Characterization and Coexpression Analysis of the TIFY Family Genes in Euryale ferox Related to Leaf Development

TIFYs are plant-specific transcription factors that contain the TIFY structural domain and play an important role in plant leaf growth and development. However, the role played by TIFY in E. ferox (Euryale ferox Salisb.) leaf development has not been investigated. In this study, 23 TIFY genes were identified in E. ferox. Phylogenetic analyses of the TIFY genes showed clustering into three groups (JAZ, ZIM, and PPD). The TIFY domain was shown to be conserved. JAZ was mainly expanded via wholegenome triplication (WGT) in E. ferox. Based on analyses of the TIFY genes in nine species, we found that JAZ has a closer relationship with PPD, in addition to appearing the most recently and expanding most rapidly, leading to the rapid expansion of TIFYs in Nymphaeaceae. In addition, their different evolution types were discovered. Different gene expressions showed the distinct and corresponsive expression patterns of the EfTIFYs in different stages of tissue and leaf development. Finally, The qPCR analysis revealed that the expression of EfTIFY7.2 and EfTIFY10.1 showed an upward trend and high expression throughout leaf development. Further co-expression analysis indicated that EfTIFY7.2 might be more important for the development of E. ferox leaves. This information will be valuable when exploring the molecular mechanisms of EfTIFYs in plants.


Introduction
TIFY family members play important roles in plant growth and development and stress responses [1,2]. In Arabidopsis, JAZ is involved in JA hormonal responses. The subfamily that contains JAZ also contains the Jas domain [3,4], which involves interactions with basic helix-loop-helix (bHLH) and myeloblastosis (MYB) transcription factors that regulate different JA-dependent responses [5][6][7][8][9]. The jasmonate-insensitive phenotype is a result of the overexpression of a truncated JAZ1 or the mutant of JAI3/JAZ3 gene (AT3G17860) in Arabidopsis [10]. The importance of CORONATINE-INSENSITIVE1 (COI1) in JA signaling, such as null mutations at this locus, is that it abolishes JA responses in diverse plant species [11]. In the presence of JA, JAZ proteins are targeted by the SCFCOI1 complex for ubiquitination and degradation, which consequently relieves the repression and rapid activation of JA responses [10,12,13]. AtTIFY10b/JAZ1 is degraded to stop the repression of JA responses by the SCFCOI1-dependent 26S proteasome pathway [14]. In addition, JAZ6 (AT1G72450) and JAZ9 (ATIG70700) have been shown to be redundant with JAZ1 in the JA signaling pathway [13,15]. AtTIFY9 is silenced by RNAi, which increases MeJA sensitivity [16]. Recently, regarding JAZ10, it was discovered that homoand heteromeric interactions between most Arabidopsis JAZs are mediated by the conserved TIFY motif within the domain [17].
