2015 Ecological risk assessment of PAHs - Ghent University Library [PDF]

Faculty of Bioscience Engineering Academic year 2014 – 2015

Ecological risk assessment of PAHs: integration of regression-based models and experimental approaches

Oscar Julian Velasquez Ballesteros Promoters: Prof. dr. Colin Janssen Prof. dr. ir. Peter Goethals Tutor: dr. ir. Gert Everaert

Master’s dissertation submitted in partial fulfilment of the requirements for the degree of Master of Science in Environmental Sanitation

i COPYRIGHT The author, the promoter and the tutor give permission to use this thesis for consultation and to copy parts of it for personal use. Any other use is subjected to the laws of Copyright. Permission to produce any material contained in this work should be obtained from the author.

© Ghent University, June 2015

The Promoters:

Prof. dr. Colin Janssen

Prof. dr. ir. Peter Goethals

The Tutor:

Gert Everaert

The Author:

Oscar Julian Velasquez Ballesteros

ii

iii Acknowledgement First of all I want to offer thankfulness to God for giving me to give me the necessary strength to live every day and for helping me in every part of this successful work. I take this opportunity to express my gratitude to all the people who helped me during these two years full of great academic and personal experiences. A special thanks to my promoters, Professor Prof. dr. Colin Janssen and Prof. dr. ir. Peter Goethals for giving me the opportunity to being part of the project and work in the laboratories. To my tutor Gert Everaert for his patience, great guidance and for the crucial feedback in moments it was really needed. Special thanks to my parents Amparo Ballesteros and Esteban Velasquez, without their support and love this would not be possible. To all my dear family loves, Gloria Ballesteros, Cristina Ballesteros, Felipe Ballesteros and Roberto Ballesteros. To my best friends John Isaac Medina, Michael Avila, Nathalie Guarnizo, Diego Oliveros and Carol Garcia for their constant support and encouragement along my studies. To all professors and assistants that during this 2 years have taught me many things that would be useful during my professional development. Thanks to Sylvie and Veerle for their support and advice during these 2 academic years, to Andrea Lopez Karolys and Laura Espinoza Garcia for being my support and family, and to all my friends in Gent who were my support and with whom I experienced many rewarding experiences.

iv

v

Abstract The present study is focused on the possible adverse effect of a mixture of 8 Polycyclic Aromatic Hydrocarbons (anthracene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(a)pyrene, chrysene, fluorene, phenanthrene, pyrene) on the growth of a marine phytoplankton species. PAHs are organic pollutants derived from oil, its derivatives and the incomplete but high-temperature combustion of organic matter, they are largely prevalent in aquatic environments and are frequently associated with soot carbon. They can accumulate in different environmental matrices because of their very low water solubility (hydrophobicity) and strong affinity for organic matter, which allows them to bound to particulate organic matter and/or organism’s tissues. Moreover they have being found to exert some ecotoxicological effects (i.e. genetic damage and different physiological cellular change). To determine the possible negative effect of such mixture of PAHs, a 72h algal growth inhibition test with Phaeodactylum tricornutum with controlled conditions (20ºC and 60 µmol m-2 to 120 µmol m-2 photon fluence) was performed. This test resulted in a 50% grow inhibition (effect concentration) of 3,300 ng L-1 (3 µg L-1). On the exposure side, data from the International Council for the Exploration of the Sea (ICES) were used to characterize time trends and quantify the exposure concentrations of the sixteen USEPA priority PAHs. Based on 28 years (1985-2013) of data it was found that the sum of aqueous concentrations of the 16 PAHs decreased from 4.98 μg L-1 (4.86-5.09 μg L-1) in 1987 to 0.88 μg L-1 (0.81-0.95 μg L-1) in 2013. To determine the potential risk of 8 PAHs (anthracene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(a)pyrene, chrysene, fluorene, phenanthrene, pyrene), both exposure concentrations and the effect concentration (3.3 µg L-1) were compared. Overall it was concluded that the risk quotients were lower than 1, meaning that there is no potential risk of the 8 PAHs on the marine ecosystems including phytoplankton and other superior species of the food web. Furthermore, it was found that, ambient concentration of 6 out of the 8 PAHs were lower than the Environmental Assessment criteria (Mussels; anthracene: 290 µg kg-1 d.w., benzo(a)anthracene: 80 µg kg-1 d.w., benzo(b)fluoranthene: NA, benzo(a)pyrene: 600 µg kg-1 d.w., chrysene: NA, fluorene: NA, phenanthrene: 1700 µg kg-1 d.w., pyrene: 100 µg kg-1 d.w.), Effect Range Low (Sediment; anthracene: 85 µg kg-1 d.w., benzo(a)anthracene: 261 µg kg-1 d.w., benzo(b)fluoranthene: NA, benzo(a)pyrene: 430 µg kg-1 d.w., chrysene: 384 µg kg-1 d.w., fluorene: NA, phenanthrene: 240 µg kg-1 d.w., pyrene: 665 µg kg-1 µg kg-1 d.w.) and Background Assessment concentrations (Sediment; anthracene: 5 µg kg-1 d.w., benzo(a)anthracene: 16 µg kg-1 d.w., benzo(b)fluoranthene: NA, benzo(a)pyrene: 30 µg kg-1 d.w., chrysene: 20 µg kg-1 d.w., fluorene: NA, phenanthrene: 32 µg kg-1 d.w., pyrene: 24 µg kg-1 d.w.; Mussels; anthracene: 2.7 µg kg-1 d.w., benzo(a)anthracene: 3.6 µg kg1 d.w., benzo(b)fluoranthene: NA, benzo(a)pyrene: 2.1 µg kg-1 d.w., chrysene: 21.8 µg kg-1 d.w.,

