C33 ACTIVATED MICROGLIA IN ALS: A DOUBLE‐EDGED SWORD
Zhao W, Xiao Q, Henkel JS, Beers DR, Simpson EP, Yen AA & Appel SH
Methodist Neurological Institute, Houston, USA
E‐mail address for correspondence: [email protected]
Background: Recent studies of ALS suggest that motor neuron death is non‐cell autonomous, with cell injury possibly mediated by interactions with non‐neuronal cells such as microglia. Our recent in vitro data demonstrated that activated microglia release free radicals, including nitric oxide (NO) and superoxide (O2• −), and initiate motor neuron (MN) injury by a glutamate‐mediated effect. Microglia can also release neurotrophic factors, including IGF‐1, as well as free radicals. A key question is whether microglia can be induced to protect against motor neuron injury in vitro, and whether mSOD1 microglia are less neuroprotective than wild‐type microglia.
Objective: To determine whether microglia can be neuroprotective, and if so, by what mechanisms.
Methods: Primary motor neuron cultures were prepared from embryonic rat spinal cords. Microglial cultures were prepared from Sprague Dawley rats and mSOD1 and wild‐type littermate mice. The anti‐inflammatory cytokines, IL‐4 or IL‐10, were applied to cultures, followed by LPS. The production of NO was estimated in the culture media and iNOS expression was measured by Western blots. IGF‐1 mRNA was detected by RT‐PCR, and protein levels of IGF‐1 and IL‐4 were measured by ELISA. MN survival was quantified by direct counting of all p75 positive live cells displaying intact neurites longer than three cell diameters.
Results: IL‐4 and IL‐10 significantly reversed motor neuron injury induced by microglia activated by LPS. The increased motor neuron survival correlated directly with the decrease in NO production. IL‐4 and IL‐10 suppressed NO release by down‐regulating microglial iNOS. Although IL‐4 and IL‐10 enhanced IGF‐1 protein levels in motor neuron and resting microglia co‐cultures (in the absence of LPS), IL‐4 and IL‐10 had minimal effects on IGF‐1 protein levels (<10 pg/ml) in motor neuron and microglia co‐cultures treated with LPS. IGF‐1 at 2–50 ng/ml did not protect against the cytotoxic effects of peroxynitrite and glutamate on motor neuron cultures. Thus, IGF‐1 does not explain the neuroprotective effects of IL‐4. mSOD1 microglia expressed higher NO levels, lower levels of iNOS, and lower levels of IGF‐1, and were less neuroprotective than wild‐type microglia.
Conclusions: The anti‐inflammatory cytokines, IL‐4 and IL‐10, protect against motor neuron toxicity induced by activated microglia. This neuroprotection is mediated by down‐regulation of iNOS expression and lowered production of NO. It does not appear to depend upon IGF‐1 production or release. These anti‐inflammatory cytokines may have possible therapeutic value in ALS primarily by their ability to down‐regulate the production and release of free radicals. Furthermore mSOD1 microglia appear to have significantly less neuroprotective capacity.
C34 MICROGLIAL ACTIVATION IN CX3CR1 NULL‐SOD1(G93A) ALS MICE WORSENS DISEASE PHENOTYPE
Pioro EP, Kostenko V, Cardona A & Ransohoff RM
Cleveland Clinic Foundation, Cleveland, USA
E‐mail address for correspondence: [email protected]
Background: Microglial mediated inflammation has been implicated in the loss of motor neurons (MNs) of amyotrophic lateral sclerosis (ALS). The fractalkine receptor, CX3CR1, is localized to microglia and suppresses their activation via its ligand, fractalkine. The transgenic (Tg) mouse model of ALS expressing the G93A‐mutated human SOD1 gene (SOD1G93A), exhibits age‐dependent degeneration of MNs, limb weakness, and early death.
Objectives: To determine if TgSOD1G93A mice without fractalkine receptor, i.e. null for the CX3CR1 gene (CX3CR1−/−), experience more microglial activation, greater MN loss and limb weakness, and shorter survival than Tg mice with at least one copy of the gene (CX3CR1−/+).
