Multi-Gene Phylogenetic Approach for Identification and Diversity Analysis of Bipolaris maydis and Curvularia lunata Isolates Causing Foliar Blight of Zea mays

Bipolaris species are known to be important plant pathogens that commonly cause leaf spot, root rot, and seedling blight in a wide range of hosts worldwide. In 2017, complex symptomatic cases of maydis leaf blight (caused by Bipolaris maydis) and maize leaf spot (caused by Curvularia lunata) have become increasingly significant in the main maize-growing regions of India. A total of 186 samples of maydis leaf blight and 129 maize leaf spot samples were collected, in 2017, from 20 sampling sites in the main maize-growing regions of India to explore the diversity and identity of this pathogenic causal agent. A total of 77 Bipolaris maydis isolates and 74 Curvularia lunata isolates were screened based on morphological and molecular characterization and phylogenetic analysis based on ribosomal markers—nuclear ribosomal DNA (rDNA) internal transcribed spacer (ITS) region, 28S nuclear ribosomal large subunit rRNA gene (LSU), D1/D2 domain of large-subunit (LSU) ribosomal DNA (rDNA), and protein-coding gene-glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Due to a dearth of molecular data from ex-type cultures, the use of few gene regions for species resolution, and overlapping morphological features, species recognition in Bipolaris has proven difficult. The present study used the multi-gene phylogenetic approach for proper identification and diversity of geographically distributed B. maydis and C. lunata isolates in Indian settings and provides useful insight into and explanation of its quantitative findings.


