Genome-Wide Identiﬁcation and Characterization of TCP Gene Family Members in Melastoma candidum

: It has been conﬁrmed


Introduction
TCP is a kind of plant-specific transcription factor (TF) with conserved sequences of 55 or 59 amino acids as a characteristic domain that is responsible for activating or repressing the transcription process and is involved in protein-protein interactions [1]. The TCP gene family was named after the TCP domain from the first identified members: TB1 substitution (Ks) and nonsynonymous substitution rates (Ka) and their ratio (Ka/Ks) were calculated to confirm the divergence time and major force to promote the evolution of TCP genes pairs in M. candidum. Evolutionary analyses among Arabidopsis, Populus and M. candidum were performed to illustrate the gene relationships of the three species. qRT-PCR was employed to investigate the expression pattern of the selected TCP genes in different tissues and developmental stages. Our research will assist in better comprehending the classification and expression pattern of TCP genes in M. candidum and provide valuable information for studying the functions of the TCP gene family during development or abiotic stress to further use them in the molecular breeding of M. candidum.

Chromosome Distribution and Evolution Relationship of TCP Genes
We first conducted an HMMER search through the genome of M. candidum and a total of 35 TCP gene family members were obtained. As a result of the distribution analysis of the 35 identified TCP gene members, it was found that all 35 members were unevenly distributed within 11 out of 12 chromosomes, with the exception of Chr11. With seven members and five members, Chr07 and Chr12 were the main TCP carriers with the highest proportion of TCP gene distribution, respectively occupying 20% and 14.3% of the total chromosomes. There were fewer TCP genes present in Chr02, Chr03, Chr04 and Chr06, as compared with other chromosomes; only two TCP genes were distributed ( Figure 1

Motif, Domain, Gene Structure and Promoter Analysis of TCP Genes in M. candidum
To further verify the identified 35 TCP gene members, we performed motif, domain and gene structure analysis. In the promoter region, genes in four clades (C1-C4) showed obvious motif characteristics: adjacent motif 1 and motif 2; genes in clade five and clade six displayed obvious adjacent motifs of 1 and 3. According to domain analysis, the most common characteristics of TCP gene family members were TCP superfamily proteins, which was a key characteristic that distinguishes the TCP gene family from other gene families. This indicates that all the 35 members belonged to the TCP gene family. There was, however, a great deal of variation in the TCP superfamily proteins between the different types of TCP individuals. The results of gene structure analysis showed that all the TCP genes in M. candidum contained only one exon, and more than 50% of TCP genes lack UTR annotation, which might be caused by the assembling quality of the genome (Figure 3).

Motif, Domain, Gene Structure and Promoter Analysis of TCP Genes in M. candidum
To further verify the identified 35 TCP gene members, we performed motif, domain and gene structure analysis. In the promoter region, genes in four clades (C1-C4) showed obvious motif characteristics: adjacent motif 1 and motif 2; genes in clade five and clade six displayed obvious adjacent motifs of 1 and 3. According to domain analysis, the most common characteristics of TCP gene family members were TCP superfamily proteins, which was a key characteristic that distinguishes the TCP gene family from other gene families. This indicates that all the 35 members belonged to the TCP gene family. There was, however, a great deal of variation in the TCP superfamily proteins between the different types of TCP individuals. The results of gene structure analysis showed that all the TCP genes in M. candidum contained only one exon, and more than 50% of TCP genes lack

