/ . Embryol. exp. Morph. Vol. 60, pp. 359-372, 1980
Printed in Great Britain © Company of Biologists Limited 1980
359
Paternal gene expression in developing hybrid
embryos of Xenopus laevis and Xenopus borealis
ByH. R. WOODLAND AND J. E. M. BALLANTINE 1
From the Department of Biological Sciences, University of Warwick
SUMMARY
We have studied protein synthesis in the viable hybrid Xenopus laevis (?) x Xenopus
borealis (#) using 2D gel electrophoresis. Fourteen borealis-specific proteins were studied.
Two of these proteins appeared by the gastrula stage, five in the gastrula and the rest later.
Where homologous laevis proteins were tentatively identified, androgenetic haploid hybrids
were used to study whether the protein was encoded by stored maternal mRNA, and how
long this mRNA persisted. The two proteins appearing in blastulae were probably initially
coded by stored maternal mRNA. This was not detectable by the tailbud-tadpole stage, and
presumably had been destroyed.
INTRODUCTION
Development is continuously directed by the expression of genetic information, but also by prior gene activity during oogenesis. In the early stages of
amphibian development the number of nuclei is too small to have a major
quantitative impact on gene activity during the rapid progression to the late
blastula stage of development (to form a 20000 cell blastula takes only 9 h).
This is not to say that genes can have no impact at all between the egg and
blastula stages, but rather that the effect can only be developmentally significant
where the product represents a small, but very active fraction of total protein or
RNA synthesis, or is coded by highly re-iterated genes (see Woodland & Wilt,
1980). The consequence of these constraints is that proteins made in very early
development, except for a hypothetical very rare class, are encoded in an oogenetic store of RNA. These conclusions are generally true of embryos that have
large eggs and/or develop rapidly (see Davidson, 1976) but even seem to apply
to mammals, where development is much slower (Braude, 1979). The experiments described below provide information about, the timing of new gene
expression in amphibian development and about the longevity and overall
impact of the stored maternal mRNA.
One approach to detecting new gene action is to look for the appearance of
proteins absent from the oocyte and other early developmental stages. A disadvantage of this approach is that it does not distinguish new gene activity from
1
Author's address: Department of Biological Sciences, University of Warwick, Coventry
CV4 7AL, U.K.
360
H. R. WOODLAND AND J. E. M. BALLANTINE
the mobilization of previously untranslated, stored mRNA. This possibility can
largely be eliminated by extracting the total cellular RNA and translating it in
vitro. A second disadvantage is that it yields information only about the synthesis
of species of mRNA that were not present at earlier stages of development,
whereas many genes might be continuously active throughout early development.
Nevertheless this approach has already provided useful information about
Xenopus development. Brock & Reeves (1978), Ballantine, Woodland &
Sturgess (1979) and Bravo & Knowland (1979) have used 2D electrophoresis
(O'Farrell, 1975) to study a range of proteins synthesized as development
progresses. However, the oocyte makes many more proteins than the egg, and
Ballantine et al. (1979) were unable to identify, with any certainty, proteins
consistently made by eggs and blastulae that were not already made in the
oocyte.1 In addition, translation of the mRNA in vitro indicated that at early
stages, changes in protein synthesis involved post-transcriptional control. These
conclusions had previously been drawn independently with respect to a single
protein class, the histones (Adamson & Woodland, 1974, 1977; Ruderman,
Woodland & Sturgess, 1979). As far as new types of proteins are concerned the
first examples were not detected until the late gastrula stage (Ballantine et al.
1979). Brock & Reeves (1978) obtained essentially the same result, but they
started their study at the egg stage and therefore saw the appearance of what
they called 'new proteins', i.e. relative to the egg, not the oocyte.
An alternative approach for detecting gene activity is to study viable interspecies hybrids. An appropriate example is the cross of X. laevis (?) and
X. borealis (<?). The hybrids develop to adults (Blackler & Gecking, 1972) and
in oocytes of the two species about 30 % of the newly synthesized proteins can be
distinguished by 2D electrophoresis (De Robertis & Black, 1978). We have
used this approach to study paternal gene expression in crosses of X. laevis (?) x
X. borealis ($). Since the sperm cannot carry a store of mRNA, the appearance
of borealis-specific proteins shows that the genes that encode them must be
active. One advantage of this approach is that it reveals the activity of
genes specifying mRNA molecules already present in the oogenetic store of
mRNA.
