Physiological impacts of acute Cu exposure on deep-sea vent mussel Bathymodiolus azoricus under a deep-sea mining activity scenario

Accepted Manuscript Title: mPhysiological impacts of acute Cu exposure on deep-sea vent mussel Bathymodiolus azoricus under a deep-sea mining activity scenario Authors: Inˆes Martins, Joana Goulart, Eva Martins, Rosa Morales-Rom´an, Sergio Mar´ın, Virginie Riou, Ana Colac¸o, Raul Bettencourt PII: DOI: Reference:

S0166-445X(17)30282-5 https://doi.org/10.1016/j.aquatox.2017.10.004 AQTOX 4766

To appear in:

Aquatic Toxicology

Received date: Revised date: Accepted date:

3-5-2017 6-9-2017 9-10-2017

Please cite this article as: Martins, Inˆes, Goulart, Joana, Martins, Eva, Morales-Rom´an, Rosa, Mar´ın, Sergio, Riou, Virginie, Colac¸o, Ana, Bettencourt, Raul, mPhysiological impacts of acute Cu exposure on deep-sea vent mussel Bathymodiolus azoricus under a deep-sea mining activity scenario.Aquatic Toxicology https://doi.org/10.1016/j.aquatox.2017.10.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

mPhysiological impacts of acute Cu exposure on deep-sea vent mussel Bathymodiolus azoricus under a deep-sea mining activity scenario

Inês Martinsa,b,*, Joana Goularta,b, Eva Martinsa,b, Rosa Morales-Románb, Sergio Marínb, Virginie Rioub, Ana Colaçoa,b,c, Raul Bettencourta,b,c a

MARE – Marine and Environmental Sciences Centre, 9901-862 Horta, Portugal

b

IMAR- Department of Oceanography and Fisheries, University of Azores, 9901-862 Horta, Portugal

c

OKEANOS- Research Unit- Faculty of Science and Technology, University of the Azores, 9901-862 Horta,

Portugal

*corresponding author: +351292200400; e-mail: [email protected]

Highlights: . B. azoricus was exposed to high Cu concentrations under pressurized conditions .High Cu concentrations cause serious changes in cellular defense mechanisms of B. azoricus . Enzymatic and transcriptional response to high Cu concentration exposure is tissuespecific . Exposure to high Cu concentrations causes a decrease in enzyme activity and transcriptional suppression in mussel gills

Abstract Over the past years, several studies have been dedicated to understanding the physiological ability of the vent mussel Bathymodiolus azoricus to overcome the high metal concentrations present in their surrounding hydrothermal environment. Potential deep-sea mining activities at Azores Triple junction hydrothermal vent deposits would inevitably lead to the emergence of new fluid sources close to mussel beds, with consequent emission of high metal concentrations and potential resolubilization of Cu from minerals formed during the active phase of the vent field. Copper is an essential metal playing a key role in the activation of metalloenzymes and metalloproteins responsible for important cellular 1

metabolic processes and tissue homeostasis. However, excessive intracellular amounts of reactive Cu ions may cause irreversible damages triggering possible cell apoptosis. In the present study, B. azoricus was exposed to increasing concentrations of Cu for 96 hours in conditions of temperature and hydrostatic pressure similar to those experienced at the Lucky Strike hydrothermal vent field. Specimens were kept in 1L flasks, exposed to four Cu concentrations: 0 µg/L (control), 300, 800 and 1600 µg/L and pressurized to 1750 bar. We addressed the question of how increased Cu concentration would affect the function of antioxidant defense proteins and expression of antioxidant and immune-related genes in B. azoricus. Both antioxidant enzymatic activities and gene expression were examined in gills, mantle and digestive gland tissues of exposed vent mussels. Our study reveals that stressful short-term Cu exposure has a strong effect on molecular metabolism of the hydrothermal vent mussel, especially in gill tissue. Initially, both the stress caused by unpressurization or by Cu exposure was associated with high antioxidant enzyme activities and tissue-specific transcriptional up-regulation. However, mussels exposed to increased Cu concentrations showed both antioxidant and immune-related gene suppression. Under a mining activity scenario, the release of an excess of dissolved Cu to the vent environment may cause serious changes in cellular defense mechanisms of B. azoricus. This outcome, while adding to our knowledge of Cu toxicity, highlights the potentially deleterious impacts of mining activities on the physiology of deep-sea organisms. Keywords: Bathymodiolus azoricus; copper; deep-sea mining; molecular biomarkers; physiological stress Introduction

Deep-sea mining activities at hydrothermal vent deposits are imminent (Nautilus in Solwara, www.nautilusminerals.com) and have been the focus of concern of the scientific community, since the scale and nature of the impacts on deep-sea ecosystems are still unknown (MIDAS project, www.eu-midas.net). It seems more than likely that mineral exploration will promote the emergence of new fluid sources and the remobilization of metals complexes trapped in the sediments, which, when in contact with oxygenated water, may be desorbed and become readily bioavailable (Roberts, 2012). This increase in metal bioavailability in marine ecosystems has serious implications for organism physiology and cellular integrity (Borgmann, 2000; Cajaraville et al., 2000). The mussel B. azoricus is the dominant species in the Mid-Atlantic Ridge hydrothermal vents and its ability to regulate high metal concentrations in the tissues (Company et al., 2010; Martins et al., 2009; 2

Martins et al., 2011b) makes this species a suitable bio-indicator of metal exposure and metabolic response to metals in the hydrothermal ecosystems (Cosson et al., 2008; Martins et al., 2016). Trace metals like Cu, are found in high concentrations in hydrothermal vent ecosystems, with concentrations ranging from 0.02 to 5.15 µM in fluid-seawater mixing zone (Sarradin et al., 2009). Copper acts as a micronutrient in seawater and plays essential roles in cellular function and metabolism (Turski and Thiele, 2009). However, excessive amounts of metal ions stimulate reactive oxygen species (ROS) production via the Fenton reaction, in which hydrogen peroxide interacts with iron to generate hydroxyl radicals (Gaetke and Chow, 2003; Leonard et al., 2004). Increased levels of ROS within the cell promote lipid peroxidation, DNA damage, altered calcium homeostasis, and cell death (Pulido and Parrish, 2003). To overcome cellular damage, organisms activate an antioxidant defense system, composed of antioxidant enzymes such as superoxide dismutase (SOD; EC 1.15.1.1), catalase (CAT; EC 1.11.1.6) and glutathione peroxidase (GPx; EC 1.11.1.9), which ensure optimal protection against oxidative stress (Lushchak, 2011). These and other biomarkers were already monitored in the deep-sea vent mussel B. azoricus to evaluate the effects of metal exposure on lipid peroxidation and metallothionein and iron storage protein concentrations (Company et al., 2007; Martins et al., 2008; Martins et al., 2016). Further studies with B. azoricus reported the transcriptional modulation of antioxidant and immune gene families upon metal exposure (Bougerol et al., 2015) and other environmental stressors (Barros et al., 2015; Boutet et al., 2011). However, the effect of acute Cu exposure under in situ pressurized conditions was never studied. The hydrothermal vent mussel B. azoricus lives harmoniously with two endosymbiotic chemosynthetic gram-positive, sulfide-oxidizing and methanotrophic bacteria, harbored in the gills (Duperron et al., 2006). The mytilid host supplies the carbon sources needed by both thiotrophic and methanotrophic symbionts, as well as essential biosynthetic tricarboxylic acid (TCA) cycle enzymes lacking in the thiotrophic symbionts. In return, the symbionts compensate the host’s putative deficiency in amino acids and cofactors (Ponnudurai et al., 2017). The gills, mantles and digestive gland of B. azoricus are metabolically very active tissues, implicated in both endosymbiont and metal metabolisms (Martins et al., 2011a; Riou et al., 2008). They are, therefore, expected to respond markedly to metal exposure under 3

experimental conditions. In the present study, we measured antioxidant enzyme activities in these tissues after exposure to increased Cu concentrations, under pressurized conditions, chosen for being representative of new sources of Cu under a scenario of mining in the vicinity of Lucky Strike vent field. We hypothesize that acute exposure to high Cu concentrations will inflict a short-term physiological stress in the vent mussel. We also examined the transcriptional modulation of antioxidant and immune genes in response to Cu stress. We chose twelve genes known for being involved in different physiological functions (detoxification, signaling, transcription and recognition) (Bettencourt et al., 2010; Bougerol et al., 2015) in order to have a better view of the regulation that occurs at a cellular level under acute exposure to high Cu concentration. Our main objective was to investigate B. azoricus physiological mechanisms of cellular protection against oxidative stress induced by short term exposure to high§ Cu concentrations and to discuss the potential consequences of mining activities in hydrothermal vent areas.

