PhCESA3 silencing inhibits elongation and stimulates radial expansion in petunia

Cellulose synthase catalytic subunits (CESAs) play important roles in plant growth, development and disease resistance. Previous studies have shown an essential role of Arabidopsis thaliana CESA3 in plant growth. However, little is known about the role of CESA3 in species other than A. thaliana. To gain a better understanding of CESA3, the petunia (Petunia hybrida) PhCESA3 gene was isolated, and the role of PhCESA3 in plant growth was analyzed in a wide range of plants. PhCESA3 mRNA was present at varying levels in tissues examined. VIGS-mediated PhCESA3 silencing resulted in dwarfing of plant height, which was consistent with the phenotype of the A. thaliana rsw1 mutant (a temperature-sensitive allele of AtCESA1), the A. thaliana cev1 mutant (the AtCESA3 mild mutant), and the antisense AtCESA3 line. However, PhCESA3 silencing led to swollen stems, pedicels, filaments, styles and epidermal hairs as well as thickened leaves and corollas, which were not observed in the A. thaliana cev1 mutant, the rsw1 mutant and the antisense AtCESA3 line. Further micrographs showed that PhCESA3 silencing reduced the length and increased the width of cells, suggesting that PhCESA3 silencing inhibits elongation and stimulates radial expansion in petunia.


Results
Isolation of P. hybrida PhCESA3 cDNA. The full-length cDNA of PhCESA3 was isolated from a cDNA library prepared from the corollas of petunia by homologous cloning and the RACE-PCR method. PhCESA3 was predicted to encode a protein of 1083 amino acids with a calculated molecular weight of 120.8 kDa and a pI of 7.02.
The multiple sequence alignments of PhCESA3 protein in the petunia and A. thaliana are presented in Fig. S1. PhCESA3 shares 86.5% amino acid sequence identity with AtCESA3 (Table S1). In addition to a putative zinc-binding domain (at amino acid residues 31-76) and two putative transmembrane helices in the N-terminus, PhCESA3 contains six putative transmembrane helices in the C-terminus (Fig. S1) similar to A. thaliana CESA proteins, and the C-terminus is strikingly conserved 19,20 . In the course of the publication of the study, the petunia genome sequences were released 21 (https://www.sgn.cornell.edu/). By blasting against Sol Genomics Network with A. thaliana AtCESAs, nine PhCESAs, including PhCESA3, were obtained in the petunia genome. The PhCESAs are named PhCESA1, PhCESA2A, PhCESA2B, PhCESA2C, PhCESA3, PhCESA4, PhCESA6, PhCESA7 and PhCESA8 based on the identity of their encoding amino acid sequences with the A. thaliana AtCESAs (Table S1). The phylogenetic tree of the PhCESAs and A. thaliana AtCESAs shown in Fig. 1 further reveals that the cloned PhCESA3 is the ortholog of AtCESA3.
Expression patterns of PhCESA3 mRNA. As a step toward functional analysis, we examined the spatiotemporal expression of PhCESA3 petunia 'Ultra' using quantitative real-time PCR (qPCR) with gene-specific primers, and actin was used as an internal control. PhCESA3 transcripts were present in root, stem, leaf, and corolla, with varying patterns of expression (Fig. 2). PhCESA3 mRNA was slightly more abundant in roots and stems compared to corollas.

VIGS-mediated
PhCESA3 silencing leads to plant dwarfing. To explore the function of PhCESA3, we generated a TRV-PhCESA3 vector that contains a 266 bp fragment of the 3′ untranslated sequence of PhCESA3 cDNA to test the silencing of PhCESA3 by virus-induced gene silencing (VIGS) 22 . Five weeks after TRV infection, plants infected with TRV-PhCESA3 showed a significant dwarfing phenotype. The stems and internodes were shorter and wider than those of the empty TRV-infected control plants ( Fig. 3a-e; Table 1). The silenced stems were about half the length of the control. The diameter of silenced stems was significantly increased. In addition, there were shorter stem epidermal hairs on PhCESA3-silenced plants compared to control plants (Fig. 3f). PhCESA3 silencing also resulted in smaller and thicker leaves than those of control plants ( Fig. 3g and Table 1). Analysis by qPCR showed decreased PhCESA3 mRNA levels in corollas, leaves and stems in plants treated with TRV-PhCESA3 compared to those in control plants (Figs 4 and S2 and 3).
