Identification and expression of CAMTA genes in Populus trichocarpa under biotic and abiotic stress

The calmodulin-binding transcription activators (CAMTAs) transcription factor family plays an important role in normal plant growth and development, as well as in biotic and abiotic stress resistance. In this study, we identified seven CAMTA genes across the whole genome of Populus trichocarpa and analyzed the expression patterns of PtCAMTAs in the root and leaf tissues. Promoter cis-element analysis indicated that most CAMTA genes contained stress- or phytohormone-related cis-elements. Quantitative real-time reverse transcription-PCR (qRT-PCR) indicated indicated that PtCAMTAs were induced by mannitol, NaCl, cold stress, pathogenic infection with A. alternata, and phytohormone treatments with abscisic acid, salicylic acid, and methyl jasmonate. We analyzed the expression of homologous genes between P. trichocarpa and P. ussuriensis and alternative splicing forms of PtCAMTA genes under cold stress. We also performed a network interaction analysis for PtCAMTA proteins to predict their interactions and associations. The results of the present study serve as a basis for future functional studies on the Populus CAMTA family.

in the roots. The majority of MtCAMTA genes have been proven to respond to hormones such as SA, MeJA, and ABA, suggesting that CAMTA-mediated abiotic and biotic stress tolerance may exist in different plant species.
Populus trichocarpa is widely used in functional forest tree genomics studies and has significant commercial and ecological value 13 . Populus trichocarpa is threatened by a multitude of environmental stresses and biotic stresses such as drought and fungal disease during growth and development. Although CAMTA genes have been studied in some plants, research into the CAMTA gene family in Populus is limited. We identified seven CAMTA genes in P. trichocarpa and analyzed their phylogenetic relationships, chromosomal locations, gene duplication events, and gene structures. The expression mode of PtCAMTAs under abiotic stress (mannitol, NaCl, 4 °C), biotic stress (Alternaria alternate infection), and phytohormone treatment, including abscisic acid (ABA), salicylic acid (SA), and methyl jasmonate (MeJA), were analyzed using quantitative real-time RT-PCR (qRT-PCR). We also analyzed the expression of homologous PtCAMTA genes between P. trichocarpa and P. ussuriensis and alternative splicing forms of PtCAMTA genes under cold stress. The results may support further functional gene research through the study of these candidate CAMTA genes in response to abiotic and biotic stress.

Results
Identification, phylogenetic relationships, gene structure, conserved domain and alternative splicing analyses of PtCAMTA genes. We identified a total of seven CAMTA genes in P. trichocarpa, which encoded proteins that varied in length from 907 to 1,116 amino acids (aa), with an average length of 1,000 aa. The CAMTA protein sequences showed large variations in isoelectric point (pI) values (ranging from 5.45 to 8.27) and molecular weight (ranging from 102.195 kDa to 126.102 kDa). The location of the PtCAMTA proteins was predicted to be the cell nucleus using Wolf PSORT (Table 1).
To examine the phylogenetic relationships among the CAMTA domain proteins in P. trichocarpa, an unrooted phylogenetic tree was constructed from the full-length CAMTA sequence alignments. We classified seven CAMTA genes into three subgroups according to their homology (Fig. 1A). A comparison of the exon/intron organization of the coding sequences of individual PtCAMTA genes showed a similar exon-intron structural pattern, indicating a necessary conservation of the genomic structure of PtCAMTA genes (Fig. 1B). The conserved domains of the CAMTAs, involving a CG-1 DNA binding domain, a TIG domain, ankyrin repeats, and one or two copies of IQ motifs were predicted in the PtCAMTA proteins (Fig. 1C).
We analyzed the PtCAMTAs primary cDNAs and the genomic DNA sequences. The results showed that PtCAMTA1-7 produced splice variants. The number of unigenes corresponding to splice variants was two in PtCAMTA1, 4 Figure S1). Chromosomal location and duplication of PtCAMTA genes. To verify the relationship between genetic divergence and gene duplication, we identified the chromosomal locations of PtCAMTA genes. As shown in Supplementary Figure S2, PtCAMTA genes were characterized by an obvious feature whereby all the genes were distributed on chromosomes I, III, V, VII, VIII, and X, respectively. There were two PtCAMTA genes (PtCAMTA3 and PtCAMTA6) on chromosome X. The rest of the PtCAMTA genes were detected on each of chromosomes I, III, V, VII, and VIII. In order to confirm the relationship between the CAMTA genes and potential segmental duplications, we used the duplicated blocks set up in a previous study. The distribution of the duplicate blocks related to CAMTA genes is illustrated in Supplementary Figure S2. We discovered that two of the seven PtCAMTA (PtCAMTA1 and PtCAMTA7) genes were present in both duplicated regions and were thus prioritized, as the others were only present in one of the blocks.  Table S1).
