166
Xenopus
is considered to be brought about by the accumulation of
mutations (base substitutions, deletions and insertions) in
the genome sequence, which is a continuous and gradual
change. On the other hand, allopolyploidization causes the
sudden combination of genetic diversities from two different
species. This means that allopolyploidization, given that the
resulting heterosis is in favor of natural selection, can lead to
“discontinuous evolution” whereby a new species is suddenly
born that is superior to the parent species. One can imagine that the two rounds of WGD in the common ancestor of
vertebrates were possibly allopolyploidization to make two
consecutive giant leaps forward in evolution.
Allotetraploidy of the Xenopus laevis genome is a disadvantage for genomic analysis but became an advantage to
analyze genome evolution of allopolyploidized species, since
this complex genome was successfully decoded, and moreover, subgenomes L and S were clearly identif ed. Luckily,
there are no massive reciprocal translocations between chromosomes, thereby revealing asymmetric evolution between
the subgenomes, in which subgenome L is more conserved
than subgenome S. Typical asymmetries are gene clusters
of rDNA, nodal3, nodal5, and vg1(S20), all of which were
retained only in subgenome L. The need for appropriate
gene expression dosage may be a reason for maintaining
these asymmetric clusters. By contrast, in a gene cluster of
subfunctionalized paralogs, such as the Hox clusters and the
ventx cluster, almost all homeologous genes are retained in
the subgenomes. Using updated versions of the X. laevis and
X. tropicalis WGS will give us the chance to discover other
types of asymmetric evolution of subgenomes. The WGS of
X. tropicalis and X. laevis has also provided many insights.
We now know (almost) all of the genomic composition of
these two species, and we are able to extract the information
that is necessary for detailed analysis.
10.5. FUTURE DIRECTIONS
What we learned from the genomes of X. tropicalis and X.
laevis goes beyond the analyses of these two species and
has pointed out that the other members of this genus will be
interesting subjects for evolutionary studies and much more.
We will discuss here one of those, how allopolyploidization
impacts initial and early subgenome evolution. To address
this issue, there are two alternative but complementary
ways: one is comparative genome analysis between X. laevis
and X. borealis, and the other is an experimental approach
using hybrid formation and polyploidization.
10.5.1. COMPARATIVE GENOME ANALYSIS
Currently, the completion of the WGS of X. borealis (SRA
Accession No. SRX1606064) is underway. A full analysis
of this genome and comparison with the X. laevis genome
would defnitely reveal how allopolyploid genomes evolve.
That is, because X. laevis and X. borealis speciated just ~1
million years after allopolyploidization (Session et al., 2016),
early events (mutations and rearrangements) which were
common between X. laevis and X. borealis can be separated
from later events which are not shared. Furthermore, genome
sequences and transcriptome analyses of other Xenopus species with higher ploidy levels (see Figure 10.1A) would give
information on what kinds of subgenomes are present, how
genomes cope with many subgenomes, how gene expression
is regulated among the homeologs, and so on.
10.5.2. ARTIFICIAL HYBRID ANALYSIS
To experimentally analyze allopolyploidization, the Xenopus
genus provides us a good system, because it is possible to
make hybrids and allopolyploids between them as reported
(see subsequently). When divergent genomes are merged by
hybridization or allopolyploidization, a “genomic shock”
occurs in which the subgenomes come into conf ict with
each other (McClintock, 1984; Bird et al., 2018). Studies of
plant genomes have shown gene loss, genomic rearrangements, and reactivation of previously silenced transposable
elements (reviewed in Bird et al., 2018), as well as nucleolar
dominance (reviewed in Preuss and Pikaard, 2007). These
early alterations probably lead to “subgenome dominance”
and thereby to asymmetric evolution of subgenomes.
In the Xenopus genus, naturally occurring interspecif c
hybrids have been reported (for example, Picker et al., 1996;
Fischer et al., 2000; Yager, 1996 ), exemplifying the preevents of allopolyploidization recurring during the course of
evolution. Experimentally, hybrids between Xenopus species
have been generated for biochemical analysis (for example,
Honjo and Reeder, 1973; Brown et al., 1977; De Robertis and
Black, 1979; Kobel et al., 1981; Bürki, 1985). More recently,
Xenopus interspecifc hybrids have been examined at molecular and gene levels for nucleolar dominance (Michalak et al.,
2015) or changes in the epigenome (Elurbe et al., 2017) to
identify early effects of merging of two genomes. Of note,
Elrube et al. (2017) found derepression of young DNA transposons, which is analogous to the reaction of plant genomes
at genomic shock. Further analyses of these interspecif c
hybrids of Xenopus are expected to reveal the mechanism
of (sub)genome dominance after genomic shock, which may
trigger asymmetric evolution of subgenomes if hybridization
will be followed by polyploidization.
10.5.3. ARTIFICIAL ALLOPOLYPLOID ANALYSIS
While hybrids are in general infertile, females of experimentally generated Xenopus interspecies hybrids can be
partially fertile. This is because the diploid hybrid females
produce two types of eggs, aneuploid and diploid (2x)
(Müller, 1977; Kobel and Du Pasquier, 1986). Diploid eggs
are produced from endoreduplicated (DNA is replicated in
the nucleus without division) oocytes (4x) and are larger in
size. When these diploid eggs are fertilized with haploid
sperm from a non-hybrid male, triploid individuals develop
(Kobel and Du Pasquier, 1975). The triploid females similarly produce triploid eggs (3x) that can be fertilized again
with haploid sperm to yield tetraploids (Kobel, 1996 ). On
Xenopus
is considered to be brought about by the accumulation of
mutations (base substitutions, deletions and insertions) in
the genome sequence, which is a continuous and gradual
change. On the other hand, allopolyploidization causes the
sudden combination of genetic diversities from two different
species. This means that allopolyploidization, given that the
resulting heterosis is in favor of natural selection, can lead to
“discontinuous evolution” whereby a new species is suddenly
born that is superior to the parent species. One can imagine that the two rounds of WGD in the common ancestor of
vertebrates were possibly allopolyploidization to make two
consecutive giant leaps forward in evolution.
