Ecological implications of accumulation of PTEs and PAHs deriving from fuel exhausts in coastal marine primary producers

Anthropogenic activities, mainly in the form of local fuel exhausts and inputs from the coastline, heavily affect ecosystems at the interface between terrestrial and marine realms, impairing their functionality and the services they provide. Due to the central role of primary producers in trophic webs, their sessile nature and ethical concerns implied in experiments on animals, pollutant analyses in both sediments and macrophytes assume special relevance in assessing pollutant transfers from the abiotic to biotic compartments and their possible transfer through trophic webs. With a view to clarify the accumulation of inorganic and organic pollutants deriving from fuel exhausts on primary producers, the concentrations of Cu, Fe, Zn, phenanthrene and benzo[a]pyrene were determined in sediments and macrophytes collected from sites along the Cilento coast, in western Mediterranean Sea, characterized by different levels of anthropogenic pressures. The 18 species analysed, belonging to Cyanobacteria, Chlorophyta, Rhodophyta, Heterokontophyta and Embryophyta, exhibited different accumulation capabilities toward pollutants, with average concentrations of Cu, phenanthrene and benzo[a]pyrene in all the divisions (17.6 ± 2.3 μg g−1 d.w., 34.3 ± 2.1 ng g−1 d.w., 61.5 ± 9.4 ng g−1 d.w., respectively) higher than those measured in sediments (4.0 ± 0.7 μg g−1 d.w., 11.6 ± 0.9 ng g−1 d.w., 14.8 ± 1.0 ng g−1 d.w., respectively) and more than one order of magnitude higher in Embryophyta for Cu (62.9 ± 7.1 μg g−1 d.w.) and in Cyanobacteria for benzo[a]pyrene (181 ± 2 ng g−1 d.w.). The obtained findings constitute a reference for the accumulation capabilities of different taxa and for the behaviour of different fuel exhaust pollutants in marine coastal environments, with implication for their transfer across trophic webs.


Introduction
Potentially toxic elements (PTEs) and polycyclic aromatic hydrocarbons (PAHs) are persistent and ubiquitous pollutants in marine ecosystems (Ranjbar Jafarabadi et al 2017a,b), where their sources are primarily represented by local fuel exhausts and inputs from the coastline (Ytreberg et al 2022).Due to their affinity for organic matter and reactive minerals, PTEs and PAHs tend to preferentially distribute in sediments, with longer residence times than in water (Han et al 2022).From sediments and interstitial water, PTEs and PAHs can be transferred and accumulated in the biota, either by direct absorption or indirectly through mobilization and absorption from the water column.Both physico-chemical and biologically-mediated processes, for instance bioturbation, contribute to the mobilization of contaminants from sediments (Jesus et al 2022), making them available for organisms incapable of direct absorption from sediments, such as benthic algae, filter-feeders and plankton (Yang et al 2020).The relative balance between absorption and excretion/removal processes determines pollutant bioaccumulation and, consequently, their transfer through trophic webs and their potential biomagnification (Sun et al 2020).These outcomes have been consistently viewed as a major environmental and human health concern, due to the toxicity of PTEs and PAHs and their effects on biodiversity and ecosystem functioning (Kumar et al 2021).
Due to the central role in trophic webs of primary producers, their sessile nature and ethical concerns implied in experiments on animals, the analysis of PTEs and PAHs in seagrasses and benthic algae assumes special relevance in assessing environmental contamination and distribution of pollutants in the biota.This is especially true for habitat-forming species, such as Posidonia oceanica and Cystoseyra spp. in the Mediterranean Sea (Orfanidis et al 2021) or Sargassum spp.and Macrocystis pyrifera in the Atlantic and Pacific oceans (Fernández et al 2020), whose production primarily supports large and diverse trophic webs.Investigating the variability in pollutant accumulation among several taxa, however, is similarly important in evaluating the effects of species morphology, physiology and evolutionary position on the differential distribution of PTEs and PAHs in the biota, as well as on the variations in pollutant flows among different trophic web nodes.
To evaluate the potential input of fuel-exhaust PTEs and PAHs to marine trophic webs mediated by primary producers, we investigated pollutant distribution in sediments and in several macrophyte species of the Mediterranean Sea coastal waters, which are mostly subjected to severe pressures from anthropogenic activities (Rizzi et al 2021, Rivoira et al 2022).Specifically, the concentrations of pollutants deriving from marine traffic, i.e.Cu, Fe and Zn according to Coufalík et al (2019), phenanthrene according to Borillo et al (2018), and benzo [a]pyrene according to Tancell et al (1995), were investigated in sediments and in Cyanobacteria, Chlorophyta, Rhodophyta, Heterokontophyta and Embryophyta species collected from sites, along the western Italian coast, differing in the level of anthropogenic pressure.

