WO2003074727A1

WO2003074727A1 - Rapid identification of yeasts

Info

Publication number: WO2003074727A1
Application number: PCT/GB2003/000841
Authority: WO
Inventors: Richard Owen Jenkins; Richard Edward Sherburn
Original assignee: De Montfort University
Priority date: 2002-03-01
Filing date: 2003-02-28
Publication date: 2003-09-12
Also published as: GB2387439A; GB0304596D0; AU2003208458A1

Abstract

The present invention concerns a method for generating mass spectra specific for yeast. Also provided are methods of determining the identity of a yeast sample, generating a database containing mass spectra for at least one yeast, and generating a report of the identity of a yeast sample.

Description

Rapid Identification of Yeasts

The present invention concerns the rapid identification of yeasts. More specifically, the present invention relates to the methods of treatment required to obtain characteristic spectra from whole cells by mass spectrometry. The characteristic spectra generated from yeast cells using such techniques as matrix-assisted laser desorption/ionisation time-of- flight mass spectrometry (MALDI-TOF-MS) distinguish different genera, species and strains of yeast.

The wine, brewing, and baking industries rely on particular yeast strains for controlled fermentation and consistency of final products. Rapid and accurate characterisation of such strains is therefore essential to these industries.

Yeasts are currently characterised by molecular biological methods, such as amplified ribosomal DNA restriction analysis (ARDRA) (Molina et al, 1993, FEMS Microbiology Letters 208, ρ259-264, PMID 8099889) and random amplification of polymorphic DNA (RAPD) (Xufre et al, 2000, Food Technology and Biotechnology 38 (1), p53-58; Gomes et al, 2000, Food Microbiology 17 (2), ρ217-223). These methods can take up to 24 hours to complete and are labour intensive, requiring special training and involve several processing steps, any one of which can introduce errors that invalidate the assay. Accordingly, there exists a need to develop a system resulting in a more accurate and simplified method for the identification and rapid classification of yeasts. Mass spectrometry is one such system, and the technique has been employed in attempts to characterise microbes for a variety of applications, for example, in diagnosis of disease, and monitoring of bioprocess operations. MALDI-TOF-MS has been used successfully for the rapid identification of bacterial cells (Krishnamurthy and Ross, 1996, Rapid Communications in Mass Spectrometry 10, pi 992- 1996, PMID 9004531). In this procedure, whole bacterial cells were lysed and soluble proteinaceous fractions isolated and analysed by MALDI-TOF-MS. The measured mass to charge ratios of the proteins isolated were found to be genus-specific and species- specific biomarkers, which were obtained on several repeated analyses (Krishnamurthy and Ross, 1996, Rapid Communications in Mass Spectrometry 10, i 992- 1996, PMID 9004531). Although MALDI-TOF-MS has recently been used to characterise fungal spores (Welham et al, 2000, Rapid Communications in Mass Spectrometry 14 (5), p307- 310, PMID 10700030), filamentous fungi, moulds (WO 01/92872) and intracellular yeast proteins (Amiri-Eliasi andFenselau, Analytical Chemistry, 2001, 73 p5228-5231, PMID 11721923), its application to the identification and characterisation of specific genera, species and strains of yeasts has not yet been demonstrated.

Notably, WO 01/92872 claims methods for generating biomarkers for a known genus, species or strain of yeast, but does not show that being done, nor does it suggest any way in which that might be achieved, instead providing examples for Penicillium (a filamentous fungi), pollen and bacteria. Similarly, Amiri-Eliasi and Fenselau (Analytical Chemistry, 2001, 73 p5228-5231, PMID 11721923) describe methods of lysis of whole yeast cells with consequent generation of biomarkers, and the identification of several organellular proteins. However, their methodology and data do not demonstrate that specific genera, species and strains of yeasts can be distinguished from each other - it merely shows that proteins such as protonated endogenous ubiquitin and several ribosomal proteins, which are released upon cellular lysis, can be detected by MALDI-TOF-MS.

The present inventor has now identified a method which allows for the generation of mass spectra and subsequent identification of biomarkers characteristic of the genus, species, and/or strain of yeast (i.e. at least the genus of the yeast), it being possible to identify the biomarkers using mass spectrometry. The key to the present invention is the treatment of the yeast with a yeast cell wall digesting enzyme to effect a partial digestion of the yeast cell wall - it has been found that the resulting molecules released from the yeast are reproducibly characteristic of the genus, species, and/or strain of yeast and are thus particularly useful in providing a rapid and accurate identification of a yeast sample. Zymolase treatment of whole yeast cells has been described previously (Amiri-Eliasi and Fenselau, Analytical Chemistry, 2001, 73 p5228-5231, PMID 11721923). Crucially, this Zymolase digestion of yeast cells was used to effect complete lysis of the yeast cell in order to characterise intracellular yeast proteins, instead of partially digesting the cell wall to release genus, species, and/or strain specific biomarkers. Consequently, the previously described use of Zymolase to effect total cellular lysis resulted in overly complex mass spectra comprising many protein peaks, and did not facilitate the identification of different genus, species, and/or strains of yeast. Notably, there is no suggestion nor is there any disclosure of using partial digestion.

