Detoxification of Fusaric Acid by the Soil Microbe Mucor rouxii

May 24, 2017 - Fusarium oxysporum f. sp. vasinfectum race 4 (VCG0114), which causes root rot and wilt of cotton (Gossypium hirsutum and G. barbadense)...
0 downloads 0 Views 437KB Size
Download PDF
Article pubs.acs.org/JAFC

Detoxification of Fusaric Acid by the Soil Microbe Mucor rouxii Frankie K. Crutcher,†,§ Lorraine S. Puckhaber,† Alois A. Bell,† Jinggao Liu,† Sara E. Duke,† Robert D. Stipanovic,*,† and Robert L. Nichols‡ †

USDA, Insect Control and Cotton Disease Research Unit, Southern Plains Agricultural Research Center, Agricultural Research Service, USDA, 2765 F&B Road, College Station, Texas 77845, United States ‡ Cotton Incorporated, 6399 Weston Parkway, Cary, North Carolina 27513, United States S Supporting Information *

ABSTRACT: Fusarium oxysporum f. sp. vasinfectum race 4 (VCG0114), which causes root rot and wilt of cotton (Gossypium hirsutum and G. barbadense), has been identified recently for the first time in the western hemisphere in certain fields in the San Joaquin Valley of California. This pathotype produces copious quantities of the plant toxin fusaric acid (5-butyl-2pyridinecarboxylic acid) compared to other isolates of F. oxysporum f. sp. vasinfectum (Fov) that are indigenous to the United States. Fusaric acid is toxic to cotton plants and may help the pathogen compete with other microbes in the soil. We found that a laboratory strain of the fungus Mucor rouxii converts fusaric acid into a newly identified compound, 8-hydroxyfusaric acid. The latter compound is significantly less phytotoxic to cotton than the parent compound. On the basis of bioassays of hydroxylated analogues of fusaric acid, hydroxylation of the butyl side chain of fusaric acid may affect a general detoxification of fusaric acid. Genes that control this hydroxylation may be useful in developing biocontrol agents to manage Fov. KEYWORDS: Fusarium oxysporum f. sp. vasinfectum race 4 (VCG0114), cotton, Gossypium hirsutum, Gossypium barbadense, fusaric acid, 8-hydroxyfusaric acid, California Fov race 4

■

INTRODUCTION Fusaric acid, 1 (Figure 1), is a plant toxin that may play a role in the pathogenicity of Fusarium oxysporum f. sp. vasinfectum

biocontrol agent that can detoxify fusaric acid may provide an environmentally advantageous strategy to manage this pathogen. To this end, we have investigated various soil microorganisms to identify those that metabolize fusaric acid. We were particularly interested in those that survived when challenged with a relatively high dose of fusaric acid (200 μg/ mL). Since we were ultimately interested in identifying unique genes (preferably, a single gene) that could be incorporated into a biocontrol agent to degrade/detoxify fusaric, we searched for microorganisms that produced an unknown peak in the high-performance liquid chromatography (HPLC) profile that increased as the fusaric acid peak decreased. We previously reported that Aspergillus tubengensis converts fusaric acid to fusarinol, and this process is a detoxification mechanism, but it probably involves more than one gene.5 In recent experiments, HPLC analysis of culture filtrates from Mucor rouxii and Aspergillus niger grown in the presence of fusaric acid revealed the presence of multiple compounds including fusarinolic acid, 4 (9-hydroxyfusaric acid) with a retention time (tR) of 2.00 min; fusaric acid (tR 4.94 min); and an unknown compound with a tR of 2.32 min (Figure 2). We selected M. rouxii for further study since the unknown compound was produced in relatively high amounts and the HPLC trace was relatively clean compared to that for A. niger. Among cultures of M. rouxii, the peak area of this unknown compound varied from about half to equal that of fusaric acid; the areas for other unknown compound peaks were always

Figure 1. Fusaric acid and hydroxyfusaric acid analogues.

