We used the GC gene sequence of Arabidopsis thaliana (AtGC1) as the query condition to search the jujube genome database. Only one candidate GC fragment cDNA sequence (828 base pairs) was determined by homologous cloning from the total RNA of ‘Dongzao’ jujube, and sequenced after transformation into Escherichia coli DH5α. The predicted CDS has 100% identity with the above 828 bp sequence. Finally, the CDS was isolated and sequenced by PCR and named as ZjGC.

The ZjGC protein conserved domains was used to search in NCBI and found that ZjGC belongs to Guanylate cyc 2 superfamily (GC-2) which belongs to soluble GC (sGC). We used the MEGA software to compare the amino acid sequence of the ZjGC with the known GCs of animals and plants. From Figure 1 , we can see that GCs are mainly divided into 4 main groups: group I is mainly composed of plant GCs, group II and group III are mainly composed of animal GCs, group IV includes GCs from animals and some algae and insects. Of them, the jujube GC (ZjGC) is closest to peach on the branch of group I.

To elucidate whether the ZjGC regulates cGMP synthesis, in vivo white-mature stage jujube fruits in three independent trees were used for transient overexpression analysis. After 24 h of treatment ( Figure 2A ), as shown in Figure 2B , the expression of ZjGC was increased in the transgenic fruits, and the contents of cGMP were significantly higher than those of the negative control ( Figure 2C ), proving that the ZjGC can catalyze the synthesis cGMP.

We cloned, expressed, and purified the ZjGC protein. A main band with an apparent molecular mass of 25 kDa was recognized from SDS-poly-acrylamide gel electrophoresis ( Figure 2D ). The size is consistent with the predicted theoretical value. Its enzyme activity was then determined by detecting the cGMP content of the reaction product, and the calibration curve covered the range of cGMP measurements in the experiments. Meanwhile, we figured out that in the presence of divalent cation, Mn²⁺ was most helpful for ZjGC activity ( Figure 2E ), and could assist protein to approach its highest activity of 0.098 nmol/min/μg. Mg²⁺ as a cofactor, could also provide effective support, the ZjGC activity reached 0.076 nm/min/μg. Compared to Mn²⁺ and Mg²⁺, the enzyme activity with Ca²⁺ as a cofactor was the lowest. The results showed that recombinant ZjGC can catalyze the synthesis of cGMP from GTP.

To further investigate the expression regulation mode and the biological functions of ZjGC, we generated ZjGC overexpression transgenic tomato plants using 35S::ZjGC-GFP and 35S::GFP. After the Agrobacterium-mediated genetic transformation, the positive tissue culture seedlings were verified by RT-PCR, indicating that the expression vector constructed had been integrated into the tomato genome. The transgenic plant seedlings with the highest GC expression level and healthy physiological state were selected for subsequent experiments and analyses.

From Figures 3A, B we can see that all the ZjGC overexpressing transgenic lines showed much earlier seed germination than wild type (WT), suggesting that the cGMP may function as a positive regulator for seed germination.

The flowering time of ZjGC overexpression transgenic lines was significantly earlier, and the size of the flowers and leaves did not change ( Figure 3C ). The average flowering time of WT and 35S::GFP was about 49 d after seed sowing, but 35S::ZjGC-GFP was only 43.3 d after seed sowing, nearly 6 days earlier than that of the control ( Figure 3D ). In addition, the transgenic 35S::ZjGC-GFP tomato plants were significantly shorter than the WT plants ( Figures 3E, F ), so we measured the flowering node position and inter-node spacing of all plants. As shown in Figure 3G , the flowering node position of 35S::ZjGC-GFP tomato plants was 6, which was 1.7 nodes less than that of the control. The average node length of 35S::ZjGC-GFP was also the shortest, which was only 0.85 cm, about 0.57 cm shorter than WT. These results suggest that the ZjGC gene positively regulates plant phase change and flowering.

