1: Sample Preparation

Sample Preparation Methods Under MAP2.0 (Current)

Stereotaxic Surgeries: Each animal will receive injections of two different anterograde tracers, biotin dextran amine (BDA)-conjugated with Texas Red (BDA-red) and BDA-conjugated with FITC (BDA-green) (0.5% in PBS, Vector laboratory), simultaneously either at two topologically different locations within the same brain regions (e.g., the cerebral cortex and hippocampus) or adjacent different brain regions giving topographically arranged axonal pathways (e.g., the basolateral and basomedial amygdalar nuclei). Injection sites (Table 3) will be systematically varied in overlapping injection boluses for maximal coverage of fiber projections. For the same animals, we will simultaneously also inject one or two retrograde tracers, cholera toxin subunit B (CTb) conjugated with Alexa fluor 647 (2% in 0.1M sodium PBS, Molecular probes) or fluorogold (FG, 2% in saline, Fluorochrome, LLC), into the major terminal fields of the targeted fiber tracts, which will allow us to validate topology of neurons that project to different terminal fields, e.g. injection of BDA-red into layer 2/3 and BDA-green into layer 5/6 of the somatomotor cortex labeling cortico-cortical projections (red) through the corpus callosum and descending projections through the internal capsule (green). Likewise, we will inject CTb into the contralateral somatosensory cortex and FG into the spinal cord for retrograde labeling of two different populations of cortical projection neurons. A typical surgery will last about 2-3 hours after which mice will be housed for 6-10 days before euthanasia and perfusion with phosphate buffered saline (PBS) and fixation with paraformaldehyde (PFA). MR Imaging: The intact and cleaned cranium (skull and brain) of the animals will be used first to obtain sMRI and HARDI/DTI image volumes. Craniums will be soaked in 5% ProHance® gadolinium MRI contrast mechanism for 5 days then immersed in Fomblin® oil (an embedding medium to limit tissue dehydration) for MR scanning. T2 3D spin-echo MRI images will be acquired using a 7.0T Bruker Avance imaging spectrometer based at the UCLA Brain Mapping Center. A phased array 1H mouse brain coil (see BUDGET JUSTIFICATION) will be employed to maximize MR signal strength and resolution to uniformly distribute transmitted RF signal into the tissue. MR data acquisition parameters will be: TR/TE=300ms/6.8ms, 2 NEX, FOV=50x50x50mm3, matrix=256x128x128, T=288.1ºK. High-angular diffusion volumes will be obtained having 16 diffusion weighted directions. The sequence will include three b-volumes at 0, 500, and 1000s/mm2, respectively, acquired in ~14 hours. The MR system is dedicated for small sample imaging and undergoes a detailed program of field distortion monitoring and shimming to maintain reliable field homogeneity. Mouse brain samples will be specifically placed at field iso-center to ensure positioning at the point of maximal homogeneity. Finally, the DTI sequence is specifically designed to minimize EPI signal distortions. One reason for such distortions in human studies is the relative under-sampling of k-space in order to minimize overall scan duration (e.g. <10mins). In our case, we expect distortions to be minimal since samples will be non-living, the sequence is designed to finely sample k-space, and the extended scan duration will ensure high SNR. While not possible to avoid all residual distortions, LONI possesses some of the most sophisticated and rigorously tested image warping software currently being developed and these methods will be applied to adjust the data accordingly for those minor distortion that remain (see below).

Cryosectioning: Following MR imaging, brains will be extracted from skulls, embedded in OCT, and frozen as described previously [9]. The brains will be attached to a chuck with OCT compound (Sakura), and serial sections will be cut on a modified, high-speed CM3050S cryostat (Leica). 250 brains will be cut on coronal planes at 25µm thickness and collected in 4 series. Additionally, 50 brains will be cut on sagittal planes at 25 µm and also collected in 4 series. We plan to take ~1 hour to section each mouse brain.

