J. Embryol. exp. Morph. 76, 217-234 (1983)
Printed in Great Britain © The Company of Biologists Limited 1983
217
Replacement of irradiated epidermis by migration of
non-irradiated epidermis in the newt limb: the
necessity of healthy epidermis for regeneration
By EMILE LHEUREUX 1
From the Laboratoire de Morphogenese Animate, Universite des Sciences et
Techniques de Lille
SUMMARY
An X-irradiated newt limb is able to regenerate if non-irradiated skin as well as nonirradiated muscle is transplanted to the stump. Non-irradiated epidermis is brought to the
stump with a skin graft but not with a muscle graft. In order to know whether limb regeneration
required healthy epidermis or not, a triploid skin cuff was set at the most proximal level of an
irradiated limb and muscle was transplanted to the level of the midforearm. The forearm was
then amputated through the muscle graft. A cytophotometrical analysis of DNA content of
the epidermis cell nuclei sampled from the skin of the regenerate was undertaken to detect a
migration of triploid epidermal cells. The result was a complete replacement of diploid
irradiated epidermis by triploid epidermis, during the six weeks necessary for regeneration.
Another investigation consisted of detecting a possible migration of non-irradiated triploid
epidermis along an irradiated limb which had not been amputated. Healthy epidermis was
found to migrate distally and replace irradiated epidermis in three weeks. Previous experiments involving transplantation of a non-irradiated skin cuff or muscle to an irradiated limb
stump were carried out again but on animals which had been entirely irradiated to prevent any
extra healthy epidermis cells from contaminating the regenerating limb epidermis. A
regenerate developed from the skin graft but not from muscle graft. It is concluded that
healthy epidermis must be present on the limb stump to permit the blastema to develop.
INTRODUCTION
Shortly after the amputation of a newt limb, the process of wound healing
begins. In its earliest phase, this process is characterized by the migration of
epidermis from the wound margin across the amputation surface (Singer &
Salpeter, 1961; Norman & Schmidt, 1967; Repesh & Oberpriller, 1978, 1980).
In a limb blastema, the apical epidermis has been demonstrated to be necessary for growth of the underlying mesodermal cells in vivo (Thornton, 1957;
Stocum, 1968; Michael & Faber, 1971; Stocum & Dearlove, 1972) and in vitro
(Globus Vethamany-Globus & Lee, 1980). These results can be explained if the
apical epidermis is responsible for the maintenance of the dedifferentiated state
of the mesodermal cells at the tip of the blastema (Globus et al. 1980).
1
Author's address: Laboratoire de Morphogenese Animale, Universite des Sciences et
Techniques de Lille, 59655 Villeneuve d'Ascq Cedex, France.
218
E. LHEUREUX
X-irradiation inhibits limb regeneration but allows the irradiated epidermis to
migrate over the amputation surface (Butler, 1933). Limb regeneration is inhibited because irradiated cells fail to complete mitosis (Maden, 1979). Nonirradiated limb tissues, transplanted to an irradiated stump can promote
regeneration. The tissues most effective in inducing regeneration are skin
(Umanski, 1937; Trampush, 1951; Rahmani & Kiortsis, 1961; Carlson, 1974;
Lheureux, 1975/?) and muscle (Umanski, 1937; Thornton, 1942; Trampusch,
1951; Lheureux, 19756).
It is interesting to note that in skin grafts, both mesodermal and epidermal
tissues are present while the muscle grafts contain only mesodermal tissues. The
results with the muscle grafts suggest that irradiated epidermis may be able to
support regeneration. One cannot eliminate the possibility of a participation of
either irradiated or non-irradiated epidermal cells in blastema formation. The
fate of irradiated epidermis is still unknown. Studies on changes in the irradiated
limb epidermis (Brunst, 1963; Rose & Rose, 1965; Maden & Wallace, 1976;
Wertz, Donaldson & Mason, 1976) have provided no satisfactory answer to this
question. Thus, in the muscle graft experiments, if non-irradiated epidermis is
required for blastema formation, a migration of healthy epidermis from the
shoulder region occurs.
This study was undertaken in order to follow the fate of irradiated epidermis
and to test whether there is a requirement for a healthy epidermis in limb
regeneration. We used a triploid cell marker which has previously permitted the
determination of the origin of the mesodermal tissues of X-irradiated limb
regenerates (Namenwirth, 1974; Lheureux, 1983).
MATERIALS AND METHODS
Pleurodeles waltlii larvae bred in our laboratory and ranging in length from
6-8 cm were used in all experiments.
