|
|
J Bacteriol. 2009 January; 191(1): 152–160. Published online 2008 October 17. doi: 10.1128/JB.01105-08. | PMCID: PMC2612442 |
Copyright © 2009, American Society for Microbiology Macroamphiphilic Components of Thermophilic Actinomycetes: Identification of Lipoteichoic Acid in Thermobifida fusca![[down-pointing small open triangle]](/corehtml/pmc/pmcents/x25BF.gif) † Obaidur Rahman,1 Markus Pfitzenmaier,2 Oxana Pester,1‡ Siegfried Morath,3 Stephen P. Cummings,1 Thomas Hartung,4 and Iain C. Sutcliffe1* School of Applied Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, United Kingdom,1 Department of Organic Chemistry, University of Marburg, 35043 Marburg, Germany,2 EU Joint Research Centre, Institute for Health and Consumer Protection In-Vitro Toxicology Unit/European Centre for the Validation of Alternative Methods (ECVAM), T.P. 580, Via E. Fermi 2749, I-21027 Ispra,3 EU Joint Research Centre, Institute for the Protection and the Security of the Citizen (IPSC), Traceability, Risk and Vulnerability Assessment Unit (TRiVA), 21027 Ispra, Italy4 Received August 7, 2008; Accepted October 13, 2008. The cell envelopes of gram-positive bacteria contain structurally diverse membrane-anchored macroamphiphiles (lipoteichoic acids and lipoglycans) whose functions are poorly understood. Since regulation of membrane composition is an important feature of adaptation to life at higher temperatures, we have examined the nature of the macroamphiphiles present in the thermophilic actinomycetes Thermobifida fusca and Rubrobacter xylanophilus. Following hot-phenol-water extraction and purification by hydrophobic interaction chromatography, Western blotting with a monoclonal antibody against lipoteichoic acid strongly suggested the presence of a polyglycerophosphate lipoteichoic acid in T. fusca. This structure was confirmed by chemical and nuclear magnetic resonance analyses, which confirmed that the lipoteichoic acid is substituted with β-glucosyl residues, in common with the teichoic acid of this organism. In contrast, several extraction methods failed to recover significant macroamphiphilic carbohydrate- or phosphate-containing material from R. xylanophilus, suggesting that this actinomycete most likely lacks a membrane-anchored macroamphiphile. The finding of a polyglycerophosphate lipoteichoic acid in T. fusca suggests that lipoteichoic acids may be more widely present in the cell envelopes of actinomycetes than was previously assumed. However, the apparent absence of macroamphiphiles in the cell envelope of R. xylanophilus is highly unusual and suggests that macroamphiphiles may not always be essential for cell envelope homeostasis in gram-positive bacteria. The bacterial cell envelope has a structural role in maintaining cellular integrity and is also critical as the site of interaction and traffic between the cell and its environment. Thus, it is important that we understand fully the composition and organization of the cell envelope and how they differ across bacterial taxa. In gram-positive bacteria, the cell envelope consists of the plasma membrane, the cell wall matrix (peptidoglycan and associated glycopolymers such as teichoic acids), and other components/layers (such as S-layers and capsules) ( 2, 41, 57). The membrane-wall interface is of particular interest, because this region may be considered directly analogous to the periplasm of gram-negative bacteria, and experimental evidence for the existence of a discrete periplasmic space is accumulating ( 34, 34a, 36, 41, 61). Thus, it can be envisaged that macromolecules tethered to the outer leaflet of the plasma membrane will project into and occupy this space ( 2, 34a, 41). Some components may also project further into the wall matrix, and thus the membrane-wall interface in gram-positive bacteria should be viewed as a continuum ( 41). This “gram-positive periplasm” likely includes the extracytoplasmic domains of integral membrane proteins, bacterial lipoproteins, secreted proteins, and macroamphiphilic glycopolymers, i.e., structurally diverse polymers with covalently linked lipid anchors ( 34a). The latter can typically be divided into two major classes, the lipoteichoic acids (LTA) and lipoglycans ( 11, 49, 53), although they can also be further categorized based on their electrostatic properties ( 57). Each of these classes can be further divided into various subtypes. LTA are defined as polymers with repeating units containing alditol phosphates, of which the polyglycerophosphate LTA (PGP-LTA) are the prototypical class ( 11, 32, 38, 41, 49, 57). Lipoglycans are structurally more diverse, but several structural archetypes, such as lipoglucogalactans, lipomannans, and lipoarabinomannans, have been recognized ( 18, 49, 51, 52). It has been suggested that LTA are representative of the gram-positive bacterial class Firmicutes, whereas the cell envelopes of members of the class Actinobacteria typically contain lipoglycans ( 11, 49). However, it is now clear that there are significant exceptions to this pattern of distribution, with lipoglycans present in Mollicutes ( 47) and LTA reported in representatives of two genera of Actinobacteria ( 19, 42). Nevertheless, the apparent ubiquity and seemingly mutually exclusive distribution of these two classes of macroamphiphile encourage the speculation that they are both functionally comparable and physiologically significant. Indeed, LTA has recently been shown to be essential for growth and cell division in Staphylococcus aureus ( 21). The cell envelopes of thermophilic bacteria play an important role in the adaptation of these bacteria to growth at high temperatures. Specifically, the maintenance of membrane structure and function is an important aspect of adaptation to thermophily ( 6, 29, 45). However, to our knowledge there have been no studies to date of the macroamphiphiles present in