vi fluorene: NA, phenanthrene: 12.6 µg kg-1 d.w., pyrene: 10.1 µg kg-1 d.w.) proposed by the OSPAR Co-ordinated Environmental Monitoring Programme (CEMP). The results of the present research enrich the knowledge of the current state of the phytoplankton (a key player role in ecology) in relation with the presence of organic pollutants such as PAHs. However, more experiments including different trophic levels and other environmental variables should be done.

Key words: PAHs, Exposure Assessment, Effect assessment, Risk assessment, Time trends

vii Table of contents Copyright ………………………………………………………………………

i

Acknowledgment …………………………………………………………….

iii

Abstract ………………………………………………………………………..

v

Table of contents …………………………………………………………….

vii

List of tables …………………………………………………………………..

x

List of figures …………………………………………………………………

xii

List of abbreviations …………………………………………………………

xiv

PART I

Introduction ……………………………………………………………………

1

PART II 2.1 2.1.1 2.1.2 2.2 2.2.1 2.2.2 2.2.3 2.2.4

Literature review ……………………………………………………………... Primary production: mechanisms ………………………………………... Temporal differences ………………………………………................... Primary production: currents and patterns ……………………………. Polycyclic aromatic hydrocarbons ………………………………………... Distribution of PAHs ……………………………………….................... Bioconcentration, bioaccumulation and biomagnification Ecotoxicological effects ………………………………………............... Time trends and legislation ……………………………………….........

3 3 4 5 7 10 11 14 16

PART III

Problem formulation – Rationale for this thesis ……………................

18

PART IV

Materials and methods ………………………………………..................... Effect assessment: algal growth inhibition test …………….................. Apparatus ………………………………………....................................

21 21 21 21 24 24 25 26 26 26

4.1 4.1.1 4.1.2 4.1.3 4.2 4.3 4.4 4.4.1 4.4.2

Test compounds and toxicity tests Effect concentration ……………………………………….................... Exposure assessment: time trend development Risk assessment ………………………………………............................ Statistics ………………………………………......................................... Effect assessment ………………………………………....................... Exposure assessment ……………………………………….................

viii

Results ………………………………………............................................... Effect assessment ……………………………………….......................... Growth inhibition ………………………………………......................... Effect concentration ……………………………………….................... Exposure assessment ……………………………………….................... Characterizing spatiotemporal trend of PAHs ……............................ Validating spatiotemporal trend of PAHs ……................................... Risk assessment ………………………………………............................

29 29 29 30 33

6.1 6.1.1 6.1.2 6.2 6.3

Discussion ………………………………………......................................... Effect Assessment ………………………………………......................... Effect concentration ……………………………………….................... Exposure assessment ……………………………………….................... Model selection ………………………………………........................... Risk assessment ………………………………………............................

49 49 50 51 51 54

PART VII

Conclusions and recommendations ……………………………………...

56

PART VIII

List of References ……………………………………….............................

