Molecular Mechanisms of Neuronal Apoptosis in EAE

Neurodegenerative diseases are often characterized by apoptosis of certain neuronal populations and concomitant intracellular protein depositions. In ALS, cell death of upper and lower motor neurons is associated with inclusion of proteins such as skein-like ubiquitin. In animal models of Huntington's disease, as well as in brain tissue from Huntington patients, neuronal cell death in the striatum related to intranuclear deposition of mutant huntingtin and ubiquitin was observed. Cell death in Parkinson's disease pre dominantly affects neurons of the substantia nigra and other subcortical nuclei that show so-called Lewy bodies in their cytoplasm (for review, see Jellinger, 2001). Distinct from those purely neurodegenerative disorders, neuronal cell death in MS or EAE does not show any clear distributional preferences. In human brain tissue, apoptotic neurons were identified in cerebral cortex lesions (Peterson et al., 2001), whereas apoptosis in different EAE models affected neuronal subpopulations such as RGCs or spinal cord neurons (Meyer et al., 2001; Diem et al., 2003b; Ahmed et al., 2002). Cytoskeleton changes or accumulation of abnormal protein during MS- or EAE-induced neuronal apoptosis are rather described for damaged axons than for the neuronal cell body itself (Kornek et al., 2000; Zhu et al., 1999). Despite these obvious differences resulting from distinct pathogenesis or pathophysiology of neurodegeneration, similar molecular mechanisms are involved because the subsequent biochemical events that execute apoptosis are highly conserved. These alterations in intracellular signal transduction occurring during early stages of neuronal apoptosis in EAE involve the Bcl-2 family of proteins. As described in context with the molecular features of developmental apoptotic neuronal cell death, this protein family includes both proapoptic and antiapoptotic members. In the rat model of rrMOG-induced EAE, a shift in the Bcl-2 family of proteins toward the proapoptotic side was observed during induction and manifestation of the disease (Hobom et al., 2004). At disease onset, RGCs showed an increase of the proapoptotic protein Bax simultaneously with a reduction of the antiapoptotic Bcl-2 protein. Figure 4 gives examples of representative retinal sections with increased protein expression of Bax in RGCs during development and manifestation of MOG-EAE.

In this model, Bcl-2 levels re-increased at day 8 of the disease when the rate of RGC death declined as shown in Fig. 5. Apoptosis of RGCs after mechanical lesion of the ON was accompanied by a reciprocal regulation of Bax and Bcl-2 as well (Isenmann et al., 1997), indicating that not only the kinetics but also the mechanisms of apoptotic RGC death are similar after autoimmune transection or surgical axo-tomy of the ON. From in vitro studies, it is known that high amounts of Bax can antagonize the antiapoptotic activity of Bcl-2 (Oltvai et al., 1993) such that the ratio of Bcl-2 to Bax determines survival or death of neurons after an apoptotic stimulus. The functional relevance of Bcl-2 for EAE-associ-ated neurodegeneration was demonstrated in a study using transgenic mice overexpressing the Bcl-2 gene, which showed reduced axonal damage and a less severe disease course during MOG-EAE (Offen et al., 2000).

The phosphatidylinositol 3-kinase (PI3-K)/Akt pathway has originally been described as a signal transduction step involved in growth-factor-induced neuroprotection against different apoptotic stimuli in vitro as well as in vivo (Dudek et al., 1997; Kermer et al., 2000). In addition, activation of this pathway after application of pro-inflammatory

0 1000 2000 3000 Cell Density (mm"?)

