Characterization and Comparison of the CPK Gene Family in the Apple (Malus × domestica) and Other Rosaceae Species and Its Response to Alternaria alternata Infection

As one of the Ca2+ sensors, calcium-dependent protein kinase (CPK) plays vital roles in immune and stress signaling, growth and development, and hormone responses, etc. Recently, the whole genome of apple (Malus × domestica), pear (Pyrus communis), peach (Prunus persica), plum (Prunus mume) and strawberry (Fragaria vesca) in Rosaceae family has been fully sequenced. However, little is known about the CPK gene family in these Rosaceae species. In this study, 123 CPK genes were identified from five Rosaceae species, including 37 apple CPKs, 37 pear CPKs, 17 peach CPKs, 16 strawberry CPKs, and 16 plum CPKs. Based on the phylogenetic tree topology and structural characteristics, we divided the CPK gene family into 4 distinct subfamilies: Group I, II, III, and IV. Whole-genome duplication (WGD) or segmental duplication played vital roles in the expansion of the CPK in these Rosaceae species. Most of segmental duplication pairs in peach and plum may have arisen from the γ triplication (~140 million years ago [MYA]), while in apple genome, many duplicated genes may have been derived from a recent WGD (30~45 MYA). Purifying selection also played a critical role in the function evolution of CPK family genes. Expression of apple CPK genes in response to apple pathotype of Alternaria alternata was verified by analysis of quantitative real-time RT-PCR (qPCR). Expression data demonstrated that CPK genes in apple might have evolved independently in different biological contexts. The analysis of evolution history and expression profile laid a foundation for further examining the function and complexity of the CPK gene family in Rosaceae.


Introduction
Plant growth and crop production are adversely affected by environmental stresses, such as drought, low temperature, high salinity, pathogen infection, microbial elicitors and wounding. To adapt to these environmental conditions, plants have developed a mechanism that includes the perception of stress signals, subsequent signal transduction, and the activation of various physiological and metabolic responses [1]. Calcium ions (Ca 2+ ) play a central role as a second messenger in the signal transduction pathways [2]. In plants, the perturbation of cytosolic Ca 2+ EF-hand domain (PF13499 and PF13202) were downloaded from the Pfam protein family database (http://pfam.sanger.ac.uk/). These HMM profiles were used to search proteomes of apple, pear, peach, plum and strawberry by hmmsearch with the threshold set of the Pfam GA gathering cutoff. Then, the HMMER selected proteins were used for a BLASTP query of the original proteomes. Finally, the BLASTP hits were scanned for kinase and EF-hand domains using InterProScan [33], then truncated sequences and pseudogenes were removed. Initially, 129 nonredundant putative genes were identified. After manually checking the sequences of these genes, 6 genes were removed for having diverse sequences.
In comparison with a previous study [32], in which 30 MdCPK genes have been identified, 37 MdCPK genes were found in this work, although the apple genome data used is the same version deposited in Phytozome. After carefully checking, we found that the discrepancy in number of CPKgenes lies in the methods used by these two studies. The previous work used BLASTP to search apple genome using CPK genes from Arabidopsis thaliana, then all the candidate CPKs of apple were scanned using SCAN PROSITE software to confirm the presence of the EF-hands signature motif. We adopted a HMMER-BLASTP-InterProScan strategy that mined the CPK genes based on the existence of the HMM profiles of protein kinase domain and EF-hand domain. The advantage of our method is that HMM profile was constructed from some representative CPKs from various species, which covered more domain information than that only from Arabidopsis CPKs. Our search strategy was tested against Arabidopsis and detected all previously reported 34 CPKs, with AtCPK25 removing for having truncated conserved domain. Using this strategy, we detected 37 CPK genes in apple genome, which include all the genes reported by [32] except one gene, MDP0000169895, in which no EF-hand was detected. The existence of protein kinase domain and EF-hand domain were further confirmed by SMART (http://smart.embl-heidelberg.de/), PfAM (http://pfam.xfam.org/), and SUPER-FAMILY (http://supfam.org/SUPERFAMILY/). Therefore, additional eight CPK genes were identified by this study.
Finally, a total of 123 CPK genes were determined as CPK genes in the five species. These CPK genes in each species were named based on the similarity with Arabidopsis CPKs as the nomenclature in [32]. A total of 37 apple CPK (MdCPK) genes, 37 pear CPK (PbCPK) genes, 17 peach CPK (PpCPK) genes, 16 strawberry CPK (FvCPK) genes, and 16 plum CPK (PmCPK) genes were identified (Table 1, S1-S4 Tables). At the time of analysis, nearly complete genomes were available for apple, peach, strawberry and plum, while pear only had scaffold assemblies data available. In apple genome, chromosome 16 did not contain any CPK genes. The highest number of MdCPK genes (5, or 13.6% of the total) was found on chromosome 8 (Fig 1). Only one CPK gene was located on chromosome 1, 3, 4, 5, 6, and 13, respectively. Similar to that of MdCPKs, the distribution of the CPK genes in peach, strawberry and plum genomes is random (Fig 1). CPK genes in pear can only be anchored to scaffold assemblies now (S1 Table).