57

PART IX

Appendices ………………………………………........................................ Literature review .............………………………………………..................... Materials and methods .............………………………………………........... Results ………………………………………................................................

66 66 68 70

PART V 5.1 5.1.1 5.1.2. 5.2 5.2.1 5.2.2 5.3 PART VI

Appendix A Appendix B Appendix C

34 47 47

ix

x List of tables Table 1

Composition of the Synthetic seawater used in all tests

22

Table 2

Nutrient stock solutions for culture medium

22

Table 3

Volumes of added Internal Standard for the quantification of PAHs 24 concentrations by gas chromatography-mass spectrometry (GC/MS)

Table 4

Nominal concentrations used in the dose-response tests and the average 31 values of the actual concentrations measured by the GC/MS system of the 8 PAHs mixture

Table 5

Percentage of inhibition of the control and acetone control after 72h of exposure

Table 6

Prediction of the EC50 (e parameter; 95% CI) by log-logistic models from 32 drm function in R package drc

Table 7

Akaike Information Criteria (AIC), p-values, adjusted R2, percentage of 36 deviance explained and variance of the residuals for the linear regression model

Table 8

Akaike Information Criteria (AIC), p-values and variance of the residuals for 37 the gamma regression model

Table 9

Akaike Information Criteria (AIC), p-values, adjusted R2, percentage of 38 deviance explained and variance of the residuals for the generalized additive model

Table 10

CEMP data assessment for 8 of the USEPA 16 priority PAHs

Table A.1

Concentrations of PAHs measured in the coastal sediments as ng g-1 dry 66 sediments

Table A.2

Blank corrected concentrations (ng g-1 dry wt) of individual polycyclic 66 aromatic hydrocarbons (PAHs) in surface sediments (0–1 cm) from the Chukchi shelf collected during the Chukchi Sea Offshore Monitoring in Drilling Area-Chemical and Benthos (2009)

32

42

xi Table A.3

Concentrations of PAHs (µg kg-1) in the North Sea surface sediments 66 (fraction < 63 µm)

Table A.4

Concentrations of measured PAH compounds in ng g-1 d.w. mussel sample 66

Table A.5

Comparison of the no observed effect concentrations (NOECs) and the 67 lowest observed effect concentrations (LOECs) of anthracene, naphthalene and phenanthrene to Tetraselmis chuii at 20 and 25 °C

Table B.1

Water solubility and acetone solubility of the PAHs used in the growth 68 inhibition test

Table B.2

List of data points used from the ICES database per each substance and 68 each matrix

Table B.3

Period covered for data acquisition of the 16 PAHs per each matrix

Table C.1

Estimated of Half-life of the PAHs in the sediments and time to degrade 90% 70 of PAHs.

68

xii List of figures Fig. 1

Primary productivity cycle

Fig. 2

Typical seasonal pattern of phytoplankton, zooplankton abundance and 5 nutrients

Fig. 3

Boundary currents associated with coastal upwelling

6

Fig. 4

Processes driving the environmental fate of Persistent Organic Pollutants

8

Fig. 5

Major processes affecting the fate of petrogenic and pyrogenic PAHs upon 9 introduction to aquatic environments

Fig. 6

Intramedia and intermedia transport processes. 1 (air), 5 (soil), 8(sediment) 10 are advective and dispersive Intramedia transport. 2,3,4 (air-soil-water), 6 (suspended solids-sediment), 7(water-sediment) are advective and dispersive intermedia transport

Fig. 7

Bioavailability concentration, concentration at target site, effects at 14 molecular level and observable effects of chemicals

Fig. 8

Risk characterization: a systematic procedure through estimation of exposure and effects

15

Fig. 9

Scheme representing the thesis outline

20

Fig. 10

Coulter counter for measurement of cell density

21

Fig. 11

Example of a set of experimental units

23

Fig. 12

Log daily algal cell density inhibition for acetone test at concentrations of 5, 30 20, 50 and 100 μL compared to the control (0 μL)

Fig. 13

Growth inhibitions from dose-response test

Fig. 14

Concentration-response curves from a dose-response log-logistic model 33 with 4 parameters (b, c, d and e)

Fig. 15

Cleveland plots of the distribution of data concentrations

4

31

34

xiii

Fig. 16

Model validation graphs obtained by applying a linear regression model on 37 the concentration of PAHs

