Genetically encodable calcium sensors for Magnetic Resonance Imaging

A key requirement for understanding the workings of the brain is to fill in the explanatory gap between molecular phenomena and identifiable behavior at the organismal level. Magnetic resonance imaging (MRI) provides a unique tool for bridging this gap, as it allows for imaging tissue throughout whole organisms. Although functional MRI (fMRI) is already a workhorse technique in human neuroscience research, current fMRI methods give us limited information about brain mechanisms because they rely on blood flow changes that are only indirectly coupled to cellular and molecular events. To associate cellular or molecular specificity to MRI, there is a need for genetically targeted, analyte-specific sensors. Calcium is a molecule of great interest to biology since its fluctuations are highly correlated with neural activity. While much progress has been made in pursuit of genetically encoded calcium sensors none allow for deep tissue imaging of whole rodent brains. In this thesis we demonstrate that genetically encodable calcium sensors based on known MRI gene reporter ferritin show modest sensitivity. To achieve higher amplification we leverage the hemodynamic response, which is coupled to neuronal activity through a calcium-activated enzyme, neuronal nitric oxide synthase (nNOS). We show that chemical stimulation of ectopically expressed neuronal nitric oxide synthase (nNOS) elicits an artificial hemodynamic response detectable by MRI. To distinguish signaling from endogenous nNOS we use a two-prong strategy to engineer a suite of enzymes with altered inhibition constants compared to nNOS. We demonstrate that these engineered enzymes (NOSTICs) exhibit calcium-dependent catalytic activity. One such NOSTIC was then virally delivered to rodent brains and shown to express in certain cell populations. Hemodynamic responses from these cell populations were recorded following electrical stimulation using MRI. The imaging strategy demonstrated here thus offers a novel and potentially powerful approach for cell-targeted functional imaging of the brain. Thesis Supervisor: Alan Jasanoff Title: Professor of Biological Engineering