Article is available in full to IFST members and subscribers.

Register on the FST Journal website for free

Click the button to register to FST Journal online for free and gain access to the latest news

 

If you are an IFST member, please login through the Members Area of the IFST website. 

 

 

 

 

 

 

 

 

 

 

 

 


It’s a kind of magic - Gaining time in food microbiology by the use of MALDI-ToF

Maria Karczmarczyk of Eurofins desrcibes MALDI-ToF diagnostic microbiology method.

Introduction

The repertoire of diagnostic microbiology methods includes traditional tests based on morphological, biochemical, and antigenic properties, collectively referred to as phenotypic.  Molecular identification methods, on the other hand, depend on the presence of specific nucleic acid sequences detected through PCR and/or sequencing.  The latter includes 16S RNA ribosomal gene sequencing, currently considered as the most definitive among bacterial identification methods.  Owing to their relatively high cost and complexity, the uptake of molecular methods has largely been limited to specialized clinical diagnostic, research and reference centres, despite their superior sensitivity and specificity. 

Phenotypic confirmatory tests depend on characteristics of microorganisms which are not always stable and or/specific enough to produce reliable results consistently.  In addition, bacteria evolve rapidly which can also result in biological variation leading to inconclusive or erroneous results.  This leads to relatively low accuracy of many identification methods which became the basis for many ISO standards. 

In recent years, matrix-assisted laser-desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) has been added to the spectrum of molecular methods of microbial identification.  This methodology, termed as proteotypic, looks at protein profiles which are typically unique for different species. The identification is achieved through comparison of the protein fingerprints obtained against a reference spectra library of known bacterial and fungal species.  Mass spectrometry (MS) dates back to early 1900s and was originally intended for use in chemistry.  Introduction of the MALDI method in 1987 enabled application of the technique to study large biomolecules (Clark et al., 2013). This development was honoured with a shared Nobel Prize in Chemistry awarded to John B. Fenn, Koichi Tanaka, and Kurt Wüthrich in 2002. 

Since then, the technology has evolved into automatic, user friendly, and highly accurate commercialised systems of microbial identification.  Due to its excellent robustness, MALDI-ToF MS has been integrated successfully into many industrial, pharmaceutical, and clinical microbiology laboratories providing the capability to identify a wide spectrum of microorganisms, including gram-positive, gram-negative, aerobic, and anaerobic bacteria, mycobacteria, yeasts, and moulds. Introduction of MALDI-ToF MS in diagnostic microbiology has been described as revolutionary, paradigm shifting and even the new golden standard (Clark et al., 2013; van Belkum et al., 2017). 

Method principle 

Although it may look simple, the science and algorithms behind the process are sophisticated.  As for cultural identification methods, MALDI-ToF MS typically requires enrichment, followed by isolation of colonies on solid agar media.  In principle, colony sample for MALDI-ToF analysis can be taken directly from selective agar plate, although sub-cultivation onto secondary media may be necessary for certain microorganisms.  This should be verified on a case by case basis for each medium and species/genus of interest.  

A small portion of the colony is deposited and smeared on a dedicated area (spot) of a plate to form an even layer of cells.  Such plates typically contain 48 or 96 spots for analysing individual colonies.  A solution of an organic acid referred to as ‘matrix’ is then deposited on the spot to cover the sample.  Upon drying, matrix co-crystallizes with biomolecules present in the colony material.  These biomolecules include peptides and proteins, the target analytes in the process.  The matrix forms a scaffold and provides a source of protons for the ionization.  The sample-matrix complex undergoes a UV laser irradiation.  The matrix features a maximum of energy absorbance at the same wavelength at which the laser works (337 nm), hence avoiding the destruction of the sample by the laser. The energy of the radiation is transferred to the sample, allowing desorption of the main compounds (transition of the solid state to the gas phase).  Additionally, the sample gets ionized by the slightly acidic matrix (proton transfer) upon which the journey of charged peptides and proteins starts.  After initial acceleration using electrostatic field, the proteins drift towards the detector at a speed defined by their mass to charge ratio.  The end of the journey occurs when they hit the surface of the detector lined with semi-conductive layer which results in electrical impulses produced at times specific for different proteins.  Time of flight (ToF) information is then transformed into mass spectra data depicted as a sequence of mass peaks forming a characteristic microorganisms’ profile also called peptide mass fingerprint (PMF).  The mass range between approximately 2 and 20 kDa is taken into account during the analysis.  The integrated software generates a list of closest matching species classified according to their protein fingerprint similarity, accompanied by a score value which indicates secure species or genus level identification.  

