ABSTRACT
When grown on microscope slides under controlled conditions, the marine hydroid Podocoryne carnea provides excellent material for the investigation of the origin of difference, regulation and the control of pattern during development. The growth of both stolons and hydranths is exponential and the ratio of hydranths to stolons is constant.
Qualitative analysis of colony growth during a 5-month period delimits five discrete states that arise in the center and spread peripherally in the radially organized system during the first 3 months. When colony expansion stops, the centrifugal pattern of differentiation is replaced by an apparently random one and the number and proportion of zooids in the colony change.
Extirpation and isolation experiments indicate that sexuality is not transmitted from sexual to asexual portions of the colony, and that it is associated with the dense stolon matte. A theory accounting for the origin of sexuality in terms of decreased permeability of the stolon to substances that necessitate sexual differentiation is suggested.
INTRODUCTION
The colonial hydroid, Podocoryne carnea, is admirably suited to the study of differentiation (Braverman, 1962a). A colony growing on the shell of the marine snail Nassa consists of nutritive and generative zooids rising from a highly ramified and anastomosing stolon. If, instead of a snail, the shell is occupied by a hermit crab, a third type of person, the spiralzooid, is found on the dorsal lip of the shell (Cazaux, 1958; Braverman, 1960). Good descriptions of the animal appear in Allman (1871) and Berrill (1953).
A single nutritive zooid removed from a colony and placed on a microscope slide in standing sea water attaches to the slide and gives rise to a colony. Colony growth consists of the formation of stolons attached to the substrate and of the appearance of zooids—first nutritive, then generative—on the stolons. The small number of differentiated entities at the zooid level—stolon, nutritive hydranth and generative zooid—recommend this system for investigating how differences arise, what determines pattern, and how regulation is accomplished in developing systems.
Like Hydra (Loomis, 1959), Podocoryne is rendered sexual by the experimental addition of a small excess of CO2 (2 per cent) to the culture water (Braverman, 1962b).
This paper describes the growth and development of the hydroid colony, quantitatively up to the time that sexual zooids appear and qualitatively until colony expansion is limited by the borders of the microscope slide. Extirpation and isolation experiments localized the sexualizing factor and indicate the minimum unit of its expression.
METHODS
Colony growth
The culture method was essentially that described by Crowell (1953). A clone was initiated by removing a single hydranth from a shell-grown colony and placing it on a microscope slide in standing sea water. After 2 or 3 days, when this hydranth had attached to the slide and had begun to form stolons, the slide was transferred to a glass slide holder suspended in an aquarium of filtered sea water at 18° or 23°C. In these experiments the sea water was aerated and agitated continually, and completely changed daily. The colonies were fed by immersing the slide holder in a dish of hatched brine shrimp, Artemia salina.
After approximately one month a colony can serve as a source of hydranths for about thirty additional colonies. In this manner large numbers of colonies can be obtained, all ultimately derived from a single hydranth. Five such clones (A, B, C, W and X) have been isolated and maintained over a period of 2 years.
Daily growth was recorded by camera lucida tracings or photographs. The total linear growth of stolons was measured, and the number of hydranths counted. These data provide information about the development of the colony up to the time that the sexual form appears. Thereafter, the hydranths are so dense that it is impossible to count them accurately.
For long term studies of development a new clone (X) was isolated and subcultured approximately every month, using the most recently initiated generation to start a new one. Sexual zooids did not appear in this clone until the fifth or sixth week, thus each colony was initiated with a hydranth of a’ non-sexual colony. All colonies were maintained at 23°C under the standard culture conditions. At the end of 5 months there were five generations, ranging in age from 1 to 5 months. These were analyzed and compared with each other and the differences in size, development and pattern were recorded.
Control of sexuality
Circles of about 7 ·5 mm. in diameter were removed from the sexual portions of ten colonies of clone X that varied in age from 6 weeks to 4 months (in 1 month intervals). Colonies were maintained at 23°C under standard culture conditions and were analyzed after 1, 2, and 4 weeks to determine the nature of the hydranths that appeared in the extirpated area.
Single hydranths, combinations of sexual and asexual zooids, or groups of sexual and asexual zooids alone were removed from 2-month old colonies derived from the same parent colony. When a colony is 2 months old, individual stolons in the sexual portion are barely discernible. The matte that results can be removed with the hydranths still attached. The zooids and stolon mattes were allowed to attach to microscope slides and the character of the zooids, maintained and appearing, was recorded.