2. Materials and methods

2.1. Sample collection and experimental design Samples were collected during the MoMARSAT cruise in July 2014 (with the R/V “Pourquoi Pas?”), at 1700 m depth on the Lucky Strike hydrothermal field (37°17’N; 32°16’W). Lucky Strike is an offshore hydrothermal complex located at the Mid-Atlantic Ridge, inside the Portuguese EEZ. Mussels were collected using the ROV “Victor 6000’’ arm grab and brought to the surface using a closed non-pressurized insulated box. After 1.5 h surfacing time, mussels were directly transferred to fresh cooled seawater (7-8°C) and held in plastic cool boxes (with air-oxygen supply) during the 48 hour transit to our landbased refrigerated laboratory “LabHorta” (Colaco et al., 2011). Immediately after arriving to the laboratory, the mussels were cleaned of visible adhering material and separated in 5 groups of 8 mussels (57 ± 4 mm). One group was dissected immediately after arriving to “LabHorta”, named hereafter by T0, and each of the other four groups were kept in 1 L polypropylene flasks filled with 5–6 °C seawater and exposed to different Cu concentrations: 0 µg/L (control), 300, 800 and 1600 µg/L following pressurization to 175 4

bar for 96 hours using the hyperbaric chamber IPOCAMP (Shillito et al., 2014). The selected Cu concentrations are representative of dissolved Cu concentrations measured, (1) near the Lucky Strike B. azoricus mussel beds (~300 µg/L); (2) in the reactive mixing zone (~800 µg/L) (Sarradin et al., 2009) and at end-member fluids (~1600 µg/L) (Charlou et al., 2000). The seawater for the experiment was supplied from a reservoir containing sandfiltered oceanic water from an unpolluted bay (Cruz et al., 2010) in Horta, Azores (38°58′N; 28°78′W). This seawater was treated by UV-light and filtered through an external power canister filter (Eheim 600) before use. Every 24 h the pressurized conditions were interrupted for measurements of air saturation and temperature (55.6± 9.1 % and 5.1±0,5 oC), using an OxyGuard® Handy Polaris 2, in each treatment and for seawater renewal while assuring that Cu concentrations were not altered before pressure reestablishment. The gill, mantle and digestive gland tissues were dissected from each sampled individual and preserved at -80 °C for later analysis.

2.2. Preparation of tissue extracts for antioxidant enzyme activity and lipid peroxidation measurements Frozen gill and mantle samples were homogenized at a 1:3 w/v and digestive gland at a 1:2 w/v ratio in ice-cold 10 mM phosphate-buffered saline solution (PBS pH 7,4) containing 5 mM EDTA and 1% (v/v) ProteaseArrestTM (G-Biosciences®) cocktail using an Ultra Turrax (Ystral®, D79282) slowly increasing the rotational velocity from 8000 to 20,000 rpm during the ~2 min extraction time. The homogenate was centrifuged at 16000 g for 30 min at 4 ºC and enzyme activities were measured in the supernatant fraction. All enzyme assays were tested with commercial enzymes obtained from Sigma® and each sample was run in triplicate (technical replicates).

2.2.1. Determination of antioxidant enzyme activities

Protein quantification Soluble proteins were quantified according to Bradford method (Bradford, 1976), adapted from Bio-Rad Bradford microassay set up in a 96-well microplate. Absorbance was read at

5

595 nm in a microplate reader (Thermo Scientific™). A calibration curve was created using bovine serum albumin (BSA; Bio-Rad) standards.

Glutathione Peroxidase (GPx) The total GPx activity was measured based on Paglia and Valentine (1967) UV method. GPx catalyzes the oxidation of Glutathione (GSH) by Cumene Hydroperoxide. In the presence of Glutathione Reductase (GR) and NADPH the oxidized Glutathione (GSSG) is immediately converted to the reduced form with a concomitant oxidation of NADPH to NADP+. NADPH oxidation is accompanied by a decrease in absorbance measured at 340 nm (Ransel RS 505 kit, Randox). Under conditions in which the GPx activity is rate limiting, the rate of decrease in the A340 is directly proportional to the GPx activity in the sample. The total GPx activity in all the tissues is expressed as Units (nmol min−1 mg−1) per milligram of total protein in the sample wet weight.

Superoxide dismutase (SOD) The total SOD activity was determined spectrophotometrically by an indirect method (Therond et al., 1996) based on competition of SOD with 2-(4-iodophenyl)-3-(4nitrophenol)-5- phenyltetrazolium chloride (I.N.T) for dismutation of superoxide anion (O2).

In this method, xanthine and xanthine oxidase were used to generate O-2 radicals which

react with I.N.T quantitatively to form a red formazan dye. Absorbance was measured at 505 nm and 25 ºC, 30s after the addition of xanthine oxidase as start reagent across a 180 s incubation period (Ransod SD 125 kit, Randox). One unit of SOD is defined as the amount of enzyme that inhibits the rate of formazan dye formation by 50%. Then, a SOD standard curve is used to correlate percent inhibition of samples with SOD activity. The percent inhibition of standards and samples was calculated using the following equation: 100[[ΔA505/min]/ [ΔAblank/min]] *100. The total SOD activity in all the tissues is expressed as Units per milligram of total protein (U mg−1) in the sample wet weight.

Catalase (CAT) CAT activity was measured spectrophotometrically, according to Beers and Sizer (1952), by measuring the rate of H2O2 disappearance at 240 nm (extinction coefficient, ε = 0,04 6

mM−1 cm−1) and 25 °C during a 180 s incubation period. In this assay, total reaction volume of 2.7 ml was obtained with 50 mM potassium phosphate buffer (pH 7.0), 13.5 mM H2O2 as a substrate and initiated by the addition of the sample into quartz cuvettes with a path length of 10 mm. Catalase from bovine liver (Sigma®) was used as a positive control (1524 U ml-1) for validation of the assay. Catalase activity was calculated using the following equation: [ΔA240/min/0.04]* [total volume/sample volume]. CAT enzymatic activity in all the tissues is expressed as Units per milligram of total protein (nmol min−1 mg−1) in the sample wet weight.

2.2.2. Lipid peroxidation (LPO) Lipid peroxidation was determined by the quantification of a specific end-product of the oxidative degradation process of lipids, malondialdehyde (MDA) (Kalghatgi et al., 2013). Concentrations of MDA were analyzed using a colorimetric reaction which uses 1-methyl2-phenylindole (MPI) as chromogen (Randox Ltd.). Condensation of one molecule of MDA with 2 molecules of MPI under acidic conditions results in the formation of a chromophore with an absorbance maximum at 586 nm. Concentrations of MDA in each tissue were calculated using a standard curve prepared with freshly prepared solutions of malondialdehyde bis [dimethyl acetal] (ACROS Organics™) and values were expressed as nmol mg−1 of total protein in the sample wet weight.

