Skip main navigation
Close Drawer MenuOpen Drawer Menu
Previous
Next
Postharvest PathologyFree Access icon

Diversity of Botryosphaeriaceae and Diaporthe Species Associated with Postharvest Apple Fruit Decay in Serbia

    Affiliations
    Authors and Affiliations
    • Nina Vučković1
    • Ivana Vico1
    • Bojan Duduk2
    • Nataša Duduk1 †
    1. 1University of Belgrade-Faculty of Agriculture, Belgrade, Serbia
    2. 2Institute of Pesticides and Environmental Protection, Belgrade, Serbia
    Published Online:16 Mar 2022https://doi.org/10.1094/PHYTO-07-21-0304-R

    Abstract

    Family Botryosphaeriaceae and the genus Diaporthe (family Diaporthaceae) represent diverse groups of plant pathogens, which include causal agents of leaf spot, shoot blight, branch and stem cankers, dieback, and pre- and postharvest apple fruit decay. Apple fruit with symptoms of light to dark brown decay were collected during and after harvest from 2016 to 2018. Thirty selected isolates, on which pathogenicity was confirmed, were identified and characterized based on multilocus phylogeny and morphology. Five species from the family Botryosphaeriaceae and two from the genus Diaporthe (fam. Diaporthaceae) were discovered. The most commonly isolated was Diplodia seriata followed by Botryosphaeria dothidea. In this work, Diaporthe rudis is described as a new postharvest pathogen of apple fruit. Diplodia bulgarica, Diplodia sapinea, Neofusicoccum yunnanense, and Diaporthe eres are initially described as postharvest apple and D. sapinea as postharvest quince and medlar fruit pathogens in Serbia. Because species of the family Botryosphaeriaceae and the genus Diaporthe are known to cause other diseases on their hosts, have an endophytic nature, and have a wide host range, findings from this study imply that they may become a new challenge for successful fruit production.

    Species of the family Botryosphaeriaceae and the genus Diaporthe (family Diaporthaceae) are known as saprobes, endophytes, and pathogens of plants around the world (Crous et al. 2006; Phillips et al. 2012; Slippers and Wingfield 2007; Udayanga et al. 2011). Most of them have the ability to infect a wide range of host plants. They are also known to persist endophytically within healthy tissue until the hosts become vulnerable under the influence of stressful environmental conditions, which can lead to rapid development of severe plant disease (Mehl et al. 2013; Slippers and Wingfield 2007). The importance of these fungi as pathogens has been escalating in recent years, which is mainly associated with climate changes, contributing not only to the increase of plant susceptibility but also to the expansion of these potentially aggressive species (Sakalidis et al. 2021; Slippers and Wingfield 2007; Slippers et al. 2017; Zlatković et al. 2016, 2017).

    Some Botryosphaeriaceae and Diaporthe species are important apple pathogens causing leaf spot, shoot blight, branch and stem cankers, dieback, and pre- and postharvest fruit rot (Abreo et al. 2012; Brown and Britton 1986; Brown-Rytlewski and McManus 2000; Cloete et al. 2011; Delgado-Cerrone et al. 2016; Jurick et al. 2013; Kaiser et al. 2002; Latorre and Toledo 1984; Phillips et al. 2012; Rosenberger 2007; Santos et al. 2017b; Sessa et al. 2017; Slippers et al. 2007; Türkölmez et al. 2016; Vasić et al. 2013). Apple fruit can be infected at any stage, but soft, light to dark brown rot symptoms mainly appear on ripe fruit during or after harvest, which indicates the occurrence of latent infection (Kim et al. 2001; Slippers and Wingfield 2007). Botryosphaeriaceae species described as apple pathogens are Botryosphaeria dothidea, B. rosaceae, Diplodia bulgarica, D. intermedia, D. malorum, D. mutila, D. seriata, D. pseudoseriata, Dothiorella iberica, Lasiodiplodia theobromae, Neofusicoccum algeriense, N. australe, N. italicum, N. luteum, N. nonquasitum, N. parvum, and N. ribis (Brown and Britton 1986; Brown-Rytlewski and McManus 2000; Delgado-Cerrone et al. 2016; Laundon 1973; Marin-Felix et al. 2017; Phillips et al. 2005, 2012; Rooney-Latham and Soriano 2016; Slippers et al. 2004, 2007; Zhang et al. 2021; Zhou et al. 2017), whereas Diaporthe species pathogenic to apple include D. actinidiae, D. ambigua, D. amygdali, D. cynaroidis, D. eres, D. foeniculina, D. mali, D. malorum, D. nobilis, D. padi, D. perniciosa, D. pomigena, D. rudis, D. serafiniae, D. tanakae, D. virgiliae, Phomopsis perniciosa, and P. truncicola (Farr and Rossman 2021; Gomes et al. 2013; Santos et al. 2017b; Udayanga et al. 2014a, b).

    The difficulty of reliable identification is common for species of the family Botryosphaeriaceae and the genus Diaporthe because of infrequent formation of the teleomorph, overlapping phenotypic characteristics of the anamorph, wide host range, and cryptic diversification (Slippers and Wingfield 2007; Slippers et al. 2004, 2014, 2017; Udayanga et al. 2011, 2012). The use of DNA sequencing and phylogenetic analyses has enabled rapid and dramatic changes in our understanding of the taxonomy and diversity of the Botryosphaeriaceae and Diaporthe species over the past decade. Multigene phylogenetic analysis is now the standard approach in species delineation (Gomes et al. 2013; Li et al. 2020; Phillips et al. 2013; Santos et al. 2017a; Slippers et al. 2013, 2017; Udayanga et al. 2012, 2014a, b; Zhang et al. 2021). Identification of species in the Botryosphaeriaceae is mainly based on internal transcribed spacer region (ITS), translation elongation factor 1-alpha gene (TEF1), and beta-tubulin gene (TUB) (Delgado-Cerrone et al. 2016; Giambra et al. 2016; Jami et al. 2012; Marin-Felix et al. 2017; Phillips et al. 2012, 2013; Slippers et al. 2013; Zlatković et al. 2016), whereas for accurate identification of species in the genus Diaporthe, histone H3 (HIS), calmodulin (CAL), and DNA-lyase (Apn2) genes are also used (Gomes et al. 2013; Guarnaccia et al. 2018; Santos et al. 2017a, b; Udayanga et al. 2014a).

