Genome-Wide Characterization of Calmodulin and Calmodulin-like Protein Gene Families in Paulownia fortunei and Identification of Their Potential Involvement in Paulownia Witches’ Broom

As significant Ca2+ sensors, calmodulin (CaM) and calmodulin-like proteins (CML), have been associated with a variety of environmental conditions in plants. However, whether CaMs/CMLs are related to the stress of phytoplasma infection has not been reported in Paulownia fortunei. In the current study, 5 PfCaMs and 58 PfCMLs were detected through a genome-wide investigation. The number of EF-hand motifs in all PfCaMs/CMLs varied. Bioinformatics analyses, including protein characteristics, conserved domain, gene structure, cis-elements, evolutionary relationship, collinearity, chromosomal location, post-translation modification site, subcellular localization and expression pattern analyses, represented the conservation and divergence of PfCaMs/CMLs. Furthermore, some PfCaMs/CMLs might be involved in plants’ reaction to phytoplasma infection and exogenous calcium therapy, indicating these genes may play a role in abiotic as well as biotic stress responses. In addition, subcellular localization analysis showed that PfCML10 was located in the cell membrane and nucleus. In summary, these findings establish a stronger platform for their subsequent functional investigation in trees and further characterize their roles in Paulownia witches’ broom (PaWB) occurrence.


Introduction
Calcium (Ca 2+ ), a second messenger, is crucial for the growth as well as development of plants, and is involved in responses to inner and outer stimuli [1][2][3][4].Evidence shows that stimulations can rapidly induce an increase in intracellular Ca 2+ concentrations, which then bind to the protein sensors and typically amplify the calcium signals [5,6], changing their behavior or capacity for interaction with subsequent proteins, and thus the signal by causing a change in protein structure [7][8][9].The majority of Ca 2+ -binding protein sensors are EFhand-domain-containing proteins such as calmodulins (CaMs), CaM-like proteins (CMLs), calcineurin B-like proteins (CBLs), and Ca 2+ -dependent protein kinases (CDPKs) [2,10,11].
CaMs are highly preserved in eukaryotes, which typically have four EF-hand motifs, and each EF-hand domain can bind one Ca 2+ ion [12].Previous studies demonstrated that in order to control the growth and development of plants and their ability to react to challenges, CaMs can bind to the target proteins and alter their function [13,14].Among the various Ca 2+ -binding proteins, CMLs are a plant-specific class of Ca 2+ sensors that feature one to six EF-hand motifs and share 16-75% amino acid identity with normal CaMs [12,[15][16][17].Additionally, they are distinct from CDPK and CBL members, and most CaMs/CMLs do not have any other known functional domain, with the exception of CaM7, a transcription factor that can directly control the expression of HY5 in Arabidopsis [18].Different CaMs/CMLs differ in how they bind to and control target proteins, and even minor structural variations can have a big impact on how well they attach to their targets [19].There is proof that CMLs may not be functionally redundant [17].
CaMs/CMLs are notable Ca 2+ -binding proteins sensors in higher plants, including Vitis vinifera, Lotus japonicas, Solanum lycopersicum, Fragaria vesca, Malus domestica, and Triticum aestivum, which can mediate flower development, seed germination and biotic stress tolerance [14,[20][21][22][23][24][25].For instance, the reduction function of AtCML24 and AtCML25 has a significant impact on pollen development, and AtCML24 can work together with ATG4b to influence autophagy growth, whereas CML25 takes part in the Ca 2+ -mediated regulation of K + influx during pollen development [26][27][28].AtCML39 may play a role in controlling seed germination, growth, and early establishment [29,30].A rice-calmodulin-like gene, OsMSR2, can increase resistance to drought and salt [31], while GmCaM4 can strengthen the binding ability of DNA and MYB2, and overexpression of GmCaM4 in soybean increases its resistance to pathogens and salt stress [32].Compared to the wild type, the Arabidopsis cml9 mutant is more sensitive to ABA and exhibits improved salt and drought tolerance [33].Tolerance to drought and ABA-induced stomatal closure are adversely regulated by Arabidopsis CML20 [34].Moreover, many previous studies have also indicated that the expression of CaMs/CMLs strongly affects the plant immune system [13].Reduced salicylic acid levels and disease resistance are the outcomes of CaM induced binding and activation of the calmodulin-binding transcription activator CAMTA3 in Arabidopsis [35].Overexpression of AtCML43 can accelerate the hypersensitive response [36].AtCML9 expression is induced by pathogens (Pseudomonas syringae) and involved in modulating plant defense [33,37].AtCML8 has a beneficial effect on plant immunity and is also strongly and transiently induced by P. syringae [38].Additionally, recent studies on the CaM/CML gene family in grapes demonstrated that the expression of CaMs/CMLs was altered under various abiotic stresses, which suggests the functional diversity in grapevine stress resistance [20].However, up to now, CaMs/CMLs and their relationship to phytoplasma infection had yet to be found and defined in Paulownia fortunei.Therefore, it was vital to investigate the function of Ca 2+ as a second messenger in the pathogenesis of PaWB disease.
P. fortunei is an important deciduous tree of Paulownia (Paulowniaceae), that is widely cultivated for its high adaptability and rapid growth [39].They have high ecological, economic, and medicinal value, which make them popular among the general public [39].However, pathogen phytoplasma can easily infect these trees, leading to the occurrence of Paulownia witches' broom (PaWB), which has a significant negative impact on the tree's productivity and economic benefits.To explore its role in PaWB, here, we identified the PfCaM/CML gene family in the P. fortunei genome and investigated their phylogenetics, gene structure, chromosomal locations, protein motifs, subcellular location, syntenic and promoter cis-acting elements.Moreover, the spatiotemporal expression patterns and expression profiles of healthy P. fortunei seedlings (PF), PaWB phytoplasma-infected P. fortunei seedlings (PFI), and PFI treated with rifampin (Rif) and methyl methane sulfonate (MMS) were also analyzed.Additionally, six PfCMLs with a response to calcium signaling under calcium treatment were examined.To better understand the Ca 2+ -mediated signal transductions in P. fortunei, as well as to aid future gene cloning and functional investigation, it would be helpful to conduct a systematic examination of the entire set of CaMs/CMLs in P. fortunei.

