Download What a couple of dimensions can do for you

151
What a couple of dimensions can do for you: Comparative
developmental studies using 4D microscopy—examples from
tardigrade development
Andreas Hejnol1,2 and Ralf Schnabel
Technische Universität Braunschweig, Spielmannstrasse 7, 38106 Braunschweig, Germany
Synopsis The development of an organism consists of processes occurring in space and time. To analyze this 4-dimensional
development in embryogenesis, an appropriate method should be chosen. We present here a sophisticated method, 4D
microscopy (3D time-lapse microscopy), initially developed to analyze the cell lineage of wild-type and mutant embryos
of the nematode Caenorhabditis elegans. Our method records the entire development of an embryo and allows detailed
analyses of events such as cleavage, cell migration, cell death (apoptosis), and cell differentiation during development. The 4D
microscopy system has 3 main parts: a motorized microscope, trigger software, and a database that facilitates the analysis of
recordings. Adopting the 4D microscopy technique for uses beyond the analysis of C. elegans makes it possible to discern
the cell lineage of other small embryos. Our method fills a gap in the study of the development of diverse organisms that are
impossible to observe with fluorescent labeling techniques using single blastomeres. The use of this technique to investigate
the development of organisms such as tardigrades, acoelomorphs, rotifers, and gastrotrichs provides fresh insight into the
evolution of developmental processes and the phylogenetic relationships between such taxa. Using tardigrade development as
an example, we demonstrate that the use of 4D microscopy can reveal new characters and corroborate or disapprove
old characters. We discuss the results in the light of recent phylogenetic hypotheses regarding the Arthropoda and their
probable sister group, the Cycloneuralia, which together form the Ecdysozoa.
Introduction
Resolution of metazoan phylogenetic relationships is
one of the most long standing and fascinating goals
of biology, in which modern techniques from the
fields of molecular biology, bioinformatics, paleontology, developmental biology, and morphology are
integrated. Consistent technological advances, driven
primarily by medical applications in model organisms,
can also be fruitful when applied to the study of
other organisms. Here we present a new form of
microscopy, termed 4D microscopy (3D time-lapse
microscopy), that was originally developed to investigate the molecular basis of the cell lineage of the
nematode Caenorhabditis elegans (Schnabel and others
1997). Progress in analog and computational imaging
technologies has made it possible to adapt this method
to investigate other embryos. Questions raised by the
publication of alternative phylogenetic hypotheses
concerning the protostomian clades guided our choice
of key organisms whose embryos are suitable for
analysis using the 4D microscope system. We present
here the phylogenetic implications of our study of the
development of the eutardigrade Thulinia stephaniae
(Hejnol and Schnabel 2005) in the light of recently
published phylogenetic hypotheses. Not only do we
either corroborate or reject prior assumptions concerning tardigrade development, but we also identify
new characters (for example, cleavage pattern)
that could be used for morphological data matrices
in phylogenetic analysis. Our results contribute to
the reconstruction of the arthropod and the ecdysozoan ground pattern.
The 4D microscopy system
The 4D microscopy system, first introduced as a
multiple focal plane time-lapse recording system by
Minden and colleagues (1989) and Hird and White
(1993), was designed to analyze the development of
living early embryos. The use of high-resolution
enhanced Nomarski optics facilitates the detection
of single cells of the embryo at different focus levels
(Thomas and others 1996; Schnabel and others 1997).
From the symposium “The New Microscopy: Toward a Phylogenetic Synthesis” presented at the annual meeting of the Society for Integrative and
Comparative Biology, January 4–8, 2005, at San Diego, California.
1 E-mail: [email protected]
2 Present address: Kewalo Marine Laboratory, University of Hawaii at Manoa, 41 Ahui Street, Honolulu, HI 96813, USA.
Integrative and Comparative Biology, volume 46, number 2, pp. 151–161
doi:10.1093/icb/icj012
Advance Access publication February 16, 2006
Ó The Society for Integrative and Comparative Biology 2006. All rights reserved. For permissions, please email: journals.permissions@
oxfordjournals.org.
152
A. Hejnol and R. Schnabel
Fig. 1 The 4D microscopy system. (A) Motorized microscope used for this study. The microscope is connected to a
Hamamatsu analog camera. The pictures are digitized using a frame grabber in the connected computer. The software
on the computer triggers the motor of the microscope and the collection of the single pictures. (B) Screenshot of the
trigger software. In the right window the photograph collected by the camera is shown. All major modification buttons
are visible on the screen and can be easily used during the recording. (C) Screenshot of the SIMI BioCell software for
the analysis of the recordings. The left-hand window shows the cell lineage of the embryo; the central window shows
the 3D reconstruction of the embryo. The right-hand window shows the picture of the corresponding level. Other
small windows facilitate navigation of the lineage tree and show information about single cells.
To handle the amount of data produced by a detailed
recording of a complete embryogenesis and to facilitate
the tracing of cell lineage, the software SIMI BioCell
was developed (Schnabel and others 1997).
Modifications to the system have allowed it to be successfully applied to other embryos, as demonstrated by
investigations of Drosophila melanogaster (Urbach and
others 2003), an isopod crustacean (Hejnol 2002;
Dohle and others 2004), other nematode embryos
(Dolinski and others 1998; Houthoofd and others
2003) and the tardigrade embryo presented here
(Hejnol and Schnabel 2005).
The system has 3 parts: a motorized microscope,
control software for recording, and analysis software
(Fig. 1). The microscope is a modified Zeiss compound microscope with a motorized light shutter
and focus (Fig. 1A). Photographs are taken using a
high-resolution analog camera, digitized with a
frame grabber, and automatically saved onto the
computer hard drive in a lossless wavelet compressed
format. To ensure constant temperature during recording, the microscope is equipped with a thermostat set
to a range of 10–37 C. The microscope software was
programmed in Cþþ by A.-K. Schulz and R.S. and
controls the whole capturing process from opening the
light shutter, to capturing individual pictures step by
step through the z-levels, to saving the z-stack onto the
hard disk, to closing the light shutter (Fig. 1B). This
process is repeated at defined time intervals during the
entire development of the embryo. Parameters can be
changed during the recording to correct for possible
sinking of the embryo within the z-stack, to increase
153
4D microscopy of tardigrade development
the number of optical sections when cells become
smaller, and to increase time intervals when cell cycles
are prolonged as development proceeds. The hardware
and software are able to capture z-stacks of up to 70
levels every 40 s to ensure that the development of fast
developing embryos with asynchronous cleavage
patterns can be recorded. The tardigrade embryo
was recorded with 45 optical sections every 45 s at a
temperature of 24 C. To handle the large number of
recorded digital pictures (for example, 1 recording of
the tardigrade embryo consists of 140,000 individual
images) we used the database SIMI BioCell, software
specifically programmed for this type of cell lineage
analysis (Fig. 1C).
SIMI BioCell allows easy visualization of single
pictures using the computer keyboard (up/down cursors to move through the z-stack; left/right cursors to
move forward and backward in time). Cell nuclei are
marked by clicking the mouse at appropriate time
intervals, which saves their coordinates and the time
to a database. By following and marking cells before
and after cell divisions, a cell lineage tree is built
in which each cell can be followed to its final cell
fate. The software is able to represent the data as a
cell lineage tree or as a 3D representation in which
the position of every cell (nucleus) is shown as a
sphere (Fig. 1C). The spheres can be colored to demonstrate cell clones and their localization. SIMI BioCell
also includes several functions that facilitate the
tracing of cells in the embryo. Because the resolution
of Nomarski optics improves at high magnifications,
4D microscopy is ideal for recording small embryos
(<60 mm) with hard eggshells, which are otherwise
inaccessible via injection of blastomeres. A future
goal is to combine 4D microscopy with fluorescent
gene expression reporters and to visualize gene expression patterns over time at a single cell resolution. A 4D
microscope system with 3 channels has been developed
for this purpose. To exemplify the power of the system
for comparative studies we discuss our results on
tardigrade development in the light of the currently
conflicting phylogenetic hypotheses.
Tardigrade development: bringing light into a
obscure field
Tardigrades—also known as water bears—are tiny protostomes that live in marine and freshwater environments. Historically, their development has received
little attention because the overwhelming interest
has been in the developmental mechanisms in model
organisms. With the rebirth, as “evo-devo,” of the
field of comparative developmental biology and
major rearrangements in the protostome phylogeny,
some of the minor phyla are attracting renewed attention. One of these neglected taxa is the Tardigrada,
which have gained attention as a result of their intermediate position between Euarthropoda and groups
of the former Nemathelminthes in the Ecdysozoa
hypothesis (Gould 1995; Aguinaldo and others 1997;
Kinchin 2000; Goldstein and Blaxter 2002). Hence, a
reinvestigation of the development of tardigrades was
advisable given the large gaps in our knowledge about
their development.
After an initial investigation by Kaufmann (1851),
the majority of studies on tardigrade development
were carried out in the early 20th century (von
Erlanger 1895; von Wenck 1914; Marcus 1928,
1929). These studies describe cleavage, gastrulation,
and germ layer formation of eutardigrade species
and form the basis for the descriptions of tardigrade
development in recent textbooks and of characteristics in current morphological data matrices. More
recently, Eibye-Jacobsen (1997) added descriptions
of the development of heterotardigrade embryos
based on transmission electron microscopy (TEM)
studies. These descriptions often conflict with the
previous studies regarding the cleavage pattern
and gastrulation, raising certain questions: Do tardigrade embryos show traits of spiral cleavage? Do
they show a stereotypical cleavage pattern? When are
the fates of the cells determined? Is the mesoderm
really formed as outpouchings from the gut
(enterocoely)?
Eibye-Jacobsen (1997) cast doubt on the descriptions by Marcus (1928, 1929) and von Erlanger
(1895) of the mesoderm being formed by enterocoely, and a careful comparison of the 2 publications
by Marcus further confuses the discussion. Marcus
(1928) stated that he could not detect cell borders
after gastrulation in his histological slices although he
included them in his drawings. One year later this
statement is missing (Marcus 1929). R. M. Kristensen
looked again at Marcus’ original sections, which are
archived in the Smithsonian Institute, and confirmed that it is not possible to detect cell borders
and furthermore that the sections are too thick
(5–8 mm) to reconstruct enterocoely (personal communication from R. M. Kristensen; unreferenced). A
reinvestigation of tardigrade development using
modern techniques is necessary to resolve these
concerns.
