- Email: [email protected]
Genetic analysis of developmental mechanisms in hydra
DEVELOPMENTAL
BIOLOGY
126.263-269 (1988)
Genetic Analysis of Developmental Mechanisms in Hydra XVIII. Mechanism for Elimination of the Interstitial Cell Lineage in the Mutant Strain Sf-1 HIROYUKI
TERADA,
TSUTOMU SUGIYAMA,*
AND YOSHINOBU SHIGENAKA
Faculty of Integrated Arts and Sciences, Hiroshima University, l-l-89 Higashi-Sendamachi, Naka-ku, Hiroshima 7.90,Japan; and *Laboratory of Developmental Genetics, National Institute of Genetics, 1111 Yata, Mishimu 411, Japan Accepted November 20, 1987 The interstitial cell lineage in mutant strain sf-1 of hydra is temperature sensitive and is lost rapidly from tissue when the animal is cultured at a restrictive temperature of 23’C or higher. The mechanism responsible for this cell elimination process was investigated. Sf-1 polyps were treated at a restrictive temperature of 27°C for varying lengths of time, their tissues were macerated, and the resultant dissociated cells were examined for evidence of phagocytosis after Feulgen staining. It was found that large phagocytic vacuoles were present in the cytoplasm of some epithelial cells. These vacuoles contained partially degraded cells, whose nuclei had highly-condensed and intensely Feulgen-positive chromatin granules. This indicated that, as in colchicine-treated (Campbell, 1976) or starved (Bosch and David, 1984) wild-type hydra, the epithelial cells in strain sf-1 engulfed and disintegrated other cells in the phagocytic vacuoles. The incidence of phagocytosis was higher in sf-1 tissue maintained at elevated temperature than in sf-I tissue maintained at normal temperature. However, the observed incidence was relatively low (maximally 0.14 phagocytosed cells per epithelial cell) and appeared to be too low to account for the very rapid interstitial cell loss occurring in this strain. We concluded that elimination of the interstitial cell lineage at a restrictive temperature in strain sf-I takes place in part by phagocytosis and in part by other yet-unidentified mechanisms (~3, Marcum et al, 1980). o 1988 Academic Press, Inc.
INTRODUCTION
Epithelial cells in hydra have the ability to engulf and eliminate other cells in tissue. This was initially demonstrated in Hydra attenuata by Campbell (1976). In response to colchicine treatment, interstitial cells and their differentiation products (nematoblasts, nematocytes, nerve cells, and gland cells) disappeared from the tissue at varying rates. Examination by light and electron microscopy revealed that this cell loss occurred through phagocytosis by both ectodermal and endodermal epithelial cells. Recently, a simple and rapid procedure to monitor cell phagocytosis in hydra was developed by Bosch and David (1984). When hydra tissue was macerated and the resultant dissociated cells were examined after Feulgen staining, the engulfed cells present in the phagocytic vacuoles of epithelial cells were readily identified by their distinct staining and morphological characteristics. Using this procedure, Bosch and David showed that extensive cell elimination by phagocytosis occurred in the tissue of starved H. attenuata (Bosch and David, 1984), and also at the junction of two tissues when two different species of hydra were grafted together (Bosch and David, 1986). The same procedure was also adopted by Kobatake and Sugiyama (1986) to examine phagocytosis occurring in a mutant strain of Hydra magnipapillata (L4) which produced “excess” epithelial cells. 263
In the present study, we adopted the same procedure to examine phagocytosis in another mutant strain, sf-1 (Sugiyama and Fujisawa, 1978; Marcum et ak, 1980). This mutant strain, whose interstitial cell lineage is temperature sensitive, grows normally when cultured at 18°C. However, it rapidly loses its interstitial cell lineage when the culture temperature is shifted up to 23°C (or higher). When maintained at this temperature for a week or longer, it loses all its interstitial cells and practically all of their differentiation products (nematoblasts, nematocytes, nerve cells, and gland cells) and turns into “epithelial hydra” which lack the ability to move or feed (Marcum and Campbell, 1978; Sugiyama and Fujisawa, 1978). At present, the mechanism for this cell elimination in sf-1 is not known. In a previous investigation, Marcum et al. (1980) used electron microscopy to examine sf-1 tissue maintained at an elevated temperature, but failed to detect phagocytosis. This, however, did not necessarily rule out the possibility of phagocytosis in sf-1 since the number of cells examined by electron microscopy was limited, and a low incidence of phagocytosis may have been undetected. To reexamine this possibility, sf-1 animals were treated at an elevated temperature for varying periods of time, and the rate of interstitial cell lineage elimination in their tissues was determined. Using the same tissues, the incidence of cell phagocytosis was systemat0012-1606/88 $3.00 Copyright All rights
0 1988 by Academic Press, Inc. of reproduction in any form reserved.
264
DEVELOPMENTAL BIOLOGY
ically examined by the maceration-Feulgen procedure of Bosch and David (1984). Similar analyses were also carried out in parallel for sf-1 animals treated with colchicine. The experiments revealed that cell phagocytosis occurred in sf-1 tissue maintained at the elevated temperature, but that the incidence (maximally 0.14 per epithelial cell) was significantly lower than that in colchitine-treated animals (maximally 1.1 per epithelial cell) and probably. too low to account for the very rapid cell loss occurring in sf-1. These observations suggested that interstitial cell lineage elimination in sf-1 took place in part by phagocytosis and in part by other yetunidentified mechanisms.
