A few hundred bacteria rest atop a lone grain of sand. Suddenly, a gust of wind scoops up the particle and its occupants, hurtling them toward an unsuspecting plant. The impact causes a tiny abrasion on a leaf and the bacterial passengers disembark to colonize their new home. The plant is ready to defend itself—approximately 15 percent of its genome is dedicated to immune responses. Microscopic pores will clamp shut to prevent bacteria from entering the plant. Proteins will guard particularly valuable molecular targets. Leaves and stems will sacrifice diseased cells to prevent the microbes from spreading. The bacteria won’t back down without a fight, however. They pack syringe-like weaponry that injects dozens of toxins directly into plant cells. These molecules will shut down the plant’s immune system and take control of its cellular machinery. What initially seemed like a quick and easy conquest for the bacteria could go either way. Every day, plants are confronted by a daunting array of pathogens, from microscopic viruses to single-celled protozoa. Generally, the plant emerges victorious. But when it doesn’t, the food supply can be hit hard. “Disease and insects account for up to a 30 percent [agricultural] yield loss globally,” says HHMI-GBMF Investigator Sheng Yang He at Michigan State University. “Every year you can just count on it.” Preventing even a portion of those crop losses is a critical goal for feeding our expanding global population. Much of what occurs in the plant-pathogen arms race is a black box, but slowly, researchers like He are uncovering the details to learn how to enhance plant resistance to invaders. Some of the facts they’re unearthing are already being used to help protect the world’s crops. Cellular Sentries Unlike animals, plants do not have armies of circulating immune cells to fend off invaders. Instead, each cell of the plant must face down every intruder it encounters. As a result, much of the research that goes into boosting plant immunity centers on the events that occur inside the cells that are under attack. Player for HHMI Bulletin website (based on New HHMI.org Player) This video shows what happens to Arabidopsis plants when they are inoculated with Pseudomonas syringae (left pot) or water (right pot) over a five-day period. Video courtesy of Bian Kvitko, Jin Chen, Alec Bonifer, and Sheng Yang He.

Page 2 of 4 Pathogens typically invade by breaching a plant’s outer protective “skin” through surface wounds or via pores called stomata—tiny, mouth-like pores that regulate the exchange of gases between the plant and the atmosphere as part of photosynthesis. These entry points lead to cavities between the cells where gas exchange occurs. It is here that the pathogen encounters the plant’s first line of defense—a series of sentinel-like molecules called pattern-recognition receptors that stand watch on the surface of the plant cell. Jutting from the cell’s membrane, these proteins detect general biological characteristics of pathogens. Known as “microbe-associated molecular patterns,” or MAMPs, the telltale traits can be anything from the sugar molecules that make up a bacterium’s cell wall to the flagellin proteins that form its whip-like tail. “MAMPs are compounds that plants, for the most part, don’t make,” explains Fred Ausubel, a geneticist at Harvard Medical School. “So [plants] can use these molecules to differentiate their own cells from pathogen cells.” After spotting a pathogen’s MAMP, the pattern-recognition receptors will sound the alarm by sending signals to the interior of the plant cell. What follows is a series of events—known as MAMP-triggered immunity—aimed at preventing the invading pathogen from colonizing the plant. For example, stomata, which coat the underside of leaves, close to keep additional microbes from entering the plant. Generally, stomata are closed at night but open during the day, when they provide a perfect entry point for opportunistic microbes. Using the plant model organism Arabidopsis thaliana, Sheng Yang He and his Michigan State team showed that the cells that form stomata contain pattern-recognition receptors in their membranes. When they detect a pathogen, these “guard” cells swell, closing up the pore. He’s group is doing genetic screens to pinpoint the proteins that control this opening and closing, with the aim of eventually fortifying the response in plants. Cristian Danna, a postdoctoral fellow in Ausubel’s lab, recently discovered another MAMP-triggered cellular defense tactic. The presence of bacterial flagellin causes plant cells to suck nutrients out of their apoplastic space—the area between cells. “This makes sense when you think about the mode of infection of most bacterial pathogens,” says Ausubel. “They don’t actually enter plant cells. They grow and multiply in the apoplastic space where they are dependent on the plant for nutrients to grow.” By turning on amino acid and sugar transporters, the plant can starve the bacteria by removing nutrients from where the bacteria live. A Clandestine Campaign If starvation tactics and stomata on lockdown were enough to deter all pathogens, researchers would have a pretty easy time fortifying plant defenses to boost crop yield. However, that’s not the case. In the late 1990s, He and other researchers discovered that some bacteria, such as Pseudomonas syringae, contain a syringe-like apparatus for injecting toxic proteins into plant cells. Called the type III secretion system (T3SS), this impressive machine is also found in animal pathogens. It consists of about 20 proteins assembled into a hollow complex that stretches from a bacterium’s cytoplasm, through its membranes, across the plant’s thick cell wall, and into the plant’s plasma membrane. The resulting channel allows a wave of toxic foot soldiers, known as “effectors,” to march directly from the bacterium into the plant cell’s cytoplasm.

