Genome-Wide Analysis of BpYABs and Function Identification Involving in the Leaf and Silique Development in Transgenic Arabidopsis

YABs play an important role in the leaf development of the paper mulberry (Broussonetia papyrifera) and of the heterophylly. Thus, we investigated the function of BpYABs. Gene cloning, phylogenetic analysis, motif identification, subcellular localization, transactivation activity assay, qRT-PCR, in situ hybridization, and ectopic expression were used in our study. Six BpYABs were isolated, and four of them had transcriptional activity. BpYAB1, BpYAB3, BpYAB4, and BpYAB5 were localized to the nucleus. BpYAB1 was only expressed in the flower, while BpYAB6 was not expressed in any detected tissues; the four remaining BpYABs were expressed in the bud, leaf and flower, and their expression level decreased with leaf development. Further in situ hybridization showed that BpYAB3 and BpYAB5 were expressed in the vascular tissues and lamina, but neither showed the adaxial–abaxial polarity distribution pattern in the mature leaf lamina. Ectopic expression of BpYAB2, BpYAB3, BpYAB4 and BpYAB5 induced increased expression of AtWOX1 and caused the leaf of Arabidopsis to become smaller and curl downwards. Ectopic expression also led to shorter siliques and smaller seeds, but not for BpYAB5. These results suggest that BpYABs have functional divergency and redundancy in regulating leaf and silique development.

In gymnosperms, the YABBYs are grouped into four clades, called A, B, C, and D, while they are divided into five subfamilies in angiosperms, named FIL-like, YAB2like, YAB5-like, CRC-like, and INO-like [12]. Reports from core eudicots suggested the evolution of the YABBY family coincides with the origin of leaves in seed plants, and YABBY activity is required for initiating the lamina and maintenance of leaf polarity, not the initial establishment of polarity [13]. Hence, it is necessary to explore the evolutionary and functional differences of YABBY genes in plant species. In order to explore the evolutionary relationships among YABBY genes, a neighborjoining tree based on 17 species, including 127 protein sequences, was constructed ( Figure 1). All species were from dicotyledons, including the base species of angiosperms, Amborella trichopoda, and the protein sequences are displayed in Table S1. The results indicated that all members of YABBY were divided into five clades, named FIL-like, INO-like, CRC-like, YAB2-like and YAB5-like. There were five YABBY genes in A. trichopoda belonging to the five subfamilies (Figure 1), indicating that YABBY members from other species in the five subfamilies might have originated from these five YABBY genes in A. trichopoda. Six YABBY members of paper mulberry were also categorized into five subfamilies. Both BpYAB3 and BpYAB4 belonged to the FIL-like clade, while BpYAB2 and BpYAB5 belonged to the YAB2-like and YAB5-like clades, respectively. These four BpYAB genes may participate in leaf development according to the function of their homologous gene in Arabidopsis. BpYAB1 belonged to the CRC-like clade, implying that BpYAB1 might be involved in flower development. BpYAB6 belonged to the INO-like clade, suggesting that the function of BpYAB6 might be associated with ovule development (Figure 1). The number of YABBY genes in the selected woody plants, Prunus persica, Ziziphus jujuba, paper mulberry, M. notabilis and Eucalyptus grandis, was five or six, indicating that YABBY members in woody plants did not undergo expansion during the species diversification and evolution. Among the five clades, the number of FIL-like groups was highest, at 39, and there were at least two members in each species except A. trichopoda. The numbers of the remaining four groups were 28,23,19, and 18, belonging to the YAB5-like, YAB2-like, CRC-like, and INO-like groups, respectively. It was interesting to find that both the INO-like and CRClike clades contained only one gene in each species, implying that these two groups did not undergo expansion. The FIL-like, YAB2-like, and YAB5-like groups had obviously undergone expansion, but the expansion rate was relatively slow. We found that the YABBY genes in mulberry (M. notabilis) displayed a close relationship with paper mulberry (Figure 1). The number of YABBY genes in A. thaliana, paper mulberry, and M. notabilis was the same. BpYABs clustered into five groups by phylogenetic analysis. The phylogenetic tree of YABBYs is divided into five clades, named FIL-like, CRC-like, INO-like, YAB2-like and YAB5-like, containing 127 amino acid sequences from 17 species. The paper mulberry (red filled circle) has two members in the FIL-like group, and other four groups contain only one member. The YABBYs with the green boxes are from A. thaliana and the red circles are from A. trichopoda. All the YABBY proteins were aligned by MUSCLE in MEGA 7.0 and adjusted manually.

Sequence Analysis and Structural Characterization of BpYAB Members
The BpYAB gene sequences are displayed in Table S2. The length of BpYAB genes varied from 1 to 7 kb. The longest one was BpYAB2 with 6482 bp, while the shortest was BpYAB6, with 1197 bp. Although their genomic length was over a large span, the coding sequences were all around 600 bp, and the protein length was about 200 amino acids (Table 1). BpYAB2 and BpYAB6 had six exons, and the remaining BpYABs all had seven exons ( Figure S2a). The lengths of exon 4 and exon 5, which were 49 and 76 bp, were the same in all BpYABs (Table S3). Each of the BpYAB proteins contained two highly conserved domains, the C 2 C 2 and YABBY domains, and their positions were relatively consistent among the different BpYAB members ( Figure S2b). In the nucleotide sequence coding, the C 2 C 2 domain was distributed in part of exon 1 and exon 2, and the YABBY domain spread in exon 4, exon 5, and exon 6 ( Figure S2a). The C 2 C 2 domain of BpYAB proteins contained 43 amino acids, including 16 highly conserved amino acids. The length of the YABBY domain was 56 amino acids, including 28 highly conserved amino acids. The conserved amino acids are marked with asterisks in Figure S2c,d.
Besides, we focused on three groups of these cis-elements related to development, hormone signaling, and environment response ( Table 2). The RY-element, involved in seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions in cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormone-related cis-element, indicating that the expression of BpYAB may be regulated by these hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were predicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate these two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, and BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibberellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA)-responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN-motif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses. ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUNmotif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ responsive element, the TCA-element, was found in BpYAB2 and Bp ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ responsive element, the TCA-element, was fou ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box ellin (GA)-responsive elements and were found in four BpYABs. The salicyli responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. So ers also contained several abiotic stress response elements. For instance, all BpY for BpYAB1 had the ARE cis-element, which is essential for anaerobic induc low-temperature-responsive cis-regulator, was only found in BpYAB2. As for motif, the wound-responsive element was located in BpYAB5 specifically (Tab case, the cis-element analysis illustrated that BpYAB genes participated in pla ment, hormone signaling, and response to environmental stresses. BpYAB4, and seven of ABRE were in BpYAB1 p ellin (GA)-responsive elements and were foun responsive element, the TCA-element, was fou ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi specifically, which implied that auxins may regulate these two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, and BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibberellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA)responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUNmotif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may re two genes. ABRE, the ABA-responsive element, was found in BpYAB1, Bp BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box ellin (GA)-responsive elements and were found in four BpYABs. The salicyli responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. So ers also contained several abiotic stress response elements. For instance, all BpY for BpYAB1 had the ARE cis-element, which is essential for anaerobic induc low-temperature-responsive cis-regulator, was only found in BpYAB2. As for motif, the wound-responsive element was located in BpYAB5 specifically (Tab case, the cis-element analysis illustrated that BpYAB genes participated in pla ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ dicted in BpYAB4 and BpYAB2 specifically, which implied that auxi two genes. ABRE, the ABA-responsive element, was found in B BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and ellin (GA)-responsive elements and were found in four BpYABs. T responsive element, the TCA-element, was found in BpYAB2 and Bp ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses. hormones. The auxin-related cis-regulators, AuxRR-core and TGA dicted in BpYAB4 and BpYAB2 specifically, which implied that auxi two genes. ABRE, the ABA-responsive element, was found in B BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and ellin (GA)-responsive elements and were found in four BpYABs. T responsive element, the TCA-element, was found in BpYAB2 and Bp ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses. related cis-element, indicating that the expression of BpYAB may be regulated by these hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were predicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate these two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, and BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibberellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA)responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUNmotif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses. seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormon related cis-element, indicating that the expression of BpYAB may be regulated by the hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were pr dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate the two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, an BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibbe ellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically ( Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses.

