Faculty Bibliography

The presence of amyloid-beta (Abeta) plaques in the brain is a hallmark pathological feature of Alzheimer's disease (AD). Transgenic mice overexpressing mutant amyloid precursor protein (APP), or both mutant APP and presenilin-1 (APP/PS1), develop Abeta plaques similar to those in AD patients, and have been proposed as animal models in which to test experimental therapeutic approaches for the clearance of Abeta. However, at present there is no in vivo whole-brain imaging method to detect Abeta plaques in mice or men. A novel method is presented to detect Abeta plaques in the brains of transgenic mice by magnetic resonance microimaging (muMRI). This method uses Abeta1-40 peptide, known for its high binding affinity to Abeta, magnetically labeled with either gadolinium (Gd) or monocrystalline iron oxide nanoparticles (MION). Intraarterial injection of magnetically labeled Abeta1-40, with mannitol to transiently open the blood-brain barrier (BBB), enabled the detection of many Abeta plaques. Furthermore, the numerical density of Abeta plaques detected by muMRI and by immunohistochemistry showed excellent correlation. This approach provides an in vivo method to detect Abeta in AD transgenic mice, and suggests that diagnostic MRI methods to detect Abeta in AD patients may ultimately be feasible

The aim of this study was to develop an MRI protocol to evaluate the growth and vascularity of implanted GL261 mouse gliomas on a 7T microimaging system. Both conventional T(1)- and T(2)-weighted imaging and dynamic, contrast-enhanced T(2)*-weighted imaging were performed on 34 mice at different stages of tumor development. MRI measurements of relative cerebral blood volume (rCBV) were compared to histological assessments of microvascular density (MVD). Enhancement on postcontrast T(1)-weighted images was compared to histological assessments of Evan's blue extravasation. Conventional T(2)-weighted and postcontrast T(1)-weighted images demonstrated tumor growth characteristics consistent with previous descriptions of GL261 glioma. Furthermore, measurements of rCBV from MRI data were in good agreement with histological measurements of MVD from the same tumors. Postcontrast enhancement on T(1)-weighted images was observed at all stages of GL261 glioma progression, even before evidence of angiogenesis, indicating that the mechanism of conventional contrast enhancement in MRI does not require neovascularization. These results provide quantitative support for MRI approaches currently used to assess human brain tumors, and form the basis for future studies of angiogenesis in genetically engineered mouse brain tumor models. Magn Reson Med 49:848-855, 2003

Amyloid deposition in Alzheimer's disease (AD) occurs many years before cognitive impairment. Brain imaging techniques targeting plaques will have an important diagnostic value and may help in identifying individuals in preclinical stages of AD. Magnetic resonance imaging (MRI) has a much higher resolution than positron enhanced tomography (PET) imaging and, therefore, is a more sensitive method to detect amyloid plaques. In our initial proof-of-concept studies (Magnetic Resonance in Medicine, in press), we utilized Abeta1-40 peptide, labeled with gadolinium or monocrystalline iron oxide nanoparticles (MION). When either of these ligands is injected in vivo systemically with mannitol to transiently open the blood-brain-barrier, we are able to image ex vivo the majority of Abeta plaques in Tg mice. Using Gd labeled Abeta1-40 and in vivo muMRI, we can also detect a substantial percentage of amyloid lesions. There is a high correlation between the numerical density of Abeta plaques detected by muMRI and by immunohistochemistry. Clinical use of Abeta1-40 is not feasible because it may add to the plaque burden. As a safer approach, we are using gadolinium labeled K6Abeta1-30, a non-toxic Abeta derivative with low propensity to form beta-sheet, while maintaining high affinity for Abeta. Our initial findings indicate that this compound has a similar effect as gadolinium labeled Abeta1-40 in allowing in vivo detection of amyloid plaques in Tg mice. We are currently exploring various ways to enhance the uptake of this compound into the brain. This approach may lead to a diagnostic MRI method to detect Abetaplaques in AD patients

