Unraveling NPR-like Family Genes in Fragaria spp. Facilitated to Identify Putative NPR1 and NPR3/4 Orthologues Participating in Strawberry-Colletotrichum fructicola Interaction

The salicylic acid receptor NPR1 (nonexpressor of pathogenesis-related genes) and its paralogues NPR3 and NPR4 are master regulators of plant immunity. Commercial strawberry (Fragaria × ananassa) is a highly valued crop vulnerable to various pathogens. Historic confusions regarding the identity of NPR-like genes have hindered research in strawberry resistance. In this study, the comprehensive identification and phylogenic analysis unraveled this family, harboring 6, 6, 5, and 23 members in F. vesca, F. viridis, F. iinumae, and F. × ananassa, respectively. These genes were clustered into three clades, with each diploid member matching three to five homoalleles in F. × ananassa. Despite the high conservation in terms of gene structure, protein module, and functional residues/motifs/domains, substantial divergence was observed, hinting strawberry NPR proteins probably function in ways somewhat different from Arabidopsis. RT-PCR and RNAseq analysis evidenced the transcriptional responses of FveNPR1 and FxaNPR1a to Colletotrichum fructicola. Extended expression analysis for strawberry NPR-likes helped to us understand how strawberry orchestrate the NPRs-centered defense system against C. fructicola. The cThe current work supports that FveNPR1 and FxaNPR1a, as well as FveNPR31 and FxaNPR31a-c, were putative functional orthologues of AtNPR1 and AtNPR3/4, respectively. These findings set a solid basis for the molecular dissection of biological functions of strawberry NPR-like genes for improving disease resistance.


Introduction
The sessile plants are constantly threatened with various biotic and abiotic stresses. They have evolved multiple defense mechanisms to ensure survival and development. The expression of pathogenesis-related (PR) genes in distal uninfected tissues is commonly a part of systemic acquired resistance (SAR) after infection with pathogens. The npr1 (NONEXPRESSOR OF PATHOGENESIS-RELATED GENES 1) mutant was first identified in Arabidopsis with reduced resistance [1]. NPR1 was characterized as a protein consisting of a N-terminal BTB domain, followed by ankyrin repeats and C-terminal nuclear localization signal [2]. Several cysteine residues within and beyond the BTB domain contribute to the reversible structural switching of AtNPR1 upon redox changes, and monomeric NPR1 is translocated to the nucleus to regulate gene expression [3]. NPR1 was confirmed to be a positive regulator of plant defense, while its paralogues NPR3/4 were largely the negative regulators of plant defense [4]. SUMOylation of NPR1 promotes its phosphorylation and interaction with TGAs to facilitate PRs expression [5]. Both NPR1 and NPR4 are required

Diversity and Phylogeny of NPR-like Members in Fragaria Species
Based on BLAST search with AtNPRs as queries and monitoring the presence of conserved domains via PFAM/CDD analysis, a total of 48 putative NPR-like genes were identified from five diploid and one octoploid Fragaria genomes. Specifically, there were 23 loci in octoploid F. × ananassa, six in diploid F. vesca, six in F. viridis, five in F. iinumae, three in F. nipponica, and five in F. nilgerrensis (Table 1). It was worth noting that the information of NPR-like genes in F. nipponica was incomplete due to the absence of high-quality genome annotation. The isoelectric points (pI) of strawberry NPR-like members were less than 7.0, ranging from 5.29 (FxaNPR32f) to 6.94 (FxaNPR32d). The molecular weights varied between 12.1 (FxaNPR32f) and 66.3 kDa (FnilNPR32, FvirNPR32a, and FxaNPR32d). PSORT prediction suggested the subcellular localization of 16 strawberry NPR-like proteins in chloroplast and 12 in endoplasmic reticulum. Strikingly, there were 25 Fragaria NPR-like proteins predicted in cytoplasm, 25 proteins in nucleus, and 13 members potentially as shuttles between nucleus and cytoplasm or plasma membrane, which hints at the versatile nature of NPRs, especially in signaling and expression regulation. To understand the evolutionary relationships among NPR-likes from diverse Fragaria species and Arabidopsis, an unrooted neighbor joining phylogenetic tree was constructed using the protein sequences of 48 strawberry NPR-like members and six AtNPRs ( Figure 1). This analysis supported the division of all proteins into three clades (Clade I, Clade II, and Clade III), with each clade headed by two Arabidopsis orthologs. Accordingly, strawberry NPR-like proteins were named following the nomenclature of Arabidopsis orthologs in certain clade. Each NPR-like protein in diploid Fragaria species corresponds to three to five homologous proteins in octoploid F. × ananassa. Clade I NPR-like proteins were not identified in F. nipponica. Moreover, the percentage of identical amino acids shared among Fragaria NPR-like proteins and AtNPRs was listed in Supplementary tables (Table S1 for all  54 members, S2-S4 for Clade I-III). Among all 54 members, the percentage of identity in amino acids ranged from 12.5% to 100%. Among clade I, II, and III, the common identity varied between 44.78 and 99.32%, 25.81 and 99.66%, 31.02 and 100%, respectively. The percentage lower than 40% largely occurred with the truncated members predicted in F. nipponica.
Interestingly, all Fragaria NPR-like members of clade I were predicted as shuttles between the nucleus and cytoplasm, and all members of clade III beyond FnipNPR5 might share locations in chloroplast and endoplasmic reticulum. Members of clade II were predicted in many subcellular locations, including the cytoplasm, endoplasmic reticulum, chloroplast, and nucleus. Three members of clade II might also be shuttles between the nucleus and cytoplasm (FveNPR31) or plasma (FxaNPR32e and FvirNPR32b). Furthermore, pI of clade I members varied between 6.17 and 6.74, while pI of all clade III members except for FnipNPR5 ranged from 6.20 to 6.30. Clade II could be grouped into three subclades. pI of NPR31-likes is between 6.15 and 6.69, and of NPR33-likes between 5.55 and 5.93. NPR32-like members from all diploid species beyond F. vesca have a pI ranging from 5.47 to 5.88. F. × ananassa and its dominant sub-genome progenitor F. vesca possess NPR32-like members with pI ranging from 6.07 to 6.94, with the exception of FxaNPR32f (5.29). Interestingly, all Fragaria NPR-like members of clade I were predicted as shuttles between the nucleus and cytoplasm, and all members of clade III beyond FnipNPR5 might share locations in chloroplast and endoplasmic reticulum. Members of clade II were predicted in many subcellular locations, including the cytoplasm, endoplasmic reticulum, chloroplast, and nucleus. Three members of clade II might also be shuttles between the nucleus and cytoplasm (FveNPR31) or plasma (FxaNPR32e and FvirNPR32b). Furthermore, pI of clade I members varied between 6.17 and 6.74, while pI of all clade III members except for FnipNPR5 ranged from 6.20 to 6.30. Clade II could be grouped into three subclades. pI of NPR31-likes is between 6.15 and 6.69, and of NPR33-likes between 5.55 and 5.93. NPR32-like members from all diploid species beyond F. vesca have a pI ranging from Figure 1. The phylogenetic relationships of NONEXPRESSOR OF PATHOGENESIS-RELATED GENES 1 (NPR1)-like proteins from Arabidopsis thaliana, five diploid strawberries, and the octoploid strawberry. The tree was clustered into three clades (I, II, and III) shaded with different colors. The length of branches indicates the relative phylogenetic relationship, and the bootstrap values near branches for confidence. Geometries with different shapes and colors are used to symbolize different NPR members, with a star for Arabidopsis, and cycles in red, yellow, green, purple, pink, and gray for Fragaria vesca, F. iinumae, F. viridis, F. nipponica, F. nilgerrensis, and F. × ananassa, respectively.