Motif, Domain, Gene Structure and Promoter Analysis of TCP Genes in M. candidum
To further verify the identified 35 TCP gene members, we performed motif, domain and gene structure analysis. In the promoter region, genes in four clades (C1-C4) showed obvious motif characteristics: adjacent motif 1 and motif 2; genes in clade five and clade six displayed obvious adjacent motifs of 1 and 3. According to domain analysis, the most common characteristics of TCP gene family members were TCP superfamily proteins, which was a key characteristic that distinguishes the TCP gene family from other gene families. This indicates that all the 35 members belonged to the TCP gene family. There was, however, a great deal of variation in the TCP superfamily proteins between the different types of TCP individuals. The results of gene structure analysis showed that all the TCP genes in M. candidum contained only one exon, and more than 50% of TCP genes lack The promoter analysis results showed that the promoter region of some TCP genes contained plant hormone-response elements. For example, 60% (21 out of 35) of TCP genes promoters contained a TGA element, which is an auxin-responsive element; 62.86% (22 out of 35) carried a TCA element, which is a salicylic-acid-responsiveness element; 60% (21 out of 35) contained an O2 site element, which is a zein-metabolism-regulation element; 91.43% (32 out of 35) harbored an ABRE-binding site, which is an abscisic-acid-responsiveness element; 80% (28 out of 35) harbored an MeJA motif, which is an MeJA-responsiveness element, etc. Promoters of some genes contained a stress-responsive element. For instance, 65.71% (23 out of 35) of TCP gene promoters carried LTR, a low-temperature-responsiveness element; 54.29% (19 out of 35) had an MBS-binding site, a drought-inducibility binding site; 48.57% (17 out of 35) contained a TC-rich repeat, a defense stress-responsiveness element; only 5.71% (2 out of 35) contained a WUN motif, a wound-responsive element (Figures 4 and 5). All sequences of the bind sites are listed in Table S1. We also observed that the promoter of TCP2b included an MBSI element, which controls flavonoid biosynthesis (Table  S1).  The promoter analysis results showed that the promoter region of some TCP gen contained plant hormone-response elements. For example, 60% (21 out of 35) of TCP gen promoters contained a TGA element, which is an auxin-responsive element; 62.86% out of 35) carried a TCA element, which is a salicylic-acid-responsiveness element; 6 (21 out of 35) contained an O2 site element, which is a zein-metabolism-regulation e ment; 91.43% (32 out of 35) harbored an ABRE-binding site, which is an abscisic-acidsponsiveness element; 80% (28 out of 35) harbored an MeJA motif, which is an Me responsiveness element, etc. Promoters of some genes contained a stress-responsive e ment. For instance, 65.71% (23 out of 35) of TCP gene promoters carried LTR, a low-te perature-responsiveness element; 54.29% (19 out of 35) had an MBS-binding site drought-inducibility binding site; 48.57% (17 out of 35) contained a TC-rich repeat, a fense stress-responsiveness element; only 5.71% (2 out of 35) contained a WUN moti wound-responsive element (Figures 4 and 5). All sequences of the bind sites are listed Table S1. We also observed that the promoter of TCP2b included an MBSI element, wh controls flavonoid biosynthesis (Table S1).  The promoter analysis results showed that the promoter region of some TCP gen contained plant hormone-response elements. For example, 60% (21 out of 35) of TCP gen promoters contained a TGA element, which is an auxin-responsive element; 62.86% out of 35) carried a TCA element, which is a salicylic-acid-responsiveness element; 6 (21 out of 35) contained an O2 site element, which is a zein-metabolism-regulation e ment; 91.43% (32 out of 35) harbored an ABRE-binding site, which is an abscisic-acidsponsiveness element; 80% (28 out of 35) harbored an MeJA motif, which is an MeJ responsiveness element, etc. Promoters of some genes contained a stress-responsive e ment.  Table S1. We also observed that the promoter of TCP2b included an MBSI element, wh controls flavonoid biosynthesis (Table S1).

Conserved Region Analyses of the Identified TCP Proteins
We aligned the protein sequences of 35 TCPs to find their conserved region by the ClustalW method. The results showed that these 35 TCPs were not conserved well enough ( Figure 6), indicating that there are functional differences between members of the TCP gene family (Figure 3). This was consistent with the results of the domain analysis of TCPs in other species. An earlier study has shown that TCP TFs can be classified into two classes, Class I and Class II [6]. In both of the classes, the N-terminus of the proteins is characterized by a basic-helix-loop-helix structure motif [40]. According to our aligned sequences of these genes, we found that they could be roughly divided into two classes.  Figure S1b) and belong to the TCP-C type with a 59 aa-long conserved region. In both groups, their conservation parts were comprised of Basic Helix I-Loop-Helix II structures. Our conservation analysis results illustrated that these TCPs may have evolved into two functionally different groups in M. candidum. In addition, we observed that seven TCPs in CYC/TB1 also contained another conserved domain, namely the R domain, with six of these genes having an R domain with a length of 18 aa, whereas McTCP18 only had a 14 aa-length R domain (Figure 6b).

