From Farm to Fork: Streptococcus suis as a Model for the Development of Novel Phage-Based Biocontrol Agents

Abstract

Bacterial infections of livestock threaten the sustainability of agriculture and public health through production losses and contamination of food products. While prophylactic and therapeutic application of antibiotics has been successful in managing such infections, the evolution and spread of antibiotic-resistant strains along the food chain and in the environment necessitates the development of alternative or adjunct preventive and/or therapeutic strategies. Additionally, the growing consumer preference for "greener" antibiotic-free food products has reinforced the need for novel and safer approaches to controlling bacterial infections. The use of bacteriophages (phages), which can target and kill bacteria, are increasingly considered as a suitable measure to reduce bacterial infections and contamination in the food industry. This review primarily elaborates on the recent veterinary applications of phages and discusses their merits and limitations. Furthermore, using Streptococcus suis as a model, we describe the prevalence of prophages and the anti-viral defence arsenal in the genome of the pathogen as a means to define the genetic building blocks that are available for the (synthetic) development of phage-based treatments. The data and approach described herein may provide a framework for the development of therapeutics against an array of bacterial pathogens.

Keywords: Streptococcus suis; anti-viral defence; food; phage; prophages; zoonosis.

Figure 1

Schematic representation of (A) conventional preclinical phage characterisation and (B) novel approaches to phage characterisation.

Figure 2

Analysis of anti-viral defence systems detected among 133 publicly available S. suis genomes using PADLOC. (A) Abundance of defence systems calculated as the percentage of genomes (out of the total 133) that encode a specific defence system. (B) Total number of anti-viral systems per genome. (C) Categories of anti-viral systems based on molecular mechanism.

Figure 3

Prevalence of prophages in S. suis genomes. Predicted regions were categorised into full-length (42.9%), putative full-length (54.1%), and incomplete prophages (84.2%). Of the 133 genomes, 84.2% are polylysogens—strains harbouring more than one phage region.

Figure 4

Whole proteomic heatmap of the 71 full-length S. suis prophages. Nine clusters (C1–C9) and four singletons were identified. Members within each cluster share ≥ 50% (≥0.5) proteomic identity. Scale bar on the right represents sequence identity. Blue (1.0) represents highest identity, and white (0) represents no proteomic pairwise identity.

Figure 5

Phylogenetic tree of S. suis (pro)phages. Tree was constructed from DNA sequence of the 71 full-length prophages with phage SMP as root. Computations of distance matrix and proteomic tree generation were performed in ViPtree. The final tree was visualised using iTOL. Clades are shown in different colours.

Figure 6

Schematic representation of (pro)phage CDS similarity and organisation. Representative full-length (pro)phage genomes from each clade (of similar genome size) were selected and aligned. Horizontal colour-coded arrows indicate gene function (key on the bottom). Coloured diagonal and vertical lines (alignment) represent percent identity of genes. Dot plot of pairwise alignment indicated on the far left. Colour scale for alignments and dot plots presented in the top left.

Figure 7

Association between number of prophages and defence systems or host genome size. (A) Average number of defence systems correlated with number of prophages identified in a genome (Spearman’s r = 0.40, p < 0.0001). (B) Average number of prophages per host genome size (Spearman’s r = 0.56, p < 0.0001).