The study was conducted in the Eucalyptus growing areas of Bedelle (BE), Jima (JI), Abru (AB), Woreta (WO), Humbo (HU), Ana lamo (AN), and Masken (MA), districts in Oromia, Amhara and Southern Nations Nationalities and Peoples (SNNP) regions of Ethiopia. In the Amhara region, the average annual minimum and maximum tem-peratures range from15°C and 21°C, the average annual rainfall is 1,194 mm, and the altitude ranges from 500 to 4,620 meters above sea level (Bewket, 2009). The average annual minimum and maximum temperatures; and rainfall ranged from 18°C to 39°C and 450 to 820 mm, respectively, while the altitude ranged from 500 to 4,607 above sea level in the Oromia region (Bewket, 2009). In similar cases, the mean annual rainfall in the SNNP region ranges between 200 and 2,200 mm, and the mean annual temperature ranges from 15°C to 26°C while the altitude range in the region is 375 to 4,207 meters above sea level (Gebremichael et al., 2014). Details of the study areas are given in Fig. 1.

The disease survey was conducted in the Eucalyptus plantation areas in Ethiopia during 2019/2020 and 2020/2021. The survey and sampling were carried out at nine locations in the regions. Two plantation sites located 3‒5 km apart, were randomly selected for disease assessment. Plantation sites were selected on the basis of agroecology, road access, and productivity. Five Eucalyptus trees were randomly selected from each plantation and examined for disease symptoms. Samples were collected from the Eucalyptus plant with symptoms of cracking of the stem and branches, cankers, and tip dieback as described by Li et al. (2018), Masood et al. (2010), and Pillay et al. (2013). The collected samples were placed in paper bags and transported to the laboratory for identification.

The incidence and severity of stem and branch canker disease were determined using the disease index described by El-Gali and El-Zahaf (2015) and Masood et al. (2010). The percentage of disease incidence was calculated using the following formula: Percentage of the canker disease incidence=(No. of diseased trees/Total No. of trees)×100.

Disease severity of stems and branch canker was determined using a scale of 0‒7, where 0=no sign of canker, 1= 1‒10% of canker over the entire observed area, 2=11‒20%, 3=21‒30%, 4=31‒40%, 5=41‒50%, 6=51‒60%, and 7=61% and above were considered as maximum canker severity. Disease severity was calculated using the formula:

Percentage of the canker diseases severity=(Σ(n×v)×100)/(N×S), where Σ=Sum, n=number of trees in each category, v= numerical value of the observed trees, N=total number of examined trees, and S=the maximum canker severity numerical value.

For the isolation and purification of fungal cultures, diseased sample segments were sterilized in 70% ethyl alcohol for 1 min and in 2% sodium hypochlorite for 2 to 4 min, washed three times with sterile distilled water, dried on sterile filter paper, aseptically transferred to malt extract agar (MEA) and kept at room temperature (25‒27°C) for 5 to 15 days (Ahmadpour et al., 2017). The hy-phal tip transfer method was used to subculture and obtain pure isolates. Alternatively stems and branches with clear cankers symptoms were sequentially surface sterilized in 70% ethyl alcohol for 1 min and in 2% sodium hypochlorite for 2 to 4 min, washed three times with sterile distilled water, dried on sterile filter paper and incubated in moist chambers to induce the development of a fungal fruiting body. After appearance and maturity of fruiting bodies monosporic cultures were generated by plating a spore taken from pycnidia suspended in sterile water on MEA. Pycnidia and conidia produced on stems and branch samples were mounted on microscope slides using distilled water and the structures of conidiomata, ascospores, and conidia were examined under the Olympus BX53 digital fluorescence microscope (Tokyo, Japan) as described by Chen et al. (2015).

Isolates were cultured and subcultured to obtain the desired purity for morphological studies. In cultures, characteristics such as colony diameter, color, texture, concentric zone, fruiting body shape, and conidia; shape, septation, pigmentation, and size were studied (Aveskamp et al., 2010; Noordeloos et al., 1953; Silva et al., 2020). All pure fungal isolates were classified according to their morphological characteristics and finally identified at the species level using internal transcribed spacer (ITS) rDNA sequence analysis.

DNA extraction was performed from 10-day-old cultures of fungal isolates, as described by Li et al. (2020). The harvested fungal mycelia were pulverized in liquid nitrogen, lysed with buffer, then dis-solved in phenol, chloroform, and isoamyl alcohol, and finally, the desired DNA was recovered by an ethanol precipitation method. The DNA base concentration was determined using a NanoDrop spectrophotometer (ND-1000, NanoDrop Technologies, Wilmington, DE, USA). To measure the quality of the obtained DNA, electrophoresis was performed on a 1% agarose gel (Raja et al., 2017). The ITS region of the DNA sample was amplified using the first internal transcribed (ITS1) and the second ITS region (ITS2) primers as described by White et al. (1990).

