Chemosphere 93 (2013) 931–938

Contents lists available at SciVerse ScienceDirect

Chemosphere

journal homepage: www.elsevier .com/locate /chemosphere

Screening for antioxidant and detoxification responsesin Perna canaliculus Gmelin exposed to an antifouling bioactiveintended for use in aquaculture

0045-6535/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.chemosphere.2013.05.058

Abbreviations: BB, biochemical biomarker; CAT, catalase; EtOH, ethanol; FSW,filtered seawater; GPOX, glutathione peroxidase; GSH, glutathione; GSH-t, totalglutathione; GSSG, glutathione disulphide; GST, glutathione S-transferase; IC99, 99%inhibition of metamorphosis; LHPO, lipid hydroperoxides; PB, physiologicalbiomarker; SOD, superoxide dismutase.⇑ Corresponding author at: Cawthron Institute, Private Bag 2, Nelson 7042, New

Zealand. Tel.: +64 3 548 2319; fax: +64 3 546 9464.E-mail address: Patrick.Cahill@cawthron.org.nz (P.L. Cahill).

Patrick Louis Cahill a,b,⇑, David Burritt c, Kevin Heasman a, Andrew Jeffs b, Jeanne Kuhajek a

a Cawthron Institute, Private Bag 2, Nelson 7042, New Zealandb Department of Marine Science, University of Auckland, P.O. Box 349, Warkworth, Northland 0941, New Zealandc Department of Botany, University of Otago, 464 Great King Street, Dunedin 9016, New Zealand

h i g h l i g h t s

� Used biochemical biomarkers to screen for effects of polygodial on Perna canaliculus.� Examined markers of oxidative stress and a detoxification pathway.� Exposure to the IC99 against fouling ascidians had no effect in P. canaliculus.� Antioxidant enzyme activity increased in P. canaliculus exposed to 10� the IC99.

a r t i c l e i n f o

Article history:Received 21 September 2012Received in revised form 13 May 2013Accepted 25 May 2013Available online 2 July 2013

Keywords:AntioxidantBiofoulingBiomarkerDetoxificationPerna canaliculusPolygodial

a b s t r a c t

Polygodial is a drimane sesquiterpene dialdehyde derived from certain terrestrial plant species thatpotently inhibits ascidian metamorphosis, and thus has potential for controlling fouling ascidians inbivalve aquaculture. The current study examined the effects of polygodial on a range of biochemical bio-markers of oxidative stress and detoxification effort in the gills of adult Perna canaliculus Gmelin. Despitehigh statistical power and the success of positive controls, the antioxidant enzymes glutathione reductase(GR), glutathione peroxidase (GPOX), catalase (CAT), and superoxide dismutase (SOD); thiol status, asmeasured by total glutathione (GSH-t), glutathione disulphide (GSSG), and GSH-t/GSSG ratio; end prod-ucts of oxidative damage, lipid hydroperoxides (LHPO) and protein carbonyls; and detoxification path-ways, represented by GSH-t and glutathione S-transferase (GST), were unaffected in the gills of adultP. canaliculus exposed to polygodial at 0.1 or 1 � the 99% effective dose in fouling ascidians (IC99). Simi-larly, GR levels, thiol status, and detoxification activities were unaffected in mussels exposed to polygo-dial at 10 � the IC99, although GPOX, CAT, and SOD activities increased. However, the increases weresmall relative to positive controls, no corresponding oxidative damage was detected, and this concentra-tion greatly exceeds effective doses required to inhibit fouling ascidians in aquaculture. These findingscompliment a previous study that established the insensitivity to polygodial of P. canaliculus growth, con-dition, and mitochondrial functioning, providing additional support for the suitability of polygodial foruse as an antifouling agent in bivalve aquaculture.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction fouling remedy for the costly problem of fouling ascidians in bi-

Polygodial is a drimane sesquiterpene dialdehyde produced byseveral terrestrial plant species that has potential as a specific anti-

valve aquaculture. Polygodial potently inhibits metamorphosis inascidian larvae (Cahill et al., 2012), and initial screens of organ-ism-level PB (growth, survival, and mitochondrial functioning) de-tected no negative effects in cultured bivalves exposed topolygodial at the 99% effective dose against ascidian metamorpho-sis (i.e., IC99; Cahill et al., 2013). However, while PB respond to abroad range of xenobiotic stressors (Depledge et al., 1995; Lamand Gray, 2003) and are relevant to whole-organism consequences(De Coen et al., 2000), they can have relatively limited sensitivityfor detecting low-level negative effects within a practicable time-frame (Van der Oost et al., 2003; Venturino et al., 2003).

932 P.L. Cahill et al. / Chemosphere 93 (2013) 931–938

Conversely, BB – biochemical components, processes, and func-tions – are inherently sensitive proxies that often react to stressorslong before organism-level impacts become apparent (Walker,1995; Madden and Gallagher, 1999; Van der Oost et al., 2003).Although the relevance of BB to whole-organism consequencesmust be interpreted carefully (Forbes et al., 2006), they can re-spond to a broad range of environmental stressors and are rou-tinely utilized for ecotoxicological studies on marine bivalves(e.g., Regoli and Principato, 1995; Walker, 1995; De Lafontaineet al., 2000). Thus, BB provide an ideal secondary screen to compli-ment the previously established lack of detectable effects on PBand further evaluate possible negative effects of polygodial onadult bivalves. While the exact mechanism by which polygodialinhibits ascidian metamorphosis is unknown, a range of biochem-ical effects have been reported in other species, including non-ionicsurfactant properties (Kubo et al., 2001); ATP synthase antagonism(Lunde and Kubo, 2000); tachykinin NK2 receptor agonism (ElSayah et al., 1998); and interaction with the opioid system (Men-des et al., 2000), nitric oxide, endogenous prostaglandins, sulfhy-dryl compounds, and vanilloid receptors (Matsuda et al., 2002).In light of the plethora of conceivable biochemical modes of actionfor polygodial in bivalves, the current study calls for widelyresponsive and generally applicable BB that reflect general envi-ronmental stress.