Study area and sample collections
The study was carried out along a 20 Km stretch of the Cilento coast, in western Mediterranean Sea (figure 1).The area is dominated by Quaternary submarine deposits insisting upon Middle-Miocene syn-orogenic units (D'Angelo et al 2020) and the substrate is mainly characterized by limestone and siliciclastic turbidite sequences, named 'Cilento Flysch' (Guida and Valente 2019).
Macrophyte samples were collected from six sites (figure 1) characterized by rocky substrates and differing in the level of alleged anthropogenic pressure.Four sites were located within a marine protected area (Santa Maria di Castellabate MPA), managed by the 'Cilento, Vallo di Diano e Alburni' Italian National Park, at different levels of restrictions (from full to partial reserve).Two sites were located at the mouth of harbours characterized by moderate (300 moorings) and high (1000 moorings) ship traffics, respectively.
In February 2023, at each site, macrophyte community was characterised by visual census and identification of the difficult taxa was carried out through morphological analysis using a M165C (Leica, Germany) stereomicroscope with 1x objective and a Dialux 20 (Leitz, Germany) optical microscope with 25x-100x objectives.According to their availability and abundance, Cyanobacteria and fully developed thalli of macroalgae (Chlorophyta, Rhodophyta and Heterokontophyta), as well as 3-6 months old leaves of Embryophyta, were handpicked in the eulittoral and upper infralittoral zone, at a depth of up to 2 m.In the same occasion, sediment samples (0-5 cm depth) were manually collected at each of the six sites.After collection, samples were stored in glass jars and transferred to the laboratory in cooled boxes, for further processing.

