In silico and in vitro comparative analysis of 79 Acinetobacter baumannii clinical isolates

Abstract

Acinetobacter baumannii is a significant nosocomial bacterial pathogen that poses a substantial infection risk due to its high resistance to antibiotics and ability to survive in hospital environments. In this study, we performed comprehensive in silico and in vitro analyses on 79 A. baumannii clinical isolates from different geographical locations to uncover their genomic and epidemiological characteristics as well as their antibiotic and phage susceptibilities. Our findings revealed considerable genomic diversity among the isolates, as shown by average nucleotide identity (ANI) heat maps, multilocus sequence typing (MLST), and core genome MLST (cgMLST). We identified several international clones known for their high antibiotic resistance and global prevalence. Surprisingly, we also observed that the number of antimicrobial resistance genes (ARGs) was higher in isolates containing CRISPR-Cas systems. Plaque assays with 13 phages indicated that Acinetobacter phages have a narrow host range, with capsule loci (KL) serving as a good indicator of phage-bacteria interactions. The presence of CRISPR-Cas systems and other antiviral defense mechanisms in A. baumannii genomes also appears to play a key role in providing phage resistance, regardless of the phage receptors. We also found that spacers associated with subtypes I-F1 and I-F2 CRISPR-Cas systems predominantly target prophages, suggesting a role in maintaining genomic stability and contributing to phage-bacteria co-evolution. Overall, this study provides a set of highly characterized A. baumannii clinical isolates for future studies on antibiotic-phage-bacteria interactions.IMPORTANCEAcinetobacter baumannii poses a significant challenge to the healthcare system due to its antibiotic resistance and strong survival mechanisms. This study examines a diverse collection of 79 clinical isolates to deepen our understanding of A. baumannii's genetic characteristics and its defense mechanisms against both antibiotics and phages. Genomic analysis revealed globally prevalent, highly resistant clones and uncovered a complex role for CRISPR-Cas systems. Although CRISPR-Cas systems were not widespread among these isolates, they primarily targeted prophages. Additionally, the study emphasizes the importance of capsule types as indicators of phage susceptibility. Together, these findings provide insights into the pathogen's resilience and evolutionary adaptations, potentially guiding future research on infection control strategies and new therapeutic approaches to combat A. baumannii infections.

Keywords: Acinetobacter baumannii; CRISPR-Cas; antimicrobial resistance (AMR); bacteriophages; capsule loci (KL); multilocus sequence typing (MLST); phages.

Fig 1

Genomic comparison of 79 A. baumannii isolates. Heat map of the average nucleotide identity (ANI). ANI values were determined using FastANI, and the heatmap was hierarchically clustered using the complete linkage method. Clonal complexes CC1, CC2, and CC15 are indicated on the left. *, isolate that does not belong to CC1.

Fig 2

Minimum spanning tree analysis of 79 A. baumannii isolates. Representation of the genetic relationships among the cgMLST profiles of the 79 A. baumannii isolates. Each circle (node) in the tree corresponds to an individual cgMLST profile, colored according to the sequence type (ST, Pasteur scheme) as indicated in the legend, with the number of isolates shown in brackets. The edges connecting the nodes indicate genetic distances between profiles, with shorter edges representing smaller genetic differences. The numbers on the edges denote the number of allelic differences out of 2,390 alleles. CC1, CC2, and CC15 are circled, each comprising STs that share a single allelic mismatch according to the MLST Pasteur scheme (seven alleles).

Fig 3

Comparative analysis of 79 A. baumannii isolates. (A) The maximum likelihood phylogenetic tree was inferred from the concatenated alignment of a set of 577 core loci. The tree is rooted at the midpoint, and ultra-fast bootstrap support ≥95% is shown as filled circles. The country of origin of each A. baumannii isolate, when available, is shown adjacent to the leaf labels as a three-letter country code: ARG (Argentina), BEL (Belgium), CAN (Canada), CZE (Czech Republic), DEU (Germany), ESP (Spain), FRA (France), GBR (United Kingdom), GEO (Georgia), ITA (Italy), NLD (Netherlands), and USA (United States). Additionally, the designated ST and the presence of subtype I-F1 (purple dot) or subtype I-F2 (fuchsia dot) CRISPR-Cas systems are indicated. (B) Susceptibility of the A. baumannii isolates to 17 antibiotics is shown as a heat map, colored according to the key on the right side of the figure. The bar chart denotes the predicted number of antimicrobial resistance genes (ARGs) from the in silico analysis. (C) The designated KL for each A. baumannii isolate is shown, as well as the sensitivity to 13 phages as a heat map, colored according to the key also on the right. The bar charts denote the predicted number of phage defense systems from the in silico analysis. *, low confidence of prediction; ND, not determined.

Fig 4

Schematic representation of spacers from CRISPR arrays in A. baumannii isolates. Each colored square corresponds to a spacer, ordered from 5′ to 3′ end. Two spacers with the same diamond–square color combination share homologous sequences. The designated ST and KL for each A. baumannii isolate are shown adjacent to their arrays. “*”: Low confidence in prediction. Isolates were grouped by CRISPR-Cas subtype: (A) I-F1, (B) I-F2, and (C) I-F2 containing two CRISPR arrays.

Fig 5

Distribution of ARGs in A. baumannii isolates with or without a subtype I-F CRISPR-Cas system. Violin plot depicting the number of ARGs in isolates with the presence of the subtype I-F CRISPR-Cas system (blue) and without any CRISPR-Cas system (orange). Mann-Whitney U test was performed to determine the P-value of 0.0001 (significant P-value < 0.05).