Markers of oxidative stress are among the most widely respon-sive and generally applicable class of BB (Almeida et al., 2007) be-cause of the close relationship between general environmentalstress and the generation rate of cellular reactive oxygen and nitro-gen species (i.e., oxidative stress) in an organism (Storey, 1996).The three main classes of BB of oxidative stress are antioxidant en-zymes, thiol status, and end products of oxidative damage. Antiox-idant enzymes of interest in mussels include GR, GPOX, CAT, andSOD (e.g., Solé and Albaigés, 1995; Lionetto et al., 2003; Bocchettiand Regoli, 2006; Box et al., 2007; Vlahogianni et al., 2007). Theseenzymes are integral components of the cellular antioxidant de-fense system: GR catalyzes the conversion of GSSG to GSH (Man-nervik, 1987); GPOX, the reduction of LHPO and hydrogenperoxide (Battin and Brumaghim, 2009); CAT, the decompositionof hydrogen peroxide (Chelikani et al., 2004); and SOD, the dismu-tation of superoxide (McCord and Fridovich, 1988). Thiol statusalso plays a central role in antioxidant defense (Finkel and Hol-brook, 2000; Dafre et al., 2004). Integral measures of thiol statusthat have been identified as useful BB of oxidative stress in bivalvesare GSH-t, GSSG, and the GSH-t/GSSG ratio (e.g., Regoli and Princ-ipato, 1995; Cheung et al., 2001; Dafre et al., 2004; Franco et al.,2006). Likewise, the end products of oxidative damage LHPO andprotein carbonyls reflect the ultimate consequences of oxidativestress for cells (Doyotte et al., 1997; Almeida et al., 2007; Verlecaret al., 2007).

Comparable to oxidative stress, detoxification effort is a usefulindicator of environmental stress that provides direct evidence ofexposure to toxins (Lagadic et al., 1994). In addition to its antioxi-dant roles, GSH is a central component of cellular detoxification,owing to its ability to conjugate some toxins (Viarengo and Nott,1993). Consequently, levels of GSH-t, and the associated conjuga-tion enzyme GST (Douglas, 1987), are effective BB of detoxificationeffort in bivalves (e.g., Hai et al., 1997; Kaaya et al., 1999; Akchaet al., 2000; Manduzio et al., 2004).

The aim of this study was to quantify the effects of polygodialon the BB outlined above in the gills of an economically importantaquaculture species in New Zealand, the green-lipped mussel (Per-na canaliculus Gmelin). This mussel is the main target aquaculturespecies for the application of polygodial and the optimal cultureparameters for this species are well established (Vakily, 1989). Fur-thermore, gills are the ideal target tissue for biomonitoring andtoxicological studies in mussels because they are the first barrier

to potential contaminants, owing to their role as the filter feedingapparatus and respiratory organ (Bolognesi et al., 2004).

Due to the previously established lack of detectable effects onthe physiological health of bivalves (Cahill et al., 2013), it washypothesized that polygodial would not affect antioxidant en-zymes, thiol status, end products of oxidative damage, or detoxifi-cation pathways in adult P. canaliculus. The results from this studyprovide an important next step towards developing polygodial asan antifouling agent for use in bivalve aquaculture.

2. Methods

2.1. Experimental organisms

Adult P. canaliculus (24.2 ± 8.3 mm shell height) were obtainedfrom a commercial mussel farm in Pelorus Sound, Marlborough,New Zealand and held in a 5000-L recirculating seawater systemaccording to Cahill et al. (2013).

2.2. Experimental design

Randomly selected mussels were housed in pairs in high den-sity polyethylene containers (130 � 130 � 150 mm), preparedaccording to Cahill et al. (2013) and each containing 2.0 L of 0.4-lm FSW (salinity: 34 ± 1 ppt). The culture containers were heldat 18 ± 1 �C and mussels were fed daily with 50 mL of a 1:1 mixof Isochrysis galbana Parke (8–9 � 106 cells mL�1) to Pavlova lutheriGreen (10–12 � 106 cells mL�1).

Treatments consisted of polygodial (ENZO Lifescience, Farming-dale, NY; yellow solid comprising 97% polygodial; extracted fromPolygonum hydropiper L.) dosed at 0.1, 1, or 10� the IC99 in ascidianlarvae, corresponding to 0.3 (Pglow), 3.0 (Pgmid), and 30.0 ng mL�1

(Pghigh), respectively (Cahill et al., 2012). Stock solutions of polyg-odial were prepared in EtOH and dosed into the culture vesselsat a final concentration of 0.1 lL EtOH mL�1 of FSW. Positive con-trols utilized zinc dosed at and above concentrations previouslyshown to negatively affect antioxidant status in a closely relatedmussel species, Perna perna L. (Franco et al., 2006). Accordingly,zinc was dosed at 0.7 (Znlow), 7.0 (Znmid), and 70.0 lg mL�1 (Znhigh).Negative controls consisted of 0.1 lL EtOH mL�1 of FSW and FSWonly. Ten replicates (n = 10) were performed simultaneously forall treatments and controls. The experiment was run for 14 d fol-lowing the first introduction of the chemical treatments. Chemicaltreatments in all containers were refreshed daily with FSWchanges.

2.3. Gill tissue preparation

At 7 and 14 d after the chemical treatments began, one musselwas removed from each culture vessel and immediately dissectedon ice. Gills were excised, divided into 100–150 mg subsamples,and placed in Nunc™ CryoTube vials (Thermo Fisher Scientific,Waltham, MA), which were flushed with nitrogen gas and thensealed. Subsamples were frozen in liquid nitrogen, and stored at�80 �C until further analysis.

All extraction steps were performed at 4 �C. Total protein wasextracted for determination of enzyme activities and protein car-bonyls according to Franco et al. (2006) with minor modifications.Briefly, gill tissue subsamples were homogenized in 900 lL ofHEPES buffer using a bead beater (3 � 2.4 mm BioSpec™ zirconiabeads; BioSpec Products, Bartlesvile, OK; 3 � 15 s cycles). Thehomogenate was centrifuged at 20,000 g for 30 min and the super-natant was stored in 150-lL aliquots at – 80 �C. For cellular thiolanalysis, the same procedure as outlined for total protein extrac-tion was used, except 5% salicylic acid was substituted for HEPES

P.L. Cahill et al. / Chemosphere 93 (2013) 931–938 933

buffer. Lipid extracts were obtained from gill tissue subsamplesaccording to Lister et al. (2010b).

