Ochraceocephala foeniculi gen. et sp. nov., a new pathogen causing crown rot of fennel in Italy

Abstract A new disease of fennel is described from Sicily (southern Italy). Surveys of the disease and sampling were conducted during spring 2017 and 2018 in Adrano and Bronte municipalities (Catania province) where this crop is widely cultivated. Isolations from the margin of symptomatic tissues resulted in fungal colonies with the same morphology. Pathogenicity tests with one isolate of the fungus on 6-month-old plants of fennel reproduced similar symptoms to those observed in nature. Inoculation experiments to assess the susceptibility of six different fennel cultivars to infection by the pathogen showed that the cultivars ‘Narciso’, ‘Apollo’, and ‘Pompeo’ were more susceptible than ‘Aurelio’, ‘Archimede’, and ‘Pegaso’. Phylogenetic analyses based on a matrix of the internal transcribed spacer (ITS), the large subunit (LSU), and the small subunit (SSU) rDNA regions revealed that the isolates represent a new genus and species within the Leptosphaeriaceae, which is here described as Ochraceocephala foeniculi gen. et sp. nov. This study improves the understanding of this new fennel disease, but further studies are needed for planning effective disease management strategies. According to the results of the phylogenetic analyses, Subplenodomus iridicola is transferred to the genus Alloleptosphaeria and Acicuseptoria rumicis to Paraleptosphaeria.


Introduction
Fennel (Foeniculum vulgare Mill.), native in arid and semi-arid regions of southern Europe and the Mediterranean area, is used as a vegetable, herb, and seed spice in the food, pharmaceutical, cosmetic, and healthcare industries. Italy is the leading world producer of fennel (around 85% of the world production), with 20,035 ha of area cultivated and a total production of 537,444 tons. Fennel represents an important crop widely cultivated in Sicily (southern Italy) with 1,620 ha harvested and a production of 35,930 tons (ISTAT 2018). Several diseases caused by fungi have been reported from this crop throughout the world (Table 1). Amongst soilborne diseases, brown rot and wilt caused by Phytophthora megasperma and crown rot caused by Didymella glomerata (syn. Phoma glomerata) were reported in Italy (Cacciola et al. 2006;Lahoz et al. 2007).
In 2017, a new disease was first observed on fennel in a farm of Adrano area (Catania province, eastern Sicily, Italy). The disease symptoms were necrotic lesions on the crown, root, and stem of fennel plants. Disease incidence initially was about 5% on 'Apollo' cultivar. However, in 2018 different surveys conducted in the same area showed a high increase of the incidence on three different cultivars with yield losses of about 20-30%. The aims of the present study were to identify the causal agent obtained from symptomatic fennel plants, using morphological characteristics and DNA sequence analyses, to evaluate the pathogenicity of one representative isolate and to evaluate the susceptibility of different cultivars of fennel to the newly described disease.

Collection of samples and fungal isolates
In order to identify the causal agent of the fennel disease, 30 samples were collected during several surveys in Adrano and Bronte area (Catania province, eastern Sicily). Pieces of tissue obtained from different parts of fennel plants (crown, root, and stem) were surface disinfected for 1 min in 1.5% sodium hypochlorite solution, rinsed in sterile water, placed on potato dextrose agar (3.9% PDA, Oxoid, Basingstoke, UK) amended with 100 mg/L of streptomycin sulfate (Sigma-Aldrich, USA) to prevent bacteria growth, and then incubated at 25 ± 1 °C for seven days. Fungal colonies consistently grown from symptomatic tissues were subcultured on new PDA plates. Subsequently, single-spore isolates were obtained from these pure cultures and stored at -20 °C in sterile 15% glycerol solution. The fungal isolates were provisionally identified by cultural and morphological characteristics, and they were deposited in the culture collection of the Department of Agriculture, Food and Environment, University of Catania. One representative isolate (Di3A-F1; ex holotype culture) was deposited at the Westerdijk Fungal Biodiversity Institute (CBS), Utrecht, the Netherlands. The holotype specimen of the new pathogen species was deposited in

Morphology
For culture characteristics, cultures were grown on 2% (w/v) malt extract agar (MEA, VWR) and on corn meal agar (CMA, Sigma-Aldrich) supplemented with 2% w/v dextrose (CMD). Colony diameters and morphologies were determined after seven days of incubation at room temperature (22 ± 1 °C) and daylight.
Microscopic observations were made in tap water. Methods of microscopy included stereomicroscopy using a Nikon SMZ 1500 equipped with a Nikon DS-U2 digital camera, and Nomarski differential interference contrast (DIC) using a Zeiss Axio Imager.A1 compound microscope equipped with a Zeiss Axiocam 506 colour digital camera. Images and data were gathered using the  or Zeiss ZEN Blue Edition software packages. Measurements are reported as maxima and minima in parentheses and the range representing the mean plus and minus the standard deviation of a number of measurements given in parentheses.

