Identification of the 14-3-3 Gene Family in Bamboo and Characterization of Pe14-3-3b Reveals Its Potential Role in Promoting Growth

The 14-3-3 protein family plays an important role in regulating plant growth and development. The genes of the 14-3-3 family have been reported in multiple species. However, little is known about the 14-3-3 gene family in bamboo. In this study, a total of 58 genes belonging to the 14-3-3 family were identified in three representative bamboo species, i.e., Olyra latifolia, Phyllostachys edulis, and Bonia amplexicaulis, whose encoding proteins were grouped into ε and non-ε groups by phylogeny analysis with 14-3-3 proteins from Arabidopsis thaliana and Oryza sativa. The 14-3-3s had diverse gene structures and motif characteristics among the three bamboo species. Collinearity analysis suggested that the genes of the 14-3-3 family in bamboo had undergone a strong purification selection during evolution. Tissue-specific expression analysis showed the expression of Pe14-3-3s varied in different tissues of P. edulis, suggesting that they had functional diversity during growth and development. Co-expression analysis showed that four Pe14-3-3s co-expressed positively with eight ribosomal genes. Yeast two-hybrid (Y2H) assays showed that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6, and qPCR results demonstrated that Pe14-3-3b/d and Pe_ribosome-1/5/6 had similar expression trends with the increase in shoot height, which further confirmed that they would work together to participate in the shoot growth and development of bamboo. Additionally, the transgenic Arabidopsis plants overexpressing Pe14-3-3b had longer roots, a larger stem diameter, an earlier bolting time and a faster growth rate than wild-type Arabidopsis, indicating that Pe14-3-3b acted as a growth promoter. Our results provide comprehensive information on 14-3-3 genes in bamboo and highlight Pe14-3-3b as a potential target for bamboo improvement.


Scaffold Localization and Collinearity Analysis of 14-3-3 Genes
The genome location analysis revealed that 14-3-3 genes were unevenly distributed on the scaffolds. The identified 14-3-3 genes were located in 8, 20 and 13 scaffolds of O. latifolia, P. edulis, and B. amplexicaulis, respectively. To fully elucidate the genetic evolution in bamboo, three typical circular maps were constructed by comparing 14-3-3 genes of P. edulis and other three plants (O. sativa, O. latifolia and B. amplexicaulis). The intraspecific collinearity results showed that 15 14-3-3 orthologous gene pairs were identified in P. edulis, and interspecific collinearity results revealed that 21, 27, and 40 14-3-3 orthologous gene pairs between P. edulis and O. sativa, O. latifolia and B. amplexicaulis (Figure 2a-c, Table S3) were identified, respectively, suggesting that Pe14-3-3s may have existed before the differentiation of the other three species. To explore the selection pressure of the 14-3-3 genes in these four plants, the Ka (non-synonymous substitution)/Ks (synonymous substitution) ratios of duplicated gene pairs were calculated. All Ka/Ks ratios of 14-3-3 orthologous gene pairs within P. edulis, and between P. edulis and the three other species, were less than 1 (Figure 2d, Table S4), indicating that these 14-3-3 genes had undergone a purifying selection during evolution [27].

Tissue Expression Patterns of Pe14-3-3s in Moso Bamboo
To reveal the role of Pe14-3-3s, we investigated their spatiotemporal expression patterns based on the published transcriptome data [28], and an expression heatmap was drawn ( Figure 3). The results showed that Pe14-3-3s exhibited a tissue-specific expression and were further divided into three groups. In group I, 16 Pe14-3-3s were nearly highly expressed in all the samples of moso bamboo and showed different expression patterns.

Tissue Expression Patterns of Pe14-3-3s in Moso Bamboo
To reveal the role of Pe14-3-3s, we investigated their spatiotemporal expression patterns based on the published transcriptome data [28], and an expression heatmap was drawn ( Figure 3). The results showed that Pe14-3-3s exhibited a tissue-specific expression and were further divided into three groups. In group I, 16 Pe14-3-3s were nearly highly expressed in all the samples of moso bamboo and showed different expression patterns. For example, some members presented a constitutive expression pattern, such as Pe14-3-3g and Pe14-3-3i, suggesting that they played an important role in different tissues during growth and development. Some members, such as Pel4-4-3b/c/a/d/f /like1, had similar patterns both in roots and shoots at different growth stages, and were highly expressed in the top part of the roots, shoots and buds, suggesting that these genes played an important role in different growth stages, especially involved in the development of actively differentiated tissues. Meanwhile, Pe14-3-3k and Pe14-3-3l were highly expressed in specific tissues, including rhizome roots with a 0.5 cm length, the tip of bamboo shoots and the buds, suggesting that they might function in particular tissues. Of 36 genes, 5 Pe14-3-3s belonged to group II, which was only expressed in certain tissues and had a very low expression level, such as Pe14-3-3like7, which was expressed relatively higher in rhizome roots with a 10 cm length and leaf sheath. In group III, the remaining 14 Pe14-3-3s, except Pe14-3-3n, could hardly be detected in all tissues, suggesting that they may be functionally redundant genes. could hardly be detected in all tissues, suggesting that they may be functionally redundant genes.

Co-expression Analysis of Pe14-3-3s
To further investigate the co-expression genes of Pe14-3-3s and the metabolic pathways they might be involved in, the prediction of co-expression analysis was performed with 16 Pe14-3-3s (group I family members), which were nearly expressed in all detected tissues. A total of 271 genes were found to be co-expressed with these Pe14-3-3s, in which the genes that participated in the ribosome pathway accounted for a higher proportion in biological processes based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis ( Figure 4a, Table S5). Furthermore, Gene Ontology (GO) enrichment analysis also found that these co-expressed genes participated in the ribosome pathway ( Figure 4b, Table S6). Therefore, we speculated that these Pe14-3-3s may be involved in the ribosome pathway in P. edulis. Further correlation analysis showed six of sixteen Pe14-3-3 genes had a high correlation with most Pe_ribosomes in different height shoots of P. edulis (Figure 4c). Similarly, six of sixteen Pe14-3-3s also showed a great correlation with most Pe_ribosomes (Figure 4d) in the root of P. edulis. Finally, there were four Pe14-3-3s that had a strong correlation with eight Pe_ribosomes, both in the shoots and roots of P. edulis (Pearson's correlation coefficients (PCC) > 0.75, p-value < 0.05)) (Table

Co-expression Analysis of Pe14-3-3s
To further investigate the co-expression genes of Pe14-3-3s and the metabolic pathways they might be involved in, the prediction of co-expression analysis was performed with 16 Pe14-3-3s (group I family members), which were nearly expressed in all detected tissues. A total of 271 genes were found to be co-expressed with these Pe14-3-3s, in which the genes that participated in the ribosome pathway accounted for a higher proportion in biological processes based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis (Figure 4a, Table S5). Furthermore, Gene Ontology (GO) enrichment analysis also found that these co-expressed genes participated in the ribosome pathway ( Figure 4b, Table S6). Therefore, we speculated that these Pe14-3-3s may be involved in the ribosome pathway in P. edulis. Further correlation analysis showed six of sixteen Pe14-3-3 genes had a high correlation with most Pe_ribosomes in different height shoots of P. edulis ( Figure 4c). Similarly, six of sixteen Pe14-3-3s also showed a great correlation with most Pe_ribosomes (Figure 4d) in the root of P. edulis. Finally, there were four Pe14-3-3s that had a strong correlation with eight Pe_ribosomes, both in the shoots and roots of P. edulis (Pearson's correlation coefficients (PCC) > 0.75, p-value < 0.05)) (Table S7). Furthermore, the intragroup correlation analysis of these twelve genes (four Pe14-3-3s and eight Pe_ribosomes) also showed a strong correlation ( Figure S4, Table S8).

