BysR, a LysR-Type Pleiotropic Regulator, Controls Production of Occidiofungin by Activating the LuxR-Type Transcriptional Regulator AmbR1 in Burkholderia sp. Strain JP2-270

We report for the first time that occidiofungin production is regulated by the global transcriptional factor BysR, by directly targeting the specific regulator ambR1, which further promotes the transcription of ocf genes. BysR also acts as a pleiotropic regulator that controls various cellular processes in Burkholderia sp. strain JP2-270. ABSTRACT Occidiofungin is a highly effective antifungal glycopeptide produced by certain Burkholderia strains. The ocf gene cluster, responsible for occidiofungin biosynthesis, is regulated by the cluster-specific regulators encoded by an ambR homolog(s) within the same gene cluster, while the extent to which occidiofungin biosynthesis is connected with the core regulation network remains unknown. Here, we report that the LysR-type regulator BysR acts as a pleiotropic regulator and is essential for occidiofungin biosynthesis. Magnaporthe oryzae was used as an antifungal target in this study, and deletion of bysR and ocfE abolished the antagonistic activity against M. oryzae in Burkholderia sp. strain JP2-270. The ΔbysR defect can be recovered by constitutively expressing bysR or ambR1, but not ambR2. Electrophoretic mobility shift assays (EMSAs) collectively showed that BysR regulates ambR1 by directly binding to its promoter region. In addition, transcriptomic analysis revealed altered expression of 350 genes in response to bysR deletion, and the genes engaged in flagellar assembly and bacterial chemotaxis constitute the most enriched pathways. Also, 400 putative BysR-targeted loci were identified by DNA affinity purification sequencing (DAP-seq) in JP2-270. These loci include not only genes engaged in key metabolic pathways but also those involved in secondary metabolic pathways. To conclude, the occidiofungin produced by JP2-270 is the main substance inhibiting M. oryzae, and BysR controls occidiofungin production by directly targeting ambR1, an intracluster transcriptional regulatory gene that further activates the transcription of the ocf gene cluster. IMPORTANCE We report for the first time that occidiofungin production is regulated by the global transcriptional factor BysR, by directly targeting the specific regulator ambR1, which further promotes the transcription of ocf genes. BysR also acts as a pleiotropic regulator that controls various cellular processes in Burkholderia sp. strain JP2-270. This study provides insight into the regulatory mechanism of occidiofungin synthesis and enhances our understanding of the regulatory patterns of the LysR-type regulator.

Occidiofungin, a glycopeptide, was first identified from Burkholderia contaminans MS14 (8), which possesses significant antifungal and antiparasitic activities (2,8,9). Occidiofungin is biosynthesized by ocfD, ocfE, ocfF, ocfH, and ocfJ, which encode nonribosomal peptide synthetases (10,11). Previous comparative genomics analysis suggested that the occidiofungin biosynthesis genes ocfD to ocfJ constitute a gene cluster and are present in plant growth-promoting Burkholderia strains but absent in pathogenic strains or non-plant-associated soil strains (12). The expression of ocfD to ocfJ is suggested to be specifically regulated by two LuxR-type regulatory genes, ambR1 and ambR2, located in the gene cluster with ocfD to ocfJ. AmbR1 plays a more critical role than AmbR2 (10,11,13). However, the integration mechanism of this secondary metabolism pathway with the core regulation network remains unknown.
In the elite biocontrol strain Burkholderia sp. strain JP2-270, we recently identified a transcriptional regulator, BysR, that is essential for the fungal-inhibitory activity of JP2-270 (14). BysR belongs to the LysR-type transcriptional regulators (LTTRs). LTTRs are the most common kind of transcription regulators in prokaryotes and can act as positive or negative modulators for various genes involved in symbiotic nitrogen fixation, virulence, quorum sensing, motility, and secondary metabolism (15)(16)(17)(18)(19)(20)(21)(22)(23). The LTTR StgR inhibits the production of actinorhodin and prodigiosin by interacting with their pathway-specific regulators in Streptomyces coelicolor (23), while the LTTR FinR was recently reported to serve as an activator of phenazine and pyoluteorin biosynthesis in Pseudomonas chlororaphis G05 (22). In addition, ScmR was identified as a global LTTR controlling virulence and diverse secondary metabolites in Burkholderia thailandensis (17).
Our work aimed to investigate the novel LTTR BysR, which is essential for the antifungal activity of Burkholderia sp. strain JP2-270. By inactivating ocfE, we demonstrated that occidiofungin produced by JP2-270 mainly contributes to the suppression activity against M. oryzae. In addition, we found that BysR is required to inhibit the mycelial growth of M. oryzae. To understand the functions and regulatory mechanisms of BysR in occidiofungin production, as well as other pathways, we performed electrophoretic mobility shift assays (EMSAs), RNA sequencing (RNA-seq), and DNA affinity purification sequencing (DAP-seq). The EMSAs showed that occidiofungin synthesis was regulated by BysR, which directly targeted ambR1, a pathway-specific regulatory gene of occidiofungin production. Aside from secondary metabolism, a series of regulatory targets were identified by DAP-seq, including genes engaged in core cellular processes and secondary metabolites. In general, our results identified a novel upstream component of the occidiofungin synthesis regulatory network and provided new insight into the regulatory mechanisms of occidiofungin synthesis. The regulatory models have been proposed to help us understand the global regulatory role of the LTTR BysR in diverse cellular processes.