VOLUME 126.1988
volume of the acetic acid-glycerol macerating solution and macerated according to David (1973). A fraction of the dissociated cell suspension was then spread evenly on a slide glass and allowed to dry in air, and the area on which the cells were spread was measured. The number of each cell type was counted per unit area on the slide glass under a phase-contrast microscope. Cell type identification was done according to David (1973), and a total of at least 500 epithelial cells were counted per sample. The results obtained were used to compute the average number of each cell type per polyp and the relative ratio of each cell type to epithelial cells. Phagoqtosis
MATERIALS
AND METHODS
Strains and Culture Two strains belonging to H. magnipapillata were used. Strain 105 is the standard wild-type strain. Strain sf-1 is a mutant strain which contains temperaturesensitive interstitial cells. This strain grows normally at the permissive temperature of WC, but it loses its interstitial cell lineage when its culture temperature is shifted up to 23°C or higher (Sugiyama and Fujisawa, 19’78;Mar-cum et al., 1980). Mature budding polyps of both strains were cultured under the rigorously controlled mass culture conditions described by Takano and Sugiyama (1983) at 18 + 0.5”C. Young polyps newly dropped from the mature polyps were collected daily and used for experiments. Temperature and Colchicine Treatment For temperature treatment, a group of 10 polyps was placed in a small plastic petri dish (50 mm in diameter), containing about 10 ml of the culture solution, and set in an incubator maintained at 27 4 0.5”C for varying periods of time. For colchicine treatment, 3-4 groups (30-40 polyps) were placed together in a plastic petri dish containing about 10 ml of 0.4% colchicine in the culture solution. These animals were maintained at 18°C and groups of 10 animals per sample were harvested after varying periods of time. For treatment lasting longer than 24 hr, the colchicine solution was replaced with a fresh one at 24 hr. Animals were not fed during temperature or colchitine treatment. Maceration and Cell Counting Immediately upon termination of temperature or colchicine treatment, hypostomes with tentacles and basal disks were removed from the 10 treated animals, and the remaining body columns were pooled in a small
To examine phagocytosis, the dissociated cells spread and air-dried on a slide glass were stained with Feulgen, counterstained with fast green, and the engulfed cells present in the phagocytic vacuoles of the epithelial cells were counted under a bright-field microscope according to the procedure of Bosch and David (1984). The total number of phagocytosed cells scored was divided by the total number of epithelial cells examined to obtain the frequency of phagocytosis. RESULTS
Elimination
of the Interstitial
Cell Lineage
Table 1 shows the effects of temperature and colchitine treatments on strain sf-1 and the standard wildtype strain 105. Young sf-1 polyps originally had an average of approximately 4.6 X lo4 cells per polyp, including approximately lo4 epithelial cells. Treatment at the elevated temperature of 27°C for 48 hr reduced this number to approximately 59% of the original value. This drop was produced largely by decreases in the number of big interstitial cells, little interstitial cells, and nematoblasts. These three cell types represented more than half of the total cells in the original animals but less than 5% in the treated animals. A similar reduction also took place in sf-1 animals with colchicine treatment for 48 hr at 18°C. Colchicine treatment, however, also produced some reduction in the number of nematocyte, gland, and nerve cells, whereas temperature treatment had little or no effect on these cell types. Control experiments carried out with the standard wild-type strain 105 showed that cell elimination took place with the colchicine treatment but not the temperature treatment (lower half of Table 1). These results are similar to those of previous studies by Marcum et al. (1980) and Campbell (1976). The time course of cell elimination in the sf-1 tissue is shown in Figs. 1 and 2. Figure 1 shows that different
TERADA, SUGIYAMA, AND SHIGENAKA
265
Cell Phngocytosis in Hydra
TABLE 1 EFFECTS OF TEMPERATURE AND COLCHICINE ON TISSUE COMPOSITION
Number of cells per polyp (X1000) f SD No. of determinations
Big
Nematoblast
Nematocyte
Gland
Nerve
Total
5.4 * 0.7
5.9 * 1.8
15.2 f 2.6
6.2 + 1.8
1.5 f 0.1
1.9 + 0.2
46.2 + 5.7
15.8 f 1.0
0.1 f 0.03
0
1.0 f 0.3
7.4 f 1.7
0.9 + 0.3
1.8 f 0.3
27.1 f
3.6
3
12.3 311.2
0.5 * 0.02
0
0.2 * 0.1
0.4 f 0.1
0.5 ?I 0.2
0.7 k 0.3
14.6 f
1.8
None” (18’C, 0 hr)
3
12.7 f 2.3
7.8 f 1.3
3.4 lk 0.3
18.1 + 2.3
4.5 k 0.7
1.9 + 0.3
2.7 iz 0.3
51.0 f
5.7
27°C (48 hr)
3
27.5 f 7.7
13.4 f 3.4
6.1 -t 0.7
28.7 k 6.0
8.0 _+2.3
4.0 + 1.2
5.1 t- 1.0
92.9 + 21.4
Colchicine (48 hr)
3
11.9 + 1.5
1.1 f 0.5
0.02 + 0.04
0.2 f 0.1
0.5 5 0.3
0.7 f 0.2
2.4 f 0.4
16.9 f
Treatment
Sf-1
None” (WC, 0 hr)
3
9.9 f 1.5
2PC (48 hr)
3
Colchicine (48 hr) 105
Little
interstitial
Strain
Epithelial
interstitial
1.9
a Original polyps collected from the stock culture.
cell types were lost from the tissue at different rates. Figure 2 shows that overall loss of the total interstitial cell lineage occurred at a faster rate in response to colchicine treatment (Fig. 2B) than to temperature treatment (Fig. 2A). The following three features should be noted in these two figures. (1) Cell d$erentiatim The nematoblast number increased slightly during the first 6 hr of temperature treatment (Fig. lC), indicating that nematoblast differentiation from little interstitial cells occurred actively during this period. The rate of this increase, computed from the data shown in Table 1 and Fig. lC, was about 0.05 cells per epithelial cell per hour. This indicated that the rapid reduction in the number of little interstitial cells shown in Fig. 1B (0.08 cells per epithelial cell per hour) was achieved in part (about 60%) by this differentiation process. The rest was presumably achieved by cell elimination. As shown by this example, cell differentiation may have also affected the observed rates of loss of other cell types. This, however, could not be assessed from the available results. In contrast to this, the loss of total interstitial cell lineage shown in Fig. 2 was not affected by cell differentiation since cell differentiation affected the cell types but not the total number of cells involved. (2) Division of interstitial cell lineage. Since cell differentiation occurred in the early part of temperature treatment, cells of the interstitial cell lineage and having mitotic activity may have divided to some extent in a similar period. If this occurred, the number of cells actually lost from the tissue was greater than indicated in Figs. 1 and 2. However, this also could not be assessed from the available data.
(3) Division of epithelial cells. Epithelial cells of sf-1 proliferate at 18°C (Takano et al., 1980) and at 27’C (Table 1) at a similar rate, producing about 23% more epithelial cells per day. This has the effect of reducing the number of other cell types per epithelial cell by about 19% per day even when their absolute number remains unchanged. This has a significant effect on the cell types which decrease slowly. For example, nerve cell differentiation from interstitial stem cells must have ceased when the latter cell type was eliminated within the initial 12 hr of treatment (Figs. 1A and 1B). During the next day, the nerve cell number decreased at the rate of about 40% per day (Fig. 1E). About half of this decrease is attributable to epithelial cell proliferation. Similarly, about one-third of the decrease of the total interstitial cell lineage at 27°C (Fig. 2A) is also attributable to the epithelial cell proliferation. Epithelial cells apparently did not proliferate in the colchicine-treated tissue (Table 1) and therefore did not affect the rate of cell loss (Figs. lG-1L and 2B). It is clear from these observations that the data shown in Figs. 1 and 2 should be regarded not as precise but only as rough estimates of cell elimination rates. Nevertheless, these figures show some important features. First, big and little interstitial cells were largely eliminated from the temperature-treated tissue within the first 12-16 hr of treatment (Figs. 1A and lB), whereas decreases in other cell types from the same tissue occurred largely after this period (Figs. lC-1F). Second, nematocytes decreased very slowly in response to temperature treatment (Fig. lD), but decreased rapidly in response to colchicine treatment (Fig. 15). Third,
266
DEVELOPMENTALBIOLOGY
1 .
