The Frq-Frh Complex Light-Dependently Delays Sfl1-Induced Microsclerotia Formation in Verticillium dahliae

Abstract

The vascular plant pathogenic fungus Verticillium dahliae has to adapt to environmental changes outside and inside its host. V. dahliae harbors homologs of Neurospora crassa clock genes. The molecular functions and interactions of Frequency (Frq) and Frq-interacting RNA helicase (Frh) in controlling conidia or microsclerotia development were investigated in V. dahliae JR2. Fungal mutant strains carrying clock gene deletions, an FRH point mutation, or GFP gene fusions were analyzed on transcript, protein, and phenotypic levels as well as in pathogenicity assays on tomato plants. Our results support that the Frq-Frh complex is formed and that it promotes conidiation, but also that it suppresses and therefore delays V. dahliae microsclerotia formation in response to light. We investigated a possible link between the negative element Frq and positive regulator Suppressor of flocculation 1 (Sfl1) in microsclerotia formation to elucidate the regulatory molecular mechanism. Both Frq and Sfl1 are mainly present during the onset of microsclerotia formation with decreasing protein levels during further development. Induction of microsclerotia formation requires Sfl1 and can be delayed at early time points in the light through the Frq-Frh complex. Gaining further molecular knowledge on V. dahliae development will improve control of fungal growth and Verticillium wilt disease.

Keywords: Verticillium dahliae; clock gene homologs; development; microsclerotia formation; plant pathogen.

Figure 1

V. dahliae Frequency (Frq) and Frq-interacting RNA helicase (Frh) repress microsclerotia formation and induce aerial hyphae production. (a) Schematic depiction of Frq and Frh proteins present in respective strains. (b–f) The ex planta phenotype was analyzed of V. dahliae wild-type (WT), a FRQ deletion strain (∆FRQ), strains expressing FRQ–GFP under native- (FRQ–GFP) or gpdA-promoter control (FRQ–GFP OE), strains with a FRH point mutation in wild-type (FRH^R806H) or mutant strain background (FRQ–GFP/FRH^R806H, FRQ–GFP OE/FRH^R806H), as well as respective complementation strains (FRQ-C, FRH-C). 50,000 spores were point-inoculated onto (b) potato dextrose medium (PDM), (c) simulated xylem medium (SXM), and (d–f) Czapek-Dox medium (CDM) agar and incubated at 25 °C in the light for ten days. Deletion of FRQ and point mutation of FRH resulted in darker, stronger melanized colonies, suggesting enhanced microsclerotia formation compared with wild-type. (e) Representative pictures of the colony center after removal of aerial hyphae, cross-sections, and microscopic images of microsclerotia of CDM cultures are depicted. Black scale bar: 500 µm; blue scale bar: 50 µm. (f) Colony melanization was quantified on CDM after ten days. The experiment was performed three times with two independent transformants of each mutant strain and one to three independent WT cultures (N = 6). Depicted is the mean of six biological replicates with standard deviation. WT melanization was set as one. Statistical differences were calculated using t-tests (n.s.: not significant, **: p < 0.01, ***: p < 0.001). Differences compared with WT are indicated on top of the bars and non-significant differences between FRQ deletion and FRH point mutation strains are shown by the connecting lines. Deletion of FRQ and FRH point mutation in all different backgrounds led to significantly increased colony melanization.

Figure 2

V. dahliae White collar 1 (Wc1) is required for repression of microsclerotia formation and induction of aerial hyphae production. A total of 50,000 spores of the V. dahliae wild-type (WT), WC1 deletion (∆WC1), complementation (WC1-C), and WC1–GFP fusion protein producing (WC1–GFP) strains, as well as the FRQ deletion (∆FRQ) and FRH point mutation (FRH^R806H) strains, were point-inoculated onto (a) PDM, (b) SXM, and (c–e) CDM. The phenotype was investigated after incubation in the light for ten days. Similar to the FRQ deletion and FRH^R806H point mutation, WC1 deletion resulted in increased colony melanization. (d) Depicted are representative pictures of the colony center after removal of aerial hyphae, cross-sections through the colony center, and microscopic images of microsclerotia of CDM-grown colonies. Black scale bar: 500 µm, blue scale bar: 50 µm. (e) The melanization of the colony center was quantified relative to WT (set as one) after incubation on CDM in the light for ten days. The experiment was performed three times with two independent mutant strain transformants or independent WT cultures (N = 6). Only one WC1–GFP-expressing transformant was analyzed (N = 3). Depicted is the mean of biological replicates with standard deviation. Statistical differences were calculated using t-tests (n.s.: not significant, ***: p < 0.001). Differences compared with WT are indicated above the bars and the result from statistical comparison between WC1 deletion and FRQ deletion or FRH point mutation strains are shown by the connecting lines. Similar to FRQ deletion and FRH^R806H strain colonies, the ∆WC1 strain colony was stronger melanized than the WT colony. FRQ, FRH, and WC1 are similarly involved in repression of microsclerotia formation.

