Nicotinic Acid Catabolism Modulates Bacterial Mycophagy in Burkholderia gladioli Strain NGJ1

ABSTRACT Burkholderia gladioli strain NGJ1 exhibits mycophagous activity on a broad range of fungi, including Rhizoctonia solani, a devastating plant pathogen. Here, we demonstrate that the nicotinic acid (NA) catabolic pathway in NGJ1 is required for mycophagy. NGJ1 is auxotrophic to NA and it potentially senses R. solani as a NA source. Mutation in the nicC and nicX genes involved in NA catabolism renders defects in mycophagy and the mutant bacteria are unable to utilize R. solani extract as the sole nutrient source. As supplementation of NA, but not FA (fumaric acid, the end product of NA catabolism) restores the mycophagous ability of ΔnicC/ΔnicX mutants, we anticipate that NA is not required as a carbon source for the bacterium during mycophagy. Notably, nicR, a MarR-type of transcriptional regulator that functions as a negative regulator of the NA catabolic pathway is upregulated in ΔnicC/ΔnicX mutant and upon NA supplementation the nicR expression is reduced to the basal level in both the mutants. The ΔnicR mutant produces excessive biofilm and is completely defective in swimming motility. On the other hand, ΔnicC/ΔnicX mutants are compromised in swimming motility as well as biofilm formation, potentially due to the upregulation of nicR. Our data suggest that a defect in NA catabolism alters the NA pool in the bacterium and upregulates nicR which in turn suppresses bacterial motility as well as biofilm formation, leading to mycophagy defects. IMPORTANCE Mycophagy is an important trait through which certain bacteria forage over fungal mycelia and utilize fungal biomass as a nutrient source to thrive in hostile environments. The present study emphasizes that nicotinic acid (NA) is important for bacterial motility and biofilm formation during mycophagy by Burkholderia gladioli strain NGJ1. Defects in NA catabolism potentially alter the cellular NA pool, upregulate the expression of nicR, a negative regulator of biofilm, and therefore suppress bacterial motility as well as biofilm formation, leading to mycophagy defects.


RESULTS
Transposon mutagenesis screen identified nicC gene to be important for mycophagous ability of B. gladioli strain NGJ1. We carried out Tn5gusA11-based transposon mutagenesis in B. gladioli NGJ1 to identify gene(s) required for bacterial mycophagy. Out of 84 mutant strains that were transposon-tagged (GUS positive), 12 were defective in mycophagy against R. solani, at 7 DPI (Fig. S1A). We selected one of them (NGJ1_6) wherein the Tn5gusA11 insertion was in the nicC gene (ACI79_RS00345; Burkholderia genome database Locus ID; https://www.burkholderia.com/) that encodes 6-hydroxynicotinate 3-monooxygenase, an important enzyme in nicotinic acid (NA) catabolism (16) for further characterization. Notably, the NGJ1_6 was compromised in foraging over R. solani mycelia, even at 7 DPI (Fig. S1B). On the other hand, wild-type NGJ1 foraged over fungal mycelia, at both 3 and 7 DPI (Fig. S1).
nicC gene is a part of NA catabolic gene cluster in B. gladioli strain NGJ1. Previous studies had identified various genes associated with NA catabolic (nic) pathway in Pseudomonas putida (16) and Bordetella bronchiseptica (15). As the KEGG pathway of NGJ1 is not available in the database, we used the genome sequence of B. gladioli strain BSR3 (18) to identify NA catabolic pathway genes (Fig. S2). Similar to B. bronchiseptica and P. putida, most of the genes associated with NA catabolism are encoded in an operonic fashion in B. gladioli strain BSR3 (Fig. 1A). BLAST analysis revealed that B. gladioli strain NGJ1 also harbors a complete nic gene cluster (Table S1, Fig. 1A).
