Purpose: To evaluate manganese (Mn 2ϩ )-enhanced MRI in a longitudinal study of normal and injured rat visual projections. Materials and Methods:MRI was performed 24 hours after unilateral intravitreal injection of MnCl 2 (150 nmol) into adult Fischer rats that were divided into four groups: 1) controls (N ϭ 5), 2) dose-response (N ϭ 10, 0.2-200 nmol), 3) time-response with repeated MRI during 24 -168 hours post injection (N ϭ 4), and 4) optic nerve crush (ONC) immediately preceding the MnCl 2 injection (N ϭ 7). Control and ONC animals were reinjected with MnCl 2 20 days after the first injection, and MRI was performed 24 hours later. Results:In the control group, the optic projection was visualized from the retina to the superior colliculus, with indications of transsynaptic transport to the cortex. There was a semilogarithmic relationship between the Mn 2ϩ dose and Mn 2ϩ enhancement from 4 to 200 nmol, and the enhancement decayed gradually to 0 by 168 hours. No Mn 2ϩ -enhanced signal was detected distal to the ON crush site. In the control group, similar enhancement was obtained after the first and second MnCl 2 injections, while in the ONC group the enhancement proximal to the crush site was reduced 20 days post lesion (20dpl). Conclusion:Mn 2ϩ -enhanced MRI is a viable method for temporospatial visualization of normal and injured ON in the adult rat. The observed reduction in the Mn 2ϩ signal proximal to the ONC is probably a result of retrograde damage to the retinal ganglion cells, and not of Mn 2ϩ toxicity.

Purpose:To provide dose-response data for the safe and effective use of MnCl 2 for manganese (Mn 2ϩ ) -enhanced MRI (MEMRI) of the visual pathway. Materials and Methods:Retinal ganglion cell (RGC) toxicity, CNR in MEMRI, axon density resolution for MEMRI, mode of axonal transport and clearance of Mn 2ϩ from the vitreous after ivit were investigated. After 0, 30, 150, 300, 1500, and 3000 nmol ivit MnCl 2 , neural toxicity was measured by counting surviving RGC back-filled with FluroGold (FG), CNR of the vitreous body and visual pathway by three-dimensional (3D) MEMRI, resolution of ON axon density by correlating CNR with axon density, and axonal transport of Mn 2ϩ by studying CNR in 3D MEMRI of the ON after ion of 200 nmol MnCl 2 .Results: There were no changes in RGC density after ivit MnCl 2 Յ 150 nmol, and reductions of 12%, 57%, and 94% occurred after 300, 1500, and 3000 nmol MnCl 2 . CNR increased in the visual pathway with MnCl 2 Յ 300 nmol, and decreased when the dose was raised further. Minimum detectable ON axon densities were 125,000/mm 2 . After 200 nmol ion MnCl 2 , CNRϾ0 were recorded distally from the ion site, but there was no signal in the retina. At ivit doses Ͼ1500 nmol, clearance from the vitreous body was impaired. Conclusion:The optimal dose for MEMRI of the rat visual pathway was found to be 150 -300 nmol ivit MnCl 2 . Higher doses are toxic, causing RGC death, impair active clearance from the vitreous, and loss of Mn 2ϩ enhancement throughout the visual pathway. Mn 2ϩ traffic within RGC axons is mediated mainly by anterograde transport.