Structural Features of Fragaria NPR-like Members in DNA and Amino Acid Sequences
The phylogeny trees among Fragaria NPR-like members generated based on amino acid sequences ( Figure 1) and those based on nucleotide sequences in the coding region ( Figure 2) were largely identical with the exception of FnilNPR5. Generally, the exon-intron organization of strawberry NPR-like genes were conserved within a certain clade. All genes of clade I beyond FvirNPR1 and FxaNPR1e contain four exons separated by three short introns, with the second and third exons being the longest and shortest, respectively. Surprisingly, both FvirNPR1 and FxaNPR1e possess five exons with a short exon (second) followed by an extraordinarily long intron. NPR-like genes in clade III uniformly possess two exons, and the relative length of intron and exon in coding region is conserved. For clade II, most genes contain four exons of relative length similarly observed in clade I, although variations in intron sizes occur. The subgroup represented by NPR31-like genes possesses three introns of a similar length. The first intron in NPR32a-like genes is longer than that in NPR31-likes. Both the first and the third introns in NPR33-likes genes are longer than those in NPR31-likes. Five Fragaria NPR32b-like genes, including FveNPR32b, FvirNPR32b, and FxaNPR32e-f-g, similarly consist of three exons, which was not observed in Arabidopsis. In addition, the predicted FnipNPR5 and FnipNPR33 consist of one and two exons, respectively, which again triggered doubt about the accuracy of annotation via FGENESH in Softberry for F. nipponica genomic DNA.
like members with pI ranging from 6.07 to 6.94, with the exception of FxaNPR32f (5.29).

Structural Features of Fragaria NPR-like Members in DNA and Amino Acid Sequences
The phylogeny trees among Fragaria NPR-like members generated based on amino acid sequences ( Figure 1) and those based on nucleotide sequences in the coding region ( Figure 2) were largely identical with the exception of FnilNPR5. Generally, the exonintron organization of strawberry NPR-like genes were conserved within a certain clade. All genes of clade I beyond FvirNPR1 and FxaNPR1e contain four exons separated by three short introns, with the second and third exons being the longest and shortest, respectively. Surprisingly, both FvirNPR1 and FxaNPR1e possess five exons with a short exon (second) followed by an extraordinarily long intron. NPR-like genes in clade III uniformly possess two exons, and the relative length of intron and exon in coding region is conserved. For clade II, most genes contain four exons of relative length similarly observed in clade I, although variations in intron sizes occur. The subgroup represented by NPR31-like genes possesses three introns of a similar length. The first intron in NPR32a-like genes is longer than that in NPR31-likes. Both the first and the third introns in NPR33-likes genes are longer than those in NPR31-likes. Five Fragaria NPR32b-like genes, including FveNPR32b, FvirNPR32b, and FxaNPR32e-f-g, similarly consist of three exons, which was not observed in Arabidopsis. In addition, the predicted FnipNPR5 and FnipNPR33 consist of one and two exons, respectively, which again triggered doubt about the accuracy of annotation via FGENESH in Softberry for F. nipponica genomic DNA. An analysis of protein domain module revealed that most Fragaria NPR-like proteins contain a typical N-terminal Bric-a-brac, tramtrack, broad-complex (BTB)/poxvirus, zinc finger (POZ) domain (BTB/POZ), Ankyrin repeats (Ank), and a NPR1/NIM1 like defense protein C terminal (NPR1_like_C) (Figure 3). The Ankyrin repeats domain is absent in FxaNPR1e of clade I. With the exception of FnipNPR5, all proteins of Clade III have no NPR1_like_C domain. The aforementioned five NPR32b-like genes composed of three exons encode proteins only possessing NPR1_like_C domain. Of the three NPR-like members deduced in F. nipponica, FnipNPR31 is the unique member possessing all three functional domains, while FnipNPR33 and FnipNPR5 only contain Ankyrin repeats and NPR1_like_C domain without BTB/POZ domain. The potentially misleading grouping of FnipNPR5 in phylogenetic clade III might result from the presence of a complete ANK domain along with a partial NPR1_like_C domain in this candidate.  Comparing the motif compositions in NPR1 and NPR-like proteins between Fragaria spp. and Arabidopsis revealed the presence of Fragaria-specific motifs at the N-terminal before BTB/POZ domain in most members of clade I and clade II ( Figure S1). Multiple sequence alignment showed that Fragaria NPR-like proteins of clade I possess a variant of IkB-like phosphodegron motif (DSxxxxS) ( Figure S2) different from the motif 'DSxxxS' in Arabidopsis, which was related to the degradation of NPR1 [4]. Six cysteine residues (C82, C150, C155, C160, and C216) beyond C156 crucial for NPR1 oligomer formation [3] were conserved in five NPR1-like members identified in octoploid strawberry and each one in four diploid species. C156 in Arabidopsis NPR1 was substituted by a serine residue in all Fragaria NPR1-likes beyond that in F. nilgerrensis (alanine residue). Virtually, clade I NPRlike proteins in Fragaria contain nearly identical Ankyrin repeats, the NIMIN1/2-binding region, the nuclear localization sequence NLS, and the SUMO-interaction motif (SIM) [2,5]. The putative hinge region (LENRV) for SA binding is conserved in all clade I NPR-likes except for FiinNPR1 [39]. In addition, this region is also conserved in all members in Fragaria NPR31 and NPR33 subgroups but not in NPR32 subgroup.

