DEYELOPMENTAL
BIOLOGY
138,243-245
(1990)
BRIEF NOTE Terminal Differentiation of Head- and Foot-Specific Epithelial Cells Occurs at the Same Location in Hydra Tissue without Polarity STEFAN
Zextrnrn
,fiir Molekulare
Biologic Heidelberg,
D~TBEL’
Im Neuenh~eimw
Feld 282,
Accepted November
6900
Heidelberg,
Federal Republic of Gewnany
7, 198.9
The reappearance of terminally differentiated ectodermal epithelial cells was studied in reaggregates of Hydra cells. These cells first occur separated from undifferentiated gastral tissue in mixed clusters consisting of cells which in normal animals are restricted to opposite body poles. Tentacles containing foot-specific basal disc cells as well as feet containing head-specific battery cells were formed from these clusters. This indicates that a positive cross-reaction of head- and foot-forming mechanisms exists at the cellular level and that induction of terminal differentiation precedes the establishment of polarity. % lwo Academic press, 1”~. INTRODUCTION
The fresh water coelenterate Hydra shows a bipolar organization along a single axis of symmetry. Polarity is generated by a set of two pairs of gradients (for review see Bode and Bode, 1984) and maintained even during regeneration of new animals from very small tissue pieces (see Javois et al., 1988, for references). At the level of cellular composition, polarity is reflected by the occurrence of cell types specific for the respective body poles. For example, the battery cells which are differentiation products of the ectodermal epithelial stem cells of the gastric column occur exclusively in the tentacles around the oral pole. They are characterized by their content of up to 20 nematocytes (stinging cells). The ectodermal epithelial stem cells of the gastric column can also differentiate into foot mucus cells which solely occur in the basal disc at the opposite body pole (Diibel et al., 1987). These foot mucus cells can be identified in the tissue by a peroxidase activity which occurs specifically in this cell type (Hoffmeister and Schaller, 1985). In this study, the inclusion of nematocytes into the battery cells of the head and the peroxidase activity specific for foot mucus cells wcrc used as markers to pursue the reappearance of terminally differentiated tissue in reaggregates of Hydra cells. Such reaggregates are formed from Hydra tissue dissociated into single, living cells (Gierer et ah, 1972). Due to the initially random cell arrangement, these reaggregates are well suited for investigating the reappearance of polarity at the morphological and cellular levels. In following the fate of cells in such reaggregates I found that wherever head-specific ectodermal epithelial cells occurred, they were colocalized with foot-spe’ Present address: German Cancer Research hpimrr Feld 280, 6900 Heidelberg, FR(;.
Center,
Im
Neuen-
243
cilic ectodermal epithelial cells, indicating that induction to new terminal differentiation occurs independent from the expression of polarity. METHODS
Hydra vuLgcwis (ex Hydra attenuata) were grown under standard conditions including daily feeding (Hoffmeister and Schaller, 1985). For all experiments nonbudding animals were collected 24 hr after the last feeding. The feet were removed before the dissociation into cells by cutting between upper peduncle and gastric regions. Since tentacles do not easily disaggregate under the used conditions, their removal prior to the disaggregation was not necessary to obtain reaggregates free of battery cells. Dissociation of IIydra tissue and reaggregation was accomplished according to Gierer et al. (1972). After different periods of time, aggregates were fixed for 30 min at room temperature with 4% (w/v) formaldehyde in PBS (phosphate-buffered saline, 140 mMNaC1, 10 mMsodium phosphate, pH 7.2). After three washes in PBS for 15 min each, the peroxidase activity was visualized by an incubation in 0.02% diaminobenzidine, 0.003% HzG2, 0.1 M sodium citrate, pH 4.0, for 30 min at room temperature (Hoffmeister and Schaller, 1985). The reaction was stopped by three washes in water. Battery cells were identified by a content of more than five nematocytes arranged side by side. RESULTS
Hydra without feet were dissociated into single cells and reaggregated. The cell aggregates were not cut into pieces after reaggregation, with the result that “sausages” were obtained consisting of several Hydra. Within the first day of regeneration, the separation of the cell aggregate into ectoderm and endoderm occurred as described (Graf and Gierer, 1980), resulting in 0012-1606/90 (‘opyright All rights
$3.00
6 1990 by Academic Press, Ine of reproduction in any fnrm reserved
b
C
d
h
