/ . Embryol. exp. Morph. Vol. 34, 1, pp. 93-112, 1975
Printed in Great Britain
93
The developmental capacity of nuclei transplanted
from keratinized skin cells of adult frogs
By J. B. GURDON, 2 R. A. LASKEY 2 AND O. R. REEVES 1
From the Medical Research Council,
Laboratory of Molecular Biology,
Cambridge, U.K.
SUMMARY
1. Nuclei from keratinized skin cells of adult Xenopus foot-webs have been transplanted
to enucleated eggs of the same species.
2. The cells used to provide donor nuclei were obtained as a monolayer outgrowth from
cultured foot-web explants. When explants were cultured without plasma for 3 days, over
99-9 % of the outgrowth cells contained keratin as revealed by the binding of monospecific
fluorescent antibody prepared against purified Xenopus keratin. Nuclei were transplanted
from cells which had been cultured for 3-| days.
3. None of the first transfer embryos developed as far as tadpoles. Eleven clones of
embryos were prepared from the nuclei of partial first-transfer blastulae by use of serial
nuclear transplantation. Eight of these clones contained swimming tadpoles with functional
muscle and nerve cells, and six clones contained tadpoles with beating hearts, well differentiated eyes, and other organs.
4. To prove that the nuclei of nuclear-transplant tadpoles were derived from the transplanted skin cell nuclei and not from a failure of ultraviolet light to inactivate the recipient
egg nucleus, l-nu skin cell nuclei were transplanted to eggs laid by 2-nu frogs. Several
advanced tadpoles from six clones were analysed for nucleolar and chromosome number
and found to be l-nu diploids.
5. The six clones of advanced tadpoles which were proved to carry the donor nuclear
marker represent six first-transfer nuclei in a total sample of 129 skin cell nuclei originally
transplanted. The probability that all six nuclei were derived from the 01 % of the donor cell
population not proved to contain keratin is less than one in 1010.
6. We conclude that cell specialization does not involve any loss, irreversible inactivation
or permanent change in chromosomal genes required for development.
INTRODUCTION
One of the original purposes of developing the technique of nuclear transplantation in animals was to determine the extent to which nuclei transplanted
from specialized cells to enucleated eggs can support normal development. The
aim was to find out whether nuclei undergo irreversible changes as cells specialize. For this purpose, nuclear transfers have been carried out with nuclei from
1
Author's address: Department of Zoology, University of British Columbia, Vancouver 8,
Canada.
2
Author's address: Medical Research Council, Laboratory of Molecular Biology, Hills
Road, Cambridge, U.K.
94
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
intestinal epithelium cells of feeding larvae (Gurdon, 1962), and from melanophores of young tadpoles (Kobel, Brun & Fischberg, 1973). Using adult frog
tissues, nuclei have been successfully transplanted from cells of kidney, heart,
skin and lung of Xenopus laevis (Laskey & Gurdon, 1970), as well as from
adenocarcinoma tissue of metamorphosed Ranapipiens (McKinnell, Deggins &
Labat, 1969). In the case of adult donor cells, the frequency with which advanced
tadpoles were obtained by nuclear transplantation was not sufficiently high to
exclude the possibility that these successful cases resulted from the transfer of
nuclei from fibroblasts or other cells which are not obviously specialized but
which are usually present in normal and neoplastic adult tissues.
The experiments to be reported here have been carried out with the nuclei of
overtly differentiated cells from adult frog skin. When an explant of frog skin
is cultured on collagen for a few days, a monolayer of cells migrates out, and
well over 99 % of these cells can be shown to contain keratin. The frequency
with which tadpoles are obtained by the transplantation of nuclei from these
cells into enucleated eggs is much higher than 1 %. These experiments therefore
prove that the nuclei of specialized adult skin cells possess the genetic information required to promote the development of tadpoles containing such diversely
differentiated cell-types as those that compose functional muscle, nerve, blood
and eyes.
MATERIAL AND METHODS
Source of animals. Xenopus laevis, imported as wild-type specimens from
South Africa, were used to provide recipient eggs. Animals used to provide
donor nuclei were laboratory-reared individuals which were heterozygous for
the anucleolate mutation, i.e. l-«w* (Elsdale, Fischberg & Smith, 1958).
Preparation of donor skin cells. Two methods have been used to obtain monolayer outgrowths from teased, 2 mm x 2 mm explants of skin from Xenopus
foot-webs. Throughout the text these methods are referred to as 'with plasma'
and 'without plasma'. The previous study, by Reeves & Laskey (1975), of the
synchrony of keratin synthesis and of the proportion of cells synthesizing
keratin, applies only to cultures without plasma.
1.' With plasma.' Explants were cultured in two-thirds diluted chicken plasma
(Difco Laboratories), i.e. 2 vols. of plasma and 1 vol. of H 2 O, which was
allowed to clot by contact with the explant. The preparation was covered with
two-thirds diluted Leibovitz L-15 medium (Flow Laboratories) containing
10 % (v/v) foetal calf serum, and incubated at 25 °C. Under these conditions the
sheets of cells which migrate out from the explant may be more than one cell
thick; some of the cells may divide and keratin is synthesized asynchronously.