X-irradiation
After anaesthetization in 1:1000 MS 222 (Sandoz) the right forelimb of the
animals was irradiated from the shoulder to the tip with 4000 rads of X-rays. In
the earliest experiments (series 1 of this report) however, a dose of 2000 rads was
used. When an irradiation of the entire body was carried out, each animal was
irradiated for the same length of time as were the limbs that received 4000 rads.
Triploid animals
Triploid animals were obtained by a cold treatment of freshly spawned eggs
(Beetschen, 1960). Fertilized eggs were collected from a spawning female and
placed in a 0°C water bath for 10 h, then returned to room temperature to
develop. Among surviving larvae, 85 % were triploid. They were distinguished
Epidermal migration and limb regeneration
219
from diploid animals by a comparative analysis of DNA content of their bloodcell nuclei.
Analysis of cell ploidy
In order to establish the percentage of diploid and triploid cells in the epidermis of the regenerates, squashes of epidermal samples were carried out and the
DNA content of cell nuclei was measured. Epidermis was first separated from
dermis by soaking pieces of skin for 20 min in 0 • 1 % EDT A in modified Steinberg
solution lacking calcium and magnesium and adjusted to pH 8-3 with Tris buffer.
The epidermis was then fixed in Carnoy's fixative for 15 min. Control diploid
epidermis and regenerate epidermis were each placed in a drop of 45 % acetic
acid in water on the same microslide and squashed with the thumb between two
microslides and allowed to freeze for a few minutes at — 60 °C. When the fluid
film between the slides was frozen, the slides were separated from each other
with a razor blade. After pretreatment in 1 % collodion both slides were
hydrolysed in 1 N - H C I at 60 °C for 7 min and stained in Feulgen for 1 h. Measurements of DNA content were made by the two-wavelengths method (560 nm,
498 nm) using a Leitz MPV cytophotometer. From the values obtained in arbitrary units by measurements of 100 control diploid cell nuclei a mean value m\
was calculated. From this value a second value mi corresponding to DNA content of triploid cell nuclei was calculated {mi =1-5 mi). Measurements of DNA
content of experimental epidermal cell nuclei were compared with both m\ and
mi values. Histograms were drawn so that diploid and triploid cell populations
could be easily distinguished.
Experimental series
Three series were carried out to follow the fate of irradiated epidermal cells
of the newt limb (series 1-3) and two series to test the need of healthy epidermis
in limb regeneration (series 4-5).
Series 1. The irradiated skin of the proximal upper arm of the right forelimb was
replaced by a non-irradiated skin cuff from a triploid newt. The dorsoventral axis
of the skin graft was reversed to distinguish it from the host skin for several
months by examination of the pigmentation contrast between dorsal and ventral
dermis. In addition two pieces of muscle from both dorsal and ventral sides of
the left non-irradiated forelimb were grafted under the skin in the middle of the
right forearm. Two days later, the right limb was amputated through the graft.
When regeneration was finished, the ploidy of epidermal cells from different
levels of the limb was determined (Fig. 1).
Series 2. The results of series 1 established that the irradiated epidermis disappeared and the regenerate was covered with non-irradiated epidermis which
migrated distally. To follow the progress of healthy epidermis ten animals
220
D
E.
D
LHEUREUX
D
Fig. 1. Experimental procedure of series 1. (A) The right limb is irradiated: shaded
area. (B) The more proximal irradiated skin is removed and replaced by a nonirradiated skin cuff from a triploid animal. (C) In the middle of the forearm two
muscle fragments from ventral and dorsal sides of the contralateral limb are transplanted under the stump skin to induce regeneration. (D) When regeneration is
finished the analysis of ploidy of epidermis from four zones is carried out. D: diploid;
T: triploid; vm: ventral muscle; dm: dorsal muscle.
underwent the same operation as the first series, except that no muscle grafts
and no amputations were performed. At regular intervals of time ranging from
the 4th day to the 22nd day after the operation, samples of epidermis from the
distal stylopodium and from the hand were analysed to follow the progress of
triploid epidermal cells towards the tip of the limb (Fig. 2).
Series 3. Because irradiated diploid cells in S or G2 could be confused with
triploid cells, additional measurements of DNA content of irradiated epidermal
cell nuclei were carried out. Two animals were entirely irradiated to eliminate
the possibility of contamination by migrating non-irradiated cells and the percentage of irradiated epidermal cells in S and G2 was calculated at regular intervals
of time. Epidermal samples were studied at various times from 4 to 23 days after
irradiation. Animals died 24 and 26 days after irradiation.
