Ren et al. 2025 — Cadmium–enrofloxacin co-exposure toxicology in zebrafish

Controlled-exposure zebrafish study (AB strain; 6,000 fish purchased, of which 3,900 were used in the chronic 20-d arm and an additional ~570 across the acute 96-h Cd-alone, ENR-alone, and combined arms) characterizing the joint toxicity of cadmium (Cd) and the fluoroquinolone antibiotic enrofloxacin (ENR). Co-exposure approximately doubled acute lethality of each compound (LC50 Cd 89.12 mg/L → 46.35 mg/L; LC50 ENR 190.11 mg/L → 99.39 mg/L) with the Concentration Addition model classifying the interaction as additive (MDR 0.96, 95% CI 0.58–1.04). Under chronic exposure, ENR enhanced Cd accumulation in zebrafish liver, intestine, and muscle by 1.11–2.33-fold over single-Cd controls and exacerbated hepatic oxidative stress (SOD, CAT, GSH-Px elevated 1.34–7.06-fold at day 8 then declining 9.9–48.98% by day 20; MDA rising to 4.06-fold above controls). 16S rRNA sequencing of gut microbiota showed Cd alone reduced α-diversity (Shannon −57 to −63%) with a 16–20% drop in beneficial Cetobacterium and a 44–114% increase in Aeromonas; co-exposure further amplified the dysbiosis. The paper measures no food-matrix contamination; its HMI relevance is mechanistic (Cd toxicokinetics, antibiotic-driven Cd bioavailability enhancement, and Cd-driven gut dysbiosis).

Key numbers

All Cd and ENR values below are exposure concentrations (mg/L water-borne) or tissue accumulations (ng/g wet weight) measured in zebrafish under controlled laboratory conditions, not food-matrix contamination. Page and figure references are to Toxics 2025, 13, 378 (21 pp.).

Acute 96-h LC50 (Section 3.1, Figure 1, p. 6):

• Cd alone: 96-h LC50 = 89.12 mg/L (six exposure concentrations 31–122 mg/L, n = 30 fish per concentration in triplicate).
• ENR alone: 96-h LC50 = 190.11 mg/L (six exposure concentrations 120–320 mg/L).
• Combined Cd + ENR at fixed Cd:ENR ratio 1:2.1: 96-h LC50 for the Cd component = 46.35 mg/L; for the ENR component = 99.39 mg/L. Combined exposure was 1.92× more toxic than Cd alone and 1.91× more toxic than ENR alone.
• Concentration Addition model predicted combined effect concentration = 139.91 mg/L (sum); Model Deviation Ratio (MDR) = 0.96 with 95% CI 0.58–1.04 — classifies the interaction as additive, not synergistic, by the CA criterion. (Note: the abstract’s phrase “synergistic effects” describes the empirical LC50 shift; the formal CA/MDR classification in Section 3.1 is additive.)

Chronic 20-d exposure design (Section 2.2, pp. 3–4):

• Five groups: CK (control); L-Cd (Cd 0.046 mg/L); H-Cd (Cd 0.46 mg/L); L-CE (Cd 0.046 + ENR 0.099 mg/L); H-CE (Cd 0.46 + ENR 0.99 mg/L). Co-exposure ratios set at 1/100 and 1/1000 of the combined 96-h LC50.
• 3,900 fish total; three replicates per group; 260 fish per tank; water replaced every 4 days; Cd by ICP-MS (Agilent 7800), ENR by UHPLC-MS/MS (Agilent 1290-6470A); experimental concentrations held within 80–120% of theoretical values; controls below instrument detection limits.

