Deciphering Common Genetic Pathways to Antibiotic Resistance in Escherichia coli Using a MEGA-Plate Evolution System

Abstract

Background. Antimicrobial resistance (AMR) poses a significant global health threat, necessitating a deeper understanding of bacterial adaptation mechanisms. Introduction. This study investigates the genotypic and phenotypic evolutionary trajectories of Escherichia coli under meropenem and gentamicin selection, and it benchmarks these findings against florfenicol-evolved strains. Methodology. Utilizing a downsized, three-layer acrylic modified "Microbial Evolution and Growth Arena (MEGA-plate) system"-scaled to 40 × 50 cm for sterile handling and uniform 37 °C incubation-we tracked adaptation over 9-13 days, enabling real-time visualization of movement across antibiotic gradients. Results. Meropenem exposure elicited pronounced genetic heterogeneity and morphological remodeling (filamentous and circular forms), characteristic of SOS-mediated division arrest and DNA-damage response. In contrast, gentamicin exposure produced a uniform resistance gene profile and minimal shape changes, suggesting reliance on conserved defenses without major morphological adaptation. Comprehensive genomic analysis revealed a core resistome of 22 chromosomal loci shared across all three antibiotics, highlighting potential cross-resistance and the central roles of baeR, gadX, and marA in coordinating adaptive responses. Gene ontology enrichment underscored the positive regulation of gene expression and intracellular signaling as key themes in resistance evolution. Discussion. Our findings illustrate the multifaceted strategies E. coli employs-combining metabolic flexibility with sophisticated regulatory networks-to withstand diverse antibiotic pressures. This study underscores the utility of the MEGA-plate system in dissecting spatiotemporal AMR dynamics in a controlled yet ecologically relevant context. Conclusions. The divergent responses to meropenem and gentamicin highlight the complexity of resistance development and reinforce the need for integrated, One Health strategies. Targeting shared regulatory hubs may open new avenues for antimicrobial intervention and help preserve the efficacy of existing drugs.

Keywords: Escherichia coli evolution; MEGA-plate system; One Health approach; antibiotic resistance genes (ARGs); antimicrobial resistance (AMR).

Figure 1

MEGA-plate experiments displaying bacterial colonization and adaptation across gradients of increasing antibiotic concentrations (0× to 1000× the minimum inhibitory concentration [MIC]) for meropenem (A) and gentamicin (B). (A) shows results from a 9-day incubation with meropenem, while (B) shows results from a 13-day incubation with gentamicin. Both images reveal how bacterial populations progressively adapt and spread into higher antibiotic concentrations over time.

Figure 2

(A) A hierarchical clustering heatmap of antibiotic resistance genes (ARGs) detected in eight meropenem-exposed samples (M1–M8). The blue-to-red color scale represents the relative abundance or presence frequency of each ARG, revealing two distinct clusters: samples M2, M4, M6, and M7 share similar ARG profiles, whereas M1, M3, M5, and M8 display greater genetic variability. (B) An association network of the same samples further illustrates these relationships, with the closely connected cluster (M2, M4, M6, and M7) reflecting similar ARG compositions and the more isolated samples (M1, M3, M5, and M8) indicating higher ARG diversity in response to meropenem exposure.

Figure 3

(A) A hierarchical clustering heatmap showing antibiotic resistance genes (ARGs) detected in ten gentamicin-exposed samples (M1–M10). The blue-to-red color scale indicates the relative abundance or presence frequency of each ARG. Samples M1, M2, and M6 form a distinct cluster with identical ARG profiles, while M9 and M10 exhibit another shared cluster. Other samples (M3, M4, M5, M7, and M8) correlate closely with these clusters, indicating a more uniform ARG distribution in response to gentamicin exposure. (B) An association network of the gentamicin-exposed samples further illustrates the relationships among their ARG profiles. M1, M2, and M6 are highly interconnected, reflecting identical ARG profiles, while M9 and M10 are similarly grouped. The remaining samples show a strong correlation with these clusters, confirming a more uniform resistance response across all ten samples.

Figure 4

Gram-stained micrographs (100× magnification) of E. coli samples taken from the meropenem control lane (C1), which contained no antibiotic. Panels (A–C) correspond to samples M1, M3, and M6, respectively, all originating from the same antibiotic-free lane. MER = meropenem, C1 = lane without antibiotic, and M = sample number from that lane.

Figure 5

Gram-stained micrographs (100× magnification) of E. coli samples. (A–D) display samples from meropenem lane 2 (C2) at an MIC of 0.125 μg/mL, with corresponding micrographs of samples M4, M6, M1, and M5, respectively. (E–I) show samples from lanes 3, 4, and 5 (C3–C5), with micrographs of M2 from lane 5 at an MIC of 125 μg/mL, M5 and M6 from lane 3 at an MIC of 1.25 μg/mL, and M3 and M4 from lane 4 at an MIC of 12.5 μg/mL. MER = meropenem, C2–C5 = lane numbers, and M = sample number from the respective lane.

Figure 6

Gram-stained micrographs (100× magnification) of E. coli samples obtained from gentamicin-treated lanes. (A–C) represent samples from lanes 1 (C1), 2 (C2), and 3 (C3): panel (A) corresponds to sample M4 from antibiotic-free lane 1, panel (B) to sample M5 from lane 3 at an MIC of 40 μg/mL, and panel (C) to sample M6 from lane 2 at an MIC of 4 μg/mL. (D–F) show samples from lanes 4 (C4) and 5 (C5): panel (D) represents sample M1 from lane 5 at an MIC of 4000 μg/mL, panel (E) represents sample M3 from lane 4 at an MIC of 400 μg/mL, and panel (F) represents sample M9 from lane 5 at an MIC of 4000 μg/mL. GEN denotes gentamicin, C indicates the lane number, and M refers to the sample number from each lane.

Figure 7

Network representation of biological functions associated with meropenem-related antibiotic resistance genes (ARGs). Each numbered node corresponds to a specific function (listed below), and the edges indicate relationships or co-occurrence between these functions. The clusters of nodes (indicated by distinct colors) represent groups of related processes—regulation of gene expression, biosynthetic activities, and antibiotic transport or response mechanisms—highlighting the complex functional landscape that emerges under meropenem pressure.

Figure 8

Network representation of biological functions associated with gentamicin-related antibiotic resistance genes (ARGs). Each numbered node corresponds to a specific function (listed below), and the edges indicate relationships or co-occurrence among these functions. The clusters of nodes (color-coded) represent interconnected groups of biological processes—metabolic regulation, biosynthetic pathways, and antibiotic transport or response mechanisms—highlighting a tightly coordinated functional network under gentamicin exposure.

Figure 9

Network representation of biological functions associated with florfenicol-related antibiotic resistance genes (ARGs). Each numbered node corresponds to a specific function (listed below), and the edges indicate relationships or co-occurrence among these functions. The large, densely connected cluster represents a complex network of metabolic and biosynthetic processes, while the smaller groups highlight additional functions such as protein modifications, signal transduction pathways, and antibiotic transport or response. Together, these clusters illustrate the intricate functional landscape influenced by florfenicol exposure.

Figure 10

Network visualization of distinct biological functions associated with antibiotic resistance genes (ARGs) identified under exposure to meropenem, gentamicin, and florfenicol. Each numbered node represents a unique biological function (listed below), and the edges indicate functional relationships or co-occurrences. The densely interconnected cluster, along with the multiple smaller groups, highlights a wide range of metabolic, biosynthetic, regulatory, and transport-related processes that collectively shape the complex resistance landscape across these antibiotics.