A single engineered peptide switched a dormant tumor-suppressor back on in colorectal cancer cells, and two of the cancer's main growth engines collapsed at the same time. The peptide is called i7-01s-20, and the result, published July 13 in the Journal of Translational Medicine ↗, is the kind of clean two-for-one mechanism that most cancer-peptide papers only gesture at.
The switch the cancer had turned off
The tumor-suppressor is Beclin1, a protein that tells a cell to recycle its own worn-out parts, a housekeeping process called autophagy. Beclin1 also helps route unwanted proteins to the cell's disposal system. In colorectal tumors it tends to get shut down, and the more it is silenced, the more advanced the tumor and the worse the patient tends to do. The authors confirmed that pattern again across public datasets and human tissue samples before touching a single drug.
i7-01s-20 is a stapled peptide. That means a short chain of amino acids has been locked into a fixed, coiled shape with a chemical brace. The staple lets it hold its grip and survive in the body far longer than a loose peptide would. This one is built to bind Beclin1 and hold it in its active form, flipping the switch the cancer had flipped off.
Two pedals, one brake
Here is where the paper earns its title. Turning Beclin1 back on did two things that a cancer cell does not want.
First, it dragged the epidermal growth factor receptor, EGFR, into the cell's recycling machinery and had it degraded. EGFR is a classic accelerator pedal for tumor growth. Losing it quieted the AKT and GSK-3 beta signaling that runs downstream. Second, it triggered the tagging and proteasomal destruction of beta-catenin, the protein at the heart of the Wnt pathway that colorectal cancers lean on almost universally. One peptide, acting through one restored switch, knocked out two separate oncogenic pathways that are usually attacked with two different drugs.
Why the controls are the story
Plenty of compounds kill cancer cells in a dish for boring reasons. Two design choices make this result harder to dismiss. When the researchers silenced Beclin1 itself, the peptide stopped working entirely. The effect runs through the intended target, then, and not through some off-target toxicity. A scrambled version of the same peptide, the same amino acids shuffled into a meaningless order, did essentially nothing at matching doses.
In mice carrying DLD-1 human colorectal tumors, systemic i7-01s-20 shrank the tumors at increasing doses without obvious whole-body toxicity. Proteomics on the tumor tissue showed the same AKT and beta-catenin suppression the cell work predicted.
What it is not
This is preclinical work in cell lines and mouse xenografts. There is no human data and no dosing schedule for a patient. There is no readout on how a stapled peptide of this size holds up over weeks of real treatment rather than a controlled experiment. Xenograft tumors also live under a suppressed immune system, so a mouse cure is a starting line, not a finish. Activating autophagy is itself double-edged in cancer, protective in some contexts and fuel in others. A longer study will have to show the balance stays on the right side.
On peptidemodel the relevant through-line is the anticancer ↗ shelf. The recurring theme there is exactly this: a short, stabilized peptide reaching a target that small molecules and antibodies struggle to drug cleanly. A tumor-suppressor you switch on, rather than an oncogene you try to block, is a rarer move, and worth watching for whether it holds up outside a dish.
Most cancer-peptide papers claim a hit and a hand-wave. This one names the switch, degrades two drivers through it, and shows the whole thing evaporates when you take the switch away.