Cd tissue accumulation under chronic co-exposure (Section 3.2, Figure 2, p. 7):

• Tissue accumulation rank order at day 20: intestine > liver > muscle.
• Cd-only liver at day 20: rose with prolonged exposure; the H-Cd liver accumulated ≈1,700 ng/g at day 20 (Figure 2a bar chart).
• Cd-only intestine at day 20: H-Cd intestine ≈50,000 ng/g (Figure 2b bar chart).
• Cd-only muscle at day 20: H-Cd muscle ≈1,150 ng/g (Figure 2c bar chart).
• Co-exposure vs single-Cd at day 20, low-concentration arm (L-CE vs L-Cd): liver +1.13-fold, intestine +1.20-fold, muscle +1.27-fold.
• Co-exposure vs single-Cd at day 20, high-concentration arm (H-CE vs H-Cd): liver +2.33-fold, intestine +2.20-fold, muscle +2.26-fold.
• Overall co-exposure increased Cd accumulation 1.11–2.33-fold across tissues and time points relative to single Cd; tissue concentrations of Cd in the intestine at day 20 were 9.54–31.60 times those in liver and 16.93–38.49 times those in muscle (Section 4.1, p. 14); liver Cd 1.22–1.77 times muscle Cd.

Hepatic oxidative-stress markers (Section 3.3, Figure 3, p. 7; days 1, 8, 20):

• All exposure groups had significantly elevated SOD, CAT, GSH-Px activities and MDA content vs CK (p < 0.05); the three enzymes showed a biphasic time course (initial rise then decline); MDA accumulated monotonically.
• Initial elevation across all exposure groups (paper-level summary, abstract and Section 3.3): SOD ×1.34–7.06, GSH-Px ×0.98–3.28, CAT ×1.53–3.65 at day 8.
• Later decline at day 20: SOD, CAT, GSH-Px reduced 9.9–48.98% relative to day-8 peaks.
• MDA at day 20: up to 4.06-fold above controls.
• Low-concentration co-exposure increment (L-CE vs L-Cd) at days 1/8/20: SOD activity +20.36%, +36.62%, +28.82%; CAT +11.24% to +17.15%; GSH-Px +20.78% to +28.53%; MDA +12.37% to +14.40%.
• High-concentration co-exposure increment (H-CE vs H-Cd) at the same time points: SOD +15.12% to +33.80%; CAT +29.87% to +31.06%; GSH-Px +18.80% to +25.11%; MDA +16.66% to +34.45%.
• Abstract-level summary: co-exposure elevated oxidative stress 11.24–34.48% relative to single exposures.

Gut microbiota — α-diversity and composition (Section 3.4, Figures 4–7, pp. 9–11):

• 15 samples sequenced; 263,591 optimized sequences, 55,328,742 bp; average read 1,049 bp; 1 domain, 1 kingdom, 14 phyla, 28 classes, 79 orders, 123 families, 258 genera, 407 species, 541 OTUs.
• Shannon α-diversity index decreased in single-Cd groups (L-Cd, H-Cd) vs CK; increased in combined groups (L-CE, H-CE) vs single Cd. Simpson index dropped further in combined groups than in single Cd. Chao and ACE indices rose under co-exposure (H-CE highest).
• Abstract-level summary: Cd exposure reduced α-diversity Shannon index by 57–63% relative to control.
• PCA: significant separation between L-Cd vs L-CE and H-Cd vs H-CE (R² = 0.6588, p = 0.002); PC1 explained 39.41%.
• PCoA / PERMANOVA between single and combined exposure groups: R² = 0.6692, p = 0.002; PCo1 60.84%, PCo2 28.10% (cumulative 88.94%).
• Phylum-level dominants: Pseudomonadota (≡Proteobacteria), Fusobacteriota, Bacillota (≡Firmicutes), Actinomycetota (≡Actinobacteriota), Bacteroidota (≡Bacteroidota).
• CK and L-Cd groups dominated by Fusobacteriota (49.80% and 41.72% respectively); H-Cd, L-CE, H-CE dominated by Pseudomonadota (40.64%, 49.53%, 40.44%).
• Compared to CK, Fusobacteriota decreased and Bacillota increased in Cd-alone groups. L-CE vs L-Cd showed increases in Pseudomonadota and Actinomycetota and decreases in Fusobacteriota and Bacillota; H-CE vs H-Cd showed a significant decrease in Fusobacteriota and increases in Bacillota and Actinomycetota.
• Top genera: Aeromonas, Cetobacterium, ZOR0006, Shewanella, Mycoplasma.
• Abstract-level summary: 16–20% reduction in beneficial Cetobacterium; 44–114% increase in pathogenic Aeromonas under Cd exposure; co-exposure exacerbated this dysbiosis.
• Genus-level comparisons: vs CK, Aeromonas and ZOR0006 increased and Cetobacterium decreased in Cd-alone groups. L-CE vs L-Cd: Aeromonas and ZOR0006 increased, Cetobacterium decreased. H-CE vs H-Cd: Aeromonas and Cetobacterium decreased significantly, ZOR0006 increased significantly.
• LEfSe (LDA > 3.0): 40 indicator species identified — CK (10), L-Cd (5), H-Cd (7), L-CE (8), H-CE (10).
• Co-occurrence network: CK 50 nodes/183 edges (modularity 0.59); L-Cd 43/197 (0.60); L-CE 41/119 (0.50); H-Cd 45/155 (0.48); H-CE 43/162 (0.51). Network degree and centrality decreased under most exposures vs CK (except low-Cd group); network complexity rose under combined exposure (Figure 8H).

