Manganese enhanced MRI (MEMRI): neurophysiological applications

Abstract

Manganese ion (Mn(2+)) is a calcium (Ca(2+)) analog that can enter neurons and other excitable cells through voltage gated Ca(2+) channels. Mn(2+) is also a paramagnetic that shortens the spin-lattice relaxation time constant (T(1)) of tissues where it has accumulated, resulting in positive contrast enhancement. Mn(2+) was first investigated as a magnetic resonance imaging (MRI) contrast agent approximately 20 years ago to assess the toxicity of the metal in rats. In the late 1990s, Alan Koretsky and colleagues pioneered the use of manganese enhanced MRI (MEMRI) towards studying brain activity, tract tracing and enhancing anatomical detail. This review will describe the methodologies and applications of MEMRI in the following areas: monitoring brain activity in animal models, in vivo neuronal tract tracing and using MEMRI to assess in vivo axonal transport rates.

Figure 1

Comparison of AIM (A), diffusion weighted image (DWI) (B), T2-weighted (T2W) image (C), and apparent diffusion coefficient (ADC) map (D) on the same stroke model. The high density area which is localized within the lateral side of the caudate-putamen and lateral cortex on AIM was smaller than both DWI hyperintensity and decreased ADC. A slight decrease in signal intensity was observed on the T2W image. [Reproduced from Aoki et al. (2004a). Copyright © 2004 John Wiley & Sons, Inc.]

Figure 2

Activation of the rat brain by the intra-carotid arterial injection of hypertonic NaCl. The top and bottom panels show Fos immunoreactivity in the cortex, lateral habenular nucleus, paraventricular hypothalamic nucleus and supraoptic nucleus. The middle panel shows T1-weighted images. Areas in which the signal intensity increased significantly (p<0.05) are shown in color. The bottom panel shows spatial overlaps of activation maps from the same two rats shown in Figure 1. The first column shows the combined maps of BOLD (red), CBF (green), and the overlap between the two (yellow). The second column shows the combined maps of BOLD (red), T1-weighted AIM (green) and the overlap between the two (yellow). The third column shows the combined maps of CBF and T1-weighted AIM. [Reproduced from Aoki et al. (2004a). Copyright © 2004 John Wiley & Sons, Inc.]

Figure 3

Typical DAIM time courses of relative signal intensities in the cortex and ventricle, obtained from the glutamate administration (solid symbols) and control (open squares) groups. The inset indicates typical regions-of-interest. The first set of images (1–16) was obtained before mannitol administration, but after the start of MnCl₂ infusion. The second set (17–32) was obtained before injection of glutamate after breaking the BBB with mannitol. The signal increases significantly in a nonspecific manner following opening of the BBB. The third set (33–48) was obtained after glutamate injection. The signal intensity in the cortex of the glutamate administration group (solid circles) increases substantially after glutamate injection, but not in the control group following saline administration (open squares). The fourth set (49–64) was obtained during rest after stopping the MnCl₂ infusion. [Reproduced from Aoki et al. (2002). Copyright © 2002 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 4

Comparison between DAIM and AIM MRI. The left column shows (A) typical DAIM functional map and (B) AIM images from the glutamate administration group. The right column shows (C and D) corresponding maps for the forepaw electrical stimulation group. While the (A and C) DAIM maps only show the regions specifically activated by the stimulation paradigm, the (B) and (D) AIM images show contamination from nonspecific activation, such as ventricles (arrows). [Reproduced from Aoki et al. (2002). Copyright © 2002 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 5

Timetable of MR scanning and drug administration. A series of high resolution DWI, T2WI, T1WI were repeatedly acquired before and at 0.1, 1.2, 2.3. 24, 48, 72, 120, and 168 h after KA administration. The arrow indicates the time of KA i.v. injection. Arrowheads indicate the times of two Gd-DTPA i.v. injections (1 h 45 min and 2 h 45 min post KA). The dashed line indicates the time of Mn²⁺ i.v. infusion. [Reproduced from Hsu et al. (2007). Copyright © 2007 Oxford University Press.]

