/. Embryol. exp. Morph. Vol. 28, 2, pp. 367-384, 1972
Printed in Great Britain
367
The appearance and quantitation of cytoplasmic
ribonucleic acid in the early chick embryo
By C. C. WYLIE 1
From the Department of Anatomy and Embryology,
University College London
SUMMARY
This paper seeks to extend our knowledge about RNA synthesis in early embryogenesis to
the domestic fowl, Gallus domesticus. Using this species for research, apart from increasing
our knowledge of higher vertebrate embryology, has certain advantages such as rapid uptake
of isotopic precursors and ease of microdissection in culture.
The following results are presented:
(1) The cell number in the whole chick embryos is shown to be increasing logarithmically
between the time of laying and the early neurula stage; with a doubling time of 7-4 h.
(2) The onset of ribosomal RNA synthesis has been shown to be during mid-cleavage of
the chick embryo, while development is taking place in the oviduct and uterus of the mother.
(3) In a cumulative labelling experiment, embryos were labelled at the unincubated-egg
stage, allowed to develop to various morphological stages up to neurulation, and their cytoplasmic RNA prepared and analysed by gel electrophoresis.
(4) The specific activity of the precursor pool for RNA synthesis was measured at several
stages, using the same labelling conditions, and the results were used to quantitate the RNA
synthesis from the incorporated radioactivity.
(5) Using these techniques, it was found that newly synthesized cytoplasmic RNA accumulates steadily in the whole chick embryo, reaching a level of 104 /*g by the early neurula stage.
On a per cell basis, however, the amount of newly synthesized cytoplasmic RNA seems to
decrease slightly.
These findings are discussed in the light of present knowledge about embryos of other
vertebrates and certain invertebrates.
INTRODUCTION
This paper seeks to extend recent work on ribonucleic acid (RNA) metabolism
in early embryos (see Davidson, 1969, for review) to that of the domestic fowl
Gallus domesticus.
There seems to be no obvious phylogenetic pattern to the easily measurable
parameters concerning RNA metabolism in early embryos, e.g. time of onset of
ribosomal RNA (rRNA) synthesis. This first becomes visible in profiles of
embryonic, radioactively labelled RNA at a variety of times in different species
(see Davidson, 1969, for table). However, there must be an evolutionary basis
1
Author's address: Department of Anatomy and Embryology, University College London,
Gower Street, London WC1E 6BT, U.K.
368
C. C. WYLIE
for these molecular events. Things likely to determine the start of rRNA synthesis, e.g. amount of ribosome stockpiling during oogenesis, or quantity of
available nutrient, are presumably themselves the results of natural selection.
A preliminary study of RNA synthesis in early avian embryos (Lerner, Bell &
Darnell, 1963) has demonstrated that stable RNA is already being synthesized at
the onset of mesoderm formation. This paper demonstrates that the start of
rRNA synthesis is sometime during mid-cleavage of the embryo. RNA synthesis
during the first 24 h of incubation is studied by labelling with radioactive uridine.
Newly synthesized cytoplasmic RNA is then prepared and analysed quantitatively and qualitatively at morphological stages up to and including the beginning of neurulation. The absolute amounts of RNA synthesis are calculated by
measuring the specific activity of the nucleotide precursor pool. The number of
cells in the chick embryo, at each of the stages used, has been measured in order
to quantitate the RNA synthesis per cell.
MATERIALS AND METHODS
(1) Labelling of embryos. Newly laid, fertilized, White Leghorn eggs were
inoculated 'in ovo' with high specific activity (29 Ci//tM) [5-3H]uridine (Radiochemical Centre, Amersham) in isotonic buffered saline (Pannett & Compton,
1924). The radioactive solution, 10-20 /^Ci in 0-1 ml, was injected into the yolk
beneath the blastoderm. The embryos were allowed to develop for the required
time in a humidity-controlled incubator at 37-5 °C. They were then explanted
into ice-cold saline, staged carefully under a dissecting microscope, pooled, and
stored at - 7 0 °C.
Embryos at pre-laying stages were removed from the oviducts and uteri of
first-year laying White Leghorns. Some of these were at such an early stage of
development that neither shell nor shell membrane had been applied to the egg
surface. In these cases the yolk, together with adherent albumen, was incubated
in sterile saline at 37-5 °C and the isotope injected beneath the cleaving embryo
as above.
(2) RNA Preparation. The frozen batches of pooled embryos (usually from
3-12 embryos/batch) were thawed into a Tris buffer (TEP2 = 0-01M Tris,
10"3M EDTA, 20 /tg/ml polyvinyl sulphate, buffered to pH 7-2 with HC1) containing 1 % (w/v) of the non-ionic detergent Brij 58. The lysate thus obtained
was spun at 10000 rev/min for 10 min on a Misco centrifuge to remove nuclei,
mitochondria, cytoplasmic membranous debris, and as much yolk as possible.
To the supernatant, sodium dodecyl sulphate (SDS) was added to a final concentration of 1 % and the preparation warmed to 37 °C for 3 min. Cold phenol
extractions were carried out until a clear interphase was obtained, followed by
further extractions with 1 % iso-amyl alcohol in chloroform. The aqueous phase
was purified from low-molecular-weight impurities by gel filtration through
a 5 ml G 50 Sephadex column. The low-molecular-weight fraction was kept for
RNA metabolism in the early chick embryo
369
further analysis and the RNA fraction was analysed by gel electrophoresis on
either 1-8 % or 4 % agarose gels (Evans, 1969).
(3) Chick-embryo cell counts. The method used was that of Solomon (1957).
Batches of 2-3 embryos were explanted at various times after the start of
incubation, and homogenized in 0-5 ml of ice-cold 1 % citric acid; 0-5 ml of
1 % aqueous methyl green was then added and the staining mixture left on ice
for 15 min. The cell nuclei were centrifuged off in a bench centrifuge and the
pellet resuspended in a known volume of tap-water. The nuclei were counted in
a Neubauer haemocytometer.
(4) The specific activity of the precursor pool. This was measured in batches of
20-50 embryos labelled at stage 1 (Hamburger & Hamilton, 1951) with 5/tCi
[3H]uridine each, and incubated to various stages up to stage 6. The frozen
batches of embryos were thawed into 0-32 M sucrose, 0-003 M magnesium chloride,
buffered to pH 7-2 with bicarbonate, and containing 1 % (w/v) Brij 58. The
nuclei were spun from the lysate and the supernatant was made to 5 % trichloracetic acid. 5'-Nucleoside monophosphates were prepared by the method of
Brown & Littna (1966) by hydrolysing the di- and tri-phosphates with N-HC1,
while adsorbed on to active Norit A (Hopkin & William).
Two methods, using column chromatography, were used to separate the four
5'-mononucleotides.
(i) Anion exchange using QAE Sephadex, was found to give good separation of
5' UMP. The following parameters were found to give the best results:
Sephadex
Column
Starting buffer
Salt gradient
Pump speed
QAE A 25 anion exchange
20 x 0-9 cm
0-025 M CO 3 /HCO 3 buffer, pH 9
300 ml 0-0-4 M sodium chloride in starting buffer,
convex gradient
36 ml/h
(ii) Separation of 5'-CMP from the other mononucleotides was achieved by
the method of Blattner & Erickson (1967), using a Dowex formate column and
elution at 0-5 ml/min with 0-1 M ammonium formate pH 3-2. The column length
was 45 cm.
RESULTS
Cell counts in early chick embryos
The results of this experiment are presented in Fig. 2. The number of cells in
the embryos is seen to be increasing logarithmically over the period studied,
with a mean doubling time of 7-5 h.
370
C. C. WYLIE
UMP
o.D. 260
x 10 2
O.D. 260
20 - cpm
x 10
cpm
xlO3
CMP
16
6
12
4
:
UMP
5
4
3
2
2
0
i
50
160 140 120 100 80 60 40 20
ml of eluate
i
i
40 30 20
ml of eluate
i
II
10
Fig. 1. (a) Separation on QAE Sephadex of a preparation of 5'-mononucleotides
from 28 embryos at stage 5, incubated for 23 h with [3-H]uridine. The dotted line
represents optical density at 260 nm (O.D. 260), the unbroken line represents incorporated radioactivity in counts/minute (cpm). (b) Separation on Dowex of a preparation of 5'-mononucleotides from 30 embryos at stage 4, incubated with [3H]uridine for 20 h. The O.D. 260 scale refers to the peak of 5'-uridine monophosphate
(5'-UMP) in (a) and 5'-cytidine monophosphate (5'-CMP) in (b).