2.4. Biochemical biomarkers

Enzyme activities were determined from total protein extractsand standardized against protein content, which was determinedas per Fryer et al. (1986) with minor modifications (Lister et al.,2010a). Glutathione reductase activities were evaluated accordingto Cribb et al. (1989) with minor modifications, and GPOX activitywas determined according to Ursini et al. (1985), as modified byContreras et al. (2005). Catalase was assayed using a chemilumi-nescence method (Maral et al., 1977), as modified by Janssenset al. (2000). Superoxide dismutase was analyzed using the micro-plate assay described by Banowetz et al. (2004) with minor modi-fications (Lister et al., 2010b). Thiol status (GSH-t, GSSG, and GSH-t/GSSG ratio) was determined from the cellular thiol extracts usinga microtitre plate-based enzymatic recycling method (Rahmanet al., 2006), and standardized against wet tissue weight. Lipidhydroperoxide levels in the lipid extracts were quantified usingferric thiocyanate (Mihaljevic et al., 1996) and standardizedagainst wet tissue weight, whereas protein carbonyl levels weredetermined from total protein extracts via reaction with 2,4-dini-trophenylhydrazine (Reznick and Packer, 1994) and standardizedagainst protein content. Glutathion S-transferase was assayedusing the photometric 1-chloro-2,4-dinitrobenzene method of Ha-big et al. (1974) with minor modifications. All assays were per-formed using a PerkinElmer (Wallac) 1420 multilabel counter(Perkin Elmer, San Jose, CA) fitted with a temperature-control celland an auto-dispenser. Data were acquired and processed usingthe WorkOut 2.0 software package (Perkin Elmer, San Jose, CA).

2.5. Statistical analyses

For the ten biomarkers examined, differences among treat-ments were analyzed using one-way ANOVA with Tukey’s HSDpost hoc test. Prior to one-way ANOVA analysis, Levene’s test forhomogeneity of variance was applied and heterogeneous data setswere LOG10 transformed. Where this transformation was unsuc-cessful in equalizing variances, the non-parametric Kruskal–Wallisrank sum test with post hoc pairwise comparisons using Wilcoxonrank sum test was applied (Holm p-adjustment). If applicable (one-way ANOVA test only) statistical power was determined using apost hoc power test for one-way ANOVA.

ANOVA analyses were performed using R version 2.13.0 (R: ALanguage and Environment for Statistical Computing. R Foundationfor Statistical Computing, Vienna, AT. Available from: http://R-pro-ject.org/). Power analyses were performed using G*Power 3.1.3(G*Power 3. Heinrich Heine Universitat Dusseldorf, Kiel, DK. Avail-able from: http://www.psycho.uniduesseldorf.de/abteilungen/aap/gpower3/). The level of significance was set at a = 0.05 and stan-dard error is reported in all cases.

3. Results

3.1. Antioxidant enzymes

The experiment had high statistical power to detect differencesamong treatments for the four antioxidant enzymes (k = 16, n = 10;GR: f = 17.8, P > 0.9; GPOX: f = 17.5, P > 0.9; CAT: f = 11.4, P > 0.9;SOD: f = 18.0, P > 0.9). Enzyme activities did not differ betweenthe FSW and EtOH controls, or between the two sampling timesfor all treatments and controls (Fig. 1).

Although the assay successfully detected decreased GR activi-ties in the 7.0 and 70.0 lg zinc mL�1 positive controls, the three

polygodial treatments had no effect on GR activities relative tothe corresponding negative control (Fig. 1A, Table 1). Glutathioneperoxidase, CAT, and SOD activities increased in the 0.7 and7.0 lg zinc mL�1 positive controls, yet decreased to near undetect-able levels in the 70.0 lg zinc mL�1 positive control (Fig. 1B–D, Ta-ble 1). While the 0.3 and 3.0 ng polygodial mL�1 treatments had noeffect on GPOX, CAT, or SOD activities, exposure to polygodial at10� the IC99 in ascidians increased the activities of these enzymesby 21 ± 1%, 19 ± 1%, and 32 ± 1%, respectively (Fig. 1B–D, Table 1).

3.2. Thiol status

Thiol status (GSH-t, GSSG, and GSH-t/GSSG ratio) did not differbetween the two negative controls, or between any of the threepolygodial treatments and the negative controls (Fig. 2, Table 1).This is in spite of the high statistical power for GSSG (k = 16,n = 10, f = 1.0, P > 0.9) and GSH-t/GSSG ratio (k = 16, n = 10,f = 1.3, P > 0.9), and the significant effects detected for GSH-t,GSSG, and GSH-t/GSSG ratio in the zinc positive controls (Fig. 2,Table 1).

3.3. End products of oxidative damage and markers of detoxificationeffort

Protein carbonyls (k = 16, n = 10, f = 1.7, P > 0.9) and GST (k = 16,n = 10, f = 2.7, P > 0.9) had high statistical power to detect differ-ences among treatments. In contrast, statistical power could notbe calculated for LHPO because the data were heterogeneous,necessitating the use of the Kruskal–Wallis test. No differenceswere apparent in LHPO, protein carbonyls, or GST between thetwo negative controls, or the two sampling times, regardless oftreatment (Fig. 3).

Zinc exposure had a positive dose dependent effect on LHPOand protein carbonyl levels, yet polygodial had no detectable ef-fects on these end products of oxidative damage (Fig. 3A–B, Ta-ble 1). Similarly, exposure to polygodial at 0.3, 3.0, or30.0 ng mL�1 had no effect on GST, despite GST activities increas-ing in mussels exposed to 0.7 or 7.0 lg zinc mL�1, and decreasingin mussels exposed to 70.0 lg zinc mL�1 (Fig. 3C, Table 1).

