Subfunctionalization of duplicated genes as a transition state to neofunctionalization.

Rastogi S, Liberles DA. · 2005

Read Subfunctionalization of duplicated genes as a transition state to neofunctionalization. by Rastogi S, Liberles DA. (2005) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc1112588.txt.

Introduction

Rastogi and Liberles model what happens to spare gene copies after duplication in this 2005 study. Gene duplication is a key source of new traits in evolution. One fate, called neofunctionalization, gives one copy a wholly new job. Another, subfunctionalization, splits the old job between the two copies. Which fate dominates was unclear.

The authors built lattice model genes, simple simulated proteins that fold and bind two targets. Each gene was duplicated in a small simulated population with no mixing of genes. Models that allowed job splitting reached new functions through it, they found. Job splitting acted as a transition state toward a new function, not as a final fate. No pressure kept both copies identical, they note. The result reframes how duplicated genes are retained.

Catalog record

CREATORRastogi S, Liberles DA.
YEAR2005
COLLECTIONPAPERS
FORMATTXT
FILE SIZE18 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC1112588

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Subfunctionalization of duplicated genes as a transition state to neofunctionalization By Shruti Rastogi, David A Liberles Published in: BMC Evolutionary Biology DOI: 10.1186/1471-2148-5-28 ABSTRACT Background Gene duplication has been suggested to be an important process in the generation of evolutionary novelty. Neofunctionalization, as an adaptive process where one copy mutates into a function that was not present in the pre-duplication gene, is one mechanism that can lead to the retention of both copies. More recently, subfunctionalization, as a neutral process where the two copies partition the ancestral function, has been proposed as an alternative mechanism driving duplicate gene retention in organisms with small effective population sizes. The relative importance of these two processes is unclear. Results A set of lattice model genes that fold and bind to two peptide ligands with overlapping binding pockets, but not a third ligand present in the cell was designed. Each gene was duplicated in a model haploid species with a small constant population size and no recombination. One set of models allowed subfunctionalization of binding events following duplication, while another set did not allow subfunctionalization. Modeling under such conditions suggests that subfunctionalization plays an important role, but as a transition state to neofunctionalization rather than as a terminal fate of duplicated genes. There is no apparent selective pressure to maintain redundancy. Conclusion Subfunctionalization results in an increase in the preservation of duplicated gene copies, including those that are neofunctionalized, but never represents a substantial fraction of duplicate gene copies at any evolutionary time point and ultimately leads to neofunctionalization of those preserved copies. This conclusion also may reflect changes in gene function after duplication with time in real genomes. BACKGROUND A number of mechanisms can generate duplicate copies of genes, ranging from single gene duplications to regional and whole genome duplications [1-3]. Large increases in gene number have been coupled to increases in organismal complexity and radiative divergence at several points in the history of metazoans including during the chordate/vertebrate transition and during the teleost fish divergence [1,4,5]. Metazoans differ from prokaryotes in their much smaller effective population sizes, where theory predicts that neutral stochastic processes will be relatively more important than adaptive processes in the expected case that adaptive mutations are rarer than nearly neutral mutations [6]. Large scale analyses, based upon the ratio of nonsynonymous to synonymous nucleotide substitution rates [7] or MacDonald-Kreitman statistics [8] have indicated small to intermediate degrees of positive selection (adaptive substitutions) in mammals, but these clearly do not represent the majority of substitutions. In such studies, it appears to be specific positions in protein-encoding genes, rather than the genes as a whole that are under positive selection [7]. Even examining substitution as a neutral walk through sequence in a folded protein (ignoring positive selection) has shown such a process to have fairly complex dynamics [9]. From this, it is relevant to examine population genomic phenomena, like the fates of duplicated genes, in the context of physical models of proteins. Further, it is not possible to systematically identify fates of real genes (subfunctionalization to the exclusion of neofunctionalization or vice-versa), so modeling under increasingly realistic conditions is likely to be the best way to understand evolutionary mechanisms. Pseudogenization or nonfunctionalization is a purely neutral process that ultimately eliminates one of the duplicated copies as a functional gene and is the most common fate. Subfunctionalization, is an alternative neutral process that leads to an increase in organismal gene number for genes or functions that show modularity (one representative type of modularity is modeled here, but other types are also possible). Neofunctionalization is an alternative process leading to an increase in organismal gene number, but dependent upon rarer adaptive mutations. Neofunctionalization can include the evolution of a completely new binding capability (as modeled here) or modification/improvement of existing binding capabilities under positive selection after removal of pleiotropic constraint. These alternative fates are presented in the context of a lattice model in Figure 1. [Figure: Eight different fates are possible in the two simulations, nonfunctionalization (pseduogenization) without cellular death, subfunctionalization, pleiotropic neofunctionalization plus either nonfunctionalization of the other copy or subfunctionalization, non-pleiotropic neofunctionalization, non-pleiotropic neofunctionalization also involving subfunctionalization, redundancy, and cellular death. Some of the fates have additional combinations of activity that are not represented in this figure. It should also be noted that some of the characterizations overlap. For example, pleiotropic neofunctionalization occurs in combination with nonfunctionalization, subfunctionalization, or is redundant.] Lattices are models of folded proteins in square or cubic shapes (a cubic lattice was employed here). The folding of a lattice is dictated by the contacts from amino acids that are not adjacent in the primary sequence (these contacts are present in the folded and unfolded states). Because lattices are small and the folding rules are simple, they can be used for evolving populations of proteins to study their structural properties. Lattice models have previously been used to make important predictions about the behavior of proteins in evolutionary contexts, including their metastability [10] and the evolvability of new folds [11]. Lattices that bind to peptides [12] and small hydrophobic molecules [13] have been described and the latter used to show that subfunctionalization can lead to an increase in duplicate gene retention rates. Here, model genes that fold into lattices and bind peptides were duplicated, with neofunctionalization and subfunctionalization (simulation A) or just neofunctionalization (simulation B) as possible events that would preserve duplicated copies in a genome, with nonfunctionalization (pseudogenization) as an alternative fate (see Figure 1 and Table 1). The relative levels of duplicate gene preservation and the importance of both neofunctionalization and subfunctionalization were assessed. [Table: The initial amino acid sequences including binding sites and folding energies (RT units) are shown for the 10 proteins. Additionally, the three ligands for each protein are shown with their initial binding energies (also RT units) at the two sites (A in italics and B in bold). It should be noted that the two binding sites are adjacent in three dimensions and overlap by one amino acid.] RESULTS AND DISCUSSION A set of 10 stably folded lattices was designed to each bind to 2 different ligands at overlapping sites. A third ligand was present in the cell, but did not bind at either site at the start of the simulation. The lattice was duplicated in a constant population of 1000 cells, where those cells that bound the third ligand were 5% more likely to appear in the next generation (a selection coefficient of 5% is arbitrary, but only serves as a scaler of the results). In each generation, 10% of molecules became nonfunctional at random through transcriptional knock-out. The fitness function required molecules to fold and genomes to have binding capabilities for the first two ligands. Cells were selected under the constraint that the first and second ligands needed to be bound, but could be bound by either molecule (subfunctionalization possible) in simulation A. The second simulation (simulation B) tightened this constraint and required both ligands to be bound to the same molecule (subfunctionalization not possible). In simulation B, only neofunctionalization is possible as a mechanism to preserve both copies non-redundantly. Neofunctionalization can occur through two mechanisms, a pleiotropic mechanism where the third ligand binds at a site that is also

Read the complete text