Summary of the invention

In one embodiment of the present invention, a method is provided for generating biomarkers for yeasts at the genus, species or strain taxonomic level employing mass spectrometry. It is also an objective to provide such a method that uses a MALDI-TOF mass spectrometer to generate these biomarkers. In another embodiment of the present invention, an unknown yeast sample can be analysed by mass spectrometry and spectra obtained for that sample compared with the foregoing biomarkers to quickly and easily identify the unknown yeast. The present invention thus permits one to more quickly and easily identify and classify these bioorganic compounds than prior art methods.

Using the methods of the present invention, a library of biomarkers can advantageously be constructed from known samples to aid in future identification of yeasts. The term "biomarker" as used herein refers to an ion or charged molecular fragment produced by mass spectrometry that produces a unique peak and/or peaks on a mass spectrum at the genus, species or strain level.

The term "bioorganic compounds" as used herein refers to whole cells or cellular components of yeasts that will generate biomarkers when the cells or cellular components are subjected to mass spectrometry.

The term "identity" as used herein refers to the genus or species or strain of a yeast.

Thus, according to a first aspect of the present invention there is provided a method for generating mass spectra and biomarkers specific for a known genus, species, or strain of yeast comprising the steps of: i) providing a sample of yeast comprising a suspension of whole yeast cells, wherein said sample comprises a known genus, species, or strain of yeast; ii) contacting said sample with a yeast cell wall digesting enzyme to partially digest the cell wall of said yeast cells; iii) placing an aliquot of said sample into a mass spectrometer; iv) subjecting the sample to an ion source to produce charged molecular ions; v) propelling the ions into a mass analyser to obtain a mass spectrum; vi) repeating steps (i)-(v) with at least one other non-identical yeast sample comprising the same genus, species, or strain of yeast; vi) comparing the mass spectra obtained for each yeast; vii) identifying at least one peak on the spectra that is common to each sample; and viii) assigning an m/z measurement of at least one peak as a genus, species, or strain specific biomarker. According to a second aspect of the present invention there is provided a method for determining the genus, species and/or strain of an unknown yeast which comprises: i) generating a mass spectrum of the unknown yeast according to steps (i)-(v) of the first aspect of the present invention; ii) comparing the mass spectrum of the unknown yeast to a plurality of yeast genus, species, or strain specific biomarkers, said biomarkers being generated according to the method of the first aspect of the present invention; and iii) correlating the results of comparison step (ii) to determine the genus, species, and/or strain of the unknown yeast.

Mass spectrometers are commercially available and can include, for example, single or multiple quadrupole, single or multiple magnetic sector, Fourier Transform ion cyclotron resonance (FTICR), ion trap, and combinations thereof (e.g. ion-trap/time-of-flight). Mass spectrometers and ionisation sources, and their uses are discussed in further detail by WO 01/92872.

Mass spectrometry (MS) is an analytical technique, most often used for the identification of chemical structures (e.g tandem-MS), determining the composition of mixtures (e.g. gas chromatography-MS), or quantitative elemental analysis (e.g. inductively coupled plasma-MS). The principle of mass spectrometry relies on the generation, separation and detection of ions, according to their mass and charge, and relative abundance. Upon analysis, a given analyte will generate a unique spectral "fingerprint", corresponding to the mass distribution of its component ion fragments. The ions may be identified from the centroid of the peak of their mass to charge ratio (m/z) from the resulting mass spectrum.

One type of mass spectrometer is a time-of-flight (TOF) mass spectrometer which separates ions according to their mass-to-charge ratio by measuring the time it takes generated ions to travel to a detector. See, e.g. US 5045694 and US 5160840. TOF mass spectrometers are advantageous because they are relatively simple instruments with a virtually unlimited mass-to-charge ratio range. TOF mass spectrometers have potentially higher sensitivity than traditional scanning instruments because they can record all the ions generated from each ionisation event. TOF mass spectrometers are particularly useful for measuring the mass-to-charge ratio of large organic molecules where conventional magnetic field mass spectrometers lack sensitivity.

Typically, TOF mass spectrometers include an ionisation source for generating ions of sample material under investigation. The ionisation source contains one or more electrodes or electrostatic lenses for accelerating and properly directing the ion beam. In the simplest case the electrodes are grids. A detector is positioned at a predetermined distance from the final grid for detecting ions as a function of time. Generally, a drift region exists between the final grid and the detector. The drift region allows the ions to travel, in free flight, a predetermined distance before they impact the detector.