The phylogenetic position of the tardigrades
Tardigrades are generally associated with the
Euarthropoda (chelicerates, myriapods, crustaceans,
and insects) based on morphological data and
154
A. Hejnol and R. Schnabel
Fig. 2 Embryonic cleavage patterns mapped onto alternative phylogenetic trees. (A) Evolution of cleavage patterns based
on the Articulata hypothesis. Molluscs and annelids show spiral cleavage. Consequently, in the Arthropoda the spiral
cleavage is plesiomorphic and was modified to a radial or syncytial cleavage pattern. (B) The alternative Ecdysozoa
hypothesis, which postulates a sister relationship of Arthropoda and Cycloneuralia, predicts a radial, indeterminate
cleavage in the ground pattern of this clade, since the Nematomorpha and Priapulida show a radial irregular cleavage
pattern. Nematode development shows variable cleavage patterns. However, an outgroup comparison with the
development of nematomorphs suggests that the stereotypical cell lineage, for example of C. elegans, is derived in the
nematodes and that an indeterminate cleavage is thus part of the ground pattern of nematodes. This suggests that an
indeterminate cleavage pattern is plesiomorphic for the Arthropoda and that the stereotypical cell lineages observed in
some arthropods are a derived characteristic in the Arthropoda as in nematodes. Therefore, we propose that an
indeterminate cleavage is part of the ground pattern in the Ecdysozoa.
molecular data (Garey and others 1996, 1999; Giribet
and others 1996; Garey 2001; Peterson and Eernisse
2001; Mallatt and Winchell 2002; Mallatt and others
2004). A 2005 molecular phylogenetic analysis
(Philippe and others 2005) placed the tardigrades as
a sister group to the nematodes. However, as there is no
significant data to the contrary, we assume the position
of the tardigrades as sister to arthropods (Fig. 2B).
Thus, tardigrades share a common ancestor with the
euarthropods, and we will discuss our results regarding
the ground pattern of the Arthropoda. To specify the
plesiomorphic characteristics, it is necessary to include
the sister group of the arthropods in the discussion. As
new phylogenies based on molecular data favor the
Cycloneuralia (Priapulida, Loricifera, Kinorhyncha,
Nematomorpha, and Nematoda) as the sister group
(Aguinaldo and others 1997; Giribet and others
2000; Garey 2001; Giribet 2003; Philippe and others
2005), rather than the Annelida (Fig. 2), we will discuss
some characteristics as possible plesiomorphies for
the Arthropoda as well as the Ecdysozoa. [The morphological data matrices of these analyses that support
the Ecdysozoa (Zrzavy and others 1998; Peterson
and Eernisse 2001; Zrzavy 2001, 2003; Eernisse and
Peterson 2004) have been criticized by Jenner and
Scholtz (2005).]
Results and discussion
Axis determination and cleavage pattern
We investigated the cell lineage of the eutardigrade T.
stephaniae using the 4D microscopy system and a subset
of developmental stages of the heterotardigrade
Echiniscoides sigismundii. Cleavage is total and equal
in both embryos. The cleavage program of T. stephaniae
is irregular, and early blastomeres show an unexpected
potential for regulation when early cells are ablated
(Hejnol and Schnabel 2005). It was not possible to identify corresponding blastomeres between embryos owing
to their equal size, the lack of morphological characteristics, and the variable spindle directions (Fig. 3D–F).
The timing of cell divisions also differed among individual embryos (Fig. 3A–C). The embryo appears to lack
a predetermined egg axis, and the localization of the
polar body does not correspond to any of the future
dorsal/ventral or anterior/posterior axes (Hejnol and
Schnabel 2005). The observed absence of predetermined
cytoplasm is supported by cell ablation experiments in
which the embryo was able to form all main axes after 1
cell of the 2-cell and 2 cells of the 4-cell stage were ablated
(Hejnol and Schnabel 2005).
Related euarthropods show a gradient of cleavage
types ranging from total to syncytial cleavage, and
4D microscopy of tardigrade development
155
Fig. 3 Cleavage and gastrulation in the eutardigrade T. stephaniae. (A–C) Cell lineages and cell pattern of 3 different
embryos of the tardigrade T. stephaniae (126 cell stage). The lineage trees show different cell division timings in all
3 embryos from early on. (D–F) Arrangements of the descendants of 2 corresponding blastomeres (determined by
the future egg axes) differ in each embryo (dark blue versus pink). The 2 bright yellow cells show the 2 primordial
germ cells. (G) Gastrulation of a tardigrade embryo: ventral view, anterior is left. The blastopore is located at the future
position of the mouth. (H) Arrangement of the germ layer precursors before cell migration at the site of gastrulation
shown with the 3D reconstruction of SIMI BioCell. The germ cells are the first to migrate into the embryo, followed by
the endodermal and mesodermal precursors.
including intermediate forms, which begin with total
cleavage and switch to syncytial cleavage (Anderson
1973; Scholtz 1997). Both cleavage type extremes are
found scattered in the arthropod lineages, and related
taxa show opposite cleavage patterns (for example,
isopods with syncytial and amphipods with holoblastic
cleavage).
Compared with the euarthropods, the tardigrade
cleavage program is most similar to the holoblastic
and irregular cleaving pycnogonids (Dogiel 1913),
some crustaceans (Fuchs 1914; Benesch 1969), and
myriapods (Tiegs 1940, 1947; Dohle 1964). Syncytial
cleavage appears to have evolved independently
in different arthropod lineages, perhaps as an
156
adaptation to a terrestrial life cycle. As an outgroup
comparison would help to determine the cleavage type
of the stem species of the Arthropoda, tardigrades may
reflect the ancestral type of cleavage in the arthropod
clade. In light of the Ecdysozoa hypothesis, irregular
and indeterminate tardigrade cleavage would be a
plesiomorphic condition (Fig. 2B). The clades that
are regarded as the new adelphotaxon of the
Arthropoda, either the whole Cycloneuralia or some
members, show holoblastic cleavage. In addition, their
cleavage pattern appears to be irregular based on
the study of certain nematodes (Malakhov and
Spiridonov 1984; Voronov and Panchin 1998;
Voronov 1999; Schierenberg 2005) and priapulids
(Zhinkin 1949; Zhinkin and Korsakova 1953). As
annelids are the sister group to Arthropoda according to the Articulata hypothesis (Fig. 2A), the cleavage
type in the arthropods should be derived from a
spiral cleaving ancestor (Scholtz 1997; Nielsen 2001).
No spiral cleaving arthropod embryo has been properly
described, and Anderson’s wideranging description
(1969) of the cirripede cleavage as spiral is based on
dubious homologization of single cells and badly needs
reinvestigation using modern methods (Zilch 1979).
Under the Ecdysozoa hypothesis, holoblastic irregular
cleavage appears to be the plesiomorphic condition
for arthropods as well as for all of the Ecdysozoa
(Fig. 2B). Which molecular mechanisms regulate cell
fate and axis determination in the tardigrade embryo
remains enigmatic, as no known mechanism from protostome model organisms can explain this ancestral
type of development.
Gastrulation
Gastrulation in T. stephaniae starts with the immigration of cells at the position of the future oral opening
(Hejnol and Schnabel 2005). The embryo does not
form a clear blastocoel. Simultaneously, some cells
migrate through the blastopore and other cells sink
temporarily into the embryo at a nearby area that
later in development gives rise to the proctodaeum
of the adult (Fig. 3G). In T. stephaniae, the first
evidence of the anterior/posterior and dorsal/ventral
axes of the embryo is recognizable at gastrulation
with the blastopore being anterior and ventral
(Hejnol and Schnabel 2005). In all embryos recorded,
the germ cells were the first to migrate, followed by
the endoderm precursors and mesoderm. During
migration these cells are indistinguishable, with only
future cell fate in terms of their position and differentiation allowing us to reconstruct their identity
during this stage. The 2 germ cells migrate into the
gonad and the mesodermal cells form the somites
A. Hejnol and R. Schnabel
and later the muscles of the limbs. The arrangement
of the cells prior to migration is shown in Figure 3H, in
which 2 germ cells are surrounded by the endoderm
and mesoderm precursors. We were unable to
determine whether there is a stereotypical pattern in
the arrangement of the mesodermal and endodermal
precursors owing to the low number of embryos in
which the cell lineage was traced (n ¼ 3). Which
mechanism induces the fate of the germ cells or the
fate of the germ layer precursors remains unknown,
but the arrangement of these cells at the blastopore is
reminiscent of some total cleaving crustaceans (Kühn
1913; Fuchs 1914; Benesch 1969; Gerberding and
others 2002; Gerberding and Patel 2004) and thus
suggests the presence of a common mechanism.
Owing to the variation in gastrulation across the arthropods (Weygoldt 1979) it is difficult to determine
whether gastrulation in tardigrades reflects an ancestral
type for the arthropods. A comparison with the putative sister group of arthropods, the Cycloneuralia, is
rather uninformative because of the lack of thorough
investigations of the gastrulation of loriciferans,
kinorhynchs, and priapulids. The descriptions of gastrulation in nematodes demonstrate the variety present
in this taxon (Sulston and others 1983; Voronov and
Panchin 1998; Schierenberg 2005) and the description of a nematomorph gastrulation (Inoue 1958) is
completely incongruent with other cycloneuralians, in
that the mouth is formed at a site other than the
blastopore.
Mesoderm development
The development of the third germ layer of tardigrades has been controversial from the beginning of
tardigrade investigation. Von Erlanger (1895) stated
that the mesoderm is formed by outpouching of the
gut. His illustrations clearly show coelomic hollows
and a gut lumen in the developing embryo. We did
not observe these lumens in living embryos of either
the heterotardigrade E. sigismundii (Fig. 4A) or the
eutardigrade T. stephaniae (Fig. 4B and C).
Enterocoelic formation was corroborated by Marcus
(1928, 1929). More recently, Eibye-Jacobson (1997)
expressed doubts concerning the formation of the
mesoderm by enterocoely in trardigrades.
Tracing back the cell lineage of single cells of
the mesodermal somites of T. stephaniae shows that
their origin is not the endodermal gut; rather they
develop from mesodermal precursors that migrate
through the blastopore into the embryo (Hejnol and
Schnabel 2005). These mesoderm precursor cells form
mesodermal cords ventrolateral to the gut (Fig. 4B).
The precursor cells do not form a clear growth zone nor
4D microscopy of tardigrade development
157
Fig. 4 Later stages of tardigrade embryos showing the mesodermal somites. The schematic drawings on the right show the
orientation of the embryo in the egg. (A) Embryos of the heterotardigrade E. sigismundii. The black arrows show the
epithelial border surrounding somites. (B) Embryo of T. stephaniae before elongation. The spheres indicate cells of
different germ layers. One part of the pharynx (ph) consists of endodermal cells. The mesodermal cells form lateral
mesodermal bands. The inner cells are surrounded by a 1-cell layered ectodermal epithelium. (C) Embryo of the eutardigrade T. stephaniae after elongation: view from the left side of the embryo. Four somites are visible, indicated by
roman numbers; each is separated by an epithelium from the other cells. The embryo forms a ventral fold (vf).
teloblastic cells (Hejnol and Schnabel 2005). Furthermore, single cells are distributed along the egg axis
throughout migration along the inner surface of the
embryo to their final destinations. During this movement, the cells proliferate and extend the mesodermal
layer. The mesodermal germ layer forms lateral bands
on either side of the entodermal gut and pharynx
(Fig. 4B). These bands separate into groups of cells
after 22 h of development. Each of these somites
appears to have an epithelium that separates them
from other cells (Fig. 4). We could not detect a hollow
inside these somites during the development because
of the naturally compressed embryo inside the eggshell
(Fig. 4). In her work, Eibye-Jacobsen (1997) could
detect hollows, which become visible as a result of
shrinking during the preparation technique used for
TEM. These results combined suggest that the coelom
is formed by schizocoely in tardigrades. This reflects a
158
A. Hejnol and R. Schnabel
Table 1 Morphological characters changed in the data matrix of Eernisse and Peterson (2004)
No.
Character
Modification
17
Position of polar bodies
Added characteristic state 4 (irregular position)
24
One axis specified during oogenesis
“?” to “0” (absent)
28
Stereotypical cleavage pattern
“?” to “0”
29
Spiral cleavage
“?” to “0”
32
Gastrulation
“?” to “1” (present)
33
Blastopore associated with larval/adult mouth
“?” to “1”
35
Endomesodermal muscle cells
“?” to “1”
36
Endomesoderm derived from gut
“?” to “0”
37
Ectomesenchyme
“?” to “0”
38
4d endomesoderm
“?” to “0”
39
Mesodermal bands from 4d
“?” to “0”
40
Lateral coelom derived from mesodermal bands
“0” to “1” (also corrected for
arthropods and onychophorans)
42
Teloblastic growth
“?” to “0”
43
Somatoblast
“?” to “0”
Tetrahedral 4-cell stage
“?” to “1”
157
well-known phenomenom in arthropods, including
the onychophorans (Anderson 1973; Dohle 1979).
Later in development, these somites disappear and
some of the cells migrate into the limb buds to form
the muscles. The other cells could not be traced further,
but they may form the “mixocoel,” which is common
to all euarthropod taxa and the onychophorans
(Weygoldt 1986; Nielsen 2001; Scholtz 2002; Mayer
and others 2004). The establishment of mesodermal
bands that separate into segmental, paired somites
to later form coelomic cavities by schizocoely is a
mechanism shared between annelids and arthropods
(Weygoldt 1963; Anderson 1973; Dohle 1979) and
has been used to unite annelids and arthropods in
defense of the Articulata hypothesis (Scholtz 2002,
2003). In nematodes and nematomorphs, mesodermal
bands are also present (Inoue 1958; Schierenberg
2005), but they do not form segmental somites.
Segmentation and posterior growth zone
Tardigrades have 4 segmental leg pairs and a brain
consisting of 3 parts (Dewel and Dewel 1996), although
we could not corroborate the presence of a tripartite
brain in our study. Our study of the development of
the eutardigrade T. stephaniae shows no sequential
development of the 4 trunk segments, including the
somites and limb buds (Fig. 4A and C). Teloblasts
in the ectoderm or mesoderm are not present in the
embryo and a posterior growth zone is also absent
(Hejnol and Schnabel 2005). A common mechanism
to elongate the body during development in annelids
and arthropods is the sequential addition of segments from a posterior growth zone. The absence of
a growth zone in tardigrades may be interpreted as a
derived state resulting from the small size of the
embryo and the low number of segments. Some
authors regard a teloblastic growth zone as an
apomorphy supporting a close relationship between
the Annelida and Arthropoda (Anderson 1973; Ax
1999; Nielsen 2003). However, recent evidence rejects
teloblasts in the ground pattern of the Arthropoda
(Dohle and Scholtz 1988; Scholtz 2002). The description of the presence of ectoteloblasts in annelids
(Anderson 1973) is only true for the clitellates
(Dohle 1999). Our cell lineage studies for the polychaete Trilobodrilus axi show that the cells that form
the ectoteloblasts in clitellates do not form ectoteloblasts in polychaetes and the “mesoteloblast” derived
from the 4d cell does not divide in a teloblast fashion
(unpublished data, A.H.). If the Ecdysozoa hypothesis
is correct and no member of the Cycloneuralia displays
an embryological posterior growth zone, then it
has evolved in the arthropods convergently. Further
investigations of priapulid, kinorhynch, and loriciferan
development are needed before an understanding of the
evolution of segmentation in the protostomes will be
complete.
Phylogenetic analysis
Our investigations of the development of the eutardigrade T. stephaniae yielded new insights and useful
characters for the reconstruction of phylogenetic
4D microscopy of tardigrade development
trees. To determine the extent to which the new
findings corroborate or weaken the position of the
Tardigrada, we modified the morphological data
matrix of Eernisse and Peterson (2004) and recalculated the Bootstrap and Bremer support with Paup*
(Swofford 2002) of a parsimony analysis. Of 166
morphological characters, 15 were changed in the matrix according to Table 1, and both trees were recalculated with 2000 replicants. The topology of the
relationship of Tardigrada to Onychophora and
Arthropoda was not changed, but the Bootstrap values
and Bremer indices differed. The node separating
Tardigrada from Onychophora/Arthropoda is supported in both trees with a Bremer index of 4, and the
Bootstrap value varies marginally between the original
tree (87) and the modified one (86). Interestingly, the
sister group relationship between onychophorans and
arthropods receives a Bootstrap value of 83 and Bremer
index of 3 in the original tree and is less supported after
introducing our modifications (Bootstrap 69, Bremer
index 2). Although support for a sister group relationship of the tardigrades to the Onychophora/
Arthropoda remains relatively high in both trees, the
introduction of our findings to the data matrix of
Eernisse and Peterson (2004) could not help to clarify
the phylogenetic relationships between these taxa.
Acknowledgments
We thank Reinhardt M. Kristensen for species determination and his help and support in collecting
E. sigismundii. The field collection trip for E. sigismundii to Denmark for A.H. was financed by the COBICE
program of the EU. We thank Amy Maxmen, Kevin
Pang, and 3 anonymous referees for critically reading
and improving the manuscript and Kevin Peterson for
sending us the nexus file.
References
Aguinaldo AM, Turbeville JM, Linford LS, Rivera MC, Garey JR,
Raff RA, Lake JA. 1997. Evidence for a clade of nematodes,
arthropods and other moulting animals. Nature 387:489–93.
Anderson DT. 1969. On the embryology of the cirripede
crustaceans Tetraclita rosea (Krauss), Tetraclita purpurascens
(Wood), Chthalamus antennatus (Darwin) and Chamaesipho
columna (Spengler) and some considerations of crustacean
phylogenetic relationships. Philos Trans R Soc B 256:183–235.
Anderson DT. 1973. Embryology and phylogeny in annelids
and arthropods. Oxford: Pergamon Press.
Ax P. 1999. Das System der Metazoa II. Stuttgart: Gustav
Fischer.
Benesch R. 1969. Zur Ontogenie und Morphologie von
Artemia salina L. Zool Jb Anat 86:307–458.
Dewel RA, Dewel WC. 1996. The brain of Echiniscus viridissimus
Peterfi 1956 (Heterotardigrada): a key to understanding the
159
phylogenetic position of tardigrades and the evolution of
the arthropod head. Zool J Linn Soc 116:35–49.
Dogiel V. 1913. Embryologische Studien an Pantopoden. Z Wiss
Zool 107:575–741, plates 17–22.
Dohle W. 1964. Die Embryonalentwicklung von Glomeris
marginata (Villers) im Vergleich zur Entwicklung anderer
Diplopoden. Zool Jb Anat 81:241–310.
Dohle W. 1979. Vergleichende Entwicklungsgeschichte des
Mesoderms bei Articulaten. Fortschr Zool Syst
Evolutionsforsch 1:120–40.
Dohle W. 1999. The ancestral cleavage pattern of the clitellates
and its phylogentic deviations. Hydrobiologia 402:267–83.
Dohle W, Gerberding M, Hejnol A, Scholtz G. 2004. Cell
lineage, segment differentiation and gene expression in crustaceans. In: Scholtz G, editor. Evolutionary developmental
biology of Crustacea. Lisse: A.A. Balkema. p 95–133.
Dohle W, Scholtz G. 1988. Clonal analysis of the crustacean
segment: the discordance between genealogical and segmental
borders. Development 104 (Suppl):147–60.
Dolinski C, Borgonie G, Schnabel R, Baldwin JG. 1998. Buccal
capsule development as a consideration for phylogenetic
analysis of Rhabditida (Nemata). Dev Genes Evol
208:495–503.
Eernisse D, Peterson K. 2004. The history of animals. In:
Cracraft J, Donoghue M, editors. Assembling the tree of
life. New York: Oxford University Press. p 197–208.
Eibye-Jacobsen J. 1997. New observations on the embryology of
the Tardigrada. Zool Anz 235:201–16.
Fuchs F. 1914. Die Keimblätterentwicklung von Cyclops viridis
Jurine. Zool Jb Anat 38:103–56.
Garey JR. 2001. Ecdysozoa: the relationship between
Cyloneuralia and Panarthropoda. Zool Anz 240:321–30.
Garey JR, Krotec M, Nelson DR, Brooks J. 1996. Molecular
analysis supports a tardigrade-arthropod association.
Invertebr Biol 115:115–21.
Garey JR, Nelson DR, Mackey LY, Li J. 1999. Tardigrade phylogeny: congruency of morphological and molecular evidence.
Zool Anz 238:205–10.
Gerberding M, Browne WE, Patel NH. 2002. Cell lineage analysis
of the amphipod crustacean Parhyale hawaiensis reveals an
early restriction of cell fates. Development 129:5789–801.
Gerberding M, Patel NH. 2004. Gastrulation in crustaceans. In:
Stern C, editor. Gastrulation. Cold Spring Harbor, NY: CSHL
Press. p 79–89.
Giribet G. 2003. Molecules, development and fossils in the study
of metazoan evolution; Articulata versus Ecdysozoa revisited.
Zoology 106:303–26.
Giribet G, Carranza S, Baguna J, Riutort M, Ribera C. 1996. First
molecular evidence for the existence of a Tardigrada
Arthropoda clade. Mol Biol Evol 13:76–84.
Giribet G, Distel DL, Polz M, Sterrer W, Wheeler WC. 2000.
Triploblastic relationships with emphasis on the acoelomates and the position of Gnathostomulida, Cycliophora,
Plathelminthes, and Chaetognatha: a combined approach
of 18S rDNA sequences and morphology. Syst Biol 49:539–62.
160
Goldstein B, Blaxter M. 2002. Tardigrades. Curr Biol 12:R475.
Gould SJ. 1995. Of tongue worms, velvet worms and water bears.
Nat Hist 95:6–15.
Hejnol A. 2002. Der postnaupliale Keimstreif von Porcellio
scaber und Orchestia cavimana (Crustacea, Peracarida):
Zellinie, Genexpression und Beginn der Morphogenese
[thesis]. Berlin: HU Berlin.
Hejnol A, Schnabel R. 2005. The eutardigrade Thulinia
stephaniae has an indeterminate development and the potential to regulate early blastomere ablations. Development
132:1349–61.
Hird SN, White JG. 1993. Cortical and cytoplasmic flow polarity
in early embryonic cells of Caenorhabditis elegans. J Cell Biol
121:1343–55.
Houthoofd W, Jacobsen K, Mertens C, Vangestel S, Coomans A,
Borgonie G. 2003. Embryonic cell lineage of the marine
nematode Pellioditis marina. Dev Biol 258:57–69.
Inoue I. 1958. Studies on the life history of Chordodes japonensis,
a species of Gordiacea. I, The development and structure of
the larva. Jap J Zool 12:203–18.
Jenner RA, Scholtz G. 2005. Playing another round of metazoan
phylogenetics: historical epistemology, sensitivity analysis,
and the position of Arthropoda within the Metazoa on the
basis of morphology. In: Koenemann S, Jenner RA, editors.
Crustacea and arthropod relationships. Lisse: A.A. Balkema.
p 355–86.
Kaufmann J. 1851. Ueber die Entwicklung und systematische
Stellung der Tardigraden. Z Wiss Zool 3:220–32.
Kinchin IM. 2000. On the significance of a “minor” phylum (the
Tardigrada) in the context of a constructivist view of
knowledge. Perspect Biol Med 43:243–51.
Kühn A. 1913. Die Sondierung der Keimesbezirke in der
Entwicklung der Sommereier von Polyphemus. Zool Jb
Anat 35: 243–340
Malakhov VV, Spiridonov SE. 1984. The embryogenesis of
Gordius sp. from Turkmenia, with special reference to the
position of the Nematomorpha in the animal kingdom.
Zool Zh 63:1285–97.
Mallatt JM, Garey JR, Shultz JW. 2004. Ecdysozoan phylogeny
and Bayesian inference: First use of nearly complete 28S and
18S rRNA gene sequences to classify the arthropods and their
kin. Mol Phylogenet Evol 31:178–91.
Mallatt JM, Winchell CJ. 2002. Testing the new animal phylogeny: first use of combined large-subunit and small-subunit
rRNA gene sequences to classify the protostomes. Mol Biol
Evol 19:289–301.
Marcus E. 1928. Zur Embryologie der Tardigraden. Verh
Deutsch Zool Ges 32:134–46.
Marcus E. 1929. Zur Embryologie der Tardigraden. Zool Jb Anat
50:334–79.
Mayer G, Ruhberg H, Bartolomaeus T. 2004. When an epithelium ceases to exist: an ultrastructural study on the fate of the
embryonic coelom in Epiperipatus biolleyi (Onychophora,
Peripatidae). Acta Zool 85:163–70.