RESULTS
BysR displays conserved domains of LTTR regulators. bysR (DM992_17470) has an open reading frame of 984 bp and encodes a putative 327-amino-acid protein. The phylogenetic analysis of BysR with its most relevant LTTRs revealed that BysR is a novel LTTR in the genus Burkholderia, and the protein that is most homologous to BysR is BcaI3178 of Burkholderia cenocepacia H111, showing 93% identity (96% similarity) (see Fig. S1A in the supplemental material). The LTTRs associated with secondary metabolite synthesis are ScmR of Burkholderia thailandensis and ShvR of Burkholderia cenocepacia (17,24) (Fig. S1A). A domain homology search with Pfam (25) revealed a typical helix-turn-helix (HTH) DNA binding domain (Pfam accession number PF00126) at the N terminus and a typical C-terminal receiver domain (PF03466) (Fig. S1B). In addition, the homology comparison at the amino acid level between BysR and other identified LTTRs revealed that the high degree of similarity was within the N-terminal helix-turn-helix domain involved in binding DNA and the less conserved region was in the C terminus, where the coinducer domain was located (Fig. S1C).
BysR positively regulates multiple pathways in Burkholderia sp. strain JP2-270. Here, we performed RNA-seq transcriptome analysis to identify the potential targets regulated by BysR. The good grouping in principal-component analysis (PCA) and high Pearson correlation coefficient indicated that the data had good discriminability and reproducibility ( Fig. 1A and Fig. S2). Furthermore, the expression of some genes was similar in RNA-seq and quantitative reverse transcription-PCR (qRT-PCR) (Fig. S3), suggesting that the data were reliable. RNA-seq analysis revealed 350 differentially expressed genes (DEGs), consisting of 48 upregulated and 302 downregulated genes [with adjusted P (P adj )  Table S2)]. Among the 12 most significant downregulated genes, 9 were related to occidiofungin production (DM992_33325, DM992_33335, DM992_33345, DM992_33360 to DM992_33380, and DM992_33390). As shown in Fig. 2A, the FPKM (fragments per kilobase of transcript sequence per million base pairs sequenced) values of genes related to occidiofungin synthesis in a DbysR mutant were significantly lower than those in wild-type JP2-270, and these genes were dramatically downregulated 3.6-to 7.9-fold in the DbysR mutant (Fig. 1B). Among them, besides the dramatically affected occidiofungin production-associated gene cluster, four other secondary metabolite biosynthesis gene clusters, including an unknown nonribosomal peptide synthetase (NRPS) type gene cluster (DM992_33030 to DM992_33110, region 3.  identified ( Fig. 2A and Table S2). These results suggest that BysR regulates the production of secondary metabolites in Burkholderia sp. strain JP2-270. Otherwise, among the upregulated genes, those responsible for the biosynthesis of pyrroloquinoline quinone (PQQ) were affected significantly (Table S2), suggesting that BysR might act as a negative regulator of PQQ production. Moreover, KEGG pathway enrichment analysis of DEGs revealed that a variety of cellular processes involved in carbon metabolism, chloroalkane and chloroalkene degradation, two-component systems, chemotaxis, and flagellar assembly were significantly enriched ( Fig. 1C and Table S3). Notably, 20 of the 38 genes engaged in flagellar assembly were downregulated in the DbysR mutant ( Fig. 1D; Tables S2 and S3), indicating that BysR plays an important role in regulating flagellar assembly.
The DEGs identified by RNA-seq include genes that are regulated directly and indirectly by BysR. To verify the genes that are directly regulated by BysR, we used DNA affinity purification sequencing analysis (DAP-seq) for genome-wide recognition of BysR binding sites in vitro (26). After affinity purification and sequencing, at least 22 million double-end reads per sample were generated, and the ones with .99% of reads were uniquely mapped to the JP2-270 genome. A total of 400 enriched common peaks of two replicates with a 2log 10 (P value) of $2 were called (Table S4). The mean width of DAP-seq peaks was ,1,000 bp (Fig. 3A). In total, 89.6% of BysR binding peaks were distributed in the promoter, promotertranscription start site (TSS), or intergenic regions, while only 8.3% and 2.3% were in the transcription termination site (TTS) and exon regions, respectively (Fig. 3B). Among the identified peaks, 232 (58%) were located in the 2700 bp-to-100 bp regions, as determined by analysis of peak summit positions relative to the start codons of JP2-270 open reading frames (or the first gene in operons) ( Fig. 3C; Table S4). These results suggest that the expression of these loci might be directly regulated by BysR. Among the 232 peaks, 134 peaks with a length greater than 500 bp were used for motif prediction using MEME-ChIP (27) and the conserved-motif-like AT-N 11 -AT box (E value = 3.8e2021) was found in 95 of 134 peaks (71%) ( Fig. 3D and Table S5). This conserved A1T-rich box was similar to the LTTR OxyR consensus binding sequence (28), and the typical motif of LTTRs (T-N 11 -A) was also recognized in this conserved box (18). Thus, we propose that the AT-N 11 -AT box is the binding feature of BysR (Fig. 3D). Further, the potential BysR binding sequences in 95 peaks were searched by using the regular expression module in Python, and at least one AT-N 11 -AT box sequence was found in each of these 95 peaks (Table S5), implying that BysR binds around the proposed consensus loci to regulate the transcription of the target genes. In order to confirm that the AT-N 11 -AT box sequences are essential for BysR binding, the promoter region fragments of genes DM992_38905 and DM992_31485, with wild-type and mutated sequences of motif-like consensus sequences, were labeled with Cy5 and used for EMSA analysis. The presence of shift bands in the wild-type probes (probe 4 and probe 6) and the absence of BysR binding to the mutated probes (probe 5 and probe 7) indicated that the predicted consensus-binding sequences were necessary for BysR to bind to the promoter region and regulate the transcription of targeted genes The LTTR BysR Regulates the Synthesis of Occidiofungin Microbiology Spectrum (Fig. S4). In addition, the identification of the ambR1 promoter locus (peak ID: Merged-Chr3-148813-2 [Table S4; Fig. 3E]) indicated that BysR may regulate the expression of the ocf gene cluster by directly binding to the ambR1 promoter region. Our data also suggested that BysR may direct regulate genes involved in core metabolisms, such as the tricarboxylic acid (TCA) cycle, amino acid metabolism, and DNA replication ( Fig. 3E and Table S4). Similarly, the growth rate of the DbysR mutant was slightly lower that of wild-type JP2-270, and the biomass of the DbysR mutant was less than that of JP2-270 in a stable period when cultured in rich medium (Fig. S5), indicating that BysR affect the core metabolism of JP2-270. In addition, some genes related to drug resistance, transport, mobility, transposase activity, transcriptional regulation, and quorum sensing were also identified as the direct targets of BysR ( Fig. 3E and Table S4).