Table 1
Morphological
stage
1
2
3
4
5
5
6
Specific activity of
No
of
embryos
47
46
29
30
28
12
22
(h)
5
9-7
15
22
23
25
25
5'-UMP
5'-CMP
3
6-78 x 10
3-92 xlO3
5-55X103
7-12xlO3
l-67xlO 3
—
3-01 x 103
1-65 xlO 3
0-85 x 103
0-79xl0 3
0-67 x 103
0-20xl0 3
0-303 x 103
0-32 xlO 3
The specific activity of the precursor pool
Using the methods described, the specific activities of 5'-UMP and 5'-CMP
were measured at various times over a period of 24 h after a single injection
of [3H]uridine at stage 1 (Hamburger & Hamilton, 1951). Fig. 1 shows two sample
separations of 5'-mononucleotide mixtures using both QAE Sephadex and
Dowex. As well as the four 5' monophosphates normally found (arrowed in
Fig. 1) there are several unlabelled peaks whose nature is unknown. Table 1
gives the specific activities calculated from separations of this type. Fig. 1
shows that 5'-UMP and 5'-CMP are labelled whereas 5'-AMP and 5'-GMP
are not. This suggests the conversion of the injected uridine into cytidine
RNA metabolism in the early chick embryo
371
Table 2
Stage
Incubation
(h)
Ratio
5'-CMP:5'-UMP
Specific
activity of
5'-mononucleotide pool
= (UMP + CMP)/4
1
2
3
4
5
6
5
9-7
15
22
23
25
0-143
0165
0192
0171
0085
0109
2108-8
1092-5
1580-3
1944-2
467-1
8351
logio number of
20 - nuclei x 103
1-8 per embryo
7
1-6
6
8r
1-4
Specific activity
(cpm///g) x 1O3
5
•
1-2
4
10
3
0-8
2
0-6
1
5
1
1
1
10
15
20
Hours of incubation
Fig. 2
1
25
5
10
15
20
25
Hours of incubation
Fig. 3
Fig. 2. The number of cells per embryo. Each point represents the number of isolated, stained nuclei per embryo in batches of two or three embryos.
Fig. 3. The change in specific activity (cpm/^g) of the 5'-monophosphate pool,
following a single dose of [3H]uridine at the unincubated egg stage.
nucleotides. This is further suggested by the fact that there is always less 5'-CMP
found in these preparations than 5'-UMP, and its specific activity is always
higher. There is also a fairly constant ratio of 5'-CMP:5'-UMP in these
preparations (see Table 2).
These results demonstrate therefore the conversion of uridine into cytidine
nucleotides in the early chick embryo.
5'-GMP was always found to be in minute quantities or not demonstrable at
all in these experiments (see Fig. 1 a), suggesting that it may be a rate-limiting
step in the biosynthesis of RNA in these early embryos. In calculating the
specific activity of the overall 5'-nucleotide pool and therefore that of the RNA
made from it, it was assumed that a negligible amount of radioactivity enters the
372
C. C. WYLIE
Table 3
Developmental
stage
At laying
18-20 h
0-18 h
18-20 h
33 h
48 h
72 h
No. of cells
5
Doubling time
0-74 x 10
2-0±10xl0 5
—
1-93 xlO5
406 x 105
907 xlO5
33-5 xlO5
7-4 h
—
80 h
—
10-7 h
—
—
Reference
Emanuelsson, 1965
Richenbacher, 1956
Emanuelsson, 1962
—
Woodard, 1948
• — •
—
•
RNA via adenine or guanine nucleotides, since these were both found to be
unlabelled in these experiments. It is also assumed that cytidine and uridine
nucleotides contribute approximately 25 % each to the RNA synthesized. The
specific activity of the 5' nucleotide pool will therefore roughly equal half of the
average specific activity of CMP and UMP (i.e. i(CMP + UMP)) (see Table 3).
The change in this overall figure is plotted in Fig. 3. The specific activity of the
5'-mononucleotide pool was found to change very little during the first 24 h of
embryogenesis.
The onset of stable RNA synthesis in chick embryos
In this experiment, two batches of embryos were taken from the oviducts and
uteri of adult hens. The first batch were taken from the upper oviducts; no shell
or shell membrane had been applied to the albumen-covered embryo. They were
therefore considered to be 6 h old or less (Patterson, 1910). The second batch
was taken from the uteri and had both shell and shell membrane attached. The
embryos were incubated in ovo with 50 JLLCI each of [5-3H]uridine for 3 h, and
their cytoplasmic RNA prepared. Morphological controls from both batches
were fixed and micrographs of these are shown in Fig. 4. This shows that the
embryos of the first batch are in early cleavage, with about ten and 100 cells
respectively. The second batch have finished cleavage and are at the stage of
epiblast and hypoblast formation. Note the prominent nucleoli in these embryos
(Fig. 4).
Gel-electrophoresis profiles of the cytoplasmic RNA from these two batches
of embryos are presented in Fig. 5. There is very little RNA synthesis in the
early cleavage embryos (Fig. 5 a) and this slight incorporation is spread hetero-
FlGURE 4
Fig. 4. (A, B) Surface views of morphological control embryos from the early
cleavage stages used. The embryos have about 10 and 100 cells respectively. (C)-(F)
Light micrographs of sections through the late cleavage embryos used; they are either
in the process of epiblast/hypoblast formation (C) or have completed the process
(D-F). Note the prominent nucleoli and mitotic figures in (E) and (F).
RNA metabolism in the early chick embryo
B
373
374
c c.
