Cadmium contamination of rice paddies is one of the most stubborn food safety problems in the world, and a new study suggests the solution may live in the soil itself. Researchers at the Hunan Academy of Agricultural Sciences report in BMC Genomics that a genetically engineered strain of the soil bacterium Citrobacter sp. XT1-2-2 can reshape the entire microbial community surrounding rice roots while simultaneously triggering protective changes inside the plant, with the combined effect of reducing the amount of cadmium available to the crop. The work, led by Zhudong Liu and colleagues including corresponding authors Yilu Li, Shandong Wu and Shiping Shan, offers a detailed molecular portrait of how a single bacterial inoculant can act on both the environment and the plant at once.
The team started with a wild-type Citrobacter strain originally isolated from rice rhizosphere soil, the thin zone of earth immediately surrounding living roots where chemistry, biology and plant physiology collide. From this native microorganism they constructed engineered variants by boosting the expression of the bacterium’s own functional genes rather than introducing foreign DNA. Two over-expression lines, designated Citrobacter sp. XT1-2-2-::APS and Citrobacter sp. XT1-2-2-::SiR, carry enhanced copies of key reductase genes, enzymes that drive reduction reactions capable of altering the chemical state of heavy metals. The choice of genes matters: reductive transformations are central to how certain bacteria immobilize cadmium, converting more mobile or bioavailable forms of the metal into species that plant roots struggle to absorb.
To test the concept, the researchers ran pot experiments comparing rice grown with the wild-type strain against rice inoculated with the two over-expression strains. At the maturity stage of the crop, the decisive measurement was the concentration of available cadmium in the soil, the fraction of the metal that plants can actually take up. Both engineered strains produced a significant reduction in available cadmium relative to the wild-type inoculant, indicating that amplifying the reductase genes translated directly into stronger immobilization of the metal in the rhizosphere. Because the engineered strains were built by enhancing endogenous genes, the biosafety profile of the construct is comparatively straightforward; the authors note that hemolysis assays were negative and that all work with the engineered microorganisms followed institutional containment guidelines, with animal toxicity studies and field trials planned before any environmental release.
The cadmium result alone would be noteworthy, but the study’s real ambition was to understand the mechanism, and that required looking at the soil as an ecosystem rather than a chemical reservoir. Using high-throughput sequencing of the 16S rRNA gene for bacteria and the internal transcribed spacer region for fungi, the team mapped the structure and diversity of the rhizosphere microbial communities under each treatment. The analysis, processed through standard pipelines such as QIIME with operational taxonomic units as the unit of counting, revealed that inoculation with the over-expressing strains significantly altered the community structure of both bacteria and fungi compared with the wild-type treatment, and that community abundance and diversity increased in the engineered-strain soils.
Those shifts were not random. Several bacterial phyla rose in relative abundance in soils treated with the over-expression strains, including Bacillota, Bacteroidota, Pseudomonadota, Thermodesulfobacteriota and Acidobacteriota. Each of these groups has known associations with soil health and metal chemistry: Bacillota include many spore-forming, plant-growth-promoting species; Pseudomonadota encompass numerous rhizosphere colonizers and metal-tolerant taxa; and sulfate- and sulfur-metabolizing lineages within the Thermodesulfobacteriota can influence metal speciation through the production of metal sulfides. On the fungal side, the community was dominated by Ascomycota, Chytridiomycota, Basidiomycota and unclassified fungi, with the engineered inoculants again shifting the balance. The picture that emerges is of a rhizosphere whose resident ecology is nudged into a configuration less favorable to cadmium mobility.
The second half of the investigation moved inside the plant. The researchers performed a proteomic analysis of rice roots using tandem mass tag labeling combined with liquid chromatography-tandem mass spectrometry, a technique that allows many protein samples to be compared quantitatively in a single experiment. TMT labels are attached to peptides from each sample, the samples are pooled and analyzed together, and the relative intensity of each tag reports how abundant the corresponding protein was in the original tissue. Applied to rice roots, this approach identified differentially expressed proteins between treatments, giving a systems-level view of how the plant responds to the engineered bacteria at the molecular level.
