Reverse development in Cnidaria

Abstract

Cnidarians have long been considered simple animals in spite of the variety of their complex life cycles and developmental patterns. Several cases of developmental conversion are known, leading to the formation of resting stages or to offspring proliferation. Besides their high regenerative and asexual-reproduction potential, a number of cnidarians can undergo ontogeny reversal, or reverse development: one or more stages in the life cycle can reactivate genetic
programs specific to earlier stages, leading to back-transformation and morph rejuvenation. The switch is achieved by a variable combination of cellular processes, such as transdifferentiation, programmed cell death, and proliferation of interstitial cells. The potential for ontogeny reversal has limited ecological meaning and is probably just an extreme example of a more general strategy for withstanding unfavourable periods and allowing temporal persistence ofspecies in the environment

Full text

Reverse development in Cnidaria
S. Piraino, D. De Vito, J. Schmich, J. Bouillon, and F. Boero
Abstract: Cnidarians have long been considered simple animals in spite of the variety of their complex life cycles and
developmental patterns. Several cases of developmental conversion are known, leading to the formation of resting
stages or to offspring proliferation. Besides their high regenerative and asexual-reproduction potential, a number of cni
-
darians can undergo ontogeny reversal, or reverse development: one or more stages in the life cycle can reactivate ge
-
netic programs specific to earlier stages, leading to back-transformation and morph rejuvenation. The switch is
achieved by a variable combination of cellular processes, such as transdifferentiation, programmed cell death, and pro
-
liferation of interstitial cells. The potential for ontogeny reversal has limited ecological meaning and is probably just an
extreme example of a more general strategy for withstanding unfavourable periods and allowing temporal persistence of
species in the environment.
Résumé : Depuis longtemps, on considère les cnidaires comme des animaux simples, malgré la variété de leurs cycles
biologiques et de leurs patterns de développement complexes. Il existe plusieurs cas de conversions au cours du déve
-
loppement qui mènent à la formation de stades de repos ou à la prolifération de rejetons. En plus de leur fort potentiel
de régénération et de reproduction asexuée, plusieurs cnidaires peuvent subir un renversement de leur ontogenèse ou un
développement inversé : un ou plusieurs stades dans le cycle biologique peuvent réactiver des programmes génétiques
spécifiques de stades antérieurs, ce qui donne une rétro-transformation et un rajeunissement des morphes. Le revire
-
ment est provoqué par une combinaison variable de processus cellulaires, tels que la transdifférentiation, la mort cellu
-
laire programmée et la prolifération de cellules interstitielles. Le potentiel de renversement ontogénique a une
signification écologique limitée; il s’agit probablement d’un état extrême d’une stratégie plus générale pour survivre
aux périodes défavorables et permettre à l’espèce de se maintenir dans le temps dans son milieu.
[Traduit par la Rédaction] Piraino et al. 1754
The diversity of cnidarian ontogeny
The complexity of hydrozoan life cycles and the
“immortal” jellyfish
The life cycle of the Hydrozoa is typically characterized
by the alternation of three life stages: the planula larva, the
postlarval polyp stage (mainly colonial), and the adult stage,
i.e., the medusa (Boero et al. 2002). The ciliated planula has
been also interpreted as a motile gastrula, and the polyp as
the true larval stage (Bouillon and Boero 2000). The planula
lives only from a few hours to a few days before metamor
-
phosis, whereas both polyp and medusa stages can have a
much longer life-span (from days to years) (Bouillon 1994).
This life cycle is usually invoked as being paradigmatic for
the Medusozoa group (Hydrozoa, Scyphozoa, Cubozoa, and
Staurozoa; see Marques and Collins 2004). However, side
-
ways deviations from the norm are common (Fig. 1). Ancestral
taxa (e.g., Actinulidae, Narcomedusae, and Trachymedusae)
lack a polyp stage (Bouillon and Boero 2000); in contrast,
members of derived families of the Hydroidomedusa group
(e.g., Aglaopheniidae, Plumulariidae, and Sertulariidae) re
-
duced, in part or totally, the medusa stage by paedomor-
phosis (Boero and Bouillon 1987). In many cases, the polyp
becomes the adult stage, and gametes can differentiate either
in the hydranth body column, (e.g., in Hydra spp.) or within
highly reduced medusae (fixed gonophores) retained by
polymorphic polyp colonies. Other species can produce
short-lived, free medusoid morphs characterized by reduced
swimming and (or) feeding potential (e.g., Macrorynchia
philippina (Kirchenpauer, 1872) (Gravier 1970) or Nemale
-
cium lighti (Hargitt, 1924) (Gravier-Bonnet and Migotto
2000)). These morphs, called swimming gonophores by
Boero and Bouillon (1989), are liberated with already mature
gonads, and their dispersal from the polyp colony is usually
limited (Boero et al. 2002).
