Host Range of Hymenoscyphus fraxineus in Slovak Arboreta

The health of 34 different Fraxinus taxa in association with the pathogenic fungus Hymenoscyphus fraxineus was assessed in four Slovak arboreta. Averaged across all arboreta, nearly one-quarter (24.9%) of all evaluated trees showed ash dieback symptoms. The damage was most serious on the common ash F. excelsior, a native species. The percentage of dead trees did not exceed 2% for all evaluated trees. Generally, ash trees of all ages were affected, though the intensity of the damage varied among the sites. The identity of H. fraxineus was confirmed by conventional PCR targeting the internal transcribed spacer (ITS) sequences of the nuclear ribosomal DNA, as well as the 18S gene/ITS-2 region of the rDNA operon. In Slovakia, the pathogen has expanded its host range from native species not only to their ornamental cultivars, but also to introduced North American (F. cinerea, F. latifolia, F. pennsylvanica, F. quadrangulata) and Asian (F. bungeana, F. chinensis ssp. rhynchophylla, F. mandshurica) ash species. H. fraxineus was also observed on the previous year’s leaf petioles of the native European species F. ornus, considered a weakly susceptible host. In Slovak arboreta, H. fraxineus was found on 23 Fraxinus taxa; 21 of them represent first records for the country. F. bungeana is recorded as a new host species of H. fraxineus.


Introduction
Fraxinus L. (Oleaceae) is widely distributed over North and Central America, Asia, Europe, and Northwest Africa, from tropical to temperate climates [1]. Although Fraxinus often grows in forests, at their edges, and along tree-lined roads, it is also planted in urban greenery and parks as an ornamental tree. Since the early 1990s, ash dieback is a serious disease that has been killing ash across Europe [2][3][4][5]. The disease is caused by the invasive fungus Hymenoscyphus fraxineus (T. Kowalski) Baral, Queloz & Hosoya (syn. Hymenoscyphus pseudoalbidus Queloz, Grünig, Berndt, T. Kowalski, T.N. Sieber & Holdenr.), the anamorph of which is Chalara fraxinea T. Kowalski. Currently, it is one of seven known species of the genus Hymenoscyphus Gray (Helotiales) occurring on ash trees. The non-pathogenic species Hymenoscyphus albidus (Gillet) W. Phillips has been known from Europe since 1851 [6]. Gross and Han [7] and Gross et al. [8] [8], there is currently no information on the pathogenicity of H. koreanus and H. occultus. Recently, Kowalski and Bilanski [9] have identified another species, H. pusillus T. Kowalski & P. Bilański, from F. pennsylvanica Marshall in Poland, but the pathogenicity of this fungus identified another species, H. pusillus T. Kowalski & P. Bilański, from F. pennsylvanica Marshall in Poland, but the pathogenicity of this fungus has not yet been confirmed. H. caudatus (P. Karst.) Dennis, occurring on various plant species, has also been recorded on ash in China [10] and the British Isles [11].
H. fraxineus is widespread in both East Asia (Northeastern China, Japan, Korea, and the Russian Far East) and Europe [12]. In Europe, this pathogen mainly attacks common ash (F. excelsior L.) and narrow-leaved ash (F. angustifolia Vahl), but it is also recorded on North American (F. americana L., F. nigra Marshall, F. pennsylvanica) and Asian (F. mandshurica Rupr., F. sogdiana Bunge) ash species in Estonia [13,14]. H. fraxineus has been recorded from F. ornus L. only in Austria [15]. In Danish arboretum, H. fraxineus apothecia were equally well-formed on many introduced host Fraxinus species as it was on common ash [16]. Most recently, the fungus has been found on three new oleaceous host plant species (Chionanthus virginicus L., Phillyrea angustifolia L., Phillyrea latifolia L.) at Westonbirt Arboretum in the United Kingdom [17]. These were the first findings on hosts other than Fraxinus. In Slovakia, two native European ash species (F. excelsior, F. angustifolia) are attacked by H. fraxineus [18][19][20][21]. Previous papers have only recorded ash dieback on these two native Fraxinus species in forest stands, seed orchards, and tree alleys in Slovakia. The health of non-native ash trees and ornamental cultivars of native species, however, has not yet been investigated in Slovakia.
The aims of this study were (i) to evaluate the health of introduced ash species and ornamental cultivars of native Fraxinus species in terms of the occurrence of H. fraxineus in Slovak arboreta and (ii) to determine the host range of H. fraxineus in the arboreta.