Figure 6

Representative high-resolution images and a corresponding histological section in normal control rats showing layer organization in the intact hippocampus and selected regions of interest (ROIs). (A) DWI showing distinct hippocampal layers in vivo (n=6). (B) T2WI did not allow for identification of hippocapal layers. (C) A magnified view of the hippocampus region in the white rectangle shown in (A) indicating six hippocampal layers, including the CA3 stratum oriens (subregion 1), CA3 pyramidal cell layer (subregion 2), CA 3 stratum radiatum (subregion 3), hippocampal fissure (subregion 4), CA1 stratum radiatum (subregion 5) and CA1 pyramidal cell layer (subregion 6). (D) Immunostaining of VGCC in brain section showing the layers in the hippocampus (n=5). The ROIs were shown as black circles. Arrows indicated hippocampal layers and the corresponding ROIs. The arrowhead indicates the control ROI in the secondary visual cortex. The scale bar represents 500 μm. [Reproduced from Hsu et al. (2007). Copyright © 2007 Oxford University Press.]

Figure 7

Statistical mapping in the mouse IC from MEMRI images. 3D MEMRI images of the IC were segmented and analyzed via voxel-by-voxel statistical comparisons. t-Test analysis was performed, comparing the signal intensities (SI) in each voxel, between 3D images from two groups of mice: V₁ array of SI values for Group 1=stimulated, V₂ array for Group 2=quiet controls. For each voxel (red cube), the resulting p-value was assigned after 8-bit gray scale conversion. In this way, a 3D p-map was created to encode the statistically significant different IC regions between the two groups of mice. [Reproduced from Yu et al. (2008). Copyright © by Elsevier, Inc.]

Figure 8

Gaussian filtering is a method for reducing the number of false positive errors. Multiple voxel-wise t-tests were made, comparing MEMRI images of mice maintained in a quiet environment and mice exposed to 40 kHz. The resulting statistical maps demonstrated difference between the two groups (quiet vs. 40 kHz=red) using a variable p-value threshold (p_th): p_th=0.05 (A–C), p_th=0.02 (D–F), and p_th=0.01 (G–I). Some scattered active voxels were observed in the original statistical maps, which are more likely to be the false positive voxels (A, D, G). A 3D Gaussian filter was employed to smooth the 3D p-map by convolving the data with a 3×3×3 Gaussian kernel (B, E, H). A constrained smoothing function (Amira) was also employed for display purposes only, generating sub-voxel weights to naturally smooth the surface of the activity contour. In the example shown, this smoothing function has been applied to the filtered p-map data (C, F, I) to show the effect of the smoothing operation. [Reproduced from Yu et al. (2008). Copyright © by Elsevier, Inc.]

Figure 9

MEMRI signal intensities were analyzed within the active IC regions defined by the 3D p-maps. Signal intensity (SI) differences (ΔSI) between sound stimulated and quiet control images were calculated in each voxel with p-value ≤ 0.05. Color-coded 2D maps show ΔSI in coronal sections of the IC (A, B, C; color bar below). (D) The mean SI was significantly higher in 40 kHz stimulated mice than in quiet controls (65 dB vs. quiet, p=0.00032, 77 dB vs. quiet, p=0.00090, 89 dB vs. quiet, p=0.00022). (E) Scatter plot of the peak ΔSI vs. peak sound stimulus amplitude, showing the regression line (center) and 95% confidence bands (outer curves). [Reproduced from Yu et al. (2008). Copyright © by Elsevier, Inc.]

Figure 10

(A) Three sagittal slices of a mouse treated with Mn²⁺ in the naris from a representative 3D T₁-weighted MRI sequence. Note the highlighting of the olfactory bulb as well as the primary olfactory cortex leading from the bulbs. (B) Four axial slices from the same mouse treated with Mn²⁺ in the naris from a 3D T₁-weighted MRI sequence. Note the highlighting of the outer layers of the olfactory bulbs where the olfactory glomeruli are located. In addition, the enhanced contrast continues caudally into the primary olfactory cortex. Due to the length of the scan, mice were sacrificed before 3D imaging. [Reproduced from Pautler et al. (1998). Copyright © 1998 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 11

Sequential, oblique horizontal slices from a 3D T1-weighted MRI set of a mouse 24 h after Mn²⁺ was injected intravitreally into one eye. Note the enhanced contrast of the eye on the right, as well as the optic nerve, as compared with the control eye. In the sequence of slices from panels (A) to (F), the enhanced contrast crosses over to the contralateral side of the brain to the area of the superior colliculus. Due to the length of the scan, mice were sacrificed before 3D imaging. [Reproduced from Pautler et al. (1998). Copyright © 1998 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 12