TIFY is a particular plant-specific class, which has a conserved TIFY domain to encode proteins (Zinc finger protein expressed in Inflorescence Meristem) [18]. The highly conserved TIFY domain contains a core motif, TIF [F/Y]XG, in the protein [18]. AT4G24470 was the first identified TIFY gene, expressed in reproductive tissue in Arabidopsis thaliana [19]. ZIM contains a C2C2-GATA zinc-finger (CX2CX20CX2C) [19]. The definition of the TIFY, ZIM, and C2C2-GATA gene families was unclear for a long time [20,21], until comprehensive system analyses of sequences were conducted to clarify this issue. Finally, the TIFY domain-containing members were identified and named TIFY. Based on phylogenetic and gene structure analyses, the TIFY proteins were divided into three groups, namely, ZML, PPD, and JAZ, in Arabidopsis [18,22]. While all TIFY proteins bear a TIFY domain, those in the ZML subfamily contain CCT (CONSTANS, CO-like, TOC1) and ZML (GATA-zinc finger domain) domains. In contrast, both the PEAPOD (PPD) and JASMONATE ZIM-DOMAIN (JAZ) subfamilies lack GATA and CCT domains [23]. In addition to this, the JAZ subfamily also contains the conserved sequence of the Jas motif, which bears the characteristic motif SLX2FX2KRX2RX5PY [24]. PPD proteins bear a unique N-terminal PPD domain, as well as a divergent Jas motif that lacks the conserved PY at its C-terminus.
Although this family has been reported in many species, only a few of genes have been functionally analyzed in Arabidopsis. The first discovered TIFY gene, ZIM/TIFY1 (AT4G24470), is overexpressed using CaMV 35S, resulting in petiole and hypocotyl elongation [21]. ZML1/TIFY2b (AT3G21175) and ZML2/TIFY2a (AT1G51600) are highly homologous with ZIM/TIFY1 (AT4G24470), identified as having similar domain compositions in terms of their encoded proteins and expression patterns compared to ZIM; however, their functions still remain unclear [22]. In Arabidopsis, the mutations of PPD1/TIFY4a and PPD2/TIFY4b show a leaf dome shape and an increased size. Overexpression of PPD could reduce the lamina size and promote silique development [24,25].
Euryale ferox is part of the angiosperm basal plant family Nymphaeaceae and has nutritional value for humans. It has also undergone triplication events dating back to~17 million years ago (MYA). Recent completion of the Euryale ferox genome sequencing can now allow the genome-wide identification of the TIFY gene family [26]. The TIFY gene family has been extensively investigated in many sequenced plants [18,27]. However, further analyses of the evolutionary divergence, gene retention and fractionation, and genome-wide expression patterns of this gene family in Euryale ferox are needed.
In the present study, we identified 23 EfTIFY genes in E. ferox via a database search and analysis of the phylogenetic relationships, conserved motifs, and retention TIFY genes and N. colorata TIFY genes. We further studied the syntenic gene pairs, gene structure, duplication, and tissue-specific expressions of EfTIFYs in different tissues. Through transcriptome and qPCR analyses, key candidate TIFY genes closely related to leaf development were identified, which will provide a theoretical basis for analyzing the molecular mechanism of leaf development in E. ferox.

