Dissertation/Thesis Abstract

Biophysical Properties of Ancestrally Resurrected Proteins Are Unbiased
by Beckett, Brian Clifford, Ph.D., Brandeis University, 2019, 102; 22589040
Abstract (Summary)

The evolutionary history of a protein is necessary to fully understand its modern form. Ancestral sequence reconstruction (ASR) methods have been widely used to determine the biophysical properties of ancestral proteins. Despite the success of ASR, the accuracy of its reconstructions is currently unknown as it is largely impossible to compare reconstructed proteins to the true ancestors that have gone extinct. Proteins reconstructed by ASR have been found to be highly thermal stable, which has then been used to argue for an ancient, hot environment for early life. However, it is currently controversial whether there is a significant bias in the biophysical properties of ancestrally resurrected proteins, such as thermal stability and enzyme activity. One way to test for bias in the properties of the single-most probable (SMP) ancestrally reconstructed protein is by comparison with an unbiased estimate of the ancestral properties, such as the average of ancestor proteins sampled from the ancestral posterior distribution. I reconstructed the SMP and sampled possible alternate ancestors at four nodes in the phylogeny of Apicomplexan malate and lactate dehydrogenases, resurrected these enzymes, and characterized their biophysical properties (molecular weight, extinction coefficient at 280nm, log(kcat), log(KM), log(kcat/KM), Tm, and ΔH). For all ancestral nodes, there was no significant bias in any global property when comparing the SMP to the sampled proteins. Similarly, evolutionary age was not correlated with bias nor with sequence probability. Neither was there any correlation between the probability of the sequence and its global property, nor between the thermal stability and function, at any ancestral node. These data support the widespread use of the SMP ancestor as representative of the true ancestor at any given ancestral node and suggests that ASR methodology is more resilient to bias than researchers have anticipated.

Indexing (document details)
Advisor: Theobald, Douglas L.
Commitee: Kern, Dorothee, Miller, Chris, Fournier, Gregory P.
School: Brandeis University
Department: Biochemistry
School Location: United States -- Massachusetts
Source: DAI-B 81/3(E), Dissertation Abstracts International
Subjects: Biochemistry, Bioinformatics, Biophysics
Keywords: Ancestral sequence reconstruction, Apicomplexa, Enzyme kinetics, Phylogenetics, Thermal stability
Publication Number: 22589040
ISBN: 9781088364673
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy