Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Apr 24;122(2):e70155. doi: 10.1111/tpj.70155

In planta genome editing in citrus facilitated by co‐expression of CRISPR/Cas and developmental regulators

Gilor Kelly 1, Elena Plesser 1, Eyal Bdolach 1, Maria Arroyave 1, Eduard Belausov 1, Adi Doron‐Faigenboim 1, Ada Rozen 1, Hanita Zemach 1, Yair Yehoshua Zach 2, Livnat Goldenberg 1, Tal Arad 1, Yossi Yaniv 1, Nir Sade 2, Amir Sherman 1, Yoram Eyal 1,✉, Nir Carmi 1,✉
PMCID: PMC12022391  PMID: 40275470

SUMMARY

Recent advances in the field of genome editing offer a promising avenue for targeted trait improvements in fruit trees. However, the predominant method taken for genome editing in citrus (and other fruit trees) involves the time‐consuming tissue culture approach, thereby prolonging the overall citrus breeding process and subjecting it to the drawbacks associated with somaclonal variation. In this study, we introduce an in planta approach for genome editing in soil‐grown citrus plants via direct transformation of young seedlings. Our editing system, abbreviated here as IPGEC (in planta genome editing in citrus), is designed to transiently co‐express three key gene groups in citrus tissue via Agrobacterium tumefaciens: (i) a genome‐editing catalytic group, (ii) a shoot induction and regeneration group, and (iii) a T‐DNA enhanced delivery group. This integrated system significantly improves de novo shoot induction and regeneration efficiency of edited tissue. By incorporating single‐guides RNA's (sgRNA's) targeting the carotenoid biosynthetic gene PHYTOENE DESATURASE (CsPDS), the IPGEC system effectively produced mutated albino shoots, confirming its ability to generate homozygous/biallelic genome‐edited plants. By using high throughput screening, we provide evidence that transgene‐free genome‐edited plants could be obtained following the IPGEC approach. Our findings further suggest that the efficiency of specific developmental regulators in inducing transformation and regeneration rates may be cultivar‐specific and therefore needs to be optimized per cultivar. Finally, targeted breeding for specific trait improvements in already successful cultivars is likely to revolutionize fruit tree breeding and will pave the way for accelerating the development of high‐quality citrus cultivars.

Keywords: genome editing, in planta, de novo meristem induction, transformation, citrus, transgenic plants, RUBY, technical advance

Significance Statement

This study presents an in planta genome‐editing system for citrus, which utilizes developmental regulators to enhance regeneration efficiency and production of genome‐edited plants. By streamlining the editing process and providing a cost‐effective alternative to tissue culture, this approach accelerates targeted trait improvements, with significant implications for citrus breeding.

INTRODUCTION

Conventional breeding based on hybridization and selection, or on random identification of bud mutants, has been the predominant methods for developing new elite varieties in fruit trees (Alvarez et al., 2021). Yet, these approaches rely on screening large numbers of trees and inherently have a very low success rate and are therefore lengthy and require expensive resources to cultivate and screen thousands of trees (Alvarez et al., 2021; Song et al., 2019). Consequently, targeted breeding approaches are essential for more efficient and accelerated cultivar production in fruit trees such as citrus. The welcomed introduction of marker‐assisted breeding, which allows screening for/against specific traits at the seedling stage, has been late in coming in fruit trees and the current repertoire of markers associated with specific traits in fruit trees is very limited due to the extensive genetic studies required to develop reliable and reproducible molecular markers. Thus, a major new approach enabling targeted breeding of fruit trees would be of great benefit and application alongside the classical and marker‐assisted approaches currently used.

Clustered Regulatory Interspaced Short Palindromic Repeats (CRISPR)/Cas9‐genome editing technology offers a promising opportunity for targeted trait modification to supplement the conventional breeding approaches (Alvarez et al., 2021; Zhu et al., 2020). Nevertheless, its integration into different fruit tree species, and particularly its development as a non‐GMO breeding approach, remains an ongoing challenge. Currently, the process in plants typically requires regeneration via tissue culture (Dutt et al., 2018; Huang et al., 2022), which is time‐consuming, laborious, expensive, and is subject to the drawbacks associated with somaclonal variation.

To overcome some of the challenges to genome editing in plants, and particularly to circumvent the extensive tissue culture steps, an in planta genome editing strategy was introduced in recent years (Belanger et al., 2024; Hao et al., 2024; Lian et al., 2022; Mei et al., 2024; Rizwan et al., 2021; Zhang, Zhang, et al., 2017). The in planta approach encompasses co‐expression of guide RNAs (genome editing [GE] cassette) and shoot regeneration factors (developmental regulators [DRs] cassette) in transgenic Cas9‐expressing plants to form de novo genome‐edited shoots (Maher et al., 2020). The expression of DRs, such as WUSCHEL (WUS), SHOOT MERISTEMLESS (STM) and PLETHORA (PLT5), was shown to stimulate developmental changes in somatic cells, leading to the acquisition of meristematic identity (Eshed Williams, 2021; Lian et al., 2022; Maher et al., 2020; Nasti et al., 2021; Zhang, Lian, et al., 2017). In tobacco, the combined effects of GE and DRs facilitate de novo edited meristem induction, which can later develop into fully edited shoots (Maher et al., 2020). Yet, this technique was shown to be effective when using transgenic plants pre‐expressing high levels of Cas9 (Maher et al., 2020).

In this study we present in planta genome editing in citrus (IPGEC), an in planta approach designed to enable the development of transgenic and non‐transgenic de novo edited shoots when applied to young citrus seedlings. Our approach involves Agrobacterium delivery‐based co‐expression of a combination of gene cassettes including: (i) Cas and gRNAs, (ii) regeneration and shoot induction factors, and (iii) factors enhancing Agrobacterium T‐DNA delivery. The approach yielded high‐efficiency editing as well as a subset of chimeric shoots containing non‐GMO edited sectors that provide the first indication for genome‐edited transgene‐free shoots via the in planta approach.

RESULTS

Designing a system for IPGEC

In planta genome editing in citrus was designed to promote de novo edited shoots by transiently co‐expressing de novo meristem formation factors (referred to as DRs), together with CRISPR/Cas9 genome editing components in citrus tissue. We selected the PHYTOENE DESATURASE (CsPDS, Orange1.1g009508m.g) gene as an experimental target for genome editing, since impaired PDS conveniently produces a visually distinct albino phenotype in plants (Qin et al., 2007) including citrus (Huang et al., 2020).

The in planta application step was designed based on a transformation method described previously (Rizwan et al., 2021; Zhang, Zhang, et al., 2017). Briefly, genetically identical seedlings [polyembryonic (Wang et al., 2017; Yadav et al., 2023)] of Carrizo Citrange [Citrus sinensis (L.) Osb. × Citrus trifoliata (L.) Raf] were grown in soil to a 5–6 leaf stage, truncated below the first true leaf (above epicotyl, Figure 1A), infected with Agrobacterium containing IPGEC constructs (described below), and incubated in the dark for 10 days (Figure 1C) before transferring back to the light (Figure 1D). Shoot regeneration was observed starting from ~30 days after the treatment (Figure 1E).

Figure 1.

Figure 1

Open in a new tab

Method overview.

The in planta transformation involves (A) cutting the stem of a 4‐month‐old seedlings below the first true leaf (B) and applying a mixture of agrobacteria to the cut site for 1 h using a 10 μl tip filled with Agro solution (C). Then, seedlings are incubated in the dark for 10 days before transferring them back to the light (D). Three days following co‐culturing, parafilm was removed and cotton balls submerged in agar trap solution containing antibiotics were used to immerse the wound sites three times. Wound sites with a remaining agar trap solution were wrapped again with parafilm and were kept in the dark for the remaining 7 days. Shoot emergence was monitored 30–60 days' post‐transformation (E, gray arrow).