4. Discussion

Despite the known sensitivity of BB (Walker, 1995; Van derOost et al., 2003), the high statistical power of the experiment,and the observed sensitivity for detecting oxidative stress anddetoxification effort in the positive controls, no effects were de-tected on the antioxidant enzymes, thiol status, end products ofoxidative damage, or detoxification pathways in the gills of adultP. canaliculus exposed to polygodial at, or below, the known IC99

in ascidian larvae. These results are consistent with previous re-search that established a lack of effects of polygodial, dosed atthe IC99 in ascidians, on the growth, condition, or mitochondrialfunctioning of adult P. canaliculus (Cahill et al., 2013).

Conversely, exposure to polygodial at 10� the IC99 in ascidiansincreased GPOX, CAT, and SOD activities. While these effects mayreflect ecologically relevant consequences of polygodial exposurefor P. canaliculus, the increases were small relative to those seenin the 0.7 and 7.0 lg zinc mL�1 positive controls, and comparedto most previous reports for chemically induced effects on GPOX,CAT, and SOD activities in mussels from the genus Perna (e.g., Che-ung et al., 2004; Franco et al., 2006; Verlecar et al., 2007). Further-more, the GR (e.g., Cheung et al., 2004; Franco et al., 2006), GPOX(Cheung et al., 2002; Verlecar et al., 2008), CAT (e.g., Lau andWong, 2003; Franco et al., 2006), and SOD (e.g., Lau and Wong,2003; Verlecar et al., 2007) activities in the polygodial treatments

Fig. 1. (A) Glutathione reductase (GR), (B) glutathione peroxidase (GPOX), (C) catalase (CAT), and (D) super oxide dismutase (SOD) activities in Perna canaliculus Gmelin gilltissue exposed to polygodial at 0.3 (Pglow), 3.0 (Pgmid), or 30.0 ng mL�1 (Pghigh); zinc positive controls at 0.7 (Znlow), 7.0 (Znmid), or 70.0 lg mL�1 (Znhigh); or negative controls(FSW: filtered seawater, ETOH: ethanol) for 7 and 14 d. Bars are means of replicate determinations (n = 10) ± standard error; ⁄p < 0.05 or ⁄⁄p < 0.0001 relative to correspondingnegative controls.

934 P.L. Cahill et al. / Chemosphere 93 (2013) 931–938

and negative controls are comparable to values previously re-ported for the gill tissue of unexposed mussels from the genusPerna. Thus, while an oxidative challenge was experienced by P.canaliculus exposed to polygodial at 30.0 ng mL�1, this challengeappears to have been small. Regardless, concentrations of polygo-dial used in aquaculture should not exceed the IC99 in ascidians,precluding possible oxidative stress in the target bivalve aquacul-ture species.

Unlike the antioxidant enzymes, thiol status, as measured byGSH-t, GSSG, and GSH-t/GSSG ratio, was unaffected by exposureto the three polygodial concentrations. This is despite the high sta-tistical power of the assays and the strong positive effects detectedfor GSH-t in the 0.7 and 7.0 lg zinc mL�1 positive controls, andGSSG and GSH-t/GSSG ratio in the 7.0 and 70.0 lg zinc mL�1 posi-tive controls. Like the antioxidant enzymes, the values observed forGSH-t in the negative controls and polygodial treatments corre-spond to previous reports for unexposed mussels from the genusPerna (e.g., Almeida et al., 2004; Dafre et al., 2004). Conversely,GSSG levels (54.0 ± 4.5–162.4 ± 14.2 nmol g wet tissue�1) werehigh relative to a previous study where GSSG ranged from9.5 ± 2.5 to 28.0 ± 5.0 nmol g wet tissue�1 in P. perna (Dafre et al.,2004). This resulted in lower GSH-t/GSSG ratios relative to the pre-vious report, and may reflect differences in holding conditions, in-ter-species variation in thiol status, or the use of gill tissue versusdigestive gland tissue between the two studies, respectively. Nev-ertheless, cellular thiol status plays a central role in antioxidant de-fense (Dafre et al., 2004), allowing the insensitivity of GSH-t, GSSG,or GSH-t/GSSG ratio to polygodial to support a corresponding lackof oxidative stress in P. canaliculus.

Akin to thiol status, levels of the end products of oxidative dam-age LHPO and protein carbonyls were unaffected by exposure topolygodial. Both LHPO and protein carbonyl levels in the negativecontrols and polygodial treatments corresponded to previous re-ports for unexposed mussels from the genus Perna (e.g., Almeidaet al., 2005; Verlecar et al., 2008; Jena et al., 2010). Statisticalpower to detect differences among treatments was high and thezinc positive controls had a strong dose dependent effect. Further-more, LHPO was more sensitive to zinc exposure in the currentstudy compared to a previous report for P. perna (Franco et al.,2006), highlighting the capacity of the assay to detect oxidativestress in P. canaliculus. The lack of effects of polygodial on LHPOand protein carbonyls contradicts the impacts of 30.0 ng polygo-dial mL�1on GPOX, CAT, and SOD activities. This finding further im-plies that the oxidative challenge experienced by mussels exposedto polygodial at 30.0 ng mL�1 was minor and, as no measurableoxidative damage resulted, this challenge was likely to have beeneffectively counteracted by cellular antioxidant defenses.

Similarly, detoxification effort, as measured by GSH-t and GST,was unaffected by exposure to polygodial despite these BB increas-ing in the zinc positive controls and the high statistical power ofthe experiment. As with GSH-t, GST activities in the negative con-trols and polygodial treatments were comparable to previous re-ports for gill tissue of untreated mussels from the genus Perna(e.g., Alves et al., 2002; Lau and Wong, 2003; Cheung et al.,2004). The strong effects observed in the zinc positive controls con-trast with a previous study that demonstrated the insensitivity ofGST to zinc exposure in P. perna (Franco et al., 2006). While thisdisparity may reflect differences in exposure time (7 and 14 d vs.

Table 1Results from one-way ANOVA analyses (within subjects df: 9, between subjects df: 7) with Tukey’s HSD post hoc test or Kruskal–Wallis rank sum test with Wilcoxon post hoc testcomparing biomarkers of oxidative stress (GR: glutathione reductase, GPOX: glutathione peroxidase, CAT: catalase, SOD: superoxide dismutase, GSH-t: total glutathione, GSSG:glutathione disulphide, LHPO: lipid hydroperoxides, Prot Carb: protein carbonyls) and detoxification effort (GSH-t, GST: glutathione S-transferase) in Perna canaliculus Gmelin gilltissue exposed to polygodial at 0.3 (Pglow), 3.0 (Pgmid), or 30.0 ng mL�1 (Pghigh) or zinc positive controls at 0.7 (Znlow), 7.0 (Znmid), or 70.0 lg mL�1 (Znhigh) for 7 and 14 d, relativeto unexposed controls. Significant values shown.

