C79 EMERGING ROLE FOR ASTROCYTES IN THE PATHOGENESIS OF ALS
Barbeito L
Institut Pasteur de Montevideo, Montevideo, Uruguay
E‐mail address for correspondence: [email protected]
The pathogenesis of ALS is likely to involve disturbed intercellular communications between motor neurons and other cell types. Fibroblast growth factor‐1 (FGF) can contribute to induce the sustained astrocyte activation characteristic of ALS. FGF‐1 is highly expressed in motor neurons. It can be released from cells during oxidative stress, which might occur from SOD‐1 aberrant function. FGF‐1 potently stimulates NGF expression and release in astrocytes. NGF in turn induces apoptosis in motor neurons expressing the p75 neurotrophin receptor (p75NTR) by a mechanism involving nitric oxide (NO) and peroxynitrite formation. In addition, FGF‐1 increased the expression of inducible nitric oxide synthase and NO production in astrocytes, making adjacent motor neurons vulnerable to NGF‐induced apoptosis. Since p75NTR is re‐expressed only by affected motor neurons in ALS, it is proposed that astrocytic NGF or high molecular weight NGF species, may serve to eliminate damaged neurons. Spinal cord astrocytes isolated from transgenic SOD1G93A rats displayed increased NO production and spontaneously induced apoptosis of co‐cultured motor neurons.
FGF‐1 also activates the redox‐sensitive transcription factor nuclear factor erythroid 2‐related factor 2 (Nrf2) in astrocytes. Accordingly, both Nrf2 and Heme oxygenase‐1 levels were increased and colocalized with reactive astrocytes in the degenerating lumbar spinal cord of rats expressing the ALS‐linked SOD1 G93A mutation. Because Nrf2 increases glutathione (GSH) biosynthesis, we investigated the role of GSH production by astrocytes on p75NTR‐dependent motor neuron apoptosis. The combined treatment of astrocytes with FGF‐1 and t‐butylhydroquinone (tBHQ) increased GSH production and secretion, preventing motor neuron apoptosis. Activation or overexpression of Nrf2 in astrocytes expressing the SOD‐1 mutations completely prevented motor neuron apoptosis in co‐cultures and prevented p75‐dependent apoptosis. We hypothesizes that SOD‐1 mutations linked to ALS sensitize astrocytes to FGF‐induced oxidative damage, leading to a decrease in the neurotrophic and antioxidant support for motor neurons.
C80 GLUTAMATE EXCITOTOXICITY IN ALS: THE SEARCH FOR PROTEIN INTERACTORS OF EAAT2
Dong YY, Goodall EF, Ahmed Z, Adu J, Douglas MR, Morrison KE
University of Birmingham, Birmingham, UK
E‐mail address for correspondence: [email protected]
Background: Many theories for the pathogenesis of amyotrophic lateral sclerosis (ALS) have been proposed including that of glutamate mediated excitotoxicity. Glutamate is the major excitatory neurotransmitter in the mammalian CNS and in excess can be toxic to neuronal cells. Glutamate levels are tightly controlled by glutamate transport proteins, the main functional transporter being the astrocytic glutamate transporter EAAT2. Evidence for glutamate excitotoxicity in ALS comes from reports of raised levels of glutamate in the cerebrospinal fluid and the finding of reduced levels of EAAT2 in the CNS of ALS patients. The cause of reduced astrocytic expression of EAAT2 in ALS patients is unknown. Investigations into the mRNA expression and genomic sequence of EAAT2 have found no differences between patients and controls. An alternative theory is that proteins interacting with EAAT2 are abnormal in ALS and it is the abnormalities in these proteins that lead to impaired EAAT2 function and glutamate excitotoxicity in the disease. No proteins that interact with EAAT2 have as yet been reported in humans, while only a single protein, Ajuba, a group 3 LIM protein, has been reported to interact with glt‐1, the homologue of EAAT2, in rats. We are investigating this interaction in humans and have undertaken a yeast two hybrid screen to identify additional interacting proteins.
Objectives: To identify proteins that interact with EAAT2 and confirm the EAAT2/Ajuba interaction in human cells. To investigate whether alterations in any of these interacting proteins are responsible for the reduced expression or altered function of EAAT2 in ALS.
