Featured Researchers

Robert Dickson, MD


Please describe the research questions of your lab.

My laboratory studies the microbiome's role in critical illness: how communities of bacteria in our lungs and guts contribute to the pathogenesis of sepsis, pneumonia, and the acute respiratory distress syndrome (ARDS). Specific interests include:

  1. The gut translocation hypothesis: how in critically ill patients, bacteria and bacterial products from the lower gastrointestinal tract "escape" through a leaky gut wall and travel to the lungs and other end-organs. We're interested in how this happens, what its biological significance is, and whether there are therapies we can use to prevent it or minimize the harm it causes.

  2. How diseases and therapies alter the lung microbiome, contributing to lung inflammation and injury. Many common exposures in the ICU influence the lung microbiome: antibiotics, aspiration, even stress ulcer prophylaxis. My laboratory studies how changes in respiratory microbiota contribute to lung injury. A specific focus is hyperoxia: oxygen is the most commonly administered drug in the ICU, and we have recently discovered that it has profound effects on lung microbiota, potentially contributing to lung injury.

  3. Bringing molecular microbiology to the bedside. Until recently, the sequencing techniques we use for microbiome research have been impractical for clinical use: too expensive, too slow, too informatically-demanding. But that is changing, and quickly. We are now using handheld sequencers that can perform real-time metagenomics on patient specimens within hours, for less than the cost of a multiplex PCR. My laboratory is working on protocols and analytic tools to put these sequencing tools to clinical use.

What genetics/genomics techniques do you utilize in your lab?

Our workhorse is bacterial community sequencing using 16S rRNA gene amplicon sequencing. This is the primary approach used in most microbiome studies. We first use PCR to amplify a short stretch of the bacterial genome; then we sequence those amplicons and figure out which bacteria they represent. This gives us a "population survey" of the bacteria present in a specimen. It's a powerful and attractive first step for understanding bacterial communities: it's high-throughput, it's relatively cheap, and it has taught us a tremendous amount about how bacterial communities change in disease and in response to our therapies.

Increasingly, my lab has been using nanopore sequencing to characterize specimens. We use the MinION (Oxford Nanopore) to perform true metagenomics (no amplification step) directly from patient specimens. This is exciting, because it gives us a tremendous amount of genomic data in real time, bringing us one step closer to clinical relevance. Lung specimens are challenging to study with metagenomics because the ratio of host DNA to microbial DNA is so high (even in infection), but there are a lot of appealing features to real-time, long-read metagenomic sequencing.

Describe a key technique/assay/instrument utilized in your lab, and what novel insights does it bring to your research question?

One important realization we've had recently is that we need to consider absolute abundance of bacterial DNA, not just relative abundance. The community sequencing techniques we use tell us a lot about who is there, but nothing about how much is there. This is like knowing that Denver and Ann Arbor have similar racial and ethnic demographics, but not knowing that Denver is seven times the size of Ann Arbor.

We've recently been addressing this by quantifying the bacterial DNA burden in respiratory specimens using droplet digital PCR. This is an ultrasensitive quantification platform that is already in common use in other fields (e.g. in oncology to measure circulating tumor DNA in blood). We had previously used conventional qPCR, but with respiratory specimens we are dealing with very small quantities of bacterial DNA, right around the limit of detection. ddPCR is giving us much better resolution.

Incorporating total bacterial DNA burden into our analyses has been quite clarifying. Among other advantages, it helps us determine whether specific bacteria we find are real or instead due to reagent contamination (a major problem in low-biomass microbiome studies).

At what point in your life did you decide you wanted to be a scientist/physician?

I studied classics and liberal arts in college, and especially loved thinking about the history of progress in scientific disciplines. Within Biology, I was most excited by Microbiology and the idea of homeostasis, both of which had obvious applicability in medicine. (Sure enough, I now make a living studying how microbiota disrupt homeostasis in the lungs.) While I've loved patient care and clinical education from the start, my favorite part of the job has always been untangling complex pathophysiology to make sense of why patients get sick. It's such a thrill and privilege to be able to ask and answer fundamental questions of disease pathogenesis. It really takes a lucky combination of resources, training, teammates, and collaborators. I actually didn’t like the preclinical/basic science part of med school; it seemed like a dry recitation of established facts. I wish we did a better job of giving students the experience of actual scientific discovery. I suspect we’d have a lot more physician-scientists.

In your opinion, what is one of the most important discoveries in the field of respiratory illness/disease/function that was dependent on genomics or similar techniques?

My entire field (lung microbiome studies) owes its existence to sequencing technology. For over a century, we were taught that the lungs are sterile. It was the advent of culture-independent microbiology (largely the amplicon sequencing I described above) that revealed that even healthy lungs have diverse and dynamic communities of bacteria. We've spent the past decade learning to think ecologically about the lungs. None of this would have happened if we were still restricted to microscopy, cultivation, and PCR.

Briefly describe your favorite publication involving genomics/omics that you were involved with.

I'm proud of our paper showing that the lung microbiome is enriched with gut bacteria in sepsis and ARDS (Nat Microbiol. 2016 Jul 18;1(10):16113.).

We've known since the 1950s that somehow the bacteria in our lower guts play a role in why patients develop lung injury in shock. For decades, it was speculated that bacteria must translocate across the gut wall and travel to the lungs, but this idea was largely abandoned in the 1990s after studies failed to detect bacterial translocation in at-risk patients. But these studies were limited by their use of standard culture-based techniques, which lack the breadth of what we can identify with our sequencing tools.

In our paper, we showed that both in animal models of sepsis and in human patients with ARDS, we could detect bacteria from the lower GI tract in the lungs. This observation re-ignited interest in the gut translocation hypothesis. In a subsequent paper we showed that we can also detect gut bacteria in brain tissue in shock (Am J Respir Crit Care Med. 2018 Mar 15;197(6):747-756.), and another group has found that Enterobacteriaceae in the lungs is associated with development of ARDS (Am J Respir Crit Care Med. 2018 Mar 1;197(5):621-631.) We've got promising data suggesting that gut translocation is relatively common, correlated with severity of illness, and may be modifiable by therapies.

I think critical care research needs fresh ideas, and the one I'm selling is that the three pounds of bacteria in our body may have something to do with why our organs shut down when we get sick. For members of the Genetics and Genomics section, I'd point out that there's far more genetic diversity in our patient's microbiota than within their own chromosomes.

What is your favorite aspect of ATS?

It's got to be the people: friends and collaborators, fellow investigators. There aren't a lot of people who study the lung microbiome, but I count on seeing nearly all of them at ATS each year. I get a tremendous amount out of hearing their perspectives and comparing notes on what we're working on. My head is always buzzing with new ideas when I head home

How could your research assist scientists and clinicians in other assemblies at ATS?

The microbiome either influences or is influenced by nearly everything related to Pulmonary and Critical Care Medicine. Our lungs evolved to cope with and adapt to constant microbial exposures, and I don't think we can understand lung immunology or cell biology without considering the microbial milieu. Our recent work has been exploring how variation in lung immunity is explained by variation in lung microbiota (Am J Respir Crit Care Med. 2018 Aug 15;198(4):497-508.). For clinicians, my hope is that before too long we can put these sequencing tools in their hands, along with a means of coherently interpreting the results.

Contact information for Dr. Dickson:  rodickso@med.umich.edu