Neoscytalidium novaehollandiae causes dieback on Pinus eldarica and its potential for infection of urban forest trees

Neoscytalidium novaehollandiae is one of the most important pathogens on woody plants which has increasingly been reported as a pathogen in different hosts in recent years. The pine trees are widely cultured in many cities of Iran. In recent years, dieback symptoms were observed on Pinus eldarica trees in Tehran and Qazvin provinces. The aim of this study was to investigate the dieback causal agent on P. eldarica trees in Iran. The branches and cones of P. eldarica trees were sampled for fungal isolation. The morphological and molecular characterizations (ITS, LSU, and TEF1-α regions) identified N. novaehollandiae as a dieback causal agent. This is the first report of N. novaehollandiae disease of P. eldarica trees in Iran. Furthermore, disease severity was assayed on 19 urban forest trees under three different temperature and relative humidity (RHs) regimes. C regime (29 °C and 15% RH) displayed more disease severity on detached branches than B (24 °C and 80% RH) and A (19 °C and 35% RH) ones. This study presents the host range of this pathogen, and showed that these potential hosts are prone to this pathogen under high temperature and low humidity which urban forest trees experienced in recent decades.

Despite similarity among four Neoscytalidium species, the diagnostic morphological characters of N. novaehollandiae are having both two-septate ( Fig. 2j-n) and muriform (Fig. 2o,p) conidia, while N. dimidiatum, N. orchidacearum and N. oculus lack these characters. So, morphological investigations showed that isolated fungi belonged to N. novaehollandiae. Phylogenetic analysis. To specify the phylogenetic affinities of strain "PTD-MA" and the closest species and genus (Table 2), two separate phylogenetic trees with combined sequences of ITS_TEF-1α (dataset1), ITS_LSU (dataset2) and ITS_TEF-1α_LSU (combined sequences of dataset1 and 2) were organized for Bayesian inference (BI). For ITS_TEF-1α phylogeny (Bayesian tree), a total number of 28 sequences were selected including the newly generated sequences and sequences of N. novaehollandiae, N. dimidiatum, Neofusicoccum parvum, N. austral, Macrophomina phaseolina, Pseudofusicoccum adansoniae, Diplodia mutila, Botryosphaeria_ramosa, and Aplosporella longipes (as outgroup). Additionally, for ITS_LSU phylogeny (Bayesian tree), a total number of 30 sequences were selected including the newly generated sequences and sequences of N. novaehollandiae, N. dimidiatum, N. oculus, N. orchidacearum, Neofusicoccum grevilleae, N. arbuti, Macrophomina phaseolina, Lasiodiplodia pseudotheobromae, Dothiorella sarmentorum, Diplodia mutila, Botryosphaeria dothidea, Pseudofusicoccum stromaticum, and Saccharata proteae (as out group). Comparisons of combined sequences of ITS_TEF-1α ( Fig. 3), combined sequences of ITS_LSU (Fig. 4) and combined sequences of ITS_TEF-1α_LSU ( Fig. 5) of isolate "PTD-MA" with sequences in GenBank affirmed that the present species in this study belongs to Neoscytalidium novaehollandiae. ITS, TEF-1α and LSU sequences of isolate "PTD-MA" were deposited at NCBI under accession numbers MW605153, MW605154, and MW605155, respectively. Pathogenicity tests. Forty days post inoculation, the pathogenicity tests on the detached branches and fruits in pine showed the brownish chlorotic lesions and chlorotic tissues around inoculation point with black spores, respectively (Fig. 6a-c). The average lesion length was 19 ± 1 cm one month after inoculation, while no disease symptoms were observed in controls. Two months after artificial inoculation on pine saplings, the tree dried, and then the tree bark was torn, and black spores were appeared on the bark (Fig. 6d). All of the controls displayed no symptoms. The causal agent was successfully recovered, approving Koch's postulates. The high mass of black spores produced. It is concluded that they can be air-borne and infect the same hosts or non-host plants in natural conditions. Quantitative analyses of disease. The results of three temperature and RH regimes (A, B, and C) showed the significant effects of high temperature and low humidity on the detached branches in 19 hosts. In C regime, the pathogen affected all plant species, and the inoculated tissues changed to brownish lesions (Fig. 7). The highest percentage of disease severity (PDS) discerningly belonged to willow, fig, apricot and mulberry with 100%. In B regime, 12 hosts displayed the symptoms which highest PDS belonged to apricot with 98%. Seven hosts including ash, grape, chinaberry, poplar, maple, elm and mulberry showed no symptoms, resistant to this pathogen in this regime. Furthermore, in A regime, 13 hosts showed symptom which highest PDS belonged to apricot with 89%. Additionally, 5 hosts including ash, grape, chinaberry, poplar, and elm showed no symptoms and considered as resistant hosts to this pathogen in this regime. Based on our knowledge, the results proposed that the high temperature and low humidity can play an essential role in host range of this fungal pathogen, so that temperature increment and relative humidity decrement contribute to establish it on other hosts.
