Transcriptome analysis of Phytophthora cactorum infecting strawberry identified RXLR effectors that induce cell death when transiently expressed in Nicotiana benthamiana

Phytophthora cactorum is a plant pathogenic oomycete that causes crown rot in strawberry leading to significant economic losses every year. To invade the host, P. cactorum secretes an arsenal of effectors that can manipulate host physiology and impair its defense system promoting infection. A transcriptome analysis was conducted on a susceptible wild strawberry genotype (Fragaria vesca) 48 hours post inoculation with P. cactorum to identify effectors expressed during the early infection stage. The analysis revealed 4,668 P. cactorum genes expressed during infection of F. vesca. A total of 539 secreted proteins encoded by transcripts were identified, including 120 carbohydrate-active enzymes, 40 RXLRs, 23 proteolytic enzymes, nine elicitins, seven cysteine rich proteins, seven necrosis inducing proteins and 216 hypothetical proteins with unknown function. Twenty of the 40 RXLR effector candidates were transiently expressed in Nicotiana benthamiana using agroinfiltration and five previously unreported RXLR effector genes (Pc741, Pc8318, Pc10890, Pc20813, and Pc22290) triggered cell death when transiently expressed. The identified cell death inducing RXLR effectors showed 31–66% identity to known RXLR effectors in different Phytophthora species having roles in pathogenicity including both activation and suppression of defense response in the host. Furthermore, homology analysis revealed that these cell death inducing RXLR effectors were highly conserved (82 - 100% identity) across 23 different strains of P. cactorum originating from apple or strawberry.


Introduction
Phytophthora cactorum is a devastating soil-borne oomycete pathogen that infects more than 200 plant species from 154 genera, including some of the most valuable horticultural plants such as apple, pear, and strawberry (Erwin and Ribeiro, 1996;Hantula et al., 2000;Sanchez et al., 2019;Chen et al., 2023).The pathogen can persist as resting oospores in soil for many years, even without a host plant, and during extreme environmental conditions.Phytophthora cactorum is challenging to control even with the use of chemicals due to oomyceticide (fungicide) resistance and the inefficiency of chemicals to different life stages of the pathogen (Marin et al., 2021;Ali et al., 2022).In the strawberry host, P. cactorum causes crown rot and leather rot, limiting the plant growth and quality of strawberry fruits, respectively, resulting in significant economic losses worldwide (Ellis and Grove, 1983;Stensvand et al., 1999).Crown rot symptoms include brown necrotic lesions in the rhizome (crown), which in severe cases result in wilting of the whole plant, whereas the leather rot affects the strawberry fruits, imparting an off-flavor taste and a pungent smell (Maas, 1998).
Plants possess a two layered immune system that responds to invading pathogens and impedes their growth.The first layer of immunity, known as pattern-triggered immunity (PTI), is activated when specific epitopes, called microbe-or pathogen associated molecular patterns (MAMPs/PAMPs), are recognized by the plant's pattern recognition receptors (PRRs) (Medzhitov and Janeway, 1997;Jones and Dangl, 2006).The second layer of immunity, referred to as effector-triggered immunity (ETI), involves host resistance proteins that interact either directly or indirectly with specific effectors secreted by the pathogen, activating several defense signaling pathways (Jones and Dangl, 2006;Dodds and Rathjen, 2010;Dou and Zhou, 2012;Giraldo and Valent, 2013).Both PTI and ETI activation result in transcriptional reprogramming of the defenserelated genes, which lead to the production of reactive oxygen species (ROS), secondary metabolites, hydrolytic enzymes, phytohormones and pathogenesis-related proteins (Tsuda and Katagiri, 2010;Naveed et al., 2020).The activation of defense responses in the host plant via PTI and ETI can also initiate a hypersensitive response or localized programmed cell death that can restrict growth of biotrophic and hemibiotrophic pathogens including Phytophthora species (Naveed et al., 2020).
Identifying and studying effector genes is crucial for understanding their roles in the interaction with host plants.Previous transcriptomic studies of different life stages of P. cactorum (mycelium, sporangia, zoospores, cysts, germinating cysts), and during infection of Nicotiana benthamiana and strawberry have identified several candidate effector genes (Chen et al., 2014(Chen et al., , 2018;;Nellist et al., 2021).Furthermore, in silico analyses of the sequenced genomes of P. cactorum have predicted hundreds of effector genes including RXLRs and CRNs (Armitage et al., 2018;Yang et al., 2018;Nellist et al., 2021;Gogoi et al., 2023b).To uncover the effectors produced by P. cactorum during the early and important phase of strawberry infection, the transcriptome was studied 48-hours post-inoculation of the rhizome (crown) of the susceptible Fragaria vesca genotype NCGR1218.Twenty candidate RXLR effector genes identified in the transcriptome study were transiently expressed in N. benthamiana leaves to examine potential cell death inducing responses.