Overall, the results suggest that BysR might be a pleiotropic transcriptional factor that directly and indirectly regulates gene expression in a variety of cellular processes in Burkholderia sp. strain JP2-270.
BysR plays an important role in the broad-spectrum antifungal activity of JP2-270. In our previous research, we showed that JP2-270 could suppress the growth of Rhizoctonia solani GD118 (14). Furthermore, in this study, we also assessed the inhibitory activity of JP2-270 against M. oryzae Guy11. As shown in Fig. 4A and J, the mycelial growth of Guy11 was obviously inhibited by JP2-270, with the mycelial growth radius of Guy11 being 0.72 6 0.08 cm and 3.44 6 0.10 cm after 7 days of incubation with and without JP2-270, respectively (P , 0.01) (Fig. 4A, B, and J). The strong antagonistic activity against M. oryzae and R. solani indicated that JP2-270 had a wide-spectrum antifungal activity.
As a pleiotropic transcriptional factor, BysR could regulate various cellular processes, including secondary metabolism. In this study, we also showed that the DbysR mutant almost totally lost inhibitory activity against M. oryzae ( Fig. 4C and J). The reduced suppressive activity of the DbysR strain could be restored by introducing pBBR2-bysR expressing wild-type bysR but not by introducing the empty vector pBBR1MCS-2 (29) (Fig. 4D, E, and J). These results further confirmed that bysR is necessary for antifungal activity of JP2-270.
Occidiofungin is the main secondary metabolite with antifungal activities. As mentioned above, we found that the expression levels (based on FPKM value) of several genes involved in occidiofungin (region 3.2), gladiostatin (region 3.7), and pyrrolnitrin (region 3.6) biosynthesis were dramatically decreased in the DbysR mutant ( Fig. 2A; Table  S2). To understand the role of these secondary metabolites in inhibiting the mycelial growth of fungi, markerless mutants with deletion of ocfE, encoding nonribosomal peptide synthetase, gdsE, encoding polyketide synthase, and prnC, encoding FAD-dependent oxidoreductase, were constructed, and the inhibitory activities of derivates were also tested. Based on the inhibition assays, we observed that the mycelium growth radius of Guy11 was 3.18 6 0.13 cm for the DocfE strain (compared to 0.72 6 0.08 cm for JP2-270), 1.65 6 0.09 cm for the DgdsE strain (compared to 0.84 6 0.11 cm for JP2-270), and 2.31 6 0.09 cm for the DprnC strain (compared to 1.44 6 0.11 cm for JP2-270) (Fig. 2B). The mutant with ocfE deletion exhibited weaker antifungal activity against M. oryzae than the DgdsE and DprnC mutants, indicating that occidiofungin produced by JP2-270 plays a more important role than gladiostatin and pyrrolnitrin in inhibiting mycelial growth of M. oryzae. Pyrrolnitrin produced by JP2-270 was the main metabolite that inhibited R. solani, with mycelium growth radii of 0.98 6 0.08 cm and 2.14 6 0.24 cm for JP2-270 and the DprnC mutant, respectively. Compared to pyrrolnitrin, occidiofungin and gladiostatin showed lower activity against R. solani, with radii of 1.52 6 0.08 cm for the DocfE mutant (compared to 1.14 6 0.09 cm for JP2-270) and 1.13 6 0.07 cm for the DgdsE mutant (compared to 0.92 6 0.13 cm for JP2-270). Overall, occidiofungin and pyrrolnitrin had better inhibitory activity against M. oryzae and R. solani, respectively, while gladiostatin showed moderate inhibitory activity against these fungi (Fig. 2B). Collectively, these results showed that JP2-270 could produce a variety of secondary metabolites, among which occidiofungin is critical for inhibiting M. oryzae. Comparative evolutionary analysis based on genome and Ocf proteins. Phylograms inferred from genome sequences revealed that the species closest to JP2-270 is Burkholderia pyrrocinia, and both belong to the Bcc (Fig. 5A). Occidiofungin was originally isolated from B. contaminans MS14, and we showed that the ocf gene cluster also occurred in 8 other species of Burkholderia (Fig. 5B). The evolutionary trajectory of Ocf proteins from some species was inconsistent with the genome-wide evolution of the corresponding host strains. B. contaminans and Burkholderia vietnamiensis were clustered in subclade B3-1 (Fig. 5B), but they were distributed separately in subclades A2 and  In detail, we compared the ocf genes between Burkholderia sp. strain JP2-270 and B. contaminans MS14 and found that ocfC, encoding glycosyl transferase, was absent in the ocf gene cluster of JP2-270 (Fig. 5C), while a gene encoding a hypothetical protein was predicted between ocfA and ambR2 in the JP2-270 ocf gene cluster ( Fig. 5C and Table 2). In addition, the ocfH gene in MS14 was highly homologous with three independent