What does your stomach have in common with the roots of a plant? Both are home to a complex and distinctive community of microbes, many of which benefit their hosts by boosting immunity and health. Collectively known as the microbiome, these microbes can help plants fend off disease, stimulate growth, crowd out space that would be taken up by pathogens, promote resistance to drought, or influence crop yield.
For years, microbiologists have investigated these stowaways one at a time—think back to high school biology, and you may remember the classic example of symbiotic nitrogen-fixing bacteria that take up residence in the roots of legumes. But thanks to a dramatic decline in the cost of sequencing, researchers are now taking a comprehensive look at the community. Similar to recent efforts to catalog the more than 10,000 microbial species living in humans, these studies provide a systems-level view of what has been called the second genome of plants. The discoveries could fuel advances in agriculture and medicine.
“The era of the microbiome is upon us,” says Jonathan A. Eisen, a microbiologist at the University of California, Davis. Only in the past year, he explains, has high-throughput sequencing become so cheap and fast that a genomic survey of the thousands of bacterial species associated with plants is feasible. As a result, microbiologists have begun treating this community as a trait of the host, similar to height, yield, or carbon fixation. Eisen, who also researches the human microbiome, sees plants like corn and rice as great model organisms for investigating how the genetics of a host influences its microbial community.
In the most in-depth examination of a plant microbiome so far, two research groups recently surveyed the bacterial community associated with what is sometimes termed a lab rat of the plant world, the flowering plant Arabidopsis thaliana. A group led by Jeffery L. Dangl of Howard Hughes Medical Institute and the University of North Carolina, Chapel Hill, took a census of the bacteria within and around the roots of more than 600 of these plants, as well as in the soil surrounding them, by sequencing the 16S ribosomal gene of each of the bacteria (Nature, DOI: 10.1038/nature11237).
As they had expected from earlier studies, Dangl’s group found that only a subset of the bacterial community in the soil is present around the plant’s roots, in a region called the rhizosphere. An even more select community occupies the roots themselves. The main root inhabitants represent two bacterial phyla—Proteobacteria, which includes many growth-promoting members, and Actinobacteria, many of which are known for producing antimicrobial compounds. The plant may be “recruiting them as guard dogs,” says Derek S. Lundberg, a graduate student in Dangl’s lab and the lead author of the study.
Dangl and colleagues then looked at the differences in the microbial communities from Arabidopsis growing in two soils that differ in mineral and nutrient composition. They found a group of root bacteria common to both soil types as well as some groups enhanced in each soil. They hypothesize that Arabidopsis recruits a core group of microbes to benefit its basic functions, and an additional subset to help it thrive in specific environments. Similar results were found in a companion study led by Paul Schulze-Lefert of Max Planck Institute for Plant Breeding Research (Nature, DOI: 10.1038/nature11336).
The method highlighted in the Arabidopsis microbiome studies is also being applied to major crops with a goal of improving their yield or resistance to disease or drought. Ruth E. Ley at Cornell University is probing the corn microbiome, as is Eisen, who is also studying rice. Philip Hugenholtz of the Australian Centre for Ecogenomics is looking at sugarcane. And the Joint Genome Institute of the U.S. Department of Energy, which carried out the sequencing in Dangl’s Arabidopsis study, is investigating the microbiome of plants with potential as biofuels, including poplar, in a study led by Gerald A. Tuskan at Oak Ridge National Laboratory.
Many plant microbiologists dream of discovering novel plant genes that recruit beneficial microbes, explains Lundberg. Such genes would be attractive targets to cross into a plant or to introduce transgenically to enable the plant to better use its environment, he says.
To his surprise, the two recent Nature studies found that Arabidopsis plants with different genetic backgrounds appear to have very similar microbiomes. However, the sequencing technique used in the studies has a blind spot, he notes: Bacteria with very different physiology can share an identical 16S gene sequence. Eisen says metagenomic studies, which can distinguish differences in closely related bacterial strains, offer a way forward in the quest for novel microbe-curating plant genes.
