Cross-species activation of hydrogen cyanide production by a promiscuous quorum-sensing receptor promotes Chromobacterium subtsugae competition in a dual-species model

Many saprophytic bacteria have LuxR-I-type acyl-homoserine lactone (AHL) quorum-sensing systems that may be important for competing with other bacteria in complex soil communities. LuxR AHL receptors specifically interact with cognate AHLs to cause changes in expression of target genes. Some LuxR-type AHL receptors have relaxed specificity and are responsive to non-cognate AHLs. These promiscuous receptors might be used to sense and respond to AHLs produced by other bacteria by eavesdropping. We are interested in understanding the role of eavesdropping during interspecies competition. The soil saprophyte Chromobacterium subtsugae has a single AHL circuit, CviR-I, which produces and responds to N-hexanoyl-HSL (C6-HSL). The AHL receptor CviR can respond to a variety of AHLs in addition to C6-HSL. In prior studies we have utilized a coculture model with C. subtsugae and another soil saprophyte, Burkholderia thailandensis . Using this model, we previously showed that promiscuous activation of CviR by B. thailandensis AHLs provides a competitive advantage to C. subtsugae . Here, we show that B. thailandensis AHLs activate transcription of dozens of genes in C. subtsugae, including the hcnABC genes coding for production of hydrogen cyanide. We show that hydrogen cyanide production is population density-dependent and demonstrate that the cross-induction of hydrogen cyanide by B. thailandensis AHLs provides a competitive advantage to C. subtsugae . Our results provide new information on C. subtsugae quorum sensing and are the basis for future studies aimed at understanding the role of eavesdropping in interspecies competition.


INTRODUCTION
Quorum sensing is a population density-dependent signalling system that relies on small diffusible signalling molecules. Once a critical population threshold is reached, the molecules interact with cytoplasmic transcription factors to cause activation of specific genes and coordinate gene expression across the population. In Proteobacteria, one type of quorum sensing is carried out by acyl-homoserine lactone (AHL) signals. These molecules have a lactone ring and a fatty acid tail that varies in length from 4 to 20 carbons and can also have modifications at the third carbon position, which may be unsubstituted, hydroxylated, or have an oxo-substitution. Typical AHL quorum-sensing gene circuits consist of a LuxR-family signal receptor and a LuxI-family signal OPEN ACCESS synthase, which produces a cognate signal. A bacterium may possess multiple LuxR-I circuits that typically produce and respond to different AHLs. Although the AHLs generally have similar structures, the receptors are thought to interact with AHLs in a highly specific and selective manner [1]. However, there are exceptions such as the CviR receptor of Chromobacterium subtsugae (formerly C. violaceum), which can promiscuously detect and respond to a range of AHLs [2,3]. The existence of such receptors suggests that receptor promiscuity might have certain advantages in some conditions. Quorum-sensing systems control different behaviours in different bacteria, such as the production of toxins, exoproducts and biofilm matrix components. Many of these behaviours might provide a benefit to individuals within groups or complex communities. For example, toxin production might be important for competition with other strains or species of bacteria. There is an increasing interest in understanding the specific benefits of quorum sensing in dynamic polymicrobial communities. A major barrier to understanding quorum sensing in this context is the need for laboratory models that enable studies of mixed-strain and mixed-species interactions in a controlled setting. Such models are increasingly being utilized, and are providing significant new insights into the role and mechanisms of quorum sensing in mixed microbial communities (for a review, see [4]).
This study utilizes a laboratory dual-species model that was previously developed where two soil bacteria, Burkholderia thailandensis and C. subtsugae, use AHL-dependent quorum sensing to compete with the other species ( Fig. 1) [5]. B. thailandensis

