Optimized R2 Retroelement Complexes Enable Precise and Efficient DNA Insertion into Plant Genomes

Optimized R2 Retroelement Complexes Enable Precise and Efficient DNA Insertion into Plant Genomes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Optimized R2 Retroelement Complexes Enable Precise and Efficient DNA Insertion into Plant Genomes Gozde Demirer, Kimberley Muchenje, Yunqing Wang, Tufan Oz, Eugene Li, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7436428/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Precise, targeted insertion of multi-kilobase DNA sequences is critical for studying gene function, ensuring robust transgene expression, and stacking traits in crops, but remains difficult to achieve. Existing targeted insertion methods in plants relying on programmable nucleases are inefficient and can generate unwanted mutations. Newer technologies based on prime editors, transposases, and site-specific recombinases extend capabilities but remain constrained with low efficiencies, off-target integration, silencing, or limited cargo DNA size. R2 non-long terminal repeat (non-LTR) retrotransposons integrate via target-primed reverse transcription specifically targeting the 25S ribosomal DNA multicopy site and enabling double-strand-break-free installation of gene-sized DNA sequences. Here, we adapted the avian Taeniopygia guttata R2 protein (R2Tg) for targeted DNA insertion into plant genomes and defined design rules for efficient integration and functional protein expression from the 25S rDNA locus. We engineered R2Tg expression cassettes and RNA payloads carrying an intron-disrupted mCherry retrotransposition reporter with length-optimized rDNA homology arms. We also optimized the R2 editor delivery format and incubation conditions. In Nicotiana benthamiana leaves, Arabidopsis thaliana protoplasts, and Solanum lycopersicum seedlings, the optimized R2Tg editor system achieved targeted insertion with efficiencies up to 24% for 2-5kb size payloads, which were carefully validated at the molecular level. This work establishes a compact, ribonucleoprotein-based platform for targeted DNA insertion into plant genomes, leveraging a multicopy genomic safe harbor site to enable efficient multi-kilobase precise gene addition. Biological sciences/Biotechnology/Plant biotechnology/Molecular engineering in plants Biological sciences/Plant sciences/Plant molecular biology R2 retrotransposon targeted DNA integration safe harbor site plant bioengineering synthetic biology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Crop yields are increasingly threatened by rising global temperatures, intensifying drought, and the expanding range of plant diseases 1 . Genome engineering offers a powerful, timely strategy for enhancing climate resilience, food security, and sustainable biomanufacturing 2 , 3 . The development of improved crops relies heavily on the incorporation of foreign genes and heterologous metabolic pathways into plant genomes, where most approaches use Agrobacterium tumefaciens for DNA delivery and insertion. However, the random DNA integration of Agrobacterium can often cause variable gene expression due to local chromatin context and nearby regulatory elements, lead to silencing, and prevent precise tagging of specific genes or loci, all of which are limitations complicating functional genomics studies and requiring extensive screening across generations to identify lines with desired, stable, and heritable expression 4–7 . Emerging targeted insertion technologies allow controlled placement of DNA sequences at predetermined genomic loci, eliminating risks of unintended gene expression changes and genomic instability 8 . Furthermore, precise DNA insertion at defined genomic sites is essential for probing gene function and regulation and enables robust outputs in synthetic gene circuits, efficient trait stacking, and accelerated development of engineered crops for agriculture and bioproduction 9,10 . To date, multiple strategies have been developed to achieve site-specific DNA integration in plants. Programmable nucleases, such as CRISPR-Cas9, introduce targeted double-strand breaks (DSBs), which are repaired by host machinery 11,12 . Although homology-directed repair can result in precise integration using donor templates with flanking homology arms at the DSB site 13 , this pathway is not preferred in most plant cells, where DNA breaks are predominantly repaired by non-homologous end joining, often leading to low efficiency, imprecise insertions, and indel generation 14 . Moreover, DSBs can lead to unpredictable outcomes, such as large deletions or chromosomal rearrangements 15 . As an alternative, prime editing offers a DSB-free approach for introducing precise nucleotide changes and short sequence insertions into the genome 16,17 . Despite broad applicability, its efficiency to insert large DNA segments in plants remains limited, restricting its use for integrating full-length genes or metabolic pathways 18 . Expanding the capabilities of prime editing, PrimeRoot employs optimized prime editing guide RNA (pegRNA) architectures, prime editors, and site-specific recombinases to achieve DSB-free targeted insertion of large DNA sequences into plant genomes 19 . Although promising with up to 6.3% efficiency, its broader application may be limited by lack of generalizable pegRNA design rules to ensure robust activity across diverse loci, and the low efficiency of the recombinase recognition site installation, which is a critical first step in PrimeRoot. These challenges highlight the need for continued engineering of more effective editors to fully realize the potential of precise, programmable DNA insertion in plants. Expanding the toolkit further, transposon-based systems offer another alternative for programmable insertion. Transposons are mobile elements that can be broadly categorized as DNA transposons or retrotransposons based on their integration intermediates. They possess sophisticated insertion mechanisms setting them up as powerful tools for targeted insertion 20,21 . A recent discovery of CRISPR-associated DNA transposases (CAST) in bacterial genomes enabled the development of editors that use programmable guide RNAs to direct transposon insertion to defined genomic sites 22,23 . Unlike nuclease-based editors, they lack DNA cleavage activity and instead rely entirely on transposase machinery to perform the integration. Even though their translation to eukaryotic human cells has been successful – albeit with a low efficiency 24 – native bacterial CAST system has yet to demonstrate feasibility in plants. However, building on this discovery, transposase-assisted target-site integration (TATSI) system was developed, which co-expresses Cas9 nuclease with the rice Pong DNA transposase to enable high efficiency, targeted DNA insertion in Arabidopsis thaliana and soybean 25 . While TATSI offers a promising approach for targeted DNA insertion in plants, its performance can be affected by high rates of off-target integration at native transposase recognition motifs, reliance on DSBs, and increased susceptibility to silencing due to the use of plant-derived transposases. In addition to DNA transposons, retrotransposons have also shown recent promise for targeted gene insertion in eukaryotes. R2 non-long terminal repeat (LTR) retrotransposons are a class of sequence-specific mobile elements that insert into the conserved 28S ribosomal DNA (rDNA) tandem repeat loci and have so far been identified in the genomes of multicellular animals including insects, crustaceans, and non-mammalian vertebrates 26–28 . R2 proteins have been successfully repurposed as high-efficiency genome editors in mammalian systems, both at their conserved target sites and at other RNA-guided genomic loci through Cas9 fusion, leveraging their target-primed reverse transcription (TPRT) mechanism 29–32 . Engineered R2 retrotransposons mediate RNA-templated insertions at the native 28S rDNA locus with efficiencies up to 80% in HEK293T cells and 60% in mouse embryos, and they achieve up to 40% at Cas9-directed, non-rDNA sites 30–32 . R2 retrotransposons provide a unique, DSB-free approach for targeted DNA insertion into the 28S rDNA site, which is referred to as 25S rDNA site in plants, offering broad adaptability for genome editing across plant species. The required domains for function, DNA binding, endonuclease, and reverse transcriptase, are encoded within a single open reading frame, enabling a compact editor design that can be co-delivered with its RNA payload as a ribonucleoprotein (RNP) complex. Moreover, the high copy number of rDNA repeats in plants (~800 copies in A. thaliana 33 , ~1247 copies in Nicotiana benthamiana 34 , ~1200 copies in Solanum lycopersicum 35,36 and ~6650 copies in wheat 37 ) can further increase integration frequency and expression, and simplify downstream screening for precise insertion events. Here, we established and optimized a plant genome editor based on the R2 retrotransposon that enables targeted, high-efficiency DNA insertion into the 25S rDNA loci. This system combines: i) an optimized R2 expression, ii) RNA donor carrying mCherry and RUBY retrotransposition reporters with length-optimized rDNA-homology arms, and iii) systematically tuned 5′ and 3′ UTRs in the RNA template. The optimized system achieves integration of payloads up to 5kb size in Nicotiana benthamiana leaves, Arabidopsis thaliana protoplasts, and Solanum lycopersicum seedlings with efficiencies reaching 24% of transformed cells quantified by imaging and 0.9-1 copies per genome quantified by droplet digital PCR on both 5′ and 3′ integrated junctions. R2-mediated insertion events were validated at the 25S rDNA loci with molecular evidence confirming both high efficiency and target specificity, establishing a robust platform for precise DNA insertion into plant genomic safe-harbor loci for various fundamental plant biology or biotechnology applications. Results Establishing an R2 retrotransposon-mediated platform for targeted DNA insertion in plants. Building on the well-characterized target-primed reverse transcription (TPRT) activity of R2 proteins in mammalian cells 27,28 , we investigated whether the same machinery could mediate precise DNA insertion in the plant nuclear genome. In the canonical mechanism, an R2 ribonucleoprotein complex composed of an R2 protein and its RNA template binds the 28S rDNA locus (25S rDNA in plants) in a manner that depends on the secondary structures of RNA sequences within the 5′ and 3′ untranslated regions (UTRs) 27,28 . The complex then nicks the bottom strand, exposing a 3′-hydroxyl primer, and synthesizes first-strand complementary DNA (cDNA) from its RNA template 27 . Subsequently, the R2 protein nicks the top strand, and initiates second-strand synthesis 28 , where the integrated element is flanked by the R2 5′ and 3′ UTRs ( Fig. 1A ). We deployed the R2 retrotransposon system (R2 system in short) to leverage this double-strand-break-free integration mechanism, combined with the high copy number 34,37 and transcriptional activity 38,39 of the 25S rDNA array, to achieve robust expression of integrated gene sequences in plants. Efficient TPRT requires high amounts of nuclear-localized R2 protein 40 . We therefore codon-optimized R2 open reading frames for plant expression, appended an N-terminal strong bipartite nuclear localization signal (BP NLS), and fused either a high affinity bioluminescence tag (HiBiT) for luminescence complementation or a single-copy YPet fluorescent protein to the N-terminus for quantifying and/or imaging protein expression ( Fig. 1B ). Each construct was driven by the strong pCmYLCV6 promoter and terminated by the A. thaliana tHSP 18.2 terminator. To further increase intracellular R2 copy number, long and short intergenic regions (LIR-SIR) together with the RepA replication-initiation protein were incorporated as a second set to test, generating self-replicating geminiviral (GV) replicons that boost transient expression 41 ( Fig. 1B ). To enable simple phenotypic detection of R2-mediated insertion events, we built an RNA payload to function as a retrotransposition reporter following a classic GFP reporter 42,43 , in which either the Solanum tuberosum ST-LS1 intron 44 or an Arabidopsis thaliana intron 45 -disrupted mCherry cassette was embedded between the 5′ and 3′ untranslated regions (UTRs) of the R2 element ( Fig. 1C) . Following transcription and intron splicing, the resulting RNA forms a contiguous template for reverse transcription via R2 protein. Hence, precise integration via R2 system restores an intron-less mCherry expression cassette, producing a fluorescent signal that enables visualization and quantification by confocal microscopy. To evaluate R2 system activity across different contexts, we introduced R2 protein and RNA payload constructs into Nicotiana benthamiana leaves and Solanum lycopersicum seedlings via Agrobacterium tumefaciens infiltration, and into A. thaliana protoplasts using polyethylene glycol (PEG)-mediated transfection ( Fig. 1D ). BFP fluorescent marker was included in the constructs for efficiency normalization. Phenotypic detection was achieved via confocal microscopy for N. benthamiana and S. lycopersicum leaves and flow cytometry for A. thaliana protoplasts. Molecular characterization of insertion junctions and copy number was performed using amplicon PCR, Sanger sequencing, high-depth next-generation sequencing (NGS), and droplet digital PCR (ddPCR) ( Fig. 1E ). R2 expression optimization and functional screening identify an active R2 system in plants To identify a functional R2 editor in plants, we first compared the expression profiles of four codon-optimized R2 proteins: R2Bm from Bombyx mori , R2Za from Zonotrichia albicollis , R2Tg from Taeniopygia guttata , and a rationally improved variant of R2Tg, R2Tg-OPT 30 . These proteins were expressed in N. benthamiana leaves using Agro-infiltration. All proteins were tested with and without geminiviral (GV) replicon elements to evaluate the impact of expression cassette copy number on R2 protein accumulation ( Fig. 1B ). 3 days after infiltration, we detected the expression of all YPet-tagged R2 proteins using confocal microscopy, where nuclear-localized YPet signal was consistently stronger in GV versions ( Fig. 2A ). Luminescence quantification using a plate reader for the HiBiT-tagged set of proteins in the plant nuclear fraction confirmed that GV parts significantly increased R2 protein levels across all variants ( Fig. 2B ). These results established our ability to express high levels of all tested R2 proteins in plants cells and that GV replicons are an effective strategy for increasing R2 protein levels in the nuclei of N. benthamiana leaf cells. Previous mammalian studies have shown that autocatalytic cleavage of the 5′ UTR is required to generate competent RNA templates for R2-mediated integration via TPRT 30 . To assess whether the 5′ UTR sequences used in our donor constructs enable autocleavage, we performed in vitro transcription and cleavage assays using the wild-type and minimal length T. castaneum and T. guttata 5’ UTR , each fused to 100 bp of the rDNA sequence upstream of the insertion site, referred to here as the left homology arm (LHA) ( Supplementary Fig. 1A ). Distinct cleavage products were observed by agarose gel electrophoresis at expected band sizes, confirming that all tested UTRs retain autocatalytic activity in vitro ( Supplementary Fig. 1B ). Next, we performed a systematic R2 editor function screen to identify an optimal pairing of R2 protein and donor RNA architecture. Each configuration combined one of the four R2 proteins with a donor cassette encoding the mCherry-based retrotransposition reporter and different UTR combinations ( Fig. 2C ). We evaluated four UTR combinations, all previously shown to be active in mammalian systems, derived from three species: minimal T. castaneum 5′ UTR with wild-type Geospiza fortis 3′ UTR; wild-type T. castaneum 5′ UTR with wild-type G. fortis 3′ UTR; wild-type T. guttata 5′ and 3′ UTRs; and wild-type T. guttata 5′ UTR paired with a minimal T. guttata 3′ UTR 29,30 ( Fig. 2C ), generating a library of 16 R2-UTR combinations to assess. Each all-in-one construct for the R2 editor function screen also encoded a constitutive mTagBFP2 reporter to normalize for transformation efficiency ( Fig. 2D ). To evaluate functional activity, 7 days after infiltration, we imaged N. benthamiana leaves infiltrated with constructs expressing all-in-one constructs for each of the 16 R2 protein-UTR combinations and scored mCherry fluorescence as a direct readout of successful retrotransposition into the 25S rDNA ( Fig. 2E ). R2Bm and R2Za proteins did not yield detectable integration activity when paired with any of 4 UTR combinations in our editor screen ( Supplementary Fig. 1C, D ). The highest integration efficiency, defined as the percentage of leaf cells with detectable mCherry signal, was obtained with wild-type R2Tg paired with its wild-type 5′ and 3′ UTRs at ~0.5% efficiency ( Fig. 2E, F ). Two additional R2Tg editor configurations with different UTRs also yielded successful integration ( Fig. 2E, F; Supplementary Fig. 1D ): R2Tg enabled detectable integration when paired with wild-type 5′ and minimal 3′ UTRs from T. guttata (0.11%), and wild-type T. castaneum 5′ UTR and wild-type G. fortis 3′ UTR (0.05%) ( Fig. 2E, F ). In contrast, no mCherry signal was observed when R2Tg was paired with the minimal T. castaneum 5′ UTR and wild-type G. fortis 3′ UTR ( Fig. 2E, F ). R2Tg-OPT exhibited minimal activity only when paired with wild-type T. guttata UTRs, and none of the other UTR combinations supported detectable integration by this variant. Our negative control samples consisting only of the payload without R2 protein did not generate any mCherry signal as expected ( Fig. 2F ; Supplementary Fig. 2) This preliminary R2 editor function screen established the wild-type R2Tg protein, paired with its wild-type 5′ and 3′ UTRs, as the most efficient R2 editor configuration, showing potential for further optimization via R2 protein and payload engineering. We next focused on increasing the integration efficiency of the R2Tg editor paired with the mCherry payload containing its wild-type 5′ and 3′ UTRs with a payload length of 2.9kb, which we refer to as the R2Tg editor system. R2Tg system optimization via improved protein & payload expression and heat shock treatment To ensure that mCherry fluorescence in our optimization assays only reflects bona fide R2-mediated integration events, we used an intron-disrupted mCherry retrotransposition reporter, same as above ( Fig. 3A ). This payload comprises an mCherry coding sequence in the antisense orientation relative to the RNA payload, interrupted by a plant intron in the sense orientation, flanked by R2TG wild-type 5′ and 3′ UTRs and 100 bp homology arms corresponding to the 25S rDNA target site. Transcription of the integrated cassette can only yield mCherry signal if the payload transcript is precisely spliced to remove the intron prior to R2-mediated insertion at the target locus. Agrobacteria T-DNA insertions or leaky expression cannot generate a functional, intron-free mCherry transcript because the cassette remains intron-bearing or incorrectly oriented, with the splice donor and acceptor motifs are present only on the sense strand, thereby ensuring the absence of false-positive signal in downstream fluorescence readouts. Using this reporter, we next tested whether targeted integration efficiency could be improved. Retrotransposon sequences are rapidly targeted by RNA-directed DNA methylation and converted to a stably silenced, heterochromatic state in plant cells 46–49 . We suspected that transcriptional interference 50 from all cassettes contained in the all-in-one construct and silencing of R2 editor encoding sequences may be limiting the efficiencies in our initial R2 editor screen. To mitigate this effect, we split the system into two plasmids: one encoding the R2Tg protein and the other carrying the R2Tg payload, a strategy intended to reduce transcriptional interference ( Fig. 3B ). We further inserted two A. thaliana introns 45 into the R2Tg protein coding sequence to suppress silencing of the R2 editor system ( Fig. 3B ). Initial efforts to express R2 proteins in N. benthamiana showed that protein expression could be significantly improved by the introduction of geminiviral replicons leveraging autonomous replication. In line with this finding, overexpressing the complete all-in-one editor system could boost integration efficiency by increasing the overall R2 editor DNA copy number. Following these insights, we set out to test whether separating the R2Tg editor components into two constructs (RT2g-2), introducing introns into the R2Tg coding sequence (RT2g-i), and flanking the full editor system with geminiviral replicons (GV-RT2g) would improve targeted integration efficiency at the 25S rDNA site in plants ( Fig. 3B ). Following the same method of N. benthamiana leaf Agro -infiltration and quantifying number of mCherry expressing cells from confocal images, we found that the separation of R2Tg editor components into 2 plasmids improved integration efficiency by 1.7-fold, while the introduction of introns into the R2Tg coding sequence combined with the 2-plasmid delivery approach resulted in a 2.7-fold improvement compared to the 1-plasmid delivery