One-way ANOVA [F(p)] Tukey’s HSD (p)

Pglow7 Pglow14 Pgmid7 Pgmid14 Pghigh7 Pghigh14 Znlow7 Znlow14 Znmid7 Znmid14 Znhigh7 Znhigh14

GR 3305.2(<0.001) – – – – – – – – <0.0001 <0.0001 <0.0001 <0.0001GPOX 2989.3(<0.0001) – – – – 0.001 0.0009 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001CAT 1320.0(<0.0001) – – – – 0.01 0.004 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001SOD 3222.1(<0.0001) – – – – <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001GSH-t 80.0(<0.0001)a – – – – – – – 0.008b 0.008b – – –GSSG 6.6(<0.0001) – – – – – – – – <0.0001 – – 0.0003GSH-t/GSSG 4.7(<0.0001) – – – – – – – – 0.0005 – – <0.0001LHPO 121.1(<0.0001)a – – – – – – 0.004b 0.02b 0.02b 0.001b 0.02b 0.02b

Prot Carb 80.3(<0.0001) – – – – – – 0.002 0.0006 <0.0001 <0.0001 <0.0001 <0.0001GST 73.0(<0.0001) – – – – – – <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

a Kruskal–Wallis test[H(p)].b Wilcoxon rank-sum test (p).

Fig. 2. (A) Total glutathione (GSH-t), (B) glutathione disulphide (GSSG), and (C) GSH-t/GSSG ratio in Perna canaliculus Gmelin gill tissue exposed to polygodial at 0.3 (Pglow),3.0 (Pgmid), or 30.0 ng mL�1 (Pghigh); zinc positive controls at 0.7 (Znlow), 7.0 (Znmid), or 70.0 lg mL�1 (Znhigh); or negative controls (FSW: filtered seawater, ETOH: ethanol) for7 and 14 d. Bars are means of replicate determinations (n = 10) ± standard error; ⁄p < 0.05 or ⁄⁄p < 0.0001 relative to corresponding negative controls.

P.L. Cahill et al. / Chemosphere 93 (2013) 931–938 935

2 d), culture conditions, or differences in oxidative defenses or zinctoxicity between the two species, it highlights the inherent sensi-tivity of the GST assay used here for detecting detoxification effortin P. canaliculus. Because the GSH-t/GST detoxification pathway ispart of the repertoire of adaptive response mechanisms to

chemical stress (Viarengo and Nott, 1993; Sharma and Davis,1994; Norppa, 2003) and its expression responds to a broad spec-trum of chemical stressors (Gadagbui and James, 2000), the lack ofeffects provides additional support for the hypothesized non-toxicnature of polygodial in P. canaliculus.

Fig. 3. (A) Lipid hydroperoxides (LHPO), (B) protein carbonyls (Prot Carbonyls), and (C) glutathione S-transferase (GST) activity in Perna canaliculus Gmelin gill tissue exposedto polygodial at 0.3 (Pglow), 3.0 (Pgmid), or 30.0 ng mL�1 (Pghigh); zinc positive controls at 0.7 (Znlow), 7.0 (Znmid), or 70.0 lg mL�1 (Znhigh); or negative controls (FSW: filteredseawater, ETOH: ethanol) for 7 and 14 d. Bars are means of replicate determinations (n = 10) ± standard error; ⁄p < 0.05 or ⁄⁄p < 0.0001 relative to corresponding negativecontrols.

936 P.L. Cahill et al. / Chemosphere 93 (2013) 931–938

Although not directly relating to the effects of polygodial on P.canaliculus, several discrepancies were apparent for the 70.0 lgzinc mL�1 positive control. First, GR, GPOX, CAT, SOD, and GSTactivities decreased to near undetectable levels. Although thelethal concentration of zinc in P. canaliculus is unknown, 70.0 lgzinc mL�1 exceeds the 50% lethal maxima for Perna viridis L.(8.8 lg mL�1; Chan, 1988) and P. perna (7.0 lg mL�1; Francoet al., 2006), suggesting that the lethal threshold for zinc in P. can-aliculus was exceeded. Accordingly, the cellular antioxidant de-fenses may have been overwhelmed, potentially accounting forthe very low enzyme activities. Several previous reports have de-tected similar decreases in antioxidant enzyme activity in organ-isms that have been exposed to a toxin at concentrationsapproaching their lethal maxima (e.g., David et al., 2008; Kavithaand Rao, 2008). This unimodal or hormetic response could havewider implications for the understanding of the antioxidant de-fences of bivalves and ecotoxicological testing, warranting furtherexamination to fully elucidate the effects of zinc on P. canaliculus.Second, LHPO and protein carbonyl levels in the 70.0 lg zincmL�1 positive control exceeded all previous reports for musselsfrom the genus Perna. While the unusually high values may beattributable to the estimates for these end products of oxidativedamage having been extrapolated from outside the standard curveused (in the case of the 70.0 lg zinc mL�1 positive control only),this finding further implies that the toxic threshold for zinc in P.canaliculus had been met. These inconsistencies, concerning thehighest concentration of zinc only, do not affect the overall

conclusions drawn regarding the effects of polygodial on bivalvesbecause the current study is concerned with detecting effects,rather than their magnitude or direction.