Figure 3 Correlation between retinal ganglion cell (RGC) density and visual function in rats with acute experimental autoimmune encephalomyelitis (EAE) induced by myelin oligodendrocyte glycoprotein (MOG). (A) Normal cell density of RGCs in a healthy control rat. (B) The numbers of remaining RGCs are significantly reduced in a rat with acute optic neuritis during MOG-EAE. (C) The decrease of RGC density in rats with optic neuritis correlates with a reduction of visual acuity determined by measurements of electroretinograms. Filled circles indicate visual acuities and RGC densities of rats with histopathologically proven optic neuritis, whereas filled triangles represent those of healthy controls. Visual acuity values and RGC counts of MOG-immunized rats without optic neuritis are given as open squares. Note that RGC densities in animals suffering from optic neuritis must be markedly decreased to lead to a reduction of visual acuity. (Reprinted from Meyer et al., 2001.)

t iC"

Figure 4 Increased expression of the pro-apoptotic protein Bax in retinal ganglion cells (RGCs) during development and manifestation of myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE) in rats. (A, B) Double staining of a retinal section from a healthy rat. No specific staining of Bax protein (A) was detected in RGCs identified by retrograde labeling with the fluorescent dye Fluorogold (FG) (B). (C, D) Retinal section from a rat at day 7 after immunization with MOG. A significant upregulation of Bax protein expression (C) was detected in some of the FG-positive RGCs indicated by the vertical arrow (D). Horizontal arrows indicate Bax-negative RGCs. (E, F) On day 1 of the disease, high expression levels of Bax protein in all RGCs identified by FG-labeling were discovered. (G, H) The amount of Bax protein in RGCs on day 8 of the disease was similar when compared to day 1 of MOG-EAE: 50mm. (Reprinted from Hobom et al., 2004.)

cytokines such as tumor necrosis factor-a or interleukin-1ß led to rescue of RGCs after surgical axotomy of the ON (Diem et al., 2001; 2003a). In general, interaction of neu-rotrophins or cytokines with their highly specific receptors activates PI3-K, which generates various phosphorylated phosphatidylinositides. These second messengers lead to activation of protein kinase B, also called Akt (Coffer et al., 1998). Phospho-Akt, in turn, can phosphorylate and thereby inactivate the proapoptotic protein Bad (Datta et al., 1997), as well as unprocessed or active caspase-9 (Cardone et al., 1998), which leads to decreased levels of the downstream effecter caspase-3 (Li et al., 1997). In MOG-EAE, phospho-Akt levels in RGCs were decreased at day 1 of the disease as shown in Fig. 5a, which corresponds to the time point when the most prominent RGC apoptosis occurred, as well as cas-pase-3 activation in RGCs peaked (Hobom et al., 2004). Simultaneously with the re-increase of Bcl-2 as described above, protein concentration of phospho-Akt normalized at day 8 of MOG-EAE, indicating a temporary breakdown of intracellular endogenous rescue mechanisms in neurons exposed to EAE-associated proapoptotic stimuli. Figure 5 compares the kinetics of phospho-Akt and Bcl-2 protein expression in RGCs during development, manifestation, and disease progression of MOG-EAE in BN rats.

Another pathway characterized in context with neu-rotrophin-induced neuroprotection involves phosphoryla-tion of mitogen-activated protein kinases (MAPKs). MAPKs regulate neuronal cell growth and morphological differentiation and promote neuronal survival after neuro-transmitter or neurotrophic factor stimulation (for review, see Fukunaga and Miyamoto; 1998). Multiple second messengers, such as cyclin adenosine monophosphate, protein kinase A, or calcium, control MAPK signaling via a small G protein called "Ras" (for review, see Grewal et al., 1999). MAPK phoshorylation during the time period of acute neuronal apoptosis in MOG-EAE may serve as an endogenous rescue mechanism (Diem et al., 2003b), as suggested for other neuronal cell types, which upregulate the phosphory-lated, active form of MAPKs during exposure to chronic stress, brain injury, or development of neurodegenerative disorders (Ferrer et al., 2001; Dash et al., 2002; Trentani et al., 2002). In rrMOG-induced EAE in BN rats, it has been shown recently that treatment with methylprednisolone, the standard therapy of acute MS relapses, increased apoptosis of RGCs secondary to severe optic neuritis by inhibiting the phosphorylation of MAPKs (Diem et al., 2003b). This steroid effect was mimicked by treatment with a pharmacological inhibitor of MAPK kinase (MEK), which in turn phosphorylates and thereby activates MAPKs. In a study on cortical neurons under hypoxic conditions, it has been demonstrated that inhibition of MAPK phosphorylation by MEK blockade reduced the rate of surviving neurons via mechanisms such as decreasing the ability to phosphorylate and thereby inactivate the proapoptotic protein Bad (Jin et al., 2002). The negative steroid effect on RGC survival during MOG-induced EAE was further mediated by calcium influx via voltage-gated calcium channels (Diem et al., 2003b), indicating a possible involvement of transmembrane ion currents in EAE-associated neurodegeneration as well.

