Two PAAR Proteins with Different C-Terminal Extended Domains Have Distinct Ecological Functions in Myxococcus xanthus

The type VI secretion system (T6SS) is a bacterial cell contact-dependent weapon capable of delivering protein effectors into neighboring cells. The PAAR protein is located at the top of the nanomachine and carries an effector for delivery.

ABSTRACT

Bacterial proline-alanine-alanine-arginine (PAAR) proteins are located at the top of the type VI secretion system (T6SS) nanomachine and carry and deliver effectors into neighboring cells. Many PAAR proteins are fused with a variable C-terminal extended domain (CTD). Here, we report that two paar-ctd genes (MXAN_RS08765 and MXAN_RS36995) located in two homologous operons are involved in different ecological functions of Myxococcus xanthus. MXAN_RS08765 inhibited the growth of plant-pathogenic fungi, while MXAN_RS36995 was associated with the colony-merger incompatibility of M. xanthus cells. These two PAAR-CTD proteins were both toxic to Escherichia coli cells, while MXAN_RS08765, but not MXAN_RS36995, was also toxic to Saccharomyces cerevisiae cells. Their downstream adjacent genes, i.e., MXAN_RS08760 and MXAN_RS24590, protected against the toxicities. The MXAN_RS36995 protein was demonstrated to have nuclease activity, and the activity was inhibited by the presence of MXAN_RS24590. Our results highlight that the PAAR proteins diversify the CTDs to play divergent roles in M. xanthus.

IMPORTANCE The type VI secretion system (T6SS) is a bacterial cell contact-dependent weapon capable of delivering protein effectors into neighboring cells. The PAAR protein is located at the top of the nanomachine and carries an effector for delivery. Many PAAR proteins are extended with a diverse C-terminal sequence with an unknown structure and function. Here, we report two paar-ctd genes located in two homologous operons involved in different ecological functions of Myxococcus xanthus; one has antifungal activity, and the other is associated with the kin discrimination phenotype. The PAAR-CTD proteins and the proteins encoded by their downstream genes form two toxin-immunity protein pairs. We demonstrated that the C-terminal diversification of the PAAR-CTD proteins enriches the ecological functions of bacterial cells.

INTRODUCTION

The type VI secretion system (T6SS) is a macromolecular weapon assembled in many Gram-negative bacteria to inject protein effectors/toxins into eukaryotic hosts or prokaryotic competitors for survival and provides advantages in various types of ecological competition (1, 2). The proline-alanine-alanine-arginine (PAAR) proteins are conical in structure and are located at the T6SS tip to sharpen the injector for piercing the target cell envelope (3). The PAAR protein is also a carrier of T6SS effectors and uses several loading modes (4), the most common mode of which is to directly fuse the toxin protein as the C-terminal extended domain (CTD), forming a PAAR-CTD protein, such as Tse7 of Pseudomonas aeruginosa, Rhs-CT1 to CT10 of Escherichia coli, and Tne2 of Pseudomonas protegens (5–8). The C-terminal toxin domain of PAAR-CTD is diversified. Diverse CTDs are predicted to have various enzymatic activities that are toxic to prokaryotic or eukaryotic cells (3).

Myxococcus xanthus is a predatory bacterium that is able to grow by lysing other bacterial or fungal cells and that possesses complex social behaviors (9–11). M. xanthus is known for kin discrimination, and recent studies have shown that kin discrimination is formed by delivering effectors of the type VI secretion, rearrangement hot spots, or outer membrane exchange system (12, 13). The M. xanthus strain DK1622 genome encodes a T6SS and six PAARs (14). With a correlated method of cryogenic photoactivated localization microscopy and cryo-electron tomography, the T6SS structure has been identified and visualized in M. xanthus cells (14, 15). What is interesting is how multiple PAAR copies encoded in the genome affect the ecological function of the host bacterium.

Kin discrimination by M. xanthus can manifest as visible boundaries between two encountered swarming colonies, termed colony-merger incompatibility (16). We previously performed random insertion mutations in the genome of M. xanthus DK1622 and obtained 11 mutants deficient in colony-merger formation with the wild-type strain (17). We also demonstrated that the deficiency in one mutant is associated with a novel nuclease toxin and immunity protein pair, and the toxin protein is delivered via the T6SS with the help of a single-domain PAAR protein (18). Here, we investigated the functional differences between the two PAAR-CTDs in M. xanthus DK1622. The two paar-ctd genes were located in two gene clusters with highly similar organization and composition, one of which was related to colony-merger incompatibility and the other of which was not. We analyzed the toxicity, toxin immunity, and enzymatic characteristics of the two PAAR-CTD proteins, as well as their neighboring homologs. Our results highlight that the paar-ctd genes diversify the C-terminal sequences to enrich bacterial ecological functions.

RESULTS AND DISCUSSION

Two PAAR loci with different CTDs are highly similar in composition and organization.

Bioinformatics analysis revealed that the genome of M. xanthus DK1622 encodes six PAAR proteins away from the T6SS gene cluster and divided these proteins into two phylogenetic groups; one consists of short proteins with single PAAR domains, and the other consists of large proteins containing the N- and C-terminal domains (Fig. 1A). The two large proteins, MXAN_RS08765 and MXAN_RS36995, contained 417 and 380 amino acids, respectively, of which 127 N-terminal amino acids were highly conserved with 88% similarity at the amino acid level and were predicted to be the PAAR domain. Their C-terminal fragments were diverse, with unknown structural and functional information. Notably, among the six paar genes, five, all except for MXAN_RS08765, were located at the loci individually involved in the colony-incompatible phenotype of M. xanthus DK1622, and one, MXAN_0044, has been demonstrated to be related to the delivery of an AHH nuclease toxin (MXAN_0050) (18). However, the C-terminal fragments of MXAN_RS08765 and MXAN_RS36995 did not contain the AHH motif.

FIG 1.

Two proline-alanine-alanine-arginine (PAAR) loci with different C-terminal extended domains (CTDs) are highly similar in composition and organization. (A) Phylogenetic relationships of the PAAR proteins or the N-terminal domains of the PAAR-CTD proteins in M. xanthus DK1622. The bootstrap value (1,000 replicates) for the phylogenetic tree is given at each node. (B) Genetic organization of the six loci containing the paar genes in M. xanthus DK1622. Homologous regions and genes are shown in the same color. Gray indicates nonhomologous genes. Arrows indicate the direction of transcription. Genes are drawn to scale. The locus tags for gene annotation in the NCBI GenBank database and reannotation in the NCBI Reference Sequence (RefSeq) database are provided. (C) Reverse transcription-PCR (RT-PCR) cotranscriptional verification of the genes at the MXAN_RS08765 and MXAN_RS36995 loci. Labeled on the left are the lengths of the amplified regions containing the 3′ end of the upstream gene, the intergenic sequence, and the 5′ end of the downstream adjacent gene. The right panel shows the amplification results of RT-PCR. The RNA template was extracted from cells grown in CTT liquid medium for 24 h. The 16S rRNA sequence was amplified from the RNA template to evaluate the RNA quality. The cDNA template was used to amplify the 16S rRNA gene as the positive control. Yields of the RT-PCR products with the expected sizes indicated cotranscription of two neighboring genes. The migration positions of the molecular weight marker are indicated.

The MXAN_RS08765- and MXAN_RS36995-containing loci were found to have five genes, which were homologous and organized in the same way (Fig. 1B). We demonstrated that the five genes in the two gene loci were cotranscribed by the reverse transcription-PCR (RT-PCR) technique (19, 20) (Fig. 1C), suggesting that they might be functionally related. Moreover, the three upstream genes in the two operons encoded proteins with the PRK06147 domain (similarity, 56%), the DUF2169 domain (similarity, 70%), and the TIGR02270 domain (similarity, 59%), and the downstream gene of paar is structurally unknown. Among these proteins, Bondage et al. demonstrated that DUF2169 proteins are involved in the delivery of T6SS toxins in Agrobacterium tumefaciens, probably as adaptor proteins (21).

