1 0 . D N A A N D R N A S Y N T H E S I S
329
in cell population. In later experiments, Harris (1959b) used cultures
of rapidly-dividing fibroblasts freshly prepared from rat connective
tissue. The cells were incubated for 20 min in a mixture of tritiated
adenosine and tritiated thymidine, followed by further incubations
(up to 12 h) in medium containing excess unlabelled precursors.
This procedure was adopted in order to attempt to select cells not
in D N A synthesis; autoradiographs were prepared and cells with
heavy nuclear labelling were discarded from the experiments on the
assumption that these were synthesizing D N A , i.e. labelled with both
precursors, at the time of incubation. For the remaining cells, the
autoradiographs showed that the grain counts over the cytoplasm
increased at a time before those over the nucleus had dropped, and
that more grains appeared subsequently over the cytoplasm than had
disappeared from the nucleus. It was concluded that the cytoplasmic
R N A was in the main not derived from the nuclear R N A .
In his paper, Harris apparently neglects the fact that, because of the
short range (^1/*) of the tritium ft particles and the smaller average
thickness of the cytoplasm, one would expect a greater efficiency of
detection of cytoplasmic label than of nuclear label.
An interesting point brought out by Harris (1959a) was that in
experiments utilizing short incubation times in cytidine, the ratio of
nucleolar to chromatin uptake was much greater than with adenosine.
In later experiments with HeLa cells, Harris and Watts (1962) provided
additional evidence that in these cells there is a breakdown of nuclear
RNA. In principle, although there may be disagreement on the
magnitude of the effect, a breakdown of a fraction o f the nuclear R N A
here is not inconsistent with the suggestion of Feinendegen et al.
(1961b, c) that a part of the nuclear R N A turns over and is subsequently
re-incorporated. The experiments of Thomson et al. (1956, 1957, 1958)
and Graham and Siminovitch (1957) have already been cited in this
respect.
In spite of some R N A breakdown, which may tend to obscure a
nucleus-to-cytoplasm transfer, the recent work of Reich, Franklin,
Shatkin and Tatum (1962) and of Shatkin (1962) does indeed indicate
that in both HeLa and L-strain cells, the synthesis of cellular R N A
is dependent on the presence of D N A and takes place in the nucleus.
In these experiments, an almost complete inhibition of R N A synthesis
was produced by the addition to the cultures of actinomycin D , an
antibiotic that binds strongly to primer D N A but much less readily to
R N A (Rauen, Kersten, and Kersten, 1960; Kawamata and Imanishi,
1961). Similar results showing the dependence of R N A synthesis on
the presence of D N A have been obtained for fresh cultures of embryomouse fibroblasts (Seed, unpublished experiments).
329
in cell population. In later experiments, Harris (1959b) used cultures
of rapidly-dividing fibroblasts freshly prepared from rat connective
tissue. The cells were incubated for 20 min in a mixture of tritiated
adenosine and tritiated thymidine, followed by further incubations
(up to 12 h) in medium containing excess unlabelled precursors.
This procedure was adopted in order to attempt to select cells not
in D N A synthesis; autoradiographs were prepared and cells with
heavy nuclear labelling were discarded from the experiments on the
assumption that these were synthesizing D N A , i.e. labelled with both
precursors, at the time of incubation. For the remaining cells, the
autoradiographs showed that the grain counts over the cytoplasm
increased at a time before those over the nucleus had dropped, and
that more grains appeared subsequently over the cytoplasm than had
disappeared from the nucleus. It was concluded that the cytoplasmic
R N A was in the main not derived from the nuclear R N A .
In his paper, Harris apparently neglects the fact that, because of the
short range (^1/*) of the tritium ft particles and the smaller average
thickness of the cytoplasm, one would expect a greater efficiency of
detection of cytoplasmic label than of nuclear label.
An interesting point brought out by Harris (1959a) was that in
experiments utilizing short incubation times in cytidine, the ratio of
nucleolar to chromatin uptake was much greater than with adenosine.
In later experiments with HeLa cells, Harris and Watts (1962) provided
additional evidence that in these cells there is a breakdown of nuclear
RNA. In principle, although there may be disagreement on the
magnitude of the effect, a breakdown of a fraction o f the nuclear R N A
here is not inconsistent with the suggestion of Feinendegen et al.
(1961b, c) that a part of the nuclear R N A turns over and is subsequently
re-incorporated. The experiments of Thomson et al. (1956, 1957, 1958)
and Graham and Siminovitch (1957) have already been cited in this
respect.
In spite of some R N A breakdown, which may tend to obscure a
nucleus-to-cytoplasm transfer, the recent work of Reich, Franklin,
Shatkin and Tatum (1962) and of Shatkin (1962) does indeed indicate
that in both HeLa and L-strain cells, the synthesis of cellular R N A
is dependent on the presence of D N A and takes place in the nucleus.
In these experiments, an almost complete inhibition of R N A synthesis
was produced by the addition to the cultures of actinomycin D , an
antibiotic that binds strongly to primer D N A but much less readily to
R N A (Rauen, Kersten, and Kersten, 1960; Kawamata and Imanishi,
1961). Similar results showing the dependence of R N A synthesis on
the presence of D N A have been obtained for fresh cultures of embryomouse fibroblasts (Seed, unpublished experiments).