Conceptually RNA should be able to
self replicate without the help of proteins. This is shown in figure 10.1. The original
strand serves as a template. New base pairs arrive and form weak bonds with their
complement. A can form a bond with U, and G can form a bond with C. After one replication,
two complementary strands exist. Another round of replication is necessary to duplicate
the original strand. The complement to the original strand is also free to make more
copies. While this is still the most promising theory for life's origin, this theory
seems to offer more problems than solutions. This is why the origin of life remains a
Figure 10.1: Conceptual Model for RNA Self Replication
On paper, this model is great. Nevertheless, it does not work all that well in the lab.
The problems were described by Joyce and Orgel as follows:
Most strands of RNA are unsuitable templates. The original RNA
molecule that serves as the template must contain a very high concentration of cytosine to
make process 1 in figure 10.1 viable.2,3 This situation is unlikely to be met
because as discussed earlier cytosine has no plausible prebiotic synthesis pathway and it
decays rapidly. Nevertheless, the original strand depicted in figure 10.1 meets the high C
The chain will not grow correctly unless a very specific
activation agent is used to activate the nucleotides. The activation agent of choice is
not ATP (GTP, CTP or UTP). While life uses these, if these activation agents are used
without proteins the phosphate bonds usually attach to the wrong carbon atom in ribose.2,3
ImpA, impG, impU and impC are the activation agents of choice. These activation agents
contain the same side group as the amino acid histidine, which is one of the three amino
acids that have not been synthesized in prebiotic experiments. Thus, it is unlikely that
these activation agents where present in the primordial soup.
The complement of the original chain will have a high G content.
This is inevitable due to the requirement for high C in the original chain. This is
problematic because RNA with a high concentration of guanine tends to fold up in such a
way that it cannot be an effective template for replication.2,3 Thus, the
second round of growth in figure 10.1 does not happen.
If different isomers of ribose are present, these isomers will
terminate the growing chain.2,3
Joyce and Orgel comment that In light of the available evidence, it seems unlikely
that a pair of complementary sequences can be found each of which facilitates the
synthesis of the other . . . 3
Just to add to the difficulties, if too many steps form in the replication
ladder (complementary bonds between base pairs), then the strands will never separate.4
Furthermore, figure 10.1 is oversimplified in that it does not show that in order for the
RNA strands to grow, an RNA enzyme is required to catalyze the reaction. Because a growing
chain cannot catalyze its own replication, two identical RNA molecules must arise
simultaneously in the soup. Each capable of replicating the other.
A pattern is beginning to emerge for the RNA world. The RNA world is a
speculative world without proteins where RNA is the most important molecule. RNA regulates
all chemical reactions and contains all of the molecular knowledge for life. The pattern
that is emerging is that perhaps this world is too speculative in that it may have never
Again Joyce and Orgel put it best: Scientists interested in the origins
of life seem to be divided neatly into two classes. The first, usually but not always
molecular biologists, believe that RNA must have been the first replicating molecule and
that chemists are exaggerating the difficulties of nucleotide synthesis . . . The second
group of scientists are much more pessimistic. They believe that the de nova appearance of
oligonucleotides on the primitive earth would have been a near miracle. The authors
subscribe to this latter view. Time will tell which is correct.3
One last point, RNA replication in the lab makes use of extensive
investigator interference. Chemicals like amino acids, aldehydes, and sugars (other than
ribose) are arbitrarily excluded. Very specific activation agents are used to encourage
replication (ImpA for adenine, ImpG for guanine, ImpC for cytosine, and ImpU for uracil).
The concentration of the chemicals (especially cytosine and ribose) is billions and
billions of orders of magnitude higher than what one would expect under plausible
Dynamite is being used to blow the door open in figure 9.4, and the door is
just too solid. It remains closed and the scientist remains trapped. Fortunately, many
scientists understand this, and they no longer claim that the door is open.
How Much Knowledge is Required to Create a Ribozyme
RNA molecules capable of facilitating chemical reactions do exist. Because such RNA
molecules perform a role traditionally carried out only by protein enzymes, they are
called ribozymes. Ribozymes have been shown to facilitate the creation of both peptide
bonds in proteins, and the bonds between phosphate and ribose in RNA. This discovery is
very significant in that it means RNA can both store and implement knowledge. It also
explains the popularity of RNA as the first living molecule.
Bartel carried out a very relevant experiment. In this experiment. 65
ribozymes were isolated from a pool of 1x1015 RNA molecules. All ribozymes
isolated contained at least 200 bases. This result allows for a direct calculation of the
knowledge in ribozymes. If 65 sequences have some minimal enzymatic activity out of a pool
containing 1015 random sequences, then one in every 15 trillion sequences is a
ribozyme. Thus the molecular knowledge is as follows: knowledge = 3.32 x log (15 trillion)
or 44 bits. Note that knowledge and not information is used because the 65 ribozymes were
not yet optimized. The experiment also subjected the ribozymes to several rounds of
selection in which only the best were chosen. Selection dramatically improved their
catalytic efficiency. Thus, Bartels experiment proves that both information and
knowledge can evolve under the guidance of natural selection.
Given the extreme difficulties associated with synthesizing an RNA molecule
containing 200 or more bases, it is unlikely that even one such molecule ever existed on
the primitive earth, and 15 trillion are needed to just get 65 functional ribozymes.
Furthermore, ribozymes are not self replicators. The knowledge required for self
replication is certainly many orders of magnitude more than the 44 bits required for a
marginally functional ribozyme. Finally, the 44 bits calculated above is in a test tube
where all competing side reactions are eliminated. If the real primordial soup contains
free amino acids, aldehydes, and undesirable isomers of ribose, then the 44 bits will
increase by a factor similar to the increase seen for the protein insulin in chapter 5.
Taking this last factor into account, the 44 bits is at least one order of magnitude too
Next: Molecular Knowledge in Primordial Soup
Previous: Self Replicating Peptides