/. Embryo/, exp. Morph. Vol. 26, 2, pp. 253-270, 1971
253
Printed in Great Britain
Tentacular and oral-disc regeneration in the
sea anemone, Aiptasia diaphana
III. Autoradiographic analysis of patterns of
tritiated thymidine uptake
By IRWIN I. SINGER 1
From the Department of Biology, New York University
SUMMARY
Autoradiography with [ H]thymidine and electron microscopy were used to determine
(a) the patterns of cellular division exhibited by intact anemones, (b) if measurable increases
in cellular proliferation accompany oral-disc regeneration, (c) whether interstitial cells are
present in Aiptasia, and (d) if these cells could be responsible for the latter proliferative
patterns.
An oral-aboral gradient in cellular proliferation was exhibited by the epidermis of uncut
anemones, with the highest levels in the tentacles.
Wound healing did not require cell proliferation and did not immediately stimulate
cellular division which was associated with subsequent morphogenetic events.
Indices of presumptive oral-disc [3H]thymidine uptake into nuclei increased tenfold with the
outgrowth of the new tentacles. This increase occurred in the epidermis, while only small
amounts of gastrodermal proliferation were detected. It is hypothesized that the epidermis
contributes new cells to the expanding gastrodermis during tentacle budding.
Most of the [3H]thymidine-labeled nuclei were localized in the basal portions of the
epidermis of intact anemones and 1- to 2-day-old regenerates; very few gastrodermal nuclei
accumulated the label.
Nests of interstitial cells and transforming interstitial cells were localized in the exact
epidermal regions where nuclear labeling took place, suggesting that the proliferative patterns
of intact and regenerating Aiptasia are a function of their interstitial cell distribution.
3
INTRODUCTION
The successive structural events culminating in the formation of a new oraldisc and tentacle crown in the sea anemone, Aiptasia diaphana, include two
important steps: (1) the fusion of the cut pharyngeal and columnar tissues,
referred to in this paper as wound healing, and (2) the appearance of the
primordia of the tentacles, 3 days after amputation, which is designated as
tentacle budding (Singer & Palmer, 1969). The purpose of the study reported
here is to determine if measurable increases in cellular division accompany
these regenerative events.
The solution of this problem would shed considerable light on the cellular
1
Author's address: Department of Biology, Saint Louis University, 1504 South Grand
Boulevard, Saint Louis, Missouri, 63104, U.S.A.
17
EMB
26
254
I. I. SINGER
basis of the regeneration in Aiptasia. For example, if no increases in the rates of
cellular division occur before or during tentacle budding, then one would expect
that the cells which are responsible for the building of the tentacles are distally
migrating columnar cells which have transformed into tentacular cells. Similar
mechanisms have been used to explain the reconstitution of complete dwarfs
from isolated gastrodermal fragments (Davis, Burnett, Haynes & Mumaw,
1966) and hypostomal mucous cell regeneration (Rose & Burnett, 1968) in
Hydra viridis. It is also noteworthy that neither increased mitotic activity nor
heightened rates of nuclear [3H]thymidine uptake have been detected in budding
or regenerating Hydra (Li & Lenhoff, 1961; Spagenberg, 1961; Breslin-Spagenberg & Eakin, 1962; Clarkson & Wolpert, 1967; Park, Ortmeyer & Blankenbaker, 1967; Clarkson, 1969).
Alternatively, tentacle buds could be formed by the rapid and repeated
division of all columnar cell types (or a single totipotent cell) thus resulting in
tissue outgrowth from the spaces between the gastric septae on the presumptive
oral-disc of Aiptasia. Comparable increases in mitotic rates have been demonstrated during the reconstitution of polyps from fragments of Amelia aurita
scyphistomae larvae (Steinberg, 1963).
Quantitative autoradiography with [3H]thymidine was used in an attempt to
clarify this problem in Aiptasia. In addition, an electron-microscopic study was
performed to determine if Aiptasia contains undifferentiated interstitial cells
similar to those present in Hydra (Slautterback & Fawcett, 1959; Lentz, 1965)
and in Metridium senile (Westfall, 1966), and if such cells could be responsible
for any patterns of proliferative activity that would emerge from the above study.
MATERIALS AND METHODS
Brown Aiptasia diaphana, containing a symbiotic infestation of zooxanthellae,
were used exclusively in this study. These anemones have been in culture since
1966 in 15 gal (ca. 57 1) fiberglass aquaria using filtered aerated sea water. The
anemones were fed daily with freshly hatched Artemia salina. The density
of the sea water was kept constant by the addition of distilled water; the
culture water was changed periodically. Animals were amputated transversely at
the lower portion of the capitulum; the details of this procedure, as well as the
method used in selecting the anemones, have been described previously (Singer
& Palmer, 1969).
One mCi of [3H]thymidine-methyl, specific activity 3-47 Ci/mM, 99 % radio
purity, was purchased from Nuclear of Chicago, and diluted to a concentration
of 200 /tCi/ml using Millipore-filtered (pore size = 0-45 /im) sea water. This
stock solution was sealed in water-tight flasks and frozen until later use. Immediately before an incubation period some of the stock solution was allowed
to thaw, and was then diluted to a final dosage concentration of either 5 or 10
/tCi/ml with Millipore-filtered sea water.