Study Sites
Study material was collected in three arboreta situated in central Slovakia and one arboretum in Southwestern Slovakia ( Figure 1, Table 1). The Borová Hora Arboretum (Zvolen District) is spread over the southwestern slopes of Zvolenská vrchovina upland, which is one of the geomorphological units of the Zvolenská kotlina basin; it is characteristically upland. From a climatic point of view, this area is warm and slightly humid ( Figure  2), with a cold winter [22].
The Kysihýbel Arboretum (Banská Štiavnica District) is located near the town of Banská Štiavnica. The area is protected by the Štiavnické vrchy Mountains Protected Landscape Area. This forestry arboretum is a unique experimental forestry area specialized in rapidly growing exotic wood species [23]. It is in a warm temperate zone with a humid climate (Figure 2), without extreme temperature changes.  The Borová Hora Arboretum (Zvolen District) is spread over the southwestern slopes of Zvolenská vrchovina upland, which is one of the geomorphological units of the Zvolenská kotlina basin; it is characteristically upland. From a climatic point of view, this area is warm and slightly humid (Figure 2), with a cold winter [22].
The Kysihýbel Arboretum (Banská Štiavnica District) is located near the town of Banská Štiavnica. The area is protected by the Štiavnické vrchy Mountains Protected Landscape Area. This forestry arboretum is a unique experimental forestry area specialized in rapidly growing exotic wood species [23]. It is in a warm temperate zone with a humid climate (Figure 2), without extreme temperature changes.
The Liptovský Hrádok Arboretum (Liptovský Mikuláš District) is situated in an urban area in the town of Liptovský Hrádok. This area has warm, humid summers and cold winters. The Liptovský Hrádok Arboretum is the highest elevation arboretum in Central Europe and is registered as a protected area under the Nature and Landscape Protection Act of Parliament [24].
The Mlyňany Arboretum (Zlaté Moravce District) is a large park situated in one of the warmest and driest areas of Slovakia in the southern foothills of the Carpathian Mountains at the northern edge of the hottest part of Slovakia. The area is characterized by prolonged dry periods during the height of summer; during some years, there is sometimes no summer rain at all. The Mlyňany Arboretum manages the largest collection of non-native trees and shrubs in Slovakia, which also happens to be one of the richest collections in Central Europe [25]. The Liptovský Hrádok Arboretum (Liptovský Mikuláš District) is situated in an urban area in the town of Liptovský Hrádok. This area has warm, humid summers and cold winters. The Liptovský Hrádok Arboretum is the highest elevation arboretum in Central Europe and is registered as a protected area under the Nature and Landscape Protection Act of Parliament [24].
The Mlyňany Arboretum (Zlaté Moravce District) is a large park situated in one of the warmest and driest areas of Slovakia in the southern foothills of the Carpathian Mountains at the northern edge of the hottest part of Slovakia. The area is characterized by prolonged dry periods during the height of summer; during some years, there is sometimes no summer rain at all. The Mlyňany Arboretum manages the largest collection of non-native trees and shrubs in Slovakia, which also happens to be one of the richest collections in Central Europe [25].

Ash Dieback Damage Assessment in Study Sites
Ash trees in the arboreta were assessed for ash dieback symptoms (necrotic leaf spots, brownish to orange colored bark necrosis and cankers on stems and branches, shoot dieback, apothecia on the previous year's leaf petioles lying in the litter under the tree) during the growing season (July-August) in Mlyňany in 2015 and in Borová hora, Kysihýbel, and Liptovský Hrádok in 2018. The extreme droughts caused by precipitation deficits in Mlyňany in 2015 were also common during the four years before the assessment. Two years prior to the assessments in Borová hora, Kysihýbel, and Liptovský Hrádok in 2018, the annual precipitation was higher and the average annual temperature was lower than in the year of the assessment. The dieback damage assessments were made before leaf fall and covered a total of 470 trees (20-118 years old) of 34 taxa. Ash dieback crown damage was assessed visually by the same person. The extent of tree crown defoliation was reported as the percentage of defoliation [27].

