Fungal Species Causing Maize Leaf Blight in Different Agro-Ecologies in India

Foliar diseases of maize cause severe economic losses in India and around the world. The increasing severity of maize leaf blight (MLB) over the past ten years necessitates rigorous identification and characterization of MLB-causing pathogens from different maize production zones to ensure the success of resistance breeding programs and the selection of appropriate disease management strategies. Although Bipolaris maydis is the primary pathogen causing MLB in India, other related genera such as Curvularia, Drechslera, and Exserohilum, and a taxonomically distant genus, Alternaria, are known to infect maize in other countries. To investigate the diversity of pathogens associated with MLB in India, 350 symptomatic leaf samples were collected between 2016 and 2018, from 20 MLB hotspots in nine states representing six ecological zones where maize is grown in India. Twenty representative fungal isolates causing MLB symptoms were characterized based on cultural, pathogenic, and molecular variability. Internal Transcribed Spacer (ITS) and glyceraldehyde-3-phosphate dehydrogenase (GADPH) gene sequence-based phylogenies showed that the majority of isolates (13/20) were Bipolaris maydis. There were also two Curvularia papendorfii isolates, and one isolate each of Bipolaris zeicola, Curvularia siddiquii, Curvularia sporobolicola, an unknown Curvularia sp. isolate phylogenetically close to C. graminicola, and an Alternaria sp. isolate. The B. zeicola, the aforesaid four Curvularia species, and the Alternaria sp. are the first reports of these fungi causing MLB in India. Pathogenicity tests on maize plants showed that isolates identified as Curvularia spp. and Alternaria sp. generally caused more severe MLB symptoms than those identified as Bipolaris spp. The diversity of fungi causing MLB, types of lesions, and variation in disease severity by different isolates described in this study provide baseline information for further investigations on MLB disease distribution, diagnosis, and management in India.


Phylogenetic Analyses
The final ITS alignment file had thirty taxa and 483 characters of which 84 were parsimony informative representing 17% of the total characters. The GADPH file had 23 taxa and 512 characters of which 118 were parsimony informative representing 23% of the total characters. For both files, reference sequences representing mostly sequences of ex-type cultures of Bipolaris or Curvularia species were obtained from GenBank and were included in Supplementary Material (Table S1). The phylogenetic trees obtained with Bayseian analyses had essentially identical topology to the trees obtained by maximum likelihood (ML). Therefore, for each locus, only one ML tree was selected for this publication, but the posterior probability (PP) support for the branches from the Bayesian trees is reflected on the maximum likelihood tree branches. The maximum likelihood tree shown in Figure 1A based on GAPDH sequence data reveals that three taxa under study formed a highly supported subclade (A) with a sequence of ex-type culture B. maydis. As a matter of fact, there were 10 additional isolates with identical sequences to taxa in clade A which were not included in the analysis due to the tree size. In other words, 13 of 20 isolates were identified as B. maydis. The tree based on ITS sequence data ( Figure 1B) was congruent with the GAPDH tree and revealed the same clade (A) within a large all Bipolaris species clade. One isolate (BmMdKa6) had identical sequence to the ex-type culture of B. zeicola and they together formed a highly supported (BS = 97%; PP = 0.98) subclade B ( Figure 1A). This subclade was sister to the B. maydis subclade with BS support of 100% and PP of 1. Similar to subclade A, isolate BmMdKa6 was also identified as B. zeicola in the ITS tree (subclade B; Figure 1B). In brief, among the 20 isolates identified morphologically and pathologically as fungi causing MLB, 13 isolates (viz. BmPhRj4, BmBjUa1, BmBsRj4, BmDhBh3, BmCgRj4, BmPnDl2, BmDnRj4, BmKgUa1, BmKrHr2, BmKtRj4, BmLdPj2, BmPtUa1, and BmMyKa6) representing 13 hotspots were identified as B. maydis and one isolate (BmMdKa6) was identified as B. zeicola, based on phylogenetic trees of two loci, GAPDH, and ITS.
The remaining taxa in the tree were Curvularia sp. isolates and they formed three main supported clades C, D, and E. ( Figure 1A). Clades C, D, and E had a polytomy relationship to each other and to Bipolaris clades B and A. ( Figure 1A). Within clade E, isolates BmAdGj5 and BmGdGj5 and ex-type culture of C. papendorfii had identical sequences, and together they formed a highly supported subclade (BS = 98% and PP = 0.99 subclade). Isolate BmLhRj4 had an identical sequence to an ex-type culture of C. sporobolicola, and together they formed a highly supported subclade. Isolate BmSkRj4 formed a subclade with two isolates of C. siddiquii, with high support (BS = 0.99 and PP = 1). In the ITS tree, however, the isolates that fell into clade E (BmLhRj4, BmGdGj5, BmAdGj5, and BmSkRj4) formed a highly supported clade with C. lunata and they all had identical ITS sequences. Clearly, phylogeny based on ITS sequence data appears to be unsuccessful in resolving Curvularia isolates to the species level like the GAPDH tree. Isolate BmSmBh3 ( Figure 1A) formed a clade with the sequence of the unidentified species of Curvularia (Accession No. KU552166). The species C. graminicola was shown to be the closest relative to them (Clade D, Figure 1A). Therefore, this unidentified species remained ambiguous, needs additional markers-based phylogeny for clarity, and was named C. graminicola-like fungus for this report. In the ITS tree, isolate BmSmBh3 fell into the large Curvularia clade but did not form a clade with a sequence of any known species of Curvularia.
The isolate BmAmRj4 did not fall into the Bipolaris or Curvularia clades but formed a clade with the ex-type culture of Alternaria alternata in the ITS tree ( Figure 1B). The two sequences had 100% homology. However, we could not obtain the GAPDH sequence for this isolate, and thus it was excluded from the phylogenetic analyses based on GAPDH. In summary, in addition to the 14 isolates of B. maydis and B. zeicola that were causative agents for MLB, we identified two isolates of C. papendorfii, and one isolate each of C. siddiquii, C. sporobolicola, a C. graminicola-like fungus, and an Alternaria sp. as MLB causing fungi in India.