Co-expression Network Construction and Validation of the Relationship between Pe14-3-3b/d and Pe_ribosomes
Based on the above analysis, a co-expression network was constructed with four Pe14-3-3s and eight Pe_ribosomes with a high correlation (Figure 5a). To further confirm how Pe14-3-3s played a role in the ribosome pathway, we checked the relationship between Pe14-3-3b/d and Pe_ribosome-1/4/5/6 using a targeted Y2H assay. As shown in Figure 5b, positive controls Pe14-3-3b/d and Pe_ribosome-1/5/6 co-transformed into yeast colonies could grow on both synthetic defined (SD)/-Leu/-Trp plates and synthetic defined (SD)/-Ade/-His/-Leu/-Trp/X-α-Gal plates, while the negative controls and the yeast colonies with other co-transformed genes only grew on the synthetic defined (SD)/-Leu/-Trp plates and did not grow on the SD/-Ade/-His/-Leu/-Trp/X-α-Gal plates. These results indicated that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6 instead of Pe_ribosome-4 in yeast. Moreover, the qPCR results showed that Pe14-3-3b/d and Pe_ribosome-1/5/6 had similar up-regulated expression trends with the increase in bamboo shoot height ( Figure  5c): this further supported the positive relationship between Pe14-3-3b/d with Pe_ribosome-1/5/6. Therefore, we speculated that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6 and they could work together to participate in the ribosomal metabolic pathway and affect the growth and development of P. edulis.

Co-expression Network Construction and Validation of the Relationship between Pe14-3-3b/d and Pe_ribosomes
Based on the above analysis, a co-expression network was constructed with four Pe14-3-3s and eight Pe_ribosomes with a high correlation (Figure 5a). To further confirm how Pe14-3-3s played a role in the ribosome pathway, we checked the relationship between Pe14-3-3b/d and Pe_ribosome-1/4/5/6 using a targeted Y2H assay. As shown in Figure 5b, positive controls Pe14-3-3b/d and Pe_ribosome-1/5/6 co-transformed into yeast colonies could grow on both synthetic defined (SD)/-Leu/-Trp plates and synthetic defined (SD)/-Ade/-His/-Leu/-Trp/X-α-Gal plates, while the negative controls and the yeast colonies with other co-transformed genes only grew on the synthetic defined (SD)/-Leu/-Trp plates and did not grow on the SD/-Ade/-His/-Leu/-Trp/X-α-Gal plates. These results indicated that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6 instead of Pe_ribosome-4 in yeast. Moreover, the qPCR results showed that Pe14-3-3b/d and Pe_ribosome-1/5/6 had similar up-regulated expression trends with the increase in bamboo shoot height (Figure 5c): this further supported the positive relationship between Pe14-3-3b/d with Pe_ribosome-1/5/6. Therefore, we speculated that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6 and they could work together to participate in the ribosomal metabolic pathway and affect the growth and development of P. edulis.

Ectopic Expression of Pe14-3-3b in Arabidopsis
To further elucidate the possible biological roles of Pe14-3-3s, Pe14-3-3b was selected for the recombinant overexpression (OE) vector construction. Pe14-3-3b, driven by 35S, was transformed into Arabidopsis, meditated by Agrobacterium tumefaciens. A total of four hygromycin-resistant T3 plants were generated, of which three homozygous lines (OE-1, OE-2, and OE-3) were selected for further investigation. Under normal conditions, the root length of all transgenic lines was significantly longer than that of the wild type (WT) (Figure 6a-b), especially that of OE-3, wherein the mean root length increased by 58.94% compared with the WT after germination for nine days. Meanwhile, the bolting time of OE-1/2/3 was prior to that of the WT (Figure 6c), and the stem diameter of OE-1/2/3 was significantly larger than that of the WT (Figure 6d). The microscopic observation showed that OE-1/2/3 plants had larger cortical cells than those of the WT (Figure 6e). Additionally, the size of rosette leaves was bigger than those of the WT ( Figure S5). Furthermore, qPCR analysis showed that the expression levels of Pe14-3-3b were significantly higher in OE-1/2/3 than that in the WT, especially in OE-3, which was 36.51 times of that in the WT. Similarly, compared with the WT, the expression levels of AtRPL30e, AtRPL4 and AtRPLS4A, which were homologous genes of Pe_ribosome-1/5/6, respectively, showed higher expression levels in OE-1/2/3 than those in the WT (Figure 6f). These results indicated that overexpressing Pe14-3-3b increased the expression of Pe_ribosomes, resulting in a promotion of the growth and development of transgenic Arabidopsis.