The universality of the method is achieved by cross-checking proteins common to particular genus and/or species.  This is why high abundance housekeeping proteins are targeted.  The major part of this information comes from the fraction of ribosomal proteins in addition to DNA-binding proteins as well as heat shock proteins, among others.  To ensure good sample integrity for mass spectrometric analysis, freshly grown colonies containing actively dividing cells with intact proteins should be subjected to testing.  This repertoire of proteins is indispensable for the cells to maintain their viability and hence is minimally affected by environmental conditions.  Thus, they are the markers of choice for MALDI-ToF MS which is designed to be a universal method, ideally not influenced by factors such as composition of the media etc.  The way computational part of the analysis is performed constitutes proprietary information of the equipment and software providing company.

Advantages 

Studies have shown that the performance characteristics including accuracy, speed, and cost efficiency, by far exceed those typical for classical phenotypic tests (Bessède et al, 2011; Thouvenot et al., 2017). 

The currently available MALDI-ToF MS - based microbial identification systems are highly accurate.  This means that the number of both false positive as well as false negative results is greatly reduced by switching from traditional methods of confirmation to MALDI-ToF MS.  The accuracy of testing method is of paramount importance in food microbiology.  Large batches of food products are routinely tested for the presence of pathogens.  Consequences of reporting any false negative result can be grim when no action is taken and consumers become exposed to foodborne illness.  False positive results on the other hand, can lead to product recall, associated with substantial financial cost, and even compromised brand integrity.  In addition, MALDI-ToF MS dramatically reduces confirmation time.  In food microbiology, it is common practice that laboratory releases notification about so-called ‘presumptive positive’ result after observation of typical pathogen colonies growing on selective agar media.  Confirmation of such colonies by using traditional tests may take up to several days (depending on the microorganism).  By utilizing MALDI-ToF MS, this time can be reduced to minutes, allowing for timely withdrawal of any potentially contaminated food batches and preventing unnecessary product recalls. 

The systems are evolving rapidly to address any remaining gaps and inaccuracies.  Updates to the reference strains library are a pivotal part of this process.  Additional software modules, designed to aid identification, may be added by exploiting more subtle differences in the mass peak spectra.  Certain systems, like MALDI Biotyper, allow for adding mass spectra to the existing proprietary libraries by users themselves.  Such customisation is an exciting feature opening up new possibilities of enriching the library with species of interest.  The process of adding new spectra has to be controlled rigorously.  Strains used for adding their mass spectra have to be identified with confidence in order to prevent potential pollution of the library. 

Preparedness and capability of an average food microbiology laboratory is usually limited to the confirmations of common foodborne pathogens, spoilage and indicator organisms.   Due to its non-targeted nature, MALDI-ToF MS can address this issue to a large extent.  In clinical microbiology, knowledge gained through MALDI-ToF MS - based identification is shedding new light on the organisms present which would otherwise not have been identified using traditional tests.  This has been recognised as invaluable source of information pointing towards the importance of hitherto overlooked bacterial species.  This newly acquired knowledge challenges previous assumption leading to reconsideration of the clinical significance of certain microorganisms (Lagier et al., 2013).  Superiority of MALDI-ToF in that respect has been gaining recognition also in the field of food microbiology.  It represents an invaluable tool in hygiene investigations where unusual or rare species are implicated. 