Laboratory analyses
Laboratory analyses on sediments and on each macrophyte species collected from the six sites were carried out in triplicates.
Sediments were sieved at 2 mm to retrieve the granulometric fraction and were characterized for the organic matter content by calcination (550 °C for 4 h) in a muffle furnace (Controller B 170, Nabertherm GmbH; Lilienthal, Germany).Macrophytes were gently washed with a 35‰ NaCl solution and manually cleaned of exogenous material through plastic hand tools, avoiding contamination.Afterward, all the samples were ovendried (at 75 and 40 °C for PTE and PAH analyses, respectively) to constant weights and pulverized by hand in china mortars or by a planetary ball mill in agate mortars (PM4, Retsch; Haan, Germany).
Total copper (Cu), iron (Fe) and zinc (Zn) concentrations were determined in both sediments and macrophytes according to the method of Baldantoni et al (2009).For this purpose, an acid mixture was employed to digest each subsample in a microwave oven (Ethos, Milestone; Shelton, CT, USA).The mineralization program consisted of six steps (Baldantoni et al 2019): 250 W for 2′, 0 W for 2′, 250 W for 5′, 400 W for 5′, 0 W for 2′, 500 W for 5′.In particular, 1 mL 49% HF (Merck; Darmstadt, Germany) and 2 mL 65% HNO 3 (Carlo Erba; Cornaredo, MI, Italy) were added to 0.125 g of each pulverized sample.After digestion, the solutions were diluted to a final volume of 25 mL, using milli-Q water (Elix 10, Millipore; Darmstadt, Germany).PTE concentrations were quantified by inductively coupled plasma optical emission spectrometry (Optima 7000DV, PerkinElmer; Wellesley, MA, USA), using a cross-flow nebulizer with a Ryton Schott spray chamber and an alumina injector.In order to verify the method accuracy, standard reference materials were also analysed: 1646a estuarine sediment (NIST 2004) for sediments and 1547 peach leaves (NIST 2019) for macrophytes.The recovery percentage of each PTE in the standard reference materials (ranging from 96 to 98%) was used to correct the quantification of the investigated PTEs.The method precision, calculated as relative standard deviation, based on n = 9 sequential measurements of the same sample for each PTE, ranged from 4 to 6%.The limits of quantification for Cu, Fe and Zn were equal to 2.0, 0.1 and 0.4 μg g −1 d.w., respectively.
For phenanthrene (Phe) and benzo[a]pyrene (BaP) extraction from sediments, samples (2 g) were treated with 10 mL of a 90:10 mix of hexane:acetone (Merck; Darmstadt, Germany) in an ultrasonic bath (Argo Lab DU −100; Carpi, MO, Italy) for 10 min, repeating the extraction for 3 times with fresh solvents and then pooling the extracts (De Nicola et al 2015 modified).The extracts were then transferred into glass tubes with 5 g deactivated alumina, 2 g primary secondary amine bonded silica (PSA, Supelco; Bellefonte, PA, USA), 500 mg anhydrous sodium sulphate and 100 mg Cu (Merck; Darmstadt, Germany).The tubes were vortexed for 5 min at 350 rpm and centrifuged for 5 min at 1578 g, according to De Nicola et al (2019).The supernatants were evaporated to ∼ 0.3 mL under a N 2 stream (MultiVap 10, LabTech; Torre Boldone, BG, Italy) and filtered (PTFE filters 0.45 μm pore size, Restek; Cernusco sul Naviglio, MI, Italy).For macrophytes, PAHs were extracted by matrix solid-phase dispersion (MSPD) according to Concha-Graña et al (2015).Specifically, samples (250 mg) were blended with 500 mg of C18 (Supelclean-Envi 18, Supelco; Bellefonte, PA, USA) and the dispersion deposited in the top of a Envi-Florisil (1 g) SPE glass tube (Supelco; Bellefonte, PA, USA) with 500 mg of anhydrous sodium sulphate (Merck; Darmstadt, Germany).PAHs were eluted with 10 mL of hexane followed by 10 mL of a 20:80 mix of dichloromethane:hexane (Merck; Darmstadt, Germany) using a Visiprep vacuum distribution manifold (Supelco; Bellefonte, PA, USA).The eluates were concentrated to ∼ 0.3 mL under a N 2 stream (MultiVap 10, LabTech; Sorisole, BG, Italy).Both sediment and macrophyte extracts were transferred in 1 mL graduate flasks and diluted to 1 mL with hexane, transferred to amber vials and analysed for Phe and BaP quantification by programmed temperature vaporization-gas chromatography-tandem mass spectrometry (PTV-GC-MS/MS, TRACE 1310 GC, ITQ 900 Thermo Fisher Scientific; Waltham, MA, USA).GC was equipped with a TR-5MS GC column (60 m × 0.25 mm, 0.25 μm film thickness; Thermo Fisher Scientific; Waltham, MA, USA), and the oven temperature program consisted of a ramp from 50 °C (5 min) to 325 °C (held for 15 min), increasing temperature by 8 °C min −1 .The system was operated in electron impact mode (70 eV), with tandem mass spectrometry as the detection mode and selected ion monitoring (SIM) as the acquisition mode.Transfer line temperature was set at 300 °C and ion source temperature at 270 °C.Helium was used as the carrier gas at a constant flow rate of 1 mL min −1 .Phe and BaP were quantified after performing a calibration curve for each PAH.Instrumental and procedural blanks were systematically evaluated (one blank for each batch).Two standard solutions containing the two studied PAHs (Dr Ehrenstorfer GmbH; Augsburg, Germany) were injected for each batch for chromatographic control.The quality control of the complete procedure was performed using D-labeled PAHs as surrogate standards (phenanthrene-d10 and benzo[a]pyrene-d12; Dr Ehrenstorfer GmbH; Augsburg, Germany) added before extraction for quantification.Surrogate recoveries fell in the range 70%-120%.The limits of quantification for Phe and BaP were equal to 1.7 and 0.5 ng g −1 d.w., respectively.

