Introduction

Esca is a grapevine trunk disease (Bertsch et al. 2013), a vascular fungal infection of the grapevine, as the possibly most relevant member of this group. Several fungal species are associated with this syndrome, while Phaeomoniella chlamydospora (Pch) and Phaeoacremonium minimum ascomycetous species, as well as members of the basidiomycetous genus Fomitiporia are considered to be the main causal agents (Beris et al. 2023). Pch and P. minimum are believed to be “pioneer” pathogens in Esca, initiating the disease development (Larignon and Dubos 1997). The pathogens enter the grapevine through damaged wood (e.g., pruning wounds) and colonize the xylem of the host. By the action of secreted cell wall degrading enzymes (Fleurat-Lessard et al. 2023), as well as the non-enzymic production of reactive hydroxyl radicals (Perez-Gonzalez et al. 2022), the fungi damage the infected woody tissues resulting in black necrosis. This necrosis becomes a white rot when Fomitiporia species (as secondary colonizers) reach a sufficient degree of colonization, resulting in the so-called Esca proper syndrome (Beris et al. 2023). Besides the local symptoms in the wood, Esca is also expressed in the annual parts of the infected plants. Leaves show characteristic tiger stripe-shaped discoloration and necrosis, while necrotic spots appear on the berries (Bertsch et al. 2013). Occasionally, Esca can cause the sudden death of the host called apoplexy (Beris et al. 2023).

The background of symptom development on the annual parts of the host is unclear. Several secreted phytotoxins are proposed to take part in the process, like scytalone and isosclerone pentaketide metabolites (Bruno and Sparapano 2006), as well as extracellular proteins (Luini et al. 2010), and even secreted polysaccharides (e.g., pullulan, Bruno and Sparapano 2006). According to another hypothesis, the obstruction of the flow of xylem sap in the infected wood results in a nutrient deficiency in the foliage and is responsible for foliar symptoms (Bortolami et al. 2023). Besides the direct damage and occlusion of the vascular tissue caused by the pathogens, the plant responses to their action may also contribute to the dysfunction of the xylem. The host tries to prevent the spreading of the pathogens by occluding the vessel with the formation of tyloses and gels, resulting in decreased flow of xylem sap (Bortolami et al. 2023).

The pathogenesis of Esca disease—or grapevine trunk diseases in general—is still unexplained. Some characteristics of Esca make its development hard to understand. One such trait is the erratic expression of external symptoms (Maher et al. 2012), as well as the very long latency of the disease, which can take several years (Beris et al. 2023). Another confusing fact that the expected correlation between the abundance of the pathogens in the host and symptom expression often can not be demonstrated in naturally infected plants (Bruez et al. 2014; Mondello et al. 2018), suggesting that the pathogens are members of the healthy endophytic microbiome of grapevine.

Because of the above controversies, external factors and rare events are considered crucial promoters for Esca symptom development. Some researchers are focusing on external triggering factors like rainfall and temperature (Calzarano et al. 2018), or pruning practices (Lecomte et al. 2018). Others emphasize the role of microbial interactions established between the pathogens (Sparapano et al. 2001), or even with the contribution of non-pathogenic bacteria (Haidar et al. 2021) or fungi (Karácsony et al. 2023). Another theory explains the characteristics of Esca development by hypothesizing that critical pathogen biomass triggers symptom expression (Hrycan et al. 2020). This can be explained by either the increment of pathogenicity factors with increasing pathogen biomass or by the shift of the fungi to pathogenic behavior at a given cell density.

Cell density is perceived in microorganisms by the so-called quorum sensing (QS) mechanism, which is widely studied in bacteria (Whiteley et al. 2017) and to a lesser extent in fungi (Tian et al. 2021). The process is based on the sensing of secreted quorum sensing molecules (QSMs) by the producing microbes. QSMs are responsible for the expression of some pathogenic traits including the yeast-to-hypha transition in dimorphic fungi like the phytopathogenic Ophiostoma ulmi (Berrocal et al. 2012) and the human pathogenic Candida albicans (Han et al. 2011). Yeast forms are believed to aid the spreading of the pathogen, while mycelial growth promotes their local action. QS also regulates biofilm formation in C. albicans (Rodrigues and Černáková 2020) as the most extensively studied fungal example for this process. Biofilm growth proposes several advantages for this pathogen by decreasing the effectiveness of antimicrobial compounds (Silva et al. 2017), and as a multicellular structure, allowing the specialization of cells to different functions like persistence (Wuyts et al. 2018), or dispersion (Uppuluri et al. 2018). In the case of microorganisms with diffuse growth in a liquid environment (yeasts and bacteria), their aggregated growth on an interface defines biofilm. This multicellular structure is embedded in an extracellular matrix (ECM) consisting of polysaccharides, proteins, nucleic acids, and lipids (Karygianni et al. 2020). Biofilm formation is often accompanied by changes in cellular morphology, like the abovementioned yeast-to-hypha transition in C. albicans (Baillie and Douglas 1999).