Phytophthora cactorum
Diploid strawberry Fragaria vesca genotype NCGR1218 (susceptible to Phytophthora cactorum) was clonally propagated from runners and maintained in a greenhouse with a 16-hour photoperiod at 18°C.Nicotiana benthamiana plants were grown from seeds and were kept in a growth chamber at 21°C with a 16hour photoperiod.Phytophthora cactorum strain 10300 previously isolated from a crown rot infected strawberry plant (Fragaria × ananassa) (Armitage et al., 2018) was routinely cultured on vegetable juice (V8) agar plates (Ferguson and Jeffers, 1999) at room temperature (~21°C) in the dark.A zoospore suspension was prepared from one-week old culture plates as described by Eikemo et al. (2000).The zoospores released from sporangia were counted using a hemocytometer and the concentration was adjusted to 2 × 10 5 zoospores/ml for inoculation of strawberry plants.The plants were gently wounded in the rhizome (crown) with a sterile scalpel and inoculated with 2 ml of the zoospore suspension or water (mock/control).Four biological replicates, with each replicate consisting of four individual plants were used for the inoculation experiment as well as the control treatment with water.The rhizome samples were harvested 48 hours after inoculation, flash frozen in liquid nitrogen and stored at -80°C until RNA isolation.The 48 hour time point represents the early infection stage of P. cactorum based on a temporal expression study of defense related genes in the resistant and susceptible F. vesca genotypes as previously described (Toljamo et al., 2016;Chen et al., 2016a;Gogoi et al., 2023a).No visible symptoms were observed at the time of harvest.Some additional plants, both zoospore inoculated and water control, were kept up to four weeks post inoculation to study the disease progression.Wilting and necrotic lesions were observed in the P. cactorum inoculated plants after two weeks while no symptoms were observed in the control plants (Gogoi et al., 2023a).

RNA extraction, transcriptome sequencing and analysis
Total RNA was extracted from the inoculated strawberry rhizomes using the Spectrum ™ Plant Total RNA Kit (Sigma-Aldrich, USA) according to the manufacturer's instructions.Oncolumn DNase digestion was performed for 30 minutes on the isolated RNA to remove traces of genomic DNA contamination (Sigma-Aldrich, USA), as described by Gogoi et al. (2023a).Sequencing libraries were prepared using the TruSeq ™ stranded total RNA library prep Kit (Illumina), and sequencing was performed using four lanes on an Illumina HiSeq 3/4000 System (2 × 150 bp) at the Norwegian Sequencing Centre, Oslo, Norway.An average of 54.8 million high-quality trimmed reads (SD 5.3 million) were obtained from the four replicates in this study.Transcriptome assembly and expression analysis were carried out as previously described (Gogoi et al., 2023a).Briefly, the transcripts were de novo assembled instead of mapping to the reference genome (P.cactorum, assembly ASM1686465v1), to recover both F. vesca and P. cactorum transcripts.Transcripts were quantified using the pseudo-alignment method Kallisto (Bray et al., 2016), and the normalization of the transcript counts was performed in the CLC genomic workbench v11.01 (Qiagen, Aarhus, Denmark) using the transcripts per million (TPM) method (Wagner et al., 2012).The longest transcript isoforms obtained from the de novo assembly were assigned a P. cactorum gene ID using BLASTN against the sequenced genome of the P. cactorum strain 10300 (GCA_003287315.1_Pcac_10300_v1_cds_from_genomic),with an expectation value e < 10 -10 as a threshold.Transcripts having ≥ 99% identity to the P. cactorum 10300 genome were selected for the downstream analysis.The transcripts were annotated using Blast2GO v5.0 (Götz et al., 2008).The full-length sequences of the proteins encoded by the genes were retrieved using NCBI protein sequences (GCA_003287315.1_Pcac_10300_v1_protein),and these were used for further analysis.The full-length protein sequences encoded by transcripts were analyzed using the STRING database V11.5 (Szklarczyk et al., 2021), with gene ontology (GO) classification for functional annotation and the Kyoto encyclopedia of genes and genomes (KEGG) to predict biological pathways.

Secretome prediction and in-silico functional analysis of effector proteins
Effector proteins must be secreted in order to reach their cellular targets at the intercellular interface between the plant and pathogen or inside the host cell (Torto et al., 2003), and therefore the proteins encoded by transcripts were analyzed for signal peptides using SignalP5 (Armenteros et al., 2019b).The proteins predicted to have a signal peptide were further analyzed for transmembrane domains using Phobius (Käll et al., 2004), mitochondrial transit peptides using TargetP2.0(Armenteros et al., 2019a), and endoplasmic reticulum retention signals (KDEL/HDEL motif) (Stornaiuolo et al., 2003) using PROSITE-Scan (de Castro et al., 2006).All proteins with a signal peptide in their N-termini with no more than one transmembrane domain and no mitochondrial transit peptide or endoplasmic reticulum retention motif were considered to be secreted proteins of P. cactorum.
Additionally, all proteins characterized as secreted were searched against the pathogen-host interaction database (PHIbase) 4.14 using BLASTP (e-value <1e-5; ≥50% query coverage; bit score ≥50; identity ≥30%).The PHI-base comprises experimentally validated genes associated with pathogen virulence, which can aid in uncovering the role of secreted proteins in pathogenicity or disease development (Urban et al., 2022).The PHI-phenotypes 'increased virulence', 'lethal', 'loss of pathogenicity', and 'reduced virulence' are produced as a result of a mutation or altered expression of a specific gene in the pathogen, while the PHI-phenotype 'plant avirulence determinant' represents the effector gene required for the recognition of a pathogen in resistant hosts (Urban et al., 2022).

