Antifungal compound from marine Serratia marcescens BKACT and its potential activity against Fusarium sp.

Ecofriendly biocontrol agents to control of pathogenic fungi are in demand globally. The present study was evaluated the antifungal potentials of marine bacteria Serratia marcescens BKACT against eight different Fusarium species. A highest 75.5 ± 0.80% of mycelial inhibition was observed against Fusarium foetens NCIM 1330. Structural characterization of the puried compound was analyzed by GC-MS and NMR techniques, based on the analysis it is conrmed as 2, 4- di-tert butyl phenol (2, 4-DTBP) with chemical structure C 14 H 22 O. at 0.53mM concentration puried compound inhibited complete spore germination of F. foetens NCIM 1330. In vitro assay showed complete inhibition of F. foetens NCIM 1330 on the wheat seeds. Tested concentration doesn’t show any toxic effect on germination of the seeds. By this study we conclude that, 2, 4- DTBP a suitable candidate to be used as biocontrol agent against Fusarium infection.


Introduction
Considerable interest has grown in nding alternatives to chemical pesticides for suppression of soil borne plant pathogens and plant parasitic nematodes. Relatively few of these antagonistic microbes have been commercialized as biocontrol agents due to problems such as inconsistent performance in the eld, lack of broad-spectrum disease suppression activity, or slower or less complete suppression when compared with chemical pesticides (Roberts et al. 2005). Fusarium is a genus of lamentous fungi which colonizes many host plants and crops worldwide. It includes around 70 species, some of which causes plant diseases such as wilts, seedling blights, rots and cankers. Fusarium is also known to produce mycotoxins in many economically important crops, resulting in high loss of yield and quality of the crop (Munkvold et al. 2017 and Aoki et al. 2014). Microorganisms have always been regarded as a treasure source of useful compounds and are considered as green and sustainable alternatives for chemical fungicides. Several studies have been carried out with respect to use of soil and freshwater microbes as biocontrol agents. Despite having colossal diversity and potential to produce a novel class of compounds, marine bacteria and their active compounds are not much explored. Marine microbial antagonist strains can produce a wide range of secondary metabolites (Imhoff et al. 2011). Serratia genus is one of the diverse families of Enterobacteriaceae. It is a highly diverse group of bacteria capable of occupying many different habitats such as water, soil, plants, vertebrates and humans. It produces prodigiosin and various bioactive metabolites, such as althiomycin, oocydin A, serrawettins, rubiwettin, and carbapenem (Soenens et al. 2020). In the present study, we have demonstrated the detailed antifungal potential of 2, 4-DTBP produced by marine Serratia marcescens BKACT against Fusarium species.

Isolation, identi cation and antifungal screening
In the previous study, around 150 marine bacteria were isolated from Chorao Island, Goa, India, and preserved. In the present study, all these isolates were subjected for antagonistic activity against eight different Fusarium species. All the test fungi were procured from the NCIM-Resource Center, CSIR-National Chemical Laboratory Pune, India. All these strains were maintained on potato dextrose agar (PDA), pH-7.0 (Hi-media, Mumbai) throughout the study.
Screening of antifungal activity was carried out using the dual culture method on PDA (Tchameni et al. 2020). Test bacteria were streaked in a straight line at the center of a plate. Spot inoculation of fungi was made using a sterile loop, and placed 1.0 cm away from the inoculated test bacteria. A plate inoculated with the fungal pathogen alone was used as a control. Three replicates of each plate were incubated at 28°C for nine days and the growth inhibition of fungal pathogens was observed at 3, 5, 7 and 9 days. The per cent (%) inhibition growth of the test fungi was calculated by using the formula: Where R is the radial growth of fungal pathogens in the control plate and "r" is the radial growth of fungal pathogens in the dual culture plate. All the experiments were performed in triplicates.