Morphological Characterization of Fungal sp. Causing Maize Leaf Blight
The 20 candidate fungal isolates maintained on Potato Dextrose Agar (PDA) plates showed striking variability in culture morphology and conidial dimensions (Table 2; Figure 2A). The length and width of conidia of different isolates varied between 7.6 µm (BmgGdGj5, Godhra) to 19.3 µm (BmPnDl12, Patel Nagar) and 6.1 µm to 3.1 µm (BmSmBh3, Samastipur and BmgGdGj5, Godhra or BmSkRj4, Sukher), respectively. The isolate BmP-nDl12 (Patel Nagar) had the largest sized conidia with mean length of 19.3 µm and width of 5.0 µm and BmgGdGj5 (Godhra) had the smallest conidia, 7.6 µm × 3.1 µm. The conidial dimensions of the rest of the 18 isolates were in the order BmAmRj4 (Amberi) > BmLhrj5  Table 2).  The number of septa of various isolates ranged between one to three (BmgGdGj5, Godhra isolate) to five to eight (BmBjUa1, Bajaura isolate) (Table 2; Figure 2A). For the rest of the isolates, it was in the order Table 2).
The culture morphology of C. papendorfii showed a rough surface with irregular margins and no zonation, conidial dimensions 7.6 µm × 3.1 µm to 16.9 µm × 4.1 µm, and one or more septa (Table 2) which falls into Group-2 (Table S2); whereas B. zeicola showed smooth appressed surface with regular margins and zonation, conidial dimensions 7.8 µm × 6.1 µm to 13.1 µm × 3.8 µm slightly larger than C. papendorfii, and 3 to 4 septa (Table 2) formed Group-5 (Table S2). The C. papendorfii isolate from Godhra showed a rough raised surface while the Anand isolate showed a rough appressed surface with no zonation and smooth margins. The conidial size was also very small, with 1 to 2 septa, and ranged from 7.6 to 12.1 µm × 3.1 to 4.3 µm. Ecologically both B. zeicola and C. papendorfii prevailed in different parts of the plateau and hill regions which are at high altitudes with cool and moist climates close to the sea (C. papendorfii from the western part having sandy loam soil and B. zeicola from the southern part of the plateau and hill region having laterite soils). Isolates from Lakhawali, Sukher, and Anand showed variable morpho-pathogenic profiles and represented different groups being different species in the genus Curvularia. The conidia of the C. sporobolicola (Lakhawali) isolate were slightly curved with 2 to 3 septa, conidial size ranging from 17.2 µm × 3.7 µm, forming a rough colony with zonation and irregular margins. However, the C. siddiquii (Sukher) isolate had conidial sizes smaller than the C. sporobolicola isolate (i.e., 16.8 µm × 3.1 µm in size) with a slight curvature, 3 to 4 septa, smooth appressed colony, no zonation, and rough margins. The conidial morphology of C. papendorfii was oval or obpyriform with 1-3 septa. The C. graminicola-like sp. (Samastipur) isolate produced sickle-shaped conidia with 2 to 3 septa, ranging in size from 7.8 to 6.1 µm and 2.6 µm average diameter. Alternaria sp. was also isolated from maize in some regions from the Amberi hotspot in Rajasthan.

