can be very rapidly corrected. Thus, in a mutant of
Drosophila melanogaster which possessed only 190
histone genes because of a heterozygous chromosome mutation, the normal count of 275 genes was
almost fully restored in eight generations [64]. In
view of the effectiveness of gene conversion, the
question arises of which mechanism hinders the
equalization of sequences in neighbouring genes of
certain gene families, e.g. the globin genes. The
answer to this problem may lie in the fact that the
formation of the heteroduplex required for gene
conversion is dependent upon the sequence similarity of non-coding sequences, e.g. the introns.
Divergent evolution of these sequences, e.g. the
frequent insertion of Alu sequences in mammalian
globin gene clusters, reduces the probability of
gene conversion and allows the development and
maintenance of adaptive differences between the
sequences [39, 365].
As long as sufficient genes of a gene family
remain functional, individual genes may be inactivated by negative mutation without causing an
increase in selection pressure. Non-expressed
members of a gene family (pseudo genes) may
also arise by the insertion into the genome of
DNA copies of an RNA (processed or retropseudogenes; p. 25). In contrast to active genes, pseudogenes display many genetic alterations, including frame-shift and nonsense mutations, changes
in splicing signals, mutations of the initiation
codon, and small insertions or deletions. Pseudogenes have the highest known rates of substitution, e.g. in the case of the globin genes it is
approximately 1.9 times higher than the rate of
synonymous substitution. Furthermore, they
approach most closely the frequently expressed
idea of functionless DNA ("junk DNA")
[210, 212]. However, in many pseudo genes synonymous substitutions are more frequent than
amino acid exchanging substitutions, and they
display the same characteristic difference in the
substitution rate of non-coding regions as do active genes, albeit to a lesser degree (see Table 4.12). This might be explained by such genes
becoming inactive only after a certain period of
evolution [210, 212].
In contrast to the pseudo genes of other gene
families, the many pseudogenes of the immunoglobulin-VH family differ only to a minor
degree from the active genes, and mostly contain
only one inactivating mutation. This is undoubtedly due to continuous sequence correction by
horizontal evolution. However, there are also
individual pseudogenes in this family with many
inactivating mutations [73]. The processes of hori4.2.6 Polyploidization
125
zontal evolution also affect the pseudo genes of
multi-gene families. They can, for example, be
duplicated. The known examples of this phenomenon include the bovine ~-globin cluster (p. 263),
the mouse urinary protein (MUP) family, and the
human GAPDH genes [10, 132]. Because pseudogenes are mutated more often by deletion than
by insertion, these genes become shorter during
evolution [149]. Pseudo genes also represent a
potentially useful reserve of genetic information;
there is evidence for the formation of hybrid
genes by sequence transfer from pseudo- to active
genes, for example in certain allotypes of the
mouse IgCy gene and the rabbit IgCx gene [209].
4.2.6 Polyploidization
A doubling of the chromosome number polyploidization and thus the total genetic information is apparently an important mechanism of
evolution [304, 307]. The event can happen in
two ways: (1) the omission of meiotic reduction
produces diploid gametes which fuse to give tetraploid zygotes (autopolyploidy); (2) hybridization of two species results in the failure of
chromosome pairing and reduction in the next
meiosis (aUopolyploidy). Problems arise here in
obligatory out-crossing organisms, and thus polyploidy occurs less frequently in animals than in
flowering plants, where self-fertilization is quite
common. In the animal kingdom, polyploidy is
found in particular in the hermaphroditic oligochaetes, turbellarians and gastropods, and in parthenogenetic species of e.g. the Coleoptera,
Lepidoptera and shrimps. Animals with the XY/
XX mechanism of genotypic sex determination
have special problems with polyploidy; for example, an egg with two sets of auto somes and two Xchromosomes (2AXX egg) together with a 2AXY
sperm would form a 4AXXXY zygote with a disturbed XYbalance; for this reason, polyploidy is
quite often found in vertebrates such as fish and
amphibians, which have less stringent mechanisms for sex determination. It is almost completely absent in reptiles, birds and mammals (Table 4.6). Polyploidization in the cartilaginous fish
is difficult to demonstrate because of the basically
high chromosome numbers; however, the existence of tetra- and octoploid species may be concluded from the bimodal distribution of DNA
contents and renaturation kinetics [308].
