Â鶹ÒùÔº

In a new study, scientists have shown that chemical receptors that plants use to recognize nitrogen-fixing bacteria have developed the same function independently on at least three separate occasions through a process called convergent evolution.

"Beans are in other ways not a burdensome crop to the ground, they even seem to manure it… wherefore the people of Macedonia and Thessaly turn over the ground when it is in flower."

The Greek philosopher Theophrastus, widely considered to be the father of botany, wrote the preceding excerpt in what is possibly the first book exclusively dedicated to the discussion of plants sometime between 350 and 287 B.C. The ancient Greeks, along with many other early agricultural civilizations, had caught on to one of botany's best kept secrets: Bean plants are a great source of fertilizer.

A little more than 2,000 years later, humans figured out why. Many bean plants and other closely related species cultivate a type of bacteria in their roots that actively turn atmospheric nitrogen—which plants cannot access—into something plants are capable of using.

For centuries, humans have used this biological trick to their advantage by growing bean plants to boost crop yields. Now that they know how it works, scientists want to take this practice a step further by imparting the symbiosis to other species through genetic engineering.

But getting there won't be easy. Scientists don't even know how many times this symbiosis has evolved, which is a crucial bit of information for understanding the complex relationship shared between plants and bacteria. Many plant species in this group lack the symbiosis, meaning it either evolved once and was subsequently lost dozens of times, or it originated separately through convergent evolution.

Researchers from several institutions, including the Florida Museum of Natural History, have recently determined that the chemical receptors plants use to recognize nitrogen-fixing bacteria in the soil independently evolved at least three times. The results support a long-standing theory that bean plants and their relatives inherited a predisposition for symbiosis from a distant ancestor that lived more than 100 million years ago.

"That was probably followed by a series of separate origins, as well as recurring loss events throughout this group," said study co-author Douglas Soltis, a distinguished professor at the Florida Museum. "This study shows, for the first time, that there are other roads to Rome. There are multiple ways to get to nitrogen-fixing symbiosis."

The authors have published in the Proceedings of the National Academy of Sciences.

Nitrogen first limited, then unlocked the potential for life on Earth

About 3.2 billion years ago, long before the first photosynthetic bacteria filled the air with oxygen, life on Earth hit a serious roadblock. The single-celled organisms that lived during this time grew in thick mats that often accreted into living boulders, called stromatolites, but they were hindered by a limited supply of building material.

Specifically, they lacked nitrogen, which was—and is—abundant in the atmosphere, but difficult to digest. Elemental nitrogen has a neutral charge, being made of seven protons and seven electrons. Four of these electrons pair up with each other, creating a stable partnership, but the remaining three orbit the nucleus of the atom alone, all magnetically yearning for a partner.

This is why you will never find elemental nitrogen in the wild. Whenever two nitrogen atoms come into close enough proximity, the three single electrons connect to each other in an incredibly strong triple bond, forming dinitrogen.

Dinitrogen is merely one among millions of neutral molecules, but of these, it has the electrical bond. This is a major inconvenience for life on Earth, which cannot use nitrogen in its diatomic form.

To be biologically useful, dinitrogen must first be converted to something that's easier for enzymes to work with. The earliest forms of life completely lacked the ability to do this, which means they had to rely on the limited nitrogen supply chain of lightning strikes and the occasional meteor impact, two of the only that generate enough heat to break dinitrogen's triple bond.

It takes a single dinitrogen molecule a to cycle out of the atmosphere, and this was simply not fast enough to satisfy the growing demand of microbes.

Fortunately, dinitrogen has a weak spot; otherwise, it's possible life would never have made it as far as multicellularity.

While it's true that dinitrogen's three pairs of interlinked electrons can only be pried apart under extreme environmental conditions, they can, in rare circumstances, be coaxed into letting go and bonding with other elemental partners.

With all its electrons spoken for, dinitrogen is mostly inert. You could throw it in a test tube with all sorts of other elements and molecules, and in most cases, dinitrogen would not participate in any of the ensuing chemical reactions.

When combined with certain types of metals, however, dinitrogen's triple bond loses some of its strength. Early prokaryotes exploited this weakness by sticking a bunch of iron atoms and a single molybdenum atom together to form an enzyme and shooting protons at it. The enzyme, called nitrogenase, traps dinitrogen in a weakened state, and the volley of protons cleaves the bond, ultimately producing ammonia, which organisms can easily process and incorporate into their cells.

