12. G E N E T I C C Y T O L O G Y
445
answered only when we understand more about the molecular architecture of the chromosomes. It is true that various models of nucleoprotein have been suggested in regard to the relationship between
DNA and protein (Mirsky and Ris, 1947; Mazia, 1954; Taylor,
1959; Zubay and Doty, 1960; cf. Busch, 1962, for pictorial models),
but not one has yet been accepted generally by the scientific public.
There is little doubt that a chromosome is not structurally uniform
throughout its length. The centromere is definitely a unique organelle
of each chromosome. Unfortunately, we do not know of what a
centromere is composed. Actually, the centromere region, besides
the centromere per se, is different from the rest of the chromosome
because of its relatively achromatic property. The nucleolar organizer
is another example of differentiated chromosome region. There are
probably more regions along the chromosome that are molecularly
different from one another but they cannot be demonstrated with
ordinary cytological techniques. Nor could physicochemical analysis
reveal such minute characteristics, at least at the present time, because
chemical study requires extraction of D N A from a large quantity of
nuclei. The D N A of the crab, Cancer borealis, for instance, contains
about 30°/o deoxy-A-T (adenine-thymine) polymers (Sueoka, 1961;
Sueoka and Cheng, 1962). How are these A - T polymers distributed?
Can there be G-C (guanine-cytosine) polymers? The work of Hsu and
Somers (1961) and of Somers and Hsu (1962), summarized below, is
somewhat related to these questions.
In the Chinese hamster, certain regions of the chromosomes, under
the influence of BUDR, may show constriction-like appearance. These
localities, together with the telomeric regions, were most susceptible
to break after B U D R treatment. It has been shown by many workers
that B U D R is actually incorporated into the D N A of the test systems,
preferably replacing thymidine (Zamenhof and Griboff, 1954; Dunn
and Smith, 1954; Djordjevic and Szybalski, 1960; Kit and Hsu, 1961).
However, the postulation made by Hsu and Somers that "the regions
where high frequencies of breaks were found correspond to what is
conventionally known as heterochromatin" requires clarification.
These regions do not correspond to conventional heterochromatin, which
should be the Heitz type. They resemble, in a way, the heterochromatin
described by Darlington and La Cour, because they show negative
heteropyknosis instead of positive heteropyknosis. Hsu and Somers also
suggested, based on the consideration that more thymine in an area of
a chromosome would mean more B U D R incorporation, hence more
damage, that these loci may be rich in A - T pairs, or that they may even
be short sections of A - T polymers.
An armchair theory proposed by Herskowitz (1961) stating that some
445
answered only when we understand more about the molecular architecture of the chromosomes. It is true that various models of nucleoprotein have been suggested in regard to the relationship between
DNA and protein (Mirsky and Ris, 1947; Mazia, 1954; Taylor,
1959; Zubay and Doty, 1960; cf. Busch, 1962, for pictorial models),
but not one has yet been accepted generally by the scientific public.
There is little doubt that a chromosome is not structurally uniform
throughout its length. The centromere is definitely a unique organelle
of each chromosome. Unfortunately, we do not know of what a
centromere is composed. Actually, the centromere region, besides
the centromere per se, is different from the rest of the chromosome
because of its relatively achromatic property. The nucleolar organizer
is another example of differentiated chromosome region. There are
probably more regions along the chromosome that are molecularly
different from one another but they cannot be demonstrated with
ordinary cytological techniques. Nor could physicochemical analysis
reveal such minute characteristics, at least at the present time, because
chemical study requires extraction of D N A from a large quantity of
nuclei. The D N A of the crab, Cancer borealis, for instance, contains
about 30°/o deoxy-A-T (adenine-thymine) polymers (Sueoka, 1961;
Sueoka and Cheng, 1962). How are these A - T polymers distributed?
Can there be G-C (guanine-cytosine) polymers? The work of Hsu and
Somers (1961) and of Somers and Hsu (1962), summarized below, is
somewhat related to these questions.
In the Chinese hamster, certain regions of the chromosomes, under
the influence of BUDR, may show constriction-like appearance. These
localities, together with the telomeric regions, were most susceptible
to break after B U D R treatment. It has been shown by many workers
that B U D R is actually incorporated into the D N A of the test systems,
preferably replacing thymidine (Zamenhof and Griboff, 1954; Dunn
and Smith, 1954; Djordjevic and Szybalski, 1960; Kit and Hsu, 1961).
However, the postulation made by Hsu and Somers that "the regions
where high frequencies of breaks were found correspond to what is
conventionally known as heterochromatin" requires clarification.
These regions do not correspond to conventional heterochromatin, which
should be the Heitz type. They resemble, in a way, the heterochromatin
described by Darlington and La Cour, because they show negative
heteropyknosis instead of positive heteropyknosis. Hsu and Somers also
suggested, based on the consideration that more thymine in an area of
a chromosome would mean more B U D R incorporation, hence more
damage, that these loci may be rich in A - T pairs, or that they may even
be short sections of A - T polymers.
An armchair theory proposed by Herskowitz (1961) stating that some