Introduction
Maize (Zea mays L.) is one of the most diverse crops with great adaptability under a range of broad agro-climatic zones. It is the third-most commonly cultivated cereal after wheat and rice, is known as the 'Queen of the Cereals' globally because of its significant potential for genetic yield among cereals, and is grown in approximately 150 m ha across about 160 countries with wide soil properties, climate, biodiversity, and management practices [1]. Maize is cultivated on 9.2 million hectares in India, with an average yield of 2965 kg/ha and a total production of 27.8 million tons [2,3]. The major producing states of maize in India are, by zone, East: Bihar, Orissa; North: Uttar Pradesh, Punjab, Haryana, Himachal Pradesh, Jammu, Kashmir; West: Rajasthan, Gujarat, Maharashtra; South: Andhra Pradesh; Central: Madhya Pradesh [4,5].
On a global basis, maize is affected by 112 diseases, but in Indian agroclimatic conditions, there are 35 main diseases that attack maize [6]. Worldwide yield losses in maize up to 9% have been estimated due to fungal, bacterial, and nematode diseases [7]. The severe maize diseases are spread across a broad disease spectrum and include downy mildew, rust, leaf blight, and stalk rot [8]. These diseases also include maydis leaf blight (MLB), one of the most devastating diseases that has emerged as an economically significant problem [9].
Worldwide, Curvularia and Bipolaris species attack cereal and grasses, causing leaf spot diseases on rice, maize, sorghum, wheat, and barley, resulting in severe grain yield losses [10][11][12][13]. An important genus of plant pathogens is Bipolaris (an anamorph of the ascomycetes genus Cochliobolus), with more than 100 species. Bipolaris species mostly attacks members of the poaecae family and causes root rot, leaf spot or blight, seedling blight, and ear rot and blight [14]. Due to being seed-borne in nature, Bipolaris spp. causes significant damage to field crops, mostly as post-harvest damage that ultimately impairs market value [15,16]. Bipolaris spp. has caused severe epidemics, such as the Great Bengal famine, Northern leaf spot blight, and Southern corn leaf blight, which ultimately caused significant economic losses in recent decades [17].
Worldwide, in maize producing regions, the maydis leaf blight, also known as Southern corn leaf blight (SCLB), is a severe maize crop disease caused by Bipolaris maydis, an anamorph of Cochliobolus heterostrophus. It is considered as a graminicolous species of Helminthosporium [11,17,18]. In USA and UK, Bipolaris maydis caused severe epidemics in 1970. The decline in the development of maize is possibly due to maydis leaf blight in India.
Maydis leaf blight (MLB or SCLB) generally occurs in moderately humid and warm conditions with temperatures of 20-32 • C and causes potential damage and yield losses in major maize-growing regions of India. During the Kharif season, the disease spreads rapidly throughout India and is most prevalent in plains, highlands, and peninsular regions [19]. Sharma and Rai (2000) reported that maydis leaf blight caused by B. maydis qualifies as a major disease of maize is capable of inflicting significant losses in productivity, to the extent of 41 percent [20].
Bipolaris and Curvularia species mainly have been characterized based on cultural and morphological characteristics in the past few decades. Dothideomycetes fungi, the most diverse class of fungi, include Curvularia and Bipolaris species [21]. All members of Drechslera and Exserohilum once grouped to the graminicolous Helminthosporium species [12]. Species of Bipolaris were previously known as Helminthosporium. The genus Helminthosporium is divided into four genera based on taxonomic refinements: Drechslera, Bipolaris, Exserohilum, and Curvularia. They are morphologically similar, and particularly, these four genera are called helminthosporioid fungi [12,18]. Some of these related genera have caused maize leaf spot and mixed infections, caused by the genera Curvularia and Exserohilum, C. intermedia, C. lunata, C. eragrostidis, C. pallescens, C. clavata, C. spicifera, C. papendorfii and C. inaequalis [22,23], while leaves, sheaths, and maize bracts were reportedly infected by E. rostratum and Exserohilum turcicum, which have also been reported to cause extreme leaf spot in maize [24,25].
Under field conditions, the symptoms of maydis leaf blight caused by those species were prevalent and complex. It is difficult to identify these species reliably based on symptoms alone. Despite their significance, species identification remains problematic in Bipolaris and Curvularia since numerous species have morphological traits that overlap, making it difficult to differentiate between these two genera under different conditions [12,26]. It might be difficult to distinguish between the genera and species of Bipolaris and Curvularia. Due to substantial intraspecies and interspecies variability and the difficulty of generating sexual phases under laboratory circumstances, asexual morphological traits are insufficient to correctly identify species [26,27]. However, combined study can differentiate these two genera based on sexual, conidia morphology, and molecular studies.
Curvularia has conidia with overly enlarged intermediate cells, which contribute to the curvature, whereas Bipolaris has a continuous curvature throughout the conidium duration. In addition, conidia are usually longer in Bipolaris than those in Curvularia [28]. Based on the morphology of ascospores, Curvularia forms loosely packed ascospores in the ascus, while Bipolaris produces supercoiled ascospores [29].
Representatives of the Consortium for the Barcode of Life's Fungal Working Group (FWG) evaluated different markers, three nuclear ribosomal DNA regions (ITS, LSU, SSU), and one protein-coding gene (RPB1), and proposed the ITS region as the key fungal barcode marker [30]. Species identification [31], species level phylogenies [32], and DNA barcoding studies [33] have all been conducted in the region. ITS data in the International Nucleotide Sequence Database (INSD: GenBank, EMBL, and DDBJ) demonstrated that this region is not equally variable in all fungi taxa, although most studies have designated it as the standard barcode marker for fungi.
These taxa have minimal or no barcode gaps in the ITS sequence, according to Stielow et al., 2015 [34]. The ITS region alone may not be sufficient for some species identification. Mycologists, particularly those working with ascomycetous fungi, frequently explore the feasibility of a two-marker barcoding system for fungi [7]). LSU [35,36] and protein-coding genes can be employed with ITS [36]. Using slowly evolving protein-coding genes, in particular, can significantly improve the phylogenetic resolution of deep divergences [37].
Bipolaris species are now typically identified based on morphological traits; however, other species exhibit numerous similarities, and conidial characteristics are frequently varied depending on isolates and culture circumstances. Due to ambiguities in morphological characteristics, DNA sequences of multiple loci are preferred to molecular identification and define Bipolaris and Curvularia species [28,38]. Currently, phylogenetic analysis studies are based on protein-coding gene-glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Ribosomal markers of rDNA internal transcribed spacer (ITS) and a large subunit of nuclear ribosomal DNA (LSU) have been used to evaluate the phylogenetic relationships among the Bipolaris genus isolates and with their sister genera [11,28,38]. The ITS, LSU regions, and GAPDH gene phylogeny provided a better resolution for taxa at the terminal clades, which has also been reported for other Dothideomycetes [39][40][41]. Based on these loci, phylogenetic studies have allowed Cochliobolus (sexual morph) species to be reallocated to either Bipolaris or Curvularia [29].
RAxML (Randomized Axelerated Maximum Likelihood) is a tool for inferring complex phylogenetic trees using maximum likelihood in a sequential and parallel way. It can also be used to perform post-analysis on phylogenetic tree sets, alignment analyses, and short-read evolutionary placement. RAxML is among the most widely used and effective software methods for creating a maximum likelihood (ML) tree. This would be the most accurate method for evaluation, but it is also the most resource-intensive. The model developed by RAxML is believed to be a good evolution model. Under this model, there was no major change in the likelihood of the inferred tree fitting the genuine tree [42].
It is difficult to distinguish between species solely based on their morphological characteristics because many of their morphological features overlap. Bhunjan et al., 2020 estimated 132 species epithets in Index Fungorum under Bipolaris, of which 15 had been moved to Curvularia, and only 43 are approved as Bipolaris species [43].
Based on the above studies, the combined study of morphological and molecular data gave a precise and correct method for identifying Bipolaris spp. Thus, the objectives of this study were as follows: (i) identification of Bipolaris maydis as the cause of maydis leaf blight, and Curvularia lunata as the cause of leaf spot of maize based on morphological analysis, (ii) precise identification and phylogenetic analysis based on ITS, GAPDH, LSU, and D1 and D2 regions of LSU to know the diversity of Bipolaris maydis and Curvularia lunata isolates in different geographical locations of major maize-growing regions in India.  (Table 1). Around 186 symptomatic samples of maydis leaf blight were collected based on the characteristic disease symptoms of diamondshaped lesions that vary in size, with 3-22 mm length and 2-6 mm width, that become rectangular-or spindle-shaped lesions with chlorotic halos at later stages. In the case of leaf spot of maize, 129 maize-leaf-spot-symptomatictic samples in the early stages of maize leaves that had yellow necrotic spots, which eventually expand to circular, spindle-shaped, oval, or strip lesions, were collected. Twenty sampling districts were covered during sampling, and their district locations are depicted in Supplementary Figure S1.

