Elsevier

NeuroImage

Volume 78, September 2013, Pages 196-203
NeuroImage

A segmentation protocol and MRI atlas of the C57BL/6J mouse neocortex

https://doi.org/10.1016/j.neuroimage.2013.04.008Get rights and content

Highlights

  • We present a methodology for delineation of the C57BL/6J mouse neocortex in MRI.

  • We successfully delineated 74 neocortical and cortical-related regions.

  • We calculated mean volumes and contrast intensities for each structure.

Abstract

The neocortex is the largest component of the mammalian cerebral cortex. It integrates sensory inputs with experiences and memory to produce sophisticated responses to an organism's internal and external environment. While areal patterning of the mouse neocortex has been mapped using histological techniques, the neocortex has not been comprehensively segmented in magnetic resonance images. This study presents a method for systematic segmentation of the C57BL/6J mouse neocortex. We created a minimum deformation atlas, which was hierarchically segmented into 74 neocortical and cortical-related regions, making it the most detailed atlas of the mouse neocortex currently available. In addition, we provide mean volumes and relative intensities for each structure as well as a nomenclature comparison between the two most cited histological atlases of the mouse brain. This MR atlas is available for download, and it should enable researchers to perform automated segmentation in genetic models of cortical disorders.

Introduction

The neocortex is the largest component of the telencephalon in the mouse. The histology of neocortical areas, typically composed of six layers of neurons, distinguishes them from the olfactory cortical areas (which contain only three distinct layers) and from the hippocampal regions. Large areas of the mouse neocortex serve as primary receiving areas for somatosensory, visual, auditory, vestibular, taste, and visceral sensations. Cortical areas subserving motor function are also relatively large, and are located rostral to the somatosensory cortex. The remainder of the mouse neocortex comprises the orbitofrontal, the cingulate/retrosplenial, and the parietal association cortical areas.

Segmentation of the mouse neocortex is challenging, but there are obvious transitions between many of the functionally distinct regions. Taken together, the many histological and electrophysiological studies of mouse cortex over the past century reveal a clear picture of areal patterning (Kirkcaldie, 2012). These studies have been based on cytoarchitectonics (Caviness, 1975, Wree et al., 1983), chemoarchitectonics (Franklin and Paxinos, 2008, Hof et al., 2000, Watson and Paxinos, 2010), electrophysiology (Tennant et al., 2011), thalamocortical connections (Jones, 2007) cortical efferent connections (Larsen et al., 2007, Meltzer and Ryugo, 2006) and gene expression studies (Bohland et al., 2010, Hawrylycz et al., 2010, Ng et al., 2009). The most comprehensive modern maps of the C57BL mouse cortex are found in the stereotaxic atlases of Paxinos and Franklin (2013) and Dong (2008). The boundaries identified in these two atlases are very similar in all but a few instances. The present study makes extensive use of the maps in these two atlases.

The shape and size of the primary regions (such as the visual, auditory and somatosensory regions) in mouse neocortex can vary significantly across strains (Hof et al., 2000). For example, inter-strain variations have been found in total cortical volume (Gaglani et al., 2009), and there are more localized differences in the size of the barrel field (Jan et al., 2008, Li et al., 2005) and the visual cortex (Airey et al., 2005, Airey et al., 2006). More marked alterations to cortical morphometry and connectivity occur in natural mutants and transgenic models of neurological disorders such as Huntington's (Lerch et al., 2008a, Lerch et al., 2008b, Zhang et al., 2010) and Alzheimer's disease (Benveniste et al., 2007, Lau et al., 2008b).

Magnetic resonance imaging (MRI) has become an important technique for examining changes in brain structure in mouse models of neurological disorders (Benveniste et al., 2007, Lau et al., 2008b, Nieman et al., 2005, Pitiot et al., 2007). The combination of image registration methods and voxel-based statistical parametric mapping techniques has been used to identify phenotypic differences between disease and control animals (Lerch et al., 2008b, Sawiak et al., 2009). While comparisons at the cortical level can already be performed using established whole-brain mouse atlases, a detailed atlas of the C57BL/6J neocortex is needed to allow researchers to map the changes at a regional and functional level. Therefore, in this paper we present a detailed protocol for segmenting the C57BL/6J cortex using high-resolution MRI and a minimum deformation atlas made up of averaged data from 18 individuals. This atlas is made freely available to assist future researchers in automatic segmentation of the mouse neocortex.