The IPGEC constructs were designed to comprise gene cassettes specializing in three essential functions to be co‐expressed: (i) genome editing machinery, (ii) meristem development (and shoot induction) factors, and (iii) Agrobacterium T‐DNA delivery factors. Four vectors were designed (Figure 2A): (a) V1 contains a genome‐editing cassette consisting of three units driving the expression of nine different sgRNAs, a Cas9 gene downstream of a constitutive 2 × 35S promoter, a fluorescent marker (GFP) and the developmental regulators WUSCHEL (WUS, ZmWUS2) and SHOOT MERISTEMLESS (STM, AtSTM) (Figure 2A, V1). The sgRNA's expression system includes the AtU6‐gRNA‐scaffold‐terminator sequence, followed by two tRNA‐mediated multiplex genome editing units, expressed under the endogenous CsU6 promoter (Figure 2A, V1 and V2). (b) V2 is similar to V1 but contains a gene encoding isopentenyl transferase (IPT) instead of the WUS‐STM unit (Figure 2A, V2). (c) V3 includes expression cassettes of five genes (LATERAL SUPPRESSOR [AtLAS, AT1G55580], REGULATOR OF AXILLARY MERISTEMS1 [AtRAX1, AT5G23000], EXCESSIVE BRANCHES1 [AtEXB1, AT1G29860], REGULATOR OF AXILLARY MERISTEM FORMATION [AtROX, AT5G01305], and RESPONSE REGULATOR1 [AtARR1, AT3G16857]), which act upstream of WUS and STM (Figures [Link], [Link] and [Link], [Link], Cao & Jiao, 2020), factors that serve as a key hub in the developmental process of meristem formation (Figure 2A, V3). (d) V4 contains a Cas9 gene downstream of the embryonic and cell division‐specific YAO promoter (Yan et al., 2015; Zhang, LeBlanc, et al., 2017) and expression cassettes of the VirE2‐INTERACTING PROTEIN (VIP1) and the Agrobacterium effector protein VirE2, involved in T‐DNA delivery to the nucleus (Figure 2A, V4). These vectors (Figure 2A, V1–V4) which constitute the core of the IPGEC system, were separately transformed into Agrobacterium and were co‐expressed in different combinations during our experiments. Detailed sequences of the IPGEC editing system are provided in Data S1.

Figure 2.

Figure 2

Open in a new tab

The IPGEC genetic system.

(A) A simplified schematic representation of the four vectors used in this study. Vectors 1 and 2 (V1 and V2) incorporate SpCas9 under a constitutive promoter, a genome editing gRNA's cassettes (include pU6‐sgRNA‐scaffold‐terminator unit followed by two tRNA‐based multiplexed editing units expressed under the endogenous CsU6 promoter), selection markers (antibiotic resistance, GFP) and a cassette containing developmental regulators (V1; WUS and STM, V2; IPT). Vector 3 (V3) includes expression cassettes for enhanced de novo meristem induction factors (LAS, RAX1, EXB1, ROX, and ARR1). Vector 4 (V4) contains SpCas9 under the YAO promoter, VIP1, and VirE 2 genes for improved T‐DNA delivery. Detailed sequences for each vector are provided in Data S1.

(B) Illustration of the CsPDS gene and the positions of the gRNAs. Exon positions are marked with black rectangles.

Co‐expression efficiency

The IPGEC system requires co‐expression of genes directing editing together with a variety of genes encoding DRs, which can either be located on the same construct or on different constructs. Therefore, the efficiency of IPGEC depends on simultaneous nuclear incorporation of T‐DNAs delivered by different Agrobacterium strains with a sufficiently high levels of expression. Hence, Agrobacterium co‐infiltration of Carrizo leaves was performed to monitor the co‐expression levels of two representative vectors (V2 and V3, Figure 3), which were tracked at three time points. The expression levels of the three genes tested in each vector were significantly elevated compared with control (non‐treated) leaves (Figure 3A,B). Peak expression was observed 6 days' post‐infiltration, followed by a reduction on day 10. At the end of the experiment, GFP fluorescence was detected, further validating the efficient transient expression encoded by the V2 vector (Figure 3C).

Figure 3.

Figure 3

Open in a new tab

Gene expression analysis and confocal imaging at three time points following agroinfiltration of plasmids V2 and V3 into Carrizo WT leaves.

(A) Expression analysis of Cas9, GFP, and IPT (V2) at 3, 6, and 10 days following agroinfiltration.

(B) Expression analysis of LAS, EXB1, and ARR1 (V3) at 3, 6, and 10 days following agroinfiltration. (A, B) Data points are means ± SE (n = 6). Different letters indicate significant differences (Tukey's HSD test, P < 0.05).

(C) Confocal microscopy image showing GFP distribution on day 10 after agroinfiltration. Panels are GFP fluorescence (stained green, top left), white‐light (top right), chlorophyll‐autofluorescence (stained magenta, bottom left) and merged (bottom right). Bar = 50 μm.

We further verified that co‐expression of genes from several constructs harbored by different Agrobacterium strains occurs in a single cell. To address this issue, we monitored fluorescence of three distinct fluorescent proteins (GFP, mCherry, and CFP) in tobacco (Nicotiana tabacum) epidermal cells, each delivered by an Agrobacterium harboring a different construct (Figure S1). Our results demonstrate that expression of the three fluorescent proteins encoded by different constructs indeed occurs in the same cell, as verified by a merged image of GFP, mCherry, and CFP (Figure S1). Thus, transient co‐expression in the same cell is obtained following co‐cultivation of three Agrobacterium isolates harboring different constructs.

Regeneration efficiency

Next, we evaluated the de novo regeneration and shoot production efficiency in response to treatment with the IPGEC system (Figure 4). Seedlings cut below the first true leaf were either not treated, infected with control (empty) Agrobacterium (Agro) lacking any DRs, or with Agro harboring IPGEC system constructs (Figure 4A–C). We noted lateral shoot induction originating from the side of epicotyls (below the cut site, green square; Figure 4), as well as shoots regenerated from the top cut site (orange square). We hypothesized that the lateral shoots represent a natural shoot induction process due to release of apical dominance and are not related to the IPGEC system, while those regenerated from the top cut site are more likely to be affected by Agrobacterium‐derived expression of the DRs (Figure 4A–C). Under conditions designed to monitor the effect of the cut (no Agro added), 38.1% of the seedlings regenerated from the top (Figure 4C, “Top” column), with most producing only a single shoot. However, when infected by control (vectorless) Agro, the regeneration efficiency dropped to 5.6% (Figure 4A,C). This suggests that, in general, the presence of Agrobacterium is not beneficial and limits regeneration, perhaps due to induction of the plant–pathogen response at the expense of developmental processes (He et al., 2022). In contrast, treatment with Agrobacterium harboring IPGEC system; V1–V4 constructs significantly enhanced regeneration efficiency, attaining rates as high as 75% (Figure 4B,C, “Top” column). Furthermore, 62.5% of these seedlings displayed an elevated shoot number with at least two shoots regenerated per plant (Figure 4C, right column). Overall, expression of the IPGEC system constructs significantly enhanced regeneration efficiency (Figure 4D), yet, occasionally excessive shoot production and a tumor‐like tissue irregularities were observed (Figure 4E).

Figure 4.

Figure 4

Open in a new tab

Regeneration efficiency of the IPGEC system.

Regeneration efficiency was tested 60 days following exposure to Agrobacterium (Agro). (A, B) Regeneration efficiency of plants treated with Agrobacterium lacking any developmental regulators (A), or agrobacteria containing the IPGEC genetic system (B). Each column represents an independent plant, and filled squares denote the number of shoots per plant, emerging either from the top (orange) or the side (green). An Illustration of seedlings with top shoots (orange squares) and side shoots (green squares) is provided (top right).

(C) Summary of regeneration efficiency. Evaluation of plants either not treated (shoots removed, no Agro), treated with Agrobacterium (without developmental regulators) or with Agrobacterium harboring IPGEC plasmids. The rates of plants with no regeneration, plants with shoots emerging from the side only, or plants with shoots emerging from the top are provided. The right column presents plants containing at least two shoots that have emerged from the top.

(D, E) Representative images of plants with top shoot regeneration. (D) Representative images of regenerated plants displaying none (left)‐ to multiple (right) shoots. (E) Over‐regeneration resulting in excessive shoot production and tumor‐like formation.

