NEW YORK (GenomeWeb News) – For a protein that has been around for about 4 billion years, thioredoxin has gone through remarkably few changes, researchers from the University of Granada in Spain reported today in the journal Stucture.
Rather than taking the more standard approach of examining a number of extant proteins from a range of organisms, Granada's Jose Sanchez-Ruiz and his colleagues turned to comparing the X-ray crystal structures of seven Precambrian-era, laboratory-resurrected thioredoxin proteins to several modern-day versions of that protein.
"So far, attempts to understand protein structure evolution have been based on the comparison between structures of modern proteins. This is equivalent to trying to understand the evolution of birds by comparing several living birds," said Sanchez-Ruiz, a co-senior author of the study, in a statement. "But it is most useful to study fossils so that changes over evolutionary time are apparent. Our approach comes as close as possible to 'digging up' fossil protein structures."
To uncover those long-buried protein structures, Sanchez-Ruiz and his colleagues built on previous work they did constructing a phylogenetic tree based on some 200 modern-day thioredoxin amino acid sequences. Using the resulting phylogeny, they were able to 'resurrect' a number of Precambrian-era protein sequences, with a focus on the thioredoxin proteins traced to the last bacterial, archael, archael-eukaryotic, and eukaryotic common ancestors.
They crystallized those ancient sequences through either counter-diffusion or hanging drop vapor diffusion and generated X-ray crystal structure data that was then compared to crystal structures for a number of modern-day proteins spanning the various domains of life.
Overall, the thioredoxin proteins contained the canonical thioredoxin fold and adhered to the general pattern of β1α1β2α2β3α3 at the N-terminus and β4β5α4 at the C-terminus, with a central section of three parallel and two anti-parallel pleated β sheets, surrounded by four helices.
Some variation, though, was apparent in the length of the helix α1. For example, the modern human and modern Escherichia coli helix α1 lengths differ; most bacterial versions of thioredoxin are much shorter than their eukaryotic counterparts, according to Sanchez-Ruiz and his colleagues.
This, they added, "leads to one obvious question: which of the structural features (long helix versus short helix) is ancestral and which is derived?"
The resurrected proteins have a short helix α1, indicating to the researchers that that is likely the ancestral state, while the long helix α1 common to eukaryotes is a derived state.
"The putative ancestral structures reported here are consistent with the thioredoxin fold being an approximate four billion-year-old molecular fossil of sorts and confirms that protein structures can evolve slowly," Sanchez-Ruiz and his colleagues wrote.
The investigators noted that this structural similarity exists despite a number of sequence differences.
"[We] speculate that the evolution of protein structures may be sometimes described as a type of punctuated equilibrium, with long periods of stasis while switch-like structural transitions occur over comparatively short periods," they added.