* 2-nu, \-nu and 0-nu, sometimes designated as + / + , +/0 and 0/0, refer to individuals
which are wild-type, heterozygous and homozygous respectively for the anucleolate mutation
in which all ribosomal RNA genes have been lost. \-nu and 2-nu nuclei are seen in Figs.
1A, C, E and 3D,
Skin cell nuclear transfers
95
2. ' Without plasma.' Explants were cultured by the method of Reeves &
Laskey (1975), on a substrate of reconstituted rat tail collagen (Ehrman & Gey,
1956). The medium was two-thirds diluted Dulbecco's modified Eagle's medium
(Flow Laboratories) containing 10% foetal calf serum and the incubation
temperature was 22 °C. Under these conditions, the cell sheets which arise by
migration from the explant are only one cell layer thick. These cells do not
divide, as demonstrated by their failure to incorporate [3H]thymidine and by
the absence of mitoses even after exposure to colchicine for up to 5 days
(Reeves & Laskey, 1975).
Nuclear transplantation. To prepare donor cells from skin cultures the
original explants were completely dissected away under a stereomicroscope
and discarded so as to leave only the monolayer sheets of cells which had
migrated out. These keratinizing cells were dissociated by incubation for
20mins in 88 mM-NaCl, 1 mM-KCl, 2-4 mM-NaHC03, 0-5% trypsin, and
20 mM trisodium citrate, pH 7-8. Dissociated cells were washed twice in 88 mMNaCl, 1 mM-KCl, 2-4 mM-NaHC03, 7-5 mM Tris, pH 7-6 and were suspended
in the same solution for nuclear transplantation. In some cases spermine (BDH)
was added to give a 5 mM concentration, following Hennen's (1970) observation that it improves the success of nuclear transplantation. Fraction V bovine
serum albumin (Armour) was also added to a final concentration of 0001 %,
because it was found to make dissociated cells less sticky.
Recipient unfertilized eggs were exposed for about 30 sees to an ultraviolet
light source (mercury arc lamp, UVS 100, 125 W, from Hanovia, Slough) so
as to receive a dose of about 3000 ergs at 2600 A per mm2 (which is approximately the exposed area of a Xenopus egg). Care was taken to expose the animal
pole of the eggs to the u.v. source. This procedure has been found to be at least
99% effective in inactivating the egg nucleus (Gurdon, 1960a).
Nuclear transplantation was carried out as described by Gurdon & Laskey
(19706), using microforge-sharpened pipettes (Gurdon, 1974). Injected eggs
were incubated at 19 °C in full-strength modified Barth solution* (Elsdale,
Gurdon & Fischberg, 1960) during early cleavage, and in a one-tenth dilution
of the same solution after this.
Nucleolar and chromosome counts. To distinguish 2-nu from \-nu individuals,
the number of nucleoli per diploid nucleus was counted. The nucleolar number
of recipient frogs was determined by preparing outgrowths from skin explants
of the foot web. The outgrowths were fixed and stained for counting nucleoli.
The nucleolar number of tadpoles obtained by nuclear transplantation was
determined by phase-contrast microscopy of small fragments, usually the tailtip. At least 50, and usually 100, nuclei were scored in each preparation. In
genetically 2-nu swimming tadpoles, about 25 % of nuclei contain two nucleoli
(Wallace, 1963). For chromosome counts similar pieces of tissue were incubated
•Solution: 88 mM-NaCl, 1 mM-KCl, 2-4 mM-NaHCO3, 0-82 mM-MgSO4, 0-33 mMCa(NO 3 ) 2 , 0-41 niM-CaCl2, penicillin and streptomycin at 10 /^g/ml, and 7-5 mM Tris, pH 7-6.
7
'?-
M B
34
96
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
in modified Barth saline containing 10~4 M colchicine (Sigma Chemical Co.)
for 8-15 h, swollen in a one-tenth dilution of the same medium for 15 mins,
fixed in ethanol: acetic acid (3:1) for 10 mins, stained in aceto-orcein, and
squashed. Nucleolar counts were usually made again on these stained preparations, and always confirmed the conclusions reached from phase-contrast
examinations.
Histological preparations. Specimens for electron microscopy were fixed in
3 % glutaraldehyde, post-fixed in 1 % OsO4, and embedded in Araldite. Sections
were stained with uranyl acetate and lead citrate. Specimens prepared for
histological examination by the light microscope were usually fixed in Bouin's
fluid, sectioned at 10 fiva, and stained in Mayer's haemalum and light green.
Statistical calculation. To determine the probability that our nucleartransplant tadpoles might have been obtained by the transplantation of nuclei
from the few non-keratinized skin cells that exist in the donor cell population,
we have used the Poisson approximation to the binomial distribution, where
probability (P) =
p-nv
k represents the number of successes (in our case, six or four according to the
test carried out); n represents the number of trials (129 in our case); p represents
the proportion of non-keratinized donor cells in our skin cell population,
namely 3 out of 6700 (Reeves & Laskey, 1975).
PLAN OF EXPERIMENTS
The aim of the principal experiment is to obtain the normal development of
a tadpole by transplanting the nucleus of a keratinized skin cell into an
enucleated egg. We first summarize an extensive series of experiments in which
nuclei were taken from cells cultured 'with plasma'. Under these conditions,
FIGURE 1
Differentiation of adult Xenopus skin cells, used as nuclear donors. A, C, E, Phasecontrast photographs (x 325) of living cells: A = skin; C = lung and E =
cultured cell line. B, D, F, ultraviolet fluorescence photographs (x 375) of cells
after binding of fluorescent antibody prepared against purified frog keratin. B =
skin, D = lung and F = cultured cell line.