Methods: We used a yeast two‐hybrid system to identify proteins that interact with EAAT2. A C‐terminal, predicted cytoplasmic fragment of EAAT2 was used as bait to screen a pretransformed adult human brain cDNA library (Clontech). The full coding sequences of the strongest potential positives from the yeast screen, Ajuba and EAAT2 were cloned into mammalian expression vectors with epitope tags. Transfected cells were visualized using immunocytochemistry and the tagged proteins extracted and used in co‐immunoprecipitation experiments.
Results: We have identified 54 proteins that interact with our EAAT2 bait in a yeast two hybrid system. Co‐immunoprecipitation and colocalization experiments to date show that three of the potential positives and Ajuba each colocalise and interact with EAAT2 in mammalian cells.
Discussion and conclusions: We have discovered three proteins that interact with EAAT2 and confirmed the EAAT2/Ajuba interaction in mammalian cells. The next challenge is to investigate if and how these interactions might influence EAAT2 function and elucidate whether changes in any of these proteins have relevance in ALS pathogenesis.
C81 ONSET AND PROGRESSION IN INHERITED ALS DETERMINED BY MOTOR NEURONS AND MICROGLIA
Yamanaka K1, Boillee S2, Lobsiger CS2, Kassiotis G3, Kollias G3, Cleveland DW2
1RIKEN Brain Science Institute, Wako, Saitama, Japan, 2University of California, San Diego and Ludwig Institute, La Jolla, CA, USA, 3Biomedical Sciences Research Center Al. Fleming, Vari, Greece
E‐mail address for correspondence: [email protected]
Background: Dominant mutations in the ubiquitously expressed Cu/Zn superoxide dismutase (SOD1) lead to amyotrophic lateral sclerosis (ALS), a neurodegenerative disease affecting adult motor neurons. Although ubiquitous expression of mutant SOD1 provokes progressive, selective motor neuron degeneration in human and rodents due to an acquired toxic property(ies) of the mutant, the cell types that contribute to the onset and progression of the motor neuron disease are not known.
Objectives: To determine the cell types in which mutant SOD1 acts to generate toxicity to the motor neurons in ALS.
Methods: We have generated a mouse ubiquitously expressing a ‘floxed’ mutant SOD1G37R transgene (LoxSOD1G37R) which can be removed within specific cell populations by the action of Cre recombinase. To delete mutant SOD1 within motor neurons or microglia/macrophages, LoxSOD1G37R mice were mated with Islet1‐Cre or CD11b‐Cre mice, respectively. The disease onset, duration, and survival of these cohorts were monitored.
Results: Removing mutant SOD1 from motor neurons extends the age of disease onset, and slows the early phase of disease progression, indicating mutant action in neurons as an initiating factor in triggering disease. More importantly, silencing of SOD1 mutant expression selectively within microglial cells and peripheral macrophages has minimal effect on age of disease onset, but dramatically slows disease progression (extending survival after onset by 75 days).
Discussion and conclusions: Microglia and/or macrophages play a central role as a determinant of the timing of disease progression in ALS. Mutant toxicity within microglia and/or macrophages leads to the non‐cell autonomous killing of motor neurons, following the disease initiation caused by the toxicity within motor neurons. Contributions of other non‐neuronal cell types, including astrocytes and muscles are being tested.
C82 GENE EXPRESSION CHANGES IN MICROGLIAL CELLS CARRYING THE HUMAN MUTANT SOD1 GENE
Blackburn DJ, Sargsyan SA, Monk PN, Shaw PJ
University of Sheffield, Sheffield, UK
E‐mail address for correspondence: [email protected]
Background: Microglia are often referred to as the macrophages of the brain. They play an active role in surveillance of CNS tissue homeostasis and respond to perturbations with acquisition of a more activated phenotype. Reactive microglia are seen in inflammatory, infective and neurodegenerative conditions. They are present in affected tissues in human cases of ALS and in mutant SOD1 transgenic mice. The hypothesis of non‐cell autonomous damage to motor neurons in ALS proposes that altered properties of glial cells, carrying the mutant SOD1 gene, could be responsible for the toxicity towards or failure to support the neighbouring motor neurons.