Isolation and Morphological Characterization of Pathogens
The symptomatic tissue was cut from the margins of lesions and surface-sterilized with 1% sodium hypochlorite for 1 min and then rinsed with distilled water three times. The surface-sterilized leaf tissues (5-6 mm 2 ) with lesions were placed on Petri plates containing potato dextrose medium (HiMedia, Mumbai, India) amended with chloramphenicol (0.05 g/L) (HiMedia, Mumbai, India) and incubated at 28 ± 2 • C for three days [44]. The agar disc (7 mm) containing the mycelia from the 3-day-old colonies was placed on the center of a Petri plate with fresh PDA and then incubated at 28 ± 2 • C. The pure cultures were obtained through single-spore isolation [29] and maintained on agar slants with regular subculturing.
Observations based on cultural characteristics of the different isolates were recorded over a 15-day incubation period of PDA containing sporulating cultures. Three replications of each isolate were maintained. Microscopic features of isolates, i.e., conidia shape, size, number of septa, color were taken as parameters by using a Olympus BX41 Epi and Trans Fluorescence Microscope (Olympus, Tokyo, Japan) for differentiation of genera [45]. While measuring the morphological features of different genera, thirty observations were randomly documented. The isolates were used to study the morphological characteristics described earlier [10,12,45,46].
Conidia and conidiophores were mounted in distilled water and viewed in an Olympus BX41 Phase Contrast and Dark Field Microscope (Olympus, Tokyo, Japan). Measurements of conidial width were recorded from the widest part of each conidium. Olympus BX41TF Epi and Trans Fluorescence Microscope (Olympus, Hamburg, Germany) measured their lengths and widths based on previous studies [10,12]. On average, 30 measurements of conidial length and width of each isolate was recorded. Mean ± standard deviation for length and width ranges of each conidium was recorded.
Bipolaris maydis isolates were cultured on PDA (Himedia, Mumbai, India) for 15 days at 28 • C. Actively growing fungal cultures of agar blocks were cut and fixed overnight at 0.1 M sodium phosphate buffer (pH 7.3) containing 2% glutaraldehyde. In phosphate buffer, each fungal mat was thoroughly washed three times for 15 min and samples were dehydrated for 15 min by using a series of graded ethanol (10,20,30,40,50,60,70,80,90, and 100%).The dehydrated and fixed samples of Bipolaris maydis isolates were further dried for 5 min with CO 2 and then immediately fixed on aluminum stubs and sputter-coated with carbon in a Polaron E-500 sputter coater (Polaron Equipment, Watford, England) and viewed under a visible scanning electron microscope (S3400N-Hitachi, Tokyo, Japan) immediately.
Koch's postulates test [47] for pathogenicity of the isolates of Bipolaris maydis and Curvularia lunata were carried out in the glass house. In glass house, five maize seeds of Kanchan variety were sown in 45 cm plastic pots containing sterilized soil and were kept in glass house at 26 • C with 18 h photoperiod. One plant per pot was maintained. Bipolaris maydis and Curvularia lunata conidial suspension was prepared by scraping all conidia from the culture plate after 10 days of incubation followed by suspension in deionized water containing 0.02% tween 20 (Himedia, Mumbai, India). The final pathogen concentration was adjusted to 1 × 10 6 conidia L −1 , as suggested in earlier reports [48]. The final spore suspension concentration (10 6 spores/mL) was inoculated on the leaves of 25-day-old maize plants using a hand atomizer until leaf run-off. Uninoculated plants were sprayed with sterile distilled water in the same way. Each treatment had three replicates, and the maize plants were kept in a glass house at 26 ± 2 • C with 12 h of darkness and 12 h of light, with the treated plants covered with polythene bags to maintain a relative humidity of more than 90% as per the method described by Manzar et al. [49]. After the infection appeared on the leaves, the fungus was re-isolated from the infected leaves and the identity was reconfirmed, to prove Koch's postulates. The pathogenicity test was carried out thrice (Supplementary Figure S5a,b). The cultures of the isolate were deposited to the IDAapproved National Agriculturally Important Microbial Culture Collection (NAIMCC) and the accession number was obtained.

