Harnessing axonal transport to map reward circuitry: Differing brain-wide projections from medial forebrain domains

Abstract

Neurons project long axons that contact other distant neurons. Projections can be mapped by hijacking endogenous membrane trafficking machinery by introducing tracers. To witness functional connections in living animals, we developed a tracer detectible by magnetic resonance imaging (MRI), Mn(II). Mn(II) relies on kinesin-1 and amyloid-precursor protein to travel out axons. Within 24h, projection fields of cortical neurons can be mapped brain-wide with this technology. MnCl2 was stereotactically injected either into anterior cingulate area (ACA) or into infralimbic/prelimbic (IL/PL) of medial forebrain (n=10-12). Projections were imaged, first by manganese-enhanced MRI (MEMRI) live, and then after fixation by microscopy. MR images were collected at 100{micro}m isotropic resolution ([~]5 neurons) in 3D at four time points: before and at successive time points after injections. Images were preprocessed by masking non-brain tissue, followed by intensity scaling and spatial alignment. Actual injection locations, measured from post-injection MR images, were found to be 0.06, 0.49 and 0.84mm apart between cohorts, in R-L, A-P, and D-V directions respectively. Mn(II) enhancements arrived in hindbrains by 24h in both cohorts, while co-injected rhodamine dextran was not detectible beyond immediate subcortical projections. Data-driven unbiased voxel-wise statistical maps after ACA injections revealed significant progression of Mn(II) distally into deeper brain regions: globus pallidus, dorsal striatum, amygdala, hypothalamus, substantia nigra, dorsal raphe and locus coeruleus. Accumulation was quantified as a fraction of total volume of each segment containing significantly enhanced voxels (fractional accumulation volumes), and results visualized in column graphs. Unpaired t-tests between groups of brain-wide voxel-wise intensity profiling by either region of interest (ROI) measurements or statistical parametric mapping highlighted distinct differences in distal accumulation between injection sites, with ACA projecting to periaqueductal gray and IL/PL to basolateral amygdala (p<0.001 FDR). Mn(II) distal accumulations differed dramatically between injection groups in subdomains of the hypothalamus, with ACA targeting dorsal medial, periventricular region and mammillary body nuclei, while IL/PL went to anterior hypothalamic areas and lateral hypothalamic nuclei. Given that these hypothalamic subsegments communicate activity in the central nervous system to the body, these observations describing distinct forebrain projection fields will undoubtedly lead to newer insights in mind-body relationships.