Epidermal migration and limb regeneration
D
221
D
Fig. 2. Experimental procedure of series 2. (A) The right limb is irradiated. (B) The
more proximal irradiated skin is removed and replaced by a triploid skin cuff. (C) At
regular intervals of time ranging from the 4th day to the 22nd day after the
X-irradiation the analysis of ploidy of epidermis from four zones is carried out. D:
diploid; T: triploid.
Series 4 and 5. To test whether there is a requirement for a non irradiated
epidermis in regeneration other experiments were carried out. Twenty four
animals in each series were entirely irradiated to eliminate any source of nonirradiated epidermis other than the epidermis brought with a graft. The grafts
consisted either of a skin cuff (series 4) or of muscle (series 5) from nonirradiated limbs. They were transplanted to both right and left upper arms of
animals which had been irradiated a few hours earlier. The following day the
limbs were amputated through the grafts. Fig. 3 illustrates the experimental
procedure.
RESULTS
Series 1
To find out whether the epidermis which covers the regenerate corresponds to
EMB76
222
E.
LHEUREUX
Fig. 3. Experimental procedure of series 3. (A) The animal is entirely irradiated. (B)
A non-irradiated skin graft is transplanted to the forelimb stump. (C) Two fragments
of dorsal and ventral forelimb muscles from a non-irradiated animal are transplanted
under the skin of the forelimb stump. D, d: dorsal; A, a: anterior; V, v: ventral; P,
p: posterior.
the diploid irradiated epidermis which was present between the level of amputation and the grafted triploid skin cuff or whether non-irradiated triploid epidermal
cells are present on the regenerate, the ploidy of epidermal cells was determined in
Epidermal migration and limb regeneration
Percentage
of cell nuclei
223
30 •
IV
Epidermis from
regenerating hand
III
Epidermis
from distal
upper arm
II
Epidermis from 10
grafted triploid
skin cuff
Control diploid
epidermis
42 Days
114 Days
144 Days
Fig. 4. Results of series 1. Histograms were drawn from measurements of epidermal
cell DNA content. Samples from four different zones of animal number 1 were
analysed 42 days after the operation. The epidermis of the cuff and the epidermis of
the distal zones are triploid. The same four zones of the epidermis of animal number
4 were successively analysed 114 days and 144 days after the operation. It is shown
that a slow migration of healthy epidermis occurred. m\\ mean diploid value; mi:
triploid value.
DNA
content
224
E. LHEUREUX
Table 1. Ploidy of epidermal cells from different levels of irradiated regenerating
limbs (series 1)
Zones
Animal No.
1
2
3a
4
3b
5
6
Number of days
after the operation
I
II
III
IV
42
112
114
141
144
143
200
2X
2X
2X
2X
2X
2X
2X
3X
3X
2X
2X
2X
2X
2X
3X
3X
3X
2X + 3X
2X
2X
2X
3X
3X
3X
3X
3X
2X
2X
The analysed samples originated from four zones: (I) flank; (II) triploid skin cuff; (III)
irradiated skin between cuff and regenerate; (IV) regenerate. The animals were classed
according to the decreasing area of the limb covered with triploid epidermis. 2X'. diploid;
3x: triploid.
samples taken from four different regions: (I) the flank; (II) the triploid skin cuff;
(III) the irradiated skin and (IV) the regenerating hand (Fig. 1). Six animals were
used in the study. The epidermis of each animal was analysed at a different time,
ranging from 42 to 200 days after amputation. The epidermis of one animal
(animal number 3 in Table I) was analysed at two different times. From the DNA
content measurements of epidermal cell nuclei in the four regions defined above,
histograms were drawn. All histograms but one were unimodal, comprising
either diploid or triploid cells. Only one comprised a mixed population of diploid
and triploid cells. Table I represents an analysis of the histograms. The first two
animals exhibited a regenerating limb entirely covered with triploid epidermis
(Fig. 4,42 days). This result is explained by a migration of the triploid epidermis
which replaced the irradiated diploid epidermis. The third animal showed
triploid epidermis in the distal zones III and IV but not at the level of the triploid
skin graft as was expected. On acccount of this, a second analysis was carried out
30 days later (Fig. 4, 114 and 144 days). The results of the two analyses of the
third animal as well as the results of animals 4,5 and 6 demonstrate that the limb
epidermis slowly migrated as a whole and that in the end the diploid epidermis
from the flank replaced the triploid epidermis of the graft after a period of five
months.
