The plant materials were C. reticulata ‘Zipao’ and wild species callus. ‘Zipao’ callus were white and loose ( Supplementary Figure 1a ); callus of the wild species were green and hard ( Supplementary Figure 1b ). Tobacco was used as the model plant for heterologous genetic transformation. Tobacco and C. reticulata callus were cultivated in a tissue culture room maintained at a constant 25°C, with a daily light exposure of 10-12 h and a humidity range of 65-75%, light intensity was 5000 LX. The red and blue light intensity required for subsequent experiments were 5000 LX.

RNA was isolated using Vazyme Biotech RNA extraction kit (Nanjing, China), according to manufacturer’s protocol. Utilizing the NCBI online platform, specific cloning primers P1-F (AAGCAGTGGGAGTAGAGGTG) and P1-R (TTTGGCAAAGGCGAGTAGTT) were designed. The cDNA synthesized via reverse transcription was employed as the template to amplify the full-length sequence of CrGST24. The RT-PCR parameters were established as: 1 cycle of 5 min at 94°C, 35 cycles of 30 s at 94°C, 35 cycles of 30 s at 58°C, 35 cycles of 1 min at 72°C, followed by 1 cycle of 10 min at 72°C and 1 cycle of 10 min at 10°C, within a 50 μL reaction buffer. Subsequently, a gel recovery kit (Biomed, Beijing, China) was applied to purify the target bands. Eventually, the purified target band was ligated into a cloning vector and introduced into E. coli recipient cells, and then the plasmid isolated and sequenced.

Three culture medium YS2, YS3, and YS4 containing different concentrations of IBA and 6-BA were prepared by increasing the concentrations of IBA and 6-BA individually or simultaneously on the basis of YS1 medium ( Supplementary Table 1 ). Callus of C. reticulata ‘Zipao’ and wild species were inoculated on four different culture media and cultured under white light for 2 h and one week. The 2^-(ΔΔCt) method was used to analyse the expression of CrGST24 gene in different types of healing tissues and under different concentrations of IBA and 6-BA treatments.

The callus of C. reticulata ‘Zipao’ and wild species were inoculated onto YS1 medium and put into light and dark environment for 8h. The 2^-(ΔΔCt) method was used to analyse the expression of CrGST24 gene in different light environments.

The callus of C. reticulata ‘Zipao’ and wild species were inoculated on the YS1 medium and placed under white, red, and blue light for 8 h and three weeks. The 2^-(ΔΔCt) method was used to analyse the expression of CrGST24 gene in conditions with white, red, or blue light.

The CrGST24 gene was subjected to double digestion using the restriction enzymes BamHI and PstI. Agarose gel electrophoresis, followed by gel cutting and DNA recovery with the TAKARA Gel DNA Recovery Kit (Beijing, China), linearized the cloning vectors. The cleaved product was purified via a DNA purification kit. The target fragment was inserted into the pCAMBIA1300-35S vector. This construct was introduced into E. coli to screen positive clones and conduct sequencing. Finally, the extracted plasmids were transferred into Agrobacterium tumefaciens GV3101.

The heterologous genetic transformation of tobacco via Agrobacterium was performed according to Wang et al. (2024). Ten transgenic lines were randomly selected. RNA was extracted from the transgenic tobacco following the RNA extraction kit’s instructions, reverse transcribed into cDNA, and then verified by qPCR to screen for high-expressing positive lines.

Phenotypic analysis was conducted on high expressing transgenic tobacco lines and WT tobacco to observe any phenotypic differences between them.

WT tobacco and CrGST24 transgenic tobacco leaves were cut into 0.5 cm*0.5 cm leaf discs and inoculated into normal MS medium, which was observed at 5 d intervals to analyse the in vitro regeneration of transgenic tobacco leaves.