Plant Materials and Growth Conditions
All plant samples utilized in this study were from Henan Agricultural University's Laboratory of Forestry Biotechnology in Zhengzhou, Henan Province, China.The tissuecultured seedlings were cultured in greenhouse at 25 ± 2 • C under a 16 h day/8 h night cycle with an irradiance of 130 µmol•m −2 •s −1 .The apical buds, stem, and leaf from PF and PFI were collected for further study.For calcium treatment, PF and PFI which grew approximately 1 cm root were transfered to 10 mM CaCl 2 and normal medium (control).The apical buds were collected at 3, 6, 9, and 12 h after transformation.All samples were rapidly frozen in liquid nitrogen and kept at −80 • C for RNA extraction.For each treatment, three biological duplicates were established.Nicotiana benthamiana was grown in an infant incubator that had a constant temperature with 16 h/8 h light/dark.

Identification of the CaM and CML Family in P. fortunei
The protein sequences of Arabidopsis thaliana CaM and CML were used to retrieve the TAIR website (http://www.Arabidopsis.org/) in order to determine the CaM and CML proteins in P. fortunei.The detailed Arabidopsis thaliana genes used for identification are listed in Table S1.These sequences served as search criteria for the P. fortunei genome's candidate members using BLASTP (E cut-off value < 1 × 10 −5 ).Coding sequences without ATG were considered partial and defective, which were eliminated.Then these nonredundancy sequences were verified using SMART (http://smart.embl-heidelberg.de/)and NCBI-CDD (http://www.ncbi.nlm.nih.gov/cdd)programs.Sequences possessing only the EF-hand domains were considered to be CaMs/CMLs, similarity greater than 90 to AtCaM2 was considered to be CaMs.

Chromosomal Locations, Motifs and Gene Structure Analyses
Based on P. fortunei genome databases, chromosomal locations and corresponding sequence of PfCaMs/CMLs were retrieved."Gene Location Visualize (advanced)" program of TBtools was used for gene mapping (v.1.09868)[40].The exon-intron structure of PfCaM/CML genes was conducted using the GSDS2.0(Gene structure display server; http://gsds.gao-lab.org/)program with default settings.The MEME suite (http: //meme-suite.org/tools/meme)was used to identify preserved domains.Normal search settings were a maximum of 10 motifs and an ideal motif width between 6 and 50.
For the gene expression verification, total RNA was isolated using the TIANGEN Simple Total RNA kit (Beijing, China).For each sample, 1 µg of total RNA was reversetranscribed into total DNAs using the GeneStar starscript III All-in-one RT, and 2×RealStar Green Fast Mixture with ROX II (Beijing, China) were used to Real-Time qPCR.PfActin as an internal control, and relative expression levels were calculated according to the 2 −∆∆CT (cycle threshold) method.The Table S2 lists the primer sequences utilized in the qRT-PCR analysis.

Vector Construction and Subcellular Localization Analysis
Utilizing the ClonExpress II One Step Cloning Kit from Vazyme, the entire PfCML10 coding sequence was cloned into the pSAK277-eGFP at EcoRI restriction sites for subcellular localization study.The recombinant plasmids 35S::PfCML10-eGFP were transferred into Agrobacterium tumefaciens strain GV3101 for Nicotiana benthamiana leaf transient expression.Then, these transformed N. benthamiana were cultured at 25 • C for 72 h, using confocal laser scanning microscope (Zeiss LSM 710, Göttingen, Germany), and eGFP signals were seen and captured using camera.The primer sequences are listed in Table S2.

Statistical Analysis
The data were analyzed, and all statistical tests were run using SPSS 19.0 (Chicago, IL, USA).Asterisks denote statistical significance at various levels (* p < 0.05 and ** p < 0.01), and values mean SD.