WYLIE
cpm
18S 28 S
Jl
ft H
[
n
•
80
1
AV
— Deleft
cpm
28S
U S
4S
•
1
i
i
50
40
400
60
300
50
i
30
20
Gel slices
500
70
\l\
40
0
_
'
V:
I
i
i
I
I
I
I
10
60
50
40
30
20
10
200
100
Gel slices
Fig. 5. Electrophoresis on 1-8% agarose gels of RNA from early cleavage-stage
embryos (a) and late cleavage embryos (b).
geneously along the gel. There is no increased incorporation into the r R N A
regions of the gel, which were shown up by the use of high optical density, unlabelled cytoplasmic RNA from 15-day chick-embryo liver.
The level of radioactivity in Fig. 5 (a) was demonstrated statistically to be
over the background level by use of the Student's t test. For this purpose the gel
was considered to be divisible into two parts: that between the origin and the
4S region containing RNA, and that beyond the RNA containing only natural
radioactivity of the agarose. A series of bottles containing only scintillation
fluid was first counted for 40 min/bottle. Gel slices were then added and the
bottles recounted for the same time. The groups of results were then compared
as populations using the Student's / test as follows: (1) RNA-containing gel v,
(2) empty bottles, (3) beyond the 4 S region v, (4) empty bottles. If there is RNA
present in these gels, the probability (Pa) that populations (1) and (2) are identical
will be much lower than the probability (Pb) that populations (3) and (4) are
dentical. This was in fact the result obtained: For Fig. 5(a) Pa = 0-0005,
Pb = 0-01.
Fig. 5(b) demonstrates that by late cleavage or delamination a large proportion of the incorporated radioactivity is found in the stable RNA types: about
62 % of the counts per minute (cpm) found in electrophoresis gel.
It can be concluded therefore that synthesis of the stable RNA types starts
sometime during cleavage in the chick, while the embryo is still in the hen, and
at least 7-10 h before mesoderm formation starts.
Synthesis of cytoplasmic RNA during the period of embryogenesis
between laying of the fertilized egg and neurulation
In this experiment embryos were labelled with [5-3H]uridine at the laid, unincubated egg stage; and incubated for various lengths of time before explanting
and preparing their cytoplasmic RNA.
375
RNA metabolism in the early chick embryo
28 S
S
T f 16
12
18 S
14
10 &
8
-i 25
12
I
10
6 2
6
4
^
4
4S
ss
50
40
30
20
10
Gel slices
50
40
1/
30
20
10
Gel slices
18 S
70
60
50 40
30 20
10
Gel slices
Fig. 6. (a, 6)1-8 % agarosegel-electrophoresis profiles of cytoplasmic RNA extracted
from stage 1 (a) and stage 6(6) embryos; explanted 2-5 and 25-5 h respectively from
a single dose of [3H]uridine at the unincubated egg stage; (c) 4 % agarose gel profile
to show complete 5 S and 4 S RNA separation. Embryos explanted at stage 1, 4 h
after isotope administration.
Electrophoresis profiles of two of these preparations are shown in Fig. 6 (a, b).
It can be seen that in one electrophoresis period (about 2-5 h) on 1-8 % gels the
28 S and 18 S rRNA species are well separated from each other but the 5 S and
4 S species are not. Electrophoresis of another sample of the same RNA preparation on 4 % agarose gels, however, allows a complete separation of 5 and 4 S
RNA (Fig. 6 c).
The stable RNA species are being synthesized and transported to the cytoplasm at all the stages studied, up to 26 h after the start of incubation, i.e. the
376
C. C. WYLIE
(
(«)
0-5 - cpm
200 - c p m
.embryo
1X0
"xlO3
160
_ /cell
0-4 _
140
/
/
120
100
f
/
I
0-2
60
20
\
\
\
•f
80
40
i
0-3
_/
\
i
-
nL
•
*
ffai
5
i
i
01
-
f
\
—A
i
i
i
10
15
20
Hours of incubation
25
5
|
I
10
15
20
Hours of incubation
A
0-2 - ng
cell
00 - //g RNA
/embryo
80 -
i
1
60 -
j
01
•\l
40 A
20
J
5
I
1
1
10
15
20
25
Hours of incubation
5
10
15
20
Hours o( incubation
Fig. 7. Radioactivity accumulating in the cytoplasmic RNA (A
A) and low
molecular weight pool (A
A) per embryo (a) and per cell (b). The specific activity of the precursor pool has been used to convert these into absolute quantities in
(c) and (d).
neurula stage. Some kinetics of this process were studied by measuring such
parameters as the radioactivity incorporated into the RNA and into different
components of RNA at different times after labelling.