In many hydrozoan species, the normal pace of the life
cycle can be modified at some stages. Both planula and
polyp stages can shrink and rest in low-metabolic-cost states
(cysts, dormant hydrorhizae) to survive critical periods, be
-
ing ready to reactivate cell proliferation and morphogenesis
at the onset of favourable seasons (Boero et al. 1992). Polyp
colonies can last for several years, annually producing large
batches of sexually competent medusae, followed by peri
-
odic colony shrinkage and recurrent cycles of regenerative
processes, without any obvious sign of senescence (Forrest
1963; Müller 1996). This process, also known as renovation,
is a type of regeneration. In nature, renovation is a normal
process in the hydroid colonies of many Leptomedusae, and
may be repeated several times during their life-span, depend
-
ing upon intrinsic factors (i.e., age of the colony, health of
the colony, etc.) or sometimes upon the ecological condi
-
Can. J. Zool. 82: 1748–1754 (2004) doi: 10.1139/Z04-174 © 2004 NRC Canada
1748
Received 28 May 2004. Accepted 23 November 2004.
Published on the NRC Research Press Web site at
http://cjz.nrc.ca on 2 February 2005.
S. Piraino,
1
D. De Vito, J. Schmich, J. Bouillon, and
F. Boero. Dipartimento di Scienze e Tecnologie Biologiche
ed Ambientali, Università di Lecce, via per Monteroni,
73100 Lecce, Italy.
1
Corresponding author (e-mail: stefano.piraino@unile.it).

© 2004 NRC Canada
Piraino et al. 1749
tions they face. Renovation is usually accomplished by
changes in morphology and functional commitment of al
-
ready specialized cells, a process known as transdifferenti
-
ation (Schmid 1992). Interstitial cells (I-cells) may partly
contribute to the renovation process, but they are not a fun
-
damental requisite. In fact, not all those species undergoing
colony or polyp renovation have I-cells (Brien 1941). More
-
over, cell replacement and differentiation along the body
column, regeneration, and asexual reproduction are equally
active processes in Hydra spp. populations experimentally
deprived of I-cells (Brien and Reniers-Decoen 1955; Diehl
and Bouillon 1966; Marcum and Campbell 1978). Renova
-
tion is rare, or perhaps less easy to observe, in Antho
-
medusae polyps, but it is known to occur (e.g., in Tubularia
spp.; Tardent 1963). In marine Hydrozoa, regeneration and
cell replacement occur by activation of epithelial smooth
muscle cells, which then become stem cells by dediffer
-
entiation (Schmid 1974).
At different levels, both polyp and medusa stages of
Hydroidomedusae can exhibit a combination of budding and
regenerative processes (Boero et al. 2002). It is known that
medusae of several species have the ability to bud (but not
to metamorphose into) polyp structures before or even after
initiation of processes of sex-cell determination (Boero et al.
2002). In addition, direct medusa budding from liberated
medusae is known for several species. Large populations can
be built up in time and space, allowing the species to cross,
by following the currents, deep oceanic waters, where
planulae may have few or no chances to settle. This type of
medusa budding involves transdifferentiation of preexisting
cell types on the manubrium, tentacular bulbs, radial canals,
exumbrellar rim, marginal canal, subumbrellar rim, or even
gonads (for further discussion and a species list see Boero et
al. 2002 and references therein).
An additional potential has been discovered in a few spe
-
cies whose polyp stage can also be reformed from regressing
tissues of sexual stages (Figs. 1C–1D). Müller (1913) first
described such “reverse ontogeny” in Hydrozoa, having ob
-
Fig. 1. Models of some hydrozoan life cycles. (A) Hydra spp. have no medusa stage and no planula larva. The gonads originate on the
polyp body column; the new polyp hatches directly from a cuticled embryo. Asexual budding of new polyps is the most frequent mode
of reproduction. (B) Hydractinia echinata (Fleming, 1828) has a highly reduced medusa stage that remains attached to the polyp in the
form of fixed gonophores. Fertilization leads to the formation of a planula larva, which will undergo metamorphosis to develop into a
polyp colony. (C) The life cycle of Hydractinia (Podocoryna) carnea includes a swimming medusa stage and a planula larva. Artificial
detachment of late medusa buds leads to the development of medusae that are complete but reduced in size. In contrast, artificially de
-
tached early medusa buds are capable of transformation back into polyps. Such reverse transformation is not usually achieved by late
medusa buds or liberated medusae. (D) Turritopsis nutricula has a typical three-stage life cycle: planula, polyp, and medusa. However,
medusae at all stages of development retain the potential for life-cycle reversal; even spent medusae do not die, but transform back
spontaneously into new polyp colonies.

served the formation of stolons and polyps from regressing
medusa buds mechanically detached from the gonozooids of
Hydractinia (Podocoryna) carnea (M. Sars, 1846). Only early
medusa buds could undergo reverse development, whereas
late buds developed into small but fully formed medusae
(Frey 1968; Schmid 1972). Hauenschild (1956) observed the
same potential in Eleutheria dichotoma Quatrefages, 1842
and Kakinuma (1969) reported a comparable process in
Cladonema sp. and Cladonema uchidai Hirai, 1958, with
the formation of hydrorhizal stolons from early medusa buds
experimentally detached from the gonozooids. Also,
Bavestrello et al. (2000) described a complex fragmentation
process in polyp colonies of Hydractinia (Stylactaria)
pruvoti (Motz-Kossovska, 1905) that is induced by strong
water movements: fragmentation of gonozooids bearing
medusa buds led to the early liberation of developing
eumedusoids. As in H. carnea, mature eumedusoids of
H. pruvoti started swimming and spawning, while early
stages underwent a regressive transformation giving rise to
stolons and polyps.
