Pseudomonas entomophila 23S Produces a Novel Antagonistic Compound against Clavibacter michiganensis subsp. michiganensis, a Pathogen of Tomato Bacterial Canker

: Biopesticides, which consist of antagonistic microorganisms, or the bioactive compounds they produce offer attractive alternatives to synthetic agrochemicals as a means of reducing plant diseases in crop production systems. In our previous study, a strain of Pseudomonas entomophila 23S was reported to have a strong antagonistic activity toward Clavibacter michiganensis subsp. michiganensis ( Cmm ) , a pathogen of tomato bacterial canker disease, which causes serious damage to the tomato production industry. This study focused on puriﬁcation and characterization of the anti- Cmm compound(s) produced by P. entomophila 23S. The cell-free supernatant was ﬁrst subjected to butanol extraction, fractioned by solid-phase extraction (SPE), and puriﬁed by high-performance liquid-chromatography (HPLC). The semi-preparative puriﬁed compounds retained anti- Cmm activity. These compounds were then characterized by liquid chromatography coupled with high-resolution tandem mass spectrometry (LC-HRMS/MS) and nuclear magnetic resonance (NMR). Two anti- Cmm compounds, C 15 H 16 N 2 O and C 16 H 18 N 2 O, were isolated by this procedure. From the LC/MS spectra, the compounds were predicted to have an isoquinoline ring, the presence of which was conﬁrmed by NMR. The compounds differed with respect to the presence of a methyl group in a side chain. In addition, P. entomophila 23S grown in different culture media produced the greatest amounts of these anti- Cmm compounds when in Nutrient Broth (NB), where bacterial growth was the slowest. NB composition is more favorable in stimulating P. entomophila 23S to produce higher amounts of these anti- Cmm compounds.


Introduction
Plant diseases caused by viruses, bacteria, and fungi are responsible for significant crop losses in agriculture. Current control measures rely largely on the application of chemical pesticides [1]. However, chemical pesticides have been criticized for detrimental effects on human and environmental health [2]. Their widespread use has also resulted in development of pest resistance [3]. Because of these concerns, many pesticides are no longer available or are being targeted for removal from the marketplace [4]. Therefore, eco-friendly alternatives are urgently needed to control crop disease organisms and other pests.
Microorganism-based biological control offers an effective alternative to synthetic chemical treatment of plant diseases. Biopesticides, which consist of antagonistic microorganisms, or the bioactive compounds they produce have been increasingly researched and

Butanol Extraction of Bacterial Culture
Pseudomonas entomophila 23S was prepared in 2 L of broth culture, in a 4-L flask at 100 rpm, for 5 days to reach stationary phase, with a total of 16 L of bacterial culture produced, following the methods described in [30]. In preliminary experiments with different organic solvents, butanol extracts showed the greatest amount of anti-Cmm activity, the butanol was selected for the extraction of anti-Cmm compounds from the bacterial culture. Eight hundred mL of 1-butanol were then added to each 2 L of bacterial culture, shaken occasionally and left overnight to allow for complete phase partitioning. The upper butanol layers were collected from all the flasks and combined, and the butanol was evaporated using a rotary evaporator (90 rpm, 50 • C; Yamato, NJ, USA). The dried material remaining after the evaporation was resuspended in 30 mL of 10% acetonitrile (ACN/H 2 O, v/v). The extract was then centrifuged at 4 • C and 18,000 rpm for 30 min to remove large particles, and the supernatant was then filtered through a 0.22 µm syringe membrane (Mixed Cellulose Ester; MilliporeSigma, Darmstadt, Germany). The filtered extract was tested for anti-Cmm activity by an agar well diffusion assay, as described below.

Agar Well Diffusion Assay
The agar well diffusion assay followed the general method of [31]. One hundred µL of Cmm culture (10 8 cfu mL −1 ) were spread onto a Nutrient Broth Yeast Extract (NBYE; per L distilled H 2 O: 8.0 g nutrient broth, 2.0 g yeast extract, 2.0 g K 2 HPO 4 , 0.5 g KH 2 PO 4 , 5.0 g glucose, 0.25 g MgSO 4 •H 2 O, and 15 g agar) agar plate. Wells with a diameter of 6 mm were made in the plate agar with a cork borer, and 80 µL of the sample extracted from P. entomophila 23S culture were loaded into each well, except the control, which was loaded with 10% acetonitrile solution. The petri plates were sealed with parafilm and incubated for at least 48 h at 28 • C. The zone of inhibition was measured (in millimeters) as the straight-line distance from the edge of the well to the edge of the zone of inhibition.