Identification Analysis of CaM/CML Family Genes in P. fortunei
Seven AtCaM and fifty AtCML proteins were used as query sequences to find the CaM and CML genes in the P. fortunei genome.Finally, the P. fortunei genome contained 58 PfCMLs and a total of 5 PfCaMs.Based on the chromosomal positions, the PfCaM/CML genes were called sequentially consecutively from PfCaM1 to PfCaM5 and PfCML1 to PfCML58.All PfCaMs had four EF-hands, their protein length (aa), isoelectric points (pI), theoretical molecular weights (MW) ranged from 149 (PfCaM1/2/5) to 188 (PfCaM4) aa, 4.07 (PfCaM3) to 4.8 (PfCaM4), and 16.85 (PfCaM1/2/5) to 21.78 (PfCaM4) kDa, respectively (Table 1).With one to four EF-hands, PfCMLs differed more from PfCaMs in terms of their features.PfCML proteins ranged in length from 84 (PfCML9/30) to 275 (PfCML51) AA residues, their molecular weights ranged from 9.20 (PfCML30) to 31.07 (PfCML7) KDa, and pI varied from 3.67 (PfCML27) to 8.94 (PfCML51), respectively (Table 1).The other information for the PfCaM/CML proteins, including functional feature sites and methionine percentage, is shown in Table 1.We constructed a phylogenetic tree using the 63 PfCaMs/CMLs, 62 MdCaMs/CMLs, and 57 AtCaMs/CMLs to examine the evolutionary relationship.According to the findings, there are nine groups that can be made up of these 63 PfCaMs/CMLs.(Figure 1), comparable to those identified in A. thaliana, Malus domestica, and Hordeum vulgare [24,41,43].PfCaMs were placed in group V, and PfCMLs were assigned to all of the groups.Group VI had the most PfCaM/CML members (11), followed by groups IX (10) and II (9), whereas groups I (5) and III (4) had few genes.Additionally, the dendrogram showed that there was no clear single homolog CaMs/CMLs among P. fortunei, Malus domestica and A. thaliana, suggesting that CaMs/CMLs from P. fortunei, A. thaliana, and Malus domestica have obvious evolutionary differences.

Conserved Motifs and Gene Structure of PfCaMs/CMLs
Phylogeny, conserved motifs, and gene structures of PfCaMs/CMLs were further analyzed (Figure 2).Family members with close evolutionary ties have substantial sequence similarity and similar motifs in general (Figure 2A).A total of 10 domains were identified and confirmed via the MEME tool.According to the findings, PfCaM/CML motifs were shown to be diverse among PfCMLs but relatively conserved among PfCaMs (Figure 2B);

Conserved Motifs and Gene Structure of PfCaMs/CMLs
Phylogeny, conserved motifs, and gene structures of PfCaMs/CMLs were further analyzed (Figure 2).Family members with close evolutionary ties have substantial sequence similarity and similar motifs in general (Figure 2A).A total of 10 domains were identified and confirmed via the MEME tool.According to the findings, PfCaM/CML motifs were shown to be diverse among PfCMLs but relatively conserved among PfCaMs (Figure 2B); more information about conserved motifs is provided in Figure S1.Furthermore, the existence of the same type of conserved domains may indicate functional similarities in the PfCaM/CML families.In addition, the gene structures were analyzed to examine the structural diversity of PfCaM/CML genes (Figure 2C).Members from groups II, IV, and V were intron-rich, while members from the three groups I, III, and VI were intron-less.Notedly, groups VII, VIII, and IX contained no intron structure.The phylogenetically closely related genes in other plant species shared the same or a very comparable exonintron structure, although PfCaMs/CMLs in various groups exhibited varied exon-intron structural characteristics.

Chromosomal Location and Duplication Analysis
To identify the distribution of PfCaMs/CMLs in P. fortunei chromosomes, we examined their chromosomal position.Of the 63 PfCaMs/CMLs, 53 genes were distributed throughout all of the P. fortunei chromosomes, except chromosome 4 (Figure 3), and the distribution appeared to be uneven.Chromosome 11 contained the most PfCML genes (7 members).Chromosome 13 comes next, which had six PfCML genes.Chromosomes 7 and 6 contained four genes.Chromosomes 3, 5, 12, 15, and 19, each had three PfCMLs.In addition, there is only one PfCML gene that was simultaneously distributed on chromosomes 9, 18, and 20.To understand the evolution of PfCaMs/CMLs, collinearity diagrams among PfCaMs/CMLs were analyzed.The findings demonstrated that the P. fortunei genome exhibited gene duplication in some PfCaMs/CMLs (Figure 3, Table S3).The majority of PfCaMs/CMLs included 1-3 paralogous genes in P. fortunei, while PfCML11 had four paralogous genes in P. fortunei.The Ka/Ks ratios of gene pairs were <1.0, indicating that these PfCaM/CML pairs were subjected to purifying selection following duplication (Table S3).We also carried out a synteny study of the CaM/CML genes in P. fortunei and A. thaliana.Thus, 50 orthologous pairs were found in P. fortunei and A. thaliana (Figure S2).Eight genes (PfCML10/16/20/22/31/47/54/56) had two orthologous genes in A. thaliana, while three genes (PfCML11/12/48) had three orthologous genes in A. thaliana.However, less than 50% identity existed in around one-third of the orthologous pairs, indicating that the function of gene pairs may have been differentiated.According to some significant percentages of identity between PfCaMs/CMLs and AtCaMs/CMLs, PfCaM/CML protein functions and sequences may be substantially conserved.