(i) Quantity of RNA synthesized and transported to the cytoplasm
This is shown as the radioactivity incorporated into the cytoplasmic RNA per
embryo and per cell after different labelling periods (Fig. 7 a, b). On a per embryo
basis the incorporation continues to rise throughout the labelling period,
RNA metabolism in the early chick embryo
311
10 -
i
5
i
i
i
r
10
15 20 25
Hours of incubation
l
5
r
10
15 20 25
Hours of incubation
Fig. 8. Quantities of various cytoplasmic RNA components synthesized (a) per
embryo and (b) per cell; after a single dose of [3H]uridine at the unincubated-egg
stage.
whereas on a per cell basis there is an initial rise up to the 10 h stage, i.e. the
appearance of the primitive streak, followed by a gradual decline over the next
16 h.
These progress curves reflect the quantity of RNA appearing in the embryonic
cell cytoplasm during the first 26 h of incubation. However, the specific activity
of the precursor pool does not remain constant over this period, but falls off
slowly (Fig. 3). Therefore the radioactivity incorporated at later stages represents
proportionally more RNA synthesis than that at early stages. The absolute
amounts of RNA synthesis were calculated from the equation:
jug RNA synthesized during time 0-t =
RNA cpm at time t
average specific activity over time'
0-t in cpm//tg.
The progress curve of the quantity of RNA (in /ig) synthesized per embryo
(Fig. 7 c) shows that this rises steadily to reach a level of about 100/tg by the
onset of neurulation. Expressing the results on a per cell basis (Fig. Id) shows
that the level of newly synthesized cytoplasmic RNA rises steeply from the 5 h
stage to about 0-2 ng RNA/cell at the 10 h stage and falls off slowly for the rest
of the labelling period.
The separation of the embryonic RNA by gel electrophoresis allows quantitation by the same process, to be applied to each component of the RNA. Since
the heterogeneous RNA is obviously not fractionated by this procedure, this is
378
C. C. WYLIE
•o
o
700
600
500
400
300
200
100
10
15
Hours of incubation
20
25
Fig. 9. Synthesis of cytoplasmic RNA, expressed as percentage of the low molecular
weight pool, over the first 26 h of incubation.
all lumped together as one class. Fig 8 shows the appearance of newly synthesized RNA components in the embryonic cytoplasm and their accumulation
during the 26 h labelling period.
Considerable variation is seen in these progress curves, which may reflect the
in vivo situation or the range of biological or experimental scatter. This scatter is
probably located primarily at the level of uptake of uridine into the cellular low
molecular weight pool. When this source of scatter is eliminated by expressing
the RNA radioactivity as a percentage of the low molecular weight pool radioactivity, most of the variation is eliminated (Fig. 9). The straight-line graph
suggests that there is a cumulative increase in the amount of labelled precursor
taken up into RNA from a precursor pool, which is kept fairly constant (Fig. 7).
(ii) The change in relation to each other of the stable components
This is demonstrated by calculating the cpm under each stable RNA peak,
over the level of heterogeneous labelling in the gel profiles; and plotting these as
percentages of the total RNA cpm (Fig. 10). Three points from this graph bear
further discussion.
(a) Initially the percentage of the total cytoplasmic RNA in 18 S is greater
than that in 28 S. This situation then reverses until the ratio of 28 S: 18 S is about
2:1. This suggests, especially in the light of previous work (see Discussion) that
the 18 S species of rRNA enters the cytoplasm before 28 S, during the biosynthesis
of early chick embryo ribosomes.
(b) The proportion of the total cytoplasmic RNA occupied by each stable
RNA species reaches a plateau level by about 10 h of incubation. If the percentage of the total cytoplasmic RNA occupied by all the stable RNA species is
379
RNA metabolism in the early chick embryo
50 _ Percentage of total
RNA radioactivity
Percentage of total
RNA radioactivity
40
30
60
20
40
10
20
5
10
15
20
25
Hours of incubation
5
10
15
20
25
Hours of incubation
Fig. 10
Fig. 11
Fig. 10. The percentage, plotted against time, of the total RNA radioactivity occupied
by the stable RNA components, following a single dose of [3H]uridine at the unincubated egg stage.
Fig. 11. The percentage, plotted against time, of the total RNA radioactivity occupied
by the sum of the stable RNA components, during the cumulative labelling period.
5S:28S
8r-
80 r-
60
40
20
10
15
20
Hours of incubation
Fig. 12
25
10
15
20
25
Hours of incubation
Fig. 13
Fig. 12. Stable RNA radioactivity minus heterogeneous RNA radioactivity in the
embryonic cytoplasm during the cumulative labelling period.
Fig. 13. Ratio of 5S:28S RNA radioactivity in the embryonic cytoplasm during the
cumulative labelling period.
25
380
C. C. WYLIE
plotted against incubation time (Fig. 11) this too rises to a plateau level at about
10 h. This plateau is not horizontal however - a fact which is demonstrated by
plotting the difference between stable and heterogeneous RNA radioactivity.
