Introduction

Vitis (family Vitaceae) is a plant genus that includes the economically important grapes, and thus because of its importance, its pathogens have received a considerable amount of attention during the past decade (Yan et al. 2015; Chethana et al. 2017). The importance of this fruit is associated with its multiple uses; as a source of nutrition, health and medicinal value, as well as its high economical significance (Dohadwala and Vita 2009; Bokulich et al. 2014). About 90% of cultivated grapes in the world are V. vinifera, which comprises wine, as well as table grapes. This genus comprises 79 accepted species of perennial woody and herbaceous vines. There are thousands of cultivars of V. vinifera that has been grown successfully around the globe (Terral et al. 2009). Species of Vitis are valued for their decorative foliage providing ornamental value for the genus. Their ability to cover walls and arches, as well as providing shade has made them important in domestic cultivation.

Numerous diseases of grapes have been identified which reduce the yield and quality of this fruit crop (Úrbez-Torres et al. 2009). Among various pathogens known on grapevine, the damage caused by fungi is significant (Úrbez-Torres 2011). Most studies on fungal pathogens in grape have focused on their pathogenic phase, which relies on direct observation and isolations of fungal pathogens from infected grape material. Fungi that may live within the host tissue are known as endophytes and are considered to cause symptomless infections (Lane and Kirk 2012). Plant pathogenic fungi can survive by changing their biotrophic mode from pathogenic to saprotrophic or at least can remain dormant on the decaying plant materials and become active when suitable conditions for infection exist (Hoppe et al. 2016; Purahong et al. 2018). For example, Botrytis cinerea causing gray mold disease in grape is able to live as a parasite in green tissues and as a saprotroph in dead plant material (Armijo et al. 2016). Unfortunately, our knowledge on saprotrophic fungal community associated with V. vinifera is limited, especially those obtained by high-resolution culture-independent techniques. The percentage of potential fungal pathogens hidden in saprotrophic community is still unclear.

Saprotrophs are organisms that derive nourishment from dead or decaying organic matter (Hyde et al. 2007). Saprotrophs are heterotrophic organisms that break down the complex compounds of dead organisms (Deighton 2016). They play an important role as decomposers of dead organic matter in natural ecosystems by releasing enzymes from hyphal tips (Duarte et al. 2006; Bucher et al. 2004). Saprotrophic fungi can be either macrofungi (Agaricus sp., Phallus sp.) (García et al. 1998) or microfungi (Aspergillus sp., Dothiorella sp., Mucor sp., Neomassaria sp., Rhizopus sp.) (Vohník et al. 2011; Hyde et al. 2016). Hyde et al. (2007) and Purahong et al. (2018), provided evidence that some plants accommodate large numbers of saprotrophic taxa and that some might be host- or organ-specific. Other than being decomposers, saprotrophs can also provide other eco-system services, such as soil formation, defence against pathogens, as a food source and modification of pollutants (Deighton 2016). Promputtha et al. (2007, 2010) provided evidence that not only fungal pathogens, but also fungal endophytes, can switch lifestyles to saprotrophs. Thus, the study of saprotrophic fungal communities associated with V. vinifera can provide information on strict saprotrophism as well as potential endophytes and pathogens with saprotrophic ability.

Fungal species identification has traditionally been based on direct observation, microscope examination, culture dependent isolation and phylogenetic analysis (Cai et al. 2011; Hyde et al. 2010, 2017; Rastogi and Sani 2011; Fadrosh et al. 2014, Tibpromma et al. 2017). Such studies have investigated the microbial ecology of various environments (Rastogi and Sani 2011). However, it has been recognized that the actual number of microbes in nature, often exceed the number of microbes identified by traditional methods (Fadrosh et al. 2014). Traditional approaches rely on growing the organisms in media, but many of the microbes in the environment may not be cultivatable (Stewart 2012). Artificial medium typically allows growth of only a small fraction of, often fast growing organisms. Therefore, traditional techniques do not provide a total community resolution (Hoppe et al. 2016).

During the past decade, microbial research made a shift from phylogenetic analyses to experimental characterization of communities through the use of complex experimental designs (Kozich et al. 2013). This shift focused on relatively inexpensive next-generation sequencing approaches (NGS) and powerful bio-informatics tools to analyse the microbial ecology (Carraro et al. 2011; Rastogi and Sani 2011). High-throughput DNA sequencing allows us to understand the presence of microbes and how their communities are structured in complex ecosystems. Microbiome analysis is a culture-independent technique which requires a low quantity of sample, but with a high sequencing depth. The term microbiome refers to the entire habitat including the microorganisms, their genomes and the surrounding environment. It is characterized by the application of one or combinations of metagenomics, metabonomics, metatranscriptomics and metaproteomics (Marchesi and Ravel 2015). Analysis of the plant microbiome involves linking microbial ecology and the plants biology and functions, and at the same time viewing microorganisms as a reservoir of additional genes and functions for their host (Vandenkoornhuyse et al. 2015). Mycobiome refers to the fungal component in a habitat (Underhill and Iliev 2014). However, these techniques may also have disadvantages. For example, the conditions that we select to do the PCR can give us biased results. Sometimes it is difficult to understand whether the fungi identified by this technique actually exist in the natural system (Mitchell and Zuccaro 2006). Therefore, in order to obtain a better resolution in species identification, richness and distribution patterns of microbes, a combination of both approaches (i.e. traditional and culture-independent) are needed. However, we are aware of no studies that have been conducted using both these approaches.

There have been many studies of major pathogens from Vitis using both morphology and phylogeny (Úrbez-Torres et al. 2012, 2013a, b; Dissanayake et al. 2015; Jayawardena et al. 2015, 2016a; Chethana et al. 2017). Even though a number of sexual and asexual fungi have been reported on Vitis species, updated information of the taxa present in this genus is lacking. Only some have good illustrations and gene sequence data. Our knowledge on saprotrophic fungal communities associated with V. vinifera is limited and data based on high-resolution culture-independent technique is lacking. Besides the percentage of potential fungal pathogens hidden in saprotrophic community is still unknown.

In this study, we aim to (i) provide taxonomic information on the saprotrophic microfungi collected from China, Italy, Thailand and Russia, (ii) compare traditional and culture-independent approaches for characterizing the saprotrophic fungal communities associated with two cultivars of V. vinifera in China, (iii) quantify plant pathogens and endophytes hidden in the saprotrophic fungal community and (iv) provide a worldwide checklist for the fungi on Vitis species based on previous and current research.