Diversified Expression Patterns of PfCaMs/CMLs
We employed transcriptome sequencing data and thoroughly examined these expression levels to investigate the expression patterns of PfCaMs/CMLs in sapling trees.In total, 57 PfCaMs/CMLs are expressed in trunk phloem, leaf, and trunk cambium (Figure 4A).Compared with other genes, five genes (PfCaM4, PfCML8/17/32/36) were highly expressed in trunk phloem.In terms of the leaf transcriptome, PfCML27 had the greatest expression level, next to PfCML8/4/46/16.Particularly, the expression level of PfCML27 was almost 2-fold higher than that of PfCML8.Six genes (PfCML48/20/8/17/49/36) appeared to have a relatively high expression in trunk cambium, which indicated that their function may relate to secondary growth.Notably, PfCML8 was highly expressed in all three tissues, which implied that PfCML8 is positively involved in developmental processes of trunk phloem, leaf, and trunk cambium.The discovery suggested that PfCaMs/CMLs may be engaged in a variety of biological processes throughout the growth and development of P. fortunei.ana.Thus, 50 orthologous pairs were found in P. fortunei and A. thaliana (Figure S2).Eight genes (PfCML10/16/20/22/31/47/54/56) had two orthologous genes in A. thaliana, while three genes (PfCML11/12/48) had three orthologous genes in A. thaliana.However, less than 50% identity existed in around one-third of the orthologous pairs, indicating that the function of gene pairs may have been differentiated.According to some significant percentages of identity between PfCaMs/CMLs and AtCaMs/CMLs, PfCaM/CML protein functions and sequences may be substantially conserved.Additionally, expression analysis of PfCaM/CML genes that responded to PaWB phytoplasmas infection was performed.Compared with PF, expression of 35 PfCaMs/CMLs was up-regulated, and 17 PfCaMs/CMLs were down-regulated in PFI.Among them, 15 genes (PfCML10/13/15/20/22/26/27/28/29/31/33/34/43/44/47) were up-regulated significantly in the PFI (fold change > 2), while 4 genes (PfCML38/42/49/58) were observably downregulated (fold change > 2) (Figure 4B,C).After MMS or Rif treatment, the relative expression levels of PfCaMs/CMLs were changed, which met the trend of phytoplasma concentration, and qRT-PCR results further confirmed the expression patterns (Figure 4D).Based on these, our data suggested that PfCaMs/CMLs may be associated with PaWB occurrence.

Analysis of the Cis-Elements in the Promoter Region of the PfCaMs/CMLs
The coordinated regulation of stress-induced gene expression in stressful situations was carried out via transcriptional networks.We assumed that PfCaMs/CMLs would be targets of stress-regulated transcription factors and may be downstream of the transcriptional network regulating stress resistance.In light of this, we investigated the cis-elements associated with plant growth, hormones, and stress responses in the promoter region of the PfCaMs/CMLs.As shown in Figure 5, the mainly cis-acting elements of PfCaMs/CMLs were grouped into four types.Cis-acting elements are associated with responses to growth, such as meristem expression (CAT-box) and endosperm expression (GCN4-motif); phyto-hormones, such as abscisic acid (ABRE), salicylic acid (TCA-element), gibberellin (p-box), methyl jasmonate (CGTCA-motif), ethylene (ERE) and auxin (TGA-element and AUXRRcore); adversity, such as anoxia stress (ARE), low temperature (LTR), drought (MBS); defense, and stress-related elements (TC-rich repeats/WUN-motif) distributed in the promoter regions of the PfCaMs/CMLs.The fact that PfCaMs/CMLs had the same or distinct cis-acting areas suggests that these genes can occasionally be regulated simultaneously in response to stress or a particular external stimulus.Therefore, our data indicated that PfCaMs/CMLs should respond to P. fortunei stress stimulus.

Diversified Expression Patterns of PfCaMs/CMLs
We employed transcriptome sequencing data and thoroughly examined these expression levels to investigate the expression patterns of PfCaMs/CMLs in sapling trees.In total, 57 PfCaMs/CMLs are expressed in trunk phloem, leaf, and trunk cambium (Figure 4A).Compared with other genes, five genes (PfCaM4, PfCML8/17/32/36) were highly expressed in trunk phloem.In terms of the leaf transcriptome, PfCML27 had the greatest expression level, next to PfCML8/4/46/16.Particularly, the expression level of PfCML27 was almost 2-fold higher than that of PfCML8.Six genes (PfCML48/20/8/17/49/36) appeared to have a relatively high expression in trunk cambium, which indicated that their function may relate to secondary growth.Notably, PfCML8 was highly expressed in all three tissues, which implied that PfCML8 is positively involved in developmental processes of trunk phloem, leaf, and trunk cambium.The discovery suggested that PfCaMs/CMLs may be engaged in a variety of biological processes throughout the growth and development of P. fortunei.