Ambiguity is a major factor in case of traditional phenotypic tests most of which depend on visual assessment by microbiologists.  One major advantage of MALDI-ToF, which should not be overlooked, is removal of subjectivity of results interpretation – the software produces identification results falling into different colour-coded categories based on the probability of the result being correct at species or genus level. 

The sheer quantity of data which can be generated by traditional methods in a given timeframe will never be matched with those obtained by MALDI-ToF MS.  This opens up new opportunities for wider pathogen/spoilage organism monitoring in a food production environment.  A more comprehensive picture of the microflora present translates into better insight into hygiene status and trends allowing for more efficient tracing of contamination sources and other investigative work. 

Last but not least, replacement of traditional confirmatory tests with MALDI-ToF MS leads to considerable reduction of running costs.  Except for obvious savings (removal of need for biochemical galleries, latex agglutination reagents, in some cases secondary agar media, less labour) resulting from switching from hierarchical, step-wise phenotypic ISO tests, additional cost savings are easily overlooked.  When considering a contract laboratory as part of a business, additional savings resulting from the following: reduced ordering, stock check, receiving, and processing invoices, storage and quality control of reagents and additional culture media, but also saving laboratory management time, training, auditing, potentially reduced events of non-conforming work, and complaints caused by delayed and/or questionable results.  Obviously, the more individual target organism tests are replaced with MALDI-ToF MS confirmations, the greater the impact of savings and efficiency of the laboratory as a whole. 

Limitations 

One potential method limitation is the composition of the reference spectra library.  This aspect continues to be addressed by library expansions with each periodic update performed.  Historically, the systems available have been very clinically-oriented and this continues to be the main application of MALDI-ToF MS gaining more and more recognition among professionals year by year. 

There are instances whereas species level identification is currently impossible for reasons other than absence of the representative reference mass spectrum in the library.  This is usually due to the high degree of mass spectra similarity between closely related species.  Such species are typically very closely related based on ribosomal DNA sequence.  A classic example is inability of the currently available systems to differentiate Escherichia coli and Shigella spp., which have been postulated by taxonomists to constitute a single species.

MALDI-ToF can only be used once the presumptive colonies are available.  As a consequence, in the current form it does not replace rapid PCR methods which are typically applied after or during the enrichment step without the need for plating on selective agar media.  Thus, the overall time saving is limited to confirmatory steps alone, although the process itself is simpler and faster compared with PCR tests, which typically require extraction step, followed by setting up enzymatic reaction and several hours of incubation.  The implication of that is that in some cases PCR may well be a more suitable method when time is a critical factor.  Overall, both approaches offer different advantages which should be considered when deciding which performance characteristics of the method are critical in a specific case. 

Optimal sample preparation has not yet been achieved for certain groups of microorganisms.  In case of most gram-negative bacteria, whose cell wall is relatively thin, sufficient quantity of proteins is released from the cells without additional preparation steps.  Additional treatment with formic acid prior to deposition of matrix solution is usually necessary in case of gram-positive bacteria.  This procedure is simply done by overlaying spotted colony sample with formic acid prior to matrix solution.  Preparation of moulds and Mycobacteria currently requires full extraction which involves multiple manipulation steps. 

Since MALDI requires primary enrichment and selection just as any traditional method, the bias arising from this step cannot be eliminated and becomes a part of the uncertainty of the method as a whole.  Although it is not associated with MALDI-ToF analysis itself, the magnitude of this contribution to the overall uncertainty should not be overlooked and ought to be under control.  This is achieved by appropriate quality controls and use of validated cultivation methods.  Equally, the selection of the colonies shall be carried out competently as this is one of the vulnerable steps where personal judgement is exercised.  This is particularly important for samples with background microflora and poor selectivity of the media used, as well as unspecific colony morphology of the target microorganism.  No matter how accurate the confirmatory test, failure to recover suitable colonies or missing presumptives, which should undergo testing, will lead to invalid results.  As a consequence, one of the main limitations is the one stemming from the fact that MALDI-ToF MS is a confirmation method embedded within a larger process and thus highly dependent on it. 