the cell envelopes of thermophilic actinomycetes, despite the fact that macroamphiphiles may represent as much as 10 mol% of the lipid in the outer leaflet of the plasma membrane in mesophilic bacteria ( 11, 12). The aim of the present study was therefore to investigate the presence of macroamphiphiles (notably lipoglycans or LTA) in the cell envelopes of the thermophilic actinomycetes Thermobifida fusca and Rubrobacter xylanophilus. T. fusca is of considerable biotechnological interest as a source of novel heat-stable enzymes ( 33, 59) and is of interest from a comparative-genomics perspective as a sporulating, filamentous actinomycete ( 7). R. xylanophilus is a less intensively studied organism, although its genome sequence has recently been completed and it is of interest with regard to the mechanisms of radiation resistance ( 4, 10; http://genome.jgi-psf.org/finished_microbes/rubxy/rubxy.home.html). Moreover, the phylogenetic position of the subclass Rubrobacteridae is of interest, because this lineage may represent one of the earliest branches of the class Actinobacteria ( 4, 16, 17, 30, 48). Interestingly, our studies have shown that whereas T. fusca synthesizes a PGP-LTA, the cell envelope of R. xylanophilus appears to lack a macroamphiphile. Bacterial cultivation. T. fusca strain YX ( 33) was very kindly supplied by David Wilson and Diane Irwin (Department of Molecular Biology and Genetics, Cornell University). Biomass was prepared from cultures grown in the semidefined medium of Hägerdal et al. ( 24), modified so that glucose or cellobiose was used as a carbon source. Cultures were incubated at 55°C with shaking (150 rpm) and were harvested by centrifugation (4,000 × g, 20 min, 4°C) when the cells had reached the early-stationary phase (typically after 48 h of culture). Cells were washed with phosphate-buffered saline, harvested by centrifugation, and lyophilized. R. xylanophilus DSM 9941 T ( 4) was very kindly supplied by Milton da Costa (Departamento de Bioquímica, Universidade de Coimbra, Coimbra, Portugal). Biomass was prepared from cultures grown in yeast extract-malt extract (YEME) medium. Cultures were incubated at 55°C with shaking (150 rpm) and were harvested by centrifugation (4,000 × g, 20 min, 4°C) when the cells had reached the early-stationary phase (typically after 48 h of culture). Cells were washed with phosphate-buffered saline, harvested by centrifugation, and lyophilized. Macroamphiphile extraction. For hot-phenol-water extraction ( 14, 58), lyophilized bacterial cells were resuspended at 50 mg/ml in distilled water, mixed with an equal volume of hot (68°C) 90% (wt/vol) phenol, and extracted for 1 h at 68°C in a shaking water bath (120 rpm). The single-phase extract was separated into distinct aqueous and phenol phases by centrifugation (1 h, 4,000 × g, 4°C), and the upper aqueous phase was withdrawn. The phenol phase was washed with an equal volume of water, and the aqueous wash was recovered by centrifugation as before. The combined aqueous extracts were extensively dialyzed to remove phenol traces and were freeze-dried. All dialysis steps were performed with low-molecular-weight cutoff dialysis tubing (SnakeSkin pleated dialysis tubing; Pierce). For butanol extraction ( 37, 55), wet bacterial cells were suspended in 0.1 M sodium citrate buffer (pH 4.7) at 0.66 mg/ml. The cells were disrupted by sonication (10-s pulse for 2 min on ice). The disrupted cells were mixed with an equal volume of n-butanol and stirred for 30 min at room temperature. After centrifugation at 13,000 × g for 20 min, the lower aqueous phase was recovered, subjected to vacuum rotary evaporation, extensively dialyzed, and freeze-dried. Chloroform-methanol extraction of R. xylanophilus was performed using an adaptation of the method of Behr et al. ( 1). Briefly, cells were suspended at 50 mg/ml in sodium acetate buffer (pH 4.7) and disrupted by sonication (10-s pulse for 2 min). After the pH was adjusted to 5.5 with 1 M NaHCO 3, 2 volumes of methanol (MeOH) and 1 volume of CHCl 3 were added, and the mixture was stirred at room temperature for 3 h. After centrifugation (600 × g, 30 min), the supernatant was withdrawn, and the pellet was resuspended in 4 volumes of 50 mM sodium acetate (pH 5.5)-MeOH-CHCl 3 (0.8:2.0:1.0; by volume) and stirred overnight. After centrifugation, the supernatant was withdrawn, combined with the first supernatant extract, and adjusted to a final CHCl 3-MeOH-H 2O proportion of 1.0:1.0:0.9. After centrifugation, the water phase was extracted twice with chloroform, freed from methanol by rotary evaporation at 20°C, dialyzed against three 5-liter changes of distilled water, and freeze-dried. Macroamphiphile purification. Crude cell extracts were subjected to hydrophobic interaction chromatography (HIC) using methods described previously ( 13, 50). The crude extracts were loaded onto a 1.75- by 20-cm Octyl-Sepharose CL-4B (Sigma-Aldrich) column in 8 ml of equilibration buffer (100 mM sodium acetate [pH 4.5] containing 15% [vol/vol] n-propanol), and the column was eluted with 48 ml of this buffer. Hydrophobically retained material was then eluted with a 192-ml gradient of 15-to-65% (vol/vol) n-propanol in 100 mM sodium acetate (pH 4.5) buffer using an automated fast performance liquid chromatography system (Pharmacia Biotech). Fractions (4 ml) were collected. Following fast performance liquid chromatography-HIC, column fractions were assayed for carbohydrate by the method of Fox and Robyt ( 15) and for phosphate by the method of Chen et al. ( 8) and by dot immunoblotting. Dot immunoblotting was performed by spotting 1-μl samples from the HIC fractions onto nitrocellulose blotting membranes, followed by blocking, washing, and development with a monoclonal antibody for the detection of LTA as described below. Gradient-eluted peak fractions of interest were pooled, dialyzed