Oscillating and pulsed gradient diffusion magnetic resonance microscopy over an extended b-value range: implications for the characterization of tissue microstructure.

Oscillating gradient spin-echo (OGSE) pulse sequences have been proposed for acquiring diffusion data with very short diffusion times, which probe tissue structure at the subcellular scale. OGSE sequences are an alternative to pulsed gradient spin echo measurements, which typically probe longer diffusion times due to gradient limitations. In this investigation, a high-strength (6600 G/cm) gradient designed for small-sample microscopy was used to acquire OGSE and pulsed gradient spin echo data in a rat hippocampal specimen at microscopic resolution. Measurements covered a broad range of diffusion times (TDeff = 1.2-15.0 ms), frequencies (ω = 67-1000 Hz), and b-values (b = 0-3.2 ms/µm2). Variations in apparent diffusion coefficient with frequency and diffusion time provided microstructural information at a scale much smaller than the imaging resolution. For a more direct comparison of the techniques, OGSE and pulsed gradient spin echo data were acquired with similar effective diffusion times. Measurements with similar TDeff were consistent at low b-value (b < 1 ms/µm(2) ), but diverged at higher b-values. Experimental observations suggest that the effective diffusion time can be helpful in the interpretation of low b-value OGSE data. However, caution is required at higher b, where enhanced sensitivity to restriction and exchange render the effective diffusion time an unsuitable representation. Oscillating and pulsed gradient diffusion techniques offer unique, complementary information. In combination, the two methods provide a powerful tool for characterizing complex diffusion within biological tissues.

Magnetic resonance microscopy and immunohistochemistry of the CNS of the mutant SOD murine model of ALS reveals widespread neural deficits.

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that primarily affects motor neurons and descending motor tracts of the CNS. We have evaluated the CNS of a murine model of familial ALS based on the over-expression of mutant human superoxide dismutase (mSOD; G93A) using magnetic resonance microscopy (MRM) and immunohistochemistry (IHC). Three-dimensional volumetric analysis was performed from 3D T2*-weighted images acquired at 17.6 T at isotropic resolutions of 40 mum. Compared to controls, mSOD mice had significant reductions in the volumes of total brain, substantia nigra, striatum, hippocampus, and internal capsule, with decreased cortical thickness in primary motor and somatosensory cortices. In the spinal cord, mSOD mice had significantly decreased volume of both the total grey and white matter; in the latter case, the volume change was confined to the dorsal white matter. Increased apoptosis, GFAP positive astrocytes, and/or activated microglia were observed in all those CNS regions that showed volume loss except for the hippocampus. The MRM findings in mSOD over-expressing mice are similar to data previously obtained from a model of ALS-parkinsonism dementia complex (ALS-PDC), in which neural damage occurred following a diet of washed cycad flour containing various neurotoxins. The primary difference between the two models involves a significantly greater decrease in spinal cord white matter volume in mSOD mice, perhaps reflecting variations in degeneration of the descending motor tracts. The extent to which several CNS structures are impacted in both murine models of ALS argues for a reevaluation of the nature of the pathogenesis of ALS since CNS structures involved in Parkinson's and Alzheimer's diseases appear to be affected as well.

Diffusion tensor microscopy indicates the cytoarchitectural basis for diffusion anisotropy in the human hippocampus.

BACKGROUND AND PURPOSE: Observing changes to water diffusivity and fractional anisotropy (FA) for particular hippocampal regions may improve the sensitivity and specificity of diffusion tensor MR imaging for hippocampal pathologies like Alzheimer disease and mesial temporal sclerosis. As a first step toward this goal, this study characterized the cytoarchitectural features underlying diffusion anisotropy in human hippocampus autopsy specimens at 60-microm in-plane resolution. MATERIALS AND METHODS: Eight-millimeter coronal segments of the hippocampal body were dissected from 5 autopsy specimens (mean = 55.6 +/- 6.2 years of age) with short postmortem intervals to fixation (21.2 +/- 5.7 hours) and no histologic evidence of neuropathology. Diffusion tensor microscopy data were collected from hippocampal specimens by using a 14.1T magnet with a protocol that included 21 unique diffusion gradient orientations (diffusion time = 17 ms, b = 1250 s/mm(2)). The resulting images were used to determine the mean diffusivity, FA, and principal fiber orientation for manually segmented hippocampal regions that included the stratum oriens, stratum radiatum, stratum pyramidale (CA1 and CA3), stratum lacunosum-molecular, hilus, molecular layer, granule cell layer, fimbria, and subiculum. RESULTS: Diffusion-weighted images had high signal-to-noise ratios (31.1 +/- 13.0) and delineated hippocampal anatomy well. Water diffusivity ranged from 1.21 +/- 0.22 x 10(-4) mm(2)/s in the fimbria to 3.48 +/- 0.72 x 10(-4) mm(2)/s in granule cells (analysis of variance, P<.001). Color fiber-orientation maps indicated the underlying microstructures responsible for diffusion anisotropy in the hippocampal lamina. CONCLUSION: Diffusion tensor microscopy provided novel microstructural information about the different lamina of the human hippocampus. These ex vivo data obtained at high-magnetic-field strengths can be used to study injury-specific diffusion changes to susceptible hippocampal regions and may lead to more specific MR imaging surrogate markers for Alzheimer disease or epilepsy.