FIG. 1. Reoccurrence of terminally differentiated ectodermal epithelial cells in reaggregates of Hydra vul~uris cells. All reaggregates were stained for peroxidase activity to visualize foot mucus cells. (a) Reaggregate after 42 hr, no foot mucus cells were detected. (b) Reaggregate after 70 hr, when the first “tentacle” buds emerged. Foot mucus cells are visible almost exclusively in the “tentacle” buds. Arrow, one of the few patches of foot mucus cells which are not located in “tentacle” buds at this stage. (c) Higher magnification of a 7%hr reaggregate showing several foot mucus cells scattered in the tissue of two “tentacle” buds, but not in the tissue between these buds. (d) Tentacle regenerated by a 5-day reaggregate containing several foot mucus cells (arrows). (e) Basal disc regenerated by a ?-day reaggregate containing two patches of battery cells (arrows) as indicated by their content of nematocytes. (f) Higher magnification of a tentacle regenerated by a 5-day reaggregate showing clusters of stinging capsules typical for battery cells and three foot mucus cells with the typical granular distribution of peroxidase activity. (g) “Tentacle” bud on a 7-day reaggregate which had not developed into a tentacle and consists of half foot mucus cells and half battery cells. (h) Low magnification of a ‘i-day reaggregate demonstrating that foot mucus cells at this stage were mostly restricted to foot structures. Scale bars in a, b, h, 500 pm; in c, d, e, f, g, 50 pm. 244
245
BRIEF N(>TE
hollow tissue tubes. No battery cells or foot mucus cells were detected in such tissue. After varying periods of time (53-70 hr), protrusions emerged from the so far smooth surface of the reaggregate (Figs. la and lb). The first detection of foot mucus cells by their specific peroxidase activity coincided with the appearance of these protrusions (Figs. lb and lc). Clusters of nematocytes indicating new differentiated battery cells were not found at this stage but emerged within the following day in the protrusions. Both cell types represented ectodermal epithelial cells which were determined to terminally differentiate after the disaggregation, since none of these cells were present in the early reaggregates. Up to 10 new differentiated foot mucus cells were found almost exclusively in the protrusions (Figs. lb and lc). The tissue between the protrusions was essentially free of foot mucus cells. Within the protrusions, their arrangement appeared to be random. In reaggregates of a size comparable to those shown in Figs. la and lb, only one or very few patches of foot mucus cells were found outside of the protrusions. All of these patches consisted of clusters of more than 20 foot mucus cells (Fig. lb). Most of the protrusions grew out subsequently to form tentacles (for exeption see Fig. lg). Several foot mucus cells occur randomly scattered in several of these tentacles (Figs. Id and If). They contain the peroxidase activity in a granular arrangement typical for foot mucus cells. Nematocytes were never detected in the cells containing peroxidase activity (Fig. If). After 1 week of regeneration, foot mucus cells were found near the tips of the tentacles (indicating the normal tissue turnover) and in larger patches resembling basal discs (Fig. lh). Battery cells were found to be included into some of these new feet (Fig. le). Few structures were found at this time which were composed of half battery cells and half foot mucus cells. These structures maintained the morphology of the early tentacle protrusions (Fig. lg). In summary, head- and foot-specific terminal differentiation products of ectodermal epithelial cells occurred in mixed clusters disregarding their normal occurrence at opposite body poles. The tissue between these clusters was essentially free of these terminally differentiated cells. DISCUSSION
The new differentiation of battery cells and foot mucus cells in reaggregates required a time which was about two to three times longer than in normal regenerates (Dtibel et al., 1987). This may be due to the need for tissue reorganization into ectoderm and endoderm (Graf and Gierer, 1980), which was totally disturbed in reaggregates, and may have to be reestablished prior to
the induction of differentiation of ectodermal epithelial cells. During regeneration of reaggregates made from Hydra from which the foot was not removed prior to the dissociation, the arrangement of foot mucus cells in 3to 7-day reaggregates was essentially the same as described above, despite the presence of randomly distributed foot mucus cells derived from the dissociated basal discs in the early reaggregates (data not shown). This suggests a selective mechanism which leads to either an elimination of cells from the “wrong” location or an individual movement of differentiating cells into clusters. This sorting mechanism established a subdivision into differentiated and nondifferentiated tissue. To my surprise, this sorting mechanism did not discriminate between head- and foot-specific differentiation products. Only later, when tentacles and basal discs were morphologically established, is the differentiation in these structures restricted to the “right” cell type. The occurrence of intermediate structures (Fig. lg), which appeared to get