One example of an ionisation source frequently used in mass spectrometry is Electron Ionisation (El). In El, the sample is vaporized within the mass spectrometer prior to passing into an electron ionisation region, where it is subjected to an electron beam. While databases have been created that contain the El mass spectra of over 100,000 compounds, facilitating the structural determination of molecules, El is limited in its application to small molecules below the range of common bioorganic compounds, (see, e. g., Siuzdak, Mass Spectrometry for Biotechnology, pp. 6-7 (1996).) El is inadequate in treating compounds above a molecular weight of about 400 Da because the thermal treatment causes decomposition of the analyte to be tested prior to vaporization and can lead, in many cases, to excessive fragmentation.

Other ionisation techniques used in mass spectrometry are laser desorption (LD) and other "soft" ionisation techniques, such as fast atom bombardment (FAB), plasma desorption and electrospray ionisation (ESI). These techniques were developed to address the problem of ionizing polar, thermally labile, nonvolatile compounds, such as bioorganic molecules, for mass spectrometric analysis. These properties are typical of bioorganic molecules (e.g. proteins, nucleic acids, oligosaccharides) and preclude or interfere with the acquisition of their spectra using a "hard" ionisation technique such as electron impact.

In FAB, the sample to be analysed is added to a matrix, usually a non- volatile solvent in which the sample is dissolved, prior to analysis. FAB typically uses a direct insertion probe for sample introduction and a high-energy beam of Xe atoms, Cs⁺ ions, or NH₄ ⁺ clusters to sputter the sample and matrix from the probe surface. The matrix replenishes the surface with new sample as the ion beam bombards the surface and absorbs most of the incident energy thus minimizing damage to the sample from the high-energy particle beam. The matrix is also believed to facilitate the ionisation process. The two most common matrices used are m-nitrobenzyl alcohol and glycerol. The ion beam desorbs the sample of interest from the matrix solution into the gas phase; charged molecules are then electrostatically propelled into the mass analyser. Massive Cluster Impact (MCI), a form of FAB produces multiply charged ions making it more suitable for use with high-molecular weight biopolymers.

Electrospray ionisation is used to produce gaseous ionized molecules from a liquid solution containing the analyte. A fine spray of highly charged droplets is created in the presence of a strong electric field at the tip of a metal nozzle maintained at about 4000 V, which are then attracted to the mass spectrometer inlet. Dry gas, heat, or both are applied to the droplets before they enter the mass spectrometer so that the solvent evaporates from the surface. The electric field density on the surface of the droplet increases as its size decreases, eventually leading to the expulsion of ions, which are directed to an orifice leading to the mass analyser. ESI is conducive to the formation of multiply charged molecules, making it possible to obtain spectra from high molecular weight compounds. Of all the desorption techniques, plasma desorption and laser desorption result in the smallest initial spatial distributions, and thus less broadening of mass spectral peaks, because ions originate from well defined areas on the sample surface and the initial spatial uncertainty of ion formation is negligible. These methods, especially laser ionisation methods, make it possible to preserve useful structural information and directly obtain mass spectra from polar, high molecular weight compounds encountered in microbiological studies without resorting to pyrolysis and other more destructive techniques.

Laser desorption is considered a "soft" ionisation technique because the resulting spectra are dominated by molecular ions instead of fragment ions. Conventional LD typically employs sufficiently short pulses (frequently less than 10 nanoseconds) to minimize temporal uncertainty. However, in some cases, ion generation may continue for some time after the laser pulse terminates causing loss of resolution due to temporal uncertainty. The main requirement, and perhaps the main drawback, of LD/MS is that analytes must absorb at the wavelength emitted by the laser. These requirements limit the range of compounds studied using LD/MS.

The performance of LD may be substantially improved by the addition of a small organic matrix molecule to the sample material. This technique, known as matrix-assisted laser desorption/ionisation (MALDI), is based on the discovery that the desorption/ionisation of large, non-volatile molecules such as proteins can be effected when a sample of such molecules is irradiated after being co-deposited with a large molar excess of an energy-absorbing "matrix" material, even though the molecule does not strongly absorb at the wavelength of the laser radiation. The abrupt energy absorption initiates a phase change in a microvolume of the absorbing sample from a solid to a gas while also inducing ionisation of the sample molecules. Once the sample molecules are vaporized and ionized they are transferred electrostatically to the flight tube of a time-of-flight mass spectrometer. MALDI is particularly advantageous in biological applications since it facilitates desorption and ionisation of large biomolecules in excess of 100,000 Da while keeping them intact. MALDI has been successfully coupled with TOF mass spectrometers to analyse large molecules.

Ionisation of the analyte is effected by pulsed laser radiation focused onto the probe tip which is located in a short (4 cm) source region containing an electric field. The molecular ions and/or fragment ions formed at the probe tip are accelerated by the electric field toward a detector through a flight tube, which is a long field free drift region.

Since all molecular ions receive the same amount of energy, the time required for ions to travel the length of the flight tube is dependent on their mass. Thus, low-mass ions have a shorter time of flight (TOF) than heavier ions. All the ions that reach the detector as the result of a single laser pulse produce a transient TOF signal. Typically, ten to several hundred transient TOF mass spectra are averaged to improve ion counting statistics.