The NCBI Primer-Blast was used to design qPCR primers for the growth hormone-related genes NtYUCCA8, NtPIN1c, NtGH3.3, NtIAA27, NtARF6, NtWOX5, and NtSAUR20; and for the cytokinin-related genes NtARR5, NtARR12, NtCRE1, NtAHP6, NtCRF6, and NtWUS qPCR primers, the NtEF1α as a reference gene ( Table 1 ).

Twelve auxin and cytokinin related genes were screened from the callus transcriptome data of C. reticulata and their expression was analyzed

Auxin and cytokinin content in transgenic tobacco and in WT were examined by ELISA enzyme-linked immunoassay. Absorbance values of auxin and cytokinin should be measured at 450 nm.

Preparation of NBT staining solution: Dissolve 50 mg NBT in 100 mL Tris buffer (pH: 7.4), and obtain NBT staining working solution after full dissolution. The prepared staining solution needs to be stored at 4°C away from light. The leaves of high expression positive transgenic tobacco and WT were cut off, and the leaves were cut into circular leaf discs with a diameter of 2 cm using a hole punch, the cut circular leaf discs were immersed in NBT staining solution, and the leaf discs were taken out after being kept away from light for 48 h. The leaf discs were immersed in 95% alcohol, and the discs were decolored with stirring while heating, and the blue areas on the leaf discs after decolorization were the areas where the ROS were accumulated. ROS, GSH, and AsA content in transgenic tobacco and in WT were examined by ELISA enzyme-linked immunoassay. Absorbance values of ROS should be measured at 450 nm; absorbance values of GSH should be measured at 412 nm; absorbance values of AsA should be measured at 265 nm.

Primary antibodies specific to the target protein were added and incubated, followed by incubation with secondary antibodies conjugated with fluorescent dyes to label the target protein, the nuclei were stained with DAPI. The coding sequences of CrGST24, CrGSHB, and CrDHAR2 were fused with the green fluorescent protein (GFP) to construct the corresponding fusion protein expression vectors. These vectors were then transformed into tobacco leaf cells to achieve transient expression, enabling the synthesis of the fusion proteins within the plant cells. Finally, confocal laser scanning microscopy was used to observe the distribution of fluorescent signals in the tobacco leaf cells, so as to determine the subcellular localization of the target proteins.

The full-length CDS of the CrDHAR2 (Accseeion No. PV920956 at the NCBI BioProject database) and CrGSHB (PV920957) were amplified and recombined into the pGADT7 vector to construct the prey vector. The CrGST24 (PV920955) was intercepted and inserted into the pGBKT7 vector to construct the bait vectors. The prey and bait vectors were co-transferred into the Y2H Gold yeast strain as described in the instructions (Clontech, Shanghai, China). Yeast strains carrying the indicated vectors were plated on SD/-Leu-Trp growth medium. The positive clones were diluted with 0.9% (w/v) NaCl to OD₆₀₀ of 0.2, after which 10 μL of each suspension was dropped on SD/-Leu-Trp, SD/-Leu-Trp-His-Ade with X-α-Gal medium. The experiments were performed three times with similar results.

The target gene’s DNA sequence was cloned into a vector containing the luciferase (LUC) reporter gene to form a recombinant plasmid. This plasmid was then transformed into tobacco leaf cells via Agrobacterium-mediated transformation and transiently expressed there. The luciferase detection system, with the substrate added to induce luminescence, was used, and the imaging system recorded the luminescent signals.

The C.reticulata ‘Zipao’ and wild species callus were placed in YS1, YS2, YS3 and YS4 callus medium, and they were placed in white light, dark, red and blue light environments cultured for 30 d.

The callus of C.reticulata ‘Zipao’ and wild species were inoculated onto YS3 medium and placed under red, blue, red-blue ratio of 2:1 and blue-red light ratio of 2:1 for one month.

The C.reticulata ‘Zipao’ and wild species callus were put into AsA1, GSH1 and A1G1 liquid medium, and cultured at 28°C and 220 R for 48 h, then inoculated into YS3 medium, and cultured for 48 h with dark place, and then put into the light environment of blue-red light ratio of 2:1. Callus inoculated into YS3 medium and cultured in blue-red light ratio of 2:1 environment without any treatment were used as control.

