on the sample plate as 1-µL droplets. 2,5-DHB was used as matrix. The accelerating voltage was 20 kV (positive mode); peaks arising from matrix (DHB) adducts are labeled with an asterisk.

plex, contains R-, keto-, and dicarboxylicmycolic acids. This latter dicarboxylic acid derives from the cleavage of the ester bond that links this molecule to long-chain alcohol (C18-C22) in the original wax ester.4,23 At last, M. alvei, a new species isolated from the environment,24 has an original pattern consisting of R- and ω-1 methoxylated mycolate (lane a). Purified methyl mycolates of interest (lanes I-X) were used for MALDI-TOF mass spectrometry analysis. MALDI Mass Spectrometry Analyses. Mass Measurements of Mycolic Acids. Mycolic acids, corynomycolic acids, and other fatty acids occur generally in nature as series of structurally related molecules differing from one another by two methylene units (28 atomic mass units), generating complex mass spectra. Thus, the first approach in applying MALDI-TOF mass spectrometry to the analysis of mycolic acids was the use of a well-defined synthetic molecule. As shown in Figure 3, well-resolved mass spectra were obtained in the positive mode using this C32H64O3 corynomycolic acid and its methyl ester derivative. The mass spectrum of the free acid (Figure 3a) exhibited a major peak at m/z 519, and conversion of the acid group into methyl ester induced a 14-atomicmass-units (amu) shift of this peak (Figure 3b). Furthermore, trimethylsilylation of the corynomycolate methyl ester led to a shift of 72 mass units of the molecular mass (data not shown). On the basis of the calculated molecular mass of this C32 molecule, which corresponds to 496, the above observations (23) Lane´elle, M. A.; Lane´elle, G. Eur. J. Biochem. 1970, 12, 296-300. (24) Luquin, M.; Roussel, J.; Lopez-Calahorra, F.; Lane´elle, G.; Ausina, V.; Lane´elle, M. A. Eur. J. Biochem. 1990, 192, 753-759.

Figure 4. Accurate mass measurement of R-mycolic acid methyl esters by MALDI-TOFMS: a, R-mycolic acid methyl esters of M. ulcerans; b, R-mycolic acid methyl esters of M. alvei. Values indicate the masses of the corresponding sodium adducts (M + 23). Internal mass calibration was performed using C76 mycolic acid methyl ester ([M + Na]+ ) 1146.1477). The measured masses were within 1-10 ppm of the theoretical values. f, peak arising from a molecule with an unknown structure constantly associated with samples fractionated on a Florisil column (m/z 1199.75). Experimental conditions as indicated in Figure 3 were used.

suggested that the native and derivatized compounds bore a Na+ adduct, which is consistent with published data on the 2,5-DHB matrix known to produce [M + Na]+ species as the major ion in a positive ion mode analysis.19 To firmly establish this point, the samples were saturated with potassium iodide salts for 1 h and reanalyzed by mass spectrometry. As expected, the resulting MALDI spectra showed a 16-mass-units shift, as compared with the masses observed in Figure 3a,b, indicating that the peaks really corresponded to [M + K]+ peaks (data not shown). To obtain accurate mass measurements, the corynomycolic acid and its methyl ester derivative were used as references and internal standards. The insert in Figure 3b shows an expanded view of the spectrum with the value of the resolution of the peak at m/z ) 533.4904. Mass Spectrometry of the Different Types of Mycolic Acids. The various purified mycolic acid methyl esters were analyzed by MALDI and gave well-resolved mass spectra composed exclusively of pseudomolecular [M + Na]+ ions, exemplified in Figures 4 and 5. In the reflectron mode, a good signal-tonoise ratio was obtained using 10 pmol of mycolates. For the two examples presented in Figure 4 (R-mycolates of M. ulcerans and M. alvei), C76 mycolic acid methyl ester was used as an internal standard, and mass measurements could be performed with a

Figure 5. MALDI-TOF mass spectra of oxygenated mycolic acid methyl. a, ω-1 methoxymycolic acid methyl esters of M. alvei; b, ketomycolic acid methyl esters of M. ulcerans; c, methoxymycolic acid methyl esters of M. ulcerans. Experimental conditions as indicated in Figure 3 were used. f ) peak arising from a molecule with an unknown structure constantly associated with samples fractionated on a Florisil column.

good accuracy. The deviation from the theoretical mass values of [M + Na]+ ions was found to be