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

A comprehensive three-dimensional digital atlas database of the C57BL/6J mouse brain was developed based on magnetic resonance microscopy images acquired on a 17.6-T superconducting magnet. By using both manual tracing and an atlas-based semi-automatic segmentation approach, T2-weighted magnetic resonance microscopy images of 10 adult male formalin-fixed, excised C57BL/6J mouse brains were segmented into 20 anatomical structures. These structures included the neocortex, hippocampus, amygdala, olfactory bulbs, basal forebrain and septum, caudate-putamen, globus pallidus, thalamus, hypothalamus, central gray, superior colliculi, inferior colliculi, the rest of midbrain, cerebellum, brainstem, corpus callosum/external capsule, internal capsule, anterior commissure, fimbria, and ventricles. The segmentation data were formatted and stored into a database containing three different atlas types: 10 single-specimen brain atlases, an average brain atlas and a probabilistic atlas. Additionally, quantitative group information, such as variations in structural volume, surface area, magnetic resonance microscopy image intensity and local geometry, were computed and stored as an integral part of the database. The database augments ongoing efforts with other high priority strains as defined by the Mouse Phenome Database focused on providing a quantitative framework for accurate mapping of functional, genetic and protein expression patterns acquired by a myriad of technologies and imaging modalities.

Alginate assessment by NMR microscopy.

Alginate hydrogels have long been used to encapsulate cells for the purpose of cell transplantation. However, they also have been criticized because they fail to consistently maintain their integrity for extended periods of time. Two issues of critical importance that have yet to be thoroughly addressed concerning the long-term integrity of alginate/poly-L-lysine/alginate microcapsules are: (i) are there temporal changes in the alginate/poly-L-lysine interaction and (ii) are there temporal changes in the alginate gel structure. NMR microscopy is a non-invasive analytical technique that can address these issues. in this report, we present data to demonstrate the utility of (1)H NMR microscopy to (i) visualize the poly-L-lysine layer in an effort to address the first question, and (ii) to observe temporal changes in the alginate matrix that may represent changes in the gel structure.

Quantitative measurement of neurodegeneration in an ALS-PDC model using MR microscopy.

Exposure to cycad (Cycas micronesica K.D. Hill) toxins via diet has been shown to induce neurodegeneration in vivo that mimics the progressive neurological disease, amyotrophic lateral sclerosis--parkinsonism dementia complex (ALS--PDC). In previous studies, specific cortical and subcortical cell loss was measured with conventional stained sections. In the present study, magnetic resonance (MR) microscopy was used to examine neurodegeneration in three dimensions (3D) in isolated intact brains and spinal cords. Mice were fed washed cycad for 2 months and showed progressive motor deficits resembling human ALS--PDC. CNS tissue was imaged at 17.6 T. T2* scans were acquired on both spinal cord and brain samples with an isotropic resolution of 41 microm. Through MR volumetrics, cycad-fed mice showed significantly decreased volumes in lumbar spinal cord gray matter, substantia nigra, striatum, basal nucleus/internal capsule, and olfactory bulb. Cortical measurements of conventionally stained sections revealed that cycad-fed mice also showed decreased cortical thickness. These results show that MR microscopy (MRM) is sensitive enough to measure degeneration in this early stage model of a progressive neurological disease with strong correlations to behavioral deficits and histological results and may be applicable in vivo to the same model. Similar analysis may be used in the future as a diagnostic aid in tracking the early progression of neurological disorders in preclinical human subjects.

Observation of significant signal voids in images of large biological samples at 11.1 T.

Proton MRI of large biological samples were obtained on an 11.1 T / 40 cm instrument. Images were obtained of a fixed human brain and a large piece of fresh beef. The proton MR images demonstrate severe distortions within these conductive samples, indicative of shortened electrical wavelengths and wave behavior within the sample. These observations have significant implications with respect to the continuing evolution of MR to higher magnetic field strengths on large samples, particularly on humans.