Solid-Phase Extraction (SPE)
The butanol extract obtained from the above procedures and showing anti-Cmm activity in vitro was subjected to solid-phase extraction. The butanol extract (30 mL) was first diluted to 100 mL with 10% acetonitrile and then loaded onto Extract-Clean TM C18 10,000 mg (75 mL) −1 SPE cartridges (Alltech, Nicholasville, KY, USA) with 25 mL per column (using a total of four columns). As an eluent, acetonitrile was applied at increasing concentrations: starting from 10% at 50 mL, then 20, 30, 40, 50, 60, and 70% each at 25 mL and, lastly, 80% at 50 mL per column. The eluate (fractions) was collected and tested for anti-Cmm activity by the agar well diffusion assay, as described above.
For the 50, 60, 70, and 80% fractions, their effects on the Cmm growth curve were also studied. For this experiment, the fractions were lyophilized, weighed out, and resuspended in 50% acetonitrile. Ten concentrations (1.0, 2.0, 4.0, 8.0, 16, 32, 64, 128, 256, and 512 µg mL −1 ) were prepared in a 96-well plate according to the microdilution method established by [32]. The initial Cmm culture was adjusted to 5 × 10 5 cfu mL −1 in NB medium. The Cmm in the NB medium without any fractions served as a growth control while the NB media without fractions and the Cmm served as a sterile control. The plate was incubated at 28 • C for 2 days using a Cytation Gen5 TM Biotek microplate reader (Biotek Instruments Inc., Winooski, VT, USA). The plate was agitated through double-orbital shaking, and the optical density at 600 nm was measured every 6 h. The data were analyzed with the SAS package program version 9.4 (SAS institute Inc., Cary, NC, USA) using the proc-mixed model, with time points and concentrations as fixed effects and well plates as a random effect. Tukey's test was applied for multiple comparisons with least square means when there was a significant treatment effect at the 95% confidence level (p < 0.05).

Comparisons of SPE Fractions from Different Strains
Active SPE fractions were analyzed by HPLC but, to determine the active peaks, the SPE fractions were also prepared from the NB media only (no bacteria grown) and from another strain of Pseudomonas sp. (accession number KJ534477, hereafter referred to as Pseudomonas sp. KJ). Pseudomonas sp. KJ is a plant growth promoting rhizobacterium of tomato plants, as studied by Ricci [33]. Butanol extracts and SPE fractions from Pseudomonas sp. KJ, grown in NB media, were prepared in the same way (i.e., same volume with same extraction methods) as P. entomophila 23S extracts. The butanol and SPE fractions from both sources did not show anti-Cmm activity, as indicated by agar well diffusion assay results (not shown) and, thus, this strain was used to help identify the active peaks identified by HPLC from the SPE fractions prepared from P. entomophila 23S, which had anti-Cmm activity. Preparation of the SPE fractions was conducted as described above.

LC-HRMS/MS Analysis
LC-MS/MS analysis was performed on a Shimadzu Nexera UHPLC (Shimadzu, Kyoto, Japan) coupled to a high-resolution Sciex Triple TOF 5600 (quadrupole time-offlight, QqTOF) system (SCIEX, Concord, ON, Canada). Samples were injected onto a Kinetex C18 50 × 2.1-mm column, with solid core 1.7-µm particles (100 Å) (Phenomenex, Torrance, CA) using a Nexera UHPLC system with water (A) and ACN (B), both containing 0.1% formic acid, at a flow rate of 300 µL min −1 (40 • C). The gradient started at 20% B, was held for 1 min, and was then linearly increased to 50% B at 12 min, to 85% B at 14 min, and held for 2 min before column re-equilibration. MS and MS/MS spectra were collected in positive and negative ion modes using a DuoSpray ion source at 5 kV source voltage, 450 • C source temperature, and 50 psi GS1/GS2 gas flows, with a declustering potential of 80 V. The instrument performed a survey TOF-MS acquisition from m/z 80 to 1000 (150-ms accumulation time), followed by MS/MS on precursor ions m/z 241.135 and 255.1505. Each MS/MS acquisition had an accumulation time of 150 ms and with collision energy of 30 V. High-resolution MS/MS spectra were visualized using PeakView software (Sciex, version 2.1).