Ectopic Expression of Pe14-3-3b in Arabidopsis
To further elucidate the possible biological roles of Pe14-3-3s, Pe14-3-3b was selected for the recombinant overexpression (OE) vector construction. Pe14-3-3b, driven by 35S, was transformed into Arabidopsis, meditated by Agrobacterium tumefaciens. A total of four hygromycin-resistant T3 plants were generated, of which three homozygous lines (OE-1, OE-2, and OE-3) were selected for further investigation. Under normal conditions, the root length of all transgenic lines was significantly longer than that of the wild type (WT) (Figure 6a-b), especially that of OE-3, wherein the mean root length increased by 58.94% compared with the WT after germination for nine days. Meanwhile, the bolting time of OE-1/2/3 was prior to that of the WT (Figure 6c), and the stem diameter of OE-1/2/3 was significantly larger than that of the WT (Figure 6d). The microscopic observation showed that OE-1/2/3 plants had larger cortical cells than those of the WT (Figure 6e). Additionally, the size of rosette leaves was bigger than those of the WT ( Figure S5). Furthermore, qPCR analysis showed that the expression levels of Pe14-3-3b were significantly higher in OE-1/2/3 than that in the WT, especially in OE-3, which was 36.51 times of that in the WT. Similarly, compared with the WT, the expression levels of AtRPL30e, AtRPL4 and AtRPLS4A, which were homologous genes of Pe_ribosome-1/5/6, respectively, showed higher expression levels in OE-1/2/3 than those in the WT (Figure 6f). These results indicated that overexpressing Pe14-3-3b increased the expression of Pe_ribosomes, resulting in a promotion of the growth and development of transgenic Arabidopsis.

Discussion
The 14-3-3 protein family, as a class of important factors for plant growth and development, have been widely studied in many eukaryotes [29]. Although the 14-3-3 gene family have been identified in multiple species, such as O. sativa, barley (Hordeum vulgare), soybean (Glycine max) and Arabidopsis [6,[30][31][32], the molecular characteristics and evolution of 14-3-3 proteins in bamboo have not yet been studied. Therefore, the present study was conducted to identify the 14-3-3 gene family in O. latifolia, P. edulis and B. amplexicaulis, which represented diploid, tetraploid and hexaploid bamboo species. Based on the comprehensive analysis of the 14-3-3 gene family in bamboo, the function of Pe14-3-3b was further validated.

Characteristics Diversity of 14-3-3 Genes in Bamboos
In the present study, 36 Pe14-3-3s were identified in P. edulis, which was more than that identified in O. latifolia (8) and B. amplexicaulis (14) (Table S1). This might be due to the fact that the P. edulis genome had undergone gene duplication events [21]. The physical and chemical characteristics of 14-3-3s in different bamboo species indicated that they were diverse. For example, the amino acid composition of some members of Pe14-3-3s was acidic and some were alkaline, while the amino acid composition of both O. latifolia and B. amplexicaulis was acidic. Likewise, Pe14-3-3s had both hydrophilic and hydrophobic members, but the members of both O. latifolia and B. amplexicaulis were hydrophobic (Table S1). The structural diversity of exons/introns was generally considered to provide further insights into structural, evolutional and functional relationships [33]. In P. edulis, more than 60% of the members contained three or four introns ( Figure S1a). Similar results were found in O. latifolia and B. amplexicaulis (Figures S2a and S3a).
In addition, the number of exons and introns in different branches were variable ( Figure S1a, S2a), suggesting that evolution may be driving this structural diversity. This was also observed in O. sativa, Medicago truncatula and G. max [32,34,35]. Additionally, the intron lengths and arrangements were different in Pe14-3-3s, and similar results were also found in O. latifolia and B. amplexicaulis, further suggesting that the exon-intron structure could reveal the evolutionary diversity of the 14-3-3 gene family [33,36]. Furthermore, the conserved motifs of 14-3-3s in O. latifolia, P. edulis and B. amplexicaulis were highly variable ( Figures S1b, S2b and S3b). The diverse motifs in the 14-3-3 family were the key structures of 14-3-3 proteins that could bind to many ligands [37], which directly affects the interaction of 14-3-3 proteins with other proteins to play a variety of functions.

