Microbiome Opens Door to Brave New World of Therapeutics

Microbiome researchers are on the threshold of refining a technology that is profoundly more accurate at identifying pathogens than current diagnostics. This increased accuracy should improve the likelihood that any intervention will be successful.

"Right now, we're at the beginning," George Weinstock, PhD, associate director of the Genome Institute at Washington University in St. Louis, Missouri, told Medscape Medical News. "We are eventually going to be able to approach medical problems from a microbiome point of view, which is a much more desirable approach in terms of its precision and in terms of being able, for example, to reserve an antibiotic treatment for situations that can't be easily fixed."

Dr. George Weinstock

Dr. Weinstock discussed how genomics will change infectious disease diagnosis and management at the Future of Genomic Medicine VI meeting in La Jolla, California.

He described how being able to identify single-nucleotide differences between the same species of organism could change medical therapy. "When you have a premature infant who develops a blood infection, we like to know whether the bacteria came from the child's own gut — microbiome — or from the environment," Dr. Weinstock explained.

In the neonatal intensive care unit (NICU) at Washington University, every diaper is saved, bar-coded, and frozen. "If a baby gets a blood infection, we have stool samples before and after the infection and we have stool samples from the same time period and the same vicinity. We can ask: How many nucleotide differences are there between the samples?" Dr. Weinstock said.

He and colleagues Phillip Tarr, MD, and Barbara Warner, MD, assessed 4 episodes of bacteremia caused by group B Streptococcus. They found that bacteria in the blood were identical to bacteria in the stool in these 4 samples. "But you can have other blood isolates not found in stool samples," Dr. Weinstock pointed out. In addition, infants in a NICU who develop bacteremia and those who do not can harbor identical bacteria in their stool — a sign that not all strains of the same organism have equal virulence.

Researchers at Washington University have used whole-genome sequencing to identify single-nucleotide variations within 7 genes encoded in 451 methicillin-resistant Staphylococcus aureus (MRSA) isolates. "Almost all bacteria were identical when subjected to multilocus sequence typing, a standard in the field, based on 7 genes," Dr. Weinstock explained.

However, when the whole genome was analyzed, the group identified approximately 100 nucleotide differences in the bacteria, a signal that not all bacteria thought to be identical are. In all likelihood, some of these differences contribute to an increased virulence of certain MRSA strains — distinctions that would not be detected with traditional diagnostics, he added.

"This is the awesome power of genome sequencing down to the single nucleotide level," Dr. Weinstock said. "These methods are much more powerful than typical diagnostic tests. When DNA sequencing gets faster and cheaper, it will become a routine tool in the diagnostic lab."

Microbial Communities

Researchers are in the process of sequencing entire microbial communities in samples taken from defined body sites to describe the human microbiome. The potential contribution of metagenomics, as this technology is called, is significant. "Children with a high fever who present to the emergency department are treated with various diagnostics. For a fraction of these children, the infecting organism cannot be identified because the causative organisms are not being tested with current diagnostics," Dr. Weinstock explained.

In a study of febrile children, Dr. Weinstock and colleagues Gregory Storch, MD, and Kristine Wylie, PhD, were able to demonstrate that children with fever were more likely to harbor a wide range of viruses, detected with nasopharyngeal swabs and in plasma, than children without fever. "By taking a nasal or blood sample, doing DNA sequencing, and analyzing these sequences, there is the potential to detect organisms that you could not detect with current standard diagnostics," Dr. Weinstock said.

This could spare children inappropriate treatment, especially young children in whom antibiotics should be avoided, he noted. The direct application of microbiome therapy to various clinical scenarios is also in its infancy. One potential application is "fecal transplant," which is given to patients with recurrent Clostridium difficile infections to reestablish a healthy microbiome. In a recent study (N Engl J Med. 2013;368:407-415), the infusion of feces from a donor into recipients suffering from recurrent C difficile infection significantly outperformed conventional antibiotics.

"The jury is still out on this, but it's on track to establish major medical approaches for the treatment of infection by manipulating the microbiome," Dr. Weinstock said.

Microbiome therapy also has a potential role in the treatment of acne. Huiying Li, PhD, from the University of California, Los Angeles, and Dr. Weinstock compared 10 subspecies of Propionibacterium acnes that colonize human skin, and found that 5 of the 10 were present in at least 99% of patients with acne, compared with less than 1% of those without acne. Conversely, 1 subspecies of P acnes was present in 99% of individuals without acne and only 1% of those with acne.

"This suggests that we need to look not only for subspecies that cause disease, but also for subspecies that are protective against disease," Dr. Weinstock said. "Once these sequences are better understood, it will open the way to microbiome therapy — a brave new world of therapeutic intervention."

Dr. Weinstock has disclosed no relevant financial relationships.

Future of Genomic Medicine (FoGM) VI. Presented March 7, 2013.