(Fov) in cotton (Gossypium hirsutum and Gossypium barbadense).1 Recently, a Fov genotype (CA-Fov4) was identified for the first time in the western hemisphere in California cotton fields,2 and it produces copious quantities of fusaric acid compared to most Fov genotypes indigenous to the United States.1 This pathotype causes severe yield losses in California and is a potential risk to cotton production in other U.S. cotton growing states. Cotton is unusually sensitive to fusaric acid compared to other plant species.3 Knockout mutants of an Australian isolate of Fov that had lost the ability to produce fusaric acid showed weaker pathogenicity than their wild-type progenitor toward tomato (Lycopersicon esculentum) seedlings in a seed germination bioassay on agar plates.1 Multiple analogues of fusaric acid have been shown to be less toxic to cotton cotyledons than fusaric acid itself.4 Thus, a © 2017 American Chemical Society

Received: Revised: Accepted: Published: 4989

April 10, 2017 May 15, 2017 May 24, 2017 May 24, 2017 DOI: 10.1021/acs.jafc.7b01655 J. Agric. Food Chem. 2017, 65, 4989−4992

Article

Journal of Agricultural and Food Chemistry

al.5 confirming the presence of the unknown compound at tR 2.32 min. The unknown compound was isolated from M. rouxii culture via extraction and HPLC separation as follows. The M. rouxii culture filtrates (divided into 10 50 mL tubes) were frozen at −80 °C and then were freeze-dried at −30 °C over 9 days yielding gummy, yellow residues. The residues were each dissolved in 1 mL of water with sonication and vigorous shaking to ensure complete dissolution of the residues. Aliquots (10 μL) of the solutions diluted with 90 μL water were analyzed via high-performance liquid chromatography (HPLC) confirming that the freeze-drying process did not alter the culture filtrate contents as compared to the filtrate before freeze-drying. Each solution was shaken with 10 mL of EtOAc and was centrifuged at 3000 rpm for 3 min. The EtOAc layers were decanted and pooled. The extraction was repeated three times, and then the aqueous layers and pooled organic samples were analyzed, confirming that 90% of the tR 2.32 min compound had been extracted from the aqueous layers. The EtOAc fractions also contained fusarinolic acid and fusaric acid. The combined organic sample was rotoevaporated under vacuum at 30 °C to remove all the EtOAc after which the residue was dissolved in 4 mL of water. The sample was acidified to pH 1.5−2.0 with one drop of 6 N HCl and then was extracted with 5 mL of EtOAc four times which removed 80% of the fusaric acid from the aqueous sample. The aqueous sample containing the tR 2.32 min compound was rotoevaporated as before to ensure that no EtOAc was present yielding a final volume of ca. 3 mL. The tR 2.32 min compound was purified from the 3 mL aqueous sample via HPLC. The HPLC instrument and method employed was that of Crutcher et al.5 except that the column was a 250 mm × 4.6 mm i.d., 5 μm, ProteCol-GP-C18 (SGE Analytical Science, Trajan Scientific Americas Inc., Austin, TX). With this column, the tR 2.32 min unknown eluted at tR 4.90 min and was almost fully resolved from its neighboring peaks, particularly fusarinolic acid at tR 4.08 min. A 425 μL aliquot of the aqueous sample was mixed with 75 μL of acetonitrile (MeCN). Multiple 25 μL injections of this mixture were made into the HPLC. The peak for the unknown compound was manually collected from the eluent by monitoring the real-time chromatogram and the UV/vis spectrum. Each day, collected fractions were stored at 2 °C. After the entire aqueous sample had been separated, the collected fractions containing the unknown were pooled and rotoevaporated to remove the MeCN. The resulting sample was frozen at −80 °C and was freeze-dried at −30 °C over 7 days yielding ca. 250 mg of a white crystalline material. HPLC established this material to be 98% pure tR 2.32 min compound. 8-Hydroxyfusaric Acid, 2. Mass spectrum: m/z (%) 185 (2), 136 (66), 119 (13), 118 (100), 92 (13), 91 (11), 90 (47), 65 (13), and 59 (23). The HRMS provided an accurate measured mass of m/z [M + H]+ calcd for C10H14NO3 196.0979; found 196.0969. HRMS/MS provided accurate measured masses of m/z [M + 1 − H2O]+ calcd for C10H12NO2, 178.0863; found 178.0864; [M + 1 − CO]+ calcd for C9H14NO2, 168.1019; found 168.1020; [M + 1 − CO, H2O]+ calcd for C9H14NO2, 150.0913; found 150.0915. 1H and 13C chemical shifts are shown in Table 1. UV maxima were observed at 223 and 272 nm (see Figure 2 for typical chromatogram and UV spectra for 1, 2, and 4). Bioassay of Fusaric Acid and Analogues. Acid delinted Coker312 cottonseed was germinated in moist paper rolls. Individual seedlings were selected for uniformity and were transplanted into pasteurized greenhouse planting mix in 450 mL plastic cups in a growth chamber with a 13 h, 28 °C day and 11 h, 22 °C night cycle as previously described.4 Solutions of fusaric acid, 8-hydroxyfusaric acid, and 10-hydroxyfusaric acid were prepared in 0.1% Tween 80 in water. Fusaric acid and 10-hydroxyfusaric acid were tested at 0.5, 1.0, 2.0, 4.0, and 8.0 mM while 8-hydroxyfusaric acid was tested only at 1 and 2 mM because of the limited quantity of material. Plants were treated with the test solutions on the morning of the seventh day after planting. The chemicals were introduced by placing three drops on the surface of a single cotyledonary leaf and by puncturing the leaf through each drop with a needle (26 gauge). Inoculation with 0.1% Tween 80 in water with no chemical was used as a control. Forty-eight hours after inoculation, the leaves on the plant were scored for necrosis by six individuals. The degree of

