LOV story

Even nonphotosynthetic bacteria respond to light in surprising ways. Have scientists found a new ubiquitous signaling mechanism?

By Alla Katsnelson

Original image: © Yves Brun / artistically modified by The Scientist

aulobacter crescentus isn't much to look at. When you peek through a microscope at 630-times magnification, the freshwater bacteria appear as a swarm of little gray, kidney-shaped creatures flitting about in the drop of medium between slide and coverslip.

Sometimes, though, one member of the gram-negative crew will adhere to the nearest surface, exchanging its flagellum for a stalk and producing what is considered to be one of nature's strongest glues. Gradually, other individual cells will glom onto the Caulobacter stalk cell, often sharing each other's points of adhesion to form rosettes—flower-like shapes in which the bacteria extend out like petals—that, under certain conditions, expand to become biofilms.

The cells' penchant for adhesion is distinctly—and surprisingly—regulated by...

Sometimes, though, one member of the gram-negative crew will adhere to the nearest surface, exchanging its flagellum for a stalk and producing what is considered to be one of nature's strongest glues. Gradually, other individual cells will glom onto the Caulobacter stalk cell, often sharing each other's points of adhesion to form rosettes—flower-like shapes in which the bacteria extend out like petals—that, under certain conditions, expand to become biofilms.

The cells' penchant for adhesion is distinctly—and surprisingly—regulated by light. Caulobacter is a chemotrophic organism, that is, it gets its energy from electron chemical reactions such as oxidation, not from sunlight. Biologists using it as a model organism for everything from cell signaling to aging have never considered light to affect its function one way or the other. But after the discovery of a new type of photoreceptor in plants a decade ago, microbial researchers began to look for light's effects in unexpected places.

Related Articles

1 The group called the receptor phototropin, and showed that it does its business via a segment on its N terminus termed a LOV domain, a highly conserved sequence that responds to light, oxygen, or voltage (i.e., LOV).

Three years later, in May of 2000, Italian biophysicist and photobiologist Aba Losi heard Briggs describe the photochemistry of the new receptor at a conference, and right away, she recalls, she had an intuition. "I immediately thought that this new and fascinating paradigm of sensing and reacting to light was something both elegant and simple, so simple that it must be very ancient," she wrote in an email.

It was the early days of the new genomic era, and researchers were just beginning to sequence bacterial genomes. A few months after her Eureka moment, Losi, based at the University of Parma, began surfing through sequence databases and soon identified three species of chemotrophic bacteria that carried conserved LOV domains: Bacillus subtilis, Caulobacter crescentus, and the cyanobacterium Synechocystis sp. PCC6803.

Bacteria growing under strong blue light irradiation. Courtesy of Ulrich Krauss

Losi's hunch was vindicated. Still, a DNA sequence was one thing, but were these domains physiologically active as light sensors? Out of her three hits, Losi singled out Bacillus, in which the LOV domain, found within a gene called YtvA, shared the strongest homology with the plant phototropin gene, and began to investigate the question. Working with Max Planck Institute for Bioinorganic Chemistry in Mülheim, she expressed YtvA in E. coli and demonstrated that the protein underwent a structural change in response to blue light—meaning, at the very least, that it functioned as a photoreceptor.2 Bacillus has a noteworthy property—if you let it go into starvation mode, it produces a dark pigment. "Then by chance we knocked out the blue light receptor, and the pigment disappeared," Gaertner says. "This was the first case where we thought, 'Oh, this looks like a physiological or morphological readout in response to blue light.'" Could it be that the pigment is protective somehow, keeping the organism from further, light-induced stress?

"The fact is, these [receptors]—and they're cropping up all over the place—are being found in things that were never thought to have any kind of photoresponse." —Sean Crosson

Right around the time Briggs's group discovered the LOV domain, Crosson was beginning his doctorate at the University of Chicago with Keith Moffat, an expert in the kinetic protein structure of signaling molecules (see p. 48 for a profile on Moffat). When Crosson arrived in 1999, the group focused primarily on two molecules: myoglobin and a photoreceptor called photoactive yellow protein. Crosson wanted to carve out his own territory with a new sensory signaling protein; he'd seen Briggs's 1997 paper and was intrigued by the discovery of a new type of light sensor. "Given that the paper was new, we thought that we could jump onto this," he says. Moffat agreed and whipped off a note to Briggs, who invited the pair to Stanford and sent them back to Chicago with the LOV clone.