Section snippets

C57BL/6J mouse brain preparation and magnetic resonance imaging

Eighteen animals (male, 12 week old) were perfused and fixed with 4% paraformaldehyde and 0.1% Magnevist® (gadopentetate dimeglumine, Bayer HealthCare Pharmaceuticals Inc., Wayne, NJ, USA) in phosphate buffer (PB). Brains were extracted and incubated in 0.1% Magnevist/PB for 4 days, placed in Fomblin (Solvay Solexis, Milan, Italy) and imaged on a 16.4 T (89 mm bore diameter) Bruker micro-imaging system (Bruker Biospin, Karlsruhe, Germany) using a 15 mm SAW coil (M2M Imaging, USA). MRI data were

Results

The minimum deformation atlas represents the average spatial positioning and intensity of each structure. Fig. 1 demonstrates the superior signal and contrast to noise ratios in comparison to the image from a single brain, features which assisted the delineation of cortical structures. The improved clarity indicates quality of the registration.

Table 1 lists the structures, abbreviations according to Paxinos and Franklin (2013), their corresponding color, average volume, and average signal

Discussion

We have developed a segmentation protocol for the mouse neocortex with operational criteria based upon signal intensity in T2*-weighted high-resolution magnetic resonance scans. Employing our methodology, we then segmented a minimum deformation model of the C57BL/6J mouse brain (Janke et al., 2012) to create an atlas of the neocortex consisting of 74 structures. This is a substantially greater number of structures than in previous MRI-based atlases, which either left the cortex as one structure

Conclusion

We provide detailed guidelines for segmenting the isocortex on magnetic resonance images of the C57BL/6J mouse brain as well as mean volumes and relative image intensities. To facilitate its use, the minimum deformation atlas and two hierarchical label fields one with only major structural regions segmented and one containing the individual regions are available in a variety of imaging formats for download at http://www.imaging.org.au/AMBMC/Cortex. By following our image acquisition protocol,

Funding

This work was supported by the National Health and Medical Research Council (NHMRC) of Australia (grant no. 436673) and a National Computational Infrastructure Grant (grant dc0). We would like to thank the National Imaging Facility (NIF) and the Queensland NMR Network (QNN) for access to the 16.4 T scanner and technical support. In addition, we would like to thank Dr. Marianne Keller and the other members of AMBMC.

References (48)

  • J.P. Lerch et al.

    Automated deformation analysis in the YAC128 Huntington disease mouse model

    Neuroimage

    (2008)
  • Y. Ma et al.

    A three-dimensional digital atlas database of the adult C57BL/6J mouse brain by magnetic resonance microscopy

    Neuroscience

    (2005)
  • A. MacKenzie-Graham et al.

    Cerebellar cortical atrophy in experimental autoimmune encephalomyelitis

    Neuroimage

    (2006)
  • S.J. Sawiak et al.

    Voxel-based morphometry in the R6/2 transgenic mouse reveals differences between genotypes not seen with manual 2D morphometry

    Neurobiol. Dis.

    (2009)
  • A.A. Sharief et al.

    Automated segmentation of the actively stained mouse brain using multi-spectral MR microscopy

    Neuroimage

    (2008)
  • J.Y. Zhang et al.

    Longitudinal characterization of brain atrophy of a Huntington's disease mouse model by automated morphological analyses of magnetic resonance images

    Neuroimage

    (2010)
  • D.C. Airey et al.

    Variation in the cortical area map of C57BL/6J and DBA/2J inbred mice predicts strain identity

    BMC Neurosci.

    (2005)
  • D.C. Airey et al.

    Geometric morphometrics defines shape differences in the cortical area map of C57BL/6J and DBA/2J inbred mice

    BMC Neurosci.

    (2006)
  • D. Barazany et al.

    Visualization of cortical lamination patterns with magnetic resonance imaging

    Cereb. Cortex

    (2012)
  • H. Benveniste et al.

    Anatomical and functional phenotyping of mice models of Alzheimer's disease by MR microscopy

    Ann. N. Y. Acad. Sci.

    (2007)
  • V.S. Caviness

    Architectonic map of neocortex of the normal mouse

    J. Comp. Neurol.

    (1975)
  • H.W. Dong

    The Allen Reference Atlas: A Digital Color Brain Atlas of the C57BL/6J Male Mouse

    (2008)
  • K. Franklin et al.

    The Mouse Brain in Stereotaxic Coordinates

    (2008)
  • S.M. Gaglani et al.

    The genetic control of neocortex volume and covariation with neocortical gene expression in mice

    BMC Neurosci.

    (2009)
  • Cited by (150)

    View all citing articles on Scopus
    1

    Contributed equally to this work.

    View full text