DNA extraction and PCR amplification
The extraction of genomic DNA from pure cultures was performed by using the Wizard Genomic DNA Purification Kit (Promega Corporation, WI, USA). Partial regions of six loci (ITS, LSU, and SSU rDNA, RPB2, TEF1, TUB2) were amplified; for details on the primers and annealing temperatures used for PCR and sequencing, see Table 2. The PCR products were sequenced in both directions by Macrogen Inc. (South Korea) or at the Department of Botany and Biodiversity Research, University of Vienna using the ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit v. 3.1 (Applied Biosystems, Warrington, UK) and an automated DNA sequencer (3730xl Genetic Analyser, Applied Biosystems). The DNA sequences generated were assembled with Lasergene SeqMan Pro (DNASTAR, Madison, USA). Sequences generated during the present study were uploaded to Genbank (Table 3).

Phylogenetic analysis
According to the results of BLAST searches in GenBank, the newly generated ITS, LSU, and SSU rDNA sequences of the fennel pathogen were aligned with selected sequences of Leptosphaeriaceae from Gruyter et al. (2013) and complemented with a few recent additions from GenBank. The familial and generic concept of Leptosphaeriaceae implemented here follows the molecular phylogenetic studies of Gruyter et al. (2013), Ariyawansa et al. (2015), and Phookamsak et al. (2019). Due to insufficient RPB2, TEF1, and TUB2 sequence data available in Genbank for the study group, the sequences of these markers could not be included in phylogenetic analyses, but they were deposited in GenBank (Table 3). A combined SSU-ITS-LSU rDNA matrix was produced for phylogenetic analyses, with six species of Coniothyrium (C. carteri, C. dolichi, C. glycines, C. multiporum, C. telephii, C. palmarum) from Coniothyriaceae added as the outgroup according to the results of the phylogenetic analyses of Gruyter et al. (2013). As the rDNA sequences of the fennel pathogen isolates were (almost) identical (see Results section below), only a single isolate (CBS 145654 = Di3A-F1; ex holotype strain) was included in the final matrix. The GenBank accession numbers of sequences used in the analyses are given in Table 4. Sequence alignments were produced with the server version of MAFFT (http://mafft.cbrc.jp/alignment/server), checked and refined using BioEdit v. 7.2.6 (Hall 1999). The combined data matrix contained 3312 characters; i.e. 607 nucleotides of the ITS, 1333 nucleotides of the LSU and 1372 nucleotides of the SSU).
Maximum likelihood (ML) analyses were performed with RAxML (Stamatakis 2006) as implemented in raxmlGUI 1.3 (Silvestro and Michalak 2012), using the ML + rapid bootstrap setting and the GTRGAMMA substitution model with 1000 bootstrap replicates. Table 2. Primers used to amplify and sequence the nuclear internal transcribed spacer (ITS), large subunit (LSU) and small subunit (SSU) rDNA regions, the RNA polymerase II second largest subunit (RPB2) gene, the translation elongation factor 1-α (TEF1) gene and the β-tubulin (TUB2) gene.  (Swofford 2002). All molecular characters were unordered and given equal weight; analyses were performed with gaps treated as missing data; the COLLAPSE command was set to MINBRLEN. MP bootstrap analyses were performed with 1000 replicates, using 5 rounds of random sequence addition and subsequent TBR branch swapping (MULTREES option in effect, steepest descent option not in effect) during each bootstrap replicate. In the Results and Discussion, bootstrap values below 70 % are considered low, between 70-90 % medium and above 90 % high.

Pathogenicity test
To determine the ability of the representative isolate Di3A-F1 (CBS 145654) to cause disease symptoms, pathogenicity tests were conducted on 6-month-old plants of fennel grown in a growth chamber. Five plants for each of the three replicates were used. The inoculum, which consisted of a 6-mm-diameter mycelial plug from a 10-day-old culture on PDA, was inserted in four points for each crown and the wounds wrapped Table 3. Characteristics and accession numbers of isolates collected from fennel plants in Sicily.  with Parafilm to prevent desiccation. Fennel plants inoculated with sterile PDA plugs served as a control. After inoculation, plants were covered with a plastic bag for 48 h and maintained at 25 ± 1 °C and 95% relative humidity (RH) under a 12 h fluorescent light/dark regime. Five days after inoculation the presence of a lesion was evaluated in each inoculation point. To fulfill Koch's postulates, symptomatic tissues taken from the crown of each inoculated plant were plated on PDA and the identity of the fungal isolates was confirmed as described above.