Methods: We established a colony of mice that were TgSOD1G93A‐CX3CR1−/−, TgSOD1G93A‐CX3CR1+/−, and non‐TgSOD1G93A‐CX3CR1−/− (as unaffected littermate controls), and performed serial behavioral and survival assessment, with histological analysis of terminal mice at 130–140 days.
Results: Microglial reaction was increased in the lumbar spinal cords of TgSOD1G93A‐CX3CR1−/− mice compared with either TgSOD1G93A‐CX3CR1+/−, or non‐TgSOD1G93A‐CX3CR1−/− animals. There were significantly fewer Nissl‐stained lumbar cord MNs in TgSOD1G93A‐CX3CR1−/− mice (mean % reduction of MNs = 58.6±4.6%; n = 5) compared with TgSOD1G93A‐CX3CR1+/− mice (42.2±5.5%; n = 5; p = 0.03), using non‐TgSOD1G93A‐CX3CR1−/− littermates (n = 5) as a reference for both. After 7–9 weeks of age, hindlimb strength was lower in TgSOD1G93A mice compared to non‐TgSOD1G93A littermates. From 15 to 20 weeks, when weakness became progressive, the rate of decline in Tg mice was much faster in the CX3CR1−/− group than in the CX3CR1+/− group (p<0.01, random coefficient regression analysis). Furthermore, this more rapid decline resulted from male CX3CR1−/− mice (slope = −19.92±2.1, SE, n = 5) rather than female CX3CR1−/− mice (slope = −8.41±2.4, SE, n = 7). Random coefficient regression analysis also demonstrated a significant mouse‐group by gender by age interaction that indicated faster loss of body weight (p<0.01) and forelimb grip strength (p = 0.02) for male TgSOD1G93A‐CX3CR1−/− mice compared to all other TgSOD1G93A mice. Kaplan‐Meier analysis revealed survival was shorter by almost two weeks in TgSOD1G93A‐CX3CR1−/− mice (129 days) compared to TgSOD1G93A‐CX3CR1+/− mice (141 days, p = 0.01).
Conclusion: Absence of the CX3CR1 in TgSOD1G93A mice results in pronounced microglial reaction, greater MN loss, worsened neurobehavioral outcomes, and shorter survival, with the effect preferentially expressed in males. Future studies will elucidate the downstream mediators of microglial injury in an effort to identify novel therapeutic targets in ALS.
C35 TRANSCRIPTIONAL REGULATION OF INFLAMMATORY PROCESSES IN AMYOTROPHIC LATERAL SCLEROSIS: EXPRESSION OF CCAAT/ENHANCER BINDING PROTEIN (C/EBP) ß, SUPPRESSORS OF CYTOKINE SIGNALLING (SOCS) AND INTERFERON REGULATORY FACTOR‐1 (IRF‐1)
Anneser JMH, Franz S, Chahli C, Borasio GD & Lorenzl S
University of Munich, Munich, Germany
E‐mail address for correspondence: [email protected]
Background: Elevations of numerous proinflammatory cytokines and chemokines in the central nervous system of amyotrophic lateral sclerosis (ALS) patients as well as in an animal model of familial ALS (SOD‐transgenic mice) indicate that inflammatory mechanisms may contribute to motor neuron death in this disease. However, little is known about exact functions, interactions, and the regulation of these molecules in ALS pathogenesis.
Objectives: The expression of several molecules which may be involved in the regulation of inflammatory processes was investigated in spinal cords of ALS patients and SOD‐transgenic mice.
Methods: The microarray technique was used to study transcriptional changes of these molecules. Furthermore, we used immunoblotting and immunohistochemistry to investigate the amount of the corresponding protein and to identify the cell species in which these proteins are expressed.
Results: Microarray analysis revealed enhanced expression of CCAAT/enhancer‐binding protein‐ß (C/EBP ß), interferon regulatory factor‐1 (IRF‐1) and suppressor of cytokine signalling 1 (SOCS1), as well as a reduced expression of SOCS3 mRNA in the spinal cord of ALS patients and SOD‐transgenic mice. These results were confirmed by immunoblotting. Furthermore, several transcriptional targets of C/EBP ß, including interleukin 6 (IL‐6) were up‐regulated in the same specimens. Immunohistochemistry showed that C/EBP ß, IRF‐1, and SOCS proteins were mainly expressed in microglial and astroglial cells of the spinal cord ventral horn.
