Lymphatic and vascular markers in an optic nerve crush model in rat.

Only few tissues lack lymphatic supply, such as the CNS or the inner eye. However, if the scleral border is compromised due to trauma or tumor, lymphatics are detected in the eye. Since the situation in the optic nerve (ON), part of the CNS, is not clear, the aim of this study is to screen for the presence of lymphatic markers in the healthy and lesioned ON. Brown Norway rats received an unilateral optic nerve crush (ONC) with defined force, leaving the dura intact. Lesioned ONs and unlesioned contralateral controls were analyzed 7 days (n = 5) and 14 days (n = 5) after ONC, with the following markers: PDGFRb (pericyte), Iba1 (microglia), CD68 (macrophages), RECA (endothelial cell), GFAP (astrocyte) as well as LYVE-1 and podoplanin (PDPN; lymphatic markers). Rat skin sections served as positive controls and confocal microscopy in single optical section mode was used for documentation. In healthy ONs, PDGFRb is detected in vessel-like structures, which are associated to RECA positive structures. Some of these PDGFRb+/RECA+ structures are closely associated with LYVE-1+ cells. Homogenous PDPN-immunoreactivity (IR) was detected in healthy ON without vascular appearance, showing no co-localization with LYVE-1 or PDGFRb but co-localization with GFAP. However, in rat skin controls PDPN-IR was co-localized with LYVE-1 and further with RECA in vessel-like structures. In lesioned ONs, numerous PDGFRb+ cells were detected with network-like appearance in the lesion core. The majority of these PDGFRb+ cells were not associated with RECA-IR, but were immunopositive for Iba1 and CD68. Further, single LYVE-1+ cells were detected here. These LYVE-1+ cells were Iba1-positive but PDPN-negative. PDPN-IR was also clearly absent within the lesion site, while LYVE-1+ and PDPN+ structures were both unaltered outside the lesion. In the lesioned area, PDGFRb+/Iba1+/CD68+ network-like cells without vascular association might represent a subtype of microglia/macrophages, potentially involved in repair and phagocytosis. PDPN was detected in non-lymphatic structures in the healthy ON, co-localizing with GFAP but lacking LYVE-1, therefore most likely representing astrocytes. Both, PDPN and GFAP positive structures are absent in the lesion core. At both time points investigated, no lymphatic structures can be identified in the lesioned ON. However, single markers used to identify lymphatics, detected non-lymphatic structures, highlighting the importance of using a panel of markers to properly identify lymphatic structures.

Characterization of dsRed2-positive cells in the doublecortin-dsRed2 transgenic adult rat retina.

Doublecortin (DCX) is predominantly expressed in neuronal precursor cells and young immature neurons of the developing and adult brain, where it is involved in neuronal differentiation, migration and plasticity. Moreover, its expression pattern reflects neurogenesis, and transgenic DCX promoter-driven reporter models have been previously used to investigate adult neurogenesis. In this study, we characterize dsRed2 reporter protein-expressing cells in the adult retina of the transgenic DCX promoter-dsRed2 rat model, with the aim to identify cells with putative neurogenic activity. Additionally, we confirmed the expression of the dsRed2 protein in DCX-expressing cells in the adult hippocampal dentate gyrus. Adult DCX-dsRed2 rat retinas were analyzed by immunohistochemistry for expression of DCX, NF200, Brn3a, Sox2, NeuN, calbindin, calretinin, PKC-a, Otx2, ChAT, PSA-NCAM and the glial markers GFAP and CRALBP, followed by confocal laser-scanning microscopy. In addition, brain sections of transgenic rats were analyzed for dsRed2 expression and co-localization with DCX, NeuN, GFAP and Sox2 in the cortex and dentate gyrus. Endogenous DCX expression in the adult retina was confined to horizontal cells, and these cells co-expressed the DCX promoter-driven dsRed2 reporter protein. In addition, we encountered dsRed2 expression in various other cell types in the retina: retinal ganglion cells (RGCs), a subpopulation of amacrine cells, a minority of bipolar cells and in perivascular cells. Since also RGCs expressed dsRed2, the DCX-dsRed2 rat model might offer a useful tool to study RGCs in vivo under various conditions. Müller glial cells, which have previously been identified as cells with stem cell features and with neurogenic potential, did express neither endogenous DCX nor the dsRed2 reporter. However, and surprisingly, we identified a perivascular glial cell type expressing the dsRed2 reporter, enmeshed with the glia/stem cell marker GFAP and colocalizing with the neural stem cell marker Sox2. These findings suggest the so far undiscovered existence of perivascular associated cell with neural stem cell-like properties in the adult retina.