We have also studied the activity and persistence of stored, maternal mRNA
by using androgenetic haploid hybrids. In these embryos the maternal nucleus
is destroyed by u.v. light (Gurdon, 1960) so the stored mRNA must all be of the
laevis type and the newly synthesized mRNA must exclusively be of the borealis
type. Therefore the appearance and disappearance of the laevis-specific proteins
reveals the mobilization and replacement of stored mRNA.
1
Overall our conclusions are similar to those of Bravo & Knowland (1979). However, we
disagree with their conclusion that new types of proteins are made in the egg, since we have
found probable homologues in the oocyte and later developmental stages. It is possible that
some of their differences result from polymorphisms. We also disagree with some of their
results regarding actin (see Sturgess et al. 1980).
Paternal gene expression in hybrid embryos o/Xenopus
361
The study described below represents an analogous strategy to that adopted
by Woodland, Flynn & Wyllie (1979) to study histone synthesis in development.
MATERIALS AND METHODS
Methods of handling and radioactively-labelling the embryos were described
by Ballantine et al. (1979). Hybrid embryos and androgenetic haploid hybrids
were obtained by the methods of Gurdon (1960) and Woodland et al. (1979).
Developmental stages are numbered according to Nieuwkoop & Faber (1956).
The procedures of sample preparation and the 2D gel electrophoretic analyses
were described by Ballantine et al. (1979) and are based on those of O'Farrell
(1975).
RESULTS
De Robertis & Black (1978) found that all of the proteins specific to the
oocytes of both parents were present in oocytes of the hybrid X. laevis (?) x
X. borealis (<?). They estimated that about 30 % of the proteins made in X.
laevis oocytes were different from those made in X. borealis oocytes - as analysed
by two-dimensional (2D) electrophoresis. We saw fewer proteins specific to
each species during the early stages of development, probably because we
detect fewer newly synthesized proteins in early embryos compared with oocytes.
The number of proteins which can be identified as species-specific increases
during the period of development studied (stage 9-42)..
Table 1 details 14 representative and reproducible borealis-speclfic proteins
whose behaviour we have investigated in detail during the development of
X. borealis embryos and in X. laevis (?) x X. borealis (<J) hybrid embryos. The
positions of all of these proteins in two-dimensional gel electrophoretic analyses
are shown in Fig. 1 (a hybrid neurula embryo). Where proteins are marked with
a subscript they are /oev/i'-specific and may be homologues of the borealisspecific protein with the same number. Fig. 5 A and D shows the specificity of
proteins 1, 2, 3, 4, 5, 6, 8, and 13 to X. borealis (protein 12 is also normally seen
at this stage in X. borealis but is not visible in these Figs.) and proteins 1 L , 2L,
3 L , 5 L , 8L, 13L to X. laevis. The probability of the proteins being homologues is
highest where the two proteins only differ slightly in charge, both appear at the
same stage of development and are found in the same regions of the embryo
(proteins 1 and 1 L , 2 and 2 L , 3 and 3 L , 5 and 5L). In the other cases (8 and 8L,
9 and 9 L and 13 and 13L) the two proteins differ both in charge and molecular
weight; the /tfev/s-specific are slightly larger (about 10 amino acids) and are
more basic. These identifications are tentative without sequence data, but we
do have preliminary evidence that proteins 9 and 9 L have similar products of
partial proteolytic cleavage (unpublished results).
The difference in borealis represents the loss of about 10 amino acids with a
net acid charge. The results are not easily explained by postulating that borealis
and laevis simply differ in their processing enzymes, since the hybrid is like a
9-10
10-12
10-12
10-12
10-12
10-12
20-30
10-12
30-40
30-40
30-40
12-20
9-10
30-40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Persists
Persists
Persists
Persists
Persists
Persists
Persists
T o stage 30
Persists
Persists
Persists
Persists
T o stage 39
Persists
Period
presentf
Whole
Ect
Never present
Never present
20-25
10-12
30-40
9-10
Whole
7Whole||
?8-9
5J|
Whole
Ect
Whole
2L
3L
A
C
6-7
10-12
10-12
16-20
20-25
Never present
homologue
Localization
of laevis
o
O
O
ffl
* See Fig. 1.
t Experiments were not continued beyond stage 40.
t The relation between possible homologues in the two species is open to doubt. However, these proteins are laevis-speci&c and therefore
still provide information about stored mRNAs etc., regardless of their homologues in borealis.
