Aquatic Toxicology 163 (2015) 37–50

Contents lists available at ScienceDirect

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

Prevalence and intensity of pathologies induced by the toxic dinoflagellate, Heterocapsa circularisquama, in the Mediterranean mussel, Mytilus galloprovincialis Leila Basti a,∗ , Makoto Endo b , Susumu Segawa c , Sandra E. Shumway d , Yuji Tanaka e , Satoshi Nagai a a Metagenomics Research Group, National Research Institute of Fisheries Science, Fisheries Research Agency, 2-12-4 Fukuura, Kanazawa, Yokohama, Kanagawa 236-8648, Japan b Laboratory of Fish Health Management, Tokyo University of Marine Science and Technology, Minato, Tokyo 108-8477, Japan c Laboratory of Aquatic Ecology, Department of Aqua Bioscience and Industry, Faculty of Bio-Industry, Tokyo University of Agriculture, 196 Yasaka, Abashiri 099-2493, Japan d Department of Marine Sciences, University of Connecticut, Groton, CT 06340, USA e Laboratory of Planktology, Department of Ocean Science, Tokyo University of Marine Science and Technology, Minato, Tokyo 108-8477, Japan

a r t i c l e

i n f o

Article history: Received 22 December 2014 Received in revised form 10 March 2015 Accepted 11 March 2015 Available online 14 March 2015 Keywords: Harmful algal blooms Heterocapsa circularisquama Mytilus galloprovincialis Histolopathology Mechanism Cytotoxicity

a b s t r a c t The harmful dinoflagellate, Heterocapsa circularisquama, has been causing mass mortalities of bivalve molluscs in Japan, at relatively low cell densities. Although several studies have been conducted to determine the toxicity mechanisms, the specific cause of death is still unclear. In a previous study, in our laboratory, it was shown that H. circularisquama (103 cells ml−1 ) caused extensive cytotoxicity in the gills of short-neck clams, Ruditapes philippinarum. In the present study, Mediterranean mussels, Mytilus galloprovincialis, were exposed to H. circularisquama at four cell densities (5, 50, 500, 103 cells ml−1 ), three temperatures (15, 20, and 25 ◦ C), and three exposure durations (3, 24, and 48 h), and the pathologies in nine organs (gills, labial palps, mantle, hepatopancreas, stomach, intestines, exhalant siphon, adductor muscles, and foot) were assessed. Foot, adductor muscles, and exhalent siphons of mussels were not affected; however, 16 inflammatory (hemocytic infiltration and aggregation, diapedesis, hyperplasia, hypertrophy, edema, melanization, and firbrosis) and degenerative (thrombus, thrombosed edema, cilia matting and exfoliation, epithelial desquamation, atrophy, and necrosis) pathologies were identified in the gills, labial palps, mantle, hepatopancreas, stomach, and intestines. The total prevalence and total intensity of pathology in each individual mussel, and the prevalence and intensity of pathology in each organ increased significantly with increased cell density, exposure duration, and temperature. The prevalence of pathology was the highest in gills, followed by the prevalence in labial palps, mantle, stomach, and intestines. Pathology was least prevalent in the hepatopancreas. The intensity of pathology was the highest in the gills, followed by the labial palps and mantle, the stomach and intestines, and the hepatopancreas. This detailed quantitative histopathological study demonstrates that exposure to H. circularisquama induces a broad cytotoxic effect in six vital organs, even at low density (5 cells ml−1 ) and low temperature (15 ◦ C), but not in muscular organs. Combining cell density, time, and duration of exposure, the organ most affected by the harmful alga was the gill, followed by the labial palps and mantle, the stomach and intestines, and the hepatopancreas. The results of this pathological analysis show that exposure to H. ciruclarisquama severely affects the gills, the labial palps, and mantle thereby interfering with particle clearance and sorting, cleansing, and respiration, but also affects the stomach, intestines, and hepatopancreas, altering the digestive processes and possibly detoxification pathways, if mussels are able to detoxify the toxins of H. circularisquama. In the most severe cases, bivalves would most likely have died as a result of combined severe alterations of the vital functions, failure of tissue repair, and moderate to heavy hemorrhaging in both the external organs and the digestive organs concomitantly with light to moderate alterations in the detoxifying processes. © 2015 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +81 45 788 7612; fax: +81 45 788 5001. E-mail addresses: [email protected], [email protected] (L. Basti). http://dx.doi.org/10.1016/j.aquatox.2015.03.012 0166-445X/© 2015 Elsevier B.V. All rights reserved.

38

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

1. Introduction Harmful algal blooms (HAB) cause a variety of acute, chronic, and sublethal effects in aquatic organisms, and mass mortalities induced by blooms of harmful algae have been reported for several species of bivalves (reviewed in Gainey and Shumway, 1988; Shumway, 1990; Landberg, 2002). Aside from the easilyperceivable mass mortalities both in the field and under laboratory conditions, HAB affect the physiology, behavior, immune system, and reproduction of bivalves (Shumway and Cucci, 1987; Shumway, 1990; Gainey and Shumway, 1988; Pearce et al., 2005; Leverone et al., 2006; Hégaret et al., 2007; Bricelj and MacQuarrie, 2007; Galimany et al., 2008a,b; Tran et al., 2010). These effects are mediated through the production of toxins, allelochemicals, and/or secondary by-products, by the sheer increase in the microalgal biomass, by physical contact, and in some cases by a combination of two or more of these affects (Anderson et al., 2002; Landberg, 2002). In Japan, blooms of the harmful dinoflagellate Heterocapsa circularisquama (average cell size: 18–24 ␮m) have been causing recurrent mortalities of bivalves including Pacific oysters, Crassostrea gigas, Japanese pearl oysters, Pinctada fucata, shortneck clams, Ruditapes philippinarum, and Mediterranean mussels, Mytilus galloprovincialis (Horiguchi, 1995; Matsuyama et al., 1996, 2001; Matsuyama, 2012). The density of its blooms associated with shellfish mortality ranges from 1.85 × 103 to 2.5 × 105 cells ml−1 (annual reports on harmful algae of Kyushu Sea Fisheries Coordination Office, 2004–2013; Matsuyama, 2003b). Previous studies have demonstrated that exposure to H. circularisquama causes a broad range of impact in bivalves, from impaired physiology and behavior in juveniles and adults to altered early-life development; extensive cytotoxicity in bivalves and several mammal cell lines being a well-documented effect of exposure to H. circularisquama through cell contact (Matsuyama, 1996, 2003a; Nagai et al., 1996, 2006; Matsuyama et al., 1997a,b; Yamatogi et al., 2005; Basti et al., 2009, 2011a,b, 2013). The toxins associated with H. circularisquama are still not well characterized due to their instability (Oda et al., 2001; Hiraga et al., 2002; Kim et al., 2002; Sato et al., 2002; Miyazaki et al., 2005), and the inability to extract and identify its toxin(s). The expansion of the geographical distribution in Japan, and its potent and lethal effects on virtually all life stages of bivalves in Japan, warrants continued assessment of the effects of H. circularisquama on bivalves. In a previous study, it was demonstrated that exposure of adult short-neck clams, R. philippinarum, to H. circularisquama resulted in necrosis stemming from extensive cytotoxicity in the gills, despite the activation of the immune responses of clams at the early stages of exposure (Basti et al., 2011a). In the present study, the effects of exposure to H. circularisquama on nine organs of the Mediterranean mussels, M. galloprovincialis, were assessed at different temperatures, cell densities, and exposure times, in an attempt to reconstruct the sequence of organ damage and to explain the toxicity mechanisms of H. circularisquama in a representative bivalve species.

