Seventy four diseased root samples of mulberry were collected from farmers field from 2017 to 2019 in high-incidence states (Andhra Pradesh, Karnataka, Tamil Nadu, and Telangana) of South India. The root rot causative fungal pathogen was isolated following the root bit method of isolation. The infected mulberry roots showing typical root rot symptoms were chopped into small bits and surface-sterilized using 5% sodium hypochlorite for 2 min. These root bits were washed with sterile water and dried up on a filter paper (Whatmann No. 1). Dried root bits were placed on Petri plates with sterile potato dextrose agar (PDA; HiMedia, Mumbai, India) under aseptic conditions. The plates with root bits were incubated at 27 ± 1°C for 7 days to obtain fungal growth. Pure cultures were obtained by sub-culturing matured hyphae on 2% water agar plates and stored for further use at 4°C.

The morphology of 7-day-old fungal cultures grown on PDA medium incubated at 27°C under dark conditions was examined. Cultural characteristics like colony color, pigmentation, texture, and growth were noted. Conidial morphology was observed using light microscopy and conidial dimensions of 20 selected isolates were determined (Supplementary Table 1). The length and width of 10 mature and immature conidia per isolate were measured using ImageJ software.

Glutaraldehyde (2.5%) prepared in 0.2 M cacodylate buffer (pH 7.2) was used for the initial fixation of fungal cultures. The fixed cultures were washed using cacodylate buffer and then by double distilled water dehydrated in ethanol series. Critical drying of the dehydrated samples was carried out by using a Critical Point Dryer (EMS850, Electron Microscopy Sciences, Hatfield, PA, USA). Lastly, samples were coated with gold nanoparticles in Sputter Coater (EMS550, Electron Microscopy Sciences) mounted onto copper stubs using double side sticky tape. Microscopic images were obtained after scanning under JEOL 100 CX ASID-4D scanning electron microscope (JEOL Ltd., Tokyo, Japan) at 20 kV (Arunakumar et al., 2018; Bozzola and Russell, 1992).

Pure cultures of 20 fungal isolates were inoculated in 50 ml of sterile potato dextrose broth and incubated for 7 days at 27 ± 1°C, without shaking. The mycelial mats were separated by using sterilized filter paper (Whatmann No.1). DNA extraction was performed, following the method described by Gnanesh et al. (2021).

The representative fungal isolates of mulberry were identified by sequencing ITS region, ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATAT GC-3′) primers (White et al., 1990) and β-tubulin region, βt2a (5′-GGTAACCAAATCGGTGCTG CTT TC-3′) and βt2b (5′-ACCCTCAGTGTAGTGACCCTTGGC-3′) (Glass and Donaldson, 1995). The polymerase chain reaction (PCR) amplification was performed with a thermocycler (GeneAmp PCR system 9700, Applied Biosystems, Foster City, CA, USA). in a 25 μl reaction mixture consisting of 2 μl of template DNA (40 ng), 12 μl of 2× Ampliqon master mix, 1.5 μl of each primer (10 pM) synthesized from Eurofins, Bangalore, India and 8 μl of nuclease free water. The PCR parameters for ITS gene PCR amplification: initial denaturation at 94°C for 4min followed by 35 amplification cycles of 94°C for 30 s, 55°C for 30 s and 72°C for 1 min, and final extension at 72°C for 8 min. Similarly, β-tubulin gene PCR amplification was performed in a 25 μl reaction mixture which contained 10 μl of 2× Ampliqon master mix, 2 μl template DNA (40 ng), 11 μl of nuclease free water and 1 μl of each primer (10 pM). The PCR parameters for partial β-tubulin region are initial denaturation at 95°C for 5min, followed by 35 amplification cycles of 94°C for 45 s, 55℃ for 45 s, and 72℃ for 1 min, and final extension at 72°C for 7 min. Amplified products were electrophoresed using 1.5% agarose gel stained with ethidium bromide with 1× TAE running buffer, and visualized in a UV transilluminator. A 100 bp ladder was used as a marker and gels were documented using GelDoc Bio Imaging System (Syngene, Cambridge, UK). The amplified PCR products were purified and sequenced by Sanger sequencing method, Eurofins Pvt., Ltd. (Bengaluru, India).