§ 'Ect' is the part of the embryo that contains ectoderm and dorsal mesoderm. 'Whole' implies that the protein also occurs in the remainder 2
of the embryo.
|| Proteins 5 and 5L vary in different series, probably because of polymorphism. These conclusions are therefore tentative.
Stage of
appearance
Borealis
protein
number*
Time at which
Nomenclature of
Putative
Stage at which maternal homologue homologue in
laevis
putative
disappears in
Ballantine
homologuet homologue appears haploid hybrid
et al. (1979)
Table 1. Summarized properties of the species-specific proteins studied
as
to
Paternal gene expression in hybrid embryos 0/Xenopus
60
150K
I-
\
5-5
50
363
pH
/
100 K
12
80 K
10
60 K
50 K
40 K
9L
9 '
14
y
ji
a
ACTIN
30 K
X. laevis X X. borealis
Stage 16-23
2N hybrid
Fig. 1. An autofluorograph of a 2D gel separation of the [35S]methionine-labelled
proteins made by a diploid hybrid embryo {Xenopus laevis x Xenopus borealis 2N)
at the late neurula stage (labelling between stages 16 and 23). The labelled rings allow
identification of the position of all the borealis-spccific proteins and their possible
laevis homologues listed in Table 1. As can be seen, not all of them are present at
this stage. The positions of a, /? and y-actins are also indicated.
mixture of the two parents. Proteins 8, 9 and 13 could be the product of the
same gene, and it could be relevant that 8 and 9 are not present at the same
developmental stage. Again sequence data are needed to elucidate this point.
Paternal proteins whichfirstappear in blastulae
Newly synthesized protein 1 is invariably easy to detect in stage-9 hybrid
embryos (Fig. 2C). Protein 13 can usually just be seen at the stage 9-10 border.
Data concerning the appearance of the proteins studied are summarized in
Table 1.
Paternal proteins first appearing in gastrulae and neurulae
The first stage-specific proteins not made in oocytes are seen in homospecific
crosses at the gastrula stage (Ballantine et al. 1979, see Introduction). It is
364
H. R. WOODLAND AND J. E. M. BALLANTINE
pH
X. laevis X X. borealis
A
2N hybrid
1L
/
13 L
y
7 0
ACT IN
Stage 6^—9
pH
MB
X. laevis
PH
2N
X. laevis X X. borealis
2N hybrid
1u
/
*
7 0
ACT IN
ACT IN
Stage 9-11
Stage 8-10
Fig. 2. Autofluorographs comparing 2D gel separations of proteins made at early
stages of development by diploid hybrid embryos (Xenopus laevis x Xenopus
borealis) and Xenopus laevis embryos. (A) At stage 6|-9 there are no differences
between X. laevis and the hybrid. Spots 1L and 13L first appear in both at this stage.
(C) Between stages 8 and 10 the first borealis-speclfic protein spot appears (1). The
difference between X. laevis and the hybrid can be seen by comparing B and C.
i?0ra?//.s-specific protein 13 is normally also just detectable at this stage. The embryos
were injected with [35S]methionine at the first stage indicated and frozen at the
second.
therefore not surprising that we see increasing numbers of species-specific
proteins in hybrids from this stage on. In the gastrula proteins 2, 3, 4, 6 and 8
become detectable, though in some cases very faintly (Fig. 3 A and B). By stage
20 proteins 5 (Fig. 3C and D) and 12 are added (although protein 12 is not
visible in Fig. 3D its position can be seen in Fig. 3F).
Paternal gene expression in hybrid embryos o/Xenopus
365
Proteins appearing at later stages
Further proteins appear as development proceeds. These include 7, 9, 10, 11
and 14 (Fig. 3E and F). Our analysis has not been extended beyond stage 40,
which is the pre-feeding tadpole stage.