Phylogeny of CPK genes
To detect the evolutionary relationships of CPK genes in Rosaceae, we performed phylogenetic analyses on the 123 CPK genes from apple, pear, peach, plum and strawberry using Neighbor-Joining (NJ), Minimal Evolution (ME) and Maximum Parsimony (MP) algorithms, respectively. The tree topologies produced by the three methods are largely consistent, with only minor modifications (data not shown). Therefore, the NJ tree was selected for further analysis (Fig 2). Based on the phylogenetic tree topology and previous studies, we divided the CPK gene family into 4 distinct groups (subfamilies): Group I, Group II, Group III, and Group IV. Group I contains 16 CPK genes from apple, 14 from pear, and 6 from each of strawberry, peach and plum. Group II consists of 8 MdCPKs, 7 PbCPKs, 5 FvCPKs, 5 PpCPKs and 5 PmCPKs. In Group III, 12 CPK genes are from apple, 14 form pear, 4 from strawberry, 5 form peach and 4 from plum. Group IV genes constitute the smallest subfamilies in all of the five species, which contain two CPK genes from pear, one from each of other species. The number of CPK genes in Group I and Group III from apple and pear were greater than that from strawberry, peach and plum, mainly due to the expansion of CPK genes in apple and pear. And the size of Group II in apple and pear were slightly bigger than other three species. These results implied that the variance in the number of CPK genes was mainly due to the occurrence of gene gain or loss in subfamilies independently among the different species. Using a CPK gene from Ostreococcus lucimarinus as the outgroup for the five Rosaceae species and Arabidopsis CPKs, the general topology of the resulting Neighbor-Joining tree appeared similar to that of a previous study [27]: the Group IV lineage appeared to have split first from the last common ancestor. Group III formed a clade close to Groups I and II, while the divergence between Groups I and II occured at the most recent time.

Gene structure and conserved domains in MdCPK genes
Most of the apple CPK genes have six or seven introns, with clear intron phase patterns (Fig 3). For the members in each subfamily, the intron number and phase pattern on the full length protein sequences are variable, but conserved on the protein kinase domain. Of the 16 members in Group I, 11 genes contain 3 introns with the same phase pattern in kinase domain. Interestingly, three intronless CPK genes were found in Group I. These CPK genes without introns were also observed in other Rosaceae species(pear, peach, plum and strawberry; data not shown), while they have not been reported previously outside Rosaceae species, indicating introns may be lost during the evolution of these CPK genes in Rosaceae. All of the Group II members share the same intron-exon organization as most of the Group I, except MdCPK9b and MdCPK19. Group III members have 4 introns with the same phase pattern, except MdCPK13a and MdCPK13b, which contain 3 introns. Compared with the other three groups, Group IV only contains one MdCPK gene, which has 6 introns on kinase domain.
The most prominent feature of the proteins in CPK gene family is the typical signature domains. We surveyed top 10 motifs in the 37 MdCPK proteins using The Multiple EM for Motif Elicitation (MEME) motif search tool. Protein sequences were also compared with wellcharacterized signature domains in databases, such as SMART, SUPERFAMILY, Pfam, ProSite and Profiles (data not shown). As showed in Fig 4, motifs 1, 2, 3, 6 and 7 correspond to protein kinase domain; motifs 4, 5, 8 and 9 correspond to calmodulin-like domain; motif 10 corresponds to the junction domain (S5 Table).

