Aquatic Toxicology 89 (2008) 75–81

Contents lists available at ScienceDirect

Aquatic Toxicology
journal homepage: www.elsevier.com/locate/aquatox

Manganese effects on haematopoietic cells and circulating coelomocytes of

Asterias rubens (Linnaeus)
Carolina Oweson a , Helen Sköld a , Annalisa Pinsino b , Valeria Matranga b , Bodil Hernroth c,∗
a
Department of Marine Ecology, Göteborg University, Kristineberg 566, 45034 Fiskebäckskil, Sweden
b
Istituto di Biomedicina e Immunologia Molecolare “A. Monroy”, Via Ugo La Malfa 153, 90146 Palermo, Italy
c
The Royal Swedish Academy of Sciences, Kristineberg 566, 45034 Fiskebäckskil, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Manganese (Mn) is a naturally abundant metal in marine sediments where it mainly occurs as MnO2 .
Received 18 March 2008 During hypoxic conditions it is converted into a bioavailable state, Mn2+ , and can reach levels that pre-
Received in revised form 26 May 2008 viously have shown effects on immune competent cells of the crustacean, Nephrops norvegicus. Here
Accepted 27 May 2008
we investigated if Mn also affects circulating coelomocytes and their renewal in the common sea star,
Asterias rubens, when exposed to concentrations of Mn that can be found in nature. When the sea stars
Keywords:
were exposed to Mn it accumulated in the coelomic fluid and the number of circulating coelomocytes,
Echinoderm
in contrast to what was recorded in Nephrops, increased significantly. By using the substitute nucleotide,
Immuno-toxicology
Coelomocyte
5-bromo-2 -deoxyuridine, BrdU, for tracing cell division and by recording mitotic index by nuclei stain-
Proliferation ing, we found that Mn induced proliferation of cells from a putative haematopoietic tissue, the coelomic
HSP70 epithelium. In addition, the haematopoietic tissue and coelomocytes showed stress response in terms of
Oxidative stress changes in HSP70 levels and protein carbonyls, as judged by immunohistochemistry and Western blotting.
Phagocytosis Measurement of dehydrogenase activity, using MTS/PMS, revealed that Mn showed cytotoxic properties.
We also found that the phagocytotic capacity of coelomocytes was significantly inhibited by Mn. It was
concluded that the exposure of A. rubens to Mn induced renewal of coelomocytes and impaired their
immune response.
© 2008 Elsevier B.V. All rights reserved.

1. Introduction et al., 1967; Iregren, 1990). Recently, we reported that manganese

Manganese (Mn) is a naturally abundant metal in aquatic sed-
iments where it mainly occurs as MnO2 . At oxygen concentration
below 16% of saturation manganese is converted into its bioavail-
able state, Mn2+ , and can reach toxic levels for benthic biota (Hall et
al., 1996). The main reason for hypoxia in bottom waters of coastal
areas derives from anthropogenic-induced eutrophication, caused
by an increased load of nutrients in concert with cascade effects of
overfishing (Diaz and Rosenberg, 1995; Frank et al., 2005) and can
occur during periods of days to weeks in the bottom water (Baden
et al., 1990; Phil et al., 1991). Mn2+ needs particles to reoxidize and
the bioavailable fraction might therefore stay in the water column
for sometime even after hypoxia.
Manganese is an essential metal as cofactor for mitochondrial
enzymes and for activation of a large number of other enzymes.
Nevertheless, in excess Mn is a strong neurotoxin that can cause the
human disorder Manganism, similar to Parkinson’s disease (Mena
∗ Corresponding author. Tel.: +46 52318513; fax: +46 52318502.
E-mail address: bodil.hernroth@marecol.gu.se (B. Hernroth).
also severely suppresses the immune activities of the Norway
lobster, Nephrops norvegicus (Hernroth et al., 2004) and induces
apoptosis of the lobster haematopoietic stem cells (Oweson et al.,
2006). However, still it has not been proven that the effect of
manganese on haematopoiesis and immunity is general also for
other phyla. Echinoderms are among the most abundant of all ben-
thic animals and may constitute more than 90% of the benthic
biomass in the deep sea. They are widely distributed in all oceans,
at all depths and their role is structurally important (Saier, 2001).
Thus, suppression of their immunity and altered capacity to con-
trol infectious diseases could have severe effects on the benthic
ecosystem.
This study aimed to provide information about how Mn, in con-
centrations that are realistic for soft bottoms during hypoxic events
(Balzer, 1982), affects echinoderms by investigating the common
European sea star, Asterias rubens. The coelomic fluid of A. rubens
possesses large populations of circulating cells, the coelomocytes.
The definitions and the nomenclature of echinoderm coelomocytes
are still not standardized (Smith, 1981; Pinsino et al., 2007). Four
different morphotypes have been described for asterioids, with the
phagocytes (also referred to as amoebocytes), as the dominating