Promoter cis-element analysis. We identified putative cis-acting regulatory DNA elements via the promoter sequences of PtCAMTA genes (2,000 bp upstream of the translation start site) based on the Phytozome version 12.1 database. The CAMTA gene family promoter sequences demonstrated that several cis-elements were related to biotic and abiotic stress responsiveness (Fig. 2). In total, 10 types of abiotic stress elements were identified. Nearly all the PtCAMTA genes had MBS elements and five had W-box in their promoters, which showed that the MYB binding site is involved in drought inducibility. Three of the PtCAMTA genes (PtCAMTA4, PtCAMTA5, and PtCAMTA6) possessed AREB-responsive elements (ABREs). Nearly all the PtCAMTA genes possessed MeJA-responsive elements (CGTCA-motif, G-Box, TGACG-motif), except PtCAMTA6. Four of the PtCAMTA genes (PtCAMTA1, PtCAMTA3, PtCAMTA5, and PtCAMTA6) possessed SA-responsive elements (TCA-element; Supplementary Table S2).

Tissue-specific expression profile. We observed specific expression patterns in the different tissues of
PtCAMTA genes from the Affymetrix (GSE6422) microarray data in PopGeneIE version 3.0 (Supplementary Figure S3). The microarray data showed high expression levels of PtCAMTA genes in the roots, with PtCAMTA3 exhibiting particularly high expression levels among all the PtCAMTA genes. In the leaves, only PtCAMTA1 was highly expressed in the mature leaves. We found that PtCAMTA2 and PtCAMTA3 displayed low expression levels in the mature leaves and high expression levels in the roots, simultaneously. Notably, the expression level of all PtCAMTAs in the young leaves was very low.

Expression analysis of PtCAMTA genes under mannitol and NaCl stress. In order to understand
how PtCAMTA genes react under osmotic stress, we analyzed the expression of the PtCAMTAs in the roots and leaves under treatment with 200 mM mannitol and 150 mM NaCl for 3 h, 6 h, 12 h, 24 h, and 7d, respectively. Genes that were up or downregulated by more than 2.0-fold were considered significantly differentially expressed 16 . Under mannitol stress, all the genes were upregulated in both the roots and leaves in the short-term treatments (3 h, 6 h, 12 h, and 24 h). Changes in the expression of PtCAMTA3, 6, and 7 were not obvious in the roots and leaves in the long term (7 d). Notably, PtCAMTA1, 4, and 5 were significantly upregulated (>5.0-fold relative to the control) in the roots. PtCAMTA1-6 was significantly upregulated in the leaves. Under NaCl stress, all the PtCAMTA genes were suppressed in the roots at all time points. PtCAMTA genes were upregulated in the leaves under short-term stress and showed no change under long-term stress (Fig. 3).  Following infection, two CAMTA genes (PtCAMTA1 and 2) were induced, four CAMTA genes (PtCAMTA3, 4, 6, and 7) were suppressed, and PtCAMTA5 showed no change. PtCAMTA1, 2, and 4 were induced at 24 h after infection (Fig. 4).

Expression levels of PtCAMTA genes in response to phytohormone stimuli. To understand how
PtCAMTA genes participate in stress-related hormone responses, we analyzed the expression of PtCAMTA genes under 200 μM ABA, 100 μM SA, and 100 μM MeJA in the leaves and roots for 3 h, 6 h, 12 h, 24 h, and 7d using qRT-PCR. Under ABA stress, three genes were upregulated before 6 h stress, and four genes were downregulated in the roots, while in the leaves all the genes were upregulated at all time points. Five CAMTA genes (PtCAMTA1, 2, 3, 6, and 7) were significantly upregulated in the short term. Under SA stress, all genes were significantly downregulated in the roots. Four genes (PtCAMTA1, 5, 6, and 7) were upregulated in the leaves, whereas PtCAMTA2, 3, and 4 were downregulated. Under MeJA stress, four genes (PtCAMTA1, 5, 6, and 7) were upregulated, two (PtCAMTA3 and 4) were downregulated, and PtCAMTA2 exhibited no change in the roots. With respect to the leaves, three genes (PtCAMTA1, 4, and 7) were upregulated and three (PtCAMTA2, 3, and 5) were downregulated. PtCAMTA1 and 7 were induced in both the roots and leaves. Most CAMTA genes were downregulated in the roots under ABA, SA, and MeJA treatments (Fig. 5).
