Investigating Biologically Active Compounds from the Filamentous Fungi Bjerkandera adusta and Irpex lacteus That Inhibit the Growth of Marchantia polymorpha

Abstract

Background

Marchantia polymorpha is a model organism for research on evolutionary molecular-microbe interactions. Filamentous fungi pathogenic to M. polymorpha have been reported and a necrotrophic-like infection process has been proposed according to experiments conducted in the laboratory, but many aspects of M. polymorpha and microbes interaction remain unknown.

Objective

Field observations have been done and indicated that several kinds of fungi had ability to infect M. polymorpha and among of them, Bjerkandera adusta and Irpex lacteus infect M. polymorpha, and laboratory experiments have revealed that these fungi have a strong ability to infect M. polymorpha. Therefore, it was hypothesized that these fungi may produce specialized metabolites that lead to effectively invade into the liverwort. The aim of this research was to determine the chemical structures of the specialized metabolites and evaluate the biological activities of the compounds.

Materials and Methods

B. adusta and I. lacteus were obtained from GenBank Project, NARO. Male accessions of M. polymorpha, Takaragaike-1, were used for bioassay. The fungi were grown in potato dextrose broth medium. The culture filtrates were filtered and concentrated, and EtOAc (3 L) and H₂O (1 L) were added to the crude material. Crude extracts obtained from the EtOAc were subjected to several kinds of purification steps to afford methyl vanillin from B. adusta and tremutin E from I. lacteus.

Results

Methyl vanillin and tremutin E were isolated from B. adusta and I. lacteus, respectively, according to a guidance of growth inhibitory activity-guided fractionations and identified using mass spectroscopic data together with 1D and 2D NMR techniques.

Conclusion

Methyl vanillin and tremutin E inhibited the growth of M. polymorpha with almost 100-fold stronger activity than that of acetic acid, a pivotal active ingredient of commercial anti-liverwort agent.

Keywords

methyl vanillin tremutin E

Introduction

The thalloid liverwort Marchantia polymorpha is widely distributed in temperate regions and has been increasingly utilized as a model bryophyte since it is easy to grow in the laboratory, has a short life cycle, possesses the ability to transform, and is compatible with several gene editing methods.^1–2 Similar to other bryophytes, gametophyte generation dominates the life cycle of M. polymorpha and the main plant body (thallus) is produced for line maintenance; furthermore, the costs of maintaining M. polymorpha cultures are lower than those for flowering plants. Moreover, M. polymorpha is a representative of an ancient lineage of land plants that colonized our planet more than 450 million years ago.^3–4

M. polymorpha is reported to interact with different types of microorganisms such as viruses, bacteria, fungi, and oomycetes.⁵ In this study, the hemibiotrophic oomycete pathogen Phytophthora palmivora colonized M. polymorpha, enabling the study of liverwort–pathogen interactions at the molecular level.⁶ Furthermore, artificial inoculation with Pseudomonas syringae pv. tomato DC3000 leads to activation of the typical hallmarks of plant innate immunity.⁷ Regarding the system establishing defense system in M. polymorpha, Monte et al reported that dinor-iso-12-oxo-phytodienoic acid (dinor-iso-OPDA) is a pivotal ingredient, which interacts with MpCOI1.⁸ Additionally, the isolation of pathogenic fungi samples from wild M. polymorpha was performed by Matsui et al,⁹ and four fungal strains were isolated: Bjerkandera adusta MI4 and MI3, Irpex lacteus MI1, and Phanerochaete peniophoroides MI2. Among these strains, I. lacteus demonstrated the strongest pathogenic effect on M. polymorpha, followed by B. adusta.

B. adusta, a wood-rotting white-rot basidiomycete, belongs to the Polyporaceae family and parasitizes particular wood types.^10–11 This fungus secretes lignin peroxidase, manganese peroxidase, and other versatile peroxidases to degrade lignin in wood.^12–14 In addition, the pathogenic wood-decaying fungus I. lacteus also belongs to the family Polyporaceae and has been widely used in Chinese folk medicine for decades to treat inflammation, bacterial/fungal infections, and urinary retention.^15–17 A chemical investigation of this fungus revealed the presence of eight sesquiterpenoids, tremutins A-H, and three other known compounds.¹⁸ Previously, Wang et al¹⁸ investigated the toxicities of all of these compounds to five human cancer cell lines: breast cancer (MCF-7), hepatocellular carcinoma (SMMC-7721), human myeloid leukemia (HL-60), colon cancer (SW480), and lung cancer (A-549).