Physical Distribution of NPR1-like Loci in the Genomes of F. vesca and F. × ananassa
The closest diploid relatives of octoploid strawberry have been suggested as F. vesca, F. iinumae, F. nipponica, and F. viridis for A, B, C, and D sub-genomes, respectively [29]. There is no doubt that F. vesca is the dominant diploid ancestor of F. × ananassa [23]. However, the sub-genome assignment problem in F. × ananassa has persisted [26,28,29]. The current comparison of the chromosomal location of NPR-like genes was limited to F. vesca and F. × ananassa.
NPR-like family genes are unevenly dispersed on three chromosomes of F. vesca ( Figure 4A). FveNPR1 of clade I is on Fvb1, and its homoalleles in octoploid strawberry are composed of five loci located in all four sub-genomes, with one more allele in B subgenome ( Figure 4B). Two thirds NPR-likes genes are located on Fvb6 of the F. vesca genome. The FveNPR5 encoding protein of clade III has four homoalleles in four sub-genomes of F. × ananassa, distributed on the homoeologous group of chromosomes Fvb6-1 to Fvb6-4. The chromosome Fvb6 of F. vesca also harbors all clade II NPR proteins encoding genes except for FveNPR31. The homoeologous chromosome group Fvb6-1~Fv6-4 in octoploid strawberry possesses 11 NPR3-like homoalleles similarly distributed, although the C subgenome only has two loci. FveNPR31, the potential negative regulator of strawberry SAdependent defense, locates on Fvb3 in F. vesca, whose homoallele is absent in the dominant A sub-genome [14]. FveNPR31 homoalleles occur on the homoeologous chromosome group Fvb3-1~Fv3-3 in B, C, and D sub-genomes.

Cis-Elements in the Promoter Regions of NPR-like Genes from F. vesca and F. × ananassa
Characterizing the occurrence of cis-elements in gene promoter region will facilitate our understanding of its regulatory expression. The 2500-bp upstream promoter sequences of NPR-like genes in both woodland and octoploid strawberries were selected and analyzed using PlantCare ( Figure 5). The elements identified were sorted into three types, including hormone response, stress response, and others. Information for the sequences and functional categories of identified cis-elements was listed in Supplementary Table S5. It is worth noting that in the current prediction from PlantCare ( Figure 5, Table S5), the TC-rich repeats (yellow block) may be responsible for the defense and stress responses [40], which was often overlapped with the MYB binding site. The WBOXATNPR1 (TTGAC) (red block) was a fungal elicitor element identical to the binding site for WRKY involved in plant defense response [41,42]. The GT1/TCA-element (red circle) is a salicylic acid responsive element [43,44]. The activation sequence-1-like (as-1) (yellow circle with black line) interpreted as auxin and salicylic acid responsive was indeed identical to the methyl-jasmonate (MeJA) responsive element, which was composed of two CGTCA or TGACG-motifs in close proximity, making up the binding site for TGA, involved in plant responses to auxins; salicylic acid (SA); MeJA; and further stressors, including hydrogen peroxide, dehydration, salt stress, etc. [45][46][47].

Cis-Elements in the Promoter Regions of NPR-like Genes from F. vesca and F. × ananassa
Characterizing the occurrence of cis-elements in gene promoter region will facilitate our understanding of its regulatory expression. The 2500-bp upstream promoter sequences of NPR-like genes in both woodland and octoploid strawberries were selected and analyzed using PlantCare ( Figure 5). The elements identified were sorted into three types, including hormone response, stress response, and others. Information for the sequences and functional categories of identified cis-elements was listed in Supplementary  Table S5. It is worth noting that in the current prediction from PlantCare ( Figure 5, Table  S5), the TC-rich repeats (yellow block) may be responsible for the defense and stress responses [40], which was often overlapped with the MYB binding site. The WBOXATNPR1 (TTGAC) (red block) was a fungal elicitor element identical to the binding site for WRKY involved in plant defense response [41,42]. The GT1/TCA-element (red circle) is a salicylic acid responsive element [43,44]. The activation sequence-1-like (as-1) (yellow circle with black line) interpreted as auxin and salicylic acid responsive was indeed identical to the methyl-jasmonate (MeJA) responsive element, which was composed of two CGTCA or TGACG-motifs in close proximity, making up the binding site for TGA, involved in plant responses to auxins; salicylic acid (SA); MeJA; and further stressors, including hydrogen peroxide, dehydration, salt stress, etc. [45][46][47]. For Fragaria NPR-like genes encoding protein of clade I, the most abundant and ubiquitous cis-elements identified were MYB-binding sites. More than 10 motifs of this type were observed in promoters of FveNPR1, FxaNPR1a, -1b, and -1d. Each member of clade I possesses one or two SA-responsive elements. Over three as-1 like elements were detected in the promoters of all NPR1-like genes except for FxaNPR1b, whose promoter only contains two copies of this element. One copy of TC-rich repeats involved in the defense and For Fragaria NPR-like genes encoding protein of clade I, the most abundant and ubiquitous cis-elements identified were MYB-binding sites. More than 10 motifs of this type were observed in promoters of FveNPR1, FxaNPR1a, -1b, and -1d. Each member of clade I possesses one or two SA-responsive elements. Over three as-1 like elements were detected in the promoters of all NPR1-like genes except for FxaNPR1b, whose promoter only contains two copies of this element. One copy of TC-rich repeats involved in the defense and stress response was observed in the promoters of FxaNPR1a, -1b, -1d, and -1e (masked by Myb-binding sites), but it was absent in the promoters of FveNPR1 and FxaNPR1c. Surprisingly, the WBOXATNPR1 (W-box, TTGAC) motif (red block) was completely absent in the promoters of all Fragaria NPR1-like genes. The hypothetical model might not be suitable for strawberry NPR1, whose transcription factors WRKY6 and WRKY18 associate with the NPR1 promoter through the W-box motif for the transcriptional regulation of NPR1 gene in Arabidopsis [48].
Three subgroups of clade II NPR-like genes varied in terms of the composition of cis-regulatory elements in their promoter regions. One copy of W-box was only present in the upstream of FveNPR31 and FxaNPR31a, although all four NPR31-like genes possess as-1 like and SA-responsive elements in their promoter regions. However, FveNPR32a and its three homoalleles FxaNPR32a, -32b, and -32d similarly had two copies of W-box in their promoters close to translation starting code. By contrast, FveNPR33 and its three homoalleles FxaNPR33a, -33b, and -33c uniformly had two copies of W-box far away from the translation starting code. Generally, promoter regions of FveNPR33 and its homoalleles contained more SA-responsive elements (3-6 copies) than the remaining members (0-2 copies) in clade II. In addition, FxaNPR32c was the unique member in the strawberry NPR-like family whose promoter did not have an SA-responsive element.
Among NPR-like genes of clade III, FveNPR5 was the unique one possessing a Wbox motif in its promoter region, while its four homoalleles in F. × ananassa did not have that motif. The promoters of all genes in this clade contained as-1 like and SAresponsive elements. Notably, there were as many as eight SA-responsive elements in the FxaNPR5d promoter.

