4.5 Methods and Problems in the Molecular Approach to Evolutionary Relationships
147
and taken as evidence for selection at the relevant
locus. However, this conclusion is not at all certain. Firstly, identical electrophoretic mobility of
proteins does not necessarily mean structural
identity. Secondly, alterations in the frequency of
neutral alleles occur so slowly that similarities
between populations are maintained for a long
time, e.g. up to 2 million years in Drosophila,
after their separation, even in the absence of
selection [240]. Finally, all gene exchange
between the compared populations must be
excluded as, independent of the population size,
the migration of just a few individuals per generation prevents the occurrence of large differences.
It could be shown for D. pseudoobscura that
intensive gene exchange occurred even between
populations 15 km apart. Thus, the experimentally disturbed frequency pattern of esterase 5 in
one oasis population was restored already after
1 year; marked flies travelled up to 10 km in
1 day [196].
Although conclusions cannot be drawn with
any certainty from the coincidence of electrophoretic patterns of isolated populations, different
allele frequencies in closely related, sympatric,
but sexually isolated, populations are a convincing
argument against selection as the cause of polymorphism at any given locus. The nicest example
of this is found in the salmon Oncorhynchus garbuscha. This shows such a rigid biennial reproductive rhythm that two sexually isolated populations exist side-by-side in many Alaskan rivers;
these are known as "even-year" and "odd-year".
These genetically separated populations living in
the same habitat show large differences in allele
frequencies of the malate dehydrogenases Mdh-A
and Mdh-B and in a-glycerophosphate dehydrogenase [12]. Completely different heterozygosity
values (H = 0.04-0.27) and variable frequency
patterns were found at six enzyme loci in 12 sympatric species of Drosophila [355]. Furthermore, it
remains a mystery why, in human populations, the
frequency of rare alleles at 21 erythrocyte protein
loci varies between the English (0.07 %) and
native inhabitants of Australia (1.1 %) [387].
4.5 Methods and Problems
in the Molecular Approach
to Evolutionary Relationships
Homologous DNA regions in different organisms
are always found to contain numerous substitutions and rearrangements which can be used in
the analysis of the genealogical relationships of
individuals and species. Man and the chimpanzee, which are considered to be closely related
species, differ in about 2 % of their DNA sequences, i.e. in approximately 60 million nucleotides,
although, as it happens, the majority of the
sequence differences have no phenotypic effects.
Even individual humans differ in up to 5 million
nucleotides [38]. Each human gamete has, on
average, about 20 nucleotide substitutions [288].
The comparison of molecular characters need
not always make use of the costly techniques of
DNA or protein sequencing. There is a whole
series of molecular properties which are correlated to the DNA or protein sequences and are
easily compared in a large number of individuals
or species; they often involve large parts of the
genome. The following contains a description of
these various methods for comparing molecular
characters and their use in constructing phylogenetic trees [175].
The reconstruction of phylogeny is closely
associated with the classification of the organisms; this latter is the task of a special biological
discipline known as systematics or taxonomy
[198, 347]. Basically, organisms may be grouped
according to chosen characters and criteria, e.g.
their usefulness or destructiveness towards man,
their habitats or any other character appropriate
to their identification. However, as a general
frame of reference, a preferred system stems from
the organisms themselves, i.e. it can be considered a natural system. The three most important
schools of systematics strive for such a system in
quite different ways.
1. "Phenetic systematics" organizes organisms
according to the degree of their similarity. A
particular advantage of this method is the possibility for quantification. The results of a phenetic analysis can be presented in the form of a
branching scheme (a cladogram), which may
be considered to be a picture of the most likely
phylogenetic relationships. The founders of
such "numerical taxonomy" were Sneath and
Sokal [111, 385].
2. The development of "phylogenetic systematics", mainly by Willi Hennig, is based entirely
on the phylogenetic relationships of the species as depicted in a cladogram with dichotomous branching; the principle of this method is
the identification of the sibling species or
group of a particular species that share the
most recent common ancestor [347].
