Description and Pathogenicity of Colletotrichum kapreanum sp. nov, a Cherelle Wilt Pathogen Belonging to the Gigasporum Species Complex

Similar to cacao pod rot, cherelle wilt decreases production from cacao fields. Among all known fungal pathogens of the cacao, Colletotrichum spp. are common infectious agents that affect the cherelles and pods of cacao; thus, cacao diseases are often classified by stage. Therefore, knowing whether these pathogens are common in both fruit stages is necessary for implementing disease control measures. Symptoms of cherelle wilt were found in cacao plants in Pangasinan, Philippines, in 2022. The fungal strain obtained from the lesion was found to be pathogenic towards cherelles, but not towards pods. The strain was classified as an unknown species belonging to the gigasporum species complex, based on the morphological and molecular phylogenetic analyses of ITS, GAPDH, CHS1, ACT, and TUB2. We propose Colletotrichum kapreanum sp. nov. as a causal agent of cacao cherelle wilt, but not pod rot.


Introduction
Cacao (Theobroma cacao L.), the source of cacao beans used to make chocolate, is an important cash crop in tropical countries [1,2].In the Philippines, the majority of the cacao farms are smallholdings owned and managed by farmers.According to Philippine Statistics Authority (PSA) data, the Philippines' cacao production has been slowly but steadily increasing, by an average of 2743 hectares per year, from 2013 to 2020.Though production is increasing, the country cannot still meet the increasing market demand of cacao locally and globally.The low production of cacao in the Philippines is mainly due to the high mortality rate of planting materials and the lack of knowledge and training specifically on good agricultural practices (GAP) for cacao farmers, resulting in improper farm management, which can result in low yield and susceptibility of cacao plants to pests and diseases [3].
Cacao pods are classified by size, into either 'cherelles' and 'pods.'Cherelles are defined as pods with a length <10 cm, while pods are >10 cm.Cacao diseases are classified based on the pod stage or morphology.Cacao pod rot is caused by several pathogens; thus, it is difficult to identify based solely on its symptoms.However, some pathogens that cause diseases in pods have their own disease names; for example, black pod rot and brown pod rot are caused by Phytophthora spp.; frosty pod disease is caused by Moniliophthora roreri (Cif.)H.C. Evans, Stalpers, Samson & Benny; and anthracnose is caused by Colletotrichum spp., all of which are diseases that have generated research interest.However, 80% of cherelle wilt cases are thought to be caused by physiological disorders [2]; thus, little research has been conducted on the relevant infections.Therefore, cherelle wilt is a disease that farmers are unfamiliar with as a control target.
Anthracnose disease in cacao, caused by Colletotrichum spp., manifests as lesions on the leaves and pods [4,5].To date, all recorded causative species belong to the Colletotrichum gloeosporioides species complex (CGLSC), which comprises C. gloeosporioides, C. siamense, C. theobromicola, C. tropicale, and C. aeschynomenes [4,[6][7][8][9].Among these, C. tropicale is pathogenic to cacao pods at different maturation stages [5], and C. theobromicola Delacr.has been isolated from both cherelle wilt and cacao pod rot [1].There is no record of other species as common pathogens in cherelles and pods.Thus, there may be a low potential for pathogens in cherelles and pods to infect each other and cause anthracnose outbreaks in cacao-producing farms.Additionally, many diseases from Colletotrichum species are reported as pathogens causing disease on fruits and leaves, such as diseases associated with C. gloeosporioides (syn.Glomerella cingulate), which causes leaf spots on apples [10]; C. truncatum, which causes leaf spots on leaves of Citrus reticulata; [11] and C. acutatum, which causes almond anthracnose [12].However, González et al. [10] found that some of the lineages of C. acutatum and C. gloeosporioides show host organ-specific pathogenicity.Some studies show that the latent phase of pathogenic Colletotrichum spp. in fruits, such as apples, strawberries, and citrus, is present in the leaves and twigs of the plant.In cacao, Rojas et al. [9] reported that fungi belonging to the CGLSC and gigasporum species complex (CGISC) have the potential to act not only as endophytes, but also as pathogens.Therefore, these pathogens are prospective targets for control of the disease.
This study aimed to identify the pathogen causing anthracnose in cacao cherelles and to determine its pathogenicity to cherelles, pods, and leaves, as well as to elucidate its life cycle for disease control strategies.We identified an isolate, PH22T135, obtained from a cherelle wilt at the species level based on morphological and molecular phylogenetic analysis, and tested its pathogenicity to cherelles, pods, and leaves of cacao.The strain was thus identified as a new species belonging to the CGISC and pathogenic only to cherelles.