2. Materials and methods

acclimated to the experimental conditions for 2–3 weeks in 70l tanks (80 individuals/tank). Mussels were fed daily on a single ration of I. galbana (25 × 108 cells/tank). Healthy mussels were allowed to clear their gut content for 48 h prior to the experiments in static 20-l tanks (20 individuals/tank) filled with filtered (18 l/tank) and continuously aerated seawater. 2.2. Exposure experiments and sample processing for light microscopy (LM) and scanning electron microscopy (SEM) Mussels (LM: N = 225, shell length = 42.89 ± 6.41 mm, wet weight = 7.58 ± 3.32 g; SEM: N = 135, shell length = 40.23 ± 5.22 mm, wet weight = 6.16 ± 2.64 g) were exposed to H. circularisquama at densities of 0, 5, 50, 500, and 103 cells ml−1 . Experiments were conducted at three temperatures 15, 20, and 25 ◦ C. For each combination of cell density and temperature, five mussels were sampled every 3, 24, and 48 h. Replication was not considered. The adductor muscles of each mussel were severed, and the individual mussels were fixed whole for 24 h in 10% buffered seawater–formalin solution (pH 7.5). The gills, labial palps, exhalent siphons, mantle, adductor muscles, and foot were then separately excised and fixed in the buffered seawater–formalin solution for an additional 24 h, while the gastro-intestinal cavity was cut in two and further fixed in the formalin solution. All tissues were then gently washed in running tap water, distilled water, and preserved in 70% ethanol until processing. After dehydration in an ascendant series of alcohol solutions, each organ and longitudinal or transverse sections of the gastro-intestinal cavity were embedded in paraffin, and 5-␮m sections were stained with hematoxylin–eosin. Slides were observed under a light microscope (Eclipse E600, Nikon, Kanagawa, Japan) to assess and quantify the pathologies. For scanning electron microscopy, the mussels were fixed in Bouin’s solution, preserved in 70% alcohol, then dehydrated organs, and sections were freeze-dried (JFD-300, JEOL, Tokyo, Japan), ion-sputtered (E-1030, HITACHI, Tokyo, Japan), and mounted on aluminum stabs for observation (S-4000, HITACHI, Tokyo, Japan). 2.3. Quantitative assessment of the prevalence and intensity of pathology Prevalence and intensity were used as quantitative measurements of the pathologies in individual mussels and in each organ examined. The prevalence is an estimate of the proportion of individual mussels that have showed pathology at least in one of the examined organs. The intensity is an estimate of the strength of the pathology. The total prevalence of pathology (P), the total intensity of pathology (I), the prevalence of pathology in each organ (Pi), and the intensity of pathology in each organ (Ii) were assessed for each combination of cell density–temperature–exposure duration, based on a semi-quantitative scale similar to the one adopted by Basti et al. (2011a) with minor modifications for enhanced accuracy, as follows:

 P=

Path o log yIndiv1,..., Path o log yIndivN



N

2.1. Algae culture and bivalve acclimation The harmful dinoflagellate H. circularisquama (strain 92HC, isolated from Ago Bay-Mie prefecture, Japan) and the nontoxic microalga Isochrysis galbana were cultured in modified SWM-3 medium as previously described by Basti et al. (2011a). Mediterranean mussels, M. galloprovincialis, were collected from Hakkeijima, Tokyo Bay Japan (35◦ 20 11 N and 130◦ 38 04 E), and

 I=

IntensityIndiv1,..., IntensityIndivN N

 Pi = 



Path o log yo rgan i/Indiv1,..., Path o log yo rgan i/IndivN N



L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

39

Table 1 Pathologies observed in nine organs of the Mediterranean mussel, Mytilus galloprovincialis, exposed to the harmful dinoflagellate, Heterocapsa circulariquama. Pathologies

Organs Gills

Labial palps

Mantle

Hepato-pancreas

Stomach

Intestines

Adductor muscles

Siphons

Foot

Hemocytic infiltrationa Hemocyte aggregationa Thrombusb Hyperplasiaa Hypertrophya Edemab Thrombosed edemab Melanizationc Fibrosisd Filament fusionb Cilia matting/exfoliationb Epithelial desquamationb Diapedesisa Hemorrhageb Atrophyb Necrosisb

o o o o o o o o × o o o × × o o

o o o o o o o × × × o o × × o o

o o o o × × × × o × o o × × o o

o o o × × × × × o × o × × o o o

o × × o o × × × × × o o o o o o

o × × o o × × × × × o o o o o o

× × × × × × × × × × × × × × × ×

× × × × × × × × × × × × × × × ×

× × × × × × × × × × × × × × × ×

Total (%)

13 (81.25)

11 (68.75)

9 (56.25)

8 (50.00)

9 (56.25)

9 (56.25)

0 (0.00)

0 (0.00)

0 (0.00)

o: Present and ×: absent. a Defensive inflammatory pathology. b Degenerative pathology. c Innate defensive pathology. d Regenerative pathology.

 Ii = 

Intensityo rgan i/Indiv1,..., Intensityo rgan i/IndivN



N

where (N) is the number of mussels processed for each combination of cell density–temperature–exposure duration, and (i) is the organ examined (i = g for gills, lp for labila palps, m for mantle, s for exhalent siphon, st for stomach, it for intestines, h for hepatopancreas, f for foot, or a for adductor muscles). The averaged data of prevalence, (P) and (Pi), were interpreted as low (0–0.25), intermediate (0.25–0.5), moderately high (0.5–0.75), or high (0.75–1). The averaged data of intensity, (I) and (Ii), were interpreted as very light (0–0.5), light (0.5–1), lightly moderate (1–1.5), moderate (1.5–2), moderately heavy (2–2.5), or heavy (2.5–3). 2.4. Data treatment Normality (Kolmogorov–Smirnov test and Shapiro–Wilk test) and homogeneity of variance (Hartley test, Cochran test, and Bartlett test) were checked a priori, and parametric assumptions were met following data transformation. Univariate and multivariate ANOVA were performed to assess the effects of temperature, exposure duration, and density of H. circularisquama on the total prevalence and intensity of pathology, and the prevalence and intensity of pathology in each organ. When analyses showed significant effects, post-hoc tests (Fisher’s LSD test and Newman–Keuls test) were considered to assess the level at which the significance occurred. Student’s T-tests for independent samples were conducted to cross compare significant difference in the prevalence and intensity of pathology between organs. Three levels of significance were adopted at P = 0.05, 0.01, or 0.001. 3. Results 3.1. Identification of pathological alterations Pathological alterations were not observed in the exhalant siphons, adductor muscles, or foot of mussels exposed to H. circularisquama; however, 16 inflammatory, defensive, regenerative, and

