Hu et al. 2006 — Trophic magnification of triphenyltin vs tributyltin in a Bohai Bay marine food web

Hu and colleagues measured butyltins (MBT, DBT, TBT) and phenyltins (MPT, DPT, TPT) in plankton, five benthic invertebrate species, six fish species, and 14 paired surface seawater/sediment sites from Bohai Bay (North China) in 2002, and combined the biota concentrations with δ15N-derived trophic levels to compute trophic magnification factors. TPT showed strong food-web biomagnification (TMF = 3.70, r² = 0.78, p < 0.001) while TBT trended downward with trophic level (TMF = 0.59, p = 0.119, not statistically significant), so that mean TPT in fish exceeded mean TBT in fish even though seawater TPT was below detection (<6.8 ng/L) and seawater TBT was 9.8 ng/L. The authors attribute the divergence to TPT’s slower metabolic degradation in fish, not to a difference in environmental input.

Important note on metals classification. The analytes here are organotin compounds (MBT, DBT, TBT, MPT, DPT, TPT), not inorganic tin. Organotin species are distinct from total tin in toxicology, environmental behavior, and regulatory scope; they reach aquatic food chains primarily through antifouling paints (TBT, TPT) and agricultural fungicide use (TPT on potato, orchard, and cotton crops in China at the time), and they are outside the HMTc 10-analyte certification scope. They are recorded here under metals: [Sn] for element-level routing; downstream synthesis should link the toxicological discussion to organotins and must not pool these values with inorganic Sn occurrence.

Key numbers

All biota concentrations are ng/g wet weight unless noted. All concentrations expressed as cationic species per the source convention. Standard deviations follow ±. Sample counts (n) per row are from Table 2.

Trophic levels (δ15N-derived, Wan et al. 2005 Chin. Sci. Bull. 50:1021):

• phytoplankton/seston: 1.61 ± 0.14
• zooplankton: 2.00 ± 0.14 (TL reference for the food web)
• bay scallop: 2.15 ± 0.12
• short-necked clam: 2.17 ± 0.20
• veined rapa whelk: 2.79 ± 0.12
• mullet: 3.01 ± 0.44
• crab: 3.10 ± 0.17
• burrowing shrimp: 3.16 ± 0.14
• wolffish: 3.58 ± 0.11
• bartail flathead: 3.65 ± 0.24
• white flower croaker: 3.65 ± 0.39
• catfish: 3.67 ± 0.04
• weever: 3.88 ± 0.49

Table 2 — organotin concentrations (ng/g wet weight; ND = not detected):

Aggregated fish-group means (n = 22, six species combined):

• TPT: 19.9 ± 12.0 ng/g wet weight
• TBT: 5.1 ± 5.1 ng/g wet weight

Organotin profile across food-web tiers: Three butyltins (MBT, DBT, TBT) were detected in every species. Of the three phenyltins, TPT was detected in every species, DPT only in crab, bartail flathead, wolffish, and white flower croaker, and MPT in no biota sample.

Trophic magnification factors (from log organotin = a + b·TL; TMF = 10^b):

• TPT: TMF = 3.70 (log[TPT] = 0.57·TL − 0.77; r² = 0.78; p < 0.001) — significant biomagnification
• TBT: TMF = 0.59 (p = 0.119) — not statistically significant; TBT trends downward across TL
• MBT: not significant (p = 0.537)
• DBT: not significant (p = 0.116)
• MPT, DPT: TMF not calculable (insufficient detections)

The authors note that TPT TMF (3.70) is comparable to TMFs they previously measured in the same food web for DDE (3.26) and HCB (2.96).