Pathogenic Variability of Fungal Isolates on Zea mays c.v. DHM 117
Koch's postulates experiments were performed for all 20 isolates, and all were positive for causing MLB disease. The MLB symptoms appeared as small yellowish necrotic spots 3 to 4 days after inoculation of the test plants, Zea mays c.v. DHM 117. Necrotic spots coalesced with the progression of the disease resulting in blight symptoms. The severity of the disease differed between the isolates ( Figure 2C). In general, isolates identified molecularly as Curvularia spp. (SN 15 to 19, Table 2) and Alternaria sp. (SN20, Table 2) caused more severe disease with scores of 3.3 to 4.5 and an average score of 3.91, than isolates identified as B. maydis (SN 1 to 13, Table 2)  The colony texture of isolates correlated with their pathogenic characteristics resulting in the categorization of the isolates into six groups (Table S1; Figure S3). Group-1, represented 64 samples (18.28%) of the total population examined, and had rough colonies with no zonation and irregular margins as shown by isolates BmPhRj4, BmAmRj4, and BmLdPj2. The isolates of this group had mild to moderate virulence, an incubation period of 72 h on the host, and disease severity of 1.5 to 4.3 on the rating scale; the highest virulence being recorded was for BmLdPj2. Group-2 with 71 samples (20.28%), had rough colonies with no zonation and regular margins, as shown by isolates BmBjUa1, BmCgRj4, BmPtUa1, and BmGdGj5, had moderate virulence, an incubation period of 48 to 72 h and disease severity of 2.6 to 3.3. Group-3 with 26 samples (7.42%) had rough colonies with zonation and irregular margins, as shown by BmPnDl12 and BmLhRj4, which had moderate to high virulence, an incubation period of 72 h, and disease severity of 2.7 to 3.7. Group-4 with 110 samples (31.42%), had smooth appressed colonies with no zonation and regular margins, as shown by isolates BmBsRj4, BmDnRj4, BmKgUa1, BmKrHr2, BmKtRj4, and BmSmBh3, had mild to high virulence, an incubation period of 48 to 96 h and disease severity of 1.9 to 3.6. Group-5 with 40 samples (11.42%) had smooth appressed colonies with zonation and regular margins, as shown by isolates BmMdKa6, BmMyKa6, and BmSmBh3 had mild to high virulence, an incubation period of 72 to76 h, and disease severity of 1.7 to 4.3. Finally, Group-6 with 39 samples (11.14%) had rough appressed colonies with no zonation and regular margins and high virulence, a 72 h incubation period, and disease rating of 3.7 to 4.5.
Based on the expression of symptoms (Type-I to -VI, Figure 2C), the isolates were categorized into six groups. The morphological and pathological variations of B. maydis isolates were distributed among all six groups, whereas Curvularia isolates were distributed from Group-2 to Group-6, and Alternaria sp. came under Group-1 (Table S1). B. maydis was the dominant pathogenic species out of the total population of MLB collected across 20 hotspots in six maize production zones being identified in 72.85% of the 350 disease samples collected, followed by Curvularia species being identified in 20.28% of the samples. A few instances of B. zeicola (3.71%), C. papendorfii (3.42%), and Alternaria sp. (3.14%) were also detected from the MLB samples. Mixed infection of B. maydis, C. papendorfii, and Alternaria sp. were noticed from the same leaf in Mysore (Karnataka state), Amberi (Rajasthan state), and Anand (Gujarat state) (data not shown).
Visually the symptoms of B. maydis could be differentiated from B. zeicola in having oval lesions (4 to 5 mm × 7 to 9 mm) which later elongated and coalesced into larger irregular necrotic lesions along the mid rib (9 to 12 cm), whereas B. zeicola produced circular to oval dot-like yellowish necrotic lesions (2 mm × 3 mm) which did not coalesce with the progression of the disease ( Figure 2C). The lesions increased in circumference but remained distinct, increasing in size up to 1.2 cm × 2.5 cm. However, for both the B. maydis and B. zeicola isolates, no lesions could be observed on husks and leaf sheaths, and no wilting was observed in the diseased plants in field observations as well as in the greenhouse inoculation experiments. This may serve as a preliminary indication that the isolates of B. maydis observed in our survey belong to race type "O" (20). B. maydis showed an extensive distribution across all maize cropping zones.
The C. papendorfii isolates formed very minute dot-like, yellowish necrotic spots more profuse than symptoms caused by B. zeicola the lesions showing irregular purplish brown margins (3 mm × 5 mm) which remained discrete and showed no coalescence with progression in size of spots, whereas C. sporobolicola and C. siddiquii formed tan colored streaks on the leaves. While C. sporobolicola showed coalescence of streaks to form large necrotic streaks on the leaf (Figure 2A, plate q), the C. siddiquii isolate formed elongated mosaic-like streaks which remained discrete and did not coalesce. Symptoms of C. graminicola were close to C. papendorfii; however, the lesions were smaller than those formed by C. papendorfii with no purplish margins, and coalescence with the progression of the symptoms was observed. The C. papendorfii and B. zeicola isolates were weakly to moderately pathogenic on maize in the plateau and hill regions of India. Another pathogen, Alternaria sp. was detected to cause MLB in India. Symptoms of Alternaria were visible as discrete oval to irregularly elongated, yellow necrotic spots (1.3 cm × 2.7 cm to 2.5 cm × 3.7 cm), discrete and larger in size than with B. maydis, C. papendorfii, and B. zeicola. With disease progression, lesions coalesced up to 2.7 cm × 3.4 cm in size, but not as large irregular elongations. Symptoms of Curvularia sp. generally were scattered 3 to 5 mm × 6 to 7 mm lesions, visible as mosaic patterns on the leaf lamina, which coalesced to form necrotic regions on leaves.