The mass of an unknown analyte is determined by comparing its experimentally determined TOF to TOF signals obtained with ions of known mass. The MALDI- TOF-MS technique is capable of determining the mass of proteins of between 1 and 40 kDa with a typical accuracy of 0. 1%, and a somewhat lower accuracy for proteins of molecular mass above 40 kDa.

Thus, the mass spectrometer may be selected from the group consisting of linear or non-linear reflectron time-of-flight, single or multiple quadrupole, single or multiple magnetic sector, fourier transform ion cyclotron resonance, ion trap and combinations thereof. The ion source may be selected from the group consisting of laser desorption, fast atom bombardment, plasma desorption, electrospray ionisation, or massive cluster impact.

In particular, the mass spectrometer may be a time-of-flight mass spectrometer, and matrix- assisted laser desorption/ionisation may be used as the ion source.

According to a third aspect of the invention there is provided a method for generating biomarkers specific for a known genus, species, or strain of yeast comprising the steps of: i) providing a sample of yeast comprising a suspension of whole yeast cells, wherein said sample comprises a known genus, species, or strain of yeast; ii) contacting said sample with a yeast cell wall digesting enzyme to partially digest the cell wall of said yeast cells; iii) mixing a sample comprising a suspension of a known genus, species or strain of yeast with a matrix solution to generate a sample mixture; iv) placing an aliquot of said sample mixture on the probe tip of the time-of- flight mass spectrometer and allowing it to dry; v) irradiating the dried aliquot with pulsed laser radiation to form charged molecular ions; vi) accelerating the charged molecular ions by an electric field toward a detector through the flight tube of the time-of-flight mass spectrometer to obtain a mass spectrum; vii) averaging the mass spectra resulting from 10 to 500 laser pulses; viii) repeating steps (i)-(vii) with at least one other, non-identical yeast sample of the same genus, species or strain; ix) comparing the averaged mass spectra obtained for each yeast sample; x) identifying at least one peak that is common to each yeast sample; and xi) assigning an m/z measurement of the at least one peak as a yeast genus, species, or strain specific biomarker. According to a fourth aspect of the invention there is provided a method for determining the identity of an unknown yeast which comprises: i) generating mass spectra of the unknown yeast according to steps (i)-(vii) of the third aspect of the present invention; ii) comparing the averaged mass spectrum of the unknown yeast to a plurality of yeast genus, species or strain specific biomarkers, said biomarkers being generated according to the method of the third aspect of the present invention; and iii) correlating the results of comparison step (ii) to determine the genus, species, and/or strain of the unknown yeast.

Yeast cell walls are typically larger than those of bacterial cells, and are generally 80-90% polysaccharide, including predominant glucans such as 1,3-β-glucan, and also the long chain carbohydrate polymer chitin which adds rigidity and structural support to the cells. Proteins (such as mannoproteins), lipids and polyphosphates together with inorganic ions make up the cell wall cementing matrix (Welham et al, 2000, Rapid Communications in Mass Spectrometry 14 (5), p307-310, PMID 10700030).

The enzymatic treatment to partially digest the yeast cell wall digests it sufficiently to render components of the yeast cell wall susceptible to ionisation, particularly ionisation from a laser when the treated cells are mixed with a suitable matrix material, i.e. turns them into specific biomarkers, as detected by MS. The enzymatic treatment step releases cell wall proteins (such as mannoproteins) through digestion of glucan. The released biomarkers are thus likely to be a combination of glucan and mannoproteins. The yeast sample should be incubated at an appropriate temperature in the presence of the cell wall digesting enzyme for a time period sufficient to release components of the cell wall, which can be detected by MS. For example, the partial digestion may be achieved by suspending 1-5 mg wet weight yeast cells in 100 μl of a solution containing 100 units lyticase in potassium phosphate buffer, pH 7.2, followed by an incubation at 30 °C for 15 minutes. Importantly, and in direct contrast to previous reports of Zymolase treatment of yeast prior to MALDI-TOF-MS analysis (Amiri-Eliasi and Fenselau, Analytical Chemistry, 2001, 73 p5228-5231, PMID 11721923), the partial digestion should not cause the generation of an overly-complex indecipherable spectrum which can result when the yeast sample is over- digested, such that the individual cells are lysed. For each yeast genus, species, and/or strain tested, the partial enzymatic digestion, ionisation and analysis by mass spectrometry results in reproducible and characteristic spectral patterns (fingerprints), due to differences in the mannoprotein content and the glucan structure of the cell wall, which can vary considerably both in the degree of branching and molecular size.

Compared with current methods of yeast identification, the present invention is much faster (with results typically available within the hour), is not labour intensive and can be carried out by non-specialists, and can use dried solid samples that are easy to handle and store. Additionally, the present invention uses simple methodology that minimize the risk of experimental errors, and it offers potential for enhanced and automated throughput of samples.