Numerous studies have reported that plant pathogenic bacteria and fungi can invade higher plants and produce specialized compounds such as fusicoccin,¹⁹ alternaric acid,²⁰ phleichrome,²¹ victorin,²² tabtoxin,²³ and syringomycin.²⁴ These compounds are useful because they can facilitate investigations of plant-microbe interactions. Previous studies have supported possibility that identifying biologically active compounds, including those produced by infectious fungi that inhibit growth of the liverwort, may potentially help to elucidate the interactions between M. polymorpha and infectious microbes. Therefore, this study aimed to isolate and purify specialized compounds from B. adusta and I. lacteus that may facilitate invasion of M. polymorpha.

Materials and Methods

General Experimental Procedures

The optical rotations were determined using a JASCO P-2200 polarimeter. Nuclear magnetic resonance (NMR) spectra were recorded in CDCl₃ using a JNM-ECZ 400 NMR spectrometer (¹H NMR: 400 MHz, JEOL Ltd) and an AMX 500 Bruker system (¹H NMR: 500 MHz, ¹³C NMR: 126 MHz). Assignments of proton and carbon signals were performed by obtaining ¹H NMR, ¹³C NMR (referenced for CDCl₃ at δ_H 7.24 and δ_C 77.23, respectively), COSY, HSQC, HMBC, and NOESY spectra. The high-resolution fast atom bombardment mass spectrometry (HRFDMS) analysis was performed using a JMS-T100GCV instrument (JEOL, Ltd). High-performance liquid chromatography (HPLC) was performed on a HITACHI CHROMASTER HPLC system equipped with a 5410 UV detector, 5310 column oven, 5110 pump, and YMC−Pack FA column (250 × 6.0 mm). Fractions were monitored using thin−layer chromatography on silica gel 60 F254 plates, and the spots were visualized by spraying the plates with a detection reagent (anisaldehyde solution prepared according to the general method)　with heating. All chemicals used in this study were of analytical grade and were purchased from Sigma‒Aldrich (Tokyo, Japan) and Kanto Chemical Co., Inc. (Tokyo, Japan).

Fungal and Plant Material

The filamentous fungi B. adusta (MAFF 420156) and I. lacteus (MAFF 420244) were obtained from GenBank Project, NARO (National Agriculture and Food Research Organization), Japan. Male accessions of M. polymorpha, Takaragaike-1, were propagated in half-strength Gamborg's B5 medium (pH 5.5).

Extraction and Isolation of Methyl Vanillin (1)

The fungus B. adusta was grown in potato dextrose broth medium (48 L) for 50 days in the dark at 25 °C. The culture was filtered and concentrated using a rotary evaporator. Ethyl acetate (EtOAc, 3 L) and H₂O (1 L) were added to the crude material, and the mixture was allowed to stand for 48 h. The separated EtOAc layer was dried over anhydrous sodium sulfate (Na₂SO₄). The volatile components of the separated EtOAc layer were removed under reduced pressure to obtain a crude extract, which was subjected to column chromatography (CC) using silica gel and eluted using a solvent system composed of methanol (MeOH)/CHCl₃ (5:95) to obtain four fractions (A-D). The fraction showing inhibitory activity (24.2 mg) was further separated using the same solvent (CC with MeOH/CHCl₃ [5:95]) to yield compound 1 (8.8 mg) as a colorless oil.

Methyl vanillin (1): colorless oil; ¹H NMR (400 MHz, CDCl₃) and ¹³C NMR (400 MHz, CDCl₃): see Table 1; HRFDMS: m/z 166.06320 [M]⁺ (calculated for C₉H₁₀O₃, 166.06299).

Table 1.

¹H NMR and ¹³C NMR Spectroscopic Data of Methyl vanillin (1) in CDCl₃ (δ in ppm, J in Hz).