Data analysis
The accumulation of PTEs and PAHs in the different macrophyte divisions and in sediments was investigated through descriptive statistics and linear mixed models.Specifically, the distribution of concentration values within each group was analysed through violin plots, trimmed to the range of values, using a bandwith of 0.75.The differences in PTE and PAH accumulation among groups were analysed through linear mixed models using the group (macrophyte divisions and sediments) as a fixed factor and the species and site, the former nested within the latter, as random factors.The choice of the site as a random factor reflects the alleged lack of independence among sites due to water movements and the proximity among sites, as well as the research aims focusing on the differentiations among groups rather than among sites.The significance of the fixed factor was evaluated through the Kenward-Roger statistic, and the pairwise comparisons were then obtained based on the estimated marginal means.All the analyses were performed within the R 4.3.1 programming environment (R Core Team 2023) using functions from the packages 'ggplot2' (Wickham 2016), 'lme4' (Bates et al 2015) and 'emmeans' (Lenth 2023).
Moreover, the biota-sediment accumulation factor (BSAF) was calculated as the ratio of the concentration of each PTE or PAH in each macrophyte division to the respective concentration in sediments, according to Arnot and Gobas (2006).

Results and discussion
Cyanobacteria, Chlorophyta, Rhodophyta, Heterokontophyta and Embryophyta, with a total number of 57 species (table 1), composed the macrophyte community censed in the eulittoral and upper infralittoral zone of the studied Mediterranean area.Among them, according to their availability and abundance, 18 species (1 Cyanobacteria, 4 Chlorophyta, 6 Rhodophyta, 6 Heterokontophyta, 1 Embryophyta) were collected from 6 sites for Cu, Fe, Zn, Phe and BaP analyses (table 1), and pollutant concentrations in Cyanobacteria, algal thalli and seagrass leaves were compared with those measured in sediments.
Pollutant concentrations in the analyzed taxa varied both within and among divisions, but were mostly comparable or higher in respect to the concentrations measured in sediments (figures 2 and 3).The specific uptake mechanisms of the different taxa and their phylogenetic position appeared to affect the accumulation of pollutants, with Embryophyta (Posidonia oceanica) showing the highest (P < 0.05) concentrations of Cu (34-128 μg g −1 d.w.) and Zn (43-105 μg g −1 d.w.).The accumulation of Fe and PAHs showed little differences (for α = 0.05) among divisions, with the only exception of Phe reaching the highest concentrations (14-106 ng g −1 d.w.) in Heterokontophyta, comparable however to those of Cyanobacteria, Chlorophyta and Embryophyta.Fe was the only pollutant showing significantly higher concentrations in sediments than in macrophytes, an occurrence explainable in terms of its exceedingly low bioavailability in sediments of the study area (Nitopi et al 2024), where it is mainly found as a mineral constituent, a common situation determining reduced transfers of Fe from sediments to the marine biota (Ito et al 2021).In terms of the specific taxa, the higher accumulation of PTEs was on average measured in Spyridia filamentosa and Laurencia microcladia among the Rhodophyta, in Dictyota spiralis among the Heterokontophyta and in Cladophora coelothrix among the Chlorophyta (figure 2).The specific PTE accumulation degree of different algae can depend on different affinities for each PTE (Volterra and Conti 2000).The capability to Although there are reports of algae accumulating PTEs in higher concentrations than Embryophyta (Lewis and Devereux 2009), in general seagrasses are able to absorb pollutants through both leaves, from water, and roots, from sediments, the sedimentation dynamics of which are also affected by the growth of several seagrasses (Gacia et al 1999, Gacia andDuarte 2001).These mechanisms make Embryophyta usually capable of accumulating PTEs at concentrations several folds higher than those in the surrounding environment (Bonanno and Di Martino 2016), as observed in the case of Cu and Zn, whose concentrations in sediments fall in the ranges generally reported for the Mediterranean Sea (Bonanno and Orlando-Bonaca 2018).
Also for PAHs, a great variability in accumulation capability was observed among divisions and among different species of each division (figure 3).For Phe, the highest concentrations were measured in Laurencia microcladia among Rhodophyta, in Taonia atomaria among Heterokontophyta and in Ulva clathrata among Chlorophyta.Laurencia microcladia showed also high concentrations of BaP together with Rhodophyta such as Laurencia obtusa and Haliptilon virgatum, whereas Cystoseyra compressa and Chaetomorpha aerea accumulated the highest amounts of this compound among Heterokontophyta and Chlorophyta, respectively.