Materials and methods

Study site, sampling and isolation of fungi

Fungal species associated with Vitis sp. were collected from China (Beijing, Sichuan and Yunnan Province), Italy (Province of Forlì-Cesena), Russia (Rostov Region) and Thailand (Chiang Sean) (Tables 1, 2). Shoots, leaves, inflorescences, bark and root samples of Vitis vinifera were used for isolation. The same set of samples from Beijing (Red Globe cultivar) and Yunnan Province (Carbanate Gernischet cultivar) were used for the mycobiome analysis to establish the fungal communities (Fig. 1). The sample sets were randomly split into two subsamples to employ the two approaches at the same time. Specimens were incubated in a moist chamber for 3–7 days at 25 °C, if they did not sporulate. Fungi were isolated by a modified single spore/conidial isolation method (Chomnunti et al. 2014) from the samples. Growth rate, colony characteristics and sexual/asexual morph morphology were determined from cultures grown on potato-dextrose agar (PDA) at 25 °C, under 12 h light/12 h dark. Fungal mycelia and spores were examined by differential interference contrast (DIC) and photographed with an axio Imager Z2 photographic Microscope (Carl Zeiss Microscopy, Germany) (Supplementary Figs. S1a–S1d, S2). Forty conidial measurements were taken for each isolate. All microscopic measurements were recorded with ZeM PRo 2012 software. Representative herbariums are deposited in the herbarium of Mae Fah Luang University, Chiang Rai, Thailand (MFU) and in Kunming, China (KIB). Representative cultures were deposited at Mae Fah Luang Culture Collection (MFLUCC), Beijing Academy of Agriculture and Forestry Sciences, China (JZB) and Kunming Culture Collection (KUMCC).

Table 1 Taxa identified in China, Russia, Italy and Thailand by directly observing specimens
Full size table
Table 2 Taxa identified from the two grape cultivars and their life modes using the traditional approach
Full size table
Fig. 1
figure 1

Dead V. vinifera samples at collection sites

Full size image

DNA extraction, PCR amplification, sequencing and phylogenetic analysis

The methods used are presented in detail in Jayawardena et al. (2018).

Culture-independent approach: mycobiome analysis

Two cultivars were selected for this analysis; Red Globe being the table grape cultivar and Carbanate Gernischet being the Wine grape cultivar. Samples of Red Globe (RG) were collected from Yanqin District of Beijing while samples of Carbanate Gernischet (CG) were collected from Yunnan Province were used for this analysis. For each cultivar, three representative grapevine plants were sampled. The root, bark, shoot, inflorescence and leaves were homogenized and sub-sampled. For culture-independent technique of fungal communities, total DNA extraction was performed using 1 g of ground specimens using 2× CTAB method. All the DNA samples were quantitated and quality checked with the NanoDrop ND-2000C spectrophotometer (ThermoFisher Scientific, Dreieich, Germany). DNA extracts were then stored at − 20 °C for further analysis. For fungal Illumina sequencing, we targeted the Internal Transcribed Spacer 1 (ITS1) region of ribosomal RNA gene cluster. ITS1 was amplified with the forward primer ITS5-1737 (GGAAGTAAAAGTCGTAACAAGG) and reverse primer ITS2-2043R (GCTCGCTTCTTCATCGATGC) (White et al. 1990). The PCR reaction was performed in a 50 ml volume that contained approximately 10 ng of DNA, ExTaq buffer, 0.2 mM of dNTPs, 0.2 mM of each primer, and 2 units of ExTaq DNA polymerase. The cycling consisted of an initial denaturing at 94 °C for 30 s, followed by 25 cycles of denaturing at 94 °C for 30 s, annealing at 54 °C for 1 min and extension at 72 °C for 2 min, and a final extension at 72 °C for 8 min. All PCR reactions were carried out with Phusion® High-Fidelity PCR Master Mix (New England Biolabs Inc. Ipswich, MA, USA). The PCR products were mixed with same volume of 1× loading buffer (contained SYB green) and then operated electrophoresis on 2% agarose gel for quality detection. Only samples with bright main strip between 400–450 bp were chosen for further experiments. The qualified PCR products were mixed in equidensity ratios. Then, mixture PCR products were purified with Qiagen Gel Extraction Kit (Qiagen, Germany) following the manufacture’s protocol.

Sequencing libraries were generated using TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, San Diego, CA, USA) following manufacturer’s recommendations and index codes were added. The library quality was assessed on the Qubit® 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. At last, the library was sequenced on an IlluminaHiSeq2500 platform and 250 bp paired-end reads were generated.

Paired-end reads were assigned to samples based on their unique barcode and truncated by cutting off the barcode and primer sequence. Paired-end reads were merged using FLASH (V1.2.7, http://ccb.jhu.edu/software/FLASH/) (Magoč and Salzberg 2011). Quality filtering on the raw tags was performed under specific filtering conditions to obtain the high-quality clean tags (Bokulich et al. 2013) according to the QIIME (V1.7.0, http://qiime.org/index.html) quality controlled process (Caporaso et al. 2010). The tags were compared with the reference database (Unite Database, https://unite.ut.ee/) using UCHIME algorithm (UCHIME Algorithm, http://www.drive5.com/usearch/manual/uchime_algo.html) to detect chimera sequences (Edgar et al. 2011), and then the chimera sequences were removed (Haas et al. 2011). Then the Effective Tags were finally obtained. Sequences analysis was performed by Uparse software (Uparse v7.0.1001, http://drive5.com/uparse/) (Edgar 2013). Sequences with ≥ 97% similarity were assigned to the same OTUs. Representative sequence for each OTU was screened for further annotation. For each representative sequence, the Unite Database (https://unite.ut.ee/) (Kõljalg et al. 2013) was used to annotate taxonomic information based on Blast algorithm, which was calculated by QIIME software (Version 1.7.0) (http://qiime.org/scripts/assign_taxonomy.html). In order to study phylogenetic relationship of different OTUs, and the difference of the dominant species in different samples (groups), multiple sequence alignment were conducted using the MUSCLE software (Version 3.8.31, http://www.drive5.com/muscle/) (Edgar 2004).