Expression analysis of PtCAMTA alternative splicing forms under cold stress. Alternative splicing, as a post-transcriptional mechanism of precursor-mRNA, plays a significant role in transcriptome and proteome diversity as well as transcript and protein abundance 17,18 . As the CAMTA gene family is mainly involved in the cold stress response, we studied the expression of several alternative splicing forms of PtCAMTA genes under cold stress in detail. The analysis of qRT-PCR using splice variant-specific primers showed that specific amplicons could be obtained for PtCAMTA1. 2  In conjunction, these results suggest that PtCAMTAs are important in cold-regulated gene expression (Fig. 6).  Homologous PtCAMTA genes in two Populus varieties under cold stress. In order to gain further insight into the relationship between PtCAMTA gene expression and cold stress resistance, we compared the expression patterns of P. trichocarpa and P. ussuriensis (Sect. Tacamahaca: the same as P. trichocarpa. P. ussuriensis Kom), which are mainly distributed in the cold temperature zone of Northeast China and the far east region of Russia. Populus trichocarpa is a cold-tolerant species that can survive an annual average temperature of −3.8 °C as well as an extremely low temperature environment (−46.9 °C). Exhibiting strong resistance to cold, it constitutes the ideal candidate for studying the molecular mechanisms of woody plants. The results of the qRT-PCR revealed that the CAMTAs were differentially expressed in P. trichocarpa and P. ussuriensis under cold stress. For P. ussuriensis, the expression of most PuCAMTA genes was downregulated in the roots (except PuCAMTA4) and the leaves (except PuCAMTA1). The same trends were observed in P. trichocarpa, in that most of the PtCAMTAs were downregulated in the roots under cold treatment. However, most of the PtCAMTA genes were upregulated in the leaves in P. trichocarpa in the short term (Fig. 7).
Protein interaction network analysis. In order to predict the interactions and associations of all the PtCAMTA proteins, we performed a network interaction analysis based on Arabidopsis proteins using STRING software, with the confidence value set at 0.5. An Arabidopsis CAMTA proteins network was created and 38 interactive proteins (confidence value = 0.5) were identified with the STRING database 19 . Then, the homologs of these 38 proteins in Populus were identified using Phytozome version 12.1. A total of eight unique proteins were predicted as potential interactors of PtCAMTA4, 6, and 7. The partners of PtCAMTA4 and 6 were predicted to be the Ca 2+ / CaM-regulated protein kinases CIPK5 and CIPK21. Moreover, PtCAMTA4 and PtCAMTA6 were identified as homologous proteins, which were predicted by STRING to directly interact or function in the same pathway (Fig. 8).
Studies have shown that CAMTA regulates the expression of the target gene by directly binding to the CGCG cis-elements of their promoter 20 . Rahman et al. have reported AtCAMTA3 protein can interacte with 10 proteins which contain at least one CGCG cis-element in their promoters in Arabidopsis 21 . We found 16 homologous genes of the 10 AtCAMTA3 target genes in P. trichocarpa using the Phytozome version 12.1 database, six of which were homologous genes. However, we discovered that only four homologous genes possessed CGCG cis-elements in their promoters (2,000 bp upstream of the translation start site) of the AtCAMTA3 target genes in P. trichocarpa. (Supplementary Table S3)

Discussion
In this report, seven P. trichocarpa CAMTA family gene members were identified, each of which contained conserved domains related to CAMTA proteins. Previous studies have suggested that the Populus genome experienced at least three rounds of genome-wide duplication, including multiple segmental duplication, tandem duplication, and transposition events such as retroposition and replicative transposition. Tandem and segmental duplication plays an important role in genomic expansions and realignments 22,23 . In our study, all the PtCAMTA genes were located in the duplicated regions. We discovered that PtCAMTA1 and PtCAMTA7 belonged to the same branch of the phylogenetic tree and were present in the homologous regions of chromosomes I and III, respectively. This finding might be explained by the homology of the two genes. As found in the CAMTAs that have been characterized in other species, all seven of the PtCAMTAs contained conserved domains. In one of the subgroups that contained CaMBD, we identified variation in PtCAMTA2 and PtCAMTA3 that was not present in the other genes. Based on the phylogenetic tree, we discovered that a close relationship exists between PtCAMTA2 and PtCAMTA3. Interestingly, PtCAMTA2 and 3 were found to be closely associated with AtCAMTA3. These have been well studied and are known to participate together in SA-mediated defense responses and cold tolerance 24 . These results suggest that CAMTA genes have divergent functions in P. trichocarpa.
