Content uploaded by Olga I Raikova
Author content
All content in this area was uploaded by Olga I Raikova on Sep 09, 2018
Content may be subject to copyright.
Organisation of the nervous
system in the Acoela: an
immunocytochemical study
M. Reuter,
1
O. I. Raikova,
2
U. Jondelius,
3
M. K. S. Gustafsson,
1
A. G. Maule,
4
D. W. Halton
4
Abstract. In order to broaden the information about the organisation of the nervous system in taxon
Acoela, an immunocytochemical study of an undetermined Acoela from Cape Kartesh,
Faerlea
glomerata
,
Avagina incola
and
Paraphanostoma crassum
has been performed. Antibodies to 5-HT and
the native flatworm neuropeptide GYIRFamide were used. As in earlier studies, the pattern of 5-HT
immunoreactivity revealed an anterior structure composed mainly of commissures, a so-called
commissural brain. Three types of brain shapes were observed. No regular orthogon was visualised.
GYIRFamide immunoreactive cell clusters were observed peripherally to the 5-HT immunoreactive
commissural brain. Staining with anti-GYIRFamide revealed more nerve processes than did staining
with anti-FMRFamide. As no synapomorphies were found in the organisation of the nervous system of
the Acoela and that of the Platyhelminthes, the results support the view that the Acoela is not a
member of the Platyhelminthes. ß2001 Harcourt Publishers Ltd
Keywords: Platyhelminthes, Acoela, nervous system, immunocytochemistry
Introduction
Traditionally, the Acoela and the Nemertodermatida
have been regarded as sister taxa and classi®ed as belong-
ing to the Platyhelminthes. However, during the last dec-
ade their phylogenetic position has become subject of
controversy (for discussion, see Raikova et al., 1998,
2001; Reuter et al., 1998). Recently, the molecular sys-
tematics of ¯atworms has been studied intensely.
According to Katayama et al. (1996), the Acoela
diverged early from the Platyhelminthes. Carranza et al.
(1997) suggest that the Acoelomorpha (Acoela and
Nemertodermatida) should be regarded as a sister group
of the Bilateria. A similar result was obtained by Ruiz-
Trillo et al. (1999) when studying the sequences of 18S
rDNA from non-fast-clock species of Acoela, i.e. species
that have evolved slowly. They pointed out that Acoela
are not a members of Platyhelminthes. However, the
Nemertodermatida appears to be buried within the bulk
of the Platyhelminthes (Ruiz-Trillo et al., 1999). The ana-
lysis of the sequences of 18S rDNA and 28S rDNA from
the Acoela on the one hand, and from other Platyhelminth
taxa on the other, indicate a wide phylogenetic divergence
between the two groups (Katayama et al., 1996; Carranza
et al., 1997; Ruiz-Trillo et al., 1999; Litvaitis & Rohde,
1999; Littlewood et al., 1999). Littlewood et al. (1999)
suggest that further studies are needed to certify the correct
phylogenetic position of the taxa Nemertodermatida and
Acoela.
Tissue & Cell, 2001 33 (2) 119±128
ß2001 Harcourt Publishers Ltd
DOI: 10.1054/tice.2000.0134, available online at http://www.idealibrary.com
Tissue&Cell
119
1
Department of Biology, A
Êbo Akademi University, Artillerigatan 6, FIN-
20520 A
Êbo, Finland.
2
Zoological Institute, Russian Academy of Sciences,
199034 St Petersburg, Russia.
3
Swedish Museum of Natural History,
POB 50007, SE-104 05 Stockholm, Sweden.
4
Parasitology Research
Group, School of Biology & Biochemistry, The Queen's University of
Belfast, Belfast, BT9 7BL, UK.
Received 9 June 2000
Accepted 1 August 2000
Correspondence to:
M. Reuter, Department of Biology, A
Êbo Akademi
University, BIOCITY, Artillerigatan 6A, II va
Ên, FIN-20520 A
Êbo, Finland
In the discussion of the phylogenetic position of
Platyhelminthes, the organisation of the nervous system
has been used as one of the discriminating criteria. The
traditional concept stated that the nervous system and the
construction of the brain of the Acoelomorpha repre-
sented the earliest form of nervous system, and that they
were inherited from a common Platyhelminth ancestor.
Neuroanatomical data collected by many authors were
critically reanalysed by Haszprunar (1996). According to
this author, the brain of Acoela evolved independently
from that of other Platyhelminthes, i.e. they are non-
homologous structures. The great variations in the
appearance of the neural organisation and the neural
centralisation at the anterior end of Acoela indicates
that multiple events of evolution have taken place
(Haszprunar, 1996). Based on these data, Haszprunar
considers the Acoelomorpha as the earliest offshoot of
the bilaterian stem line.