All OTU abundance information was normalized using a standard of sequence number corresponding to the sample with the least sequences (45, 246 sequences). From the data set, rare OTUs (singletons), which could have potentially originated from sequencing errors (Kunin et al. 2010), were removed. We used a Mantel test based on Bray–Curtis distance measure with 999 permutations to assess the correlation between the whole matrix and a matrix excluding the rare OTUs (Hammer et al. 2001; Hoppe et al. 2016). The results indicated that the removal of rare OTUs from the fungal communities had no effect (RMantel = 1.000, P = 0.002). Subsequent analysis of alpha diversity and community composition were all performed basing on these normalized rare OTUs removal data. The fungal ITS rDNA genes Illumina sequencing data are deposited in the NCBI under the BioProject No PRJNA437133.

Statistical analysis

Mycobiome analysis

All datasets related to fungal OTU richness were tested for normality and equality of variances using the Jarque–Bera test. To assess the coverage of the sequencing depth, individual rarefaction analysis was performed for each sample using the “diversity” function in PAST (Hammer et al. 2001). In this work we used observed fungal OTU richness and Shannon diversity index as the measures for fungal diversity. The difference in fungal OTU richness between the two deadwood species was compared using a two-sample t test in PAST. To visualize the fungal community compositions, we used non-metric multidimensional scaling (NMDS) analysis based on the Bray–Curtis dissimilarity index calculated PAST. Similarity Percentages (SIMPER) analysis using PAST was used to obtain the identity and relative abundances of the fungal taxa that contributed to 92.92% of the observed pair-wise variation in the fungal community composition due to different V. vinifera cultivars. To accounting for the effect of locations when compared the fungal community compositions of the two grape cultivars which were collected from different locations, we eliminated all location specific fungal OTUs (87 OTUs). We finally retained 139 OTUs for the community composition analysis using NMDS based on the Bray–Curtis dissimilarity. The results from these reduced datasets were highly consistent compared with the total datasets (Supplementary Fig. S3a, b). Potential fungal functional groups were identified using the online Guilds application tool: FUNGuildb (Nguyen et al. 2015). The ITS1 fragments were extracted from both the Sanger sequencing (traditional) and Illumina sequencing datasets using ITS1. The output showed that both datasets have the ITS1 region except culture sequences of the genus Neopestalotiopsis (9 sequences). The sequence similarity based comparison was performed using the cd-hit-est-2d algorithm at 90% similarity level for a genus level comparison.

Diversity analysis

Taxa were recorded as either present or absent from each sample. Occurrence of a fungus was designated based on the presence of a particular fungus on the host samples. Percentage occurrence of a taxon on one sample was calculated using the following formula (Tsui et al. 2001; Yanna and Hyde 2002; Wang et al. 2008):

$$ \begin{aligned} & {\text{Percent occurrence of a taxon A (\%)}} = \frac{\text{Occurrence of taxon A}}{\text{Occurrence of all taxa in one sample}} \times 100 \\ & {\text{Species richness}} = {\text{the number of different species represented in an ecological community}} \\ \end{aligned} $$

Following diversity indices were calculated using the R software for the two cultivars and the habits.

  1. (i)
    $$ {\text{Shannon}}{-}{\text{Wiener's Index (H)}} = \sum p_{i} \ln p_{i},$$

    where p i is the frequency of fungal species I occurring on a specific sample (Begon et al. 1993; Wong and Hyde 2001; Wang et al. 2008).

    1. (ii)
      $$ {\text{S}}{\o}{\text{rensen's index of similarity (S)}} = 2c/(a + b), $$

      where a is the total number of species on host A, b is the total number of species on host B and c is the number of species on both host. Similarity is expressed with values between 0 (no similarity) and 1 (absolute similarity) (Wang et al. 2008).

Compiling the checklist

The checklist is based on, articles in referred journals, Index to Saccardo’s Sylloge fungorum, Petrak’s Lists, Index of Fungi, graduate student theses, books, and web-based resources such as annual reports on this host and the SMML database (https://nt.ars-grin.gov/fungaldatabases/) (latest accessed 14-9-2017). The mode of life, such as pathogen, endophyte or saprotroph is listed. The checklist includes species names, family, life modes, disease name if any and locality. The current name is used according to Index Fungorum (2018) and Wijayawardene et al. (2017) and the classification follows Wijayawardene et al. (2018). Genera and species are listed in alphabetical order. Identification confirmed by molecular data is marked with an asterisk (*). In some cases, the host name given in the original citation was changed to be consistent with current taxonomy. In a few cases, neither the species cited nor a proper synonym was identified and the species name was used as originally cited.

Results

Species identified from fresh collections based on morphology and phylogeny (traditional method)

Fungal saprophytic diversity and community composition of the two grape cultivars: traditional method

Examination of decaying leaves, shoots, inflorescence, berries, root and bark of two cultivars of V. vinifera from China yielded 461 collections for the Red Globe variety and 180 collections for Carbanate Gernischet. The Red Globe variety had higher species richness (41) than the Carbanate Gernischet variety (23), however the Shannon diversity was not significantly different (Table 3). The majority of the culturable saprotrophic fungi were ascomycetes. However, there were two species belonging to Agaricomycetes and one species belonging to Oomycota incertae sedis. Thirty genera and 45 taxa were identified based on morphology and phylogenetic sequence data. From the identified isolates, 32.6% were Sordariomycetes, 26.1% Dothideomycetes 19.7% Eurotiomycetes, 6.5% Mucoromycetes, 4.4% Agaricomycetes, 2.2% Leotiomycetes and 2.2% of Oomycota incertae sedis. There were four taxa belonging to Zygomycota incertae sedis, which we were unable to identify. The identified Sordariomycetes belonged to Bionectriaceae (6.7%), Diaporthaceae (6.7%), Glomerellaceae (20%), Hypocreaceae (20%), Nectriaceae (13.3%), Schizoparmaceae (6.7%), Stachybotryaceae (13.3%) and Sporocadaceae (13.3%). Dothideomycete isolates belonged to Botryosphaeriaceae (16.7%), Cladosporiaceae (16.7%), Didymellaceae (33.3%) and Pleosporaceae (33.3%). The rest of the isolates belong to Mucoraceae (6.5%), Peniophoraceae (2.2%), Pythiaceae (2.2%), Rhizopodaceae (2.2%), Sclerotiniaceae (2.2%) and Trichocomaceae (17.4%). Among those 45 taxa, we found 19 species that were common on both cultivars: Actinomucor elegans, Alternaria alternata, Aspergillus niger, A. aculeatus, Cladosporium cladosporioides, Cladosporium sp., Clonostachys rosea, Coniella vitis, Diaporthe eres, Fusarium oxysporum, Mucor racemosus, Penicillium terrigenum, Phoma medicaginis, Rhizopus oryzae, Talaromyces amestolkiae, T. pinophilus, T. purpurogenus and T. harzianum. The Sørensen’s index of similarity of the two grape cultivars was 0.58. We have identified 45 taxa to species level, although in six cases the identification is only to genus level due to lack of enough molecular data. Taxa were identified using both morphology and molecular techniques. Identified species are listed in Tables 1 and 2 (Supplementary Figs. S1a–d, S2).