Thymidine uptake in regenerating Aiptasia
255
To follow the time course of cellular proliferation during wound healing and
tentacle budding, regenerates and intact control anemones were placed in either
5 or 10 /fcCi/ml [3H]thymidine for 4-5 h at various intervals following the day of
amputation, as indicated in Table 1.
Table 1. Daily numbers of anemones used to determine the time course
of ceil division during regeneration
Days after amputation
Daily numbers of regenerates
Daily numbers of controls
1
2
3
4
5
3
3
4
2
4
3
4
2
4
3
After incubation, the anemones were washed several times with sea water and
immediately prepared for histological study as previously described (Singer &
Palmer, 1969). Stained slides containing serial sections were coated with Kodak
NTB2 liquid emulsion according to the technique of Kopriwa & Leblond (1962).
Before this was done, a test was performed to determine if any significant fogging
of the emulsion had occurred. Clean glass slides were dipped into the emulsion,
dried under controlled humidity, and developed. Forty determinations of the
number of exposed grains per 1000 jam2 were made, and values of 2-4 were
consistently obtained, indicating low initial background fog.
Several control procedures were employed for each series of autoradiographs
that was processed. Slides containing either no sections, stained unlabeled
sections, or unstained unlabeled sections were coated with the emulsion and
processed at the end of each experiment. In no case was there any effect of the
unlabeled tissue or stain upon the level of emulsion fogging.
Autoradiographs prepared from anemone tissues incubated in 5 or 10/tCi/ml
[3H]thymidine, and exposed for 2 or 7 days respectively (in light-tight slide
boxes containing Drierite at 4 °C) exhibited adequate levels of exposure. The
slides were processed according to Kodak's instructions, and subsequently
covered with cover-glasses.
The number of labeled nuclei per 1000 /mi2 of tissue was determined using a
net reticule inserted into the ocular of the microscope. This value was defined as
the thymidine-tabeling index, and it was assumed to reflect the rate of cellular
division of a given tissue region in Aiptasia. To be counted as labeled, a nucleus
must have been covered by a dense spherical cluster of exposed grains. All
measurements were made at x 100.
Thymidine-labeling indices were computed for the epidermis of the tentacles,
oral-disc, capitulum, and the pedal disc of non-regenerating Aiptasia. Likewise,
measurements were completed for the epidermis of the presumptive oral-disc
and tentacle buds (where appropriate), the new capitulum, and the pedal disc
of regenerates. The individual mean thymidine-labeling index for each tissue
17-2
256
I. I. SINGER
region of every anemone, be it a regenerate or control, was obtained by averaging
at least 20 random determinations of the thymidine-labeling index derived from
sampling at least five sections selected haphazardly. The Student t test was
applied to determine if the individual mean thymidine-labeling indices obtained
M
Thymidine uptake in regenerating Aiptasia
257
for successive axial regions of an anemone differed significantly; a P value of
less than 0-05 was considered significant.
In addition, the thymidine-labeling indices for each tissue region within a
daily common population were pooled and averaged - by region - and group
mean thymidine-labeling indices were thus derived. The Student t test was again
used to see whether the group mean thymidine-labeling indices derived for
each successive body region of a given population were significantly different;
the above confidence levels were used.
For purposes of ultrastructural study, non-regenerating anemones were fixed
by immersion in 5 % glutaraldehyde made up with pH 7-4 phosphate buffer for
1-5 h at 4 °C. The tentacles were then removed, macerated, and washed with
10 % sucrose in pH 7-4 phosphate buffer for 1 h at 4 °C. Postfixation was
performed for 1 h at 4 °C with a 1 % OsO4 solution made up in the above buffer
containing 4 % sucrose. Tentacular tissues were rapidly dehydrated and subsequently embedded in Epon 812 (A:B = 1:1) which was then cured for 3 days
at the following temperatures on each day: 35, 45 and 60 °C. Sections 1 /mi
thick were cut on a Reichert ultramicrotome and later stained with 0-5 %
toluidine blue in 2 % sodium tetraborate buffer. Silver-gold thin sections were
cut with a diamond knife, mounted on Formvar carbon-coated grids, and
stained using a saturated solution of uranyl acetate for 30 min followed by lead
citrate (Reynolds, 1963) for 5 min at room temperature. Electron micrographs
were taken through a Zeiss 9 S electron microscope.
Fig. I. Autoradiograph of a tentacle of non-regenerating Aiptasia incubated in
[3H]thymidine and stained with hematoxylin and eosin. The epidermis contains
many intensely labeled nuclei (N) located basally, just above the processes of the
epitheliomuscular cells (P). The gastrodermis (G) exhibits a diffuse labeling, and is
filled with zooxanthellae. S, spirotrich nematocyst; Mu, muscular portion of
epidermis; M, mesoglea.