Fermentation and extraction of antifungal compound
The production of the antifungal compound was accessed at ask level fermentation. A single colony of antagonistic bacteria from fresh nutrient agar plate were inoculated into the nutrient broth as a seed culture and kept on a rotary shaker at 28°C with 140 rpm for 18h. Subsequently, 5% inoculum was added to the 100 mL king's modi ed broth ((g/L) glycerol 30, peptone, K 2 HPO 4 0.5, MgSO 4 . 7H 2 O 0.5 and pH 7.0) as a production medium and incubated for ve days at 28°C. Cells were harvested after ve days of incubation by centrifugation at 10,000 rpm for 10 minutes. The resultant supernatant was acidi ed to pH 2.0 with 4N HCl, and subsequently extracted with an equal volume of ethyl acetate twice. The organic phase was collected and concentrated using a rotary evaporator. The crude extract was checked for its antifungal activity against all fungal strains. Based on the preliminary antifungal activity in crude extract, the antifungal compound was produced in a 14.0L lab scale fermenter (BioFlo/CelliGen 115) with 10.0L working volume. Dissolved oxygen was maintained above 50% by adjusting agitation 250 to 350 rpm till the end of fermentation batch.
The antifungal activity of the crude extract from the fermentation batch was evaluated by well diffusion method. The spore suspension of all the test fungi were prepared in 0.01% Tween 80 solutions from a 7day old grown culture. The nal concentration of spore suspension was adjusted to have 1.0x10 6 CFU/mL by hemocytometer. To sterile PDA media spore suspension was added with proper mixing. Spore mixed media was poured into the sterile petri dishes. After solidi cation, 100µL of crude extract from the 10.0 mg/mL stock was loaded in each well and 100µL of methanol was used as a solvent control, plates were incubated at 28°C for 72-96h, and antifungal activity was measured by the zone of inhibition in millimeter (mm).

Thin-layer chromatography (TLC) based Bio-autography
Bio-autography was performed according to (Grzelak et al. 2016) on TLC plate silica gel 60 (Merck, Darmstadt, Germany). The crude extract (30µL) was spotted 1.0 cm apart from the baseline into the silica gel plate and allowed to dry. The plate was then developed with Ethyl acetate: Pet ether (70:30) in a previously saturated glass chamber at room temperature. The developed plate is dried at room temperature, and the spots were visualized in a UV chamber at 254 nm. Bio-autographic evaluation of the crude extract is performed to check the antifungal activity of the separated compounds on the TLC plate. TLC developed plates were UV sterilized for 30 min in laminar air ow and placed in a petri dish, and 20.0 ml of 0.8% semi-solid potato dextrose agar having 1.0×10 6 spore/mL of F. foetens NCIM 1330 is poured over the TLC plate placed in the petri dish. After proper solidi cation, the petri dishes were incubated at 28°C for 72-96h to observe the zones of inhibition of active compounds separated on the TLC plates.