Finally, as with any microbiological analysis, every MALDI-ToF MS result should be interpreted within the context and by a competent microbiologist.  Perhaps the most critical is the interpretation of unexpected and no identification results.  There can potentially be several reasons for that situation and competence in both microbiology and MALDI-ToF technique is essential in order to judge what should be done in that scenario.  This situation may necessitate repeated analysis to include an extraction step(s), if not performed initially.  At times, falling back on traditional identification procedures may be the only remaining option. 

Standardisation

In April 2017, The Clinical and Laboratory Standards Institute has published M58-Ed1 document Methods for the Identification of Cultured Microorganisms Using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry.  Although intended for clinical laboratory scientists, the document provides a practical source of information for any laboratory implementing MALDI-ToF MS for microbial identification.  The guideline contains recommendations for integration of the method in the existing laboratory operation, and includes overview of performance characteristics, as well as considerations for verification, reporting, quality assurance, and training, among others (CLSI). 

Food industry applications of MALDI-ToF MS method are lagging behind its implementation in medical laboratories.  This can largely be attributed to food microbiology standard methods still depending on phenotypic tests, although some progress has been made recently and work on standardisation of the methodology is on-going.  One notable example is the latest revision of ISO 10272-2:2017 Microbiology of the food chain -- Horizontal method for detection and enumeration of Campylobacter spp., which now mentions MALDI-ToF MS as an alternative confirmation tool.  Another reason for slow uptake is the significant initial investment in the equipment required, although overall financial benefit upon implementation justifies it as the cost is considerably reduced in comparison with classical methods. 

Extensive method verification by a testing laboratory is a prerequisite to obtain accreditation to ISO/IEC 17025:2005 standard.  Such studies need to cover a range of parameters within a realistic laboratory scenario of testing, e.g. agar media types, pre-existing methods (up to the confirmation step), and different users.  All factors contributing to uncertainty of identification results should be captured during validation/verification which will show the overall accuracy of the method under different conditions which can potentially introduce variation.  Such studies frequently expose weaknesses of traditional tests when the two approaches are compared with those based on reference 16S ribosomal gene sequencing. 

Emerging new applications

Strain typing

Both phenotypic and genetic strain typing methods are time-consuming and costly.  Strain typing based on MALDI-ToF MS explores the mass spectra information generated and used for the initial identification meaning that no additional sample manipulation is required.  DNA-based typing methods, on the other hand, require investment in dedicated equipment and are relatively complex to perform and interpret.  However, the diversity of techniques targeting various DNA markers allows for differentiation of bacterial lineages of a wide range of microbial species and has proven its utility for outbreak investigations and other epidemiological studies carried out mainly by reference and research laboratories. 

Information contained within peptide mass fingerprint is limited to a number of markers of a certain mass range.  Published studies have shown varied performance of MALDI-ToF MS-based subtyping attempts, including utilisation of the technique for serotyping, largely dependent on the species of interest, system and software used.  There are published examples of successful use of the MALDI-ToF for determination of phylogenetic relationship of isolates within one species, a feature useful for epidemiological studies and tracing sources of contamination in various environments, including food production and distribution facilities (Ojima-Kato et al., 2017).  In some instances this information may not be sufficient to reliably differentiate different lineages of the same species.  As the systems and algorithms are being perfected over time, it is likely that we will see continued improvements in this area, though. 

Owing to simple workflow and low cost, the possibility of using MALDI-ToF MS for strain typing could be an attractive option for a wider range of laboratories, including food microbiology, for which cost and labour associated with DNA-based typing methods are still prohibitory. 