Based on the total average of lesion lengths in three conditions, the highest amounts belonged to C (10.75), A (3.11), B (1.74) regimes, respectively. Additionally, the total percentage of disease severity were recorded for C (15.63), A (8.87), and B (54.17) regimes, respectively. The highest PDS on detached branches of 19 hosts was recorded in C regime (high temperature and low humidity) (Fig. 8A). Total PDS were calculated for three regimes which C regime had the highest mean (Fig. 8B). The heatmap of 19 hosts displayed more PDS of N. novaehollandiae in C regime than A and B ones (Fig. 8C). Total scales related to all host responses to three regimes showed the susceptible and resistant condition in hosts affected by pathogen (Fig. 8A). These graphs clearly verified that high temperature and low humidity led to more PDS i.e., more damage to inoculated detached branches.

Discussion
This study was performed for the first time to identify the dieback causal agent in pine trees in Tehran and Qazvin provinces. Causal agent of pine was identified as N. novaehollandiae according to morphological and molecular (ITS, TEF-1α and LSU sequences) studies 1,34,35,39,71 . The interesting point of this pathogen is that it may survive in spilled fruits at the foot of trees. Since this pathogen produces high populations of black powdery spores under torn skin, it can be air-born and infect other susceptible, injured or weak hosts.  [36][37][38] . Furthermore, the pathogenicity of this species was confirmed on plants in China, Turkey and Australia which infected baobab (Adansonia gibbosa), bardi bush (Acacia synchronica), blue grevillea (Grevillia agrifolia), rattlepod (Crotalaria medicaginea) 39 , mango (Mangifera indica) 40,41 , elm (Ulmus densa) 42 , Grapevine (Vitis vinifera) 43 , pistachio (Pistacia vera) 44 , almond (Prunus dulcis) 45 , japanese persimmon (Diospyros kaki) 46 , tomato (Solanum lycopersicum( 47 , pear (Pyrus communis) 48 and sage (Salvia officinalis) 49 . Based on the results, the host range of N. novaehollandiae was confined to 12 species of the plant hosts and most of them were woody plants. This study represents the first report of this species on P. eldarica worldwide. The present study showed the pathogen potential for attacking pine trees in Tehran and Qazvin provinces. So, to date, this pathogen was reported in six provinces of Iran including Tehran, Qazvin, Kerman, Kermanshah, Gilan and Sistan-Baluchestan [36][37][38] .
The pathogenicity tests on the detached branches and five-year-old tree confirmed that this pathogen can make brown lesions and decline symptoms associated with black spores of fungi at 29 °C. In another study, the pathogenicity of N. novaehollandiae was confirmed on Quercus brantii detached branches (incubated on a 12/12 h light/dark cycle at 25 °C for 25 days) and greenhouse pathogenicity tests on 2-year-old seedlings (natural day/night length, 25-27 °C, 60-70% RH), while no symptoms were observed on the mock-inoculated controls including Ficus carica, Acer monspessulanum, Crataegus aronia, Amygdalus scoparia, Cornus mas, Pistacia atlantica and Pinus eldarica 37 . The pathogenicity tests on branches of selected forest trees (12-15 years old) showed that lesion lengths varied between species including Punica granatum, Alnus glutinosa, Pterocarya fraxinifolia, Figure 3. Bayesian tree inferred under the GTR + G model from the combined sequences of ITS and TEF1-α for Neoscytalidium novaehollandiae and related species of closest genus, using Aplosporella longipes 71 as outgroup. The sequences aligned using MAFFT software and edited using Gblocks program. Bayesian posterior probabilities more than 0.50 are given for appropriate clades. Newly obtained sequence and other isolates of N. novaehollandiae indicated by yellow framework. PTD-MA is the strain of this study.  [36][37][38][39][40][41][42][44][45][46][47][48][49] . This study assessed the response of the detached branches of pine, willow, cypress, apricot, magnolia, fig, ash, grape, chinaberry, poplar, pomegranate, ginkgo, sycamore, maple, elm, mulberry, catalpas, olive, and ailanthus against N. novaehollandiae under three temperature and relative humidity regimes for the first time. The application of several growth chambers made it possible to set three temperature and relative humidity regimes to gain reliable results. Likewise, more investigations affirmed the susceptibility of various cultivars and hosts to pathogenicity assays in controlled conditions. Mideros et al. 50 51 . Lasiodiplodia theobromae, as decline pathogen, was evaluated to diagnose the susceptible varieties in mango trees, and the results revealed