Finding criteria for biofilm in the case of filamentous fungi is more complicated since they grow as a mycelial network and tend to attach to solid surfaces. In their case, changes in cellular morphology and ECM production may define biofilm (Harding et al. 2009). Biofilm formation is poorly studied in phytopathogenic fungi and demonstrated only in a limited number of species (reviewed by Motaung et al. 2020), like Fusarium graminearum (Shay et al. 2022), Fusarium verticilloides (Miguel et al. 2015), Fusarium oxysporum (Li et al. 2014), Verticillium dahliae, and Botrytis cinerea (Harding et al. 2010), as well as Dydimella bryoniae and Sclerotiana sclerotiorum (Harding et al. 2019). However, most of the above studies are based on the results of in vitro experiments and the role of biofilm formation in the pathogenesis of fungal plant diseases needs to be explained.

The present study is the first report on the possible role of QS and biofilm formation in a grapevine trunk disease, the Esca syndrome. The dimorphic growth of the pioneer pathogen Pch was demonstrated with the observation of yeast-like cells in mature colonies on solid media. An ECM with a proposed high polysaccharide content was also detected in association with the yeast-like cells. The fungus was found to produce acetate which promoted the yeast-like cellular morphology, as well as the ECM production, acting as a possible QSM in the process. The dimorphic growth of the fungus was also demonstrated on autoclaved grapevine wood, suggesting that Pch biofilms may be formed in the infected host.

Materials and methods

Fungal isolation and identification

Strains belonging to Pch were isolated from grapevine wood as described previously (Karácsony et al. 2023). The trunks of grapevines showing the foliar symptoms of Esca disease were sampled, and thin disks were cut from the infected wood. The disks were surface sterilized with 70% (v/v) ethanol for 2 min, in 4% (m/v) sodium hypochlorite for 2 min, and finally in 70% (v/v) ethanol for 2 min. The disks were dried on a sterilized paper towel and cut into small pieces. Five wood pieces were placed on the surface of PDA (potato dextrose agar) plates supplemented with 10 μg/ml oxytetracycline to prevent bacterial growth. The plates were incubated at room temperature (21 ± 2 °C) for 4 weeks and small portions of emerging mycelia were sequentially subcultured on PDA plates three times to obtain monoclonal isolates. The isolates were identified according to morphological characteristics as well as by sequencing the ITS (internal transcribed spacer) gene region amplified by PCR using the ITS1-F (Gardes and Bruns 1993) and ITS4 (White et al. 1990) primers. The obtained amplicons were sequenced (BaseClear B.V. Netherlands) and the sequences were compared in online databases using BLAST (www.ncbi.nlm.nih.gov). Fungal strains were cryopreserved as mycelial/conidial suspensions in 50% (v/v) glycerol at − 80 °C.

Fungal growth conditions

Strains of Pch were grown in a nutrient-rich medium with a composition of 5% (m/v) sucrose and 1% (m/v) yeast extract (YS5), solidified by 2% (m/v) bacteriological grade agar. The YS5 medium was amended with congo red indicator at 80 mg/ml concentration (YS5C80 medium) or 8 mg/ml concentration (YS5C8 medium) for the detection of extracellular polymeric substances indicating biofilm growth. To study the effects of acetate on Pch colony development YS5C8 medium was supplemented with sodium acetate in the 6.25–100 mM range. Fungi were also cultivated on autoclaved grapevine wood chips (~ 5 × 5 × 2 mm, cut from 1-year-old canes of Cabernet Sauvignon) placed on 2% (m/v) water-agar. All the media and the wood chips were inoculated with 7 μl of conidial suspensions (5 × 106 cells/ml) and incubated at 25 °C in the dark for 2–12 days before examinations.

