From StemCell科学观想

During cancer progression, tumor-resident bacteria have been shown to reduce cell stiffness by remodeling the host cell’s actin cytoskeleton, thereby endowing circulating tumor cells (CTCs) with enhanced deformability. This allows CTCs to evade hemodynamic shear stress in circulation and successfully colonize distant organs. Although CTC detection has become a crucial tool for tumor monitoring, conventional methods struggle to efficiently identify these highly metastatic "bacteria-infected CTCs."

Professor Song Yujun’s team at Nanjing University’s School of Biomedical Engineering developed a single-cell mechanophenotyping microfluidic chip named LesM, designed to efficiently capture and distinguish bacteria-infected CTCs based on their biomechanical properties. The core of the chip lies in its ingeniously designed L-shaped trapping units, each consisting of a hook-like structure for single-cell capture and a narrow microchannel for deformability assessment. The working principle exploits biomechanical differences: stiffer native cancer cells are blocked at the hook, whereas bacteria-infected cancer cells—due to cytoskeletal remodeling and reduced Young’s modulus—exhibit high deformability and can squeeze into the narrow microchannel.

To achieve optimal performance, the researchers systematically optimized the chip parameters through engineering design. Experiments demonstrated that by precisely controlling microchannel width (6–7 μm) and flow rate (5–10 μL/min), LesM achieves an ideal balance between capture efficiency and specificity. When processing complex samples, the chip exhibited exceptional performance, achieving a single-cell capture efficiency of 95.42%, enabling sensitive isolation of extremely rare tumor cells from whole blood containing vast numbers of blood cells. Cells are classified based on whether they "squeeze into" the channel, allowing quantification of their metastatic potential.

The study further investigated how intratumoral bacteria—such as Staphylococcus xylosus and Lactobacillus animalis—affect the mechanical properties of breast cancer cells. Confocal imaging and flow cytometry revealed that bacterial invasion induces disassembly of stress fibers in host cells. This structural change results from bacterial suppression of the RhoA-ROCK signaling pathway. Comparative experiments confirmed that S. xylosus-infected MDA-MB-231 and MCF-7 cells exhibited significantly reduced Young’s modulus, showing a softening effect similar to that induced by ROCK inhibitors. This biomechanical "camouflage" is key to CTC survival in circulation by helping them evade mechanical damage.

Using fluorescent labeling, the LesM chip successfully correlated intracellular microbial load with cellular mechanical phenotype (squeezing depth) at the single-cell level, validating mechanophenotype as a reliable biomarker for metastatic risk.

In vivo experiments using mouse models confirmed that bacteria-infected tumor cells possess stronger lung metastatic capacity. Monitoring with the LesM chip revealed a positive correlation between the number of infected CTCs detected in blood samples and the formation of lung metastatic nodules in mice.

Of significant clinical relevance, the study found that antibiotic administration before surgery (but not after) effectively reduced the number of infected CTCs in blood and decreased lung metastasis. This suggests that preoperative eradication of intratumoral microbes can block bacterial-induced mechanical remodeling of CTCs, thereby cutting off a key "metastatic booster."

In a retrospective clinical study of 69 breast cancer patients, the LesM chip demonstrated high clinical potential. Data showed that patients in the distant metastasis group had significantly higher counts of "bacteria-carrying, highly deformable" CTCs in their blood compared to those in the non-metastasis or lymph-node-only metastasis groups. ROC curve analysis yielded an area under the curve (AUC) of 0.904 for predicting distant metastasis, proving the chip’s exceptional value in clinical decision support, assessing the protective effects of antibiotic therapy, and predicting tumor recurrence.

Literature Link: https://doi.org/10.1038/s41467-025-68152-y