Expansion of the CPK gene family in Rosaceae
It is thought that genes in a family usually evolved from multiple gene duplication events. Gene duplication mechanisms mainly include tandem duplication and large segmental/wholegenome duplication (WGD). To examine the relative contribution of these kinds of mechanisms in the expansion of the CPK gene family in Rosaceae, we detected the origins of duplicate genes for the CPK gene family using the MicroSyn package [34]. Different patterns of gene duplication contributed differentially to the expansion of the CPK gene family in the investigated species (Table 2). Remarkably, 15CPK genes in apple were duplicated and retained from WGD events, compared to 10 in peach and 6 in plum. No WGD events can be detected between CPK genes in strawberry. We did not find tandem duplication events for CPK genes in strawberry, peach and plum, however, 10 CPK genes in apple genome were found to form 4 tandem duplicated clusters. These results showed that WGD or segmental duplication played vital roles in the expansion of the CPK genes in apple, peach and plum. However, for the CPK genes in strawberry, other duplication events, such as dispersed gene duplication, is the main source of expansion.
With the availability of genome sequence, studies of CPK gene family duplication history become possible in part by detecting synteny or clustering and ordering of neighboring matching gene pairs [35]. Using the information of neighboring matching gene pairs between two interested genes, synteny can usually detect duplication event of two chromosome fragments containing interested genes. For the five Rosaceae species investigated here, apple and pear belong to the Maloideae, peach and plum belong to the Prunoideae, and strawberry belongs to the Rosoideae. Owing to the lack of complete genome sequence for pear and computational limit, we select apple from the Maloideae, peach from the Prunoideae and strawberry from the Rosoideae to further survey the origin and evolution of CPK genes in Rosaceae. The synteny analysis showed that synteny relationship was mainly detected within subgroups, indicating that if the synteny between two members of a gene family is more significant, these two members evolved from a duplication event more recently (Fig 5, S6 Table). Two syntenies were also detected between subgroups, which indicates that some duplication history can be traced between subgroup. As illustrated in Fig 5, one synteny relationship was detected between subgroup I and II, and subgroup II and III, respectively, which support the relationship of 4 subgroups inferred by the feature of phylogenetic tree [27].
The synteny analysis showed that CPK genes can be divided into two types in Rosaceae. The first type of syntenic genes has a single strawberry and plum gene that corresponds to two apple gene, such as FvCPK1/PpCPK1/MdCPK1a/MdCPK1b, FvCPK20/PpCPK20/MdCPK20a/ MdCPK20b (Fig 5). The second type has a single strawberry and peach gene that corresponds to a single apple gene, such as MdCPK29/PpCPK29/FvCPK29 and FvCPK28/MdCPK28/ PpCPK28. These results provide insights that will assist in understanding of orthologous relationship among CPK genes in Rosaceae.

Ks value and Ka/Ks ratio of CPK genes
To estimate the evolutionary dates of the segmental duplication events among the CPK gene family, we calculated the synonymous substitution (Ks) values between each pair of duplicated genes. The mean Ks of the duplicated CPK gene pairs in the syntenic region are shown in Table 2. The Ks values for the CPK gene pairs ranged from 0.13 to 1.15. We found that most of segmental duplication pairs in peach and plum and one pair in apple (MdCPK13a/MdCPK2; Ks = 1.15) may have arisen from the γ triplication (~140 million years ago [MYA]). While in apple genome, many duplicated gene pairs had lower Ks values (0.19-0.55), suggesting that these duplications may have been derived from the recent WGD (30~45 MYA).
To further detect which selective force has been acted on the evolution of the CPK gene family, we also calculated the ratio of non-synonymous to synonymous substitution ratio (Ka/Ks) for 8, 11 and 3 pair of duplicated CPK genes in apple, peach and plum, respectively. All the Ka/ Ks ratios of gene pairs were less than one, implying that purifying selection was the primary influence on the functional evolution of CPK family genes.