Protein homology and substitution rates analyses of the RXLR effectors
The predicted protein structures of 39 of the 40 RXLR effectors detected in this study (PC110_g6139 was unavailable) were downloaded as PDB files from the AlphaFold Protein Structure Database (Varadi et al., 2022).The predicted structure of the RXLR effectors were aligned using template modeling (TM)-align (Zhang and Skolnick, 2005) and a TM-score matrix was constructed for topological structural similarity assessment of the protein structures.Protein pairs with a TM-score > 0.5 are considered to share the same global fold, while those with a TM-score < 0.5 do not (Xu and Zhang, 2010).Additionally, homologs of the five RXLR effectors that induced cell death in N. benthamiana were identified from 23 different strains (Supplementary Material S5) of P. cactorum using OrthoFinder v2.4.0 (Emms and Kelly, 2019).The protein sequences of the identified homologs for each RXLR were aligned using Clustal Omega (Madeira et al., 2022) and phylograms were constructed using the neighbor-joining method and visualized using PRESTO (http://www.atgc-montpellier.fr/presto).
The rate of synonymous (a) and nonsynonymous (b) substitutions in the homologs of cell death inducing RXLR effectors were calculated for the four RXLRs that had variation within available sequences using FUBAR (Fast, Unconstrained Bayesian AppRoximation) from the HyPhy software v2.5.29 (MP) (Murrell et al., 2013;Kosakovsky Pond et al., 2020).Amino acids sites with b > a and posterior probability > 0.9 were considered under positive (diversifying) selection, whereas b < a (with posterior probability > 0.9) were considered under negative (purifying) selection.

Cloning of P. cactorum RXLR effector genes and their transient expression in Nicotiana benthamiana
To understand the role of RXLR effectors in inducing plant immunity, RXLR candidate genes were cloned in a plant expression vector and were transiently expressed in N. benthamiana leaves.For cloning, cDNA was used as a template for amplification of the target RXLR genes.Briefly, one microgram of total RNA isolated from the infected rhizome samples of the susceptible strawberry genotype NCGR1218 was used for cDNA synthesis using the iScript ™ cDNA Synthesis Kit (Bio-Rad, USA).The full-length genes without the signal peptide region were amplified in a 50 µl PCR reaction mix with Phusion ® High-Fidelity DNA Polymerase (5 U/µl) (New England Biolabs, USA) using gene-specific primers (Supplementary Table S1).PCR was performed in the T100 Thermal Cycler (BioRad, USA) with an initial denaturation at 95°C for five minutes followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 60°C for 30 seconds, and polymerization at 72°C for 1 minute with a final extension of 12 minutes at 72°C.The amplified products were gel purified using QIAquick Gel Extraction Kit (Qiagen, USA), cloned into the Gateway entry vector pDONR ™ /Zeo (Thermo Fisher Scientific, USA) by BP recombination, and subsequently moved into the plant expression vector pK7WG2 (Karimi et al., 2002) through an LR recombination reaction (Thermo Fisher Scientific, USA).The constructs were transformed into chemically competent Escherichia coli DH5a cells (Library Efficiency ™ DH5a competent cells, Invitrogen, USA) using the heat-shock transformation method (Froger and Hall, 2007).Transformants were selected using Luria-Bertani agar (LA) plates containing 50 µg/ml of zeocin and 100 µg/ml of spectinomycin for BP and LR transformants, respectively, followed by colony PCR in a 20 µl reaction mix with AmpliTaq DNA Polymerase (5 U/µl) (Applied Biosciences, USA) and PCR conditions as described above with primer specific annealing temperature (Supplementary Table S1).The recombinant plasmids were isolated using the Qiagen ™ Plasmid Mini Kit (Qiagen, Germany), and the target insert sequence was verified using Sanger sequencing at Eurofins Genomics (Germany).The freeze-thaw transformation method was used to introduce the recombinant plasmid constructs into chemically competent cells of Agrobacterium tumefaciens strain AGL1 (Weigel and Glazebrook, 2006).The transformed bacterial colonies were selected on LA plates containing spectinomycin (100 µg/ml), carbenicillin (50 µg/ml) and rifampicin (15 µg/ml) and confirmed by colony PCR using gene-specific primers (Supplementary Table S1).A suspension of A. tumefaciens with an OD 600 of 0.8 prepared as described by Du et al. (2014) was infiltrated into the top 3 rd to 5 th leaf of 3-4 weeks old N. benthamiana plants.The two P. cactorum genes, Pc16451 (PC110_g16451) and Pc22254 (PC110_g22254), previously reported to induce cell death in N. benthamiana, as RXLR6 and RXLR27, respectively (Chen et al., 2014) were included as positive controls.In addition INF1, which is a known elicitor of cell death from P. infestans (Kamoun et al., 1997(Kamoun et al., , 1998)), was used as a positive control, while the empty vector pK7WG2 was used as a negative control.The expression of the recombinant RXLR genes in N. benthamiana were confirmed three days after the agroinfiltration by reverse transcription PCR (RT-PCR).Briefly, the cDNA was synthesized from the total RNA isolated from agroinfiltrated leaves and PCR was performed using 2 µl cDNA as a template, and primers targeting short fragments of the RXLR genes (Supplementary Table S1).RNA was used as a template for a negative control (-RT control).