In addition, PhCESA3 silencing has no significantly effect on the expression of the other 8 PhCESAs in corollas (Fig. 4), leaves (Fig. S2) and stems (Fig. S3). PhCESA2B, PhCESA2C, and PhCESA6 mRNA levels were not detected by qPCR in either TRV-PhCESA3-treated or control corollas and PhCESA6 mRNA levels were not also detected in either TRV-PhCESA3-treated or control leaves.
PhCESA3 silencing alters flower organ growth. PhCESA3 silencing reduced the height of stems but did not change the flowering time. However, altered flower organ growth was observed in PhCESA3-silenced plants. PhCESA3 silencing reduced the size but increased the thickness of petal limbs, and it reduced the height of petal tubes ( Fig. 3h and i; Table 1).
The pedicels (fruit stalks), filaments and styles were shorter and wider in PhCESA3-silenced plants than those in control plants ( Fig. 3j-l; Table 1). The lengths of pedicels (fruit stalks), styles, and filaments in PhCESA3-silenced plants were about two-thirds, half and half of those of control plants, respectively ( Table 1). The diameters of pedicels and styles of PhCESA3-silenced plants were about one and half, and two fold of those of control plants, respectively (Table 1). In addition, the styles and filaments were semi-transparent and loose, and their surface was rough in PhCESA3-silenced plants ( Fig. 3n and o). In PhCESA3-silenced plants, the stigma was small and deformed (Fig. 3m), and the unpollinated ovaries and young fruits were often deformed ( Fig. 3p and q).
PhCESA3 silencing leads to a significant reduction in cellulose content. To further characterize the function of PhCESA3, cellulose content was analyzed in plants treated with TRV and TRV-PhCESA3 treatment plant. PhCESA3-silenced plants showed a significant decrease in total cellulose content of the stems and leaves ( Fig. 5a and b). In 5-week-old plants, mature stems in PhCESA3-silenced plants had about 80% and mature leaves had about 70% of the control level of cellulose ( Fig. 5a and b).
PhCESA3 silencing leads to a significant reduction in fertility. The effects of PhCESA3 silencing on fertility were examined. PhCESA3 silencing caused a significant reduction in fertility, resulting in fewer than 100 seeds per fruit. More seeds per fruit were obtained in reciprocal crosses between PhCESA3-silenced plants and control plants when PhCESA3-silenced plants were used as the pollen donor or receptor (Table 2) but fewer than crosses between both control plants as parents, indicating reduced female and male fertility. Reciprocal crosses with control plants showed that the female reproductive capacity was more strongly affected than male reproductive capacity in PhCESA3-silenced plants ( Table 2).
PhCESA3 silencing inhibits cell elongation and causes cell swelling. Because PhCESA3 silencing altered the growth of stems, leaves and flowers, we further examined these organs in detail. Scanning electron micrographs showed reduced cell length and increased cell width of the epidermal cells of stems, pedicels, filaments and styles in PhCESA3-silenced plants (Fig. 6a-d; Table 3). Optical micrographs of the epidermis of stems and pedicels showed similar results ( Fig. S4a and b).
Scientific RepoRts | 7:41471 | DOI: 10.1038/srep41471 As visualized in transverse and longitudinal sections, PhCESA3 silencing also reduced the length and increased the width of the cortical cells of stems, pedicels, and filaments, but the number of cortical cells did not change (Fig. 7a,b,e,f and l; Table 3), indicating that the swollen stems were attributed to an increased cell width rather than a decrease in cell number. Hand-cut sections also showed similar results ( Fig. S5a and b). The transverse sections showed wider pedicels, filaments and styles in PhCESA3-silenced plants than those in control plants (Fig. 7d,j,m; Table 1). In addition, the arrangement of cortical cells in stems, pedicels and filaments was irregular and close in PhCESA3-silenced plants compared with that of control plants (Fig. 7b,e,l). The cortical cells were often round or oval shaped in control plants, while the cortical cells were often irregular polygon shaped in PhCESA3-silenced plants (Fig. 7b,e and l).