    In Serbia, species of Botryosphaeriaceae have been studied as important pathogens of ornamental and forest trees (Karadžić et al. 2000; Milijašević 2009; Zlatković et al. 2016), whereas Diaporthe spp. have been investigated on field crops (Muntañola-Cvetković et al. 1981; Petrović et al. 2015, 2016; Vidić et al. 2011). There are only a few reports about the presence of these pathogens on apple or other pome fruit, although apple (Malus domestica Borkh.) is one of the most widely grown and economically significant fruit crops in Serbia. In 2019, the harvested area of apple fruit in Serbia was 26,089 ha, with production of ∼499,578 tonnes (FAOSTAT 2019). The limited research about species belonging to Botryosphaeriaceae and Diaporthe as postharvest pathogens has resulted in only three reports on apple fruit in Serbia to date: B. obtusa (Stojanović et al. 2003), Phomopsis perniciosa (Arsenijević and Gavrilović 2005), and B. dothidea (Vasić et al. 2013), and two on quince fruit: Diplodia seriata (Vico et al. 2017) and Diaporthe eres (Ristić et al. 2016). These results suggested sporadic presence of Botryosphaeriaceae and Diaporthe as postharvest fruit pathogens. Our preliminary results revealed a more frequent occurrence of Botryosphaeriaceae and Diaporthe species as causal agents of apple fruit decay characterized by symptoms of light to dark brown rot. Therefore, this investigation originated as a result of the lack of data regarding postharvest decay caused by species of Botryosphaeriaceae family and the genus Diaporthe (family Diaporthaceae) on apple fruit in Serbia and worldwide, facts that Botryosphaeriaceae are important pathogens of ornamental and forest trees in Serbia (Zlatković et al. 2016) and that Botryosphaeriaceae and Diaporthe can cause the same type of disease/symptoms on their hosts (Chen et al. 2014a, b).

    The main hypothesis of this work was that causal agents of the observed postharvest apple fruit rot are species from the family Botryosphaeriaceae and the genus Diaporthe and that a more diverse group of these pathogens is involved in the etiology of the disease than currently known. To test the hypotheses, a comprehensive study was performed: apple fruit with light to dark brown rot were collected during and after harvest, from 2016 to 2018 and causal agents were isolated and obtained isolates were identified and characterized based on detailed multilocus analyses and morphological investigation. To fulfill Koch’s postulates, the pathogenicity of the isolates was tested on apple fruit, and association of symptoms observed both naturally and on inoculated apple fruit with corresponding causal agents was evaluated. As a result, findings from this study show the importance of Botryosphaeriaceae and Diaporthe species as causal agents of apple fruit decay and contribute to the global knowledge of these fungal groups as postharvest pathogens. Additional knowledge of the presence, symptomatology, morphology, and genetic characteristics of these pathogens will be the first step in improving apple fruit decay management and will enable apple growers and industry to produce high-quality fruit.

    MATERIALS AND METHODS

    Sampling, isolation, and pathogenicity test.

    For this study, apple fruit with symptoms of light to dark brown rot with no evident sporulation were collected from different storages (controlled atmosphere, normal atmosphere, and cellar) or orchards during harvest. Apple fruit were collected from December 2016 to April 2018 at eight locations in Serbia (Avala, Bela Crkva, Grocka, Ivanovo, Jazak, Mladenovac, Radmilovac, and Svilajnac). Fruit collected from cellars and orchards (30 fruit) were symptomatic, whereas some fruit collected from controlled or normal atmosphere storages (78 out of 173 fruit) exhibited symptoms 7 to 14 days after transfer to indoor environment (temperature 19 to 24°C). After preliminary examination of sporulation or isolations from 108 symptomatic apple fruit, fungi resembling other postharvest pathogens (Alternaria sp., Botrytis sp., Colletotrichum sp., Monilinia sp., or Penicillium sp.) were disregarded and 26 Botryosphaeriaceae and Diaporthe isolates were obtained from 25 fruit. Twenty isolates were obtained from stored apple fruit and six isolates from apple fruit collected during harvest. Similar symptoms were observed on quince (Cydonia oblonga) and medlar (Mespilus germanica) fruit after harvest in 2015, so one isolate from quince and three isolates from medlar were also included in the study. Isolates were maintained long term at −80°C. The origin of investigated isolates (location, apple cultivar, time of isolation, and storage type) is shown in Table 1.

    TABLE 1. Fungal isolates obtained in this study

    Isolation was done from symptomatic fruit after surface sanitization with 70% ethanol and aseptic removal of the skin. Fragments of tissue from the margin between symptomatic and healthy tissue (2 × 2 mm) were transferred onto potato dextrose agar (PDA; EMD Chemicals Inc., Gibbstown, NJ) and incubated at 25°C in the dark for 4 days. Subsequently, developed fungal colonies were subcultured onto fresh PDA plates and incubated under the same conditions for 7 days.

    Pathogenicity of all 26 isolates was tested on mature ‘Idared’ apple fruit to fulfill Koch’s postulates. The apple fruit was washed in commercial detergent, rinsed in sterile distilled water, dried, and then surface-sanitized using 70% ethanol. For inoculation, two wounds per fruit were made using a 5-mm cork borer. Same-size mycelial plugs were taken from the margin of 7-day-old cultures grown on PDA and placed in the wounds. Wounded fruit inoculated with sterile PDA plugs was used as a control. Three replicates were performed for each isolate and control. All inoculated fruit was kept in sterilized plastic containers lined with wet paper towels and incubated at room temperature for 14 days in the lab. Observation of symptom development was performed after 3, 7, and 14 days. Fungi were reisolated from symptomatic inoculated fruit as described above. Pathogenic isolates were purified by hyphal tip transfers to PDA for molecular and morphological characterization. Isolates were preserved in Eppendorf tubes containing 15% glycerol solution at −80°C as mycelial plugs.

    DNA extraction, PCR amplification, and sequencing.

    Genomic DNA was extracted from mycelia of 7-day-old cultures and grown on PDA at 25°C in the dark, following the cetyltrimethylammonium bromide protocol of Day and Shattock (1997). The ITS region, including the 3′ end of the 18S ribosomal RNA (rRNA) gene, the internal transcribed spacer 1, the complete 5.8S rRNA gene, the internal transcribed spacer 2, and the 5′ end of the 28S rRNA gene, was amplified for all isolates using primer pairs ITS1/ITS4 (White et al. 1990). Partial translation elongation factor 1-alpha gene (TEF1) and partial β-tubulin gene (TUB) were amplified for all Diaporthe and selected isolates of Botryosphaeriaceae using the primer pairs EF1-728F/EF1-986R (Carbone and Kohn 1999) and Bt2a/Bt2b (Glass and Donaldson 1995), respectively. Additionally, partial CAL, HIS, and Apn2 genes were amplified using primer pairs CAL228F/CAL737R (Carbone and Kohn 1999), CYLH3F/H3-1b (Crous et al. 2004; Glass and Donaldson 1995), and apn2fw2/apn2rw2 (Udayanga et al. 2014a), respectively, for isolates of Diaporthe. The PCR mix (25 µl) contained 1 µl of template DNA, 1× PCR Master Mix (Thermo Scientific, Vilnius, Lithuania), and 0.4 µM of each primer. The PCR conditions were as follows: initial denaturation at 95°C for 2 min, followed by 39 cycles of denaturation at 95°C for 30 s, annealing at 54°C for 50 s (Apn2) or 55°C for 30 s (ITS) or 55°C for 50 s (TUB, CAL, HIS) or 58°C for 50 s (TEF1), and elongation at 72°C for 1 min, and a final elongation at 72°C for 10 min. The PCR products (5 µl) were separated in 1.2% agarose gel, stained with ethidium bromide (0.5 µg/ml), and visualized with an ultraviolet transilluminator.