Cultivar susceptibility
To evaluate the susceptibility of six different cultivars of fennel to infection by the pathogen, one experiment was conducted on 1 to 2-month-old seedlings of fennel in a growth chamber. Eight plants for each of three replicates were used. The inoculum, which consisted of a 6-mm-diameter mycelial plug from a 10-day-old culture on PDA, was inserted at the crown of each plant and wrapped with Parafilm to prevent desiccation. Fennel plants inoculated with sterile PDA plugs served as a control. All the replicates were enclosed in plastic bags and maintained at 25 ± 1 °C and 95% relative humidity (RH) under a 12 h fluorescent light/dark regime in a growth chamber until the symptoms were observed. Plant mortality (PM), disease incidence (DI) and symptom severity (SS) were evaluated. Symptom severity was rated using a category scale from 0 to 5, where 0 = healthy plant; 1 = necrotic lesion on crown from 0.1 to 0.2 cm; 2 = from 0.3 to 1 cm; 3 = from 1.1 to 2 cm; 4 = from 2.1 to 3.5 cm; 5 = dead plant. The experiment was performed twice.

Statistical analysis
Data about disease susceptibility of examined fennel cultivars from the repeated experiments were analysed by using the Statistica package software (v. 10; Statsoft Inc., Tulsa, OK, USA). The arithmetic means of PM, DI, and SS were calculated, averaging the values determined for the single replicates of each treatment. Percentage data concerning PM and DI were transformed into the arcsine (sin -1 square rootx) prior to analysis of variance (ANOVA), whereas SS values were not transformed. Initial analyses of PM and DI were performed by calculating F and P values associated to evaluate whether the effects of single factor (cultivar) and cultivar × trial interactions are significant. In the post hoc analyses, the corresponding mean values of PM and DI were subsequently separated by the Fisher's least significant difference test (P = 0.05). Because ordinal scales were adopted for SS data calculation, different nonparametric approaches were used. Kendall's coefficient of concordance (W) was calculated to assess whether the rankings of the SS scores among fennel cultivars are similar within each trial (cultivar × trial interactions). Since in the susceptibility experiment W was higher than 0.9, the SS scores were at first analysed by using Friedman's nonparametric rank test, and subsequently followed by the all possible pairwise performed with the Wil-coxon signed-rank at P < 0.05. On the other hand, when only the cultivar effects were examined, the Kruskal-Wallis non parametric one-way test was preliminarily applied, calculating χ 2 and P value associated.

Collection of samples and isolates
Symptoms referable to infection (Fig. 1a, b) were detected in five commercial farms surveyed in eastern Sicily, Italy. The disease was observed on 3 different cultivars of fennel (4 to 6-month-old) in open fields. The symptoms consisted of depressed necrotic lesions formed near the soil line and affected crown, root, and stem. The lesion was first light brown with wet appearance, becoming dark brown to black with age and sometimes appearing dry. Under favourable conditions (high humidity), the lesion extended and the infection resulted in a crown and root rot. Fungal colonies representing the new fennel pathogen were consistently obtained from symptomatic tissues. A total of 32 single-spore isolates were collected (Table 3). Preliminary identity of the fungal isolates was based on cultural and morphological characteristics. Among these, 17 isolates were obtained from 'Apollo', 14 from 'Narciso', and one from 'Aurelio' cultivars.

Sequencing
All strains of the new fennel pathogen sequenced had identical LSU, SSU, RPB2, TEF1, and TUB2 sequences. Also all ITS sequences were identical, except for a single nucleotide polymorphism (A/G) towards the end of the ITS2 region. All sequences generated during this study were deposited at GenBank; for GenBank accession numbers, see Table 3.