Post-Translation Modification and Subcellular Localization Prediction
To uncover the relative regulatory mechanisms of PfCaMs/CMLs, we predic translational modification (PTM) sites, such as methylation, acetylation, ubiqui and phosphorylation (Figure 6).We found up to 22 potential phosphorylation PfCML52, with at least 1 putative phosphorylation site found in PfCaMs/CMLs (T Moreover, seven and six CaMs/CMLs were found putatively at methylation s ubiquitization sites.More than 50% of the family members contained acetylation s

Post-Translation Modification and Subcellular Localization Prediction
To uncover the relative regulatory mechanisms of PfCaMs/CMLs, we predicted posttranslational modification (PTM) sites, such as methylation, acetylation, ubiquitization, and phosphorylation (Figure 6).We found up to 22 potential phosphorylation sites in PfCML52, with at least 1 putative phosphorylation site found in PfCaMs/CMLs (Table S4).Moreover, seven and six CaMs/CMLs were found putatively at methylation sites and ubiquitization sites.More than 50% of the family members contained acetylation sites, and three members had up to six acetylation sites (Figure 6, Table S4).These findings laid the groundwork for further research into how the PTM affects the structure and roles of PfCaM/CML proteins.
x FOR PEER REVIEW 13 of 18 The potential subcellular localization of the PfCaM/CML proteins was predicted, as shown in Table S5.Except for PfCML23 and PfCML37, which were located outside the cell, most PfCaMs/CMLs were located at extranuclear sites, such as cytoplasm, vacuole, and chloroplasts.To further substantiate the subcellular localization of the PfCaM/CML proteins, we produced a temporary expression construct (35S::PfCML10-GFP).The PfCML10-GFP fluorescence signal was found in the cytoplasm and nucleus, while the GFP-free control signal was present throughout the entire population of cells (Figure 7).The potential subcellular localization of the PfCaM/CML proteins was predicted, as shown in Table S5.Except for PfCML23 and PfCML37, which were located outside the cell, most PfCaMs/CMLs were located at extranuclear sites, such as cytoplasm, vacuole, and chloroplasts.To further substantiate the subcellular localization of the PfCaM/CML proteins, we produced a temporary expression construct (35S::PfCML10-GFP).The PfCML10-GFP fluorescence signal was found in the cytoplasm and nucleus, while the GFP-free control signal was present throughout the entire population of cells (Figure 7).

Expression Analysis of PfCMLs in Response to Calcium Signaling
To explore the potential regulation of PfCaM/CML genes in response to calcium signaling, the expression levels of six candidate PfCMLs (PfCML10/15/28/34/38/57) that were selected according to the RNA-seq result were analyzed under a high concentration of calcium.Six PfCMLs' expression levels changed in response to calcium, and there were differences in the calcium response time between PF and PFI (Figure 8).Among them, the expression levels of four genes (PfCML10/15/34/38) and PfCML28 increased within 6 h and 9 h after the calcium treatment in PFI, respectively.Nevertheless, three genes (PfCML15/34/57) increased to the highest expression level at 9 h in PF, and PfCML57 revealed the greatest expression levels, showing 50-times greater expression at 9 h in the CaCl 2 treatment than control.These results suggested that the expression patterns were modified by increased Ca 2+ concentration., and six PfCMLs should be variously involved in PaWB occurrence.
The potential subcellular localization of the PfCaM/CML proteins was predicted, shown in Table S5.Except for PfCML23 and PfCML37, which were located outside t cell, most PfCaMs/CMLs were located at extranuclear sites, such as cytoplasm, vacuo and chloroplasts.To further substantiate the subcellular localization of the PfCaM/CM proteins, we produced a temporary expression construct (35S::PfCML10-GFP).T PfCML10-GFP fluorescence signal was found in the cytoplasm and nucleus, while t GFP-free control signal was present throughout the entire population of cells (

Expression Analysis of PfCMLs in Response to Calcium Signaling
To explore the potential regulation of PfCaM/CML genes in response to calcium signaling, the expression levels of six candidate PfCMLs (PfCML10/15/28/34/38/57) that were selected according to the RNA-seq result were analyzed under a high concentration of calcium.Six PfCMLs' expression levels changed in response to calcium, and there were differences in the calcium response time between PF and PFI (Figure 8).Among them, the expression levels of four genes (PfCML10/15/34/38) and PfCML28 increased within 6 h and 9 h after the calcium treatment in PFI, respectively.Nevertheless, three genes (PfCML15/34/57) increased to the highest expression level at 9 h in PF, and PfCML57 revealed the greatest expression levels, showing 50-times greater expression at 9 h in the CaCl2 treatment than control.These results suggested that the expression patterns were modified by increased Ca 2+ concentration., and six PfCMLs should be variously involved in PaWB occurrence.