Scanning electron micrographs of adaxial and abaxial epidermal cells of leaves and petal limbs showed reduced cell size in PhCESA3-silenced plants ( Fig. 6e and h). The reduced proportion of leaves and petal limbs was similar to that of the epidermal cells of leaves and petal limbs. Optical micrographs of the epidermis of leaves and petals showed similar results (Fig. S4c-f). As visualized in transverse sections, PhCESA3 silencing increased the width  of the mesophyll cells of leaves and petals ( Fig. 7c and g), while the number of mesophyll cells did not change, indicating that the thickened leaves and petals were attributed to an increased cell size rather than cell number.
In addition, the sections of petal limbs revealed that the outside shape of the epidermal cells of the adaxial side of petal limbs from control plants was often sharp, while the outside shape of the same cells from PhCESA3-silenced plants was curved ( Fig. 7g and h). The epidermal cells of the abaxial side of petal limbs of control plants showed a similar size and good arrangement, but the same cells in PhCESA3-silenced plants had different sizes, different shapes, and irregular arrangement ( Fig. 7g and i). Transmission electron micrographs of cells from petal limbs confirmed these results ( Fig. 8e-g).
Scanning electron micrographs showed that the epidermal cells of petals in PhCESA3-silenced plants were often collapsed while those of control plants were plump (Fig. 6g). The collapsed cells may have been due to thin and fragile epidermal cell walls in PhCESA3-silenced plants compared to those of control plants as revealed by transmission electron microscopy (Fig. 8e).
The transverse sections of filaments and styles showed the wide marrow in PhCESA3-silenced plants compared with that of the control ( Fig. 7k and n). Moreover, the myeloid cells of control plants were similar in size and had a good arrangement while those of PhCESA3-silenced plants had different sizes, different shapes and irregular arrangement ( Fig. 7k and n). In addition, wider epidermal cells of style were observed in the PhCESA3-silenced plants compared to those in the control plants (Fig. 7o).
Scanning electron and optical micrographs of hand-cut sections and stereomicroscope micrographs all showed that PhCESA3 silencing reduced the length and increased the width of cells of epidermal hairs in stems and pedicels ( Fig. S5c and d). Scanning electron microscopy showed the increase of number of epidermal hairs of pedicels in the PhCESA3-silenced plants (Fig. 6i). In addition, scanning electron micrographs showed collapsed epidermal hairs on PhCESA3-silenced plants, which may be attributed to the thin and fragile cell walls in PhCESA3-silenced plants (Fig. 6j).
In addition, scanning electron micrographs of pollen grains showed that parts of the pollen cell wall were collapsed in PhCESA3-silenced plants, but the pollen cells in control plants were plump (Fig. 6k), which may partially explain the sterility of PhCESA3-silenced plants.
PhCESA3 silencing alters the ultrastructure of cells. To investigate if PhCESA3 silencing affects the ultrastructure of cells, we examined the thickness of cell walls. Transmission electron micrographs of cell walls revealed that the epidermal cell walls of stems, petal limbs and filaments in PhCESA3-silenced plants were thin compared to those of control plants (Fig. 8a,e,f,g and h; Table 4), while the thickness of the cell walls of cortical cells and xylem cells of stems was not significantly different between the plants ( Fig. 8b and d; Table 4). In addition, we observed that the starch granules in chloroplasts of stem cortical cells in PhCESA3-silenced plants were rare compared to control plants (Fig. 8c).

Discussion
CESA3, which encodes the catalytic subunit 3 of cellulose synthase, has been shown to be involved in the biosynthesis of primary cell walls in A. thaliana [12][13][14][15] . We isolated a petunia homolog of the A. thaliana CESA3 gene (PhCESA3), and we analyzed the spatial and temporal regulation of PhCESA3 expression. Furthermore, we produced VIGS-mediated PhCESA3-silenced plants and compared these plants to control plants to examine the role of PhCESA3 in plant growth and development.