Microscopic examinations

For the examination of fungal colonies growing on solid media, small sections were directly hand-cut from the cultures using a scalpel, to obtain 3–5-mm-wide longitudinal or cross-section with 1–2-mm thickness. Fungal cells were examined directly after placing on a glass slide, or subjected to fluorescent staining by immersing in 100 μg/ml acridine orange in distilled water for 5 min, followed by 10 min destaining in distilled water. Thereafter, samples were placed on a glass slide and covered with a coverslip before examination. In some cases, the longitudinal sections were placed in a microcentrifuge tube with 1.5 M NaCl and disrupted by a dissecting needle and vigorous vortexing. About 10 μl of the obtained cell suspensions were pipetted on a glass slide, covered with a coverslip, and visualized immediately or after 16-h incubation at room temperature (21 ± 2 °C). Grapevine wood chips with inoculated fungi were stained by congo red (8 mg/ml, 10 min), destained in distilled water for 10 min, and placed on a glass slide before microscopic examination. In some cases, fungal biomass was scraped by a scalpel from the wood chips into a drop of distilled water, covered with a coverslip, and observed thereafter.

All examinations were done using an Olympus BX53F2 (Olympus Corporation, Tokyo, Japan) microscope equipped with differential interference contrast (DIC) and phase contrast (PC) optical accessories. Fluorophores were excited by an LED light source (λ = 360–665 nm). The following filter sets were used for the different fluorophores: congo red (λex = 540–550 nm, λem = 575–625 nm, red emitted color); acridine orange (λex = 470–495 nm, λem = 510–550 nm, green color); wood chips autofluorescence (λex = 360–370 nm, λem = 420–460 nm, blue color). Photomicrographs were recorded by the use of a DP47 camera controlled by CellSens Entry software (Olympus Corporation, Tokyo, Japan).

Measurement of fungal metabolites

Fungal strains were cultivated on YS5 solid medium as described above. Agar plugs (d = 5 mm) with the fungal biomass were cut by a cork borer from the center of colonies and the intact disks were extracted two times with 0.5 ml ethyl acetate at room temperature for 4 h, with 20-rpm shaking. Fractions were pooled and used for analysis. All inoculations and extractions were done in triplicates. Metabolites were detected by combined gas chromatography-mass spectrometry (GC-MS) analysis using Clarus 600 GC, Clarus 600 T MS device (Perkin Elmer, Waltham, USA) equipped with a built-in autosampler. A 1 μl volume of samples was injected at 250 °C. For chromatography, a polyethylene glycol column (VF-WAXms, Agilent Technologies, Santa Clara, USA) was used in split mode (50 split ratios). Helium was used as a carrier with a 1 ml/min flow rate. The temperature of the column was programmed as follows: 60 °C for 2 min, heated to 120 °C at 10 °C/min, and to 240 °C at 5 °C/min rates. Acetate was quantified using acetic acid as a standard, and values measured in the fungal cultures were normalized to media (agar plug) volumes.

Softwares

Statistical analysis and data visualization were done by GraphPad Prism 5 software demo version (GraphPad Software, San Diego California, USA, www.graphpad.com) using ANOVA analysis with the Tukey post hoc test.

Photographs and microscopic images were processed by Adobe Photoshop CS6 demo version and image analyses were done by Fiji (Schindelin et al. 2012).

Results

Isolation and identification of P. chlamydospora

Pch strains were identified according to morphological characteristics and by sequencing the ITS gene region. Three Pch isolates were selected for further studies and are listed in Table 1.