Sequence alignments of ITS and β-tubulin were carried out using the MEGA X software (Kumar et al., 2018). The obtained nucleotide sequences were edited using BioEdit software and submitted to BLAST search analysis to identify the sequence homology in NCBI database. Finally, the sequences were deposited in GenBank and obtained the accession numbers. The phylogenetic analysis of concatenated data was done by selecting the best nucleotide substitution model using Bayesian Information Criterion (BIC) in Mega X software (Kumar et al., 2018). Tamura 3-parameter (T92) model was selected for phylogeny construction and the tree was rooted to Barriopsis fusca as an outgroup. Type strain and reference sequences were obtained from GenBank and the trees were analyzed by the bootstrap method with 1,000 replications.

Pathogenicity tests of 20 representative isolates were performed on two susceptible mulberry accessions, Victory-1 and Thailand Male, following the root dip method under glasshouse conditions (Arunakumar and Gnanesh, unpublished). Four months old saplings grown in raised nursery beds were gently uprooted and dipped in the conidial suspension (1 × 10⁶ spore load per ml load) for 12 h. Control plants were also dipped in sterile distilled water for 12 h. Later, saplings were transplanted into the earthen pots having sterile soil arranged with three replications following a completely randomized design (CRD). The infected plants were watered regularly to maintain sufficient moisture. BRR symptoms, wilting, and rotting percentage were recorded after 120 days of post inoculation. The pathogenicity of the isolates was classified using the wilting and rotting percentage as described in Table 1.

Forty five mulberry germplasm accessions including exotic and indigenous collections were used for screening. The accessions represent eight different countries including India (33), Indonesia (3), Japan and Thailand (2 accessions each), Philippines, Australia, China and France one accessions each. Among the 45 accessions 17 accessions belonged to collections, eight were open pollinated hybrids, six were from the survey, five accessions were from selections, and four each accession were from cross pollinated hybrids and clonal selection respectively and only one accession was developed. These accessions belonged to six different Morus spp. including a known popular susceptible variety Victory-1, were chosen for identification of resistance against L. theobromae causing BRR of mulberry (Table 2). Four months old saplings of each accession were uprooted gently and inoculated following the root dip method as mentioned above. The inoculated pots with four replications (single plant/replication) were arranged in a CRD. The experiment was conducted under glasshouse conditions with two control plants free from inoculation and repeated twice. The aerial observation of the total number of leaves, shoots (wilted and healthy), and the number of plants dead was recorded at 30 days intervals in each accession. Plants were uprooted after 120 days of post inoculation and observations on weight of the whole root, healthy root and above ground biomass per plant (g) were recorded. These disease indices like wilting and rotting percentages were calculated according to formulae described by Sowmya et al. (2018). The disease severity index and disease reaction were categorized for each accession based on the wilting and rotting percentage using the modified scale as mentioned in Table 1 (Ahmed et al., 1999).

The collected data were statistically processed using IBM SPSS Statistics software version 23.0 (IBM Corp., Armonk, NY, USA). One-way ANOVA analysis was performed to study the difference in the pathogenicity of the selected L. theobromae isolates and means of the BRR disease scores were compared using Tukey test (P < 0.05).

Symptomatic mulberry plants in the field exhibited varying degrees of BRR symptoms, characterized by yellowing and wilting of leaves (Fig. 1A), browning of the severely infected roots and in some cases appearance of newly regenerated roots were observed (Fig. 1B). The stem of the symptomatic plants was browning of root xylem and rotting of root cortex (Fig. 1C). Fifty seven isolates of L. theobromae were successfully recovered from 74 root rot samples collected from Southern India. The isolated fungi were grown on PDA media at 27°C in dark conditions. The cultures showed initially off-white to grey with fluffy, round to irregular margin, flat aerial mycelia were grown radially and uniformly in all directions and covered the surface of the media (90 mm diameter Petri dishes) within 4-5 days after incubation (Fig. 2A). Initially, the reverse side of the culture showed algae green dot patches and longitudinal striations towards the periphery. After 14-15 days of incubation, the reverse side turned to black color (Fig. 2B) and the upper surface formed dark grey colonies with fluffy aerial mycelium (Fig. 2C). Liquid exudates were produced and became dry after a few days (Fig. 2D).

