Skip to main content
Premium Trial:

Request an Annual Quote

NIH Team Using MALDI Biotyper to Identify Antibiotic-Resistant Bacteria


NEW YORK (GenomeWeb) – Researchers from the National Institutes of Health have managed to identify antibiotic-resistant bacteria using Bruker's MALDI Biotyper clinical microbiology system.

Detailed in a study published this September in the Journal of Clinical Microbiology, the work represents an advance in application of MALDI mass spec to clinical microbiology. While Bruker's MALDI Bioyper and BioMérieux's Vitek MS systems have for several years now been used for microbial identification in clinical settings, they have made less headway in detecting antibiotic resistance in these organisms. 

Significantly, the NIH researchers were able to detect antibiotic resistance using data already collected by the MALDI Biotyper in the course of microbial identification work. And while the method is applicable only to certain cases of resistance, in these cases it provides clinicians with information they can use to, for example, guide patient therapy or implement infection control programs, Anna Lau, an NIH scientist and senior author on the paper, told GenomeWeb.

The NIH researchers' approach is able to detect resistance to carbapenems, a class of antibiotics commonly used in hospitals and in patients with infections already resistant to other antibiotics. A common agent of carbapenem resistance is the blaKPCgene, which encodes the Klebsiella pneumoniae carbapenemase (KPC) protein, which hydrolyzes these antibiotics.

Other researchers have attempted to use MALDI mass spec to detect the hydrolysis products generated by KPC, however, Lau and her co-authors noted in the JCM paper, "there are limited data on the clinical utility of these assays to date."

The NIH team, on the other hand, detect KPC-mediated resistance via a proxy, the protein pKpQIL_p019, which is encoded by a gene found on some blaKPC-containing plasmids and which can be identified in spectra collected by the MALDI Biotyper during routine identification assays.

"If [the p019] protein peak is present in the data that we have already collected we can infer the presence of the p019 protein and that then allows us to infer, at least in our dataset, the presence of the blaKPCgene," said John Dekker, a colleague of Lau's at NIH and co-author on the paper.

"The beauty of this is that you can look for the p019 peak in spectra that is already being collected as part of your routine identification," Lau said.

The NIH researchers first presented the link between p019 and blaKPC in another JCM paper published last year. In that paper, they followed a blaKPC-containing plasmid involved in a Carbapenem-resistant Enterobacteriaceae outbreak at the NIH Clinical Center in 2011, discovering that the p019 protein was detectable via MALDI-MS in all 18 isolates from the outbreak that they analyzed.

In their recent study, the researchers followed up on this finding with a blinded analysis of 140 previously characterized Enterobacteriaceae representing eight genera and at least 14 different species along with a retrospective analysis of 720 clinical Enterobacteriaceae spectra.

In the case of the 140 Enterobacteriaceae, the researchers found that all 26 isolates that were positive for p019 also contained blaKPC and were carbapenem resistant. Additionally, the mass spec assay did not detect p019 in any isolates where the presence of the gene had not been confirmed by PCR.

In the retrospective analysis, they detected what appeared to be p019 in nine of the 720 spectra, seven of which were actually from p019-containing, carbapenem-resistant isolates.

The approach was not, however, able to detect resistance in Enterobacteriaceae that did not code for p019. For instance, while in the 140-sample set only 26 isolates were positive for p019 (which the researchers confirmed via PCR), the set contained a total of 86 carbapenem-resistant isolates.

This, noted Dekker, is the drawback of relying on a proxy as opposed to direct detection of the actual molecule responsible for resistance.

"In terms of how many blaKPC-carrying plasmids contain p019, we don’t have a general answer to that," he said. "We know that certain blaKPC-carrying plasmids contain p019, but many don't."

Additionally, there is the question of whether the protein is associated with other carbapenemases besides blaKPC. "The association that has been found with the p019 protein is so far only for the blaKPC gene," Dekker said.

Even so, Lau said, the method's ability to quickly identify some carbapenem-resistant organisms in real time using only the mass spec data collected as part of routine identification assays is a boon for clinicians. She and her colleagues have written a script that automatically searches their clinical MALDI spectra for the presence of p019 that she said they use routinely at their facility.

While traditional susceptibility testing is still required to confirm the results, the assay "means that physicians can act immediately to ensure that the patient is receiving appropriate antimicrobial coverage and infection control practitioners can rapidly put strategies in place to help minimize the spread of resistant bacteria in the hospital environment," she said.

The researchers are also interested in finding additional proteins that could similarly serve as proxies for other agents of antibiotic resistance, Dekker said. "Ideally, you would want to be able to detect [mass spec] peaks that are fragments of the actual molecules that are mediating resistance, but the next best thing is to detect molecular proxies."

Detecting the actual mediators of resistance themselves is a challenge due to the technical limitations of the MALDI instruments used in Bruker's and BioMérieux's respective clinical microbiology platforms, Dekker noted. These instruments, he said, are most effective at detecting proteins 20,000 Daltons and smaller, but many of the proteins that confer antibiotic resistance are larger than this. And because these platforms typically look at intact, rather than digested proteins, this has proven an obstacle.

One possible way around the issue would be to look at smaller proteolytic fragments of the proteins involved in resistance, Dekker said, adding that this is something he and his colleagues "are very interested in exploring."

Lau and her colleagues have collaborated with Bruker in the past on developing methods for MALDI-based identification of microorganisms, most notably on a database for mold identification.


The p019 work was done independently, though, Dekker said, adding that the method is freely available to any other laboratories that are interested in implementing it.

Both Bruker and BioMérieux have expressed interest in developing methods for detecting antibiotic resistance using their MALDI systems. For instance, Bruker's MALDI-Biotyper-Spectrum-Beta-Lactamase, MSBL, workflow, allows researchers using the research-use-only version of the system to perform functional beta-lactam antibiotic resistance testing for selected antibiotics. The company is also looking into using the Biotyper to detect other forms of antibiotic resistance, such as those mediated by mechanisms like efflux pumps and porin changes.

BioMérieux researchers have presented data on use of the RUO version of that company's Vitek MS instrument for detection of beta-lactam antibiotic resistance.

Beyond these MALDI clinical microbiology platforms, a number of other firms have developed or are developing molecular tests aimed at detecting antibiotic resistance, and carbapenem-resistant Enterobacteriaceae, in particular.

For instance, Dutch firm Check-Points offers a multiplex PCR test for detection of a variety of genes that confer carbapenem-resistance. As previously reported by GenomeWeb, the company also plans next year to file for US Food and Drug Administration 510(k) clearance for an expanded version of the test.

Cepheid also offer a PCR-based CRE test, its Xpert CARBA-R product, which, the company said, is able to rapidly detect five types of CRE in around an hour. The company received a CE mark for the test in 2014 and submitted it for FDA clearance in September.

Other firms including BioFire Diagnostics, Great Basin, Becton Dickinson, Renishaw, Mobidiag, and Curetis also offer or are developing PCR-based tests targeted CRE.