Heterotrimeric Gα-subunit regulates flower and fruit development in CLAVATA signaling pathway in cucumber

Abstract Flowers and fruits are the reproductive organs in plants and play essential roles in natural beauty and the human diet. CLAVATA (CLV) signaling has been well characterized as regulating floral organ development by modulating shoot apical meristem (SAM) size; however, the signaling molecules downstream of the CLV pathway remain largely unknown in crops. Here, we found that functional disruption of CsCLV3 peptide and its receptor CsCLV1 both resulted in flowers with extra organs and stumpy fruits in cucumber. A heterotrimeric G protein α-subunit (CsGPA1) was shown to interact with CsCLV1. Csgpa1 mutant plants derived from gene editing displayed significantly increased floral organ numbers and shorter and wider fruits, a phenotype resembling that of Csclv mutants in cucumber. Moreover, the SAM size was enlarged and the longitudinal cell size of fruit was decreased in Csgpa1 mutants. The expression of the classical stem cell regulator WUSCHEL (WUS) was elevated in the SAM, while the expression of the fruit length stimulator CRABS CLAW (CRC) was reduced in the fruit of Csgpa1 mutants. Therefore, the Gα-subunit CsGPA1 protein interacts with CsCLV1 to inhibit floral organ numbers but promote fruit elongation, via repressing CsWUS expression and activating CsCRC transcription in cucumber. Our findings identified a new player in the CLV signaling pathway during flower and fruit development in dicots, increasing the number of target genes for precise manipulation of fruit shape during crop breeding.


Introduction
Flowers and fruits are the reproductive organs in plants and play essential roles in natural beauty and the human diet.The CLAVATA (CLV)-WUSCHEL (WUS) signaling pathway regulates f lower and fruit development by modulating shoot apical meristem (SAM) size [1].The SAM is located at the shoot tip and is responsible for the continuous organogenesis of all aerial organs.The central zone of the SAM contains a mass of pluripotent stem cells that maintain themselves and replenish cells for organogenesis.The peripheral zone of the SAM produces leaves during the vegetative stage and f lowers during the reproductive phase [2].CLV3 is a peptide that belongs to the CLAVATA3/EMBRYO SURROUNDING REGION (ESR)-related (CLE) peptide family, secreted by stem cells and predominantly expressed in the central zone [3].The CLV3 peptide can be perceived by a leucine-rich repeat receptor kinase, CLV1 [4,5].WUS encodes a homeodomain transcription factor that is expressed at the organization center of the SAM [6,7].CLV1 binding to CLV3 will result in suppression of WUS activity, whereas WUS stimulates the expression of CLV3, thus forming a negative feedback loop [8,9].CLV2 encodes a leucine-rich repeat-like protein without a kinase domain that plays a parallel role to CLV1 during SAM regulation [10].
Mutations in the CLV-WUS signaling pathway genes generally lead to alteration of SAM size, changes in f loral organs, and fruit development.For example, functional disruption of CLV3homologous genes resulted in enlarged meristem, increased f loral organ numbers, and greater fruit diameter in Arabidopsis, rice, and tomato [11][12][13][14][15].Likewise, enhanced expression of ZmWUS1 may cause a larger SAM in maize [16], while a gain-of-function mutation of SlWUS underlying the locule number (lc) locus led to fruits with more locules [17].Therefore, functions of the CLV3-WUS pathway genes are widely conserved in f lowering plants.Fine tuning of the CLV3-WUS pathway genes is of significant potential for crop breeding to produce f lowers with extra organs and enlarged fruits.
Several genes have been found to function downstream of CLV signaling during SAM development and maintenance.POLTERGEIST (POL) encodes a Protein Phosphatase 2C (PP2C)that functions downstream of CLV1 receptor by transmitting signals from CLV1 to the nucleus [18].Recent studies indicated that receptor-like cytoplasmic kinase (RLCK) PBLs can undergo phosphorylation by CLV1 to mediate Arabidopsis meristem size [19,20].Another reported downstream component of the CLV signaling pathway are heterotrimeric G proteins, which are composed of Gα, Gβ, and Gγ subunits.Mutations of G proteins result in rounded and wrinkled leaves, reduced organ size and decreased hypocotyl length in Arabidopsis [21][22][23][24][25]. Gβ was shown to interact with RECEPTOR-LIKE PROTEIN KINASE 2 instead of CLV1 to regulate SAM size in Arabidopsis [26].In maize, Gα protein COMPACT PLANT2 (CT2) transmits the stem-cellrestrictive signal from CLV receptors to modulate meristem size and f loral development.Mutations in CT2 resulted in fasciated ears, increased tassel branches and spikelet density, resembling the clv mutant phenotypes [27].However, whether Gα functions in the CLV signaling pathway to regulate SAM size and f loral organ numbers remains unknown in dicots.
Cucumber is an important vegetable crop cultivated worldwide for over 3000 years [28].Cucumber f lowers and fruits are produced from the leaf axil, and fruits develop from the inferior ovary with three fused carpels.Cucumber fruit is generally harvested when immature (1-2 weeks after anthesis) and consumed fresh or after processing by pickling [29].Fruit shape consists of fruit length, fruit diameter, and fruit shape index, which directly affect cucumber appearance, quality, and commercial value [30].Here, we found a new player, a Gα-subunit in the CLV signaling pathway that regulates f lower and fruit development in cucumber.We further demonstrated that the Gα-subunit interacts with CLV1 to modulate f loral organ number and fruit length by inhibiting CsWUS expression while promoting CsCRC expression in cucumber.

Functional disruption of CsCLV3 by gene editing resulted in increased floral organ numbers and stumpy fruits in cucumber
Our previous studies using RNAi showed that the CsWUS-CsCLV3 pathway plays an essential role in carpel number determination in cucumber [31].Due to the intrinsic feature of instability of RNAi lines, stable Csclv3 mutants were obtained, using the CRISPR/Cas9 system, to dissect how CsCLV3 peptide transmits the signal during f lower development.Two independent homozygous lines were chosen for further characterization.Csclv3#1 has a 1bp insertion and Csclv3#2 has a 1-bp deletion, leading to premature termination of protein translation and a frameshift mutation, respectively (Fig. 1A and B).No mutations were detected in the potential off-target sites of the PCR sequencing products (Supplementary Data Table S2).Compared with wild type (WT), the Csclv3 mutants displayed dramatically increased numbers of f loral organ, including sepals, petals, stamens, and carpels (Fig. 1C-I).WT f lowers generally produce 5 petals, 5 stamens, and 3 carpels, while Csclv3 f lowers exhibit 9-11 petals, 10-12 stamens, and 8-10 carpels.The f loral phenotype is much severe than that of RNAi lines previously reported [31].More interestingly, the fruits of Csclv3 mutants were shorter and wider, displaying stumpy fruit with significantly reduced fruit shape index compared with WT (Fig. 1J-M).

CsCLV1 negatively regulates floral organ number but positively controls fruit shape index in cucumber
To investigate the downstream signaling components of CsCLV3 peptide, we chose its receptor CLV1 homolog (CsCLV1) in cucumber for characterization.qRT-PCR demonstrated that CsCLV1 was expressed ubiquitously in different organs (Supplementary Data Fig.S1A).In situ hybridization showed that CsCLV1 was enriched in the SAM and f loral meristem (FM), especially in the deep layer of the central zone (Fig. 2A, Supplementary Data Fig.S1C), similar to CLV1 in Arabidopsis [4].CsCLV1 signals were also detected in f loral organ primordia, with strong signals in developing carpels, nectaries, and ovules (Fig. 2B and C, Supplementary Data Fig.S1C-H).No signals were detected in the negative control (Supplementary Data Fig.S1B).
To explore the biological function of CsCLV1 in cucumber, we engineered mutations in CsCLV1 with two single-guide RNAs (sgRNAs) using the CRISPR/Cas9 system.Two homozygous mutants were obtained, Csclv1#1 (deletions of 4 and 2 bp) and Csclv1#2 (deletions of 7 and 6 bp), both resulting in premature termination (Fig. 2D, Supplementary Data Fig.S1I).No mutations were detected in the potential off-target sites of the PCR sequencing products (Supplementary Data Table S3).Similarly to Csclv3 mutants, Csclv1 mutants showed significantly increased f loral organ numbers.The average numbers of sepals, petals, stamens, and carpels were 5.0, 5.1, 5.1, and 3.2 in WT, while in the Csclv1 mutants they were 5.6, 6.0, 6.4, and 5.5, respectively (Fig. 2E-K).
Notably, fruit length was significantly reduced in Csclv1 mutants (Fig. 3A and B), with an average of 42.7% decrease at anthesis (Fig. 3C), and 26.8% reduction at 40 DAA compared with WT (Fig. 3D).However, the average fruit diameter of Csclv1 mutants increased by 24.0% at 40 DAA (Fig. 3E).Consequently, the fruit shape index was significantly reduced in Csclv1 mutants (Fig. 3F).To investigate the reason for reduced fruit length in Csclv1 mutants, longitudinal sections of fruits at 40 DAA were analyzed.Our data indicated that cell size was greatly decreased in the Csclv1 mutants compared with WT (Fig. 3G and H), implying that CsCLV1 promotes fruit elongation through cell expansion in cucumber.

CsCLV1 physically interacts with CsGPA1 at protein level
To further identify the signaling components of the CsCLV3-CsCLV1 pathway, we screened for CsCLV1 companions using a yeast two-hybrid library based on its intracellular kinase domain (Supplementary Data Fig.S2A).Sequencing and annotation of 116 positive clones were captured (Supplementary Data Table S1).Among them, the guanine nucleotide-binding protein α-subunit (GPA1) and GTP-binding nuclear protein Ran were identified (Supplementary Data Table S1).The Gα-subunit (CT2) had been previously documented as a participant in the CLV signaling pathway by affecting meristem size in maize [27].We therefore speculated whether the Gα-subunit (designated CsGPA1 hereinafter) transmits CLV signaling in cucumber.The interaction between the kinase domain of CsCLV1 and CsGPA1 was further verified by yeast two-hybrid assay (Supplementary Data Fig.S2B).The bimolecular f luorescence complementation (BiFC) assay indicated that CsCLV1 and CsGPA1 interact at the plasma membrane.Subsequently, firef ly luciferase complementation imaging (LCI) and co-immunoprecipitation (Co-IP) assays further confirmed the interaction of CsCLV1 and CsGPA1 (Fig. 4A-C).Therefore, CsGPA1 could physically interact with CsCLV1 at protein level.

Functional disruption of CsGPA1 produced a similar phenotype to Csclv1 mutants
Phylogenetic analysis showed that GPA1 homologs in dicotyledons were clustered into a clade distinct from than those in monocotyledons (Supplementary Data Fig.S2C).CsGPA1 was expressed ubiquitously in various tissues of cucumber (Supplementary Data Fig.S2D).In situ hybridization demonstrated that CsGPA1 signals were strongly enriched in the   SAM, FM, vascular bundles of leaf primordia and f loral organs (Fig. 5A-C, Supplementary Data Fig.S2E-H).Transcripts of CsGPA1 were also detected in developing nectaries and ovules (Supplementary Data Fig.S2I).No signals were detected in the negative control (Supplementary Data Fig.S2E).Notably, the expression pattern of CsGPA1 largely overlapped with that of CsCLV1, implying that they may function in the same pathway in cucumber.
Next, we explored the biological functions of CsGPA1 via the CRISPR/Cas9 system.Two homozygous mutant lines, Csgpa1#1 and Csgpa1#2, were obtained, both leading to frameshift mutations (Fig. 5D, Supplementary Data Fig.S2J).No mutations were detected in the potential off-target sites of the PCR sequencing products (Supplementary Data Table S4).Phenotypic analyses revealed a significantly elevated number of f loral organs in Csgpa1 mutants compared with WT.In WT, the average numbers of sepals, petals, stamens, and carpels were 5.0, 5.1, 5.1 and 3.2, while they were 5.3, 5.7, 5.5, and 4.1 in Csgpa1 mutants, respectively (Fig. 5E-K).Furthermore, both Csgpa1 mutants produced significantly shorter and wider fruits compared with WT (Fig. 6A and B).Fruit length decreased by 21.0% at anthesis and 21.7% at 40 DAA in Csgpa1 mutants (Fig. 6C and D).However, fruit diameter increased by 12.5% at anthesis and 13.2% at 40 DAA compared with WT (Fig. 6E and F), and the fruit shape index was significantly decreased in both stages of Csgpa1 mutants (Fig. 6G and H).Fruit longitudinal sections indicated that cell size was greatly reduced in Csgpa1 mutants (Fig. 6I and J).These phenotypes were similar to that of Csclv1, supporting the idea that CsGPA1 may function in the same pathway of CsCLV1 during f lower and fruit development in cucumber.In addition, the hypocotyls of Csgpa1 mutants were significantly shorter than those of WT (Supplementary Data Fig.S3A  and B).Hormonal measurements demonstrated that auxin and gibberellin levels were greatly decreased in Csgpa1 mutant hypocotyls (Supplementary Data Fig.S3C and D).Similar to the phenotype of gpa1 (Gα-subunit) and agb1 (Gβ-subunit) mutants in Arabidopsis.We also observed the round leaf shape phenotype in Csgpa1 mutants compared with the palmate leaf in WT (Supplementary Data Fig.S3E) [32].

Expression of CsWUS was elevated in the meristem of Csgpa1 mutants
Given that the maize GPA1-homologous gene CT2 regulates shoot meristem size by participating in the CLV signaling pathway [27,33], we explored shoot meristem development in Csgpa1 mutants using light microscopy and sectioning.Our data showed that shoot meristem size was significantly enlarged (an average of 36.3% increase) in diameter in Csgpa1 mutants compared with WT (Fig. 7A-D).Considering that CsWUS was reported to be the key determinant of SAM size [34], the expression of CsWUS was detected by in situ hybridization and qRT-PCR (Fig. 7E and F).As expected, the CsWUS signal was greatly enhanced in the FM and SAM of Csgpa1 compared with WT (Fig. 7E).qRT-PCR data indicated that the expression level of CsWUS increased 2.0-fold in the Csgpa1 shoot apex (Fig. 7F).Therefore, the increment of f loral organ numbers in Csgpa1 may be due to elevated CsWUS expression and enlarged SAM size in cucumber.

Expression of CsCRC was decreased in Csgpa1 and Csclv1 mutant fruits
To explore the underlying factors of shorter fruit phenotype in the Csgpa1 mutant, RNA-seq analysis was conducted on ovaries at anthesis from Csgpa1 mutants and WT.The results revealed that 61 and 143 differentially expressed genes (DEGs) were up-and downregulated in Csgpa1 mutants versus WT (fold change ≥ 2, FDR < 0.01), respectively (Supplementary Data Table S2, Supplementary Data Fig.S4A).Among the 204 DEGs, transcription factors were enriched, including YABBY, MYB, and zinc finger family genes (Supplementary Data Fig.S4B and C).Notably, the YABBY transcription factor CRABS CLAW (CsCRC) was significantly downregulated in Csgpa1 mutant fruits (Fig. 7G).
Our previous study showed that CsCRC is a major-effect fruit length regulator that stimulates fruit elongation by cell expansion in cucumber [35].qRT-PCR demonstrated that the expression level of CsCRC declined 1.9-fold in Csgpa1 and 1.6-fold in Csclv1 (Fig. 7H and I).Therefore, CsCLV1 and CsGPA1 might participate in fruit elongation by modulating CsCRC transcription in cucumber.

Discussion
The Gα-subunit (CsGPA1) is a novel regulator of floral organ number and fruit shape in cucumber Heterotrimeric G proteins are important signaling molecules that help transfer signals from outside to inside the cell by interacting with G-protein-coupled receptors (GPCRs) [24].However, due to lack of GPCRs in plants, G proteins interact with single transmembrane receptors to regulate plant growth and development [36].The functions of G proteins have been intensively explored in stress responses [25,37], but there are relatively few reports in plant organ development [36].In dicots, the function of GPA1 in organ development has only been investigated in Arabidopsis, in which mutants of gpa1 displayed stumpy hypocotyls, smaller f lowers, and slightly shortened siliques, but no changes in SAM size and f loral organ numbers [22,23,32].In our study, we found that CsGPA1 is expressed in the SAM, FM, vascular bundles of leaf primordia, and f loral organs (Fig. 5A-C, Supplementary Data Fig.S2E-H), similar to GPA1 in Arabidopsis [38].In cucumber gpa1 mutants, we also found the round leaf shape and short hypocotyl phenotypes as found in Arabidopsis (Supplementary Data Fig.S3A, B, and E) [23].Interestingly, we showed that CsGPA1 could interact with receptor kinase CsCLV1 at protein level (Fig. 4, Supplementary Data Fig.S2B).Functional disruption of CsCLV1 resulted in f lowers with extra organs and stumpy fruits (Figs 2E-K and 3A-F).Loss of function of CsGPA1 led to phenotypes similar to Csclv1 mutant (Figs 5E-K and 6A-H), suggesting that CsGPA1 is a novel player in f loral organ number and fruit shape development in the CLV signaling pathway in cucumber.

