MIT bioengineers have built sensors that listen for the brain's chemical whispers, turning a notoriously tricky imaging challenge into something measurably visible. The new sensors, called liposomal nanoparticle reporters or LisNRs, can brighten or dim MRI signals in response to specific molecular targets—a breakthrough that could transform how doctors and scientists track the neurochemicals that make our minds and bodies work.
For years, researchers have struggled with a stubborn problem: MRI is excellent at imaging the brain's structure, but detecting the small molecules that actually drive neural activity has been nearly impossible. Neurotransmitters, neuropeptides, and metabolites exist in such low concentrations that traditional contrast agents simply don't respond visibly. As Alan Jasanoff, the Eugene McDermott Professor in the Brain Sciences and Human Behavior at MIT, explains it, "The signal change that you see in the imaging will be very modest. It won't let us detect physiological events." His team's new approach solves that by amplifying the effect—one target molecule can now activate many contrast agents at once, rather than just one.
The innovation came from rethinking the basic structure of the sensor itself. Postdoc Sayani Das and graduate student Jacob Cyert Simon designed LisNRs by packaging gadolinium, a magnetic material, inside tiny lipid-walled nanoparticles. Gadolinium normally brightens MRI signals by interacting with water molecules, but Das and Simon trapped it inside the nanoparticles where it stays inactive—unless water can get in. They engineered the particle walls with water channels that open or close depending on whether the target molecule is present.
When a target molecule arrives, it can knock aside a protein blocker guarding the channel, letting water rush in and activating the gadolinium's magnetic properties. The MRI signal brightens noticeably at that spot. Alternatively, the team also designed LisNRs that work in reverse: channels stay open and water flows freely until the target molecule arrives and blocks it, dimming the signal. This flexibility means the sensors could eventually detect dozens of different neurochemicals simultaneously.
To prove the concept works, Miranda Dawson and the Jasanoff team tested their LisNRs in living rats, detecting micromolar-scale levels of biotin—a test molecule—in both the brain and body with roughly tenfold greater sensitivity than earlier approaches allowed. The work appeared in May 2024 in Nature Biomedical Engineering, signaling that this fundamental advance is ready for real-world application.
Jasanoff's vision for what comes next is ambitious: "We want to be able to measure distinct chemical signals like neurotransmitters, neuropeptides, and metabolites as they fluctuate across the whole brain. These chemicals are important ingredients in neural computations, and we want to use the types of probes that we developed to detect these signals dynamically." The research team included key contributions from lab members Vinay Sharma, Samira Abozeid, and Gregory Thiabaud, along with collaborators at the University of Tokyo who helped engineer even more potent channels.
What makes this moment significant is that it bridges the gap between what we want to see in the brain and what our tools can actually show us. For neuroscientists trying to understand psychiatric conditions, neurological diseases, or even how a healthy brain processes information, this new generation of MRI sensors promises to reveal the chemical dynamics that have, until now, remained frustratingly invisible.