Vertebrate evolution: doubling and shuffling with a full deck

doi:10.1016/S0168-9525(02)00008-2

Trends in Genetics

Volume 19, Issue 1, January 2003, Pages 2-5

https://doi.org/10.1016/S0168-9525(02)00008-2 Get rights and content

Abstract

The number and role of whole-genome duplications in vertebrate evolution has intrigued evolutionary biologists since Ohno first proposed genome duplication as the force driving the ‘big leap’ in vertebrate morphological innovation. Attempts to resolve these issues have been thwarted by small and noisy datasets, and by lack of computational accuracy and statistical rigor. Recently, Ken Wolfe and colleagues presented a genome-scale, statistically rigorous analysis of evidence based on the spatial organization of duplicated genes, as well as estimates of duplication times. Their results provide the strongest evidence to date of large-scale duplication throughout the vertebrate genome, consistent with at least one whole-genome duplication.

Section snippets

Testable predictions of 2R

Proposed testable predictions of the 2R hypothesis fall into two categories, spatial and temporal. Spatial analyses are based on the contention that local similarities in gene content should still be discernible in the modern genome, despite extensive rearrangement and gene loss following polyploidization. However, conserved synteny (see Glossary) between paralogous genes, presented in early studies, is not constitute rigorous proof of polyploidization [4]. More recent studies 4, 7, 8 report

Computational genome scale analysis

In the absence of a complete vertebrate genome sequence, previous studies relied on partial datasets. Now that the whole-genome sequence for human is available, can 2R be resolved? The careful study presented by McLysaght et al. [14], based on a comprehensive spatial and temporal approach, addresses many of the problems that plagued earlier analyses. McLysaght et al. use Monte Carlo significance testing to validate the spatial results, and they validate the temporal evidence by estimating gene

New results and remaining challenges

The picture that emerges is of widespread, large-scale duplication, be it polyploidy, aneuploidy (chromosomal duplication) or many sub-chromosomal duplications, followed by substantial gene loss. These conclusions are supported by both the spatial and the temporal evidence. Forty-four percent of the genome was covered by 96 paralogons with sm≥6, and a majority of the genes analyzed were duplicated during the period 0.4D–0.7D, where D is the time since the fly–vertebrate split. This is

Can 2R be resolved?

The results presented by McLysaght et al., as well as those of other genome-scale studies 6, 7, 9, are consistent with at least one round of polyploidization, but could also be explained by many smaller block duplications and do not strongly support 2R. Resolution of the alternative hypotheses will be helped by more sequence data of greater accuracy, including better assembly and gene prediction for the human genome, additional vertebrate genomes, additional invertebrate outgroup genomes and

Beyond 2R

Although recent studies have focused on 2R, Ohno's seminal work [1] also presented a broader vision of whole-genome duplication as a force that could enable the leap in morphological and developmental complexity seen in modern vertebrates. Since 1970, genomics has revealed several other mechanisms that promote diversity of protein function on a genomic scale, including alternative splicing [19] and domain shuffling [6]. Widespread segmental duplication and domain accretion (recently reported by

Acknowledgements

I thank C. Beshers, L. Hurst, E.W. Jones, A.J. Lopez and D. Sankoff for helpful comments. I was supported by NIH grant 1 K22 HG 02451-01 and a David and Lucille Packard Foundation fellowship.

Glossary

Conserved synteny:: In a comparative map, the co-occurrence of homologous genes on the same chromosome in both genomes.
Gene conversion:: Nonreciprocal recombination resulting in identical subsequences on both chromosomes.
Molecular clock:: The hypothesis that the accepted mutation rate is roughly constant in all lineages.
Monte Carlo significance testing:: An approach to hypothesis testing in which the distribution of the test statistic under the null hypothesis is simulated using randomization.

References (25)

J.S. Taylor et al.
2R or not 2R?
Trends Genet.
(2001)
S. Li
Comparative analysis of human genome assemblies reveals genome-level differences
Genomics
(2002)
E.E. Eichler
Recent duplication, domain accretion and the dynamic mutation of the human genome
Trends Genet.
(2001)
S. Ohno
Evolution by Gene Duplication
(1970)
W. Makalowski
Are we polyploids? A brief history of one hypothesis
Genome Res.
(2001)
L. Skrabanek et al.
Eukaryote genome duplication – where's the evidence?
Curr. Opin. Genet. Dev.
(1998)
K.H. Wolfe
Yesterday's polyploids and the mystery of diploidization
Nat. Rev. Genet.
(2001)
Initial sequencing and analysis of the human genome
Nature
(2001)
J.C. Venter
The sequence of the human genome
Science
(2001)
D. Durand et al.
Tests for gene clustering

X. Gu

Age distribution of human gene families shows significant roles of both large- and small-scale duplications in vertebrate evolution

Nat. Genet.