HPLC Semi-Preparative Purification
Purification of extracts was conducted on a semi-preparative scale (5 mL min −1 ), Agilent 1260 Infinity HPLC system (Agilent Technologies, Santa Clara, CA, USA), with UV detection at 254 nm (see results for more information of fractions purified). Samples were concentrated by drying under nitrogen until approximately one-half of the original volume remained. Concentrated samples were centrifuged (5 min; 140,000 rpm) and the supernatant was transferred to vials. An Agilent Pursuit XRs C18 column (5 µm; 250 × 10 mm) was employed at a flow rate of 5 mL min −1 . Mobile phases A and B were water and acetonitrile, both containing 0.1% formic acid. The gradient started at 30% B, which was held for 1 min after sample injection. The mobile phase B was then increased to 70% from 1-4 min and then to 85% for 4-14 min. Next, the column was washed at 95% B from 14.5-16 min, prior to re-equilibration at 30% B. Sample injection volume was 900 µL. Fractions collected were analyzed by LC-UV-MS and then subjected to biological testing to confirm activity.

LC-UV-MS Analysis of HPLC Purified Fractions
LC-UV-MS analysis was performed on an Agilent 1200 HPLC coupled to an Agilent 6210 ESI-TOF mass spectrometer. HPLC fractions were injected onto an Eclipse Plus C18 column (50 × 3 mm, 1.8-µm particles) (Agilent Technologies, Santa Clara, CA, USA) with water (A) and ACN (B), both containing 0.1% formic acid, at a flow rate of 400 µL min −1 . The gradient started at 30% B, was held for 1 min, and was then linearly increased to 50% B over 4 min, to 75% B at 11 min, to 95% at 11.5 min, and held for 2.5 min before column reequilibration. The UV detector was set to 254 nm and MS spectra were collected in positive ion mode from m/z 100-1000. High-resolution MS spectra were internally calibrated for accurate mass measurements.

NMR Characterization of HPLC Fraction 1
NMR spectra were acquired on a Bruker Avance III-HD using a double-resonance, 5-mm BBFO probe. Samples were dissolved in methanol-d 4 and 1 H, COSY, 13 C, and 1 H-13 C HSQC spectra with multiplicity edition spectra recorded at room temperature. The 1 H and 13 C 90 • pulses were 12.9 and 12 µs, respectively. Recycle delays were 1 s for the 1H spectrum and 1.4 s for all other spectra. Acquisition times were 2.7 s ( 1 H), 0.128 s (COSY), 0.9 s ( 13 C), and 0.166 s (HSQC). Sixty-four transients were accumulated for the 1 H spectrum and 1024 for the 13 C spectrum. For the COSY spectrum, two scans were accumulated for each of the 256 increments in the indirect dimension and the spectral widths were 13.36 ppm. For the HSQC spectrum, two scans were accumulated for each of the 648 increments in the indirect dimension and the spectral widths were 10.29 ppm ( 1 H) and 180 ppm ( 13 C). All spectra were referenced with respect to the residual methanol peak calibrated at 3.31 ppm.

Effects of Media on Anti-Cmm Activity of Butanol Extracts
In addition, butanol extracts were also prepared from three additional media (other than NB): (1) King's B (KB), (2) Luria Broth (LB), and (3) Tryptic Soy Broth (TSB). Preparation of KB, LB, and TSB were prepared as follows: for KB, per L of distilled H 2 O-proteose peptone No.3 20 g, glycerol 10 mL, K2HPO4 1.5 g, MgSO4 1.5 g; for LB, per L of distilled H 2 O-tryptone 10 g, NaCl 10 g, yeast extract 5 g; and for TSB, per L of distilled H 2 Otryptic soy broth 30 g. Pseudomonas entomophila 23S was grown in each medium and extracted with butanol as described above and the four extracts were tested for anti-Cmm activity. At the same time, the effect of each medium on the growth rate of P. entomophila 23S was studied by using a 96-well plate, which was read for cell density at regular intervals. The initial culture was adjusted to 5 × 10 5 cfu mL −1 for each medium and 100 µL was applied in each well. Each medium, without the culture, served as a sterile control. The plate was incubated at 28 • C for 2 days using the Cytation Gen5 TM Biotek microplate reader. The plate was double orbitally shaken, and the optical density at 600 nm was measured every 2 h. The data were analyzed with the SAS package program version 9.4 (SAS institute Inc., Cary, NC, USA) using the proc-mixed model, with time points and concentrations as fixed effects. Tukey's test was applied for multiple comparisons for the least square means when there was a significant treatment effect at the 95% confidence level (p < 0.05).