Objectives: Our objective is to investigate the effects of mutant SOD1 on the microglial transcriptome by comparing the mRNA profile of non‐transgenic and mutant SOD1 G93A transgenic microglial cells isolated from animal models of the disease.
Methods: We established a protocol to obtain highly purified microglial cultures from non‐transgenic and human SOD1 G93A transgenic neonatal mice. The purity of cultures was determined by immunocytochemistry using a macrophage specific marker F4/80. The mRNA of the purified microglial cells was used in rounds of amplification, in vitro transcription, labelling, fragmentation and hybridized onto the mouse genome 430 2.0 microarray using Affymetrix technology. Pathway Architect software was used to pull out the key affected pathways in the transgenic microglial cells. Differences in gene expression identified in this analysis have been validated using quantitative RT‐PCR, measurement of protein expression and relevant functional assays.
Results: With respect to non‐transgenic cells, the transgenic microglia showed 61 differentially expressed genes, most of which were down‐regulated. The key pathways affected in the transgenic cells appeared to be those related to the actin cytoskeleton, cellular adhesion, phagocytosis and motility and genes involved in the calcium signalling system. Preliminary validation assays confirm the microarray results at both the mRNA and protein level.
Discussion and conclusion: In keeping with other gene expression profiling studies investigating the cellular pathways affected by the presence of mutant SOD1, our data show a predominant down‐regulation of differentially expressed genes. The effects of human SOD1 G93A on the microglial transcriptome provide novel and interesting insights into which cellular properties might be altered in transgenic cells to hinder their neuroprotective function. The inability of microglial cells to recognize and support damaged motor neurons may prove essential for ALS pathology.
C83 THE ‘DOUBLE‐EDGED SWORD’ EFFECT OF MICROGLIAL ACTIVATION IN SOD1G93A ALS MICE NULL FOR CX3CR1: SPINAL CORD AND CORTICAL PATHOLOGY
Kostenko V, Cardona A, Ransohoff RM, Pioro EP
Cleveland Clinic, Cleveland, OH, USA
E‐mail address for correspondence: [email protected]
Background: Motor neuron (MN) loss in ALS is believed to be influenced by microglial mediated inflammation. Activation of microglia is normally suppressed by neuronally produced fractalkine (CX3CL1) acting via its receptor (CX3CR1) found exclusively on microglia. Our preliminary findings indicate that knocking out CX3CR1 in transgenic (Tg) mice expressing the mutated human SOD1G93A gene may initially improve MN function but ultimately worsens MN loss in the lumbar spinal cord, resulting in more rapid limb weakness, and early death.
Objectives: 1) To confirm our preliminary findings that SOD1G93A mice without fractalkine receptor (CX3CR1−/−) experience more microglial activation, greater MN loss, and more rapid disease progression; 2) To examine whether microglia can be neuroprotective in early disease; 3) To determine if onset of disease is earlier in SOD1G93A‐CX3CR1−/− mice; 4) To determine if microglial neurotoxicity is enhanced in cortex – all relative to SOD1G93A mice with one or two copies of CX3CR1 (CX3CR1+/−, CX3CR1+/+).
Methods: We studied a colony of mice that were SOD1G93A‐CX3CR1−/−, SOD1G93A‐CX3CR1+/−, SOD1G93A‐CX3CR1+/+, and non‐Tg SOD1G93A‐CX3CR1+/−, ‐CX3CR1−/−, with serial behavioral and survival assessments, and histological analysis of spinal cord and cortex in terminal mice at 18–20 weeks.