Table 3 Richness and diversity (mean ± SD, n = 3) of fungi detected in the two Vitis vinifera cultivars
Full size table

Fungal saprophytic diversity and community composition of the two grape cultivars: culture-independent technique

Bioinformatics processing of the sequence data sets

A total of 638,146 quality-filtered fungal ITS reads were obtained after removal of chimeric and the unique tag (3703 sequences) sequences. After normalizing all data sets to a smallest sequence read (45, 246 sequences) and removing all rare taxa, the final analyse data sets contained 226 fungal OTUs. Phylogenetic trees for the top 20 species in different samples of the two cultivars Carbanate Gernischet and Red Globe of Vitis vinifera are given in Fig. 2. With the high number of sequence reads per sample obtained in this study, the sample-based rarefaction curves almost reached saturation for all samples (Fig. 3a). We used the observed OTU richness and Shannon diversity directly for further analyses.

Fig. 2
figure 2

Phylogenetic tree of top 20 species in different samples of the two cultivars Carbanate Gernischet and Red Globe of Vitis vinifera

Full size image
Fig. 3
figure 3

Rarefaction curves (a), and Venn diagrams show distribution of OTUs across different samples (1–3): (b) in both Carbanate Gernischet (CG) and Red Globe (RG) cultivars, (c) only Carbanate Gernischet and (d) only Red Globe

Full size image

Fungal saprotrophic OTU diversity and distribution in the two cultivars of Vitis vinifera

Diverse fungi colonized the debris samples derived from Carbanate Gernischet and Red Globe cultivars. Fungal OTU richness was not significantly different between the two V. vinifera cultivars tested in this study, ranging from 122–137 (127.33 ± 4.84 (mean ± SD); Carbanate Gernischet) and 116–141 (126.33 ± 7.54 (mean ± SD); Red Globe) (t = 0.11, P = 0.916). Shannon diversity also showed a similar trend ranging from 1.98–2.35 (2.16 ± 0.11 (mean ± SD); Carbanate Gernischet) and 1.78–1.98 (1.86 ± 0.06 (mean ± SD); Red Globe) (t = 2.41, P = 0.07). In total we detected 226 fungal OTUs with 176 and 189 belonging to the cultivars Carbanate Gernischet and Red Globe, respectively. There were 139 Fungal OTUs shared between the two V. vinifera cultivars and 37 and 50 OTUs were specific to Carbanate Gernischet and Red Globe cultivars. When we took each replicate into account, we detected only moderate proportion of fungal OTUs shared across different replicates (31–44%, 51 OTUs, Fig. 3b). Distributions of fungal OTUs across replicates for the two cultivars and for each specific cultivar are shown in Fig. 3b–d.

Fungal saprophytic community composition: culture-independent technique

The NMDS ordination plot and SIMPER analysis revealed distinct fungal communities in the two cultivars of V. vinifera samples (Table 4 and Supplementary Fig. S3a, b). Overall, fungal community composition of the two cultivars had the overall average dissimilarity of 94.29% (based on Bray–Curtis distance measure) and 30 fungal OTUs mostly responsible for differences in fungal community composition were all together accounting for 94.41% of the overall average dissimilarity (Table 4). The difference in fungal community composition between the two cultivars of V. vinifera was detected across different taxonomic levels (Supplementary Figs. S4, S5).

Table 4 Similarity percentages (SIMPER) analysis showing the top 30 fungal OTUs mostly responsible for differences in fungal community composition between Carbanate Gernischet(CG) and Red Globe (RG) cultivars; OA Dissimilarity = overall average dissimilarity
Full size table

In Carbanate Gernischet, members of Ascomycota were commonly detected accounting for 77% (46% Sordariomycetes, 19% Eurotiomycetes and 7% Dothideomycetes) of total sequences in this cultivar followed by unidentified phylum (23%; Fungal OTU-7) and Basidiomycota and Zygomycota (less than 0.1%). In Red Globe, almost all sequences were assigned to Ascomycota (97%; 51% Eurotiomycetes, 42% Sordariomycetes, and 3% Dothideomycetes) followed by Basidiomycota (3%; 1% Agaricomycetes and 1% Tremellomycetes) and unidentified phylum (Fungal OTU-7) and Zygomycota were negligible (altogether less than 0.5%). Phylogenetic tree for the abundance at genus level using the top 35 genera detected in the two cultivars are shown in Fig. 4. The difference between the fungal community composition of the two cultivars of V. vinifera were clearly demonstrated at OTU level: Trichothecium roseum OTU-1, Fungal-OTU-7, Aspergillus piperis OTU-5 and Nectriaceae OTU-3 were commonly detected (10–20%) in Carbanate Gernischet, but almost absent in Red Globe (represented by Aspergillus OTU-11, Ilyonectria macrodidyma OTU-4, Aspergillus cibarius OTU-2 and Diaporthaceae OTU-10; 11–19%; Table 4).

Fig. 4
figure 4

Abundance phylogenetic tree at genus level (top 35 genera) in two cultivars [Carbanate Gernischet (CG1-3) and Red Globe (RG1-3)] of Vitis vinifera

Full size image

Using presence/absence data, we found that for both Vitis vinifera cultivars Ascomycota (Carbanate Gernischet = 151 OTUs and Red Globe = 162 OTUs) was the richest OTU phylum followed by unidentified phylum, Basidiomycota (11–14 OTUs) and Zygomycota (2 OTUs). Patterns of the richest OTU classes and orders were similar for both cultivars: Sordariomycetes (Hypocreales (Carbanate Gernischet = 32 OTUs and Red Globe = 39 OTUs), Sordariales (Carbanate Gernischet = 13 OTUs and Red Globe = 13 OTUs), Microascales (Carbanate Gernischet = 10 OTUs and Red Globe = 11 OTUs)], Eurotiomycetes (Eurotiales, Carbanate Gernischet = 41 OTUs and Red Globe = 38 OTUs) and Dothideomycetes (Pleosporales, Carbanate Gernischet = 7 OTUs and Red Globe = 7 OTUs).

