I have a few problems with Mark Siddall's reply to Jan Ulrik Thorup.
1. Thorup asked about experimental observation of intron loss. The
5.8S-LSU rRNA spacer is *not* an intron - no religation occurs.
2. The situation in Microsporidia is not well established as, to my
knowledge, only in V. necatrix has the absence of 5.8S/LSU processing
been shown directly - it is possible that this is a Vairimorpha-
3. In any case, the absence of 5.8S/LSU processing in V. necatrix is
almost certainly a derived condition. This is based not on phylogeny
but rather on comparative secondary structure analysis. (I mentioned
this in a 1987 paper - J Mol Evol, 25, 343). The rRNA stem and loop
in which the processed spacer is found in other eukaryotes is also
found in eubacteria and archaes (without being processed) but it is
absent in Vairimorpha. The most parsimonious explanation is a lineage
specific deletion of this stem (and thus of the processing site).
Several conserved rRNA structural elements are also missing from the
V. necatrix SSU-rRNA, indicating that the missing 5.8S/LSU stem is
part of an overall reduction in size of this organism's ribosomal
Getting back to Jan Ulrik Thorup's original question, I do not
believe that anyone has observed intron loss *experimentally*, i.e.
comparison of parental and derived lines of the same organism showing
loss of an intron in the latter. However, I believe there are
examples of intron distributions for which the most parsimonious
explanation is a lineage specific loss, for example the rat insulin
I'd be interested to hear what it is Jan Ulrik Thorup has seen.
C. Graham Clark, Ph.D.
Laboratory of Parasitic Diseases,
National Institutes of Health,
Bethesda, MD 20892
e-mail: cge at cu.nih.gov