Five finfish and one shellfish species (Table 1) were collected from one or two sources: vessel-retrieved or/and retail-purchased from Oregon waters (see Supplementary Appendix SA1); the numbers from each source, categorized by species, are provided in Table 1. “Vessel-retrieved individuals” were whole-body fish or crustaceans caught by a fishing vessel in Oregon waters, the majority by the National Oceanic and Atmospheric Administration’s Northwest Fisheries Science Center Observers program during the 2021–2022 collection season, with the exception of riverine and ocean-phase Pacific lamprey that were collected during the 2017–2018 season (due to constraints imposed by the COVID-19 pandemic). Retail-purchased individuals were either fish fillets (finfish) or gutted shrimp purchased from a supermarket or seafood vendor. For one finfish species, lingcod, and the crustacean species, pink shrimp, samples were acquired from both fishing vessels (NOAA) and retail market (seafood counters at grocery stores), the source most accessible to the general public, and analyzed. Pacific lamprey species, listed on Oregon’s threatened and endangered species list, were acquired from ODFW and collected under permit.

Vessel-retrieved individuals were humanely euthanized by placing them in ice water baths, then double-wrapped in aluminum foil, and placed whole in plastic bags (IACUC-37, 2022). These individuals were kept frozen during transport to the Applied Coastal Ecology (ACE) Laboratory at Portland State University. Retail-purchased individuals (fillets and shelled shrimp) were purchased and transported to the ACE Laboratory as packaged by the retail store for a typical consumer. The packaging material included plastic-lined butcher paper, plastic takeout containers, and plastic freezer bags.

A total of 182 samples (122 finfish and 60 crustaceans) were collected from the Oregon coast and Oregon retail markets in 2021, while Pacific lamprey samples, collected in 2017, were frozen for later analysis.

For vessel-retrieved individuals, biological measurements, including full body length (in) and weight (g), and, if available, muscle tissue length (in) and weight (g) were recorded for all individuals. Whole-body individuals were dissected from behind the pectoral fin to the caudal fin to extract a filet of tissue corresponding to the parts of the animal typically consumed by humans.

For vessel-retrieved lingcod and Chinook salmon, approximately 220 g of muscle tissue (but ranging from 71–702 g) were randomly dissected from the front, middle, and end of the muscle tissue; for retail-purchased black rockfish and lingcod, filets were rinsed and then 32–133 (mean = 88) g of tissue were randomly dissected from the fillet. For shrimp (vessel and retail) and herring, individuals were rinsed and gutted; then all muscle tissue was dissected as all individuals sampled (75) were under 125 g (range 0.1–28 g). On average, vessel pink shrimp were 2 g (0.18–5.4 g), retail pink shrimp were 3.7 g (0.16–28 g), and vessel pacific herring were 10.77 g (3.5–19). For lamprey, riverine juveniles were defined and headed, and a small amount of muscle tissue was digested; for ocean-phase adult lamprey, we received and digested sections of muscle tissue (range 3–21 g) (see Supplementary Appendix Table SA1).

Dissected tissue from each individual was placed in its own 250 mL beaker and covered with a watch glass. Each sample was digested using a 10% potassium hydroxide (KOH) solution heated for 24–48 h at 40 C (except lamprey = 60 C), as outlined by Baechler et al. (2020), although density separation was not needed. A second digestion was needed for all lamprey samples. After digestion, the samples were vacuum-filtered through a 20-micron brass sieve (Hogentogler) to collect APs and remove the remaining liquefied tissue. The samples were then rinsed from the sieve into a vacuum filtration apparatus (Millipore Sigma) using a 47 mm diameter x 10-micron polycarbonate filter (10 μm, Millipore Sigma). Filters were then enclosed in PetriSlides (47 mm, Millipore Sigma).

Filters were examined to enumerate and measure APs using a Leica ICC50 HD with LAS V4.13 software and a ZEISS Primostar 3 with Labscope v3.3 software. Filters were examined under ×40 and ×10 magnification (depending on AP size) and photographed. Suspected APs were counted following the method outlined by Lusher et al. (2020) and classified based on their morphology, color, maximum width, and maximum length. Fiber bundles were separated into individual fibers when possible and classified accordingly. When fiber bundles could not be separated, the ends of fibers were used to count the total number of fibers in the bundle.

