./ Embryol. exp. Morph. Vol. 39, pp. 23-43, 1977
Printed in Great Britain
23
Redifferentiation, cellular elongation and the cell
surface during lens regeneration
By SHARON HORNSBY 1 AND SARA E. ZALIK 1
From the Department of Zoology, University of Alberta, Edmonton
SUMMARY
(1) Dorsal irises from normal and lentectomized eyes of the newt Notophthalmus viridescens
were cultured in vitro in the presence of colchicine (1-25 x 10~3 to 2-5 x 10~6 M), vinblastine
sulfate (4-4xlO~8) and cytochalasin B (20/tg/ml). Explants were treated for 24 h and
implanted into host eyes of previously lentectomized newts.
(2) In regenerates treated with colchicine at stages prior to the onset of lens fiber formation,
cellular elongation was inhibited or considerably delayed. Cytochalasin B has no effect on
lens fiber differentiation when compared with regenerates treated with its solvent, dimethyl
sulfoxide.
(3) Colchicine-treated regenerates showed no detectable fluorescence when stained with
anti-newt lens fluorescent antibodies, suggesting that cells in which cellular elongation was
inhibited were unable to synthesize lens specific proteins.
(4) Ultrastructural observations of elongating lens regenerate cells showed microtubules
present in the cytoplasm parallel to the axis of elongation. Microtubules were not observed
in depigmented cells from colchicine-treated regenerates. Characteristic vinblastine-induced
microtubule crystals were present in depigmented cells from vinblastine-treated regenerates.
(5) Experiments using cell electrophoresis showed that neuraminidase-sensitive groups
were detectable at the ionic double layer of cells from implanted regenerates at stages where
lens fiber differentiation occurs. Cells from colchicine-treated regenerates showed no significant reduction of their electrophoretic mobility after neuraminidase treatment. This
suggests that the appearance of some cell surface components in the redifferentiated cell may
also be colchicine sensitive and probably mediated by microtubules.
INTRODUCTION
Surgical removal of the lens from the eye of the newt Notophthalmus viridescens results in the initiation of a sequence of events that cause the cells of
the dorsal iris to increase their RNA synthetic activity, initiate DNA synthesis,
depigment, elongate, and become organized into a functional new lens. During
lens redifferentiation from the depigmented iris cells, the cuboidal depigmented
iris cells elongate into lens fibers. In the lens regenerating system, cellular
elongation occurs concurrently with a marked increase in protein synthesis
which has been associated with the synthesis of lens crystallins (Yamada &
Takata, 1963). Wolfiian lens regeneration therefore represents a system useful
1
Authors' address: Department of Zoology, University of Alberta, Edmonton, Alberta,
Canada.
24
S. HORNSBY AND S. E. ZALIK
for the study of the relationship between cell elongation and cellular processes
involved in lens fiber differentiation.
It has been suggested that cytoplasmic microtubules have a role in the
acquisition of an asymmetrical cell shape (Wessels et ah 1971); these organelles
have also been associated with processes accompanying cellular elongation.
Cytoplasmic microtubules have been observed during the invagination and
accompanying cellular elongation of the lens placode in the chick embryo
(Byers & Porter, 1964). It has also been suggested (Piatigorsky, Webster &
Wolberg, 1972a) that de novo synthesis of microtubular protein is necessary for
the continued elongation of cultured chick lens epithelial cells; the initial
elongation of differentiating lens fibers involving the use of pools of microtubular protein already available in the cytoplasm. The involvement of microtubules in other morphogenetic processes involving changes in cell shape has
also been proposed (Arnold, 1966; Grandholm & Baker, 1970; Handel & Roth,
1971;Burnside, 1973).
In the present study, several compounds which have been reported to interfere
with microtubule or microfilament integrity have been tested at various stages
of lens regeneration. The effects of these compounds on cell elongation and on
other processes of lens fiber differentiation have been studied. The results
obtained so far indicate that microtubules may play a role in lens fiber elongation
and that this process may be associated with the tissue specific protein synthesis
which occurs during lens fiber differentiation. Our results also suggest that
microtubules play a role in the reappearance of certain cell surface components
during redifferentiation.
MATERIALS AND METHODS
Adult newts, Notophthalmus viridescens, were lentectomized according to the
method of Eisenberg & Yamada (1966). Prior to lentectomy the animals were
kept at 8 °C; after the operation they were kept at 22 °C and fed weekly.
In vitro treatment. At various intervals after lens removal dorsal irises were
obtained as described by Zalik & Scott (1969). Irises were cleaned from
adhering material and trimmed with iridectomy scissors so that only the middorsal portion of the iris remained. The irises were then placed in 50 % L-15
medium (Grand Island Biological Company) diluted with sterile distilled water
and containing 50 units Penicillin-streptomycin (Grand Island Biological
Company). This medium will be referred to as control medium. In other
experiments specific drugs were incorporated in the control medium in order to
test their effects on irises at various stages of regeneration; colchicine (Sigma)
ranging in concentrations from 1-25 xlO~ 3 to 2 - 5 X 1 0 ~ 6 M and vinblastine
sulfate (Lilly), used at concentrations ranging from 4-4 x 10~8 to 4-4 x 10~6 M,
were used as antimitotic microtubular disrupting drugs; lumicolchicine was
prepared according to the method of Wilson & Friedkin (1966) and used at
Cell elongation during lens regeneration
3
25
a concentration of 3-5 x 10~ M. The microfilament disrupting agent cytochalasin B (I.C.I. Laboratories, Cheshire, England) was dissolved in dimethyl
sulphoxide (DMSO) and added to the medium at a concentration of 20 jag /ml.
Controls for this treatment contained the appropriate amount of DMSO.
Usually 10-20 irises were placed in 5 ml of medium in a Petri dish 60 mm in
diameter and incubated at 22 °C for 24 h.
In vivo culture. After in vitro incubation irises were rinsed in control medium
and implanted into the eyes of newts lentectomized immediately prior to the
insertion of the tissue; one iris was implanted in each lentectomized eye. One to
twenty-five days after the iris implantation, host newts were anesthetized,
sacrificed and eyes with the implanted irises were fixed and processed for
microscopical examination. Light microscopy was carried out on host eyes
fixed in Bouin's fixative, wax embedded and stained with hematoxylhveosin.