We further screened 12 other myxobacterial genomes and revealed 24 genes encoding PAAR-CTD proteins (see Fig. S1 in the supplemental material). The myxobacterial strains had at least a single paar-ctd gene copy, e.g., Myxococcus macrosporus DSM 14697; many possessed two paralogs, such as that in M. xanthus DK1622, while some had multiple paar-ctd gene copies, e.g., five paralogs in Myxococcus virescens DSM 2260. These PAAR protein sequences were all 365 to 482 amino acids (aa) in length, comprising a conserved N-terminal PAAR domain and a variable C-terminal sequence. Similarly to the MXAN_RS08765 and MXAN_RS36995 loci, these paar-ctd genes were found to be adjacent upstream with homologous genes encoding proteins with the PRK06147, DUF2169, and TIGR02270 domains. The highly similar organization and composition suggest that these loci are probably derived from the same ancestor and are highly related in function.

Two PAAR-CTDs involve different phenotypes.

In our previous work, a transposon mutation at MXAN_RS24590, the downstream gene of MXAN_RS36995, caused a colony boundary between the mutant and the wild-type strain (17). To determine whether MXAN_RS24590 is responsible for the colony-incompatible phenotype, this gene was subjected to an in-frame deletion, as were the other four individual genes in the operon. The five genes in the MXAN_RS08765 locus were also deleted individually to check whether this locus was similarly involved in the colony-incompatible phenotype. As expected, the MXAN_RS24590 deletion (ΔMXAN_RS24590) mutant formed a visible colony boundary with DK1622, and the deletion of MXAN_RS36995 (ΔMXAN_RS36995) eliminated the incompatibility (Fig. 2A), which was similar to the phenotype caused by mutation of a toxin-immunity protein system (18). Deletion of the other three genes in the operon caused the mutants to form merged colony boundaries with either DK1622 or ΔMXAN_RS24590 strains (see Fig. S2 in the supplemental material), which suggests that these three genes were also involved in colony-merger incompatibility and probably participated in toxin delivery. However, deletion of MXAN_RS08760 or the other genes in the MXAN_RS08765 locus did not produce any colony-incompatible phenotype (Fig. 2A and Fig. S2).

FIG 2.

The MXAN_RS36995 gene participates in colony-merger incompatibility. (A) The MXAN_RS08760 deletion mutant merged colonies with DK1622 and the ΔMXAN_RS08765 mutant, while the ΔMXAN_RS24590 mutant formed colony boundaries with DK1622, and deletion of MXAN_RS36995 merged colonies with either the wild-type strain or the ΔMXAN_RS24590 mutant. (B) Complementation of the MXAN_RS24590 gene corrected by us with the ΔMXAN_RS24590 mutant (ΔMXAN_RS24590::MXAN_RS24590), rather than the MXAN_RS24590 gene predicted by NCBI (C) with the ΔMXAN_RS24590 mutant (ΔMXAN_RS24590::MXAN_RS24590-NCBI), recovered the colony-merger phenotype of the ΔMXAN_RS24590 mutant. Complementation of MXAN_RS36995 with the ΔMXAN_RS36995 mutant recovered the boundary-forming phenotype as the wild-type strain, i.e., merging colonies with DK1622 but forming boundaries with the ΔMXAN_RS24590 mutant. The corresponding strain complemented with the MXAN_RS08765 or MXAN_RS08760 gene merged colonies with either DK1622 or the ΔMXAN_RS08760 mutant. All photos are representative of three biological replicates. Bars, 5 mm (A, B, and C).

To confirm this conclusion, we further complemented genes ectopically at the attP site in their corresponding deletion mutants. The strain complemented with MXAN_RS24590 (ΔMXAN_RS24590::MXAN_RS24590) retrieved the incompatible phenotype of the ΔMXAN_RS24590 mutant, and the strain complemented with MXAN_RS36995 (ΔMXAN_RS36995::MXAN_RS36995) retrieved the phenotype of the ΔMXAN_RS36995 mutant, while complementation of MXAN_RS08760 or MXAN_RS08765 in the corresponding deletion mutant exhibited no phenotypic change (Fig. 2B). The results indicate that the MXAN_RS36995 and MXAN_RS24590 gene pair is responsible for the incompatible phenotype, probably as a pair of toxin-immunity proteins (18).

M. xanthus is a predatory bacterium that is able to grow by lysing prey cells for nutrient uptake (22). We found that M. xanthus DK1622 inhibited the growth of the tested phytopathogenic fungi Colletotrichum orbiculare, Curvularia lunata, and Septoria lycopersici, and the inhibition was lost by deleting the T6SS gene cluster (Fig. 3A and B). Interestingly, the ΔMXAN_RS08765 mutant also partly lost the ability to inhibit the three tested phytopathogenic fungi, while the ΔMXAN_RS36995 mutant had no effect. Complementation of the ΔMXAN_RS08765 mutant with MXAN_RS08765 restored the antifungal ability. The above results indicate that the T6SS gene cluster and the MXAN_RS08765 gene are functionally related to the antifungal ability of M. xanthus. Therefore, these two paar-ctd genes have evolved different ecological functions; MXAN_RS36995 is associated with colony-merger incompatibility, while MXAN_RS08765 is involved in antifungal activity.

FIG 3.

The MXAN_RS08765 gene is associated with antifungal activity. (A) Effects of DK1622 and the Δt6ss, ΔMXAN_RS08765, and ΔMXAN_RS36995 mutants on the cellular growth of S. lycopersici, C. lunata, and C. orbiculare. (B) The hyphal growth of S. lycopersici was inhibited by DK1622 and by the ΔMXAN_RS36995 mutant, but not by the Δt6ss or ΔMXAN_RS08765 mutants. Complementation of MXAN_RS08765 with the ΔMXAN_RS08765 mutant recovered inhibition of S. lycopersici. All the photos are representative of three biological replicates. Bar, 5 mm.

The paar-ctd genes and their downstream genes encode two toxin-immunity protein pairs.

Bacteria usually have an immunity gene adjacent to the toxin gene in the genome (23–25). Our aforementioned results suggested that the paar-ctd genes and their downstream genes probably encode pairs of toxin and immunity proteins. To investigate their in vivo functions, we constructed the PAAR-CTD protein gene in plasmid pMAL c5X under an isopropyl β-d-1-thiogalactopyranoside (IPTG)-inducible promoter, and the immunity protein gene in plasmid pACYC Duet-1 under a constitutive promoter. The two compatible plasmids were cotransformed into Escherichia coli BL21 to assay their effects on cellular growth by expressing the PAAR-CTD proteins with or without their corresponding immunity proteins. The two empty plasmids of pMAL c5X and pACYC Duet-1 were introduced into the BL21 strain as the control. When the expression of MXAN_RS08765 or MXAN_RS36995 was induced by IPTG, which was added in Luria-Bertani (LB) medium in advance, the growth of the E. coli cells with the MXAN_RS08765- or MXAN_RS36995-harboring pMAL c5X plasmid and the empty pACYC Duet-1 plasmid was significantly inhibited (Fig. 4A). The results suggested that although the addition of IPTG exhibited some inhibition on E. coli cells (26), the PAAR-CTD proteins were both obviously toxic to E. coli cells. However, the toxin effect of the PAAR-CTD protein could be alleviated by the coexpression of protein encoded by the corresponding downstream gene (the cells containing both the MXAN_RS08765- or MXAN_RS36995-harboring pMAL c5X plasmid and the MXAN_RS08760- or MXAN_RS24590-harboring pACYC Duet-1 plasmid; Fig. 4A). The results suggest that the paar-ctd gene and its downstream gene formed toxin-immunity systems. When the two paar-ctd genes were crossly constructed with the downstream genes, no cross-immune reaction appeared (Fig. 4B).

FIG 4.