Discussion and conclusions: C/EBP ß is a transcription factor that couples extracellular signals to various intracellular processes but also participates in inflammatory processes. IL‐6 is not only a transcriptional target, but also a strong inductor of C/EBP ß expression. Elevated levels of these molecules may be part of a feed‐forward mechanism causing the spreading of the inflammation. The SOCS family comprises proteins induced on cytokine stimulation, which blocks further signalling in a feedback loop. SOCS 1 is the most potent inhibitor of cytokine signalling. Therefore, the elevated expression of SOCS1 in ALS may be an attempt to limit inflammatory processes in ALS. SOCS3 is a negative regulator of IL‐6. SOCS 3 deficiency, leading to sustained IL‐6 activation, has been associated with the development of chronic inflammatory diseases. IRF‐1 activates the expression of various downstream targets; for example, it has been associated with the induction of inducible nitric oxide synthase (iNOS). The astrogliosis and microgliosis observed in ALS are associated with altered levels of C/EBP ß, SOCS proteins and IRF‐1. These molecules may act as regulators of inflammatory signalling in this disease.
C36 DEFINITION OF A TRANSCRIPTIONAL SIGNATURE OF PERIPHERAL MYELOID CELLS ASSOCIATED WITH SEVERE NEURODEGENERATIVE DISEASE
Do H1, Yu S1, Mass J2, Katz j2, Zhang J3, Gascon R3, Lancero M3, Miller RG2, McGrath MS3 & Hadlock KG1
1Pathologica LLC, Burilingame, 2Forbes Norris MDA/ALS Research Center, San Francisco, and 3University of California San Francisco, San Francisco, USA
E‐mail address for correspondence: [email protected]
Background: Amyotrophic lateral sclerosis (ALS), Alzheimer's disease (AD), macular degeneration (MDgn) and HIV associated dementia are all characterized by elevated levels of activated (CD14/16++) macrophages in peripheral blood. This common phenotype suggests that these disparate diseases may share common mechanistic features. One indication of a common mechanism would be similar alterations in gene expression patterns relative to healthy controls.
Objective: To evaluate individuals with elevated levels of activated macrophages in their peripheral blood for common gene expression signatures.
Methods: Blood samples were obtained from individuals with ALS (n = 32), AD (n = 9), MDgn (n = 22), HIV infection (n = 5), and healthy controls (n = 20) after obtaining informed consent. Mononuclear cells were isolated by Percoll gradient centrifugation. The cells were then cultured overnight at 37°C under non‐adherent conditions, collected, lysed, and total RNA prepared. RNA expression data were obtained using an Affymetrix scanner and protocols. Selected genes identified by the microarray data were selected for quantitative RT‐PCR analysis using a Light‐Cycler. Genes were evaluated for coordinated transcription by determining the Pearson correlation coefficient (R) for each probe set with other probes expressed in ALS patients (∼23,000). Pairs of probes with an R>0.7 (p<0.0001) were said to exhibit correlated transcription.
Results: Samples from individuals with ALS (30 of 32), AD (9 of 9) and HIV infected individuals (4 of 5) all had a group of ∼500 genes that had R values of 0.7 or more with the majority of other members of the group. The genes in the signature were expressed at low levels and did not exhibit correlated transcription in healthy age‐matched controls. A significant fraction of the signature included genes known to be expressed in myeloid cells. Another prevalent group in the signature was genes associated with the α/β interferon‐mediated antiviral response. Other signature genes included genes associated with the innate immune response and regulation of apoptosis. Individuals with MDgn did not exhibit significant up‐regulation and correlated transcription of interferon response genes but did show elevation of some myeloid cell genes.
Conclusions: Peripheral blood cells from HIV infected individuals, ALS patients, and AD patients have a common transcriptional signature that is not seen in controls or individuals with MDgn. Although detection of a strong α/β interferon‐mediated antiviral response would be expected in HIV infected individuals, its presence in individuals with ALS or AD is surprising and suggests that one feature of the chronic inflammation seen in ALS and AD is an unregulated or inappropriate antiviral response.