5. Conclusions

Despite the high statistical power to detect effects and successof the positive controls, polygodial dosed at or below the IC99 inascidians had no detectable effects on antioxidant enzymes, thiolstatus, end-products of oxidative damage, or detoxification path-ways in the gills of adult P. canaliculus. While exposure to polygo-dial at 10� the IC99 in ascidians increased antioxidant enzymesexpression, the magnitude of the impacts were small and therewas a lack of corresponding oxidative damage. Furthermore, 10�the IC99 greatly exceeds the effective dose of polygodial requiredto control fouling ascidians in aquaculture. These findings compli-ment two previous studies reporting the high potency of polygo-dial against ascidian metamorphosis (Cahill et al., 2012) and thelack of measurable effects on the physiological health of bivalves(Cahill et al., 2013), but the suite of biomarkers tested to date isfar from exhaustive. Accordingly, future research aiming to fullyevaluate the suitability of polygodial for use as an antifouling agentin aquaculture should combine developmental, intergenerational,genetic, neurological, and reproductive screening in P. canaliculuswith robust ecotoxicological evaluations of possible environmentaleffects of polygodial.

P.L. Cahill et al. / Chemosphere 93 (2013) 931–938 937

Acknowledgement

This study was supported by the New Zealand Ministry for Busi-ness, Innovation and Employment (CAWX0802).

References

Akcha, F., Izuel, C., Venier, P., Budzinski, H., Burgeot, T., Narbonne, J.F., 2000.Enzymatic biomarker measurement and study of DNA adduct formation inbenzo[a]pyrene-contaminated mussels, Mytilus galloprovincialis. Aquat. Toxicol.49, 269–287.

Almeida, E.A., Miyamoto, S., Bainy, A.C.D., Medeiros, M.H.G., Mascio, P.D., 2004.Protective effect of phospholipid hydroperoxide glutathione peroxidase(PHGPx) against lipid peroxidation in mussels, Perna perna, exposed todifferent metals. Mar. Pollut. Bull. 49, 386–392.

Almeida, E.A., Bainy, A.C.D., Dafre, A.L., Gomes, O.F., Medeiros, M.H.G., Mascio, P.D.,2005. Oxidative stress in digestive gland and gill of the brown mussel (Pernaperna) exposed to air and re-submersed. J. Exp. Mar. Biol. Ecol. 318, 21–30.

Almeida, E.A., Bainy, A.C.D., Melo Loureiro, A.P., Martinez, G.R., Miyamoto, S., Onuki,J., Barbosa, L.F., Garcia, C.C.M., Prado, F.M., Ronsein, G.E., Sigolo, C.A., Brochini,C.B., Martins, A.M.G., Medeiros, M.H.G., Mascio, P.D., 2007. Oxidative damage inPerna perna and other bivalves as indicators of environmental stress in theBrazilian marine environment: antioxidants, lipid peroxidation and DNAdamage. Comp. Biochem. Physiol. 146, 588–600.

Alves, S.R.C., Severino, P.C., Ibbottson, D.P., Silva, A.Z., Lopes, F.R.A.S., Saenz, L.A.,Bainy, A.C.D., 2002. Effects of furadan in the brown mussel Perna perna and inthe mangrove oyster Crassostrea rhizophorae. Mar. Environ. Res. 54, 241–245.

Banowetz, G.M., Dierksen, K.P., Azevedo, M.D., Stout, R., 2004. Microplatequantification of plant leaf superoxide dismutases. Anal. Biochem. 332, 314–320.

Battin, E.E., Brumaghim, J.L., 2009. Antioxidant activity of sulphur and selenium: areview of reactive oxygen species scavenging, glutathione peroxidase, andmetal binding antioxidant mechanisms. Cell Biochem. Biophys. 55, 1–23.

Bocchetti, R., Regoli, F., 2006. Seasonal variability of oxidative biomarkers,lysosomal parameters, metallothioneins and peroxisomal enzymes in theMediterranean mussel Mytilus galloprovincialis from Adriatic Sea.Chemosphere 65, 913–921.

Bolognesi, C., Frenzilli, G., Lasagna, C., Perrone, E., Roggieri, P., 2004. Genotoxicitybiomarkers in Mytilus galloprovincialis: wild versus caged mussels. Mutat. Res.Fund. Mol. M. 552, 153–162.

Box, A., Sureda, A., Galgani, F., Pons, A., Deudero, S., 2007. Assessment ofenvironmental pollution at Balearic Islands applying oxidative stressbiomarkers in the mussel Mytilus galloprovincialis. Comp. Biochem. Phys. C146, 531–539.

Cahill, P., Heasman, K., Jeffs, A., Kuhajek, J., Mountfort, D., 2012. Preventing ascidianfouling in aquaculture: screening selected allelochemicals for anti-metamorphic properties in ascidian larvae. Biofouling 28, 39–49.

Cahill, P., Heasman, K., Hickey, A., Mountfort, D., Jeffs, A., Kuhajek, J., 2013. Screeningfor negative effects of candidate antifoulant compounds on a target aquaculturespecies. Biofouling 29, 29–37.

Chan, H.M., 1988. Accumulation and tolerance to cadmium, copper, lead and zinc bythe green mussel Perna viridis. Mar. Ecol. Prog. Ser. 48, 295–303.

Chelikani, P., Fita, I., Loewen, P.C., 2004. Diversity of structures and propertiesamong catalases. Cell. Mol. Life Sci. 61, 192–208.

Cheung, C.C.C., Zheng, G.J., Li, A.M.Y., Richardson, B.J., Lam, P.K.S., 2001.Relationships between tissue concentration of polycyclic aromatichydrocarbons and antioxidative responses of marine mussels, Perna viridis.Aquat. Toxicol. 52, 189–203.

Cheung, C.C.C., Zheng, G.J., Lam, P.K.S., Richardson, B.J., 2002. Relationships betweentissue concentrations of chlorinated hydrocarbons (polychlorinated biphenylsand chlorinated pesticides) and antioxidative responses of marine mussels,Perna viridis. Mar. Pollut. Bull. 45, 181–191.

Cheung, C.C.C., Sui, W.H.L., Richardson, B.J., De Luca-Abbott, S.B., Lam, P.K.S., 2004.Antioxidant responses to benzo[a]pyrene and aroclor 1254 exposure in thegreen-lipped mussel, Perna viridis. Environ. Pollut. 128, 393–403.

Contreras, L., Moenne, A., Correa, J.A., 2005. Antioxidant responses in Scytosiphonlomentaria (Phaeophyceae) inhabiting copper-enriched coastal environments. J.Phycol. 41, 1184–1195.