The paraffin section method was used to observe the internal cellular structure of the callus of ‘Zipao’ and wild species C. reticulata after treatment with different light intensities, light qualities, GSH, and AsA. Making paraffin sections mainly includes: sampling, fixation, dehydration, clearing, wax infiltration, embedding, sectioning, staining (using safranin - fast green staining), dewaxing, and sealing. Microscopic observation was performed using an optical microscope (Leica DM2500), and the magnification used was 200×.

The GST24 was cloned from callus of Camellia reticulata. Multiple sequence alignment and phylogenetic analysis revealed high similarity between this gene and GST gene in Theaceae, leading to its designation as CrGST24. Based on protein homology and gene structure characteristics, the GST gene is divided into seven subfamilies: TAU, Zeta, Lambda, DHAR, Tchqd, Theta, and Phi. The CrGST24 belongs to the TAU, full length is 693 bp and contains an open reading frame encoding 213 amino acids ( Figure 1a ). In order to predict which proteins CrGST24 can interact with, we used the STRING database for analysis, which revealed that CrGST24 may have potential interactions with GSHB and DHAR2 ( Figure 1b ).

The callus of wild species and ‘Zipao’ of C. reticulata were placed in medium containing different concentrations of IBA and 6-BA, and treated for 2 h and a week. The expression of CrGST24 gene was significantly higher than that of the control after 2 h of treatment in medium containing different concentrations of IBA and 6-BA. The expression of CrGST24 gene in the callus of wild species was 1.55, 1.87, and 1.38 folds higher than that of the control; the expression of CrGST24 in the callus of ‘Zipao’ was 1.33, 2.03, and 1.23 folds higher than that of the control ( Figure 2a ). The expression of CrGST24 gene was significantly higher than that of the control after a week of treatment. The expression of CrGST24 gene in the callus of wild species was 1.30, 1.79 and, 1.33 folds higher than that of the control; the expression of CrGST24 gene in the callus of ‘Zipao’ was 1.40, 2.27, and 1.57 folds higher than that of the control ( Figure 2b ). The CrGST24 gene was most highly expressed in the medium of 0.1 mg/L IBA and 1.0 mg/L 6-BA.

The callus of wild species and ‘Zipao’ of C. reticulata were placed in YS1 mediun and cultured in dark and white light conditions for 8 h. The dark condition as a control. The expression of CrGST24 gene was significantly higher than that of control after 8h of treatment under different light conditions. The expression of the CrGST24 gene in the wild species was 1.53 folds higher than that of the control; the expression of the CrGST24 gene in the ‘Zipao’ was 1.76 folds higher than that of the control ( Figure 2c ). The CrGST24 gene responds to light induction.

The callus of wild species and ‘Zipao’ of C. reticulata were placed in YS1 medium and cultured in white, red and blue light conditions for 8 h, and 3 d. The white light as a control. The expression of CrGST24 gene was significantly higher than that of the control after 8 h of treatment in red, and blue light conditions. The expression of CrGST24 gene in the callus of wild species was 1.31 and 1.75 folds higher than that of the control; the expression of CrGST24 gene in the callus of ‘Zi Pao’ was 1.28 and 1.63 folds higher than that of the control ( Figure 2d ). The expression of CrGST24 gene was significantly higher than that of the control after 3 d of treatment. The expression of CrGST24 gene in the callus of wild species was 2.32 and 4.55 folds higher than that of the control; the expression of CrGST24 gene in the callus of ‘Zipao’ was 1.87 and 3.85 folds higher than that of the control ( Figure 2e ). The expression of the CrGST24 gene increased with treatment time, and the CrGST24 gene was responsive to both red and blue light induction.

