Portal de Periódicos da CAPES

Microextraction of bacterial lipid A: easy and rapid method for mass spectrometric characterization

2005; Elsevier BV; Volume: 46; Issue: 8 Linguagem: Inglês

10.1194/jlr.d500014-jlr200

1539-7262

Asmaa El Hamidi, Alina Tîrşoaga, А. В. Новиков, Ahmed Hussein, Martine Caroff,

Bacterial Identification and Susceptibility Testing

Endotoxins (lipopolysaccharides) are the main components of Gram-negative bacterial outer membranes. A quick and simple way to isolate their lipid region (lipid A) directly from whole bacterial cells was devised. This method using hot ammonium-isobutyrate solvent was applied to small quantities of cells and proved to be indispensable when a rapid characterization of lipid A structure by mass spectrometry was required. Biological activities of endotoxins are directly related to the lipid A structures, which vary greatly with cell growth conditions.This method is suitable for rough- and smooth-type bacteria and very efficient for screening variations in lipid A structures. Data are acquired in a few hours and avoid the use of phenol in extraction. Endotoxins (lipopolysaccharides) are the main components of Gram-negative bacterial outer membranes. A quick and simple way to isolate their lipid region (lipid A) directly from whole bacterial cells was devised. This method using hot ammonium-isobutyrate solvent was applied to small quantities of cells and proved to be indispensable when a rapid characterization of lipid A structure by mass spectrometry was required. Biological activities of endotoxins are directly related to the lipid A structures, which vary greatly with cell growth conditions. This method is suitable for rough- and smooth-type bacteria and very efficient for screening variations in lipid A structures. Data are acquired in a few hours and avoid the use of phenol in extraction. Lipopolysaccharides (LPSs) are the major components of the external membrane of almost all Gram-negative bacteria (1Raetz C.R. Biochemistry of endotoxins.Annu. Rev. Biochem. 1990; 59: 129-170Google Scholar, 2Brade Rietschel, E. T, O. Holst, A. J. Ulmer, H. Flad H. Biochemistry and cell biology of bacterial endotoxin.FEMS Immunol. Med. Microbiol. 1996; 16: 83-104Google Scholar); they are known as endotoxins and may cause several pathophysiological symptoms, such as fever, diarrhea, blood pressure decrease, septic shock, and death (3Morrison D.C. Ryan J.L. Endotoxins and disease mechanisms.Annu. Rev. Med. 1987; 38: 417-432Google Scholar).The LPS molecular architecture consists of three different regions. The innermost, hydrophobic region, lipid A, is responsible for the major toxic and beneficial properties of bacterial endotoxins (4Caroff M. Karibian D. Structure of bacterial lipopolysaccharides.Carbohydr. Res. 2003; 338: 2431-2447Google Scholar). Lipid A is the least variable part of the molecule among the different species of a genus, and its structure generally consists of a diglucosamine backbone substituted with varying numbers (usually four to seven) of ester- or amide-linked fatty acids. Phosphate and/or other substituents are linked to carbons at the C-1 and C-4′ positions of the glucosamine disaccharide (5Takayama K. Qureshi N. Chemical structure of lipid A.in: Morrison D.C. Ryan J.L. Bacterial Endotoxic Lipopolysaccharides. Molecular Biochemistry and Cellular Biology. CRC Press, Boca Raton, FL1992: 43-65Google Scholar). A 2-keto-3-deoxyoctonate (Kdo) unit links the lipid A to a core oligosaccharide (OS) composed of ∼10 sugar residues divided into two regions: a best conserved inner core part and a distal outer core. The core is linked to a third outermost region of a highly immunogenic and variable O-chain polysaccharide (PS) or O-antigen made up of repeating OS units. The latter region of the LPS molecule is responsible for bacterial serological strain specificity (6Weynants V. Gilson D. Cloeckaert A. Tibor A. Denoel P.A. Godfroid F. Limet J.N. Letesson J.J. Characterization of smooth lipopolysaccharides and O polysaccharides of Brucella species by competition binding assays with monoclonal antibodies.Infect. Immun. 