Introduction

The analysis of functional brain network using resting-state fMRI (rs-fMRI) is extensively employed to study and characterize cognitive impairment present in patients affected by neurodegenerative disease¹. Moreover, alteration in functional connectivity seems to occur in early stages of the diseases before the manifestation of gross morphological damage or other obvious neurological synptoms². For this reason, investigating resting state connectivity in animal models of cognitive impairment offers a precious tool to gain insight into progression of neurodegenerative disease. Due to the sensitivity of MRI to motion artifacts, awake imaging can be challenging in preclinical studies involving animals. Awake imaging is not only necessary for task-based fMRI, but also anesthesia or sedation can alter the BOLD signal and functional connectivity^3,4 creating a confounding effect when interpreting group differences. In this perspective, rabbits have been shown to be ideal subjects for awake studies because of their tolerance to restraint⁵. Using a 3-channel coil specifically designed to enable head-posting and restraint we conducted experiments to advance our previous work⁶ and to further enhance investigation of resting-state functional networks.

Material and Methods

All experimental procedures involving animals complied with Northwestern’s IACUC guidelines. Four Rabbits underwent habituation to MRI environment. Acquisitions were performed on Bruker 7T ClinScan MRI using a three-channel custom made receiver coil (RAPID MR International, Columbus (OH)) designed with a central aperture that allows animal head fixation. Volume coil was used for transmission (Fig1). The animal cradle was designed to integrate the 3-channel receiver coil in a cross-bar fixed to the holder (Fig2). Prior to MRI surgery was performed on each rabbit. Two nylon bolts were fixed on the skull to lock the head into the 3-channel coil limiting head motion during MRI. Each rabbit was inserted in a Lomir Snuggle Sack to reduce stress and then placed in the cradle (Fig3) with the head locked in the cross-bar using the skull fixed bolts. Physiological parameters were recorded throughout the experiment inside the MRI. The acquisition protocol included a coronal 3D-GRE multi-echo scan (TR=68ms; TEs=2.7,6.83,11.26,16,10.13,25ms; Flip-Angle=15; voxel size= 0.29x0.29x0.5mm3; FOV=29.6x55.6x24mm3) and an transverse EPI for rs-fMRI (TR=1800ms; TE=25; Flip-Angle=70; voxel size=0.5x0.5mm2; slice-thickness=1.5; slices=20; FOV= 34x26mm2; GRAPPA=2; echo spacing=0.25ms; volumes=500). The rs-fMRI was repeated twice. Data analysis was performed using FMRIB Software Library v6.0 (Analysis Group, FMRIB, Oxford, UK). A high-resolution 3D image of each brain was generated from the 5 echo average. The 3D images from each rabbit were co-registered (non-rigid 12 degree-of-freedom transformation) and used to generate the rabbit brain template (Fig4) after skull removal. For the resting-state analysis, EPIs were first preprocessed: i)all volumes were registered by a rigid transformation to the central volume, ii)a brain mask was generated by the bias field corrected mean of the volumes and used as an inclusive mask for EPI, iii)a common origin was selected for all subjects’ EPIs, iv)slice timing correction was performed, v)time course was high-pass filtered with a threshold of 0.02 Hz and vi)images were smoothed using a 0.5 mm gaussian kernel. Visual inspection of EPI volumes and motion correction reports enabled overall quality control of data. Functional volumes were registered to high-resolution 3D images and then to the common template before independent components analysis (ICA). ICA group-analysis was run using a multi-session temporal concatenation pre-selecting 40 desired components. Resulting component were finally inspected to identify resting state network and spurious components.

Results

The second EPI repetition of one subject was excluded because of excessive body motion. The main functional networks previously described in the literature were identified and are shown in Figure 5.

Discussion and Conclusion

Reduction of motion artifacts in awake imaging requires head-posting of animals. This limits the types of rf coils which can be used for imaging. Most experiments employ open single channel surface coils with limited temporal resolution and coverage. To the best of our knowledge this is the first small animal awake rs-fMRI study using a customized phase-array 3 channel coil compatible with head-posting. With this coil we were able to use parallel imaging scheme (GRAPPA acceleration factor) increasing temporal and spatial resolution and extending whole brain coverage. Our setup allowed enhanced localization accuracy and improved temporal characterization of resting state fuctuations⁷. The ability to the delineate the expected independent components despite the small group of animals verified the good sensitivity of the described setup.

Acknowledgements

No acknowledgement found.