that most of the mango varieties were susceptible to this pathogen except Bagan Pali, Saroli and Saleh Bhai 52 . Also, the cultivars of olive were exposed to Verticillium dahlia to determine the resistance or susceptible cultivars. Most cultivars exhibited the susceptibility except two genotypes (Kalamon and Koroneiki) 53 . Similar investigations have shown the pathogenicity test of Diplodia bulgarica on apple cultivars divided the www.nature.com/scientificreports/ cultivars into most susceptible and moderately resistant groups 54 . Although the optimum conditions for most fungal pathogens are the combination of high temperature and high humidity 55 . This research and other studies showed that N. dimidiatum [56][57][58][59][60] and even N. novaehollandiae can progressively grow and make symptoms in high temperatures and low relative humidity. An interesting point of this study is the determination of the susceptible hosts. So, the plantation of these susceptible hosts will be avoided in regions with high infections. Conclusively, this study presents the valuable information towards novel researches on both P. eldarica and N. novaehollandiae. More investigations are required to characterize factors promoting diseases caused by N. novaehollandiae. This can notify the strategies to prevent and manage diseases caused by this pathogen not only in P. eldarica but also in other woody plants and crops in the different regions of Iran.

Material and methods
Sampling and pathogen isolation. The symptoms of the decline and dieback were observed on the pine trees in Tehran (35°43′48″ N, 51°12′35″ E and 35°44′16″ N, 51°09′52″ E) and Qazvin (36°18′50″ N, 49°58′26″ E and 36°19′15″ N, 49°58′31″ E) provinces in 2019 (Fig. 9). The symptomatic fruits and branches of the pines were sampled in paper envelopes and transferred to the laboratory and kept at 5 °C. The infected tissue was cut into 1 × 2 cm pieces by sterilized gardening scissors, dipped in 70% ethanol for 80 s, and washed two times with sterile distilled water for 1 min. These pieces were placed in sterilized blotting paper to eliminate excess water. Four to six pieces of surface-sterilized infected parts were cultured on Potato dextrose agar (PDA) plates supplemented Figure 5. Bayesian tree inferred under the GTR + G model from the combined sequences of ITS, TEF1-α and LSU for Neoscytalidium novaehollandiae and related species of closest genus, using Saccharata proteae as outgroup. The sequences aligned using MAFFT software and edited using Gblocks program. Bayesian posterior probabilities more than 0.50 are given for appropriate clades. Newly obtained sequence and another isolate of N. novaehollandiae indicated by yellow framework. PTD-MA is the isolate of this study.   DNA extraction, PCR amplification and sequencing. Genomic DNA was extracted from one representative isolate as described previously [61][62][63] . Briefly, the isolate was grown on potato dextrose broth (PDB) for 8 days at 27 °C and genomic DNA was extracted from the fungal colony. The fungal mycelia were ground under liquid nitrogen in a mortar and pestle. Then 40-100 mg of ground mycelia transferred to a 1.5 mL microtube and 400 μL of DNA salt solution (Tris-HCl, 100 Mm; EDTA, 5 Mm; NaCl, 1.4 M; pH, 7.5-8) was added and mixed thoroughly with a vortex. This microtube was transferred to a 65 °C water bath for 15 min, and then was incubated at ice for 15 min. Then the homogenate was centrifuged for 10 min at 10,000 rpm at 4 °C. The supernatant was transferred to a new tube and then cold isopropanol (0.7 its volume) was added to it. The tube was slowly vortexed several times for 4 s. The tube was centrifuged at 20 °C for 15 min at 10,000 rpm. The supernatant was  Internal transcribed spacer (ITS) region, large-subunit (LSU) region and translation elongation factor 1-α (TEF1-α) region were amplified using ITS1/ITS4 64 , LROR/LR7 65 and EF1-728F/EF1-986R 66 primers, respectively (Table 1). PCR reaction (25 µl) included 12.5 µl Mastermix, 9 µl deionized water, 1 µl Forward primer, 1 µl Reverse primer, and 1.5 µl DNA. The PCR procedure for ITS and LSU was as follows: initial denaturation at 95 °C for 3 min, followed by 35 cycles at 94 °C for 40 s, 58 °C for 45 s, and 72 °C for 1 min, and a final extension of 72 °C for 10 min. Furthermore, the PCR procedure for EF1-α was as follows: initial denaturation at 96 °C for 3 min, followed by 35 cycles at 95 °C for 30 s, 54 °C for 45 s, and 72 °C for 45 s, and a final extension of 72 °C for 7 min. The PCR product was analyzed on a 1% agarose gel stained with GreenView Ultra. The PCR products were purified and sequenced by Biomagic Company (Karaj-Iran).
