Fig 1.
Pathogenicity and transcriptome of FocTR4 morphotypes.
A Hyphal cells of FocTR4, expressing the plasma membrane marker Foc_GFP-Sso1 grown in liquid medium. Scale bar = 5 μm. B Chlamydospores, expressing the plasma membrane marker Foc_GFP-Sso1 (green) on the surface of a root (red) at 7 dpi. Plant cell walls are visualised by their auto-fluorescence. Germ tubes are indicated by arrowheads. Scale bar = 50 μm. C Lipid droplets (LD) in electron micrograph of a germinating chlamydospores and in a chlamydospores, stained with BODIPY 493/503 (green in inset; cell wall stained using calcofluor). Germination was triggered by incubation in PDB for 5 h. Scale bars = 2 μm and 5 μm (inset). D FocTR4 hyphae colonising banana root tissue at 26 dpi. Fungal cells express a fluorescent plasma membrane-located syntaxin (eGFP-Sso1, green), plant cell walls are visualised by their auto-fluorescence (red). Scale bar = 50 μm. E Morphology of purified FocTR4 hyphae, microconidia, macroconidia and chlamydospores. Scale bar = 20 μm. F Principal component analysis of RNA-preparations from 4 morphotypes of FocTR4. See also S1 Video. G Banana plants at 56 days after inoculation (dpi) with hyphae, microconidia, macroconidia or chlamydospores. All morphotypes cause leaf wilting, chlorosis and necrosis. H Necrosis of banana corms at 56 dpi with hyphae, microconidia, macroconidia or chlamydospores. Scale bar = 3 cm. I Quantitative assessment of necrosis in corm tissue after inoculation with the morphotypes. Bars in (I) represent mean ± SEM; sample sizes are indicated; statistical analysis in (I) used one-way ANOVA; n.s. = non-significant difference amongst data sets to control at two-tailed P = 0.2590.
Fig 2.
The effect of various fungicides on growth of FocTR4 and FocR1 on agar.
A FocTR4 colonies on PDA plates after 2 days growth at 25°C. Note that colonies contain of central region of densely grown cells and a "corona" of more diffuse spreading hyphae. Numbers of inoculated microconidia per colony are indicated. B, C Scanning electron micrographs of cells at the edge (B) and the centre (C) of a colony of FocTR4, grown on PDA agar plates. The central region shows yeast-like cells, which are occasionally multi-cellular (septa indicated by arrowheads), thus may resemble micro- or macroconidia; the peripheral region consists of hyphae. Scale bars = 10 μm (C) and 50 μm (B). D FocR1 colonies on PDA plates after 2 days growth at 25°C. Numbers of inoculated microconidia per colony are indicated. E, F Scanning electron micrographs of cells at the edge (E) and the centre (F) of a colony of FocR1, grown on PD agar plates. The central region shows yeast-like cells, which are occasionally multi-cellular (septa indicated by arrowheads), thus may resemble micro- or macroconidia; the peripheral region consists of hyphae. Scale bars = 10 μm (F) and 50 μm (E). G, H Morphology of FocTR4 (G) and FocR1 (H) persister cells, harvested from PDA plates after 2 days grown in the presence of 3 μg ml-1 epoxiconazole, 500 μg ml-1 fluxapyroxad or 500 μg ml-1 azoxystrobin. Colonies from where cells were taken are shown in the upper right. Scale bars = 10 μm. I Colony growth of FocTR4 persisters on fungicide-supplemented agar plates. Pre-treated: Cells were pre-grown for 2d in liquid cultures, supplemented with 3 μg ml-1 epoxiconazole, 500 μg ml-1 fluxapyroxad or 500 μg ml-1 azoxystrobin, harvested and plated on fungicide-containing agar plates; Untreated: Cells were pre-grown for 2d in liquid cultures, supplemented with 0.06–2.5% (v v-1) of the solvent methanol, harvested and plated on fungicide-containing agar plates. Note that colony formation on fungicide-supplemented agar plates by liquid-grown persister cells is almost identical to that of control cells. Non-linear regressions in was done using GraphPad Prism 6 (equation "[Inhibitor] vs. response—variable slope (four parameters)"). All data points are given as mean ± SEM, sample size is 4 colonies from 2 experiments; statistical testing used Student’s t-test, only significant differences are indicated (* represents two-tailed P-value <0.05; ** represents two-tailed P-value <0.01).