Evaluation of genome editing efficiency

Next, we evaluated the IPGEC system's ability to impact genome editing. We used multiple single‐guide RNA's (sgRNA's) targeting the carotenoid biosynthetic gene PHYTOENE DESATURASE (CsPDS) at independent sites spanning several exons (Figure 2B). Expression of IPGEC system constructs resulted in a spectrum of outcomes, with shoots that were transgenic (displaying GFP fluorescence) and green vs. those that were transgenic as well as albino, suggesting editing (Figure 5A,B). We were able to identify shoots that were either entirely or partly (chimeric) transgenic, as indicated by GFP fluorescence (Figure 5A, lines #1 and #12, respectively), and shoots that underwent CsPDS gene editing, displaying an albino phenotype (Figure 5B,D). While some of the edited shoots were chimeric for editing (Figure 5B, #17, #27, #36, #40, Figure 5E) others were fully mutated, showing a complete albino phenotype (Figure 5B, #39; Figure S3). These results confirm the potential of the IPGEC system to generate fully edited shoots (Figure 5E). Genome‐editing efficiency was further verified by sequencing (Figure 5C). For example, samples #36 and #40 displayed albino sectors adjacent to a green sector throughout the developing shoot (Figure 5B). Sequencing of the albino segments revealed a deletion of 1 or 2 bases in exon4 of the CsPDS gene in line #36, and an insertion (+T) in exon3 of line #40 (Figure 5C). Interestingly, in the case of shoot #39, a biallelic mutation was detected in exon13, characterized by deletion and insertion, and more importantly, lacking any traces of wild‐type sequences (Figure 5C). Furthermore, this sample displayed a multiallelic editing in exons 3, 5, and 13 (Figure 5C), suggesting that the IPGEC system can promote multiplex editing.

Figure 5.

Figure 5

Open in a new tab

In planta transformation and genome editing using the IPGEC genetic system.

(A) Representative images of transgenic shoots expressing GFP from two independent transgenic lines #1 and #12. Panels display merged images of GFP (stained green) and chlorophyll autofluorescence (stained magenta). The bar scale is indicated for each image.

(B) Representative images of regenerated shoots displaying a complete or a partial (chimeric) albino phenotype, indicating a CsPDS‐impaired gene.

(C) Representative CsPDS sequences displaying editing events in specific exons of regenerated shoots. The gRNA sequences (bold), PAM site (green), Deletions (pink) and insertions (blue) for exons 3–5 and 13 of the CsPDS are indicated (an illustration of the gRNA's positions in the CsPDS gene is available in Figure 2B). Sample #39 demonstrates multiallelic editing in four different exons.

(D) Representative images of WT and albino plants following tissue lysis.

(E) Summarizing table of three experiments using the IPGEC system.

In addition to the Carrizo experimental system, we also tested efficiency of the IPGEC system on “Duncan” grapefruit (Citrus paradisi Macf.) seedlings. Similar to our findings in Carrizo, we identified transgenic shoots displaying GFP fluorescence throughout the entire leaf (stomata, epidermis, and mesophyll) (Figure 6A). These results provide additional support for the IPGEC system's ability to target the initial (progenitor) cells prior to cell division and shoot emergence, in both, Carrizo and Duncan cultivars. Based on visualization, we have identified mutated (chimeric) shoots in Duncan as well (Figure 6B,C). For example, the leaves of line #38‐6 displayed white and green sectors, divided by the primary mid‐vein. The albino sector of the leaf displayed GFP fluorescence, while the green sector, harboring a functional CsPDS gene, was not transgenic (Figure 6B). In some cases, chimerism was less pronounced, displaying a milky‐yellowish coloration confirmed by confocal imaging to possess a mixture of GFP‐expressing‐ and WT cells throughout (Figure 6C, line #38‐7). These examples are indicative of a partial transgenic and partial edited state, with both edited and non‐edited cells coexisting throughout the leaf.

Figure 6.

Figure 6

Open in a new tab

In planta transformation using the IPGEC system in grapefruit Duncan cultivar.

(A) Representative images of Wild‐type and transgenic shoots expressing GFP.

(B) Transgenic plant displaying a sectorial chimeric morphology divided by the central leaf vein. GFP fluorescence is observed in the white‐albino section of the leaf.

(C) Transgenic plant displaying partial CsPDS mutation with a milky‐yellowish phenotype and a mix of WT and GFP‐expressing cells. (A) Panels are merged images of GFP (stained green) and chlorophyll‐autofluorescence (stained magenta). (B, C) Panels are GFP fluorescence (stained green, top left), white‐light (top right), chlorophyll‐autofluorescence (stained magenta, bottom left) and merged (bottom right). Bar scales are indicated for each image.

IPGEC and the potential for transgene‐free genome editing

The ultimate goal for an editing system in fruit trees is a system that gives rise to edited, yet non‐transgenic shoots. We speculated that the enhanced T‐DNA delivery in the IPGEC system may generate transient expression levels that are sufficient to lead to genome editing, without integration of T‐DNA, resulting in transgene‐free genome‐edited shoots. To address this possibility, a large population of 200 Carrizo shoots was generated following IPGEC treatment, and the resulting shoots were screened for editing mutants using the sensitive high‐resolution melting (HRM) method (Figure 7A). We were able to identify chimeric CsPDS‐impaired shoots, further verified by next‐generation sequencing (NGS). Among the chimeric edited lines, we detected sample #81 that was found to harbor editing‐based deletions in exon13 (−2 bp, −5 bp) in 30% of the sequenced reads and editing‐based deletions/insertions in exon3 (+1 bp, −5 bp) in 45% of the sequences (Figure 7B; Table S4). A PCR reaction and the sensitive TaqMan analyses were used to verify that these shoots were transgene‐free (Figure 7C). Thus, we demonstrate that transgene‐free editing is achievable; however, further optimization is required for obtaining non‐chimeric, non‐transgenic edited shoots.

Figure 7.

Figure 7

Open in a new tab

Identifying transgene‐free, genome‐edited plants.

(A) Genotyping workflow. Following DNA extraction, HRM was used to identify candidates potentially harboring gene editing events. Candidates were then sent for sequencing for mutation verification, followed by PCR and TaqMan analysis for testing transgene presence. Samples #23 and #81 represent plants displaying CsPDS gene editing events that are either transgenic or transgene‐free, respectively.

(B) CsPDS sequences of samples #23 and #81 displaying gene editing events in exon3 and exon13 of regenerated shoots. The gRNA sequences (bold), PAM site (green), deletions (pink) and insertions (blue) are indicated. Read counts for each sample are specified in Table S4.

(C) PCR analysis of samples #23 and #81 for the detection of SpCas9. CsACTIN served as a reference gene. NTC: no template control. Primers are listed in Table S3.

Cultivar‐to‐developmental regulator interaction determines regeneration and transgenic efficiencies

As demonstrated, the use of developmental regulators significantly enhances regeneration rates (Figure 4) and genome‐edited shoot production (Figures 5 and 6; Figure S3). We therefore tested the potential impact of additional developmental factors on shoot emergence efficiency. PLT5 and GRF4‐GIF1 were previously reported to improve other species' transformation and regeneration efficiencies (Debernardi et al., 2020; Lian et al., 2022). Hence, we studied the effect of these factors, alongside IPT and WUS‐STM DRs used in the IPGEC system, on the regeneration and transformation efficiency of four different citrus cultivars (Carrizo, Duncan, Foster, and Hudson). For convenient screening of the positively transformed shoots, we used a cassette expressing the betalain biosynthetic genes (RUBY) that lead to the accumulation of the pink/purple metabolite betalain (He et al., 2020; Kumar et al., 2022; Polturak et al., 2016). Four independent vectors that contain the RUBY cassette together with either IPT, WUS‐STM, GRF4‐GIF1, or PLT5 transcriptional units, driven by different promoters, were assembled (Figure 8A), and expressed in planta using the IPGEC protocol. Regeneration and transformation rates were analyzed 60 days following infection with Agrobacterium (Figure 8B,C). Carrizo generally displayed relatively low regeneration rates compared with the other cultivars in all DRs tested (Figure 8C). IPT was found to be the most efficient DR in Carrizo and Hudson, while in Duncan and Foster, it was the GRF4‐GIF1, with rates above 90% (Figure 8C). In contrast, using the PLT5 in Carrizo and Foster displayed limited efficiency where regeneration rates were relatively low (15.6% and 25%, respectively).