Fungal Isolation
A cacao fruit sample showing "cherelle wilt" was collected in June 2022 from Pangasinan, Philippines.The fruit was blighted and dried with salmon-pink conidiomata (Figure 1A).Approximately 5 × 5 mm segments of symptomatic pod tissues were soaked into 0.6% (v/v) sodium hypochlorite for 30 s for surface sterilization, blot-dried using sterile tissue paper, and placed on water agar (agar 15 g/L) plates.Hyphae emerging from the tissues were transferred onto potato dextrose agar (PDA; 39 g/L; Eiken Kagaku, Tokyo, Japan) plates and maintained at room temperature (25 • C) to induce sporulation.A monoculture was obtained by isolating a single conidium from a conidial mass of the fungal colony.The isolate was maintained on a PDA plate.Our isolate was preserved in a tube with 10% sterile glycerol at −80 • C, as the ex-type living culture and type specimen were deposited at Central Luzon State University.

Morphological Observation
After a ten-day inoculation period, the characteristics of colonies on oatmeal agar (OA; oatmeal 20 g/L, agar 15 g/L) plate and synthetic nutrient-poor agar (SNA) [13] were observed at 25 • C in natural light conditions.Conidial masses produced on OA plates and pine needle placed on SNA plates were observed under an MDG-17 stereo microscope (Leica Microsystems GmbH, Wetzlar, Germany).Morphological characteristics of asci, ascospores, and conidia on OA plates or pine needles placed on SNA plates as well as the appressoria that formed on the bottom surface of the SNA plates were examined under a light microscope (BX51, Olympus, Tokyo, Japan).Each structure (n = 30) was

Morphological Observation
After a ten-day inoculation period, the characteristics of colonies on oatmeal agar (OA; oatmeal 20 g/L, agar 15 g/L) plate and synthetic nutrient-poor agar (SNA) [13] were observed at 25 °C in natural light conditions.Conidial masses produced on OA plates and pine needle placed on SNA plates were observed under an MDG-17 stereo microscope (Leica Microsystems GmbH, Wetzlar, Germany).Morphological characteristics of asci, ascospores, and conidia on OA plates or pine needles placed on SNA plates as well as the appressoria that formed on the bottom surface of the SNA plates were examined under a light microscope (BX51, Olympus, Tokyo, Japan).Each structure (n = 30) was measured using ImageJ software 1.53o (https://imagej.nih.gov/ij/index.html,accessed on 20 January 2022).

Molecular Phylogenetic Analyses
To identify PH22T135, 18 sequences of strains belonging to 10 species under the CGISC were used to construct molecular phylogenetic trees (Table 1).Sequences from Colletotrichum acutatum (CBS112996) and C. gloeosporioides (ICMP17821) were used as outgroups.The sequences of ITS, GAPDH, CHS1, ACT, and TUB2 were aligned independently with ClustalW [20] and the gap-including sites at both ends were pruned using MEGA 10.2 software [21].ML analyses were performed using MEGA 10.2 software.The best substitution model for ML analysis was decided based on the Akaike information criterion in the complete edition of the sequence using model search; the Tamura-Nei model + G was chosen for the substitution model.All gap-including sites were eliminated from the dataset, after which bootstrap (BS) analysis was performed in MEGA 10.2 software.Bayesian probability (BP) analysis was performed using MrBayes (version 3.1.1)[22].The best-fit models were separately searched for each DNA region using MrModeltest v.2.3 [23], and performed using PAUP* v.4.0b10 [24].SYM + G, HKY + I, and HKY + G were selected for ITS, GAPDH, and CHS1, respectively, while SYM + I was selected for ACT and TUB2.Analysis of the two Markov and Monte Carlo chains of the four runs was run for 1,000,000 generations.Data were rated every 100 generations, and the first 6500 generations were burned-in, because the average standard deviation of the split frequency rate in these generations was less than 0.01.