As a postdoc in Lucy Shapiro's lab, Crosson began playing with the LOV system in Caulobacter, an aquatic bacterium with a unique cell cycle. When he started up his own lab focused on environmental signaling in the organism, he began to search for what process, if anything, in the bacterium is triggered by light. One by one, his lab tested factors such as growth rate, cell-cycle parameters, and cell morphology for blue-light effects, all to no avail. "It was a depressing year, I've got to say," Crosson recalls.

Bacillus subtilis © Dennis Kunkel Microscopy, Inc.

As these things usually go, the discovery of a light-induced phenotype came when they weren't looking. One of Crosson's graduate students, Dan Siegal-Gaskins, was using a 50-micron–wide microfluidic channel to examine cell growth and cell adhesion in various Caulobacter mutants. The gene containing the bacteria's LOV domain also codes for a signaling protein called a histidine kinase. When he tested a mutant in which the gene was knocked out, it didn't seem to stick as well. Finally, thought Crosson, they were on to something. The group quickly made some new mutants, overexpressing the light-sensing LOV histidine kinase and its downstream signaling partner, and found it had enhanced cell adhesion. The LOV-containing gene, and presumably the light signaling through it, seemed to be telling the cells to stick.3

It was a striking story—a clear, in vivo demonstration that light regulates a process central to the cell's physiology. But before Crosson's group managed to publish their findings, Briggs, Roberto Bogomolni of the University of California, Santa Cruz, and a group of Argentinian collaborators published a similar study in a different species of bacteria that, as Losi likes to say, "created a minor earthquake" in the photoreceptor world.

Losi had initially found just three species of bacteria carrying the LOV domain sequence, but as more bacterial genomes were sequenced, the roster quickly grew. Briggs and Bogomolni had gone through the updated list of LOV-carrying bacteria, picking out the highly toxic Brucella abortus because of its importance to public health. The bacterium is a major cause of illness and spontaneous abortion in cattle, and along with its genus cousins, causes a violent fever when it infects humans. By 2005 the scientists had cloned its LOV receptor and were working to identify its physiology. When they knocked out a key element of the domain, virulence went down. That suggested to Briggs and Bogomolni that virulence might be the light phenotype, but they had trouble convincing their collaborators, who, unlike them, were experts in Brucella. "Nobody believed me," recalls Bogomolni, who'd gone down to Buenos Aires to wrap up the project. "We had to beg them to do the infection experiments under aluminum foil." Indeed, the bacterium turned out to be about 10 times more virulent when grown in the light than when grown in the dark—an effect they traced to its LOV domain.4

Brucella abortus
© Dennis Kunkel Microscopy, Inc.

If light has such a strong effect, says Briggs, now director emeritus of Carnegie Institution's Department of Plant Biology at Stanford, he can picture a whole slew of experiments on other LOV-containing pathogenic bacteria. Is virulence in Listeria, for example, also light sensitive? Would that give clues about controlling food-borne illnesses? And how does the pathogen's light sensitivity interact with its host's physiology? "I'm dying to do an experiment in which we see how much light penetrates a cow," he says.

Some researchers are speculating that light may actually be affecting results in microbial research. "Since we study global gene regulation, we take all of these precautions" to control for environmental factors, says Sean Cartinhour at Cornell University, such as growing cells in bioreactors. "We thought one day, 'what if we're not controlling for light and it's important? Then wouldn't we have to change all of our protocols?'"