Methods (brief)

Static-renewal aquatic-toxicology design under China’s “Chemicals — Acute Toxicity Test for Fish” GB/T 27861-2011 protocol (Approval No. BMY-ZY-202402, Guizhou Provincial Analytical Testing Research Institute). Zebrafish (AB strain, 3–4 months) acclimatized 15 days; rearing water aerated 48 h, quartz-filtered, UV-sterilized; 25 ± 1 °C; pH 7.2–7.5; 14 h light/10 h dark. Acute test: 96-h static exposure, six concentration gradients per compound and per mixture, 10 fish per replicate × 3 replicates per concentration, mortality scored at 24, 48, 72, 96 h, LC50 via probit-concentration regression and Fieller’s theorem in SPSS 2019. Chronic test: 20-d exposure at 1/100 and 1/1000 of combined LC50; water replaced every 4 days; Cd in tissues by Agilent 7800 ICP-MS after polytetrafluoroethylene-vessel microwave digestion (Mars 6, CEM) in 8 mL HNO₃ + 2 mL H₂O₂ for 4 h; ENR by Agilent 1290-6470A UHPLC-MS/MS. Antioxidant enzymes (SOD, CAT, GSH-Px), MDA, and total protein measured on 10% liver homogenates (Sorvall Legend Micro 21R, Thermo Fisher) using Nanjing Jiancheng Bioengineering Institute kits read on a Multiskan GO microplate reader (Thermo Fisher). 16S rRNA V3–V4 region (primers 338F/806R, Sangon Biotech) amplified, library-prepped, sequenced on Illumina NextSeq2000 (Majorbio); OTUs clustered at 97% UPARSE; taxonomy annotated against SILVA v138; analyses in R packages vegan (α-diversity), PCA/PCoA (β-diversity), LEfSe (LDA > 3.0, p < 0.05), and Hmisc + igraph for Spearman co-occurrence networks (|r| > 0.6, p < 0.05). Concentration Addition model and Model Deviation Ratio used to classify joint toxicity. Joint toxicity assessment formulas: EC_x,mix = (Σ pᵢ / EC_x,i)⁻¹; MDR = EC_x,PRE / EC_x,OBS; MDR_Upper = EC_x,Upper/EC_x,OBS; MDR_Lower = EC_x,Lower/EC_x,OBS. Synergy if MDR > MDR_Upper; antagonism if MDR < MDR_Lower; addition otherwise.

Limitations the authors acknowledge: exposure concentrations are higher than typical environmental levels (the cited environmental ENR range is 0.019–11.9 µg/L; Cd freshwater EU limit 0.25 µg/L); zebrafish responses extrapolate to other teleosts and humans only with caution; no molecular-mechanism work was performed.

