UW News

August 25, 2017

As Tolstoy noted (sort of), all unhappy microbiomes are unhappy in their own way

tie-dye-microbesThe bacterial communities that live inside each of our guts are relatively similar when times are good, but when stress enters the equation, those communities can react very differently from person to person.

This microbiological version of the “Anna Karenina principle” is a new paradigm suggested by scientists at the University of Washington Bothell and Oregon State University. It may suggest who would benefit most from screens to identify the microbes that reside in their gut, with implications for drug therapy, management of chronic diseases and other aspects of medical care.

On Aug. 24, the researchers published a perspective piece in Nature Microbiology outlining their adaptation of the Anna Karenina principle for the microbial realm. The principle gets its name from the opening line of the novel “Anna Karenina” by Leo Tolstoy: “All happy families are alike; each unhappy family is unhappy in its own way.”

It turns out that this observation applies to perturbed microbiotas of humans and animals. When these microbiotas are unhappy, each is unhappy in its own way.

“This line of thinking started with studies of the microbiology of threatened corals,” said lead and corresponding author Jesse Zaneveld, an assistant professor of biological sciences at UW Bothell. “We found that several stressors made the types of bacteria on corals more variable, allowing blooms of different harmful bacteria on each coral.”

“We were struck by similarities to HIV/AIDs. After HIV suppresses the immune system, patients become vulnerable to opportunistic pathogens — but you can’t predict which one will infect any particular patient. It turns out that this microbial variation is a pattern common to many — though certainly not all — stressors and diseases, and occurs in helpful microbes as well as harmful ones.”

Before joining the UW Bothell faculty, Zaneveld was a postdoctoral researcher at OSU, working with assistant professor of microbiology Rebecca Vega Thurber. It was there that they formulated the idea that microbial communities might behave more in line with Tolstoy’s words than scientists had previously thought.

“When microbiologists have looked at how microbiomes change when their hosts are stressed from any number of factors — temperature, smoking, diabetes, for example — they’ve tended to assume directional and predictable changes in the community,” said Vega Thurber, who is also a corresponding author on the perspective. “After tracking many datasets of our own we rarely seemed to find this pattern but rather found a distinct one where microbiomes actually change in a stochastic, or random, way.”

An undersea coral

Collecting a microbiome sample from a marine coral.Oregon State University

Zaneveld and Vega Thurber worked with OSU doctoral student Ryan McMinds to survey the academic and research literature on microbial changes caused by perturbation. They found those stochastic — or random — changes to be a common occurrence, but one that researchers have tended to discard or bury deep in supplementary materials, rather than highlight in their reports.

“What’s amazing is how obvious these Anna Karenina principle effects are — if you’re looking for them — and how easy they are to miss if you’re searching for a more conventional pattern,” said Zaneveld. “When researchers have reported them, they’ve often assumed that they are a unique quirk of the microbiology of their disease of interest, rather than a more general phenomenon.”

Their work drew together diverse ideas and experiments from microbiome research — including observations from humans and other animals and across multiple human diseases. They propose new methods for analyzing microbiome data to identify situations where the Anna Karenina principle might be at work.

“When healthy, our microbiomes look alike, but when stressed each one of us has our own microbial ‘snowflake,'” said Vega Thurber. “You or I could be put under the same stress, and our microbiomes will respond in different ways — that’s a very important facet to consider for managing approaches to personalized medicine. Stressors like antibiotics or diabetes can cause different people’s microbiomes to react in very different ways.”

Humans and animals are filled with symbiotic communities of microorganisms that often fill key roles in normal physiological function and also influence susceptibility to disease. Predicting how these communities of organisms respond to perturbations — anything that alters the systems’ function — is one of microbiologists’ essential challenges.

Studies of microbiome dynamics have typically looked for patterns that shift microbiomes from a healthy, stable state to a “dysbiotic,” stable state; dysbiosis refers to any unusual configuration of the microbiome with negative consequences for the health of the host. By the Anna Karenina principle, the microbial communities of dysbiotic individuals vary more in composition than in healthy individuals.

The researchers found patterns consistent with Anna Karenina effects in other systems as well, such as the lungs of smokers. Since microbiomes also influence how patients respond to medical drugs, conditions that make the microbiome more variable — such as inflammatory bowel disorders — may also make more variable patients’ responses to drugs from digoxin to asprin.

But, to consider and test these possibilities, scientists must first discuss the Anna Karenina effect among themselves.

“This is the start of a conversation, and not all diseases will show these patterns,” said Zaneveld. “But when you see the same pattern everywhere — from corals enduring high temperatures to wild chimpanzees with suppressed immunity — it suggests we should pay very close attention to the mechanisms that produce it.”

“I hope that by drawing together these research findings from diverse areas, we accelerate the development of common tools and language to understand the role of chance in shaping the microbial part of ourselves.”

The research was funded by the National Science Foundation.


For more information, contact Zaneveld at 425-352-3789 or zaneveld@uw.edu and Vega Thurber at 541-737-185 or Rebecca.Vega-Thurber@oregonstate.edu.

Adapted from a press release by the OSU Office of News and Research Communications.