Skin cultures A and B were monolayer outgrowths of \-nu cells from foot web
epidermis cultured according to 'without plasma' conditions specified in Methods.
These cultures were photographed and tested for antibody-binding 3 days after the
explant was made. The original explant was removed and all cells shown are those
which migrated out from the explant. Only cells in such outgrowth areas were used
to provide nuclei for transplantation (see Methods, p. 4). In B, all cells bound the
anti-keratin antibody. C, D, Cells grown out from 1-nu lung tissue, photographed
live (C) and tested for antibody-binding 3 days after the explant was made (D).
E, F, A cultured line of cells which originated from wild-type, 2-nu embryos.
PHASE
Skin cell nuclear transfers
CONTRAST
FLUORESCENCE
97
No detectable
fluorescence
7-2
98
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
cells which migrate out from a skin explant become keratinized within 10 days,
but cells which grow out from other organs such as kidney or lung, under the
same conditions, do not. These skin cells are described as 'determined', since
they seem committed to keratinize. However, they do not accumulate keratin
synchronously, and at the time of nuclear transfer (6 days after the time of
explantation), some cells may not be differentiated in the sense of containing
keratin. These experiments on cells cultured with plasma are described because
they establish the fact that advanced tadpoles can be prepared by transplanting
nuclei from cells which are determined for skin differentiation. These results
also show that different experiments with cultured skin cell nuclei regularly
yield advanced tadpoles.
To prove that the nucleus of a differentiated cell can support normal transplant-embryo development, we have used donor cells cultured 'without plasma'.
Under these conditions, all cells which migrate out from the explants undergo
keratinization synchronously. By the 3rd day of culture more than 99 % (all
but 3 out of 6700 in a test series) of the cells in the extending monolayer contain
keratin as shown by the binding of fluorescent antibody prepared against
purified Xenopus keratins. Detailed evidence has been presented previously
that the antibody-binding material is keratin (Reeves, 1975) and that it is
present in such a high proportion of cultured skin cells (Reeves & Laskey,
1975). The use of monolayer cell cultures rather than of directly dissociated
skin cells makes it easier to demonstrate the presence of keratin in individual
cells. The use of fluorescent antibody for this purpose is illustrated in Fig. 1.
The absence of keratin in cells grown out from other organs under the same
conditions as skin is described by Reeves & Laskey (1975), and is also illustrated
in Fig. 1. In this paper we describe, in detail, one experiment in which advanced
tadpoles were obtained from the nuclei of keratinized skin cells cultured
without plasma. Donor nuclei were taken from cells which had been cultured
for 3\ days-that is, well after the time by which uniform keratinization is
demonstrable.
It is essential to prove directly that nuclear-transplant tadpoles contain
nuclei which are derived exclusively from the transplanted nucleus and not
from a failure to enucleate the recipient egg. Although the reliability of ultraviolet irradiation for enucleating unfertilized eggs of Xenopus has been demonstrated (Gurdon, 1960a), an important feature of the experiments described
here is the use of a genetic marker to distinguish the products of the transplanted
nucleus from those of a recipient egg nucleus which had escaped inactivation.
For all the experiments to be described donor skin cells have been taken from
frogs which were heterozygous (l-«w)* for the anucleolate mutation (Elsdale
et al. 1958). Nuclei were transplanted into unfertilized eggs of wild-type (2-«w)*
animals. The mitotic progeny of the transplanted nucleus can be recognized by
the presence of only one nucleolus per diploid chromosome set.
* See footnote on p. 2.
99
Skin cell nuclear transfers
Original explant
removed
Adult frog of \-nu strain
as nuclear donor
Parent of 1st transfer
recipient eggs
Enucleation of
recipient eggs
Foot web outgrowth
prove frog was 2-nu
Outgrowth of
epidermal cells
Donor cells for
nuclear transfer
1st nuclear transfer
Cells trypsinized
and washed
Uncleaved
Completely cleaved
(70 V) Martially cleaved
/c o/\
(25%)
Dissociated cells for
serial transfer
I
* ^ i / KJpZ*
Parent of serial titransfer
Enucleation of
1
recipient eggs
recipient eggs
Serial nuclear transfer
Foot web outgrowth
prove frog was 2-nu
Uncleaved
Completely cleaved
(40/O Partially cleaved
(30/0
(30%)
Nuclear transplant tadpole:
l-nu diploid from nucleolus and chromosome counts
(present in 36% of serial clones)
Fig. 2. Plan of serial nuclear transfer experiments, using nuclei from adult skin celJs.
Reasons for the various steps are explained in the text (p. 99). The percentages of
injected eggs which cleave in different ways are those typically obtained with adult
skin cell nuclei. The actual results of the experiments reported here are shown in
Tables 1 and 3.
