Dose-Dependent Effect of Tilmicosin Residues on erm A Rebound Mediated by IntI 1 in Pig Manure Compost

Abstract

The impact of varying antibiotic residue levels on antibiotic resistance gene (ARG) removal during composting is still unclear. This study investigated the impact of different residue levels of tilmicosin (TIM), a common veterinary macrolide antibiotic, on ARG removal during pig manure composting. Three groups were used: the CK group (no TIM), the L group (246.49 ± 22.83 mg/kg TIM), and the H group (529.99 ± 16.15 mg/kg TIM). Composting removed most targeted macrolide resistance genes (MRGs) like ereA, ermC, and ermF (>90% removal), and reduced ermB, ermX, ermQ, acrA, acrB, and mefA (30-70% removal). However, ermA increased in abundance. TIM altered compost community structure, driving succession through a deterministic process. At low doses, TIM reduced MRG-bacteria co-occurrence, with horizontal gene transfer via intI1 being the main cause of ermA rebound. In conclusion, composting reduces many MRG levels in pig manure, but the persistence and rebound of genes like ermA reveal the complex interactions between composting conditions and microbial gene transfer.

Keywords: bacteria; composting; horizontal gene transfer; macrolide resistance genes; pig manure; tilmicosin.

Figure 1

Composting process monitoring: (A) system design, (B) moisture content, (C) temperature changes, (D) TIM levels. Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg).

Figure 2

Temporal variations in relative abundance of 10 identified MRGs during composting (A–J). Removal rates of MRGs at the end of composting (K). The asterisk (*) in the upper right of the MRGs indicates that it is a multiple-antibiotic resistance gene. Different letters (a, b, c) denote significant inter-group differences (p < 0.05), while shared letters indicate non-significance. This suggests that the treatment had a significant effect on the experimental results. Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg).

Figure 3

Dynamics of MGEs (intI1, intI2, tnpA and Tn916/1545) during composting (A–D). Different letters (a, b, c) denote significant inter-group differences (p < 0.05), while shared letters indicate non-significance. Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg). Correlations between the relative abundances of MRGs and MGEs (E–I).

Figure 4

Microbial diversity dynamics across composting treatments. Alpha diversity dynamics (Shannon/Simpson) during composting (A). PCoA (Bray–Curtis) of bacterial community succession during composting (95% C.I. ellipses) (B). Data points are colored and shaped by treatment group: orange circles for CK, light blue squares for L, and dark blue triangles for H. Each subplot corresponds to a sampling time point from D0 to D20. The light-colored shaded areas surrounding each group of points represent 95% confidence ellipses, which visualize the distribution range of samples within each group and highlight the dissimilarities between groups. The weighted abundance β-average nearest classification distance (C). Asterisks indicate statistical significance: * for p < 0.05, ** for p < 0.01, and *** for p < 0.001. Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg).

Figure 5

Phylum-level community dynamics across composting groups (A). Differentially abundant microorganisms at the phylum level (B,C). Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg).

Figure 6

Co-occurrence network linking ten MRGs to dominant host genera (abundance > 0.1%) (correlation > 0.6, p < 0.05). In the network, nodes represent the following: (1) green nodes—MRGs with >90% removal efficiency; (2) orange nodes—MRGs with 30–70% removal efficiency; and (3) red nodes—MRGs showing increased abundance. The edges represent significant co-occurrence relationships between MRGs and their potential microbial hosts. (A) represents the co-occurrence network of the CK group, (B) represents that of the L group, and (C) represents that of the H group.

Figure 7

Random forest search for biomarkers and environmental drivers. In the left panel, color depth (saturation) indicates the relative importance of bacteria in each group (CK, L, H), with darker shades representing higher importance. Blue and red hues denote abundance levels: blue indicates lower abundance or negative correlation, while red indicates higher abundance or positive correlation. Asterisks indicate statistical significance: * for p < 0.05, ** for p < 0.01, and *** for p < 0.001. Groups: CK (control), L (246.49 ± 22.83 mg/kg), H (529.99 ± 16.15 mg/kg). T, temperature. EC, electrical conductivity. OM, organic matter. TN, total nitrogen. ${\mathrm{NO}}_3^{-}$, nitrate nitrogen. ${\mathrm{NH}}_4^{+}$, ammonium nitrogen.