Acute and chronic evolution of MRI findings in a case of posterior spinal cord ischemia.

STUDY DESIGN: Case report. OBJECTIVE: Reveal the evolution of the magnetic resonance imaging (MRI) pattern in a patient with a posterior spinal artery infarction, which belongs to a subgroup of spinal cord ischemia syndromes and presents a rare cause of spinal cord injury. Our report underlines that diagnosis of spinal cord ischemia and thus clinical decision making remains challenging. SETTING: University Hospital of Innsbruck and University Hospital of Salzburg, Austria. METHODS: Here we present clinical, electrophysiological and imaging data in the acute, subacute and chronic phase of a woman who developed signs and symptoms related to a bilateral posterior spinal cord infarction. RESULTS: At the clinical nadir (24 h after symptom onset), MRI did not exhibit T2 hyperintensities. However, such MRI changes were detected 8 days after symptom onset and persisted until the latest follow-up at 5 months. CONCLUSIONS: Repeated MRI constitutes an indispensable diagnostic and follow-up tool for spinal cord ischemia. The imaging data in accordance with the electrophysiological measurements correlated well with the clinical presentation in the subacute und chronic phase. Therefore, further studies might allow using MRI following spinal cord ischemia as a prognostic marker for an individual outcome.

Ageing abolishes the effects of fluoxetine on neurogenesis.

Depression constitutes a widespread condition observed in elderly patients. Recently, it was found that several drugs employed in therapies against depression stimulate hippocampal neurogenesis in young rodents and nonhuman primates. As the rate of neurogenesis is dramatically reduced during ageing, we examined the influences of ageing on neurogenic actions of antidepressants. We tested the impact of fluoxetine, a broadly used antidepressant, on hippocampal neurogenesis in mice of three different age groups (100, 200 and over 400 days of age). Proliferation and survival rate of newly generated cells, as well as the percentage of cells that acquired a neuronal phenotype were analyzed in the hippocampus of mice that received fluoxetine daily in a chronic manner. Surprisingly, the action of fluoxetine on neurogenesis was decreasing as a function of age and was only significant in young animals. Hence, fluoxetine increased survival and the frequency of neuronal marker expression in newly generated cells of the hippocampus in the young adult group (that is 100 days of age) only. No significant effects on neurogenesis could be detected in fluoxetine-treated adult and elderly mice (200 and over 400 days of age). The data indicate that the action of fluoxetine on neurogenesis is highly dependent on the age of the treated individual. Although the function of neurogenesis in the clinical manifestation of depression is currently a matter of speculation, this study clearly shows that the therapeutic effects of antidepressants in elderly patients are not mediated by neurogenesis modulation.

Molecular mechanisms of neuronal migration disorders, quo vadis?

Following terminal mitosis, neuronal precursor cells leave their site of origin and migrate towards their definitive site of residency. In order to establish the intricate cytoarchitecture described in the adult human brain, neuronal migration must be finely regulated. In humans, brain malformations can result from neuronal migration defects. The spectrum of migration disorder severity extends from few heterotopic neurons, as observed in periventricular heterotopia, to a complete cortical disorganization, as observed in cases of lissencephaly. Recently, specific migration disorders have been linked to mutations/deletions in the doublecortin, filamin-1, LIS1 and reelin genes. These proteins act at different levels of the signaling cascades transducing extracellular guiding cues into cytoskeletal reorganization. Here, we summarize the data concerning these four molecules and speculate on their functions and interaction partners during neuronal development.

Extra neurofilament NF-L subunits rescue motor neuron disease caused by overexpression of the human NF-H gene in mice.

Previous studies demonstrated that transgenic mice overexpressing human neurofilament heavy (hNF-H) protein develop a progressive motor neuron disease characterized by the perikaryal accumulations of neurofilaments resembling those found in amyotrophic lateral sclerosis (ALS). To further investigate this neurofilament-induced pathology, we generated transgenic mice expressing, solely or concomitantly, the hNF-H and the human neurofilament light (hNF-L) proteins. We report here that the motor neuron disease caused by excess hNF-H proteins can be rescued by overexpression of hNF-L in a dosage-dependent fashion. In hNF-H transgenic mice, the additional hNF-L led to reduction of perikaryal swellings, relief of axonal transport defect and restoration of axonal radial growth. A gene delivery approach based on recombinant adenoviruses bearing the hNF-L gene also demonstrated the possibility to reduce perikaryal swellings after their formation in adult mice. The finding that extra NF-L can protect against NF-H-mediated pathogenesis is of potential importance for ALS, particularly for cases with NF-H abnormalities.