0166-445X/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.aquatox.2008.05.016
76 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81
type, comprising approximately 95% of the population (Pinsino
et al., 2007). These cells undergo a petaloid to filopodial transi-
tion, where the latter type is able to attach to glass and form
network. Other sub-populations have been described as vibratile
cells, haemocytes, and slow-moving cells with an irregular shape,
recently called amoebocytes (Matranga et al., 2006). The coelo-
mocytes are able to efficiently clear bacteria from the coelomic
cavity and in case of injury they take part in wound healing by
migrating to the injured site, prevent bleeding by clothing and
interact with the extracellular matrix during the healing process
(Smith, 1981). The recruitment of circulating coelomocytes is not
fully understood. The coelomic epithelium (coelothelial cells) has
been suggested as one of the most probable potential source of the
coelomocytes of echinoderms (Muñoz-Chápuli et al., 2005; Holm
et al., 2008).
It has been found in recent papers that stresses such as injury
and polluted water affect the immune response in echinoderms
(Matranga et al., 2000; Coteur et al., 2003). Based on a pilot study
where we observed that the coelomocyte numbers and their level
of heat shock protein (HSP70) increased after manganese exposure
we hypothesized that the accumulation of Mn in coelomic fluid
of A. rubens affects proliferation of haematopoietic cells, effecting
the numbers of circulating coelomocytes, their stress response and
their phagocytic capacity.
2. Material and methods
2.1. Animal handling and experimental setup
Specimens of A. rubens were collected by scuba divers from
approximately 5–15 m depth at the mouth of the Gullmar Fjord,
adjacent to Kristineberg Marine Research Station, situated at the
West coast of Sweden. The sea stars were maintained in basins
supplied with running seawater of ambient temperature and
salinity until used for the experiments. Once a week they were
fed with blue mussels, Mytilus edulis. All sea stars used for the
study were 10–14 cm across, from arm tip to most distant arm
tip.
Sea stars were kept in containers with seawater, 3 l per ani-
mal, at constant salinity and temperature, 33 PSU and 12 ◦ C, and
exposed to different concentrations of Mn, achieved by using man-
ganese(II)chloride tetrahydrate (GR, Merck, Germany). First a study
was performed when the sea stars were exposed to 5.5 mg Mn/l.
This study raised questions that made it relevant to increase the
concentration to 15 mg Mn/l, which was comparably to previous
studies on a crustacean (Hernroth et al., 2004; Oweson et al., 2006).
The water, continuously and gently mixed, was exchanged daily and
the sea stars were not fed during the experiment. Sea stars used as
controls were treated in the same way but in seawater without Mn
addition.
2.2. Manganese analysis
Coelomic fluid was collected from sea stars pre-exposed to
5.5 mg Mn/l for 5, 10, 15 and 25 days and from control animals
(n = 8), not exposed to Mn. Manganese concentration was deter-
mined by atomic absorption spectrophotometry using both flame
and graphite furnace (GBC 932 AA). Prior to the metal analysis 1 ml
coelomic fluid was dissolved in an equal volume of nitric acid (65%,
pro analysis, Merck) and digested in a CEM microwave extraction
system (MES 1000). Standards were made up in 20% nitric acid
and samples were diluted to appropriate concentrations. A certi-
fied reference material made of non-defatted lobster midgut gland
(LUTS-1, National Research Council, Canada) was used as internal
standard.
2.3. Body volume
In order to investigate if changes in the total coelomocyte counts
(TCC) could be related to an adjustment of the volume of the
coelomic fluid by the sea stars, Mn effect on the body volume was
investigated. Sea stars were exposed to 15 mg Mn/l for 5 and 10
days (n = 8) and the weight (g) as well as the distance between
the madreporite and the tip of the nearest arm (cm) were deter-
mined before and after the exposure. The mean value of body index
[weight (g) × 1000/length (cm)3 ] was calculated.
2.4. Total coelomocyte counts (TCC)
Sea stars were exposed to 15 mg Mn/l for 5 and 10 days (n = 10)
and to 5.5 mg Mn/l for 25 days (n = 8) in containers as described
above and TCC were compared to that of control animals. The
coelomic fluid was collected by cutting the outermost edge of an
arm and the fluid was drained into an Eppendorf tube, kept in ice.
One hundred microliters of the fluid was immediately fixed in 4%
formalin. The cell numbers were counted in a microscope using
Bürker chamber.
2.5. Cell proliferation
Two different methods, as described below, were used to inves-
tigate the influence of Mn on cell proliferation of circulating
coelomocytes and of coelothelial cells. For both methods the coelo-