2.2.3 Total RNA extraction and quantification of gene expression Total RNA was extracted from frozen gill, mantle and digestive gland tissues with TriReagent® (Ambion®) and further purified with GeneJet™ RNA Kit (Fermentas), following the manufacturer's specifications, and re-suspended in nuclease-free DEPCtreated water. Total RNA quality and concentrations were assessed by the A260/280 and A260/230 spectrophotometric ratios using the NanoVue spectrophotometer (GE, Healthcare Life Sciences). Single-strand cDNA was synthesized from total RNA with Thermo Scientific Maxima First Strand cDNA Synthesis Kit for RT-qPCR, according to the manufacturer's instructions, using 2 µg of total RNA per sample. The cDNA concentration was measured using the NanoVue spectrophotometer as above. Gene expression analyses for gill, mantle and digestive gland tissues in every copper treatment 7

were carried out by means of quantitative PCR (qPCR) following the MIQE guidelines (Bustin et al., 2009), using a mixture of two samples for gills and mantles corresponding to three biological replicates (n=3 with pool of 2 samples in each replicate) per each tissue per copper treatment. Digestive glands were analyzed individually with a total of three samples (n=3) per copper treatment. Gene sequences were based on query analyses using the DeepSeaVent, B. azoricus transcriptome database (http://transcriptomics.biocant.pt/deepSeaVent) (Bettencourt et al., 2010) and selected to target both antioxidant an immune metabolic pathways. Gene name, primers sequences and metabolic pathways are provided in Table 1. Real-time PCR was carried out in the CFX96TM Real-Time (Bio-Rad) using the Power SYBR®green PCR Master Mix (Applied Biosystems). The reaction was performed in triplicate in a total volume of 10 µl containing 5 µl SYBR Power SYBR®green PCR Master Mix, 1 µl cDNA, 1 µl (10 µM) of each of the primers and 3 µl PCR-grade water. The PCR program used was 95 ºC for 10 min, followed by 40 cycles at 95 ºC for 15 s, 60 ºC for 1 min. Negative control without cDNA was included in each assay. Melting curve analysis of amplification products was performed at the end of each PCR to confirm that only one product was amplified and detected. The 18S and 28S ribosomal genes (sequences are available in Table 2) were examined for the potential to serve as qPCR references in our experimental study. Both genes were validated as reference genes in previous gene expression studies with B. azoricus mussel (Barros et al., 2015; Bettencourt et al., 2014; Martins et al., 2015). The 28S ribosomal gene had the highest PCR efficiency (101% Table 2) and therefore was chosen as the reference gene for our qPCR experiment. Data analyses were based on the ΔΔCt method with normalization of the raw data to the reference gene expression values (Livak and Schmittgen, 2001; Pfaffl, 2004). PCR efficiency (E%) and correlation coefficient (R2) were determined based on the slopes of the standard curves generated using serial 5-fold dilutions (assayed in duplicate) of sample cDNA (transcribed from 50 ng of total RNA). The efficiency (E) was calculated as follows: E = [10(-1/slope) – 1] *100 (Kubista et al., 2006). PCR amplification efficiencies are present in Table 2. Three technical replicates were obtained from qPCR experiments. Fold change units were calculated by dividing the normalized expression in different copper treatments by the

8

normalized expression values from time zero (T0). Data were expressed as normalized gene expression (mean ± SD).

2.3. Statistical analysis All values are reported as mean ± standard deviation (SD). Enzymes activities, lipid peroxidation and gene expression data did not meet the normality and homoscedasticity assumptions, even after transformation. Therefore, to test the influence of high copper concentrations on the antioxidant and immune defense mechanisms of the vent mussel, a PERMANOVA multivariate analysis of variance test (unrestricted permutation of raw data) was applied (Anderson et al., 2008). This statistical analysis is a powerful non-parametric approach that uses a permutational technique to enable significance tests for small sample sizes to be conducted (Walters and Coen, 2006) and was used to test the null hypotheses: (1) high copper concentrations had no effect on the antioxidant enzyme activity of B. azoricus; (2) high copper concentrations had no effect on transcriptional modulation of antioxidant and immune genes of B. azoricus. The analyses were conducted using the software PRIMER 6 & PERMANOVA using a resemblance matrix based on Euclidean distance (Anderson et al., 2008) and treatments as fixed effects. The PERMANOVA was run using 9999 permutations to produce p values using the Monte Carlo (MC) method. When the main test produced a significant result (p < 0.05), a pairwise test was conducted to identify the individual differences between treatments.

3. Results During the exposure period, a small mortality was recorded (2 mussels in non-contaminated group and 300 µg L-1 and 1 mussel in 800 and 1600 µg L-1) while all other specimens presented a healthy appearance during the dissection process.

3.1. Enzymatic activity of GPx, SOD, CAT and LPO levels Figures 1, 2 and 3 show the results of GPx, SOD and CAT mean specific activities (Units per mg of protein) and LPO levels (MDA nmol per mg of protein) found in gills, mantles

9

and digestive glands, respectively, of B. azoricus mussels sampled before exposure (T0) and after 96h Cu exposure (0-1600 µg/L) under pressurized conditions. Gills The antioxidant enzymatic activities of GPx and SOD showed no significant difference between treatments (PERMANOVA, p>0.05) in gill tissues. CAT activity decreases after exposure to the highest Cu concentration, 1600 µg/L, showing a significant decrease of activity when compared with CAT activity in gills from T0 mussels (PERMANOVA, pairwise test p<0.05) (Fig.1 A). An increasing (although not statistically significant (PERMANOVA, p>0.05)) trend of lipid peroxidation (LPO) was found in mussel gills after exposure to Cu (Fig.1 B). Mantle The antioxidant enzymatic activities of SOD and CAT showed no significant difference between treatments (PERMANOVA, p>0.05) in mantle tissues. GPx activity decreased significantly in mussels exposed to 800 and 1600 µg/L concentrations of Cu, (PERMANOVA, pairwise test p<0.01) (Fig.2 A). There is a trend (although not statistically significant (PERMANOVA, p>0.05)) for LPO increase in mantles from exposed mussels (Fig.2 B).

Digestive gland The antioxidant enzymatic activities of GPx and SOD showed no significant difference between treatments (PERMANOVA, p>0.05) in digestive glands. The digestive glands from the non-contaminated treatment showed the lowest CAT activity (PERMANOVA, pairwise test p<0.01) (Fig.3 A). There is a trend for LPO increase in digestive glands from exposed mussels with a following decrease in mussels exposed to the highest, 1600 µg/L (PERMANOVA, p>0.05), Cu concentrations (Fig.3 B).

An overall analysis of the antioxidant enzymatic specific activities and LPO levels between tissues shows that gills have the highest activities of GPx (0.09 U/mg) and SOD (3.7 U/mg) 10

activities and the lowest LPO levels (0.11 nmol MDA/mg) in T0 and Cu exposure treatments for 96h. The digestive glands presented the highest CAT activity (15.4 U/mg) and the lowest GPx activity (0.03 U/mg). The mantles showed the highest levels of LPO (0.52 nmol MDA/mg), similar to digestive glands (0.48 nmol MDA/mg), and the lowest CAT activity (4.3 U/mg). 3.2. Antioxidant and immune related gene expression Figures 4, 5 and 6 show the results of normalized expression of 6 antioxidant (CAT; SOD; Ferritin; GPx; HSP70; MT) and 6 immune (ILR2; PGRP; RTK; TLR2; LITAF; NF-kB) related genes, found in gills, mantles and digestive glands, respectively, of B. azoricus mussels after 96h Cu exposure (0-1600 µg/L) under pressurized conditions. Standardization of expression ratios took into consideration a fold change unit equal to 1 in T0 samples. Fold changes above 1 were considered as up-regulation of genes (higher mRNA abundance than in T0) and those below 1 were considered as down-regulation of genes (lower mRNA abundance than in T0).

Gills After pressurization, there is an evident up-regulated expression of antioxidant and immune genes in gills from mussels from the non-contaminated treatment (Fig. 4). Nevertheless, the transcriptional levels of antioxidant and immune genes were down-regulated by Cu exposure as mussels were exposed to increased levels of Cu. Furthermore, 1600µg/L of Cu exposure inhibited the transcriptional expression of both groups of genes (Fig. 4). Among Cu treatments, mussels in the non-contaminated treatment show the highest expression of genes encoding Ferritin, GPx, HSP70 and MT (Fig. 4A) and the 6 immune genes (PERMANOVA, pairwise test p<0.05) (Fig. 4B). The mussels exposed to the highest Cu concentrations, 1600 µg/L, presented a significant expression decrease of both antioxidant and immune-related genes (PERMANOVA, pairwise test p<0.01) (Fig.4). Mantle No significant regulation of antioxidant (Fig. 5A) and immune (Fig. 5B) gene expressions was observed in mantle tissue for non-contaminated and Cu exposed mussels after 11

pressurization (PERMANOVA, p>0.05). Also, no significant variation of antioxidant and immune gene expressions was observed in mantle tissues among Cu treatments (PERMANOVA, p>0.05) (Fig. 5).

