Objective

Passalora sequoiae (Ellis & Everh.) Y.L. Guo & W.H. Hsieh (syn. Cercosporidium sequoiae (Ellis and Everh.) Baker and Partridge) is a fungus that causes needle blight on genera in the Cupressaceae, mainly Leyland cypress (x Cupressocyparis leylandii) [1, 2]. Disease symptoms of brown to gray needles appear during the spring and progressively appear throughout the tree canopy to result in unmarketable trees (Fig. 1). Annual fungicide application and crop loss inflict significant costs on the ornamental tree and Christmas tree industries [3,4,5].

Fig. 1
figure 1

Leyland cypress tree showing Passalora twig blight symptoms

Full size image

The objective of this work was to sequence the whole genome of P. sequoiae using PacBio and Illumina to assemble contigs. A lack of genetic information for this fungus prevents utilization of genetic tools to determine genetic diversity of isolates, potential differences in virulence, and ultimately the development of control practices. Currently, only three entries are listed for Passalora spp. in GenBank (NCBI), corresponding to the 18S rDNA gene of this fungus, a total of 5476 base pairs (bp).

A problem in sampling P. sequoiae populations is that numerous dematiaceous hyphomycetes with morphologically similar conidia and conidioma are found in many regions (Figs. 2 and 3). Proper identification of these organisms is further complicated by the numerous name revisions over the last two decades [1, 6,7,8,9,10,11,12]. A further constraining factor is that only a small number of dematiaceous hyphomycetes have been included in genetic phylogenies using DNA loci, mRNA and proteins [7, 10,11,12,13,14,15,16,17,18,19,20]. Mycosphaerellaceae was recently narrowed to 120 genera based on phylogenetic data [12].

Fig. 2
figure 2

Infected Leyland cypress leaf with sporulating conidioma of Passalora sequoiae

Full size image
Fig. 3
figure 3

Conidia of Passalora sequoiae

Full size image

Data description

A single spore isolate of P. sequoiae 9LC2 was recovered from a Christmas tree near Hattiesburg, MS, USA. DNA was extracted [21] and sheared to approximately 20 kb fragments. SMRTbell library was prepared, then sequenced on a PacBio Sequel sequencer at USDA-ARS, Stoneville, MS, USA. Bam files were processed using Finishing Module 20.0 of CLC_Bio Workbench v.12 (Qiagen LLC, Hilden, Germany). A total of 519,499 subreads with 6,612,712,889 nucleotides (nt) total, average length 14,247 nt, N50 21,720, were generated. Subreads were corrected and de novo assembled. The initial 19 contigs were manually split when necessary, rendering 44 contigs of 722,016 nt average and 44 x coverage. A total of 244,368,646 reads with an average length of 148 nt after trimming were obtained from Illumina sequencing. These reads were mapped to the PacBio assembled contigs resulting in 1011 x average coverage. A small percentage of gaps, 2–4 nt in length, approximately 2–3 gaps every 150,000 nt were observed using Illumina reads on the PacBio assembly, and they corresponded to microsatellites; thus, in all cases, the PacBio assembly was chosen (Table 1).

Table 1 Overview of data files/data sets
Full size table

Basic Local Alignment Search Tool (BLAST) [22] of a 9360 nt contig containing the 18S rDNA gene and internal transcribed spacers of P. sequoiae isolate 9LC2 showed a 99.65% identity with the 5476 nt NCBI entry Passalora sequoiae GU214667.1 [10]. The 5476 bp region of 9LC2 was used to retrieve 20 closely related sequences with 100% coverage. A Neighbor Joining [23] phylogenetic radial tree was constructed [24] using CLC Genomics Workbench 20.0 (Fig. 4), using NCBI accessions: GU214655.1; GU214656.1; GU214658.1; GU214661.1; GU214662.1; GU214664.1; GU214665.1; GU214666.1; GU214667.1; GU214668.1; GU214670.1; GU214671.1; GU214673.1; GU214678.1; GU214684.1; GU214686.1; GU214688.1; GU214697.1; GU214698.1; GU214699.1. Passalora sequoiae 9LC2 showed 99.7% identity to P. sequoiae CPC 11258, and 99.2 identity to P. brachycarpa CBS 115124. Though the taxonomy of Passalora is still being debated [12], P. sequoiae 9LC2 grouped with previously reported Passalora spp.

Fig. 4
figure 4

Phylogeny of Passalora sequoiae 9LC2 and closely related species based on Neighbor-Joining analysis of 5465 nt of 18S ribosomal RNA (rRNA) gene, Internal transcribed spacer (ITS) 1, 5.8S rRNA gene, ITS2 and 28S ribosomal RNA gene partial sequence. Bootstrap of 100 resampling are shown at the nodes; scale is nucleotide substitution rate

Full size image

Structural annotation of the genome assembly was determined using MAKER v.2.31.8 [25]. The MAKER pipeline included programs 1) RepeatMasker v.4.0.6 [26] to mask interspersed repeats and low complexity DNA sequences; 2) three gene predictors: GeneMark-ES [27]; SNAP [28], trained with Sordariomycetidae proteins from the Uniprot database; and Augustus [29]; and 3) tRNAscan [30] to identify tRNA genes in the genomic sequence. The total number of genes identified by Maker was 10,657. Of those, 10,576 genes were predicted to have proteins ≥ 50 amino acids. Maker also identified 81 tRNA and 3.42% of the genome corresponded to short repetitive sequences.

DbCAN2 [31] identified 331 predicted proteins that had signatures as carbohydrate active enzymes (CAZymes). Of those 52, 9, 186, 3, 79 and 9 corresponded to auxiliary activity enzymes, carbohydrate esterases, glycoside hydrolases, polysaccharide lyases, glycosyl transferases and carbohydrate binding modules, respectively. Thirty-four proteins had blast hits to the phi-database [32].

This whole-genome project has been deposited in DDBJ/ENA/GenBank under the accession number WSQC00000000 [33]. The version described in this paper is the first version, WSQC01000000.

Limitations

The genome sequence of a single isolate of P. sequoiae is being reported; thus, sequences of additional isolates would be needed to perform comparative genomics. Mapping of the Illumina sequences to PacBio contigs resulted in small gaps of low frequency; therefore, no serious limitation of data quality was evident. Reconstruction of whole chromosomes showing predicted genes and their annotation would provide characterization of the structural and functional levels.