Conserved Region Analyses of the Identified TCP Proteins
We aligned the protein sequences of 35 TCPs to find their conserved region by the ClustalW method. The results showed that these 35 TCPs were not conserved well enough ( Figure 6), indicating that there are functional differences between members of the TCP gene family ( Figure 3). This was consistent with the results of the domain analysis of TCPs in other species. An earlier study has shown that TCP TFs can be classified into two classes, Class I and Class II [6]. In both of the classes, the N-terminus of the proteins is characterized by a basic-helix-loop-helix structure motif [40]. According to our aligned sequences of these genes, we found that they could be roughly divided into two classes.  Figure  S1b) and belong to the TCP-C type with a 59 aa-long conserved region. In both groups, their conservation parts were comprised of Basic Helix I-Loop-Helix II structures. Our conservation analysis results illustrated that these TCPs may have evolved into two functionally different groups in M. candidum. In addition, we observed that seven TCPs in CYC/TB1 also contained another conserved domain, namely the R domain, with six of these genes having an R domain with a length of 18 aa, whereas McTCP18 only had a 14 aa-length R domain (Figure 6b).

Duplication Events and Divergence Time Estimation of TCP Gene Pairs
To investigate the duplication events of identified TCP genes, we conducted a collinearity analysis of these TCP genes from the genome level by using TBtools software. Figure 7a shows the associated gene pairs of TCP genes. A total of 34 out of 35 genes were found to have corresponding genes, indicating that TCP genes have gone through extensive duplication events. To understand the divergence time of gene pairs, we calculated the synonymous substitution rate (Ks/dS) and estimated the divergence time by using a divergence rate of 6.5 × 10 −9 per synonymous site per year [41]. There was a wide range of divergence times between 12.8 and 99.82 million years ago for the TCP genes ( Figure 7b Combining phylogenetic trees, we found that the longer the divergence time, the more distant evolutionary relationships were. A substitution ratio mutation (Ka/Ks) reflects the selection method experienced by gene pairs. When Ka/Ks < 1, the genes experience purifying selection, which means the selection process could neutralize mutation to maintain the stability of the protein; in contrast, when Ka/Ks > 1, the genes experience positive selection, which means great mutation happens in genes and eventually leads to a change in coded proteins. Our identified TCP gene pairs had Ka/Ks values ranging from 0.1 to 0.33, proving that all of these genes experienced a purification selection process in M. candidum (Figure 7b). This reflected that M. candidum experienced little severe mutation disturbance during its life cycle on earth.

Duplication Events and Divergence Time Estimation of TCP Gene Pairs
To investigate the duplication events of identified TCP genes, we conducted a collinearity analysis of these TCP genes from the genome level by using TBtools software. Figure 7a shows the associated gene pairs of TCP genes. A total of 34 out of 35 genes were found to have corresponding genes, indicating that TCP genes have gone through extensive duplication events. To understand the divergence time of gene pairs, we calculated the synonymous substitution rate (Ks/dS) and estimated the divergence time by using a divergence rate of 6.5 × 10 −9 per synonymous site per year [41]. There was a wide range of divergence times between 12.

