Phylogenetic Distribution of Trans-Splicing

arlin at ac.dal.ca arlin at ac.dal.ca
Wed Apr 29 10:52:24 EST 1992


In article <1992Apr28.113028.11678 at husc3.harvard.edu>, robison1 at husc10.harvard.edu (Keith Robison) writes:
> 
> 
> Okay, this one came up in a discussion today.  Would anyone care
> to speculate as to why trans-splicing of mRNAs have such a wierd
> wierd phylogenetic distribution.  As far as I know, the only proven
> cases are in Trypanosome, C.elegans, and the chloroplast of Chlamydomonas.
> It probably also occurs in Euglena.  In Trypanosomes and C.elegans, trans
> splicing appears to be widespread and it may also be common in Euglena, so
> why is it seemingly present in only these lineages?
> 
A distinction has to be made between different types of trans-
splicing.  In chlamydomonas chloroplasts (and also wheat
mitochondria), some genes with group II self-splicing introns are
divided into two transcription units, and the split occurs in an
intron.  When transcribed, secondary structure interactions between
the two halves of the split intron are presumably sufficient to bring
the two parts together; once the intron is re-assembled, it can splice
itself out.

The form of trans-splicing in kinetoplasts and nematodes is distinct
in that a) spliceosomal introns are involved, not self-splicing
introns; b) one of the partners in the reaction is always the same
sequence; and c) this sequence is a 5' leader-- it is always the first
exon spliced onto the first exon of the other partner in the reaction.
The splicing machinery employed (snRNPs, etc) and the reaction
(forming a "Y" structure, like a cleaved lariat) show that the
reaction is homologous to canonical spliceosomal splicing of introns.
It has been suggested that the 5' leader exon spliced onto transcripts
in kinetoplasts and nematodes is actually a modified form of the U1
snRNA, part of the normal splicing machinery.  The 5' leader sequence 
provides the mRNAs with the 5' cap that is essential for transport 
out of the nucleus.

{for review on mechanisms, see Laird, 1989 (TIG 5: 204) or Green, 1991
(Ann Rev Cell biol 7: 559)}

But this mechanistic distinction does little to answer Keith's
question, about the distribution of trans-splicing. In the case of
organellar group II introns participating in trans-splicing, one could
suggest that such abberations could arise by neutral evolution in
organelles. Dividing a gene into two transcription units would
normally be difficult to achieve, since the halves would be unlikely
to be co-regulated.  However, in organelles there is very little gene
regulation at the transcriptional level, so that most transcriptional
units are transcribed in the same amounts. A gene that is divided by a
DNA rearrangement into two transcription units might still function if
it divides the group II intron felicitously.  Such divided genes would
tend to accumulate in organelles if it is difficult to generate the
specific allele fusing the genes back together.  Therefore, my guess
is that trans-splicing of divided group II introns will have a broad
distribution in organellar genomes-- if we look hard enough, we'll
find it in many phyla, especially in chloroplast genomes, which tend
to be bigger and more prone to rearrangement than mitochondrial ones.

I have little understanding of why trans-splicing of leader sequences
occurs in trypanosomes and nematodes, or whether it is limited to
them.  Perhaps the first type of splicing to evolve was trans-splicing, 
and perhaps this was used to distinguish mRNAs that had to be exported 
(if the nucleus was originally an endosymbiont, then there was at some
time a transition period in which some of the translation occurred in
the nucleus and some occurred in the cytoplasm). Or perhaps trans-
splicing of leaders evolved for some other reason as an alternative 
method of capping mRNAs.

Arlin Stoltzfus 
arlin at ac.dal.ca



More information about the Mol-evol mailing list