Colletotrichum is widely recognized for inducing mango disease with potential yield losses of up to 100% (Li et al., 2019; Sharma and Shenoy, 2016). In Indonesia, mango dieback cases linked to Colletotrichum infection have been particularly unreported where reviews have predominantly focused on mango fruit anthracnose disease particularly in Java Island, as the mango production center (Benatar et al., 2020). Despite the Colletotrichum affecting various plant parts, significant yield losses still manifest during the fruit-ripening stage when anthracnose symptoms become apparent (Widiastuti et al., 2023). This study aims to identify and characterize Colletotrichum spp. causing mango dieback in Indonesia which needs further concern for biosecurity.

For this study, symptomatic dieback twig samples were purposefully collected from various mango production centers in Java, Indonesia (Fig. 1). The sample had necrosis on twigs, petioles, and leaves, and the necrotic lesion initiated with irregular blackish-brown discoloration, expanding to cause dried areas (Fig. 2A-C).

Infected twig pieces were subjected to disinfection with 70% ethanol for 1 min, followed by 1% sodium hypochlorite (NaOCl) for 1 min. They were then rinsed twice with sterile ddH₂O and dried using sterile filter paper. Two pieces were placed into potato dextrose agar (PDA) medium with 5 μl lactic acid, and the mixture was incubated at 28°C for 3 days. The resulting colonies, representing a single conidium of Colletotrichum, were recultured and incubated under the same conditions (Choi et al., 1999). In this study, 11 Colletotrichum spp. isolates were obtained from different cultivars, consisting of Manalagi, Arumanis, Gadung, Golek, and Gedong Gincu (Table 1).

Genomic DNA was extracted from pure mycelia of Colletotrichum isolates grown in PDA at room temperature 25°C for 7 days using the Taiwan Geneaid Genomic DNA Mini Kit (Plant) (Geneaid Biotech Ltd., New Taipei, Taiwan), following the manufacturer’s protocol. The genetic diversity fingerprints of Colletotrichum spp. were assessed through rep-PCR (repetitive element sequence-based PCR), using two primer sets, including BOX and ERIC. The PCR amplification and visualization method is based on Pramunadipta et al. (2022). PCR mixture consisted of 12.5 μl (dNTPs; MgCl₂; 2× MyTaq HS Red Mix, Bioline, Meridian Bioscience, Cincinnati, OH, USA), 1 μl for each primer (100 μM), 2 μl template DNA isolates, and 8.5 μl ultrapure water, resulting in a total volume of 25 μl. Rep-PCR using BOX (5′-CTACGGCAAGGCGACGCTGACG-3′), ERIC1 (5′-ATGTAAGCTCCTGGGGATTCAC-3′), and ERIC2 (5′-AAGTAAGTGACTGGGGTGAGCG-3′) was conducted with the annealing temperatures set to 53°C for BOX and 52°C for ERIC. Polymorphic bands, observed between 100 and 3,000 base pairs, were shown in Fig. 3. A dendrogram was constructed using Unweighted Pair Group Method with Arithmetic Mean (UPGMA) analysis on rep-PCR based on the presence bands amplified with BOX and ERIC primers in Ntsys 2.2 software. Several isolates represented each banding pattern identical group, and 11 isolates were divided into four major groups based on a similarity coefficient of 70% (Fig. 4). The first group, containing only one fungal, was distinct from others (groups 2, 3, and 4), suggesting that this fungal isolate belonged to different clades of Colletotrichum species complex. The representative isolates from each group were chosen for molecular identification. Notably, for the 2nd and 4th groups, multiple isolates were selected to represent the origin of sample accurately. MLG-TJB, HM-IJB, HM-BDIY, MLG-TJT, MLG-PAJT, and HM-PAJT were selected for phylogenetic analysis using partial sequences of four gene regions (internal transcribed spacer [ITS], partial actin [ACT], glyceraldehyde 3-phosphate dehydrogenase [GAPDH], and β-tubulin [TUB2]). Patricia et al. (2021) mentioned that the absence and presence of band information were used to build UPGMA tree to compare the amplification profile of Colletotrichum strains. Genetic variability accessed by rep-PCR analysis constructed into a dendrogram phylogenetic also supports the genetic differences among the Colletotrichum isolates.

