Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jul 1.
Published in final edited form as: Neuroimage. 2015 Apr 11;114:303–310. doi: 10.1016/j.neuroimage.2015.04.010

3D mapping of neuronal migration in the embryonic mouse brain with magnetic resonance microimaging

Abby E Deans 1,†, Youssef Zaim Wadghiri 2, Orlando Aristizábal 1, Daniel H Turnbull 1,2,3,*
PMCID: PMC4446241  NIHMSID: NIHMS680371  PMID: 25869862

Abstract

A prominent feature of the developing mammalian brain is the widespread migration of neural progenitor (NP) cells during embryogenesis. A striking example is provided by NP cells born in the ventral forebrain of mid-gestation stage mice, which subsequently migrate long distances to their final positions in the cortex and olfactory bulb. Previous studies have used two-dimensional histological methods, making it difficult to analyze three-dimensional (3D) migration patterns. Unlike histology, magnetic resonance microimaging (micro-MRI) is a non-destructive, quantitative and inherently 3D imaging method for analyzing mouse embryos. To allow mapping of migrating NP cells with micro-MRI, cells were labeled in situ in the medial (MGE) and lateral (LGE) ganglionic eminences, using targeted in utero ultrasound-guided injection of micron-sized particles of iron-oxide (MPIO). Ex vivo micro-MRI and histology were then performed 5-6 days after injection, demonstrating that the MPIO had magnetically labeled the migrating NP populations, which enabled 3D visualization and automated segmentation of the labeled cells. This approach was used to analyze the distinct patterns of migration from the MGE and LGE, and to construct rostral-caudal migration maps from each progenitor region. Furthermore, abnormal migratory phenotypes were observed in Nkx2.1−/− embryos, most notably a significant increase in cortical neurons derived from the Nkx2.1−/− LGE. Taken together, these results demonstrate that MPIO labeling and micro-MRI provide an efficient and powerful approach for analyzing 3D cell migration patterns in the normal and mutant mouse embryonic brain.

Keywords: lateral ganglionic eminence, LGE; medial ganglionic eminence, MGE; magnetic resonance imaging, MRI; micron-sized particles of iron-oxide, MPIO; ultrasound biomicroscopy, UBM

INTRODUCTION

Cell migration is a key feature of mammalian brain development, with NP cells arising in a number of progenitor regions and subsequently migrating via distinct pathways to take up their final positions in the mature brain. Studies of neural cell migration have largely been performed in the mouse, taking advantage of the vast resource of genetically modified strains to investigate the roles of defined molecular factors on cell migration and specification during brain development. In particular, a number of investigations have revealed that two distinct populations of NP cells are born ventrally in the MGE and LGE, and undergo extensive migration before differentiating into cortical and olfactory bulb interneurons, respectively (Marin et al., 2000, Anderson et al., 2001, Wichterle et al., 2001) (Fig. 1).

Fig. 1. MGE and LGE NP migration patterns.

Fig. 1

Open in a new tab

MGE NP cells (purple) undergo tangential migration (arrows) to the dorsal cortex (Cx), starting around embryonic day E12.5. By E18.5, the globus pallidus (GP) is the remnant of the MGE in the ventral forebrain. Over similar embryonic stages, LGE NP cells (green) move locally within the developing striatum (St) and migrate along the rostral migratory stream (RMS) into the olfactory bulb (OB). Other labels: lateral ventricle, lv. Orientation directions: dorsal, D; ventral, V; caudal, C; rostral, R.

Lacking in classical studies of NP cell movements in the embryonic mouse brain have been efficient methods for analyzing the complex 3D migration patterns. Numerous methods are available for optical labeling of cells in the developing mouse brain, including direct injection of fluorescent dyes or viruses, or genetic labeling via expression of reporter genes using cell-specific promoter-enhancer elements in transgenic or mutant mice. Despite these advances in cell labeling, 3D visualization of cell migration still relies largely on histology, which is inherently two-dimensional and prone to artifacts produced during sectioning and staining. Optical imaging methods such as two-photon microscopy and optical projection tomography provide 3D capability, but are currently limited in penetration to relatively early stages of embryonic brain development, or to superficial regions of later stage embryonic brains.

Micro-MRI provides a combination of relatively high spatial resolution (~50 µm), together with sufficient penetration for whole brain imaging at all developmental stages up to adulthood in mice (Turnbull and Mori, 2007, Nieman and Turnbull, 2010). For cell labeling applications in the adult mouse and rat brain, micron-sized particles of iron oxide (MPIO) have been the focus of a number of studies, being efficiently taken up by stem cells in culture (Hinds et al., 2003, Shapiro et al., 2005) and in situ by NP cells in the subventricular zone (Shapiro et al., 2006). By combining T2*-weighted MRI with MPIO labeling in situ in the adult SVZ, migration of NP cells in the rostral migratory stream have been investigated in a number of recent studies (Shapiro et al., 2006, Sumner et al., 2009, Nieman et al., 2010, Vreys et al., 2010, Granot et al., 2011).. Interestingly, MPIO labeling has also been used to study altered NP migration following injury in the early postnatal rat brain (Yang et al., 2009). However, this approach has yet to be applied to the much more extensive neurodevelopmental migration events occurring prior to birth.