Figure 8.

Figure 8

Open in a new tab

The effect of developmental regulators on the regeneration and transformation efficiencies of different cultivars.

(A) Schematic description of the vectors assembled for this assay. A RUBY unit, together with the NptII transcriptional unit, was assembled together with four different developmental regulators (DRs; IPT, WUS‐STM, GRF4‐GIF1 and PLT5). A detailed description of the sequences is provided in Data S1.

(B) Representative images of betalain‐accumulating plants following transformation. Plants were either chimeric or completely transformed.

(C) Regeneration and transformation efficiency of Carrizo, Duncan, Foster, and Hudson cultivars following transformation with RUBY‐IPT, RUBY‐WUS‐STM, RUBY‐GRF4‐GIF1, and RUBY‐PLT5. Gray columns represent regeneration efficiency, and pink columns (inset) represents transformation efficiency (betalain‐accumulating shoots). The number of biological replicates (total seedlings per treatment) is detailed in Table S5.

Concerning transformation rates, variable results were observed throughout this survey. Generally, the highest number of shoots displaying betalain accumulation was observed when IPT served as a DR. This was the case for Carrizo, Duncan, and Hudson (Figure 8C). In Foster, however, RUBY‐expressing shoots were most prominent when GRF4‐GIF1 was used. Overall, these results demonstrate the capability of the in planta methodology to yield transgenic plants in a variety of species and emphasize the need to match the best performing DR for every species. Thus, a prior optimization step assessing the efficiency of the different DRs can enhance the rate of success of IPGEC towards the end‐game of producing transgene‐free edited plants (Figure 7).

DISCUSSION

Genome editing, mostly based on plant tissue culture methodologies, transgenesis, and segregating‐out of transgenes, has been successfully applied to model plants and annual crop plants, yet is much more challenging for application in improving perennial fruit tree varieties due to the long generation time and since many fruit trees are predominantly heterozygous due to various outcross‐promoting mechanisms (i.e., pollen–pistil self‐incompatibility in rosaceae [apple, pear, almond, apricot, peach, plum, and others], in Coffea [canephora] and in Citrus; synchronous dichogamy in avocado; dioecy in figs; etc.). Thus, segregating‐out of transgenes while maintaining the original qualities of the variety would not be possible due to both limitations in self‐hybridization and due to genetic segregation of any offspring obtained. In this study, we introduce an in planta approach for genome editing, which shows potential to overcome the specific challenges of editing in fruit trees. This comprehensive system involves co‐expression of a genome editing cassette, developmental regulators/meristem induction factors, and an improved Agrobacterium delivery cassette. The simultaneous activation of these factors induces gene editing of meristematic cells, leading to edited shoot outgrowths, which can then be grafted to form fully grown plants displaying the desired trait.

Transformation and genome editing through tissue culture are well‐established methodologies in citrus, with several studies reporting trait improvement via genome editing (Campa et al., 2024; Huang et al., 2022; Jia et al., 2017; Mahmoud et al., 2022; Nerva et al., 2023; Parajuli et al., 2022). However, the tissue culture step is time‐consuming, laborious, and potentially subject to somaclonal variation, thus limiting its application. To overcome the limitations of tissue culture, ongoing efforts are dedicated to the development of efficient in planta transformation techniques in various plant species (Cody et al., 2023; Hamada et al., 2017; Hao et al., 2024; Khadgi et al., 2024; Lian et al., 2022). In some species, the transformation efficiency, a combination of transgene integration and regeneration efficiencies, is relatively high without adding DRs. For passion fruit, for example, transformation efficiency is reported to be 71.1–81.4% (Rizwan et al., 2021). In citrus, however, transformation efficiency is species‐dependent (Zhang, Zhang, et al., 2017). Developmental regulators emerge as a powerful tool to improve transformation rates. When the DRs – WUS, STM, and IPT were – employed, the transformation efficiency went up from none, without DR's, to 6–10% in tobacco (Maher et al., 2020). Expression of the PLT5 developmental regulator enabled transformation in several species (Lian et al., 2022); in snapdragon (Antirrhinum majus), tomato (Solanum lycopersicum), and in two brassica cabbage varieties (Brassica rapa var. Bok choy and Pei‐Tsai), the transformation efficiency increased from none, without PLT5, to 8.8, 13.3, 12, and 12%, respectively, when PLT5 was expressed (Lian et al., 2022). In line with the developmental regulator's positive effect, we obtained relatively high transformation rates when IPT was expressed, displaying the best performance in Carrizo, Duncan, and Hudson cultivars (12.5–30%) and when the GFR4‐GIF1 was employed in Foster (8.5%, Figure 8). Taken together, applying developmental regulators while using the in planta technique can significantly enhance transformation efficiency. Taking a significant step forward would be to attain genome editing in fruit trees via the in planta application, as was recently demonstrated in Cas9 pre‐expressing tobacco, in which the transgene was segregated out (Maher et al., 2020). However, Cas9 pre‐expression is not a possibility for producing transgene‐free edited shoots in fruit trees, and thus obtaining high levels of Cas9 activity that are critical for the editing step is a major challenge. Previous studies provide an indication that transient gene expression can be an efficient tool to alter gene function (Jelly et al., 2014; Tyurin et al., 2020). Our data demonstrate that the transient expression of Cas9 under two different promoters, p2 × 35S (constitutively expressed) and pYAO (active during cell division), delivered by two independent agrobacteria concomitantly, drives sufficient expression for attaining effective gene editing (Figures 3, 5, and 6), including multiallelic editing in some shoots (Figures 5D, sample #39 and 7B, samples #23 and #81). These results demonstrate the potential for targeting different genes simultaneously, thus providing a powerful tool for multiplex genome editing in citrus.

The de novo meristem regeneration/induction is a hallmark of in planta genome editing, and we therefore considered various factors to improve the efficiency of this step. Beyond the key factors, WUS, STM, and IPT known to serve as key regulators of meristem development (Figure 2A, V1–V2), we have studied additional factors shown to induce the expression of WUS and STM (Figure 2A, V3). These factors, hypothesized to further support meristem formation, include LAS, RAX1, EXB1, ROX, and ARR1, which were previously shown to regulate axillary meristem initiation, resulting in shoot overbranching (Cao & Jiao, 2020; Greb et al., 2003; Guo et al., 2015; Keller et al., 2006; Yang et al., 2012, Figure S2). Overall, the IPGEC system promotes a regeneration and shoot emergence level of 75% (Figure 4C). Having said this, in some cases, uncontrolled proliferation was observed that did not develop into shoots (Figure 4E), emphasizing the need to fine‐tune the application of these factors to avoid developmental dead ends.

Strategies to obtain transgene‐free genome‐edited plants continue to evolve. In annual crops, this process is relatively straightforward whereby, following validation of editing, transgenes can be segregated out through self‐pollination or backcrossing. However, this method is impractical for perennial trees such as citrus due to their high heterozygosity, the lengthy juvenile phase, and, in many elite varieties, impaired or aborted sexual reproduction resulting in seedless fruit. Recent innovations have sought to address these challenges. One approach involves protoplast editing by the Cas9‐sgRNA ribonucleoprotein system, followed by regeneration (Lin et al., 2018; Su et al., 2023). Another potentially effective strategy combines the co‐editing of a selectable gene (the acetolactate synthase gene for herbicide resistance) together with the gene of interest (Huang et al., 2023; Jia et al., 2024). Our study provides an indication that transgene‐free editing can potentially be achieved via transient expression of DRs and genome editing components in planta, thus saving both time and effort as well as being devoid of the risks associated with somaclonal variation (Mei et al., 2024). In our current work, we demonstrated the ability to produce chimeric plants that are both transgene‐free and edited. For example, sample #81 shoot exhibits multiallelic editing, with 45% and 30% editing observed in exons 3 and 13, respectively, while genetic screens verified it to be transgene‐free (Figure 7C). We note that shoot chimerism is not a limiting factor in producing fully edited plants, as the conversion of chimeric radiation‐mutagenesis‐derived mutants to fully mutated ones is a common practice in citrus breeding programs achieved by successive bud grafting (Goldenberg et al., 2014; Vardi et al., 2008). Furthermore, high‐throughput screening techniques facilitating much larger editing experiments are likely to yield a variety of edited events, including uniformly edited shoots. Therefore, although the generation of transgene‐free, genome‐edited citrus plants requires additional effort to ensure methodological reproducibility, it is a feasible approach that paves the way to develop commercially viable non‐GMO edited cultivars.