Effects of the Two Purified Anti-CMM Compounds on Cmm Growth Curves
The two isolated anti-Cmm compounds were used to monitor their effects on Cmm growth. The experiment was performed in the same manner as the study of the effects of SPE fraction on the Cmm growth curve, as described above, by using a Gen5 TM Biotek microplate reader. Due to the limited amounts of purified compounds obtained, the tested concentration was 0.5-32 µg mL −1 for the first compound (m/z 241 (MH + )) and was 0.5-256 µg mL −1 for the second compound (m/z 255 (MH + )).

Anti-Cmm Activity of Butanol Extract and Anti-Cmm Activity of Solid-Phase Extraction (SPE) Fractions
The butanol extract, obtained from P. entomophila 23S grown in NB media, was tested for anti-Cmm activity with a disk diffusion assay on NBY plates (Figure 1), resulting in the inhibition of Cmm growth, as indicated by zone of inhibition (10 mm).

Anti-Cmm Activity of Butanol Extract and Anti-Cmm Activity of Solid-Phase Extraction (SPE) Fractions
The butanol extract, obtained from P. entomophila 23S grown in NB media, was tested for anti-Cmm activity with a disk diffusion assay on NBY plates (Figure 1), resulting in the inhibition of Cmm growth, as indicated by zone of inhibition (10 mm).
. The butanol extract was then subjected to solid-phase extraction (SPE), using 10 to 80% ACN as eluent, and each fraction was assessed for anti-Cmm activity. The disk diffusion assay showed that among the eight fractions, the 50, 60, 70, and 80% SPE fractions inhibited the growth of Cmm, as indicated by zone of inhibition (Table 1). In addition, the 50, 60, 70, and 80% SPE fractions were also used to study their effect, at different concentrations, on Cmm-growth curves using spectrophotometry ( Figure 2). When statistical tests were performed, the concentrations that significantly inhibited the growth of Cmm, as compared with control, were: 64, 128, 256, and 512 μg mL −1 for the 50% fraction; 128, 256, and 512 μg mL −1 for the 60% fraction; 128, 256, and 512 μg mL −1 for the 70% fraction; and 64, 128, 256, and 512 μg mL −1 for the 80% fraction. The effects of other concentrations were not significantly different from those of the control and some actually enhanced the growth of Cmm. The butanol extract was then subjected to solid-phase extraction (SPE), using 10 to 80% ACN as eluent, and each fraction was assessed for anti-Cmm activity. The disk diffusion assay showed that among the eight fractions, the 50, 60, 70, and 80% SPE fractions inhibited the growth of Cmm, as indicated by zone of inhibition (Table 1). In addition, the 50, 60, 70, and 80% SPE fractions were also used to study their effect, at different concentrations, on Cmm-growth curves using spectrophotometry ( Figure 2). When statistical tests were performed, the concentrations that significantly inhibited the growth of Cmm, as compared with control, were: 64, 128, 256, and 512 µg mL −1 for the 50% fraction; 128, 256, and 512 µg mL −1 for the 60% fraction; 128, 256, and 512 µg mL −1 for the 70% fraction; and 64, 128, 256, and 512 µg mL −1 for the 80% fraction. The effects of other concentrations were not significantly different from those of the control and some actually enhanced the growth of Cmm.