The differentially expressed proteins were enriched mainly in biosynthetic pathways, metabolic pathways, glutathione metabolism, cell wall organization and antioxidant defense pathways, as annotated against the Kyoto Encyclopedia of Genes and Genomes. Each of these categories tells a coherent story about cadmium stress. Glutathione is a tripeptide that binds heavy metals and feeds the synthesis of phytochelatins, peptides that sequester cadmium inside plant cells and limit its toxicity. Cell wall organization matters because the wall is the first barrier a metal ion encounters on its way into a root cell, and modifications to its composition can reduce metal entry. Antioxidant defenses counter the oxidative damage that cadmium provokes indirectly by disrupting electron transport and depleting cellular thiols. Together, the proteomic signature suggests that rice roots inoculated with the engineered strains mount a coordinated defensive and metabolic reprogramming, consistent with the observed reduction in cadmium uptake.
What makes the study conceptually interesting is the dual mechanism it documents. The engineered bacteria act externally, lowering the pool of available cadmium in the rhizosphere and recruiting or enriching a microbial community that supports that state, while acting internally, priming the plant’s own stress-response and sequestration machinery through changes in root protein expression. The authors frame this as illuminating the complex interactions between rice and its associated microorganisms, and specifically as showing how over-expression of key reductase genes in XT1-2-2 propagates outward to the endophytic and rhizosphere microbiome and inward to the host proteome. In other words, a targeted genetic edit in one soil bacterium produces cascading effects across a whole ecological network, and those cascades converge on the practical outcome of less cadmium in the crop.
The practical implications extend well beyond one pot experiment. Rice is the dietary staple for billions of people and is unusually efficient at accumulating cadmium from flooded soils, which is why cadmium-tainted rice remains a recurring public health scandal in several rice-growing regions. Conventional remediation, such as soil replacement, chemical immobilizers or growing non-food crops on contaminated land, is expensive and slow. A living inoculant that can be applied to paddies, derived from a native soil organism and engineered only to over-produce its own enzymes, represents a potentially cheaper and more scalable intervention. The authors position the work as theoretical support for applying functional microorganisms to promote rice growth while reducing cadmium accumulation, and the combined microbiome and proteome datasets provide the mechanistic grounding that regulators and agronomists would need before such an approach could move outdoors.
Caution is still warranted. The experiments were conducted under controlled laboratory conditions, and the authors themselves state that further animal toxicity studies and field trials are planned to evaluate the biosafety of the engineered strain prior to any potential environmental application. Rhizosphere communities vary enormously with soil type, water management, climate and cropping history, so the community shifts observed in pots will need to be reproduced in real paddies. Yet the study stands as a striking demonstration of synthetic biology meeting soil ecology: by tuning the expression of two reductase genes in a bacterium that rice roots already know, the researchers changed the chemistry of the soil, the composition of the microbial ecosystem and the protein inventory of the plant, and in doing so kept a toxic heavy metal out of the food chain. If the field results hold, the humble rhizosphere may become the most important battleground in the fight against cadmium in rice.
Subject of Research: Engineered Citrobacter bacteria reducing cadmium uptake in rice through rhizosphere microbiome modulation and root proteomic changes
Article Title: A genetically engineered Citrobacter sp. XT1-2-2 modulates rhizosphere soil microbiome and induces root proteomic changes to reduce cadmium uptake in rice
Article References: Liu, Z., Du, D., Cheng, W., Yan, Z., Wei, X., Zhang, M., Li, Y., Wu, S., & Shan, S. (2026). A genetically engineered Citrobacter sp. XT1-2-2 modulates rhizosphere soil microbiome and induces root proteomic changes to reduce cadmium uptake in rice. BMC Genomics. https://doi.org/10.1186/s12864-026-13465-5
Image Credits: AI Generated
DOI: 10.1186/s12864-026-13465-5
Keywords: Citrobacter sp. XT1-2-2, cadmium, rice, rhizosphere microbiome, soil microbiology, genetic engineering, proteomics, TMT labeling, glutathione metabolism, heavy metal immobilization, plant growth-promoting bacteria, BMC Genomics
News Source: Morgan Morrow. (October 19, 2026). Engineered Soil Bacteria Rewire Rice Roots to Block Cadmium Uptake. Scienmag.
