Dual-species proteomics and targeted intervention of animal-pathogen interactions

Abstract

Introduction: Host-microbe interactions are important to human health and ecosystems globally, so elucidating the complex host-microbe interactions and associated protein expressions drives the need to develop sensitive and accurate biochemical techniques. Current proteomics techniques reveal information from the point of view of either the host or microbe, but do not provide data on the corresponding partner. Moreover, it remains challenging to simultaneously study host-microbe proteomes that reflect the direct competition between host and microbe. This raises the need to develop a dual-species proteomics method for host-microbe interactions.

Objectives: We aim to establish a forward + reverse Stable Isotope Labeling with Amino acids in Cell culture (SILAC) proteomics approach to simultaneously label and quantify newly-expressed proteins of host and microbe without physical isolation, for investigating mechanisms in direct host-microbe interactions.

Methods: Using Caenorhabditis elegans-Pseudomonas aeruginosa infection model as proof-of-concept, we employed SILAC proteomics and molecular pathway analysis to characterize the differentially-expressed microbial and host proteins. We then used molecular docking and chemical characterization to identify chemical inhibitors that intercept host-microbe interactions and eliminate microbial infection.

Results: Based on our proteomics results, we studied the iron competition between pathogen iron scavenger and host iron uptake protein, where P. aeruginosa upregulated pyoverdine synthesis protein (PvdA) (fold-change of 5.2313) and secreted pyoverdine, and C. elegans expressed ferritin (FTN-2) (fold-change of 3.4057). Targeted intervention of iron competition was achieved using Galangin, a ginger-derived phytochemical that inhibited pyoverdine production and biofilm formation in P. aeruginosa. The Galangin-ciprofloxacin combinatorial therapy could eliminate P. aeruginosa biofilms in a fish wound infection model, and enabled animal survival.

Conclusion: Our work provides a novel SILAC-based proteomics method that can simultaneously evaluate host and microbe proteomes, with future applications in higher host organisms and other microbial species. It also provides insights into the mechanisms dictating host-microbe interactions, offering novel strategies for anti-infective therapy.

Keywords: Caenorhabditis elegans; Host-microbe interactions; Iron competition; Pseudomonas aeruginosa; SILAC proteomics.

Graphical abstract

Fig. 1

Workflow of SILAC proteome analysis of Caenorhabditis elegans-Pseudomonas aeruginosa host-pathogen interaction model. (A) P. aeruginosa population numbers in intestine at 6 h.p.i and 24 h.p.i. (B) Fixed C. elegans with gfp-tagged PAO1 in intestine at 6 h.p.i and 24 h.p.i. (C) The workflow and experimental design of this project, including the P. aeruginosa – C. elegans infection model and the SILAC proteomics assay. ΔlysA mutant labeled with ¹³C- _L-lysine was cultivated on agars which contained ¹³C- _L-lysine for two days to form biofilm. The samples were then collected for the SILAC proteomics assay. H.p.i = hours post-infection. Scale bar: 100 μm. ***P < 0.001.

Fig. 2

Proteomic profiling of proteins in Pseudomonas aeruginosa and Caenorhabditis elegans during interactions. (A) The PCA score plots showed that the proteins of P. aeruginosa samples from 6 h.p.i and 24 h.p.i groups were clustered separately. (B) Volcano plot representation of differential expression analysis of proteins in P. aeruginosa samples from 6 h.p.i and 24 h.p.i groups. (C) The PCA score plots showed that the proteins of C. elegans samples from 6 h.p.i and 24 h.p.i groups were clustered separately. (D) Volcano plot representation of differential expression analysis of proteins in C. elegans from 6 h.p.i and 24 h.p.i groups. (E) Functional groups of upregulated proteins of P. aeruginosa from 6 h.p.i to 24 h.p.i. The Motility & Attachment took a large proportion in upregulated groups. (F) Significantly enriched KEGG pathways of the upregulated proteins from C. elegans.