The application of immunocytochemistry (ICC) and
in particular the study of the immunoreactivity (IR) pat-
terns using antibodies to serotonin (5-HT) and
FMRFamide-related peptides (FaRPs) (Shaw et al.,
1996) have signi®cantly increased the knowledge of the
organisation of the Platyhelminth nervous system. The
reliability of the technique has been proven in numerous
studies (for reviews see Gustafsson, 1992; Reuter &
Gustafsson, 1995; Halton & Gustafsson, 1996; Reuter
& Halton, 2001; Reuter & Gustafsson, 2000). Two
immunocytochemical studies of the patterns of 5-HT-
and FMRFamide-IR in the nervous system of four spe-
cies of Acoela were recently performed by Raikova et al.
(1998) and Reuter et al. (1998). These studies showed that
the construction of the brain of the Acoela and the
organisation of the longitudinal nerve cords differ
signi®cantly from the corresponding IR patterns in
species of the taxon Platyhelminthes. Recently, the 5-
HT- and FMRFamide-IR patterns of the nervous
systems in two species of the Nemertodermatida were
studied by Raikova et al. (2000, 2001). They show distinct
differences from the corresponding patterns obtained
from Acoela.
This study aims to broaden the information about the
organisation of the nervous system in the Acoela by ex-
amining the patterns of 5-HT-IR and GYIRFamide-IR
in four additional species. The results will be discussed
against the background of earlier studies.
Materials and methods
Animals
Specimens of a small 0.5 mm long undetermined species
of Acoela were collected from tidal pools near the White
Sea Biological Station at Cape Kartesh (Russia).
Specimens of Faerlea glomerata Westblad, 1945 and
Paraphanostoma crassum Westblad, 1942 were collected
at 40±60 m depth from sand/muddy bottoms in the vicin-
ity of Kristineberg Biological Station (West coast of
Sweden). Specimens of Avagina incola Leiper, 1902 were
obtained from the gut of the sea urchin Spatangus purpur-
eus O.F. Mu
Èller, 1776 at 100 m depth in the vicinity of
Bergen, Norway.
Immunocytochemistry
Specimens of undertermined Acoela from Cape Kartesh,
F. glomerata,P. crassum and A. incola were ®xed in
Stefanini's ®xative (2% paraformaldehyde and 15% picric
acid in 0.1 M sodium phosphate buffer) at pH 7.6, stored
for several weeks in ®xative at 48C and rinsed for 24±48 h
0.1 M sodium phosphate buffer (pH 7.6) containing
10±20 % sucrose. The worms were handled as whole
mounts on poly-L-lysine coated glass slides, allowed to
dry and frozen at ÿ708C. Prior to staining, the whole
mounts were thawed and immersed in phosphate-
buffered saline (PBS) containing 1% bovine serum
albumin (BSA) and 0.2 % Triton X-100 (PBS-T).
Immunostaining was performed according to the indirect
immuno¯uorescence method of Coons et al. (1955).
The concentration of rabbit anti-5-HT was 1:500.
Incubations were performed with a cocktail of rabbit
anti-5-HT (Sigma) antiserum (1:100) and guinea pig anti-
serum against the native ¯atworm neuropeptide GYIRF
(Bdelloura candida) (1:500±1:1000) raised in guinea pigs
by Johnston et al. (1995). Incubation times were 8 h, 24 h,
36 h, 48 h or 72 h. Incubation was performed at RT or
48C and followed by rinsing 3 5 min in PBS-T,
incubation for 1 h with the secondary antibody FITC
labelled goat anti-guinea pig immunoserum (Cappel)
(dilution 1:30). Thereafter the preparations were washed
35 min in PBS-T and incubated for 1 h with the sec-
ondary antibody TRITC labelled swine anti-rabbit im-
munoserum (DAKO) (dilution 1:30) followed by washing
35 min in PBS-T. The whole mounts were then
mounted in 50 % glycerol and stored in the dark at
ÿ208C. The controls for speci®city included: (1) omitting
the primary antibody and (2) using non-immune serum.
Microscopy
The preparations were examined in a Leitz Orthoplan
microscope combined with ®lter blocks I2 and N2.
Photomicrographs were produced by an Olympus auto-
matic photomicrograph system, model PM 10ADS (®lm:
TMAX 400). A confocal scanning laser microscope
(CSLM: LEICA TCS 4D) was used to better visualise
details of the nervous system.
Computer processing of immunochemistry micrographs
Files obtained from CSLM were processed with Adobe
Photoshop 4.0. Only the commands `mode RGB-
Grayscale', `level of grey', `brightness' and `contrast'
were used. This was to avoid any distortion of the
information contents of the image.
120 REUTER ET AL.
Results
Undetermined Acoela from Cape Kartesh
5-HT immunoreactivity
This animal was stained only with anti-5-HT. The im-
munoreaction revealed four pairs of longitudinal nerve
cords, dorsal (dc), lateral (lc), ventrolateral (vlc) and
ventral (vc) (Fig. 6 A±C). The staining reaction was
equally strong on the dorsal and ventral sides. Thin trans-
versal ®bres connecting the cords were observed in the
frontal end of the worm. In other parts of the body an
irregular mesh of transverse ®bres was observed.
Faerlea glomerata
5-HT immunoreactivity
Brain 5-HT-IR revealed two transverse commissures (c1,
c2,) on the dorsal side in front of the statocyst (Fig. 1, Fig.
7A. The commissures connect the strongly stained ante-
rior parts of the dorsal longitudinal nerve cords (dc).
Branches from the dorsal cords connect them with the
lateral longitudinal cords (1c) (Fig. 7A, B). Cell bodies
occur at the crossing points of the transverse commissures
and the longitudinal nerve cords. On the ventral side thin
transverse commissures (c3, c4) occur at the same level as
the dorsal in front of the statocyst (Fig. 1, Fig. 7D). The
commissures connect the ventral longitudinal nerve cords
(vc). The most anterior dorsal transverse commissures
are connected to a ventral nerve net (Fig. 7C). 5-HT
immunoreactive frontal cells occur just behind the
anterior commissures. 5-HT immunoreactive cell bodies
are also observed at crossing points of ventral com-
missures and the nerve net. A very thin transverse
commissure behind the statocyst (s) forms part of a ven-
tral nerve net.
Longitudinal nerve cords Four pairs of nerve cords
have been observed: dorsal (dc), lateral (lc), ventrolateral
(vlc) and ventral (vc). The strongly stained anterior parts
of the longitudinal cords in the brain continue caudally as
dorsal nerve cords. They can be followed to the middle of
the body. Branches starting at the foremost dorsal trans-
verse commissure (c2) bend backwards forming lateral
longitudinal nerve cords (1c) (Fig. 7B). On the ventral
side, ventral (vc) and ventrolateral longitudinal nerve
cords (vlc) originate at cell bodies of the most anterior
transverse commissures (long arrow). Additionally, mar-
ginal nerve ®bres are observed. They seem to be part of
the ventral nerve net (nn).
GYIRFamide immunoreactivity
No GYIRF-IR was detected in this species. Changes of
antibody dilutions, incubation times and temperature did
not improve the staining.
Avagina incola
5-HT immunoreactivity
Brain 5-HT-IR revealed a bilateral bridge-like structure
composed of ®ne nerve ®bres (Fig. 2, Fig. 8A). Thin
anterior transversal commissures (tc) represent the arch
of the bridge. The lateral abutments of the bridge are
composed of masses of loosely interwoven ®bre loops
(Fig. 8A, large arrow). While a few large loops were
observed dorsally, only a network of ®ner loops occur
on the ventral side. The ®bres are most concentrated at
the ventral and basal parts of the abutments. A pair of
symmetrically located 5-HT-immunoreactive cells occur
latero-posteriorly of the ®bre loops, where the caudal
dorsal longitudinal nerve cords originate (dc).
Longitudinal nerve cords 5-HT-IR revealed two pairs
of longitudinal caudal nerve cords, the dorsal (dc) and the
c1
c3
c2
c4
nn
vc dc vlc lc
s
Fig. 1 Faerlea glomerata, schematic drawing of the organisation of 5-HT
immunoreactive elements in the nervous system. Barrel-shaped brain
composed of longitudinal connectives linked dorsally (c1, c2) and ventrally
(c3, c4) by transversal commissures. Symmetrical cells occur at crossing
points of transverse and longitudinal connectives (arrow). Longitudinal
nerve cords are formed as a continuation of longitudinal brain connectives,
dorsal cords (dc), ventral cords (vc), ventro-lateral cords (vlc), lateral cords
(lc), ventral nerve net (nn), statocyst (s).
tc
dc
lc
Fig. 2 Avagina incola, schematic drawing of the organisation of 5-HT
immunoreactive elements in the nervous system. Symmetrically arranged
concentrations of fibre loops connected by thin transverse commissures (tc)
form the brain. Cell bodies occur at the origin of dorsal nerve cords (dc)
and lateral nerve cords (lc) branching into three branches on each side. Two
rows of sensilla (arrows) in the frontal region are connected by fine
processes to the brain.
ORGANISATION OF THE NERVOUS SYSTEM IN THE ACOELA 121
lateral nerve cords (lc) (Fig. 2, Fig. 8A). Both nerve cords
originate at the cell pair located latero-posteriorly in the
brain. The lateral cords form a continuation of the trans-
verse brain commissure and more laterally they branch
into two thin longitudinal nerve ®bres, a dorsolateral and
a marginal one. The dorsal and lateral nerve cords extend
to the middle of the body. Meshes of a ®ne 5-HT im-
munoreactive nerve network were occasionally observed
in the tail end on the ventral side (Fig. 8C).
Sensilla Two rows of pear-shaped cell bodies (size
10 mm) occur in the frontal end (Fig. 2, Fig. 8A small
arrows). Thin nerve ®bres connect them with the tangled
®bre loops of the brain.
GYIRFamide immunoreactivity
Brain The GYIRFamide-IR revealed a brain composed
of symmetrical clusters of about three closely packed cells
on each side (Fig. 3, Figs 8B, E±F). Nerve processes (tc)
run towards the centre from each cluster. After a short
distance the processes send several very ®ne branches
anteriorly(Fig. 8B, arrows).These ®ne processessubmerge
into a subepidermal nerve net forming large irregular
meshes (Fig. 8D). Laterally, at the base of the cell clusters,
are processes that run latero-posteriorly (1c).
Longitudinal nerve cords The aforementioned pro-
cesses originating at the cell clusters form short caudal
longitudinal lateral nerve cords (Fig. 3, 1c). Branches
from the cords to the periphery were observed. They
submerge into meshes of the nerve net (Fig. 8E).
Paraphanostoma crassum
5-HT immunoreactivity
Brain-like structure The 5-HT-IR revealed an anterior
nerve centralisation composed of anterior ®bre loops
(arrows) and a connecting bridge-like construction
surrounding the statocyst (Fig. 4, Figs 9 A±C).
Posterior to the statocyst, a strongly stained transversal
commissure (tc) gives rise to a pair of ®bres that form a
semicircle or possible full circle around the statocyst (s) ±
as the anterior commissure (ac) (Fig. 9B). Nerve nets are
located at the lateral margins of the semicircle. Three
cells (arrow) are observed frontally (Fig. 9, Fig. 9B).
Symmetrically located cells (Fig. 4, Fig. 9B) (arrow)
occur, associated with the transversal brain commissure
(tc) at the beginning of the dorsal longitudinal nerve
cords (dc). Frontally, nerve ®bres form two pairs of ante-
rior loops (Fig. 4). Basally the transversal commissure
forms an arch which bends backwards into the strongly
stained lateral nerve cords (lc).
Longitudinal nerve cords Four pairs of posterior 5-HT
immunoreactive longitudinal nerve cords were observed
(Fig. 4, Fig. 9A): dorsal (dc), ventral (vc), lateral (lc) and
marginal (mc) nerve cords. Symmetrically located cell
bodies (Fig. 4) occur at the beginning of the posterior
dorsal cords along the transversal brain commissure. An
additional cell pair is observed at the origin of the ventral
cords. The lateral and the dorsal cords showed strong 5-
HT-IR and could be followed caudally to the middle of
the body. Anteriorly, they continue as anterior dorsal
(adc) and lateral (alc) nerve cords, respectively. The ante-
rior dorsal cords join into a large anterior arch while
branches from the lateral cords form the marginal nerve
cords (mc). A 5-HT immunoreactive nerve net occurs
close to the body surface (Fig. 9E).
Sensilla A row of pear-shaped 5-HT immunoreactive
sensory cell bodies (Fig. 4) was observed close to the
surface in the periphery along the lateral frontal cords.
adc
ac
mc tc
dc vc lc
alc
s
Fig. 4 Paraphanostoma crassum, schematic drawing of the
organisation of 5-HT immunoreactive elements in the nervous system.
Bridge-shaped brain composed of transverse posterior commissure (tc)
and anterior semicircle (ac) connecting anterior fibre loops and fibre
nets. Cell bodies (arrows) frontally and at the origin of longitudinal
nerve cords, dorsal cord (dc), ventral cord (vc), lateral cord (lc), and
marginal cord (mc) formed as branches from lateral cords. Lateral and
dorsal cords extend anteriorly as anterior dorsal (adc) and anterior lateral
(alc) cords.
tc cl
lc
Fig. 3 Avagina incola, schematic drawing showing organisation of
GYIRFamide immunoreactive elements in the nervous system. Symmet-
rical cell clusters (cl), anterior processes (arrow), processes parallel to the
5-HT immunoreactive lateral cords (lc) and thin transverse nerve fibre (tc)
connecting the cell clusters.
122 REUTER ET AL.
GYIRFamide immunoreactivity
Brain The GYIRFamide-IR revealed two lateral
groups of loosely packed cells (Fig. 5, Fig. 9D). Latero-
posteriorly, three large cells are particularly noticeable.
The cells are pear-shaped and connect to a network of
®bre loops. The ®bre network gives rise to short ®bres
running towards the frontal end of the body. A thin
transversal commissure (tc) behind the statocyst (s) con-
nects the cell groups.
Longitudinal nerve cords A pair of short dorsal lon-
gitudinal nerve cords (dc) forms the continuations of the
thin transverse commissure (tc). The cords can be fol-
lowed backwards to about the middle of the body. The
transversal commissure and the dorsal nerve cords run
parallel to the corresponding 5-HT immunoreactive
structures. Some nerve ®bres were observed close to the
body surface (Fig. 9 F).