Cis-Element
Function AuxRR-core auxin responsiveness ◆ seed-specific regulation, was found only in BpYAB3, and MSA-lik cell cycle regulation, was specific to BpYAB6. All the BpYABs had a related cis-element, indicating that the expression of BpYAB may b hormones. The auxin-related cis-regulators, AuxRR-core and TGA dicted in BpYAB4 and BpYAB2 specifically, which implied that auxi two genes. ABRE, the ABA-responsive element, was found in B BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and ellin (GA)-responsive elements and were found in four BpYABs. T responsive element, the TCA-element, was found in BpYAB2 and Bp ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses.

Cis-Element
Function AuxRR-core auxin responsiveness ◆ seed-specific regulation, was found only in Bp cell cycle regulation, was specific to BpYAB6. A related cis-element, indicating that the express hormones. The auxin-related cis-regulators, A dicted in BpYAB4 and BpYAB2 specifically, wh two genes. ABRE, the ABA-responsive eleme BpYAB4, and seven of ABRE were in BpYAB1 p ellin (GA)-responsive elements and were foun responsive element, the TCA-element, was fou ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi hormone signaling, and environment response ( Table 2). The RY-element, involved seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormon related cis-element, indicating that the expression of BpYAB may be regulated by the hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were pr dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate the two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, an BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibbe ellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses.

Cis-Element
Function AuxRR-core auxin responsiveness ◆ hormone signaling, and environment response (Table 2 seed-specific regulation, was found only in BpYAB3, and cell cycle regulation, was specific to BpYAB6. All the BpY related cis-element, indicating that the expression of BpY hormones. The auxin-related cis-regulators, AuxRR-core dicted in BpYAB4 and BpYAB2 specifically, which implied two genes. ABRE, the ABA-responsive element, was fo BpYAB4, and seven of ABRE were in BpYAB1 promoter. P ellin (GA)-responsive elements and were found in four B responsive element, the TCA-element, was found in BpYA ers also contained several abiotic stress response elements for BpYAB1 had the ARE cis-element, which is essential low-temperature-responsive cis-regulator, was only foun motif, the wound-responsive element was located in BpYA case, the cis-element analysis illustrated that BpYAB gene ment, hormone signaling, and response to environmental sition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to developmen hormone signaling, and environment response ( Table 2). The RY-element, involved seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormon related cis-element, indicating that the expression of BpYAB may be regulated by the hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were pr dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate the two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, an BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibbe ellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically (Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses. that BpYABs participates in these same regulatory processes, but the quantitative comp sition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to developmen hormone signaling, and environment response ( Table 2). The RY-element, involved seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormon related cis-element, indicating that the expression of BpYAB may be regulated by the hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were pr dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate the two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, an BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibbe ellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically ( Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses.