Ash Dieback Damage Assessment in Study Sites
Ash trees in the arboreta were assessed for ash dieback symptoms (necrotic leaf spots, brownish to orange colored bark necrosis and cankers on stems and branches, shoot dieback, apothecia on the previous year's leaf petioles lying in the litter under the tree) during the growing season (July-August) in Mlyňany in 2015 and in Borová hora, Kysihýbel, and Liptovský Hrádok in 2018. The extreme droughts caused by precipitation deficits in Mlyňany in 2015 were also common during the four years before the assessment. Two years prior to the assessments in Borová hora, Kysihýbel, and Liptovský Hrádok in 2018, the annual precipitation was higher and the average annual temperature was lower than in the year of the assessment. The dieback damage assessments were made before leaf fall and covered a total of 470 trees (20-118 years old) of 34 taxa. Ash dieback crown damage was assessed visually by the same person. The extent of tree crown defoliation was reported as the percentage of defoliation [27].  were placed in plastic tubes and brought to the laboratory for molecular analysis to confirm species identity. In arboreta, individual ash species are spatially separated from each other. Particular ash species grow either solitarily or in monoculture groups surrounded by non-ash tree species, thus minimizing the possibility that the leaf litter under a tree would contain leaves of another ash species.

DNA Extraction, PCR Amplification, and Sequencing
DNA was extracted from the samples by grinding apothecia in liquid nitrogen with a disposable polypropylene pellet pestle directly in 1.5 mL microtubes. DNA extractions were performed using the EZ-10 Spin Column Fungal Genomic DNA Mini-Preps Kit (Bio Basic Inc., Markham, ON, Canada) according to the protocol of the manufacturer. DNA was resuspended in 50 µL of elution buffer and stored at −20 • C.
The internal transcribed spacer (ITS) region of ribosomal DNA (ITS1-5.8S-ITS2) was amplified with the universal primer pair ITS1F/ ITS4 [28,29]. For identification of H. fraxineus, species-specific primers targeting the 18S gene and the ITS-2 region of rDNA operon [30] were used in conventional PCR.
All PCRs were performed in a total volume of 20 µL. The PCR mixtures consisted of 4 µL 5× HOT FIREpol ® BlendMaster Mix (Solis BioDyne, Tartu, Estonia), 0.4 µL of each primer (final concentration 0.2 µM), deionized water of molecular grade (Pro injection, B. Braun), and~10 ng of template DNA. Every reaction was performed with a negative control containing the master mix without template DNA. The PCRs were carried out on a T100™ Thermal Cycler (Biorad, Applied Biosystems, Waltham, MA, USA). For the ITS region, the PCR cycling conditions included an initial denaturation for 14 min at 95 • C followed by 35 cycles for 15 s at 95 • C, annealing for 30 s at 50 • C, elongation for 80 s at 72 • C, and a final elongation step for 10 min at 72 • C. PCR conditions for the specific primer 18S gene/ITS2 region included an initial denaturation step for 15 min at 95 • C, followed by 35 cycles for 20 s at 95 • C, annealing for 30 s at 62 • C, elongation for 1 min at 72 • C, and a final elongation step for 8 min at 72 • C. All PCR products were visualized by electrophoresis on 0.8% (w/v) TBE agarose gels stained with SimplySafe stain (EURx, Gdansk, Poland).
The target PCR fragments were purified using a QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sequenced by the Sanger method using an ABI 3130xl sequencer (Applied Biosystems) at SEQme Ltd. (Dobříš, Czech Republic). The obtained DNA sequences were compared by BLAST (Basic Local Alignment Search Tool, available at https://blast.ncbi.nlm.nih.gov/Blast.cgi) against ITS sequences in the NCBI GenBank Sequence Database.

Phylogenetic Analysis
The ITS sequences obtained were trimmed and edited using Molecular Evolutionary Genetics Analysis software Version 7.0 (MEGA7) [31]. Final datasets were aligned with the BioEdit Sequence Alignment Editor Version 7.2.5 [32]. To demonstrate differences in sequences of the studied material of H. fraxineus from Slovakia, we performed a phylogenetic analysis using corresponding GenBank sequence deposits of several known species of the Hymenoscyphus genus on Fraxinus as references, i.e., H. albidus, H. fraxineus, H. occultus, H. pusillus, H. koreanus, H. linearis, and H. caudatus. Hymenoscyphus menthae (W. Phillips) Baral was used as the outgroup. All samples used in our analysis are represented by ITS sequence. The ITS dataset was analyzed using the Maximum Likelihood (ML) method based on the Tamura-Nei model [33]. The phylogenetic tree was reconstructed using MEGA7.

Statistical Analysis
The data on crown defoliation of different Fraxinus taxa inspected in four Slovak arboreta were tested with one-way analysis of variance (ANOVA). Taxa with low sample size (n < 2) were excluded from the test. Significance of the differences in defoliation means among arboreta was evaluated by Duncan test. The statistical analysis was performed using STATISTICA software Version 10 (StatSoft Inc., Tulsa, OK, USA, 2011).