The average time for WT and 35S::GFP to bear fruit after seed sowing were 58.7 d and 56.7 d, respectively, but only 51.3 d for 35S::ZjGC-GFP, 5–7 d earlier ( Figure 3H ). The average time from seed sowing to fruit coloring of WT and 35S::GFP was 98.3 d and 97.3 d ( Figure 3I ), while 35S::ZjGC-GFP was only 88 d, 9–10 days earlier. From fruit setting to fruit color change, the development period of WT fruit was 39.6 d, 35S::GFP was 40.6 d, and ZjGC overexpression transgenic fruit was only 36.7 d. The results showed that the ZjGC gene has the function of accelerating fruit growth and development. Also, we can see that ZjGC affects the fruit size and appearance ( Figure 3J ), the ZjGC overexpression transgenic fruit showed significant smaller fruit ( Table 1 ), thinner mesocarp and lighter skin color in mature fruit.

Besides, it was found that the number of seeds in the transgenic tomato fruit was significantly higher than that in the WT. The average number of seed in the WT and 35S:: GFP tomato fruits was 26 and 31.7, while the 35S:: ZjGC-GFP tomato fruit was 36.5, increasing over 40% compared to the WT ( Figure 3K ). Therefore, the results demonstrated that the ZjGC has a biological function in promoting seed formation. In addition, ZjGC also affects seed size ( Figure 3L ). The transverse and longitudinal diameters of WT tomato seeds were 2.2 mm and 3.4 mm, respectively, while those of 35S::GFP seeds were 2.1 mm and 3.2 mm. In contrast, the seeds of the transgenic tomatoes were significantly smaller than the former two, with transverse and longitudinal diameters of 1.7 mm and 2.7 mm, respectively.

On the 10th day after treatment, the medium added with cGMP of 200 and 300 μM had the greatest effect on the calli growth rate, both of them were all reached 68% ( Table 2 ). On the 20th day, the best one was 100μM cGMP, with calli growth rate reaching 145%, 1.79 times of the control. The results showed that cGMP could promote the growth of calli.

As shown in Figure 4 , exogenous cGMP could cause morphological changes of jujube calli. For example, on the 20th day after treated with 100μM cGMP, jujube calli were tight and rigid with many small cells, presenting a type I callus shape that is easy to differentiate into buds.

It can be seen from Figure 5 that the content of cGMP in the overexpressing transgenic tomato plants was extremely higher than that of the control, indicating that ZjGC can catalyze the generation of cGMP in transgenic tomato plants. The hormone content analysis showed that IAA content and ABA content in 35S::ZjGC-GFP overexpressing transgenic fruits were significantly lower than those in WT by 0.005 ng/g (7.39%) and 0.07 ng/g (16.21%), respectively. GA4 content was also lower than WT. However, GA3 content in overexpressed tomato fruits was 0.92 ng/g (2.08-fold) higher than that in WT and 0.74 ng/g (1.71-fold) higher than that in 35S::GFP.

From Figure 5 , we can see that exogenous cGMP can cause a significant increase in cGMP content in jujube calli. However, on the day 20, IAA contents in calli treated with cGMP of 200μM and 300μM were significantly decreased by 15.15% and 16.41%, respectively. Exogenous cGMP also increased the content of ABA on the 10th day, with 100μM and 300μM more efficient, increasing 26.05% and 32.27%, respectively. The highest effect of cGMP was found in GA3. On the 10th day, the content of GA3 in the calli treated with 200μM and 300μM cGMP increased by 74.68% and 134.22% compared with CK, respectively. On the 20th day, the content of GA3 in the jujube calli treated with 100μM cGMP even increased 293.87%. Furthermore, exogenous cGMP treatment at concentrations of 200μM and 300μM on the 10th day increased the content of GA4 in jujube calli by 21.13% and 17.93%, respectively. In addition, we found that exogenous cGMP did not affect ZT content in jujube calli on the 10th day, and only a slight increase on the 20th day.

We extracted RNA from ‘Dongzao’, a leading variety of Chinese jujube whose genome fully sequenced and assembled by our group (Liu et al., 2014; Yang et al., 2023), and used it as a template to clone candidate GC genes. White-ripe stage fruits were used for transient overexpression analysis. The wild type Micro-Tom was obtained from Biorun Biotechnology Company (Wuhan, China), and the Micro-Tom genetic transformation system was used to obtain transgenic tomato plants. Jujube calli were cultured at the Research Center of Chinese Jujube, Hebei Agricultural University, China, and used for exogenous cGMP treatment.

Total RNA was isolated directly from jujube fruit and tomato tissues or organs using the RNA prep Pure Plant Plus Kit (TIANGEN Biotech Beijing Co., Ltd.) according to the manufacturer’s protocol. The ZjGC gene was amplified and cloned using “Dongzao” as a template.

