Characterization of Phytopythium Species Involved in the Establishment and Development of Kiwifruit Vine Decline Syndrome

Since 2012, the kiwifruit vine decline syndrome (KVDS) has progressively compromised Italian kiwifruit orchards. Different abiotic and biotic factors have been associated with the establishment and development of KVDS. During monitoring of orchards affected by KVDS in north-western Italy during 2016–2019, 71 Phytopythium spp. were isolated. Based on maximum likelihood concatenated phylogeny on the ITS1-5.8S-ITS2 region of the rDNA, large subunit rDNA, and cytochrome oxidase I, isolates were identified as P. vexans (52), P. litorale (10), P. chamaehyphon (7) and P. helicoides (2). Phytopythium litorale and P. helicoides are reported for the first time as agents of KVDS in Italy. To demonstrate pathogenicity and fulfil Koch’s postulates, representative isolates of P. vexans, P. litorale, P. chamaehyphon and P. helicoides were inoculated in potted plants. In these trials, waterlogging was applied to stress plant with a temporary anoxia and to favour the production of infective zoospores by the oomycetes. In experiments in vitro, the four species showed the highest growth at 25–30 °C, depending on the media used. P. helicoides was able to grow also at 40 °C. The four species were able to grow in vitro at a pH ranging from 5.0 to 8.0, showing that pH had less effect on growth than temperature. The present study suggests a strong role of different species of Phytopythium in the establishment and development of KVDS. Phytopythium spp. could be favoured by the average increase in soil temperatures during summer, associated with global warming.


Introduction
Kiwifruit production is around 3.5 million tons worldwide [1], and Italy is the second kiwifruit producer after China, with 316,443 tons in 2019, whose 250,000 tons are exported [2]. Since 2012, kiwifruit vine decline syndrome (KVDS) has affected over 10% (almost 2900 ha) of Italian kiwifruit orchards, in various regions of northern (Veneto, Piedmont, Friuli Venezia Giulia), central (Lazio, Emilia Romagna) and southern Italy (Calabria) [1]. However, it is estimated that over 25% of the Italian kiwifruit orchards are compromised [3]. In the areas where KVDS is observed, the most common irrigation system consists of temporary flooding [4,5].
Up to now, the syndrome has been reported only in Italy. Similar vine decline disorders were reported also in other countries but were attributed to a specific pathogen [6,7], to waterlogging [8], to root rotting, or to root asphyxia [9].
Typical KVDS symptoms are a reduction in plant vigour, progressive leaf desiccation starting from the margin to the centre of the leaves, leaf curling with a progression from the basal to the upper leaves [3,10]. Canopy symptoms are generally related to damage of structural roots with brown rotting areas and absence of feeder roots, even though vines with compromised roots could still have asymptomatic canopy [11]. The symptomatology is more evident during summer, when high temperatures and transpiration are recorded. Once the syndrome becomes visible, the plant dies within one year [12].
Different research has aimed to understand the aetiology of KVDS, suggesting that different abiotic and biotic factors may be involved. Sorrenti et al. [1] reported a high disease frequency in silty-sandy soil, where temporary waterlogging occurs. Soil drainage seems to represent one important factor involved in the syndrome. Waterlogging could promote KVDS, even though the disease is reported also on sandy or well-drained soils [3,5,13,14]. Soil structure and its management are also involved in KVDS development [15], as well as the interaction between these factors and weather conditions [12]. As reported by Tacconi et al. [16], a delay in the development of KVDS was obtained using soil convexing and compost in order to improve water drainage and soil texture.
The involvement of biotic factors in KVDS was demonstrated by Savian et al. [3], where symptom development was obtained using soil from infected KVDS orchards, whereas no symptoms were observed using sterilised soil. Different pathogens have been associated with KVDS. Among fungal species isolated from symptomatic tissues, the pathogenicity on kiwifruits plants was demonstrated for Phytophthora cryptogea, P. citrophthora, Phytopythium vexans, P. chamaehyphon, and Desarmillaria tabescens [5,10,11,17]. Among bacteria, the genus Clostridium was associated with the disease [18] even though Donati et al. [11] did not isolate bacterial pathogens neither from affected KVDS orchards nor from healthy ones.
The aim of the present work was to investigate the presence of microorganisms associated with KVDS and their role in the syndrome development, with a focus on oomycetes. The strains were identified by morphological and molecular approaches. Biological characterisation of the isolates included pathogenicity tests performed to reproduce typical symptoms of KVDS and the evaluation of the effect of temperature and pH on their growth.