Figure 3

V. dahliae Frq and wild-type Frh positively affect spore production. A total of 200,000 spores of the V. dahliae wild-type (WT), FRQ deletion (∆FRQ), Frq–GFP fusion protein producing (FRQ–GFP; FRQ–GFP OE), FRH^R806H point mutation (FRH^R806H, FRQ–GFP/FRH^R806H, FRQ–GFP OE/FRH^R806H), WC1 deletion (∆WC1), WC1–GFP-expressing (WC1–GFP), and respective complementation strains (FRQ-C, FRH-C, WC1-C) were inoculated into 50 mL liquid SXM and incubated under agitation at 25 °C in the light. The number of produced spores was quantified after five days. The experiment was performed three times with two transformants of the mutant strains and independent wild-type cultures (N = 6), except for WC1–GFP, of which only one transformant was used (N = 3). Bars represent the mean of biological replicates with standard deviation. WT spore production was set as one. Statistical differences were calculated using t-tests (n.s.: not significant, *: p < 0.05, **: p < 0.01, ***: p < 0.001). Results of statistical analyses compared with the WT are displayed above the bars, and other comparisons are indicated with the connecting lines. (a) Significantly fewer conidia were produced by FRQ deletion and FRH point mutation strains compared with WT. (b) WC1 deletion also resulted in significantly reduced conidiation compared with WT. However, the positive effect on conidiation mediated by WC1 was significantly smaller than that mediated by FRQ and FRH.

Figure 4

Clock gene homologs, FRQ, Frh amino acid residue arginine 806, and WC1 are dispensable for V. dahliae-mediated symptom induction in tomato plants. Ten-day-old tomato plant seedlings were wounded and infected with spores of V. dahliae wild-type (WT), (a) FRQ deletion (∆FRQ), FRH^R806H point mutation (FRH^R806H), or their respective complementation strains (FRQ-C, FRH-C), as well as (b) WC1 deletion (∆WC1) or complementation (WC1-C) strains. Two independent transformants of mutant strains served as biological replicates. Control plants were inoculated with water (mock). After incubation under long-day-conditions for 21 days, plants were classified as being healthy (mean disease level = 1–1.99; green), or showing mild (mean disease level = 2–2.99; yellow), strong (mean disease level = 3–3.99; orange), or very strong (mean disease level = 4; red) symptoms based on the height of the plant, the length of the longest leaf, and the fresh weight. Depicted are results from (a) three or (b) two independent experiments with 11 to 15 plants per biological replicate. The number of total plants (N) is indicated above the bars. Statistical significances compared with WT infection were calculated with Mann–Whitney U tests (n.s: not significant, ***: p < 0.001). Pictures of plant trays, representative individual plants, and hypocotyl cross sections are depicted below the diagram. V. dahliae FRQ, Frh amino acid residue arginine 806, and WC1 were dispensable for wild-type-like symptom development in tomato plants.

