LINKING SIMPLE CHEMISTRY TO SOMETHING LIKE LIFE.

A computer model of a growing membrane.

 

Origin of life researchers have made impressive progress in recent years, showing that simple chemicals can combine to make nucleotides, the building blocks of DNA and RNA. Given the right conditions, these nucleotides can combine into ever-longer stretches of RNA. A lot of work has demonstrated that RNAs can perform all sorts of interesting chemistry, specifically binding other molecules and catalyzing reactions.

So the case for life getting its start in an RNA world has gotten very strong in the past decade, but the difference between a collection of interesting RNAs and anything like a primitive cell—surrounded by membranes, filled with both RNA and proteins, and running a simple metabolism—remains a very wide chasm. Or so it seems. A set of papers that came out in the past several days suggest that the chasm might not be as large as we'd tend to think.

Ironing out metabolism

A lot of the basic chemistry that drives the cell is based on electron transport, typically involving proteins that contain an iron atom. These reactions not only create some of the basic chemicals that are necessary for life, they're also essential to powering the cell. Both photosynthesis and the breakdown of sugars involve the transfer of electrons to and from proteins that contain an iron atom.
DNA and RNA tend to have nothing to do with iron, interacting with magnesium instead. But some researchers at Georgia Tech have considered that fact a historical accident. Since photosynthesis put so much oxygen into the atmosphere, most of the iron has been oxidized into a state where it's not soluble in water. If you go back to before photosynthesis was around, the oceans were filled with dissolved iron. Previously, the group had shown that, in oxygen-free and iron rich conditions, RNAs would happily work with iron instead and that its presence could speed up their catalytic activity.
Now the group is back with a new paper showing that if you put a bunch of random RNAs into the same conditions, some of them can catalyze electron transfer reactions. By "random," I mean RNAs that are currently used by cells to do completely unrelated things (specifically, ribosomal and transfer RNAs). The reactions they catalyze are very simple, but remember: these RNAs don't normally function as a catalyst at all. It wouldn't surprise me if, after a number of rounds of evolutionary selection, an iron-RNA combination could be found that catalyzes a reaction that's a lot closer to modern metabolism.
All of which suggests that the basics of a metabolism could have gotten started without proteins around.

Proteins build membranes

Clearly, proteins showed up at some point. They certainly didn't look much like the proteins we see today, which may have hundreds or thousands of amino acids linked together. In fact, they may not have looked much like proteins at all, if a paper from Jack Szostak's group is any indication. Szostak's found that just two amino acids linked together may have catalytic activity. Some of that activity can help them engage in competition over another key element of the first cells: membrane material.
The work starts with a two amino acid long chemical called a peptide. If that peptide happens to be serine linked to histidine (two amino acids in use by life today), it has an interesting chemical activity: very slowly and poorly, it links other amino acids together to form more peptides. This weak activity is especially true if the amino acids are phenylalanine and leucine, two water-hating chemicals. Once they're linked, they will precipitate out of a water solution.
The authors added a fatty acid membrane, figuring that it would soak up the reaction product. That definitely worked, with the catalytic efficiency of serine-histidine going up as a result. But something else happened as well: membranes that incorporated the reaction product started growing. It turns out that its presence in the membrane made it an efficient scrounger of other membrane material. As they grew, these membranes extended as long filaments that would break up into smaller parts with a gentle agitation and then start growing all over again.
In fact, the authors could set up a bit of a Darwinian competition between membranes based on how much starting catalyst each had. All of which suggests that proteins might have found their way into the cell as very simple chemicals that, at least initially, weren't in any way connected to genetic and biochemical functions performed by RNA. But any cell-like things that evolved an RNA that made short proteins could have a big advantage over its competition.

Proteins go big

How do you go from short peptides to long, complex proteins that rely on a specific sequence of amino acids in order to function? It turns out that you need a fairly specific sequence in order to perform an equally specific function (although even that can be very flexible). However, you may not need to be very specific at all if you just care about having any function at all. In other words, it may have been useful for a cell to just randomly link amino acids together.
This work grew out of a previous study in which someone specified enough of the amino acid sequence to allow a protein to fold up into a small spherical structure. Anything beyond that was made random. The surprising thing was that these partly random sequences ended up doing all sorts of different things, binding some chemicals, catalyzing reactions, and so forth. But it's hard to make and test a huge complement of random sequences, so the authors turned to computational modeling, trying out a comprehensive set of potential proteins.
It turns out that a lot of them would potentially stick to interesting chemicals and form various binding pockets. In fact, a number of proteins that appear to be entirely unrelated on the sequence level happily formed very similarly shaped binding pockets. All of which suggests that making just any protein, with little regard for its actual sequence, could have a positive impact on the cell's fitness. Once in place, it could be adapted to be a bit more sequence-specific.
Although all of these results are exciting, it's important to place them in context. We're never going to know precisely how life first arose, since the actual evidence for the events no longer exists. But we can come up with plausible pathways from basic chemistry to simple biochemistry, something these studies seem to provide. That's not to say that another study won't come up with something that's even more plausible in the future.
The other thing to note is that these studies may start filling in what seems to be a big chasm between life and non-life, and they have a habit of taking a single gap and dividing it in two. They may identify landmarks on the journey towards life, but they open up new questions about exactly what route was taken between them.