NA catabolic genes (nicC and nicX) are required for bacterial mycophagy by B. gladioli strain NGJ1. In order to validate whether mycophagy defect in NGJ1_6 is due to transposon-mediated disruption of the nicC gene, we created DnicC, an independent insertional mutant in NGJ1. Similar to the transposon-tagged mutant, the DnicC was deficient in bacterial mycophagy on R. solani (Fig. 1B). Considering that nicC is a part of the nic operon (Fig. 1A), we envisaged that insertion in the gene may lead to disruption of downstream genes associated with NA catabolism. Hence, to explore the importance of NA catabolic pathway during mycophagy, we investigated the involvement of a downstream gene i.e. nicX (2,5-dihydroxypyridine dioxygenase; ACI79_RS00335). The qRT-PCR analysis reflected that both nicC and nicX were upregulated during mycophagous interaction with R. solani (Fig. 1C). We created an insertion mutant in the nicX gene and observed that mutant strain (DnicX) was defective in foraging over R. solani (Fig. 1 B). The bacterial abundance (log CFU/ mL) on R. solani mycelia was negligible during the confrontation of DnicC and DnicX, while abundant bacterial growth on the mycelia was observed in case of wild-type NGJ1 (Fig. S3 A). Further, we used a colorimetric MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay to detect the viability of bacterial-treated R. solani mycelia. It is known that viable cells convert MTT into a colored compound, while nonviable cells remain colorless (19). The treatment with DnicC and DnicX mutant bacterial strains failed to induce cell death in R. solani mycelia, as evident by the formation of color (Fig. S3 B). On the other hand, the wild-type NGJ1-treated R. solani The interaction of B. gladioli NGJ1 strains (10 3 cells/mL density) with R. solani on PDA plates at 3 and 7 DPI. (C) qRT-PCR-based expression of nicC and nicX genes of NGJ1 during 48h of mycophagous interaction with R. solani. The fold change was estimated by 2 2DDCt method using 16S rRNA of NGJ1 as the endogenous control. (D) Growth of serially diluted bacterial cultures on M9 minimal media (MM), with and without supplementation of R. solani mycelial extract (RS extract). The experiments were independently repeated three times with a minimum of three technical replicates. Graphs show mean values 6 standard error of three biological replicates.

Role of Nicotinic Acid in Bacterial Mycophagy
Microbiology Spectrum mycelia were colorless, emphasizing that they were nonviable. Further, the spectrophotometric measurement reflected that DnicC and DnicX mutant treated mycelia had significantly more formazon, compared to those treated with wild-type bacterium ( Fig. S3 C). Overall, this suggested that disruption in either nicC or nicX genes of NA catabolic pathway compromises NGJ1 to forage over R. solani. NA is required for the growth of B. gladioli strain NGJ1 in minimal media. We observed that the wild-type as well as DnicC and DnicX mutant NGJ1 strains were unable to grow in M9 minimal media that is devoid of any carbon source, whereas the bacterial growth was comparable in rich media (PDB) (Fig. S3 D). The supplementation of NA restored the growth of wild-type NGJ1 (to a certain extent), whereas other N-heterocyclic aromatic compounds i.e. picolinic acid (PA) and isonicotinic acid (INA) failed to promote bacterial growth in minimal media (Fig. S3 D). In contrast, the addition of NA/N-heterocyclic aromatic compounds was unable to support the growth of DnicC and DnicX mutant strains (Fig. S3 D). Moreover, we observed that supplementation of cell extract of R. solani, promoted the growth of wild-type but not DnicC and DnicX mutant strains (Fig. 1D). Taken together, our data suggest that B. gladioli strain NGJ1 may sense R. solani as a source of NA, while DnicC/DnicX mutants are defective in this process.
Supplementation of NA restores mycophagy in DnicC and DnicX mutants of B. gladioli strain NGJ1. Considering that the end product of NA catabolism is fumaric acid (FA), a TCA cycle intermediate, we tested whether supplementation with FA (20 mM) restores mycophagy in the DnicC/DnicX mutants. The analysis revealed that DnicC and DnicX are still compromised in mycophagy against R. solani on FA supplemented plates ( Fig. 2A). On the other hand, supplementation of NA (20 mM), restored mycophagy in both DnicC as well as DnicX mutant strains ( Fig. 2A), and the bacterial abundance on fungal mycelia was enhanced (Fig. 2B). It is to be noted that supplementation of NA has no apparent effect on the growth of the wild-type, DnicC, and DnicX The experiments were independently repeated three times with a minimum of three technical replicates. Graphs show mean values 6 standard error of three biological replicates. Asterisks * and ** indicate a significant difference at P , 0.05 and P , 0.01, respectively (estimated using oneway ANOVA). NS, nonsignificant.