Digestive gland No significant regulation of antioxidant (Fig. 6A) and immune (Fig. 6B) gene expressions was found in digestive glands for non-contaminated and Cu exposed mussels after pressurization (PERMANOVA, p>0.05). Also, among Cu treatments, no significant variation of antioxidant and immune gene expressions was observed for this tissue (PERMANOVA, p>0.05) (Fig. 5). An overall analysis of the antioxidant and immune gene expressions shows that gills are more responsive to Cu exposure for 96h than mantles and digestive glands. Moreover, the transcriptional levels of the analyzed genes in gill tissues are most noticeably influenced by Cu concentration. On the other hand, mantles and digestive glands show minor variations of gene expression across Cu treatments.

4. Discussion While mining companies might avoid active vents (due to hazardous conditions), direct impacts on peripheral biological communities and indirect impacts on vent communities themselves are anticipated (Boschen et al., 2013). Studying the ecological risks of the deepsea mining activities is a difficult task because there are several important variables to consider in an environmental impact study, such as the mining technology used, the degrees of sediment suspension, and the toxicant properties. Nevertheless, the possibility of new vent sources arising near hydrothermal vent fields is real and the consequent impacts on vent organisms need to be evaluated. The hydrothermal vent biota at the Mid-Atlantic Ridge, including the mixotrophic vent mussel B. azoricus, does not live close to the hot fluid exits but rather at the interface where the venting water meets the ambient seawater (Cuvelier et al., 2009). This distribution is explained by the fact that chemosynthetic vent organisms require both oxidized seawater and reduced chemicals (H2S and CH4) that are 12

their metabolic source of energy (Fiala-Médioni et al., 2002). Oxygen and H2S only coexist at low temperatures around the vent because of the high redox capacity of the hydrothermal reduced chemicals (Sarradin et al., 1999). The survival of aerobic animals above these temperatures is due to the small-scale turbulence that delivers parcels of cooler oxygenated water to the animals (Johnson et al., 1988; Tunnicliffe, 1991). Consequent mining activities possibly causing the outflow of hot, acidic fluid rich in sulfide and metals will create a more toxic and anoxic environment to hydrothermal vent biological communities (Halfar and Fujita, 2007). Impact of mining activity is now the subject of growing concerns among deep sea biologists alarmed by the possible harmful impacts that such human activities may bring about on susceptible biological vent communities. Most importantly, research studies are just beginning to address such environmental impacts. In this study we simulated the exposure of B. azoricus to acutely elevated Cu concentrations under temperature and hydrostatic pressure conditions close to those experienced by the vent mussel at Lucky Strike hydrothermal vent field (Sarradin et al., 1999) in order to investigate the molecular response of B. azoricus to a dissolved Cu excess in a short-term exposure. Our results show that short-term Cu exposure induces an antioxidant and immune response in B. azoricus tissues. Such response seems to be tissue-specific, but out of the studied tissues gills are the most responsive one to Cu exposure under high hydrostatic pressure conditions.

Gills In B. azoricus gills enzymatic activity and lipid peroxidation are high in individuals from T0, which represents their physiological state after 48h acclimation at atmospheric pressure. Moreover, CAT and GPx seem to be the primary antioxidant defense agents in gills against atmospheric pressure acclimatization. This may indicate that H2O2 accounts for a large proportion of reactive oxygen species (Gutierrez et al., 2003; Pallavi Sharma et al., 2012) generated in mussels gills under unpressurized conditions. Therefore, the retrieval of animals from and subsequent decrease of hydrostatic pressure for B. azoricus from Lucky Strike high depth vent field (1700 m depth) may possibly represent an enormous physiological stress, since Lucky Strike mussels are unable to survive more than a few days at atmospheric pressure (I. Martins, personal communication). In fact after pressurization, the enzymatic activity and LPO decreases in mussel gills. However, Cu exposure triggered 13

the activation of antioxidant defense system and increase of lipid peroxidation. In B. azoricus mussels, as in other marine organisms, Cu is considered to be a strong inducer of intracellular oxidative stress (due to its high redox activity) causing the stimulation of antioxidant enzyme activity as cellular defense mechanism (Company et al., 2008; Kim et al., 2014), which bring evidence supporting our observed result. Among the different analyzed tissues, gills presented the highest amounts of antioxidant activity and the lowest amounts of LPO reflecting an efficient short-term defense response against stressful conditions caused by unpressurization and Cu contamination. The transcript expression observed in gills shows a significant up-regulation of both antioxidant and immune-related genes after pressurization in noncontaminated conditions, followed by a general decrease of expression with increased Cu concentrations. In fact, high Cu concentrations of 1600 µg/L seem to suppress the expression of both antioxidant and immune genes. This outcome, only observed in gills, can be related with a dual impact of Cu exposure and symbiont depletion resulting in a worsened condition of gill tissues. Previous studies have been suggesting a metabolic interdependence between the two partners (Bettencourt et al., 2010; Detree et al., 2016; Sayavedra et al., 2015). Although B. azoricus is able to acquire amino acids by direct uptake from seawater of from breakdown of organic matter ingested through filter feeding at the 840 m deep Menez Gwen vent site (Riou et al., 2010), specimens living at deeper vent fields as the Lucky Strike vent site, experience a lower particle flux and rely more on symbionts to meet their carbon needs (Boutet et al., 2011). Moreover, Sayavedra et al. (2015) suggest that endosymbiotic bacteria of Bathymodiolus have an important role in host protection against pathogens and may be requisitioned to counteract harmful interactions. A strong transcriptional up-regulation of both antioxidant and immune-related genes in gill tissues under non-contaminated conditions is noteworthy and might indicate that under pressurized conditions B. azoricus recovers its ability to trigger basal defense mechanisms to overcome symbiont loss. In this study histological analyses were not performed to confirm the presence of endosymbionts in mussel gills before and after the exposure experiment. Nevertheless, it is known from previous studies that a gradual loss of symbionts in B. azoricus gills occurs in the absence of reduced compounds (H2S, CH4, H2) (Detree et al., 2016; Kádár et al., 2005).

14

Among the most up-regulated antioxidant-related genes, ferritin presented a predominant level of expression. Bacteria are known to have a key role in cellular iron cycle (Verkhovtseva et al., 2001), benefiting the host in regulating Fe ions that in excess may disturb iron homeostasis causing iron toxicity and tissue abnormalities (Gonzalez et al., 2010). Other studies also suggest that ferritin is implicated in bivalve’s immune defense against bacterial pathogens (Zhou et al., 2014) by sequestering free Fe from the cellular environment into the ferritin core to suppress the pathogen growth, since Fe acquisition is necessary for pathogenicity and bacterial growth (Bullen et al., 2005). Therefore, B. azoricus may activate the transcription of the gene encoding ferritin to overcome the potential cellular stress driven by symbiont loss. After exposure to Cu, the levels of expression for both antioxidant and immune-related genes decrease significantly to the point that they are almost completely suppressed by toxic Cu concentrations (1600 µg/L). This might indicate that the transcription of genes is inhibited by high concentration of Cu in damaged gills. These results are in accordance with those found by Bougerol et al. (2015) who found that exposure to high metal concentrations strongly induce a decrease of gene expression regulation in B. azoricus gills.

Mantles In mantle tissues the expression of antioxidant enzymes activity is lower than in gills, especially for CAT activity. Interestingly, other metal exposure studies with B. azoricus demonstrated the reduced expression of GPx, SOD and CAT activity in mantle tissues in comparison with gills (Company et al., 2008; Company et al., 2010). This may suggest that either other ROS rather than H2O2 are present in mantle cells or other antioxidant enzymes may yet intervene in this tissue. Amongst the analyzed antioxidant enzymes only GPx activity showed differences between the experimental treatments. A decrease of GPx activity was observed in mantles from mussels exposed to the highest Cu concentrations. Under such toxic conditions, the observed decrease can reflect GPx degradation caused by the excessive amounts of reactive oxygen species. Oxidative degradation of proteins under conditions of oxidative stress was already described in previous studies (Dubinina et al., 2002; Soldatov et al., 2007). This fact leads us to assume that the complex action of ROS, triggered by the excessive Cu concentrations, may disturb the function of GPx in mantle 15