(1) Identification of the complete coding sequence (CDS) of ZjGC. The partial GC gene sequence of Arabidopsis thaliana was used as the query condition to search ZjGC from jujube (Ludidi and Chris, 2003). (2) The amplification of the cDNA sequence of the ZjGC fragment. The ZjGC fragment cDNA sequence was obtained from jujube genome using homologous cloning. Primer pairs were designed using Primer Premier 5.0 software. (forward: 5 ‘ATGTGGCCTTTATATATCCTTTTCA - 3, the reverse: 5’ - CTAAGCAGAAAGCTGTAGAGAATTG - 3, N is degenerate base), according to the conserved nucleotide sequences of putative cloned GCs in the NCBI database (National Center for Biotechnology Information). The PCR products were ligated into the pMD19-T vector Cloning Kit (Takara) and sequenced. (3) To confirm the CDS of ZjGC, PCR amplification was performed with specific primers designed according to the predicted ZjGC CDS. The primer pairs were as follows: 5′- TACAGCATTCAGTCTTTGGC-3′ (forward) and 5′- GAATAGCGTCTCCACTCGTA-3′ (reverse). The PCR product ligation and sequencing were the same as those described above.

Phylogenetic analysis of the catalytic domains of GCs. The ML (Maximum likelihood) tree was constructed from the amino acid sequences of ZjGC using MEGA 5.2 with 1000 bootstrap replicates, using the online website (https://www.chiplot.online/) to render the color and shape of evolutionary trees (Xie et al., 2023).

The complete coding sequences of ZjGC genes were ligated into the green fluorescent protein (GFP) vector pCambia1302 with 35S promoter (GENCEFE Biotech Wuxi, China), and transferred into Agrobacterium tumefaciens strains GV3101 (Keylab Biotech, Jiangsu, China) by the freeze-thaw method. After that screened for positive transformants in LB agar plate supplemented with kanamycin (100 μg/mL) and rifampicin (100 μg/mL). Single colonies were picked and incubated into 20 mL LB medium at 28 °C, 200 rpm, then detected by PCR and harvested at an OD600 of 1.5 to 2. A. tumefaciens was resuspended for transient overexpression of ZjGC in jujube fruit in vivo. At the same time, PCR-positive colonies were inoculated into a 20 mL LB medium containing 100 μM acetosyringone (Sigma-Aldrich) at 28 °C, with shaking at 200 rpm. After centrifugation, the bacteria were re-suspended at an OD600 of 0.6 to 0.8 with MS liquid medium for the transformation of cotyledons (Wang et al., 2024).

Three independent Z. jujuba “Dongzao” trees cultivated at the ZanHuang jujube Experiment Station, were used for transient overexpression analysis. Each fresh white mature fruit was injected with 2 mL of A. tumefaciens strain GV3101 solution containing 35S::ZjGC-GFP using a syringe. 35S::GFP was injected as a negative control. After infiltration, the fruits were covered with aluminum foil for 24 h before removal. Samples were collected before injection (0 h), 24 h, 48 h, and 72 h after injection, when collecting, it is necessary to select fruits with consistent growth, each treatment required 5 fruits to be collected from each tree. Samples then were ground in liquid nitrogen for analysis of ZjGC expression level and cGMP content.

CDSs of ZjGC were synthesized by Anhui General Biological Systems Co., LTD and inserted into the pET15a vector. The construct was transferred to BL21 (DE3) E. coli receptor cells (TIANGEN Biotech, Beijing, China), and grown at 36 °C in LB medium with 1 mM isopropyl-1-thio-β-D-galactoside (IPTG) for protein induction. Proteins were purified with nickel-nitrogen triacetic acid resin according to the manufacturer’s protocol, and then the appropriate amount of sample was to determine protein purity by SDS-polyacrylamide gel electrophoresis (Yuan et al., 2022).

The activity of GC proteins was determined by monitoring the release of cGMP from GTP. All reactions were carried out in a 500 μL mixture, recombinant protein was incubated for 20 min at 30 °C in 50 mM Tris−HCl (pH 7.5) with 0.5 mM GTP concentrations, 10 mM divalent cation (Mn²⁺, Mg²⁺, Ca²⁺) cofactor and 20 μg purified protein. Generation of cGMP was determined by ELISA Kit (LE-Y1602, Hefei Lai Er Bio-Tech Co., Ltd).

