Inhibition of stenotele commitment by an endogenous factor in Hydra

Nematocytes of hydra are stinging cells that are involved in the capture of prey, defence and locomotion. On the basis of the distinct morphology of the large complex structures (nematocysts) present in their cytoplasm, four types of nematocytes have been characterized in hydra: stenotele, desmoneme, holotrichous isorhiza and atrichous isorhiza (Weill, 1934). Nematocytes are differentiated from multipotent undifferentiated interstitial stem cells (Lehn, 1951; David & Gierer, 1974). Interstitial stem cells entering the nematocyte differentiation pathway undergo two to three rounds of synchronous cell division (Lehn, 1951; Rich & Tardent, 1969). The daughter cells remain connected to each other by cytoplasmic bridges to form clusters or nests of 4, 8, 16 and 32 cells (Slautterback & Fawcett, 1959). Following terminal cell division, all cells in a nest differentiate synchronously to produce the same type of capsule (Lehn, 1951; Rich & Tardent, 1969). After completion of capsule maturation, the nests break up and individual nematocytes migrate to the tentacles. Precursor cells are committed to differentiate into a specific capsule type near the S/G₂ boundary in the terminal cell cycle prior to capsule differentiation (Fujisawa & David, 1981, 1982).

Differentiation of stenotele nematocytes appears to be controlled by an endogenous factor that specifically inhibits stenotele differentiation, and which is distributed in a gradient from the head to foot (Fujisawa, 1987a,b). When additional heads are laterally grafted to the site of stenotele differentiation in the body column of hydra or when hydra are treated with this endogenous factor, the number of differentiating stenoteles decreases 4 days Printed in Great Britain © The Company of Biologists Limited 1988 after treatment (Fujisawa, 1987a,b). Since the time required for differentiation of stenotele nematocytes is about 3 days (David & Gierer, 1974; Fujisawa & Sugiyama, 1978; Fujisawa & David, 1982), it appears likely that precursor cells in the terminal cell cycle are affected by the factor.

In the present report, the precise timing of stenotele inhibition by an endogenous factor was determined within the terminal cell cycle of precursor cells. In addition, the fate of precursor cells that were prevented from entering the stenotele pathway by the factor was examined.

A standard wild-type strain (105) of Hydra magnipapillata (Sugiyama & Fujisawa, 1977) was cultured as described (Fujisawa, 1987a). All animals used in the present study were fed daily with freshly hatched brine shrimps.

Tentacles were removed from a polyp at their bases and the remaining tissue was macerated to analyse cell types, according to David (1973). Differentiating nematocytes scored in macerations using phase-contrast optics were post-mitotic cells in which developing nematocyst capsules were identifiable (Fujisawa & David, 1981).

Polyps were labelled with [³H]thymidine (20 μCiml⁻¹ of culture solution) by injecting 0 ·1 μl of the radioisotope directly into the gastric cavity through the mouth using fine-tipped polyethylene tubing. Labelled animals were macerated and cells were spread on gelatin-coated microscopic slides. After drying, the slides were washed briefly in water, dried and then dipped in Sakura NR-M2 autoradiographic emulsion. After 2-3 weeks of exposure at 4°C, the slides were developed and labelled cells were scored.

Nests of differentiating nematocytes were stained with a thiolacetic acid/lead nitrate reagent (David & Chailoner, 1974). The reagent specifically stains nematocyst capsules for a short period of time near the end of capsule development (David & Gierer, 1974; Fujisawa & David, 1981). Nests are intact at this stage and thus the nest size (the number of nematocytes per nest) and the nematocyte type can be determined in whole-mount preparations.

The crude extract of hydra was prepared as described (Fujisawa, 1987b). Animals showing the first sign of bud protrusions (stage II polyps) were homogenized in culture solution. The concentration of the homogenate was adjusted to 40 μg protein ml⁻¹ (1 stage II polyp ml⁻¹), which is the lowest concentration to give a maximum inhibition (Fujisawa, 1987b), and then centrifuged to remove cell debris at 1000gfor 10 min. The resulting supernatant was filtered through a membrane filter, UK-10 (Toyo Roshi, Japan), which effectively retains molecules larger than 10000M_r. The filtrate was used as crude extract.