Results: After 7–9 weeks, limb strength was lower in SOD1G93A mice compared to non‐Tg SOD1G93A‐CX3CR1−/− littermate controls, and from 15 to 20 wks the decline was more rapid in SOD1G93A‐CX3CR1−/− mice for hindlimb (p<0.01) and forelimb (p = 0.02) grip strength compared to SOD1G93A‐CX3CR1+/− mice. The male CX3CR1−/− group had the largest decline and accounted for the group differences for hindlimb (−19.9±2.1, estimated slope ± SE) and forelimb strength (−23.6±3.8) compared to all other groups. However, between 14 and 15 weeks, SOD1G93A‐CX3CR1‐/‐ mice were significantly stronger. Disease onset was slightly sooner and survival was shorter in SOD1G93A‐CX3CR1−/− mice (18.4 weeks) compared to SOD1G93A‐CX3CR1+/− mice (20.1 weeks, p = 0.01). Microglial reaction was increased, Nissl‐stained lumbar cord MNs were reduced, and neuronal injury was greater in the cortex of SOD1G93A‐CX3CR1−/− mice compared with SOD1G93A‐CX3CR1+/− mice.
Discussion and conclusions: Absence of the CX3CR1 in SOD1G93A mice results in microglial activation which initially enhances MN function. This neuroprotective effect is transient and eventually gives way to worsened neurobehavioral outcomes, greater MN loss, and shorter survival, especially in males. Onset of disease also appears slightly earlier in CX3CR1 null mice. Exaggeration of cortical pathology with degenerating motor projection neurons indicates an extra‐spinal effect of microglial neurotoxicity. Ongoing studies at pre‐terminal and terminal ages are exploring downstream mediators of microglial neuroprotection and neurotoxicity in an effort to develop pharmacotherapies for ALS that would enhance the former and suppress the latter for improved MN survival and function.
Acknowledgement: Funding was provided by the Robert Packard Center for ALS Research at Johns Hopkins University.
C84 PROTECTION BY WILD‐TYPE IMMUNE CELLS: A CENTRAL AND PERIPHERAL EVENT AND THEIR THERAPEUTIC POTENTIAL
Beers DR, Henkel JS, Zhao W, Xiao Q, Appel SH
Methodist Neurological Institute, Houston, USA
E‐mail address for correspondence: [email protected]
Background: Dominant mutations in superoxide dismutase (SOD1) cause amyotrophic lateral sclerosis, a progressive, paralytic disease characterized by loss of motor neurons. To address the question of whether central nervous system microglia and the peripheral immune system are involved in this disease, we utilized PU.1−/− mice, which are unable to develop myeloid and lymphoid cells. Wild‐type bone marrow transplants (BMT) into mSOD1/PU.1−/− mice repopulated the central and peripheral immune systems, slowed the loss of motor neurons, prolonged survival, and increased disease duration by 40%. These data demonstrate the importance of the immune system in mediating motor neuron protection.
Objectives: As the results with mSOD1/PU.1−/− mice may have been due to the central and/or peripheral immune systems, we determined whether wild‐type microglia, peripheral immune cells, or both, were responsible for the observed neuroprotection.
Methods and results: Using primary motor neuron/microglia co‐cultures, mSOD1 microglia were found to be more neurotoxic than wild‐type microglia. When mSOD1 mice were irradiated followed by BMT from wild‐type or green fluorescent protein donor animals (which replaces the peripheral immune system, but only a few parenchymal immune cells) there was no protection. In contrast with our earlier study using mSOD1/PU.1−/− mice, when mSOD1/PU.1−/− mice were transplanted with CCR2−/−‐derived bone marrow, the receptor for MCP‐1 involved in immune cell recruitment, the protective effects of wild‐type microglia and immune cells were again lost. To determine the contribution of the peripheral immune system to disease, mSOD1 was expressed in the RAG2−/− mice, which lack functional T‐ and B‐cells. Compared with mSOD1/RAG2+/− mice, disease progression was significantly accelerated in mSOD1/RAG2−/− mice.
Discussion and conclusions: These data emphasize the importance of the neuroprotection provided by wild‐type BMT in mSOD1/PU.1−/− mice, which involves both the central and peripheral immune systems. The toxic effect of mSOD1 microglia in mice corresponds to toxic effects observed in culture. Furthermore, in the absence of functional T‐ and/or B‐cells, there is an enhanced toxicity, suggesting that T‐ and/or B‐cells may alter the state of microglial activation. The data using CCR2−/− bone marrow suggest that the lack of active recruitment of neuroprotective cells hastens motor neuron injury. These studies are in agreement with the optic (1) or facial nerve (2) injury models suggesting that neuroprotection involves the communication between central microglia and peripheral immune cells. Additionally, our data suggest that following irradiation, wild‐type or GFP BMT does not alter disease onset, survival, or duration in mSOD1 mice. Because wild‐type peripheral immune cells enter sites of CNS injury, these cells may be manipulated to enhance their neuroprotective attributes and possibly to be used therapeutically.