mocytes were collected in calcium and magnesium free saline
buffer CMFSS–EGTA (435 mM NaCl, 10.7 mM KCl, 27 mM Na2 SO4 ,
16.6 mM C6 H12 O6 , 12 mM HEPES, 5 mM EGTA; pH 7.4) and washed
twice in CMFSS without EGTA by centrifugation at 2000 × g for
5 min at 4 ◦ C. The coelomic epithelium was dissected from two
arms of each sea star and then prepared in accordance with the
study of Söderhäll et al. (2003). Briefly, the epithelium was trans-
ferred to 600 ␮l 0.1% collagenase Type I and 0.1% collagenase Type
IV, dissolved in CMFSS. After approximately 1 h incubation at room
temperature (RT) the remaining epithelium was gently crushed and
removed before the released cells were washed twice as described
above. The cells were allowed to adhere to SuperFrost®Plus slides
(Menzel GmbH & Co KG, Braunschweig, Germany) in a final concen-
tration of 50 mM CaCl2 for 45 min. After rinsing the slides in CMFSS
the cells were fixed in 70% ethanol in 50 mM glycine for 5 min.
The cell monolayers were submitted to the following proliferation
analyses.
2.5.1. BrdU incorporation
Cell proliferation was analyzed and compared by using the
substitute nucleotide, 5-bromo-2 -deoxyuridine (BrdU; B 5002,
Sigma–Aldrich, Stockholm, Sweden; Moss et al., 1998) in Mn-
exposed, 15 mg Mn/l seawater for 5 days, and control sea stars
(n = 8). Four hours prior to the sampling and performance of cell
monolayers (as described above) the sea stars were injected with
50 mM BrdU (10 ␮l/g wet tissue). Fixed cells were permeabilized
by incubation in 2 M HCl with 0.2 mg pepsin (Merck, Dramstadt,
Germany) per ml for 30 min at 30 ◦ C and then neutralized in 0.1 M
Na-borate (pH 8.5) for 5 min at RT. Bovine serum albumin (0.5% BSA;
Intergen, Toronto, Ontario, Canada) in PBS (phosphate-buffered
saline)–Tween (0.05% Tween-20; Merck) was used as blocking
medium, 1 h at RT before incubation for 30 min with anti-BrdU-FITC
(BD Bioscience, 347583, Becton and Dickinson, USA) diluted 1:10
in blocking medium. To enhance the FITC labeling the slides were
incubated with a secondary anti-mouse IgG-FITC (Sigma F3008),
diluted 1:100 in blocking medium for 30 min before mounted in
Vectashield with propidiumiodide (H-1300; Vector). During the
anti-BrdU and FITC labeling the incubations were done in a humid
 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81
chamber and each step of the procedure was followed by rinsing
the slides several times in PBS–Tween. Percentage of BrdU incor-
porating nuclei was examined by using a Leica DMRBE fluorescent
microscope at 40× magnification (filters: N2.I-green, to detect pro-
pidiumiodide, and I3-blue, to detect FITC).
2.5.2. Mitotic index
Cell proliferation was compared by microscopical determina-
tion of the rate of mitotic stages of nuclei found in Mn-exposed,
15 mg Mn/l seawater for 5 and 10 days, and control sea stars (n = 8).
The slides with cell monolayers were rinsed for 5 min in sterile
water before permeabilization of cells in 0.1 ␮g Proteinase K (PCR-
grade, Roche), in TES-buffer (50 mM Tris–HCL, 10 mM EDTA, 10 mM
NaCl; pH 7.4) for 12 min in a humid chamber at RT. Then the slides
were rinsed again in sterile water and mounted in Vectashield with
the nuclei staining dye, propidium iodide (H-1300; Vector). The per-
centage of mitotic cells, MI (%), was examined using a Leica DMRBE
fluorescent microscope at 40× magnification (filters: N2.I-green).
2.6. Stress response
2.6.1. Immunohistochemistry to detect HSP70
Pieces of arms (10–15 mm) from sea stars that had been exposed
to 5.5 mg Mn/l for 5, 10, 15 and 25 days and control sea stars
were removed by scalpel after being anaesthetized in 3% MgCl2
in seawater for up to 30 min (Moss et al., 1998). Samples were
fixed in 4% paraformaldehyde for 4 h at RT and then de-calcified
in 0.5 M EDTA in filtered sea water, pH 8, for 48 h. Thereafter
the tissue was embedded in paraffin wax (Oxoid) and sectioned
(6 ␮m) by standard methods. Sections were de-waxed, rehydrated
through an alcohol series, rinsed with PBS and incubated for 1 h
in 5% (v/v) heat inactivated normal goat serum (NGS) diluted in
PBS. The specimens were then incubated overnight with primary
antibody, anti-HSP70 antibody (H-5147, Sigma), diluted 1:500 in
5% NGS in PBS (Pinsino et al., 2007). Slides were rinsed in PBS
and incubated in the secondary antibody goat anti-mouse IgG
alkaline phospatase conjugate (Sigma), diluted 1:600 in 5% NGS
in PBS for 1 h. Cross-reactivity was revealed by incubation with
5-bromo-4-chloro-3-indolyl phospate/nitroblue tetrazolium sub-
strate (BCIP/NBT stock solution, Sigma: 18.75 mg/ml NBT, 9.4 mg/ml
BCIP in 67% v/v, DMSO). Controls were made by omitting the first
antibody and these showed no appreciable staining. Photos were
taken using a Leica DMRBE microscope, equipped with a digital
camera.
2.6.2. Western blotting to detect HSP70
Coelomocytes from sea stars, exposed to 15 mg Mn/l for 5 and 10