A subsample of suspected plastics (n = 209, 10% of total suspected APs encountered) was sent to the Ecotox and Environmental Stress Laboratory at Oregon State University to undergo micro-Fourier transform infrared (μFTIR) spectroscopy analysis to identify specific polymers and validate total counts. Six to fifteen suspected APs (depending on the number of individuals per species) were randomly selected from each species group and analyzed via FTIR spectroscopy. OpenSpecy (Cowger et al., 2021) was used to calibrate and confirm sample material types, following methods described by Caldwell et al. (2022), Talbot et al. (2022), and Lasdin et al. (2023).

Quality control protocols were adapted from Baechler et al. (2020) and Brander et al. (2020). Pink 100% cotton laboratory coats and facemasks were used during all processing steps. Pink material was chosen to easily identify AP input from researchers in samples and controls. In addition, 100% cotton clothing was worn at all times throughout sample processing. All containers, glassware, sieves, and beakers were triple-rinsed with DI water, inverted, covered with aluminum foil, and then air-dried to minimize paper towel fibers on glass surfaces. Three DI water procedural blanks were run in conjunction with each species batch (a total of 18 blanks; 3 per every 30 samples processed) and subjected to the same digestion, sieving, and filtering protocols. Air control blanks (total of 18, three for every species batch regardless of batch size), consisting of a clean polycarbonate filter inside a clean PetriSlide, followed each species batch through the process. A 1:1 sample-to-blank ratio was used to quantify APs entering samples during microscopy. Additionally, a snorkel hood was positioned over the sample on the microscope to minimize airborne contamination.

Using R-studio (version 4.1.2.) statistical software, analysis of variance (ANOVA) (alpha = 0.05) was performed to test for differences in AP load among species. A Tukey honestly significant difference post hoc test was used to confirm significant findings. ANOVA and Welch’s two-sample t-tests were performed to test for differences between source types of the same species (pink shrimp, lingcod, and Pacific lamprey). Spearman correlation coefficients were used to evaluate relationships between the total body weight or total filet weight and overall AP tissue burden. Plots were generated using the ggplot2 and vegan packages in R.

In organisms: Through microscope search, 1,806 suspected APs were identified across 180 of 182 individuals (averages varied drastically among species; Table 2). Fibers (1,466; 81.17%) were the most abundant, followed by fragments (332; 18.38%) and films (8; 0.44%) (see Supplementary Appendix Figure SA3B). The most common colors were blue (234; 12.95%), black (234; 12.95%), and clear or white (1,297; 71.81%) (see Supplementary Appendix Figure SA3C). The maximum length of APs ranged from 2.00–3,619 μm (mean = 911.78 μm ± 633.01), and the maximum width ranged from 0.477–1757.5 μm (mean = 26.56 μm ± 71.35) (Table 3; Supplementary Appendix Figure SA1).

In controls: Through visual search, a total of 190 suspected APs were identified in procedural, air, and microscopy controls (see Table 4 and Supplementary Appendix Table SA2 for additional details). Fibers (160; 84.21%) were the most abundant shape found, followed by fragments (25, 13.15%) (Supplementary Appendix Figure SA3A). The most common colors were clear (132; 69.74%), blue (26; 13.68%), and black (32; 16.84%). The maximum length of all APs found in controls ranged from 29.75 to 2,969.18 μm (mean of 744.05 μm ± 515.64), and the maximum width ranged from 3.59 to 367.33 μm (mean of 29.91 μm ± 44.12) (Supplementary Appendix Figure SA1).

APs were found in the muscle tissue of all species of finfish and shellfish sampled; of the 182 individuals sampled, only two individuals (1%) had no APs in the section of tissue sampled (one vessel-retrieved lingcod and one vessel-retrieved herring). Among the species sampled, pink shrimp contained the most APs per individual, regardless of source type (retail: 25 (12.6 ± 1.67) per individual; vessel: 36 (11.9 ± 1.22) per individual) (Figure 1), with the most particles (36) found in a single pink shrimp weighing 4.9 g (7.35 AP/g of tissue; Tables 2, 3). Vessel-retrieved Chinook contained the smallest abundance and concentration of APs (1–11 per individual and 0.028 AP/g; Table 3). AP ranges by species and source type varied (Figure 1; Table 3). Muscle tissue weight and AP burden were inversely correlated (Spearman rank = −0.23), indicating that smaller individuals are more likely to contain APs (ANOVA: f = 9.2 and p = 0.0028). Biological measurements were not obtainable for retail-purchased individuals.