Progress in high field MRI at the University of Florida.

In this article we report on progress in high magnetic field MRI at the University of Florida in support of our new 750MHz wide bore and 11.7T/40cm MR instruments. The primary emphasis is on the associated rf technology required, particularly high frequency volume and phased array coils. Preliminary imaging results at 750MHz are presented. Our results imply that the pursuit of even higher fields seems warranted.

MR microscopy of multicomponent diffusion in single neurons.

This study examines multicomponent diffusion in isolated single neurons and discusses the implications of the results for macroscopic water diffusion in tissues. L7 Aplysia neurons were isolated and analyzed using a 600 MHz Bruker wide-bore instrument with a magnetic susceptibility-matched radiofrequency microcoil. Using a biexponential fit, the apparent diffusion coefficients (ADCs) from the cytoplasm (with relative fraction) were 0.48 +/- 0.14 x 10(-3) mm2 x s(-1) (61 +/- 11%) for the fast component, and 0.034 +/- 0.017 x 10(-3) mm2 x s(-1) (32 +/- 11%) for the slow component (N = 10). Diffusion in the nucleus appears to be primarily monoexponential, but with biexponential analysis it yields 1.31 +/- 0.32 x 10(-3) mm2 x s(-1) (89 +/- 6%) for the fast component and 0.057 +/- 0.073 x 10(-3) mm2 x s(-1) (11 +/- 6%) for the slow (N = 5). The slow component in the nucleus may be explained by cytoplasmic volume averaging. These data demonstrate that water diffusion in the cytoplasm of isolated single Aplysia neurons supports a multiexponential model. The ADCs are consistent with previous measurements in the cytoplasm of single neurons and with the slow ADC measurement in perfused brain slices. These distributions may explain the multiple compartments observed in tissues, greatly aiding the development of quantitative models of MRI in whole tissues.

MR imaging measurement of compartmental water diffusion in perfused heart slices.

Myocardial tissue slices were isolated from the left ventricular free wall (7 slices) and left ventricular papillary muscle (3 slices) of New Zealand White male rabbits (n = 4) and were subsequently superfused with a modified St. Thomas' Hospital cardioplegic solution at 19 degrees C. The diffusion-weighted images were obtained with a 600-MHz nuclear magnetic resonance spectrometer using diffusion gradient b-values that ranged from 166 to 6,408 s/mm(2); the apparent diffusion coefficient of water in the tissues were subsequently calculated. All of the tissue samples that were studied exhibited nonmonoexponential diffusion. Data from seven slices were mathematically fitted by a biexponential expression with a fast diffusion component of 0.72 +/- 0.07 x 10(-3) mm(2)/s, and a slow diffusion component of 0.060 +/- 0.033 x 10(-3) mm(2)/s. The fast component dominated the calculated apparent diffusion coefficient of the tissue, composed of 82 +/- 3% of the overall diffusion-dependent signal decay. Thus myocardial tissue exhibits characteristics consistent with multiple compartments of diffusion. This work has important implications for myocardial diffusion tensor imaging, as well as the changes in diffusion that have been reported following myocardial ischemia.

Nuclear magnetic resonance spectroscopy and imaging in animal research.

Nuclear magnetic resonance (NMR) spectroscopy and imaging can be used to investigate, noninvasively, a wide range of biological processes in systems as diverse as protein solutions, single cells, isolated perfused organs, and tissues in vivo. It is also possible to combine different NMR techniques enabling metabolic, anatomical, and physiological information to be obtained in the same experiment. This review provides a simple overview of the basic principles of NMR and outlines both the advantages and disadvantages of NMR spectroscopy and imaging. A few examples of potential applications of NMR spectroscopy and imaging are presented, which demonstrate the range of questions that can be asked using these techniques. The potential impact of using NMR techniques in a biomedical research program on the total number of animals required for specific investigations, as well as the number of animals used in research, are discussed. The article concludes with a personal perspective on the impact of continuing improvements in NMR technology for future applications in animal research.

Two-component diffusion tensor MRI of isolated perfused hearts.

Nonmonoexponential MR diffusion decay behavior has been observed at high diffusion-weighting strengths for cell aggregates and tissues, including the myocardium; however, implications for myocardial MR diffusion tensor imaging are largely unknown. In this study, a slow-exchange-limit, two-component diffusion tensor model was fitted to diffusion-weighted images obtained in isolated, perfused rat hearts. Results indicate that there are at least two distinct components of anisotropic diffusion, characterized by a "fast" component whose principal diffusivity is comparable to that of the perfusate, and a highly anisotropic "slow" component. It is speculated that the two components correspond to tissue compartments and have a general agreement with the orientations of anisotropy, or fiber orientations, in the myocardium. Moreover, consideration of previous studies of myocardial diffusion suggests that the presently observed fast component may likely be dominated by diffusion in the vascular space, whereas the slow component may include the intracellular and interstitial compartments. The implications of the results for myocardial fiber orientation mapping and limitations of the current two-component model used are also discussed.

