WO2014035270A1

WO2014035270A1 - A method of identifying gram-negative bacteria using the maldi-tof ms method

Info

Publication number: WO2014035270A1
Application number: PCT/PL2013/000112
Authority: WO
Inventors: Sabina KOJ; Tomasz Niedziela; Czesław ŁUGOWSKI
Original assignee: Wrocławskie Centrum Badań Eit + Spółka Z Organiczona
Priority date: 2012-08-31
Filing date: 2013-08-30
Publication date: 2014-03-06
Also published as: PL223203B1; PL400598A1; WO2014035270A8

Abstract

The present invention relates to the identification of lipopolysaccharide chemotypes of Gram-negative bacteria based on the endotoxin profile of whole bacteria using MALDI-TOF MS. The profile derived from lipopolysaccharides (LPS) for bacterial preparation is obtained in the negative mode of a MALDI-TOF spectrometer using DHB as a matrix. The subject invention is useful in the analysis of clinical samples, diagnosis of infections caused by Gram-negative bacteria and can be useful as a tool in the therapy directed against the O-antigen using specific antibodies.

Description

A method of identifying Gram-negative bacteria using the MALDI-TOF MS method

The subject of the present invention is a method of identifying Gram-negative bacteria that uses the MALDI-TOF MS mass spectroscopy method.

The O-antigen, as one of the surface antigens of Gram-negative bacteria is used to differentiate and identify bacteria [1]. The O-antigen, consisting of many repeating subunits constitutes the saccharide portion of lipopolysaccharide (LPS), which covers about 70% of the bacterial cell surface. Due to the biological activity of LPS, it is defined as an endotoxin. During an infection, LPS is released from the surface of the bacterial cell into the bloodstream, leading to a systemic inflammatory reaction.

The O-antigen exhibits a large variability due to the presence of many sugar residues, their arrangement in the subunit as well as connections within and between the subunits, which makes this the most variable cell component [2]. The structural variability of this antigen makes it unique for a given bacterial strain. At the same time, being a highly immunogenic component exposed on the bacterial cell surface, the O-antigen is a basis for the serotyping of Gram- negative bacteria. Based on the affinity analysis of antibodies against O- antigens, serotypes have been assigned to bacterial strains. For example, among 180 serotypes of Escherichia coli, the O-antigen is the main differentiating antigen. Within this species, there are 170 O-antigens, about 80 K antigens and about 50 H antigens. Antibodies generated in response to the O-antigen are widely used in the diagnostics of Gram-negative infections. In the case of Gram-negative bacteria with an LPS lacking the O-antigen, serological specificity is determined by the core oligosaccharide (core OS, R antigen). Termed a lipooligosaccharide (LOS, R-LPS), this component is commonly found among mucosal pathogens, i.e. Neisseria meningitidis [3]. Aside from the saccharide portion, the composition of lipopolysaccharides includes a phospholipid portion (lipid A) that anchors LPS in the bacterial cell membrane. Thus, three parts may be differentiated in the composition of LPS: the O-specific chain, core oligosaccharide and lipid A. The complete structure of isolated LPS of various bacterial strains is determined using a combination of structural analysis methods such as nuclear magnetic resonance and mass spectrometry of complete LPS molecules (HR-MAS NMR, MS), deacylated LPS (NMR, MS and GC) and fragments of its oligosaccharide (NMR, MS and GC). In this way, data has been made available on the structure of bacterial LPS, and its structural variability among bacterial strains [3-8].

A few publications are known in which spectra of isolated LPS have been obtained using 2,4,6-THAP as a matrix and in negative mode of matrix- assisted laser desorption/ionisation time of flight mass spectrometry (MALDI- TOF MS). The MALDI-TOF MS spectra of unmodified LPS have been obtained for Gram-negative bacteria with R-LPS i.e. Erwinia carotovora and Shewanella pacifica [7,9]. With optimised measurement conditions using this method, isolated LPS spectra have been observed to exhibit groups of signals derived from the heterogeneity of LPS molecules. The heterogeneity of LPS molecules is usually limited to saccharides and phosphates. Signal groups in MALDI-TOF MS spectra constitute unique profiles of lipopolysaccharides. Due to the limited detection scope of isolated LPS MALDI-TOF MS spectra, ions from the repeating O-specific chain subunits are not visible. Components of bacterial cells such as lipopolysaccharides and oligosaccharides are being evaluated as chemotaxonomic biomarkers specific for bacterial strains in order to identify microorganisms.