Purpose:To evaluate manganese (Mn 2ϩ )-enhanced MRI (MEMRI) and diffusion tensor imaging (DTI) as tools for detection of axonal injury and regeneration after intravitreal peripheral nerve graft (PNG) implantation in the rat optic nerve (ON). Materials and Methods:In adult Fischer rats, retinal ganglion cell (RGC) survival was evaluated in Flurogold (FG) back-filled retinal whole mounts after ON crush (ONC), intravitreal PNG, and intravitreal MnCl 2 injection (150 nmol) at 0 and 20 days post lesion (dpl). MEMRI and echoplanar DTI (DTI-EPI) was obtained of noninjured ON one day after intravitreal MnCl 2 injection, and at 1 and 21 dpl after ONC, intravitreal PNG, and intravitreal MnCl 2 injections given at 0 and 20 dpl. GAP-43 immunohistochemistry was performed after the last MRI.Results: ONC reduced RGC density in retina by 94% at 21 dpl compared to noninjured ON without MnCl 2 injections. Both intravitreal PNG and intravitreal MnCl 2 injections improved RGC survival in retina, which was reduced by 90% (ONCϩMnCl 2 ), 82% (ONCϩPNG), and 74% (ONCϩPNGϩMnCl 2 ) compared to noninjured ON. DTIderived parameters (fractional anisotropy [FA], mean diffusivity, axial diffusivity , and radial diffusivity Ќ ) were unaffected by the presence of Mn 2ϩ in the ON. At 1 dpl, CNR MEMRI and were reduced at the injury site, while at 21 dpl they were increased at the injury site compared to values measured at 1 dpl. GAP-43 immunoreactive axons were present in the ON distal to the ONC injury site. Conclusion:MEMRI and DTI enabled detection of functional and structural degradation after rat ON injury, and there was correlation between the MRI-derived and immunohistochemical measures of axon regeneration. UNLIKE AXONS IN THE PERIPHERAL NERVOUS SYS-TEM, those in the central nervous system (CNS) of adult mammals do not regenerate after injury (1,2). Failure to regenerate is attributed to a combination of axon growth arrest by myelin-associated and scar-derived inhibitory molecules (3,4), and to the absence of growth-promoting neurotrophic factors in the adult CNS (5). Several therapeutic interventions have been tested in animal models to try to promote axon regeneration in the adult mammalian CNS, including neutralizing inhibitory molecules by bacterial enzyme chondroitinase ABC (6) and administration of growthpromoting factors released, for example, from olfactory ensheathing cells and stem cells (7,8). One method that has been shown to have an effect in an optic nerve (ON) animal model is the intravitreal implantation of a peripheral nerve graft (PNG) after ON transection. Schwann cells in the PNG produce trophic factors that promote both retinal ganglion cell (RGC) survival and axon growth through the putative inhibitory environment of the injured ON, as documented in several studies (9 -11). Most studies detect regenerating axons in the CNS using traditional axon tracing techniques (e.g.,

Magnetic resonance imaging (MRI)-based tracking is increasingly attracting attention as a means of better understanding stem cell dynamics in vivo. Intracellular labeling with micrometer-sized particles of iron oxide (MPIOs) provides a practical MRI-based approach due to superior detectability relative to smaller iron oxide particles. However, insufficient information is available about the general utility across cell types and the effects on cell vitality of MPIO labeling of human stem cells. We labeled six human cell types from different sources: mesenchymal stem cells derived from bone marrow (MSCs), mesenchymal stem cells derived from adipose tissue (ASCs), presumptive adult neural stem cells (ad-NSCs), fetal neural progenitor cells (f-NPCs), a glioma cell line (U87), and glioblastoma tumor stem cells (GSCs), with two different sizes of MPIOs (0.9 and 2.84 µm). Labeling and uptake efficiencies were highly variable among cell types. Several parameters of general cell function were tested in vitro. Only minor differences were found between labeled and unlabeled cells with respect to proliferation rate, mitotic duration, random motility, and capacity for differentiation to specific phenotypes. In vivo behavior was tested in chicken embryos and severe combined immunodeficient (SCID) mice. Postmortem histology showed that labeled cells survived and could integrate into various tissues. MRI-based tracking over several weeks in the SCID mice showed that labeled GSCs and f-NPCs injected into the brain exhibited translocations similar to those seen for unlabeled cells and as expected from migratory behavior described in previous studies. The results support MPIO-based cell tracking as a generally useful tool for studies of human stem cell dynamics in vivo.

Several studies have demonstrated that untreated tumors may show significant fluctuations in tissue oxygen tension (pO(2)). Radiation treatment may induce changes in the tumor microenvironment that alter the pO(2) fluctuation pattern. The purpose of the present study was to investigate whether pO(2) fluctuations may also occur in irradiated tumors. A-07 human melanoma xenografts were irradiated with single doses of 0, 5 or 10 Gy. Fluctuations in pO(2) were recorded with OxyLite probes prior to irradiation and 24 and 72 h after the radiation exposure. Radiation-induced changes in the tumor microenvironment (i.e. blood perfusion and extracellular volume fraction) were assessed by dynamic contrast-enhanced magnetic resonance imaging. Seventy-two hours after 10 Gy, tumor blood perfusion had decreased to approximately 40% of that prior to irradiation, whereas the extracellular volume fraction had increased by approximately 25%. Fluctuations in pO(2) were seen in most tumors, irrespective of radiation dose and time after irradiation. The mean pO(2), the number of fluctuations around the mean pO(2), the number of fluctuations around threshold pO(2) values of 1, 2, 3, 5, 7 and 10 mmHg, and the amplitude of the fluctuations were determined for each pO(2) trace. No significant differences were detected between irradiated and unirradiated tumors. The results showed that pO(2) fluctuations may occur in irradiated tumors and that the pO(2) fluctuation pattern in A-07 tumors exposed to 5 or 10 Gy is similar to that in untreated tumors. Consequently, these doses did not induce changes in the tumor microenvironment that were sufficient to cause detectable alterations in the pO(2) fluctuation pattern.