Abstract

The impact of developmental ablation of Pax6 function on morphology and functional connectivity of the adult cerebrum was studied in cortex-specific Pax6 knockout mice (Pax6cKO) using structural magnetic resonance imaging (MRI), manganese-enhanced MRI, and diffusion tensor MRI in conjunction with fiber tractography. Mutants presented with decreased volumes of total brain and olfactory bulb, reduced cortical thickness, and altered layering of the piriform cortex. Tracking of major neuronal fiber bundles revealed a disorganization of callosal fibers with an almost complete lack of interhemispheric connectivity. In Pax6cKO mice intrahemispheric callosal fibers as well as intracortical fibers were predominantly directed along a rostrocaudal orientation instead of a left–right and dorsoventral orientation found in controls. Fiber disorganization also involved the septohippocampal connection targeting mostly the lateral septal nucleus. The hippocampus was rostrally extended and its volume was increased relative to that of the forebrain and midbrain. Manganese-induced MRI signal enhancement in the CA3 region suggested a normal function of hippocampal pyramidal cells. Noteworthy, several morphologic disturbances in gray and white matter of Pax6cKO mice were similar to observations in human aniridia patients. The present findings indicate an important role of Pax6 in the development of both the cortex and cerebral fiber connectivity.

Introduction

The mammalian neocortex is radially organized in 6 layers and tangentially arrayed in numerous functional areas. The molecular and cellular mechanisms leading to this organization have only recently begun to be understood (Guillemot 2005; reviewed by Guillemot et al. 2006; Mallamaci and Stoykova 2006; O'Leary and Sahara 2008; Pontious et al. 2008). The main population of neurons making up the cerebral neocortex are the glutamatergic projection pyramidal neurons. These neurons include 2 main classes: one class of neurons whose axons predominantly cross to the contralateral hemisphere, thus building the corpus callosum, and a second class of neurons projecting to subcortical targets (Molyneaux et al. 2007). The first group of projection neurons are distributed across all cortical layers, most abundantly located in layers II–IV. The second class is primarily present in cortical layers V and VI. The neurons with distinct layer identity are born during a specific developmental time window: the neurons of layer V and VI are generated by progenitors in cortical ventricular zone (VZ) at early stages (in mouse between embryonic day E10-E14), whereas layer II to layer IV neurons are produced predominantly by late progenitors (E15-E18) located in the second germinative zone of the cortex, the subventricular zone (McConnell 1985, 1988); reviewed by (Gotz and Huttner 2005).

In addition to the timing of neuronal birth, neuronal layer identity seems to depend on the expression of distinct transcription factors in cortical progenitors. Analyses of mouse mutants for the bHLH transcription factors Ngn1/Ngn2 and for the paired-domain transcription factor Pax6 revealed a severe malformation specifically of the lower or upper neuronal layers, implicating that the differentiation of these 2 neuronal types is under specific control of Ngn2 and Pax6 in the pallial progenitors, respectively (Schuurmans et al. 2004). Furthermore, recent evidence indicates that the projection trajectory of the pyramidal neurons depends on the expression of specific sets of transcription factors in the mature neurons. Thus, although the expression of transcription factors Fezf2 and Ctip2 defines the fate of the subcortical projection neurons, transcription factor Satb2 appears to act as a postmitotic determinant of neurons with callosal projections (Arlotta et al. 2005; Britanova et al. 2005, 2008; Chen, Rasin, et al., 2005; Chen, Scaevitz, et al., 2005; Molyneaux et al. 2005).

In developing cortex transcription factor Pax6 is highly expressed in VZ of the rostrolateral cortex and only faintly expressed in progenitors of the mediocaudal cortex (Stoykova et al. 1997; Walther and Gruss 1991). When Pax6 is not functional as in the homozygous Small eye (Sey/Sey) embryo, rostral cortical domains are shrunken in expense of expanded caudal domains suggesting that Pax6 plays a role in the regionalization of cortical functions (Bishop et al. 2002; Muzio et al. 2002). However, despite being changed in size, the rostrocaudal cortical domains show a normal thalamocortical connectivity indicating that Pax6 does not affect the area identity (Pinon et al. 2008). In Pax6 deficiency, the production of mutant cells in developing cortex is severely reduced possibly due to premature exits of the pallial progenitors from the mitotic cycle causing a depletion of the progenitor pool (Quinn et al. 2007). Notably, even haploinsufficiency of Pax6 in aniridia patients causes multiple neurologic and cognitive disabilities the underlying mechanisms of which are still unknown (Sisodiya et al. 2001).

Although the cortical neurons are born during embryogenesis, they reach their final location in distinct layers only postnatally (in mouse at postnatal day P8). Because homozygous Pax6-deficient mice die at birth, the study of consequences of the developmental Pax6 mutation on cerebral structure is severely hampered. In order to examine the abnormalities of the adult brain after elimination of Pax6 function during development, we took advantage of the recently generated cortex-specific conditional Pax6 knockout mice (Pax6cKO) reported to reach adolescence (Pinon et al. 2008). The brain phenotype of the Pax6cKO mutant was studied by high-resolution 3D T1- and T2-weighted MRI, manganese-enhanced MRI, and diffusion tensor MRI combined with fiber tractography.

MRI of the brain is widely used in humans for medical diagnosis as well as for neurobiologic research of both humans and mice, for example see (Natt and Frahm 2005). Among other advantages, in vivo 3D MRI allows for the quantification of volume changes of specific brain structures without shrinking artifacts which may occur during fixation for histopathology. Manganese-enhanced MRI exploits the fact that Mn2+ ions shorten the T1 relaxation time of affected water protons in tissue and most importantly behave like Ca2+ ions because of their similar valence and size. It has been shown that Mn2+ ions can enter neurons via voltage-gated calcium channels and cross the intact blood–brain barrier by diffusion and carrier-mediated processes (Yokel and Crossgrove 2004). Systemic application leads to a heterogeneous tissue-specific signal enhancement in T1-weighted 3D MRI which most likely reflects an activity-dependent intracellular enrichment of Mn2+ (Watanabe et al. 2004). Diffusion tensor MRI, also termed diffusion tensor imaging (DTI), maps the molecular mobility of water molecules in tissue which may be quantified as diffusion properties. The MRI-accessible directional dependence of the water diffusivity in highly organized structures like neuronal fiber bundles allows for a 3D virtual reconstruction of major axonal connections in vivo.