Pathogenicity Test
Pathogenicity tests were conducted using a conidial suspension adjusted to 1 × 10 5 conidia/mL.Twenty microliters of the conidial suspension were dropped onto three areas per pod, some of which had wounds, and some did not, on healthy young and mature cacao pods, and on young and old leaves.Sterile water was used as control.The samples were covered with plastic bags for five days to maintain high humidity.After 9-14 d, the inoculated samples were observed, and re-isolation was conducted from symptomatic pods and leaves.

Mycelial Growth Test
To determine the optimal temperature for the growth of the isolate, a mycelial growth test was conducted twice with 5 replicates each.The mycelial plugs (φ6 mm) were obtained from the colonies of our isolate, which was grown for 7 d at 25 • C, and placed at the center of PDA, SNA, and OA plates (φ9 cm).The plates were incubated at 4, 10, 15, 20, 25, 30, 35, and 40 • C for 4 d.Five replications were conducted for each temperature.Mycelial growth was measured daily, and the average colony diameter at each temperature was calculated.
Identification of our isolate at the species level was conducted using the data of species in CGISC (Figure 2).The final dataset contained 1725 bp, including gaps, comprising 545, 286, 280, 228, and 391 positions from ITS, GAPDH, CHS1, ACT, and TUB2, respectively.As the ML and BP trees exhibited similar topologies, the ML tree shown in Figure 2 had the highest log likelihood score of −6928.70.Our isolate was independent from the other 11 clades, including all 9 ex-type strains belonging to CGISC (Figure 2), and most closely related to C. zahaoqingense, as evidenced by high BS values and BP (BS/BP: 99/1.0).

Pathogenicity Test
Our isolate produced symptoms in cherelles, resulting in black lesions in all inoculated areas four days after inoculation (Figure 3A).In mature pods, black spots appeared only in wounded areas (Figure 3C).These were slightly darker than those in control (Figure B), and no expansion of lesions was observed.No symptoms developed in the pods or leaves treated with sterile water as controls (Figure 3B,D,F,H).The inoculum was re-isolated from all symptomatic tissues.Regarding pathogenicity on wounded young and mature leaves, the isolate did not show pathogenicity as seen in cherelles (Figure 3E,G).However, over time, the inoculated parts were slightly bigger as lesions than the control one, and the inoculum was re-isolated from the lesion.

Pathogenicity Test
Our isolate produced symptoms in cherelles, resulting in black lesions in all inoculated areas four days after inoculation (Figure 3A).In mature pods, black spots appeared only in wounded areas (Figure 3C).These were slightly darker than those in control (Fig- ure B), and no expansion of lesions was observed.No symptoms developed in the pods or leaves treated with sterile water as controls (Figure 3B,D,F,H).The inoculum was reisolated from all symptomatic tissues.Regarding pathogenicity on wounded young and mature leaves, the isolate did not show pathogenicity as seen in cherelles (Figure 3E,G).However, over time, the inoculated parts were slightly bigger as lesions than the control one, and the inoculum was re-isolated from the lesion.

Pathogenicity Test
Our isolate produced symptoms in cherelles, resulting in black lesions in all inoculated areas four days after inoculation (Figure 3A).In mature pods, black spots appeared only in wounded areas (Figure 3C).These were slightly darker than those in control (Fig- ure B), and no expansion of lesions was observed.No symptoms developed in the pods or leaves treated with sterile water as controls (Figure 3B,D,F,H).The inoculum was reisolated from all symptomatic tissues.Regarding pathogenicity on wounded young and mature leaves, the isolate did not show pathogenicity as seen in cherelles (Figure 3E,G).However, over time, the inoculated parts were slightly bigger as lesions than the control one, and the inoculum was re-isolated from the lesion.

Taxonomy
Colletotrichum kapreanum Takata & Kyoko Watan., sp.nov.(Figure 1).MycoBank no.: MB 852300.Etymology: The species name was derived from the name 'kapre', a giant living on a tree based on Philippine mythology.