tissues. Moreover, this tissue presented the highest MDA concentrations, almost five times higher than those in gills. This observation is contrary to what was found in previous studies for B. azoricus from Menez Gwen vent field exposed to Cu (Company et al., 2008; Company et al., 2006). Nevertheless, our study differs from the former ones in several key factors such as collection depth and location, experimental pressure and Cu levels in acute exposure. All of these factors can be responsible for a dissimilar tissue response. Moreover, in mantles a trend for an increase of MDA concentrations during Cu exposure followed by a decrease under exposure to toxic concentrations (1600 µg/L) was observed. In response to membrane lipid peroxidation, cells promote their maintenance and survival through constitutive antioxidant defense system, yet under toxic conditions oxidative damage overwhelms repair capacity and the cells induce apoptosis and consequent cell death (Ayala et al., 2014). In view of this, the observed trend of LPO in mantles may reflect this specific cellular process of survival and death. The transcript expression in mantle tissues presented a transcriptional up-regulation of both antioxidant and immune-related genes after pressurization, probably a basal response to former atmospheric pressure acclimation for 48h. No expression changes were found as a result of the Cu treatments. It seems that Cu exposure does not promote a transcriptional overexpression of the genes analyzed in mantles, at least not after a 96h period. Nevertheless is noteworthy that, although not significantly up-regulated, transcriptional upregulation of the gene encoding Nuclear Factor-kappaB (NF-B) stands out among the immune-related genes expressed between Cu treatments. NF-B is a dimeric transcription factor that regulates diverse biological processes, including immune responses, inflammation, cell proliferation, and apoptosis (Morgan and Liu, 2011). It is possible that MDA functions as a signaling molecule regulating transcription factors sensitive to stress, such as NF-kB, particularly when the basal levels of antioxidant enzyme activities are not sufficient to prevent MDA concentrations from increasing (Ayala et al., 2014).

Digestive gland CAT showed the highest activity of the three analyzed tissues in digestive glands. The protective role of CAT in mussel’s digestive gland was already described in previous studies (Gonzalez-Rey and Bebianno, 2013; Martins et al., 2016). The mussel digestive 16

gland is the key organ of metabolism for producing digestive enzymes, for absorption of nutrients, for the activity of ROS-generating biotransformation enzymes handling xenobiotes, and for metal storage and detoxification (Geret and Cosson, 2002; Livingstone et al., 1992; Martins et al., 2011a). For this reason, the significantly increased CAT activity observed in digestive glands from mussels exposed to atmospheric pressure (T0) and to elevated Cu concentrations may be a consequence of cellular response to oxidative stress triggered by starvation, lack of pressure and Cu ion toxicity. MDA concentrations found in digestive gland are similar to the ones found in mantle tissue. Indeed, LPO levels in digestive glands also follow an increasing trend in mussels undergoing Cu exposure followed by a decrease during exposure to toxic concentrations (1600µg/L). Such a decline of LPO was accompanied by a decrease of antioxidant enzyme activity, which might be associated with a reduced ability of digestive gland in neutralizing the excessive amount of oxyradicals, whose cellular formation is enhanced by exposure to toxic levels of Cu. The uncontrolled oxidative damage induces first a cellular survival system and then apoptosis (Ayala et al., 2014) and therefore, similar to what was observed for mantles, the trend of LPO in digestive glands may reflect this process. The transcript expression in digestive glands, showed a general up-regulation of both antioxidant and immune genes. Nevertheless, no specific or unusual pattern of regulation was observed among Cu treatments. Similar to what was observed for mantle tissues, Cu does not seem to promote an overexpression of the analyzed genes in digestive glands during a 96h period of exposure. The fact that this tissue is not on the interface between the organism and the surrounding Cu-contaminated seawater should not be disregarded as this tissue might be internally protected from a direct exposure to a short-term dissolved Cu contamination. Moreover, the constantly up-regulated gene expression observed in this tissue may suggest that digestive glands of B. azoricus mussel are able to cope with high concentrations of Cu for a brief exposure period. Since this species lives in a highly fluctuating environment and is exposed for brief periods to anoxic hot fluids (correlated to high metal concentrations) (Le Bris et al., 2001), we could hypothesize that adaptation strategy of vent mussels includes the regulation of their basal energy costs in storage organs such as digestive glands and mantles to limit cell damages. 17

5. Conclusion Overall, our results indicate that B. azoricus mussels from Lucky Strike vent field are able to cope with high concentrations of Cu for a short period of time but with a consequent increase in basal energy expenditure and cellular injury. The B. azoricus metabolic responses to toxic concentrations of Cu show to be tissue-specific. The gills are the most responsive and impaired organ among the tissues studied, probably due to their prominent role in harboring chemosynthetic symbionts and because they are located at the interface between the internal milieu and external vent environment. These outcomes must be taken into account when discussing species sustainability, the potential impact of mining exploitation, as well as measures of protection and restoration, before starting deep-sea mining activities in the vicinity of hydrothermal vent fields. This era of new technologies allowing for the use of aggressive and high-tech machinery (www.nautilusminerals.com) will potentially disturb, in a large scale, the seabed and the associated deep-sea surrounding communities. Therefore it is predicted that such invasive exploitation will produce new sources of hydrothermal activity with consequent increase of trace metal, specifically Cu, levels near the mussel communities of the vents. Under this context, our study reveals that such sudden source of bioavailable Cu may have serious repercussions on B. azoricus basal antioxidant and immune defense system. This outcome, while adding to our knowledge for the Cu toxicity, highlights the potential impacts of mining activities in deep-sea organism physiology. To our knowledge it is the first time that an integrative approach including enzymatic and molecular analyses are used to investigate acute metal exposure impacts in cellular stress and immune status of Bathymodiolus azoricus vent mussel. Such approached proved to be insightful and should be applied in future investigations to examine how vent mussel responds to a prolonged metal exposure.

Acknowledgments IM is financed by SFRH / BPD / 73481 / 2010 grant. AC is financed by Programa Investigador FCT (IF/00029/2014/CP1230/CT0002). The authors gratefully acknowledge the captain and crew of the R/V‘‘PourquoiPas?’’and Victor 6000 ROV team, during the MoMARSAT cruise (IFREMER). The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under 18

the MIDAS project, grant agreement n° 603418. This study had the support of Fundação para a Ciência e Tecnologia (FCT), through the strategic project UID/MAR/04292/2013 granted to MARE. The authors wish to thank Renato Bettencourt, Teresa Cerqueira, Pedro Mesquita, Luis Pires and Valentina Costa for their collaborative support. IM performed the experiment and analyzed the data. JG, EM, RM and SM helped to perform laboratory analysis. VR and AC provided the samples and experimental design assistance. RB helped on experimental design and laboratory execution. IM wrote the paper with co-authors.

6. References

Anderson MJ, Gorley RN, Clarke KR, 2008. PERMANOVA + for PRIMER: guide to software and statistical methods. PRIMER-E, Plymouth, UK. Ayala, A., Munoz, M.F., Arguelles, S., 2014. Lipid Peroxidation: Production, Metabolism, and Signaling Mechanisms of Malondialdehyde and 4-Hydroxy-2-Nonenal. Oxid Med Cell Longev 2014, 31. Barros, I., Divya, B., Martins, I., Vandeperre, F., Santos, R.S., Bettencourt, R., 2015. Postcapture immune gene expression studies in the deep-sea hydrothermal vent mussel Bathymodiolus azoricus acclimatized to atmospheric pressure. Fish. Shellfish. Immunol. 42, 159-170. Beers, R.F., Sizer, I.W., 1952. A spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase. J. Biol. Chem. 195, 133-140. Bettencourt, R., Pinheiro, M., Egas, C., Gomes, P., Afonso, M., Shank, T., Santos, R.S., 2010. High-throughput sequencing and analysis of the gill tissue transcriptome from the deep-sea hydrothermal vent mussel Bathymodiolus azoricus. BMC Genomics 11. Bettencourt, R., Rodrigues, M., Barros, I., Cerqueira, T., Freitas, C., Costa, V., Pinheiro, M., Egas, C., Santos, R.S., 2014. Site-related differences in gene expression and bacterial densities in the mussel Bathymodiolus azoricus from the Menez Gwen and Lucky Strike deep-sea hydrothermal vent sites. Fish. Shellfish. Immunol. 39, 343-353. Borgmann, U., 2000. Methods for assessing the toxicological significance of metals in aquatic ecosystems: bio-accumulation-toxicity relationships, water concentrations and sediment spiking approaches. Aquatic Ecosystem Health & Management 3, 277-289. Boschen, R.E., Rowden, A.A., Clark, M.R., Gardner, J.P.A., 2013. Mining of deep-sea seafloor massive sulfides: A review of the deposits, their benthic communities, impacts from mining, regulatory frameworks and management strategies. Ocean & Coastal Management 84, 54-67. Bougerol, M., Boutet, I., LeGuen, D., Jollivet, D., Tanguy, A., 2015. Transcriptomic response of the hydrothermal mussel Bathymodiolus azoricus in experimental exposure to heavy metals is modulated by the Pgm genotype and symbiont content. Marine Genomics 21, 63-73. Boutet, I., Ripp, R., Lecompte, O., Dossat, C., Corre, E., Tanguy, A., Lallier, F.H., 2011. Conjugating effects of symbionts and environmental factors on gene expression in deep-sea hydrothermal vent mussels. BMC Genomics 12, 530.