The purpose of this study was to investigate the impact of cortical Pax6 embryonic deficiency on brain morphology and neuronal connectivity in the adult mouse. Some of our findings are similar to deficiencies reported in aniridia patients and may give further insight into the role of Pax6 in cerebral development.

Methods

Animals

In order to generate mice in which Pax6 was inactivated at the onset of neurogenesis of developing cortex, homozygous Pax6 floxed mice (Pax6fl/fl) (Ashery-Padan 2002) and Emx1IRES-Cre mice (Gorski et al. 2002) were bred to produce Pax6fl/f;Emx1IRES-Cre mutant progeny, (named Pax6cKO), as described by (Pinon et al. 2008). Litter-mates with genotypes Pax6fl/fl or Pax6fl/+g were used as controls. As shown previously, the residual Pax6 expression in the cortex of Pax6cKO mice was less then 10% of control sample (Pinon et al. 2008).

All experiments were performed in compliance with relevant laws and institutional guidelines. Animal experiments were approved by local authorities. For MRI the mice were initially anesthetized using a chamber pervaded with 5% isoflurane in oxygen. Subsequently, the mice were intubated and kept under anesthesia with 1–1.5% isoflurane in a 1:1.5-mixture of oxygen and ambient air. Respiration was monitored by a signal derived from a home-made pressure transducer fixed to the animal's chest. The animals were placed in a prone position on a purpose-built plate holder with an adjustable nose cone. The rectal temperature was held constant at 36 ± 0.5 °C by water blankets.

In Vivo MRI

Three Pax6 null mutants and 3 control litter-mates underwent structural MRI, DTI, and manganese-enhanced MRI at the age of 4 months. Two animals of each group were also studied at the age of 11 months using structural MRI and DTI.

Structural MRI was performed at 2.35 T using an MRBR 4.7/400 magnet (Magnex, Scientific, Abingdon, UK) equipped with a DBX spectroscopy and imaging system (Bruker Biospin, Ettlingen, Germany). A Helmholtz coil (diameter 100 mm) was used for radiofrequency excitation and combined with an elliptical surface coil (20 × 12 mm) for signal detection. T1-weighted images (3D FLASH, repetition time/echo time [TR/TE] = 17/7.6 ms, flip angle 25°, matrix 128 × 256 × 256) and T2-weighted images (3D FSE, TR/TE = 3000/61 ms, interecho spacing = 14.4 ms, matrix 128 × 128 × 128) were acquired at 117-μm isotropic spatial resolution.

DTI was performed at 9.4 T (Bruker Biospin) with the use of a birdcage resonator (inner diameter 70 mm) for excitation and a saddle-shaped surface coil for signal detection (Bruker Biospin). Acquisitions were based on half Fourier diffusion-weighted single-shot STEAM MRI yielding 125 × 125 × 500 μm3 resolution (Boretius et al. 2007). Diffusion preparation involved 2 b values (0 and 1000 s/mm2) in 12 different directions (6 different gradients and their opposite values).

Manganese-enhanced MRI was performed at both 2.35 and 9.4 T using T1-weighted 3D FLASH MRI. After the acquisition of precontrast images all animals received 40 mg/kg body weight MnCl2 (Sigma-Aldrich) intraperitoneally. Postcontrast images were acquired at 24 and 48 h after injection (for 2.35 T parameters see above). T1-weighted MRI at 9.4 T was performed with the use of an adapted sequence (3D FLASH, TR/TE = 17/4.0 ms, flip angle 25°) yielding 100-μm isotropic resolution.

Data Analysis

Volumes of total brain, forebrain (excluding the olfactory bulb) and midbrain, olfactory bulb, hippocampal formation, lateral ventricles, and cerebellum were determined using T1-weighted manganese-enhanced 3D MRI and manual segmentation of respective areas (Amira Software, Visage Imaging, Berlin, Germany). Due to the lack of a clear MRI separation of forebrain and midbrain the volumes of these areas were combined. Mean cortical thickness relative to brain height was assessed on 2 parasagittal T1-weighted images (about 0.5 mm left and right to the midline) using multiple measurements in rostral, intermediate, and caudal regions of the cortex (Supplementary Fig. 1).

Preprocessing of DTI data involved a linear interpolation by doubling the acquisition matrix to twice the resolution in all 3 dimensions followed by the application of a mild 3D Gaussian filter (half width corresponding to one voxel). The calculation of the diffusion tensor employed a weighted linear least-squares algorithm. Further details about the DTI analysis (Küntzel 2007) as well as the reconstruction of fiber tracks in mouse brain have been described elsewhere (Boretius et al. 2007). Briefly, maps of the main diffusion direction were color-coded and superimposed onto corresponding maps of the fractional anisotropy. Estimates of axonal fiber projections were computed by the fiber assignment by continuous tracking algorithm (Mori et al. 1999) after subdividing each start voxel to contain 9 seed points. Criteria for terminating the tracking of individual fibers included an anisotropy threshold (values below 0.15) and a maximum stiffness condition, so that the tracking was terminated when the diffusion directions in consecutive steps differed by more than 40°. Region-of-interest (ROI) analyses of the fractional anisotropy, axial diffusivity (largest eigenvalue of the diffusion tensor), and radial diffusivity (mean value of the eigenvalues perpendicular to the largest eigenvalue) were performed in the lateral and medial part of the corpus callosum on an axial section at the level of the anterior commissure.

Results

Brain Anatomy

Adult Pax6cKO mutants presented with marked structural changes of the brain. MRI demonstrated a flatter and smaller neocortex as compared with controls (Fig. 1) in line with a generalized reduction of cortical thickness (Fig. 2) and total brain size including a reduced volume of the fore- and midbrain, and the olfactory bulb. The size of the cerebellum was similar to that of controls. The shape of the hippocampal formation was altered by propagation into a more rostral direction. The septal region as well as the fornix, hippocampal fimbria, and lateral ventricles were shifted into a rostroventral direction. The septum was compressed and deformed, whereas the septo-hippocampal connection via the fornix and fimbria appeared reduced or even interrupted (Fig. 2, circle).

Figure 1.

MRI-based surface reconstructions of the brain of a control and Pax6cKO mouse in vivo. Mutants exhibit a clear reduction of total brain size including a reduced volume of the forebrain and midbrain (yellow) as well as olfactory bulb (red, red arrow). The size of the cerebellum (violet) is similar to that of controls. The shape of the hippocampal formation (dark yellow) is altered as it propagates into a more rostral direction (white arrow). Cortical thickness is generally reduced (yellow arrows).

Figure 1.

