Jumbo phage killer immune system targets early infection of nucleus-forming phages

Abstract

Jumbo bacteriophages of the ϕKZ-like family assemble a lipid-based early phage infection (EPI) vesicle and a proteinaceous nucleus-like structure during infection. These structures protect the phage from nucleases and may create selective pressure for immunity mechanisms targeting this specific phage family. Here, we identify "jumbo phage killer" (Juk), a two-component immune system that terminates infection of ϕKZ-like phages, suppressing the expression of early phage genes and preventing phage DNA replication and phage nucleus assembly while saving the cell. JukA (formerly YaaW) rapidly senses the EPI vesicle by binding to an early-expressed phage protein, gp241, and then directly recruits JukB. The JukB effector structurally resembles a pore-forming toxin and destabilizes the EPI vesicle. Functional anti-ϕKZ JukA homologs are found across bacterial phyla, associated with diverse effectors. These findings reveal a widespread defense system that specifically targets early events executed by ϕKZ-like jumbo phages prior to phage nucleus assembly.

Keywords: anti-phage defense; jumbo phage; phage nucleus; phage vesicle; ϕKZ.

Figure 1:. Discovery of the jumbophage killer (Juk) system.

(A) Bacterial growth curves (OD600) of Pseudomonas aeruginosa isolate PAO1 or PA14 across a range of multiplicities of infection (MOI) of ⏀KZ. Error bars in (A) represent one standard deviation calculated from three technical replicates. (B) Spot titration of ⏀KZ input and output (10-fold serial dilutions) from PAO1 or PA14 infection at MOI 0.01 or MOI 10 (from A) on a lawn of sensitive bacteria PAO1. (C) Cartoon schematic of how the PA14 transposon (Tn) mutant library was constructed and used to identify ⏀KZ sensitive or resistant mutants. Each dot represents the Tn disruption of a unique gene. The fitness and read depth of Tn mutants are shown after being exposed to ⏀KZ. Genes of interest are highlighted. Fitness (s) is calculated as ln(‘mutant_frequency_with_⏀KZ_infection’ / ‘mutant_frequency_without_⏀KZ_infection’). (D, E) Growth curves measuring OD600 during ⏀KZ infection in (D) PA14 and indicated mutants or (E) PAO1 and strains heterologously expressing jukA and jukB via either plasmid (+pjukAB) or chromosome integration (::jukAB), both from jukAB native promoter. Error bars in (D) and (E) represent one standard deviation calculated from two technical replicates. (F) Spot titration (10-fold serial dilutions) of indicated phage on indicated bacterial lawns. See also Fig. S1, S2.

Figure 2:. Juk immune system acts on early ⏀KZ infection.

⏀KZ infection of indicated PAO1 cells with DAPI staining at (A) 50 or (B) 10 minutes post infection. Bacterial cells and ⏀KZ were incubated at 30 °C for 10 minutes prior to imaging. Ejected ⏀KZ genomes are stained by DAPI and marked by arrows. (C) Quantification of the DNA level of two ⏀KZ genes over the course of infection at MOI 0.5. (D) Transcription level of ⏀KZ early (kz054 and kz241), middle (kz180) and late (kz153) genes at MOI 0.1. Points below the assay detection limit were eliminated. “ND” indicates that the transcripts were not detected above the detection limit. Error bars in (C) and (D) represent one standard deviation calculated from two technical replicates. (E) The level of phage proteins in ⏀KZ infected Juk- and Juk+ cells and uninfected cells. The protein level was calculated by normalizing ⏀KZ protein intensity against the uninfected cells. Each dot represents one technical replicate. Cells were infected by ⏀KZ at MOI 2.5 and collected 5 minutes post infection. See also Fig. S2, S3.

Figure 3:. JukA is the infection sensor in the two-component Juk immune system.