The resulting graph (Fig. 12) shows a continual increase of the stable RNA over
the heterogeneous RNA level.
(c) The proportion of the total cytoplasmic RNA occupied by both 5 S and
4S RNA are more or less constant throughout the labelling period. The 28 S
RNA, however, rises to a plateau level. This means that the 5 S: 28 S ratio in the
cytoplasm drops over the labelling period (Fig. 13). Presumably therefore the
28 S and 5S species become associated in the cytoplasm during the formation of
the large ribosomal subunit.
DISCUSSION
Despite certain obvious disadvantages, e.g. large pool size and manipulative
difficulties, the avian embryo is probably the most amenable of the higher
vertebrate embryos to a molecular study of this kind, e.g. the embryo grows
initially as a flat disc, which, combined with its successful growth in vitro (New,
1955) makes micro-dissection relatively simple.
Cell counts
Surprisingly few studies of cell number and cell generation time have been
carried out on chick embryos at this early age. Results taken from the literature
can be seen in Table 3.
The chick embryo develops for about 22 h in the oviduct of the hen (Olsen,
1942; Patterson, 1910). In order to achieve the number of cells seen at laying
(see above), the average doubling time must be about 1-4 h. It does not maintain
this average during the pre-laying stages; the generation time falls off to 7-4 h
just before laying (Emanuelsson, 1965). The results presented here suggest that
a doubling time of 7-5 h is maintained for the first 24 h, when the number of
cells reaches about 6 x 105 and then falls off to about 10-7 h (Woodard, 1948).
The onset of stable RNA synthesis
This has already been demonstrated to be under way at the onset of mesoderm
formation in the chick embryo (Lerner, et al. 1963), i.e. about 7 h after
incubation of the laid egg. This last-mentioned paper has been misinterpreted in
the literature (Davidson, 1969) to mean that rRNA is being synthesized 4-18 h
after fertilization. In fact, there is no previously published study which demonstrates synthesis of rRNA before the early primitive streak stage, i.e. gastrulation,
an event which takes place about 27 h after fertilization. This paper demonstrates
that rRNA and 4S RNA synthesis and transport to the cytoplasm are occurring
many hours before this, while the embryo is still undergoing cleavage in the
oviduct or uterus of the hen. This approximately coincides with the time of
nucleolar appearance in the chick embryo (Bechtina, 1960).
RNA metabolism in the early chick embryo
381
Among the vertebrate species studied, this time of onset of stable RNA synthesis is morphogenetically earlier than that seen in amphibians (Brown &
Littna, 1964; Waddington & Perkowska, 1965) and fishes (Aitkhozin, Belitsina
& Spirin 1964); but later than that in mammals (Mintz, 1964; Ellem & Gwatkin,
1968; Woodland & Graham, 1969; Piko, 1970). It has to be remembered,
however, that the times of onset cited above are from a very small number of
species, which tend to be assumed to represent typically the class of vertebrates
to which they belong. The comparison between chick and mouse embryos is by
no means accurate, since it rests on evidence from only six chick embryos. It is
also open to the criticism that the synthesis of small quantities of rRNA may be
obscured by heterogeneous RNA synthesis (Emerson & Humphreys, 1970).
Repeated gel electrophoresis of the 28 S area, using a non-radioactive marker
28 S rRNA, would probably elucidate this point.
It would be interesting to look for the start of rRNA synthesis in other
vertebrates to see if any phylogenetic basis for this phenomenon could be established. For example, the earlier time of onset of rRNA synthesis in chick embryos
may reflect the fact that the store of maternal ribosomes in the fertilized egg is
not as great as in the amphibian species cited above, or in the echinoderm species
studied (Gross, 1967).
It would be interesting for this reason to see if rRNA synthesis in the chick
oocyte proceeds on amplified rDNA, and to measure the store of maternal
ribosomes in the mature egg. This work is already progressing in this laboratory.
Precursor pool size in the early chick embryo
Although these were measured primarily to quantitate the RNA synthesis
from the incorporation of [3H]uridine, there are two interesting points to be
made from them: (1) the rapid conversion of uridine to cytidine in the chick
embryonic cells; (2) the vanishingly small pool of guanosine phosphates in early
chick embryo cells. This fact has also been reported in X. laevis embryos (Brown
& Gurdon, 1966) and may indicate that the passage of guanosine across the cell
membrane, or its de novo synthesis within the cell, represents a quantitative
control mechanism for RNA synthesis. Its very small pool size suggests that
guanosine may be a more efficient isotopically labelled precursor than uridine
in RNA studied. In fact, further experiments are already demonstrating this
(C. C. Wylie, in preparation).