MRI-based surface reconstructions of the brain of a control and Pax6cKO mouse in vivo. Mutants exhibit a clear reduction of total brain size including a reduced volume of the forebrain and midbrain (yellow) as well as olfactory bulb (red, red arrow). The size of the cerebellum (violet) is similar to that of controls. The shape of the hippocampal formation (dark yellow) is altered as it propagates into a more rostral direction (white arrow). Cortical thickness is generally reduced (yellow arrows).

Figure 2.

T1-weighted sagittal images of a control and Pax6cKO mouse in vivo. In mutants the cortical thickness is reduced (arrows) and the septal area is altered (circle). ls = lateral septal nucleus, fi = hippocampal fimbria, hp = hippocampus.

Figure 2.

T1-weighted sagittal images of a control and Pax6cKO mouse in vivo. In mutants the cortical thickness is reduced (arrows) and the septal area is altered (circle). ls = lateral septal nucleus, fi = hippocampal fimbria, hp = hippocampus.

MRI of the brain of Pax6cKO mice also revealed pronounced alterations of the corpus callosum (Fig. 3). This major white matter structure not only appeared smaller in frontal areas, but was almost entirely missing in caudal locations (Supplementary Fig. 2). The ventral hippocampal commissure appeared rostrally shifted (Fig. 3 and Supplementary Fig. 2). These findings are further supported by DTI and fiber tractography (see below).

Figure 3.

T1-weighted (T1w) and T2-weighted (T2w) horizontal images of a control and Pax6cKO mouse in vivo. In mutants the entire cortex is thinner, the genu of the corpus callosum (ccg) is hardly detectable, and the hippocampal fimbria (fi) and lateral ventricles are rostrally displaced (arrows). vhc = ventral hippocampal commissure.

Figure 3.

T1-weighted (T1w) and T2-weighted (T2w) horizontal images of a control and Pax6cKO mouse in vivo. In mutants the entire cortex is thinner, the genu of the corpus callosum (ccg) is hardly detectable, and the hippocampal fimbria (fi) and lateral ventricles are rostrally displaced (arrows). vhc = ventral hippocampal commissure.

In mutants quantitative determinations of the mean cortical thickness revealed a reduction which attenuated from rostral (−30%) to intermediate (−20%) and caudal areas (−9%) (Table 1 and Supplementary Fig. 1). Quantitative analyses also confirmed volume reductions of the forebrain/midbrain region (−27%) as well as olfactory bulb (−43%) in Pax6cKO mice (Table 1). The hippocampal formation showed similar volumes compared with controls, although the size varied considerably across mutants. However, the ratio of the hippocampal volume to that of forebrain and midbrain was increased from 0.07 in controls to 0.11 in mutants. The size of the lateral ventricles varied considerably across mutants, but the mean value was similar to that of controls.

Table 1

Cortical thickness and cerebral volumes of Pax6cKO mice and controls

 Control
 
Pax6cKO
 
 Mean Range Mean Range 
Cortical thickness (mm) 
    (Relative units)a 
        Rostralb 1.89 1.82–1.94 1.15 1.12–1.19 
 (0.307) (0.302–0.315) (0.216) (0.216–0.218) 
        Intermediate 1.29 1.24–1.34 0.89 0.87–0.94 
 (0.210) (0.206–0.213) (0.168) (0.165–0.173) 
        Caudal 1.07 1.03–1.12 0.84 0.73–0.96 
 (0.174) (0.170–0.176) (0.158) (0.138–0.176) 
Volume (mm3
        Forebrain and midbrain 324 307–338 235 228–248 
        Olfactory bulb 23 22–24 13 12–14 
        Hippocampal formation 23 22–24 25 22–28 
        Cerebellum 60 59–61 60 56–61 
        Lateral ventricles 3–7 
        Total brain 407 390–408 307 297–319 
 Control
 
Pax6cKO
 
 Mean Range Mean Range 
Cortical thickness (mm) 
    (Relative units)a 
        Rostralb 1.89 1.82–1.94 1.15 1.12–1.19 
 (0.307) (0.302–0.315) (0.216) (0.216–0.218) 
        Intermediate 1.29 1.24–1.34 0.89 0.87–0.94 
 (0.210) (0.206–0.213) (0.168) (0.165–0.173) 
        Caudal 1.07 1.03–1.12 0.84 0.73–0.96 
 (0.174) (0.170–0.176) (0.158) (0.138–0.176) 
Volume (mm3
        Forebrain and midbrain 324 307–338 235 228–248 
        Olfactory bulb 23 22–24 13 12–14 
        Hippocampal formation 23 22–24 25 22–28 
        Cerebellum 60 59–61 60 56–61 
        Lateral ventricles 3–7 
        Total brain 407 390–408 307 297–319 
a

Cortical thickness relative to anterior–posterior brain dimension (dotted line in Supplementary Fig. 1).

b

Regions indicated in Supplementary Figure 1.

Manganese-Enhanced MRI

Manganese-induced MRI signal enhancements at 24 and 48 h after MnCl2 administration indicated a similar Ca2+ uptake in the hippocampal formation of mutants and controls. In agreement with previous studies (Watanabe et al. 2002), the CA3 region but not CA1 showed a marked manganese accumulation. Noteworthy, the location of the enhanced CA3 region provided further evidence for a rostral prolongation of the hippocampus in Pax6cKO mice (Fig. 4).

Figure 4.

Manganese-enhanced T1-weighted in vivo MRI of a control and Pax6cKO mouse 24 h after MnCl2 administration in (A) horizontal, (B) sagittal, and (C, D) axial sections. Apart from a rostral displacement of the hippocampus (AC, arrows), mutants exhibit a signal enhancement of the dentate gyrus (dg) and hippocampal CA3 region similar to that of controls.

Figure 4.

Manganese-enhanced T1-weighted in vivo MRI of a control and Pax6cKO mouse 24 h after MnCl2 administration in (A) horizontal, (B) sagittal, and (C, D) axial sections. Apart from a rostral displacement of the hippocampus (AC, arrows), mutants exhibit a signal enhancement of the dentate gyrus (dg) and hippocampal CA3 region similar to that of controls.

The alteration of the septal area already found in T1- and T2-weighted images was confirmed by differences in manganese-enhanced MRI. Although controls exhibited a clear signal increase in the lateral septal nuclei, Pax6cKO mice presented with a much less prominent enhancement. Instead, the entire nucleus appeared rostrally shifted and reduced in volume as demonstrated by a 3D surface reconstruction (Supplementary Fig. 3).