Effects of the paar-ctd genes and their associated genes in Escherichia coli. (A) Growth of serial dilutions of E. coli BL21 cells containing MXAN_RS08765, MXAN_RS08760, MXAN_RS36995, and/or MXAN_RS24590 on LB agar, to which IPTG was added in advance. An equal cell number of each dilution was spotted on the plate. The expression of MXAN_RS08765 and MXAN_RS36995, induced by the addition of IPTG, was toxic to E. coli cells and was recovered by the coexpression of the downstream genes MXAN_RS08760 and MXAN_RS24590, which we corrected, respectively, rather than by that of MXAN_RS08760-NCBI and MXAN_RS24590-NCBI. (B) The MXAN_RS08765-MXAN_RS08760 and MXAN_RS36995-MXAN_RS24590 pairs were cross-constructed into E. coli BL21 cells to check the cross-immunity reactions. Control, E. coli cells with the empty pMAL c5X and pACYC Duet-1 plasmids; MXAN_RS08765/MXAN_RS36995, E. coli cells with the MXAN_RS08765- or MXAN_RS36995-harboring pMAL c5X and empty pACYC Duet-1 plasmids; MXAN_RS08760/MXAN_RS24590, E. coli cells with the empty pMAL c5X and MXAN_RS08760- or MXAN_RS24590-harboring pACYC Duet-1 plasmids; MXAN_RS08765/MXAN_RS36995 & MXAN_RS08760/MXAN_RS24590, E. coli cells with the MXAN_RS08765- or MXAN_RS36995-harboring pMAL c5X and MXAN_RS08760- or MXAN_RS24590-harboring pACYC Duet-1 plasmids. Each dilution was spotted on the plate at an optical density at 600 nm (OD₆₀₀) of 1, 0.75, or 0.5. Photographs were taken after 15 h of incubation. The images are representative of three biological replicates. (C) Comparison of the NCBI predictions and our predictions of the gene positions of MXAN_RS08760 and MXAN_RS24590.

It is worth noting that the positions of the MXAN_RS08760 and MXAN_RS24590 genes predicted by NCBI are incorrect (Fig. 4C). The MXAN_RS24590 gene in the genome of DK1622 was predicted at the position of 6,337,957 to 6,338,418 bp in the reverse strand (462 bp in size, MXAN_RS24590-NCBI in this study) in the NCBI database, which is different from our correction (6,337,957 to 6,338,685 bp, 729 bp in size, MXAN_RS24590). Similarly, the NCBI position of MXAN_RS08760 was 2,143,618 to 2,143,965 bp (348 bp in size, MXAN_RS08760-NCBI), while our corrected position of the gene was 2,143,618 to 2,144,346 bp (729 bp in size, MXAN_RS08760). We also constructed MXAN_RS08760-NCBI and MXAN_RS24590-NCBI under a constitutive promoter in the pACYC Duet-1 plasmid and coexpressed them with the corresponding PAAR-CTD, but the growth of E. coli cells was still inhibited (Fig. 4A). However, our corrected immunity protein genes restored the growth of E. coli. In addition, since the ΔMXAN_RS24590 mutant can form a colony boundary with DK1622, we complemented the MXAN_RS24590-NCBI gene ectopically with the ΔMXAN_RS24590 mutant (ΔMXAN_RS24590::MXAN_RS24590-NCBI). The results show that the boundary phenotype had not been changed (Fig. 2C). In comparison, the strain complemented with the corrected MXAN_RS24590 retrieved the incompatible phenotype of the ΔMXAN_RS24590 mutant. Therefore, the corrected MXAN_RS08760 and MXAN_RS24590 genes are complete and can exercise real immunity protein functions. The positions of all MXAN_RS08765 and MXAN_RS24590 genes used in this study are the corrected positions.

We cocultured the immunity gene deletion mutants with other strains to evaluate the in vivo toxicity of PAAR-CTDs in M. xanthus cells. We constructed kanamycin (Km)-resistant strains of the ΔMXAN_RS08760 (ΔMXAN_RS08760^R) and ΔMXAN_RS24590 (ΔMXAN_RS24590^R) immunity gene deletion mutants using a previous protocol (17) and cultured them by mixing with the Km-sensitive strains of wild-type DK1622, the t6ss deletion mutant, and the corresponding ΔMXAN_RS08765 or ΔMXAN_RS36995 toxin gene deletion mutant (at a 1:10 ratio), respectively. The cocultures of the Km-resistant and Km-sensitive ΔMXAN_RS08760 (or ΔMXAN_RS24590) immunity gene deletion mutants were used as controls. After 4 days of incubation in Km-free medium, the mixed cells were recovered, plated on Km-containing medium, and the CFU count of the resistant strains were calculated to evaluate their competitive abilities in cocultures. We found that the CFU of the ΔMXAN_RS24590^R and ΔMXAN_RS08760^R strains in the cocultures with DK1622 were approximately 50 and 2.5 times lower than those in the cocultures with the sensitive ΔMXAN_RS24590 and ΔMXAN_RS08760 strains, respectively (Fig. 5A and B). The mutations of the corresponding paar-ctd genes or the T6SS apparatus in DK1622 weakened its superiority in competition to a great extent. In comparison, the CFU count of the ΔMXAN_RS08760^R (or ΔMXAN_RS24590^R) strain showed no significant difference in the cocultures with the Δt6ss, ΔMXAN_RS08765 (or ΔMXAN_RS36995) and ΔMXAN_RS08760 (or ΔMXAN_RS24590) mutants. The results suggest that MXAN_RS36995 is more toxic to M. xanthus cells than MXAN_RS08765, which probably led to the inability of the ΔMXAN_RS08760 mutant to form a colony boundary with DK1622.

FIG 5.

In vivo functional assays of paar-ctd. (A) Competitive growth assay of the ΔMXAN_RS08760^R mutant cocultured with the Km-sensitive DK1622, ΔMXAN_RS08760, ΔMXAN_RS08765, and Δt6ss strains. (B) Competitive growth assay of the ΔMXAN_RS24590^R mutant cocultured with the Km-sensitive DK1622, ΔMXAN_RS24590, ΔMXAN_RS36995, and Δt6ss strains. Three dilutions and three replicates were performed for each experiment. The error bars represent the standard deviations, and asterisks denote the P values of the t test for the differences, **, P < 0.01; NS, not significant. (C) Growth of S. cerevisiae 102-5B strains containing MXAN_RS08765 or MXAN_RS36995 on synthetic complete (SC)-Ura agar supplemented with 0.02% galactose to induce the expression of MXAN_RS08765 or MXAN_RS36995 compared with growth on 0.02% glucose. The control was the S. cerevisiae strain containing the empty pYES2 plasmid. An equal number of S. cerevisiae 102-5B cells were serially diluted and spotted on the plate at an OD₆₀₀ of 2.5, 2, 1.5, 1, or 0.1. The images are representative of three biological replicates.

When the two paar-ctd genes were expressed in Saccharomyces cerevisiae cells by using the galactose (Gal)-inducible expression system, induced expression of MXAN_RS08765 inhibited the growth of S. cerevisiae, while the expression of MXAN_RS36995 had no effect (Fig. 5C). The in vivo toxicity experiment results in different hosts were consistent with the phenotypes of these two paar genes in M. xanthus.

The two PAAR-CTDs and their corresponding immunity proteins interact specifically.

We demonstrated the characteristics of the MXAN_RS36995 and MXAN_RS24590 proteins in vitro. To facilitate the assays, MXAN_RS24590 was cloned into the pACYC Duet-1 plasmid with the His₆ tag sequence (His₆-24590) and MXAN_RS36995 was cloned into pMAL c5X plasmid with the maltose-binding protein (MBP) tag sequence (MBP-36995). The two plasmids were expressed in E. coli. After IPTG induction, the expressed His₆-24590 and MBP-36995 proteins were extracted and purified through Ni²⁺ beads and amylose resin for the pulldown experiment. Because MBP-36995 lacks the His₆-epitope tag, its retention on Ni²⁺ beads will depend on its potential binding to His₆-24590. Similarly, the retention of His₆-24590 on amylose resin will depend on binding to MBP-36995. SDS-PAGE showed that when the mixtures of His₆-24590 and MBP-36995 were eluted through the column, the proteins bound to either the Ni²⁺ beads or the amylose resin contained both the His₆-24590 and MBP-36995 proteins (Fig. 6A). To confirm this result, the MBP tag protein was expressed in the pMAL c5X plasmid, purified by amylose resin, and mixed with the His₆-24590 protein. The MBP and His₆-24590 proteins were present separately in the bound and unbound fractions with either the Ni²⁺ beads or the amylose resin (Fig. 6A), i.e., they were unable to bind to each other. Similarly, the MBP-36995 protein alone could not bind to the Ni²⁺ beads. The results indicated that the MXAN_RS24590 protein was able to bind to the MXAN_RS36995 protein.