Prior art includes the known biotyping of bacteria using MALDI-TOF MS based on the generation of MALDI-TOF MS protein profiles of bacteria or protein extracts thereof (US6177266, US7684934, US8155892). The protein profile consists of a group of signals from proteins observed in MALDI-TOF MS spectra for bacterial sample and their extracts using positive-mode MALDI-TOF MS with HCCA as a matrix. A MALDI-TOF MS analysis is performed on whole bacteria or protein extracts obtained following lysis or on cell component fractions. Such profiles are usually obtained from proteins released from the cytosol during the preparation of the bacterial sample for MALDI-TOF MS analysis (Ryzhov V., ^"FeTisela rCTr''Characterization of the Protein Subset Desorbed by MALDI from Whole Bacterial Cells." Analytical Chemistry, 73 (2001): 746-750).

According to the instructions "MALDI-TOF Fingerprinting of bacteria" by Bruker Daltonics, there is a procedure of identifying bacteria using a reference base of MALDI-TOF MS protein profiles directly for bacteria. MALDI-TOF MS based on soft ionisation makes it possible to analyze peptides and proteins desorbed from whole microorganism. Appropriate ions are separated and detected based on molecular mass and charge. The resulting spectrum is composed of signals in the range from 2000 to 20000 m/z. Such an analysis is usually performed on an isolated bacterial colony from a clinical sample cultured on an appropriate growth medium and incubated at an appropriate temperature. A small amount of biological material is transferred directly onto a MALDI plate, alpha-cyano-4- hydroxy-cinammic acid (HCCA) is added to the sample. Next, the sample is analyzed on a MALDI-TOF spectrometer in linear positive mode. The identification makes use of a mass range of 2000 to 20000 Daltons. In order to identify unknown microorganisms, the program makes pattern matching analyses and compares the generated list of signals with a library of reference spectra, which contains characteristic spectral information. The identification in the described method uses the so-called "fingerprinting" which in terms of bacteria constitutes a protein pattern facilitating strain differentiation.

A MALDI-TOF MS analysis of protein profiles directly from bacterial preparations makes it possible to obtain information on the antigens of the unknown bacteria. However, in addition to the MALDI-TOF MS protein profile for a bacterial mass there is an extant need to deliver additional information regarding bacterial cells and their unique biomarkers, which may be used in the biotyping of microorganisms and in the design of antibacterial therapeutic tools. Information regarding the surface antigens of bacterial cells (such as the O- antigen) in the preparation of an unknown bacterial colony will enable the facile and precise identification of a bacterial strain since surface antigens constitute specific chemotaxonomic biomarkers of bacterial strains. Additionally, the O- antigen qualifies as a useful biomarker for identifying bacteria due to its considerable chemical and thermal stability. An analysis of the surface antigens of whole bacteria also makes it possible to collect information essential to the design-of-a therapy targ^te l^{^}a^ins the O-antigen. The therapy may be based on antibodies or sera directed against the O-antigen prepared in the form of a conjugate with a carrier protein.

Unexpectedly, this goal was realized by the present invention.

The subject of the present invention is a method of identifying Gram-negative bacteria using the MALDI-TOF MS method, according to which

a) a bacterial preparation is made for analysis using the MALDI-TOF MS method

b) the bacterial preparation is analyzed using the negative mode of

MALDI-TOF MS in the range of signals from ions of LPS molecules found on the bacterial cell surface

c) an endotoxin spectral profile is obtained of whole bacteria characteristic for a given bacterial strain

d) the resulting spectral endotoxin profile of the bacterial preparation is compared to a reference database containing MALDI-TOF MS spectral profiles of bacterial strains.