Figure 2. HPLC chromatogram of culture filtrate from M. rouxii grown in the presence of fusaric acid and analyzed using the method described in Crutcher et al.5 The inset shows the UV/vis spectrum for fusarinolic acid, 4 (dotted line); fusaric acid, 1 (dashed line); and the unknown, 2 (2.32 m, solid line).

individually no more than 25% of the fusaric acid peak. Because the UV/vis spectrum of the tR 2.32 min unknown was almost identical to that of fusarinolic acid and fusaric acid (Figure 2), it was isolated in sufficient quantities for NMR analysis. Herein, we report the chemical characterization of this compound and that of commercially available 10-hydroxyfusaric acid and the phytotoxicity of these compounds compared to that of fusaric acid using a cotyledonary leaf assay that has been used to ascertain the phytotoxicity of other fusaric acid analogues.4

■

MATERIALS AND METHODS

General. 1H NMR, 13C NMR, two-dimension NMR proton-proton homonuclear correlation spectroscopy (1H−1H COSY), two-dimension NMR heteronuclear multiple-bond correlation (HMBC), and inverse two-dimension NMR heteronuclear single quantum correlation (HSQC) spectra were acquired on an Avance III 500 instrument (Bruker, Billerica, MA) equipped with a cryoprobe operating at 500 MHz for 1H and 125 MHz for 13C; spectra were determined in D2O with a trace of CD3OD, which was used as an internal standard (1H: δ 3.35; 13C: δ 49.3). One- and two-dimensional 1H and 13C NMR spectra (1H−1H COSY, HMBC, and HSQC) were used to assign specific proton and carbon assignments. Direct exposure probe mass spectrometry (MS) was acquired on a Thermo-Electron DSQ instrument (West Palm Beach, FL) in positive ion electron ionization (EI mode) (70 EV, source 180 °C, scan rate 300 Da/s, scan 40−300 Da). High-resolution mass spectrometry (HRMS) and HRMS/MS were obtained on a ThermoScientific Orbitrap Elite (Bremen, Germany) in positive electrospray ionization (ESI) mode [needle voltage +4500 v, capillary temperature 275 °C, capillary volts (ground), sheath gas N2 (8 psi), scan m/z 50−700 Da, resolution 120 000, collisional induced dissociation energy 30; HRMS/MS scan 50−300 Da for ion m/z 196.10, resolution 120 000, isolation window 1 Da, collision energy 30]. 10-Hydroxyfusaric acid, 3, was purchased from Chem-Space (Riga, Latvia). Culturing M. rouxii. M. rouxii spores collected from potato dextrose agar (PDA) plates were cryopreserved in 30% glycerol at −80 °C; spores for the current experiment were transferred from the glycerol to PDA plates and were maintained on PDA at 27 °C until use. For fusaric acid detoxification, M. rouxii was grown on PDA, and 20 3 mm plugs from the actively growing edge of the colony were used to inoculate 100 mL cultures of potato dextrose broth containing 200 μg/mL fusaric acid. In total, five cultures were inoculated and allowed to shake at 135 rpm and 27 °C. After 4 days, the culture filtrate was filtered through Miracloth (Calbiochem, San Diego, CA) and was stored at −20 °C until further analysis. Isolation of the tR 2.32 min Unknown Compound. M. rouxii cultures were prepared as described earlier. Filtrates from five 100 mL M. rouxii cultures were subjected to HPLC as described by Crutcher et 4990

