The presence and biodistribution of perfluorooctanoic acid (PFOA) in Paracentrotus lividus highlights its potential application for environmental biomonitoring

Collecting samples

Three sampling campaigns were carried out at the two sampling sites (A and B) on the northwest coast of Sicily (Fig. 1a) chosen for this study. The main site characteristics and sampling details are summarized in Table S1 (Additional Information). A total of 90 specimens of sea urchins Paracentrotus livid (45 specimens per site), 30 l of seawater (15 per site), 40 samples (20 per site) of seagrass Oceanic posidonia (less than 5 cm of leaf fragments, according to institutional and national ethical guidelines) were collected and analyzed with 30 l of brackish water from site B (15 l from each stream).

The sampling activity has been authorized by the Capitaneria di Porto de Palermo with the protocol number: 0029430. In the absence of data on PFOA contamination in the most recent report on chemical contamination in the region coastal subject to this study31, the choice of sampling sites was based on supposedly different pollution statuses depending on the location of the site or the proximity of human activities (eg restaurants, pipes, sewers, etc.).

Site A (see Fig. 1d), was chosen assuming a lower pollution state due to its position near Capo Zafferano, at the northern end of the bay of S. Elia, with an average depth of 11 m and rocky seabed (see Fig. 1b and additional information: Table S1). Conversely, site B (see Fig. 1e was chosen in the same coastal area (only 4.7 km from site A) assuming a higher pollution status due to its position on the south side. from the promontory of Solanto, near a pipeline and the mouths of two small inland coves, with a shallow sandy bottom (3 m) and where a bathing ban is in force32 (see Fig. 1b, c and Additional Information: Table S1).

The biodistribution of PFOA in the different matrices was evaluated by analyzing sea urchin coelomocytes (CC) (90 samples) and coelomic fluid (CF) (90 samples), as well as the gonads (G) (63 samples of 32 sea urchins collected from site A and 31 sea urchins collected from site B), or mixed organs (MIX) (27 samples from 13 sea urchins collected from site A and 14 sea urchins collected from site B) consisting of a homogenized mixture of sea urchin internal matrices when the gonads were not sufficiently developed for sampling. Due to their mutually exclusive nature, the last two datasets (G and MIX) were merged and labeled as “Gonads or Mixed Organs” (GoM) for statistical analysis and graphical representations which required a uniform dataset of. 45 items per site. Further details on the collection of matrices and their labeling are described in the Supplementary Information.

The size of the sea urchins (horizontal diameter without thorns) varied between 30 and 51 mm indicating specimens that lived in their respective sites for approximately 3 to 5 years25.

PFOA extraction and analysis

The hardware, equipment and software are described in the Additional Information.

PFOA extraction procedures have been adapted33 the type of matrix to be analyzed. The percentages of recovery (R%) were checked for each batch of analyzes by adding blank samples with different amounts of PFOA analytical standard before the extraction procedure.33.

The spiked samples underwent the same extraction procedure as the unenriched samples and the percent recovery R was calculated according to Eq. 1, where Cpeak is the known concentration of enriched PFOA, Dsharp is the instrumental analytical response (LC-MS) of the spiked sample (i.e. the “detected” concentration), Dpointless is the analytical response of the unenriched sample. R was then used in the equation. 2 to calculate the actual values, [PFOA], the concentrations of PFOA in the non-enriched analyzed samples.

$${ text {R}} = 100 times left ({{ text {D}} _ {{{ text {spiked}}}} – { text {D}} _ {{{ text {unspike}}}}} right) / { text {C}} _ ​​{{{ text {spike}}}}$$

(1)

$$left[ {{text{PFOA}}} right] = 100 times { text {D}} _ {{{ text {unspiked}}}} / { text {R}}$$

(2)

With the exception of [PFOA]sea ​​water and [PFOA]stream, which are expressed in nanograms per liter (ppt), all other concentrations of PFOA are expressed in nanograms per gram of matrix (ppb).

The PFOA standard was used for calibration before each batch of analyzes and a linear response (R2> 0.99) was recorded in the concentration range of 0.1 to 1000 ppb. RSDs on three replicates were less than 10%. LOD (0.1 ppb) and LOQ (1.0 ppb) were quantified by the IUPAC method. LC-MS analyzes were performed in negative ion monitoring mode (see Additional Information).

For the analysis of P. livid samples, an estimate of the total concentration of PFOA, [PFOA]EARLY in ng / g, in each sea urchin was calculated considering the sampled weight (W) in grams of each matrix (Eq. 3):

$$left[ {{text{PFOA}}} right]_ {{{ text {TOT}}}} = left ({{ text {W}} _ {{{ text {CF}}}} left[ {{text{PFOA}}} right]_ {{{ text {CF}}}} + { text {W}} _ {{{ text {CC}}}} left[ {{text{PFOA}}} right]_ {{{ text {CC}}}} + { text {W}} _ {{{ text {GoM}}}} left[ {{text{PFOA}}} right]_ {{{ text {GoM}}}}} right) / left ({{ text {W}} _ {{{ text {CF}}}} + { text {W}} _ { {{ text {CC}}}} + { text {W}} _ {{{ text {GoM}}}}} right)$$

(3)

Water analysis

During each of the 3 sampling campaigns, 2 seawater samples (5 l from site A and 5 l from site B) and 2 brackish water samples (5 l from each creek mouth of site B) were collected for a total of 6 seawater samples and 6 brackish water samples.