Puri cation and characterization
The crude extract of 9.6 gm obtained from fermentation is subjected for puri cation by column chromatography using 45× 7.5 cm column packed with 120-200-mesh silica. The column was eluted with Dichloromethane (DCM) to ensure that the column was properly packed. The crude extract was mixed with 60-120 mesh silica bed and applied uniformly from the top of the column with the combination of Dichloromethane and methanol (MeOH) (100 to 0% DCM with 0 to 100% MeOH ratio) in increasing order of the polarity. Forty mL fractions were collected. A total of 126 fractions were collected and checked for TLC pro les. All the 126 tubes were pooled into nine fractions based on their Rf value, these pooled fractions were further con rmed for their antifungal activity. The active fraction no. 3, was partially puri ed using a smaller silica column (35×3.0 cm). Gradient elution of the column was carried out with a combination of pet-ether and ethyl acetate by increasing the order of the polarity with ethyl acetate. 25mL each of the fractions were collected in 86 tubes, TLC pro ling was again based on their TLC similarity, and all the tubes were pooled into ve fractions. Subsequently, all ve fractions were recon rmed for their antifungal activity. The partially puri ed active fraction number F-3(II) was subjected to further puri cation by preparative thin-layer chromatography (PTLC) on pre-coated Silica Gel 60 254 plates (20×20 cm, Merck). The PTLC plates were developed in pet ether: ethyl acetate (70:30), and after air-drying, the plates were visualized under UV light (254 nm). The desired band was scratched from the PTLC plate and extracted with ethyl acetate. The solvent was evaporated by vacuum, and the weight of a puri ed compound was recorded.
The structure of the puri ed compound was established by nuclear magnetic resonance (NMR). The 1 H and 13 C NMR spectra were recorded on Bruker AV 500 MHz and 125 MHz, respectively, in deuterated chloroform (CDCl 3 ). The chemical shifts were given in δ, and ppm (parts per million) values referenced to the chloroform solvent at δ 7.27 in 1 H and 77.00 ppm in 13 C NMR.
1.0 mg/mL stock of puri ed compound and the crude extract were prepared in GC-MS grade methanol. From the stock, 1.0 µL was injected and GC-MS was carried out by using a 7890A gas chromatograph with a 5975C inert XL quadrupole mass spectrometer detector (MSD) (Agilent Technologies, USA) operated in electron ionization (EI) mode with a kinetic energy of the impacting electrons of 70 eV. The Restek Rtx®-5MS fused silica capillary column (30 m × 0.25 mm × 0.25 µm) with the non-polar stationary phase of 5% diphenyl / 95% dimethyl polysiloxane was used. The data was analyzed with the ChemStation software and validated with the NIST mass spectral library (Agilent Technologies). The oven temperature for the column was programmed (total 51.71 min), starting from 40°C withhold for 2 min, then rising with a ramp of 5°C/ min up to 180°C. Then further increased with the ramp of 7°C/min up to 220°C and nally ramped with 10°C per min up to 28°C with having 10 min hold. Helium (99.9% pure) was used as a carrier gas with a constant ow of 1.0 mL/min. The inlet temperature was kept at 250°C in split less mode. The axillary temperature was kept at 28°C. The EI ion source and quadrupole temperature were kept at 230°C and 15°C, respectively. Mass spectra and reconstructed total ion chromatograms (TIC) were obtained after 4 min solvent delay by automatic scanning in the uni ed mass range of 50-600 u. The retention time and mass fragmentation pattern compared with reference compounds identi ed as the possible compounds.
A standard of 2, 4-di-tert butyl-phenol (2, 4-TBP, CAS: 96-76-4, Sigma-Aldrich, Switzerland) (1.0 mg/mL) and 0.1mg/mL stock of puri ed compound were prepared in HPLC grade methanol for the comparative analysis. The puri ed compound and standard 2, 4-DTBP were analyzed by HPLC (Thermo Scienti c Dionex Ultimate 3000) using a C18 column (4.6×250mm, 5µm particle size thermo hypersil gold) with methanol: water with 0.1% tri uoroacetic acid (TFA) as a mobile phase, ow rate of 1.0 mL/min and detection at 254 nm. The 5.0µL sample from both standard 2, 4-DTBP and puri ed compounds were injected in isocratic mode with 80:20 of methanol and water for 14 minutes. Further, both compounds were compared with TLC using ethyl acetate: pet ether 70:30) as the solvent system, and detection was completed at UV 254 nm to nd the Rf values.

In-vitro antifungal effect
The compound 2, 4-DTBP is con rmed as a volatile in nature. Further, the compound is evaluated for its antifungal potential using the bi-compartment petri dish method (Massawe et al. 2018). For the antifungal activity, pure compound with 20.0 mg/mL working stock was prepared in methanol (Hi-media, Mumbai). The fungal mycelial plug of F. foetens NCIM 1330 from 7-day old culture was inoculated on the one compartment of potato dextrose agar. In another compartment having different concentrations, of compound i.e, 0.07, 0.26 and 0.53 mM were loaded on the 9.0 mm size disc in individual plates. Methanol is served as a control. Further, plates were sealed with para lm and incubated at 28°C for 5, 7 and 9 days.
The percent mycelial growth inhibition is calculated by using the below mentioned formula: Where R is the radial growth of fungal pathogens in the control plate and 'r' is the radial growth of the fungal pathogen in the treated plate. The experiment was conducted in triplicates.
Spore germination assay was carried according to (Wang et al, 2020). Spore suspension of 1.0 × 10 6 spores/mL was prepared in 0.01% tween 80 solution, 20µL was added to potato dextrose agar smear (diameter-12mm) on the glass slide for spore germination assay. Slides were placed in a petri dish and moist lter paper at the bottom to maintain 90-92% humidity. A sterile paper disc containing 2, 4-DTBP (0.07, 0.26 and 0.53 mM) were placed on the other side of the plate. Water and methanol were served as the negative and solvent control respectively. All the plates were prepared in triplicate and sealed with para lm were incubated at 28°C for 3, 6, 12 and 24h. At least 200 spores of each sample were examined at 40x under light microscopy (Nikon, Japan). Observation were made on germination by checking the germ tube length exceeded half of the diameter of spores.

