Diversity and Pathogenicity of Six Diaporthe Species from Juglans regia in China

Walnut (Juglans regia L.) is cultivated extensively in China for its substantial economic potential as a woody oil species. However, many diseases caused by Diaporthe greatly affect the health of Juglans regia trees. The present study revealed the presence of Diaporthe species from Juglans regia. A total of six species of Diaporthe were isolated from twigs of Juglans regia in three provinces in China, including two known species (Diaporthe gammata and D. tibetensis) and four novel species (D. chaotianensis, D. olivacea, D. shangluoensis and D. shangrilaensis). Phylogenetic relationships of the new species were determined by multilocus phylogenetic analyses based on partial sequences of the internal transcribed spacer (ITS) region, calmodulin (cal) gene, histone H3 (his3) gene, translation elongation factor 1-α (tef1-α) gene and β-tubulin (tub2) gene. Pathogenicity tests indicated that all Diaporthe species obtained in this study were confirmed as pathogens of Juglans regia. This study deepens the understanding of species associated with several disease symptoms in Juglans regia and provides useful information for effective disease control.


Introduction
The walnut tree (Juglans regia L.), a perennial deciduous species, stands out as an economically significant hardwood tree cultivated worldwide for its nutritious nuts and valuable timber.Leading in global production are the United States and China, accounting for 30% (611,280 tons) and 43% (0.88 million tons) of the total fruit production worldwide, respectively [1].Walnut seeds are also a high-density source of nutrients, particularly rich in proteins and essential fatty acids.The production of walnut seeds has increased rapidly worldwide in recent years.However, various diseases affect the condition of walnuts, thereby diminishing their economic potential.For example, twelve species of genera Botryosphaeria, Diaporthe, Diplodia, Dothiorella, Lasiodiplodia and Neofusicoccum cause cankers and blights of Juglans regia [2]; species of genera Colletotrichum and Fusarium cause serious leaf-spot disease [3,4]; and species of genera Diaporthe and Neofusicoccum can also cause fruit blight disease [5].
The identification of Diaporthe has traditionally relied mainly on host associations and morphological characteristics such as the shape and size of ascomata, asci, ascospores, conidiomata, conidia and conidiophores [7,[18][19][20][21].The initial concept of Diaporthe species was founded on the premise of host specificity [19], which has given rise to the designation of nearly 2000 species names for Diaporthe and Phomopsis.Nonetheless, the validity of identifying species within this genus based solely on host associations and morphological features is contentious.Previous studies have shown that the morphological characters of many Diaporthe species are not always stable, as they may vary with the environment [10,18,22].Recent studies demonstrated that most Diaporthe species could be found on diverse hosts and could co-occur on the same host or lesion with different life patterns [20,21,23,24].Therefore, identification and description of species based on host association and morphological characters are not reliable within Diaporthe [10,25,26].Currently, a polyphasic taxonomic approach combining phylogenetic and morphological analyses is widely employed [10,13,17,[25][26][27][28].Five genetic sequences (ITS, cal, his3, tef1-α and tub2) are widely used for phylogenetic analyses [27][28][29].
During the investigation of pathogens causing tree cankers or dieback diseases in Shaanxi, Sichuan and Yunnan provinces of China, branches with typical canker symptoms were collected and subsequently identified combining modern taxonomic concepts (Figure 1).Therefore, the objectives of the present study were to (i) identify Diaporthe taxa associated with dieback diseases of Juglans regia collected in this study and (ii) test the pathogenicity of species collected on Juglans regia.alpha conidia are fusiform, usually biguttulate; beta conidia are filiform, straight or more often hamate, and lack guttules; and gamma conidia are fusiform to subcylindrical, and multiguttulate [7,10].
The identification of Diaporthe has traditionally relied mainly on host associations and morphological characteristics such as the shape and size of ascomata, asci, ascospores, conidiomata, conidia and conidiophores [7,[18][19][20][21].The initial concept of Diaporthe species was founded on the premise of host specificity [19], which has given rise to the designation of nearly 2000 species names for Diaporthe and Phomopsis.Nonetheless, the validity of identifying species within this genus based solely on host associations and morphological features is contentious.Previous studies have shown that the morphological characters of many Diaporthe species are not always stable, as they may vary with the environment [10,18,22].Recent studies demonstrated that most Diaporthe species could be found on diverse hosts and could co-occur on the same host or lesion with different life patterns [20,21,23,24].Therefore, identification and description of species based on host association and morphological characters are not reliable within Diaporthe [10,25,26].Currently, a polyphasic taxonomic approach combining phylogenetic and morphological analyses is widely employed [10,13,17,[25][26][27][28].Five genetic sequences (ITS, cal, his3, tef1-α and tub2) are widely used for phylogenetic analyses [27][28][29].
During the investigation of pathogens causing tree cankers or dieback diseases in Shaanxi, Sichuan and Yunnan provinces of China, branches with typical canker symptoms were collected and subsequently identified combining modern taxonomic concepts (Figure 1).Therefore, the objectives of the present study were to (i) identify Diaporthe taxa associated with dieback diseases of Juglans regia collected in this study and (ii) test the pathogenicity of species collected on Juglans regia.