It is now known that the reverse transformation of
H. carnea is due not only to regeneration by proliferation of
I-cells, but also to the occurrence of cell-transdifferentiation
processes (Schmid 1972; Schmid et al. 1982; Schmid and
Alder 1984; Alder and Schmid 1987). Transdifferentiation is
defined as a change in commitment and gene expression of
well-differentiated, non-cycling somatic cells to other cell
types directly or through their reversion to undifferentiated
cells (Okada 1991). In Hydroidomedusae, transdifferenti-
ation might be a common phenomenon in budding and re-
generation, especially where no or few I-cells are located
(Brien 1941). Manubrial budding of secondary medusae in a
few species belonging to different families (Bougainvillidae,
Hydractiniidae, Rathkeidae) originates only from differenti-
ated ectodermal cells, with no aid from the I-cell compart-
ment (Bouillon 1962; see also Boero et al. 2002 and
references therein). All tissues of the newly formed medusae
derive from transdifferentiating ectoderm. However, in all
other cases, the contribution of transdifferentiation processes
is still unstudied.
The fate of developing buds of Sarsia tubulosa (M. Sars,
1835) can be altered by sudden changes in temperature
(Berrill 1953). Medusa buds can develop, depending on wa
-
ter temperature, either in normal free medusae or in sessile
eumedusoids with rudimentary tentacles and lacking both
mouth and ocelli. Moreover, when rearing of S. tubulosa is
shifted to the temperature range 6–8 °C, medusa buds can be
transformed back into polyp buds (Werner 1963). This pro
-
cess suggests the involvement of cell transdifferentiation and
programmed cell death, as occurs in H. carnea.
In most cases, ontogeny reversal occurs in incompletely
developed medusa buds with no sign of gonads. The poten
-
tial for reverse ontogeny is suppressed in late bud stages
with developing gonads, and in liberated medusae. The onset
of sexual reproduction might thus be regarded as the point of
no return in the ontogenetic sequence in any living organ
-
ism, ultimately leading to aging and death (Stearns 1992). In
cnidarians, as in most other organisms, germ-cell differentia
-
tion seems to represent a signal for irreversible somatic de
-
termination, preventing back-transformation of late medusa
buds. However, studies on Turritopsis nutricula Mccrady,
1859 (Anthomedusae, Clavidae) (Fig. 2) demonstrated that
this rule can be broken. Bavestrello et al. (1992) showed that
newly liberated medusae of this species can revert to polyp
stages; later, Piraino et al. (1996) showed that ontogeny re
-
versal occurs at all stages of medusa growth, including the
adult stage with mature gonads (Fig. 2d). Thus, rejuvenation
of the medusa morph into the polyp stage is always possible
in this species, and is not constrained by gamete differentia
-
tion. In T. nutricula, rejuvenation involves the contribution
of both transdifferentiation and I-cell-proliferation processes
(Piraino et al. 1996), together with the activation of cell-
death programs (Carlà et al. 2003). These medusae react to
unfavourable conditions (e.g., starvation, mechanical stress,
temperature and salinity changes), starting reverse transfor
-
mation into the polyp stage: a complete reduction of all
medusa-specific organs and tissues, followed by differentia
-
tion of new, polyp-specific cell types and the formation of
the polyp morph (Piraino et al. 1996). In T. nutricula, even
senescence (started by the sexual maturation of the gonads)
is a stress factor that consistently induces reverse transfor
-
mation of medusae. This process would be hardly more re
-
markable if a butterfly were able to revert to its caterpillar
stage. It must be considered a true metamorphosis, but in the
opposite direction to larval metamorphosis. At the molecular
level, it is reflected in a distinct pattern of stage-specific
gene expressions (Spring et al. 2000; Yanze et al. 2001). Re-
cently, young medusae of T. nutricula and even H. carnea
were shown to undergo reverse transformation and rejuvena-
tion to the polyp stage by exposure to cesium, a chemical
inducer of metamorphosis (J. Schmich, unpublished data).
In addition, reverse development has been recently demon-
strated to occur also in Leptomedusae: young, immature
medusae of Laodicea undulata (Forbes and Goodsir, 1851)
are able to revert to the polyp stage (De Vito et al.
2
). Steps
leading to back-transformation in L. undulata are compara
-
ble to those observed in young T. nutricula medusae
(Piraino et al. 1996). The first signals of ontogeny reversal
are thinning of the mesoglea and shortening of the tentacles.
The medusa then becomes unable to swim and settles to the
bottom of the culture vessel. The umbrella is reduced within
a few hours, and the whole medusa is transformed into a
cyst-like rounded mass of poorly differentiated cells.
Periderm-covered hydrorhizae develop soon and the first
“regenerated” polyp appears within 48–72 h of the initiation
of the process. Within a month, new gonothecae are already
reformed from the rejuvenated colony and, from there, sec
-
ondary medusae are newly budded. Medusae of L. undulata
with mature gonads were not observed and it is still not
known if they retain the potential for ontogeny reversal also
in sexually competent stages, as in T. nutricula.
The formation of stolons by back-transformation was also
observed in mechanically detached medusoids of Cytaeis
schneideri (Motz-Kossowska, 1905) (S. Piraino, unpublished
observations) and from a newly liberated medusa of Clytia
hemisphaerica (L., 1767) (De Vito et al.