LC-MS Analysis of SPE Fractions from Two Strains of Pseudomonas
To characterize the SPE anti-Cmm peaks produced by P. entomophila 23S, SPE fractions were prepared from three sources: (1) NB media only, (2) P. entomophila 23S grown in NB, and (3) Pseudomonas sp. strain KJ grown in NB. Because the fractions prepared from the NB media only, or from the Pseudomonas sp. strain KJ grown in NB, did not possess anti-Cmm activities, the unique peaks seen in the LC-MS analysis of the three fractions (60, 70, and 80% ACN) from the 23S strain were tentatively deemed responsible for the anti-Cmm biological activity. Two peaks of interest were found only in the 23S strain fractions and are shown in Figure 3a, where extracted ion chromatograms were overlaid for the three samples (NB media control and the two strains grown in NB). These two peaks of interest yielded protonated molecules with mass-to-charge ratios (m/z) of 241.1341 and 255.1499, corresponding to neutral formulae of C15H16N2O (2.4 ppm mass error) and C16H18N2O (2.9 ppm), respectively. Figure 3b shows the high-resolution MS/MS spectra for these two protonated molecules.

LC-MS Analysis of SPE Fractions from Two Strains of Pseudomonas
To characterize the SPE anti-Cmm peaks produced by P. entomophila 23S, SPE fractions were prepared from three sources: (1) NB media only, (2) P. entomophila 23S grown in NB, and (3) Pseudomonas sp. strain KJ grown in NB. Because the fractions prepared from the NB media only, or from the Pseudomonas sp. strain KJ grown in NB, did not possess anti-Cmm activities, the unique peaks seen in the LC-MS analysis of the three fractions (60, 70, and 80% ACN) from the 23S strain were tentatively deemed responsible for the anti-Cmm biological activity. Two peaks of interest were found only in the 23S strain fractions and are shown in Figure 3a, where extracted ion chromatograms were overlaid for the three samples (NB media control and the two strains grown in NB). These two peaks of interest yielded protonated molecules with mass-to-charge ratios (m/z) of 241.1341 and 255.1499, corresponding to neutral formulae of C 15 H 16 N 2 O (2.4 ppm mass error) and C 16 H 18 N 2 O (2.9 ppm), respectively. Figure 3b shows the high-resolution MS/MS spectra for these two protonated molecules. From this data, the formulae for fragment ions were confirmed (Tables 2 and 3) and, hence, elucidating certain structural features of these compounds was possible. It was apparent that the two compounds were structurally very similar, with only one CH 2 group difference. By studying the fragmentation pathways, these compounds were tentatively determined to contain an isoquinoline ring system as well as an alkyl side chain, where the mentioned extra CH 2 group was present in the protonated molecule at m/z 255. Two of the most active SPE fractions (60 and 70%) were subjected to HPLC purification to better characterize the putative biologically active compounds. From the 60% SPE fraction, three fractions were collected. From the 70% SPE fraction, two fractions were collected. Representative HPLC traces from this purification showing the fractionation windows collected are provided as Supplementary Figure S1. The fractions were tested for purity by LC-UV-TOF-MS (see Methods) and five out of six were found to contain only one peak of interest in the UV and total ion chromatogram traces. Fraction 3 from the 60% sample and Fraction 1 from the 70% samples had a protonated molecule at m/z 241.1348 and Fraction 2 from the 70% sample showed a protonated molecule at m/z 255.1490 (Supplementary Figure S2). After multiple HPLC injections, appropriate fractions were pooled, concentrated under nitrogen, and then subjected to LC-MS/MS and NMR analysis, as well as testing for biological activity (Supplementary Figure S3). Peaks eluting at 9.7 and 11.3 min were collected and pooled into Fractions 1 and 2, respectively. LC-MS/MS analysis confirmed these two fractions to be the two compounds of interest at m/z 241 and 255, respectively. NMR analysis was conducted on the first collected fraction, as the concentration and purity of the second fraction were not high enough for proper structural characterization. The 1 H and COSY NMR spectra were acquired (Supplementary Figure S4.1), and HSQC analysis (Supplementary Figure S4.2) confirmed the presence of the quinolone ring as well as the terminal alkyl chain, as proposed in Figure S3. While an absolute structure of this compound could not be confirmed at this time, there was compelling evidence for it. The NMR spectra for the first compound (m/z 241) were consistent with the high-resolution MS/MS data. The proposed fragmentation of both protonated and deprotonated precursors are shown in supplemental figures (Supplementary Figures S5 and S6). The second compound (m/z 255) differed from the first compound (m/z 241) by CH 2 , as suggested by the MS/MS data, in the terminal alkyl chain of the molecule.