Increased calcium influx into neurons after activation of AMPA (a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid)/kainate receptors may be another important mechanism for neuronal degeneration in MS and EAE. These receptors mediate toxicity induced by the excitatory neuro-transmitter glutamate, which has been shown to contribute to neuronal apoptosis in neurodegenerative disorders such as Alzheimer's disease, Huntington's disease, and ALS (for review, see Mattson, 2000a). Overactivation of glutamate

Figure 5 Western blot analysis of the anti-apoptotic proteins phospho-Akt (pAkt) (A) and Bcl-2 (C) in retinal ganglion cells during development and manifestation of myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE) in rats. On days 1 and 2 of MOG-EAE, phospho-Akt and Bcl-2 expression decreased to almost no detectable protein levels. The expression level of phospho-Akt and Bcl-2 re-increased on day 8 of the disease. As a control, protein concentrations of the inactive, unphosphorylated form of the Akt protein were not regulated (B). (Reprinted from Hobom et al., 2004.)

receptors, especially under conditions of decreased energy availability and oxidative stress, has also been proposed for pathophysiological conditions during stroke and epileptic seizures (for review, see Mody and MacDonald, 1995). In a rat model of EAE induced by immunization with myelin basic protein (MBP), it has been shown that administration of different AMPA/kainate receptor antagonists improved neurological outcome and limited apoptotic cell death of spinal cord neurons (Smith et al., 2000). Thereby it could be demonstrated that the beneficial effect of these drugs cannot be attributed to possible anti-inflammatory or immunomod-ulatory actions, as inflammatory histopatholological changes were not influenced. The relevance of glutamate toxicity for the human disease has also been suggested by the observation of increased glutamate levels in the CSF of patients during acute MS relapse (Stover et al., 1997). Histopathological analysis of human brain and spinal cord tissue obtained at autopsy revealed a high expression of glutaminase, a marker for glutamate production, in active MS lesions. These elevated glutaminase levels were detected in macrophages and microglia in close proximity to damaged axons (Werner et al., 2001). In an adoptive transfer model in mice induced by injection of MBP-reactive T-cells, gluta-mine synthetase and glutamate dehydrogenase activity were downregulated in astrocytes during acute EAE, resulting in decreased ability to metabolize the excitatory neurotrans-mitter glutamate (Hardin-Pouzet et al., 1997). Glutamate targets not only neuronal cell types during autoimmune CNS inflammation but also oligodendrocytes that express the AMPA/kainate type of excitatory glutamate receptors are subjected to excitotoxicity. Increased oligodendrocyte survival was seen under treatment with the AMPA/kainate antagonist NBQX in adoptive transfer EAE in mice (Pitt et al., 2000), suggesting an involvement of glutamate toxic-ity in the pathophysiology of at least two important histopathological features of MS or EAE.