Discussion
Ca 2+ is a crucial second messenger that performs a key role in plants [6].A plethora of Ca 2+ binding proteins that act as Ca 2+ sensors decipher complicated Ca 2+ signals [44].Ca 2+ binding proteins react to the fluctuating intracellular Ca 2+ level and boost later signaling, triggering a physiological reaction to the signal.Among Ca 2+ sensors, CaMs and CMLs are the most universally conserved major calcium sensor proteins in plants.However, no report on participation of plant CaMs/CMLs in P. fortunei has been found.Additionally, most CaM/CML investigations have mostly focused on herbaceous plants, particularly A. thaliana and Oryza sativa [26][27][28][29][30][31]33,38], and have ignored this family in angiosperm deciduous tree species.We completely characterized the CaM/CML families in P. fortunei in the current work, including the phylogeny, gene structures, conserved motifs, gene duplications, synteny analysis, expression patterns, cis-acting elements, subcellular localization, and PTM sites.Additionally, six PfCMLs were proposed to be involved in PaWB occurrence.These findings offer a whole comprehension of the CaM/CML gene families in deciduous tree species within the ngiosperm and establish important groundwork for subsequent research on how they affect tree growth and development.
According to our investigation, we discovered 58 CML genes in P. fortunei (Table 1),

Discussion
Ca 2+ is a crucial second messenger that performs a key role in plants [6].A plethora of Ca 2+ binding proteins that act as Ca 2+ sensors decipher complicated Ca 2+ signals [44].Ca 2+ binding proteins react to the fluctuating intracellular Ca 2+ level and boost later signaling, triggering a physiological reaction to the signal.Among Ca 2+ sensors, CaMs and CMLs are the most universally conserved major calcium sensor proteins in plants.However, no report on participation of plant CaMs/CMLs in P. fortunei has been found.Additionally, most CaM/CML investigations have mostly focused on herbaceous plants, particularly A. thaliana and Oryza sativa [26][27][28][29][30][31]33,38], and have ignored this family in angiosperm deciduous tree species.We completely characterized the CaM/CML families in P. fortunei in the current work, including the phylogeny, gene structures, conserved motifs, gene duplications, synteny analysis, expression patterns, cis-acting elements, subcellular localization, and PTM sites.Additionally, six PfCMLs were proposed to be involved in PaWB occurrence.These findings offer a whole comprehension of the CaM/CML gene families in deciduous tree species within the ngiosperm and establish important groundwork for subsequent research on how they affect tree growth and development.
According to our investigation, we discovered 58 CML genes in P. fortunei (Table 1), somewhat more than those in A. thaliana (50 CML-encoding genes), even if certain clusters of gene duplication were visible.The total amount of members of the PfCaM/CML gene families falls short of what we anticipated when we took into account the ratio of 3.0 putative P. fortunei genome size for A. thaliana.At least two substantial genome-wide duplications were performed on the P. fortunei genome [39], which could lead to numerous imperfect and incomplete coding sequences.Moreover, prior research has shown that numerous P. fortunei chromosomal pairs have the same fusion pattern and that substantial gene loss happened in an asymmetrical and reciprocal manner [39].As a result, many fewer PfCaM/CML genes than we anticipated were found in P. fortunei.Nevertheless, approximately 84% (53 of 63) of the PfCaM/CML genes were linked to duplications, suggesting these genes were functionally redundant.According to our analysis of selective pressure, purifying selection was the dominant force during the expansion of the CaM/CML gene families in P. fortunei, implying that these duplicate genes may still maintain their original functions.
The majority of PfCaMs/CMLs from the same group shared comparable gene structures and conserved domains (Figure 2), as well as the subcellular localization prediction (Table S5).PfCaMs/CMLs were classified into nine groups, consistent with A. thaliana [41].Moreover, the intron-exon structural analysis revealed that the majority of PfCML genes lack an intron (Figure 2).This intron-exon structure may have already existed in common ancestor genes [43,45].Despite not contributing to protein sequences, introns' relative placements provide some hints for forecasting how genes and the proteins they encode change, which helps the gene family become more diverse [46,47].Therefore, our data revealed that CaM/CML families shared a conserved gene structure, supporting the classification and evolutionary relationships in P. fortunei.This evidence also suggested plant CaMs/CMLs belonging to the same subgroup might play roles that are comparatively preserved.However, the protein composition and quantity of EF-hand domains in CaM/CML members, which are associated with PTM sites and Ca 2+ -binding characteristics, determine how well they operate.In actuality, different plant species have different numbers of EF-hand patterns.Two to six EF-hand motifs are generally present in A. thaliana CML proteins [41], but PfCMLs contained one to four preserved EF-hand motifs.In addition, most PTM sites were also different in the same group.Combined with collinearity analysis, these data suggested that the function of partial gene pairs may have been differentiated.
The patterns of gene expression and function were closely connected.According to RNA-seq and qRT-PCR analyses, the genes in the PfCaM/CML families have a variety of expression patterns (Figure 4).Previous studies indicated that MMS and Rif were confirmed to significantly inhibit PaWB in two different ways, and PFI treated with certain concentrations of MMS or Rif could transform healthy morphology; PaWB phytoplasma was undetectable in apical buds after 20 days of treatment [39].Based on RNA-Seq of PfCaMs/CMLs, almost all PfCaMs/CMLs displayed higher transcript levels in buds of PFI, except for PfCML38, for which the expression level was down-regulated in PFI.Even so, these gene expression patterns were also positively correlated with detectable levels of PaWB phytoplasma.These findings suggest that PfCaM/CML genes should interact with or function in the PaWB.For this purpose, the stress-related regulatory elements in the promoter regions were examined.Numerous cis-elements related to hormones and stress reaction were present in almost all members (Figure 5), which have been associated with biotic stress.Furthermore, the PfCML10 was located in the cell membrane and nucleus within N. benthamiana cells, similar to that in Arabidopsis AtCML24, modulating the actin cytoskeleton; then, it regulates pollen tube growth and affects autophagic progression [27,28].
The involvement of PfCML10 in the autophagic process during phytoplasma infection is assumed, such as degradation of chlorophyll and chlorophyll-protein complexes.Presumably, these proteins may be signaling proteins, which possibly contribute to the occurrence of PaWB.Co-immunoprecipitation, yeast two-hybrid, and proteomics techniques could be used to identify the target proteins of PfCML10.
The variation in the expression of PfCaMs/CMLs in response to calcium treatment may be relevant in examining its functional significance in plant growth and development, considering that Ca 2+ plays a significant part in plant stress response.We analyzed the expression of six PfCMLs in apical buds via qRT-PCR.In PFI, these six selected PfCMLs were up-regulated after CaCl 2 treatment in comparison with PF, indicating that Ca 2+ -permeable channels may be activated by high concentrations of exogenous calcium ion, increasing the intracellular Ca 2+ concentration, thus generating signal transduction.
By reducing BRI1-mediated reactions, AtCML8 could positively control defense systems, enabling full PTI establishment [48].The expression level of PfCML15/28, which is closely related to AtCML8 in evolution and high homology (43.7% and 64.5%), were significantly up-regulated in PFI.In this scenario, the significant up-regulation of PfCML15/28 might be connected to the efficient suppression of BRI1 in P. fortunei.Ma et al. [49] found AtCML24 favorably regulates defense resistance to the bacterial pathogen P. syringae.As a collinear gene of AtCML24, PfCML10 shares more than 50% identity with AtCML24 and, in response to phytoplasma infection, these indicated that it might be involved in regulating resistance to pathogens.However, in view of the short treatment time of calcium ions, we did not notice a change in the growth and development of PFI.Whether the PaWB occurrence is mediated by these PfCaMs/CMLs needs to be further investigated in the transgenic P. fortunei.It will be possible to learn more about the genetic functions and underlying molecular mechanisms of these P. fortunei transgenic plants through subsequent efforts to obtain mutants and overexpression lines.

