A highly contiguous genome sequence of Alternaria porri isolate Apn-Nashik causing purple blotch disease in onion

Abstract

Objectives

Purple blotch, caused by the necrotrophic pathogen Alternaria porri, is one of the most economically significant diseases of onion and allied crops. While the virulent nature of many Alternaria spp. has been identified, the pathogenic repertoire of A. porri is still unknown. The objective of this work was to sequence the genome of A. porri using the PacBio SMRT sequencing strategy and analyse the repertoire of CAZymes, secondary metabolites, secretome and effectors in A. porri. Our research group is working to identify onion germplasm with purple blotch resistance and to understand the genetics of the pathogen. The reported de-novo assembly will contribute to the analysis of potential variants and the gene repertoire contributing to the virulence and pathogenicity of the purple blotch pathogen.

Data description

Long-read sequencing on a PacBio Sequel II system resulted in a 32.98 Mb (20 contigs) assembly with an N50 of 2, 657, 264 bp, the longest contig length of 5.05 Mb, and a GC content of 51.06%. The Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis resulted in 99.7% genome completeness at the Dothideomycetes lineage, representing a high-quality genome assembly. AUGUSTUS ab initio analysis resulted in 9875 protein-coding genes. Of the 6776 pathogenicity-related genes, 537 genes with effector functions were identified. Likewise, the glycoside hydrolases (434) were the most dominant group of the total 837 predicted CAZymes. The assembled genome of A. porri showed distinctive similarities to the genomes of A. alternata and A. brassicicola, the causal agents of leaf blight of onion and leaf spot of Brassica crops, respectively.

Objective

Purple blotch (PB), caused by A. porri (Ellis) Ciferri, is one of the most destructive diseases of onion and allied Allium spp. [1]. The disease often affects the leaves and bulbs of the plants leading to heavy yield losses ranging from 76 to 97% [2]. Under favourable conditions, the small sunken lesions enlarge and produce brownish-purple necrotic lesions in the leaves, and subsequently wilt, delaying the creation and maturation of the bulb. Purple Blotch disease is especially severe in onion grown in high humidity and moderate temperature of 80–90% and 25–30 °C, respectively. Management of PB through the application of fungicides is only partially effective besides accruing environmentally hazardous residues in the crop and ecosystem [2]. Therefore, the identification of resistant cultivars through host resistance breeding is the most efficient method to control PB disease. However, wide-range screening for resistance to A. porri in cultivated onion germplasms has not resulted in any potential source of resistance [3].

The molecular determinants responsible for the pathogenicity of A. porri are fairly unknown. Similar to other pathogenic fungi, Alternaria spp secretes a broad range of host-specific (HSTs) and non-host-specific (nHSTs) phytotoxins that act as significant virulence factors during pathogenesis [4]. These are mostly the low molecular weight secondary metabolites that causes leaf necrosis of susceptible cultivars. The HSTs have been reported in at least 12 A. alternata pathotypes [5]. Likewise, nHSTs like tenuazonic acid, brefeldin A, zinniol and tentoxin have been reported from many other Alternaria species [4]. Alternaria porri has been reported to produce secondary metabolites such as altersolanol A, alterporriol, macrosporin, erythroglaucin and tentoxin [6]. However, their role in the pathogenicity of PB fungus is yet to be established.

The advent of genome sequencing and comparative analysis has largely contributed to the identification of species-specific pathogenicity factors in closely related species [7]. Currently 169 Alternaria genomes of 31 species and nine unclassified species including A porri is available in the GenBank, that has contributed immensely to the taxonomy and molecular physiology of the genus [4]. However, the species-specific pathogenicity factors that could confer the ability to infect onion and allied crops are yet to be carried out. The currently available genomic information for A. porri has been generated through short-read sequencing and thus is highly fragmented and laconically inconclusive [4]. The third generation long-read sequencing approaches, including the single molecule real time (SMRT) sequencing and Oxford nanopore sequencing, has facilitated the generation of high quality and complete genome assemblies with greater accuracy [8]. Therefore, the objective of this work was to sequence the genome of A. porri using the PacBio SMRT sequencing strategy and analyse the repertoire of pathogenic effectors in A. porri.

Data description

The detail description of the methodology is represented in data file 1. The Alternaria porri super virulent isolate Apn-Nashik was obtained as a hyphal culture (Table 1, Data file 2) from infected leaves of onion variety Agrifound Rose (Table 1, Data file 3) through the tissue segment approach was used for whole genome sequencing. The purified genomic DNA isolated from the fungal mycelium (diploid) was used for generating long read sequencing libraries through the PacBio single molecule real time (SMRT) sequencing technology. The sequencing resulted in 199 GB of data comprising 2,115,917 polymerase raw reads with 113x coverage. A total of ~ 30 Gb filtered polymerase read of unique single molecules were assembled through Improved Phased Assembler (IPA) that produced 33.5 Mb primary (33 contigs) assembly. After the elimination of low-quality reads and adapter sequences using the SMRTlink tool, an assembly of 20 contigs was generated. The final assembly consisted of 32,989,564 bp with N50 of 26,57,264 bp, longest contig length of 5.05 Mb, and GC content of 51.06% (Table 1, Data file 4). These data were found well within the predicted genome size but more than those reported by Dang et al. [4]. A total of 10,699,75 circular consensus sequence (CCS) reads were aligned against the nearest Alternaria genome (Accession ID: GCA_014751505.1_ASM1475150v1) using the minimap2 program and resulted in 100% mapped reads. The BUSCO analysis resulted in 99.7% genome completeness (including 99.5% complete + 0.2% duplicated) at the Dothideomycetes lineage (Table 1, Data file 5), representing a high-quality genome assembly. A total of 3786 BUSCO genes were searched against the Dothideomycetes lineage, from which 3768 were categorized as complete and single copy, six genes as complete and duplicated, and one fragmented, while 13 genes were classified as missing.

