Magnetic Bacteria Eaten as Pills Destroy Colon Tumors and Rewire Gut Microbiome
Engineered magnetotactic bacteria delivered orally penetrate colorectal tumors, trigger immune attack, and reshape gut microbial metabolism in mice.
Summary
Researchers engineered a strain of naturally magnetic bacteria (Magnetospirillum gryphiswaldense) to carry two cancer-fighting agents — the chemotherapy drug camptothecin and an immune checkpoint-blocking peptide — then coated them in a protective lipid shell for oral delivery. In mouse models of colorectal cancer, these bacteria survived the harsh gastrointestinal tract, reached tumors, and when activated by an alternating magnetic field, generated localized heat that killed cancer cells. The treatment triggered a broad immune response, activated cancer-killing T cells, and simultaneously shifted the gut microbiome toward beneficial bacteria that produce short-chain fatty acids. Both primary and distant tumors shrank significantly, suggesting systemic immune memory was established.
Detailed Summary
Colorectal cancer (CRC) remains one of the most common and deadly cancers worldwide, and standard treatments — surgery, chemotherapy, and immunotherapy — frequently fail due to immune suppression and gut microbiome disruption. This study, published in Bioactive Materials (2025), introduces a novel living therapeutic platform: oncolytic magnetotactic bacteria engineered to deliver combination chemo-immunotherapy directly to colorectal tumors via oral administration, while simultaneously remodeling the gut microbiota and its metabolic output.
The platform, called MSR-CPT/APP@LPs, was built on Magnetospirillum gryphiswaldense (MSR), a facultative anaerobe that naturally biomineralizes chains of iron-oxide magnetosomes (30–50 nm) inside its helical body (2–5 μm long). Camptothecin (CPT), a topoisomerase I inhibitor, was conjugated to the bacterial surface via a glutathione (GSH)-responsive linker for tumor-selective release, while an anti-PD-L1 peptide (APP) was attached via a hypoxia-responsive linker. The bacteria were then encapsulated in a protective shell of mulberry-leaf lipids and Pluronic F127 (LPs) to survive gastric acid, intestinal enzymes, and colonic fluid. Growth curve experiments confirmed that MSR-CPT/APP@LPs maintained viability comparable to unmodified MSR after sequential exposure to simulated gastric (2 h), small intestinal (8 h), and colonic (12 h) fluids.
Upon reaching the tumor, exposure to an alternating magnetic field (AMF, 4.5 kW, 375 kHz) caused the magnetosomes to generate heat — suspensions of 1.2 × 10⁹ CFU/mL reached 84 °C within 5 minutes — while the lipid shell degraded at 42 °C, releasing the bacteria and their drug payloads. In CT-26 colorectal cancer cell experiments, MSR-CPT@LPs + AMF produced the highest cytotoxicity and induced robust DNA damage (γ-H2AX staining), activated the cGAS-STING pathway (elevated p-STING, p-TBK1, p-IRF3), triggered immunogenic cell death (ICD) markers including calreticulin exposure and HMGB1 release, and promoted macrophage polarization toward the pro-inflammatory M1 phenotype.
In AOM/DSS-induced primary CRC mouse models (n = 6 per group), oral MSR-CPT/APP@LPs + AMF significantly suppressed primary tumor growth and, in tumor rechallenge experiments (subcutaneous CT-26 injection on day 59), nearly abolished distant tumor formation — demonstrating durable systemic immune memory. Flow cytometry of spleens and lymph nodes revealed increased mature dendritic cells (CD80⁺CD86⁺), elevated CD8⁺ cytotoxic T cells, reduced regulatory T cells (Foxp3⁺), and expanded effector memory T cells (CD44⁺CD62L⁻). Serum levels of TNF-α, IFN-γ, and IFN-β were all significantly elevated in the MSR-CPT/APP@LPs + AMF group versus controls.
A particularly striking finding was the treatment's impact on gut microbiota and metabolism. 16S rRNA sequencing of feces showed that MSR-CPT/APP@LPs + AMF increased relative abundances of beneficial taxa including Lachnospiraceae (unclassified) and Alistipes, while reducing Bacteroides. Metabolomic analysis revealed elevated short-chain fatty acids (butyric, valeric, and caproic acids), which activate CD8⁺ T cells and suppress the immunosuppressive kynurenine pathway by downregulating IDO1, IDO2, and TDO2. Citrulline — an immune-activating metabolite — increased, while L-glutamine (which fuels tumor growth) and kynurenic acid (which suppresses T cells) decreased. Lachnospiraceae downregulated glycerophospholipids (PC, PE, PS), further relieving T cell inhibition. These microbiome-metabolome shifts represent a secondary, indirect antitumor mechanism operating in parallel with direct tumor killing.
Key Findings
- MSR suspensions (1.2 × 10⁹ CFU/mL) reached 84 °C within 5 minutes under AMF (4.5 kW, 375 kHz), demonstrating potent magnetothermal capacity sufficient to kill tumor cells
- MSR-CPT/APP@LPs maintained bacterial viability comparable to unmodified MSR after sequential exposure to simulated gastric, small intestinal, and colonic fluids, confirming GI tract survival
- CPT was released in a GSH-dependent manner (accelerated release at 10 mM GSH vs. 0 mM over 72 h), and APP was released under hypoxic conditions, enabling tumor-selective drug delivery
- In tumor rechallenge models (n = 6/group), MSR-CPT/APP@LPs + AMF nearly abolished distant subcutaneous CT-26 tumor growth, indicating durable systemic immune memory
- Treatment significantly elevated serum TNF-α, IFN-γ, and IFN-β, increased CD8⁺ T cells and mature dendritic cells, and reduced immunosuppressive Foxp3⁺ regulatory T cells vs. controls
- 16S rRNA sequencing showed increased Lachnospiraceae and Alistipes with decreased Bacteroides; metabolomics confirmed elevated butyric, valeric, and caproic acids alongside reduced L-glutamine and kynurenic acid
- Butyric acid was shown to suppress the kynurenine pathway by downregulating IDO1, IDO2, and TDO2, relieving CD8⁺ T cell inhibition as an indirect antitumor mechanism
Methodology
Mouse colorectal cancer was induced using AOM (10 mg/kg i.p.) plus two cycles of 3% DSS in drinking water; treatment groups (n = 6 biologically independent samples each) received oral MSR-CPT/APP@LPs (MSR: 5×10⁸ CFU/mouse; CPT: 2 mg/kg; APP: 1–2 mg/kg) on days 42, 45, 48, 51, and 54, with AMF irradiation (4.5 kW, 3 min) on alternating days. Distant tumor rechallenge was performed by subcutaneous injection of 3×10⁶ CT-26 cells on day 59. Outcomes included tumor volume, histology (H&E, Ki67, TUNEL), flow cytometry (T cell subsets, dendritic cells), ELISA (cytokines, STING pathway proteins), 16S rRNA gut microbiome sequencing, and untargeted metabolomics. Statistical comparisons used mean ± s.e.m. with significance thresholds at p<0.05, p<0.01, and p<0.001.
Study Limitations
This study was conducted entirely in mouse models (AOM/DSS-induced CRC and CT-26 subcutaneous tumors), and the translation of magnetothermal therapy requiring external AMF application to human colorectal tumors presents substantial engineering and safety challenges. The paper does not report long-term toxicity data or assess the fate of the engineered bacteria after tumor treatment, leaving biosafety questions open. No conflicts of interest were declared, but the study was performed by the same groups that developed the platform, and independent replication has not yet been reported.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