The overall plan of our experiments is illustrated in Fig. 2. Serial transfers
were carried out in all cases, and the following comments summarize the
reasons for this. A serial nuclear transfer experiment is one in which a donor
nucleus is taken from an embryo which has itself resulted from a previous,
first-transfer, experiment (Fig. 2). Most of the blastulae obtained from first
100
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
Table 1. Nuclear transfers from 'determined' adult skin cells cultured in
plasma clots
First transfers
Results
expressed
as
Serial transfers
Partial
blastulae
Total
used to
Partial or make serial no. of
clones
First
complete transfer
transfers blastulae clones prepared
Numbers
461
As % of clones
As % of total
first transfers 100 %
124f
26-9 %
14
30 %
14
100%
No. of clones containing one or more
embryos which reached:
Blastula
(10 h)
13
93%
2-8%
Neurula
(24 h)
10
71%
2-2%
Muscular
response
(40 h)
6
43%
Heartbeat
(60 h)
6*
43%
1-3%
1-3%
The results in this table were obtained by transplanting nuclei from four separate preparations of skin
cells into eggs of six different females.
* These 6 clones included a total of 15 tadpoles of which 7 were examined for numbers of chromosomes
and nucleoli. All seven were found to be \-nu 2N, and therefore valid products of the transplanted nucleus.
The most advanced of these tadpoles survived for over 1 week at 19 °C, reaching morphological stage
41-42. They then became oedematous and died without feeding.
t Embryos not used for serial transplantation were allowed to develop as far as they could. Partial
blastulae died before gastrulation; no complete blastulae developed beyond the neurula stage.
transfers of nuclei from larval or adult cells are only partially cleaved (Gurdon,
1962; Gurdon & Laskey, 1970a). In all experiments reported here we have
used partial blastulae as donors of nuclei for serial transfers. This is because
our previous results have shown that nuclei from partially cleaved first-transfer
embryos generally support more normal development than the nuclei of completely cleaved first-transfer embryos (Gurdon & Laskey, 1970 a). This effect
can be accounted for in the following way. Partially cleaved embryos frequently
arise from eggs in which the transplanted nucleus passes undivided into only
one of the two blastomeres at the first mitosis, thereby gaining about twice the
normal amount of time in which to complete replication of its chromosomes
(Gurdon & Laskey, 1970 a). Complete first-transfer blastulae are formed from
eggs in which the two daughters of the transplanted nucleus enter the first two
blastomeres and are thought, in many cases, to have been incompletely replicated. Partially cleaved blastulae may therefore contain nuclei which have been
protected from the chromosomal damage which commonly follows nuclear
transplantation (Briggs, King & DiBerardino, 1960; Hennen, 1963). We have
discussed the origin of partial blastulae in more detail elsewhere (Gurdon &
Laskey, 1970a).
Skin cell nuclear transfers
101
Table 2. First transfers of nuclei from differentiated adult skin cells cultured
without plasma
Donor
preparation
I
H
III
IV
V
Total
transfers
15
11
19
30
54
129
Uncleaved and
abortive
cleavage
Partial
cleavage
13
9
16
16
35
1
1
3
11
12
Complete
blastulae
1
1
3
7
12f
Partial blastulae
used for
serial transfers
1
1
1
2
6
11 clones
Total
89
28*
Alias %of
100
9%
total transfers
69%
22%
Donor preparations: 1-V were prepared from different animals. Cells were prepared from
cultures as described in Methods, and were resuspended, for nuclear transfer, in 5 mM
spermine and 0001 % bovine serum albumin. Donor cells were prepared 3^ days after the
foot-web explants were made.
* Partial blastulae not used for serial transplantation died at the late blastula stage.
t Complete blastulae died as late blastulae or during gastrulation.
RESULTS
Development of nuclear transplant-embryos using ''determined'' donor skin cells
cultured with plasma
Table 1 summarizes the results of first and serial nuclear transfers. In agreement with our previous results (Laskey & Gurdon, 1970), about 25 % of the
first transfers resulted in partial or complete blastulae suitable for use as serial
transfer donors. It was not possible within the time available to prepare serial
transfer clones from all 124 partial or complete blastulae obtained. Altogether
14 first-transfer embryos were used to prepare clones, of which 43 % (6 out of
14) contained tadpoles which reached the heart-beat stage of development, and
which carried the donor nuclear marker (Table 1). These clones represent
1 -3 % of the originally transplanted nuclei (Table 1). There is, however, no
reason to doubt that many of the other first-transfer blastulae would also have
yielded tadpoles if it had been possible to perform serial transfers from them
in the time available.
The results summarized in Table 1 are representative of those generally
obtained from cultured skin cell nuclei. In preliminary experiments, 29 clones
were prepared from over 1000 first transfers, and nine of these clones contained
heart-beat tadpoles. Among the variables tested in these experiments was the
survival of first and serial transfer embryos prepared from footweb nuclei of
adolescent frogs (3 months after metamorphosis) and of adult frogs (several
years old). No difference in normality of development was observed when
this comparison was made.
11
9
11
9
12
16
52
25
27
24
22
22
23
93
V:l
V:2
V:3
V:4
V:5
V:6
2
2
4
22
14
1
2
3
5
2
3
4
1
5
15
Neurula
3
(l-nu1)
1
(l-nu: 2N)
(\-nu: 2N)
2
1
13
(l-nu1)
3
Muscular
response
2
2N 4 )
(l-nu)
l
_
_
(2-/?M 2 :4N 1 )
(2-nu2: 4N2)
l-nu10:
(l-nu: 2N)
(2-nu: 4N)
11
Heart-beatf
108(30%)
371(100%) 151(40%)
112(30%)
Total
10
4
Superscripts: e.g. l-nu : 2N indicates that 10 of those tadpoles were shown to be l-nu and 4 shown to be diploid. Some of the embryos
which did not reach the muscular response stage were also found to be l-nu: 2N.