Phylogeny and Evolution of 14-3-3 Proteins
Previously, genome duplication studies of 141 sequenced plant genomes demonstrated that, compared with A. thaliana, P. edulis was easier to evolve with O. sativa on shorter branches [28,38], and the relationship between P. edulis and O. sativa, O. latifolia and B. amplexicaulis was closer [39]. In our study, 37 14-3-3s from O. latifolia, P. edulis and B. amplexicaulis were clustered into ε and non-ε groups by phylogenetic analysis (Figure 1), which was consistent with previous reports on Arabidopsis, wheat (Triticum aestivum) and G. soja [7,30,40]. Our study also suggested the same phenomenon that most members of Pe14-3-3s, along with 14-3-3s in O. sativa, O. latifolia and B. amplexicaulis, clustered in close branches, while the evolutionary distance between P. edulis and A. thaliana was relatively far. It was crucial for us to infer the phylogeny and evolutionary direction of the 14-3-3 gene family between P. edulis and other plants (O. sativa, O. latifolia and B. amplexicaulis).
During evolution, plants undergo gene duplication events, among which wholegenome duplication and tandem duplication often promote the expansion of gene families [41,42], resulting in the diversity of gene function [43]. A previous study showed that P. edulis had undergone at least one round of whole-genome duplication, followed by multiple-segment duplication [21], which was considered to be the cause of the expansion of 14-3-3 family members in P. edulis during evolution. The differences of 14-3-3 gene members among three bamboo species might be mainly caused by the occurrence of genome-replication events [44], including series replication, such as fragment replica-tion [45]. In our study, gene duplication and syntenic analyses of 14-3-3s supported this assumption as 15 segmental duplication pairs were found within Pe14-3-3s, and 21 pairs were found between Pe14-3-3s and Os14-3-3s (Figure 2a). Similarly, 27 and 40 pairs were found between Pe14-3-3s and Ol14-3-3s, and Ba14-3-3s, respectively (Figure 2c,d). In addition, the Ka/Ks ratio can be used to indicate the selecting direction of one gene and measure the historical choice of coding sequences [44][45][46][47]. In our study, the Ka/Ks ratios (Figure 2d, Table S4) showed that Pe14-3-3s had undergone a purification selection after duplication with limited functional divergence [27], which provided a better insight into the evolution of the 14-3-3 gene family.

