Reference Genome Resource for the Citrus Pathogen Phytophthora citrophthora

Phytophthora citrophthora is an oomycete pathogen that infects citrus. Its occurrence in citrus-growing regions worldwide is considered a major contributor to crop losses. This study presents a high-quality genome resource for P. citrophthora, which was generated using PacBio HiFi long-read high-throughput sequencing technology. We successfully assembled a 48.5 Mb genome containing 16,409 protein-coding genes from high-quality reads. This marks the first complete genome assembly of P. citrophthora, providing a valuable resource to enhance the understanding of pathogenic behaviour and fungicide sensitivity of this destructive citrus pathogen.


Introduction
Phytophthora citrophthora is the causal agent of root rot, gummosis, and branch canker in citrus trees, and brown rot in citrus fruit [1].This soil-borne pathogen was first described in 1906 by Smith and Smith.Along with Phytophthora nicotianae, it currently represents the most destructive Phytophthora species causing disease in citrus [2].This oomycete pathogen is widespread, causing significant tree and crop losses in all tropical and subtropical citrus regions worldwide [2].P. citrophthora was the first Phytophthora species reported in South African citrus [3] and has since then been reported in citrus orchards in various provinces of the country, including the Western Cape, Eastern Cape, Limpopo, and Mpumalanga [4].Despite P. citrophthora being classified as a threat, limited genetic information is available for this pathogen and no complete genome sequence has been published.A more comprehensive understanding of the molecular mechanisms of P. citrophthora will increase knowledge of its pathogenicity and aid in the improvement of current disease management practices.This communication presents a complete genome sequence to aid in this matter.

Materials and Methods
The P. citrophthora isolate STE-U-9442 was isolated through soil baiting from a citrus nursery in the Eastern Cape Province of South Africa.It was grown in a 250 mL Erlenmeyer flask containing 100 mL potato dextrose broth (Difco TM ).The culture was grown in a shaking incubator (± 120 rpm) at 27°C for three to five days.After incubation, mycelia were harvested, washed with distilled water, and frozen at -80°C.The frozen mycelia were ground to a fine powder in liquid nitrogen using a mortar and pestle.High-quality DNA (approximately 5,000 ng) was extracted from mycelia using a CTAB/PVP pre-extraction followed by the Qiagen DNeasy Plant Mini Kit protocol (QIAGEN, Hilden, Germany).For the pre-extraction, 75 mg of ground tissue was added to a 2 mL Eppendorf tube containing sterilised glass beads.The samples were disrupted in a TissueLyser (QIAGEN, Hilden, Germany) twice for 30 sec at high Ivyspring International Publisher speed, after which 1 mL of CTAB/PVP extraction buffer (prewarmed to 60°C) was added to each sample (1.4 M NaCl, 2% CTAB (w/v), 0.1 M Tris (pH 8), 0.02 M EDTA (pH 8), 1% PVP (pH 8)).The PVP was added to the extraction buffer shortly before use.The samples were again disrupted twice for 30 sec at high speed.Then, 4 µL of proteinase K (10 mg/mL) was added to the solution and incubated at 60°C for 30 min, inverting tubes every 10 min.Thereafter, 3 µL of Rnase (100 mg/mL) was added to the solution and incubated at 60°C for 30 min, inverting tubes every 10 min.After centrifugation for 10 min at 13,000 rpm, the lysate was transferred to new 2 mL tubes and the samples were further treated according to the Qiagen DNeasy Plant Mini Kit protocol from step 2 onwards.The quality of the DNA was determined using a NanoDrop spectrophotometer (ThermoFisher Scientific, Waltham, Massachusetts, USA), Qubit (ThermoFisher Scientific, Waltham, Massachusetts, USA), and BioAnalyzer (Agilent Technologies, Santa Clara, California, USA).
The genomic DNA library was constructed with a PacBio HiFi Library kit and was subjected to circular consensus sequencing on a PacBio Sequel II instrument by Macrogen (Seoul, South Korea) to generate HiFi reads.Preprocessing of reads was performed using SMRT Link software (Pacific Biosciences) whereby adapter sequences were removed and consensus sequences were generated through multiple passes around a circularised single DNA molecule (SMRTbell template).The Genome Assembly application, powered by the Improved Phase Assembler HiFi genome assembler (SMRT Link v11.0), was used to generate a de novo genome assembly using HiFi reads.Firstly, Pancake was used to overlap reads and the overlapped reads were phased using Nighthawk.Chimeras and duplicates were eliminated from the overlapped reads and a string graph was constructed, which resulted in the generation of primary contigs.Racon [5] was used to polish contigs with phased reads.Default parameters were used for all Genome Assembly application processes.Following assembly, the depth of coverage was determined by mapping the HiFi reads to the assembled contigs.During this step, contigs shorter than 1 kb were excluded.Genome completeness was evaluated with BUSCO (v5.3.0)[6] using lineage eukaryota_odb10.2019-11-20(number of genomes: 70, number of BUSCOs: 255).MAKER (v3.01.03) was used to predict gene location.Protein BLAST+ (v2.7.1+) was performed against UniProt Swiss-Prot (201806) to identify proteins using various databases, including GO [7], Interpro (v69.0)[8], Pfam (v31.0)[9], and EggNOG (v4.5.1) [10] to determine their function.
Using the HMMER and DIAMOND tools on the dbCAN server (https://bcb.unl.edu/dbCAN2/index.php)[11,12], the predicted proteins of P. citrophthora were searched against the dbCAN, dbCAN-sub, and CAZy databases.Proteins selected by at least two of the searches were defined as carbohydrate-active enzymes (CAZymes).
Proteins with a signal peptide, predicted using signal version 6 [13], but without transmembrane helices, predicted using TMHMM version 1.0.20 [14], were defined as candidate effectors.These candidate effectors were subjected to screening with EffectorP version 3 [15].
Protein sequences were subjected to a BLAST search (percent query coverage and identity cut-off of 35, E-value cut-off 1.0 x 10 -5 ) against the Pathogen Host Interactions base [16,17] to identify proteins associated with pathogenicity.