Identification of TIFY Genes in E. ferox
A HMM for the TIFY domain identified 25 gene candidates in the E. ferox genome. Subsequently, by using Pfam and SMART, we found that two of the 25 putative TIFY genes lacked TIFY domains. Therefore, 23 genes in E. ferox were identified as TIFY family members (Supplementary Table S2). Three conserved domains were identified in the EfTIFY genes, including TIFY, Jas, and GATA. In particular, TIFY was more conserved, comprising approximately 40 amino acids. Jas and GATA were located in the C-terminus of the proteins. TIFYs can be divided into three subgroups according to their structural domains. JAZ contains TIFY and JAZ structural domains, PPD contains TIFY and truncated JAZ structural domains, and ZIM contains TIFY, CCT and ZIM structural domains (Figure 1b). proteins (EfTIFY3.2, EfTIFY4.4, EfTIFY10.2, EfTIFY10.1, and EfTIFY3.1) showed a low pI of nine. The other classes had complex theoretical pI values, ranging from four to eight. EfTIFY proteins are unstable proteins, based on calculation of the average instability index being approximately 68.11. All EfTIFY proteins are hydrophilic, as their hydropathicity values were shown to be <0.

Phylogenetic and Classification Analysis of EfTIFY
To investigate the classification and expansion of the TIFY genes in E. ferox, N. colorata, and A. thaliana, we conducted phylogenetic analyses of all of the TIFY genes. The phylogenetic analyses indicated that the EfTIFY family can be divided into three subfamilies (ZIM and ZML, PPD, and JAZ). JAZ has the largest number of EfTIFY genes (50%) (Figure 1a). To investigate the structural characteristics of the TIFY family in E. ferox, we chose TIFY of A. thaliana as a comparison. In the E. ferox and Arabidopsis TIFY proteins, we identified 10 conserved motifs ( Figure 2). Genes from the same subfamily that share similar motif compositions are likely to share similar functions. Through a comparison of the Arabidopsis and E. ferox TIFY proteins, we found that they have similar structures in every subgroup. Moreover, each class of the EfTIFY proteins had several special motifs, such as motif 5, located at the N-terminal, which only exists in JAZV, and motif 4, located at the C-terminal, which only exists in PPD. All of the EfTIFY proteins contained motif 1, corresponding to the TIFY conserved domain. Motif2 corresponds to the CCT structural domain and motif5 corresponds to the GATA domain. These domains are the key to the function of TIFY family genes. The gene structure analysis indicated that most EfTIFY genes have at least three introns. As shown in Figure S1, the ZML and PPD subfamilies contained relatively more introns, meaning that they were stable and conserved during the evolution process [28]. The chemical characteristics of EfTIFY by ExPASy were predicted (Supplementary Table S2); the theoretical pI of the JAZ subfamily was approximately nine, but five EfTIFY proteins (EfTIFY3.2, EfTIFY4.4, EfTIFY10.2, EfTIFY10.1, and EfTIFY3.1) showed a low pI of nine. The other classes had complex theoretical pI values, ranging from four to eight. EfTIFY proteins are unstable proteins, based on calculation of the average instability index being approximately 68.11. All EfTIFY proteins are hydrophilic, as their hydropathicity values were shown to be <0.

Phylogenetic and Classification Analysis of EfTIFY
To investigate the classification and expansion of the TIFY genes in E. ferox, N. colorata, and A. thaliana, we conducted phylogenetic analyses of all of the TIFY genes. The phylogenetic analyses indicated that the EfTIFY family can be divided into three subfamilies (ZIM and ZML, PPD, and JAZ). JAZ has the largest number of EfTIFY genes (50%) (Figure 1a). To investigate the structural characteristics of the TIFY family in E. ferox, we chose TIFY of A. thaliana as a comparison. In the E. ferox and Arabidopsis TIFY proteins, we identified 10 conserved motifs ( Figure 2). Genes from the same subfamily that share similar motif compositions are likely to share similar functions. Through a comparison of the Arabidopsis and E. ferox TIFY proteins, we found that they have similar structures in every subgroup. Moreover, each class of the EfTIFY proteins had several special motifs, such as motif 5, located at the N-terminal, which only exists in JAZV, and motif 4, located at the C-terminal, which only exists in PPD. All of the EfTIFY proteins contained motif 1, corresponding to the TIFY conserved domain. Motif2 corresponds to the CCT structural domain and motif5 corresponds to the GATA domain. These domains are the key to the function of TIFY family genes. The gene structure analysis indicated that most EfTIFY genes have at least three introns. As shown in Figure S1, the ZML and PPD subfamilies contained relatively more introns, meaning that they were stable and conserved during the evolution process [28].

Differential Copy Number and Retention of EfTIFY
To investigate the copy number variation in TIFY during E. ferox-specific W events, we compared the homologous TIFY genes in N. colorata and the three E. ferox s genomes (Sub1, Sub2, and Sub3) and found most EfTIFYs on the conserved collin blocks (Supplementary Table S3). Specifically, compared to the PPD and JAZ genes, o the ZIM genes have been retained in three copies. The majority of the ZIM (100%) ge were retained in two or three copies, compared to only 83.33% of the JAZ genes a 50.00% of the PPD genes. It is known that the JAZ and ZIM proteins have many import functions that negatively regulate JA signaling via direct interaction with several tr scription factors, according to the gene dosage hypothesis. In addition, the percentage homeologs varies among the three sub-genomes. Compared to the other group of gen significantly more TIFY gene homologs were retained in the Sub2 and Sub3 sub-genom AKE (ancestral karyotype of E. ferox) (AKE1-12) has been inferred in E. ferox [29,30]. M of the 23 EfTIFY genes belong to AKE3 (43.4%), followed by AKE5 (21.7%), while only of the EfTIFY genes are assigned to AKE11.