Impact Statement
In quorum sensing, population density-dependent changes in gene regulation are the result of a cytoplasmic transcription regulator binding to a quorum sensing signal. The signal-receptor interaction is considered to be specific to ensure fidelity of the system. However, some quorum-sensing receptor proteins have relaxed specificity and can recognize and respond to a range of signals. These promiscuous receptors might provide some benefit by enabling interspecies activation of quorum sensing by 'eavesdropping', although the potential benefits of eavesdropping are not well studied. The current study utilizes a dual-species laboratory competition model, where one species has a promiscuous signal receptor and can respond to signals produced by the other species. In our study, we identify the signals that enable quorum sensing cross-talk and show that cross-talk promotes competition by inducing hydrogen cyanide production. Our results highlight how quorum sensing-enabled interspecies cross-talk might provide an advantage during competition and inform our understanding of how receptor-signal pairs might evolve in natural environments. Fig. 1. Burkholderia thailandensis-Chromobacterium subtsugae coculture model. The B. thailandensis quorum-sensing system BtaR2-I2 produces and responds to 3OHC8-HSL and 3OHC10-HSL, and activates production of bactobolin antibiotic. B. thailandensis also has two other quorum-sensing systems, BtaR1-I1 and BtaR3-I3, which produce C8-HSL and 3OHC8-HSL, respectively. BtaR2-I2 is important for B. thailandensis to compete with C. subtsugae due to activation of bactobolin production. The C. subtsugae quorum-sensing system CviR-I produces and responds to C6-HSL. CviR-I is important for C. subtsugae to compete with B. thailandensis, through an unknown mechanism. produces a ribosome-targeting antibiotic called bactobolin [6]. Bactobolin is regulated by a LuxR-I system called BtaR2-I2, which senses and responds to 3OHC8-HSL and 3OHC10-HSL [7]. Both bactobolin and the BtaR2-I2 system are important for B. thailandensis to compete with C. subtsugae [5]. In addition to BtaR2-I2, B. thailandensis has two other complete quorum-sensing circuits, BtaR1-I1 and BtaR3-I3, which produce and respond to C8-HSL and 3OHC10-HSL, respectively [8]. C. subtsugae has a single LuxR-I-type system, CviR-I, which produces and responds to C6-HSL [2]. CviR-I is important for competition with B. thailandensis [5]. CviR is known to activate production of a purple pigmented antibiotic, violacein, although mutational disruption of violacein had no effect on competition [5]. We also showed that CviR can be activated by B. thailandensis AHLs by 'eavesdropping' to promote the competitive ability of C. subtsugae [5].
In this study, we identify C8-HSL as the B. thailandensis AHL best able to activate C. subtsugae production of quorum-dependent antimicrobials. We use RNAseq to identify the full set of C. subtsugae genes that can be activated by either C8-HSL or the native signal, C6-HSL. Among the genes most highly activated by both signals are those coding for hydrogen cyanide biosynthesis. We use our coculture model to validate the importance of hydrogen cyanide for C. subtsugae competition. We also demonstrate that the hydrogen cyanide genes can be transcriptionally activated by CviR and both native and non-native signals. The results provide new information on promiscuous AHL responses in C. subtsugae and further support the idea that AHLs drive important interactions through interspecies cross-talk.

Bacterial culture conditions and reagents
All strains were grown in lysogeny broth (LB) (10 g tryptone, 5 g yeast extract and 10 g NaCl in 1 l water), LB with morpholinepropanesulfonic acid (LB-MOPS, 50 mM; pH 7), or in LB with 1.5 % (w/v) Bacto-Agar. All broth cultures were incubated at 30 °C (C. subtsugae or C. subtsugae-B. thailandensis cocultures) or 37 °C (B. thailandensis or Escherichia coli). Synthetic acyl-homoserine lactone signals were purchased from Cayman Chemicals (MI, USA) and stored in ethyl acetate acidified with 0.1 % of glacial acetic acid. In all cases, synthetic signals were added to the culture flask and dried down completely using a stream of nitrogen gas prior to addition of the culture media. For strain constructions, we used gentamicin at 50 µg ml −1 (C. subtsugae) or 15 µg ml −1 (E. coli). For selection from cocultures, we used gentamicin at 100 µg ml −1 (B. thailandensis) and trimethoprim at 100 µg ml −1 (C. subtsugae).
Genomic DNA, PCR and DNA fragments and plasmid DNA were purified by using a Puregene Core A kit, plasmid purification miniprep kit, or PCR cleanup/gel extraction kits (Qiagen or IBI-MidSci) according to the manufacturer's protocol. RNA was isolated using the Qiagen RNEasy Minikit.