DNA Extraction, PCR, and Sequencing
Via genomic DNA extraction, the isolates were cultured on PDA medium for 15 days at 28 • C for DNA isolation. Mycelial scrapings (50-60 mg) were obtained from the fully grown cultures in the Petri plate. Harvested mycelium was transferred into an autoclaved mortar and then powdered in a pestle mortar with liquid nitrogen. To extract genomic DNA, Nucleopore GDNA Fungus Kit-NP-7006D (Genetix Biotech Asia Pvt. Ltd., New Delhi, India) was used, as described in a previous report [29]. Quantification and assessment of the quality of genomic DNA were checked using a spectrophotometer (Shimadzu, Tokyo, Japan) by recording the absorbance at 245 nm.

Sequence Alignment and Phylogenetic Analyses
Newly generated ITS, GAPDH, and LSU sequences were analyzed separately with all available ex-type available sequences downloaded from GenBank and DDBJ to determine preliminary identifications of all the isolates. Raw sequences were assembled with Sequencher v. 4.9 for Windows (Gene Codes Corp., Ann Arbor, MI, USA). The assembled consensus sequences were initially aligned with MAFFT v. 7 using default settings (http://mafft.cbrc.jp/alignment/server/; accessed on 26 June 2021) and adjusted manually where necessary [53]; poorly aligned regions of the alignments and hyper-variable regions where alignment was ambiguous were removed using Trimal [54] and exported in a standard PHYLIP file. To fully resolve closely related species, all isolates were subjected to a multi-gene combined analysis. Phylogenetic reconstructions of concatenated and individual gene trees were performed using maximum likelihood (ML) criteria. Maximum likelihood trees were generated using RAxML 8.1.15 [55] under a GTRGAMMA model. Bipartitioned information was also input into RAxML software [56] for maximum likelihood (ML) analysis. A rapid bootstrap (BS) analysis was conducted with 1000 replications, the software produced a best-scoring ML tree with BS probabilities, and the results were obtained in 'TRE' files. The combined multi-marker dataset (ITS, GAPDH and LSU) was bipartitioned by gene region and was analyzed using RAxML as described above. Phylogenetic trees were viewed in MEGA v. 11 [57]. All sequences generated were deposited in GenBank (Table 2).

Statistical Analysis
Statistical analysis of length and width of each isolate were analyzed by using a oneway analysis of variance [58] and Duncan's multiple range test (DMRT) using Statistical Product and Service Solution (SPSS) version 16.0 software Developed by SPSS Inc., now IBM SPSS Armonk, NY, USA. On average, 30 measurements of conidial length and width of each isolate were recorded. Mean ± standard deviation for length and width ranges of each conidium was recorded. Table 2. Details of the isolates used in this study, including the hosts, locations, and GenBank accession numbers of the generated sequences of the ribosomal marker (ITS, LSU, and D1 and D2 region of LSU) and protein-coding gene GAPDH.

Morphological Identification of Bipolaris Isolates (Maydis Leaf Blight)
Twenty sampling district sites were chosen from major maize-growing regions of India, from which a total of 186 leaf blight symptomatic samples were collected. Seventy-seven isolates of Bipolaris were identified from the total collected maydis-leaf-blight-symptomatic samples ( Table 1). The 77 isolates were divided into five groups as per cultural and morphological characteristics (Table 3). Group A consisted of 20 isolates, representing 25.97%, Group B consisted of 9 isolates, representing 11.68%, Group C consisted of 25 isolates accounting for 32.46%, Group D had 11 isolates accounting for 14.28%, and Group E consisted of 12 isolates which accounted for 15.58%. Table 3, give a brief overview of the morphological data for these Bipolaris isolates in groups A-E. Characteristics of the colony: variable colony morphology on PDA was found after 15 days for each group. Group A isolates had gray colonies with spots, raised mycelia was observed with irregular margin; in Group B isolates, colony color was gray with smooth raised mycelia and irregular margins. The isolates in Group C produced gray colonies with appressed mycelia and irregular margin. The colonies from Group D were blackish gray with white spot, smooth appressed mycelia, and regular margin, whereas Group E had whitish gray colonies, rough raised mycelia, and irregular margin. The conidia length-width ratio, color and shape, and conidial morphology was grouped into five groups. Group A had a larger length-width ratio than Group C and Group D. All the isolates of the five groups produced slightly curved, light-brown to brown, fusiform conidia (Supplementary Figure S2). SEM observations: Bipolaris conidia had smooth walls, were fusiform or slightly curved, and placed on a geniculate conidiophore. Bipolaris spp. conidial attachment to conidiophores, conidial hilum, and conidia were all seen (Supplementary Figure S3).

