AI-tuned nanoparticle recipe keeps mRNA vaccines stable for a year at room temperature

RNA falls apart easily on its own, so vaccine makers wrap it in lipid nanoparticles that shield the molecule and carry it into cells ^[8]. To push those particles to tolerate heat, formulators mix in excipients, the sugars, salts and polymers that hold a formulation together ^[11]. The MIT team started with excipients that had stabilized nanoparticles in its earlier work, and stalled.

"We were trying to use and screen excipients that we've previously used successfully to stabilize LNPs, but it just wasn't working. It was really frustrating for the team," Jaklenec said ^[13].

So they turned to software, adjusting the formulation around the nanoparticles until it held up in heat ^[1]. Built with MIT's CSAIL, the algorithm scored nearly 50 FDA-approved excipients. For each one, the team packed a nanoparticle with mRNA coding for firefly luciferase and measured the light the cells emitted, a proxy for how much intact RNA got through ^[16]. It then predicted ratios for the five most promising excipients, which cut the number of bench experiments the team had to run ^[17][4]. Konakovic Lukovic, a CSAIL assistant professor and co-author, said she was surprised by how quickly the algorithm converged on a stable formulation, getting there in just a handful of iterations rather than the exhaustive search that would normally be required ^[15][18].

The functional test was in mice, where the reformulated COVID vaccine raised an immune response as strong as a shot similar to Moderna's ^[3]. The storage figures and that mouse comparison are reported separately, and the account does not say the mice received heat-aged doses. So the work shows the particles survive heat, and that they immunize, without yet demonstrating in one animal that a dose held warm for months still protects ^[2][3].