degenerative pathologies were identified in the gills, labial palps, mantle, hepatopancreas, stomach, and intestines of exposed mussels (Table 1). More than 80% of the 16 pathologies were observed in the gills, nearly 70% in the labial palps, 56% in the mantle, stomach and intestines, and 50% in the hepatopancreas. All six affected organs (i.e., gills, labial palps, mantle, stomach, intestines, and hepatopancreas) exhibited an inflammatory response that translated into very light to heavy hemocytic infiltrations of the connective and/or epithelial tissues (Figs. 1G, 2I, 3A, 4B, 5B, and 6B ). In the gills, labial palps, and especially the mantle and hepatopancreas, small to very large hemocytic aggregations were frequently observed (Figs. 3B–D and 4C). The infiltrations/aggregations of hemocytes were followed by the appearance of hyperplasia and/or hypertrophy in the epithelia of the gills, labial palps, mantle, digestive diverticulae of the hepatopancreas, stomach, and intestines (Figs. 1H, 2G, 5C, and 6C). Small to very large thrombi formed in the gills, labial palps, mantle, and to a lesser extent the hepatopancreas (Figs. 1I, 2G, and 3D), and small to relatively large edemas that occasionally became thrombosed were frequently observed in the gills (Fig. 1F and I), labial palps (Fig. 2G and H), mantle (Fig. 3D), and sometimes in the epithelium of the intestines (Fig. 6C). These inflammatory pathologies were accompanied by a progressive onset of degenerative pathologies in the gills, labila palps, mantle, stomach, intestines, and hepatopancreas. The matting and exfoliation of the ciliary structures of the affected organs was followed by epithelial desquamation (Figs. 1B–D and I, 2C, D, and I, 3E, 4D, 5D, 6D), the appearance of necrotic epithelial and/or connective tissues, then atrophy (Figs. 1J, 2J, 3E and F, 4D and F, 5D, 6D). Preceding the atrophy of the stomach and intestine epithelia, and the functional atrophy of the digestive canals of the hepatopancreas, hemocytes migrated via diapedesis to the lumen of these organs and formed light hemorrhage in the primary canals and digestive diverticulae of the hepatopancreas (Fig. 4D and E), and light to heavy hemorrhaging in the lumen of the stomach and intestines (Figs. 5C and D, 6B). Only in a very few individuals, were defensive melanization in the epithelium of mussel gills (Fig. 1H), and the regenerative fibrosis of the connective tissue of the mantle (Fig. 3B) and the basal epithelium of the digestive diverticulae observed (Fig. 4D).

40

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Fig. 1. Scanning (A–D) and light (E–J) micrographs of gills of Mytilus galloprovincialis exposed to Heterocapsa circularisquama. (A and E) Controls and (B–D and F–J) exposed. A: atrophy, BL: basal lamina, CE: cilia exfoliation, E: epithelium, ED: epithelial desquamation, FC: frontal cilia, Ft: filament thinning, H: hemocyte, HI: hemocytic infiltration, HP: hyperplasia, HT: hypertrophy, IFJ: interfilamentar junction, LC: lateral cilia, LFC: latero-frontal cilia, M: mucus, N: necrosis, TH: thrombosed edema, ( ) edema, and () melanization.

3.2. Effects of duration of exposure, density of H. ciruclarisquama, and temperature on the total prevalence and total intensity of pathology The total prevalence of pathology (P) and total intensity of pathology (I) were significantly affected by the cell density of H. circularisquama, the exposure duration, and temperature (Fig. 7 and Table 2). The three evaluation parameters showed paired synergistic effects on (I), but did not have any significant interaction-effect on (P). At 15 ◦ C, 3 h of exposure to H. circularisquma had no effect on either (P) or (I) (Fig. 7A and D). After 24 h, exposures to 500 and 1 × 103 cells ml−1 induced a significant increase of (P) and (I) to intermediate prevalence with very light intensity. A 48 h-exposure to 50 cells ml−1 induced a significant increase of (P) to an intermediate prevalence, but not (I). Exposures to 500 and 1 × 103 cells ml−1 for 48 h induced (P) and (I) to increase to intermediate prevalence with very light intensity and moderately-high prevalence with light intensity, respectively. A further increase in temperature by 5 and 10 ◦ C caused a significant increase in the (P) and (I) for all cell densities and exposure durations as compared to the control groups (Fig. 7B–F), with the highest (P) and (I) registered following 48 h of exposure to 1 × 103 cells ml−1 at 25 ◦ C. At 20 ◦ C, exposures to 5–1 × 103 cells ml−1 induced the pathologies to occur at low (P) with very light (I) to moderately-high (P) with light (I), intermediate (P) with very light (I) to moderately-high (P) with light (I), and intermediate (P) with light (I) to high (P) with moderate (I), following 3 h, 24 h, and 48 h, respectively. At 25◦ C, exposures to 5–1 × 103 cells ml−1 induced

the pathologies to occur at intermediate (P) with very light (I) to intermediate (P) with light (I), moderately-high (P) with light (I) to high (P) with lightly-moderate (I), and high (P) with light (I) to high (P) with moderate (I), following 3 h, 24 h, and 48 h of exposure, respectively. 3.3. Effects of duration of exposure, density of H. ciruclarisquama, and temperature on the prevalence and intensity of pathology in each organ The prevalence (Fig. 8) and intensity (Fig. 9) of pathology in each of the six affected organs (Pi and Ii) were significantly affected by the duration of exposure, cell density of H. circularisquama, and temperature as compared to their respective controls (Table 2), with a clear trend of increase at higher levels of the three parameters of evaluation. The latter three interacted synergistically in (Pg) and (Ist). A few paired synergistic interactions between these parameters of evaluation were found but with no clear trend. At 15 ◦ C, the (Pi) were significantly increased following 24 h and 48 h of exposures to 500 and 1 × 103 cells ml−1 in all organs, except for the (Ph) that was significant only for exposure at 1 × 103 cells ml−1 . At 50 cells ml−1 , the (Plp) and the (Pit) were significantly increased following 24 h, and 24 h, and 48 h, respectively. The highest prevalence was registered for (Pm) (high), followed by (Pg) and (Plp) (moderately-high), the (Ph) (intermediate), and (Pst) and (Pit) (intermediate). The (Ii) followed a similar trend of increase with cell densities and exposure durations, the heaviest intensity was registered for (Im) (light), but followed by (Ilp) and (Iit) (light), then (Ig), (Ih) and (Ist) (very light).

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

41

Fig. 2. Scanning (A–D) and light (E–J) micrographs of labial palps of M. galloprovincialis exposed to H. circularisquama. (A, B and E, F) Controls and (C, D and G–J) exposed. CE: cilia exfoliation, CT: connective tissue, CuE: cubic epithelium, CyE: cylindrical epithelium, ED: epithelium desquamation, HI: hemocytic infiltration, Hp: hyperplasia, LC: long cilia, N: necrosis, SC: short cilia, TE: thrombosed edema, and TH: thrombus.

At 20 ◦ C, exposure to 50–1 × 103 cells ml−1 induced a significant increase in the (Pi) in all affected six organs within 3 h. Following 24 h and 48 h of exposure to 5 cells ml−1 , the prevalence of pathology did also increase significantly, except for the (Pm). The

highest prevalence was registered for (Pg) (high), (Plp), (Pm), and (Ph) (high), followed by (Pst) and (Pit) (high). The (Ii) followed a trend similar to the (Pi) except that it was not significant in the intestines following exposure to 5 cells ml−1 . The heaviest at

Fig. 3. Light micrographs of mantle of M. galloprovincialis exposed to H. circularisquama. (A) Control and (B–F) exposed. A: reproductive acini, At: atrophy, F: fibrosis, Ha: hemocytic aggregation, HI: hemocytic infiltration, Th: thrombus, and N: necrosis.

42

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Fig. 4. Light micrographs of digestive diverticula of M. galloprovincialis exposed to H. circularisquama. (A) Control and (B–F) exposed. A: atrophy, BC: basophilic cells, C: cilia, CE: cilia exfoliation, DD: digestive diverticulum, DT: digestive tubule, F: fibrosis, H: hemocyte, Ha: hemocytic aggregation, HI: hemocytic infiltration, L: lumen, N: necrosis, PC: primary canal, and SC: secretory cells.