Behaviour of \a.Qvis-specific proteins in androgenetic haploid hybrids
In androgenetic haploids from the mating Xenopus laevis (?) x X. borealis (c?)
the cytoplasm, and hence any mRNA stored during oogenesis, will be exclusively
laevis. On the other hand the only nuclear genes present will be from the borealis
sperm. Thus as development proceeds, all newly synthesized mRNA molecules
will be of the borealis type. Any newly synthesized /aev/s-specific proteins can
only have been made on maternal mRNA. These androgenetic haploid hybrids
thus enable us to follow the activity of stored mRNA and its replacement by
new transcripts (see also Woodland et al. 1979).
Androgenetic haploids were prepared by the method of Gurdon (1960) as
described by Woodland et al. (1979). The efficiency with which the maternal
nuclei were destroyed was measured by counting nucleoli in control haploid
X. laevis x X. laevis embryos, as well as by following the disappearance of
newly synthesized maternal-type HI histone. The experiments were performed
on the same matings already described by Woodland et al. (1979), who provide
the detailed data on enucleation. The success of enucleation was at least 98 %.
mRNAs coding for /tfev/s-specific proteins could behave in one of three ways:
(1) They could be synthesized and stored exclusively in oogenesis. This would
mean that the proteins would be made in all early embryos derived from a
laevis mother, but unless the mRNA was very stable they would cease synthesis
by later stages.
(2) They could be stored during oogenesis and also made through early
development. This would mean that the proteins they encode would be made at
early stages in all eggs laid by laevis females and would continue to be made
both in laevis embryos and in laevis (?) x borealis (S) diploid hybrids. They
would cease synthesis in laevis (?) x borealis (S) androgenetic hybrids.
(3) They could be made only in embryos, and never in oocytes. Therefore the
proteins that they encode would be absent from early embryos of all types and
would appear at specific developmental stages in laevis (?) x laevis ($) embryos
or laevis (?) x borealis (<$) diploid hybrids. They would never be synthesized
in androgenetic haploid hybrids.
No proteins of the first type were found, but there were various proteins of
categories (2) and (3). Thus in a mixture of radioactive protein preparations
from stage-37 X. laevis and borealis tadpoles, 39 /tfevw-specific proteins could
be seen. Ten were detectable in embryos labelled with [35S]methionine from
stage 91 to 10| and 30 in embryos labelled between stages 12 and 20. The seven
/tfev/s-specific proteins indicated in Fig. 1 and Table 1, column 4, illustrate both
24
EMB 60
Molecular weiqht
\
,TO
.•
^
I
*
CT
O
f
i
«X.
V
-
» Molecular weight
:
O1
X
CO
\
N
Molecular weight
X
•o
Paternal gene expression in hybrid embryos o/Xenopus
367
of these categories and are singled out because they may be homologous with
borealis proteins (see above). There is little to be gained by describing the other
26 in detail.
Protein 1 L is clearly of type 2 (Fig. 6), as are the histones (Woodland et al.
1979). In the haploid hybrid, lLbecomes detectable in the mid-blastula (Fig. 4A)
though it can hardly be seen in the newly synthesized proteins of the egg or
early morula (Ballantine et al. 1979, Fig. 2, protein A). This shows that 1 L
represents the product of a mobilized maternal mRNA under translational
control. Protein 1 L becomes fainter in the haploid hybrid gastrula and is
indetectable after stage 20 (Fig. 4C). Since protein I seems to be its borealis
homologue we interpret our observations to mean that the stored transcripts
encoding protein 1 L are replaced by new mRNAs synthesized from the blastula
stage onwards. The status of protein 5 L is uncertain (see Table 1) but proteins
13L and 2 L also seem to be of type (2). They appear between stages 9 and 12 in
normal laevis embryos, hybrids and androgenetic haploid hybrids, and then
disappear between stages 20 and 25 in haploid, but not diploid hybrids. Their
putative borealis homologues are first detectable at about the same stage, but
much more maternal than paternal-type protein is seen in early diploid hybrids.
The genes for 2 L and 13L seem to be active in oogenesis, then through early
development, from at least the late blastula stage onwards. Unlike 1 L and 2 L
protein 13L may not be made after stage 40.
Proteins 3 L , 8L and 9L are representatives of class 3. They were not detected
in androgenetic haploids at any stage. They are therefore absent from the
oogenetic mRNA store, at least at detectable levels, and reveal new gene activity
at a range of stages from the gastrula onwards (Table 1, column 5).