days, were collected into Eppendorf tubes on ice and homogenized
in cold RIPA lysis buffer (Hernebring et al., 2006) supplemented
with 1 mM Pefa-block as a reducing agent. The homogenates were
centrifuged at 14,000 × g, for 10 min at 4 ◦ C and the obtained super-
natants were stored at −80 ◦ C until used. Total protein content
was measured using the BCA kit (Pierce, Rockford, IL, USA). About
15 ␮g of protein from each sample was mixed with 6 ␮l sample
buffer (Invitrogen) and distilled water. Samples were denatured
at 100 ◦ C for 5 min and loaded on gel (Invitrogen Minigel 4–12%).
The gel electrophoresis was run (200 V, 50 min) using Invitrogen
running buffer. The cassette was stacked for Western Blot with
Hybond-P membranes (GE Healthcare Bio-Sciences AB, Uppsala,
Sweden) loaded with Invitrogen transfer buffer and run at 30 V for
1 h. The blot was blocked for 1 h, in 5% dry milk TBS-T solution. After
washing in TBS-T the membrane was incubated with the primary
antibody, anti-HSP70 (Sigma) for 1 h at RT, and then after washing
in TBS-T by the secondary antibody, peroxidase-conjugated anti-
mouse IgG, for 1 h at RT. The antibodies were diluted in TBS-T,
77
1:3000 in accordance to the manufacturer’s instruction. For chemi-
luminescence detection ECL Western Blotting Detection Reagents
(RPN2106 Amersham Bioscience) were used. Blots were analyzed
in a Geldoc 2000 and the intensity of the HSP70 bands was judged
by measuring the peak density (mean of three measurements) with
the software Quantity One (Bio-Rad Laboratories AB, Sundbyberg,
Sweden). The electrophoresis analyses were repeated at least two
times for each set of samples.
2.6.3. Oxyblot
Proteins are one of the major targets for oxygen-derived free
radicals. In order to recognize such oxidation, proteins from coelo-
mocytes and coelothelial cells were extracted as described above
and analyzed by measuring the levels of dinitrophenylhydrazone
derivatives of protein carbonyls by using the Oxyblot S7150 kit
(Chemicon International, Inc., Temecula, CA, USA). Twenty micro-
grams of protein from Mn-exposed (15 mg/l for 5 and 10 days) and
control sea stars were derivatized and then stored at −80 ◦ C to be
used for electrophoresis analyses within a week. Precasted 10%
Tris–HCl NuPage gels (Invitrogen Corporation, Carlsbad, CA, USA)
and Hybond-P membranes (GE Healthcare Bio-Sciences AB) were
used for the electrophoresis and blotting procedure. The antibody
labeling pattern was visualized using ECL plus (GE Healthcare Bio-
Sciences AB) and a Geldoc 2000 (software Quantity one; Bio-Rad
Laboratories) and the intensity of the 45 kDa band was analyzed
as described for HSP70. The membrane was post-stained in 0.5%
Ponceu S (P3504) to verify equal protein loading and to check that
no transfer between samples had occurred. All reagents not further
specified were from Sigma–Aldrich Corporate (St. Louis, MO, USA).
The electrophoresis analyses were repeated at least two times for
each set of samples.
2.6.4. Cytotoxicity
In order to investigate the cytotoxicity of Mn a colorimetric
method, based on enzymatic reduction of tetrazolium (MTS) and
phenylmetasulfazone (PMS) to formazan (CellTiter 96® aQuenous
Non-Radioactive Cell Proliferation Assay G5421, Promega Corpo-
ration, WI, USA) was used. Coelomocytes were harvested from
un-exposed sea stars (n = 4) and the concentration adjusted to
approximately 107 ml−1 of homologous cell free coelomic fluid.
Fifty microliters aliquots of the suspension were transferred to
96-wells microplates and exposed to 0, 1, 7 and 15 mg Mn/l of physi-
ological saline buffer (PSB; 20 mM HEPES, 436 mM NaCl, 10 mM KCl,
10 mM CaCl2 , 53 mM MgSO4 ; pH 7.4) in a final volume of 100 ␮l.
Coelomocytes exposed to 7.2 and 30 ␮g Zn (PH Tamm Laborato-
ries AB, Altuna, Sweden) per liter, was used as a cation reference to
the Mn exposure, as previously reported in studies on sea urchin
embryos (Kobayashi and Okamura, 2005). After in vitro incubation
on a shaking table for 16 h at RT 20 ␮l of the MTS/PMS dyes were
added, followed by another 2 h of incubation. The formazan product
was measured (490 nm) in a microplate reader (Labsystems iEMS
Reader MF). The survival index, SI (%), of metal-exposed cells was
calculated in relation to cells incubated with PSB only.
2.7. Phagocytosis
To test the functional response in Mn-exposed A. rubens, the
phagocytic capacity of coelomocytes was investigated 5 days
after exposure to 15 mg Mn/l and compared to control sea stars
(n = 16). Following the protocol of Andersson and Mora (1995)
heat-killed yeast cells (Saccharomyces cerevisiae) were labeled
with fluorescein-5-isothiocyanate (FITC, Merck, Darmstadt, Ger-
many) and diluted with PSB to a concentration of 108 cells/ml.
Coelomic fluid was sampled and the numbers of coelomocytes
were adjusted in PSB to a concentration of 107 cells/ml. Fifty
78 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81