Oomycete Isolation
Phytopythium spp. isolates were collected from 18 A. chinensis orchards during the period August-October of 2016, 2018 and 2019, in Piedmont, north-western Italy (Tables 1 and S1). The strains were isolated from kiwifruit plants showing typical KVDS symptoms, i.e., reduction in plant vigour, leaf curling, or complete decline, focusing on microorganisms associated with infected tissues. Isolation was carried out from symptomatic rotten roots at the margin between infected and healthy tissue to fulfil the first postulate of Koch, as previously described by Prencipe et al. [10]. Briefly, roots were first surface-disinfected with 1% sodium hypochlorite, washed in sterile deionised water and air-dried. Five fragments of each root were cut at the symptom edges and plated onto Potato Dextrose Agar (PDA, Merck, Germany) and semi-selective oomycete PARP medium (17 g corn meal agar, 0.01 g Pimaricin, 0.01 g Ampicillin 0.01 g, Rifampicin and 0.07 g Pentachloronitrobenzene,) Petri dishes. After 4 days of incubation at 25 ± 1 • C, 71 representative isolates were selected (Table 1), based on colony morphology, and they were maintained in tubes of PARP medium.

Molecular Identification
Genomic DNA was extracted from isolates grown in Potato dextrose broth (PDB, Merck, Darmstadt, Germany) at 25 ± 1 • C on rotary shaker for 8 days, using an Omega E.Z.N.A. Fungal DNA Mini Kit (VWR, Radnor, USA), according to manufacturer's instructions. The ITS1-5.8S-ITS2 region of the rDNA was amplified using primers ITS1 and ITS4 or ITS4 and ITS6 and protocols reported in White et al. [19] and Cooke and Duncan [20]. The large subunit (LSU) rDNA, was amplified using primers (NL1/NL4) and protocol reported in Baten et al. [21] Finally the cytochrome oxidase I gene (COI) was amplified using primes (FM85mod/OomCOILevup) and protocol reported in Robideau et al. [22]. PCR were carried in a volume of 25 µL using: 2.5 µL of Qiagen PCR Buffer, 0.5 µL of MgCl 2 , 0.75 µL of dNTPs (10 mM), 1 µL of each primer (10 µM), 0.2 µL of Taq DNA polymerase and 1 µL (20 ng) of template DNA. The amplicons were checked by electrophoresis at 110 V/cm for 40 min in 1% agarose gel stained with 1 µL of GelRed™ (VWR). Single PCR fragments were purified using QIAquick© PCR purification Kit (Qiagen, Hilden, Germany), and sequenced in both directions by Macrogen, Inc. (Amsterdam, The Netherlands). The DNA Baser program (Heracle Biosoft S.R.L., Arges, Romania) was used to obtain the consensus sequences and alignment was performed using CLUSTALW through Molecular Evolutionary Genetics Analysis (MEGA6) software, version 6.0. After cutting the trimmed regions and manual correction, a dataset of 761 bp, 723 bp and 658 bp for ITS, LSU and COI, respectively, was obtained. The best-fit nucleotide model for the concatenated dataset was obtained using MEGA version 6, as well as to perform the phylogenetic analysis with the Maximum likelihood (ML) algorithm. Reference sequences used for phylogeny, according to the last revision of the genus [23] and the latest species descriptions, are reported in Table S2. All sequences were deposited in GenBank (Table 1).

