Dispatches from the Creation Wars

Let’s take a break from litigation and turn our eyes on ligation. The type of ligation I am talking about is connecting two strands of nucleic acids (RNA in this case; a similar process with the same name takes place with amino acids) to make a longer strand. This is an important concept in origins-of-life research (and in biology), because it allows long strands with high information content to be assembled in shorter segments, kind of like a chemical assembly line. (Note that I am using “information content” in the sense of compressibility). In essence this allows Nature to reduce the odds against producing the right sequence of bases in a long strand. It’s generally much easier to reliably produce short strands than it is to reliably produce long strands.

Of course, it doesn’t do you any good if the shorter strands are simply connecting at random – this doesn’t reduce the probability. So what you want is a process that reliably connects the correct strands in the correct order. The process doesn’t have to be perfect, just better than random. One way to do this is by speeding up the ligation process for the right strands in the right order – in other words, use a catalyst. RNA has an interesting property – it has a “backbone” that strongly connects linearly, as well as matching base pairs that connect weakly. This means that RNA can act as a catalyst for itself – an autocatalyst. Are there combinations of short RNA strands that reliably catalyze into longer strands?

What would be even cooler is if the longer strand could act as a catalyst that takes the short strands and makes another long strand just like itself. That’s self-replication, the first step towards life.

Chemists at the Scripps Research Institute did just that.

The basic idea is simple. Take two strands, A and B. When strung together, they create a longer strand E. If A, B, and E have the right properties, E will catalyze A+B, making another E (note that A+B can join spontaneously, the catalyst just makes it happen faster).

It wasn’t so easy in practice. Apparently, the candidates for a straight A+B=E reaction also catalyzed some byproducts that interfered with the main process, causing it to stop. So what they did instead was to tweak it so that four short strands, A, B, A’, and B’, created two long strands, E and E’. E acted as the catalyst for A+B=E’, and E’ was the catalyst for A’+B’=E. A little bit of tweaking, and one day, the reaction not only kept going, it experienced exponential growth. Side note: the tweaking was done using mutagenic PCR. In other words, random mutation! PZ has a detailed explanation of this over at Pharyngula.

But they didn’t stop there. They actually developed several sets that could sustain the reaction. They then put these replicators in direct competition to see what would happen. The procedure is pretty simple, and may sound quite familiar to those who heard about the cit+ evolution in E. coli published by Lenski’s team at Michigan State University last year. Place the experimental “population” in a container with a known quantity of resources (short strings, in this case). Let population grow until resources are gone, then transfer a small number into a new container with fresh resources. Wash, rinse, repeat. This procedure, called serial transfer, is quite useful. First, it allows you to accurately track the number of doublings that occur based on the transfer volume. Here, 1/25 of the volume is transferred, so the number of doublings (equivalent to generations) between transfers is log2 25 = 4.64, or about 14 “generations” every three transfers. The population that is not transferred can be analyzed or stored for future reference. It also provides a selective pressure based solely on how quickly a replicator can copy itself.

A really cool result came out of this part of the experiment. The replicators occasionally had minor mutations, some of which proved to be faster than the original replicators. By the 77th generation, (about 15 transfers after it was started) all of the original replicators had gone extinct, and three mutant replicators dominated the population, with a slew of other less successful mutants.

That’s right. We have experimentally observed chemical evolution.

Lincoln TA, Joyce GF (2009) Self-sustained replication of an RNA enzyme. Science Jan 8. [Epub ahead of print].