Establishment of an efficient Agrobacterium‐mediated transformation system for chilli pepper and its application in genome editing

Capsicum annuum (chilli pepper) is one of the most important vegetables worldwide. However, recalcitrance to Agrobacterium-mediated transformation has become a bottleneck in the functional characterization of genes and the genetic improvement of pepper. Recently, significant progress has been made in the genome editing of pepper. Zhao et al. (2024) utilized a tomato spotted wilt virus (TSWV)-mediated CRISPR/Cas nuclease transient delivery system to establish an efficient, transgene-free gene-editing technology in chilli pepper. Despite this advancement, the stable overexpression of transgenes in chilli pepper remains a significant challenge. To establish a stable transformation system in chilli pepper, we initially attempted the well-established method we previously reported for tomato, which employs a combination of zeatin and auxin as plant hormones, coupled with a two-stage regeneration process characterized by a significantly reduced level of cytokinin in the later stage. To facilitate regeneration and elongation of adventitious shoots, an ethylene inhibitor (silver nitrate) and gibberellin (GA3) were supplemented (Gammoudi et al., 2018). Initially, we tested a dwarf variety, MiniPep (Shi et al., 2022), which was used to construct a mutant library via ethyl methanesulfonate (EMS) chemical mutagenesis, following our tomato transformation procedure. However, we found that only rare positive transgenic plants could be obtained (a few transformants among hundreds of cotyledon explants), which is insufficient for practical application. Since genotype is the most critical factor affecting Agrobacterium susceptibility and plant regeneration (Li et al., 2019), we screened for better genotypes. We used the efficiency of virus-induced gene silencing (VIGS) as a selection criterion for Agrobacterium susceptibility (Zhou et al., 2021) and assessed the ability of plants to produce lateral branches as an indicator of regeneration capacity, leading to the identification of a more transformable genotype, PC69 (see more details in Appendix S1). In the transformation of PC69, the optimal concentration of kanamycin for selecting resistant shoots was identified as 75 mg L−1. To facilitate the visual observation of transformation events, we compared the effectiveness of the DsRed, green fluorescent protein (GFP) and RUBY reporter genes (He et al., 2020) (Figure 1a, Table S1). We found that no obvious fluorescence was observed in the transformed callus tissue with DsRed. In contrast, green fluorescence was detectable in the transformed callus tissue, but not in the regenerated shoots, suggesting that the transformed cells either failed to regenerate or that GFP expression was challenging to observe in the regenerated shoots (Figure 1b). However, potential GFP signals were observed in the roots of the transformed plants (Figure 1b), using a portable GFP detector. Notably, RUBY, as a visible reporter, could be seen in the transformed callus tissue, as well as in the young shoots, leaves, roots, flowers and fruits of the regenerated plants (Figure 1c,e). This indicates that RUBY is a suitable reporter gene for pepper transformation. As vacuum treatment and avoidance of pre-culture have been shown to enhance transformation efficiency, we further investigated the effects of these two factors (Figure 1d, Table S2) using RUBY as a phenotypic indicator. Our findings indicate that the combination of vacuum treatment and non-pre-culture significantly increases the transformation efficiency of cotyledon explants, measured as the number of explants exhibiting callus formation with the RUBY phenotype relative to the total number of explants. However, this improvement was not observed in hypocotyl explants. In addition, we examined the inheritance of RUBY in the offspring (T1 generation). All T0 plants produced offspring with the RUBY phenotype, and some lines exhibited phenotypic segregation fitting a 3:1 ratio (Figure 1e, Appendix S1), consistent with Mendel's law of segregation. In summary, through the optimization of various factors, we have developed an effective transformation system for pepper, as briefly described below (Figure 1f, see more details in Appendix S1). Twelve-day-old seedlings were used to prepare explants of cotyledon and hypocotyl segments. The explants were directly inoculated with Agrobacterium at an OD600 of 0.6 under −0.6 Mpa air pressure, followed by a two-day co-culture. The explants were then cultured in a callus-inducing medium (CIM) comprising 4.4 g L−1 Murashige & Skoog (MS) medium, 7.4 g L−1 agar, 30 g L−1 sucrose, 2 mg L−1 zeatin riboside (ZR), 0.1 mg L−1 indole-3-acetic acid (IAA), 360 mg L−1 timentin, 75 mg L−1 kanamycin sulfate (Kan) and 4 mg L−1 AgNO3. Upon the appearance of green bud primordia, the explants were transferred to a shoot-inducing medium (SIM) similar to CIM, except that the ZR was decreased to 0.5 mg L−1, IAA was replaced with 0.17 mg L−1 gibberellic acid (GA3) and 100 mg L−1 activated carbon was included. Elongated shoots were excised and cultured in a root-inducing medium (RIM) comprising 4.4 g L−1 MS medium with agar, sucrose, timentin and 2 mg L−1 Indole-3 butyric acid (IBA). The effective transformation efficiency (number of explant with RUBY phenotype shoots /total explant) in our system is approximately 5% (Table S3), making it ready for practical use. Interestingly, when we applied this transformation system to a gene-editing vector (Figure 1a), we observed albino leaves and regenerated shoots at a low frequency (Figure 1g). Sequence amplification and sequencing of the albino regenerated shoots revealed the presence of edits at the target sites, indicating that this transformation system can also be used for pepper genome editing. Recent studies have highlighted the importance of developmental regulators in post-transformation regeneration efficiency (Lian et al., 2022). We tested the effect of a tomato gene encoding the growth-regulating factor (GRF) interacting factor (GIF), SlGIF1, which has been shown to improve tomato transformation in our test (Table S4). We found that overexpression of SlGIF1 further improved the transformation efficiency of pepper (Table S5), indicating that SlGIF1 may be useful for promoting genome editing in pepper. With the ongoing discovery of developmental regulators, such as peptides (Yang et al., 2024), we believe that the pepper transformation system we have established will undergo further optimization, and this system can also be utilized to explore novel genome-editing strategies (Liu et al., 2024). This research was supported by the National Natural Science Foundation (U21A20230), the National Key R&D Program (2022YFE0100900) and the High-Quality Development Project for the Seed Industry in Hubei Province (HBZY2023B004), China. Y.T. and X.S. designed the experiments; B.O., Y.T. and X.S. wrote the manuscript; Y.T., X.S., Y.S., Y.Z. and X.D., performed the experiments; Y.L. and F. L. provided valuable suggestions and revised the manuscript. Figure S1–S3. Table S1–S5. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.