Saturday, 9 May 2009

The Central Dogma of Molecular Biology

I thought I'd venture just a little outside of medicine today.  After his co-discovery of the structure of DNA in 1953, Francis Crick's next keen interest was in elucidating how the DNA-RNA-protein sequence unfolded.  At that stage, it was still unclear as to which of the 9 'transformations' were possible.  For instance, could DNA transfer its information to RNA directly, or did it have to go through the intermediate of a protein? Could a cell subsist on RNA alone (without DNA)?  Could proteins generate DNA?  Could you create RNA from other RNA, and could you make DNA from an RNA template?  The nine possibilities for information transfer between each of the three actors had varying degrees of experimental backing, but little in the way of theoretical guidance.  In Crick's own words:

All we had to work on were certain fragmentary experimental results, themselves often rather uncertain and confused, and a boundless optimism that the basic concepts involved were rather simple and probably much the same in living things.  In such a situation well constructed theories can play a really useful part in stating problems clearly and thus guiding experiment. 

With roughly that desire in mind, Crick famously argued in 1958 that the 9 possible transformations could be divided into three groups.

The first group consisted of those for which there was positive evidence.  At the time, this was limited to:
  • DNA → DNA
  • DNA → RNA
  • RNA → Protein
  • RNA → RNA
The second group dealt with those transfers for which there lacked any evidence or theoretical requirement, but which were nonetheless felt to be be plausible, even if they turned out to be rare or absent from the world.  These were:
  • RNA → DNA
  • DNA → Protein
The last group were felt, on theoretical reasons (rather than on lack of experimental support), to be extremely unlikely to occur.  They were:
  • Protein → Protein
  • Protein → RNA
  • Protein → DNA
As you can see, Crick was very skeptical of the idea that information could be transferred from a protein to anything else. (Briefly, he thought that the complicated machinery for translating the DNA/RNA 'alphabet' into a completely different amino acid 'alphabet' couldn't work backwards, and there was no evidence of any alternative complicated machinery to perform this back-translation.) Proteins were thus dead ends for information, and this insight constituted his 'Central Dogma of Molecular Biology': "once (sequential) information has passed into protein it cannot get out again."

Just as Crick had hoped, this insight proved to be both valid and enormously useful.  Incidentally, the first of the second group's constituents has now found experimental support, largely in the form of the retroviruses (like HIV), which initially use their own RNA to back-translate their information into DNA.  Will we one day discover a "DNA → Protein" transformation?

Crick wrote an extremely accessible defense of his Central Dogma in the journal Nature in 1970. You can access the PDF by clicking here; I strongly recommend it for its wonderful clarity and its knack of hitting nails on heads.  It's superb.


  1. genes are regulated by proteins so in some ways information does travel from proteins -> DNA, though it's not the same kind of information. i seem to remember that barbara mcclintock had some difficulty presenting her research on gene regulation because the central dogma held that information traveled in only one direction from DNA -> RNA -> protein

  2. Yes, indeed - thanks for the great comment. The central dogma describes direct information transfer between various 'alphabets' (DNA/RNA/amino acid). It says nothing of the various regulation processes (i.e. different information) that interact with such transfers.

    The Central Dogma doesn't for a second outlaw feedback systems from protein to DNA; it only says that the protein's "code" can't be directly converted back to DNA or RNA "code".

    This distinction (which you also clearly see) is rather subtle, I suppose.