VOLUME126,1988
respectively. In the temperature-treated tissue, phagocytosis occurred mostly from the third to the eighth hour of treatment, forming a small peak with the highest point at about the fourth hour. Thereafter, no significant level of phagocytosis was observed (Fig. 2A). In contrast, phagocytosis was observed at significant frequencies throughout the entire period of treatment in the colchicine-treated tissue (Fig. 2B). It should be pointed out here that the actual levels of phagocytosis may have been somewhat higher than those shown in Fig. 2. This was suggested by the presence of partially degraded cells in the macerationFeulgen preparations made from the treated animals (Figs. 3F and 3G). These cells had condensed and strongly Feulgen-positive nuclei and resembled the cells present in the phagocytic vacuoles of epithelial cells. The origin of these cells was uncertain (see Discussion). One possibility is that they were artificially produced from the phagocytosed cells by the breakup of the epithelial cells containing them. Whether or not such a breakup did occur to any significant extent during the maceration procedure is not certain. If it did,
(D) ne~tq;yteA
7
0 , 0
27°C 24
I 48 0
24
48
n
a .
. . -i
. Time
of treatment
- 1
(A) 27%
.
(hour)
z ‘L-,
FIG. 1. Elimination of various cell types from sf-1 tissue. The abscissa shows time after transfer of animals from 18 to 27°C (A-F) or from the normal to the colchicine-containing culture solution (G-L). The ordinate shows the ratio of each cell type to epithelial cells, normalized to the values of the original animals before treatment. The pooled results of three independent experiments are shown using different symbols for each experiment.
-0.
\ .
-C
13
(6) Colchicine
elimination of total interstitial cell lineage occurred largely within the initial 12 hr of colchicine treatment (Fig. 2B), whereas it occurred throughout the entire period of temperature treatment (Fig. 2A).
- 1 0
-0,
Phagocytosis Typical examples of the sf-1 epithelial cells containing phagocytic vacuoles in their cytoplasm are presented in Fig. 3. These phagocytotic structures observed in the present study appeared to be similar in morphology and staining property to those previously observed by Bosch and David (1984) and by Kobatake and Sugiyama (1986). The frequency of phagocytosis (number of phagocytosed cells per epithelial cell) observed in sf-1 tissue is shown by the open symbols in Fig. 2. It was generally much higher in colchicine-treated tissue (Fig. 2B) than in temperature-treated tissue (Fig. 2A). It was maximally about 1.1 and 0.14 in the former and latter tissues,
L
,
24
0 Time
of treatment
)- (
48 (hour)
FIG. 2. Cell elimination and phagocytosis. The abscissa shows time of temperature treatment (A) and colchicine treatment (B). The ordinate on the left (closed symbols) shows the total number of cells belonging to the interstitial cell lineage (big and little interstitial cells, nematoblasts, nematocytes, nerve cells, and gland cells) per epithelial cell. The ordinate on the right (open symbols) shows the number of phagocytosed cells per epithelial cell.
267
FIG. 3. Epithelial cells containing engulfed cells in their phagocytic vacuoles (A-E) and partially degraded cells (F, G). (A, B) Ectodermal and endodermal epithelial cells, respectively, from temperature-treated sf-1 tissue. (C, D) Ectodermal epithelial cells and (E) an endodermal epithelial cell from colchicine-treated sf-1 tissue. Phagocytic vacuoles are indicated by arrows, and the nuclei of the host cells are indicated by n. The cell shown in (C) has two engulfed cells within one vacuole, the cell shown in (D) and (E) has three vacuoles, and the cell shown in (E) contains an engulfed stenotele capsule (middle arrow). (F) and (G) show partially degraded cells from temperature- and colchicine-treated sf-1 tissues, respectively. These cells resemble the engulfed cells present in the vacuoles, but appear to occur outside of the epithelial cells (see main text). The dissociated cells were stained with Feulgen, counterstained with fast green, and examined by bright-field microscopy (see Materials and Methods). The bar represents 25 pm.
then the actual incidence of phagocytosis would be somewhat higher than that shown in Fig. 2. Estimation of Time Necessary for Phagocytosis The cells engulfed by epithelial cells were presumably digested and disintegrated completely by the digestive enzyme system of the phagocytic vacuoles. If one assumes this to be true, and that this process alone is solely responsible for interstitial cell lineage removal, the average length of time the engulfed cells remained in the phagocytic vacuoles “phagocytic duration” could be calculated from the results shown in Fig. 2. For example, at the fourth hour of temperature treatment, there were 3.4 cells belonging to the interstitial cell lineage per epithelial cell (Fig. 2A). During the next 4 hr, this number dropped to 2.3, giving an average drop rate of 0.23 cells per hour per epithelial cell. Since about one-third of this decrease was produced by epithelial proliferation (see preceding section), the actual rate of
decrease by cell elimination was approximately 0.19. At the midpoint of this time period (sixth hour), the frequency of phagocytosis was 0.15 per epithelial cell. Dividing this value by the hourly elimination rate of 0.19 gives 0.8 hr as the average phagocytic duration for this period. Phagocytic duration calculated in this manner for the entire period of temperature treatment is presented in Fig. 4A. It shows that it fluctuated significantly during the early part of treatment (O-16 hr). This fluctuation was probably produced by errors made in estimating cell elimination and/or phagocytosis rates described in the preceding sections. The most important source of error was probably the division of interstitial cell lineage occurring in the early part of temperature treatment but not accounted for in Fig. 2A. This error, if present, would produce an overestimation of phagocytic duration (Fig. 4A). In spite of the inaccuracy and fluctuation of the results, Fig. 4A shows one very important feature. The
268
DEVELOPMENTALBIOLOGY
calculated time for complete digestion and disintegration of the phagocytosed cells was 25-200 min in the early period (O-16 hr) of temperature treatment. However, it became very short (10 min or less) after this period. Figure 4B shows phagocytic duration in the colchitine-treated animals calculated in the same manner, except that no correction was made for epithelial cell proliferation. It shows that the phagocytic duration in the colchicine-treated tissue was significantly longer than that in the temperature-treated tissue (Fig. 4A). (Note the difference in the scale in Figs. 4A and 4B.) It was about 1 hr initially, became gradually longer, and reached the calculated value of over 100 hr late in treatment. It is interesting to note in this respect that cell elimination in the colchicine-treated animals took place mostly in the initial 12 hr and little after this period (Fig. 2B), but that phagocytosis incidence remained high even late in the treatment (see Discussion). 0
Both ectodermal and endodermal epithelial cells have phagocytic capacity (Campbell, 1976) (Fig. 3). Figure 5
1
c
24 Time of treotment
48 (hour)
FIG. 5. Relative phagocytic activity of the ectodermal and endoderma1 epithelial cells. The abscissa represents time of temperature (A) and colchicine (B) treatments. The ordinate represents the proportion of phagocytosed cells present in ectodermal epithelial cells. Proportions greater (smaller) than 0.5 indicate that more phagocytosis occurs with ectodermal (endodermal) epithelial cells. Asterisks indicate points where few phagocytic cells were found (less than 15, see Fig. 2A). However, the pooled results of these three time points (7132) indicate that the proportion was generally low late in treatment.
shows the results of comparing the relative phagocytic activity of the two epithelial cell types in sf-1. It shows that phagocytosis was more active in the ectodermal than in the endodermal epithelial cells in the early part of both temperature (Fig. 5A) and colchicine treatments (Fig. 5B). This, however, gradually changed and the situation was reversed later in both types of treatment.