Sampling and Isolation
During the survey conducted in 2022 and 2023, 122 diseased (branches and twigs with canker symptoms) Juglans regia branch samples were collected from five Juglans regia plantations in Shaanxi, Sichuan and Yunnan in China.Approximately 15-25 Juglans regia trees were sampled from each site, and cankered tissues were collected from a single branch of each tree showing symptoms typical of branch canker and dieback for detailed examination and fungal isolation.In total, 45 samples were collected.A total of 31 Diaporthe isolates were obtained from 45 specimens by removing a mucoid spore mass from conidiomata and/or ascomata, spreading the suspension over the surface with potato dextrose agar (PDA) (200 g potatoes, 20 g glucose and 20 g agar/L water) in a Petri dish and incubating at 25 • C for up to 24 h.Hyphal tips were removed to a new PDA plate twice to obtain a pure culture.Specimens were deposited in the Museum of Beijing Forestry University (BJFC).Axenic cultures are maintained in the China Forestry Culture Collection Centre (CFCC).

Morphological Analyses
Species identification was based on morphological characteristics of the ascomata or conidiomata formed on infected host materials.Macromorphological features (structure and size of conidiomata, ascomata, ectostromatic disc and ostioles) were photographed using a Leica stereomicroscope (M205 FA) (Leica Microsystems, Wetzlar, Germany).Micromorphological features (conidiophores, conidiogenous cells, asci and conidia/ascospores) were photographed using a Nikon Eclipse 80i microscope (Nikon Corporation, Tokyo, Japan), equipped with a Nikon digital sight DS-Ri2 high-resolution colour camera with differential interference contrast.Over 20 conidiomata were sectioned and 50 conidia were selected randomly to measure their lengths and widths.Colony diameters were measured and the colony colours described after 3 days and 14 days according to the colour charts of Rayner (1970) [35].

DNA Extraction, PCR Amplification and Sequencing
Mycelium used for DNA extraction was grown on PDA for three days and obtained from the cellophane surface by scraping.The genomic DNA was extracted from axenic cultures using the modified CTAB method [36].Sequences were amplified by PCR from the ITS, cal, his3, tef1-α and tub2 genetic regions.The PCR mixtures for all genes included 10 µL Mix (Promega, Madison, MI, USA), 7 µL double deionized water, 1 µL of each primer and 1 µL template DNA.All primers and PCR conditions are listed in Table 1.PCR products were electrophoresed in 1% agarose gel, and the DNA was sequenced by the Sino Geno Max Biotechnology Company Limited (Beijing, China).DNA sequences generated by the forward and reverse primers combination were used to obtain consensus sequences using Seqman v. 7.1.0(DNASTAR Inc., Madison, WI, USA).