Fig. 2. Autoradiograph of the pedal-disc epidermis (E) of non-regenerating Aiptasia
prepared as above. Several labeled nuclei (JV) are found in the region of the epidermis
above the mesoglea (M).
Fig. 3. Autoradiograph of the gonads of a non-regenerating sexually mature anemone
prepared as Fig. I. Many [3H]thymidine-tagged nuclei (N) are situated in the
peripheral regions of each gonadal segment.
Fig. 4. Autoradiograph of the presumptive oral-disc and the healing wound of a
.1-day-old regenerate prepared as indicated above. Very little epidermal nuclear
labeling is displayed by the healing epidermis (He), and none by the healing gastrodermis (Hg). The mesoglea (M) is thin in the wounded region. A few basally located
nuclei (N) can be found in the epidermis of the presumptive oral-disc (Pod).
Fig. 5. Autoradiograph of the presumptive oral-disc of a 2-day-old regenerate
prepared as above. Thymidine-labeled nuclei (N) are situated in the basal part of the
epidermis (£). Zooxanthellae (Z) are present in both the epidermis and the gastrodermis (G).
Fig. 6. Photomicrograph of the gastrodermis of a non-regenerating tentacle prepared
using ultrastructural procedures and stained with toluidine blue. Zooxanthellae (Z)
are surrounded by thin but continuous cytoplasmic extensions (Ce) of gastrodermal
cells.
258
I. I. SINGER
RESULTS
Autoradiography of non-regenerating anemones
Most of the thymidine-labeled nuclei are localized in basal regions of the
epidermis, i.e. immediately above the epitheliomuscular cell processes of the
tentacles (Fig. 1), or adjacent to the mesoglea of other columnar regions (Fig. 2).
The gastrodermis of these animals rarely exhibits intensely labeled nuclei;
instead, individual silver grains are spread diffusely throughout this body layer
(Figs. 1, 5). Because the epidermis is coated with, a rather impermeable mucous
layer, and since the gastrodermis is the primary trophic layer in coelenterates,
this pattern of labeling suggests that the gastrodermis binds the thymidine and
then passes it along to the epidermis where it is incorporated into the DNA of
cells preparing to divide. This type of labeling may also be due to bacterial
contamination in part. The peripheral regions of the gonads are the only
gastrodermal regions which exhibit nuclear localization of thymidine, presumably associated with germ cell formation (Fig. 3).
2 10.
"i
8
1
2
3
4
5
Consecutive days of isolation
Fig. 7. A graphic representation of the group mean thymidine-labeling indices of
successive axial epidermal tissue regions in non-regenerating Aiptasia: the tentacles
(O—O), oral-disc ( • — • ) , capitulum (O
O), and pedal disc ( •
#).
The mean thymidine-labeling indices for each of the various epidermal regions
(tentacles, oral-disc, capitulum, and pedal disc) of non-regenerating anemones
are presented in Table 2.
Inspection of the individual and group mean thymidine-labeling indices in
Table 2 reveals a distal-proximal proliferative gradient (with the highest values
in the tentacular epidermis, and the lower indices in the pedal-disc epidermis) in
non-regenerating anemones. The differences between the mean thymidinelabeling indices of successive axial epidermal regions are statistically significant
Days
C7/25
P
C7/34
P
C7/8
P
C7/31
P
C7/27
P
C7/33
P
C7/32
P
C7/3
P
C7/30
P
C7/28
P
C7/29
P
C7/26
P
C7/9
P
Animals;
P values
3-81
901
9-75
906
6-39
10-28
919
711
10-46
9-62
8-24
8-29
50
100
50
50
50
50
100
50
50
50
50
100
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
3-61
410
5-59
4-78
3-40
4-56
2-72
4-49
4-81
616
4-77
2-58
3-26
4-92
50
0001
Oraldisc
Tentacles
0001
0001
0 001
0001
0 001
0001
0001
0001
001
0 001
0001
002
001
2-30
2-47
2-70
1-87
2-25
203
1-79
2-90
3-75
3-49
2-43
206
2 61
005
0001
0001
0001
001
005
0001
0001
001
0001
0001
0001
0001
Capitulum
1-64
1-45
1-87
1-37
1-59
114
1-27
1-84
1-94
1-80
1-20
1-66
1 -81
Pedal
disc
Individual mean thymidine-labeling indices
Concentration
[3H]thymidine
(/iCi/ml)
•71
•03
8-55
9-54
0001
0001
0001
0001
4-45
3-98
3-84
5-49
0001
0001
0001
0001
2-48
210
2-26
3-54
0001
0001
0001
0001
1-65
1-50
1-43
1 87
Group mean thymidine-labeling indices
\.
OralPedal
Tentacles
disc
Capitulum
disc
4-62
3-27
1 59
2-29
0001
0001
0001
(The P values represent the confidence levels of the differences between the mean thymidine-labeling
indices of successive regions of the epidermis; P is less than its indicated values in all cases.)