In-vivo antifungal effect
Mature and healthy wheat seeds were purchased from the local market and used for the experiment. The assay was carried out according to method of (Zhang et al. 2021), seeds were cleaned by washing and soaked in water for 1.0h and autoclaved. All the seeds were spiked with a 1.0 × 10 6 conidia/mL suspension. In each petri dish, 20 seeds were placed at one corner of the plate. Antifungal volatile effect of 2, 4-DTBP is evaluated by different concentrations, i.e., 0.07, 0.26 and 0.53 mM, whereas water and methanol were used as negative and solvent control. All the plates were sealed with para lm and incubated at 28°C for seven days. All the experiments were performed in three experimental replications.
The protection of fungal infection on the wheat seeds was quanti ed based on infected seed counts as minutes, then rinsed twice with sterile distilled water for one minute and air-dried. Two hundred seeds were added per plate keeping the sterile disc containing 2, 4-DTBP at lowest 0.07 and highest of 1.06 mM concentration, methanol used as solvent control and incubated for 7 days at 28°C. A hundred seeds were randomly withdrawn after seven days of incubation and placed on a lter paper (pre-soaked in sterile water) and cultivated at 28°C. The germination percentages of wheat seeds were calculated.
Germination was considered, when the radicle protruded by 2 mm from the seeds.

Statistical analysis
Statistical analysis was carried by one-way ANOVA followed with Post hoc Tukey's (HSD) honestly signi cant difference test. The p-values < 0.05 were considered as statistically signi cant. Statistical analysis is performed using SPSS software version 26.0 (SPSS Inc., USA). Graph Pad Prism 8.0.2 software is used for plotting the graphs. Detailed statistical analysis data has been provided in the supplementary information. All experiments were carried out in triplicates and data were presented as mean ± standard deviation.

Screening for antagonistic activity
Among one hundred and fty marine bacteria were screened for antifungal activity, strain BKACT was identi ed as the potential antifungal candidate which inhibited > 50 percent of mycelial growth against all the Fusarium species tested (Table. 1). Under light and scanning electron microscopy, the hyphal morphology of treated fungi showed an abnormal, degraded and deformed shape in comparison to the control test ( Figure. 1a-f and Supplementary gure. 1). The highest percentage of mycelial growth inhibition of 75.56 ± 0.80 was observed against Fusarium foetens NCIM 1330 (Table. 1). Further, strain BKACT demonstrated a signi cant mycelial growth inhibition of 50 ± 2.0, 53.03 ± 2.14, 69.56 ± 1.22 and 75.56 ± 0.80% on 3rd, 5th, 7th and 9th day of incubation (Figure. 1g).

Molecular characterization
Identi cation of antifungal strain BKACT based on 16S rRNA gene sequencing analysis and NCBI-BLASTn con rmed it belongs to genus Serratia. The strain BKACT showed the highest sequence similarity to Serratia marcescens sub sp. ATCC 13880 and its phylogenetic analysis also con rms by forming similar clade with Serratia marcescens sub sp. ATCC 13880 ( Figure. 2). The 16S rRNA gene sequence of strain BKACT has been deposited in NCBI GenBank under accession number MT186165.

Antifungal activity and thin layer chromatography based bio-autography
The crude fermented extract of strain BKACT showed a signi cant antifungal activity when compared with solvent control against all the Fusarium species tested. The highest zone of inhibition (40 ± 1.0 mm) was observed against Fusarium foetens NCIM 1330 (Table. 1). For further identi cation and con rmation of antifungal fraction from the crude extract, thin layer chromatography (TLC) based bio autography is performed. At UV 254 nm, total six bands were observed on the TLC plate, out of these six, band number 5 showed antifungal activity against F. foetens NCIM 1330 (Supplementary gure. 2).