We have shown that the IPGEC system is effective when applied to soil‐grown seedlings of polyembryonic citrus varieties (e.g., Duncan, Hudson and Foster grapefruit and Carrizo citrange), which produce multiple genetically identical seedlings (clones) originating from maternal tissue (Wang et al., 2017; Yadav et al., 2023). However, many citrus varieties are mono‐embryonic, producing hybrid non‐uniform seedlings, which therefore cannot be used for genome editing purposes using the current IPGEC methodology. Further improvements of our methodology to allow the use of mature plants for in planta genome editing could extend the applicability of the IPGEC system to a wider range of citrus varieties and shorten breeding time even further.

We show that when applying DRs using the IPGEC system, genetic transformation rates do not necessarily correspond to regeneration efficiency. For example, in Carrizo, regeneration efficiency was relatively low compared with the high transformation rates obtained when IPT served as the DR (Figure 8C). In contrast, in Duncan grapefruit treated with GRF4‐GIF1, regeneration rates were high, but shoots lacked any transformation events. Therefore, matching the optimal DR for each cultivar is essential to obtain higher transformation/editing rates. However, when aiming for transgene‐free edited plants, future experiments may be directed to explore the benefit of using high regeneration‐inducing DRs that display low transformation rates, as in the case of Duncan‐WUS‐STM, Duncan‐GRF4‐GIF1, Foster‐WUS‐STM, and Hudson‐PLT5 (Figure 8C).

Finally, this study demonstrates the potential for performing in planta genome editing in citrus and for obtaining transgene‐free edited plants based on existing commercial cultivars. While various components of the system still require further optimization and fine‐tuning, the in vivo editing approach provides tools for targeted breeding to the toolkit of citrus breeders as well as for fruit tree breeders in general.

MATERIALS AND METHODS

Plant material and growth conditions

Plants used in this study were Carrizo citrange (Carizzo, Citrus Osb sinensis “Washington” sweet orange × Citrus trifoliata Raf), Duncan, Hudson and Foster grapefruits (Citrus paradisi Macf.) or Nicotiana benthamiana, as indicated in each experiment. Plants were grown in a 16 h light/8 h dark photoperiod at 25°C, 70 μmol m−2 sec−1 light intensity (except where noted), in a soil potting mix containing (w/w) 70% peat, 30% perlite, supplemented with slow‐release fertilizer (Even‐Ari, Israel).

The IPGEC genetic system's building blocks

The IPGEC system is comprised of four vectors. The first two vectors include a genome‐editing cassette and transcriptional unit of Cas9, selection markers, and developmental regulators (Figure 2A; Table S1). The regulators used are either WUSCHEL (ZmWUS2) together with SHOOT MERISTEMLESS (AtSTM, Figure 2A, V1), or a gene encoding isopentenyl transferase (IPT, Figure 2A, V2). The editing cassettes were designed to target the CsPDS gene. Nine guide RNAs (sgRNAs) were designed to target the coding region of eight different exons of CsPDS. These gRNAs were expressed under the control of a short AtU6 promoter (pU6) as well as a double polycistronic‐tRNA‐gRNA system (PTG), activated under the control of the endogenous CsU6 promoter, as described in Huang et al. (2020) (Figure 2A). The third vector is a shoot induction and regeneration cassette, featuring the genes: AtLAS, AtRAX1, AtEXB1, AtROX, and AtARR1 (Cao & Jiao, 2020; Greb et al., 2003; Guo et al., 2015; Keller et al., 2006; Yang et al., 2012; Zhang, Lian, et al., 2017). These genes act upstream of WUS and STM, while EXB1, RAX1, and LAS function specifically in axillary meristem formation (Cao & Jiao, 2020; Figure 2A, V3; Figure S2). The combination of these genes was designed to further enhance de novo meristem acquisition and to promote the overall shoot induction process. The final vector is aimed at improving Cas9 gene delivery and expression. To increase the spatial, temporal, and overall transient activity of Cas9, we employed both the constitutive 2 × 35S promoter and the cell division‐specific YAO promoter to drive the expression of Cas9 [Figure 2A, V4 (Yan et al., 2015; Zhang, LeBlanc, et al., 2017)]. We expected that a combination of 2 × 35S and YAO promoters (Figure 2A, V1, V2 and V4) would provide better coverage for the Cas9 in our transient expression system to improve editing. In addition, the VIrE2‐INTERACTING PROTEIN (VIP1) and the Agrobacterium effector protein VirE2 genes, driven by Ubi10 and 2 × 35S promoters, respectively, were included to facilitate more efficient Agrobacterium T‐DNA delivery (Figure 2A, V4, Gelvin, 2017; Lacroix & Citovsky, 2013, 2019). These four vectors (Figure 2A; V1–V4), each transformed into independent Agrobacterium lines, were simultaneously applied to the decapitated site of young seedlings. A detailed description and sequences of the IPGEC system are provided in Data S1 and Table S1.

Plasmid construction

The vectors used in this study were assembled using the Golden braid (GB) modular cloning system (Sarrion‐Perdigones et al., 2013). Transcriptional units were synthesized by GeneArt (Thermo Fisher Scientific, Regensburg, Germany) and served for GB cloning. For a detailed list of all transcriptional units and plasmids constructed in this study (see Table S1; Data S1). The GB assembly was conducted according to the GB protocol using T4 ligase and either BsaI‐HF or BsmBI enzymes (New England Biolabs, Ipswich, MA, USA). The p35S‐CFP plasmid was kindly provided by the lab of Prof. Shaul Yalovsky (TAU, Israel), and the NptII‐RUBY plasmid (Kumar et al., 2022) was kindly provided by the lab of Dr. Samuel Bocobza (Agricultural Research Organization, Israel). The p35S‐GFP plasmid was kindly provided by the lab of Dr. Arthur Schaffer (Agricultural research organization, Israel). All vectors were inserted into Escherichia coli (DH5α)‐heat‐shock competent cells and selected using suitable antibiotics and X‐gal, verified by PCR and Sanger sequencing. Agrobacterium tumefaciens (EHA105 strain) was transformed using 100 ng plasmid by heat shock.

Agrobacterium infiltration for transient expression and fluorescence

For transient expression assays, we used 8‐month‐old Carrizo plants or 3‐week‐old Nicotiana benthamiana plants. For Carrizo (Figure 3), two Agrobacterium lines, harboring the V2 or the V3 plasmids at OD600 = 0.2 each, were mixed and immersed in infiltration medium (IM) that contained 10 mm MES [2‐(N‐Morpholino) ethanesulfonic acid sodium salt; Duchefa, Haarlem, The Netherlands], 10 mm MgCl2, 5% sucrose, and 200 μm acetosyringone (pH 5.6). After 1 h at 28°C and 220 rpm, a gentle incision (2–5 mm) was made at the base of the abaxial leaf near the midrib, and Agrobacterium suspension was injected using a 1 ml syringe (Artsana Group, Como, Italy) without a needle. Six plants, 3 leaves each, were injected and samples were collected at three time points: 3, 6, and 10 days following agro‐infiltration, harvested and snap‐frozen in liquid nitrogen, and later used for expression analysis. At the end of the experiment, samples were also imaged by confocal microscopy. For tobacco agro‐infiltration (Figure S1), three agrobacteria, harboring the p35S‐GFP, pNos‐mCherry, or the p35S‐CFP at OD600 = 0.2 each, were mixed and immersed in IM as described above and infiltrated into leaves using a 1 ml syringe (without a needle or incision). Confocal imaging was performed 3 days post‐infiltration.