extensively, and freeze-dried. Protein contamination was assayed using a commercial kit (Bradford reagent; Sigma-Aldrich) and judged to be minimal. Reference LTA. Streptococcus agalactiae strain A909 (group B streptococcus [GBS]) was used as a source of PGP-LTA ( 9). S. agalactiae was cultured in Todd-Hewitt broth and harvested by centrifugation. Washed cells were phenol-water extracted, and the LTA was purified by HIC as described above. Electrophoresis and Western blotting procedures. Macroamphiphile preparations were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a discontinuous electrophoresis system with 15% resolving gels in a minigel format (Mini-Protean II; Bio-Rad). Gels were stained with Alcian blue 8GX (0.05% in distilled water; Sigma). Macroamphiphiles were further analyzed following electrophoretic transfer (Western blotting; Transblot apparatus; Bio-Rad, United Kingdom) to nitrocellulose membranes (pore size, 0.2 μm; Bio-Rad). After transfer, the membranes were incubated overnight with a blocking solution of 5% skim milk in phosphate-buffered saline containing 0.05% (wt/vol) Tween 80 (PBST). Blots were subsequently incubated for 2 h with a monoclonal anti-LTA antibody diluted 1/2,000 in 5% (wt/vol) skim milk in PBST. After thorough washing with five changes of PBST, the blots were incubated with an alkaline phosphatase-conjugated anti-human immunoglobulin G (Dako A/S, Denmark) secondary antibody diluted 1/2,000 in 5% (wt/vol) skim milk in PBST. After through washing with five changes of PBST, the blots were developed with a 5-bromo-4-chloro-3-indolylphosphate (BCIP)-nitroblue tetrazolium (NBT) alkaline phosphatase substrate solution (Zymed, CA). The monoclonal anti-LTA antibody BSYX-A110 (Pagibaximab) ( 56) was very kindly provided by Biosynexus Incorporated (Gaithersburg, MD). Fatty acid derivatization for GC. Fatty acids in the macroamphiphiles (typically 1 to 2 mg) and lyophilized bacterial whole cells (ca. 30 mg) were analyzed by gas chromatography (GC) following derivatization to their fatty acid methyl esters (FAMEs) by acid-catalyzed methanolysis using 1 ml of 1.5% (vol/vol) sulfuric acid in anhydrous methanol (16 h, 50°C). FAMEs were recovered by three extractions into 3 ml hexane. The pooled hexane phases (ca. 9 ml) were backwashed with an equal volume of water, removed, and dried over anhydrous sodium sulfate. Finally, the pooled hexane phases were concentrated under nitrogen. Major fatty acids were identified by comparison of retention times with authentic FAME standards (Sigma Chemical Co.). Alditol acetate derivatization. For analysis of carbohydrate composition, samples (typically 1 to 2 mg of macroamphiphile) were acid hydrolyzed with 2 M hydrochloric acid (2 h at 120°C) in sealed ampoules. Hydrolysates were neutralized by drying in vacuo over sodium hydroxide, and the released carbohydrates were converted to their alditol acetate derivatives by the method of Saddler et al. ( 46). Sugars were identified by comparison of GC retention times with authentic sugar standards (Sigma Chemical Co., United Kingdom) derivatized by the same method. GC analysis. GC was carried out using an ATI-Unicam 610 series gas chromatograph fitted with a DB225 (J&W Scientific, Folsom, CA) fused silica capillary column (length, 30 m; internal diameter, 0.25 mm) with detection by flame ionization. Helium was used as the carrier gas. Major fatty acids were identified by comparison of retention times with those of authentic FAME standards (Sigma Chemical Co.). NMR methods. T. fusca LTA (LTA Tf) was analyzed by 1H and 2-dimensional nuclear magnetic resonance (NMR) as described previously ( 25, 44). Briefly, all spectra were recorded on a Bruker Avance 600 spectrometer at 300 K using a 5-mm BBI probe head. Samples were prepared as solutions in D 2O with sodium 3-trimethylsilyl-3,3,2,2-tetradeuteropropanoate (TSP) added as an internal standard for 1H NMR (δ H 0.00 ppm) and acetone for 13C NMR (δ C 30.02 ppm). Homonuclear assignments were based on 2-dimensional double-quantum-filtered correlation spectroscopy (DQF-COSY), total correlated spectroscopy (TOCSY), and rotational nuclear Overhauser effect spectroscopy (ROESY) experiments using presaturation for water suppression. TOCSY and ROESY experiments were performed in the phase-sensitive mode using mixing times of 100 ms for TOCSY and a 200-ms spin lock for ROESY. 13C chemical shift assignments were obtained from gradient-enhanced heteronuclear single-quantum correlation (HSQC) spectra. Data were acquired and processed by using standard Bruker software. The average number of repeating units in the polyglycerophosphate backbone and the percentage of substitution were determined by integration of the corresponding peak volumes in the 1H NMR. Phylogenetic analysis. A 16S rRNA gene sequence phylogenetic tree was constructed using the neighbor-joining method and the MEGA4 program ( 54) with 16S rRNA gene sequences obtained from Ribosomal Database Project II ( http://rdp.cme.msu.edu/). The tree was selected after consideration of 1,000 bootstrap replicates. Extraction and purification of a macroamphiphile from T. fusca. We used the standard hot-phenol-water procedure ( 14, 58) to extract freeze-dried cells of T. fusca. Following dialysis and lyophilization, extracts were purified by HIC. As shown in Fig. , hydrophilic contaminants (typically nucleic acids, proteins, etc.) were not retained by the column, while the propanol gradient elution recovered a macroamphiphilic component rich in phosphate and with appreciable carbohydrate content. This type of HIC profile is typical of LTA, which are phosphate rich due to the alditol phosphates in their polymer repeat units. In contrast, lipoglycans typically exhibit elution profiles with a substantial carbohydrate content and lesser or minimal amounts of phosphate ( 13, 18, 50, 52). A comparable