Defoliation Level of Ash Trees in Slovak Arboreta
In Borová hora Arboretum, 28 trees of 12 Fraxinus taxa were evaluated ( Table 2). The percentage of defoliation varied among taxa. Out of six F. angustifolia trees, three showed less than 10% defoliation, one was slightly defoliated (up to 25%), and two were moderately defoliated (30% and 40%). A single specimen of F. angustifolia ssp. pannonica Soó & T. Simon was heavily defoliated (85%). Variation in the level of defoliation was observed also among cultivars of F. excelsior. The 'Aurea Pendula', 'Diversifolia' (one out of three trees), 'Heterophylla Pendula', and 'Nana' cultivars showed less than 20% defoliation, and two 'Diversifolia', 'Pendula' and 'Pruhoniciana' were heavily defoliated (60%-95%). F. ornus trees showed no defoliation. Three trees of F. mandshurica, the only Asian species growing in the arboretum, were moderately defoliated (40%). Three trees of the North-American species, F. pennsylvanica, showed defoliation (one had up to 10% defoliation, two showed moderate defoliation (30% and 50%)); two trees of 'Aucubifolia' cultivar had a defoliation level of 20%. No ash tree had the crown completely defoliated. Average percent defoliation of ash trees in Borová hora was 31.8%. All Fraxinus taxa, except F. ornus, showed ash dieback symptoms.
In Kysihýbel Arboretum, 54 trees of 9 taxa were inspected ( Table 2). Out of 25 F. excelsior trees, two showed slight defoliation (up to 20%), 12 had moderate defoliation (30%-60%), nine were heavily defoliated (70%-95%), and two were completely defoliated. It should be noted that several of the dying F. excelsior trees had also been heavily attacked by the ash bark beetle, Hylesinus sp. (Curculionidae, Coleoptera). All trees of F. ornus showed up to 10% defoliation. Of the assessed Asian species, F. bungeana A. DC. and F. chinensis ssp. rhynchophylla (Hance) A.E. Murray showed up to 5% defoliation. Out of four F. mandshurica trees, three had up to 10% defoliation and one showed 20% defoliation. The North American species F. americana, F. cinerea Bosc, and F. latifolia Benth. were slightly defoliated (15%-30%). All trees of F. pennsylvanica had up to 30% defoliation. Average percent defoliation of ash trees in Kysihýbel was 35.7%.
A total of 470 trees of 34 taxa were evaluated. In all arboreta, 69.6% of evaluated trees showed no, or less than 10%, defoliation. The mean value of defoliation reached 14.5%. Dead trees accounted for 1.7% of all assessed trees. Figure 3 shows the estimated crown defoliation of ash trees in Mlyňany in 2015 and in Borová hora, Liptovský Hrádok, and Kysihýbel in 2018.   Defoliation percentage or range of defoliation percentage (if more than 2 trees of the same taxon on the same site showed different levels of defoliation) of individual Fraxinus taxa in each arboretum is given in Table 2. Severe to extreme defoliation was common in F. excelsior and most of its cultivars and rarely in F. mandshurica, F. pennsylvanica, and two subspecies of F. angustifolia (Figure 4).  Defoliation percentage or range of defoliation percentage (if more than 2 trees of the same taxon on the same site showed different levels of defoliation) of individual Fraxinus taxa in each arboretum is given in Table 2. Severe to extreme defoliation was common in F. excelsior and most of its cultivars and rarely in F. mandshurica, F. pennsylvanica, and two subspecies of F. angustifolia (Figure 4).   Only three species of ash, F. excelsior, F. mandshurica, and F. pennsylvanica, were present in all Slovak arboreta. Percentage of defoliation of F. excelsior (including taxa of lower rank) varied significantly among arboreta (F (3, 64) = 3.35, p = 0.024; Figure 5). The crown damage of F. excelsior trees in Liptovský Hrádok was significantly lower than in Mlyňany and Kysihýbel. There was also a significant difference in the percentage of defoliation of F. pennsylvanica among arboreta (F (3, 63) = 20.93, p < 0.001; Figure 5). The highest defoliation was observed in Liptovský Hrádok, the lowest in Mlyňany. The average defoliation showed a similar level in Borová hora and Kysihýbel, but it was significantly higher than in Mlyňany and lower than in Liptovský Hrádok. Although F. mandshurica grew in all arboreta, we did not include it in the analysis due to the low sample size in Liptovský Hrádok.  