In a preferable embodiment of the present invention, the material for obtaining the preparation is a bacterial colony from a culture medium or the precipitate following the centrifugation of a bacteria from a liquid medium.

Preferably, the bacterial preparation is obtained as a result of the inactivation and extraction of bacteria with formic acid.

In another preferable embodiment of the present invention, the bacterial preparation is obtained as a result of the treatment of bacteria with a proteolytic enzyme.

Preferably, in order to obtain an endotoxin spectral profile a benzoic acid derivative is used as a matrix, preferably 2,4-dihydroxybenzoic acid (DHB).

Preferably, the analysis is performed on signals corresponding to LPS-derived core oligosaccharides, O-antigen oligosaccharides, lipid A or full-lenght LPS. Preferably, the MALDI-TOF MS spectra are analysed in the m/z range from 700 to 50000.

Preferably, the MALDI-TOF MS spectra are analysed in the m/z range from 700 to 4000. The subject invention relates to a method of identifying strains of Gram-negative bactefia^"based on the spectral endotoxin profile of whole bacteria using MALDI- TOF MS. The endotoxin profile for bacterial preparations is obtained in the negative mode of a MALDI-TOF MS using DHB as a matrix. The method encompasses the preparation of a bacterial sample for MALDI-TOF MS analysis, obtaining unique MALDI-TOF endotoxin profiles for these preparations and the identification of bacterial strains by comparing the resulting spectra with a reference database of MALDI-TOF MS endotoxin profiles.

A bacterial sample for biotyping using the method according to the present invention is a bacterial preparation. The preparation may be a bacterial colony from a culture medium or the precipitate following the centrifugation of a bacteria from a liquid medium. Two procedures of preparing the bacterial sample are set out in the present invention: one, due to the pathogenicity of bacteria, encompasses stages of inactivation and extraction, whereas the second encompasses the preparation of the bacterial mass using treatment with a proteolytic enzyme. In the case of pathogenic bacteria the preparation of a bacterial mass is based on the inactivation of bacteria with 80% ethanol, and subsequently extraction using 70% formic acid and an equal volume of acetonitrile. The MALDI-TOF MS analysis of endotoxin profiles using the method according to the present invention makes use of bacterial masses separated from the extract. The preparation procedure of a mass for obtaining an endotoxin profile comprises a modified procedure according to the usage note: MT-80 "Microorganism identification and classification based on MALDI- TOF MS fingerprinting with MALDI-Biotyper" Bruker Daltonics (Fig. 1).

The preparation of an enterobacterial mass MALDI-TOF MS analysis is based on the treatment of a bacterial preparation with a proteolytic enzyme. The enzymatic preparation of an enterobacterial mass makes it possible to remove molecules that generate ions "disruptive" to MALDI-TOF MS spectra and to the observation of endotoxin profiles (Fig. 5 and 6). MALDI-TOF analysis of endotoxin profiles using the method according to the present invention makes use of bacterial preparation purified of the enzyme and salts prior to the analysis. Next, a method according to the present invention is based on the addition of benzoic^~acid^"defivatives as a matrix to the bacterial sample prior to the MALDI- TOF MS analysis. Preferably, the benzoic acid derivative comprises 2,4- dihydroxybenzoic acid (DHB).

A small amount of sample, most preferably 1-2 μΙ, is used for MALDI-TOF MS analysis. A method according to the present invention is based on the acquisition of MALDI-TOF MS spectra in the negative mode for an m/z range encompassing ions corresponding to LPS-derived oligosaccharide molecules. The range of masses usually encompasses a m/z from 700 to 4000. The MALDI-TOF MS spectra are averaged of 100-200 laser shots, collected and compared to reference MALDI-TOF MS spectra of whole bacteria in order to identify unknown bacterial samples.