1997; 65: 1939-1943Google Scholar) and is present only in smooth-type bacteria. The so-called rough-type bacteria produce LPSs lacking O-antigens.Endotoxin can be isolated from Gram-negative bacteria by different methods, the most efficient and commonly used one being the hot phenol-water extraction procedure introduced by Westphal and Lüderitz (7Westphal O. Lüderitz O. Chemische Erforschung von Lipopolysacchariden Gram-negativer Bakterien.Angew. Chem. 1954; 66: 407-417Google Scholar). It was later modified by different authors (8Westphal O. Jann K. Bacterial lipopolysaccharide. Extraction with phenol-water and further application of the procedure.Methods Carbohydr. Chem. 1965; 5: 83-91Google Scholar, 9Johnson K.G. Perry M.B. Improved techniques for the preparation of bacterial lipopolysaccharides.Can. J. Microbiol. 1997; 1: 29-34Google Scholar), and specific methods were developed for rough-type endotoxin extractions (10Galanos C. Luderlitz O. Westphal O. A new method for the extraction of R lipopolysaccharides.Eur. J. Biochem. 1968; 9: 245-249Google Scholar, 11Nurminen M. Vaara M. Methanol extracts LPS from deep rough bacteria.Biochem. Biophys. Res. Commun. 1996; 219: 441-444Google Scholar). However, each method requires several days for the extraction and purification of endotoxins and further steps to isolate the lipid A moiety. New methods have been described to extract LPS from small quantities of cells, such as by mini phenol extraction (12Li J. Thibault P. Martin A. Richards J.C. Wakarchuk W.W. Van der Wilp W. Development of an on-line preconcentration method for the analysis of pathogenic lipopolysaccharides using capillary electrophoresis-electrospray mass spectrometry. Application to small colony isolates.J. Chromatogr. A. 1998; 817: 325-336Google Scholar) or using an RNA-isolating reagent, but these methods still require 2 or 3 days and the use of phenol (13Yi C.E. Hackett M. Rapid isolation method for lipopolysaccharide and lipid A from Gram-negative bacteria.Analyst. 2000; 125: 651-656Google Scholar).In early experiments, mineral acid hydrolysis was used to liberate lipid A from endotoxins, splitting the acido-labile ketosidic bond of Kdo. It was followed in 1963 by a milder hydrolysis treatment using acetic acid (14Osborn M.J. Studies on the Gram-negative cell wall. I. Evidence for the role of 2-keto-3-deoxyoctonate in the lipopolysaccharide of Salmonella typhimurium.Proc. Natl. Acad. Sci. USA. 1963; 50: 499-506Google Scholar). Excessively long and strong hydrolytic conditions, which are sometimes necessary to release the lipid A-PS, results in some dephosphorylation and O-deacylation of lipid A (15Karibian D. Deprun C. Caroff M. Use of plasma desorption mass spectrometry in structural analysis of endotoxins: effect on lipid A of different acid treatments.in: Levin J. Alving C.R. Munford R.S. Redl H. Bacterial Endotoxins: Lipopolysaccharides for Genes to Therapy. Wiley-Liss, New York1995: 103-111Google Scholar). Such modifications strongly diminish the biological activities of the molecule. Milder hydrolysis conditions, such as pH 4.5 sodium acetate buffer, were often shown to be efficient for lipid A liberation (16Rosner M.R. Tang J. Barzilay I. Khorana H.G. Structure of the lipopolysaccharide from an Escherichia coli heptose-less mutant. I. Chemical degradations and identification of products.J. Biol. Chem. 1979; 254: 5906-5917Google Scholar) and were usually improved by adding SDS (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar) when the hydrolysis kinetics were too slow or totally ineffective.It is often necessary to check the reproducibility of bacterial LPS extractions and to characterize structural modifications having potential effects on biological activities. Such characterization may also be necessary to determine the consequences of variability in cell growth conditions (temperature, salt concentration, etc.).We recently developed new conditions for extracting rough-type endotoxins with ammonium isobutyrate in a quick and simple way (Caroff, M. patent 2004/062690 A1). The method used the solubility capacities of a solvent originally used for paper chromatography of peptidoglycan fragments or 14C-labeled LPSs (18Boman H.G. Monner D.A. Characterization of lipopolysaccharides from Escherichia coli K-12 mutants.J. Bacteriol. 