After Agrobacterium-mediated heterologous genetic transformation of tobacco, forty T1 tobacco lines were obtained. Seeds were collected and sown to generate forty T2 lines. Using DNA from the T2 transgenic tobacco as a template, primers targeting the hygromycin transferase gene were designed for PCR, screening out ten positive lines. Through qRT-PCR, three high-expression lines were identified: CrGST24-OE6, CrGST24-OE7, and CrGST24-OE22, with expression levels 1402.37, 1292.74, and 971.02 folds higher than that in WT ( Figures 3a-h ).

The length and density of stem trichomes in transgenic tobacco were superior to those in WT ( Figures 3i-l ). Electron microscopy of freehand sections revealed that WT had approximately 35-40 trichomes per field of view (20x), with lengths ranging from 1.02 to 1.51 mm and an average length of 1.32 mm. In contrast, transgenic tobacco exhibited 75-80 trichomes, with lengths ranging from 2.11 to 3.64 mm and an average length of about 3.13 mm. This resulted in trichome lengths and densities that were 2.37 folds higher than those of the WT ( Figures 3m-p ).

The leaves of WT and transgenic tobacco were cut into uniform sizes, and then placed in normal MS medium to observe the growth. The WT leaves started to differentiate callus on the 10th day and as the treatment time increased, more callus were differentiated from the WT and by the 20th day, four callus were differentiated from the WT leaves. Leaves of CrGST24 transgenic tobacco showed signs of differentiation of adventitious roots and callus at the 5th day, with adventitious roots of about 0.1 mm in length; on the 10th day adventitious roots continued to grow and were about 0.2-0.5 mm in length; The length of the adventitious roots was about 0.4-0.7 mm on the 15th day; the length of the adventitious roots was about 0.5-0.8 mm on the 20th day. On the 35th day, both WT and transgenic tobacco plants produced adventitious buds. Compared with the WT, the adventitious buds produced by the transgenic tobacco were more obvious. On the 50th day, the adventitious buds of WT gradually grew larger, while those of transgenic tobacco had grown into independent plants ( Figure 4a ).

Based on the leaf regeneration of transgenic tobacco in vitro, we found auxin and cytokinin-related genes in tobacco at the NCBI online website, and analysed the expression of these genes in transgenic tobacco by q-PCR. The expression of auxin-related genes in transgenic tobacco were all significantly increased and significantly higher than the WT. The expression of NtYUCCA8 in the three transgenic tobacco lines were 3.98, 6.07, and 3.54 folds higher than the WT ( Figure 4b ); NtPIN1c were 6.11, 3.87, and 7.97 folds higher than the WT ( Figure 4c ); NtGH3.3 were 2.32, 4.03, and 3.12 folds higher than the WT ( Figure 4d ); NtIAA27 were 3.68, 4.97, and 4.33 folds higher than the WT ( Figure 4e ); NtARF6 were 4.18, 2.89, and 1.93 folds higher than the WT ( Figure 4f ); NtWOX5 were 6.36, 4.02, and 3.47 folds higher than the WT ( Figure 4g ); NtSAUR20 were 5.87, 7.92, and 5.71 folds higher than the WT ( Figure 4h ).

The expression of cytokinin-related genes were significantly increased in transgenic tobacco and was significantly higher than the WT. NtARR5 were 3.24, 4.27, and 2.18 folds higher than the WT ( Figure 4i ); NtARR12 were 6.22, 4.12, and 5.28 folds higher than the WT ( Figure 4j ); NtCRE1 were 2.48, 3.67, and 5.69 folds higher than the WT ( Figure 4k ); NtCRF6 were 3.76, 5.71, and 3.98 folds higher than the WT ( Figure 4l ); NtAHP6 were 4.07, 7.39, and 5.24 folds higher than the WT ( Figure 4m ); NtWUS were 4.32, 6.26, and 3.77 folds higher than the WT ( Figure 4n ). The CrGST24 gene promoted the expression of both auxin and cytokinin-related genes.