Expression profiles of the MdCPK genes in response to Alternaria alternata
Recent studies have provided compelling evidence for the involvement of CPKs in most of the immune signaling events. Alternaria blotch disease of apple is one of the most serious fungal diseases, which is caused by the apple pathotype of Alternaria alternata. To examine the expression pattern of MdCPK genes responding to this pathogen, we examined their expression information using quantitative real-time RT-PCR (qPCR) analysis with RNA from apple leaves sampled at 18 h, 36 h, and 72 h after A. alternata inoculation (hai). Among the 37 MdCPK genes, the qPCR products have been confirmed by sequencing except 9 genes for their unspecific amplification or undetectable expressions, providing a measure of the reliability of the qPCR results of MdCPK expression. The sequences of primers were listed in S7 Table. As shown in Fig 6, the expression pattern of MdCPK genes in response to A. alternata can be divided into 4 clusters. Cluster A contains 12 (42.9%) members of detectable genes, which were significantly up-regulated at 18, 36 and 72 hai, except MdCPK32a, which showed slight downregulation at 18 hai. Cluster B contains 3 genes, which exhibited lower abundance of expression at 18 hai, while were highly induced at 36 and 72 hai compared with control. In Cluster C (3 genes), all the MdCPK genes were down-regulated after pathogen infection throughout the three time points. Genes in Cluster D are up-regulated at 36 hai, whereas down-regulated at18 and 72 hai, except two genes (MdCPK1b and MdCPK17b), which displayed no significant changes of expression at 36 hai.

Discussion
CPKs are encoded by a multigene family, which have been identified and analyzed in many land plant species. The number and composition of CPK family members differ in various plants. Ancient polyploidy events (hexaploidization) and additional recent lineage-specific WGDs have presumably resulted in varying numbers of gene family within flowering plants. In this study, the number of the CPK gene families identified from the five Rosaceae genomes is diverse. The number of CPK genes in apple (37) and pear (37) are greater than that in peach (17), strawberry (16), and plum (16). Apple and pear were suggested to have undergone a recent WGD, while peach, strawberry, and plum did not undergo this event [36]. Therefore, this recent WGD event likely led to the different size of CPK genes in these Rosaceae species.
For the five Rosaceae species investigated here, apple and pear belong to the Maloideae, peach and plum belong to the Prunoideae, and strawberry belongs to the Rosoideae. The specification of the Rosoideae occurred prior to the split of the Maloideae and Prunoideae. The phylogenetic analysis showed that MdCPKs were closest to PbCPKs in the phylogenetic tree, while PpCPKs and PmCPKs had a closer relationship, which was consistent with the evolutionary history among the three subfamilies.
Different expansion mechanisms, such as genome-wide, tandem, and dispersed duplications, were thought to play a significant role in the expansion of specific gene families in plant genomes [37]. It has been shown that the three whole-genome duplications in Arabidopsis were responsible for more than 90% of the increase in transcription factors, signal transducers, and developmental genes [38]. Recently, genome-wide studies have revealed that the apple and pear genomes have experienced at least two genome duplications, one ancient and one before the apple-pear divergence [39]. In this study, the size of apple and pear CPK gene family is as twice as that of peach, strawberry and plum. The results of the synteny analysis verified that the expansion of the CPK gene family in apple, peach and plum was mainly derived from WGD or segmental duplications and tandem duplication. In strawberry, only a few or no significant segmental duplications were detected. However, the result of synteny analysis across apple, strawberry and peach revealed that almost all CPK genes in strawberry have orthologous genes in apple and peach genome. In strawberry, other duplication events, such as dispersed duplications might be the major drivers for CPK gene family expansion. The mean Ks value of the duplicated blocks of CPK genes in peach and plum are greater than that of apple, indicating that expansion of PpCPK and PmCPK genes may have arisen from the γ triplication (~140 MYA). It is worth noticing that the synteny detection algorithm determines the relationship between two members in a gene family by searching for conserved, flanking collinear homologous gene pairs between two genomic fragments. For the ancient duplication events, the original flanking homologous gene pairs were hard to determine, leading to uncertain result of synteny analysis. Therefore, it might be that CPK genes in strawberry were evolved from the ancient WGD or segment duplications, but they are difficult to be detected now. Other mechanisms, such as genome rearrangement, gene loss, gene transposition and retrotransposition after the ancient polyploidy event, may also have affected the evolution of the CPK gene family in strawberry.
The genes duplicated through WGD might experience three alternative fates during the process of evolution, including (i) one copy may become silenced and lost original functions (nonfunctionalization), (ii) one copy may acquire a novel, beneficial function, while the other copy retained the original function (neofunctionalization) and (iii) both copies may become partition of original functions (subfunctionalization) [40]. It has been shown that the retention of genes duplicated through WGD is biased in plant genomes [41]. In Arabidopsis, genes encoding transcription factors, protein kinases, and ribosomal proteins have been preferentially retained after WGD [42]. It is hypothesized that the overretention of duplicated genes through WGD is strongly correlated with greater structural complexity, highly conserved domains, and lower evolutionary rates in the plant genome. Multiple models may simultaneously drive the evolution of genes duplicated after WGDs. In this study, the CPK gene family has undergone specific expansion and been preferentially retained in Rosaceae. Rosaceae CPK family genes contain several highly conserved functional domains, and present lower Ka/Ks ratios, corresponding to a slower evolutionary rate. These results implied that functions of the duplicated CPK genes in Rosaceae did not diverge much during subsequent evolution. These stable function of CPK genes over recent years may serve as good targets for dosage balance selection [43].
To resist pathogen, plants have evolved two defense mechanisms to sense pathogens invasion [44]. On the cell-surface of host, pathogen-associated molecular patterns (PAMPs) are recognized by pattern recognition receptors (PRRs), which are subsequently stimulated to trigger PAMP-triggered immunity (PTI).In response, pathogens have evolved the means to suppress PTI by secreting effectors inside the plant cell. These pathogen effectors are recognized by intracellular nucleotide-binding leucine-rich repeat (NB-LRR) immune sensors, which activate the second type of immune defense mechanism, effector-triggered immunity (ETI). PAMPs initiate an influx of calcium ions and an oxidative burst, followed by activation of MAPK and calcium-dependent protein kinase [45,46]. In these studies, the expression pattern of MdCPK genes after A. alternata infection was surveyed by qPCR. These genes included 10 MdCPKs from Group I, 8 from Group II, 9 from Group III and 1 from Group IV. In Arabidopsis, the expression of AtCPK1 from Group I is rapidly induced by fungal elicitors. Overexpression of AtCPK1 confers broad-spectrum resistance to bacteria and fungi. Long-term AtCPK1 over-expression triggered salicylic acid (SA) accumulation and constitutive expression of SA-regulated defense and resistance genes [14]. MdCPK2, a closest homolog of AtCPK1, displayed continuously up-regulation during the three time points (18, 36 and 72 hai). A functional genomic screen identified four related AtCPKs in Group I (AtCPK4, AtCPK5, AtCPK6 and AtCPK11), as early transcriptional regulators in MAMP signaling [47]. The closest homolog of AtCPK4 and AtCPK11 are MdCPK4b and MdCPK4a. The transcript accumulation of MdCPK4b reduced at early stage of infection, and subsequently increased 36 hai, and then declined 72 hai, whereas MdCPK4a showed continuously up-regulation. These results indicate that some CPKs from Group I exhibittransient and sustained transcriptional modifications upon pathogen infection. Interestingly, in this study, most of the members (77.8%) in Group III exhibited continuously up-regulation at the three time points, suggesting their conserved function in plant defense response. By contrast, several MdCPKs (MdCPK17b, MdCPK1b, MdCPK28, MdCPK5b, and MdCPK20b) showed down-regulation throughout all the treatments. These results suggest that CPK genes in apple might have evolved independently in different biological contexts.