Table 2. The individual and group mean epidermal thymidine-labeling indices of non-regenerating
Aiptasia diaphana sacrificed daily during a 5-day period at 20 °C
I'
260
I. I. SINGER
in all cases. The data on the right side of Table 2 are portrayed graphically in
Fig. 7, where it can easily be seen that an axial gradient exists, and that the group
mean thymidine-labeling indices of each epidermal region remain rather constant throughout the 5-day period of observation. The relatively slight increases
250//m
Pb
12
250 fim
Thymidine uptake in regenerating Aiptasia
261
in the group mean thymidine-labeling indices of the tentacular and oral-disc
epidermis observed between days 1 and 2 are probably due to cell division
associated with the stimulation of nematocyte production during the selection
procedures performed before day 1.
Autoradiography of regenerating anemones
Autoradiographs prepared from anemones labeled and fixed 1 day after
excision of the oral-disc show little if any localization of [3H]thymidine in the
immediate vicinity of the healing wound. Wound healing, which is still in
progress at this time, therefore requires little or no cell division. The epidermal
tissues adjacent to the wound which are destined to become the new oral-disc
similarly show little labeling. Any localization of the label which may occur in
the presumptive oral-disc during wound healing is usually exhibited by basal
portions of the epidermis (Fig. 4).
Two-day-old regenerates exhibit increased numbers of labeled nuclei in only
the basal epidermal regions of the presumptive oral-disc. In addition, these
animals have zooxanthellae located in both the epidermis and the gastrodermis
of their forming oral-disc (Fig. 5). Since the zooxanthellae are normally
surrounded by the cytoplasmic extensions of gastrodermal cells (Fig. 6), their
Fig. 8. Autoradiograph of a longitudinal section of a tentacle bud {Tb) and the
pharyngeal tissues (Ph) of a 3-day-old regenerate incubated in [3H]thymidine and
stained with hematoxylin and eosin. The epidermis of the growing tentacle contains
many labeled nuclei, while its gastrodermis {G) displays relatively little labeling.
Fig. 9. Autoradiograph of the presumptive oral-disc {Pod) and pharyngeal tissues
{Ph) of the same anemone depicted in Fig. 8. Again, note that the presumptive oraldisc epidermis is filled with [3H]thymidine-labeled nuclei, whereas the gastrodermis
is not.
Fig. 10. Higher magnification of the presumptive oral-disc epidermis shown in Fig. 9.
The labeled nuclei {N) are distributed in the basal {B), middle (M), and apical {A)
portions of the epithelium.
Fig. 11. Autoradiograph of the presumptive oral-disc {Pod), tentacle bud (Tb), and
presumptive capitulum {Pc) of a 5-day-old regenerate prepared as the anemone in
Fig. 8. The epidermis of all of these regions contains large numbers of labeled nuclei;
the gastrodermis (G) exhibits only sparse amounts.
Fig. 12. Photomicrograph of a 1/tm section of non-regenerating tentacular epidermis
prepared using electron-microscopic techniques and stained with toluidine blue.
Both apical nuclei {Na) and basal nuclei {Nb) are present. Several basophilic cells,
which apparently are cnidoblasts {Cn), and their precursors, are localized in a basal
position, just above the network of epitheliomuscular cell processes {P). One of these
cells contains a coiled tubular structure cut in cross-section {T). S, spirotrich nematocyst; Mu, muscle fibers of epitheliomuscular cells; M, mesoglea.
Fig. 13. Photomicrograph of a 5/tm section of non-regenerating pharyngeal epidermis prepared using the paraffin technique and stained with hematoxylin and
eosin. The epidermis contains numerous basally situated nuclei (TV) adjacent to the
network of epitheliomuscular cell processes (P). Gl, glandular cell; M, mesoglea;
G, gastrodermis.
Days
2-61
2-14
4-63
11-62
11-58
10-21
9-88
1608
1510
22-23
28-70
1807
22-39
2207
27-42
26-97
31-41
17-39
3115
50
50
100
50
50
50
100
50
50
50
100
50
50
50
100
50
50
50
100
cue
p
C7/13
P
C7/23
P
C7/18
P
C7/5
P
C7/19
P
C7/21
P
C7/15
P
C7/1
P
C7/12
P
C7/17
P
C7/4
P
C7/16
P
C7/20
P
C7/24
P
C7/2
P
C7/14
P
C7/11
P
C7/22
P
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
0001
--«
***
• 0-80
- 0-30
515
7-28
5-96
6-84
306
2-64
1-62
2-30
406
5-57
3-80
3-36
5 05
3 08
2-57
2-80
4-68
1-93
0001
0001
0001
0001
0001
002
0001
0001
0001
0001
0001
0001
0001
0001
001
0001
0001
0001
1-78
1-27
211
2-56
1-32
1-42
0-86
1-38
1-35
206
1-90
1-75
1-70
1-73
1-69
1-83
1-98
1-25
—
—
—
26-76
—
—
—
2206
—
—
—
19-78
—
—
—
10-90
—
—
0001
—
—
0001
—
—
0001
—
0001
—
—
—
—
6-31
—
—
—
2-29
—
—
—
4-21
—
—
—
3-32
—
—
2-84
New
capitulum
> 0-70
2-91
0-93
2-47
0001
Presumptive
oral-disc
Pedal
disc
0001
—
0001
—
0001
—
0001
—
0001
—
—
1-93
_-.