FIGS. A and B. (A) Side and (B) top views of slide-grown Podocoryne carnea colonies. The colony consists of two kinds of zooids, nutritive and generative. Medusa buds (arrows) are released and produce gametes that are shed into the water. Generative zooids become as large as they appear in (A) when the colonies grown at 23°C are transferred to 18°C. Clone X; magnified about × 10. Fig. A was photographed by Giacomelli Sante.
FIGS. A and B. (A) Side and (B) top views of slide-grown Podocoryne carnea colonies. The colony consists of two kinds of zooids, nutritive and generative. Medusa buds (arrows) are released and produce gametes that are shed into the water. Generative zooids become as large as they appear in (A) when the colonies grown at 23°C are transferred to 18°C. Clone X; magnified about × 10. Fig. A was photographed by Giacomelli Sante.
RESULTS
Colony growth
The stolons and zooids in a Podocoryne colony seem to be spaced irregularly, with no apparent pattern (Plate 1, Figs. A and B). Camera lucida tracings (Text-fig. 1) of the growth of a colony of clone A (18°C) show how the stolons grow out from the original hydranth. Upon these stolons more hydranths form. Stolons branch and anastomose, as well as growing terminally. During this early period they seem to grow at random throughout the colony.
Camera lucida tracings of colony A –9, maintained at 18°C. Measurements of the length of stolons and the number of hydranths can be easily taken from such tracings. The date of each tracing is shown.
The apparent irregularity of zooid distribution notwithstanding, a constant proportion is maintained between stolon length and hydranth number throughout the measurable growth period (Text-figs. 2 and 3).
Average growth of six colonies of clone B, grown at 18°C. Stolon length is plotted in arbitrary units. A. Plotted on a linear scale. The ratio of stolon length to hydranth number is constant. B. Plotted on a semi-log scale. The growth of clone B was exponential, the doubling time about 3 days.
Average growth of six colonies of clone B, grown at 18°C. Stolon length is plotted in arbitrary units. A. Plotted on a linear scale. The ratio of stolon length to hydranth number is constant. B. Plotted on a semi-log scale. The growth of clone B was exponential, the doubling time about 3 days.
Average growth of four colonies of clone W, grown at 23°C. These figures were obtained from photographs of the colonies. Stolon length is expressed in the same units as Text-fig. 2. The rate of hydranth increase falls off just before the appearance of generative zooids. After the second deflection, three of the four colonies were sexual.
Average growth of four colonies of clone W, grown at 23°C. These figures were obtained from photographs of the colonies. Stolon length is expressed in the same units as Text-fig. 2. The rate of hydranth increase falls off just before the appearance of generative zooids. After the second deflection, three of the four colonies were sexual.
After 2 to 6 weeks, depending on the clone, sexual zooids began to appear in the central portion of the colony. These first sexual zooids were always new zooids; the medusa buds were visible almost as soon as the zooid appeared. It is not known whether nutritive hydranths ever become sexual under normal culture conditions. Almost all sexual zooids were distinctly smaller than the nutritive hydranths.
Meanwhile the colony continued to expand peripherally as stolons grew out and hydranths appeared on them. In a similar manner sexual zooids also spread out from the center of the colony.
Diagrams of colonies of clone X (23 °C) illustrate their growth from a presexual state to the time when the colony has completely covered the microscope slide and has no further room for expansion (Text-fig. 4). Sexuality appeared much later in clone X than in clone W (18°C).
This shows how the colonies would look if they were cut through the center and at right angles to the slide. For clarity only two tentacles are shown on each zooid, and the size of the zooids is exaggerated. The relative size of the hydranths is correct. Shadings show the division of the colony into sexual, asexual, and stolon areas.
This shows how the colonies would look if they were cut through the center and at right angles to the slide. For clarity only two tentacles are shown on each zooid, and the size of the zooids is exaggerated. The relative size of the hydranths is correct. Shadings show the division of the colony into sexual, asexual, and stolon areas.
One month
The hydranths in the pre-sexual colony are of three size groups. In the center of the colony is a single hydranth, probably the initiating hydranth, larger than any other in the colony. Surrounding it are a number of hydranths all the same size, about two-thirds that of the large single hydranth. In addition, many small hydranths can be seen in this area. Between this region, where the hydranths reach the two-thirds maximum size and the peripheral stolon, the hydranths grade in size, the largest being centrally located. At the periphery of the colony is a small area of stolon on which hydranths have not formed (Text-fig. 4A).