Role of Nicotinic Acid in Bacterial Mycophagy
Microbiology Spectrum strains in laboratory media, in the absence of fungal mycelia (Fig. 2C). Overall, our data suggest that NA promotes the mycophagous ability of B. gladioli strain NGJ1 on R. solani. We observed that expression of the nicT gene (ACI79_RS00310), which encodes a major facilitator superfamily (MFS) transporter protein (potentially involved in NA uptake [16]), was upregulated in wild-type and DnicC/DnicX mutant bacterium during 48 h of interaction with R solani (Fig. S4). We anticipate that upregulation of NicT may facilitate the NGJ1 bacterium to take up NA during mycophagy.
NA regulates motility and biofilm formation ability in B. gladioli strain NGJ1. Considering that NA but not FA restored mycophagy in DnicC and DnicX mutant strains, we envisaged that NA may regulate mycophagy through modulation of bacterial physiological processes. Motility and biofilm formation ability play important roles during bacterial interaction with eukaryotic organisms (20). We observed that wildtype NGJ1 is proficient in swimming motility on laboratory plates (0.25% PDA), which gets enhanced in the presence of NA ( Fig. 3A and B). The wild-type NGJ1 also produced biofilm on polystyrene plates and NA enhanced biofilm formation (Fig. 3C). Confocal laser scanning microscopic analysis further reflected that NA enhances biofilm formation on glass slides ( Fig. 3D and E). Moreover, FA supplementation had no significant effect on the bacterial motility ( Fig. 3A and B) and biofilm forming ability on polystyrene plates (Fig. 3C) as well as glass slides ( Fig. 3D and E). (E) Average biofilm thickness formed by NGJ1 strains, estimated from the CLSM images. Graphs show mean values 6 standard error of three biological replicates. Asterisks * and ** indicate a significant difference at P , 0.05 and P , 0.01, respectively (estimated using one-way ANOVA). NS, nonsignificant.

Role of Nicotinic Acid in Bacterial Mycophagy
Microbiology Spectrum Swimming motility as well as biofilm formation ability are compromised in DnicC and DnicX mutant strains. We further tested the swimming motility and biofilm formation ability in DnicC and DnicX mutant strains. Compared to the wild-type NGJ1, DnicC and DnicX were less motile ( Fig. 3A and B). Also, the mutants were compromised in biofilm formation on polystyrene plates (Fig. 3C) and glass slides ( Fig. 3D and E). NA supplementation enhanced the motility ( Fig. 3A and B) and biofilm formation ability in the mutant bacteria ( Fig. 3C to E). On the other hand, supplementation of FA has no significant effect on the motility ( Fig. 3A and B) and biofilm forming ability of DnicC/ DnicX on polystyrene plates (Fig. 3C) as well as glass slides ( Fig. 3D and E).
nicR, encoding a MarR type of transcriptional repressor, is upregulated in DnicC and DnicX mutants of B. gladioli strain NGJ1. It was intriguing why the mutation in NA catabolic genes leads to defects in bacterial motility and biofilm formation. Generally, NA catabolism is regulated by a MarR type of transcriptional repressor in bacteria which maintains the cellular NA pool (16). We observed that NGJ1 encodes NicR (ACI79_RS00315), a homolog of MarR and the expression of the gene was upregulated in DnicC and DnicX mutants during 48 h of confrontation with R. solani, compared to the wild-type NGJ1 (Fig. 4A). However, upon NA supplementation, nicR expression was comparable in DnicC, DnicX and wild-type NGJ1 during mycophagy (Fig. 4B). We anticipate that upon NA supplementation reduced expression of nicR restores mycophagous ability in the DnicC/DnicX mutant strains.
