Hypoxic preconditioning and cell death from oxygen/glucose deprivation co-opt a subset of the unfolded protein response in hippocampal neurons.

The state of protein folding in the endoplasmic reticulum (ER), via the unfolded protein response (UPR), regulates a pro- or anti-apoptotic cell fate. Hypoxic preconditioning (HPC) is a potent anti-apoptotic stimulus, wherein ischemic neural injury is averted by a non-damaging exposure to hypoxia. We tested if UPR modulation contributes to the pro-survival/anti-apoptotic phenotype in neurons preconditioned with hypoxia, using organotypic cultures of rat hippocampus as a model system. Pharmacologic induction of the UPR with tunicamycin increased mRNA of 79 of 84 UPR genes and replicated the pro-survival phenotype of HPC, whereas only small numbers of the same mRNAs were upregulated at 0, 6 and 24h after HPC. During the first 24h after HPC, protein signals in all 3 UPR pathways increased at various times: increased ATF4, phosphorylation of eif2α and IRE1, cleavage of xbb1 mRNA and cleavage of ATF6. Pharmacologic inhibition of ATF6 and IRE1 blocked HPC. Ischemia-like conditions (oxygen/glucose deprivation, OGD) caused extensive neuron cell damage and involved some of the same UPR protein signals as HPC. In distinction to HPC and tunicamycin, OGD caused widespread suppression of UPR genes: 55 of 84 UPR gene mRNAs were numerically downregulated. We conclude that although HPC and ischemic cell death in hippocampal neurons involve protein-based signaling in all 3 UPR pathways, these processes co-opt only a subset of the genomic response elicited by agents known to cause protein misfolding, possibly because of persistent transcription/translation arrest induced by hypoxia and especially OGD.

Transcriptional expression patterns triggered by chemically distinct neuroprotective molecules.

Glutamate-mediated excitotoxicity has been purported to underlie many neurodegenerative disorders. A subtype of glutamate receptors, namely N-methyl-d-aspartate (NMDA) receptors, has been recognized as potential targets for neuroprotection. To increase our understanding of the mechanisms that underlie this neuroprotection, we employed a mouse model of glutamate receptor-induced excitotoxic injury. Primary cortical neurons derived from postnatal day-0 CD-1 mice were cultured in the presence or absence of neuroprotective molecules and exposed to NMDA. Following a recovery period, whole genome expression was measured by microarray analysis. We used a combination of database and text mining, as well as systems modeling to identify signatures within the differentially expressed genes. While molecules differed in their mechanisms of action, we found significant overlap in the expression of a core group of genes and pathways. Many of these molecules have clear links to neuronal protection and survival, including ion channels, transporters, as well as signaling pathways including the mitogen-activated protein kinase (MAPK), the Toll-like receptor (TLR), and the hypoxic inducible factor (HIF). Within the TLR pathway, we also discovered a significant enrichment of interferon regulatory factor 7 (IRF7)-regulated genes. Knockdown of Irf7 by RNA interference resulted in reduced survival following NMDA treatment. Given the prominent role that IRF7 plays in the transduction of type-I interferons (IFNs), we also tested whether type-I IFNs alone functioned as neuroprotective agents and found that type-I IFNs were sufficient to promote neuronal survival. Our data suggest that the TLR/IRF7/IFN axis plays a significant role in recovery from glutamate-induced excitotoxicity.

Hypothermia and rewarming injury in hippocampal neurons involve intracellular Ca2+ and glutamate excitotoxicity.