* Example of terminology: clone V: 1 refers to a serial transfer clone prepared from nuclei of a blastula which was itself a first-transfer embryo
from donor preparation V (see Table 2).
f The most advanced heart-beat tadpoles survived for over a week at 19 °C. They reached a morphological stage approximately equivalent to
stage 41 or 42 (Fig. 3 A), but then became oedematous and died without feeding.
19
11
8
3
10
10
6
3
5
IV:2
11
14
30
IV: 1
18
3
22
12
37
111:1
15
5
32
11:1
> Gastrula
No. of embryos reaching each stage:
19
13
17
36
1:1
Complete
blastulae
Uncleaved
and
abortive
cleavage
Partial
blastulae
Total
serial
transfers
in clone
Serial
transfer
clone*
Table 3. Serial transfers of nuclei from differentiated adult skin cells cultured without plasma
^
on
'
p
Z
d
m
>
z
o
O
cj
d
Skin cell nuclear transfers
103
Table 4. Summary of serial transfer experiments with nuclei from differentiated
adult frog foot-web skin cells, cultured without plasma
Number of eggs or embryos
Partial
blastulae
used to
clones containing
embryos which developed
to:
Partial
malrp Qprial
Total 1st
transfers
blastulae
obtained
transfer
clones
129
—
28
—
11
—
—
—
—
—
—
—
4-6%
3-1%
—
—
—
p=<io-10
p=<io-6
STCLY l l c t l
No. of clones containing tadpoles
which were proved to be valid
(l-m/:2Nor2-m/:4N)*
Clones proved to be valid
(as % of clones prepared)
Clones proved to be valid
(as % of all first transfers)
Statistical probability of clones
arising from the <0-l % cells
which were not shown to contain
keratin (see Methods)
1>LJ. KJl oCI la. I II.dl 1MCI
(
N
Muscular
response
Heart-beat
8
6
6
4
54 %
36 %
* Clones 1:1, 11:1,111:1,1V: 1, V:1 and V:2 (see Table 3).
As was explained under 'plan of experiments', skin cells cultured with plasma
are determined but not necessarily differentiated. Therefore the conclusion
from these 'with plasma' experiments is that nuclei from determined skin cells
can promote the development of swimming tadpoles. Heart-beat tadpoles are
contained in nearly half of all clones prepared from the nuclei of such cells.
Development of nuclear transplant-embryos using differentiated donor skin cells
cultured without plasma
First transfers of nuclei yielded partially cleaved embryos in 22% of all
transfers (Table 2). The results of serial transfers are summarized in Table 3
and the overall results in Table 4. Of the 11 serial transfer clones prepared,
eight contained tadpoles with contracting myotomes and therefore with functional muscle and nerve cells; six of these clones contained swimming tadpoles
with differentiated cell-types of various kinds described later.
When interpreting these results, it is important to appreciate that it is the
most normal tadpole or embryo in a serial transfer clone which provides the
best indication of the developmental capacity of the single original skin cell
nucleus from which all members of the clone are derived. There are several
reasons why embryos often develop abnormally even though prepared by the
transplantation of genetically normal nuclei (details in Gurdon, 1963, 1974).
104
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
Table 5. Validity of nuclear transfer embryos reaching the muscular response stage
Numbers of embryos
Type of donor
cell preparation
Recipient
frogs
6 2-nu
\-nu adult skin cultured
with plasma (Table 1
frogs
for details)
(see Table 6)
5 2-nu
\-nu adult skin cultured
frogs
without plasma (Tables
2,3 and 4 for details)
(see Table 6)
Nucleolar
Examined for chromosomes
phenotype Examined
and nucleoli
K
x
of
for
,
embryos nucleoli IN 2N* 3N 4N* Other
1-/7W
.14
2-nu
1-nu
17
1
2-nu
* See text (p. 105) for explanation of these categories.
It is essential to be certain that the most normal members of a clone contain
nuclei which are demonstrated by a nuclear marker to have originated solely
from the originally transplanted nuclei; the evidence for this is presented in the
next section. In drawing conclusions, we consider only those clones which
were demonstrated to be valid in this way. The number of clones containing
genetically marked muscular response tadpoles represent 4-6% of all differentiated skin cell nuclei initially transplanted (Table 4). Table 4 also shows that
at least 3-1 % of all specialized adult skin cells carry the genetic information
required for the development of more advanced heart-beat tadpoles (Fig. 3A)
containing several other kinds of differentiated cells. Over 99-9 % of the donor
skin cell population were demonstrated to contain keratin, and the chance that
all the swimming nuclear-transplant tadpoles obtained in these experiments
could have been derived from the remaining 0-1 % of the donor cell population
is P < 0-000000 000 05 for muscular response tadpoles and P < 0-0000005 for
heart-beat tadpoles (see Methods for the basis of these calculations).
We conclude from these experiments that transplanted nuclei of differentiated
skin cells can promote the development of swimming tadpoles with cells of
many different types.