Position	Type	δ _C	δ _H
1	CH	191.0	9.84, s
2	C	154.8
3	C	148.5
4	C	130.5
5	CH	126.9	7.46, dd (8.0, 2.0)
6	CH	110.4	6.96, d (8.4)
7	CH	109.0	7.40, d (2.0)
8	CH₃	56.2	3.98, s
9	CH₃	56.1	3.93, s

Extraction and Isolation of Tremutin E (2)

The fungus I. lacteus was grown in potato dextrose broth (48 L) for 50 days in the dark at 25 °C. The culture was filtered and concentrated using a rotary evaporator. EtOAc (3 L) and H₂O (1 L) were added to the crude extract, and the mixture was left to stand for 48 h. The separated EtOAc layer was dried over Na₂SO₄. After removing volatiles under reduced pressure, the residue was subjected to silica gel CC and eluted with a solvent system composed of MeOH/CHCl₃ (5:95) to yield 40 fractions. The fraction showing inhibitory activity (275.6 mg) was further purified by HPLC (YMC-Pack FA, f 6.0 × 250 mm, MeOH:H₂O:acetic acid (AcOH) = 58:42:0.1, 1 mL/min, A_{210 nm}) to give compound 2 (4.2 mg).

Tremutin E (2): Colorless oil; $[α]_{D}^{27}$ + 69.146 (c 1.0, MeOH); ¹H NMR (500 MHz, CDCl₃) and ¹³C NMR (125 MHz, CDCl₃); see Table 2; HRF-DMS: m/z 264.13492 [M]⁺ (calculated for C₁₅H₂₀O₄, 264.13616)

Table 2.

¹H NMR and ¹³C NMR Spectroscopic Data of Tremutin E (2) in CDCl₃ (δ in ppm, J in Hz).

Position	Type	δ _C	δ _H
1	C	67.5
2	C	156.1
3	C	129.3
4	CH	66.6	4.57, br d
5	CH₂	35.9	1.81, m
6	CH	29.1	1.88, m
7	CH	47.7	2.59, q (8.2, 5.1, 8.3)
8	CH₂	43.8	1.75, q (4.7, 9.1)1.65, dd (9.0, 4.7, 9.1)
9	C	38.2
10	CH	77.9	3.25, s
11	CH₂	70.1	4.66, dd (17.9, 1.9)4.36, d (17.8)
12	C	174.5
13	CH₃	19.8	0.93, d (6.8)
14	CH₃	26.9	1.15, s
15	CH₃	24.7	1.08, s

Diffusion Assay

M. polymorpha grown on half-strength Gamborg's B5 medium for 14 days was used to analyze the inhibitory effects of the crude extracts from B. adusta and I. lacteus (Figure 1). The crude extracts were diluted with ethanol (EtOH, 1000 µL), and these solutions were applied to a paper disk for duplicate antibiotic assays (ϕ: 8 mm, 1.5 mm thickness; ADVANTEC, Taipei City, Taiwan). Each paper disk received 500 µL of solution, and residual solvent was removed by air-drying on a stainless-steel needle point holder for 15 min. Each paper disk was then placed in the middle of a medium containing four 14-day-old M. polymorpha individuals. Paper disks treated with EtOH (1000 µL) served as mock controls. Each treatment was replicated thrice. The growth area was measured using the ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Figure 1.

Schematic Illustration of the Diffusion Assay.

Biological Activity Test of Methyl Vanillin (1) and Tremutin E (2) Using a Dilution Assay

To evaluate the biological activities of methyl vanillin (1) and tremutin E (2), AcOH (3) was used as a positive control. Methyl vanillin (1, 0.32 mg), tremutin E (2, 0.53 mg), and AcOH (3, 1.4 mg) were dissolved in MeOH (0.2 mL). Each solution was added to half-strength Gamborg's B5 medium (20 mL), which was then poured into a Petri dish (ϕ: 8.5 cm, h: 2.0 cm). Seven-day-old thalli were transplanted into the media and maintained at 23 °C under light conditions for up to 10 days. Subsequently, the area of growth was visually examined and measured using ImageJ software. The medium treated with MeOH served as a mock control. Five replicate samples were analyzed at each concentration. This assay was referred to as the dilution assay (Figure 2) in this study.

Figure 2.

Schematic Illustration of the Dilution Assay.

Statistical Analysis

The data were subjected to analysis using Tukey's test. Statistical significance was defined as P value less than 05. Values were presented as mean ± standard deviation (SD).

