Agrobacterium is the workhorse of plant transformation, ferrying foreign DNA into plant cells through a process that has been harnessed for decades in laboratories worldwide. Dozens of engineered strains exist, each with distinct genotypes and phenotypes suited to different plant species, yet the molecular details of how the bacterium infects plants and integrates T-DNA into host genomes remain incompletely understood. Screening mutant strains and deliberately constructing engineered variants are considered key routes to unraveling these mechanisms, but the tools for doing so have been cumbersome.

Traditional Agrobacterium gene editing relies on plasmids carrying suicide genes, such as the widely used pK18mobSacB system, and typically demands two rounds of homologous recombination. In the first round, an antibiotic like kanamycin forces a tool plasmid that cannot replicate autonomously in Agrobacterium to integrate into the bacterium’s endogenous Ti or Ri plasmid through homologous recombination. In the second round, the SacB suicide gene enables negative selection against cells still carrying the plasmid, allowing researchers to isolate products of the second recombination event. The problem is that these steps yield large numbers of unmodified bacteria, driving overall modification efficiency down to very low levels.

The new method, described for the first time in detail by Yu Liu, Geng Sun, and colleagues, sidesteps much of this complexity by requiring only a single round of homologous recombination. The trick lies in turning the bacterium’s own biology against itself, so to speak: instead of relying solely on an introduced template, the researchers induce the production of endogenous T-DNA within the Agrobacterium cell and use this single-stranded DNA as the homologous repair template. Because single-stranded DNA is known to be a more effective substrate for homologous recombination than double-stranded DNA, this endogenous supply of T-DNA provides a ready-made molecular scaffold for precise editing.

Getting tool plasmids into Agrobacterium has historically been its own bottleneck, often requiring triparental mating, a three-strain conjugation dance, or expensive electroporation equipment. The Bozhou team showed that a simple freeze-thaw method suffices for introducing their plasmids into competent GV3101 cells, a convenience that could make the protocol accessible to laboratories without specialized transformation apparatus. The researchers constructed a backbone vector called pSG-SacB05, combining pBBR1 origin and replication sequences from pBBR1MCS-2, the SacB gene from pK18mobSacB, a pBR322 origin for easier plasmid preparation, a kanamycin resistance gene, and T-DNA repeat sequences with multiple cloning sites for inserting template sequences.

To demonstrate the method, the team targeted VirD2, a crucial protein in Agrobacterium-mediated genetic transformation that plays a pivotal role in transferring T-DNA from bacterium to plant cell and is essential for integrating T-DNA into the host genome. Using a tool plasmid called pSG-SacB10, they engineered simultaneous modifications at both the N-terminal and C-terminal ends of VirD2, which resides on the Ti plasmid of GV3101. The design inserted a 72-base-pair SpyTag sequence flanked by a 487-base-pair left homology arm derived from VirD1, and deleted 714 base pairs of the VirD2 C-terminus while inserting 152 base pairs of VirF sequence, all flanked by a 574-base-pair right homology arm from VirD3. A 630-base-pair N-terminal segment of VirD2 was preserved and could act as a middle homology arm.

The modification protocol unfolds across eleven carefully defined steps, beginning with vector construction using homologous arms longer than 400 base pairs, followed by freeze-thaw transformation onto plates containing rifampicin, streptomycin, and kanamycin. The critical induction phase involves culturing kanamycin-resistant clones in liquid medium adjusted to pH 5.5 with 20 milligrams per liter kanamycin and 300 micromolar acetosyringone, a phenolic compound that triggers Agrobacterium’s virulence machinery, at 25 degrees Celsius for three days. An optional extended culture step can push modification efficiency higher, and dual PCR detection with specific primer pairs confirms whether edits occurred at both target sites before the bacteria are plated on sucrose-containing medium to eliminate cells still carrying the tool plasmid.

The results revealed a striking dependence on culture conditions. When five independent modification cultures were grown at pH 5.5 with acetosyringone, four produced clear 572-base-pair and 793-base-pair electrophoretic bands indicating edits at both the upstream and downstream regions. Among 22 monoclonal colonies tested from each culture, 13.6 percent carried the downstream modification, 22.7 percent carried the upstream modification, and 4.5 percent carried both simultaneously. After secondary colony purification and SacB residue testing, all 27 sucrose-surviving colonies were SacB negative, and Sanger sequencing of four candidate modified strains confirmed their sequences matched the designed target with perfect fidelity. Kanamycin sensitivity tests further verified that no residual antibiotic resistance from the tool plasmid remained in the successfully edited bacteria.

In contrast, parallel cultures grown at pH 7.0 without acetosyringone told a different story. While 39.1 percent of colonies showed the upstream modification, not a single colony among 110 tested displayed the downstream edit, despite weak 793-base-pair signals in liquid culture. This opposite trend, with upstream modification rising while downstream modification vanished, suggests that small-fragment and long-fragment recombination operate through fundamentally different mechanisms. The researchers propose that the medium-copy-number pSG-SacB10 plasmid provides enough template for short-fragment recombination on its own, but long-fragment modification depends on the abundance of single-stranded endogenous T-DNA, which is only produced under the acidic, acetosyringone-induced conditions that activate Vir gene expression and promote T-DNA processing and stability.

The authors acknowledge that the method’s efficiency, while promising, remains modest, and they point to several contributing factors. High concentrations of streptomycin and kanamycin may suppress Vir gene expression, and more importantly, commonly used Agrobacterium strains including GV3101 carry a function-less recA mutation, removing a key enzyme in DNA repair that would normally boost recombination. The team suggests that introducing a functional recA gene will be crucial for improving efficiency and expanding the host range of the technique in future work. Even so, with nine Sanger-confirmed colonies across multiple experiments all showing 100 percent sequence accuracy, the precision of the method is clear, and given how rapidly bacteria grow, the researchers argue the approach can meet practical demands. They position the protocol as a straightforward, reproducible tool for Agrobacterium genome engineering with potential applications across biotechnology and agricultural sciences, from creating integration-less strains for cleaner plant transformation to building custom variants that illuminate the still-mysterious mechanics of nature’s own genetic engineer.

Subject of Research: A one-step homologous recombination method for precise Agrobacterium genome editing using induced endogenous T-DNA

Article Title: Precise genome modification of agrobacterium based on one-step homologous recombination via endogenous transfer DNA

Article References: Liu, Y., Sun, G., Gao, C., Gao, R., Tang, J., Wang, M., Guo, J., & Guo, R. (2026). Precise genome modification of agrobacterium based on one-step homologous recombination via endogenous transfer DNA. Acta Biochimica Polonica, 73, Article 16599. https://doi.org/10.3389/abp.2026.16599

Image Credits: AI Generated

DOI: 10.3389/abp.2026.16599

Keywords: Agrobacterium, genome editing, T-DNA, homologous recombination, VirD2, SacB, acetosyringone, plant transformation, biotechnology, GV3101, genetic engineering, homology-directed repair