FIG 6.

The nuclease activity of MXAN_RS36995 and interactions between the PAAR-CTD and immunity proteins. (A) A pulldown assay on the in vitro binding activities between the MXAN_RS36995 (MBP-36995) and MXAN_RS24590 (His₆-24590) proteins. The His₆-24590 and MBP-36995 protein complexes were captured by either Ni²⁺ beads or amylose resin. The MBP protein was incubated with His₆-24590 protein as the negative control, and the two proteins were separately present in the bound and unbound fractions of either the Ni²⁺ beads or amylose resin. The MBP-36995 protein was incubated with Ni²⁺ beads as another control, and the MBP-36995 protein was present in the unbound fractions. The resin-bound and unbound samples were analyzed by SDS-PAGE and Coomassie blue staining. (B) A pulldown assay of MXAN_RS08765 (MBP-08765) and MXAN_RS08760 (His₆-08760) proteins. The His₆-08760 and MBP-08765 protein complexes were captured by either Ni²⁺ beads or amylose resin. Bound, protein was captured by either Ni²⁺ beads or amylose resin; unbound, protein was not bound to Ni²⁺ beads or amylose resin and was eluted by buffer. (C) Yeast two-hybrid analysis of the interaction between PAAR-CTD and the immunity protein. Bait constructs of PAAR-CTD/immunity protein, fused with GBD in the pGBKT7 vector, were cotransformed with prey constructs of immunity protein/PAAR-CTD, fused with GAD in the pGADT7 vector, into the Y2HGold yeast cells. The positive control (a) was Y2HGold with pGADT7-T-antigen and pGBKT7-p53, while the negative control (f) was Y2HGold with two empty vectors. “AD-08760/BK-08765” in subpanel b indicates the Y2HGold cells with MXAN_RS08760 fused to GAD and MXAN_RS08765 fused to GBD, “AD-08765/BK-08760” in subpanel c indicates the Y2HGold cells with MXAN_RS08765 fused to GAD and MXAN_RS08760 fused to GBD, and so on. These strains were grown on synthetic dropout (SD)-Trp-Leu medium (left lanes), SD-Trp-Leu + X-α-Gal (middle lanes) or SD-Trp-Leu-His-Ade (right lanes). (D) The MXAN_RS36995 (MBP-36995) protein (40 μM) cleaved the genomic DNA of E. coli DH5α (0.5 μg), and the MXAN_RS24590 (His₆-24590) protein (40 μM) inhibited the nuclease activity. The incubation was performed at 37°C for 15 min. DNase I was used as the positive control, and no protein was added to the negative control. The migration positions of the λ-HindIII-digested DNA standard markers are indicated in kilobase pairs (kbp). The reaction products were detected by gel electrophoresis and ethidium bromide staining.

The in vitro interaction of the MXAN_RS08765 and MXAN_RS08760 proteins was also assayed. However, the MXAN_RS08765 protein could not be purified alone, but could be copurified together with the MXAN_RS08760 protein. Thus, the extracted and purified proteins from the E. coli cells containing the MXAN_RS08760-harboring pACYC Duet-1 (His₆-08760) and MXAN_RS08765-harboring pMAL c5X (MBP-08765) were employed for the pulldown experiment (Fig. 6B).

We further investigated the interaction between the PAAR-CTD and immunity proteins in vivo by yeast two-hybrid (Y2H) assay. In this system, the Gal4 activation domain (GAD) and the DNA binding domain (GBD) were separately present in the pGADT7 and pGBKT7 plasmids. We fused the PAAR-CTD and immunity proteins to the C terminus of GAD and the N terminus of GBD, respectively, and the two plasmids were transformed simultaneously into the Y2HGold strain. If PAAR-CTD and the immunity protein were able to bind to one another, the GBD and the GAD were brought into proximity to allow the cells to grow on synthetic dropout (SD)-Trp-Leu-Ade-His minimal medium and make the cells form blue colonies on SD-Trp-Leu + X-α-Gal agar medium. Our results showed that the cells containing coexpressed PAAR-CTD and the corresponding immunity protein yielded blue colonies on the SD-Trp-Leu + X-α-Gal agar plate and were able to grow on the SD-Trp-Leu-His-Ade agar plate (Fig. 6C). Thus, the PAAR-CTD protein is able to bind to its corresponding immunity protein, but unable to bind to another immunity protein. The results further demonstrated that the two PAAR-CTD proteins formed toxin-immunity protein pairs with the proteins encoded by the downstream genes of paar-ctd, with no cross-immune reaction between the two pairs.

In our previous work, we determined that the toxin gene (MXAN_0050) of the SI01 mutant deficient in self-identification encodes an AHH motif-containing nuclease delivered by a PAAR-mediated T6SS mechanism (17, 18). However, the C-terminal fragment of the toxin MXAN_RS36995 did not contain any known domain. We assayed the in vitro nuclease activity using prepared MBP-tagged MXAN_RS36995, which showed that the protein, similarly to DNase I, completely digested the genomic DNA of E. coli DH5α (Fig. 6D). The nuclease activity was blocked by the presence of the His₆-tagged MXAN_RS24590 protein. However, the MXAN_RS24590 protein was unable to block the nuclease activity of DNase I. These results indicate that MXAN_RS36995 is a new kind of nuclease and that MXAN_RS24590 is its specific immunity protein.

In summary, M. xanthus possesses two PAAR-CTD proteins that are conserved at the N-terminal PAAR domain but polymorphic at the C-terminal domain. These two proteins are in highly similar composition and organization but perform distinct ecological functions; MXAN_RS08765 has antifungal activity, while MXAN_RS36995 is associated with the formation of colony boundaries. The two PAAR-CTD proteins specifically interact with the proteins encoded by the downstream genes of paar-ctd, respectively, forming two independent toxin-immunity protein pairs. The C-terminal diversification of the PAAR-CTD proteins enriches the ecological functions of bacterial cells.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions.

The bacterial strains and plasmids used in this study are listed in Table 1. Myxococcus strains and mutants were cultivated at 30°C in liquid CTT medium (10 g/liter casein peptone, 1.97 g/liter MgSO₄·7H₂O, 1 mM KH₂PO4/K₂HPO₄ buffer, and 10 mM Tris HCl buffer [pH 7.6]) with shaking at 200 rpm or on CTT plates supplemented with 1.5% agar (27). The E. coli strains in this study included XL1-Blue MR for deletion plasmid preservation, DH5α λpir for maintenance of the suicide vector pSWU19, BL21(DE3) for protein expression, and DH5a for the conservation of other plasmids. The E. coli strains were cultivated in liquid Luria-Bertani (LB) medium with shaking at 200 rpm or on solid LB medium plates (10 g/liter tryptone, 5 g/liter yeast extract, and 10 g/liter NaCl) at 37°C (28). Where required, the CTT and LB media were supplemented with kanamycin (Km; 40 μg/ml), ampicillin (Amp; 100 μg/ml) or chloramphenicol (Cm; 34 μg/ml).

TABLE 1.