Result

Isolation and Structural Determination of Growth Inhibitory Compounds

First, we established a bioassay to evaluate M. polymorpha growth inhibition. A schematic illustration of the bioassay used to isolate low-molecular-weight bioactive compounds is presented in Figure 1, whose method is named as diffusion assay in this report. Procedural details of the assay are provided in the Experimental Section. Biological-activity-guided fractionation was performed using the diffusion assay to obtain compounds 1 and 2 from B. adusta and I. lacteus, respectively.

Compound 1 (8.8 mg) was a colorless oil isolated from 48 L of the B. adusta culture filtrate. Its molecular formula was deduced to be C₉H₁₀O₃ from its molecular ion ([M]⁺) at m/z 166.06320 detected using high-resolution field desorption mass spectrometry (HRFDMS) analysis, indicating five degrees of unsaturation. According to the ¹H NMR data (Table 1), two methoxy groups were identified by their characteristic shifts at δ_H 3.98 (3H, s, H-8) and δH 3.93 (3H, s, H-9). Three aromatic protons were observed at δ_H 7.46 (1H, dd, J = 8.0, 2.0 Hz, H−5), 6.96 (1H, d, J = 8.4 Hz, H−6), and 7.40 (1H, d, J = 2.0 Hz, H−7). The resonance at δ_H 9.84 was attributed to a proton from an aldehyde moiety. Additionally, nine resonances appeared in the ¹³C NMR spectrum of 1 (Table 1), corresponding to two CH₃ groups, four CH groups, and three quaternary carbons, as indicated by heteronuclear single quantum correlation (HSQC) analysis. Finally, the chemical structure of 1 was determined and identified as methyl vanillin (Figure 3) after the individual parts were connected via ¹H−¹H correlation spectroscopy (COSY) and heteronuclear multiple bond correlation (HMBC) analyses. Methyl vanillin, a derivative of vanillin, is a volatile component of natural vanilla. The instrumental analysis data obtained in this study for 1 were consistent with those of a commercially available standard sample.

Figure 3.

Chemical Structures of Methyl Vanillin (1) and Tremutin E (2).

Compound 2 (16.2 mg) was also a colorless oil isolated from 40 L of the I. lacteus culture filtrate. HRFDMS analysis revealed that its molecular formula was C₁₅H₂₀O₄ according to its [M]⁺ ion at m/z 264.13492, suggesting six degrees of unsaturation. The ¹H NMR data of 2 (Table 2) revealed three methyl signals at δ_H 0.94 (3H, d, J = 6.8 Hz, H-15), 1.08 (3H, s, H−14), and 1.16 (3H, s, H−13). In addition, the signals at δ_H 4.36 (1H, d, 17.8 Hz, H−11), 4.57 (1H, br. d, H−10), and 4.66 (1H, dd, J = 17.9, 1.9 Hz, H−4) were attributed to the protons on oxygenated carbons. The ¹³C NMR spectrum of 2 (Table 2) revealed 15 carbon resonances attributed to three CH₃ groups, three CH₂ groups, four CH groups, and five quaternary carbons; these data were supported by HSQC analysis. The chemical structure of compound 2 was determined to be tremutin E (Figure 3) after the individual parts were connected via. ¹H−¹H COSY and HMBC analyses. The spectral data obtained were similar to those obtained in previous studies ¹⁸ including the optical value [α]D.

Growth Inhibitory Activity of Compounds 1 and 2 on M. polymorpha

Since compounds 1 and 2 were isolated via biological activity-guided fractionation using the diffusion assay (Figure 1), their biological activities were examined in detail using a dilution assay. The details of the method are described in the Experimental section and presented in Figure 2. Compounds 1 and 2 (final concentration of 10 μM for both) were independently added to 1/2 Gamborg media (20 mL), onto which 7-day-old M. polymorpha were transplanted to observe the growth of the liverwort. AcOH (3) was used as the positive control because it is commercially used as an anti-liverwort agent. Compounds 1 and 2 exhibited almost the same biological activity as compound 3 at 100-fold lower concentrations (Figure 4).

Figure 4.

Growth Inhibitory Test Using the Isolated Compounds. Each Value is Represented as the Mean Value ± SD. The Number of the Independent Replicates were Five for Mock and Ten for the Treatmets of AcOH, Methyl Vanillin, and Tremutin E. Different Letters Indicate Statistically Significant Differences (P < .05, Tukey's Test).