DOI: 10.1021/acs.jafc.7b01655 J. Agric. Food Chem. 2017, 65, 4989−4992

Article

Journal of Agricultural and Food Chemistry Table 1. 1H-NMR and 13C-NMR Chemical Shifts for 8Hydroxyfusaric Acid (2) C# 2 3 4 5 6 7a 7b 8 9a 9b 10 11

#H

δ (1H)

mult.

1 1

8.34 8.53

d d of d

1 1 1 1 1 1 3

8.64 3.17 2.95 3.89 1.65 1.55 1.01

bs m d of d d of d m m t

J (Hz) 8.12 8.12, 1.49

14.19, 3.76 14.19, 8.66

7.45

δ (13C)

mult.

145.7 127.1 148.6 142.5 143.1 40.3

s d d s d t

73.9 30.5

d t

10.5 165.6

q

These fragments would be expected if the hydroxyl group were attached at C-8. An HRMS ESI provided a [M + H]+ at m/z 196.0969 (calcd 196.0979) confirming a molecular formula of C10H13NO3 for the tr 2.32 peak. An HRMS/MS spectrum was also recorded showing loss of H2O (m/z 178.0864) and CO (m/z 168.1020) as well as H2O and CO (m/z 150.0915). To further elucidate its structure, the sample was submitted to NMR analysis. The 13C NMR spectrum of the metabolite was especially informative, indicating the presence of 10 carbon atoms. Five were sp2 carbon atoms (δ 127.1, 142.5, 143.1, 145.7, and 148.6) as expected for a pyridine ring; a peak at δ 165.6 supported the presence of a carboxylic acid group. Four aliphatic carbons were also in evidence with chemical shifts of δ 10.5, 30.5, 40.3, and 73.9; the chemical shift of the latter is characteristic of an aliphatic carbon to which a hydroxyl group is attached. The 1H NMR spectrum showed protons attached to three sp2 carbon atoms (δ 8.34, 8.53, and 8.64), a methyl group at δ 1.01 (t, 3H), two methylene groups with protons at δ 2.95 (1H) and 3.17 (1H) and at δ 1.55 (1H) and 1.65 (1H), and one aliphatic methine carbon (δ 3.89, 1H). In the 1H−1H COSY spectrum, the methine proton was coupled only to the protons on the two methylene groups, and the protons at δ 1.65 and 1.55 were the only protons coupled to the methyl. Thus, the peaks at δ 1.55 and 1.65 are on carbon 9, while those at δ 2.95 and 3.17 are on carbon 7. Furthermore, the peaks at δ 8.34 and 8.53 were strongly coupled, while the peaks at δ 8.53 and 8.64 were weakly coupled. This established that the protons on carbons 3, 4, and 6 had chemical shifts of δ 8.34, 8.53, and 8.64, respectively. In the HMBC spectrum, the proton on carbon 8 showed weak coupling to carbons 5, 9, and 10 and very weak coupling to carbon 7. The proton on carbon 3 showed coupling to the carboxyl carbon (C-11). All chemical shift assignments and other pertinent data are shown in Table 1. The 1H and 13C NMR chemical shift assignments and other pertinent data for the commercially acquired 10-hydroxyfusaric acid, which has not been published, are shown in Table 2.