19

Bradford, M.M., 1976. Rapid and sensitive method for quantification of microgram quantities of protein utilizing principle of protein-dye binding. Anal. Biochem. 72, 248-254. Bullen, J.J., Rogers, H.J., Spalding, P.B., Ward, C.G., 2005. Iron and infection: the heart of the matter. FEMS Immunol. Med. Microbiol. 43, 325-330. Bustin, S.A., Benes, V., Garson, J.A., Hellemans, J., Huggett, J., Kubista, M., Mueller, R., Nolan, T., Pfaffl, M.W., Shipley, G.L., Vandesompele, J., Wittwer, C.T., 2009. The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clin. Chem. 55, 611-622. Cajaraville, M.P., Bebianno, M.J., Blasco, J., Porte, C., Sarasquete, C., Viarengo, A., 2000. The use of biomarkers to assess the impact of pollution in coastal environments of the Iberian Peninsula: a practical approach. Sci. Total Environ. 247, 295-311. Charlou, J.L., Donval, J.P., Douville, E., Jean-Baptiste, P., Radford-Knoery, J., Fouquet, Y., Dapoigny, A., Stievenard, M., 2000. Compared geochemical signatures and the evolution of Menez Gwen (37º50'N) and Lucky Strike (37º17'N) hydrothermal fluids, south of the Azores Triple Junction on the Mid-Atlantic Ridge. Chem. Geol. 171, 49-75. Colaco, A., Bettencourt, R., Costa, V., Lino, S., Lopes, H., Martins, I., Pires, L., Prieto, C., Santos, R.S., 2011. LabHorta: a controlled aquarium system for monitoring physiological characteristics of the hydrothermal vent mussel Bathymodiolus azoricus. ICES J. Mar. Sci. 68, 349-356. Company, R., Serafim, A., Cosson, R.P., Fiala-Médioni, A., Camus, L., Colaço, A., Santos, R.S., Bebianno, M.J., 2008. Antioxidant biochemical responses to long-term copper exposure in Bathymodiolus azoricus from Menez-Gwen hydrothermal vent. Science of the Total Environment, 407-417. Company, R., Serafim, A., Cosson, R.P., Fiala-Médioni, A., Camus, L., Santos, R.S., Bebianno, M.J., 2010. Sub-lethal effects of cadmium on the antioxidant defense system of the hydrothermal vent mussel Bathymodiolus azoricus. Ecotoxicol. Environ. Saf. 73, 788795. Company, R., Serafim, A., Cosson, R.P., Fiala-Médioni, A., Dixon, D., Bebianno, M.J., 2006. Temporal variation in the antioxidant defense system and lipid peroxidation in the gills and mantle of hydrothermal vent mussel Bathymodiolus azoricus. Deep-Sea Research Part I 53, 1101-1116. Company, R., Serafim, A., Cosson, R.P., Fiala-Médioni, A., Dixon, D., Bebianno, M.J., 2007. Adaptation of the antioxidant defence system in hydrothermal-vent mussels (Bathymodiolus azoricus) transplanted between two Mid-Atlantic Ridge sites. Mar. Ecol. 28, 93-99. Cosson, R.P., Thiebaut, E., Company, R., Castrec-Rouelle, M., Colaço, A., Martins, I., Sarradin, P.M., Bebianno, M.J., 2008. Spatial variation of metal bioaccumulation in the hydrothermal vent mussel Bathymodiolus azoricus. Mar. Environ. Res. 65, 405-415. Cruz, J.V., Pacheco, D., Cymbron, R., Mendes, S., 2010. Monitoring of the groundwater chemical status in the Azores archipelago (Portugal) in the context of the EU water framework directive. Environ. Earth. Sci. 61, 173-186. Cuvelier, D., Sarrazin, J., Colaço, A., Copley, J., Desbruyères, D., Glover, A.G., Tyler, P., Santos, R.S., 2009. Distribution and spatial variation of hydrothermal faunal assemblages at Lucky Strike (Mid-Atlantic Ridge) revealed by high-resolution video image analysis. DeepSea Research Part I 56, 2026-2040.

20

Detree, C., Chabenat, A., Lallier, F.H., Satoh, N., Shoguchi, E., Tanguy, A., Mary, J., 2016. Multiple I-Type Lysozymes in the Hydrothermal Vent Mussel Bathymodiolus azoricus and Their Role in Symbiotic Plasticity. Plos One 11, e0148988. Dubinina, E.E., Gavrovskaya, S.V., Kuzmich, E.V., Leonova, N.V., Morozova, M.G., Kovrugina, S.V., Smirnova, T.A., 2002. Oxidative Modification of Proteins: Oxidation of Tryptophan and Production of Dityrosine in Purified Proteins Using Fenton's System. Biochemistry (Moscow) 67, 343-350. Duperron, S., Bergin, C., Zielinski, F., Blazejak, A., Pernthaler, A., McKiness, Z.P., DeChaine, E., Cavanaugh, C.M., Dubilier, N., 2006. A dual symbiosis shared by two mussel species, Bathymodiolus azoricus and Bathymodiolus puteoserpentis (Bivalvia : Mytilidae), from hydrothermal vents along the northern Mid-Atlantic Ridge. Environ. Microbiol. 8, 1441-1447. Fiala-Médioni, A., McKiness, Z.P., Dando, P., Boulegue, J., Mariotti, A., Alayse-Danet, A.M., Robinson, J.J., Cavanaugh, C.M., 2002. Ultrastructural, biochemical, and immunological characterization of two populations of the mytilid mussel Bathymodiolus azoricus from the Mid-Atlantic Ridge: evidence for a dual symbiosis. Mar. Biol. 141, 10351043. Gaetke, L.M., Chow, C.K., 2003. Copper toxicity, oxidative stress, and antioxidant nutrients. Toxicology 189, 147-163. Geret, F., Cosson, R.P., 2002. Induction of specific isoforms of metallothionein in mussel tissues after exposure to cadmium or mercury. Arch. Environ. Contam. Toxicol. 42, 36-42. Gonzalez-Rey, M., Bebianno, M.J., 2013. Does selective serotonin reuptake inhibitor (SSRI) fluoxetine affects mussel Mytilus galloprovincialis? Environ. Pollut. 173, 200-209. Gonzalez, P.M., Abele, D., Puntarulo, S., 2010. Exposure to excess dissolved iron in vivo affects oxidative status in the bivalve Mya arenaria. Comp Biochem Phys C 152, 167-174. Gutierrez, J.C., Martin-Gonzalez, A., Diaz, S., Ortega, R., 2003. Ciliates as a potential source of cellular and molecular biomarkers/biosensors for heavy metal pollution. Eur J Protistol 39, 461-467. Halfar, J., Fujita, R.M., 2007. Danger of Deep-Sea Mining. Science 316, 987-987. Johnson, K.S., Childress, J.J., Beehler, C.L., 1988. Short-term temperature variability in the Rose Garden hydrothermal vent field- an unstable deep-sea environment. Deep-Sea Research Part I 35, 1711-1721. Kádár, E., Bettencourt, R., Costa, V., Santos, R.S., Lobo-Da-Cunha, A., Dando, P., 2005. Experimentally induced endosymbiont loss and re-acquirement in the hydrothermal vent bivalve Bathymodiolus azoricus. J. Exp. Mar. Biol. Ecol. 318, 99-110. Kalghatgi, S., Spina, C.S., Costello, J.C., Liesa, M., Morones-Ramirez, J.R., Slomovic, S., Molina, A., Shirihai, O.S., Collins, J.J., 2013. Bactericidal Antibiotics Induce Mitochondrial Dysfunction and Oxidative Damage in Mammalian Cells. Science Translational Medicine 5. Kim, B.M., Rhee, J.S., Jeong, C.B., Seo, J.S., Park, G.S., Lee, Y.M., Lee, J.S., 2014. Heavy metals induce oxidative stress and trigger oxidative stress-mediated heat shock protein (hsp) modulation in the intertidal copepod Tigriopus japonicus. Comp Biochem Phys C 166, 65-74. Kubista, M., Andrade, J.M., Bengtsson, M., Forootan, A., Jonák, J., Lind, K., Sindelka, R., Sjöback, R., Sjögreen, B., Strömbom, L., Ståhlberg, A., Zoric, N., 2006. The real-time polymerase chain reaction. Mol. Aspects Med. 27, 95-125. 21