Multiple Functions of 14-3-3s in Plant Growth and Development
The 14-3-3s are of vital importance during plant growth and development, which mainly manifests in reproductive and vegetative growth [4]. The function in reproductive and vegetative growth and expression patterns of 14-3-3s have been studied in many plant species [48][49][50][51][52], but the possible function of Pe14-3-3s remains unclear. Gene expression patterns are an important manifestation of gene function. The expression patterns of Pe14-3-3s in various organs, and at different plant growth stages in moso bamboo (Figure 3), suggested that they had different functions, which was consistent with previous studies [53,54]. In line with our results shown in Figure 6a,b, the overexpression of Pe14-3-3b in Arabidopsis showed a longer root length than WT plants. Usually, shoot and stem diameter correspond to the number of cortical cells, and the increased number of cortical cells could improve shoot and stem growth [55]. The Pe14-3-3b transgenic Arabidopsis plants showed a larger stem diameter than WT plants (Figure 6d), which was supported by the larger size, instead of the number of cortical cells in transgenic Arabidopsis plants. Therefore, we deduced that Pe14-3-3b may improve shoot growth by increasing the size of cortical cells. The 14-3-3 protein family could also contribute to the development of leaves [51]. Similar results were found in our study ( Figure S5): an overexpression of Pe14-3-3b in Arabidopsis led to the formation of bigger rosette leaves than WT plants, suggesting that Pe14-3-3b could accelerate the development of rosette leaves.
Furthermore, 14-3-3s could combine with multiple proteins to perform their functions in plant growth and development. Proteomic analyses revealed that 14-3-3 proteins were ribosomal-related [17]. Our study confirmed that Pe14-3-3b/d could interact with Pe_ribosome-1/5/6, among which Pe_ribosome-1 was the homologue of 40S ribosomal protein (AT5G58420) with an important role in seed germination and seedling transition [56]. The qPCR results showed that, with the increase in shoot height, the expression levels of Pe14-3-3b/d and Pe_ribosome-1/5/6 all showed an up-regulated trend, consistent with previous studies, wherein 14-3-3 proteins could up-regulate the ribosome biogenesis [57]. Meanwhile, the expression level of Pe_ribosome-4 showed a trend of increasing first, and then decreasing with the increase in shoot height, which may be a cause of its inability to interact with 14-3-3 proteins. Moreover, higher expression levels of Pe14-3-3b and Pe_ribosomes homologue genes were found in the transgenic Arabidopsis than those in WT plants (Figure 6f), which further suggested that these Pe_ribosomes work together with Pe14-3-3b to play an important role in the growth and development of moso bamboo. However, how the Pe14-3-3b is involved in ribosome pathways to regulate the development of moso bamboo shoots needs to be further studied.

Plant Materials
The moso bamboo (Phyllostachys edulis) samples were taken from the Jiangxi Academy of Forestry; the 15th internode was collected from representative shoots with heights of 0.5 m, 1.0 m, 4.0 m and 6.0 m. After treating with liquid nitrogen, the samples were stored in the refrigerator at −80 • C for further processing.

Scaffold Localization and Collinearity Analysis of 14-3-3 Genes
The scaffold distribution and syntenic relationships of the 14-3-3 genes of P. edulis, O. sativa, O. latifolia and B. amplexicaulis were obtained by MCScanX and visualized using TBtools [66]. The TBtools software was also used to calculate the Ka (non-synonymous substitution)/Ks (synonymous substitution) ratios of paralogous pairs to deduce selective pressure.

Co-expression, KEGG, GO and Correlation Analysis of Pe14-3-3s
The BambooNET (http://bioinformatics.cau.edu.cn/bamboo/index.html/ (accessed on 23 August 2021)) [67] was used to carry out a co-expression prediction analysis of Pe14-3-3s. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses were performed on an online website (https://www.omicshare.com/tools/ (accessed on 24 August 2021)) for genes co-expressed with Pe14-3-3s. Based on the RNA-seq data of moso bamboo in different tissues (roots with different lengths and shoots with different heights), the correlation analysis was conducted by calculating pairwise Pearson's correlation coefficients (PCC) and p-values on an online website (https://www.omicshare.com/tools/ (accessed on 24 August 2021)) [68] between Pe14-3-3s and Pe_ribosomes: the results were visualized with TBtools [66]. In addition, a co-expression network was constructed using Cytoscape 3.7.1 [69] based on the results of the intra-group correlation analysis of Pe14-3-3s and Pe_ribosomes. PCC > 0.75 and p-values < 0.05 were set in the co-expression relationship analysis.

RNA Extraction and qPCR Analysis
The total RNA of the samples from moso bamboo was extracted using RNA Plus (Jianshi, Shanghai, China) following the instructions, and was reverse transcribed into cDNA with a PrimeScript™ RT Reagent Kit (TaKaRa, Kyoto, Japan) according to the manufacturer's instructions. The qPCR experiments were performed using SYBR Green chemistry (Roche, Mannheim, Germany) on a qTOWER 2.2 system (Analytik, Jena, Germany) according to the manufacturer's directions. PeTIP41 was used as a reference gene for moso bamboo [70] to calculate the relative expression of the selected genes using the 2 −∆∆CT method [71]. The specific primers were designed by the Primer 5.0 software (Jin Wang, Soochow, China) (Table S9).