    Amplified PCR products were commercially sequenced in both directions (Macrogen Inc., Seoul, South Korea) using the same primers as for amplification, and the obtained sequences were assembled using Pregap4 from the Staden program package (Staden et al. 2000). For species identification, the obtained sequences were compared with those publicly available in NCBI’s GenBank database using the MegaBLAST algorithm, and identical and closely related sequences were included in the analyses. Sequence alignments for each locus were performed using Clustal X (Thompson et al. 1997) under MEGA version 7 (Kumar et al. 2016) and manually adjusted in Bioedit version 7.2.5 (Hall 1999) where necessary. The sequences obtained in this study were deposited in GenBank (Table 1).

    Phylogenetic analyses.

    To infer evolutionary relationships among Botryosphaeriaceae species and confirm the identity of the obtained isolates, phylogenetic trees were constructed based on individual and concatenated sequences of three loci (ITS, TEF1, and TUB) with Melanops tulasnei as an outgroup. Botryosphaeriaceae isolates used in phylogenetic analyses are shown in Supplementary Table S1. Additionally, for resolving the relationship within Neofusicoccum spp., in multilocus phylogeny B. dothidea was an outgroup. To clarify Diplodia spp., phylogeny was based on concatenated two (ITS and TEF1) and three loci (ITS, TEF1, and TUB) with Lasiodiplodia theobromae as an outgroup.

    To identify and determine the phylogenetic position of the obtained Diaporthe isolates, phylogenetic analysis based on six loci (ITS, TEF1, CAL, TUB, HIS, and Apn2) was conducted. The sequences of these six loci were analyzed individually and in a combination of concatenated four (ITS-TEF1-CAL-TUB and TEF1-CAL-TUB-HIS), five (ITS-TEF1-CAL-TUB-HIS and TEF1-CAL-TUB-HIS-Apn2), and all six loci. Obtained and closely related Diaporthe sequences (including ex-types), with D. citri as an outgroup, were used in phylogenetic analyses (Supplementary Table S1).

    Phylogenetic analyses were conducted in MEGA X (Kumar et al. 2018) using the maximum parsimony (MP) and maximum likelihood (ML) methods. The MP trees were obtained using the tree-bisection-reconnection algorithm with search level 3, in which the initial trees were obtained by the random addition of sequences (10 replicates). For trees construction based on ML, the best nucleotide substitution model for each sequence data set was determined using MEGA X. Nearest-neighbor-interchange heuristic method, with initial tree obtained automatically by neighbor-join and BioNJ algorithms, was used for tree inference. To statistically evaluate the reliability of the inferred phylogenies, each tree was bootstrapped 1,000 times.

    Morphological characterization.

    Colony morphology and growth rate of the obtained isolates were determined on PDA. For each isolate, 5-mm mycelial plugs were taken from the margins of 7-day-old cultures grown on PDA and placed in the center of fresh PDA plates. Three replicate plates per isolate were incubated in the dark at 25°C. Colony morphology (color, form, elevation, margin, and presence of reproductive structures) was observed daily and recorded after 3, 6, 9, and 15 days of growth on PDA. Two colony diameters perpendicular to each other were measured daily for 5 days, and growth rate was calculated. The analysis of variability in growth was estimated using one-way analysis of variance followed by Student-Newman-Keuls test at P < 0.05.

    To induce sporulation, isolates were plated on 2% water agar with autoclaved pine needles on the agar surface (PNA) and incubated under continuous artificial light at room temperature for at least 30 days. One isolate (JBotRd2), because of inability to sporulate on PNA, was cultured on PDA plates acidified with lactic acid (APDA) and incubated under the same conditions. Mature pycnidia crushed by hand or conidia released from pycnidia were mounted in distilled water on microscope slides for morphological examination. Conidial observations (shape, color, presence of septa, or number of guttules) and measurements of the length and width of ∼100 conidia per isolate were performed using a compound microscope Carl Zeiss Axio Lab.A1, equipped with camera Axiocam ErC.5s and ZEN 2.1/ZEN 2 microscopy imaging software (Carl Zeiss GmbH). Minimum, maximum, 95% confidence intervals, mean, standard deviation (SD), and length and width ratio (L/W) were calculated from measurements.

    RESULTS

    Symptoms, isolation, and morphological grouping.

    Apple fruit exhibited light, medium, or dark brown rot with no evident sporulation. Symptoms varied from small circular lesions with uneven or defined margins to large sunken or swollen areas to completely decayed fruit. In some cases, exudate was present on the fruit surface, or the fruit leaked when pressure was applied. Decayed tissue color and consistency also varied from light to dark brown and from soft and watery to gelatinous.

    From symptomatic fruit collected after storage in cellars on which, after transfer to room temperature, rot rapidly progressed, nine isolates were obtained. From symptomless fruit collected in conventional cold storage facilities that developed symptoms 7 to 14 days after transfer to an indoor environment, 11 isolates were obtained. From fruit exhibiting rot symptoms in the orchard at harvest time, six isolates were obtained (Table 1).

    Based on colony morphology, the isolates obtained from apple fruit (26 isolates) as well as isolates from quince and medlar fruit included in this study (four isolates) were separated into two groups: (i) 23 isolates with fast-growing, olive-brown or gray-brown mycelium that turned dark brown to black on the reverse of PDA with age, which is typical of Botryosphaeriaceae species, and (ii) seven isolates with slower-growing, white to brown rosulate colonies resembling cultures of Diaporthe spp. This grouping was further confirmed by molecular analyses.

    Pathogenicity.

    Pathogenicity of the isolates was confirmed because they were able to cause rot symptoms on inoculated apple fruit, thus fulfilling Koch’s postulates. Three days after inoculation, symptoms started developing on all inoculated fruit, and after 14 days most of the fruit decayed completely. Color of the lesions varied from light to dark brown, with watery to gelatinous consistency similar to symptoms observed in naturally infected fruit. Seven days postinoculation, cottony mycelial growth was present on the surface of the decayed area, and it was more prominent after inoculation with Botryosphaeriaceae than Diaporthe species. Pycnidia formation was observed on some inoculated apple fruit after 14 days of incubation. All tested isolates were successfully reisolated from inoculated apples. Reisolates expressed the same morphological features as isolates, thus fulfilling Koch’s postulates. No symptoms were observed on control fruit.

    Molecular identification and phylogeny of Botryosphaeriaceae.

    ITS1/ITS4 primer pair generated amplicons of expected size and sequences of 540 to 544 nt were obtained in all 23 isolates. EF1-728F/EF1-986R and Bt2a/Bt2b primer pairs amplified the corresponding loci (TEF1 and TUB), and sequences of 246 to 260 nt and 411 to 413 nt, respectively, were obtained in the selected 12 isolates. Blast analyses and multiple sequence alignment showed that the obtained Botryosphaeriaceae isolates belong to three genera: Diplodia, Botryosphaeria, and Neofusicoccum.

    Eleven D. seriata isolates obtained in this study (JBotMl1, JBotRd4, JBotRd12, JBotGr3, JBotRd22, JBotAv10, JBotRd43, JBotSv5, JBotSv6, JBotSv7, and JIva1) were identical in ITS, and, when compared with the available sequences, were also identical with the ex-epitype of D. seriata (CBS 112555) and D. seriata from ornamental trees in Serbia (e.g., CMW 39384, CMW 39377) (Zlatković et al. 2016). TEF1 and TUB of three selected isolates from this group (JBotMl1, JBotRd4, and JBotRd12) were identical and were identical with sequences of D. seriata (CBS 112555, CMW 39384, and CMW 39377).