DISCUSSION
Onset of gene activity in development
Once a gene becomes active, our ability to detect its protein will always depend
on four variables: the sensitivity of the assay, the background, the intensity of
activity of the gene and the reiteration frequency of the gene. For example, the
HI histone genes are reiterated 20-50 times (Jacob, Malacinski & Birnstiel,
1976). If all are active in development their protein should be detectable at
Fig. 3. Autofluorographs comparing the 2D gel separation of proteins from embryos
of X. laevis x X. borealis diploid hybrids with those of X. laevis at different stages
of development. At stage 10-12 (A, B) 6oreo/w-specific spots 2, 3, 4, 6, 8 appear and
spots 1 and 13 are still present. At stage 16-22/23 (C, D) borealis-specific spot 5
appears and borealis-sptcxfiQ spot .12 is also normally seen at this stage. At stage 42
(E, F) it is possible to see borealis-spQcific spots 7, 9, 10, 11, 14 though some of the
earlier appearing borealis-^Qc\fYC spots (8,13) are no longer detectable. The embryos
were injected with [35S]methionine at the first stage indicated and frozen at the
second.
24-2
368
H. R. WOODLAND AND J. E. M. BALLANTINE
pH
a A
XJpevis X X. borealis
PH
1N hybrid
C
X laevis X X. borealis
1N hybrid
1
1L
13
'A
Stage 7-8
Stage 10-12
ACTIN
B X. laevis * X. borealis
•
-
~
Stage 9-10
/f
ACTIN
1N hybrid*
D
n
Stage 1 2 - 2 0
X. laevis x X. borealis
W
h
Ybrid
*
ACTIN
ACTIN
Fig. 4. Autofluorographs showing changes in the patterns of 2D gel separations of
proteins made by haploid hybrid (X. laevis x X. borealis 1N) embryos during early
development. Stage 7-8 (A) shows the appearance of the laevis protein 1L followed
by the appearance at stage 9-10 (B) of the borealis-specific homologues (1). By stage
10-12 (C) the relative intensity of spots 1 and 1L have changed compared to B and
to the diploid hybrid (Fig. 3B). By stage 12-20 (D) spot 1L has disappeared. The
absence of other /aevw-specific spots at this stage (2L, 3L, 8L, 13L) is also indicated
in D by labelled arrows. Radioactive labelling was as in Fig. 2.
about the 500-cell stage (Woodland & Wilt, 1979) and it can, in fact, be detected
as a radioactive protein in X. laevis x X. borealis hybrids at the 1000-10000cell stage, only an hour or so later (Woodland et al. 1979). HI histone represents
about 2 % of protein synthesis at this stage (Adamson & Woodland, 1977;
Flynn & Woodland, 1980). For analyses of comparable sensitivities and
against similar backgrounds other proteins made at the same efficiency should
be detected at the same stage, if their genes are also reiterated 20-50 times.
On the other hand there would need to be about 20000-50000 cells if the gene
were single copy, i.e. the protein would first be detected at some time between
stage 9 and stage 16.
The first new gene activity that we detect appears at stage 9 (10000-30000
cells, Dawid, 1965) which is as soon as one could reasonably expect, given the
Paternal gene expression in hybrid embryos o/Xenopus
2N
X. borealis
369
X. laevis X X. borealis
\
/
2N
/
hybrid
.V
2L
3-L
Stage 16-23
7
-ACTIN
7 P «
ACTIN
Stage 16-22
X. laevis X A", borealis
B
1N
2N
X. laevis
hybrid
5L
\
1L
8L
Stage 19-24
*CTIN
13L
y
Stage 16-23
2L
\
ACTIN
Fig. 5. Autofluorographs of 2D gel separations of protein from late neurulae
(stage 16-23) of X. borealis and X. laevis non-hybrids compared with their diploid
and haploid hybrids. A and D show the presence of the spots specific to the parental
types. C shows that thefow«?//j-specificspots are present in the 2N hybrid together
with a possible laevis homologue in some cases. B shows that the /oev/j-specific spots
are absent from the haploid hybrid and it resembles the borealis parent (spot 12 is
usually seen in auto-fluorographs of A, B and C at this stage). Other details are as in
Fig. 2.
methods used. This means that if such genes were equally active at earlier stages
their products would not have been detected. The same is true of the more
extensive gene expression detectable in the gastrula (Table 1).