Fig 3.
Quantitative assessment of colony formation of FocR1 and FocTR4 on PDA plates, supplemented with fungicides.
Growth was monitored by measuring cell formation in images (see Methods) after 2d incubation at 25°C. 7 fungicide class were tested, each represented by 3 compounds; SDHIs = succinate dehydrogenase inhibitors; MALCs = Mono-alkyl cations. Note that a small population of "persister cells" of FocTR4 and FocR1 are highly tolerant in the major fungicide classes (azoles, SDHIs and strobilurins). See also S1 Fig for inhibition curves of Z. tritici cells and Table 1 for estimated concentrations at 50%, 90% and >99.5% growth inhibition (EC50, EC90, MIC). Non-linear regressions analysis used GraphPad Prism 9.
Table 1.
Inhibition of plate growth of FocTR4 and FocR1.
Fig 4.
Fungicide binding sites in target enzymes and fungicide-induced transcriptional changes in FocTR4 persisters.
A Azole binding site in Candida albicans Erg11 (Ca_ERG11, UniProt ID: P10613), Z. tritici Erg11 (Zt_Erg11, UniProt ID: F9XG32), FocTR4 Erg11_1 (Foc_Erg11_1; UniProt ID: X0JWG9), FocTR4 Erg11_2 (Foc_Erg11_2, UniProt ID: X0LR56) and FocTR4 Erg11_3 (Foc_Erg11_3, UniProt ID: X0JM02). Residues known in C. albicans Cyb51, reported to bind the azoles posaconazole and oteseconazole [41], are shown in red-bold; asterisk indicate potential involvement in resistance [35]; the impact of all residue substitutions in the Z. tritici and FocTR4 homologues of Ca_ERG11 was estimated using the SIFT server (https://sift.bii.a-star.edu.sg/); sequence alignment and determination of conserved amino acid exchanges used Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). B SDHI-binding site in succinate dehydrogenase subunits of Z. tritici (ZtSdh2, UniProt ID: O42772; ZtSdh4, UniProt ID: F9X9V6; ZtSdh3_1, UniProt ID: F9XH52; ZtSdh3_2, UniProt ID: F9XLX9 (partial); complete ZtSdh3_2 sequence at FungiDB, https://fungidb.org/fungidb/app/record/gene/ZTRI_10.476;) and FocTR4 (FocSdh2, UniProt ID: X0JUP8; FocSdh4, UniProt ID: X0JC07; FocSdh3_1, UniProt ID: X0JTM7; FocSdh3_2, UniProt ID: X0K132). Residues known in Z. tritici and other fungi to bind SDHI molecules [36] are shown in red-bold; asterisks indicate potential involvement in resistance [36]; the impact of all residue substitutions in FocTR4 homologues was estimated using the SIFT server (https://sift.bii.a-star.edu.sg/); sequence alignment used Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). C Comparison of the amino acid sequence of quinol oxidation (Qo) site of mitochondrial cytochrome b in Z. tritici (ZtCytb; UniProt ID: Q6X9S4) and FocTR4 (FocCytb, UniProt ID: A0A1A7TD23). Residues known to bind strobilurin molecules are shown in red-bold; asterisks indicate potential involvement in resistance; the impact of all residue substitutions in FocTR4 was estimated using the SIFT server (https://sift.bii.a-star.edu.sg/); sequence alignment used Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). Note that all substitutions in FocCytb are predicted to be of low impact ("tolerated"). Information on strobilurin molecule-interacting (Cpd-interacting) and resistance-conferring residues from [33,34]. D Venn diagram showing numbers of FocTR4 genes, up-regulated significantly (>5-fold) in the presence of fungicides. Note that a transcription factor (UniProt ID: X0JJ38) and an ABC transporter (UniProt ID: X0JJA9) are up-regulated in all conditions. E Number of up-regulated genes in FocTR4 persister cells, putatively involved in providing fungicide tolerance. Putative detoxifying enzymes (red), efflux transporter and pumps (green) and proteins involved in the targeted cellular pathway (blue) that are significantly upregulated (P<0.05) in epoxiconazole, fluxapyroxad and azoxystrobin-persisters are included. F Regulation of FocTR4 mitochondrial respiration complex subunits in the presence of 500 μg ml-1 azoxystrobin. Dotted green line indicates non-significant difference to methanol-treated control cells. Median of expression of nucleus-encoded subunits (red) is indicated by black lines. Data shown in (D) were derived from RNA-preparations of cells, grown for 2 days in medium, supplemented with 3 μg ml-1 epoxiconazole, 500 μg ml-1 azoxystrobin or 500 μg ml-1 fluxapyroxad.