Cis-Element
Function AuxRR-core auxin responsiveness ◆ that BpYABs participates in these same regulatory processes, but the quantita sition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to de hormone signaling, and environment response ( Table 2). The RY-element, seed-specific regulation, was found only in BpYAB3, and MSA-like, which f cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least on related cis-element, indicating that the expression of BpYAB may be regulat hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may re two genes. ABRE, the ABA-responsive element, was found in BpYAB1, Bp BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box ellin (GA)-responsive elements and were found in four BpYABs. The salicyli responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. So ers also contained several abiotic stress response elements. For instance, all BpY for BpYAB1 had the ARE cis-element, which is essential for anaerobic induc low-temperature-responsive cis-regulator, was only found in BpYAB2. As for motif, the wound-responsive element was located in BpYAB5 specifically (Tab case, the cis-element analysis illustrated that BpYAB genes participated in pla ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ that BpYABs participates in these same regulatory processes, but th sition was different, indicating their ability to function was differen Besides, we focused on three groups of these cis-elements rel hormone signaling, and environment response ( Table 2). The RY seed-specific regulation, was found only in BpYAB3, and MSA-lik cell cycle regulation, was specific to BpYAB6. All the BpYABs had a related cis-element, indicating that the expression of BpYAB may b hormones. The auxin-related cis-regulators, AuxRR-core and TGA dicted in BpYAB4 and BpYAB2 specifically, which implied that auxi two genes. ABRE, the ABA-responsive element, was found in B BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and ellin (GA)-responsive elements and were found in four BpYABs. T responsive element, the TCA-element, was found in BpYAB2 and Bp ers also contained several abiotic stress response elements. For instan for BpYAB1 had the ARE cis-element, which is essential for anaero low-temperature-responsive cis-regulator, was only found in BpYA motif, the wound-responsive element was located in BpYAB5 specif case, the cis-element analysis illustrated that BpYAB genes participa ment, hormone signaling, and response to environmental stresses. AuxRR-core auxin responsiveness ◆ that BpYABs participates in these same regulatory process sition was different, indicating their ability to function wa Besides, we focused on three groups of these cis-ele hormone signaling, and environment response (Table 2 seed-specific regulation, was found only in BpYAB3, and cell cycle regulation, was specific to BpYAB6. All the BpY related cis-element, indicating that the expression of BpY hormones. The auxin-related cis-regulators, AuxRR-core dicted in BpYAB4 and BpYAB2 specifically, which implied two genes. ABRE, the ABA-responsive element, was fo BpYAB4, and seven of ABRE were in BpYAB1 promoter. P ellin (GA)-responsive elements and were found in four B responsive element, the TCA-element, was found in BpYA ers also contained several abiotic stress response elements for BpYAB1 had the ARE cis-element, which is essential low-temperature-responsive cis-regulator, was only foun motif, the wound-responsive element was located in BpYA case, the cis-element analysis illustrated that BpYAB gene ment, hormone signaling, and response to environmental AuxRR-core auxin responsiveness that BpYABs participates in these same regulat sition was different, indicating their ability to f Besides, we focused on three groups of t hormone signaling, and environment respons seed-specific regulation, was found only in Bp cell cycle regulation, was specific to BpYAB6. A related cis-element, indicating that the express hormones. The auxin-related cis-regulators, A dicted in BpYAB4 and BpYAB2 specifically, wh two genes. ABRE, the ABA-responsive eleme BpYAB4, and seven of ABRE were in BpYAB1 p ellin (GA)-responsive elements and were foun responsive element, the TCA-element, was fou ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi POX1, GT1CONSENSUS, ARR1AT and CTRMCAMV35S is the majority, which show that BpYABs participates in these same regulatory processes, but the quantitative comp sition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to developmen hormone signaling, and environment response ( Table 2). The RY-element, involved seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormon related cis-element, indicating that the expression of BpYAB may be regulated by the hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were pr dicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate the two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, an BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibbe ellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promo ers also contained several abiotic stress response elements. For instance, all BpYABs exce for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUN motif, the wound-responsive element was located in BpYAB5 specifically ( Table 2). In an case, the cis-element analysis illustrated that BpYAB genes participated in plant develo ment, hormone signaling, and response to environmental stresses.  (17), and so on. It can be seen from the above statistics that the number of cis-elements such as CACTFTPPCA1, DOFCOREZM, ROOTMOTIFTA-POX1, GT1CONSENSUS, ARR1AT and CTRMCAMV35S is the majority, which shows that BpYABs participates in these same regulatory processes, but the quantitative composition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to development, hormone signaling, and environment response ( Table 2). The RY-element, involved in seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions in cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormonerelated cis-element, indicating that the expression of BpYAB may be regulated by these hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were predicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate these two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, and BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibberellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA)responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUNmotif, the wound-responsive element was located in BpYAB5 specifically ( Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses.  (17), and so on. It the number of cis-elements such as CACTFTP POX1, GT1CONSENSUS, ARR1AT and CTRM that BpYABs participates in these same regulat sition was different, indicating their ability to f Besides, we focused on three groups of t hormone signaling, and environment respons seed-specific regulation, was found only in Bp cell cycle regulation, was specific to BpYAB6. A related cis-element, indicating that the express hormones. The auxin-related cis-regulators, A dicted in BpYAB4 and BpYAB2 specifically, wh two genes. ABRE, the ABA-responsive eleme BpYAB4, and seven of ABRE were in BpYAB1 p ellin (GA)-responsive elements and were foun responsive element, the TCA-element, was fou ers also contained several abiotic stress respons for BpYAB1 had the ARE cis-element, which is low-temperature-responsive cis-regulator, was motif, the wound-responsive element was locat case, the cis-element analysis illustrated that Bp ment, hormone signaling, and response to envi  (17), and so on. It can be see the number of cis-elements such as CACTFTPPCA1, DO POX1, GT1CONSENSUS, ARR1AT and CTRMCAMV35 that BpYABs participates in these same regulatory process sition was different, indicating their ability to function wa Besides, we focused on three groups of these cis-ele hormone signaling, and environment response (Table 2 seed-specific regulation, was found only in BpYAB3, and cell cycle regulation, was specific to BpYAB6. All the BpY related cis-element, indicating that the expression of BpY hormones. The auxin-related cis-regulators, AuxRR-core dicted in BpYAB4 and BpYAB2 specifically, which implied two genes. ABRE, the ABA-responsive element, was fo BpYAB4, and seven of ABRE were in BpYAB1 promoter. P ellin (GA)-responsive elements and were found in four B responsive element, the TCA-element, was found in BpYA ers also contained several abiotic stress response elements for BpYAB1 had the ARE cis-element, which is essential low-temperature-responsive cis-regulator, was only foun motif, the wound-responsive element was located in BpYA case, the cis-element analysis illustrated that BpYAB gene ment, hormone signaling, and response to environmental  (17), and so on. It can be seen from the above statistics that the number of cis-elements such as CACTFTPPCA1, DOFCOREZM, ROOTMOTIFTA-POX1, GT1CONSENSUS, ARR1AT and CTRMCAMV35S is the majority, which shows that BpYABs participates in these same regulatory processes, but the quantitative composition was different, indicating their ability to function was different. Besides, we focused on three groups of these cis-elements related to development, hormone signaling, and environment response ( Table 2). The RY-element, involved in seed-specific regulation, was found only in BpYAB3, and MSA-like, which functions in cell cycle regulation, was specific to BpYAB6. All the BpYABs had at least one hormonerelated cis-element, indicating that the expression of BpYAB may be regulated by these hormones. The auxin-related cis-regulators, AuxRR-core and TGA-element, were predicted in BpYAB4 and BpYAB2 specifically, which implied that auxins may regulate these two genes. ABRE, the ABA-responsive element, was found in BpYAB1, BpYAB3, and BpYAB4, and seven of ABRE were in BpYAB1 promoter. P-box and TATC-box are gibberellin (GA)-responsive elements and were found in four BpYABs. The salicylic acid (SA)responsive element, the TCA-element, was found in BpYAB2 and BpYAB5. Some promoters also contained several abiotic stress response elements. For instance, all BpYABs except for BpYAB1 had the ARE cis-element, which is essential for anaerobic induction. LTR, a low-temperature-responsive cis-regulator, was only found in BpYAB2. As for the WUNmotif, the wound-responsive element was located in BpYAB5 specifically ( Table 2). In any case, the cis-element analysis illustrated that BpYAB genes participated in plant development, hormone signaling, and response to environmental stresses.

The Conserved Motif Identification of BpYAB Family
The phylogenetic relationships were reconstructed, and the conserved motifs of YAB proteins were compared among the three species ( Figure S3). Consistently, these YABBYs were divided into five clades ( Figure S3a). A total of ten motifs were observed ( Figure S3b), and most of these motifs are not yet well-characterized. The sequences of the ten motifs are displayed in Figure S4. Motif 1 was located in the YABBY domain, while motif 2 was located in the C 2 C 2 domain. Some motifs were only found in genes within the same clade. Motif 3 was found in three clades, FIL-like, YAB2-like, and YAB5-like, inferring that these three clades were closer in the evolutionary relationship. Motif 4 was only identified in the FIL-like clade in three species, and motif 9 was also specifically identified in the FIL-like clade, except in AtYAB3. Motifs 5, 6, 8, and 10 were found only in the paper mulberry and M. notabilis, revealing that these two species have a very close evolutionary relationship.

Transactivation Activity and Subcellular Localization of the BpYAB Family
The transactivation activity of each BpYAB protein was tested by a yeast one-hybrid system in the synthetic dropout nutrient medium-Trp-His (SD-Trp-His) growth condition. The transactivation activity was characterized by the growth ability of the transformed yeast on the SD plate containing 0-30 mM 3-AT (3-Aminotriazole). The pBridge-BpWUS was used as a positive control, and the pBridge (BD) was taken as a negative control [28]. The results indicated that four BpYAB proteins-BpYAB2, BpYAB3, BpYAB4, and BpYAB5-have transcriptional activity, while BpYAB1 and BpYAB6 did not exhibit transcriptional activity ( Figure 2). None of the YABBYs in paper mulberry showed high-level transcriptional activity, as they could not grow on the selective medium containing 5-30 mM 3-AT. Figure 2. The transactivation activity of YABBY proteins in paper mulberry. The full-length ORF of each BpYAB gene was fused with pBridge and transformed into AH109 yeasts. Then AH109 yeasts were selected from the SD-Trp-His medium. The pBridge--BpWUS was used as a positive control, and the empty vector pBridge (BD) was used as a negative control.
The pCAMBIA1300-GFP was used as a positive control to illustrate the subcellular localization of BpYABs. The GFP signal of four BpYABs (BpYAB1, BpYAB3, BpYAB4, and BpYAB5) GFP fusion proteins was observed exclusively in the nucleus, suggesting that these four BpYABs were localized in the nucleus. The GFP signal of the BpYAB2-GFP and Comm 2，但我 嘛 Figure 2. The transactivation activity of YABBY proteins in paper mulberry. The full-length ORF of each BpYAB gene was fused with pBridge and transformed into AH109 yeasts. Then AH109 yeasts were selected from the SD-Trp-His medium. The pBridge-BpWUS was used as a positive control, and the empty vector pBridge (BD) was used as a negative control.
The pCAMBIA1300-GFP was used as a positive control to illustrate the subcellular localization of BpYABs. The GFP signal of four BpYABs (BpYAB1, BpYAB3, BpYAB4, and BpYAB5) GFP fusion proteins was observed exclusively in the nucleus, suggesting that these four BpYABs were localized in the nucleus. The GFP signal of the BpYAB2-GFP and BpYAB6-GFP fusion proteins was in both the nucleus and cytoplasm (Figure 3), indicating that these two BpYABs were not unique nuclear localization proteins.