25 ◦ C, the (Pi) were significant for all densities and exposure durations, except for the (Ph) and (Pst) following 3 h of exposure to 5 cells ml−1 . A 48 h of exposure to 5 cells ml−1 and 24 h of exposure to 1 × 103 cells ml−1 resulted in the occurrence of all pathologies in the gills and intestines of mussel, respectively. Subsequently, the highest prevalence was recorded for (Pg) and (Pit) (high), followed by (Plp), (Pm), and (Pst) (high), then (Ph) (high). The (Ig), (Ilp), and

(Iit) were all significantly different from the respective control for all combinations of cell density-exposure durations. A 3 h of exposure to 5 cells ml−1 and 5–500 cells ml−1 had no effect on the (Iit) and (Ist), respectively. Exposure to 5 cells ml−1 and 3 h-exposure to 50 cells ml−1 did not affect the (Ih). The heaviest intensity was recorded for (Ig) (moderately-heavy), followed by (Ilp) and (Im) (moderate), (Ih) and (Iit) (moderate), then (Ist) (lightly-moderate).

Fig. 5. Light micrographs of stomach of M. galloprovincialis exposed to H. circularisquama. (A) Contro and (B–D) exposed. C: cilia, CE: cilia exfoliation, CT: connective tissue, E: epithelium, ED: epithelial desquamation, H: hemocyte, HI: hemocytic infiltration, Hm: hemorrhage, Hp: hyperplasia, Ht: hypertrophy, L: lumen, and N: necrosis.

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

43

Fig. 6. Light micrographs of intestines of M. galloprovincialis exposed to H. circularisquama. (A) Control and (B–E) exposed. E: epithelium, C: cilia, CE: cilia exfoliation, D: diapedesis, ED: epithelial desquamation, H: hemorrhage, HI: hemocytic infiltration, L: lumen, M: muscle fiber, Mv: microvilli, N: necrosis, and T: thrombus.

3.4. Comparisons between exposed groups of the effects of duration of exposure, density of H. ciruclarisquama, and temperature on the prevalence and intensity of pathology Cross-comparisons of the effects of temperature, exposure duration, and cell density of H. circularisquama between exposed groups were conducted. For temperature, all exposed groups were

significantly different in prevalence and intensity, except for the (Pst) and (Ist) for 20 ◦ C vs 25 ◦ C. For the exposure duration, all groups were significantly different in prevalence and intensity, except for (Pit), (Ist), and (Iit) for 24 h vs 48 h. The (P) and (Pi) were not significantly different between the 500 cells ml−1 and 1 × 103 cells ml−1 groups, except for (Pg) and (Ph) (Fig. 10), but the (I) and (Ii) were significantly different among

Fig. 7. Time-, temperature-, and cell density-dependent changes in the total prevalence (P), (A–C) and total intensity (I), (D–F) of pathology in M. galloprovincilais following exposure to H. circularisquama. (A, D) Exposure at 15 ◦ C, (B and E) exposure at 20 ◦ C, and (C and F) exposure at 25 ◦ C. (*) Indicates significant difference from respective controls (P < 0.05).

44

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Fig. 8. Time-, temperature-, and cell density-dependent changes in the prevalence of pathologies in each organ (Pi) of M. galloprovincilais following exposure to H. circularisquama. (A–C) Gills, (D–F) labial palps, (G–I) mantle, (J–L) hepatopancreas, (M–O) stomach, and (P–R) intestines. (A, D, G, J, M, and P) Exposure at 15 ◦ C, (B, E, H, K, N, and Q) exposure at 20 ◦ C, and (C, F, I, L, O, and R) exposure at 25 ◦ C. (*) Indicates significant difference from respective controls (P < 0.05).

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

45

Fig. 9. Time-, temperature-, and cell density-dependent changes in the intensity of pathologies in each organ (Ii) of M. galloprovincilais following exposure to H. circularisquama. (A–C) Gills, (D–F) labial palps, (G–I) mantle, (J–L) hepatopancreas, (M–O) stomach, and (P–R) intestines. (A, D, G, J, M, and P) Exposure at 15 ◦ C, (B, E, H, K, N, and Q) exposure at 20 ◦ C, and (C, F, I, L, O, and R) exposure at 25 ◦ C. (*) Indicates significant difference from respective controls (P < 0.05).

46

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Table 2 Factorial analysis (ANOVA) of the total prevalence, total intensity, and the prevalence and intensity of pathologies in each of the six organs of the Mediterranean mussel, M. galloprovincialis, exposed to the harmful dinoflagellate, H. circulariquama. Parameter of evaluation

Prevalence

Intensity

T

DH

ED

T × DH

T × ED

DH × ED

T × DH × ED

Gills (Pg) Labial palps (Plp) Mantle (Pm) Hepatopancreas (Ph) Stomach (Pst) Intestines (Pi)

*** *** *** *** *** ***

*** *** *** *** *** ***

*** *** *** *** *** ***

NS NS NS NS NS NS

NS NS * NS NS NS

NS NS NS * NS *

* NS NS NS NS NS

Total (P)

***

***

***

NS

NS

NS

NS

Gills (Ig) Labial palps (Ilp) Mantle (Im) Hepatopancreas (Ih) Stomach (Ist) Intestines (Ii)

*** *** *** *** *** ***

*** *** *** *** *** ***

*** *** *** *** *** ***

*** NS NS ** * NS

NS *** * * NS NS

NS * NS *** NS ***

NS NS NS NS * NS

Total (I)

***

***

***

**

*

***

NS

T: temperature, DH: cell density of H. circularisquama, ED: exposure duration, ×: interaction, and NS: Non-significant effect. *P < 0.05, **P < 0.01, and ***P < 0.001.

all cell density groups, except for (Im) 5 cells ml−1 vs 50 cells ml−1 , (Ih) 50 cells ml−1 vs 500 cells ml−1 , and (Ist) 500 cells ml−1 vs 1 × 103 cells ml−1 . The densities of 5 and 50 cells ml−1 equally affected the (Pg), (Plp), (Pm), and (Im), and the densities of 50 and 500 cells ml−1 similarly affected the (Pm), (Ph), and (Ih).

3.5. Comparisons of the prevalence and intensity of pathology between organs Cross-comparisons of the prevalence and intensity of pathology between the six affected organs, for all treatments, are given in Table 3. The (Pg) and (Ig) were significantly higher than those

Fig. 10. Cross-comparison of the effects of the density of H. circularisquama on the total prevalence (P) and intensity (I) of pathologies and the prevalence (Pi) and intensity (Ii) of pathologies in each organ. (A) Prevalence and (B) intensity. (*) Indicates significant difference between exposure groups (P < 0.05). (**) Indicates significant difference between exposure groups (P < 0.01). (***) Indicates significant difference between exposure groups (P < 0.001).