The predicted peptide sequence of PhCESA3 shows high similarity to A. thaliana CESAs throughout the protein with a putative zinc-binding domain and two putative transmembrane helices in the N-terminus as well as six putative transmembrane helices in the C-terminus (Fig. S1). PhCESA3 has the homology with AtCESA3 in A. thaliana, which indicates that PhCESA3 is the homolog of AtCESA3. PhCESA3 mRNA was detected in all organs examined but was present at different levels in different organs (Fig. 2), indicating that PhCESA3 expression is spatially regulated.
In the flowers of plants treated with TRV-PhCESA3, PhCESA3 mRNA levels showed decreased compared to those of control flowers, while the mRNA levels of other PhCESAs were not significantly changed compared to those of control flowers (Fig. 4), which showed the specific silencing of PhCESA3 in plants treated with TRV-PhCESA3. In A. thaliana, mutants of AtCESA4, 7, and 8 specifically show a cellulose defect in the secondary wall of the xylem 15,23,24 , whereas AtCESA1, 3 and 6 are mainly involved in cellulose synthesis in the primary cell wall 5,25,26 . AtCESA1 and AtCESA3 are required for cellulose synthesis in the primary cell wall, whereas AtCESA2 and AtCESA6 may be at least partially redundant 9 . Null alleles of AtCESA3 and AtCESA1 are embryo lethal 27,28 . In A. thaliana, the rsw1 mutant (the temperature-sensitive allele of AtCESA1) the antisense AtCESA3 line, and the cev1 mutant (a mild mutation of AtCESA3) results in dwarfing of plants, small leaves, small petals, short pedicels, short filaments, short styles, and reduced fertility compared to control plants [12][13][14][15] . The short stamen filament surface of the rsw1 mutant is crumpled 29 . In tobacco, using the VIGS-method silencing of NtCESA1, the homolog of AtCESA3, leads to shorter internode lengths, small leaves, and a dwarf phenotype 30 . In this study, similar results were obtained in the PhCESA3-silenced plants (Fig. 3a-e,g,i-l). Both the antisense AtCESA3 line and rsw1 mutant affect female reproductive competence more severely than male competence in A. thaliana 13,29 , which is consistent with the results in this study (Table 2). However, PhCESA3 silencing caused swollen stems, pedicels, filaments, styles and epidermal hairs, which were not observed in the AtCESA3 antisense line, the cev1 mutant and the rsw1 mutant in A. thaliana [12][13][14][15]29 . Further, scanning, transmission and optical micrographs showed that PhCESA3 silencing caused short and swollen cells of these organs or tissues, which was consistent with the corresponding phenotypes. Thus, PhCESA3 silencing resulted in primary cell wall extension in length and width. Importantly, growing the rsw1 mutant at its restrictive temperature inhibits root and hypocotyl elongation as well as promotes radial swelling 29 . Mutations in A. thaliana AtCESA3 and AtCESA6 confer resistance to isoxaben (an herbicide that  specifically inhibits cellulose) and isoxaben, which causes radial swelling of roots 15 . Taken together, with strong inhibition of cellulose synthesis, whether genetic or chemical (i.e., with herbicides such as isoxaben), organ elongation is inhibited and radial expansion is stimulated. However, the organs that exhibit radial expansion induced by CESA silencing are different in various species. Among the ten AtCESAs in A. thaliana, AtCESA3 (IXR1/CEV1) is suggested to be essential for depositing cellulose in primary walls 10 . When changes to primary walls alter cell expansion and/or cell divisions, changes to cell shape, growth, and morphogenesis occur. Changes to thickening after growth stops occur too late to affect these processes, resulting in normal morphology but changes in mechanical properties 13 . Similar to the A. thaliana AtCESA3 mutant, PhCESA3 silencing resulted in significant changes in the cell size and shape in several organs. Moreover, the cellulose content of mature stems and leaves from PhCESA3-silenced plants was significantly decreased compared to that of control plants. These results suggested that PhCESA3 mainly functions for catalyzing the biosynthesis of cellulose deposited to the primary wall. Ultrastructural studies clearly showed that the cell walls of the epidermis in stems and petals, which are heavily thickened during primary growth 31 , were thinner, thus further supporting the involvement of PhCESA3 in depositing cellulose in the primary wall. The thickness of the cell walls of cortical cells and xylem cells of stems did not significantly change ( Fig. 8b and d; Table 4), suggesting that PhCESA3 is not involved in depositing cellulose in thickening walls after cell growth stops.