For electron microscopic observations, 15-day regenerates cultured in vitro for
24 h in either control or drug-containing media were fixed in 3 % glutaraldehyde
in 0-1 M phosphate buffer pH 7-4 followed by 2 % osmium tetroxide in the same
buffer. The material was dehydrated in graded ethanol solutions embedded in
Epon, sections stained with uranyl acetate and lead citrate and observed with
a Philips 200 electron microscope.
For immunofluorescent studies newt lens homogenates (five lenses/ml) were
prepared. Rabbits were injected three times at 2-week intervals with 2 ml of
lens homogenate mixed 1:1, v/v, with Freund's adjuvant. Two weeks after the
last injection animals were bled and the antisera were obtained. Before use the
antiserum was absorbed with newt tissue powder obtained from lentectomized
newts. Reaction of the antiserum with newt lens extract after immunoelectrophoresis revealed the presence of a, /? and y crystallins. Tissues werefixedin 95 %
alcohol at 4 °C and processed according to Sainte-Marie (1962). Deparaffinized
sections of control and drug-treated regenerates were treated first with lens
antiserum, washed and then exposed to fluorescein-labelled goat anti-rabbit
immunoglobulin (Microbiological associates). Controls consisted of lens
regenerates exposed to serum from a non-injected rabbit. Observations were
performed on a Leitz fluorescence microscope with the appropriate filters.
Some experiments were designed to investigate the effects of colchicine on the
incorporation of uridine or thymidine on lens regenerates. Fifteen-day regenerates were cultured in vitro in either control medium or in medium containing
colchicine at a concentration of 2-5 x 10~5 M. Irises were then implanted into
host newt eyes and animals were injected with either 10 jaCi of uridine
(5-[H3] 29-5 Ci/mM) or 30 /id thymidine (methyl-[H3] 16-9 Ci/mM, New England
Nuclear Corp.). Animals were sacrificed 12 h later and eyes were fixed. In
experiments designed to measure amino acid incorporation into protein,
10- and 15-day regenerates were cultured in either control or colchicine containing medium as described previously. Irises were then implanted into host
lentectomized newts and were allowed to develop in the host eyes for 12 h, 5,10,
26
S. HORNSBY AND S. E. ZALIK
15 and 20 days after implantation. Three hours before sacrifice, animals were
injected with 5 /A of D,L-leucine (4,5-[H3] 35-4 Ci/mM, New England Nuclear).
Tissues were fixed in Bouin's fixative, embedded and sectioned. Sections were
treated with cold 5 % TCA for 10 min, washed, and slides were covered with
Kodak NTB-2 liquid autoradiographic emulsion. Slides were exposed for
3 weeks, developed with D-ll, fixed for 5 min in Kodak fixer and stained with
hematoxylin-eosin. Slides with sections from [H3]thymidine- and [H3]uridinetreated regenerates were placed in 10 % H 2 O 2 overnight and washed
several times in distilled water before being exposed to the autoradiographic
emulsion.
In some experiments cell electrophoresis was performed on cell suspensions
from regenerate irises. Fourteen-day regenerates were treated for 24 h in vitro
either in control medium or in media containing 2-5 x 10~5 M colchicine.
Regenerates were implanted into lentectomized host newt eyes and cultured
in vivo for 1-8 days. At the desired time intervals the implanted irises were
removed from the host newt eyes and cell dissociation was accomplished by the
method of Zalik & Scott (1972). In some experiments aliquots of the cell
suspension were incubated in 0-118 M-NaCl containing 25 u. /ml of neuraminidase
(from Vibrio cholera, Behringswerke) and 0-005 M-CaCl2 pH 7-2. Controls consisted of cells placed in saline without enzyme. Cell suspensions were incubated
at 37 °C for 30 min, washed and suspended in 0-118 M-NaCl pH 7-2; cell
electrophoresis was determined in a cylindrical cell electrophoresis apparatus
(Bangham, Flemans, Heard & Seaman, 1958), equipped with the micro cell
attachment.
RESULTS
The extent of development undergone by the cultured regenerates was determined by comparison with the stages proposed by Yamada (1967) according to
the following criteria: (a) ratio of pigmented to depigmented cells in the area
making up the regenerate; (b) degree of elongation and extent of nuclear
degeneration in primary lens fibers; (c) presence and ratio of secondary lens
fibers. In all of the experiments performed the staging was done blind.
Retardation due to experimental manipulation. Intrinsic retardation. The
methods chosen for in vitro drug treatment and in vitro culture offer certain
advantages. They allow for a better control of the time and drug concentration
to which the regenerate is exposed, and they eliminate drug effects on the
experimental animals which in turn may influence the process under investigation. However, these conditions do have several drawbacks: the iris tissue is
subject to a certain degree of injury imposed by the manipulation of the tissue,
and the iris, although implanted into the optic chamber of a lentectomized newt,
may not be surrounded by the same conditions existing in the original
lentectomiz;ed eye at the developmental stage of the iris implant. A question
arising in these experiments is the extent to which our experimental conditions
Cell elongation during lens regeneration
27
Table 1. Intrinsic retardation experienced by control iris regenerates
implanted into lentectomized eyes
Days
Actual f
postage
lentectomy
0
2
5
6
7
10
15
20
16
18
21
22
23
26
31
36
Stages
IV-VI
IV-VI
V-VI
VI-VIII
vii-vin
VII-IX
VIII-IX
X
Apparent ageAverage
Intrinsic
apparent agej retardation
9-16—14
9-16—14
12-16—15
12-19—16
15-19—17
15-20—18
15-20—20
.18-25—22
2
4
6
6
6
8
11
14
t Actual age represents the sum of the days post-lentectomy, one day for the in vitro
culture and 15 days for the in vivo culture.
% Apparent ages were obtained by comparing the observed stages with those described by
Yamada (1967). The average age in days for each regenerate was obtained and these values
were used to calculate the average apparent age for each of the groups studied. Therefore, the
average apparent age is not necessarily the numerical mean of the apparent age. These values
as well as those for intrinsic retardation represent only estimates and no further quantitative
assessment of the data is intended.
influence the progress of lens regeneration. To obtain an estimate of this
parameter, sections of each implanted regenerate were examined under the light
microscope and assigned a stage (Yamada, 1967) according to the differentiation
it had attained. Although a considerable variation in developmental stages
occurs within time intervals after lentectomy, an approximate estimate of the
average time at which a certain regeneration stage may occur can be obtained.
Each implanted iris was assigned an 'apparent age' in days, by this procedure.