    Out of five D. sapinea isolates, four (JBotAv1, DBA1, MRI2, and MRI3) were identical in the three assessed loci, whereas isolate MRI5 differed from them only in ITS in 2 nt and one gap. ITS of four isolates and TEF1 of all isolates were identical with those of the ex-type isolate CBS 141915 and other reference isolates of D. rosacearum (e.g., NB8, BN-67, CAP 330). Based on TUB, they were identical to the ex-epitype isolate of D. sapinea (CBS 393.84), D. sapinea isolates from Serbia (e.g., CMW 39338, CMW 39346), and D. intermedia isolate CBS 112556. However, no TUB sequence data are publicly available for D. rosacearum for comparison. Additionally, it was noticed that our isolates and isolates of D. rosacearum differed from the ex-type of D. intermedia (CBS 124462) and the ex-epitype of D. sapinea (CBS 393.84) in TEF1 in 4 nt and 3 nt, respectively. Similarly, four out of five D. sapinea isolates and D. rosacearum isolates differed in ITS from the ex-type of D. intermedia and the ex-epitype of D. sapinea, having one and two gaps, respectively. According to recent evaluation of species in Botryosphaeriales, which reduces D. rosacearum as synonym of D. sapinea (Zhang et al. 2021), the five isolates obtained in this study were assigned to D. sapinea.

    Diplodia bulgarica isolate (JBotRd6) was identical to the ex-type isolate CBS 124254 and other reference isolates of D. bulgarica for the three assessed loci.

    Five B. dothidea isolates from this study (JBotRd1, JBotGr1, JBotRd31, JBotAv7, and JBotSv3) had identical ITS and were in line with the ex-epitype (CBS 115476) and other publicly available reference isolates of B. dothidea (e.g., CMW 39302, CMW 39304). TEF1 and TUB of selected B. dothidea isolate (JBotRd1) were identical with corresponding loci of B. dothidea (CBS 115476, CMW 39302, CMW 39304).

    Neofusicoccum yunnanense isolate (JBotRd7) was identical in all three loci to several isolates of N. yunnanense (e.g., CSF 5706, CSF 5793) and to N. parvum isolates (CMW 39318, CMW 39325) collected from ornamental trees in Serbia (Zlatković et al. 2016). It differed from the ex-type isolate of N. yunnanense CSF 6142 in TEF1 (1 nt difference) and from ex-type isolate of N. parvum ATCC 58191 in ITS, TEF1, and TUB (1, 2, and 2 nt, respectively). Based on the identity of the obtained isolate with several N. yunnanense in all loci and greater similarity with the ex-type of N. yunnanense, the isolate was named N. yunnanense.

    Each single locus and multilocus phylogeny confirmed the identity of the obtained isolates identified as D. seriata, D. bulgarica, and B. dothidea, positioning them in the corresponding well-supported clades (Figs. 1 and 2). Obtained isolates identified as D. sapinea, resided in clade with reference isolates of D. intermedia, D. rosacearum, and D. sapinea based on the individual phylogenies of ITS and TUB, whereas TEF1 phylogeny showed clustering of the obtained isolates with D. rosacearum in the distinct but weakly supported clade (data not shown). Multilocus phylogeny (ITS-TEF1-TUB) of different Botryosphaeriaceae genera provided stronger resolution for the D. rosacearum clade (including our isolates) than TEF1 phylogeny, whereas D. intermedia and D. sapinea clustered together (Fig. 1). Multilocus phylogeny (ITS-TEF1 and ITS-TEF1-TUB) of congeneric species showed the delineation of D. intermedia, D. sapinea, and D. rosacearum (including our isolates) into three clades. The same clustering was obtained in both ITS-TEF1 (Fig. 2) and ITS-TEF1-TUB phylogenies (slightly lower level of confidence when TUB was included; data not shown). Obtained N. yunnanense isolate clustered with reference isolates of N. parvum/N. ribis complex: N. algeriense, N. italicum, N. parvum, N. ribis, and N. yunnanense, based on ITS phylogeny. Species from this cluster could not be reliably separated in TEF1 or TUB phylogeny. Combined analysis of three loci revealed four clades (Fig. 1) that support values increased when congeneric species were analyzed (Fig. 3). Obtained N. yunnanense isolate clustered in N. yunnanense clade, which also contained two Serbian isolates previously identified as N. parvum.

    Fig. 1.

    Fig. 1. Phylogenetic tree resulting from analysis of concatenated internal transcribed spacer, translation elongation factor 1-alpha, and beta-tubulin sequences of Botryosphaeriaceae. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates. Values <60% are marked with ‘‘-’’. Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Melanops tulasnei. The scale bar represents 20 nucleotide substitutions.

    Download as PowerPoint
    Fig. 2.

    Fig. 2. Phylogenetic tree resulting from analysis of concatenated internal transcribed spacer and translation elongation factor 1-alpha sequences of Diplodia spp. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates. Values <60% are marked with ‘‘-’’. Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Lasiodiplodia theobromae. The scale bar represents five nucleotide substitutions.

    Download as PowerPoint
    Fig. 3.

    Fig. 3. Phylogenetic tree resulting from analysis of concatenated internal transcribed spacer, translation elongation factor 1-alpha, and beta-tubulin sequences of Neofusicoccum spp. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates. Values <60% are marked with ‘‘-’’. Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Botryosphaeria dothidea. The scale bar represents 10 nucleotide substitutions.

    Download as PowerPoint

    Statistical parameters describing the sequence alignments and phylogenetic analyses are shown in Supplementary Table S2. MP and ML analyses of the same single or multilocus dataset produced trees with similar topologies.

    Morphology of Botryosphaeriaceae.

    All obtained isolates of the Botryosphaeriaceae produced mature pycnidia on PNA within 3 to 4 weeks of incubation. Colony and conidial characteristics of the species are shown in Figure 4, whereas dimensions of conidia and microconidia are shown in Supplementary Table S3. Teleomorph structures were not observed.

    Fig. 4.

    Fig. 4. Morphological characteristics of Botryosphaeriaceae isolated from apple fruit in Serbia: A1 and A2, Diplodia seriata, 6-day-old culture on PDA, A3 to A7, conidia, and A8, microconidia; B1 and B2, Diplodia sapinea, 6-day-old culture on PDA, B3 to B5, conidia, and B6, microconidia; C1 and C2, Diplodia bulgarica, 6-day-old culture on PDA, C3, conidia discharging from pycnidia, and C4 to C6, conidia from crushed pycnidia; D1 and D2, Botryosphaeria dothidea, 6-day-old culture on PDA, D3, conidia, D4, Dichomera-like conidia, and D5, microconidia; E1 and E2, Neofusicoccum yunnanense, 6-day-old culture on PDA, E3 and E4, conidia, and E5, microconidia. Scale bars = 10 µm.