Bacterial strains and plasmids
The strains are summarized in Table 1. For cocultures, we used B. thailandensis strains JBT125 and BD20. Both are derived from strain E264 [9]; BD20 is an bactobolin-deficient btaK mutant [7], and JBT125 is an AHL-and bactobolin-deficient bta1-3, btaK quadruple mutant [8]. C. subtsugae strain CV017 (referred to as wild-type, previously known as C. violaceum CV017) is a derivative of strain ATCC 31532 with a transposon insertion in gene CV_RS05185 causing overexpression of violacein [2]. All C. subtsugae mutant strains were constructed from CV017 using allelic exchange and the methods described below. For plasmid construction, we used E. coli strain DH10B (Invitrogen).
Unmarked in-frame deletions of cviR, cviI and hcnB (Cv017_06925) were constructed using the following method. DNA fragments were generated by PCR or DNA synthesis (GenScript, Piscataway, NJ, USA) containing ~500-1000 bp DNA flanking each gene and fused together, creating an unmarked, non-polar deletion of each gene with incorporated XbaI and SacI restriction enzymes sites. These fragments were digested with XbaI and SacI and cloned into XbaI, SacI-digested pEXG2 [10]. The relevant pEXG2 gene deletion plasmid was subsequently used to make the mutant C. subtsugae strain using previously described methods [5]. Briefly, the relevant pEXG2 plasmid was transformed into C. subtsugae by electroporation or conjugation. Merodiploids were selected on LB agar containing gentamicin, and deletion mutants were counterselected on LB+15 % sucrose. Mutant strains were verified by testing for gentamicin sensitivity and by PCR-amplifying the deletion region and sequencing the PCR product.

RNA isolation
To isolate RNA for RNAseq, cultures were grown to an OD 600 of 4.0 in 10 ml LB-MOPS in 125 ml non-baffled flasks. Bacteria (~1×10 9 cells) were removed from the cultures and suspended in RNAprotect bacteria reagent (Qiagen), pelleted by centrifugation and stored at −80 °C. Thawed cells were suspended in 1 ml RLT buffer (Qiagen) containing 2-mercaptoethanol and lysed by bead beating. RNA was purified using the RNAeasy minikit (Qiagen). Contaminating DNA was removed with Turbo DNase (Ambion), and RNA was obtained by using RNeasy MinElute cleanup kit (Qiagen).

RNA-seq library construction, sequencing, mapping and analysis
Libraries for RNA-seq were prepared by Novogene using the Novogene mRNA-seq services project workflow. cDNA libraries were sequenced by Novogene using an Illumina platform and PE-150 reads. Sequences were aligned and mapped to the C. subtsugae CV017 genome (accession NZ_JAHDTB000000000) using Featurecounts and the Rsubread package in R [12]. We determined differentially regulated genes for biological replicates using function edgeR [13] and using a false discovery rate (FDR) cutoff of 0.05. We proceeded with genes with >15 aligned reads and 2-fold or more regulation by C6-or C8-HSL relative to no AHL. Functions and their sources can be found at https://github.com/ChandlerLabKU/Cs_rnaseq. The data have been deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) [BioProject identification (ID) PRJNA894552]. For Fig. 6, promoter sites and DNA recognition motifs were predicted using RSAT [14] and the transcription start site in the DNA region upstream of the C. subtsugae hcnA gene was predicted from the RNAseq data generated in this study using the TSSr program in R [15].

Coculture experiments
Coculture experiments were conducted in 20 ml LB-MOPS medium in 125 ml non-baffled flasks. The inoculum was from logarithmic-phase pure cultures of C. subtsugae and B. thailandensis. The initial OD 600 in the coculture was 0.05 for B. thailandensis (2-4×10 7 cells ml −1 ) and 0.005 for C. subtsugae (2-4×10 6 cells ml −1 ). After inoculation, cocultures were incubated at 30 °C with shaking at 250 r.p.m. for 24 h. Colony-forming units (c.f.u.) of each species were determined by using differential antibiotic selection on LB agar plates. B. thailandensis was selected with gentamicin and C. subtsugae was selected with trimethoprim.

Transcription reporter assays
To assess activation of the hcnA promoter in C. subtsugae, overnight cultures of C. subtsugae strains carrying the pMP220_hcn-lacZ reporter plasmid were used as starters by diluting to an OD 600 of 0.05. When experimental cultures reached an OD 600 of 0.5  [11] pMP220_hcn-lacZ pMP220 with the hcnA promoter extending from −1 to −500 relative to the translation start site, Tc R This study *Tc R , tetracycline-resistant; Gm R , gentamicin-resistant.
they were distributed to 2 ml Eppendorf tubes containing different concentrations of dried C6-HSL, C8-HSL, 3OHC8-HSL and 3OHC10-HSL. The volume in each tube was 0.5 ml. After 5 h with shaking at 30 °C, β-galactosidase activity was measured using the Tropix Galacto-Light Plus chemiluminescent kit according to the manufacturer's protocol (Applied Biosystems, Foster City, CA, USA).