Discussion
Maize leaf blight (MLB) is listed as a major biotic stress on maize in India, and every year monitoring visits are undertaken to survey the disease-prone areas to examine yield losses due to the disease [1,2]. Precise characterization of the species infecting the maize crop is needed for developing effective strategies to manage the economic losses. In the global climate change scenario, it is very important to examine the trends in yield losses and the severity of MLB on the crop within different cropping zones. In the current investigation, for molecular identification of fungi obtained from MLB samples, we depended primarily on the section of the GAPDH gene, which has been regarded as the best single marker for delineating species of the genus Bipolaris [29]. Additionally, we used phylogeny based on ITS to further support of the results. B. maydis was identified in 13 of 20 hotspots surveyed. This observation supports B. maydis as the dominant MLB pathogen in India [3]. However, B. zeicola was present in one hotspot. These two species were the only Bioplaris species detected in our relatively small sample size for characterizations. Our results are in agreement with the survey in China where they found that these two species accounted for about 97% of Bipolaris species causing diseases in maize [20]. In Yunnan, Sichuan, and Shaanxi Provinces of China, B. zeicola isolates were reported to produce long, narrow linear lesions [8,20,30,31]. However, the pathogenicity of the Indian isolate of B. zeicola is slightly different from the Chinese isolates. Isolates of B. maydis from different hotspots across six maize production zones of India showed mainly Type-I, -II, and -III symptoms (Table 2; Figure 2C). Therefore, elongated, necrotic lesions were the typical symptoms caused by B. maydis, as reported in previous studies [3,28]. Although the association of B. zeicola (synonym B. carbonum) in healthy maize seeds [32] and on healthy rice leaves [33] in India were noted earlier, the reports lacked pathogenicity data of the respective organisms.
B. zeicola is also reported to be a pathogen of rice and maize in China [34] and a pathogen of rice in Pakistan [35].
Of the four reported races of B. maydis viz. O, T, C, and S [14,36], lesions of race O were tan in color with buff to brown borders. They began as small, diamond-shaped lesions and sometimes elongated within the veins to become larger and rectangular. However, race O lesions were contained within the leaves. Lesion size ranges from 2 to 6 mm × 2 to 22 mm. Lesions produced by race T were oval and larger than those produced by race O and isolates of race T commonly affected husks and leaf sheaths. Lesions caused by race C were necrotic and found to be about 5 mm long. They also tended to cause wilt [14]. While it is beyond the scope of this study, based on symptomatology we conclude that the B. maydis isolates observed in our survey belong to race O. Race O was also found to be the predominant pathogen of maize in India by other workers [27,37]. Studies on the virulence of B. maydis and B. zeicola isolates revealed that both are adapted to distinct ecological conditions [31,38,39]. It was reported that race 3 of B. zeicola with narrow linear lesions on the leaves of mature maize plants may have been a mountain ecotype, favoring high humidity and cool temperatures at high elevations for infection. However, in our study B. zeicola produced dot-like, spherical lesions.
In recent years variations in MLB symptoms on maize have been noticed in India. This points towards the likely establishment of other species/races also infecting maize. In substantiation, we report for the first time that along with B. zeicola, four Curvularia pathogens (C. papendorfii, C. sporobolicola, C. siddiquii, C. graminicola-like fungus), and Alternaria sp. are pathogens causing MLB disease in India. Earlier, a leaf spot disease of maize caused by C. clavata and maize leaf spot caused by C. geniculata were recorded in India [25,40]. The presence of C. papendorfii in rice soil [41] has been documented from India but the report lacked any pathogenicity data for the organism. However, the presence of four Curvularia species in five out of 20 molecularly identified pathogens raises concern that Curvularia is an emerging threat to maize in India. C. sporobolicola was shown to be a pathogen of the grass Sporobolus australasicus in Australia [42] and C. siddiquii a pathogen of Pennisetum americanum in India [43], C. graminicola was isolated from Aristida ingrata (Poaceae) in Australia [44], and a taxonomically close relative of the fungus (Curvularia sp. BRIP 61674) was found to be a pathogen of Oryza spp. in Australia [45]. Furthermore, C. papendorfii (synonym: B. papendorfii) was shown to be a pathogen of maize in China [46]. Alternaria species including A. tenuissima, A. alternata, and A. burnsii were shown to cause disease in maize [47,48]. Migration of these pathogenic fungal species to maize may have happened from adjacent sugarcane or rice fields or due to secondary inoculum developed on previous crops in the rotation schedule, which necessitates further investigations.