147
and taken as evidence for selection at the relevant
locus. However, this conclusion is not at all certain. Firstly, identical electrophoretic mobility of
proteins does not necessarily mean structural
identity. Secondly, alterations in the frequency of
neutral alleles occur so slowly that similarities
between populations are maintained for a long
time, e.g. up to 2 million years in Drosophila,
after their separation, even in the absence of
selection [240]. Finally, all gene exchange
between the compared populations must be
excluded as, independent of the population size,
the migration of just a few individuals per generation prevents the occurrence of large differences.
It could be shown for D. pseudoobscura that
intensive gene exchange occurred even between
populations 15 km apart. Thus, the experimentally disturbed frequency pattern of esterase 5 in
one oasis population was restored already after
1 year; marked flies travelled up to 10 km in
1 day [196].
Although conclusions cannot be drawn with
any certainty from the coincidence of electrophoretic patterns of isolated populations, different
allele frequencies in closely related, sympatric,
but sexually isolated, populations are a convincing
argument against selection as the cause of polymorphism at any given locus. The nicest example
of this is found in the salmon Oncorhynchus garbuscha. This shows such a rigid biennial reproductive rhythm that two sexually isolated populations exist side-by-side in many Alaskan rivers;
these are known as "even-year" and "odd-year".
These genetically separated populations living in
the same habitat show large differences in allele
frequencies of the malate dehydrogenases Mdh-A
and Mdh-B and in a-glycerophosphate dehydrogenase [12]. Completely different heterozygosity
values (H = 0.04-0.27) and variable frequency
patterns were found at six enzyme loci in 12 sympatric species of Drosophila [355]. Furthermore, it
remains a mystery why, in human populations, the
frequency of rare alleles at 21 erythrocyte protein
loci varies between the English (0.07 %) and
native inhabitants of Australia (1.1 %) [387].
4.5 Methods and Problems
in the Molecular Approach
to Evolutionary Relationships
Homologous DNA regions in different organisms
are always found to contain numerous substitutions and rearrangements which can be used in
the analysis of the genealogical relationships of
individuals and species. Man and the chimpanzee, which are considered to be closely related
species, differ in about 2 % of their DNA sequences, i.e. in approximately 60 million nucleotides,
although, as it happens, the majority of the
sequence differences have no phenotypic effects.
Even individual humans differ in up to 5 million
nucleotides [38]. Each human gamete has, on
average, about 20 nucleotide substitutions [288].
The comparison of molecular characters need
not always make use of the costly techniques of
DNA or protein sequencing. There is a whole
series of molecular properties which are correlated to the DNA or protein sequences and are
easily compared in a large number of individuals
or species; they often involve large parts of the
genome. The following contains a description of
these various methods for comparing molecular
characters and their use in constructing phylogenetic trees [175].
The reconstruction of phylogeny is closely
associated with the classification of the organisms; this latter is the task of a special biological
discipline known as systematics or taxonomy
[198, 347]. Basically, organisms may be grouped
according to chosen characters and criteria, e.g.
their usefulness or destructiveness towards man,
their habitats or any other character appropriate
to their identification. However, as a general
frame of reference, a preferred system stems from
the organisms themselves, i.e. it can be considered a natural system. The three most important
schools of systematics strive for such a system in
quite different ways.
1. "Phenetic systematics" organizes organisms
according to the degree of their similarity. A
particular advantage of this method is the possibility for quantification. The results of a phenetic analysis can be presented in the form of a
branching scheme (a cladogram), which may
be considered to be a picture of the most likely
phylogenetic relationships. The founders of
such "numerical taxonomy" were Sneath and
Sokal [111, 385].
2. The development of "phylogenetic systematics", mainly by Willi Hennig, is based entirely
on the phylogenetic relationships of the species as depicted in a cladogram with dichotomous branching; the principle of this method is
the identification of the sibling species or
group of a particular species that share the
most recent common ancestor [347].