The apparent age of the iris was then compared to its 'actual age'. The actual
age of the regenerate was calculated as the sum of the numbers of days after
lentectomy, plus one day for in vitro culture and the number of days it was
cultured in vivo. The intrinsic retardation was obtained by subtracting the
apparent age from the actual age of the implanted irises. Table 1 summarizes
the intrinsic retardation experienced by irises at succeeding stages of regeneration cultured for 24 h in vitro and subsequently implanted into host newt eyes
for 15 days. For normal irises as well as for regenerates at 2, 5, 6, and 7 days
after lentectomy, ten explants were used while, for 10-, 15-, and 20-day regenerates, 20 explants were examined, in these studies. It can be seen, from the data
in Table 1, that the intrinsic retardation shown by the implants increases with
the age of the regenerate. This would seem to indicate that irises at later stages
of regeneration are more sensitive to experimental manipulation and surrounding environmental conditions in the ocular environment.
28
S. HORNSBY AND S. E. ZALIK
Table 2. Effect of colchicine concentration on lens regeneration
Treatment
Dose
Stage
Control
Colchicine
Colchicine
Colchicine
Colchicine
—
2-5X10-6M
2-5 x 10-5 M
2-5 x 10-4 M
VIII-X
VI-VII
l-25xlO"3M
V
V
III
Apparent ageAverage
apparent age
15-25—20
12-18—16
12-15—13
12-15—13
8-11—10
Retardation
11
15
18
18
21
Fifteen-day regenerate irises were treated in vitro for 24 h then incubated in vivo for
15 days; the actual age of regenerates was 31 days. Average apparent age was calculated as in
Table 1.
Effects of colchicine on lens regeneration
Colchicine-treated regenerates were evaluated and staged in the same manner
as the controls. The 'actual age' and the 'apparent age' were calculated and
a retardation time, in days, was estimated. The intrinsic retardation time of the
controls was subtracted from that experienced by the drug-treated regenerates
to obtain a retardation time which is presumed to be due to drug effects.
A series of experiments were first performed in order to investigate the
colchicine concentrations that would affect regeneration without being lethal
to the cells. To determine this parameter 15-day regenerates were used and the
effects of several colchicine concentrations on subsequent cellular elongation
was determined. For these experiments regenerates were cultured in the presence
of colchicine (2-5 x 10~6 to 1-25 x 10~3) for 24 h and incubated in vivo for 15 days.
The results of these experiments are presented in Table 2. From the data in this
table it can be observed that colchicine at all of the doses used retarded regeneration. At drug concentrations of 2-5 x 10~6 the regeneration process was retarded,
but some cellular elongation occurred as evidenced by the fact that regenerates
attained stage VI and VII, in which some cells have elongated into primary lens
fibers. Concentrations of colchicine of 2-5xlO~ 5 and 1-25 xlO~ 3 completely
inhibited elongation; regenerates of this series were composed of masses of
Fig. 1. Appearance of 15-day regenerates treated for 25 h in vitro in the presence or
absence of colchicine. (A) Stage X control iris cultured for 15 days in vivo; x250.
(B) Late stage VII control iris cultured for 15 days in vivo, x250. (C) Colchicinetreated iris cultured for 15 days in vivo; this regenerate was classified as early stage VI;
arrows points to elongating cells. (D) Colchicine-treated regenerate cultured for
15 days in vivo, composed by a mass of depigmented cells (arrows); x250. (E)
Colchicine-treated regenerate incubated for 15 days in vivo, regenerate is composed
of pigmented and depigmented cells; x 125. (F) Colchicine regenerate incubated for
25 days in vivo. This regenerate attained stage IX; primary (p) and some secondary
(s) lens fibers are evident. This was the most advanced colchicine-treated iris observed.
The host regenerate is also present, x 125.
29
Cell elongation during lens regeneration
EMB 39
30
S. HORNSBY AND S. E. ZALIK
Table 3. Retardation experienced by colchicine-treated 15-day regenerates
at different time intervals after implantation in host lentectomized eyes
In
vivo
incubation
Control
\ r*tt ISI 1
Age-Average age
(days)
Stages
5
10
15
20
25
V-VII
VII-IX
VII-X
X
X
12-18—15
15-20—18
15-25—20
18-25—22
18-25—22
21
26
31
36
41
Rlt
Stages
6
8
11
14
19
V
V
V-VI
VI-VII
VIII-IX
Age-Average R2J
12-15—12
12-15—12
12-16—14
12-18—17
15-20—19
9
13
17
19
22
Rl§
3
5
6
5
3
Regenerates were cultured in vitro in the presence of 2-5 x 10~5 M colchicine for 24 h and
subsequently implanted into host lentectomized newts. Age and Average refer to apparent
age and average apparent age respectively. For further information about constraints of these
values see Table 1.
t Retardation in days experienced by the control due to experimental procedures intrinsic retardation.
t R2, Retardation in days experienced by the colchicine-treated irises due to experimental
conditions and colchicine treatment.
§ R2 — Rl, Retardation due to the effect of colchicine as compared to controls.
Table 4. Effect of colchicine on various stages of lens regeneration
Age of iris
after
lentectomy
0
2
5
6
7
10
15
20
Total age
Stage of operated
control/av.
apparent age
16
18
21
22
23
26
31
36
IV-VI/14
IV-VI/14
VI/14
VI-VIII/17
VII-VIII/17
VII-X/18
VIII-X/19
X/22
Stage of
Colchicine
colchicine-treated av. retardation
apparent age
(days)
IV-VI/14
IV-VI/14
VI/14
V-VI/14
IV-V/13
V-VI/14
V-VI/14
VII-VIII/18
0
0
0
3
4
4
5
4
Irises were cultured in vitro in the presence of 2 X 1 0 ~ 5 M colchicine and subsequently
implanted in host lentectomized eyes for 15 days. See Table 1 for calculation of average
apparent age.
depigmented cells showing no evidence of elongation (Fig. 1). Under histological
examination no morphological evidence of cytoplasmic or nuclear damage was
observed. Moreover, when regenerates treated with these colchicine concentrations were cultured for 25 days in vivo, they were able to recover and form
lens fibers (Table 3). The histological appearance of control and colchicinetreated regenerates is presented in Fig. 1. For subsequent work a colchicine
Cell elongation during lens regeneration
31
B
D
Fig. 2. Effects of colchicine on 7- and 10-day regenerates cultured for one day in
vitro and 15 days in vivo. (A) Seven-day regenerate control late stage VI. (B) Colchicine-treated 7-day regenerate stage V cells at the onset of elongation are pointed at
(arrow). (C) Ten-day regenerate control; primary lens fibers are evident, stage VII.