To investigate further, we created a DnicR mutant of NGJ1 and observed that it is compromised in mycophagous ability on R. solani (Fig. 4C, S5A). The supplementation of NA failed to enhance mycophagy in the DnicR (Fig. 4C, S5A), potentially due to the inability of the mutant bacterium to sense NA, as in case of DbpsR mutant of B. bronchiseptica (21). Notably, DnicR mutant bacterium was completely defective in swimming motility (Fig. 4D, S5 B); however, it had enhanced ability to form biofilm (Fig. 4E, S5 C and D), irrespective of the presence or absence of NA. Considering that DnicR bacterium was defective in motility, we analyzed the expression of flhDC genes that regulate flagellar biosynthesis (22,23). The qRT-PCR analysis reflected that flhC (ACI79_RS06945) and flhD (ACI79_RS06940) genes were induced in wild-type as well as in DnicC but not in DnicR mutant strains during confrontation with R. solani (Fig. 4F). Moreover, expression of flhC and flhD genes were upregulated upon NA supplementation during confrontation of wild-type and DnicC strains with R. solani (Fig. 4G). On the other hand, the flhDC genes were not induced in DnicR during interaction with R. solani, even in the presence of NA (Fig. 4G). This suggested that DnicR mutant is compromised in flagellar biosynthesis. Transmission electron microscopic analysis reflected that wild-type and DnicC mutant bacteria possess a single polar flagellum while no flagella were seen in case of DnicR mutant during growth in laboratory media (Fig. 4H). Overall, our data suggest that impaired motility and excessive biofilm formation ability impart mycophagy defects in DnicR mutant.

DISCUSSION
Mycophagy is an important trait that enables bacteria to survive under nutrient-limiting conditions by utilizing fungal biomass as a nutrient source. Previously we have demonstrated that a prophage tail-like protein (Bg_9562) is deployed by the B. gladioli strain NGJ1 to forage over fungi (3). In the present study, we demonstrate the importance of NA catabolism for mycophagous ability of B. gladioli strain NGJ1. Previous studies have shown that NA catabolism is important for bacterial virulence and growth under diverse environments (24,25). Members of gammaproteobacteria (Pseudomonadales) and betaproteobacteria (Burkholderiales), including Pseudomonas putida and Bordetella bronchiseptica possess complete NA catabolic pathway (15,16,21). The operonic arrangement of NA catabolic (nic) genes differ between the Pseudomonas and Bordetella species. We observed that NA hydroxylase enzymes are tripartite (NicA, NicB1, NicB2) in B. gladioli (NGJ1 and BSR3) and Bordetella sp. (15). Similar to P. putida, B. gladioli strains harbor putative major facilitator superfamily transporter (NicT) at the nic locus, while Bordetella sp. harbors putative ATP binding cassette transporter, potentially for uptake of NA (15).
Our data reflected that NGJ1 is auxotrophic to NA, whereas supplementation with R. solani mycelial extract enables bacterial growth in minimal media without an additional carbon source. On the other hand, neither NA nor R. solani extract restored the growth of DnicC/DnicX mutants of NGJ1. This reinforced that NA catabolism is important for the growth of NGJ1 under carbon-deficient conditions. Notably, the DnicC/ DnicX mutants were mycophagy defective and supplementation of NA promoted mycophagy. This reflects that NA is important for mycophagy and we anticipate that during confrontation NGJ1 may derive NA from R. solani. Considering that NA catabolism leads to the formation of fumaric acid (FA), a TCA cycle intermediate (16), we speculated that supplementation of FA may restore mycophagy in the DnicC and DnicX mutant strains. However, FA did not restore mycophagous ability in DnicC/DnicX mutants. This implies that during mycophagy NA might have additional regulatory functions, other than being a carbon source. NA catabolism is tightly regulated in bacteria by a MarR-type of transcriptional regulator i.e NicR in P. putida and BpsR (NicR homolog) in B. bronchiseptica (17,21). Depending upon NA pool, the expression of The experiments were independently repeated three times with a minimum of three technical replicates. The fold change was estimated by 2 2DCt method, using 16S rRNA of NGJ1 as the endogenous control. Graphs show mean values 6 standard error of three biological replicates. Asterisk * indicates a significant difference at P , 0.05 (estimated using one-way ANOVA). NS, nonsignificant.