Evolutionary Relationship of TCP Genes among Different Species
Eukaryotic genomes differ in the degree to which genes remain on corresponding chromosomes (synteny) and in corresponding orders (collinearity) [42]. Comparative analysis of species genomes could illustrate genomic evolution. Species relationships could be studied by searching for conserved genes pairwise among them [7]. To understand the evolutionary relationships of TCP genes among different species, we conducted a multi-collinearity analysis by selecting herbal species A. thaliana and woody species P. trichocarpa. A total of 71.4% (25 out of 35) TCP genes of M. candidum had a collinearity connection with 18 TCP genes of A. thaliana, and 74.3% (26 out of 35) TCP genes of M. candidum had a collinearity connection with 20 TCP genes of P. trichocarpa (Figure 8). According to the phylogenetic tree, these TCP genes were divided into three main groups. The first clade was mainly comprised of TCP-P group members, such as TCP9 (At2g45680, Potri.003G120201, Potri.001G111800, McTCP9), TCP19 (At5g51910, McTCP19, Potri.012G135900), TCP6 (At5g41030, McTCP6), TCP20 (At3g27010, Potri.001G327100, Potri.003G167900, Potri.001G060000, McTCP20), TCP14 (At3g47620, McTCP14), TCP15 (At1g69690, McTCP15), TCP22 (At1g72010, Potri.019G081800, McTCP22) and TCP23 (At1g35560, Potri.013G110700) (Figure 9). It was interesting to note that McTCP9 is highly homologous to At2g45680 (AtTCP9), whereas it was closely clustered with TCP19 of Arabidopsis and P. trichocarpa in the phylogenetic tree. McTCP19 was highly homologous to AT5G51910 (AtTCP19), but it was closely clustered with TCP9 of Arabidopsis and P. trichocarpa. This evidence showed that TCP9 and TCP19 had relatively close evolutionary relationships. The second clade mainly comprised CYC/TB1-type TCP genes, such as TCP1 (Potri.017G112000), TCP12 (Potri.015G050500, Potri.012G059900, Potri.008G115800, Potri.010G130200, At1g68800, McTCP12b-e) and TCP18 (At3g18550 ,  McTCP18). Interestingly, in this clade, we only obtained one TCP1 gene in P. trichocarpa, indicating a low collinearity relationship for this gene among three species and high similarity of TCP1 in P. trichocarpa with other TCP genes such as McTCP12e in the other two species (Figure 9). More TCP12 genes were found in Populus and M. candidum, suggesting that more complex gene-duplication events for TCP12 happened in these two species than in Arabidopsis. The third clade mainly comprised CIN-type TCP genes, such as TCP4 (Potri.011G096600, Potri.  (Figure 9). Amongst them, McTCP4 and McTCP5 had more members, suggesting that these two subfamilies experienced more complex gene-duplication events in M. candidum than in Arabidopsis and P. trichocarpa. TCP2, TCP4, TCP10, TCP14, TCP15 and TCP18 were only found in collinearity relations between M. candidum and Arabidopsis, reflecting that these genes were conserved in these two species. In each clade, we observed a closer evolutionary relationship intraspecies than interspecies. In all, our results still proved the similar characteristics of TCP genes in different species, and the TCP genes identified in M. candidum were reliable. Combining phylogenetic trees, we found that the longer the divergence time, the more distant evolutionary relationships were. A substitution ratio mutation (Ka/Ks) reflects the selection method experienced by gene pairs. When Ka/Ks < 1, the genes experience purifying selection, which means the selection process could neutralize mutation to maintain the stability of the protein; in contrast, when Ka/Ks > 1, the genes experience positive selection, which means great mutation happens in genes and eventually leads to a change in coded proteins. Our identified TCP gene pairs had Ka/Ks values ranging from 0.1 to 0.33, proving that all of these genes experienced a purification selection process in M. candidum ( Figure  7b). This reflected that M. candidum experienced little severe mutation disturbance during its life cycle on earth.

Evolutionary Relationship of TCP Genes among Different Species
Eukaryotic genomes differ in the degree to which genes remain on corresponding chromosomes (synteny) and in corresponding orders (collinearity) [42]. Comparative analysis of species genomes could illustrate genomic evolution. Species relationships could be studied by searching for conserved genes pairwise among them [7]. To understand the evolutionary relationships of TCP genes among different species, we conducted members, suggesting that these two subfamilies experienced more complex gene-duplication events in M. candidum than in Arabidopsis and P. trichocarpa. TCP2, TCP4, TCP10, TCP14, TCP15 and TCP18 were only found in collinearity relations between M. candidum and Arabidopsis, reflecting that these genes were conserved in these two species. In each clade, we observed a closer evolutionary relationship intraspecies than interspecies. In all, our results still proved the similar characteristics of TCP genes in different species, and the TCP genes identified in M. candidum were reliable.

Expression Patterns of the Identified TCP Genes in Different Tissues of M. candidum
To understand the expression pattern of TCP genes, we took part of them to perform qRT-PCR experiments with cDNA as templated from nine kinds of tissues of M. candidum. These nine tissues included young leaves (YL), adult leaves (AL), young stems (YS), adult stems (AS), seeds (S), roots (R), early-stage flowers (EF), middle-stage flowers (MF) and blooming flowers (BF) (Figure 9). Interestingly, our study suggested that more than half

Expression Patterns of the Identified TCP Genes in Different Tissues of M. candidum
To understand the expression pattern of TCP genes, we took part of them to perform qRT-PCR experiments with cDNA as templated from nine kinds of tissues of M. candidum. These nine tissues included young leaves (YL), adult leaves (AL), young stems (YS), adult stems (AS), seeds (S), roots (R), early-stage flowers (EF), middle-stage flowers (MF) and blooming flowers (BF) (Figure 9). Interestingly, our study suggested that more than half of the selected TCP genes, including   Duncan's test was used to evaluate significant difference levels. Lowercase letters mean p ≤ 0.05, capital letters mean p ≤ 0.01.