Turku, Finland). Phagocytic index (PI, %) was calculated in rela-

tion to yeast-FITC, incubated in buffer without coelomocytes:
[PI = 100 − ((Absyeast × 100)/Abs(yeast+coelomocytes) )], where the Abs
values represented the mean value of five wells.

2.8. Statistics

Kruskal–Wallis one way analysis of variance on ranks (ANOVA)

Fig. 1. The concentration of manganese (mg/l ± S.E.) in the coelomic fluid of A. rubens
after 5, 10, 15 and 25 days of exposure to 5.5 mg Mn/l seawater.
was performed to investigate Mn effect on TCC data and Dunn’s test
was used for pairwise multiple comparison. The power of data was
tested before entering the ANOVA. Two-tailed Student’s t-test was
used to compare the immunocytochemistry data of the percent-
age of BrdU-incorporated as well as of mitotic cells. The same test
was used for comparing carbonylated proteins (oxyblot), expres-
sion of HSP70, cytotoxicity and phagocytic capacity, respectively,
between Mn-exposed and control sea stars. The significance was
set to p = 0.05. The statistical analyses were performed using Sigma
Stat (version 3.5; Jandel Scientific software, San Rafael, CA).

3. Results
microliters of coelomocytes and yeast cells, respectively, were
incubated in 96-well microplates for 45 min before adding trypan 3.1. Manganese analysis
blue solution for quenching extracellular FITC-yeast signal. After
8 min of quenching the microplates were analyzed at 485/535 nm In control animals the mean manganese concentration of
in a Wallac 1420 VICTORTM multilabel counter (EG&G Wallac, coelomic fluid was 0.22 ± 0.13 mg/l. After 5 days of exposure to

Fig. 2. Graphs showing the effects of manganese on numbers of circulating coelomocytes and proliferation of cells from coelomic epithelium (coelothelial cells) of A. rubens.
Un-exposed sea stars are used as controls (Cont). (A) Total coelomocyte counts 106 ml−1 (±S.E.) of coelomic fluid of sea stars exposed to 15 mg/l for 5 and 10 days (grey
columns) and to 5.5 mg/l for 25 days (white columns), respectively. (B) Percentage (±S.E.) of anti-BrdU-labeled coelothelial cells of sea stars exposed to 15 mg Mn/l for 5 days.
(C) Percentage of mitotic coelothelial cells of sea stars exposed to 15 mg Mn/l for 5 and 10 days. *p ≤ 0.05; **p ≤ 0.01 and ***p ≤ 0.001.
 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81 79

5.5 mg Mn/l the concentration rose to 8.26 ± 0.58 mg/l and did not sea stars after 5 days of exposure (Diff of means: 5.3%; p < 0.001)
become significantly different after 10, 15 and 25 days of exposure (Fig. 2B).
(Fig. 1).

3.4.2. Mitotic index

3.2. Body index
The mean MI (%) (n = 8) of cells released from coelomic epithe-
lium was 4–5% in control animals and slightly but significantly
The mean of the calculated body index varied between 152 and
elevated in Mn-exposed sea stars both after 5 and 10 days (Diff of
223 when analyzing all the different treatment groups (control 5
means 5 days: 3.4%, p < 0.004; Diff of means 10 days: 2.7%, p = 0.008)
and 10 days, Mn-exposed 5 and 10 days). The difference was not
(Fig. 2C). Mitotic cells were not found in circulating coelomocytes.
significant for any of the groups when comparing body index before
and after the treatments.
3.5. Stress response
3.3. Total coelomocyte counts
3.5.1. HSP70
When investigating HSP70 in coelomocytes with the Western
After 5 days of exposure to 15 mg Mn/l the mean cell number
blot technique no pronounced expression was observed after 5 days
was 5.3 ± 2.3 × 106 ml−1 and significantly higher than in controls
of exposure to 15 mg Mn/l but after 10 days there was a tendency
(3.0 ± 2.4 × 106 ml−1 ; Diff of ranks: 6.3; p = 0.014). After 10 days the
of higher levels in Mn-exposed sea stars. This was, however, not
mean of cell numbers in the Mn-exposed specimens was still sig-
statistically significant (Fig. 3A and D). The immunohistochemi-
nificantly higher than that of the controls (<0.001) but not higher
cal staining of tissue sections showed an overexpression of HSP70
than after 5 days. After 25 days of exposure to 5.5 mg Mn/l, TCC was
when sea stars were exposed to 5.5 mg Mn/l. This was confirmed
significantly higher (7.5 ± 1.7 × 106 ml−1 ) compared to that of the
for all sampling occasions (5, 10, 15 and 25 days) but the highest
control animals (3.6 ± 0.7 × 106 ml−1 ; Diff of ranks: 3.9; p < 0.001)
tissue expression was found after 5 days (Fig. 3B and C).
and also significantly higher compared to the value in animals
exposed to 15 mg/l for 5 and 10 days (Fig. 2A).
3.5.2. Oxyblot
3.4. Cell proliferation Oxyblot analyses of coelomocytes showed a pronounced car-
bonylation of a protein with a molecular size of about 45 kDa. After
3.4.1. BrdU incorporation 10 days of Mn exposure (15 mg/l) there was a reduction in levels
The mean percentage of BrdU-labeled coelomic epithelium cells of protein carbonyls compared to those of the controls (p = 0.004,
was <2% in control animals and significantly higher in Mn-exposed n = 7; Fig. 3E and F). This was not shown for the shorter exposure