without introns ( Fig. 3C, E; Supplementary Fig. 3A for more replicates). Addition of geminiviral elements to the complete 1-plasmid R2Tg editing system led to a 3.3-fold increase in integration efficiency ( Fig. 3C, E; Supplementary Fig. 3A ), achieving highest integration efficiency of ~1.8% in this set (GV-RT2g-1). Together, these findings support the hypothesis that R2Tg protein silencing and transcriptional interference in the initial all-in-one constructs limit integration efficiency, and that strategies aimed at reducing both effects and increasing the copy number of both the R2Tg protein and payload cassettes can significantly enhance targeted integration at the 25S rDNA site. Successful applications of R2 editors in mammalian cells consistently incubate cells transformed with editors at 37°C before detecting editing events 29–32 . The R2 editor functional screen in N. benthamiana , however, was conducted under standard plant growth conditions of 22°C during the light period and 20°C during the dark period in a 16-hour light/8-hour dark photoperiod. Given the high body temperature of native organisms with R2, we hypothesized that the lower temperatures used in plant studies may be limiting the catalytic efficiency of R2-mediated TPRT, and that elevating the temperature via heat shock may enhance integration activity. To test this, Agrobacterium tumefaciens –infiltrated plants were first maintained at 25°C for 48 hours under a 16-hour light/8-hour dark cycle, allowing sufficient time for T-DNA delivery and accumulation of the R2 protein and RNA payload prior to the induction of heat shock. Plants were then exposed to a heat shock regimen of 25°C during the light period and 37°C during the dark period for an additional four days before analysis. Heat shock treatment of the single-plasmid R2Tg-1 system resulted in a 1.9-fold increase in targeted integration efficiency compared to the non-heat-shocked single-plasmid R2Tg-1 system. When applied to the geminiviral-amplified GV-R2Tg-1 system, heat shock yielded a 1.27-fold improvement, corresponding to a 4.2-fold increase relative to the baseline R2Tg-1 construct, reaching up to ~2% targeted integration efficiency (Fig. 3D, E; Supplementary Fig. 3A for more replicates). These findings revealed that the elevated temperature indeed enhances R2Tg-mediated TPRT in planta . The R2Tg editor system achieves integration into A. thaliana and N. benthamiana protoplasts To enable a more streamlined and quantitative workflow for assessing R2Tg-mediated integration at single-cell resolution using flow cytometry, we delivered all-in-one R2Tg editor constructs, with or without geminiviral elements, into A. thaliana protoplasts using polyethylene glycol (PEG)-mediated transformation. The baseline R2Tg-1 system yielded a targeted integration efficiency of 0.765%, while the geminiviral-amplified GV-R2Tg-1 system achieved 1.167% (Supplementary Fig. 3B) , as reported by flow cytometry referring to % cells with mCherry fluorescence. Although these results demonstrate that the R2Tg system is active in A. thaliana protoplasts, in addition to N. benthamiana leaves, protoplast viability declines significantly after 48 hours, limiting the number of intact cells and window for achieving and detecting integration events. To address this limitation, we established a complementary assay transforming N. benthamiana leaves via Agro -infiltration, and then extracting protoplasts from this transformed tissue, which allows longer incubation of the R2Tg editor system in planta . R2Tg-1 constructs were delivered by Agrobacterium infiltration, and targeted integration was detected and quantified 5 days post-infiltration in protoplasts extracted from infiltrated leaves using flow cytometry measuring mCherry fluorescence. This system yielded substantially higher integration efficiencies, reaching 4.5% with the R2Tg-1 construct (Supplementary Fig. 3C) . These findings confirm that the R2Tg editor system drives targeted integration into the 25S rDNA locus in both protoplast and leaf cells of two plant species. They further demonstrate that A. thaliana protoplasts provide a functional platform for early-stage detection, while intact leaves support extended assay durations and higher integration efficiencies for downstream applications. Compact RNA template with minimal homology length and 3 ′ ribozyme achieves the highest integration efficiency Molecular characterization of integration junctions at the 25S rDNA locus is critical to confirm the activity and specificity of R2Tg-mediated targeted insertion. However, the repetitive structure of rDNA tandem arrays and the presence of identical homology arms within the retrotransposition reporter generate false-positive chimeric PCR products, complicating unambiguous identification of genuine integration events 51 . In mammalian systems, such artifacts were successfully eliminated by delivering both components of the R2 editor system as RNA through the PRINT (precise RNA-mediated insertion of transgenes) method 29 . Given the technical limitations of the RNA delivery in plants, we designed a modified mCherry reporter harboring minimal homology arms of 33 base pairs upstream (R33) and 4 base pairs downstream (R4) relative to the 25S rDNA cleavage site (Fig. 4A) . We reasoned that reducing the homology arm length would minimize PCR-generated chimera formation, thereby enabling more accurate quantification of authentic integration events without compromising efficiency, consistent with observations in mammalian cells 29,32 . We also replaced the Cassava vein mosaic virus promoter (pVmv) that drives the expression of the mCherry reporter with the stronger Cauliflower mosaic virus 35S promoter (p35s) (Fig. 4A) . Additionally, to generate a homogeneous and precise 3′ RNA end, we incorporated the hepatitis delta virus (HDV) ribozyme 52 downstream of the payload. This ribozyme enables autocatalytic cleavage, resulting in precise processing of the RNA transcript and retention of the R4 sequence at the 3′ terminus of the R2Tg retrotransposition reporter. These modifications yielded two distinct RNA template designs, termed the R4 minimal payload and the R4-HDV minimal payload, respectively (Fig. 4A) . We evaluated the integration efficiencies of R2Tg editor system with the R4 and R4-HDV payload designs in N. benthamiana leaves. Constructs were delivered as a 1-plasmid or a 2-plasmid system separating protein and payload cassettes (Fig. 4A) . Confocal microscopy performed 7 days post-infiltration revealed consistent mCherry signals, indicative of successful R2-mediated integration for both payload designs when paired with the wild-type R2Tg protein or an intronized version of the protein (Fig. 4B, Supplementary Fig. 4A for more replicates). Quantification of the percentage of cells expressing mCherry from confocal images revealed that the new designs have substantially improved the integration efficiencies. In the case of R4 minimal payload design, efficiency of R2Tg-1 increased from 0.5% to 10.4%, R2Tg-2 increased from 0.85% to 12.2%, and R2Tg-2i increased from 1.6% to 15.1%. In the case of R4-HDV minimal payload design, efficiency of R2Tg-1 reached up to 10.7%, R2Tg-1i up to 12.6%, R2Tg-2 up to 20.3%, and R2Tg-2i up to 23.6%. Notably, R2Tg-mediated integration was markedly increased in samples expressing the R4-HDV payload compared to the R4 payload (Fig. 4B, C; Supplementary Fig. 4A) . In addition, intronized R2Tg variants delivered via either 1- or 2-plasmid strategies consistently generated the strongest mCherry signals, while the R2Tg RT-dead mutant 29 did not produce any detectable fluorescence neither the payload without the R2 protein (Fig. 4B, C; Supplementary Fig. 4A) . Target-junction PCR assays corroborated imaging results at the molecular level, producing amplicons of expected sizes (425 bp for 5′ junction, 454 bp for 3′ junction) in samples harboring functional R4-HDV R2Tg minimal payload variant (Fig. 4D) . Both 5′ and 3′ junction PCR products were detected for intronized and wild-type R2Tg constructs paired with the R4-HDV minimal payload, whereas RT-dead controls lacked the correct-sized band. Sanger sequencing junction amplicons from R4-HDV payload paired with the intronized R2Tg (R2Tg-1i) revealed precise junction sequences, perfectly matching the expected integration architecture at both 5′ and 3′ ends (Fig. 4E) . These results validate the fidelity of the integration events generated by the optimized R2Tg editor system. To rigorously quantify integration efficiency, we implemented a droplet digital PCR (ddPCR) assay targeting the unique 3′ and 5’ integration junctions and normalized it against the endogenous NbEF1α locus (Fig. 4F; Supplementary Fig. 4B) . ddPCR analysis across 3 biological replicates showed that R2Tg editor comprising of the intronized R2Tg with the R4-HDV minimal payload in the 1-plasmid delivery format (R2Tg-1i) significantly exceeds efficiencies observed with all other R2 editor configuration tested in N. benthamiana , reaching an average of approximately 1.2 integrated 3′ junction copy per genome, while ~80-90% of those copies show full-length integration events (Fig. 4G; Supplementary Fig. 4C, D) . ddPCR at both junction sites further revealed that, across R2Tg editors tested in N. benthamiana , the proportion of integrated copies containing both 3′ and 5′ junctions (full-length integration events) ranged from 73% to 87% (Supplementary Fig. 4D ). Deep next-generation sequencing (NGS) of 3′ junction amplicons revealed 72.14% of reads harboring integrated junctions exclusively in samples expressing the R4-HDV payload in combination with intronized R2Tg (R2Tg-1i), while samples containing the payload alone lacked any integration-specific reads (Supplementary Fig. 4E-G) . Alignment of sequencing reads confirmed high-fidelity, precise integration accompanied by a minority of indel variants comprising non-templated adenine additions and subtle shifts in endonuclease cleavage, occurring one to two nucleotides upstream of the canonical nick position, consistent with the established activity profile of R2 endonuclease (Supplementary Fig. 4H) . Collectively, these findings demonstrate that compact R2Tg payloads designs, with minimized homology arm length and 3′ ribozyme substantially enhance R2Tg-mediated targeted integration efficiencies at the 25S rDNA site in plants and maintains precise, targeted DNA insertion. To evaluate whether the optimized R2Tg system could mediate targeted DNA insertion at the 25s rDNA target site in the non-model crop Solanum lycopersicum (tomato), we tested the minimal R4-HDV mCherry payload delivered in either a one-plasmid or two-plasmid format together with the intronized R2Tg using a transient transformation method developed in our lab 53 (Supplementary Fig. 5A) . Confocal microscopy of seedling leaves transformed with these constructs revealed strong mCherry fluorescence in nuclei of cells co-expressing the intronized R2Tg, consistent with R2-mediated retrotransposition, while mTagBFP2 served as a transformation control (Supplementary Fig. 5B) . PCR assays specifically targeting integrated junctions at the molecular level, further validated precise insertion of the mCherry payload at the 25S rDNA locus with the expected 425 bp and 454 bp products detected for the 5′ and 3′ junctions, respectively, while no bands were obtained when only template is delivered without the R2 protein (Supplementary Fig. 5C-E) . These results demonstrate that the R2Tg editor achieves efficient targeted integration in S. lycopersicum , establishing its functionality beyond model systems and highlighting its potential as a broadly applicable technology for precise genome engineering in crop species. Next, we assessed whether the optimized R2Tg editor could support efficient targeted insertion of larger DNA sequences into the 25s rDNA loci of N. benthamiana . For this, we constructed a 5 kb R4-HDV RUBY reporter cassette 54 (Fig. 5A) . The RUBY reporter encoded a polycistronic open reading frame linking 3 enzymes: cytochrome P450 CYP76AD1 that oxidizes tyrosine to L-DOPA and cyclo-DOPA; L-DOPA 4,5-dioxygenase (DODA) that converts L-DOPA to betalamic acid; and a glucosyltransferase that glycosylates betanidin to produce the red betalain pigment. The three enzymes were separated by 2A peptides to allow coordinated expression from a single promoter 54,55 . The R2Tg RUBY reporter was interrupted by a plant intron in the sense orientation, similar to the mCherry payload, but with the intron placed within the glucosyltransferase gene so that betalain production occurs only upon successful R2-mediated integration of the full 5 kb polycistronic payload. The system was delivered either as a single plasmid encoding both the intronized R2Tg and the RUBY payload, or as two separate plasmids carrying R2Tg and the payload separately (Fig. 5A) . Expression of the RUBY payload with intronized R2Tg in N. benthamiana leaves produced visible betalain pigmentation within infiltrated areas, with stronger pigmentation observed for R2Tg-1i compared to R2Tg-2i, while leaves infiltrated with the RUBY payload alone, without the R2 protein, did not show pigmentation (Supplementary Fig. 6B-D) . Target-junction PCR using the genomic DNA extracted from infiltrated N. benthamiana leaves confirmed successful integration, producing the expected 5’ (425 bp) and 3’ (454 bp) amplicons (Fig. 5B) . To quantify insertion efficiencies, we applied a droplet digital PCR (ddPCR) assay specific to the integrated 5′ and 3′ junctions (Fig. 5C) . ddPCR measurements showed that the one-plasmid system achieved the highest efficiencies, with an average of one copy per genome at the 5′ junction and 1.5 copies per genome at the 3′ junction, significantly outperforming the two-plasmid format which had 0.26 copies per genome at the 5′ junction and 0.33 copies per genome at the 3′ junction (Fig. 5D, E) . These findings validate that the intronized R2Tg system efficiently inserts multi-kilobase metabolic pathways into the 25S rDNA safe harbor of N. benthamiana . Discussion In this work, we established targeted DNA insertion in plants using the Taeniopygia guttata R2 non-LTR retroelement protein (R2Tg editor), which natively mediates site-specific integration through target-primed reverse transcription (TPRT) 27 . The R2 ribonucleoprotein complex achieves precise, multi-kilobase DNA integration by tightly coupling sequence-specific 25S rDNA recognition, bottom-strand nicking, primer hand-off, and reverse transcription of an RNA donor into cDNA 28 . Unlike CRISPR-Cas9 homology-directed repair (HDR), which depends on host DNA repair machinery to resolve a double-strand break using a donor DNA template, often resulting in low insertion efficiencies, the R2 mechanism bypasses the need for DSB induction entirely, avoiding competing end-joining pathways and the risk of large deletions or chromosomal rearrangements 15 . R2 editors use RNA payloads in a similar manner to prime editing. In plants, prime editors are most effective for base substitutions and short insertions, and efficiency declines as insertion length and pegRNA design complexity increases. However, unlike prime editors that are constrained by insertion length, and guide RNA design complexity, R2 editors circumvent these constraints by directly synthesizing long cDNAs at the genomic target site during integration, leveraging the high processivity of their reverse transcriptase domains 27,28 . It remains to be demonstrated whether protein engineering can sufficiently increase processivity to support TPRT of even much larger payloads than the 5 kb demonstrated here. Another unique property of R2 editors is their targeting to the high-copy 25S rDNA locus, which embeds integrated genes in a specialized chromatin environment linked to stable transgene expression and minimal risk of disrupting essential host genes 56–58 . Here, we harnessed these unique capabilities to achieve targeted insertion in Nicotiana benthamiana and Arabidopsis thaliana and Solanum lycopersicum . We optimized efficiencies through engineered RNA payload architectures and improved R2 protein expression systems and validated high-efficiency targeted insertion via molecular detection of integrated junctions and functional expression of fluorescent reporters. Our results illuminated key determinants of integration efficiency in plants. The highest measured mCherry positive cell counts were achieved with the intronized R2Tg protein and R4-HDV minimal payload while using the 2-plasmid delivery approach. In contrast, the highest integrated 3’ junction copy numbers detected, approaching one copy per genome, was obtained with an intronized R2Tg and R4-HDV minimal payload in a 1-plasmid delivery. Additionally, this R2Tg editor configuration achieved targeted insertion in Solanum lycopersicum expanding the tool’s capabilities to crop engineering. Notably, the R2Tg editor system also enabled efficient integration of a 5 kb metabolic pathway cassette, into the 25S rDNA safe-harbor locus, demonstrating that R2 retrotransposons can efficiently install complex metabolic pathways at a targeted genomic site. The divergence in mCherry signals and ddPCR detected copy numbers per genome may reflect differences in RNA payload processing and the completeness of integrated products, as R2-mediated integration can yield incomplete insertion events 29,30,32 , 59 and the fluorescence readout from the mCherry retrotransposition reporter requires successful intron splicing before integration. Incomplete integration events or unspliced integration events would not yield detectable mCherry signal but molecular-level detection copy number allows us to obtain a comprehensive picture of integration events at the 25S rDNA site. These observations are consistent with the possibility that R2Tg silencing and transcriptional interference in the all-in-one setup may affect efficient splicing and integration of the mCherry reporter. Intronization of R2Tg and separation of protein and payload cassettes may alleviate these constraints, improve the expression and splicing of payloads prior to integration at the 25S rDNA site to yield detectable mCherry reporter expression. We observed improved mCherry signal after heat shock treatment of N. benthamiana plants, expressing the R2 editor system after Agrobacterium -mediated leaf infiltrated. This improvement in integration efficiency may be a result enzymatic kinetics or ribonucleoprotein (RNP) interactions required for integration. Other temperature-sensitive host processes may also contribute to this effect. In particular, heat stress has been shown to activate endogenous transposable elements in plants through global chromatin relaxation, reduction of DNA methylation, and suppression of RNA interference 46–48,50,60 . Such mechanisms may alleviate epigenetic repression of R2 editor components, thereby facilitating higher integration efficiencies. Disentangling these direct catalytic effects from host epigenetic changes will be important for understanding the mechanism at the molecular level. In future studies, it will be important to assess whether transcriptional regulation of R2 protein and RNA template expression impose plant-specific constraints on R2 integration efficiencies not seen mammalian systems. To complement our work further, comprehensive genome-wide mapping is required to confirm the absence of off-target integrations, and long-term studies of stability and heritability will be crucial for translating this platform into breeding and crop improvement pipelines. Although in this study we delivered the R2 system using Agrobacterium tumefaciens , which introduces DNA randomly into the genome, future efforts will focus on direct delivery of the R2 system as a ribonucleoprotein complex using nanoparticles or biolistics. This strategy would eliminate unintended T-DNA insertions, ensuring that only precise edits at the target rDNA locus are recovered. Such delivery advances will not only improve editing fidelity but also accelerate generation of clean, marker-free lines, reduce the need for laborious screening, and broaden applicability across plant species that are recalcitrant to Agrobacterium . Together, these improvements could establish R2-based editors as a practical and widely deployable platform for precise, large-fragment DNA insertion in plants, an advancement with far-reaching implications for both fundamental genome studies and crop engineering. Together, these findings establish R2 retrotransposons as a distinctive and versatile addition to the plant genome engineering toolkit. By leveraging the precision of a native retroelement and augmenting it through improved protein expression platforms and engineered RNA payloads designs, this technology enables efficient, DSB-free integration of gene-sized cassettes into a defined genomic safe harbor locus in plants. With continued refinement, R2-based systems have the potential to complement and, in certain applications, surpass existing technologies for precise and targeted introduction of complex genetic traits into plant genomes. Methods Plasmid construction Coding sequences for 4 R2 non-LTR retrotransposases, R2Tg ( Taeniopygia guttata ), R2Bm ( Bombyx mori ), R2Tg-opt (rationally engineered T. guttata ) and R2Za ( Zonotrichia albicollis ), were codon-optimized for Nicotiana benthamiana and Arabidopsis thaliana using GenSmart Codon Optimization (GenScript) and synthesized as double-stranded DNA fragments (Twist Bioscience). Each open reading frame contained an N-terminus bipartite SV40 nuclear localization signal (KRTADGSEFESPKKKRKV). 