Transcriptional Responses of Strawberry NPR-like Genes to C. fructicola
To reveal whether and how strawberry motivates its NPR-like family members in responses to biotic stress, strawberries, including F. vesca 'Hawaii4', F. × ananassa cultivars 'Camarosa', and 'Benihoppe', were inoculated with C. fructicola (strain CGMCC3.17371) of C. gloeosporioides complex, a dominant species causing strawberry anthracnose crown rot worldwide. Whole plants were spray-inoculated, and clear necrosis lesions appeared in 'Benihoppe' at 2 days post inoculation (dpi), while in 'Camarosa' and 'Hawaii4', small lesions only appeared at 4 dpi (data omitted). The detached leaves were inoculated on wounded sites for phenotyping ( Figure 6A). Significantly larger lesions were observed on leaves of 'Benihoppe' than those of 'Hawaii4' and 'Camarosa' since 4 dpi, which was consistent with the field observation that 'Benihoppe', is more susceptible than most other genotypes to C. fructicola. Concerning the potential central regulatory importance of clade I NPR-like genes in plant immunity, allele-specific primers were designed for expression analysis in octoploid strawberry. FxaNPR1a expression was always detected in 'Camarosa' leaves either infected or mock-treated, and a pathogen-induced transcription decrease in this allele was observed at 0 (about 15 min or less post inoculation), 6, 24, and 48 hpi. However, FxaNPR1a expression was only occasionally detected in 'Benihoppe' at a lower level, although at even lower levels, expression of the remaining three alleles FxaNPR1b-d was detected in 'Camarosa' while scarcely in 'Benihoppe'. The transcription of FxaNPR1e was never detected in the current work. Variations in the expression patterns of FxaNPR1a-d observed in two cultivars might be caused by different genetic backgrounds and less primer efficiency influenced by potential single nucleotide polymorphisms in 'Benihoppe'. Expression responses of strawberry NPR-like family genes were investigated in RNA samples from the leaves of whole plants spray-inoculated with C. fructicola or mock-treated. First, semi-quantitative RT-PCR was carried out ( Figure 6B). The results showed that the transcripts of whole NPR-like family genes could be detected in "Hawaii4' leaves, although FveNPR1 and FveNPR5 were expressed at much lower levels than FveNPR31-33. Indeed, five-time more cDNA templates combined with 40-cycle repeats in PCR settings were required for detecting the transcripts of FveNPR1 and -5. The same primerset matching FveNPR5 was used for analysis in octoploid strawberry. It was revealed that the total expression of several FxaNPR5s alleles was significantly higher than that of a single FveNPR5 in "Hawaii4'. As compared with mock treatment, there was a mild pathogen invasion-induced expression of FxaNPR5s at 6 hpi in 'Benihoppe'.
Concerning the potential central regulatory importance of clade I NPR-like genes in plant immunity, allele-specific primers were designed for expression analysis in octoploid strawberry. FxaNPR1a expression was always detected in 'Camarosa' leaves either infected or mock-treated, and a pathogen-induced transcription decrease in this allele was observed at 0 (about 15 min or less post inoculation), 6, 24, and 48 hpi. However, FxaNPR1a expression was only occasionally detected in 'Benihoppe' at a lower level, although at even lower levels, expression of the remaining three alleles FxaNPR1b-d was detected in 'Camarosa' while scarcely in 'Benihoppe'. The transcription of FxaNPR1e was never detected in the current work. Variations in the expression patterns of FxaNPR1a-d observed in two cultivars might be caused by different genetic backgrounds and less primer efficiency influenced by potential single nucleotide polymorphisms in 'Benihoppe'.
Indeed, the expression of FveNPR1, -5, and their homoalleles in octoploid strawberry was too weak to be investigated via quantitative RT-PCR (qPCR). Accordingly, qPCR analysis was only performed for FveNPR31-33 in "Hawaii4' and their homoalleles in 'Benihoppe' (Figure 7). It was revealed that FveNPR31 expression was quickly suppressed upon inoculation at 0 hpi (less than 15 min after inoculation) and significantly induced at 96 hpi in "Hawaii4'. Similarly, the total expression of FxaNPR31abc was up-regulated at 48 hpi (significantly) and 96 hpi in 'Benihoppe'. The expression of FveNPR32a, -32b, and -33 was not significantly altered in "Hawaii4' infected with C. fructicola, while the total transcription of their homoalleles in 'Benihoppe' was significantly suppressed at certain stages post inoculation. In addition, a clear induction of the stress-responsive defense gene FvePR10a and its homoalleles FxaPR10a was detected in both "Hawaii4' and 'Benihoppe' at 48 hpi and 96 hpi. The well-known SA-pathway marker gene FvePR1 was suppressed upon C. fructicola infection in two strawberries (data omitted). Indeed, the expression of FveNPR1, -5, and their homoalleles in octoploid strawberry was too weak to be investigated via quantitative RT-PCR (qPCR). Accordingly, qPCR analysis was only performed for FveNPR31-33 in 'Hawaii4′ and their homoalleles in 'Benihoppe' (Figure 7). It was revealed that FveNPR31 expression was quickly suppressed upon inoculation at 0 hpi (less than 15 min after inoculation) and significantly induced at 96 hpi in 'Hawaii4′. Similarly, the total expression of FxaNPR31abc was up-regulated at 48 hpi (significantly) and 96 hpi in 'Benihoppe'. The expression of FveNPR32a, -32b, and -33 was not significantly altered in 'Hawaii4′ infected with C. fructicola, while the total transcription of their homoalleles in 'Benihoppe' was significantly suppressed at certain stages post inoculation. In addition, a clear induction of the stress-responsive defense gene FvePR10a and its homoalleles FxaPR10a was detected in both 'Hawaii4′ and 'Benihoppe' at 48 hpi and 96 hpi. The well-known SA-pathway marker gene FvePR1 was suppressed upon C. fructicola infection in two strawberries (data omitted).  Figure 6 were analyzed. The same primer pairs for all NPR3-like genes except for NPR32b were used in F. vesca 'Hawaii4′ and F. × ananassa 'Benihoppe'. FvePR10a (FvH4_4g19120) was used as a marker gene of SA-depending defense signaling pathway. The primers and corresponding target alleles amplified were shown in Supplementary Table S6. The gene name followed with '-s' indicates multiple alleles detected in 'Benihoppe'. The relative transcript levels of NPR3-like genes were normalized with two reference genes EF1α and GAPDH2 [49] and reported as the mean of three biological replicates ± SE. The asterisks indicate significant differences based on a Student t-test analysis (*, p < 0.05).