Strains and plasmids used in this study

Designation	Genotype or descriptiona	Source
Strains
M. xanthus
DK1622	Wild type	D. Kaiser, Stanford University
ΔMXAN_RS08760	Deletion of MXAN_RS08760, from 2143618–2144346	This study
ΔMXAN_RS08765	Deletion of MXAN_RS08765, from 2144359–2145612	This study
ΔMXAN_RS08770	Deletion of MXAN_RS08770, from 2145600–2146691	This study
ΔMXAN_RS08775	Deletion of MXAN_RS08775, from 2146693–2147697	This study
ΔMXAN_RS08780	Deletion of MXAN_RS08780, from 2147726–2149021	This study
ΔMXAN_RS24590	Deletion of MXAN_RS24590, from 6337957–6338685	This study
ΔMXAN_RS36995	Deletion of MXAN_RS36995, from 6338694–6339836	This study
ΔMXAN_RS24595	Deletion of MXAN_RS24595, from 6339850–6340899	This study
ΔMXAN_RS24600	Deletion of MXAN_RS24600, from 6340896–6341906	This study
ΔMXAN_RS24605	Deletion of MXAN_RS24605, from 6341913–6343220	This study
ΔMXAN_RS08760::MXAN_RS08760	Complementation strain, Kmr; ΔMXAN_RS08760+ pSWU19 containing the pilA promoter and the MXAN_RS08760 gene	This study
ΔMXAN_RS08765::MXAN_RS08765	Complementation strain, Kmr; ΔMXAN_RS08765+ pSWU19 containing the pilA promoter and the MXAN_RS08765 gene	This study
ΔMXAN_RS24590::MXAN_RS24590	Complementation strain, Kmr; ΔMXAN_RS24590+ pSWU19 containing the pilA promoter and the MXAN_RS24590 gene	This study
ΔMXAN_RS36995::MXAN_RS36995	Complementation strain, Kmr; ΔMXAN_RS36995+ pSWU19 containing the pilA promoter and the MXAN_RS36995 gene	This study
Δt6ss	Deletion of MXAN_4800–MXAN_4813, from 6004071–6024574	This study
ΔMXAN_RS08760R	Kmr; ΔMXAN_RS08760+ pSWU19	This study
ΔMXAN_RS24590R	Kmr; ΔMXAN_RS24590+ pSWU19	This study
E. coli
DH5α	F− endA1 glnV44 thi-1 recA1 relA1 gyrA96 deoR nupG Φ80dlacZΔM15 Δ(lacZYA-argF)U169, hsdR17(rK−mK+), λ−	Life Technologies, Inc.
XL1-Blue MR	Δ(mcrA)183 Δ(mcrCB-hsdSMR-mrr)173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F′ proAB lacIqZΔM15 Tn10 (Tetr)	Stratagene Co.
DH5α λpir	Φ80dlacZΔM15 ΔlacU169 recA1 endA1 hsdR17 supE44 thi-1 gyrA relA1 λpir	H. B. Kaplan, University of Texas
BL21(DE3)	E. coli strain B F− ompT gal dcm lon hsdSB(rB−mB−) λ(DE3 [lacI lacUV5-T7 gene 1 ind1 sam7 nin5]) [malB+]K-12(λS)	Stratagene Co.
Fungi
C. orbiculare ACCC36065	Wild type	Agricultural Culture Collection of China
C. lunata ACCC38967	Wild type	Agricultural Culture Collection of China
S. lycopersici ACCC36106	Wild type	Agricultural Culture Collection of China
S. cerevisiae
CEN.PK 102-5B	MATa; ura3-52, his3Δ1, leu2-3, 112	40
Y2HGold	MATa, trp1-901, leu2-3, 112, ura3-52, his3-200, gal4Δ, gal80Δ, LYS2::GAL1UAS-Gal1TATA-His3, GAL2UAS-Gal2TATA-Ade2 URA3::MEL1UAS-Mel1TATA, AUR1-C MEL1	Clontech
Plasmids
pBJ113	Gene replacement vector with KG cassette; Kmr	41
pSWU19	Site-specific integration vector with Mx8 attP integration site; Kmr	42
pMAL c5X	Expression vector, maltose-binding protein fusions, the pBR322 replicon; Ampr	Biolabs
pACYC Duet-1	Expression vector, the P15A replicon; Cmr	Novagen
pYES2	Expression vector in S. cerevisiae, URA, Ampr	Invitrogen
pGADT7	Ampr; Leu	Clontech
pGADT7-T	Ampr; Leu	Clontech
pGBKT7	Kmr; Trp	Clontech
pGBKT7-p53	Kmr; Trp	Clontech
08760LR-pBJ113	MXAN_RS08760 gene replacement vector with KG cassette; Kmr	This study
08765 LR-pBJ113	MXAN_RS08765 gene replacement vector with KG cassette; Kmr	This study
08770 LR-pBJ113	MXAN_RS08770 gene replacement vector with KG cassette; Kmr	This study
08775 LR-pBJ113	MXAN_RS08775 gene replacement vector with KG cassette; Kmr	This study
08780 LR-pBJ113	MXAN_RS08780 gene replacement vector with KG cassette; Kmr	This study
24590 LR-pBJ113	MXAN_RS24590 gene replacement vector with KG cassette; Kmr	This study
36995LR-pBJ113	MXAN_RS36995 gene replacement vector with KG cassette; Kmr	This study
24595LR-pBJ113	MXAN_RS24595 gene replacement vector with KG cassette; Kmr	This study
24600LR-pBJ113	MXAN_RS24600 gene replacement vector with KG cassette; Kmr	This study
24605LR-pBJ113	MXAN_RS24605 gene replacement vector with KG cassette; Kmr	This study
pilA-08760-pSWU19	pilA promoter, MXAN_RS08760 gene insertion in pSWU19; Kmr	This study
pilA-08765-pSWU19	pilA promoter, MXAN_RS08765 gene insertion in pSWU19; Kmr
pilA-24590-pSWU19	pilA promoter, MXAN_RS24590 gene insertion in pSWU19; Kmr	This study
pilA-36995-pSWU19	pilA promoter, MXAN_RS36995 gene insertion in pSWU19; Kmr	This study
08765-pMAL c5X	Ampr, tac promoter, maltose-binding protein fusions, MXAN_RS08765 insertion in pMAL c5X	This study
36995-pMAL c5X	Ampr, tac promoter, maltose-binding protein fusions, MXAN_RS36995 insertion in pMAL c5X	This study
08760-pACYC Duet-1	Cmr, T7 promoter, His tag, MXAN_RS08760 insertion in pACYC Duet-1	This study
24590-pACYC Duet-1	Cmr, T7 promoter, His tag, MXAN_RS24590 insertion in pACYC Duet-1	This study
PKm-08760-pACYC Duet-1	Cmr, Kmr gene promoter, His tag, MXAN_RS08760 insertion in pACYC Duet-1	This study
PKm-24590-pACYC Duet-1	Cmr, Kmr gene promoter, His tag, MXAN_RS24590 insertion in pACYC Duet-1	This study
PKm-08760-NCBI-pACYC Duet-1	Cmr, Kmr gene promoter, His tag, MXAN_RS08760-NCBI insertion in pACYC Duet-1	This study
PKm-24590-NCBI-pACYC Duet-1	Cmr, Kmr gene promoter, His tag, MXAN_RS24590-NCBI insertion in pACYC Duet-1	This study
08765-pYES2	MXAN_RS08765 insertion in pYES2, URA, Ampr	This study
36995-pYES2	MXAN_RS36995 insertion in pYES2, URA, Ampr	This study
AD-08765	MXAN_RS08765 insertion in pGADT7, Ampr; Leu	This study
AD-08760	MXAN_RS08760 insertion in pGADT7, Ampr; Leu	This study
AD-36995	MXAN_RS36995 insertion in pGADT7, Ampr; Leu	This study
AD-24590	MXAN_RS24590 insertion in pGADT7, Ampr; Leu	This study
BK-08765	MXAN_RS08765 insertion in pGBKT7, Kmr; Trp	This study
BK-08760	MXAN_RS08760 insertion in pGBKT7, Kmr; Trp	This study
BK-36995	MXAN_RS36995 insertion in pGBKT7, Kmr; Trp	This study
BK-24590	MXAN_RS24590 insertion in pGBKT7, Kmr; Trp	This study

^aTet, tetracycline; Km, kanamycin; Amp, ampicillin; Cm, chloramphenicol; ^r, resistance.

Fungal strains, plasmids, and culture conditions.