3 additional versions of the R2Tg protein were generated: i) R2Tg-intronic, which contains 2 A. thaliana introns 45 ; ii) R2Tg RT-dead, carrying D660A and D661A substitutions in the reverse-transcriptase domain; and iii) R2Tg EN-dead, carrying D1057A and D1070A substitutions in the endonuclease domain 29 . The retrotransposition reporter followed a classic GFP-intron design 42,43 . A Cauliflower mosaic virus (CaMV) 35S promoter and an A. thaliana HSP18.2 terminator flank a reverse-oriented nested cassette that contains the Cassava vein mosaic virus (CsVMV) promoter, mCherry open reading frame interrupted by either the Solanum tuberosum ST-LS1 intron or an A. thaliana intron, and a CaMV t35S terminator 44 , 45 . Plasmids carrying R2 protein expression cassettes and donor DNA templates were combined with Loop Assembly using BsaI- and SapI-mediated Golden Gate reactions (NEB catalog numbers R0569S and E1601L respectively) and cloned into pCAMBIA backbones 61 . All plasmids were propagated in NEB Turbo Escherichia coli competent cells (NEB catalog no. C2984I) and sequence-verified by Plasmidsaurus. A geminiviral replicon based on the Bean yellow dwarf virus (BeYDV) was used to transiently overexpress either the R2 proteins or the complete R2 editing system 41 . Constructs generated in this work can be found in Supplementary Table 1. Plant growth and transformation Nicotiana benthamiana and Arabidopsis thaliana (Col-0) plants were grown in a Conviron controlled-environment chamber set to 22 °C during the day, 20 °C during the night with a 16 h light / 8 h dark photoperiod and 55 % relative humidity. 4-weeks old N. benthamiana plants were used for leaf infiltrations for transient experiments. 4-weeks old A. thaliana plant leaves were harvested for protoplast isolation. Agrobacterium tumefaciens strain GV3101 (GOLDBIO catalogue no. CC-125) carrying the binary R2 construct, the p19 silencing-suppressor plasmid, and the pSOUP helper plasmid was cultured overnight at 28°C in 2YT medium (Invitrogen catalogue no. 12-780-052) supplemented with the appropriate antibiotics. Cells were collected and resuspended in induction media (10 mM MES pH 5.5, 10 mM MgCl 2 , 200 mM Acetosyringone (PhytoTech Labs catalogue no. A104-5G) in DMSO), at OD 600 0.4. Suspensions were allowed to shake at 70 rpm for 4 hours at room temperature. The Agrobacterium suspension was infiltrated into the abaxial surface of fully expanded leaves using a 1 mL needleless syringe until the tissue was saturated. 3 leaves per plant were treated. Plants were returned to the growth chamber immediately and sampled 3 to 7 days post infiltration depending on the experiment. Transient Transformation of Solanum lycopersicum (cv. Micro-Tom) Transient transformation was performed using a protocol adapted from the VAST (Vacuum and Sonication-Assisted Transient expression) method 53 . Seeds of S. lycopersicum (cv. Micro-Tom) were surface-sterilized with 33% (v/v) bleach containing 0.1% Tween-20 for 6 min, rinsed 4–5 times with sterile water, and germinated on ½ MS agar at 22 °C under a 16 h light/8 h dark photoperiod until cotyledons were fully expanded (~9 days). Agrobacterium tumefaciens strain GV3101 (GOLDBIO, CC-125) carrying the binary R2 construct, p19 silencing suppressor, and pSOUP helper plasmid was cultured in 2X-YT medium with antibiotics at 28 °C, 200 rpm, 2 days prior to co-cultivation. Cultures were pelleted (8,000 g, 5 min), resuspended in AB-MES induction medium (200 µM Acetosyringone, no antibiotics) to OD₆₀₀ = 0.3, and incubated overnight at 28 °C, 200 rpm. On the day of co-cultivation, cultures were pelleted again and resuspended in co-culture medium (1:1 AB-MES and ½ MS supplemented with 200 µM Acetosyringone) and adjusted to OD₆₀₀ = 0.3. For infection, 7–8 seedlings were immersed in 8 mL of Agrobacterium suspension in 12 mL tubes, sonicated for 30 s (Branson CPX-952-116R), and subjected to three cycles of vacuum infiltration (5 min vacuum followed by rapid release per cycle) with gentle mixing between cycles. Seedlings were then transferred to six-well plates containing 4 mL of fresh Agrobacterium suspension per well and co-cultivated for 2 days at 22 °C under a 16 h light/8 h dark cycle. Following co-cultivation, seedlings were washed three times with sterile water and transferred to ½ MS agar plates containing 100 µM timentin to suppress bacterial growth. Plants were maintained under standard growth conditions, and transient expression was assessed 3 days post-infection by confocal microscopy. Confocal microscopy Leaf discs ~5 mm in diameter were collected from infiltrated Nicotiana benthamiana leaves using a sterile hole puncher, mounted in distilled water, and imaged immediately on a Leica Stellaris 8 confocal laser-scanning microscope (HC PL APO 20×/0.75 NA water-immersion objective) housed in the Biological Imaging Facility (BIF) at Caltech. The blue fluorescent protein (BFP) reporter was excited at 405 nm and its emission collected from 430 nm to 470 nm, while mCherry was excited at 561 nm and its emission collected from 580 nm to 630 nm. A single optical section was acquired for each field with the pinhole set to 1 Airy unit, using identical laser power, detector gain, and offset for all samples within an experiment. Images were exported from Leica LAS X v4.3 as 16-bit TIFF files. A 100 µm scale bar was added to exported images using ImageJ Fiji. Protein expression and detection R2Tg, R2Bm, R2Tg-opt, and R2Za coding sequences were fused to an N-terminal YPet tag for detection and confirmation of nuclear localization via confocal microscopy. Complementarily, R2 proteins were fused to an N-terminal HiBiT peptide tag for higher throughput luminescence detection using the Nano-Glo® HiBiT Lytic Detection System. 3 days post infiltration (dpi), 3 infiltrated Nicotiana benthamiana leaves were harvested, flash-frozen in liquid nitrogen, and ground to a fine powder before extraction in lysis buffer (20 mM Tris-HCl pH 7.4 (ThermoFisher Scientific catalogue no. 15567027), 25 % glycerol (Sigma-Aldrich catalogue no. G5516), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 2 mM EDTA (Sigma-Aldrich catalogue no. E9884), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 250 mM sucrose (ThermoFisher scientific catalogue no. J65148.A1), 0.1 % PMSF (ThermoFisher Scientific catalogue no. 36978)). After centrifugation at 1500 × g for 10 min at 4 °C to pellet nuclei, the pellet was washed up to 5 times with wash buffer (20 mM Tris-HCl pH 7.4 (ThermoFisher Scientific catalogue no. 15567027), 25 % glycerol (Sigma-Aldrich catalogue no. G5516), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 0.2 % Triton X-100 (Sigma-Aldrich catalogue no. S5886)) and lysed in high-salt buffer (20 mM HEPES-KOH pH 7.9 (Sigma-Aldrich catalogue no. H3375), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 100 mM NaCl (Sigma-Aldrich catalogue no. S5886), 20 % glycerol (Sigma-Aldrich catalogue no. G5516), 0.2 mM EDTA (Sigma-Aldrich catalogue no. E9884), 0.5 mM DTT (GOLDBIO catalogue no. DTT), one cOmplete Mini EDTA-free protease-inhibitor tablet (Sigma-Aldrich catalogue no. 11836170001) per 10 mL). For HiBiT detection, 30 µL of nuclear lysate was combined with 30 µL Nano-Glo HiBiT Lytic reagent (Promega catalogue no. N3030) containing LgBiT and furimazine, and luminescence was recorded on a Tecan Spark plate reader with a 1000 millisecond integration time. Isolation and transfection of Nicotiana benthamiana and Arabidopsis thaliana protoplasts Mesophyll protoplasts were isolated from fully expanded leaves of 4-week-old N. benthamiana or A. thaliana (Col-0) following Yoo et al. (2007) 62 with minor modifications. On the day of the experiment, a fresh enzyme solution (20 mL) containing 20 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933), 0.4 M mannitol (Millipore Sigma catalogue no. 443907), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 1.5% (w⁄v) cellulase onozuka R-10 (Yakult Pharmaceutical IND. CO., LTD) and 0.4% (w⁄v) macerozyme R-10 (Yakult Pharmaceutical IND. CO., LTD) was prepared; the MES–mannitol–KCl base was pre-heated to 70°C for 4 min before the enzymes were dissolved. After cooling, 10 mM CaCl₂ (Sigma-Aldrich catalogue no. C7902) and 0.1% (w⁄v) BSA (Fisher bioreagents catalogue no. BP9703) were added, and the solution was passed through a 0.45 µm syringe filter into a Petri dish. N. benthamiana leaf strips or A. thaliana full leaves were incubated in this solution without agitation for 3 h at room temperature in the dark, with gentle swirling once per hour. The digest was diluted 1:1 with W5 solution (154 mM NaCl (Sigma-Aldrich catalogue no. S5886), 125 mM CaCl₂ (Sigma-Aldrich catalogue no. C7902), 5 mM KCl (Sigma-Aldrich catalogue no. P5405), 2 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)), passed through a 70 µm nylon mesh (Fisherbrand catalogue no. 22363548), and the flow-through was centrifuged at 200 × g for 2 min. The pellet was gently resuspended in 5 mL W5, chilled on ice for 30 min, then the supernatant was removed and the protoplasts were adjusted to 5 × 10⁵ cells mL⁻¹ in MMG solution (0.4 M mannitol (Millipore Sigma catalogue no. 443907), 15 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 4 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)). For PEG-mediated transfection a 40 % PEG 4000 solution (Sigma-Aldrich catalogue no. 81240) was freshly prepared (4 g PEG 4000, 3 mL H₂O, 2.5 mL 0.8 M mannitol (Millipore Sigma catalogue no. 443907), 1 mL 1 M CaCl₂ (Sigma-Aldrich catalogue no. C7902)). In round-bottom Eppendorf tubes, 20 µg plasmid DNA (≤ 20 µL) was combined with 100 µL protoplast suspension, mixed gently, and 120 µL PEG solution was added. After 15 min at room temperature, 1 mL W5 was added and the mixture was centrifuged (50 × g, 2 min); the wash was repeated twice. The pellet was resuspended in WI solution (0.5 M mannitol (Millipore Sigma catalogue no. 443907), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 4 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)), transferred to BSA-blocked well plates (0.1 % BSA (Fisher bioreagents catalogue no. BP9703)), and incubated at room temperature in the dark for 24-72 h before downstream analyses.After incubation in the dark at room temperature for 24 to 72 h, transformed protoplasts were used for confocal imaging and/or flow cytometry. In a modified protocol, protoplasts were extracted from 5-week-old Nicotiana benthamiana leaves 7 days after Agrobacterium infiltration and analyzed directly by flow cytometry. Flow cytometry analysis Protoplasts were analyzed on a CytoFLEX flow cytometer (Beckman Coulter) provided by the Caltech Flow Cytometry and Cell Sorting Facility. Live protoplast gates were defined using fluorescein diacetate (FDA) staining (Fisher scientific catalogue no. F1303), detected in the FITC channel. BFP was excited with a 405 nm laser and detected in the 450/45 nm channel, while mCherry was excited with a 561 nm laser and detected in the 610/20 nm channel. Data was exported as FCS files and analyzed using FloReada for gating, quantification, and visualization. Amplicon PCR and sanger sequencing Genomic DNA (gDNA) was extracted from Nicotiana benthamiana leaf tissue infiltrated with R2 constructs using the DNeasy Plant Mini Kit (Qiagen catalogue no. 69106) according to the manufacturer’s protocol. For 3′-junction analysis, 25 µL PCRs were assembled with 1× Phusion High-Fidelity Master Mix (NEB catalogue no. M0531S), 0.5 µM each primer 2F and 2R, and 20 ng gDNA. Cycling conditions: 98 °C for 30 s; 35 cycles of 98 °C for 10 s, 60 °C for 20 s, 72 °C for 30 s; final extension 72 °C for 2 min. The expected 450 bp amplicons were verified on a 1 % agarose gel, purified with the QIAquick Gel Extraction Kit (Qiagen catalogue no. 28704) and submitted to Laragen Inc. for Sanger sequencing. Chromatograms were aligned to reference sequences in SnapGene v6.2 to confirm precise junction formation. Next generation sequencing (NGS) The 450 bp 3′-junction PCR products (primer pair 2F/2R) were gel-purified with the QIAquick Gel Extraction Kit (Qiagen catalogue no. 28704) and eluted in 35 µL nuclease-free water. DNA concentration was measured on a NanoDrop 2000 spectrophotometer, and each amplicon was adjusted to 50 ng/µL, sealed in Eppendorf microcentrifuge tubes, and shipped overnight to the Massachusetts General Hospital CCIB DNA Core (Harvard) for Illumina sequencing. The core performed NGS and provided (i) de novo-assembled contigs representing variants present at > 1 % abundance and (ii) the complete demultiplexed FASTQ files. In Galaxy, paired-end reads were processed with DADA2: adapters were trimmed, reads were quality-filtered, and overlapping pairs were merged. Merged reads were aligned to a custom reference comprising the payload integrated at the 25S rDNA locus using Bowtie2 v2.5.0 to capture low-frequency variants (< 1 %). The subset of sequence variants mapping perfectly to the insertion junction were imported into Geneious Prime 2024.1, aligned to the same reference and the multiple alignment was exported as FASTA. Alignment visualization was carried out in Google Collab with Biopython 1.81, pandas 2.2.2, and matplotlib 3.7.3; the resulting grid plot of base-by-base variation was exported as high-resolution PNG files for publication. Droplet digital PCR (ddPCR) for detecting targeted insertion efficiency Genomic DNA was extracted from 100 mg of Nicotiana benthamiana leaf tissue collected from the Agrobacterium infiltration site 7 days after infiltration using the DNeasy Plant Mini Kit (Qiagen). Genomic DNA was predigested with FastDigest KpnI (Thermo Fisher Scientific, catalog no. FD0524) according to the manufacturer’s protocol. Duplex 22 µl ddPCR reactions were prepared by mixing 11 µl of ddPCR Supermix for Probes (no dUTP; Bio-Rad, catalog no. 1863024), 900 nM forward and reverse primers for the target and reference loci, 250 nM FAM-labeled probe targeting the NbEF1α reference gene, 250 nM HEX-labeled probe targeting the integrated R2 junction, and 50 ng of predigested gDNA. Oligonucleotide sequences were synthesized by Integrated DNA Technologies (IDT) and are listed in Supplementary Table 2. Reaction mixtures were loaded into a DG8 cartridge (Bio-Rad, catalog no. 1864007) along with 70 µl of droplet generation oil (Bio-Rad, catalog no. 1863005), and droplets were generated using a Bio-Rad QX200 Droplet Generator. Following droplet generation, 40 µl of emulsion was transferred to a 96-well plate, heat-sealed with pierceable foil, and thermal-cycled under the manufacturer’s conditions with an annealing/extension temperature of 55. o C for 3’ junction and 60 o C for 5’ junction. Droplets were read using a Bio-Rad QX200 Droplet Reader. Data were analyzed using QX Manager software (Bio-Rad). All Bio-Rad ddPCR equipment was provided by the CLARITY, Optogenetics and Vector Engineering Research (CLOVER) Center at Caltech. Insertion efficiency was calculated as the ratio of HEX-labeled target copy number to the FAM-labeled reference gene ( NbEF1α ) copy number assuming 4 copies of the target sequence. Declarations Data Availability Statement All data are included either in the manuscript or supplementary files. Conflict of Interest Statement Authors declare no conflict of interest. Author Contributions Conceptualization: KTM, GSD Methodology: KTM, YW, TMO, AS, EL, GSD Data Acquisition: KTM, YW, TMO, AS, EL Supervision: KTM, TMO, GSD Writing: KTM, GSD Funding Acquisition: GSD Acknowledgements We thank Professor Kathleen Collins (University of California, Berkeley) for valuable discussions and insights on experimental design. Imaging was performed in the Biological Imaging Facility, with the support of the Caltech Beckman Institute and the Arnold and Mabel Beckman Foundation. We acknowledge the Beckman Institute CLOVER Center for providing access to droplet digital PCR instrumentation and support with experiments. Flow cytometry data used in this manuscript was generated in the Caltech Flow Cytometry & Cell Sorting Facility (Research Resource Identifier number is RRID: SCR_025087). Schematics were created with BioRender.com. Funding This work was supported by the Caltech startup funds, Caltech Space-Health Innovation Fund, Henry Luce Foundation, and Shurl and Kay Curci Foundation. References Dasgupta, S. & Robinson, E. J. Z. Attributing changes in food insecurity to a changing climate. Sci. Rep. 12 , 4709 (2022). Bradbury, A. et al. Integrating genome editing with omics, artificial intelligence, and advanced farming technologies to increase crop productivity. Plant Commun. 6 , 101386 (2025). Scheben, A., Yuan, Y. & Edwards, D. Advances in genomics for adapting crops to climate change. Curr. Plant Biol. 6 , 2–10 (2016). Tzfira, T. & Citovsky, V. Agrobacterium-mediated genetic transformation of plants: biology and biotechnology. Curr. Opin. Biotechnol. 17 , 147–154 (2006). Kim, S.-I. & and Gelvin, S. B. Genome-wide analysis of Agrobacterium T-DNA integration sites in the Arabidopsis genome generated under non-selective conditions. Plant J. 51 , 779–791 (2007). Rajeevkumar, S., Anunanthini, P. & Sathishkumar, R. Epigenetic silencing in transgenic plants. Front. Plant Sci. 6 , 693 (2015). Gupta, M. & Ram, R. Development of Genetically Modified Agronomic Crops. in The GMO Handbook: Genetically Modified Animals, Microbes, and Plants in Biotechnology (ed. Parekh, S. R.) 219–241 (Humana Press, Totowa, NJ, 2004). doi:10.1007/978-1-59259-801-4_8. Dong, O. X. & Ronald, P. C. Targeted DNA insertion in plants. Proc. Natl. Acad. Sci. 118 , e2004834117 (2021). Khan, A. & Lister, R. Synthetic gene circuits in plants: recent advances and challenges. Quant. Plant Biol. 6 , e6 (2025). Chen, W. & Ow, D. W. Precise, flexible and affordable gene stacking for crop improvement. Bioengineered 8 , 451–456 (2017). Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337 , 816–821 (2012). Cong, L. et al. Multiplex Genome Engineering Using CRISPR/Cas Systems. Science 339 , 819–823 (2013). Dong, O. X. et al. Marker-free carotenoid-enriched rice generated through targeted gene insertion using CRISPR-Cas9. Nat. Commun. 11 , 1178 (2020). Lu, Y. et al. Targeted, efficient sequence insertion and replacement in rice. Nat. Biotechnol. 38 , 1402–1407 (2020). Kosicki, M., Tomberg, K. & Bradley, A. Repair of double-strand breaks induced by CRISPR–Cas9 leads to large deletions and complex rearrangements. Nat. Biotechnol. 36 , 765–771 (2018). Zong, Y. et al. An engineered prime editor with enhanced editing efficiency in plants. Nat. Biotechnol. 40 , 1394–1402 (2022). Anzalone, A. V. et al. Programmable deletion, replacement, integration and inversion of large DNA sequences with twin prime editing. Nat. Biotechnol. 40 , 731–740 (2022). Jiang, Y.-Y. et al. Prime editing efficiently generates W542L and S621I double mutations in two ALS genes in maize. Genome Biol. 21 , 257 (2020). Sun, C. et al. Precise integration of large DNA sequences in plant genomes using PrimeRoot editors. Nat. Biotechnol. 42 , 316–327 (2024). Wells, J. N. & Feschotte, C. A Field Guide to Eukaryotic Transposable Elements. Annu. Rev. Genet. 54 , 539–561 (2020). Hickman, A. B. & Dyda, F. DNA Transposition at Work. Chem. Rev. 116 , 12758–12784 (2016). Gelsinger, D. R. et al. Bacterial genome engineering using CRISPR-associated transposases. Nat. Protoc. 19 , 752–790 (2024). Lampe, G. D. et al. Targeted DNA integration in human cells without double-strand breaks using CRISPR-associated transposases. Nat. Biotechnol. (2023) doi:10.1038/s41587-023-01748-1. George, J. T. et al. Mechanism of target site selection by type V-K CRISPR-associated transposases. Science 382 , eadj8543 (2023). Liu, P. et al. Transposase-assisted target-site integration for efficient plant genome engineering. Nature 631 , 593–600 (2024). Fujiwara, H. Site-specific non-LTR retrotransposons. Microbiol. Spectr. 3 , MDNA3-0001–2014 (2015). Wilkinson, M. E., Frangieh, C. J., Macrae, R. K. & Zhang, F. Structure of the R2 non-LTR retrotransposon initiating target-primed reverse transcription. Science 380 , 301–308 (2023). Thawani, A. et al. Structures of vertebrate R2 retrotransposon complexes during target-primed reverse transcription and after second-strand nicking. Sci. Adv. 11 , eadu5533 (2025). Zhang, X. et al. Harnessing eukaryotic retroelement proteins for transgene insertion into human safe-harbor loci. Nat. Biotechnol. (2024) doi:10.1038/s41587-024-02137-y. Chen, Y. et al. All-RNA-mediated targeted gene integration in mammalian cells with rationally engineered R2 retrotransposons. Cell 187 , 4674-4689.e18 (2024). Fell, C. W. et al. Reprogramming site-specific retrotransposon activity to new DNA sites. Nature 1–10 (2025) doi:10.1038/s41586-025-08877-4. Edmonds, K. K. et al. Structure and biochemistry-guided engineering of an all-RNA system for DNA insertion with R2 retrotransposons. Nat. Commun. 