Fig. 17

Model validation graphs obtained by applying a generalized linear model 38 with gamma distribution on the concentration of PAHs

Fig. 18

Validation tool for the generalized additive model that contains a smoother for concentration, year, month, substance and fraction

40

Fig. 19

Estimated smoothing curves after changing the basis dimension for year and month

40

Fig. 20

Temporal trends of the sediment primary producer. The BCF increased as trophic level also increase, being the highest in birds. The BCF for PAHs with lower molecular weight was smaller than those with higher ones(52). A research performed in 2013 by Almeda et al. in the northern Gulf of Mexico showed the bioaccumulation of polycyclic aromatic hydrocarbons (PAHs) in mesozooplankton communities. These BAFs ranged from 3 to 2,570 µg L-1 depending on the type of PAH, the crude oil concentration and the copepod community. Being fluoranthene and pyrene in one of the stations (10 µg L-1), the one with the highest bioaccumulation factors reaching a value of 1,000(53). In other study done in the Bohai Bay (North of China) Wan et al. in 2007 analyzed 18 PAHs in phytoplankton/seston, zooplankton, five invertebrate species, five fish species, and one seabird species and found that PAHs biomagnification factors (BMFs) in the food web ranged from 0.11 for fluoranthene to 0.45 for acenaphthylene(54). PAHs can also be biotransformed or degraded. Elimination of PAHs from organisms is a combination of passive, thermodynamically driven diffusion and enzymatic transformation followed by excretion. In marine polychaetes, biotransformation of PAHs appears to be a two-step process (similar as in vertebrates). In Phase I enzymes primarily Cytochrome P450 (CYP enzymes) catalyze the introduction of a functional group into the PAH, which somewhat increases water solubility. Afterwards in phase II, the water solubility increases further since an enzymes catalyze covalent attachment of large polar groups. For example nereis virens is considered an efficient

14 biotransformer since scientist have found that more than 80% of the pyrene derived compounds extracted from gut tissue were present as phase II metabolites after 5 days of exposure(55). 2.2.3 Ecotoxicological effects Toxicity can be defined as the relative ability of a substance to cause adverse effects in living organisms. This ability is dependent on some conditions such as the quantity or the dose of the substance, which determines whether the effects of the chemical are toxic, nontoxic or beneficial, the route of entry, duration, frequency of exposure, variations between different species (interspecies) and variations among members of the same species (intraspecies)(56). The bioavailability of chemicals that dependent on biogeochemical processes is an important factor often neglected in ecotoxicological evaluation and hazard assessment. The amount of bioavailable fraction and its chemical available form are critical factors for toxicity in terms of uptake and ultimately for the concentration at the target sites in organisms. In Fig. 7, the bioavailability, toxicity and some adverse effects of chemicals is depicted.

Fig. 7. Bioavailability concentration, concentration at target site, effects at molecular level and observable effects of chemicals(57).

PAHs have the ability to bioaccumulate in the tissues of living organisms and can be biotransformed throughout the food chain and may cause damage in the organic metabolism(58). For that reason, identifying the intrinsic characteristics of such compounds (hazard), measuring and predicting their exposure concentrations once they are produced used and emitted (predicted environmental concentration), and estimating the extent of their toxic effect or disease through dose-response analysis (predicted no effect concentration) is very important to characterize a risk. The risk is determined by means of a risk quotient that corresponds to the ratio between the predicted environmental concentration (PEC) and the predicted no effect concentration (PNEC) as shown in Fig. 8.

15

Fig. 8. Risk characterization: a systematic procedure through estimation of exposure and effects(39). Among the PAHs (a group of more than 10,000 compounds with two or more fused benzene rings)(59) the U.S. Environmental Protection Agency (USEPA) listed sixteen as priority pollutants due to their adverse toxic effects on organisms (acenaphthene, acenaphthylene, anthracene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo[a]pyrene, benzo[ghi]perylene, chrysene, dibenzo[a,h]anthracene, fluoranthene, fluorene, indeno[1,2,3‐ cd]pyrene, naphthalene, phenanthrene, pyrene). There are also over twenty alkylated homologs, that are more abundant, persistent, sometimes more toxic than their parent PAHs, and tend to bioaccumulate in a greater extend(40). Alebic-Juretic suggests that only sediments with high concentrations of PAHs (>10 mg kg-1 dry weight) have been found to be mutagenic, the majority of which effect is attributed to benzo(a)pyrene(47). Marine ecosystems are increasingly exposed to complex mixtures of organic chemicals. Despite this, mixtures have been scarcely studied since the main attention have been focused on few individual substances such as the aforementioned sixteen priority PAHs(60). As these mixtures can exert and/or induce joint effects, more realistic studies should be performed to have broader and complete information about the composition of water in terms of the pollution(61). Only few studies (e.g. Echeveste et al. in 2010) on the effect of mixtures of organic pollutants to marine phytoplankton growth have been performed. The main conclusion of this study was that there is no ecotoxicological effect of the exposure of PAHs mixtures since they found toxic effect on