Implications

• Certification (HMTc): The paper measures no food-matrix Cd contamination and proposes no certification threshold. The relevance to HMTc is indirect — it documents an antibiotic-driven mechanism (Cd-ENR complexation enhancing Cd absorption 1.11–2.33-fold) that strengthens the case for considering co-pollutant context when interpreting Cd biomonitoring data from aquaculture-sourced ingredients. No threshold contribution.
• Courses: Teachable as a case study in joint-toxicity classification (CA model vs empirical LC50 doubling) and in the distinction between “synergistic” as a colloquial descriptor and as a formal CA/MDR classification.
• App: Does not contribute primary occurrence values to any ingredient’s contamination_profile. Belongs in the cadmium mechanism evidence layer, not the contamination layer.
• Microbiome: Direct evidence for Cd-driven gut microbial dysbiosis in a teleost model (Shannon −57 to −63% under Cd; Cetobacterium −16 to −20%; Aeromonas +44 to +114%). Relevant to cadmium-gut-axis crosswalk for WikiBiome federation and to the broader Cd-microbiome literature accumulating on this wiki.

Wiki pages this source may touch

• Cadmium

Verification notes

• The paper’s abstract calls the joint toxicity “synergistic” based on the LC50 doubling, but Section 3.1 explicitly classifies the interaction as additive under the Concentration Addition model (MDR 0.96, within the 0.58–1.04 confidence band). Both phrasings are preserved here verbatim; the apparent contradiction is internal to the source.
• Audit subagent 2026-06-02 flagged the original “n≈10,000 across acute and chronic arms” prose as overstated; verified against PDF pp. 3–4 (6,000 fish purchased; 3,900 chronic; 360 + ~210 acute = ~570 acute) and corrected the opening paragraph to the actual cohort breakdown.
• Audit subagent 2026-06-02 flagged a missing wikilink target cadmium-gut-axis; verified that wiki/microbiome/ contains only index.md and nickel-microbial-pathogenesis.md as of 2026-06-02. Removed the dead wikilink from “Wiki pages this source may touch”. A Cd-gut-axis microbiome page does not yet exist; surfacing here as a new-page candidate for Karen rather than auto-creating per Part 10. This source contributes the following empirical anchors if such a page is later scaffolded: Shannon α-diversity −57 to −63%; Cetobacterium −16 to −20%; Aeromonas +44 to +114%; co-exposure with antibiotic amplifies the dysbiosis.
• Audit subagent 2026-06-02 flagged a paper-internal inconsistency in the ENR acute exposure ladder: the methods text on PDF p. 3 states “six gradient concentrations were established,” but the explicit enumeration on the same page lists only five ENR concentrations (120, 160, 240, 280, 320 mg/L) — the value near ~200 mg/L appears to be omitted from the printed list. Verified against the source; the wiki page retains the paper’s “six exposure concentrations 120–320 mg/L” wording in Key numbers but flags the source-internal list/text inconsistency here.
• No food matrix is measured; ingredients: [] and products: [] reflect the controlled model-organism scope. matrices: [fish-tissue] captures the analytical matrix (zebrafish liver, intestine, muscle) without overspecifying.
• Speciation: total Cd (Cd²⁺ from CdCl₂ AR-grade dosing, ICP-MS detection of total Cd in digested tissue). No Cd species fractionation; this is total Cd throughout.
• No brand names attached to contamination values (per Part 12). Scientific instrument/reagent vendor names retained per the Exception 2 carve-out: Agilent 7800 ICP-MS, Agilent 1290-6470A UHPLC-MS/MS, CEM Mars 6 microwave digestion, Sorvall Legend Micro 21R, Thermo Fisher Multiskan GO, Nanjing Jiancheng Bioengineering Institute assay kits, Shanghai Macklin Biochemical Co., Tianjin Komiou Chemical Reagent Co., Sinopharm Chemical Reagent Co., Shanghai Hongye Ornamental Fish Farm, Sangon Biotech, Illumina NextSeq2000, Majorbio Cloud Platform, SILVA v138, SPSS 2019, Origin 2020.
• Cite key derived from first author + year + topic (Ren + 2025 + cadmium-enrofloxacin-zebrafish). No triage-manifest cite key was pre-populated.
• Jurisdiction: CN (Guizhou University and Guizhou Academy of Testing and Analysis, Guiyang). Funding/approval at Guizhou Provincial Analytical Testing Research Institute (Approval No. BMY-ZY-202402).

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.