16 , 6079 (2025). Fultz, D., McKinlay, A., Enganti, R. & Pikaard, C. S. Sequence and epigenetic landscapes of active and silent nucleolus organizer regions in Arabidopsis . Sci. Adv. 9 , eadj4509 (2023). Chen, W. et al. The complete genome assembly of Nicotiana benthamiana reveals the genetic and epigenetic landscape of centromeres. Nat. Plants 10 , 1928–1943 (2024). Shirasawa, K. & Ariizumi, T. Near-complete genome assembly of tomato (Solanum lycopersicum) cultivar Micro-Tom. Plant Biotechnol. advpub , (2024). Zhou, X. et al. Tandem Repeat Detection Using Next-Generation Mapping and its Application in Detecting rDNA Copy Number Variation in Plant and Human Genomes. Tulpová, Z. et al. Fine structure and transcription dynamics of bread wheat ribosomal DNA loci deciphered by a multi-omics approach. Plant Genome 15 , e20191 (2022). Hori, Y., Engel, C. & Kobayashi, T. Regulation of ribosomal RNA gene copy number, transcription and nucleolus organization in eukaryotes. Nat. Rev. Mol. Cell Biol. 24 , 414–429 (2023). Layat, E., Sáez-Vásquez, J. & Tourmente, S. Regulation of Pol I-Transcribed 45S rDNA and Pol III-Transcribed 5S rDNA in Arabidopsis. Plant Cell Physiol. 53 , 267–276 (2012). Christensen, S. M. & Eickbush, T. H. R2 Target-Primed Reverse Transcription: Ordered Cleavage and Polymerization Steps by Protein Subunits Asymmetrically Bound to the Target DNA. Mol. Cell. Biol. 25 , 6617–6628 (2005). Baltes, N. J., Gil-Humanes, J., Cermak, T., Atkins, P. A. & Voytas, D. F. DNA Replicons for Plant Genome Engineering. Plant Cell 26 , 151–163 (2014). Ostertag, E. M., Luning Prak, E. T., DeBerardinis, R. J., Moran, J. V. & Kazazian, H. H. Determination of L1 retrotransposition kinetics in cultured cells. Nucleic Acids Res. 28 , 1418–1423 (2000). Tao, J., Wang, Q., Mendez-Dorantes, C., Burns, K. H. & Chiarle, R. Frequency and mechanisms of LINE-1 retrotransposon insertions at CRISPR/Cas9 sites. Nat. Commun. 13 , 3685 (2022). Johansen, I. E. & Lund, O. S. Insertion of Introns: A Strategy to Facilitate Assembly of Infectious Full Length Clones. in Plant Virology Protocols (eds. Foster, G. D., Johansen, I. E., Hong, Y. & Nagy, P. D.) vol. 451 535–544 (Humana Press, Totowa, NJ, 2008). Grützner, R. et al. High-efficiency genome editing in plants mediated by a Cas9 gene containing multiple introns. Plant Commun. 2 , 100135 (2021). Bond, D. M. & Baulcombe, D. C. Epigenetic transitions leading to heritable, RNA-mediated de novo silencing in Arabidopsis thaliana . Proc. Natl. Acad. Sci. 112 , 917–922 (2015). Sigman, M. J. & Slotkin, R. K. The First Rule of Plant Transposable Element Silencing: Location, Location, Location. Plant Cell 28 , 304–313 (2016). Nuthikattu, S. et al. The Initiation of Epigenetic Silencing of Active Transposable Elements Is Triggered by RDR6 and 21-22 Nucleotide Small Interfering RNAs1[W][OA]. Plant Physiol. 162 , 116–131 (2013). Burgess, D., Li, H., Zhao, M., Kim, S. Y. & Lisch, D. Silencing of Mutator Elements in Maize Involves Distinct Populations of Small RNAs and Distinct Patterns of DNA Methylation. Genetics 215 , 379–391 (2020). Hedtke, B. & Grimm, B. Silencing of a plant gene by transcriptional interference. Nucleic Acids Res. 37 , 3739–3746 (2009). Omelina, E. S., Ivankin, A. V., Letiagina, A. E. & Pindyurin, A. V. Optimized PCR conditions minimizing the formation of chimeric DNA molecules from MPRA plasmid libraries. BMC Genomics 20 , 536 (2019). Wichłacz, A., Łęgiewicz, M. & Ciesiołka, J. Generating in vitro transcripts with homogenous 3′ ends using trans-acting antigenomic delta ribozyme. Nucleic Acids Res. 32 , e39 (2004). Li, E. et al. Vacuum and sonication treatment enables efficient transient gene expression in various monocot and eudicot plant seedlings. 2025.03.13.642903 Preprint at https://doi.org/10.1101/2025.03.13.642903 (2025). He, Y., Zhang, T., Sun, H., Zhan, H. & Zhao, Y. A reporter for noninvasively monitoring gene expression and plant transformation. Hortic. Res. 7 , 152 (2020). Yuan, G. et al. Split selectable marker systems utilizing inteins facilitate gene stacking in plants. Commun. Biol. 6 , 567 (2023). Schenkwein, D., Afzal, S., Nousiainen, A., Schmidt, M. & Ylä-Herttuala, S. Efficient Nuclease-Directed Integration of Lentivirus Vectors into the Human Ribosomal DNA Locus. Mol. Ther. 28 , 1858–1875 (2020). McStay, B. Nucleolar organizer regions: genomic ‘dark matter’ requiring illumination. Genes Dev. 30 , 1598–1610 (2016). Lisowski, L. et al. Ribosomal DNA Integrating rAAV-rDNA Vectors Allow for Stable Transgene Expression. Mol. Ther. 20 , 1912 (2012). Rodríguez-Vargas, A. & Collins, K. Distinct and overlapping RNA determinants for binding and target-primed reverse transcription by Bombyx mori R2 retrotransposon protein. Nucleic Acids Res. 52 , 6571–6585 (2024). Nguyen, V. H., Mittelsten Scheid, O. & Gutzat, R. Heat stress response and transposon control in plant shoot stem cells. Plant Physiol. 197 , kiaf110 (2025). Pollak, B. et al. Loop assembly: a simple and open system for recursive fabrication of DNA circuits. New Phytol. 222 , 628–640 (2019). Yoo, S.-D., Cho, Y.-H. & Sheen, J. Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat. Protoc. 2 , 1565–1572 (2007). Additional Declarations There is NO Competing Interest. Supplementary Files R2suppfigures.docx Supplementary Figures Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7436428","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":505205755,"identity":"ab22f400-a926-40b2-a360-5e48069c7dba","order_by":0,"name":"Gozde Demirer","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIie2OIQvCQBSAn8WVA+uZ9heeCENBVv0bGwdLQwVBDIaBMKPVn6FlyfDGUMv9gNm0LBlMYlJv2gznouE+rtzjfXwPwGD4S7z3gwZAjT5fwGpKMwIgqqyUIFVV7L7Yrc/b3rB9yE7ZdeoOwJonXKe0ZBHkXhF0ExkgkRRjYPuJXlmFTu5Rhk6ubktj8iMeOj+U4U0pT2yvrCulD6XYF71i87CuFELkTFWissL0CrKiPEwgl+GI5F74MQvGHW1lIYrjnVxsLA6b03Tm+ksr2+TaCn1P6rr1dyX6tWEwGAyGFxSrVuV9HHNdAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-3007-1489","institution":"California Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Gozde","middleName":"","lastName":"Demirer","suffix":""},{"id":505205756,"identity":"08c69433-2b2b-4d21-891b-27e003027ebe","order_by":1,"name":"Kimberley Muchenje","email":"","orcid":"https://orcid.org/0009-0005-9419-7469","institution":"California Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Kimberley","middleName":"","lastName":"Muchenje","suffix":""},{"id":505205757,"identity":"afbcf0aa-cca4-4f9e-9d9a-5ff24e24abf8","order_by":2,"name":"Yunqing Wang","email":"","orcid":"","institution":"California Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yunqing","middleName":"","lastName":"Wang","suffix":""},{"id":505205758,"identity":"d31f2fa4-5237-41d6-80bf-ff38a35edb43","order_by":3,"name":"Tufan Oz","email":"","orcid":"","institution":"California Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Tufan","middleName":"","lastName":"Oz","suffix":""},{"id":505205759,"identity":"e5885bc8-edf1-4731-9afe-ad75e6c28d07","order_by":4,"name":"Eugene Li","email":"","orcid":"","institution":"Caltech","correspondingAuthor":false,"prefix":"","firstName":"Eugene","middleName":"","lastName":"Li","suffix":""},{"id":505205760,"identity":"344a5002-5c99-42d3-906f-c865cfb05058","order_by":5,"name":"Amelia Saffron","email":"","orcid":"","institution":"California Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Amelia","middleName":"","lastName":"Saffron","suffix":""}],"badges":[],"createdAt":"2025-08-22 17:00:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7436428/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7436428/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89884601,"identity":"c989c7c4-418c-42e2-939c-d3757a6cff7c","added_by":"auto","created_at":"2025-08-26 06:20:36","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":140366,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverview of R2 retrotransposon-mediated targeted DNA insertion in plants. (A)\u003c/strong\u003e Mechanism of R2 integration at the 25S rDNA locus via site-specific Target Primed Reverse Transcription. \u003cstrong\u003e(B)\u003c/strong\u003e Constructs for detecting R2 protein expression in plants, with geminiviral (GV) parts added for enhanced expression. Proteins were tagged at the N-terminus with HiBiT tag for luminescence detection or 1x YPEt for fluorescence detection. All proteins were also tagged at the N-terminus with bipartite (BP) NLS tag for nuclear localization. \u003cstrong\u003e(C)\u003c/strong\u003e Payload design: An mCherry intron-containing cassette embedded within 5′ and 3′ UTRs is spliced to generate a mature RNA template for retrotransposition, enabling phenotypic detection of integration by confocal microscopy. \u003cstrong\u003e(D)\u003c/strong\u003e R2 system expression was performed via \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003einfiltration of \u003cem\u003eNicotiana benthamiana \u003c/em\u003eleaves and \u003cem\u003eSolanum lycopersicum \u003c/em\u003eseedlings as well as PEG-mediated \u003cem\u003eArabidopsis thaliana \u003c/em\u003eprotoplast transfection. \u003cstrong\u003e(E)\u003c/strong\u003e Integration events were evaluated by microscopy, flow cytometry, amplicon PCRs, NGS, and droplet digital PCR (ddPCR). HiBiT: high affinity bioluminescence tag, YPet: yellow fluorescent protein, BFP: blue fluorescent protein, BP NLS: Bipartite nuclear localization signal, UTR: untranslated region, rDNA: ribosomal DNA, LHA: left homology arm, RHA: right homology arm, pCmY: pCmYLCV6 promoter derived from Cestrum yellow leaf-curling virus, tHSP: \u003cem\u003eArabidopsis thaliana\u003c/em\u003e heat shock protein 18.2 terminator, GV: geminiviral replicon, LIR: long intergenic region of the geminiviral replicon, SIR: short intergenic region of the geminiviral replicon, RepA: trans-acting replication-initiation protein. SD: splice donor, SA: splice acceptor.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/cd6eacc4af50b811bbcc09ae.png"},{"id":89884603,"identity":"9971030d-d1f6-4f29-8e2b-6843e41f9cfe","added_by":"auto","created_at":"2025-08-26 06:20:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":179234,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProtein expression and function screening identify optimal R2 proteins and UTR variant combinations. (A) \u003c/strong\u003eRepresentative\u003cstrong\u003e \u003c/strong\u003econfocal fluorescence microscopy images of \u003cem\u003eN. benthamiana\u003c/em\u003e leaves expressing 1×YPet-tagged R2 proteins with or without geminiviral (GV) replication elements. Scale bars, 100 µm. \u003cstrong\u003e(B)\u003c/strong\u003e R2 protein expression quantified by HiBiT luminescence (n=3 biological replicates). Ordinary one-way ANOVA Tukey HSD multiple comparison test. Different letters denote statistically significant differences. Values are mean ± S.E.M. See Source Data File for details. \u003cstrong\u003e(C)\u003c/strong\u003e R2 editor screen setup with combinations of four UTRs and four R2 proteins. \u003cstrong\u003e(D)\u003c/strong\u003e All-in-one screening construct containing an mCherry payload cassette, R2 protein cassette, and an mTagBFP2 transformation marker. \u003cstrong\u003e(E)\u003c/strong\u003e Representative confocal microscopy images of \u003cem\u003eN. benthamiana\u003c/em\u003e leaves showing mTagBFP2 (cyan) indicating transfection and mCherry (magenta) indicating R2-mediated targeted insertion. Scale bars, 100 µm. \u003cstrong\u003e(F)\u003c/strong\u003e R2-mediated targeted insertion efficiency quantified from confocal microscopy images (n=3 biological replicates). Ordinary one-way Anova with Tukey HSD multiple comparison test. Different letters denote statistically significant differences. Values are mean ± S.E.M. See Source Data File for details. UTR: untranslated region, min: minimal sequence variant, wt: wild-type sequence, p35S: Cauliflower mosaic virus 35S promoter, pCmY: pCmYLCV6 promoter, tHSP: heat shock protein 18.2 terminator, mTagBFP2: blue fluorescent protein, mCherry: red fluorescent protein, a.u: arbitrary units.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/c27d2d4466814831e1717538.png"},{"id":89884602,"identity":"61ca3ae0-5779-4499-8563-dcd5e48ff814","added_by":"auto","created_at":"2025-08-26 06:20:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":143552,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHigher R2 system copy number and heat shock treatment increases integration efficiency.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e Schematic of the R2Tg payload cassette with 100 bp homology arms. \u003cstrong\u003e(B)\u003c/strong\u003e Schematic of the wild-type R2 editors using either 1-plasmid or 2-plasmid delivery approaches, intronized R2Tg-2i setup, and geminiviral-amplified R2 editor (GV-R2Tg-1). Representative confocal microscopy images of leaves infiltrated with R2 constructs \u003cstrong\u003e(C)\u003c/strong\u003e under no heat shock conditions and \u003cstrong\u003e(D) \u003c/strong\u003eheat shock; mTagBFP2 (cyan) is a transfection control; mCherry (magenta) reports successful R2-mediated integration at 25S rDNA loci. Scale bars, 100 µm. (\u003cstrong\u003eE)\u003c/strong\u003e Quantification of mCherry positive nuclei as a measure of targeted insertion efficiency. Bars represent mean ± S.E.M. of 4 independent biological replicates. P values were obtained using ordinary one-way ANOVA with Tukey HSD multiple comparison test. Different letters denote statistically significant differences. See Source Data File for details. LHA: left homology arm, RHA: right homology arm, p35S: Cauliflower mosaic virus 35S promoter, t35S: Cauliflower mosaic virus 35S terminator, pVmV: Cassava vein mosaic virus promoter, SD: splice donor, SA: splice acceptor, pCmY: pCmYLCV6 promoter, tHSP: \u0026nbsp;heat shock protein 18.2 terminator, mTagBFP2: blue fluorescent protein variant, mCherry: red fluorescent protein, SIR: short intergenic region and LIR: long intergenic region of the geminiviral replicon, RepA: geminiviral replication-associated protein.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/fe31e0c3edc619db7482baa6.png"},{"id":89884604,"identity":"6c08ffc8-f6f9-4a51-8125-5363395a794e","added_by":"auto","created_at":"2025-08-26 06:20:36","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":214381,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIncreased integration efficiency and molecular validation via minimal homology length RNA templates with HDV ribozyme. (A) \u003c/strong\u003eSchematics of R4 and R4-HDV minimal payload RNA templates and two delivery formats. \u003cstrong\u003e(B) \u003c/strong\u003eRepresentative\u003cstrong\u003e \u003c/strong\u003econfocal microscopy images of \u003cem\u003eN. benthamiana\u003c/em\u003e leaves infiltrated with constructs encoding different R2Tg variants and donor RNA templates. mTagBFP2 (cyan, reporter for transformation), mCherry (magenta, R2-mediated integration reporter), and brightfield overlay. Scale bars, 100 µm.\u003cstrong\u003e (C) \u003c/strong\u003eQuantification of mCherry positive nuclei as a measure of targeted insertion efficiency. Bars represent mean ± S.E.M. of 3 independent biological replicates. P values were obtained using ordinary one-way ANOVA with Tukey HSD multiple comparison test. Different letters denote statistically significant differences. See Source Data File for details. \u003cstrong\u003e(D) \u003c/strong\u003eTarget-junction PCR on genomic DNA isolated from infiltrated leaves. Expected band sizes for the 5′ and 3′ junctions are 425 bp and 454 bp. \u003cstrong\u003e(E) \u003c/strong\u003eRepresentative Sanger sequencing chromatograms of 5’ and 3’ integrated junction amplicons for the R4-HDV payload paired with intronized R2Tg (R2Tg-1i). \u003cstrong\u003e(F) \u003c/strong\u003eDroplet digital PCR (ddPCR) assay schematic for quantification of integration efficiency using primers and a probe specific to the integrated 3′ junction, normalized against an endogenous \u003cem\u003eNbEF1α\u003c/em\u003e control locus. \u003cstrong\u003e(G) \u003c/strong\u003eQuantification of targeted integration efficiencies by ddPCR. Bars represent mean ± S.E.M. from 3 independent biological replicates. Statistical analysis is done using ordinary one-way ANOVA with Tukey HSD multiple comparison test. Different letters denote statistically significant differences. See Source Data File for details. R2Tg: \u003cem\u003eTaeniopygia guttata\u003c/em\u003e R2 element, RT: reverse transcriptase, rDNA: ribosomal DNA, UTR: untranslated region, HDV: hepatitis delta virus ribozyme, R4: 4-nucleotide sequence derived from the 25S rDNA region immediately downstream of the target site, R33: 33-nucleotide sequence derived from the 25S rDNA region immediately upstream of the target site, wt: wild type, pVmv: Cassava vein mosaic virus promoter, tHSP: \u003cem\u003eArabidopsis thaliana\u003c/em\u003e heat shock protein 18.2 terminator, pCmY: pCmYLCV6 promoter derived from \u003cem\u003eCestrum\u003c/em\u003e yellow leaf-curling virus, p35s: Cauliflower mosaic virus 35S promoter, t35s: Cauliflower mosaic virus 35S terminator, SD: splice donor, SA: splice acceptor.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/e5b5416ba6ed4cc8113a45b1.png"},{"id":89884605,"identity":"15166962-8137-4cb8-b0aa-40f951c6e2e0","added_by":"auto","created_at":"2025-08-26 06:20:36","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":101825,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOptimized R2Tg system achieves efficient targeted insertion of a large DNA payload expressing the RUBY biosynthetic pathway. (A) \u003c/strong\u003eSchematics of R4-HDV RUBY payload (5 kb) with two delivery setups. \u003cstrong\u003e(B) \u003c/strong\u003eTarget-junction PCR on genomic DNA isolated from infiltrated leaves. Expected band sizes for the 5′ and 3′ junctions are 425 bp and 454 bp. \u003cstrong\u003e(C) \u003c/strong\u003eSchematic of ddPCR on the 3’ and 5’ integrated junctions. \u003cstrong\u003e(D) \u003c/strong\u003eQuantification of targeted integration efficiencies by ddPCR on the 5’ and \u003cstrong\u003e(E)\u003c/strong\u003e 3’junction.Bars represent mean ± S.E.M. from 3 independent biological replicates. Statistical analysis is done using ordinary one-way ANOVA with Tukey HSD multiple comparison test. Different letters denote statistically significant differences. See Source Data File for details. RT: reverse transcriptase, rDNA: ribosomal DNA, UTR: untranslated region, HDV: hepatitis delta virus ribozyme, R4: 4-nucleotide sequence derived from the 25S rDNA region immediately downstream of the target site, R33: 33-nucleotide sequence derived from the 25S rDNA region immediately upstream of the target site, wt: wild type, pVmv: Cassava vein mosaic virus promoter, tHSP: \u003cem\u003eArabidopsis thaliana\u003c/em\u003e heat shock protein 18.2 terminator, pCmY: pCmYLCV6 promoter, p35s: Cauliflower mosaic virus 35S promoter, t35s: Cauliflower mosaic virus 35S terminator, SD: splice donor, SA: splice acceptor. RUBY: An artificial polycistronic open reading frame that expresses the three enzymes required for Betalain biosynthesis, CYP76AD1: cytochrome P450 enzyme converts tyrosine to L-DOPA (L-3,4-dihydroxyphenylalanine), DODA: l-DOPA 4,5-dioxygenase enzyme cleaves L-DOPA to produce betalamic acid, Glucosyl transferase: glycosylates betanidin to generate the red colored betalain compound, 2A: self-cleaving 2A peptide sequence.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/7f6c11f8212a174590f2a09d.png"},{"id":89885481,"identity":"1a686d8f-494f-4a5e-84cd-5ad3770158f8","added_by":"auto","created_at":"2025-08-26 06:28:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2314809,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/b6c23b03-f45c-4a22-817a-35ea310e38b2.pdf"},{"id":89884607,"identity":"a5a73b7a-bb1d-4a7f-9f0f-57ae569a211a","added_by":"auto","created_at":"2025-08-26 06:20:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":61971016,"visible":true,"origin":"","legend":"Supplementary Figures","description":"","filename":"R2suppfigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7436428/v1/97f8d773b32fd44e216a7dcc.