Transcriptome Profile of NPR-like and Related Genes in Strawberry Infected with C. fructicola
The pivotal role of NPR-mediated defense response lies in the interaction with other proteins, including many transcription factors, since NPR1/3/4 are transcription cofactors without a DNA binding domain [6]. Potential-interacting proteins were sought for FveNPRs at STRING (https//cn.string-db.org, accessed on 26 February 2022). Accordingly, a protein-protein interaction network consisting of 65 modes and 371 edges was generated largely based on the known interactions from curated databases and experimentally determined interactions (Figure 8).  Figure 6 were analyzed. The same primer pairs for all NPR3-like genes except for NPR32b were used in F. vesca "Hawaii4' and F. × ananassa 'Benihoppe'. FvePR10a (FvH4_4g19120) was used as a marker gene of SA-depending defense signaling pathway. The primers and corresponding target alleles amplified were shown in Supplementary Table S6. The gene name followed with '-s' indicates multiple alleles detected in 'Benihoppe'. The relative transcript levels of NPR3-like genes were normalized with two reference genes EF1α and GAPDH2 [49] and reported as the mean of three biological replicates ± SE. The asterisks indicate significant differences based on a Student t-test analysis (*, p < 0.05).

Transcriptome Profile of NPR-like and Related Genes in Strawberry Infected with C. fructicola
The pivotal role of NPR-mediated defense response lies in the interaction with other proteins, including many transcription factors, since NPR1/3/4 are transcription cofactors without a DNA binding domain [6]. Potential-interacting proteins were sought for FveNPRs at STRING (https//cn.string-db.org, accessed on 26 February 2022). Accordingly, a proteinprotein interaction network consisting of 65 modes and 371 edges was generated largely based on the known interactions from curated databases and experimentally determined interactions (Figure 8). Following functional enrichment analysis, it was revealed that FveNPR1 and FveNPR31, the potential shuttles between nuclear and cytoplasm predicted in PSORT, might directly interact with four TGA transcriptional factors for gene expression regulation, with one NIMIN-2 and two CUL3a for regulating protein stability, and with MAPK for signaling magnification. Furthermore, two CUL3a proteins could bridge FveNPR1 and -31 with three Ring-box 1a-like proteins, BPM (adapter protein containing BTB/POZ and MATH domains that link cell surface receptors and downstream kinase cascades), and CSN (COP9 signalosome complex subunit like). Later, two UBE (Ubiquitin-conjugating enzyme E27-like), three BTB/POZ domain-containing proteins, and 26S proteasome were recruited into this FveNPR1/31-centored network bridged by CUL3a. In another direction, MAPK might link FveNPR1 and -31 with NDPK (nucleoside diphosphate kinase 2, chloroplastic-like), additional MAPK, and PP2C (protein phosphatase 2C and cyclic nucleo-

Discussion
Generally, NPR-like genes belong to a small family that is highly conserved in many plants. Arabidopsis harbors six NPR-like genes [50]. There are five NPR-like members in rice [51] and Persea americana [13]. Notably, the genomes of Chinese pear (Pyrus bretschneideri Rehd, cv. Dangshan Suli) and apple (Malus × domestica) consist of nine and eight NPRlike members, respectively [20]. Tetraploid emmer wheat and hexaploidy wheat embrace 12 and 17 NPR-like genes, respectively [15]. The allotetraploid genomes of oilseed rape (Brassica napus) [52] and mustard (Brassica juncea) [19] consistently harbor 19 NPR-like Figure 9. RNA-seq data showing the differential expression of genes coding proteins potentially directly or indirectly interacting with NPR1-like proteins (A) and the other known members in SA signaling (B) in strawberry during the C. fructicola invasion. The heatmap was generated using RPKM (reads per kilobase per million mapped reads) values normalized via Log2-transformation for each transcript in a moderate susceptible strawberry cv. 'Jiuxiang' mock-treated or infected by C. fructicola at 24, 72, or 96 hpi. The black star symbols indicate NPR family members. Strawberry materials and C. fructicola inoculation conditions for RNA-seq data generation have been reported previously [35].
Following functional enrichment analysis, it was revealed that FveNPR1 and FveNPR31, the potential shuttles between nuclear and cytoplasm predicted in PSORT, might directly interact with four TGA transcriptional factors for gene expression regulation, with one NIMIN-2 and two CUL3a for regulating protein stability, and with MAPK for signaling magnification. Furthermore, two CUL3a proteins could bridge FveNPR1 and -31 with three Ring-box 1a-like proteins, BPM (adapter protein containing BTB/POZ and MATH domains that link cell surface receptors and downstream kinase cascades), and CSN (COP9 signalosome complex subunit like). Later, two UBE (Ubiquitin-conjugating enzyme E27like), three BTB/POZ domain-containing proteins, and 26S proteasome were recruited into this FveNPR1/31-centored network bridged by CUL3a. In another direction, MAPK might link FveNPR1 and -31 with NDPK (nucleoside diphosphate kinase 2, chloroplastic-like), ad-ditional MAPK, and PP2C (protein phosphatase 2C and cyclic nucleotide-binding/kinase). MAPK and PP2C further linked FveNPR1/31 with PTP (protein-tyrosine-phosphatase), GNB (guanine nucleotide-binding protein), LRR-RLK (leucine-rich repeat receptor-like serine/threonine/tyrosine-protein kinase), WRKY, and HY5-like transcription factors.
Since transcriptome profiling could provide a comprehensive view of a certain regulatory network at a transcription level, we revisited previous RNA sequencing (RNA-seq) data in the current lab to determine how strawberry activates related defenses against the invasion of C. fructicola via motivating NPR-like and related genes in a susceptible cv. 'Jiuxiang' [35]. In control strawberry (mature healthy 'Jiuxiang' mock-treated with water), the relative levels of FveNPRs transcripts ranked as follows: Figure 9A). C. fructicola infection significantly induced the expression of FveNPR31, moderately upregulated FveNPR32 and -33, and inhibited FveNPR5 in 'Jiuxiang' (Figure 9A). FveNPR1 expression was extremely weak, with minor fluctuations. The expression profile of FveNPR31 revealed in RNA-seq was congruent with our RT-PCR analysis in 'Camarosa' and 'Benihoppe'. A set of potential interacting protein ( Figure 8) coding genes were revealed to be differently expressed during invasion with C. fructicola. The transcription of genes encoding the remaining proteins in that PPI network were not differently altered in response to C. fructicola infection (Supplementary Figure S3). In addition, the expression profiles of 14 genes involved in SA-related signaling were identified ( Figure 9B). Upon C. fructicola infection, strawberry SA biosynthesis-related genes ICS and PALs were differentially regulated. The SA pathway marker genes PR1 and PR10b were suppressed, while PR10a was upregulated, largely matching the current RT-PCR analysis.