Fig. 5. Main results of series 2. The histograms drawn from the measurements of the
DNA content of epidermal cells show a fast migration of the healthy triploid epidermis which completely replace the irradiated diploid epidermis 3 weeks after the
X-irradiation. The analyses of the epidermis of four animals illustrate the progress
of triploid epidermis migration. As shown in zone III 13 days and 21 days after the
irradiation, the migrating epidermis exhibits many cells in S or G2. m\\ mean diploid
value, mi\ triploid value.
Control
diploid epidermis
10
20
Epidermis from grafted
triploid skin cuff
Epidermis from
distal upper arm
Epidermis
from hand
Percentage
of cell nuclei 0
I
II
III
IV
10
10
10
10
DNA content
3
o
3a-,
S
3
Si.
3
3
».
CTQ
226
E. LHEUREUX
Table 2. Ploidy of epidermal cells from different levels of irradiated limbs (series 2)
Zones
Animal No.
1
2
3
4
5
6
7
8
9
10
Number of days
after irradiation
I
II
III
IV
4
5
5
6
7
11
13
14
21
22
2X
2X
2X
2X
2X
2X
2X
2X
2X
2X
3X
3X
3X
3X
3X
3X
3X
3X
3X
3X
2X
2X + 3X
2X + 3 X
2X + 3X
2X + 3X
2X + 3X
2X + 3X
2X + 3X
3X
3X
2X
2X
2X
2X
2X
2X + 3X
2X + 3X
2X + 2X
3X
3X
The analysed samples originated from four zones: (I) hindlimb; (II) triploid skin cuff; (III)
irradiated skin of distal upper arm; (IV) irradiated skin of hand. 2\: diploid; 3x: triploid.
Series 2
This series was undertaken to follow the successive stages of replacement of
irradiated epidermis by healthy epidermis. Four zones of epidermis were simultaneously analysed: (I) control diploid epidermis from hindlimb; (II) epidermis
from triploid skin graft; (III) epidermis from distal stylopodium and (IV) epidermis from the hand. Histograms drawn from measurements of cell nuclei DNA
content of every epidermal sample showed whether the epidermis was diploid,
triploid or chimaeric. Results are shown in Table 2. A different animal was
analysed at each timepoint. Fig. 5 consists of histograms of DNA content
distribution which illustrate the main stages of migration of the triploid epidermis
and correlative disappearance of the irradiated epidermis. Such a migration is
rapid. Triploid epidermal cells are observed to have migrated to the distal part
of the stylopodium on the fifth day after irradiation (Table 2). A sample is
considered to correspond to a mixed population of diploid and triploid cells if
5 % or more of the cell nuclei have a DNA content reaching 3C value as allowed
by results of series 3 (see below). Triploid epidermis reaches the hand of the
irradiated limb 11 days after the operation. At this point, the cell populations at
the level of the distal stylopolium are still mixed. The percentage of diploid and
Fig. 6. Changes of percentage of cells in S and G2 in the irradiated epidermis. The
epidermis of the entire body is irradiated. No difference in the percentage of cells in
S + G2 is observed between control non-irradiated epidermis (2-4 %) and irradiated
epidermis during the two weeks following X-irradiation. However, the percentage
of irradiated epidermal cells in S + G2 increases to 10 %, 19 days after irradiation.
(C) Control: irradiation and no amputation; (1) to (5): irradiated limb epidermis; (1)
4 days; (2): 7 days; (3): 12 days; (4): 19 days; (5): 23 days.
Epidermal migration and limb regeneration
<J
E
227
228
E. LHEUREUX
triploid cells are different at distal and proximal levels of the irradiated limb (Fig.
5,13 days). These observations show that the progress of triploid epidermis and
the elimination of irradiated diploid epidermal cells are not synchronous
phenomena. Irradiated epidermal cells were probably eliminated by shedding
which occurs once a week in Pleurodeles.
Moreover, many migrating triploid epidermal cells were in S or G2 and many
mitoses could be observed in a few squashes. Healthy epidermal cells divided
intensively until they covered all the limb. After three weeks the non-irradiated
epidermis which had migrated from the triploid skin cuff had replaced the diploid
irradiated epidermis (Fig. 5, 21 days).
Series 3
The animals which were entirely irradiated ceased to eat and died 25 to 35 days
after X-irradiation. To determine the percentage of irradiated epidermal cells in
S or G2 a square of forelimb skin, 2 or 3 mm2, was removed at regular intervals
of time. It was observed that the ability of epidermis to migrate over the wound
surface was progressively reduced as the time after irradiation increased. One
night is sufficient for epidermis of a normal limb to cover such a wound. On
irradiated limbs, wound healing lasted more than one day if the removal of skin
occurred seven days after irradiation, and more than two days if removal of skin
occurred 14 days after irradiation. Not only was the migration of epidermis
inhibited by X-rays but also irradiated epidermis adhered less and less well to the
underlying dermis. Usually epidermis is separated from dermis after pieces of
normal skin are soaked for 20min in EDTA. On the contrary, the skin of an
animal which had been irradiated two weeks earlier needed to be soaked for only
3min in EDTA to separate epidermis from dermis.
