Chinese Researchers Unveil SmartPE System to Overcome Gene Editing Bottlenecks in Dicot Crops

While prime editing has demonstrated high utility in model organisms and key cereal grasses over recent years, its adoption in dicotyledonous crops has faced structural hurdles. Low baseline editing efficiencies in broadleaf plants have frequently rendered the technology too labor-intensive or unpredictable for commercial breeding pipelines. By refining the machinery that delivers and executes these edits, the team behind SmartPE aims to bring major vegetable, legume, and oilseed crops into parity with staple grains that have historically responded more readily to genetic engineering tools.

What happened

According to reporting by phys.org, agricultural scientists in China engineered the SmartPE architecture to address the stubborn inefficiency that has constrained prime editing in dicotyledonous species. In standard plant prime editing, the molecular machinery often suffers from poor expression levels, sub-optimal reverse transcription dynamics, or rapid degradation of the guide RNAs within plant cell environments. SmartPE modifies these functional components to stabilize the editing complex and improve its interaction with plant genomic targets.

During validation trials, the research group deployed the system across several distinct plant models and food crops. The tests focused on crops of high economic importance, primarily tomatoes, soybeans, and rapeseed. In these plants, traditional prime editors frequently register single-digit or near-negligible percentage rates of successful editing at target sites, compelling breeders to regenerate thousands of tissue samples to secure a single desired plant line.

By adjusting the regulatory sequences driving the editor and refining the core prime editing cassette, SmartPE demonstrated marked improvements in on-target mutation yields across the evaluated species. The researchers applied the tool to introduce defined sequence alterations, confirming that the enhanced framework maintains the characteristic accuracy of prime editing while significantly raising the proportion of successfully edited cells. The system functioned effectively across multiple target loci within these crops, demonstrating that the improvement is not limited to a single favorable genomic site.

Modern agriculture faces concurrent pressures from climate volatility, shifting pest pressures, and rising global demand for plant-based proteins and vegetable oils. Dicotyledonous crops supply the majority of the world's plant proteins through legumes such as soybeans, critical cooking oils and biofuels via brassicas such as rapeseed, and essential micronutrients through horticultural staples like tomatoes. Accelerating the pace at which breeders can adapt these species is vital to stabilizing agricultural output.

Until now, precision breeding in these crops has relied heavily on standard CRISPR-Cas9 or base editors. Conventional CRISPR-Cas9 functions primarily by cleaving both strands of the DNA helix, triggering blunt cellular repair pathways that typically produce random insertions or deletions. While effective for knocking out unwanted genes, it cannot easily rewrite specific sequences to create superior alleles. Base editors, introduced in 2016, permit direct single-nucleotide conversions, but they are biochemically constrained: they can only convert certain nucleotides into others (such as cytosine to thymine, or adenine to guanine) and frequently generate bystander mutations if unintended target bases lie within the editing window.

Prime editing resolves both issues by acting like a molecular word processor, capable of searching for a genomic address and typing in precise replacements of varying lengths. However, because its utility in dicots was severely depressed by low recovery rates, agricultural firms and academic labs often relegated prime editing to monocot cereals like rice and maize. By making prime editing commercially viable in dicots, SmartPE opens the door to creating non-transgenic crops with optimized flowering times, enhanced drought tolerance, customized seed-oil profiles, and built-in resistance to pathogens without leaving foreign bacterial genes in the plant genome.

The background

The trajectory of molecular plant breeding was transformed in 2012 following the characterization of the CRISPR-Cas9 immune mechanism from bacteria as a programmable genome-editing tool. Within five years, laboratories worldwide were deploying Cas9 nucleases to edit livestock, human cell lines, and agricultural cultivars. Yet the reliance of classical CRISPR on double-stranded breaks presented fundamental risks, including off-target chromosome rearrangements and cellular toxicity.

In October 2019, a research team led by David Liu at the Broad Institute of MIT and Harvard unveiled prime editing. The original system coupled an engineered Cas9 nickase — which cuts only one strand of the DNA double helix — to an engineered reverse transcriptase enzyme. The construct is guided by a prime editing guide RNA (pegRNA), an extended RNA molecule that not only identifies the target genomic sequence but also carries the template sequence specifying the desired edit. Once the nickase creates a single-strand incision, the reverse transcriptase reads the pegRNA template and directly writes the corrected genetic code into the DNA.

