Organization of the mycobacterial cell wall: a nanoscale view
- 1.6k Downloads
The biosynthesis of the Mycobacterium tuberculosis cell wall is targeted by some of the most powerful antituberculous drugs. To date, the molecular mechanisms by which these antibiotics affect the cell wall characteristics are not well understood. Here, we used atomic force microscopy – in three different modes – to probe the nanoscale surface properties of live mycobacteria and their modifications upon incubation with four antimycobacterial drugs: isoniazid, ethionamide, ethambutol, and streptomycine. Topographic imaging, combined with quantitative surface roughness analysis, demonstrated that all drugs induce a substantial increase of surface roughness to an extent that correlates with the localization of the target (i.e., synthesis of mycolic acids, arabinogalactans, or proteins). Chemical force microscopy with hydrophobic tips revealed that the structural alterations induced by isoniazid and ethambutol were correlated with a dramatic decrease of cell surface hydrophobicity, reflecting the removal of the outermost mycolic acid layer. Consistent with this finding, tapping mode imaging, combined with immunogold labeling, showed that the two drugs lead to the massive exposure of hydrophilic lipoarabinomannans at the surface. Taken together, these structural, chemical, and immunological data provide novel insight into the action mode of antimycobacterial drugs, as well as into the spatial organization of the mycobacterial cell wall.
KeywordsAntimycobacterial drugs Cell walls Chemical force microscopy Immunogold labeling Hydrophobicity Live cell imaging
In the past few years, atomic force microscopy (AFM) has opened up a range of unprecedented opportunities for imaging and manipulating microbial cells in their native environment . Of particular interest in the pharmacological context is the possibility to directly visualize the effect of drugs on cell surfaces. Early investigations performed in air demonstrated the ability of AFM to visualize drug-induced alterations in the cell walls of Escherichia coli , Helicobacter pylori , and Staphylococcus aureus . However, attempts to probe cell–drug interaction in situ, i.e., in relevant hydrated conditions, have been very limited so far .
The biosynthesis of mycobacterial cell envelope constituents is the site of action of powerful antimycobacterial agents. Ethionamide (ETH) and isoniazid (INH) inhibit the enoyl ACP reductase InhA involved in the biosynthesis of mycolic acids , while the synthesis of the glycosylated portion of the wall is efficiently inhibited by ethambutol (EMB), which targets arabinosyltransferases . Streptomycin (STR) targets the 30S subunit of ribosomes, leading to an inhibition of protein synthesis .
The aim of the present study was to gain a better understanding of how major antimycobacterial antibiotics alter the mycobacterial cell envelope and, in turn, to refine our molecular view of the mycobacterial wall architecture. To address these issues, three complementary AFM modes were used, i.e., topographic imaging, chemical force microscopy (CFM), and immunogold detection. AFM imaging combined with quantitative roughness analysis revealed that INH, ETH, EMB, and STR induce major alterations of the mycobacterial surface. The CFM technique allowed us to correlate these structural changes with variations of chemical properties. Lastly, immunogold detection enabled us to map the distribution of LAM, both on native and treated cells. Our data show that the combination of antibiotic treatments with AFM imaging/force spectroscopy constitutes a powerful tool to reveal the very complex architecture of the mycobacterial cell wall. Ultimately, these noninvasive nanoscale investigations may help us to understand how structural alterations of the envelope may lead to cell death.
Materials and methods
Mycobacterium bovis BCG (strain 1173P2, World Health Organization, Stockholm, Sweden) was grown in Sauton medium. Mycobacteria were cultured at 37°C for about 2 weeks in static conditions using 50-cm2 Roux flasks that contained 50 mL of Sauton medium supplemented with Triton WR1339 (Sigma-Aldridch, St. Louis, MO, USA). For some experiments, cells were resuspended for 24 h in Sauton medium containing INH, ETH, EMB, and STR at concentrations corresponding to minimum inhibitory concentration (MIC)/10, MIC, and 10 × MIC. MIC values were 0.02, 2, 10, and 15 μg/mL for INH, ETH, EMB, and STR, respectively. All cells were harvested by centrifugation, washed three times with milliQ water, and resuspended to a concentration of ∼108 cells per milliliter.