Morphological Observation
For the macro-morphology, two representative isolates per species (Table 1) were grown onto PDA, Corn Meal Agar (CMA) and Potato Carrot Agar (PCA) Petri dishes [24], as described by de Cock & Lévesque [25]. Each plate was inoculated with a 6 mm mycelial plug in the centre of the plate and observed for colony characteristics (growth, colour, margin shape and texture), after 5 days of incubation at 25 ± 1 • C in the dark.
For the micro-morphology, sporangia and zoospore productions were induced for two isolates per species (Table 1), as described by de Cock & Lévesque [25]. Briefly, pieces of sterilised grass blades were placed onto CMA Petri dishes colonised by actively growing mycelium of the oomycetes. After 24 h, grass blades were transferred into Petri dishes filled with 10 mL of sterile soil broth (100 g sandy soil, 1 L deionised water), incubated at 20 ± 1 • C and exposed to 16 h daylight. Observations were carried out using a Nikon (Eclipse 55t) microscope (Tokyo, Japan) at 40× magnification after 7-14 days, depending on the strain. Twenty measurements were made for each isolate.

Effect of Temperature and pH on In Vitro Growth
For the assessment of the growth cardinal temperatures, the same isolates used for morphological observation (Table 1) were inoculated onto PDA, PCA and CMA, and incubated at 10, 15, 20, 25, 30, 35, 40 and 45 ± 1 • C for 5 days according to de Cock & Lévesque [25] and de Cock et al. [23]. Radial growth was measured daily, along two lines intersecting the centre of the plate, where the inoculum plug (6 mm) was positioned, and data were expressed as growth rate (mm/day).
To assess the effect of pH on the growth of colonies, a mycelial plug (6 mm) of each isolate was inoculated on three plates containing PCA adjusted at specific pH and incubated at 25 ± 1 • C for 5 days. To obtain PCA at specific pH (5.0, 5.5, 6.0, 6.5, 7.0, 7.5, and 8.0), NaoH 1M and HCl 1M solutions were used. The pH was measured using FiveEasy pH meter (Mettler Toledo, Milano, Italy). Radial growth was measured daily along two lines intersecting the centre of the mycelial plug, and data were expressed as growth rate (mm/day).

Pathogenicity Test
The pathogenicity test was carried out on 1-year-old plants of A. chinensis var. deliciosa 'Hayward' grown in 3 L pots containing a sterilised agriperlite-peat mixture. One representative isolate strain per species was used for the inoculation of five kiwifruit plants: P. vexans strain PP1, P. chamaehyphon strain PH6, P. helicoides strain CA2 and P. litorale strain R3A. The inoculum was prepared according to Prencipe et al. [10]. Briefly, each strain was grown for 7 days on a wheat and hemp mixture (100 g wheat, 50 g hemp and 170 mL of water), and used to inoculate the soil at a rate of 6 g/L per pot. Similarly, negative control plants were inoculated with a sterile seed mixture, whereas positive control plants were potted onto diseased-infected soil taken from an orchard of A. chinensis var. deliciosa 'Hayward' located in Saluzzo (Piedmont, north-western Italy), showing typical KVDS symptoms. Plants were kept in greenhouse at 28 ± 5 • C. In order to simulate the field capacity, 2 days post inoculation (dpi), all of the plants were submerged up to the crown level for 72 h. Based on symptoms observed, the disease severity (DS) on Actinidia plants was scored at 12 and 26 dpi using a scale of 0 to 4 ( To fulfil Koch's postulates, re-isolations were performed from the roots of all the plants and the isolates were identified by sequencing the COI gene, as described above. Furthermore, soil pH and temperature were measured at the beginning of the trial and at 26 dpi with a pH-meter and a temperature probe. After 26 dpi, the dry weight of the roots was also measured.

Statistical Analysis
Statistical analyses on the in vitro test and the pathogenicity test were carried out by using IBM SPSS (Chicago, IL, USA) statistics version 25 for the normal distribution analysis using Shapiro-Wilk normality test, the homogeneity of variance using Levene's test and one-way analysis of variance using Tukey's test (p ≤ 0.05).