Two months
Generative zooids have appeared in the central part of the colony. They are smaller than the adjacent nutritives and continue to be so at 23°C. No single hydranth is larger than the others. In the central portion of the colony hydranths of maximum size are found interspersed with the sexual zooids. These bear medusae that grow, free themselves from the zooid and develop mature gametes. Surrounding the sexual area, consisting of sexual zooids and nutritive hydranths of maximum size, is an area containing only maximum size nutritive hydranths. Peripheral to this are the same three areas found at 1 month. The area of stolon devoid of hydranths is at least as large as the rest of the colony. On these peripheral stolons are scattered a few tiny hydranths. Some of the colonies bear clusters of six to twelve of these diminutive nutritive hydranths on the most peripheral parts of the stolon area (Text-fig. 4B).
Three months
The colony is similar to the 2-month stage. Differences are: the slide is almost completely covered with stolons; the sexual center has expanded so that it includes most of the colony; no area of graded hydranth size can be recognized; the peripheral stolon area, with no hydranths, is now very much smaller than it was at 2 months. By 3 months the stolons in the central part of the colony have ramified and anastomosed to the extent that it is impossible to separate a single stolon from the matte (Text-fig. 4c).
Four months
Between 3 and 4 months the colony completely covers the glass area. As might be expected the limitation in further overall growth in area has profound consequences for colony pattern. Sexual zooids are now distributed evenly throughout the entire colony. There are no remaining areas of asexual hydranths alone, nor areas of only stolon. The zooid density of the colony is less than it was in the sexual center at 2 or 3 months, but the proportion of sexual to nutritive zooids is higher. Unlike the 2- and 3-month old colonies, which bear buds mainly at the periphery, the colony at 4 months contains many small hydranths scattered throughout.
Five months
Between the fourth and fifth month no changes in the size or distribution of zooids occur. The colony still consists of a sparsely covered stolon matte bearing nutritive and sexual zooids. Small local areas consisting of a tangle of bright yellow stolons can now be seen randomly distributed on the stolon matte.
Control of sexuality
When circular portions of colonies containing sexual zooids were removed, stolons grew into the bare areas from the surrounding colony. In every case the zooids that appeared on these new stolons were nutritive hydranths (Textfig. 5). Only after about 1 month (slightly less than the time it took for a colony of this clone growing from a single hydranth to become sexual) did sexual zooids appear on the new stolons. The rate of regeneration was about the same in all colonies and was not affected by the age of the colony.
When a circular portion was cut from the center of a sexual colony stolons grew into the bare area. The zooids appearing on the stolons could be either sexual, asexual, or both. In fact, they are asexual, demonstrating that new hydranths were not forming under the same influences as did sexual zooids in the rest of the colony.
When a circular portion was cut from the center of a sexual colony stolons grew into the bare area. The zooids appearing on the stolons could be either sexual, asexual, or both. In fact, they are asexual, demonstrating that new hydranths were not forming under the same influences as did sexual zooids in the rest of the colony.
Single hydranths, combinations of sexual and asexual zooids, and groups of sexual and asexual zooids alone, removed from a 3-month old colony, will attach to microscope slides. Stolon outgrowth, and therefore attachment, took a day or two longer if the stolon matte was included. Stolons grew out from single hydranths and from clumps; nutritive hydranths formed on these new stolons. On every explant that included a piece of stolon matte, generative zooids were either maintained or appeared if none was present initially (Textfig. 6).
In the left-hand column are illustrated the different combinations of zooids, both with and without the connecting stolon matte, that were cut from a colony and allowed to attach to microscope slides. Columns two and three show the typical appearance of the colonies that grew from these explants after 9 and 15 days. In column three the upper legend describes the number and kind of zooids growing on the explanted matte; the lower legend, the zooids on the stolons that grew out from that matte. If a piece of matte was included in the explant, sexual zooids were maintained or appeared on it. New stolons growing either from the matte or from hydranths bore only nutritive zooids.
In the left-hand column are illustrated the different combinations of zooids, both with and without the connecting stolon matte, that were cut from a colony and allowed to attach to microscope slides. Columns two and three show the typical appearance of the colonies that grew from these explants after 9 and 15 days. In column three the upper legend describes the number and kind of zooids growing on the explanted matte; the lower legend, the zooids on the stolons that grew out from that matte. If a piece of matte was included in the explant, sexual zooids were maintained or appeared on it. New stolons growing either from the matte or from hydranths bore only nutritive zooids.