CsGPA1 negatively regulates floral organ numbers by modulating shoot apical meristem size
Variations in f loral organ numbers are usually due to two reasons, one being change in the size of the meristem, and the other being homeotic conversion of f loral organs [39].Genes in the CLV-WUS signaling pathway were shown to regulate f loral organ numbers by affecting meristem size [34].Elevated WUS expression led to an enlarged meristem and extra f loral organs in tomato and maize, and overexpression of CsWUS resulted in more carpels in cucumber [16,17,31].Here, we found that CsGPA1 interacts with the CsCLV1 receptor kinase in cucumber.Functional disruption of CsGPA1 resulted in extra f loral organs, increased meristem size, and enhanced CsWUS expression (Figs 5E-K and 7A-F), suggesting that CsGPA1 negatively regulates f loral organ numbers probably by inhibiting CsWUS-mediated meristem size in cucumber (Fig. 8).

The CsCLV1-CsGPA1 module positively regulates cucumber fruit elongation
The CRC gene belongs to the YABBY family of transcription factors containing the C 2 C 2 zinc-finger and DNA-binding domain, which has been reported to be involved in carpel and nectary development in Arabidopsis [40][41][42][43].In our previous study, CsCRC was found to promote fruit elongation through cell expansion [35,44].Here, we found that fruit length was significantly decreased in Csclv1 and Csgpa1 mutants (Figs 3A-D and 6A-D).Sectioning results showed that the reduced fruit length was caused by smaller cell (Figs 3G and H and 6I and J).Notably, CsCRC expression was significantly reduced in both Csclv1 and cytosol GPA1 Csgpa1 mutants (Supplementary Data Table S2, Fig. 7H and I).

WUS CRC nucleus
The expression patterns of CsCRC overlap with those of CsGPA1 and CsCLV1, with particularly enriched signal in developing carpel primordia and nectaries (Figs 2B and C and 5B and C, Supplementary Data Figs S1D-J and S2G-I [35].These data suggest that CsCLV1 and CsGPA1 act as a module to stimulate fruit elongation, probably through CsCRC-mediated cell expansion in cucumber (Fig. 8).
In conclusion, we identified a new player, CsGPA1, in the CLV signaling pathway during f lower and fruit development in cucumber.Specifically, the CsCLV3 peptide was received by receptor kinase CsCLV1.Gα-subunit CsGPA1 protein interacts with CsCLV1 to inhibit f loral organ number but promote fruit elongation, by repressing CsWUS expression and activating CsCRC transcription (Fig. 8).Our findings not only enrich the gene regulatory network downstream of the CLV signaling pathway, but also provide target genes for precise manipulation of fruit shape during cucumber breeding.

Plant materials
Cucumber (Cucumis sativus L.) inbred line 9930 of the North China type was used in this study.Soaked cucumber seeds were transferred to plugs after 36 h of germination in an incubator at 28 • C under dark conditions and then transplanted to the greenhouse of China Agricultural University in Beijing when the third true leaf appeared.Standard water and fertilizer management as well as pest control were applied for all cucumber seedlings.Nicotiana benthamiana plants were grown in a light incubator with 16 h light/8 h dark at 25 • C. Tobacco seedlings at 6 weeks old grown under standard management were used for biochemical analysis.

