Identification of a new effector-immunity pair of Aeromonas hydrophila type VI secretion system

Abstract

The type VI secretion system (T6SS) is a multiprotein weapon that kills eukaryotic predators or prokaryotic competitors by delivering toxic effectors. Despite the importance of T6SS in bacterial environmental adaptation, it is still challenging to systematically identify T6SS effectors because of their high diversity and lack of conserved domains. In this report, we discovered a putative effector gene, U876-17730, in the whole genome of Aeromonas hydrophila NJ-35 based on the reported conservative domain DUF4123 (domain of unknown function), with two cognate immunity proteins encoded downstream. Phylogenetic tree analysis of amino acids indicates that AH17730 belongs to the Tle1 (type VI lipase effector) family, and therefore was named Tle1^AH. The deletion of tle1^AH resulted in significantly decreased biofilm formation, antibacterial competition ability and virulence in zebrafish (Danio rerio) when compared to the wild-type strain. Only when the two immunity proteins coexist can bacteria protect themselves from the toxicity of Tle1^AH. Further study shows that Tle1^AH is a kind of phospholipase that possesses a conserved lipase motif, Gly-X-Ser-X-Gly (X is for any amino acid). Tle1^AH is secreted by T6SS, and this secretion requires its interaction with an associated VgrG (valine-glycine repeat protein G). In conclusion, we identified a T6SS effector-immunity pair and verified its function, which lays the foundation for future research on the role of T6SS in the pathogenic mechanism of A. hydrophila.

Figure 1

Tle1^AHis a potential T6SS effector inA. hydrophilaNJ-35. A Genetic organization of T6SS-related proteins containing the DUF4123 domain in A. hydrophila NJ-35. The numbers below refer to the gene locus tag (U876-XXXXX). Sequencing data for NJ-35 can be obtained from the National Center for Biotechnology Information (accession number: CP006870). B Sequence alignment of conserved catalytic motifs (labeled in red) compared between Tle families. Sequence logos were generated from alignments of the catalytic motifs from the families Tle1-4 (Gly-X-Ser-X-Gly, X is for any amino acid). * represents the catalytic residues. C Phylogenetic analyses of Tle1^AH (AKJ35788.1) with representative members of the families Tle1-4. Figure was prepared using MEGA7.0.

Figure 2

Tle1^AHis required for the interbacterial antagonism ofA. hydrophilaNJ-35. Predator and prey cells at a ratio of 5:1 were cocultured to assay the recovery of surviving prey cells by determining colony forming unit (CFU). A. hydrophila NJ-35 and its mutant derivatives ∆clpV, ∆tle1^AH or C∆tle1^AH were used as the predator strains. ClpV, which encodes a putative ATPase required for T6SS function, was deleted to construct the T6SS⁻ strain (∆clpV). “LB” indicates incubation of E. coli with sterile LB medium alone and serves as the control. AE. coli BL21 as the prey strain. BV. parahaemolyticus RIMD 2210633 as the prey strain. CAeromonas strains as the preys, including A. hydrophila strains ATCC 7966, J-1 and NJ-3, A. sobria CS-2, A. media NJ-8 and A. veronii XH-14. Lane 1, the wild-type A. hydrophila NJ-35; Lane 2, ∆clpV (T6SS⁻); Lane 3, ∆tle1^AH. Data are presented as the mean ± standard deviation (error bars) of three independent experiments. ***P < 0.001, **P < 0.01. *P < 0.05.

Figure 3

Tle1^AHis a phospholipase effector secreted by T6SS ofA. hydrophilaNJ-35. A T6SS-dependent secretion of Tle1^AH was confirmed by Western blot on whole cells and supernatants of A. hydrophila NJ-35 and the ∆clpV strain. ClpV, which encodes a putative ATPase required for T6SS function, was deleted to construct the T6SS⁻ strain (∆clpV). The anti-His antibody was used to measure the production of Tle1^AH and anti-GroEL antibody served as an internal reference. GroEL: heat shock protein Hsp60. B Growth of E. coli TOP10 producing peri-Tle1^AH and peri-Tle1^AHS303A in LB broth. pBAD/His was used for construction of the expression vectors for tle1^AH and its point mutant tle1^AHS303A (the catalysis site of Tle1^AH at position 303 mutated from serine to alanine). To achieve periplasmic localization, the PelB leader sequence was fused in front of the tle1^AH and tle1^AHS303A. Cultures were induced by l-arabinose (l-Ara) at the indicated time by the arrow. A growth curve was drawn by measuring the OD₆₀₀ every 30 min. Data are presented as the mean ± standard deviation (error bars) of three independent experiments. The expression of peri-Tle1^AH and peri-Tle1^AHS303A was detected in E. coli TOP10 by Western blot using anti-His antibody.

Figure 4

Tli1Tli2^AHare the cognate immunity proteins to Tle1^AH. A. hydrophila NJ-35 and the ∆clpV strain were used as the predator strains. ClpV, which encodes a putative ATPase required for T6SS function, was deleted to construct the T6SS⁻ strain (∆clpV). The prey strains included the gene deletion mutant ∆tle1-tli1tli2^AH and its single or double restoration strains of immunity genes, they are, C∆tli1^AH (∆tle1-tli1tli2^AH/pMMB-tli1^AH), C∆tli2^AH (∆tle1-tli1tli2^AH/pMMB-tli2^AH) and C∆tli1tli2^AH (∆tle1-tli1tli2^AH/pMMB-tli1tli2^AH). The predator and prey strains were cultured at a ratio of 5:1, and surviving prey cells were serially diluted and determined on the LB plate containing antibiotics. Data are presented as the mean ± standard deviation (error bars) of three independent experiments. ^##P < 0.01 indicates a significant difference between this group and the NJ-35 group. ***P < 0.001 or *P < 0.05 indicate significant differences between the two specified groups.

Figure 5

The interaction of Tle1^AHwith VgrG was verified by an in vitro pull-down assay. Purified GST, GST-Tle1^AH and His-VgrG were incubated with magnetic beads for the GST pull-down assay. Controls included the incubation with GST-Tle1^AH alone, His-VgrG alone, or GST with His-VgrG. Bound proteins were washed off the beads with the elution buffer and detected by immunoblotting using anti-GST or anti-His antibodies.

Figure 6

Tle1^AHis required for the virulence and colonization ofA. hydrophilaNJ-35. A Determination of the LD₅₀ values of the wild-type and tle1^AH mutant strains in zebrafish. Zebrafish were intraperitoneally (i.p.) injected with tenfold serially diluted bacterial suspensions. The control group was i.p. injected with sterile PBS only. B Competitive assays of NJ-35 and Δtle1^AH in crucian carp. Strains were mixed at a ratio of 1:1 and inoculated to fish by intraperitoneal injection. After 24 h, heart, hepatopancreas, spleen and kidney were harvested for counting of the number of colony-forming units (CFU) per gram of sample. Data are presented as the mean ± standard deviation (error bars) of three independent experiments. ***P < 0.001, **P < 0.01, *P < 0.05.

Figure 7

Biofilm formation of the wild-type andtle1^AHmutant strains. Biofilm formation was determined by crystal violet staining using 96-well plates, and the values were measured at OD₅₉₅. Data are presented as the mean ± standard deviation (error bars) of three independent experiments. *P < 0.05.

Figure 8

Predicted DUF4123-associated T6SS effector-immunity (EI) gene modules in 14A. hydrophilastrains. All the strains used to search for putative effectors are listed in Table 2. The gray arrows indicate DUF4123, the purple arrows indicate the putative effectors, and yellow arrows indicate immunity genes. The arrows indicate the direction of transcription.