DISCUSSION
Colony growth
When a single Podocoryne hydranth is placed on a microscope slide, stolons grow from its proximal end and new hydranths appear on the stolons. The rate of growth of both stolon length and hydranth number is exponential and the ratio of stolon length to hydranth number is constant during the time that these two quantities can be measured. Exponential growth has also been described in Hydra (Loomis, 1950) and Cordylophora (Fulton, 1960).
As yet these studies have given no indication of the mechanism of this regulation. Studies with Tubularia (Rose, 1940; Steinberg, 1954; Tardent, 1955; and Tweedell, 1958) suggest that an inhibitor of hydranth formation is present in hydranths. Miller (1942), pointing out that the perisarc can limit diffusion both in and out of Tubularia stems, suggests that carbon dioxide may act as an inhibitor of hydranth formation.
For the first four of the 5 months that the colony was studied qualitatively, it grew symmetrically around its point of origin. Differentiation also followed a radial pattern, each of the five recognizable differentiations appearing first at the center, then extending toward the periphery. The five differentiations in the order of their appearance in the colony are:
Stolon: The outer portion of the colony consists of stolons devoid of hydranths. The proportion of the colony consisting of stolon alone varies, but is extremely high at 2 months.
Graded hydranth growth: An area of hydranths of graded sizes, the smallest at the periphery.
Hydranths two-thirds maximum size: An area of hydranths all the same size, but smaller than the largest in the colony.
Maximum size hydranths : An area of nutritive hydranths all of maximum size.
Sexual zone: An area of nutritive and generative zooids. The nutritives are the same maximum size as the hydranths of area 4.
Each zone spreads peripherally. It is only when the sexual area approaches the edge of the colony that the entire stolon system is covered with zooids. This gives the appearance of the sexual area ‘catching up’ with the other differentiations.
The distribution of maximum size hydranths among the sexual and adjacent areas suggests either that generative zooids play a mediating role in the control of hydranth size or that some common principle is associated with hydranth growth and generative zooid appearance.
Between the third and fourth months the zooid-bearing area expands to cover the entire stolon system. At the same time, or not long after, the expanding sexual center catches up with the hydranth area which had, up to that time, been a peripheral ring about it. As a result the radial pattern disappears and is replaced by an homogeneous distribution of nutritive and generative zooids. The change from heretogeneity to homogeneity is a consequence of expanding peripheral differentiation. It is not yet known if the time sequence that effects this change—in which the sexual area expands over the entire zooidbearing portion at the same time as this zooid-bearing portion covers the entire stolon system—is a coincidence resulting from the particular size to which these colonies were limited, or if this is an unavoidable consequence of growth and pattern control in the colony.
When the zooid-bearing area has completely covered the glass the total number of zooids in the colony is reduced and the ratio of nutritive zooids to generatives drops. As yet there is no way of knowing if this change in the zooid population is a consequence of the disappearance of the growing edge, or the arrival of the colony at a homogeneous, perhaps steady, state. It is noteworthy that, in this system, only the final stage of differentiation persists; that is, the result of the initial period of simultaneous expansion and differentiation is an apparently homogeneous colony.
During the fifth month new entities appear in the colony in an entirely new pattern. The localized areas in which stolons take on a glassy yellow appearance are distributed at random throughout the colony. This random distribution suggests considerable reorganization of the colony.
Berrill’s recent survey (Berrill, 1961) of developmental mechanisms, which draws much of its material from studies with colonial animals, concludes with a number of hypotheses regarding development. Berrill (1961, p. 401) points out that these conclusions should be of general application to vertebrate animals and to histogenesis, as well as to morphogenesis in the coelenterates, sponges, worms and tunicates that were the specific objects of the studies. Two of his conclusions are particularly relevant to this study. Berrill (1961, pp. 408 –9) predicts (1) radial expansion of growth and differentiation (although he would have the highest growth rate at the center), and (2) radical changes associated with stopping points in expansion. Both of these features are demonstrated during the development of P. carnea.
The control of sexuality
Generative zooids appear as the newest individual entities in the central portion of colonies of about 1 month of age. Sexual zooids appear precociously (Braverman, 1962b) in colonies treated with CO2. Moreover, in the colonies exposed to a CO2 excess, nutritive zooids transformed to generatives. The initial appearance of sexual differentiation at the center of the colony prompts us to ask how this differentiation is localized. We must not only demonstrate that a substance can engender sexuality, but also show how it can be locally concentrated in such a way as to account for the concentric pattern.
