It goes without saying that both patients and their doctors want to treat infections as quickly and successfully as possible. Unfortunately, when it comes to traditional diagnostic methods, culturing a microorganism can take several days – time during which the microbes may be multiplying and the infection spreading. Slower-growing organisms, such as fungi, can take weeks to identify with culture methods. Acquiring the resistance profile of these organisms takes additional precious time, during which the patient caregivers must make their best guess at the infection source and susceptibility.
Clinical microbiologists are working hard to change this turnaround time. Two clinical microbiology experts, Dr. John Dekker and Dr. Anna Lau, have recently published results in the Journal of Clinical Microbiology suggesting mass spectrometry (often referred to as mass spec) may help improve clinical diagnostics. Dekker and Lau, the codirectors of Bacteriology, Parasitology, and Molecular Epidemiology at the NIH Clinical Center Department of Laboratory Medicine, have used mass spec to simultaneously identify microorganisms and a resistance-associated marker.
Many technologies, including mass spec, have advanced the speed of organism identification. Mass spec detects an organism’s protein fingerprint that is then challenged against a microbial database to identify the organism. While this method does require culturing of the samples, “as soon as a single colony is growing on the plate, it can be picked and analyzed on instrument,” explains Lau. “Our particular method here uses mass spec to get rapid identification, and on top of that, we have found a marker – a particular protein marker – that is associated with a plasmid that can carry genes that confer antibiotic resistance.”
“We are studying a protein biomarker detectable by mass spectrometry, which can be used in certain instances as a proxy for resistance to a class of antibiotics called carbapenems,” explains Dekker. The proxy protein used, p019, is encoded by a gene present on a group of plasmids that also carry the blaKPC gene that confers the actual resistance. “By detecting the spectral peak associated with this protein, we can infer the presence of the carbapenemase gene in certain cases,” he continues.
The need for a proxy protein biomarker, rather than detection of the carbapenemase protein itself, is due to instrument constraints. The carbapenemase protein is much larger than the 2000-20,0000 dalton size that the instrument used, called a MALDI TOF mass spectrometer, can detect. “If we could detect the carbapenemase itself, that’s what we would want to do,” says Dekker. Because of its smaller size, p019 is a better candidate for detection using MALDI TOF mass spectrometers.
Dekker and Lau used a collection of clinical samples to validate the clinical application of the method they developed in 2014, also published in Journal of Clinical Microbiology. The collection included 86 carbapenem-resistant bacterial isolates, of which two-thirds carried the blaKPC gene. Of these, half also carried p019. All but one of these p019-positive strains were detected by two independent technicians. This exception highlighted the importance of proper machine settings, as the experiment was repeated after the routine annual instrument maintenance. Once the settings had been recalibrated, all p019-positive strains were correctly identified.
The ability of simultaneous culture ID and resistance profiling is appealing, but so far remains limited to recognizing a specific cause of carbapenem resistance. “Remember, carbapenem resistance can be caused by a very large number of different types of mechanisms. Classes of carbapenemases other than KPC, like NDM or VIM or IMP, are not necessarily associated on a plasmid with the p019 protein,” explains Dekker. Molecular epidemiology studies can be used to complement this diagnostic technique. “The likelihood of capturing a carbapenem-resistant isolate by just looking for p019 will depend on the exact geographic distribution of plasmids and the association of the p019 gene with the blaKPC gene in those plasmids.”
However, for those p019-positive samples, the identification of antibiotic resistance comes at the same moment as bacterial identification. “The beauty of our method is that we are already collecting spectra as part of our routine organism identification for clinical care,” says Lau. “We built an automated program in the computer that will scan all the collected spectra in real-time and flag those that contain a peak similar to the size of p019. This then triggers the technologists to perform additional tests to confirm the presence of a carbapenemase and antibiotic resistance. What we highly stress is that the presence or absence of the p019 peak does not negate the need to do antimicrobial resistance testing; that is still the gold standard in clinical micro for providing standardized information to the care provider for how to treat the patient. What is advantageous about our method is the potential for earlier detection of an organism that may be carbapenem resistant.”
Both Lau and Dekker are quick to emphasize the collaborative efforts of the clinical microbiology community toward this goal. “There were many people involved in the development of the method: sequencing, proteomics – it was a group effort,” says Dekker. “There are scientists in many, many groups around the world who are trying to push the capabilities of this technology just a little bit further,” adds Lau.
Despite worldwide efforts to increase its capabilities, the use of mass spec for identification of infectious microbes is still “the new kid on the block,” says Lau. Two recent FDA approvals of mass spec for rapid identification of bacteria and yeast have increased its popularity in the United States and around the world. One hurdle is the initial outlay: the instrument cost is higher than differential agars or even thermocyclers for PCR, but the savings add up when considering rapid turnaround time, better patient care, decreased length of stay, and better infection control. “It all adds up in the end as being more cost effective,’ reasons Lau.
Now the challenge is to expand the capabilities of this machine even further. The use of mass spectrometry for drug resistance was not the original purpose of these instruments. The MALDI TOF mass spectrometer gives very reliable identification of microorganisms, but it’s a crude way of really measuring protein spectra, getting a fingerprint, and matching that against a database. Identifying peaks that can give any additional information will help those suffering infections, the goal of all clinical microbiologists. “It’s an exciting time,” says Lau. “We’ve found one protein that can be used as a marker. We are definitely hunting for more. There is more to come.”
-- Julie Wolf