Expansion and Evolution of the TIFY Genes in Plants
Comparative genomic analysis confirmed that E. ferox underwent genome tripli tion since its divergence from N. colorata [26]. For the analyses of the EfTIFY genes' trip cation in N. colorata from E. ferox, the syntenic gene pairs were analyzed. We identified syntenic orthologous gene pairs using the MCScanX program among all of the TIFY p teins of E. ferox. Additionally, 33 syntenic orthologous TIFY gene pairs were identifi between E. ferox and N. colorata ( Figure 3). Among the syntenic orthologous gene pa we found that each NcTIFY gene had two or three E. ferox syntenic orthologous gen demonstrating that the TIFY genes in E. ferox underwent duplication accompanied by nome triplication. However, the gene number in the E. ferox genome was notably th times lower than that of the N. colorata, indicating the gene loss that occurred during polyploid process. The visualization of the syntenic orthologous genes among the t

Differential Copy Number and Retention of EfTIFY
To investigate the copy number variation in TIFY during E. ferox-specific WGT events, we compared the homologous TIFY genes in N. colorata and the three E. ferox sub-genomes (Sub1, Sub2, and Sub3) and found most EfTIFYs on the conserved collinear blocks (Supplementary Table S3). Specifically, compared to the PPD and JAZ genes, only the ZIM genes have been retained in three copies. The majority of the ZIM (100%) genes were retained in two or three copies, compared to only 83.33% of the JAZ genes and 50.00% of the PPD genes. It is known that the JAZ and ZIM proteins have many important functions that negatively regulate JA signaling via direct interaction with several transcription factors, according to the gene dosage hypothesis. In addition, the percentage of homeologs varies among the three sub-genomes. Compared to the other group of genes, significantly more TIFY gene homologs were retained in the Sub2 and Sub3 sub-genomes. AKE (ancestral karyotype of E. ferox) (AKE1-12) has been inferred in E. ferox [29,30]. Most of the 23 EfTIFY genes belong to AKE3 (43.4%), followed by AKE5 (21.7%), while only 4% of the EfTIFY genes are assigned to AKE11.

Expansion and Evolution of the TIFY Genes in Plants
Comparative genomic analysis confirmed that E. ferox underwent genome triplication since its divergence from N. colorata [26]. For the analyses of the EfTIFY genes' triplication in N. colorata from E. ferox, the syntenic gene pairs were analyzed. We identified 37 syntenic orthologous gene pairs using the MCScanX program among all of the TIFY proteins of E. ferox. Additionally, 33 syntenic orthologous TIFY gene pairs were identified between E. ferox and N. colorata (Figure 3). Among the syntenic orthologous gene pairs, we found that each NcTIFY gene had two or three E. ferox syntenic orthologous genes, demonstrating that the TIFY genes in E. ferox underwent duplication accompanied by genome triplication. However, the gene number in the E. ferox genome was notably three times lower than that of the N. colorata, indicating the gene loss that occurred during the polyploid process. The visualization of the syntenic orthologous genes among the two species was carried out using the TBtool software ( Figure 3). Among the syntenic orthologous gene pairs between E. ferox and N. colorata, we found more E. ferox TIFY genes in N. colorata chromosomes 1

Expansion and Evolution Pattern of the TIFY Genes in Plants
To investigate the evolution of the TIFY family in the plant kingdom, we selected eight Angiospermae (seven eudicots and one basal angiosperm), one Pteridophyta, and one Bryophyta species for comparative analysis (Figure 4a). These plants have played an important role in evolution, such as polyploidization and duplication events [31][32][33]. The phylogenetic tree showed that TIFY also formed three distinct clades (JAZ, ZIM, and PPD). JAZ and ZIM were found to exist in P. patens, which indicates that these two groups originated from before Bryophyta diverged from Chlorophyta. However, the PPD subfamily was lost in the species of Bryophyta. Meanwhile, we found that no PPD was detected in P. patens. JAZ has a closer relationship with PPD, meaning that JAZ and PPD may share a common evolutionary origin, based on previous reports [18].
Furthermore, the number of TIFY genes in polyploidization plants was higher than that in the other species, and was most likely due to WGDs, which lead to gene family expansion. The TIFY gene family incurred WGT events in E. ferox and B. rapa, accounting for more than other species. Among the TIFY gene family, the JAZ subgroup has an important role in expansion, with the greatest rate and extent of expansion (Figure 4b). Compared to other groups, the number of PPDs is more stable. In summary, we infer a possible evolutionary footprint, with WGD being the main driving force in terms of the expansion and evolution of the TIFY gene family in plant genomes (Figure 4c).