2
). In both cases,
© 2004 NRC Canada
1750 Can. J. Zool. Vol. 82, 2004
2
D. De Vito, S. Piraino, J. Schmich, J. Bouillon, and F. Boero. Laodicea undulata (Forbes and Goodsir 1851) (Cnidaria, Hydrozoa), a
leptomedusa with the potential for reverse ontogeny. Submitted for publication.

however, stolon formation was not followed — at least in
our laboratory experiments — by polyp development.
Back-transformation (Rückbildung) in Scyphozoa
In Scyphozoa, the paradigm of the medusozoan life cycle
(planula–polyp–medusa) is retained, but with the distinctive
addition of a pre-adult stage, the ephyra, produced by trans
-
verse fission of the polyp (strobilation; see Spangenberg
1965, 1968). The medusa phase predominates in the life cy
-
cle of most scyphozoans, and the polyp stage (scyphistoma)
is in some cases significantly reduced or even completely
suppressed (e.g., in the genera Atolla and Pelagia). Species
of two coronate genera (Nausithoe, Thecoscyphus) lack a
free medusa stage (Jarms 1997) but, unlike many Hydrozoa,
Scyphozoa never completely lose a medusoid stage. Modifi
-
cations of the basic scyphozoan cycle are the production of
polyp resting stages, called podocysts (known in both
Semeostomeae and Rhizostomeae; for a list of species see
Arai 1997), and of “planula-cysts” (in the genus Cyanea;
Brewer 1976 and references therein). Regeneration can be as
powerful in the Scyphozoa as in the Hydrozoa. Scypho
-
medusae can regenerate lost organs and tissues, and new
coronate polyp colonies can grow from individually excised
polyps. A striking example of regeneration and asexual re
-
production by means of “pseudoplanulae” was first discov
-
ered in Chrysaora sp. by Hérouard (1909), who described
the autotomy of polyp tentacles leading to a ciliated
“planula-like” stage, which was capable of locomotion and
settlement to produce secondary polyps. Afterwards, Lesh-
Laurie and Corriel (1973) demonstrated consistent regenera
-
tion of complete polyps from isolated tentacles of Aurelia
aurita (L., 1758) scyphistomae. The process of propagule
formation by autotomy of tentacles is also known in Hydro
-
zoa (e.g., the limnomedusa Armorhydra janowiczi Swedmark
and Teissier, 1958 (Campbell 1974); the hydroidomedusa
H. pruvoti (Bavestrello et al. 2000)).
The first reports of true reverse development in Cnidaria
were given by Hadzi (1909a, 1909b, 1912), who described
back-transformation (Rückbildung) of ephyrae of Chrysaora
hysoscella (L., 1766) to scyphistoma polyps under unfavour
-
able environmental conditions. Reverse transformation of
ephyrae into scyphopolyps was later observed also in Rhizo
-
stoma pulmo (Macri, 1778) (Paspaleff 1938) and A. aurita
(Thiel 1963). Comparably, under unfavourable rearing con
-
ditions, the ephyrae of eight species of Coronatae may un
-
dergo ontogeny reversal, first regressing to planuloid masses
of cells, and growing into polyps later (from weeks to
months) (reviewed by Jarms 1997). The ability of Nausi
-
thoidae to produce planuloids is regarded as an ontogenetic
strategy to minimize energy losses and enable survival and
propagation in low-energy habitats, such as caves or deep
seas (Jarms 1997). In this framework, Nausithoe planulo
-
© 2004 NRC Canada
Piraino et al. 1751
Fig. 2. Reverse development of T. nutricula.(a) Free-living healthy medusa. (b) Transforming medusa, with the everted subumbrellar
muscle layer (at left) and outer manubrium (at right). BrdU staining of DNA replicating nuclei. (c) Ball-like or cyst-like stage of a
transforming medusa. BrdU staining of DNA replicating nuclei. (d ) Butterfly-shaped remnant of an adult medusa of T. nutricula (set
-
tled on a glass slide) with four mature male gonads (white arrows), producing a hydrorhizal stolon characteristic of the polyp stage
(black arrow). (e) A newly formed polyp from reverse development of a medusa (within 36–48 h at 24 °C). Scale bars: (a)1mm;
(b) 500
µ
m; (c) 300
µ
m; (d and e) 500
µ
m.

phorus (Werner, 1971) shows the highest degree of special
-
ization by retaining within the polyp peridermal tube all
strobilae, which are eventually transformed into the
planuloid stage, crowling out of the tube and settling a short
distance from the parental polyp. As noted by Jarms (1997),
this apogamic life cycle precludes the benefits of sexual re
-
combination and should be considered an evolutionary dead-
end.
Reversible metamorphosis in Anthozoa
Anthozoa are considered the ancestral group of Cnidaria
(Bridge et al. 1992; Collins 2002). Their simple life-cycle
pattern consistently lacks a medusa stage, whereas dispersal
is mainly delegated to planula larvae or occurs by fragmen
-
tation and drift of asexual propagules. Recently, Fautin
(2002) reviewed the modes of cnidarian reproduction: two
main modes of asexual reproduction, scissiparity (either
transverse and longitudinal fission) and budding (sensu
Bouillon 1994) are mainly found in Anthozoa (with the ex
-
ception of Ceriantharia). Secondary modes of asexual repro
-
duction are less common, like laceration of the pedal disk in
some Actiniaria, or pinnitomy in some octocorals, whose
tentacular pinnules can autotomize and fall from the parent
colony to give rise to new polyps and, then, new colonies
(Gohar 1940). Comparably, the tentacles of boloceroidid sea
anemones can reconstitute new polyps (Cutress 1979; Pearse
2002). Tentacle or pinnule autotomy is a homologue of the
above-mentioned formation of pseudoplanulae in Scypho-
zoa, with regeneration of complete polyps from autotomized
or isolated tentacles of scyphistomes.