Fig. 3. Stress response to manganese in coelomocytes and coelomic epithelium of A. rubens. Un-exposed sea stars are used as controls Holm et al., 2008. (A) Graph showing
the peak density mm−1 (±S.E.) from Western Blot analyses of the expression of HSP70 in coelomocytes (D) of A. rubens after exposure to 15 Mn mg/l, for 10 days compared
to that of control animals. (B) Immunohistochemical analyze of coelomic epithelium of control sea stars did not show expression of HSP70. (C) After exposure to 5.5 mg Mn/l
for 5 days the expression was visual as shown by the dark colored tissue. (E) Graph of the oxyblot analyses of coelomocytes showing reduction in levels of carbonylation of
protein after 10 days of Mn exposure (15 mg/l) compared to that of the controls. (F) Blot showing the carbonylation of proteins in the coelomocytes. Bands of the molecular
size of about 45 kDa, indicated with asterisk, were used for judging the peak density mm−1 (±S.E.) of the graph shown in (E).
80 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81
Fig. 4. Cytotoxic response of coelomocytes to manganese. Graph showing an
increasing reduction of coelomocyte dehydrogenase activity after in vitro exposure
to 1, 7 and 15 mg Mn/l. The response to the reference cation Zn in concentrations of
7.2 and 30 ␮g/l showed an opposite pattern. The viability was measured as enzy-
matic reduction of MTS/PMS to formazan, calculated in relation (% ± S.E.) to control
sea stars (n = 4) and presented as survival index (SI, %).
time (5 days). The protein carbonyls of coelomic epithelium did not
differ after 5 or after 10 days Mn exposure.
3.5.3. Cytotoxicity
The viability of coelomocytes was not significantly affected
when exposed in vitro to 1 and 7 mg Mn/l, respectively, but was
reduced by approximately 20% when exposed to 15 mg Mn/l (Fig. 4).
The exposure to Zn did not have such an effect. Instead the
metabolic activity tended to increase in the presence of the con-
centration of Zn used in this study (Fig. 4).
3.6. Phagocytosis
When coelomocytes from control animals were interacting with
yeast cells in the in vitro experiment the mean PI (%) (n = 16) was
20.2%. For coelomocytes from Mn-exposed sea stars (15 mg/l, 5
days) it was significantly lower (14.4%; Diff. of mean: 5.8%, p < 0.037)
(Fig. 5).
Fig. 5. Phagocytic response of coelomocytes to manganese. Graph showing reduced
phagocytic capacity of coelomocytes after 5 days of Mn exposure (15 mg/l) compared
to that of controls (n = 16) when incubated with heat-killed FITC-labeled yeast cells
(Saccharomyces cerevisiae). Phagocytotic index was calculated as the percentage of
the added yeast that was engulfed after 45 min incubation (±S.E.).
4. Discussion
Despite the observation that manganese accumulates in the
coelomic fluid of A. rubens the coelomocyte counts were not, as
hypothesized, reduced. Already after 5 days of exposure the cell
numbers had increased with approximately 50%. This was con-
trary to what previously has been found in the crustacean Neprophs
norvegicus (Hernroth et al., 2004) where manganese significantly
reduced the cell numbers and inhibited the proliferation and dif-
ferentiation of haematopoietic stem cells. In high concentrations
Mn is known to interact with calcium and in that way interrupt
the synaptic transmission (Luk and Culotta, 2001). Thus Mn might
neurologically affect A. rubens ectoderm derivatives such as the
hydrostatic organs, or tube feet, and thereby disturb the home-
ostasis of the coelom. Such a homeostatic change rather than an
induction of cell proliferation might be the cause for the observed
elevated concentrations of coelomocytes. However, the present
study revealed that the coelomic fluid density did not change as
indicated by the unchanged body index of Mn-exposed A. rubens.
Instead, the increase of dividing coelothelial cells suggests that
manganese-induced cell proliferation and renewal of circulating
coelomocytes.
We found that the coelomocyte numbers increased more after
the animals were exposed for 25 days to the lower concentration
of Mn (5.5 mg/l) than when they were exposed to the higher con-
centration (15 mg/l) for a shorter time. This might be explained by
the cytotoxic properties of Mn that were recognized in the in vitro
study at this concentration. The coelomic fluid has been suggested
as a barrier that prevents manganese to accumulate in the deeper
tissues of A. rubens (Nordahl Hansen and Bjerregaard, 1995). The
role of coelomocytes behind such possible barrier function is cur-
rently unclear. Detoxification through metallothioneins, which are
known to regulate the sequestration of a variety of metals such
as cadmium and copper, does not appear as a pathway for the
elimination of manganese (Viarengo, 1985). However, a possible
pathway might be to trap the metal in the lysosomes of the phago-
cytes to be eliminated by the diapedesis process, which has been
described for indigestible particles in molluscs (George and Pirie,
1980; Suryawanshi and Langekar, 2006).
The HSP70 chaperones have frequently been used as stress
markers (Matranga et al., 2002, 2006; Pinsino et al., 2007; Yang
et al., 2008). They are regarded as intracellular molecules but
can also be released extracellularly in response to stressful con-
ditions (Tsan and Gao, 2004). We found increased HSP70 levels in
coelomic epithelium as well as a tendency for overexpression also