Digital Blockface Imaging: Dry objectives of 2-40x magnification will be used to image the face of the embedding block producing images with an in-plane resolution of 8.8 to 0.4µm, respectively. We will employ a fluorescence imaging modality that relies on endogenous autofluorescence labeling of the specimen. To prevent out-of-focus fluorescence from compromising image quality, specimens are embedded in highly light absorbent black paraffin or dark embedding resins; the opacity of the embedding medium can be adjusted to match the in-plane resolution of the objective being used and the section thickness, producing isotropic image volumes. A Polaroid DMCIe digital camera will obtain blockface images prior to each section at a resolution of 1600×1200 (~6.7µm/pixel) at 24-bits. The image stack will be reconstructed into a high-resolution isotropic image volume of ~0.5mm3. The advantage of this approach is that this imaging of modality bridges the mesoscopic scales of MRI, blockface, and subcellular resolution that can be captured by confocal microscopy.

Histology: Four series of sections will be obtained to perform multiple fluorescent immunostaining and other protein markers for cyto-architecture and myelin fibers with different combinations (Fig. 11): 1) BDA-red, BDA-green, CTb-Alexa fluor 647, and fluorogold; 2) NeuN, as a protein marker for cytoarchitecture, will be stained with mouse primary antiserum (Invitrogen, mouse, 1: 1000) and goat anti-mouse Alexa fluor 647); 3) Myelin basic protein (MBP), stained with rat anti-MBP serum (1:1000, abcam) and donkey anti-rat IgG conjugated with Alexa fluor 647); and 4) other neuronal transmitters (e.g., tyrosine hydroxylase or 5-HT) as markers for other major fiber tracts. For the 4 series of sections, the first will be used to visualize BDA-red, BDA-green and fluorogold (or CTb-Alexa fluor 647), with DAPI (blue) count staining. The second and third series will be used to reveal BDA-red and BDA-green, co-stained with either NeuN or MBP. NeuN staining provides improved cytoarchitectonic background over DAPI, revealing more detailed neuronal morphology (e.g., proximal dendrites) of individual neurons (under-resolved in Nissl-stained sections), while MBP can reveal myelin fiber tracts. For certain dopaminergic pathways (e.g., the nigrostriatal tract), MBP staining can be replaced with more specific stains with anti-serum of tyrosine hydroxylase (TH). Finally, the last series will be used to perform Nissl stain to match the cytoarchitecture in the original MAP atlas. Immunohistological processing will take ~4 days.

Microscopy: All histological sections will be digitized using the Olympus NanoZoomer Digital Pathology System (Hamamatsu), which convert slides at high speed under a 20X objective lens with 4 fluorescent colors (DAPI), FITC, Rhodamine, and far-red (Alexa Fluor 647) at 5 z-plane focuses. The resulting images have a spatial resolution of >460 nm/pixel in 24-bit color. Mouse brain sections so digitized constitute datasets of about 60TB and these will be immediately transferred to LONI servers to be viewed soon after scanning via Nanozoomer-compatible WebSlide viewing software. Images will be automatically available in MBAT 4.0 and registered with corresponding MR/DTI and blockface images. Neurolucida ( will be used to assist 3D reconstruction and quantification. Finally, we will also use custom software to measure color in each traced section, pixel distribution, and mean color centroid for each reconstructed axon. Results will be made available via MBAT 4.0 plugins, developed under SA3, and via the MAP2.0 project website.

Sample Preparation Methods Under the Original MAP Protocols (2001-2005)

In vivo MRI
Twelve week-old male C57BL/6 mice (Jackson Laboratories) were initially anesthetized with ketamine/xylzaine and then maintained on isofluorane for the duration of the imaging experiment.

Ex vivo MRI
Twelve week-old male C57BL/6 mice were sacrificed by an overdose of halothane (Sigma) according to procedures approved by the UCLA Animal Research Committee. The animals were intracardially perfused using a Minipuls II peristaltic pump (Gilson) at very low pressure with chilled PBS for approximately two minutes and FormaldeFresh (Fisher) for 15 minutes. The animals were decapitated, soft tissue removed, and the skulls were post-fixed in FormaldeFresh for 16 hours and then scanned.

Blockface and Histology
The brains were removed from post-fixed skulls and further post-fixed in FormaldeFresh for 16 hours. After post-fixation the brains were dipped in a mixture of india ink (Pelikan) and 5% gelatin (Sigma) to simplify segmentation of tissue from background later. The brains were cryoprotected in a solution of 20% sucrose for 16 hours to prevent freezing artifacts. The brains were then embedded in OCT compound (Sakura) at 4° C and snap-frozen at -70° C in a 2-methylbutane/dry ice bath.