Le Bris, N., Sarradin, P.-M., Pennec, S., 2001. A new deep-sea probe for in situ pH measurement in the environment of hydrothermal vent biological communities. Deep-Sea Research Part I 48, 1941-1951. Leonard, S.S., Harris, G.K., Shi, X.L., 2004. Metal-induced oxidative stress and signal transduction. Free Radic. Biol. Med. 37, 1921-1942. Livak, K.J., Schmittgen, T.D., 2001. Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods 25, 402-408. Livingstone, D.R., Lips, F., Martinez, P.G., Pipe, R.K., 1992. Antioxidant enzymes in the digestive gland of the common musselMytilus edulis. Mar. Biol. 112, 265-276. Lushchak, V.I., 2011. Environmentally induced oxidative stress in aquatic animals. Aquat. Toxicol. 101, 13-30. Martins, E., Santos, R.S., Bettencourt, R., 2015. Vibrio diabolicus challenge in Bathymodiolus azoricus populations from Menez Gwen and Lucky Strike hydrothermal vent sites. Fish. Shellfish. Immun. 47, 962-977. Martins, I., Bettencourt, R., Colaco, A., Sarradin, P.-M., Santos, R.S., Cosson, R., 2011a. The influence of nutritional conditions on metal uptake by the mixotrophic dual symbiosis harboring vent mussel Bathymodiolus azoricus. Comp. Biochem. Physiol. C. 153, 40-52. Martins, I., Colaco, A., Lesongeur, F., Godfroy, A., Sarradin, P.-M., Casson, R.P., Santos, R.S., 2008. Metal exposure and physiological condition of the mussel, Bathymodiolus azoricus, from the Eiffel Tower hydrothermal vent site. Mar. Environ. Res. 66, 42-42. Martins, I., Colaço, A., Santos, R.S., Lesongeur, F., Godfroy, A., Sarradin, P.M., Cosson, R.P., 2009. Relationship between the occurrence of filamentous bacteria on Bathymodiolus azoricus shell and the physiological and toxicological status of the vent mussel. J. Exp. Mar. Biol. Ecol. 376, 1-6. Martins, I., Cosson, R.P., Riou, V., Sarradin, P.-M., Sarrazin, J., Santos, R.S., Colaco, A., 2011b. Relationship between metal levels in the vent mussel Bathymodiolus azoricus and local microhabitat chemical characteristics of Eiffel Tower (Lucky Strike). Deep-Sea Res. Pt I 58, 306-315. Martins, I., Romao, C.V., Goulart, J., Cerqueira, T., Santos, R.S., Bettencourt, R., 2016. Activity of antioxidant enzymes in response to atmospheric pressure induced physiological stress in deep-sea hydrothermal vent mussel Bathymodiolus azoricus. Mar. Environ. Res. 114, 65-73. Morgan, M.J., Liu, Z.-g., 2011. Crosstalk of reactive oxygen species and NF-kB signaling. Cell Res. 21, 103-115. Paglia, D., Valentine, W., 1967. Studies on quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J. Lab. Clin. Med. 70, 158-169. Pallavi Sharma, Ambuj Bhushan Jha, Rama Shanker Dubey, Pessarakli, M., 2012. Reactive Oxygen Species, Oxidative Damage, and Antioxidative Defense Mechanism in Plants under Stressful Conditions. Journal of Botany 2012, 26. Pfaffl, M.W., 2004. Quantification strategies in real-time PCR, in: Bustin, S.A. (Ed.), The Real-Time PCR Encyclopedia A–Z of Quantitative PCR. International University Line (IUL) La Jolla, pp. 87–112. Ponnudurai, R., Kleiner, M., Sayavedra, L., Petersen, J.M., Moche, M., Otto, A., Becher, D., Takeuchi, T., Satoh, N., Dubilier, N., Schweder, T., Markert, S., 2017. Metabolic and physiological interdependencies in the Bathymodiolus azoricus symbiosis. ISME J 11, 463477. 22

Pulido, M.D., Parrish, A.R., 2003. Metal-induced apoptosis: mechanisms. Mutation Research-Fundamental and Molecular Mechanisms of Mutagenesis 533, 227-241. Riou, V., Colaço, A., Bouillon, S., Khripounoff, A., Dando, P.R., Mangion, P., Chevalier, E., Korntheuer, M., Serrão Santos, R., Dehairs, F.A., 2010. Mixotrophy in the deep sea: a dual endosymbiotic hydrothermal mytilid assimilates dissolved and particulate organic matter. Mar Ecol Prog Ser 405. Riou, V., Halary, S., Duperron, S., Bouillon, S., Elskens, M., Bettencourt, R., Santos, R.S., Dehairs, F., Colaço, A., 2008. Influence of CH4 and H2S availability on symbiont distribution, carbon assimilation and transfer in the dual symbiotic vent mussel Bathymodiolus azoricus. Biogeosciences 5, 1681-1691. Roberts, D.A., 2012. Causes and ecological effects of resuspended contaminated sediments (RCS) in marine environments. Environ. Int. 40, 230-243. Sarradin, P.-M., Caprais, J.-C., Riso, R., Kerouel, R., Aminot, A., 1999. Chemical environment of the hydrothermal mussel communities in the Lucky Strike and Menez Gwen vent fields, Mid Atlantic Ridge. Cah. Biol. Mar. 40, 93-104. Sarradin, P.M., Waeles, M., Bernagout, S., Le Gall, C., Sarrazin, J., Riso, R., 2009. Speciation of dissolved copper within an active hydrothermal edifice on the Lucky Strike vent field (MAR, 37 degrees N). Sci Tot Environ 407. Sayavedra, L., Kleiner, M., Ponnudurai, R., Wetzel, S., Pelletier, E., Barbe, V., Satoh, N., Shoguchi, E., Fink, D., Breusing, C., Reusch, T.B.H., Rosenstiel, P., Schilhabel, M.B., Becher, D., Schweder, T., Markert, S., Dubilier, N., Petersen, J.M., 2015. Abundant toxinrelated genes in the genomes of beneficial symbionts from deep-sea hydrothermal vent mussels. Elife 4. Shillito, B., Gaill, F., Ravaux, J., 2014. The IPOCAMP pressure incubator for deep-sea fauna. Journal of Marine Science and Technology-Taiwan 22, 97-102. Soldatov, A.A., Gostyukhina, O.L., Golovina, I.V., 2007. Antioxidant enzyme complex of tissues of the bivalve Mytilus galloprovincialis Lam. under normal and oxidative-stress conditions: A review. Applied Biochemistry and Microbiology 43, 556-562. Therond, P., Gerbaud, P., Dimon, S., Anderson, W.B., EvainBrion, D., Raynaud, F., 1996. Antioxidant enzymes in psoriatic fibroblasts and erythrocytes. J. Invest. Dermatol. 106, 1325-1328. Tunnicliffe, V., 1991. The Biology of Hydrothermal Vents - Ecology and Evolution. Oceanography and Marine Biology 29, 319-407. Turski, M.L., Thiele, D.J., 2009. New Roles for Copper Metabolism in Cell Proliferation, Signaling, and Disease. The Journal of Biological Chemistry 284, 717-721. Verkhovtseva, Y.V., Filina, Y.Y., Pukhov, D.E., 2001. Evolutionary role of iron in metabolism of prokaryotes and in biogeochemical processes. J Evol Biochem Phys+ 37, 444-450. Walters, K., Coen, L.D., 2006. A comparison of statistical approaches to analyzing community convergence between natural and constructed oyster reefs. J. Exp. Mar. Biol. Ecol. 330, 81-95. Zhou, Q., Zhang, Y., Peng, H.F., Ke, C.H., Huang, H.Q., 2014. Toxicological responses of the hard clam Meretrix meretrix exposed to excess dissolved iron or challenged by Vibrio parahaemolyticus. Aquat. Toxicol. 156, 240-247.