16 phytoplankton abundances, viability and concentrations of Chlorophyll a at concentrations that were 20–40 folds than those found in the open ocean. They also suggest that the influence of complex mixtures of organic hydrophobic pollutants should be studied on oceanic phytoplankton communities(62). PAHs can exert ecotoxicological effects. These effects range from genetic damage to physiological cellular change. These compounds have been studied in the laboratory and in the field where relations between pathologies and PAHs presence in in vitro and in vivo organisms have being found(27). For example, PAHs cause a damage of the DNA in organisms resulting in mutagenic, teratogenic and carcinogenic properties. This direct damage could delay or even interrupt its replication within the cell division cycle specifically during the S-phase(63,64,65). Another physiological function that can be affected by the action of the PAHs is the one done by cellular membranes. They attack plasma membranes and change their permeability, this disruption may alter lipid arrangement in membranes causing the formation of pores and increasing the cells permeability(66,67). Ecotoxicological effects of PAHs in food chain organisms have been often reported. For example Cerezo and Agustí in 2015 found that PAHs in a Prochlorococcus populations (a marine cyanobacteria) affected cell division, altering the percentage of cells that enter the S and G2M phases, resulting as well in the increase of the time needed to complete the cell division(68). In 2007 Chung et al. investigated the responses of the benthic microalga Chlorococcum meneghini to PAH-spiked sand and found that this exposure diminished cell densities and chlorophyll a concentrations(66). Vieira and Guilhermino (2012) found that phenanthrene, anthracene and naphthalene inhibited the growth rate of Tetraselmis chuii (a marine algae) in concentrations ranging from between 0.070 and 0.360 mg L-1 and that phenanthrene induced significant effects at lower concentrations than anthracene and naphthalene(69). More detailed information can be found in Table A.5. At higher trophic levels for instance, in mullet (Mugil cephalus) and sea bass marine fish species (Dicentrarchus labrax) from the Red Sea, Abdel et al. (2014) have found alterations in the hepatic mRNA levels of CYP1A, MT-A and GST alpha genes caused by the 16 USEPA PAHs(70). Other study in 2000 by Garçon et al. suggests that in in vitro human epithelial lung cells exposed to benzo(a)pyrene suffered from induced lipid peroxidation, which may cause a firmly impairing membrane integrity(71). 2.2.4 Time trends and legislation Legislative efforts aim to decrease the use and emission of PAHs in order to reduce their adverse ecological effects. To assess the effectivity of the imposed legislation it is important to evaluate whether concentration of PAHs change over time. To do so, time trend development is crucial as it would give important information about the distribution and behavior of those substances into the marine environment and provide elements for the development of solid