Morphological Identification of Curvularia Isolates (Maize Leaf Spot)
A total of 129 maize-leaf-spot-symptomatic samples were collected from 20 sampling districts from five major maize-growing regions and 74 Curvularia isolates were identified based on their morphological characteristics (Table 4). Seventy-four isolates were divided into five groups based on their cultural and morphological characteristics. Group F consisted of 15 isolates, representing 20.27%, Group G included 8 isolates, accounting for 10.81%, Group H consisted of 21 isolates, accounting for 28.37%, Group I had 14 isolates, accounting for 18.91%, and Group J consisted of 16 isolates, which accounted for 21.62%. A brief description of the cultural and morphological characteristics of Curvularia isolates as described in F-J groups can be found in Table 4. Different colony morphology was found on PDA for each distinct group after 7 days of incubation. Group F isolates had black colonies and smooth velvety mycelia with regular margin; Group G isolates had colonies that were gray in color with smooth floccose mycelia and irregular margins. The isolates in Group H produced gray colonies with smooth appressed mycelia and irregular margin. The colonies from Group I were blackish gray with smooth appressed mycelia and regular margin. The colonies produced by Group J isolates were whitish gray with smooth velvety mycelia and regular margin. Conidial morphology: conidial types were observed among the five groups based on the following conidia shapes, color and length-width ratio: The length-width ratio of Group I was larger than those of Group G and Group J. All Group isolates produced 3-5 distoseptate conidia with light-brown to brown color (Supplementary Figure S4).

Pathogenicity of the Isolate
Koch's postulates experiments were performed for the isolates and were positive for causing MLB disease and leaf spot of maize. The maydis leaf blight symptoms appeared as small yellow necrotic spots that later become spindle-or elliptical-shaped lesions and were observed in the inoculated plants, Zea mays c.v. Kanchan (Supplementary Figure S5a). In maize leaf spot, the symptoms appeared as yellow necrotic spots in the early stages of maize leaves, which eventually expanded to circular, oval, or strip lesions formed after 10 days of inoculation, identical to those seen in the field for leaf spot of maize. (Supplementary Figure S5b).

Phylogenetic Study of Bipolaris Isolates (Causal Organism of Maydis Leaf Blight)
PCR-amplified product of Bipolaris spp. and Curvularia spp. using primer pairs ITS1/ITS4, GAPDH1/GAPDH2, D1/D2 LSU yielded specific amplified products of approximately 650 bp (ITS), 472 bp (gapdh), and 917 bp (LSU). Maximum likelihood analyses were performed both individually and in combination based on the collected sequenced data to produce phylograms. Forty-three ex-type closely related species of Bipolaris and Curvularia were used as reference sequences. The isolates were characterized at molecular level using thee three above-mentioned gene loci; upon NCBI BLAST search, they showed 99-100% similarity with concern pathogen and were identified as Bipolaris maydis and Curvularia lunata. These isolates were submitted to GenBank and IDA (International Depositary Authority)-approved culture collection center NAIMCC after approval of cultures authenticity from the authority. Accession no. of culture collection is mentioned and described in Table 2.

Figure 1.
Dendrogram constructed from maximum likelihood method based on combined internal transcribed spacer (ITS). Sequenced data of the Bipolaris maydis isolates were inferred with Raxml based on GTR + Gamma model. Numerical value presented in the node indicates bootstrap value with a 1000 non-parametric bootstrap replicate analysis. The symbol • denotes as 'ex-type' closely related species sequences obtained from GenBank. Alternaria alternata used as an outgroup.