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Optimized R2 Retroelement Complexes Enable Precise and Efficient DNA Insertion into Plant Genomes","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCrop yields are increasingly threatened by rising global temperatures, intensifying drought, and the expanding range of plant diseases\u003csup\u003e1\u003c/sup\u003e. Genome engineering offers a powerful, timely strategy for enhancing climate resilience, food security, and sustainable biomanufacturing\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e3\u003c/sup\u003e. The development of improved crops relies heavily on the incorporation of foreign genes and heterologous metabolic pathways into plant genomes, where most approaches use \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e for DNA delivery and insertion. However, the random DNA integration of \u003cem\u003eAgrobacterium\u0026nbsp;\u003c/em\u003ecan often cause variable gene expression due to local chromatin context and nearby regulatory elements, lead to silencing, and prevent precise tagging of specific genes or loci, all of which are limitations complicating functional genomics studies and requiring extensive screening across generations to identify lines with desired, stable, and heritable expression\u003csup\u003e4–7\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEmerging targeted insertion technologies allow controlled placement of DNA sequences at predetermined genomic loci, eliminating risks of unintended gene expression changes and genomic instability\u003csup\u003e8\u003c/sup\u003e. Furthermore, precise DNA insertion at defined genomic sites is essential for probing gene function and regulation and enables robust outputs in synthetic gene circuits, efficient trait stacking, and accelerated development of engineered crops for agriculture and bioproduction\u003csup\u003e9,10\u003c/sup\u003e. To date, multiple strategies have been developed to achieve site-specific DNA integration in plants. Programmable nucleases, such as CRISPR-Cas9, introduce targeted double-strand breaks (DSBs), which are repaired by host machinery\u003csup\u003e11,12\u003c/sup\u003e. Although homology-directed repair can result in precise integration using donor templates with flanking homology arms at the DSB site\u003csup\u003e13\u003c/sup\u003e, this pathway is not preferred in most plant cells, where DNA breaks are predominantly repaired by non-homologous end joining, often leading to low efficiency, imprecise insertions, and indel generation\u003csup\u003e14\u003c/sup\u003e. Moreover, DSBs can lead to unpredictable outcomes, such as large deletions or chromosomal rearrangements\u003csup\u003e15\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs an alternative, prime editing offers a DSB-free approach for introducing precise nucleotide changes and short sequence insertions into the genome\u003csup\u003e16,17\u003c/sup\u003e. Despite broad applicability, its efficiency to insert large DNA segments in plants remains limited, restricting its use for integrating full-length genes or metabolic pathways\u003csup\u003e18\u003c/sup\u003e. Expanding the capabilities of prime editing, PrimeRoot employs optimized prime editing guide RNA (pegRNA) architectures, prime editors, and site-specific recombinases to achieve DSB-free targeted insertion of large DNA sequences into plant genomes\u003csup\u003e19\u003c/sup\u003e. Although promising with up to 6.3% efficiency, its broader application may be limited by lack of generalizable pegRNA design rules to ensure robust activity across diverse loci, and the low efficiency of the recombinase recognition site installation, which is a critical first step in PrimeRoot. These challenges highlight the need for continued engineering of more effective editors to fully realize the potential of precise, programmable DNA insertion in plants.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExpanding the toolkit further, transposon-based systems offer another alternative for programmable insertion. Transposons are mobile elements that can be broadly categorized as DNA transposons or retrotransposons based on their integration intermediates. They possess sophisticated insertion mechanisms setting them up as powerful tools for targeted insertion\u003csup\u003e20,21\u003c/sup\u003e. A recent discovery of CRISPR-associated DNA transposases (CAST) in bacterial genomes enabled the development of editors that use programmable guide RNAs to direct transposon insertion to defined genomic sites\u003csup\u003e22,23\u003c/sup\u003e. Unlike nuclease-based editors, they lack DNA cleavage activity and instead rely entirely on transposase machinery to perform the integration. Even though their translation to eukaryotic human cells has been successful – albeit with a low efficiency\u003csup\u003e24\u003c/sup\u003e – native bacterial CAST system has yet to demonstrate feasibility in plants. However, building on this discovery, transposase-assisted target-site integration (TATSI) system was developed, which co-expresses Cas9 nuclease with the rice Pong DNA transposase to enable high efficiency, targeted DNA insertion in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e and soybean\u003csup\u003e25\u003c/sup\u003e. While TATSI offers a promising approach for targeted DNA insertion in plants, its performance can be affected by high rates of off-target integration at native transposase recognition motifs, reliance on DSBs, and increased susceptibility to silencing due to the use of plant-derived transposases. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition to DNA transposons, retrotransposons have also shown recent promise for targeted gene insertion in eukaryotes. R2 non-long terminal repeat (LTR) retrotransposons are a class of sequence-specific mobile elements that insert into the conserved 28S ribosomal DNA (rDNA) tandem repeat loci and have so far been identified in the genomes of multicellular animals including insects, crustaceans, and non-mammalian vertebrates\u003csup\u003e26–28\u003c/sup\u003e. R2 proteins have been successfully repurposed as high-efficiency genome editors in mammalian systems, both at their conserved target sites and at other RNA-guided genomic loci through Cas9 fusion, leveraging their target-primed reverse transcription (TPRT) mechanism\u003csup\u003e29–32\u003c/sup\u003e. Engineered R2 retrotransposons mediate RNA-templated insertions at the native 28S rDNA locus with efficiencies up to 80% in HEK293T cells and 60% in mouse embryos, and they achieve up to 40% at Cas9-directed, non-rDNA sites\u003csup\u003e30–32\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eR2 retrotransposons provide a unique, DSB-free approach for targeted DNA insertion into the 28S rDNA site, which is referred to as 25S rDNA site in plants,\u0026nbsp;offering broad adaptability for genome editing across plant species. The required domains for function, DNA binding, endonuclease, and reverse transcriptase, are encoded within a single open reading frame, enabling a compact editor design that can be co-delivered with its RNA payload as a ribonucleoprotein (RNP) complex. Moreover, the high copy number of rDNA repeats in plants (~800 copies in \u003cem\u003eA. thaliana\u003c/em\u003e\u003csup\u003e33\u003c/sup\u003e, ~1247 copies in \u003cem\u003eNicotiana benthamiana\u003c/em\u003e\u003csup\u003e34\u003c/sup\u003e, ~1200 copies in \u003cem\u003eSolanum lycopersicum\u003c/em\u003e\u003csup\u003e35,36\u003c/sup\u003e and ~6650 copies in wheat\u003csup\u003e37\u003c/sup\u003e) can further increase integration frequency and expression, and simplify downstream screening for precise insertion events.\u003c/p\u003e\n\u003cp\u003eHere, we established and optimized a plant genome editor based on the R2 retrotransposon that enables targeted, high-efficiency DNA insertion into the 25S rDNA loci. This system combines: i) an optimized R2 expression, ii) RNA donor carrying mCherry and RUBY retrotransposition reporters with length-optimized rDNA-homology arms, and iii) systematically tuned 5′ and 3′ UTRs in the RNA template. The optimized system achieves integration of payloads up to 5kb size in \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaves, \u003cem\u003eArabidopsis thaliana\u003c/em\u003e protoplasts, and \u003cem\u003eSolanum lycopersicum\u003c/em\u003e seedlings with efficiencies reaching 24% of transformed cells quantified by imaging and 0.9-1 copies per genome quantified by droplet digital PCR on both 5′ and 3′ integrated junctions. R2-mediated insertion events were validated at the 25S rDNA loci with molecular evidence confirming both high efficiency and target specificity, establishing a robust platform for precise DNA insertion into plant genomic safe-harbor loci for various fundamental plant biology or biotechnology applications.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEstablishing an R2 retrotransposon-mediated platform for targeted DNA insertion in plants.\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBuilding on the well-characterized target-primed reverse transcription (TPRT) activity of R2 proteins in mammalian cells\u003csup\u003e27,28\u003c/sup\u003e, we investigated whether the same machinery could mediate precise DNA insertion in the plant nuclear genome. In the canonical mechanism, an R2 ribonucleoprotein complex composed of an R2 protein and its RNA template binds the 28S rDNA locus (25S rDNA in plants) in a manner that depends on the secondary structures of RNA sequences within the 5\u0026prime; and 3\u0026prime; untranslated regions (UTRs)\u003csup\u003e27,28\u003c/sup\u003e. The complex then nicks the bottom strand, exposing a 3\u0026prime;-hydroxyl primer, and synthesizes first-strand complementary DNA (cDNA) from its RNA template\u003csup\u003e27\u003c/sup\u003e. Subsequently, the R2 protein nicks the top strand, and initiates second-strand synthesis\u003csup\u003e28\u003c/sup\u003e, where the integrated element is flanked by the R2 5\u0026prime; and 3\u0026prime; UTRs (\u003cstrong\u003eFig. 1A\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe deployed the R2 retrotransposon system (R2 system in short) to leverage this double-strand-break-free integration mechanism, combined with the high copy number\u003csup\u003e34,37\u003c/sup\u003e and transcriptional activity\u003csup\u003e38,39\u003c/sup\u003e of the 25S rDNA array, to achieve robust expression of integrated gene sequences in plants. Efficient TPRT requires high amounts of nuclear-localized R2 protein\u003csup\u003e40\u003c/sup\u003e. We therefore codon-optimized R2 open reading frames for plant expression, appended an N-terminal strong bipartite nuclear localization signal (BP NLS), and fused either a high affinity bioluminescence tag (HiBiT) for luminescence complementation or a single-copy YPet fluorescent protein to the N-terminus for quantifying and/or imaging protein expression (\u003cstrong\u003eFig. 1B\u003c/strong\u003e). Each construct was driven by the strong pCmYLCV6 promoter and terminated by the \u003cem\u003eA. thaliana\u003c/em\u003e tHSP 18.2 terminator. To further increase intracellular R2 copy number, long and short intergenic regions (LIR-SIR) together with the RepA replication-initiation protein were incorporated as a second set to test, generating self-replicating geminiviral (GV) replicons that boost transient expression\u003csup\u003e41\u003c/sup\u003e (\u003cstrong\u003eFig. 1B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTo enable simple phenotypic detection of R2-mediated insertion events, we built an RNA payload to function as a retrotransposition reporter following a classic GFP reporter\u003csup\u003e42,43\u003c/sup\u003e, in which either the \u003cem\u003eSolanum tuberosum\u003c/em\u003e ST-LS1 intron\u003csup\u003e44\u003c/sup\u003e or an \u003cem\u003eArabidopsis thaliana\u003c/em\u003e intron\u003csup\u003e45\u003c/sup\u003e-disrupted mCherry cassette was embedded between the 5\u0026prime; and 3\u0026prime; untranslated regions (UTRs) of the R2 element (\u003cstrong\u003eFig. 1C)\u003c/strong\u003e. Following transcription and intron splicing, the resulting RNA forms a contiguous template for reverse transcription via R2 protein. Hence, precise integration via R2 system restores an intron-less mCherry expression cassette, producing a fluorescent signal that enables visualization and quantification by confocal microscopy.\u003c/p\u003e\n\u003cp\u003eTo evaluate R2 system activity across different contexts, we introduced R2 protein and RNA payload constructs into \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaves and \u003cem\u003eSolanum lycopersicum\u003c/em\u003e seedlings via \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e infiltration, and into \u003cem\u003eA. thaliana\u003c/em\u003e protoplasts using polyethylene glycol (PEG)-mediated transfection (\u003cstrong\u003eFig. 1D\u003c/strong\u003e). BFP fluorescent marker was included in the constructs for efficiency normalization. Phenotypic detection was achieved via confocal microscopy for \u003cem\u003eN. benthamiana\u003c/em\u003e and \u003cem\u003eS. lycopersicum\u003c/em\u003e leaves and flow cytometry for \u003cem\u003eA. thaliana\u0026nbsp;\u003c/em\u003eprotoplasts. Molecular characterization of insertion junctions and copy number was performed using amplicon PCR, Sanger sequencing, high-depth next-generation sequencing (NGS), and droplet digital PCR (ddPCR) (\u003cstrong\u003eFig. 1E\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eR2 expression optimization and functional screening identify an active R2 system in plants\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo identify a functional R2 editor in plants, we first compared the expression profiles of four codon-optimized R2 proteins: R2Bm from \u003cem\u003eBombyx mori\u003c/em\u003e, R2Za from \u003cem\u003eZonotrichia albicollis\u003c/em\u003e, R2Tg from \u003cem\u003eTaeniopygia guttata\u003c/em\u003e, and a rationally improved variant of R2Tg, R2Tg-OPT\u003csup\u003e30\u003c/sup\u003e. These proteins were expressed in \u003cem\u003eN. benthamiana\u0026nbsp;\u003c/em\u003eleaves using Agro-infiltration. All proteins were tested with and without geminiviral (GV) replicon elements to evaluate the impact of expression cassette copy number on R2 protein accumulation (\u003cstrong\u003eFig. 1B\u003c/strong\u003e). 3 days after infiltration, we detected the expression of all YPet-tagged R2 proteins using confocal microscopy, where nuclear-localized YPet signal was consistently stronger in GV versions (\u003cstrong\u003eFig. 2A\u003c/strong\u003e). Luminescence quantification using a plate reader for the HiBiT-tagged set of proteins in the plant nuclear fraction confirmed that GV parts significantly increased R2 protein levels across all variants (\u003cstrong\u003eFig. 2B\u003c/strong\u003e). These results established our ability to express high levels of all tested R2 proteins in plants cells and that GV replicons are an effective strategy for increasing R2 protein levels in the nuclei of \u003cem\u003eN. benthamiana\u003c/em\u003e leaf cells.\u003c/p\u003e\n\u003cp\u003ePrevious mammalian studies have shown that autocatalytic cleavage of the 5\u0026prime; UTR is required to generate competent RNA templates for R2-mediated integration via TPRT\u003csup\u003e30\u003c/sup\u003e. To assess whether the 5\u0026prime; UTR sequences used in our donor constructs enable autocleavage, we performed \u003cem\u003ein vitro\u003c/em\u003e transcription and cleavage assays using the wild-type and minimal length \u003cem\u003eT. castaneum\u003c/em\u003e and \u003cem\u003eT. guttata 5\u0026rsquo; UTR\u003c/em\u003e, each fused to 100 bp of the rDNA sequence upstream of the insertion site, referred to here as the left homology arm (LHA) (\u003cstrong\u003eSupplementary Fig. 1A\u003c/strong\u003e). Distinct cleavage products were observed by agarose gel electrophoresis at expected band sizes, confirming that all tested UTRs retain autocatalytic activity \u003cem\u003ein vitro\u003c/em\u003e (\u003cstrong\u003eSupplementary Fig. 1B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eNext, we performed a systematic R2 editor function screen to identify an optimal pairing of R2 protein and donor RNA architecture. Each configuration combined one of the four R2 proteins with a donor cassette encoding the mCherry-based retrotransposition reporter and different UTR combinations (\u003cstrong\u003eFig. 2C\u003c/strong\u003e). We evaluated four UTR combinations, all previously shown to be active in mammalian systems, derived from three species: minimal \u003cem\u003eT. castaneum\u003c/em\u003e 5\u0026prime; UTR with wild-type \u003cem\u003eGeospiza fortis\u003c/em\u003e 3\u0026prime; UTR; wild-type \u003cem\u003eT. castaneum\u003c/em\u003e 5\u0026prime; UTR with wild-type \u003cem\u003eG. fortis\u003c/em\u003e 3\u0026prime; UTR; wild-type \u003cem\u003eT. guttata\u003c/em\u003e 5\u0026prime; and 3\u0026prime; UTRs; and wild-type \u003cem\u003eT. guttata\u003c/em\u003e 5\u0026prime; UTR paired with a minimal \u003cem\u003eT. guttata\u003c/em\u003e 3\u0026prime; UTR\u003csup\u003e29,30\u003c/sup\u003e (\u003cstrong\u003eFig. 2C\u003c/strong\u003e), generating a library of 16 R2-UTR combinations to assess. Each all-in-one construct for the R2 editor function screen also encoded a constitutive mTagBFP2 reporter to normalize for transformation efficiency (\u003cstrong\u003eFig. 2D\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTo evaluate functional activity, 7 days after infiltration, we imaged \u003cem\u003eN. benthamiana\u003c/em\u003e leaves infiltrated with constructs expressing all-in-one constructs for each of the 16 R2 protein-UTR combinations and scored mCherry fluorescence as a direct readout of successful retrotransposition into the 25S rDNA (\u003cstrong\u003eFig. 2E\u003c/strong\u003e). R2Bm and R2Za proteins did not yield detectable integration activity when paired with any of 4 UTR combinations in our editor screen (\u003cstrong\u003eSupplementary Fig. 1C, D\u003c/strong\u003e). The highest integration efficiency, defined as the percentage of leaf cells with detectable mCherry signal, was obtained with wild-type R2Tg paired with its wild-type 5\u0026prime; and 3\u0026prime; UTRs at ~0.5% efficiency (\u003cstrong\u003eFig. 2E, F\u003c/strong\u003e). Two additional R2Tg editor configurations with different UTRs also yielded successful integration (\u003cstrong\u003eFig. 2E, F; Supplementary Fig. 1D\u003c/strong\u003e): R2Tg enabled detectable integration when paired with wild-type 5\u0026prime; and minimal 3\u0026prime; UTRs from \u003cem\u003eT. guttata\u003c/em\u003e (0.11%), and wild-type \u003cem\u003eT. castaneum\u003c/em\u003e 5\u0026prime; UTR and wild-type \u003cem\u003eG. fortis\u003c/em\u003e 3\u0026prime; UTR (0.05%) (\u003cstrong\u003eFig. 2E, F\u003c/strong\u003e). In contrast, no mCherry signal was observed when R2Tg was paired with the minimal \u003cem\u003eT. castaneum\u003c/em\u003e 5\u0026prime; UTR and wild-type \u003cem\u003eG. fortis\u003c/em\u003e 3\u0026prime; UTR (\u003cstrong\u003eFig. 2E, F\u003c/strong\u003e). R2Tg-OPT exhibited minimal activity only when paired with wild-type \u003cem\u003eT. guttata\u003c/em\u003e UTRs, and none of the other UTR combinations supported detectable integration by this variant. Our negative control samples consisting only of the payload without R2 protein did not generate any mCherry signal as expected (\u003cstrong\u003eFig. 2F\u003c/strong\u003e; \u003cstrong\u003eSupplementary Fig. 2)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis preliminary R2 editor function screen established the wild-type R2Tg protein, paired with its wild-type 5\u0026prime; and 3\u0026prime; UTRs, as the most efficient R2 editor configuration, showing potential for further optimization via R2 protein and payload engineering. We next focused on increasing the integration efficiency of the R2Tg editor paired with the mCherry payload containing its wild-type 5\u0026prime; and 3\u0026prime; UTRs with a payload length of 2.9kb, which we refer to as the R2Tg editor system.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eR2Tg system optimization via improved protein \u0026amp; payload expression and heat shock treatment\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo ensure that mCherry fluorescence in our optimization assays only reflects bona fide R2-mediated integration events, we used an intron-disrupted mCherry retrotransposition reporter, same as above (\u003cstrong\u003eFig. 3A\u003c/strong\u003e). This payload comprises an mCherry coding sequence in the antisense orientation relative to the RNA payload, interrupted by a plant intron in the sense orientation, flanked by R2TG wild-type 5\u0026prime; and 3\u0026prime; UTRs and 100 bp homology arms corresponding to the 25S rDNA target site. Transcription of the integrated cassette can only yield mCherry signal if the payload transcript is precisely spliced to remove the intron prior to R2-mediated insertion at the target locus. \u003cem\u003eAgrobacteria\u003c/em\u003e T-DNA insertions or leaky expression cannot generate a functional, intron-free mCherry transcript because the cassette remains intron-bearing or incorrectly oriented, with the splice donor and acceptor motifs are present only on the sense strand, thereby ensuring the absence of false-positive signal in downstream fluorescence readouts.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUsing this reporter, we next tested whether targeted integration efficiency could be improved. Retrotransposon sequences are rapidly targeted by RNA-directed DNA methylation and converted to a stably silenced, heterochromatic state in plant cells\u003csup\u003e46\u0026ndash;49\u003c/sup\u003e. We suspected that transcriptional interference\u003csup\u003e50\u003c/sup\u003e from all cassettes contained in the all-in-one construct and silencing of R2 editor encoding sequences may be limiting the efficiencies in our initial R2 editor screen. To mitigate this effect, we split the system into two plasmids: one encoding the R2Tg protein and the other carrying the R2Tg payload, a strategy intended to reduce transcriptional interference (\u003cstrong\u003eFig. 3B\u003c/strong\u003e). We further inserted two \u003cem\u003eA. thaliana\u003c/em\u003e introns\u003csup\u003e45\u003c/sup\u003e into the R2Tg protein coding sequence to suppress silencing of the R2 editor system (\u003cstrong\u003eFig. 3B\u003c/strong\u003e). Initial efforts to express R2 proteins in \u003cem\u003eN. benthamiana\u003c/em\u003e showed that protein expression could be significantly improved by the introduction of geminiviral replicons leveraging autonomous replication. In line with this finding, overexpressing the complete all-in-one editor system could boost integration efficiency by increasing the overall R2 editor DNA copy number. Following these insights, we set out to test whether separating the R2Tg editor components into two constructs (RT2g-2), introducing introns into the R2Tg coding sequence (RT2g-i), and flanking the full editor system with geminiviral replicons (GV-RT2g) would improve targeted integration efficiency at the 25S rDNA site in plants (\u003cstrong\u003eFig. 3B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eFollowing the same method of \u003cem\u003eN. benthamiana\u003c/em\u003e leaf \u003cem\u003eAgro\u003c/em\u003e-infiltration and quantifying number of mCherry expressing cells from confocal images, we found that the separation of R2Tg editor components into 2 plasmids improved integration efficiency by 1.7-fold, while the introduction of introns into the R2Tg coding sequence combined with the 2-plasmid delivery approach resulted in a 2.7-fold improvement compared to the 1-plasmid delivery without introns (\u003cstrong\u003eFig. 3C, E; Supplementary Fig. 3A\u0026nbsp;\u003c/strong\u003efor more replicates). Addition of geminiviral elements to the complete 1-plasmid R2Tg editing system led to a 3.3-fold increase in integration efficiency (\u003cstrong\u003eFig. 3C, E; Supplementary Fig. 3A\u003c/strong\u003e), achieving highest integration efficiency of ~1.8% in this set (GV-RT2g-1). Together, these findings support the hypothesis that R2Tg protein silencing and transcriptional interference in the initial all-in-one constructs limit integration efficiency, and that strategies aimed at reducing both effects and increasing the copy number of both the R2Tg protein and payload cassettes can significantly enhance targeted integration at the 25S rDNA site.