Agrobacterium tumefaciens Deploys a Versatile Antibacterial Strategy To Increase Its Competitiveness

Peptidoglycan-related modifications in E. coli recipient cells lead to enhanced T6SS-dependent susceptibility. (a) Schematic diagram of the experimental design to study the rationale of the differential susceptibility of E. coli cells harboring different antibiotic-resistant genes. (b) T6SS-dependent susceptibility of E. coli DH10B recipient cells harboring different antibiotic-resistant genes (Cb^r, carbenicillin resistant; Sp^r, spectinomycin resistant; Gm^r, gentamicin resistant). As demonstrated in panel a, two types of recipient cells were coincubated with A. tumefaciens with (i.e., WT) or without (i.e., ΔtssL) active T6SS on AK medium agar plates. E. coli harboring the carbenicillin-resistant gene was generally more susceptible than E. coli harboring spectinomycin- or gentamicin-resistant genes. Representative data with three biological replicates of two independent experiments are shown, and the samples are grouped by one-way analysis of variance (ANOVA) with a significant P value of <0.05. T6SS-dependent susceptibility index (SI) (17) was designated the logarithm-recovered CFU of that attacked by ΔtssL subtracted by that attacked by WT C58. The higher SI value indicates stronger A. tumefaciens T6SS killing. (c) Recovery of E. coli DH10B harboring vector (pTrc200) or mepS-expressing plasmid (pMepS) after coincubation with A. tumefaciens strain C58 with (T6SS⁺, i.e., WT) or without (T6SS⁻, i.e., ΔtssL) active T6SS on an AKG medium agar plate with IPTG. Recovery of DH10B overexpressing mepS (pMepS) was significantly lower than DH10B harboring vector (pTrc200). (d) Cell morphology of E. coli DH10B strains after growing in LB medium with or without IPTG for 4 h. (e) Recovery of cephalexin (3 μg/ml)-pretreated E. coli DH10B harboring pTrc200 after coincubation with A. tumefaciens with (T6SS⁺, i.e., WT) or without (T6SS⁻, i.e., ΔtssL) active T6SS on AKG agar. (f) Cell morphology of E. coli DH10B after treating with cephalexin (3 μg/ml) at log phase for 3 h. Representative results of at least two independent experiments are shown.

AtTae is a peptidoglycan amidase causing cell elongation and enhances T6SS susceptibility when ectopically expressed in E. coli. (a) E. coli DH10B harboring vector control (pTrc200) or Tae-expressing vector (pTae) was induced by IPTG (0.5 mM) in LB medium for 4 h. The cells were observed under a light microscope, and the longitudinal cell lengths was recorded. A representative experiment of three independent experiments is shown (n = 28; one-way ANOVA; P < 0.001). (b) Three-dimensional structure comparison between AtTae (orange; PDB ID 6IJF) and SmSsp1 (grey; PDB ID 4BI3) or EcTae4 (green; PDB ID 4HFL) based on the jFATCAT_rigid algorithm (43) and drawn by UCSF Chimera (44). Residues substituted in the mutant TaeMX (AtTae residues C47, C144, and C148) are highlighted. (c) UPLC-MS analysis of E. coli peptidoglycan after digestion AtTae. Peak area of disaccharide dipeptide (Di) product (m/z, 699.29; apex RT, 3.0 min) was massively enriched after incubation with WT Tae, but the peak area of disaccharide tetrapeptide (Tetra; m/z, 942.42; apex RT, 2.96 min) was significantly reduced. The pattern was not observed when peptidoglycan was inoculated with nonfunctional TaeMX protein. This indicates that Tae targets and cleaves the bonds d-Glu and mDAP in peptidoglycan. Data are means ± standard deviations (SD) from of three technical replicates. Similar results were obtained from two independent experiments. (d) Expression of the tae mutant-producing TaeMX (C47A, C144A, and C148A) did not promote a significant difference in cell elongation in E. coli from that of the WT Tae. Cell length was the mean of three independent experiments (n = 100; one-way ANOVA; P < 0.001). Susceptibility index (SI) of IPTG-induced E. coli DH10B cell-expressing Tae is enhanced. Recipient E. coli cells harboring vector control (pTrc200), WT Tae (pTae), and TaeMX (pTaeMX) were coincubated with donor cells A. tumefaciens WT and ΔtssL mutant (donor-to-recipient ratio, 30:1) on AKG agar. T6SS-dependent SI was designated the logarithm-recovered CFU of that attacked by ΔtssL subtracted by that attacked by WT C58. The higher SI value indicates stronger A. tumefaciens T6SS killing. Data are means ± SD from three biological replicates by Student’s t test with a significant P value of < 0.05. Representative data of three independent experiments are shown.