RNA extraction, cDNA preparation, and quantitative real‐time PCR

RNA was extracted from leaves using the Total RNA (plant) Mini Kit (Geneaid, New Taipei, Taiwan). The RNA integrity was validated using NanoDrop (MaestroNano, New Taipei, Taiwan) and gel electrophoresis. cDNA strands were synthesized using a designated kit (qScript cDNA Synthesis Kit; QuantaBio, Beverly, MA, USA) and subjected to PCR (SimpliAmp™ Thermal Cycler; Thermo Fisher Scientific), according to the manufacturer's instructions. Quantitative PCR was performed using the PikoReal 96 Real‐Time PCR System (Thermo Fisher Scientific) and qPCRBIO SyGreen Blue Mix Hi‐ROX kit by PCR Biosystems (London, UK). Primers used for real‐time PCR are specified in Table S2.

GFP screening, confocal microscopy imaging

Screening for GFP‐expressing plants was performed using the Xite‐RB fluorescence flashlight system (Nightsea; Electron Microscopy Sciences, Hatfield, PA, USA). Confocal image acquisition was done using a Leica SP8 laser scanning microscope (Leica, Wetzlar, Germany) equipped with a solid‐state laser with 488 nm light, a HCPL APO CS 63×/1.2 water immersion objective (Leica) and Leica Application Suite X software (LASX; Leica). Images of GFP signal were acquired using the 488‐nm laser light, and the emission was detected with a HyD (hybrid) detector in a range of 500–525 nm. Autofluorescence of the chloroplasts was detected in a range of 650–750 nm with a PMT detector.

In planta transformation and genome editing of citrus seedlings

In planta transformation in citrus seedlings was conducted following the work of (Rizwan et al., 2021; Zhang, Zhang, et al., 2017) with some modifications. Agrobacterium strain EHA105 harboring either the V1, V2, V3, or the V4 plasmids was grown in 50 ml LB with 50 mg L−1 kanamycin (Kan; Duchefa) overnight at 28°C (220 rpm). In the morning of the experiment, the different agrobacteria were calculated to form a mixed OD600 of 0.35 for V1, 0.35 for V2, 0.25 for V3, and 0.2 for V4 (total OD600 of 1.15) in a single 50 ml tube (Agro‐mix). The Agro‐mix was centrifuged at 3000 g for 10 min, and the cell pellet was re‐suspended with MS + 5% Suc solution; a Murashige and Skoog medium (MS; Duchefa) supplemented with 5% sucrose (Suc; Duchefa) and 200 μm acetosyringone [3′,5′‐dimethoxy‐4′‐hydroxyacetophenone; Sigma‐Aldrich, Jerusalem, Israel], adjusted to pH 5.8. The suspension was incubated for 1.5 h (28°C, 220 rpm) prior to the experiment. Seedlings of 4‐month‐old Carrizo hybrid or Duncan, Hudson, and Foster grapefruit seedlings were cut 4–5 cm above ground (above cotyledons and under the first true leaf) and a 10 μl tip filled with Agro‐mix was placed on the decapitated site for 1 hour in the dark to induce infection (Figure 1). Following tip removal, the wound infected sites were wrapped with parafilm and kept in the dark for 10 days. Three days following co‐culturing, parafilm was removed and cotton balls submerged in agar trap solution (0.5 MS × 0.5 Suc, pH 5.8, 0.4% plant agar), containing 50 mg L−1 Kan and 250 mg L−1 Claforan (Cla; Cefotaxim, Duchefa), were used to immerse the wound sites at least two times. Wound sites with a remaining agar trap solution were wrapped again with parafilm and were kept in the dark for another 7 days. Plants were transferred to the light (70 μmol m−2 sec−1) starting day 11, and were kept in high humidity using plastic dome covers. The induction of new shoots was notable starting from 3 weeks after infection and up to 2 months.

Regeneration efficiency measurements

Regeneration efficiency was assayed 60 days following infection. The number of shoots emerging from the side and those emerging from the top cut site (referred to as “Top”) was counted for each plant. Regeneration efficiency was determined as the number of plants having at least one shoot emerging from the top, out of the total plants assayed. In addition, we calculated the percentage of plants displaying two shoots or more that regenerated from the top cut site (referred to as “+2 top shoots”).

Mutant screening and DNA analyses

For next‐generation sequencing (NGS), Sanger‐ and amplicon sequencing, DNA was extracted using the GenElute plant genomic DNA kit (Sigma‐Aldrich) according to the manufacturer's protocol. For NGS, the CsPDS sgRNA's target sites (nine overall; gRNA1, gRNA2, gRNA3, gRNA4, gRNA5, gRNA8, gRNA11, gRNA13a, gRNA13b) were amplified using primers that contain adapters designed for library preparation (Table S3). We created a pool of all PCR products for each sample in a single tube, which later served for library preparation. Sequencing was performed using the DNBSEQ‐G400 system (MGI) at Syntezza Bioscience Ltd. (Jerusalem, Israel). To sequence large deletions, we amplified the CsPDS gene from exon 1 to exon 13 using the high‐fidelity proof‐reading enzyme Phanta flash master mix (Vazyme Biotech, Nanjing, China) according to the manufacturer's protocol. Amplified products were subjected to gel or column purification (NucleoSpin kit, Macherey‐Nagel, Düren, Germany) and amplicon sequencing was performed by outsourcing (Plasmidaurus, Oxford NANOPORE Technologies, Oxford, UK). DNA alignments were performed using the Benchling platform (San Francisco, CA, USA). To screen a large population (Figure 7), we used the high‐resolution melt (HRM) analysis. DNA was extracted using a crude extraction method as previously described (Kim et al., 2016) with slight modifications. Young citrus leaves were snap‐frozen in liquid nitrogen and powdered using the TissueLyser II (Qiagen, Hilden, Germany). DNA was extracted with 200 μl of extraction buffer (100 mm Tris–HCl [pH 9.5], 1 m KCl, 10 mm EDTA [pH 8]). After extraction, DNA was incubated at 65°C for 1 h, 1 ml water was added, and samples were centrifuged (13 000 g , 15 min). The upper phase was carefully transferred to a new tube, and the DNA was stored at −20°C. HRM for a representative target site was performed using the CFX Duet real‐time PCR system (Bio‐Rad Laboratories Ltd., Hercules, CA, USA), using the AccuMelt HRM superMix (Quantabio, Beverly, MA, USA), according to the manufacturer's protocols. Primers were designed to flank the guides: gRNA3, gRNA4, gRNA5, gRNA13a, and gRNA13b. Data were analyzed using the Bio‐Rad precision melt analysis software (Bio‐Rad), and selected plant DNA's were sent for Sanger sequencing or NGS analysis for verification. TaqMan analysis was performed using the AriaMX Real‐Time PCR System (Agilent, Santa Clara, CA, USA) with FAM and CY5 optical modules, following the manufacturer's protocols. Reactions were carried out using AccuStart Genotyping ToughMix (Quantabio) according to the manufacturer's instructions. Primers were designed to flank the fluorescent probes targeting GFP and ACTIN sequences. Primers and probes were designed by Agentek (Israel) and synthesized by Biosearch Technologies (Hoddesdon, UK). Primers used for NGS, Sanger sequencing, amplicon sequencing, HRM, and TaqMan analysis (including probe sequences) are listed in Table S3.

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.

Supporting information

Data S1. Detailed sequences of the vectors used in this study.

TPJ-122-0-s001.pdf (333KB, pdf)

Table S1. IPGEC and RUBY genetic building blocks, served for golden braid assembly in this study.

Table S2. Quantitative real‐time PCR primers used in this study.

Table S3. Primers used in this study for genotyping.

Table S4. Detailed read counts for samples presented in Figure 7.

Table S5. Number of biological replicates (total number of seedlings) for each treatment used in Figure 8(C).

Figure S1. Confocal imaging of tobacco leaves following agro‐infiltration of three fluorescent proteins.