So far, researchers have shown physiological evidence for LOV's light sensitivity in just three bacterial species: Caulobacter crescentus, Brucella abortus, and Bacillus subtilis. But Losi, Gaertner, Crosson, Bogomolni, and others are going out on a limb, concluding that light represents a hitherto-unknown general—and widespread—signaling mechanism. Already starting to find hints of blue-light signaling in other bacterial species, Briggs' and Bogomolni's Brucella study, for example, also identified three other species of bacteria with functional, blue-light–sensing LOV domains. The duo just obtained a three-year National Science Foundation grant to study Rhizobium, an agricultural symbiote that helps plant roots fix nitrogen, and Shewanella, which binds oxides in metal and may be usable for toxic waste clean-up. Last year, Losi's and Gaertner's groups showed in vitro that the plant pathogen Pseudomonas syringae has a photoactive LOV domain as well. Another group at Kalrsuhe University in Germany reported that Agrobacterium tumefaciens, also a prevalent plant pathogen, like Brucella seems to become more virulent under blue light, though they have not identified the relevant receptor. "There's a whole area of bacteriology that just didn't exist two years ago," Briggs says. "My guess is there's going to be a whole explosion of this stuff."

In Rhodobacter sphaeroides, the BLUF domain containing protein, called AppA, senses both oxygen and light and regulates the switch between oxygen-based respiration and light-based photosynthesis. Mutants lacking AppA (top) are locked in default "high O2" mode, producing no photosynthetic pigments. PpsR, AppA's signaling target, represses photosynthesis genes when AppA is inactive or mutated. However, when PpsR is knocked out (bottom), the bacterium falls into "no O2" mode, producing photosynthetic pigments indiscriminately, regardless of the presence of oxygen or light. Courtesy of Mark Gomelsky

It's not surprising that the microbial research community has managed to miss the effect of light on bacterial physiology for so long; traditionally, microbes are grown in dark incubators, and the question of light sensitivity rarely arose. "We were doing everything possible to avoid seeing the effects of light on microbes—that's basically the story of microbiology," says Mark Gomelsky at the University of Wyoming, who recently identified the BLUF domain, another type of blue-light protein that may be even more widely prevalent in chemotrophic bacteria than LOV.5 In fact, he almost missed it himself in his work on Rhodobacter sphaeroides, a purple bacterium often studied as a model system for photosynthesis, whose complex metabolism allows it to survive in an especially large range of environments.

Rhodobacter, explains Gomelsky, "solved its energy problems long ago—it can use all known means to derive energy—including oxygen-based respiration and light-based photosynthesis." The latter, though, is a back-up mechanism that kicks in only in the absence of oxygen. He and Sam Kaplan from the University of Texas Medical School assumed that the gene at the core of this process, AppA, contained an oxygen sensor, but they couldn't figure out how it worked. "And then I got a strange call from Gabriele Klug at the University of Giessen in Germany, who had this bizarre idea, or so I thought at the time, that [AppA] was a light sensor." She turned out to be right, and in 2002, the duo demonstrated that the protein controlled Rhodobacter's responses to both oxygen and light. Meanwhile, groups led by Masakatsu Watanabe from Japan and Carl Bauer at the University of Illinois published studies showing that the flavin-containing domain was indeed a light sensor. Although technically BLUF's name is an acronym for "blue light using flavin adenine dinucleotide," "it got its name because at the time I got a sense it was deceiving me with its function," Gomelsky says.

He's not the only microbiologist to be perplexed by the phenomenon; so far, the idea of light as a specific physiological signal in chemotrophic bacteria is still on the fringes of research—it's simply not something that microbiologists are used to considering. "When you come to a scientific conference with a story about photoreceptors," says Klaas Hellingwerf at the University of Amsterdam, "they say, well, this is interesting for plants and photosynthesis, but not for the rest." Hellingwerf, who is collaborating with Losi and Gaertner on investigating the role of LOV in Bacillus, is convinced that even the studied-to-death bacterium Escherichia coli contains a photoreceptor. The organism has the gene sequence and it seems to play a role in biofilm formation, but his group has not yet succeeded in uncovering the molecular mechanism. "In Bacillus, we knew the output of the response, and so we could work our way through" to demonstrate the phenotype, he says. But in E. coli "the biological function is more complicated, and we didn't have a good readout."