Previous investigations on the characterization of maize pathogens from MLB disease prone maize production zones indicated the presence of the disease caused by B. maydis in all the maize production zones, especially in Kharif maize [3,28]. C. papendorfii and B. zeicola were characterized by Godhra and Mandya hotspots which were in the plateau and hill region. In addition to the widely distributed B. maydis and minor reports of B. zeicola, some other Bipolaris species such as B. sorokiniana have also been reported in other countries as harmful pathogens of maize [4,49,50]. Interestingly, B. sorokiniana that causes wheat root rot and leaf spot was the dominant species infecting wheat in India [51]. Therefore, the chance establishment of this species on maize, particularly in traditional wheat-maize rotations is probable. B. sorokiniana has been reported from maize fields under wheat-maize rotation in Sichuan [20]. Here it is worth mentioning that reports of B. sorokiniana as a dominant species infecting wheat in India can be an emerging threat to maize because the crop cycle of spring maize has some overlap with the wheat season. Similarly, B. sacchari a pathogen of sugarcane in India [43], was also reported as a pathogen of maize in China [52]. Recently, a new sheath spot disease of maize caused by Waitea circinata var. prodigus has been reported from eastern India [53]. Therefore, regular monitoring in maize fields for the possible presence of new or emerging pathogens along with B. maydis, B. zeicola, Curvularia spp., and Alternaria sp. (this report) are necessary to document various fungi causing MLB in India.
Our studies on pathogenicity supported previous reports, however, severity varied for different isolates. Although isolates of B. zeicola, C. papendorfii, C. sporobolicola, C. siddiquii, and C. graminicola-like pathogens showed weak, moderate to severe virulence on maize, their occurrence is a concern for the Western and Southern Plateau and Hill region. Their incidence may rise in the future with changing environments. Hence, it is essential to test the virulence on maize lines and to establish the lines showing severe infection for advisories to avoid huge maize losses. We suggest a more rigorous screening of maize germplasm under simulating epiphytotic conditions to examine the pathogenicity of species of Bipolaris, Curvularia, and Alternaria at high altitudes and cold regimes. The race diversity of B. maydis, B. zeicola, Curvularia spp., and Alternaria sp. on maize also needs to be investigated to avoid complications of possible mixed infection.
Taken together, we explored the species diversity of fungi causing MLB in maize production zones of India based on cultural morphology and symptoms on the host. We found that B. maydis was the dominant species infecting maize in all geographical locations surveyed. The isolates of B. maydis also showed a variation in symptoms on Zea mays c.v. DHM 117. The B. zeicola, C. papendorfii, C. sporobolicola, C. siddiquii, C. graminicola-like fungus, and Alternaria sp. are six new species we identified to cause MLB in India. Of these C. sporobolicola, C. siddiquii, C. graminicola-like fungus are probably the first worldwide reported as pathogens of maize. The symptoms of the four Carvularia species and the Alternaria sp. recorded on maize in India were more severe than the dominant species B. maydis probably due to more rigorous crop management strategies against the target species. These species recorded in the study reported here were occasionally able to cause mixed infections in the field but were distinguished by pure culture isolations and symptom expressions on test plants. These findings may contribute greatly to the understanding of the species diversity in the maize production zones of India and aid in the diagnosis of MLB pathogens and management.