    Download as PowerPoint

    Diplodia seriata De Not., Micr. Ital. Dec. 4:6. 1942. Cultures were initially white, turning pale olive-brown from the center of the colony after 3 days, becoming dark olive-brown to gray-brown toward the margin of the colony after 6 days, and becoming dark gray on the surface and black on the reverse after 15 days of incubation. Colonies with moderately dense to dense, fluffy aerial mycelium had the entire margin. Conidia were ovoid or oblong-elliptical to subcylindrical, truncate or rounded at the base, rounded at the apex, aseptate, initially hyaline, becoming pale brown to brown and occasionally one-septate with age, in some isolates two- or even three-septate, (17.01-) 21.76 to 24.27 (-30.95) µm long, (8.16-) 9.81 to 10.95 (-13.6) µm wide (mean ± SD of 963 conidia = 23.02 ± 1.91 × 10.41 ± 0.88 µm), with an L/W ratio of 2.23 ± 0.29. Septation of unpigmented conidia was also observed. Microconidia, which were observed for two of 10 isolates, were hyaline, aseptate, cylindrical, with rounded or truncate apices, straight or slightly curved, (4.49-) 5.53 to 6.47 (-8.23) µm long, (1.02-) 1.35 to 1.59 (-2.0) µm wide (mean ± SD of 101 conidia = 6.06 ± 0.71 × 1.48 ± 0.18 µm), with an L/W ratio of 4.14 ± 0.62.

    Diplodia sapinea (Fr.) Fuckel, Jb. Nassau. Ver. Naturk. 23-24:393. 1870 (Syn. Diplodia rosacearum S. Giambra, A. Alves, J. Armengol & S. Burruano, Mycosphere 7: 983. 2016). Cultures were initially white, turning light brown from the center of the colony after 3 days, becoming gray-brown in the center and light brown to beige toward the margin of the colony after 6 days, and finally becoming dark gray-brown on the surface and dark brown to black on the reverse after 15 days of incubation. Colonies with moderately dense to dense, fluffy aerial mycelium grew in concentric rings forming a rosette with a slightly undulate margin. Conidia were oblong-elliptical to ovoid, truncate or rounded at the base, rounded at the apex, initially hyaline, becoming pale brown, then brown with age, aseptate, less commonly one- or two- or even three-septate, (20.08-) 22.77 to 25.12 (-29.36) µm long, (9.15-) 11.17 to 12.35 (-14.22) µm wide (mean ± SD of 488 conidia = 23.95 ± 1.69 × 11.72 ± 0.84 µm), with an L/W ratio of 2.05 ± 0.21. Microconidia, which were observed in four out of five isolates, were hyaline, aseptate, elliptical to cylindrical, with rounded ends, (3.21-) 3.88 to 4.34 (-5.92) µm long, (1.15-) 1.4 to 1.64 (-2.09) µm wide (mean ± SD of 364 conidia = 4.15 ± 0.38 × 1.52 ± 0.17 µm), with an L/W ratio of 2.75 ± 0.31.

    Diplodia bulgarica A.J.L. Phillips, J. Lopes & S.G. Bobev, Persoonia 29:33. 2012. Cultures were initially white, turning light olive-brown from the center of the colony after 3 days, becoming light gray-brown at the surface and light brown with brown concentric rings at the reverse after 6 days, and finally turning dark gray-brown on the surface and dark brown on the reverse after 15 days of incubation. Colonies with moderately dense to dense, fluffy aerial mycelium grew in concentric rings forming a rosette with the entire margin. Conidia were elliptical to ovoid, widest in the middle, rounded at the ends, initially hyaline and aseptate, soon becoming light tan, and then dark brown and one-septate, (21.11-) 25.0 to 26.78 (-29.66) µm long, (11.3-) 13.2 to 14.24 (-15.56) µm wide (mean ± SD of 113 conidia = 25.76 ± 1.47 × 13.65 ± 0.81 µm), with an L/W ratio of 1.89 ± 0.16. It has been observed that conidia can form the septum even before they become dark.

    Botryosphaeria dothidea (Moug.: Fr.) Ces. & De Not., Comm. Soc. Crittog. Ital. 1:212. 1863. Cultures were initially white, turning pale olive-brown from the center of the colony after 3 days, becoming dark olive-brown to gray-brown toward the colony margin after 6 days, and finally turning dark gray on the surface and black on the reverse after 15 days of incubation. Isolates formed abundant fluffy aerial mycelium and occasionally grouped in tufts reaching the lid of Petri dishes. The margin of the colonies of all B. dothidea isolates was irregular (undulate to lobate), and this feature distinguished B. dothidea from other Botryosphaeriaceae species found in this study. Conidia were hyaline, fusiform, subtruncate at the base, subobtuse at the apex, mostly aseptate, less commonly one-septate, in some isolates two-septate, (18.31-) 22.32 to 24.75 (-30.09) µm long, (4.56-) 5.64 to 6.58 (-7.73) µm wide (mean ± SD of 488 conidia = 23.44 ± 1.87 × 6.13 ± 0.62 µm), with an L/W ratio of 3.87 ± 0.53. Besides typical fusiform conidia, isolate JBotAv7 also produced Dichomera-like conidia, which were mostly obpyriform and aseptate, rarely one-septate, (10.88-) 12.01 to 14.05 (-16.66) µm long, (5.29-) 6.22 to 6.63 (-7.2) µm wide (mean ± SD of 50 conidia = 13.21 ± 1.43 × 6.4 ± 0.44 µm), with an L/W ratio of 2.07 ± 0.23 µm. Microconidia, which were observed in two of five isolates of B. dothidea, were hyaline, aseptate, cylindrical, with rounded or truncate apices, (4.49-) 5.29 to 6.51 (-8.26) µm long, (1.46-) 1.79 to 2.33 (-3.04) µm wide (mean ± SD of 109 conidia = 5.93 ± 0.83 × 2.12 ± 0.43 µm), with an L/W ratio of 2.85 ± 0.42.

    Neofusicoccum yunnanense G.Q. Li & S.F. Chen, IMA Fungus 11:39. 2020. Cultures were initially white, turning pale olive-brown from the center of the colony after 3 days, becoming dark olive-brown to gray-brown toward the margin of the colony after 6 days, and turning dark gray on the surface and black on the reverse after 15 days of incubation. Colonies with dense, woolly aerial mycelium, frequently grouped in tufts reaching the lid of Petri dishes, had the entire margin. Conidia were hyaline, fusiform to ellipsoidal, subtruncate or subobtuse at the base, subobtuse at the apex, aseptate, frequently with darker middle cell, (11.9-) 15.58 to 17.92 (-21.3) µm long, (4.75-) 6.27 to 6.86 (-8.06) µm wide (mean ± SD of 100 conidia = 16.89 ± 1.77 × 6.56 ± 0.54 µm), with an L/W ratio of 2.59 ± 0.41. Microconidia were hyaline, aseptate, cylindrical, with rounded or truncate apices, (3.81-) 4.62 to 5.36 (-5.95) µm long, (1.35-) 1.67 to 1.94 (-2.45) µm wide (mean ± SD of 105 conidia = 4.97 ± 0.5 × 1.81 ± 0.22 µm), with an L/W ratio of 2.77 ± 0.37.

    Molecular identification and phylogeny of Diaporthe spp.