Role of Nicotinic Acid in Bacterial Mycophagy
Microbiology Spectrum nicC and nicR genes is modulated (15). In wild-type B. bronchiseptica the nicC expression was enhanced in an NA concentration dependent manner, while the DbpsR mutant strain was unable to sense NA, exhibiting constant upregulation of nicC (15). We hypothesized that defects in nicC/nicX genes may alter the cellular NA pool in NGJ1. In this regard, we observed that compared to the wild-type NGJ1, nicR gene was upregulated in DnicC as well as DnicX mutants during confrontation with R. solani. On the other hand, supplementation of NA led to the basal level expression of nicR in DnicC/ DnicX mutant strains. Considering the above, we anticipate that supplementation of NA restores NA pool, maintains basal level expression of nicR and therefore promotes mycophagy in DnicC/DnicX mutants. It is known that bacterial motility and biofilm formation play an important role during host colonization. The fungal hypha serves as means for the long-distance movement of bacteria in rhizosphere (26). The soil-dwelling bacterium, Bacillus subtilis produces biofilms on the fungal mycelia of Aspergillus niger and Agaricus bisporus (27). We envisaged that mycophagous bacteria need to be efficient in establishing contact with fungi, colonizing and spreading over the mycelia. In this regard, we observed that wild-type NGJ1 is proficient, while DnicC/DnicX mutants are compromised in swimming motility as well as biofilm formation. The presence of NA but not FA enhanced the bacterial motility as well as biofilm forming ability and thereby promotes mycophagy in wild-type/DnicC/DnicX strains. It is noteworthy that DnicR was compromised in mycophagy and supplementation of NA did not restore mycophagous ability. We anticipate this is due to the inability of DnicR mutant of NGJ1 to sense NA, similar to the DbspR mutant of Bordetella (21). A previous study has shown that DbpsR mutant of B. bronchiseptica produces excessive biofilm (28). We observed that although the DnicR mutant of NGJ1 produced excessive biofilm but it was completely defective in swimming motility, irrespective of the presence or absence of NA in the media. Our data emphasize that the swimming motility in DnicR is compromised due to defects in flagellar biosynthesis. The transmission electron microscopic studies revealed that DnicR mutant was not able to produce flagella, while prominent flagellum was visible in wild-type NGJ1. The qRT-PCR analysis reflected that during mycophagy, the expression of flhDC genes that encode the master regulator of flagellar biosynthesis (23,29,30), was upregulated in wild-type NGJ1. On the other hand, the DnicR mutant was compromised in the expression of flhDC genes, even in the presence of NA. Bacterial flagella are important for adherence, maturation and dispersal of biofilm-forming cells to newer sites (31,32). Generally, a transition from motility to biofilm is important for bacterial pathogenesis and an optimal biofilm consists of both motile and nonmotile cells (33,34). Our data points out that due to the lack of flagella, DnicR mutant remains nonmotile and its ability to disperse from biofilm is compromised. Therefore, DnicR mutant exhibits reduced mycophagy compared to wild-type NGJ1.
Overall, our study suggests that NGJ1 requires NA for optimal growth. During mycophagous interaction, NGJ1 potentially senses fungal mycelia as a NA source and maintains the basal level expression of nicR. This in turn enables the bacterium to produce optimal biofilm, which is important for effective colonization on fungal mycelia. Further, swimming motility is an important trait that enables the NGJ1 bacteria to spread over fungi during mycophagy. We highlight that disruption of NA catabolic genes (nicC and nicX) alters the cellular NA pool in NGJ1, which in turn upregulates nicR, the negative regulator of biofilm formation. The upregulation of nicR impairs biofilm formation and bacterial motility, rendering mycophagy defects in the DnicC/DnicX mutant strains (Fig. 5).

MATERIALS AND METHODS
Growth conditions. B. gladioli strains were grown on potato dextrose agar (PDA; HiMedia Laboratories, India) plates at 28°C. The Tn mutant strains of NGJ1 were grown on PDA supplemented with rifampicin and spectinomycin. The insertional mutants, DnicC, DnicX and DnicR were grown on PDA supplemented with rifampicin and kanamycin. R. solani AG1-IA strain BRS1 (35) was cultured on PDA plates at 28°C, and freshly collected sclerotia were used for mycophagy assays. E. coli strains were grown on LBA (Luria Bertani Agar media; HiMedia Laboratories, India) at 37°C. Whenever required, medium was

Role of Nicotinic Acid in Bacterial Mycophagy
Microbiology Spectrum supplemented with appropriate concentration of antibiotics (kanamycin: 50 mg/mL; rifampicin:50 mg/mL; ampicillin: 50 mg/mL; and spectinomycin: 50 mg/mL). Various bacterial and fungal strains, as well as plasmids used in this study are enlisted in Table S2.