RNA extraction and quantitative real-time PCR
Different tissue samples were collected and quickly placed in liquid nitrogen for extracting total RNA using the Eastep ® Super Total RNA Extraction Kit (Promega).Two micrograms of RNA was used for synthesizing cDNA with the FastKing gDNA Dispelling RT SuperMix Kit (Tiangen, Beijing, China).The qRT-PCR assay was performed using TB Green ® Premix Ex TaqTM II (Takara, Kyoto, Japan) on the CFX384 Real-Time PCR Detection System (Bio-Rad) for gene expression analysis.Each gene had three biological replicates and three technical replicates.CsUBI was used as an internal reference gene [45].Detailed information on the primers is listed in Supplementary Data Table S6.

In situ hybridization
The shoot apexes of 20-and 35-day-old cucumber seedlings and young ovaries were harvested.Samples were fixed, sectioned, and hybridized for detecting gene expression.The detailed process was performed as previously reported [46].Specific regions of target genes were used to design probe primers, after which sense and antisense probes were synthesized by SP6 and T7 RNA polymerase using the DIG RNA Labeling Kit (Roche, Basel, Switzerland), respectively.Different f lower developmental stages were classified with reference to a previous study [47].The probe primer sequence is listed in Supplementary Data Table S6.

Cucumber genetic transformation
For the construction of knockout vectors, sgRNA-specific sequences of target genes were selected firstly on the web (http://crispr.hzau.edu.cn/CRISPR2/).PCR fragments containing two targets were generated by two pairs of primers amplified using pCBC-DT1T2 vector as template, and then inserted into the binary CRISPR/Cas9 vector pKSE402 with the BsaI site and T4 ligase.After the recombinant pKSE402 vector was chemically transferred to Agrobacterium tumefaciens strain EHA105, transgenic cucumber plants were obtained by Agrobacterium-mediated cucumber cotyledon transformation as previous described [48].T 0 transgenic plants were selected by screening cucumber seedlings containing GFP f luorescence using a f luorescence microscope, and the transgenic mutant form was verified by sequencing the PCR products of target genes.The primer information is listed in Supplementary Data Table S6.

Yeast two-hybrid assay
The intracellular kinase domain of CsCLV1 was integrated into the pGBKT7 vector, then employed as a bait protein for hybridization with a cucumber yeast library of pGADT7, following a self-activation assay.Positive clones were identified by PCR amplification and sequencing after screening on selective medium SD/−Trp/−Leu/-His/−Ade with X-α-Gal.In yeast twohybrid assays, the coding sequence of CsGPA1 was introduced into the pGADT7 vector.AD and BK recombinant vectors were co-transfected into the yeast strain AH109 and grown on SD/−Trp/−Leu medium.Protein interaction was screened on selective medium SD/−Trp/−Leu/−His/−Ade.The specific procedure was performed as described in the Matchmaker™ GAL4 Two-Hybrid System 3 & Libraries (Clontech) protocol.The primer information can be found in Supplementary Data Table S6.

Bimolecular fluorescence complementation assay
The full-length coding sequences without the stop codon of CsCLV1 and CsGPA1 were introduced into pSPYCE-35S and pSPYNE-35S vectors, respectively.Recombinant vectors were transferred into Agrobacterium strain GV3101 and co-injected into 6-weekold N. benthamiana leaves together with the p19 Agrobacterium and plasma membrane PM-mCherry as described [49].The YFP signals were detected by confocal microscopy (Olympus BX51, Japan) under an excitation wavelength of 488 nm for interaction signals and 562 nm for PM-mCherry at 48 h after infiltration.The primers used are shown in Supplementary Data Table S6.

Luciferase complementation imaging assay
The complete coding sequence without the stop codon of CsCLV1 was inserted into pCAMBIA1300-nLUC and the full-length coding sequence of CsGPA1 was integrated into pCAMBIA1300-cLUC.The recombinant vectors were then introduced into Agrobacterium strain GV3101 and co-injected with p19 Agrobacterium into 6week-old N. benthamiana leaves.After 48 h of infiltration, the abaxial surfaces of the leaves were treated with a 1 mM solution of d-Luciferin Potassium Salt (Biovision).After a period of 5 min in darkness, the interactions were captured using a CCD imaging system (MiniChemi 610, Sagecreation).Details of the primer sequences have been provided in Supplementary Data Table S6.