1-23
. .
—
1-79
—
—
1-75
—
—
1-35
Pedal
disc
Group mean thymidine-labeling indices
New
capitulum
—•0-50
.
*-*
Presumptive
oral-disc
Animals;
P values
Individual mean thymidine-labeling indices5
Concentration
[3H]thymidine
(/iCi/ml)
(The P values represent the confidence levels of the differences between the mean thymidine-labeling indices of
successive regions of the epidermis; P is less than its indicated values, except where shown otherwise.)
Table 3. The individual and group mean epidermal thymidine-labeling indices of regenerating
Aiptasia diaphana sacrificed daily during a 5-day period at 20 °C
tn
o
ON
Thymidine uptake in regenerating Aiptasia
263
presence in the epidermis implies that gastrodermal cells have migrated into the
epidermis shortly after the completion of wound healing.
Anemones which have regenerated for 3 days display even greater increases in
the quantity of labeled epidermal nuclei in the tentacle buds (Fig. 8) and the rest
of the new oral-disc (Fig. 9). Further, non-regenerates and 1- to 2-day-old
regenerates only show labeled nuclei in basal epidermal regions, while anemones
which have regenerated for 3 days exhibit labeled nuclei in the apical, middle
and basal regions of their regenerating epidermal epithelia (Fig. 10). This
suggests that basal cells are being forced apically as they proliferate, or that cells
already located in apical epidermal regions have begun to duplicate at this time.
The gastrodermis of 3-day-old regenerates shows relatively little nuclear
labeling (Figs. 8, 9).
Increased quantities of labeled nuclei are likewise found throughout the
epidermis of the presumptive oral-disc, tentacle buds, and presumptive capitulum
in 4- and 5-day-old regenerates; the gastrodermis in these regions does not
contain many labeled nuclei (Fig. 11).
The individual and group mean thymidine-labeling indices computed for the
epidermis of the presumptive oral-disc and tentacle buds (where present), the
presumptive capitulum, and the pedal disc of regenerating Aiptasia are presented
in Table 3.
The data in Table 3 show that a statistically significant distal-proximal
gradient in epidermal thymidine-labeling indices exists in the regenerating
anemones. The index values for the presumptive capitulum and the pedal disc
are similar to those found in the non-regenerates, while the values for the presumptive oral-disc steadily increase due to the regenerative process. On day 1
the epidermal mean thymidine-labeling indices of the presumptive oral-disc
and the new capitulum are almost identical, but the indices of these two regions
become significantly different on day 2 when the group mean thymidine-labeling
index of the presumptive oral-disc is comparable to that found in the tentacular
epidermis of non-regenerates. The differences between the mean thymidinelabeling indices of the presumptive oral-disc and the capitulum continue to
enlarge progressively, and the group mean thymidine-labeling index of the
regenerating oral-disc epidermis increases tenfold by the end of the experiment.
The mean thymidine-labeling indices of the new capitulum and the pedal disc
remain relatively constant during the first 4 days of regeneration, but the indices
of the former region increase noticeably on day 5 while those of the latter tissue
region are still unchanged. These trends are shown graphically in Fig. 14, which
is composed from the values on the right side of Table 3.
Tentacular ultrastructure of non-regenerates
Histological studies using the paraffin technique show that most epidermal
nuclei of non-regenerating Aiptasia are localized in the basal portions of the
epithelium, just above the network of epitheliomuscular cell processes (Fig. 13).
264
I. I. SINGER
In contrast, sections 1 /im thick prepared according to ultrastructural procedures
reveal that the nuclei of the tentacular epidermal cells are indeed found in the
apical, middle and basal regions of the epithelium. Further, basophilic cells
which apparently are interstitial cells and cnidoblasts are exclusively localized
in clumps between the bases of the other epithelial cells (Fig. 12).
25
•E
20
i15
10
1 -
1
2
3
4
Days after amputation
5
Fig. 14. A graphic portrayal of the group mean thymidine-labeling indices of successive axial epidermal tissue regions in regenerating Aiptasia: the presumptive oral-disc
and tentacles ( • - - - • ) , the new capitulum ( • — # ) , and the pedal disc ( •
•).
A significant difference between the indices of the presumptive oral-disc epidermis
and the new capitular epidermis isfirstobserved on day 2; the new tentacles become
visible on day 3.
Electron-microscopic examination of the basally located basophilic cells
confirms the hypothesis that they are interstitial cells and cnidoblasts. Spindleshaped interstitial cells containing large nuclei, prominent nucleoli and relatively
small amounts of cytoplasm lacking the various types of cellular organelles
(except for ribosomes) are found adjacent to cnidoblasts containing nematocyst
primordia; these two cell types are usually found clumped together (Fig. 15).
