Hypermagnetic Engineering Proteins for Functional Neuroimaging

Noninvasive measurement and manipulation of the nervous system could be possible using magnetic techniques, but a missing link is the availability of highly magnetic "handles" on cellular or molecular function. We therefore take on the challenge of engineering proteins that spontaneously become highly magnetic-so-called hypermagnetic ferritins (hFts)-and incorporating them into genetically encodable sensors for comprehensive brain activity mapping studies using magnetic resonance imaging (MRI). The magnetic proteins we produce will have impact in three broad areas: First and foremost, they will serve as building blocks for functional molecular neuroimaging probes, our principal interest and the application we focus on in this proposal. Second, magnetic proteins we produce will be uniquely powerful reporters for additional magnetic neuroimaging approaches. Third, highly magnetic proteins will be tools for magnetic manipulation of cells in the nervous system. These applications relate to nearly all areas of neuroscience, so the engineered proteins are likely to have very general significance. Our approach focuses on high throughput directed evolution of ferritin iron storage proteins for enhanced iron loading (Aim 1) and magnetization (Aim 2). Selected proteins will then be incorporated into sensors. This approach is justified because preliminary results already show that wild type or minimally engineered ferritins can be used as imaging agents, MRI sensors, and even magnetic manipulation approaches. Aim 1 makes use of a novel cytosolic iron reporting system to screen for hFt variants with abnormally high iron accumulation.

Preliminary results already show efficacy, and since brain Ft is normally only ~15% iron loaded, a considerable dynamic range should be accessible by applying this screening method over repeated cycles. Aim 2 applies a magnetic column based screen to identify even more highly magnetic hFt variants from yeast expressing mutant libraries. The ideal outcome of this approach will be an hFt variant that spontaneously acquires a superparamagnetic iron oxide core with specific magnetization and resulting magnetic properties up to 100-fold greater than the ferrihydrite core of wild type ferritins. In Aim 3, we wll incorporate hFt variants into calcium sensitive genetically encoded MRI contrast agents. Potentiating the magnetic moment of Ft-based calcium sensors by incorporating hypermagnetic building blocks will be make a critical difference to the sensors' ability to support robust functional imaging in vivo, and to applications in which these probes are genetically targeted to discrete cell types or circuit elements.

The proposal is consistent on multiple levels with the EUREKA program goals. Although the proposed research is "high risk," it builds on promising preliminary results, as well as the PI's record in developing novel neuroimaging probes. We believe that it is feasible to consider completely the proposed research within the four year EUREKA grant period.