The partial digestion of the cell wall of the yeast may be achieved using a solution containing a yeast cell wall digesting enzyme, either a pure or crude preparation of a β- glucanase, for example a β-l,3-glucanase, such as Zymolase or Lyticase. Other suitable enzymes which can be used to partially digest the yeast cell wall to release such biomarkers will be readily apparent to a person skilled in the art and include other glucanases such as β-l,6-glucanase. The efficacy of a given enzyme can be readily determined by using it in the method of the present invention and comparing resultant spectra to confirm whether or not they are useful for identifying different yeast. The solution containing the yeast cell wall digesting enzyme may be buffered to a pH suitable for activity of the yeast cell wall digesting enzyme, for example by using potassium phosphate buffer, pH 7.2.

The solution containing the yeast cell wall digesting enzyme may also contain a chemical for disruption of thiol groups, for example β-2-mercaptoethanol. By placing the digestion mixture under reducing conditions, the disulphide bridges of cell wall proteins are disrupted. This disruption increases the exposure of β-l,3,D-endoglucan to the cell wall digesting enzyme, facilitating digestion of the yeast cell wall.

The partial enzymatic digestion of the yeast cell wall may be achieved by incubation at a temperature between 4-50 °C.

A wide range of matrix solutions are well known in the field of MALDI-TOF-MS and can be used in the present invention. For example, the matrix solution may be 4-hydroxy-α- cyano-cinnamic acid, or 3,5-dimethoxy-4-hydroxycinnamic acid, dissolved at a concentration of lOmg/ml. Other matrix solutions which can be used include 2,5- dihydroxybenzoic acid, 2',4',6'-trihydroxyacetophenone (THAP), 3-hydroxy-2- pyridinecarboxylic acid, dithranol (DIT), 2,-(4-hydroxy-phenylazo)-benzoic acid (HABA), trans-3-indoleacrylic acid (IAA), 4-hydroxy-3-methoxycinnamic acid, nicotinic acid-N- oxide, 2'-6'-dihydroxyacetophenone, 2-pyridine carboxylic acid, and 6-aza-2-thiothymine (ATT).

The ratio of the treated yeast sample and the matrix solution may be between 1 : 1 and 1 :50 - optimal ratios can be readily determined by routine experimentation.

Ionisation may be induced with a nitrogen UV laser (wavelength 337nm) and positive ion spectra may be obtained with a Lasermat 2000 linear time-of-flight mass spectrometer (Finnegan Mat Ltd, Hemel Hempstead, Hertfordshire) using high resolution acquisition with filters in linear mode. The positive ions may be recorded by the inbuilt detector and the signal averaged on a PC using the dedicated Lasermat 2000 software package. After acquisition, the spectra may be baseline corrected and smoothed using the proprietary software. The sample port may be kept at 4.5 x 10^"7 mBar pressure.

Naturally the present invention is not linked to the use of specific wavelengths of ionisation lasers, or specific modes of ionisation, or specific types of mass spectrometry devices such as MALDI-TOF-MS devices, and instead simply requires that mass spectrometry is used to detect the generated ions.

In the case of using the mass spectra data to define spectral characteristics not for a specific yeast strain but for higher taxonomic levels, e.g. species or genus, then the spectrum can be correlated with those obtained from samples of other yeasts of the same genus or species (as appropriate) in order that spectral patterns or characteristics specific to the genus or species are identified.

Thus a spectrum for a sample yeast can be generated and used to interrogate a database of spectra or other characteristics for given strains, species or genera of yeast in order to identify the yeast e.g. to confirm it as being of a given strain, or as being of an unknown strain and species but of a known genus, or as being totally unknown.

According to a fifth aspect of the invention there is provided a method for generating a database containing mass spectra for at least one yeast comprising the steps of: i) generating at least one mass spectrum for said at least one yeast according to the method of any one of the first, second, or third aspects of the invention; ii) compiling said at least one mass spectrum on said database; According to a sixth aspect of the invention there is provided a method for generating a report of the identity of a yeast sample comprising the steps of: i) generating at least one mass spectrum for said yeast sample according to the method of any one of the first, second, or third aspects of the invention; ii) comparing the at least one mass spectrum of said yeast sample with mass spectra obtained according to the method of any one of claims 1-11 for at least one yeast having a known identity; iii) correlating the results of comparison step (ii) to determine the identity of said yeast sample; and iv) generating a report of the identity of said yeast sample incorporating the results of correlation step (iii).

According to a seventh aspect of the invention there is provided a biomarker library, the library comprising yeast genus, species, and/or strain specific biomarkers for known yeasts generated by the method of any one of the first, second, or third aspects of the invention. The yeasts in the library may be selected from the taxonomic order Endomycetales.

The invention will be further apparent from the following description, with reference to the several figures of the accompanying drawings, which show, by way of example only, one form of rapid identification of yeast using mass spectrometry.