Table 2.
Strongly induced genes, putatively involved in fungicide tolerance in FocTR4.
Fig 5.
The effect of azoxystrobin on Z. tritici IPO323 and FocTR4.
A Predicted effect of strobilurin on fungal respiration. Electrons are generated at complexes I and II, and by alternative NADH dehydrogenases. Coenzyme Q delivers them cytochrome b (Cytb) at complex III, from where they pass to cytochrome C (CytC) and arrive at complex IV (CIV) for neutralisation. This generates a proton gradient for ATP synthesis. Strobilurins block electron transfer Cytb (blue blunt arrow), which could form reactive oxygen species (ROS), known to trigger fungal apoptotic cell death [58]. Strobilurin-induced ROS was reported to be neutralised by fungal alternative oxidase (AOX, [55]), yet this enzyme is down-regulated in strobilurin persister cells. B Reactive oxygen species in Z. tritici IPO323 and FocTR4 cells, visualised with DHR123 (green). Both cell types were treated with 500 μg ml-1 azoxystrobin for 24 h. Scale bar = 10 μm. C ROS levels in azoxystrobin-treated Z. tritici and FocTR4 cells, visualised by DHR123 staining. D Apoptotic cells in Z. tritici IPO323, visualised with FITC-VAD(OMe)-FMK (green, open arrowhead). No apoptotic cells were found in FocTR4. Both cell types were treated with 500 μg ml-1 azoxystrobin for 24 h. Dead cells appear yellow due to co-staining with red-fluorescent propidium iodide (closed arrowhead). Note that these cells were not included in (E). Scale bar = 10 μm. E Apoptotic cells after treatment with azoxystrobin in Z. tritici and FocTR4, visualised by FITC-VAD(OMe)-FMK staining. F Mitochondrial membrane potential in azoxystrobin-treated Z. tritici and FocTR4 cells, treated with 500 μg ml-1 azoxystrobin and visualised by staining with the voltage-sensitive dye TMRM.FocTR4 persister cells maintained 66.1±4.0% (n = 80) of their mitochondrial polarization. IPO323 cells 35.5±3.1% (n = 50 cells). In (C—F) 500 μg ml-1 azoxystrobin was applied for 24 h. Control indicates the use of an equivalent volume of the solvent methanol. Non-normally distributed data in (C, F; Shapiro-Wilk test, P<0.05) shown as Whiskers’ plots with 25/75 percentiles (blue lines) and median (red lines); bars in (E) represent mean ± SEM, red dots represent independent experiments; sample sizes indicated; statistical analysis used non-parametric Mann-Whitney testing (C, F), or Student’s t-testing with Welch correction (E); n.s. = non-significant difference to control at two-tailed P = 0.2144 (C) and P = 0.4264 (E); **** = significant difference to control at two-tailed P<0.0001 (C, E, F).