Expression Patterns of BpYAB Genes in Different Tissues
None of the BpYAB members were expressed in the root, but were mainly expressed in the bud, stem, leaf and flower (Figure 4a), indicating that the YABBY family functioned during aboveground organ development. However, neither BpYAB1 nor BpYAB6 was expressed in the vegetative organs, including the bud, stem, and leaf (Figure 4a). This expression pattern was consistent with their counterparts in Arabidopsis, CRC, and INO, which were not expressed in vegetative organs as well. Except for the two abovementioned genes, the remaining four BpYABs were highly expressed in the bud (Figure 4a). Moreover, BpYAB3 and BpYAB5 were strongly expressed in the mature leaf and lowly expressed in the stem, implying that these two genes may be involved in leaf development. On the other hand, BpYAB4 had a relatively higher expression in the stem and lower expression in the leaf, indicating that BpYAB4 may function in stem growth (Figure 4a). Except for BpYAB6, the BpYABs were all expressed in the flowers, especially in the female flower and bisexual flower (Figure 4b). BpYAB3 showed a higher expression in the female flower than in the bisexual flower (p < 0.001) among the rest of the BpYABs, and this implied that BpYAB3 was specific for female flower development (Figure 4b). In addition, only three members were expressed in the male flower, and BpYAB5 showed a higher expression than BpYAB2 and BpYAB4 (Figure 4b). To check the expression dynamics of four vegetative BpYAB genes in leaves, we obtained the FPKM values of three developmental stage leaves from the transcriptome sequencing data of paper mulberry. Leaf 1 means the newly grown leaf visible to the human eye, Leaf 2 is the tender leaf, and Leaf 3 represents the mature leaf. All the BpYAB genes showed a medium level in the bud and reached the highest expression in the Leaf 1 stage, then decreased with the leaf growing. The expression levels of BpYAB5 were twice as high as those of the other three genes during leaf development (Figure 4c). It was interesting to find that BpYAB5 showed high activity in many tissues, including both vegetative organs and reproductive organs, but mostly in leaves (Figure 4). used as a control. From left to right, the images show the green fluorescent field, bright field, and overlap of two illuminations. Bar = 50 µm.

Expression Patterns of BpYAB Genes in Different Tissues
None of the BpYAB members were expressed in the root, but were mainly expressed in the bud, stem, leaf and flower (Figure 4a), indicating that the YABBY family functioned during aboveground organ development. However, neither BpYAB1 nor BpYAB6 was expressed in the vegetative organs, including the bud, stem, and leaf (Figure 4a). This expression pattern was consistent with their counterparts in Arabidopsis, CRC, and INO, which were not expressed in vegetative organs as well. Except for the two abovementioned genes, the remaining four BpYABs were highly expressed in the bud (Figure 4a). Moreover, BpYAB3 and BpYAB5 were strongly expressed in the mature leaf and lowly expressed in the stem, implying that these two genes may be involved in leaf development. On the other hand, BpYAB4 had a relatively higher expression in the stem and lower expression in the leaf, indicating that BpYAB4 may function in stem growth (Figure 4a). Except for BpYAB6, the BpYABs were all expressed in the flowers, especially in the female flower and bisexual flower (Figure 4b). BpYAB3 showed a higher expression in the female flower than in the bisexual flower (p < 0.001) among the rest of the BpYABs, and this implied that BpYAB3 was specific for female flower development (Figure 4b). In addition, only three members were expressed in the male flower, and BpYAB5 showed a higher expression than BpYAB2 and BpYAB4 (Figure 4b). To check the expression dynamics of four vegetative BpYAB genes in leaves, we obtained the FPKM values of three developmental stage leaves from the transcriptome sequencing data of paper mulberry. Leaf 1 means the newly grown leaf visible to the human eye, Leaf 2 is the tender leaf, and Leaf 3 represents the mature leaf. All the BpYAB genes showed a medium level in the bud and reached the highest expression in the Leaf 1 stage, then decreased with the leaf growing. The expression levels of BpYAB5 were twice as high as those of the other three genes during leaf development (Figure 4c). It was interesting to find that BpYAB5 showed high activity in many tissues, including both vegetative organs and reproductive organs, but mostly in leaves (Figure 4). . * means p less than 0.05, ** represents p < 0.01 and *** represent p < 0. 001.Transcript levels of (a) and (b) were determined by qRT-PCR with three independent biological replications, and the expression of BpGAPDH was used as an internal control. (c) The expression levels (values of FPKM) of four BpYAB genes in the apical bud and leaf at three developmental stages. Leaf 1 is the newly grown leaf, visible to the human eye; Leaf 2 is the tender leaf; and Leaf 3 is the mature leaf. The data were obtained from the transcriptome of the paper mulberry. . * means p less than 0.05, ** represents p < 0.01 and *** represent p < 0. 001.Transcript levels of (a) and (b) were determined by qRT-PCR with three independent biological replications, and the expression of BpGAPDH was used as an internal control. (c) The expression levels (values of FPKM) of four BpYAB genes in the apical bud and leaf at three developmental stages. Leaf 1 is the newly grown leaf, visible to the human eye; Leaf 2 is the tender leaf; and Leaf 3 is the mature leaf. The data were obtained from the transcriptome of the paper mulberry.

BpYAB3 and BpYAB5 Do Not Show a Polarity Distribution in the Leaf
YABBY genes are expressed in a polar pattern and function in promoting abaxial polarity in Arabidopsis, including leaf, flower and ovule. However, YABBY genes do not show a polar pattern in rice. To check whether the YABBY genes of the paper mulberry showed polar distribution in leaves, we conducted the RNA in situ hybridization analyses of BpYAB3 and BpYAB5 in the apical bud and mature leaf because these two genes were highly expressed in the bud and leaf. The results showed that both were expressed in the vascular bundle of primary shoot and leaf primordium, including xylem and phloem (Figure 5a,d). In the mature leaf, BpYAB3 and BpYAB5 were both expressed throughout the entire leaf blade and did not show a polar distribution (Figure 5b,e). Both BpYAB3 and BpYAB5 were expressed in the xylem of the mature leaf midrib, and BpYAB3 was also expressed in the phloem of the mature leaf midrib. A negative control of the sense probe for BpYAB3 and BpYAB5 was used (Figure 5c,f). Overall, both BpYAB3 and BpYAB5 were expressed not only in the adaxial or abaxial lamina, specifically, but throughout the entire leaf blade, demonstrating that YABBYs in the paper mulberry may play a role in initiating outgrowth of the lamina. YABBY genes are expressed in a polar pattern and function in promoting abaxial polarity in Arabidopsis, including leaf, flower and ovule. However, YABBY genes do not show a polar pattern in rice. To check whether the YABBY genes of the paper mulberry showed polar distribution in leaves, we conducted the RNA in situ hybridization analyses of BpYAB3 and BpYAB5 in the apical bud and mature leaf because these two genes were highly expressed in the bud and leaf. The results showed that both were expressed in the vascular bundle of primary shoot and leaf primordium, including xylem and phloem (Figure 5a,d). In the mature leaf, BpYAB3 and BpYAB5 were both expressed throughout the entire leaf blade and did not show a polar distribution (Figure 5b,e). Both BpYAB3 and BpYAB5 were expressed in the xylem of the mature leaf midrib, and BpYAB3 was also expressed in the phloem of the mature leaf midrib. A negative control of the sense probe for BpYAB3 and BpYAB5 was used (Figure 5c,f). Overall, both BpYAB3 and BpYAB5 were expressed not only in the adaxial or abaxial lamina, specifically, but throughout the entire leaf blade, demonstrating that YABBYs in the paper mulberry may play a role in initiating outgrowth of the lamina.