Fig. 3

Pseudomonas aeruginosa upregulates pyoverdine expression via repression of AlgR and enhancing of PvdA in intestinal infection of Caenorhabditis elegans. (A) Representative fluorescent images of the fixed nematode’s intestine after the infection of PAO1/p_pvdA-gfp, ΔpvdA/p_pvdA-gfp and ΔalgR/p_pvdA-gfp from 6th to 24th hour. GFP is green, and Pyoverdine channel is blue. Scale bars, 50 μm. (B) Relative GFP fluorescence density and pyoverdine density from wild-type (N2) nematodes infected by PAO1/p_pvdA-gfp, ΔpvdA/p_pvdA-gfp and ΔalgR/p_pvdA-gfp from 6th to 24th hour. (C) Nematodes were cultivated on different bacterial mutants’ lawn for 5 days. The survival rate of nematodes was recorded. H.p.i = hours post-infection. Scale bar: 100 μm. ***P < 0.001.

Fig. 4

Role of FTN-2 by Caenohabditis elegans in iron competition during the bacterial infection interaction compared to wild type nematodes. (A) Representative fluorescent images of the fixed wild type (N2) and ftn-2 (ok404) nematodes intestine after the infection of PAO1/Tn7-gfp and PAO1/p_pvdA-gfp from two time points (6 h.p.i and 24 h.p.i). (B) Relative GFP density and Pyoverdine density of the nematodes intestinal PAO1/p_pvdA-gfp biosensor between wild type nematodes and ftn-2 (ok404) mutant nematodes over time. Both GFP promoted by PvdA and pyoverdine of P. aeruginosa had high expression in wild-type C. elegans intestine than ftn-2 (ok404) C. elegans. (C) The survivability of wild-type C. elegans compared to ftn-2 (ok404) C. elegans infected by PAO1 for 5 days. Compared to the wild type nematodes, ftn-2 (ok404) nematodes had less competition with bacteria. H.p.i = hours post-infection. Scale bar: 100 μm. ***P < 0.001.

Fig. 5

Galangin serves as PvdA inhibitor that abrogates Pseudomonas aeruginosa intestinal infection. (A) Molecular docking revealed Galangin (blue) and ONH (pink) binding to amino acid residues in active sites of PvdA. (B) Galangin inhibits in vitro pvdA gene expression by testing GFP level of PAO1/p_pvdA-gfp. (C) Galangin inhibits in vitro pyoverdine production by testing fluorescent density. (D) The half-maximal inhibitory concentration (IC₅₀) graph of Galangin to pvdA expression and pyoverdine production inhibition. (E) Representative fluorescent images of fixed C. elegans intestine with Galangin and Ciprofloxacin treatment. (F) Relative pyoverdine density from wild type nematodes infected by PAO1 and then with the combinatorial treatment of Galangin and Ciprofloxacin. This combinatorial treatment can help to eliminate the intestinal P. aeruginosa infection in C. elegans in 156 h with several times of drug administration. (G) P. aeruginosa CFU in nematodes intestine after Galangin and/or Ciprofloxacin treatment for 5 days. Bacteria numbers show an alleviation of biofilm infection after combinatorial treatment. (H) Survival rate of wild type C. elegans with Galangin and Ciprofloxacin treatment. This combinatorial treatment could increase survival rate of host. Scale bar: 100 μm. ***P < 0.001, **P < 0.01, *P < 0.05.

Fig. 6

Combinatorial treatment of ciprofloxacin and Galangin eliminates the wound infection caused by Pseudomonas aeruginosa on medaka fish tail. (A) The establishment of medaka fish – P. aeruginosa infection model for testing novel antimicrobial strategies in vivo. (B) The bacterial populations in the wound infection after treatment by monotherapy and combinatorial therapy of Galangin and ciprofloxacin. (C) Relative GFP fluorescence expression level of PAO1/p_pvdA-gfp on the wound infection after treatment by monotherapy and combinatorial therapy of Galangin and ciprofloxacin. (D) Pyoverdine expression level of in wound biofilm infection after monotherapy and combinatorial therapy of Galangin and ciprofloxacin. (E) Representative fluorescent images of wound bacterial (PAO1/p_pvdA-gfp) biofilms infection on medaka fish after treatment. GFP: green, and pyoverdine: blue. Scale bar: 100 μm. The mean and ± sd. from three experiments are shown. ***P < 0.001, **P < 0.01, *P < 0.05.