(8) Colchicine
0
(6) Colchicine
L
Relative Phagocytic Activity of Ectodermal and Endodermal Epithelial Cells
J
(A) 27°C
DISCUSSION
Colchicine Treatment
24 Time of ireotment
48 (hour)
FIG. 4. Calculated time that engulfed cells remained in phagocytic vacuoles. The abscissa represents time of temperature (A) and col-
chicine (B) treatments. The ordinate represents the average length of time the phagocytosed cells remain in the phagocytic vacuoles from engulfment to complete digestion and disintegration. This was calculated from results shown in Fig. 2 as explained in the main text.
In response to colchicine treatment, the interstitial cell lineage is eliminated rapidly through phagocytosis in If. Attenuate (Campbell, 1976) and in strain sf-1 (present study). An interesting aspect was revealed in this process. The average length of time necessary for digestion of the phagocytosed cells (phagocytic duration) was about 1 hr soon after the start of the colchicine treatment. However, it became gradually longer later in treatment, reaching more than 100 hr (Fig. 4B). This suggests that
TERADA, SUGIYAMA,AND SHIGENAKA
the epithelial cells in tissue treated with colchicine for a long time are able to engulf other cells but unable to digest them promptly. Colchicine has apparently little effect on the engulfing process but has a delayed effect on the digesting process. Similar effects of colchicine were described in other organisms (Freed and Lebowitz, 1970; Bhisey and Freed, 1971). Temperature Treatment Phagocytosis was also found to occur in sf-1 tissue maintained at an elevated temperature (Figs. lA-1F and 3). This phagocytosis was undetected in a previous study by Marcum et al. (1980), probably because they used electron microscopy and did not examine a large enough number of epithelial cells to detect the low incidence of phagocytosis. An important question is whether or not phagocytosis is also primarily responsible for the elimination of the interstitial cell lineage at an elevated temperature in sf-1. To answer this question, it is necessary to divide the 48 hr of temperature treatment into early and late phases, and discuss the events in these two phases separately. In the early phase which lasted for 12 to 16 hr from the start of the temperature shift-up, the majority of big and little interstitial cells were eliminated, but the majority of the other cell types remained intact (Figs. lA-1F). The phagocytic duration calculated for this period fluctuated significantly, ranging from 25 to 200 min and giving an average of about 60 min (Fig. 4A). This average value is similar to the duration in starved hydra (l-2 hr, maximally 3 hr) estimated by Bosch and David (1984). It therefore appears reasonable to assume that a large part of cell elimination was achieved through phagocytosis in this phase. In the late phase starting 12 to 16 hr after the temperature change, the majority of nematoblasts and some nematocytes, nerve cells, and gland cells were eliminated. However, the incidence of phagocytosis was very low, and phagocytic duration calculated for this period was very short, particularly late in the phase (10 min or less). This value may be too short for complete digestion and disintegration of the engulfed cells. An obvious explanation for this is to assume that phagocytosis alone is not solely responsible for cell elimination and another, as yet undetected, mechanism is also involved. For example, if the actual phagocytic duration is 30 (or 60) min, only about one-third (or one-sixth) of the cell elimination is carried out by cell phagocytosis in this phase. The rest, then, must be carried out by some other mechanism.
Cell Phagocytosis in Hydra
269
The identity of such a mechanism is uncertain at present. On the basis of electron microscopic observation, Marcum et al. (1980) suggested that self-disintegration by necrosis may play an important role in cell elimination in sf-1. Another possibility is the extrusion of the partially degraded (or normal) cells into the gastric cavity. The existence of such a process was suggested in the late stage of colchicine treatment (Fig. 6C in Campbell, 1976). The extrusion of wasted cells into extracellular milieu is known to occur in amphibians, mammals, and insects (Beaulaton and Lockshin, 1982). Whether or not such mechanisms are responsible for the interstitial cell lineage elimination in sf-1 at the restrictive temperature remains to be investigated. The authors are greatly indebted to Dr. R. D. Campbell for his critical reading of the manuscript. REFERENCES
BEAULATON, J., and LOCKSHIN, R. A. (1982). The relation of programmed cell death to development and reproduction: Comparative studies and an attempt at classification. Int. Rev. @to!. 79,215-235. BHISEY, A. N., and FREED, J. J. (1971). Altered movement of endosomes in colchicine-treated cultured macrophages. Exp. Cell Res. 64, 430-438. BOSCH,T. C. G., and DAVID, C. N. (1984). Growth regulation in Hydra: Relationship between epithelial cell cycle length and growth rate. Dev. Biol. 104,161-171. BOSCH, T. C. G., and DAVID, C. N. (1986). Immunocompetence in Hydra: Epithelial cells recognize self-nonself and react against it. J Exp. Zool. 238,225-234. CAMPBELL,R. D. (19’76).Elimination of Hydra interstitial and nerve cells by means of colchicine. J. CelESci. 21,1-13. DAVID, C. N. (1973). A quantitative method for maeeration of hydra tissue. Wilhelm Rax’s Arch. Dev. Biol. 171,259-268. FREED,J. J., and LEBOWITZ,M. E. (1970). The association of a class of saltatory movements with microtubules in cultured cells. J. Cell Biol. 45, 334-354. KOBATAKE,E., and SUGIYAMA,T. (1986). Genetic analysis of developmental mechanisms in hydra. XVII. Elimination of excess epithelial cells by phagocytosis in a mutant strain L4. Dev. Biol. 115, 249-255. MARCUM, B. A., and CAMPBELL,R. D. (1978). Development of hydra lacking nerve and interstitial cells. J. Cell Sci. 29,17-33. MARCUM,B., FUJISAWA,T., and SUGIYAMA,T. (1980). A mutant hydra strain (sf-1) containing temperature-sensitive interstitial cells. In “Developmental and Cellular Biology of Coelenterates” (P. Tardent and R. Tardent, Eds.), pp. 429-434. Elsevier/North-Holland, Amsterdam. SUGIYAMA,T., and FUJISAWA,T. (1978). Genetic analysis of developmental mechanisms in hydra. II. Isolation and characterization of an interstitial cell-deficient strain. J. Cell Sci. 29, 35-52. TAKANO, J., FUJISAWA,T., and SUGIYAMA,T. (1980). Growth and cell cycle of hydra. In “Developmental and Cellular Biology of Coelenterates” (P. Tardent and R. Tardent, Eds.), pp. 429-434. Elsevier/ North-Holland, Amsterdam. TAKANO,J., and SUGIYAMA,T. (1983). Genetic analysis of developmental mechanisms in hydra. VIII. Head-activation and head-inhibition potentials of slow-budding strain (L4). J. Embqol. Exp. Marphol. 78, 141-168.