Only three species of ash, F. excelsior, F. mandshurica, and F. pennsylvanica, were present in all Slovak arboreta. Percentage of defoliation of F. excelsior (including taxa of lower rank) varied significantly among arboreta (F(3, 64) = 3.35, p = 0.024; Figure 5). The crown damage of F. excelsior trees in Liptovský Hrádok was significantly lower than in Mlyňany and Kysihýbel. There was also a significant difference in the percentage of defoliation of F. pennsylvanica among arboreta (F(3, 63) = 20.93, p < 0.001; Figure 5). The highest defoliation was observed in Liptovský Hrádok, the lowest in Mlyňany. The average defoliation showed a similar level in Borová hora and Kysihýbel, but it was significantly higher than in Mlyňany and lower than in Liptovský Hrádok. Although F. mandshurica grew in all arboreta, we did not include it in the analysis due to the low sample size in Liptovský Hrádok.   Only three species of ash, F. excelsior, F. mandshurica, and F. pennsylvanica, were present in all Slovak arboreta. Percentage of defoliation of F. excelsior (including taxa of lower rank) varied significantly among arboreta (F(3, 64) = 3.35, p = 0.024; Figure 5). The crown damage of F. excelsior trees in Liptovský Hrádok was significantly lower than in Mlyňany and Kysihýbel. There was also a significant difference in the percentage of defoliation of F. pennsylvanica among arboreta (F(3, 63) = 20.93, p < 0.001; Figure 5). The highest defoliation was observed in Liptovský Hrádok, the lowest in Mlyňany. The average defoliation showed a similar level in Borová hora and Kysihýbel, but it was significantly higher than in Mlyňany and lower than in Liptovský Hrádok. Although F. mandshurica grew in all arboreta, we did not include it in the analysis due to the low sample size in Liptovský Hrádok.

Sequencing Analysis
DNA was extracted from 64 apothecia samples. The amplicons of ITS region using the ITS1F/ITS4 primers ranged from 700 to 875 bp. The obtained ITS1-5.8S-ITS2 rDNA sequences exhibited between 99.07% (GenBank accession no. KY593990) to 100% (GenBank accession no. MN833901) similarity to previously deposited sequences of H. fraxineus in GenBank. The obtained ITS sequences (a single sequence per Fraxinus taxon and per site) were deposited in the NCBI GenBank; accession numbers are given in Table 2.
In 57 of the 64 samples, the presence of H. fraxineus was detected by species-specific PCR, which produced an amplicon of size 456 bp and showed positive results for all samples except three samples from F. pennsylvanica v. subintegerrima, three from F. pennsylvanica v. lanceolata, and one from F. americana that were collected in Mlyňany ( Table 2).

Phylogenetic Analysis
The ITS dataset contained 65 sequences representing eight Hymenoscyphus species. Among them, 32 sequences were newly obtained and 33 were retrieved from GenBank. In the phylogenetic analysis, Hymenoscyphus menthae from Solidago canadensis L. was chosen as the outgroup.
Phylogenetic analysis of 32 samples of H. fraxineus from different Fraxinus species collected in Slovak arboreta showed high sequence similarity with all reference sequences of H. fraxineus from GenBank. They all grouped together in one strongly supported (81% bootstrap support) clade with H. fraxineus from different Fraxinus species collected in other European localities and Asia ( Figure 6).
The obtained ITS sequences of Slovak samples were identical irrespective of the host and concurred with those of European samples but differed from Asian samples at some positions. All Slovak ITS sequences differed from the three ITS sequences (KF188725, KJ511194, KP068051) from Japan, Korea, and China at two nucleotide positions, 433 (a gap instead of G) and 474 (T instead of C), and from one ITS sequence (KU323580) from Russia at three positions, 424 (a gap instead of G), 465 (T instead of C), and 679 (C instead of T).

Host Range of H. fraxineus in Slovak Arboreta
In four Slovak arboreta, the identity of the fungus H. fraxineus was molecularly confirmed in 23 out of 34 evaluated Fraxinus taxa. Ash dieback was recorded in 11 out of 12 taxa in Borová hora, in 6 out of 9 taxa in Kysihýbel, in 6 out of 7 taxa in Liptovský Hrádok, and in 9 out of 23 taxa in Mlyňany. The numbers of ash trees (at species level), of both symptomatic and asymptomatic individuals, in each of the examined arboreta are given in Table 3.