Oomycete Isolation and Identification
During the summer-autumn period of three years (2016, 2018 and 2019), several isolates of Phytopythium spp. were collected from plants of A. chinensis showing symptoms of KVDS. Based on colony morphology 26, 12, and 33 representative isolates were selected in 2016, 2018, and 2019, respectively, to obtain a total of 71 isolates ( Table  1). The isolates collected in 2016 were all from 'Hayward' kiwifruit, except for one isolate from the cultivar 'Soreli'. The isolates collected in 2018 were from 'Hayward' plants, while those collected in 2019 were from different cultivars: twenty-three from 'Hayward', a green pulp variety, eight from 'Soreli', a yellow pulp variety, and two from 'Dong Hong', a red pulp variety.
Sequencing of the ITS1-5.8S-ITS2 region, the large subunit (LSU) rDNA, and cytochrome oxidase I gene (COI) were used for species assignation. The best-fit model used for the concatenated dataset was Tamura Nei + Gamma distribution. Based on concatenated phylogeny, 52 isolates were identified as P. vexans, 10 as P. litorale, seven as P. chamaehyphon, and two as P. helicoides (Figure 2).
A great intraspecific variability was observed. P. vexans strains were divided into three main groups: one with 39 strains, isolated in the three years of sampling, the second group with three strains, two isolated in 2016 and one in 2018, and 10 strains in the last group, all isolated in 2016 ( Figure 2, Table 1). P. chamaehyphon strains clustered into two groups: one with four strains isolated during 2016 and the second group with three strains isolated in 2019 ( Figure 2, Table 1). P. litorale strains were also divided into To fulfil Koch's postulates, re-isolations were performed from the roots of all the plants and the isolates were identified by sequencing the COI gene, as described above. Furthermore, soil pH and temperature were measured at the beginning of the trial and at 26 dpi with a pH-meter and a temperature probe. After 26 dpi, the dry weight of the roots was also measured.

Statistical Analysis
Statistical analyses on the in vitro test and the pathogenicity test were carried out by using IBM SPSS (Chicago, IL, USA) statistics version 25 for the normal distribution analysis using Shapiro-Wilk normality test, the homogeneity of variance using Levene's test and one-way analysis of variance using Tukey's test (p ≤ 0.05).  Table 1). The isolates collected in 2016 were all from 'Hayward' kiwifruit, except for one isolate from the cultivar 'Soreli'. The isolates collected in 2018 were from 'Hayward' plants, while those collected in 2019 were from different cultivars: twenty-three from 'Hayward', a green pulp variety, eight from 'Soreli', a yellow pulp variety, and two from 'Dong Hong', a red pulp variety.

Oomycete Isolation and Identification
Sequencing of the ITS1-5.8S-ITS2 region, the large subunit (LSU) rDNA, and cytochrome oxidase I gene (COI) were used for species assignation. The best-fit model used for the concatenated dataset was Tamura Nei + Gamma distribution. Based on concatenated phylogeny, 52 isolates were identified as P. vexans, 10 as P. litorale, seven as P. chamaehyphon, and two as P. helicoides (Figure 2). isolated in all the orchards while P. chamaehyphon and P. litorale were isolated only in orchards 1 and 2, respectively. In 2018, all isolates (12) were identified as P. vexans, whereas in 2019 the sampling in eight orchards (Table S1) yielded 22 strains: 11 identified as P. vexans, six as P. litorale, three as P. chamaehyphon, and 2 as P. helicoides. Only P. vexans was isolated from six orchards sampled in 2019. In orchard 12 both P. chamaehyphon and P. helicoides were isolated, whereas in orchard 16 both P. vexans and P. litorale were found.  A great intraspecific variability was observed. P. vexans strains were divided into three main groups: one with 39 strains, isolated in the three years of sampling, the second group with three strains, two isolated in 2016 and one in 2018, and 10 strains in the last group, all isolated in 2016 ( Figure 2, Table 1). P. chamaehyphon strains clustered into two groups: one with four strains isolated during 2016 and the second group with three strains isolated in 2019 ( Figure 2, Table 1). P. litorale strains were also divided into two groups: the first comprises three strains isolated in 2016 and one in 2019, and the second group has one strain isolated in 2016 and five strains isolated during 2019 ( Figure 2, Table 1).
In 2016, 10 orchards were sampled (Table S1) and enabled the isolation of 26 strains: 18 identified as P. vexans, four as P. litorale and four as P. chamaehyphon. P. vexans was isolated in all the orchards while P. chamaehyphon and P. litorale were isolated only in orchards 1 and 2, respectively. In 2018, all isolates (12) were identified as P. vexans, whereas in 2019 the sampling in eight orchards (Table S1) yielded 22 strains: 11 identified as P. vexans, six as P. litorale, three as P. chamaehyphon, and 2 as P. helicoides. Only P. vexans was isolated from six orchards sampled in 2019. In orchard 12 both P. chamaehyphon and P. helicoides were isolated, whereas in orchard 16 both P. vexans and P. litorale were found.