genes, DM992_33365, DM992_33360, and DM992_33355, in JP2-270 ( Fig. 5C and Table 2). The other genes of the JP2-270 ocf gene cluster are the same as those of B. contaminans MS14, with amino acid identity ranging from 88.28% to 95.86% ( Table 2). The typical 235 and 210 boxes were predicted in the regions upstream of ocfN, ocfJ, ambR2, and ambR1 in JP2-270, indicating that the putative promoters exist upstream of the respective genes (Fig. 6E). Notably, the putative promoters of ocfN and ocfJ were not identified in MS14, but a promoter was predicted upstream of ocfL in MS14 (10). Overall, the genes in the ocf gene cluster of JP2-270 have relatively high similarity with the corresponding genes in MS14 ( Fig. 5C; Table 2).
AmbR1 is a downstream component of the BysR regulation system and is directly regulated by BysR. It is known that ambR1 is a pathway-specific regulatory gene which positively regulates occidiofungin production. As the DocfE mutant displays significantly decreased inhibitory activity against M. oryzae compared to JP2-270 and the expression of ambR1 is significantly downregulated in the DbysR mutant (Fig. 1B, 2A, and 4F), we inferred that BysR might regulate occidiofungin production by modulating ambR1 expression. Therefore, we enhanced the expression of ambR1 in the DbysR mutant by introducing the ambR1 expression vector controlled by a constitutive promoter, which resulted in increased suppressive activity against M. oryzae (Fig. 4G and J). The observed radius of mycelial growth was 0.86 6 0.15 cm for the DbysR pBBR2pCS-ambR1 strain and 0.72 6 0.08 cm for JP2-270, indicating that ambR1 could almost fully restore the inhibitory activity of the DbysR mutant ( Fig. 4G and J). However, similar to pBBR2pCS, which was used as the control, overexpression of ambR2 could not recover the inhibitory  The LTTR BysR Regulates the Synthesis of Occidiofungin Microbiology Spectrum activity of the DbysR strain ( Fig. 4H to J). These results suggested that ambR1 but not ambR2 is a downstream component of the BysR system in JP2-270. To further assess whether transcriptional regulation of ambR1 was mediated by direct binding of BysR to the promoter, we performed electrophoretic mobility shift assays (EMSAs). BysR fused to an N-terminal glutathione S-transferase (GST) tag was purified using affinity chromatography, and then an ;39-kDa BysR protein was obtained by excising the GST tag (Fig. 6A). Three DNA fragments of the 59-Cy5-labeled ambR1 promoter, obtained by PCR amplification, were used as the probes. The probe with a fragment with positions 2677 to 2844 relative to the translation initiation site of ambR1 was designated probe 1 (2677 to 2844) (Fig. 6B). Similarity, the other two probes were named probe 2 (27 to 2336) and probe 3 (2160 to 2692) (Fig. 6B). The EMSAs showed that the complex of BysR and probe 3 migrated more slowly than the unbound probe, while there were no shift bands observed for probe 1 and probe 2 (Fig. 6B). Moreover, probe 3 binding to BysR was significantly enhanced with increasing amounts of BysR protein (Fig. 6C). These results indicated that BysR could bind to the promoter region of ambR1 and that the promoter region from 2160 to 2692 is essential for binding with BysR. Bioinformatics analysis also revealed that the consensus binding sequence with an AT-N 11 -AT motif (59-ATCGGCGATTTTCAT-39) was present in probe 3 but not in probes 1 and 2 (Fig. S6). These results confirmed that BysR could bind to the promoter region of ambR1 and that the consensus binding sequence is required for binding with BysR. In conclusion, BysR positively regulates the expression of ambR1 by directly binding to its promoter.
BysR regulates occidiofungin biosynthesis at the transcriptional level. To clarify the regulatory effect of BysR on the ocf operon, we used qRT-PCR analysis to assess the transcriptional changes of ocfJ, ocfE, and ambR1 in wild-type JP2-270 and the DbysR, DbysR pBBR2pCS-ambR1, and DbysR pBBR2pCS mutants. As shown in Fig. 6D, the deficiency of bysR resulted in a 7.2-to 28.6-fold reduction in the transcriptional levels of these three selected genes, relative to wild-type JP2-270. Moreover, the overexpression of ambR1 in the DbysR mutant increased the expression levels of ocfJ and ocfE, resulting in 5.45-fold and 4.79-fold upregulation of ocfJ and ocfE, respectively, relative to the wild type (WT) (Fig. 6D), while the expression levels of ocfJ, ocfE, and ambR1 in the DbysR pBBR2pCS strain were similar to those in wild-type JP2-270. The 26.56-fold-upregulated ambR1 transcriptional level was observed in the DbysR strain harboring pBBR2pCS-ambR1 (Fig. 6D), suggesting that ambR1 was successfully overexpressed in the DbysR strain.