In total, 64.3% of the trees of native ash species showed symptoms of ash dieback in the arboreta. The fungus was confirmed on the common ash F. excelsior as well as its eight cultivars in all evaluated arboreta; on F. angustifolia and its two subspecies and one variety in Borová hora and Mlyňany; and on F. ornus in Mlyňany.
Analysis of samples from Asian Fraxinus species confirmed the presence of H. fraxineus on F. mandshurica in all arboreta, on F. chinensis ssp. rhynchophylla in Kysihýbel, and on F. bungeana in Liptovský Hrádok. Four Asian species, F. longicuspis, F. paxiana, F. pubinervis, and F. sogdiana, were not attacked by the pathogen. Less than 6% of the trees of Asian species showed ash dieback symptoms in all examined arboreta.
The causal agent of ash dieback was recorded on four North American ash species: F. cinerea and F. latifolia in Kysihýbel; on F. pennsylvanica and its one cultivar and one variety in Borová hora, Kysihýbel, and Liptovský Hrádok; and on F. quadrangulata in Mlyňany. Although F. velutina showed slight defoliation, no other symptoms of ash dieback or production of apothecia were observed. F. americana was moderately defoliated in Mlyňany but without clear symptoms and with a negative result for a single sample from this host tested using H. fraxineus-specific PCR primers. In Slovak arboreta, 25% of the trees of North American species were attacked by H. fraxineus. Forests 2020, 11, x FOR PEER REVIEW 10 of 18   Abbreviations: BH-Borová hora, K-Kysihýbel, LH-Liptovský Hrádok, M-Mlyňany. * including taxa of lower rank.

Host Range of H. fraxineus in Slovak Arboreta
The North African ash tree species F. xanthoxyloides v. dimorpha, an endangered species listed in the IUCN Red List of Threatened Species 2017, had no symptoms of ash dieback in Mlyňany.
The highest number of assessed trees was in Mlyňany Arboretum, and the number of trees of introduced ash taxa greatly exceeded the native species in this locality. In Mlyňany, 9.5% of evaluated trees showed ash dieback symptoms in 2015. Most of the asymptomatic trees in Mlyňany belong to introduced species. The percentage of symptomatic ash trees in Kysihýbel, Borová hora, and Liptovský Hrádok ranged from 70% to 95% in 2018. The native species predominated in these three arboreta with lower total numbers of ash trees.
Ash dieback symptoms were recorded in ash trees of all ages (Table 2). Together in all arboreta, 24.9% of all evaluated trees showed ash dieback symptoms (  [35] on the basis of ITS rDNA sequences. None of the samples examined in the present study were identified as H. albidus. Hymenoscyphus pusillus, recently discovered and described from F. pennsylvanica in Poland [9], has not yet been recorded in Slovakia. The pathogenic H. fraxineus, the cause of massive mortality of ash trees in forest stands throughout Europe, mainly attacks the native ash species F. excelsior and F. angustifolia [2,36,37]. In Slovakia, the invasive fungus H. fraxineus causes much damage to both F. excelsior and F. angustifolia in forest stands and seed orchards [19][20][21]. The third native European species, F. ornus, seemed to be tolerant to ash dieback until recently. However, Kirisits and Schwanda [15] recently isolated H. fraxineus from leaf rachises of F. ornus seedlings in Austria. We also recorded F. ornus as a host of the pathogen in the Mlyňany Arboretum.
The present study of introduced ash species in Slovak arboreta confirmed that the pathogen has a much larger host range in Slovakia. H. fraxineus is considered a weak pathogen of the Asian species F. mandshurica in Japan [38,39], China [40], the Russian Far East [10,41], and South Korea [42]. However, F. mandshurica trees were moderately, and occasionally badly, affected with symptoms of dieback of shoots and twigs in the evaluated Slovak arboreta. Drenkhan and Hanso [13] also recorded F. mandshurica trees affected with ash dieback in Estonian parks. Cleary et al. [43] confirmed the presence of H. fraxineus in asymptomatic leaves of F. mandshurica from Eastern Russia.
In South Korea [42] and in the Russian Far East [10], apothecia of H. fraxineus have been recorded on fallen leaves of F. chinensis ssp. rhynchophylla. Nielsen et al. [16] did not detect H. fraxineus DNA in the leaves of F. chinensis ssp. rhynchophylla after intensive spore exposure, and no bark lesions were developed after direct stem inoculation. As a result, they suggested that F. chinensis ssp. rhynchophylla may be less susceptible to leaf infection by H. fraxineus than F. mandshurica. When exposed to natural infection in a Danish arboretum, some of the additional Asian Fraxinus species examined were also asymptomatic or showed only a few dead shoots. In the present study, F. chinensis ssp. rhynchophylla trees in Kysihýbel Arboretum showed slight symptoms, including minor shoot dieback and bark necrosis.