The aim of this study was to magnetically label MGE and LGE NP populations for micro-MRI analysis of migration during embryonic brain development in wildtype (WT) mouse embryos and in Nkx2.1−/− embryos, which lack the physical structure and molecular markers of the MGE. Mid-gestational MGE or LGE cells were labeled in situ by targeted injection of MPIO under the guidance of ultrasound biomicroscopy (UBM), after which labeled NP cell migration was assessed on fixed brain samples using 3D T2*-weighted micro-MRI and histology. Our results demonstrate the utility of this approach for 3D analysis of NP cell migration patterns in the normal and mutant embryonic mouse brain.

MATERIALS & METHODS

Animals

All mice used in these studies were maintained under protocols approved by the Institutional Animal Care and Use Committee at New York University School of Medicine. Experiments were first performed in timed-pregnant ICR strain mice, where embryonic day (E)0.5 was defined to be noon on the day of vaginal plug detection after overnight mating. For mutant mouse analysis, Nkx2.1+/− heterozygous mutant breeding pairs were used to produce null mutants (Nkx2.1−/−), as well as heterozygous (Nkx2.1+/−) and wildtype (Nkx2.1+/+) littermate controls. PCR on adult or embryo tail DNA was used to identify genotype, using primers for Neo and Nkx2.1 as described previously (Kimura et al., 1996).

In situ NP cell labeling

The micron-sized particles of iron oxide (MPIO) contrast agent used was a (dragon) green fluorescent polystyrene-divinylbenzene coated particle with a mean diameter of 1.63 μm (Bangs Laboratories), and mean iron content of 1pg per particle (Shapiro et al., 2005). In utero injection of MPIO into the embryonic mouse brain were performed using UBM-guidance (Olsson et al., 1997, Wichterle et al., 2001), targeting the MGE at E11.5 to E13.5 or the LGE at E12.5 to E13.5. Pregnant mice were anesthetized with intraperitoneal (IP) pentobarbital, the abdomen shaved and sterilized, and midline laparotomy performed. Each uterine horn was externalized to assess the number and distribution of embryos, and then carefully replaced within the abdomen. The mouse was placed supine in a custom holder allowing for positioning of a modified petri dish over the abdomen, as described in detail previously (Liu et al., 1998). The 65-mm plastic petri dish was modified, removing a 30-mm diameter circular aperture from the center, and attaching a thin rubber membrane over the aperture. A 15 mm slit was cut in the rubber membrane to allow access to the uterus, through the laparotomy. The rubber membrane provided a seal between the plastic plate and the shaved abdomen, allowing the dish, which was filled with sterile phosphate buffered saline (PBS), to continuously bathe the uterus. The embryos (within the uterus) were carefully pulled through the slit in the membrane and positioned with gentle stabilization by the membrane itself. Embryos were visualized within the uterus using a commercial UBM system (Vevo 770, VisualSonics) with a 40MHz transducer. Once the MGE or LGE was identified morphologically, a 50μm diameter beveled glass needle loaded with particle solution was advanced, under real-time UBM guidance, through the uterus and into the selected brain region. Particle solution (10-25 nL; approximately 10,000 particles per nL) was delivered using a micro-injector system (Narishige), and the needle was withdrawn. After all selected embryos were injected, the laparotomy was closed with continuous silk sutures in the abdominal muscle and staples or interrupted silk sutures in the skin. The mouse was allowed to recover in a warm cage until fully awake.

Ex vivo Micro-MRI

To prepare E18.5 embryos for ex vivo micro-MRI, the mother was euthanized with an overdose of IP pentobarbital and cervical dislocation, and the embryos were removed from the uterus and cardioperfused with 4% paraformaldehyde (PFA). Embryos were then post-fixed for approximately 3 hours in 4% PFA at 4°C, washed in PBS, and mounted in a syringe phantom surrounded by Fomblin perfluoropolyether (Ausimont), which gives no proton NMR signal.

Ex vivo MRI data were acquired on a 7-Tesla micro-imaging system (SMIS), consisting of a horizontal bore magnet (Magnex Scientific) and an actively shielded 250mT/m gradient insert (Magnex). Imaging was done using either a custom saddle coil (ID= 22mm, length= 20mm) or a commercial quadrature Litzcage coil (ID= 25mm, length= 22mm; Doty Scientific). T2*-weighted 3D gradient-echo imaging was performed in 10h overnight sessions including both 100-μm (2h) and 50-μm (8h) isotropic resolution scans with the following sequence parameters: repetition time, TR=50-ms; echo time, TE=15-ms; flip angle, FA=18°; field of view, FOV=(25.6mm)3; Image matrix (100-μm)=2563; matrix (50-μm)=5123.

Image analysis

Image analysis and segmentation was done using Analyze (AnalyzeDirect) and Amira (Visage Imaging) software packages. Each brain was segmented from the 3D micro-MRI dataset using a semi-automated method of seeding with manual editing to remove non-neural tissues when necessary. The remnant of the injection site (IS) was identified as the largest dark region in the ventral brain. The center of the IS was identified using the “Magic Wand Selection” function to define the dark injection site, and decreasing the region in all dimensions 3 times using the “Shrink Selection” function of the Amira Segmentation Editor, where the “Magic Wand Selection” and “Shrink Selection” are proprietary functions of the Amira software. Dark regions representing MPIO-labeled migratory cells were segmented automatically in Amira using a threshold defined by Rose’s Criterion, where signal intensity 5 standard deviations (SD) or more below the mean background signal (Haacke et al., 1999). For these analyses, the mean signal intensity and standard deviation were measured in the (unlabeled) brain region contralateral to the MPIO injection. On micro-MRI, particles were often observed to have leaked into and spread within the cerebral ventricles. To improve meaningful appreciation of MPIO-labeled cell migration in the segmented volumes, regions of signal loss within the ventricular system were manually subtracted from the segmented volume.