Comparing and matching of traditional and culture-independent approaches

Several commonly detected fungal genera (Aspergillus, Clonostachys and Fusarium) were detected in both approaches. However, there are many highly or frequently detected genera in the mycobiome that were not detected in the traditional method. These include Acrostalagmus, Aureobasidium, Ceratobasidium, Chrysosporium, Ilyonectria, Lasiodiplodia, Microascus, and Trichothecium (Supplementary Table S1). Some frequently isolated fungi, especially the fast growing ones (Rhizopus and Mucor) were not detected in mycobiome analysis.

ITS sequences obtained from both traditional and culture-independent methods were compared using the query and cluster cover. This showed that the saprotrophs detected from the two approaches are consistent in most cases. However, in few cases we found inconsistent identifications which have arisen from the lower level of taxonomic assignment in the culture-independent (amplicon sequencing) as compared with traditional approaches. For example, Diaporthe eres and Alternaria identified via the traditional approach were identified as Diaporthaceae and Pleosporaceae in the culture-independent analyses. A mismatch was also found between Albifimbria viridis and Myrothecium sp., which are classified in the same order (Hypocreales). The sexual morph genus Talaromyces was matched with its potential asexual morph (Penicillium). We found that twelve taxa detected from traditional method form a cluster (91–100% similarity) with 25 fungal OTUs from the culture-independent methods (Table 5). We were able to assign 25 OTUs from NGS: 20 OTUs to genus and 5 OTUs (similarity 99–100%) to species level respectively (Table 5). We removed two OTUs as singletons (Botryosphaeria OTU-178 and Ascomycota OTU-213) as they were detected only once. However, in the direct matching of ITS sequences, these fungal OTUs showed 97 and 100% similarity to Botryospaeria dothidea and Coniella vitis, respectively. The other fungi that we were able to identify to the species level are Aspergillus niger, Clonostachys rosea, Botrytis cinerea, and Albifimbria viridis. Most of the frequently detected genera in the traditional approach (i.e. with relative abundance higher than 5%; Alternaria, Clonostachys, Fusarium) were also detected in culture-independent approach. Rhizopus sp. and Talaromyces sp. were frequently detected in the traditional approach, but exhibited low relative abundances or disappeared in the culture-independent approach.

Table 5 Matching of fungal isolates to the saprotrophic mycobiome of Vitis vinifera
Full size table

Fungal functional groups identified using traditional and culture-independent approaches

Among the 45 identified taxa based on traditional method, 17 are well known pathogens on V. vinifera causing severe yield as well as economic loss to viticulture around the world (Table 2). Six species of secondary pathogens of V. vinifera were also identified in this study. Most of the pathogens tend to survive or overwinter on dead plant material as saprotrophs and act as the primary inoculums once the conditions are favourable (Armijo et al. 2016).

In total, 143 fungal OTUs (63% of total fungal OTUs) were successfully assigned for their functions (Supplementary Table S1). We identified six functional groups of fungi associated with dead materials of V. vinifera: saprotrophs, plant pathogens, endophytes, fungal parasites–saprotrophs (mycoparasites–saprotrophs), ectomycorrhizae and animal pathogens. The fungal community was dominated by saprotrophs (102 OTUs) and plant pathogens (22 OTUs), which accounted for 71% and 15% of the function assigned to fungal OTUs in this study. Clonostachys, Lasiodiplodia and Trichothecium, were the most commonly detected plant pathogen genera with relative abundances 1–10%. Botrytis sp., an important fungal pathogen in grape, was also detected with low relative abundance. Endophytes together with endophyte–saprotrophs and endophyte–plant pathogens (9 OTUs) contributed little and most OTUs were detected with low relative abundance, except, Acrostalagmus luteoalbus. All fungal OTUs with their potential functions are listed in Supplementary Table S1.

Checklist of fungi on Vitis

Nine-hundred and six fungal taxa have been reported on Vitis species and are listed in Table 6, although the actual number of fungal taxa associated with this host is likely much higher. It is not possible to reconfirm all previous reports by re-examining collections to confirm their identities. In many cases no fungarium material is linked to the reports, while examining nearly 900 specimens would be an almost impossible task. Even if it was were possible, it would most likely be futile, since molecular data would be needed to establish correct names. This is extremely difficult based on the presently available techniques and not permitted by many fungaria. Most of the 905 taxa reported from Vitis species do not have sequence data. Therefore, recollecting and sequencing these taxa are essential to establish and accurate species list associated with Vitis species.

Table 6 Check list of fungi on Vitis sp. (classification follows Wijayawardene et al. 2017, 2018)
Full size table

Discussion

Before the advent of molecular data in taxonomy, studies on the fungi on Vitis were based on traditional methodology and have resulted in hundreds of records of fungi from this host genus (Table 6). Most recent studies have been related to pathogens that affect grape yield and production (Úrbez-Torres et al. 2012, 2013a, b; Dissanayake et al. 2015; Liang et al. 2016; Jayawardena et al. 2015, 2016a; Yan et al. 2015; Chethana et al. 2017) and have resulted in well-resolved taxonomy as they have used molecular data. However, studies on saprobes using molecular data and culture-independent techniques have not been used to identify the fungi on Vitis to date. In this study, we therefore provide the first work comparing saprobes on Vitis sp. using both traditional and culture-independent approaches, with well-resolved taxonomic identifications based on molecular analyses. The taxa derived from both approaches are compared as the same samples were used in the study. We have also established the saprotrophic communities associated with both wine and table grapevine cultivars and demonstrate cultivar specific communities for each grapevine cultivar. A checklist of fungi of Vitis is also provided which is an important resource for viticulture.

Microfungi collected from China, Italy, Russia and Thailand

Sixty-seven saprotrophic taxa from 46 genera were identified in this study (Table 1). Using traditional methodology and analyses of molecular data, we identified two new species, and 41 new host or distribution records for V. vinifera. Taxonomic details, descriptions, photographic plates and phylogenetic analyses are provided in Jayawardena et al. (2018). Some of these genera have a wide distribution. For example, botryosphaerious and Colletotrichum taxa have a wide distribution. These taxa are well-known pathogens and can be spread to other countries undetectable through the exportation of rootstocks. Some genera are only known from one or two countries. This may be due to the lack of data on the fungi associated with this host.

