¿ìè¶ÌÊÓƵ

In the 1980s, molecular biology’s founding creed – that DNA encodes
RNA which encodes protein – suffered a series of hefty blows. One of the
heftiest came from a puzzling genetic phenomenon discovered in the flagellum
of Trypanosoma brucei, an insect-borne parasite that causes sleeping sickness
in humans. In 1986 it emerged that the parasite has an unorthodox talent:
the ability to edit the information stored in some of its genes. Today,
after five years of hectic research and the discovery of many more examples
of gene editing, molecular biologists are close to unravelling its complex
molecular origins.

Robert Benne and his colleagues at the University of Amsterdam in the
Netherlands stumbled upon the phenomenon while examining genes belonging
to the parasite’s mitochondrion. (Like all so-called ‘kinetoplastid’ protozoans,
T. brucei has a single, large mitochondrion at the base of its flagellum.)
One of these genes, coding for a protein fragment, called COII, of the enzyme
cytochrome oxidase, was odd. One of its hundreds of linked nucleotides was
missing, something which should have rendered its coded information meaningless.
Yet – and here was the mystery – the parasite produced flawless copies of
the COII protein.

The key to the puzzle lay in RNA. The parasite was cannily correcting
the information by adding the missing nucleotide – a uridine unit – but
only after the gene had been copied into RNA. The concept of RNA editing
was thus born, shattering the dogma that genetic information always flows
unaltered from DNA into proteins. Certain cells, it seemed, could tamper
with their genetic information, and even add to it, en route.

Benne’s finding was by no means unique. Further examples of RNA editing
have since come to light, both in T. brucei and in two other kinetoplastid
protozoa, Leishmania tarentolae and Crithidia fasciculata. Most involve
genetic information being altered by the insertion of one or more uridines
into the RNA copy of a defective gene. In rare cases, editing results in
the deletion of excess uridines from RNA.

But RNA editing is not always done on a modest scale. In 1988, in the
first of a series of such finds, researchers at the Seattle Biochemical
Research Institute in the US discovered a gene in T. brucei that is edited
by the addition of no less than 398 uridine units at 158 different sites.
The gene coded for the third protein fragment (COIII) of the parasite’s
cytochrome oxidase enzyme. More than half the information for making the
protein came not from its gene but from added code.

Molecular biologists were stunned. Here was a ‘cryptogene’ – something
that by all normal criteria was unrecognisable as a gene, yet which coded
for a functional protein. Where did the extra information come from? How
did the molecular machinery responsible for the editing know how and where
to add the extra code? The mystery deepened when it became clear that there
is no pattern to the sequences of RNA that are selected for editing.

Today, the emerging consensus is that the extra genetic information
is supplied by special guide molecules. These molecules, themselves made
of RNA, are presumed to act as templates, guiding enzymes to remove or insert
uridine units at specific points along an RNA chain. The key to the operation
is that the sequence of each guide molecule is almost complementary to that
of its target RNA chain. So guide molecules can home in on and entwine with
RNA chains that need editing.

Though not yet enshrined as doctrine, this theory was bolstered last
year by Larry Simpson and his colleagues at the University of California
in Los Angeles. The researchers discovered small genes that appear to code
for guide molecules in the mitochondrion of L. tarentolae. Other studies
have shed light on exactly how the different parasites deploy their own
particular guide molecules.

In L. tarentolae, it seems that the editing machinery slides along an
RNA chain, until it encounters a region with a sequence that complements
that of a guide molecule. It then grabs the guide molecule, cuts the RNA
chain and inserts uridines according to the instructions held in the guide
molecule (see Figure). Enzymes finish the job by splicing the two halves
of the edited RNA chain. By contrast, the editing machinery of T. brucei
inserts uridines randomly, producing a plethora of different edited versions.
Guide RNA molecules then select from these the ones that have been edited
correctly.

All this begs a question: how and why did such baroque genetic mechanisms
evolve? One theory (but by no means the only one) invokes gene duplication.
Millions of years ago, it says, certain genes belonging to an ancestral
parasite became duplicated. Some of the duplicates then split into fragments,
becoming the precursors of the genes that today encode guide RNA molecules.
Later on, some of the remaining intact genes evolved defects, which the
parasite was able to correct using RNA templates made from the fragmented
genes.

According to this theory, the molecular mechanisms that underpin RNA
editing must have evolved before the three kinetoplastid parasites diverged
from a common ancestor 200 to 300 million years ago. But some of the gene
defects which are corrected by editing must have arisen more recently. In
the case of the COIII gene, for example, most of the defects occur only
in L. tarentolae – a sign that they evolved after T. brucei and L. tarentolae
diverged between 100 and 150 million years ago.

Some researchers suspect that RNA editing might be derived from a much
earlier evolutionary epoch. Molecular templates similar to guide molecules
probably played a key part in enabling primeval genes to replicate. One
idea is that RNA editing evolved as a way of maximising the diversity of
information that could flow from a single such gene. Along with RNA catalysts,
self-splicing introns and RNA viruses, RNA editing may be a fascinating
relic from an ancient world dominated by RNA.

Another important but as yet unresolved question is, why RNA editing
has persisted in the mitochondria of kinetoplastid protozoans. What evolutionary
advantage does it confer?

The unproven assumption is that RNA editing provides kinetoplastid parasites
with a subtle way of regulating the production of certain proteins. To make
functional versions of the proteins encoded by their defective genes, each
parasite must first switch on the production of guide molecules – and this
gives an extra level of control. Intriguingly, RNA editing occurs in only
one of the two stages of T. brucei’s life cycle: when the parasite is in
its insect host rather than in a mammal.