Histology and Immunohistochemistry

Following micro-MRI, fixed samples were removed from the phantom, equilibrated sequentially with 15% and 30% sucrose solution at 4°C for cryoprotection, embedded in “Optimal Cutting Temperature”, OCT compound (Tissue-Tek®, Sakura Finetek USA) and frozen for cryo-sectioning (12-20μm sections). Sections used for correlation with MRI were washed and stained with 4',6-diamidino-2-phenylindole (DAPI) nuclear counterstain. For correlation, fluorescence images from three sequential 20-μm thick coronal sections were acquired and overlayed on Photoshop (Adobe) for comparison with a matched coronal image from the 50-μm resolution MRI dataset.

Immunohistochemistry (IHC) for detection of Tuj1 (pan-neuronal marker), Iba1 (microglial marker) and PDGFRα (oligodendrocyte marker) used the following general protocol: slides were washed in phosphate-buffered saline (PBS) followed by blocking in PBS with 0.1% triton X-100 (PBS-T) and 10% normal goat serum (NGS). Primary antibody solution (PBS-T with 1% NGS and antibody [anti-tuj-1 (Covance) at 1:1000 dilution; anti-Iba1 (Wako) 1:1000; anti-PDGFRα (BD Pharminigen) 1:500] was pipetted onto slides, which were kept in a wet box at 4°C overnight. Slides were washed in PBS followed by application of the secondary antibody solution: PBS-T with 1% NGS and antibody [Fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (Jackson) 1:200; Cyanine3 (Cy3)-conjugated goat anti-rabbit IgG (Jackson) 1:200]. After 1 hour at room temperature in a dark wet box, slides were washed (PBS × 3) and coverslipped with water-soluble mounting medium (Biomeda). Nuclear DAPI counterstain was used where noted in the first wash (1:10,000 in PBS). Fluorescence images were acquired on a Leica (Leitz DM RXE) microscope with a CCD camera (Pixelfly).

Quantitative histological analysis was done by identifying a total of 50-100 MPIO-labeled cells on single or double-labeled sections from three separate embryos and counting whether the MPIO-labeled cells were positive or negative for the marker used. Percentages were calculated from each embryo; error bars represent the standard deviation of values from the three embryos in each set: WT MGE, WT LGE, and Nkx2.1 LGE.

RESULTS

In situ injections of MPIO effectively targeted MGE or LGE progenitor zones

UBM image-guidance enabled real-time visualization and placement of the glass micro-capillary needle tip in either the MGE or LGE for accurate injection of MPIO (Fig. 2a,b). Although the needle produced a large reflection artifact on UBM, the tip of the needle was reliably identifiable during injection. Correct targeting of the MPIO was verified using micro-MRI of embryos, fixed and imaged immediately after injection, which showed prominent dark artifacts from the MPIO injected into the MGE or LGE (N=3 each; Fig. 2c,d). These results were further validated using fluorescence microscopy of matched histological sections, showing the localization of (green) MPIO particles in the injected regions (N=3; Fig. 2e,f). Comparison between histology and micro-MRI showed the expected blooming effect of the MPIO, with the region of signal loss much larger than the actual size of the particle deposit.

Fig. 2. MPIO labeling of NP cells in the MGE and LGE.

Fig. 2

Open in a new tab

(a,b) In utero UBM images were used to monitor a 50-μm diameter glass needle as it was advanced into the MGE (a) or LGE (b) for injection of MPIOs. The outline of the MGE and LGE were obvious from their protrusions into the (dark) lateral ventricle (lv). The tip of the needle is marked *; note the reflection artifact (A) caused by the glass injection needle. Dashed lines were added to enhance the border between MGE and LGE. (c,d) After injection, E12.5 embryos were harvested and prepared for ex vivo micro-MRI. In each case the injection site (*) was obvious from the T2*-blooming effect. The protrusions of MGE and LGE into the lv, as well as the outline of the lv were enhanced by dashed lines. (e,f) There was good qualitative correlation between the location of the MPIO-induced darkening on ex vivo micro-MRI and the fluorescent particles (*, green) in histological sections from the same brains. The counter stain (blue) in panels was non-specific nuclear stain, DAPI. Orientation directions: dorsal, D; ventral, V. Scale bars = 1-mm (panel b for a-d; panel f for e-f).

