1. IntroductionWheat, rice, and maize form an essential food component worldwide and account for approximately 80% of total food grains production in the world. Wheat is the second most widely grown and consumed food crop globally after rice and is the staple food of approximately 35% of the world’s population. Current global wheat production is about 760.9 million tons (https://www.fao.org/faostat/en/#data/QCL; accessed on 3 September 2022). To feed the ever-increasing population of the world from diminishing agricultural land, it is essential to effectively manage the biotic stresses affecting wheat [1]. The major fungal diseases of wheat include rusts, fusarium head blight, powdery mildew, spot blotch, Septoria tritici blotch, tan spot, and glume blotch [2,3].Spot blotch is a foliar disease caused by the air and soil-borne fungal pathogen Bipolaris sorokiniana, the anamorph (asexual stage) of Cochliobolus sativus (teleomorph, sexual stage). The estimates of yield losses in wheat due to spot blotch vary from 15.5 to 19.6% [4], 20 to 80% [5], and may reach up to 100% under severe infection conditions [6]. Losses of up to 85% in Zambia [7] and 40% in the Philippines [8] have been reported. In highly susceptible wheat varieties, Hetzler et al. [9] reported yield losses of up to 87% due to B. sorokiniana. It is the predominant pathogen of foliar blight disease in the Indian states of Bihar, Delhi, Gujarat, Haryana, Karnataka, Maharashtra, Rajasthan, Uttar Pradesh, West Bengal, as well as in the neighboring countries of Bangladesh and Nepal [10].B. sorokiniana isolates vary considerably in their morphology, pathogenicity, and virulence. Variation in pathogenicity of B. sorokiniana under different conditions or locations was reported from Pakistan and Nepal [11,12]. Pandey et al. [13] employed random amplified polymorphic DNA (RAPD) markers to identify pathogen isolates belonging to different groups based on their morphological variation. The authors recommended that the genes for developing resistant cultivars should be selected according to the virulence genes present in pathogen isolates prevalent in a particular geographical region. This selection requires the information about the pathogen population structure in a specific area, which would differ in various regions with respect to different virulence genes [14].Over the past ten years, advances in Next Generation Sequencing (NGS) technologies and decreasing sequencing costs have led to an increase in the sequencing of several plant pathogenic fungi hampering crop production. Sequencing pathogenic fungi helps in exploring the host–pathogen interactions at the molecular level in several pathosystems and leads to the identification of several genes and virulence factors that play essential roles in infection and disease progression [15,16]. The whole genome sequence of several Bipolaris species belonging to different species complexes, viz., Bipolaris sorokiniana, B. maydis, B. oryzae, B. zeicola, B. cookei, and B. victoriae, are publicly available, providing an impetus to the Bipolaris research. The comparative genomics of important gene classes among different Bipolaris species and related fungi can help reveal the core genes in the genus Bipolaris and the expansion and deletion of vital gene classes attributed to their specific lifestyles.Till now, whole-genome sequences of eight B. sorokiniana isolates are available. The genome size of the B. sorokiniana strain ND90Pr is 34.42 Mb with 12,250 genes. In ND90Pr and ND93-1 isolates, 60,448 SNPs and 121 polymorphic SSRs were detected. The genes encoding enzymes such as NRPSs and PKSs involved in synthesizing several secondary metabolites, which control virulence, have been identified. Some of these genes are unique and conserved in B. sorokiniana [17,18]. Recently, Aggarwal et al. [19] announced the whole genome of the B. sorokiniana strain BS112, an Indian isolate (RCTM00000000) with a genome assembly size of 35.64 Mb. This whole-genome de novo sequence represents the first Indian genome-scale assembly for B. sorokiniana. However, detailed sequence analysis regarding pathogenicity-related genes, carbohydrate-active enzymes, and comparative genome analysis with other isolates is not available.

In the present study, we collected 12 B. sorokiniana isolates exhibiting high morphological variations from wheat fields in three different geographical locations of India and evaluated their pathogenicity on a spot blotch-susceptible wheat variety under greenhouse conditions. The three most virulent isolates were identified and sequenced. Comparative genomic analysis was performed to understand the factors governing the differences in their aggressiveness in causing the spot blotch disease. Interestingly, the isolate D2 had a relatively larger genome with an expanded set of pathogenicity-related genes, which might contribute to its higher virulence. The study provides the pathogen genomic resources for further identification and characterization of key pathogenesis-related genes and would aid in the development of better disease control measures.