Discussion
Generally, NPR-like genes belong to a small family that is highly conserved in many plants. Arabidopsis harbors six NPR-like genes [50]. There are five NPR-like members in rice [51] and Persea americana [13]. Notably, the genomes of Chinese pear (Pyrus bretschneideri Rehd, cv. Dangshan Suli) and apple (Malus × domestica) consist of nine and eight NPR-like members, respectively [20]. Tetraploid emmer wheat and hexaploidy wheat embrace 12 and 17 NPR-like genes, respectively [15]. The allotetraploid genomes of oilseed rape (Brassica napus) [52] and mustard (Brassica juncea) [19] consistently harbor 19 NPR-like genes. In this study, a total of 23 NPR-like members were identified in octoploid commercial strawberry F. × ananassa. With the exception of F. nipponica, five to six NPR-like genes were identified in the genomes of diploid Fragaria species, including F. vesca (6), F. viridis (6), F. iinumae (5), and F. nilgerrensis (5). The NPR-like family in F. vesca genome has been previously reported [14], which was largely consistent with the current work, except for that one more member was recruited due to the division of gene28770 (previous FvNPRL-3 in strawberry v1.0-hybrid version) into FvH4_6g38830 (FveNPR32a) and FvH4_6g38821 (FveNPR32b), which might have resulted from alternative splicing. Apart from FveNPR32a and FveNPR32b in F. vesca, there were an additional three pairs of genes physically linked, which might be potentially one locus with alternative splicing patterns. These were FvirNPR32a and FvirNPR32b in F. viridis, FxaNPR32a and FxaNPR32e, as well as FxaNPR32b and FxaNPR32g in F. × ananassa. Moreover, it is worthwhile to note that current FveNPR1 corresponds to the FvH4_1g07810 (gene12668), previously referred to as FvNPR1 [53] or FvNPRL-4 [14]. Current FveNPR31 is virtually the FvH4_3g11950 (gene20070), previously referred to as FvNPR31 [53], FvNPRL-1 [14], or FvNPR1 [35,36]. Furthermore, the primers for detecting FaNPR1 expression previously [37] fit with all four alleles of FxaNPR33a-d in the current work.
Phylogenetic analysis unambiguously clustered Fragaria NPR-like genes into three clades as in all other plants reported. Among NPRs of a certain clade from diverse plants, the conservation is widely observed in the relative length of the exons; the exon-intron organization; and the functionally important residues, motifs, and domains. Interestingly, the pI values of NPRs belonging to a certain clade might be an exceptional case, not conserved among diverse plants. For example, Arabidopsis NPR1 protein has a pI value of 5.86 (www.arabidopsis.org), and the pI values of clade I NPR-likes in three wheat species with distinct chromosome ploidy ranged from 5.18 to 5.88 [15], while in six strawberry species this parameter varied between 6.17 and 6.74. NPR-like genes in Fragaria species might have a unique evolutionary history after the divergence from Arabidopsis. More structural evidence supported that NPR-like proteins in strawberry might have functional divergence or a function somewhat partially different from their orthologs in Arabidopsis. MEME analysis revealed the presence of additional motifs before the BTB/POZ domain in strawberry NPRs of clade II and III (Supplementary Figure S1), which were not observed in Arabidopsis. Sequences of potential IkB-like phosphodegron in clade I NPR-like proteins were not completely conserved in Arabidopsis and strawberry (Supplementary Figure S2B). The Cysteine residue C156 in AtNPR1 related to facilitating NPR1 oligomerization was substituted by Alanine (F. nilgerrensis) or Serine (the rest five strawberry species) residue (Supplementary Figure S2B) [54]. A subgroup represented by FveNPR32a in clade II similarly was composed of three exons, while the other members of clade II consistently have four exons like Arabidopsis orthologues. In addition, the deduced amino acid sequences of FveNPR32a and its subgroup members have a variant form of the putative hinge region 'FENRV' (SA binding pocket) in the C terminal, while all Fragaria members of clade I except for FiinNPR1 ('LENR' followed by a deletion of three amino acids) and the other members of clade II uniformly harbor 'LENRV' (Supplementary Figure S2C), which was identical to that in AtNPR1 [39]. Strikingly, the putative hinge region in AtNPR3/4 is composed of 'LEKRV', which was not observed in Fragaria spp. Previous work suggested that AtNPR4 has an affinity with SA five times higher than that of AtNPR3 and AtNPR1 [4,55]. Apparently, the SA-binding core (SBC) is wider than the hinge region [9].
In contrast to the NPR-like members of clade III, which are mainly involved in leaf and flower development [56,57], NPR family members of clade I and II play broad roles in plant immunity [6,7,58]. The degradation of the transcription cofactor nonexpresser of PR genes 1 (NPR1) acts as a molecular switch in many plants, and NPR4 is indispensable for the function of NPR1 [6]. Previously, the transcription of FveNPR1, the NPR1-like gene (gene12668, FvH4_1g07810, and FvNPRL-4), had never been detected in strawberry except for in an RNA sequencing analysis [14,[34][35][36]53]. RNAseq in early-stage fruits and achenes of woodland strawberry [59] revealed FveNPR1 was largely considered as non-expressed, with transcripts having average reads per kilobase per million (RPKM) lower than 0.3 in all reproductive and vegetative organs except for a peak in achene ghost at stage 5 (10-13 days post anthesis, RPKM: 6.79) and achene wall at stage 1-4 (RPKM: 0.45-3.04). In the current work, based on a comprehensive phylogeny analysis for NPR-like family in octoploid and diploid strawberry, allele specific primers were available for the RT-PCR analysis of NPR1-like genes in strawberry. The expression of FveNPR1 and one allele FxaNPR1a (FxaC_1g08070) was clearly detected in the mature leaves of "Hawaii4' and 'Camarosa' infected or not infected with C. fructicola ( Figure 6B). A fast and obvious inhibition effect of pathogen inoculation was observed for FxaNPR1a transcription, and a similar but weak tendency for FveNPR1 was also detected. Still, its transcript level was too weak to be quantitatively analyzed. Reasonable explanations would be that NPR1 is strictly regulated at both transcription and protein levels, together with a highly efficient translation system for this master regulator. Virtually, it has been suggested that upon infection with pathogens, the translational induction of resistance-related protein may precede transcriptional alteration in plants [60]. Together, both molecular characterization and expression analysis indicated that FveNPR1 and FxaNPR1a were active in strawberry and could respond to C. fructicola infection at a transcriptional level, implying that they probably function in strawberry like their orthologue NPR1 in Arabidopsis.
Among NPR-like members of clade II in strawberry, FveNPR31 has been functionally studied [14,53]. Although RNA interference and the overexpression of FveNPR3 in 'Camarosa' strawberry did not result in decisive conclusions on its role, the overexpression of this gene in wild-type and mutant (npr1-1, npr3-1, and npr4-3) Arabidopsis strongly evi-denced that it might function as a negative regulator of plant immunity such as AtNPR3 or AtNPR4 in Arabidopsis [53]. The transcription of FveNPR31 and its homoalleles could be induced by exogenous SA and Colletotrichum spp., or suppressed by exogenous MeJA and Podosphaera aphanis, dynamically varying with the cultivar genotype [14,[34][35][36][37]53]. Based on our previous [35,36] and the current work under the same inoculation conditions with the same C. fructicola strain, it was concluded that in a susceptible cultivar such as 'Jiuxiang' or 'Camarosa', the transcription of FxaNPR31s could be quickly motivated upon C. fructicola inoculation but soon relaxed at 6 hpi and again up-regulated in the later stage when the pathogens experienced a necrotrophic way for colonization (simultaneously with necrosis syndromes visible). All these observations triggered speculation that the transcription of strawberry NPR31, the potential negative regulator of SA-dependent immunity, might be hijacked by Colletotrichum pathogen at an early stage (1-2 hpi), but it was soon quenched (6 hpi) when strawberry begun to frustrate this invasion via motivating multiple defense strategies such as PTI, ETI, and SAR. Furthermore, when necrosis syndrome appeared and the hemibiotrophic pathogen Colletotrichum initiated a necrotrophic life (2 dpi and later), there was a transition of defense strategy in strawberry to largely rely on the JA-mediated network, while SA-mediated pathways might be restricted through the activation of the negative regulator FveNPR31.
Finally, to summarize the transcriptional changes observed in both RNAseq and RT-PCR for strawberry NPR1, NPR31, and related partners suggested from the PPI prediction, schematic representations were proposed to facilitate our understanding of molecular mechanisms underlying NPR-related immunity during strawberry-Colletotrichum spp. interaction ( Figure 10A,B). A glimpse of these transcriptional events would shed some light on how strawberry orchestrates NPRs-centered defense system against C. fructicola.