Comparisons of traditional and culture-independent approaches for characterizing the saprotrophic fungal communities associated with two cultivars of Vitis vinifera

Most previous studies on fungi on grapevine have relied on traditional approaches (Table 6). Some recent identification of isolated taxa have incorporated analyses of ITS sequence data (Guo et al. 2003; Promputtha et al. 2007), but this was shown not to be accurate (Ko et al. 2011). The most recent studies on fungal pathogens of grapevine have incorporated multigene analysis to accurately resolve taxa (Dissanayake et al. 2015; Jayawardena et al. 2015; Yan et al. 2015; Chethana et al. 2017).

Most previous studies did not address the total community of fungi on Vitis vinifera. Pancher et al. (2012) carried out an extensive study on endophytes on this host, showing that how various anthropic and nonanthropic factors shape microbial communities. There have been extensive studies on the disease causing agents with more than 150 taxa known to cause various diseases of grapevine. For example, Colletotrichum species cause grape ripe rot of Vitis vinifera worldwide (Jayawardena et al. 2016b). There have however, been no investigation on the saprobes of grapevines using molecular identification and there has been no study using mycobiome analysis to reveal saprotrophic communities. The study of saprotrophs is important, as they not only decay dead leaves and branches, thus beneficial recyclers, but they may also become pathogens when conditions are suitable.

This study therefore fills this void by establishing the saprotrophic fungi on Vitis vinifera using both traditional and culture-independent approaches. In this study we did not obtain similar results from the two methods. In the traditional method, 45 species belonging to 30 genera were identified (Table 2), while in culture-independent method 226 OTUs’ and 72 genera were identified. Even though we isolated directly from the fruiting bodies, some fungi were not able to grow on media. Several single spore isolations were unsuccessful. This may be due to the availability of nutrient content, pH, temperature, and presence of inhibitors and the time of incubation. The number of isolates obtained was less than the actual fungal community and can be misinterpreted (Hugenholtz et al. 1998). These conditions make it difficult to accurately identify and document the vast number of unrecognized taxa (Lücking and Moncada 2017). For example, in this study the total identified taxa from the traditional method were 45. Therefore, to overcome the constraints of traditional methods, culture-independent techniques are proposed as an alternative technique (Hoppe et al. 2016).

The aim of environmental sequence nomenclature is to place names of species of fungi that would otherwise be left undescribed (Lücking and Moncada 2017). These techniques can provide sequence reads almost 1000 times more than the traditional DNA sequencing methods (Lücking and Moncada 2017). Lücking and Moncada (2017) showed that a formally recognized unnamed lichenicolous basidiomycete can be considered as a new genus, with seven new species, although there is no physical type specimens are available. These authors also suggested that this would allow the recognition of thousands of species of voucher less taxa detected through environmental sequencing techniques. However, there are several constrains to NGS methods. DNA may not be recovered from all genotypes and the results of NGS can be biased towards the most abundant organisms at the time of sampling (Ward et al. 1990). The reason for this is that the relative abundance (Fig. 5) of microbial species in a natural habitat is rarely equal. Usually, with a few species being predominant among a larger group of common species makes it difficult to identify the species that are actually present. NGS are mainly based on analysis of ITS regions (Schoch et al. 2012). However, due to the high variability of ITS regions (ITS1 and ITS2), reliable sequence alignments are difficult to obtain for some fungal taxa. Therefore, this method is not reliable for species level identification. The identification levels are usually reported at the genus level or even higher taxonomic levels, such as family or order (Purahong et al. 2018). Another constraint of NGS is that the correspondence of OTU with species can be unreliable. OTUs are defined based on the similarity threshold, usually with a 97% (Sneath and Sokal 1973). However, some species have genes that are 97% similar, which will result in merged OTUs containing multiple species. In the same way, a single species may have paralogs that are < 97% similar, causing the species to be split across two or more species. Some identified clusters, even when a majority, may be false, due to the artifacts including reading errors and chimeras (Sneath and Sokal 1973). Assessing species richness and diversity of a microbial community using culture-independent method (rarefaction curves), suggests that OTUs are observations of organisms with ‘negligible error’. Also, it suggests that the number of reads correlates well with the total number of individuals present in the community. However, if the majority of OTUs are experimental artifacts, the traditional species richness estimations cannot be applied. The measures between sample variations will tend to reflect differences in artifact frequencies rather than biological differences (Sokal and Sneath 1963).

Fig. 5
figure 5

Relative abundance of the top 10% phylum from different samples of the cultivars Carbanate Gernischet and Red Globe of Vitis vinifera

Full size image

Artifacts can be occurred due to several reasons. PCR amplification steps can be affected by preferentially/differentially that can hinder the detection of some genotypes when analysing bulk DNA extracts from a substrate (Kanagawa 2003). Primer mismatches, a lower rate of primer hybridization, occurrence of heteroduplexes and chimeric amplicons can generate additional signals that do not correspond to the genotypes in the same samples (Suzuki and Giovannoni 1996; Kanagawa 2003). Also, the analysis of fungal rRNA genes limits identification to the genus or family level (Anderson and Cairney 2004).

Dissanayake et al. (2018) in her study using paired-end Illumina sequencing with 55, 822 high quality sequences per endophyte sample (saturated rarefaction curves for all samples) revealed 59 OTUs (the majority containing genera level identification) that were similar to genera revealed by the traditional method (28 species).

Traditional versus culture-independent methods: can matching of these two approaches enable us to identify correct fungal taxonomic information at genus and species levels?

In this study, taxa (OTUs) of Aspergillus, Botrytis, Cladosporium, Clonostachys, Fusarium, Penicillium, Phoma and Talaromyces were identified using both traditional and culture-independent approaches. Some fast growing fungi may be dominant in the culture plates, even though in the natural habitat they may be minorities. Many hyphomycetes tend to grow faster than the other groups of fungi. So, they may suppress the growth of other important, dominant fungi. The fast growing fungi (hyphomycetes) identified in the traditional method such as Mucor and Rhizopus were not recognized in the culture-independent method. The majority of the genera identified in the traditional method are phytopathogens, while in the culture-independent method the majority are saprotrophs. In the traditional method using both morphological and molecular approaches, we were able to identify many taxa to species level, although in six cases the identification is only up to the genus level due to lack of data. We have generated a phytogenetic tree for the genus Colletotrichum using the strains identified in the traditional approach as well as the OTU identified in the culture-independent method. OTU-234, was identified as Glomerellaceae sp. in the culture-independent approach. The blastn result of OTU-234 in NCBI shows 100% similarity to many strains of C. gloeosporioides. Two-hundred and fifteen basepairs of OTU-234 were used in the alignment (Supplementary Fig. S6). In the phylogenetic tree constructed using the ex-type strains of gloeosporioides complex and the truncatum complex, OTU-234 cluster within the truncatum complex, closely to C. curcumae (Supplementary Fig. S7). This example also provides evidence that the NGS sequences is not reliable to identify an organism to the species level. However, in the culture-independent technique we were able to identify 90 out of 226 OTUs to species level, and the rest were identified to genera or family level. These may not be correctly identified as in general NGS fungal taxa identification may be only accurate to the genus level (Purahong et al. 2018), which suggests that the sequence data from the culture-independent approach is inadequate to accurately identify species. The overlap between the two methods in identifying the taxa to the species level is negligible.