The use of the nuclear marker to demonstrate the validity of nuclear-transplant
tadpoles
To prove that the embryos obtained are valid products of nuclear transplantation, i.e. that their nuclei are derived only from transplanted nuclei, it is
necessary to demonstrate that these embryos have only one nucleolus in each
diploid nucleus and therefore carry the donor nuclear marker. Since nuclear
transplantation is known to result in tetraploidy in 10-20 % of all cases (Gurdon,
1959), an embryo with two nucleoli per nucleus is also valid, but only if it is
shown to be tetraploid.
Skin cell nuclear transfers
105
Table 6. Nucleolar constitution offemale frogs used to provide recipient eggs for
the first and serial nuclear transfers of adult skin cell nuclei
Type of donor cell
preparation for which
frogs were used
Recipient frog
No. of
nuclei scored
Nuclei with
2-nucleoli(%)
Cultures with plasma
(see Table 1 for details)
A
B
C
D
E
F
549
514
542
525
545
527
24
15
8-5
181
22
16-5
Cultures without plasma
(see Tables 2, 3, 4 for details)
A
B
C
D
E
111
109
123
266
131
22-5
19-2
10-6
15-4
160
100 nuclei in
each of 32 frogs
100 nuclei in
each of 29 frogs
14-5 ± 4-7
2-nu control standards (±SEM)
l-«w control standards (± SEM)
1-4 ± 1-0
Nucleolar counts were performed on stained preparations of foot-web skin cultures grown
in plasma clots.
Control standards refer to 2-nu and l-nu frogs not involved directly in the experiments
reported here. It is likely that the 1-2 % of two-nucleolated cells found in \-nu frog cultures
were tetraploid.
There are technical difficulties in making exact chromosome counts in
amphibian embryos of some species. Yolk may prevent the complete squashing
of preparations so that chromosomes are not distributed in a single focal plane.
For this reason we have not attempted to look for minor karyotypic abnormalities but only to determine the ploidy class so as to distinguish l-nu diploids
(l-nu 2N) from wild-type haploids (l-nu IN), and 2-nu tetraploids (2-nu 4N)
from wild-type diploids (2-nu 2N). Embryos were classed as diploid only when
it was certain that 36 ± 5 chromosomes were present in several (on average six)
mitoses. Similarly chromosome counts of 72 ± 10 were required to classify an
embryo as tetraploid. In several cases where clear resolution of all chromosomes
was possible, exactly 36 were found (seven mitoses in four tadpoles). No cases
of aneuploidy were found nor did haploids or triploids occur. For nucleolar
counts, at least 50, and usually 100, nuclei were scored. It has been shown
previously (Elsdale et al. 1960) that l-nu and 2-nu (wild-type) embryos are
readily distinguished in this way.
Table 5 shows that 31 of the 37 nuclear-transplant tadpoles examined were
l-nu. Six 2-nucleolated tadpoles were obtained and all three of those in which
chromosomes could be counted were tetraploid. Furthermore, chromosome
106
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
preparations were examined in 14 of the 1-nu tadpoles and all were found to
be diploid. Table 3 shows which of the serial transfer clones were proved to be
valid in this way.
l-nu frogs are encountered occasionally in nature at a very low frequency
(Blackler, 1968). To eliminate the remote possibility that the females used to
provide recipient eggs for these experiments could have been l-nu, we have
counted nucleoli in outgrowths from explants of their foot-web skin. Table 6
shows that all of the recipient frogs were 2-nu; since they were diploid, the
nuclei of their eggs could not therefore have contributed to the l-nu diploid or
2-nu tetraploid nuclear-transplant tadpoles. Previous work (Gurdon & Laskey,
1910a) has tested and excluded the possibility that u.v. irradiation of recipient
eggs can convert a genetically 2-nu egg to a l-nu condition.
We consider these tests to provide conclusive proof that the nuclei of nucleartransplant tadpoles described were derived from transplanted nuclei and not
from incompletely inactivated egg pronuclei.
The range and organization of differentiated cell-types present in nuclear transplant
tadpoles
The wide developmental capacity of skin cell nuclei is shown most convincingly
by the range of differentiated cell-types present, and the normality of their
organization, in swimming nuclear-transplant tadpoles. The presence of
functional muscle and nerve cells is evident in tadpoles which swim spontaneously in the culture dish. A beating heart, circulating blood, pigmented
eyes and melanophores can be seen by observation through a stereomicroscope.
A tadpole showing these characteristics is seen in Fig. 3 A, and an enlarged
view of its myotomes in Fig. 3B.
Histological examination of fixed tadpoles confirms the presence and normal
construction of all main tissues. These include heart, striated muscle, brain,
nerve cord, notochord, sucker, pronephros, intestine, and eyes with lens, nerve
fibres, three-layered retina, and pigmented tapetum (Fig. 3F, G). Electron
FIGURE 3
Histological structure of a tadpole obtained by serial nuclear transplantation from
a skin cell grown from a foot web under the 'without plasma' conditions specified
in the Methods section. (A) Whole tadpole showing eye, heart, intestine, myotomes
and melanophores. x 22.
(B) Enlarged view of myotomes showing regular arrangement, x 62.
(C) Transverse section through head region of tadpole showing nerve cord (nc),
notochord (noto), otocysts (oto), and myotomes (myo). x 88.