Cribb, A.E., Leeder, J.S., Spielberg, S.P., 1989. Use of a microplate reader in an assay ofglutathione-reductase using 5,5’-dithiobis(2-nitrobenzoic acid). Anal. Biochem.183, 195–196.

Dafre, A.L., Medeiros, I.D., Müller, I.C., Ventura, E.C., Bainy, A.C.D., 2004. Antioxidantenzymes and thiol disulphide status in the digestive gland of the brown musselPerna perna exposed to lead and paraquat. Chem. Biol. Interact. 149, 97–105.

David, M., Munaswamy, V., Halappa, R., Marigoudar, S.R., 2008. Impact of sodiumcyanide on catalase activity in the freshwater exotic carp, Cyprinus carpio(Linnaeus). Pest. Biochem. Physiol. 92, 15–18.

De Coen, W.M., Janssen, C.R., Giesy, J.P., 2000. Biomarker applications inecotoxicology: bridging the gap between toxicology and ecology. In:Persoone, G., Janssen, C., De Coen, W. (Eds.), New Microbiotests for RoutineToxicity Screening and Biomonitoring. Kluwer Academic/Plenum Publishers,New York, pp. 13–26.

De Lafontaine, Y., Gagné, F., Blaise, C., Costan, G., Gagnon, P., Chan, H.M., 2000.Biomarkers in zebra mussels (Dreissena polymorpha) for the assessment and

monitoring of water quality of the St Lawrence River (Canada). Aquat. Toxicol.50, 51–71.

Depledge, M.H., Aagaard, A., Gyorkos, P., 1995. Assessment of trace metal toxicityusing molecular, physiological and behavioural biomarkers. Mar. Pollut. Bull.31, 19–27.

Douglas, K.T., 1987. Mechanism of action of glutathione-dependent enzymes. Adv.Enzymol. Relat. Areas. Mol. Biol. 59, 103–167.

Doyotte, A., Cossu, C., Jacquin, C.M., Babut, M., Vasseur, P., 1997. Antioxidantenzymes, glutathione and lipid peroxidation as relevant biomarkers ofexperimental or field exposure in the gills and the digestive gland of thefreshwater bivalve Unio tumidus. Aqua. Toxicol. 39, 93–110.

El Sayah, M., Cechinel Filho, V., Yunes, R.A., Pinheiro, T.R., Calixto, J.B., 1998. Actionof polygodial, a sesquiterpene isolated from Drymis winteri, in the guinea-pigileum and trachea in vitro. Eur. J. Pharmacol. 344, 215–221.

Finkel, T., Holbrook, N.J., 2000. Oxidants, oxidative stress and the biology of ageing.Nature 408, 239–247.

Forbes, V.E., Palmqvist, A., Bach, L., 2006. The use and misuse of biomarkers inecotoxicology. Environ. Toxicol. Chem. 25, 272–280.

Franco, J.L., Trivella, D.B.B., Trevisan, R., Dinslaken, D.F., Marques, M.R.F., Bainy,A.C.D., Dafre, A.L., 2006. Antioxidant status and stress proteins in the gills of thebrown mussel Perna perna exposed to zinc. Chem. Biol. Interact. 160, 232–240.

Fryer, H.J.L., Davis, G.E., Manthorpe, M., Varon, S., 1986. Lowry protein assay usingan automatic microtiter plate spectrophotometer. Anal. Biochem. 153, 262–266.

Gadagbui, B.K.M., James, M.O., 2000. Activities of affinity-isolated glutathione S-transferase (GST) from channel catfish whole intestine. Aquat. Toxicol. 49, 27–37.

Habig, W.H., Pabst, M.J., Jakoby, W.B., 1974. Glutathione S-transferases – firstenzymatic step in mercapturic acid formation. J. Biol. Chem. 249, 7130–7139.

Hai, D.Q., Varga, S.I., Matkovics, B., 1997. Organophosphate effects on antioxidantsystem of carp (Cyprinus carpio) and catfish (Ictalurus nebulosus). Comp.Biochem. Physiol. C 117, 83–88.

Janssens, B.J., Childress, J.J., Baguet, F., Rees, J.F., 2000. Reduced enzymaticantioxidative defense in deep-sea fish. J. Exp. Biol. 203, 3717–3725.

Jena, K.B., Jagtap, T.G., Verlecar, X.N., 2010. Antioxidative potential of Perna viridisand its protective role against ROS induced lipidperoxidation and proteincarbonyl. Curr. Trends Biotechnol. Pharm. 4, 862–870.

Kaaya, A., Najimi, S., Ribera, D., Narbonne, J.F., Moukrim, A., 1999. Characterizationof glutathione S-transferase (GST) activities in Perna perna and Mytilusgalloprovincialis used as a biomarker of pollution in the Agadir Bay (SouthMorocco). Bull. Environ. Contam. Toxicol. 62, 623–629.

Kavitha, P., Rao, J.V., 2008. Toxic effects of chlorpyrifos on antioxidant enzymes andtarget enzyme acetylcholinesterase interaction in mosquito fish, Gambusiaaffinis. Environ. Toxicol. Pharm. 26, 192–198.

Kubo, I., Fujita, K.I., Lee, S.H., 2001. Antifungal mechanism of polygodial. J. Agric.Food Chem. 49, 1607–1611.

Lagadic, L., Caquet, T., Ramade, F., 1994. The role of biomarkers in environmentalassessment (5). Invertebrate populations and communities. Ecotoxicology 3,193–208.

Lam, P.K.S., Gray, J.S., 2003. The use of biomarkers in environmental monitoringprograms. Mar. Pollut. Bull. 46, 182–186.

Lau, P.S., Wong, H.L., 2003. Effect of size, tissue parts and location on six biochemicalmarkers in the green-lipped mussel, Perna viridis. Mar. Pollut. Bull. 46, 1563–1572.

Lionetto, M.G., Caricato, R., Giordano, M.E., Pascariello, M.F., Marinosci, L., Schettino,T., 2003. Integrated use of biomarkers (acetylcholinesterase and antioxidantenzyme activities) in Mytilus galloprovincialis and Mullus barbatus in an Italiancoastal marine area. Mar. Pollut. Bull. 46, 324–330.