Interstitial cells assumed to be in the process of transforming into more
specialized cells are sometimes found in the latter cellular aggregates. These
Thymidine uptake in regenerating Aiptasia
265
15
Fig. 15. Electron micrograph of a nest of interstitial cells and differentiating cnidoblasts located in the basal epidermal region (see Fig. 12) of a non-regenerating
tentacle. Two interstitial cells (/c) containing small amounts of cytoplasm and some
ribosomes are located towards the upper left. One of these cells has a prominent
granular nucleolus. A slightly more differentiated interstitial cell containing a
secretory vesicle (Sv) is shown at the lower right. Two cnidoblasts (Cn) are wedged
between these interstitial cells. Each cnidoblast contains rough surfaced endoplasmic
reticulum(er), and cross-sections of a tubular nematocyst primordium (T) surrounded
by microtubules in several regions. Thecnidoblast at the bottom of the micrograph
has a large Golgi apparatus (C7) whose lamellae contain a flocculent material. The
close association of an active Golgi apparatus and the nematocyst primordium
suggests that nematocyst proteins are derived from this organelle. N, nucleus;
tn, mitochondrion.
Fig. 16. An electron micrograph of the secreting Golgi apparatus of a transforming
interstitial cell in the basal region of the non-regenerating tentacular epidermis. The
vesicle at the center of this organelle (arrow) is filled with a dense substance, and is
probably the precurser of a nematocyst.
266
I. I. SINGER
transforming cells contain slightly increased amounts of cytoplasm, rough
endoplasmic reticulum, secretory vesicles (Fig. 15), and a well-defined active
Golgi apparatus exhibiting evidence of secretory vesicle formation (Fig. 16).
The interstitial cells, their derivatives, and cnidoblasts have the exact location as
the cells whose nuclei show large amounts of [3H]thymidine uptake in the autoradiographs of non-regenerating tentacles; comparable cells could not be found
in the gastrodermis.
DISCUSSION
Wound healing
The observation that no increases in the number of [3H]thymidine-labeled
nuclei accompany wound healing in A. diaphana demonstrates that this process
does not require cell division, and that wounding does not immediately stimulate
cell proliferation, which is associated with subsequent morphogenetic events.
This conclusion is supported by studies on Hydra which show that subhypostomal
amputation does not stimulate DNA synthesis (Clarkson, 1969) and that mitotic
activity does not increase during any stage of distal-end regeneration (Park et al.
1967). The exact mechanism whereby the cut epidermal and gastrodermal tissues
fuse during wound healing in the above coelenterates is unknown. However,
healing in Aiptasia could occur as it does in the hydrozoan Tubularia (Stevens,
1903), where the cells at the edge of the cut tissue elongate and stretch, thus
forming a membrane that closes the wound aperture. Electron-microscopic
studies are presently being conducted in this laboratory to determine if similar
events are responsible for wound closure in Aiptasia.
Proliferating cell types
The autoradiographs of Aiptasia presented in this paper show that nuclear
DNA synthesis in non-regenerates and 1- to 2-day-old regenerates is localized in
a narrow basal portion of the epidermal epithelium; the gastrodermis in these
anemones contains relatively few labeled nuclei. Further, electron-microscopic
observations show that epidermal interstitial cells and their derivatives form
aggregates situated in the exact location where epidermal nuclear labeling takes
place; no interstitial cells could be found in the gastrodermis. It is therefore
postulated that epidermal interstitial cells are primarily responsible for the
dramatic increases in the epidermal mean thymidine-labeling indices of early
regenerates.
Since the tentacle buds of regenerating Aiptasia are formed by the column,
which lacks the specialized glandular, sensory, and nematocyst-containing cells
characteristic of the tentacles (Stephenson, 1928), an analysis of tentacular
regeneration must explain how these cells are formed. In addition, one must
account for the production of increased numbers of epitheliomuscular cells
during the growth of tentacle buds. Consequently, the early phases of tentacular
formation in Aiptasia would involve the proliferation of undifferentiated
Thymidine uptake in regenerating Aiptasia
267
epidermal interstitial cells, followed by their differentiation into epitheliomuscular cells as well as the other characteristic tentacular cell types. It is
reasonable to claim that the interstitial cells of Aiptasia can form a variety of
differentiated cell types, since they are very unspecialized cells with little cytoplasm. Also, similar cells are found in the hydrozoa, and are of considerable
importance during morphogenesis (Tardent, 1960); Lentz (1965) has also
presented evidence at the ultrastructural level which supports the totipotentiality
of interstitial cells m Hydra Iittoralis. Experiments are now being done to verify the
hypothesis that the epidermal interstitial cells of Aiptasia actively proliferate
and subsequently differentiate into the specialized cell types required for tentacular regeneration. Regenerating anemones are given a short pulse of [3H]thymidine (thus tagging the interstitial cells), and subsequently fixed at various times
during the regenerative period; electron-microscopic autoradiography is being
employed to determine the cell types that contain the label (which are derived
from the interstitial cells).
The finding that the nuclei of the cells located in the middle and apical
regions of the epidermis rarely incorporate [3H]thymidine during normal growth
or within the first 2 days of regeneration, suggests that such cells do not proliferate actively at these times. However, they do concentrate [3H]thymidine in
3-day-old regenerates, indicating that the nuclei of the differentiated cells
located in the above regions (i.e. epitheliomuscular cells) have begun to divide.
