The Newest Technology Coming (Soon?) to a Lab Near You: MALDI-TOF MS and PCR ESI-MS
By Ellen Jo Baron, PhD, D(ABMM), Professor of Pathology and Medicine, Stanford University; Medical School Director, Clinical Microbiology Laboratory, Stanford University School of Medicine, is Associate Editor for Infectious Disease Alert.
Dr. Baron reports no financial relationship to this field of study.
It took polymerase chain reaction at least 30 years (by my reckoning) to evolve from its inception as a promising method for laboratory detection of infectious agents in patient samples to its widespread utilization in diagnostic clinical microbiology laboratories. The speed of adoption of some new technologies, in contrast, appears to be three times faster. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) is one of these.1
As a tool for use in clinical microbiology, the method was first exploited to create a database for identifying anaerobic bacteria back in 2002.2 Today, there are at least three commercial instruments with two different formats available. The Shimadzu FLEXIMASS and Bruker MALDI Biotyper employ MALDI-TOF technology and the Abbott PLEX-ID uses an initial amplification of nucleic acids by PCR, followed by electrospray ionization mass spectrometry (PCR ESI-MS). The Bruker method is fast becoming the standard in European laboratories for organism identification from colonies grown on culture. A few laboratories in the United States have begun to use one of these systems for culture identification as well and, within the next few years, I believe that many of the larger laboratories in the United States will probably be moving from biochemical systems (which require 8 hours to overnight incubation for organism identification by growth characteristics in various biochemical solutions) to the more rapid method of MALDI-TOF MS.
The original investment, on the order of more than half a million dollars, is radically more expensive than buying, say, a Vitek2 (bioMerieux) or Phoenix (BD) or MicroScan (Siemens) instrument, but individual identifications from isolated colonies cost only pennies in reagents for MALDI-TOF, compared to around $3-$4 per organism on a Vitek2 or MicroScan. The PCR ESI-MS final results are quite a bit more expensive (requiring first PCR and more labor and reagents), but results are more complete, too. Of course, susceptibilities will still need to be performed for most organisms for which resistance-determining genetic sequences are not all known, probably on the lab's old Vitek2, Phoenix, or MicroScan. The huge cost and relatively large size of the mass spec instruments will initially limit early adoption, but in time the cost will drop and the instruments will become smaller.
The publication named above, by Schmidt and colleagues, focuses on another exciting use of these technologies: direct identification of bacteria from positive blood culture broths, bypassing at least a day of incubation in obtaining final organism names. This is not the first publication on the topic. In the United States, one laboratory currently uses this technology for routine identification of the organisms in positive blood cultures daily.3 In the near future, a variant of the technology also may be employed to detect some resistance mechanisms.4
MALDI-TOF processing begins with a cell paste or concentrate of bacterial cells in suspension (such as a centrifuged sediment from a positive blood culture broth). The suspension undergoes a simple extraction in one or several steps involving heating and quick centrifugations, and the resulting cell mass is treated with an organic solvent (the matrix) and deposited (or dropped) onto a grid on a plate (which can accommodate a number of extracts at one time), and allowed to dry down. For positive blood culture broths, this may take only 20 minutes of hands-on time. For colonies, a toothpick can be used to spread colony paste on the metal grid plate, and preparation time is less than 5 minutes. The plate is placed into the instrument, the chamber is placed under vacuum pressure, and a laser beam bombards the spots, causing protein molecules from bacterial cell walls and other structures to vaporize and ionize in that state to be dispersed and moved in the vacuum toward a detector. The ions move through the system based on mass and charge, and the ratio of mass to charge determines their speed (or time of flight) to reach the detector, a type of mass spectrometer. The results are displayed as peaks on a mass/charge scale. And the ions move through the system to the detector in nanoseconds! Once the grid has been placed on the instrument the whole process takes no more than 20 minutes, primarily for computer algorithms to run through their paces. And almost every organism species has a unique pattern. Building databases from known organisms is one challenge for the technology, but the current instrument manufacturers have been working on their databases for several years now and the accuracy for organism identification (from colonies) usually is excellent.1,5
The ESI-MS system first requires amplification of bacterial, viral, or fungal DNA using broad primers and then extraction of the DNA. The resulting suspensions (up to 96 per plate) are placed into the instrument, electrospray ionization (ESI) occurs to move ionized nucleic acid particles into the detector module, which also uses mass to charge ratio to develop patterns of recognition. Although in contrast to MALDI-TOF's current capabilities, the ESI-MS can identify mixtures of organisms including viruses and resistance determinants as well.1,4 The key difference from a previously used method that relied on patterns developed by molecules traveling through a matrix via high pressure liquid chromatography (which, although I used it as recently as 1997, now seems positively ancient), known as cell wall fatty acid methyl ester analysis, is that the atmosphere and medium on which the organisms are grown does not influence the results.6 Thus, any colony or any suspension of microbes, including direct specimens from infected sites, can be analyzed by MALDI-TOF if enough organism mass is present in the sample. This is not a limitation of ESI-MS, as the microbial DNA or RNA is first amplified before detection, which does slow down the process and delay final results for up to 6-8 hours.