Phylogenetic Analyses
The sequences obtained from this study were analyzed with the NCBIs GenBank nucleotide datasets.Alignments based on ITS, cal, his3, tef1-α and tub2 sequence data, including sequences obtained from this study and those downloaded from GenBank (Supplementary Table S1), were first aligned using MAFFT v. 6 [37] and edited manually using MEGA v. 6.0 [38].Diaporthella corylina (CBS 121124) was used as the outgroup in polygenic Diaporthe analyses.Phylogenetic analyses were performed with PhyML v. 3.0 for the maximum likelihood (ML) method [39] and MrBayes v. 3.1.2for the Bayesian inference (BI) method [40].
The best-fit evolutionary models for each partitioned locus were estimated by MrModeltest v. 2.3 following the Akaike information criterion (AIC) in ML and BI analyses [41].For ML analysis, RAxML-NG was used [42].Bootstrap supports were estimated with 100 pseudoreplicates and the appropriate models for each gene.BI analyses were completed using a Markov chain Monte Carlo (MCMC) algorithm with Bayesian posterior probabilities (BPPs) [43].Trees were sampled every 100th generation after two MCMC chains were run from random trees for 10 million generations, which stopped when the average standard deviation of split frequencies fell below 0.01.For the burn-in phase of each analysis, the first 25% of the trees were discarded and the remaining trees were assessed to calculate BPP [43].FigTree v. 1.3.1 [44] was used to show phylograms.The sequence data of 31 isolates were deposited in GenBank; their accession numbers, together with those of the other species used for the analysis, are listed in Table S1.The multilocus sequence alignment was deposited in TreeBASE (www.treebase.org,accessed on 1 July 2024; study ID 31531).

Pathogenicity Test
Three representative isolates from each identified Diaporthe species were selected for pathogenicity testing in this study (D. chaotianensis: CFCC 70718-70720; D. gammata: CFCC 70722-70724; D. olivacea: CFCC 70713, 70715, 70716; D. shangluoensis: CFCC 70728, 70729, 70731; D. shangrilaensis: CFCC 70703, 70705, 70706; D. tibetensis: CFCC 70702, 70710, 70711).The pathogenicity tests were conducted on 2-year-old Juglans regia trees 1.3 m high and 1.5 cm thick which were planted in the field at a nursery of the Forest Protection Lab (Beijing, China).After the leaves grew, they were inoculated under natural conditions to determine pathogenicity during early April 2024 (mean air temperature = 15 • C).Sterilized 5 mm diameter inoculation rings were used to scald the bark surface of each branch to a depth of 2 mm.Agar plugs of the same size were removed from 6-day-old colonies of selected isolates, inserted into the wounds, sealed with moistened cotton wool and protected with parafilm.Six replications were made for each isolate.One plant per isolate was used as the negative control, and an equal number of plants inoculated with PDA agar plugs without colonies were used as the positive control.After one week, the parafilm and cotton wool were removed.These inoculated plants were maintained in the field.Fourteen days after inoculation, the lengths of the spots on the bark surface were measured from the inoculation point upwards and downwards using a digital calliper and then averaged.All spots from the experimental and control groups were reisolated to verify that the morphological characteristics and DNA sequences were consistent with the original isolates, thus fulfilling the Koch hypothesis.Differences in lesion length between isolates were analyzed by one-way analysis of variance (ANOVA) followed by least significant difference (LSD) tests.Statistical analysis was carried out by SPSS v. 20.0 and considered as significant at p < 0.05.