The problem was solved, and life proliferated, except that the many multicellular organisms that came after never bothered to evolve a nitrogen-fixing enzyme of their own. This may have something to do with the fact that nitrogen fixation ranks among one of biology's most energetically expensive metabolic processes, commensurate with the strength of its bond.

As per the chemical equation for photosynthesis, it takes three molecules of adenosine triphosphate (ATP) to fix a single molecule of carbon dioxide. It takes 16 ATPs to do the same for a single molecule of dinitrogen. This is not the only associated expense.

Recall that the nitrogenase enzyme is made of mostly iron, and as anyone who has left a steel tool out in the rain can attest, iron is highly reactive to oxygen. So, nitrogen-fixing bacteria have the added task of keeping oxygen away from the enzyme to prevent it from rusting, which requires a lot of work. In one , researchers found that oxygen removal at certain times of the day used up the largest allocation of the cell's energy, even more so than the actual process of nitrogen fixation itself.

As a consequence, prokaryotes are still the only living things capable of splitting dinitrogen after 3 billion years of evolution on our planet. This had serious repercussions for the direction life took when it evolved roots and legs and made the transition onto land. In modern terrestrial environments, a lack of nitrogen is the primary limitation to plant growth, after which comes water, then phosphorus.

But plants are clever, and over the years, they've developed a number of strategies to catch nitrogen in areas where there's not enough to go around. Most notably—and gruesomely—plants like sundews, Venus flytraps, bladderworts and pitcher plants have resorted to violence, taking the nitrogen they need from insects they trap and consume.

Other plants have opted for a more peaceful resolution by striking a deal with their ancient prokaryotic relatives in which the two form a mutually beneficial partnership. In exchange for room and board, nitrogen-fixing bacteria relinquish a portion of the ammonia they produce to their plant hosts.

Some feather mosses, for example, fold the tips of their leaves, providing an insulated environment for nitrogen-fixing bacteria to grow. There are aquatic ferns with secret, inner chambers full of nitrogen-fixers and strange plants called cycads whose roots seem to have lost their way and grown up from the ground, flooding the photosynthetic nitrogen fixers that live inside with light.

A rare type of corn plant native to Mexico has evolved its own version of the latter strategy, but its aerial roots are covered in a generous layer of mucous. But these are, for the most part, isolated instances.

The real botanical champions of nitrogen fixation are plants that house bacteria in subterranean nodules produced by their roots. The function of these nodules was initially discovered in the bean family by a pair of German botanists, Hermann Hellriegel and Hermann Wilfarth, working in the 1880s. Nodules have since been confirmed in thousands of species in the bean family and hundreds of other species in seemingly random and unrelated families. These include the birch, hemp, wax myrtle, buckthorn and rose families.

The only thing these wildly different groups seemed to have in common was the presence of nodules, which implied they had all hit upon the same innovation independently of each other through a phenomenon called convergent evolution.

Then, in the 1990s, the Florida Museum's Pamela and Douglas Soltis found something strange. Up to that point, most of the work done by scientists to figure out how things are related to each other had been based on morphology. Using this approach, two species that looked similar to each other were considered to be more closely related than to species that looked different. When scientists figured out how to sequence DNA in the 1970s, it quickly became apparent just how insufficient this classification system had been. All sorts of things that were thought to belong to one group turned out to be in another.

In one of the more , the Soltises showed that all the disparate families with nodulating species were actually each other's closest relatives and belonged together in one group.

"We cleverly named it the nitrogen-fixing clade," Douglas Soltis reminisced.

Everything in the nitrogen-fixing clade evolved from a single ancestor, which lived in the realm of 110 million years ago. This, in turn, implied that the basic underlying genetic framework for making nodules had evolved just once in this ancient ancestor, which it then passed down to its descendants.

There were other implications as well. Not everything in the nitrogen-fixing clade is capable of fixing nitrogen. Of the 30,000 or so species in this group, scientists have only observed nodules in 8,000. If nodules had only evolved once, as the data seemed to suggest, then they must have been lost multiple times in the lineages that no longer had them.