Effects of Different Media on Anti-Cmm Activity of Butanol Extracts
Butanol extracts were prepared with P. entomophila 23S grown in different culture media, King's B (KB), Luria broth (LB), and tryptic soy broth (TSB), to compare the relative amounts of putative anti-Cmm compounds produced when the bacterium was grown in NB. Only the butanol extracts made from NB-grown culture showed anti-Cmm activity (Supplementary Figure S7).
To investigate whether differences in the anti-Cmm activity of butanol extracts prepared from different media were due to the number of cells, growth of the P. entomophila 23S was compared when it was grown in NB, KB, LB, and TSB (Supplementary Figure S7e). Statistical analysis indicated that the growth of the P. entomophila 23S began to differ at 8 h. The value of optical density (OD 600 ) was significantly higher for the TSB than other media. At the same time, the value for NB was significantly lower than for other media. This trend continued to develop during the 20 h of growth.
In addition, the four butanol extracts were analyzed by LC-MS and checked for the presence of the two active compounds (C 15 H 16 N 2 O and C 16 H 18 N 2 O). As shown by the overlaid, extracted ion chromatograms from LC-MS analysis (Figure 4), peaks for protonated ions at m/z 241 and 255 were both present for all four media. However, the peak intensities for the NB-prepared butanol extracts were much higher than those for KB-, LB-, and TSB-prepared butanol extracts.

Effects of Different Media on Anti-Cmm Activity of Butanol Extracts
Butanol extracts were prepared with P. entomophila 23S grown in different culture media, King's B (KB), Luria broth (LB), and tryptic soy broth (TSB), to compare the relative amounts of putative anti-Cmm compounds produced when the bacterium was grown in NB. Only the butanol extracts made from NB-grown culture showed anti-Cmm activity (Supplementary data S7).
To investigate whether differences in the anti-Cmm activity of butanol extracts prepared from different media were due to the number of cells, growth of the P. entomophila 23S was compared when it was grown in NB, KB, LB, and TSB (Supplementary data S7(e)). Statistical analysis indicated that the growth of the P. entomophila 23S began to differ at 8 h. The value of optical density (OD600) was significantly higher for the TSB than other media. At the same time, the value for NB was significantly lower than for other media. This trend continued to develop during the 20 h of growth.
In addition, the four butanol extracts were analyzed by LC-MS and checked for the presence of the two active compounds (C15H16N2O and C16H18N2O). As shown by the overlaid, extracted ion chromatograms from LC-MS analysis (Figure 4), peaks for protonated ions at m/z 241 and 255 were both present for all four media. However, the peak intensities for the NB-prepared butanol extracts were much higher than those for KB-, LB-, and TSBprepared butanol extracts.

Effects of Two Purified Anti-Cmm Compounds on Cmm Growth Curve
The two isolated anti-Cmm compounds were also studied in terms of their effects on the Cmm growth curve by spectrophotometry ( Figure 5). For the first compound (C 15 H 16 N 2 O, m/z 241), all the concentrations used, from 0.5 to 32 µg mL −1 , significantly inhibited the growth of Cmm when compared with control. For the second compound (C 16 H 18 N 2 O, m/z 255), concentrations ranging from 16 to 254 µg mL −1 significantly suppressed the Cmm growth, as compared with the control, while concentrations between 1.0 to 8.0 µg mL −1 promoted Cmm growth.

Effects of Two Purified Anti-Cmm Compounds on Cmm Growth Curve
The two isolated anti-Cmm compounds were also studied in terms of their effects on the Cmm growth curve by spectrophotometry ( Figure 5). For the first compound (C15H16N2O, m/z 241), all the concentrations used, from 0.5 to 32 μg mL −1 , significantly inhibited the growth of Cmm when compared with control. For the second compound (C16H18N2O, m/z 255), concentrations ranging from 16 to 254 μg mL −1 significantly suppressed the Cmm growth, as compared with the control, while concentrations between 1.0 to 8.0 μg mL −1 promoted Cmm growth.