A total of 12 auxin and cytokinin-related genes were screened in the transcriptome data of C. reticulata callus. There are seven auxin-related genes, including: CrYUCCA, CrPIN, CrGH3, CrWOX, CrARF, CrSAUR and CrIAA. The most significant differences in the expression of CrYUCCA2, CrPIN3, CrGH3.9, CrWOX13, CrARF4, CrSAUR20 and CrIAA9 were found in the callus of the wild species, which were higher than ‘Zipao’ by 1.36, 2.41, 4.17, 8.25, 11.67, 17.24 and 7.73 folds; the most significant differences in the expression of CrYUCCA10, CrPIN1, CrGH3.1, CrWOX5, CrSAUR15 and CrIAA17 were found in the callus of ‘Zipao’, which were 4.88, 2.05, 10.12, 8.85, 52.51, and 1.43 folds higher than that of the wild species ( Figures 5a-g ).

There were five cytokinin-related genes, including CrARR, CrCRE, CrCRF, CrAHP, and CrWUS. The most significant differences in the expression of CrARR1, CrCRE1, CrCRF4, CrAHP1, and CrWUS14 were found in the callus of the wild species, which were 188.95, 1.25, 205.32, 237.02, and 3.11 folds higher than that of ‘Zipao’. The most significant differences in the expression of CrARR4, CrCRE14, CrCRF6, CrAHP2, and CrWUS9 were found in ‘Zipao’ callus, which were 201.14, 2.51, 3.58, 5.67, and 2.48 folds higher than that of the wild species ( Figures 5h-l ).

When the auxin and cytokinin contents within transgenic tobacco was examined, the auxin contents in transgenic tobacco were significantly higher than that in WT, which were 1.55, 2.09, and 1.47 folds higher than that in WT ( Figure 6a ); cytokinin contents in transgenic tobacco were significantly higher than that in WT, which were 1.13, 2.16, and 1.54 folds higher than that in WT ( Figure 6b ). The contents of auxin were significantly higher than that of cytokinin in transgenic tobacco, and the ratios of auxin and cytokinin contents in WT and transgenic tobacco were analysed, and the ratios of auxin and cytokinin contents in CrGST24 transgenic tobacco were significantly higher than those in WT, which were 1.24, 1.13, and 1.15 folds higher than those in WT ( Figure 6c ).

NBT staining followed by decolourisation of WT and transgenic tobacco revealed that the area of ROS was significantly larger in WT than in transgenic tobacco ( Figures 6d-g ). The detection of ROS contents in both WT and transgenic tobacco revealed that the ROS contents of transgenic tobacco were significantly lower than that of WT, 1.15, 1.07 and 1.31 folds lower than that of WT ( Figure 6h ); the GSH contents of transgenic tobacco were all significantly higher than those of WT, 1.27, 1.42 and 1.35 folds higher than those of WT ( Figure 6i ); the AsA content of transgenic tobacco were all significant higher than those of WT, 6.09, 4.47, and 2.92 folds higher than those of WT ( Figure 6j ).

The subcellular localization results showed that CrGST24, CrGSHB, and CrDHAR2 are all located in the nucleus ( Figure 7a ). We screened key genes involved in AsA and GSH synthesis by yeast two-hybrid assay, and showed that the CrGST24 gene could interact with CrGSHB and CrDHAR2 ( Figures 7b, c ). The LUC experiment further confirmed that CrGST24 interacted with CrGSHB and CrDHAR2 ( Figures 7d, e ).

The callus of ‘Zipao’ and wild species were placed on YS1, YS2, YS3 and YS4 callus medium containing different concentrations of IBA and 6-BA, and they were incubated with white light, darkness, red light, or blue light for 30 d to observe the growth of the callus. In comparison with the white loose tissue structure before treatment and other different treatments, ‘Zipao’ callus in YS3 medium and treated with red light or blue light respectively had obvious changes in colour from white to green, and the tissue structure changed from loose to denser, and the newly differentiated callus also showed obvious green colour ( Figure 8a ). The wild species grew better in YS3 medium and under red or blue light treatments, with denser tissue structure and subsequent newly differentiated healing tissues that were also distinctly green in colour ( Figure 8b ). The above results indicated that YS3 medium, as well as red or blue light were good for the C. reticulata callus growth.