HIC profile (data not shown) was obtained when T. fusca cells were extracted by the butanol-water method, which can also be used to extract LTA (see below). Using a monoclonal anti-PGP-LTA antibody, dot immunoblotting across the peak fractions identified by chemical analysis gave a positive reaction indicative of PGP-LTA (data not shown). This result was further substantiated after pooling, dialysis, and freeze-drying of the macroamphiphilic component. Western blot analysis showed the presence of a component that cross-reacted with the monoclonal anti-PGP-LTA antibody and that had an electrophoretic mobility similar to that of the reference PGP-LTA purified from S. agalactiae (Fig. ) and Streptococcus sanguinis (data not shown). The broad, smeared appearance of macroamphiphilic components in SDS-PAGE analysis has been observed previously with both LTA ( 21, 35) and various lipoglycans (see, e.g., references 18 and 50). Moreover, the macroamphiphilic component purified by HIC could also be stained with the cationic dye Alcian blue (Fig. ). Alcian blue has been used previously to detect various polyanionic components, including LTA ( 35). These data strongly suggested that the macroamphiphilic component extracted and purified from T. fusca was a PGP-LTA. To confirm these findings, the material was subjected to acid hydrolysis, and the products released were derivatized to alditol acetates and analyzed by GC. As expected, significant amounts of glycerol were detected, along with minor amounts of glucose (data not shown). The fatty acid composition of the whole bacterial cells was consistent with that reported previously for T. fusca ( 20, 60); it was dominated by branched-chain saturated fatty acids, notably 14-methylpentadecanoate (iso-C 16:0 [56%]), 15-methylhexadecanoate (anteiso-C 17:0 [10%] and iso-C 17:0 [6%]), and 16-methylheptadecanoate (iso-C 18:0 [8%]). In contrast, iso-C 16:0 was predominant in the macroamphiphilic component, along with traces of fatty acids consistent with the whole-cell profile, suggesting selective enrichment of iso-C 16:0 in the macroamphiphile. Cumulatively, these data suggested the presence of a PGP-LTA with a fatty acylated lipid anchor unit containing glucose (and possibly glucosyl substitution on the PGP polymer backbone). This material is referred to below as LTATf. The cell wall of T. fusca contains polyglycerophosphate teichoic acids ( 43). Teichoic acids are biosynthesized at the cytoplasmic face of the plasma membrane on a polyprenol-phosphate carrier lipid before they are flipped to the outer leaflet of the plasma membrane and transferred onto the growing cell wall ( 2, 41). Consequently, we wished to exclude the possibility that the LTA Tf described here was in fact a polyprenol-linked biosynthetic intermediate from teichoic acid biosynthesis. Several lines of evidence argued against this possibility. First, LTA Tf represents ca. 0.5% of the cell (dry weight), whereas biosynthetic intermediates in teichoic acid synthesis are expected to turn over rapidly and not accumulate to these levels. Second, LTA Tf contained fatty acids, consistent with a glycolipid-based anchor rather than a polyprenol-linked intermediate. Finally, we measured the absorbance spectrum of LTA Tf and detected no significant absorbance at 232 nm, indicating that the C=C double bonds that would be characteristic of a polyprenol-phosphate lipid carrier were absent. NMR analysis of LTATf. Advances in the field of high-resolution NMR in recent years have allowed the analysis of native LTA with an estimated mass between 5 and 10 kDa ( 25, 44). However, structural investigations of LTA remain challenging because of microheterogeneity in the fatty acid chains, the length of the repeating unit, and the glycosylation and acylation patterns. NMR analysis of HIC-purified LTA Tf confirmed the presence of a typical PGP-LTA structure (Fig. ; see also Tables S1 and S2 and Fig. S1 in the supplemental material) with an average chain length of approximately 19 glycerophosphate units. The proton spectrum in Fig. shows the typical line broadening of all signals belonging to the amphiphilic polymer and is clearly comparable to that previously described for the PGP-LTA of Lactobacillus delbrueckii ( 44). This larger half-width of NMR signals is characteristic for LTA and is caused by the microheterogeneity of these macromolecules and the micellar structure of native LTA in water. In the alkyl region, the TOCSY spectrum showed signals belonging to three separate spin systems, which were identified as three different types of fatty acids. Exact assignment of individual signals was possible using DQF-COSY. The corresponding 13C shifts were taken from the HSQC spectrum. These signals were consistent with straight-chain and branched (iso- and anteiso-) fatty acids in the glycolipid anchor (see Table S3 in the supplemental material), confirming the results of the GC analysis. Due to the line broadening in the NMR of the native LTA in aqueous solution, no further details about the structure of the membrane anchor could be obtained. The signals in the proton spectrum at δ H 1.33 and δ H 4.16 showing 3J coupling in DQF-COSY were assigned to lactic acid, also because of the chemical shifts of the appropriate signals from the HSQC spectrum. In the ROESY spectrum no further signals could be observed, and therefore at present the localization of this lactyl group within the LTA Tf molecule cannot be precisely determined. A heteronuclear multiple-bond correlation spectrum gave only very weak signals, mainly due to the high micellar aggregation, so no further information about the linkage of the lactyl groups could be obtained. Because of the high signal intensity, the lactic acid is most likely connected to the PGP backbone of LTA Tf. If it is assumed that this substituent is also linked to the C-2 hydroxyl