(D) Colchicine-treated 10-day regenerate in which many depigmented cells are
evident. x250.
concentration of 2-5 x 10~5 M was selected since this was the lowest dose that
would inhibit cellular elongation under these conditions.
Another parameter to be investigated was the time of in vivo culture at which
the effects of colchicine on the treated regenerate would become evident. To
3"3
32
S. HORNSBY AND S. E. ZALIK
Fig. 3. Fifteen-day regenerates incubated for one day in vitro in the presence (B) or
in the absence (A) of colchicine, and subsequently implanted into host lentectomized
eyes. Sections were treated with lens antiserum and stained with fluorescein-labelled
goat antirabbit immunoglobulin. Control-treated regenerates are heavily stained
(A) while no detectablefluorescenceis observed in the cells of the colchicine-treated
regenerate, x 170, arrow points to the lens.
answer this question experiments were performed in which 15-day regenerates
were cultured for 24 h in the presence of 2-5 x 10~5 M colchicine and then
implanted in the eyes of host newts for 5, 10, 15, and 25 days. The results of
these experiments are presented in Table 3. Five to ten regenerates were studied
for each time interval. It can be observed that a retardation of the regeneration
process is already evident in regenerates cultured in vivo for 5 days after
colchicine treatment. The most pronounced retardation was observed in
regenerates cultured in vivo for 15 days after treatment; no evidence for lens
fiber differentiation was observed in these implants. That regenerates were
able to recover from the inhibition of regeneration was evident in irises maintained for 20 and 25 days in host eyes where the presence of primary and
secondary fibers was observed in implanted irises that had attained stages
VI-IX (Fig. 1).
To investigate whether other regeneration stages were also affected by
colchicine treatment, experiments were performed in which normal irises and
regenerates at 2, 5, 6, 7, 10, 15, and 20 days after lentectomy were used. Tissues
were cultured for 24 h in the presence of 2-5 x 10~5 M colchicine and implanted
in host newt eyes for 15 days. Five irises were examined for each period after
lentectomy and drug-induced retardation was estimated as before. The results
of these experiments are shown in Table 4. It can be observed that colchicine
at this concentration has no noticeable effect on the further development of
implants of normal dorsal irises and regenerates at 2 days after lentectomy.
Colchicine treatment of regenerates at 6-15 days after lens removal inhibits the
further morphogenesis of the implanted irises. The development of the regener-
Cell elongation during lens regeneration
33
ates is arrested at stages V or VI where onset of cellular elongation occurs. Regenerates treated at 20 days after lentectomy, when elongated primary lens fibers
are already present, also undergo a delay in development; however, all of the
regenerates in these series attained the morphogenesis of stages VII-VIII
characterized by presence of elongated acidophilic lens fiber cells. The scatter
for each group of regenerates varied in a random fashion; this was also the case
for the experiments mentioned in the previous section. The effects of colchicine
on 7-, 10-, and 15-day regenerates cultured for 15 days in vivo can be observed
in Figs. 1 and 2. Lumicolchicine treatment had no appreciable effect on the
development of implanted regenerates when compared to the controls.
Effects of colchicine on the presence of tens specific protein
In order to determine whether lens specific protein synthesis could occur in
the absence of cell elongation, fluorescent antibody studies were performed on
15-day regenerates cultured 1 day in vitro in the presence or absence of colchicine, and then cultured 15 days in vivo. Figure 3 shows the results obtained
when regenerates treated with absorbed anti-lens serum were stained with
fluorescein-conjugated antirabbit immunoglobulin. In control irises a bright
fluorescence was present in all of the lens fiber cells, while a slight fluorescence
occurred in the lens epithelium; although not as bright as that in the fiber cells
(Fig. 3 A). No fluorescence was present in other tissues of the host eye. Under
these conditions colchicine-treated irises showed no detectable fluorescence
when compared to controls (Fig. 3B). One or two minute spots of fluorescence
were present in a few of the internal cells of one of the colchicine-treated irises
studied. Four regenerates were examined for the control and three for the
colchicine treatments. Newt liver tissue, treated with lens antiserum and stained
with fluorescein-conjugated anti-rabbit immunoglobulin, showed no detectable
fluorescence. Tests with serum from uninjected rabbits showed no fluorescence
in control lens regenerates in situ.
Effects of colchicine on the incorporation of [H3]uridine, [H3]thymidine and
[H3]leucine
Colchicine has been reported to inhibit nucleoside incorporation in mammalian
systems (Hell & Cox, 1963; Creasey & Markiw, 1965). Experiments were conducted to investigate whether incorporation of uridine and thymidine was
affected in colchicine-treated regenerates. When autoradiographs of [H3]uridinetreated implants were examined, no significant differences in grain counts per
area were detected in control and colchicine-treated regenerates. Similarly the
synthetic index did not vary appreciably in control and colchicine-treated
implants as measured by [H3]thymidine incorporation.
In implanted regenerates labelled with [H3]leucine differences were observed
in autoradiographs of control and drug-treated tissues. Since increased protein
synthesis occurs first in a localized number of cells at the onset of lens fiber
34
S. HORNSBY AND S. E. ZALIK
>100 r-
x x
90-100 -
o
o
10-20 -
16
21
26
31
Age (days)
Fig. 4. Grain counts per area in control (x — x) and colchicine-treated implanted
regenerates (O
O) as a function of age after lentectomy. The counts were
tabulated from sections of the central portion of the regenerate. Six cells (three
with heavy label and three with light label) were counted for each section. The same
number for sections was used for each regenerate. Each point represents one
regenerate. Age (days) represents days after lens removal.
differentiation (Yamada & Takata, 1963), to obtain an estimate of the incorporation of [H3]leucine into protein, autoradiographs of control and colchicinetreated implants were investigated in the following manner. Grain counts were
performed in cells which were heavily labelled; similarly grain counts were
performed in cells which appeared weakly labelled. Counts were performed for
the same number of cells in each class, in randomly chosen sections. The sum of
averages of these counts was obtained for several control and colchicine-treated
implants. In Fig. 4 differences in incorporation of [H3]leucine between control
and colchicine-treated regenerate implants is presented. It can be seen that
colchicine-treated regenerates exhibit a lower incorporation of this amino acid
into protein as evidenced by the lower grain counts present in their autoradiographs. Similarly, the increase in average grain counts per area observed in
regenerates cultured for 5 days in vivo is delayed slightly in colchicine-treated
regenerates. This result suggests that although colchicine treatment affects
amino acid incorporation, it does not completely inhibit protein synthesis in
implanted regenerates.