Discussion
TCP TFs play important roles in diverse biological processes [43,44], making them promising candidates in molecular breeding. Although the TCP genes have been identified in some model species, such as Arabidopsis, rice, maize, Populus, etc., the identification of TCP genes in M. candidum has still not been reported because of the absence of complete genome resources. In this study, we integrated the HMMER search and BLASTp method to identify TCP gene family members based on the genome of M. candidum. Combining domain analyses, a total of 35 TCP TFs were identified and were classified into two typical classes of TCPs, TCP-P and TCP-C. Most TCP genes were found to have been duplicated, resulting in their expansion in M. candidum. The expansion of TCP genes may reflect the involvement of these genes in a more complicated transcription process in the perennial shrub species.
The analysis of the phylogenetic trees of TCPs-in M. candidum and multi-speciessupports the results of earlier related research [2][3][4], which has declared two subfamilies are included in TCP gene family. In addition, we have also detected conserved regions in our identified TCP proteins that are similar to those reported previously, which illustrates that there is a high degree of conservation of TCPs across species. However, there were also some subtle differences. For example, several studies have suggested that the R domain of the R proteins is primarily responsible for mediating the interactions of proteins [45]. In contrast, in CIN-and TCP-P-type TFs, we did not yet find the R domain; this is different from TCP identification in bananas [7,45], indicating different characteristics of TCPs between M. candidum and bananas. It is important to note, however, that there are different numbers or types of TCPs in different species. A number of previous studies have also revealed that there are a large number of differences between different species, ranging from 17 to 75 family members [6,31,32,35,37,39]. During the course of our study, a total of 35 TCP gene members were identified. However, comparing 24 TCP genes in Arabidopsis, we observed that TCP15 and TCP23 were not present in our results. When we used 24 TCP genes of Arabidopsis as a query to blast against the database of the M. candidum genome, we found that all 24 TCP genes of Arabidopsis were able to find the best match in M. candidum (Table S2). Furthermore, all of the best matches in this study belonged to the TCP genes we had identified in M. candidum. Some TCP genes in Arabidopsis, such as AT1G53230 (AtTCP3) and AT3G15030 (AtTCP4), were best matched with McTCP3, and AT2G37000 (AtTCP11), AT3G47620 (AtTCP14) and AT1G69690 (AtTCP15) were best matched with McTCP11, illustrating that gene-duplication events also happened in Arabidopsis. Hence, it is normal that we might not be able to find all TCP genes in our results because of the different rules for naming genes. In regards to the different number of genes identified in a specific family, it has been suggested that it may be due to the differences in the species themselves; for example, gene-duplication events and the size of the genome in one species could have an effect on the number of genes identified in a specific family [45]. The second reason is because of the threshold that was chosen when performing the HMMER search. There are some researchers who broaden the definition of the E value from 0.01 to 0.10 [10]. There is no doubt that the results will be different if the threshold is set at a different level. As of now, researchers only name genes numerically, which results in the loss of gene structure information, as well as the existence of gene duplications. Therefore, in a related field, authoritative and reasonable naming standards are urgently needed.
The TCP transcription factors are ancient proteins that are specific to plants. Although there have been no reports in unicellular algae, they have been reported in pluricellular green algae, moss, ferns and lycophytes, typically with five to six members in them [10,43]. Gene family expansion and evolution are mostly attributed to gene-duplication events such as segmental, tandem, transposition and whole-genome-duplication (WGD) events [46,47]. Segmentally duplicated genes are those that are present on different chromosomes and show similar expression patterns [48]. According to this concept, we found that many gene pairs were also distributed across different chromosomes as well, such as candidum (Figure 7a). The results of this study are in agreement with previous investigations of the TCP gene family in other species [35,38,49]. In most plants, WGD is also the dominant cause of genome diversity [7]. Based on synonymous substitution rate (Ks) analysis, previous studies had defined different types of WGD events such as α + β WGD and γ WGD events [45,50]. Using this method, we were able to determine that the genome of M. candidum has undergone three WGD events during evolution. Amongst them, approximately 41.7% (10 out of 24) TCP gene pairs experienced WGD α events (Ks < 0.45), 16.7% (4 out of 24) of TCP gene pairs experienced WGD β events (0.45 ≤ Ks ≤ 0.85) and 41.7% (10 out of 24) TCP gene pairs experienced WGD γ events (Ks > 0.85) (Figure 7b), suggesting WGD is responsible for the expansion of the TCP gene family in M. candidum. A gene pair with a Ks greater than 0.85 indicates that the gene may have originated from a more ancient duplication event and has since undergone multiple rounds of WGD. Earlier studies have shown that TCP gene expansion is not uniformly distributed in various classes of bananas. For example, the CIN subclade exhibited more gene duplication than any other subclade [7]. However, in our results, we did not observe uneven distribution of geneduplication events in three types of TCP genes. There were duplication events detected in most of the TCP genes of different types (Figure 7a,b). It is a possibility that this may be caused by the species itself.
There is a relationship between the expression profile of genes and the function of those genes [51]. The genes had distinct expression profiles in diverse organs, suggesting their role in the development of various organs. There is a great deal of evidence that TCP transcription factors play an important role in plant growth and development, including the development of all types of branches, leaves and flowers [21,45], as well as fruit development and ripening [37]. Some research had proved that more than two thirds of TCP-C subfamily genes have organ-specific expressions [10] and could inhibit plant growth and cell differentiation [22]. TCP-P-and some CIN-genes were detected in the flower, leaf and stem of Prunus mume [10]. In Arabidopsis, eight CIN-type genes, including AtTCP2, AtTCP3, AtTCP4, AtTCP5, AtTCP10, AtTCP13, AtTCP17 and AtTCP24, are highly transcribed in the leaf and are responsible for the regulation of leaf growth [16,52]. CYC genes may be related to changes in petal size [53] and floral zygomorphy [54]. In accordance with the previous study, our results of qRT-PCR assays revealed that some members of the TCP-P sub-family-such as  Figure 9). These results indicate that TCP genes may be functionally conserved among different species.