Polyp bail-out (Sammarco 1982) is an adaptive response
to environmental stress, providing a way of escape and adult
dispersal in a pocilloporid coral (also observed in Hydrozoa;
see Gravier-Bonnet 1992). Polyps are able to leave their
hard skeleton and detach from the parent colony. This pro-
cess implies a partial rearrangement of the running genetic
machinery, switching off the progressive assembly of inor
-
ganic skeletal components and partly reshaping the polyp
column. A further step towards reactivation of early develop
-
mental programs in Anthozoa is found again within the fam
-
ily Pocilloporidae. Richmond (1985) described “reversible
metamorphosis” in Pocillopora damicornis (L., 1758), whose
primary polyps may, under unfavourable conditions, aban
-
don the primary corallites shortly after metamorphosis and
regress to a planula-like stage. Such secondary larvae could
search for appropriate sites of settlement and metamorphose
again. However, nothing is known about such reverse trans
-
formation at the cellular level, including the potential
contribution from the I-cell compartment or from trans
-
differentiation processes.
Discussion
Many metazoans, namely Porifera, Cnidaria, Platyhel
-
minthes, Placozoa, Nemertea, Rotifera, Nematoda, Acantho
-
cephala, Arthropoda, Tardigrada, Ectoprocta, and Tunicata,
are able to withstand unfavourable conditions by producing
resting stages through different processes (e.g., encystment,
degrowth or reduction in size, survival budding; see Piraino
et al. 1996 and references therein). These can be envisaged
as developmental conversions along the normal ontogenetic
path. Reverse development, however, seems to be a distinc
-
tive feature of cnidarians. A growing mass of information
suggests that the potential for ontogeny reversal is more
widespread in Cnidaria than was previously thought.
In most cases, the transformation of body portions into
earlier structures represents a way to produce either resting
or dispersive stages without sexual reproduction. This possi
-
bility exists in both polyps (which give rise to planula-like
structures) and young medusae (which give rise to polyps).
In the case of T. nutricula, and a small group of other spe
-
cies, however, a completely different interpretation is called
for. The medusae of T. nutricula, in fact, invariably trans
-
form into polyps under laboratory conditions (Piraino et al.
1996) and have been called immortal by the nonscientific
press. From each medusa undergoing reverse ontogeny, a
new colony can potentially originate, which in turn will pro
-
duce new copies of the original medusa, leading to an expo
-
nential increase in the total number of clonemate offspring.
To date, this potential has been repeatedly demonstrated in
the laboratory with a large number of observations in
T. nutricula (Piraino et al. 1996). By implication, they could
attain immortality. But if this potential for avoiding death
were expressed in the field, these animals would saturate the
world’s oceans! Instead, we see that they suffer mortality,
and population growth is kept under control. It is probable
that such a great potential for ontogeny reversal has limited
meaning and is just an extreme example of a more general
strategy for withstanding unfavourable periods and allowing
temporal persistence of species in the environment.
As already mentioned, the potential for reverse develop-
ment could be present, but hidden, in other cnidarians. In
laboratory experiments the lack of unknown field factors
(e.g., species-specific relationships with marine bacteria) may
obscure possible back-transformation, but the genetic ma-
chinery may remain functional, as is also demonstrated by
the artificial activation of reverse development in adult medu
-
sae of H. carnea by chemical inducers of metamorphosis.
To date, it seems that only scyphozoan ephyrae, but not
adult medusae, can revert to the polyp stage and, within
Anthozoa, only coral primary polyps can revert to planula
larvae. However, available information on Staurozoa and
Cubozoa is scant and the life cycles of many cnidarians still
need to be elucidated. Therefore, we can reasonably assume
that new cases of reverse development could be discovered.
Other species can retain the potential for transformation, but
(i) it can be transiently expressed only under extraordinary
conditions, or (ii) the genetic program involved in the trans
-
formation process can be activated, but only few species re
-
tain the whole signalling cascades needed for the completion
of morph reversal.
As often happens in experimental biology, extreme cases
can become a paradigm for uncovering general, albeit less
extreme, patterns. Boero et al. (1997) theorized that the ge
-
nome of hydromedusan species with complex life cycles
might be (functionally) divided into two main portions, one
coding for the polyp and one coding for the medusa stage.
The two main stages of the life cycle, in fact, possess differ
-
ent cell types and organs. As was noted by Boero et al.
(1998), hydrozoan polyps are diploblastic, whereas their
medusae are triploblastic and possibly coelomate, since their
subumbrellar cavity originates by cavitation (i.e., schizo
-
© 2004 NRC Canada
1752 Can. J. Zool. Vol. 82, 2004

coely) within a third cell layer that, in turn, differentiates
into the striated muscle lining the subumbrella (striated mus
-
cle is mesodermic in all other Metazoa). With ontogeny re
-
versal by transdifferentiation, medusa cells transform into
polyp cells, with a total reprogramming of their fate. The
one-way route of developmental programs, always leading
to death, is inverted! In T. nutricula, ontogeny reversal is
caused by both environmental cues (sublethal stress) and se
-
nescence, activating a trigger that can switch on or restart
the “reading” of the genetic information almost from the
very beginning. As is often the case, epigenetic and genetic
cues concur in defining developmental patterns. The discov
-
ery of such switching mechanisms is a further step towards
clarifying the mechanisms of senescence and rejuvenation,
and might represent a breakthrough in developmental biol
-
ogy in general (Boero 1998).
Acknowledgments
Many thanks are extended to Prof. Volker Schmid (Basel),
who guided us to an understanding of reverse development
and transdifferentiation in Hydrozoa and provided valuable
comments on the manuscript. Financial support was pro
-
vided by Ministero dell’Istruzione, dell’Università e della
Ricerca (60%, Cofinanziamento progetti di ricerca di inter-
esse nazionale projects and Fondo per gli investimenti dell
ricerca di base), the Administration of the Province of Lecce,
Istituto centrale per la ricerca scientifica e tecnologica
applicata al mare (“Identificazione e distribuzione delle spe-
cie non indigene nel Mediterraneo” project), the European
Union (Marie Curie contract HPMD-CT-2001-00099, and
the Marine Biodiversity and Ecosystem Functioning Net-
work of Excellence), the National Science Foundation of the
USA (Partnerships for Enhancing Expertise in Taxonomy:
NSF DEB-9978131 project on the Hydrozoa).
References
Alder, H., and Schmid, V. 1987. Cell cycles and in vitro trans
-
differentiation and regeneration of isolated, striated muscle of
jellyfish. Dev. Biol. 124: 358–369.
Arai, M.N. 1997. A functional biology of Scyphozoa. Chapman
and Hall, London.
Bavestrello, G., Sommer, C., and Sarà, M. 1992. Bi-directional
conversion in Turritopsis nutricula (Hydrozoa). In Aspects of
hydrozoan biology. Edited by J. Bouillon, F. Boero, F. Cicogna,
J.M. Gili, and R.G. Hughes. Sci. Mar. 56: 137–140.
Bavestrello, G., Puce, S., Cerrano, C., Castellano, L., and Arillo A.
2000. Water movement activating fragmentation: a new dis
-
persal strategy for hydractiniid hydroids. J. Mar. Biol. Assoc.
U.K. 80: 361–362.
Berrill, N.J. 1953. Growth and form in gymnoblastic hydroids. VII.
Growth and reproduction in Syncoryne and Coryne. J. Morphol.
92: 273–302.
Boero, F. 1998. Ontogeny. In 1999 McGraw–Hill yearbook of sci
-
ence and technology. McGraw–Hill, New York. pp. 265–258.
Boero, F., and Bouillon, J. 1987. Inconsistent evolution and paedo
-
morphosis among the hydroids and medusae of the Athecatae/
Anthomedusae and the Thecatae/Leptomedusae (Cnidaria, Hydro
-
zoa). In Modern trends in the systematics, ecology and evolution
of hydroids and Hydromedusae. Edited by J. Bouillon, F. Boero,
F. Cicogna, and P.F.S. Cornelius. Clarendon Press, Oxford.
pp. 229–250.
Boero, F., and Bouillon, J. 1989. An evolutionary interpretation of
anomalous medusoid stages in the life cycle of some
Leptomedusae (Cnidaria). In Reproduction, genetics and distri
-
butions of marine organisms. Edited by J.S.Ryland and P.A. Ty
-
ler. Olsen and Olsen, Fredensborg, Denmark. pp. 37–41.
Boero, F., Bouillon, J., and Piraino, S. 1992. On the origins and
evolution of hydromedusan life cycles (Cnidaria, Hydrozoa). In
Sex origin and evolution. Edited by R. Dallai. Sel. Symp.
Monogr. U.Z.I. 6: 59–68.
Boero, F., Bouillon, J., Piraino, S., and Schmid, V. 1997. Diversity
of hydromedusan life cycles: ecological implications and
evolutionary patterns. In Proceedings of the 6th International
Conference on Coelenterate Biology, Noordwijkerhout, the
Netherlands, 16–21 July 1995. Edited by J.C. den Hartog,
L.P. van Ofwegen, and S. van der Spoel. Nationaal Natuur
-
historisch Museum, Leiden, the Netherlands. pp. 53–62.
Boero, F., Gravili, C., Pagliara, P., Piraino, S., Bouillon, J., and
Schmid, V. 1998. The cnidarian premises of metazoan evolution:
from triploblasty, to coelom formation, to metamery. Ital. J.
Zool. 65: 5–9.
Boero, F., Bouillon, J., Piraino, S., and Schmid, V. 2002. Asexual
reproduction in the Hydrozoa (Cnidaria). In Reproductive biol
-
ogy of invertebrates. Vol. XI. Progress in asexual reproduction.
Edited by R.N. Hughes. Oxford & IBH Publishing Co. Pvt. Ltd.,
New Delhi. pp. 141–158.
