Why Genomes in Pieces?
Laura F. Landweber
Microbial
eukaryotes take ample de-tours along the route from DNA to messenger RNA (mRNA)
and protein. Some of their tricks continue to erode the notion of a gene beyond
its natural sub-division into functional exons and noncoding introns. Two
discontinuous genetic systems described in this issue further challenge this
dogma. Marande and Burger report a fully scrambled mitochondrial genome in Diplonema
papillatum, a free-living relative
of disease-causing trypanosomes, and Soma et al. describe a set of scrambled transfer RNA (tRNA) genes
in the nuclear genome of the red alga Cyanidioschyzon merolae. The findings are reminders that a genome sequence
can be a far cry from knowledge of gene products.
Marande
and Burger explode the notion of a gene with mRNA building blocks present as
“modules” of ~165 base pairs, each on a separate chromosome in the
mitochondria of the protist D. papillatum.
Construction of a com-plete mRNA requires joining up to nine modules through a
mechanism that appears distinct from known forms of RNA splicing, the processes
that join exons in eukaryotic mRNA. Although split genes occur in other systems
(including Chlamydomonas, Euglena,
Alveolata, plants, and Diptera), rarely are the scrambled
pieces “sewn” back together to cre-ate a contiguous gene or RNA. Some
excep-tions are gene unscrambling in ciliates and the bursicon gene in mosquitoes. Trans-splicing of
RNA and even proteins can also merge functional regions located on dispersed
elements of prokaryotic or eukar-yotic genomes.
The pathway for
gene assembly in diplone-mid mitochondria may provide a clue to the origin of
U-insertional RNA editing, which makes a modest appearance in the report by
Marande and Burger as six non–DNA-encoded uracil (U) residues that join two RNA
modules. Perhaps the ancestral role of guide RNAs that direct U insertion and
deletion in the related kinetoplastid protozoa was to provide a tem-plate
scaffold to link modules. Such small anti-sense RNAs may later have gained a
role in RNA editing, possibly under selective pressure to repair a region or
restore a reading frame after loss or erosion of a module.
Soma
et
al. describe a new layer of tRNA
processing in the red alga C. merolae:
circularly permuted tRNAs, with the coding region for the
RNA 3'
end
located upstream of the coding region for its 5' end.
Although circular permutation has been a laboratory tool for the study of RNA
structure and function for years, true biological occurrences were previously
known only in phage and ciliate mitochondria. Maturation of tRNA is an
elaborate RNA- and protein-driven cascade of clipping, coiffing, and adorning
an initial RNA tran-script. Soma et al. add one
more decryption step to this assembly line.
Why do quirky genetic
architectures emerge and persist? Some genetic systems may pro-vide a source of
evolutionary novelty. For example, module recycling or shuffling could generate
new gene products without destroying the old ones or requiring duplication. So
far, the gene products in D. papillatum seem conventional, but examination of the mitochondrial proteome may tell
otherwise. Some genetic systems may be molecular fossils—neutral vestiges of
the past without any special benefits. It is parsimonious to assume that the
earliest organisms had split genes, for instance, but because all extant life
has been evolving for the same length of time from a common ancestor, one
cannot infer the preservation of ancestral genome organization without detailed
mapping of ancestral and derived characters on a reliable phylogeny.
Another possible
explanation for rococo genetic systems is atavism, in which some biological
mechanisms revert back to an ancestral state, although presumably with
modification, in a new, derived genetic back-ground. Some of these events may
appear to recapitulate features of primitive genomes, providing indirect clues
as to how early genetic systems could have functioned.
There is also pure
chance, a scenario that is probably slightly deleterious. Unconstrained by
dogma and size, why shouldn’t microbial life explore a broad range of
possibilities? Protists, often reproducing asexually in the wild, would
gradually accumulate small mutations and genome rearrangements that would be
crippling without a mechanism to mitigate the effect. Acquisition of a new
mechanism may be successful if the organism can recruit a preexisting cellular
function or template for repair or rearrangement and then elaborate on the
basic mechanism, leading to fixation and expansion of a complex genetic system.
Reductive
evolution could account for the svelte genome size (16.5 Mb) of C. merolae and perhaps even some of its quirky
genome architecture, if a few spandrels arose as by-products of genome
compaction. This is consistent with its recent placement as a derived lineage
within an outgroup of red algae. Genome reduction may lead to intrachromosome
rearrangement or overlapping genes in related protists, as either a consequence
of, or adaptation to, small size. Germ-line rearrangements can also yield gene
duplications, which would be trimmed back under the sword of reductive
evolution. Thus, the model that Soma et al. propose for the origin of a per-muted tRNA gene is feasible,
albeit via secondary acquisition: tRNA gene duplications could emerge along
with other germ-line rearrangements, and then the 5' end of the upstream gene would be lost,
as well as the 3' end of a
downstream gene, leaving the organ-ism no choice but to exploit such a
resulting permuted gene, if it can. The only option would be to rescue it by
adding a few more acrobatic steps to the already-complex tRNA-processing
cascade. Clearly there is a need for a suite of tRNA sequences, or better yet,
comparative genomes, both closely and distantly related to C. merolae, to decipher the evolutionary history
of its permuted tRNA genes.
Evolution is a
tinkerer, and its products are not necessarily neat or elegant. Like a Rube
Goldberg invention, it builds upon existing parts, embracing all their
gawkiness but gradually smoothing out operations with optimization over time.
The biological results are often robust systems that, in the case of protists,
may not seem so at first glance.
Click here to read original articles