Independent of the reduced size of the olfactory bulb, Pax6cKO mice showed a similar manganese-induced MRI signal enhancement of the glomerular and mitral cell layers as in controls (Fig. 5A–C). These layers were clearly separated from the less enhanced external plexiform layer. The piriform cortex, however, apparently lost its 3-layer structure in Pax6cKO mutants (Fig. 5E). The distinction of the pyramidal cell layer from both the unenhanced adjacent molecular layer of the piriform cortex and the enhanced lateral olfactory tract in controls was not seen in Pax6cKO mice. Instead, mutants exhibited a diffuse signal enhancement in the piriform area that precluded any reliable layer discrimination.

Figure 5.

Manganese-enhanced T1-weighted in vivo MRI of a control and Pax6cKO mouse 24 h after MnCl2 administration in (A) horizontal and (BE) axial sections. In both controls and mutants the glomerular layer (gl) and mitral cell layer (mi) of the olfactory bulb (AC) are clearly enhanced and separated from the darker external plexiform layer (epl). The granule layer (gr) is slightly enhanced, whereas white matter structures such as the rostral fibers of the anterior commissure (ac) appear dark. Furthermore, the olfactory nerve layer (onl) and the lateral olfactory tract (lot) are enhanced. In mutants, the size of the olfactory bulb and the prefrontal cortex are dramatically reduced. The piriform cortex (E) lost its 3-layer structure in Pax6cKO mutants. The distinction of the pyramidal layer (pcp) from both the unenhanced adjacent molecular layer (pcm) and the enhanced lateral olfactory tract in controls was not seen in Pax6cKO mice. ao = accessory olfactory bulb, aon = anterior olfactory nucleus.

Figure 5.

Manganese-enhanced T1-weighted in vivo MRI of a control and Pax6cKO mouse 24 h after MnCl2 administration in (A) horizontal and (BE) axial sections. In both controls and mutants the glomerular layer (gl) and mitral cell layer (mi) of the olfactory bulb (AC) are clearly enhanced and separated from the darker external plexiform layer (epl). The granule layer (gr) is slightly enhanced, whereas white matter structures such as the rostral fibers of the anterior commissure (ac) appear dark. Furthermore, the olfactory nerve layer (onl) and the lateral olfactory tract (lot) are enhanced. In mutants, the size of the olfactory bulb and the prefrontal cortex are dramatically reduced. The piriform cortex (E) lost its 3-layer structure in Pax6cKO mutants. The distinction of the pyramidal layer (pcp) from both the unenhanced adjacent molecular layer (pcm) and the enhanced lateral olfactory tract in controls was not seen in Pax6cKO mice. ao = accessory olfactory bulb, aon = anterior olfactory nucleus.

White Matter Fiber Organization

Maps of the fractional anisotropy revealed a reduced directionality of water diffusion (less bright) in the corpus callosum mainly for central regions in frontal brain (Fig. 6). In the more caudal sections no fibers were detectable in the central corpus callosum. The corresponding color-coded maps of the main diffusion direction demonstrate a left–right fiber orientation (red) in the corpus callosum of controls, whereas respective fibers of mutants were preferentially directed in a rostrocaudal direction (blue). The anterior commissure which emerges as a small left–right connection in controls could only be found in one mutant and with much reduced size. A quantitative analysis (Table 2) confirmed reduced anisotropy values, a reduced axial diffusivity, and an increased radial diffusivity in the frontomedial and frontolateral parts of mutants indicating a reduced number of callosal fibers.

Table 2

Fractional anisotropy, axial diffusivity, and radial diffusivity of the frontal corpus callosum of Pax6cKO mice and controls

 Fractional anisotropy
 
Axial diffusivity/10−6 mm2/s
 
Radial diffusivity/10−6 mm2/s
 
 Lateral Medial Lateral Medial Lateral Medial 
Pax6cKO 
    #1 0.38 0.41 890 1085 512 573 
    #2 0.31 0.41 853 1051 543 559 
    #3 0.38 0.42 882 1070 500 562 
    Mean 0.36 0.41 875 1068 518 565 
Control 
    #1 0.50 0.50 933 1156 433 553 
    #2 0.50 0.50 891 1328 396 611 
    #3 0.49 0.55 1005 1241 462 561 
    Mean 0.50 0.52 943 1242 430 575 
 Fractional anisotropy
 
Axial diffusivity/10−6 mm2/s
 
Radial diffusivity/10−6 mm2/s
 
 Lateral Medial Lateral Medial Lateral Medial 
Pax6cKO 
    #1 0.38 0.41 890 1085 512 573 
    #2 0.31 0.41 853 1051 543 559 
    #3 0.38 0.42 882 1070 500 562 
    Mean 0.36 0.41 875 1068 518 565 
Control 
    #1 0.50 0.50 933 1156 433 553 
    #2 0.50 0.50 891 1328 396 611 
    #3 0.49 0.55 1005 1241 462 561 
    Mean 0.50 0.52 943 1242 430 575 
Figure 6.

Consecutive rostral to caudal axial maps of the fractional anisotropy (FA) and color-coded main diffusion direction of a control and Pax6cKO mouse in vivo. In mutants, most of the interhemispheric fiber connections via the corpus callosum (cc) are missing, whereas FA values are reduced in rostromedial parts of the cc and similar to gray matter in caudomedial parts. Instead of a normal left–right orientation (red), remaining midline callosal fibers in frontal sections project along a rostrocaudal direction (blue). The very small red fiber bridge (arrow) most likely belongs to the rostrally displaced ventral hippocampal fissure. The anterior commissure (ac) is reduced in Pax6cKO mutants. cpd = cerebral peduncle, cing = cingulum, fi = hippocampal fimbria, int = internal capsule, opt = optic tract. Directional color code: red = left–right, blue = rostral–caudal, green = anterior–posterior.

Figure 6.

Consecutive rostral to caudal axial maps of the fractional anisotropy (FA) and color-coded main diffusion direction of a control and Pax6cKO mouse in vivo. In mutants, most of the interhemispheric fiber connections via the corpus callosum (cc) are missing, whereas FA values are reduced in rostromedial parts of the cc and similar to gray matter in caudomedial parts. Instead of a normal left–right orientation (red), remaining midline callosal fibers in frontal sections project along a rostrocaudal direction (blue). The very small red fiber bridge (arrow) most likely belongs to the rostrally displaced ventral hippocampal fissure. The anterior commissure (ac) is reduced in Pax6cKO mutants. cpd = cerebral peduncle, cing = cingulum, fi = hippocampal fimbria, int = internal capsule, opt = optic tract. Directional color code: red = left–right, blue = rostral–caudal, green = anterior–posterior.

