18
2 Nucleic Acids and Nuclear Proteins
phila proteins that bind specifically to the
sequence (GT/CA)n have now been found [453].
The proportion of highly repetitive DNA, determined with a radiolabelled DNA probe for example in rats, is higher than the estimate from .reassociation experiments [480]. In contrast to hIghly
repeated sequences, middle repetitive DNA is
always widely distributed throughout the
genome. It is now clear that many of these
sequences are transposable, i.e. they can be introduced into other parts of the genome.
2.3.1 Satellite DNA
The name satellite DNA derives from the specific
bands that form during density-gradient centrifugation due to the deviations in. density ca,!~ed
by the typical composition of hIghly repetItIve
DNA. The term is today applied to all tandemly
clustered, highly repeated sequences, ev~n ,:hen
the base composition, and thus the densIty, IS no
different from that of the rest of the DNA (cryptic
satellites) [31]. With the exception of the large
copy number and the tandem ar~angement s~tellite DNAs in animals show consIderable vanety.
The length of the repeat, for example, can vary
between 2 and more than 1000 bp. Satellite DNA
is mainly to be found in the highly condensed,
strongly staining areas of the nucleus (heterochromatin) bound to specific proteins [13]. The
amount may differ even between closely related
species. The short satellites of Drosophila melanogaster, for example, make up 21 % of the
genome, those of D. simulans constitute 5 %, but
in D. erecta the value is only 0,5 % [269]. In the
flour beetle, Tenebrio molitor, no less than 49 %
of the genome is made up of satellite DNA, corresponding to the very large amount of hete~ochromatin seen in this species [343]. Satelhte
DNA is completely absent from the somatic cells
of several species, which leads to the question of
whether it has any somatic function at all. It is
interesting in this connection that such DNA can
be highly methylated in somatic cells but unmethylated in the germline; this is very different to
the genes that are normally more highly methylated in the germline and are inactive [179].
The relationship between base composition
and density. can be exploited for DNA fractionation. Depending upon the molar fraction of guanosine and cytidine (G+C), DNA in a caesium
chloride solution at 25°C has a density of
p = 1.660 + 0.098 (G+C).
In mammals and birds, density-gradient centrifugation results in four bands; two have densities
less than 1.703, and two are GC-rich, "heavy"
bands with densities of more than 1.704. Lower
vertebrates usually show a complex pattern in the
lower density region; only a very few species also
have "heavy" DNA [345]. The simplest satellites
are found, for example, in Cancer pagurus and
other decapod Crustacea, where 10-30 % of the
total genome consists of poly(AT). Although
satellite DNA is mostly absent from lower eukaryotes, very complicated satellites with repeats of
177 and 196 bp are to be found in the flagellates
Trypanosoma brucei and T. cruzi, respecti~ely
[400]. In Drosophila melanogaster, four satelhtes
may be isolated from density gradients; one satellite is made up of repeats of more than 300 bp,
and the other three consist of 10 or 11 different
repeats of 5-9 bp [269]. Unusually complex
repeats of 1460 bp are seen in the satellite DNA
of the sea snail, Rapana thomasiana [282].
Satellite DNA shows rapid evolution, and even
closely related species have different repe~ted
sequences [111]. There are, however, excep~lOns
to this rule but these have not yet been explamed.
For example, the same 6-bp repeat is found in the
kangaroo rat Dipodomys and the guinea-pig;
seven of the ten short satellites of Drosophila
melanogaster are found in the twin species (sibling species) D. simulans, whereas all ten, in fact,
occur in the more distantly related D. erecta
[269]. The mechanisms of "horizontal evolution:'
lead, on the one hand, to more sequence UnIformity within individual tandem clusters, but, on
the other hand, to the rapid spread of single-point
mutations. Thus, satellite DNA shows a higher
degree of variability between individuals of a species than was suspected earlier from measurements of denaturation temperature or density: in
both Drosophila and mammals, over 10 %
sequence difference is found between in~ividual
repeats. In addition to sequence alteratIons by
substitution, addition/deletion and rearrangement, changes in the amount through amplification and the elimination of whole regions appear
to be frequent events in the evolution of satellite
DNA [31].
The multiple satellites of a species often
develop one from the other. In man and other primates, one finds a family of repeats of about
171 bp, called a-satellite or alphoid-DNA, in the
centromere region of all chromosomes. The alphoid sequences of humans deviate, on average, by
16 % from a common consensus sequence and
may be assigned to five groups of closely related
2 Nucleic Acids and Nuclear Proteins
phila proteins that bind specifically to the
sequence (GT/CA)n have now been found [453].