Ectopic Expression of BpYABs Affects Arabidopsis Leaf Development
Ectopic overexpression of BpYAB2, BpYAB3, BpYAB4, and BpYAB5 in Arabidopsis affected the morphology of Arabidopsis leaves. Compared with the wild-type Arabidopsis (Col-0), the transgenic line showed that the cotyledons curled downward to form an inverted bowl-shaped leaf, and the leaf area was relatively smaller. The leaves of the wildtype Arabidopsis were flat ( Figure 6). When the seedlings were transplanted to nutrient soil

Ectopic Expression of BpYABs Affects Arabidopsis Leaf Development
Ectopic overexpression of BpYAB2, BpYAB3, BpYAB4, and BpYAB5 in Arabidopsis affected the morphology of Arabidopsis leaves. Compared with the wild-type Arabidopsis (Col-0), the transgenic line showed that the cotyledons curled downward to form an inverted bowl-shaped leaf, and the leaf area was relatively smaller. The leaves of the wildtype Arabidopsis were flat ( Figure 6). When the seedlings were transplanted to nutrient soil and cultured for 10 days, it was found that in the transgenic line, the true leaves of Arabidopsis also appeared curled at the leaf edge, and some of them showed a twisted phenotype. The growth of the entire plant was slower than that of the wild-type ( Figure S5). and cultured for 10 days, it was found that in the transgenic line, the true leaves of Arabidopsis also appeared curled at the leaf edge, and some of them showed a twisted phenotype. The growth of the entire plant was slower than that of the wild-type ( Figure S5). The ectopic overexpression of the BpYAB2, 3-5 genes in Arabidopsis did not affect the expression of the AtYAB genes (YAB2, YAB3, FIL, and YAB5) in Arabidopsis ( Figure  7). Expression levels of KAN2, KAN3 and KAN4 genes functioning on the abaxial surface of the leaves, and of the genes PHB and REV specifically expressed in the paraxial plane, also did not change significantly. There was no difference in the related factors that controlled the development of the adaxial-abaxial axis in Arabidopsis (AS1 and AS2). These results suggested that the ectopic overexpression in Arabidopsis did not cause the loss of the dorsoventral polarity. The expression of the CUC gene that controls the leaf margin development was significantly decreased in the 35S:BpYAB3 and 35S:BpYAB5 strains (p < 0.001), but slightly increased in the 35S:BpYAB2 strain. The expression of the downstream regulatory gene WOX1 of YAB increased significantly in all the transgenic strains, especially in 35S:BpYAB4, increasing 48 times that in the WT. In addition, the expression of auxin polar transport PIN1 doubled in the 35S:BpYAB4 strain (p < 0.05).
Therefore, we speculated that the abnormal leaf development caused by the ectopic expression of BpYABs may be due to the fact that BpYABs promote the expression of their downstream gene WOX1 and involve changes in the distribution of auxin, although the influence differed in different ectopic expressions of BpYAB strains. The ectopic overexpression of the BpYAB2, 3-5 genes in Arabidopsis did not affect the expression of the AtYAB genes (YAB2, YAB3, FIL, and YAB5) in Arabidopsis (Figure 7). Expression levels of KAN2, KAN3 and KAN4 genes functioning on the abaxial surface of the leaves, and of the genes PHB and REV specifically expressed in the paraxial plane, also did not change significantly. There was no difference in the related factors that controlled the development of the adaxial-abaxial axis in Arabidopsis (AS1 and AS2). These results suggested that the ectopic overexpression in Arabidopsis did not cause the loss of the dorsoventral polarity. The expression of the CUC gene that controls the leaf margin development was significantly decreased in the 35S:BpYAB3 and 35S:BpYAB5 strains (p < 0.001), but slightly increased in the 35S:BpYAB2 strain. The expression of the downstream regulatory gene WOX1 of YAB increased significantly in all the transgenic strains, especially in 35S:BpYAB4, increasing 48 times that in the WT. In addition, the expression of auxin polar transport PIN1 doubled in the 35S:BpYAB4 strain (p < 0.05).
Therefore, we speculated that the abnormal leaf development caused by the ectopic expression of BpYABs may be due to the fact that BpYABs promote the expression of their downstream gene WOX1 and involve changes in the distribution of auxin, although the influence differed in different ectopic expressions of BpYAB strains.

Ectopic Expression of BpYABs Affects Arabidopsis Silique Length and Seed Size
The ectopic expression of BpYAB2, BpYAB3, and BpYAB4 will cause the Arabidopsis silique length to be significantly shorter (Figure 8). The length and width of the wildtype and transgenic Arabidopsis siliques were measured, and the results of statistical analysis of differences found that the lengths in the transgenic of BpYAB2 (p < 0.001), BpYAB3 (p < 0.05) and BpYAB4 (p < 0.01) line were significantly shorter than those of the wild-type and 35S:BpYAB5 lines (Figure 8a,b), and the widths were slightly greater than those of the wild-type and 35S:BpYAB5 lines (Figure 8c). Therefore, the ectopic expression of BpYAB2, BpYAB3, and BpYAB4 can cause the length of the Arabidopsis silique to be significantly shorter.
Additionally, the number of seeds in a single pod was counted. The results showed that the number of seeds did not decrease in all the transgenic plants ( Figure S6a). Furthermore, observation and measurement of the sizes of mature seeds ( Figure S7) showed that the length and width of seeds in the BpYAB2, BpYAB3, and BpYAB4 strains were smaller than those of the wild-type and 35S:BpYAB5 lines ( Figure S6b). Moreover, the ectopic expression of BpYAB2, BpYAB3 and BpYAB4 can shorten the silique and seeds of Arabidopsis more than that of WT or 35S:YAB5, but does not affect the number of seeds in a single silique.  Table S5.

Discussion
Members of YABBY are seed-plant-specific transcription factors and function in lateral organ development. With the availability of the whole genome, YAB families have been identified in Arabidopsis, rice, tomato, soybean, and cotton [3,29,30]. However, few studies have been performed on YABBY genes in trees.