Phylogeny
Each gene region and the concatenated sequences of five genetic regions (ITS, cal, his3, tef1-α and tub2) were analyzed to infer the interspecific relationships within Diaporthe.The topological structures derived from each gene region were found to be consistent with those of the combined dataset (Figure 2, Figures S1-S5).The combined sequences dataset comprised 481 isolates (480 ingroup taxa including 31 new isolates in this study and one outgroup taxa, Diaporthella corylina CBS 121124).The sequence fragments were 3954 characters including gaps (548 for ITS, 952 for cal, 663 for his3, 785 for tef1-α and 1006 for tub2).The topologies resulting from ML and BI analyses of the concatenated dataset were similar.ML bootstraps (ML BS ≥ 60%) and Bayesian posterior probabilities (BPP ≥ 0.95) have been shown above the branches (Figure 2).For ML analysis, the matrix had 3044 distinct alignment patterns.The model parameters were as follows: A = 0.217335, C = 0.286382, G = 0.258070, T = 0.218908: substitution rates: AC = 0.866066, AG = 3.129585, AT = 0.966855, CG = 0.698073, CT = 3.794326, GT = 1.000000; gamma distribution shape parameter α = 0.538403; and likelihood value of ln: −124,733.462770.
The current 31 isolates clustered in six clades representing six species.Two represented known species (D. gammata and D. tibetensis) and four new clades.Isolates in clades 1, 2, 4 and 6 were separated from all other species and were also highly supported (ML/BI = 100/1) (Figure 2), representing four novel species (D. chaotianensis, D. olivacea, D. shangluoensis and D. shangrilaensis), which are detailed in the following sections.
Typification: China, Sichuan Province, Guangyuan City: Chaotian District, Zhongzi Town, 32 Cultural characteristics: Colonies initially white, grown to 63 mm after 3 days, compact at the centre and sparse at the surroundings, becoming honey after 7 days.Colonies are flat with a uniform texture, lacking aerial mycelium.Conidiomata were sparse, black, distributed irregularly (Figure 4a).
Additional materials examined: China, Sichuan Province, Guangyuan City: Chaotian District, Zhongzi Town, 32 • 41 ′ 21 ′′ N, 106 • 02 ′ 25 ′′ E, from branches of Juglans regia, 10 October 2023, Y.X.Li, L. Lin and X.L. Fan (BJFC-S2347, living culture CFCC 70719, 70721).Cultural characteristics: Colonies initially white, grown to 63 mm after 3 days, compact at the centre and sparse at the surroundings, becoming honey after 7 days.Colonies are flat with a uniform texture, lacking aerial mycelium.Conidiomata were sparse, black, distributed irregularly (Figure 4a).-S5).Morphologically, three types of conidia were observed in the present study, which is consistent with the description of D. gammata [53].Therefore, five isolates collected in this study are identified as D. gammata.Additionally, this is the first report of D. gammata being responsible for Juglans regia shoot canker.Culture characteristics: Cultures on PDA incubated at 25 • C in darkness, colony originally flat with a white felty aerial mycelium after 3 days, becoming a white compact aerial mycelium at the centre with a smoke-grey aerial mycelium at the margin after 7-10 days (Figure 4e), margin irregular.Conidiomata are sparse, irregularly distributed over the agar surface after 30 days.

Analysis of Pathogenicity Test
For pathogenicity tests via Juglans regia shoot inoculations, the results showed that all the tested Diaporthe isolates could induce discoloured and necrotic lesions 14 d post inoculation (Figure 10; Table 2).Sunken cankers were obvious on the stems and produced brown lesions upward and downward from the point of inoculation.At the same time, no lesions were observed on the branches of the control (Figure 10g1,g2).Koch's postulates were fulfilled and confirmed that all the tested Diaporthe species in this study are pathogens of Juglans regia.Diaporthe Shangrilaensis was shown to be the most aggressive species:

Analysis of Pathogenicity Test
For pathogenicity tests via Juglans regia shoot inoculations, the results showed that all the tested Diaporthe isolates could induce discoloured and necrotic lesions 14 d post inoculation (Figure 10; Table 2).Sunken cankers were obvious on the stems and produced brown lesions upward and downward from the point of inoculation.At the same time, no lesions were observed on the branches of the control (Figure 10(g1,g2)).Koch's postulates were fulfilled and confirmed that all the tested Diaporthe species in this study are pathogens of Juglans regia.Diaporthe Shangrilaensis was shown to be the most aggressive species: isolates CFCC 70703, 70705, 70706 of D. shangrilaensis caused larger lesions (Figure 10(e1,e2)), followed by the isolates CFCC 70713, 70715, 70716 of D. olivacea (Figure 10(c1,c2)), and the remaining isolates CFCC 70728, 70729, 70731 of D. shangluoensis induced smaller lesions (Figure 10(d1,d2)).Isolates of D. gammata caused only slight discolouration around the inoculation points (Figure 10(b1,b2)).In contrast, the remaining Diaporthe isolates induced very limited lesions (5 < mean lesion length < 6 mm).Although the difference in mean lesion length between D. chaotianensis and D. tibetensis was not significant, the disease incidence of D. chaotianensis (89%) was higher than that of D. tibetensis (56%).Therefore, D. chaotianensis was found to be more virulent than D. tibetensis.Diaporthe gammata was significantly lower than the other five species in lesion length, with canker length averaging 9.1 ± 0.6 mm.ANOVA revealed significant (p < 0.05) differences among the treatment means in all six species (Table 2).

Analysis of Pathogenicity Test
For pathogenicity tests via Juglans regia shoot inoculations, the results showed that all the tested Diaporthe isolates could induce discoloured and necrotic lesions 14 d post inoculation (Figure 10; Table 2).Sunken cankers were obvious on the stems and produced brown lesions upward and downward from the point of inoculation.At the same time, no lesions were observed on the branches of the control (Figure 10g1,g2).Koch's postulates were fulfilled and confirmed that all the tested Diaporthe species in this study are pathogens of Juglans regia.Diaporthe Shangrilaensis was shown to be the most aggressive species: isolates CFCC 70703, 70705, 70706 of D. shangrilaensis caused larger lesions (Figure 10e1,e2), followed by the isolates CFCC 70713, 70715, 70716 of D. olivacea (Figure 10c1,c2), and the remaining isolates CFCC 70728, 70729, 70731 of D. shangluoensis induced smaller lesions (Figure 10d1,d2).Isolates of D. gammata caused only slight discolouration around the inoculation points (Figure 10b1,b2).In contrast, the remaining Diaporthe isolates induced very limited lesions (5 < mean lesion length < 6 mm).Although the difference in mean lesion length between D. chaotianensis and D. tibetensis was not significant, the disease incidence of D. chaotianensis (89%) was higher than that of D. tibetensis (56%).Therefore, D. chaotianensis was found to be more virulent than D. tibetensis.Diaporthe gammata was significantly lower than the other five species in lesion length, with canker length averaging 9.1 ± 0.6 mm.ANOVA revealed significant (p < 0.05) differences among the treatment means in all six species (Table 2).