Transgenic mice in the study of ALS: the role of neurofilaments.

Amyotrophic lateral sclerosis (ALS) is an adult-onset neurological disorder of multiple etiologies that affects primarily motor neurons in the brain and spinal cord. Abnormal accumulations of neurofilaments (NFs) in motor neurons and a down-regulation of mRNA for the NF light subunit (NF-L) are associated with ALS, but it remains unclear to what extent these NF perturbations contribute to human disease. Transgenic mouse studies demonstrated that overexpression of normal and mutant NF proteins can sometimes provoke a motor neuronopathy characterized by the presence of abnormal NF accumulations resembling those found in ALS. Remarkably, the motor neuronopathy in transgenic mice overexpressing human NF heavy (NF-H) subunits was rescued by the co-expression of a human NF-L transgene at levels that restored a correct stoichiometry of NF-L to NF-H subunits. Transgenic approaches have also been used to investigate the role of NFs in disease caused by Cu/Zn superoxide dismutase (SOD1) mutations, which is responsible for approximately 2% cases of ALS. Studies with transgenic mice expressing low levels of a fusion NF-H/lacZ protein, in which NFs are withheld from the axonal compartment, suggested that axonal NFs are not toxic intermediates required for SOD1-mediated disease. On the contrary, overexpression of human NF-H proteins was found to confer an effective protection against mutant SOD1 toxicity in transgenic mice, a phenomenon that may be due to the ability of NF proteins to chelate calcium. In conclusion, transgenic studies showed that disorganized NFs can sometimes have noxious effects resulting in neuronopathy. However, in the context of motor neuron disease caused by mutant SOD1, there is emerging evidence that NF proteins rather play a protective role.

LC-MS/MS identification of doublecortin as abundant beta cell-selective protein discharged by damaged beta cells in vitro.

There is a clinical need for plasma tests that can directly detect injury to pancreatic beta cells in type 1 diabetes. Such tests require biomarkers that are abundantly and selectively released into plasma by damaged beta cells. We combined LC-MS/MS proteomics and tissue-comparative transcriptomics of FACS-purified beta cells for bottom-up identification of candidate markers. Less than 10% of 467 proteins detected in beta cells showed endocrine-enriched expression. One surprising candidate was the neuronal migration marker doublecortin: in situ analysis revealed uniform doublecortin expression in the cytoplasm of all beta cells. Western blotting and real-time PCR confirmed its strong beta cell-selectivity outside the brain and its high molar abundance, indicating promising biomarker properties in comparison to GAD65, a more established marker of beta cell injury. DCX potential was validated in vitro: chemically-induced necrosis of rat and human beta cells led to a discharge of intracellular doublecortin into the extracellular space, proportionate to the amount of injured cells, and similar to GAD65. In vivo, recombinant DCX showed favorable pharmacokinetic properties, with a half-life in plasma of around 3h. Combined, our findings provide first proof-of-principle for doublecortin as biomarker for beta cell injury in vitro, advocating its further validation as biomarker in vivo.

Extra axonal neurofilaments do not exacerbate disease caused by mutant Cu,Zn superoxide dismutase.

A recent report by T. L. Williamson et al. (1998, Proc. Natl. Acad. Sci. USA 95, 9631-9636) showed that disease caused by expression of mutant Cu,Zn superoxide dismutase (SOD1) in mice was slowed down by disruption of the neurofilament light (NF-L) gene. This led to the conclusion that decreasing the axonal amount of neurofilaments reduces the vulnerability of motor neurons to toxicity mediated by mutant SOD1. We report here that, unexpectedly, overexpression of human NF-L proteins resulting in extra axonal neurofilaments does not shorten the life span of transgenic mice expressing a mutant SOD1 (SOD1(G37R)). Microscopic examination of spinal cord and ventral roots even shows modest protective effects of NF-L overexpression. These results suggest that axonal neurofilaments are not an exacerbating factor in motor neuron disease mediated by mutant SOD1 and that perikaryal neurofilaments may even have beneficial effects.