The poly-merase chain reaction (PCR) products were cleaned using a High Pure PCR Product Purification Kit (Roche Molecular Biochemicals, Mannheim, Germany) and sequenced in both directions with the same primers as used for PCR. The ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer Applied Biosystems, Foster City, CA, USA) was used to perform the sequencing reactions and the sequences were run on an ABI PRISM 377/3100 Autose-quencer (Perkin-Elmer Applied Biosystems) as described by White et al. (1990) at the Netherlands Microbiology Research Centre (MRC) and the Center for Agriculture and Bioscience International (CABI), UK using forward and reverse primers. ITS sequences obtained were trimmed using BioEdit v. 7.0.9.0 (Hall, 1999), and alignments for each sequence were determined using the Basic Local Alignment Search Tool (BLAST) in the NCBI database (National Center for Biotechnology In-formation, National Library of Medicine, Bethesda, MD, USA) for identification. Consensus sequences were assembled in Molecular Evolution Genetic Analysis (MEGA) version 10 software (Kumar et al., 2018). Additional known sequences of genera Didymellaceae and outgroup were obtained from GenBank and the phylogenetic tree was constructed using the neighbor-joining analysis algorism. All positions containing gaps and missing data were eliminated. The Nearest-Neighbor-Interchange maximum likelihood heuristic search option was used for tree inference. Bootstrap analysis was performed using 1,000 replicates and evolutionary distances were calculated using the maximum composite likelihood method (Tamura et al., 2004)

The pathogenicity potential of the fungal isolates was tested using a 7-day culture in MEA. The mycelial plugs of each fungal species, cut to a size of 5.0 mm, were inoculated into five seedlings of the E. camaldulensis host plants through small wounds (2 to 3 cm) drilled into the stems with a scalper, at a depth sufficient to expose the xy-lem, approximately 30 cm above the collar root. Inoculation was done by placing the mycelial plugs upside down into the wound and covering it with masking tape. Sterile MEA discs free of fungal samples were used to inoculate the corresponding controls for comparison. The inoculated seedlings were carefully examined for symptoms of canker disease. After 4 months, canker symptoms were fully observed on the stems of the corresponding Eucalyptus seedlings. All resulting lesions were measured, sampled, cultured, re-isolated, and identified to confirm Koch's postulate (Pérez et al., 2009).

The fields being studied were domi-nated by E. camaldulensis grown at their respective altitudes and often in non-agricultural rocky areas with infertile and compacted soils. Stem and branch canker diseases were observed, widely incident and severe in all plantation fields (Fig. 2). The E. camaldulensis plants were found to be relatively stressed in low-altitude areas. The minimum and maximum altitudes of the survey fields are 1,563 meters Abru (Abyotfire) and 2,254 meters Masken (Dirama) above sea level (Table 1). The highest, mean severity (90%) and incidence (100%) were recorded in the Amhara region, North Wollo, Abru district, and Abyot fire locality while the lowest average severity (45%) and incidence (66.2%) were recorded in the SNNP region, Gurage zone, Masken district, Dirama locality (Table 1). Stem and branch canker disease was recorded as more severe and incident at lower altitudes (Table 1).

The occurrence and severity of stem and branch canker diseases were found to vary significantly (P<0.01) in the survey fields (Table 1). The results of the correlation test between the disease parameters (severity and incidence) of the canker diseases and the altitude of the study area show a significant negative relationship (r=‒0.82) at the 1% probability level. The result implies that the incidence and severity of stem canker diseases increase by 0.82 and 0.89 with a unit decrease in field altitude (Table 2).

The result of the regression analysis using a linear model that relates the locations of the study areas as independent variables with the disease parameters as dependent variables show variables such as altitude in the surveyed areas influence (P<0.01) the incidence and severity of the disease with adjusted R²=0.87 and R²=0.99, respectively (Table 2). The result implies that environmental variables in the plantation fields have a strong impact on the significant variations in disease parameters in the study areas.