\u003c/p\u003e\n\u003cp\u003eSuccessful applications of R2 editors in mammalian cells consistently incubate cells transformed with editors at 37\u0026deg;C before detecting editing events\u003csup\u003e29\u0026ndash;32\u003c/sup\u003e. The R2 editor functional screen in \u003cem\u003eN. benthamiana\u003c/em\u003e, however, was conducted under standard plant growth conditions of 22\u0026deg;C during the light period and 20\u0026deg;C during the dark period in a 16-hour light/8-hour dark photoperiod. Given the high body temperature of native organisms with R2, we hypothesized that the lower temperatures used in plant studies may be limiting the catalytic efficiency of R2-mediated TPRT, and that elevating the temperature via heat shock may enhance integration activity.\u003c/p\u003e\n\u003cp\u003eTo test this, \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e\u0026ndash;infiltrated plants were first maintained at 25\u0026deg;C for 48 hours under a 16-hour light/8-hour dark cycle, allowing sufficient time for T-DNA delivery and accumulation of the R2 protein and RNA payload prior to the induction of heat shock. Plants were then exposed to a heat shock regimen of 25\u0026deg;C during the light period and 37\u0026deg;C during the dark period for an additional four days before analysis. Heat shock treatment of the single-plasmid R2Tg-1 system resulted in a 1.9-fold increase in targeted integration efficiency compared to the non-heat-shocked single-plasmid R2Tg-1 system. When applied to the geminiviral-amplified GV-R2Tg-1 system, heat shock yielded a 1.27-fold improvement, corresponding to a 4.2-fold increase relative to the baseline R2Tg-1 construct, reaching up to ~2% targeted integration efficiency \u003cstrong\u003e(Fig. 3D, E; Supplementary Fig. 3A\u0026nbsp;\u003c/strong\u003efor more replicates). These findings revealed that the elevated temperature indeed enhances R2Tg-mediated TPRT \u003cem\u003ein planta\u003c/em\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eThe R2Tg editor system achieves integration into A. thaliana and N. benthamiana protoplasts\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo enable a more streamlined and quantitative workflow for assessing R2Tg-mediated integration at single-cell resolution using flow cytometry, we delivered all-in-one R2Tg editor constructs, with or without geminiviral elements, into \u003cem\u003eA. thaliana\u003c/em\u003e protoplasts using polyethylene glycol (PEG)-mediated transformation. The baseline R2Tg-1 system yielded a targeted integration efficiency of 0.765%, while the geminiviral-amplified GV-R2Tg-1 system achieved 1.167% \u003cstrong\u003e(Supplementary Fig. 3B)\u003c/strong\u003e, as reported by flow cytometry referring to % cells with mCherry fluorescence. Although these results demonstrate that the R2Tg system is active in \u003cem\u003eA. thaliana\u003c/em\u003e protoplasts, in addition to \u003cem\u003eN. benthamiana\u003c/em\u003e leaves, protoplast viability declines significantly after 48 hours, limiting the number of intact cells and window for achieving and detecting integration events.\u003c/p\u003e\n\u003cp\u003eTo address this limitation, we established a complementary assay transforming \u003cem\u003eN. benthamiana\u0026nbsp;\u003c/em\u003eleaves via \u003cem\u003eAgro\u003c/em\u003e-infiltration, and then extracting protoplasts from this transformed tissue, which allows longer incubation of the R2Tg editor system \u003cem\u003ein planta\u003c/em\u003e. R2Tg-1 constructs were delivered by \u003cem\u003eAgrobacterium\u003c/em\u003e infiltration, and targeted integration was detected and quantified 5 days post-infiltration in protoplasts extracted from infiltrated leaves using flow cytometry measuring mCherry fluorescence. This system yielded substantially higher integration efficiencies, reaching 4.5% with the R2Tg-1 construct \u003cstrong\u003e(Supplementary Fig. 3C)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eThese findings confirm that the R2Tg editor system drives targeted integration into the 25S rDNA locus in both protoplast and leaf cells of two plant species. They further demonstrate that \u003cem\u003eA. thaliana\u003c/em\u003e protoplasts provide a functional platform for early-stage detection, while intact leaves support extended assay durations and higher integration efficiencies for downstream applications.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCompact RNA template with minimal homology length and 3\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e\u0026prime;\u003cstrong\u003e\u0026nbsp;ribozyme achieves the highest integration efficiency\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMolecular characterization of integration junctions at the 25S rDNA locus is critical to confirm the activity and specificity of R2Tg-mediated targeted insertion. However, the repetitive structure of rDNA tandem arrays and the presence of identical homology arms within the retrotransposition reporter generate false-positive chimeric PCR products, complicating unambiguous identification of genuine integration events\u003csup\u003e51\u003c/sup\u003e. In mammalian systems, such artifacts were successfully eliminated by delivering both components of the R2 editor system as RNA through the PRINT (precise RNA-mediated insertion of transgenes) method\u003csup\u003e29\u003c/sup\u003e. Given the technical limitations of the RNA delivery in plants, we designed a modified mCherry reporter harboring minimal homology arms of 33 base pairs upstream (R33) and 4 base pairs downstream (R4) relative to the 25S rDNA cleavage site \u003cstrong\u003e(Fig. 4A)\u003c/strong\u003e. We reasoned that reducing the homology arm length would minimize PCR-generated chimera formation, thereby enabling more accurate quantification of authentic integration events without compromising efficiency, consistent with observations in mammalian cells\u003csup\u003e29,32\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWe also replaced the Cassava vein mosaic virus promoter (pVmv) that drives the expression of the mCherry reporter with the stronger Cauliflower mosaic virus 35S promoter (p35s) \u003cstrong\u003e(Fig. 4A)\u003c/strong\u003e. Additionally, to generate a homogeneous and precise 3\u0026prime; RNA end, we incorporated the hepatitis delta virus (HDV) ribozyme\u003csup\u003e52\u003c/sup\u003e downstream of the payload. This ribozyme enables autocatalytic cleavage, resulting in precise processing of the RNA transcript and retention of the R4 sequence at the 3\u0026prime; terminus of the R2Tg retrotransposition reporter. These modifications yielded two distinct RNA template designs, termed the R4 minimal payload and the R4-HDV minimal payload, respectively \u003cstrong\u003e(Fig. 4A)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eWe evaluated the integration efficiencies of R2Tg editor system with the R4 and R4-HDV payload designs in \u003cem\u003eN. benthamiana\u003c/em\u003e leaves. Constructs were delivered as a 1-plasmid or a 2-plasmid system separating protein and payload cassettes \u003cstrong\u003e(Fig. 4A)\u003c/strong\u003e. Confocal microscopy performed 7 days post-infiltration revealed consistent mCherry signals, indicative of successful R2-mediated integration for both payload designs when paired with the wild-type R2Tg protein or an intronized version of the protein \u003cstrong\u003e(Fig. 4B, Supplementary Fig. 4A\u0026nbsp;\u003c/strong\u003efor more replicates). Quantification of the percentage of cells expressing mCherry from confocal images revealed that the new designs have substantially improved the integration efficiencies. In the case of R4 minimal payload design, efficiency of R2Tg-1 increased from 0.5% to 10.4%, R2Tg-2 increased from 0.85% to 12.2%, and R2Tg-2i increased from 1.6% to 15.1%. In the case of R4-HDV minimal payload design, efficiency of R2Tg-1 reached up to 10.7%, R2Tg-1i up to 12.6%, R2Tg-2 up to 20.3%, and R2Tg-2i up to 23.6%. Notably, R2Tg-mediated integration was markedly increased in samples expressing the R4-HDV payload compared to the R4 payload \u003cstrong\u003e(Fig. 4B, C; Supplementary Fig. 4A)\u003c/strong\u003e. In addition, intronized R2Tg variants delivered via either 1- or 2-plasmid strategies consistently generated the strongest mCherry signals, while the R2Tg RT-dead mutant\u003csup\u003e29\u003c/sup\u003e did not produce any detectable fluorescence neither the payload without the R2 protein \u003cstrong\u003e(Fig. 4B, C; Supplementary Fig. 4A)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eTarget-junction PCR assays corroborated imaging results at the molecular level, producing amplicons of expected sizes (425 bp for 5\u0026prime; junction, 454 bp for 3\u0026prime; junction) in samples harboring functional R4-HDV R2Tg minimal payload variant \u003cstrong\u003e(Fig. 4D)\u003c/strong\u003e. Both 5\u0026prime; and 3\u0026prime; junction PCR products were detected for intronized and wild-type R2Tg constructs paired with the R4-HDV minimal payload, whereas RT-dead controls lacked the correct-sized band. Sanger sequencing junction amplicons from R4-HDV payload paired with the intronized R2Tg (R2Tg-1i) revealed precise junction sequences, perfectly matching the expected integration architecture at both 5\u0026prime; and 3\u0026prime; ends \u003cstrong\u003e(Fig. 4E)\u003c/strong\u003e. These results validate the fidelity of the integration events generated by the optimized R2Tg editor system.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo rigorously quantify integration efficiency, we implemented a droplet digital PCR (ddPCR) assay targeting the unique 3\u0026prime; and 5\u0026rsquo; integration junctions and normalized it against the endogenous \u003cem\u003eNbEF1\u0026alpha;\u003c/em\u003e locus \u003cstrong\u003e(Fig. 4F; Supplementary Fig. 4B)\u003c/strong\u003e. ddPCR analysis across 3 biological replicates showed that R2Tg editor comprising of the intronized R2Tg with the R4-HDV minimal payload in the 1-plasmid delivery format (R2Tg-1i) significantly exceeds efficiencies observed with all other R2 editor configuration tested in \u003cem\u003eN. benthamiana\u003c/em\u003e, reaching an average of approximately 1.2 integrated 3\u0026prime; junction copy per genome, while ~80-90% of those copies show full-length integration events \u003cstrong\u003e(Fig. 4G; Supplementary Fig. 4C, D)\u003c/strong\u003e. ddPCR at both junction sites further revealed that, across R2Tg editors tested in \u003cem\u003eN. benthamiana\u003c/em\u003e, the proportion of integrated copies containing both 3\u0026prime; and 5\u0026prime; junctions (full-length integration events) ranged from 73% to 87% \u003cstrong\u003e(Supplementary Fig. 4D\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDeep next-generation sequencing (NGS) of 3\u0026prime; junction amplicons revealed 72.14% of reads harboring integrated junctions exclusively in samples expressing the R4-HDV payload in combination with intronized R2Tg (R2Tg-1i), while samples containing the payload alone lacked any integration-specific reads \u003cstrong\u003e(Supplementary Fig. 4E-G)\u003c/strong\u003e. Alignment of sequencing reads confirmed high-fidelity, precise integration accompanied by a minority of indel variants comprising non-templated adenine additions and subtle shifts in endonuclease cleavage, occurring one to two nucleotides upstream of the canonical nick position, consistent with the established activity profile of R2 endonuclease \u003cstrong\u003e(Supplementary Fig. 4H)\u003c/strong\u003e. Collectively, these findings demonstrate that compact R2Tg payloads designs, with minimized homology arm length and 3\u0026prime; ribozyme substantially enhance R2Tg-mediated targeted integration efficiencies at the 25S rDNA site in plants and maintains precise, targeted DNA insertion.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo evaluate whether the optimized R2Tg system could mediate targeted DNA insertion at the 25s rDNA target site in the non-model crop \u003cem\u003eSolanum lycopersicum\u0026nbsp;\u003c/em\u003e(tomato), we tested the minimal R4-HDV mCherry payload delivered in either a one-plasmid or two-plasmid format together with the intronized R2Tg using a transient transformation method developed in our lab\u003csup\u003e53\u003c/sup\u003e \u003cstrong\u003e(Supplementary Fig. 5A)\u003c/strong\u003e. Confocal microscopy of seedling leaves transformed with these constructs revealed strong mCherry fluorescence in nuclei of cells co-expressing the intronized R2Tg, consistent with R2-mediated retrotransposition, while mTagBFP2 served as a transformation control \u003cstrong\u003e(Supplementary Fig. 5B)\u003c/strong\u003e. PCR assays specifically targeting integrated junctions at the molecular level, further validated precise insertion of the mCherry payload at the 25S rDNA locus with the expected 425 bp and 454 bp products detected for the 5\u0026prime; and 3\u0026prime; junctions, respectively, while no bands were obtained when only template is delivered without the R2 protein \u003cstrong\u003e(Supplementary Fig. 5C-E)\u003c/strong\u003e. These results demonstrate that the R2Tg editor achieves efficient targeted integration in \u003cem\u003eS. lycopersicum\u003c/em\u003e, establishing its functionality beyond model systems and highlighting its potential as a broadly applicable technology for precise genome engineering in crop species.\u003c/p\u003e\n\u003cp\u003eNext, we assessed whether the optimized R2Tg editor could support efficient targeted insertion of larger DNA sequences into the 25s rDNA loci of \u003cem\u003eN. benthamiana\u003c/em\u003e. For this, we constructed a 5 kb R4-HDV RUBY reporter cassette\u003csup\u003e54\u003c/sup\u003e \u003cstrong\u003e(Fig. 5A)\u003c/strong\u003e. The RUBY reporter encoded a polycistronic open reading frame linking 3 enzymes: cytochrome P450 CYP76AD1 that oxidizes tyrosine to L-DOPA and cyclo-DOPA; L-DOPA 4,5-dioxygenase (DODA) that converts L-DOPA to betalamic acid; and a glucosyltransferase that glycosylates betanidin to produce the red betalain pigment. The three enzymes were separated by 2A peptides to allow coordinated expression from a single promoter\u003csup\u003e54,55\u003c/sup\u003e. The R2Tg RUBY reporter was interrupted by a plant intron in the sense orientation, similar to the mCherry payload, but with the intron placed within the glucosyltransferase gene so that betalain production occurs only upon successful R2-mediated integration of the full 5 kb polycistronic payload. The system was delivered either as a single plasmid encoding both the intronized R2Tg and the RUBY payload, or as two separate plasmids carrying R2Tg and the payload separately \u003cstrong\u003e(Fig. 5A)\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExpression of the RUBY payload with intronized R2Tg in \u003cem\u003eN. benthamiana\u003c/em\u003e leaves produced visible betalain pigmentation within infiltrated areas, with stronger pigmentation observed for R2Tg-1i compared to R2Tg-2i, while leaves infiltrated with the RUBY payload alone, without the R2 protein, did not show pigmentation \u003cstrong\u003e(Supplementary Fig. 6B-D)\u003c/strong\u003e. Target-junction PCR using the genomic DNA extracted from infiltrated \u003cem\u003eN. benthamiana\u003c/em\u003e leaves confirmed successful integration, producing the expected 5\u0026rsquo; (425 bp) and 3\u0026rsquo; (454 bp) amplicons \u003cstrong\u003e(Fig. 5B)\u003c/strong\u003e. To quantify insertion efficiencies, we applied a droplet digital PCR (ddPCR) assay specific to the integrated 5\u0026prime; and 3\u0026prime; junctions \u003cstrong\u003e(Fig. 5C)\u003c/strong\u003e. ddPCR measurements showed that the one-plasmid system achieved the highest efficiencies, with an average of one copy per genome at the 5\u0026prime; junction and 1.5 copies per genome at the 3\u0026prime; junction, significantly outperforming the two-plasmid format which had 0.26 copies per genome at the 5\u0026prime; junction and 0.33 copies per genome at the 3\u0026prime; junction \u003cstrong\u003e(Fig. 5D, E)\u003c/strong\u003e. These findings validate that the intronized R2Tg system efficiently inserts multi-kilobase metabolic pathways into the 25S rDNA safe harbor of \u003cem\u003eN. benthamiana\u003c/em\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this work, we established targeted DNA insertion in plants using the \u003cem\u003eTaeniopygia guttata\u003c/em\u003e R2 non-LTR retroelement protein (R2Tg editor), which natively mediates site-specific integration through target-primed reverse transcription (TPRT)\u003csup\u003e27\u003c/sup\u003e. The R2 ribonucleoprotein complex achieves precise, multi-kilobase DNA integration by tightly coupling sequence-specific 25S rDNA recognition, bottom-strand nicking, primer hand-off, and reverse transcription of an RNA donor into cDNA\u003csup\u003e28\u003c/sup\u003e. Unlike CRISPR-Cas9 homology-directed repair (HDR), which depends on host DNA repair machinery to resolve a double-strand break using a donor DNA template, often resulting in low insertion efficiencies, the R2 mechanism bypasses the need for DSB induction entirely, avoiding competing end-joining pathways and the risk of large deletions or chromosomal rearrangements\u003csup\u003e15\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eR2 editors use RNA payloads in a similar manner to prime editing. In plants, prime editors are most effective for base substitutions and short insertions, and efficiency declines as insertion length and pegRNA design complexity increases. However, unlike prime editors that are constrained by insertion length, and guide RNA design complexity, R2 editors circumvent these constraints by directly synthesizing long cDNAs at the genomic target site during integration, leveraging the high processivity of their reverse transcriptase domains\u003csup\u003e27,28\u003c/sup\u003e. It remains to be demonstrated whether protein engineering can sufficiently increase processivity to support TPRT of even much larger payloads than the 5 kb demonstrated here. Another unique property of R2 editors is their targeting to the high-copy 25S rDNA locus, which embeds integrated genes in a specialized chromatin environment linked to stable transgene expression and minimal risk of disrupting essential host genes\u003csup\u003e56–58\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHere, we harnessed these unique capabilities to achieve targeted insertion in \u003cem\u003eNicotiana benthamiana\u003c/em\u003e and \u003cem\u003eArabidopsis thaliana\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;Solanum lycopersicum\u003c/em\u003e. We optimized efficiencies through engineered RNA payload architectures and improved R2 protein expression systems and validated high-efficiency targeted insertion via molecular detection of integrated junctions and functional expression of fluorescent reporters. Our results illuminated key determinants of integration efficiency in plants. The highest measured mCherry positive cell counts were achieved with the intronized R2Tg protein and R4-HDV minimal payload while using the 2-plasmid delivery approach. In contrast, the highest integrated 3’ junction copy numbers detected, approaching one copy per genome, was obtained with an intronized R2Tg and R4-HDV minimal payload in a 1-plasmid delivery. Additionally, this R2Tg editor configuration achieved targeted insertion in \u003cem\u003eSolanum lycopersicum\u003c/em\u003e expanding the tool’s capabilities to crop engineering. Notably, the R2Tg editor system also enabled efficient integration of a 5 kb metabolic pathway cassette, into the 25S rDNA safe-harbor locus, demonstrating that R2 retrotransposons can efficiently install complex metabolic pathways at a targeted genomic site.