Page 3 of 4 Each of the roughly 15 to 30 effectors that bacteria deliver into the plant cell carries a tag that directs it to a certain cellular location—for example, the plasma membrane, chloroplasts, nuclei, or mitochondria. “It’s like a zip code that gets them to the various places,” explains plant biologist Brian Staskawicz. His group at the University of California, Berkeley, was the first to identify a bacterial effector. Once deployed, the effectors set about terminating MAMP-triggered immune responses, diverting nutrients, and making it easier for the pathogen to colonize the rest of the plant, often by mimicking or inhibiting the cell’s functions. Several effectors from P. syringae target the MAMP receptors at the plant cell membrane. Another effector, called HopM1, goes after an Arabidopsis protein that ups a plant’s defenses, possibly by moving antimicrobial compounds to the cell wall. T3SS may be the chink in a pathogen’s armor that researchers have been looking for. Taking out the syringe-like structure could effectively halt the delivery of all toxic proteins into the plant cell. Many research groups are screening libraries of chemical compounds that could inhibit T3SS. He’s team is taking a different approach, screening 4,000 medicinal plant extracts to see if any affect this particular weapon. “You would think that if the T3SS is so important, and if plants are smart, they would have evolved something to target it for defense,” He says. “The hypothesis may be wrong, but I think it’s worth trying.” If he’s right, his extracts could yield a very effective pesticide. The Permanent Arms Race Once their primary blockades have been breached, plants have a second line of defense: They unleash a squad of “resistance” proteins that disarm the invading effectors. Scientists now have a good handle on the identities of many of these molecules. “People have been selecting for disease-resistant plants since the dawn of agriculture,” explains Jeff Dangl, an HHMI-GBMF investigator at the University of North Carolina at Chapel Hill. “What in fact they were selecting for, and what’s been bred into all of our food, are disease-resistance genes.” These plants have allowed our crop yields to remain high in the face of pathogen invasion, explains Dangl. With the genes that produce disease-resistance proteins in hand, scientists like Dangl hope to deploy them in a more targeted fashion to improve crop yield. Most disease-resistance genes code for a family of molecules called nucleotide-binding leucine-rich repeat receptor, or NLR, proteins. These proteins are found in everything from moss to tomatoes. NLR proteins detect effectors. In some cases, they do it by sensing the action of the pathogen effector on its host target, like a surveillance antenna. In other cases, they bind directly to the effector. Both cases activate the NLR protein, producing a suite of cellular responses that block pathogen replication. “There’s a permanent arms race between the pathogen and the plant,” explains HHMI-GBMF Investigator Jorge Dubcovsky. Pathogens are constantly evolving new effectors and effector combinations that are not detected by NLR proteins, and plants are continuously evolving new NLR proteins that recognize the novel effectors. Dubcovsky, who uses genetics to build stronger varieties of wheat, is entering this arms race. A major project in his University of California, Davis, lab focuses on helping wheat build its defenses against rust fungus. In the 1950s, an outbreak of the fungus wiped out about 40 percent of the U.S. wheat harvest. Scientists have since bred disease-resistant wheat cultivars (strains), but a new race of rust able to defeat the deployed defense genes showed up in Uganda in 1999. The new rust variant, called Ug99, has now spread to South Africa and Iran, among other countries. “This is kind of personal because when I released my first wheat variety as a breeder, the very next year it was destroyed by rust,” explains Dubcovsky. Since then, he’s developed several varieties of wheat containing combinations of resistance genes that protect against current rust races. He’s also cloned two wheat genes—Yr36 and Sr35—that confer resistance to rust. The Yr36 gene, cloned in 2009, has already been added to several varieties of wheat released in California and other parts of the world. Dubcovsky published the identity of the Sr35 gene in Science in August 2013. In the same issue of the journal, Australia’s Commonwealth Scientific and Industrial Research Organisation reported on the identification of another Ug99 resistance gene. Both genes can now be bred into wheat to decrease the chances of a new strain of Ug99 from cropping up. “The combination of rust resistance genes is similar to the AIDS strategy where you attack the pathogen with a cocktail of things that target different viral pathways,” explains Dubcovsky. “It’s very unlikely that the pathogen can mutate simultaneously at all these pathways.” Staskawicz, at Berkeley, is hoping to create a similar cocktail to protect tomatoes against bacterial spot disease, one of the most destructive pathogens affecting field and greenhouse crops. His team is searching for effectors that are conserved in several strains of the bacteria and using them to find the corresponding disease-resistance genes. They’ve also taken a bacterial spot resistance gene from pepper plants and placed it in tomatoes. Field tests in Florida showed promising results. Eventually, farmers growing these modified plants will not have to use harmful copper-based pesticides to protect tomatoes against bacterial spot disease. Tactical Defense Although plant breeders have been tweaking disease-resistance genes to boost plant immunity for over a hundred years by using the simple rules of genetics, exactly how their NLR proteins stop effectors remained a mystery until about 13 years ago. Plants have far fewer disease-resistance proteins than bacteria and other pathogens have effectors, so it’s unlikely that each NLR protein recognizes just one specific effector. In 2001, Dangl and Jonathan Jones, a plant biologist at the Sainsbury Laboratory in the United Kingdom, proposed an alternative theory.