Bouillon, J. 1962. Sur le bougeonnement médusaire manubrial de
Rathkea octopunctata. Ann. Soc. R. Zool. Belg. 92: 7–25.
Bouillon, J. 1994. Embranchement des Cnidaires (Cnidaria). In
Traité de Zoologie. Vol. 3(2). Edited by P.P. Grassé and D.
Doumenc. Masson, Paris. pp. 1–28.
Bouillon, J., and Boero, F. 2000. Phylogeny and classification of
Hydroidomedusae. The Hydrozoa: a new classification in the
light of old knowledge. Thalassia Salentina, 24: 1–46.
Brewer, R.H. 1976. Larval settling behavior in Cyanea capillata.
Biol. Bull. (Woods Hole), 150: 183–199.
Bridge, D., Cunningham, C.W., Schierwater, B., DeSalle, R., and
Buss, L. 1992. Class-level relationships in the phylum Cnidaria:
evidence from mitochondrial genome structure. Proc. Natl.
Acad. Sci. U.S.A. 89: 8750–8763.
Brien, P. 1941. Remarques au sujet des conceptions relatives à
l’existence et à la pérennité, chez les hydroïdes, d’une réserve
embryonaire et d’une lignée germinale, à propos du bourge
-
onnement et des potentialités de l’ectoderme de Cladonema
radiatum (Duj.). Ann. Soc. R. Zool. Belg. 72: 37–62.
Brien, P., and Reniers-Decoen, M. 1955. La signification des cellules
interstitielles des Hydres d’eau douce et le problème de la
réverse embryonnaire. Bull. Biol. Fr. Belg. 89: 258–325.
Campbell, R. 1974. Cnidaria. In Reproduction of marine inverte
-
brates. Vol. I. Acoelomate and pseudocoelomate metazoans.
Chap. 3. Edited by A.C. Giese and J.S. Pearse. Academic Press,
New York. pp. 133–199.
Carlà, E.C., Pagliara, P., Piraino, S., Boero, F., and Dini, L. 2003.
Morphological and ultrastructural analysis of Turritopsis
nutricula during life cycle reversal. Tissue Cell, 35: 213–222.
Collins, A.G. 2002. Phylogeny of Medusozoa and the evolution of
cnidarian life cycles. J. Evol. Biol. 15: 418–432.
Cutress, C.E. 1979. Bunodeopsis medusoides Fowler and Actino
-
discus neglectus Fowler, two Tahitian sea anemones: redescrip
-
tion and biological notes. Bull. Mar. Sci. 29: 96–109.
Diehl, F.A., and Bouillon, J. 1966. Observations sur les potenti
-
alités morphogénétiques des tissus irradiés de Cordylophora
caspia (Pallas). Bull. Cl. Sci. Acad. R. Belg. 52: 138–146.
© 2004 NRC Canada
Piraino et al. 1753

Fautin, D.G. 2002. Reproduction in Cnidaria. Can. J. Zool. 80:
1735–1754.
Forrest, H. 1963. Immortality and Pearl’s survivorship curve for
Hydra. Ecology, 44: 609–610.
Frey, J. 1968. Die Entwicklungsleistungen der Medusenknospen
und Medusen von Podocoryne carnea M. Sars nach Isolation
und Dissoziation. Wilhelm Roux’ Arch. Entwicklungsmech.
Org. 160: 428–464.
Gohar, H.A.F. 1940. Studies on the Xeniidae of the Red Sea: their
ecology, physiology, taxonomy and phylogeny. Publ. Mar. Biol.
Stn. Al Ghardaqa, 3: 27–78.
Gravier, N. 1970. Libération des médusoides par Macrorhynchia
philippina Kirchenpauer, 1872 (Hydrozoa, Plumulariidae). Recl.
Trav. Stn. Mar. Endoume Fac. Sci. Mars. 10: 253–257.
Gravier-Bonnet, N. 1992. Cloning and dispersal by buoyant auto
-
tomised hydranths of a thecate hydroid (Cnidaria; Hydrozoa).
Sci. Mar. 56: 229–236.
Gravier-Bonnet, N., and Migotto, A.E. 2000. Gonangium develop
-
ment and medusoid of Nemalecium lighti (Hargitt, 1924)
(Cnidaria: Hydrozoa, Haleciidae). Sci. Mar. 64: 207–213.
Hadzi, J. 1909a. Einige Kapitel aus der Entwicklungsgeschichte
von Chrysaora. Arb. Zool. Inst. Univ. Wien, 17: 17–44.
Hadzi, J. 1909b. Rückgängig gemachte Entwicklung einer Scypho
-
meduse. Zool. Anz. 34.
Hadzi, J. 1912. Die Reduktion des Scyphopolypen und der Ephyra
von Chrysaora. In Proceedings of the VIII International Zoolog-
ical Congress, Graz, Austria, 1910. G. Fisher Verlag, Jena.
pp. 578–594.
Hauenschild, C. 1956. Experimentelle Untersuchungen über die
Entstehung asexueller Klone bei der Hydromeduse Eleutheria
dichotoma. Z. Naturforsch. 11: 394–402.
Hérouard, E. 1909. Sur les cycles évolutifs d’un Scyphistome. C.