Epitheliomuscular cells in regenerating hydrozoans are capable of division
(Diehl & Burnett, 1965; Diehl, 1969), thus supporting this observation. Increased
numbers of this cell could therefore be formed via autoreproduction during the
later stages of regeneration.
Gradients in cell division and morphogenesis
The interesting observation that the number of nuclei synthesizing DNA in
non-regenerating anemones was highest in the tentacular epidermis and
decreased proximally in graded fashion is probably related to the morphogenetic
potentials of various axial levels along the column. Experiments designed to
measure the ability of different columnar levels to undergo tentacle-crown
regeneration in Aiptasia revealed that an oral-aboral gradient exists concerning
the time required for the initiation of tentacle budding: the tentacle primordia
of distally transected anemones appeared sooner than those of anemones
amputated through more proximal columnar levels (Singer & Palmer, 1969).
Since large increases in the numbers of labeled epidermal nuclei accompany
regeneration in Aiptasia, the above results suggest that the relative ability of any
given columnar region to initiate tentacle-bud outgrowth is governed by its
vegetative rate of epidermal cell division. Thus, proximal columnar levels
cannot begin to form tentacles until their epidermal rates of cell proliferation
reach a certain threshold. Regional differences in cell division rates may in turn
be regulated by metabolic gradients (Child, 1941).
268
I. I. SINGER
Epidermal proliferation during regeneration
During tentacle-crown and oral-disc regeneration in Aiptasia there is a
tenfold increase in the number of [3H]thymidine-labeled nuclei in the presumptive oral-disc epidermis. This augmented epidermal cell proliferation is
first noticed 2 days after amputation - one full day before tentacle budding. The
latter finding is in marked contrast to data collected on regenerating Hydra.
showing that the formation of a new hypostome is neither accompanied by
significant increases in DNA synthesis (Clarkson, 1969), nor growing numbers of
mitotic figures (Spagenberg, 1961; Breslin-Spagenberg & Eakin, 1962; Park
et ah 1967). Workers on Hydra have therefore proposed that a primary mechanism which accounts for increasing the local cell number of a growing or
regenerating region is the process of cellular migration (Shostak, Patel & Burnett,
1965; Shostak & Globus, 1966; Clarkson & Wolpert, 1967; Rose & Burnett,
1968); new cells are presumably produced elsewhere during the post-regenerative
period. Such a situation does not obtain in Aiptasia where there appears to be
a direct association between morphogenesis and cellular division within a
regenerating region.
Another important finding of this study is that while the epidermis of Aiptasia
exhibits measurable amounts of cell division during growth and regeneration,
the gastrodermis displays relatively little proliferative activity when these events
occur. Other mechanisms must be invoked to explain how increased quantities
of gastrodermal cells accumulate in the growing and regenerating tentacles of
Aiptasia. The formation of an oral 'pigmented ring' of alga-containing gastrodermal cells in the pretentacular area (Singer & Palmer, 1969) suggests that
gastrodermal cells are capable of migrating to regenerating regions in A.
diaphana. Relatively independent movements of gastrodermal tissues have also
been observed in various hydroids during regeneration (Breslin-Spagenberg &
Eakin, 1962; Tardent, 1962). In addition, the actively proliferating epidermis of
Aiptasia may produce cells capable of migrating into the gastrodermis and
redifferentiating in accordance with their new surroundings. The production of
gastrodermal cells by the epidermis is common during growth and morphogenesis in various coelenterates - Amelia aurita (Steinberg, 1963), Clytia
johnstoni (Hale, 1964), Cordylophora caspia (Diehl, 1969), Podocoryne carnea
(Braverman, 1969), Hydra oligactis and H. pseudoligactis (Lowell & Burnett,
1969) - supporting the idea that a similar phenomenon may occur in Aiptasia.
The observation that alga-containing gastrodermal cells are capable of migrating
into the epidermis also lends credence to this hypothesis, because it shows that
the mesoglea is not a barrier preventing the passage of cells between the epidermis
and gastrodermis. Experiments where [3H]thymidine-labeled epidermis is
recombined with non-labeled gastrodermis during regeneration, and subsequently studied autoradiographically, should corroborate the epidermal origin
of at least some gastrodermal cell types.
Thymidine uptake in regenerating Aiptasia
269
J am indebted to Professor John D. Palmer for his advice, support and encouragement
during this study. 1 also thank Mrs Christine Happ for training in ultrastructural techniques
and Professor George Happ for use of his laboratory facilities. This work was sustained by
N.S.F. grant GB-5045 to J.D.P. and N.I.H. Bio-Medical Sciences Support Grant to N. Y.U.
This work forms part of a dissertation submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy at New York University, June 1970.
I thank Mrs Tana Muellner for secretarial assistance.
REFERENCES
M. (1969). Studies on hydroid differentiation. IV. Cell movements in Podocoryne
carnea hydranths. Growth 33, 99-111.
