Identification and characterisation of two functional antibiotic MATE efflux pumps in the archaeon Halorubrum amylolyticum

Abstract

Multidrug efflux pumps have been found to play a crucial role in drug resistance in bacteria and eukaryotes. In this study, we investigated the presence of functional multidrug and toxic compound extrusion (MATE) efflux pumps, inferred from whole genome sequencing, in the halophilic archaeon Halorubrum amylolyticum CSM52 using Hoechst 33342 dye accumulation and antimicrobial sensitivity tests in the presence and absence of efflux pump inhibitors (EPIs). The whole genome sequence of H. amylolyticum CSM52 contained two putative MATE-type efflux pump genes, which may contribute to the inherent resistance to conventional antimicrobial agents reported in archaea. Antimicrobial susceptibility of the wild-type H. amylolyticum CSM52 testing revealed a lack of sensitivity to a wide range of antimicrobials, including glycopeptides, aminoglycosides, macrolides, fluoroquinolones, tetracycline, and chloramphenicol. However, the presence of EPIs, such as thioridazine, fluoxetine, and chlorpromazine, significantly increased the susceptibility of H. amylolyticum CSM52 to a number of these antimicrobials, indicating the potential involvement of efflux pumps in the observed resistance. A molecular modelling study with EPIs and substrate antimicrobials provided important insights into the molecular interactions with the putative transporter. It suggests that the occupancy of the transporter channel by EPIs has the potential to impact the efflux of antimicrobials. Phylogenetic analysis of the amino acid sequences of both MATE pumps showed low similarity with bacterial representatives, suggesting the presence of novel and distinct MATE efflux pumps in archaea. Our findings provide the first experimental evidence of active antibiotic efflux mechanisms in archaea and their potential roles in antimicrobial resistance, broadening our understanding of mechanisms of archaeal antimicrobial resistance, an overlooked aspect of AMR research.

Fig. 1. Sequence alignment of MATE-996 and MATE-2898 proteins.

The alignment was performed using the ClustalW algorithm and visually represented using Boxshade. Identical residues between the two sequences are highlighted in black, and differences in the sequences are shown in white. The overall sequence identity and similarity provide insights into the evolutionary relationship and potential functional similarities between the two MATE proteins.

Fig. 2. Variation in Hoechst dye accumulation in H. amylolyticum CSM52 in the presence of ½ MIC and ¼ MIC of three EPIs thioridazine, chlorpromazine and fluoxetine, over 3 h incubation time.

Results are based on eight replicates (n = 8).

Fig. 3. Increase in Hoechst fluorescence in H. amylolyticum CSM52 in the presence of ½ MIC and ¼ MIC of thioridazine, chlorpromazine and fluoxetine.

Each time point (a 12 min, b 60 min, c 120 min, d 174 min) contrasts fluorescence in the presence and absence of EPIs. Error bars indicate the mean ± standard deviation determined from replicates (n = 8). Asterisks denote significant differences between the relevant EPI compared to control, with **** p < 0.0001, *** p < 0.001, and ** p < 0.01.

Fig. 4. Quantification of Hoechst dye efflux from E. coli clones harbouring MATE-996 or MATE-2898 efflux genes.

Variation in Hoechst dye accumulation in E. coli clones harbouring either MATE-996 or MATE-2898 genes compared to E. coli BL21DE3 (control) (a). The effect of EPIs on Hoechst dye accumulation in E. coli cells expressing MATE-996, treated with thioridazine, fluoxetine, and chlorpromazine at 12, 60, 120, and 174 min (b). Hoechst dye accumulation in E. coli cells expressing MATE-2898, treated with thioridazine, fluoxetine, and chlorpromazine at 12, 60, 120, and 174 min (c). (n = 3). Asterisks indicate statistical significance compared to the control cells, with **** p < 0.0001, *** p < 0.001, and ** p < 0.01.

Fig. 5. Interactions of the EPIs, fluoxetine, thioridazine and chlorpromazine and the substrate antibiotics rifampicin, chloramphenicol, and tetracycline within the channel region of MATE996 transporter.

3D structure of MATE996 with EPIs (a) and substrate antibiotics (b). Binding sites and key interactions between MATE996 and EPIs (a i–iii) and substrate antibiotics (b i–iii).

Fig. 5. Interactions of the EPIs, fluoxetine, thioridazine and chlorpromazine and the substrate antibiotics rifampicin, chloramphenicol, and tetracycline within the channel region of MATE996 transporter.

3D structure of MATE996 with EPIs (a) and substrate antibiotics (b). Binding sites and key interactions between MATE996 and EPIs (a i–iii) and substrate antibiotics (b i–iii).

Fig. 6. Interactions of the EPIs, fluoxetine, thioridazine and chlorpromazine and the substrate antibiotics rifampicin, chloramphenicol, and tetracycline within the channel region of MATE2898 transporter.

3D structure of MATE-2898 with EPIs (a) and substrate antibiotics (b). Binding sites and key interactions between MATE-2898 and EPIs (a i–iii) and substrate antibiotics (b i–iii).

Fig. 6. Interactions of the EPIs, fluoxetine, thioridazine and chlorpromazine and the substrate antibiotics rifampicin, chloramphenicol, and tetracycline within the channel region of MATE2898 transporter.

3D structure of MATE-2898 with EPIs (a) and substrate antibiotics (b). Binding sites and key interactions between MATE-2898 and EPIs (a i–iii) and substrate antibiotics (b i–iii).

Fig. 7

Molecular phylogenetic analysis based on the amino acid sequences of the MATE proteins from of H. amylolyticum CSM52 (in red) in comparison with MATE protein sequence from various bacterial species. The evolutionary history was inferred by using the Maximum Likelihood method based on the JTT matrix-based model. The tree with the highest log likelihood (−49977.93) is shown. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 83 amino acid sequences. There was a total of 310 positions in the final dataset. Evolutionary analyses were conducted in FastTree (v5).