PURPOSE: To examine, using blood oxygen level dependent (BOLD) MRI and EPR oximetry, the changes in oxygenation of intracranial tumors induced by carbogen breathing. MATERIALS AND METHODS: The 9L and CNS-1 intracranial rat tumor models were imaged at 7T, before and during carbogen breathing, using a multi-echo gradient-echo (GE) sequence to map R(2)*. On a different group of 9L tumors, tissue pO(2) was measured using EPR oximetry with lithium phthalocyanine as the oxygen-sensitive material. RESULTS: The average decline in R(2)* with carbogen breathing was 13 +/- 1 s(-1) in the CNS-1 tumors and 29 +/- 4 s(-1) in the 9L tumor. The SI vs. TE decay curves indicate the presence of multiple components in the tumor. Tissue pO(2) in the two 9L tumors measured was 8.6 +/- 0.5 and 3.6 +/- 0.6 mmHg during air breathing, and rose to 20 +/- 7 and 16 +/- 4 mmHg (mean +/- SE) with carbogen breathing. Significant changes were observed by 10 minutes, but changes in pO(2) and R(2)* continued in some subjects over the entire 40 minutes. CONCLUSION: EPR results indicate that glial sarcomas may be radiobiologically hypoxic. Both EPR and BOLD data indicate that carbogen breathing increases brain tumor oxygenation. These data support the use of BOLD imaging to monitor changes in oxygenation in brain tumors

Huntingtin is an essential protein that with mutant polyglutamine tracts initiates dominant striatal neurodegeneration in Huntington's disease (HD). To assess the consequences of mutant protein when huntingtin is limiting, we have studied three lines of compound heterozygous mice in which both copies of the HD gene homolog (Hdh) were altered, resulting in greatly reduced levels of huntingtin with a normal human polyglutamine length (Q20) and/or an expanded disease-associated segment (Q111): Hdh(neoQ20)/Hdh(neoQ20), Hdh(neoQ20)/Hdh(null) and Hdh(neoQ20)/Hdh(neoQ111). All surviving mice in each of the three lines were small from birth, and had variable movement abnormalities. Magnetic resonance micro-imaging and histological evaluation showed enlarged ventricles in approximately 50% of the Hdh(neoQ20)/Hdh(neoQ111) and Hdh(neoQ20)/Hdh(null) mice, revealing a developmental defect that does not worsen with age. Only Hdh(neoQ20)/Hdh(neoQ111) mice exhibited a rapidly progressive movement disorder that, in the absence of striatal pathology, begins with hind-limb clasping during tail suspension and tail stiffness during walking by 3-4 months of age, and then progresses to paralysis of the limbs and tail, hypokinesis and premature death, usually by 12 months of age. Thus, dramatically reduced huntingtin levels fail to support normal development in mice, resulting in reduced body size, movement abnormalities and a variable increase in ventricle volume. On this sensitized background, mutant huntingtin causes a rapid neurological disease, distinct from the HD-pathogenic process. These results raise the possibility that therapeutic elimination of huntingtin in HD patients could lead to unintended neurological, as well as developmental side-effects

Although shimming can improve static field inhomogeneity, local field imperfections induced by tissue susceptibility differences cannot be completely corrected and can cause substantial signal loss in gradient echo images through intravoxel dephasing. Dephasing increases with voxel size so that one simple method of reducing the effect is to use thin slices. Signal-to-noise ratio (SNR) can then be increased by averaging over the subslices to form the final, thick slice. We call this method subslice averaging or SSAVE. Alternatively, a range of different amplitude slice select rephase gradients can be used to compensate for different susceptibility induced gradient offsets. The final image can then be formed by combining individual images in a variety of ways: summation, summation of the squares of the images, forming the maximum intensity projection of the image set, and Fourier transformation followed by summation. We show here that, contrary to previous claims, the theoretical sensitivity (i.e., SNR divided by the square root of the imaging time) of all these alternative methods is very similar. However, performance time (i.e., minimum-imaging time) of the simplest method, SSAVE, is much shorter than that of alternatives. This is confirmed experimentally on phantoms and anesthetized mice. Magn Reson Med 45:470-476, 2001.