Callus of ‘Zipao’ and wild species were inoculated on YS3 medium, or under four different light quality treatments, including red, blue, red-blue ratio of 2:1, and blue-red ratio of 2:1 for one month with white light as the control. Under the four different light quality treatments, the callus of ‘Zipao’ changed significantly from the original white loose tissue to the green firm healing tissue, and the growth condition was better under blue-red ratio of 2:1. The paraffin sections showed that the cells of callus were compact under the red or blue light treatment compared to the white light; the cells were tightly packed under red-blue ratio of 2:1 or blue-red ratio of 2:1, and compared with red-blue ratio of 2:1, the cells were much more tightly under blue-red ratio of 2:1 ( Figure 8c ).

There was no significant difference in the morphological characteristics of wild species callus under red or blue light compared with white light, but the morphological characteristics changed significantly after red-blue ratio of 2:1 and blue-red ratio of 2:1, with the colour changing from dark green to bright green. The paraffin sections showed that the degree of cell tightness under red or blue light was not significantly different from that after white light treatment; cells treated under red-blue ratio of 2:1 or blue-red ratio of 2:1 were tight, and the tightness of the cells was made higher under the blue-red ratio of 2:1 treatment ( Figure 8d ).

The wild species callus treated with GSH1 changed from light green to green and became tighter, and the cells were more tight according to the paraffin sections; the wild species callus treated with AsA1 did not change significantly; the best growth of wild species callus were observed after A1G1 treatment, with bright green and tight cells ( Figure 8e ).

The ‘Zipao’ callus treated with GSH1 changed from white to green and became tighter, and the cells of ‘Zipao’ callus treated with GSH1 were more tight according to the paraffin sections; there were no significant changes in the external morphology and internal structure of ‘Zipao’ callus treated with AsA1 compared to those before treatment; ‘Zipao’ callus treated with A1G1 showed the best growth, in terms of external colour the ‘Zipao’ callus treated with A1G1 were bright green in colour compared to the pre-treatment and GSH1 treatment, and in terms of cellular structure the cultivar healing cells were highly aggregated in the A1G1 treatment ( Figure 8f ).

In this study, we found that CrGST24 could promote the synthesis of endogenous auxin and cytokinin. It also caused changes in the expression of auxin and cytokinin - related transcription factors such as YUCCA, ARF, GH3, WOX, WUS, CRF, AHP, CRE, and ARR. Overexpression of the YUCCA could significantly increase the auxin content in plants, thereby affecting their growth and development (Cao et al., 2019). In Arabidopsis thaliana, knocking out the YUCCA resulted in diminished adventitious bud formation capacity, whereas overexpression of the YUCCA enhanced the ability of adventitious bud differentiation (Miguel et al., 2020). In rice, overexpression of OsWOX3A led to significantly upregulated expression of PIN1. The resulting change in auxin transport direction caused by PIN1 led to local concentration gradients. This process activated the expression of YUCCA genes through feedback regulation, thereby promoting the synthesis of endogenous auxin (Sung and Paek, 2016). The WOX5 gene interacted with auxin and cytokinin signaling pathways. Specifically, by regulating the ARR genes, it affected the sensitivity to cytokinins, thereby promoting the differentiation of adventitious buds from callus (Zhai and Xu, 2021). In plant biology, the ARF transcription factor activated GH3 and SAUR genes, thereby promoted the formation of IAA conjugates and increased the levels of endogenous auxin (Dong et al., 2024). The cytokinin signal was transmitted through the AHK-AHP-ARR cascade (To and Kieber, 2008). Once activated, specific types of ARR directly bound to the promoter of CRF and upregulated their expression. Meanwhile, AHP interacted with CRF, promoting their nuclear accumulation. This formed a coordinated regulatory module between CRF and ARR, which together activated downstream cytokinin - response genes, thereby promoted cytokinin synthesis (Cutcliffe et al., 2011). Previous studies have found that the transcription factors YUCCA, PIN, WOX, SAUR, ARF, ARR, AHP, and CRF have extremely important functions in promoting the synthesis of auxin and cytokinin. In the transcriptome data of C. reticulata callus, the auxin and cytokinin-related transcription factors were also found. Combined with the results of this study, it is hypothesized that these transcription factors may bind to the promoter of the CrGST24 gene and activate its expression. Along with exogenous IBA and 6-BA, as well as red - blue light induction, this process promotes the synthesis of endogenous auxin and cytokinin.

GSH and AsA have extremely important roles in plant growth, especially in the scavenging of ROS (Hasanuzzaman et al., 2020). In this study we found that the CrGST24 gene promoted the synthesis of GSH and AsA in transgenic tobacco, and that the content of glutathione and ascorbic acid in CrGST24 transgenic tobacco was significantly lower than that in WT tobacco. Better growth of C. reticulata callus treated with GSH and AsA. GSH has an extremely important role in the differentiation of adventitious buds (Elaziem and Ahmed, 2020). In the induction of adventitious buds of Phoenix dactylifera, the addition of 30 mg/L GSH to the proliferation medium significantly promoted the differentiation of adventitious buds. After four generations of subculture, the number of adventitious buds significantly increased in the medium containing 30 mg/L GSH and 0.5 mg/L TDZ (Elaziem and Ahmed, 2020). GSH enhanced plant adventitious shoot differentiation by regulated intracellular iron homeostasis, participated in signal transduction pathways and improved the redox environment of the medium (Ranjana et al., 2021). Adding a certain amount of GSH to the culture medium reduced the secretion of phenolic substances, prevented browning of plant materials, and improved the induction efficiency of adventitious buds in bananas (Skogerboe and Jenderek, 2022). Glutathione is an important antioxidant in plant cells, and the dynamic balance between its reduced (GSH) and oxidised (GSSG) states (GSH/GSSG) directly affects the redox state of cells. During adventitious bud differentiation, accumulation of reactive oxygen species (ROS) inhibited cell dedifferentiation or redifferentiation. Glutathione protected DNA, proteins and membrane systems from oxidative damage by scavenging ROS, thus provided a stable microenvironment for callus formation and bud primordial differentiation (Müller et al., 2020). AsA plays an important role in cell division and growth, and AsA has an extremely important role in the growth of plant roots and adventitious shoots (Akram et al., 2017). When ascorbate peroxidase (APX1) was knocked out in oilseed rape, the differentiation rate of adventitious shoots was drastically reduced (Liang et al., 2024). Addition of AsA to the culture medium significantly increased the efficiency of inducing adventitious shoots in sorghum (Baskaran and Jayabalan, 2005). After blue light treatment of wheat seedlings, the contents of POD, SOD, GSH and AsA in wheat leaves significantly increased, while the ROS content markedly decreased. This indicated that blue light treatment could enhance the antioxidant capacity of wheat seedlings (Li et al., 2022). GST genes were closely related to GSH and AsA synthesis. In this study we found that CrGST24 promoted GSH and AsA synthesis in transgenic tobacco. The callus treated with GSH and AsA grew better. Excessive ROS could severely inhibit dedifferentiation and redifferentiation of plant callus. We hypothesised that when the GSH and AsA contents were elevated in C. reticulata callus then excessive ROS would be scavenged; exogenous addition of GSH and AsA, on the other hand, inhibits the secretion of phenolics and keeps the browning of healing tissues in culture under control. When the GSH and AsA contents in the healing tissue increased significantly, and at the same time GSH and AsA were added exogenously, a double safeguard was formed, which promoted the differentiation of adventitious shoots in the C. reticulata callus. The synthesis of GSH and AsA is a complex mechanism that requires further study.

GST gene enhances plant resistance to biotic and abiotic stresses by scavenging excess ROS. In this study, overexpression of the CrGST24 gene promoted the synthesis of GSH and AsA in transgenic tobacco, and the ROS content in transgenic tobacco was significantly lower than that in WT tobacco. We found that the CrGST24 gene could interact with CrGSHB, a key gene for GSH synthesis, and CrDHAR2, a key gene for AsA synthesis, by Y2H assay. GSHB genes have an important role in plant resistance to adversity stress (Anjuman et al., 2025). Numerous studies have shown that DHAR and GSHB genes have extremely important roles in plant resistance to biotic and abiotic stresses, and in scavenging ROS. DHAR has an important role in the synthesis of AsA, DHAR maintains intracellular AsA levels in the reduced state by reducing dehydroascorbic acid (DHA) to AsA. DHAR protected cells from oxidative damage by promoted AsA synthesis and scavenged intracellular ROS (Petrova and Smith, 2015). When the DHAR gene was knocked out, the content of AsA was significantly reduced and the antioxidant capacity of the plant was reduced (Terai et al., 2020). In rice, overexpression of the DHAR gene increased rice resistance to salt stress by reducing H₂O₂ levels (Kim et al., 2017). GSHB is a glutathione synthetase that adds glycine to γ-glutamylcysteine to eventually produce GSH. In Brassica campestris, overexpression of the GSHB gene significantly promoted GSH synthesis, which improved antioxidant capacity and stress tolerance in Brassica campestris (Yan et al., 2013). GSHB combines γ-glutamylcysteine (γ-Glu-Cys), which is produced by catalyzing glutamate, cysteine, and γ-glutamyltransferase (GshA), with glycine to ultimately form glutathione (GSH). This process is an important mechanism for cellular antioxidation and maintenance of redox homeostasis (Copley and Dhillon, 2002). In cyanobacteria, GSH cannot be synthesized when the GSHB gene was knocked out (Okumura et al., 1997). Overexpression of the GSHB gene in Arabidopsis promoted GSH synthesis and also improved tolerance to Cd stress in Arabidopsis (Kumar and Chattopadhyay, 2018). In AsA-GSH cycle, DHAR used reduced glutathione (GSH) as a hydrogen donor to catalyzed DHA production of AsA. Under oxidative stress, the AsA synthesis pathway is limited, and DHAR compensated for the synthesis deficit by regenerated AsA. For example, under drought conditions, elevated DHAR activity promoted AsA synthesis (Hasanuzzaman et al., 2020). In transgenic tobacco, overexpression of DHAR gene increased AsA content by 2-4 folds (Wang et al., 2010). Based on the previous studies and the results of this research, we speculated that CrGST24 could utilize GSH as a cofactor to convert harmful ROS into harmless substances. CrGSHB enhanced the efficiency of the entire antioxidant system by synthesizing more GSH to provide sufficient substrates for CrGST24. When the CrGST24 gene was overexpressed, the expression of the CrGSHB gene may be promoted, thereby increased the synthesis of GSHB enzyme. More GSHB enzymes could catalyze more γ-Glu-Cys and Gly to synthesize GSH, thereby led to an increase in GSH content. When the CrGST24 gene was overexpressed, it may promote the synthesis of GSH or reduce the consumption of GSH through some mechanism, thereby providing more GSH as a substrate for CrDHAR2. Sufficient supply of GSH is conducive to CrDHAR2 catalyzing the reduction reaction of DHA more effectively, thereby increasing the content of AsA. The interaction between CrGST24 gene and CrDHAR2 gene may activate the entire AsA-GSH cycle system. Besides CrDHAR2, the activities of other related enzymes in the cycle (such as APX, GR, etc.) may also be coordinately regulated, thereby promoting the synthesis and regeneration of AsA. When the CrGST24 gene interacted with the CrGSHB and CrDHAR2 genes, it activated the entire AsA-GSH cycle system, regulated the expression of some key genes and thereby promoted the synthesis of GSH and AsA.