Atomic force microscopy
AFM images and force–distance curves were recorded at room temperature (20°C) in deionized water using a Nanoscope IV Multimode AFM (Veeco Metrology Group, Santa Barbara, CA, USA). We used oxide-sharpened microfabricated Si3N4 cantilevers with spring constants of 0.01 N/m (Microlevers, Veeco Metrology Group). Mycobacteria were immobilized onto Isopore polycarbonate membranes (Millipore, Billerica, MA, USA) [17, 20]. After filtering a concentrated cell suspension, the filter was gently rinsed with deionized water, carefully cut (1 × 1 cm), and attached to a steel sample puck (Veeco Metrology Group) using a small piece of adhesive tape, and the mounted sample was transferred into the AFM liquid cell.
For CFM, cantilevers were coated by electron beam thermal evaporation with a 5-nm-thick Cr layer followed by a 30-nm-thick Au layer. Gold-coated cantilevers were immersed for 14 h in 1-mM solutions of HS(CH2)11CH3 in ethanol and then rinsed with ethanol. Adhesion maps and histograms were obtained by recording 16 × 16 or 32 × 32 force–distance curves on areas of given size and calculating the adhesion force for each force curve.
The procedure for immunogold detection was as follows: Cells immobilized on membranes were preincubated for 1 h and 30 min with 10% BSA in phosphate-buffered saline (PBS) to minimize nonspecific adsorption. Samples were then incubated for 2.5 h with monoclonal mouse anti-LAM antibodies (04.CS.40.1.21.LAM.mm, Colorado State University, USA) diluted in PBS (1:20). Samples were washed three times with PBS and then further incubated for 2.5 h with the corresponding goat antimouse second antibody conjugated to 10-nm gold particles (Goat-anti-Mouse IgG + IgM, 810.044, Aurion, Wageningen, The Netherlands). After washing three times with PBS, samples were mounted onto steel sample pucks as describe above.
Results and discussion
Antimycobacterial drugs induce major ultrastructural changes
For the three drugs, cells became significantly rougher upon treatment with MIC and 10 × MIC concentrations. No major differences were observed between the two concentrations, suggesting that increasing the concentration at values larger than the MIC does not change the drug effects. In summary, the above quantitative surface roughness analysis demonstrates that all drugs induced major cell-surface alterations to an extent that correlates with the localization of the target.
Structural changes correlate with differences in chemical properties
These nanoscale measurements are consistent with earlier macroscopic water contact angle data , which showed that the presence of mycolic acids of bacteria from Corynebacterium and Mycobacterium genera is related to cell surface hydrophobicity as well as to cell adhesion to defined surfaces. Mycobacterial species were shown to exhibit water contact angles of 85 to 98°, reflecting remarkably strong hydrophobic properties, while bacteria lacking mycolic acids were much less hydrophobic. Although the water contact angle approach is very useful for assessing the hydrophobic qualities of microorganisms, it provides averaged information obtained on large ensembles of cells. Accordingly, AFM is a powerful complementary approach to such traditional methods, providing for the first time spatially resolved and quantitative measurements of hydrophobicity on single live cells.
Notably, Fig. 6d–i shows that treatment with INH and EMB markedly decreased the hydrophobic character of the cells, adhesion forces of only 137 ± 31 and 167 ± 56 pN being recorded in these conditions. We believe the measured strong hydrophilic properties reflect the exposure of inner-cell-wall carbohydrates, such as arabinose, galactose, and mannose, resulting from the removal of the outermost mycolic acid layer. This observation is consistent with the action modes of the drugs, i.e., inhibition of mycolic acid (INH) and arabinan (EMB) synthesis, as well as with the structural changes observed by topography imaging.
Antimycobacterial drugs lead to the exposure of LAM
The feasibility of using immunogold labels as cell-surface markers in AFM studies has been demonstrated earlier. In a pioneering study , AFM was used to image the surface of immunogold-labeled human lymphocytes. AFM images revealed colloidal gold particles on the cell surface with and without silver enhancement. Individual immunogold particles were clearly resolved from the cell surface, thus determining the location of antigens. More recently, a similar AFM-based immunogold technique allowed the revealing of types I and II collagen fibers on rat fibroblasts and human chondrosarcoma cells . Because the above cellular studies were performed in the dried state, it is unclear whether the information provided is more relevant than that obtained using conventional immunogold electron microscopy. To our knowledge, the present study represents the first attempt to apply immunogold AFM to hydrated bacterial cells.
Summary and biological implications
AFM – used in the topography, CFM, and immunogold modes – is a powerful multifunctional tool for exploring the interactions between live mycobacteria and antimycobacterial drugs. Quantitative topographic imaging shows that INH, ETH, EMB, and STR induce a major increase of cell surface roughness. Using CFM, these structural alterations are found to correlate with a dramatic decrease of cell surface hydrophobicity, an effect that we attribute, at least for INH and ETH treatments, to the removal of the mycolic acid layer. Consistent with these findings, immunogold detection shows that treatment with these two drugs leads to the massive exposure of hydrophilic LAM at the surface.
