identity and hence lead to homeotic phenotypes
upon mutation (for a review, see Gramzow and
Theißen 2010; Smaczniak et al. 2012; Theißen
et al. 2016). Five classes (A–E) of floral organ
identity genes have been identified by mutant
analysis, with class A + E genes specifying
sepals, A + B + E petals, B + C + E stamens,
C + E carpels, and C + D + E ovules (Theißen
et al. 2016). Almost all of these genes encode
MIKC
C -group MTFs.
The phylogeny of MIKC
C
-group genes is
characterized by preferential retention of duplicate genes after whole genome duplications, followed by sequence divergence, sub- and
neo-functionalization of the paralogs (Gramzow
and Theißen 2013, 2015, 2016). Increases in gene
numbers occurred independently in different
groups of land plants. Interestingly, the floral
homeotic genes are all members of gene clades
that are seed plant- or flowering plant-specific.
Phylogeny reconstructions revealed that all the
MIKC
C -group genes of angiosperms are members of 11 seed plant-specific superclades which
had been established already in the MRCA of
extant seed plants (spermatophytes) about 300
million years ago (MYA). None of these gene
clades may have existed already in the MRCA of
extant euphyllophytes, i.e. monilophytes (ferns
and their allies such as horsetails) and seed plants
(gymnosperms + angiosperms) about 400 MYA
(Gramzow et al. 2014). Among these, superclades
are those containing the genes providing the floral
homeotic A-function (FLC/SQUA-like, or FLC/
AP1-like genes), containing class B genes (DEF/
GLO/OsMADS32-like, or AP3/PI/OsMADS32like genes), containing class C and D genes (AGlike genes), and containing class E genes
(SEP/AGL6-like genes). Due to gene duplications
in the stem group of angiosperms, these genes
evolved into 17 clades that had already been
established in the MRCA of extant angiosperms
(Fig. 9.1). These angiosperm-specific clades
include distinct DEF- (AP3-) and GLO- (PI-) like
genes (class B), the AG-like and STK-like genes
(class C and D), the AGL2-like (SEP1-like),
AGL9-like (SEP3-like) and AGL6-like genes
(class E), and the SQUA- (AP1-) like genes (class
A) (Gramzow et al. 2014). Genes of the
remaining clades are involved in diverse developmental processes ranging from root to fruit
development. These clades comprise AGL12-like,
AGL15-like, AGL17-like, FLC-like, GGM13-like
(B sister ), OsMADS32-like, StMADS11-like (SVPlike), TM3-like (SOC1-like) and TM8-like genes
(alternative clade names given in brackets)
(Smaczniak et al. 2012).
9.3 When Less Is More: Loss
of MADS-Box Genes During
Evolution
Somewhat similar to living beings also genes are
born (e.g. by gene duplication or de novo from
non-genic DNA), and eventually, they die (even
though that may take millions of years). Mutational gene death (nonfunctionalization) and loss
is probably the most frequent fate of duplicated
genes, but compared to gene birth, it has found
relatively little scientific interest (Panchy et al.
2016). This ignorance has probably at least two
reasons, a scientific and a technical one. On the
one hand, gene loss might be easily viewed as the
trivial outcome of random mutations in sequences without a function on which hence purifying
selection is not acting anymore. On the other
hand, gene loss is much more difficult to
demonstrate than gene birth; while one can
conclude that a gene birth must have happened at
some time from the simple presence of a gene in
an organismic lineage, demonstration of gene
loss in a rigorous way requires solid evidence of
the absence of the gene in an organismic lineage
and hence typically depends on reliable
whole-genome information. Comparably little is
known, therefore, about the mechanisms and
dynamics of the loss of genes during evolution.
Nevertheless, the interest in gene loss has
rapidly increased recently. One reason is the
availability of rising numbers of whole-genome
sequences of high quality from diverse taxa,
which facilitates the identification of gene loss.
Another reason is mounting evidence that gene
loss can be of considerable adaptive value (for a
review, see Albalat and Cañestro 2016; Hoffmeier et al. 2018).