In 2015, Drenkhan et al. [14] recorded F. sogdiana, a Central Asian ash species, as a new host of H. fraxineus in Estonia, indicating that some additional Asian ash species planted in Europe can serve as hosts for this invasive pathogen. Although H. fraxineus has not yet been recorded in North America, some North American Fraxinus species (F. albicans Buckley, F. americana, F. latifolia, F. nigra, F. pennsylvanica, F. profunda (Bush) Bush, F. quadrangulata, F. velutina) seem to be susceptible to shoot infection [13,16,44,45]. Fraxinus pennsylvanica naturally infected by the ash dieback pathogen was recorded in Estonia and Austria [13,46]. The results from Slovak arboreta confirmed F. cinerea, F. latifolia, F. pennsylvanica, F. pennsylvanica 'Aucubifolia', F. pennsylvanica v. lanceolata, and F. quadrangulata as hosts of H. fraxineus. On the other hand, F. americana and F. velutina planted in Slovak arboreta were not infected by the pathogen. Hymenoscyphus fraxineus has already been reported on these two host species from European parks and arboreta [13,16]. Although several F. pennsylvanica v. lanceolata and v. subintegerrima trees in Mlyňany Arboretum showed severe defoliation, the presence of H. fraxineus was not confirmed by our molecular work. Hymenoscyphus-like apothecia collected under these trees will be subjected to further analysis.
Four ash species, F. longicuspis, F. paxiana, F. pubinervis, and F. xanthoxyloides v. dimorpha, not yet recorded as hosts of H. fraxineus, showed no symptoms of dieback in Slovak arboreta. Nielsen et al. [16] assumed that failure of H. fraxineus to infect two Asian ash species, F. sieboldiana Blume and F. paxiana, may reflect cases of non-host resistance. We recommend, however, that monitoring and analysis of all asymptomatic exotic ash trees be continued; their resistance to ash dieback should not be assumed.
The phylogram ( Figure 6) shows that ITS sequences of all Slovak samples of H. fraxineus are grouped in a separate clade together with 16 sequences from other European samples retrieved from the GenBank database. The Slovak ITS sequences concur with sequences of European samples and form a subgroup within a H. fraxineus clade with a strong nodal support (ML = 82). The phylogram also demonstrates an intraspecific variation in H. fraxineus. The four GenBank sequences of Asian strains are clearly separated from the Slovak and all European samples due to differences at two nucleotide positions for strains from F. mandshurica (KF188725, KJ511194, KP068051) and at the other three positions for a strain from F. chinensis ssp. rhynchophylla (KU323580). These observations are in accordance with several recent studies and confirm that the ITS region of H. fraxineus from Asian populations (from Japan, China, and Korea) as well as samples from the Russian Far East are genetically divergent from European population of H. fraxineus [10,39,40,42].
The artificial inoculations as well as field observations of recent studies have demonstrated that Asian ash species tended to be immune to ash dieback disease compared to European ash species [10,13,43,47]. Field observations of ash dieback symptoms in this study also suggest that H. fraxineus is more pathogenic to European ash species (e.g., F. excelsior and F. angustifolia) than to Asian ash species (F. mandshurica and F. chinensis). It is important to note that no pathogenicity tests were carried out in this study and the above findings are solely based on the field observations of symptoms.
There are clear differences in the progression of ash dieback in trees of different age, irrespective of stand disease history, but there are also differences among stands in different regions, depending on their disease histories [48]. Trees are subject to dieback at all ages [3]; our results confirm this. Although semi-mature trees (20-60 years old) of introduced ash species growing in arboreta situated in temperate and humid areas displayed ash dieback symptoms, they were not infected (or only rarely infected) in the warmest and driest arboretum (Mlyňany). It has been shown that a high number of the youngest and intermediate-aged future crop trees in forest stands in Norway were dead, while development of the disease in large, dominant trees was slower [48]. The higher mortality of young trees and nursery saplings causes immense problems for establishing and tending young stands [49]. Due to the different age structure of individual Fraxinus taxa in Slovak arboreta, it was not possible to evaluate the influence of tree age on susceptibility to infection. However, the influence of tree age is considered a secondary factor [50]. We should note that the number of dead trees can be affected by regular tree maintenance (removal of dead and/or dying trees) in the arboreta. For example, during our investigations, no ash tree had its crown completely defoliated in Liptovský Hrádok Arboretum-however, seven dead trees (5 F. excelsior, 1 F. excelsior 'Pendula', 1 F. pennsylvanica 'Aucubifolia') were felled 1-2 years before our assessment [51]. In addition, from 2012 to 2019, 31 dead ash trees (26 F. excelsior cultivars, 4 F. ornus, 1 F. angustifolia) were removed in Borová hora Arboretum. In Mlyňany Arboretum, 9 dead trees of F. excelsior and 1 F. mandshurica were felled from 2019-2020. The cause of the death of ash trees that were felled before our survey was not determined. From the time of our survey to 2019, no ash trees were felled due to H. fraxineus attack in Kysihýbel Arboretum [52].