4. DiscussionLike many phytopathogenic fungi, B. sorokiniana also exhibits high morphological and pathological variations [42]. In the present study, we observed morphological variability among the isolates of B. sorokiniana. Of the 7 isolates, 3 (A, L, and HD3069) had similar morphology and white mycelia with similar melanization on PDA. However, these isolates varied in their growth patterns. The isolates BS52 and D2 had similar growth patterns on PDA but showed differences in melanization. In an earlier study [43], no correlation was observed between the genetic similarity of groups and geographical origins of B. sorokiniana isolates, inferring that the morphological characteristics are not conditioned solely by genetic composition. As isolates from the same species differ in their morphology and genetic compositions, this variability must be attributed to interactions between genetic and environmental conditions, such as ecology and climatic variations in the areas from where the isolates have been obtained.Variation in environmental conditions, production of toxins, and other metabolites are among the few conditions that often result in variation in pathogenicity [23]. Studies that characterized pathogenic variations among the B. sorokiniana isolates from distinct geographical areas showed significant differences in host response [44]. In the present study, the pathogenicity of 12 B. sorokiniana isolates collected from three geographical regions of India was assessed on the wheat variety DDK 1025. Based on this, the 3 most virulent isolates (BS52, D2, and SI) were selected, which incidentally belonged to the three geographical regions. Among them, D2 was the most virulent. Milus et al. [45] also observed significant differences in aggressiveness and infection efficiency of 2 B. sorokiniana isolates belonging to the same pathotype. Sultana et al. [46] performed a similar investigation where 169 isolates of B. sorokiniana were assessed for pathogenic variability on the spot blotch-susceptible variety ‘Kanchan.’ They found a positive relationship between pathogenic variability and aggressiveness with agroclimatic conditions.Genome sequencing and assembly of 3 isolates of B. sorokiniana, BS52, D2, and SI, revealed genome sizes of 35.19 MB, 39.32 MB, and 32.76 MB, respectively. The genome assembly of BS52, D2, and SI showed N50 lengths of 1054399, 803483, and 1063709 and GC contents of 48.48%, 50.43%, and 49.42%, respectively. The genome of the isolate D2 was found to be comparatively larger than that of BS52 and SI. Similar efforts to sequence the genomes of the pathogenic fungal isolates collected from different regions showed variation in genome size. The genome assembly of the 19 isolates of the fungal wheat pathogen Zymoseptoria tritici ranged from 37.13 Mb to 41.76 Mb [47]. Similarly, 2 isolates of the oat crown rust pathogen, Puccinia coronata f.sp. avenae, showed genome assembly of 99.16 Mb (12SD80) and 105.25 Mb (12NC29) [48]. It has been reported that a larger genome is more likely to contain a high number of secondary metabolite-related genes, virulence genes, and stress-tolerance genes [49].BUSCO was used to assess the completeness of genome assembly utilizing a set of core single-copy orthologous genes [27]. The BUSCO alignment showed that final assemblies of BS52, D2, and SI were complete to approximately 99.08%, 99.06%, and 90.87%, respectively. The BUSCO prediction with >90% completeness indicated coverage of most of the genomic sequence space [16]. A phylogenetic tree constructed based on the single-copy orthologs of BS52, D2, and SI, with the other 11 genome sequences inferred that the isolates BS52, D2, and SI formed a clade with the 4 B. sorokiniana strains, indicating their genetic relatedness. Further, synteny analysis showed that all 3 B. sorokiniana isolates share a high degree of genomic synteny with the reference genome B. sorokiniana strain ND90Pr.Orthologous gene clusters were identified between the B. sorokiniana strain ND90Pr, and the 3 isolates using OrthoVenn2. The isolates BS52, D2, and SI formed a large number of common clusters along with the reference strain ND90Pr. Among the 3 isolates, D2 showed several unique clusters (318) containing the genes involved in essential activities, such as sequence-specific DNA binding transcription factor activity, transcription regulation, and protein secretion, which might contribute to the increased pathogenicity of D2. The predicted secretome for D2 was also relatively large compared to that of the isolates BS52 and SI. The secreted proteins are known to impart virulence characteristics to phytopathogenic fungi. During plant–fungi interaction, the phytopathogenic fungi secrete several proteins, which play a crucial role in fungal penetration, colonization, and lesion formation [49]; thus, imparting higher virulence to the phytopathogenic fungi.The Cluster of Orthologous Groups (COGs) results revealed that the genes for defense mechanisms were higher in the isolate D2 than