The loss of interhemispheric connectivity in Pax6cKO mice is further documented by DTI-based fiber tractography. In controls the reconstruction of major fiber pathways revealed extensive connections between the 2 hemispheres as well as projections into the cingulate cortex and motor cortex (Fig. 7). In mutants, however, no fibers could be detected that cross the midline or project into frontal cortical areas. Instead, Pax6cKO mice exhibited strong callosal fiber pathways in a rostrocaudal orientation. Fiber disorganization was also seen in the septal area which in mutants was characterized by an enormous sprouting of fibers (Fig. 7). Further alterations of fiber connectivity were detected in the enlarged and rostrally displaced hippocampal fimbria. This observation became even more apparent by region-to-region tracking from the caudal edge to the rostral part of the fimbria (Fig. 7). Although controls presented with only a small curved fiber bundle, mutants exhibited a much thicker and less curved structure.

Figure 7.

3D views of major fiber pathways of a control and Pax6cKO mouse in vivo. (Corpus callosum) Although controls present with a pronounced connection of the 2 hemispheres via the corpus callosum (cc) and projections into the cingulate and motor cortex, mutants reveal a strong rostrocaudal fiber orientation and a nearly complete lack of interhemispheric projections. (Septal area) Mutants are characterized by a vast disorganization of fibers crossing the septal region. (Fimbria) In mutants the enlarged hippocampal fimbria (fi) is shifted rostrally and exhibits a much thicker and less curved fiber structure as demonstrated by region-to-region fiber tracking (bottom). cing = cingulum, fx = fornix. Directional color code: red = left–right, blue = rostral–caudal, green = anterior–posterior.

Figure 7.

3D views of major fiber pathways of a control and Pax6cKO mouse in vivo. (Corpus callosum) Although controls present with a pronounced connection of the 2 hemispheres via the corpus callosum (cc) and projections into the cingulate and motor cortex, mutants reveal a strong rostrocaudal fiber orientation and a nearly complete lack of interhemispheric projections. (Septal area) Mutants are characterized by a vast disorganization of fibers crossing the septal region. (Fimbria) In mutants the enlarged hippocampal fimbria (fi) is shifted rostrally and exhibits a much thicker and less curved fiber structure as demonstrated by region-to-region fiber tracking (bottom). cing = cingulum, fx = fornix. Directional color code: red = left–right, blue = rostral–caudal, green = anterior–posterior.

In the cortex, maps of the main diffusion direction indicated a predominance of rostrocaudal fiber directions (blue) in the Pax6cKO mutants as opposed to dorsoventral (radial) directions (green) in controls (Fig. 6). This is further illustrated by a pixelwise visualization of fiber directions in the form of lines or points (Supplementary Fig. 4).

Follow-Up Studies

T1-weighted MRI of a Pax6cKO mouse at 4 and 11 months of age demonstrated the persistence of cerebral alterations in mutants over a period of 7 months (Supplementary Fig. 5). In particular, old mutants still show a thinner cortex and lack a clear interhemispheric connectivity (Fig. 8). On the other hand, the preferential rostrocaudal fiber orientation in the cortex at 4 months of age slightly changed to a more dorsoventral orientation at 11 months, although still different to those of age-matched controls (Fig. 8). Small changes were also seen in the septal region of Pax6cKO mice where fibers changed from a rostrodorsal to ventrocaudal orientation at 4 months of age to a dorsoventral orientation at 11 months (Fig. 9).

Figure 8.

Maps of the main diffusion direction of a Pax6cKO mouse in vivo at 4 and 11 months of age in comparison to an 11-month-old control. The maps either display directions with the use of a color code or as blue vectors in each voxel. The white rectangles identify magnified areas. In mutants, the rostrocaudal orientation of callosal fibers and reduced size of the anterior commissure (ac) persists over a period of 7 months, whereas in the cortex the diffusion direction changes from a predominantly rostrocaudal orientation (4 months) to a slightly more dorsoventral orientation (11 months). cc = corpus callosum. Directional color code: red = left–right, blue = rostral-caudal, green = anterior–posterior.

Figure 8.

Maps of the main diffusion direction of a Pax6cKO mouse in vivo at 4 and 11 months of age in comparison to an 11-month-old control. The maps either display directions with the use of a color code or as blue vectors in each voxel. The white rectangles identify magnified areas. In mutants, the rostrocaudal orientation of callosal fibers and reduced size of the anterior commissure (ac) persists over a period of 7 months, whereas in the cortex the diffusion direction changes from a predominantly rostrocaudal orientation (4 months) to a slightly more dorsoventral orientation (11 months). cc = corpus callosum. Directional color code: red = left–right, blue = rostral-caudal, green = anterior–posterior.

Figure 9.

3D views of major fiber pathways of a control and Pax6cKO mouse in vivo in the septal area at 4 and 11 months of age. Although controls present with reproducible fiber structures over a period of 7 months, mutants reveal a partial reorganization which refers to a strengthening of dorsoventrally oriented fibers in conjunction with a rostral displacement of the ventral part.

Figure 9.

3D views of major fiber pathways of a control and Pax6cKO mouse in vivo in the septal area at 4 and 11 months of age. Although controls present with reproducible fiber structures over a period of 7 months, mutants reveal a partial reorganization which refers to a strengthening of dorsoventrally oriented fibers in conjunction with a rostral displacement of the ventral part.

Discussion

In vivo MRI of adult mice lacking the transcription factor Pax6 in the cortex demonstrated morphological changes of both gray and white matter which parallel findings in aniridia patients, that are adult humans with heterozygous PAX6 mutations.