Discussion
The distinct differences in the 5-HT-IR and FaRP-IR
patterns of the four species previously studied by
Raikova et al. (1998) and Reuter et al. (1998) and the
four new species of the Acoela studied here indicate that
the two IR patterns should be discussed separately.
Thereafter, the phylogenetic implications of the patterns
are discussed.
The 5-HT immunoreactivity pattern
Common for the 5-HT-IR patterns of the Acoela is an
anterior symmetrical brain-like structure composed of
commissural nerve ®bres associated with a few cell
bodies. According to Westblad (1948), the brain-like
structure in the Acoela varies considerably in shape and
undergoes an evolutionary development from a ring-like
to a bridge-like structure. The results of the present in-
vestigation con®rm his observations.
In the present study, the varying shapes of the brain of
the Acoela are united into three main categories: the
barrel shape, the rosette shape and the bridge shape.
The most primitive structure occurs in the undetermined
Acoela from Cape Kartesh, in which no brain-like struc-
ture was seen, only equally strongly stained longitudinal
cords close to the body surface and weakly stained trans-
verse elements were observed.
Westblad (1948) described the brain of several Acoela
species `Diopisthoporus, Haploposthia, Faerlea fragilis,
Anaperus tvaerminnensis and Mecynostomum', which he
considered primitive. He used the term ring-formed to
describe the brain shape. However, judging from his
schematic drawings, the term is slightly misleading as
the longitudinal elements forming the brain closely re-
semble a barrel, with strong ribs and weak transverse
elements (barrel hoops). Thus, we suggest that the term
barrel-shape should be used for the brains of the Acoela
species Anaperus biaculeatus and Actinoposthia bekle-
mischevi (Raikova et al., 1998) and for the undetermined
Acoela from Cape Kartesh and F. glomerata from the
present study. In the undetermined acoel from Cape
Kartesh, all longitudinal cords appear to be equally
developed with poorly developed commissures. In A.
beklemischevi, the commissures are more pronounced.
In F. glomerata, the transverse elements and the long-
itudinal ones are more developed on the dorsal side.
However, no nerve ®bres occur in the parenchyma.
In A. biaculeatus, the dorsal part of the brain dominates
but the ventral part can still be easily distinguished
(Raikova et al., 1998).
The brain of Childia groenlandica and Mecynostomum
sp. (Raikova et al., 1998) is rosette-shaped. Here, the
anterior loops join a common transversal commissure
on the dorsal side. In these two species the loops are
large and loosely connected (Raikova et al., 1998).
A trend towards a bridge-shaped brain can be seen in
A. incola where the loops are small and concentrated to a
bridge-like construction. In P. crassum, a strongly stained
dorsal transverse commissure gives rise to a semicircle
around the statocyst. The construction of the brain and
the close association between the nerves and the statocyst
indicate a distinct evolutionary development from super-
®cial centralisation close to the body surface to true
cephalisation.
The FaRP immunoreactivity pattern
The peptidergic component of the nervous system has
been recognised as a fundamentally important part of
the metazoan nervous system (Shaw et al., 1996). The
major difference between classical transmitters and
neuropeptides is their mode of synthesis. While classical
s
tc
dc
cl
Fig. 5 Paraphanostoma crassum, schematic drawing of the
organisation of GYIRF immunoreactive elements in the nervous system.
Symmetrical groups of cells connected by transversal commissure (tc).
Anterior processes (arrows) and networks of fibre loops. Dorsal
longitudinal nerve cords (dc).
ORGANISATION OF THE NERVOUS SYSTEM IN THE ACOELA 123
transmitters are synthesised in nerve terminals, the neu-
ropeptides are synthesised in the cell bodies of the pep-
tidergic neurons, processed within vesicles and then
transported along the axons to the nerve terminals. In
conformity with the IR pattern obtained by antibodies of
the molluscan cardioactive neuropeptide FMRFamide
(Reuter et al., 1998), the GYIRFamide-IR pattern re-
vealed an anterior centralisation dominated by cell clus-
ters. More nerve processes were identi®ed using an
antibody to the native ¯atworm neuropeptide
GYIRFamide. So far, four FaRPs have been isolated
from ¯atworms: 1. GNFFRFamide from the cestode
Moniezia expansa (Maule et al., 1993); 2. RYIRFamide
from the terrestrial planarian Artioposthia triangulata
(Maule et al., 1994); 3. GYIRFamide from Girardia
(Dugesia) tigrina and Bdelloura candida (Johnston et al.,
1995, 1996); and 4. YIRFamide from B. candida
(Johnston et al., 1996). It appears that the FaRP structure
of the ancestral ¯atworm may have been XYIRFamide,
where X is a variable residue (Shaw et al., 1996).
Nothing is known about the corresponding FaRP of the
Acoela. However, the IR pattern obtained with
anti-GYIRFamide is more complete than with anti-
FMRFamide, indicating the occurrence of an endogen-
ous FaRP which is more structurally related to
GYIRFamide than to FMRFamide.
The results of this study show that the patterns of 5-
HT-and FaRP-IR in the Acoela differ clearly from those
in the taxon Platyhelminthes. However, when discussing
the neuroanatomy and the construction of the brain it has
to be kept in mind that the present view is based on the
patterns of only two neurotransmitter substances.
Ultrastructural studies have revealed abundant nerve
cells close to the statocyst (Ehlers, 1985; Bedini &
Lanfranchi, 1991), but these cells did not react with the
5-HT and the FaRP antibodies. To clarify this fact,
staining for other neuronal signal substances should be
carried out.
6A 6B 6C
7A 7B
7D7C
c1
c3
c4
vlc
lc
dc
nn vc
c2
dc
lc
lc
vc
vlc
dc
dc
lc lc
s
Fig. 6 A±C Undetermined acoel from Cape Kartesh. Optical sections showing organisation of 5-HT immunoreactive elements. 180. (A) dorsal side,
dorsal longitudinal nerve cord (dc); (B) mid body, lateral cord (lc); (C) ventral side, ventro-lateral cord (vlc), ventral cord (vc). Fig. 7 A±D Faerlea glomerata,
optical sections of anterior end showing: (A) dorsal side, (B) central part, (C) surface nerve net, (D) ventral side. 180. Longitudinal brain connectives
continuing backwards as dorsal nerve cord (dc), ventro-lateral cord (vlc), lateral cord (lc), ventral cord (vc), transverse brain commissures, dorsal
commissures (c1, c2), ventral commissures (c3, c4), surface nerve net (nn), statocyst (s).
124 REUTER ET AL.
8A
8C
8E 8F
8B
8D
tc tc cl
lc
lc dc
cl cl
Fig. 8 A±F Avagina incola. (A) Optical section showing organisation of 5-HT immunoreactive elements in anterior end. Symmetrical concentrations
of interwoven fibre loops (long arrow) connected by transverse commissures (tc). Branching lateral nerve cords (lc). Two frontal rows of sensilla
(short arrows) and processes connecting them to the brain. 180. (B) Optical sections showing organisation of GYIRFamide immunoreactive
elements in anterior end. Symmetrical cell clusters (cl), anterior processes (arrows) and lateral processes continuing in lateral nerve cord (lc). 180.
(C) Optical section showing 5-HT immunoreactive nerve net in posterior end (arrow). 230. (D) Optical section showing GYIRFamide immunoreactive
nerve net (arrow). 230. (E, F) Larger magnifications of GYIRFamide immunoreactive cell clusters showing the compact construction of 3±4 cell
bodies (cl). 360.
ORGANISATION OF THE NERVOUS SYSTEM IN THE ACOELA 125
9A
9C
adc
adc
adc
dc
alc
alc
mc tc
ac tc
tc cl
cl
tc
dc
dc
vc
lc
lc
9B
9D
9F9E
s
Fig. 9 A±F Paraphanostoma crassum. (A±C) Optical sections showing 5-HT immunoreactive elements.(A) Anterior end overview showing
bridge-shaped brain and longitudinal nerve cords. Transversal commissure (tc), dorsal nerve cord (dc), ventral nerve cord (vc) lateral nerve cord (lc),
marginal nerve cord (mc), anterior dorsal nerve cord (adc), anterior lateral nerve cord (alc). 180. (B) Details of brain and nerve cords.
Transversal commissure (tc), anterior commissure forming semicircle (ac) around statocyst (s), cell bodies (arrows), lateral nerve cord (lc), anterior
dorsal nerve cord (adc), anterior lateral nerve cords (alc). 280. (C) 5-HT immunoreactive transverse commissure (tc), dorsal nerve cord (dc) and
anterior dorsal nerve cord (adc), 280, in optical section corresponding to (D) showing GYIRFamide immunoreactive elements in anterior end.
Symmetrical groups of cells (cl), connecting transverse commissure (tc), and processes running anteriorly (arrows), dorsal longitudinal nerve
cord (dc). 280. (E) Optical section showing 5-HT immunoreactive nerve net (arrow). 380. (F) GYIRFamide immunoreactive nerve fibres
visualised near body surface (arrow). 380.
126 REUTER ET AL.
Phylogenetic implications
According to Ivanov and Mamkaev (1973), the brain of
the Acoela is composed of two parts ± an outer part, the
orthogonal brain, formed by the thickened anterior
parts of the longitudinal nerve cords and the foremost
transverse commissures and an inner part, the endonal
brain, concentrated around the statocyst. The endonal
brain was considered homologous in all Platyhelminthes.
The ®ndings by Reuter et al. (1998), Raikova et al. (1998)
and the present study bring forward the question about
the presence of a double brain. Staining with anti-5-HT
shows the presence of a brain composed of longitudinal
cords and commissures ± an orthogonal brain. However,
the cell groups, reacting positive with anti-FMRFamide
(Raikova et al., 1998) and anti-GYIRFamide, do not
surround the statocyst and can therefore not be con-
sidered to correspond to the endonal brain (Ivanov &
Mamkaev, 1973). The observations of Ehlers (1985) and
Bedini and Lanfranchi (1991) of nerve cells around the
statocyst of acoels, indicates the presence of other neuro-
transmitters.
In the central nervous system of the Catenulida and
the Rhabditophora the patterns of 5-HT and FaRP-IR
are basically similar. The pattern consists of a neuropile
surrounded by cell bodies forming symmetrical lobes and
a pair of main longitudinal nerve cords (MCs) joined by
5-HT immunoreactive marker neurones (for references
see Gustafsson, 1992; Reuter & Gustafsson 1995, 2000;
Halton & Gustafsson, 1996; Reuter & Halton, 2001).
In the peripheral nervous system, FaRP-IR occurs in
the pharynx and the gut (for references see Kreshchenko
et al., 1999). In addition, a strong association of FaRP-IR
with the reproductive system has been observed (Shaw
et al., 1996; Reuter & Gustafsson, 2000). 5-HT-IR
appears in the nerve plexus close to the body musculature
and in the pharynx of more advanced Platyhelminthes
(for review see Gustafsson, 1992; Reuter & Gustafsson
1995; Halton & Gustafsson, 1996; Reuter & Halton
2001).
Compared to the aminergic and peptidergic IR pat-
terns obtained in Platyhelminthes (references as above)
the following differences in the patterns characterise the
Acoela:
1. The presence of a structure named commissural
brain, showing positive reactions for anti-5-HT,
catecholamines and acetylcholinesterase
(Raikova et al., 1998) but with no resemblance to
the bilobed brain structure in Plathyhelminthes,
with its nerve cells surrounding a neuropile.
2. The presence of clusters of peptidergic FaRP
positive cells that are not integrated into a brain of
the common ¯atworm type.
3. The absence of a regular orthogon. Only
longitudinal nerve cords, connected by irregular
nerve ®bres forming a subepidermal nerve plexus,
were observed.
4. The absence of 5-HT immunoreactive marker
neurones along the MCs.
5. The absence of an inner (stomatogastric) FaRP
immunoreactive nervous system.
Thus, no synapomorphies in the organisation of the
nervous system of Acoela and previously studied
Platyhelminthes were found. Our results support the
view of Katayama et al. (1996), Carranza et al. (1997),
Ruiz-Trillo et al. (1999), Litvaitis & Rohde (1999) and
Littlewood et al. (1999), based on molecular systematics,
that the taxon Acoela is separate from the taxon
Platyhelminthes.
ACKNOWLEDGEMENTS
The authors wish to thank the staff of the marine
biological station at Kristineberg (Sweden) for their
help in collecting the material. Thanks are extended to
Mr E. Nummelin and Mr J. Korhonen for valuable assist-
ance, to the Research Institute of the A
Êbo Akademi
University Foundation, the Foundation for Swedish
Culture in Finland and the Swedish Natural Science
Research Council for ®nancial support. Olga Raikova
was bene®ciary of the Visby Scholarship of the Swedish
Institute and of a Russian Basic Research Foundation
grant RFFI-99-04-49783.
REFERENCES
Bedini, C. and Lanfranchi, A. 1991. The central and peripheral
nervous system of Acoela (Platyhelminthes). An electron micro-
scopical study. Acta Zool. (Stockholm), 72, 101±106.
Carranza, S., Baguna
Á, J. and Riutort, M. 1997. Are the
Platyhelminthes a monophyletic primitive group? An assessment
using 18S rDNA sequences. Mol. Biol. Evol., 14, 485±497.
Coons, A.H., Leduc, E.H. and Connolly, J. M. 1955. Studies on
antibody production. I. A method for the histochemical demon-
stration of specific antibody and its application to a study of a
hyperimmune rabbit. J. Exp. Med., 102, 49±60.
Ehlers, U. 1985. Das phylogenetische System des Plathelminthes.
Gustav Fischer Verlag, Stuttgart.
Gustafsson, M.K.S. 1992. The neuroanatomy of parasitic flatworms.
Adv. Neuroimmunol., 2, 267±286.
Halton, D.W. and Gustafsson, M.K.S. 1996. Functional morphology
of the platyhelminth nervous system. Parasitology, 113, S47±S72.
Haszprunar, G. 1996. Platyhelminthes and Platyhelminthomorpha ±
paraphyletic taxa. J. Zool. Syst. Evol. Res. 34, 41±48.
Ivanov, A.V. and Mamkaev, Yu. V. 1973. Turbellaria, their origin and
evolution. Nauka, Leningrad (in Russian), 1±221.
Johnston, R.N., Shaw, C., Halton, D.W., Verhaert, P. and Baguna
Á, J.