Phagocytosed cells in epithelial cells were identified by the method of Fujisawa & David (1984). Polyps either untreated or treated with the stenotele inhibitor were macerated and cells were spread on gelatin-coated slides. After drying, the cells were stained using the Feulgen method with Fast Green counterstain (Pearse, 1960). Phagocytosed cells or cell fragments occur in phagocytic vacuoles in epithelial cells. Their condensed chromatin is stained intensely red by the Feulgen reagent and the cytoplasm pale green by the Fast Green.

In order to examine whether treatment with crude extract of hydra has any effect on the differentiation time of stenoteles, the differentiation kinetics of stenoteles as well as the other three types of nematocytes were examined in animals treated with extract. Groups of 10 newly detached buds (stage I polyps) were placed for 24 h in a plastic Petri dish (diam. 60 mm) containing 10 ml of crude extract. After incubation, animals were thoroughly washed, transferred to fresh culture solution containing no extract and further cultured. Animals were continuously labelled for 120 h with [³H]thymidine, starting at the same time as the onset of the treatment with extract (time 0). At various times, starting from 40 h, all animals in a dish were macerated and processed for autoradiography to score for labelled nematocytes. Control animals were continuously labelled with [³H]thymidine for the same period of time but not treated with extract. The results of such experiments are shown in Fig. 1. Two points are immediately noticeable: (1) no significant difference was detected between the control and treated animals in the differentiation kinetics of all nematocyte types; (2) no significant difference was detected in differentiation kinetics between desmonemes and atrichous isorhizas, and between stenoteles and holotrichous isorhizas. Labelled desmonemes and atrichous isorhizas started to appear at 48 h and the labelling indices increased rapidly to 100% during the next 12h. Thus, the differentiation time of desmonemes and atrichous isorhizas is 48 h from the end of the final S phase in the differentiation pathway and the time required for capsule development is 12h. Labelled stenoteles and holotrichous isorhizas started to appear at about 74 h, and the labelling indices increased rapidly up to about 50% during the next 24 h. Thereafter, the rate of labelling slowed down giving a biphasic labelling pattern. The rapidly labelled stenoteles and holotrichous isorhizas represent differentiating cells, most of which migrate to the tentacles upon completion of capsule development. The slowly labelled nematocytes are mature ones that remain in the body column and are mounted in ectodermal epithelial cells of the body column, and turn over slowly (David & Gierer, 1974). Thus, the differentiation time of stenoteles and holotrichous isorhizas is 74 h and the time required for capsule development is 24 h. These results agree with those previously reported (David & Gierer, 1974; Fujisawa & Sugiyama, 1978; Fujisawa & David, 1982).

In order to determine the precise timing of stenotele inhibition by crude extract, the kinetics of inhibition were examined in detail. Groups of 10 stage I polyps were treated for 24 h with crude extract, then washed and transferred to fresh culture solution containing no extract. At various times, starting 48 h after the beginning of the experiment, all treated animals as well as untreated control animals in a group were macerated together in order to determine the stenotele density (stenoteles/epithelial cells). The results are shown in Fig. 2. Stenotele density started to decrease at 72 h and reached its lowest level of about 0 ·02 at 96 h. After a short delay of 6 h or less, the stenotele density started to increase gradually and reached the control level at 126h. Considering the differentiation time of stenoteles from the end of final S phase to be 74h (Fig. 1), these results appear to suggest that precursor cells at the S/G₂ boundary in the terminal cell cycle responded to the extract and stenotele differentiation was blocked. The gradual decrease in stenotele density over a period of 24 h followed by a gradual increase over the same period of time was expected if one considers the following facts: (1) the development time of stenotele capsules detected by maceration was 24 h (David & Gierer, 1974; Fig. 1); (2) during 24 h of treatment with extract, a cohort of randomly cycling cells were blocked in their stenotele differentiation. It was also expected that longer treatment with crude extract would not affect the level of inhibition, but would prolong the time of the maximum level of inhibition. In order to examine this, groups of 10 stage I polyps were treated for 48 h with extract, washed and transferred to a fresh culture solution. Every 24h, starting 24 h after the transfer, animals were macerated to determine stenotele density. The results are shown in Fig. 3. Stenotele density started to decrease at 72 h and reached its lowest level of about 0·02 at 96h, at which it continued for 24h. It started to increase at 120h and returned to a level close to the control level at 144 h. The results indicate that longer treatment only increased the duration of maximum inhibition.