Pax6 is Required for Normal Cortex Development and Interhemispheric Connectivity

The MRI observation that cortical areas of adult Pax6cKO mutants are severely diminished is in agreement with similar findings in the frontal cortex of Sey/Sey (Bishop et al. 2002), SeyNeu−/− and SeyNeu+/− embryos (Schmahl et al. 1993) as well as in the cortex of juvenile Pax6cKO mutants at P10 (Pinon et al. 2008). The regionalized disturbances of the mutant cortex parallel the endogenous rostrocaudal expression gradient of Pax6 in cortical radial glia cells and support previous data that Pax6-deficient phenotype in developing eye and cortex depends on Pax6 dosage (Schmahl et al. 1993; Glaser et al. 1994). In the same line of evidence, MRI studies of aniridia patients revealed strongly reduced gray matter especially in the anterior cingulate cortex (Ellison-Wright et al. 2004). The reduced cortical thickness in Pax6cKO mutants, most prominently in rostral areas (Table 1), may be ascribed to the almost complete loss of cortical layers II to IV (Pinon et al. 2008), which are the main contributors of the corpus callosum. In fact, in Pax6cKO mice the interhemispheric connectivity via the corpus callosum and anterior commissure was found to be severely reduced. These findings together with decreased anisotropies and axial diffusivities as well as increased radial diffusivities suggest a reduced density of fibers even in lateral parts of the corpus callosum (Table 2). Similarly, histologic examinations of juvenile (P10) (Pinon et al. 2008), and adult (TC Tuoc, K Radyushkin, AB Tonchev, MC Piñon, R Ashery-Padan, Z Molnár, MS Davidoff, and A Stoykova, in preparation) homozygous Pax6cKO mice, as well as SeyNeu/− (Schmahl et al. 1993), and Sey/Sey embryos (Stoykova et al. 1996) revealed the presence of only few corpus callosum fiber structures. Importantly, also MRI studies of aniridia patients have shown a dysgenesis of the corpus callosum, and an absence or hypoplasia of the anterior commissure (Sisodiya et al. 2001; Ellison-Wright et al. 2004; Bamiou et al. 2007).

In Pax6 deficiency, the generation of the low-layer neurons is preserved (Tarabykin et al. 2001; Schuurmans et al. 2004), and some layer V lineages appear to be augmented, specifically in the rostral cortex of Pax6cKO mice (TC Tuoc, K Radyushkin, AB Tonchev, MC Piñon, R Ashery-Padan, Z Molnár, MS Davidoff, and A Stoykova, in preparation). Because layer V neurons also send interhemispheric projections, the detected residual fibers in the rostral and middle regions of the corpus callosum of Pax6cKO mice might be formed by such layer V neuronal subsets. However, apart from a lower fiber density, the remaining callosal fibers were mainly oriented in a rostrocaudal direction, probably due to a lack of their respective cortical target regions.

Moreover, the rostrocaudal direction of projections within the cortical layers of Pax6cKO mice at the age of 4 months differed from the mainly radial fiber direction of controls, the latter reflecting the normal direction of the dendrite extensions of layer V and VI neurons towards their targets in layers III–I. Immunohistochemistry for NF- SMI-32 antibody revealed that the apical dendrites of the pyramidal neurons in mutants were much more branched instead of extending radially (Pinon et al. 2008). Intriguingly, fiber tractography (Fig. 8) in the Pax6cKO mutants at the age of 11 months revealed a tendency of fiber reorganization amplifying the dorsoventral orientation. Further immunhistochemical analyses are necessary to elucidate the question whether this MRI finding may be taken as an indicator of plasticity in the adult Pax6cKO cortex.

Lack of Cortical Pax6 Alters the Septal Area and Septo-Hippocampal Connection

Structural MRI as well as DTI-based fiber tractography revealed an alteration of the septal area and septo-hippocampal connection in Pax6cKO mice. In controls, neuronal fibers crossing the area of the medial and lateral septal nucleus remain in the midline which separates the lateral and medial septal complex and parts of the fornix. In Pax6cKO mice, however, a diffuse fiber sprouting was observed primarily into the direction of the hippocampus. Furthermore, manganese-enhanced MRI indicated changes of the septum. Noteworthy, a tendency for fiber reorganization could be observed in mutants at the age of 11 months.

In developing and adult brain Pax6 shows moderate or strong expression in lateral and medial septal nuclei, respectively (Stoykova and Gruss 1994). Results from Cre-LoxP based cell lineage mapping indicate that the Emx1Cre line drives recombination in the septum as well as in the septohippocampal, septofimbrial, and lateral septal nuclei, whereas no recombination was found in the medial septal nucleus (Gorski et al. 2002). Thus, cholinergic neurons in the medial septal nuclei that project to the hippocampal formation via the fimbria–fornix tract seem to be unaffected in Pax6cKO mutants. The results suggest that the altered septohippocampal connections in Pax6cKO mutants are a consequence of eliminated function of Pax6 in the lateral septal nucleus, leading to a compensatory sprouting of the remaining fibers to the hippocampus. The expression of Pax6 in specific structures of the adult brain was suggested to play a role in maintenance and/or functioning of specific circuits as the limbic system (Stoykova and Gruss 1994). In controls, a pronounced manganese-induced MRI signal enhancement most likely reflects an extensive Ca2+ uptake in the septal area, whereas in Pax6cKO mice the corresponding signal enhancement appeared smaller and rostrally displaced suggesting a functional deficit.

Aniridia patients exert mild to severe learning disabilities (Heyman et al. 1999; Malandrini et al. 2001) and a local excess of the gray matter volume (Ellison-Wright et al. 2004). Although the absolute volume of the hippocampus did not show a significant difference between controls and Pax6cKO mice, the relative size of the hippocampus was augmented and rostrally extended in mutants. Moreover, the mutant brain contained enlarged hippocampal fimbriae diffusely projecting into the septum, which might contribute to the enlargement of the striatum as described for aniridia patients (Ellison-Wright et al. 2004). Manganese-enhanced MRI in the hippocampal CA3 region was comparable in Pax6cKO and controls. The selective enhancement of CA3 but not CA1 is in line with previous studies (Watanabe et al. 2002) which indicated the CA3 region as a highly excitable area with numerous low-voltage–activated Ca2+ channels. Thus, despite the expanded relative size of the hippocampus, the manganese enhancement suggests normal activity of the hippocampal pyramidal cells in Pax6cKO mice. Similarly, although both hippocampus and occipital cortex in aniridia patients showed a relative gray matter excess, these structures were normally activated by verbal executive tasks (Ellison-Wright et al. 2004).

Pax6cKO Exhibit a Smaller Olfactory Bulb and Alterations of the Piriform Cortex

The volume of the olfactory bulb was reduced in Pax6cKO mice compared with controls. However, the layer structure detectable by manganese-enhanced MRI appeared similar to that found in normal mice (Lee et al. 2005).

Pax6 is known to play an important role in the development of the olfactory system (Dellovade et al. 1998). However, because the olfactory phenotype of Pax6cKO mice depends mostly on how effectively EmxCre lines recombine, it is not completely clear what portion of olfactory progenitor cells was affected in our mutants (Gorski et al. 2002).