In this study, the shorter stems, pedicels, filaments and styles in PhCESA3-silenced plants were attributed to reduction in cell length rather than cell number, and these swollen organs in PhCESA3-silenced plants were attributed to increased cell width rather than cell number (Fig. 6a-d; Fig. 7a,b,e,f,j and l). These results implicated that PhCESA3 is mainly involved in cell elongation rather than cell division, which is consistent with results obtained with the antisense AtCESA3 line in A. thaliana [12][13][14][15] . In addition, plant organ shaping requires control over cell wall expansion anisotropy, which is characterized by the direction and degree of anisotropy 32 . In this study, the inhibited elongation and stimulated radial expansion of several organs in PhCESA3-silenced plants showed the change of the direction and degree of anisotropy of cell wall.
Here, PhCESA3 silencing resulted in changes of plant height, flower size and fertility, which are the important features of ornamental plants. These results suggested the important role of PhCESA3 in controlling plant stature and form. By using specific promoters, changes in CESA3 expression in different organs may create an excellent variety for ornamental plants. Therefore, PhCESA3 may be utilized in petunia breeding for changing the plant height, flower size and fertility.  RNA extraction, RT-PCR and cloning of the petunia PhCESA3 gene. Total RNA was extracted and reverse-transcribed according to the methods of Liu et al. 33 . The partial sequences of PhCESA3 were obtained using a computational identification approach 34 . Briefly, TBLASTN analysis against the GenBank EST database (http://www.ncbi.nlm.nih.gov) with A. thaliana CESA3 identified 1 petunia clone, FN016492, which encodes putative proteins that displayed conservation with A. thaliana CESA3. The remaining 5′ and 3′ cDNA sequences of PhCESA3 were isolated by RACE using a specific primer (Table S2), and its full-length cDNA was isolated by RT-PCR. Sequence analysis. Alignments were performed, and a phylogenetic tree was generated using the DNAMAN software. Identity search for nucleotides and translated amino acids was performed using the National Center for Biotechnology Information (NCBI) BLAST network server (http://www.ncbi.nlm.gov/BLAST).
Quantitative real-time PCR assays. Quantitative real-time PCR (qPCR) assays were performed according to previous methods 33 . Analyses were conducted following the Minimum Information for Publication of Quantitative Real-Time PCR Experiments guidelines 35 . Petunia Actin (accession no. FN014209) and Cyclophilin (CYP) (accession no. EST883944) genes were used as the internal reference genes to quantify the cDNA abundance 36 . Similar results were obtained for both reference genes. The data presented in the body of the text represent relative expression values calculated using Actin, and the data presented in the attachment represent relative expression values calculated using CYP. The sequences of all primers used for qPCR analysis are described in Table S3.

Agroinoculation of TRV vectors.
To generate pTRV2 containing the 3′ untranslated region of PhCESA3 (TRV2-PhCESA3), the gene sequence of 266 bp was PCR amplified using forward primers and reverse primers (Table S4), and the PCR products were inserted into the pTRV2 vector. Agrobacterium tumefaciens (strain GV3101) transformed with pTRV1 and pTRV2 derivatives were prepared as previously described 22 Cellulose measurement. Cellulose content in mature stems and leaves was measured by a previously published method 38,39 . Analysis was conducted with material from a single plant using three to five fully expanded leaves (from 5-week-old plants) or a 3 cm segment from the base of the primary inflorescence stem. Stems were chopped with a razor blade, and a crude cell wall fraction was obtained by extracting the soluble material with two changes of 70% ethanol at 70 °C for 1 h each 40 . The ethanol was removed, and the samples were dried under vacuum. The dry weight of the wall material was recorded. Cellulose was measured according to a previously published method 39 .