MEMRI images of mouse injections in the basolateral amygdala. All tracings due to Mn²⁺ accumulation within the brain are seen as hyperintense regions. From the site of injection in the amygdala (A), tracings to the entorhinal cortex (EC), subiculum (S), and the hippocampus (HC) are apparent. Starting ventrally (approx. 4.5 mm below the top of the brain) and moving dorsally in horizontal sections, in the top two panels, the aforementioned structures are clearly visible (left side of the figure). Homotopic connections to the hippocampus on the right side of the brain are also evident. In the lower left panel, projections to the fornix (F) from the hippocampus are visible. Fornix enhancements, the lateral septal nucleus (SN), and the habenulla (H) also exhibited positive contrast enhancement due to Mn²⁺ accumulation. In the lower right panel the final closure of the hippocampal formation (DHC) is visible. Note that all homotopic connections are of less signal intensity than connections on the side ipsilateral to injection (see Table 2). [Reproduced from Pautler et al. (2003). Copyright © 2003 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 13

Sagittal (A, C), horizontal (D) and coronal (E, F) in vivo MEMRI of the male starling brain obtained 6 h after MnCl₂ injection into HVc. The injection area is indicated by the arrow in panel (B) and by the signal-enhanced areas on the corresponding sagittal MRI slices in panels (A) and (C). In the images, the brighter signal enhanced areas demonstrate the labeling by Mn²⁺ as is the case for the RA and area X. A schematic illustration of the song-control system redrawn based on a figure in Nottebohm (1991) is provided in (B). Panel (C) (a duplicate of panel A) illustrates the different planes of imaging for panels (D), (E) and (F). Image resolution in the coronal plane is 97 μm (pixel size). Scale bar=10 mm. lMAN, Lateral magnocellular nucleus of the anterior neostriatum; DLM, dorsolateral nucleus of the medial thalamus; xNIIts, nucleus hypoglossus pars tracheosyringealis. [Reproduced from Van der Linden et al. (2002). Copyright © 2002 Elsevier Science Ltd.]

Figure 14

Schematic of manganese chloride (MnCl₂) injection sites and the actual injection sites in the striatum of Case 2 (F97). (A) Schematic lateral views of the monkey brain show the injection sites in the rostral part of the right putamen (pu) and the head of the left caudate nucleus (cd). (B–D) Stereotaxic coordinates for the caudate and putamen injections with respect to the Horsley-Clark (Hc) stereotaxic coordinate system in a sagittal MR image (B) and dorsal view of the rendered brain [(C), 45 h post-injection] are shown. The horizontal and vertical dashed lines labeled Hc0 show the horizontal and frontal zero planes, respectively, in the Horsley-Clark system. The solid yellow lines indicate the coordinates of the centers of the caudate and putamen injection sites shown in (D). The red structures in (C) show the brain vasculature as determined by angiography. Blood vessels appear bright white (high signal) in the anatomical scans, as does the manganese (Mn2) signal (see D). To minimize the error in identifying tracer-containing regions, the signal originating from the vasculature was removed from the anatomical scans by subtracting the corresponding angiography scans. The green structures represent the extents of the Mn2 signal at the sites of the left caudate (cd) and right putamen (pu) injections (spherical regions) and the subsequent path of the tracer in the basal ganglia. The injection coordinates for the caudate were anterior (cdA) 22.0 mm, lateral (cdL) 5 mm, and dorsal (cdD) 13.5 mm. The injection coordinates for the putamen were anterior (puA) 22.0 mm, lateral (puL) 12.5 mm and dorsal (puD) 9 mm. In (D), the bright regions indicate the spatial extent of the Mn²⁺ signal at the caudate and putamen injection sites in coronal, horizontal, and sagittal MR images at 45 h post-injection. Both injections were well localized within the corresponding structures. The green arrow (sagittal image, bottom left) indicates the needle penetration tract for the putamen injection. The green arrowhead (sagittal image, bottom right) indicates axonal transport of Mn2 from the caudate injection site. Major blood vessels are labeled with red arrows in the coronal and sagittal images. Fr0, Frankfurt zero plane; CA, carotid artery; MCA, middle cerebral artery; OA, ophthalmic artery; and PCA, posterior cerebral artery. [Reproduced from Saleem et al. (2002). Copyright © 2002 Elsevier Science Ltd.]

Figure 15

The mouse hippocampal formation in (top) a histologic section and corresponding horizontal T1-weighted MR images (left) 2 h and (right) 6 h after intrahippocampal injection of (middle) 200 mM MnCl₂ (no. 7). Pronounced signal enhancements in the dentate gyrus (DG) and CA3 subfield are complemented by only mild signal increases in the CA1 subfield and subiculum (S). [Reproduced from Watanabe et al. (2004). Copyright © 2002 Elsevier Science Ltd.]