In contrast to the normal-appearing layer structure of the olfactory bulb, the diffuse manganese enhancement in the piriform cortex of Pax6cKO mice indicates an alteration of the 3-layered paleocortex. In controls only the cell-rich pyramidal layer II and the lateral olfactory tract were enhanced and clearly separated from the moderate density cell layer III as well as from layer I, which consists mainly of fiber systems and dendrites. The diminishing piriform cortex in Pax6cKO mice is in accordance with the recombination of EmxCre (Gorski et al. 2002) and the reduced Pax6 level in this domain confirms the important role of Pax6 in cortex development.

Conclusion

Complementary to histological analyses at the cellular level, MRI allows for a more general view of the consequences of gene mutation in the entire brain including morphologic and microstructural as well as functional aspects in vivo. Based on this methodology, the present study unraveled multiple disturbances of the cortical organization and neuronal fiber connectivity in Pax6cKO mice. Most of the abnormalities are comparable with those of human Pax6 heterocygotes. Thus, MRI analyses of adult, selectively conditioned Pax6 mouse mutants may contribute to bridge the gap between basic research and clinical application.

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/

Funding

DFG Research Center for Molecular Physiology of the Brain in Göttingen, Germany.

Conflict of Interest: None declared.

References

Arlotta
P
Molyneaux
BJ
Chen
J
Inoue
J
Kominami
R
Macklis
JD
Neuronal subtype-specific genes that control corticospinal motor neuron development in vivo
Neuron.
 , 
2005
, vol. 
45
 (pg. 
207
-
221
)
Ashery-Padan
R
Somatic gene targeting in the developing and adult mouse retina
Methods.
 , 
2002
, vol. 
28
 (pg. 
457
-
464
)
Bamiou
DE
Free
SL
Sisodiya
SM
Chong
WK
Musiek
F
Williamson
KA
van Heyningen
V
Moore
AT
Gadian
D
Luxon
LM
Auditory interhemispheric transfer deficits, hearing difficulties, and brain magnetic resonance imaging abnormalities in children with congenital aniridia due to PAX6 mutations
Arch Pediatr Adolesc Med.
 , 
2007
, vol. 
161
 (pg. 
463
-
469
)
Bishop
KM
Rubenstein
JL
O'Leary
DD
Distinct actions of Emx1, Emx2, and Pax6 in regulating the specification of areas in the developing neocortex
J Neurosci.
 , 
2002
, vol. 
22
 (pg. 
7627
-
7638
)
Boretius
S
Würfel
J
Zipp
F
Frahm
J
Michaelis
T
High-field diffusion tensor imaging of mouse brain in vivo using single-shot STEAM MRI
J Neurosci Methods.
 , 
2007
, vol. 
161
 (pg. 
112
-
117
)
Britanova
O
Akopov
S
Lukyanov
S
Gruss
P
Tarabykin
V
Novel transcription factor Satb2 interacts with matrix attachment region DNA elements in a tissue-specific manner and demonstrates cell-type-dependent expression in the developing mouse CNS
Eur J Neurosci.
 , 
2005
, vol. 
21
 (pg. 
658
-
668
)
Britanova
O
de Juan Romero
C
Cheung
A
Kwan
KY
Schwark
M
Gyorgy
A
Vogel
T
Akopov
S
Mitkovski
M
Agoston
D
, et al.  . 
Satb2 is a postmitotic determinant for upper-layer neuron specification in the neocortex
Neuron.
 , 
2008
, vol. 
57
 (pg. 
378
-
392
)
Chen
JG
Rasin
MR
Kwan
KY
Sestan
N
Zfp312 is required for subcortical axonal projections and dendritic morphology of deep-layer pyramidal neurons of the cerebral cortex
Proc Natl Acad Sci USA.
 , 
2005
, vol. 
102
 (pg. 
17792
-
17797
)
Chen
B
Schaevitz
LR
McConnell
SK
Fezl regulates the differentiation and axon targeting of layer 5 subcortical projection neurons in cerebral cortex
Proc Natl Acad Sci USA.
 , 
2005
, vol. 
102
 (pg. 
17184
-
17189
)
Dellovade
TL
Pfaff
DW
Schwanzel-Fukuda
M
Olfactory bulb development is altered in small-eye (Sey) mice
J Comp Neurol.
 , 
1998
, vol. 
402
 (pg. 
402
-
418
)
Ellison-Wright
Z
Heyman
I
Frampton
I
Rubia
K
Chitnis
X
Ellison-Wright
I
Williams
SC
Suckling
J
Simmons
A
Bullmore
E
Heterozygous PAX6 mutation, adult brain structure and fronto-striato-thalamic function in a human family
Eur J Neurosci.
 , 
2004
, vol. 
19
 (pg. 
1505
-
1512
)
Glaser
T
Jepeal
L
Edwards
JG
Young
SR
Favor
J
Mass
RL
PAX6 gene dosage effect in a family with congenital cataracts, aniridia, anophthalmia and central nervous system defects
Nat Genet.
 , 
1994
, vol. 
7
 (pg. 
463
-
471
)
Gorski
JA
Talley
T
Qiu
M
Puelles
L
Rubenstein
JL
Jones
KR
Cortical excitatory neurons and glia, but not GABAergic neurons, are produced in the Emx1-expressing lineage
J Neurosci.
 , 
2002
, vol. 
22
 (pg. 
6309
-
6314
)
Gotz
M
Huttner
WB
The cell biology of neurogenesis
Nat Rev Mol Cell Biol.
 , 
2005
, vol. 
6
 (pg. 
777
-
788
)
Guillemot
F
Cellular and molecular control of neurogenesis in the mammalian telencephalon
Curr Opin Cell Biol.
 , 
2005
, vol. 
17
 (pg. 
639
-
647
)
Guillemot
F
Molnar
Z
Tarabykin
V
Stoykova
A
Molecular mechanisms of cortical differentiation
Eur J Neurosci.
 , 
2006
, vol. 
23
 (pg. 
857
-
868
)
Heyman
I
Frampton
I
van Heyningen
V
Hanson
I
Teague
P
Taylor
A
Simonoff
E
Psychiatric disorder and cognitive function in a family with an inherited novel mutation of the developmental control gene PAX6
Psychiatr Genet.
 , 
1999
, vol. 
9
 (pg. 
85
-
90
)
Küntzel
M
Parallele Datenakquisition zur Beschleunigung diffusions-gewichteter Kernspintomographie mit stimulierten Echos [Parallel data acquisition for diffusion-weighted MRI using stimulated echoes] [dissertation] Göttingen (Germany): Georg-August-Universität. Available at
2007
 