Phylogenetic Analysis of ITS + GAPDH Gene of Bipolaris maydis
A phylogenetic tree was constructed based on the aligned sequences of ITS + GAPDH region to ascertain the phylogenetic relationship between the Bipolaris maydis isolates. The cladogram was divided into two clades. In clade 1, there are two subclades: in subclade 1Aa, the Bipolaris maydis isolates AR5182, AR5183, F28 were clustered together with a bootstrap value of 100% and CPC28823, E19, E15, E10, E25, E30, E38, E26, E7, E24 formed a sister clade with them, whereas subclade 1Ab consists of E50 isolate, which formed a distinct clade. Subclade 1Ba consists of E27 isolate with a bootstrap value of 80%, subclade 1Bb consists of CBS307.64, CBS136.29, CBS130.26, which were clustered together with a bootstrap value of 100%, whereas C5, C4 formed a sister clade with them, with a bootstrap value of 100%. Subclade 2 consists of E3 isolate. Alternaria alternata E11 was used as an outgroup in clade 2 ( Figure 2).

Phylogeny Based on the ITS + GAPDH of Curvularia lunata
Based on the ITS + GAPDH region, the Curvularia lunata isolates formed two clades formed. In clade 1, there are two subclades, 1 and 2. In subclade 1A, E32, E33 isolates were clustered together with a bootstrap value of 81%, whereas E36, E37 formed a sister clade with them with a bootstrap value of 73%; E13, E41 were clustered together with a bootstrap value of 94% and formed a sister clade with them. In subclade 1B, E35, E14, E39, were clustered together. In subclade 1C, E28, E6, E12, E34 were clustered together with a bootstrap value of 98%. Subclade 1Da consists of E16 with a bootstrap value of 91% and subclade 1Db consists of E31 with a bootstrap value of 84%, whereas in subclade 1Dc, CBS 730.96, CBS 157-34 were clustered together with a bootstrap value of 99% and MFLUCC 10-0706, MFLUCC 10-0695 were clustered together with a bootstrap value of 100%. Subclade 2 consists of Curvularia lunata isolate E43. In clade 2, Alternaria alternata (E11) was used as an outgroup that formed a separate clade ( Figure 6).

Phylogeny Based on the ITS + LSU of Curvularia lunata
Based on the ITS + LSU region, two clades were formed; in clade 1, two subclades formed, subclade 1 and subclade 2. In subclade 1Aa, Curvularia lunata isolates E31, CBS 730.96, CBS 157-34, MFLUCC 10-0695 were clustered together with a bootstrap value of 90%, MFLUCC 10-0706 formed a sister clade with a bootstrap value of 94%, and E16 formed a sister clade with a bootstrap value of 97%. In subclade 1Ab, E32, E41, E33 were clustered together with a bootstrap value of 76% and E39, E35, E36 formed a sister clade with them, whereas E37, E13, E14 formed a distinct clade with a bootstrap value of 88%. Subclade 1Ca consists of E28 with a bootstrap value of 61%, whereas in subclade 1Cb, E34, E12 were clustered together with a bootstrap value of 97% and E43, E6 were clustered together with a bootstrap value of 69%. In clade 2, Alternaria alternata (E11) was used as an outgroup to form a distinct clade (Figure 7).

Phylogeny Based on the Multigene of Curvularia lunata
A cladogram was constructed based on the concatenated sequences of ITS, LSU, and GAPDH gene regions. Two clades were formed, clade 1 and clade 2. In clade 1 there are two subclades, subclade 1 and subclade 2. In subclade 1, Curvularia lunata isolate E35 formed a distinct clade. In subclade 2Aa, E28 formed a distinct clade with a bootstrap value of 100%. In subclade 2Ab, E34 formed a distinct clade, E43, E6 were clustered together with a bootstrap value of 76%, and E12 formed a sister clade with them. In subclade 2Aca, two clades formed. In subclade 2Ac, E31 formed a clade with a bootstrap value of 99%; in subclade 2Acb, MFLUCC 10-706, MFLUCC 10-0695 were clustered together with a bootstrap value of 99% and CBS 157-34, CBS 730.96 were clustered with each other and formed a distinct clade with a bootstrap value of 96%, whereas E16 formed a sister clade with them with a bootstrap value of 83%. In subclade 2Ad, E14, E33 were clustered together and E32 formed a sister clade with them, whereas E41, E13 were clustered together with a bootstrap value of 66%, and E36, E39 were clustered together with a bootstrap value of 71% and E37 formed a sister clade with them. In clade 2, Alternaria alternata (E11) was used as an outgroup formed a distinct clade (Figure 8).

Phylogenetic Studies between Bipolaris and Curvularia Complex
Based on the combined sequences of ITS gene regions of Bipolaris maydis and Curvularia lunata, a cladogram was constructed based on the multilocus sequences to know the phylogenetic relationship among the 26 isolates of Bipolaris maydis and 21 isolates of Curvularia lunata. Rhizopus oryzae was used as an outgroup, and 43 ex-type closely related Curvularia and Bipolaris species were used as a reference sequence.
In clade 2, Rhizopus oryzae (CBS 330.53) is considered as an outgroup that is entirely different from Helminthosporoid genera ( Figure 11).

