Self-catalytic RNA enzymes

GumbyTheCat recently posted about a study in which researchers created self-replicated RNA enzymes. There was a bit of discussion in the comments about what those enzymes are all about and whether they are, indeed, "true" enzymes (i.e., proteins).

As I have access to Science magazine online, I was able to read the actual article (Gumby's post was based only on the abstract, I think), and so I can now answer the question of what, exactly, an RNA enzyme is, and how the research group got it to replicate itself.

An RNA enzyme, it turns out, is not a true enzyme. That is, it isn't a protein made up of amino acids. It's actually a strand of RNA. The particular RNA enzymes this group made look kind of like a capital T with one side of the crossbar a lot longer than the other. Like all RNA, they're made up of nucleotides (a nucleotide is a molecule consisting of a sugar molecule--ribose, in the case of RNA--a phosphate group, and a nitrogenous base). (I am forced to conclude that the "RNA" in "RNA enzyme" is an adjectival form, rather than a description of what the enzyme catalyzes.)

To understand how the enzyme works, you first need to know a bit about bonding in nucleic acids (DNA and RNA). What follows is a brief discussion; details can be found in any introductory biology textbook.

A single strand of a nucleic acid is a polymer (a really big molecule made up of a lot of similar, smaller subunits called monomers). As mentioned above, the monomers in nucleic acids are nucleotides. When nucleotides join together to form a nucleic acid, the sugars and phosphates bond together to form a "backbone." The nitrogenous bases stick off one side of the backbone. There are five nitrogenous bases that can form nucleotides: thymine, adenine, uracil, guanine, and cytosine. They are abbreviated T, A, U, G, and C, respectively. A, T, G, and C are found in DNA; RNA contains A, U, G, and C. So, a single strand of RNA looks kind of like half a ladder; the rungs are A, U, C, and G molecules. A double-stranded nucleotide (such as DNA) looks like a full ladder; the base in each "rung" is bonded to another base on a rung on the other side of the ladder. The bonded bases form a full rung. (Of course, a DNA molecule really looks like a twisted ladder, but the physics of why it twists isn't important for our purposes here.)

These bases aren't just any random molecules, though. As it turns out, their molecular structures force them to bond together in specific ways: A can bind only with T or U, and G can bind only with C (and vice versa, in each case). In a double-stranded nucleotide, therefore, each rung is made up of either a C-G pair or an A-T (or A-U if it's RNA) pair. You can probably see the beauty of this arrangement: it means that if you have one half of a double strand of RNA or DNA, you can construct the other half.

As I mentioned before, the RNA enzymes in this study look like lopsided Ts. The stem of the T is actually a double strand of RNA: part of the RNA molecule has bonded to itself. (A similar structure is found in some kinds of RNA that take part in transcription and translation in eukaryotic cells.) The crossbars of the T are single strands of RNA.

Each enzyme forms from two smaller pieces of RNA: a straight piece (called "B") and a piece that looks like a regular (i.e., not lopsided) T (called "A"). The straight piece binds to one of the crossbars of the T-shaped piece to form the lopsided T (which the researchers refer to as "E", for enzyme).

Each enzyme (and each sub-enzyme piece) actually exists in two "mirror-image" forms (i.e., E and E', A and A', and B and B'). The mirror-image forms can bind to each other because of the way the bases pair. However, A doesn't bind to A', or B to B'. Instead, A binds to B', and B bonds to A'. The A-B' combination forms E; the A'-B combination forms E'. [EDIT: the previous sentences should read "Instead, A binds to B, and B' binds to A'. The A-B combination forms E; the A'-B' combination forms E'."] The drawing below shows my lame attempt to summarize.

Essentially, when the researchers put some E into a mixture of A, B, A', and B', the A' and B' pieces bonded to the E to form molecules of E'. Once there was some E' in the mixture, the A and B molecules could bond to it to form new E molecules, and Presto! self-replicating RNA.

Of course, it wasn't really that simple. And actually, the not-simple part is kind of cool: The original E that the researchers used wasn't very efficient at catalyzing its own formation. So, basically, the researchers evolved it. They generated new A and B with mutations--variations in the sequences of bases on the backbone--and selected the ones that formed E that could replicate itself most quickly.

Because they have groovy tools (such as polymerase chain reaction machines) and computer to do the analysis, they were able to try a whole lot of different combinations in order to find the set of A and B that produced the most efficient E.

All in all, a really groovy little study!

Lincoln, Tracey A., and Gerald F. Joyce, 2009. "Self-sustained replication of an RNA enzyme." Sciencexpress. published online 8 January 2009; 10.1126/science.1167856.