Collection and Maintenance of Isolates
Maize leaves showing characteristic maize leaf blight (MLB)-like symptoms (n = 350) were collected from maize fields across the different agro-climatic zones of India covering the states of Uttaranchal, Himachal Pradesh, Karnataka, Haryana, Delhi, Rajasthan, Bihar, Gujarat, and Punjab ( Figure S1). Surveys were undertaken in disease prone areas to collect different fungal isolates from maize showing leaf spots and blights (Table 1). Symptomatic samples were thoroughly washed in sterile water and 1-2 mm bits of infected leaf tissue showing lesions were cut and surface sterilized using 2% sodium hypochlorite for 5 min, washed with sterile water, and blotted dry. The sterilized bits were then transferred aseptically into Petri plates containing PDA These plates were incubated at 25 ± 1 • C in a BOD Incubator (REMI Cl-10, Mumbai, India). Pure cultures of isolates were established by single spore isolation and examined under a light microscope (Olympus BX-53, Tokyo, Japan) to study characteristic features. The fungal isolates were maintained at standard storage conditions on PDA slants for further studies. From a total of 350 samples analyzed, 20 fungal isolates representing candidate isolates for each location surveyed were further examined to assess the MLB causing fungal species diversity.

Fungal DNA Extraction, Amplification, Sequencing and Phylogenetic Analysis
Growth plugs (10 mm diameter) from actively growing 7-day-old cultures of 20 fungal isolates (Table 1) maintained on PDA were inoculated into 100 mL of potato dextrose broth media in Erlenmeyer flasks and incubated at 25 • C in a BOD incubator, with shaking at 100 rpm. Mycelia were harvested with a sterile Whatman No. 4 filter disk and Buchner funnel attached to a vacuum flask. Then, mycelia were washed with sterile distilled water, blotted dry between layers of tissue paper, wrapped in aluminum foil, frozen in liquid nitrogen, and stored at −80 • C until needed.
Genomic DNA was extracted using the Cetyltrimethylammonium bromide (CTAB) method [54]. DNA concentrations were determined using a Pico-drop Spectrophotometer (Pico-drop Ltd., Cambridge, UK), and the concentration adjusted to 200 ng/µL. DNA solutions were stored at −80 • C until used.
PCR was performed to amplify the internal transcribed region of the rRNA gene using primers ITS1 and ITS4 [55]. The PCR mixture (50 µL) contained 10 µL of 5× PCR Go TaqFlexi buffer (Promega Corp., supplied by Pragati Biomedical, New Delhi, India), 0.2 mM dNTPs, 0.2 µM of each primer, and 1.25 units of Taq Polymerase GoTaq Flexi (Promega Corp., Pvt. Ltd., Mumbai, India). Reaction mixtures in PCR vials were placed in a thermocycler (CT-100 Bio-Rad, Gurugram, India). The program used for amplification consisted of an initial 2 min denaturation at 94 • C, followed by 35 cycles of 1 min denaturation at 94 • C, 1 min annealing at 55 • C, and 1 min extension at 72 • C. The program was concluded with a 10 min extension at 72 • C. PCR amplification was checked by gel electrophoresis on 1% agarose gels stained with ethidium bromide. PCR products of 439 to 587 bp were excised from the gel and cleaned using the Wizard ® SV Gel and PCR Clean-Up System (Promega Corp., Madison, WI, USA). Cleaned PCR products were sequenced using an ABI 3730 XL Sequencer (Xcleris Labs Pvt. Ltd., Ahmedabad, India).and the BigDye Terminator cycle sequencing kit (Applied Biosystems, Foster City, CA, USA). Products were analyzed directly on a 3730 XL DNA sequencer (Applied Biosystems). Both DNA strands were sequenced with primers ITS1 and ITS4 in separate reactions. PCR for amplification of GAPDH was performed as described for ITS above except that the annealing temperature used was 52 • C and the primers were GPD-1 and GPD-2 (12). All sequences were submitted to GenBank (Table 1) and subjected to the Basic Local Alignment Search Tool (BLAST) available at http://blast.ncbi.nlm.nih.gov/Blast.cgi (accessed on 12 October 2021).