Of the Figures:

Figure 1 shows an example of mass spectra obtained by MALDI-TOF- MS analysis of Saccharomyces cerevisiae NCYC22118 without prior enzymatic digestion (Panel a), or following a 20 minute partial enzymatic digestion in the presence of lyticase (Panel b); Figure 2 shows an example of mass spectra obtained by MALDI-TOF- MS analysis following partial enzymatic digestion with lyticase for three species of Candida - Candida albicans NCYC 597 (Panel a), Candida boidinii NCYC1513 (Panel b), and Candida tropicalis NCYC 1503 (Panel c); and

Figure 3 shows an example of mass spectra obtained by MALDI-TOF- MS analysis following partial enzymatic digestion with lyticase for three strains of Saccharomyces cerevisiae - Saccharomyces cerevisiae NCYC 19201 (Panel a), Saccharomyces cerevisiae NCYC22118 (Panel b), and an unknown strain of Saccharomyces cerevisiae (Panel c).

EXPERIMENTS

The following experiments demonstrate the differences in mass spectra for various yeast samples, obtained with and without partial enzymatic digestion of the yeast cell wall. The results show that a greater number of distinct peaks are obtained with an enzymatic digestion step than are obtained without such a treatment step (Figure 1). The results also demonstrate that different mass spectra are obtained following an enzymatic digestion step for different species of yeast (Figure 2), and different strains of yeast (Figure 3). Overall, these experiments show clear differences in the distribution of peaks within strains, between species and between genera after treatment with lyticase. The finding that each strain, species or genus of yeast has a unique spectral fingerprint upon enzymatic digestion with lyticase, ionisation and analysis by mass spectrometry can be exploited to identify an unknown yeast sample by comparison with mass spectra obtained from known samples of yeast.

The contents of each of the references discussed herein, including the references cited therein, are herein incorporated by reference in their entirety.

Where "PMID:" reference numbers are given for publications, these are the PubMed identification numbers allocated to them by the US National Library of Medicine, from which full bibliographic information and abstract for each publication is available at www.ncbi.nlm.nih.gov. This can also provide direct access to electronic copies of the complete publications, particularly in the case of e.g. PNAS, JBC and MBC publications.

Example 1

Cells of Candida albican$ ^~NCYC 597, Candida boidiniXNCYC1513, Candida tropicalis NCYC 1503, Saccharomyces cerevisiae NCYC 19201, Saccharomyces cerevisiae NCYC22118 and an unknown strain of Saccharomyces cerevisiae were from the culture collection held by De Montfort University, Leicester, U.K. (NCYC strains available from National Collection of Yeast Cultures, Institute of Food Research, Norwich, UK). Cells were grown under standard conditions in malt extract broth for 18-24 hours at 30 °C, harvested and suspended (1-5 mg wet weight) in 100 μl of a solution containing 100 units lyticase in potassium phosphate buffer, pH 7.2. This reaction mixture was incubated at 30 °C for 15 minutes and the cells were harvested by centrifugation at 13 000 rpm for one minute. Cell pellets were washed once in 0.1 % trifluoroacetic acid and once in sterile distilled water before being suspended in sterile distilled water (50 μl). A 5 μl aliquot of this suspension was mixed with 45 μl of a 10 mg/ml solution of 3,5-dimethoxy-4- hydroxycinnamic acid in 0.1 % trifluoroacetic acid/acetonitrile (ratio 2:1). A 1 μl aliquot of this suspension was deposited onto a stainless steel sample probe and allowed to air dry. Controls were also prepared for the yeast cells by omitting the lyticase treatment step, for the lyticase solution and for the matrix. All of the samples prepared were then ionized using MALDI-TOF-MS at 26kV acceleration in linear mode, giving an ion flight path of 1.25 metres in a Lasermat 2000 linear time-of-flight mass spectrometer (Fi negan Mat Ltd, Hemel Hempstead, Hertfordshire).

Mass spectral analysis of yeast cells produced clear differences in the number of distinct peaks before and after enzymatic digestion with lyticase. Figure 1 shows the effect of a 20 minute enzymatic digestion with lyticase on Saccharomyces cerevisiae NCYC22118. Two distinct peaks were obtained without enzyme digestion (Panel a) compared with eight distinct peaks after partial digestion with lyticase (Panel b). Compared to undigested samples, lyticase treatment also increased the number of peaks obtainable for Candida albicans NCYC 597, Candida boidinii NCYC1513, Candida tropicalis NCYC1503, Saccharomyces cerevisiae NCYC19201, Saccharomyces cerevisiae NCYC22118 and the unknown strain of Saccharomyces cerevisiae (data not shown). These data demonstrate the efficacy of the enzymatic pretreatment specified in this invention for producing fingerprint spectra from yeast cells. Figure 2 shows the mass spectra obtained by MALDI-TOF-MS analysis following a 15 minute enzymatic digestion with lyticase for three species f Candida - Candida albicans NCYC 597 (Panel a), Candida boidinii NCYC1513 (Panel b), and Candida tropicalis NCYC1503 (Panel c). The data sets obtained show distinct differences in the mass spectra obtained for each species, illustrating the application of the invention to the identification of yeasts of different species.

