Sensitive Microbial Test Identifies Previously Undetectable Bacteria

Spacecraft Assembly Facility at NASA’s Jet Propulsion Laboratory at California Institute of Technology. [NASA JPL/Caltech]

Scientists at the University of California, San Diego (UCSD) have developed a scalable, highly sensitive microbial detection protocol that can identify as few as 50 bacteria present on surfaces in cleanrooms and other low-biomass settings. The technology, which the researchers have called KatharoSeq, has been validated by testing swabs from hundreds of surfaces in three very different low-biomass environments—a spacecraft assembly facility, a neonatal intensive care unit (NICU), and a rare shellfish breeding facility. 

“The more we know about the microbial communities in a given environment, the more likely it is we can reshape them to improve environmental and human health,” comments Rob Knight, Ph.D., professor of pediatrics and computer science and engineering, and director of the Center for Microbiome Innovation at UCSD. Knight and colleagues report on the results of tests with the KatharoSeq technology, in mSystems. Their paper is entitled “KatharoSeq Enables High-Throughput Microbiome Analysis from Low-Biomass Samples.”

Microbial contamination of environments such as hospitals and spacecraft assembly facilities (SAFs) must be monitored, but detecting very low numbers of microbes in low-biomass environments is challenging due to factors including contamination and inefficient testing methods, the authors explain. “Many studies that attempt to evaluate these low-biomass microbiome samples are riddled with erroneous results that are typically false positive signals obtained during the sampling process.”

SAFs are an example of environment where maintaining a very low biomass is critical and cleanroom surfaces must be tested. “The National Aeronautics and Space Administration (NASA) takes extreme steps to avoid the transfer of any terrestrial contaminants to other planets,” the authors write. Other environments that need to maintain a low microbial biomass include neonatal intensive care units (NICUs) and aquaculture facilities.

The UCSD developed their new testing protocol, KatharoSeq (the name is derived from the Greek word katharos, meaning clean, or pure), to combine high sensitivity and low contamination, so that the nature and distribution of even very low numbers of microbes that are present in low-microbial-biomass settings can be evaluated. The testing protocol integrates a sample pooling strategy with positive and negative controls, high-throughput DNA extraction, and magnetic bead–based DNA clean up.

“KatharoSeq consists of a commercial off-the-shelf high-throughput DNA extraction protocol, combined with carefully arranged titrations of positive and negative controls at the DNA extraction and library construction phase to assess cell counts and well-to-well contamination, together with an integrated bioinformatics pipeline for calculating and applying sample exclusion that is compatible with either amplicon sequencing or shotgun metagenomics,” the team explains.

The team suggests that the KatharoSeq protocol results in a level of microbial detection that is about two orders of magnitude better, and a rate of sample processing that is about five-fold higher than standard sampling and sequencing approaches. “We can get results as quickly as 48 hours after receiving a biological sample,” adds first study author Jeremiah Minich, who works in the lab of Dr. Knight and also in that of Eric E. Allen, Ph.D., associate professor at UCSD Scripps Institute of Oceanography. “And we think we could do it even faster once we scale it up and incorporate more automation into the KatharoSeq process.”

The researchers applied the KatharoSeq protocol to test hundreds of swabs taken from multiple sites within three very different low-biomass-built environments: the SAF at NASA's Jet Propulsion Laboratory (JPL) at California Institute of Technology; the NICU at Jacobs Medical Center, UC San Diego Health; and an endangered white abalone-rearing facility at the National Oceanic and Atmospheric Administration’s (NOAA) Southwest Fisheries Science Center in La Jolla, CA.

Of the three facilities, the SAF demonstrated the lowest microbial diversity, but bacteria were still present in cleanrooms at the site, which aim to be completely sterile. Acinetobacter lwoffii, a species associated with human foot traffic, was the most abundant species present among the 32 different types identified in cleanroom areas at the SAF.

A greater number of bacterial species were identified at the 52-room Jacobs Medical Center NICU. Harmless staphylococcal bacteria, including the skin bacterium Staphylococcus epidermis, topped the list of the most prominent types. Interestingly, there was a higher culture rate from surfaces tested in the high-acuity wing of the NICU than in the lower-acuity unit. Tests in one of the NICU rooms identified the bacterium Serratia marcescens. The USCD team was unaware that the infant occupant of the room at the time of testing harbored a lung infection with that bacterium. “Finding S. marcescens in this room in multiple sample types 48 days after the initial positive culture, while not observing the organism in two adjacent rooms, indicates the potential utility of noninvasive built-environment sampling for monitoring and discovering infectious agents in a clinical setting,” the authors write.

“All hospitals have bacteria,” Dr. Knight comments. “But this is the kind of information we don't yet have—which bacteria are found where, and for how long.” Current methods for monitoring pathogens involves culturing patient samples, which is time consuming. “Being able to monitor and predict pathogens by routine, noninvasive sampling of the built environment and sequencing to identify the bacteria, rather than waiting for cultures to grow, could be a useful approach for identifying potential hotspots of transmission that are currently unknown,” Knight notes.

The third facility tested using KatharoSeq was the white abalone-rearing unit at the NOAA Southwest Fisheries Science Center. Here, KatharoSeq detected a diverse microbial community in the shellfish tanks, including symbiotic algae. Encouragingly, the test didn’t detect any of the Rickettsia bacteria that have been largely responsible for declining populations of the white abalone.

“Taken together, our results demonstrate that the KatharoSeq protocol provides compelling microbial community analyses down to limits of detection of 50 to 500 cells in a high-throughput setting,” the authors conclude. “The total processing time from when biological samples are received to when sequencing data are obtained is approximately 48 h. The pipeline can be easily scaled to increase throughput, because multiple steps, including plate loading, DNA extraction, PCR, and sequencing, can be automated by using robotic liquid handlers.”

Knight’s team plans to collaborate with all three facilities to monitor bacterial species over time, which could help scientists understand how and where transmission occurs, which bacterial species are commonly present in different environments, and how long they persist.

Work with the JPL unit aims to map microbial communities over time, including those on the Mars 2020 Rover. The ultimate aim is to send a sterile rover to Mars. In the NICU, clinicians will investigate whether probiotics could help prevent colonization with pathogenic bacteria, both in the built environment and in patients. Knight’s team also aims to work with the abalone-rearing facility to identify the optimum bacterial makeup in the shellfish and their environment, and understand the effects of bacterial colonization on survival of captive-reared abalone released into the wild.