17 management plans that includes biological, social and legal strategies, which may help in the cleaning, restoration and/or banning of toxic substances. Some researchers have identified that peaks distribution of the PAHs are related with their emission time from particular sources, like periods of the wars in the twentieth century or some other events socio-economic advancement(72). For example, the results of Sakary et al. (2012) suggest that PAHs input started soon after World War II and increased exponentially from 1980 onwards by 310 ng g -1 d.w(73). Others examples show that in a Feitsui Reservoir in Taiwan in the period between 1986 and 2002, PAHs concentrations increased due to increased traffic(73). In England between 2008 and 2013 in some remote sampling locations, PAHs concentrations varied depending on the location (northeast, east, south and southwest). The concentrations in 2008 varied in concentration across different geographical regions as follows: 1,020-4,612 µg kg-1 in the Northeast, 101-5,230 µg kg-1 in the East and 35,4-1,554 g kg-1 in the Southwest with a subsequent decrease in all the regions by year 2011(74). To protect the marine environment from pollution effects due to PAHs, different legal frameworks have been developed at national, regional and European levels. The most important of these include the Convention for the Protection of the Marine Environment of the North-East Atlantic (OSPAR), the Helsinki Commission (HELCOM), the Barcelona Convention for the Protection of the Mediterranean Sea along with the Mediterranean Action Plan (MAP), and the Mediterranean Marine Pollution Monitoring and Research Program (MED POL). There are also a series of protocols including the Land Based Sources (LBS) protocol, as well as European Directives such as the Water Framework Directive (2000/60/EC) and the Marine Strategy Framework Directive (2008/56/EC)(38). According to the Marine Strategy Framework Directive (MSFD) of the European Union (2008 and 2010), monitoring programs of marine litter should monitor the success of implemented measures to combat the pollution of European coastal areas. Examination of the existing monitoring data from the marine compartments, such as the one from the Oslo and Paris convention for the protection of the marine environment of the North-East Atlantic (OSPAR) is a prerequisite to define the good environmental status (GES) of marine waters and to identify indicators for the achievement of GES(75). The Division of Environmental Law and Conventions (DELC) of the United Nations Environment Programme (UNEP) have Regional Seas Programmes and Conventions that are working for the protection of marine and coastal environments in 18 regions of the world to ensure the progressive development of environmental law across different environmental sectors and levels of governance(76). In the Baltic Sea area, In the Baltic Sea area, the Helsinki Commission (HELCOM) is working to ensure that the signed (1974) and revised (1992) Helsinki convection is well implemented throughout the coastal states, taking precautions and implementing necessary correctness to prevent and eliminate pollution from land-based and marine sources and promote the ecological restoration of the Baltic Sea area for the preservation of its ecological balance(77).

18 PART III Problem formulation – Rationale for this thesis PAHs originate from natural and anthropogenic sources and from point and/or diffuse sources. During many years, their concentrations in the environment were increasing due to the human economic activities. From the 1960s onwards more attention has been paid on the assessment of the ecotoxicological potential of PAHs. Time trends indicate that concentrations of PAHs have increased from 1940s during war events with high peaks between the early 1980s and 1900s due to industrialization and urbanization(72) and a decrease in recent times from 2009 when a series of laws were set to banned the production and use of PAHs-related products(76). However, to date the ecological risk of PAHs to marine phytoplankton has been poorly quantified. The present study is a first step to quantify the risk of eight persistent, bioaccumulative and toxic PAHs (listed amongst the 16 priority PAHs by the USEPA). To do so, existing data of marine sediments and tissues of mussels are used to infer spatiotemporal trends of PAH concentrations. Next, an effect assessment is performed in which the toxicity of a mixture of eight PAHs on the growth of a marine diatom is tested. Finally, the results of the exposure assessment and the effect assessment are integrated and the potential risk of the PAHs quantified. Specific aims of the research Many reports and scientific papers have reported the presence of PAHs in significant concentrations and found their possible influence in the appearance of some adverse effects, in this context this research aims to: 1) test the dose-response effect of a mixture of 8 PAHs in the growth of Phaeodactylum tricornutum. For this purpose, the marine diatom was exposed to different concentrations (ranging between 211 to 7,178 ng L-1) of a mixture of PAHs and the algal growth rate was monitored and compared with a blank with no PAHs concentration. By means of mathematical algorithms doseresponse graphs were obtained, being able to predict the concentration at which PAHs cause an effect in 50% of the algae (EC50). Following the study of Echeveste et al. (2010), reporting no ecotoxicological effect on the exposure of mixtures of different organic pollutants(62), we hypothesized that the concentrations of the PAHs mixture used in this study do not interfere or influence the growth of the used marine diatom. 2) investigate and model the concentrations of the 16 USEPA priority PAHs in different environmental compartments in order to infer temporal trends. This was done by means of gathering information from the International Council for the Exploration of the Sea (ICES) in different matrices such as sediments and the whole body of a mussel (Mytilus edulis) in the 5 OSPAR regions. Based on the European Chemicals Agency Legislation (2009), the use of PAHscontaining oils such as extender oils (for making car tyres) and coal tar pitch (for corrosion protection coatings and wood preservatives) is banned(78). Following this statement, and based on the fact that PAHs emissions have been decreasing during past years, we hypnotized that the

