Does salinity modulates the response of Mytilus galloprovincialis exposed to triclosan and diclofenac?

In the present study Mytilus galloprovincialis mussels were exposed for 28 days to three salinities: 30 (control), 25 and 35. Simultaneously, organisms at each salinity were exposed to either the antimicrobial agent Triclosan (TCS) or the pharmaceutical drug Diclofenac (DIC) at 1 μg/L. Salinity alone and exposure to PPCPs changed mussel's metabolic capacity and oxidative status, but no additive or synergetic effects resulting from the combined exposures were observed. Overall, the metabolic capacity of mussels was decreased when exposed to TCS and DIC under control salinity, which was less pronounced at salinities out of the control level. TCS had a notorious effect over glutathione peroxidase activity while DIC exposure enhanced catalase response. Such defence mechanisms were able to prevent cellular damage but still a clear reduction in GSH/GSSG ratio after PPCPs exposures indicates oxidative stress which could compromise bivalve's performance to further stressing events.

Decreased salinity • Mussels lowered their metabolic rate after drug exposures at control salinity • Mussels increased antioxidant defences when exposed to drugs at all salinities • GSH/GSSG ratio was consistently reduced when mussels were exposed to TCS and DIC. 57 Matozzo et al., 2012) and DIC (Fontes et al., 2018;Gonzalez-Rey and Bebianno, 2014; 58 Goodchild et al., 2016;Mezzelani et al., 2016;Mezzelani et al., 2018;Munari et al., 2018; In the aquatic environment pollutants are not acting alone, with environmental changes 61 playing an important role on their fate and impacts, with changes on species sensitivity and 62 pollutants toxicity. In particular, recent studies already demonstrated that increased sensitivity of 63 invertebrates to pharmaceuticals, nanoparticles or metals may result from exposure to climate 64 change related factors, such as seawater acidification and warming (Coppola et al., 2018; M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 4 example, the transformation behaviour of ibuprofen differed between freshwater and seawater 70 (Weigel et al., 2004) and prochlorperazine was more stable in seawater than freshwater 71 (Spongberg et al., 2011). Still, little is known on the impact of climate change events, in 72 particular salinity changes, on the toxicity of PPCPs within the marine environment.

73
Biomarkers related to metabolic capacity, available energy reserves, oxidative stress 74 defences and damage are very informative as they reveal the capacity of organisms to face 75 challenging situations derived from chemical exposures and/or physical unfavourable conditions 76 (Monserrat et al., 2007;Regoli and Giuliani, 2014). Marine bivalves and mussels in particular

134
The extracts where then analyzed with a GC-MS method (Tohidi and Cai, 2015).

135
Calibration curve was performed with TCS standard (Sigma-Aldrich) in dichloromethane. All 136 samples were analyzed by the use of a GC Trace 1300 (Thermo Scientific) coupled to a TriPlus 137 RSH autosample and a triple quadrupole mass spectrometer TSQ Duo with an electron impact 138 ionization source (EI) (Thermo Scientific); the column was an Agilent DB-5MS. The detection 139 limit (LOD), calculated as a signal-to-noise ratio of 3:1, was 0.008 µg/L for water samples and 140 0.13 ng/g d.w. for soft tissues. The percent of recovery was >91 for water samples and >89 for 141 soft tissues (see Table 1).

142
High performance liquid chromatography-ultraviolet (HPLC-UV) detection was used for

157
UV detector, which was set at 280 nm, was used as HPLC system. The mobile phase consisted 158 of acetonitrile and 0.2% formic acid in water, at a ratio of 60:40 (v:v). The reversed-phase 159 column was a Haisil, LC column (5 µm, 150x4.60 mm, Higgins). The column was kept at room 160 temperature. Turbochrome software was used for data processing. The DIC recovery was 161 >80% for water samples and >77% for soft tissues. The detection limit, calculated as a signal-162 to-noise ratio of 3:1, was 0.10 µg/L for water samples and 5 ng/g d.w. for soft tissues (see Table   163 1).

Biochemical parameters 166
After 28 days exposure, mussels used for biomarker analysis (2 per replicate, 8 per

Triclosan and Diclofenac concentrations 203
In Table 1

Metabolic capacity and energy reserves 213
ETS activity in control and exposed mussels is illustrated in Figure 1A. At salinity 25 214 mussels exposed to DIC significantly increased ETS activity in respect to control organisms. By 215 contrast, at salinities 30 and 35 contaminated mussels tended to decrease ETS activity in 216 comparison to unexposed specimens, with significantly lower values in mussels exposed to 217 TCS and DIC at salinity 30 and mussels exposed to TCS at salinity 35. ETS values observed in 218 unexposed mussels were significantly lower at salinity 25 compared to salinities 30 and 35 219 (Table 2).

220
GLY content ( Figure 1B) increased in mussels contaminated with TCS and DIC at salinity 221 30 and mussels exposed to DIC at salinity 35 in respect to their controls. Except between 222 CTL25 and CTL35, GLY values varied significantly among salinities for non-contaminated 223 mussels, with significantly lower values at salinity 30 (Table 2).

224
At salinities 25 and 30 mussels exposed to DIC showed significantly higher PROT 225 content in comparison to mussels exposed to TCS or unexposed ( Figure 1C). At salinity 35 226 mussels exposed to TCS showed significantly lower PROT content in comparison to CTL 227 organisms. PROT values were significantly different between control mussels maintained at 10 SOD activity (Figure 2A) between non-contaminated (CTL) and contaminated mussels 232 (TCS and DIC) was only enhanced at the control salinity (30). SOD activity also varied 233 significantly among salinities for unexposed mussels, with the lowest values attained in mussels 234 held at salinity 30 (Table 2).

