Why Is There a Need for More Advanced Diagnostic Microbiology Techniques?
Especially in the emergency and critical care patient, timely information on whether bacterial infection is present or not, and if so, which pathogen is present and which antimicrobial agent is expected to have good clinical result, is of vital importance (Kollef et al. 2021). The classic C&ST approach does not meet the current expectations for efficient critical care management. Speeding up the C&ST process can result in shorter hospitalization periods, decreased mortality and more appropriate antimicrobial use. To speed up this process, various tools are currently available, including chromogenic agars, automated detection of positive hemocultures, bacterial identification with MALDI-TOF MS and susceptibility testing directly from positive hemocultures, rapid diagnostic tests able to detect either specific pathogens or resistance mechanisms within few hours (e.g., real-time PCR panels, LAMP, nanotechnology, laboratory-in-a-cartridge solutions; Peri et al. 2021). Even though the venue of these techniques clearly holds promising opportunities, veterinary applications are not always at hand or affordable and time loss due to both logistic (e.g., time needed to get sample to the lab) or organizational issues (24/7 availability in lab) very often prohibit that significant time savings can be achieved. This lecture will focus on MALDI-TOF MS and third-generation sequencing as cost-efficient and broadly applicable rapid veterinary diagnostic techniques that are probably “future-proof.”
Matrix Assisted Laser Desorption and Ionization—Time-of-Flight Mass Spectrometry
MALDI-TOF MS has revolutionized bacterial identification in both human and veterinary medicine. This method identifies bacteria by comparing the “fingerprint” of the masses of a bacterium’s most abundant proteins with a database of fingerprints as provided by the manufacturer and/or in-house created databases. It is a simple, rapid, cost-effective, and robust tool for the identification of most bacteria and fungi to the species level, and sometimes even beyond. Because the fixed costs (mass spectrometer and maintenance) are high, but the consumable cost per sample is very low, the price per sample drops considerably when large numbers of samples are processed. In addition, in critically ill patients, the use of hemocultures can have an added value as specific protocols allow direct identification from positive hemocultures or after short subculture and can result in accelerated identification of sepsis-inducing bacteria (Ulrich et al. 2020).
Even though MALDI-TOF MS is mainly used for identification, it can also be used to detect antimicrobial resistance, using different approaches. First of all, specific proteins (peaks) can be identified that are linked with a certain resistant subpopulation. This peak profiling often has no direct link with the resistance mechanism, so may only predict resistance in specific subgroups of a bacterial species in regions where these subgroups are highly prevalent. Secondly, specific kits allow the detection of strains able to enzymatically degrade antibiotics (for example ESBL or carbapenemase-producing Enterobacteriaceae) in a ± 2-hour protocol. Finally, protocols have been described as able to discriminate susceptible from resistant strains for various antibiotics in various bacterial species after short incubation with the respective antimicrobial agents using 1- to 6-hour protocols (Idelevich, Becker 2021). Increasing quantity and quality of database entries will ensure that identification of bacteria will ameliorate continuously. In addition, machine learning may increase possibilities related to identification, AMR detection and, to some extent, strain typing and detection of host biomarkers (Mortier et al. 2021, Weis et al. 2022).
Third-Generation Sequencing
Third-generation sequencing (3GS) allows high-throughput native DNA or RNA sequencing that can be processed in real-time (i.e., already during the sequencing process), providing an important opportunity to reduce sample-to-result time. Even though there are various providers on 3GS, Oxford Nanopore Technologies currently are most probable to break through in veterinary medicine. The technology is based on the use of very small pores (nanopores) in lipid bilayers for sequencing. As the native DNA/RNA strand passes through the pore, the nucleotides can be identified due to the change in electrical flow and results in very large, sequenced DNA fragments making it easier to “put the genomic puzzle together.” This also means that it is possible to assemble the genomic data of most of the prevalent bacteria in a sample in 1 single sequencing run. This means, no targeted molecular and/or culture diagnostic choices should be made and also metagenomic/viral/parasitic/host-related data may be generated using a single technique in a semi-quantitative way. This may be of special benefit for samples in which there is no clue on which pathogens to expect or when a complex of various viral and/or bacterial pathogens may be involved. Apart from identification of pathogens, 3GS also holds possibilities for DNA-based antimicrobial resistance detection and strain typing.