1975; 121: 455-464Google Scholar). Here, we present a new lipid A microisolation procedure performed directly on bacterial cells using this solvent. The method is broadly applicable to different Gram-negative bacteria and is appropriate for the analysis of small amounts of cells.MATERIALS AND METHODSBacterial strainsBacterial strains used were Haemophilus influenzae (strain b Eagan) and the smooth-type bacteria Bordetella holmesii (ATCC 51541), both from National Research Council (Ottawa, Canada), Escherichia coli (strain C60) from Insitut de Biochimie, Biophysique Moléculaire et Cellulaire (Orsay, France), and Bordetella pertussis (strain Bp1414) from the Institut Mérieux (Lyon, France).Thin-layer chromatographyChromatography was performed on aluminum-backed silica TLC plates (Merck), and compounds were visualized by charring (145°C after spraying with 10% sulfuric acid in ethanol). Solvent 1 was a mixture of isobutyric acid-molar ammonium hydroxide [3:5 (v/v) for OSs and 5:3 (v/v) for LPS] (19Caroff M. Karibian D. Several uses for isobutyric acid-ammonium hydroxide solvent in endotoxin analysis.Appl. Environ. Microbiol. 1990; 56: 1957-1959Google Scholar). The latter was also used for kinetics when lipid A, LPS, and PSs had to be seen on the same plate. Solvent 2 used for lipid A sample migration was a mixture of chloroform-methanol-water-triethylamine (3:1.5:0.25:0.1, v/v), as described previously (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar).LPS hydrolysesAcetic acid hydrolysisThe LPSs were cleaved by hydrolysis in 1% acetic acid at 100°C for 1.5 h at a concentration of 5 mg/ml. Lipid A was recovered from the lyophilized residue by two extractions with a volume of 0.5 ml of a chloroform-methanol-water (3:2:0.25, v/v) mixture.Detergent-promoted hydrolysisThe LPSs were cleaved by hydrolysis in 20 mM Na acetic acid-sodium acetate buffer, pH 4.5, and 1% Na dodecyl sulfate at 100°C for 1 h at a concentration of 5 mg/ml, and lipid A was isolated as described previously (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar).Matrix-assisted laser desorption/ionization negative ion mass spectrometryAnalyses were performed on a PerSeptive Voyager-DE STR model time-of-flight mass spectrometer (Applied Biosystems; Institut de Biochimie, Biophysique Moléculaire et Cellulaire, IFR46, Université de Paris XI) in linear mode with delayed extraction. The ion-accelerating voltage was set at 20 kV. Dihydroxybenzoic acid (Sigma Chemical Co., St. Louis, MO) was used as a matrix. A few microliters of lipid A suspension (1 mg/ml) was desalted with a few grains of ion-exchange resin (Dowex 50W-X8; H+) either in an Eppendorf tube or as a single droplet onto a piece of Parafilm® for small samples. A 1 μl aliquot of the suspension (50–100 μl) was deposited on the target and covered with the same amount of the matrix suspended at 10 mg/ml in a 0.1 M solution of citric acid (20Thérisod H. Labas V. Caroff M. Direct microextraction and analysis of rough-type lipopolysaccharides by combined thin-layer chromatography and MALDI mass spectrometry.Anal. Chem. 2001; 73: 3804-3807Google Scholar). Different ratios between the samples and dihydroxybenzoic acid were tested when necessary. B. pertussis or E. coli lipids A were used as external standards.Cell washThis procedure was advisable with E. coli but not B. pertussis lipid A, so it should be used only when a high content of phospholipid contaminants appears in the matrix-assisted laser desorption/ionization (MALDI) spectra. To remove most of the membrane phospholipids, lyophilized bacterial cells (10 mg) were washed twice with 400 μl of a fresh, single-phase mixture of