Was it Colonel Mustard in the library with a lead pipe? Or Mrs. Peacock in the ballroom with a candlestick? No, it was deadly, drug-resistant Klebsiella pneumoniae from a 43-year-old woman spreading to 17 other patients, killing 6 of them and sickening 5 others, at the National Institutes of Health’s (NIH) Clinical Center in June 2011.

In a biotech version of the classic board game "Clue," researchers from the National Human Genome Research Institute (NHGRI) used genome sequencing to solve the medical mystery of how the infection spread. The story, recounted today in Science Translational Medicine, is a fabulous teaming of classic epidemiological sleuthing and genome sequencing of the pathogen. At the same time, the saga glimpses microevolution in action.

K. pneumoniae is one nasty microbe. It kills half of those it infects. It resists many antibiotics, especially in the drug haven that is a hospital, stays alive on the hands of hospital staff, and lives happily in the guts of some people who are unaware that they are infected, even as they unknowingly spread the infection.

These bacteria are so alike genetically that standard ways to type them are useless. Yet changes in their DNA sequences do arise, even in as short a time as the 4 weeks when investigators probed various nooks and crannies of the index case’s body to sample the bacterial genome. They found telltale single gene variants (SNVs, the same as SNPs) at 41 sites in the 6-million-base genome. Some of the variants may have little or no effect, the consequence of a DNA replication error. It happens. But some may reflect the bacterium mutating towards yet another drug resistance. This is the essence of evolution: genetic change over time that affects the phenotype, thereby offering fodder for natural selection.

To deduce who passed the bug to whom, the researchers, with the aid of an algorithm to sort through the possibilities, compared bacterial sequences. "We thought we could use genome sequencing to tell whether the K. pneumoniae from the first patient was the same strain as the one that infected the second patient," said Julie Segre, who led the team.

The 17 patients fell into 2 groups and one loner. One cluster shared variants with bacteria from the lungs and groin of the index patient (who recovered), another with variants from her throat. The outlier patient got the infection from a contaminated ventilator. To reconstruct these events, because they didn’t connect quite as directly as the suspects in a game of Clue, the researchers sequenced the genomes of all 1,115 patients in the hospital at the time, finding 5 typhoid Marys to fill in the gaps.

And so like the 6 weapons used in 9 rooms of a mansion in the board game, wedding genome sequencing to the who-what-where-when-and-how of an epidemiological investigation can tell infectious disease specialists where to focus their attention.

According to the Centers for Disease Control and Prevention, about 1.7 million people acquire infections in hospitals in the U.S. each year, factoring into 99,000 deaths. "By marshalling the ability to sequence bacterial genomes in real-time to accurately trace the bacteria, our researchers successfully elucidated what happened, which in turn has taught us some important lessons. This study gives us a glimpse of how genomic technologies will alter our approach to microbial epidemics in the future," said NHGRI Director Eric Green.

Added Segre, "Genome sequencing and analysis is our best hope for anticipating and outpacing the pathogenic evolution of infectious agents. Though our practice of genomics didn’t change the way patients were treated in this outbreak, it did change the way the hospital practiced infection control."

And provided a compelling example of evolution happening right now.


Evan S. Snitkin, Adrian M. Zelazny, Pamela J. Thomas, Frida Stock, NISC Comparative Sequencing Program, David K. Henderson, Tara N. Palmore, and Julia A. Segre, Tracking a Hospital Outbreak of Carbapenem-Resistant Klebsiella pneumoniae with Whole-Genome Sequencing. Sci Transl Med 22 August 2012: Vol. 4, Issue 148, p. 148ra116 Sci. Transl. Med. DOI: 10.1126/scitranslmed.3004129