Figure S2. Summary of transcription factors used in the V3 vector, which function upstream of WUS and STM.

Figure S3. Additional Carrizo lines (#54, #73) exhibiting albino phenotype, as shown in Figure 5.

TPJ-122-0-s002.pdf (467.4KB, pdf)

ACKNOWLEDGEMENTS

We would like to thank Prof. Vitaly Citovsky (State University of New York, Stony Brook, USA) for fruitful discussions and idea exchanges.

Contributor Information

Yoram Eyal, Email: eyalab@agri.gov.il.

Nir Carmi, Email: vhncarmi@agri.gov.il.

DATA AVAILABILITY STATEMENT

All relevant data can be found within the manuscript and its supporting materials.

REFERENCES

  1. Alvarez, D. , Cerda‐Bennasser, P. , Stowe, E. , Ramirez‐Torres, F. , Capell, T. , Dhingra, A. et al. (2021) Fruit crops in the era of genome editing: closing the regulatory gap. Plant Cell Reports, 40(6), 915–930. [DOI] [PubMed] [Google Scholar]
  2. Belanger, J.G. , Copley, T.R. , Hoyos‐Villegas, V. , Charron, J.B. & O'Donoughue, L. (2024) A comprehensive review of in planta stable transformation strategies. Plant Methods, 20(1), 20–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Campa, M. , Miranda, S. , Licciardello, C. , Lashbrooke, J.G. , Dalla Costa, L. , Guan, Q. et al. (2024) Application of new breeding techniques in fruit trees. Plant Physiology, 194(3), 1304–1322. [DOI] [PubMed] [Google Scholar]
  4. Cao, X. & Jiao, Y. (2020) Control of cell fate during axillary meristem initiation. Cellular and Molecular Life Sciences, 77(12), 2343–2354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cody, J.P. , Maher, M.F. , Nasti, R.A. , Starker, C.G. , Chamness, J.C. & Voytas, D.F. (2023) Direct delivery and fast‐treated Agrobacterium co‐culture (fast‐TrACC) plant transformation methods for Nicotiana benthamiana . Nature Protocols, 18(1), 81–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Debernardi, J.M. , Tricoli, D.M. , Ercoli, M.F. , Hayta, S. , Ronald, P. , Palatnik, J.F. et al. (2020) A GRF‐GIF chimeric protein improves the regeneration efficiency of transgenic plants. Nature Biotechnology, 38(11), 1274–1279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dutt, M. , Zambon, F.T. , Erpen, L. , Soriano, L. & Grosser, J. (2018) Embryo‐specific expression of a visual reporter gene as a selection system for citrus transformation. PLoS One, 13, e0190413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Eshed Williams, L. (2021) Genetics of shoot meristem and shoot regeneration. Annual Review of Genetics, 55, 661–681. [DOI] [PubMed] [Google Scholar]
  9. Gelvin, S.B. (2017) Integration of Agrobacterium T‐DNA into the plant genome. Annual Review of Genetics, 51, 195–217. [DOI] [PubMed] [Google Scholar]
  10. Goldenberg, L. , Yaniv, Y. , Porat, R. & Carmi, N. (2014) Effects of gamma‐irradiation mutagenesis for induction of seedlessness, on the quality of mandarin fruit. Food and Nutrition Sciences, 5(10), 943–952. [Google Scholar]
  11. Greb, T. , Clarenz, O. , Schafer, E. , Muller, D. , Herrero, R. , Schmitz, G. et al. (2003) Molecular analysis of the LATERAL SUPPRESSOR gene in Arabidopsis reveals a conserved control mechanism for axillary meristem formation. Genes & Development, 17(9), 1175–1187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Guo, D. , Zhang, J. , Wang, X. , Han, X. , Wei, B. , Wang, J. et al. (2015) The WRKY transcription factor WRKY71/EXB1 controls shoot branching by transcriptionally regulating RAX genes in Arabidopsis. Plant Cell, 27(11), 3112–3127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hamada, H. , Linghu, Q. , Nagira, Y. , Miki, R. , Taoka, N. & Imai, R. (2017) An in planta biolistic method for stable wheat transformation. Scientific Reports, 7(1), 11443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hao, S.Y. , Zhang, Y.Y. , Li, R.D. , Qu, P.Y. & Cheng, C.Z. (2024) Agrobacterium‐mediated in planta transformation of horticultural plants: current status and future prospects. Scientia Horticulturae, 325, 112693. Available from: 10.1016/j.scienta.2023.112693 [DOI] [Google Scholar]
  15. He, Y.B. , Zhang, T. , Sun, H. , Zhan, H.D. & Zhao, Y.D. (2020) A reporter for noninvasively monitoring gene expression and plant transformation. Horticulture Research, 7(1), 152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. He, Z. , Webster, S. & He, S.Y. (2022) Growth–defense trade‐offs in plants. Current Biology, 32(12), R634–R639. [DOI] [PubMed] [Google Scholar]
  17. Huang, X.E. , Jia, H.G. , Xu, J. , Wang, Y.C. , Wen, J.W. & Wang, N. (2023) Transgene‐free genome editing of vegetatively propagated and perennial plant species in the T0 generation via a co‐editing strategy. Nature Plants, 9(10), 1591–1597. [DOI] [PubMed] [Google Scholar]
  18. Huang, X.E. , Wang, Y.C. & Wang, N. (2022) Highly efficient generation of canker‐resistant sweet orange enabled by an improved CRISPR/Cas9 system. Frontiers in Plant Science, 12, 769907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Huang, X.E. , Wang, Y.C. , Xu, J. & Wang, N. (2020) Development of multiplex genome editing toolkits for citrus with high efficacy in biallelic and homozygous mutations. Plant Molecular Biology, 104(3), 297–307. [DOI] [PubMed] [Google Scholar]
  20. Jelly, N.S. , Valat, L. , Walter, B. & Maillot, P. (2014) Transient expression assays in grapevine: a step towards genetic improvement. Plant Biotechnology Journal, 12, 1231–1245. [DOI] [PubMed] [Google Scholar]
  21. Jia, H.G. , Omar, A.A. , Xu, J. , Dalmendray, J. , Wang, Y.C. , Feng, Y. et al. (2024) Generation of transgene‐free canker‐resistant cv. Hamlin in the T0 generation through Cas12a/CBE co‐editing. Frontiers in Plant Science, 15, 1385768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jia, H.G. , Zhang, Y.Z. , Orbovic, V. , Xu, J. , White, F.F. , Jones, J.B. et al. (2017) Genome editing of the disease susceptibility gene CsLOB1 in citrus confers resistance to citrus canker. Plant Biotechnology Journal, 15, 817–823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Keller, T. , Abbott, J. , Moritz, T. & Doerner, P. (2006) Arabidopsis REGULATOR OF AXILLARY MERISTEMS1 controls a leaf axil stem cell niche and modulates vegetative development. Plant Cell, 18(3), 598–611. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Khadgi, A. , Sagawa, C.H.D. , Vernon, C. , Mermaz, B. & Irish, V.F. (2024) Optimization of in planta methodology for genome editing and transformation in citrus. Frontiers in Plant Science, 15, 1438031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kim, S.‐R. , Yang, J. , An, G. & Jena, K.K. (2016) A simple DNA preparation method for high quality polymerase chain reaction in rice. Plant Breeding and Biotechnology, 4, 99–106. [Google Scholar]
  26. Kumar, M. , Tripathi, P.K. , Ayzenshtat, D. , Marko, A. , Forotan, Z. & Bocobza, S. (2022) Increased rates of gene‐editing events using a simplified RNAi configuration designed to reduce gene silencing. Plant Cell Reports, 41(10), 1987–2003. [DOI] [PubMed] [Google Scholar]
  27. Lacroix, B. & Citovsky, V. (2013) Characterization of VIP1 activity as a transcriptional regulator in vitro and in planta . Scientific Reports, 3, 2440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lacroix, B. & Citovsky, V. (2019) Pathways of DNA transfer to plants from Agrobacterium tumefaciens and related bacterial species. Annual Review of Phytopathology, 57, 231–251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lian, Z.Y. , Nguyen, C.D. , Liu, L. , Wang, G.L. , Chen, J.J. , Wang, S.H. et al. (2022) Application of developmental regulators to improve in planta or in vitro transformation in plants. Plant Biotechnology Journal, 20(8), 1622–1635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lin, C.S. , Hsu, C.T. , Yang, L.H. , Lee, L.Y. , Fu, J.Y. , Cheng, Q.W. et al. (2018) Application of protoplast technology to CRISPR/Cas9 mutagenesis: from single‐cell mutation detection to mutant plant regeneration. Plant Biotechnology Journal, 16(7), 1295–1310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Maher, M.F. , Nasti, R.A. , Vollbrecht, M. , Starker, C.G. , Clark, M.D. & Voytas, D.F. (2020) Plant gene editing through de novo induction of meristems. Nature Biotechnology, 38(1), 84–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mahmoud, L.M. , Kaur, P. , Stanton, D. , Grosser, J.W. & Dutt, M. (2022) A cationic lipid mediated CRISPR/Cas9 technique for the production of stable genome edited citrus plants. Plant Methods, 18(1), 33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mei, G.G. , Chen, A. , Wang, Y.R. , Li, S.Q. , Wu, M.Y. , Hu, Y.L. et al. (2024) A simple and efficient in planta transformation method based on the active regeneration capacity of plants. Plant Communications, 5(4), 100822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Nasti, R.A. , Zinselmeier, M.H. , Vollbrecht, M. , Maher, M.F. & Voytas, D.F. (2021) Fast‐TrACC: a rapid method for delivering and testing gene editing reagents in somatic plant cells. Frontiers in Genome Editing, 2, 621710. Available from: 10.3389/fgeed.2020.621710 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nerva, L. , Dalla Costa, L. , Ciacciulli, A. , Sabbadini, S. , Pavese, V. , Dondini, L. et al. (2023) The role of Italy in the use of advanced plant genomic techniques on fruit trees: state of the art and future perspectives. International Journal of Molecular Sciences, 24, 977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Parajuli, S. , Huo, H.Q. , Gmitter, F.G. , Duan, Y.P. , Luo, F. & Deng, Z.N. (2022) Editing the CsDMR6 gene in citrus results in resistance to the bacterial disease citrus canker. Horticulture Research, 9, uhac082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Polturak, G. , Breitel, D. , Grossman, N. , Sarrion‐Perdigones, A. , Weithorn, E. , Pliner, M. et al. (2016) Elucidation of the first committed step in betalain biosynthesis enables the heterologous engineering of betalain pigments in plants. The New Phytologist, 210, 269–283. [DOI] [PubMed] [Google Scholar]
  38. Qin, G. , Gu, H. , Ma, L. , Peng, Y. , Deng, X.W. , Chen, Z. et al. (2007) Disruption of phytoene desaturase gene results in albino and dwarf phenotypes in Arabidopsis by impairing chlorophyll, carotenoid, and gibberellin biosynthesis. Cell Research, 17, 471–482. [DOI] [PubMed] [Google Scholar]
  39. Rizwan, H.M. , Yang, Q. , Yousef, A.F. , Zhang, X.X. , Sharif, Y. , Kaijie, J. et al. (2021) Establishment of a novel and efficient Agrobacterium‐mediated in planta transformation system for passion fruit (Passiflora edulis). Plants (Basel), 10(11), 2459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sarrion‐Perdigones, A. , Vazquez‐Vilar, M. , Palací, J. , Castelijns, B. , Forment, J. , Ziarsolo, P. et al. (2013) GoldenBraid 2.0: a comprehensive DNA assembly framework for plant synthetic biology. Plant Physiology, 162, 1618–1631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Song, G.Q. , Prieto, H. & Orbovic, V. (2019) Agrobacterium‐mediated transformation of tree fruit crops: methods, progress, and challenges. Frontiers in Plant Science, 10, 226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Su, H. , Wang, Y.C. , Xu, J. , Omar, A.A. , Grosser, J.W. , Calovic, M. et al. (2023) Generation of the transgene‐free canker‐resistant Citrus sinensis using Cas12a/crRNA ribonucleoprotein in the T0 generation. Nature Communications, 14(1), 3957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tyurin, A.A. , Suhorukova, A.V. , Kabardaeva, K.V. & Goldenkova‐Pavlova, I.V. (2020) Transient gene expression is an effective experimental tool for the research into the fine mechanisms of plant gene function: advantages, limitations, and solutions. Plants (Basel), 9(9), 1187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vardi, A. , Levin, I. & Carmi, N. (2008) Induction of seedlessness in citrus: from classical techniques to emerging biotechnological approaches. Journal of the American Society for Horticultural Science, 133(1), 117–126. [Google Scholar]
  45. Wang, X. , Xu, Y.T. , Zhang, S.Q. , Cao, L. , Huang, Y. , Cheng, J.F. et al. (2017) Genomic analyses of primitive, wild and cultivated citrus provide insights into asexual reproduction. Nature Genetics, 49(5), 765–772. [DOI] [PubMed] [Google Scholar]
  46. Yadav, C.B. , Rozen, A. , Eshed, R. , Ish‐Shalom, M. , Faigenboim, A. , Dillon, N. et al. (2023) Promoter insertion leads to polyembryony in mango – a case of convergent evolution with citrus. Horticulture Research, 10(12), uhad227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yan, L.H. , Wei, S.W. , Wu, Y.R. , Hu, R.L. , Li, H.J. , Yang, W.C. et al. (2015) High‐efficiency genome editing in Arabidopsis using promoter‐driven CRISPR/Cas9 system. Molecular Plant, 8(12), 1820–1823. [DOI] [PubMed] [Google Scholar]
  48. Yang, F. , Wang, Q. , Schmitz, G. , Müller, D. & Theres, K. (2012) The bHLH protein ROX acts in concert with RAX1 and LAS to modulate axillary meristem formation in Arabidopsis. The Plant Journal, 71(1), 61–70. [DOI] [PubMed] [Google Scholar]
  49. Zhang, F. , LeBlanc, C. , Irish, V.F. & Jacob, Y. (2017) Rapid and efficient CRISPR/Cas9 gene editing in Citrus using the YAO promoter. Plant Cell Reports, 36(12), 1883–1887. [DOI] [PubMed] [Google Scholar]
  50. Zhang, T.Q. , Lian, H. , Zhou, C.M. , Xu, L. , Jiao, Y.L. & Wang, J.W. (2017) A two‐step model for de novo activation of WUSCHEL during plant shoot regeneration. Plant Cell, 29(5), 1073–1087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhang, Y.Y. , Zhang, D.M. , Zhong, Y. , Chang, X.J. , Hu, M.L. & Cheng, C.Z. (2017) A simple and efficient in planta transformation method for pommelo (Citrus maxima) using Agrobacterium tumefaciens . Scientia Horticulturae, 214, 174–179. [Google Scholar]
  52. Zhu, H.C. , Li, C. & Gao, C.X. (2020) Applications of CRISPR‐Cas in agriculture and plant biotechnology. Nature Reviews. Molecular Cell Biology, 21(12), 661–677. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1. Detailed sequences of the vectors used in this study.

TPJ-122-0-s001.pdf (333KB, pdf)

Table S1. IPGEC and RUBY genetic building blocks, served for golden braid assembly in this study.

Table S2. Quantitative real‐time PCR primers used in this study.

Table S3. Primers used in this study for genotyping.

Table S4. Detailed read counts for samples presented in Figure 7.

Table S5. Number of biological replicates (total number of seedlings) for each treatment used in Figure 8(C).

Figure S1. Confocal imaging of tobacco leaves following agro‐infiltration of three fluorescent proteins.

Figure S2. Summary of transcription factors used in the V3 vector, which function upstream of WUS and STM.

Figure S3. Additional Carrizo lines (#54, #73) exhibiting albino phenotype, as shown in Figure 5.

TPJ-122-0-s002.pdf (467.4KB, pdf)

Data Availability Statement

All relevant data can be found within the manuscript and its supporting materials.

Articles from The Plant Journal are provided here courtesy of Wiley

RESOURCES