Figure 5

V. dahliae Frh amino acid residue arginine 806 is largely dispensable for FRQ, WC1, and WHITE COLLAR 2 (WC2) transcript and Frq protein levels. As many as 1 × 10⁶ spores of the V. dahliae wild-type (WT), FRQ deletion strain (∆FRQ) as well as FRQ–GFP-expressing strains with either wild-type FRH (FRQ–GFP) or point mutated FRH (FRQ–GFP/FRH^R806H) were grown in (a,c) liquid SXM or (b,d) on SXM agar covered with a nylon membrane. Cultures were incubated at 25 °C in the light for two (2d), four (4d), and six days (6d) before (a,b) RNAs or (c,d) proteins were extracted. (a,b) Transcript levels of FRQ, WC1, and WC2 were investigated via quantitative reverse transcription PCR for indicated strains and time points. Depicted is the mean of three biological replicates with standard deviation (N = 3). Reference genes H2A and EIF2B were used for normalization and the two-day WT transcript levels were set as one. Significant differences to WT samples of respective time points are labeled above the bar. Non-significant differences are indicated by connecting lines (calculated with t-tests, n.s.: not significant, *: p < 0.05). Differences in FRQ transcript levels between WT and the ∆FRQ strain could not be statistically analyzed as no FRQ transcript was detected in the latter strain. FRQ transcript levels were not significantly changed upon Frh^R806H amino acid exchange. The WC2 transcript level was only significantly decreased upon FRH mutation during cultivation on SXM agar for two days. (c,d) The protein amount of Frq–GFP was quantified using a GFP antibody in western experiments. Images of representative replicates are depicted below the graphs. The pixel density of signals was normalized to the Ponceau S staining. The protein signal intensity of the FRQ–GFP strain after two days of growth was set as one. The mean of three biological replicates with standard deviation is depicted (N = 3). Statistical significances were determined with t-tests (n.s.: not significant, *: p < 0.05, **: p < 0.01). Differences to FRQ–GFP after two days are indicated above the bars. As shown by the connecting lines, no significant difference was observed between four-day or six-day Frq–GFP protein levels. The amino acid exchange in the Frh protein did not significantly affect the Frq–GFP protein levels.

Figure 6

Nuclear localization of Frq–GFP and Frh–GFP is unaffected by Frh amino acid substitution. For the analysis of subcellular localization of Frq–GFP, Frh–GFP, or Frh^R806H–GFP fusion proteins, spores of the indicated strains were inoculated into PDM and incubated overnight in the light. (a) Fluorescence signals of V. dahliae WT with RFP-labeled histones (WT/RFP–H2B), as well as a strain expressing FRQ–GFP under control of the native promoter either without (FRQ–GFP) or with expression of the RFP–H2B fusion construct (FRQ–GFP/RFP–H2B), were compared. Frq–GFP and RFP–H2b co-localized in the nuclei. Scale bar: 10 µm. (b,c) The GFP signal localization of strains producing (b) Frq–GFP with wild-type FRH (FRQ–GFP) or point-mutated FRH (FRQ–GFP/FRH^R806H) or (c) Frh–GFP (FRH–GFP) or Frh^R806H–GFP fusion proteins (FRH^R806H–GFP) was analyzed. V. dahliae wild-type ectopically overexpressing GFP (WT/GFP OE) served as control. Localization of Frq–GFP and Frh–GFP was unaffected by Frh^R806H amino acid substitution. Scale bar: 10 µm.