Discussion
Butanol extracts, prepared from cultures of the P. entomophila 23S clearly showed anti-Cmm activity, implying that the solvent extraction was efficient. Following the butanol extraction, solid-phase extraction was conducted and the results indicated that the 50, 60, 70, and 80% fractions possessed anti-Cmm activity, suggesting that the more hydrophobic fractions had inhibitory effects toward Cmm. The size of the zone of inhibition was larger for the 60, 70, and 80% fractions than the 50% fraction, implying that the superior anti-Cmm activities for the 60, 70, and 80% fractions were due to greater concentrations of the active material(s). In addition to the anti-Cmm compounds, some growth-promoting compounds might have been present in the fractions and, at certain concentrations, these growth-promoting effects of the fractions were greater than the growth-inhibiting effects, resulting in enhanced Cmm growth. This concentration-dependent effect on Cmm growth was also observed when Cmm was grown with purified compounds.
As a result of purification and analyses, we isolated two anti-Cmm compounds whose structures were identified as having isoquinoline as their largest part. Isoquinoline alkaloids containing a heterocyclic isoquinoline backbone represent a diverse group of compounds, such as antibacterial berberine and sanguinarine, antitussive and anti-cancer noscapine, anti-gout colchicine, and analgesic compounds. These compounds are valued as natural medicinal and pharmaceutical natural products [34,35]. They can be found commonly in plants, although a few fungal (e.g., Aspergillus fumigatus [36], Penicillium sp. [37]) and bacterial species (Streptomyces sp. [38]) produce isoquinoline alkaloids with new antibacterial activities at this time. Isoquinoline alkaloids are well known for their antimicrobial activities against a wide range of human pathogenic bacteria and phytopathogenic fungi [34,39]. For instance, recently discovered, novel isoquinoline alkaloids were confirmed for bioactivities against human pathogens, such as Bacillus cereus, Staphylococcus aureus, Salmonella typhi, and Escherichia coli 100 [40,41], as well as phytopathogenic fungi such as Fusarium culmorum and Geotrichum candidum [42]. Two modes of action have been proposed, inhibiting cell division and inhibiting nucleic acid synthesis [43], although exact