Discussion

It was reported that several kinds of fungi had ability to infect M. polymorpha and among of them, B. adusta and I. lacteus showed ability to infect M. polymorpha.⁹ In this study, methyl vanillin (1) and tremutin E (2) were isolated from culture filtrate of B. adusta and I. lacteus, respectively (Figure 3), as growth inhibitor of the liverwort (Figure 4). During this research, a simpler bioactivity assessment method (Figure 1) was used for the isolation of active compounds (1 and 2). The method enabled us to assess a large number of crude samples at once. To assess the growth inhibitory activity of the isolated compounds (1 and 2) on M. polymorpha in detail, a dilution assay (Figure 2) was used. This method has made it possible to set the exact concentration of the sample. It was revealed that methyl vanillin (1) and tremutin E (2) exhibited almost the same biological activity as AcOH at 100-fold lower concentrations (Figure 4). AcOH was one of pivotal components composing commercial anti-liverwort agents. Since numerous studies have reported that plant pathogenic bacteria and fungi can invade higher plants and produce specialized compounds giving impacts upon the plants, it was assumed that methyl vanillin (1) and tremutin E (2) might have the activity to scramble the metabolic and other regulatory mechanisms of M. polymorpha. However, the underlying mechanisms by which the isolated compounds impact M. polymorpha remain unclear, and the details of how these pathogens invade M. polymorpha is a subject for future research.

Conclusion

This project was initiated by the isolation and purification of specialized compounds produced by infectious fungi in M. polymorpha, such as those that inhibit liverwort growth. In this study, two bioactive compounds, methyl vanillin (1) and tremutin E (2), were isolated from two species of filamentous fungi that infect liverwort. Both compounds inhibited the growth of M. polymorpha with almost 100-fold stronger activity than that of acetic acid, a pivotal active ingredient of commercial anti-liverwort agents.

Footnotes

Acknowledgements

We acknowledge the assistance of Dr Eri Fukushi and Mr Yusuke Takata (GC‒MS & NMR Laboratory, Research Faculty of Agriculture, Hokkaido University) for obtaining the spectroscopic data, and Dr Hirofumi Nakagami (Plant Proteomics Research Unit, RIKEN CSRS, Yokohama, Kanagawa, 230-0045 Japan) for providing information on fungal strains. We also acknowledge a scholarship for S.I. from Hitachi Corporative Education and Research Support (Hitachi Ltd). The authors also thank the Program for Forming Japan's Peak Research Universities (J-PEAKS) of the Japan Society for the Promotion of Science (JSPS) for financial support.

ORCID iD

Hideyuki Matsuura

Ethical Considerations

Ethical Approval is not applicable for this article.

Statement of Informed Consent

There are no human subjects in this article and informed consent is not applicable.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Statement of Human and Animal Rights

The experiments described in this article were conducted in compliance with laws that protect Human and Animal Rights.

References

Poveda

. Marchantia polymorpha as a model plant in the evolutionary study of plant-microorganism interactions. Curr Plant Biol. 2020:23. doi: 10.1016/j.cpb.2020.100152

Ishizaki

Nishihama

Yamato

, et al. Molecular genetic tools and techniques for Marchantia polymorpha research. Plant and Cell Physiol. 2016;57(2):262‐270. doi: 10.1093/pcp/pcv097

Shimamura

. Taxonomy, phylogeny and morphology of a model system. Plant and Cell Physiol. 2016;57(2):230‐256. doi: 10.1093/pcp/pcv192

Berger

Bowman

Kohchi

. Marchantia. Curr Biol. 2016;26(5):R186‐RR87. doi: 10.1016/j.cub.2015.12.013

Poveda

. Analysis of Marchantia polymorpha-microorganism interactions: basis for understanding plant-microbe and plant-pathogen interactions. Front Plant Sci. 2024;15:1301816. doi: 10.3389/fpls.2024.1301816

Carella

Gogleva

Tomaselli

, et al. Establishes tissue-specific intracellular infection structures in the earliest divergent land plant lineage. P Natl Acad Sci USA. 2018;115(16):E3846‐E3E55. doi: 10.1073/pnas.1717900115

Gimenez-Ibanez

Zamarreño

García-Mina

, et al. An evolutionarily ancient immune system governs the interactions between Pseudomonas syringae and an early-diverging land plant lineage. Curr Biol. 2019;29(14):2270‐2281. doi: 10.1016/j.cub.2019.05.079

Monte

Ishida

Zamarreno

, et al. Ligand-receptor co-evolution shaped the jasmonate pathway in land plants. Nat Chem Biol. 2018;14(5):480‐488. doi: 10.1038/s41589-018-0033-4

Matsui

Iwakawa

Hyon

, et al. Isolation of natural fungal pathogens from Marchantia polymorpha reveals antagonism between salicylic acid and jasmonate during liverwort–fungus interactions. Plant and Cell Physiol. 2020;61(2):442‐442. doi: 10.1093/pcp/pcz214

10.