The fungal strains and plasmids used in this study are listed in Table 1. S. cerevisiae 102-5B strain was used for the toxicity assay study in yeast and was cultured at 30°C in yest extract-peptone-dextrose (YPD; 10 g/liter yeast extract, 20 g/liter peptone, and 20 g/liter glucose) or synthetic complete (SC; 6.9 g/liter yeast nitrogen base complete with ammonium sulfate and amino acids) media (29). SC medium lacking the respective amino acid or nucleotide was used for auxotrophy selection and toxicity assays. For the plates, 2% agar was supplemented. The C. orbiculare ACCC36065, C. lunata ACCC38967, and S. lycopersici ACCC36106 strains (Agricultural Culture Collection of China [ACCC]) were cultivated on potato dextrose agar (PDA) medium plates and grown at 25°C. The PDA medium (6 g/liter potato powder, 20 g/liter glucose, and 20 g/liter agar) was purchased from Beijing Solarbio Science and Technology, China.

Construction of the M. xanthus mutants.

In-frame deletion of the gene was performed in M. xanthus DK1622 using positive-negative KG cassettes (30). The cassettes consisted of a Km resistance gene for positive screening and a galactokinase gene (galK) for negative screening to replace the genomic target gene by double crossover homologous recombination. Genomic DNA from DK1622 served as the template for PCR amplification of the upstream and downstream regions of target gene. The upstream and downstream regions were fused to create internal deletion fragments and cloned into the plasmid pBJ113 using the overlap PCR technique. The deletion plasmids were electroporated into DK1622, in which they integrated into the genome by single homologous recombination, and colonies grown on CTT agar plates containing Km were selected. These strains were then cultured on 1% d-Gal CTT agar plates in the absence of Km, in which the unstable tandem duplication excised the plasmid by homologous recombination. The screened strains retained either the original wild-type locus or the deleted locus, depending on where the recombination occurred. The obtained strains were further screened by colony PCR and sequencing (18).

The ΔMXAN_RS08760^R mutant and the ΔMXAN_RS24590^R mutant were generated by electroporation of the plasmid pSWU19 (containing the Km resistance gene), which is able to integrate into the Mx8 attP site of the DK1622 genome. The attR primer pair (31) was used for integration.

Complemented deletion strains were constructed. The promoter of the pilA gene (32) was amplified from the DK1622 genome and fused with the genes MXAN_RS24590, MXAN_RS36995, MXAN_RS08760, or MXAN_RS08765 using the overlapping PCR technique. Then, the connected DNA sequences were cloned into the site-specific integration plasmid pSWU19. The resulting plasmids were electroporated into the M. xanthus strains. The complementary strains were growing on CTT agar plates containing Km and further verified by colony PCR and sequencing.

All of the primers used in this study are listed in Table 2.

TABLE 2.

Primers used in this study

Primer name	Sequencea (5′–3′)
Primers for gene deletion
KO-08760-L-up	tgaattcgagctcggtacccGATGGCATACGCACTGGATGA
KO-08760-L-down	AAGAGCGTCATCACGCGG
KO-08760-R-up	taccgcgtgatgacgctcttGCCCGCGTCTTCCTCCAG
KO-08760-R-down	gtcgactctagaggatccccCCCAACATCGGCAACAGC
KO-08765-L-up	GCTCTAGACGAAGACGATGGGATTTAGG
KO-08765-L-down	AGGACGCCCGTCACCAAGCAAGAAAAACTCCCTGCCAT
KO-08765-R-up	ATGGCAGGGAGTTTTTCTTGCTTGGTGACGGGCGTCCT
KO-08765-R-down	CCCAAGCTTCATTACGGAAAGGACGGACTGA
KO-08770-L-up	tgaattcgagctcggtacccGCGTCTTCCTCCAGCATTGA
KO-08770-L-down	TACTTGCGAAACCATGGCCAAGG
KO-08770-R-up	TGGCCATGGTTTCGCAAGTACCGCCTCACCGCTC
KO-08770-R-down	gtcgactctagaggatccccCGAAGAGGACTTGGACGCAG
KO-08775-L-up	tgaattcgagctcggtacccCGTCGCCGCTGTTGCCGA
KO-08775-L-down	TTCGTTGCCAACCCATGAGCGGTGAGGCGGT
KO-08775-R-up	GCTCATGGGTTGGCAACGAAGCGAGG
KO-08775-R-down	gtcgactctagaggatccccCTGCAAGCAAGGCCGAAA
KO-08780-L-up	tgaattcgagctcggtacccCGTCCTTTCCGTAATGGCTG
KO-08780-L-down	AAGAAGACACCCAACAACCCCCTCGCTTCG
KO-08780-R-up	GGGTTGTTGGGTGTCTTCTTCCCCCTCTGG
KO-08780-R-down	gtcgactctagaggatccccCGAGGCATACCTGTTTCCCA
KO-24590-L-up	GCTCTAGAGACGATGTCGCTCCGAGGCT
KO-24590-L-down	AGCACACGAACTAGCAGCCGAGCTCGCAGTCCTCCGGACGATTAC
KO-24590-R-up	GTAATCGTCCGGAGGACTGCGAGCTCGGCTGCTAGTTCGTGTGCT
KO-24590-R-down	CCCAAGCTTTTGCAGTCCGGGCATTCG
KO-36995-L-up	GCTCTAGACCGGTCGCTCATACAATCT
KO-36995-L-down	AAGACCCCGGTAACTGAACTCTGTCCTGAAGACAAGGTACT
KO-36995-L-up	AGTACCTTGTCTTCAGGACAGAGTTCAGTTACCGGGGTCTT
KO-36995-L-down	CCCAAGCTTGACGAGACGGAGATTGCAGAAT
KO-24595-L-up	tgaattcgagctcggtacccCGCAGGAACCACAGATAGAACG
KO-24595-L-down	GGAATGAGCTCGTAAGGAGAGGATGCCCA
KO-24595-R-up	TCTCCTTACGAGCTCATTCCGCCTCCTTCA
KO-24595-R-down	gtcgactctagaggatccccAGGACGCAGGGCATACACG
KO-24600-L-up	tgaattcgagctcggtacccCGATATTGGGTAGCGGTGTTG
KO-24600-L-down	TTCCATGTGGGAATGAGCACCTCCCTGGAGG
KO-24600-R-up	GTGCTCATTCCCACATGGAACTTCACCCCC
KO-24600-R-down	gtcgactctagaggatccccTGCCCACGCTGAGGATAGAC
KO-245605-L-up	tgaattcgagctcggtacccAAGGGAAACTCGGCGTAGCG
KO-245605-L-down	ACTGACGTGAGTTCCATGTGGGCACTGAAGA
KO-245605-R-up	CACATGGAACTCACGTCAGTTGACCTGCCCC
KO-245605-R-down	gtcgactctagaggatccccTGGTGCTGCTGGCTGCGT
KO-4800-13-L-up	GGAATTCGCACTCGTCCTTGTCGTTGTCG
KO-4800-13-L-down	GCGTTCGTTTCATGGATTCTCCCACAGTCCCCCAGCGCG
KO-4800-13-R-up	CGCGCTGGGGGACTGTGGGAGAATCCATGAAACGAACGC
KO-4800-13-R-down	CCCAAGCTTCCGCACGACGACTTCCTTCA
Primers for gene complementation
com-08765-up	ATGGCCAAGGTAACAGTCAACTTC
com-08765-down	CTATTTGCAGATTGTCTTGCAGTACA
com-08760-NCBI-up	ATGTACGTGGGGCTGGAGGAAT
com-08760-NCBI-down	CTACCGCGTGATGACGCTCT
com-08760-729-up	ATGAAACCCTTTGTTGAATTTGTCG
com-08760-729-down	CTACCGCGTGATGACGCTCT
com-36995-up	ATGCCCAAGGTCTCAGTGAATG
com-36995-down	CTAGTTCGTGTGCTGACATTCCA
com-24590-NCBI-up	ATGATGCCACTGCATCTCTTCTAC
com-24590-NCBI-down	TCATCCAAGCGCTGGGGT
com-24590-729-up	ATGGAAGATCTACTCCTGAGGATCG
com-24590-729-down	TCATCCAAGCGCTGGGGT
Primers for gene expression in E. coli
36995-up	ATGCCCAAGGTCTCAGTG
36995-down	CTAGTTCGTGTGCTGACA
24590-up	ATGGAAGATCTACTCCTGAG
24590-down	TCATCCAAGCGCTGGGGTC
24590-NCBI-up	ATGATGCCACTGCATCTCTTCTAC
24590-NCBI-down	TCATCCAAGCGCTGGGGT
08765-up	ATGGCCAAGGTAACAGTCAAC
08765-down	CTATTTGCAGATTGTCTTG
08760-up	ATGAAACCCTTTGTTGAATTTGTCG
08760-down	CTACCGCGTGATGACGCTCT
08760-NCBI-up	ATGTACGTGGGGCTGGAGG
08760-NCBI-down	CTACCGCGTGATGACGCTCT
Primers for cotranscription PCR detection
27F	AGAGTTTGATCCTGGCTCAG
1492R	GGTTACCTTGTTACGACTT
Inter08780-08775-up	CTCGAAGAGGACTTGGACGCAGAC
Inter08780-08775-down	TGGCCTTGACGGCGACAACC
Inter08775-08770-up	CATCGAGCCCGAGGTGAAGAAG
Inter08775-08770-down	TGGCGAGGACTTCGTCCAACG
Inter08770-08765-up	CTCAACTTGGGCGAGGGGATG
Inter08770-08765-down	GCCGCTGTTGCCGATGTT
Inter08765-08760-up	CCTTGCCGAGCGACTATTT
Inter08765-08760-down	CGCTTGAAGCCCTTCTCC
Inter24605-24600-up	GGCACTGAGGACGCAGGGCATACA
Inter24605-24600-down	CTCCGGTATCAGCACAGGAGGA
Inter24600-24595-up	ACGCCTGACCACGGTTCTCG
Inter24600-24595-down	ATCCCGTCCGATGCAATCA
Inter24595-36995-up	GTCTGGAAGTCCACCAGCTACAGGG
Inter24595-36995-down	GATATTGGGTAGCGGTGTTGG
Inter36995-24590-up	TATCCAAGAGGCATCACCG
Inter36995-24590-down	TGAGGCATCGTGTTGACC
Primers for Y2H plasmid construction
08765AD-F	aggccagtgaattccacccgATGGCCAAGGTAACAGTCAACTTC
08765AD-R	atcccgtatcgatgcccaccCTATTTGCAGATTGTCTTGCAGTACA
08760AD-F	aggccagtgaattccacccgATGAAACCCTTTGTTGAATTTGTCG
08760AD-R	atcccgtatcgatgcccaccCTACCGCGTGATGACGCTCT
36995AD-F	aggccagtgaattccacccgATGCCCAAGGTCTCAGTGAATG
36995AD-R	atcccgtatcgatgcccaccCTAGTTCGTGTGCTGACATTCCA
24590AD-F	aggccagtgaattccacccgATGGAAGATCTACTCCTGAGGATCG
24590AD-R	atcccgtatcgatgcccaccTCATCCAAGCGCTGGGGT
08765BK-F	ccatggaggccgaattcccgATGGCCAAGGTAACAGTCAACTTC
08765BK-R	gctgcaggtcgacggatcccCTATTTGCAGATTGTCTTGCAGTACA
08760BK-F	ccatggaggccgaattcccgATGAAACCCTTTGTTGAATTTGTCG
08760BK-R	gctgcaggtcgacggatcccCTACCGCGTGATGACGCTCT
36995BK-F	ccatggaggccgaattcccgATGCCCAAGGTCTCAGTGAATG
36995BK-R	gctgcaggtcgacggatcccCTAGTTCGTGTGCTGACATTCCA
24590BK-F	ccatggaggccgaattcccgATGGAAGATCTACTCCTGAGGATCG
24590BK-R	gctgcaggtcgacggatcccTCATCCAAGCGCTGGGGT