Transgenic Micro-Tom plants overexpressing ZjGC genes were regenerated from leaf disks after Agrobacterium tumefaciens-mediated transformation. The seeds were surface-sterilized by immersion for 30 s in 70% (v/v) ethanol followed by 15 min in 5% (v/v) NaClO and 5 rinses with sterile distilled water. Sterilized seeds were cultured on 0.5 × MS medium, when the cotyledons expanded fully, cut the cotyledons and placed on MS + 1 mg/L zeatin (ZT) for a 48 h preculture. Then, the cotyledon explants were subjected to Agrobacterium tumefaciens GV3101 infection solution for 15 min and transferred to MS + 2 mg/L ZT medium for a 48 h co-cultivation. After transformation, the cotyledon explants were put on screening media (MS + 2 mg/L ZT + 10 mg/L hygromycin + 300 mg/L cefotaxime) for regeneration. When calli with shoot buds formed from the cotyledon fragment, the cotyledons were cut off and transferred to rooting medium, MS + 1 mg/L IAA (Indole acetic acid) + 10 mg/L hygromycin + 300 mg/L cefotaxime. Finally, rooted plants were transferred to soil and grown to maturity. The Micro-Tom plants that regenerated from untransformed cotyle-dons were taken as control. The regenerated transgenic plants confirmed by PCR were planted in the greenhouse. Only the homologous T2 lines were used for phenotyping, Tomato RNA was extracted using RNA prep Pure Plant Plus Kit. All fresh plant samples were ground in liquid nitrogen completely and stored at -80 °C for further use.

The jujube seed-induced calli were selected as the material for the calli test. The subculture medium of calli was MS 4.23 g/L, AGAR 5.5 g/L, maltose 20 g/L, trace elements (M2, M3, M4, M5), and the pH value was adjusted by NaOH (5.8). Plant growth regulators: naphthalene acetic acid (NAA), thidiazuron (TDZ). The culture medium was treated with different concentrations of cGMP (0, 50,100,200,300 μM) (Ma, 2007; Jia and Ma, 2020), and the calli was cultured in Petri dishes. For each group, three dishes were set up and repeated three times. The materials were incubated at 25 °C for 10 d and 20 d. Before sampling, the morphological changes of calli were observed and photographed. All samples were carefully picked up with tweezers and put into new petri dishes, weighed and recorded. Rinsed briefly with small water flow, and used dust-free paper to repeatedly absorb the water, then frozen in liquid nitrogen and powdered.

To measure cGMP in plant samples, 1 g of fresh samples were ground in liquid nitrogen completely, and mixed with 5 mL of water. The mixture was subjected to a 60 °C water bath for 3h, after centrifugation at 12000 rpm for 10 min, the supernatants were filtered through a 0.22 μm filter and tested. The cGMP content was determined by HPLC (Agilent 1200 Series) (Liu et al., 2023).

In the experiment, methanol was used as the extraction agent to prepare a standard mixture, the mixed liquids included IAA (Indoleacetic acid), ABA (Abscisic acid), GA3 (Gibberellic acid 3), GA4 (Gibberellic acid 4) and ZT (Zeatin). A total of 0.5 g frozen plant samples were ground to powder in liquid nitrogen, suspended in a 3 mL methanol mixture (Methanol/water/formic acid, 15/4/1), and extracted at 4 °C for 16 h. After 30 min of low-temperature ultrasound oscillation, centrifuged for 5 min at 4 °C, 12000 rpm, following the supernatants were transferred and concentrated in new centrifuge tubes. Then, the concentrated liquids were dissolved in 1 mL 80% methanol and passed through a 0.22 μm filter for LC - QQQ (Agilent1290-6495C) analysis (Gong et al., 2016).

The material of the tomato model was provided by Figdraw 2.0.

ZjGC promoter was analyzed online by the PlantPAN 4.0 website (https://plantpan.itps.ncku.edu.tw/plantpan4/index.html) and visualized using ChiPlot (https://www.chiplot.online/).

The data were collated using Excel software and the mean was used to represent the overall level of each trait in the sample. One-way ANOVA was used for statistical evaluations by SPSS Statistics 26.0 software. Graph Pad Prism 10 was used for Charting.