    With ITS1/ITS4 primer pair the expected amplicons were obtained for all seven isolates, and ITS sequences of 538 to 552 nt were determined. Four isolates (JBotRd3, JBotRd5, JBotSv1, and JBotVT3) had genetically variable ITS sequences (543 to 545 nt) and three ITS variants were observed, which differed from each other in a maximum of seven nucleotides. The first variant (JBotVT3 and JBotSv1) was identical with D. eres isolates CBS 283.85, CBS 109767, CBS 129168; the second variant (JBotRd3) was identical with isolate DAN001A; and the third variant (JBotRd5) was identical with D. eres ex-types CBS 138594 and CBS 439.82. Two isolates (BC2 and JBotAv5) showed no variability in ITS sequences (538 nt) and were identical to D. nobilis isolate CBS 200.39 (Gomes et al. 2013). Additionally, these two isolates differed in ITS1 (positions 79 to 123) and ITS2 region (positions 382-395) from D. eres ex-types CBS 138594 and CBS 439.82 in 28 nt and seven gaps. For all six isolates, five more loci were amplified and sequences of 330 nt for TEF1, 362 to 363 nt for CAL, 493 nt for TUB, 464 to 465 nt for HIS, and 751 nt for Apn2 were obtained. Analysis of these sequences showed genetic variability in all loci. The greatest similarity was detected between isolates JBotRd3 and JBotVT3, which were identical in TEF1, CAL, and TUB but differed in ITS, HIS, and Apn2 (3, 1, and 5 nt, respectively). The next most similar pair were isolates BC2 and JBotAv5 with identical ITS, TUB, and HIS and differences in CAL, TEF1, and Apn2 (1, 6, and 12 nt, respectively). In all other comparisons, a maximum of one locus was found to be identical for two isolates. Generally, the greatest similarity among the obtained isolates was detected in CAL locus, from which the third intron was excluded because of ambiguous alignment (Udayanga et al. 2012), whereas the greatest variability was observed in ITS, followed by Apn2.

    Single locus phylogeny based on different loci produced trees with inconsistent topologies, which made it impossible to draw a unique phylogenetic inference. The only separation of the isolates into D. eres and D. nobilis can be achieved based on ITS phylogeny, but it was not supported by any other single locus phylogenetic analysis. The TEF1 locus, followed by HIS, has shown to be the most informative for species delimitation within D. eres complex (Fig. 5), based on TEF1 phylogeny obtained isolates clustered with D. eres. Therefore, the isolates were identified as D. eres.

    Fig. 5.

    Fig. 5. Phylogenetic tree resulting from analysis of translation elongation factor 1-alpha sequences of Diaporthe eres species complex. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates. Values <60% are marked with ‘‘-’’. Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Diaporthe citri. The scale bar represents two nucleotide substitutions.

    Download as PowerPoint

    Multilocus phylogeny, including ITS as one of four, five, or six loci, showed that the four isolates obtained in this study clustered among D. eres, whereas two isolates clustered in a clade with previously recognized D. nobilis but with low support (Fig. 6). By excluding ITS, these three multilocus analyses D. nobilis and D. eres did not separate, and by including the isolates obtained in this study, all were clustered into a single clade (Fig. 7). This confirmed identification of isolates obtained in this study as D. eres.

    Fig. 6.

    Fig. 6. Phylogenetic tree resulting from analysis of concatenated internal transcribed spacer, translation elongation factor 1-alpha, calmodulin, and beta-tubulin sequences of Diaporthe spp. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates (only values >60% are shown). Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Diaporthe citri. The scale bar represents 20 nucleotide substitutions.

    Download as PowerPoint
    Fig. 7.

    Fig. 7. Phylogenetic tree resulting from analysis of concatenated translation elongation factor 1-alpha, calmodulin, beta-tubulin, and histone sequences of Diaporthe eres species complex. Numbers on the branches represent maximum parsimony and maximum likelihood bootstrap values from 1,000 replicates (only values >60% are shown). Ex-type and ex-epitype isolates are shown in bold. The tree was rooted to Diaporthe citri. The scale bar represents 10 nucleotide substitutions.

    Download as PowerPoint

    Multiple sequence alignment showed that one isolate (JBotRd2) with ITS sequence 552 nt long was the most similar with ex-epitype strain CBS 113201 of D. rudis, barring 2 nt differences. For this isolate, in addition to ITS, three more loci were amplified, and sequences of 316 nt for the TEF1, 450 nt for the CAL, and 474 nt for the TUB were obtained. TEF1 and CAL were identical with the same loci of the ex-epitype strain CBS 113201 of D. rudis, whereas TUB differed in 2 nt. Additionally, comparing JBotRd2 with D. rudis isolate (PS76) previously found in Serbia from soybean (Petrović et al. 2016), 1 nt difference was detected within the ITS region, whereas the TEF1 locus was identical. Isolate JBotRd2 was identified as D. rudis. Multilocus phylogenetic analysis of concatenated ITS, TEF1, CAL, and TUB sequences confirmed the preliminary identification of JBotRd2 isolate, positioning it in the strongly supported D. rudis clade (Fig. 6).

    Statistical parameters describing the sequence alignments and phylogenetic analyses are shown in Supplementary Table S2. MP and ML analyses of the same single or multilocus dataset produced trees with similar topologies.

    Morphology of Diaporthe spp.

    Based on the colony morphology described below, all Diaporthe isolates obtained in this study could be separated into three groups (two within D. eres and one for D. rudis) which was in agreement with ITS phylogeny. Colony and conidial characteristics of the species are shown in Figure 8, and dimensions of conidia in Supplementary Table S4.

    Fig. 8.

    Fig. 8. Morphological characteristics of Diaporthe spp. isolated from apple fruit in Serbia: A1 and A2, Diaporthe eres morphotype 1 15-day-old culture on PDA, A3, sporulation of pycnidia, A4, beta conidia, and A5, alpha conidia; B1 and B2, D. eres morphotype 2 15-day-old culture on PDA, B3, sporulation of pycnidia, B4, beta conidia, and B5, alpha conidia; C1 and C2, D. rudis 15-day-old culture on PDA, C3 and C4, pycnidia on plates acidified with lactic acid, and C5, alpha conidia. Scale bars = 10 µm.

    Download as PowerPoint

    Diaporthe eres Nitschke, Pyrenomyc. Germ. 2:245. 1870. Morphotype 1 (brown colony): Colonies on PDA were initially white with fluffy aerial mycelium, turning pale brown from the center after 3 days of incubation, becoming surface and reverse gray-brown to brown with aging. Colonies grew in concentric rings that formed a dense rosette with an irregular margin. Pycnidia formed on PNA exuded cream to yellow conidial masses after 30 days of incubation. Alpha conidia were ovoid to elliptical, subtruncate or rounded at the base, rounded at the apex, hyaline, aseptate, mostly biguttulate, (5.22-) 6.35 to 7.23 (-10.19) µm long, (1.62-) 2.36 to 2.91 (-4.1) µm wide (mean ± SD of 361 conidia = 6.83 ± 0.7 × 2.63 ± 0.4 µm), with an L/W ratio of 2.64 ± 0.42. Beta conidia were filiform, straight or slightly curved, subtruncate at the base, subacute at the apex, hyaline, aseptate, eguttulate, (17.5-) 22.77 to 32.23 (-38.87) µm long, (1.4-) 1.69 to 1.95 (-2.35) µm wide (mean ± SD of 274 conidia = 27.5 ± 5.5 × 1.82 ± 0.2 µm).