Polyploidization produces an excess of genetic
information that must be processed during evolution and there are basically two possibilities:
Drosophila melanogaster which possessed only 190
histone genes because of a heterozygous chromosome mutation, the normal count of 275 genes was
almost fully restored in eight generations [64]. In
view of the effectiveness of gene conversion, the
question arises of which mechanism hinders the
equalization of sequences in neighbouring genes of
certain gene families, e.g. the globin genes. The
answer to this problem may lie in the fact that the
formation of the heteroduplex required for gene
conversion is dependent upon the sequence similarity of non-coding sequences, e.g. the introns.
Divergent evolution of these sequences, e.g. the
frequent insertion of Alu sequences in mammalian
globin gene clusters, reduces the probability of
gene conversion and allows the development and
maintenance of adaptive differences between the
sequences [39, 365].
As long as sufficient genes of a gene family
remain functional, individual genes may be inactivated by negative mutation without causing an
increase in selection pressure. Non-expressed
members of a gene family (pseudo genes) may
also arise by the insertion into the genome of
DNA copies of an RNA (processed or retropseudogenes; p. 25). In contrast to active genes, pseudogenes display many genetic alterations, including frame-shift and nonsense mutations, changes
in splicing signals, mutations of the initiation
codon, and small insertions or deletions. Pseudogenes have the highest known rates of substitution, e.g. in the case of the globin genes it is
approximately 1.9 times higher than the rate of
synonymous substitution. Furthermore, they
approach most closely the frequently expressed
idea of functionless DNA ("junk DNA")
[210, 212]. However, in many pseudo genes synonymous substitutions are more frequent than
amino acid exchanging substitutions, and they
display the same characteristic difference in the
substitution rate of non-coding regions as do active genes, albeit to a lesser degree (see Table 4.12). This might be explained by such genes
becoming inactive only after a certain period of
evolution [210, 212].
In contrast to the pseudo genes of other gene
families, the many pseudogenes of the immunoglobulin-VH family differ only to a minor
degree from the active genes, and mostly contain
only one inactivating mutation. This is undoubtedly due to continuous sequence correction by
horizontal evolution. However, there are also
individual pseudogenes in this family with many
inactivating mutations [73]. The processes of hori4.2.6 Polyploidization
125
zontal evolution also affect the pseudo genes of
multi-gene families. They can, for example, be
duplicated. The known examples of this phenomenon include the bovine ~-globin cluster (p. 263),
the mouse urinary protein (MUP) family, and the
human GAPDH genes [10, 132]. Because pseudogenes are mutated more often by deletion than
by insertion, these genes become shorter during
evolution [149]. Pseudo genes also represent a
potentially useful reserve of genetic information;
there is evidence for the formation of hybrid
genes by sequence transfer from pseudo- to active
genes, for example in certain allotypes of the
mouse IgCy gene and the rabbit IgCx gene [209].
4.2.6 Polyploidization
A doubling of the chromosome number polyploidization and thus the total genetic information is apparently an important mechanism of
evolution [304, 307]. The event can happen in
two ways: (1) the omission of meiotic reduction
produces diploid gametes which fuse to give tetraploid zygotes (autopolyploidy); (2) hybridization of two species results in the failure of
chromosome pairing and reduction in the next
meiosis (aUopolyploidy). Problems arise here in
obligatory out-crossing organisms, and thus polyploidy occurs less frequently in animals than in
flowering plants, where self-fertilization is quite
common. In the animal kingdom, polyploidy is
found in particular in the hermaphroditic oligochaetes, turbellarians and gastropods, and in parthenogenetic species of e.g. the Coleoptera,
Lepidoptera and shrimps. Animals with the XY/
XX mechanism of genotypic sex determination
have special problems with polyploidy; for example, an egg with two sets of auto somes and two Xchromosomes (2AXX egg) together with a 2AXY
sperm would form a 4AXXXY zygote with a disturbed XYbalance; for this reason, polyploidy is
quite often found in vertebrates such as fish and
amphibians, which have less stringent mechanisms for sex determination. It is almost completely absent in reptiles, birds and mammals (Table 4.6). Polyploidization in the cartilaginous fish
is difficult to demonstrate because of the basically
high chromosome numbers; however, the existence of tetra- and octoploid species may be concluded from the bimodal distribution of DNA
contents and renaturation kinetics [308].
Polyploidization produces an excess of genetic
information that must be processed during evolution and there are basically two possibilities: