Integrative and Comparative Biology
Integrative and Comparative Biology, volume 53, number 1, pp. 27–38
doi:10.1093/icb/ict050
Society for Integrative and Comparative Biology
SYMPOSIUM
C-opsin Expressing Photoreceptors in Echinoderms
Esther M. Ullrich-Lüter,*,† Salvatore D’Aniello‡ and Maria I. Arnone1,‡
*Universität Bonn, Institut für Evolutionsbiologie und Ökologie, An der Immenburg 1, 53121 Bonn, Germany; †Museum
für Naturkunde, Invalidenstr. 43, 10115 Berlin, Germany; ‡Stazione Zoologica Anton Dohrn, Cellular and Developmental
Biology, Villa Comunale, 80121 Naples, Italy
From the symposium ‘‘Integrating Genomics with Comparative Vision Research of the Invertebrates’’ presented at the
annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2013 at San Francisco, California.
1
E-mail: [email protected]
Synopsis Today’s progress in molecular analysis and, in particular, the increased availability of genome sequences have
enabled us to investigate photoreceptor cells (PRCs) in organisms that were formerly inaccessible to experimental
manipulation. Our studies of marine non-chordate deuterostomes thus aim to bridge a gap of knowledge regarding
the evolution of deuterostome PRCs prior to the emergence of vertebrates’ eyes. In this contribution, we will show
evidence for expression of a c-opsin photopigment, which, according to our phylogenetic analysis, is closely related to an
assemblage of chordate visual c-opsins. An antibody raised against sea urchins’ c-opsin protein (Sp-Opsin1) recognizes
epitopes in a variety of tissues of different echinoderms. While in sea urchins this c-opsin is expressed in locomotory and
buccal tube feet, spines, pedicellaria, and epidermis, in brittlestars and starfish we found the immuno-reaction to be
located exclusively in cells within the animals’ spines. Structural characteristics of these c-opsinþ PRC types include the
close vicinity/connection to nerve strands and a, so far unexplored, conspicuous association with the animals’ calcite
skeleton, which previously has been hypothesized to play a role in echinoderm photobiology. These features are discussed
within the context of the evolution of photoreceptors in echinoderms and in deuterostomes generally.
Introduction
Research on metazoan eyes and photoreceptor cells
(PRCs) has provided major insights into general
evolutionary mechanisms. However, most of the
attention previously was focused on optical systems
capable of high-resolution vision, in particular those
of terrestrial species (Nilsson 2009). In contrast,
many marine invertebrates have, primarily due to
the lack of available genomic information, remained
largely unexplored. Echinoderms as non-chordate
deuterostome representatives constitute one such
underinvestigated phylum and just recently have
been shown to possess a sophisticated photoreceptor
system (Ullrich-Lüter et al. 2011). Investigating functional molecular components of PRCs and structural
characteristics of echinoderm light-sensing organs,
we could demonstrate that PRCs of the purple sea
urchin, Strongylocentrotus purpuratus, express a rhabdomeric-type opsin protein (Sp-Opsin4) and fulfill
the requirements for functioning in directed phototaxis (Ullrich-Lüter et al. 2011). This finding has
been surprising from different perspectives. First, a
majority of investigated PRCs engaged in the visual
processes of deuterostomes is of the ciliary type,
expressing ciliary-type opsins (c-opsin), for instance
in the cephalochordate, Branchiostoma floridae
(Vopalensky et al. 2012), in the tunicate, Ciona intestinalis (Kusakabe et al. 2001), and in vertebrates
(reviewed by Lamb et al. 2007). In this context, a
microvillar PRC-type expressing a rhabdomeric-type
opsin (r-opsin) and functioning in vision in an early
branching deuterostome raises questions about the
evolutionary ‘‘ancestral’’ function of that PRC-type
in all deuterostomes. Second, the deployment of
the sea urchin skeleton as a directional light-shielding device supports the involvement of skeletal elements in echinoderms’ photoreception, probably an
apomorphic feature of the whole echinoderm
Advanced Access publication May 10, 2013
ß The Author 2013. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved.
For permissions please email: [email protected].
28
phylum. In fact, the use of skeletal components
in echinoderms’ photoreception has been proposed
already by other authors (Hendler and Byrne 1987;
Johnsen 1997; Johnsen and Kier 1999; Aizenberg
et al. 2001). The study of different brittlestars of
the genus Ophiocoma showed that those that display
phototactic behavior possess lens-like structures
within their calcite skeleton that should facilitate
focusing incoming light on a hypothetical optical
plane located within the arms of the animal. Nerve
strands located within the area of maximum focus,
but lacking distinct typical PRC membrane enlargements, have been hypothesized as intrinsically photosensitive neural components, thus mediating a
photonegative behavior. Other authors have suggested a possible involvement of the echinoid skeleton in visual function, based on the optical
properties of the calcite material (Raup 1966).
Under given natural conditions of polarization of
light, different amounts of light might be transmitted
through adjacent skeletal plates, depending on the
axis of the orientation of calcite for each plate.
Raup (1966) pointed out the possibility that, by
analogy with the use of polarized light by honeybees,
echinoids could use such information as the basis for
a crude system of navigation. However, we should
stress that no molecular or structural evidence for a
PRC functioning in neither of those contexts has
been identified so far.
Considering the huge variety of echinoderms’
photobehaviors, which include shade-seeking, diurnal
migrations, collective spawning, phototaxis, and even
spatial vision (Yoshida 1966; Millott 1975; Yerramilli
and Johnsen 2010), it is not surprising that different
PRC-types might be involved in mediating these
differentiated organismic reactions. Accordingly,
S. purpuratus has been shown to express at least six
different opsin proteins as well as a whole repertoire
of mammalian retinal gene orthologs (Burke et al.
2006; Raible et al. 2006; Agca et al. 2011; Lesser et al.
2011; Ullrich-Lüter et al. 2011). Here, we report our
most recent findings on the expression of a ciliary-type
opsin photopigment (Sp-Opsin1) associated with skeletal elements in the purple sea urchin, S. purpuratus,
and in different species of sea stars and brittlestars.
Materials and methods
Animal care and dissection of adults’ tissues
Strongylocentrotus purpuratus: adult animals were collected by Patrick Leahy, Kerchoff Marine Laboratory
(California Institute of Technology, Pasadena, CA,
USA) from a location at 25–30 m depth in San
Diego Bay (San Diego, CA, USA). The animals
E. M. Ullrich-Lüter et al.
were maintained for a week in an open tank
system and then shipped to the SZN in Naples,
Italy. After arrival they were housed in sea water
aquaria at 128C under 12/12 h day/night light cycle.
Asterias rubens and Amphiura filiformis: Juveniles
and adults were collected at the Sven Lovén Centre
for Marine Sciences in Kristineberg, Sweden. The
material was freshly fixed and stored in phosphatebuffered saline (PBS) with sodium azide.
Ophiocoma nigra: adult animals were collected at
Concarneau Marine Station France, transferred to
the Free University Berlin, Germany, and housed in
sea water aquaria at 158C under 12/12 h day/night
light cycle.
A variety of external appendages were dissected
for immunohistology and in situ hybridization. After
dissection under a stereomicroscope using micro-surgical tools, tissues were immediately placed into the
respective fixatives (see description of methods below).
Production and purification of antibodies
Purified polyclonal antibodies against the c-terminal
tail (residues 314–461) of Sp-Opsin1 were developed
upon request by PRIMM (www.primm.it). Recombinant proteins were produced as His-Tag fusions.
Immunohistochemistry
Prior to some experiments in immunostaining, live
animals were anesthetized and their muscles relaxed
with 7% MgCl in a 1:1 mixture of FSW and distilled
water. Specimens or dissected tissues were directly
transferred to the fixative. Tissues from adults as
well as whole juvenile specimens were fixed in 4%
paraformaldehyde in filtered sea water (FSW) or PBS
(0.05 M PB/0.3 M NaCl, pH 7.4) at room temperature (RT). Fixation times varied among different
treatments with antibodies; all stainings involving
anti-Synaptotagmin B (1E11; Nakajima et al. 2004)
required a short fixation time of 15 min and subsequent post-fixation with ice-cold methanol for 1
min; tissues prepared for other antibody treatments
were fixed for 30–60 min at RT. If necessary, decalcification was obtained by treatment with 2% ascorbic acid/0.15 M NaCl (after initial fixation) for 2–6
days on a slow rotator at RT. Fixed specimens were
rinsed in PBS and blocked in 0.25% bovine serum
albumin containing 0.1% triton X-100 and 0.05%
NaN3 for 30 min at RT. Anti-Synaptotagmin B,
anti-acetylated -tubulin (SIGMA Chemie GmbH,
Munich, Germany), and anti-Sp-Opsin1 were diluted
in PBS with final concentrations of 1:200, 1:250, and
1:50, respectively. After overnight incubation at 48C,
tissues were rinsed in PBS and then incubated in a
29
C-opsin expression in echinoderms
1:500 dilution of Alexa Fluor488-conjugated goat
anti-rabbit IgG and Alexa Fluor568-conjugated goat
anti-mouse IgG (Molecular Probes) for 2 h at RT.
Immunostaining of juvenile sea urchins both with
anti-Sp-Opsin1 and anti-SpOpsin4 antibodies was
performed in two steps. Incubation of antiSpOpsin4 in a 1:50 dilution was followed by several
PBS washes and incubation of secondary antibody
Alexa Fluor555-conjugated goat anti-rabbit IgG in a
1:500 dilution. After several washes in PBS, the specimens/tissues were post-fixed for 10 min in 2% paraformaldehyde in PBS. Afterwards, a second round of
incubation of primary antibody namely anti-acetylated -tubulin (SIGMA) and anti-Sp-Opsin1 diluted
in PBS with final concentrations of 1:250 and 1:50
(overnight at 48C) followed. Specimens were rinsed
in PBS and then incubated in a 1:500 dilution of
Alexa Fluor488-conjugated goat anti-rabbit IgG and
Alexa Fluor633-conjugated goat anti-mouse IgG
(Molecular Probes) for 2 h at RT. Phalloidin labeling
was performed in a one-step method deploying Alexa
Fluor488-conjugated phalloidin (Molecular Probes).
After several washes in PBS, specimens were
mounted in an anti-fading medium containing a
mixture of glycerin and PBS and then examined
using a Leica TCS-SPE or a Zeiss 510Meta confocal
microscope. Projections shown in this study
were produced by recording confocal image stacks
and projecting them in the z-axis using
MacBiophotonics ImageJ or Fiji.
Transmission electron microscopy
Tissues were fixed in 2.5% glutaraldehyde in PBS at
48C for 1 h and subsequently rinsed in PBS several
times. After post-fixation in 1% OsO4 in PBS, specimens were dehydrated in a graded acetone series and
embedded in plastic resin (Araldite M, Fluka, art. no.
10951) for 48 h at 608C. Serial sections were cut
using a diamond knife and a Reichert ultramicrotome S and automatically stained with 2.0% uranyl
acetate for 15 s and lead citrate after Reynolds (1963)
for 20 s in a Phoenix ultrastain (Phoenix staining
technologies, Berlin). Examination was carried out
using a Philips CM 120 Biotwin transmission electron microscopy (TEM). Three-dimensional reconstruction of PRCs and surrounding tissues was
performed using Photoshop CS2 for alignment of
the TEM serial section photographs and Amira 2.0
for subsequent 3D surface rendering.
Whole-mount in situ hybridization
Both antisense- and sense-digoxigenin-labeled RNA
probes were prepared and labeled as described by
Ullrich-Lüter et al. (2011). Fragments of Sp-opsin1
were amplified from genomic DNA by PCR, using
specific primers (Op1IL: GCACTGAGCAGAT
TGTGATCTT/Op1IR: AAGATGATGATGCCGAAA
GG). Probes were labeled using DIG-RNA labeling
KitÕ (Roche). Whole-mount in situ hybridization
(WMISH) of S. purpuratus pedicellaria was performed according to Ullrich-Lüter et al. (2011).
Accuracy of the WMISH results was confirmed by
(negative) control experiments using exactly the
same procedure except deploying sense probes
instead of antisense probes.
Phylogenetic analysis
Opsin proteins were aligned using MEGA5 (Tamura
et al. 2011) and the sequence dataset was truncated
at the extremities to avoid unreliable aligned regions.
All sequences upstream of residue T66 and downstream of residue K343 of the S. purpuratus
Ospin1, taken as a reference protein, were excluded.
The GTR þ G model was the best-fit model of protein evolution for our opsin sequence dataset, according to Feuda et al. (2012). We performed an
unrooted Bayesian analysis with MrBayes v3.1.2
(Ronquist and Huelsenbeck 2003): four independent
runs of 2 000 000 generations were performed, saving
trees every 100 generations and reaching a standard
deviation value 50.01. All sequences used in the phylogenetic reconstruction are listed in Supplementary
Table S1, which provides the name, organism, accession number, and one related literature reference for
each listed sequence (when available).
Results
Phylogenetic relationships of sea urchins’ C-opsins
A Bayesian analysis was used to identify the phylogenetic position of the S. purpuratus ciliary opsin
(Sp-Opsin1). Since the WAG model resulted to be
not adequate for opsin alignment, as recently
reported in Feuda et al. (2012), we used the
GTR þ G model. Sp-Opsin1, together with the encephalopsin protein previously isolated from another
species of sea urchin Hemicentrotus pulcherrimus
(Ooka et al. 2010), branch basally to a subfamily of
chordate c-opsins, including vertebrates’ rhodopsins
and pinopsins (Fig. 1). Noteworthy, sea urchin
c-opsins are the only opsin representatives of the
phylum Ambulacraria in our analysis. To be able to
better resolve the c-opsin phylogeny in deuterostomes, it will be essential to add sequences of representative hemichordates, Ptychodera flava and
Saccoglossus kowalevskii. However, our current analysis does not include sequences from these genomes
30
as no validated opsin sequences are yet available in
the literature. Regarding the phylogeny of deuterostome c-opsins, our analysis is partly in accordance
with the recent analyses of Porter et al. (2012) in
presenting two distinct groups of deuterostome
c-opsins, one that gave rise to the chordate visual
opsins and another one including Branchiostoma
c-opsins, teleost multiple tissue opsins, and encephalopsins (Fig. 1). In opposition to the analysis of
Porter et al. (2012), sea urchin c-opsins in our analysis are not nested within the sistergroup of chordate
visual and non-visual opsins but instead are more
closely related to chordate visual c-opsins than to
the formerly mentioned group (Fig. 1). Our phylogeny also shows similarities to previous ones of
Plachetzki et al. (2007, 2010), who found the same
two groups of chordate opsins as in the abovementioned study, but, as Plachetzki and collaborators
did not include data on sea urchins in their matrices,
no further conclusions on c-opsin evolution can be
drawn from this comparison.
E. M. Ullrich-Lüter et al.
Expression of C-opsin in different tissues of sea
urchins
A combined approach using WMISH to detect Spopsin1 mRNA and immunohistochemistry to detect
localization of Sp-Opsin1 protein was applied to
identify cells expressing c-opsins in the sea urchin,
S. purpuratus. Examination of c-opsin expression
(Sp-Opsin1) in whole mounts of juveniles as well
as in different tissues of adults showed that the
photopigment is present in a variety of the animals’
organs. Those include locomotory and buccal tube
feet, spines, pedicellaria, and portions of the epidermis of the body wall. In the tissues of adults, the
most prominent Sp-opsin1 mRNA expression is
observed in the so-called tridentate pedicellaria
(Fig. 2A). These organs are composed of a threeclawed head region, followed by a neck and the
stalk (Fig. 2B). The head and stalk are supported
by calcite skeletal elements, whereas the neck is free
of calcite and thus is contractile and highly mobile.
Fig. 1 Unrooted phylogenetic reconstruction of bilaterian opsins generated by Bayesian inference method supports Sp-Ospin1
orthology assignment to c-opsin group. The opsins of Strongylocentrotus purpuratus are in bold face.
C-opsin expression in echinoderms
This type of echinoid pedicellaria is involved in the
defense and feeding mechanisms of the animal
(Campbell and Laverick 1968; Ghyoot et al. 1987)
and has been demonstrated to host many sensory
cells (Peters and Campbell 1987). To verify the specificity of our antibody, designed to be against the
sea urchin c-opsin photopigment of S. purpuratus
31
(Sp-Opsin1), double labeling via immunostainings
and conventional in situ hybridization was performed. We could thus demonstrate co-localization
of Sp-opsin1 RNA and protein in ciliated cells of the
stalk epidermis of tridentate pedicellaria (Fig. 2Ai
and ii). In addition, Sp-Opsin1 is also expressed in
epidermal cells of buccal tube feet, with some of
Fig. 2 Sea urchin c-opsin (Sp-opsin1) RNA and protein expression in adult tissues of S. purpuratus. (A) Sp-opsin1 RNA expressed in
proximal stalk region of tridentate pedicellaria (in purple). (Ai) Sp-Opsin1 protein in ciliated cells of the proximal stalk region (hot red).
(Aii) Co-labeling via in situ hybridization and anti-Sp-Opsin1 immunostaining of cells in the proximal stalk. (B) Tridentate pedicellaria
composed of head-region (h), neck-region (n), and stalk-region (st) with skeletal claws (cl) in the head and supporting skeleton (ssk) in
the stalk. Note the relaxed neck region in comparison to (A). (C) Expression of sea urchin c-opsin protein in the head and neck region.
Nerve cells (ns) of the pedicellaria immunolabeled via anti-acetylated -tubulin (green) accompany the Sp-Opsin1þ cells in the neck
region and within fractions of the head region (overlay of both immunostainings in yellow). (D) Double immunolabeling against
Sp-Opsin1 and acetylated -tubulin in buccal tube feet shows presence of the c-opsin in large portions of the tube-foot disc (purple)
and in some specialized cilia (overlay of both stainings in white) (ci ¼ cilia). (E) Sp-opsin1 RNA expressing cells in tridentate pedicellarian
head region accompanied by nerve cells (green). (F) Identical region of another specimen showing presence of Sp-Opsin1 protein in
fiber-like cells spanning between the head ossicles.
32
these cells showing the presence of c-opsin protein in
morphologically modified external cilia, which are
co-labeled with anti-acetylated -tubulin (Fig. 2D,
co-labeling with anti-Sp-Opsin1 and anti-acetylated
-tubulin in white).
Irrespective of the findings on c-opsin expression
in ciliated sensory cells, the presence of Sp-Opsin1
positive (þ) cells in sea urchins is not limited to a
single morphologically distinct cell type. Although
conventional in situ hybridization had confirmed
Sp-opsin1 expression in the proximal stalk region of
pedicellariae, the use of more sensitive double-fluorescent in situ hybridization, as well as immunolabeling against Sp-Opsin1, has revealed the presence of
additional Sp-Opsin1þ cells in the pedicellarian head
and neck region (Fig. 2C, E, and F). The c-opsinþ
cells there show a fiber-like morphology. It has to be
noted that pedicellaria of other echinoid species have
been shown to possess a neuropil within each valve
of the pedicellarian head. This neuropil receives
input from primary sensory cells and, in turn, supplies the stalk with up to 12 longitudinal nerves
(Peters and Campbell 1987; Ghyoot and Jangoux
1988; reviewed by Cavey and Märkel 1994). The
Sp-Opsin1þ cells in the pedicellarian head are arranged as a compact tissue associated with the
valve ossicles, in the same region of the described
neuropil (Fig. 2C: overlay in yellow). The presence
of neuronal projections from the neuropil through
the pedicellarian neck in S. purpuratus corroborates
the organization of nerve tissue described by the
above-mentioned authors (Fig. 2C). The fiber-like
Sp-Opsin1þ cells in the neck which seemingly connect the head ossicles with the stalk ossicle (Fig. 2C)
are accompanying those nerve strands, although the
analysis of confocal z-stack images clearly show that
the Sp-Opsin1þ cells are located more internally.
Immunohistochemical attempts to identify the
tissue type expressing the sea urchin ciliary opsin
(e.g., phalloidin staining to detect f-actin as an
essential component of muscle-cell fibers) failed to
label Sp-Opsin1þ cells.
A general feature of the investigated c-opsinþ cell
types in S. purpuratus is their association with skeletal
elements of the animal. Using reflection microscopy,
we found this association in pedicellaria, developing
spines and buccal tube feet, locomotory tube-foot
discs (Fig. 3A–C), and the aboral body surface
(Fig. 3B). Some c-opsinþ cells have a compact cell
body placed between the trabeculae of the stereom
(mesh-like arrangement of echinoderms’ calcite skeleton) and are positioned in close vicinity to the nerve
strands of the spine epidermis, the tube-foot disc,
and the body surface (see arrows in Fig. 3Ai and Bii).
E. M. Ullrich-Lüter et al.
In locomotory tube feet, small aggregations of ciliated,
slender cells with axonal projections connecting to the
nervous system are the only candidates for the expression of c-opsin in the outer disc region (Fig. 4A–D).
This cell type has been hypothesized to be a sensory cell
before, but their function could not be determined
(Agca et al. 2011; Ullrich-Lüter et al. 2011).
Additionally to the formerly described cell types, another Sp-Opsin1þ fiber-shaped cell type, could be detected in spines, tube feet, and pedicellaria. Those
fiber-shaped cells are also associated with skeletal elements either by direct connections to each other or by
closely accompanying such connective elements. The
fiber-like Sp-Opsin1þ cells encircle the spine bases and
connect the proximal part of the spine via the ball and
socket joint (see Millott 1966; Stauber and Märkel
1988) to the body wall (Fig. 3Bii and iii). In the
region of the tube-foot disc, they run from the skeletal
rosettes toward the stalk of the tube foot, and within
the stalk run parallel with its lumen toward the body
wall of the animal (Fig. 3C). This fiber-like cell type is
also present in the head and neck of the formerly mentioned pedicellaria of S. purpuratus (Fig. 2C and F).
Finally, immunostaining of dissected S. purpuratus
radial-nerve tissue revealed the presence of
Sp-Opsin1þ cells within the nerve strand itself. The
majority of Sp-Opsin1þ cells in the radial nerve are
concentrated within certain areas although some cells
or cell processes occupy longer portions of the radial
nerve (Fig. 3D). A scheme displaying the position,
relative to the sea urchin’s nervous system, of putative
c-opsin expressing (Sp-Opsin1þ) photoreceptors
and of the previously described r-opsin positive
(Sp-Opsin4þ) PRC is reported in Fig. 4E.
Localization of C-opsin protein in ophiuroid
and asteroid spines
Although in sea urchins the ciliary opsin Sp-Opsin1
is expressed in a variety of tissues and cell types, our
studies on starfish and brittlestars show that in the
species examined, expression of this photopigment
seems to be restricted to the animals’ spines. AntiSp-Opsin1 immunostainings on decalcified whole
mounts of adult A. filiformis (brittlestar) show that
the Sp-Opsin1þ cells are located in the proximal
region of the spines, their morphology resembling
that of sea urchin spine PRCs (Fig. 5A and B). The
Sp-Opsin1þ cells lie in close vicinity to the nervous
system as shown by immunolabeling against acetylated -tubulin (see inset in Fig. 5B). By contrast, in
the brittlestar O. nigra, the Sp-Opsin1þ epidermal
cells are equally distributed over the entire dermis
of the spine (Fig. 5C and D), with the highest concentration of protein in the apical region of the cell.
C-opsin expression in echinoderms
33
Fig. 3 Sea urchin c-opsin protein in whole mounts of juveniles and in the tube feet and radial nerves of adults. Immunostainings and
reflection microscopy: c-opsin (red/hot red), r-opsin (blue), acetylated -tubulin (green), skeleton (white). (A) Juvenile, oral view.
C-opsin protein present in primary spines (psp), spines (sp), primary tube feet (ptf), buccal tube feet (btf), and epidermis. Reflection
microscopy reveals c-opsin cells to be embedded between skeletal trabeculae. (Ai) C-opsinþ cells lying close to nerve strands
(arrows). (Aii) C-opsinþ cells in the region of the spine connecting to its base and in buccal tube feet. (Aiii) C-opsinþ cells embedded
between skeletal rosettes of primary tube foot. (B) Juvenile aboral view. C-opsinþ cells in primary spines, spines, pedicellariae, and
epidermis. (Bi) C-opsinþ cells embedded between skeletal trabeculae of primary spine. (Bii) C-opsinþ cells lying close to nerve strands
(arrow). (Biii) The c-opsinþ cells within the region of the base of the spine have a fiber-like morphology. (C) C-opsinþ cells in the
locomotory tube foot of an adult. The cells run from the skeletal rosette (ros) in the disc (ds) toward the stalk (st) in a triangular shape
and within the stalk, accompanied by nerve strands (ns), parallel to its lumen. (D) C-opsin protein in the radial nerve (rn) of an adult.
The c-opsinþ cells cluster in distinct regions but extensions run along the nerve strand (ci ¼ cilia of the epineural sinus).
High-magnification images reveal an immunoreactive cell extension, presumably of a ciliary type
(Fig. 5D inset). However, anti-acetylated -tubulin
immunolabeling failed to co-stain these SpOpsin1þ cell extensions. The Sp-Opsin1þ PRCs in
O. nigra contacts the nervous system via axonal projections into single nerve strands (Fig. 5D).
In the starfish A. rubens, the pattern of Sp-Opsin1
protein resembles that of the brittlestar O. nigra. The
Sp-Opsin1þ cells in both species are distributed over
34
E. M. Ullrich-Lüter et al.
Fig. 4 Expression c-opsin protein and the ultrastructure of c-opsin candidate cells in locomotory tube feet of adult S. purpuratus.
Scheme on spatial distribution of c-opsin versus r-opsin expression in adult S. purpuratus. (A) C-opsin protein expression (red).
Tube foot in top view. Slim cells run from the rim of the disc ‘‘hillock’’ towards tube-foot ring nerve that also shows immunopositive
reaction to anti-Sp-Opsin1 (red). (B) Detail of A (square), showing the slender morphology of two single c-opsinþ photoreceptors.
(C) TEM serial section of the rim of the disc of the tube foot with c-opsin expression in candidate sensory cells. Several of the slender
ciliated cells (arrowheads) are associated with a rhabdomeric photoreceptor (black arrow). (D) 3D reconstruction of TEM serial
sections. The slender green cells represent c-opsinþ candidate sensory cells associated with two rhabdomeric photoreceptors (purple
and violet), a portion of a skeletal rosette (gray) and nerve tissue (yellow). (E) Scheme displaying the spatial distribution of photoreceptors expressing c-opsin (yellow) and r-opsin (red) relative to the nervous system (green). Bns, basiepithelial nervous system; btf,
buccal tube foot; btfn, btf nerve; onr, oral nerve ring; ped, pedicellaria; radn, radial nerve; ske, skeleton; sp, spine; tf, tube foot; tfn, tubefoot nerve.
C-opsin expression in echinoderms
35
Fig. 5 C-opsin protein in brittlestars and starfish. Immunostainings: c-opsin (hot red), acetylated -tubulin (green). (A and B) Adult,
decalcified A. filiformis. (A) Arm region (ar) with c-opsinþ cells in spines (sp). (B) Higher magnification reveals c-opsinþ cells in the
proximal portion of the spine and close association of the cells with nerves in the spine (ns) (inset). (C and D) Adult, decalcified
O. nigra. (C) Portion of an arm showing spines and a tube foot (tf). C-opsinþ cells are dispersed over the epidermis of the spine.
(D) C-opsin protein is present in the whole c-opsinþ cell body but most dominantly within apical extensions of the cells, presumably of
a ciliary type (inset). Again, c-opsinþ cells show close association with nerve strands in the spine. (E and F) Juvenile A. rubens. (E) Distal
portion of an arm showing c-opsinþ cells in spines. Tube feet and terminal tentacle (tt) with optic cushions show no immunoreactivity
against sea urchin c-opsin. The c-opsinþ cells are distributed over the epidermis of the spine. (F) C-opsin protein in extensions of cells
intermingled between other cilia (ci) of the spine’s epidermis.
the entire epidermis of the spine with no indication
of other tissues expressing this photopigment
(Fig. 5E and F).
Discussion
Since it was known from analysis of the S. purpuratus
genome (Sodergren et al. 2006) and from different
expression analyses (Burke et al. 2006; Raible et al.
2006) that this echinoid species possesses more than
one expressed opsin protein, it could be assumed
that the animals also possess more than one PRCtype. The potential function of those different photoreceptor systems, however, remained largely
enigmatic. Besides the involvement of r-opsin PRCs
in directed phototaxis (Yoshida and Ohtsuki 1966,
1968; Ullrich-Lüter et al. 2011), no other function
has been attributed to any certain type of opsin/
PRC. With the aim of finding out more about the
evolutionary origin and potential function of sea
urchin c-opsin, we performed a phylogenetic analysis
using a large set of bilaterian opsins. The resulting
tree shows that the sea urchin c-opsins (Sp-Opsin1
and H. pulcherrimus encephalopsin) constitute a
sistergroup to the group that gave rise to the chordate visual opsins. This group includes rhodopsins
with visual function in rods and cones as well as,
for example, pineal opsins, some of them expressed
in photoreceptors mediating non-retinal visual
behaviors such as shadow responses in Xenopus and
tunicate larvae (Roberts 1978; Korf et al. 1989;
Kusakabe et al. 2001). According to our analysis,
sea urchins’ c-opsins are clearly separated from different groups of c-opsins involved in non-visual photoreception, like the one including Homo
encephalopsin (Blackshaw and Snyder 1999) and another one including Platynereis C-opsin (Arendt
et al. 2004). Since sea urchin c-opsins form a sistergroup to chordate visual opsins (and few with
36
still unproven function), our analysis suggests a potential visual function, sensu lato, for this type of
echinoderm opsin.
Our initial studies on spatial and temporal expression of Sp-Opsin1-expressing cells in S. purpuratus
yielded a seemingly overall organization (Fig. 3A and
B) comparable to the formerly described system of
distinct r-opsinþ PRCs (Ullrich-Lüter et al. 2011).
One major difference concerns the levels of expression of c-opsin, which in former quantitative analyses (Burke et al 2006; Raible et al 2006) were
demonstrated to be 10-fold lower (e.g., in tube
feet), than those of sea urchin r-opsin. As a consequence, amplification in RNA/protein detection
signal had to be boosted by modification of the experimental procedures and the recording of the data,
thus inevitably resulting in a lower resolution of the
images. One striking structural characteristic of the
sea urchin c-opsinþ PRCs is their close association
with the calcite skeleton (in white in Fig. 3). Many
Sp-Opsin1þ cells have their cell bodies embedded
within pores of the calcite skeleton of the tube feet,
pedicellaria, or test (body wall). This conspicuous
association is also evident in other echinoderms,
for instance the two investigated brittlestars A. filiformis and O. nigra and the starfish A. rubens.
However, in these echinoderms, the Sp-Opsin1 immunoreactive cells could be found exclusively within
the animals’ spines. The immunolabeling in spines of
A. rubens corresponds to the staining obtained by
Johnsen (1997) with bovine rhodopsin antibodies
in the tips of spines in Asterias forbesi. In
A. rubens, the c-opsin immunopositive PRCs seem
to be located within the epidermis of the spine,
whereas in A. filiformis at least the cell bodies lie
deeper within the tissue of the spine. Many ophiuroids possess spines that show a ‘‘conspicuous
hollow’’ (Byrne 1994) with the central space occupied by the axial nerve. The documented nerve
strands and the c-opsinþ cells in A. filiformis
might well correspond to a structure positioned
inside the cylinder of the calcite spine. For echinoids
(that also possess hollow spines in many species), it
has been demonstrated that the long anatomical axis
of the spine and the optical axis (c-axis) coincide
(Raup 1966). PRCs embedded within those spines,
as in the ophiuroid A. filiformis, may be reached
by polarized light transmitted through the ossicle
of the spine. This is reminiscent of the results obtained by Aizenberg et al. (2001) who (i) propose a
light-focusing function for specialized structures of
the ophiuroid skeleton and (ii) assume that the hypothesized PRCs mediate a shadow response in
Ophiocoma wendtii. Since A. filiformis and O. wendtii
E. M. Ullrich-Lüter et al.
exhibit a completely different lifestyle, the absence
of c-opsin protein from the ossicles of the arms of
A. filiformis does not exclude the possibility that this
PRC-type might function in association with the
ossicles of the arms of O. wendtii.
The expression pattern of Sp-Opsin1þ PRCs in
S. purpuratus, widely dispersed over the majority of
the body wall, makes them ideal candidates for the
PRC-type mediating the long-proposed dermal sense
of light in echinoderms (for review, see Yoshida
1966; Millott 1975). This dermal light-sense was hypothesized on the basis of experiments showing that
a large portion of sea urchins’ epidermis is sensitive
(to different degrees) to changes in light intensity. A
prominent photoreaction attributed to that dermal
light-sense and best investigated in diadematid sea
urchins (for review, see Yoshida 1966; Millot 1975)
is the so-called shadow reflex, a rapid jerking of the
spine occurring upon the sudden shading of the
animal. As the photoreceptors themselves in those
previous studies were not known, conclusions from
those experiments often remained unsatisfying and
ambiguous. However, it became clear that epidermal
PRCs, in conjunction with the integrative nervoussystem centers of the animals, the so-called radial
nerves, were involved in the shadow reflex (see also
Yoshida and Millott 1959). Our experiments show
expression of the c-opsin photopigment within a distinct receptor cell type of the sea urchins’ epidermis
and within the radial nerve of the animals. Another
cell type of S. purpuratus, which has a fiber-like morphology, also expresses the c-opsin. Those fiber-like
c-opsinþ cells seem to correspond, spatially, to adductor and abductor muscles within the pedicellarian
valves and to longitudinal flexor muscles in the necks
of pedicellaria (scheme by M. Stauber, cited by Cavey
and Märkel 1994). Further investigations will have to
be carried out in order to clarify whether this type of
opsin might be co-expressed in sea urchins’ muscle
cells, a finding that might account for some of the
contradictory results on affector and effector reactions obtained in former physiological studies (for
review, see Yoshida 1966), or if sea urchins’ copsin is expressed in nerve cells accompanying
those muscle cells.
Compared with sea urchins’ photoreceptor system
expressing a rhabdomeric opsin (Ullrich-Lüter et al.
2011), the c-opsin expressing photoreceptors are not
likely to be involved in directional vision. The
r-opsin PRCs form distinct aggregations of reasonable cell numbers at the base of the animals’ tube
feet and are embedded in a skeletal invagination providing a partial shielding from light. The photoreceptors integrate into the animals’ internal, ring-like
C-opsin expression in echinoderms
interconnected nervous system, which finally enables
them to detect the direction of incoming light.
Contrasting to those r-opsin PRC clusters, c-opsin
is expressed in many different types of sea urchins’
tissues and thus in sum dispersed over the entire
body surface. We have no indication that cells expressing c-opsins in echinoids have any kind of lightshielding mechanism that might enable them to
distinguish the direction of a light stimulus.
Additionally, spatial distribution and form of these
cells do not support a function in directional detection of light. It is thus reasonable to speculate that
the two types of sea urchin opsins are involved in
different photoreceptive contexts. Although r-opsin
PRCs serve directional vision, c-opsin PRCs might
be involved in light detection not relying on directional information such as in the above-mentioned
shadow reflex.
Supplementary Data
Supplementary Data available at ICB online.
Acknowledgments
We thank Patrick Leahy and the Southern California
Sea Urchin Company (USA) for constant supply of
S. purpuratus; Sam Dupont for invaluable assistance
in collecting A. rubens, O. nigra, and A. filiformis;
Remo Sanges (SZN) for assistance with bioinformatic
analysis, Giovanna Benvenuto (SZN) for support
with confocal microscopy, and anonymous reviewers
for useful comments and suggestions.
Funding
This work was supported by German Research
Foundation (Deutsche Forschungsgemeinschaft)
Grant (HA 4443/5-1) and in part by Marie Curie
Research Training Network Zoonet MRTN-CT2004-005624 and Marie Curie Initial Training
Network Neptune PITN-GA-2012-317172.
References
Agca C, Elhajj MC, Klein WH, Venuti JM. 2011.
Neurosensory and neuromuscular organization in tube
feet of the sea urchin Strongylocentrotus purpuratus.
J Comp Neurol 519:3566–79.
Aizenberg J, Tkachenko A, Weiner S, Addadi L, Hendler G.
2001. Calcitic microlenses as part of the photoreceptor
system in brittlestars. Nature 412:819–22.
Arendt D, Tessmar-Raible K, Snyman H, Dorresteijn AW,
Wittbrodt J. 2004. Ciliary photoreceptors with a vertebrate-type opsin in an invertebrate brain. Science
306:869–71.
Blackshaw SJ, Snyder S. 1999. Encephalopsin: a novel mammalian extraretinal opsin discretely localized in the brain. J
Neurosci 19:3681–90.
37
Burke RD, Angerer LM, Elphick MR, Humphrey GW,
Yaguchi S, Kiyama T, Liang S, Mu X, Agca C, Klein WH,
et al. 2006. A genomic view of the sea urchin nervous
system. Dev Biol 300:434–60.
Byrne M. 1994. Ophiuroidea. In: Harrison FW, Chia FS, editors. Microscopic anatomy of invertebrates. Vol. 14.
Echinodermata. New York: Wiley-Liss. p. 274–343.
Campbell AC, Laverick MS. 1968. The responses of pedicellaria from Echinus esculentus. J Exp Mar Biol Ecol
2:191–214.
Cavey MJ, Märkel K. 1994. Echinoidea. In: Harrison FW,
Chia FS, editors. Microscopic anatomy of invertebrates.
Vol. 14. Echinodermata. New York: Wiley-Liss. p. 345–400.
Feuda R, Hamilton SC, McInerney JO, Pisani D. 2012.
Metazoan opsin evolution reveals a simple route to
animal vision. Proc Natl Acad Sci USA 109:18868–72.
Ghyoot M, Jangoux M. 1988. The fine structure of the
globiferous pedicellariae of Shpherechinus granularis
(Echinoidea): innervation of the stalk. In: Burke RD,
Mladenov PV, Lambert P, Parsley RL, editors.
Echinoderm biology. Rotterdam: Balkema. p. 713–7.
Ghyoot M, De Ridder C, Jangoux M. 1987. Fine structure and
presumed functions of the pedicellariae of Echinocardium
cordatum (Echinodermata, Echinoida). Zoomorphology
106:279–88.
Hendler G, Byrne M. 1987. Fine structure of the dorsal arm
plate of Ophiocoma wendti: evidence for a photoreceptor
system (Echinodermata, Ophiuroidea). Zoomorphology
107:261–72.