Estrogens decrease reperfusion-associated cortical ischemic damage: an MRI analysis in a transient focal ischemia model.

BACKGROUND AND PURPOSE: Early identification of irreversible cerebral ischemia is critical in defining strategies that influence neuronal survival after stroke. We used MRI to investigate the effects of 17beta-estradiol (E2) on the temporal evolution of focal ischemia. METHODS: Female rats were ovariectomized and divided into 1 of 2 groups: ovariectomy alone (OVX; n=4) or ovariectomy with estrogen replacement (OVX+E2; n=3). Both groups were then subjected to 1-hour middle cerebral artery occlusion (MCAO), with the use of a standardized endovascular monofilament model, followed by reperfusion. Sequential diffusion-weighted (DWI) and T2-weighted (T2WI) MRI were obtained during and after the MCAO. In separate groups of animals (n=5 for OVX and OVX+E2), cerebral blood flow (CBF) was measured by laser-Doppler methods before, during, and after occlusion. RESULTS: DWI detected similar lesion characteristics during MCAO in both groups. In the OVX group, lesion size did not change during reperfusion, but the signal intensity ratio increased early and stabilized during the latter stages. In contrast, DWI lesion size decreased during reperfusion in OVX+E2 rats by 50% to 60% (P<0.05), a size reduction almost exclusively limited to cortical regions. During MCAO, the signal intensity ratio in OVX+E2 rats was reduced compared with OVX rats. Reperfusion further attenuated the signal intensity ratio in cortical but not subcortical regions (P<0.05 versus OVX). T2WI revealed no lesions in either group during MCAO, but it detected lesion sizes similar to that of DWI during reperfusion. Furthermore, similar patterns and magnitudes of estrogen treatment-related decrease in lesion size were noted after reperfusion. T2WI demonstrated less intense signal intensity ratio changes in both groups compared with DWI. There were no differences in CBF between groups either during occlusion, early reperfusion, or 1 day after reperfusion. CONCLUSIONS: This study strongly suggests that estrogens selectively protect cortical tissue from ischemic damage during MCAO and that this protection is exerted during both the occlusion and reperfusion phases of ischemia and does not involve an estrogen-related change in CBF.

NMR spectroscopy of single neurons.

The first spatially localized NMR spectra of osmolytes and metabolites from single isolated neurons have been obtained using a combination of high magnetic field strengths and NMR radio frequency (RF) microcoils. The proton spectra display peaks at high concentrations (100-300 mM) assigned to betaine and choline, and other metabolite resonances including lactate at lower concentrations in the order of 10s of millimoles. The volumes examined were approximately 10 nl, over two orders of magnitude less than previously possible. In these initial experiments; the cells were unperfused and the signal intensities of the osmolytes decrease with time, a phenomenon consistent with cell swelling. This work demonstrates the technical feasibility of NMR spectroscopy of single cells, further broadening the scope of NMR spectroscopy of living tissues from application to entire living organisms (man and animal models) and isolated tissues (perfused organs and cultured assemblies of cells) and now to single cells. Magn Reson Med 44:19-22, 2000.

MRI measurement of cell volume fraction in the perfused rat hippocampal slice.

T(1)-weighted NMR imaging of the isolated perfused rat hippo-campal slice was used to estimate cell volume fraction. Eight brain slices were studied in artificial cerebrospinal fluid (aCSF) using a 600 MHz narrow bore spectrometer and a home built perfusion chamber. Cell volume fraction was calculated as 1 - f(ECS), where f(ECS) is the distribution volume of gadodiamide in the slice. This was determined by measuring the T(1) of the slice before and after perfusion with gadodiamde. A mean cell volume fraction of 0.66 +/- 0. 04 was estimated. The addition of 60 mM mannitol to three of the brain slices produced a 26% decrease in the cell volume fraction. The technique affords a simple means of estimating cell volume fraction and can be extended to produce images reflecting cell density. Magn Reson Med 42:603-607, 1999.

Nuclear magnetic resonance imaging measurements of water diffusion in the perfused hippocampal slice during N-methyl-D-aspartate-induced excitotoxicity.