Scientists have since waffled back and forth on the answer. Did nodules evolve just one time and later disappear in some families? Or, despite their close relationship, did some families in the nitrogen-fixing clade evolve nodules more or less independently from others?

The question might seem arcane, but there's a lot riding on the outcome. Humans figured out how to split dinitrogen on an industrial scale in the 20th century, and we've since dumped tons of it—in the form of fertilizer—onto agriculture fields. This innovation has allowed our population to reach its current record high of 8 billion people and caused a slew of unintended environmental catastrophes.

If scientists can figure out a way to genetically alter crop plants to produce nodules for bacteria, farmers would save an enormous amount of money on fertilizer, and the environmental cost of industrial agriculture would be significantly reduced.

Before that can happen, scientists need to know whether there's one set of genes that control the production of nodules, passed down by a single ancestor, or if nodules can be made willy nilly with all sorts of different genes. There has, so far, been no consensus.

Plants call for aid, and bacteria answer

There has, however, been progress. Scientists have tackled the question from multiple different angles, and they've learned a lot about how plants and nitrogen-fixing bacteria interact and how nodules are formed.

New plants capable of producing nodules do not come equipped with bacteria straight out of the seed. Instead, they have to find it for themselves. Most species in the nitrogen-fixing clade form relationships with bacteria in the genus Rhizobium. A few others pair up with Frankia, a genus of bacteria named after Albert Bernhard Frank, one of the first pioneering botanists to study plant nodules and the person generally attributed with coining the term "symbiosis."

The amount of Rhizobium and Frankia in the soil is low compared to other bacterial and fungal groups. One estimate puts the number at just per gram of soil. This isn't much, considering that a single gram of soil can contain a total of more than .

To find each other in the dark, the plant first secretes flavonoids, which function as a chemical signal alerting nitrogen-fixing bacteria to the presence of what will potentially be their new home. To begin the process of assimilation and symbiosis, Rhizobium and Frankia must respond in kind with a chemical password known only to them. This password takes the form of what are called nod factors, long strings of complicated polymers that plants recognize upon contact.

The correct identities established, the root opens to let the bacteria inside. That is, whenever a root hair is given the signal, it begins to coil around itself—sometimes in a half spiral of 180 degrees, sometimes in a full 360-degree twist, thereby scooping up a pinch of the waiting bacteria and holding them in a tight embrace. The bacteria, meanwhile, begin secreting an enzyme that weakens the cell wall of their chamber. Once softened enough, the cell wall caves in on itself, though whether it does this because it's being pushed from the outside or pulled from within the cell, scientists don't know.

This cavity inside the root hair, complete with its own membrane that keeps bacteria from wandering off in the wrong direction, opens into the inner cortex of the root, allowing the bacteria to bypass the plant's defenses without setting off any alarms. Once in the root proper, the presence of bacteria initiates the growth of nodules.

Scientists who study the genetics behind this intricate process tend to favor the idea that nodulation evolved just once, reasoning that it'd be easier to lose a complicated trait than it would be to build it from scratch. As evidence, they point to a suite of genes with lengthy names, such as the symbiosis receptorlike kinase that are present in all nodulating species and absent or highly modified in closely related non-nodulators. The likeliest explanation for having a common set of genes that all have the same or similar functions, they contend, is that they all came from a single ancestor.

On the other side of the debate are the phylogeneticists, including Pamela and Douglas Soltis, who are primarily interested in how species are related to each other. Results from their work generally suggest that a predisposition for nodulation evolved in the ancestor of the nitrogen-fixing clade. Critically, this ancestor would not have been able to produce nodules. Rather, it benefited in some less direct way from having the genes that would later evolve—on several separate occasions—into the genes that code for nodulation.

According to this line of reasoning, the mere fact that there are two bacterial lineages—Rhizobium and Frankia—that have made similar arrangements seems to hint at separate origins. The nodules also have different shapes and are located in different parts of the root depending on what family produces them.

There are multiple roads to nitrogen fixation, most of them uncharted

In the present study, both groups combined their methodologies and worked together.

"No one has been able to reconcile the genomic data with this pattern of potential parallel gains," Pamela Soltis said. "That's essentially what Christina did."

When she started working on the study, Christina Finegan was a graduate student in the Plant Molecular and Cellular Biology program at the University of Florida. She led the study as part of her doctoral dissertation. Finegan is a crop biologist by trade who specializes in quantitative genetics, the field of science that concerns itself with how genes work and what they do.