Plant Materials, Growth Conditions, and Inoculation with C. fructicola
The woodland strawberry Fragaria vesca "Hawaii4' and cultivated strawberry F. × ananassa cultivars 'Camarosa' and 'Benihoppe' were used. Healthy plants with 6-8 compound leaves were developed from stolon-derived seedlings after three months' cultivation in a greenhouse. These plants were transplanted into pots and cultivated in a growth chamber (GXZ-1000, Jiangnan Instrument Factory, Ningbo, China) under a 12 h-light/12 h-dark cycle at constant 25 • C for two more weeks before inoculation. C. fructicola strain CGMCC3.17371 was used for preparing inoculum. Inoculation was carried out in two ways [61]. For RNA extraction and gene expression analysis, whole plants were spray-inoculated, largely as previously described [35]. For "Hawaii4' and 'Benihoppe', 18 plants divided into three groups were sprayed with C. fructicola conidial suspensions at a concentration of 2 × 10 6 per ml with 0.01% (v/v) Tween-20. Simultaneously, additional 9 plants were mock-treated with sterile ddH 2 O containing 0.01% (v/v) Tween-20. For 'Camarosa', 12 plants were inoculated and 6 plants were mock-treated. The third to fifth fully expanded leaves from three (mock-treated) or six plants (C. fructicola inoculated) were harvested and pooled for one replicate at 0, 6, 24, 48, and 96 hpi. A total of three biological repeats were obtained for each treatment × time combination. For 0 hpi, sampling was accomplished within 15 min after spaying. Leaf blades were immediately frozen in liquid nitrogen and then stored at −80 • C before RNA purification.
In addition, in-vitro inoculation on wounded detached leaves was performed as previously described with minor modifications [62]. The third or fourth fully expanded leaves were detached with the cut end wrapped with paraffin ceresin and placed on wet sterile filter paper in sealed petri dishes (diameter, 15 cm) to maintain humidity. Each leaf blade was injured with a pipette tip at four sites. Two 10-µL drops of freshly prepared conidia solutions at a concentration of 2 × 10 6 per mL with 0.01% (v/v) Tween 20 were deposited on the injured sites on the right side, with mock treatment (sterile water with 0.01% (v/v) Tween 20) on the left side. A total of nine leaf blades were inoculated for each strawberry variety. The typical necrosis of anthracnose syndromes was photographed at 4dpi and 5 dpi.
NPR-like proteins in six Fragaria species were symbolized with specific numbers followed by FxaC, FvH4, evm.model.scaf, FnYN, evm.model.ctg, and FNI, representing those from F. × ananassa, F. vesca, F. iinume, F. nilgerrensis, F. viridis, and F. nipponica, respectively. Phylogenetic analysis was performed for a set of 54 NPR-like proteins from Arabidopsis and aforementioned six Fragaria species using a neighbor-joining (NJ) statistical method in MEGA v7.0 [64]. The corresponding tree was constructed using 1000 re-samplings in MEGA and visualized via using the interactive Tree Of Life (iTOL) [65].
The in silico prediction of biochemical characteristics was performed for each strawberry NPR-like member. The subcellular localization was predicted at https://wolfpsort. hgc.jp/, accessed on 12 October 2021. The isoelectric point (pI) and molecular weight of the protein sequences were calculated using the online Compute pI and Mw tool at https://web.expasy.org/, accessed on 12 October 2021.