Fig 6.
Fungicide sensitivity of FocTR4 spores in liquid culture.
A Examples of LIVE/DEAD dye-stained microconidia and chlamydospores after 10 days in ddH2O, supplemented with carbendazim, mancozeb or captan. Living cells exclude the dye and often accumulate it in the cell wall (open arrowheads); dead cells fluoresce red. Scale bars = 10 μm. B Relative number of living microconidia (LIVE/DEAD staining negative) after 10 d-treatment with fungicides; fungicide classes are indicated above parenthesis. C Relative number of living macroconidia (LIVE/DEAD staining negative) after 10 d-treatment with fungicides; fungicide classes are indicated above parenthesis. D Relative number of living chlamydospores (LIVE/DEAD staining negative) after 10 d-treatment with fungicides; fungicide classes are indicated above parenthesis. E Morphology and LIVE/DEAD staining of microconidia, macroconidia and chlamydospores after 10 day-incubation in SDW, supplemented with epoxiconazole, fluxapyroxad and azoxystrobin (all 100 μg ml-1). Note that all spore forms can persist at high concentration of fungicides. Scale bar = 10 μm. Bars shown in (B—D) represent the mean proportion (±SEM) of cells that did not take up LIVE/DEAD staining, thus were considered alive; red dots represent averages of 3 independent experiments. All spore types in (B—D) were incubated for 10 days in SDW at 25°C and under rotation; concentrations used were in category I (= inhibition on agar plates at <10 μg ml-1): thiabendazole, 10 μg ml-1; carbendazim, 4.5 μg ml-1; chlorothalonil, 5 μg ml-1; captan, 10 μg ml-1; mancozeb, 35 μg ml-1; in category II (= inhibition on agar plates at >10 μg ml-1): copper, 100 μg ml-1 (applied as Copper(II) sulfate pentahydrate); LMW chitosan, 100 μg ml-1 (applied as a lactate salt); garlic oil, 100 μg ml-1; CTAB (Cetrimonium bromide), 100 μg ml-1; dodine, 100 μg ml-1; C18DMS; 100 μg ml-1; triticonazole, 100 μg ml-1; epoxiconazole, 100 μg ml-1: tebuconazole, 100 μg ml-1; azoxystrobin, 100 μg ml-1; trifloxystrobin, 100 μg ml-1; pyraclostrobin, 100 μg ml-1; bixafen, 100 μg ml-1; fluxapyroxad, 100 μg ml-1; boscalid, 100 μg ml-1; thiophanate methyl, 100 μg ml-1; concentrations in (e) are 100 μg ml-1.
Fig 7.
Fungicides and protection of bananas against Panama disease.
A Whole-plant symptoms of Panama disease at 56 days after root inoculation with chlamydospores, followed by 2 applications of fungicides or the solvents 0.14% (v v-1) DMSO or 0.16% (v v-1) methanol in water (Control (+)). The negative control (Control (-)) was only treated with the solvent. B Corm necrosis in bananas at 56 days after root inoculation with chlamydospores, followed by 2 treatments with fungicides. Control (+) indicates inoculation with spores, followed by treatment with 0.16% (v v-1) methanol (positive control). Scale bar = 2 cm. C Quantitative assessment of darkening of corm tissue after inoculation with chlamydospores followed by 2 treatments with fungicides. Banana corm necrosis was analysed 56 days after the first treatment. Light blue: Controls; dark blue: fungicide treatments. Bars in (C) show mean ± SEM from 18–24 measurements of 9–12 plants from 3–4 experiments; statistical comparison in (C) used Student’s t-testing with Welch correction; n.s. = non-significant difference to respective control at two-tailed error probability of P = 0.7074 (Mancozeb) and P = 0.1871 (Copper); * = significant difference to control at two-tailed at P = 0.0452 (Chitosan); ** = significant difference to control at two-tailed at P = 0.006 (Captan); **** = significant difference to control at two-tailed P<0.0001. Plant infection assays used LMW chitosan, 200 μg ml-1 (applied as 333 μg ml-1 lactate salt); copper, 200 μg ml-1 (applied as 786 μg ml-1 copper (II) sulfate pentahydrate); mancozeb, 70 μg ml-1; captan, 20 μg ml-1; CTAB, 200 μg ml-1; dodine, 200 μg ml-1; C18DMS, 200 μg ml-1.