A total of 400 (320 branches and 80 stems) diseased samples of E. camaldulensis were collected from the survey fields. The fungal cultures were subcultured using the single-hyphal transfer method and several pure fungal isolates were obtained. A total of 120 fungal species were recovered from the collected samples. Isolation and morphological characterization of the fungal isolates were done using colony color, shape, zo-nation, structures of conidiomata, ascospores, and conidia morphs like shape and septation as indicated by Gezahgne et al. (2003). The bryoid and dendritic crystals formations associated with cultures were also used for morphological study as presented by Noordeloos et al. (1953). Most of the isolates (80%) were found Botryosphaeria, while the remaining 40% were Aplosporella and Didymella species described in this study. The Botryosphaeria and Aplosporella species studied in the research were submitted for publication. Research reports indicate that Didymella fungal species are currently the cause of emerging tree diseases in the world and that is why due consideration was given to the study (Chen et al., 2015). The fungal isolates in this study were cultured on MEA and the morphological characteristics were studied for three weeks as described by Noordeloos et al. (1953). The results of the fungal morphological character study revealed that the fungal growth was slow and the average radial diameter was recorded 75‒85 mm after 2 weeks. The colony was pale olivaceous gray to gray olivaceous, zonate, and/or with sectors; aerial mycelium was compact, white to gray, and flocculent on the front and reverse sides. The bryoid and dendritic crystals associated with the olivaceous black centres were formed after one week on the MEA (Fig. 3). Micromorphological observations were made using an Olympus BX53 digital fluorescence microscope. Pycnidia were abundant in concentric zones after crystals 3 weeks, but also scattered in the aerial mycelium and agar; black, globose or irregular; ostiolate solitary or mixed, sphere to pyriform, apical openings, superficial, brown to black with variable shape and size (Fig. 3). It was found that the conidia were light, ovate, aseptate, variable, smooth, and with slightly point-shaped formations at the tip (Fig. 3). The observed morphological characteristics were consistent with the genera of Didymella. The result agrees with the morphological features reported by Ahmadpour et al. (2017), Aveskamp et al. (2010), and Chen et al. (2015). The identification of the isolates to the key species level was performed by sequence analysis and phylogenetic study (Fig. 4).

The species, host, origins, and GenBank accession numbers, strains used for the phyllo-genetic study are displayed in Table 3. The phylogenetic tree was generated using the analysis of maximum likelihood (ML) algorithm (Fig. 4). The topology of the phylogenetic tree showed consistent relationships between the supported clades. Tree topology and phylogenetic analysis revealed that the fungal isolates in this study were associated with Didymella species (Fig. 4). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates) are shown next to the branches. The tree is drawn to scale, with lengths next to the branches. The evolutionary distances were calculated using the maximum composite likelihood method (Tamura et al., 2004). Fungal species with close overlapping morphs and known outgroups were also included in the phylogeny for the reliability of the identification. This analysis was performed using 23 nucleotide sequences, and evolutionary analysis was performed in MEGA 10 (Kumar et al., 2018).

The best-fitting ML nucleotide substitution model for phylogenetic analysis of three-locus and four-locus com-bined datasets were selected Tamura-Nei (TN93+G+I) and General Time Reversible (GTR+G+I), respectively (Tamura, 1992). The phylogenetic tree of the dataset based on the ML analysis is shown in Fig. 4. The topology of the phylogenetic tree showed consistent relationships among the strongly supported clades. The tree topology of the phylogenetic analysis revealed that the isolates under study were associated with the species D. coffeae-arabicae and D. pinodella. In the tree, the isolates in our study (E864010FR002, 10B03-1; E864010FR003, 10B03-2; E864010FR004, 10B03-3) and (Q203241044, G11042; Q203241045, G11043) generated supportive monophyletic clades of 91 and 95% confidence levels with a reference strain from GenBank, Didymella coffeae-arabicae (MH863293, CBS 123380), and D. pinodella (MT274580:18-484, B36) (Fig. 4).

Based on the ITS sequence alignment in the BLAST search, the Didymella isolates in this study were 100% matched in sequence with D. coffeae-arabicae (GenBank no. MH863293) while the other fungal sequence in this study showed 99.79% identity with reference Didymella species in the NCBI GenBank (GenBank no. MT274580:18-484) (Fig. 4).

In this study, both D. pinodella and D. coffeae-arabicae isolates in the study caused lesions (Fig. 5) a month after inoculation of E. camaldulensis. There were no significant differences in the length of the lesion caused between the two species (P>0.01) (Table 4). Samples were collected, cultured in MEA, isolated, morphological characteristics were compared, and the results were consistent with the morphologies of isolation obtained from field samples. After the full onset of canker symptoms, small pieces of necrotic samples taken from the edges of the lesions were surface sterilized, cultured on the MEA, re-isolated and compared with morphological features of cultures obtained from diseased E. camaldulensis samples in the survey fields. The results were found consistent with the morphologies of the isolates obtained from the surveyed fields.