Taken together, the above structural, chemical, and immunological data provide novel insight into the 3-D organization of the mycobacterial cell wall. The uniform distribution of cell surface hydrophobicity measured on native cells and the absence of any substantial labeling with anti-LAM antibodies strongly suggest that LAM is not abundant at the cell surface, the latter being essentially composed of highly hydrophobic molecules such as mycolic acids or wall-associated lipids. This finding makes it very unlikely that LAM is anchored in the mycolic acid leaflet via its lipid tail and exposed at the surface of the bacteria. This may seem surprising because ELISA tests have shown the presence of LAM at the cell surface . However, this apparent discrepancy may easily be explained as follows: (1) unlike AFM, ELISA assays are not specific to the outermost cell surface but probe the entire cell wall, and (2) as opposed to here, ELISA measurements may have been performed in conditions favoring a nonspecific anchoring of LAM to the cell wall. The question as to how these glycolipids, as well as other components of the external portion of the envelope, cross the outer lipid barrier remains to be addressed.
To our knowledge, this is the first time that modifications of the physicochemical properties of the cell wall due to antibiotic treatments were observed on hydrated mycobacteria. The data confirm the idea that the dual property of the cell wall composed of an outer homogenously hydrophobic leaflet of mycolic acid and an inner hydrophilic arabinogalactan matrix should eventually limit the passage of both hydrophobic and hydrophilic antibiotics. Consequently, the development of new antimycobacterial compounds should take into account their future association with drugs that modify the ultrastructure of the cell wall and that may influence their capacity to cross the remaining envelope. On the other hand, this work also suggests that the efficiency of a given therapy will progressively change with the erosion of the envelope, and not necessarily in a positive manner.
This work was supported by the National Foundation for Scientific Research (FNRS); the Université catholique de Louvain (Fonds Spéciaux de Recherche), the Région wallonne; the Federal Office for Scientific, Technical and Cultural Affairs (Interuniversity Poles of Attraction Programme); Institut National de la Santé et de la Recherche Médicale; and the Research Department of the Communauté française de Belgique (Concerted Research Action). Y.F.D. and D.A. are Research Associate and Research Fellow of the FNRS, respectively. Anti-LAM antibodies were kindly provided by the Colorado State University as part of NIH, National Institute of Allergy and Infectious Diseases contract no. HHSN266200400091C, entitled “Tuberculosis Vaccine Testing and Research Materials.”
- 3.Baulard AR, Besra GS, Brennan PJ (1999) The cell wall core of Mycobacterium: structure, biogenesis, and genetics. In: Ratledge C, Dale J (eds) Mycobacteria: molecular biology and virulence. Blackwell, Oxford, pp 240–259Google Scholar
- 4.Belanger AE, Besra GS, Ford ME, Mikusova K, Belisle JT, Brennan PJ, Inamine JM (1996) The embAB genes of Mycobacterium avium encode an arabinosyl transferase involved in cell wall arabinan biosynthesis that is the target for the antimycobacterial drug ethambutol. Proc Natl Acad Sci USA 93:11919–11924PubMedCrossRefGoogle Scholar
- 6.Boyle-Vavra S, Hahm J, Sibener SJ, Daum RS (2000) Structural and topological differences between a glycopeptide-intermediate clinical strain and glycopeptide-susceptible strains of Staphylococcus aureus revealed by atomic force microscopy. Antimicrob Agents Chemother 44:3456–3460PubMedCrossRefGoogle Scholar
- 16.Draper P (1998) The outer parts of the mycobacterial envelope as permeability barriers. Front Biosci 3:1253–1261Google Scholar
- 21.Khoo KH, Douglas E, Azadi P, Inamine JM, Besra GS, Mikusova K, Brennan PJ, Chatterjee D (1996) Truncated structural variants of lipoarabinomannan in ethambutol drug-resistant strains of Mycobacterium smegmatis-inhibition of arabinan biosynthesis by ethambutol. J Biol Chem 271:28682–28690PubMedCrossRefGoogle Scholar
- 23.OrtaloMagne A, Lemassu A, Laneelle MA, Bardou F, Silve G, Gounon P, Marchal G, Daffe M (1996) Identification of the surface-exposed lipids on the cell envelopes of Mycobacterium tuberculosis and other mycobacterial species. J Bacteriol 178:456–461Google Scholar