Expansion and Evolution Pattern of the TIFY Genes in Plants
To investigate the evolution of the TIFY family in the plant kingdom, we selected eight Angiospermae (seven eudicots and one basal angiosperm), one Pteridophyta, and one Bryophyta species for comparative analysis (Figure 4a). These plants have played an important role in evolution, such as polyploidization and duplication events [31][32][33]. The phylogenetic tree showed that TIFY also formed three distinct clades (JAZ, ZIM, and PPD). JAZ and ZIM were found to exist in P. patens, which indicates that these two groups originated from before Bryophyta diverged from Chlorophyta. However, the PPD subfamily was lost in the species of Bryophyta. Meanwhile, we found that no PPD was detected in P. patens. JAZ has a closer relationship with PPD, meaning that JAZ and PPD may share a common evolutionary origin, based on previous reports [18].
Furthermore, the number of TIFY genes in polyploidization plants was higher than that in the other species, and was most likely due to WGDs, which lead to gene family expansion. The TIFY gene family incurred WGT events in E. ferox and B. rapa, accounting for more than other species. Among the TIFY gene family, the JAZ subgroup has an important role in expansion, with the greatest rate and extent of expansion (Figure 4b). Compared to other groups, the number of PPDs is more stable. In summary, we infer a possible evolutionary footprint, with WGD being the main driving force in terms of the expansion and evolution of the TIFY gene family in plant genomes (Figure 4c).

Comparative Expression Pattern Analysis of the TIFY Genes between A. thaliana and E. ferox
The TIFY genes play important roles in plant growth involving physiological activity [18]. We investigated the different expression pattern of the TIFY genes in different tissues (roots, stems, leaves, and flowers) between A. thaliana and E. ferox. The seeds and fruits were studied only in E. ferox, and the siliques and mature pollen were studied only in A. thaliana ( Figure 5 and Supplementary Table S7). The transcript levels (TPKM values) of 20 of the 23 EfTIFYs were obtained from at least one of the six tissues (Supplementary Table  S6 and Figure 5). Interestingly, the TPKM values of five EfTIFYs (EfTIFY3.2, EfTIFY11, EfTIFY9.3, EfTIFY7.1, and EfTIFY7.2) in the JAZ subfamily were highly expressed, indicating that they may participate in the development of E. ferox. In addition, EfTIFY8.2 was highly expressed in the stems (Figure 6a). In A. thaliana, AtTIFY8, AtTIFY85b, and At-TIFY10b were highly expressed in the roots. Meanwhile, eight TIFY genes (AtTIFY2a, At-TIFY11a, AtTIFY3a, AtTIFY4a, AtTIFY5a, AtTIFY6a, and AtTIFY6b) were highly expressed in the stems. However, the homologs of EFTIFY were not highly expressed in the stems, such as EfTIFY7.3, EfTIFY10.2, and EfTIFY10.2. Furthermore, the paralogous genes were differently expressed due to the differentiation of function.