Based on our results, we concluded that (i) BysR binds directly to the promoter region of ambR1 and activates the transcription of ambR1 and (ii) AmbR1 promotes transcription of ocf genes located downstream of ambR1 and thus controls the production of occidiofungin (Fig. 6E).

DISCUSSION
Occidiofungin has promising applications in agriculture and biomedicine. Occidiofungin is a natural product produced by Burkholderia spp. and possesses broad-spectrum antifungal, antiparasitic, and anticancer activities with limited toxicity and chemical stability (2,9,(30)(31)(32). Besides its marked potential for application in agricultural practices for controlling fungal plant diseases, occidiofungin has been developed as a lead compound for clinical antifungal therapeutics (e.g., candidiasis) and antiparasitic treatment owing to its distinctive mode of action (2,9,31,32). In addition, occidiofungin and its derivatives could also be developed as anticancer drugs due to their targeting effect on some cancer cell lines (30). However, the lack of knowledge of the regulatory mechanisms of occidiofungin production significantly hinders effective utilization of the occidiofungin-producing strains, such as identification of highly productive strains and manipulation of the occidiofungin production pathway.
Our studies have shown that BysR was required in secondary metabolism and antagonism (14). Here, the antifungal activity of three metabolites produced by JP2-270 was analyzed, and we found that occidiofungin and pyrrolnitrin play important roles in inhibiting M. oryzae and R. solani, respectively ( Fig. 2B and 4). Gladiostatin, recently identified from Burkholderia gladioli, is a novel glutarimide antibiotic with promising anticancer activity (33). However, the gladiostatin-like compound produced by JP2-270 was not as effective as occidiofungin and pyrrolnitrin in inhibiting fungi (Fig. 2B). In order to study and make full use of microbial secondary metabolites for plant disease control, it is necessary to conduct research on the synthesis and regulation of secondary metabolites with potential as biological control agents.
The ocf gene cluster was found in 9 species of Burkholderia but not in any other genus, implying that vertical transmission is the main evolutionary mode of the ocf gene cluster within this genus. However, the ocf gene cluster was transferred in these 9 species ( Fig. 5A and B), indicating that there were both horizontal and vertical transfers in the acquisition of this gene cluster in these species. The ocf gene cluster of JP2-270 was slightly different from that of B. contaminans MS14, such as ocfC (Fig. 5C). The absence of ocfC implies that the occidiofungin-like compound produced by JP2-270 may not have a glycosylation modification and is a new derivative of occidiofungin.
BysR, as a global transcriptional regulator, regulates various biological processes. LTTRs are multifunctional transcription regulators in prokaryotes and play an important role in carbon catabolism, amino acid metabolism, antibiotic resistance and motility, etc. (15,18,19,21). BysR has been identified as a member of the LTTR family and highly homologous to BcaI3178, which controls biofilm formation and protease production (34).
In this study, the isolated beneficial strain Burkholderia sp. strain JP2-270 was used as a model. We identified the genes and associated pathways directly or indirectly regulated by BysR by exploring the RNA-seq and DAP-seq results. Previous studies suggested that ambR1 and ambR2 upregulated the expression of ocfD to ocfJ, which are responsible for occidiofungin biosynthesis (10,11,13,35). Our results indicated that ambR1 but not ambR2 is a direct downstream target gene of BysR. Interestingly, BysR might directly regulate the expression of ocfJ and ocfL, as revealed by DAP-seq ( Fig. 3E; Tables S4 and S5). These results suggested more complex and well-tuned regulation maps of occidiofungin production in Burkholderia. Of note, diaminobutyrate-2-oxoglutarate transaminase (DABAT; EC 2.6.1.76), encoded by the gene ocfL (DM992_33335), presumably catalyzes the formation of L-2,4-diaminobutyric acid (DABA) and 2-oxoglutarate (10,36). DABA was added to the intermediates of occidiofungin production by OcfE (10), while 2-oxoglutarate may participate in the TCA cycle. The ocfL may not only provide aminobutyric acid during the synthesis of occidiofungin but also supplement the source of 2-oxoglutarate in the TCA cycle. Thus, we predicted that ocfL might be involved in both primary and secondary metabolic processes.
RNA-seq results showed that the genes associated with various cellular processes were regulated by BysR, among which the genes related to flagellar assembly and bacterial chemotaxis constitute the two most enriched pathways in the KEGG analysis (Fig. 1C). Moreover, the Lrp family transcriptional factor and H-NS histone family proteins are known as regulators modulating flagellar production (37), and their encoding genes were directly regulated by BysR in our DAP-seq ( Fig. 3E; Table S4), implying that BysR might indirectly regulate flagellar assembly by targeting Lrp or H-NS genes. The previous study showed that ocf genes (including ocfA to N and ambR2), flagellar assembly-related genes (including fliT, fliD, and motB), and the genes associated with the type VI secretion system (including hcp, tssC, and tssB) were significantly downregulated in an ambR1 mutant (MS455MT38) compared to the wild-type Burkholderia sp. strain MS455, as revealed by RNA-seq analysis (2). The results are consistent with our study showing that a bysR-defective mutant has altered expression of ambR1, which in turn affects the expression of the ocf gene cluster, flagellar assembly genes, and type VI secretion system-related genes. Therefore, integrating previous studies and our results, we inferred that BysR is involved in various cellular processes, including secondary metabolites production, TCA cycle, flagellar assembly, bacterial chemotaxis, and secretion.