Table 1 Data on the identification and origin of P. chlamydospora strains used in the present study
Full size table

Developmental characteristics of P. chlamydospora isolates growing on congo red-amended solid medium

The possible biofilm formation of the examined Pch strains (P46, P201, P621) was tested by culturing them on a nutrient-rich medium amended with congo red (YS5C80). Congo red is routinely used for this purpose, as it forms a black precipitate with biofilm extracellular matrix components (mainly polysaccharides) in both bacteria and fungi (Moreno et al. 2023). All three tested strains produced such black pigment, not in the colonies themselves, but around them (Fig. 1A). This suggests that this color reaction is more likely due to the presence of secreted soluble polysaccharides, rather than an ECM covering the cells. It is also notable that P201 and P621 formed dense aerial mycelia, masking the red coloration of the colonies, which later can be observed in the case of P46. The production of abundant aerial mycelia is probably due to the cell wall stress caused by congo red (Ram and Klis 2006). Because of the above limitations of YS5C80, a decreased concentration of congo red was used in further experiments (8 mg/ml; YS5C8 medium). All three Pch strains cultured on this medium lacked the aerial mycelia and showed significant absorption of congo red, indicating high polysaccharide content in the colonies (Fig. 1A). The wrinkled surface of the colonies is also notable, whose phenotype is characteristic of biofilm-forming bacteria growing on solid surfaces (Hofer 2022; Ray et al. 2012). Microscopic examination of the development of colonies revealed that P46 and P201 established a dense mycelial network at 2 dpi, while P621 produced elongated hyphae with fewer branches (Fig. 1B). It was also notable, that dense cellular patches formed on the mycelial mats of all strains at early stages (2 dpi for P46 and P201, 4 dpi for P621) of development, with a high affinity toward congo red, suggesting high polysaccharide content (Fig. 1B). These patches fused into a homogenous layer at 4 dpi in the case of P46 and P201, and 6 dpi in the case of P621.

Fig. 1
figure 1

Development of Pch (strains P46, P201, and P621) colonies on YS5 medium in the presence of congo red at 25 °C. A Six-day-old colonies growing on media amended with 80 mg/ml (YS5C80) or 8 mg/ml (YS5C8) congo red. B Merged PC and red fluorescence (congo red staining) photomicrographs of colonies growing on YS5C8 medium. Scale bars represent 100 μm. C Photomicrographs of the cross-section of a P46 colony at 6 dpi (days post-inoculation) growing on YS5C8 medium, after staining with acridine orange. Fluorescence was collected with the red filter set for visualization of congo red and green filter set for observation of acridine orange. Their composite image was also prepared. Scale bars represent 100 μm

Full size image

Visualization of the congo red fluorescence in the cross-sections of the mature colonies suggested that this layer is a thick, biofilm-like structure (Fig. 1C). Visualization of fungal cells with acridine orange staining revealed the presence of several layers of fungal cells in this biofilm-like structure, and expansively growing mycelia at the bottom of the colonies (Fig. 1C), which latter showed weak staining with congo red (Fig. 1C). This result suggests that most of the polysaccharide content of the colonies is concentrated to the biofilm-like structure.

To get a deeper insight into the biofilm-like structure, mature colonies were mechanically disrupted and examined. Cellular aggregates were released from the cultures (Fig. 2A), containing mycelia (Fig. 2A), and several swollen, yeast-like cells (Fig. 2A, inset). These cells could be clearly distinguished from the conidia of the fungus (Fig. 2A).

Fig. 2
figure 2

Detailed microscopic examination of mechanically disrupted Pch colonies growing on YS5C8 at 25 °C, at 6 dpi (days post-inoculation). A PC photomicrograph of a cellular aggregate from a colony of P201. White arrows mark hyphae, and white asterisks mark conidia. The inset shows cells with yeast-like morphology. Scale bars represent 10 μm. B Photomicrograph of a mechanically disrupted colony of P46. The sample was dried for 16h at room temperature and photographs were taken with DIC optics and by the visualization of congo red fluorescence with the red filter set. Scale bars represent 20 μm

Full size image

Observation of dried, disrupted biofilms suggested, that the cellular aggregates are embedded in an extracellular matrix which strongly stained by congo red, probably due to its polysaccharide content, and it was absent in the case of mycelial cells (Fig. 2B).

Production of acetate by P. chlamydospra isolates

All three examined Pch strains produced a substantial amount of acetate when cultivated on a solid YS5 medium (Fig. 3, Table S1) with a maximum of 13.4 mM measured in the case of P46 at 7 dpi. All three strains showed a plateauing effect in the acetate production within 5–7 dpi, with P621 as the poorest producer with a maximum of 10 mM at 7 dpi.