Cyanide measurements
Cyanide concentrations were measured using a CN − ion-selective electrode (Cole-Parmer, USA). Cells were clarified from culture fluid using centrifugation followed by filter sterilization and the cell-free fluid was brought to a pH of 12 using NaOH prior to measuring the CN − ion by conductivity. A standard curve using KCN was similarly measured and used to calculate the CN − concentration in samples.

C. subtsugae response to B. thailandensis AHLs in coculture
Previously, we showed that extracted fluid from B. thailandensis cultures containing all three B. thailandensis AHLs (C8-HSL, 3OHC8-HSL and 3OHC10-HSL) increases C. subtsugae competitiveness in a manner dependent on the C. subtsugae quorumsensing receptor CviR [5]. These results suggest that C. subtsugae CviR is responsive to one or several of the B. thailandensis AHLs. Therefore, we carried out coculture experiments to determine the individual contribution of each of the B. thailandensis AHL(s) to C. subtsugae competitiveness. We used signal synthase mutants of C. subtsugae and B. thailandensis for the competition experiment and added each of the B. thailandensis synthetic AHLs individually at 1 uM AHLs, which is a concentration in the range of that reported for similarly grown pure cultures of B. thailandensis [16]. The B. thailandensis strain JBT125 was also bactobolin-defective (ΔbtaK) so that C. subtsugae would not be killed by AHL-induced bactobolin in the experiment. In our coculture experiment we observed that the C. subtsugae native signal C6-HSL improved C. subtsugae competitiveness by ~100-fold compared with the no signal condition (Fig. 2), similar to our prior study [5]. Adding C8-HSL and 3OHC8-HSL also significantly improved competitiveness of C. subtsugae, although to a lesser degree than the native C6-HSL. 3OHC10-HSL had a more variable effect than the other AHLs and was not significantly different from no AHLs, although this signal also appeared to improve competitiveness. Together, these results show that C. subtsugae is responsive to the B. thailandensis AHLs C8-HSL and 3OHC8-HSL, and possibly also to 3OHC10-HSL.

Production of hydrogen cyanide is induced by quorum sensing and hcnB
We used RNAseq transcriptomic analysis to identify AHL-activated C. subtsugae genes as a first step to identify the factor important for competition with B. thailandensis. We compared transcripts in our ΔcviI strain grown with or without exogenously added synthetic AHLs. We grew cultures with either C6-HSL or C8-HSL, with the goal of identifying a set of genes with the ability to respond to both signals. We anticipated that the gene(s) required for C. subtsugae to compete via eavesdropping would be among this gene set. We collected cells for our analysis during the transition from logarithmic growth to stationary phase, which corresponds to an optical density at 600 nm (OD 600 ) of 4.0. AHLs were added at 2 µM to ensure robust responses for detecting differences in gene expression.
Overall, 348 C. subtsugae genes were activated by C6-HSL, and 97 of these genes were also activated by C8-HSL (Table S1, available in the online version of this article). In the set of genes most highly activated by both C6-HSL and C8-HSL (Table 2), we identified hcnA, hcnB and hcnC, which are predicted to code for biosynthesis of hydrogen cyanide. Hydrogen cyanide is important for C. subtsugae to protect itself from predation by other bacteria and kill mosquito larvae [17,18], and is also important for Pseudomonas aeruginosa to compete with B. multivorans in cocultures [19]. Therefore, we focused our attention on hydrogen cyanide. To determine whether C. subtsugae produces hydrogen cyanide, we measured cyanide ion (CN − ) in cell-free fluid from wild-type C. subtsugae cultures grown to various growth stages (early logarithmic, late logarithmic, stationary phase). The maximum concentration of CN − was 3.3±1.1 mM at an OD 600 of ~8 (Fig. 3a). When adjusted for growth, the production of CN − was population density-dependent (Fig. 3a), consistent with quorum-sensing activation of these genes. To determine if either the hcn genes or quorum sensing was required for hydrogen cyanide production, we introduced a ΔhcnB mutation to the wild-type genome and measured CN − in culture fluid of the ΔhcnB and ΔcviR mutants and wild-type. Growth-adjusted CN − levels were low in both ΔhcnB and ΔcviR culture fluid compared with that of the wild-type (Fig. 3b). These results support the conclusion that hcnB and quorum sensing are important for hydrogen cyanide production in C. subtsugae.