Histochemical assays
Agroinfiltrated N. benthamiana leaves showing visible cell death were stained using trypan blue to confirm the cell death response, as trypan blue is readily taken up by dead cells but not by living cells (Keogh et al., 1980).Briefly, leaves were boiled for 5-10 minutes in the trypan blue staining solution [10 ml DL-lactic acid (90%), 10 ml phenol (equilibrated with 10 mM Tris-HCL, pH 8.0), 1 mM EDTA, 10 ml glycerol (98%), 10 ml distilled water, 20 mg trypan blue] diluted with 96% ethanol (1:1 v/v), until the green color of the leaf disappeared (Bach-Pages and Preston, 2018).The leaves were further incubated in the solution at room temperature for 30 minutes, and destained overnight in chloral hydrate solution (2.5 g/ ml) on a rotary shaker (40 rpm).Additionally, 3, 3´diaminobenzidine (DAB) and aniline blue staining (Bach-Pages and Preston, 2018) were performed to assess oxidative burst/H 2 O 2 accumulation and callose deposition, respectively, 3 days after agroinfiltration.The insoluble brown precipitate formed after DAB staining is a result of a peroxidase dependent polymerization of DAB with H 2 O 2, indicating H 2 O 2 accumulation, which is connected to cell death response (Alvarez et al., 1998;Steffens et al., 2013).Callose depositions are formed by plants upon recognition of non-self components and these fluoresce under UV light after binding with aniline blue (Jin and Mackey, 2017).

Transcriptome profile and functional classification
From our previous defense transcriptome study of Fragaria vesca genotypes inoculated with Phytophthora cactorum (Gogoi et al., 2023a), a total of 412,970 de novo assembled transcript isoforms were obtained.Blasting the longest isoforms of de novo assembled transcripts (308,070) to the P. cactorum strain 10300 genome (GCA_003287315.1)resulted in 4,808 transcripts, which represent ca 1.6% of the total transcripts, of which 4,665 had more than 99% identity to P. cactorum genes (Supplementary Material S1).Three candidate RXLR effector genes, Pc741 (PC110_g741), Pc12148 (PC110_g12148), and Pc19826 (PC110_g19826), with less than 99% nucleotide identity to the genome of P. cactorum strain 10300, were manually curated based on the RXLR motif and their expression during infection.These genes were included in the final list of P. cactorum genes, amounting to a total of 4,668 genes (Supplementary Material S1).
The 4,668 proteins encoded by the P. cactorum transcripts were classified using gene ontology (GO) categories, and 3,365 of them were represented in 848 ontologies (517 biological process, 188 molecular function, and 143 cellular component) while KEGG analysis identified 2,009 functional transcripts assigned to 78 pathways (Figure 1; Supplementary Material S2).

Secretome analysis of P. cactorum genes expressed during strawberry infection
Of the 4,668 proteins encoded by the detected transcripts, 539 (~12%) were predicted to have signal peptides for secretion and are thus referred to as secreted proteins (Supplementary Material S1).The 539 secreted proteins belonged to several effector families including 120 carbohydrate active enzymes (CAZymes), 40 RXLRs, 23 proteolytic enzymes, nine elicitins, seven cysteine rich proteins, seven necrosis inducing proteins and three crinklers (CRN) (Table 1; Supplementary Table S1).The 120 identified CAZymes belonged to 52 sub-families potentially targeting different substrates of the plant cell (Table 2).The majority of the sub-families belonged to glycoside Gene ontology (GO) classification and Kyoto encyclopedia of genes and genome (KEGG) analysis of proteins encoded by transcripts of Phytophthora cactorum (accession GCA_003287315.1) detected in the rhizome (crown) of the susceptible Fragaria vesca NCGR1218, 48 hours after inoculation.The GO terms and KEGG pathways were categorized using the STRING V11.5 database (Szklarczyk et al., 2021).The top 10 most represented terms in each of the GO categories cellular component, molecular function, and biological process, as well as KEGG pathways are shown.
In silico analysis of the secreted proteins of P. cactorum using the pathogen host interaction database (PHI-base) showed 129 proteins having 30-99% sequence identity with proteins known to influence host-pathogen interaction in oomycete, fungal, bacterial, or nematode species.These 129 proteins were categorized in different PHI-phenotypes, of which 70 were assigned to the 'reduced virulence' phenotype, 45 to 'plant avirulence determinant', 29 to 'increased virulence', and nine to 'loss of pathogenicity' (Figure 2) (Supplementary Material S3).Twenty-four of the proteins were assigned to more than one PHI-phenotype.