Samples were checked for the presence of PFOA by solid phase extraction (SPE) (see Additional Information) followed by LC-MS analysis34.

The percentage recovery, calculated according to Eq. 1, was R = 120%. [PFOA]sea ​​water and [PFOA]stream the concentrations (ng / L) were determined from the analytical data according to Eq. 2.

Oceanic posidonia analysis

A total of 40 leaf samples were taken from different individuals of P. oceanic (20 samples from site A and 20 samples from site B). Each sample was cut into small pieces and homogenized using an agate mortar and pestle, weighed (0.5 g) and transferred to a glass tube for extraction (see Additional Information).

The percentage recovery, calculated according to Eq. 1, was R = 70%. [PFOA]P. oceanic the concentrations (ng / g) were determined from the analytical data according to Eq. 2.

Analysis of coelomocytes and coelomic fluid

Coelomic fluid, also containing the coelomocyte population, was withdrawn from the ninety samples collected (45 per site) by inserting an ultrafine, sharp needle (32G 0.26 mm × 12 mm) from a syringe of 1 ml through the peristomal membrane.35 . All samples were centrifuged at 4 ° C and 1500 rpm for 5 min in a 5804R refrigerated centrifuge (Eppendorf, Germany) thereby separating the supernatant coelomic fluid (FC) coelomocytes (CC). CF and CC were then weighed and placed in different glass tubes for subsequent PFOA extractions (see Additional Information).

The percentage of recovery, calculated according to Eq. 1, was R = 28% for CF and R = 68% for CC. [PFOA]FC and [PFOA]CC the concentrations (ng / g) were determined from the analytical data according to Eq. 2.

PFOA extraction from 63 gonad samples (32 from site A and 31 from site B) was performed with LC-MS grade methanol following the same procedure used for extraction from of CF and CC (5 mL for samples greater than 0.5 g; 2.5 mL for samples between 0.1 g and 0.5 g). In case of undetected PFOA (considered zero values ​​in graphs and processing of statistical data), analyzes were repeated for confirmation on concentrated sample extracts.

Twenty spiked samples were prepared from the most abundant gonad samples (10 spiked samples per site), by adding 25 µL of a 1 mg / L aqueous stock solution of PFOA to 0.25 g of samples of gonads. The percent recovery of PFOA from the gonads, calculated according to Eq. 1, was R = 73%. [PFOA]g the concentrations (ng / g) were determined from the analytical data according to Eq. 2.

Mixed organ analysis

In 27 sea urchin specimens (13 from site A and 14 from site B), the state of development was not sufficient to collect at least 0.1 g of gonad sample. For these individuals, the organs remaining after the collection of CF and CC, mainly the intestine and the undeveloped gonads, were mixed and extracted in the same way as the other matrices.

Spiked samples were prepared by adding 25 µL of a 1 mg / L aqueous stock solution of PFOA to 0.25 g of mixed organs (TO MIX TOGETHER) samples. The percentage of PFOA recovered from TO MIX TOGETHER, calculated according to Eq. 1, was R = 20%. [PFOA]TO MIX TOGETHER the concentrations (ng / g) were determined from the analytical data according to Eq. 2.

Statistical analyzes and graphical representation of data

The distribution of PFOA concentrations in all matrices sampled from collected sea urchins is graphically represented by a box and a jitter diagram (Fig. 2) where the 25 to 75 percentiles are plotted using a box. ; the minimum and maximum are indicated at the end of the thin lines (whiskers), while the median is marked as a horizontal line in the boxfitting. Statistical tests and linear fits were used to assess the significance and correlations of the data (see Additional Information).

A permutational multivariate analysis of variance PERMANOVA36was carried out to assess the differences in the concentrations of PFOA between the two groups of sea urchins collected from site A and site B. The experimental setup included a factor (Site) two levels (fixed and orthogonal) and four variables corresponding to the concentrations of PFOA in each type of sample analyzed (coelomocytes, coelomic fluid, gonad or mixed organs) including the estimated total concentration of PFOA. Each term of the analysis was tested by 999 random permutations.

Finally, principal component analysis (PCA) (see Additional information: PCA tables and graphs) was performed on a data set containing five variables. More precisely the size of the sea urchins and the concentrations of PFOA in each type of sample (CF, CC and GoM) as well as in the whole sea urchin (TOT), in order to verify the multivariate character of the data in a number relatively small in size, thus limiting the loss of information.