BIOLOGY
126.263-269 (1988)
Genetic Analysis of Developmental Mechanisms in Hydra XVIII. Mechanism for Elimination of the Interstitial Cell Lineage in the Mutant Strain Sf-1 HIROYUKI
TERADA,
TSUTOMU SUGIYAMA,*
AND YOSHINOBU SHIGENAKA
Faculty of Integrated Arts and Sciences, Hiroshima University, l-l-89 Higashi-Sendamachi, Naka-ku, Hiroshima 7.90,Japan; and *Laboratory of Developmental Genetics, National Institute of Genetics, 1111 Yata, Mishimu 411, Japan Accepted November 20, 1987 The interstitial cell lineage in mutant strain sf-1 of hydra is temperature sensitive and is lost rapidly from tissue when the animal is cultured at a restrictive temperature of 23’C or higher. The mechanism responsible for this cell elimination process was investigated. Sf-1 polyps were treated at a restrictive temperature of 27°C for varying lengths of time, their tissues were macerated, and the resultant dissociated cells were examined for evidence of phagocytosis after Feulgen staining. It was found that large phagocytic vacuoles were present in the cytoplasm of some epithelial cells. These vacuoles contained partially degraded cells, whose nuclei had highly-condensed and intensely Feulgen-positive chromatin granules. This indicated that, as in colchicine-treated (Campbell, 1976) or starved (Bosch and David, 1984) wild-type hydra, the epithelial cells in strain sf-1 engulfed and disintegrated other cells in the phagocytic vacuoles. The incidence of phagocytosis was higher in sf-1 tissue maintained at elevated temperature than in sf-I tissue maintained at normal temperature. However, the observed incidence was relatively low (maximally 0.14 phagocytosed cells per epithelial cell) and appeared to be too low to account for the very rapid interstitial cell loss occurring in this strain. We concluded that elimination of the interstitial cell lineage at a restrictive temperature in strain sf-I takes place in part by phagocytosis and in part by other yet-unidentified mechanisms (~3, Marcum et al, 1980). o 1988 Academic Press, Inc.
INTRODUCTION
Epithelial cells in hydra have the ability to engulf and eliminate other cells in tissue. This was initially demonstrated in Hydra attenuata by Campbell (1976). In response to colchicine treatment, interstitial cells and their differentiation products (nematoblasts, nematocytes, nerve cells, and gland cells) disappeared from the tissue at varying rates. Examination by light and electron microscopy revealed that this cell loss occurred through phagocytosis by both ectodermal and endodermal epithelial cells. Recently, a simple and rapid procedure to monitor cell phagocytosis in hydra was developed by Bosch and David (1984). When hydra tissue was macerated and the resultant dissociated cells were examined after Feulgen staining, the engulfed cells present in the phagocytic vacuoles of epithelial cells were readily identified by their distinct staining and morphological characteristics. Using this procedure, Bosch and David showed that extensive cell elimination by phagocytosis occurred in the tissue of starved H. attenuata (Bosch and David, 1984), and also at the junction of two tissues when two different species of hydra were grafted together (Bosch and David, 1986). The same procedure was also adopted by Kobatake and Sugiyama (1986) to examine phagocytosis occurring in a mutant strain of Hydra magnipapillata (L4) which produced “excess” epithelial cells. 263
In the present study, we adopted the same procedure to examine phagocytosis in another mutant strain, sf-1 (Sugiyama and Fujisawa, 1978; Marcum et ak, 1980). This mutant strain, whose interstitial cell lineage is temperature sensitive, grows normally when cultured at 18°C. However, it rapidly loses its interstitial cell lineage when the culture temperature is shifted up to 23°C (or higher). When maintained at this temperature for a week or longer, it loses all its interstitial cells and practically all of their differentiation products (nematoblasts, nematocytes, nerve cells, and gland cells) and turns into “epithelial hydra” which lack the ability to move or feed (Marcum and Campbell, 1978; Sugiyama and Fujisawa, 1978). At present, the mechanism for this cell elimination in sf-1 is not known. In a previous investigation, Marcum et al. (1980) used electron microscopy to examine sf-1 tissue maintained at an elevated temperature, but failed to detect phagocytosis. This, however, did not necessarily rule out the possibility of phagocytosis in sf-1 since the number of cells examined by electron microscopy was limited, and a low incidence of phagocytosis may have been undetected. To reexamine this possibility, sf-1 animals were treated at an elevated temperature for varying periods of time, and the rate of interstitial cell lineage elimination in their tissues was determined. Using the same tissues, the incidence of cell phagocytosis was systemat0012-1606/88 $3.00 Copyright All rights
0 1988 by Academic Press, Inc. of reproduction in any form reserved.
264
DEVELOPMENTAL BIOLOGY
ically examined by the maceration-Feulgen procedure of Bosch and David (1984). Similar analyses were also carried out in parallel for sf-1 animals treated with colchicine. The experiments revealed that cell phagocytosis occurred in sf-1 tissue maintained at the elevated temperature, but that the incidence (maximally 0.14 per epithelial cell) was significantly lower than that in colchitine-treated animals (maximally 1.1 per epithelial cell) and probably. too low to account for the very rapid cell loss occurring in sf-1. These observations suggested that interstitial cell lineage elimination in sf-1 took place in part by phagocytosis and in part by other yetunidentified mechanisms.