A method according to the present invention encompasses a reference database of MALDI-TOF MS endotoxin profiles. Signal series in reference MALDI-TOF MS spectra constitute unique profiles obtained for bacteria preparations (Fig. 2 and Fig. 5) and confirmed as corresponding to lipopolysaccharide molecules by comparing z MALDI-TOF MS profiles of isolated LPS (Fig. 3 and 7). In MALDI-TOF MS spectra of isolated LPS, we observe ions corresponding to lipid A, core oligosaccharide (core OS) and full- lenght LPS. We have also observed that the endotoxin profiles of the analysed Gram-negative bacteria (profile of whole bacteria) are consistent with the isolated LPS profiles in the mass range for core OS. The acquired endotoxin profile of isolated LPS and of whole bacteria consists of many spectral signals due to the heterogeneity of LPS-derived core oligosaccharide molecules. The heterogeneity of core OS molecules is due to the non-stoichiometric substitution with phosphate groups, phosphoethanolamine and sugar residues. Signals in an endotoxin profile are usually obtained in the m/z range from 700 to 4000. In MALDI-TOF MS spectra of the analysed LPS and preparations of bacteria with a smooth-type LPS (LPS containing the O-antigen) we observed only ions corresponding to the core oligosaccharide. We observed no ions corresponding to the repeating O-specific chain subunits. In the case of bacteria possessing S- type LPS, the differentiation of strains is also possible based on signals corresponding to the core OS which are observed also in MALDI-TOF MS spectra of LPS (Fig. 9).

In the present invention, the identification of Gram-negative bacteria encompasses the stage of comparing the endotoxin profiles of MALDI-TOF MS spectra from bacterial preparations with a database containing characteristic MALDI-TOF MS spectra of bacterial strains, which makes it possible to differentiate strains differing in terms of their LPS-derived cores. The resulting endotoxin profiles according to the present invention encompass signals corresponding to LPS-derived core oligosaccharides. The core oligosaccharide is one of the LPS components that determines specificity of Gram-negative bacteria. For example, among Enterobacteriaceae, the R1-type structure is dominant among strains causing intestinal infections, whereas an OS of R3 type is dominant among isolates producing verotoxin (i.e. E. coli O157:H7) [3]. A MALDI-TOF MS analysis of endotoxin profiles makes it possible to differentiate Enterobacteriaceae possessing LPS differing in the structure of the core OS. As described above, a method according to the present invention is based on the generation of unique spectral mass profiles of whole bacteria using the negative mode of MALDI-TOF MS. This method makes it possible to observe unique components of oligosaccharides within bacterial antigens and to obtain information on their profiles. The subject invention makes it possible to evaluate specific biomarkers with an oligosaccharide structure, which differentiate the bacteria of various species and strains, facilitating the effective identification of the lipopolysaccharide chemotypes of microorganisms. The subject solution can be used in the analysis of clinical samples, diagnosis of infections caused by Gram-negative bacteria and also serve as a tool for therapy directed against the O-antigen via specific antibodies.

The invention is illustrated by the following figures.

Fig. 1. Preparation procedure of the Gram-negative bacterial sample for the MALDI-TOF MS analysis of its endotoxin profile (modified sample preparation protocol for MALDI-TOF MS analysis according to the MT80 instruction, Bruker Daltonics), facilitating the direct observation of LPS/LOS.

Fig. 2. The endotoxin profile of Bordetella pertussis cells. The MALDI-TOF MS spectrum of B. pertussis using DHB (in ACN/ 0,1 % TFA at a ratio of 30/70) as a matrix. The spectrum was recorded using an Autoflex III instrument from BrukerDaltonicsT

Fig. 3. MALDI-TOF MS spectrum of the LOS of Bordetella pertussis 186 recorded using the negative mode and using DHB (A) linear mode, (B) reflectron mode; where: Hep - heptose, PPEtN - pyrophosphoethanolamine residue, P - phosphate group, Ac - acetyl group, H₂O - water molecule, C14OH - fatty acid residue. The spectrum was recorded using an Autoflex ill instrument from Bruker Daltonics.

Fig. 4. Interpretation of the ions corresponding to molecules observed in the MALDI-TOF MS spectra of B. pertussis 186 LOS or OS.

