L-Arabinose Alters the E. coli Transcriptome to Favor Biofilm Growth and Enhances Survival During Fluoroquinolone Stress

Abstract

Environmental conditions, including nutrient composition and temperature, influence biofilm formation and antibiotic resistance in Escherichia coli. Understanding how specific metabolites modulate these processes is critical for improving antimicrobial strategies. Here, we investigated the growth and composition of Escherichia coli in both planktonic and biofilm states in the presence of L-arabinose, with and without exposure to the fluoroquinolone antibiotic levofloxacin, at two temperatures: 28 and 37 °C. At both temperatures, L-arabinose increased the growth rate of planktonic E. coli but resulted in reduced total growth; concurrently, it enhanced biofilm growth at 37 °C. L-arabinose reduced the efficacy of levofloxacin and promoted growth in sub-minimum inhibitory concentrations (25 ng/mL). Transcriptomic analyses provided insight into the molecular basis of arabinose-mediated reduced susceptibility of E. coli to levofloxacin. We found that L-arabinose had a temperature- and state-dependent impact on the transcriptome. Using gene ontology overrepresentation analyses, we found that L-arabinose modulated the expression of many critical antibiotic resistance genes, including efflux pumps (ydeA, mdtH, mdtM), transporters (proVWX), and biofilm-related genes for external structures like pili (fimA) and curli (csgA, csgB). This study demonstrates a previously uncharacterized role for L-arabinose in modulating antibiotic resistance and biofilm-associated gene expression in E. coli and provides a foundation for additional exploration of sugar-mediated antibiotic sensitivity in bacterial biofilms.

Keywords: E. coli; biofilm; fluoroquinolone; transcriptomics.

Figure 1

E. coli growth with arabinose and levofloxacin. Swimming cell growth was measured at (A) 28 °C and (B) 37 °C at 600 nm over 15 h. N = 3. (C) Biofilm was grown on glass wool over the course of 48 h and stained using crystal violet. Significance was determined using a two-way ANOVA with post hoc Tukey test and reported as ** p < 0.01 and * p < 0.05 compared to 0% sugar added.

Figure 2

Biofilm colony growth and morphology. (A) Representative images of biofilm colonies grown for 48 h at varied experimental conditions. Cartoon depicting our definition of the center of the biofilm colony versus the outer rim. (B) Total area measurements and (C) outer rim measurements of the biofilm colonies. Note that no outer rim could be determined for levofloxacin biofilm colonies at 28 °C. Significance was determined using a two-way ANOVA with a post hoc Tukey test.

Figure 3

Biofilm growth on agar. (A) Wet mass of biofilm formed after 48 h growth at indicated temperature. (B) Protein concentration of the EPS determined by bicinchoninic acid assay and normalized to wet biofilm mass. (C) Carbohydrate concentration of the EPS determined by phenol-sulfuric acid assay and normalized to wet biofilm mass. Statistical significance was determined by a two-way ANOVA with post hoc Tukey test.

Figure 4

RNA-seq transcriptomics. (A) Principal component analysis of eight experimental conditions with three biological replicates each (24 samples). (B) Venn diagram highlighting overlap of significantly differentially expressed genes (FC > 4, false discovery rate (FDR) < 0.01) for different conditions, scaled to overlap size.

Figure 5

Scatter plots of log₂ (Fold Change) values of DEGs for planktonic versus biofilm conditions in the presence of 0.5% (w/w) arabinose compared to control at (A) 37 °C and (B) 28 °C. Each point represents a DEG (FC > 4 and FDR < 0.01), notable genes are indicated by name.

Figure 6

Heat maps of log₂ (Fold Change) values for genes found in the (A) pilus (GO:0009289) and (B) antibiotic response (GO:0044667) gene ontology biological process categories. Hashed and faded cells represent genes that did not have significance as determined by our DESeq2 differential expression analysis cutoffs of log₂(FC) > 2 and p < 0.01.