VOLUME 126.1988
volume of the acetic acid-glycerol macerating solution and macerated according to David (1973). A fraction of the dissociated cell suspension was then spread evenly on a slide glass and allowed to dry in air, and the area on which the cells were spread was measured. The number of each cell type was counted per unit area on the slide glass under a phase-contrast microscope. Cell type identification was done according to David (1973), and a total of at least 500 epithelial cells were counted per sample. The results obtained were used to compute the average number of each cell type per polyp and the relative ratio of each cell type to epithelial cells. Phagoqtosis
MATERIALS
AND METHODS
Strains and Culture Two strains belonging to H. magnipapillata were used. Strain 105 is the standard wild-type strain. Strain sf-1 is a mutant strain which contains temperaturesensitive interstitial cells. This strain grows normally at the permissive temperature of WC, but it loses its interstitial cell lineage when its culture temperature is shifted up to 23°C or higher (Sugiyama and Fujisawa, 19’78;Mar-cum et al., 1980). Mature budding polyps of both strains were cultured under the rigorously controlled mass culture conditions described by Takano and Sugiyama (1983) at 18 + 0.5”C. Young polyps newly dropped from the mature polyps were collected daily and used for experiments. Temperature and Colchicine Treatment For temperature treatment, a group of 10 polyps was placed in a small plastic petri dish (50 mm in diameter), containing about 10 ml of the culture solution, and set in an incubator maintained at 27 4 0.5”C for varying periods of time. For colchicine treatment, 3-4 groups (30-40 polyps) were placed together in a plastic petri dish containing about 10 ml of 0.4% colchicine in the culture solution. These animals were maintained at 18°C and groups of 10 animals per sample were harvested after varying periods of time. For treatment lasting longer than 24 hr, the colchicine solution was replaced with a fresh one at 24 hr. Animals were not fed during temperature or colchitine treatment. Maceration and Cell Counting Immediately upon termination of temperature or colchicine treatment, hypostomes with tentacles and basal disks were removed from the 10 treated animals, and the remaining body columns were pooled in a small
To examine phagocytosis, the dissociated cells spread and air-dried on a slide glass were stained with Feulgen, counterstained with fast green, and the engulfed cells present in the phagocytic vacuoles of the epithelial cells were counted under a bright-field microscope according to the procedure of Bosch and David (1984). The total number of phagocytosed cells scored was divided by the total number of epithelial cells examined to obtain the frequency of phagocytosis. RESULTS
Elimination
of the Interstitial
Cell Lineage
Table 1 shows the effects of temperature and colchitine treatments on strain sf-1 and the standard wildtype strain 105. Young sf-1 polyps originally had an average of approximately 4.6 X lo4 cells per polyp, including approximately lo4 epithelial cells. Treatment at the elevated temperature of 27°C for 48 hr reduced this number to approximately 59% of the original value. This drop was produced largely by decreases in the number of big interstitial cells, little interstitial cells, and nematoblasts. These three cell types represented more than half of the total cells in the original animals but less than 5% in the treated animals. A similar reduction also took place in sf-1 animals with colchicine treatment for 48 hr at 18°C. Colchicine treatment, however, also produced some reduction in the number of nematocyte, gland, and nerve cells, whereas temperature treatment had little or no effect on these cell types. Control experiments carried out with the standard wild-type strain 105 showed that cell elimination took place with the colchicine treatment but not the temperature treatment (lower half of Table 1). These results are similar to those of previous studies by Marcum et al. (1980) and Campbell (1976). The time course of cell elimination in the sf-1 tissue is shown in Figs. 1 and 2. Figure 1 shows that different
TERADA, SUGIYAMA, AND SHIGENAKA
265
Cell Phngocytosis in Hydra
TABLE 1 EFFECTS OF TEMPERATURE AND COLCHICINE ON TISSUE COMPOSITION
Number of cells per polyp (X1000) f SD No. of determinations
Big
Nematoblast
Nematocyte
Gland
Nerve
Total
5.4 * 0.7
5.9 * 1.8
15.2 f 2.6
6.2 + 1.8
1.5 f 0.1
1.9 + 0.2
46.2 + 5.7
15.8 f 1.0
0.1 f 0.03
0
1.0 f 0.3
7.4 f 1.7
0.9 + 0.3
1.8 f 0.3
27.1 f
3.6
3
12.3 311.2
0.5 * 0.02
0
0.2 * 0.1
0.4 f 0.1
0.5 ?I 0.2
0.7 k 0.3
14.6 f
1.8
None” (18’C, 0 hr)
3
12.7 f 2.3
7.8 f 1.3
3.4 lk 0.3
18.1 + 2.3
4.5 k 0.7
1.9 + 0.3
2.7 iz 0.3
51.0 f
5.7
27°C (48 hr)
3
27.5 f 7.7
13.4 f 3.4
6.1 -t 0.7
28.7 k 6.0
8.0 _+2.3
4.0 + 1.2
5.1 t- 1.0
92.9 + 21.4
Colchicine (48 hr)
3
11.9 + 1.5
1.1 f 0.5
0.02 + 0.04
0.2 f 0.1
0.5 5 0.3
0.7 f 0.2
2.4 f 0.4
16.9 f
Treatment
Sf-1
None” (WC, 0 hr)
3
9.9 f 1.5
2PC (48 hr)
3
Colchicine (48 hr) 105
Little
interstitial
Strain
Epithelial
interstitial
1.9
a Original polyps collected from the stock culture.
cell types were lost from the tissue at different rates. Figure 2 shows that overall loss of the total interstitial cell lineage occurred at a faster rate in response to colchicine treatment (Fig. 2B) than to temperature treatment (Fig. 2A). The following three features should be noted in these two figures. (1) Cell d$erentiatim The nematoblast number increased slightly during the first 6 hr of temperature treatment (Fig. lC), indicating that nematoblast differentiation from little interstitial cells occurred actively during this period. The rate of this increase, computed from the data shown in Table 1 and Fig. lC, was about 0.05 cells per epithelial cell per hour. This indicated that the rapid reduction in the number of little interstitial cells shown in Fig. 1B (0.08 cells per epithelial cell per hour) was achieved in part (about 60%) by this differentiation process. The rest was presumably achieved by cell elimination. As shown by this example, cell differentiation may have also affected the observed rates of loss of other cell types. This, however, could not be assessed from the available results. In contrast to this, the loss of total interstitial cell lineage shown in Fig. 2 was not affected by cell differentiation since cell differentiation affected the cell types but not the total number of cells involved. (2) Division of interstitial cell lineage. Since cell differentiation occurred in the early part of temperature treatment, cells of the interstitial cell lineage and having mitotic activity may have divided to some extent in a similar period. If this occurred, the number of cells actually lost from the tissue was greater than indicated in Figs. 1 and 2. However, this also could not be assessed from the available data.
(3) Division of epithelial cells. Epithelial cells of sf-1 proliferate at 18°C (Takano et al., 1980) and at 27’C (Table 1) at a similar rate, producing about 23% more epithelial cells per day. This has the effect of reducing the number of other cell types per epithelial cell by about 19% per day even when their absolute number remains unchanged. This has a significant effect on the cell types which decrease slowly. For example, nerve cell differentiation from interstitial stem cells must have ceased when the latter cell type was eliminated within the initial 12 hr of treatment (Figs. 1A and 1B). During the next day, the nerve cell number decreased at the rate of about 40% per day (Fig. 1E). About half of this decrease is attributable to epithelial cell proliferation. Similarly, about one-third of the decrease of the total interstitial cell lineage at 27°C (Fig. 2A) is also attributable to the epithelial cell proliferation. Epithelial cells apparently did not proliferate in the colchicine-treated tissue (Table 1) and therefore did not affect the rate of cell loss (Figs. lG-1L and 2B). It is clear from these observations that the data shown in Figs. 1 and 2 should be regarded not as precise but only as rough estimates of cell elimination rates. Nevertheless, these figures show some important features. First, big and little interstitial cells were largely eliminated from the temperature-treated tissue within the first 12-16 hr of treatment (Figs. 1A and lB), whereas decreases in other cell types from the same tissue occurred largely after this period (Figs. lC-1F). Second, nematocytes decreased very slowly in response to temperature treatment (Fig. lD), but decreased rapidly in response to colchicine treatment (Fig. 15). Third,
266
DEVELOPMENTALBIOLOGY
1 .