Figure 7

V. dahliae Frh arginine at position 806 is dispensable for Frh–GFP protein levels but is required for Frh-Frq interaction. (a) A total of 50,000 spores of V. dahliae wild-type (WT), an FRH point mutation strain (FRH^R806H), and strains expressing either FRH–GFP or FRH^R806H–GFP (FRH–GFP, FRH^R806H–GFP) under control of the native promoter were point-inoculated onto PDM, SXM, and CDM. The phenotype was investigated after incubation at 25 °C in the light for ten days. The strain producing the Frh–GFP fusion protein grew similar to the WT. (b,c) Spores of V. dahliae strains expressing FRH–GFP and FRH^R806H–GFP were inoculated (b) into liquid SXM (1 × 10⁶ spores) or (c) spread on solid SXM covered with a nylon membrane (4 × 10⁶ spores). Cultures were grown at 25 °C for two (2d), four (4d), and six days (6d). Proteins were extracted and the protein amount of Frh–GFP and Frh^R806H–GFP (≥152 kDa) was quantified via western experiments using a GFP antibody. The pixel density of detected signals was quantified and normalized to the respective Ponceau S staining. Images of one replicate are depicted below the graphs. The fusion protein amount of the FRH–GFP strain after two days of cultivation was set as one. Depicted is the mean of three biological replicates with standard deviation (N = 3). Statistical significances were determined with t-tests (n.s.: not significant, *: p < 0.05, **: p < 0.01). Significances compared with Frh–GFP 2d fusion protein amount are indicated on top of the bars. Comparisons between the 4d or 6d time points are shown by connecting lines. The amino acid exchange does not significantly affect the fusion protein levels. (d,e) As many as 5 × 10⁹ spores of V. dahliae WT, V. dahliae WT with ectopically overexpressed GFP, FRH–GFP, and FRH^R806H–GFP strains were inoculated into 500 mL SXM and grown at 25 °C for two days. One to three cultures were combined per sample (N = 1). As control, mycelium of wild-type strains with and without GFP overexpression were mixed approx. 1/60 (with GFP/without GFP). GFP trap pull-down was performed with total protein extracts. Peptides obtained through trypsin digestion were analyzed by LC/MS. The experiment was performed in triplicates (N = 3). During subsequent analysis, missing values were replaced by imputation four times. Interaction candidates found to be significant in all four imputation repetitions are colored in the volcano plots—Frh (bait): blue; Frq: orange; other significant interaction partners: brown. GFP is colored green. Proteins in the upper right part are significant interactors. (d) Frh interacted with six proteins, including Frq. (e) Frh^R806H did not significantly interact with any other protein.

Figure 8

High abundance of the Suppressor of flocculation 1 (Sfl1)–GFP fusion protein in the FRQ deletion mutant does not lead to enhanced microsclerotia formation. GFP–Sfl1 protein localization and microsclerotia production of a FRQ deletion strain ectopically overexpressing GFP–SFL1 (∆FRQ (NAT^R)/GFP–SFL1 OE) were analyzed. The FRQ deletion strain with nourseothricin resistance cassette (∆FRQ (NAT^R)) was used as background strain. (a) Spores were inoculated into PDM and grown at 25 °C overnight. Subcellular localization of GFP–Sfl1 was analyzed by fluorescence microscopy. Wild-type ectopically overexpressing GFP (WT/GFP OE) served as control. GFP–Sfl1 is predominantly localized in nuclei. Scale bars = 10 µm. (b,c) A total of 50,000 spores of the V. dahliae wild-type (WT) and ∆FRQ (NAT^R), as well as ∆FRQ (NAT^R)/GFP–SFL1 OE strains, were point-inoculated onto minimal medium (CDM). Plates were incubated at 25 °C in the light (left) or in the dark (right) for ten days. (b) Colonies and cross sections of the colony center are depicted. Scale bar = 500 µm. (c) Melanization of the colonies was quantified. The experiment was performed three times with one to two independent transformants or cultures as biological replicates (N = 4–6). Depicted is the mean of biological replicates with standard deviation. WT melanization was set as one. Significant differences were calculated with t-tests (n.s.: not significant, *: p < 0.05). Significant differences compared with WT are depicted above the bars. Comparisons between other strains are indicated by connecting lines. Light-incubated cultures of ∆FRQ (NAT^R) and ∆FRQ (NAT^R)/GFP–SFL1 OE strains were significantly more melanized than the WT. When grown in the dark, FRQ deletion did not affect colony melanization, but GFP–SFL1 overexpression led to significantly reduced colony melanization compared with WT.

Figure 9

SFL1 acts epistatically on FRQ in the regulatory pathway of microsclerotia formation. Microsclerotia production of V. dahliae wild-type (WT), FRQ and SFL1 single-deletion strains (∆FRQ, ∆SFL1), ectopic complementation strains (FRQ-eC, SFL1-eC), and double-deletion strain (∆FRQ/∆SFL1) was analyzed. (a) A total of 50,000 spores of the strains were point-inoculated onto SXM or CDM and incubated at 25 °C in the light (left panel) or in the dark (right panel) for ten days. Top-view pictures of the colonies are shown. Of CDM cultures, the colony center without aerial hyphae, cross sections of colony centers, and microscopic views are depicted. The FRQ and SFL1 double-deletion strain grew similar to the SFL1 single-deletion strain. Black scale bars = 1 mm, blue scale bars = 100 µm. (b) A total of 50,000 spores were point-inoculated onto CDM and incubated at 25 °C for 14 days. Melanization of the colonies was quantified and WT was set to one. Two independent transformants (N = 2) of the deletion strains were tested, and one transformant (N = 1) was used for each of the complementation strains and wild-type. Mean values with standard deviation are depicted (N = 3–6). Significant differences compared with WT were calculated with t-tests and are indicated above the bars (n.s.: not significant, *: p < 0.05, ***: p < 0.001). The SFL1 single-deletion and FRQ/SFL1 double-deletion strains were significantly less melanized compared with wild-type.