For this study, she and her colleagues chose to focus on the proteins that handle password recognition in nodulating plants, both for their importance and their antiquity. As with all other complex traits, these proteins didn't spring from the forge of evolution fully formed. Rather, they were co-opted from genes that control a similar microbial interaction that occurs in the roots of more than 90% of all plants and dates back to a time before trees.

In this partnership, fungi called mycorrhizae do all the hard work of absorbing water and nutrients from the soil in exchange for a cut of the sugar plants make with it. At some unknown point in the history of the nitrogen-fixing clade, a plant (or plants) underwent one (or multiple) genome duplication events that left them with an extra copy of the genes that recognize mycorrhizal fungi.

Since they only needed one copy to get the job done, the others were free to accumulate mutations and take on new functions, which is how they became part of the genetic toolkit for nodulation.

"The machinery was kind of just sitting there, and that made a lot of room for evolution to play with and repurpose it," Finegan said.

Rather than trying to pinpoint the origin of the nodule symbiosis, which is made of multiple moving parts, the authors reasoned they could get at the same question by figuring out how many times the password recognition genes had been duplicated and reassigned. If the switch occurred just once, then the co-opted password recognition genes should share basic similarities among all nodulating species in the nitrogen-fixing clade.

They could do this effectively because they'd recently constructed a massive tree of life for the nitrogen-fixing clade with DNA from more than 12,000 species. Using this as a framework, they looked for duplications among the password recognition genes in 28 species for which complete genomes were available. This included 11 species with rhizobial symbionts, five with Frankia symbionts and 12 that did not produce nodules.

The results show that within these 28 species, the genes in question had been duplicated nine times. Such DNA duplications are common, and for the most part, nothing seems to come of them. This is an active area of research, and there are currently more questions than answers. But many organisms—especially plants—are saddled with the accumulated waste of ancient duplications that may do little more than weigh them down. Occasionally, however, they confer an advantage, as appears to be the case for three of the nine duplications in the nitrogen-fixing clade. Two occurred in the bean family and one in the ancestor of the rose, pumpkin and other related families.

Tellingly, species in the bean family that lacked the duplication were those that did not produce nodules. This is highly suggestive of multiple origins through duplication.

There was, however, one group of nodulators that showed no evidence of duplication, which were represented by two species: the common alder, a tree of middling height with a reputation for thriving in poor soils throughout Europe and Asia; and a tree in the birch family called the swamp she-oak which, despite the name, looks like it's doing its best impression of a pine tree. The latter species grows in disturbed and waterlogged environments in Australia and is also invasive in Florida.

These species are two of the few nodulators that form a relationship with Frankia, and because they're trees, scientists know comparatively little about their symbiosis.

"It's hard to get down to the right part of the root to study the nodule," Pamela Soltis said.

Most of the research on nodule symbiosis has been done on a handful of model organisms, which in this case are small plants that are easy to grow in a greenhouse.

The authors say that it's likely the common alder and swamp she-oak communicate with their bacterial partners using a completely different set of genes, which strongly aligns with the theory that nodules evolved multiple times.

"By taking this broad evolutionary perspective, we were able to take the power of biodiversity and find evidence of processes that you can't find if you're just sticking with model organisms," Pamela Soltis said.

As to what this portends for the potential to genetically engineer nitrogen-fixing crop plants, Finegan is optimistic. On the one hand, a single origin might make things easier. If all nodulating plants in the nitrogen-fixing clade are using the same set of genes they received from a single ancestor, then using a model organism that's easy to study would be a more sensible strategy for scientists than spreading their efforts thinly among dozens of other species, many with nodules that are deeply rooted in the earth.

On the other hand, if there's more than one way of making an inviting nodule for microbes, then there are presumably just as many ways to do so artificially.

"One good thing about convergent evolution is you can narrow down on what's truly needed because you've arrived at the same thing through independent pathways," she said.

Other authors of the study are Heather Kates, Marcio Resende, Jr. and Matias Kirst of the University of Florida; Robert Guralnick of the Florida Museum of Natural History; Jean-Michel Ané of the University of Wisconsin-Madison; and Ryan Folk, former postdoc of the Florida Museum of Natural History and now at Mississippi State University.