(D) Pronephros, showing nuclei with single nucleoli. x 308.
(E) Intestine, with epithelium, x 616.
(F, G) Sections through eye showing three-layered retina (G), lens body (lens), lens
epithelium (epithel), inner and outer (ipl and opl) plexiform layers of retinal ganglion
axons (Fig. G), retina, visual layers, and pigmented tapetum. x 308.
Skin cell nuclear transfers
melanophores
myotome
eye
intestine
heart
107
108
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
Skin cell nuclear transfers
109
microscopy shows the expected fine structure of muscle and nerve (Fig. 4A-D).
The presence of histologically normal muscle, lens and blood implies the
synthesis of actin, myosin, crystallins and haemoglobin, and of many other
kinds of proteins usually found in these cells in smaller amounts.
Fig. 3 shows that different tissues and cell-types are arranged normally with
respect to each other. At the ultrastructural level, molecules contained in
striated muscle are also assembled in their characteristic way (Fig. 4B, C).
The normal organization of cells in a tadpole, and of molecules in a cell,
probably depends upon the normal genetic activity of nuclei.
All these specialized cell-types were present in tadpoles which were \-nu
diploids and must therefore have been derived genetically from the nucleus
of a skin cell. In the two sets of experiments described (those with and without
plasma), 36 tadpoles reached the developmental stage shown in Fig. 3A.
DISCUSSION
We consider three types of reservation which might affect our conclusions.
The first concerns the donor cell population. It might be argued that some of
the cells shown to contain keratin have not synthesized it but have received it
by transmission from other cells in the population. We consider this extremely
unlikely for several reasons. When cultured without plasma, skin cells are
strikingly similar not only in structure and in the absence of DNA synthesis but
also in the timing of keratin appearance (Reeves & Laskey, 1975). In these
cells [3H]cysteine is incorporated mainly into keratin and this serves as a measure
of keratin synthesis; autoradiography shows that [3H]cysteine is incorporated
very uniformly (Reeves, unpublished observations). Efficient transmission of
keratins through the medium is most unlikely because of their very low solubility. Furthermore there is no known precedent for the transfer of large
amounts of a cell-type specific protein between specialized cells (see review of
molecular communication between cells by Pitts, 1974). In growing oocytes,
into which large amounts of yolk protein are transported, the surrounding
follicle cells appear to be structurally adapted for this purpose. Small amounts
of protein hormones and viruses may pass between cells, but special mechanisms
Fig. 4. Sections of a swimming tadpole similar to that shown in Fig. 3 A, which
resulted from the serial transplantation of a nucleus from keratinized skin cells
cultured without plasma.
(A) Organized blocks of muscle in the myotomes, and typically vacuolated cells of
the notochord. x 4480.
(B) A longitudinal section through a body muscle showing the characteristic cytological zones, x 21600.
(C) A transverse section through tail muscles showing thick myosin filaments, each
surrounded by six thin actin filaments, x 43 200.
(D) Two sections through the nerve chord showing the characteristic microtubules
(about 25 nm in diameter) in each sectioned axon. x 72000.
110
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
are also thought to exist in these cases. It seems clear that keratin synthesis by
cultured skin cells represents the specialized activity of each of their nuclei.
We wish to emphasize the point that, in all our experiments, the original skin
explant which was several cell layers thick was removed, leaving only the monolayer outgrowth. In each experiment, several parallel cultures were set up;
some were used for antibody tests, and others for the provision of donor
nuclei, but all cultures differentiated very uniformly.
Another type of reservation concerns the relatively low proportion (6 %) of
skin cell nuclei which yielded tadpoles after transplantation. This may seem
surprising in view of the uniformity of the donor cell population. However
there are several reasons why this is a minimum estimate of the developmental
capacity of the donor cells.
In the particular experiments reported here, we have excluded from the
percentage calculated in Table 4 two heart-beat clones because they were not
tested for the presence of the donor nuclear marker. There is, however, no
reason to doubt that the embryos in these two clones were valid products of
nuclear transplantation. Another important point is that it was possible, using
serial transfers, to test the developmental capacity of only 40% (11 out of 28)
of the partially cleaved blastulae derived from first-transfers. It is likely that
the remaining 60% of partial blastulae would also have yielded muscular
response and heart-beat tadpoles, if clones could have been made in the available time.
There are, in addition, several general reasons why nuclear transplantation
does not test the developmental capacity of all nuclei which are genetically
totipotent. For example, there is a fine and not easily controllable difference
between the optimal distortion of donor cells in the injection pipette, and
insufficient distortion which fails to rupture the donor cell. Another possible
source of variation lies in the ability of nuclei to make the immensely rapid
transition from a slow or non-existent division cycle to the initiation and completion of chromosome replication within 90 mins of transplantation to an
egg. Factors of this kind have been discussed in detail by Gurdon (19606) and
by Gurdon & Laskey (1970a). As emphasized previously (Gurdon, 1963)
nuclear transfer experiments can provide only a minimum estimate of the
developmental capacity of a nucleus or of a population of nuclei. It certainly
cannot be concluded from experiments of the kind reported here that most
skin cell nuclei are incapable of supporting normal development.