Significant changes in the apparent diffusion coefficient of water are observed in nuclear magnetic resonance images of patients with acute ischemic stroke. However, the underlying mechanisms of these apparent diffusion coefficient changes are still unresolved. To analyse possible mechanisms, this study applies nuclear magnetic resonance imaging on a 14.1 Tesla narrow-bore magnet to quantitatively study water diffusion in individually perfused brain slices following exposure to N-methyl-D-aspartate excitotoxicity. The results indicate that brain slices have at least two distinct diffusing water compartments with apparent diffusion coefficients of 0.96+/-0.10x10(-3) mm2/s and 0.06+/-0.01x10(-3) mm2/s. When excitotoxicity was induced with N-methyl-D-aspartate, there was a significant decrease in the fraction of the fast diffusing water component in the slices (P<0.001). However, neither apparent diffusion coefficient changed significantly. Prior treatment with dizocilpine maleate (MK-801) depressed the effects of N-methyl-D-aspartate (P<0.01, ANOVA). The results demonstrate brain slice compartmental changes resulting from direct receptor stimulation and provide evidence for tissue water redistribution as an important mechanism for changes in apparent diffusion coefficient seen in clinical magnetic resonance imaging. The brain slice preparation affords a well-controlled method to study the mechanisms of tissue nuclear magnetic resonance contrast, bridging the gap between basic nuclear magnetic resonance studies and clinical magnetic resonance imaging. The brain slice model also offers a new way to test the utility of potential anti-stroke drugs using high field nuclear magnetic resonance imaging.

The effect of ouabain on water diffusion in the rat hippocampal slice measured by high resolution NMR imaging.

High resolution NMR imaging of the isolated perfused rat hippocampal slice was used to quantitate ADC changes following ouabain-induced cell swelling. Hippocampal slices were studied in artificial cerebrospinal fluid and then in ouabain using a 600-MHz narrow bore spectrometer and a home-built perfusion chamber. The brain slices demonstrated biexponential diffusion behavior. After perfusion with 1 mMouabain, there was an increase in the fraction of slowly diffusing water. The ADCs of the two fractions did not change. These data support the hypothesis that the decrease in the ADC of brain water following an ischemic attack is caused by cell swelling. The relative amplitudes of the two diffusing fractions do not match the expected ratio of intracellular and extracellular fractions. This discrepancy may be principally due to the difference in T2 relaxation rates of the two compartments.

NMR microscopy--beginnings and new directions.

In this paper we briefly review the origins of NMR microscopy, and in the spirit of the Sir Peter Mansfield Symposium of which this presentation was a part, point out especially Sir Mansfield and his co-workers contributions in this area. We then review some recent studies applying magnetic resonance (MR) microscopy focusing on our own contributions in these regards, in particular with reference to imaging of single neurons and more recent microimaging studies on isolated perfused brain slices. Finally we briefly describe recent preliminary studies on the feasibility of spectroscopic experiments that may be performed at the single cell level, further illustrating the growing scope and potential of magnetic resonance imaging (MRI) in general as a tool for examining biological systems non-invasively.

MR microscopy of perfused brain slices.

To study the origins of signal changes in clinical MRI we have previously studied isolated single neuronal cells by MR microscopy. To account for the extracellular environment of the cells, we have developed a prototype perfusion chamber for MR microimaging of perfused rat hippocampal brain slices. To demonstrate the utility of this model, brain slices were initially perfused in isotonic solutions and then subjected to osmotic perturbations via perfusate exchange with 20% hypertonic and 20% hypotonic solutions. In diffusion weighted images, signal intensity changes of +16(sigma(n-1) = 11)% (hypotonic) and -26(sigma(n-1) = 10)% (hypertonic) were observed. No significant variation in response was observed across the slice when several subregions were examined. These observations are consistent with the view that contrast changes are driven primarily by changes in the intra- and extracellular compartmentation of water. This is the first report of MR microimaging of the isolated brain slice. The technique will enable the correlation of MR microimaging measurements with microscopic changes using other modalities and techniques to provide a better understanding of signals in clinical MRI.

A study of diffusion isotropy in single neurons by using NMR microscopy.

The apparent diffusion coefficient (ADC) of water was measured in single Aplysia californica neurons by using NMR microscopy encoded in each of two perpendicular gradient directions. Comparisons of the mean ADCs of the gross nuclear and cytoplasmic compartments in five cells, and 50 subregions within these cells, showed no significant difference between the diffusion measurements in the majority of cases. Since anisotropic diffusion would make the ADC dependent on the encoding direction, the results indicate that the ADC in these single neurons is isotropic at the spatial and temporal resolutions used in these studies. Consequently, a single scalar ADC measurement is sufficient for characterizing the ADC in these cells, hence reducing the acquisition time and measurement complexity that would have been required had the ADC been anisotropic.