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

47

Table 3 Comparison (Student’s T-test) of the prevalence and intensity of pathology between organs of the Mediterranean mussel, Mytilus galloprovincialis, exposed to the harmful dinoflagellate, Heterocapsa circulariquama. Prevalence Organs

Gills

Labial palps

Gills Labial palps Mantle Hepatopancreas Stomach Intestines

* ** *** ** *

NS *** NS NS

Intensity Mantle

*** NS NS

Hepato-pancreas

** **

Stomach

NS

Intestines

Gills

Labial palps

Mantle

Hepato-pancreas

Stomach

*** ** *** *** ***

NS *** ** **

*** *** **

NS **

NS

Intestines

NS: non-significant effect. *P < 0.05, **P < 0.01, and ***P < 0.001.

of the other five organs. The (Plp), (Pm), (Pst), and (Pit) were not significantly different, but were significantly higher than (Ph). The (Ilp) and (Im) were not significantly different, but were higher than (Ist) and (Iit) which were equal. The (Ih) was the lowest. 4. Discussion The detailed histopathological investigation revealed that H. circularisquama affects the gills, labial palps, mantle, stomach, intestines, and hepatopancreas of mussels, and that these effects are influenced by cell density, duration of exposure, and temperature. A significant increase in the total prevalence of pathology and total intensity of pathology is reported at 15 ◦ C within 24 h of exposure to 500 cells ml−1 . At 20 ◦ C and 25 ◦ C, a significant increase in the prevalence and intensity of pathology were observed at as low as 5 cells and within just 3 h of exposure. The gill was the organ most affected by exposure to H. circularisquama, followed by the labial palps and mantle (equally affected), the stomach and intestines (equally affected), and the hepatopancreas (the least affected). Several species of harmful dinoflagellates are known to cause pathological damage in bivalve molluscs. Pathologies in gills of various bivalve species range from inflammatory responses to necrosis following exposure to the harmful dinoflagellates Alexandrium minutum, A. tamarense, A. fundyense, A. monilatum, Prorocentrum minimum, Karenia brevis, K. mikimotoi, Gymnodinium catenatum, and Cochlodinium polykrikoides (reviewed in Basti et al., 2011a). Laboratory exposures of bivalves to harmful algae also induced pathologies in other organs. Inflammation in the kidney and marked cellular damage to the guts of bay scallops, Argopecten irradians, with decreased height of absorptive cells and increased lumen diameter (Smolowitz and Shumway, 1997), mantle and gill lesions in eastern oysters, Crassostrea virginica (Nielsen and Strømberg, 1991), and marked cellular damage in the guts of blue mussels, Mytilus edulis, were caused by exposure to K. mikimotoi, formerly known as Gyrodinium aureolum (Widdows et al., 1979). Digestive gland lesions were also observed in scallops, Pecten alba, induced by exposure to the diatom, Rhizosolenia chunii (Parry et al., 1989). The stomach, intestines, digestive diverticula, and mantle of M. edulis, showed hemocytic infiltration, focal inflammatory responses, red granules, increased infestation with crenoids and trematods, and ova degeneration following exposure to A. fundyense (Galimany et al., 2008a). Exposure to the STX-producer (saxitoxin) G. catenatum induced melanization in the labial palps and mantle, and aggregation of hemocytes in the mantle and digestive gland of giant lion’s paw, Nodipecten subnodosus (Estrada et al., 2007a,b; Hallegraeff et al., 2012). Hemocytic infiltration in the digestive glands of M. edulis, was induced by exposure to Karlodinium veneficum (Galimany et al., 2008b). Increased hemocytes, presence of brown cells, phagocytosis, pycnosis, thin and dilated tubules, and sloughing of cells were observed in the guts of spats of Pacific oyster, C. gigas, exposed to a simulated bloom of P. rhathymum for 21 days at 104 cells ml−1 (Pearce et al., 2005). Five-day exposure of hard clams, Mercenaria mercenaria, to P. minimum at

2 × 104 cells ml−1 caused hemocytic infiltration and aggregations with granulomas in the kidneys, gills, foot, heart, and pericardial sac (Wikfors and Smolowitz, 1993; Landberg, 2002). In M. edulis, exposure to P. minimum resulted in hemocytic infiltration, diapedesis, and elevated bacteria in the stomach and intestines, and hemocytic aggregation in the gonads (Hégaret and Wikfors, 2005a,b; Galimany et al., 2008b; Hégaret et al., 2010). Hemorrhage and squamation in the digestive tracts of A. irradians, were observed following exposure to C. polykrikoides (Gobler et al., 2008). A broad spectrum of pathologies is thus, induced by a variety of harmful dinoflagellates, each of which produces a set of different toxins, alleochemicals, and bioactive secondary metabolites. Seemingly, a high number of HAB species studied to date causes a wide range of pathologies in several organs of bivalves denoting a high variability of end effects in relation with the harmful compounds produced by the harmful alga. Based on the results of this study and our previous studies, H. circularisquama induces highly potent and conserved cytotoxic activity in all non-muscular organs, at least in the six organs affected in this study, specifically targeting bivalve filter feeders and to a lesser extent filter-feeding gastropods (Basti et al., 2011a; Matsuyama, 2012). As previously described by Basti et al. (2011a) for gills of clams, R. philippinarum, following exposure to H. circularisquama (1 × 103 cells ml−1 , 3–72 h of exposure, 20 ◦ C), all affected organs showed an onset of defensive and inflammatory immune responses that ultimately failed to counteract the extensive cytotoxicity and/or hemolytic activity of H. circularisquama which resulted in cilia exfoliation, epithelial desquamation, necrosis, and atrophy. Other studies of the toxicity of H. circularisquama showed that bivalves are impacted at all their life stages, from gametes, fertilization, embryos, larvae, juveniles, to adults (Matsuyama, 1999a,b, 2003a,b, 2012; Matsuyama et al., 1996, 1997a,b, 1998, 1999, 2001; Matsuyama and Honjo, 2001; Basti and Segawa, 2010; Basti et al., 2011a,b, 2013). In addition, H. circularisquama was shown to affect the valve movement behavior of several bivalve species with an increased valve adduction related to an altered activity of the adductor muscles (Nagai et al., 2006; Basti et al., 2009). These studies strongly suggested that the broad negative effects of H. circularisquama could be related to disrupted cellular homeostasis, especially cytosolic calcium regulation (Matsuyama, 2003a, 2012; Basti et al., 2013). The fact that neither the adductor muscles, nor the exhalent siphons and the foot were affected in our study suggests that either effects of H. circularisquama on calcium regulations were not strong enough to induce observable pathologies in the three muscular organs, and/or that other/additional cellular receptor(s), more sensitive than the ones implicated in Ca2+ regulation, are targeted by the toxins of H. circularisquama in the gills, labial palps, mantle, stomach, intestines, and hepatopancreas. The active compounds from H. circularisquama have not been well characterized due to high instability of the extracts hampering attempts to clarify its toxicity. Cell extracts of H. circularisquama with 80% methanol, acetone, hexane, 0.5 M-acetic acid, 1:1 v chloroform:methanol, and water were not toxic to M. galloprovincialis, in spite of the high toxicity of the cell culture (Matsuyama, 2003b). No