RXLR effectors that trigger cell death in Nicotiana benthamiana
Twenty of the 40 candidate RXLR effector genes with high expression values during infection of strawberry or with high similarity to known Phytophthora effectors were selected for transient expression in N. benthamiana (Table 3).The expression of the agroinfiltrated RXLR effector genes in N. benthamiana were confirmed three days post agroinfiltration using RT-PCR (Supplementary Figure S1), while the phenotypic responses of the expressed RXLR genes were recorded five days post agroinfiltration.Five previously unreported RXLR effector genes, Pc741 (PC110_g741), Pc8318 (PC110_g8318), Pc10890 (PC110_g10890), Pc20813 (PC110_g20813), and Pc22290 (PC110_g22290) induced cell death when expressed in N. benthamiana leaves, which was confirmed with trypan blue staining (Figure 3).The cell death induced by the five RXLR effectors were further confirmed by 3, 3'-diaminobenzidine (DAB) staining, which showed a strong brown color precipitation in the RXLR agroinfiltrated regions, indicating an accumulation of H 2 O 2 compared to the regions agroinfiltrated with the empty vector control (Supplementary Figure S2).Furthermore, aniline blue staining showed fluorescence in the RXLR agroinfiltrated regions with induced cell death, demonstrating accumulation of callose, in contrast to the regions agroinfiltrated with empty vector control (Supplementary Figure S3).These results indicate that the five RXLR effectors could activate the immune system of N. benthamiana.
The protein template modeling (TM)-score matrix for RXLR effectors revealed that the TM-scores for 99.6% of the pairwise comparisons of effectors were 0.5 or below (Supplementary Material S4), suggesting that they have different structures.For the five RXLRs that induced cell death in N. benthamiana (Pc741, Pc8318, Pc10890, Pc20813, and Pc22290), the TM scores were less than 0.4 in pairwise comparison with each other.Three RXLR effector pairs gave TM scores slightly above 0.5, which could indicate some structural similarity.However, the scores were all below 0.6, so not very confident.Phylogenetic analysis of these five RXLRs with homologs from 23 different P. cactorum strains revealed that most of the homologs from strawberry strains were highly similar.Some of the homologs of Pc741, Pc8318, and Pc22290 from the apple strains were phylogenetically distinct from particularly the crown rot strains (Supplementary Figure S4).Homologs of Pc10890 were detected only in 16 of the P. cactorum genomes available, and these were all identical (Supplementary Material S5).
In the cell death inducing RXLR effectors, from zero to four sites under positive selection were detected and from zero to six sites under negative selection (Supplementary Material S6).The RXLR effector Pc20813, which was among the P. cactorum genes with highest expression (13 to 17 fold higher than Pc741, Pc8318 and 22290), had exclusively positive selection sites compared to its homologs.