Most remarkably, the genes encoding transcriptional regulators constitute the largest group directly regulated by BysR (Fig. 3E), such as LTTRs, LuxR family regulators, MarR family regulators, H-NS family regulators, IclR regulators, GntR family regulators, Lrp/AsnC family regulators, TerR/AcrR family regulators, and so on. Obviously, BysR is the upstream regulator of other transcriptional factors. Therefore, in addition to the regulatory mechanisms related to secondary metabolite synthesis, we need more studies to identify other functions of BysR. Revealing more biological functions of BysR will provide a better understanding of how Bcc isolates adapt to environmental conditions.
The feature of BysR regulation. As an alternative and complementary method to chromatin immunoprecipitation sequencing (ChIP-seq), DAP-seq is powerful and convenient. As DAP-seq is not limited to one specific growth condition, it usually can reveal the binding events under conditions that are not suitable for ChIP-seq (26). DAP-seq has successfully been applied to identify transcription factor binding sites (TFBSs) in eukaryotes and prokaryotes (26,(38)(39)(40)(41). In this study, DAP-seq was used to identify the TFBSs of BysR in the genome of JP2-270, and 400 putative binding loci were uncovered (Table S4). Based on the sequences of BysR binding loci, the consensus binding motif AT-N 11 -AT was predicted, which also displays the dyad symmetry pattern and has the typical T-N 11 -A motif of LTTRs (18). Although the classical LTTR binding locus is a dyad symmetry consensus T-N 11 -A motif (18), the binding sequences of LTTRs can vary significantly in base composition and length, such as the T-N 11 -A motif, which is found mainly in Pseudomonas (42), the ATC-N 9 -GAT motif in Rhizobium (43), and the TTA-N 7 -TAA motif in Lactobacillus plantarum (44). Furthermore, bioinformatics analysis revealed that the promoter region of ambR1 possesses a few putative BysR binding loci (matching the consensus in Fig. 3D and Table S5), with a potential site (ATCGGCGATTTTCAT) downstream of the predicted 210 promoter region of ambR1, which is located close to the BysR DAP-seq peak summit (Merged-Chr3-148813-2) (Fig. S6). In addition, BysR binds to a variety of sites in JP2-270 (Table S4), and the majority of BysR binding loci (about 71%) contain the AT-N 11 -AT box (Table  S5). This confirmed that the predicted consensus motif AT-N 11 -AT was the typical binding feature of BysR.
Based on our results, we summarized the biological processes that are potentially regulated by BysR and proposed a model of BysR-regulated occidiofungin synthesis ( Fig. 3E and 6E). BysR not only regulates secondary metabolites but also participates in the regulation of various core metabolic pathways, such as the TCA cycle, amino acid synthesis and metabolism, and DNA replication. BysR is responsible for regulating the expression of multiple transcription factors, and it is possible that BysR occupies a higher regulatory position in the regulatory network of Burkholderia sp. strain JP2-270. For example, ambR1, as a downstream target gene, was confirmed to be directly regulated by BysR, and AmbR1 was reported to regulate flagellar assembly and the type VI secretion system in addition to ocf gene cluster expression.

MATERIALS AND METHODS
Bacterial/fungal strains and culture conditions. Bacterial strains, fungal strains, and plasmids used in this study are listed in Table 3. Burkholderia sp. strain JP2-270 and derivates were routinely cultured in Luria-Bertani (LB) medium at 28°C (45). Escherichia coli strains were routinely maintained in LB medium at 37°C. The concentrations of antibiotics, when necessary, used for E. coli were 50 mg/mL for carbenicillin, 50 mg/mL for kanamycin, 50 mg/mL for streptomycin, and 30 mg/mL for nalidixic acid. M. oryzae isolate Guy11 and R. solani GD118 were routinely cultivated on complete medium (CM) (46) and potato dextrose agar (PDA) plates (200 g potato, 20 g glucose, 20 g agar, 1 L water; natural pH), respectively, at 25°C.
In vitro inhibition assay. The petri dish assay was used to test in vitro antagonistic activity of JP2-270 and derivates against M. oryzae Guy11 and R. solani GD118. Overnight cultures of JP2-270 and derivates were streak inoculated at both sides, 2 cm away from the center of CM or PDA plates. After incubation for 24 h at 25°C, mycelium plugs (5-mm-diameter) from the fresh edge of M. oryzae Guy11 or R. solani GD118 were placed at the center of the CM (for Guy11) and PDA (for GD118) medium plates. Following 2 to 3 days for GD118 and 7 to 10 days for Guy11 coculture at 25°C, the radii of mycelium growth were measured to evaluate the inhibitory effect. Five biological replicates were performed, and an average value was calculated to evaluate the inhibitory activity.
Growth curve. The overnight cultures (15 h at 30°C) in LB were inoculated into 100 mL of rich medium (LB medium) (45) at a ratio of 1:100 and incubated at 30°C with shaking at 200 rpm. Sampling was performed at appropriate intervals up to 72 h according to the growth conditions. The growth rates were determined by measuring absorbance at 600 nm. The results reported here are averages for three replicate samples.

