ATLANTA (GenomeWeb) – Metagenomics, and even metatranscriptomics, are enabling clinicians and researchers to tease out pathogens from a mishmash of microbes, according to speakers at the ASM Microbe 2018 meeting.
Sequencing gets to results faster than culturing microbial samples, researchers said during a session held Friday on metagenomics for clinical infectious disease. Researchers are beginning to show that metagenomics can pick up the same microbes that culturing does and may even spot ones missed by culturing. At the same time, metagenomic sequencing can capture which antibiotic resistance genes a sample may harbor — and metatranscriptomics is beginning to give researchers a glimpse into RNA viruses as well as which genes are actually transcribed.
"Microbiology is changing rapidly," session chair Jacques Schrenzel, an associate professor at Geneva University Hospital in Switzerland, said.
Still, both metagenomics and metatranscriptomics have their challenges, including cost.
During his talk, Schrenzel said that metagenomics could be used to identify suspected infections. He recounted a case from 2012 in which a 29-year-old Portuguese woman was found to have an abscess. While her sample tested negative for pathogens by culture and qPCR, the source of the infection was suspected to be Brucella.
Though she was treated, the patient returned four years later with the same symptoms, and once again had negative results by culture and qPCR. With metagenomic sequencing of a liver sample — part of which the patient had removed to treat the liver abscess, he and his colleagues were able to uncover a handful of Brucella reads — at a ratio of 1 Brucella read per 43,000 human reads — that were not present in the controls, suggesting it was the source of her infection.
Similarly, the Mayo Clinic's Morgan Ivy has used metagenomic sequencing to uncover pathogens in patients with infections affecting prosthetic joints. She and her colleagues collected synovial fluid from 168 failed total knee replacements, which they both cultured and subjected to metagenomic sequencing.
Metagenomic sequencing, she reported, homed in on the same pathogen in 73 of the 82 culture-positive cases analyzed. At the same time, it uncovered potential pathogens in four of the culture-negative cases. This, she added, suggests that shotgun metagenomic sequencing could be useful for identifying pathogens, even in cases where culturing could not.
Metagenomics, Schrenzel said, can also be used to dive deeper into a pathogen genome to pick up antimicrobial resistance genes, which can then be used to direct treatment approaches.
At Johns Hopkins University, Patricia Simner has been exploring the use of both short-read and nanopore sequencing to detect antimicrobial resistance genes. Currently, she said, it takes two to three days to complete antimicrobial susceptibility testing, but sequencing — particularly nanopore sequencing — has the potential to shorten that timeframe.
She described the case of a 64-year-old woman who developed sepsis following a liver transplant and who was started on vancomycin and meropenem. Blood cultures then indicated she was positive for Klebsiella pneumoniae,and the samples were sent for further antimicrobial susceptibility testing. But, Simner said, had a metagenomic-based approach been applied first, they could have predicted the bacteria's phenotype some 20 hours sooner and prevented the need for additional testing.
In a further study of resistant and susceptible K. pneumoniae, she and her colleagues found that nanopore sequencing with direct read mapping could uncover bacterial resistance within a few hours.
"Nanopore sequencing has the potential to be real-time," Simner said. It is, however, error-prone, though she and her colleagues have used and have developed tools to predict and spot common errors.
Other researchers are beginning to turn to the metatranscriptome. The University of Sydney's Eddie Holmes has been collecting samples from a range of Australian animals to catalogue the viruses and bacteria infecting them — including influenza among fish, respiratory viruses among humpback whales, and tularemia among ringtail possums.
"RNA sequencing is an extremely powerful tool for pathogen detection," he said. However, he added that it is not cheap and the bioinformatics can be challenging. Still, he's now beginning to apply it to study human diseases, particularly tick-borne ones.