The proportion of highly repetitive DNA, determined with a radiolabelled DNA probe for example in rats, is higher than the estimate from .reassociation experiments [480]. In contrast to hIghly
repeated sequences, middle repetitive DNA is
always widely distributed throughout the
genome. It is now clear that many of these
sequences are transposable, i.e. they can be introduced into other parts of the genome.
2.3.1 Satellite DNA
The name satellite DNA derives from the specific
bands that form during density-gradient centrifugation due to the deviations in. density ca,!~ed
by the typical composition of hIghly repetItIve
DNA. The term is today applied to all tandemly
clustered, highly repeated sequences, ev~n ,:hen
the base composition, and thus the densIty, IS no
different from that of the rest of the DNA (cryptic
satellites) [31]. With the exception of the large
copy number and the tandem ar~angement s~tellite DNAs in animals show consIderable vanety.
The length of the repeat, for example, can vary
between 2 and more than 1000 bp. Satellite DNA
is mainly to be found in the highly condensed,
strongly staining areas of the nucleus (heterochromatin) bound to specific proteins [13]. The
amount may differ even between closely related
species. The short satellites of Drosophila melanogaster, for example, make up 21 % of the
genome, those of D. simulans constitute 5 %, but
in D. erecta the value is only 0,5 % [269]. In the
flour beetle, Tenebrio molitor, no less than 49 %
of the genome is made up of satellite DNA, corresponding to the very large amount of hete~ochromatin seen in this species [343]. Satelhte
DNA is completely absent from the somatic cells
of several species, which leads to the question of
whether it has any somatic function at all. It is
interesting in this connection that such DNA can
be highly methylated in somatic cells but unmethylated in the germline; this is very different to
the genes that are normally more highly methylated in the germline and are inactive [179].
The relationship between base composition
and density. can be exploited for DNA fractionation. Depending upon the molar fraction of guanosine and cytidine (G+C), DNA in a caesium
chloride solution at 25°C has a density of
p = 1.660 + 0.098 (G+C).
In mammals and birds, density-gradient centrifugation results in four bands; two have densities
less than 1.703, and two are GC-rich, "heavy"
bands with densities of more than 1.704. Lower
vertebrates usually show a complex pattern in the
lower density region; only a very few species also
have "heavy" DNA [345]. The simplest satellites
are found, for example, in Cancer pagurus and
other decapod Crustacea, where 10-30 % of the
total genome consists of poly(AT). Although
satellite DNA is mostly absent from lower eukaryotes, very complicated satellites with repeats of
177 and 196 bp are to be found in the flagellates
Trypanosoma brucei and T. cruzi, respecti~ely
[400]. In Drosophila melanogaster, four satelhtes
may be isolated from density gradients; one satellite is made up of repeats of more than 300 bp,
and the other three consist of 10 or 11 different
repeats of 5-9 bp [269]. Unusually complex
repeats of 1460 bp are seen in the satellite DNA
of the sea snail, Rapana thomasiana [282].
Satellite DNA shows rapid evolution, and even
closely related species have different repe~ted
sequences [111]. There are, however, excep~lOns
to this rule but these have not yet been explamed.
For example, the same 6-bp repeat is found in the
kangaroo rat Dipodomys and the guinea-pig;
seven of the ten short satellites of Drosophila
melanogaster are found in the twin species (sibling species) D. simulans, whereas all ten, in fact,
occur in the more distantly related D. erecta
[269]. The mechanisms of "horizontal evolution:'
lead, on the one hand, to more sequence UnIformity within individual tandem clusters, but, on
the other hand, to the rapid spread of single-point
mutations. Thus, satellite DNA shows a higher
degree of variability between individuals of a species than was suspected earlier from measurements of denaturation temperature or density: in
both Drosophila and mammals, over 10 %
sequence difference is found between in~ividual
repeats. In addition to sequence alteratIons by
substitution, addition/deletion and rearrangement, changes in the amount through amplification and the elimination of whole regions appear
to be frequent events in the evolution of satellite
DNA [31].
The multiple satellites of a species often
develop one from the other. In man and other primates, one finds a family of repeats of about
171 bp, called a-satellite or alphoid-DNA, in the
centromere region of all chromosomes. The alphoid sequences of humans deviate, on average, by
16 % from a common consensus sequence and
may be assigned to five groups of closely related