of glycerol units, it can be calculated that about 24% of all C-2 hydroxyl groups are esterified with lactic acid. Again, consistent with the GC analysis, glucosyl substituents were detected by NMR methods. In the proton spectrum, the signals at δ H 4.64 and 4.68 were assigned to anomeric protons of two different glucose moieties. Interestingly, these glucosyl substituents were identified as being β-glycosidically linked to the C-2 hydroxyl of the PGP repeating unit. β-Glycosidic pyranoses have been reported previously as components of a heavily substituted PGP wall teichoic acid in T. fusca ( 43). A nuclear Overhauser effect (NOE) contact between the anomeric protons and the methine protons at δ H 4.16 and 4.19, respectively, confirmed this assignment. For further assignment of the glucosyl ring protons, DQF-COSY and TOCSY were used. Glucose B has a free hydroxyl group at C-6, whereas glucose A is acetylated at the primary OH group based on the carbon shift of its methylene group at δ C 64.9. A cross-peak in the NOE spectrum between the H6,H6′ cluster of glucose A and the methyl group of the acetate at δ H 2.15 supports this conclusion. A further signal for a second acetyl group that is directly linked to the C-2 of the PGP repeat unit was identified at δ H 2.17. This linkage was proven by an appropriate NOE signal. The average number of repeating units in the PGP backbone and the overall composition of the different substituents were determined by integration of the corresponding peak volumes in the 1H NMR spectrum and calculation of the ratios between the signal intensities of the backbone and the signals of the fatty acids. d-Alanine substituents, which are typical of many PGP-LTA ( 11, 41), were not detected. In Firmicutes such as Bacillus subtilis and Lactobacillus rhamnosus, the Dlt system provides a dedicated mechanism for the activation, relay, and ligation of d-alanine to teichoic acids and LTA ( 41). The absence of d-alanine in LTA Tf is consistent with our failure to detect significant homologues of the DltB, DltC, and DltD proteins encoded in the T. fusca genome (data not shown). Also consistent with this was the absence of d-alanine substituents on the teichoic acids of T. fusca ( 43). Cumulatively, the chemical, electrophoretic, and NMR analyses are consistent with the structure of LTA Tf being a PGP-LTA predominantly bearing β-glucosyl substituent groups (Fig. ) and membrane anchored by covalent attachment to a glycolipid, which we presume to be a glucosyl-containing diacylglyceride glycolipid. We note that Thermobifida spp. synthesize phosphatidylinositol ( 60) and that the genome sequence suggests an operon (Tfu_2101-Tfu_2102) consistent with the synthesis of phosphatidylinositol mannoside, the anchor unit of the lipoarabinomannan family of lipoglycans ( 18, 51). Our NMR analyses gave no suggestion of a phosphatidylinositol-based anchor in LTA Tf. Extraction and purification of extracts from R. xylanophilus. We also used the hot-phenol-water procedure to extract freeze-dried cells of R. xylanophilus. Following dialysis and lyophilization, the extract was purified by HIC. As shown in Fig. , hydrophilic contaminants were washed straight through the column. However, unusually, the propanol gradient failed to elute any significant carbohydrate- and/or phosphate-containing material, suggesting that no macroamphiphilic component was present (compare Fig. and ). To explore this possibility further, freeze-dried cells of R. xylanophilus were also extracted using the butanol-water method, which has been used previously to extract LTA from gram-positive bacteria and lipopolysaccharide from various gram-negative bacteria ( 37, 39, 55). Again, following application of the extract to the HIC column, an increasing propanol gradient failed to elute any significant carbohydrate- and/or phosphate-containing material (data not shown). Finally, the F antigen (LTA) of Streptococcus pneumoniae has been shown to partition anomalously during standard phenol-water extraction ( 1). We therefore used a modification of the chloroform-methanol-based method used to extract pneumococcal F antigen ( 1). As with our previous attempts, chemical analysis of fractions following HIC purification indicated that this method had also failed to recover macroamphiphilic carbohydrate- and/or phosphate-containing material from freeze-dried cells of R. xylanophilus (data not shown). We therefore conclude that R. xylanophilus does not synthesize significant quantities of a carbohydrate- and/or phosphate-containing macroamphiphile. This is a highly unusual finding: virtually all studies that have examined extracts of gram-positive bacteria for LTA or lipoglycans have recovered one or the other of these macroamphiphiles. To our knowledge, the only previous reports of the absence of LTA or a related macroamphiphile from gram-positive bacteria are studies of some Bacillus spp. Bacillus sp. strain A007 failed to turn over radiolabeled phosphatidylglycerol into PGP-LTA, although the presence of an alternative macroamphiphile was not investigated ( 28). Likewise, strains of Bacillus polymyxa (now Paenibacillus polymyxa) and Bacillus circulans lack PGP-LTA ( 27). However, neither of these studies used HIC to distinguish macroamphiphiles from other nonlipidated cellular components, and so the absence of structurally variant (non-PGP) LTA or lipoglycan in these species needs definitive confirmation. Implications for membrane composition and organization in thermophiles. Our finding that the cell envelope of T. fusca contains a PGP-LTA is of interest given the evident need for thermophiles to carefully regulate their membrane composition in order to counteract the fluidizing effects on membrane lipids of growth at high temperatures ( 6, 29, 45). Notably, LTA are not inherently bilayer-forming lipids, and they form micelles