Climate and site conditions have a fundamental influence on H. fraxineus sporulation, spore release and dispersal, as well as on infection rate throughout the growing season [53][54][55][56][57]. The dynamics of apothecia production varied depending on weather conditions during the years of our survey. Longer periods with low rainfall adversely affected the fructification of the pathogen. Thus, apothecia were not found under some trees with ash dieback symptoms in Mlyňany in 2015. A lower infection rate (9.5%) was recorded in Mlyňany Arboretum, while the proportion of symptomatic ash trees in Kysihýbel, Borová hora, and Liptovský Hrádok ranged from 70% to 95%. We assume that the higher percentage of infected trees in the three arboreta from central Slovakia was influenced by more favorable microclimatic conditions at the localities. Borová hora, Kysihýbel, and Liptovský Hrádok experience lower average temperatures and higher amounts of precipitation compared to Mlyňany Arboretum, which is situated in one of the warmest and driest areas of Slovakia. Likewise, Ogris [58], Kessler et al. [49], Hauptman et al. [59], and Gross et al. [60] recorded higher severity of ash dieback on sites with high relative air humidity and lower temperatures. The differences in the infection rates of ashes among Slovak arboreta can be caused not only by environmental factors (e.g., tree competition, shading, microclimate, etc.) or crown and canopy management, but also by ash species composition and tree size. The lesser amount of damage found in Mlyňany might be due to the fact that the investigation in Mlyňany took place in 2015, whereas the other three arboreta were surveyed in 2018. The disease progression in these arboreta will be further monitored.
Insect pests and other plant pathogenic fungi represent a noticeable threat to stressed and weakened ash trees and may contribute to the progress of ash dieback disease in ash stands. Numerous fungal taxa on ash stems and twigs in the initial and advanced stages of dieback were recorded by Bakys et al.
[3], Kowalski et al. [61], and Pastirčáková et al. [62]. The development and severity of ash dieback may also be correlated with soil moisture in stands and with Armillaria root infections [48,63]. In Slovak seed orchards, Longauerová et al. [20] observed frequent occurrence of Armillaria cepistipes Velen. and Armillaria gallica Marxm. & Romagn. on trees affected by ash dieback. However, we did not find any Armillaria fruiting bodies on assessed ash trees in Slovak arboreta. Post-dieback invasion by secondary pests also accelerates the death of dieback-weakened trees. In Kysihýbel Arboretum, we observed that common ash trees with symptoms of ash dieback had been heavily attacked by the ash bark beetle, Hylesinus sp., which affected the degree of crown defoliation. In older ash stands in Slovakia, Kunca et al. [64] and Zúbrik et al. [65] noted an increased number of the secondary invaders Hylesinus fraxini Panzer and Hylesinus crenatus Fabricius on trees already weakened by H. fraxineus.
Recent research studies have focused on finding resistance in host trees. The evidence for genetic control of resistance was provided by studies of clonal seed orchards for the native species F. angustifolia and F. excelsior [66,67]. In addition, species-specific variation in disease susceptibility has been previously documented among European, North American, and Asian Fraxinus species [16]. Recently, Stocks et al. [68] identified ash genes that may confer resistance to ash dieback. Given the importance of ash in forest management and urban greenery, it is crucial to continue developing resistant trees for the purpose of safeguarding European ash populations for future generations.

Conclusions
The invasive fungus Hymenoscyphus fraxineus decimates not only native European Fraxinus species, but introduced ash species as well. A survey of four Slovak arboreta revealed that the host range of H. fraxineus in Slovakia also includes Asian and North American ash species. Ash dieback was recorded from F. bungeana for the very first time. Since many introduced ash species have been attacked in arboreta in Northern and Central Europe, we assume that a similar situation can be found in arboreta and botanical gardens in other parts of Europe.