The percentage of epidermal cells in S and G2 in entirely irradiated animals
was estimated from histograms drawn from DNA content measurements of
epidermis samples. Fig. 6 shows DNA content distribution in non-irradiated
epidermis and in epidermis analysed 4 days, 7 days, 10 days, 14 days, 19 days,
and 23 days after irradiation. 96 to 98 % of the cells of the epidermis of a nonirradiated and non-amputated limb were in GO or Gl. The percentage of cells in
S and G2 was therefore equal to 2 to 4 %. These percentages were also observed
in irradiated cell populations at least during the two weeks following X-irradiation of the animals. The percentage of cells in S and G2 slightly increased in
epidermis 19 and 23 days after irradiation.
Series 4
The animals used as recipients were entirely irradiated. A few animals lived
for four or five weeks. This was sufficient time to observe a small regenerate.
Thirty operated limbs exhibited blastemata at the stages of cone or paddle, but
only five of them developed a miniature limb with four individualized digits (Fig.
Epidermal migration and limb regeneration
229
lmm
Fig. 7. Results of series 4. Two regenerating forelimbs were produced by two newts
which had been entirely irradiated. The skin cuff from the limb of a normal animal
was transplanted to the irradiated limb stump. This graft was the only non-irradiated
tissue of the experimental animal, and allowed a miniature forelimb to regenerate.
G: skin graft.
Series 5
Wound healing occurred normally after amputation of the operated limbs but
no blastema was observed. Moreover, in a few cases, the cut end of the humerus
protruded through the soft tissues of the stump three weeks after the operation.
These limbs underwent an extended regression whereas the operated limbs of
the animals in series 4 showed well-developed blastemata.
DISCUSSION
The major findings in the present study are the fast replacement of irradiated
epidermis by the nearest non-irradiated epidermis and the need of a healthy
epidermis for regeneration. Several questions arising from these results will be
discussed. They concern the fate of irradiated epidermis, the origin of the epidermis which participates in replacement of irradiated epidermis, the slow migration of non-irradiated epidermis and the basic role of healthy epidermis in
regeneration.
1) Fate of irradiated epidermis
Unlike irradiated mesodermal tissues, irradiated epidermis of a limb disappears and is replaced by migrating non-irradiated epidermis. Because amputation or reamputation of an irradiated limb several months after irradiation does
not allow such a limb to regenerate (Lheureux, 1915b) it is assumed that no
migration of healthy mesenchymatous cells occurs. This study shows that nonirradiated epidermis migrates and replaces irradiated epidermis along the length
230
E. LHEUREUX
of the limb in three weeks. Brunst (1963) observed a replacement of skin
epithelium of the head of the axolotl which had been X-irradiated with doses
ranging from 500 rads to 3000 rads, but after irradiation with 6000 rads, because
the animals lived only 29 days, the replacement of irradiated epidermis was not
observed.
As shown in series 2, the disappearance of the irradiated epidermis from the
limb is not synchronous with the distal progression of non-irradiated epidermis
(Fig. 5). It is likely that the healthy epidermis insinuates itself between irradiated
dermis and irradiated epidermis while establishing a more or less regular front
of migration (Fig. 8). This idea may be explained by a differential adhesivity of
the two types of epidermis to the dermis. It was observed in series 3 that
irradiated epidermis was more and more easily separated from dermis as time
passed. This suggests that the damages caused by X-irradiation result in a loss of
adhesivity to the dermis by the epidermis. However this damage is not immediate; the results of series 2 show that the migration of triploid epidermis does
not begin until 5 days after the operation. This absence of migration during the
first few days after the operation may correspond to the maintenance of equal
adhesivities of each epidermis to the dermis. As the adhesivity of the irradiated
epidermis to the underlying dermis weakens, the migration of the triploid epidermis can begin. The irradiated epidermal cells can remain over the migrating
triploid epidermis until shedding eliminates them as a whole.
\ZJ
[\
Fig. 8. The two types of epidermis migration over an irradiated limb demonstrated
by the use of non-irradiated triploid epidermis. (A) Irradiation of the limb of a
diploid animal. (B) A triploid skin graft replaces a proximal cuff of irradiated skin.