in aqueous solutions ( 12, 22, 23, 31). However, biophysical studies have suggested that appropriately regulated amounts of PGP-LTA may help stabilize the membrane surface due to interactions between the PGP and membrane lipid headgroups ( 23). The absence of d-alanine and the presence of acetyl and lactyl substituents will affect the biophysical properties of LTA Tf (notably its high overall negative charge) and thus will be an interesting subject for future study. The present study, along with previous studies of Bacillus coagulans ( 26) and Bacillus stearothermophilus (now Geobacillus stearothermophilus) ( 3), is significant in confirming that the presence of LTA is clearly compatible with effective membrane function during growth at high temperatures. However, it is interesting that the fatty acid composition of the LTA anchor is enriched in iso-C 16:0 compared to the whole-cell fatty acid profile. Conversely, our failure to recover LTA or lipoglycan from the membrane of R. xylanophilus suggests that a macroamphiphile is not per se needed as a plasma membrane component in either mesophilic or thermophilic gram-positive bacteria. The latter finding is of obvious significance when one considers the physiological functions of LTA and lipoglycans, which as yet remain to be fully determined ( 11, 21, 32, 49, 51). Chemotaxonomic considerations. The distribution of LTA and lipoglycans has some chemotaxonomic utility, typically at the generic and suprageneric levels ( 18, 49, 52). Our finding of PGP-LTA in T. fusca corroborates previous findings that a Streptomyces sp. ( 42) and four representatives of the genus Agromyces ( 19) produce PGP-LTA, i.e., that some actinobacteria synthesize LTA. We have also identified a putative LTA in the model actinomycete Streptomyces coelicolor M145 (O. Rahman and I. C. Sutcliffe, unpublished data). To place our data in a phylogenetic context, we constructed a 16S rRNA gene phylogenetic tree of representative gram-positive bacteria (Fig. ). The phylogenetic analysis demonstrates the well-established separation of the Tenericutes, the low-G+C Firmicutes, and the high-G+C Actinobacteria, with Rubrobacter an outlier to the latter ( 16, 17, 30, 48). Consideration of the distribution of LTA and lipoglycans in these taxa suggests that the previous assumption that LTA are underrepresented in the Actinobacteria may simply represent a sampling bias, since many major lineages remain to be investigated to determine the nature of the macroamphiphile present (Fig. ). Studies of lipoglycans have tended to focus on closely related taxa, such as the Mycolata and the Micrococcaceae ( 18, 51, 52) (Fig. ). Future studies of the Actinobacteria may therefore uncover further diversity in macroamphiphile composition. Interestingly, the cell envelopes of all the actinobacteria thus far shown to contain LTA also contain teichoic acids ( 19, 40, 43), consistent with the view that these components may interact in the formation of a “continuum of anionic charge” ( 41). Indeed, as noted previously ( 57), there is a striking distinction between two gram-positive cell envelope archetypes: one (polyanionic) dominated by the presence of LTA and teichoic acids and the other dominated by lipoglycans and other types of cell wall glycan. However, we have also observed that the recently described LtaS polymerase, which is essential for PGP-LTA synthesis in S. aureus ( 21), lacks clear orthologues in the sequenced actinobacterial genomes, including that of T. fusca, although more distantly related proteins that (like LtaS) also belong to the PF00884 sulfatase family are present. Thus, LTA biosynthesis may proceed by an alternative pathway in actinomycetes. These findings are important in understanding the phylogeny of gram-positive organisms, since the precise relationship between the Firmicutes and the Actinobacteria has been difficult to define (for an example, including an impassioned reappraisal of the use of the term “ Firmicutes,” see reference 5). Chemotaxonomic data, such as those presented here, suggest that many taxa of both Firmicutes and Actinobacteria have significant shared cell envelope characteristics. Just as representatives of both Firmicutes and Actinobacteria synthesize LTA, it is notable that representatives of some lineages phylogenetically placed within the Firmicutes synthesize lipoglycans (Fig. ), notably the Mollicutes ( 47), although this lineage warrants a separate status as the phylum “ Tenericutes” ( http://www.bacterio.cict.fr/classifphyla.html#Tenericutes). The finding that R. xylanophilus apparently lacks a macroamphiphile is also notable given that previous analyses, and the present study, have shown that the Rubrobacteridae represent one of the deepest branches of the class Actinobacteria ( 4, 16, 17, 30, 48). However, it remains a formal possibility that R. xylanophilus produces a highly atypical macroamphiphile (i.e., lacking both carbohydrate and/or phosphate), which would not have been detected in our study. Concluding comments. Our data have shown that T. fusca synthesizes a PGP-LTA, a finding that highlights the need for further studies of the distribution and biosynthesis of this macroamphiphile in Actinobacteria and also the need for further consideration of the roles of macroamphiphiles in the membrane adaptation of thermophiles. However, the apparent absence of both LTA and lipoglycan in the membranes of R. xylanophilus is intriguing given the presumed functional importance of macroamphiphiles in the membranes of gram-positive bacteria and is notable given the phylogenetic position of the genus Rubrobacter. We are grateful to Biosynexus Incorporated (Gaithersburg, MD) for the generous gift of the monoclonal anti-LTA antibody used in this study. We thank David Wilson and Diane Irwin (Cornell University) and Milton da Costa (Universidade de Coimbra, Coimbra, Portugal) for providing the bacterial strains used in this study. 