94
L. Gramzow and G. Theißen
upon mutation (for a review, see Gramzow and
Theißen 2010; Smaczniak et al. 2012; Theißen
et al. 2016). Five classes (A–E) of floral organ
identity genes have been identified by mutant
analysis, with class A + E genes specifying
sepals, A + B + E petals, B + C + E stamens,
C + E carpels, and C + D + E ovules (Theißen
et al. 2016). Almost all of these genes encode
MIKC
C -group MTFs.
The phylogeny of MIKC
C
-group genes is
characterized by preferential retention of duplicate genes after whole genome duplications, followed by sequence divergence, sub- and
neo-functionalization of the paralogs (Gramzow
and Theißen 2013, 2015, 2016). Increases in gene
numbers occurred independently in different
groups of land plants. Interestingly, the floral
homeotic genes are all members of gene clades
that are seed plant- or flowering plant-specific.
Phylogeny reconstructions revealed that all the
MIKC
C -group genes of angiosperms are members of 11 seed plant-specific superclades which
had been established already in the MRCA of
extant seed plants (spermatophytes) about 300
million years ago (MYA). None of these gene
clades may have existed already in the MRCA of
extant euphyllophytes, i.e. monilophytes (ferns
and their allies such as horsetails) and seed plants
(gymnosperms + angiosperms) about 400 MYA
(Gramzow et al. 2014). Among these, superclades
are those containing the genes providing the floral
homeotic A-function (FLC/SQUA-like, or FLC/
AP1-like genes), containing class B genes (DEF/
GLO/OsMADS32-like, or AP3/PI/OsMADS32like genes), containing class C and D genes (AGlike genes), and containing class E genes
(SEP/AGL6-like genes). Due to gene duplications
in the stem group of angiosperms, these genes
evolved into 17 clades that had already been
established in the MRCA of extant angiosperms
(Fig. 9.1). These angiosperm-specific clades
include distinct DEF- (AP3-) and GLO- (PI-) like
genes (class B), the AG-like and STK-like genes
(class C and D), the AGL2-like (SEP1-like),
AGL9-like (SEP3-like) and AGL6-like genes
(class E), and the SQUA- (AP1-) like genes (class
A) (Gramzow et al. 2014). Genes of the
remaining clades are involved in diverse developmental processes ranging from root to fruit
development. These clades comprise AGL12-like,
AGL15-like, AGL17-like, FLC-like, GGM13-like
(B sister ), OsMADS32-like, StMADS11-like (SVPlike), TM3-like (SOC1-like) and TM8-like genes
(alternative clade names given in brackets)
(Smaczniak et al. 2012).
9.3 When Less Is More: Loss
of MADS-Box Genes During
Evolution
Somewhat similar to living beings also genes are
born (e.g. by gene duplication or de novo from
non-genic DNA), and eventually, they die (even
though that may take millions of years). Mutational gene death (nonfunctionalization) and loss
is probably the most frequent fate of duplicated
genes, but compared to gene birth, it has found
relatively little scientific interest (Panchy et al.
2016). This ignorance has probably at least two
reasons, a scientific and a technical one. On the
one hand, gene loss might be easily viewed as the
trivial outcome of random mutations in sequences without a function on which hence purifying
selection is not acting anymore. On the other
hand, gene loss is much more difficult to
demonstrate than gene birth; while one can
conclude that a gene birth must have happened at
some time from the simple presence of a gene in
an organismic lineage, demonstration of gene
loss in a rigorous way requires solid evidence of
the absence of the gene in an organismic lineage
and hence typically depends on reliable
whole-genome information. Comparably little is
known, therefore, about the mechanisms and
dynamics of the loss of genes during evolution.
Nevertheless, the interest in gene loss has
rapidly increased recently. One reason is the
availability of rising numbers of whole-genome
sequences of high quality from diverse taxa,
which facilitates the identification of gene loss.
Another reason is mounting evidence that gene
loss can be of considerable adaptive value (for a
review, see Albalat and Cañestro 2016; Hoffmeier et al. 2018).
94
L. Gramzow and G. Theißen