Pilot Scale (10L) production, puri cation and characterization
Based on its preliminary antifungal activity from crude extract, and active fraction from TLC bioautography. Strain BKACT is further subjected for pilot scale production at a 10.0L lab scale fermenter. From the 10.0L fermentation batch, 9.6 gm of crude extract was extracted, and the antifungal activity is recon rmed against the test fungi. The detailed information on puri cation of antifungal compound is schematically presented in (Supplementary Fig. 3). The puri ed antifungal compound further subjected characterized and structural elucidation by HPLC, NMR and GC-MS.
The 1 H NMR spectra of a puri ed compound showed three signals of aromatic protons at 7.31 (1H, d, J = 2.29 Hz), 7.09 (1H, dd, J = 8.39, 2.29 Hz), and 6.62 (1H, d, J = 8.39 Hz) which con rms the trisubstituted benzene and six methyl's at δ 1.43 (9H, s, 3 × CH 3 ) and 1.31 (9H, s, 3 × CH 3 ) con rmed the two di-tert butyl groups present on the benzene ring. The 13 C NMR spectra indicate the ten carbon signals in which six signals were in the aromatic region, which con rms the presence of the benzene ring. The 151.6 ppm of phenol substitution on the benzene ring and 29.6 and 31.6 ppm indicates the methyl signals of the tertiary butyl group. Using NMR and literature reports, we con rmed that the puri ed compound structure is 2, 4-di-tert butyl phenol (Supplementary gure. 4).
The puri ed compound from BKACT strain was further identi ed as the 2, 4-di-tert butyl-phenol by analyzing at GC-MS. The mass spectra of the identi ed peaks using pure standard substances were compared with peaks of the NIST mass spectral data to con rm the chemical structures of the detected compound as 2, 4-di-tert butyl-phenol (16.5 min) Ion [M] + at m/z 206, the fragment ion [M -CH 3 ] + at m/z 191 are characteristic for 2, 4-di-tert butyl-phenol ( Figure. 3). The same compound has been con rmed in the crude extract of the BKACT (Supplementary Fig. 5).
Based on the structure of a puri ed compound by NMR and GC-MS, the same compound was analyzed with TLC and HPLC. The puri ed compound was again con rmed as the 2, 4-di-tert butyl-phenol compared to the synthetic compound by peak pro les at the same retention time The volatile effect of the 2, 4-DTBP was qualitatively analyzed on spore germination. At the presence of solvent control and lower concentration 0.07mM, spores germinated normally and formed visible white mycelia as the incubation time increased. Even at 0.26mM concentration, the spore germination inhibition rate was initially signi cantly higher as compared to the control, but it decreases as the incubation time increases. At 0.53mM concentration of 2, 4-DTBP spore germination was completely (100%) suppressed at all the incubation times ( Figure. 5a). The per cent germination inhibition rate of the spores in the control and 0.07mM treatment groups was 28.74 ± 0.85 and 27.5 ± 5.14 respectively after 3h of incubation. As the time increases, the inhibition rate decreases to zero at 24 hrs. And at the 0.26mM concentration of compound, the rate of spore germination inhibition was 63.01 ± 5.6, 46.05 ± 2.94, 15.35 ± 1.4 and 8.33 ± 2.05 at 3, 6, 12 and 24 h of incubation, respectively ( Figure. 5b).

In-vivo antifungal volatile effect
The volatile antifungal effect of the 2, 4-DTBP was checked against F. foetens NCIM 1330 to protect wheat kernels. As the concentration of the compound increases, the visual growth of Fusarium foetens NCIM 1330 decreases. At 1.0 mM concentration, the growth was completely suppressed even after seven days of incubation ( Figure. 6a). The per cent seed contamination index (PSCI) was analyzed at a different compound concentration. The seed contamination index (SCI) in control and 0.07 mM concentration was 100 percent and 91.66 ± 2.35 at 0.26 mM concentration. At 0.53 mM concentration, the SCI was 46.66 ± 6.23, signi cantly less than the control. The PSCI was zero at 1.0 mM, which is con rmed as the effective concentration for protecting the wheat seed from F. foetens NCIM 1330 ( Figure.