1. Behr, T., W. Fischer, J. Peterkatalinić, and H. Egge. 1992. The structure of pneumococcal lipoteichoic acid—improved preparation, chemical and mass-spectrometric studies. Eur. J. Biochem. 2071063-1075. [PubMed] 2. Bhavsar, A. P., and E. D. Brown. 2006. Cell wall assembly in Bacillus subtilis: how spirals and spaces challenge paradigms. Mol. Microbiol. 601077-1090. [PubMed] 3. Card, G. L., and D. J. Finn. 1983. Products of phospholipid metabolism in Bacillus stearothermophilus. J. Bacteriol. 154294-303. [PubMed] 4. Carreto, L., E. Moore, M. F. Nobre, R. Wait, P. W. Riley, R. J. Sharp, and M. S. Da Costa. 1996. Rubrobacter xylanophilus sp. nov.: a new thermophilic species isolated from a thermally polluted effluent. Int. J. Syst. Bacteriol. 46460-465. 5. Cavalier-Smith, T. 2006. Rooting the tree of life by transition analysis. Biol. Direct. 119. [PubMed] 6. Charlier, D., and L. Droogmans. 2005. Microbial life at high temperature, the challenges, the strategies. Cell. Mol. Life Sci. 622974-2984. [PubMed] 7. Chater, K. F., and G. Chandra. 2006. The evolution of development in Streptomyces analyzed by genome comparisons. FEMS Microbiol. Rev. 30651-672. [PubMed] 8. Chen, P. S., T. Y. Toribara, and H. Warner. 1956. Microdetermination of phosphorus. Anal. Chem. 281756-1758. 9. Doran, K. S., E. J. Engelson, A. Khosravi, H. C. Maisey, I. Fedtke, O. Equils, K. S. Michelsen, M. Arditi, A. Peschel, and V. Nizet. 2005. Blood-brain barrier invasion by group B Streptococcus depends upon proper cell-surface anchoring of lipoteichoic acid. J. Clin. Investig. 1152499-2507. [PubMed] 10. Ferreira, A. C., M. F. Nobre, E. Moore, F. A. Rainey, J. R. Battista, and M. S. Da Costa. 1999. Characterization and radiation resistance of new isolates of Rubrobacter radiotolerans and Rubrobacter xylanophilus. Extremophiles 3235-238. [PubMed] 11. Fischer, W. 1994. Lipoteichoic acids and lipoglycans, p. 119-215. In J. M. Ghuysen and R. Hakenbeck (ed.), Bacterial cell wall. Elsevier Science, B.V., Amsterdam, The Netherlands. 12. Fischer, W. 1994. Lipoteichoic acid and lipids in the membrane of Staphylococcus aureus. Med. Microbiol. Immunol. 18361-76. [PubMed] 13. Fischer, W. 1996. Molecular analysis of lipid macroamphiphiles by hydrophobic interaction chromatography. J. Microbiol. Methods 25126-144. 14. Fischer, W., H. U. Koch, and R. Haas. 1983. Improved preparation of lipoteichoic acids. Eur. J. Biochem. 133523-530. [PubMed] 15. Fox, J. D., and J. F. Robyt. 1991. Miniaturization of three carbohydrate analyses using a microsample plate reader. Anal. Biochem. 19593-96. [PubMed] 16. Gao, B., and R. S. Gupta. 2005. Conserved indels in protein sequences that are characteristic of the phylum Actinobacteria. Int. J. Syst. Evol. Microbiol. 552401-2412. [PubMed] 17. Gao, B., R. Paramanathan, and R. S. Gupta. 2006. Signature proteins that are distinctive characteristics of Actinobacteria and their subgroups. Antonie van Leeuwenhoek 9069-91. [PubMed] 18. Gilleron, M., N. J. Garton, J. Nigou, T. Brando, G. Puzo, and I. C. Sutcliffe. 2005. Characterization of truncated lipoarabinomannan from the actinomycete Turicella otitidis. J. Bacteriol. 187854-861. [PubMed] 19. Gnilozub, V. A., G. M. Streshinskaya, L.I. Evtushenko, A.S. Shashkov, and I. B. Naumova. 1994. Lipoteichoic acids of Agromyces species. Microbiology USSR [Mikrobiologiya] 63275-279. 20. Greneir-Mai, E., R. M. Kroppenstedt, F. Korn-Wendisch, and H. J. Kutzner. 1987. Morphological and biochemical characterization and emended descriptions of thermophilic actinomycetes species. Syst. Appl. Microbiol. 997-109. 21. Grundling, A., and O. Schneewind. 2007. Synthesis of glycerol phosphate lipoteichoic acid in Staphylococcus aureus. Proc. Natl. Acad. Sci. USA 1048478-8483. [PubMed] 22. Gutberlet, T., K. Milde, H. Bradaczek, H. Haas, and H. Möhwald. 1994. Miscibility of lipoteichoic acid in dipalmitoylphosphatidylcholine studied by monofilm investigations and fluorescence microscopy. Chem. Phys. Lipids 69151-159. [PubMed] 23. Gutberlet, T., J. Frank, H. Bradaczek, and W. Fischer. 1997. Effect of lipoteichoic acid on thermotropic membrane properties. J. Bacteriol. 1792879-2883. [PubMed] 24. Hägerdal, B. G., J. D. Ferchak, and E. K. Pye. 1978. Cellulolytic enzyme system of Thermoactinomyces sp. grown in microcrystalline cellulose. Appl. Environ. Microbiol. 36606-612. [PubMed] 25. Henneke, P., S. Morath, S. Uematsu, S. Weichert, M. Pfitzenmaier, O. Takeuchi, A. Müller, C. Poyart, S. Akira, R. Berner, G. Teti, A. Geyer, T. Hartung, P. Trieu-Cuot, D. L. Kasper, and D. T. Golenbock. 2005. Role of lipoteichoic acid in the phagocyte response to group B streptococcus. J. Immunol. 1746449-6455. [PubMed] 26. Iwasaki, H., A. Shimada, and E. Ito. 1986. Comparative studies of lipoteichoic acids from Bacillus strains. J. Bacteriol. 167508-516. [PubMed] 27. Iwasaki, H., A. Shimada, K. Yokoyama, and E. Ito. 1989. Structure and glycosylation of lipoteichoic acids in Bacillus strains. J. Bacteriol. 171424-429. [PubMed] 28. Koga, Y., M. Nishihara, and H. Morii. 1984. Products of phosphatidylglycerol turnover in two Bacillus strains with and without lipoteichoic acid in the cells. Biochim. Biophys. Acta 79386-94. [PubMed] 29. Konings, W. N., S. V. Albers, S. Koning, and A. J. M. Driessen. 2002. The cell membrane plays a crucial role in survival of bacteria and archaea in extreme environments. Antonie van Leeuwenhoek 8161-72. [PubMed] 30. Kunisawa, T. 2007. Gene arrangements characteristic of the phylum Actinobacteria. Antonie van Leeuwenhoek 92359-365. [PubMed] 31. Labischinski, H., D. Naumann, and W. Fischer. 1991. Small and medium-angle X-ray analysis of bacterial lipoteichoic acid phase-structure. Eur. J. Biochem. 2021269-1274. [PubMed] 32. Lambert, P. A., I. C. Hancock, and J. Baddiley. 