Phylogenetic Analysis
ITS sequences for 20 isolates under study plus additional sequences of reference strains obtained from GenBank (Table S1) were aligned using MEGA X [56] and then manually adjusted if needed in Mesquite [57]. Ends of the alignment were cut to make the analysis on common regions for all the taxa. Phylogeny trees derived from ITS sequences were constructed using maximum likelihood with substitution model K2 + g obtained by MEGA. Support for the branches was obtained with bootstrap, 500 replicates. Phylogeny trees were also obtained with Mrbayes (3.2.7a) [58]. The Bayesian analysis used the DNA substitution model of Kimara 2 parameter (K2 + G, nst = 2) with gamma distribution determined using MEGA X. Four chains and 1,000,000 Markov chain-Monte Carlo generations were run, and the current tree was saved to a file every 1000 generations. The stability of likelihood scores was confirmed with the plot of likelihood score versus generation number in Microsoft Excel. 25% of the initial trees were discarded as the burn-in phase. Posterior probabilities of above 95 were considered significant support for the clades. The maximum likelihood and the Bayesian trees were rooted to Pyrenophora chaetomioides.
GADPH trees of the twelve GenBank accessions (Table S1) and the 19 isolates under investigation were also obtained with two methods, maximum likelihood with MEGA X and Bayesian, as described for ITS. The models used for both trees were K2 + G, obtained with MEGA X. Character status of the data was obtained with MEGA X.

Cultural and Morphological Variation
Single-spore-purified fungal cultures from maize leaf spots were maintained on PDA in 100 mm × 15 mm sterile polystyrene Petri plates (Fisher Scientific, Thane, India). For observations of morphological variability, 5 mm plugs of the seven-day-old culture of the isolates were placed in the center of the PDA plates and incubated at 27 ± 1 • C in a BOD incubator with alternate light and dark for 12 h daily. Observations were recorded in triplicate. Morphometric variations in the size of conidia and number of septa were examined at 100× magnification with the BX-53 microscope fitted with a camera and imaging software. The length and width of conidia as well as the number of septa were observed microscopically and compared with the identification key of Manamgoda et al., 2014 [4]; measurements being taken using Biovis Image Plus Software with advanced image analysis and image processing tools (developed by Expert Vision Labs, Mumbai, India). Averages of 10 conidia in a microscopic field are presented in Table 2. The observations on colony color and texture were recorded for 10 days after inoculation (DAI) from the top and bottom sides of the culture plates. The isolates were designated to different groups based on cultural characteristics. Observations on radial growth patterns were recorded at 24 h intervals and the final growth at 10 DAI is presented in Table 2. Average radial growth was recorded, and the cultures were assigned as fast (+++), medium (++), and slow (+) groups (Table 2).