    Morphotype 2 (beige colony): Colonies on PDA were initially white with fluffy aerial mycelium, becoming surface and reverse cream to beige with aging. The colonies grew in concentric rings, with an ill-defined rosette appearance. Pycnidia formed on PNA exuded cream to yellow conidial masses after 30 days of incubation. Alpha conidia were ovoid to elliptical, subtruncate or rounded at the base, rounded at the apex, hyaline, aseptate, biguttulate or multiguttulate, (6.04-) 6.98 to 7.59 (-9.11) µm long, (2.18-) 2.49 to 2.69 (-3.07) µm wide (mean ± SD of 207 conidia = 7.29 ± 0.47 × 2.59 ± 0.16 µm), with an L/W ratio of 2.83 ± 0.26. Beta conidia were filiform, straight or slightly curved, subtruncate at the base, subacute at the apex, hyaline, aseptate, eguttulate, (22.7-) 26.83 to 30.01 (-33.59) µm long, (1.29-) 1.58 to 1.82 (-2.16) µm wide (mean ± SD of 198 conidia = 28.22 ± 2.36 × 1.7 ± 0.17 µm).

    Diaporthe rudis (Fr.) Nitschke, Pyrenomyc. Germ. 2:282. 1870. Colonies on PDA were initially white with fluffy aerial mycelium, turning red-brown from the center after 4 days of incubation, becoming surface and reverse darker in the center and red-brown toward the slightly irregular margin with age. Stromata appeared after ∼10 days on APDA, but exudation of cream to yellow conidial masses from pycnidia was observed after 30 days of incubation. Alpha conidia were ovoid to elliptical, subtruncate to obtuse at the base, rounded at the apex, hyaline, aseptate, biguttulate, (5.33-) 6.26 to 6.74 (-7.56) µm long, (2.1-) 2.27 to 2.46 (-2.7) µm wide (mean ± SD of 100 conidia = 6.5 ± 0.4 × 2.37 ± 0.14 µm), with an L/W ratio of 2.76 ± 0.27. Beta and gamma conidia were not observed.

    Botryosphaeriaceae grow faster than Diaporthe spp.

    Fungal isolates from apple fruit showed a difference in growth rate after 7 days on PDA at 25°C in the dark (Fig. 9). When these species were compared, the fastest growth rate was demonstrated by isolate of N. yunnanense (mean ± SE = 20 ± 0.59 mm/day), followed by isolates of Diplodia: D. seriata (12.29 ± 0.18 mm/day), D. sapinea (9.09 ± 0.16 mm/day), and D. bulgarica (8.36 ± 0.02 mm/day) (P < 0.05). Isolates of B. dothidea showed variability but had a generally slow growth rate of 4.74 ± 0.82 mm/day. Isolates of Diaporthe grew slower compared with isolates of Diplodia species, with a mean growth rate of 6.29 ± 0.07, 5.6 ± 0.3, and 4.56 ± 0.42 mm/day for Diaporthe rudis, D. eres morphotype 1, and D. eres morphotype 2, respectively (P < 0.05).

    Fig. 9.

    Fig. 9. Colony growth rate of Botryosphaeriaceae and Diaporthe isolates on PDA at 25°C in the dark. Vertical bars indicate standard error of mean. Means followed by the same letter are not significantly different based on Student-Newman-Keuls test at P < 0.05.

    Download as PowerPoint

    DISCUSSION

    Based on multilocus phylogeny and morphology, five species from the family Botryosphaeriaceae, Diplodia seriata, D. sapinea, D. bulgarica, B. dothidea, and N. yunnanense, and two from the genus Diaporthe, D. eres and D. rudis, were identified as causal agents of soft, light to dark brown rot of apple fruit in Serbia. Observed symptoms on naturally infected as well as on inoculated fruit were not species specific, and no correlation could be made between the causal agent and the observed symptoms that a given isolate caused on apple fruit. This observation is in accordance with the reports of other authors who described that Botryosphaeriaceae and Diaporthe species cause the same symptoms/diseases on their hosts (Chen et al. 2014a, b). In this study, occurrence of the disease was noticed during harvest but also after harvest, which implies the ability of both groups to cause latent infection on their hosts (Kim et al. 2001; Slippers and Wingfield 2007). Also, it has been noted in this study that warmer storage conditions favor disease development because decay developed in cellars or after transfer from cold storage to indoor environment.

    Among identified apple rot causal agents, the two most commonly isolated were D. seriata and B. dothidea, which at the same time are the only two previously reported Botryosphaeriaceae on apple fruit in Serbia (Stojanović et al. 2003; Vasić et al. 2013). The obtained D. seriata corresponded morphologically to other D. seriata isolated from apple (Delgado-Cerrone et al. 2016; Slippers et al. 2007) and to the morphotype B of B. obtusa from apple in Serbia (Stojanović et al. 2003). Also, the septation of unpigmented conidia and presence of 3-septate brown conidia are consistent with the description of Zlatković et al. (2016). Regarding B. dothidea morphology, conidia size was in accordance with the description by Delgado-Cerrone et al. (2016); however, compared with the ex-epitype (Slippers et al. 2004), a lower L/W ratio was observed. The shape and size of Dichomera-like conidia in this study were similar to those of B. dothidea isolate CMW 39304 (Zlatković et al. 2016) but with the presence of septa, which is consistent with Slippers et al. (2014). We observed genetically uniform population of D. seriata and B. dothidea in Serbia, regardless of the host. Analysis of three loci (ITS, TEF1, TUB) revealed the identity of obtained and previously described Serbian D. seriata and B. dothidea from pome fruit as well as ornamental and forest trees (Vasić et al. 2013; Vico et al. 2017; Zlatković et al. 2016).

    Diplodia bulgarica is a novel find in Serbia after being described in Bulgaria and Iran as an apple pathogen (Phillips et al. 2012). Molecular data and morphological characteristics of obtained D. bulgarica are in accordance with the description of this species by Phillips et al. (2012), apart from conidial width, which was slightly smaller than the conidia of the Serbian isolate. Furthermore, as a contribution to D. bulgarica morphology, septation of unpigmented conidia was observed in this study.

    Some host association to family Rosaceae in Diplodia species was observed by Phillips et al. (2012). In resolving the Diplodia complex on Rosaceae hosts, Phillips et al. (2012) separated D. intermedia from D. sapinea based on host affiliation, smaller conidia, and phylogeny of ITS and TEF1. Also, a pathogen of Rosaceae, D. rosacearum, described as a new species in Italy, Spain, and Bulgaria, was separated from D. intermedia based on smaller conidia and phylogeny of ITS and TEF1 (Giambra et al. 2016). In the latest evaluation of Botryosphaeriales, Zhang et al. (2021) reduced D. intermedia and D. rosacearum to synonyms of D. sapinea based on phylogeny. In accordance with this newest taxonomy, isolates obtained in this study from apple, quince, and medlar fruits were identified as D. sapinea. Nevertheless, morphology and phylogeny of the Serbian isolates, which are consistent with those of D. rosacearum (Giambra et al. 2016), differed from D. sapinea and D. intermedia. Concerning morphology, D. sapinea conidia exceed 50 µm in length and are aseptate (rarely 1-septate), D. intermedia conidia are >28 µm long and are also aseptate (Delgado-Cerrone et al. 2016; Phillips et al. 2012, 2013), whereas conidia of Serbian and D. rosacearum isolates (Giambra et al. 2016) are shorter (23.95 µm), mostly aseptate, but also 1, 2 and 3-septate. Observed differences in conidial morphology were supported by multilocus phylogeny (ITS, TEF1, and TUB). Additionally, isolates from different pome fruits obtained in this study differed molecularly from typical D. sapinea from ornamental and forest trees in Serbia (Zlatković et al. 2016). Keeping in mind the observed differences, there might be some divergence within D. sapinea that is reflected in the affiliation toward specific hosts. Additionally, the description of microconidia is a contributive, diagnostic finding for this species.

