Resolving of Crossing Pathways in the Optic Chiasm of Marmoset Monkey using Diffusion Tractography with High Spatial and Angular Resolution

Introduction The structure of the optic chiasm varies widely between the primate and the rodent [1]. The fiber architecture of the marmoset (Callithrix jacchus), a New World primate, is close to that of humans [2]; therefore, the marmoset has been used to study the development of the optic chiasm. Further, the body weight of the marmoset (around 300 g) is similar to that of the rat, but its brain-to-body mass ratio is more than 4 times greater than that of the rat (man, 7.5 > marmoset, 1.7 > rat, 0.4 [3]). In addition, the marmoset has a well-developed visual pathway. Because of these factors, the marmoset is an ideal animal model for performing visual research by using a small magnetic resonance imaging (MRI) scanner. In a previous study, we used manganese-enhanced MRI and diffusion tensor tractography (DTT) to reveal the neuroanatomic features of the visual pathway in the marmoset [4]. However, the DTT technique has some limitations with regard to intravoxel fiber crossing. Recently, increased angular resolution of diffusion MRI, namely, high angular resolution diffusion imaging (HARDI), has been introduced to overcome this limitation. HARDI tractography is expected to be a method for visualizing the fiber pathway more precisely, particularly in the region of the fiber crossing. We performed HARDI with increased spatial resolution of the ex vivo optic chiasm in the marmoset in order to evaluate the fiber architecture of the optic chiasm. Materials & Methods Optic chiasm ex vivo models were obtained from post-mortem fixed brains of adult common marmosets (4 females). Each animal was perfused intracardially with 4% paraformaldehyde, and subsequently, the optic chiasm tissues were removed. Two specimens were soaked in phosphate buffered saline solution with 1 mM Gd-DTPA to optimize diffusion MRI [5]. Histological analysis was performed by staining 1 specimen with luxol fast blue (LFB) for evaluation of the myelinated area. Further, another specimen was used for the evaluation of the axon diameter by means of an electron microscope. All the studies were performed in accordance with animal protection laws and were approved by the Animal Ethical Committee of the Central Institute for Experimental Animals. MRI experiments were performed using a 7T PharmaScan 70/16 system with high gradient strength (300 mT/m). A saddle coil (internal diameter, 22 mm) tuned to 300.5 MHz for proton resonance was used. Diffusion-weighted images (DWIs) were acquired in a 3D DW-SE sequence. The parameters were as follows: TR/TE = 2000 ms/38 ms, spatial resolution = 150(μm), MPG duration time (δ)/separation time (Δ) = 12 ms/20 ms, different 8 b-values of 0–12200 s/mm, and MPG directions = 81 (third-order tessellation of the icosahedrons). The software construct used to perform HARDI analysis was an in-house IDL program based on spherical harmonics [6]. HARDI Tractography was performed using TrackVis [7] with the fiber assignment by continuous tracking (FACT) algorithm.