Effects ofvinb las tine in implanted regenerates
Although considerable retardation was induced by vinblastine sulfate, we
regard studies on the effect of this alkaloid as inconclusive because of evidence
35
Cell elongation during lens regeneration
Table 5. The effect of cytochalasin B on various stages of lens regeneration
Age of
iris
postlentectoray
(days)
0
2
6
10
15
Retardation
Total
age
(days)
Stages of
control/average
apparent age
16
18
22
26
31
IV-VI/14
IV-VI/14
VI-VIII/17
VH-X/18
VII-X/20
Stages of treated/average
apparent age
Cytochalasin B
IV-VI/14
IV-VI/14
VI-VIII/17
VII-X/18
IX-X/20
DMSO
IV-VI/14
VI-VI/14
vi-vm/n
VIII-X/20
CytochalasinControl
B
DMSO
2
4
5
8
11
2
4
5
8
11
4
4
5
11
Irises were cultured in vitro in the presence of cytochalasin B (20 /tg/ml) for 24 h and subsequently
implanted into host lentectomized eyes for 15 days. See Table 1 for estimates of average apparent age.
of pronounced toxic effects exerted by this compound. At the lowest concentrations used in this study (4-4 x 10~7 M) pronounced damaging effects were
observed in cells of the implanted regenerates. Cellular outlines were irregular,
numerous pyknotic nuclei were evident and the organization of the regenerate
was disturbed. In a few cases a group of cells in the implanted irises appeared to
have recovered from the treatment, these cells appeared to have undergone
depigmentation.
Effects of cytochalasin B on lens regeneration
A series of experiments was conducted in order to investigate the effect of this
chemical on irises at increasing time intervals after lentectomy. Regenerates
were treated in vitro with cytochalasin B and maintained as implants for 15 days.
The results of these studies are presented in Table 5. The data show that under
the experimental conditions of this study there is no observable effect of cytochalasin B on lens regeneration.
Ultrastructural observations were performed in 15-day regenerates and were
confined to regions of the regenerates where cells involved in the process of
depigmentation or depigmenting cells at the onset of elongation were present.
In control regenerates, depigmented cells were associated rather loosely; they
appeared to send cellular processes which intertwined and came close to each
other. In some regions, cells were present in close apposition to one another and
desmosomes occurred. In agreement with earlier studies (Dumont & Yamada,
1972) the cytoplasm was rich in mitochondria, ribosomes and profiles of rough
endoplasmic reticulum; microfilaments were observed located immediately
below the cell surface. Microtubules were present in all areas of the thin layer
of cytoplasm surrounding the nucleus of the elongating cell. They were located
just below the cell surface, in the intermediate areas of the cytoplasm and close
to the nuclear membrane. Microtubules were oriented for the most part parallel
36
S. HORNSBY AND S. E. ZALIK
Cell elongation during lens regeneration
37
Table 6. Effect of neuraminidase on electrophoretic mobilities {[im\sec\Vlcm)
of 14-day regenerates cultured for different time intervals
Incubation
time
Saline
(S)
Neuraminidase
(N)
Difference
(S-N)
0a
2b
1-875
1-904
1-951
1-821
1-94
1-992
20
1-77
1-775
1-73
1-919
1-40
1-48
1-636
1-384
1-38
0100
0174
0032
0-421**
0-46**
0-356**
0-616**
0-39**
3
4
5
Fourteen-day regenerates were cultured for 24 h in vitro and subsequently implanted into
host lentectomized eyes. Cell suspensions were obtained by try psinization of irises and electrophoretic mobilities were determined: a, determined in cell suspension prior to culture;
b, EPM determined in cell suspensions 24 h after in vivo culture.
** When compared by a t test the differences were significant by a probability < 0-5 %.
Where probability values are missing the differences are not significant.
to the axis of elongation (Fig. 5), although in many sections some appeared to
run perpendicular to this axis. Sections of several colchicine-treated regenerates
were observed; although the cellular organization of regenerates was similar to
that of controls, no microtubules were observed in the cytoplasm of depigmented
cells while microfilaments were still visible in regions of the cytoplasm. Lattices
presumably composed of microtubule crystals were regularly observed in the
cytoplasm of depigmented cells of vinblastine-treated regenerates (Fig. 5).
These lattices were similar to those described by Wilson, Bryan, Ruby & Mazia
(1970) and identified as precipitated lattices of microtubular protein (Bensch &
Malawista, 1968; Bensch, Marantz, Wisniewski & Shelanski, 1969).
Cell electrophoresis
The purpose of these experiments was to determine whether colchicine
treatment would affect the appearance of certain cell surface groups in the
regenerating iris. Neuraminidase sensitive groups present in cell suspensions of
normal irises cannot be detected at the cell surface when dedifferentiation is
completed. These groups reappear as ^differentiation occurs, at 15 days after
lentectomy and later. It was of interest to determine whether colchicine treatment
Fig. 5. Electron micrographs of sections from 15-day regenerates. (A) Section through
an elongating cell from the central portion of the regenerates. Note the sub surface
microtubules (Smt) and a desmosome (D) between two elongating pigmented cells,
x 42000. (B) Section through a depigmented cell from a 15-day regenerate treated
with vinblastine sulfate for 24 h. A typical vinblastine-induced crystal (V) is
present; microfilaments (F) can also be discerned in the cytoplasm (x 80000).
38
S. HORNSBY AND S. E. ZALIK
Table 7. Effect of neuraminidase on the electrophoretic mobilities
) of colchicine-treated regenerates
Control
Age of
regenerate
8
13
14
15
16
21
Colchicine
K
K
r
Saline
Enzyme
Saline
Enzyme
1-50
1-553
1-49
1-824
1-532
1-622
1-994
1-526
1-806
1-556
1-549
1-65
1-07**
1-256*
1-34
1-376**
1-172**
1-178**
1-57**
1085**
1-343**
1-236***
1-492
1-526
1-76
1-69
1-53
1-649
1-962
1-478
1-776
1-546
1-68
1-592
1-113**
1-262*
1-55
1-707
1-55
1-621
1-776
1-340
1-534
1-452
1-20***
1-22***
1-25***
112**
Regenerates at different days after lentectomy were cultured in vitro in the absence (control)
or in the presence of colchicine (2 x 10~5 M). They were then implanted into host lentectomized
eyes for 4 days. After this interval implanted irises were removed and cell suspensions were
obtained. For each experiment aliquots were incubated in the presence or absence of neuraminidase (25 u./ml).