The Sequence Structures of YABBY Transcription Factors in Paper Mulberry Are Conserved
In this study, six YABBY genes were identified from the paper mulberry genome (Figure 1 and Table 1), and subsequent analysis showed that the structure of the YABBYs was relatively conserved among different plants. BpYAB2 and BpYAB6 had six exons, while the remaining four BpYAB genes had seven exons in the genomic DNA sequence ( Figure S2a). The lengths of exon 4 and exon 5 were the same in the six YABBY members of paper mulberry: 49 and 76 bp, respectively. These two exons were conserved in different plants, possibly because they code the YABBY domain. Previous studies have shown that the core element of the C 2 C 2 -type zinc-finger domain with Zn 2+ is the C-X 2 -C-X 9 -P-X 11 -C-X 2 -C motif, and this structure has also been found in the YABBY protein sequences of paper mulberry ( Figure S2c). The putative helix-loop-helix "YABBY" domain has been reported to have a sequence similar to the first two helices of the high mobility group (HMG) box [31], and we found that BpYABs also contain typical helices of HMG boxes ( Figure S2d), suggesting that the structure of the YABBYs is conserved among seed plants. Except for two conserved domains, many conserved motifs found in variable regions from each sub-family also showed similarities. Three motifs (motif 3, motif 4, and motif 9) were found in the C-terminal of the YABBY protein. Motif 3 was found in the vegetative clades, including FIL-like, YAB2-like, and YAB5-like clades, while motif 4 and motif 9 were only specific to the FIL-like clade ( Figure S3b). Five motifs, located in the variable region between the two conserved domains (C 2 C 2 zinc-finger and YABBY), were found only in M. notabilis and paper mulberry, but not in A. thaliana ( Figure S3). Paper mulberry and M. notabilis, which both belong to the Moraceae family, shared the same conserved motif in the YABBY protein sequence, suggesting they may have a very close evolutionary relationship and may have evolved from a common ancestor.

Three BpYAB Proteins Are Typical Transcription Factors
The YABBY acts as a transcription factor, including a typical DNA binding site (YABBY domain) and an oligomerization site (C 2 C 2 zinc-finger domain) [32]. We studied the subcellular localization and transactivation activity of the BpYAB family for the first time in a comprehensive way. Three BpYABs (BpYAB3, BpYAB4, and BpYAB5) showed typical characteristics of the transcription factor, including the transactivation activity and nuclear localization signal, as well. These results were consistent with VpYAB1 in grapes and BraYAB1 in cabbages, which belong to the FIL-like clade, as well as BpYAB3 and BpYAB4 [8,10]. MsYAB5 is the homologous gene of BpYAB5 in spearmint; both of them are in the YAB5-like clade and localized in the nucleus [11]. BpYAB2 could not localize in the nucleus specifically but could activate the downstream gene expression, implying that it needs to form complexes with other proteins to function in the nucleus. VpYAB2 and IaYAB2, which are members of the YAB2-like clade and are localized in the nucleus [8,9], differed from BpYAB2 in the paper mulberry, implying the diverse characteristics of YABBYs in different species. Additionally, BpYAB1 is a nuclear localization protein without the transactivation activity, implying that it might function as a transcriptional repressor. This result was in accordance with the CRC gene in Arabidopsis, which is localized in the nucleus and acts as a transcriptional repressor in the floral meristem [2]. Notably, BpYAB6 had neither the nuclear localization signal nor the transactivation activity, and the expression level was very low in the tissues detected, inferring that its function may not be as important as others in the BpYAB family.

YAB Family Underwent Expansion in the FIL-like Clade
The YABBY genes from seed plants can be divided into five groups: the CRC-like, INOlike, FIL-like, YAB2-like, and YAB5-like groups [12]. Members of YABBYs from 17 species, including 127 protein sequences, were evenly divided into these five groups. The resulting phylogenetic tree of dicotyledons showed that YAB2-like and YAB5-like clades had a close relationship, while CRC-like and INO-like clades might have originated from a common ancestor (Figure 1). In addition, the five YABBYs from the base angiosperm A. trichopoda were found in these exact five groups. These results indicated that YABBYs may diverge from a common ancestor within the same group, and the biological function of each group might be similar in different species. The INO-like and CRC-like clades contained 18 and 19 YABBY genes among the 17 species, and each group contained at least one gene from the two groups. One or two YABBY genes were found in the YAB2-like and YAB5-like clades, the number of YABBY genes in these two groups were 23 and 28, respectively. There were five more members in the YAB5-like clade than in the YAB2-like clade, and this may have resulted from the expansion of GhYABs in the YAB5-like clade (Figure 1). However, the number of YABBY genes in the FIL-like group, which was 39, was twice more than that in the CRC-like and INO-like groups. There was only one YABBY gene in the FIL-like clade from the base species of angiosperm A. trichopoda, but at least two members in other dicotyledon plants from Vitaceae to Solanaceae, indicating that there was an expansion event of YABBY in the FIL-like clade. We also found that WOX genes in the WUS clade underwent an expansion event in the paper mulberry as well, demonstrating that the expansion event is common in the transcription factor family in the paper mulberry.

The Function of BpYAB Genes Was Related to the Evolutionary Relationship
In Arabidopsis and other species, expressions of CRC-like and INO-like groups are restricted to reproductive organs, such as the carpel and ovule [14,32]. In contrast, members of the FIL-like, YAB2-like, and YAB5-like groups are mostly expressed in the leaves and are termed the "vegetative YABBYs" [20,33]. The vegetative angiosperm YABBY genes are exclusively expressed in leaf-homologous organs, including the leaf and flower. Evidence has shown that vegetative YABBYs act in several aspects of leaf development [34,35]. We identified six YABBY genes in the paper mulberry, and the number of YABBY families in the paper mulberry was the same as in the model plant, Arabidopsis. YABBY genes in Arabidopsis and paper mulberry can have a one-to-one correspondence in the phylogenetic tree, and their biological function may be similar, as well. The expression of BpYAB family members was not detected in the root, suggesting they were mainly expressed in aboveground organs such as the bud, leaf, stem, and flower. BpYAB1 was only expressed in the female flower and bisexual flower in paper mulberry but was not expressed in the male flower or other vegetative organs (Figure 4a,b). This was consistent with its counterpart CRC gene in Arabidopsis, which functions in carpel development [14]. BpYAB6, the counterpart of the INO gene in Arabidopsis, was not expressed in the organs detected (Figure 4a,b), inferring that it may participate in other types of organ development that we did not detect in the paper mulberry. The remaining four BpYAB genes, belonging to the FIL-like, YAB2-like, and YAB5-like clades, were expressed in both vegetative and reproductive organs (Figure 4), and their functions were in line with corresponding AtYAB genes [19,36]. BpYAB3 and BpYAB5 were especially highly expressed in the bud and leaf (Figure 4a), suggesting they are significant to leaf formation. In addition, the four BpYAB genes played a meaningful role in the leaf development process, as well (Figure 4c). BpYAB5 showed a high activity in all tissues, including the male flower and leaf, revealing that this gene is fundamental to the growth of paper mulberry. Altogether, the functions of the BpYAB gene are related to its evolution, demonstrating the conserved biological role in dicotyledons.