Conclusions
A total of 123 CPK genes were identified in the five Rosaceae genomes. Based on the phylogenetic tree topology and structural characteristics, these CPK genes were divided into 4 distinct subfamilies (Group I, II, III and IV). Collinearity analysis showed that many duplicated genes in apple genome may have been evolved from a recent WGD (30~45 MYA), whereas, most of segmental duplication genes in peach and plum may have arisen from the γ triplication (~140 MYA). Purifying selection is the major force driving the function evolution of CPK family genes. qPCR evidence showed that MdCPKs genes might have evolved independently in different biological contexts. These results in this study laid a foundation for further examining the function and complexity of the CPK gene family in the Rosaceae.

Gene Identification
The genome sequences of apple, peach, and strawberry were downloaded from Phytozome (http://phytozome.jgi.doe.gov/pz/portal.html#). The pear genome sequence was downloaded from the pear genome project(http://peargenome.njau.edu.cn/), and the plum genome sequence was downloaded from the Prunus mume Genome Project (http://prunusmumegenome.bjfu.edu. cn/index.jsp).The complete genome, proteome sequences of Arabidopsis was obtained from The Arabidopsis Information Resource (version 10; http://www.arabidopsis.org). In proteome datasets, if two or more protein sequences at the same locus were identical where they overlapped, we selected the longest sequence. The HMM profiles of protein kinase domain Pkinase (PF00069) and EF-hand domain (PF13499 and PF13202) were downloaded from the Pfam protein family database (http://pfam.sanger.ac.uk/). HMMER was used to search a customized database containing the proteome with the threshold set of the Pfam GA gathering cutoff. The HMMER selected proteins were used for a BLASTP query of the original protein database. Finally, the BLASTP hits were scanned for protein kinase domain Pkinase and EF-hand domains using InterProScan.