(C), (D), (E) Stages of distal migration of the triploid epidermis which replaced the
diploid irradiated epidermis three weeks after the irradiation. (F) Another migration
is observed several months after the irradiation. It is a slow distal migration of the
whole non-irradiated epidermis. In five months the non-irradiated triploid epidermis
was replaced by diploid epidermis from the shoulder.
Epidermal migration and limb regeneration
231
Although it is not known how X-irradiation causes this loss of adhesivity, it
appears to act at the level of interaction between dermis and epidermis. In the
above experiments where epidermis was separated from dermis, a loss of adhesivity between epidermal cells was not observed. The epidermis was always
removed as a single sheet. Moreover, Lassalle (personal communication) observed a normal number and structure of desmosomes in electron microscopic
studies of X-irradiated epidermis of Pleurodeles. It would be interesting to investigate this effect of X-irradiation on adhesion of epidermis to underlying
mesodermal tissues.
2) Origin of the epidermis participating in replacement of the irradiated epidermis
In the present study it was demonstrated that diploid irradiated epidermis was
entirely replaced by triploid epidermis which originated from a non-irradiated
triploid skin graft. The following observations suggest that only a narrow distal
part of the grafted triploid epidermis supplies the irradiated distal limb with
healthy epidermal cells. Firstly, healthy epidermis from the cuff began migrating
and at the same time began dividing intensively (Fig. 5: 13 days, zone III) to
produce daughter cells which also migrated and divided. Secondly, the triploid
epidermal cells of the cuff divided less intensively than the migrating triploid
cells. By taking into account these observations, it is possible to explain the
abnormal increase of mitosis observed in the epidermis of an irradiated axolotl
limb (Maden & Wallace, 1976) seven weeks after the irradiation, by migration
of healthy epidermis from the shoulder.
3) Slow migration of non-irradiated epidermis
As a whole the epidermis slides distally at the rate of lcm per five or six
months. During the slow migration of epidermis resulting in the elimination of
triploid epidermis at the tip of the limb, the boundary between diploid and
triploid cells seems to remain well defined as shown by the paucity of mixed
populations found in the epidermal samples in the first series (Table 1).
Immunological rejection was never observed when homografts were transplanted to irradiated recipients. On the contrary, it was not possible to follow the
fate of triploid epidermis when grafted on a non-irradiated limb because the graft
was rejected six or eight weeks later. However, additional experiments must be
done because in a few cases homografts of Pleurodeles tissues were not
eliminated. Nevertheless it is likely that the slow migration of the limb epidermis
is a normal phenomenon.
4) Epidermis and regeneration
The results of both series 4 and 5 demonstrated that a non-irradiated epidermis
was necessary for regeneration. When the animals used as recipients were entirely irradiated, a non-irradiated skin cuff was effective in inducing blastema formation and growth whereas non-irradiated muscle fragments alone were not. These
232
E. LHEUREUX
results may be compared with results of experiments consisting of transplantation of skin cuff or muscle fragments to irradiated limb stumps of newts, the rest
of whose body had been protected from X-irradiation (Lheureux, 1975b). In
these experiments, both skin grafts and muscle grafts were effective in inducing
regeneration of irradiated limb stumps. Muscle grafts did not bring healthy
epidermis but the latter probably migrated from the flank which had been protected from X rays. This possibility was excluded in series 5 of this report where the
animals were completely irradiated. The results further confirm that epidermis
is not simply a common protective epithelium but is also an active component in
blastema formation as has been shown by other investigators (see Introduction).
Histological studies of irradiated limb stumps provided with grafts of various
combinations of non-irradiated muscle dermis and epidermis are in progress with
the goal of determining what phases of regeneration, if any, are able to occur in
the presence of an irradiated epidermis.