Fig. 3
figure 3

Mean values of acetate production of P46 (red circle), P201 (green square), and P621 (blue triangle) Pch strains on YS5 solid medium, at 25 °C. Letters mark significantly (p < 0.05) differing datasets. The color of the letters corresponds to the Pch strain

Full size image

Effects of acetate on the colony development of P. chlamydospora isolates

Pch isolates were grown for 12 days on YS5C8 medium amended with 0–100 mM sodium acetate to study the effects of acetate on colony development (Fig. 4). All strains in the case of all acetate concentrations formed wrinkled colonies on the inoculated surface, surrounded by radially striated, smooth margins with different width, corresponding to the radial growth of the colonies. With the increasing amount of acetate, the radial growth at the colony margins was decreased, resulting in colonies with a yeast-like morphology in the presence of 100 mM acetate (Fig. 4A). Quantification of the growth rates in triplicate experiments showed that P621 has a significantly lower radial growth compared to P46 and P201, even in the absence of acetate (Fig. 4B, Table S2). All colonies showed a red coloration—originating from the adsorption of congo red—which significantly increased with the increasing concentration of acetate supplement (Fig. 4C, Table S3) in all tested strains, according to quantification of the density of red pixels by image analysis in three replicate experiments. It is also notable, that in the absence of acetate, P621 colonies absorbed a significantly higher amount of congo red compared to P46 and P201.

Fig. 4
figure 4

Effects of acetate supplementation (0–100 mM) on the development of Pch colonies on YS5C8 medium, grown for 12 days, at 25 °C. A Photographs of Pch colonies. B Mean values of the radial growth of Pch colonies. C Mean values of the red pixel density of Pch colonies measured by image analysis. Red circles show P46, green squares show P201, and blue triangles show P621 strain values in B and C. Letters mark significantly differing (p < 0.05) datasets. The color of the letters corresponds to the Pch strain. D DIC photomicrographs of Pch colony margins in the absence (a)–(c), or the presence (d)–(f) of 100 mM acetate supplement. Scalebars represent 20 μm

Full size image

Microscopic examination of the colony margins showed only mycelial growth in the absence of acetate in the case of P46 and P201 (Fig. 4a and b on panel D), while the previously observed yeast-like cells (Fig. 2A) can be occasionally detected on P621 colonies (Fig. 4c on panel D). These yeast-like cells became abundant in the presence of 100 mM acetate in the case of all Pch strains (Fig. 4d–f on panel D).

Growth of P. chlamydospora isolates on grapevine wood

Pch strains were grown on sterilized grapevine wood chips and visualized after congo red staining (Fig. 5). In situ examination of fungal biomass growing on the wood, resulted in the detection of fungal cellular aggregates with high affinity to congo red and yeast-like morphology (Fig. 5a), similar to the aggregates observed in disrupted Pch colonies (Fig. 2A and B). Scraping of fungal biomass from the inoculated wood chips allowed a detailed observation, further reinforcing this similarity (Fig. 5b). Meanwhile, most of the detected fungal biomass formed mycelia.

Fig. 5
figure 5

Photomicrographs of P46 Pch strain growing on autoclaved grapevine wood chips at 25 °C, for 7 days, after staining with congo red. a Merged blue fluorescence (plant tissue autofluorescence) and red fluorescence (congo red) images of fungal cellular aggregates on wood. b Merged DIC and red fluorescence (congo red) images of fungal cellular aggregates scraped from wood. Scale bars represent 10 μm

Full size image

Discussion

While biofilm formation is in the focus of several studies on human pathogenic bacteria (Vestby et al. 2020), there is very little known about the role of fungal biofilms in the pathogenesis, except for the dimorphic Candida species affecting human health (Cavalheiro and Teixeira 2018; Ramage et al. 2023). The criterium that can distinguish biofilm growth from mycelial growth in filamentous fungi would be the presence of ECM previously reported for example in Aspergillus fumigatus (Reichhardt et al. 2015) or F. graminearum (Shay et al. 2022). In the present study, the formation of a multicellular layer embedded in a polysaccharide-rich ECM was demonstrated in three Pch strains on a solid medium, whose structure is suggested to be considered a biofilm.