Discussion
The current study reveals the diversity and pathogenicity of Diaporthe species from Juglans regia in China.Six Diaporthe species were identified from the collected specimens in this study (Diaporthe chaotianensis, D. gammata, D. olivacea, D. shangluoensis D. shangrilaensis and D. tibetensis).Among these, D. chaotianensis, D. olivacea, D. shangluoensis and D. shangrilaensis were described as four novel species and D. gammata was reported for the first time in Juglans regia trees.These findings suggest a higher level of diversity among Diaporthe species responsible for cankers on Juglans regia than has been previously recognized.Moreover, Koch's postulates confirmed that those species were pathogens of Juglans regia.
The taxonomy of Diaporthe species is increasingly attracting attention from researchers.Most research focuses on the identification and descriptions of novel species and new host records, as well as on the regulation of pathogenicity in important Diaporthe species, indicating that genus Diaporthe has a high potential for rapid evolution [14].Previous studies have revealed that Diaporthe species have high genetic diversity on a single host.For example, Gao et al. (2016) [58] reported nine species of Diaporthe isolated from Camellia in China, and Wan et al. (2023) [59] revealed three new Diaporthe species on Acer palmatum in China [58,59].This study collected an extensive number of Diaporthe isolates from areas of Juglans regia cultivation to reveal the genetic diversity of Diaporthe species.Previous studies have reported the presence of Diaporthe species in Juglans regia, and the results of this study further support these findings.Indeed, more novel species will likely be found in the future because several species of Diaporthe have a wide host range and can move between hosts among geographic regions.For example, D. gammata was originally reported on Citrus in Chongqing Municipality [53], but the fungus has also been found on Juglans regia in this study.
The current pathogenicity tests showed that D. chaotianensis, D. gammata, D. olivacea, D. shangluoensis, D. shangrilaensis and D. tibetensis are pathogens, which could pose threats to the Juglans regia industry in China.Furthermore, D. shangrilaensis, D. olivacea and D. shangluoensis are more aggressive among the six species, and this should be considered in the development of disease control measures.Among these species, Diaporthe shangrilaensis demonstrated greater aggressiveness, evidenced by its more potent virulence and the larger lesions it induced.To clarify the reproductive characteristics of this pathogen, additional quantitative experimental analysis is warranted.The virulence of Diaporthe species could be affected by environmental factors such as moisture content, rainfall intensity and temperature [60,61].Manawasinghe et al. (2018) [15] found that environmental factors could alter the life mode of the fungi, from endophytic or saprophytic to pathogenic, thus enabling the colonization of new hosts.Diaporthe tibetensis was first reported by Fan et al. (2018) [31] on Juglans regia in Tibet Autonomous Region, and was also found on Juglans regia branches in Yunnan in this study, where the climates are quite different from the place where the pathogenicity tests were conducted.The pathogenicity tests showed that the majority of these species obtained in this study are weakly aggressive or non-aggressive to Juglans regia branches (Figure 10f1,f2; Table 2).This may be because the experimental conditions were different from those in the natural environment and because of differences in climate between the north and south of China.This implies that D. tibetensis may become highly aggressively pathogenic to Juglans regia under favourable environmental conditions, which need to be determined by further research in order to prevent Juglans regia cankers.
This study provides novel information on the ability of those species to cause disease in Juglans regia.In addition, many studies have reported that most Diaporthe species had a wide host range.For example, Diaporthe eres could cause shoot blight, leaf necrosis and branch canker on different hosts [62][63][64][65][66]. Diaporthe sojae was confirmed as the pathogen of fungal diseases on pears, sunflowers, honeybush, kiwi fruit and soybeans [16,[67][68][69][70].Those studies indicated that species obtained in this study may be capable of infecting other plants.In this study, pathogenicity tests were conducted exclusively on Juglans regia.Therefore, it would be recommended that pathogenicity tests be conducted on other plants in future studies.In addition, among the eight species from Juglans regia in China previously

Figure 1 .
Figure 1.Canker and dieback diseases caused by Diaporthe species in Juglans regia.(a,b) Disease of the Juglans regia caused by Diaporthe in the field.(c-f) The trees infected by Diaporthe.

Figure 2 .
Figure 2. Phylogenetic tree of Diaporthe based on multiple gene loci (ITS, cal, his3, tef1-α and tub2) derived from ML analysis.ML bootstrap support values above 60% and Bayesian posterior probabilities above 0.95 are shown near nodes.Ex-type isolates are in bold.Strains in the current study are in blue.Isolates in this study are highlighted in two different colours.Clade 1-2, 4, 6 represent novel species.Clade 3, 5 represent known species.

Figure 2 .Figure 3 .
Figure 2. Phylogenetic tree of Diaporthe based on multiple gene loci (ITS, cal, his3, tef1-α and tub2) derived from ML analysis.ML bootstrap support values above 60% and Bayesian posterior probabilities above 0.95 are shown near nodes.Ex-type isolates are in bold.Strains in the current study are in blue.Isolates in this study are highlighted in two different colours.Clade 1-2, 4, 6 represent novel species.Clade 3, 5 represent known species.

Table 1 .
Genes used in this study with PCR primers, primer DNA sequence, optimal annealing temperature and corresponding references.

Table 2 .
Disease incidence and lesion size on Juglans regia branches one month after inoculation with isolates of Diaporthe species.
Different lowercase letters indicate significant differences among six fungal species (LSD test, p < 0.05).