References:
- Shaked, et al. J Neuroimmunol. 2004;146:84–93.
- Byram, et al. J Neurosci. 2004;24:4333.
C85 OBSERVED ELEVATED CYTOKINES AND MYELOID / INTERFERON‐INDUCED (MIFN) SIGNATURE PROTEIN LEVELS IN ALS: A POSSIBLE ALS EARLY DETECTION METHOD
Lancero HL1, Do H1, Gascon R2, Lancero MG2, Narvaez A2, Miller RG3, Mass J3, Katz J3, McGrath M2, Hadlock KG1
1Pathologica, LLC, Burlingame, CA, USA, 2University of CA, San Francisco/ San Francisco General Hospital, AIDS and Cancer Specimen Resource (ACSR), San Francisco, CA, USA, 3California Pacific Medical Center, San Francisco, CA, USA
E‐mail address for correspondence: [email protected]
Background: We have observed that ALS patients have elevated levels of activated macrophages in the peripheral blood and proceeded to investigate the gene expression pattern using microarrays and quantitative RT‐PCR. Peripheral blood mononuclear cells from ALS and healthy individuals were found to express a distinct myeloid/interferon‐induced (MIFN) transcriptional signature. The MIFN Signature is a characteristic transcriptional response of myeloid cells occurring immediately after cell isolation from the bloodstream. To further evaluate the MIFN Signature, a number of cytokines and genes that were highly up‐regulated in ALS patients were selected for analysis of their protein expression.
Objective: To investigate the expression of MIFN Signature proteins and pro‐inflammatory cytokines by peripheral blood myeloid cells from individuals with ALS.
Methods: Plasma and whole blood cells were obtained from ALS patients and healthy controls. Mononuclear cells were purified from blood and incubated in cell culture medium for 24 h at 37°C. Tissue culture supernatants (TCS) from cultured mononuclear cells were tested for secreted cytokine and MIFN Signature proteins by ELISA, as was plasma from ALS patients and controls. Cell surface and intracellular expression of MIFN Signature proteins were analysed by flow cytometry.
Results: In TCS from ALS patient mononuclear cells, TNF‐alpha levels increased from 97 pg/ml to 159 pg/ml within 3 h before decreasing to 40 pg/ml after 24 h; TNF‐alpha levels from controls remained at 2 pg/ml through all time points. IL‐6 protein levels in ALS TCS increased 7.8‐fold in 24 h, while the increase in IL‐6 levels from healthy individuals were negligible. The MIFN signature proteins, PI3 (Elafin) and IL‐1 receptor antagonist (IL‐1RN), increased 53‐ and 67‐fold, respectively, after 24 h in TCS from ALS patients but not controls. Mean plasma protein levels of PI3 were approximately 2‐fold higher in ALS patients than in healthy individuals (p<0.05). Mean IL‐1RN levels were reduced in ALS patients relative to controls. TRAIL and IL‐6 plasma levels were similar in both groups. Expression of 10 MIFN Signature proteins was evaluated by flow cytometry and all 10 proteins were expressed primarily in polymorphonuclear cells and monocytes. Of the MIFN Signature proteins evaluated, IL‐1RN, FPRL1, and Chitinase 3‐like 1 (CHI3L1), were found to have a higher mean fluorescent staining at both 0 and 24 h among ALS patients relative to controls.
Conclusions: Peripheral blood myeloid cells from patients with ALS are committed to a pro‐inflammatory program dominated by secretion of TNF‐alpha and IL‐6 prior to leaving the bloodstream. Additionally, peripheral blood myeloid cells from ALS patients express a distinct protein profile that can be employed to identify individuals with ALS. This can benefit both the patient and the medical community by allowing for earlier treatment of this neurodegenerative disorder.