* Probability < 1 % or less; ** Probability < 0-5 % or less; *** Probability < 0-1 %
or less. Where probability values are missing the differences between saline and enzyme
treated cells are not significant.
would affect the appearance of neuraminidase sensitive groups, detectable by
cell electrophoresis during ^differentiation.
A first series of experiments were performed to determine the sequence of
appearance of neuraminidase sensitive groups in cultured iris implants. Fourteenday regenerates were cultured in vitro in control medium and the effects of
neuraminidase on the electrophoretic mobilities was determined in cell suspension of regenerates implanted into host newts for 1-4 days. From the data
presented in Table 6 it can be seen that surface groups sensitive to neuraminidase
treatment appear in cells of cultured regenerates 2-4 days after implantation
into host newt eyes.
Regenerates 8-21 days after lentectomy were then cultured in vitro in the
presence of colchicine and implanted into newt host eyes for 4 days. The results
presented in Table 7 indicate that enzyme sensitive groups are present in cells
from 14- and 15-day control regenerates; however, the electrophoretic mobilities
of cells from colchicine-treated regenerates are not affected by neuraminidase
treatment. The presence of enzyme sensitive groups in 8-day colchicine-treated
regenerates is to be expected, since these groups are still present in 7- and some
10-day regenerates during the normal course of regeneration (Zalik & Scott,
Cell elongation during lens regeneration
39
1973). The results obtained with 16- and 21-day regenerates suggest that
neuraminidase sensitive groups already present at the cell surface (Zalik &
Scott, 1972) were not affected by colchicine treatment.
DISCUSSION
Lentectomy stimulates the iris epithelial cells of the newt's eye to undergo
DNA synthesis (Eisenberg & Yamada, 1966; Reyer, 1971; Yamada & Roesel,
1969) and proliferate. Concomitantly with these processes, melanosomes disappear from the pigmented iris epithelium either by dilution during cell division
or by an active process on which the iris epithelial cells discharge the melanosomes
invested in portions of the cell membrane (Eguchi, 1964; Dumont & Yamada,
1972). After completion of this phase of dedifferentiation, some cells retreat
from the cell cycle, elongate, proceed to synthesize lens specific proteins and
transform into lens fiber cells (Yamada, 1967). One of the questions pertinent to
our studies is how many of the above mentioned processes are affected by the
colchicine treatment. Cell cycle parameter studies on the dedifferentiating iris
epithelium (Yamada, Roesel & Beauchamp, 1975) have indicated the presence
of two types of cells in the dorsal iris; those which transform into lens cells
after the complete loss of melanosomes and follow the pathway of conveision,
and those involved in the pathway of retrieval. The latter are represented by
cells who enter mitosis, undergo a partial loss of melanosomes, and subsequently
redifferentiate into pigmented iris cells (Yamada, 1976). Colchicine binds to
microtubular subunits (Borisy & Taylor, 1967a, b) and evidence suggests that
slow morphogenetic movements that involve assembly of these subunits tend to
be colchicine sensitive (Margulis, 1973). In our experiments, retardation induced
by this alkaloid was first observed in 6-day regenerates, and increased in 7- and
10-day regenerates. Mitotic figures first become evident in the iris epithelium
4-5 days after lentectomy, while in 7- and 10-day regenerates a large number of
mitotic cells occur (Eisenberg & Yamada, 1966; Yamada & Roesel, 1971). It is
conceivable that some of the colchicine effects observed at these stages were due
to its interference with mitosis, since it has been suggested that at least six cell
cycles are necessary for the iris melanocyte to transform into a lens fiber
(Yamada, 1976). The appearance of depigmented cells in 7- and 10-day colchicine-treated regenerates cultured for 15 days could be explained by extrusion
of melanosomes in cells in which cell division is halted. Colchicine effects on
15-day regenerates, however, cannot solely be explained by its interaction with
microtubular subunits during formation of the mitotic spindle, since at this
stage many cells in the regenerate have left the cell cycle and are beginning to
elongate to follow the pathway of conversion. A more likely possibility in this
case is that colchicine affects lens cell elongation by its interaction with microtubular subunits. That the effects of colchicine are not due to cell death is
indicated by the general healthy appeaiance of the affected regenerates and their
40
S. HORNSBY AND S. E. ZALIK
3
incorporation of [H ]uridine and [H3]thymidine into high molecular weight
compounds. The ability of colchicine-treated regenerates to recover from the
effects of this alkaloid after a suitable interval, argues against an irreversible
non-specific damage induced in this system.
A noticeable effect of colchicine on 15-day regenerates was the inhibition of
cellular elongation which occurs during lens fiber formation. The notion that
the inhibitory effects of this alkaloid are due to its interaction with the microtubular apparatus is supported by ultrastructural observations. Microtubules
normally oiiented along the axis of elongation, disappeared after colchicine
treatment (Hornsby, 1975) and characteristic vinblastine-induced crystals were
detected in non-pigmented cells of 15-day regenerates. Our results are in
agreement with those of other investigators (Piatigorsky et ah 1912a, b) studying
lens fiber formation in the cultured chick lens epithelium. The presence of lens
fibers in 20-day colchicine-treated regenerates suggests that once elongation has
occurred it cannot be reversed by colchicine, and in this respect our results are
in agreement with those of Pearce & Zwaan (1970) in the developing chick lens.
Yamada & Takata (1963) and Takata, Albright & Yamada (1964) have observed that onset of crystallin synthesis in the prospective primary lens fiber of
15-day regenerates coincides with a noticeable increase in incorporation of amino
acids into protein observable by autoradiography. Our data with implanted regenerates show that this secondary enhancement of protein synthesis is retarded
when compared to regenerates in situ. In colchicine-treated regenerates the increase in amino acid incorporation is more gradual and never reaches the magnitude of the increase for controls. When absorbed anti-lens serum was applied
to control and colchicine-treated regenerates, a bright fluorescence was observed
in the lens-cell aggregates formed by the implanted regenerates, while no detectable fluorescence was observed in colchicine-treated irises. An interpretation of
these results is that in the latter [H3]leucine was incorporated into proteins that
are not lens-specific, although the possibility that the level of lens-specific proteins
in colchicine-treated regenerates was not high enough to be detected by the
fluorescent antibody method also has to be considered. An attractive explanation of these results is that elongation and synthesis of tissue-specific protein
occur in concert and that inhibition of elongation resulted in the inhibition of
tissue-specific protein synthesis. A relationship between microtubules and orientation of tissue-specific proteins has been reported during myogenesis (Fishman,
1967; Warren, 1968). However, the possibility that colchicine inhibits tissue
specific protein synthesis directly cannot be discarded.