235
CAT activity ( Figure 2B) was enhanced regardless of the salinity tested, in mussels 236 exposed to TCS (at salinity 30 and 35) and DIC (at 25 and 30) in comparison to their controls.

237
CAT activity in unexposed organisms also varied significantly as a function of salinity, with the 238 lowest value also at control salinity (30). Organisms exposed to DIC showed significantly higher 239 values at salinity 25 compared to salinities 30 and 35 (Table 2).

240
GPx activity ( Figure 2C) was strongly enhanced at all salinities in mussels exposed to 241 TCS but also DIC in respect to their controls, with higher values in contaminated mussels 242 maintained at salinity 25. Unexposed mussels at salinity 25 showed significantly lower GPx 243 values than non-contaminated organisms at salinities 30 and 35. Mussels exposed to TCS and 244 salinity 35 showed significantly lower GPx values than organisms at salinities 25 and 30 (Table   245 2).

246
GSTs activity (Figure 3), contaminated mussels tended to increase their activity in 247 comparison to CTL organisms, with significantly higher values in organisms exposed to DIC (at 248 salinity 25) and to TCS and DIC (at salinity 30). Unexposed mussels at salinity 35 showed 249 significantly higher GSTs values than non-contaminated organisms at salinities 25 and 30. A 250 similar trend was observed in organisms exposed to TCS while exposure to DIC was 251 responsible for higher GSTs values at salinity 30 in respect to the others (Table 2).

Indicators of cellular damage 254
LPO levels ( Figure 4A) in mussels at salinity 25 and exposed to TCS and DIC 255 significantly increased in comparison to non-contamianted organisms. By contrast at salinities 256 30 and 35 contaminated mussels tended to decrease their LPO levels in comparison to non-257 contaminated mussels, with significantly lower values in mussels exposed to TCS and DIC at 258 salinity 30 and mussels exposed to DIC at salinity 35. Salinity levels and PPCP exposures affected LPO levels in mussels, with the exception for organisms exposed to DIC at salinities 30 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 11 and 35 where no significant differences were observed between mussels exposed to these 261 conditions (DIC 30 vs DIC 35) ( Table 2).

262
Regardless of salinity, exposure to PPCPs caused significantly lower GSH/GSSG 263 values in comparison to their respective controls, with the highest reduction attained after 264 exposure to DIC ( Figure 4B). Unexposed mussels showed significantly higher GSH/GSSG ratio 265 at salinity 35 in comparison to salinity 30. Less evident although still significant, mussels 266 exposed to TCS and salinity 30 showed significantly lower ratio values than those at salinities 267 25 and 35. Exposure to DIC caused significant lower GSH/GSSG values in mussels at salinity 268 25 in respect to those held at salinity of 35 (Table 2).

DISCUSSION
levels were more variable than those corresponding to DIC (Table 1)

280
Bioaccumulated TCS in whole mussels tissue was highly dependent on the salinity of the

286
Salinity influences sorption and therefore bioavailability of hydrophobic chemicals, 287 including TCS (Wu et al., 2016;Xie et al., 2008). In this case, the lower salinity could have 288 determined the higher bioavailability of TCS that explains its higher bioaccumulation in mussels

310
In order to assess if PPCPs levels in mussels tissues were able to modify their responses 311 in a salinity-dependent manner, biomarkers related to energy balance, oxidative defences and

325
Our findings further revealed that under control salinity the impacts of PPCPs were 326 noticed, especially with the reduction of mussel's metabolism and increased energy reserves 327 content, increased antioxidant and biotransformation enzymes activities and lower LPO.

328
Although the magnitude of ROS production was not measured in the present study, all the 329 antioxidant defences considered (SOD, CAT, GPx and GSTs) were enhanced at the natural M A N U S C R I P T

A C C E P T E D ACCEPTED MANUSCRIPT
14 which probably results from the combined reduced metabolism (lower ETS) and the efficient 333 action of the antioxidant responses in comparison to control organisms. Moreover, a strong 334 decrease in the ratio GSH/GSSG was observed in mussels exposed to TCS and DIC in 335 comparison to unexposed ones, revealing a general increase of the oxidative status in M. 336 galloprovincialis exposed to those PPCPs. This ratio is considered as a reliable biomarker for 337 monitoring the effects of xenobiotics (van der Oost et al., 2003). In the present study the ratio

373
indicating that the effects of both PPCPs were also observed at freshwater conditions. 374 375

376
The present results clearly revealed metabolic and oxidative stress impacts of both TCS 377 and DIC in M. galloprovincialis, regardless the salinity tested. In fact, salinity changes alone 378 were responsible for more metabolic and oxidative parameter responses in mussels than the 379 PPCPs themselves. DIC showed preferentially enhanced CAT activity while TCS strongly 380 increased GPx activity and both PPCPs caused enhanced GSTs activities. Damage measured 381 as increased LPO levels was evident only at the lowest salinity while the GSH/GSSG balance 382 was the parameter more consistently affected by salinity changes and PPCPs exposures.               Table 1. Water and tissue concentrations of Triclosan and Diclofenac. Water samples were analysed soon after spiking while tissue samples were analysed after 28-days exposure period.
Water and tissue samples at control conditions presented PPCPs lower than the LOQ. LOD: limit of detection; LOQ: limit of quantification. Different letters represent significant differences among salinity levels, for each PPCP (Triclosan or Diclofenac) and sample type (water or tiss ue). M A N U S C R I P T