Identifying Pathogens

A mass-spectrometry-based method developed by researchers in London can rapidly identify the species and strain of bacteria and fungi responsible for infections (Anal. Chem. 2014, DOI: 10.1021/ac501075f). Doctors could use such a technique to find the microbial culprits behind life-threatening infections in patients, allowing them to save precious time and pick the most effective antibiotics, the researchers say.

The standard way to identify a microorganism in the clinic is to grow cultures in the laboratory and look for characteristic physical properties, a process that can take days. In recent years, culture-free methods have emerged that identify microbes based on their characteristic 16S ribosomal RNA sequences. However, these approaches often require extensive sample preparation and still sometimes fail to make an unequivocal designation.

Zoltan Takats of Imperial College London, who developed the new method with colleagues, says that mass spectrometry could be a quick alternative to gene sequencing. They envisioned using the technique to obtain profiles of the lipids in different microbes and then using those molecular signatures to identify species.

To test their idea, Takats’s team used a method called rapid evaporative ionization mass spectrometry (REIMS) to analyze microbes grown under a variety of conditions. They picked up colonies they wanted to analyze with a device that looks like a pair of forceps. Sending a current through the device generates a cloud of mostly lipids from the selected colony. The airborne lipids are then drawn into a fluid line embedded in the body of the forceps, directing the molecules to the spectrometer for detection.

A computer program analyzed the spectra and compared the colonies’ lipid profiles, allowing Takats’s team to successfully distinguish 28 species of bacteria and five species of Candida fungi. They could make the identifications regardless of growth conditions and often at greater than 95% accuracy. The identification process, from picking the colony to running the program, took as little as three to five seconds. To see if they could identify different strains of the same type of bacteria, the researchers analyzed seven strains of Escherichia coli and found they could distinguish them with around 90% accuracy.

The technique’s ability to distinguish microorganisms based on their lipid profiles is surprising and exciting, says Charles Y. Chiu of the University of California, San Francisco. However, he points out that the study was done with pure colonies, which have a high concentration of organisms. “In clinical samples, it’s more of a needle and a haystack,” Chiu says.

Chiu also wonders if the method will be able to distinguish between dangerous and harmless versions of the same microorganism. Takats says that is his goal; he is currently working to discriminate between pathogenic and innocuous strains of Clostridium difficile.