DNA Detectives: Using Genomic Sequencing in Infection Control

Crime scene investigators use many tools to figure out the events leading up to a crime, how the crime was committed, and how to apprehend the criminal. They draw on their knowledge of physics, psychology, statistics, and many other fields in order to glean as much information as possible from the evidence. In many ways, hospital epidemiologists operate in the same way - with very similar tools. In these cases, the victims of the crime are vulnerable patients. The criminals? Infectious pathogens leading to infection. Today we will explore one tool they share in common: DNA sequencing.

When there is an outbreak in a hospital, or when they are trying to reduce HAIs, epidemiologists must look at many of the same things as a crime scene investigator. The scene of the crime could be the patient room, or the device reprocessing unit, or even a shared common area. The victims could be clustered in the same department, or spread throughout the hospital or even the same network. The criminals, the pathogens, might be endemic - present all the time - or epidemic - present only in outbreaks. Regardless of the setting, patients, or pathogens, bacterial genome sequencing plays a significant role.

DNA sequencing has been around for over 50 years. At first, the methods used were extremely time-consuming and expensive, so only a handful of laboratories could handle sequencing a genome. Today, however, newer methods make sequencing much faster and cheaper, enabling small laboratories such as those found in hospitals to perform genomic sequencing in-house and on a large scale. This has allowed hospital epidemiologists to use DNA to help solve outbreaks of infections just as police departments use it to solve crimes.

Police investigators want to find the individual who committed a crime. Luckily for these investigators, each of us (except for identical twins) has a unique DNA made up of a combination of genes from our mother and father. So a crime investigator can take one DNA sample and match it to one individual, end of story. Hospital epidemiologists don't have it so easy. First, knowing which pathogen is infecting the patient is only the first step. The next step is figuring out where that bacteria came from - and who else might be infected by it. But bacteria split into two identical cells during reproduction - each individual bacterial cell does not have a unique sequence! How can they possibly find out if the infections between certain patients are more related than others?

Thankfully, bacteria are different from humans in another critical way: They reproduce way, way, WAY faster. Entire new generations occur within minutes, compared to 20-30 years for humans. This rapid reproduction means that the slight errors or alterations in tiny fragments of DNA that happen after hundreds and thousands of copies (something called genetic drift) happen far more frequently. Bacteria within one species (say, Clostridium difficile) that are closely related will have fewer of these differences, while those who are very distantly related will have many more differences. Closely-related bacterial samples are considered the same strain of a pathogen, while those who are distantly-related, with more differences between them, would be considered different strains. Both will be the same species, just a different strain. (Think of the diversity within the human race: We are all human, despite tremendous variety in shapes, sizes, and characteristics.)

For example, an epidemiologist sequences three samples of MRSA. She sees that the DNA sequence between two of them is much more similar than the third sample. She can deduce that the related samples share a more recent common ancestor, possibly at the original source of the contamination. The third, more distantly related, may not share that same source, indicating a possible second source of contamination. Now consider if the samples that were related came from patients at opposite ends of the hospital. The epidemiologist, like the crime scene investigator, would then have to look deeply into the records of both patients to see where they overlap - did they both visit the same room? Did they share a healthcare worker? Were they both assessed on the same piece of equipment? In this way, the epidemiologists can begin to hone in on the possible source of contamination - what exactly was the pathway of transmission from one patient to the other. Without genomic sequencing, the hospital epidemiologist would know that there was an outbreak of a certain bacteria, but nothing about how the outbreak samples are connected. The ability to sequence multiple samples of pathogens relatively quickly means that infection preventionists can become DNA detectives, tracking down the evidence that can lead to the eradication of an outbreak.

Whole genome sequencing in epidemiology is a relatively new addition to standard procedures. Newer technologies are cheaper, faster, and even more precise than the earlier methods. New studies are being published every day demonstrating how epidemiologists used whole genome sequencing to stop outbreaks or prevent HAIs. Current research may shed further light on HAIs by helping us learn more about antibiotic resistance, virulence, and modes of transmission. It is an exciting time to be a DNA detective!