Promoter cis-elements play crucial roles in the response to biotic and abiotic stresses 25 . In this study, we identified many abiotic stress responsiveness cis-elements in the promoters of PtCAMTA family genes, including MBS, ABRE, TCA-element, G-Box, and W-Box. In particular, PtCAMTA5 was found to possess seven abiotic stresses responsiveness cis-elements, suggesting important functions under abiotic stress. Interestingly, PtCAMTA1, 2, and 3 did not contain ABRE elements in their −2kb promoters, but their responsiveness to ABA treatment was clear. We can thus speculate that it is inaccurate to assume a correlation between the existence and responsiveness to the related stress treatments. The same conditions were found in soybean, in that GmCAMTA7 and GmCAMTA9 contained ABRE elements in their promoters, but their responsiveness to ABA treatment was not obvious 7 . In particular, most PtCAMTA genes have SA and MeJA related cis-elements (TCA-element, and G-Box, CGTCA-motif, and TGACG-motif), but no fungus related cis-element. The qRT-PCR results showed that PtCAMTA2 and PtCAMTA3 were induced not only by SA and MeJA, but also by infection with A. alternata. This result suggests that a close relationship exists between SA and MeJA with regards to biotic stress.
Plants are often affected by abiotic and biotic stress in the process of growth and development. Under mannitol and NaCl stress, all the PtCAMTAs were induced in the leaves. Conversely, all the PtCAMTA genes were downregulated in the roots under NaCl stress. We suggested that all the PtCAMTA genes have significant functions under mannitol and NaCl stress. Previous studies have shown that all the GmCAMTA genes and the majority of ZmCAMTA genes are induced in the roots under drought and NaCl conditions, indicating that different expression patterns exist between woody plants and crops 8 . Alternaria alternata is a fungus that has been recorded to cause leaf spot and other diseases on over 380 host species of plant 26 . In the present study, all PtCAMTA genes were differentially expressed in the leaves under A. alternata infection. Three PtCAMTA genes were upregulated under the 24 h infection period, while four were downregulated. A previous study showed that the expression of all but one of the MtCAMTA genes was significantly downregulated under biotic stress (Sinorhizobium meliloti infection) 6 . These results indicate different expression patterns of CAMTA genes under biotic stress across a variety of species. In particular, four PtCAMTA genes that were downregulated under A. alternata infection exhibited similar expression patterns under SA and MeJA stress. This result indicates that PtCAMTA genes are involved in the pathogen expression network related to SA and MeJA pathways.
Research has indicated that a variety of hormone responses to biotic and abiotic stress could mediate the expression of CAMTA in Arabidopsis, including ABA (AtCAMTA2, 4, 5, and 6), SA (AtCAMTA2, 4, 5, and 6), and MeJA (AtCAMTA1, 3, and 4) 27 . It has been reported that ABA is involved in the mediation of drought stress, and that it is an important regulatory factor during drought stress 28 . Research has also demonstrated that SA and MeJA constitute the two major plant hormones that regulate plant biotic stress signal transduction 29 . In our study, most of the PtCAMTA genes were differentially expressed in the roots and leaves under ABA, MeJA, and SA treatments. This result indicates that phytohormones regulate the expression of PtCAMTA genes. PtCAMTA2 and PtCAMTA3, as homologous genes, were downregulated in the roots under ABA, MeJA, and SA treatment. Conversely, two soybean homolog genes (Glyma05g31190 and Glyma08g14370) in PtCAMTA2 and PtCAMTA3 were induced in the roots  Cold stress is a major environmental factor that can affect plant growth and development. Previous studies have shown that CAMTA transcription factors play an important role in cold regulation. AtCAMTA genes have been reported to be involved in cold stress, and the AtCAMTA1 and AtCAMTA3 double mutant in particular was found to have a negative effect on freezing tolerance 30 . Previous studies have shown that AtCAMTA genes might regulate freezing tolerance via the SA signaling pathway. AtCAMTA3 was found to be a negative regulator of the SA signaling pathway, and elevated levels of endogenous SA can enhance plant defense responses 31 . A study has shown that AtCAMTA3 is induced under cold stress, while AtCAMTA1 and AtCAMTA2 are downregulated under cold stress. Conversely, PtCAMTA2 and PtCAMTA3, the