Table 1 Overview of data files/data sets

Estimation of repeat content using the RepeatMasker revealed 2784 repetitive elements consisting of retroelement (209), LTRs (186), DNA transposons (52), Gypsy/DIRS1 (140), Ty1/Copia (44) and simple repeats (3065) (Table 1, Data file 6). Further, the simple repeats included 1536-di, 932- tri, 618-tetra, 249-penta and 103- hexanucleotide repeats. While a majority of filamentous plant pathogens have higher repeat content, A. porri genome accounted for only ~ 1.5% repeats, which is significantly lower compared to all the other Alternaria species sequenced to date [4, 9]. In addition, tRNAscan identified 289 tRNAs in the A. porri genome. This suggests that A. porri is a relatively new species in the evolutionary development of the order Pleosporales [9]. Also, the lower repeat content could be attributed to repeat induced point mutations (RIP) which prevent the development of highly repeated sequences [10, 11].

AUGUSTUS ab initio analysis resulted in the identification of 9875 protein-coding genes (Table 1, Data set 1). Among them, 6031 (61%) genes and 3438 (34.8%) genes exhibited homologies with known functions in the GO and KEGG databases, respectively (Table 1, Data file 7). The short-read sequence assembly of A. porri BMP0178 as reported by Dang et al. [4] could assign GO terms and KEGG IDs to only 5238 and 3176 genes, respectively. The categorization of functional annotation in the predicted genes showed that 4419 genes (73.3%) were assigned with molecular functions, followed by 3915 (64.9%) genes with cellular components and 3970 (65.8%) genes with biological processes (Table 1, Data file 8). Catalytic (25.3%) and binding (21.6%) activities dominated the molecular function category, whereas, under the category of cellular component, cells (17.3%) and cell components (14.4%) formed the major groups. Likewise, KEGG enrichment showed that the majority of enzymes belong to hydrolases (1148), followed by transferases (1064) and oxidoreductases (588) (Table 1, Data file 7). Previous studies have shown that hydrolases are a major factor of pathogenicity as they help the fungus to break down plant cell walls, acquire nutrients from plant tissues, and evade plant immune responses [11, 12].

The pathogen-host interactions (PHI)-database identified a total of 6776 (74.03%) pathogenicity-related genes in A. porri genome, which could be crucial for the virulence of the purple blotch pathogen (Table 1, Data file 9). Among them, 3605 genes were associated with reduced virulence, 554 with loss of pathogenicity, 537 genes were identified with effector functions (plant avirulence determinant), 479 genes related to increased virulence, 157 related to lethal factors, 17 genes associated with chemical resistance, and nine genes related to chemical sensitivity. A majority of them belong to secreted proteases, CAZymes and transporters that play an important role in the interaction of the fungal pathogen with its host [13]. The enrichment of CAZymes using three algorithms, viz., Diamond, HMMER, and Hotpep in the dbCAN2 Meta server resulted in the identification of 837 genes (Table 1, Data set 2) in different CAZymes families including 434 glycoside hydrolases (GHs), 157 enzymes with auxiliary activities (AAs), 145 glycosyltransferases (GTs), 49 carbohydrate esterases (CEs), 25 polysaccharide lyases (PLs), and 27 carbohydrate-binding modules (CBMs) (Table 1, Data file 10). The higher number of CAZymes suggests that many of them might be involved in the degradation of plant cell wall components such as cellulose and hemicellulose, while at the same time utilizing the plant polysaccharides during colonization of the host plant [14]. Interestingly, the auxiliary activity enzyme 9 (AA9) was significantly represented (30 genes) in the A. porri genome. AA9, formerly known as GH61 (copper-dependent lytic polysaccharide monooxygenases) has been reported as over-represented in multiple Alternaria spp [9]. This family of CAZymes exhibited dual role as a degrader of cell-wall polysaccharides and contributed to host necrosis through the generation of reactive oxygen species (ROS) [15].

Alignment of A. porri genome with the genomic sequences of A. alternata, A. brassicae and A. brassicicola showed distinctive inversions (flipped genomic segments), duplication (similar copy segments) and translocation (DNA fragments moving to different locations) events across the four genomes (Table 1, Data file 11). The presence of continuous coloured blocks showed that conserved sequences are shared among the compared genomes. A. porri was found evolutionarily closer to A. alternata followed by A. brassicicola. The gaps or mismatches within the aligned regions represented the insertion, deletions and substitutions among the closely related species. While, there was no significant gene insertion or deletion of large regions in A. porri genome, the location of homologous genes was different. Orthology analysis reported 8914 orthologous genes among the four Alternaria spp. (Table 1, Data file 12). A. porri displayed 207 unique genes, while it shared 604 genes with A. alternata, 3558 genes with A. brassicicola, and only nine genes with A brassicae.

Limitations

Using the latest genome sequencing platform, we have developed the nearly complete and contiguous genome assembly of A. porri in the present study. A greater variability has been represented in the PB isolates from different onion-growing regions of the world. As such, the assembled data from this study might not be fully accurate. If and when more funds are available, multiple isolates of PB could be sequenced using long-read sequencing technologies with greater coverage to increase the completeness and accuracy of the PB genome.