Heinfling

Martinez

, et al. Purification and characterization of peroxidases from the dye-decolorizing fungus. Fems Microbiol Lett. 1998;165(1):43‐50. doi: 10.1016/S0378-1097(98)00255-9

11.

Sugawara

Sugano

. Genome sequence of Bjerkandera adusta strain Dec 1, a basidiomycete secreting DyP-Type peroxidase. Microbiol Resour Ann. 2023;12(1). doi: 10.1128/mra.01049-22

12.

Gold

Kuwahara

Chiu

, et al. Purification and characterization of an extracellular H2O2-requiring diarylpropane oxygenase from the white rot basidiomycete, phanerochaete chrysosporium. Arch of Biochem Biophys. 1984;234(2):353‐362. doi: 10.1016/0003-9861(84)90280-7

13.

Glenn

Gold

. Purification and characterization of an extracellular Mn(II)-dependent peroxidase from the lignin-degrading basidiomycete, phanerochaete-chrysosporium. Arch of Biochem Biophys. 1985;242(2):329‐341. doi: 10.1016/0003-9861(85)90217-6

14.

Camarero

Sarkar

Ruiz-Dueñas

, et al. Description of a versatile peroxidase involved in the natural degradation of lignin that has both manganese peroxidase and lignin peroxidase substrate interaction sites. J Biol Chem. 1999;274(15):10324‐10330. doi: 10.1074/jbc.274.15.10324

15.

Novotny

Cajthaml

Svobodová

, et al. Irpex lacteus, a white-rot fungus with biotechnological potential - review. Folia Microbiol. 2009;54(5):375‐390. doi: 10.1007/s12223-009-0053-2

16.

Feng

, et al. Isolation, colonization, and chlorpyrifos degradation mediation of the endophytic bacterium Sphingomonas strain HJY in Chinese chives (Allium tuberosum). J Agric Food Chem. 2017;65(6):1131‐1138. doi: 10.1021/acs.jafc.6b05283

17.

Tang

Zhao

, et al. Irpexoates A-D, four triterpenoids with malonyl modifications from the fruiting bodies of the medicinal fungus. Nat Product Bioprosp. 2018;8(3):171‐176. doi: 10.1007/s13659-018-0160-3

18.

Wang

Hui-Xiang

, et al. Sesquiterpenoids from cultures of the basidiomycetes. J Nat Prod. 2020;83(5):1524‐1531. doi: 10.1021/acs.jnatprod.9b01177

19.

de Boer

. The fusicoccin story revisited. J Exp Bot. 2024;75(18):5531‐5546. doi: 10.1093/jxb/erae300

20.

Hassan

Okuta

Kato

, et al. Alternaric acid stimulates phosphorylation of his-tagged RiCDPK2, a calcium-dependent protein kinase in potato plants. Genet Mol Res. 2012;11(3):2381‐2389. doi: 10.4238/2012.May.10.1

21.

Seto

Kogami

Shimanuki

, et al. Production of phleichrome by as stimulated by diketopiperadines of. Biosci Biotechnol Biochem. 2005;69(8):1515‐1519. doi: 10.1271/bbb.69.1515

22.

Tada

Ohura

Betsuyaku

, et al. Victorin triggers programmed cell death and the defense response via interaction with a cell surface mediator. Plant Cell Physiol. 2005;46(11):1787‐1798. doi: 10.1093/pcp/pci193

23.

Manning

Danson

Calderone

. Functional chararacterization of the enzymes TabB and TabD involved in tabtoxin biosynthesis by. Biochem Biophys Res Commun. 2018;496(1):212‐217. doi: 10.1016/j.bbrc.2018.01.028

24.

Mott

Takemoto

. Syringomycin, a bacterial phytotoxin, closes stomata. Plant Physiol. 1989;90(4):1435‐1439. doi: 10.1104/pp.90.4.1435