C. Role of Inflammation, Demyelination, Axonal Pathology, and Electrophysiological Dysfunction in the Pathogenesis of Neuronal Apoptosis

Currently, investigations of EAE- and MS-associated neurodegeneration devote much effort to understanding the nature of neuronal cell death in chronic inflammatory autoimmune CNS diseases. Degeneration of the axon may precede the death of the cell body and make a more important contribution to the development of chronic disability with consecutive conversion into the secondary progressive form of the disease in patients with MS. Transection and swelling of axons, changes in the axonal cytoskeleton, and impaired axonal transport have been described in acute MS (Ferguson et al., 1997; Trapp et al., 1998) and EAE lesions (Kornek et al., 2000; Meyer et al., 2001) and might induce secondary apoptotic cell death of the neuron itself. Secondary or retrograde neuronal apoptosis after surgical axotomy or crush injury of the axon is a well-characterized process in different animal models (Isenmann et al., 1997; Diem et al., 2001; Ugolini et al., 2003), which shows similarities in kinetics, extent, and mechanisms when compared with neuronal cell death in EAE (Hobom et al., 2004). Apoptotic cell death of CNS neurons in turn can lead to a reduced number of remaining projection targets for other neuronal subpopulations and thereby induce further cell death as a result of target deprivation (Theoret et al., 1997). However, apoptotic neuronal cell death during chronic inflammatory autoimmune CNS disease might have additional triggers unrelated to axonal damage.

Comparing the development of axonal injury of the ON with the time kinetics of RGC apoptosis in the rat model of MOG-induced EAE revealed that apoptosis of neurons can precede degeneration of their respective axons (Hobom et al., 2004). In this animal model, it has been demonstrated that electrophysiological dysfunction of the ON occurs simultaneously with the onset of RGC apoptosis and before histopathological abnormalities of the ON can be detected. Functional integrity can be an important survival factor for neuronal cell types as demonstrated for purified RGCs (Meyer-Franke et al., 1998), as well as for hippocampal neurons (Neumann et al., 1997). Neurons with impaired elec-trophysiological function might become a target for the action of cytotoxic T-cells due to cell surface expression of P2-microglobulin and major histocompatibility complex class I molecules, whereas in electrically active neurons, expression of these proteins was suppressed (Neumann et al., 1997). Furthermore, it has been demonstrated that depolarization of purified RGCs increased cell surface levels of the neurotrophic receptor TrkB and thereby improved their trophic responsiveness and cell survival (Meyer-Franke et al., 1998; Shen et al., 1999). Depolarization was supposed to increase intracellular Ca2+ levels with a consecutive augmentation of intracellular cAMP concentration, which in turn promotes transport of TrkB to the cell surface. Also in the intact retina, physiological levels of electrical activity enhance the responsiveness of RGCs to peptide trophic factor stimulation (Shen et al., 1999). In MS or EAE, impairment of electrophysiological function of axons even before demyelination is detectable might be caused by elevated levels of pro-inflammatory cytokines such as tumor necrosis factor-a or interleukin-1p. For these factors, cytokine-induced alterations of voltage-gated ion channel function have been demonstrated in a rat model of ON transection (Diem et al., 2001; 2003a). For patients with MS, it has been shown that an increase of circulating pro-inflammatory cytokines in the blood after therapeutically evoked lym-phopenia led to a reversible exacerbation of neurological symptoms for several hours, a time period certainly too short for demyelination and remyelination (Moreau et al., 1996). The authors suggested that this reawakening of preexisting symptoms might be due to alterations in axonal conduction induced by direct electrophysiological effects of pro-inflammatory cytokines.