Quantification of micro-MRI signal changes due to MPIO particles

After injection of MPIO particles into the E12.5 MGE, embryos were analyzed at E18.5 with both micro-MRI and histology to determine the effect of MPIO on MRI signal intensity (Fig. 3). Particles were often observed within the ventricular system at MRI and histology; however, labeled cells were only found within the brain parenchyma on the side ipsilateral to the injection. Control injections into the ventricle at time points between E10.5 and E14.5, failed to result in cell labeling within the brain parenchyma in our experiments. Comparison of micro-MRI and matched histological sections after MGE injections showed good qualitative agreement between locations of the largest particles on fluorescence microscopy (Fig. 3a) with the largest dark spots on micro-MRI (Fig. 3b). Rare cases where dark spots on micro-MRI did not correlate with particles on histology were likely due to particle loss during tissue sectioning and staining. To further analyze the relationship between MPIO and micro-MRI signal change, the Rose criterion (signal change ≥ 5SD of background signal) was tested as a quantitative method for determining the threshold for segmenting MPIO particles on micro-MRI (Haacke et al., 1999), measuring line profiles through dark spots confirmed to be particles on histology (N=10) and an equal number of non-particle regions (N=10) (Fig. 3c). All confirmed particles were identified with thresholding based on the Rose criterion, while none of the non-particle regions were identified. Based on these results, the threshold for segmenting MPIOs on micro-MRI was set at 5SD below the (contralateral) background signal for all subsequent analyses.

Fig. 3. Correlation between micro-MRI and histology of MPIO particles.

Fig. 3

Open in a new tab

(a) Fluorescence-microscopy images of histological sections (overlay of three 20-μm sections) were compared to (b) matched (50-μm) micro-MRI images from the same region (dashed rectangle on b shows approximate location of a). There was good qualitative correlation between the largest green particles (a: locations 1-10 shown at higher magnification below) and the largest regions of signal loss on micro-MRI (b: 1-10). Two regions of signal loss on micro-MRI without clear correlation to microscopy (b: *) were likely the result of tissue distortion or particle loss during staining. (c) Line profiles drawn through confirmed particles 1-10 (black tracings) and 10 other non-particle (control) regions (white tracings) showed that signal intensity (SI) loss greater than 5 standard deviations (-5SD; Rose’s Criterion) from the mean background (Mean) was specific for MPIO. The background SI (mean and standard deviation) was measured on the contralateral side of the brain (solid box in b). Label: lateral ventricle, lv. Orientation directions: dorsal, D; ventral, V. Scale bar = 1-mm (b).

Micro-MRI demonstrated 3D cell migration patterns after MGE and LGE labeling

To assess the ability of micro-MRI to visualize patterns of NP cell migration, MPIO were injected into E11.5 to E13.5 MGE and E12.5 to E13.5 LGE using UBM image guidance. The embryos were then allowed to develop normally in utero for 5-7 days, and micro-MRI was performed at E18.5, one day before birth. Following MGE injection, micro-MRI showed wide dispersion of particles away from the injection site into the ipsilateral cortex (N=16; Fig. 4a,b). In contrast, after LGE injection micro-MRI showed local dispersion of particles throughout the striatum, as well as anterior movement in the (ipsilateral) rostral migratory stream (RMS) to the olfactory bulb (OB) (N=12; Fig. 4c,d). Although MPIO particles were frequently observed bilaterally in the cerebral ventricles, most likely due to leakage during the injection, it is notable that MPIO-labeling in the brain tissue was invariably only observed ipsilateral to the injection site, consistent with the known NP migration pathways that do not cross the midline. Segmentation of the labeled voxels in the MGE and LGE datasets clearly revealed the expected 3D patterns of NP cell migration from the MGE (Fig. 4e) and LGE (Fig. 4f) in individual mouse embryos. Although higher resolution (50-µm isotropic) imaging was used for segmenting the MPIO particles in this study, it was interesting that lower resolution (100-µm isotropic) was effectively able to demonstrate the same overall patterns of particle distributions in a time scale (2-h acquisition time) compatible with in vivo imaging (Suppl. Fig. 1).

Fig. 4. T2*-weighted micro-MRI reveals 3D migration patterns of MPIO-labeled cells.

Fig. 4

Open in a new tab

Ex vivo micro-MRI of fixed E18.5 embryos showed dispersion of MPIO-labeled cells from the injection sites (*). Following MGE injection, labeled cells migrated towards and throughout the dorsal cortex (Cx) (a, coronal; b, sagittal); following LGE injection, labeled cells migrated ventrally into the striatum (St) (c; coronal) and forward along the rostral migratory stream (RMS) to the olfactory bulb (OB) (d; sagittal). Segmentation of the dark voxels (shown in false-color, green) for 3D visualization allowed appreciation of tangential migratory pathways (arrowheads) from the MGE to the Cx (e) and along the RMS from the LGE to the OB (f). For 3D visualization of the migration patterns, see Suppl. Video 1 (MGE, e) and Suppl. Video 2 (LGE, f). Other labels: lateral ventricle, lv. Orientation directions: dorsal, D; ventral, V; caudal, C; rostral, R. Scale bar = 1-mm (panel b for a-f).