Experimental induction of sexuality by CO2 is in itself no proof that the same mechanism operates in the developing animal. Nevertheless it is tempting to consider this gas as the sexualizing agent, since endogenous mechanisms for its local concentration do exist. During the early period of colony growth it is likely that the rate of diffusion of CO2 out of the stolons is such that the critical level required for sexuality is never reached.
The extirpation and isolation experiments show that only nutritives regenerate on stolons growing into an extirpated area of the colony. Sexual zooids will grow from a small piece of stolon matte removed from the sexual center, but not from new stolons growing from this matte. Sexuality is associated only with the dense stolon matte that exists in the older colonies. The same conclusions were drawn by Hauenschild (1954), who found a variety of Hydractinia echinata, a similar marine hydroid, which formed a dense stolon matte only if its growth was restricted. Furthermore, generative zooids appeared only on restricted colonies.
This association of the dense stolon matte with sexual zooids suggests a mechanism for concentrating CO2 locally. In the older and more central portions of the colony there are more stolons and they lie closer together than in the younger peripheral portions. Clearly, CO2 would diffuse more easily from an isolated stolon than from a stolon matte. It is possible, too, that the chitinous perisarc that surrounds the stolon might thicken with age, reducing diffusion out of the coenosarc. Thus the crowding of stolons that occurs first in the central part of the colony would tend to raise the CO2 concentration there above the concentration in the more diffuse peripheral stolons.
Nutritive hydranths can transform into generative zooids when subjected to a high concentration of CO2. But, in the central portions of the colony where a high CO2 concentration can build up even when no excess CO2 has been added to the water, nutritives did not affect this transformation. Why did only new zooids evidence sexuality? The first and most obvious answer is that young zooids might be more sensistive to the sexualizing effect of CO2 than mature zooids.
There is another possible explanation that does not require this assumption of selective sensitivity. Mature hydranths have a mechanism lacking in young buds for lowering their internal CO2 concentration. During the first stages of its formation, a hydranth bud has no mouth. Before the mouth breaks through, it is a cul-de-sac (Berrill, 1949) exposed to the CO2 of the stolon with no means of ventilation. The CO2 concentration in a mature hydranth’s gastrovascular cavity, on the other hand, is equilibrated with the surrounding water, not only through its body wall but also through the open mouth. In water containing excess CO2, mature hydranths can be seen with their mouths open and their gastroderms everted.
If these assumptions regarding equilibration of external and internal environment are true, only under circumstances of high external CO2 would the mature nutritive hydranths be exposed to sexualizing concentrations of carbon dioxide.
In terms of this hypothesis, stolons are crowded more and more closely together as the colony grows. Diffusion out of the stolons is impeded and the internal carbon dioxide concentration is raised. Buds have no mouths and are spherical, a shape that offers the least possible ratio of surface area to volume. Not having the means of equilibration and diffusion the bud develops under a local concentration of CO2. If this is below a threshold value the bud develops as a nutritive hydranth; if it is above it, it develops as a generative zooid.
RÉSUMÉ
Etudes sur la différenciation des hydráires. La croissance de la colonie et P acquisition de la sexualité
Lorsque l’hydraire marin Podocoryne carnea est cultivé sur des lames porte-objets dans des conditions contrôlées, il constitue un excellent matériel pour l’étude des facteurs de différenciation, de la régulation et du contrôle au cours du développement, et de la conformation générale. La croissance des stolons et des hydranthes est exponentielle pour les uns comme pour les autres et la proportion d’hydranthes par rapport aux stolons est constante.
Par l’analyse qualitative de la croissance de la colonie pendant une période de 5 mois, on arrive à délimiter cinq états distincts qui, au cours des trois premiers mois, apparaissent au centre de la colonie et s’étendent vers la périphérie du système organisé radiarement. Lorsque l’expansion de la colonie prend fin, le mode centrifuge de différenciation est remplacé par une distribution apparemment accidentelle et le nombre ainsi que la proportion des zoïdes formant la colonie se modifient.
Les expériences d’extirpation et d’isolement indiquent que la sexualité n’est pas transmise des parties sexuées de la colonie aux parties asexuées, et qu’elle est associée au feutrage dense des stolons. Il est suggéré une théorie qui rendrait compte de l’origine de la sexualité en invoquant une diminution de la perméabilité des stolons à des substances qui entraînent la différenciation sexuelle.
ACKNOWLEDGEMENTS
The research reported here was carried out at the Stazione Zoológica, Naples, Italy, during the tenure of a NATO post-doctoral fellowship. Particluar thanks are due to Professor J. Runnstrôm for his criticism of the manuscript and to Miss Hanne Grunbaum for preparing the graphs.