Morphological Observations
For the macro-morphology, two representative isolates per species were grown onto PDA, CMA and PCA Petri dishes and observed after 5 days. The same two strains were also grown on soil broth containing grass blades and their micro-morphology was observed under microscope after 7-14 days.
Phytopythium litorale colonies showed aerial mycelium and rosette mycelial pattern on CMA and PDA ( Figure 3B; mean diameter of 79 and 63 mm, respectively), whereas submerged mycelium and radiate chrysanthemum pattern on PCA ( Figure 3B; mean diameter: 79 mm). Hyphae were hyaline, 6 to 15.95 µm wide. Sporangia and oogonia were not produced. No differences between the two strains were observed.

Effect of Temperature and pH on In Vitro Growth
The optimal growth for P. vexans strains occurred at 25 • C on CMA and PCA, with an average radial growth of 21 and 29 mm/24 h, respectively (Figure 4a). Onto PDA, it was 25 • C for the strain R1A, with an average radial growth of 13 mm/24 h, whereas it was at 30 • C for the strain PPA, with an average radial growth of 17 mm/24 h (Figure 4a). The minimum and maximum growth temperature were 10 • C and 30 • C, respectively, on all media.
The optimal growth for P. litorale strains occurred at 30 • C on CMA, with an average radial growth of 20 mm/24 h (Figure 4b). Onto PDA, it was at 25 • C for the strain R3A, with an average radial growth of 12 mm/24 h, and at 30 • C for the strain P8G, with an average radial growth of 12 mm/24 h. Onto PCA, for both strains, it was at 25 • C with an average radial growth of 21 mm/24 h. The minimum and maximum growth temperature were 10 • C and 35 • C, respectively, on all media. without visible pattern on PDA ( Figure 3A; mean diameter: 46 mm), and a submerged mycelium and a slight radiate chrysanthemum pattern on PCA ( Figure 3A; mean diameter: 79 mm). Hyphae were hyaline, 7 to 15.19 µm wide. Sporangia were subglobose (20.17 ± 3.56 µm × 19.59 ± 3.26 µm) non-papillate. Oogonia were not produced. No differences between the two strains were observed. Phytopythium litorale colonies showed aerial mycelium and rosette mycelial pattern on CMA and PDA ( Figure 3B; mean diameter of 79 and 63 mm, respectively), whereas submerged mycelium and radiate chrysanthemum pattern on PCA ( Figure 3B; mean diameter: 79 mm). Hyphae were hyaline, 6 to 15.95 µm wide. Sporangia and oogonia were not produced. No differences between the two strains were observed.
Phytopythium chamaehyphon colonies showed aerial mycelium and a radiate chrysanthemum pattern on CMA and PDA ( Figure 3C; mean diameter: 79 mm), whereas The optimal growth for P. chamaehyphon strains occurred at 25 • C on CMA, PDA and PCA, with an average radial growth of 25, 23, and 23 mm/24 h, respectively (Figure 4c). The minimum and maximum growth temperature were 10 • C and 30 • C, respectively, on all media.
The optimal growth for P. helicoides strains occurred at 25 • C on CMA and PCA media with an average radial growth of 40 mm/24 h and 42 mm/24 h, respectively (Figure 4d). Onto PDA media was 30 • C, with an average radial growth of 23 mm/24 h. The minimum temperature for growth was 10 • C in all the media, whereas the maximum growth temperatures was 35 • C on PDA and 40 • C on CMA and PCA.
The four species tested were able to grow on PCA at pH from 5.0 to 8.0, with different growth rate ( Figure 5). Since no statistical different growth rate (p ≥ 0.05) was found for both strains of the same species, the values shown are the mean of the two strains. P. vexans showed the highest growth rate at pH 8.0 and pH 5.5 (p ≤ 0.05), whereas P. litorale at pH from pH 6.5 to 8.0 (p ≤ 0.05). For P. chamaehyphon, there was no statistical different growth rate from pH 5.5 to 8.0, whereas a lower growth rate was shown at pH 5.0 (p ≤ 0.05). The optimal pH was between pH 5.0 and 5.5 (p ≤ 0.05) for P. helicoides.