Lee
JH
Silva
AC
Merkle
H
Koretsky
AP
Manganese-enhanced magnetic resonance imaging of mouse brain after systemic administration of MnCl2: dose-dependent and temporal evolution of T1 contrast
Magn Reson Med.
 , 
2005
, vol. 
53
 (pg. 
640
-
648
)
Malandrini
A
Mari
F
Palmeri
S
Gambelli
S
Berti
G
Bruttini
M
Bardelli
AM
Williamson
K
van Heyningen
V
Renieri
A
PAX6 mutation in a family with aniridia, congenital ptosis, and mental retardation
Clin Genet.
 , 
2001
, vol. 
60
 (pg. 
151
-
154
)
Mallamaci
A
Stoykova
A
Gene networks controlling early cerebral cortex arealization
Eur J Neurosci.
 , 
2006
, vol. 
23
 (pg. 
847
-
856
)
McConnell
SK
Migration and differentiation of cerebral cortical neurons after transplantation into the brains of ferrets
Science.
 , 
1985
, vol. 
229
 (pg. 
1268
-
1271
)
McConnell
SK
Fates of visual cortical neurons in the ferret after isochronic and heterochronic transplantation
J Neurosci.
 , 
1988
, vol. 
8
 (pg. 
945
-
974
)
Molyneaux
BJ
Arlotta
P
Hirata
T
Hibi
M
Macklis
JD
Fezl is required for the birth and specification of corticospinal motor neurons
Neuron.
 , 
2005
, vol. 
47
 (pg. 
817
-
831
)
Molyneaux
BJ
Arlotta
P
Menezes
JR
Macklis
JD
Neuronal subtype specification in the cerebral cortex
Nat Rev Neurosci.
 , 
2007
, vol. 
8
 (pg. 
427
-
437
)
Mori
S
Crain
BJ
Chacko
VP
van Zijl
PC
Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging
Ann Neurol.
 , 
1999
, vol. 
45
 (pg. 
265
-
269
)
Muzio
L
DiBenedetto
B
Stoykova
A
Boncinelli
E
Gruss
P
Mallamaci
A
Conversion of cerebral cortex into basal ganglia in Emx2(-/-) Pax6(Sey/Sey) double-mutant mice
Nat Neurosci.
 , 
2002
, vol. 
5
 (pg. 
737
-
745
)
Natt
O
Frahm
J
In vivo magnetic resonance imaging: insights into structure and function of the central nervous system
Meas Sci Technol.
 , 
2005
, vol. 
16
 (pg. 
R17
-
R36
)
O'Leary
DD
Sahara
S
Genetic regulation of arealization of the neocortex
Curr Opin Neurobiol.
 , 
2008
, vol. 
12
 (pg. 
14
-
25
)
Pinon
MC
Tuoc
TC
Ashery-Padan
R
Molnar
Z
Stoykova
A
Altered molecular regionalization and normal thalamocortical connections in cortex-specific Pax6 knock-out mice
J Neurosci.
 , 
2008
, vol. 
28
 (pg. 
8724
-
8734
)
Pontious
A
Kowalczyk
T
Edlund
C
Hevner
RH
Role of intermediate progenitor cells in cerebral cortex development
Dev Neurosci.
 , 
2008
, vol. 
30
 (pg. 
24
-
32
)
Quinn
JC
Molinek
M
Martynoga
BS
Zaki
PA
Faedo
A
Bulfone
A
Hevner
RF
West
JD
Price
DJ
Pax6 controls cerebral cortical cell number by regulating exit from the cell cycle and specifies cortical cell identity by a cell autonomous mechanism
Dev Biol.
 , 
2007
, vol. 
302
 (pg. 
50
-
65
)
Schmahl
W
Knoedlseder
M
Favor
J
Davidson
D
Defects of neuronal migration and the pathogenesis of cortical malformations are associated with Small eye (Sey) in the mouse, a point mutation at the Pax-6-locus
Acta Neuropathol.
 , 
1993
, vol. 
86
 (pg. 
126
-
135
)
Schuurmans
C
Armant
O
Nieto
M
Stenman
JM
Britz
O
Klenin
N
Brown
C
Langevin
LM
Seibt
J
Tang
H
, et al.  . 
Sequential phases of cortical specification involve neurogenin-dependent and -independent pathways
EMBO J.
 , 
2004
, vol. 
23
 (pg. 
2892
-
2902
)
Sisodiya
SM
Free
SL
Williamson
KA
Mitchell
TN
Willis
C
Stevens
JM
Kendall
BE
Shorvon
SD
Hanson
IM
Moore
AT
, et al.  . 
PAX6 haploinsufficiency causes cerebral malformation and olfactory dysfunction in humans
Nat Genet.
 , 
2001
, vol. 
28
 (pg. 
214
-
216
)
Stoykova
A
Fritsch
R
Walther
C
Gruss
P
Forebrain patterning defects in small eye mutant mice
Development
 , 
1996
, vol. 
122
 (pg. 
3453
-
3465
)
Stoykova
A
Gotz
M
Gruss
P
Price
J
Pax6-dependent regulation of adhesive patterning, R-cadherin expression and boundary formation in developing forebrain
Development.
 , 
1997
, vol. 
124
 (pg. 
3765
-
3777
)
Stoykova
A
Gruss
P
Roles of Pax-genes in developing and adult brain as suggested by expression patterns
J Neurosci.
 , 
1994
, vol. 
14
 (pg. 
1395
-
1412
)
Tarabykin
V
Stoykova
A
Usman
N
Gruss
P
Cortical upper layer neurons derive from the subventricular zone as indicated by Svet1 gene expression
Development.
 , 
2001
, vol. 
128
 (pg. 
1983
-
1993
)
Walther
C
Gruss
P
Pax6, a murine paired box gene, is expressed in the developing CNS
Development.
 , 
1991
, vol. 
113
 (pg. 
1435
-
1449
)
Watanabe
T
Frahm
J
Michaelis
T
Functional mapping of neural pathways in rodent brain in vivo using manganese-enhanced three-dimensional magnetic resonance imaging
NMR Biomed.
 , 
2004
, vol. 
17
 (pg. 
554
-
568
)
Watanabe
T
Natt
O
Boretius
S
Frahm
J
Michaelis
T
In vivo 3D MRI staining of mouse brain after subcutaneous application of MnCl2
Magn Reson Med.
 , 
2002
, vol. 
48
 (pg. 
852
-
859
)
Yokel
RA
Crossgrove
JS
Manganese toxicokinetics at the blood-brain barrier
Res Rep Health Eff Inst.
 , 
2004
(pg. 
7
-
58
discussion 59–73