Immunohistochemistry shows that most migrating MPIO-labeled cells are neurons

IHC analysis of histological sections was performed after E18.5 micro-MRI to assess whether the particles were internalized in neural cells, and to determine the cell types labeled with MPIO (Fig. 5). Quantitative analyses after IHC of MGE-injected brains demonstrated expression of Tuj1 in the vast majority (> 85%) of MPIO labeled cells in the cortex, while LGE-injected brains demonstrated Tuj1 expression in most (>80%) MPIO labeled cells in the striatum and external layers of the olfactory bulb. These results show that most MPIO labeled migrating cells were Tuj1-expressing neurons. After both MGE and LGE injections, labeled cells near the IS were either Tuj1-expressing neurons (Tuj1+) or Iba1-expressing microglia. Although it has been shown that a subset of cortical oligodendrocytes also arise from the MGE during embryogenesis (Kessaris et al., 2006), MPIO was almost never found in PDGFRα-expressing oligodendrocytes in MGE-injected brains. In addition, MPIO-labeled cells in the RMS or central OB of LGE-injected brains were often not Tuj1-positive, possibly because they were insufficiently mature to express this neuronal marker. These results demonstrate that MPIO labeled the migrating neuronal populations known to arise from the MGE and LGE at the developmental stages investigated.

Fig. 5. Immunohistochemical (IHC) analysis of MPIO-labeled cells.

Fig. 5

Open in a new tab

Following MGE injection (a-c), the majority of MPIO-labeled cells in the cortex (Cx) were Tuj1-positive (a,g), indicating neuronal cell types. Near the injection site (IS), a number of MPIO particles were also found in Iba1-positive microglia (b,g). MPIO was almost never found in PDGFRα positive oligodendrocytes (c,g). Following LGE injection (d-f), the majority of MPIO-labeled cells away from the IS, in both the striatum (St; d) and olfactory bulb (OB; e) were neuronal (h), with microglia also being labeled near the IS (f,h). Within the subventricular zone (SVZ) and rostral migratory stream (RMS), including the central part of the OB, many labeled cells were neither Iba1-positive nor Tuj1-positive (h). These cells likely represent immature migrating neurons. Scale bars in panels (a) and (e) are 30 µm. Data in panels (g,h) were plotted as mean ± SD. In panels (a-f), MPIO particles are green, cell-specific markers (Tuj1, Iba1, PDGFRα) are red, and non-specific nuclear stain (DAPI) is blue.

Micro-MRI reveals rostral-caudal differences in migration within the MGE and LGE

MPIO labeling and 3D micro-MRI was used to investigate differences in NP cell migration patterns from different sub-regions within the MGE and LGE, an area about which little has been reported. 3D MPIO distributions were segmented from micro-MRI images of MGE- and LGE-injected embryos, and each IS determined retrospectively from the densest regions of MPIO labeling within the MGE or LGE (Fig. 6). Cell migration patterns from injections along the rostral-caudal extent of each progenitor zone were then registered to a common 3D space for comparison. Migration from within the MGE showed a similar wide distribution throughout the dorsal cortex from each IS (typical N=3 distributions, Fig. 6a,b; Suppl. Fig. 2a-c). Although subtle, the data suggested that there might be a slight increase in rostral migration from more caudal regions of MGE, a result that requires further investigation to verify. In contrast, migration from within the LGE demonstrated a clear rostral-caudal organization, with more rostral sub-regions giving rise to NP cells that migrate in the RMS to the OB, while more caudal sub-regions demonstrated mostly local migration within the striatum (typical N=3 distributions, Fig. 6c,d; Suppl. Fig. 2 d-f).

Fig. 6. Distinct rostral-caudal migration patterns of NP cells were detected by micro-MRI.

Fig. 6

Open in a new tab

MPIO particles were segmented automatically (shown in color) from 3D micro-MRI images, and co-registered to visualize the distributions (N=3 mice shown for each region) of migrating MPIO-labeled NP cells originating from distinct injection sites (IS) within the MGE (a,b) and LGE (c,d). MGE NP cells (a; ISs shown from caudal to rostral, colored yellow, blue, red), distributed widely throughout the cortex (b), with more caudal NP cells (yellow) having a tendency to migrate rostrally. LGE NP cells (c; ISs shown from caudal to rostral, colored yellow, red, blue) showed a clear band-like distribution in the striatum, directly ventral to each IS, and preferential migration to the olfactory bulb from more rostral ISs (d). See Suppl. Fig. 2 for individual segmentations of each mouse shown in panels (b) and (d). Orientation directions: dorsal, D; ventral, V; caudal, C, rostral, R. Scale bar = 1-mm (panel d for a-d).

Micro-MRI demonstrates altered migration patterns in Nkx2.1−/− mutant mice

Nkx2.1 is expressed in the MGE, and not the LGE, during embryonic brain development, and Nkx2.1−/− mutant embryos do not form a morphological MGE, being described as having a ventral to dorsal transformation of (mutant) MGE into an LGE fate (Sussel et al., 1999). UBM images clearly demonstrated the Nkx2.1−/− mutant phenotype of the Nkx2.1−/− embryo, showing a single ganglionic eminence, presumably the LGE (Fig. 7a,b). After in utero UBM-guided injection of MPIO into the E12.5 Nkx2.1−/− LGE, micro-MRI at E18.5 sometimes showed normal LGE migration patterns (N=5/16; Fig. 7c), but often demonstrated abnormal migration, suggesting a combination of the normal MGE and LGE patterns (N=11/16; Fig. 7d). In addition, embryos with more rostral MPIO injections resulted in more labeling in the RMS, OB and frontal cortex, while more caudal injections resulted in a more disperse (MGE-like) pattern of cortical labeling (typical distributions, N = 4; Fig. 7e,f; Suppl. Fig. 3).

Fig. 7. Micro-MRI detected altered migration patterns of LGE NP cells in Nkx2.1−/− mice.