Phylogenetic analysis. The newly sequenced regions were observed with chromas software (www. techn elysi um. com. au/ chrom as. html) for picks of nucleotides and then were manually edited. The Basic Local Alignment Search Tool (BLAST) (https:// blast. ncbi. nlm.nih.gov/Blast.cgi) was employed to compare the new ITS, TEF-1α, and LSU sequences with other sequences accessible in GenBank database. The datasets of combined sequences of ITS_TEF-1α and ITS_LSU sequences were separately aligned using the Q-INS-i algorithm of an online version of MAFFT v.7.205 (https:// mafft. cbrc. jp/ align ment/ server/) 67 . The Gblocks program (version 0.91b) with all the three less stringent parameters (Allow smaller final blocks, Allow gap positions within the final blocks, and Allow gap positions within the final blocks) a server tool at the Castresana Lab (http:// molev ol. cmima. csic. es/ castr esana/ Gblocks_server.html) was utilized for editing of the alignments, i.e., to remove the regions with poor align or divergent positions. The combined sequences of ITS_TEF-1α and ITS_LSU aligned and edited in this manner were called dataset1 and dataset2, respectively. The best-fitting models for dataset1 and dataset2 were chosen using PAUP*/MrModeltest.2 68 . Bayesian analyses were carried out using MrBayes 3.1.2 69 with a starting random tree and GTR + G model for ITS_TEF-1α (dataset1) and SYM + G model for ITS_LSU (dataset2) with four million generations. Dendroscope V.3.2.8 was used for the visualization of output files prepared by phylogenetic programs 70 . Dataset1 and dataset2 are indicated at Table 2.
Pathogenicity tests. Inoculating detached branches is a standard test described by Afek et al. 72 and researchers [73][74][75][76] . To clarify terminology, these was cited as in vitro, in vivo or in planta tests. To gain consistent test material, the detached branches (2 cm diameter) from healthy 2-year-old shade trees (pine, willow, walnut, cypress, apricot, magnolia, ash, grape, chinaberry, poplar, pomegranate, ginkgo, sycamore, maple, elm, mulberry, fig, catalpas, olive, and ailanthus) were selected and cut into 19-20 cm long pieces. A total of 342 fresh and healthy branches (285 pieces for pathogenicity tests and 57 pieces as negative control) were used for the pathogenicity tests 72 . Fungal isolates were grown on PDA medium for 8 days, and then the skin and a little part of the wood of the detached branches were removed and inoculated with a fungal disc (5 mm diameter). The control was inoculated with PDA. The lesions inoculated with fungal disc and PDA were sealed with Parafilm, and both ends were also sealed with wet cotton to prevent the desiccation of the detached branches and maintained in the growth chamber at 19 °C with 35% relative humidity (RH) for 20 days. After 20 days, these detached branches were divided into three parts, and they were placed in three different conditions in regard to temperature and RH, designated as A, B and C regimes, in the growth chamber for 20 days. These regimes were as follow: A regime (19 °C and 35% RH), B regime (24 °C and 80% RH) and C regime (29 °C and 15% RH). The cotton pieces were wetted every other day to prevent the desiccation of the detached branches. Furthermore, a pathogenicity test was performed on five-year-old pine trees. Artificial lesions (5 cm long and 2 cm wide) were made on the skin and a part of wood in tree and inoculated as described above. Koch's postulates were done to confirm fungal pathogenicity.
Disease measurement. Lesion length was measured in all hosts based on color change to brownish necrosis on the detached branches. For the evaluation of the percentage of disease severity (PDS), lesion length was divided by branch length (Table 3.).

Statistical analysis.
The experiment, containing all controls and different treatments, was performed in quintuplicate. The hypothesis of normality and equal variance were tested. Conventional parametric statistics Table 1. Sequences of the primers used for PCR analysis.