Phage-based biocontrol of Porphyromonas gingivalis through indirect targeting

Abstract

Bacteriophages offer an opportunity for chemical-free, precise control of problematic bacteria, but this approach can be limited when lytic phages are difficult to obtain for the target host. In such cases, phage-based targeting of cooperating or cross-feeding bacteria (e.g., Streptococcus gordonii) can be an effective approach to control the problematic bacteria (e.g., Porphyromonas gingivalis). Using a dual-species biofilm system, phage predation of S. gordonii (10⁸ PFU·mL^-1) decreased the abundance of pathogenic P. gingivalis by >99% compared with no-treatment controls, while also inhibiting the production of cytotoxic metabolic end products (butyric and propionic acids). Phage treatment upregulated genes associated with interspecies co-adhesion (5- to 8-fold) and quorum sensing (10-fold) in residual P. gingivalis, which is conducive to increased potential to bind to S. gordonii. Counterintuitively, lower-titer phage applications (10⁴ PFU·mL^-1) increased the production of extracellular polymeric substance (EPS) by 22% and biofilm biomass by 50%. This overproduction of EPS may contribute to the phenomenon where the biofilm separated into two distinct species layers, as observed by confocal laser scanning microscopy. Although more complex mixed-culture systems should be considered to delineate the merits and limitations of this novel biocontrol approach (which would likely require the use of phage cocktails), our results offer proof of concept that indirect phage-based targeting can expand the applicability of phage-based control of pathogenic bacteria for public health protection.

Importance: Lytic phages are valuable agents for targeted elimination of bacteria in diverse applications. Nevertheless, lytic phages are difficult to isolate for some target pathogens. We offer proof of concept that this limitation may be overcome via indirect phage targeting, which involves knocking out species that interact closely with and benefit the primary problematic target bacteria. Our target (P. gingivalis) only forms a periodontal pathogenic biofilm if the pioneer colonizer (S. gordonii) offers its surface for P. gingivalis to attach. Phage predation of the co-adhesive S. gordonii significantly reduced abundance of the target pathogen by >99%, decreased the total biofilm biomass by >44%, and suppressed its production of cytotoxic metabolic byproducts. Thus, this research extends the scope of phage-based biocontrol for public health protection.

Keywords: biofilm; indirect targeting; pathogen control; phage therapy; virulence.

Fig 1

Biofilm biomass of dual-species system treated with different levels of lytic phage ΦSG005. S. gordonii was inoculated in plate wells on the first day to form the primary biofilm, P. gingivalis was inoculated on the next day to form the dual-species biofilm system, and phages ΦSG005 with various titers were treated on the third day. Crystal violet optical assay was used to assess biomass changes after 24 h of phage treatment. A phage treatment with at least 10⁸ PFU·mL⁻¹ (MOI≈1) was needed for controlling biofilm biomass in the dual-species system. Error bars indicate ±1 standard deviation from the mean of the triplicate independent experiments. Individual data points are given in the supplemental material.

Fig 2

Absolute abundance of P. gingivalis (A) and S. gordonii (B) treated with different levels of lytic phage ΦSG005. The abundance of both species remaining in the dual-species biofilm after 24 h of phage treatment was quantified by qPCR. Phage treatment with at least 10⁶ PFU·mL⁻¹ (MOI≈0.01) significantly decreased the total abundance of both species, demonstrating that indirect targeting may be a feasible strategy to control pathogens for which lytic phages are difficult to obtain. Error bars indicate ±1 standard deviation from the mean of the triplicate independent experiments. Individual data points are given in the supplemental material.

Fig 3

Confocal laser scanning microscopy analysis of the distribution of S. gordonii and P. gingivalis and the structure of the biofilm in the dual-species system after various phage treatments. Higher-titer phage treatment decreased the abundance of P. gingivalis cells (stained green with carboxyfluorescein diacetate succinimidyl ester), whereas lower-titer treatment counterintuitively promoted the production of biofilm biomass of S. gordonii cells (stained red with hexidium iodide). Top view of the biofilm under different phage titers (A) shows that the abundance of P. gingivalis cells decreased as phage titers increased. Profile view of the biofilm under varying phage titers (B) illustrates that lower levels of phage titers promote the production of biofilm biomass by S. gordonii, thereby influencing the overall biofilm structure. With the use of low phage titers (10² and 10⁴ PFU·mL⁻¹), the biofilm separated into two layers due to the overproduction of biofilm biomass by the S. gordonii layer, resulting in the upward movement of the P. gingivalis layer.

Fig 4

Expression of genes related to interspecies co-adhesion and quorum sensing increased with the phage titer used. The quantification of each gene is expressed as the fold change (fold changes are Log₂ transformed) relative to unexposed controls. Error bars indicate ±1 standard deviation from the mean of the triplicate independent experiments. Individual data points are given in the supplemental material.

Fig 5

Effect of phage treatment (10⁸ PFU·mL⁻¹) on the production of short-chain fatty acids (propionic acid and butyric acid). Phage treatment was applied after forming dual-species system, 10% of liquid culture was extracted every 2 h after adding phages to analyze acid production using GC-MS, fresh culture was replenished at the same time. Mitigation of propionic acid (A) and butyric acid (B) accumulation by phage treatment was observed compared with the control group. These cytotoxic volatile fatty acids are produced by P. gingivalis. Error bars indicate ±1 standard deviation from the mean of the triplicate independent experiments. Individual data points are given in the supplemental material.

Fig 6

Concept summary of indirect phage treatment using different ΦSG005 titters. In controls without phage addition, P. gingivalis (green) forms a pathogenic biofilm after the pioneer colonizer S. gordonii (red) offers its surface for P. gingivalis to attach. Low phage titers counterintuitively promote EPS production by host S. gordonii, displacing upwards the biofilm layer of P. gingivalis and forming double-layer structure. High phage concentrations significantly decrease the abundance of nonhost target P. gingivalis and suppress its production of cytotoxic metabolic byproducts.