Spongiform encephalopathies or prion diseases are fatal neurodegenerative disorders that have prolonged asymptomatic incubation periods. The key event in pathogenesis of prion disease is conversion of normal prion protein into the pathogenic state due to conformational changes occurring in the native protein structure (fig.1)[1]. Chemical Exchange Saturation Transfer (CEST) imaging of NOE effects mediated by exchange-relayed signals or intermolecular pathways has been proposed as a new imaging mechanism to monitor protein folding by MRI[2,3]. Changes in protein conformation are expected to alter accessibility to water, which might affect slow chemical exchange processes occurring through exchange-relayed NOEs. Altered peptide chain dynamics might also affect dipole-dipole interactions through intramolecular NOEs.

The objective of this study was to map changes in NOE levels in the brains of prion infected mice using CEST MRI. Diseased animals were imaged at three stages (asymptomatic, early-stage and late-stage) to investigate whether prion propagation could be detected in their brains[4].

Two groups of 7-week-old FVB mice were intracerebrally inoculated with 30μl of 1% brain homogenate from Rocky Mountain Laboratory prion-infected mice (n=19) or brain homogenate from uninfected mice as controls (n=11). The prion-infected group was separated into three groups of mice scanned at different stages of prion disease: 80 days post injection (dpi) – asymptomatic-stage (n=6), 130dpi – early-stage (n=6), and 160dpi – late-stage (n=7). Control mice were separated into two groups: 80dpi (n=5) and 160dpi (n=6). All mice were anaesthetized (1.5-1.8% isoflurane in air enhanced with 1.5% oxygen) and scanned on a 9.4 T Agilent system using a 33-mm-diameter transmit/receive coil (Rapid Biomedical). All images were acquired in a single slice (thickness=2mm), centered on the thalamus. Anatomical scans were acquired using a fast spin-echo sequence (matrix 256x128, TR=3000ms, TE=20ms, FOV=20x20mm2). CEST measurements were acquired using a gradient-echo sequence (matrix 64x64, TR=2.11ms, TE=1.07ms, FOV=20x20mm²) with a saturation train of 80 Gaussian pulses at 0.6μT power before the readout (pulse length=50ms, flip angle=360°, 99% duty cycle). Saturation was applied at 71 frequency offsets between -5.0 and 5.0ppm.

Data analysis: Regions-of-interest (ROIs) were drawn on the anatomical image of each mouse in the cortex and thalamus for analysis of the corresponding NOE* maps. NOE* values were calculated between -3.7 and -3.3ppm by finding the difference between the Z-spectrum and a linear interpolation[5]. B0 correction was performed assuming zero the minimum of the fitted spline on each z-spectrum. Two-tailed t-tests were used to evaluate changes in NOE* values in prion-infected and control mice. Significant changes in NOE* values are indicated by p<0.05.

NOE* values showed no significant difference (p>0.05) between the two control groups scanned at 80dpi and 160dpi, therefore they were merged into one group when compared with prion-infected mice (fig.2). NOE* was significantly reduced in the thalamus and cortex of prion-infected mice at 130dpi (p<0.001 for thalamus, p=0.001 for cortex) and 160dpi (p<0.001 for thalamus, p=0.003 for cortex) compared with the control group (figs.2 & 3). NOE* was significantly reduced (p=0.001) in the thalamus of asymptomatic mice (80dpi) compared with the control group, however there was no significant change in the cortex (p>0.05) (figs.2 & 3).In this study we applied CEST to detect changes related to prion protein misfolding in prion-infected mice at three stages of prion disease. There were no significant changes in NOE* between the control groups, indicating that potential changes in NOE* in prion-infected mice between 80dpi and 160dpi are not age-related and follow prion propagation. The significant reduction of NOE* in the thalamus of prion-infected mice is in line with pathological changes that occur in this region throughout the disease course as has been shown in histological studies[4]. Significant reduction of NOE* in the cortex of prion-infected mice was observed only at 130dpi and 160dpi. This is consistent with observations that abnormal prions show a global pattern of deposition at later stages of the disease, while at 80dpi pathological changes are found mostly in the thalamus and brainstem, not in the cortex[4]. We conclude that NOE* values at different stages of prion disease provide additional evidence of prion protein misfolding occurring in the brains of diseased mice.