23

Fig. 1 – Panel A: mean specific activities of glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT) and Panel B: lipid peroxidation (LPO) (as given by MDA concentration) in gills of B. azoricus mussels from: unpressured and non-contaminated (T0) (n=8); pressurized with no Cu contamination (0 µg/L) (n=6) and Cu exposure with 300 (n=6), 800 (n=7) and 1600 µg/L (n=7). Vertical bars represent the standard deviation of the mean. Symbol (*) indicates statistically significant differences among treatments (PERMANOVA, pairwise test, p <0.05)

Fig. 2 - Panel A: mean specific activities of glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT) and Panel B: lipid peroxidation (LPO) (as given by MDA concentration) in mantles of B. azoricus mussels from: unpressured and non-contaminated (T0) (n=8); pressurized with no Cu contamination (0 µg/L) (n=6) and Cu exposure with 300 (n=6), 800 (n=7) and 1600 µg/L (n=7). Vertical bars represent the standard deviation of the mean. Symbol (**) indicates statistically significant difference among treatments (PERMANOVA, pairwise test p <0.01)

24

Fig. 3 - Panel A: mean specific activities of glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT) and Panel B: lipid peroxidation (LPO) concentration in digestive gland of B. azoricus mussels from: unpressured and non-contaminated (T0) (n=8); pressurized with no Cu contamination (0 µg/L) (n=6) and Cu exposure with 300 (n=6), 800 (n=7) and 1600 µg/L (n=7). Vertical bars represent the standard deviation of the mean. Symbol (**) indicates statistically significant difference among treatments (PERMANOVA, pairwise test p <0.01)

Fig. 4 – Mean normalized expression (with 28S rRNA as the reference gene) of antioxidant related genes (A) and immune related genes (B) in gills (n=3, pool of 2 gills in each replicate) of B. azoricus mussels from no Cu contamination (0 µg/L) and Cu contamination with 300, 800 and 1600 µg/L. Vertical bars represent the standard deviation of the mean. Symbols (*) and (**) indicate significant statistical difference among treatments (PERMANOVA, pairwise test p <0.05 and p <0.001, respectively). Abbreviations: CAT the gene encoding catalase; SOD the gene encoding superoxide dismutase; Ferritin the gene encoding ferritin; GPx the gene encoding glutathione peroxidase; HSP70 the gene encoding heat shock protein 70; MT the gene encoding a metallothionein; ILR2 the gene encoding immune lectin receptor 2; PGRP the gene encoding peptidoglycan recognition protein; RTK genes encoding receptor tyrosine kinases; TLR2 the gene encoding toll-like receptor 2; LITAF the gene encoding lipopolysaccharide-induced tumor necrosis factor-alpha factor; NF-kB the gene encoding nuclear-Factor kappa B.

25

Fig. 5 – Mean normalized expression (with 28S rRNA) of antioxidant related genes (A) and immune related genes (B) in mantles (n=3, pool of 2 mantles in each replicate) of B. azoricus mussels from no Cu contamination (0 µg/L) and Cu contamination with 300, 800 and 1600 µg/L. Vertical bars represent the standard deviation of the mean. Abbreviations: CAT the gene encoding catalase; SOD the gene encoding superoxide dismutase; Ferritin the gene encoding ferritin; GPx the gene encoding glutathione peroxidase; HSP70 the gene encoding heat shock protein 70; MT the gene encoding a metallothionein; ILR2 the gene encoding immune lectin receptor 2; PGRP the gene encoding peptidoglycan recognition protein; RTK genes encoding receptor tyrosine kinases; TLR2 the gene encoding toll-like receptor 2; LITAF the gene encoding lipopolysaccharide-induced tumor necrosis factor-alpha factor; NF-kB the gene encoding nuclear-Factor kappa B.

Fig. 6 – Mean normalized expression (with 28S rRNA) of antioxidant related genes (A) and immune related genes (B) in digestive gland (n=3) of B. azoricus mussels from no Cu contamination (0 µg/L) and Cu contamination with 300, 800 and 1600 µg/L. Vertical bars represent the standard deviation of the mean. Abbreviations: CAT the gene encoding catalase; SOD the gene encoding superoxide dismutase; Ferritin the gene encoding ferritin; GPx the gene encoding glutathione peroxidase; HSP70 the gene encoding heat shock protein 70; MT the gene encoding a metallothionein; ILR2 the gene encoding immune lectin receptor 2; PGRP the gene encoding peptidoglycan recognition protein; RTK genes encoding receptor tyrosine kinases;

26

TLR2 the gene encoding toll-like receptor 2; LITAF the gene encoding lipopolysaccharide-induced tumor necrosis factor-alpha factor; NF-kB the gene encoding nuclear-Factor kappa B.

Table 1. Gene names, molecular pathway and GenBank accession numbers for each gene used in qPCR analyses. Gene name (symbol) 28S ribosomal RNA (28S) 18S ribosomal RNA (18S) Catalase (CAT) Superoxide dismutase (SOD) Ferritin Glutathione peroxidase (GPx) Heat shock protein (HSP70) Metallothionein (MT)

Molecular pathway* RNA binding

Accession nº

RNA binding

AY649822.1

Oxidative damage protection Oxidative damage protection Iron metabolism Oxidative damage protection Stress response

HM756152.1

Metal detoxification

HM756147.1

AY781148.1

JN863296.1 FJ767378.1 HM756144.1 HM756159.1

27

Immune lectin receptor 2 (ILR2) Signaling Peptidoglycan recognition protein (PGRP) Receptor tyrosine kinases (RTK) Toll-like receptor 2 (TLR2) Lipopolysaccharide-induced tumor necrosis factor-alpha factor (LITAF)

Recognition

mussel_rep_c 70917a HM756116.1

Signaling Signaling

HM756159.1 HM756129.1

Signaling

HM756126.1

Signaling HM756140.1 Nuclear-Factor kappa B (NF-κB) *According to http://www.geneontology.org/ and http://www.uniprot.org/ a

DeepSea vent (http://transcriptomics.biocant.pt/deepSeaVent) accession code

28

Table 2. Primers sequences, amplicons size (bp), efficiency (E%) and correlation coefficients (R2) from the standard curves generated for each primer pair to estimate efficiencies. Gene

Primer Forward primer (5´- 3´)

sequences Reverse primer (5´- 3´)

Product size (bp)

PCR efficiency (E%)

28S rRNA 18S rRNA CAT SOD Ferritin GPx HSP70 MT ILR2 PGRP RTK TLR2 LITAF NF-κB

TTCTTTTCACTTTCCCTCACG TCAACACGGGAAAACTCACC CATGTTAGCAGGCACTCCAGA GATGAGACGATCAGCCTTC TTCGATAGGGATGACGTAGC AACAGTTTAAGATTTCC TGAAGAAAATGTGTGGTGACTTG TGCCCAAAGAACAAAGGATG TGGACACTGCTACCATTATGGGACC TGTTGGTGAAGATGGCAAAA TGAAGAAATGTGTGGTGACTTG CAGGAGGACTCGGATGACAC ATGAGAGATACCCCCGTGAA AGTGGCGTATCACCGTTACA

CTTGGAGTCGGGTTGTTGA AACCAGACAAATCGCTCCAC AAGCTGACCCAGAATATGGACA GTCCAGCATTACCTCCCTTT TTTCATCAGCTTTTCAGCATGT TGGCTTCTCTCTGAGGAACAACTG CCCTACCAGAACGACCTCAT TTTCCACAACCACTTCCACA CGATTGGTCATAGCTCCAACGCC CCGTGAGTGGTGTGTGTGT CCCTACCAGAACGACCTCAT TCTCCAGTCAGTGGTGCAAG CACAAAACAACACCCAGCAT GGCTGTGTTTGGTTGGACAT

113 121 95 124 93 150 109 104 208 131 109 107 144 149

101 53.1 115 105 115 90 97 115 97 115 115 110 105 105

Correlation Coefficient (R2) 0.950 0.967 0.979 0.998 0.998 0.985 0.992 0.988 0.997 0.996 0.995 0.995 0.997 0.999

29