Pathogenic Variability
Large-scale growth of inoculum was done on sorghum grains [59]. Sorghum grains were thoroughly washed in tap water, sterilized in 2 % sodium hypochlorite for 5 min, and soaked overnight in distilled water. Excess water from the soaked grains was drained off through several layers of cheesecloth, the grains dispensed into aliquots of 100 g each in 250 mL Erlenmeyer flasks, and autoclaved at 120 • C for 40 min. The sorghum grains were inoculated with bits (5 mm diameter) of actively growing culture of each of the fungal isolates. Cultures were shaken after every two days and incubated for 10 to 15 days at 28 • C. Colonized sorghum grains were dried under shade for 7 to 10 days and powdered. Simultaneously, seeds were planted in 20 inch-diameter pots in a sterilized soil mixture containing vermiculite, coco peat, and sand (2:2:1) in four replications of six pots each.
The Koch's Postulates test [60] for pathogenicity of the 20 fungal isolates was conducted on the susceptible maize inbred line DHM117 in the greenhouse under controlled conditions (28 ± 2 • C temperature, relative humidity 85%, and 16 h photoperiod). Inoculation of maize seedlings was done using powdered sorghum grains. Twelve seedlings in six pots were inoculated twice viz., at the seven-leaf stage and eight days after the first inoculation, with a small pinch of inoculum (about 100 mg) applied in the leaf whorl as per standard techniques for disease resistance screening [61] (Figure S2). Twelve control plants were treated with sterilized healthy sorghum seed powder only. Pots were covered with polythene bags to maintain the desired humidity. The plants were observed daily, and disease scoring was done up to 20 days after inoculation. The inoculation experiment was performed a total of three times. Disease scoring was done using the rating scale of Payak and Sharma [58], which is based upon the severity of the infected leaves after 20 days of inoculation (1 to 5 rating scale) ( Figure S2B).

1.
Very mild infection, as 1 to 2 or more scattered lesions on lower leaves of the host.

2.
Moderate infection showing few lesions on lower leaves only of the host.

3.
Moderate infection, with abundant lesions on lower leaves, spreading up to middle leaves and extending to upper leaves of the host.

4.
Severe infection showing abundant lesions on lower and middle leaves, extending to upper leaves of the host.

5.
Intense severity with abundant lesions on almost all the leaves showing premature drying or necrosis of infected leaf tissue.
Pathogens from the diseased leaf spots following inoculations were reisolated and observed to have similar morphology of the respective fungal inoculum.

Statistical Analysis
The data from cultural, morphological, and pathogenic variability was analyzed statistically to derive significance by SAS Ver 10.0. (SAS Institute, Cary, NC, USA), with desired statistic estimates such as Means, Standard Error (SE), Standard Deviation (SD), and Coefficient of Variation (CV). Percent Disease Index (PDI) was calculated based on an average of 10 replications using the formula: PDI = (Sum of all the ratings/maximum disease rating) × 100.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/pathogens10121621/s1, Figure S1, Hotspot locations covered under survey and surveillance for monitoring maize leaf blight symptoms, Figure S2A, Inoculation of fungal isolates to the leaf whorl in test plants, Zea mays at 7-leaf stage to examine the symptoms and pathogenicity; Figure S2B, Disease rating scale of Payak and Sharma [59], Figure S3, Distribution of morphological and pathogenic variability in the total population of fungal species sampled from 6 maize production zones of India; Table S1: Description of fungal species, strain numbers, hosts, counties, and GenBank accessions of ITS and GAPDH used as references in the study. Table S2, Distribution of morphological and pathogenic variability in the total population of maize leaf blight isolates sampled from six maize production zones of India.