The results shown in Fig. 2 appear to show that the maximum inhibition of stenotele differentiation was of a short duration before it started to decrease. In order to estimate the length of the effect more precisely, the inhibition kinetics of stenotele differentiation were examined by changing the time of treatment with extract. Groups of 10 stage I polyps were treated with extract for 4, 6, 8, 12, 18 and 24 h, starting from time 0. All samples were macerated at 98 h to determine stenotele density. The results are shown in Fig. 4. The ordinate is expressed as percentage inhibition of stenotele differentiation: i.e. the percentage decrease in stenotele density in treated animals compared with untreated control animals sampled at 96 h in Fig. 2. The abscissa shows the length of treatment with extract. The level of inhibition increased linearly as treatment time increased (Fig. 4). The regression line intersects with the ordinate at 5·2% and at —3 h with the abscissa. These results along with those shown in Fig. 2 indicate that the inhibition of stenotele differentiation lasts 3-6 h longer than the actual

The effect of extract of hydra on nematocyte differentiation was also examined by quantifying the nests of differentiating nematocytes stained by using the thiolacetic acid/lead nitrate method (David & Challoner, 1974). Groups of 10 stage I polyps were treated with extract for 24 h, washed and transferred to fresh culture solutions containing no extract. Animals were sampled every 24 h, starting 48 h after the beginning of the treatment with extract and stained. Untreated control animals were also stained. The results are shown in Fig. 5. The number of nests for all four nematocyte types in control animals increased exponentially, with a doubling time of about 72 h (Fig. 5). In contrast, in treated animals, the nests did not increase exponentially in number. The number of stenotele nests decreased at 96 h with recovery close to the control level occurring at 120 h (Fig. 5A). The numbers of desmoneme and atrichous isorhiza nests increased slightly over the control level at 72h (Fig. 5B and D, respectively). The number of holotrichous isorhiza nests increased similarly in control and treated animals.

In animals treated with extract of hydra, the number of stenotele nests decreased at 96 h and the numbers of desmoneme and atrichous isorhiza nests increased at 72 h (Fig. 5). The differentiation time for stenoteles, 26h longer than that for desmonemes and atrichous isorhizas (Fig. 1), suggests that precursor cells shifted their differentiation pathways from stenoteles to other nematocytes. In order to test this quantitatively, the numbers of precursor nests (4s and 8s) blocked from entering the stenotele pathway during extract treatment were compared with the numbers above control levels that entered the desmoneme and atrichous isorhiza pathways during the same period of time. These numbers were estimated in the following way. First, the number of nests observed in Fig. 5 was divided into subclasses according to nest size (Table 1). Since nests of 8 and 16 nematocytes are derived directly from 4s and 8s, following mitosis (David & Gierer, 1974; Fujisawa & David, 1981), these numbers represent the numbers of 4s and 8s entering each differentiation pathway. Second, in order to examine which precursors are affected by the treatment, their differentiation pathways were reconstructed according to Fujisawa & David (1981) using the following parameters: the cell cycle parameters of 4s and 8s (12 h of S phase, 6 h of G₂ phase and virtually no G₁ phase; Campbell & David, 1974), the differentiation time and capsule development time of each nematocyte type (see Fig. 1) and the duration of the staining period (10 h; David & Gierer, 1974; Fujisawa & David, 1981). Finally, the numbers of 4s and 8s that were blocked from the stenotele pathway were estimated by correcting the decreased numbers of nests of 8 and 16 stenoteles (referred to as S₈ and S₉, respectively) observed at 96 h (Table 1) for the duration of the staining period and the fraction of cell cycle occurring during treatment. This resulted in estimates of 29·0/polyp per day and 28·6/polyp per day, respectively. Similarly, the numbers of 4s and 8s that entered the desmoneme and atrichous isorhiza pathways above control levels were estimated from the increased numbers of nests of 8 and 16 cells (referred to as D₈ and D₁₆, and A₈ and A₁₆, respectively) observed at 72h (Table 1), resulting in estimates of 23·0/polyp per day and 34·7/polyp per day, respectively. The estimated numbers of 4s or 8s that were blocked from the stenotele pathway matched reasonably well with those that entered the desmoneme and atrichous isorhiza pathways. These quantitative comparisons support the idea that precursor cells were forced to switch their pathways from stenotele to desmoneme and atrichous isorhiza differentiation by an endogenous factor. However, the precursor cells that were blocked from the stenotele pathway were not equally distributed into other pathways, because essentially all of the 4s that did not enter the stenotele pathway appeared to give rise to A₈s, and 8s gave rise to D_l6s.