necrosis was scored on a scale of 0 to 5 with 0 being no effect and 5 being severe necrosis and leaf curling around the inoculation site. The scores from the individual researchers at each compound concentration were averaged to provide the toxicity ratings. These mean ratings were then analyzed in an LSMeans Differences Tukey HSD comparison (n = 6, α < 0.05, Q = 3.5361) with concentration (mM) and compound as fixed factors. Mean differences between compounds within each mM concentration were compared by statistical contrasts. To provide a permanent record for later referral, the leaves were also photographed.

■

RESULTS AND DISCUSSION Previous evidence indicates that fusaric acid is a significant contributor to the pathogenicity of Fov.1,3 In addition to enhancing pathogenicity, fusaric acid assists Fov to compete and survive in soil.6 A biocontrol agent that degrades or detoxifies fusaric acid might provide an environmentally sound disease management option for cotton growers. Microorganisms isolated from a cotton field and several laboratory isolates were surveyed to identify isolates that could survive in cultures containing fusaric acid and concomitantly that could decrease the concentration of fusaric acid. Among those investigated was an M. rouxii isolate that decreased the concentration of fusaric acid while producing a new metabolite (HPLC tR 2.32 min, Figure 3) whose UV spectrum was nearly identical to that of fusaric acid (Figure 2). The mass spectrum of the purified tR 2.32 min compound provided a parent ion at m/z 195 (2%), characteristic of a hydroxylated fusaric acid metabolite. The position of the hydroxyl group was indicated by an ion at m/z 136 (66%, +M-59, +M-C3H7O) and m/z 59 (23%, +C3H7O).

Table 2. 1H-NMR and 13C-NMR Chemical Shifts for 10Hydroxyfusaric Acid (3) C# 2 3 4 5 6 7 8 9 10 11

#H

δ (1H)

mult.

1 1

8.26 8.45

d d

1 2 2 2 2

8.57 2.89 1.74 1.56 3.59

s t mult. mult. t

J (Hz) 8.1 8.1

7.7

6.50

δ (13C)

mult.

143.2 125.7 146.5 144.1 140.3 31.2 25.6 30.2 60.8 163.9

s d d s d t t t t s

Fusarinolic acid is less toxic than fusaric acid.4 Thus, the phytotoxicity of 8-hydroxyfusaric acid and that of 10hydroxyfusaric acid were compared to that of fusaric acid in a leaf bioassay as previously described.4 A control treatment with no chemical showed no necrosis; the relative toxicity results of 8-hydroxyfusaric acid and 10-hydroxyfusaric acid versus fusaric acid are given in Table 3. At concentrations of 1 and 2 mM, 8hydroxyfusaric acid and 10-hydroxyfusaric acid were significantly less phytotoxic than fusaric acid at these concentrations. In addition, 8-hydroxyfusaric acid was slightly less phytotoxic than 10-hydroxyfusaric acid. At 4 and 8 mM, 10-hydroxyfusaric

Figure 3. Concentration (mg/mL) of fusaric acid, 1 (solid line), and 8-hydroxyfusaric acid, 2 (dashed line), in M. rouxii culture filtrates over time. Data points are the means of three cultures. The standard errors (vertical bars) are insignificant except at 72 and 96 h. 4991

DOI: 10.1021/acs.jafc.7b01655 J. Agric. Food Chem. 2017, 65, 4989−4992

Journal of Agricultural and Food Chemistry

■

Table 3. Relative Toxicity Ratings [Mean Score 0−5a (±Standard Deviation)] of Fusaric Acid (1) and Analogues

ABBREVIATIONS USED Ca-Fov4, Fusarium oxysporum f. sp. vasinfectum race 4 (VCG0114); PDA, potato dextrose agar; 1H−1H COSY, twodimension NMR proton−proton homonuclear correlation spectroscopy experiment; HMBC, two-dimension NMR heteronuclear multiple-bond correlation experiment; HSQC, inverse two-dimension NMR heteronuclear single quantum correlation experiment

concentration (mM) 1.0b

2.0

4.0

8.0

fusaric acid (1)