Differences in average AP/g of tissue were inconsistent between retail and vessel-caught individuals across species (lingcod and pink shrimp). Retail lingcod contained more AP/individual and more APs/g of tissue (7.33 AP/individual, 0.091 AP/g) than vessel-retrieved lingcod (3.91 AP/individual, 0.022 AP/g; Welch’s t-test for AP/g tissue: t = −5.1, p = 8.79^–5) (Figure 1). However, retail pink shrimp contained slightly more APs/individual but fewer APs/g of tissue (12.6 AP/individual, 7.62 AP/g) than vessel-retrieved pink shrimp (11.9 AP/individual, 10.67 AP/g; Welch’s t-test for AP/g tissue: t = −1.2, p = 0.227) although the difference was not significant. An individual retail pink shrimp contained the most particles across all species and source types in a single individual (36 particles) (Figures 1C, D).

Comparing late-stage riverine juveniles and ocean phase-adult Pacific lamprey, the adults had marginally higher AP loads (15.9 versus 8.13 particles per individual) but marginally lower concentrations (0.6 AP/g versus 0.99 AP/g; Welch’s t-test for AP/g tissue: t = −1.32, p = 0.08) than juveniles (Figure 1B).

Of the 270 (∼10%) suspected APs tested using FTIR, 230 suspected APs were from individuals and 40 were from controls (detailed FTIR results in Supplementary Appendix Table SA3). OpenSpecy (Cowger et al., 2021) was used to calibrate and confirm sample material types according to methods described by Caldwell et al. (2022), Talbot et al. (2022), and Lasdin et al. (2023). In addition, 17.06% of suspected APs were fully synthetic materials, 9.47% were semi-synthetic, 8.05% were natural materials, and the overwhelming majority, 65.40%, were identified as anthropogenically modified (Supplementary Appendix Figures SA3A, SA3C). Synthetic and semi-synthetic material types included polyethylene terephthalate (PET; n = 33), polypropylene (PP; n = 3), high-density polyethene (HDPE; n = 1), low-density polyethylene (LDPE; n = 4), polyethylene vinyl acetate (PEVA; n = 1), fiberglass (n = 16), and semi-synthetic cardboard (n = 22). There was a single aramid fiber, a common material used in marine rope, flame-retardant fabrics, and military applications (Gong and Chen, 2016). Cellulose (n = 52), cotton fiber (n = 41), and cellulose acetate filter (n = 55) were the most common anthropogenically modified particles found.

AP contamination in procedural controls (average: 4.80 particles), fume hood blanks (average: 2.51 particles), and microscope blanks (0.82 particles) (Table 4) was averaged for each species batch, as per previous studies (Li et al., 2015; Rochman et al., 2015; Abbasi et al., 2018), and reported in Table 4 to provide an estimate of total contamination at each sample processing step. Pink MF contamination from laboratory clothing ranged from 0.41 to 1.25 particles per sample and was excluded from all counts. Supplementary Appendix Table SA1 details AP contamination across individuals.

The array of AP types and colors found across the six taxa highlights the complexity of identifying AP pollution sources in aquatic environments. Li et al. (2021) described using “microplastic communities”—APs of various colors, shapes, and polymer types that accumulate together in the environment—to elicit a potential number of AP pollution sources (Li et al., 2021). Of the published studies on APs in muscle tissue that share size categories, translocation into the muscle tissue is facilitated by shape (most frequently fibers) and size although this study found slightly larger fibers and fragments than those in fish muscle tissue of other studies (Barboza et al., 2020; McIlwraith et al., 2021). This may be due to differences in protocols, AP presence in the environment at each study area, or other factors not measured in these studies. While some organisms have a tendency to ingest APs of certain colors, shapes, and sizes (Q. Chen et al., 2020; Okamoto et al., 2022), further research is needed to understand how these variables affect translocation into the muscle tissue and toxicity to organisms (Mehito et al., 2022). A recent synthesis indicated that particles <80 μm could translocate in aquatic organisms (Mehinto et al., 2022), and given that the diameter of the typical microfiber is 10–15 μm, indications are that these particles could translocate if in the proper orientation.