BRESLIN-SPAGENBERG, D. & EAKIN, R. E. (1962). Histological studies on the mechanisms
involved in Hydra regeneration. /. exp. Zool. 151, 85-94.
CHILD, C. M. (1941). Patterns and Problems of Development. University of Chicago Press.
CLARKSON, S. G. (1969). Nucleic acid and protein synthesis and pattern regulation in hydra.
1. Regional patterns of synthesis and changes of synthesis during hypostome formation.
/. Embryol. exp. Morph. 21, 33-54.
CLARKSON, S. G. & WOLPERT, L. (1967). Bud morphogenesis in hydra. Nature, Lond. 214,
780-783.
DAVIS, L. E., BURNETT, A. L., HAYNES, J. E. & MUMAW, V. R. (1966). A histological and
ultrastructural study of dedifferentiation and redifferentiation of digestive and gland cells
in Hydra viridis. Devi Biol. 14, 307-329.
DIEHL, F. A. (1969). Cellular differentiation and morphogenesis in Cordylophora. Wilhelm
Roux Arch. EntwMech. Org. 162, 309-335.
DIEHL, F. A. & BURNETT, A. L. (1965). The role of the interstitial cells in the maintenance of
Hydra. 111. Regeneration of the hypostome and tentades. /. exp. Zool. 158, 299-318.
HALE, L. J. (1964). Cell movements, cell division, and growth in the hydroid, Clytia johnstoni.
J. Embryol. exp. Morph. 12, 517-538.
KOPRIWA, B. & LEBLOND, C. P. (1962). Improvements in the coating technique of autoradiography. /. Histochem. Cytochem. 10, 269-284.
LENTZ, T. L. (1965). The fine structure of differentiating interstitial cells of hydra. Z. Zellforsch. mikrosk. Anat. 67, 547-560.
Li, Y. & LENHOFF, H. M. (1961). Nucleic acid and protein changes in budding Hydra littoralis.
In The Biology of Hydra, ed. W. F. Loomis and H. M. Lenhoff. Coral Gables, Florida:
University of Miami Press.
LOWELL, R. D. & BURNETT, A. L. (1969). Regeneration of complete hydra from isolated
epidermal explants. Biol. Bull. mar. biol. Lab., Woods Hole 137, 3.12-320.
PARK, H. D., ORTMEYER, A. & BLANKENBAKER, D. (1967). Mitotic activity in Hydra pseudoligactis during regeneration of apical structures. Am. Zool. 7, 750-751. (Abstr.)
REYNOLDS, E. S. (1963). The use of lead citrate at high pH as an electron-opaque stain in
electron microscopy. /. Cell Biol. 17, 208-212.
ROSE, P. G. & BURNETT, A. L. (1968). An electron microscopic and histochemical study of the
secretory cells in Hydra viridis. Wilhelm Roux Arch. EntwMech. Org. 161, 281-298.
SHOSTAK, S. & GLOBUS, M. (1966). Migration of epitheliomuscular cells in Hydra. Nature,
Lond. 210, 218-219.
SHOSTAK, S., PATEL, N. G. & BURNETT, A. L. (1965). The role of the mesoglea in mass cell
movements in Hydra. Devi Biol. 12, 434-450.
SINGER, 1.1. & PALMER, J. D. (1969). Tentacular and oral-disc regeneration in the sea anemone,
Aiptasia diaphana. I. Sequential morphological events in distal-end restitution. /. Morph.
127, 373-382.
SLAUTTERBACK, D. B. & FAWCETT, D. W. (1959). The development of cnidoblasts of Hydra.
J. biophys. biochem. Cytol. 5, 441-453.
SPAGENBERG, D. B. (1961). A study of normal and abnormal regeneration of Hydra. In The
Biology of Hydra, ed. W. F. Loomis and H. M. Lenhoff. Coral Gables, Florida: University
of Miami Press.
BRAVERMAN,
l8
EMB 26
270
I. I. SINGER
STEINBERG, S. (1963). The regeneration of whole polyps from ectodermal fragments of
Scyphistomae larvae of Aurelia aurita. Biol. Bull. mar. biol. Lab., Woods Hole 124, 337-343.
STEPHENSON, T. A. (1928). The British Sea Anemones, vol. i. Monogr. Ray Soc. London,
no. 113.
STEVENS, N. M. (1903). Regeneration in Tubularia mesembryanthemum: II. Wilhelm Roux
Arch. EntwMech. Org. 15, 319-326.
TARDENT, P. E. (1960). Principles governing the process of regeneration in hydroids. In
Developing Cell Systems: Eighteenth Growth Symposium, ed. D. Rudnick. New York:
Ronald Press.
TARDENT, P. E. (1962). Morphogenetic phenomena in the hydrocaulis of Tubularia. Pubbl.
Staz. zool, Napoli 33, 50-63.
WESTFALL, J. A. (1966). The differentiation of nematocysts and associated structures in the
Cnidaria. Z. Zellforsch. mikrosk. Anat. 75, 381-403.
{Received 17 November 1970, revised 10 March 1971)