BpYAB Did Not Show the Adaxial-Abaxial Polarity Distribution in Lamina
It has been reported that YABBY genes show the abaxial expression pattern in the Arabidopsis leaf and the adaxial expression pattern in the maize leaf [1,23]. However, YABBYs in rice are fully expressed in the spikelet and midrib in the leaf, which do not show the polarity pattern. These results indicate that the polarity expression pattern may diverge in different species. The in-situ results of BpYAB3 and BpYAB5 showed that they were expressed in the vascular bundle of the primary shoot and leaf primordium and were not expressed in the lamina of the leaf primordium. In the mature leaf, BpYAB3 and BpYAB5 were both expressed throughout the leaf blade and did not show a polar distribution ( Figure 5). BpYAB3 was expressed in both the xylem and phloem of the midrib in the mature leaf, but BpYAB5 was only expressed in the xylem, and this result was consistent with the expression levels of two BpYAB genes in the stem, where BpYAB3 was higher than BpYAB5 ( Figure 4a). As mentioned earlier, the abaxial expression pattern may diverge in different species, which makes sense in paper mulberry, as well.

BpYABs Involved in Leaf and Silique Development
AtCRC specifically regulates carpel and nectary development [14]. According to the phylogenetic tree analysis, BpYAB1 and AtCRC were homologous genes. Meanwhile, BpYAB1 was only expressed in the female flowers but not in the vegetative organs of paper mulberry, which indicated that the genes of the CRC sub-clade mainly function in the development of floral organs, and their functions are relatively conservative. BpYAB6 is homologous to AtINO, which is expressed in the outer integument of the ovule [37]. Sequence analysis revealed that BpYAB6 had no nuclear localization signal and lacked the ability to activate downstream reporter gene expression (Figure 2). BpYAB6 was not expressed in the vegetative organs but only in the female flowers, and the expression level was very low. In this study, without the positive transgenic plants, it was not possible to speculate what growth and development processes BpYAB6 and BpYAB1 might perform in paper mulberry.
In Arabidopsis, FIL, YAB3, YAB2, and YAB5 have redundant functions and jointly regulate the formation of the abaxial surface of leaves. BpYAB2 and BpYAB5 are homologous with AtYAB2 and AtYAB5, and the genes homologous to AtFIL are BpYAB3 and BpYAB4. The expression patterns of these four YAB genes of Arabidopsis are similar, and all of them are specifically expressed in the abaxial face cells of lateral organs in the aboveground part [36]. The overexpression of FIL and YAB3 genes in Arabidopsis can promote cell development at the distal end of leaves [1,19].
OsYAB3, OsYAB4, and OsYAB5 are closely related to the FIL/YAB3 of Arabidopsis. The expression of OsYAB3 was detected in both the leaves and floral organs, and the RNAi transgenic rice leaves of OsYAB3 became twisted, knotted, or leafless ears, but this does not affect the polarity development of the leaves. At the same time, YAB3, WOX3, and KNOX1 are involved in the formation of apical organs in rice [38]. The overexpression of OsYAB4 in Arabidopsis shows irregular leaf veins. It has been speculated that OsYAB4 is related to the development of rice vascular tissue [6]. TOB1 (OsYAB5) is mainly involved in rice spikelet development [21]. OsYAB6 is closely related to OsYAB1 and belongs to the YAB2-like sub-clade. OsYAB6 regulates the morphology of the paraxial end of the leaf by affecting the development of vesicle cells [39]. Except for OsYAB7, other rice YABBY genes are involved in the growth and development of the lateral organs, but they do not have a distribution pattern of dorsoventral polarity in the leaf.
The expression patterns of BpYAB3 and BpYAB5 in paper mulberry were similar, and both of them were highly expressed in mature leaves. Additionally, they were fully expressed in the lamina of paper mulberry without the adaxial-abaxial polarity. We speculated that BpYAB might play a role in the later development of leaves and affect the development of the leaf shape by regulating the expansion of the leaf blade. BpYAB4 was highly expressed in the paper mulberry stems, but lower in the leaves. Furthermore, BpYAB2, BpYAB3, BpYAB4, and BpYAB5 were all expressed in the leaf and bud, while BpYAB3 and BpYAB5 exhibited the higher level of expression (Figure 4a). The ectopic overexpression of BpYAB2, BpYAB3, BpYAB4, and BpYAB5 caused the Arabidopsis leaves to become smaller and curl downwards. This phenotype was similar to that of Arabidopsis auxin mutant axr1, in which the lack of auxin causes the leaves to become smaller and curl downwards, but in 35S:BpYAB4, the expression of the auxin polar transport factor PIN1 increased, suggesting that the auxin content of 35S:BpYAB4 might be increased. Meanwhile, the expression pattern of CUC1 and CUC2 in the transgenic Arabidopsis was similar in that they showed slightly more upregulation in 35S:BpYAB2 than that in WT, and slightly more downregulation in 35S:BpYAB4 than that in WT, but almost undetectable expression in 35S:BpYAB3 and 35S:BpYAB5. CUC1 and CUC2 are essential for shoot meristem initiation, as well as to promote carpel margin meristem formation during the Arabidopsis gynoecium development [40]. Furthermore, CUC1 and CUC2 can regulate cytokinin homeostasis to determine the ovule number in Arabidopsis [41]. Besides, previous studies and our experimental results showed that both an increase and decrease in auxin levels caused the leaves to curl downwards and shrink. Therefore, we speculated that the lack of auxin homeostasis led to the downward shape and curl shrinks of the leaves. In addition, the YAB gene might also indirectly participate in the auxin regulation pathway. The silique of the Arabidopsis crc mutant appeared to be shorter and thicker, and the tips could not be fused [14]. The Arabidopsis siliques were also found to be shorter and thicker in the 35S:BpYAB2, 35S:BpYAB3, and 35S:BpYAB4 ectopic expression lines (Figure 8), but the tip could be fused. The mature seeds were smaller than the wild-type ( Figure S6), but the number did not change significantly, indicating that the abnormal expression of these BpYABs could also affect the morphology of the Arabidopsis silique.
Overall, the YABBY gene expression pattern in dicotyledonous plants is relatively conservative, mainly distributed at the abaxial end of the leaf, but it has other functions in different dicotyledonous plants. Although YABBY is defined as specifically expressed on the abaxial surface of the leaf, more and more studies have shown that YABBY also plays a role in the lateral growth of later leaves. In monocots, the function of YABBY genes is more diversified, mainly involved in the development of the floral organs. Thus, we speculated that YABBY has produced a functional differentiation in the evolution of dicotyledonous and monocots plants, not only limited to the abaxial cells of the organ development but presenting a diversified phenomenon.

Identification and Gene Sequence Analysis of the BpYAB in Paper Mulberry
To identify YABBY candidates in paper mulberry, we used protein sequences of A. thaliana and M. notabilis to search the paper mulberry protein database by employing the blastp. Then the blastn was used to identify potential genes in the paper mulberry genome [26]. The putative YABBY members were aligned with other YABBYs in A. thaliana and M. notabilis to check the conserved amino acid, and the predicted protein sequences lacking the C 2 C 2 domain or YABBY domain were excluded. Finally, RACE-PCR was used to obtain and check the full length of YAB genes from paper mulberry.
The chromosomal positions and relative distances of the BpYAB genes were drawn with the online tool MG2C (Map Gene2 Chromosome V2) in Figure S1 (http://mg2c.iask. in/mg2c_v2.0, accessed on 20 August 2018). The amino acid sequences of putative YABBY proteins were submitted to the ProtParam of Expasy http://web.expasy.org/protparam, accessed on 5 August 2018) to calculate the molecular weights and theoretical isoelectric points in Table 1. The gene structure information of six BpYAB genes was obtained from the genome of the paper mulberry [26] and was drawn by a software package called Illustrator of Biological Sequences (IBS) [42] in Figure S2a. The conserved C 2 C 2 domain and YABBY domain were aligned by MUSCLE in Geneious software and adjusted manually, and the relative positions in Figure S2b were drawn with IBS. For the analysis of cis-elements in the promoters of BpYAB genes, the upstream sequences (2 kb) from the translation start sites of the BpYAB genes were submitted to PLACE database (https://www.dna.affrc.go. jp/PLACE/, accessed on 5 January 2022) to identify the cis-acting regulatory elements in Tables 2 and S4.