Allotetraploidy of the Xenopus laevis genome is a disadvantage for genomic analysis but became an advantage to
analyze genome evolution of allopolyploidized species, since
this complex genome was successfully decoded, and moreover, subgenomes L and S were clearly identif ed. Luckily,
there are no massive reciprocal translocations between chromosomes, thereby revealing asymmetric evolution between
the subgenomes, in which subgenome L is more conserved
than subgenome S. Typical asymmetries are gene clusters
of rDNA, nodal3, nodal5, and vg1(S20), all of which were
retained only in subgenome L. The need for appropriate
gene expression dosage may be a reason for maintaining
these asymmetric clusters. By contrast, in a gene cluster of
subfunctionalized paralogs, such as the Hox clusters and the
ventx cluster, almost all homeologous genes are retained in
the subgenomes. Using updated versions of the X. laevis and
X. tropicalis WGS will give us the chance to discover other
types of asymmetric evolution of subgenomes. The WGS of
X. tropicalis and X. laevis has also provided many insights.
We now know (almost) all of the genomic composition of
these two species, and we are able to extract the information
that is necessary for detailed analysis.
10.5. FUTURE DIRECTIONS
What we learned from the genomes of X. tropicalis and X.
laevis goes beyond the analyses of these two species and
has pointed out that the other members of this genus will be
interesting subjects for evolutionary studies and much more.
We will discuss here one of those, how allopolyploidization
impacts initial and early subgenome evolution. To address
this issue, there are two alternative but complementary
ways: one is comparative genome analysis between X. laevis
and X. borealis, and the other is an experimental approach
using hybrid formation and polyploidization.
10.5.1. COMPARATIVE GENOME ANALYSIS
Currently, the completion of the WGS of X. borealis (SRA
Accession No. SRX1606064) is underway. A full analysis
of this genome and comparison with the X. laevis genome
would defnitely reveal how allopolyploid genomes evolve.
That is, because X. laevis and X. borealis speciated just ~1
million years after allopolyploidization (Session et al., 2016),
early events (mutations and rearrangements) which were
common between X. laevis and X. borealis can be separated
from later events which are not shared. Furthermore, genome
sequences and transcriptome analyses of other Xenopus species with higher ploidy levels (see Figure 10.1A) would give
information on what kinds of subgenomes are present, how
genomes cope with many subgenomes, how gene expression
is regulated among the homeologs, and so on.
10.5.2. ARTIFICIAL HYBRID ANALYSIS
To experimentally analyze allopolyploidization, the Xenopus
genus provides us a good system, because it is possible to
make hybrids and allopolyploids between them as reported
(see subsequently). When divergent genomes are merged by
hybridization or allopolyploidization, a “genomic shock”
occurs in which the subgenomes come into conf ict with
each other (McClintock, 1984; Bird et al., 2018). Studies of
plant genomes have shown gene loss, genomic rearrangements, and reactivation of previously silenced transposable
elements (reviewed in Bird et al., 2018), as well as nucleolar
dominance (reviewed in Preuss and Pikaard, 2007). These
early alterations probably lead to “subgenome dominance”
and thereby to asymmetric evolution of subgenomes.
In the Xenopus genus, naturally occurring interspecif c
hybrids have been reported (for example, Picker et al., 1996;
Fischer et al., 2000; Yager, 1996 ), exemplifying the preevents of allopolyploidization recurring during the course of
evolution. Experimentally, hybrids between Xenopus species
have been generated for biochemical analysis (for example,
Honjo and Reeder, 1973; Brown et al., 1977; De Robertis and
Black, 1979; Kobel et al., 1981; Bürki, 1985). More recently,
Xenopus interspecifc hybrids have been examined at molecular and gene levels for nucleolar dominance (Michalak et al.,
2015) or changes in the epigenome (Elurbe et al., 2017) to
identify early effects of merging of two genomes. Of note,
Elrube et al. (2017) found derepression of young DNA transposons, which is analogous to the reaction of plant genomes
at genomic shock. Further analyses of these interspecif c
hybrids of Xenopus are expected to reveal the mechanism
of (sub)genome dominance after genomic shock, which may
trigger asymmetric evolution of subgenomes if hybridization
will be followed by polyploidization.
10.5.3. ARTIFICIAL ALLOPOLYPLOID ANALYSIS
While hybrids are in general infertile, females of experimentally generated Xenopus interspecies hybrids can be
partially fertile. This is because the diploid hybrid females
produce two types of eggs, aneuploid and diploid (2x)
(Müller, 1977; Kobel and Du Pasquier, 1986). Diploid eggs
are produced from endoreduplicated (DNA is replicated in
the nucleus without division) oocytes (4x) and are larger in
size. When these diploid eggs are fertilized with haploid
sperm from a non-hybrid male, triploid individuals develop
(Kobel and Du Pasquier, 1975). The triploid females similarly produce triploid eggs (3x) that can be fertilized again
with haploid sperm to yield tetraploids (Kobel, 1996 ). On