Y2H Assay
Firstly, the full-length coding sequences of Pe14-3-3b/d and Pe_ribosome-1/4/5/6 were obtained using specific primers (Table S9) and cloned into pGBKT7 and pGADT7 vectors to form recombinant plasmids pGBKT7-Pe14-3-3s and pGADT7-Pe_ribosomes, respectively. According to the Yeast Protocols Handbook (Clontech, Mountain View, CA, USA), these combination constructs, including the positive controls of pGBKT7-53 and pGADT7-T, negative controls of pGBKT7-lam and pGADT7-T7 and the experimental groups of pGBKT7-Pe14-3-3s and pGADT7-Pe_ribosomes, were co-transformed into the yeast strain Y2HGold (Saccharomyces cerevisiae) and then plated on an SD solid-selection medium, including an SD/-Leu/-Trp medium and an SD/-Leu/-Trp/-His/-Ade/X-α-Gal medium, which were incubated at 30 • C until the appearance of colonies. Photographs were taken to record the growth of yeast colonies.

Transformation and Validation of Transgenic Arabidopsis Plants
The obtained CDS sequence of Pe14-3-3b was cloned into the overexpression (OE) vector Super1300 to form recombinant expression vector Super1300-Pe14-3-3b. The Super1300-Pe14-3-3b vector was introduced into the Agrobacterium tumefaciens strain GV3101 for Arabidopsis transformation using the floral dipping method [72]. Positive T1 transgenic plants were identified by PCR analysis and selected on a 1/2 MS solid medium with hygromycin (50 mg/mL); homozygous T3 seeds were used for subsequent experiments.
The seeds of wild-type (WT) Arabidopsis and transgenic Arabidopsis were surfacesterilized and seeded on a 1/2 MS solid medium. After vernalization, kept at 4 • C for two days, the seeds were transferred to an artificial climate chamber (temperature 23 • C, humidity 60-80%, light duration 16 h). About seven days later, the seedlings were transplanted into the nutrient soil (humus: vermiculite: 7:3) for further cultivation and subsequent experiment. For the measurement of root length, the seeds of transgenic lines and the wild type were transformed into vertical plates after germination for three days. The leaf phenotype was observed before bolting (around three weeks after germination). The stem diameter of all transgenic lines and WT plants were determined after germination for 30 days. Three technical and biological replicates were performed for all experimental data. The cDNA of transgenic lines and WT plants were used as templates for Pe14-3-3b expression analysis using AtActin2 as an internal reference [73]. The primer sequences are listed in Table S9.

Statistical Analysis
The statistical analyses were performed using IBM SPSS Statistics 22.0 (Armonk, NY, USA), and the mean and standard deviation of three biological replicates were presented. Significant differences were indicated at * p < 0.05, ** p < 0.01. For the measurement of the root length, 40 seedlings were statistically analyzed. For the measurement of the stem diameter, 30 seedlings were statistically analyzed.

Conclusions
In this study, we identified 8, 36 and 14 genes encoding 14-3-3 proteins from O. latifolia, P. edulis and B. amplexicaulis, respectively. The analyses of gene structures and conserved motifs indicated that 14-3-3s were diverse. Differentially expressed patterns of Pe14-3-3s in different tissues of P. edulis supported their diversity. A co-expression network of Pe14-3-3s revealed that most Pe14-3-3s were ribosomal-related. Pe14-3-3b/d could interact with Pe_ribosome-1/5/6 in yeast two-hybrid assays, and they had similar expression patterns in shoots with different heights, which suggested that Pe14-3-3s worked together with Pe_ribosomes in the growth of P. edulis. Moreover, the overexpression of Pe14-3-3b resulted in the promotion of Arabidopsis growth, which further supported that Pe14-3-3b might play an important role in the rapid growth of P. edulis.