Co-immunoprecipitation assay
The entire coding sequence of CsCLV1, excluding the stop codon, was inserted into pCAMBIA1300-GFP vector and the full-length coding sequence without the stop codon of CsGPA1 was cloned into pCAMBIA1300-FLAG vector.The recombinant vectors were then transferred into Agrobacterium strain GV3101 and coinfiltrated with p19 Agrobacterium into N. benthamiana leaves.Samples were collected and ground into powder using liquid nitrogen after a 48-h infiltration.Following this, the samples were homogenized in extraction buffer.Immunoprecipitation was carried out using anti-GFP antibody-coated agarose beads (KTHEALTH, China; KTSM1301) at 4 • C for 3 h.The beads underwent six washes with washing buffer (50 mM HEPES [pH 7.5], 150 mM KCl, 1 mM EDTA, 0.2% [v/v] Triton-X 100, 1 mM DTT).The immunoprecipitates were separated using SDS-PAGE and detected by immunoblot with either anti-GFP (TransGen Biotech, Beijing, China; Catalog No. HT801) or anti-FLAG (Sigma-Aldrich, Burlington, MA, USA; Catalogue No. F3165) antibodies.The primer information can be found in Supplementary Data Table S6.

Histology
Fruit tissue samples were prepared at 40 days after anthesis (DAA) by fixation, embedding, and sectioning to 10-μm thickness, followed by dewaxing.Subsequently, images were captured using an Olympus light microscope.Cell counting was performed in various fields, and the number of cells per unit area was calculated.For each sample, three fields were observed, and five biological samples were examined for every representative line.

Measurement of endogenous hormones
To analyze the amounts of auxin (IAA) and gibberellic acid (GA3), we harvested and homogenized 0.1-0.3g of hypocotyls from 10day-old seedling samples of WT and Csgpa1 mutants.This was carried out in 4 ml of 80% methanol containing an antioxidant.Enzyme-linked immunosorbent assays (ELISAs) were performed to extract and quantify phytohormones, as previously described [50].Five biological replicates were measured for each line.

Transcriptome analysis
Ovaries at anthesis from WT and Csgpa1 were utilized for RNA-seq, with each sample consisting of three biological replicates.The Biomarker Technologies Corporation (Beijing, China) performed RNA library construction and sequencing with an Illumina NovaSeq 6000 platform.Transcriptomic data were analyzed on the BMKCloud platform (www.biocloud.net)

Figure 1 .
Figure 1.Phenotype identification of Csclv3 mutants in cucumber.A, B Two mutation sites in CsCLV3 generated by CRISPR/Cas9.A Genotype identification of Csclv3 mutants indicated the Csclv3#1 mutant with a 1-bp insertion and the Csclv3#2 mutant with a 1-bp deletion, leading to premature termination of protein translation and a frameshift mutation, respectively.B Schematic illustration of the two mutation forms in CsCLV3.Red arrows represent the two target sites and the red star indicates where the protein's translation is prematurely terminated.Orange horizontal bars represent amino acids translated in Csclv3 mutants, different from WT. C-E Representative phenotypes of petals (C), stamens (D), and carpel numbers (E) in Csclv3#1 mutants and WT.The numbers in the pictures represent f loral organ numbers.Scale bars = 1 cm in (C) and 100 μm in (D and E).F-I Quantification for percentages of sepals (H), petals (I), stamens (J), and carpel numbers (I) in WT and Csclv3 lines (n = 25).J Representative images of cucumber ovaries at anthesis from WT and Csclv3 mutants (scale bars, 1 cm).K-M Ovary length (K), ovary diameter (L), and ovary shape index at anthesis (M) in WT and Csclv3 mutants.Values are means ± standard deviation (n = 8).* * P < 0.01 (two-tailed Student's t-test).

stage 4 stage 7 Figure 2 .Figure 3 .
Figure 2. Expression analysis of CsCLV1 and phenotype identification of f loral organ numbers in Csclv1 mutants in cucumber.A-C In situ hybridization analysis of CsCLV1 in SAM and FM of different stages.le, leaf or leaf primordium; fm, f loral meristem; se, sepal primordium; pe, petal primordium; st, stamen primordium; ca, carpel primordium.Scale bars, 100 μm.D Two mutation sites in CsCLV1 generated by CRISPR/Cas9, both of which result in premature termination of protein translation, with red arrows representing the two target sites and red stars indicating where the protein's translation is prematurely terminated.E-G Representative phenotypes of petals (E), stamens (F), and carpels (G) in Csclv1#1 mutants and WT.Scale bars = 1 cm in (E) and 100 μm in (F and G).H-K Percentages of sepals (H), petals (I), stamens (J), and carpels (K) in WT and Csclv1 lines (n = 50).