^aRestriction sites are underlined. The lowercase parts were designed to overlap the recombination site sequences of the vector, and the uppercase sequences were for specific amplification.

RNA extraction and reverse transcription-PCR.

Total RNA from Myxococcus cells was obtained using an SV total RNA purification kit (Promega, United States). Residual genomic DNA was removed, and cDNA was synthesized from the RNA template by using the PrimeScript RT reagent kit with gDNA Eraser (TaKaRa, China). The cDNA product was used as the template to amplify a fragment containing the 3′ sequence of the upstream gene, the intergenic region, and the 5′ sequence of the downstream gene with specific primer pairs to evaluate cotranscription.

Boundary assays.

The strains were cultured in liquid CTT medium with shaking at 200 rpm. Cells were harvested by centrifugation after 20 to 24 h and suspended in TPM buffer (1.97 g/liter MgSO₄·7H₂O, 1 mM KH₂PO₄/K₂HPO₄, and 10 mM Tris HCl [pH 7.6]) at a final concentration of 5 × 10⁹ cells/ml. Then, 3-μl aliquots of the cells were inoculated on CTT agar plates at an intercolony distance of 7 mm. After 3 to 5 days of cultivation, the boundaries were observed under an SMZ18 dissection microscope (Nikon, Japan).

Toxicity assay in E. coli.

The toxicity assay was performed as described previously (33). The MXAN_RS08765 and MXAN_RS36995 genes were expressed under the control of an IPTG-inducible promoter in the pMAL c5X plasmid. The MXAN_RS08760 and MXAN_RS24590 genes were constitutively expressed from the pACYC Duet-1 plasmid, in which the IPTG-inducible promoter was replaced with a promoter of the Km resistance gene. The MXAN_RS08765-harboring pMAL c5X plasmid was cotransformed with the MXAN_RS08760-harboring pACYC Duet-1 plasmid, MXAN_RS08760-NCBI-harboring pACYC Duet-1 plasmid, MXAN_RS24590-harboring pACYC Duet-1 plasmid, or empty pACYC Duet-1 plasmid into the E. coli BL21 strain. The MXAN_RS36995-harboring pMAL c5X plasmid was cotransformed with the MXAN_RS24590-harboring pACYC Duet-1 plasmid, MXAN_RS24590-NCBI-harboring pACYC Duet-1 plasmid, MXAN_RS08760-harboring pACYC Duet-1 plasmid, or empty pACYC Duet-1 plasmid into the E. coli BL21 strain. The two empty plasmids of pMAL c5X and pACYC Duet-1 were transformed into the BL21 strain as a control. Empty pMAL c5X and MXAN_RS08760-harboring pACYC Duet-1 or MXAN_RS24590-harboring pACYC Duet-1 were introduced into the BL21 strain together as the other control. The E. coli strains were cultured overnight at 37°C in LB liquid medium supplemented with Amp and Cm. The optical density at 600 nm (OD₆₀₀) values of all strains were measured, and then the strains were adjusted to the same cell density in fresh LB liquid medium. All strains were adjusted to an OD₆₀₀ of 1 and serially diluted, resulting in tubes containing 1, 0.75, or 0.5 OD₆₀₀ units/ml. Next, 3-μl aliquots of the cell dilutions were spotted on LB agar plates containing Amp, Cm and IPTG. This assay was performed with three biological replicates each time and repeated more than three times. Photographs were obtained after incubation.

Toxicity assay in S. cerevisiae.

The MXAN_RS08765 and MXAN_RS36995 genes were expressed under the control of a Gal-inducible promoter in the pYES2 plasmids, and the two plasmids and the empty pYES2 plasmid (control) were transformed into the S. cerevisiae 102-5B strains, respectively. S. cerevisiae cells were pregrown in noninducing liquid minimal medium (SC plus 2% glucose) at 30°C. All strains were adjusted to an OD₆₀₀ of 2.5 and serially diluted, resulting in tubes containing 2.5, 2, 1.5, 1, and 0.1 OD₆₀₀ units/ml, and 5 μl was spotted onto minimal medium with agar. Heterologous protein expression was induced by the addition of Gal (0.02%). Agar plates were incubated at 30°C. This assay was performed with three biological replicates each time and was repeated more than three times. Photographs were obtained after incubation.

Coincubation experiments.