Fig 8.
Protective effect of C18DMS and dodine in FocTR4-infected bananas.
A Cavendish banana plants, infected with FocTR4 chlamydospores, at 56 days after 2 treatments (day 0 and day 7) with C18DMS, dodine or solvent control (0.16% MetOH (v v-1)). B Bar chart showing plant health, assessed by measuring all aerial tissues in photographic images of 56 d-old plants, infected with chlamydospores and treated twice with C18DMS or dodine. Control indicates healthy plants (100% plant health) that were not infected by FocTR4, but which were treated with the solvent methanol. Data in (B) are given as mean ± SEM; sample sizes indicated in graph; statistical comparison with control and between data sets used Student’s t-testing with Welch correction; * = significant difference at two-tailed P = 0.0480; *** = significant difference at two-tailed P = 0.0007; **** = significant difference at two-tailed P<0.0001. Treatments were solvent control: 0.16% MetOH (v v-1), dodine: 200 μg ml-1 and C18DMS: 200 μg ml-1.
Fig 9.
The MoA of C18DMS in germinating chlamydospores of FocTR4.
A Developmental stages of chlamydospore germination in potato dextrose medium. Dormant spores appear smooth (stage I); prior to germination they appear granular (stage II). Single germ tubes emerge, which form septa (arrowhead), extend at their growing tip and develop into hyphae (stage IV). Scale bar = 10. B Mitochondrial membrane potential, visualised by the cell-penetrating dye TMRM, in chlamydospores at different stages of germination. Note that almost no TMRM staining is seen in un-germinated spores (stage I), but mitochondria-associated fluorescence appear shortly before the germ tube emerges (stage II; arrowhead). Scale bar = 10 μm. C Mitochondrial membrane potential, indicated by TMRM fluorescence, in chlamydospores at stage (I) and (II) and in the germ tube (IV). D Intensity profile of TMRM fluorescence, indicative of mitochondrial membrane potential, along germ tubes of 7 h-old germlings of FocTR4, treated with the solvent (red curve), C18DMS (blue curve) and dodine (green curve). Values are shown as mean ± SEM of 14–15 germlings. Note that both MALCs depolarises mitochondria. E Abundance of ROS, labelled with DHR123 in control, C18DMS- and dodine-treated FocTR4 cells. C18DMS induces ROS generation, whereas dodine reduces ROS. F Induction of cellular apoptosis, visualised by FITC-VAD(OMe)-FMK staining, in control, C18DMS- and dodine-treated FocTR4 cells. C18DMS alone induces apoptotic cell death. This activity was reported in Z. tritici [58]. Data in (C, E) are not normally distributed (Shapiro-Wilk test, P<0.05) and are shown as Whiskers’ plots with 25/75 percentiles (blue line) and median (red line); data in (D, F) are given as mean ± SEM; red dots in (F) represent independent experiments; sample sizes indicated in all graphs; statistical comparison with control in (F) used Student’s t-testing with Welch correction and in (C, E) non-parametric Mann-Whitney testing; n.s. = non-significant difference at two-tailed P = 0.0932 (C), P = 0.0993 (E) and two-tailed P = 0.1016 (F); **** = significant difference at two-tailed P<0.0001 (C, E, F). Treatments were 10 μg ml-1, 30 min, for C18DMS and dodine (D, E) and 10 μg ml-1 of both compounds for 24 h (F). Control in (D, E, F) indicates the use of an equivalent volume of the solvent methanol.