Discussion
In this study, the transcriptome of the strain P. cactorum 10300 was investigated during infection of the susceptible Fragaria vesca genotype NCGR1218 at 48 hours post inoculation, which is an early and presumably important phase of the infection (Toljamo et al., 2016;Gogoi et al., 2023a).No visible necrotic lesions were observed at this time point, which is consistent with the previous report on the Hawaii4 genotype, where only hyphal growth was visible on the root surfaces after inoculation with P. cactorum (Toljamo et al., 2016).Our study  focused on the transcriptome profile of P. cactorum during infection of its natural host, while previous transcriptomics studies were either based on different life stages of P. cactorum, infection of model hosts like Nicotiana benthamiana, N. tabacum, and Solanum lycopersicum, or strawberry in vitro (Chen et al., 2014(Chen et al., , 2018;;Nellist et al., 2021).In total 4668 P. cactorum transcripts encoding proteins were detected.GO and KEGG analysis indicated that most of the proteins encoded by the transcripts had binding and catalytic activity and were involved in the metabolism of the pathogen (Figure 1), but 539 (˜12%) were predicted to encode secreted effector proteins belonging to different apoplastic and cytoplasmic effector groups (Table 1; Supplementary Material S1).
Based on analysis of the secreted proteins encoded by the detected P. cactorum transcripts in the pathogen-host interaction (PHI)-database (Urban et al., 2022), 129 of 539 secreted proteins were 30-99% identical to proteins previously demonstrated to have influence on virulence of different pathogens, including Phytophthora spp.(Figure 2; Supplementary Material S3).
As much as 40% of the predicted secreted proteins encoded by P. cactorum transcripts detected during strawberry infection were hypothetical proteins with unknown function.Of the predicted secreted proteins with a putative function, most belonged to typical apoplastic effector groups of which the largest was carbohydrate active enzymes (CAZymes) that are known to participate in host cell wall degradation and metabolism to facilitate infection (Ospina- Giraldo et al., 2010).Of the 120 proteins belonging to the CAZymes, the majority were glycoside hydrolases (GHs).The GHs secreted by plant associated fungi and oomycetes represent the largest class of CAZymes (Bradley et al., 2022).Numerous CAZymes belonging to different families including AA17, CBM25, GH3, GH5, GH6, GH7, GH10, GH12, GH16, GH17, GH28 detected in this study have previously been described as virulence factors in different pathogens (Hardham and Blackman, 2018;Rafiei et al., 2021;Sabbadin et al., 2021;Bradley et al., 2022).For instance, the xylanases PpXYn1 and PpXyn2, belonging to the GH10 family, are virulence factors that degrade xylan and are upregulated during Phytophthora parasitica infection (Lai and Liou, 2018).However, some of the CAZymes that act as virulence factors may also elicit defense responses in the host.For example, the cellobiohydrolase PsGH7 and the xyloglucanase PsXEG1 (GH12 family) from Phytophthora sojae are virulence factors that promote infection, while they also elicit PL4_e12 Pectin Pc14972, Pc19106, Pc19107, Pc19109 1 -indicate no available information about enzyme substrate in the dbCAN3 metaserver.
2 Full length sequences of the proteins encoded by detected transcripts were retrieved using NCBI protein sequences (GCA_003287315.1_Pcac_10300_v1_protein),and these were used for CAZyme analysis in the dbCAN3 metaserver.hypersensitive response in the soybean host, due to recognition as pathogen-associated molecular patterns (PAMPs) (Ma et al., 2015;Tan et al., 2020).
In addition to CAZymes, twenty-three proteolytic enzymes including proteases and peptidases were identified in our study.Proteolytic enzymes are important virulence factors in plantpathogen interactions and may disrupt the host defense for successful invasion (Jashni et al., 2015).For instance, the cysteine protease genes PpCys44 and PpCys45 of P. parasitica have been shown to act as virulence factors during infection in N. benthamiana (Zhang et al., 2020).Three enzyme inhibitors, Kazal-type serine protease inhibitor, protease inhibitor epic4 and elastase-like inhibitors, were also identified.This group of proteins, particularly the protease inhibitors are known to bind and inhibit the function of host apoplastic proteases during plant colonization (Jashni et al., 2015).For instance, an extracellular kazal-like serine protease inhibitor EP1 from P. infestans has been shown to inhibit and interact with the pathogenesis-related P69B subtilisin-like serine protease of tomato (Tian et al., 2004).Similarly, protein inhibitors like EPIC genes were shown to be upregulated during plant infection and their products are known to inhibit papain-like cysteine proteases (Tian et al., 2007;Kaschani et al., 2010).
Furthermore, nine elicitins, seven cysteine rich proteins, seven necrosis inducing proteins (NPPs), four transglutaminase elicitors, and one phytotoxin protein were identified.Cysteine rich proteins, elicitins, and NPPs in general are considered PAMPs that may Full length sequences of the proteins encoded by detected transcripts were retrieved using NCBI protein sequences (GCA_003287315.1_Pcac_10300_v1_protein),and these were used to identify RXLR effectors using the effectR program.
2 Pc is the abbreviated form of PC110_g in the RXLR ID and was obtained from Phytophthora cactorum accession GCA_003287315.1.Secreted Phytophthora cactorum proteins with similarity to proteins that influence pathogen host interaction (PHI), based on the PHI-base (Urban et al., 2022).The proteins are deduced from transcripts detected in the rhizome (crown) of a susceptible Fragaria vesca genotype (NCGR1218), 48 hours after inoculation with P. cactorum.The PHI-phenotypes are resulting from a mutation or altered expression of the specific gene in the pathogen while 'plant avirulence determinant' represents an effector required for recognition of a pathogen by the resistant host.Pc is the abbreviated form of PC110_g in the gene identifier and is obtained from P. cactorum accession GCA_003287315.1.The gene IDs in bold are assigned to more than one PHI-phenotype.
Of cytoplasmic effector groups, only three CRNs were identified in our study, of which PC110_g7969 showed 81% identity (86% similarity) to CRN1 from P. infestans that trigger necrotic responses in N. benthamiana and the host plant tomato (Torto et al., 2003).CRNs are known to result in cell death and chlorosis when expressed in the plant, however some of the CRNs can also suppress cell death, thereby promoting virulence.For instance, P. parasitica CRN effector PpCRN7 enhances INF1 induced cell death in N. benthamiana, while PpCRN20 suppresses it.Despite their contrasting functions, both PpCRN7 and PpCRN20 increase plant susceptibility to P. parasitica (Maximo et al., 2019).
Forty RXLR candidates were detected in the transcripts, which constitutes only 20% of the predicted RXLRs in the genome of P. cactorum (Armitage et al., 2018).This was 10% less than detected by Nellist et al. (2021) at 48h after inoculation, but they used tissue culture plants which are much more fragile than the plants used in our study thus promoting a more rapid infection process.Twenty of the 40 detected RXLR candidates were transiently expressed in N. benthamiana to investigate their cell death inducing properties and five previously unreported RXLR effector genes, Pc741, Pc8318, Pc10890, Pc20813, and Pc22290, induced cell death.Interestingly, Pc22290, an RXLR candidate without the EER motif induced cell death.The EER motif supposedly helps in translocation of RXLR effectors into the host cell, but an exact RXLR-EER sequence is not a requirement for its translocation (Dou et al., 2008;Van West et al., 2010;Chen et al., 2020).RXLR effectors that lack the EER motif have also previously been shown to induce cell death, e.g., the avirulence protein ATR13 from the oomycete pathogen Hyaloperonospora parasitica induced cell death in Arabidopsis (Allen et al., 2004).