VOLUME126,1988
respectively. In the temperature-treated tissue, phagocytosis occurred mostly from the third to the eighth hour of treatment, forming a small peak with the highest point at about the fourth hour. Thereafter, no significant level of phagocytosis was observed (Fig. 2A). In contrast, phagocytosis was observed at significant frequencies throughout the entire period of treatment in the colchicine-treated tissue (Fig. 2B). It should be pointed out here that the actual levels of phagocytosis may have been somewhat higher than those shown in Fig. 2. This was suggested by the presence of partially degraded cells in the macerationFeulgen preparations made from the treated animals (Figs. 3F and 3G). These cells had condensed and strongly Feulgen-positive nuclei and resembled the cells present in the phagocytic vacuoles of epithelial cells. The origin of these cells was uncertain (see Discussion). One possibility is that they were artificially produced from the phagocytosed cells by the breakup of the epithelial cells containing them. Whether or not such a breakup did occur to any significant extent during the maceration procedure is not certain. If it did,
(D) ne~tq;yteA
7
0 , 0
27°C 24
I 48 0
24
48
n
a .
. . -i
. Time
of treatment
- 1
(A) 27%
.
(hour)
z ‘L-,
FIG. 1. Elimination of various cell types from sf-1 tissue. The abscissa shows time after transfer of animals from 18 to 27°C (A-F) or from the normal to the colchicine-containing culture solution (G-L). The ordinate shows the ratio of each cell type to epithelial cells, normalized to the values of the original animals before treatment. The pooled results of three independent experiments are shown using different symbols for each experiment.
-0.
\ .
-C
13
(6) Colchicine
elimination of total interstitial cell lineage occurred largely within the initial 12 hr of colchicine treatment (Fig. 2B), whereas it occurred throughout the entire period of temperature treatment (Fig. 2A).
- 1 0
-0,
Phagocytosis Typical examples of the sf-1 epithelial cells containing phagocytic vacuoles in their cytoplasm are presented in Fig. 3. These phagocytotic structures observed in the present study appeared to be similar in morphology and staining property to those previously observed by Bosch and David (1984) and by Kobatake and Sugiyama (1986). The frequency of phagocytosis (number of phagocytosed cells per epithelial cell) observed in sf-1 tissue is shown by the open symbols in Fig. 2. It was generally much higher in colchicine-treated tissue (Fig. 2B) than in temperature-treated tissue (Fig. 2A). It was maximally about 1.1 and 0.14 in the former and latter tissues,
L
,
24
0 Time
of treatment
)- (
48 (hour)
FIG. 2. Cell elimination and phagocytosis. The abscissa shows time of temperature treatment (A) and colchicine treatment (B). The ordinate on the left (closed symbols) shows the total number of cells belonging to the interstitial cell lineage (big and little interstitial cells, nematoblasts, nematocytes, nerve cells, and gland cells) per epithelial cell. The ordinate on the right (open symbols) shows the number of phagocytosed cells per epithelial cell.
267
FIG. 3. Epithelial cells containing engulfed cells in their phagocytic vacuoles (A-E) and partially degraded cells (F, G). (A, B) Ectodermal and endodermal epithelial cells, respectively, from temperature-treated sf-1 tissue. (C, D) Ectodermal epithelial cells and (E) an endodermal epithelial cell from colchicine-treated sf-1 tissue. Phagocytic vacuoles are indicated by arrows, and the nuclei of the host cells are indicated by n. The cell shown in (C) has two engulfed cells within one vacuole, the cell shown in (D) and (E) has three vacuoles, and the cell shown in (E) contains an engulfed stenotele capsule (middle arrow). (F) and (G) show partially degraded cells from temperature- and colchicine-treated sf-1 tissues, respectively. These cells resemble the engulfed cells present in the vacuoles, but appear to occur outside of the epithelial cells (see main text). The dissociated cells were stained with Feulgen, counterstained with fast green, and examined by bright-field microscopy (see Materials and Methods). The bar represents 25 pm.
then the actual incidence of phagocytosis would be somewhat higher than that shown in Fig. 2. Estimation of Time Necessary for Phagocytosis The cells engulfed by epithelial cells were presumably digested and disintegrated completely by the digestive enzyme system of the phagocytic vacuoles. If one assumes this to be true, and that this process alone is solely responsible for interstitial cell lineage removal, the average length of time the engulfed cells remained in the phagocytic vacuoles “phagocytic duration” could be calculated from the results shown in Fig. 2. For example, at the fourth hour of temperature treatment, there were 3.4 cells belonging to the interstitial cell lineage per epithelial cell (Fig. 2A). During the next 4 hr, this number dropped to 2.3, giving an average drop rate of 0.23 cells per hour per epithelial cell. Since about one-third of this decrease was produced by epithelial proliferation (see preceding section), the actual rate of
decrease by cell elimination was approximately 0.19. At the midpoint of this time period (sixth hour), the frequency of phagocytosis was 0.15 per epithelial cell. Dividing this value by the hourly elimination rate of 0.19 gives 0.8 hr as the average phagocytic duration for this period. Phagocytic duration calculated in this manner for the entire period of temperature treatment is presented in Fig. 4A. It shows that it fluctuated significantly during the early part of treatment (O-16 hr). This fluctuation was probably produced by errors made in estimating cell elimination and/or phagocytosis rates described in the preceding sections. The most important source of error was probably the division of interstitial cell lineage occurring in the early part of temperature treatment but not accounted for in Fig. 2A. This error, if present, would produce an overestimation of phagocytic duration (Fig. 4A). In spite of the inaccuracy and fluctuation of the results, Fig. 4A shows one very important feature. The
268
DEVELOPMENTALBIOLOGY
calculated time for complete digestion and disintegration of the phagocytosed cells was 25-200 min in the early period (O-16 hr) of temperature treatment. However, it became very short (10 min or less) after this period. Figure 4B shows phagocytic duration in the colchitine-treated animals calculated in the same manner, except that no correction was made for epithelial cell proliferation. It shows that the phagocytic duration in the colchicine-treated tissue was significantly longer than that in the temperature-treated tissue (Fig. 4A). (Note the difference in the scale in Figs. 4A and 4B.) It was about 1 hr initially, became gradually longer, and reached the calculated value of over 100 hr late in treatment. It is interesting to note in this respect that cell elimination in the colchicine-treated animals took place mostly in the initial 12 hr and little after this period (Fig. 2B), but that phagocytosis incidence remained high even late in the treatment (see Discussion). 0
Both ectodermal and endodermal epithelial cells have phagocytic capacity (Campbell, 1976) (Fig. 3). Figure 5
1
c
24 Time of treotment
48 (hour)
FIG. 5. Relative phagocytic activity of the ectodermal and endoderma1 epithelial cells. The abscissa represents time of temperature (A) and colchicine (B) treatments. The ordinate represents the proportion of phagocytosed cells present in ectodermal epithelial cells. Proportions greater (smaller) than 0.5 indicate that more phagocytosis occurs with ectodermal (endodermal) epithelial cells. Asterisks indicate points where few phagocytic cells were found (less than 15, see Fig. 2A). However, the pooled results of these three time points (7132) indicate that the proportion was generally low late in treatment.
shows the results of comparing the relative phagocytic activity of the two epithelial cell types in sf-1. It shows that phagocytosis was more active in the ectodermal than in the endodermal epithelial cells in the early part of both temperature (Fig. 5A) and colchicine treatments (Fig. 5B). This, however, gradually changed and the situation was reversed later in both types of treatment.