The LTTR BysR Regulates the Synthesis of Occidiofungin Microbiology Spectrum
Overexpression of ambR1 and ambR2 in the DbysR mutant. The cold shock protein gene DM992_01545 (CP029824) was highly expressed in JP2-270 and the expression level was not affected by bysR mutation based on the RNA-seq results (GSE193778). The promoter region of DM992_01545 was amplified using pCS-F/pCS-R and cloned into the multiple-cloning site (MCS) of pBBR1MCS-2 to obtain the modified vector pBBR1MCS-2pCS. To construct a plasmid expressing ambR1, the fragment containing coding sequences of ambR1 (DM992_33410) from the JP2-270 genome (CP029826) was cloned downstream of the constitutive promoter of pBBR1MCS-2pCS. The plasmid pBBR2pCS-ambR1, expressing ambR1, was obtained. The primers CambR1-F and CambR1-R were used to amplify the corresponding fragment. All constructs were verified by PCR and sequencing. Similarly, pBBR2pCS-ambR2, expressing ambR2, was constructed. The obtained plasmids were electroporated into the DbysR mutant to obtain the ambR1-and ambR2-overexpressing DbysR pBBR2pCS-ambR1 and DbysR pBBR2pCS-ambR2 strains, respectively. pBBR1MCS-2pCS was also transferred into the DbysR mutant as a negative control (DbysR pBBR2pCS strain). The primers used in this study are listed in Table S1.
In-frame ocfE, gdsE, and prnC gene deletion in JP2-270. Marker-free site-directed deletion was carried out using pK18mobSacB (47). The upstream and downstream regions of ocfE, gdsE and prnC to be deleted were fused using overlap extension PCR. The fusion products were then subcloned into the suicide vector pK18mobSacB. The resultant recombinant plasmids were introduced into JP2-270 by electroporation transformation, and subsequently, the plasmids were integrated into the target gene via homologous recombination. Burkholderia sp. strain JP2-270 is not sensitive to nalidixic acid. Burkholderia isolates containing the plasmids were selected on LB with 50 mg/mL kanamycin and 30 mg/mL nalidixic acid. The Burkholderia isolates containing plasmids were subsequently cultured in LB without any antibiotic for several generations. The deletion mutants that grew on LB plates with 30 mg/mL nalidixic acid but not on LB plates with 50 mg/mL kanamycin and 30 mg/mL nalidixic acid were selected, and the resultant ocfE (DocfE), gdsE (DgdsE), and prnC (DprnC) allelic exchange mutants were verified by PCR and subsequent DNA sequencing (Fig. S7). The primers used in this study are listed in Table S1.
RNA isolation, RNA-seq, and quantitative RT-PCR. The single-colony bacteria of JP2-270 (WT) and the DbysR mutant (in-frame bysR deletion isolate) (14) were precultured in LB medium at 28°C overnight. The overnight cultures were inoculated into CM in a ratio of 1:100 and cultured at 28°C with shaking at 200 rpm. The cells were harvested after 24 h of incubation. Three biological replicates of each strain were used. The collected cells were sent directly to Beijing Novogene Bioinformatics Technology Co., Ltd., for further treatments. Briefly, a total of 3 mg RNA per sample was obtained, and rRNA was depleted using a Ribo-Zero rRNA removal kit (Gram-negative bacteria). The mRNAs were fragmented using divalent cations at an elevated temperature in NEBNext first-strand synthesis reaction buffer (5Â). Then, the cDNA libraries were generated using a NEBNext Ultra directional RNA library preparation kit for Illumina (New England Biolabs [NEB], USA) following the manufacturer's instructions. The cDNA fragments were purified with an AMPure XP system (Beckman Coulter, Beverly, MA, USA), and then 3 mL USER enzyme (NEB, USA) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 min followed by 5 min at 95°C before PCR. Then, PCR was performed with Phusion high-fidelity DNA polymerase, universal PCR primers, and an index primer. Last, products were purified (AMPure XP system), and library quality was assessed on the Agilent Bioanalyzer 2100 system. The resultant samples were sequenced on an Illumina HiSeq 2000 platform. For qRT-PCR analysis, an RNA preparation pure cell/bacterial kit (Tiangen Biotech, Beijing, China) was used to prepare total RNA according to the instructions. The quality and concentration were analyzed by 1% agarose gel electrophoresis. Genomic DNA (gDNA) was digested with gDNA remover (Toyobo) at 37°C for 5 min. Total RNA (;0.5 mg) was reverted to cDNA using a ReverTra Ace qRT-PCR master mix (Toyobo). Samples were diluted, about 25 ng of cDNA was added to the PCR system in a total volume of 20 mL, and qPCR was conducted using Thunderbird SYBR qPCR mix (Toyobo). The qRT-PCR was performed in an Applied Biosystems 7500 instrument according to the manufacturer's instructions. The recA gene was used as an internal standard, and relative expression was quantified using the 2 2DDCT threshold cycle (C T ) method. Primers are listed in Table S1. Three independent experiments were conducted, each with three replicates.