Comparative Expression Pattern Analysis of the TIFY Genes between A. thaliana and E. ferox
The TIFY genes play important roles in plant growth involving physiological activity [18]. We investigated the different expression pattern of the TIFY genes in different tissues (roots, stems, leaves, and flowers) between A. thaliana and E. ferox. The seeds and fruits were studied only in E. ferox, and the siliques and mature pollen were studied only in A. thaliana ( Figure 5 and Supplementary Table S7). The transcript levels (TPKM values) of 20 of the 23 EfTIFYs were obtained from at least one of the six tissues (Supplementary  Table S6 and Figure 5). Interestingly, the TPKM values of five EfTIFYs (EfTIFY3.2, EfTIFY11, EfTIFY9.3, EfTIFY7.1, and EfTIFY7.2) in the JAZ subfamily were highly expressed, indicating that they may participate in the development of E. ferox. In addition, EfTIFY8.2 was highly expressed in the stems (Figure 5a). In A. thaliana, AtTIFY8, AtTIFY85b, and AtTIFY10b were highly expressed in the roots. Meanwhile, eight TIFY genes (AtTIFY2a, AtTIFY11a, AtTIFY3a, AtTIFY4a, AtTIFY5a, AtTIFY6a, and AtTIFY6b) were highly expressed in the stems. However, the homologs of EFTIFY were not highly expressed in the stems, such as EfTIFY7.3, EfTIFY10.2, and EfTIFY10.2. Furthermore, the paralogous genes were differently expressed due to the differentiation of function.

Differential Expression of the EfTIFY Genes under Leaf Development
In the present study, we investigated the response of the E. ferox ZML, PPD, and JAZ subfamily genes to different leaf development processes (Figure 6, Supplementary Table  S8). We further explored the gene pathways involved in the development of the early adult leaves (EA1-EA4) and adult leaves (A1-A7) development of E. ferox.

Differential Expression of the EfTIFY Genes under Leaf Development
In the present study, we investigated the response of the E. ferox ZML, PPD, and JAZ subfamily genes to different leaf development processes (Figure 6, Supplementary  Table S8). We further explored the gene pathways involved in the development of the early adult leaves (EA1-EA4) and adult leaves (A1-A7) development of E. ferox.
All EfTIFY genes were lowly expressed at EA1 ( Figure 6). Moreover, three EFTIFY genes (EfTIFY4.1, EfTIFY10.1, and EfTIFY11) were upregulated at EA2. As the leaves grow, more of the EfTIFY genes were highly expressed in EA3 and EA4. Specifically, we observed that the JAZ subgroup was highly expressed in the early adult leaves. For the adult leaves, we found that almost all of the EfTIFY genes had high expression levels at A1-A3, especially the JAZ subfamily, indicating a potentially important function in leaf development processes. In addition, EfTIFY8.4, EfTIFY4.3, EfTIFY4.1, and EfTIFY10.2 were upregulated at A7. Furthermore, the expression pattern of EfTIFYs at critical stages of leaf development was analyzed via qPCR (Figure 7a,b). The expression of EfTIFY9.2 was higher at the mature stage of leaf development (A4,A7), while the expression of EfTIFY8.5 was high in submerged leaves (A1-A3), and the expression of EfTIFY8.4 and EfTIFY3.2 remained high during leaf development. The expression of EfTIFY7.2, EfTIFY10.1 and EfTIFY11 increased with leaf development, while the expression of EfTIFY8.2 increased significantly in A1-A3 and decreased significantly in A3-A7 (Figure 7a, Supplementary  Table S9). Notably, EfTIFY7.2 and EfTIFY10.1 had higher expression in A1 to A7 compared with other genes (Figure 7a, Supplementary Table S9), suggesting that these two genes may have critical roles in the growth and development of E. ferox leaves.