Delayed maturation of regenerating myelinated axons in mice lacking neurofilaments.

Using the technique of homologous recombination in embryonic stem cells, we generated mice bearing a targeted disruption of the gene encoding the neurofilament light (NF-L) protein. The absence of NF-L protein in mice resulted in dramatic declines of approximately 20-fold in the levels of neurofilament medium and heavy proteins in the brain and sciatic nerve while increases were detected for other cytoskeletal proteins such as tubulin and GAP-43. Despite a lack of neurofilaments and hypotrophy of axons, the NF-L knockout mice develop normally and do not exhibit overt phenotypes. However, in both NF-L -/- and NF-L +/- mice, the regeneration of myelinated axons following crush injury of peripheral nerves was found to be abnormal. In the second week after axotomy, the number of newly regenerated myelinated axons in the sciatic nerve and facial nerve of NF-L -/- mice corresponded to only approximately 25 and approximately 5% of the number of myelinated axons found in normal mice, respectively. At this early postaxotomy stage, electron microscopy of nerve segments distal to the crush site in NF-L -/- mice revealed abundant clusters of axonal sprouts that were indicative of retarded maturation of regenerating fibers. The analysis of the distal sciatic nerve at 2 months after crush indicated that neurofilament-deficient axons have the capacity to regrow for a long distance and to remyelinate, albeit at a slower rate. These results provide the first direct evidence for a role of neurofilaments in the maturation of regenerating myelinated axons.

Disruption of type IV intermediate filament network in mice lacking the neurofilament medium and heavy subunits.

To clarify the role of the neurofilament (NF) medium (NF-M) and heavy (NF-H) subunits, we generated mice with targeted disruption of both NF-M and NF-H genes. The absence of the NF-M subunit resulted in a two- to threefold reduction in the caliber of large myelinated axons, whereas the lack of NF-H subunits had little effect on the radial growth of motor axons. In NF-M-/- mice, the velocity of axonal transport of NF light (NF-L) and NF-H proteins was increased by about two-fold, whereas the steady-state levels of assembled NF-L were reduced. Although the NF-M or NF-H subunits are each dispensable for the formation of intermediate filaments, the absence of both subunits in double NF-M; NF-H knockout mice led to a scarcity of intermediate filament structures in axons and to a marked approximately twofold increase in the number of microtubules. Protein analysis indicated that the levels of NF-L and alpha-internexin proteins were reduced dramatically throughout the nervous system. Immunohistochemistry of spinal cord from the NF-M-/-;NF-H-/- mice revealed enhanced NF-L staining in the perikaryon of motor neurons but a weak NF-L staining in axons. In addition, axonal transport studies carried out by the injection of [35S]methionine into spinal cord revealed after 30 days very low levels of newly synthesized NF-L proteins in the sciatic nerve of NF-M-/-;NF-H-/- mice. The combined results demonstrate a requirement of the high-molecular-weight subunits for the assembly of type IV intermediate filament proteins and for the efficient translocation of NF-L proteins into the axonal compartment.

Protective effect of neurofilament heavy gene overexpression in motor neuron disease induced by mutant superoxide dismutase.

To investigate the role of neurofilaments in motor neuron disease caused by superoxide dismutase (SOD1) mutations, transgenic mice expressing a amyotrophic lateral sclerosis-linked SOD1 mutant (SOD1(G37R)) were mated with transgenic mice expressing human neurofilament heavy (NF-H) subunits. Unexpectedly, expression of human NF-H transgenes increased by up to 65%, the mean lifespan of SOD1(G37R) mice. Microscopic examination corroborated the protective effect of NF-H protein against SOD1 toxicity. Although massive neurodegeneration occurred in 1-yr-old mice expressing SOD1(G37R) alone, spinal root axons and motor neurons were remarkably spared in doubly SOD1(G37R);NF-H-transgenic littermates.

Somatic mosaicism and variable penetrance in doublecortin-associated migration disorders.

X-linked isolated lissencephaly sequence (XLIS) and subcortical band heterotopia (SBH) are allelic disorders caused by mutations in the doublecortin (DCX) gene. This genetic analysis of seven families revealed four novel mutations in the DCX gene. The authors detected a high rate of somatic mosaicism in male and female patients with variable penetrance of bilateral SBH including nonpenetrance in a heterozygous woman. In addition, the authors implemented prenatal diagnosis in a family with SBH/XLIS.