48

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

effects were observed in M. galloprovincialis following exposure to culture filtrate or cells of H. circularisquama that lost their cell wall after physical treatments, and following chemical treatment with several surfactants, enzymes, and inhibitors of synthesis. Nonetheless, the toxicity of H. circularisquama to M. galloprovincialis was lost when treated with glycol-chain scavengers, protein decomposers, and glycol–protein synthesis inhibitors. These results indicate that the toxins of H. circularisquama are labile glycol–protein agents located on the cell surface (Matsuyama et al., 1997; Matsuyama, 2003b; reviewed in Matsuyama, 2012). Cell contact between H. circularisquama and other phytoplankton species and bivalves/early development stages was shown to be the process by which H. circularisquama induces toxicity (Uchida et al., 1995; Yamasaki et al., 2011; Basti et al., 2011b, 2013). In a previous study, timedependent, extensive necrotic effects of H. circuarisquama in gills of R. philippinarum as a result of cell-contact was demonstrated (Basti et al., 2011a), which seems to be additionally confirmed, in this study, for the other two external pallial organs, the labial palps and the mantle. The effects in the internal organs of mussels (the stomach, intestines, and hepatopancreas), however, suggest that additional processes should have been associated with the cytotoxicity observed, since cells of H. circularisquama easily lose their cell scales following physical contact or disturbance, which would have been the case following contact with the pallial organs of the mussels, and then following ingestion. Potent hemolytic activity of H. circularisquama towards erythrocytes of three mammalian lines was demonstrated, and both cell cultures and ethanol extracts of H. circularisquama (1%, 50%, and 100% ethanol extracts) showed time-dependent hemolytic activity towards the mammalian erythrocytes, affected membrane of eggs of the Pacific oyster, C. gigas, and were lethal to the rotifer, Brachionus plicatilis; however, the effects of the ethanol extracts were far less potent than those of the whole culture (Oda et al., 2001). A photosensitizing hemolytic and cytotoxic compound, a porphyrin derivative labeled H2-a, was purified and characterized from methanol extracts of H. circularisquama and found to be toxic to both the rotifer B. plicatilis and the short-neck clam, R. philippinarum (Miyazaki et al., 2005). The hemolytic activity in rabbit erythrocytes, the most sensitive mammalian erythrocytes, was confirmed in several cell culture strains of H. circularisquama (Kim et al., 2002). The culture supernatant and the resuspended cell pellet of the strain of H. circularisquama with the highest hemolytic activity were further tested to determine the location of the hemolytic toxin. The whole culture showed the highest hemolytic activity, followed by the supernatant which activity, contrary to the whole culture, decreased rapidly with time. The cell pellets on the other hand showed no such hemolytic activity and the cells of H. circularisquama were morphologically changed following discharge of the cell scales; however, within 24 h both the original morphology and the hemolytic activity were recovered showing that the hemolytic toxins were unstable, loosely attached to the cell surface of H. circularisquama, and continuously released in the medium during cell growth (Kim et al., 2002). Hemolytic effects of H. circularisquama in bivalves, however, have not been fully demonstrated, although exposure of adult R. philippinarum to cell cultures of H. circularisquama altered the hemocytic responses of clams (Basti et al., 2011a). In 2002, a new digalactosyl diacylglycerol and two known monogalactosyl diacylglycerols were successfully extracted and purified from H. circularisquama, and shown to induce cytolysis of heart cell suspension of C. gigas (Hiraga et al., 2002). Potent, cell density-dependent cytotoxicity of cell culture and culture filtrate of H. circularisquama was later confirmed in Vero cells, providing new clues supporting both cell-contact dependent toxicity of H. circularisquama but also the production of cytotoxic exotoxins. Additionally, in 2008, the necrotic activity of H2-a was confirmed in HeLa cells (Kim et al., 2008). Although the mediation of necrosis

via cell-contact, and the trade-offs between the hemolytic toxins, the cytotoxins, and other uncharacterized exotoxins await further clarifications, it is clear that, in the present study, all the organs associated with vital physiological functions were affected, and that the gills were the most affected, followed by the mantle and labial palps, the stomach and intestines, and the hepatopancreas. The sequence of organ damage suggests that the mussels, and probably other bivalve species, die following extensive toxicity that alters mainly respiration, then feeding and digestion, and possibly detoxification. Further studies should explore alternative pathways of the toxins in bivalves. In this laboratory study, the cytotoxicity of H. circularisquama is once again shown to be dependent of the cell density, duration of exposure, and water temperature, and deleterious effects were observed at low levels of all three factors. Until recently, mortalities of bivalves in the wild were believed to occur when water temperatures exceeded 23 ◦ C (Matsuyama, 2003a, 2012); however, both recent laboratory study (Basti and Segawa, 2010) and recent field observations reported die-offs of Pacific oyster at 15 ◦ C (Kondo et al., 2012; Fujimoto et al., 2013). The first bloom of H. circularisquama occurred in Uranochi bay in 1988, and a rapid expansion of its recurrent blooms and associated bivalve kills have been subsequently reported from several localities in Western Japan (Matsuyama, 2012). In 2009, the first bloom of H. circularisquama in Lake Kamo, in Niigata Prefecture (located in Eastern Japan), caused mass mortality of bivalves at 15 ◦ C and proved that not only H. circularisquama is lethal at lower temperatures than previously described, but that its geographical distribution is on a Northward expansion to colder waters. The sea surface temperature (SST) in Japan has increased by 0.8–1.7 ◦ C over the past century, and is projected to further increase by 0.6–3.1 ◦ C over the 21st century (Sato et al., 2006; Japan Meteorological Agency, 2008). Such an increase of the SST around Japan, with the well-documented expansion and increased intensity and duration of H. circularisquama blooms with increased SST during winter (Matsuyama, 2003a, 2012), are expected to exhibit effects on wild and cultured bivalve populations around Japan. Other ecologically and economically important species of bivalves, whose geographical distributions are located North of H. circularisquama bloom distribution, seem vulnerable to near-future temperatures associated with climate change. Indeed, several species of harmful and toxic Alexandrium have already been shown to have increased in Japanese coastal waters in correlation with increased SST (Nagai et al., 2010) and H. circularisquama seems to be following a similar trend.

5. Conclusion Exposure to the harmful dinoflagellate, H. circularisquama, was shown to affect Mediterranean mussels, M. galloprovincialis, at low density, low temperature, and short duration of exposure. Several inflammatory, defensive, regenerative, and degenerative pathologies were identified in the gills, labial palps, mantle, stomach, intestines, and hepatopancreas of exposed mussels, but exhalent siphons, adductor muscles, and foot were not affected. The prevalence and intensity of the pathologies were significantly increased with increasing cell density, temperature, and duration of exposure. The organ most affected was the gill, followed by the labial palps and mantle, the stomach and intestines, and the hepatopancreas. This study showed that compromised respiration, feeding, and digestion advanced by the broad cytotoxicity translating into necrosis and atrophy, paralleled with hemorrhaging in the affected organs, would have been the mechanisms by which the highly toxic H. circularisquama causes death in mussels, and other mollusc bivalves. The cell-contact dependent toxicity of H. circularisquama does not seem to be the only mechanism of toxicity.