Plant scientists quickly sought to adapt Liu’s human-optimized prime editors for agriculture. The earliest plant prime editors, reported in 2020 by institutions including the Chinese Academy of Sciences, functioned reasonably well in rice and wheat. Monocots possess cellular environments, promoter dynamics, and tissue regeneration characteristics that supported modest but functional prime editing frequencies, typically between 5% and 20%.

Dicots, however, proved markedly less hospitable. Broadleaf species exhibit distinct chromatin architecture, variable codon preferences, and different endogenous DNA repair enzymes. When early plant prime editors were applied to species such as *Arabidopsis thaliana*, tobacco, tomato, and soybean, editing efficiencies routinely lingered below 2%, and often beneath 0.1%. While specialized modifications — such as melting-temperature optimizations, structured RNA motifs, and dual-pegRNA configurations — incrementally raised performance in laboratory settings, a dependable, universally robust system for dicot crops remained an elusive goal for geneticists.

Reaction

The development of SmartPE is expected to generate significant interest across public and commercial agricultural biotechnology sectors, particularly among research institutes focused on crop resilience and seed development.

Academic plant geneticists have long argued that the imbalance between monocot and dicot editing tools was skewing basic plant research toward cereal models at the expense of legume and horticultural biology. The introduction of a dedicated dicot tool provides university departments and international agronomy consortia, such as the Consultative Group on International Agricultural Research (CGIAR), with a workable protocol for functional genomics in complex broadleaf species.

Commercial seed developers in North America, South America, and East Asia are also monitoring technical improvements in prime editing closely. Unlike early-generation genetically modified organisms (GMOs) that incorporated foreign DNA sequences from bacteria or viruses, prime-edited crops frequently fall into regulatory categories that exempt them from costly transgenic oversight. Regulatory authorities in jurisdictions such as the United States, through the Department of Agriculture’s SECURE rule, and China, under guidelines issued by the Ministry of Agriculture and Rural Affairs in 2022, have created streamlined evaluation pathways for gene-edited crops lacking foreign genetic inserts. A more efficient editing platform lowers the cost and time required to generate commercial-grade parental lines under these modernized frameworks.

What we don't know yet

While the initial demonstration of SmartPE confirms enhanced performance across tomato, soybean, and rapeseed tissues, several critical scientific and commercial parameters remain unverified in the public domain. The available reporting does not disclose the precise quantitative efficiencies achieved by the system, leaving it unclear whether average editing frequencies have reached double-digit percentages across all tested crops or if they remain concentrated at specific, highly accessible target genes.

Furthermore, the long-term fidelity of SmartPE requires independent validation. Prime editing is generally characterized by low off-target activity, but high-expression systems can inadvertently cause non-specific edits or unintended insertions of partial reverse-transcription templates at the target site. Data detailing the off-target mutation profile across whole-genome sequencing of SmartPE-treated lines has not yet been thoroughly compared to existing benchmarks.

Crucially, practical plant breeding requires that genetic edits be stable, heritable, and free from the editing machinery itself. The current accounts do not specify whether the reported efficiencies were measured in transient protoplast assays (isolated cells without walls) or in stably regenerated, fertile T1 and T2 generation plants that pass the edited trait through conventional seed reproduction.

What to watch

The primary benchmark to monitor will be the publication of peer-reviewed data providing comprehensive molecular characterizations of SmartPE, including detailed plasmid maps, codon optimization tables, and exact mutation frequencies across varied chromosomal contexts. Independent academic laboratories will seek to replicate the findings in other challenging dicots, such as potatoes, cotton, and woody fruit crops.

Observers should also track field-testing applications submitted to national biosafety authorities. If the developers or corporate partners move to translate these laboratory edits into practical agronomic traits, applications for environmental release will appear in public regulatory registries, particularly in China. Regulatory filings will reveal the specific traits prioritized by the team, such as resistance to soybean stem rot, shatter-resistant rapeseed pods, or shelf-life extensions in tomatoes.

Finally, the intellectual property framework surrounding SmartPE will influence its global deployment. Because foundational patents for prime editing are concentrated within institutions such as the Broad Institute, Harvard University, and commercial spinouts like Prime Medicine, the degree to which SmartPE constitutes a distinct, patentable architecture or relies on existing cross-licensing arrangements will determine how easily western agribusinesses can adopt the platform.

This report is based on original reporting by phys.org.