Hannah M. Gibbons

Microglia induce neural cell death via a proximity-dependent mechanism involving nitric oxide

Microglial cells play a major role in the pathogenesis of many neurological diseases by exacerbating neuronal and non-neuronal cell death, but the mechanisms involved are unclear. To investigate the microglial-neuronal interactions, we used the murine BV-2 microglial cell line and the human neuronal-like SK-N-SH neuroblastoma cell line in a co-culture system that enabled proximity-dependent interaction and communication, a trans-well system that allowed proximity-independent communication through diffusible molecules only, and a conditioned media system through which no proximity-dependent interactions or cell-to-cell communication is possible. Activation of BV-2 cells with lipopolysaccharide and interferon-gamma (LPS/IFN-gamma) decreased viability of the BV-2 cells alone and in co-cultures with SK-N-SH cells, but not SK-N-SH cells grown alone. In contrast, activation of BV-2 cells in the trans-well and conditioned media system did not have any effect on the viability of SK-N-SH cells, suggesting that microglia must be in close proximity to the neural cells to elicit cytotoxicity. To determine the molecules involved in proximity-dependent cell death, inhibitors of microglial activation were investigated. Only the specific inducible nitric oxide synthase (iNOS) inhibitor S-methylisothiourea, and hypothermia, which is known to suppress microglial iNOS expression, prevented cell death after LPS/IFN-gamma activation. These results suggest that activated microglia release nitric oxide that is, at least partially, responsible for proximity-dependent microglial-mediated neural toxicity.

c-Jun promotes neurite outgrowth and survival in PC12 cells.

We investigated the function of c-Jun in PC12 cells by transfecting them with a plasmid containing a c-Jun cDNA transcription cassette. Transfected cells expressed high levels of c-Jun mRNA and protein and demonstrated an increase in both AP-1 DNA binding and gene activation. The c-Jun over-expressing cells showed marked neurite outgrowth but no evidence of spontaneous cell death. In fact, c-Jun over-expressing cells were more resistant to okadaic acid-induced apoptosis. The process outgrowth was not indicative of a full neuronal differentiation response as the transfected PC12 cells did not display action potentials when examined with whole-cell patch-clamping. The phosphorylation of c-Jun on serine 73 appears to be important for this neurite sprouting effect as mutagenesis at this site reduced sprouting whereas a serine 63 mutant tended to increase sprouting. Thus, in PC12 cells c-Jun expression does not induce apoptosis, but rather functions as a neurite outgrowth and neuronal survival signal.

Doublecortin expression in the normal and epileptic adult human brain

Mesial temporal lobe epilepsy (MTLE) is a neurological disorder associated with spontaneous recurrent complex partial seizures and hippocampal sclerosis. Although increased hippocampal neurogenesis has been reported in animal models of MTLE, increased neurogenesis has not been reported in the hippocampus of adult human MTLE cases. Here we showed that cells expressing doublecortin (Dcx), a microtubule‐associated protein expressed in migrating neuroblasts, were present in the hippocampus and temporal cortex of the normal and MTLE adult human brain. In particular, increased numbers of Dcx‐positive cells were observed in the epileptic compared with the normal temporal cortex. Importantly, 56% of Dcx‐expressing cells in the epileptic temporal cortex coexpressed both the proliferative cell marker, proliferating cell nuclear antigen and early neuronal marker, TuJ1, suggesting that they may be newly generated neurons. A subpopulation of Dcx‐positive cells in the epileptic temporal cortex also coexpressed the mature neuronal marker, NeuN, suggesting that epilepsy may promote the generation of new neurons in the temporal cortex. This study has identified, for the first time, a novel population of Dcx‐positive cells in the adult human temporal cortex that can be upregulated by epilepsy and thus, raises the possibility that these cells may have functional significance in the pathophysiology of epilepsy.

Valproic acid induces caspase 3-mediated apoptosis in microglial cells

Valproic acid is widely used for the treatment of epilepsy and mood disorders, but its mode of action is unclear. Treatment of neuronal cells with valproic acid promotes neurite sprouting, is neuroprotective and drives neurogenesis; however its effects on non-neuronal brain cells are less clear. We report that valproic acid induces apoptosis in the mouse microglial cell line, BV-2, at concentrations within the therapeutic range. When BV-2 cells were incubated for 24 h with 500-1000 microM valproic acid we observed a reduction in cell number, the appearance of apoptotic morphology and increased caspase 3 cleavage. Exposure of a macrophage cell line (RAW 264.7) to similar concentrations of valproic acid also led to reduced cell number but no caspase 3 cleavage, suggesting these cells responded to valproic acid with reduced proliferation rather than apoptosis. This was confirmed using bromodeoxyuridine incorporation studies. Similar concentrations of valproic acid added to Neuro-2a, SK-N-SH and C6 cell lines as well as human NTera-2 astrocytes did not evoke cell death. The caspase 3 inhibitor DEVD-CHO inhibited valproic acid-induced apoptosis in BV-2 cells whereas the MEK inhibitor U0126 potentiated valproic acid-mediated apoptosis. These results demonstrate that valproic acid selectively induces apoptosis in BV-2 cells by way of a caspase 3-mediated action. As activated microglia secrete neurotoxins in neurodegenerative diseases such as Alzheimers, Parkinsons, and HIV dementia, valproic acid may alleviate these diseases by selectively killing microglia.