Fig. 5. MALDI-TOF MS spectrum obtained directly from an enterobacterial preparation (preparation of Salmonella without enzymatic digestion). The spectrum was recorded using an Autoflex III instrument from Bruker Daltonics. Fig. 6. MALDI-TOF MS spectrum obtained for a bacterial preparation of Salmonella following digestion with a proteolytic enzyme. The spectrum was recorded using an Autoflex III instrument from Bruker Daltonics.

Fig. 7. MALDI-TOF MS spectrum of Salmonella LPS after digestion with a proteolytic enzyme. The spectrum was recorded using an Autoflex III instrument from Bruker Daltonics.

Fig. 8. The MALDI-TOF MS spectra of whole E. coli cells. Comparison of the endotoxin profiles of E. coli differing in their type of core OS: R1 and R3. The spectrum was recorded using an Autoflex III instrument from Bruker Daltonics. Fig. 9. The MALDI-TOF MS spectra LPS E. coli with a S-type LPS in the m/z range from 1600 to 2100, demonstrating signals corresponding to the core OS and lipid A. The MALDI-TOF MS spectra of LPS were performed for E. coli serotypes: O5, O6, 039, 064 and O111. The spectra were recorded using an Autoflex III instrument from Bruker Daltonics.

The subject of the present invention is shown in example embodiments that do not limit the scope of its protection. Example 1. The MALDI-TOF MS endotoxin profile of Bordetella pertussis

Colonies of Bordetella pertussis were collected from a culture medium and inactivated with 80% ethanol, and subsequently extracted with formic acid while using the bacterial mass for MALDI-TOF MS analysis according to Fig. 1. The extraction is made by using 70% formic acid and an equal volume of acetonitrile at the stage of precipitation by centrifugation (modified sample preparation procedure for MALDI-TOF MS analysis according to the instruction note MT80, Bruker Daltonics). The mass of B. pertussis was mixed with DHB and a MALDI- TOF MS spectrum was recorded of whole bacteria (Fig. 2). This spectrum was obtained in the negative ion mode of the Autoflex III MALDI-TOF spectrometer, Bruker Daltonics. The origin of profiles from LPS molecules was confirmed by comparing with the MALDI-TOF MS spectra of isolated LPS (Fig. 3).

We obtained the MALDI-TOF MS endotoxin profile of B. pertussis consistent with the MALDI-TOF MS profile of isolated LPS. Signals in the m/z range from 2200 to 2600 correspond to ions originating from molecules of the LPS-derived core oligosaccharide (Fig. 4). For core OS, we observed ions corresponding to anhydrododecasaccharide (m/z 2291 ,63) and its pyrophosphorylated form, and substituted with a pyrophosphoethanolamine molecule (m/z 2452,57 and 2497,13). The most intensive ion (m/z 2231 ,43) indicates the loss of a neutral CO₂ molecule (-44 Da) from the anhydrododecasaccharide. We also observed several glycoforms lacking a terminal heptose identified among the ions with m/z of 2231 and 2039, 2274 and 2028, 2452 and 2260. In the spectrum of isolated LOS, signals in the lower mass range correspond to lipid A (m/z 1558,95), whereas in the higher range we observed ions originating from the full-lenght lipooligosaccharide (m/z 4056,65). The observation of ions corresponding to the lipid A molecule in the preparation of the bacteria may derive from the presence of LOS that has dissociated from the bacterial cell. A comparison of the endotoxin profiles of B. pertussis cells with the LOS molecule profile of B. pertussis makes it possible to identify the bacterial colony as B. pertussis. The resulting endotoxin profile of whole bacterial cells constitutes the reference spectrum in the database and serves as a reference for the analysis of unknown bacterial colonies.

Example 2. MALDI-TOF MS endotoxin profiles of enterobacteria In the case of evaluated enterobacteria, we observe the presence of ions "disruptive"-to^_MALDI=TOF^~MS spectra of whole bacteria in the m/z range, corresponding to the LPS molecule (Fig. 5). The preparation of bacterial preparations for MALDI-TOF MS analysis is based on the treatment of isolated enterobacterial preparations with a proteolytic enzyme, such as Proteinase K. An enzymatic preparation of an enterobacterial mass makes it possible to remove molecules which generate "disruptive" ions and enables the observation of endotoxin profiles (Fig. 6).