Identification of TCP Transcription Factors
The genome resource of M. candidum was provided by Zhou's group of Sun Yat-sen University (unpublished data). To identify TCP gene family members in M. candidum, the TCP domain HMM profile (PF03634) was used as a query to perform a HMMER search with an E-value cut-off of 1 × 10 −3 through the M. candidum genome by following the HMMER User Guide. All the motif analyses for the obtained proteins were conducted on MEME Suite Version 5.5.0. Available online: https://meme-suite.org/meme/tools/meme (accessed on 24 August 2022) and domain analyses were performed on the PFAM website (https://www.ebi.ac.uk/Tools/pfa/pfamscan/) with an E-value 1 × 10 −5 to further verify the identified TCP gene members.

Phylogeny Tree Construction and Location of TCP Gene Family Members in Chromosome
All the identified TCP genes were aligned by MUSCLE method in MEGA 6.01, and then a phylogenetic tree was constructed by the ML (maximum likelihood) method based on LG models with 1000 bootstrap replications. The phylogeny tree was visualized on the Interactive tree of life (iTOL). Available online: https://itol.embl.de/ (accessed on 26 August 2022). By utilizing the Gene Location Visualize from the GTF/GFF function module of TBtools software, we visualized the distribution of TCP genes along the chromosomes through the gtf annotation of the genome and the gene density file.

Visualization of Motif, Domain, Gene Structure and Promoter of TCP Genes
Motif, domain, gene structure and promoter analyses were conducted for all the identified TCP genes. The upstream 2000 bp of the TCP CDs were extracted for the purpose of conducting promoter analyses on PlantCARE. Available online: https://bioinformatics.psb. ugent.be/webtools/plantcare/html/ (accessed on 28 August 2022). With the .gff file of the genome, gene structures including CDs and untranslational region (UTR) were displayed in the gene structure view of TBtools. Based on the identification of TCP transcription factor parts, the motifs and domains of TCP genes were visualized by the gene structure view and Batch SMART module of the TBtools software, respectively.

Conserved Region Analysis of the TCP Proteins
The protein sequences of identified TCPs were first aligned by the ClustalW method in MEGA software. The aligned fasta file was input into Jalview software to show the conserved region of all the aligned TCPs. The Seqlogo graph of the conserved region was produced by the Amazing Simple SeqLogo module of TBtools to show the conservation of amino acids in the corresponding region.