Differential Expression of the EfTIFY Genes under Leaf Development
In the present study, we investigated the response of the E. ferox ZML, PPD, and JAZ subfamily genes to different leaf development processes ( Figure 6, Supplementary Table  S8). We further explored the gene pathways involved in the development of the early adult leaves (EA1-EA4) and adult leaves (A1-A7) development of E. ferox.
All EfTIFY genes were lowly expressed at EA1 ( Figure 6). Moreover, three EFTIFY genes (EfTIFY4.1, EfTIFY10.1, and EfTIFY11) were upregulated at EA2. As the leaves grow, Plant cis-elements are important ways to regulate gene expression to participate in plant growth and development and adapt to the environment. Thus, we used the PlantCare online tool to identify TIFY genes of the cis-regulatory elements in Euryale ferox. Two cis-regulatory elements, ARBE and the CGTCA-motif, the GARE-motif, were responsive to plant hormones, including ABA and JA, in addition to cis-regulatory lightresponsive element (AE-box and G-Box), low-temperature responsiveness (LTR), and drought-responsive element (MBS). These suggested that they could affect TIFY genes expression to regulate development in Euryale ferox ( Figure S2).
In addition, a co-expression network was established for these TIFY genes with a Pearson correlation coefficient of exceed 0.8. EfTIFY7.2 was observed to be significantly correlated with the expression of several genes (EfTIFY11, EfTIFY3.2, EfTIFY8.2, EfTIFY8.4), further confirming that this gene is likely to be a key gene in leaf development. Moreover, EfTIFY11 was associated with several genes and its relative expression was high during leaf development. These results indicated that it is probable that multiple TIFY genes coordinated with each other to promote the growth and development of E. ferox leaves.

Discussion
E. ferox belongs to the basal Nymphaeaceae family, which is evolutionarily important for plants. It has recently been reported that the TIFY gene family plays an important role in plant development, which is tightly linked to stress responses [34]. A genome-wide

Discussion
E. ferox belongs to the basal Nymphaeaceae family, which is evolutionarily important for plants. It has recently been reported that the TIFY gene family plays an important role in plant development, which is tightly linked to stress responses [34]. A genome-wide identification method was performed to identify the TIFY gene in E. ferox and provide clues about its evolutionary history and expression diversification. Meanwhile, the expansion of TIFY in E. ferox and other plants was revealed. Through the expression profile data of different tissues and leaf development stages, it was revealed that the family genes have different tissue expression specificities and participate in E. ferox leaf development.
In this study, we identified 23 TIFY family genes in E. ferox and other species, including nine plants, one moss, one lycophyte, one mocnocotyledonous angiosperm, and six eudicotyledonous angiosperms. A total of 164 TIFY family genes were identified and analyzed in our study. The number of each subgroup in E. ferox is different to that in N. colorata, which is due to differences in replication and retention during evolution. Most of plant species have to undergo polyploidization [35,36]. Polyploidization is a major force in plant adaptive evolution [37,38]. Plants such as White lupin (Lupinus albus) have evolved from a whole-genome triplication (WGT) event. Papaver [39], Acorus tatarinowii [40] and Saccharum spontaneum [41] have also undegone polyploidization. As reported previously, Euryale ferox has also undergone triplication events during the last~17 million years [26]. We found that angiosperms contain a relatively large number of TIFY genes. The TIFY genes exist in land plants, in which the TIFY domain originated after the divergence of algae from land plants and might have been essential in land plant emergence [18,22]. In this study, the number of JAZ subfamily genes in E. ferox was more than that in N. colorata, and the JAZ subfamily genes were preferentially retained relative to the other subfamily genes. Duplication events are the main factor driving gene family expansion in plants [19]. At the same time, gene retention and loss always occur in the evolutionary process, which varied in each TIFY subfamily gene.
The TIFY family is a novel, plant-specific gene family. In plants, the TIFY gene family plays very important roles in development, as well as in hormonal regulation and stress responses [1]. Gene expression patterns can provide important clues for gene function. An alternative splicing form of JAZ10 (JAZ10.4) plays an essential role in the regulation of JA-induced degradation in Arabidopsis thaliana [42]. We speculate that EfJAZ transcription factors may also experience alternative splicing, and have an effect on the stability of the JAZ family and its function. Due to the close relationship between E. ferox and Arabidopsis, highly homologous genes between the two species were identified and used to predict the functions of the TIFY genes in E. ferox. For example, AtTIFY1 (Supplementary Table S6) was reported to be involved in petiole and hypocotyl elongation. These results show that homologous EfTIFY genes may also function in the response to those abiotic and biotic stresses, which needs to be explored in further studies.
Furthermore, via qPCR analysis, we identified that the expression of EfTIFY7.2 and EfTIFY10.1 was closely related to the developmental process of the leaf, and these two genes are likely to play important roles in E. ferox leaf development.