Matching between the traditional and culture-independent data allows us to have a better understanding concerning the functional information of the fungal OTUs resulting from culture-independent methods. Next generation sequencing often results in sequences that are associated with taxa, which have not been reported in previous studies (Tejesvi et al. 2010; Ko et al. 2011; Taylor et al. 2016). However, as most of these OTUs are identified to genus or family level, it makes it difficult to relate whether these are actually correctly identified and whether the use of this method is important. In the preparation of the checklist of fungi on Vitis species, the authors had to eliminate most of the taxa that were identified using NGS, as those data can be unreliable. In our study, we compared the sequence similarity between the two cultivars using a 90% similarity of ITS1 sequence data, followed by a manual BLAST based identification of the respective OTUs. We considered the ITS similarity at 98–99% as the same species (Garnica et al. 2016; Jeewon and Hyde 2016). In this criterion, identification of genera can be bias/difficult as some of the data in databases have mistakes or they may be inaccurate. The increase of the ITS similarity to 99–100% can give us better and reliable identification of the species. However, ITS sequence data alone will not be able to identify the complexes genera such as Colletotrichum and Diaporthe to their species level. For a better resolution of these genera protein coding gene regions are required.

Can direct matching between traditional and culture-independent methods help to identify the rare taxa?

Our results show that some singletons, which were usually removed as artifacts or errors of the NGS may actually be real OTUs. In this study, we found two OTUs (Botryospaeria OTU-178 and Ascomycota OTU-213) as singletons and removed them from the analysis. However, with direct matching, we found that these OTUs are Botryospaeria dothidea and Coniella vitis. Therefore, we can assume that not all the singletons are artifacts and matching between traditional and culture-independent methods can help to identify the real rare taxa in the fungal community.

Potential effect of grape cultivars (table grape (Red Globe) and wine grape (Carbanate Gernischet) on fungal saprotrophic community composition and richness

Another aspect of this study was to study whether there is any difference in the fungal communities based on cultivars. In the present study, traditional and culture-independent approaches allows the identification of potential roles of the saprotrophs in the two grapevine cultivars. In this study we identified more than 10 main and important fungal pathogens of grapevine using both methods. With the evaluation of both community composition and community diversity we were able to identify that the fungal communities of the two grape cultivars appear to be different. Alternaria vitis, Albifrimbria viridis, Bipolaris maydis, Botryosphaeria dothidea, Botrytis cinerea, Colletotrichum hebeinse, C. truncatum, C. viniferum, Didymella pomorum, Dothiorella sarmentorum, Epiccocum nigrum, Fusarium sp., Mucor circinelloides, Paraphoma chrysanthemicola, Neopestalotiopsis clavispora, Stagonosporopsis sp.1, Minimedusa sp., Peniophora sp., Penicillium brevicompactum and P. citrinum were recorded only from Red Globe cultivar while Albifimbria verrucaria, Neopestalotiopsis vitis, Pythium amasculinum, Stagonosporopsis sp.2, Trichoderma lixii and Septoriella allojunci were recorded from Carbanate Gernischet cultivar in the traditional method. In the culture-independent approach, Acremonium chrysogenum (OTU-195), Apodus sp. (OTU-235), Ascomycota (OTU-80, 99, 182, 222, 253), Aspergillus sp. (OTU-116, 199), Candida mucifera (OTU-227), Cylindrocarpon sp. (OTU-171), Dactylellina phymatopaga (OTU-58), Davidiella tassiana (OTU-86), Deroxomyces sp. (OTU-225), Fungal (OTU-245, 254), Fusarium cf. dimerum (OTU-162), Helotiales (OTU-91), Hypocreales (OTU-65), Kernia nitida (OTU-184), Kernia pachypleura (OTU-211), Lecanicilium dimorphum (OTU-101), Lentinus squarrosulus (OTU-249), Lophiostoma sp. (OTU-142), Microascales (OTU-138), Metarhizium pinghaense (OTU-210), Myceliophthora fergusii (OTU-198), Myrothecium sp. (OTU-238), Nectriaceae (OTU-97, 194), Papulospora equi (OTU-148), Phialosimplex caninus (OTU-187), Psathyrellaceae (OTU-145), Pseudallescheria angusta (OTU-183), Pyronemataceae (OTU-237), Remersonia sp. (OTU 108), Sordariomycetes (OTU-128), Sordariales (OTU-196, 200, 214), Xylaria sp. (OTU-159) were found only in association with the Red Globe cultivar while Acremonium sp. (OTU-188), Apllosporella yalgorensis (OTU-85), Arachnomyces kanei (OTU-170), Aspergillus melleus (OTU-143), Aspergillus wentii (OTU-175), Cadophora luteo-olivaceae (OTU-146), Ceratobasidiaceae (OTU-219), Chaetomium carinthiacum (OTU-177), Chysisporium lobatum (OTU-150), Cladosporium grevilleae (OTU-121), Dothideomycetes (OTU-140), Eurotiales (OTU-76), Fungal (OTU-82, 104, 165), Gymnascella aurantiaca (OTU-151), Hansfodia sp. (OTU-232), Hypocreales (OTU-49, 92, 202), Lasiophaeriaceae (OTU-189), Leptosphaeria sp. (OTU-205), Magnoporthaceae (OTU- 141, 149), Microascales (OTU-114), Microascus sp. (OTU-185), Microdochium sp. (OTU-225), Nectriaceae (OTU-218), Penicillium ilerdanum (OTU-163), Penicillium neocrassum (OTU-168), Podospora communis (OTU-190), Scopulariopsis sp. (OTU-164), Sordariales (OTU-133), Spiromastix princeps (OTU-139), Thielavia basicola (OTU-155), Trichomaceae (OTU-156, 166) were recorded on from Carbanate Gernischet cultivar.