Figure 4 .
Figure 4. CsCLV1 physically interacts with CsGPA1 at protein level.A BiFC assay.CsGPA1-YFP N , CsCLV1-YFP C , and PM-mCherry were co-infiltrated into N. benthamiana leaves.The other combination was used as a control.Green f luorescence represents the interaction signal, red f luorescence represents the plasma membrane marker signal, and orange f luorescence represents the merged outcome.B Firef ly LCI assay.CsCLV1-nluc and CsGPA1-cluc were co-infiltrated into N. benthamiana leaves and the remainder of the combinations were used as controls.Imaging in red represents strong interactions and blue indicates weak interactions.C Co-IP assay.The constructs specified were expressed in leaves of N. benthamiana, and Co-IP was performed using anti-GFP antibody.The band indicated by the red arrow represents the in vivo interaction between CsCLV1 and CsGPA1.

Figure 5 .Figure 6 .
Figure 5. Expression analysis of CsGPA1 and phenotypic analysis of f loral organ numbers in Csgpa1 mutants.A-C In situ hybridization analysis of CsGPA1 in SAM and FM at different stages.le, leaf or leaf primordium; fm, f loral meristem; se, sepal primordium; pe, petal primordium; st, stamen primordium; ca, carpel primordium; vb, vascular bundle.Scale bars, 250 μm.D Mutant alleles generated in Csgpa1 mutants by CRISPR/Cas9 in cucumber.The red inverted triangles represent the two targets located in the fourth and sixth exons of CsGPA1.The fourth exon resulted in a 92-bp deletion including the intron in both mutants, leading to loss of the fourth exon.A base was added in the sixth exon (+G and +A, respectively) and resulted in translation into different amino acids, D (aspartic acid) and N (asparagine) at the 111th position, respectively.Both forms led to frameshift mutation.Blue bars represent amino acids translated in Csgpa1 mutants, different from WT. E-G Representative phenotypes of petals (E) and stamens (F) and carpel numbers (G) in WT and Csgpa1#2 mutants.Scale bars = 1 cm in (E) and 100 μm in (F and G).H-K Percentages of sepals (H), petals (I), stamens (J), and carpels (K) in WT and Csgpa1 lines (n = 50).

Figure 7 .
Figure 7. CsGPA1 affects the expression of CsWUS and CsCRC in cucumber.A Stereoscopic images of SAM of 20-day-old cucumber seedlings from WT and Csgpa1 mutants (scale bars, 0.25 mm).lp, leaf primordium.B SAM diameter in stereoscopic images (scale bars, 100 μm).C Longitudinal sections of shoot apexes from 20-day-old WT and Csgpa1 cucumber seedlings.D SAM diameter in longitudinal sections from shoot apexes of WT and Csgpa1 mutants.E, F CsWUS expression was detected in the shoot apex of 20-day-old cucumber seedlings from WT and Csgpa1 mutants by in situ hybridization (E) and qRT-PCR (F) (scale bars, 100 μm).G Heat map of the related DEGs of transcription factors in WT and Csgpa1 lines.Members of different gene families are separated by vertical lines.H, I Gene expression analysis of CsCRC in Csgpa1 mutants (H) and Csclv1 mutants (I).* P < 0.05, * * P < 0.01 (two-tailed Student's t-test).

Figure 8 .
Figure 8.A model for the regulation of f loral organ number and fruit length by CsGPA1 in the CLV signaling pathway.The CsCLV3 peptide is received by receptor kinase CsCLV1.Gα-subunit CsGPA1 protein interacts with CsCLV1 to inhibit f loral organ number but promote fruit elongation, by repressing CsWUS expression and activating CsCRC transcription in cucumber.