Effect of Temperature and pH on In Vitro Growth
The optimal growth for P. vexans strains occurred at 25°C on CMA and PCA, with an average radial growth of 21 and 29 mm/24 h, respectively (Figure 4a). Onto PDA, it was 25°C for the strain R1A, with an average radial growth of 13 mm/24 h, whereas it was at 30°C for the strain PPA, with an average radial growth of 17 mm/24 h (Figure 4a). The minimum and maximum growth temperature were 10°C and 30°C, respectively, on all media. The optimal growth for P. litorale strains occurred at 30°C on CMA, with an average radial growth of 20 mm/24 h (Figure 4b). Onto PDA, it was at 25°C for the strain R3A, with an average radial growth of 12 mm/24 h, and at 30°C for the strain P8G, with an average radial growth of 12 mm/24 h. Onto PCA, for both strains, it was at 25°C with an average radial growth of 21 mm/24 h. The minimum and maximum growth temperature were 10°C and 35°C, respectively, on all media.
The optimal growth for P. chamaehyphon strains occurred at 25°C on CMA, PDA and PCA, with an average radial growth of 25, 23, and 23 mm/24 h, respectively (Figure ). The minimum and maximum growth temperature were 10°C and 30 respectively, on all media. The optimal growth for P. helicoides strains occurred at 25°C on CMA and PC media with an average radial growth of 40 mm/24 h and 42 mm/24 h, respectiv (Figure 4d). Onto PDA media was 30°C, with an average radial growth of 23 mm/24 The minimum temperature for growth was 10°C in all the media, whereas maximum growth temperatures was 35°C on PDA and 40°C on CMA and PCA.
The four species tested were able to grow on PCA at pH from 5.0 to 8.0, w different growth rate ( Figure 5). Since no statistical different growth rate (p ≥ 0.05) w found for both strains of the same species, the values shown are the mean of the t strains. P. vexans showed the highest growth rate at pH 8.0 and pH 5.5 (p ≤ 0.0 whereas P. litorale at pH from pH 6.5 to 8.0 (p ≤ 0.05). For P. chamaehyphon, there was statistical different growth rate from pH 5.5 to 8.0, whereas a lower growth rate w shown at pH 5.0 (p ≤ 0.05). The optimal pH was between pH 5.0 and 5.5 (p ≤ 0.05) for helicoides.

Pathogenicity Test
All Phytopythium species under investigation were able to induce leaf curling, r rot, and decline of inoculated Actinidia plants. The first symptoms occurred after 12 da post-inoculation ( Table 2) in all of the inoculated plants, whereas negative contr remained symptomless. P. helicoides showed the highest disease index (3.67 ± 0.58) a no statistical differences were observed when compared to the positive control (infec soil). The other species showed a slow progression of symptoms (Table 2). After 26 d the species with the highest virulence remained P. helicoides (4.00 ± 0.00) compared to other species tested, and no statistical differences were observed when compared to positive control (Table 2). P. vexans and P. chamaehyphon showed similar virulence, w a mean disease index of 2.17 ± 0.58 and 2.67 ± 0.29, respectively. These species showe

