Gut microbiome changes nanomedicine delivery to tumors in mice

In these experiments, nanomedicine delivery to tumors depended on more than the design of the drug carrier. Researchers changed it by altering the gut microbiome: after metronidazole treatment, the liver captured less liposomal doxorubicin, the drug remained in circulation longer and more of it accumulated in tumors.

Nanomedicines package an active drug inside a microscopic carrier such as a liposome or an albumin particle. This can change where the drug travels in the body and may reduce exposure of healthy tissues. Yet a major obstacle appears before the particles reach a tumor: the liver removes a substantial fraction of them from the bloodstream.

Much of this filtering is performed by Kupffer cells, the resident macrophages of the liver. These cells recognize and engulf foreign material as part of the body’s normal defense system. That function is useful for clearing potentially harmful particles, but it also means that a nanoparticle drug may disappear from circulation before it reaches its intended target.

The gut tunes the liver’s filter

The investigators compared conventional specific-pathogen-free mice with germ-free animals, altered microbial communities with several antibiotics and transferred gut microbiota between animals. They also used single-cell RNA sequencing and metabolomic profiling to examine changes in liver immune cells and microbial metabolites.

Metronidazole produced the strongest reduction in nanoparticle uptake by liver macrophages. The same treatment had no such effect in germ-free mice, which lacked a microbiome to alter. Directly exposing isolated macrophages to metronidazole also did not reduce their uptake of nanoparticles or liposomes.

A fecal microbiota transfer experiment provided stronger evidence that the microbial community itself carried the effect. Microbiota from metronidazole-treated donors were transferred into germ-free recipients. No residual antibiotic was detected in the transferred material, yet the recipients still showed lower liver uptake, greater tumor delivery and a stronger response to nanomedicine.

Bile acids provided the mechanistic link. Metronidazole-associated microbiome remodeling reduced the availability of cholic acid and deoxycholic acid. When these molecules were added to primary mouse or human Kupffer cells in vitro, the cells became more efficient at engulfing nanoparticles and liposomes. The microbiome was therefore not guiding drugs toward tumors directly. It was changing chemical signals that helped keep the liver’s macrophages in a more phagocytic state.

The drug stayed in circulation longer

Pharmacokinetics were examined in mice bearing MC38 tumors and treated with Doxil, a liposomal formulation of doxorubicin. After metronidazole, the 48-hour area under the concentration–time curve increased from about 420 to 665 μg·h/ml. The apparent elimination half-life rose from 32.9 to 70.8 hours.

Doxorubicin accumulation fell in the liver and increased in tumors across several time points. The effect was not limited to Doxil. It was also observed with Abraxane, in which paclitaxel is carried by albumin particles, and with an experimental polymer nanoparticle formulation.

Microbiome remodeling improved treatment responses in mouse models of colon cancer, breast cancer, melanoma and pancreatic cancer. In several models, combining metronidazole with nanomedicine slowed tumor growth more effectively than the nanoparticle drug alone. In models where survival was measured, survival also improved. The benefit persisted in T-cell-deficient mice, indicating that it did not depend solely on adaptive T-cell antitumor immunity.

Why this is not yet a treatment strategy

Most of the evidence comes from mice. The human component was limited to experiments with commercially obtained primary Kupffer cells in culture. Those cells responded to bile acids in the predicted direction, but this does not show that microbiome manipulation will improve drug delivery in patients.

The mice also received metronidazole for 21 days. Antibiotics can disrupt microbial ecosystems, produce adverse effects and select for antimicrobial resistance. The study therefore does not support taking an antibiotic before chemotherapy.

Its main contribution is conceptual. Hepatic removal of nanoparticle drugs is not necessarily a fixed physiological barrier determined only by particle chemistry. It can be influenced through a gut microbiome–bile acid–Kupffer cell pathway. If that mechanism is confirmed in humans, more selective approaches might eventually target specific microbes, bile-acid signals or macrophage states without broadly depleting the microbiome.