Exon-Intron Structure in NPR-like Genes and Domain Organization in Deduced Proteins
Information for the exon-intron structure of the NPR-like genes was obtained by aligning the coding sequences (CDS) from SGD or GDR with the corresponding genomic DNA sequences from GDR. A systematic analysis was performed via using the gene structure display serve (GSDS) 2.0 (http://gsds.cbi.pku.edu.cn/, accessed on 27 February 2022). A Newick-type file was generated in MEGA7.0 with the CDS nucleotide sequences of all Fragaria NPR-like genes after MUSCLE clustering. This file was further introduced into GSDS as a neighbor-joining tree to arrange the exon-intron structures of strawberry NPR-like genes.
Conserved protein domains were identified at PFAM and illustrated with IBS1.0 (Illustrator for Biological Sequences 1.0) [66]. To identify the conserved motifs in strawberry NPR proteins, the MEME program at http://meme-suite.org/ (accessed on 11 October 2021) was employed using the following settings: optimum width, 15-60; number of repetitions, any; and maximum number of motifs, 10. Multiple sequence alignment was performed via using Clustal Omega at https://www.ebi.ac.uk/Tools/msa/clustal/ (accessed on 21 December 2021) and further visualized using Jalview (https://www.jalview.org/, downloaded on 25 September 2021) [67]. The percentage of identical amino acids among strawberry NPR-like proteins was calculated after ClustalW analysis.

Chromosomal Locations of NPR-like Genes in F. vesca and F. × ananassa
The genomic locations of NPR-like genes were obtained from F. vesca and F. × ananassa genome databases at GDR [23] and assigned to distinct chromosomes or subgenomes [29]. Mapchart1.0. software was used to draw the physical maps showing the distribution of NPR-like genes [68].

Cis-Elements in Promoters of NPR-like Loci in F. vesca and F. × ananassa
The promoter sequence of 2500-bp in length was extracted from GDR at the upstream of the ATG translation start codon for each NPR-like genes in F. vesca and F. × ananassa. Cis-elements were predicted in the PlantCare database at http://bioinformatics.psb.ugent. be/webtools/plantcare/html (accessed on 9 February 2022) [69] and illustrated via using TBtools package [70].

Protein-Protein Interaction Prediction for Strawberry NPR-like Proteins
Potential protein-protein interactions (PPI) were predicted for the diploid strawberry NPR-like proteins from F. vesca using the STRING database version 11.0 (http: //string-db.org/cgi/input.pl, accessed on 26 February 2022) [71]. Functional enrichment analysis started with all FveNPR members beyond FveNPR5 and resulted in a network harboring 65 nodes linked with 165 expected edges (expected interactions) and a total of 371 edges (potential interactions included) with a p-value less than 1.0 × 10 −16 . The interaction network was built via seven times of PPI enrichment analysis, resulting in ever increasing proteins of known interactions from curated databases and experimentally determined, predicted interactions due to the gene neighborhood, gene fusions, and gene co-occurrence, as well as other potential interactions based on text-mining, co-expression, and protein homology.

RNA Purification, RT-PCR, and RNA Sequencing
The infected or mock-treated plant materials were homogenized by liquid nitrogen, and total RNA was extracted using the OMEGA Plant RNA kit (OMEGA Bio-tek, Inc. Cat#R6827, Norcross, GA, USA) no longer than two weeks after sampling. For RT-PCR, cDNAs synthesis was achieved by using the HiScript III RT SuperMix for qPCR kit with gDNA wiper (Vazyme, Lot#R323, Nanjing, China).
RT-PCR primers matching each FveNPR and its homoeologous FxaNPRs were first designed based on sequence alignment to avoid single nucleotide polymorphisms. Furthermore, allele-specific primers were designed for clade I NPR1-like loci in octoploid strawberry. The information for all primers was shown in Supplementary Table S6. For NPR-like family genes in F. vesca and NPR1-like genes in F. × ananassa, gene-specific primer pairs were used to distinguish each allele. For NPR-like genes of clade II and III, the SA signaling marker gene PR10a, and thge two reference genes EF1α (FvH4_3g33150, gene28639) and GAPDH2 (FvH4_4g24420, gene07104) [49], the same primer set was used for analysis in F. vesca and F. × ananassa, which means that the homologous alleles were not distinguished in octoploid strawberry.
Semi-quantitative RT-PCR was accomplished in an ETC 821 DNA amplifier (Eastwin, Beijing, China) via using the 2xTaq PCR Master Mix (Cat #PM1101, Biosune, Shanghai, China) in a 20-µL cocktail. PCR cycles for the internal control EF1α and NPR-likes were 28 and 40, respectively. For EF1α and strawberry NPR3-like genes of clade II, 0.2-µL original cDNAs was used as a template, while for NPR-like genes of clade I and III, 1-µL original cDNAs was used. After thermal cycles, 6.5 or 7.5-µL PCR products were detected on 2% agarose gel. Three biological replicates were tested.
Real-time quantitative PCR (qPCR) assay was performed for NPR3-like genes of clade II in F. vesca "Hawaii4' and F. × ananassa cv. 'Benihoppe'. The ChamQ TM Universal SYBR qPCR Master Mix (Vazyme, Lot#Q711, Nanjing, China) was used in a 12-µL cocktail following user manuals. qPCR was conducted on a Light Cycler 480 (Roche, USA). For all NPR3-like genes, primer amplification efficiency (E) was first examined with a gradient of diluted cDNA templates. FveEF1a and FveGAPDH2 were used for expression normalization. qPCR analysis was performed using the (1 + E) −∆∆Ct method combined with a normalization factor calibrated with geNorm software [72]. Three biological replicates were examined with each in three technical repeats. Student t-test analysis was performed to reveal the significance of differences.
RNA sequencing for C. fructicola-infected strawberry cv. 'Jiuxiang' in the current laboratory has been reported previously [35].

Conclusions
In summary, this study reports a comprehensive identification of NPR-like family genes in the genomes of octoploid strawberry and its five diploid relatives based on updated whole genome information. Phylogeny analysis and molecular characterization enabled us to recognize all 23 NPR-like alleles in commercial strawberry. The mystery of putative NPR1 orthologue in strawberry may be due to the fact its transcripts in most organs are at an extremely low level and had been never detected before this work due to impaired efficiency in primers confused with distinct homoalleles. We provided experimental evidence for the view that FveNPR1 and FxaNPR1a were active genes transcriptionally responsive to pathogen invasion in strawberry. Furthermore, FveNPR31 and three homoalleles FxaNPR31a-c in octoploid strawberry were transiently induced (perhaps hijacked by pathogens) at an early stage and finally motivated to reduce the antagonizing effect of the SA-dependent pathway on JA-mediated resistance, which is required for plants defending a necrotrophic pathogen. Thus, FveNPR1 and FxaNPR1a, as well as FveNPR31 and FxaNPR31a-c, are putative functional orthologues of AtNPR1 and AtNPR3/4, respectively, serving as important candidate genes for future functional study aimed at improving strawberry disease resistance.