R. Hebd. Seanc. Acad. Sci. (Paris), 148: 320–323.
Jarms, G. 1997. The polyps of Coronatae (Scyphozoa):a review
and some new results. In Coelenterate Biology: Proceedings
of the 6th International Congress of Coelenterate Biology,
Nordwijkerhout, the Netherlands, 16–21 July 1995. Edited by
J.C. den Hartog, L.P. van Ofwegen, and S. van der Spoel.
Nationaal Natuurhistorisch Museum, Leiden, the Netherlands.
pp. 271–278.
Kakinuma, Y. 1969. On the differentiation of the isolated medusa
bud of the hydrozoans, Cladonema uchidai and Cladonema sp.
Bull. Mar. Biol. Stn. Asamushi, 13: 169–172.
Lesh-Laurie, G.E., and Corriel, R. 1973. Scyphistoma regeneration
from isolated tentacles in Aurelia aurita. J. Mar. Biol. Assoc.
U.K. 53: 885–894.
Marcum, B.A., and Campbell, R.D. 1978. Development of Hydra
lacking nerve and interstitial cells. J. Cell Sci. 29
: 17–33.
Marques, A.C., and Collins, A.G. 2004. Cladistic analysis of
Medusozoa and cnidarian evolution. Invertebr. Biol. 123: 23–42.
Müller, H.C. 1913. Die Regeneration der Gonophore bei den
Hydroiden und anschliessende biologische Beobachtungen I.
Athecata. Arch. Entwicklungsmech. Org. 37: 319–419.
Müller, W.A. 1996. Pattern formation in the immortal Hydra.
Trends Genet. 12: 91–96.
Okada, T.S. 1991. Transdifferentiation. Clarendon Press, Oxford.
Paspaleff, G.V. 1938. Ueber die Entwicklung von Rhizostoma
pulmo Agassiz Arb. Biol. Meeres Stn. Varna, 7: 1–17.
Pearse, V.B. 2002. Prodigies of propagation: the many modes of
clonal replication in boloceroidid sea anemones (Cnidaria,
Anthozoa, Actiniaria). Invertebr. Reprod. Dev. 41: 201–213.
Piraino, S., Boero, F., Aeschbach, B., and Schmid, V. 1996. Re
-
versing the life cycle: medusae trasforming into polyps and
cell transdifferentiation in Turritopsis nutricula (Cnidaria,
Hydrozoa). Biol. Bull. (Woods Hole), 190: 302–312.
Richmond, R.H. 1985. Reversible metamorphosis in coral planula
larvae. Mar. Ecol. Prog. Ser. 22: 181–185.
Sammarco, P.W. 1982. Polyp bail-out: an escape response to envi
-
ronmental stress and a new means of reproduction in corals.
Mar. Ecol. Prog. Ser. 10: 57–65.
Schmid, V. 1972. Untersuchungen über Dedifferenzierungsvorgänge
bei Medusenknospen und Medusen von Podocoryne carnea M.
Sars. Wilhelm Roux’ Arch. Entwicklungsmech. Org. 169: 281–
307.
Schmid, V. 1974. Regeneration in medusa buds and medusae of
Hydrozoa. Am. Zool. 4: 773–781.
Schmid, V. 1992. Transdifferentiation in medusae. Int. Rev. Cytol.
142: 213–261.
Schmid, V., and Alder, H. 1984. Isolated, mononucleated, striated
muscle can undergo pluripotent transdifferentiation and form a
complex regenerate. Cell, 38: 801–810.
Schmid, V., Wydler M., and Alder, H. 1982. Transdifferentiation
and regeneration in vitro. Dev. Biol. 92: 476–488.
Spangenberg, D.B. 1965. A study of strobilation in Aurelia aurita
under controlled conditions. J. Exp. Zool. 160: 1–10.
Spangenberg, D.B. 1968. Recent studies of strobilation in jellyfish.
Oceanogr. Mar. Biol. Ann. Rev. 6: 231–247.
Spring, J., Yanze, N., Middel, A.M., Stierwald, M., Gröger, H., and
Schmid, V. 2000. The mesoderm specification factor twist in the
life cycle of jellyfish. Dev. Biol. 228: 363–375.
Stearns, S.C. 1992. The evolution of life histories. Oxford Univer
-
sity Press, New York.
Tardent, P. 1963. Regeneration in the Hydrozoa. Biol. Rev. Camb.
Philos. Soc. 38: 293–333.
Thiel, H. 1963. Untersuchungen über die Entstehung abnormer
Scyphistomae, Strobilae und Ephyrae von Aurelia aurita Lam.
und ihre theoretische Bedeutung. Zool. Jahrb. Abt. Anat. Ontog.
Tiere, 81: 311–358.
Werner, B. 1963. Effect of some environmental factors on differen
-
tiation and determination in marine Hydrozoa, with a note on
their evolutionary significance. Ann. N.Y. Acad. Sci. 105: 461–
488.
Werner, B. 1971. Stephanoscyphus planulophorus n.spec., ein neuer
Scyphopolyp mit einem neuen Entwicklungsmodus. Helgol.
Wiss. Meeresunters. 22: 120–140.
Yanze, N., Spring, J., Schmidli, C., and Schmid, V. 2001. Conser
-
vation of Hox/ParaHox-related genes in the early development
of a cnidarian. Dev. Biol. 236: 89–98.
© 2004 NRC Canada
1754 Can. J. Zool. Vol. 82, 2004