chloroform-methanol (1:2, v/v) and once with 400 μl of chloroform-methanol-water (3:2:0.25, v/v). The insoluble material, corresponding to washed cells, was recovered by centrifugation in the pellet, and the supernatants were discarded.Lipid A isolation from whole cellsLyophilized crude or freshly washed cells (10 mg) were suspended in 400 μl of isobutyric acid-ammonium hydroxide 1M (5:3, v/v) and were kept for 2 h at 100°C in a screw-cap test tube under magnetic stirring. The mixture was cooled in ice water and centrifuged (2,000 g for 15 min). The supernatant was diluted with water (1:1, v/v) and lyophilized. The sample was then washed twice with 400 μl of methanol and centrifuged (2,000 g for 15 min). Finally, the insoluble lipid A was solubilized and extracted once in 100–200 μl of a mixture of chloroform-methanol-water (3:1.5:0.25, v/v). For 1 mg samples, 100 μl of the different solvent mixtures were used at each step.Monitoring the kinetics of lipid A release by TLCAfter repeating the above procedure up to the methanol wash and centrifugation, the residues were suspended in water or small volumes of isobutyric acid-ammonium hydroxide (1M) (5:3, v/v), deposited on the plate, chromatographed in the same mixture of solvents, dried, and charred (19Caroff M. Karibian D. Several uses for isobutyric acid-ammonium hydroxide solvent in endotoxin analysis.Appl. Environ. Microbiol. 1990; 56: 1957-1959Google Scholar).RESULTS AND DISCUSSIONThe need for a rapid method to check the variability of lipid A structures with growth conditions of bacteria initiated our search for new isolation methods. The mild SDS promoted hydrolysis at pH 4.5 (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar) and had been applied directly on bacterial cells with success (21Que N.L. Lin S. Cotter R.J. Raetz C.R. Purification and mass spectrometry of six lipid A species from the bacterial endosymbiont Rhizobium etli. Demonstration of a conserved distal unit and a variable proximal portion.J. Biol. Chem. 2000; 275: 28006-28016Google Scholar). However, this method was too time-consuming for our purpose.The extraction of LPS with methanol as described in the literature was limited to deep-rough LPSs (11Nurminen M. Vaara M. Methanol extracts LPS from deep rough bacteria.Biochem. Biophys. Res. Commun. 1996; 219: 441-444Google Scholar). The tri-reagent extraction method (13Yi C.E. Hackett M. Rapid isolation method for lipopolysaccharide and lipid A from Gram-negative bacteria.Analyst. 2000; 125: 651-656Google Scholar) used for all types of LPS was also time-consuming and did not avoid the use of phenol. In this work, we examined the capacity of ammonium-isobutyrate to simultaneously extract and hydrolyze the endotoxin directly from bacterial cells.Different conditions of concentrations, temperatures (40, 60, 80, and 100°C), time (30 min, 1 h, 1.5 h, 2 h, 3 h, and 4 h), and mixtures of solvents [isobutyric acid-water (1:1), isobutyric acid: (1M and 2 M)-ammonium hydroxide (5:3), and pure isobutyric acid] were tested. Splitting kinetics were visualized by thin-layer chromatography (19Caroff M. Karibian D. Several uses for isobutyric acid-ammonium hydroxide solvent in endotoxin analysis.Appl. Environ. Microbiol. 1990; 56: 1957-1959Google Scholar), followed by more accurate analysis by MALDI mass spectrometry (20Thérisod H. Labas V. Caroff M. Direct microextraction and analysis of rough-type lipopolysaccharides by combined thin-layer chromatography and MALDI mass spectrometry.Anal. Chem. 2001; 73: 3804-3807Google Scholar).The best results were obtained with the 5:3 (M) mixture for the endotoxins tested. Good splitting results were obtained at 100°C in a few hours without degradation of LPS components. If any dephosphorylation occurred during the testing of the conditions, it was indicated in the spectra by the appearance of peaks at m/z −80 Da, and the duration of hydrolysis was reduced. At lower temperatures, no or incomplete splitting was observed. It is highly recommended that the proper mixture of solvents and hydrolysis conditions be determined for each bacterium.The isobutyrate TLC solvent system 1 (19Caroff M. Karibian D. Several uses for isobutyric acid-ammonium hydroxide solvent in endotoxin analysis.Appl. Environ. Microbiol. 