Similarly to Cu and Zn, also Phe and BaP were preferentially accumulated in the biota, with concentrations in primary producers up to one order of magnitude higher than those observed in sediments (figures 2 and 3),  1).Divisions are indicated in different colours for ease of comparison.Different letters indicate significant differences (for α = 0.05) among the groups according to the linear mixed models.
which showed limited variations (∼3x between the lowest and highest values).Such a finding is remarkable, considering that PAHs usually tend to be sequestered within sediments according to their hydrophobic nature (Mille et al 2007) and may be explained by the low organic matter content (0.88%-1.59% d.w.) in the sediments of the study sites.Such a hypothesis is supported by the concentrations of BaP (8-21 ng g −1 d.w.) always below the threshold (30 ng g −1 d.w.) set by the Italian legislation (Legislative Decree 152/2006 and further modifications) for marine sediments, even in the case of sediments collected from sites neighboring harbours.
The biota-sediment accumulation factors (BSAFs) calculated for both PTEs and PAHs (table 2) showed exceedingly low values for Fe (0.03-0.18) and, generally, for Zn (0.19-2.47), highlighting the limited transfer of these inorganic pollutants from the abiotic to biotic compartments (Arnot andGobas 2006, Nedjimi 2022).The unique BSAF value greater than 1 calculated for Zn (2.47) was observed in Embryophyta, which also showed a tenfold higher Cu-BSAF value (15.80) than the other divisions (1.64-3.59).The pattern of BSAF values for PAHs among divisions differed from the one observed for PTEs: the highest BSAFs were observed in Heterokontophyta for Phe (4.37) and in Cyanobacteria for BaP (12.30), confirming different mechanisms of uptake and accumulation in macrophytes for the two contaminant classes.The different BSAFs observed for Phe (3.17) and BaP (1.59) in Embryophyta may be related to the different number of aromatic rings of the two PAHs (3 for Phe and 5 for BaP) affecting their translocation from roots to shoots (the higher the number of rings, the lower the translocation; Minkina et al 2022).Overall, according to the obtained results, both inorganic and organic pollutants deriving from fuel exhausts accumulate in marine primary producers to a different extent, often at high degrees if compared to PTE-BSAFs reported for Chlorophyta (Ulva lactuca) from the Lebanese coast, in the eastern Mediterranean Sea (Ghosn et al 2020).
In general, high BSAF values for specific pollutants may also guide the choice of taxa useful for environmental remediation applications, ensuring the immobilization of pollutants into macrophyte biomass (González Fernández et al 2023).In this context, however, the possible transfer to the higher trophic levels should be minimized, which raises the need of suitable engineering solutions for macrophyte exposure, removal and final disposal.Indeed, the effects of pollutant accumulation can cascade up the trophic web (Mahmoudi et al 2005, Louati et al 2013), affecting consumers, including pelagic species (Roberts 2012), thus altering community structure biodiversity and ecosystem functioning and services (De La Fuente et al 2019, Bakiu et al 2023).In this context, the consumption of macrophytes by both indigenous and non-indigenous herbivores (Marić et al 2016), obviously plays a key role in mediating the transfer of fuel exhaust pollutants to higher trophic levels.Moreover, herbivory from sea urchins such as Paracentrotus lividus and Arbacia lixula as well as from fishes such as Sarpa salpa is one of the main factors shaping macrophyte communities in the Mediterranean Sea (Ruitton et al 2000, Tomas et al 2011), together with several abiotic factors, such as waves, sea currents and nutrient availability (Zabala and Ballesteros 1989).Among the different algae divisions, on an annual basis, the herbivore fish Sarpa salpa feeds for 60.8% on Heterokontophyta, for 25.9% on Chlorophyta and for 13.3% on Rhodophyta (Tomas et al 2011).Among Heterokontophyta, Cystoseyra spp., one of the main habitat-forming species of the Mediterranean Sea, are highly predated by Sarpa salpa, whose gut contents can contain up to the 60% of this algal genus, percentages comparable to those observed in other herbivore fishes and sea urchins (Vergés et al 2009).Sarpa salpa and Paracentrotus lividus are the main predators also of Posidonia oceanica, consuming up to the 50% of its annual primary production (Prado et al 2007, Planes et al 2011, Buñuel et al 2023).From a mechanistic point of view, it is thus possible to hypothesize that the transfer of PTEs and PAHs from macrophytes to higher trophic levels may exert adverse effects on trophic webs but, to the best of our knowledge, few attempts have been made to actually measure such effects.