in the coelomocytes. Carbonylated proteins, recognized as mark-
ers of oxidative stress (Banerjee et al., 2007; Almroth et al., 2005)
were, on the other hand, reduced in these cells after 10 days of
Mn-exposure. The reduction in protein carbonyls was surprising,
but can theoretically be explained by upstream chaperone effects
of up-regulated HSP72 (Bose et al., 1996), or possibly by stimu-
lated MnSOD activities which reduce protein carbonyls (Wang et
al., 2004). The response using these kinds of stress markers gives,
however, little information concerning possible negative effects of
Mn on A. rubens. At a functional level, however, there was a negative
effect on the phagocytic capacity of coelomocytes of Mn-exposed
animals. HSP70 may suppress the reactivity of immune system
cells as shown in studies on sea urchin coelomocytes (Browne et
al., 2007) and the cytotoxic properties of Mn might also impair
their functionality. Another explanation for the reduced phago-
cytic capacity could be a possible change of the coelomocyte
composition with reduced numbers of phagocytes. Although fur-
ther studies are needed to clarify the reason behind it could be
concluded that Mn suppresses the phagocytic capacity of the sea
stars.
 C. Oweson et al. / Aquatic Toxicology 89 (2008) 75–81
When exposing coelomocytes in vitro to Mn it showed, contrary
to the cation Zn, cytotoxic properties at levels that can be found in
nature. “No observed effect concentration” (NOEC) of zinc on sea
urchin embryos has been determined at the level of 7.2 ␮g/l, which
was considered the normal level of non-contaminated seawater.
However, at abnormal levels (>14 ␮g/l) the effects were obvious
(Kobayashi and Okamura, 2005). In the present study neither the
normal nor the abnormal concentrations of Zn negatively affected
the viability of coelomocytes after in vitro exposure. Certainly, dif-
ferent cell types have different resistance to Mn. Hirata (2002)
found, by using the same method as in our study, that the viability of
a neuronal cell line (PC2) was significantly reduced when exposed
to 5 and 55 mg/l of Mn for 48 h. Opposite, Mn (20 mg/l) was not
cytotoxic to haemocytes of N. norvegicus as shown in the study of
Oweson et al. (2006). At the organismal level it has been pointed out
that A. rubens does not show mortality at Mn concentrations used in
our study, but when increasing the concentration 10-fold, the mor-
tality becomes significant (Nordahl Hansen and Bjerregaard, 1995).
Accumulation of the metal in A. rubens has shown linearity with
time up to 23 days at low concentration (0.1 mg/l) but saturation
kinetics at higher concentrations (Nordahl Hansen and Bjerregaard,
1995). This was also the case in the present study where steady-
state levels were reached in the coelomic fluids already after 5 days
of exposure to 5.5 mg/l.
This study indicated that manganese has immunotoxicologi-
cal effects when A. rubens are exposed to levels as high as those
that could be found in seawater affected by hypoxia or mine
effluent. Although manganese acts as a trigger for proliferation
of coelothelial cells and thus increased the number of circulating
coelomocytes, the cells showed stress response and their phago-
cytic capacity was negatively affected. The increase of coelomocyte
numbers may compensate for this functional disruption. However,
as being cytotoxic to these cells it seems possible that Mn affects
also other immune mechanisms and suppresses their defense
against pathogens.
Acknowledgements
We thank the Biomedical Scientist Eva Nilsson for excellent
technical support. The project has partly been financed by The
Swedish Research Council for Environmental, Agriculture and Spe-
cial Planning and The Memory Foundation of Carl Tryggers (C.O.
and B.H), The Swedish Research Council (H.S.), the visiting pro-
gram of the European Network of Excellence “Marine Genomics
Europe” and Italian Space Agency (MoMa Project) (V.M.). A.P. has
been supported by a scholarship from the University of Palermo,
Italy.
References
Almroth, B.C., Sturve, J., Berglund, Å., Förlin, L., 2005. Oxidative damage in eelpout
(Zoarces viviparus), measured as protein carbonyls and TBARS, as biomarkers.
Aquat. Toxicol. 73, 171–180.
Andersson, R.S., Mora, L.M., 1995. Phagocytosis: a microtiter plate assay. In: Stolen,
J.S., Fletcher, T.C., Smith, S.A., Zelikoff, J.T., Kaattari, S.L., Andersson, R.S., Söderhäll,
K., Weeks-Perkins, B.A. (Eds.), Techniques in Fish Immunology-4: Immunology
and Pathology of Aquatic Invertebrates. SOS Publications, Fair Haven, NJ, USA,
pp. 109–115.
Baden, S.P., Pihl, L., Rosenberg, R., 1990. Effects of oxygen depletion on the ecology,
blood physiology and fishery of the Norway lobster, Nephrops norvegicus. Mar.
Ecol. Prog. Ser. 67, 141–155.
Balzer, W., 1982. On the distribution of iron and manganese at the sediment/water
interface: thermodynamic versus kinetic control. Geochim. Cosmochim. Acta 46,
1153–1161.
Banerjee, S., Maity, P., Mukherjee, S., Sil, A.K., Panda, K., Chattopadhyay, D., Chatter-
jee, I.B., 2007. Black tea prevents cigarette smoke-induced apoptosis and lung
damage. J. Inflamm. (London) 4:3 (14).
81
Bose, S., Weikl, T., Bügl, H., Buchner, J., 1996. Chaperone function of Hsp90-associated
proteins. Science 274, 1715–1717.
Browne, C.L., Swan, J.B., Rankin, E.E., Calvert, H., Griffiths, S., Tytell, M., 2007. Extra-