Fig. 7

Open in a new tab

During in utero (E12.5) image-guided injections of MPIOs, UBM images of both Nkx2.1+/+ and Nkx2.1+/− brains (morphologically wildtype, WT) showed two distinct eminences (dashed lines), the MGE and LGE (a), whereas the Nkx2.1−/− (mutant) brains had a single large eminence, presumed to be the LGE (b). LGE injections into Nkx2.1−/− mice demonstrated the expected ventral migration within the striatum (c), and often a striking amount of tangential migration to the cortex (d, arrows). Similar to the results following WT injections, a rostral-caudal pattern of migration was observed in the Nkx2.1−/− mice, with more caudal injection sites (ISs) demonstrating predominant cortical migration and more rostral ISs resulting in OB and frontal cortex migration. Segmentation and co-registration of 2 typical caudal (green, blue) and 2 typical rostral (red, yellow) ISs illustrate the results (e, ISs shown in inset). See Suppl. Fig. 3 for individual segmentations of each mouse shown in panel (e). Orientation directions: dorsal, D; ventral, V. Scale bars = 1-mm (panel b for a-b; panel d for c-d).

IHC analysis in E18.5 Nkx2.1−/− embryos showed that MPIO-labeled cells in both cortex and striatum were predominantly Tuj1-expressing neurons, while labeled cells in the IS were both neurons and Iba1-expressing microglia (Fig. 8). Similar to WT embryos, MPIO-labeled cells in the RMS and central OB, expected to be the most immature NP cells, were often not Tuj1-positive. Taken together, these data are consistent with the proposed ventral origin of a population of cortical interneurons in Nkx2.1−/− mutants, approximately half the WT number, still present in the mutant cortex at late embryonic stages (Sussel et al., 1999).

Fig. 8. IHC analysis of MPIO-labeled cells originating in the Nkx2.1−/− LGE.

Fig. 8

Open in a new tab

Following injection into the Nkx2.1−/− LGE, the majority of MPIO-labeled cells observed in the cortex (Cx), striatum (St) and injection site (IS) were Tuj1-positive neurons (dark grey bars). In addition to neurons, Iba1-positive microglia (light grey bars) were also observed in each region, with a larger proportion of microglial uptake around the IS. Data were plotted as mean ± SD.

DISCUSSION

In this study, we demonstrated effective ultrasound-guided spatially and temporally targeted delivery of MPIO cellular contrast agent to two neuronal progenitor zones (MGE, LGE) in the mid-gestation embryonic mouse brain. At late embryogenesis, labeled progenitor migration was assessed on fixed brain samples using 3D T2*-weighted micro-MRI and histology. In WT embryos, robust migration from the MGE to the dorsal cortex, and from the LGE to the striatum and olfactory bulb were observed. The vast majority of MPIO-labeled cells in the cortex and olfactory bulb were confirmed by IHC to be neuronal, as expected, with only rare labeling of phagocytic microglia. We used a semi-automated, threshold-based segmentation method to visualize the 3D distributions of MPIO labeled cells. Comparison of multiple WT data sets demonstrated sub-populations of migratory cells along the rostral-caudal axis within both the MGE and LGE, which have not previously been described. Interestingly, migration from the LGE in Nkx2.1−/− mutant embryos showed an expanded pattern of migration including a marked migration to the dorsal cortex, not observed in WT LGE.

This is the first MRI study of NP cell migration patterns in the embryonic mouse brain, specifically from the ventral MGE and LGE progenitor zones. Several molecular markers of ventral progenitor zones have been described, including Nkx2.1 (Sussel et al., 1999, Marin et al., 2000, Xu et al., 2004, Cobos et al., 2006); however, it is likely that other factors have yet to be identified. Our approach may be helpful for future studies in identifying and characterizing the molecular origins of these subpopulations. Ongoing studies have expanded MRI to the in utero setting, which could allow future investigations exploring the temporal nature of these migratory processes longitudinally in living embryos (Deans et al., 2008, Berrios-Otero et al., 2012). Preliminary data at 100-µm isotropic resolution (Suppl. Fig. 1), which has recently been demonstrated to be feasible for in utero MRI (Parasoglou et al., 2013), together with previous reports of cell tracking in postnatal mice at the same resolution (Nieman et al., 2010, Granot et al., 2011), are encouraging that the overall migration pattern would still be appreciable at the lower resolution required for in vivo imaging. However, technical challenges remain for in vivo cell migration mapping in mouse embryos, including correction of artifacts due to physiologic motion of the mother and fetus during imaging, differentiating MPIO signal from the endogenous T2* effect of the fetal vasculature, potential toxicity related to multiple gestational exposures to anesthesia, and the challenge of reproducibly identifying a specific embryo within each litter.

Mapping the migrating NP cells in this study depended on the sensitivity of detecting MPIO particles with T2*-weighted gradient-echo MRI. This sensitivity is not without cost, since the T2* blooming effect also caused the distortion of the images around the injection site (e.g., Fig. 2). In our application, MPIO uptake and ex vivo T2*-weighted micro-MRI provided an excellent method for visualizing the relatively sparse patterns of migrating NP cells within the surrounding anatomy of the cortex, striatum and olfactory bulbs, which were sufficiently removed from the injection site, where anatomical details were obscured by the blooming effect. In future, a multi-echo gradient echo sequence (using a shorter echo to visualize the injection site and a longer echo to detect the isolated MPIO particles), or a combination of T2-weighted spin-echo (injection site) and T2*-weighted gradient-echo (MPIO particles), might enable more accurate assessment of anatomical detail and particle distribution near the injection site.