Identification of Pathogenic Isolate
The PH22T135 strain extracted from cherelle wilt was identified as a new sp pathogenic to cacao, belonging to the CGISC based on morphological and phylog analyses; it was named Colletotrichum kapreanum.Although there are some reports of  2).Two species produce ascospores with septa in the CGISC, including C. gigasporum [18,25] and C. taiwanense [26]; however, the number of septa in C. taiwanense is larger than that of C. kapreanum (Table 3).The size of the conidia of C. kapreanum, however, is similar to that of C. gigasporum (Table 2).Meanwhile, the size of appressoria of C. gigasporum is larger than those of C. kapreanum (Table 4).Recorded sizes of ascospores of C. gigasporum and C. taiwanense were measured with PDA plates and rice straw, but PH22T135 asci and ascospores were not observed on PDA plates.

Identification of Pathogenic Isolate
The PH22T135 strain extracted from cherelle wilt was identified as a new species pathogenic to cacao, belonging to the CGISC based on morphological and phylogenetic analyses; it was named Colletotrichum kapreanum.Although there are some reports of other pathogens belonging to the genus Colletorichum that affect cacao, C. kapreanum is independent of not only these species, including C. gloeosporioides, C. siamense, C. theobromicola, C. tropicale, and C. aeschynomenes [4,[6][7][8][9], but also other reported species in the molecular phylogenetic tree, per morphological analysis.CGISC comprises 17 species, typically with longer conidia, characterized by a straight shape with a round end, than those of other Colletotrichum species, belonging to other species complexes.These species can also be distinguished from other species in other species complexes using phylogenetic analyses [19,20].The anamorph of C. kapreanum is consistent with the morphology of species belonging to CGISC.Focusing on morphologies of conidia and appressoria, C. kapreanum is distinct from C. zhaoqingense in the size of conidia and the shape and size of appressoria.Although C. gigasporum produces similarly sized conidia to that of C. kapreanum, their appressoria are significantly different in size.Additionally, we morphologically compared C. kapreanum with C. taiwanense, the latter of which has no molecular data, because the anamorphs of C. taiwanense are alike those of the species in CGISC, and we found no significant difference in the size of conidia between C. kapreanum with C. taiwanense.However, these two species also are distinguished from each other in the size and shape of appressoria.Regarding morphologies of ascospores, C. kapreanum produces smaller ascospores than those of C. gigasporum, while those produced by C. taiwanense are similar in size to those of C. kapreanum.However, C. kapreanum differs from C. taiwanense in terms of the number of septa of ascospores.C. kapreanum produces ascospores with a septum in the middle (usually one septum, rarely three septa), while C. taiwanense produces 0-3 ascospores and, rarely, septa [26,27].Therefore, we propose C. kapreanum as the new species belonging to CGISC.To the best of our knowledge, this is the third report of this species having ascospores with a septum, as belonging to the genus Colletotrichum.

Pathogenicity
The isolate was relatively pathogenic toward cherelles and produced symptoms in both wounded and unwounded parts of the organ.However, the isolate did not show pathogenicity to mature pods and leaves.C. tropicale is reported to cause anthracnose of cacao pods at different stages, including young, immature, and mature pods [5].Paguntalan et al. [27] reported Colletotrichum sp.isolated from an anthracnose-infected pod through a postharvest survey in Mindanao, Philippines.This strain belongs to CGLSC, according to the results obtained from morphological data and the pictures in their article.There are no records of Colletotrichum spp., except for CGLSC, related to anthracnose in cacao.In addition to our results, cacao pods are affected by different pathogens depending on their developmental stage which is a point to be noted when managing the relevant diseases.
In the life cycle of the plant pathogens, latent infection must be considered because this information is crucial in disease management strategies.Some studies provide information on the importance of latent infection of Colletotrichum spp. to control the disease.For example, Peres et al. [12] investigated the life cycle of the pathogenic C. acutatum in citrus, strawberries, blueberries, and almonds, which was detected as the causal agent of latent infection stages in the tissues of leaves, twigs, and/or flowers.Eventually, these fungi produced fruit rot from acervuli that appeared on damaged tissues, infected twigs, flowers, and/or buds.Our isolate C. kapreanum produced symptoms on mature pods, and colonies could be reisolated, which indicates that, although our isolate can be regarded as a host organ-specific pathogen to cherelles, it may be able to infect leaves and mature pods quiescently.However, knowledge about the biotrophic and epiphytic phases of this pathogen is needed.Furthermore, the investigation of this species on several organs with/without symptoms in cacao trees is essential for clarifying the life cycle of the pathogen, and further investigation of pathogens causing cherelle wilt in cacao could help clarify whether C. kapreanum is the dominant pathogen of cacao cherelles.