Evidence for a direct attack of autoreactive lymphocytes as the main cause for apoptotic neuronal cell death in EAE came from electron microscopy of motoneurons of the lumbar spinal cord in rats immunized with MBP (Smith et al., 2000). At the clinical peak of the disease course, motoneu-ron death in the ventral horn of the lumbar spine was associated with lymphocyte entry. Sequestration of lymphocytes into motoneurons led to the development of neuronal vacuoles filled with cellular debris and consecutive reduction of neuronal density in that area. Immunohistochemistry revealed that these lymphocytes entering motoneurons during the acute phase of EAE were T-lymphocytes positive for CD2, whereas an additional involvement of B- or natural killer lymphocytes, or macrophages/microglia could be excluded. In contrast, acute RGC apoptosis in MOG-EAE in BN rats does not seem to be mediated by direct autoreactive lymphocyte action. In this model, inflammatory infiltrates in the retina consisting of T-cells or macrophages as an explanation for the degeneration of RGCs could not be detected by immunohistochemistry against T-cell receptors or as against lysosomal membrane-related antigens on macrophages and microglia (Meyer et al., 2001). Different from MOG-induced EAE, inflammatory infiltrates in the anterior segments of the eye leading to acute iridocyclitis or anterior uveitis together with optic neuritis have been described during acute MBP-EAE (Hayreh, 1981; Verhagen et al., 1994).

The concept of benign or neuroprotective effects of inflammatory infiltrates on neuronal survival as an opposite point of view has been suggested by a study using a crush injury model of the ON (Moalem et al., 1999). In this work, it has been shown that autoreactive T-lymphocytes specific for MBP reduced secondary neuronal cell death of RGCs because of a transient reduction of energy requirement caused by a decrease in nerve activity. Active immunization with encephalitogenic or nonencephalitogenic peptides of proteolipid protein or MOG can also exert neuroprotective effects on RGCs after ON crush injury (Fisher et al., 2001). In a different injury model based on avulsion of ventral roots in the rat lumbar spinal cord, it has been demonstrated that apoptotic cell death of motoneurons was reduced in rats suffering from MBP-induced EAE (Hammarberg et al., 2000). The mechanisms by which transferred T-cells or active immunization promote neuronal survival are thought to depend on the secretion of neurotrophic factors by cells of the immune system. High levels of BDNF, NT-3, and other glial cell line-derived neurotrophic factors were detected in T-cells and natural killer cells in the spinal cord from animals with MBP-EAE and simultaneous axotomy of spinal motoneurons (Hammarberg et al., 2000). Increased production of bioactive BDNF on antigen stimulation was shown in a study on T-helper (Th)1- and Th2-type CD4+ T-cell lines specific for myelin autoantigens such as MBP or MOG (Kerschensteiner et al., 1999). In this study, immune-cell-derived BDNF was demonstrated to support the survival of sensory neurons in vitro. BDNF immunoreactivity was also detected in inflammatory cells in lesional areas of the brain from patients with MS (Kerschensteiner et al., 1999). Although specific autoimmunity in the CNS can have a physiological neuroprotective role, the gap between benign and destructive effects of an autoreactive immune response might be small. The neuroprotective effect after immunization with immunodominant encephalitogenic epitopes, for example, could be achieved only in cases where the disease they caused was mild (Fisher et al., 2001).

Whereas the interaction between axonal and oligoden-drocyte integrity is a frequently investigated phenomenon, little is known about the direct impact of demyelination on neuronal apoptosis during autoimmune CNS inflammation. In MS and EAE where axonal damage could be the consequence of demyelination, it is difficult to differentiate between the pathogenetic relevance of these two histopatho-logical features for the induction of neuronal apoptosis. Transection of dysmyelinated ON axons in rats lacking MBP did not affect the survival rate of RGCs for time periods of up to 180 days (Phokeo and Ball, 2000), indicating that the absence of myelin did not accelerate neuronal death in this model. In a demyelinating mouse model of the human globoid cell leukodystrophy, neuronal apoptosis was reduced by lipocalin-type prostaglandin produced by perineuronal oligodendrocytes (Taniike et al., 2002), suggesting that these oligodendrocytes might be able to directly contribute to the survival of neuronal cells. At least in vitro, a reciprocal survival-promoting interaction between these two cell types exists, as it has been shown that purified sensory neurons can increase the survival of oligodendrocytes in culture (for review, see Raff et al., 1993).

How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book


Post a comment