stuck in the process of pattern expression at the decision to be tentacle or foot, suggests that the fate of the original protrusions is correlated with their initial fraction of head- or foot-specific cells. On the basis of observations on the morphological level of tandem grafts of hydra tissue, it was proposed that head- and foot-forming mechanisms cross-react positively (Ando ef ul., 1989). This study suggests a positive cross-reaction also at the cellular level. The results also indicate that the determination of ectodermal epithelial cells to terminally differentiate can occur before the establishment of polarity. I thank H. C. Schaller in whose laboratory this work was carried out (supported by the DFG, SFB 317, by the Bundesministerium fiir Forschung und Technologie BCT 365/l, and by the Fonds der Deutschen Chemischrn Industrie) and S. Hoffmeister for helpful comments during the whole work. REFERENCES Ar;uo, H., SAWAr).4, Y., SIIIMIZ~I, H., and SIXIYAMA, T. (1989). Pattern formation in Hydra tissue without developmental gradients. 11~. Bid. 133, 405-414. Bo~)E, P. M., and BODE, H. R. (1984). Pattering in Hydra. I?1 “Primers in Developmental Biology (G. Malacinski and S. Bryant, Eds.), pp. 213-241. MacMillan, New York. DOBEL, S., HWFMEISTER, S. -4. H., and SC~IALLER, H. C. (1987). Differentiation pathways of ectodermal epithelial cells in hydra. Dif.fivmfircfion 3.5, 181-189. GIERER, A., BERKINC:, S., BWE, H., DAVII~, C. N., FLICK, K., HANSMANN, G., SCIIALLER, H., and TRENKNER, E. (1972). Regeneration of Hydra from reaggreaated cells. Noflcre (Ltmdvn) 239, 98-101. GRAF, L., and GIERER, A. (1980). Size, shape and orientation of cells in budding Hydra and regulation of regeneration in cell aggregates. Wilhelm Rourk Arc+. 188, 141-151. HOFFMEISTER, S., and SC~IAI,LER, H. C. (1985). A new biochemical marker for foot-specific cell differentiation in hydra. Wilhelm R0u.r :s Arch. 194, 453461. JAWIS, L., Bong, P. M., and BODE, H. R. (1988). Pattering of the head in Hydra as visualized by a monoclonal antibody. II. The initiation and localization of head structures in regenerating pieces of tissue. Df~i~. Bid. 129, 390-399.
BIOLOGY
138,243-245
(1990)
BRIEF NOTE Terminal Differentiation of Head- and Foot-Specific Epithelial Cells Occurs at the Same Location in Hydra Tissue without Polarity STEFAN
Zextrnrn
,fiir Molekulare
Biologic Heidelberg,
D~TBEL’
Im Neuenh~eimw
Feld 282,
Accepted November
6900
Heidelberg,
Federal Republic of Gewnany
7, 198.9
The reappearance of terminally differentiated ectodermal epithelial cells was studied in reaggregates of Hydra cells. These cells first occur separated from undifferentiated gastral tissue in mixed clusters consisting of cells which in normal animals are restricted to opposite body poles. Tentacles containing foot-specific basal disc cells as well as feet containing head-specific battery cells were formed from these clusters. This indicates that a positive cross-reaction of head- and foot-forming mechanisms exists at the cellular level and that induction of terminal differentiation precedes the establishment of polarity. % lwo Academic press, 1”~. INTRODUCTION
The fresh water coelenterate Hydra shows a bipolar organization along a single axis of symmetry. Polarity is generated by a set of two pairs of gradients (for review see Bode and Bode, 1984) and maintained even during regeneration of new animals from very small tissue pieces (see Javois et al., 1988, for references). At the level of cellular composition, polarity is reflected by the occurrence of cell types specific for the respective body poles. For example, the battery cells which are differentiation products of the ectodermal epithelial stem cells of the gastric column occur exclusively in the tentacles around the oral pole. They are characterized by their content of up to 20 nematocytes (stinging cells). The ectodermal epithelial stem cells of the gastric column can also differentiate into foot mucus cells which solely occur in the basal disc at the opposite body pole (Diibel et al., 1987). These foot mucus cells can be identified in the tissue by a peroxidase activity which occurs specifically in this cell type (Hoffmeister and Schaller, 1985). In this study, the inclusion of nematocytes into the battery cells of the head and the peroxidase activity specific for foot mucus cells wcrc used as markers to pursue the reappearance of terminally differentiated tissue in reaggregates of Hydra cells. Such reaggregates are formed from Hydra tissue dissociated into single, living cells (Gierer et ah, 1972). Due to the initially random cell arrangement, these reaggregates are well suited for investigating the reappearance of polarity at the morphological and cellular levels. In following the fate of cells in such reaggregates I found that wherever head-specific ectodermal epithelial cells occurred, they were colocalized with foot-spe’ Present address: German Cancer Research hpimrr Feld 280, 6900 Heidelberg, FR(;.
Center,
Im
Neuen-
243
cilic ectodermal epithelial cells, indicating that induction to new terminal differentiation occurs independent from the expression of polarity. METHODS
Hydra vuLgcwis (ex Hydra attenuata) were grown under standard conditions including daily feeding (Hoffmeister and Schaller, 1985). For all experiments nonbudding animals were collected 24 hr after the last feeding. The feet were removed before the dissociation into cells by cutting between upper peduncle and gastric regions. Since tentacles do not easily disaggregate under the used conditions, their removal prior to the disaggregation was not necessary to obtain reaggregates free of battery cells. Dissociation of IIydra tissue and reaggregation was accomplished according to Gierer et al. (1972). After different periods of time, aggregates were fixed for 30 min at room temperature with 4% (w/v) formaldehyde in PBS (phosphate-buffered saline, 140 mMNaC1, 10 mMsodium phosphate, pH 7.2). After three washes in PBS for 15 min each, the peroxidase activity was visualized by an incubation in 0.02% diaminobenzidine, 0.003% HzG2, 0.1 M sodium citrate, pH 4.0, for 30 min at room temperature (Hoffmeister and Schaller, 1985). The reaction was stopped by three washes in water. Battery cells were identified by a content of more than five nematocytes arranged side by side. RESULTS
Hydra without feet were dissociated into single cells and reaggregated. The cell aggregates were not cut into pieces after reaggregation, with the result that “sausages” were obtained consisting of several Hydra. Within the first day of regeneration, the separation of the cell aggregate into ectoderm and endoderm occurred as described (Graf and Gierer, 1980), resulting in 0012-1606/90 (‘opyright All rights
$3.00
6 1990 by Academic Press, Ine of reproduction in any fnrm reserved
b
C
d
h
FIG. 1. Reoccurrence of terminally differentiated ectodermal epithelial cells in reaggregates of Hydra vul~uris cells. All reaggregates were stained for peroxidase activity to visualize foot mucus cells. (a) Reaggregate after 42 hr, no foot mucus cells were detected. (b) Reaggregate after 70 hr, when the first “tentacle” buds emerged. Foot mucus cells are visible almost exclusively in the “tentacle” buds. Arrow, one of the few patches of foot mucus cells which are not located in “tentacle” buds at this stage. (c) Higher magnification of a 7%hr reaggregate showing several foot mucus cells scattered in the tissue of two “tentacle” buds, but not in the tissue between these buds. (d) Tentacle regenerated by a 5-day reaggregate containing several foot mucus cells (arrows). (e) Basal disc regenerated by a ?-day reaggregate containing two patches of battery cells (arrows) as indicated by their content of nematocytes. (f) Higher magnification of a tentacle regenerated by a 5-day reaggregate showing clusters of stinging capsules typical for battery cells and three foot mucus cells with the typical granular distribution of peroxidase activity. (g) “Tentacle” bud on a 7-day reaggregate which had not developed into a tentacle and consists of half foot mucus cells and half battery cells. (h) Low magnification of a ‘i-day reaggregate demonstrating that foot mucus cells at this stage were mostly restricted to foot structures. Scale bars in a, b, h, 500 pm; in c, d, e, f, g, 50 pm. 244
245
BRIEF N(>TE
hollow tissue tubes. No battery cells or foot mucus cells were detected in such tissue. After varying periods of time (53-70 hr), protrusions emerged from the so far smooth surface of the reaggregate (Figs. la and lb). The first detection of foot mucus cells by their specific peroxidase activity coincided with the appearance of these protrusions (Figs. lb and lc). Clusters of nematocytes indicating new differentiated battery cells were not found at this stage but emerged within the following day in the protrusions. Both cell types represented ectodermal epithelial cells which were determined to terminally differentiate after the disaggregation, since none of these cells were present in the early reaggregates. Up to 10 new differentiated foot mucus cells were found almost exclusively in the protrusions (Figs. lb and lc). The tissue between the protrusions was essentially free of foot mucus cells. Within the protrusions, their arrangement appeared to be random. In reaggregates of a size comparable to those shown in Figs. la and lb, only one or very few patches of foot mucus cells were found outside of the protrusions. All of these patches consisted of clusters of more than 20 foot mucus cells (Fig. lb). Most of the protrusions grew out subsequently to form tentacles (for