The ITS region, TUB2, ACT, and GAPDH genes, were amplified using set primers of ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′), BTUB2fd (5′-GTBCACCTYCARACCGGYCARTG-3′) and BTUB4rd (5′-CCRGAYTGRCCRAARACRAAGTTGTC-3′), ACT512F (5′-ATGTGCAAGGCCGGTTTCGC-3′) and ACT783R (5′-TACGAGTCCTTCTGGCCCAT-3′), GDF1 (5′-GCCGTCAACGACCCCTTCATTGA-3′) and GDR1 (5′-GGGTGGAGTCGTACTTGAGCATGT-3′). PCR amplification and visualization adjustment were performed based on Wang et al. (2021), with annealing temperatures were set to 52°C, 58°C, 60°C, and 55°C for ITS, ACT, GAPDH, and TUB2, respectively. The results of PCR amplification on six isolates showed that all isolates amplified in the range of 600 bp for the ITS and 500 bp for the target β-tubulin region. Meanwhile, the target areas of the Actin and GAPDH were amplified in the range of 300 bp. The phylogenetic tree was constructed from concatenated ITS, ACT, GAPDH, and TUB2 sequences of Colletotrichum spp. samples and reference Colletotrichum isolates from GenBank (Table 2) using the maximum likelihood method with the Kimura 2-parameter model, and it was tested by 1,000 bootstrap replications.

The phylogenetic tree (Fig. 5) showed that six representative isolates formed two main groups with a high bootstrap value exceeding 50%. HM-BDIY, MLG-PAJT, and HM-PAJT had close relations to Colletotrichum asianum CMM4067, while HM-IJB and MLG-TJT clustered with C. asianum CMM4056. On the other hand, MLG-TJB was separated and clustered with Colletotrichum cairnsense AC11. It was observed that C. asianum belonged to Colletotrichum gloeosporioides species complex, whereas C. cairnsense belonged to Colletotrichum acutatum species complex (de Silva et al., 2017). The identification was in line with the genetic fingerprint results obtained from rep-PCR. The sequences of the six isolates (24 sequences in total) have been deposited in GenBank with the following accession numbers, ITS: OR945809-OR945814, actin: PP091732 to PP091736 and PP091747, GAPDH: PP091737 to PP0091741 and PP091748, TUB2: PP091742 to PP091746 and PP091749.

Macroscopic and microscopic morphological observations were conducted on 10-day-old colonies. In general, Colletotrichum isolates produced light to dark olive-grey mycelium, with the center of colonies covered by orange conidia masses (Uysal et al., 2022). This study recorded a mycelial growth rate of 7.24 mm/d to 7.89 mm/d, as shown in Table 3. C. asianum had cottony mycelium, predominantly white, occasionally white grayish with sparse aerial hyphae (Fig. 6A). In this study, isolates identified as C. asianum exhibited morphological characteristics that closely matched those described by Benatar et al. (2020). The authors used the morphology of the MLG-TJT to represent C. asianum isolates in this figure, as they were the closest to the reference Meanwhile, C. cairnsense had white-greenish cottony mycelium (Fig. 6F). De Silva et al. (2017) also mentioned that C. cairnsense had a pale white-grey to olivaceous grey mycelium with a whitish margin. The reverse side of both species was white, typically forming concentric rings and dark spots as mature conidiomata developed (Fig. 6B and G). Additionally, both species produced masses of conidia (conidiomata) as presented in Fig. 6C and H.

