Transcriptional Modulation of Penicillin-Binding Protein 1b, Outer Membrane Protein P2 and Efflux Pump (AcrAB-TolC) during Heat Stress Is Correlated to Enhanced Bactericidal Action of Imipenem on Non-typeable Haemophilus influenzae

Abstract

Objective: The purpose of the present study was to investigate the penicillin binding proteins (PBPs), drug influx and efflux modulations during heat stress and their effects on the bactericidal action of imipenem on non-typeable Haemophilus influenzae (NTHi). Methods: The two NTHi clinical isolates (GE47 and GE88, imipenem MICs by E-test > 32 μg/mL) examined in this study were collected at Geneva University Hospitals. The imipenem killing activity was assessed after incubation of the NTHi strains at either 37 or 42°C for 3 h with increasing concentrations of imipenem. The detection of PBPs was carried out by Bocillin-FL. Global transcriptional changes were monitored by RNA-seq after pre-incubation of bacterial cells at either 37 or 42°C, and the expression levels of relevant target genes were confirmed by qRT-PCR. Results: Quantitation of NTHi viable cells after incubation with 0.25 μg/mL of imipenem for 3 h revealed more than a twofold decrease in GE47 and GE88 viable cells at 42°C as compared to 37°C. Transcriptome analysis showed that under heat stress conditions, there were 141 differentially expressed genes with a | log2(fold change)| > 1, including 67 up-regulated and 74 down-regulated genes. The expression levels of ponB (encoding PBP1b) and acrR (regulator of AcrAB-TolC efflux pump) were significantly increased at 42°C. In contrast, the transcript levels of ompP2 (encoding the outer membrane protein P2) and acrB gene (encoding AcrB) were significantly lower under heat stress condition. Conclusion: This study shows that the transcriptional modulation of ponB, ompP2, acrR, and acrB in the heat stress response is correlated to enhanced antimicrobial effects of imipenem on non-typeable H. influenzae.

Keywords: AcrAB-TolC; Bocillin-FL; NTHi; OmpP2; PBP1b; RNA-seq; heat stress.

FIGURE 1

Viable NTHi cells after growth at 37°C in the presence of increasing concentrations of imipenem. Data are presented as means ± SD of 8 independent biological replicates.

FIGURE 2

(A) Population analysis profiles of GE47 and GE88 strains as determined by microdilution plate counts. Haemophilus influenzae ATCC 49766 strain was used as a negative control for all experiments. (B) One colony of each of two NTHi strains was picked from HTM agar containing 16 or 32 μg of imipenem/mL and passaged in imipenem free HTM agar. After overnight growth, these cultures were named GE47/32 and GE88/16.

FIGURE 3

Percentage of viable NTHi cells relative to the control condition after exposition to increasing concentration of imipenem during 3 h at either 37 or 42°C. The amounts of GE47 and GE88 viable cells after incubation with increasing concentration of imipenem ranged from 0.25 to 256 μg/mL at either 37 or 42°C were normalized based on their amount of viable cells in a control condition (i.e., growth at either 37 or 42°C with 0 μg/mL of imipenem). No significant difference was observed on viable cells between 37 and 42°C when the bacterial cells were incubated with higher imipenem concentrations (>0.25 μg/mL for GE47 and >2.0 μg/mL for GE88) (Supplementary Figure S4). (A) GE47 strain (imipenem MIC by E-test = > 32 μg/mL); (B) GE88 strain (imipenem MIC by E-test = > 32 μg/mL). Experiments were performed in eight independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01 (paired Student’s t-test).

FIGURE 4

Binding of Bocillin-FL to penicillin binding proteins (PBPs) at two different growth and labeling temperatures (37 or 42°C). (B) Shows a stained image (Coomassie) of the top gel (A), which indicates that for both strains (GE47 and GE88) the same amounts of membrane proteins were loaded in all the lanes.

FIGURE 5

Bocillin-FL binding to PBPs of the reference H. influenzae strain Rd KW20 at two different labeling temperatures (37 or 42°C).

FIGURE 6

(A) Volcano plot reporting the –log10 false discovery rate (FDR) versus the log2 transformed fold change in read counts between 37 and 42°C. Each dot represents a given gene. Fold change is computed by considering the expression level at 42°C compared to 37°C, thereby all genes at the right of 0 of x-values are meant to be up-regulated at 42°C and those at the left of 0 of x-values are down-regulated at 42°C. Gray = genes that do not change in significantly expression; black = genes with expression values associated with FDR-corrected p-values; blue = genes expression values associated with p-value < 0.05 and linear fold change between 37 and 42°C higher than 2. (B–D) Same as (A) except that in (B) genes coding for PBPs are marked in red; in (C) outer membrane protein P2 gene is in violet, (D) efflux pumps (AcrAB-TolC) and the repressor AcrR-coding genes are depicted in green. ponA gene encoding PBP1a; ponB gene encoding PBP1b; pbp2 gene encoding PBP2; ftsI gene encoding PBP3; dacA gene encoding PBP5.

FIGURE 7

qRT-PCR of differentially expressed genes. The mRNA levels of target genes were normalized based on their ribosomal RNA small subunit methyltransferase H (RsmH) transcript level, which were assayed in each round of qRT-PCR. Data were presented as means ± SD of 4 independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001 (paired Student’s t-test).