Conclusions
A new fungal pathogen, Colletotrichum kapreanum sp.nov, was identified in cacao fields in Pangasinan, Philippines.This pathogen only affects cacao cherelles, not pods, and can lead to a reduction in cacao production.Removing diseased cherelles is crucial in mitigating its impact on cacao cultivation.

Figure 2 .
Figure 2. Maximum likelihood (ML) tree of the Colletotrichum gigasporum species complex based on combined data sets of ITS, GAPDH, CHS1, ACT, and TUB2 sequences (1725 bp including gaps).ML bootstrap values and BP analysis are shown at the nodes (BS/BP).BS > 80% and BP > 0.95 are shown.Asterisks indicate ex-type strains.C. kapreanum, written in red, is a new species proposed in the present study.

Figure 3 .
Figure 3. Cacao pods and leaves inoculated with the PH22T135 strain.Black spot lesions in (A) wounded (right) and unwounded (left) cherelle and (B) the control nine days after inoculation.(C)

Figure 2 .
Figure 2. Maximum likelihood (ML) tree of the Colletotrichum gigasporum species complex based on combined data sets of ITS, GAPDH, CHS1, ACT, and TUB2 sequences (1725 bp including gaps).ML bootstrap values and BP analysis are shown at the nodes (BS/BP).BS > 80% and BP > 0.95 are shown.Asterisks indicate ex-type strains.C. kapreanum, written in red, is a new species proposed in the present study.

Figure 2 .
Figure 2. Maximum likelihood (ML) tree of the Colletotrichum gigasporum species complex based on combined data sets of ITS, GAPDH, CHS1, ACT, and TUB2 sequences (1725 bp including gaps).ML bootstrap values and BP analysis are shown at the nodes (BS/BP).BS > 80% and BP > 0.95 are shown.Asterisks indicate ex-type strains.C. kapreanum, written in red, is a new species proposed in the present study.

Figure 3 .Figure 3 .
Figure 3. Cacao pods and leaves inoculated with the PH22T135 strain.Black spot lesions in (A) wounded (right) and unwounded (left) cherelle and (B) the control nine days after inoculation.(C) Figure 3. Cacao pods and leaves inoculated with the PH22T135 strain.Black spot lesions in (A) wounded (right) and unwounded (left) cherelle and (B) the control nine days after inoculation.(C) Black spots appeared in the mature wounded pod (right) whereas no spots were found in the unwounded pod (left) 14 days after inoculation or (D) control.No lesions were found in (E) wounded (right) and unwounded (left) or (F) control young leave, nor in (G) wounded or (H) control old leave at 14 days after inoculation.

Figure 4 .
Figure 4. Mycelial growth rate of C. kspresii PH22T135 on OA, PDA, and SNA at different tem tures in the dark for 7 days.The vertical lines on each bar indicate standard deviations.

Figure 4 .
Figure 4. Mycelial growth rate of C. kspresii PH22T135 on OA, PDA, and SNA at different temperatures in the dark for 7 days.The vertical lines on each bar indicate standard deviations.Note: C. kapreanum resembles C. zahaoqingense, a sister group of C. kapreanum, in the shape of conidia and appressoria, but not in conidial size, as those of C. kapreanum are larger than those of C. zahaoqingense (Table2).Two species produce ascospores with septa in the CGISC, including C. gigasporum[18,25] and C. taiwanense[26]; however, the number of septa in C. taiwanense is larger than that of C. kapreanum (Table3).The size of the conidia

Table 1 .
Details of strains included in the phylogenetic analyses.

Table 4 .
Appressorial characteristics of Colletotrichum kapreanum and related species.

Table 3 .
Ascospore characteristics of Colletotrichum kapreanum and related species.

Table 4 .
Appressorial characteristics of Colletotrichum kapreanum and related species.