Seawater (14 sites, September 2002; concentrations from companion paper Gao et al. 2004, Bull. Environ. Contam. Toxicol. 72:945, ref. 35):

• TBT mean: 9.8 ± 8.7 ng/L (n = 20)
• TPT: below detection limit of 6.8 ng/L

Surface sediment (14 sites, dry weight):

• TBT mean: 0.7 ng/g; range BDL to 2.3 ng/g; harbor site outlier: 40.6 ng/g
• TPT: detected in 4 of 14 sites at concentrations near the 0.1 ng/g LOD

Method detection limits (S/N = 3):

• Biota (ng/g wet weight, n = 6): MBT 0.7 ± 0.1; DBT 0.6 ± 0.1; TBT 0.8 ± 0.1; MPT 2.5 ± 0.5; DPT 1.0 ± 0.5; TPT 0.6 ± 0.1
• Sediment (ng/g dry weight, n = 5): MBT 0.4 ± 0.1; DBT 0.2 ± 0.1; TBT 0.2 ± 0.1; MPT 1.0 ± 0.4; DPT 0.2 ± 0.1; TPT 0.12 ± 0.02
• Water (ng/L, n = 5): MBT 1.5 ± 0.2; DBT 2.4 ± 0.2; TBT 4.1 ± 0.4; MPT 3.6 ± 0.4; DPT 5.1 ± 0.5; TPT 6.8 ± 0.7

Deuterium-labeled surrogate recoveries:

• Biota (n = 6): MBT-d9 90 ± 8%, DBT-d18 90 ± 7%, TBT-d27 103 ± 10%, TPT-d15 83 ± 8%; MPT-d5 limited to 26%; DPT-d10 limited to 46%
• Sediment (n = 5): MBT-d9 63 ± 17%, DBT-d18 93 ± 18%, TBT-d27 104 ± 22%, DPT-d10 77 ± 9%, TPT-d15 63 ± 12%; MPT-d5 limited to 15%

The authors flag the low MPT and DPT recoveries as the reason MPT and DPT TMFs were not computed.

Methods (brief)

Samples were collected May, June, and September 2002 from Bohai Bay (six plankton-tow stations and trawl stations for benthic invertebrates and fish; 14 paired water/sediment sites in September). Frozen biota were thawed and homogenized; 5 g of tissue (wet phytoplankton/seston and zooplankton, soft tissue of invertebrates, muscle of fish) or 5 g of air-dried sediment was spiked with six deuterium-labeled surrogate analogues (MBT-d9, DBT-d18, TBT-d27, MPT-d5, DPT-d10, TPT-d15) and extracted by ultrasonic shaking with 30 mL of 0.03% (w/v) tropolone in methanol plus 2 mL of concentrated HCl. After centrifugation, the supernatant was partitioned three times into a 30% NaCl–water / dichloromethane mixture; the organic layer was concentrated and derivatized with sodium tetraethylborate (NaBEt4) in pH 5.0 acetate buffer. Biota extracts were additionally refluxed for ~1 h in 1 N KOH–ethanol for saponification, then partitioned three times into hexane. The hexane extract was passed through a 1 g Florisil cartridge layered with sodium sulfate and eluted with hexane/diethyl ether (9:1); the eluate was concentrated to 0.3 mL (biota) or 0.1 mL (sediment), and tetrabutyltin-d36 (TeBT-d36) was added as instrument internal standard. Water samples (~20 mL) were derivatized in 40 mL amber glass vials and analyzed by headspace SPME (headspace mode for TBT, DBT, MBT, MPT; immersion for TPT, DPT) following Chou and Lee 2005. Analysis was performed on a Hewlett-Packard 5890 GC coupled to a 5971 mass spectrometer in electron impact mode (70 eV) with selected ion monitoring on a 60 m × 0.25 mm i.d., 0.25 µm HP-5MS capillary column; injector 270 °C, detector source 280 °C, oven 60 (2 min) → 130 °C at 20 °C/min (26 min) → 280 °C at 20 °C/min (7 min); 1 µL splitless injection. Quantitation was by isotope dilution against the matched deuterated surrogate for each target, with TeBT-d36 as injection internal standard. Trophic levels were derived from δ15N as TL = 2 + (δ15N_consumer − δ15N_zooplankton)/3.8 using zooplankton as TL = 2, following the authors’ prior Bohai Bay food-web δ15N/δ13C work (Wan et al. 2005). TMFs were computed by linear regression of log10(concentration) on TL and taken as 10^b; correlations were tested by Pearson rank correlation with p < 0.05 as the significance criterion. Lipid content was determined gravimetrically after 24 h Soxhlet extraction in dichloromethane/methanol (7:3 v/v). The authors flag MPT and DPT recoveries below 50% as a quantitation limitation, which is why MPT and DPT TMFs were not calculated.