The cell death region induced by RXLR effectors in N. benthamiana leaves showed accumulation of reactive oxygen species (ROS) such as H 2 O 2 and callose deposits as confirmed by 3, 3'-diaminobenzidine (DAB) and aniline blue staining, respectively.Similar responses have been observed for RXLR effectors from other Phytophthora species (Yin et al., 2019;Situ et al., 2020).ROS like H 2 O 2 strengthen the plant cell walls through oxidative cross-linking and also act as signaling molecules that induce defense responses (Alvarez et al., 1998;Steffens et al., 2013;Redza-Dutordoir and Averill-Bates, 2016).Furthermore, callose deposition is one of the immune responses deployed by plants upon recognition of 'non-self' components and is induced during PTI (Du et al., 2015).It is not yet clear whether the cell death induced is the direct effect of the RXLR effectors or due to recognition of the effector by resistance proteins in N. benthamiana.However, deposition of callose in the RXLR expressing leaves suggests that cell death is a result of host defense response (Jin and Mackey, 2017).It should be noted that these RXLR effectors not necessarily induce similar effects in strawberry, but the response in N. benthamiana is an indication of their relevance.
Since the RXLR effectors are known to target a wide range of cellular processes, the protein sequences of the identified cell death inducing RXLR effectors from P. cactorum were blasted against the UniProt database to look for similarities with functionally studied proteins from other pathogens.The RXLR effector Pc741 showed 44% sequence identity (54% similarity) to the P. capsici RXLR207 that induced ROS-mediated cell death in transgenic Arabidopsis and reduced pathogen colonization.However, a mutation in RXLR207 resulted in decreased pathogen virulence, suggesting it is crucial for infection (Li et al., 2019).The Pc8318 RXLR effector showed 54% identity (68% similarity) to the RXLR effector CRE5 (PITG_06308) of P. infestans, which did not induce cell death but in contrast suppressed cell death induced by Avh238, Avh241, BAX, INF1, an elicitor gene of cell death from Phytophthora infestans was used as a positive control (+), and the empty vector pK7WG2 was used as a negative control (-) in each of the leaves.In addition, two previously reported cell death inducing RXLR genes (D) Pc16451 and (F) Pc22254 from P. cactorum (Chen et al., 2014) were included as positive controls.Images were taken five days after agroinfiltration of the gene constructs and after trypan blue staining of the same leaf.Pc is the abbreviated form of PC110_g from the gene identifier and is obtained from P. cactorum accession GCA_003287315.1.
INF1, thus indicating its potential roles in the virulence of Phytophthora spp (Yin et al., 2017).The Pc10890 RXLR effector showed 66% identity (77% similarity) with RXLR effector SFI6 from P. infestans, which like CRE5 did not induce cell death, but suppressed PAMP-triggered immunity in Arabidopsis and tomato, and increased virulence of P. infestans in N. benthamiana (Zheng et al., 2014).The Pc20813 RXLR effector showed 31% identity (47% similarity) with P. infestans RXLR effector SFI2 that induces cell death in N. benthamiana but attenuates flg22-induced immune response in tomato and Arabidopsis (Zheng et al., 2014).Furthermore, the Pc22290 RXLR effector showed 36% identity (53% similarity) with P. parasitica effector RXLR3 that did not induce cell death but significantly increased Phytophthora infection, and moderately suppressed INF1 induced cell death in N. benthamiana (Dalio et al., 2018).Although the five RXLR effectors induced cell death in N. benthamiana, it has yet to be confirmed if the induced cell death can inhibit P. cactorum infection.Thirteen of the 20 cloned RXLR effectors did not induce cell death in N. benthamiana indicating that these RXLRs do not directly activate the plant's immune system.Some RXLR effectors that do not trigger cell death, instead can suppress PTI or ETI induced cell death thereby promoting pathogen virulence (Wang et al., 2023).The cell death suppression effect of RXLRs can be studied by co-infiltration with known effectors or elicitins that induce cell death in N. benthamiana (Zhan et al., 2022).
Two of the RXLR effectors that induced cell death in N. benthamiana, Pc741 and Pc22254, were assigned to the PHIphenotype 'plant avirulence determinant', hence confirming the prediction from the PHI database and demonstrating that it can be a useful tool for selection of effector candidates for further characterization and functional studies.
Only a few P. cactorum effectors have been functionally characterized so far: four RXLRs, one NPP, two elicitins, and four cysteine rich proteins (Chen et al., 2023).The five additional cell death inducing RXLRs identified in this study can in principle play important roles in pathogenicity, despite being recognized by N. benthamiana when transiently expressed.However, the direct roles of the RXLR effectors identified in this study, in the virulence of P. cactorum remain to be elucidated.The protein structural similarity analysis of the identified RXLR effectors indicated that 99.6% of them were structurally unrelated with TM scores 0.5 or below.This included the RXLRs that induced cell death in N. benthamiana.Homology analysis revealed that all cell death inducing effectors were conserved across different strains of P. cactorum, but that some of the homologs of Pc741, Pc8318, and Pc22290 from the apple strains were phylogenetically distinct from particularly the crown rot strains (Supplementary Figure S4, Supplementary Material S5).The RXLR effector Pc20813 exhibited only sites under positive selection, indicating that these amino acid residues have an important role in the host-pathogen interaction (Sironi et al., 2015).The Pc20813 gene was among the genes with highest expression during the infection and displayed greater divergence among its homologs in P. cactorum strains than the other cell death inducing effectors.This contrasts with previous findings that suggest genes with higher expression levels tend to diverge less than those with lower expression (Paĺ et al., 2001;Yang et al., 2012;Williamson et al., 2014).In three other RXLRs (Pc741, Pc8318 and Pc22290), more negative selection sites were observed suggesting that these genes might be under strong negative selection favoring amino acid substitutions that can affect pathogen's fitness.
Previous studies have reported that pathogens drive effector evolution through mutation and sequence substitution in RXLR effector genes, thus promoting their virulence (Win and Kamoun, 2007;Jiang et al., 2008;Dong et al., 2009;Qutob et al., 2009;Cui et al., 2012).Further analysis of polymorphisms in the equivalent RXLR effector genes from other strains of P. cactorum could therefore provide more insights about their important domains and evolution of virulence mechanisms.In addition, identification of the host receptors or protein interacting partners of the identified cell death inducing RXLR effectors could help to understand signaling pathways involved in cell death and provide insights into resistance mechanism involved.These insights can further help in screening of resistant strawberry genotypes and development of future disease control strategies.