DAP-seq analysis. DAP-seq was carried out in duplicate on Burkholderia sp. strain JP2-270 genomic DNA as previously described (26). Specifically, the HaloTag-BysR protein was expressed using the TNT SP6 wheat germ protein expression system (Promega, Fitchburg, WI, USA). The HaloTag-BysR protein was purified using magnetic HaloTag beads (Promega) and verified by Western blotting with the anti-HaloTag antibody (Promega). Genomic DNA was extracted from JP2-270 using DNAiso (TaKaRa, Japan). A genomic DNA library was generated using a NEXTflex rapid DNA sequencing kit (Bioo Scientific, USA) following the manufacturer's instructions. The purified protein was incubated with DNA library at 30°C for 2 h, and then the unbound DNA fragments were washed away. Then, the BysR-bound DNA fragments were recovered and PCR amplified. The resultant products were sequenced using the Illumina NovaSeq platform with the PE150 sequencing strategy. Two technical duplicates were performed, and a DNA sample that did not undergo incubation with BysR was used as the input negative control.
Bioinformatics analysis. For RNA-seq data, clean reads were obtained by removing reads containing adapter, reads containing poly-N, and low-quality reads from raw data. Clean reads were mapped to the Burkholderia sp. strain JP2-270 reference genome (CP029824 to CP029828) using Bowtie 2-2.2.3 (48). Then, Htseq v0.6.1 was used to count the reads mapped to each gene (49). Measurement of FPKM (fragments per kilobase of transcript sequence per million base pairs sequenced) was used to estimate gene expression level (50). Differential expression analysis of two groups (three biological replicates per group) was performed using the DESeq R package (1.18.0). The square of the Pearson correlation coefficient (R 2 ) was calculated to estimate the reproducibility of the replicates. DESeq provides statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting P values were adjusted using the Benjamini-Hochberg approach for controlling the false discovery rate. Genes with an adjusted P value of ,0.05 and jlog 2 (fold change)j of .1 found by DESeq were assigned as differentially expressed. The volcano plot drawn by Origin 2022 was applied to display significantly different gene expression. KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment analysis was performed by KOBAS software (51).
For DAP-seq analysis, the clean reads were obtained by filtering the raw data using FASTP with the default parameters. The clean reads were mapped to the Burkholderia sp. strain JP2-270 reference genome (CP029824 to CP029828) using BWA-MEM (52). Peaks were generated with MACS2 (53) (P value , 0.01), and IDR software was used to merge the peaks present in the two replicates. MEME-ChIP was used to analyze the conservative motifs in the peaks (54). The HOMER software (55) was used to annotate peaks. The box plots of the peaks' width distribution, the pie chart of distribution of BysR-bound genes, and the histogram distance to the TSS were drawn using Origin 2022. The Venn plot for RNA-seq and DAP-seq data was produced with Origin 2022.
Phylogenetic analysis was performed using phyloPhlAn 3 and MEGA7 software (56,57). The open reading frames (ORFs) of the JP2-270 ocf gene cluster were predicted by the Softberry FGENESB program (Softberry, Inc., Mount Kisco, NY), and the identified ORFs were analyzed using a BLASTn search in the Burkholderia Genome Database (58). The Softberry BPROM program was used to identify the putative promoter sequences. The putative consensus binding sequences of BysR were identified by regular expression with the package in Python.
Protein expression, purification, and EMSA. The coding region of BysR was obtained by PCR with the primers BysR-F and BysR-R (Table S1) and cloned into vector pGEX-6P-1 to generate an N-terminally GST-tagged BysR protein fusion. The resulting plasmid was transformed into E. coli BL21(DE3) ( Table 3) for protein expression. The resultant strain was cultivated in LB medium (containing 100 mg/mL ampicillin) overnight at 37°C. Two milliliters of the overnight culture was transferred into 200 mL fresh LB at 37°C and grown with shaking at 200 rpm, until an optical density at 600 nm (OD 600 ) of 0.4 to 0.6 was reached. Isopropyl-b-D-thiogalactopyranoside (IPTG; Sangon, Shanghai, China) was added to a final concentration of 0.4 mM. The culture was incubated for an additional 16 h at 18°C with shaking at 100 rpm. Cells were collected by centrifugation at 4°C, resuspended in 25 mL phosphate-buffered saline (PBS) lysis buffer containing 10 mM protease inhibitor (phenylmethylsulfonyl fluoride [PMSF]; Beyotime, China), and lysed by sonication (JY92-IIDN; Scientz, Ningbo, China) (power, 200 W; ultrasonic, 5 s; intermittent, 10 s). Following centrifugation at 4,500 rpm at 4°C for 30 min, the soluble protein was collected by incubation with GST beads (Sangon, Shanghai, China) for 30 min. The purified GST-BysR protein was eluted with buffer containing glutathione. The GST tag was removed by PreScission protease according to the instructions (P2303; Beyotime, Shanghai, China). The purity of BysR protein was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and the concentration of BysR was determined with a bicinchoninic acid (BCA) assay kit (Sangon Biotech, Shanghai, China). Aliquots of the proteins were stored at 280°C.
Statistics. Data analysis was performed using Microsoft Excel 2010. Student's t test was used to compare the differences between two sets of data. The differences between results were considered statistically significant when P was ,0.05 and extremely significant when P was ,0.01.
Data availability. Raw sequence data from our RNA-seq and DAP-seq analyses can be accessed via the National Center for Biotechnology Information Sequence Read Archive server under accession numbers GSE193778 and GSE193916, respectively, and the complete genome sequence of Burkholderia sp. strain JP2-270 has been deposited in NCBI under accession numbers CP029824 to CP029828.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 2.2 MB.