Figure 3 shows the mass spectra obtained by MALDI-TOF-MS analysis following a 15 minute enzymatic digestion with lyticase for three strains of Saccharomyces cerevisiae - Saccharomyces cerevisiae NCYC 19201 (Panel a), Saccharomyces cerevisiae NCYC22118 (Panel b) and an unknown strain of Saccharomyces cerevisiae (Panel c). The data sets obtained show distinct differences in the mass spectra obtained for each strain, illustrating the application of the invention to the identification of different strains from the same species of yeast.

Example 2

In order to determine the identity of an unknown yeast, a MALDI-TOF mass spectrum is generated for the yeast as detailed above. The measured mass charge ratio (m/z) for a particular ion varies slightly, within a range of 0.5% for replicate analyses and 1.0% for separate cultures of a known yeast. Any averaged m/z measurement of an unknown yeast was considered a potential genus, species or strain biomarker if also present in mass spectra of known yeasts. These biomarkers may be common to other genera, species or strains of yeasts, or they may be specific to the yeast being analysed. The biomarkers assigned to each yeast species were the average of the m/z ratios for ions measured in 4-5 samples. All of the biomarkers for a species must be present for a positive identification.

Table 1: An example of the reproducibility of MALDI-TOF-MS for Candida species.

The m/z ratios for species specific biomarkers are shown from replicate experiments. These peaks are reproducible within 0.5-1.0%) in separate samples and within 0.5% for replicate analyses.

In some cases positive identification may be obtained by analysing trends in mass spectra e.g. the difference in m/z value between adjacent peaks in the data.

Example 3

In order to generate a database containing mass spectra for yeasts, a MALDI-TOF-MS spectrum is generated for the sample as detailed above, and the spectrum is submitted to the database. Each spectrum compiled in the database is recorded as numerical data (in both digital and analogue form) and as a graphical representation of a plot of intensity against mass to charge ratio. Preferably, each spectrum is stored on the storage medium of a computer or on a magnetic storage medium, such as a floppy disk. Data is also stored in paper form. Alternatively, storage may be optical, such as CD-ROM or laser disc, or in ROM microchips. Preferably, each spectrum is stored as a series of m/z ratios, each corresponding to the centroid of the peaks of each spectrum. By storing m/z ratios alone, the volume of data stored is minimised. To aid the search and identification process, data may be arranged in groups of data corresponding to the genus of each yeast, with subdivisions corresponding to the species and strain of yeast. Example 4

Where the database is stored in the storage medium of a computer, software may be provided to carry out the screening of the unidentified mass spectra. The software may carry out the steps of acquiring and storing the unidentified micro-organism's mass spectra by manual input, or preferably directly from the spectrometer. Additionally, by retrieving individual files from the database and overlaying them, unidentified spectra may be compared with the pre-characterised mass spectra in the database and similarities or differences may be highlighted. The software may be capable of allowing for calibration errors in the unidentified mass spectra, so that trends (e.g. the difference in m/z value between adjacent peaks in the data) may be screened with the corresponding data in the database rather than the absolute values. The computer may be incorporated into the MALDI-TOF mass spectrometer or spaced apart from the spectrometer and linked by remote means.

Where a spectrum analysed is not capable of identification, the spectra may be saved as individual files in a database, and reviewed for identification or reference purposes at a later date.

Example 5 In order to generate a report of the identity of a yeast sample a MALDI-TOF-MS spectrum is generated for the sample as detailed above (Example 1). This is then compared to spectra generated as above for a set of yeast samples of a known taxonomy. The results of the comparison step are correlated in order to determine the identity of the yeast sample. A report of the identity of the yeast sample is generated, incorporating the results of the correlation step. Example 6

In order to generate a biomarker library, MALDI-TOF mass spectra are generated for samples as detailed above (Example 1). The mass spectra together with any information relating to the identity of the yeast sample are then compiled to produce a biomarker library.

It will be understood that various modifications may be made to the embodiments disclosed herein. Therefore the above description should not be construed as limiting, but merely as exemplifications of preferred embodiments. For example, different types of yeasts other than those disclosed herein can be analysed and biomarkers generated therefrom. Those skilled in the art will envision other modifications within the scope and spirit of the claims appended hereto.

Claims

1. A method for generating mass spectra and biomarkers specific for a known genus, species, or strain of yeast comprising the steps of: i) providing a sample of yeast comprising a suspension of whole yeast cells, wherein said sample comprises a known genus, species, or strain of yeast; ii) contacting said sample with a yeast cell wall digesting enzyme to partially digest the cell wall of said yeast cells; iii) placing an aliquot of said sample into a mass spectrometer; iv) subjecting the sample to an ion source to produce charged molecular ions; v) propelling the ions into a mass analyser to obtain a mass spectrum; vi) repeating steps (i)-(v) with at least one other non-identical yeast sample comprising the same genus, species, or strain of yeast; vi) comparing the mass spectra obtained for each sample; vii) identifying at least one peak on the spectra that is common to each sample; and viii) assigning an m/z measurement of at least one peak as a genus, species, or strain specific biomarker.