Scanning electron micrographs. The stems, leaves, pedicels, petal limbs, filaments, styles and anthers from PhCESA3-silenced plants and control plants were cut into 3-5 mm 2 pieces. The samples were fixed in 4% glutaraldehyde in 0.1 mol/L PBS (pH 7.2) for 4 h at 4 °C and then washed 3 times in the same buffer, which was followed by post-fixation in 1% osmium tetroxide for 2 h at room temperature and 3 rinses using the same buffer. Samples were dehydrated in increasing grades of ethanol and then dried with a critical point drier (CPD 030, Switzerland, Bal-Tec). The dried samples were fixed on the sample stage and coated with gold by ion sputtering equipment. Samples were observed with a scanning electron microscope (XL-30-ESEM, The Netherlands, FEI) at 10 kV acceleration and photographed.  Paraffin sections. The stems, leaves, pedicels and styles from plants were cut into 5 mm × 5 mm × 5 mm pieces. The samples were fixed in FAA fixative solution (every 100 ml of FAA fixative solution contains 90 ml of 50% or 70% ethanol, 5 ml of acetic acid, and of 5 ml formalin) for 24 h at room temperature and then washed in running water for 24 h. Samples were stained by hematoxylin for 4 d and then washed in running water for 24 h, which was followed by dehydration in increasing grades of ethanol. A graded chloroform series was used for clearing, and the samples were embedded in paraffin. Paraffin sections were cut to a thickness of 8 μ m on a Leica RM2235 followed by dewaxing with xylene. Lastly, slides were sealed with neutral resin. Sections were observed and photographed with a Zeiss Scope.A1 microscope. Semi-thin sections. The pedicels, petal limbs and filaments from PhCESA3-silenced plants and control plants were cut into 1 mm × 1 mm × 0.5 mm pieces. The specimens were fixed in 2.5% paraformaldehyde/3.0% glutaraldehyde in 0.1 mol/L PBS (pH 7.2) for 4 h at 4 °C, and the specimens were then washed 3 times in the same buffer, which was followed by post-fixation in 1% osmium tetroxide for 2 h at room temperature and 3 rinses using the same buffer. The specimens were dehydrated in a graded ethanol series and embedded in Epon812 (SPI

TRV (Control) TRV-PhCESA3 TRV-PhCESA3/TRV (%)
Epidermal cell walls of stems (μ m) 5  Supplies Division of Structure Probe Inc., West Chester, PA, USA). Polymerization took place for 24 h at 40 °C, which was followed by 24 h at 60 °C. Specimens were cut to a thickness of 1 μ m on a Leica RM2155 and were stained with 0.5% toluidine blue. Sections were observed and photographed with a Leica DMLB microscope.
Transmission electron micrograph. The stems, petal limbs and filaments from PhCESA3-silenced plants and control plants were cut into < 1 mm 3 pieces. The samples were fixed in 4% glutaraldehyde in 0.1 mol/L PBS (pH 7.2) for 4 h at 4 °C and then washed 3 times in the same buffer, which was followed by post-fixation in 1% osmium tetroxide for 2 h at room temperature and 3 rinses using the same buffer. Samples were dehydrated in increasing grades of ethanol and embedded in Epon812 (containing 51.64% Epon812, 5.37% DDSA, 42.99% MNA and 1.5% DMP-30). Polymerization took place for 24 h at 45 °C, which was followed by 24 h at 60 °C. Ultrathin sections (100 nm thick) were cut with a Leica ULTACUT ultramicrotome and deposited on a 200 mesh copper net with support film. Samples were stained with a 2% uranyl acetate solution followed by a 6% lead citrate solution, and samples were visualized with a Tecnai 12 transmission electron microscope (FEI, Eindhoven, The Netherlands) at 80 kV acceleration and photographed.