Figure 16

Oblique 2D slices from the MRI 3D volumes obtained from animals in the frog (a, c, e, g) and fish (b, d, f, h) group: (A, B) Normal controls showing the Mn²⁺-enhanced (a) frog and (b) fish ON 48 h and 96 h, respectively, after ivit + injection; (C, D) Mn²⁺-enhanced ON of (b) frog and (c) fish 48 h and 96 h, respectively, after ONT and ivit MnCl₂ injection, with clear delineation of the lesion; (E, F) Mn²⁺-enhanced regenerating ON of (e) frog at 1 mpl and (f) fish at 3 mpl; (G, H) Mn²⁺-enhanced regenerating ON of (g) frog at 3 mpl and (h) fish at 6 mpl. [Reproduced from Sandvig et al. (2011). Copyright © 2003 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 17

Time-lapse MRI of optic nerve enhancement after Mn²⁺ injection into the right eye of KLC+/+ (wildtype littermate) and KLC−/− mice. (A) Representative images from two time-lapse sequences, KLC+/+ (top) and KLC−/− mice, captured at 30 min, 1, 1.5 and 2.5 h after injection of 0.25 μl of 200 mM Mn²⁺ into the right eye. Note that by 2.5 h after injection in the littermate, Mn²⁺ signal is present along the optic nerve, possibly reaching the chiasm. In contrast in the KLC−/− mouse, Mn²⁺ enhancement is only detected in a short segment at the beginning of the optic nerve. (B) At 24 h Mn²⁺ signal has progressed along the optic track in both genotypes, demonstrating that the optic tract is intact. (C) ROI intensity measurement (position indicated by circles in left panel) demonstrate that over the first 160 min after Mn²⁺ injection into the vitreous, intensity (the ratio of optic track vs. cheek muscle) increased in the optic tract by 150% in the wildtype KLL++ while no intensity increase was detected in the KLC−/− mouse. Error bars represent the sum of the standard deviations of the average voxel intensities in optic tract vs. cheek muscle ROIs. For KLC+/+ these were 10% and 6%, respectively and for KLC−/− 7% and 6%. [Reproduced from Bearer et al. (2007). Copyright © 2007 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 18

(A) Successive visualization of the region of interest (ROI) tracks changes in Mn²⁺ enhanced signal intensity during the second hour post-Mn²⁺ lavage. Muscle was used to control for background effects. (B) This figure demonstrates how the ROI was determined using the length of the olfactory bulb and 90° angle to find the midpoint on the olfactory neuronal layer. Each pixel measures 0.23 mm. [Reproduced from Smith et al. (2007). Copyright © 2003 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 19

The differences between Mn²⁺ treated and control mice (no Mn²⁺) clearly demonstrate the increased signal intensity acquired using MEMRI. Data were quantified as a function of the change in signal intensity (ΔSI) over time (min). Slope of line acquired through linear regression. t-Test of Mn²⁺ vs. no Mn^{2+ a}p-Value<0.0001. ^ap-Value is significant. (Mn²⁺ 0.00587±0.001 ΔSI/time (min); n=8; No Mn²⁺: −0.00071±0.001 ΔSI/time (min); n=5). [Reproduced from Smith et al. (2007). Copyright © 2003 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]

Figure 20

SOD-2 overexpression improves the axonal transport deficits displayed by Tg2576 mice. (A) The graph represents the rates on axonal transport as measured in vivo by MEMRI in 12 to 16 months old WT, SOD-2, Tg2576 and Tg2576/SOD-2 mice. Significance was assessed by a one-way ANOVA with Dunnett’s post-test for multiple comparisons. **p, 0.01. [Reproduced from Massaad et al. (2010)].

Figure 21

Mn²⁺ through the lateral olfactory tract is delayed with aging. Setting a statistical threshold of Z=4.0 (after Bonferroni’s correction for comparison across multiple pixels), group-wise statistical maps of changes in MR intensity indicate significant peak enhancement in both anterior (blue arrow) and posterior (purple arrow) at the EARLY (11–12 h) time point. Mid-age group statistical maps indicate sub-threshold enhancement of both anterior and posterior VOI until the 24 h time point. In contrast, aged group peak z-value exceeded the threshold by 24 h in the anterior VOI, but was still sub-thresh-old in the posterior VOI. By 36 h post MnCl₂ administration, aged group posterior VOI reached statistical threshold. Statistical maps are shown in coronal slices superimposed onto a template MRI for localization purposes. Anterior tract=+4.6 mm, posterior tract=+0.2 mm from bregma landmark. [Reproduced from Cross et al. (2008) Copyright © 2003 John Wiley & Sons, Inc. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.]