Clinically relevant pathogens on surfaces display differences in survival and transcriptomic response in relation to probiotic and traditional cleaning strategies

Abstract

Indoor surfaces are paradoxically presumed to be both colonized by pathogens, necessitating disinfection, and "microbial wastelands." In these resource-poor, dry environments, competition and decay are thought to be important drivers of microbial community composition. However, the relative contributions of these two processes have not been specifically evaluated. To bridge this knowledge gap, we used microcosms to evaluate whether interspecies interactions occur on surfaces. We combined transcriptomics and traditional microbiology techniques to investigate whether competition occurred between two clinically important pathogens, Acinetobacter baumannii and Klebsiella pneumoniae, and a probiotic cleaner containing a consortium of Bacillus species. Probiotic cleaning seeks to take advantage of ecological principles such as competitive exclusion, thus using benign microorganisms to inhibit viable pathogens, but there is limited evidence that competitive exclusion in fact occurs in environments of interest (i.e., indoor surfaces). Our results indicate that competition in this setting has a negligible impact on community composition but may influence the functions expressed by active organisms. Although Bacillus spp. remained viable on surfaces for an extended period of time after application, viable colony forming units (CFUs) of A. baumannii recovered following exposure to a chemical-based detergent with and without Bacillus spp. showed no statistical difference. Similarly, for K. pneumoniae, there were small statistical differences in CFUs between cleaning scenarios with or without Bacillus spp. in the chemical-based detergent. The transcriptome of A. baumannii with and without Bacillus spp. exposure shared a high degree of similarity in overall gene expression, but the transcriptome of K. pneumoniae differed in overall gene expression, including reduced response in genes related to antimicrobial resistance. Together, these results highlight the need to fully understand the underlying biological and ecological mechanisms for community assembly and function on indoor surfaces, as well as having practical implications for cleaning and disinfection strategies for infection prevention.

Fig. 1. Schematic describing the experimental design.

General overview of the biological question (a). Microcosm experimental design (b) with four types of inocula containing pathogen (A. baumannii or K. pneumoniae) and Bacillus spp. and sealed sterile microcosm chamber maintained under ambient temperature and relative humidity (c). The elevated temperature and humidity condition was maintained by adding 10 mL of sterilized miliQ water into the sealed chamber.

Fig. 2. Culture-based survival of pathogens.

Colony forming units (CFU) of pathogens A. baumannii (a) and K. pneumoniae (b) and composition of surface microbial community under four cleaning scenarios and two temperature and humidity conditions. CFUs of pathogens were counted on day 3 of incubation (37 ^oC) on their selective media while total CFUs (pathogen and Bacillus spp.) were obtained from TSA. The proportions of A. baumannii and K. pneumoniae were calculated as the ratio between CFU from their selective media and CFU from TSA. The proportion of Bacillus spp. was calculated as $1-\frac{\mathit{CFU}\left(\mathit{CHROM}\mathit{or}M\mathit{acConkey}\right)}{\mathit{CFU}\left(\mathit{TSA}\right)}$. Dashed lines represent a limit of detection (LOD) of 10 CFU. Error bars represent standard deviations. * Due to possible technical variation, the proportion of A. baumannii exceeded 1 at T24 for A. baumannii Germinated Bacillus scenario.

Fig. 3. Variations in transcriptomes of pathogens on surfaces.

Principal component analysis on variance stabilized A. baumannii (a) and K. pneumoniae (c) read counts and number of differentially expressed genes of 11 pairwise comparisons for A. baumannii (b) and K. pneumoniae (d) experiments. Samples were color-coded based on cleaning conditions; each point (dot or triangle) represents its corresponding temperature and humidity condition.

Fig. 4. Variations in functional pathways in pathogens on surfaces under ambient conditions based on transcriptomics.

KEGG pathway enrichment results for samples collected under ambient temperature and humidity condition (a, d), log₂ fold change of genes associated with antibiotic/antimicrobial resistance (b, e), and sulfur metabolism (c, f), compared to no-cleaning samples (A. baumannii or K. pneumoniae). KEGG pathways shown here were statistically enriched in at least one of the cleaning conditions (Benjamini-Hochberg corrected p value ≤0.05). Normalized enrichment score (NES) accounts for differences in gene set size and were used to compare enrichment results across KEGG pathways; a positive NES suggests an overall positive upregulation of genes belonging to its corresponding pathway^–. *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 5. Variations in functional pathways in pathogens on surfaces under humid conditions based on transcriptomics.

KEGG pathway enrichment results for No Cleaner and Probiotic Cleaner samples collected under elevated temperature and humidity condition (a, d), log₂ fold change of genes associated with A. baumannii cell motility (b), K. pneumoniae biofilm formation (e), and sulfur metabolism (c, f). KEGG pathways shown here were statistically enriched in at least one of the cleaning conditions (Benjamini-Hochberg corrected p value ≤0.05). *p < 0.05, **p < 0.01, ***p < 0.001.