1990; 56: 1957-1959Google Scholar) allowed quick visualization of the uncleaved native B. pertussis endotoxin and of the released OS and lipid A moieties as a function of time. As expected and shown in Fig. 1, the LPS spots decreased progressively as those of OS and lipid A increased, giving a first indication of the splitting rate. This easy test allowed a preselection of the hydrolytic conditions. Lipid A is rarely composed of a single molecular species: in the case of the B. pertussis cells used in this laboratory, it consists of two to three molecular species, as seen in the spectra (22Caroff M. Deprun C. Richards J.C. Karibian D. Structural characterization of the lipid A of Bordetella pertussis 1414 endotoxin.J. Bacteriol. 1994; 176: 5156-5159Google Scholar). Therefore, on thin-layer chromatography, the lipid A preparation gives two or three spots, more or less discrete. There is also a slight difference between the migration of pure isolated lipid A and that of lipid A liberated in the solvent system, because of the presence of salts in the latter. The charring method is also less sensitive for lipid A than for LPS or PS because it has fewer hydroxyl groups.This experiment was followed by extraction of the lipid A with a mixture of solvents (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar) and TLC of the concentrated extracts in solvent system 2, as well as MS analyses. These data confirmed the choice of 2 h of hydrolysis at 100°C in the solvent mixture isobutyric acid-ammonium hydroxide (1M) (5:3).Figure 2shows a comparison by TLC of B. pertussis lipid A thus obtained (Fig. 2A), with lipids A obtained by hydrolysis of the B. pertussis endotoxin by the SDS-pH 4.5 method (Fig. 2B) and by the 1% acetic acid method for 1.5 h at 100°C (Fig. 2C). Less degradation was observed with the methods shown in Fig. 2A, B, as shown by the absence of fast-migrating products corresponding to dephosphorylated lipid A molecular species. These results were confirmed by MALDI analyses of the lipid A samples. The spectrum shown in Fig. 2A was obtained by the present method and shows two peaks corresponding to the migrating molecular species of the penta-acyl and tetra-acyl B. pertussis lipid A (1,558Da and 1,332Da, respectively). The presence of these two molecular species in lipid A obtained by mild hydrolysis has already been described for this hypoacylated lipid A (22Caroff M. Deprun C. Richards J.C. Karibian D. Structural characterization of the lipid A of Bordetella pertussis 1414 endotoxin.J. Bacteriol. 1994; 176: 5156-5159Google Scholar). It is very similar to the spectrum shown in Fig. 2B obtained for lipid A after the mild SDS-pH 4.5 hydrolysis (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar). The lipid obtained by hydrolysis in acetic acid showed some degradation (Fig. 2C), especially by the loss of phosphate groups from the two main molecular species with peaks at m/z 1,252 and 1,478. The presence of the glycosidic phosphate in MALDI spectra was a good indicator of structure integrity, the latter being the more labile component of such molecules (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar).Fig. 2Negative-ion matrix-assisted laser desorption/ionization (MALDI) mass spectrometry and TLC comparison of B. pertussis lipid A obtained by different methods. A: Hydrolysis of B. pertussis cells by isobutyric acid-molar ammonium hydroxide (5:3, v/v) for 2 h at 100°C. B: Hydrolysis of B. pertussis LPS by SDS-pH 4.5 for 1 h at 100°C. C: Hydrolysis of B. pertussis LPS by 1% acetic acid for 1.5 h at 100°C. Molecular species visualized by TLC were labeled with the corresponding masses obtained in MALDI.View Large Image Figure ViewerDownload (PPT)Different rough-type bacteria of known lipid A structure were tested: B. pertussis (22Caroff M. Deprun C. Richards J.C. Karibian D. Structural characterization of the lipid A of Bordetella pertussis 1414 endotoxin.J. Bacteriol. 