Phylogenetic tree building and protein motif prediction
The amino acid sequences of full length sequences were aligned using the MUSCLE [48] with default parameters. Phylogenetic trees for the aligned sequences were constructed using NJ, ME and MP methods. The statistical support of the retrieved topology was assessed using a bootstrap analysis with 1000 replicates for trees. The conserved motifs in the proteins were detected by MEME (http://meme.nbcr.net/meme/cgibin/meme.cgi), with the following parameters: number of repetitions: any; maximum number of motifs: 20; the optimum motif widths: 6-200 amino acid residues.

Chromosomal distribution and gene duplication
The genes were plotted separately onto the chromosomes according to their locations on the chromosomes in the GFF file. Genes with a maximum of 5 genes distance were considered to be tandem duplicates. The syntenies between each pair of members were detected by using the MicroSyn software. The parameters were set as follows: window size of 100 genes, tandem gap value of 2, expected threshold value cut off of 1e-10, and 8 homologous pairs to define a syntenic segment. The mean Ks values of orthologous gene pairs in the same synteny block, the Ka and Ks were calculated by Microsyn.

Plant materials and pathogen treatments
To examine the expression of MdCPK genes, the leaves of cultivar, 'Red Delicious', inoculated with A.alternata were used. A. alternata was expanded on potato dextrose agar (PDA; 200 g potato extract, 20 g dextrose, 20 g agar, 1L water) medium for 5 d at 26°C under dark conditions. The inoculation method was carried out according to the protocol described previously [49]. The mycelia were punched using a hole puncher (DIA. = 5 mm). Six pieces of mycelium 'cake' were put on each side of midrib of the abaxial leaf surface, and then incubated at 25°C under a 14 h light/10 h dark cycle in sterilized plastic chambers. Leaves were sampled at 0, 18, 36, and 72 hours after inoculation.

Gene expression analysis by quantitative real-time PCR
Quantitative real-time PCR was carried out on three independent biological replicates of each sample, as well as three technical replicates for each reaction. Total RNA was extracted using the CTAB method. The qualities and quantities of RNA were determined by 1.5% agarose gel electrophoresis and Nanodrop ND-1000 Spectrophotometer (Thermo Fisher Scientific Inc.; USA), respectively. A SMART™ PCR cDNA Synthesis kit was used to synthesize the cDNA from 1μg total RNA. Primers of MdCDPK genes for qPCR were designed using Beacon Designer v7.0 (Premier Biosoft International, California, USA). The real-time quantification was performed using an ABI7300 Detection System using SYBR Green qRT PCR kits (TaKaRa, Toyoto, Japan). PCR mixtures (final volume, 20 μL) contained 10 μL of 2×SYBR Premix, 2 μL of cDNA template, 200 nM of each primer. The PCR reaction was performed as the following conditions: for 3 min denaturation at 94°C followed by 40 cycles of 94°C for 20 s, 60°C for 20 s, and 72°C for 40 s. Following amplification, melting curves were determined with the following program: 95°C for 15 s, 60°C for 1 min, and 95°C for 15 s. For the accuracy of results, each reaction was conducted in triplicate. Expression levels of the selected genes were normalized to tubulin expression. 2 -ΔΔCt method [ΔΔCt = (Ct target gene −Ct tubulin gene ) treatment −(Ct target gene -Ct tubulin gene ) control ] was used to calculate the relative expression of MdCDPK genes. The gene expression levels were visualized as the heat map and histogram based on the value of log 2 (2 -ΔΔCt ). Heatmaps were generated using R package 'gplots' (http://www.bioconductor.org/). Clustering in the heatmap was carried out using Hierarchical Clustering by hclust function in R (http://www.r-project.org/). Supporting Information S1 Table. CPK genes and related information in pear. (DOC) S2 Table. CPK genes and related information in strawberry. (DOC) S3 Table. CPK genes and related information in peach.