There is recent evidence suggesting that microtubules may play a role in the
modulation of receptors at the cell surface (Yahara & Edelman, 1972; de Petris,
1974; Oliver, Ukena & Berlin, 1974; Edelman, 1976). Previous work from this
laboratory has shown that surface components present in the pigmented iris
cell disappear during dedifferentiation and reappear at the onset of redifferentiation (Zalik & Scott, 1972, 1973; Zalik, Scott & Dimitrov, 1976). From the
Cell elongation during lens regeneration
41
data in Table 7 it can be concluded, that, while the cells in control implanted regenerates possess neuraminidase sensitive groups at their surfaces, these groups
are not detectable in cells of colchicine-treated regenerates. Although a variety of
factors may affect the detection of specific charged groups at the ionic double
layer by cell electrophoresis (see Cook & Stoddart, 1973) one possibility is that
the lack of effect of neuraminidase on the cell's EPM reflects the absence of
groups sensitive to this enzyme at the cell surface. A possible explanation for
this phenomenon is that a colchicine sensitive event presumably microtubulemediated may be involved in the appearance of certain groups at the cell surface.
An inhibition of cell aggregation by colchicine has been reported by several
investigators (Waddell, Robson & Edwards, 1974; Lackie, 1974), and it has been
suggested that microtubules may be involved in the insertion of new membrane
or in redistribution of surface groups involved in adhesion (Waddell et al. 1974;
Edwards, Campbell, Robson & Vicker, 1975). There is also evidence suggesting
that neuraminidase sensitive groups may be involved in adhesion (Vicker &
Edwards, 1972; Lloyd & Cook, 1974). The possibility exists that, during lens
fiber differentiation, cellular elongation and insertion of new adhesive sites at
the cell surface occur in synchrony and are microtubule mediated. However,
a direct effect of colchicine on the cell membrane (Wiinderlich, Miller & Speth,
1973; Furcht & Scott, 1975) cannot be excluded.
Cytochalasin B had no observable effect on elongation or morphogenesis
regardless of the developmental stage of the regenerate. However, microfilaments were present in elongating cells of regenerates; and, in other systems,
these structures have been shown to reappear rapidly after removal of this drug
(Wessells et al. 1971). Short-term experiments are needed before a role of microfilaments in cellular elongation is discarded in this system.
We thank the National Cancer Institute of Canada for their continuous support. This
paper is part of the thesis submitted by Sharon Hornsby in partial fulfillment of the requirements for a Ph.D. degree at the University of Alberta. Portions of this work were also supported by the National Research Council of Canada.
REFERENCES
J. M. (1966). On the occurrence of microtubules in the developing lens of the squid,
Loligo pealii. J. Ultrastruct. Res. 14, 534-539.
BANGHAM, A. D., FLEMANS, K., HEARD, D. H. & SEAMAN, G. Y. F. (1958). An apparatus
for microelectrophoresis of small particles. Nature, Lond. 182, 642-644.
BENSCH, K. G. & MALAWISTA, S. E. (1968). Microtubule crystals: a new biophysical phenomenon induced by Vinca alkaloids. Nature, Lond. 218, 1176-1177.
BENSCH, K. G., MARANTZ, R., WISNIEWSKI, H. & SHELANSKI, M. (1969). Induction in vitro
of microtubular crystals by Vinca alkaloids. Science, N. Y. 165, 495-496.
BORTSY, G. G. & TAYLOR, E. W. (1967 a). Mechanism of colchicine action. I. Binding of
colchicine-3H to cellular protein. /. Cell Biol. 34, 525-533.
BORISY, G. G. & TAYLOR, E. W. (19676). Mechanism of colchicine action. II. Colchicine
binding to sea urchin eggs and the mitotic apparatus. /. Cell Biol. 34, 338-348.
ARNOLD,
42
S. HORNSBY AND S. E. ZALIK
B. (1973). Microtubules and microfilaments in amphibian neurulation. Amer.
Zool. 13, 989-1006.
BYERS, B. & PORTER, K. R. (1964). Oriented microtubules in elongating cells of the developing lens rudiment after induction. Proc. natn. Acad. Sci., U.S.A. 52, 1091-1099.
COOK, G. M. W. & STODDART, R. W. (1973). Surface Carbohydrates in the Eukaryotic Cell.
London: Academic Press.
CREASEY, W. A. & MARKIW, M. E. (1965). Biochemical effects of Vinca alkaloids. III. The
synthesis of ribonucleic acid and the incorporation of amino acids in Ehrlich Ascites cells
in vitro. Biochim. Biophys. Ada 103, 635-645.
DE PETRIS, S. (1974). Inhibition and reversal of capping by cytochalasin B, vinblastine, and
colchicine. Nature, Loud. 250, 54-55.
DUMONT, J. N. & YAMADA, T. (1972). Dedifferentiation of iris epithelial cells. Devi Biol. 29,
385-401.
EDELMAN, G. M. (1976). Surface modulation in cell recognition and growth. Science, N.Y.
192, 218-226.
EDWARDS, J. G., CAMPBELL, J. A., ROBSON, R. T. & VICKER, M. G. (1975). Trypsinized
BHK-21 cells aggregate in the presence of metabolic inhibitors and in the absence of divalent
cations. /. Cell Sci. 19, 653-667.
EGUCHI, G. (1964). Electromicroscope studies on lens regeneration. II. Formation and growth
of the lens vesicle and differentiation of lens fibers. Embryologia 8, 247-287.
EISENBERG, S. & YAMADA, T. (1966). A study of DNA synthesis during the transformation of
the iris into lens in the lentectomized newt. J. exp. Zool. 162, 353-368.
FISHMAN, D. A. (1967). An electron microscope study of myofibril formation in embryonic
chick skeletal muscle. / . Cell Biol. 32, 557-571.
FURCHT, L. T. & SCOTT, R. E. (1975). Effect of vinblastine sulphate, colchicine and lumicolchicine on membrane organization of normal and transformed cells. Expl Cell Res. 96,
271-282.
GRANHOLM, N. H. & BAKER, J. R. (1970). Cytoplasmic microtubules and the mechanism of
avian gastrulation. Devi Biol. 23, 563-584.