In summary, some of the data published to date on MALDI-ToF MS for bacterial strain typing is promising but the results should be interpreted with caution.  Careful validation using strain collections characterised by established typing methods should be used to objectively assess their discriminatory power and comparability with methods considered as golden standard such as Pulsed Field Gel Electrophoresis (PFGE) and Multilocus Sequence Typing (MLST). 

Antimicrobial resistance

Growing number of research studies explore MALDI-ToF MS information to predict antimicrobial resistance of microorganisms.  Several different approaches have been evaluated for that purpose.  One of them is based on detecting the presence of specific resistance mediating proteins.  For instance, methicillin resistance of Staphylococcus aureus can be inferred from the presence of a single mass spectral peak at 2415 m/z indicating that PSM-mec is being expressed.  Despite high specificity (≥98%), the sensitivity of this approach was estimated at 37% due to the lack of detectable levels of PSM-mec in the majority of methicillin resistant Staphylococci (Rhoads et al., 2016).  A different methodology exploits known associations of certain clonal lineages of bacterial strains with antimicrobial resistance.  Taking methicillin resistant S. aureus (MRSA) as an example, this approach has been shown to reliably determine four major clonal lineages of MRSA with high sensitivity and specificity (Zhang et al., 2015).  Using MALDI-ToF MS to detect antibiotic hydrolysis is another method evaluated for that purpose.  Unlike two approaches outlined above, this one is based on visualisation of antimicrobial compounds and their degradation products and requires additional manipulation and incubation (addition of antimicrobial compound and time required for its hydrolysis).  The products of the degradation are detected as characteristic changes of the mass peaks profile signifying the manifestation of antimicrobial resistance.  The utility of this approach has been demonstrated for carbapenemase-producing Enterobacteriaceae, among others (Choquet et al., 2018). 

There are several challenges associated with antimicrobial resistance testing using MALDI-ToF MS.  Many of the proteins conferring drug resistance fall outside of the mass range taken into account during analysis and are undetectable, as a result.  Aside from that, the information inferred from MALDI-ToF MS analysis is typically limited to one specific drug or drug class, whilst knowledge about the entire resistance spectrum is typically required in clinical context.  One of the main underlying issues is the fact that in case of many antimicrobials, a plethora of different mechanisms can potentially be responsible for the development of resistance.  MALDI-ToF MS, will likely be limited to the detection of one particular mechanism associated with a relevant marker (unless antimicrobial hydrolysis assay is used).  Should a different mechanism be responsible for the manifestation of resistance, it will remain undetected. 

Conclusions

Owing to the superior specificity and robustness, MALDI-ToF MS transformed diagnostic microbiology despite its introduction into routine practice less than 10 years ago.  Clinical laboratories embraced the technology with all the advantages it offers and the change it introduced is described as revolutionary.  Due to its speed and accuracy it has led to measurable improvements in patient management and generated new knowledge on the prevalence and significance of ‘hard-to-identify’ and rare bacterial species. 

The technology is evolving rapidly.  Many of the early challenges have already been addressed by filling the gaps in the reference spectra libraries and refinements in the computational analysis.  But the progress does not stop there.  Advanced applications such as strain typing and antimicrobial susceptibility testing are now emerging as a possibility but are not yet translated to the level of simplicity and reproducibility expected in routine diagnostics.

In the meantime, standardisation bodies begin to acknowledge the value of the technology which may well be the first step to its wider implementation in food microbiology laboratories.  Ultimately, this should lead to improved consumer safety through enhancing laboratories’ capability by more accurate and faster testing, as well as by widening the spectrum of microorganisms identifiable from food samples. 

 

Figure 2. Example of proteomic fingerprints of microorganisms obtained by using the MALDI Biotyper. The characteristic spectrum pattern is used to identify a particular microorganism by matching with thousands of reference spectra available in a dedicated library database.

Dr Maria Karczmarczyk

Molecular Biologist

Eurofins Food Testing UK Limited

 

References

Bessède E, Solecki O, Sifré E, Labadi L, Mégraud F. Identification of Campylobacter species and related organisms by matrix assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry. Clin Microbiol Infect. 2011 Nov;17(11):1735-9. doi: 10.1111/j.1469-0691.2011.03468.x.