Beneficial bacteria discovered through these studies could be added to seeds as a probiotic, says Lundberg. This modern type of bacterial amendment would build on established farming traditions. “Farmers have been paying attention to the microbial qualities of soil for a long time,” Eisen notes.
Plant microbiome studies are also driving the discovery of new compounds with potential applications in agriculture or medicine. “Whether you care about microbes or not, you can use this ecosystem to find out about small molecules that manipulate host biology,” Eisen says. “If there are 2,000 species of bacteria on maize roots, plus many on the leaves and stems, and within each species, hundreds of thousands of types, and each microbe produces about 2,000 proteins, the potential for discovery here is just astonishing. That’s why we’ve moved from studying single organisms and a host to communities.”
Although the genomic tools available are powerful, “the challenge is complexity,” Eisen notes. The “conversation between bacteria and plants,” as Yale University microbiologist Jo Handelsman calls it, involves a lot of back-and-forth communication. For example, researchers have fleshed out that old story from high school biology about legumes and nitrogen-fixing bacteria to reveal an intricate dialogue.
Legumes start the conversation by secreting flavonoid compounds from their roots. Rhizobia bacteria respond to these signals by releasing chemicals called Nod factors, whose sulfated form causes changes in gene expression in the legume’s roots that eventually result in the bacteria being engulfed inside a root nodule. Once there, the bacteria fix nitrogen, providing a critical nutrient for the plant, in return for food from the plant’s roots. “It’s an exquisitely tuned system,” Handelsman says. “I’m sure there will be a lot of others like that.”
Handelsman’s lab studies plant-microbe interactions by making targeted knockouts in plant-associated bacteria to find clusters of genes responsible, for example, for a change in the mixture of small molecules released by a plant’s roots. But the cross fire of interactions can make these systems difficult to study. “A change in one microbial metabolite will change the profile of metabolites from the plant,” she says, and this can create a feedback that then affects the microbial metabolites.
Research into these microbial metabolites could pave the way for the discovery of natural products with potential medical applications, according to Handelsman. “Plants are amazingly prolific producers of metabolites,” she says, pointing out that the tobacco plant alone produces about 2,500 different ones. The difficulty is finding a way to analyze them all.
Recently, Pieter C. Dorrestein, a chemist at the University of California, San Diego, and collaborators have applied new mass spectrometry approaches to track the production of metabolites by a microbial community in real time and space. The first uses imaging mass spectrometry to make two- and three-dimensional maps of molecules present in a microbial community in petri dishes.
The second can probe live microbial communities on any surface, from an agar plate to the leaf of a plant, to monitor and generate a map of their molecular products grouped by structural similarity. The method combines nanospray desorption electrospray ionization mass spectrometry, or nanoDESI MS, with spectral MS/MS network construction (Proc. Natl. Acad. Sci. USA, DOI: 10.1073/pnas.1203689109).
Together, they are “some of the first tools that allow us to capture the complexity of the microbial community at the molecular level,” Dorrestein says.
Dorrestein and colleagues have used the nanoDESI spectral networking technique to characterize a bacterial molecule in soils that protects sugar beets against fungal infection. Other researchers had tried to identify the compound for years using gene knockout methods. With the new technique, “we found it in 10 minutes,” he says. The lipopeptide, thanamycin, has since shown potency against cultured leukemia cells.
New compounds must still be examined by nuclear magnetic resonance spectroscopy to determine their structure, creating a bottleneck in the discovery process. “We need chemists’ ingenuity to make these systems easier to study,” Handelsman says.
A further complication arises because it’s difficult to determine which compounds are made by a plant and which by its many microbes. Researchers often narrow the field by studying compounds produced by the isolated plant and cultured bacteria. But after that, they rely on their knowledge of typical plant or bacterial metabolisms, or generate a mutant in the pathway to synthesize the compound of interest.
Despite the potential of these new compounds, funding limits large-scale discovery projects in plants, Handelsman says, in contrast with greater funding of medical research.
“Agriculture suffers from its own success,” she says. “We feed people so well and can buffer against epidemics. We don’t look at our food supply as under threat.”