Discussion
Bipolaris and Curvularia species that are anamorphs of Cochliobolus species are important pathogens of the Poaceae family. Leaf spots, blight, root rot, and crown rots are diseases caused by these pathogens [11,12,38]. Payak and Sharma (1981) reported 61 diseases in maize [65]. Among the diseases, maydis leaf blight causes yield losses of up to 41 percent [20]. Considerable phenotypic, pathogenic, and genetic variations occur among the pathogens under the genus Helminthosporium [66][67][68]. Bipolaris maydis, previously called Helminthosporium maydis, also shows enormous variability. The studies on diversity analysis based on morphological variation and phylogenetic basis in B. maydis-Zea mays pathosystem have not been carried out very extensively. Keeping this fact in mind, the present study was undertaken, and the results are here discussed. The present work focused on elucidating the biodiversity of the Bipolaris maydis, causing maydis leaf blight, and Curvularia lunata, causing maize leaf spot. To this end, a vast collection of Bipolaris maydis isolates obtained mainly from maize from the major maize-growing regions of Bihar, Punjab, Uttar Pradesh, and Uttarakhand were characterized based on morphological and molecular characteristics.
Seventy-seven maydis leaf blight isolates were morphologically characterized and formed into five groups based on their morphology. The data showed that different isolate lengths and widths of conidia varied between 65.38-80. 44 Leonard and Suggs, 1974 [14,69,70]. Comparison of the conidium sizes of the isolates of Bipolaris maydis and Curvularia lunata examined in this study was consistent with results from previous studies [12,71,72]. The number of septa of various isolates ranged between four and nine, and the average number of septa was highest for Group E (5-9); in Group D, the number of septa was smallest (4)(5). The isolates showed two textural classes, namely rough and smooth, and colony pattern was either appressed or raised type. Group A and Group E isolates showed rough texture with raised colony pattern, whereas Group C showed smooth texture with an appressed colony. Group A, B, C, and E grown on PDA showed regular margins, while Group D exhibited irregular margins. Similar variations in number of septa among the isolates of Bipolaris maydis were found in the same agroclimatic zones [14,70,73].
In Curvularia lunata isolates, Group H had the largest size of conidia, having a mean length of 29.15 µm and width of 11.96 µm, whereas in Group I, the size of conidia was smallest (19.69 × 9.58 µm) and showed a distinct variation among the isolates based on conidia length and width. The distoseptate conidia were smooth-walled and curved at subterminal cell, three-septate, four-celled, and brown. The Group H, and Group I produced grey, smooth mycelia, whereas Group F produced black smooth velvety mycelia. These morphological characters are similar to the characters of Curvularia lunata-caused leaf spot of maize described earlier [74,75] and standard descriptions from Ellis (1971) [10].
Our study mainly highlights the correct way of identifying of Bipolaris maydis and Curvularia lunata isolates, as both show complex symptoms in field conditions that can cause confusion in their identification. In this study, we screened out indigenous isolates based on morphological characteristics and ribosomal markers ITS and LSU, while proteincoding GAPDH genes were used for molecular identification and phylogenetic analysis. A phylogenetic tree was constructed based on the concatenated aligned multigene sequences of ribosomal markers ITS and LSU and protein-coding GAPDH gene regions of Bipolaris maydis isolates and found that maydis leaf blight isolates of Bihar, Uttar Pradesh, and Punjab isolates were clustered together, whereas Uttatrakhand isolates were clustered only with Bihar and Uttarpradesh isolates. In other words, Upper Gangetic Plain region and Middle Gangetic Plain region isolates of Bipolaris maydis showed similarity with each other. Whiles Bipolaris maydis isolates constructed phylogeny only on the basis ITS sequences, the Bihar, Uttarpradesh, Punjab, and Uttarakhand isolates clustered together, and the aligned multigene sequences of ribosomal marker ITS and protein-coding GAPDH gene regions revealed that the isolates of Bipolaris maydis from other countries such as Japan, China, USA, and the Netherlands did not cluster with the Indian isolates, but when phylogeny was constructed based only on ITS sequences, Bipolaris maydis isolates from Japan, China, USA, and the Netherlands clustered with the Indian isolates. This indicates that nuclear ribosomal DNA (rDNA) internal transcribed spacer (ITS) sequences alone could not differentiate the isolates based on their geographical locations, but that combined ITS, LSU, and GAPDH phylogeny was able to differentiate among the isolates based on said locations. A cladogram was constructed based on the concatenated sequences of ribosomal markers ITS and LSU and protein-coding GAPDH gene regions of Curvularia lunata isolates. The combined analysis of ITS, LSU, and GAPDH phylogeny of Curvularia lunata isolates revealed that most of the isolates of Punjab and Uttarakhand were clustered together, whereas Bihar isolates only clustered with Uttarpradesh isolates. Similarly, the Curvularia lunata isolates of Thailand, Indonesia, and USA were clustered together but did not cluster with Indian-origin isolates. However, when phylogeny was made on the basis of ITS sequences or combined sequences of ITS and LSU the Curvularia lunata isolates of Thailand, Indonesia, and USA were clustered with Bihar and Uttarpradesh isolates. Similar observations were found as reported earlier [11,14,29].
The only morphological distinction between Curvularia and the helminthosporioid genus Bipolaris is the conidial curvature and length. Both genera have species with intermediate morphology, requiring sequencing data to distinguish them adequately. The overlapping morphological features of several Curvularia taxa make species identification difficult [20,66]. Only morphological evidence with the subjective determination of phenotypic characteristics such as spore size has been used to identify Bipolaris and Curvularia spp., while some were identified entirely on the basis of host relationship. Several duplicate sequences of Bipolaris spp. with different names were submitted in GenBank. When searching for generic placement, ITS BLAST search should be used because it is the most efficient. Notably, Xue et al. (2016) included eight species of Bipolaris in their phylogenetic analysis, which led to incorrect identification of two of the three strains. Therefore, sequencing data are critical for reliable species identification with ITS, GAPDH, and LSU gene regions [28], although the ITS and GAPDH alone can still resolve the majority of taxa in Curvularia [26]. Although the ITS regions alone can resolve the majority of taxa in Curvularia [26], sequences of GAPDH and LSU regions are needed to clarify if it is a different species. In our study, for accurate identification of Curvularia lunata and Bipolaris maydis isolates, phylogenetic analyses using multi-genes (ITS or LSU) or protein-coding genes such as GAPDH were performed; the multigene analysis highlights the similarity and relationships between different isolates from different geographical locations [26,38,43].