Choquet M, Guiheneuf R, Castelain S, Cattoir V, Auzou M, Pluquet E, Decroix V. Comparison of MALDI-ToF MS with the Rapidec Carba NP test for the detection of carbapenemase-producing Enterobacteriaceae. Eur J Clin Microbiol Infect Dis. 2018 Jan;37(1):149-155. doi: 10.1007/s10096-017-3115-4. Epub 2017 Oct 4.

Clark AE, Kaleta EJ, Arora A, Wolk DM. Matrix-assisted laser desorption ionization-time of flight mass spectrometry: a fundamental shift in the routine practice of clinical microbiology. Clin Microbiol Rev. 2013 Jul;26(3):547-603. doi: 10.1128/CMR.00072-12.

Clinical and Laboratory Standards Institute. Methods for the Identification of Cultured Microorganisms Using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (M58-Ed1). Wayne, Pennsylvania, USA. 28/04/201

Lagier JC, Armougom F, Million M, Hugon P, Pagnier I, Robert C, Bittar F, Fournous G, Gimenez G, Maraninchi M, Trape JF, Koonin EV, La Scola B, Raoult D. 2012. Microbial culturomics: paradigm shift in the human gut microbiome study. Clin. Microbiol. Infect. 18:1185–1193.

Ojima-Kato T, Yamamoto N, Nagai S, Shima K, Akiyama Y, Ota J, Tamura H. Application of proteotyping Strain Solution™ ver. 2 software and theoretically calculated mass database in MALDI-TOF MS typing of Salmonella serotype. Appl Microbiol Biotechnol. 2017 Dec;101(23-24):8557-8569. doi: 10.1007/s00253-017-8563-3. Epub 2017 Oct 14.

Rhoads DD, Wang H, Karichu J, Richter SS. The presence of a single MALDI-TOF mass spectral peak predicts methicillin resistance in staphylococci. Diagn Microbiol Infect Dis. 2016 Nov;86(3):257-261. doi: 10.1016/j.diagmicrobio.2016.08.001. Epub 2016 Aug 7.

Sauget M, Valot B, Bertrand X, Hocquet D. Can MALDI-TOF Mass Spectrometry Reasonably Type Bacteria? Trends Microbiol. 2017 Jun;25(6):447-455. doi: 10.1016/j.tim.2016.12.006. Epub 2017 Jan 13.

Thouvenot P, Vales G, Bracq-Dieye H, Tessaud-Rita N, Maury MM, Moura A, Lecuit M, Leclercq A. MALDI-TOF mass spectrometry-based identification of Listeria species in surveillance: A prospective study. J Microbiol Methods. 2018 Jan;144:29-32. doi: 10.1016/j.mimet.2017.10.009. Epub 2017 Oct 21.

ISO 10272-2:2017 ISO 10272-2:2017. Microbiology of the food chain -- Horizontal method for detection and enumeration of Campylobacter spp. -- Part 2: Colony-count technique. Publication date: 2017-06.

van Belkum A, Welker M, Pincus D, Charrier JP, Girard V. Matrix-Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrometry in Clinical Microbiology: What Are the Current Issues? Ann Lab Med. 2017 Nov;37(6):475-483. doi: 10.3343/alm.2017.37.6.475.

Zhang T, Ding J, Rao X, Yu J, Chu M, Ren W, Wang L, Xue W. Analysis of methicillin-resistant Staphylococcus aureus major clonal lineages by Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS). J Microbiol Methods. 2015 Oct;117:122-7. doi: 10.1016/j.mimet.2015.08.002. Epub 2015 Aug 5.

 

 

 



View the latest digital issue of FS&T or browse the archive

 

Click here

 
Become a member of the Institute of Food Science and Technology
 

 

IFST Twitter Feed