1994; 176: 5156-5159Google Scholar), H. influenzae (23Helander I.M. Lindner B. Brade H. Altmann K. Lindberg A.A. Rietschel E.T. Zähringer U. Chemical structure of the lipopolysaccharide of Haemophilus influenzae strain I-69 Rd−/b+. Description of a novel deep-rough chemotype.Eur. J. Biochem. 1988; 177: 483-492Google Scholar), and E. coli (24Karibian D. Deprun C. Caroff M. Comparison of lipids A of several Salmonella and Escherichia strains by 252Cf plasma desorption mass spectrometry.J. Bacteriol. 1993; 175: 2988-2993Google Scholar). MALDI spectra in the negative-ion mode demonstrated that no structural elements were lost upon hydrolysis. Each spectrum obtained with the new method was compared with the corresponding spectrum obtained after SDS-promoted hydrolysis of the phenol-extracted endotoxins. They were found to give similar results in terms of lipid A structural integrity.Dephosphorylation and O-deacylation of lipid A (25Caroff M. Karibian D. Haeffner-Cavaillon N. Cavaillon J-M. Structural and functional analyses of bacterial lipopolysaccharides.Microbes Infect. 2002; 4: 915-926Google Scholar) are modifications that strongly diminish the biological activities of lipid A, as shown, for example, by the loss of toxicity and pyrogenicity of the B. pertussis lipid A after release of the glycosidic phosphate (26Caroff M. Cavaillon J-M. Fitting C. Haeffner-Cavaillon N. Inability of pyrogenic, purified Bordetella pertussis lipid A to induce interleukin-1 release by human monocytes.Infect. Immun. 1986; 54: 465-471Google Scholar). The use of SDS to disrupt micellar LPS suspensions during hydrolysis (17Caroff M. Tacken A. Szabo L. Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin.Carbohydr. Res. 1988; 175: 273-282Google Scholar) proved to be effective in favoring hydrolysis of the ketosidic bond of Kdo at pH 4.5. In the present experiments, the solvent used for extraction of endotoxins combines the effect of solubilization of the endotoxin and gives a final pH of 4.2–4.3, somewhat lower than the conventional pH of acetic acid-sodium acetate buffer, but a weaker acid that is ideal for preserving linkages involving more labile structural elements. It is recommended that the conditions to be used on the same or similar types of cells be tested before their use in a routine procedure. In some cases, partial dephosphorylation can help with the structural determination of unknown lipids A, a more common difference of 80Da suggesting the loss of a glycosidic phosphate. Differences corresponding with the molecular weight of other compounds such as sugars can be expected with other structures (4Caroff M. Karibian D. Structure of bacterial lipopolysaccharides.Carbohydr. Res. 2003; 338: 2431-2447Google Scholar). Thus, as already mentioned, precise experimental conditions have to be defined for each particular purpose and bacterium.Spectra obtained for the three tested bacteria under the new conditions of hydrolysis are presented in Fig. 3. Figure 3A shows the spectrum obtained for E. coli lipid A (24Karibian D. Deprun C. Caroff M. Comparison of lipids A of several Salmonella and Escherichia strains by 252Cf plasma desorption mass spectrometry.J. Bacteriol. 1993; 175: 2988-2993Google Scholar) under the new conditions with the classical hexa-acylated molecular species at m/z 1,797 followed by penta- and tetra-acylated molecular species at m/z 1,570 and 1,361, respectively. For B. pertussis lipid A (Fig. 3B), two main peaks are present corresponding to the tetra- and penta-acylated molecular species at m/z 1,332 and 1,558, respectively (22Caroff M. Deprun C. Richards J.C. Karibian D. Structural characterization of the lipid A of Bordetella pertussis 1414 endotoxin.J. Bacteriol. 