Rhodobacter sphaeroides
Courtesy of A. Varga and S. Kaplan

The idea is gaining ground, though, in part, researchers say, because it makes so much evolutionary sense—in retrospect, the surprising thing may be that there aren't more such receptors. Blue light itself is not harmful to cells, but ultraviolet light is a major source of cellular oxidative stress, and was especially strong before the formation of the earth's ozone layer, explains Bauer, who studies photobiology and bacterial evolution. Blue light's proximity to UV on the light spectrum means it might be telling organisms that UV levels are high, signaling to motile organisms that it's time to find some cover, or switching on protective mechanisms in sessile cells that can't run away. "I would not be at all surprised if DNA repair pathways were found to be regulated by light," says Bauer.

"It's as though the whole system is an electric cable, in which a single input plugs into a whole range of outputs." —Keith Moffat

It's still not clear to Crosson why Caulobacter perceives light, or why light would affect the surface properties or adhesiveness of a bacterial cell. His group has recently discovered strong feedback regulation between the LOV-light pathway and an unrelated sensory system in Caulobacter that detects a range of other environmental signals. Light-activated LOV, he suspects, may not be tuned specifically to cell adhesion, but instead may be part of an interplay of environmental cues that modulate each other to produce different cellular effects.

Probably, researchers say, a bacterium's specific environmental niche will dictate which specific cell functions link to the blue light photoreceptor. Most LOV domains are associated with histidine kinases (key elements of two-component signaling), but others link to different genes, potentially regulating a range of cell features. "There are various environmental reasons a system could choose to have light as a signal," notes Hellingwerf—light might inform the organism about its position in the water column, the time of day, how deep in the soil it finds itself, or whether it's at increased risk of oxidative stress, caused by exposure to ultraviolet light. In pathogenic bacteria, for example, light might be signaling whether the organism is inside the host or outside in a grassy field, so that it can turn up or turn down its virulence as the surroundings demand. What's more, the photoreceptor's modularity is such that a single species of bacterium may have several LOV domains, each plugging into a different cellular function. "It's as though the whole system is an electric cable, in which a single input plugs into a whole range of outputs," says Moffat, who has continued to work with LOV domains after Crosson introduced them to his lab.

6 So convinced is Moffat of the approach, that after the paper came out, he said, "I did something I've never done before"—he sent it to about 15 researchers who he thought might be interested. One of the first replies he received was from Hellinwerf, who is also planning to engineer a LOV-based sensor to study the basic properties of diffusion in eukaryotic cytoplasm.

Meanwhile, researchers are still far from a complete understanding of where photoreceptors such as LOV domains might crop up, and what other processes in the bacterial world (and beyond) might prove to be sensitive to one of the planet's most widely present stimuli. Hellingwerf is convinced, for example, that his postdoc has found a human gene "that has all the characteristics of a LOV domain with respect to gene sequence," though his lab has not yet been able to demonstrate that it acts as a photoreceptor. Crosson, for one, describes an almost mystical rush when he wonders what other aspects of organisms and their environments remain undiscovered. "Here's a signal that we've just completely ignored," he says. "It makes me think, what else are we totally missing?"

Correction (May 7): The Scientist incorrectly identified Wolfgang Gaertner's institution as the the Max Planck Institute for Chemical Ecology in Munich. In fact, he is at the Max Planck Institute for Bioinorganic Chemistry in Mülheim.

1. E. Huala et al., "Arabidopsis NPH1: A protein kinase with a putative redox-sensing domain," Science, 278:2120–23, 1997.
2. A. Losi et al., "First evidence for phototropin-related blue-light receptors in prokaryotes," Biophys J, 82: 2627–34, 2002.
3. E.B. Purcell et al., "A photosensory two-component system regulates bacterial cell attachment," Proc Natl Acad Sci USA, 104:18241–46, 2007.
4. T.E. Swartz et al., "Blue-light-activated histidine kinases: Two-component sensors in bacteria," Science, 317:1090–93, 2007.
5. M. Gomelsky and G. Klug, "BLUF: A novel FAD-binding domain involved in sensory transduction in microorganisms," Trends Biochem Sci, 27:497–500, 2002.
6. D. Strickland et al., "Light-activated DNA binding in a designed allosteric protein," Proc Natl Acad Sci USA, 105:10709–14, 2008.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?