activity was investigated on 1000 pure lines arising from one Drosophila pair; within only 300
generations, alcohol dehydrogenase activities
were found to vary by a factor of 10-20 [283]. In
various populations of D. melanogaster, a mean
frequency of null alleles of about 0.25 % was
observed at 20 autosomal loci [224]. Whilst
sequence variants of proteins are often selectively
neutral, quantitative variants especially when
they are homozygous, do influence fitness and are
thus subject to selection. Appropriate investigations have been carried out, for example, on the
a-amylase locus of D. melanogaster. Adults and
larvae with higher amylase activities (Amy4,6)
have a significantly greater life expectancy than
those with lower activities (Amyl) when starch is
the limiting nutrient [216].
Heritable differences in enzyme activity can
arise by changes in either the catalytic properties
or the synthesis or degradation, and therefore the
concentration, of the enzyme protein. Changes in
catalytic properties are always due to amino acid
substitutions. These need not necessarily influence the active centre directly but can change the
physicochemical properties of the protein and
thereby affect enzyme function indirectly. For
example, in various Drosophila null variants, the
ability to form active heterodimers with normal
subunits has been lost. The three alleles of
glucose-6-phosphate dehydrogenase in D. melanogaster code for enzymes of different quaternary
structure: Zw A and Zw B , which are active dimers
and tetramers, and ZW lO1 , which is monomeric
and of limited activity [50, 124].
Heritable changes in the rate of enzyme synthesis can involve very different types of mutation; for example, mutations in the signals for
transcription and mRNA maturation, mutations
in regulatory genes, or the influence of gene
expression by rearrangements or translocations.
The null allele adh nAH52 of D. melanogaster has
been sequenced. It contains an insertion of 8 bp
and an adjacent deletion of 2 bp in intron 2; as a
result, the transcription rate is reduced to onetenth of the normal [134]. The rate of synthesis of
multiple genes can be influenced by the number
of active genes. A frame-shift mutation, i.e. the
addition or deletion of one or two nucleotides in
a coding sequence, or a nonsense mutation, i.e.
the production of a stop codon within a coding
sequence, prevents formation of the normal
enzyme protein; resulting anomalous polypeptides are usually immediately degraded proteolytically. Whether activity variants are based upon
changes in enzyme concentration or catalytic
4.3.7 DNA Polymorphism
137
properties can mostly be determined from immunological measurements of protein concentration; however, an amino acid substitution can
also change immunological reactivity and catalytic activity. Unambiguous analysis requires a
knowledge of the DNA sequence.
4.3.7 DNA Polymorphism
Protein comparison reveals only a part of the genetic variation present in a natural population;
synonymous nucleotide substitutions and variants
in non-coding sequences can only be recognized
by analysis of the DNA itself. Protein polymorphism normally involves the substitution of single
nucleotides in a coding sequence; DNA polymorphism can also result from deletions, insertions or duplications of longer sequences and
from other types of rearrangements. The widely
found length polymorphisms of non-coding
sequences are usually based on a variable number
of short repetitive sequences; examples are the
rDNA of D. melanogaster [71], the chicken vitellogenin gene [329], and the human albumin, globin and insulin genes [309]. In multi-gene families, the number of copies can vary or parts of a
cluster may be duplicated or missing; examples
include the rDNA of many animals, human aand ~-globin genes, human immunoglobulin
genes and the amylase genes of several rodents,
and these are described in more detail elsewhere
in this book. Polymorphism can also result from
the exchange of partial sequences between the
members of a multi-gene family, as described for
the human y-globin gene (Fig. 4.6a).
As DNA sequencing requires considerable
effort, relatively few comparative investigations
have been undertaken on sequences originating
from the two homologous chromosomes of cells
of an individual or from different individuals of a
species or population. The most extensive data on
the polymorphism of DNA sequences come from
the genes for human haemoglobin [308, 443].