homologs of AtCAMTA1, 2 and 3, were reduced in P. trichocarpa. According to the expression of several alternative splicing forms of PtCAMTA genes, we found that most of the splice variants of the PtCAMTAs were also downregulated in the roots, but induced in the leaves under cold stress. Alternative splicing plays a key regulatory role in modulating gene expression during development, and in response to environmental stimuli 32 . In this study, PtCAMTA4.2 was more clearly expressed in leaves under cold stress. However, the expression of PtCAMTA4 was not significant. These might be related to the absence motifs of PtCAMTA4.2 alternative splicing variants which shown in Supplementary Figure S4. In conjunction, we believe these results indicate that PtCAMTA genes play an important role in the cold stress response, and that there are differences between P. trichocarpa and Arabidopsis in the response of CAMTA to cold stress. Furthermore, the expression patterns of CAMTA in P. trichocarpa differ from that of P. ussuriensis, in that most of the genes were clearly downregulated in the leaves under cold stress. We suggest that the differences in expression of the CAMTA genes under cold stress may be closely related to the differential cold tolerance of P. trichocarpa and P. ussuriensis.
In this study, we predicted eight potential PtCAMTA interactors using STRING software. The most consistent and interesting finding of this analysis was the different protein interactions between PtCAMTAs and AtCAMTAs. PtCAMTA2 and PtCAMTA3 share a high degree of homology with AtCAMTA3, which is known to be involved in both plant disease resistance and abiotic stress responses 33,34 . Rahman, H. et al. reported that AtCAMTA3 possessed 10 interactors that are DNA-binding transcription factors, including SRS, CBP60G, CM2, ICE1, XLG2, RHL41/ ZAT12, CBF1, CBF2, EDS1, and EDS16/ICS1. However, no protein that interacts with PtCAMTA2 and PtCAMTA3 was predicted using STRING software in the present study. This might be related to the absence of the CGCG CAMTA-binding element in the promoter of the homologous genes of the AtCAMTA3 interactors in P. trichocarpa. This result indicates that the regulatory pathway of PtCAMTA proteins is not the same as AtCAMTA proteins. An earlier study showed that the 38 potential interactors of AtCAMTAs are related to Ca 2+ signaling components, such as Ca 2+ /CaM-regulated protein kinases, Ca 2+ -dependent phospholipids, and CaM-binding proteins 35 . According to the protein interactions analyzed in this study, none were related to Ca 2+ signaling components among the eight potential interactors of PtCAMTA proteins. We boldly speculate that other binding sites exist besides the CGCG CAMTA-binding element in the promoter of the homologous genes of the AtCAMTA3 interactors in P. trichocarpa that act with PtCAMTA proteins. Another speculation is that PtCAMTA proteins act with the other DNA binding transcription factors to regulate Ca 2+ -related biological processes. The initial results of the present study have provided essential information on the Populus CAMTA family that may serve as basis for future functional studies, and will facilitate future mechanistic research aiming to investigate the divergent roles of these genes.
Gene structure, chromosome localization, and gene duplications. The coding domain sequences (CDS) and DNA sequences of the P. trichocarpa CAMTA genes were used to reveal the gene structure using the Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/index.php) 39 . The Multiple Expectation Maximization for Motif Elucidation (MEME) system (Version 4.9.1, http://meme.nbcr.net/meme/) was used to identify conserved motifs for each CAMTA gene 40 . The Softberry (http://linux1.softberry.com/berry.phtml?topic=fgene-sh&group=programs&subgroup=gfind) was used to generate the exon/intron organization. Bioinformatics analysis of the conserved domains was conducted using the Pfam database (http://pfam.janelia.org/). The domain structures of the PtCAMTAs were drawn using Domain Illustrator software (http://dog.biocuckoo.org/) 41 . In order to confirm the chromosomal locations of the CAMTA genes, all PtCAMTA genes were obtained from the PopGenIE version 3 database (http://www.popgenie.org/). The PtCAMTA genes defined as separate by five or fewer gene loci within a genetic distance of 100 kb were considered to be tandem duplicates 42 . A schematic view of the reorganization of homologous chromosome segments was based on the most recent account of whole-genome duplication in P. trichocarpa 43 .