(8) Colchicine
0
(6) Colchicine
L
Relative Phagocytic Activity of Ectodermal and Endodermal Epithelial Cells
J
(A) 27°C
DISCUSSION
Colchicine Treatment
24 Time of ireotment
48 (hour)
FIG. 4. Calculated time that engulfed cells remained in phagocytic vacuoles. The abscissa represents time of temperature (A) and col-
chicine (B) treatments. The ordinate represents the average length of time the phagocytosed cells remain in the phagocytic vacuoles from engulfment to complete digestion and disintegration. This was calculated from results shown in Fig. 2 as explained in the main text.
In response to colchicine treatment, the interstitial cell lineage is eliminated rapidly through phagocytosis in If. Attenuate (Campbell, 1976) and in strain sf-1 (present study). An interesting aspect was revealed in this process. The average length of time necessary for digestion of the phagocytosed cells (phagocytic duration) was about 1 hr soon after the start of the colchicine treatment. However, it became gradually longer later in treatment, reaching more than 100 hr (Fig. 4B). This suggests that
TERADA, SUGIYAMA,AND SHIGENAKA
the epithelial cells in tissue treated with colchicine for a long time are able to engulf other cells but unable to digest them promptly. Colchicine has apparently little effect on the engulfing process but has a delayed effect on the digesting process. Similar effects of colchicine were described in other organisms (Freed and Lebowitz, 1970; Bhisey and Freed, 1971). Temperature Treatment Phagocytosis was also found to occur in sf-1 tissue maintained at an elevated temperature (Figs. lA-1F and 3). This phagocytosis was undetected in a previous study by Marcum et al. (1980), probably because they used electron microscopy and did not examine a large enough number of epithelial cells to detect the low incidence of phagocytosis. An important question is whether or not phagocytosis is also primarily responsible for the elimination of the interstitial cell lineage at an elevated temperature in sf-1. To answer this question, it is necessary to divide the 48 hr of temperature treatment into early and late phases, and discuss the events in these two phases separately. In the early phase which lasted for 12 to 16 hr from the start of the temperature shift-up, the majority of big and little interstitial cells were eliminated, but the majority of the other cell types remained intact (Figs. lA-1F). The phagocytic duration calculated for this period fluctuated significantly, ranging from 25 to 200 min and giving an average of about 60 min (Fig. 4A). This average value is similar to the duration in starved hydra (l-2 hr, maximally 3 hr) estimated by Bosch and David (1984). It therefore appears reasonable to assume that a large part of cell elimination was achieved through phagocytosis in this phase. In the late phase starting 12 to 16 hr after the temperature change, the majority of nematoblasts and some nematocytes, nerve cells, and gland cells were eliminated. However, the incidence of phagocytosis was very low, and phagocytic duration calculated for this period was very short, particularly late in the phase (10 min or less). This value may be too short for complete digestion and disintegration of the engulfed cells. An obvious explanation for this is to assume that phagocytosis alone is not solely responsible for cell elimination and another, as yet undetected, mechanism is also involved. For example, if the actual phagocytic duration is 30 (or 60) min, only about one-third (or one-sixth) of the cell elimination is carried out by cell phagocytosis in this phase. The rest, then, must be carried out by some other mechanism.
Cell Phagocytosis in Hydra
269
The identity of such a mechanism is uncertain at present. On the basis of electron microscopic observation, Marcum et al. (1980) suggested that self-disintegration by necrosis may play an important role in cell elimination in sf-1. Another possibility is the extrusion of the partially degraded (or normal) cells into the gastric cavity. The existence of such a process was suggested in the late stage of colchicine treatment (Fig. 6C in Campbell, 1976). The extrusion of wasted cells into extracellular milieu is known to occur in amphibians, mammals, and insects (Beaulaton and Lockshin, 1982). Whether or not such mechanisms are responsible for the interstitial cell lineage elimination in sf-1 at the restrictive temperature remains to be investigated. The authors are greatly indebted to Dr. R. D. Campbell for his critical reading of the manuscript. REFERENCES
BEAULATON, J., and LOCKSHIN, R. A. (1982). The relation of programmed cell death to development and reproduction: Comparative studies and an attempt at classification. Int. Rev. @to!. 79,215-235. BHISEY, A. N., and FREED, J. J. (1971). Altered movement of endosomes in colchicine-treated cultured macrophages. Exp. Cell Res. 64, 430-438. BOSCH,T. C. G., and DAVID, C. N. (1984). Growth regulation in Hydra: Relationship between epithelial cell cycle length and growth rate. Dev. Biol. 104,161-171. BOSCH, T. C. G., and DAVID, C. N. (1986). Immunocompetence in Hydra: Epithelial cells recognize self-nonself and react against it. J Exp. Zool. 238,225-234. CAMPBELL,R. D. (19’76).Elimination of Hydra interstitial and nerve cells by means of colchicine. J. CelESci. 21,1-13. DAVID, C. N. (1973). A quantitative method for maeeration of hydra tissue. Wilhelm Rax’s Arch. Dev. Biol. 171,259-268. FREED,J. J., and LEBOWITZ,M. E. (1970). The association of a class of saltatory movements with microtubules in cultured cells. J. Cell Biol. 45, 334-354. KOBATAKE,E., and SUGIYAMA,T. (1986). Genetic analysis of developmental mechanisms in hydra. XVII. Elimination of excess epithelial cells by phagocytosis in a mutant strain L4. Dev. Biol. 115, 249-255. MARCUM, B. A., and CAMPBELL,R. D. (1978). Development of hydra lacking nerve and interstitial cells. J. Cell Sci. 29,17-33. MARCUM,B., FUJISAWA,T., and SUGIYAMA,T. (1980). A mutant hydra strain (sf-1) containing temperature-sensitive interstitial cells. In “Developmental and Cellular Biology of Coelenterates” (P. Tardent and R. Tardent, Eds.), pp. 429-434. Elsevier/North-Holland, Amsterdam. SUGIYAMA,T., and FUJISAWA,T. (1978). Genetic analysis of developmental mechanisms in hydra. II. Isolation and characterization of an interstitial cell-deficient strain. J. Cell Sci. 29, 35-52. TAKANO, J., FUJISAWA,T., and SUGIYAMA,T. (1980). Growth and cell cycle of hydra. In “Developmental and Cellular Biology of Coelenterates” (P. Tardent and R. Tardent, Eds.), pp. 429-434. Elsevier/ North-Holland, Amsterdam. TAKANO,J., and SUGIYAMA,T. (1983). Genetic analysis of developmental mechanisms in hydra. VIII. Head-activation and head-inhibition potentials of slow-budding strain (L4). J. Embqol. Exp. Marphol. 78, 141-168.