Johnsen S. 1997. Identification and localization of a possible
rhodopsin in the echinoderms Asterias forbesi (Asteroidea)
and Ophioderma brevispinum (Ophiuroidea). Biol Bull
193:97–105.
Johnsen S, Kier WM. 1999. Shade-seeking behaviour under
polarized light by the brittle star Ophioderma brevispinum
(Echinodermata: Ophiuroidea). J Mar Biol Assoc UK
9:761–3.
Lamb TD, Collin SP, Pugh EN. 2007. Evolution of the vertebrate eye: opsins, photoreceptors, retina and eye cup.
Nat Rev Neurosci 8:960–76.
Lesser MP, Carleton KL, Böttger SA, Barry TM, Walker CW.
2011. Sea urchin tube feet are photosensory organs that
express a rhabdomeric-like opsin and PAX6. Proc Biol Sci
278:3371–9.
Korf B, Rollag MD, Korf HW. 1989. Ontogenetic development of S-antigen- and rodopsin immunoreactions in retinal and pineal photoreceptors of Xenopus laevis in relation
to the onset of melatonin-dependent color-change mechanisms. Cell Tissue Res 258:319–29.
Kusakabe T, Kusakabe R, Kawakami I, Satou Y, Satoh N,
Tsuda M. 2001. Ci-opsin1, a vertebrate type ospin gene,
expressed in the larval ocellus of the ascidian Ciona intestinalis. FEBS Lett 506:68–72.
Millott N. 1966. Co-ordination of spine movement in echinoids. In: Boolootian RA, editor. Physiology of echinodermata. New York: Interscience. p. 465–85.
Millott N. 1975. The photosensitivity of echinoids. Adv Mar
Biol 13:1–52.
Nakajima Y, Kaneko H, Murray G, Burke RD. 2004.
Divergent patterns of neural development in larval echinoids and asteroids. Evol Dev 6:95–104.
38
Nilsson DE. 2009. The evolution of eyes and visually guided
behaviour. Philos Trans R Soc Lond B Biol Sci 364:2833–47.
Ooka S, Katow T, Yaguchi S, Yaguchi J, Katow H. 2010.
Spatiotemporal expression pattern of an encephalopsin
orthologue of the sea urchin Hemicentrotus pulcherrimus
during early development, and its potential role in larval
vertical migration. Dev Growth Differ 52:195–207.
Peters BH, Campbell AC. 1987. Morphology of the nervous
and muscular systems in the heads of pedicellariae from the
sea urchin Echinus esculentus L. J Morphol 193:35–51.
Plachetzki DC, Degnan BM, Oakley TH. 2007. The origins of
novel protein interactions during animal opsin evolution.
PLoS One 10:e1054.
Plachetzki DC, Fong CR, Oakley TH. 2010. The evolution of
phototransduction from an ancestral cyclic nucleotide gated
pathway. Proc R Soc B 277:1963–9.
Porter M, Blasic JR, Bok MJ, Cameron EG, Pringle T,
Cronin TW, Robinson PR. 2012. Shedding new light on
opsin evolution. Proc R Soc B 279:3–14.
Raible F, Tessmar-Raible K, Arboleda E, Kaller T, Bork P,
Arendt D, Arnone MI. 2006. Opsins and clusters of
G-protein coupled receptors in the sea urchin genome.
Dev Biol 300:461–75.
Raup DM. 1966. The endoskeleton. In: Boolootian RA, editor.
Physiology of echinodermata. New York: Interscience.
p. 379–95.
Roberts A. 1978. Pineal eye and behaviour in Xenopus
tadpoles. Nature 273:774–5.
Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics
19:1572–4.
Sodergren E, Weinstock GM, Davidson EH, Cameron RA,
Gibbs RA, Angerer RC, Angerer LM, Arnone MI,
E. M. Ullrich-Lüter et al.
Burgess DR, Burke RD, et al. 2006. The genome of
the sea urchin Strongylocentrotus purpuratus. Science
314:941–52.
Stauber M, Märkel K. 1988. Comparative morphology of
muscle-skeleton attachments in the Echinodermata.
Zoomorphology 108:137–48.
Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S.
2011. MEGA5: molecular evolutionary genetics analysis using
maximum likelihood, evolutionary distance, and maximum
parsimony methods. Mol Biol Evol 28:2731–9.
Ullrich-Lüter E, Dupont S, Arboleda E, Hausen H,
Arnone MI. 2011. A unique system of photoreceptors
in sea urchin tube feet. Proc Natl Acad Sci USA
108:8367–72.
Vopalensky P, Pergner J, Liegertova M, Benito-Gutierrez E,
Arendt D, Kozmik Z. 2012. Molecular analysis of the amphioxus frontal eye unravels the evolutionary origin of the
retina and pigment cells of the vertebrate eye. Proc Natl
Acad Sci USA 109:15383–8.
Yerramilli D, Johnsen S. 2010. Spatial vision in the purple sea
urchin Strongylocentrotus purpuratus (Echinoidea). J Exp
Biol 213:249–55.
Yoshida M. 1966. Photosensitivity. In: Boolootian RA, editor.
Physiology of echinodermata. New York: John Wiley and
Sons. p. 435–64.
Yoshida M, Millott N. 1959. Light sensitive nerve in an
echinoid. Experimentia 15:13–4.
Yoshida M, Ohtsuki H. 1966. Compound ocellus of a starfish:
its function. Science 153:197–8.
Yoshida M, Ohtsuki H. 1968. The phototactic behaviour
of the starfish Asterias amurensis Lütken. Biol Bull
134:516–32.