The MALDI-TOF technology currently requires a minimum cell volume (about 106 organisms in a drop of liquid) to yield an answer, and there are still some technical problems to overcome, but these are bumps on the road. Schmidt and colleagues used blood cultures from two currently popular instruments and broth formulations to compare sample processing methods for testing with the Shimadzu instrument. Unfortunately, media from one of two of the blood culture systems seemed to interfere with performance, and at this time cannot be recommended. Both of the articles by Schmidt and colleagues and Stevenson et al found that the systems had difficulties differentiating viridans streptococci from Streptococcus pneumoniae.3,5 Obviously this must be improved upon. Gram-negative rods seem to be better identified than Gram-positive organisms. The rate of correct identifications from positive blood culture broths from the manufacturer with the broth that worked better was around 72% overall, but this broke down to 87% for Gram-negative bacteria and only 60% for Gram-positive organisms. The Stevenson publication evaluated true positive blood cultures sequentially using the Bruker instrument. They found that 20% of the positive blood cultures had too few organisms to yield any result; however, among the remaining 170 cultures, 95% were correctly identified and those that failed were S. mitis incorrectly identified as S. pneumoniae.3 In contrast, direct colony identifications were 97%-99% correct in at least two recent studies.1,7
In summary, utilization of one of these mass spectrometry methods for identification of all organisms detected in positive blood culture broths is still not ready for routine performance in every laboratory. However, for those laboratories able to pry loose the capital budget funds from their administrators, the use of MALDI-TOF for direct identification from isolated colonies is a real option. Results are available faster and more accurately than biochemical results are now, and I predict that the use of this technology will expand dramatically in the next few years.
- Cherkaoui A, et al. Comparison of two matrix-assisted laser desorption ionization-time of flight mass spectrometry methods with conventional phenotypic identification for routine identification of bacteria to the species level. J Clin Microbiol 2010;48:1169-1175.
- Shah HN, et al. Matrix-assisted laser desorption/ionisation time of flight mass spectrometry and proteomics; a new era in anaerobic microbiology. Clin Infectious Diseases 2002;35:58-64.
- Stevenson LG, et. Rapid identification of bacteria in positive blood culture broths by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 2010;48:444-447.
- Wolk DM, et al. Pathogen profiling: Rapid molecular characterization of Staphylococcus aureus by PCR/electro- spray ionization-mass spectrometry and correlation with phenotype. J Clin Microbiol 2009;47:3129-3137.
- Schmidt V, et al. Rapid identification of bacteria in positive blood culture by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Eur J Clin Microbiol Infect Dis 2011 June 23; Epub ahead of print.
- Sasser M, et al. Identification of Bacillus anthracis from culture using gas chromatographic analysis of fatty acid methyl esters. J AOAC International 2005;88:178-181.
- Kaleta EJ, et al. Comparative analysis of PCR-electrospray ionization/mass spectrometry (MS) and MALDI-TOF/MS for the Identification of bacteria and yeast from positive blood culture bottles. Clinical Chemistry 2011;57:1057-1067.