BS52 and SI. Prediction of Biosynthetic gene clusters (BGCs) involved in secondary metabolite synthesis showed the highest number of genes for NRPS and PKS clusters in the D2 isolate. An expansion of NRPS and PKS backbone genes leads to increased production of secondary metabolites of these groups, which may play a significant role in host invasion by enabling appressorial penetration and pathogenicity [50]. The analysis of BGCs among the 3 B. sorokiniana isolates revealed that the beta-lactone, bacteriocin, and siderophore clusters were exclusively present in D2. These secondary metabolites are reported to impart increased virulence in fungal pathogens [51].Phytopathogenic fungi use carbohydrate-active enzymes (CAZymes) to break down the cell wall and enter plant cells. CAZymes play a vital role during early infection by sequestering the chitin oligomers released by the fungus and preventing their recognition by the plant immune system [52]. Among the six categories of CAZymes, glycosyl hydrolases (GH), which efficiently break down the plant cell wall for penetration and successful infection [39], were highly abundant in all 3 isolates. However, D2 showed a high number of CAZymes in all six categories compared with isolates BS52 and SI, indicating the expansion of specific gene classes. The results were consistent with previous studies in Colletotrichum species that showed the expansion of specific gene classes potentially involved in pathogenesis, particularly CAZymes and secondary metabolites [50,53].The prediction of secretome revealed that the number of secretory proteins was relatively large in isolate D2 compared with BS52 and SI. Several genes encoding secretory proteins such as CUTI1, PME, CATB, FCK1, RED3, etc. were unique in the isolate D2. These genes were enriched in gene ontologies, such as pathogenesis, cellular responses to hydrogen peroxide, hydrogen peroxide catabolic process, oxidative stress, invasive growth, in response to glucose limitation, pseudohyphal growth, etc. Similarly, these D2-specific genes were enriched in various pathways, including secondary metabolite biosynthesis, mycotoxin biosynthesis, polyketide biosynthesis, glycan metabolism, pectin degradation, etc. According to previous studies [Lu et al. [54] (cutinase activity), Sella et al. [55] (pectinesterase activity), Hernandez et al. [56] (catalase activity), Sørenson et al. [57] (cytokinin biosynthesis), Inderbitzin et al. [58] (alcohol dehydrogenase (NAD+) activity-mycotoxin production), etc.], the gene ontology and pathways of these D2-specific genes indicate their roles in fungal virulence, which could have contributed to D2 being more virulent than the isolates BS52 and SI.The Pathogen–Host Interactions database (PHI-base) is a catalog of experimentally validated pathogenicity genes from plant and animal pathogens that are manually curated from peer-reviewed publications [41]. Most of the genes from B. sorokiniana isolates had homology to the genes from Fusarium oxysporum, Magnaporthe oryzae, and Aspergillus fumigatus, likely due to the extensive research carried out on these model fungi, leading to the highest represented pathogens in PHI-base. The comparative analysis of the PHI-base revealed more pathogenicity genes in the isolate D2 (2476), followed by SI (2018) and BS52 (2015). Additionally, the unique phytopathogenic genes were identified in each of the isolates.Since the isolate D2 showed higher virulence than BS52 and SI, it was further analyzed to understand the possible reasons for its higher virulence. A total of 4069 genes were found to be unique in the D2 isolate. Among these, 41 genes were common in at least two of the three functional categories (CAZy, PHI, and secretory proteins), highlighting their importance in virulence. Among these, the following genes were reported to be associated with fungal pathogenicity, CDS_000012722 (cell wall organization and lytic transglycosylase) [59,60], CDS_000014640 (trehalose biosynthesis) [61], CDS_000011491 (acetylglucosaminyl transferase activity) [62], CDS_000012002 (lipid A biosynthesis) [63], CDS_000013078 (polygalacturonase activity) [64], and CDS_000012490 (glutathione metabolism) [65]. Functional validation of these genes will confirm their roles in pathogenicity and the molecular basis of variation in virulence of the 3 isolates of B. sorokiniana. The validated pathogenicity-related genes could be used as key targets to develop strategies to control the fungal pathogen.

The isolate D2 exhibited a relatively larger genome with expanded arsenals of BGCs, CAZymes, pathogenicity genes, and secretome, which might be responsible for imparting the higher pathogenicity relative to the isolates BS52 and SI. These results are consistent with the fungal pathogenicity assay where the isolate D2 caused more severe disease symptoms on wheat leaves than SI and BS52. Thus, our results provide valuable genomic data to further characterize the molecular mechanisms for increased virulence in B. sorokiniana isolates.