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Acknowledgments We would like to thank the Pearl Research Laboratory of K. Mikimoto & Co. Ltd., for providing H. circularisquama. The research was funded by the Ministry of Education, Culture, Sports, Science and Technology, Japan, through a grant to Basti & Segawa, and from Tokyo University of Marine Science and Technology to Basti & Tanaka, under the young strategic researcher program. References Anderson, D.M., Glibert, P.M., Burkholder, J.M., 2002. Harmful algal blooms and eutrophication nutrient sources, composition, and consequences. Estuaries 25 (4b), 704–726. Basti, L., Nagai, K., Tanaka, Y., Segawa, S., 2013. Sensitivity of gamates, fertilization, and embryo development of the Japanse pearl oyster Pinctada fucata martensii, to the harmful dinoflagellate, Heterocapa circularisquama. Mar. Biol. 160, 211–219. Basti, L., Endo, M., Segawa, S., 2011a. Physiological, pathological, and defense alterations in Manila clams (short-neck clams), Ruditapes philippinarum, induced by Heterocapsa circularisquama. J. Shellfish Res. 30, 829–844. Basti, L.G.J., Higuchi, K., Nagai, K., Segawa, S., 2011b. Effects of the toxic dinoflagellate Heterocapsa circularisquama on larvae of the pearl oyster Pinctada fucata martensii (Dunker, 1873). J. Shellfish Res. 30, 177–186. Basti, L., Segawa, S., 2010. Mortalities of the short-neck clam Ruditapes philippinarum induced by the toxic dinoflagellate Heterocapsa circularisquama. Fish. Sci. 76, 625–631. Basti, L., Nagai, K., Shimasaki, Y., Oshima, Y., Honjo, T., Segawa, S., 2009. Effects of the toxic dinoflagellate Heterocapsa circularisquama on the valve movement behavior of the Manila clam Ruditapes philippinarum. Aquaculture 291, 41–47. Bricelj, V.M., MacQuarrie, S.P., 2007. Effects of brown tide (Aureococcus anophagefferens) on hard clam Mercenaria mercenaria larvae and implications for benthic recruitment. Mar. Ecol. Prog. Ser. 331, 147–159. Estrada, N.A., Lagos, N., García, C., Maeda-Martínez, A.N., Ascencio, F., 2007a. Effects of the toxic dinoflagellate Gymnodinium catenatum on uptake and fate of paralytic shellfish poisons in the Pacific giant lion-paw scallop Nodipecten subnodosus. Mar. Biol. 151, 1205–1214. Estrada, N., Romeao, M.J., Campa-Córdova, A., Luna, A., Ascencio, F., 2007b. Effects of the toxic dinoflagellate, Gymnodinium catenatum on hydrolytic and antioxidant enzymes, in tissues of the giant lion-paw scallop Nodipecten subnodosus. Comp. Biochem. Physiol. C 146, 502–510. Fujimoto, A., Kondo, S., Kakao, R., Tomaru, Y., Nagazaki, K., 2013. Co-occurrence of Heterocapsa circularisquama bloom and its lytic viruses in lake Kamo, Japan, 2010. JARQ 47, 329–338. Gainey, L.F.J., Shumway, S.E., 1988. A compendium of the responses of bivalve molluscs to toxic dinoflagellates. J. Shellfish Res. 7, 623–628. Galimany, E., Sulina, I., Hégaret, H., Ramón, M., Wikfors, G.H., 2008a. Experimental exposure of the blue mussel (Mytilus edulis L.) to the toxic dinoflagellate Alexandrium fundyense: histology, immune responses, and recovery. Harmful Algae 7, 702–711. Galimany, E., Sulina, I., Hégaret, H., Ramón, M., Wikfors, G.H., 2008b. Pathology and immune responses of the blue mussel (Mytilus edulis L.) after an exposure to the harmful dinoflagellate Prorocentrum minimum. Harmful Algae 7, 630–638. Gobler, C.J., Berry, D.L., Anderson, O.R., Burson, A., Koch, F., Rodgers, B.S., Moore, L.K., Goleski, J.A., Allam, B., Browser, P., Tang, Y., Nuzzi, R., 2008. Characterization, dynamics, and ecological impacts of harmful Cochlodinium polykrikroides blooms on eastern Long Island, NY, USA. Harmful Algae 7, 293–307. Hallegraeff, G.M., Blackburn, S.I., Doblin, M.A., Bolch, C.J.S., 2012. Global toxicology, ecophysiology and population relationships of the chain forming PST dinoflagellate Gymnodinium catenatum. Harmful Algae 14, 130–143. Hégaret, H., Wikfors, G.H., 2005a. Effects of natural and field-simulated blooms of the dinoflagellate Prorocentrum minimum upon hemocytes of the Eastern oysters, Crassostrea virginica, from two different populations. Harmful Algae 4, 201–209. Hégaret, H., Wikfors, G.H., 2005b. Time-dependent changes in hemocytes of Eastern oysters, Crassostrea virginica, and northern bay scallops, Argopecten irradians, exposed to a culture strain of Prorocentrum minimum. Harmful Algae 4, 187–199. Hégaret, H., Wikfors, G.H., Shumway, S.E., 2007. Diverse feeding responses of five species of bivalve mollusc when exposed to three species of harmful algae. J. Shellfish Res. 26, 549–559. Hégaret, H., Smolowitz, R.M., Sunila, I., Shumway, S.E., Alix, J., Dixon, M., Wikfors, G.H., 2010. Combined effects of a parasite, QPX, and the harmful-alga, Prorocentrum minimum on northern quahogs, Mercenaria mercenaria. Mar. Environ. Res. 69, 337–344. Hiraga, Y., Kaku, K., Omoda, D., Sugihara, K., Hosoya, H., Hino, M., 2002. A new digalactosyl diacylglycerol from a cultured marine dinoflagellate Heterocapsa circularisquama. J. Nat. Prod. 65, 1494–1496. Horiguchi, T., 1995. Heterocapsa circularisquama sp. nov. (Peridiniales Dinophyceae): a new marine dinoflagellate causing mass mortality of bivalves in Japan. Phycol. Res. 43, 129–136. Japan Meteorological Agency, 2008. Global Warming Projection, vol. 7, 59 pp.