5) Positional values in newt epidermis
The results of the present study along with earlier investigations demonstrated
that the limb epidermis may be replaced by ectopic epidermis without affecting
the morphogenesis of the regenerating limb. Belly-skin grafts (Taube, 1921),
flank-skin grafts (Droin, 1959; Lheureux, 1975a) or head-skin grafts (Thornton,
1962) do not prevent a limb stump from regenerating. Even when flank-skin
grafts inhibit the limb regeneration, the epidermis of the flank is not responsible
for this inhibition (Tank, 1983). Slices of limb devoid of skin transplanted to the
flank were covered with epidermis of flank and gave rise to a limb (Umanski &
Kudokotsev, 1949). Since epidermis can originate from any part of the newt, all
epidermal cells must be specified by the same positional value. This is consistent
with experimental data demonstrating that epidermis is a passive component in
terms of pattern formation (Carlson, 1975; Lheureux, 1976). However epidermis
would have a share in pattern regulation by setting the distal boundary value in
a growing blastema (Faber, 1976; Maden, 1977). In terms of positional values,
the role of the wound epidermis could be to transmit a message to the mesodermal tissues of the stump. According to Maden (1977) this message would result
from a difference between positional values of epidermis and mesodermal
tissues, and would be responsible for releasing dedifferentiation of mesodermal
tissues and setting positional values of distal tissue cells in the blastema through
successive mitoses. X-irradiation does not suppress positional values of tissues
in Drosophila imaginal discs (Adler & Bryant, 1977) in amphibian limb tissues
(Maden, 1979; Holder, Bryant & Tank, 1979) and in chick wing bud tissues
(Smith, Tickle & Wolpert, 1978). If the boundary value of the wound epidermis
is destroyed by X-rays, the release of dedifferentiation and the formation of
blastema could not occur according to the averaging hypothesis of Maden (1977).
This destruction of positional values would be a unique case, however, since
other tissues appear to keep their positional values after X-irradiation. It is likely
Epidermal migration and limb regeneration
233
that irradiated epidermis retains its positional value and that X-irradiation acts
upon some other factor equally indispensable for blastema formation.
In conclusion, it is clear that irradiated epidermis must be replaced either by
non-irradiated skin grafts or by migration of healthy epidermis from skin that has
been protected from X-irradiation in order to take part, together with healthy
mesodermal tissues, in blastema formation, growth and pattern regulation.
I thank Frederick Carey for improving the English of this paper.
REFERENCES
P. N. & BRYANT, P. J. (1977). Participation of lethally irradiated imaginal disc tissue
in pattern regulation in Drosophila. Devl Biol. 60, 298-304.
BEETSCHEN, J. C. (1960). Recherches sur I'h6t6roploidie exp6rimentale chez un Amphibien
Urodele Pleurodeles waltlii Michah. Bull. Biol. Fr. Belg. 96,12-127.
BRUNST, V. V. (1963). The successive changes in the skin epithelium of the young axolotl
(Siredon mexicanum) after Roentgen irradiation. Am. J. Roentg. 89, 624-647.
BUTLER, E. G. (1933). The effects of X irradiation on the regeneration of the forelimbs of
Amblystoma larvae. /. exp. Zool. 65, 271-315.
CARLSON, B. M. (1974). Morphogenetic interactions between rotated skin cuffs and underlying stump tissue in regenerating axolotl forelimbs. Devi Biol. 39, 263-285.
CARLSON, B. M. (1975). The effect of rotation and positional change of stump tissues upon
morphogenesis of the regenerating axolotl limb. Devi Biol. 47, 269-291.
DROIN, A. (1959). Potentiality morphogenes dans la peau de triton en r6g6n6ration. Rev.
Suisse Zool. 66, 641-709.
FABER, J. (1976). Positional information in amphibian limb. Ada biotheretica 25, 44-65.
GLOBUS, M., VETHAMANY-GLOBUS, S. & LEE, Y. C. I. (1980). Effect of apical epidermal cap
on mi to tic cycle and cartilage differentiation in regeneration blastemata in the newt,
Notophthalmus viridescens. Devi Biol. 75, 358-372.
HOLDER, N., BRYANT, S. V. &TANK, P. W. (1979). Interactions between irradiated and unirradiated tissues during supernumerary limb formation in the newt. 7. exp. Zool. 208,303-310.
LHEUREUX, E. (1975a). Nouvelles donn6es sur le role de la peau et des tissus internes dans la
r6g6n6ration du membre du triton Pleurodeles waltlii Michah (Amphibien Urodele). Wilhelm Roux Arch, devl Biol. 176, 285-301.
LHEUREUX, E. (19756). R6g6n6ration des membres irradi£s de Pleurodeles waltlii Michah
(Urodele). Influence des qualit6s et orientations des greffons non irradi6s. Wilhelm Roux
Arch, devl Biol. 176, 303-327.
LHEUREUX, E. (1976). Roles du derme et de l'6piderme dans la regeneration du membre du
triton Pleurodeles waltlii. Bull. Soc. Zool. Fr. 101, (suppl. 3), 109-118.
LHEUREUX, E. (1983). The origin of tissues in the X-irradiated regenerating limb of the newt
Pleurodeles waltlii. In Limb Development and Regeneration, (eds J. F. Fallon & A. I.
Caplan). Part A, pp. 455-465. New York: Alan R. Liss.
MADEN, M. (1977). The regeneration of positional information in the Amphibian limb. /.
theor. Biol. 69, 735-753.