    Neofusicoccum yunnanense, the fastest growing Botryosphaeriaceae isolate from apple fruit, was primarily identified as N. parvum. This was based on the sequence identity of the obtained isolate in ITS, TEF1, and TUB with Serbian N. parvum from ornamentals and the highest similarity with N. parvum ex-type (Phillips et al. 2013; Zlatković et al. 2016). Nevertheless, multilocus phylogeny revealed some separation of Serbian isolates from other N. parvum isolates. Numerous cryptic species within the N. parvum/N. ribis complex were discovered (Pavlic et al. 2009; Sakalidis et al. 2011; Slippers et al. 2004), and among them, N. yunnanense, a new species phylogenetically closely related to N. algeriense, N. italicum, and N. parvum, was described (Li et al. 2020). Our isolate was molecularly identical (ITS, TEF1, and TUB) with N. yunnanense, and multilocus phylogeny clustered them separately from N. parvum. In the last reassessment of the Neofusicoccum species (Zhang et al. 2021), N. yunnanense was not included in the research, but species closely related to N. yunnanense (N. algeriense and N. italicum) were reduced to synonymy with N. parvum. With regards to molecular identity, we named our isolate N. yunnanense, but with regards to the new developments in Neofusicoccum taxonomy, this species might soon be reduced to synonymy with N. parvum.

    In this study, species from the genus Diaporthe are newly described as causal agents of apple fruit rot in Serbia. Generally, species of Diaporthe cannot be separated based on conidial morphology because of overlap in the shape and size of conidia between species (Udayanga et al. 2011). Diaporthe eres and D. rudis were identified based on molecular phylogenetic analyses of six and four loci according to Udayanga et al. (2014a, b). Multigene genealogical approach is currently accepted in taxonomy and species delimitation in the genus Diaporthe. Based on this approach, the status of D. eres and closely related species was defined in the D. eres complex (Udayanga et al. 2014a). The multilocus phylogenetic tree that was generated, devoid of ITS, strongly confirmed the supported 10 clades in D. eres complex, which correspond to 10 species: nine proposed by Udayanga et al. (2014a) and one recognized by Guarnaccia et al. (2018), and clustered our isolates from apple fruit within D. eres clade. Phylogenetic inference derived from analysis of TEF1, proposed as a universal secondary DNA barcode (Stielow et al. 2015), corresponded to those obtained by ITS-free multilocus phylogenetic analyses, indicating that TEF1 has highest species resolution power within D. eres species complex, which was also observed by Santos et al. (2017a) and Udayanga et al. (2014a). The ITS region in Diaporthe eres has great variability, as observed in ITS1 and ITS2. ITS variability was found within the same host and in the same geographic region and resulted in limiting resolution phylogeny (Santos et al. 2010; Udayanga et al. 2014a). It was also observed in obtained isolates from apple, and by comparing those with Serbian isolates from soybean (Petrović et al. 2015), the same variability was noted. Interestingly, two obtained D. eres isolates that differed in colony color from other D. eres isolates had ITS identical or the most similar with previously described D. nobilis (CBS 200.39, CBS 113470, CBS 116954, CBS 124030) and differed from ex-types of D. eres (CBS 138594, CBS 439.82) in ITS1 (positions 79-123) and ITS2 (positions 382-395) in ∼30 nt. Other D. nobilis (CBS 587.79 and CBS 116953) analyzed in this study differed from ex-types of D. eres only in ITS1 in ∼20 nt (data not shown). Each of our multilocus phylogenetic analysis based on four, five, or six loci including ITS was able to resolve ex D. nobilis group from D. eres, placing them in two distinct but low supported clades. D. nobilis complex proposed by Gomes et al. (2013) based on multilocus phylogeny (ITS, TEF1, CAL, TUB, and HIS) was included later in the Diaporthe eres complex delimitation, which excluded ITS, consequently was considered superfluous and was thus reduced to a synonym of D. eres (Udayanga et al. 2014a).

    The most novel finding from this study is the description of Diaporthe rudis as a postharvest pathogen of apple fruit in the world. Investigating the association of this fungus with the apple as a host, only one report was found in the literature, and that is D. medusaea (synonym of D. rudis) on Malus pumila var. domestica in Japan (Farr and Rossman 2021). In recent years, D. rudis has been increasingly reported as a postharvest pathogen of kiwifruit, grape, and pear fruit (Diaz et al. 2017; KC and Rasmussen 2019; Lorenzini and Zapparoli 2019). Interestingly, D. rudis (JBotRd2) and D. eres (JBotRd3) were isolated from the same apple fruit, thus confirming the fact that multiple species of Diaporthe can co-occur on the same host.

    Current study showed that a diverse group of fungal species belonging to the family Botryosphaeriaceae and the genus Diaporthe can cause light to dark brown rot on apple fruit. Through detailed phylogenetic and morphological analyses, Diaporthe rudis is described as a new postharvest pathogen of apple fruit worldwide, whereas Diplodia bulgarica, D. sapinea, N. yunnanense, and Diaporthe eres are initially described as postharvest apple and D. sapinea as postharvest quince and medlar fruit pathogens in Serbia. The occurrence of two known species, Diplodia seriata and Botryosphaeria dothidea, and several new species on apple fruit during or after harvest implies the expansion and growing importance of both groups as postharvest pathogens. Additionally, species from the family Botryosphaeriaceae and the genus Diaporthe have an endophytic nature, which shifts into a pathogenic lifestyle under stress conditions (Sakalidis et al. 2021; Slippers and Wingfield 2007). Coupled with their wide host range, these pathogens are becoming a new challenge for successful fruit production. Because species of the family Botryosphaeriaceae and the genus Diaporthe are known to cause other disease symptoms on their hosts (leaf spot, shoot blight, branch and stem cankers, dieback, even tree death), some of which can become severe, future research should be aimed at the questions regarding the occurrence and importance of identified Botryosphaeriaceae and genus Diaporthe in orchards as well. Comprehensive knowledge of these pathogens will enable improvement in apple disease management and increase the quality of apple production in Serbia.

    ACKNOWLEDGMENTS

    We thank Aleksandra Žebeljan and MiljanVasić for contributing to the sample and isolate collection.

    The author(s) declare no conflict of interest.

    LITERATURE CITED

    Funding: This work was supported by Ministry of Education, Science, and Technological Development of the Republic of Serbia grant 451-03-9/2021-14/200116.

    The author(s) declare no conflict of interest.