Thus we may now say that HI histone genes are not alone in being active in
the amphibian blastula. Unfortunately we do not yet know the function of
these other proteins. In the gastrula more genes are active. Some are regionspecific and none were detectable at earlier stages. These supplement a-actin
(Ballantine et al 1979), collagen (Green, Goldberg, Schwarz & Brown, 1968),
370
H. R. WOODLAND AND J. E. M. BALLANTINE
X. laevis X X. laevis
diploid
Stage 7-9
Stage 9-10
.
S
X. laevis X X. borealis
diploid
X toev/5 X X. borealis
haploid
/
. W
•Stage 10-12
Stage 12-20
Fig. 6. Diagram illustrating the behaviour of proteins 1 (borealis-spec\f\c) and 1L
(/ams-specific) during the early development of hybrid and non-hybrid embryos.
The unlabelled spot in the diagram is common to both species.
tyrosine (Benson & Triplett 1974) and various isoenzymes (Johnson, 1971;
Johnson & Chapman, 1972; Wright & Subtelny, 1971, 1973; Wright, 1975) on
the list of proteins first detectably made at the gastrula or neurula stage of
amphibian development.
Stored maternal mRNA
Certain of the proteins studied here are /oevw-specific. It they were synthesized
by the androgenetic haploid hybrid they must have been coded by RNA
synthesized by the maternal genome during oogenesis, since the embryo will
contain only paternal X. borealis genes. If they never appear then it is most
likely that they represent proteins made on transcripts synthesized only after
fertilization. The majority come into the second category (Table 1), but two are
certainly represented in the maternal store (proteins 1 L and 13L). These proteins
have putative borealis homologues. If this identification is correct, these proteins
may be placed in a regulatory class with HI histones. Each becomes detectable
at the early blastula stage, as the result of mobilization of stored maternal
mRNA. Each manifests paternal gene expression by the late blastula stage,
although HI is detectable slightly earlier than the proteins studied here (Woodland et al. 1979). In each case the activity of the maternal transcripts disappears
after gastrulation. In the absence of maternal genes, the stored HI transcripts.
Paternal gene expression in hybrid embryos 0/Xenopus
371
vanish by the early gastrula stage, whereas protein 1 L disappears in the neurula
and 13L a little later. The functional instability of histone mRNA may be
ascribed to its lack of a 3'poly (A) tract (Huez et al, 1978; Woodland & Wilt,
1979). The polyadenylation state of the mRNA encoding proteins 1 L and 13L is
not yet known. In any case both contrast with polyadenylated globin mRNA
injected into 1- to 2-cell embryos the activity of which remains detectable to
the tadpole stage (Gurdon, 1974).
REFERENCES
D. & WOODLAND, H. R. (1974). Histone synthesis in early amphibian development: histone and DNA synthesis are not co-ordinated. / . molec. Bio/. 88, 263-285.
ADAMSON, E. D. & WOODLAND, H. R. (1977). Changes in the rate of histone synthesis during
oocyte maturation and very early development of Xenopus laevis. Devi. Biol. 43, 159-174.
BALLANTLNE, J. E. M., WOODLAND, H. R. & STURGESS, E. A. (1979). Changes in protein
synthesis during the development of Xenopus laevis. J. Embryo), exp. Morph. 51,137-153.
BENSON, S. C. & TRIPLETT, E. L. (1974). The synthesis and activity of tryosinase during
development of the frog Ranapipiens. Devi. Biol. 40, 270-282.
BLACKLER, A. W. & GECKENG, C. A. (1972). The transmission of sex cells of one species
through the body of a second species in the genus Xenopus. II Interspecific matings.
Devi Biol. 27, 385-394.
BRAUDE, P. R. (1979). Control of protein synthesis during blastocyst formation in the mouse.
Devi Biol. 68,440-452.
BRAVO, R. & KNOWLAND, J. (1979). Classes of proteins synthesised in oocytes, eggs, embryos
and differentiated tissues of Xenopus laevis. Differentiation 13, 101-108.