Of the 182 individuals sampled, only two (one vessel-retrieved Pacific herring and one vessel-retrieved lingcod) had no detected APs in their tissue although these were smaller samples 0.18–19 g of muscle (Lingcod) or full tissue (herring) sampled. The study results mirror those of Akoueson et al. (2020) and Barboza et al. (2020) and provide evidence of the widespread presence of APs in the edible tissues of Oregon’s marine and freshwater species across trophic levels and feeding modes. Pink shrimp, which filter-feed in the upper water column which contains 8–9200 AP particles/m³, had the highest concentrations of APs (ODFW, 2019; Curren et al., 2020; Dawson et al., 2021; Yang et al., 2021). On the other hand, Chinook salmon had the lowest concentrations, followed by black rockfish and lingcod.

Across different life stages of lamprey, we hypothesized riverine-phase juvenile lamprey, which follow the filter-feeding phase, would contain more APs than the parasitic-feeding ocean-phase adult lamprey; on an AP/g of fish basis, this hypothesis proved true, with ocean-phase adults containing 0.604 AP/g compared to riverine juveniles with 0.999 AP/g; however, since adult lamprey are larger than juveniles, they had a higher total number of APs per individual. As adults attach to a host with their mouthparts and feed primarily on the bodily fluids of host organisms (Goodman and Reid, 2017), they may inherit a portion of their APs from the bloodstream of the host organisms they feed on.

This study aligns with other studies in that species across various trophic levels are exposed to AP pollution (Talley et al., 2020; F; Wang et al., 2021; Yıldız et al., 2022) and uptake and translocate particles into their tissue (e.g., Akhbarizadeh et al., 2019; Bagheri et al., 2020; Hossain et al., 2023; Nan et al., 2020; Figure 2, Supplementary Appendix Table SA2 and references therein). Furthermore, as evident from comparisons with other studies (Figure 2), we found evidence of an inverse relationship between muscle tissue AP concentration and trophic level, indicating a potential relationship between AP presence and habitat or feeding mode (see Figure 1A). Similar conclusions have been drawn by others studying trophic transfer and habitat depth (Aiguo et al., 2022; Carbery et al., 2018) and in comparisons of shellfish muscle tissue to that of higher trophic level finfish (Walkinshaw et al., 2020a).

Our study bolsters existing work on the AP muscle tissue presence, but further studies are needed to understand the mechanisms by which APs translocate into muscle tissue. Barboza et al. (2020) hypothesized that APs may transit through the bloodstream and into the muscle tissue (Barboza et al., 2020). Others hypothesize that macrophages may scavenge particles, leading to immune response and inflammation, which may facilitate translocation into the muscle tissue via cells (Leslie et al., 2022; Beijer et al., 2022).

Although other studies have found APs (MPs in the literature referenced) in retail market seafood products (Ferrante et al., 2022; Nalbone et al., 2021; Thiele et al., 2021), our results raise questions about the extent to which the retail process is a source. Our source type comparisons indicate ambiguity in retail processing as a source: AP concentrations were greater for retail than vessel-retrieved lingcod (after rinsing the surface flesh of fillets) but lower for retail than vessel-retrieved pink shrimp (after rinsing). Our results suggest that, in some cases, retail market individuals may be exposed to additional APs through processing, resulting in the incorporation of additional particles into the edible portions of seafood items. These post-mortem APs could be introduced by plastic packaging meant to preserve seafood (Dawson et al., 2021; Habib et al., 2022; Jadhav et al., 2021; Kedzierski et al., 2020). It is unclear why the retail process did not add microplastics to lingcod, as was observed for pink shrimp, indicating a need for further investigation to understand where and when AP contamination occurs post-catch.

Our sample size for the larger finfish was small, limiting the generalizability of the findings for these species. Additionally, since the study focused on species that span the US West Coast but only collected organisms from the Oregon coast, the concentrations may not represent coast-wide microplastic concentrations. However, a comparison with other studies on muscle tissue microplastic concentrations demonstrates that our results fall within the range of microplastics per gram of edible fish tissue found globally (Figure 2; Supplementary Appendix Table SA4). Future research should consider collecting the same species at various points along the West Coast to determine whether there is spatial variability in microplastic contamination from northern Washington to Southern California, particularly for the species that span the entire coast.