Conclusions
It is difficult to differentiate sister genera (Bipoplaris and Curvularia) solely based on morphological examination, because many of their morphological features overlap. Therefore, other identification methods, such as molecular identification, are necessary for identification. Here, we used GAPDH, a protein-coding gene, and other ribosomal markers like ITS or LSU for phylogenetic analysis in Bipolaris and Curvularia spp. to ensure precise identification. We used precise taxon sampling for accurate results by adding 43 ex-type closely related Bipolaris and Curvularia spp. sequences from GenBank that derived from type material to identify the delineate species in Bipolaris and Curvularia. Based on the phylogenetic tree, multigene analysis (ITS + LSU + GAPDH) yielded accurate identification of fungal isolates such as B. maydis and C. lunata. While the multigene analysis highlights the similarity and relationships between the different isolates from different geographical locations, the information is comparatively degenerate in single-gene-based phylogeny. In this study, an unclear bifurcation of Drechslera genera was also observed in the case of (ITS + GAPDH) phylogeny, as Drechslera erythrospila CBS108941 were clustered with Curvularia lunata, but this case was resolved after using a multigene phylogenetic approach (ITS + LSU + GAPDH). Upon so doing, the Drechslera genera formed a separate cluster, which gives a clue that single-or oligogene-based molecular identification may be incorrect sometimes for correct identification. Nonetheless, multiple-gene-based concatenated phylogenies are much accurate and useful for identification in cases such as the sister genera Bipolaris-Curvularia complex, and they also can be utilized to differentiate isolates of fungi from different geographical locations. This research contributes to understanding Bipolaris maydis and Curvularia lunata maize disease complexity and offers an approach for correct diagnosis using concatenated gene sequence analysis. We found that B. maydis and C. lunata were the dominant species infecting maize in all geographical locations surveyed of major maize-growing regions. Future studies should focus on expanding genetic information on Bipolaris and Curvularia spp., and proper identification should continue to rely on molecular data, not on morphological parameters. The habitat and the host range of many Bipolaris species are incompletely understood. Extensive sampling and accumulation of molecular data will improve understanding of the host range and ecological significance of Bipolaris and Curvularia spp.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof8080802/s1, Figure S1: Sampling district location shown in Map of India.; Figure S2: Representative Bipolaris isolates showing the morphological description from Group A to E; Figure S3: Scanning electron microscope showing conidiophore, conidia and hilum of Bipolaris isolates; Figure S4