Our results illustrate the potential for phenotypic studies of mutant embryos with altered neuronal migration patterns such as the Nkx2.1−/− model. In the Nkx2.1 knockout mouse (Kimura et al., 1996), the MGE, thought to represent the primary source of cortical interneurons, has been described to be molecularly re-specified to LGE cell fate (Sussel et al., 1999). However, loss of the MGE in Nkx2.1−/− mice results in loss of only half of the cortical interneuron population. Further studies have demonstrated that these remaining molecularly distinct cortical interneurons likely originate in the caudal ganglionic eminence (CGE) (Nery et al., 2001, Butt et al., 2005, Miyoshi et al., 2010). The migration patterns demonstrated by micro-MRI suggest that these CGE progenitors and the molecular-LGE progenitors may be spatially co-localized within the monolithic Nkx2.1−/− ganglionic eminence, particularly caudally. This may imply that Nkx2.1 normally acts to generate a boundary to NP cells originating in the CGE, so that in the absence of Nkx2.1 the CGE progenitors can move more rostrally during embryonic brain development.

In conclusion, this study represents a significant methodological advance for mouse neurodevelopment research. Taken together, our results have broad implications for future investigation of NP cell migration in the embryonic mouse brain.

Supplementary Material

1
2
3
Download video file (12.5MB, avi)
4
Download video file (11.1MB, avi)
5

HIGHLIGHTS.

  • - Ultrasound-guided injection of micron-sized particles of iron-oxide (MPIO) in the embryonic mouse brain