Implications

• Certification (HMTc): Provides occurrence and trophic-magnification context for organotin compounds in marine fish, shellfish, and molluscs from a heavily industrialized Chinese bay sampled in 2002. The HMTc 10-analyte vocabulary covers Sn as inorganic/total tin (canned-food migration regime); organotin TBT/TPT contamination is outside that scope and should not be folded into Sn occurrence pools for HMTc threshold work. The data do, however, document that marine-fish TPT body burdens can exceed seawater TPT inputs by orders of magnitude through food-web biomagnification, which is relevant any time the certification program considers an organotin-specific seafood analyte.
• Courses: A textbook biomagnification case study. The contrast between TBT (TMF = 0.59, no significant trophic trend) and TPT (TMF = 3.70, comparable to legacy organochlorines DDE and HCB) in the same food web, combined with seawater TPT below detection while seawater TBT was 9.8 ng/L, makes the point that environmental input and biota body burden can move in opposite directions when metabolic clearance differs between congeners.
• App: For the ingredient-risk estimator, the relevant signal is that marine fish at high trophic levels (bartail flathead, wolffish, white flower croaker) carry an order of magnitude more TPT than benthic invertebrates from the same bay (~25–35 vs ~3–5 ng/g wet weight). Organotins are out of the HMI public-app metal scope by default, but the trophic-position signal will recur for any analyte whose congener-specific metabolism differs across the food web.

Wiki pages this source may touch

• fish
• shellfish
• molluscs
• seafood
• seafood
• fresh-fish
• shellfish
• fish-marine-non-predatory
• fish-marine-predatory
• organotins
• tin

Verification notes

• 2026-06-04 fresh ingest from raw/manual-fetch/Kimi_Agent_Download Corruption Issue/seafood_papers/04_Shellfish/. PDF SHA-256 517dc7ae…7e9347. DOI 10.1021/es0514747 verified against page header. Three identity checks (DOI, raw_handle, cite-key) returned no prior wiki page.
• The source measures organotin compounds (TBT, TPT, and their MBT/DBT/MPT/DPT degradation products), not inorganic tin. Frontmatter uses metals: [Sn] for element-level routing consistent with the cfs2019-organotin-aquatic-hongkong precedent; downstream synthesis must link the toxicology to organotins and must not pool these values with inorganic Sn occurrence used for canned-food / canned-tuna threshold work.
• No ingredients/crustaceans slug in the current taxonomy; crab and burrowing shrimp are routed under ingredients/shellfish for the wiki layer while preserving their specific identification in the per-species Key numbers table.
• No ingredients/plankton or matrices term for plankton in the current vocabulary; phytoplankton/seston and zooplankton are retained in the body-table for completeness as trophic-baseline rows, not as routed ingredients.
• The seawater results (TBT and TPT in 14 sites) are cited from the companion paper Gao et al. 2004 (Bull. Environ. Contam. Toxicol. 72:945, ref. 35 in this article) rather than measured in this article itself; the authors use them as the surface-water surrogate for input-vs-trophic-magnification interpretation. Reproduced here exactly as the authors report them and attributed in the table notes.
• No brand names appear in the source’s Key numbers or biota tables; species are reported with Linnaean binomials. No brand-firewall edits required. Instrument vendor names (Hewlett-Packard, ACROS, Aldrich, Wako, Hayashi, Fisher, DIKMA, Siyou, Waters) appear only in Materials and Methods and are retained per the locked-2026-05-17 scientific-method-vendor exception.
• The earlier matrix seawater is not in the documented matrices vocabulary; seawater results are confined to the Key numbers narrative and the matrices array lists sediment plus the three biota matrices that fan out to routed ingredient pages.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.