Pathogenicity Test
All Phytopythium species under investigation were able to induce leaf curling, root rot, and decline of inoculated Actinidia plants. The first symptoms occurred after 12 days postinoculation ( Table 2) in all of the inoculated plants, whereas negative controls remained symptomless. P. helicoides showed the highest disease index (3.67 ± 0.58) and no statistical differences were observed when compared to the positive control (infected soil). The other species showed a slow progression of symptoms (Table 2). After 26 dpi, the species with the highest virulence remained P. helicoides (4.00 ± 0.00) compared to the other species tested, and no statistical differences were observed when compared to the positive control ( Table 2). P. vexans and P. chamaehyphon showed similar virulence, with a mean disease index of 2.17 ± 0.58 and 2.67 ± 0.29, respectively. These species showed a lower disease index and the disease progressed more slowly during the test. P. litorale showed to be more virulent (3.17 ± 0.76) compared to P. vexans, but no statistical differences were observed compared to the positive control (Table 2). To fulfil Koch's postulates, re-isolations were performed from the roots of all the plants and the isolates were identified as P. vexans, P. helicoides, P. chamaehyphon and P. litorale.
The soil pH at the beginning of the trial was 4.35 ± 0.22, whereas after 26 dpi it was 5.33 ± 0.22. The soil temperature was 23.74 ± 0.76 • C when the trial started and it was 22.06 ± 0.81 at the end of the trial. The highest dry weight was recorded from roots sampled from healthy plants (40.05 ± 8.32 g) compared to the other roots (Table 3). No statistical differences were observed between roots inoculated with P. vexans, P. litorale and P. chamaehyphon (p ≤ 0.05), whereas a statistically lower dry weight was observed for the roots of plants inoculated with P. helicoides. No statistical differences were observed when P. helicoides was compared to the positive control.