Phylogenetic Analysis and Conserved Motif Identification of BpYAB Proteins
The 127 YAB proteins of 17 plants were obtained from the PlantTFDB (Plant Transcription Factor Database) (http://planttfdb.cbi.pku.edu.cn/index.php, accessed on 21 June 2018). Evolutionary analyses were conducted in MEGA7, and the number of bootstrap replicates was 1000 [43]. The sequence information is provided in Table S1. The evolutionary history of YABs was inferred using the NJ (Neighbor-Joining) method. The optimal tree with the sum of the branch length was 19.12922157. The Poisson correction method was used to calculate the evolutionary distances. All ambiguous positions were removed for each sequence pair. The conserved motifs of YABBY proteins in A. thaliana, paper mulberry, and M. notabilis were identified using the motif elicitation tool MEME (http://meme.nbcr.net/meme/, accessed on 21 June 2018) with the following parameters: any number of repetitions, a maximum of 10 misfits with an optimum motif width from 6 to 200 amino acid residues.

Subcellular Localization and Transactivation Activity Assay of BpYAB Proteins
The open reading frame (ORF) of BpYABs was inserted into the pCAMBIA1300-GFP vector to generate the BpYAB-GFP fusion protein. The enzymes and primers used to construct the vectors are listed in Table S6. The pCAMBIA1300-BpYAB-GFP plasmid was transformed into A. tumefaciens (EHA105) and introduced into the N. benthamiana epidermal cell. After 48 h cultivation, fluorescence was examined by laser confocal microscopy (Leica TCS SP5, Wetzlar, Germany).
To perform the transactivation activity assay, the ORF of BpYABs was cloned into the pBridge vector. The restriction enzymes and primers are provided in Table S7. The pBridge-BpYABs vectors were transformed into Saccharomyces cerevisiae (AH109, BD Biosciences, Palo Alto, CA, USA) according to the manufacturer's protocol. Transformed yeasts were screening cultured on SD medium without His and Trp. Then, the positive yeast cells were dropped on SD-Trp-His plates containing 3-AT (0 to 30 mM) and incubated at 30 • C for about 3 days. This experimental procedure was referenced in Tang et al. [28].

Plant Materials for qRT-PCR and RNA-Seq Data Analysis
To gain insight into the positions where the BpYAB genes were active, qRT-PCR was performed for different tissues, including the callus, bud, leaf, stem and root of tissue culture seedlings and the flowers of perennial paper mulberry from a wild field. For tissue specificity, plantlets, except the flower, were cultured on the MS culture media in an artificial climatic chamber at 26 • C for 14/10 h (day/night) for a month. Three kinds of paper mulberry flowers were obtained from three individual plants growing in the experimental field in the resource nursery, which belonged to our laboratory. All samples were frozen in liquid nitrogen and stored at −80 • C. Total RNAs of each sample were extracted with a kit (TransGen, Beijing, China). RNA quality and purity were evaluated with an OD260/280 ratio, and the value of RIN (RNA integrity number) was tested using the NanoDrop 2000 (Thermo Fisher, Waltham, MS, USA). First-strand cDNA synthesis was performed using PrimeScript RT Reagent Kit, (Takara, Dalian, China). Quantitative real-time PCR (qRT-PCR) was conducted on a real-time PCR system MX3000PTM, (Agilent Stratagene, Santa Clara, CA, USA). Each reaction was comprised of 20 µL, including 10 µL PCR Mix of SYBR-Green PrimeScript RT-PCR Kit (,Takara, Dalian, China), 2 µL template, 0.5 µL of each primer, 0.5 µL ROX, and 6.5 µL ddH 2 O. Three technical replicates were taken for each sample. The 2 −∆∆CT method was used to analyze the expression levels of target genes, and the data were normalized by the transcript level of the BpGAPDH gene. The p value was calculated by means of the T student test and variance analysis (ANOVA) with significance levels of 5% (p < 0.05). All the primers for qRT-PCR are listed in Table S8. The FPKM values of four BpYAB genes (Table S9) were obtained from the RNA-seq data of the paper mulberry [26].

In Situ Hybridization of BpYAB3 and BpYAB5
The freshly collected apical bud and mature leaves were fixed immediately in FAA (3.7% formaldehyde, 5% acetic acid, and 50% ethanol) solution. The plant material was treated and detected as reported in [44]. The sequence-encoding C-terminal regions of BpYAB3 (nucleotides 679-996 of ORF) and BpYAB5 (nucleotide positions 517-819 of ORF) were amplified and transcribed as probes (Digoxigenin RNA labeling kit, Roche, Shanghai, China).

Agrobacterium-Mediated Transformation of Arabidopsis
The ORF regions of BpYAB2, BpYAB3, BpYAB4, and BpYAB5 were amplified using primers with restriction sites and ligated to the pCAMBIA1300 vector with T4 ligase (Takara, Dalian, China). The constructed vector was finally transferred into Agrobacterium GV3101. The Arabidopsis inflorescence was infected by the Agrobacterium for transformation, and the obtained seeds were screened with 50 µg/mL hygromycin B. Five overexpression lines for every BpYAB were obtained, and the leaf phenotype trend was consistent, so we selected one of them and the seedlings of T2 generation were used for further analysis.

The Morphological Observation of Transgenic Arabidopsis
The transgenic Arabidopsis was germinated in a Petri dish. On the 12th day, when the cotyledons unfolded and the true leaves were just about to grow out, the first phenotypic observation and photographs were taken. Then the seedlings were transplanted to the soil in the nutrient bowl. Observations and photographs were taken on the 8th, 10th, and 13th days after transplanting. During the grouting period, the length and width of the silique of each transgenic plant were measured, and there were more than 30 plants for each line. After the seeds matured, the number of seeds and the size of the seeds in a single silique were measured, and the total number in each line was more than 30. The p value was calculated by means of the T Student test and variance analysis (ANOVA) with significance levels of 5% (p < 0.05). The wild-type plants (WT) were used as controls.

Conclusions
In this study, six YABBY genes were identified from paper mulberry and were evenly divided into five subfamilies from the phylogenetic analysis. Four members were located in chromosomes and the other two were in scaffolds. Subsequent sequence analysis revealed that BpYAB genes possessed six or seven exons and had several cis-elements that were development-, phytohormone-, and environment-related in the promotor region. Three members, BpYAB3, BpYAB4, and BpYAB5, were typical transcription factors, which were localized in the nucleus and possessed transactivation ability. Furthermore, BpYAB members were mainly expressed in aboveground organs, such as the leaf and flower. BpYAB3 and BpYAB5 showed full expression in the lamina of the mature leaf but did not show the adaxial or abaxial polarity in the paper mulberry as described in other species.
Ectopic expression of four BpYABs caused leaves to curl downwards and shrink, and three of them affected the morphology of the Arabidopsis silique. Ultimately, our work sheds light on the evolutionary history and expression patterns of BpYABs and provides fundamental information for understanding the precise function of YABBY in woody plants for further research.