A protein found in the brain could make the cancer drug Paclitaxel 3,600 times more soluble

Paclitaxel is one of the most effective anticancer drugs available. It is also one of the most difficult to deliver. The compound dissolves poorly in water, has a molecular weight of 854 daltons - large for a small-molecule drug - and distributes indiscriminately into normal tissues, causing the severe side effects that make chemotherapy so punishing for patients.

Improving how Paclitaxel reaches tumors and stays there has been a challenge for decades. A team at Osaka Metropolitan University has found an unlikely solution: an enzyme normally found in the brain.

Borrowing from cerebrospinal fluid

Lipocalin-type prostaglandin D synthase (L-PGDS) is a small protein with a barrel-shaped structure that naturally transports hydrophobic (water-fearing) molecules in the brain and cerebrospinal fluid. Professor Takashi Inui's research group recognized that L-PGDS's barrel interior could potentially accommodate drug molecules - including ones as large as Paclitaxel.

Using docking simulations and solubility testing, they confirmed that Paclitaxel binds primarily through hydrophobic interactions to the upper region of L-PGDS's beta-barrel structure. The result was dramatic: when bound to L-PGDS, Paclitaxel's solubility improved approximately 3,600-fold compared to the drug suspended in phosphate-buffered saline alone.

That's the difference between a drug that barely dissolves and one that can circulate effectively in the bloodstream.

Adding a cancer-targeting address label

Solubility is only half the delivery problem. The other half is getting the drug to tumor tissue specifically, rather than letting it damage healthy cells along the way. The research team attached a short targeting peptide called CRGDK to L-PGDS, creating a modified carrier they designated L-PGDS-CRGDK. The CRGDK peptide binds to neuropilin-1, a receptor that is overexpressed on the surface of many cancer cells.

In theory, this gives the drug-carrier complex a molecular address label - it should preferentially accumulate in tissues where neuropilin-1 is abundant, meaning tumor sites.

Mouse model results

The team tested the system in mice implanted with MDA-MB-231 breast cancer cells, an aggressive triple-negative breast cancer line commonly used in drug development research. They compared three formulations: a commercially available Paclitaxel preparation, Paclitaxel loaded into unmodified L-PGDS, and Paclitaxel loaded into the targeted L-PGDS-CRGDK carrier.

All three showed antitumor effects during active treatment. But the distinction emerged after treatment stopped. The commercial formulation lost its effectiveness once dosing ceased. In contrast, both L-PGDS formulations maintained antitumor activity after administration ended, with PTX/L-PGDS-CRGDK showing the strongest and most sustained tumor suppression.

The sustained effect suggests that the protein carrier changes the drug's pharmacokinetics - how it is absorbed, distributed, and cleared from the body - in ways that extend its activity beyond the dosing window.

From protein barrel to drug delivery platform

Professor Inui framed the significance in terms of the carrier's versatility. L-PGDS can bind relatively large drug molecules with molecular weights up to approximately 850 daltons, and introducing a targeting peptide enables selective delivery to cancer cells. This positions L-PGDS as a potential platform for delivering not just Paclitaxel but a range of poorly soluble anticancer agents.

A mouse study with the usual caveats

This is early-stage research with standard limitations. The study was conducted in a single mouse cancer model, and results in mice frequently do not translate to human patients. The number of animals per treatment group, the dosing regimen, and the specific tumor model all limit how broadly the findings can be generalized.

The targeting approach - using neuropilin-1 as the docking receptor - depends on that receptor being sufficiently overexpressed on the cancer type being treated. Not all tumors express neuropilin-1 at the same levels, and off-target binding in normal tissues that express the receptor could still occur.

Scaling production of a protein-based drug carrier from laboratory to clinical quantities presents its own manufacturing challenges - protein drugs are more expensive and complex to produce than small molecules.

Still, a 3,600-fold solubility improvement is a striking number, and the sustained antitumor effect after treatment cessation is noteworthy. If these results hold in broader preclinical testing, L-PGDS could offer a biologically elegant alternative to the synthetic nanoparticle carriers that currently dominate drug delivery research.