Figure 10

Sfl1 is, to a lesser extent than Frq, involved in induction of V. dahliae conidiospore production, and the combined actions of Frq and Sfl1 slightly enhance disease development. The spore production and disease induction of V. dahliae wild-type (WT), FRQ and SFL1 single-deletion strains (∆FRQ, ∆SFL1), ectopic complementation strains (FRQ-eC, SFL1-eC), and the FRQ and SFL1 double-deletion strain (∆FRQ/∆SFL1) were analyzed. (a) A total of 200,000 spores of the respective strains were inoculated into SXM and incubated at 25 °C for five days under agitation. Conidiospore production was quantified. One WT culture, one transformant of each complementation strain, and two independent transformants of the single- and double-deletion strains were tested. WT conidia concentration was set to one. The experiment was performed three times with all strains (complementation strains: N = 3) and a fourth time without the complementation strains (WT: N = 4, deletion strains: N = 8). Bars represent the mean relative concentration with standard deviation. Significant differences compared with WT were calculated with t-tests and are indicated above the bars (n.s.: not significant, **: p < 0.01, ***: p < 0.001). Conidiation was significantly reduced in the SFL1 deletion strain and deletion of FRQ in presence or absence of SFL1 resulted in a greater decrease in conidiation compared with WT. (b) Tomato plants were treated with spores of indicated strains and grown for 21 days under long-day-conditions. Two independent deletion or complementation transformants served as biological replicates. Water-inoculated plants (mock) served as control. Plant height, length of the longest leaf, and fresh weight were measured and compared with respective mock plant values to calculate the mean disease level of each plant. Thereby, plants were classified as being healthy (mean disease level = 1–1.99, green), or showing mild (mean disease level = 2–2.99, yellow), strong (mean disease level = 3–3.99, orange), or very strong (mean disease level = 4, red) symptoms. Depicted are the results of two (∆FRQ, FRQ-eC, ∆FRQ/∆SFL1) to four independent experiments (mock, WT, ∆SFL1, SFL1-eC) with 13 to 15 plants per biological replicate. The total number of plants (N) is indicated above the bars. Statistical significances compared with WT infection were calculated with Mann–Whitney U tests (n.s: not significant, *: p < 0.05). Representative pictures of plant trays, individual plants, and hypocotyl cross sections are depicted below the diagram. Only plants treated with the FRQ/SFL1 double-deletion strains were slightly, but significantly less diseased.

Figure 11

Schematic representation of Frq-, Frh-, Sfl1-, and Wc1-mediated control of V. dahliae development. The scheme summarizes the interplay of the investigated proteins and their functions in controlling the development of V. dahliae. Inducing functions are indicated by green arrows and a plus (+), whereas repressive functions are indicated by red arrows and a minus (−). Smaller effects are represented by dashed lines. The Frq–Frh complex is involved in enhanced conidiation. This developmental process is, to a lesser extent, also enhanced by Sfl1 and Wc1. Frq, Frh, Wc1, or Sfl1, individually, did not affect the symptom development in tomato plants, but the combined effects of Sfl1 and Frq slightly enhanced symptom induction in planta. This is potentially due to Sfl1- and Frq-mediated increased conidiation and thus enhanced colonization of the host plant. The control of microsclerotia formation was more complex. Sfl1 is required for enhanced microsclerotia formation regardless of the light conditions. In the light, the Frq–Frh complex mainly represses the Sfl1-dependently induced microsclerotia formation, but also the Sfl1-independently enhanced development of microsclerotia. Wc1 enhances transcription of FRQ and presumably thereby also represses microsclerotia formation in the light. In darkness, however, Wc1 enhances microsclerotia formation.