Two different nucleotides were found in each of
eight positions at the B-globin locus, and in one
position there were in fact three differences; the
combination of these characters defines at least
seven haplotypes that are distributed in human
populations [361]. Two types (R and T) of the
human o-globin gene found in all populations differ from each other by 16 substitutions and two
deletions, which are mainly in the non-coding 5'
non-translated region; the sequences differ by
almost 1 % in this region of 1. 7 kb [257]. 1\vo
generations, alcohol dehydrogenase activities
were found to vary by a factor of 10-20 [283]. In
various populations of D. melanogaster, a mean
frequency of null alleles of about 0.25 % was
observed at 20 autosomal loci [224]. Whilst
sequence variants of proteins are often selectively
neutral, quantitative variants especially when
they are homozygous, do influence fitness and are
thus subject to selection. Appropriate investigations have been carried out, for example, on the
a-amylase locus of D. melanogaster. Adults and
larvae with higher amylase activities (Amy4,6)
have a significantly greater life expectancy than
those with lower activities (Amyl) when starch is
the limiting nutrient [216].
Heritable differences in enzyme activity can
arise by changes in either the catalytic properties
or the synthesis or degradation, and therefore the
concentration, of the enzyme protein. Changes in
catalytic properties are always due to amino acid
substitutions. These need not necessarily influence the active centre directly but can change the
physicochemical properties of the protein and
thereby affect enzyme function indirectly. For
example, in various Drosophila null variants, the
ability to form active heterodimers with normal
subunits has been lost. The three alleles of
glucose-6-phosphate dehydrogenase in D. melanogaster code for enzymes of different quaternary
structure: Zw A and Zw B , which are active dimers
and tetramers, and ZW lO1 , which is monomeric
and of limited activity [50, 124].
Heritable changes in the rate of enzyme synthesis can involve very different types of mutation; for example, mutations in the signals for
transcription and mRNA maturation, mutations
in regulatory genes, or the influence of gene
expression by rearrangements or translocations.
The null allele adh nAH52 of D. melanogaster has
been sequenced. It contains an insertion of 8 bp
and an adjacent deletion of 2 bp in intron 2; as a
result, the transcription rate is reduced to onetenth of the normal [134]. The rate of synthesis of
multiple genes can be influenced by the number
of active genes. A frame-shift mutation, i.e. the
addition or deletion of one or two nucleotides in
a coding sequence, or a nonsense mutation, i.e.
the production of a stop codon within a coding
sequence, prevents formation of the normal
enzyme protein; resulting anomalous polypeptides are usually immediately degraded proteolytically. Whether activity variants are based upon
changes in enzyme concentration or catalytic
4.3.7 DNA Polymorphism
137
properties can mostly be determined from immunological measurements of protein concentration; however, an amino acid substitution can
also change immunological reactivity and catalytic activity. Unambiguous analysis requires a
knowledge of the DNA sequence.
4.3.7 DNA Polymorphism
Protein comparison reveals only a part of the genetic variation present in a natural population;
synonymous nucleotide substitutions and variants
in non-coding sequences can only be recognized
by analysis of the DNA itself. Protein polymorphism normally involves the substitution of single
nucleotides in a coding sequence; DNA polymorphism can also result from deletions, insertions or duplications of longer sequences and
from other types of rearrangements. The widely
found length polymorphisms of non-coding
sequences are usually based on a variable number
of short repetitive sequences; examples are the
rDNA of D. melanogaster [71], the chicken vitellogenin gene [329], and the human albumin, globin and insulin genes [309]. In multi-gene families, the number of copies can vary or parts of a
cluster may be duplicated or missing; examples
include the rDNA of many animals, human aand ~-globin genes, human immunoglobulin
genes and the amylase genes of several rodents,
and these are described in more detail elsewhere
in this book. Polymorphism can also result from
the exchange of partial sequences between the
members of a multi-gene family, as described for
the human y-globin gene (Fig. 4.6a).
As DNA sequencing requires considerable
effort, relatively few comparative investigations
have been undertaken on sequences originating
from the two homologous chromosomes of cells
of an individual or from different individuals of a
species or population. The most extensive data on
the polymorphism of DNA sequences come from
the genes for human haemoglobin [308, 443].
Two different nucleotides were found in each of
eight positions at the B-globin locus, and in one
position there were in fact three differences; the
combination of these characters defines at least
seven haplotypes that are distributed in human
populations [361]. Two types (R and T) of the
human o-globin gene found in all populations differ from each other by 16 substitutions and two
deletions, which are mainly in the non-coding 5'
non-translated region; the sequences differ by
almost 1 % in this region of 1. 7 kb [257]. 1\vo