Phylogenetic analyses
Of the 3312 characters included in the phylogenetic analyses, 294 were parsimony informative (222 from the ITS, 62 from the LSU, 10 from the SSU). The best ML tree (lnL = -14211.5558) revealed by RAxML is shown in Figure 2. In the phylogenetic tree, the Leptosphaeriaceae received high (96% ML and MP) support. Within Leptosphaeriaceae, most of the deeper nodes of the tree backbone received low to insignificant support. Highly supported genera include Alloleptosphaeria, Heterosporicola, Leptosphaeria (all three with maximum support) and Alternariaster (99% ML and 100% MP), while Sphaerellopsis received low (53%) and Paraleptosphaeria medium (75%) support only in the ML analyses, and Plenodomus and Subplenodomus were unsupported. Subplenodomus iridicola was not contained within the Subplenodomus clade, but sister species to Alloleptosphaeria italica with maximum support, and Aci- cuseptoria rumicis was embedded within the Paraleptosphaeria clade, indicating that they are generically misplaced. The new fennel pathogen was placed basal to the Plenodomus clade, however, without significant support. Although the new fennel pathogen is closely related to the genus Plenodomus, it is morphologically highly distinct. As no suitable described genus is available, a new genus is therefore established here.

Type species. Ochraceocephala foeniculi Voglmayr & Aiello.
Notes. Ochraceocephala is phylogenetically closely related to Plenodomus, from which it deviates substantially in morphology. Plenodomus species are characterised by pycnidial phoma-like asexual morphs, and while in two Plenodomus species (P. chrysanthemi, P. tracheiphilus) simple hyphomycetous, phialophora-like synanamorphs have been recorded (Boerema et al. 1994), these are very different from the complex conidiophores of the present fennel pathogen. These morphological differences, the lack of a suitable genus within Leptosphaeriaceae and its phylogenetic position therefore warrants the establishment of a new genus.

Host and substrate. Pathogenic on crown, roots and stems of living
Notes. In the phylogenetic analyses (Fig. 2) Subplenodomus iridicola is placed remote from the other species of Subplenodomus, but is sister species to Alloleptosphaeria italica with maximum support; S. iridicola is therefore transferred to the genus Alloleptosphaeria.
Notes. The monotypic genus Acicuseptoria was described by Quaedvlieg et al. (2013) as a segregate of the polyphyletic genus Septoria, and it was characterised by brown, globose pycnidia with conidiophores reduced to ampulliform conidiogenous cells bearing acicular, hyaline, euseptate conidia. However, its position within the Leptosphaeriaceae remained undetermined as no other representatives of the family were included in their phylogenetic tree (Quaedvlieg et al. 2013: fig. 2). In our phylogenetic analyses (Fig. 2), Acicuseptoria rumicis is embedded within the genus Paraleptosphaeria and placed in a highly supported subclade that also contains the generic type, P. nitschkei. Acicuseptoria rumicis is therefore transferred to the genus Paraleptosphaeria.

Pathogenicity test
The representative isolate (CBS 145654) was pathogenic to fennel plants, and produced symptoms similar to those observed in open field after five days (Fig. 1e). The pathogen was re-isolated from the artificially inoculated plants, and identified as previously described. No symptoms were observed on control plants.

Cultivar susceptibility
In the experiments on fennel susceptibility there was always a significant effect of the cultivar on all disease parameters (PM, DI and SS) of pathogen infections (p < 0.0001). Otherwise, a not significant cultivar × trial effect (p > 0.56) was observed for parametric variables (PM and DI) in this repeated experiment (Table 5). Besides, Kendall's coefficient of concordance was 0.96 for SS data, thus indicating very high concordance between the two trials (Table 5). Therefore, the two trials were combined.
Regarding susceptibility of fennel to this phytopathogenic fungus, a great variability was detected among the tested cultivars eight days after inoculation. Comprehensively, cultivar 'Narciso' was the most susceptible since all disease parameters and  its PM value were significantly the highest among the tested cultivars. 'Apollo' was also highly susceptible to infection by the new fennel pathogen, significantly differing only in a slightly lower PM value. 'Pompeo' displayed PM and DI values similar to those recorded for 'Apollo', but its SS score was significantly lower than in the former ( Table 6). In decreasing order of susceptibility, 'Aurelio' did not significantly differ from 'Pompeo' for DI and SS values, but its PM caused by the fennel pathogen was strongly reduced. No dead seedlings (PM = 0) were recorded for both 'Archimede' and 'Pegaso', that significantly differed for DI and SS from the other remaining cultivars. Altogether, 'Pegaso' was the least susceptible cultivar to fungal infection since it showed the lowest values of disease severity.