Transposon mutagenesis in NGJ1. The transposon, Tn5gusA11 having a spectinomycin resistance marker was maintained in E. coli strain TP003 as a suicidal plasmid (36). Biparental mating between the donor E. coli TP003 and the recipient B. gladioli NGJ1 strain was performed, as described before (37). Briefly, a single colony of E. coli strain TP003 was mixed with the recipient B. gladioli NGJ1 cells on a piece of Hybond N Plus nylon membrane (Amersham Life Science, Buckinghamshire, UK). Upon incubation for 48 h at 28°C, the bacterial cells were scraped with a sterile toothpick, resuspended in 100 mL sterile water and plated on KBA (King's medium B Base; HiMedia, India) supplemented with rifampicin (50 mg/mL) and spectinomycin (50 mg/mL). The plates were incubated at 28°C and positive conjugants were analyzed for gus (b-glucuronidase) activity, as described in (36). Briefly, bacteria were taken in a sterile microcentrifuge tube, mixed with 1 mL of X-gluc (5-bromo-4-chloro-3-indolyl-beta-d-glucuronic acid;100 mg/mL) and incubated at 37°C for 30 min or until the appearance of blue color in the suspension.
Screening of transposon-tagged mutants for mycophagy defects and identification of the transposon-tagged gene. The transposon-tagged mutants were screened for defects in mycophagy against R. solani on PDA plates, as described before (3). The mutants compromised in mycophagy were selected. For identification of the disrupted gene, genomic DNA of the mutant bacterium was partially digested with SauIIIA (New England Biolabs, Inc. Ipswich, USA) and cloned into pBS-KS (1) vector, linearized with BamHI. The clone having the DNA fragment corresponding to spectinomycin marker of the transposon and the flanking sequence of NGJ1 (tagged-region) were selected on LBA plates containing ampicillin (50 mg/mL) and spectinomycin (50 mg/mL). The positive transformants were sequenced using M13 forward/reverse primers (Table S3). The sequences, thus obtained, were used for BLAST (Basic Local Alignment Search Tool) searches in the Burkholderia Genome Database (https://www.burkholderia.com) to identify the disrupted genes.
Insertional mutagenesis through plasmid integration in the nic genes. Partial fragments (250 to 300bp) of nicC, nicX, and nicR genes of NGJ1 were amplified from genomic DNA using gene specific primers (Table S3), cloned into pK18mob vector and electroporated into competent cells of wild-type NGJ1 using Gene pulsar XcellTm; Bio-Rad Laboratories, Inc., CA, USA, as described previously (4). The transformants were selected on rifampicin (50 mg/mL) and kanamycin (50 mg/mL) containing KBA plates. The mutants were confirmed by PCR using gene-specific flanking forward and M13 vector specific reverse primers.
Bacterial-fungal confrontation assay. The bacterial-fungal interaction was carried out as described previously (5,38). Briefly, R. solani sclerotia were treated with 5 mL of 10 3 cells/mL cell density of various NGJ1 strains for 4 h at 28°C. After washing, the bacterial-treated sclerotia were placed on PDA plates and the bacterial foraging over fungal mycelia was monitored at 3 and 7 DPI (days postinoculation). Wherever applicable, 20 mM nicotinic acid (NA; Sigma-Aldrich, Cat. no 72309) or fumaric acid (FA; Sigma-Aldrich, Cat. No. 47910) were supplemented onto the confrontation plates. The bacterial abundance on mycelial mass after 7 DPI was estimated by serial dilution plating and colony counting. Each experiment was repeated at least thrice with minimum 4 sclerotia per treatment being analyzed in each replicate. solani) mycelia/growth media and upregulates its NA catabolic pathway. This keeps the expression of nicR (a negative regulator of NA catabolism and biofilm formation) to the basal level. This enables the bacteria to form motile colonies and produce optimum biofilm which is important for mycophagy. On the other hand, disruption of nicC/nicX modulates NA pool, leading to upregulation of NicR which alters bacterial motility and ability to produce optimal biofilm, thereby negatively impacting mycophagy.