Conclusions
We found and described 5 PfCaM and 58 PfCML proteins with functional EF-hand domains in the P. fortunei genome.According to expression studies, PfCaM/CML members exhibit a variety of expression patterns in various tissues and stress responses, and their expression levels follow the trend of phytoplasma concentration.Meanwhile, partial PfCaM/CML members' response to calcium stimulus, and six PfCMLs were selected that were suggested to be possibly differently involved in PaWB occurrence.To date, it is yet unknown how PfCaMs and CMLs work in P. fortunei.Thus, in order to clearly identify the functions of these PfCaM/CML genes in the occurrence of PaWB, we next examined the phenotypes of the CRISPR/Cas9-targeted PfCaMs/CMLs knockout P. fortunei trees.These results pave the way for further investigations into the role of PfCaMs/CMLs in calcium signaling and stress responses during PaWB phytoplasma infection.

18 Figure 1 .
Figure 1.The phylogenetic tree of CaM/CML from Paulownia fortunei, Arabidopsis thaliana, and Malus domestica.Phylogenetic distance calculate by MEGA X program and the Maximum Likelihood (ML) method, using bootstrap values of 1000.Based on the protein sequence similarities between Paulownia fortunei, Arabidopsis thaliana, and Malus domestica, the members were divided into 9 groups (I-IX).

Figure 1 .
Figure 1.The phylogenetic tree of CaM/CML from Paulownia fortunei, Arabidopsis thaliana, and Malus domestica.Phylogenetic distance calculate by MEGA X program and the Maximum Likelihood (ML) method, using bootstrap values of 1000.Based on the protein sequence similarities between Paulownia fortunei, Arabidopsis thaliana, and Malus domestica, the members were divided into 9 groups (I-IX).

Figure 2 .
Figure 2. Phylogenetic groups, motif compositions and gene structures of PfCaMs/CMLs.(A) Phylogenetic classification of PfCaM/CML proteins; (B) MEME analysis revealed a schematic representation of the conserved motifs among the PfCaM/CML proteins.Each color stood for a distinct motif; (C) Gene structures of PfCaMs/CMLs.The CDS, UTR and Intron were appeared by green boxes, yellow boxes, and grey line, respectively.