Lister, K.N., Lamare, M.D., Burritt, D.J., 2010a. Oxidative damage in response tonatural levels of UV-B radiation in larvae of the tropical sea urchin Tripneustesgratilla. Photochem. Photobiol. 86, 1091–1098.

Lister, K.N., Lamare, M.D., Burritt, D.J., 2010b. Sea ice protects the embryos of theAntarctic sea urchin Sterechinus neumayeri from oxidative damage due tonaturally enhanced levels of UV-B radiation. J. Exp. Biol. 213, 1967–1975.

Lunde, C.S., Kubo, I., 2000. Effect of polygodial on the mitochondrial ATPase ofSaccharomyces cerevisiae. Antimicrob. Agents Ch. 44, 1943–1953.

Madden, M.C., Gallagher, J.E., 1999. Biomarkers of exposure. In: Holgate, S.T., Samet,J.M., Koren, H.S., Maynard, R.L. (Eds.), Air Pollution and Health. Academic Press,London, pp. 417–431.

Manduzio, H., Monsinjon, T., Galap, C., Leboulenger, F., Rocher, B., 2004. Seasonalvariations in antioxidant defenses in blue mussel Mytilus edulis collected from apolluted river: major contributions of an inducible isoform of Cu/Zn-superoxidedismutase and of glutathione S-transferase. Aquat. Toxicol. 70, 83–93.

Mannervik, B., 1987. The enzymes of glutathione metabolism: an overview.Biochem. Soc. Trans. 15, 717–723.

Maral, J., Puget, K., Michelson, A.M., 1977. Comparative-study of superoxidedismutase, catalase and glutathione peroxidase levels in erythrocytes ofdifferent animals. Biochem. Biophys. Res. Commun. 77, 1525–1535.

Matsuda, H., Pongpiriyadacha, Y., Morikawa, T., Kashima, Y., Nakano, K., Yoshikawa,M., 2002. Protective effects of polygodial and related compounds on ethanol-induced gastric mucosal lesions in rats: structural requirements and mode ofaction. Bioorg. Med. Chem. Lett. 12, 477–482.

McCord, J.M., Fridovich, I., 1988. Superoxide dismutase: the first twenty years(1968–1988). Free Radic. Biol. Med. 5, 342–363.

Mendes, G.L., Santos, A.R., Malheiros, A., Cechinel Filho, V., Yunes, R.A., Calixto, J.B.,2000. Assessment of mechanisms involved in antinociception caused bysesquiterpene polygodial. J. Pharm. Exp. Ther. 292, 164–172.

938 P.L. Cahill et al. / Chemosphere 93 (2013) 931–938

Mihaljevic, B., Katusin-Razem, B., Razem, D., 1996. The re-evaluation of the ferricthiocyanate assay for lipid hydroperoxides with special considerations of themechanistic aspects of the response. Free Radic. Biol. Med. 21, 53–63.

Norppa, H., 2003. Genetic susceptibility, biomarker responses and cancer. MutantRes. 544, 339–348.

Rahman, I., Kode, A., Biswas, S.K., 2006. Assay for quantitative determination ofglutathione and glutathione disulfide levels using enzymatic recycling method.Nat. Protoc. 1, 3159–3165.

Regoli, F., Principato, G., 1995. Glutathione, glutathione-dependent and anti-oxidantenzymes in mussels, Mytilus galloprovincialis, exposed to metals under field andlaboratory conditions: implications for the use of biochemical biomarkers.Aqua. Toxicol. 31, 143–164.

Reznick, A.Z., Packer, L., 1994. Oxidative damage to proteins – spectrophotometricmethod for carbonyl assay. In: Ableson, J., Simon, M. (Eds.), Oxygen Radicals inBiological Systems Part C, Methods in Enzymology, vol. 233. Elsevier,Amsterdam, pp. 357–363.

Sharma, Y.K., Davis, K.R., 1994. Ozone-induced expression of stress related genes inArabidopsis thaliana. Plant Physiol. 105, 1089–1096.

Solé, M., Albaigés, C.P.J., 1995. The use of biomarkers for assessing the effects oforganic pollution in mussels. Sci. Total Environ. 159, 147–153.

Storey, K.B., 1996. Oxidative stress: animal adaptations in nature. Braz. J. Med. Biol.Res. 29, 1715–1733.

Ursini, F., Maiorino, M., Gregolin, C., 1985. The selenoenzyme phospholipidhydroperoxide glutathione-peroxidase. Biochim. Biophys. Acta 839, 62–70.

Vakily, J.M., 1989. The biology and culture of mussels of the genus Perna.International Centre for Living Aquatic Resources Management, Eschbom.

Van der Oost, R., Beyer, J., Vermeulen, N.P.E., 2003. Fish bioaccumulation andbiomarkers in environmental risk assessment: a review. Environ. Toxicol.Pharm. 13, 57–149.

Venturino, A., Rosenbaum, E., De Castro, A.C., Anguiano, O.L., Guana, L., Schroeder,T.F., D’Angelo, A.M.P., 2003. Biomarkers of effects in toads and frogs. Biomarkers8, 167–187.

Verlecar, X.N., Jena, K.B., Chainy, G.B.N., 2007. Biochemical markers of oxidativestress in Perna viridis exposed to mercury and temperature. Chem. Biol. Interact.167, 219–226.

Verlecar, X.N., Jena, K.B., Chainy, G.B.N., 2008. Modulation of antioxidant defenses indigestive gland of Perna viridis, on mercury exposure. Chemosphere 71, 1977–1985.

Viarengo, A., Nott, J.A., 1993. Mechanisms of heavy-metal cation homeostasis inmarine-invertebrates. Comp. Biochem. Physiol. C 104, 355–372.

Vlahogianni, T., Dassenakis, M., Scoullos, M.J., Valavanidis, A., 2007. Integrated useof biomarkers (superoxide dismutase, catalase and lipid peroxidation) inmussels Mytilus galloprovincialis for assessing heavy metals’ pollution incoastal areas from the Saronikos Gulf of Greece. Mar. Pollut. Bull. 54, 1361–1371.

Walker, C.H., 1995. Biochemical biomarkers in ecotoxicology – some recentdevelopments. Sci. Total Environ. 171, 189–195.