2. The method according to claim 1 wherein the mass spectrometer is selected from the group consisting of linear or non-linear reflectron time-of-flight, single or multiple quadrupole, single or multiple magnetic sector, fourier transform ion cyclotron resonance, ion trap and combinations thereof.

3. The method according to either one of claims 1 or 2 wherein the ion source is selected from the group consisting of laser desorption, fast atom bombardment, plasma desorption, electrospray ionisation, and massive cluster impact.

4. The method according to any one of claims 1-3 wherein the mass spectrometer is a time-of-flight mass spectrometer.

5. The method according to claim 4 wherein matrix-assisted laser desorption/ionisation is used as the ion source.

6. The method according to claim 5, said time-of-flight mass spectrometer comprising a probe tip, a detector, and a flight tube, the method comprising the steps of: i) providing a sample of yeast comprising a suspension of whole yeast cells, wherein said sample comprises a known genus, species, or strain of yeast; ii) contacting said sample with a yeast cell wall digesting enzyme to partially digest the cell wall of said yeast cells; iii) mixing a sample comprising a suspension of a known genus, species or strain of yeast with a matrix solution to generate a sample mixture; iv) placing an aliquot of said sample mixture on the probe tip of the time-of- flight mass spectrometer and allowing it to dry; v) irradiating the dried aliquot with pulsed laser radiation to form charged molecular ions; vi) accelerating the charged molecular ions by an electric field toward a detector through the flight tube of the time-of-flight mass spectrometer to obtain a mass spectrum; vii) averaging the mass spectra resulting from 10 to 500 laser pulses; viii) repeating steps (i)-(vii) with at least one other, non-identical yeast sample of the same genus, species or strain; ix) comparing the averaged mass spectra obtained for each yeast sample; x) identifying at least one peak that is common to each yeast sample; and xi) assigning an m/z measurement of the at least one peak as a yeast genus, species, or strain specific biomarker.

7. A method for determining the genus, species and/or strain of an unknown yeast which comprises: i) generating a mass spectrum of the unknown yeast according to steps (i)-(v) of claim 1; ii) comparing the mass spectrum of the unknown yeast to a plurality of yeast genus, species, or strain specific biomarkers, said biomarkers being generated according to the method of claim 1; and iii) correlating the results of comparison step (ii) to determine the genus, species, and/or strain of the unknown yeast.

8. A method for determining the identity of an unknown yeast which comprises: i) generating mass spectra of the unknown yeast according to steps (i)-(vii) of claim 6; ii) comparing the averaged mass spectrum of the unknown yeast to a plurality of yeast genus, species or strain specific biomarkers, said biomarkers being generated according to the method of claim 6; and iii) correlating the results of comparison step (ii) to determine the genus, species, and/or strain of the unknown yeast.

9. A method according to any one of the preceding claims, said yeast cell wall digesting enzyme being a pure or crude preparation of at least one β-glucanase.

10. A method according to any one of the preceding claims, wherein said yeast cell wall digesting enzyme is buffered to a pH suitable for activity of said yeast cell wall digesting enzyme.

11. A method according to any one of the preceding claims, wherein said yeast cell wall digesting enzyme contains a chemical for disrupting thiol groups.

12. A method according to any one of the preceding claims, wherein said yeast cell wall digesting enzyme comprises β-2-mercaptoethanol.

13. A method according to any one of the preceding claims wherein said yeast cells and said yeast cell wall digesting enzyme are incubated at a temperature between 4-50

°C.

14. A method for generating a database containing mass spectra for at least one yeast comprising the steps of: i) generating at least one mass spectrum for said at least one yeast according to the method of any one of claims 1-6, or claims 9-13 when dependent on any one of claims 1-6; and ii) compiling said at least one mass spectrum on said database;

15. A method for generating a report of the identity of a yeast sample comprising the steps of: i) generating at least one mass spectrum for said yeast sample according to the method of any one of claims 1-6, or claims 9-13 when dependent on any one of claims 1-6; and ii) comparing the at least one mass spectrum of said yeast sample with mass spectra data obtained according to the method of any one of claims 1-6, or claims 9-13 when dependent on any one of claims 1-6 for at least one yeast having a known identity; iii) correlating the results of comparison step (ii) to determine the identity of said yeast sample; and iv) generating a report of the identity of said yeast sample incorporating the results of correlation step (iii).

16. A biomarker library comprising yeast genus, species, and/or strain specific biomarkers for known yeasts generated by the method of any one of claims 1-6, or claims 9-13 when dependent on any one of claims 1-6.

17. The library of claim 16 wherein the yeasts are selected from the taxonomic order Endomycetales.