During head and foot regeneration, nematoblasts are inactivated and almost all of them are removed by phagocytosis of epithelial cells (Fujisawa & David, 1984). If the extract treatment inactivates stenotele precursor cells, this could lead to an increase in the number of phagocytosed nematoblasts in epithelial cells. The decrease in stenoteles at 96 h due to treatment with extract was 600/polyp when determined in macerated preparations. Since the extract exerted its effect near the S/G₂ boundary in the terminal cell cycle of nematoblasts, the number of affected nematoblasts was 300/polyp. Whether the number of nematoblasts was in fact phagocytosed by epithelial cells was then examined. Groups of 10 animals were macerated at 6, 12 and 24 h during the treatment with extract and processed for Feulgen staining with Fast Green counterstain. Untreated control animals were also stained. Table 2 shows the observed number of phagocytosed nematoblasts in epithelial cells of both control and treated animals. An average of 37 phagocytosed nematoblasts was observed in a treated animal at all sampling points. The value was much smaller than expected and was not significantly different from that observed in untreated control animals. The results indicate that very few precursor cells were removed by phagocytosis.

The first sign of the inhibition of stenotele differentiation by treatment with crude extract was detected 72-78 h after the addition of the extract (Fig. 2). Since the time required for stenotele differentiation, from the end of the final S phase, is 74 h (Fig. 1), the results indicate that an endogenous factor present in crude extract affects precursor cells near the S/G₂ boundary in the terminal cell cycle. The timing of the inhibition coincides well with the timing of commitment of stenotele differentiation, which occurs near the S/G₂ boundary in the terminal cell cycle of precursor cells (Fujisawa & David, 1982). This suggests, therefore, that an endogenous factor blocks the commitment process. However, since the present experiments involved extract of hydra rather than a highly purified factor, it is not clear whether a single factor or more than one factor is involved in the inhibition of stenotele commitment.

The inhibition of stenotele differentiation appears to have lasted 3-6 h longer than the actual length of treatment (Figs 2, 4). There are three possible explanations for this: (1) there is a short period in the cell cycle that is sensitive to the factor; (2) the factor remains effective for a short period of time in tissues; (3) a combination of (1) and (2). if the first possibility is the case, the sensitive period extends over 3-6 h from the S/G₂ boundary (or even from early G₂ phase) into the S phase in the terminal cell cycle of precursor cells. Since the duration of S phase of nematoblasts is 12 h (Campbell & David, 1974), commitment may be a process extending from the mid-late S phase to the S/G₂ boundary (or early G₂ phase). If the second possibility is correct, 3-6 h is the time required for the decay of the factor in tissues, and the factor affects precursor cells at one point in the terminal cell cycle near the S/G2 boundary. At present, we cannot decide which of these possibilities explains the discrepancy between differentiation time and treatment time.

The precursor cells that were prevented from stenotele differentiation could have three possible outcomes: (1) they could shift from the stenotele to other pathways; (2) they could be inactivated and removed from the hydra body; (3) they could accumulate in number. Results obtained in this study all favoured the first possibility. First, the number of stenotele nests decreased by treatment with extract matches well with the increased number of desmoneme and atrichous isorhiza nests (Table 1). Since the number of precursor nests is fixed, these results suggest a shift of pathways from stenotele to desmoneme and atrichous isorhiza differentiation. Furthermore, quantitative comparison between the numbers of 4s and 8s that were blocked from Sg and Si6 pathways during treatment with extract, and 4s and 8s that entered A_s and D|_ft pathways in numbers above the control levels during the same period of time, suggests that the precursor cells shifted their pathways from stenotele to desmoneme and atrichous isorhiza differentiation. Second, although phagocytosis appears to be a common mechanism for removing inactivated cells in hydra (Fujisawa & David, 1984; Bosch & David, 1984), no significant increase in phagocytosis of epithelial cells was found in animals treated with extract (Table 2). Third, after the removal of extract, no overshooting of stenotele differentiation was observed (Figs 2, 3). If the precursor cells accumulate, one should see a much more rapid recovery in stenotele number, thus resulting in an asymmetry in the kinetics of the decrease and increase in the stenotele density. The results of this study indicate that the kinetics were in fact symmetrical (Figs 2, 3).

The excellent technical assistance of Miss Ikuko Gotoh and Mr Nono Sugimoto is greatly appreciated. The author is also grateful to Dr C. N. David for valuable discussions and comments and to Dr Sugiyama for encouragement. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan.