Chimaeric hydra strains were produced from a normal strain (105) and a naturally-occurring mutant strain (L4) which has a large polyp size, a low budding rate and a high head-inhibition potential. Various properties of the chimaeras were then examined and compared to those of the two parental strains.

Hydra tissue consists of three cell lineages: the ectodermal epithelial, the endodermal epithelial and the interstitial cell lineages. Using the methods recently developed by Marcum & Campbell (1978b) and by Wanek & Campbell (1982), six chimaeric strains were produced which contained six different combinations of the three cell lineages from 105 and L4.

Evidence obtained from the comparison of the chimaeras and their parental strains indicates that the ectodermal epithelial cell lineage in L4 is primarily responsible for the large polyp size and the low budding rate of this strain, whereas the endodermal epithelial cell lineage is largely, and the interstitial cell lineage is also partially, responsible for the high head-inhibition potential in L4. This suggests that the mechanisms determining the occurrence and location of bud formation and the mechanisms determining the inhibition potential levels are not related to each other (cf. Takano & Sugiyama, 1983; Bode & Bode, 1983). Evidence was also obtained which suggests that the levels of the head-activation and head-inhibition potentials in the chimaeras are determined independent of each other, apparently without the cross-catalytic relationship between them assumed in the Gierer-Meinhardt model (Gierer & Meinhardt, 1972; Meinhardt & Gierer, 1974).

Hydra tissue consists of three self-proliferating cell lineages: the ectodermal epithelial cell lineage, the endodermal epithelial cell lineage and the interstitial cell lineage (which includes the interstitial stem cells and their differentiated products, nerve cells and nematocytes).

Methods have been developed recently to produce two types of chimaeric hydra strains which consist of different lineages derived from different strains (Marcum & Campbell, 1978b; Wanek & Campbell, 1982). The first type of chimaera (epithelial/interstitial chimaera) consists of the ectodermal and the endodermal epithelial cell lineages from one strain and the interstitial cell lineage from another. This type of chimaera is produced by the ‘interstitial cell elimination and réintroduction’ method described by Marcum & Campbell (1978b). The interstitial cells can be eliminated from one strain by means of colchicine treatment (Campbell, 1976) or by other means (Fradkin, Kakis & Campbell, 1978; Sugiyama & Fujisawa, 1978; Marcum, Fujisawa & Sugiyama, 1980). Réintroduction of the interstitial cells from another strain into these interstitial cell-free hydra results in the production of the epithelial/interstitial chimaera (Marcum & Campbell, 1978b; Sugiyama & Fujisawa, 1978).

The second type of chimaera (ectoderm/endoderm chimaera) consists of the ectodermal epithelial cell lineage from one strain and the endodermal epithelial cell lineage from another. This type of chimaera can be produced by the ‘epithelial migration’ method recently described by Wanek & Campbell (1982). In this method, a ring of vitally stained tissue from one strain is grafted into an unmarked polyp of another strain. Differential movement of the ectoderm and endoderm eventually results in chimaeric regions which are visible as stained ectoderm overlying unstained endoderm and unstained ectoderm overlying stained endoderm. These chimaeric regions are then excised and each is allowed to regenerate, forming a whole chimaeric animal containing the ectodermal and the endodermal epithelial cell lineages derived from two different strains.

As pointed out by Wanek & Campbell (1982), the chimaeras produced by their method probably contain a mixed population of interstitial cells from both parental strains. This is because interstitial cells are migratory cells (Brien & Reniers-Decoen, 1955; Tardent & Morgenthaler, 1966) and these cells from both parents are probably able to intermingle when the chimaeric regions are initially formed. This uncertainty, however, can be readily corrected by the method of Marcum & Campbell (1978b). The interstitial cells can be eliminated by means of colchicine treatment from the ectoderm/endoderm chimaera, and then reintroduced from one of the parental strains (or from a third strain). This results in a new chimaera consisting of all the three cell lineages of defined origins. This type of chimaera, however, has never been produced previously.

In the present study, the two methods of chimaera production were used, singly or in combination, to produce chimaeric strains containing the cell lineages of defined origins from the two parental strains, 105 and L4. The former strain is a wild-type standard strain while the latter is a naturally-occurring mutant strain which has a significantly larger polyp size, a significantly lower budding rate (Sugiyama & Fujisawa, 1979) and significantly higher levels of the head-inhibition potential than 105 (Takano & Sugiyama, 1983).

From the three cell lineages in the two parental strains, it is possible to produce six chimaeric strains containing different combinations of the cell lineages. With the two parental strains, therefore, there are eight combinations. The six chimaeric strains were produced, and following two major questions were asked on the properties of these strains:

  • (1) What are the polyp sizes and the budding rates of the chimaeric strains? Which of the three cell lineages is responsible for the large polyp size and the low budding rate of L4? Are these two characters determined by the same or different lineages?

  • (2) What are the levels of the head-inhibition potentials in the chimaeric strains, and which cell lineage is responsible for the high inhibition potential in L4? Are the inhibition potential levels related to the polyp sizes and/or the budding rates in the chimaeric strains?

The present study provided answers to most of these questions. In addition, an important and unexpected observation was also made on the mutual relationship between the head-activation and head-inhibition potentials in the chimaeric strains.

Epithelial/interstitial chimaeras

Chimaeric strains consisting of the ectodermal and endodermal epithelial cell lineages from one parental strain and the interstitial cell lineage from the other parental strain were produced by the procedure of Marcum & Campbell (1978b) with minor modifications. Elimination of the interstitial cells was carried out either by treating animals twice with 0·2% colchicine, or by treating animals once with 0·4 % colchicine followed by 4–6 weeks of starvation. In the latter case, animals were examined carefully under a dissecting microscope at the end of the starvation period, and polyps having smooth tentacles (indicating the absence of nematocytes) were selected. The treated animals were individually placed in small plastic Petri dishes, and these animals were subsequently maintained by hand feeding each animal as described by Marcum & Campbell (1978a). Buds produced from them were also similarly maintained by hand feeding. Within several weeks, stocks of non-feeding animals were established, each originating from a single treated animal. Three to four representative members of the stocks were sacrificed and cell types present in them were examined after maceration of the tissue into single cells according to David (1973). A stock was regarded as interstitial cell-free when no interstitial cells or their derivative cell types were found after counting about 1000 cells in each of the three or four polyps examined.

Interstitial cells were reintroduced into these interstitial cell-free hydra by the procedure described by Sugiyama & Fujisawa (1978). Polyps containing interstitial cells were vitally stained with Evans blue according to Wilby & Webster (1970). The upper half of an interstitial cell-free polyp and the bottom half of an interstitial cell-containing and vitally stained polyp were axially grafted together by threading a piece of nylon fishing line through them. The graft was maintained for 2 to 3 days to allow interstitial cells to migrate into the interstitial cell-free tissue. Then the upper half was separated from the stained bottom half by cutting above the graft junction. The upper half now repopulated with interstitial cells was hand fed until the animal started to feed itself. This animal and its buds constituted a chimaeric clone.

Ectoderm /endoderm chimaeras

Chimaeric strains consisting of the ectodermal epithelial cell lineage from one strain and the endodermal epithelial cell lineage from the other strain were produced by a two-step procedure.

The first step was done by the method of Wanek & Campbell (1982). A 105 polyp was vitally stained in both the ectoderm and endoderm using India ink (Campbell, 1973; Marcum & Campbell, 1978a). A ring of stained tissue was cut out from the upper gastric region of this polyp, and it was grafted into the upper gastric region of the unstained polyp of L4 using a piece of nylon fishing line. During the following several days, the ectoderm and endoderm moved relative to one another along the body column. This produced chimaeric regions which consisted of the stained ectoderm of 105 overlying the unstained endoderm of L4 (and vice versa). These chimaeric regions were then excised out and each is allowed to regenerate into a whole chimaeric polyp. This chimaeric polyp and its buds constitute an ‘intermediate’ chimaeric clone.

In the second step, the interstitial cells of undefined origins were eliminated from the intermediate chimaeras and new interstitial cells were reintroduced from either 105 or L4. This was achieved using the same method described above to produce the epithelial/interstitial chimaeras.

The six chimaeric strains produced in the present study are referred to by the notations showing the origins of the three cell lineages in them in the order of ectodermal epithelial, endodermal epithelial and interstitial cell lineage. For example, the chimaera [105ect/L4end/105int] contains the ectodermal epithelial and the interstitial cell lineages from 105 and the endodermal epithelial cell lineage from L4. The intermediate chimaeras containing the interstitial cell lineage of undefined origins are referred to by the notations using question marks for the origin of the interstitial cell lineage. For example, the chimaera [105ect/ L4end/?] contains the ectodermal epithelial cell lineage from 105, the endodermal epithelial cell lineage from L4 and the interstitial cell lineage of undefined origin.

Culture conditions

Animals were cultured in a constant temperature room maintained at 18 ± 0·5°C, in the modified ‘M’-solution (Muscatine & Lenhoff, 1965; Takano & Sugiyama, 1983). Antibiotics rifampicin (50μg/ml, Daiichi Seiyaku) and kanamycin (50 μg/ml, Banyu Seiyaku) were added to the culture solution for the colchicine-treated animals and the resultant interstitial cell-free hydra. Freshly hatched brine shrimp nauplii (Nissei brand, Nippon Jisei Sangyo, Tokyo) were used as food.

Three different culturing conditions were employed. Stock cultures of parental and chimaeric strains were maintained at the density of about 10 to 20 polyps in a 200 ml beaker containing about 200 ml of the culture solution. They were fed three or four times per week, and a few hours after each feeding transferred to new containers with fresh culture solution.

Animals used for determination of the morphogenetic potentials (see below) were cultured under the rigorously controlled and standardized mass culture conditions described previously by Takano & Sugiyama (1983). Relatively young animals showing their first bud protrusion (termed ‘standard polyps’) were collected from the mass culture and used for the grafting experiments.

Animals used for determination of polyp size and growth and multiplication parameters were cultured individually in small plastic dishes as described previously by Sugiyama & Fujisawa (1979).

Parameters of growth and multiplication

From the culture record of a population of individually cultured animals described above, the following three parameters were obtained.

(1) Population growth rate

This was expressed by the population doubling time which is the average numbers of days required for the 2-fold increase of the total numbers of animals in a population.

(2) Budding rate

This was expressed by the average numbers of buds produced per day per mature polyp.

(3) Bud developmental time

This is defined as the average numbers of days required by a newly formed bud to produce its own first bud. Completion of the basal disk while the bud is still attached to the parent is used as the criterion of the new bud formation (Sugiyama & Fujisawa, 1979).

Polyp size

Hydra size varies according to the stage of polyp development and also to the culture conditions. To compare polyps in the same stage of development and cultured under the same conditions, standard polyps (young polyps showing the first bud protrusion) were obtained from the populations of individually cultured animals described above, and their sizes were determined in the following three ways.

(1) Protein content per polyp

This was determined by the modified method of Lowry, Rosenbrough, Farr & Randoll (1951) using bovine serum albumin as the standard as described previously by Takano (1983).

(2) Total cell numbers per polyp

This was determined by macerating individual polyp in 0-4 ml of macerating solution (David, 1973) and counting the total cells in a haemocytometer.

(3) Polyp length

The total length of well-stretched polyp was determined under a dissecting microscope with an ocular scale as described previously by Takano (1983).

Endodermal pigment

Endodermal pigment was analysed on 5-day starved hydra by the methods of Krinsky & Lenhoff (1965) (also see Wanek & Campbell, 1982). Approximately 20–40 polyps were homogenized in 1 ml of 95 % ethanol, the homogenate was centrifuged at 2500 g for 10 min and the absorption spectrum of the resultant supernatant fluid was recorded with a Shimadzu UV-190 spectrophotometer.

Morphogenetic potentials

The levels of the head-activation and the head-inhibition potentials of the chimaeras were examined by the lateral tissue grafting procedure which was originally described by Webster & Wolpert (1966) and previously used by Takano & Sugiyama (1983) to compare the potential levels of 105 and L4.

The body column of a well-stretched standard polyp from the hypostome to the bud protrusion was divided into four equal lengths, the column from the bud protrusion to the basal disk was divided into the ratio of 1:2, and the four positions thus obtained were numbered from 1 to 4 and used as the sites for potential comparison (Sugiyama, 1982).

To examine the head-activation potential, a ring of tissue was excised from one of the four positions of a donor polyp, cut into two or three pieces containing about 5000 cells each, and one of the pieces was grafted onto the standard host site at position-4 of a 105 polyp. From 30 to 40 grafts were generally made using the donor tissue obtained from the same origin. The grafted animals were kept for 7–8 days and the percentage of head induction produced by the donor tissue was determined.

To examine the head-inhibition potential, the standard donor tissue obtained from position-1 of 105 was grafted to position-1, -2, -3 or -4 of a recipient polyp. From 30 to 40 grafts were generally made using the same recipient site and the percentage of head induction produced during the following 7-8 days Was determined.

The head-induction percentage values thus obtained were then used to compare the potential levels of the chimaeric and parental strains employing strain 105 as standard of the potentials. In this procedure, the potential levels of all the strains were expressed by the positions on 105 which have the same levels of the head-activation or head-inhibition potentials. For this purpose, two standard head-induction percentage lines were produced. The first was produced from the results obtained when the four donor tissues from 105 were grafted to the standard host site (position-4 of 105). This line was used to find the 105 positions which have the same levels of the head-activation potentials as the other donor tissues. For example, the donor tissue from position-1 of L4 induced heads at 54 % at the standard host site. Application of this value to the standard line indicated that the donor tissue from position-1·8 of 105 would also induce heads at the same percentage. Thus, we used position-1·8 of 105 to represent the headactivation potential level at position-1 of L4.

The second standard line, which was used to compare the head-inhibition potentials, was produced from the results of grafting the standard donor tissue (position-1 of 105) to the four recipient sites on 105. The standard donor tissue induced heads at 24 % when grafted to position-4 of L4. Application of this value to the second standard line indicated that heads would be induced at the same percentage at position-1·7 of 105 by the same donor. Thus, position-1·7 of 105 was used to represent the level of the head-inhibition potential at position-4 of L4. In this manner, the potential levels of all the strains were expressed by the 105 positions having the same potential levels, and they were presented as shown in Fig. 2 (for more details see Takano & Sugiyama, 1983).

Cell lineages in the chimaeras

Six chimaeric strains were produced from the two parental strains in the present study. Several months after their production, the cell lineages in the chimaeras produced were analysed by two different examinations.

The endodermal epithelial cell lineage was analysed by the carotenoid pigment(s) present in the cells of this lineage. On a diet of brine shrimp nauplii, strain 105 is pale pink in colour, while strain L4 appears yellow. This colour difference of the two strains is shown by the absorption spectra of the pigments in the ethanol extract from them. The spectrum for 105 has a single broad peak at 473 nm (Fig. 1A), whereas the spectrum for L4 has three peaks at 410,442 and 469nm (Fig. IB). This difference was utilized to examine the endodermal epithelial cell lineage in the chimaeras. Fig. 1C shows that three chimaeras containing the endodermal epithelial cell lineage from 105 ([105ect/105end/L4int], [L4ect/105end/105int] and [L4ect/105end/L4int]) have the same spectrum as 105, whereas Fig. ID shows that the other three chimaeras containing the endodermal epithelial cell lineage from L4 ([105ect/L4end/105int], [105ect/L4end/ L4int] and [L4ect/L4end/105int]) have the same spectrum as L4. This indicates that the former and the latter three chimaeras indeed have the same endodermal epithelial cells as 105 and L4, respectively. These colour characteristics of chimaeras remained unchanged for more than two years after their production.

Fig. 1.

Absorption spectra of the 95% ethanol extracts. (A) 105. (B) L4. (C) [105ect/105end/L4int] [L4ect/105end/105int] (– – –) and [L4ect/ 105end/L4int] (—). (D) [L4ect/L4end/105int]( – – –) [105ect/L4end/105int] (—) and [105ect/L4end/L4int] (– – –).

Fig. 1.

Absorption spectra of the 95% ethanol extracts. (A) 105. (B) L4. (C) [105ect/105end/L4int] [L4ect/105end/105int] (– – –) and [L4ect/ 105end/L4int] (—). (D) [L4ect/L4end/105int]( – – –) [105ect/L4end/105int] (—) and [105ect/L4end/L4int] (– – –).

Fig. 2.

The levels of the head-activation and the head-inhibition potentials. The abscissa represents the four axial positions along the body axis. The ordinate represents the levels of the head-activation potential (open circles with solid line) and the head-inhibition potential (closed circles with broken line). These levels are shown using 105 as the standard of the potentials. In 105, position-1 has the highest level and position-2, -3 and -4 have progressively lower levels of the potentials. These levels are represented from 1 to 4 on the ordinate (Takano & Sugiyama, 1983). (A) shows the potential levels of strain 105. Since this strain is used as the standard, its potential levels are shown by the two straight lines (solid and broken lines for the head-activation and the head-inhibition potential, respectively). (B) to (H) show the potential levels of the six chimaeric strains and L4 as indicated. The straight diagonal dotted lines are drawn in these figures to indicate the potential levels of 105 as reference.

Fig. 2.

The levels of the head-activation and the head-inhibition potentials. The abscissa represents the four axial positions along the body axis. The ordinate represents the levels of the head-activation potential (open circles with solid line) and the head-inhibition potential (closed circles with broken line). These levels are shown using 105 as the standard of the potentials. In 105, position-1 has the highest level and position-2, -3 and -4 have progressively lower levels of the potentials. These levels are represented from 1 to 4 on the ordinate (Takano & Sugiyama, 1983). (A) shows the potential levels of strain 105. Since this strain is used as the standard, its potential levels are shown by the two straight lines (solid and broken lines for the head-activation and the head-inhibition potential, respectively). (B) to (H) show the potential levels of the six chimaeric strains and L4 as indicated. The straight diagonal dotted lines are drawn in these figures to indicate the potential levels of 105 as reference.

The interstitial cell lineage in the chimaeras was examined by an indirect method. In the process of producing the six chimaeric strains, stocks of interstitial cell-free hydra were produced by colchicine treatment and starvation (see Materials and Methods). These stocks contained no detectable level of interstitial cells or their derivatives. This, however, did not rule out the possibility that a very low number of the interstitial stem cells were present in these stocks. After the interstitial cell reintroduction, the residual small numbers of interstitial cells, if present, might have multiplied vigorously to eventually become predominant in the chimaeras.

Whether such an event occurred or not was tested indirectly by using a nematocyst-deficient strain (nem-3). This strain contains a very reduced number of holotrichous isorhizas in its tentacles (Sugiyama & Fujisawa, 1977). In parallel with the reintroduction of the interstitial cells from 105 or L4 to the interstitial cell-free hydra, the interstitial cells from nem-3 were also reintroduced into different polyps of the same interstitial cell-free stocks. For example, the chimaera [105ect/L4end/nem-3int] was produced in parallel with the production of [105ect/L4end/105int] and [105ect/L4end/L4int].

Table 1 shows the nematocyst composition in the tentacles of 105, L4, nem-3 and the chimaeric strains produced from them. Nematocyst composition in 105 and L4 is very similar. As compared to them, nem-3 contains a significantly lower proportion of holotrichous isorhizas and slightly higher proportions of stenoteles and atrichous isorhizas. This altered nematocyst composition is also found in all the chimaeric strains which are produced by the reintroduction of the interstitial cell lineage from nem-3 (but not in the other chimaeras). This indicates that the defect responsible for the abnormal nematocyst composition in nem-3 is located in its interstitial cell lineage, and that the chimaeras produced by reintroduction of this lineage from nem-3 indeed contain this lineage. This suggests that recovery of the residual interstitial cells did not occur in these chimaeras. Therefore, it seems safe to assume that it similarly did not occur in the other chimaeras produced in parallel.

Table 1.

Nemotocyst composition in the tentacles

Nemotocyst composition in the tentacles
Nemotocyst composition in the tentacles

Polyp size

The standard polyps are relatively young polyps showing their first bud protrusion. The sizes of the standard polyps of the two parental and the six chimaeric strains were examined in three different ways (see Materials and Methods), and the results obtained are presented in Table 2. L4 is two to three times larger than 105 in protein content per polyp, total cell numbers per polyp and polyp length. The three chimaeric strains (No. 3, 4 and 5 in Table 2) are similar to 105 in size, whereas the other three chimaeras (No. 6, 7 and 8 in Table 2) are similar to L4 in size. The former chimaeras contain the ectodermal epithelial cell lineage of 105 origin, whereas the latter chimaeras contain the ectodermal epithelial cell lineage of L4 origin. This indicates that the sizes of the chimaeras are similar to those of their ectodermal epithelial parents.

Table 2.

Size of standard polyp

Size of standard polyp
Size of standard polyp

Parameters of growth and multiplication

A hydra population grows exponentially by asexual budding under standard laboratory culture conditions. The growth rate of a population is determined primarily by two factors; how rapidly newly formed buds mature to start producing their own buds (bud developmental rate) and how frequently mature polyps produce buds (budding rate) (Maruyama & Sugiyama, 1979).

Table 3 shows these three rates for the two parental and the six chimaeric strains. L4 has a significantly longer population doubling time, a significantly lower budding rate and a significantly longer bud developmental time than 105 (Sugiyama & Fujisawa, 1979). It can be noted that these three parameters of the chimaeras are similar to those of their ectodermal epithelial cell parents. Namely, the three chimaeras containing the ectodermal epithelial cell lineage of 105 origin (No. 3, 4 and 5 in Table 3) are similar to 105, whereas the other three chimaeras containing the ectodermal epithelial cell lineage of L4 origin (No. 6, 7 and 8 in Table 3) are similar to L4 in these rates.

Table 3.

Parameters of growth and multiplication

Parameters of growth and multiplication
Parameters of growth and multiplication

Morphogenetic potentials

Hydra tissue has two types of antagonistic morphogenetic potentials involved in head formation. One is the potential to stimulate or activate the formation of the head structure (head-activation potential), and the other is the potential to inhibit head structure formation (head-inhibition potential).

The relative levels of the two potentials in 105 and L4 were previously compared by lateral grafting of tissue by Takano & Sugiyama (1983). The results showed that L4 had nearly the same or a slightly lower head-activation and a significantly higher head-inhibition potential than 105.

The same analysis was extended in the present study to compare the levels of the two potentials in the six chimaeric and the two parental strains. The method employed is described in Materials and Methods, and the results obtained are presented in Fig. 2.

Fig. 2A shows the potential levels in 105. Since this strain is adopted as the standard, the gradients of the two potentials in this strain are represented by the two straight lines (Takano & Sugiyama, 1983).

Fig. 2H shows the potential levels at the four positions of L4. As compared to 105, its levels of the head-activation potential are significantly lower and its levels of the inhibition potential are significantly higher. The inhibition potential levels at position-1 and -2 of L4 are not shown. This is because their levels are higher than the highest level of the standard (position-1 of 105) and cannot be shown exactly in the present method of expressing the potential levels. The level of the head-activation potential at position-4 of L4 is also not shown. This is because its level is approximately at the same levels as position-4 of 105, but not exactly known (Takano & Sugiyama, 1983).

The potential levels of the six chimaeric strains are shown in Figs 2B-G. From the comparison of the potential levels in Fig. 2, following features can be observed on the relationships between the cell lineages and the potential levels.

To examine the influence of the interstitial cell lineage on the potentials, one selects pairs of strains which have the same ectodermal and endodermal epithelial cell lineages but different interstitial cell lineage, and then compares their potential levels. Strain 105 (Fig. 2A) and the chimaera [105ect/105end/ L4int] (Fig. 2B) are an example. These two strains have nearly the same levels of the head-activation potential, but the latter appears to have slightly higher levels of the inhibition potential. A similar relationship is also found in two other pairs of strains which have the same cell lineages except the interstitial cel1 lineage (C vs. D and G vs. H in Fig. 2). This indicates that removing the interstitial cell lineage of 105 origin from a strain and replacing it by the same lineage from L4 origin produces a small increase of the inhibition potential but no difference of the activation potential.

However, there appears to be an exception to this rule. A pair of strains (E vs. F in Fig. 2) show no recognizable differences in either activation or inhibition potentials although they have the same ectodermal and endodermal epithelial cell lineages but different interstitial cell lineages. The significance of this, however, is uncertain due to the limited resolving power of the analysis.

The influence of the endodermal epithelial cell lineage on the potentials can be similarly examined by comparing the potential levels in pairs of strains which have the same lineages except the endodermal epithelial cells. Strain 105 (Fig. 2A) and the chimaera [105ect/L4end/105int] (Fig. 2C) are an example. The latter has substantially higher levels of the inhibition potential than the former, but the two have nearly the same levels of the activation potential. The same relationship can be also found in the three other pairs (B vs. D, E vs. G and F vs. H in Fig. 2).

This indicates that replacement of the endodermal epithelial cell lineage from 105 to L4 produces a substantial increase of the inhibition potential but no difference of the activation potential.

In principle, the influence of the ectodermal epithelial cell lineage on the potentials can be also similarly examined by comparing the potential levels in pairs of strains which differ only in the ectodermal epithelial cell lineage in the lineage composition. Strain 105 (Fig. 2A) and the chimaera [L4ect/105end/ 105int] (Fig. 2E) have the same cell lineages except in the ectodermal epithelial cells. The latter has substantially lower levels of the activation potential than the former, but the two have nearly the same levels of the inhibition potential. Essentially the same relationship is also found in the three other pairs which have different ectodermal epithelial but the same endodermal epithelial and interstitial cell lineages (B vs. F, C vs. G and D vs. H in Fig. 2). This appears to indicate that replacement of the ectodermal epithelial cell lineage from 105 to L4 produces a substantial drop of the activation potential but no significant difference in the inhibition potential.

An important reservation, however, has to be attached to this observation. This is because replacement of this lineage (but not the other two lineages) from 105 to L4 produces a large difference in the polyp size as already shown (Table 2), and this makes the potential level comparison very difficult (see Discussion).

Origin and stability of the cell lineages in the chimaeras

Before going into main discussion, it is important to establish that the six chimaeric strains produced in the present study indeed contain the cell lineages from 105 or L4 as they are intended to contain. Uncertainty in the cell lineages in the chimaera could arise from contamination of the cell lineages occurring during the process of chimaera production, or from cell type conversion (transdifferentiation) from one lineage to another occurring after the chimaera production (Sugiyama & Fujisawa, 1978).

The evidence from the present study, however, speaks strongly against such lineage changes in the chimaeras produced in this study. The endodermal epithelial cell lineage in the chimaeras was tested directly by the absorption spectra of the carotenoid pigement(s) extracted from these cells (Fig. 1). The interstitial cell lineage was tested indirectly by using a nematocyst-deficient strain (nem-3) in parallel with 105 or L4 as the interstitial cell donors to produce the epithelial/interstitial cell chimaeras (Table 1). The results of these tests strongly suggested that there was no uncertainty in the origin and stability of these two lineages in the chimaeras.

The ectodermal epithelial cell lineage in the chimaeras was not tested in the present study because of lack of a suitable cytological marker to distinguish this lineage from 105 and L4. However, circumstantial evidence strongly supports the idea that the chimaeras indeed contain this lineage from 105 or L4 as they are intended to contain. The polyp sizes of the chimaeras are all very similar to those of their ectodermal epithelial parents (Table 2). For example, the chimaera [L4ect/105end/105int] has a large polyp size comparable to L4. This indirectly shows that the ectodermal epithelial cell lineage of this strain is derived from L4. Had it contained the ectodermal epithelial cell lineage from 105, its cell lineage composition becomes identical to 105. Its polyp size (and other characters too) should then become very similar to 105. The fact that this is not the case strongly indicates that this strain does not contain the ectodermal epithelial cell lineage of 105 origin. The same argument also applies to the chimaera [105ect/L4end/ L4int]. Its polyp size is normal and comparable to 105 (Table 2). Its polyp size (and other characters too) should become similar to L4, had it contained the ectodermal epithelial cell lineage of L4 origin. These observations indicate that the chimaeras [L4ect/105end/105int] and [105ect/L4end/L4int] have the ectodermal epithelial cells of L4 and 105 origin, respectively. Two chimaeric strains ([L4ect/105end/L4int] and [105ect/L4end/105int]) were produced in parallel with, and thus have the same epithelial cell lineages as, these two strains. Therefore, their ectodermal epithelial cells should be also of L4 and 105 origin, respectively.

In the case of the two chimaeras ([105ect/105end/L4int] and [L4ect/L4end/ 105int]), little uncertainty exists in their epithelial cell lineages since these lineages were unmanipulated in producing them.

From these observations and considerations, it seems safe to conclude that cell lineage contamination or transdifferentiation from one lineage to another did not occur in the six chimaeric strains produced and examined in the present study.

Cell lineages responsible for the L4 characters

Strain L4 is a naturally occurring mutant strain which has a significantly larger polyp size (Table 2) and significantly lower parameters of growth and multiplication (Table 3) than 105. The present study has shown that the chimaeric strains containing the ectodermal epithelial cell lineage from L4 all have the same large polyp size and slow growth and multiplication rates as L4, whereas the chimaeric strains containing the normal ectodermal epithelial cell lineage from 105 are all normal in these characters (Tables 2 and 3). This indicates that the ectodermal epithelial cell lineage in L4 is primarily responsible for the L4 characters, and that the chimaeras containing the L4 ectodermal epithelial cell lineage also show the same characters. Apparently the other two lineages in L4 are not responsible for the L4 characters since the chimaeras’ characters are little influenced by the origin of the endodermal epithelial and/or the interstitial cell lineages in them.

Wanek (1983) previously produced two ectodermal/endodermal chimaeric strains (containing the interstitial cell lineage of undefined origin) from L4 and the normal strain. She reported that the budding rates of the chimaeras were similar to those of their ectodermal parents. This agrees with our present results (Table 3). However, the polyp sizes of the chimaeras produced by her were intermediate between the two parental strains. This disagrees with our present result which showed that the chimaera’s sizes are similar to those of their ectodermal parents (Table 2). The reason for this disagreement is unclear.

Cell lineages responsible for the morphogenetic potentials

As previously shown by Rubin & Bode (1982) and Takano & Sugiyama (1983), comparison of the morphogenetic potentials of the strains which have different polyp sizes is rather complicated. To avoid this complication, comparison of the potential levels in the present discussion will be limited to strains which have approximately the same polyp sizes: namely strains having normal polyp sizes comparable to 105 and strains having large polyp sizes comparable to L4 (upper and lower halves in Fig. 2).