49

Kim, D., Oda, T., Muramatsu, T., Kim, D., Matsuyama, Y., Honjo, T., 2002. Possible factors responsible for the toxicity of Cochlodinium polykrikoides, a red tide phytoplankton. Comp. Biochem. Physiol. C 132, 415–423. Kim, D., Miyazaki, Y., Nakashima, T., Iwasi, T., Fujita, T., Yamaguchi, K., Choi, K.S., Oda, T., 2008. Cytotoxic action mode of a novel porphyrin derivative isolated from harmful red tide dinoflagellate Heterocapsa circularisquama. J. Biochem. Mol. Toxicol. 22, 158–165. Kondo, S., Nakao, R., Iwataki, M., Sakamoto, S., Itakura, S., Matsuyama, Y., Nagasaki, K., 2012. Heterocapsa circularisquama coming up north–Mass mortality of Pacific oysters due to its blooming at Lake Kamo in Sado Island, Japan. Nippon Suisan Gakkaishi 78, 719–725 (in Japanese with English abstract). Landberg, J.H., 2002. The effects of harmful algal blooms on aquatic organisms. Rev. Fish. Sci. 10, 113–390. Leverone, J.R., Blake, N.J., Pierce, R.H., Shumway, S.E., 2006. Effects of the dinoflagellate Karenia brevis on larval development in three species of bivalve molluscs from Florida. Toxicon 48, 75–84. Matsuyama, Y., 1999a. Harmful effect of dinoflagellate Heterocapsa circularisquama on shellfish aquaculture in Japan. Jpn. Agric. Res. Q. 33, 283–293. Matsuyama, Y., 1999b. The toxic effects of Heterocapsa circularisquama on bivalve mollusks. Bull. Plankton Soc. Jpn. 46, 157–160. Matsuyama, Y., 2003a. Physiological and ecological studies on harmful dinoflagellate Heterocapsa circularisquama. II. Clarification on the toxicity of H. circularisquama and its mechanisms causing shellfish kills. Bull. Fish. Res. Agen. 9, 13–117 (in Japanese with English abstract). Matsuyama, Y., 2003b. Physiological and ecological studies on harmful dinoflagellate Heterocapsa circularisquama. I. Elucidation of environmental factors underlying the occurrence and development of H. circularisquama red tide. Bull. Fish. Res. Agen. 7, 24–105 (in Japanese with English abstract). Matsuyama, Y., 2012. Impacts of the harmful dinoflagellate Heterocapsa circularisquama bloom on shellfish aquaculture in Japan and some experimental studies on invertebrates. Harmful Algae 14, 144–155. Matsuyama, Y., Honjo, T., 2001. Hemolytic activity of Heterocapsa circularisquama (Dinophyceae) and its possible involvement in shellfish toxicity. J. Phycol. 37, 509–516. Matsuyama, Y., Uchida, T., Nagai, K., Ishimura, M., Nishimura, A., Yamaguchi, M., Honjo, T., 1996. Biological and environmental aspects of noxious dinoflagellate blooms by Heterocapsa circularisquama in the west Japan. In: Yasumoto, T., Oshima, Y., Fukuyo, Y. (Eds.), Harmful and Toxic Algal Blooms. Intergovernmental Oceanographic Commission of UNESCO, Paris, pp. 247–250. Matsuyama, Y., Kimura, A., Fujii, H., Takayama, H., Uchida, T., 1997a. Occurrence of Heterocapsa circularisquama red tide and subsequent damage to shellfish in western Hiroshima Bay Seto Inland Sea Japan in 1995. Bull. Nansei Natl. Fish. Res. Inst. 30, 189–207. Matsuyama, Y., Uchida, T., Honjo, T., 1997b. Toxic effects of the dinoflagellate Heterocapsa circularisquama on clearance rate of the blue mussel Mytilus galloprovincialis. Mar. Ecol. Prog. Ser. 146, 73–80. Matsuyama, Y., Uchida, T., Honjo, T., 1998. The harmful effect of red tide dinoflagellates, Heterocapsa circularisquama and Gymnodinium mikimotoi on the clearance rate and survival of the blue mussel, Mytilus galloprovincialis. In: Reguera, B., Blanco, J., Fernandez, M., Wyatt, T. (Eds.), Harmful Algae. Xunta de Gracilia and IOC of UNESCO, Paris, pp. 422–424. Matsuyama, Y., Uchida, T., Honjo, T., 1999. Effect of harmful dinoflagellates, Gymnodinium mikimotoi and Heterocapsa circularisquama, red-tide on filtering rate of bivalve molluscs. Fish. Sci. 65, 248–253. Matsuyama, Y., Usuki, H., Uchida, T., Kotani, Y., 2001. Effects of harmful algae on the early planktonic larvae of the oyster, Crassostrea gigas. In: Hallegraeff, G., Blackburn, S., Bolch, C., Lewis, R. (Eds.), Harmful Algal Blooms. IOC, UNESCO, Paris, pp. 411–414. Miyazaki, Y., Nakashima, T., Iwashita, T., Fujita, T., Yamaguchi, K., Oda, T., 2005. Purification and characterization of photosensitizing hemolytic toxin from harmful red tide phytoplankton Heterocapsa circularisquama. Aquat. Toxicol. 73, 382–393. Nagai, K., Honjo, T., Go, J., Yamashita, H., Oh, S.J., 2006. Detecting the shellfish killer Heterocapsa circularisquama (Dinophyceae) by measuring bivalve valve activity with a Hall element sensor. Aquaculture 255, 395–401. Nagai, K., Matsuyama, Y., Uchida, T., Yamaguchi, M., Ishimaru, M., Nishimaru, A., Akamatsu, S., Honjo, T., 1996. Toxicity and LD50 levels of the red tide dinoflagellate Heterocapsa circularisquama on juvenile pearl oysters. Aquaculture 144, 149–154. Nagai, S., Yoshida, G., Tarutani, K., 2010. Change in species composition and distribution of algae in the coastal waters of Western Japan. In: Casalegno, S. (Ed.), Global Warming Impacts-case Studies on the Economy, Human Health, and on Urban and Natural Environments. InTech, pp. 209–236. Nielsen, M.V., Strømberg, T., 1991. Shell growth response of mussels (Mytilus edulis) exposed to toxic microalgae. Mar. Biol. 108, 263–267. Oda, T., Sato, Y., Kim, D., Muramatsu, T., Matsuyama, Y., Honjo, T., 2001. Hemolytic activity of Heterocapsa circularisquama (Dinophyceae) an its possible involvement in shellfish toxicity. J. Phycol. 137, 509–516. Parry, G.D., Langdon, J.S., Huisman, J.M., 1989. Toxic effects of a bloom of the diatom Rhizosolenia chunii on shellfish in Port Phillip Bay southeastern Australia. Mar. Biol. 102, 25–41. Pearce, I., Handlinger, J.H., Hallegraeff, G.M., 2005. Histopahtology in pacific oyster (Crassostrea gigas) spat caused by the dinoflagellate Prorocentrum rhathymum. Harmful Algae 4, 61–74.

50

L. Basti et al. / Aquatic Toxicology 163 (2015) 37–50

Sato, Y., Oda, T., Muramatsu, T., Matsuyama, Y., Honjo, T., 2002. Photosensitizing hemolytic toxin in Heterocapsa circularisquama, a newly identified harmful red tide dinoflagellate. Aquat. Toxicol. 56, 191–196. Sato, Y., Yukimoto, S., Tsujino, H., Ishizaki, H., Noda, A., 2006. Response of North Pacific Ocean circulation in a Kuroshio-resolving ocean model to an Arctic oscillation (AO)-like change in Northern Hemisphere Atmospheric circulation due to greenhouse-gas forcing. J. Meteorol. Soc. Jpn. 84, 295–309. Shumway, S.E., 1990. A review of the effects of algal blooms on shellfish and aquaculture. J. World Aquacult. Soc. 21, 65–104. Shumway, S.E., Cucci, T.L., 1987. The effects of the toxic dinoflagellate Protogonyaulax tamarensis on the feeding and behaviour of bivalve molluscs. Aquat. Toxicol. 10, 9–27. Smolowitz, R., Shumway, S.E., 1997. Possible cytotoxic effects of the dinoflagellate Gyrodinium aureolum, on juvenile bivalve molluscs. Aquacult. Int. 5, 291–300. Tran, D., Haberkorn, H., Soudant, P., Ciret, P., Massabuau, J.C., 2010. Behavioral responses of Crassostrea gigas exposed to the harmful algae Alexandrium minutum. Aquaculture 298, 338–345.

Uchida, T., Yamaguchi, M., Matsuyama, Y., Honjo, T., 1995. The bloom dinoflagellate Heterocapsa sp. kills Gyrodinium instriatum by cell contact. Mar. Ecol. Prog. Ser. 118, 301–303. Widdows, J., Moore, M.N., Lowe, D.M., Salkeld, P.N., 1979. Some effects of a dinoflagellate bloom (Gyrodinium aureolum) on the mussel Mytilus edulis. J. Mar. Biol. Assoc. U. K. 59, 522–524. Wikfors, G.H., Smolowitz, R.M., 1993. Detrimental effects of a Prorocentrum minimum isolate upon hard clams and bay scallops in laboratory feeding studies. In: Smayda, T.J., Shimizu, Y. (Eds.), Toxic Phytoplankton Blooms in the Sea. Elsevier Science, New York, pp. 452–477. Yamatogi, T., Sakaguchi, M., Matsuda, M., Iwanaga, S., Iwataki, M., Matsuoka, K., 2005. Effect on bivalve molluscs of a harmful dinoflagellate Heterocapsa circularisquama isolated from Omura Bay Japan and its growth characteristics. Nippon Suisan Gakkaishi 71, 746–754 (in Japanese with English abstract). Yamasaki, Y., Yanan, Z., Go, J., Shikata, T., Matsuyama, Y., Nagai, K., Shimasaki, Y., Yamaguchi, K., Oshima, Y., Oda, T., Honjo, T., 2011. Cell contact-dependent lethal effects of the dinoflagellate Heterocapsa ciruclarisquama on phytoplankton–phytoplankton interactions. J. Sea Res. 65, 76–83.