\u003c/p\u003e\n\u003cp\u003eThe divergence in mCherry signals and ddPCR detected copy numbers per genome may reflect differences in RNA payload processing and the completeness of integrated products, as R2-mediated integration can yield incomplete insertion events\u003csup\u003e29,30,32\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e59\u003c/sup\u003e and the fluorescence readout from the mCherry retrotransposition reporter requires successful intron splicing before integration. Incomplete integration events or unspliced integration events would not yield detectable mCherry signal but molecular-level detection copy number allows us to obtain a comprehensive picture of integration events at the 25S rDNA site. These observations are consistent with the possibility that R2Tg silencing and transcriptional interference in the all-in-one setup may affect efficient splicing and integration of the mCherry reporter. Intronization of R2Tg and separation of protein and payload cassettes may alleviate these constraints, improve the expression and splicing of payloads prior to integration at the 25S rDNA site to yield detectable mCherry reporter expression. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe observed improved mCherry signal after heat shock treatment of \u003cem\u003eN. benthamiana\u003c/em\u003e plants, expressing the R2 editor system after \u003cem\u003eAgrobacterium\u003c/em\u003e-mediated leaf infiltrated. This improvement in integration efficiency may be a result enzymatic kinetics or ribonucleoprotein (RNP) interactions required for integration. Other temperature-sensitive host processes may also contribute to this effect. In particular, heat stress has been shown to activate endogenous transposable elements in plants through global chromatin relaxation, reduction of DNA methylation, and suppression of RNA interference\u003csup\u003e46–48,50,60\u003c/sup\u003e. Such mechanisms may alleviate epigenetic repression of R2 editor components, thereby facilitating higher integration efficiencies. Disentangling these direct catalytic effects from host epigenetic changes will be important for understanding the mechanism at the molecular level.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn future studies, it will be important to assess whether transcriptional regulation of R2 protein and RNA template expression impose plant-specific constraints on R2 integration efficiencies not seen mammalian systems. To complement our work further, comprehensive genome-wide mapping is required to confirm the absence of off-target integrations, and long-term studies of stability and heritability will be crucial for translating this platform into breeding and crop improvement pipelines.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAlthough in this study we delivered the R2 system using \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e, which introduces DNA randomly into the genome, future efforts will focus on direct delivery of the R2 system as a ribonucleoprotein complex using nanoparticles or biolistics. This strategy would eliminate unintended T-DNA insertions, ensuring that only precise edits at the target rDNA locus are recovered. Such delivery advances will not only improve editing fidelity but also accelerate generation of clean, marker-free lines, reduce the need for laborious screening, and broaden applicability across plant species that are recalcitrant to \u003cem\u003eAgrobacterium\u003c/em\u003e. Together, these improvements could establish R2-based editors as a practical and widely deployable platform for precise, large-fragment DNA insertion in plants, an advancement with far-reaching implications for both fundamental genome studies and crop engineering.\u003c/p\u003e\n\u003cp\u003eTogether, these findings establish R2 retrotransposons as a distinctive and versatile addition to the plant genome engineering toolkit. By leveraging the precision of a native retroelement and augmenting it through improved protein expression platforms and engineered RNA payloads designs, this technology enables efficient, DSB-free integration of gene-sized cassettes into a defined genomic safe harbor locus in plants. With continued refinement, R2-based systems have the potential to complement and, in certain applications, surpass existing technologies for precise and targeted introduction of complex genetic traits into plant genomes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cu\u003ePlasmid construction\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eCoding sequences for 4 R2 non-LTR retrotransposases, R2Tg (\u003cem\u003eTaeniopygia guttata\u003c/em\u003e), R2Bm (\u003cem\u003eBombyx mori\u003c/em\u003e), R2Tg-opt (rationally engineered \u003cem\u003eT. guttata\u003c/em\u003e) and R2Za (\u003cem\u003eZonotrichia albicollis\u003c/em\u003e), were codon-optimized for \u003cem\u003eNicotiana benthamiana\u003c/em\u003e and \u003cem\u003eArabidopsis thaliana\u003c/em\u003e using GenSmart Codon Optimization (GenScript) and synthesized as double-stranded DNA fragments (Twist Bioscience). Each open reading frame contained an N-terminus bipartite SV40 nuclear localization signal (KRTADGSEFESPKKKRKV). 3 additional versions of the R2Tg protein were generated: i) R2Tg-intronic, which contains 2 \u003cem\u003eA. thaliana\u003c/em\u003e introns\u003csup\u003e45\u003c/sup\u003e; ii) R2Tg RT-dead, carrying D660A and D661A substitutions in the reverse-transcriptase domain; and iii) R2Tg EN-dead, carrying D1057A and D1070A substitutions in the endonuclease domain\u003csup\u003e29\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe retrotransposition reporter followed a classic GFP-intron design\u003csup\u003e42,43\u003c/sup\u003e. A Cauliflower mosaic virus (CaMV) 35S promoter and an \u003cem\u003eA. thaliana\u003c/em\u003e HSP18.2 terminator flank a reverse-oriented nested cassette that contains the Cassava vein mosaic virus (CsVMV) promoter, mCherry open reading frame interrupted by either the \u003cem\u003eSolanum tuberosum\u003c/em\u003e ST-LS1 intron or an \u003cem\u003eA. thaliana\u003c/em\u003e intron, and a CaMV t35S terminator\u003csup\u003e44\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e45\u003c/sup\u003e. Plasmids carrying R2 protein expression cassettes and donor DNA templates were combined with Loop Assembly using BsaI- and SapI-mediated Golden Gate reactions (NEB catalog numbers R0569S and E1601L respectively) and cloned into pCAMBIA backbones\u003csup\u003e61\u003c/sup\u003e. All plasmids were propagated in NEB Turbo \u003cem\u003eEscherichia coli\u003c/em\u003e competent cells (NEB catalog no. C2984I) and sequence-verified by Plasmidsaurus. A geminiviral replicon based on the Bean yellow dwarf virus (BeYDV) was used to transiently overexpress either the R2 proteins or the complete R2 editing system\u003csup\u003e41\u003c/sup\u003e. Constructs generated in this work can be found in Supplementary Table 1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003ePlant growth and transformation\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNicotiana benthamiana\u003c/em\u003e and \u003cem\u003eArabidopsis thaliana\u003c/em\u003e (Col-0) plants were grown in a Conviron controlled-environment chamber set to 22 °C during the day, 20 °C during the night with a 16 h light / 8 h dark photoperiod and 55 % relative humidity. 4-weeks old \u003cem\u003eN. benthamiana\u003c/em\u003e plants were used for leaf infiltrations for transient experiments. 4-weeks old \u003cem\u003eA. thaliana\u003c/em\u003e plant leaves were harvested for protoplast isolation.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e strain GV3101 (GOLDBIO catalogue no. CC-125) carrying the binary R2 construct, the p19 silencing-suppressor plasmid, and the pSOUP helper plasmid was cultured overnight at 28°C in 2YT medium (Invitrogen catalogue no. 12-780-052) supplemented with the appropriate antibiotics. Cells were collected and resuspended in induction media (10 mM MES pH 5.5, 10 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 200\u0026nbsp;mM Acetosyringone (PhytoTech Labs catalogue no. A104-5G) in DMSO), at OD\u003csub\u003e600\u003c/sub\u003e 0.4. Suspensions were allowed to shake at 70 rpm for 4 hours at room temperature. The \u003cem\u003eAgrobacterium\u003c/em\u003e suspension was infiltrated into the abaxial surface of fully expanded leaves using a 1 mL needleless syringe until the tissue was saturated. 3 leaves per plant were treated. Plants were returned to the growth chamber immediately and sampled 3 to 7 days post infiltration depending on the experiment.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eTransient Transformation of \u003cem\u003eSolanum lycopersicum\u003c/em\u003e (cv. Micro-Tom)\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eTransient transformation was performed using a protocol adapted from the VAST (Vacuum and Sonication-Assisted Transient expression) method\u003csup\u003e53\u003c/sup\u003e. Seeds of \u003cem\u003eS. lycopersicum\u003c/em\u003e (cv. Micro-Tom) were surface-sterilized with 33% (v/v) bleach containing 0.1% Tween-20 for 6 min, rinsed 4–5 times with sterile water, and germinated on ½ MS agar at 22 °C under a 16 h light/8 h dark photoperiod until cotyledons were fully expanded (~9 days). \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e strain GV3101 (GOLDBIO, CC-125) carrying the binary R2 construct, p19 silencing suppressor, and pSOUP helper plasmid was cultured in 2X-YT medium with antibiotics at 28 °C, 200 rpm, 2 days prior to co-cultivation. Cultures were pelleted (8,000 g, 5 min), resuspended in AB-MES induction medium (200 µM Acetosyringone, no antibiotics) to OD₆₀₀ = 0.3, and incubated overnight at 28 °C, 200 rpm. On the day of co-cultivation, cultures were pelleted again and resuspended in co-culture medium (1:1 AB-MES and ½ MS supplemented with 200 µM Acetosyringone) and adjusted to OD₆₀₀ = 0.3.\u003c/p\u003e\n\u003cp\u003eFor infection, 7–8 seedlings were immersed in 8 mL of \u003cem\u003eAgrobacterium\u003c/em\u003e suspension in 12 mL tubes, sonicated for 30 s (Branson CPX-952-116R), and subjected to three cycles of vacuum infiltration (5 min vacuum followed by rapid release per cycle) with gentle mixing between cycles. Seedlings were then transferred to six-well plates containing 4 mL of fresh \u003cem\u003eAgrobacterium\u003c/em\u003e suspension per well and co-cultivated for 2 days at 22 °C under a 16 h light/8 h dark cycle. Following co-cultivation, seedlings were washed three times with sterile water and transferred to ½ MS agar plates containing 100 µM timentin to suppress bacterial growth. Plants were maintained under standard growth conditions, and transient expression was assessed 3 days post-infection by confocal microscopy.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eConfocal microscopy\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eLeaf discs ~5 mm in diameter were collected from infiltrated \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaves using a sterile hole puncher, mounted in distilled water, and imaged immediately on a Leica Stellaris 8 confocal laser-scanning microscope (HC PL APO 20×/0.75 NA water-immersion objective) housed in the Biological Imaging Facility (BIF) at Caltech. The blue fluorescent protein (BFP) reporter was excited at 405 nm and its emission collected from 430 nm to 470 nm, while mCherry was excited at 561 nm and its emission collected from 580 nm to 630 nm. A single optical section was acquired for each field with the pinhole set to 1 Airy unit, using identical laser power, detector gain, and offset for all samples within an experiment. Images were exported from Leica LAS X v4.3 as 16-bit TIFF files. A 100 µm scale bar was added to exported images using ImageJ Fiji.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eProtein expression and detection\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eR2Tg, R2Bm, R2Tg-opt, and R2Za coding sequences were fused to an N-terminal YPet tag for detection and confirmation of nuclear localization via confocal microscopy. Complementarily, R2 proteins were fused to an N-terminal HiBiT peptide tag for higher throughput luminescence detection using the Nano-Glo® HiBiT Lytic Detection System. 3 days post infiltration (dpi), 3 infiltrated \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaves were harvested, flash-frozen in liquid nitrogen, and ground to a fine powder before extraction in lysis buffer (20 mM Tris-HCl pH 7.4 (ThermoFisher Scientific catalogue no. 15567027), 25 % glycerol (Sigma-Aldrich catalogue no. G5516), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 2 mM EDTA (Sigma-Aldrich catalogue no. E9884), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 250 mM sucrose (ThermoFisher scientific catalogue no. J65148.A1), 0.1 % PMSF (ThermoFisher Scientific catalogue no. 36978)). After centrifugation at 1500 × g for 10 min at 4 °C to pellet nuclei, the pellet was washed up to 5 times with wash buffer (20 mM Tris-HCl pH 7.4 (ThermoFisher Scientific catalogue no. 15567027), 25 % glycerol (Sigma-Aldrich catalogue no. G5516), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 0.2 % Triton X-100 (Sigma-Aldrich catalogue no. S5886)) and lysed in high-salt buffer (20 mM HEPES-KOH pH 7.9 (Sigma-Aldrich catalogue no. H3375), 2.5 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 100 mM NaCl (Sigma-Aldrich catalogue no. S5886), 20 % glycerol (Sigma-Aldrich catalogue no. G5516), 0.2 mM EDTA (Sigma-Aldrich catalogue no. E9884), 0.5 mM DTT (GOLDBIO catalogue no. DTT), one cOmplete Mini EDTA-free protease-inhibitor tablet (Sigma-Aldrich \u0026nbsp;catalogue no. 11836170001) per 10 mL). For HiBiT detection, 30 µL of nuclear lysate was combined with 30 µL Nano-Glo HiBiT Lytic reagent (Promega catalogue no. N3030) containing LgBiT and furimazine, and luminescence was recorded on a Tecan Spark plate reader with a 1000 millisecond integration time. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eIsolation and transfection of \u003cem\u003eNicotiana benthamiana\u003c/em\u003e and \u003cem\u003eArabidopsis thaliana\u003c/em\u003e protoplasts\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eMesophyll protoplasts were isolated from fully expanded leaves of 4-week-old \u003cem\u003eN. benthamiana\u003c/em\u003e or \u003cem\u003eA. thaliana\u003c/em\u003e (Col-0) following Yoo et al. (2007)\u003csup\u003e62\u003c/sup\u003e with minor modifications. On the day of the experiment, a fresh enzyme solution (20 mL) containing 20 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933), 0.4 M mannitol (Millipore Sigma catalogue no. 443907), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 1.5% (w⁄v) cellulase onozuka R-10 (Yakult Pharmaceutical IND. CO., LTD) and 0.4% (w⁄v) macerozyme R-10 (Yakult Pharmaceutical IND. CO., LTD) was prepared; the MES–mannitol–KCl base was pre-heated to 70°C for 4 min before the enzymes were dissolved. After cooling, 10 mM CaCl₂ (Sigma-Aldrich catalogue no. C7902) and 0.1% (w⁄v) BSA (Fisher bioreagents catalogue no. BP9703) were added, and the solution was passed through a 0.45 µm syringe filter into a Petri dish. \u003cem\u003eN. benthamiana\u003c/em\u003e leaf strips or \u003cem\u003eA. thaliana\u003c/em\u003e full leaves were incubated in this solution without agitation for 3 h at room temperature in the dark, with gentle swirling once per hour. The digest was diluted 1:1 with W5 solution (154 mM NaCl (Sigma-Aldrich catalogue no. S5886), 125 mM CaCl₂ (Sigma-Aldrich catalogue no. C7902), 5 mM KCl (Sigma-Aldrich catalogue no. P5405), 2 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)), passed through a 70 µm nylon mesh (Fisherbrand catalogue no. 22363548), and the flow-through was centrifuged at 200 × g for 2 min. The pellet was gently resuspended in 5 mL W5, chilled on ice for 30 min, then the supernatant was removed and the protoplasts were adjusted to 5 × 10⁵ cells mL⁻¹ in MMG solution (0.4 M mannitol (Millipore Sigma catalogue no. 443907), 15 mM MgCl₂ (Sigma-Aldrich catalogue no. M2393), 4 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)).\u003c/p\u003e\n\u003cp\u003eFor PEG-mediated transfection a 40 % PEG 4000 solution (Sigma-Aldrich catalogue no. 81240) was freshly prepared (4 g PEG 4000, 3 mL H₂O, 2.5 mL 0.8 M mannitol (Millipore Sigma catalogue no. 443907), 1 mL 1 M CaCl₂ (Sigma-Aldrich catalogue no. C7902)). In round-bottom Eppendorf tubes, 20 µg plasmid DNA (≤ 20 µL) was combined with 100 µL protoplast suspension, mixed gently, and 120 µL PEG solution was added. After 15 min at room temperature, 1 mL W5 was added and the mixture was centrifuged (50 × g, 2 min); the wash was repeated twice. The pellet was resuspended in WI solution (0.5 M mannitol (Millipore Sigma catalogue no. 443907), 20 mM KCl (Sigma-Aldrich catalogue no. P5405), 4 mM MES pH 5.7 (Sigma-Aldrich catalogue no. M2933)), transferred to BSA-blocked well plates (0.1 % BSA (Fisher bioreagents catalogue no. BP9703)), and incubated at room temperature in the dark for 24-72 h before downstream analyses.After incubation in the dark at room temperature for 24 to 72 h, transformed protoplasts were used for confocal imaging and/or flow cytometry. In a modified protocol, protoplasts were extracted from 5-week-old \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaves 7 days after \u003cem\u003eAgrobacterium\u003c/em\u003e infiltration and analyzed directly by flow cytometry.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eFlow cytometry analysis\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eProtoplasts were analyzed on a CytoFLEX flow cytometer (Beckman Coulter) provided by the Caltech Flow Cytometry and Cell Sorting Facility. Live protoplast gates were defined using fluorescein diacetate (FDA) staining (Fisher scientific catalogue no. F1303), detected in the FITC channel. BFP was excited with a 405 nm laser and detected in the 450/45 nm channel, while mCherry was excited with a 561 nm laser and detected in the 610/20 nm channel. Data was exported as FCS files and analyzed using FloReada for gating, quantification, and visualization.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eAmplicon PCR and sanger sequencing\u0026nbsp;\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eGenomic DNA (gDNA) was extracted from \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaf tissue infiltrated with R2 constructs using the DNeasy Plant Mini Kit (Qiagen catalogue no. 69106) according to the manufacturer’s protocol. For 3′-junction analysis, 25 µL PCRs were assembled with 1× Phusion High-Fidelity Master Mix (NEB catalogue no. M0531S), 0.5 µM each primer 2F and 2R, and 20 ng gDNA. Cycling conditions: 98 °C for 30 s; 35 cycles of 98 °C for 10 s, 60 °C for 20 s, 72 °C for 30 s; final extension 72 °C for 2 min. The expected 450 bp amplicons were verified on a 1 % agarose gel, purified with the QIAquick Gel Extraction Kit (Qiagen catalogue no. 28704) and submitted to Laragen Inc. for Sanger sequencing. Chromatograms were aligned to reference sequences in SnapGene v6.2 to confirm precise junction formation.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eNext generation sequencing (NGS)\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eThe 450 bp 3′-junction PCR products (primer pair 2F/2R) were gel-purified with the QIAquick Gel Extraction Kit (Qiagen catalogue no. 28704) and eluted in 35 µL nuclease-free water. DNA concentration was measured on a NanoDrop 2000 spectrophotometer, and each amplicon was adjusted to 50 ng/µL, sealed in Eppendorf microcentrifuge tubes, and shipped overnight to the Massachusetts General Hospital CCIB DNA Core (Harvard) for Illumina sequencing. The core performed NGS and provided (i) de novo-assembled contigs representing variants present at \u0026gt; 1 % abundance and (ii) the complete demultiplexed FASTQ files.\u003c/p\u003e\n\u003cp\u003eIn Galaxy, paired-end reads were processed with DADA2: adapters were trimmed, reads were quality-filtered, and overlapping pairs were merged. Merged reads were aligned to a custom reference comprising the payload integrated at the 25S rDNA locus using Bowtie2 v2.5.0 to capture low-frequency variants (\u0026lt; 1 %). The subset of sequence variants mapping perfectly to the insertion junction were imported into Geneious Prime 2024.1, aligned to the same reference and the multiple alignment was exported as FASTA. Alignment visualization was carried out in Google Collab with Biopython 1.81, pandas 2.2.2, and matplotlib 3.7.3; the resulting grid plot of base-by-base variation was exported as high-resolution PNG files for publication.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eDroplet digital PCR (ddPCR) for detecting targeted insertion efficiency\u0026nbsp;\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eGenomic DNA was extracted from 100 mg of \u003cem\u003eNicotiana benthamiana\u003c/em\u003e leaf tissue collected from the \u003cem\u003eAgrobacterium\u003c/em\u003e infiltration site 7 days after infiltration using the DNeasy Plant Mini Kit (Qiagen). Genomic DNA was predigested with FastDigest KpnI (Thermo Fisher Scientific, catalog no. FD0524) according to the manufacturer’s protocol. Duplex 22 µl ddPCR reactions were prepared by mixing 11 µl of ddPCR Supermix for Probes (no dUTP; Bio-Rad, catalog no. 1863024), 900 nM forward and reverse primers for the target and reference loci, 250 nM FAM-labeled probe targeting the \u003cem\u003eNbEF1α\u003c/em\u003e reference gene, 250 nM HEX-labeled probe targeting the integrated R2 junction, and 50 ng of predigested gDNA. Oligonucleotide sequences were synthesized by Integrated DNA Technologies (IDT) and are listed in Supplementary Table 2. Reaction mixtures were loaded into a DG8 cartridge (Bio-Rad, catalog no. 1864007) along with 70 µl of droplet generation oil (Bio-Rad, catalog no. 1863005), and droplets were generated using a Bio-Rad QX200 Droplet Generator. Following droplet generation, 40 µl of emulsion was transferred to a 96-well plate, heat-sealed with pierceable foil, and thermal-cycled under the manufacturer’s conditions with an annealing/extension temperature of 55.