Conclusion
The present transcriptome study provides comprehensive insights of the Phytophthora cactorum genes expressed during the early stage of infection of the rhizome of the model plant, Fragaria vesca.A total of 4668 P. cactorum transcripts were identified, and 539 of these were predicted to encode secreted proteins belonging to different effector families including CAZymes, elicitins, cysteine rich proteins, necrosis inducing proteins, proteolytic enzymes and RXLRs.Twenty of the 40 RXLR effectors identified were transiently expressed in Nicotiana benthamiana, and five previously unreported RXLR effectors triggered cell death response.The functional roles of these RXLR effectors in strawberry infection is not yet known and need further investigation.Further research on the subcellular localization of these RXLR proteins in the plant cell, and interactions of host proteins and the RXLR effectors of P. cactorum can help to develop new strategies for breeding resistance in strawberry.

Funding
The author(s) declare that financial support was received for the research, authorship, and/or publication of this article.BG was funded by the European Union's Horizon 2020 (H2020) Marie Skłodowska-Curie Actions Innovative Training Networks PROTECTA H2020-MSCA-ITN-2017 call, under the grant agreement number 766048.The work was also supported by NIBIO (basic funding) and the Research Council of Norway, grant number 326212.

3
RXLR effectors selected for cloning and transient expression in Nicotiana benthamiana.induced upon transient expression of RXLR effectors in N. benthamiana at five days post agroinfiltration; '-' indicates the RXLR effectors that were not tested.

FIGURE 3
FIGURE 3 Cell death induced by Phytophthora cactorum RXLR effector genes (A) Pc741, (B) Pc8318, (C) Pc10890, (D) Pc16451, (E) Pc20813, (F) Pc22254, and (G) Pc22290 after transient expression in Nicotiana benthamiana leaves.The lower panel confirms cell death response after trypan blue staining.INF1, an elicitor gene of cell death from Phytophthora infestans was used as a positive control (+), and the empty vector pK7WG2 was used as a negative control (-) in each of the leaves.In addition, two previously reported cell death inducing RXLR genes (D) Pc16451 and (F) Pc22254 from P. cactorum(Chen et al., 2014) were included as positive controls.Images were taken five days after agroinfiltration of the gene constructs and after trypan blue staining of the same leaf.Pc is the abbreviated form of PC110_g from the gene identifier and is obtained from P. cactorum accession GCA_003287315.1.

TABLE 1
Number and encoding properties of transcripts fromPhytophthora cactorum detected 48 hours after inoculation into the crown (rhizome) of the susceptible Fragaria vesca NCGR1218.
1Secreted proteins are proteins encoded by transcripts with ≥99% identity to P. cactorum genes (accession GCA_003287315.1) that were predicted to be secreted (see materials and methods), and in addition include three RXLR effectors Pc741, Pc12148, and Pc19826 encoded by transcripts with 93%, 98%, and 94% identity to P. cactorum 10300 genes, respectively.2 Proteins encoded by transcripts were categorized based on the functional annotations retrieved from the dbCAN3 metaserver, effectR package, and NCBI Nr database.

TABLE 2
Carbohydrate active enzymes (CAZymes) encoded by the transcripts detected 48 hours after inoculation with Phytophthora cactorum in the rhizome (crown) of the susceptible Fragaria vesca NCGR1218.

TABLE 3
RXLR effector candidates encoded by transcripts detected 48 hours after inoculation with Phytophthora cactorum in the rhizome (crown) of the susceptible Fragaria vesca NCGR1218.