Ultra-high-field fMRI mapping of layer-specific somatosensory processing in marmoset brain

2433 

Abstract Submission 

Jiajia Yang¹, Hirohiko Imai², Masaki Fukunaga³, Hiroki Yamamoto⁴, Chenyu Wang¹, Yinghua Yu¹, Kazuhiko Seki⁵, Takashi Hanakawa⁶, Tatsuya Umeda⁶

¹Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama, Japan, ²Department of Informatics, Kyoto University Graduate School of Informatics, Kyoto, Japan, ³Section of Brain Function Information, National Institute for Physiological Sciences, Aichi, Japan, ⁴Graduate School of Human and Environmental Studies, Kyoto University, Kyoto, Japan, ⁵Department of Neurophysiology, National Center of Neurology and Psychiatry, Tokyo, Japan, ⁶Department of Integrated Neuroanatomy & Neuroimaging, Kyoto University Graduate School of Medicine, Kyoto, Japan

Jiajia Yang, PhD  
Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University
Okayama, Japan

Hirohiko Imai, PhD  
Department of Informatics, Kyoto University Graduate School of Informatics
Kyoto, Japan

Masaki Fukunaga  
Section of Brain Function Information, National Institute for Physiological Sciences
Aichi, Japan

Hiroki Yamamoto, PhD  
Graduate School of Human and Environmental Studies, Kyoto University
Kyoto, Japan

Chenyu Wang  
Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University
Okayama, Japan

Yinghua Yu  
Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University
Okayama, Japan

Kazuhiko Seki, PhD  
Department of Neurophysiology, National Center of Neurology and Psychiatry
Tokyo, Japan

Takashi Hanakawa, PhD  
Department of Integrated Neuroanatomy & Neuroimaging, Kyoto University Graduate School of Medicine
Kyoto, Japan

Tatsuya Umeda, PhD  
Department of Integrated Neuroanatomy & Neuroimaging, Kyoto University Graduate School of Medicine
Kyoto, Japan

The somatosensory systems share many similarities in anatomical and functional organization across the humans and nonhuman primates. It is well accepted in neuroscience that the primary somatosensory cortex (S1) can be divided into four sub-areas: 3a, 3b, 1, and 2. Area 3b is known to receive direct cutaneous tactile input of the contralateral body parts from the ventral posterolateral nucleus [1,2]. Then, the higher-order processing of tactile information occurs in areas 1 and 2, as well as the bilateral secondary somatosensory cortex (S2) and the somatosensory region of the posterior parietal cortex (PPC). To date, however, how tactile information is processed and updated through the layer-specific circuits across the whole brain remains poorly understood. Marmosets, compared to humans, have relatively thick cerebral cortex disproportionate to their overall brain size. This suggests the possibility of recording cortical layer-specific activities throughout the entire brain in marmosets. In the present study, we used a custom-designed MRI transmit and receiver coil set for marmoset at 7T MRI and acquired whole-brain layer-specific fMRI during tactile stimulation to identify the layer-specific hierarchical somatosensory processing in the marmoset brain.

One adult male common marmoset, weighting 356 g, was used in the experiment. All fMRI experiments were performed in a 7T/20-cm bore magnet (Bruker‐Biospin). A saddle coil with an inner diameter of 80 mm was used as a transmit coil, and the MR signal was acquired from one loop coil placed around the marmoset head (Figure 1A) (Takashima Seisakusho Co., Ltd, Tokyo, Japan). On the day of the experiment, initial sedation was induced by a small dose of ketamine (20mg/kg) and isoflurane (2.5%) inhaled via a facemask. After the marmoset received cannulation of a leg vein, anesthesia was switched to a constant intravenous infusion of dexmedetomidine (5ug/kg/hr) and 0.5% isoflurane. A dose of 20 mg/kg of 30‐nm USPIO particles was injected. CBV weighted fMRI data were obtained with a GE-EPI sequence. FOV, 25.875 x 23 mm2; matrix, 72 x 64; slice thickness, 1 mm; nominal resolution, 0.36 x 0.36 x 1.0 mm3; TE, 9.9 ms; TR, 2000 ms. We used a 30s-on 30s-off block design, and the right hand of the marmoset was stimulated (200 ms air-puff of 0.2 kPa followed by an 800-ms null) during the on blocks. AFNI was used for the fMRI data preprocessing and GLM analysis. Laminar analyses were conducted with LayNii [3]. The study protocol was approved by the experimental animal committee at Kyoto University, Japan.

Initially, we confirmed that the largest CBV signal changes in the gray matter rather than in large vessel regions. Then, we found that right hand tactile stimulation activated the left thalamus, S1 (area 3b, 1/2) and bilateral S2 in the marmoset brain (Figure 1B). We determined the ROIs of area 3b, 1/2, and S2 based on the previous findings [4,5] and generated laminar masks of each ROI. The layer-specific activities of each area are shown in Figure 1C, and we found clear hand corresponding activation in the middle layers of area 3b rather than in the upper and deep layers. In contrast, we did not find a clear middle layer activity peak signature in area 1/2 and S2.

Our custom-designed MRI transmit and receive coil set combined with a USPIO imaging contrast agent at 7T MRI was designed to provide enhanced signal sensitivity and high spatial specificity than BOLD for layer-specific fMRI. We succeeded in recording the layer-specific tactile stimulation-related responses in the marmoset whole brain. The middle layers' peak activity in area 3b was thought to reflect the somatosensory input via the thalamus, and the different layer-specific activity patterns in other areas may reflect the inter-area layer-specific interaction during the somatosensory processing. Our approach is expected to provide a new viewpoint on understanding the bidirectional hierarchical somatosensory processing.

Non-BOLD fMRI ¹

Imaging Methods Other

Perception: Tactile/Somatosensory ²

Cortical Layers

FUNCTIONAL MRI

Somatosensory

Other - laminar fMRI