A. tumefaciens C58 deploys different effectors, leading to changes in the morphology of recipient E. coli cells. (a) E. coli ΔmepS expressing pRL-GFP(S65T) was inoculated with C58 with different combinations of T6SS effectors. Donor cells (C58) and recipient cells (ΔmepS) were mixed in a 9:1 ratio and spotted on 523 agar plate for 3 h at 28°C. After inoculation, cells were observed under a fluorescence microscope at ×100 magnification. The A. tumefaciens strains used were WT C58 containing all functional effectors (Tde1⁺/Tde2⁺/Tae⁺), taemx carrying functional Tde effectors and a nonfunctional Tae (Tde1⁺/Tde2⁺/TaeMX⁺), Δ2tdei plus pEML4286 carrying a functional Tae and a nonfunctional Tde1 (Tae⁺/Tde1M⁺), Δ2tdei::taemx plus pEML4286 carrying a nonfunctional Tae and a nonfunctional Tde1 (TaeMX⁺/Tde1M⁺), and Δ3TIs lacking any of the three effectors (no effectors). ΔmepS (GFP) not incubated with any A. tumefaciens donor (no donor cells) was used as a control. (b) Similar to the above, the cells were stained with Hoechst before observation under the microscope to visualize genetic materials inside the cells. On incubation with only the donor expressing effective Tde effectors (Tde1⁺/Tde2⁺/TaeMX⁺), ΔmepS cells showed heavy elongation and segmentation of genetic material, whereas donor cells expressing Tae (Tae⁺/Tde1M⁺) led to the enlargement of cells with intact genetic material. With donor lacking no effectors, ΔmepS cells remained uniform in cell size. The text in red indicates the nonfunctional effector in A. tumefaciens. (c) Cellular area of an individual ΔmepS cell as reflected from the GFP signals in the experiment of panel a was automatically measured by ImageJ and then manually confirmed. The text in red indicates the nonfunctional effector. The numbers in the brackets indicate the total number of cells counted, and the cell sizes of ±30 cells of the median are shown and grouped by one-way ANOVA with a significance value of P < 0.05.

Tae-expressing A. tumefaciens retards the growth of recipient cells during coinoculation on rich medium. (a) CFU of recovered E. coli DH10B harboring pRL662 after coinoculation with strains of A. tumefaciens on AKG agar or LB agar at time 0 and after 16 h. The data are means ± SD from six biological replicates from three independent experiments. (b) Growth curve of E. coli DH10B harboring pRL662 after coinoculation with A. tumefaciens strain with only functional AtTae secretion (Δ2tdei*, Δ2tdei plus pEML4286) or no secretion of any effectors (Δ3TIs*, Δ3TIs plus pEML4286) (see Fig S5 in the supplemental material for details) on AKG agar or LB agar at a ratio of 30:1 (donor to recipient) for 16 h at 28°C. Data are means ± SD from three biological replicates. Representative data of three independent experiments are shown (*, P < 0.05). (c) Growth curve of DH10B harboring empty vector or derivative expressing Tae immunity protein (pTai) after coinoculation with A. tumefaciens strain Δ2tdei* or Δ3TIs* on LB agar at a ratio of 10:1 (donor to recipient) for 16 h at 28°C. Data are means ± SD from three biological replicates. Representative data of three independent experiments are shown (*, P < 0.05).

Tae-expressing A. tumefaciens shows better competitiveness in a bacterial population. (a) A. tumefaciens with Tae shows better competitiveness among E. coli cells. WT (WT plus pTrc200), ΔtssL (ΔtssL plus pTrc200), and a strain with only functional AtTae secretion (Δ2tdei*) of A. tumefaciens strains were each coinoculated with E. coli BW25113 at a 1:9 ratio on an LB agar plate for 16 h. The population of ΔtssL and Δ2tdei* cells among the E. coli cells was quantified by qPCR and represented as an index relative to the population of WT. Similarly, the A. tumefaciens strains were coincubated with BW25113 harboring a control vector (−tai, pRL662) or tai-expressing vector (+tai, pTai). The population of A. tumefaciens strains among the BW25113 cells without tai (−tai) was quantified by qPCR and represented as an index relative to the population of A. tumefaciens strains among the BW25113 cells with tai (+tai). The index (competitiveness index) is 1 if there is no advantage between two strains/conditions but >1 if there is an advantage and vice versa. Each dot represents an average of three technical replicates in an experiment. Data are means ± SD from at least three independent experiments. (b) The proportion of C58 donor cells with (WT) or without functional Tae (taemx) when coinoculated with recipient C58 cells with (Δ2tdei) or without (Δ3TIs) tae-tai toxin immunity pair. The cell proportion of donor cells in a mixed population was determined by qPCR with specific primers of tde1 and 16S rRNA genes. The WT strain used is the WT tae gene recovered in Δtae that underwent the same process in generating taemx. Each dot represents an average of three technical replicates in an experiment. Data are means ± SD from three independent experiments.