It was shown that when the interstitial cell lineage from 105 origin in a strain is removed and replaced by the same lineage from L4 origin, the resultant strain nearly always shows a slightly higher head-inhibition potential than the original strain. A similar but substantially greater increase of the inhibition potential is also produced by the replacement of the endodermal epithelial cell lineage in a strain from the 105 to the L4 origin. The effects of this replacement are apparently additive since replacement of the two lineages produces approximately the combined effects of the two replacements (Fig. 2).

These observations suggest that the endodermal epithelial cell lineage of L4 is largely responsible and its interstitial cell lineage is also partially responsible for the high head-inhibition potential in this strain. These observations also suggest that these two lineages are involved in controlling the levels of the inhibition potential in the normal strain.

Whether or not the ectodermal epithelial cell lineage is also involved in controlling the inhibition potential levels cannot be determined from the present study. This is because the replacement of the ectodermal epithelial cell lineage from 105 to L4 produces a difference in the polyp size (Table 2) and makes the potential comparison difficult.

The present study also fails to provide information on the cell lineages which are responsible for controlling the head-activation potential. This is because no significant differences are found in the levels of the head-activation potential within the strains which have the same polyp sizes.

Budding and morphogenetic potentials in L4

Examination of the potential levels in the chimaeras has produced two unexpected observations. One concerns the roles of the inhibition potential on budding. Strain 105 and the three chimaeras containing the ectodermal epithelial cell lineage from 105 have all approximately the same polyp sizes (Table 2) and budding rates (Table 3). These strains, however, have significantly different levels of the inhibition potential (upper half in Fig. 2). The same is also true for L4 and the three chimaeric strains containing the ectodermal epithelial cell lineage from L4 origin.

These observations indicate that the polyp sizes and the budding rates are uninfluenced by the levels of the inhibition potentials in these strains. This is contrary to the previous suggestion made by us and others that the headinhibition potential plays important roles in controlling the occurrence and location of bud formation (Meinhardt & Gierer, 1974; Takano & Sugiyama, 1983; Bode & Bode, 1983).

Relative levels of the head-activation and head-inhibition potentials

The other interesting aspect found in the potential levels of the chimaeras is related to the mechanisms responsible for establishing the gradients of the two potentials along the body axis. Strain 105 and the three chimaeric strains containing the ectodermal epithelial cell lineage from 105 have approximately the same levels of the head-activation potential but significantly different levels of the head-inhibition potential along their body axis (upper half in Fig. 2). The same is also true for L4 and the chimaeric strains containing the ectodermal epithelial cell lineage from L4 (lower half in Fig. 2). In other words, by means of chimaera construction it is possible to change these two potentials independently. This suggests that the levels of the two potentials are determined independent of each other, and that the levels of one potential can change without producing the change of the other in these strains.

The mechanisms responsible for determining the levels of the two potentials are unknown at present. However, an important model has been proposed by Gierer & Meinhardt (Gierer & Meinhardt, 1972; Meinhardt & Gierer, 1974; also see MacWilliams, 1982). According to this model, the levels of the two potentials are strongly influenced by each other’s levels by the cross-catalytic relationship between them (see original literature for details). The present result appears to be inconsistent with this basic assumption of the model since the two potential levels appear to be independent of each other.

We thank Dr N. Wanek for useful advice and suggestions and Dr R. D. Campbell for critical reading of the manuscript. This work was supported by Grants-in-Aid for Scientific Research from Ministry of Education, Japan.