1995. GYIRFamide: a novel FMRFamide-related peptide (FaRP)
from the triclad turbellarian, Dugesia tigrina. Biochem. Biophys.
Res. Com., 209, 689±697.
Johnston, R.N., Shaw, C., Halton, D.W., Verhaert, P., Blair, K.L.,
Brennan, G.P., Price, D.A. and Anderson, P.A.V. 1996. Isolation,
localization and bioactivity of the FMRFamide-related neuropep-
tides GYIRFamide and YIRFamide from the marine turbellarian
Bdelloura candida. J. Neurochem. 67, 814±821.
Katayama, T., Nishioka, M. and Yamamoto, M. 1996. Phylogenetic
relationships among turbellarian orders inferred from 18S rDNA
sequences. Zool. Science. 13, 747±756.
Kreshchenko, N.D., Reuter, M., Sheiman, I.M., Halton, D.W.,
Johnston, R.N., Shaw, C. and Gustafsson, M.K.S. 1999.
Relationship between musculature and nervous system in the
ORGANISATION OF THE NERVOUS SYSTEM IN THE ACOELA 127
regenerating pharynx in Girardia tigrina (Platyhelminthes). Invert.
Repr. Dev. 35, 109±125.
Littlewood, D.T.J., Rohde, K. and Clough, K.A. 1999. The
interrelationships of all major groups of Platyhelminthes:
phylogenetic evidence from morphology and molecules. J. Linn.
Soc. 66, 75±114.
Litvaitis, M.K. and Rohde, K. 1999. A molecular test of platyhelminth
phylogeny: interferences from partial 28S rDNA sequences. Invert.
Biol., 118, 42±56.
Maule, A.G., Shaw, C., Halton, D.W., Curry, W.J. and Thim, L. 1994.
RYIRFamide: a turbellasian FMRFamide-related peptide (FaRP).
Regulatory Peptides, 50, 37±43.
Maule, A.G., Shaw, C., Halton, D.W. and Thim, L. 1993.
GNFFRFamide: a novel FMRFamide-immunoreactive peptide
isolated from the sheep tapeworm Moniezia expansa. Biochem.
Biophys, Res. Com., 193, 1054±1060.
Raikova, O.I., Reuter, M., Jondelius, U. and Gustafsson, M.K.S.
2000. The brain of Nemertodermatida (Platyhelminthes) as
revealed by anti-5-HT and anti-FMRFamide immunostainings.
Tissue & Cell, 32, 358±365.
Raikova, O.I., Reuter, M. and Justine, J.-L. 2001. Contributions to the
phylogeny and systematics of the Acoelomorpha. In:
Interrelationships of Platyhelminthes (eds. T.J. Littlewood and
R.A. Ray), Taylor and Francis, London, pp 13±23.
Raikova, O.I., Reuter, M., Kotikova, E.A. and Gustafsson, M.K.S.
1998. A commissural brain! The pattern of 5-HT immunoreactivity
in Acoela (Platyhelminthes). Zoomorphology, 118, 69±77.
Reuter, M. and Gustafsson, M.K.S. 1995. The flatworm nervous
system: pattern and phylogeny. In: The nervous systems of
invertebrates: an evolutionary and comparative approach
(eds. O. Breidbach and W. Kutsch). Birkha
Èuser Verlag, Basel,
Zoomorphology. 25±59.
Reuter, M. and Gustafsson, M.K. 2000. Platyhelminthes. In:
Reproductive Biology of Invertebrates, (ed. A. Dorn). John Wiley
& Sons, Chichester, New York, Weinheim, Brisbane, Singapore,
Toronto pp.73±103.
Reuter, M. and Halton, D. 2001. Comparative neurobiology of
Platyhelminthes. In: The Interrelationships of Platyhelminthes,
(eds. T. J. Littlewood and R.A. Ray), Taylor and Francis, London,
pp 239±279.
Reuter, M., Raikova, O.I. and Gustafsson, M.K.S. 1998. An endocrine
brain? The pattern of FMRF-amide immunoreactivity in Acoela
(Platyhelminthes). Tissue & Cell. 30 57±63.
Ruiz-Trillo, I., Riutort, M., Littlewood, T.J., Herniou, E.A. and
Baguna
Á, J. 1999. Acoel flatworms: earliest extant Bilaterian
Metazoans, not members of Platyhelminthes. Science, 283,
1919±1923.
Shaw, C., Maule, A.G. and Halton, D.W. 1996. Platyhelminth
FMRFamide-related Peptides. Int. J. Parasitol., 26, 335±345.
Smith, J.P.S. Tyler, S. and Rieger, R. 1986. Is the Turbellaria
polyphyletic. Hydrobiologia, 132, 13±21.
Westblad, E. 1948. Studien uber skandinavische Turbellaria Acoela.
Arkiv Zoologi, 41A. 1±82.
128 REUTER ET AL.