MADEN, M. (1979). Neurotrophic and X-ray blocks in the blastemal cell cycle. 7. Embryol.
exp. Morph. 50, 169-173.
MADEN, M. (1979). The role of irradiated tissue during pattern formation in the regenerating
limb. /. Embryol. exp. Morph. 50, 235-242.
MADEN, M. & WALLACE, H. (1976). How X-rays inhibit Amphibian limb regeneration. J. exp.
Zool. 197, 105-114.
MICHAEL, M. I. & FABER, J. (1971). Morphogenesis of mesenchyme from regeneration
blastemas in the absence of digit formation in Ambystoma mexicanum. Wilhelm Roux Arch.
EntwMech. Org. 168, 174-180.
ADLER,
234
E.
LHEUREUX
M. (1974). The inheritance of cell differentiation during limb regeneration in
the axolotl. Devi Biol. 41, 42-56.
NORMAN, W. P. & SCHMIDT, A. J. (1967). The fine structures of tissues in the amputated
regenerating limb of the adult newt Diemictylus viridescens. J. Morph. 123, 271-312.
RAHMANI, T. & KIORTSIS, V. (1961). Le role de la peau et des tissus profonds dans la re"g6n6ration de la patte. Rev. Suisse Zool. 68, 91-102.
REPESH, L. A. & OBERPRILLER, J. C. (1978). Scanning electron microscopy of epidermal cell
migration in wound healing during limb regeneration in the adult newt, Notophthalmus
viridescens. Am. J. Anat. 151, 539-555.
REPESH, L. A. & OBERPRILLER, J. C. (1980). Ultrastructural studies on migrating epidermal
cells during the wound healing stage of regeneration in the adult newt, Notophthalmus
viridescens. Am. J. Anat. 159, 187-208.
ROSE, F. C. & ROSE, S. M. (1965). The role of normal epidermis in recovery of regenerative
ability in X-rayed limbs of Triturus. Growth 29, 361-393.
SINGER, M. & SALPETER, M. M. (1961). Regeneration in Vertebrates: the role of the wound
epithelium. In Growth in Living Systems, (ed. M. X. Zarrow), pp. 277-311. New York:
Basic Books.
SMITH, J. C , TICKLE, C. & WOLPERT, L. (1978). Attenuation of positional signalling in the
chick limb by high doses of y radiation. Nature 272, 612-613.
STOCUM, D. L. (1968). The Urodele limb regeneration blastema. A Self-organizing system. II.
Morphogenesis and differentiation of autografted whole and fractional blastemas. Devi
Biol. 18, 457-480.
STOCUM, D. L. &DEARLOVE, G. E. (1972). Epidermal mesodermal interaction during morphogenesis of the limb regeneration blastema in larval Salamanders. /. exp. Zool. 181, 49-62.
TANK, P. W. (1983). Skin of non-limb origin blocks regeneration of the newt forelimb. In Limb
Development and Regeneration, (eds J. F. Fallon & A. I. Caplan). Part A, pp. 565-575.
TAUBE, E. (1921). Regeneration mit Beteiligung ortsfremder Haut bei Tritonen. Arch. f.
Entw. Mech 49, 269-315.
THORNTON, C. S. (1942). Studies on the origin of the regeneration blastema in Triturus
viridescens. J. exp. Zool. 89, 375-389.
THORNTON, C. S. (1957). The effect of apical cap removal on limb regeneration in Amblystoma
larvae. /. exp. Zool. 134, 357-382.
THORNTON, C. S. (1962). Influence of head skin on limb regeneration in Urodele Amphibians.
/. exp. Zool. 150, 5-15.
TRAMPUSCH, H. A. L. (1951). Regeneration inhibited by X-rays and its recovery. Proc. Kon.
ned. Akad. Wet. 54, 373-385.
UMANSKI, E. (1937). An investigation of the regenerative process in amphibians by the
method of exclusion of individual tissues from Roentgen beams (in Russian). Biol. Zh. 6,
739-758.
UMANSKI, E. & KUDOKOTSEV, V. P. (1949). On regional peculiarities of the skin epithelium
on limb regeneration in the axolotl (in Russian). Dokl. Akad. Nauk. SSSR. 69, 265-268.
WERTZ, R. L., DONALDSON, D. J. & MASON, M. J. (1976). X-ray induced inhibition of DNA
synthesis and mitosis in internal tissues during the initiation of limb regeneration in the adult
newt. J. exp. Zool. 198, 253-259.
NAMENWIRTH,
(Accepted 22 March 1983)