This study examines the causes of hypothermia and rewarming injury in CA1, CA3, and dentate neurons in rat hippocampal slice cultures. Neuronal death, assessed with propidium iodide or Sytox fluorescence, Fluoro-Jade labeling, and Cresyl Violet staining, depended on the severity and duration of hypothermia. More than 6 h at temperatures less than 12 °C followed by rewarming to 37 °C (profound hypothermia and rewarming, PH/RW) caused swelling and death in large number of neurons in CA1, CA3, and dentate. During PH, [ATP] decreased and [Ca(2+)](I) and extracellular [glutamate] increased, with neuron rupture and nuclear condensation following RW. The data support the hypothesis that neuronal death from PH/RW is excitotoxic, due to ATP loss, glutamate receptor activation and Ca(2+) influx. We found that antagonism of N-methyl-D-aspartate (NMDA) receptors, but not 2-amino-3-(5-methyl-3-oxo-1,2- oxazol-4-yl) propanoic acid or metabotropic glutamate receptors, decreased neuron death and prevented increases in [Ca(2+)](I) caused by PH/RW. Chelating extracellular Ca(2+) decreased PH/RW injury, but inhibiting L- and T-type voltage-gated Ca(2+) channels, K+ channels, Ca(2+) release from the endoplasmic reticulum, and reverse Na(+)/Ca(2+) exchange did not affect the Ca(2+) changes or cell death. We conclude that the mechanism of PH/RW neuronal injury in hippocampal slices primarily involves intracellular Ca(2+) accumulation mediated by NMDA receptors that activates necrotic, but not apoptotic processes.

Longitudinal system-based analysis of transcriptional responses to type I interferons.

Type I interferons (IFNs) are pleiotropic cytokines that modulate both innate and adaptive immune responses. They have been used to treat autoimmune disorders, cancers, and viral infection and have been demonstrated to elicit differential responses within cells, despite sharing a single receptor. The molecular basis for such differential responses has remained elusive. To identify the mechanisms underlying differential type I IFN signaling, we used whole genome microarrays to measure longitudinal transcriptional events within human CD4(+) T cells treated with IFN-alpha(2b) or IFN-beta(1a). We identified differentially regulated genes, analyzed them for the enrichment of known promoter elements and pathways, and constructed a network module based on weighted gene coexpression network analysis (WGCNA). WGCNA uses advanced statistical measures to find interconnected modules of correlated genes. Overall, differential responses to IFN in CD4(+) T cells related to three dominant themes: migration, antigen presentation, and the cytotoxic response. For migration, WGCNA identified subtype-specific regulation of pre-mRNA processing factor 4 homolog B and eukaryotic translation initiation factor 4A2, which work at various levels within the cell to affect the expression of the chemokine CCL5. WGCNA also identified sterile alpha-motif domain-containing 9-like (SAMD9L) as critical in subtype-independent effects of IFN treatment. RNA interference of SAMD9L expression enhanced the migratory phenotype of activated T cells treated with IFN-beta compared with controls. Through the analysis of the dynamic transcriptional events after differential IFN treatment, we were able to identify specific signatures and to uncover novel genes that may underpin the type I IFN response.

Genetic variation in nitric oxide synthase 2A (NOS2A) and risk for multiple sclerosis.

Multiple sclerosis (MS) is a chronic inflammatory disorder of the central nervous system with a strong genetic component. Variation in the major histocompatibility complex on chromosome 6p21, specifically the HLA-DRB1*15 haplotype, is the strongest genetic factor for MS, yet it is estimated to account for only a portion of risk for the disease. Previous evidence has implicated the nitric oxide synthase gene (NOS2A) encoding inducible NOS on chromosome 17q11 as a potential MS susceptibility gene. To determine whether variation in the NOS2A gene contributes to MS risk, we investigated a total of 50 polymorphisms within or flanking the locus for evidence of association using a comprehensive analytical strategy. A total of 6265 members from 1858 well-characterized MS families were utilized. No evidence for overtransmission of any individual single-nucleotide polymorphism allele or haplotype to the MS-affected individuals was observed. Furthermore, different transmission rates were not observed in either DRB1*15-positive or DRB1*15-negative family subgroups, or when extreme clinical outcomes characterizing disease progression were examined. The very largest study of NOS2A variation in MS, to date, excludes even a modest role for this locus in susceptibility.