Results and Discussion
Library sequencing resulted in 2,432,934 HiFi reads with an average read length of 10,393 bp.The final assembly product was a ~48.5 Mb genome, with coverage of 521 x.The genome consisted of 155 contigs with an N 50 length of ~908.6 Kb (Table 1).Assessment of completeness showed that out of 255 BUSCO groups searched, the assembly of STE-U-9442 contained 233 complete and single-copy BUSCOs (91.37%), 6 complete and duplicated BUSCOs (2.35%), 7 fragmented BUSCOs (2.75%), and 9 missing BUSCOs (3.53%).
A total of 16,409 protein-coding genes were predicted in the P. citrophthora STE-U-9442 genome (Table 2).The largest number of genes (630 genes) were annotated to have a function relating to the post-translational modification of proteins (Figure 1).Pathogens rely on protein changes to manipulate the plant host response, increase their activity during infection, and ultimately promote their survival [18].The high number of genes involved in post-translational modifications alludes to the complex interaction between Phytophthora and the citrus plant host and why this species is difficult to manage when infection is already established.
The mitochondrial genome of P. citrophthora was assembled into a circular molecule of 37,510 bp with a 21.94% G+C content.It was predicted to encode 39 protein-coding genes, two ribosomal RNA genes, and 25 tRNA genes.
The full genome sequence of P. citrophthora will be essential for understanding the biology of this citrus pathogen, developing diagnostic tools for pathogen detection, identifying potential targets of disease control, and understanding the genetic evolution of this pathogen.

Figure 1 .
Figure 1.EggNOG functional protein classifications of Phytophthora citrophthora culture STE-U-9442.Different protein classes are indicated with A -Y.

Table 1 .
Genome assembly statistics of Phytophthora species with available whole genome sequences on FungiDB, including Phytophthora citrophthora culture STE-U-9442 from this study.