Conclusions
The present research constitutes a reference for the capabilities of different Mediterranean marine primary producers to accumulate fuel-exhaust pollutants and mediate their transfer to higher trophic levels.Apart from Fe, whose low bioavailability limits its accumulation and transfer through the trophic webs, Cu, phenanthrene, Table 2. Biota-sediment accumulation factors (BSAFs) calculated as the ratio of the concentration of each PTE or PAH in each macrophyte division to the respective concentration in sediments (Arnot and Gobas 2006) benzo [a]pyrene and, to a lower extent, Zn raise concerns for their bioaccumulation in several macrophytes, even in a marginally impacted area, as demonstrated by sediment PTE and PAH concentrations.Seagrass interaction with sediments determines higher accumulation capabilities toward specific inorganic pollutants than algae and cyanobacteria, whose comparable pollutant bioaccumulations can be attributed to similar uptake mechanisms.

Figure 1 .
Figure 1.Map of the study area in Southern Italy, with indication of the sampling sites on the western Mediterranean Sea and their classification according to their location within the Santa Maria di Castellabate MPA or the type of the neighbouring harbour.

Figure 2 .
Figure 2. Violin plots of the concentrations of Cu (upper panel), Fe (middle panel) and Zn (lower panel) in the different macrophyte divisions as well as in sediments, with indication of each observation (circles) and the mean for each species (blue Roman numbers, coded according to table1).Divisions are indicated in different colours for ease of comparison.Different letters indicate significant differences (for α = 0.05) among the groups according to the linear mixed models.

Figure 3 .
Figure 3. Violin plots of the concentrations of phenanthrene (upper panel) and benzo[a]pyrene (lower panel) in the different macrophyte divisions as well as in sediments, with indication of each observation (circles) and the mean for each species (blue Roman numbers, coded according to table1).Divisions are indicated in different colours for ease of comparison.Different letters indicate significant differences (for α = 0.05) among the groups according to the linear mixed models.

Table 1 .
Macrophytes censed in the eulittoral and upper infralittoral zone of the studied Mediterranean area, with the indication of the division they belong to and the species collected for pollutant analyses, labelled with Roman numbers.
. Values are expressed as mean ± s.e.m., where applicable.