A. Hejnol and R. Schnabel
Minden JS, Agard DA, Sedat JW, Alberts BM. 1989. Direct cell
lineage analysis in Drosophila melanogaster by time-lapse,
three-dimensional optical microscopy of living embryos. J
Cell Biol 109:505–16.
Nielsen C. 2001. Animal evolution. New York: Oxford
University Press.
Nielsen C. 2003. Proposing a solution of the ArticulataEcdysozoa controversy. Zool Scr 32:475–83.
Peterson KJ, Eernisse DJ. 2001. Animal phylogeny and the
ancestry of bilaterians: inferences from morphology and
18S rDNA gene sequences. Evol Dev 3:170–205.
Philippe H, Lartillot N, Brinkmann H. 2005. Multigene
analyses of bilaterian animals corroborate the monophyly
of Ecdysozoa, Lophotrochozoa and Protostomia. Mol Biol
Evol 22:1246–53.
Schierenberg E. 2005. Unusual cleavage and gastrulation in a
freshwater nematode: developmental and phylogenetic
implications. Dev Genes Evol 215:103–8.
Schnabel R, Hutter H, Moerman D, Schnabel H. 1997. Assessing
normal embryogenesis in Caenorhabditis elegans using a 4D
microscope: variability of development and regional specification. Dev Biol 184:234–65.
Scholtz G. 1997. Cleavage, germ band formation and head
segmentation: the ground pattern of the Euarthropoda. In:
Fortey RA, Thomas RH, editors. Arthropod relationships.
London: Chapman & Hall. p 317–332.
Scholtz G. 2002. The Articulata hypothesis—or what is a
segment? Org Divers Evol 2:197–215.
Scholtz G. 2003. Is the taxon Articulata obsolete? Arguments in
favour of a close relationship between annelids and arthropods. In: Legakis A, Sfenthourakis S, Polymeni R, ThessalouLegaki M, editors. The new panorama of animal evolution. 18th
International Congress of Zoology Sofia: Pensoft. p 489–501.
Sulston JE, Schierenberg E, White JG, Thomson JN. 1983. The
embryonic cell lineage of the nematode Caenorhabditis
elegans. Dev Biol 100:64–119.
Swofford D. 2002. Paup? phylogenetic analysis using parsimony
(and other methods), ver. 4.0b10 for Unix. Sunderland, MA:
Sinauer Associates.
Thomas C, DeVries P, Hardin J, White J. 1996. Fourdimensional imaging: computer visualization of 3D movements in living specimens. Science 273:603–7.
Tiegs OW. 1940. The embryology and affinities of the Symphyla,
based on a study of Hanseniella agilis. Q J Micros Sci
82:1–225.
Tiegs OW. 1947. The development and affinities of the
Pauropoda, based on a study of Pauropus sylvaticus. Q J
Micros Sci 88:165–267, 275–336.
Urbach R, Schnabel R, Technau GM. 2003. The pattern of
neuroblast formation, mitotic domains and proneural gene
expression during early brain development in Drosophila.
Development 130:3589–606.
von Erlanger R. 1895. Beiträge zur Morphologie der
Tardigraden. I, Zur Embryologie eines Tardigraden:
Macrobiotus macronyx Dujardin. Morph Jb 22:491–513.
4D microscopy of tardigrade development
von Wenck W. 1914. Entwicklungsgeschichtliche Untersuchungen an Tardigraden (Macrobiotus lacustris Duj.).
Zool Jb Anat 37:465–514.
Voronov DA. 1999. The embryonic development of Pontonema
vulgare (Enoplida: Onycholaimidea) with a discussion of
nematode phylogeny. Russ J Nematol 7:105–14.
161
Zhinkin L. 1949. Early stages in the development of Priapulus
caudatus [in Russian]. Dokl Akad Nauk SSSR 65:409–12.
Zhinkin L, Korsakova G. 1953. Early stages in the development
of Halicryptus spinulosus [in Russian]. Dokl Akad Nauk SSSR
88:571–3.
Voronov DA, Panchin YV. 1998. Cell lineage in marine
nematode Enoplus brevis. Development 125:143–50.
Weygoldt P. 1963. Grundorganisation und Primitiventwicklung
bei Articulaten. Zool Anz 171:363–76.
Weygoldt P. 1979. Gastrulation in the Arthropoda? Fortschr
Zool Syst Evolutionsforsch 1:73–81.
Zilch R. 1979. Cell lineage in arthropods? Fortschr Zool Syst
Evolutionsforsch 1:19–41.
Zrzavy J. 2001. Myzostomida are not annelids: molecular and
morphological support for a clade of animals with anterior
sperm flagella. Cladistics 17:170–98.
Zrzavy J. 2003. Gastrotricha and metazoan phylogeny. Zool Scr
32:61–81.
Weygoldt P. 1986. Arthropod interrelationships: the
phylogenetic-systematic approach. Z Zool Syst Evolutionsforsch 24:19–35.
Zrzavy J, Mihulka S, Kepka P, Bezdek A, Tietz D. 1998.
Phylogeny of the Metazoa based on morphological and 18S
ribosomal DNA evidence. Cladistics 14:249–85.