Materials
The Euryale ferox plants were grown in Yangzhou University Aquatic Vegetable Test Base, normal cultivation management. Samples were taken at critical periods of leaf development (A1, A2, A3, A4, A7), snap-frozen in liquid nitrogen and stored at −80 • C in the refrigerator.

Identification of the TIFY Gene Family in Different Plants
The E. ferox genome was downloaded from NCBI [26], and we analyzed the domains using the HMM profile [43] of the TIFY domains (PF06200). The Arabidopsis TIFY sequences were used as the query to perform a BLAST search in these species, with a cutoff e-value of <10 −10 . To confirm the obtained proteins, the Pfam database (http://Pfam. sanger.ac.uk/, 1 January 2023) [44] and the SMART tool (http://smart.embl-heidelberg.de/, 1 January 2023) were used for the examination. All of the TIFY proteins were downloaded from the genome browser phytozome v13 (http://www.phytozome.net/, 1 January 2023) and were used for identification in the Pfam database.

Identification of Syntenic Pairs of TIFY Genes in E. ferox
The Multiple Collinearity Scan toolkit (MCScanX) was used to identify syntenic gene pairs [46]. These syntenic gene pairs were then demonstrated with Circos using TBtool [47].

Gene Expression Data Analysis
To analyze TIFY in different tissues and different leaf development expression patterns in E. ferox, we used previously reported RNA-seq data [26]. Finally, heat maps of the hierarchical clustering were visualized using TBtool [47].

Quantitative Real-Time PCR Analysis
Plant RNA extraction kit (Takara, Dalian, China) was used to extract total RNA from leaf of E. ferox at different developmental stages. Then, HiScript ® IIl RT SuperMixfor qPCR (Vazyme, Nanjing, China) was used to reverse transcription into cDNA. The qRT-PCR reaction was 20 µL, including 10 µL 2 × ChamQ SYBR qPCR Master Mix (Vazyme, Nanjing, China), 0.4 µL forward primer, 0.4 µL reverse primer, 1.0 µL cDNA template and 8.2 µL ddH 2 O, respectively. Primer Premier 5.0 was used for primer design. See Supplementary  Table S1 for the gene specific primer sequences. EfUBQ5-3 was used as an internal gene expression. Amplification was performed on the CFX-96 Real-time PCR system (Bio-Rad, Hercules, CA, USA) using the following real-time fluorescent quantitative PCR program: 95 • C for 30 s, then 95 • C for 10 s, and 60 • C for 30 s for a total of 40 cycles. The relative gene expression was calculated by 2 −∆∆CT [48]. Three replicates were performed for each amplification reaction.

Conclusions
Bioinformatics plays a key role in examining the molecular regulation of E. ferox development, which provides us with basic resources, and the bioinformatics analysis results ensure the creditability of experimental results. In summary, 23 TIFY genes (seven ZIM genes, four PPD genes, and 12 JAZ genes) were identified in the entire E. ferox genome. A comparison of the phylogenetic relationships between E. ferox and Arabidopsis TIFY genes suggested that although most of the basic subfamilies have been retained in E. ferox, the number of each subfamily is different. We further demonstrated that WGD or segmental duplications have contributed to the expansion of the TIFY gene family. Comparative synteny analysis between the E. ferox and N. colorata genomes indicated that the majority of E. ferox and N. colorata TIFY genes are located in syntenic regions. Multiple TIFY genes were identified via qPCR and co-expression analysis as possibly being involved in the leaf development process. These results will lay the foundation for resolving the molecular mechanism of leaf development in E. ferox.