(A, B) Fluorescence microscopy of JukA and JukB localization (A) without and (B) with ⏀KZ infection in PAO1::mCherry-jukA/jukB-mNeonGreen strain. (C) JukA localization in the absence of JukB using PAO1[pBAD::mCherry-jukA] strain. (D) JukB localization in the absence of JukA using PAO1[pBAD::jukB-mNeonGreen] strain. In (A) - (D), DAPI stained DNA are shown. Ejected ⏀KZ genomes are marked by arrows. Scale is the same in (C) and (D). (E) Time series visualization of JukA and phage genomes using PAO1[pBAD::mCherry-jukA] strain infected by ⏀KZ. JukA puncta and ⏀KZ genomes are indicated by arrows. Note that MOI 1 is used for ⏀KZ infection. Infected cells are incubated at 30 °C for 10 minutes before being subject to microscopy. See also Fig. S1, S2.

Figure 4:. Phage protein gp241 is sufficient to recruit Juk proteins and to induce JukA and JukB binding.

(A) Map of ⏀KZ gDNA fragments carried by plasmids (B1, C1, and F1) that weaken Juk immunity. (B, C) In the absence of ⏀KZ infection, JukA and JukB localization in PAO1::mCherry-jukA/jukB-mNeonGreen strains carrying (B) wild-type C1 plasmid or C1 plasmid with kz241, kz242, or kz241+kz242 deleted, and (C) plasmids with kz241, kz242, or kz241+kz242 under pBAD promoter induced by 0.25% arabinose. Weak JukA puncta at the cellular pole that are induced by gp241 alone are indicated by arrows. (D, E) In the absence of ⏀KZ infection, JukA and gp241 localization in PAO1::mCherry-jukA/jukB strains carrying C1 plasmid with (D) gp241 being tagged by mNeonGreen on its C-terminus, and (E) the transmembrane (TM) domain of gp241 deleted. Scale bars in (C) - (E) are the same as in (B). (F) In vitro immunoprecipitation (IP) assays among purified proteins JukA, JukB, and gp241. JukA contains His tag on its C terminal and is used as the bait protein. + and − indicate the presence/absence of corresponding protein. Lane 1–5 are input samples. Lane 6–10 are IP samples. See also Fig. S4.

Figure 5:. JukB tetramer structure resembles pore forming toxin.

(A) Structure of the tetramer form of SxJukB, (B) colored by the hydrophobicity (red: hydrophobic, white: hydrophilic), (C) colored by surface electrostatics (white: neutral, blue: positive, red: negative surfaces). (D) Structure of SxJukB monomer and (E) structural alignment between SxJukB and Cry3a (PDB: 4QX2). (F) Schematics of the Liposome-Calcein assay. (G, H) Fraction of liposomes that are permeabilized by purified proteins with PBS as the negative control. Purified proteins were at a final concentration of 1.2nM. Wild-type PaJukA was used in (G). (R73A,K74A) mutated PaJukA protein was used in (H). Error bars represent one standard deviation calculated from two technical replicates. See also Fig. S5.

Figure 6:. Jumbo phage killer systems encompass numerous, distinct putative effectors in diverse bacteria.

(A) Organization of jukA neighborhoods. Genes are shown as arrows. Genes and untranslated regions are proportional for their size. The domain boundaries are approximate. Bacteria strain name, genome identifier, and the coordinates of depicted genes are shown on the right. (B) Spot titration (10-fold serial dilutions) of ⏀KZ and the control phage JBD30 on the lawn of PAO1, PAO1 expressing wild-type PfJuk, and PAO1 expressing mutated PfJuk. (C) Localization of JukA homologs containing a N-terminal mCherry tag in the absence and presence of ⏀KZ infection. MOI 1 is used for ⏀KZ infection. (D) Spot titration (10-fold serial dilutions) of wild-type ⏀KZ and its mutants on indicated bacterial lawns. The expression of PfJuk and its mutants in (B) and (D) were induced by 0.2% arabinose. See also Fig. S6.