The difference of the two fungal communities can be due to the geographic variation of the cultivars. This can be a result from the interactions with specific V. vinifera varieties and its soil and climatic conditions (Bokulich et al. 2014). Red Globe cultivar was collected in Beijing, which is a region in North of China and Carbanate Gernischet cultivar was collected from Yunnan which is in the southern part of the country. The difference between the fungal communities in regions may be a function of a neutral process, where these different communities established by chance and lack of species dispersal allows these communities to persist (Martiny et al. 2006). This difference can also be due to the Baas Becking hypothesis, which states that there is no limit to the range of species but that selection sorts these species and defines community composition and diversity in any one area (Hanson et al. 2012). Climate can also co-relates with differences in fungal communities in China, as one moves North up in China, the climate becomes increasingly cold and dry so the pattern of lower fungal species richness in the northern most regions hints that selection might have a role in determining these patterns.

Plant pathogens and endophytes in the saprotrophic fungal community

Species richness and distribution patterns of saprotrophic fungi in a vineyard can provide important insights into the roles of each fungal group for the stability and functioning of its respective ecosystem (Kubartova et al. 2012). However, knowledge of saprotrophic fungi associated with grapevine is very much limited. In this study, we identified 17 primary and six species of secondary pathogens of grapevine as saprobes using the traditional method, while 27 OTUs were identified as both primary and secondary pathogens from dead material of Vitis vinifera in the culture-independent method.

Species of Alternaria are responsible in causing berry rots, raisin molds and rots as well as pedicel and rachis diseases (Barbe and Hewitt 1965; Gonzalez and Tello 2011; Tao et al. 2014, Ariyawansa et al. 2015) and also considered as wound and secondary invaders. Alternaria alternata and A. vitis were isolated in this study. Aspergillus is a causal agent of berry rots as well as a wound and secondary invader (Hewitt 2015). In our study A. aculeatus and A. niger were recorded using the traditional method, while A. aculeatus was also recorded from culture-independent method. Botryosphaerious taxa are well-known to be associated with grapevine canker and die back (Úrbez-Torres et al. 2012, 2013a, b). In our study we identified Botryosphaeria dothidea and Dothiorella sarmentorum as saprotrophs using traditional methodology. Lasiodiplodia was recorded in the culture-independent method.

Botrytis is another genus that we obtained in both traditional and independent approaches. Botrytis cinerea is a pathogen of grapevine causing Botrytis bunch rot and blight all over the world (Fournier et al. 2013; Hyde et al. 2014; Javed et al. 2017). Cladosporium was also recorded in both approaches. Species of this genus cause minor foliage diseases of grapevine, as well as bunch rots (Bensch et al. 2015). Clonostachys is another genus recorded in both approaches. Clonostachys rosea is known to cause root rot of grapevine in Switzerland (Casieri et al. 2009). Colletotrichum hebeiense, C. truncatum and C. viniferum were recorded in the traditional method. Species of this genus cause grape ripe rot affecting the quality and production of grapevine (Yan et al. 2015). Diaporthe eres is another pathogen of grapevine causing die back (Lawrence et al. 2015; Baumgartner et al. 2013; Cinelli et al. 2016; Fischer et al. 2016; Bastide et al. 2017), which was recorded via the traditional methodology as well as via the culture-independent approach.

Species of Fusarium cause wilt disease of grapevine (Castillo-Pando et al. 2001; Gonzalez and Tello 2011). This genus was recorded in both approaches. Neopestalotiopsis vitis recorded from traditional method is a pathogen causing fruit rot, die back and leaf spots of grapevine (Jayawardena et al. 2015, 2016a). Coniella vitis is a pathogen causing white rot of grapes, identified using traditional methods (Chethana et al. 2017). Species of Penicillium are wound and secondary pathogens of grapevines causing bunch rot (Kim et al. 2007). Rhizopus oryzae is another wound and secondary pathogen causing bunch rots of grapevines (Hewitt 2015).

Several genera were identified only in the culture-independent method. Aplosporella is known to cause lesions on grapevine stems in China (Tai 1979). Claviceps is known to be a pathogen on grasses and cereals, but has not been recorded as a pathogen of grapevine (Mey et al. 2002). Therefore, this study provides the first record of this genus on V. vinifera. Cylindrocarpon species are known to cause the black foot disease of grapevine (Abreo et al. 2010, 2012; Mohammadi et al. 2013a, b). Devriesia is a facultative pathogen, but there are no records of this species on V. vinifera (Seifert et al. 2004). Therefore, this study provides the first record of this genus on V. vinifera. Species of Leptosphaeria has been reported as endophytes and saprotrophs of grapevine (Crane and Shearer 1991). However, some species of this genus can be pathogenic to some economically important crops (Fitt et al. 2006). Monographella is a known leaf pathogen on rice, barley, maize and wheat (Daamen et al. 1991; Hock et al. 1992; Tatagiba et al. 2015). However, there are no records of species of this genus associated with grapevine. Therefore, this study provides the first record of Monographella associated with grapevine. Species of Phaeoacremonium are causal agents of Esca disease around the world (Garcia-Benavides et al. 2013). Species of Trichothecium are known to cause berry rot of grapevine, but this is not considered as a major pathogen on grapevine (Oh et al. 2014).

Even though genus Volutella is a facultative pathogen causing leaf spot and cankers (Henricot et al. 2000; Shi and Hsiang 2014), there is no record of this genus occurring on grapevine. Therefore, this study provides the first record of Volutella associated with V. vinifera as a saprotroph.

Among the 45 identified saprotrophic taxa, 17 are well known pathogens of Vitis vinifera causing severe yield as well as economic losses to viticulture worldwide (Table 2). Six secondary pathogens of were also identified in this study. Most of the pathogens tend to survive or overwinter on dead plant material as saprotrophs and act as the primary inoculums once the conditions are favourable (Armijo et al. 2016).

Many studies have shown that most pathogenic fungi can survive unsuitable conditions, such as cold during the winter, by changing their life mode to saprotrophs, and become active pathogens again once the conditions are suitable. Therefore, dead plant materials are the potential primary inocula for plant pathogens in vineyards. In order to avoid this problem, vineyards must be kept clean. If there are any dead grapevines they must be removed and if possible should be burned. This will reduce the pathogenic fungi from year to year.

Checklist of fungi on Vitis

Nine-hundred and five micro- and macro- fungal taxa reported on Vitis species are listed in this study. This is an updated worldwide checklist of fungi on Vitis. These taxa are distributed in 156 families and 343 genera. For each species, family, life mode, diseases caused and the known locality as well as references are provided.