(2002)

A.L. Hughes

Phylogenies of developmentally important proteins do not support the hypothesis of two rounds of genome duplication early in vertebrate history

J. Mol. Evol.

(1999)

Cited by (44)

Impact of gene gains, losses and duplication modes on the origin and diversification of vertebrates
2013, Seminars in Cell and Developmental Biology
Citation Excerpt :
Besides, due to the modular nature of regulatory sequences, the DDI model (Duplication Degeneration Innovation) argues a step forward to DDC, suggesting that recently degenerated regulatory sequences are more prone to acquire, by random mutation, new binding sites for transcription factors, hence facilitating neofunctionalization and innovation [44]. Neofunctionalization and subfunctionalization appear, therefore, as the two main processes driving the functional fate of new genes generated by duplication events [37,43–52]. In many cases, neofunctionalization and subfunctionalization events have been associated to changes in cis-regulatory elements that might lead to adaptative changes in duplicated genes [43,44].
The study of the evolutionary origin of vertebrates has been linked to the study of genome duplications since Susumo Ohno suggested that the successful diversification of vertebrate innovations was facilitated by two rounds of whole-genome duplication (2R-WGD) in the stem vertebrate. Since then, studies on the functional evolution of many genes duplicated in the vertebrate lineage have provided the grounds to support experimentally this link. This article reviews cases of gene duplications derived either from the 2R-WGD or from local gene duplication events in vertebrates, analyzing their impact on the evolution of developmental innovations. We analyze how gene regulatory networks can be rewired by the activity of transposable elements after genome duplications, discuss how different mechanisms of duplication might affect the fate of duplicated genes, and how the loss of gene duplicates might influence the fate of surviving paralogs. We also discuss the evolutionary relationships between gene duplication and alternative splicing, in particular in the vertebrate lineage. Finally, we discuss the role that the 2R-WGD might have played in the evolution of vertebrate developmental gene networks, paying special attention to those related to vertebrate key features such as neural crest cells, placodes, and the complex tripartite brain. In this context, we argue that current evidences points that the 2R-WGD may not be linked to the origin of vertebrate innovations, but to their subsequent diversification in a broad variety of complex structures and functions that facilitated the successful transition from peaceful filter-feeding non-vertebrate ancestors to voracious vertebrate predators.
Natural alcohol exposure: Is ethanol the main substrate for alcohol dehydrogenases in animals?
2011, Chemico-Biological Interactions
Citation Excerpt :
However, the increase in the number of Zn-ADH isozymes does not correlate with the total number of protein-coding genes for each taxum. Therefore, the increase in Zn-ADH isozymes is not just a consequence of whole genome duplication observed through the evolution of vertebrates [43,44]. Indeed, animal Zn-ADH expansion proceeded mainly by tandem gene duplication in the same chromosome [45].
Alcohol dehydrogenase (ADH) activity is widely distributed in all phyla. In animals, three non-homologous NAD(P)⁺-dependent ADH protein families are reported. These arose independently throughout evolution and possess different structures and mechanisms of reaction: type I (medium-chain) ADHs are zinc-containing enzymes and comprise the most studied group in vertebrates; type II (short-chain) ADHs lack metal cofactor and have been extensively studied in Drosophila; and type III ADHs are iron-dependent/-activated enzymes that were initially identified only in microorganisms. The presence of these different ADHs in animals has been assumed to be a consequence of chronic exposure to ethanol. By far the most common natural source of ethanol is fermentation of fruit sugars by yeast, and available data support that this fruit trait evolved in concert with the characteristics of their frugivorous seed dispersers. Therefore, if the presence of ADHs in animals evolved as an adaptive response to dietary ethanol exposure, then it can be expected that the enzymogenesis of these enzymes began after the appearance of angiosperms with fleshy fruits, because substrate availability must precede enzyme selection. In this work, available evidence supporting this possibility is discussed.
Phylogenetic analyses reveal that type II ADHs suffered several duplications, all of these restricted to flies (order Diptera). Induction of type II Adh by ethanol exposure, a positive correlation between ADH activity and ethanol resistance, and the fact that flies and type II Adh diversification occurred in concert with angiosperm diversification, strongly suggest that type II ADHs were recruited to allow larval flies to exploit new restricted niches with high ethanol content.
In contrast, phyletic distribution of types I and III ADHs in animals showed that these appeared before angiosperms and land plants, independently of ethanol availability. Because these enzymes are not induced by ethanol exposure and possess a high affinity and/or catalytic efficiency for non-ethanol endogenous substrates, it can be concluded that the participation of types I and III ADHs in ethanol metabolism can be considered as incidental, and not adaptive.
Costimulatory receptors in jawed vertebrates: Conserved CD28, odd CTLA4 and multiple BTLAs
2007, Developmental and Comparative Immunology