1.8 (±0.1)A

2.9 (±0.1)A

8-hydroxyfusaric acid (2) 10-hydroxyfusaric acid (3)

0.5 (±0.1)B

0.9 (±0.1)B

0.9 (±0.1)C

1.3 (±0.1)C

3.6 (±0.1)A not tested 2.3 (±0.1)B

4.1 (±0.1)A not tested 3.1 (±0.1)B

compound

■

a

Visual rating: 0 = no necrosis; 5 = severe necrosis. bEntries in a column with different superscripts differ significantly at P < 0.05.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.7b01655. Photographs of inoculated leaves showing necrosis and leaf curling around the inoculation site (PDF) HRMS/MS with proposed structures (PDF)

■

REFERENCES

(1) Liu, J.; Bell, A. A.; Stipanovic, R. D.; Puckhaber, L. S.; Shim, W. B. In A polyketides synthase gene and an aspartate kinase like gene are required for the biosynthesis of fusaric acid in Fusarium oxysporum f. sp. vasinfectum, Proc. Beltwide Cotton Conf., Atlanta, GA, Jan. 4−11, 2011; Boyd, S., Huffman, M., Robertson, B., Eds.; National Cotton Council of America: Memphis, TN, 2011; pp 183−187. (2) Kim, Y. D.; Hutmacher, R. B.; Davis, R. M. Characterization of California isolates of Fusarium oxysporum f. sp. vasinfectum. Plant Dis. 2005, 89, 366−372. (3) Gaumann, E. Fusaric acid as a wilt toxin. Phytopathology 1957, 47, 342−357. (4) Stipanovic, R. D.; Puckhaber, L. S.; Bell, A. A.; Liu, J. 2011. Phytotoxicity of fusaric acid and analogues to cotton. Toxicon 2011, 57, 176−178. (5) Crutcher, F. K.; Liu, J.; Puckhaber, L. S.; Stipanovic, R. D.; Duke, S. E.; Bell, A. A.; Williams, H. J.; Nichols, R. L. Conversion of fusaric acid by Aspergillus tubingensis: a detoxification reaction. J. Chem. Ecol. 2014, 40, 84−89. (6) Ruiz, J. A.; Berma, E. M.; Jung, K. Production of siderophores increases resistance to fusaric acid in Pseudomonas protegens Pf-5. PLoS One [Online] 2015. http://dx.doi.org/10.1371/journal.pone. 0117040 (accessed Aug 12, 2015).10.1371/journal.pone.0117040

acid remained significantly less toxic than fusaric acid. The bioassay results indicate that conversion of fusaric acid to 8hydroxyfusaric acid by M. rouxii is a detoxification process. Since fusarinolic acid and 10-hydroxyfusaric acid also are less toxic than fusaric acid, the hydroxylation of the butyl side chain of fusaric acid appears to serve as a general detoxification process. On the basis of these observations, biocontrol agents that exhibit hydroxylation of the butyl side chain of fusaric acid may provide an alternative management tool for controlling CA-Fov4 and possibly other Fov pathotypes.

■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (979) 823-5670. Fax (979) 260-9319. ORCID

Robert D. Stipanovic: 0000-0003-4455-6767 Present Address §

Eastern Agricultural Research Center, Montana State University, Sidney, MT 59270, USA. Funding

We thank Cotton Incorporated, Cary, North Carolina, USA, for partial support of this research. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The laboratory isolate of Mucor rouxii was a kind gift from Cruz Torres, Texas A&M University, Department of Plant Pathology and Microbiology. We thank Ms. Shelby Dennis for helpful assistance including collection of 8-hydroxyfusaric acid from the culture media, and Dr. Howard Williams, Texas A&M University, Department of Chemistry, for providing NMR spectra. HRMS and HRMS/MS was provided by Dr. ChauWen Chou, Proteomics and Mass Spectrometry Core Facility, University of Georgia, Athens, GA. 4992

DOI: 10.1021/acs.jafc.7b01655 J. Agric. Food Chem. 2017, 65, 4989−4992