The development of this layer was initiated by the formation of dense, polysaccharide-rich cellular patches, shortly after the establishment of mycelial mat in the colonies. In this early stage, strain P621 showed a delay in the process, probably in connection with the strain’s poor ability to form a mycelial network after germination. Examination of mechanically disrupted mature biofilms revealed cellular aggregates consisting of swollen mycelia as well as yeast-like cells. This latter morphology has not been reported previously in the literature for Pch. The differentiation in cellular morphology in connection with biofilm growth was previously reported in F. graminearum (Shay et al. 2022) by the detection of swollen bulbous hyphae. The presence of ECM, and its association with the cellular aggregates—as well as its absence in mycelia at the bottom of the colonies—was demonstrated by congo red staining, suggesting the developmental co-regulation, or a causal relationship between the morphological shift and ECM production.

Since biofilm development is regulated by QSMs in both bacteria and fungi, the identification of such molecules in Pch was straightforward to further reinforce the biofilm growth of the fungus. While well-known fungal QSMs (e.g., tyrosol, farnesol, Rodrigues and Černáková 2020) could not be detected, all three Pch strains produced acetate in high amounts with a plateauing effect. Strain P621 was the poorest producer, suggesting it is more sensitive to acetate as a QSM. Supplementation of growth media with acetate resulted in developmental changes related to biofilm formation: reduction in mycelial growth, and increased absorption of congo red from the medium, which latter is probably due to the increased production of ECM as suggested by our previous results. Acetate also promoted morphological differentiation at the cellular level, resulting in the abundant presence of yeast-like cells at 100 mM concentration, even at the colony margins. Colonies of strain P621 showed significantly reduced mycelial growth and an increased absorption of congo red, as well as a higher tendency to form yeast-like cells, even in the absence of acetate supplement. These results are in accordance with the previously proposed higher sensitivity of the strain to this molecule as a QSM. Considering the delayed development of P621 at early stages of growth, the QSM role of acetate is possibly expressed at later stages of colony formation, or affected by yet unidentified factors, like other QSMs. The function of acetate as a QSM was previously reported in bacteria as a promoter of biofilm production (Chen et al. 2015), and in Saccharomyces cerevisiae as a regulator of metabolism (Zhang et al. 2022), while there are no similar observations in filamentous fungi.

The development of structures satisfying the above-suggested definition of Pch biofilm (aggregates of yeast-like cells with high affinity to congo red) was also demonstrated on sterilized grapevine wood. This result suggests, that Pch biofilms may also occur in infected host plants, or at least the nutrients derived from the decomposing plant material are sufficient for biofilm growth.

While our knowledge of the role of biofilms in plant-fungal interactions is very limited, several studies are available in the case of bacteria. Biofilm is believed to be an important pathogenicity factor in plant-infecting bacteria, especially in vascular pathogens (Mina et al. 2019). The possibly closest analogue of Esca among the bacterial infections of grapevine is Pierce disease, caused by the biofilm-forming species Xylella fastidiosa (Rapicavoli et al. 2018). The similarities between Esca and Pierce disease are conspicuous. Both syndromes are caused by a vascular infection, both show a chronic nature and are expressed as discolorations and necroses on annual parts and eventually, both can lead to the death of the infected plant. The biofilm development of X. fastidiosa is enhanced in the presence of grapevine xylem sap, suggesting the importance of this type of growth in pathogenesis (Zaini et al. 2009). Biofilms of X. fastidiosa showed increased tolerance to copper, a widely used antimicrobial in grapevine protection (Rodrigues et al. 2008). This phenomenon is probably due to the presence of persister cells in the biofilm (Muranaka et al. 2012). Occlusion of the xylem vessel by X. fastidiosa is suggested to contribute to the development of foliar symptoms (Sun et al. 2013) like in the case of Esca, and possibly related to the biofilm formation of the pathogen.

Our results suggest, that the Esca pioneer pathogen Pch is capable of biofilm growth. The associated ECM may serve as a defense against host antifungals and may also contribute to the occlusion of xylem vessels. The presence of yeast-like cells in the biofilm suggests the dimorphic growth of the fungus, and this specialized cellular morphology may aid the spreading of the fungus in the vascular system of the host. The process of biofilm formation is proved to be positively regulated by acetate, through the QS mechanism. The QS-dependent nature of biofilm growth may partly explain the controversies between Pch occurrence in the host and expression of Esca symptoms, as well as the long latency of the disease.