The virulent isolate MRR-142 is shown in the microscopic pictures in Fig. 3. The mycelia were hyaline and septate, and some of the malformed mycelia were also observed (Fig. 3A and B). Pycnidia were semi-immersed, solitary, globose, papillate, and leaden-black forms after 15-20 days of incubation. Paraphyses were cylindrical, aseptate and hyaline. Conidiogenous cells were characterized as holoblastic, cylindrical and hyaline. Conidia produced after 15-20 days of incubation, immature conidial length measured from 13.34 μm to 7.4 (± 1.24) μm and breadth were ranged from 4.47 μm to 2.28 (± 0.51) μm (Fig. 3C and D). The mature conidial length was found in the range from 25 μm to 17 (± 3.0) μm and breadth was in the range from 14 μm to 7 (± 2.0) μm (Supplementary Table 1). Conidia were initially hyaline, aseptate, ellipsoid to obovoid, unicellular, cinnamon to sepia, dense walled and rounded with longitudinal light striations, then mature conidia form 1-2 septa (Fig. 3E and F). The conidial morphology and size indicated that the isolates were L. theobromae. The fine structure of conidial surfaces and conidia on a conidiogenous cell was observed using scanning electron microscopy (SEM). Conidial surface morphology of L. theobromae, although appeared to be smooth when viewed in a light microscope they are roughened and verruciform when viewed in high magnification in SEM (Fig. 3G-I). The ornamentation type could not be reliably distinguished with light microscopy. Conidia are broadly ovoid to ellipsoidal, presenting both ends broadly rounded or end rounded and the base narrowed or apiculate (Fig. 3H).

Twenty L. theobromae isolates were selected for phylogenetic analysis using ITS and β-tubulin regions based on the geographical region, representing the four states of South India. Also, based on virulence and morphological characters of the isolates like colony color, characters of aerial mycelia growth. The amplicon size of ITS and β-tubulin was detected between 500-600 bp and 300-400 bp, respectively (Munirah et al., 2017). Blast search homology of ITS and β-tubulin sequences revealed 98-100% similarity with type strain CBS 164.96 L. theobromae (Supplementary Table 2). GenBank accession numbers were obtained for all the isolates (Table 3). A phylogenetic tree was constructed for all the isolates of L. theobromae and the isolates were grouped into a single clade, representing the reference sequences (CBS111530 and COAD1788) and type strain (CBS 164.96) with a bootstrap value of 83% (Fig. 4). Thus confirming the identity of the isolated pathogen is L. theobromae, whereas all the other Lasiodiplodia species; L. citricola, L. crassispora, L. euphorbicola, L. mahajangana; L. missouriana; L. pseudotheobromae; L. subglobosa formed monophyletic clade individually with respect to their type strain. The closest species near to L. theobromae isolates were L. missouriana and L. viticola, whereas L. crassispora formed a separate cluster. The final pooled dataset of ITS and β-tubulin of 40 nucleotide sequences after alignment consists of 649 positions, 595 were conserved, 54 variable, 31 parsimony-informative, and 23 were singleton.

The pathogenicity test conducted on susceptible genotypes Victory-1 and Thailand Male revealed significant variation (P < 0.05) between the selected 20 isolates (Table 4). The infected plants, showed similar symptoms of BRR, like the appearance of dark colored lesions, rotted roots, wilting of leaves and browning of stems (Fig. 5). The wilting and rotting percentage of Victory-1 ranged between 3.6 to 97.6 and 4.6 to 100.0, respectively, similarly of Thailand male, 7.6 to 99.3 and 10.3 to 95.0, respectively. Sixteen isolates were highly pathogenic and four isolates, MRR-056, MRR-126, MRR-133, and MRR-153 were less pathogenic on both the susceptible genotypes exhibiting low levels of wilting and rotting. Among all the isolates, MRR-142, isolated from Andhra Pradesh, was found to be highly pathogenic which caused a maximum percentage of wilting and rotting (Table 4).

A total of 45 mulberry accessions were screened against the L. theobromae isolate MRR-142 through the root dip method of inoculation. The BRR disease reactions of resistant and susceptible accessions revealed significant variation under glasshouse conditions. The wilting and rotting percentage among the 45 accessions in the two experiments ranged between 0.0 to 96.7 and 0.0 to 100.0, respectively (Table 5). Accession, ME-0168 of Indonesian origin belonging to M. latifolia was found to be highly resistant (HR) against BRR (Fig. 6). The exotic accession ME-0168 was free from BRR symptoms and was on par with healthy control plants. Eight accessions that were found to be resistant (R) were G2, ME-0006, ME-0011, ME-0093, MI-0006, MI-0291, MI-0489 and MI-0501. Whereas six accessions showed highly susceptible (HS) reaction, 12 were susceptible, and 18 were moderately resistant (MR) to BRR (Table 5, Fig. 7).