\u003csup\u003eo\u0026nbsp;\u003c/sup\u003eC for 3’ junction and 60\u003csup\u003eo\u003c/sup\u003eC for 5’ junction. Droplets were read using a Bio-Rad QX200 Droplet Reader. Data were analyzed using QX Manager software (Bio-Rad). All Bio-Rad ddPCR equipment was provided by the CLARITY, Optogenetics and Vector Engineering Research (CLOVER) Center at Caltech. Insertion efficiency was calculated as the ratio of HEX-labeled target copy number to the FAM-labeled reference gene (\u003cem\u003eNbEF1α\u003c/em\u003e) copy number assuming 4 copies of the target sequence.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data are included either in the manuscript or supplementary files.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthors declare no conflict of interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: KTM, GSD\u003c/p\u003e\n\u003cp\u003eMethodology: KTM, YW, TMO, AS, EL, GSD\u003c/p\u003e\n\u003cp\u003eData Acquisition: KTM, YW, TMO, AS, EL\u003c/p\u003e\n\u003cp\u003eSupervision: KTM, TMO, GSD\u003c/p\u003e\n\u003cp\u003eWriting: KTM, GSD\u003c/p\u003e\n\u003cp\u003eFunding Acquisition: GSD\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Professor Kathleen Collins (University of California, Berkeley) for valuable discussions and insights on experimental design. Imaging was performed in the Biological Imaging Facility, with the support of the Caltech Beckman Institute and the Arnold and Mabel Beckman Foundation. We acknowledge the Beckman Institute CLOVER Center for providing access to droplet digital PCR instrumentation and support with experiments. Flow cytometry data used in this manuscript was generated in the Caltech Flow Cytometry \u0026amp; Cell Sorting Facility (Research Resource Identifier number is RRID: SCR_025087). Schematics were created with BioRender.com.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Caltech startup funds, Caltech Space-Health Innovation Fund, Henry Luce Foundation, and Shurl and Kay Curci Foundation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDasgupta, S. \u0026amp; Robinson, E. J. Z. Attributing changes in food insecurity to a changing climate. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 4709 (2022).\u003c/li\u003e\n\u003cli\u003eBradbury, A. \u003cem\u003eet al.\u003c/em\u003e Integrating genome editing with omics, artificial intelligence, and advanced farming technologies to increase crop productivity. \u003cem\u003ePlant Commun.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 101386 (2025).\u003c/li\u003e\n\u003cli\u003eScheben, A., Yuan, Y. \u0026amp; Edwards, D. Advances in genomics for adapting crops to climate change. \u003cem\u003eCurr. Plant Biol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 2\u0026ndash;10 (2016).\u003c/li\u003e\n\u003cli\u003eTzfira, T. \u0026amp; Citovsky, V. Agrobacterium-mediated genetic transformation of plants: biology and biotechnology. \u003cem\u003eCurr. Opin. Biotechnol.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 147\u0026ndash;154 (2006).\u003c/li\u003e\n\u003cli\u003eKim, S.-I. \u0026amp; and Gelvin, S. B. Genome-wide analysis of Agrobacterium T-DNA integration sites in the Arabidopsis genome generated under non-selective conditions. \u003cem\u003ePlant J.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 779\u0026ndash;791 (2007).\u003c/li\u003e\n\u003cli\u003eRajeevkumar, S., Anunanthini, P. \u0026amp; Sathishkumar, R. Epigenetic silencing in transgenic plants. \u003cem\u003eFront. Plant Sci.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 693 (2015).\u003c/li\u003e\n\u003cli\u003eGupta, M. \u0026amp; Ram, R. Development of Genetically Modified Agronomic Crops. in \u003cem\u003eThe GMO Handbook: Genetically Modified Animals, Microbes, and Plants in Biotechnology\u003c/em\u003e (ed. Parekh, S. R.) 219\u0026ndash;241 (Humana Press, Totowa, NJ, 2004). doi:10.1007/978-1-59259-801-4_8.\u003c/li\u003e\n\u003cli\u003eDong, O. X. \u0026amp; Ronald, P. C. Targeted DNA insertion in plants. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2004834117 (2021).\u003c/li\u003e\n\u003cli\u003eKhan, A. \u0026amp; Lister, R. Synthetic gene circuits in plants: recent advances and challenges. \u003cem\u003eQuant. Plant Biol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e6 (2025).\u003c/li\u003e\n\u003cli\u003eChen, W. \u0026amp; Ow, D. W. Precise, flexible and affordable gene stacking for crop improvement. \u003cem\u003eBioengineered\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 451\u0026ndash;456 (2017).\u003c/li\u003e\n\u003cli\u003eJinek, M. \u003cem\u003eet al.\u003c/em\u003e A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e337\u003c/strong\u003e, 816\u0026ndash;821 (2012).\u003c/li\u003e\n\u003cli\u003eCong, L. \u003cem\u003eet al.\u003c/em\u003e Multiplex Genome Engineering Using CRISPR/Cas Systems. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e339\u003c/strong\u003e, 819\u0026ndash;823 (2013).\u003c/li\u003e\n\u003cli\u003eDong, O. X. \u003cem\u003eet al.\u003c/em\u003e Marker-free carotenoid-enriched rice generated through targeted gene insertion using CRISPR-Cas9. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1178 (2020).\u003c/li\u003e\n\u003cli\u003eLu, Y. \u003cem\u003eet al.\u003c/em\u003e Targeted, efficient sequence insertion and replacement in rice. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 1402\u0026ndash;1407 (2020).\u003c/li\u003e\n\u003cli\u003eKosicki, M., Tomberg, K. \u0026amp; Bradley, A. Repair of double-strand breaks induced by CRISPR\u0026ndash;Cas9 leads to large deletions and complex rearrangements. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 765\u0026ndash;771 (2018).\u003c/li\u003e\n\u003cli\u003eZong, Y. \u003cem\u003eet al.\u003c/em\u003e An engineered prime editor with enhanced editing efficiency in plants. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 1394\u0026ndash;1402 (2022).\u003c/li\u003e\n\u003cli\u003eAnzalone, A. V. \u003cem\u003eet al.\u003c/em\u003e Programmable deletion, replacement, integration and inversion of large DNA sequences with twin prime editing. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 731\u0026ndash;740 (2022).\u003c/li\u003e\n\u003cli\u003eJiang, Y.-Y. \u003cem\u003eet al.\u003c/em\u003e Prime editing efficiently generates W542L and S621I double mutations in two ALS genes in maize. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 257 (2020).\u003c/li\u003e\n\u003cli\u003eSun, C. \u003cem\u003eet al.\u003c/em\u003e Precise integration of large DNA sequences in plant genomes using PrimeRoot editors. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 316\u0026ndash;327 (2024).\u003c/li\u003e\n\u003cli\u003eWells, J. N. \u0026amp; Feschotte, C. A Field Guide to Eukaryotic Transposable Elements. \u003cem\u003eAnnu. Rev. Genet.\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 539\u0026ndash;561 (2020).\u003c/li\u003e\n\u003cli\u003eHickman, A. B. \u0026amp; Dyda, F. DNA Transposition at Work. \u003cem\u003eChem. Rev.\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 12758\u0026ndash;12784 (2016).\u003c/li\u003e\n\u003cli\u003eGelsinger, D. R. \u003cem\u003eet al.\u003c/em\u003e Bacterial genome engineering using CRISPR-associated transposases. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 752\u0026ndash;790 (2024).\u003c/li\u003e\n\u003cli\u003eLampe, G. D. \u003cem\u003eet al.\u003c/em\u003e Targeted DNA integration in human cells without double-strand breaks using CRISPR-associated transposases. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e (2023) doi:10.1038/s41587-023-01748-1.\u003c/li\u003e\n\u003cli\u003eGeorge, J. T. \u003cem\u003eet al.\u003c/em\u003e Mechanism of target site selection by type V-K CRISPR-associated transposases. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e382\u003c/strong\u003e, eadj8543 (2023).\u003c/li\u003e\n\u003cli\u003eLiu, P. \u003cem\u003eet al.\u003c/em\u003e Transposase-assisted target-site integration for efficient plant genome engineering. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e631\u003c/strong\u003e, 593\u0026ndash;600 (2024).\u003c/li\u003e\n\u003cli\u003eFujiwara, H. Site-specific non-LTR retrotransposons. \u003cem\u003eMicrobiol. Spectr.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, MDNA3-0001\u0026ndash;2014 (2015).\u003c/li\u003e\n\u003cli\u003eWilkinson, M. E., Frangieh, C. J., Macrae, R. K. \u0026amp; Zhang, F. Structure of the R2 non-LTR retrotransposon initiating target-primed reverse transcription. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e380\u003c/strong\u003e, 301\u0026ndash;308 (2023).\u003c/li\u003e\n\u003cli\u003eThawani, A. \u003cem\u003eet al.\u003c/em\u003e Structures of vertebrate R2 retrotransposon complexes during target-primed reverse transcription and after second-strand nicking. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, eadu5533 (2025).\u003c/li\u003e\n\u003cli\u003eZhang, X. \u003cem\u003eet al.\u003c/em\u003e Harnessing eukaryotic retroelement proteins for transgene insertion into human safe-harbor loci. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e (2024) doi:10.1038/s41587-024-02137-y.\u003c/li\u003e\n\u003cli\u003eChen, Y. \u003cem\u003eet al.\u003c/em\u003e All-RNA-mediated targeted gene integration in mammalian cells with rationally engineered R2 retrotransposons. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e187\u003c/strong\u003e, 4674-4689.e18 (2024).\u003c/li\u003e\n\u003cli\u003eFell, C. W. \u003cem\u003eet al.\u003c/em\u003e Reprogramming site-specific retrotransposon activity to new DNA sites. \u003cem\u003eNature\u003c/em\u003e 1\u0026ndash;10 (2025) doi:10.1038/s41586-025-08877-4.\u003c/li\u003e\n\u003cli\u003eEdmonds, K. K. \u003cem\u003eet al.\u003c/em\u003e Structure and biochemistry-guided engineering of an all-RNA system for DNA insertion with R2 retrotransposons. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 6079 (2025).\u003c/li\u003e\n\u003cli\u003eFultz, D., McKinlay, A., Enganti, R. \u0026amp; Pikaard, C. S. Sequence and epigenetic landscapes of active and silent nucleolus organizer regions in \u003cem\u003eArabidopsis\u003c/em\u003e. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, eadj4509 (2023).\u003c/li\u003e\n\u003cli\u003eChen, W. \u003cem\u003eet al.\u003c/em\u003e The complete genome assembly of Nicotiana benthamiana reveals the genetic and epigenetic landscape of centromeres. \u003cem\u003eNat. Plants\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1928\u0026ndash;1943 (2024).\u003c/li\u003e\n\u003cli\u003eShirasawa, K. \u0026amp; Ariizumi, T. Near-complete genome assembly of tomato (Solanum lycopersicum) cultivar Micro-Tom. \u003cem\u003ePlant Biotechnol.\u003c/em\u003e \u003cstrong\u003eadvpub\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eZhou, X. \u003cem\u003eet al.\u003c/em\u003e Tandem Repeat Detection Using Next-Generation Mapping and its Application in Detecting rDNA Copy Number Variation in Plant and Human Genomes.\u003c/li\u003e\n\u003cli\u003eTulpov\u0026aacute;, Z. \u003cem\u003eet al.\u003c/em\u003e Fine structure and transcription dynamics of bread wheat ribosomal DNA loci deciphered by a multi-omics approach. \u003cem\u003ePlant Genome\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e20191 (2022).\u003c/li\u003e\n\u003cli\u003eHori, Y., Engel, C. \u0026amp; Kobayashi, T. Regulation of ribosomal RNA gene copy number, transcription and nucleolus organization in eukaryotes. \u003cem\u003eNat. Rev. Mol. Cell Biol.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 414\u0026ndash;429 (2023).\u003c/li\u003e\n\u003cli\u003eLayat, E., S\u0026aacute;ez-V\u0026aacute;squez, J. \u0026amp; Tourmente, S. Regulation of Pol I-Transcribed 45S rDNA and Pol III-Transcribed 5S rDNA in Arabidopsis. \u003cem\u003ePlant Cell Physiol.\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 267\u0026ndash;276 (2012).\u003c/li\u003e\n\u003cli\u003eChristensen, S. M. \u0026amp; Eickbush, T. H. R2 Target-Primed Reverse Transcription: Ordered Cleavage and Polymerization Steps by Protein Subunits Asymmetrically Bound to the Target DNA. \u003cem\u003eMol. Cell. Biol.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 6617\u0026ndash;6628 (2005).\u003c/li\u003e\n\u003cli\u003eBaltes, N. J., Gil-Humanes, J., Cermak, T., Atkins, P. A. \u0026amp; Voytas, D. F. DNA Replicons for Plant Genome Engineering. \u003cem\u003ePlant Cell\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 151\u0026ndash;163 (2014).\u003c/li\u003e\n\u003cli\u003eOstertag, E. M., Luning Prak, E. T., DeBerardinis, R. J., Moran, J. V. \u0026amp; Kazazian, H. H. Determination of L1 retrotransposition kinetics in cultured cells. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1418\u0026ndash;1423 (2000).\u003c/li\u003e\n\u003cli\u003eTao, J., Wang, Q., Mendez-Dorantes, C., Burns, K. H. \u0026amp; Chiarle, R. Frequency and mechanisms of LINE-1 retrotransposon insertions at CRISPR/Cas9 sites. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 3685 (2022).\u003c/li\u003e\n\u003cli\u003eJohansen, I. E. \u0026amp; Lund, O. S. Insertion of Introns: A Strategy to Facilitate Assembly of Infectious Full Length Clones. in \u003cem\u003ePlant Virology Protocols\u003c/em\u003e (eds. Foster, G. D., Johansen, I. E., Hong, Y. \u0026amp; Nagy, P. D.) vol. 451 535\u0026ndash;544 (Humana Press, Totowa, NJ, 2008).\u003c/li\u003e\n\u003cli\u003eGr\u0026uuml;tzner, R. \u003cem\u003eet al.\u003c/em\u003e High-efficiency genome editing in plants mediated by a Cas9 gene containing multiple introns. \u003cem\u003ePlant Commun.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 100135 (2021).\u003c/li\u003e\n\u003cli\u003eBond, D. M. \u0026amp; Baulcombe, D. C. Epigenetic transitions leading to heritable, RNA-mediated de novo silencing in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 917\u0026ndash;922 (2015).\u003c/li\u003e\n\u003cli\u003eSigman, M. J. \u0026amp; Slotkin, R. K. The First Rule of Plant Transposable Element Silencing: Location, Location, Location. \u003cem\u003ePlant Cell\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 304\u0026ndash;313 (2016).\u003c/li\u003e\n\u003cli\u003eNuthikattu, S. \u003cem\u003eet al.\u003c/em\u003e The Initiation of Epigenetic Silencing of Active Transposable Elements Is Triggered by RDR6 and 21-22 Nucleotide Small Interfering RNAs1[W][OA]. \u003cem\u003ePlant Physiol.\u003c/em\u003e \u003cstrong\u003e162\u003c/strong\u003e, 116\u0026ndash;131 (2013).\u003c/li\u003e\n\u003cli\u003eBurgess, D., Li, H., Zhao, M., Kim, S. Y. \u0026amp; Lisch, D. Silencing of Mutator Elements in Maize Involves Distinct Populations of Small RNAs and Distinct Patterns of DNA Methylation. \u003cem\u003eGenetics\u003c/em\u003e \u003cstrong\u003e215\u003c/strong\u003e, 379\u0026ndash;391 (2020).\u003c/li\u003e\n\u003cli\u003eHedtke, B. \u0026amp; Grimm, B. Silencing of a plant gene by transcriptional interference. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 3739\u0026ndash;3746 (2009).\u003c/li\u003e\n\u003cli\u003eOmelina, E. S., Ivankin, A. V., Letiagina, A. E. \u0026amp; Pindyurin, A. V. Optimized PCR conditions minimizing the formation of chimeric DNA molecules from MPRA plasmid libraries. \u003cem\u003eBMC Genomics\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 536 (2019).\u003c/li\u003e\n\u003cli\u003eWichłacz, A., Łęgiewicz, M. \u0026amp; Ciesiołka, J. Generating in vitro transcripts with homogenous 3\u0026prime; ends using trans-acting antigenomic delta ribozyme. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, e39 (2004).\u003c/li\u003e\n\u003cli\u003eLi, E. \u003cem\u003eet al.\u003c/em\u003e Vacuum and sonication treatment enables efficient transient gene expression in various monocot and eudicot plant seedlings. 2025.03.13.642903 Preprint at https://doi.org/10.1101/2025.03.13.642903 (2025).\u003c/li\u003e\n\u003cli\u003eHe, Y., Zhang, T., Sun, H., Zhan, H. \u0026amp; Zhao, Y. A reporter for noninvasively monitoring gene expression and plant transformation. \u003cem\u003eHortic. Res.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 152 (2020).\u003c/li\u003e\n\u003cli\u003eYuan, G. \u003cem\u003eet al.\u003c/em\u003e Split selectable marker systems utilizing inteins facilitate gene stacking in plants. \u003cem\u003eCommun. Biol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 567 (2023).\u003c/li\u003e\n\u003cli\u003eSchenkwein, D., Afzal, S., Nousiainen, A., Schmidt, M. \u0026amp; Yl\u0026auml;-Herttuala, S. Efficient Nuclease-Directed Integration of Lentivirus Vectors into the Human Ribosomal DNA Locus. \u003cem\u003eMol. Ther.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1858\u0026ndash;1875 (2020).\u003c/li\u003e\n\u003cli\u003eMcStay, B. Nucleolar organizer regions: genomic \u0026lsquo;dark matter\u0026rsquo; requiring illumination. \u003cem\u003eGenes Dev.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1598\u0026ndash;1610 (2016).\u003c/li\u003e\n\u003cli\u003eLisowski, L. \u003cem\u003eet al.\u003c/em\u003e Ribosomal DNA Integrating rAAV-rDNA Vectors Allow for Stable Transgene Expression. \u003cem\u003eMol. Ther.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1912 (2012).\u003c/li\u003e\n\u003cli\u003eRodr\u0026iacute;guez-Vargas, A. \u0026amp; Collins, K. Distinct and overlapping RNA determinants for binding and target-primed reverse transcription by Bombyx mori R2 retrotransposon protein. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 6571\u0026ndash;6585 (2024).\u003c/li\u003e\n\u003cli\u003eNguyen, V. H., Mittelsten Scheid, O. \u0026amp; Gutzat, R. Heat stress response and transposon control in plant shoot stem cells. \u003cem\u003ePlant Physiol.\u003c/em\u003e \u003cstrong\u003e197\u003c/strong\u003e, kiaf110 (2025).\u003c/li\u003e\n\u003cli\u003ePollak, B. \u003cem\u003eet al.\u003c/em\u003e Loop assembly: a simple and open system for recursive fabrication of DNA circuits. \u003cem\u003eNew Phytol.\u003c/em\u003e \u003cstrong\u003e222\u003c/strong\u003e, 628\u0026ndash;640 (2019).\u003c/li\u003e\n\u003cli\u003eYoo, S.-D., Cho, Y.-H. \u0026amp; Sheen, J. Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1565\u0026ndash;1572 (2007).\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"R2 retrotransposon, targeted DNA integration, safe harbor site, plant bioengineering, synthetic biology","lastPublishedDoi":"10.21203/rs.3.rs-7436428/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7436428/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Precise, targeted insertion of multi-kilobase DNA sequences is critical for studying gene function, ensuring robust transgene expression, and stacking traits in crops, but remains difficult to achieve. Existing targeted insertion methods in plants relying on programmable nucleases are inefficient and can generate unwanted mutations. Newer technologies based on prime editors, transposases, and site-specific recombinases extend capabilities but remain constrained with low efficiencies, off-target integration, silencing, or limited cargo DNA size. R2 non-long terminal repeat (non-LTR) retrotransposons integrate via target-primed reverse transcription specifically targeting the 25S ribosomal DNA multicopy site and enabling double-strand-break-free installation of gene-sized DNA sequences. Here, we adapted the avian Taeniopygia guttata R2 protein (R2Tg) for targeted DNA insertion into plant genomes and defined design rules for efficient integration and functional protein expression from the 25S rDNA locus. We engineered R2Tg expression cassettes and RNA payloads carrying an intron-disrupted mCherry retrotransposition reporter with length-optimized rDNA homology arms. We also optimized the R2 editor delivery format and incubation conditions. In Nicotiana benthamiana leaves, Arabidopsis thaliana protoplasts, and Solanum lycopersicum seedlings, the optimized R2Tg editor system achieved targeted insertion with efficiencies up to 24% for 2-5kb size payloads, which were carefully validated at the molecular level. This work establishes a compact, ribonucleoprotein-based platform for targeted DNA insertion into plant genomes, leveraging a multicopy genomic safe harbor site to enable efficient multi-kilobase precise gene addition.","manuscriptTitle":"Optimized R2 Retroelement Complexes Enable Precise and Efficient DNA Insertion into Plant Genomes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-26 06:20:31","doi":"10.21203/rs.3.rs-7436428/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-biotechnology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nbt","sideBox":"Learn more about [Nature Biotechnology](http://www.nature.com/nbt/)","snPcode":"","submissionUrl":"","title":"Nature Biotechnology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"90c6f737-6911-4271-8685-5646244e563e","owner":[],"postedDate":"August 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":53658709,"name":"Biological sciences/Biotechnology/Plant biotechnology/Molecular engineering in plants"},{"id":53658710,"name":"Biological sciences/Plant sciences/Plant molecular biology"}],"tags":[],"updatedAt":"2026-05-08T18:30:54+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-26 06:20:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7436428","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7436428","identity":"rs-7436428","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}