CD28 family of costimulatory receptors is comprised of molecules with a single V-type extracellular Ig domain, a transmembrane and an intracytoplasmic region with signaling motifs. CD28 and cytotoxic T lymphocyte antigen-4 (CTLA4) homologs have been recently identified in rainbow trout. Other sequences similar to mammalian CD28 family members have now been identified using teleost, Xenopus and chicken databases. CD28- and CTLA4 homologs were found in all vertebrate classes whereas inducible costimulatory signal (ICOS) was restricted to tetrapods, and programmed cell death-1 (PD1) was limited to mammals and chicken. Multiple B and T Lymphocyte Attenuator (BTLA) sequences were found in teleosts, but not in Xenopus or in avian genomes. The intron/exon structure of btlas was different from that of cd28 and other members of the family. The Ig domain encoded in all the btla genes has features of the C-type structure, which suggests that BTLA does not belong to the CD28 family. The genomic localization of these genes in vertebrate genomes supports the split between the BTLA and CD28 families.
Germline and somatic diversification of immune recognition elements in Metazoa
2006, Immunology Letters
The histories of the immune systems of Metazoa during evolution are envisaged like as many adaptations to the continuous diversification of immune receptors and effectors genes under the pressure of changing environments.
A basic diversity of potential immune receptor genes existed in primitive Metazoa. Their subsequent recruitment into immunity, their diversification revolving around the conservation of signaling cascades was paralleled by cell specialization and the introduction of regulatory networks. Polymorphism, duplication and somatic mechanisms of diversification affected independently and still affect different gene families in many phyla, creating a greater variety of immune system exhibiting sometimes little homology but much analogy to one another. Diversity and multiplicity of receptors was generated by duplication and creation of multigene families. Independently in several phyla further diversity is created somatically by alternate splicing, somatic mutation, gene conversion and gene rearrangement. In several instances combinatorial usage of polypeptide chains or genes segments increases the repertoire of the recognition structures. Metazoa had to adapt to the conditions generated by this diversity: the control of expression of multiple genes and the risk of autoimmunity.
Large-Scale Gene and Ancient Genome Duplications
2005, The Evolution of the Genome
This chapter deals with an important subset of large-scale gene duplications (versus the small-scale events) and ancient duplications of whole genomes (versus more recent polyploidy in plants and animals). Duplications of genetic elements can occur by a variety of mechanisms and at different chromosomal and temporal scales. The chapter emphasizes on the techniques used to identify, date, and otherwise investigate such events, as illustrated by some key recent example. Large-scale duplication events have featured prominently in many taxa, even those with small genomes. It is becoming increasingly apparent that large-scale duplication events have featured prominently in many taxa, even those with small genomes. As the pace of complete genome sequencing continues to quicken, detailed investigations of this issue will become possible in an ever-widening array of species. This is indeed an exciting time in the study of genome biology, and one that will undoubtedly continue to alter the understanding of the evolutionary process at both genomic and geological scales.
Novel evolutionary relationship among four fish model systems
2004, Trends in Genetics
Knowledge of the correct phylogenetic relationships among animals is crucial for the valid interpretation of evolutionary trends in biology. Zebrafish, medaka, pufferfish and cichilds are fish models for development, genomics and comparative genetics studies, although their phylogenetic relationships have not been tested rigorously. The results of phylogenomic analysis based on 20 nuclear protein-coding genes confirmed the basal placement of zebrafish in the fish phylogeny but revealed an unexpected relationship among the other three species, contrary to traditionally held systematic views based on morphology. Our analyses show that medaka (Beloniformes) and cichlids (Perciformes) appear to be more closely related to each other than either of them is to pufferfish (Tetraodontiformes), suggesting that a re-interpretation of some findings in comparative biology might be required. In addition, phylogenomic analyses show that fish typically have more copies of nuclear genes than land vertebrates, supporting the fish-specific genome duplication hypothesis.

View all citing articles on Scopus

View full text

Trends in Genetics

Research FocusVertebrate evolution: doubling and shuffling with a full deck

Abstract

Section snippets

Testable predictions of 2R

Computational genome scale analysis

New results and remaining challenges

Can 2R be resolved?

Beyond 2R

Acknowledgements

Glossary

Glossary

Trends Genet.

Genomics

Trends Genet.

Evolution by Gene Duplication

Are we polyploids? A brief history of one hypothesis

Genome Res.

Eukaryote genome duplication – where's the evidence?

Curr. Opin. Genet. Dev.

Yesterday's polyploids and the mystery of diploidization

Nat. Rev. Genet.

Initial sequencing and analysis of the human genome

Nature

The sequence of the human genome

Science

Tests for gene clustering

Age distribution of human gene families shows significant roles of both large- and small-scale duplications in vertebrate evolution

Nat. Genet.

Phylogenies of developmentally important proteins do not support the hypothesis of two rounds of genome duplication early in vertebrate history

J. Mol. Evol.

Research Focus
Vertebrate evolution: doubling and shuffling with a full deck