Enterobacterial colonies were collected from a culture medium using a loop and suspended in PBS, and subsequently digested with Proteinase K for 2 days at 4°C with mixing (20 \ig of enzyme per about 10 mg of the bacterial mass). After about 48h the suspension was centrifuged (35000xg, 20'), the supernatant was discarded, and the precipitate was rinsed with distilled water. The enterobacterial mass was lyophilized and used for the analysis using the negative mode of MALDI-TOF MS after mixing with DHB as the matrix. In the spectra of enterobacterial cells, we observed signals consistent with the signals of MALDI-TOF MS spectra of isolated LPS (Fig. 7). The resulting endotoxin profile of whole bacterial cells constitutes a reference spectrum in the database and is useful as a reference profile for analyses of unknown bacterial colonies.

Example 3. The identification of strains of E. coli differing by R1 and R3 type core

Colonies of enterobacteria were collected from a culture medium using a loop and suspended in PBS, and subsequently digested with Proteinase K for 2 days at 4°C with mixing (20 g of enzyme per about 10 mg of the bacterial mass). After about 48h the suspension was centrifuged (35000xg, 20'), supernatant was discarded, and the precipitate was rinsed with distilled water. The enterobacterial preparation was lyophilised and used for analysis using the negative mode of MALDI-TOF MS. In the spectra of enterobacteria, we observed signals consistent with the signals of MALDI-TOF MS spectra of isolated LPS. A comparison of the MALDI-TOF MS spectra in the m/z range correspond to ions corresponding to LPS makes it possible to differentiate types of core OS, i.e. R1 and R3 (Fig. 8). We also observed significant differences in profiles derive from the heterogeneity of the core OS in terms of saccharide residues and acetyl groups. A comparison of these spectra facilitates the identification of bacterial strains. The resulting endotoxin profile of whole bacteria constitutes a reference spectrum in the database and serves as a reference profile for the analysis of unknown colonies.

Example 4. Reference database of endotoxin profiles of Gram-negative bacteria possessing a S-type LPS

Lipopolysaccharides of E. coli were isolated using a water-phenol method according to Westphal. The MALDI-TOF MS spectra of isolated LPS were recorded in the negative mode MALDI-TOF MS after mixing with DHB as a matrix. In MALDI-TOF MS spectra of the analysed LPS we observed ions corresponding to the core oligosaccharide and lipid A (Fig. 9). Profiles corresponding to the core OS differ for each of the analysed bacterial strains (E coli 05, O6, 039, 064 and 0111) due to the heterogeneity of these molecules. The analysed endotoxin profiles are unique and make it possible to differentiate bacteria within the core OS. The resulting MALDI-TOF MS spectra from isolated LPS were placed in the reference database of endotoxin profiles of Gram- negative bacteria. The identification of bacteria with a S-type LPS is based on the comparison of MALDI-TOF MS spectra of bacterial preparations against spectra in the database that are based on signals corresponding to the core OS. For strains of E. coli with a common core type - R1 (E. coli 05, 06, 039, 064) we observed divergent endotoxin profiles in the MALDI-TOF MS spectra. The differences in the endotoxin profiles, despite a shared core type, may derive from the natural heterogeneity of LPS molecules. These differences are also observed in a comparison with the spectrum of E. coli 0111 which possesses a R3-type core.

Bordetella pertussis 186 was obtained from the Zakted Profilaktyki Zakazen and Zakazeii Szpitalnych Narodowego Instytutu Lekow, Warszawa. The remaining strains, Salmonella typhimurium Ra, Escherichia coli (R1 , R3, 06, 05, 064, 039) were obtained from the Polish Collection of Microorganisms, Instytut Immunologii and Terapii Doswiadczalnej PAN, Wroclaw

Applicant: Wrodawskie Centrum Badan EIT + Spoika z ograniczona. odpowiedzialnoscia.

mgr in*. Rafat Witek

Rzecznik Patentowy

References:

1. Lukasiewicz J., Lugowski C, „Biologiczna aktywnosc lipopolisacharydu" Postepy Higieny i Medycyny Doswiadczalnej, 57 (2003): 33-53.