Identification of TCP Gene Pairs and Divergence Time Estimation
Gene collinearity analysis and visualization within the M. candidum genome were conducted by TBtools [55]. In a nutshell, (1) the collinearity, CTL, and .gff file were produced by the One Step MCScanX-Super Fast module; (2) the chromosome length file was generated by the Fasta stats module; (3) a gene pair file was produced from the collinearity file by the File Merge for MCScanX module; (4) a link region file was generated by the File Transformat for Microsyteny viewer module; (5) gene pairs of TCP gene family members were produced by the Text Block Extract and Filter module. Based on the above analyses, we aligned protein sequences and ORFs of the TCP gene pairs by ClustalW method in MEGA software. The synonymous substitution (Ks) and non-synonymous substitution (Ka) rates were calculated using the CODEML program of PAML on PAL2NAL. Available online: http://www.bork.embl.de/pal2nal/ (accessed on 29 August 2022) [56]. Divergence times (DT) of the gene pairs were estimated using the formula T = Ks/2λ, with the divergence rate λ = 6.5 × 10 −9 [41,57]. TCP gene pairs were visualized in the Advanced circos module of the TBtools [55].

Collinearity and Phylogeny Analyses of TCP Genes among Different Species
Multiple Chr layout, gene link and .gff files between A. thaliana and M. candidum, P. trichocarpa and M. candidum were produced by the One Step MCScanX-Super Fast module of the TBtools with an E-value of 1 × 10 −3 . The homologous genes among the different species were obtained from the merged gene link file after merging three files of two comparison groups using the File Merge For MCScan-X module. By using the extracted protein sequence of these homologous genes, a phylogeny tree among three species was performed in MEGA by the ML method with 1000 boot replications. TCP gene collinearity plots among different species were visualized by the Multiple Synteny Plot module of the TBtools [55].

Quantitative Real-Time PCR and Statistical Analysis of the Selected TCP Genes in Different Tissues of M. candidum
A total of 22 selected TCP genes (Table S3) were blasted against the whole genome of M. candidum to find a specific region, and then the primers were designed for qRT-PCR. Nine tissues, including young leaves (YL), adult leaves (AL), young stems (YS), adult stems (AS), seeds (S), roots (R) and three-stage flowers (EF, MF, LF), were collected to extract the total RNA by OminiPlant RNA Kit (DNase I) (CW2598, CWBIO, Taizhou, China) following the manufacturer's instructions. After checking the quality of the total RNAs, 0.5 µg of total RNA was reverse-transcribed into first-strand cDNA using the PrimeScript RT reagent Kit gDNA Eraser (Takara, Dalian, China). Then, the SYBR @Premix Ex Taq TMII (Takara, Dalian, China) was used for qRT-PCR of genes, following the manufacturer's instruction, on the Illumina Eco real-time PCR system (Illumina, USA) platform. The α-tubulin gene of M. candidum was used as the internal reference gene. Ct values obtained on the thermal cycler platform were then calculated by the 2 −∆Ct algorithm [58]. Using mixtures of cDNAs from nine tissues, primer PCR amplification efficiency was evaluated by following Pfaffl's research [59]. Primers sequences, length of PCR products and PCR amplification efficiency for each pair of primers are listed in Table S3. Relative expression levels of selected genes in different tissues were analyzed using ANOVA with Duncan's test by SPSS 18.0 software. The graphs were visualized by GraphPad Prism 9 and Adobe Illustrator 2020.

Conclusions
Collectively, a total of 35 TCP gene family members were identified based on the genome-wide identification of M. candidum in our study. In the identified TCP gene family members, there was a common domain in the TCP superfamily. Despite the fact that there were differences between all the TCP genes, the same gene type showed high conservation. The number of TCP genes was more due to more frequent gene-duplication events that occurred in M. candidum. It is more likely that the majority of TCP genes had been affected by natural selection rather than human interference. TCP genes showed distinct family members from two model species. A greater focus should be placed on functional exploration to expand their application in the garden and pharmaceutical industries. Our results provided valuable information for understanding the classification and functions of TCP genes in M. candidum.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27249036/s1, Figure S1: Conservation analysis of two types of TCP genes. The same color in the column and the big size of the letters mean a highly conservative region; Table S1: Identified cis-elements in promoter region of TCP genes; Table S2: Blast results of 24 TCP genes in Arabidopsis with corresponding genes in M. candidum; Table S3: Primers used in qRT-PCR.