Discussion
Fusarium species is a devastating pathogen in agriculture that cause severe loss of economically important plants such as wheat, maize, banana, tomato, sugarcane etc. It is also known to produce mycotoxins such as fumonisins, zearalenone, deoxynivalenol, fusaric acid and trichothecenes. Different species such as F. graminearum, F. proliferatum, F. tricinctum, F. moniliforme, F. verticillioides and F. Despite having negative impact on humans and the environment, many chemical fungicides are being used to control Fusarium attacks. However, due to its excessive and frequent use, the phytopathogenic fungi are able to acquire resistance to the existing fungicides. In the sense of searching safe, eco-friendly and sustainable alternative, biocontrol bacteria and their active components are considered the best choice (Köhl et al. 2019). In the past few decades' extensive study has been conducted on terrestrial bacteria and their active compounds against various plant pathogenic fungi. Still, the discovery of potential organisms and their novel metabolites is diminishing. Oceans are the most diverse, adverse and competitive ecosystem. To survive in such a unique environment, marine bacteria have developed adaptation mechanisms to produce unique biomolecules. Consequently, marine bacteria can produce bioactive compounds generally not found in terrestrial environments (Imhoff et al Regardless of antifungal and antioxidant activity, the 2, 4-DTBP has great volatile property. The microbial, volatile organic compounds (VOCs) have a signi cant role in disease management, especially for controlling the plant pathogenic fungi. VOCs are generally effective at minimal concentration, and they are capable of spreading in the atmosphere over the large distances. VOCs exert their inhibitory activity without direct physical contact with target pathogens (Schmidt et al. 2015). In the recent study 2, 4-DTBP at one mole per liter reported as the effective volatile concentration against Colletotrichum gloeosporioides (Wang et al. 2021). However, in our present study, we found that 2, 4-DTBP from S. marcescens showed a great volatile antifungal activity against F. foetens NCIM 1330 at minimal concentration of 0.53 mM as compared to reported values. Mycelial growth inhibition was signi cantly higher when compared to control, even at lower concentrations. The highest 86.6 ± 2.0% mycelial growth was inhibited at 0.53 mM concentration. At the same concentration, 100% spore inhibition was also observed. A signi cant difference in the inhibitory concentration could be attributed to the 2, 4-DTBP produced by different microorganisms and the solvent variation of DMSO (Wang et al. 2021) and methanol (used in the present study) for dissolution of 2, 4-DTBP. Varsha et al. (2015) in their study coated a 25 mg/ml concentration of 2, 4-DTBP on wheat seed which identi ed to protect from A. niger, F. chlamydosporum and F. moniliforme infections. Considering its antifungal activity and volatile property, we believe that, this molecule could control F. foetens on the wheat seeds. To date there is only one report highlighting F. foetens to produce mycotoxin such as beauvericin and fusaric acid in the cereal like maize (González-Jartín et al. 2019). Here for the rst time we observed in the absence of compound F. foetens infect wheat seed and grew easily at above 90% relative humidity. At 0.53 mM concentration of 2, 4-DTBP, percent seed contamination index (PSCI) was signi cantly lowered when compare to control. Hundred percent controls of F. foetens NCIM 1330 was observed at 1.0 mM concentration and it was identi ed as the effective treatment. Interestingly, the compound has not shown any adverse effect on the germination of wheat seeds at 1.0 mM concentration.

Conclusion
In the present study, we have found potential antagonistic marine S. marcescens BKACT, which produces an antifungal compound against Fusarium spp. This is the rst study that emphasizes on detailed puri cation and characterization of the 2, 4-DTBP from a marine S. marcescens strain BKACT. Additionally, the potential volatile antifungal effect on the growth of mycelia and spore germination of F. foetens NCIM 1330 was observed at 0.53 mM concentration. For the rst time, F. foetens NCIM 1330 was identi ed to infect wheat seeds and the 1.0 mM concentration of 2, 4-DTBP determined an effective concentration for controlling F. foetens NCIM 1330. At the same concentration no toxic effect were observed on seed germination. The current study is important as it provides a important observations that may be instrumental in the agricultural research. We conclude that the marine bacteria S. marcescens strain BKACT and its puri ed compound has promising antifungal potentials for the control of Fusarium spp. However, further evaluation of 2, 4-DTBP is required for various formulations utilizing its bioactive potentials and volatile characteristic in plant disease control. .

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.