19 concentrations in the 5 OSPAR regions (Artic waters, Bay of Biscay and Iberian Coast, Celtic Seas, Greater North Sea and Wider Atlantic) are decreasing. 3) assess the risk of PAHs on a marine phytoplankton (Phaeodactylum tricornutum) and mussel (Mytilus edulis) species. To do so, results from the algal growth inhibition test (effect assessment) and time trends data for exposure assessment of 8 PAHs were used to calculate the risk quotient. We hypothesized that the quality thresholds are not exceeded based on the decreasing trend in the concentrations of organic contaminants in the marine environment during the past two decades(79). The novelty and relevance of the present research lies in the integration of existing data (assessed using regression-based models) and lab-based experiments (algal growth inhibition) for the determination and identification of the risk of PAHs in the marine environment. As scheme of the thesis outline is depicted in Fig. 9.

20

THESIS OUTLINE

Exposure concentrations from ICES in sediment and Mytilus edulis (§4.2)

Effect concentrations from laboratory tests with Phaeodactylum tricornutum (§4.1)

-1

Time trends development concentrations (§5.2)

of

exposure

Dose-response test (211 to 7,178 ng L ) to -1 derive effect concentration (EC50= 3,300 ng L ) (§5.1)

Risk Quotient of 8 out of the 16 USEPA priority PAHs from 1985 to 2013 (§5.3)

Risk Characterization (§5.3) Fig. 9. Scheme representing the thesis outline

21 PART IV Materials and methods 4.1 Effect assessment: algal growth inhibition test The marine diatom Phaeodactylum tricornutum Bohlin strain 1052/1A was obtained from the Culture Collection of Algae and Protozoa (Oban, United Kingdom). In a full factorial design, the diatom was exposed to a nominal concentration gradient of 0 to 200,000 ng L-1, including concentrations of 10; 100; 1000; 10,000; 25,000; 75,000; 100,000; and 200,000 ng L-1 of a mixture of eight PAHs: anthracene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(a)pyrene, chrysene, fluorene, phenanthrene and pyrene). Concentrations that were subsequent checked by gas chromatography-mass spectrometry. The test vessels were incubated under continuous and uniform white light with a photon fluence rate of 60 µmol m-2 to 120 µmol m-2 and gentle shaken twice a day at 20ºC. Test flasks were covered to avoid airborne contamination and to reduce water evaporation but letting CO2 enter to the vessels. All chemicals used during the tests were of analytical grade. 4.1.1 Apparatus The equipment used in the experimental part for measuring algae cell density were a Z™ Series COULTER COUNTER Cell and Particle Counter with temperature-controlled cabinet that provides continuous illumination (Fig. 10), culture flasks, apparatus for membrane filtration and autoclave.

Fig. 10. Coulter counter for cell density measurement

4.1.2 Test compounds and toxicity tests Precultures were prepared with synthetic sea water (contains NaCl, MgCl2.6H2O, NaSO4, CaCL2, KCl, NaHCO3, H3BO3 in different concentrations; Table 1), and three stock solutions that containing nutrients, vitamins and minerals (Table 2) to enriched the previous solution and make

22 it suitable for algal growth as stated in the ISO 10253:2006 protocol(79). Fifty mL of enriched medium per test flask were used to inoculate from the precultures a spike of the cells to obtain a cell density of about 10,000 cells per mL. Every day the cell density was measured by means of a coulter counter at the same time (+/- 2 hours). A total of 32 test flasks were used (including the acetone control), each concentration had 3 replicates and a color control. The color control consisted only of medium, acetone and the mixture of PAHs and was used just to qualitatively monitor the colors of the flask for possible contamination. Table 1. Composition of the Synthetic seawater used in all tests Salt

Concentration of salt in synthetic sea water (g L-1)

NaCl

22

.

