A Stitch in Time

By Mark Schrope

Isaac Asimov famously observed that: "The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' ('I found it!'), but rather, 'Hmmm…that's funny.'" Researchers in Professor Albert Eschenmoser's laboratory at The Scripps Research Institute recently had just such an experience in their search for the chemistry of life's origins.

Eschenmoser and his team at Scripps Research and its Skaggs Institute for Chemical Biology had set out to study the potential of novel oligomer systems to act as informational systems similar to RNA. One of the goals of the team's work has been to systematically synthesize by chemical methods potentially self-replicating chemical systems that might have been competitors to RNA in a primordial world.  In a system that they studied recently, the structure of the strands involved led them to expect that each would pair well with RNA and DNA as well as with its partner.

Then came the "Hmmm…that's funny." To the team's surprise, some of the strands unexpectedly failed to bind, leading to the discovery of an additional criterion in their search for strands that might reasonably be thought of as primordial alternatives to RNA and that may shed light on why RNA prevailed in the primordial world.

RNA World

Eschenmoser first became involved with origins-of-life research in the 1980s while at the Swiss Federal Institute of Technology in Zurich. He was studying the complex synthesis of the vitamin B12, the most structurally complex vitamin known.  From there, his interest moved to another complex structure: RNA.

There is a prevailing theory that in the primordial world, before the rise of DNA, RNA acted alone in propagating primitive life. In such an "RNA World," as the scenario is known, RNA would have had to act as both a transmitter of information and catalyst of life-sustaining reactions.

A problem with the RNA World is that, despite half a century of research, the potential for RNA to self-assemble under what are thought to be likely primordial conditions has still not been demonstrated.  "You get there and then you are stumped," says Ramanarayanan ("Ram") Krishnamurthy, a Scripps Research associate professor with Eschenmoser's group.

This problem raises questions that have driven Eschenmoser's research now for 20 years.  "Life, as we know it, is a chemical life," says Eschenmoser,  "Therefore, the problem of life's origin is above all a chemical problem, referring to chemistry as whole, yet in its heart it is a problem of synthetic organic chemistry."

As a result, some goals of the Eschenmoser team's work have been to synthesize potentially self-assembling and self-replicating chemical systems that might either have been competitors to RNA in a primordial world or feasible precursors to the initial, enigmatic generation of RNA.   

Krishnamurthy, who has worked with Eschenmoser since 1992, was hooked on the research as soon as he heard Eschenmoser discuss it at a seminar two years before that at the Ohio State University. Krishnamurthy had been trained there in synthetic methodology, but was instantly attracted to the possibility of using chemistry to address fundamental questions between chemistry and biology. "For the first time, I saw organic chemistry being used in a very different way," he says, "I just jumped at that."

Krishnamurthy says that addressing the fundamental implications of the group's work has always been a challenge. "We want to avoid giving the impression that we are trying to reconstruct the history of the origins of life," he says, "We are very well aware that that is almost impossible to do, because there is no way of going back and checking."

Novel Backbones

The Eschenmoser group's earliest origins-of-life experiments involved oligomers where sugars similar to the ribose in RNA were substituted in the backbone or sides of the DNA/RNA "ladder." These novel backbones were bound to the same nucleic acids, or bases, found in RNA and DNA, known as the canonical bases. This research revealed that the Watson-Crick base pairing seen in RNA and DNA could be achieved with a variety of backbone structures, as long as certain criteria, such as the right level of flexibility, were met.

One of the team's earlier successes was with substituting the simple sugar threose in a backbone, to form TNA, which proved itself a completely functional informational system like RNA. TNA is especially intriguing because threose is a simpler sugar than the ribose in RNA, meaning that it would be easier to form, suggesting some potential as an RNA precursor.

Despite the team's best efforts not to overstate the significance and implications of such research, reporters have nonetheless at times gone beyond the bounds of the results. Krishnamurthy remembers with a laugh being asked after the TNA studies were first published whether the compound might be found on Mars. "There was no way for me to keep a straight face and answer," he says, "because it's too speculative. We don't even know if TNA existed on Earth."

For later research, the group began to expand beyond systems structurally similar to RNA to systems that were generationally similar, meaning compounds that could feasibly be produced by processes similar to those that will generate RNA using chemical building blocks potentially present in a prebiotic world.  This allowed them to substitute potential alternative bases for the canonical bases to determine whether the alternatives too might exhibit Watson-Crick base pairing. 

The Eschenmoser team's two most recent papers, just published online in Angewandte Chemie November 17 and designated by the journal as "Very Important Papers," describe the results of this new direction. The work focused on two pairs of oligomers with the diamino and dioxo forms of triazines, and the diamino and dioxo forms of pyrimidines acting as alternative bases. The expectation, based on the compounds' modeled structures, was that the diamino and dioxo forms of each would exhibit Watson-Crick base pairing, and that the compounds would also bond well with RNA and DNA. The results of experiments to test these theories were startling.

The diamino triazine paired strongly with RNA and DNA, as expected, but the dioxo triazine paired very weakly. Stranger still, the observations were the opposite for the pyrimidines, with the dioxo pairing strongly and the diamino pairing weakly.

"How Could It Be?"

Krishnamurthy remembers clearly when postdoctoral fellow Gopi Mittapalli, lead author of the recent publications, first told him the results. "I said,  'How could it be?'" he says, "We initially thought there was something wrong with the compounds." So, the researchers rechecked their work to synthesize the compounds, and used nuclear magnetic resonance (NMR) to confirm that they had the correct structures.

Once satisfied there had not been a mistake, the team began looking for answers. What they found was a remarkable correlation between whether the base in a synthesized compound bonded well with RNA and DNA or its pair, and the value of its acid dissociation constant. Known as pKa, this is an intrinsic property of a compound that is a measure of its acidity and basicity in a given medium such as water. The group's results suggest that a base with a pKa value close to that of the base it is to pair with will have weak if any bonding, while the team found strong bonding among bases with high differences between their pKa values.

Optimal strength of bonding between bases is literally responsible for life as we know it. Bonding that is too weak would prevent self-replication from proceeding, while bonding that was too strong might do the same, because double strands would become too difficult to separate. RNA and DNA exhibit an optimum base pairing strength, and understanding the reasons why are critical to understanding how they arose.

So, though the team initially set out to identify potential RNA alternatives, what they found instead was an intriguing new criterion to better understand RNA and DNA and further the search for alternatives or precursors. Though pKa itself may not be responsible for the relative bonding strengths of different bases, the results do suggest that pKa differences are a good and relatively convenient indicator for potential bases that will have the proper bonding strength. "We see this as a correlation, not an explanation," says Krishnamurthy. "An explanation must come from the physical and chemical properties of the bases themselves." Nonetheless, he says, pKa will now be one of the first considerations in work to identify alternative bases.

Indeed, the researchers are already working to identify new compounds with favorable pKa values to study. In addition, work is under way in collaboration with the group of Julius Rebek, Jr., a professor at Scripps Research and director of the Skaggs Institute for Chemical Biology, to determine what effect the solvent used for experiments has on base pairing. 

Although no one will ever be able to say for sure what happened chemically at the dawn of life, each advance the team makes is a step toward answering the overarching question of how life could have emerged.

Send comments to: mikaono[at]scripps.edu

The Eschenmoser lab searches for the chemistry of life's origins. Pictured here are (left to right): Kondreddi Ravinder Reddy, Miguel Guerrero, Ramanarayanan Krishnamurthy, Albert Eschenmoser, Gopi Kumar Mittapalli, and Yazmin Osornio.