HANDEL, M. A. & ROTH, L. E. (1971). Cell shape and morphology of the neural tube:
implications for microtubule function. Devi Biol. 25, 78-95.
HELL, E. & Cox, D. G. (1963). Effects of colchicine and colcemid on synthesis of deoxyribonucleic acid in the skin of the guinea-pig ear in vitro. Nature, Lond. 197, 287-288.
HORNSBY, S. (1975). Ph.D. thesis. University of Alberta.
LACKIE, J. M. (1974). The aggregation of rabbit polymorphonuclear leucocytes: effects of
antimitotic agents, cyclic nucleotides and methyl xanthines. /. Cell Sci. 16, 167-180.
LLOYD, C. W. & COOK, G. M. W. (1974). On the mechanism of the increased aggregation by
neuraminidase of 16 C malignant rat dermal fibroblasts in vitro. J. Cell Sci. 15, 575-590.
MARGULIS, L. (1973). Colchicine sensitive microtubules. Int. Rev. Cytol. (ed. G. W. Bourne,
J. F. Daniel!i & K. W. Jeon), 34, 333-361. Academic Press.
OLIVER, J. M., UKENA, T. E. & BERLIN, R. D. (1974). Effects of phagocytosis and colchicine
on the distribution of lectin-binding sites on cell surfaces. Proc. natn. Acad. Sci., U.S.A. 71,
394-398.
PEARCE, T. L. & ZWAAN, J. (1970). A light and electron microscope study of cell behavior and
microtubules in the embryonic chicken lens using colcemid. /. Embryol. exp. Morph. 23,
291-507.
PIATIGORSKY, J., WEBSTER, H. DEF. & WOLBERG, M. (1972a). Cell elongation in the cultured
embryonic chick lens epithelium with and without protein synthesis. Involvement of
microtubules. /. Cell Biol. 55, 82-92.
PIATIGORSKY, J., WEBSTER, H. DEF. & CRAIG, S. P. (19726). Protein synthesis and ultrastructure
during the formation of embryonic chick lens fibers in vivo and in vitro. Devi Biol. 27,
176-189.
REYER, R. W. (1971). DNA synthesis and the incorporation of labelled iris cells into the lens
during lens regeneration in adult newts. Devi Biol. 24, 533-558.
SAINTE-MARIE, G. (1962). A paraffin embedding technique for studies employing immunofluorescence. /. Histochem. Cytochem. 10, 250-256.
BURNSIDE,
Cell elongation during lens regeneration
43
TAKATA, C , ALBRIGHT, J.
F. & YAMADA, T. (1964). Lens antigens in a lens regenerating system
studied by the immunofluorescent technique. Devi Biol. 9, 385-397.
VICKER, M. & EDWARDS, J. G. (1972). The effect of neuraminidase on the aggregation of
BHK-21 cells and BHK-21 cells transformed with Polyoma virus. /. Cell Sci. 10, 756-768.
WADDELL, A. W., ROBSON, R. T. & EDWARDS, J. G. (1974). Colchicine and vinblastine inhibit
fibroblast aggregation. Nature, Lond. 248, 239-241.
WARREN, R. H. (1968). The effect of colchicine in myogenesis in vivo in Rana pipiens and
Rhodnius prolixus. J. Cell Biol. 39, 544-555.
WESSELS, N. K., SPOONER, B. S., ASH, J. F., BRADLEY, N. O., LUDUENA, M. A., TAYLOR,
E. L., WRENN, J. T. & YAMADA, K. M. (1971). Microfilaments in cellular and developmental processes. Science, N.Y. Ill, 135-143.
WILSON, L. & FRIEDKJN, M. (1966). The biochemical events of mitosis. I. Synthesis and
properties of colchicine labelled with tritium and its acetyl moiety. Biochemistry, N. Y. 5,
2463-2468.
WILSON, L. & FRIEDKIN, M. (1967). The biochemical events of mitosis. II. The in vivo and
in vitro binding of colchicine in grasshopper embryos and its possible relation to
inhibition of mitosis. Biochemistry, N. Y. 6, 3-26.
WILSON, L., BRYAN, J., RUBY, A. & MAZIA, D. (1970). Precipitation of proteins by vinblastine and calcium ions. Proc. natn. Acad. Sci., U.S.A. 66, 807-814.
WUNDERLICH, F., MILLER, R. & SPETH, V. (1973). Direct evidence for a colchicine-induced
impairment in the mobility of membrane components. Science, N.Y. 182, 1136-1138.
YAHARA, I. & EDELMAN, G. N. (1972). Restriction of the mobility of lymphocyte immunoglobulin receptors by concanavalin A. Proc. natn. Acad. Sci., U.S.A. 69, 608-612.
YAMADA, T. (1967). Cellular and subcellular events in Wolffian lens regeneration. In Current
Topics in Developmental Biology, vol. 2 (ed. A. A. Moscona & A. Monroy), pp. 247-296.
YAMADA, T. (1976). Dedifferentiation associated with cell-type conversion in the newt lens
regenerating system. A review. In Progress in Differentiation Research (ed. Muller-Berat),
pp. 355-360. Amsterdam: North Holland Publishing Company.
YAMADA, T. & ROESEL, M. (1971). Activation of DNA replication in the iris epithelium by
lens removal. /. exp. Zool. Ill, 425-432.
YAMADA, T., ROESEL, M. & BEAUCHAMP, J. J. (1975). Cell cycle parameters in dedifferentiating iris epithelial cells. /. Embryol. exp. Morph. 34, 497-510.
YAMADA, T. & TAKATA, C. (1963). An autoradiographic study of protein synthesis in
regenerative tissue transformation of iris into lens in the newt. Devi Biol. 8, 358-369.
ZALIK, S. E. & SCOTT, V. (1969). In vitro development of the regenerating lens. Devi Biol. 19,
368-379.
ZALIK, S. E. & SCOTT, V. (1972). Cell surface changes during dedifferentiation in the metaplastic transformation of iris into lens. /. Cell Biol. 55, 134-146.
ZALTK, S. E. & SCOTT, V. (1973). Sequential disappearance of cell surface components during
dedifferentiation in lens regeneration. Nature, Lond. 244, 212-214.
ZALIK, S. E., SCOTT, V. & DIMITROV, E. (1976). Changes at the cell surface during in vivo and
in vitro dedifferentiation in cellular metaplasia. In Progress in Differentiation Research
(ed. N. Miiller-Berat), pp. 361-367. Amsterdam: North Holland Publishing Company.
(Received 5 July 1976, revised 14 December 1976)