  • - In situ magnetic labeling of embryonic neural progenitor cells

  • - Automated threshold detection and segmentation of MPIO-labeled cells

  • - 3D micro-MRI analysis of neural cell migration patterns

ACKNOWLEDGEMENTS

This work was supported by NIH grants R01NS038461 and R21NS066490 (to DHT). We thank Drs. Marc Fuccillo and Gordon Fishell for advice on MGE/LGE cell migration and assistance with the immunohistochemistry.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  1. Anderson SA, Marin O, Horn C, Jennings K, Rubenstein JL. Distinct cortical migrations from the medial and lateral ganglionic eminences. Development. 2001;128:353–363. doi: 10.1242/dev.128.3.353. [DOI] [PubMed] [Google Scholar]
  2. Berrios-Otero CA, Nieman BJ, Parasoglou P, Turnbull DH. In utero phenotyping of mouse embryonic vasculature with MRI. Magn Reson Med. 2012;67:251–257. doi: 10.1002/mrm.22991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Butt SJ, Fuccillo M, Nery S, Noctor S, Kriegstein A, Corbin JG, Fishell G. The temporal and spatial origins of cortical interneurons predict their physiological subtype. Neuron. 2005;48:591–604. doi: 10.1016/j.neuron.2005.09.034. [DOI] [PubMed] [Google Scholar]
  4. Cobos I, Long JE, Thwin MT, Rubenstein JL. Cellular patterns of transcription factor expression in developing cortical interneurons. Cereb Cortex. 2006;16(Suppl 1):i82–88. doi: 10.1093/cercor/bhk003. [DOI] [PubMed] [Google Scholar]
  5. Deans AE, Wadghiri YZ, Berrios-Otero CA, Turnbull DH. Mn enhancement and respiratory gating for in utero MRI of the embryonic mouse central nervous system. Magn Reson Med. 2008;59:1320–1328. doi: 10.1002/mrm.21609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Granot D, Scheinost D, Markakis EA, Papademetris X, Shapiro EM. Serial monitoring of endogenous neuroblast migration by cellular MRI. Neuroimage. 2011;57:817–824. doi: 10.1016/j.neuroimage.2011.04.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Haacke EMBR, Thompson MR, Venkatesan R. Magnetic Resonance Imaging: Physical Principles and Sequence Design. John Wiley and Sons; New York: 1999. [Google Scholar]
  8. Hinds KA, Hill JM, Shapiro EM, Laukkanen MO, Silva AC, Combs CA, Varney TR, Balaban RS, Koretsky AP, Dunbar CE. Highly efficient endosomal labeling of progenitor and stem cells with large magnetic particles allows magnetic resonance imaging of single cells. Blood. 2003;102:867–872. doi: 10.1182/blood-2002-12-3669. [DOI] [PubMed] [Google Scholar]
  9. Kessaris N, Fogarty M, Iannarelli P, Grist M, Wegner M, Richardson WD. Competing waves of oligodendrocytes in the forebrain and postnatal elimination of an embryonic lineage. Nature neuroscience. 2006;9:173–179. doi: 10.1038/nn1620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kimura S, Hara Y, Pineau T, Fernandez-Salguero P, Fox CH, Ward JM, Gonzalez FJ. The T/ebp null mouse: thyroid-specific enhancer-binding protein is essential for the organogenesis of the thyroid, lung, ventral forebrain, and pituitary. Genes Dev. 1996;10:60–69. doi: 10.1101/gad.10.1.60. [DOI] [PubMed] [Google Scholar]
  11. Liu A, Joyner AL, Turnbull DH. Alteration of limb and brain patterning in early mouse embryos by ultrasound-guided injection of Shh-expressing cells. Mech Dev. 1998;75:107–15. doi: 10.1016/s0925-4773(98)00090-2. [DOI] [PubMed] [Google Scholar]
  12. Marin O, Anderson SA, Rubenstein JL. Origin and molecular specification of striatal interneurons. J Neurosci. 2000;20:6063–6076. doi: 10.1523/JNEUROSCI.20-16-06063.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Miyoshi G, Hjerling-Leffler J, Karayannis T, Sousa VH, Butt SJ, Battiste J, Johnson JE, Machold RP, Fishell G. Genetic fate mapping reveals that the caudal ganglionic eminence produces a large and diverse population of superficial cortical interneurons. J Neurosci. 2010;30:1582–1594. doi: 10.1523/JNEUROSCI.4515-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Nery S, Wichterle H, Fishell G. Sonic hedgehog contributes to oligodendrocyte specification in the mammalian forebrain. Development. 2001;128:527–540. doi: 10.1242/dev.128.4.527. [DOI] [PubMed] [Google Scholar]
  15. Nieman BJ, Turnbull DH. Ultrasound and magnetic resonance microimaging of mouse development. Methods Enzymol. 2010;476:379–400. doi: 10.1016/S0076-6879(10)76021-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Nieman BJ, Shyu JY, Rodriguez JJ, Garcia AD, Joyner AL, Turnbull DH. In vivo MRI of neural cell migration dynamics in the mouse brain. Neuroimage. 2010;50:456–464. doi: 10.1016/j.neuroimage.2009.12.107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Olsson M, Campbell K, Turnbull DH. Specification of mouse telencephalic and mid-hindbrain progenitors following heterotopic ultrasound-guided embryonic transplantation. Neuron. 1997;19:761–772. doi: 10.1016/s0896-6273(00)80959-9. [DOI] [PubMed] [Google Scholar]
  18. Parasoglou P, Berrios-Otero CA, Nieman BJ, Turnbull DH. High-resolution MRI of early-stage mouse embryos. NMR Biomed. 2013;26:224–31. doi: 10.1002/nbm.2843. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Shapiro EM, Gonzalez-Perez O, Manuel Garcia-Verdugo J, Alvarez-Buylla A, Koretsky AP. Magnetic resonance imaging of the migration of neuronal precursors generated in the adult rodent brain. Neuroimage. 2006;32:1150–1157. doi: 10.1016/j.neuroimage.2006.04.219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Shapiro EM, Skrtic S, Koretsky AP. Sizing it up: cellular MRI using micron-sized iron oxide particles. Magn Reson Med. 2005;53:329–338. doi: 10.1002/mrm.20342. [DOI] [PubMed] [Google Scholar]
  21. Sumner JP, Shapiro EM, Maric D, Conroy R, Koretsky AP. In vivo labeling of adult neural progenitors for MRI with micron sized particles of iron oxide: quantification of labeled cell phenotype. Neuroimage. 2009;44:671–678. doi: 10.1016/j.neuroimage.2008.07.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sussel L, Marin O, Kimura S, Rubenstein JL. Loss of Nkx2.1 homeobox gene function results in a ventral to dorsal molecular respecification within the basal telencephalon: evidence for a transformation of the pallidum into the striatum. Development. 1999;126:3359–3370. doi: 10.1242/dev.126.15.3359. [DOI] [PubMed] [Google Scholar]
  23. Turnbull DH, Mori S. MRI in mouse developmental biology. NMR in biomedicine. 2007;20:265–274. doi: 10.1002/nbm.1146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Vreys R, Vande Velde G, Krylychkina O, Vellema M, Verhoye M, Timmermans JP, Baekelandt V, Van der Linden A. MRI visualization of endogenous neural progenitor cell migration along the RMS in the adult mouse brain: validation of various MPIO labeling strategies. Neuroimage. 2010;49:2094–2103. doi: 10.1016/j.neuroimage.2009.10.034. [DOI] [PubMed] [Google Scholar]
  25. Wichterle H, Turnbull DH, Nery S, Fishell G, Alvarez-Buylla A. In utero fate mapping reveals distinct migratory pathways and fates of neurons born in the mammalian basal forebrain. Development. 2001;128:3759–3771. doi: 10.1242/dev.128.19.3759. [DOI] [PubMed] [Google Scholar]
  26. Xu Q, Cobos I, De La Cruz E, Rubenstein JL, Anderson SA. Origins of cortical interneuron subtypes. J Neurosci. 2004;24:2612–2622. doi: 10.1523/JNEUROSCI.5667-03.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Yang J, Liu J, Niu G, Chan KC, Wang R, Liu Y, Wu EX. In vivo MRI of endogenous stem/progenitor cell migration from subventricular zone in normal and injured developing brains. Neuroimage. 2009;48:319–328. doi: 10.1016/j.neuroimage.2009.06.075. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS680371-supplement-1.tif (253.4KB, tif)
2
NIHMS680371-supplement-2.tif (233.5KB, tif)
3
Download video file (12.5MB, avi)
4
Download video file (11.1MB, avi)
5
NIHMS680371-supplement-5.pdf (2.6MB, pdf)

RESOURCES