Cyanide is important for C. subtsugae to compete with B. thailandensis
Because expression of the hydrogen cyanide biosynthesis genes is responsive to both C6-HSL and C8-HSL, we hypothesized that B. thailandensis AHLs increase the competitiveness of C. subtsugae by activating hydrogen cyanide production through eavesdropping. As a first test of this hypothesis, we determined the susceptibility of B. thailandensis strain BD20 to a range of concentrations of potassium cyanide (0-750 uM), which mimics cyanide ions [20]. We determined the 50 % lethal dose (LD 50 ) of potassium cyanide for B. thailandensis BD20 to be 119 µM (Fig. 4a). Thus, the concentration of cyanide produced by C. subtsugae (~3 mM, Fig. 3a) is more than sufficient to kill B. thailandensis.
Next, we investigated the effects of hydrogen cyanide in cocultures (Fig. 4b). We competed the bactobolin-deficient B. thailandensis (BD20) with different genetic mutants of C. subtsugae. In these conditions, B. thailandensis BD20 produces AHLs that can activate the quorum-sensing system of C. subtsugae [5]. Consistent with prior results [5], in these conditions the C. subtsugae ΔcviI mutant and wild-type are similarly competitive, because the ΔcviI mutant has CviR that can be activated by B. thailandensis AHLs. The ΔcviR mutant cannot respond to any AHLs and, as expected, was significantly less competitive than the wild-type and ΔcviI mutant. To determine the role of hydrogen cyanide in competition, we tested the competitive ability of the ΔhcnB mutant. The ΔhcnB single mutant competed poorly with BD20 and was similar to the ΔcviR mutant. These results support that hydrogen cyanide is important for the competitive ability of C. subtsugae. We also tested the ΔcviI, ΔhcnB double mutant. In this strain, the competitive ability of C. subtsugae relies on AHLs produced by B. thailandensis. The ΔcviI, ΔhcnB double mutant also competed poorly, similar to the ΔcviR and ΔhcnB single mutants. These results support that hydrogen cyanide is important for C. subtsugae to compete in response to B. thailandensis AHLs. Together, the results support the idea that hydrogen cyanide is required for the competitive advantage provided to C. subtsugae in response to native AHLs and non-native AHLs.

Sensitivity of hcnA promoter to AHLs
The hydrogen cyanide biosynthesis genes hcnA, hcnB and hcnC (the hcn genes) are adjacently encoded in the genome facing the same direction. There is a single putative promoter upstream of hcnA. We hypothesized that this promoter is activated either directly or indirectly by CviR in response to the C. subtsugae-produced signal C6-HSL, as well as the B. thailandensis signals C8-HSL, 3OHC8-HSL and possibly 3OHC10-HSL. To test our hypothesis, we created a plasmid with a DNA fragment containing the putative promoter region upstream of hcnA fused to a promoterless lacZ gene. This plasmid was introduced into the C. subtsugae ΔcviI and ΔcviR mutants. The hcnA-lacZ reporter was activated by 5 µM of each of the four AHLs (C6-, C8-, 3OHC8-and 3OHC10-HSL) in the ΔcviI mutant, but not in the ΔcviR mutant (Fig. 5a). Thus, activation of hcnA-lacZ required CviR and AHLs. We also determined the signal sensitivity of the hcnA-lacZ reporter by testing each AHL at a range of concentrations (Fig. 5b).
The hcnA-lacZ reporter responded most sensitively to the native signal C6-HSL. There was also a response to C8-HSL. The other two AHLs, 3OHC8-HSL and 3OHC10-HSL, only activated the reporter at the highest signal concentrations (1-5 µM). These results were consistent with the coculture results (Fig. 2), and demonstrate that C8-HSL and, to a lesser extent, 3OHC8-HSL and 3OHC10-HSL, can activate the hcn gene promoter in a manner dependent on CviR.