Microscopic characterization was observed using an Olympus CX31 microscope (Tokyo, Japan) and Miconos Optilab (PT Miconos, Yogyakarta, Indonesia) and it showed that C. asianum and C. cairnsense produced hyaline, smooth-walled cylindrical conidia with distinguishing features. For instance, C. asianum had obtuse ends, and the conidia of C. cairnsense had acutely rounded ends (Fig. 6D and I). The length and width of the conidia were measured by randomly selecting 50 conidia using a calibrated ImageRaster software (Ramos et al., 2016). Conidia measured 7-18.6 μm in length, 2.3-6.9 μm in width, and had a length/width ratio of 2.56-3.23 (Table 3). The formation of appressoria was observed using slide culture technique by growing the isolates on PDA (Siddiquee, 2017). Both C. asianum and C. cairsense formed appressoria originating from mycelia, characterized by dark brown, ovoid to irregular structures found at the ends of hyphae. According to Prihastuti et al. (2009), in slide culture, C. asianum formed appressoria that were brown to dark brown, ovoid, clavate to irregular in shape and often became complex with age. In contrast, de Silva et al. (2017) reported that C. cairnsense formed appressoria single that was medium brown, smooth-walled, subglobose, ovoid to ellipsoidal. In this study, only C. asianum produced setae that were perpendicular in shape and dark brown (Fig. 6E and J). C. asianum was initially reported in Indonesia by Benatar et al. (2020) in connection with mango anthracnose in Indramayu. On the other hand, C. cairsense has not been previously recorded in the country but was first identified in Australia. C. cairnsense sp. nov. was recognized as the causative agent of chili anthracnose by de Silva et al. (2017). According to the morphology observed in this study. The conidia of C. gloeosporioides species complex were cylindrical with obtuse ends, while C. acutatum species complex had cylindrical conidia with two acute ends or one end slightly obtuse.

Pathogenicity test was conducted on 8-month-old mango seedlings, specifically cv. Arumanis, with an average height of approximately 50 cm per seedling. cv. Arumanis were selected for this pathogenicity test because these cultivars are widely cultivated in Indonesia and frequently show dieback symptoms due to Colletotrichum infection, despite the lack of official reports documenting these cases. The test was performed exclusively on representative isolates identified through phylogenetic analysis, and all isolates of both C. asianum and C. cairnsense showed pathogenicity (Fig. 7B and D). Moreover, inoculation followed the method outlined by Mayorquin et al. (2019) with modifications. An active hyphal plug (approximately 5 mm in diameter) from a 14-day PDA culture of Colletotrichum spp. was placed on the twig tip, creating wounds up to 3 mm in diameter and 2 mm in depth, and covered with a moist sterile tissue then sealed. Control treatments were carried out by placing a sterile plug on the seedling with identical wounds. This process was repeated four times and incubated at greenhouse temperatures (approximately 28°C). A necrotic lesion was observed behind dieback lesion (Fig. 7C and E), with an orange-to-black conidial mass visible on the necrotic area (Fig. 7D). The control twig showed no symptoms (Fig. 7A), and the virulence assessment measured the lesion area on the 20th day after inoculation. All six representative isolates had virulence, with TJT-MLG (188 mm²) and PAJT-HM (119 mm²) identified as the most aggressive isolates (Fig. 7G).

This study found that C. asianum was in line with several observations, indicating the isolates were considered the most significant fungal associated with mango diseases. According to Benatar et al. (2020), C. asianum was a virulent pathogen causing mango anthracnose in Indramayu, Indonesia. Vitale et al. (2020) also mentioned that C. asianum was a common species responsible for mango anthracnose, reported in various countries such as Australia, Brazil, China, Ghana, Japan, Malaysia, Mexico, Panama, Philippines, South Africa, Sri Lanka, Thailand, and Florida. Other species in C. acutatum species complex, particularly C. cairnsense, were newly recognized causes of mango diseases, despite previously being identified to be pathogens on Chili in Australia.

In conclusion, dieback studies associated with Colletotrichum infection were significantly scarce, but this study provided a new examination of mango dieback in Indonesia. It represented the new documentation of C. asianum and C. cairnsense as the causative pathogens. Further investigations were required to explore and quantify the occurrence and intensity of mango dieback caused by Colletotrichum in Indonesia. Accuracy in diagnosing causal pathogens was crucial for the efficacy of disease management.