Cross-protection and cross-feeding between Klebsiella pneumoniae and Acinetobacter baumannii promotes their co-existence

Abstract

Acinetobacter baumannii and Klebsiella pneumoniae are opportunistic pathogens frequently co-isolated from polymicrobial infections. The infections where these pathogens co-exist can be more severe and recalcitrant to therapy than infections caused by either species alone, however there is a lack of knowledge on their potential synergistic interactions. In this study we characterise the genomes of A. baumannii and K. pneumoniae strains co-isolated from a single human lung infection. We examine various aspects of their interactions through transcriptomic, phenomic and phenotypic assays that form a basis for understanding their effects on antimicrobial resistance and virulence during co-infection. Using co-culturing and analyses of secreted metabolites, we discover the ability of K. pneumoniae to cross-feed A. baumannii by-products of sugar fermentation. Minimum inhibitory concentration testing of mono- and co-cultures reveals the ability for A. baumannii to cross-protect K. pneumoniae against the cephalosporin, cefotaxime. Our study demonstrates distinct syntrophic interactions occur between A. baumannii and K. pneumoniae, helping to elucidate the basis for their co-existence in polymicrobial infections.

Fig. 1. Phylogeny and synteny of A. baumannii AB6870155 and K. pneumoniae KP6870155.

Dendrograms representing Euclidean distance matrices from pairwise fastANI distances between genomes of a AB6870155 and other A. baumannii strains and, b KP6870155 and other K. pneumoniae genomes. Panel inserts show mappings of orthologous regions between each strain and its closest relative strain computed by fastANI and plotted with genoPlotR. K. pneumoniae sequence types are in parentheses next to the strain name. Source data are provided as a Source Data file.

Fig. 2. Respiration, growth and gene expression of AB6870155 and KP6870155.

a Distribution of gene expression across major pathways as measured by the log 2-fold change (Log2FC) of gene expression in co-cultures of each strain versus their pure culture counterparts grown in SLMM and harvested for RNA at the 24 h timepoint. b Co-culture AB6870155 + KP6870155 (A + K) and mono-culture growth and respiration of A. baumannii AB6870155 and K. pneumoniae KP6870155 grown in SLMM and c MH media as measured in arbitrary omnilog units (AOU). d Co-culture proportions of AB6870155 and KP6870155 during growth in SLMM (n = 3 independent bacterial cultures), and e MH media (n = 3) as measured by qPCR. Source data are provided as a Source Data file.

Fig. 3. Carbon source utilisation and cross-feeding between A. baumannii AB6870155 and K. pneumoniae KP6870155.

a Carbon-source utilisation activity calculated by AUC of AB6870155 (inner ring) and KP6870155 (outer ring) based on PM01-02 Biolog Phenotype Microarrays. b Venn diagram of C-source utilisation from phenome data showing overlap between AB6870155 and KP6870155. c Schematic of cross-feeding experiment utilising Millicell culture inserts (Merck). d Growth of AB6870155 in Millicell plates with minimal media and select compounds (glycerol, mannose, galactose, serine, maltose, sucrose) provided as the sole carbon source in the presence (denoted as AB + KP) and absence of KP6870155 (denoted as AB) (n = 3 independent bacterial cultures). Data are presented as mean values + /− SD. Source data are provided as a Source Data file.

Fig. 4. Ethanol and lactate production by KP6870155 fed with various carbon sources and KP6870155 spent media utilisation by AB6870155.

HPLC measured ethanol and lactate production by K. pneumoniae KP6870155 (abbreviated as KP) grown with various carbon sources and consumption by A. baumannii AB6870155 (abbreviated as AB) grown in the KP6870155 spent media (n = 3 independent bacterial cultures). a, b glycerol, c, d lactose, e, f mannitol, g, h sucrose. Data are presented as mean values + /− SD. Source data are provided as a Source Data file.

Fig. 5. Biofilm properties of A. baumannii AB6870155 and K. pneumoniae KP6870155 co-cultures.

a RNA-seq expression changes (log2FC) of AB6870155 and KP6870155 co-cultured biofilms relative to their respective mono-cultured biofilms. b Growth (bars) and biofilm formation (points) (measured as OD₅₅₀ after staining biofilms with crystal violet) of various ratios of AB6870155: KP6870155 grown planktonically; biofilm production detected by crystal violet assay (n = 8) and data are presented as mean values + /− SEM. c CLSM images of mono-culture and dual-species biofilms grown in flow-cell chambers for 3 days and stained using BacLight Live/Dead stain (ThermoFisher); green—live cells, red—dead cells. d SEM images of mono-culture and dual-species biofilms grown in microculture plates for 21 h (n = 3 independent bacterial cultures). e Physical appearance of single species or dual-species biofilms grown on coverslips for 21 h. f Mean cell length of pure versus co-culture biofilms grown on coverslips and imaged with SEM (AB + KP = AB6870155 + KP6870155 co-cultures); MicrobeJ software used to calculate cell lengths in SEM images. P-values were calculated using a two-sided Mann–Whitney test where significance values represent: **p ≤ 0.02, non-significant p-value = ns where p > 0.05. Boxes are bound by the first and third quartile with a horizontal line at the median and whiskers represent 1.5x the interquartile range. Dots represent individual cells (n = 172 AB + KP, n = 225 AB6870155, n = 21 KP6870155). Source data are provided as a Source Data file.

Fig. 6. Effects of co-cultures on antibiotic resistance and virulence.

a Antibiotic cross-protection assay in 512 μg/mL cefotaxime. A + K denotes AB6870155 and KP6870155 co-cultures at a 30:70 ratio respectively. Inset of MacConkey agar plate shows KP (KP6870155) colonies appearing red due to K. pneumoniae β-galactosidase activity; AB (AB6870155) lacks this enzyme and appears as white colonies. b Growth of K. pneumoniae KP6870155 in cefotaxime only (+ CEF) and cefotaxime + sulbactam (+ CEF:SUL at 2:1) as measured by optical density (OD₆₀₀). “KP only” represents K. pneumoniae KP6870155 grown in wells without sharing media with A. baumannii AB6870155 while “KP ( + AB insert)” represents KP6870155 physically separated from AB6870155 via a Millicell hanging insert and able to share media (n = 3). Data are presented as mean values + /− SEM. Significance was measured by paired Student’s t-test (two-tailed) where *represents a p < 0.05 (p = 0.03) and ns (not significant) represents a p > 0.05. c Kaplan–Meier curves of single injected AB6870155 (AB) and KP6870155 (KP) and co-injected (A + K) at a 1:1 ratio in G. mellonella (p-value based on log-rank test); PBS only controls had 100% survival (overlap with AB curve). d schematic representation created with BioRender and ChemDraw (v22.0.0) of A. baumannii AB6870155 and K. pneumoniae KP6870155 interactions. Source data are provided as a Source Data file.