DISCUSSION
The B. thailandensis-C. subtsugae model system was developed previously to examine the effects of quorum sensing and AHLdependent eavesdropping in competition [5]. In those studies, we elucidated the importance of B. thailandensis bactobolin for competition, although the C. subtsugae toxin remained elusive. Here, we identified hydrogen cyanide as the C. subtsugae toxin, filling in this gap in knowledge. We have also shown that noncognate AHLs can induce dozens of genes in addition to those coding for hydrogen cyanide. The results with our coculture model support the idea that promiscuous receptors might enable interspecies interactions by coordinating broad changes in gene expression in response to AHLs from other species. The results of our studies also expand the list of genes known to be activated by the CviR-I system in C. subtsugae.
The C. subtsugae CviR receptor can activate certain genes in response to a wide range of AHLs [2]. Although detailed structurefunction studies of CviR and different AHLs have been carried out [21,22], few studies have addressed the potential benefits of AHL receptor promiscuity for the cell. Receptor activation by non-self AHLs could enable detection of other bacteria and might provide certain benefits for interspecies competition [5] or cooperation [23]. Indeed, there is evidence of eavesdropping in natural polymicrobial communities in plants [23][24][25], and eavesdropping has also been shown to occur in communities relevant to infections [26]. Recent studies also support the idea that promiscuity might be a feature of a wider range of receptors than previously recognized [27]. These studies support the idea that AHL-dependent cross-talk might be an important but underrecognized facet of quorum-sensing communication and that it warrants further study.
We find it interesting that in our study, the majority of genes most strongly activated by both C6-HSL and C8-HSL in C. subtsugae are predicted to code for antibiotic production ( Table 2). The receptor BtaR2 in B. thailandensis is also highly promiscuous and regulates production of a potent and broad-spectrum antibiotic, bactobolin [6,7]. The role of these receptors in regulating antibiotic production supports the idea that promiscuous activation of receptors could play an important role in competition.
The promiscuous activation of CviR by C8-HSL leads to the induction of a subset of genes that overlaps with that controlled by C6-HSL. It may be that these genes are more highly sensitive to CviR, enabling response to CviR activation by non-native signals. An alternative possibility is that C8-HSL changes CviR recognition of gene promoters, possibly through allosteric changes in the DNA binding motif. Future studies focused on the regulation of particular genes by different AHLs could help to elucidate the mechanism of CviR activation of gene promoters by non-native AHLs.
Prior studies show that receptor selectivity falls on a continuum from highly selective (e.g. P. aeruginosa RhlR) to intermediate (e.g. P. aeruginosa LasR) to highly promiscuous (e.g. CviR and B. thailandensis BtaR2) [27]. The affinity of receptors for a particular AHL can be influenced by changing specific residues within the AHL binding pocket of the receptor [21,[28][29][30]. For example, although CviR is normally inhibited by the noncognate signal C10-HSL, replacing methionine at position 89 with serine converts this receptor to become C10-HSL-responsive [21]. These results suggest that a particular receptor could easily adapt to become more or less promiscuous, dependent on the benefits of promiscuity. In addition, receptor selectivity is influenced by receptor expression levels [27]. Although very little is known about the environmental conditions that regulate quorum-sensing receptors, one can imagine that conditions that increase receptor levels in the cell could increase AHL promiscuity. For example, it was recently demonstrated that the CviR receptor is activated by certain antibiotics [31], although the effects of antibiotics on receptor selectivity remain to be tested.
The results of this study are consistent with those of prior studies [17,19,32], supporting a role for hydrogen cyanide in competition. In P. aeruginosa, the hydrogen cyanide biosynthesis genes are part of an operon driven by a promoter upstream of hcnA (see Fig. 6). The P. aeruginosa hcnA promoter is quorum-sensing controlled, and has an experimentally validated binding site for the quorum-sensing receptor LasR at position −101 relative to the translation start site [33]. This region also has a binding site for the anaerobic regulator (Anr) at position −72 (Fig. 6) [33,34]. The intergenic regions upstream of hcnA in P. aeruginosa and C. subtsugae share 45 % identity; however, we were unable to find evidence of a lux box or Anr-binding site anywhere in the C. subtsugae hcnA promoter. It is possible that CviR can recognize the hcnA promoter in the absence of a recognizable lux box, which has been reported in some cases for LasR [33]. It is also possible that CviR controls the hcn genes indirectly. Either way, our results provide additional support for the idea that CviR can activate production of hydrogen cyanide in response to AHL from other species, supporting the idea that some quorum sensing-dependent competitive behaviours might be induced by interspecies cross-talk.

Funding information
This work was supported by the National Institutes of Health (NIH) R35GM133572 to J.R.C. S.B. and C.L. were supported by NIH K-INBRE fellowships (P20GM103418), K.C.E. was supported by the NIH Chemical Biology Training Program (T32 GM08545) and P.K. was supported by the KU Weaver Fellowship. We also thank the KU Genomics Core, which is supported by NIH grants P20GM103638 (COBRE CMADP) and P20GM103418 (K-INBRE).