Discussion
In the present study, 32 fungal isolates were recovered from symptomatic fennel plants in Sicily over a 2-year period. Disease symptoms were observed in three farms, and included necrotic lesions and crown and root rot on three different cultivars. The fungal species obtained from symptomatic tissues was identified based on morphological characters and molecular phylogenetic analyses of an ITS-LSU-SSU rDNA matrix, resulting in the description of the fennel pathogen as a new genus and species, Ochraceocephala foeniculi.
In the phylogenetic analyses, O. foeniculi was revealed as sister group of Plenodomus; however, without significant support (Fig. 2). As commonly observed with ITS-LSU-SSU rDNA data, support of many backbone nodes is low or absent, and additional protein-coding markers like RPB2, TEF1 and TUB2 are necessary for an improved phylogenetic resolution of genera and families in Pleosporales (Voglmayr and Jaklitsch 2017;Jaklitsch et al. 2018). Although we sequenced RPB2, TEF1, and TUB2 for O. foeniculi, it was currently not feasible to perform multi-gene analyses due to insufficient sequence data for most species of Leptosphaeriaceae, in particular for Plenodomus. However, we consider the phylogenetic and morphological evidence conclusive for establishing the new genus Ochraceocephala. Also the generic transfer of Subplenodomus iridicola to Alloleptosphaeria is well substantiated, considering its highly supported phylogenetic position as sister species of Alloleptosphaeria italica, remote from the generic type (S. violicola) and other species of Subplenodomus (Fig. 2). In the phylogenetic analyses of the LSU rDNA matrix of Crous et al. (2018: fig. 1), only few taxa of Leptosphaeriaceae were included, and the phylogenetic position of S. iridicola remained inconclusive due to low resolution; however, also in their analyses it was placed remote from the generic type, S. violicola. In addition, they did not include its closest relative, Alloleptosphaeria italica, although it was mentioned as the closest match of an ITS BLAST search . No asexual morph is known for A. italica (Dayarathne et al. 2015), but the ascomata, asci and ascospores of A. iridicola and A. italica share many traits. Our phylogenetic analyses also showed that Acicuseptoria rumicis should be included within Paraleptosphaeria (Fig. 2). Although it was correctly placed within Leptosphaeriaceae by Quaedvlieg et al. (2013), its position within the family remained undetermined as no other representatives of the family were included in their phylogenetic analyses. As for most other species of Paraleptosphaeria no asexual morphs are known, no comprehensive morphological comparison can currently be made with P. rumicis.
Within Leptosphaeriaceae, O. foeniculi is remarkable and unique by its complex hyphomycetous asexual morph composed of branched conidiophores with phialidic conidiation and conidia produced in basipetal chains. Asexual morphs in Leptosphaeriaceae are typically coelomycetous and phoma-like, which is also the case in the closest relative of Ochraceocephala, Plenodomus (Gruyter et al. 2013). Another genus of Leptosphaeriaceae with a hyphomycetous asexual morph is Alternariaster, which, however, differs significantly by tretic condiogenous cells forming large, brown, septate conidia not produced in chains (Simmons 2007;Alves et al. 2013). Therefore, the unique morphology in combination with an isolated phylogenetic position within Leptosphaeriaceae warrant the establishment of a new genus.
Other fungal species belonging to Leptosphaeriaceae, as well as the closely related Didymellaceae (Odstrčilová et al. 2002;Shaker and Alhamadany 2015) have been reported worldwide in fennel crops. In Italy, crown rot of fennel caused by Didymella glomerata (syn. Phoma glomerata) was recorded from southern Italy (Lahoz et al. 2007). As confirmed in the pathogenicity tests, O. foeniculi caused symptoms on artificially inoculated plants of the same cultivar and, moreover, also on different fennel cultivars that showed some variability in disease susceptibility. To this regard, it is noteworthy that this study also represents a preliminary evaluation of fennel germplasm according to their susceptibility to this new disease. Although these data should be confirmed by additional investigations, this study might provide very useful information for local farmers and technicians. The determination of the extent of susceptibility to O. foeniculi is a starting point for evaluating the tolerance of commercial fennel cultivars to this disease under different agronomic and phytosanitary conditions.
On the basis of the disease incidence and severity observed in the field, we believe that this disease represents a serious threat to fennel crop in Sicily and may become a major problem also to other areas of fennel production if accidentally introduced. Moreover, infected soil could represent an inoculum source for this fungus. Further studies are needed to examine the life cycle of O. foeniculi and to ascertain the cardinal temperatures of the fungus for successful infection since this pathogen is well established in this representative fennel production area. This information is required for the setup and timing of sustainable approaches for soil disinfection, including solarization and/or fumigation at low rates, to reduce the level of the primary inoculum in the soil and hence the disease amount, like successfully applied for other soilborne plant pathogens Aiello et al. 2018).
Although not always conclusive, soil disinfestation and host resistance can be considered environmentally friendly means to be included within integrated pest management (IPM) strategies against crown rot caused by O. foeniculi in order to minimize the number and intensity of fungicide applications.