exeption see Fig. lg). Several foot mucus cells occur randomly scattered in several of these tentacles (Figs. Id and If). They contain the peroxidase activity in a granular arrangement typical for foot mucus cells. Nematocytes were never detected in the cells containing peroxidase activity (Fig. If). After 1 week of regeneration, foot mucus cells were found near the tips of the tentacles (indicating the normal tissue turnover) and in larger patches resembling basal discs (Fig. lh). Battery cells were found to be included into some of these new feet (Fig. le). Few structures were found at this time which were composed of half battery cells and half foot mucus cells. These structures maintained the morphology of the early tentacle protrusions (Fig. lg). In summary, head- and foot-specific terminal differentiation products of ectodermal epithelial cells occurred in mixed clusters disregarding their normal occurrence at opposite body poles. The tissue between these clusters was essentially free of these terminally differentiated cells. DISCUSSION
The new differentiation of battery cells and foot mucus cells in reaggregates required a time which was about two to three times longer than in normal regenerates (Dtibel et al., 1987). This may be due to the need for tissue reorganization into ectoderm and endoderm (Graf and Gierer, 1980), which was totally disturbed in reaggregates, and may have to be reestablished prior to
the induction of differentiation of ectodermal epithelial cells. During regeneration of reaggregates made from Hydra from which the foot was not removed prior to the dissociation, the arrangement of foot mucus cells in 3to 7-day reaggregates was essentially the same as described above, despite the presence of randomly distributed foot mucus cells derived from the dissociated basal discs in the early reaggregates (data not shown). This suggests a selective mechanism which leads to either an elimination of cells from the “wrong” location or an individual movement of differentiating cells into clusters. This sorting mechanism established a subdivision into differentiated and nondifferentiated tissue. To my surprise, this sorting mechanism did not discriminate between head- and foot-specific differentiation products. Only later, when tentacles and basal discs were morphologically established, is the differentiation in these structures restricted to the “right” cell type. The occurrence of intermediate structures (Fig. lg), which appeared to get stuck in the process of pattern expression at the decision to be tentacle or foot, suggests that the fate of the original protrusions is correlated with their initial fraction of head- or foot-specific cells. On the basis of observations on the morphological level of tandem grafts of hydra tissue, it was proposed that head- and foot-forming mechanisms cross-react positively (Ando ef ul., 1989). This study suggests a positive cross-reaction also at the cellular level. The results also indicate that the determination of ectodermal epithelial cells to terminally differentiate can occur before the establishment of polarity. I thank H. C. Schaller in whose laboratory this work was carried out (supported by the DFG, SFB 317, by the Bundesministerium fiir Forschung und Technologie BCT 365/l, and by the Fonds der Deutschen Chemischrn Industrie) and S. Hoffmeister for helpful comments during the whole work. REFERENCES Ar;uo, H., SAWAr).4, Y., SIIIMIZ~I, H., and SIXIYAMA, T. (1989). Pattern formation in Hydra tissue without developmental gradients. 11~. Bid. 133, 405-414. Bo~)E, P. M., and BODE, H. R. (1984). Pattering in Hydra. I?1 “Primers in Developmental Biology (G. Malacinski and S. Bryant, Eds.), pp. 213-241. MacMillan, New York. DOBEL, S., HWFMEISTER, S. -4. H., and SC~IALLER, H. C. (1987). Differentiation pathways of ectodermal epithelial cells in hydra. Dif.fivmfircfion 3.5, 181-189. GIERER, A., BERKINC:, S., BWE, H., DAVII~, C. N., FLICK, K., HANSMANN, G., SCIIALLER, H., and TRENKNER, E. (1972). Regeneration of Hydra from reaggreaated cells. Noflcre (Ltmdvn) 239, 98-101. GRAF, L., and GIERER, A. (1980). Size, shape and orientation of cells in budding Hydra and regulation of regeneration in cell aggregates. Wilhelm Rourk Arc+. 188, 141-151. HOFFMEISTER, S., and SC~IAI,LER, H. C. (1985). A new biochemical marker for foot-specific cell differentiation in hydra. Wilhelm R0u.r :s Arch. 194, 453461. JAWIS, L., Bong, P. M., and BODE, H. R. (1988). Pattering of the head in Hydra as visualized by a monoclonal antibody. II. The initiation and localization of head structures in regenerating pieces of tissue. Df~i~. Bid. 129, 390-399.
