In a recent paper published in International Journal of Medical Sciences, researchers have described how they used artificial intelligence and machine learning tools to find mTOR-inhibiting molecules [1].
mTOR is a common target for lifespan extension interventions
The mechanistic target of rapamycin (mTOR) is a well-known molecule in the rejuvenation world. Previous studies have demonstrated that reducing the activity of mTOR increases the lifespan of multiple laboratory animals, including yeast, worms, flies, and mice [2].
Currently, using mTOR inhibitors such as rapamycin to extend human lifespan is a subject of debate due to the possible side effects [3], which include anemia, increased blood pressure, fever, headache, nausea, diarrhea, and even new-onset diabetes [4]. This team looked for different, effective inhibitors of mTOR activity, which may not have these side effects.
Using AI to filter a thousand molecules
The researchers used machine learning tools to generate a pool of 1,000 molecules, narrowed down the pool to 132 based on their potential for mTOR targeting, then chose 29 that were likely to have low toxicity. The researchers then ran those final candidates through an ADMET (absorption, distribution, metabolism, excretion, toxicity) profile. The winner of this molecular competition was TKA001.
This competition, of course, was only in a simulation, and AI-generated molecules must be tested in real-world models to determine if they are actually effective. The researchers began these experiments by testing the activity of mTOR in human cell lines upon adding TKA001.
mTOR binds to two different sets of proteins, thus creating the mTORC1 and mTORC2 complexes [2], which each have their own targets and impact on downstream proteins downstream of them. When active, mTORC1 attaches a phosphate group to the molecule S6K. mTORC2, on the other hand, attaches a phosphate group to the molecule AKT.
Researchers observed a reduced number of phosphate attachments to both S6K and AKT in cells that had been administered TKA001. This suggests that TKA001 inhibited both mTORC1 and mTORC2.
TKA001 inhibits cancer cell proliferation
The AI-based analysis predicted that TKA001 could be a potent agent in prostate cancer treatment, so the researchers conducted experiments to confirm its effectiveness against cancer. They began with epithelial cancer cells from patients with fibrosarcoma, a type of tumor that originates from fibrous connective tissue. They also used human cervical cancer cells.
Half maximal inhibitory concentration (IC50) is a measurement of how much of a given molecule is needed to inhibit 50% of a biological process. In this case, it refers to the proliferation of cancer cells. Rapamycin has an IC50 of 1.8 µM in fibrosarcoma cells and 0.25 µM in cervical cancer cells, but TKA001 has 200nM and 1µM, respectively, showing that it has a comparable effect on cancer.
TKA001 extends C. elegans lifespan
The researchers wanted to test TKA001 on complete living organisms, so they chose C. elegans, a small roundworm commonly used in lifespan studies. They found that different doses extended the lifespan of C. elegans when given to adult or young adult worms. However, although the lifespan extension in C. elegans was statistically significant, it appears to be rather modest.
Strong potential, but more testing needed
In general, these results are encouraging. However, they need to be confirmed first in organisms that are biologically closer to humans, such as mice, before this molecule is brought into clinical trials to test safety and effectiveness.
TKA001 could be an interesting alternative to rapamycin, especially since the authors’ AI-based analysis suggests that TKA001 has low toxicity. This suggests that its side effects should be limited, but that is also something that can only be tested in clinical trials.
Literature
[1] Vidovic T, Dakhovnik A, Hrabovskyi O, MacArthur MR, Ewald CY. AI-Predicted mTOR Inhibitor Reduces Cancer Cell Proliferation and Extends the Lifespan of C. elegans. Int J Mol Sci. 2023 Apr 25;24(9):7850. doi: 10.3390/ijms24097850. PMID: 37175557; PMCID: PMC10177929.
[2] Saxton, R. A., & Sabatini, D. M. (2017). mTOR Signaling in Growth, Metabolism, and Disease. Cell, 168(6), 960–976. https://doi.org/10.1016/j.cell.2017.02.004
[3] Salmon A. B. (2015). About-face on the metabolic side effects of rapamycin. Oncotarget, 6(5), 2585–2586. https://doi.org/10.18632/oncotarget.3354
[4] Johnston, O., Rose, C. L., Webster, A. C., & Gill, J. S. (2008). Sirolimus is associated with new-onset diabetes in kidney transplant recipients. Journal of the American Society of Nephrology : JASN, 19(7), 1411–1418. https://doi.org/10.1681/ASN.2007111202
