In the previous posting, the researchers noted that “NRs appear to be a metazoan innovation, because they are absent from the genomes of choanoflagellates, fungi, plants, and prokaryotes.”
Yet the hypothesis of front-loading would predict these nuclear receptors that play key roles in animal development and homeostasis would exist in some single-celled organisms. So I would argue that this observation of the NRs as a metazoan innovation is premature. That is, from the perspective of front-loading, I predict that NRs will be found to predate the appearance of metazoa and be found in some unicellular organisms.
There are two points to consider. First, we have just scratched the surface in terms of protozoan genomes that have been sequenced. Just because the NR gene has not been detected in the small number of sequenced genomes that currently exist does not mean it is a metazoan innovation.
Secondly, sequence analysis itself is likely to scratch the surface. That is, homologs may indeed exist that are not detectable by aligning amino acid sequences.
For example, there is a bacterial protein known as FtsZ. It binds GTP and is involved in cell division. It is homologous to the eukaryotic protein, tubulin. Yet the clear evidence of this relationship was not known until the actual physical structures of both proteins were solved and compared, as it turns out their structure is almost identical.
Yet this homologous relationship would not be discovered by BLASTing sequences. For example, FtsZ from the bacterium Staphylococcus aureus is 390 amino acids in length while tubulin from rat is 451 amino acids in length. If you use the FtsZ sequence to probe the rat genome, tubulin is not retrieved. In fact, no homolog is retrieved. That’s because if you align the two sequences with BLAST, the best match involved a stretch of only 52 amino acidsand has an E value of 0.013.
Put simply, BLASTing cannot detect all homologies.
So let me again predict that as scientists sequence more genomes, develop better alignment programs, and accumulate more biochemical data, that a homolog of metazoan nuclear receptors will be found in a unicellular organism. For front-loading does indeed expect:
We predict that, when sufficient data are gathered to allow detailed evolutionary reconstructions, it will become apparent that most protein superfamilies diversified by subtle modification and partial degradation of ancient, deeply homologous functions. Invoking the evolution of wholesale “novelty” will seldom be necessary.