The appearance of newly synthesized cytoplasmic RNA from laying to neurulation
The newly synthesized cytoplasmic RNA per embryo, after a lag period of
about 5 h, rises linearly to reach a level of about 104 fig by the onset of neurulation (Fig. 7 c). On a per cell basis, however (Fig. 7 d), the initial dramatic rise is
followed by a slow decline. This presumably reflects the fact that after about 6 h
of incubation the rate of cell production becomes greater than the rate of
accumulation of newly synthesized RNA. This fact has been reported recently
25-2
382
C. C. WYLIE
in certain echinoderms (Kijima & Wilt, 1969). Whether this change represents
an embryo-wide phenomenon or reflects differentiation within the embryo is an
interesting problem which can be tackled in the chick embryo due to the ease of
micro-dissection (C. C. Wylie, in preparation). The lag period in appearance
of cytoplasmic RNA from 0 to 6 h probably represents the time taken for the
egg to warm up in the incubator to a temperature at which development
can proceed.
The variation seen in these progress curves has been shown to be largely
a variation in the passage of labelled precursor across the membranes into the
embryonic cells. Whether this is a rate-limiting step (i.e. a quantitative control
mechanism) in the synthesis of nucleic acids is a subject requiring further
investigation. The case of guanosine is particularly interesting. The alternative is
simply that it is a process subject to considerable experimental variation,
especially in view of the fact that small batches of embryos (4-12 embryos in
each) have been used in this study.
There are large variations between different species of RNA appearing in the
embryonic cytoplasm during this labelling period. The total cytoplasmic RNA
graph is made up of numerous species of RNA whose proportions are seen to be
shifting in relation to one another (Figs. 8, 10) as the labelling period increases.
(1) The relationship of the 28 S and 18 S rRNA is as expected in the light of
their known biosynthetic pathway (Vaughan, Warner & Darnell, 1967). the
18 S being the first to appear in the cytoplasm, followed by the 28 S rRNA,
which then rises to a value double that of the 18 S (Fig. 10).
(2) The 5 S rRNA component, known to be associated with the 28 S molecule
in the structure of the larger ribosomal subunit (Knight & Darnell, 1967),
maintains a more or less constant percentage of the total cytoplasmic RNA
(Fig. 10). It therefore does not maintain a constant relationship with the 28 S
rRNA, which would be expected if these two molecules were either (a) transcribed
molecule for molecule from their respective genes and/or (b) associated in the
nucleus and transported together to the cytoplasm. The 5 S and 28 S molecules
become associated in the cytoplasm rather than the nucleus during the formation
of the large ribosomal subunit, therefore, in the early chick embryo. These
results do not demonstrate whether the genes for 28 S rRNA and 5 S rRNA are
functionally linked in chick embryos, i.e. whether there is co-ordinate synthesis
of these two RNA types. This is thought to be so in prokaryotes (Roschenthaler
et ah 1969) but not in HeLa cells (Knight & Darnell, 1967), L-cells (Perry &
Kelley, 1968), Drosophilia melanogaster (Tartof & Perry, 1970), Xenopus
laevis oocytes (Ford, 1971) and X. laevis embryos (Abe & Yamana, 1971).
(3) The heterogeneous RNA shows wide fluctuations in quantity over the
labelling period (Fig. 8), which seem to be fairly independent of the stable RNA
components. It shows a relative excess at some times compared with the 28 S
rRNA and a relative deficiency at others. Work is in progress to demonstrate
whether these represent large-scale changes in populations of messenger RNAs
RNA metabolism in the early chick embryo
383
or are extremes of practical scatter. This will only be established after careful
repetition, together with pulse-chase experiments.
I wish to thank Professor J. Z. Young, F.R.S., Dr Ruth Bellairs and Dr Martin Evans for
stimulating and helpful discussion during the course of this work. Expert technical help was
provided by Miss M. McCrossan and Miss M. McNeill. Grateful thanks are also due to the
Medical Research Council, who supported me financially during the course of this work.
REFERENCES
H. & YAMANA, K. (1971). Synthesis of 5 s RNA and its regulation during early embryogenesis of Xenopus laevis. Biochim. biophys. Acta 240, 392-406.
AITKHOZIN, M., BELITSINA, M. & SPIRIN, A. (1964). Nuclei acids during early development of
fish embryos. Biokhimiya 29, 145-152.
BECHTINA, O. G. (1960). Early stages of cleavage in the chick embryo. Arch. Anat. {Strasbourg)
38, 77-85.
BLATTNER, F. R. & ERICKSON, H. P. (1967). Rapid nucleotide separation by chromatography
on cation-exchange columns. Analyt. Biochem. 18, 220-227.
BROWN, D. D. & GURDON, J. (1966). Size distribution and stability of DNA-like RNA synthesised during development of anucleolate embryos of Xenopus laevis J. molec. Biol. 19,
399-422.