Fungal cell viability assay using MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide). The cell viability assay was performed as described previously (5). Briefly, pregrown R. solani mycelial mass was treated with NGJ1 strains (1% inoculum of 10 9 cells/mL) in liquid media (PDB) for 48 h at 28°C. R. solani mycelia without any bacterial treatment were used as a control. The bacterial cells loosely attached with fungal mycelia were washed (2 to 3 times) with PBS buffer (phosphate buffer saline; 10 mM, pH 7.4) and the fungal biomass was stained with 100 mL of MTT (HiMedia Laboratories, India; Catalog no. TC191) solution (5 mg/mL in PBS buffer). Upon incubation in the dark for 90 min, the appearance of a dark-colored formazon was observed. After washing the fungal mycelia with PBS buffer, bound formazon was extracted using absolute ethanol and quantified (OD 570 ) using a spectrophotometer (Bio-Rad Smart Spec-3000), as described previously (3).
qRT-PCR based bacterial gene expression analysis during bacterial-fungal interaction. The bacterial-treated fungal mycelia were filtered out from the liquid media using mira cloth and total RNA was isolated from the bacterial cells using RNeasy minikit (Qiagen), following manufacturer's instructions and as described before (5). One mg of total RNA was taken for cDNA synthesis, using Verso cDNA Synthesis kit (Thermo Fisher Scientific Inc.). The qRT-PCR was performed using gene specific primers (Table S3) and PowerUp SYBR green master mix (Applied Biosystems), on ABI 7900HT Real-time PCR (Applied Biosystems). The 16S rRNA of NGJ1 was used as the endogenous control and fold change in gene expression was estimated using 2 2DDCt methods (39).
Bacterial swimming motility assay. Semisolid plates were prepared with 0.25% agar in PDB and NA/FA (20 mM) were supplemented, wherever mentioned. Bacterial cultures (OD 600 = 0.5) were washed and resuspended in PBS. The bacterial cultures were spotted onto the center of the semisolid plates and incubated at 28°C for 16 h. The swim zone was monitored and the diameter (in cm) was recorded. The experiment was performed thrice independently with at least five technical replicates.
Biofilm formation assay in 96 well-based polystyrene plates. The biofilm formation assay in 96well polystyrene plates was carried out as per previously described method (40) with slight modifications. Briefly, 100 mL bacterial cultures (10 6 cells/mL cell density) were grown in polystyrene 96-well plates with or without 20 mM NA/FA supplementation. After 72 h of static growth at 28°C, the medium was gently decanted and planktonic cells were washed off from the wells. The polystyrene surface attached bacterial cells forming biofilm were stained with 1% crystal violet for 10 min at room temperature. The bound crystal violet in the bacterial cells was solubilized using absolute methanol (100 mL) and quantified by measuring the absorbance at 540 nm. Each experiment was performed with five technical replicates and the average of three biological replicates was plotted as a graph.
Visualization of biofilm under confocal laser scanning microscope. The bacterial cultures (10 6 cells/mL cell density) were grown in PDB with or without supplementation of NA or FA (20 mM) in 50-mL falcon tubes containing a sterile glass slide half-dipped in the media and incubated for 72 h at 28°C, as described (40). Upon washing with sterile water, the bacterial biofilms formed at the air-media interface at the glass slide were stained using the BacLight LIVE/DEAD Bacterial Viability kit (L7012; Invitrogen, Eugene, OR, USA), as per the manufacturer's instructions. Slides were mounted with 25% glycerol and analyzed under 63Â objective of confocal laser scanning microscope (Leica TCS-SP8). The 3D images were reconstructed using LAS AF Version: 2.6.0 build 7266 software. Biofilm thickness was quantified by measuring the average of horizontal (xz) optical sections (at 1 mm intervals) of the biofilm on the glass slides. Each experiment was performed with five technical replicates and the average of three biological replicates was plotted as a graph.
Transmission electron microscopy for visualization of flagella. Bacterial cultures were grown to log phase, cells were pelleted down and gently washed using 1 mM phosphate buffer (pH 7.2). The cells were fixed with 2.5% glutaraldehyde and 1.5% formaldehyde solution for overnight. Upon washing with phosphate buffer, cells were stained with phosphotungstic acid (PTA) and mounted on Carbon Film 200 Mesh Copper (Electron Microscopy Sciences). The grids were dried at room temperature and visualized under transmission electron microscope using SAIF-EM facility at All India Institute of Medical Sciences (AIIMS), New Delhi, India.
Statistical analysis. One-way analysis of variance was performed using Sigma Plot 11.0 software (SPSS, Chicago, IL, USA), using the Student-Newman-Keuls test considering P # 0.001, P # 0.01, and P # 0.05 as statistically significant. Where applicable, the significance is mentioned in the figure legend.
Data availability. All data related to the study are included in the article and supplemental material.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.7 MB.