BROCK, H. W. & REEVES, R. (1978). An investigation of de novo protein synthesis in the
South African clawed toad, Xenopus laevis. Devi Biol. 66,128-141.
DAVIDSON, E. H. (1976). Gene Activity in Early Development. New York: Academic Press.
DAWID, [. B. (1965). Deoxyribonucleic acid in amphibian eggs. /. molec. Biol. 12, 581-599.
DE R.OBER.T1S, E. M. & BLACK, P. (1978). Hybrids of Xenopus laevis and Xenopus borealis
express proteins from both parents. Devi Biol. 68,334-339.
FLYNN, J. M. & WOODLAND, H. R. (1980). The synthesis of histone HI during early amphibian
development. Devi Biol. (In the Press.)
GREEN, H., GOLDBERG, B., SCHWARZ, M. & BROWN, D. D. (1968). The synthesis of collagen
during the development of Xenopus laevis. Devi Biol. 18, 391-400.
GURDON, J. B. (1960). The effects of ultraviolet irradiation on the uncleaved eggs of Xenopus
laevis. Q. J. microsc. Sci. 101,299-312.
GURDON, J. B. (1974). The Control of Gene Expression in Animal Development. Oxford:
Clarendon Press.
ADAMSON, E.
HUEZ, G., MARBAIX, G., GALLWITZ,D., WEINBERG, E., DEVOS, R., HUBERT,E. &CLEUTER, Y.
(1978). Functional stabilization of HeLa cell histone messenger RNAs injected into
Xenopus oocytes by 3'-OH polyadenylation. Nature, Land. 271, 572-573.
JACOB, E., MALACINSKI, G. & BIRNSTIEL, M. L. (1976). Re-iteration frequency of the histone
genes in the genome of the amphibian, Xenopus laevis. Eur. J. Biochem. 69, 45-54.
JOHNSON, K. E. (1971). A biochemical and cytological investigation of differentiation in the
interspecific hybrid amphibian embryo Rana pipiens $ x Rana sylvatica $. J. exp. Zool.
Ill, 191-206.
JOHNSON, K. E. & CHAPMAN, V. H. (1972). Expression of paternal genes during embryogenesis
in the viable interspecific hybrid amphibian embryo Rana pipiens $ x Rana palustris <?.
Five enzyme systems. /. exp. Zool. 178,313-318.
NIEUWKOOP, P. D. &FABER, J. (1956). NormalTable o/Xenopus laevis (Daudin). Amsterdam:
North Holland.
372
H. R. W O O D L A N D AND J. E. M. BALLANTINE
P. H. (1975). High resolution two dimensional electrophoresis of proteins. J.
biol. Chem. 250,4007-4021.
RUDERMAN, J. V., WOODLAND, H. R. & STURGESS, E. A. (1979). Modulation of histone
messenger RNA during the early development of Xenopus laevis. Devi Biol. 71, 71-82.
O'FARRELL,
STURGESS, E. A., BALLANTINE, J. E. M., WOODLAND, H. R., MOHUN, P. R., LANE, C. D. &
DIMITRIADIS, G. J. (1980). Actin synthesis during the early development of Xenopus laevis.
J. Embryol. exp. Morph. 58, 303-320.
H. R., FLYNN, J. M. & WYLLIE, A. J. (1979). Utilization of stored mRNA in
Xenopus embryos and its replacement by newly synthesised transcripts: histone HI synthesis
using interspecies hybrids. Cell. 18,165-171.
WOODLAND, H. R. & WILT, F. (1980). The stability and translation of sea urchin histone
mRNA injected into Xenopus laevis eggs and developing embryos. Devi Biol. 75, 214-221.
WRIGHT, D. A. & SUBTELNY, S. (1971). Nuclear and cytoplasmic contributions to dehydrogenase phenotypes in hybrid frog embryos. Devi Biol. 24, 119-140.
WRIGHT, D. A. & SUBTELNY, S. (1973). Effects of haploidy and hybridization on the activity
of four dehydrogenases in frog embryos. Devi Biol. 32; 297-308.
WRIGHT, D. A. (1975). In Isozymes. IV. Genetics and Evolution (ed. C. L. Markert), pp. 649654. New York: Academic Press.
WOODLAND,
{Received 14 March 1980, revised 9 May 1980)