Discussion
Butanol extracts, prepared from cultures of the P. entomophila 23S clearly showed anti-Cmm activity, implying that the solvent extraction was efficient. Following the butanol extraction, solid-phase extraction was conducted and the results indicated that the 50, 60, 70, and 80% fractions possessed anti-Cmm activity, suggesting that the more hydrophobic fractions had inhibitory effects toward Cmm. The size of the zone of inhibition was larger for the 60, 70, and 80% fractions than the 50% fraction, implying that the superior anti-Cmm activities for the 60, 70, and 80% fractions were due to greater concentrations of the active material(s). In addition to the anti-Cmm compounds, some growth-promoting compounds might have been present in the fractions and, at certain concentrations, these growth-promoting effects of the fractions were greater than the growth-inhibiting effects, resulting in enhanced Cmm growth. This concentration-dependent effect on Cmm growth was also observed when Cmm was grown with purified compounds.
As a result of purification and analyses, we isolated two anti-Cmm compounds whose structures were identified as having isoquinoline as their largest part. Isoquinoline alkaloids containing a heterocyclic isoquinoline backbone represent a diverse group of compounds, such as antibacterial berberine and sanguinarine, antitussive and anti-cancer noscapine, anti-gout colchicine, and analgesic compounds. These compounds are valued as natural medicinal and pharmaceutical natural products [34,35]. They can be found commonly in plants, although a few fungal (e.g., Aspergillus fumigatus [36], Penicillium sp. [37]) and bacterial species (Streptomyces sp. [38]) produce isoquinoline alkaloids with new antibacterial activities at this time. Isoquinoline alkaloids are well known for their antimicrobial activities against a wide range of human pathogenic bacteria and phytopathogenic fungi [34,39]. For instance, recently discovered, novel isoquinoline alkaloids were confirmed for bioactivities against human pathogens, such as Bacillus cereus, Staphylococcus aureus, Salmonella typhi, and Escherichia coli 100 [40,41], as well as phytopathogenic fungi such as Fusarium culmorum and Geotrichum candidum [42]. Two modes of action have been proposed, inhibiting cell division and inhibiting nucleic acid synthesis [43], although exact actions were not completely elucidated [39]. Derived from tyrosine, quinoline synthesis in plants has been characterized in detail [35,44]. Xu et al. [16] reported 12 anti-Cmm compounds identified by screening a chemical library of yeast-active molecules. These compounds had piperidines, benzimidazoles, phenols, phenoxy isopropanolamines, or pyrrolidone as base structures. Isoquinoline seems not to have been previously studied or/and reported to have antimicrobial activity toward Cmm. When PubChem (https://pubchem.ncbi.nlm.nih.gov accessed on 1 March 2021) was used to search for this isoquinoline-based structure, the system indicated that there were "zero results found", suggesting novelty for the isolated compound.
Since the largest production of anti-Cmm compounds occurred in the NB media where bacterial growth was the slowest, greater anti-Cmm activities observed from the NB extracts were not due to greater bacterial growth, but due to higher concentrations of the active compounds. The production of anti-Cmm compounds by P. entomophila 23S might be occurring in response to less favorable culture nutrient conditions leading to stress. Microorganisms often switch on specific aspects of secondary metabolism under stressful conditions, as a strategy to remain competitive and survive in their environment [9,45]. Among the four media tested, P. entomophila 23S grew best in TSB. The main difference between TSB and other media is that TSB contains soybean meal, a plant component. Pseudomonas entomophila 23S was originally isolated from the rhizosphere of soybean plants [7]. Thus, the TSB might have been more favorable for growth of this bacterium than the other evaluated media. Specific ingredients, or the lack thereof, in the NB media may be acting to inhibit the growth of P. entomophila 23S and, at the same time, resulting in induced stress, causing the bacteria to produce anti-Cmm compounds in this medium at much higher levels.
In this study, the two isolated compounds from P. entomophila 23S inhibited Cmm growth. Although their full structures were not confirmed, the data indicated that these compounds, with molecular formulae of C 15 H 16 N 2 O and C 16 H 18 N 2 O, differ only by one CH 2 group and that both have an isoquinoline ring as the base component of structure. These appear to be novel compounds with regard to inhibition of Cmm growth and when grown in NB medium. Given the seriousness of Cmm-caused bacterial canker and lack of very effective control methods, these compounds would be of great potential as new biocontrol agents for the disease.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/applmicrobiol1010006/s1. Figure S1: Representative HPLC-UV chromatograms (at 265 nm) from semi-preparative purification of SPE fractions (A) 70% extract showing two fractions collected. Figure S2: Representative high-resolution extracted ion chromatograms (left) and mass spectra (right) for two fractions (F1 and F2) from 70% SPE sample, purified by HPLC. F1 and F2 contain molecular species, giving MH + at m/z 241.1348 and m/z 255.1490, respectively. The complete ion chromatograms and UV traces (at 254 nm) showed no other peaks present in these purified fractions. Figure S3: Anti-Cmm activities of HPLC fractions (1 and 2) collected from 60 and 70% SPE fractions: (a) 60% fraction 1, (b) 60% fraction 2, (c) 60% fraction 3, (d) 70% fraction 1, (e) 70% fraction 2, and (f) 50% acetonitrile (control). Figure S4.1: The 1 H and COSY NMR spectra of fraction 1 in MeOD. The proposed structure, which satisfies the chemical data, multiplicities, and COSY correlations, is shown in insert. The numbers in parentheses after the peak positions in the 1-D spectra identify the various protons on the structure. Peaks at 3.3 and 4.8 ppm correspond to residual methanol and water peaks. Figure S4.2: The { 1 H-13C} HSQC spectrum with multiplicity edition of fraction 1 in methanol-d4. Figure S4.3: Aromatic region of the { 1 H-13C} HSQC spectra of fractions 1 (top) and 2 (bottom), showing the conserved quinoline peaks. Figure S5. Proposed fragment ions formed from protonated precursor ion at m/z 241 based on high-resolution MS/MS data presented in manuscript. Figure S6: Proposed fragment ions and high-resolution MS/MS data from deprotonated precursor ion at m/z 239. Figure S7: Anti-Cmm activity of butanol extracts prepared from (a) nutrient broth, (b) King's B, (c) Luria broth, and (d) tryptic soy broth, and growth of P. entomophila 23S in each media (e).