Achermann
,
J.
&
Sugiyama
,
T.
(
1983
).
Genetic analysis of developmental mechanisms in hydra
.
X. Morphogenetic potentials of a regeneration-deficient strain (reg-16) (in preparation)
.
Bode
,
P. M.
&
Bode
,
H. R.
(
1983
).
Patterning in hydra
.
In Primers in Developmental Biology
, vol.
1
(in press).
Brien
,
P.
&
Reniers-Decoen
,
M.
(
1955
).
La signification des cellules interstitielles des hydra d’eau douce et le problème de la réserve embryonnaire
.
Bull. Biol. Fr. Belg
.
89
,
258
325
.
Campbell
,
R. D.
(
1973
).
Vital marking of single cells in developing tissues: India ink injection to trace tissue movements in hydra
.
J. Cell Sci
.
13
,
651
661
.
Campbell
,
R. D.
(
1976
).
Elimination of Hydra interstitial and nerve cells by means of colchicine
.
J. Cell Sci
.
21
,
1
13
.
David
,
C. N.
(
1973
).
A quantitative method for maceration of hydra tissue
.
Wilhelm Roux’ Arch. EntwMech. Org
.
171
,
259
268
.
Fradkin
,
M.
,
Kakis
,
H.
&
Campbell
,
R. D.
(
1978
).
Effects of μrradiation of hydra: Elimination of interstitial cells from viable hydra
.
Radiai. Res
.
76
,
187
197
.
Gierer
,
A.
&
Meinhardt
,
H.
(
1972
).
A theory of biological pattern formation
.
Kybernetik
.
12
,
30
39
.
Krinsky
,
N. I.
&
Lenhoff
,
H. M.
(
1965
).
Some carotenoids in hydra
.
Comp. Biochem. Physiol
.
16
,
189
198
.
Lowry
,
D. H.
,
Rosenbrough
,
N. J.
,
Farr
,
A. L.
&
Randall
,
R. J.
(
1951
).
Protein measurement with the Folin phenol reagent
.
J. biol. Chem
.
193
,
265
275
.
Macwilliams
,
H. K.
(
1982
).
Numerical simulations of hydra head regeneration using a proportion-regulation version of the Gierer-Meinhardt model
.
J. theor. Biol
.
99
,
681
703
.
Marcum
,
B. A.
&
Campbell
,
R. D.
(
1978a
).
Development of Hydra lacking nerve and interstitial cells
.
J. Cell Sci
.
29
,
17
33
.
Marcum
,
B. A.
&
Campbell
,
R. D.
(
1978b
).
Developmental roles of epithelial and interstitial cell lineages in hydra: Analysis of chimeras
.
J. Cell Sci
.
32
,
233
247
.
Marcum
,
B. A.
,
Fujisawa
,
T.
&
Sugiyama
,
T.
(
1980
).
A mutant hydra strain (sf-1) containing temperature-sensitive interstitial cells
.
In Developmental and Cellular Biology of Coelenterates
(ed.
P.
Tardent
&
R.
Tardent
), pp.
429
434
.
Amsterdam
:
Elsevier/North-Holland
.
Maruyama
,
T.
&
Sugiyama
,
T.
(
1979
).
A mathematical model for hydra multiplication by asexual budding. Appendix to “Genetic analysis of developmental mechanisms in hydra. VII Statistical analyses of developmental-morphological characters and cellular compositions.”
Devl. Growth & Differ
.
21
,
337
381
.
Meinhardt
,
H.
&
Gierer
,
A.
(
1974
).
Application of a theory of biological pattern formation based on lateral inhibition
.
J. Cell Sci
.
15
,
321
346
.
Muscatine
,
L.
&
Lenhoff
,
H. M.
(
1965
).
Symbiosis of hydra and algae. I. Effect of some environmental cations on growth of symbiotic and aposymbiotic hydra
.
Biol. Bull. mar. biol. Lab., Woods Hole
.
128
,
415
424
.
Rubin
,
D. I.
&
Bode
,
H. R.
(
1982
).
The aberrant, a morphological mutant of Hydra attenuata, has altered inhibition properties
.
Devl Biol
.
89
,
316
331
.
Sugiyama
,
T.
(
1982
).
Roles of head-activation and head-inhibition potentials in pattern formation of hydra: Analysis of a multi-headed mutant strain
.
Amer. Zool
.
22
,
27
34
.
Sugiyama
,
T.
&
Fujisawa
,
T.
(
1977
).
Genetic analysis of developmental mechanisms in hydra. I. Sexual reproduction of Hydra magnipapillata and isolation of mutants
.
Devl. Growth & Differ
.
19
,
187
200
.
Sugiyama
,
T.
&
Fujisawa
,
T.
(
1978
).
Genetic analysis of developmental mechanisms in hydra. V. Cell lineage and development of chimera hydra
.
J. Cell Sci
.
32
,
215
232
.
Sugiyama
,
T.
&
Fujisawa
,
T.
(
1979
).
Genetic analysis of developmental mechanisms in hydra. VII. Statistical analyses of developmental-morphological characters and cellular compositions
.
Devl. Growth & Differ
.
21
,
361
375
.
Takano
,
J.
(
1983
).
Genetic analysis of developmental mechanisms in hydra. IX. Effect offood on development of a slow-budding strain (L4)
.
Devl Biol., in press
.
Takano
,
J.
&
Sugiyama
,
T.
(
1983
).
Genetic analysis of developmental mechanisms in hydra. VIII Head-activation and head-inhibition potentials of a slow-budding strain (L4)
.
J. Embryol. exp. Morph
.
78
,
00
00
.
Tardent
,
P.
&
Morgenthaler
,
U.
(
1966
).
Autoradiographische Untersuchungen zum Problem der Zellwanderungen bei Hydra attenuata (Pall)
.
Rev. Suisse Zool
.
73
,
468
480
.
Wanek
,
N.
(
1983
).
Roles of ectodermal and endodermal epithelial cells in hydra morphogenesis: Analysis of chimeras
.
J. exp. Zool
.
225
,
89
97
.
Wanek
,
N.
&
Campbell
,
R. D.
(
1982
).
Roles of ectodermal and endodermal epithelial cells in hydra morphogenesis: Construction of chimeric strain
.
J. exp. Zool
.
221
,
37
47
.
Webster
,
G.
&
Wolpert
,
L.
(
1966
).
Studies on pattern regulation in hydra. I. Regional differences in time required for hypostome determination
.
J. Embryol. exp. Morph
.
16
,
91
104
.
Wilby
,
O. K.
&
Webster
,
G.
(
1970
).
Experimental studies on axial polarity in Hydra
.
J. Embryol. exp. Morph
.
24
,
595
613
.