Figure 2 .
Figure 2. Phylogenetic groups, motif compositions and gene structures of PfCaMs/CMLs.(A) Phylogenetic classification of PfCaM/CML proteins; (B) MEME analysis revealed a schematic representation of the conserved motifs among the PfCaM/CML proteins.Each color stood for a distinct motif; (C) Gene structures of PfCaMs/CMLs.The CDS, UTR and Intron were appeared by green boxes, yellow boxes, and grey line, respectively.

Figure 3 .
Figure 3. Chromosomal location and collinearity analysis of PfCaMs/CMLs in the Paulownia fortunei genome.The relative positions of the 53 PfCaMs/CMLs marked on chromosomes (Chr01-20).Red line represented the collinearity gene pair in the P. fortunei genome.The coloured boxes represented gene density, the more orange the colour, and the higher the density.

Figure 5 .
Figure 5. Representatives of the major cis-regulatory elements identified from PfCaMs/C motors.PlantCARE was used to determine the cis-acting elements of PfCaMs/CMLs.The elements were presented in four types (Growth, Defense, Stress, and Plant hormones).The represented amount of cis-acting elements, and the blank space meant the gene didn't hav ulatory element.The different colours also represented the differences with the number of

Figure 5 .
Figure 5. Representatives of the major cis-regulatory elements identified from PfCaMs/CMLs promotors.PlantCARE was used to determine the cis-acting elements of PfCaMs/CMLs.The cis-acting elements were presented in four types (Growth, Defense, Stress, and Plant hormones).The numbers represented amount of cis-acting elements, and the blank space meant the gene didn't have the regulatory element.The different colours also represented the differences with the number of elements.

Figure 6 .
Figure 6.Schematic representation of four potential post-translation modification of Paulownia fortunei CaM and CML proteins.Methylation, Acetylation, Ubiquitination, and Phosphorylation sites were predicted on the CUCKOO Workgruop and NetPhos 3.1, respectively.For each CaM and CML proteins, the putative Methylation (in blue), Acetylation (in orange), Ubiquitination (in grey), and Phosphorylation (in yellow) sites were shown.The length of colored bars corresponded to the number of expected locations.

Figure 6 .
Figure 6.Schematic representation of four potential post-translation modification of Paulownia fortunei CaM and CML proteins.Methylation, Acetylation, Ubiquitination, and Phosphorylation sites were predicted on the CUCKOO Workgruop and NetPhos 3.1, respectively.For each CaM and CML proteins, the putative Methylation (in blue), Acetylation (in orange), Ubiquitination (in grey), and Phosphorylation (in yellow) sites were shown.The length of colored bars corresponded to the number of expected locations.

Figure 7 .
Figure 7. Subcellular localization of PfCML10.35S:PfCML10-GFP and 35S:GFP constructs were dividually injected into the epidermal cells of N. benthamiana.The transient expression of PfCML1 GFP was observed and captured by a confocal laser scanning microscope.Autofluorescence of t GFP-free was also observed and photographed.mCherry-PIP2;1 as PM marker.Scale bars were µm.

Figure 7 .
Figure 7. Subcellular localization of PfCML10.35S:PfCML10-GFP and 35S:GFP constructs were individually injected into the epidermal cells of N. benthamiana.The transient expression of PfCML10-GFP was observed and captured by a confocal laser scanning microscope.Autofluorescence of the GFP-free was also observed and photographed.mCherry-PIP2;1 as PM marker.Scale bars were 20 µm.

Figure 8 .
Figure 8. Expression patterns of 6 PfCMLs under calcium treatment.Apical buds were collected at 3, 6, 9, and 12 h after treatment.The red and blue lines represented PF and PFI treated with 10 mM CaCl2, respectively, and black line represented untreated PF as control.Error bars represented SD values from three biological repeats.

Figure 8 .
Figure 8. Expression patterns of 6 PfCMLs under calcium treatment.Apical buds were collected at 3, 6, 9, and 12 h after treatment.The red and blue lines represented PF and PFI treated with 10 mM CaCl 2 , respectively, and black line represented untreated PF as control.Error bars represented SD values from three biological repeats.

:
Prediction analysis of subcellular localization of the 63 PfCaM/CMLs.Author Contributions: G.F. conceived and designed the experiments; L.L. and N.G.performed the experiments and wrote the paper; Y.C. and X.Z.contributed reagents and analyzed the data.All authors have read and agreed to the published version of the manuscript.Funding: This research was funded by the Academic Scientist Fund for Zhongyuan Scholars of Henan Province (grant 2018 [99]), the Science and Technology Promoting Forestry Project of Henan Province (grant YLK202210), the Project of Central Plains Science and Technology Innovation Leading Talents of Henan Province (224200510010) and Zhongyuan Scholars Workstation Project (234400510030).
a Presence of protein length.b Presence of molecular weight.c Isoelectric point prediction by ExpasyProtParam server.d Presence of a cysteine equivalent to Cys27 of typical plant CaMs at residue 7(-Y) of the first EF-hand.e Presence of a lysine equivalent to Lys116 of typical plant CaMs.f Percentage of methionine residues in the deduced amino acid sequence.