The last reservation of which we are aware concerns the theoretical possibility that development up to a swimming tadpole stage does not depend on
activity of the tadpole's nuclear genes; a large store of maternal mRNA might,
if brought into progressive use, promote development independently of any
specific nuclear function. In Xenopus, nuclear RNA synthesis has commenced at
the mid-blastula stage, and the expression of paternal genes has been demonstrated in neurulae. These and other reasons (for details see Gurdon, 1974)
Skin cell nuclear transfers
111
argue strongly that the formation of a swimming tadpole is dependent on gene
expression by cell nuclei throughout early development. We therefore conclude
that the formation of a swimming tadpole is a sensitive test of the developmental
capacity of a transplanted nucleus.
We wish to thank Mr N. Thomson for preparing electron micrographs, Mrs Judy Smith
for advice on statistical calculations, and Mr T. Mills and Mrs P. Patel for painstaking
technical assistance. We are very grateful to the following for detailed comments on the
manuscript: R. M. Benbow, A. Colman, C. C. Ford, J. S. Knowland, L. J. Wangh and
N. Williams.
REFERENCES
A. W. (1968). New cases of the Oxford nuclear marker in the South African
clawed toad. Revue Suisse Zool. 75, 506-509.
BRIGGS, R., KING, T. J. & DIBERARDINO, M. A. (1960). Development of nuclear-transplant
embryos of known chromosome complement following parabiosis with normal embryos.
In Germ Cells and Development, pp. 441-447. Baselli, Milano: Inst. Intern. d'Embryol.,
Fondaz.
EHRMAN, R. L. & GEY, G. O. (1956). The growth of cells on a transparent gel of reconstituted
rat-tail collagen. /. natn. Cancer Inst. 16, 1375-1403.
ELSDALE, T. R., FISCHBERG, M. & SMITH, S. (1958). A mutation that reduces nucleolar
number in Xenopus laevis. Expl Cell Res. 14, 642-643.
ELSDALE, T. R., GURDON, J. B. & FISCHBERG, M. (1960). A description of the technique for
nuclear transplantation in Xenopus laevis. J. Embryol. exp. Morph. 8, 437-444.
GURDON, J. B. (1959). Tetraploid frogs. /. exp. Zool. 141, 519-544.
GURDON, J. B. (1960o). The effects of ultraviolet irradiation on the uncleaved eggs of Xenopus
laevis. Q. Jl Microsc. Sci. 101, 299-312.
GURDON, J. B. (19606). Factors responsible for the abnormal development of embryos
obtained by nuclear transplantation in Xenopus laevis. J. Embryol. exp. Morph. 8, 327-340.
GURDON, J. B. (1962). The developmental capacity of nuclei taken from intestinal epithelium
cells of feeding tadpoles. /. Embryol. exp. Morph. 10, 622-640.
GURDON, J. B. (1963). Nuclear transplantation in Amphibia and the importance of stable
nuclear changes in cellular differentiation. Q. Rev. Biol. 38, 54-78.
GURDON, J. B. (1974). The Control of Gene Expression in Animal Development, pp. 121-123.
Oxford: Oxford University Press.
GURDON, J. B. & LASKEY, R. A. (1970a). The transplantation of nuclei from single cultured
cells into enucleate frogs' eggs. /. Embryol. exp. Morph. 24, 227-248.
GURDON, J. B. & LASKEY, R. A. (19706). Methods of transplanting nuclei from single
cultured cells to unfertilised frogs' eggs. /. Embryol. exp. Morph. 1A, 249-255.
HENNEN, S. (1963). Chromosomal and embryological analyses of nuclear changes occurring
in embryos derived from transfers of nuclei between Rana pipiens and Rana sylvatica.
Devi Biol. 6, 133-183.
HENNEN, S. (1970). Influence of spermine and reduced temperature on the ability of transplanted nuclei to promote normal development in eggs of Rana pipiens. Proc. natn. Acad.
Sci., N.Y. 66, 630-637.
KOBEL, H. R., BRUN, R. B. & FISCHBERG, M. (1973). Nuclear transplantation with melanophores, ciliated epidermal cells, and the established cell-line A-8 in Xenopus laevis.
J. Embryol. exp. Morph. 29, 539-547.
LASKEY, R. A. & GURDON, J. B. (1970). Genetic content of adult somatic cells tested by
nuclear transplantation from cultured cells. Nature, Lond. 228, 1332-1334.
MCKINNEL, R. G., DEGGINS, B. A. & LABAT, D. D. (1969). Transplantation of pluripotential
nuclei from triploid frog tumors. Science, N. Y. 165, 394-6.
PITTS, J. (1974). Junctions as channels of direct communication between cells. In Textbook of
Developmental Biology (ed. C. F. Graham & P. F. Wareing). (In the Press.) Blackwell
Scientific Publications.
BLACKLER,
8
EM B
34
112
J. B. GURDON, R. A. LASKEY AND O. R. REEVES
O. R. (1975). Adult amphibian epidermal proteins: biochemical characterization
and developmental appearance. /. Embryol. exp. Morph. 34, 55-73.
REEVES, O. R. & LASKEY, R. A. (1975). In vitro differentiation of a homogeneous cell population - the epidermis of Xenopus laevis. J. Embryol. exp. Morph. 34, 75-92.
WALLACE, H. R. (1963). Nucleolar growth and fusion during cellular differentiation.
J. Morph. Ill, 261-278.
REEVES,
(Received 9 December 1974)