1977. Occurrence and function of membrane teichoic acids. Biochim. Biophys. Acta 4721-12. [PubMed] 33. Lykidis, A., K. Mavromatis, N. Ivanova, I. Anderson, M. Land, G. DiBartolo, M. Martinez, A. Lapidus, S. Lucas, A. Copeland, P. Richardson, D. B. Wilson, and N. Kyrpides. 2007. Genome sequence and analysis of the soil cellulolytic actinomycete Thermobifida fusca YX. J. Bacteriol. 1892477-2486. [PubMed] 34. Matias, V. R. F., and T. J. Beveridge. 2006. Native cell wall organization shown by cryo-electron microscopy confirms the existence of a periplasmic space in Staphylococcus aureus. J. Bacteriol. 1881011-1021. [PubMed] 34a. Matias, V. R. F., and T. J. Beveridge. 12 September 2008. Lipoteichoic acid is a major component of the Bacillus subtilis periplasm. J. Bacteriol. doi: doi: 10.1128/JB.00581-08.. [PubMed] 35. Maurer, J. J., and S. J. Mattingly. 1991. Molecular analysis of lipoteichoic acid from Streptococcus agalactiae. J. Bacteriol. 173487-494. [PubMed] 36. Merchante, R., H. M. Pooley, and D. Karamata. 1995. A periplasm in Bacillus subtilis. J. Bacteriol. 1776176-6183. [PubMed] 37. Morath, S., A. Geyer, and T. Hartung. 2001. Structure-function relationship of cytokine induction by lipoteichoic acid from Staphylococcus aureus. J. Exp. Med. 193393-397. [PubMed] 38. Morath, S., S. von Aulock, and T. Hartung. 2005. Structure/function relationships of lipoteichoic acids. J. Endotoxin Res. 11348-356. [PubMed] 39. Morrison, D. C., and L. Leive. 1975. Fractions of lipopolysaccharide for Escherichia coli O111:B4 prepared by two extraction procedures. J. Biol. Chem. 2502911-2919. [PubMed] 40. Naumova, I. B., V. D. Kuznestov, K.S. Kudrina, and A. P. Bezzubenkova. 1980. The occurrence of teichoic acids in streptomycetes. Arch. Microbiol. 12671-75. [PubMed] 41. Neuhaus, F. C., and J. Baddiley. 2003. A continuum of anionic charge: structures and functions of d-alanyl-teichoic acids in gram-positive bacteria. Microbiol. Mol. Biol. Rev. 67686-723. [PubMed] 42. Potekhina, N. V., G. M. Streshinskaya, G. V. Novitskaya, and I. B. Naumova. 1983. Isolation of lipoteichoic acid from Streptomyces levoris. Microbiology USSR [Mikrobiologiya] 52340-343. 43. Potekhina, N. V., A. S. Shashkov, L. I. Evtushenko, and I. B. Naumova. 2003. Teichoic acids in the cell walls of Microbispora mesophila Ac-1953T and Thermobifida fusca Ac-1952T. Microbiology USSR [Mikrobiologiya] 72157-161. 44. Räisänen, L., C. Draing, M. Pfitzenmaier, K. Schubert, T. Jaakonsaari, S. von Aulock, T. Hartung, and T. Alatossava. 2007. Molecular interaction between lipoteichoic acids and Lactobacillus delbrueckii phages depends on d-alanyl and α-glucose substitution of poly(glycerophosphate) backbones. J. Bacteriol. 1894135-4140. [PubMed] 45. Russell, N. J., and N. Fukunaga. 1990. A comparison of thermal adaptation of membrane lipids in psychrophilic and thermophilic bacteria. FEMS Microbiol. Rev. 75171-182. 46. Saddler, G. S., P. Tavecchia, S. Lociuro, M. Zanol, L. Colombo, and E. Selva. 1991. Analysis of madurose and other actinomycete whole cell sugars by gas chromatography. J. Microbiol. Methods 14185-191. 47. Smith, P. F. 1985. Detection of lipoglycans in ureaplasmas. J. Bacteriol. 162611-614. [PubMed] 48. Stackebrandt, E., F. A. Rainey, and N. L. Ward-Rainey. 1997. Proposal for a new hierarchic classification system, Actinobacteria classis nov. Int. J. Syst. Bacteriol. 47479-491. 49. Sutcliffe, I. C. 1994. The lipoteichoic acids and lipoglycans of Gram-positive bacteria: a chemotaxonomic perspective. Syst. Appl. Microbiol. 17467-480. 50. Sutcliffe, I. C. 2000. Characterisation of a lipomannan lipoglycan from the mycolic acid containing actinomycete Dietzia maris. Antonie van Leeuwenhoek 78195-201. [PubMed] 51. Sutcliffe, I. C. 2005. Lipoarabinomannans—structurally diverse and functionally enigmatic macroamphiphiles of mycobacteria and related actinomycetes. Tuberculosis 85205-206. [PubMed] 52. Sutcliffe, I. C., and G. Alderson. 1995. A chemotaxonomic appraisal of the distribution of lipomannans within the genus Micrococcus. FEMS Microbiol. Lett. 133233-237. 53. Sutcliffe, I. C., and N. Shaw. 1991. Atypical lipoteichoic acids of gram-positive bacteria. J. Bacteriol. 1737065-7069. [PubMed] 54. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: molecular evolutionary genetics analysis (MEGA) software, version 4.0. Mol. Biol. Evol. 241596-1599. [PubMed] 55. Theilacker, C., Z. Kaczynski, A. Kropec, F. Fabretti, T. Sange, O. Holst, and J. Huebner. 2006. Opsonic antibodies to Enterococcus faecalis strain 12030 are directed against lipoteichoic acid. Infect. Immun. 745703-5712. [PubMed] 56. Walsh, S., J. Kokai-Kun, A. Shah, and J. Mond. 2004. Extended nasal residence time of lysostaphin and an anti-staphylococcal monoclonal antibody by delivery in semisolid or polymeric carriers. Pharm. Res. 211770-1775. [PubMed] 57. Weidenmaier, C., and A. Peschel. 2008. Teichoic acids and related cell-wall glycopolymers in Gram-positive physiology and host interactions. Nat. Rev. Microbiol. 6276-287. [PubMed] 58. Westphal, O., and K. Jann. 1965. Bacterial lipopolysaccharides. Extraction with phenol-water and further application of the procedure. Methods Carbohydr. Chem. 583-91. 59. Wilson, D. B. 2004. Studies of Thermobifida fusca plant cell wall degrading enzymes. Chem. Rec. 472-82. [PubMed] 60. Yang, L.-L., S.-K. Tang, Y.-Q. Zhang, X.-Y. Zhi, D. Wang, L.-H. Xu, and W.-J. Li. 2008. Thermobifida halotolerans sp. nov., isolated from a salt mine sample, and emended description of the genus Thermobifida. Int. J. Syst. Evol. Microbiol. 581821-1825. [PubMed] 61. Zuber, B., M. Haenni, T. Ribeiro, K. Minnig, F. Lopes, P. Moreillon, and J. Dubochet. 2006. Granular layer in the periplasmic space of gram-positive bacteria and fine structures of Enterococcus gallinarum and Streptococcus gordonii septa revealed by cryo-electron microscopy of vitreous sections. J. Bacteriol. 1886652-6660. [PubMed]
|