1994; 176: 5156-5159Google Scholar, 27Caroff M. Brisson J-R. Martin A. Karibian D. Structure of the Bordetella pertussis 1414 endotoxin.FEBS Lett. 2000; 477: 8-14Google Scholar). Some minor peaks correspond to the presence of fatty acids with two additional carbons. H. influenzae lipid A (23Helander I.M. Lindner B. Brade H. Altmann K. Lindberg A.A. Rietschel E.T. Zähringer U. Chemical structure of the lipopolysaccharide of Haemophilus influenzae strain I-69 Rd−/b+. Description of a novel deep-rough chemotype.Eur. J. Biochem. 1988; 177: 483-492Google Scholar) extracted under the new conditions gave the spectrum shown in Fig. 3C. Three main peaks are observed at m/z 1,824, 1,599, and 1,388, corresponding, respectively, to hexa-, penta-, and tetra-acylated molecular species. Small peaks at m/z 1,308 and 1,744 are attributable to a slight dephosphorylation of the main molecular species. The peak at m/z 1,163 present in the three spectra corresponds to a triacylated molecular species.Fig. 3Negative-ion MALDI mass spectra and structures of the main molecular species present in E. coli (A), B. pertussis (B), and H. influenzae (C) lipids A obtained after hydrolysis of the cells by isobutyric acid-molar ammonium hydroxide (5:3, v/v) for 2 h at 100°C. MW, molecular weight.View Large Image Figure ViewerDownload (PPT)The sensitivity of the method was tested on E. coli cells and found to be reproducible and to give good quality spectra on samples of lyophilized cells as small as 250 μg. In this case, the concentrations were decreased to facilitate handling of the samples (see Materials and Methods). No variations in the results were observed with the new concentrations. The limit of detection was found to be 50–100 μg depending on the cells tested.The experimental conditions used with the above bacteria having known structures were then tested on cells of a recently isolated smooth-type Bordetella species: B. holmesii (28Weyant R.S. Hollis D.G. Weaver R.E. Amin M.F. Steigerwalt A.G. O'Connor S.P. Whitney A.M. Daneshvar M.I. Moss C.W. Brenner D.J. Bordetella holmesii sp. nov., a new gram-negative species associated with septicemia.J. Clin. Microbiol. 1995; 33: 1-7Google Scholar). Most of the Bordetella species examined to date have displayed different lipid A structures. The B. holmesii lipid A spectrum isolated by the new conditions was compared with that of B. pertussis, the “prototype” of the genus that is responsible for whooping cough, and to those of other Bordetella species. The spectrum revealed structural differences from that of the B. pertussis lipid A. The heterogeneous spectrum, however, did give mass peaks corresponding to molecular species found in other Bordetella species, such as B. bronchiseptica (29Zarrouk H. Karibian D. Bodié S. Perry M.B. Richards J.C. Caroff M. Structural characterization of the lipids A of three Bordetella bronchiseptica strains: variability of fatty acid substitution.J. Bacteriol. 1997; 179: 3756-3760Google Scholar) and B. hinzii (30Aussel L. Brisson J.R. Perry M.B. Caroff M. Structure of the lipid A of Bordetella hinzii ATCC 51730.Rapid Commun. Mass Spectrom. 2000; 14: 595-599Google Scholar). The detailed structure will be described in a future publication.Comparisons of lipid A structures are essential when they are associated with and related to different biological activities or bacterial resistance. Good examples are the presence or absence of a single fatty acid (31Tanamoto K. Azumi S. Salmonella-type heptaacylated lipid A is inactive and acts as an antagonist of lipopolysaccharide action on human line cells.J. Immunol. 2000; 164: 3149-3156Google Scholar, 32Raetz C.R.H. Regulated covalent modifications of lipid A.J. Endotoxin Res. 2001; 7: 73-78Google Scholar, 33Preston A. Maxim E. Toland E. Pishko E.J. Caroff M. Maskell D.J. Bordetella bronchiseptica PagP is a Bvg-regulated lipid A palmitoyl transferase that is required for persistent colonization of the mouse respiratory tract.Mol. Microbiol. 2003; 48: 725-736Google Scholar) or of substituents on the C-1 and/or C-4′ phosphates (34Gunn J.S. Bacterial modification of LPS and resistance to antimicrobial peptides.J. Endotoxin Res. 2001; 7: 57-62Google Scholar). Such structural differences could b