2. Wang L., Wang Q., Reeves P. R "The variation of O antigens in Gram- negative bacteria" in Subcellular Biochemistry 53, chapter 6 (2010), 123-

152.

3. Raetz C. R, Whitfiel C. Lipopolisaccharide endotoxin" Annu Rev Biochem., 71 (2002): 635-700.

4. Banoub J. H., El Aneed A., Cohen A. M., Joly N., "Structural Investigation of Bacterial Lipopolysaccharides by Mass Spectrometry and Tandem Mass

Spectrometry." Mass Spectrometry Reviews, 29 (2010): 606-650.

5. van Baar B L., "Characterisation of Bacteria by Matrix-assisted Laser Desorption/ionisation and Electrospray Mass Spectrometry." FEMS Microbiology Reviews, 24 (2000): 193-219. 6. Caroff M., Karibian D., "Structure of Bacterial Lipopolysaccharides." Carbohydrate Research, 338 (2003): 2431-2447.

7. Harvey J. D. "Matrix-assisted laser desorption/ionization mass spectrometry of carbohydrates" Mass Spectrometry Reviews, 18 (1999): 349-451

8. Jachymek W., Niedziela T, Petersson C, Lugowski C, Czaja J., Kenne L., "Structures of the O-specific polysaccharides from Yokenella regensburgei

(Koserella trabulsii) strains PCM 2476, 2477, 2478, and 2494: high- resolution magic-angle spinning NMR investigation of the O-specific polysaccharides in native lipopolysaccharides and directly on the surface of living bacteria" Biochemistry, 38 (1999): 11788-11795. 9. Sturial L, Garozzo D., Silipo A., Lanzetta R., Parrilli M., Molinaro A., "New conditions for matrix-assisted laser desorption / ionization mass spectrometry of native bacterial R-type lipopolysaccharides" Rapid Communications in Mass Spectrometry, 19 (2005): 1829-1834.

Claims

1. A method of identifying Gram-negative bacteria using the MALDI-TOF MS method, characterised in that

a) a bacterial mass preparation is made for analysis using the MALDI-TOF MS method

b) the bacterial mass preparation is analyzed using the negative mode of MALDI-TOF MS in the range of signals corresponding to ions of LPS molecules found on the bacterial cell surface

d) the resulting spectral endotoxin profile of the bacterial mass preparation is compared to a reference database containing MALDI-TOF MS spectral profiles bacterial strains.

2. The method according to Claim 1 , characterised in that the material for producing the preparation constitutes a bacterial colony from a culture medium or a precipitate following the centrifugation of a bacterial mass from a liquid medium.

3. The method according to Claim 1 , characterised in that the bacterial mass preparation is obtained as a result of the inactivation and extraction of bacteria with formic acid.

4. The method according to Claim 1 , characterised in that the bacterial mass preparation is obtained as a result of the treatment of bacteria with a proteolytic enzyme.

5. The method according to Claim 1 , characterised in that a benzoic acid derivative is used as a matrix in order to obtain an endotoxin spectral profile for the prepared bacterial colonies as defined in w Claim 3 or 4, preferably 2,4- dihydroxybenzoic acid (DHB).

6. The method according to any of Claim 1 , characterised in that signals are analyzed corresponding to LPS-derived core oligosaccharides, O-antigen component oligosaccharides, lipid A or the complete LPS.

7. The method according to Claim 1 , characterised in that the MALDI-TOF MS spectra^~aTe analy¾e¾^~in^~the m7z range from 700 to 50000.

8. The method according to Claim 1 , characterised in that the MALDI-TOF MS spectra are analysed in the m/z range from 700 to 4000.

Applicant: Wroctawskie Centrum Badan EIT + Spotka z ograniczona.

od owiedzialnoscia.

Attorney:

mgr inz. Rafaf Witek

Rzecznik Patentowy