MgCl2 6H2O

9.7

Na2SO4

3.7

CaCl2

1.0

KCl

0.65

NaHCO3

0.20

H3BO3

0.023

Table 2. Nutrient stock solutions for culture medium Nutrient

Concentration in stock solution

Final concentration in test Solution

Stock solution 1 FeCl3.6H2O

48 mg L-1

149 µg L-1

MnCl2.4H2O

144 mg L-1

605 µg L-1

ZnSO4.7H2O

45 mg L-1

150 µg L-1

CuSO4.5H2O

0.157 mg L-1

0.6 µg L-1

CoCl2.6H2O

0.404 mg L-1

1.5 µg L-1

H3BO3

1 140 mg L-1

3.0 mg L-1

Na2EDTA

1 000 mg L-1

15-0 mg L-1

Stock solution 2 Thiamin hydrochloride

50 mg L-1

25 µg L-1

Biotin

0.01 mg L-1

0.005 µg L-1

Vitamin B12 (cyanocobalamin)

0.10 mg L-1

0.05 µg L-1

Stock solution 3 K3PO4

3.0 g L-1

3.0 mg L-1; 0.438 mg L-1 P

NaNO3

50.0 g L-1

50.0 mg L-1; 8.24 mg L-1 N

NaSiO3.5H2O

14.9 g L-1

14.9 mg L-1; 1.97 mg L-1 Si

23 A preliminary test with acetone (used to dissolve the PAHs) was done to identify a possible interference with the inhibition of the substances used; the acetone volumes were 5, 20, 50 and 100 μL. Information about solubility properties is shown in Table B.1. Additionally, an algal growth tests in which the diatom Phaeodactylum tricornutum was exposed to different concentrations of a mixture of 8 PAHs was performed. This test included a different range of nominal concentrations, i.e. 1000, 10,000, 25,000, 75,000, 100,000 and 200,000 ng L-1 (being after carefully checked by a gas chromatography) and was used to calculate the EC50 (using data from the first 72 hours). Next figure shows an example of the treatment of the samples throughout the different concentrations (Fig. 11).

Fig. 11. Example of a set of experimental units.

Once the preliminary acetone test was done, the actual concentrations of PAHs in the test flasks based on gas chromatography (trace GC 2000 series) coupled to mass spectrometry (Finnigan Trace DSQ) were calculated. To do so, all samples, including the colour controls were subjected to a dichloromethane extraction. Samples were weighted before and after the extraction to determine the volume gravimetrically and an internal standard (IE) with a known concentration was added for compensating errors in the samples (the added volumes are shown in Table 3). The first extraction (30 mL of dichloromethane) gave 2 phases; an organic and a water phase. From the water phase an extra extraction with approximately the same amount of dichloromethane was done to obtain a second organic phase, which was then combined with the first one to further percolation using over 3g NaSO4 to remove the remaining water. At this point and by using a Büchi RE111 rotary evaporator samples were concentrated to ≤ 5 mL whereupon, a further concentration to 0.5 mL for the controls and the 1000 ng L-1 concentration, and 1 mL to the higher concentrations. A recovery standard was also added at the end for checking the accuracy of the instrument.

24 Table 3. Volumes of added Internal Standard for the quantification of PAHs concentrations by gas chromatography-mass spectrometry (GC/MS). Sample concentration

Added volume

IE concentration

Up to 75,000 ng L-1

10 μL

100 μg L-1

100,000 ng L-1

20 μL

100 μg L-1

200,000 ng L-1

30 μL

100 μg L-1

After the analysis of the samples in the GC/MS system, the real concentrations of the mixture of PAHs were calculated. With the new values obtained, the growth rate was calculated as follow(76):

(eq. 2)

where NL specifies the measure cell density at time TL (time of the last measuring within the exponential growth) and N0 is the nominal initial cell density at T0 (time of test start). Once the growth rate is calculated, the percentage of inhibition for test flasks can be computed as follow: (eq. 3) where ūi represents the growth rate or test flask and ūc is the mean growth rate for the control. 4.1.3 Effect concentration The concentration vs the percentage of inhibition was plot in a graph and the EC50 was derived by means of a log-logistic model of the package drc in R. Each test was followed during 12-16 days but the calculations were made based on data from the first 72 hours. The package drc includes a drm function for fitting concentration/dose/time-response models to data. By this function, 4 different non-linear models were fitted using two, three, four and five-parameter and a binomial argument type(81). 4.2 Exposure assessment: time trend development Data of national monitoring campaigns aiming to quantify the concentration of PAHs in the marine environment are stored in data repositories. One example of such repository is ICES, an intergovernmental organization whose main objective is to increase the scientific knowledge of the marine environment and its living resources, and to use the knowledge to provide unbiased and non-political advice to competent authorities.

25 Data list from the sediments micron fraction (