The human gut microbiome is a world in miniature, populated by a chatty community of bacteria, viruses, fungi, and protozoa nestled within various gastrointestinal niches. Over the past decade, researchers have linked disturbances within this complicated microbial society to a variety of diseases. Major depressive disorder (MDD) is one such condition, but the studies have been small and the findings imprecise. A study published December 2 in Science Advances changes all that with its vivid description of a distinct microbiome associated with major depressive disorder, as well as the profile of molecules these organisms produce. The researchers were able to use this microbial “fingerprint” to distinguish between individuals with MDD and healthy controls, solely on the composition of a few microbes and compounds in their fecal matter.

“What this paper does is bring the complexity of the ecology of the microbiome into focus,” says neuroscientist John Cryan at APC Microbiome Ireland and University College Cork who was not part of the study team. “It’s a welcome addition to the field.”

“The strength of the paper is this dual approach to both the metagenomics to identify the key taxa as well as the metabolites, because in the end we need to link whatever the biosignature of the taxa are to the host,” McMaster University’s Jane Foster, who was not part of the study, tells The Scientist. She adds that it’s one of the first studies to conduct metagenomics and metabolomics in the same sample for depression.

Microbiome researchers studying MDD have been using an inexact technique called 16S ribosomal RNA sequencing, which can identify bacteria only down to the genus level within a batch of microorganisms, and it excludes viruses. But psychiatrist Shaohua Hu at Zhejiang University School of Medicine in China and his group wanted a more precise picture of the organisms present, so they gathered fecal samples from 236 people, half of whom had been diagnosed with MDD and were unmedicated, and half who were healthy. They sequenced the total genomic DNA of all the bacteria and viruses in the samples, and then used statistical programs to analyze the differences and similarities between people with MDD and healthy controls.

They found that 18 bacterial species were more abundant in people with MDD (mainly belonging to the genus Bacteroides) and 29 were less common (primarily the genera Blautia and Eubacterium) compared to healthy controls. Hu and his team also found three bacteriophages (viruses that infect bacteria) whose levels were different in MDD versus healthy controls, the first time the virome has been studied in MDD.

Sequencing entire microbial genomes allowed the team to distinguish between organisms that are genetically similar, but functionally very unique, writes University of Melbourne, Australia, researcher Carra Simpson in an email. “Compared to the commonly employed marker gene sequencing approaches, [the authors] have greater resolution to distinguish these species and within a comparatively large sample size.” This helps to “elucidate the functional implications of bacterial alterations on the host.”

To determine the effect of these changes, Hu and the group analyzed the so-called “functional readout”—the molecules the microorganisms produce—of the entire gut microbiome using gas chromatography-mass spectrometry (GC-MS). It turns out that MDD patients harbored significantly more of 16 metabolites and less of 34 compounds than did healthy controls; most of these molecules were involved in amino acid metabolism. The three most important pathways were related to gamma-aminobutyric acid (GABA) metabolism, phenylalanine metabolism, and tryptophan metabolism.

Hu’s team then created a biomarker panel consisting of two species of bacteria, two types of bacteriophage viruses, and two different metabolites. In a separate group of 75 subjects (half with MDD, half healthy controls), the biomarker panel was able to accurately pick out those with depression around 90% of the time.

Possible microbial effects on depression—and vice versa

The researchers point out that GABA is a neurotransmitter in the brain, but it’s also made by gut microbes; fecal levels of GABA and certain of its metabolites were decreased in the MDD patients, and the team also found that GABA-related microbial genes were altered in MDD patients, suggesting that microbes modulate GABA levels. Hu and his team hypothesize that this may dysregulate the function of GABA in the brain, and could lead to depressive symptoms.

In addition, the scientists hypothesize that perhaps the increase in Bacteroides bacteria, which induce cytokine production, could increase inflammation, a condition that has been linked to MDD. Also, decreased Blautia, which has been shown to have anti-inflammatory effects, could contribute to MDD. Other studies have also found that when researchers transplant the entire microbiota of a person with MDD into a germ-free rat, the rat starts to behave depressed.

These are speculative assertions, and Hu says this one cross-sectional study can’t determine causality, and rather, “I think it’s a bi-directional axis. The depressed symptoms can influence our diet behavior, so it can influence our gut characteristics and composition, and also on the other side, our bacteria can produce some special metabolites and have a special pathway that can influence our brain function.”

“This sets the scene for a lot more work to validate whether any of these pathways are actually causally related to depression,” says Cryan, who receives research funding from Dupont Nutrition, Cremo SA, and Nutricia Danone. “There’s a whole emerging field in nutritional psychiatry right now. Can we target the microbiome through diet, could we alleviate some of the effects of depression?”

J. Yang et al., “Landscapes of bacterial and metabolic signatures and their interaction in major depressive disorders,” Science Advances, 6:eaba8555, 2020.