Discussion
Field surveys carried out in 18 A. chinensis orchards of north-western Italy during 2016-2019 permitted us to collect 71 isolates of Phytopythium spp. from kiwifruit plants, showing typical symptoms of KVDS, such as reduction in plant vigour, leaf curling, or complete decline. Isolation from roots of plants affected by KVDS is a difficult process, as it involves root surface disinfection, washing, air-drying, tissue sample taking from the affected area, plating on PDA and oomycete PARP media and, after 4 days, transplant into tubes with PARP medium. All of the isolates showed to belong to the oomycete genus Phytopythium.
The role of the microbial community in KVDS development was previously demonstrated. Donati and colleagues [11] underlined the role of the rhizosphere microbial com-munity, since a high incidence of Phytophthora spp. and Phytopythium spp. was associated with plants affected with KVDS. The role of biotic components in KVDS development was also described by Savian et al. [3], where in greenhouse experiments the symptoms were reproduced using unsterilised soil from a KVDS affected orchard, whereas no symptoms were observed using the same soil sterilised. The most frequently isolated species from kiwifruit affected by KVDS were Phytophthora citrophthora, P. cryptogea, P. infestans, P. megasperma, and Cylindrocarpon spp. [5,11]. Furthermore, Phytopythium spp. were isolated from symptomatic kiwifruits plants in several investigations [5,10,11,17].
In the present study, based on ML concatenated phylogeny on the ITS1-5.8S-ITS2 region of the rDNA, the large subunit (LSU) rDNA, and cytochrome oxidase I gene (COI), 52 strains were identified as P. vexans, 10 strains as P. litorale, seven strains as P. chamaehyphon and two strains as P. helicoides. Both, P. litorale and P. helicoides are reported for the first time as agents of KVDS on kiwifruit plants in Italy. Most of the isolates showed to belong to the species P. vexans, which is known to cause root and crown root in different crops, including kiwifruit [23,26,27]. Most orchards sampled showed the presence of only one Phytopythium species, except for some exceptions where two species were found.
In the pathogenicity tests, the virulence of all of the isolated species was demonstrated. Previously, P. vexans and P. chamaehyphon resulted pathogenic on 1-year-old or 6-month-old Actinidia chinensis var. deliciosa 'Hayward' plants [10,17]. Furthermore, the symptoms reproduction was also demonstrated for Desarmillaria tabescens and one isolate of Phytopythium spp. [11]. In this work, we demonstrated for the first time the pathogenicity of P. litorale on kiwifruit plants and P. helicoides was reported for the first time on kiwifruit plants in Italy. P. litorale, P. helicoides and P. vexans were previously reported as pathogenic on other hosts, such as on Platanus orientalis [37], Rhododendron pulchrum [38], on citrus, apple, and pear [39], and on almond [40]. Moreover, P. helicoides was already reported as agent of root and collar rot on kiwifruit in China [41].
In the pathogenicity test, flooding was used to reproduce soil water content proximal to field capacity. Waterlogging was previously investigated and, when applied alone, it was unable to reproduce KVDS symptoms. However, waterlogging is able to promote the progression of KVDS symptoms [3] and oomycetes were mainly reisolated from plants subjected to high irrigations volumes [11]. It should be noted that Phytopythium spp. live in water and soil and need a high humidity to produce sporangia and zoospores that are important infective propagules [42]. Therefore, the presence of water seems an important factor to promote Phytopythium spp. propagules and the onset of the disease.
In experiments in vitro, the four species showed an optimal growth at a temperature between 25 and 30 • C, depending on the media used. The maximum growth temperature was 30 • C for P. vexans, P. chamaehyphon, and P. litorale. Only P. helicoides was able to grow at 35 • C and even at 40 • C has a slow growth, confirming that the species is tolerant to high temperatures [43]. The data confirmed what reported in literature for P. litorale [33,44]. In other papers, where different strains were tested, 35 • C was reported as the maximum growth limit for P. vexans and 38 • C was reported as the optimal temperature for P. helicoides [31,45]. All of the tested strains showed an optimal growth at temperatures, which are in accordance with what has been observed in the soil of the orchards. In a monitoring performed in kiwifruit orchard during 2019, the average soil temperature measured during summer, when the KVDS symptoms occurred, was 23.2 ± 1.3 • C, whereas the average air temperature was 23.4 ± 4.5 • C.
The test to evaluate the tolerance of the species to different pH in vitro, showed their ability to grow at all the tested pH, ranging from 5.0 to 8.0. The maximum growth (p ≤ 0.05) was observed at pH 8.0 for P. vexans, at pH 6.5, 7.5 and 8.0 for P. litorale and at 5.0 and 5.5 for P. helicoides, whereas P. chamaehyphon seems the least influenced species by pH, where the highest growth was recorded in the range from 5.5 to 8.0. The average radial growth rate was similar for the species P. vexans, P. litorale and P. chamaehyphon, whereas P. helicoides showed the highest radial growth. For all the species under investigation, results showed that pH had less effect on growth than temperature, as previously shown by Cantrell and Dowler [45] for P. vexans. It should be noted that the effects of pH on fungal growth are complex, as an initial pH may affect growth, and subsequently growth can affect pH through release of metabolites into the growing medium. The values of pH and temperatures of the substrates recorded during pathogenicity tests were around the optimal values for the growth of the species of Phytopythium tested.

Conclusions
The present study demonstrates the strong role of oomycetes in the establishment and development of KVDS. The presence of different species of oomycetes suggests that the oomycete component of the soil microbiota present in the soil is involved in the development of KVDS, and not only a single species is involved in this complex syndrome. The isolated species have a relatively high optimal and maximum temperature for growth in soil, and they could be favoured by the average increase in soil temperatures during summer, associated with global warming. Waterlogging could exert a double effect, both on stressing the plant with a temporary anoxia and on favouring the release of infective zoospores by oomycetes. Further studies should investigate the complex interactions between kiwifruit, the oomycete species, the soil environment, and the effect of different management strategies in the field. Moreover, a study of the soil and rhizosphere microbiome could help to clarify the changes occurring in the soil microbiota of kiwifruit orchards.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/microorganisms11010216/s1. Table S1: Strain name and information about orchards from where the strains were isolated (geographical location, geographical coordinates and orchard number); Table S2: List of species, strain designation and accession numbers for ITS and LSU regions and COI gene used for the phylogeny of Phytopythium spp. isolated from kiwifruit roots in this study.