The M. xanthus strains were inoculated into liquid CTT medium, grown at 30°C with shaking, and harvested by centrifugation at 8,000 rpm for 5 min (5804R centrifuge; Eppendorf, Germany) at the mid-log phase. The cells were resuspended in TPM buffer and adjusted to a density of 10 OD₆₀₀ units/ml. Then, the cell suspensions of paired strains were mixed at a ratio of 10/1 (vol/vol, killer/prey), and 5-μl aliquots of each mixed suspension were dropped onto CTT agar medium. After 4 days of incubation, all the spots were harvested, resuspended in 500 μl TPM buffer, and serially diluted 10-fold in TPM buffer. Aliquots of 50 μl of the cell diluents were mixed with molten CTT soft agar containing 0.4% agar and poured onto CTT plates containing Km for recovery of prey cells harboring the pSWU19 plasmid. Colonies were counted after 5 days of incubation at 30°C. Three dilutions and three replicates were performed.

Antifungal assays.

The C. orbiculare ACCC36065, C. lunata ACCC38967, and S. lycopersici ACCC36106 strains were inoculated into PDA medium plates and grown at 30°C. After spore formation, the spores were resuspended in TPM buffer and adjusted to a certain concentration of spore suspension. The M. xanthus strain was inoculated into liquid CTT medium, grown at 30°C with shaking, and harvested by centrifugation at 8,000 rpm for 5 min at the mid-log phase. The cells were resuspended in TPM buffer and adjusted to a density of 8 OD₆₀₀ units/ml. Then, 20 μl of spore suspension was dropped onto 10% PDA plus TPM buffer agar medium. After they were naturally air-dried, 30-μl aliquots of different M. xanthus strain suspensions were spotted on the spore plaque, and photographs were recorded for spore germination. On the other hand, 20-μl aliquots of different M. xanthus strain suspensions were spotted on both sides of the spore plaque (the distance between the two plaques was 7 to 8 mm), and the growth of fungal hyphae was photographed.

Protein purification.

The pMAL c5X plasmid harboring the MXAN_RS36995 gene (MBP-36995) and the pACYC Duet-1 plasmid containing the MXAN_RS24590 gene (His₆-24590) were transformed into E. coli BL21(DE3) cells, and the pMAL c5X plasmid harboring the MXAN_RS08765 gene (MBP-08765) and the pACYC Duet-1 plasmid containing the MXAN_RS08760 gene (His₆-08760) were cotransformed into E. coli BL21(DE3) cells. Expression of the proteins in E. coli BL21(DE3) cells was induced by the addition of 0.1 mM IPTG when the culture reached an OD₆₀₀ value of 1.0. After an additional 20 h of incubation at 16°C, the cells were collected by centrifugation at 8,000 rpm (5804R centrifuge; Eppendorf) for 10 min and then washed with resuspension buffer (20 mM Tris HCl [pH 7.4], 200 mM NaCl, and 5% glycerol). The cells were resuspended in resuspension buffer and lysed by sonication on an ice slurry. The mixture was centrifuged at 14,000 rpm for 30 min to remove intact cells and debris. For the His₆-tagged protein, the supernatant was incubated with Ni²⁺ beads (GE Healthcare, Sweden) for 2 h at 4°C, and the beads were washed with resuspension buffer supplemented with 20 mM imidazole. The bound proteins were eluted with a 50 to 250 mM imidazole gradient in resuspension buffer. For the MBP-tagged protein, the supernatant was incubated with amylose resin (New England Biolabs) for 2 h at 4°C, and then the resin was washed with resuspension buffer. The fusion protein was eluted with resuspension buffer supplemented with 10 mM maltose. If necessary, each eluted protein was concentrated using a 3-kDa centrifugal concentrator (Millipore, Germany). The protein concentration was detected using a Nano-300 microspectrophotometer (Allsheng, China).

Pulldown assay for protein-protein interactions.

The pMAL c5X plasmid harboring the MXAN_RS36995 gene (MBP-36995) and the pACYC Duet-1 plasmid containing the MXAN_RS24590 gene (His₆-24590) were transformed into E. coli BL21(DE3) cells. Proteins were expressed, purified, and incubated together. The MBP-36995 (2.9 mg/ml) and His₆-24590 (3.3 mg/ml) proteins were mixed at the 1:1 ratio, then incubated with Ni²⁺ beads (GE Healthcare) for 2 h and washed with buffer (20 mM imidazole, 20 mM Tris HCl [pH 7.4], 200 mM NaCl, and 5% glycerol). The MBP and His₆-24590 protein mixture was incubated with Ni²⁺ beads, and the MBP-36995 protein solution incubated with Ni²⁺ beads was used as the control. The bound proteins were retained on Ni²⁺ beads, and the unbound proteins were eluted. The MBP-36995 and His₆-24590 protein mixture was incubated with amylose resin (New England Biolabs) for 2 h and washed with buffer. The MBP and His₆-24590 protein mixture was incubated with amylose resin as the control. The bound proteins were retained on amylose resin, and the unbound proteins were eluted. Additionally, the pulldown assay of the MBP-08765 and His₆-08760 proteins was analyzed using a similar approach. The bound or unbound protein solutions were then tested by SDS-PAGE and Coomassie blue staining for three biological replicates.

Yeast two-hybrid (Y2H) assay for protein-protein interactions.

A yeast two-hybrid assay was conducted using the Gal4 system. The paar-ctd and immunity gene were fused to the pGADT7 and pGBKT7 vectors to generate the prey and bait plasmids, respectively. PAAR-CTD-harboring pGADT7 and immunity gene-harboring pGBKT7 or immunity gene-harboring pGADT7 and PAAR-CTD-harboring pGBKT7 were transformed into the Y2HGold strain according to the manufacturer's instructions (Clontech). The yeast cells were grown in YPDA medium containing 1% yeast extract, 2% peptone, 2% glucose, and 0.003% adenine hemisulfate (pH 6.5) at 30°C. The synthetic dropout medium (SD medium, purchased from Beijing Coolaber, China) with no specified nutrients (SD-Trp-Leu) was used for the selection of yeast transformants. The transformants were grown on the SD-Trp-Leu medium supplemented with X-α-Gal (SD-Trp-Leu + X-α-Gal) or the SD-Trp-Leu-His-Ade medium for interaction analysis.

In vitro nuclease activity of the toxin protein.

A pMAL c5X plasmid harboring the toxin gene was constructed. Then, the toxin protein fused to the MBP tag was expressed. The immunity protein was expressed with a His₆ tag fusion using a pACYC Duet-1 plasmid. The proteins were overexpressed and purified using Ni²⁺ beads or amylose resin and used for the nuclease activity test. The DNA substrate for the in vitro assay was the genomic DNA of E. coli DH5α. Genomic DNA was extracted using the TIANamp bacteria DNA kit (Tiangen, China). Subsequently, 40 μM purified toxin protein and 0.5 μg genomic DNA were incubated for 15 min at 37°C in reaction buffer containing 20 mM Tris HCl (pH 7.4), 200 mM NaCl, and 5% glycerol. In addition, the toxin protein and immunity protein complexes were assayed under the same conditions. Digestion by the known nuclease DNase I was performed as a positive control. Both incubation with the immunity protein and incubation with a mixture of the immunity protein and DNase I were performed. The reactions were repeated more than three times. The digested DNA solutions were stopped by the addition of an equal volume of phenol-chloroform–isoamyl alcohol, and the extracted DNA solutions were examined by 0.8% agarose gel electrophoresis. The gels were stained with ethidium bromide and then photographed. A λ-HindIII-digested DNA standard marker (TaKaRa, China) was used.

Bioinformatics analysis.

The coding sequences of all the M. xanthus genes and the completely sequenced bacterial genomes were acquired from the NCBI RefSeq database (34, 35). Proteins were classified by the CDD and PFAM databases (36, 37). The domain architectures of proteins were analyzed by RPS-BLAST in the CDD database (50,369 position-specific scoring matrices [PSSMs]) under the retrieval condition of an E value of less than 0.01. Multiple sequence alignments of the full-length protein sequences were implemented using MAFFT (38). A phylogenetic tree was generated using the MEGA 7 neighbor-joining method with 1,000 bootstrap replicates (39).