cellular heat shock protein 70 has novel functional effects on sea urchin eggs
and coelomocytes. J. Exp. Biol. 210, 1275–1287.
Coteur, G., Gillan, D., Joly, G., Pernet, P., Dubois, P., 2003. Field contamination of the
starfish Asterias rubens by metals. Part 2. Effects on cellular immunity. Environ.
Toxicol. Chem. 22, 2145–2151.
Diaz, R.J., Rosenberg, R., 1995. Marine benthic hypoxia: a review of its ecological
effects and the behavioural responses of benthic macrofauna. Oceanogr. Mar.
Biol.: Annu. Rev. 33, 245–303.
Frank, K.T., Petrie, B., Choi, J.S., Leggett, W.C., 2005. Trophic cascades in a formerly
cod-dominated ecosystem. Science 308, 1621–1623.
George, S.G., Pirie, B.J.S., 1980. Metabolism of zinc in the mussel, Mytilus edulis (L.):
a combined ultrastructural and biochemical study. J. Mar. Biol. Assoc. UK 60,
575–590.
Hall, I.R., Hydes, D.J., Statham, P.J., Overnell, J., 1996. Dissolve and particulate trace
metals in a Scottish Sea Loch: an example of pristine environment? Mar. Pollut.
Bull. 32, 846–854.
Hernebring, M., Brolen, G., Aguilaniu, H., Semb, H., Nystrom, T., 2006. Elimination
of damaged proteins during differentiation of embryonic stem cells. Proc. Natl.
Acad. Sci. U.S.A. 16, 7700–7705.
Hernroth, B., Baden, S., Holm, K., Andrén, T., Söderhäll, I., 2004. Manganese induced
immune suppression of the lobster Nephrops norvegicus. Aquat. Toxicol. 70,
223–231.
Hirata, Y., 2002. Manganese-induced apoptosis in PC12 cells. Neurotoxicol. Teratol.
24, 639–653.
Holm, K., Dupont, S., Sköld, H., Stenius, A., Thorndyke, M., Hernroth, B., 2008. Induced
cell proliferation in putative haematopoietic tissues of the sea star, Asterias
rubens (Linnaeus). J. Exp. Biol. 211.
Iregren, A., 1990. Physiological test performance in foundry workers exposed to low
levels of manganese. Neurotoxicol. Teratol. 12, 673–675.
Kobayashi, N., Okamura, H., 2005. Effects of heavy metals on sea urchin embryo
development. Part 2. Interactive toxic effect of heavy metals in synthetic mine
effluents. Chemosphere 61, 1198–1203.
Luk, E.E., Culotta, V.C., 2001. Manganese superoxide dismutase in Saccaromyces cere-
visiae acquires its metal co-factor through a pathway involving the Nramp metal
transporter, Smf2p. J. Biol. Chem. 276, 556–562.
Matranga, V., Toia, G., Bonaventura, R., Müller, W.E.G., 2000. Cellular and biochemical
responses to environmental and experimentally induced stress in sea urchin
coelomocytes. Cell Stress Chaperones 5, 113–120.
Matranga, V., Bonaventura, R., Di Bella, G., 2002. HSP70 as a stress marker of sea
urchin coelomocytes in short term cultures. Cell. Mol. Biol. 48, 345–359.
Matranga, V., Pinsino, A., Celi, M., Di Bella, G., Natoli, A., 2006. Impacts of UV-B
radiation on short-term cultures of sea urchin coelomocytes. Mar. Biol. 149,
25–34.
Mena, I., Marin, O., Fuenzalida, S., Cotizas, G.C., 1967. Chronic manganese poisoning,
clinical pictures and manganese turn over. Neurology 17, 128–136.
Moss, C., Hunter, A.J., Thorndyke, M.C., 1998. Patterns of bromodeoxyuridine incor-
poration in neuropeptide immunoreactivity during arm regeneration in the
starfish Asterias rubens. Philos. Trans. R. Soc. Lond. B: Biol. Sci. 353, 421–436.
Muñoz-Chápuli, R., Carmona, R., Guadix, J.A., Macías, D., Pérez-Pomares, J.M., 2005.
The origin of the endothelial cells: an evo–devo approach for the inverte-
brate/vertebrate transition of the circulatory system. Evol. Dev. 7, 351–358.
Nordahl Hansen, S., Bjerregaard, P., 1995. Manganese kinetics in the sea star Asterias
rubens (L.) exposed via food or water. Mar. Pollut. Bull. 31, 127–132.
Oweson, C., Baden, S., Hernroth, B., 2006. Manganese induced apoptosis in
haematopoietic cells of the lobster, Nephrops norvegicus. Aquat. Toxicol. 77,
322–328.
Phil, L., Baden, S.P., Diaz, R.J., 1991. Effects of periodic hypoxia on distribution of
demersal fish and crustaceans. Mar. Biol. 108, 349–360.
Pinsino, A., Thorndyke, M.C., Matranga, V., 2007. Coelomocytes and post-traumatic
response in the common sea star Asterias rubens. Cell Stress Chaperones 12,
331–341.
Saier, B., 2001. Direct and indirect effects of seastars Asterias rubens on mussel beds
(Mytilus edulis) in the Wadden Sea. J. Sea Res. 46, 29–42.
Smith, V.J., 1981. Invertebrate blood cells. In: Ratcliffe, N.A., Rowley, A.F. (Eds.), The
Echinoderms. Academic Press, New York, pp. 514–562.
Söderhäll, I., Bangyeekhun, E., Mayo, S., Söderhäll, K., 2003. Hemocyte production
and maturation in an invertebrate animal; proliferation and gene expression in
hematopoietic stem cells of Pacifastacus leniusculus. Dev. Comp. Immunol. 27,
661–672.
Suryawanshi, G.D., Langekar, Z.A., 2006. Zinc and cadmium toxicity to estuarine rock
oyster Crassostera cattuckensis on Ratnagiri coast. Nat. Environ. Pollut. Technol.
5, 463–467.
Tsan, M.-F., Gao, B., 2004. Heat shock protein and innate immunity. Cell. Mol.
Immunol. 1, 274–279.
Viarengo, A., 1985. Biochemical effects of trace metals. Mar. Pollut. Bull. 16, 153–158.
Wang, Y., Oberley, L.W., Howe, D., Jarvis, D.L., Chauhan, G., Murhammer, D.W., 2004.
Effect of expression of manganese superoxide dismutase in baculovirus-infected
insect cells. Appl. Biochem. Biotechnol. 119, 181–193.
Yang, T., Hsu, C., Liao, W., Chuang, J.S., 2008. Heat shock protein 70 expression in
epilepsy suggests stress rather than protection. Acta Neuropathol. 115, 219–230.