Bacterial Concentrations and Water Turbulence Influence the Importance of Conjugation Versus Phage-Mediated Antibiotic Resistance Gene Transfer in Suspended Growth Systems

Abstract

Despite the abundance of phage-borne antibiotic resistance genes (ARGs) in the environment, the frequency of ARG propagation via phage-mediated transduction (relative to via conjugation) is poorly understood. We investigated the influence of bacterial concentration and water turbulence level [quantified as Reynold's number (Re)] in suspended growth systems on the frequency of ARG transfer by two mechanisms: delivery by a lysogenic phage (phage λ carrying gentamycin-resistance gene, genR) and conjugation mediated by the self-transmissible plasmid RP4. Using Escherichia coli (E. coli) as the recipient, phage delivery had a comparable frequency (1.2 ± 0.9 × 10^-6) to that of conjugation (1.1 ± 0.9 × 10^-6) in suspensions with low cell concentration (10⁴ CFU/mL) and moderate turbulence (Re = 5 × 10⁴). Turbulence affected cell (or phage)-to-cell contact rates and detachment (due to shear force), and thus, it affected the relative importance of conjugation versus phage delivery. At 10⁷ CFU/mL, no significant difference was observed between the frequencies of ARG transfer by the two mechanisms under quiescent water conditions (2.8 ± 0.3 × 10^-5 for conjugation vs 2.2 ± 0.5 × 10^-5 for phage delivery, p = 0.19) or when Re reached 5 × 10⁵ (3.4 ± 1.5 × 10^-5 for conjugation vs 2.9 ± 1.0 × 10^-5 for phage delivery, p = 0.52). Transcriptomic analysis of genes related to conjugation and phage delivery and simulation of cell (or phage)-to-cell collisions at different Re values corroborate that the importance of phage delivery relative to conjugation increases under either quiescent or turbulent conditions. This finding challenges the prevailing view that conjugation is the dominant ARG transfer mechanism and underscores the need to consider and mitigate potential ARG dissemination via transduction.

Figure 1

Publication trends (a) and the reported ARG transfer frequencies (b) of plasmid-mediated conjugation and phage-mediated transduction. (a) Publications (https://pubmed.ncbi.nlm.nih.gov/) from 1981 to 2020 on conjugation and transduction-associated antibiotic resistance both show an exponential increase, with the number of papers related to conjugation 3–5 times higher than that of transduction. (b) Reported frequencies of ARG transfer via conjugation and transduction both showed high variability. Average frequencies are indicated by “×”.

Figure 2

Lower frequency but higher relative importance of genR transfer via phage delivery (relative to conjugation) occurs at lower bacterial concentration. (a) Transfer frequency was determined after 1 h of conjugation and phage delivery assays in minimal medium with 3.0 g/L glucose and shaken at an Re of 5 × 10⁴. The genR donor (E. coli DH5α harboring plasmid RP4 for conjugation assays or λ phage for phage delivery assays) and recipient (E. coli MG1655) were mixed at a 1:1 ratio. Frequency determined by the plate counting method is presented as bars with the detection limit (10^–6) indicated by the horizontal dashed line; orange dots represent the frequency determined by qPCR analysis. Asterisks (*) indicate significant differences (p < 0.05) between conjugation and phage delivery frequency based on Student’s t-test. “N.D.” refers to “not detected”. (b) Upregulation of conjugation-related genes (traI, traJ, and trbAp) and phage delivery-related genes (lamB and malT) at lower bacterial concentration (twofold gene expression change). The top X-axis depicts decreasing bacterial concentration (CFU/mL). The expression level at a bacterial concentration of 10⁸ CFU/mL was used as the control. Error bars depict ± one standard deviation from the mean of at least three independent replicates.

Figure 3

Turbulence (represented by Reynolds number) significantly impacts the relative frequency of conjugation vs phage delivery for the horizontal transfer of genR in a bell-shaped fashion. The transfer frequency was determined after 1 h of conjugation and phage delivery assays in minimal medium containing 3.0 g/L glucose with bacterial concentration at 10⁷ CFU/mL. The genR donor (E. coli DH5α harboring plasmid RP4 for conjugation assays or λ phage for phage delivery assays) and recipient (E. coli MG1655) were mixed at a 1:1 ratio. Asterisks (*) indicate significant differences (p < 0.05) between conjugation and phage delivery frequency based on Student’s t-test. Error bars depict ± one standard deviation from the mean of at least three independent replicates.

Figure 4

Higher water turbulence increases the theoretical cell (or phage)-to-cell collision frequency and interferes with the stability of ARG transfer channels. (a) Collision frequency (total collision events/s) under static (Re = 0) and dynamic conditions (Re > 0) was simulated based on the Brownian motion (eq 3) and turbulence shear (eq 4), respectively. Brownian motion was not considered in the simulation under turbulent conditions because its induced collision has a frequency that is over 3 orders of magnitude lower than that of shear-induced collision (eq S7). (b) Shear-induced detachment frequency would be higher for conjugating cells than for attached phages due to postulated weaker links formed by thinner and longer multiprotein, hair-like links.^,−