BROWN, D. D., & LITTNA, E. (1964). RNA synthesis during development of Xenopus laevis.
J. molec. Biol. 8, 669-687.
BROWN, D. D. & LITTNA, E. (1966). Synthesis and accumulation of DNA-like RNA during
embryogenesis of Xenopus laevis. J. molec. Biol. 20, 81-94.
DAVIDSON, E. H. (1969). Gene Activity in Early Development. New York: Academic Press.
ELLEM, K. & GWATKIN, R. (1968). Patterns of nucleic acid synthesis in early mouse embryos.
Devi Biol. 18,311-330.
EMANUELSSON, H. (1962). Growth and Nucleic Acid Mobilisation in the Early Chick Embryo,
pp. 1-34. Thesis of the University of Lund.
EMANUELSSON, H. (1965). Cell multiplication in the chick blastoderm up to the time of laying.
Expl Cell Res. 39, 3863-99.
EMERSON, C. P. & HUMPHREYS, T. (1970). Regulation of DNA-like RNA and the apparent
activation of ribosomal RNA synthesis in sea-urchin embryos: quantitive measurements
of newly synthesized RNA. Devi Biol. 23, 86-112.
EVANS, M. J. (1969). Studies on the Ribonucleic Acids of Early Amphibian Embryos. Ph.D.
thesis. University of London.
FORD, P. J. (1971). Non-coordinated accumulation and synthesis of 5 s RNA by ovaries of
Xenopus laevis. Nature, Lond. 233, 561-564.
GROSS, P. R. (1967). Control of protein synthesis in embryonic development and differentiation. Current Topics, in Devi Biol. 2, 1-46.
HAMBURGER, V. & HAMILTON, H. L. (1951). A series of normal stages in the development of
the chick. /. Morph. 88, 49-92.
KIJIMA, S. & WILT, F. (1969). Rate of turnover in sea-urchin embryos. /. molec. Biol. 40,235246.
KNIGHT, E. & DARNELL, J. (1967). Distribution of 5 s RNA in HeLa cells. 7. molec. Biol. 28,
491-502.
LERNER, A., BELL, E. & DARNELL, J. (1963). Ribosomal RNA synthesis in chick embryos.
Science, NY. 141, 1187-1188.
MINTZ, B. (1964). Synthesis processes and early development in the mammalian egg. /. exp.
Zool. 157, 85-100.
NEW, D. A. T. (1955). A new technique for the cultivation of the chick embryo in vitro.
J. Embryol. exp. Morph. 3, 326-331.
OLSEN, M. W. (1942). Maturation, fertilization and early cleavage in the hen's egg. /. Morph.
70, 513-533.
ABE,
384
C. C. WYLIE
C. A. & COMPTON, A. (1924). The cultivation of tissues in saline embryonic juice.
Lancet 205, 381-384.
PATTERSON, J. (1910). Early development of the hen's egg. /. Morph. 21, 101-134.
PERRY, R. P. & KELLEY, D. E. (1968). Persistent synthesis of 5 s RNA when production of
28 s and 18 s rRNA is inhibited by low doses of actinomycin D. J. CellPhysiol. 72,235-245.
PIKO, L. (1970). Synthesis of macromolecules in early mouse embryos cultured in vitro: RNA,
DNA and a polysaccharide component. Devi Biol. 21, 257-279.
RICHENBACHER, J. (1956). Die Wirkung des Stilboestrols auf die Keimscheibe des Huhnchens.
Z. Zellforsch. microsk. Anat. 45, 339-377.
ROSCHENTHALER, R., DEVYNCK, M. A., FROMAGEOT, P. & SIMON, E. J. (1969). Inhibition of the
synthesis of 5 s rRNA in E. coli by levallorphan. Biochim. biophys. Ada 182, 481-490.
SOLOMON, J. B. (1957). Nucleic acid content of early chick embryos and the hen's egg. Biochim.
biophys. Acta 24, 584-591.
TARTOF, K. D. & PERRY, R. P. (1970). The 5 s RNA genes of Drosophila melanogaster. J.
molec. Biol. 51, 171-183.
VAUGHAN, M., WARNER, J. & DARNELL, J. (1967). Ribosomal precursor particles in the HeLa
cell nucleus. J. molec. Biol. 25, 235-251.
WADDINGTON, C. H. & PERKOWSKA, E. (1965). Synthesis of RNA by different regions of the
early amphibian embryo. Nature, Lond. 207, 1244-1246.
WOODARD, T. M. (1948). The mitotic index in the chick embryo. Am. Nat. 82, 129-136.
WOODLAND, H. B. & GRAHAM, C. (1969). RNA synthesis during early development of the
mouse. Nature, Lond. 221, 327-332.
PANNETT,
(Manuscript received 27 March 1972)