Even though 3GS might lead to the most comprehensive and most reliable results ever in veterinary microbiology, this technique also holds several risks. First of all, due to the fact that all observed microbial agents that are present in the sample are reported, there is a risk of over-interpretation of obtained results: which of the reported microbial agents are clinically relevant? This is especially important in samples where a microbiome can be expected to be present. In addition, this technique may hold a risk of putting too much focus on the pathogen, while especially in multifactorial diseases also predisposing factors, non-infectious primary problems, etc. should not be neglected. Considering this technique is still in its infancy, there is still a lot of work to be done on setting up and validating workflows. A possible bottleneck may be the bioinformatics expertise for setting up and validating sequence analysis pipelines. In addition, using DNA-based detection of antimicrobial resistance determinants to predict in vivo susceptibility will also require further research and validation. Both hardware and software-related progress will result in decreasing both price and sample-to-result time of 3GS, provided that general laboratory equipment and 3GS specific consumable prices would not increase significantly. Automated and species-specific bioinformatics protocols will be able to increase quality of (deep) sequencing data and shorten processing and interpretation time (Vereecke et al. 2020). In addition, in the near future, it might become possible for veterinary clinics to have their own sequencing station without the need for high investments. While sample preparation and sequencing can be performed on-site immediately after sampling, online access to the bioinformatics protocols of specialized providers with appropriate computational power may allow high quality and even real-time processing and interpretation of the results. Such a workflow might result in a sequencing report, possibly within hours after sampling (Marcolungo et al. 2022). Validation of such point-of-care set-ups and clinically relevant interpretation will however be of major importance.
References
1. Idelevich EA, Becker K. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for antimicrobial susceptibility testing. J Clin Microbiol. 2021;59(12):e0181419.
2. Kollef MH, Shorr AF, Bassetti M, Timsit JF, Micek ST, Michelson AP, Garnacho-Montero J. Timing of antibiotic therapy in the ICU. Crit Care. 2021;25(1):360.
3. Marcolungo L, Passera A, Maestri S, Segala E, Alfano M, Gaffuri F, Marturano G, Casati P, Bianco PA, Delledonne M. Real-time on-site diagnosis of quarantine pathogens in plant tissues by nanopore-based sequencing. Pathogens. 2022;11(2):199.
4. Mortier T, Wieme AD, Vandamme P, Waegeman W. Bacterial species identification using MALDI-TOF mass spectrometry and machine learning techniques: a large-scale benchmarking study. Comput Struct Biotechnol J. 2021;19:6157–6168.
5. Peri AM, Stewart A, Hume A, Irwin A, Harris PNA. New microbiological techniques for the diagnosis of bacterial infections and sepsis in ICU including point of care. Curr Infect Dis Rep. 2021;23(8):12.
6. Ulrich S, Gottschalk C, Straubinger RK, Schwaiger K, Dörfelt R. Acceleration of the identification of sepsis-inducing bacteria in cultures of dog and cat blood. J Small Anim Pract. 2020;61(1):42–45.
7. Vereecke N, Bokma J, Haesebrouck F, Nauwynck H, Boyen F, Pardon B, Theuns S. High quality genome assemblies of Mycoplasma bovis using a taxon-specific Bonito basecaller for MinION and Flongle long-read nanopore sequencing. BMC Bioinformatics. 2020;21(1):517.
8. Weis C, Cuénod A, Rieck B, Dubuis O, Graf S, Lang C, Oberle M, Brackmann M, Søgaard KK, Osthoff M, Borgwardt K, Egli A. Direct antimicrobial resistance prediction from clinical MALDI-TOF mass spectra using machine learning. Nat Med. 2022;28(1):164–174.