Noradrenaline transporter expression in the pons and medulla oblongata of the rat : localisation to noradrenergic and some C1 adrenergic neurones

Catecholaminergic neurotransmission is normally terminated by rapid re-uptake of the neurotransmitter by a high-affinity Na+/Cl--dependent plasma membrane transporter. Specific transporters have been cloned for both dopamine (DAT) and noradrenaline (NAT) in the rat. While DAT has been studied extensively, NAT expression has received less attention, particularly at the protein level. We used an antibody generated against a 49 residue segment of an extracellular loop region of NAT to study expression of the transporter protein throughout the rat pons and medulla oblongata. NAT was expressed in over 95% of noradrenergic neurones in the A1, A2/area postrema, A5, A6/locus subcoeruleus, and A7 noradrenergic groups. Approximately 10% of C1 adrenergic neurones located in the rostral ventrolateral medulla (RVL) also expressed NAT. Expression of NAT mRNA in bulbospinal C1 cells was confirmed using single-cell reverse transcription polymerase chain reaction (RT-PCR) of acutely isolated RVL neurones. Spinally projecting neurones were identified by retrograde labelling with rhodamine beads, and C1 neurones were identified by RT-PCR using primers specific for tyrosine hydroxylase (TH) or phenylethanolamine N-methyltransferase (PNMT) mRNAs. Thirteen percent of adrenergic bulbospinal neurones tested expressed NAT mRNA. C1 neurones are potentially important in cardiovascular control and blood pressure regulation, and the identification of NAT expression in a sub-population of these neurones provides further evidence for the heterogeneity of this neuronal population.

Hypothermia suppresses inducible nitric oxide synthase and stimulates cyclooxygenase-2 in lipopolysaccharide stimulated BV-2 cells

Hypothermia is neuroprotective, possibly through suppression of microglial activation. We investigated the effects of hypothermia on lipopolysaccharide (LPS) stimulated BV-2 cells. At 37 degrees C, LPS elicited strong increases in inducible nitric oxide synthase (iNOS), nitric oxide (NO), cyclooxygenase-2 (COX-2), tumour necrosis factor-alpha (TNF-alpha), and interleukin-6 (IL-6), accompanied by translocation of nuclear factor-kappaB (NF-kappaB) to the nucleus. Hypothermia (33 degrees C) caused complete suppression of iNOS and NO, a partial reduction of IL-6 but did not prevent TNF-alpha production or NF-kappaB translocation. In contrast, LPS induced cyclooxygenase-2 (COX-2) to higher levels under hypothermic conditions. These results show that hypothermia selectively suppresses iNOS in microglia.

Effects of cyanide and hypoxia on membrane currents in neurones acutely dissociated from the rostral ventrolateral medulla of the rat

Previous reports suggested that some neurones located in the rostral ventrolateral medulla (RVL) can act as fast oxygen sensors which enhance the sympathetic activity and blood pressure independent of peripheral chemoreceptors. The aim of this study was to compare hypoxic responses of different subpopulations of RVL neurones to ascertain whether the hypoxic sensitivity is restricted to one group of these neurones. Whole-cell patch-clamp recordings were made from acutely dissociated neurones obtained from RVL of P13-P19 rats. Short-lasting hypoxia (1-2 min) was evoked by pressure injection of NaCN or lowering pO2. Cells projecting to the upper thoracic segments were retrogradely labelled with fluorescent beads. Catecholaminergic (CA) or non-catecholaminergic (non-CA) neurones were identified using single-cell reverse-transcription polymerase chain reaction (RT-PCR) or immunocytochemistry. Recordings were made from 38 neurones (26 spinally-projecting, 12 non-spinal) using Cs+/TEA or K+-containing pipettes. In most of the cells tested with slow depolarising ramp commands (78%; including spinally-projecting and non-spinal neurones, as well as CA and non-CA neurones), NaCN or hypoxia evoked a reversible increase of the sustained inward current. Extracellular application of 1 mM Co2+ or 25 nM TTX revealed three components of the hypoxia-sensitive inward current which resembled the persistent sodium (INaP), low threshold calcium (LVA Ca2+) and high threshold calcium (HVA Ca2+) currents. The NaCN or hypoxia induced increase of the current could also be observed during step commands. Recordings with K+-containing pipettes during similar depolarising ramps revealed, in addition, a reversible increase of IK in 78% of tested cells (in all four types of examined neurones). These results are consistent with the concepts that RVL neurones can act as a central oxygen sensor. However, in contrast to the previously published data demonstrating that in pentobarbital anaesthetised rats only the barosensitive and spinally projecting cells were affected by a short-lasting hypoxia, our findings obtained with dissociated RVL neurones indicate that sensitivity to hypoxia is widely distributed within this part of the medulla oblongata.

Extracellular signal-regulated kinase involvement in human astrocyte migration.

Glial scar formation occurs after virtually any injury to the brain. The migration of astrocytes into regions of brain injury underlies the formation of the glial scar. The exact role of the glial scar has yet to be elucidated, although it is likely to impair brain recovery. Understanding astrocyte migration is fundamental to understanding the formation of the glial scar. We have used human astrocytes (NT2A cells), derived from human NT2/D1 precursor cells to study astrocyte migration using an in vitro scratch wound model. Time-lapse microscopy and bromodeoxyuridine labeling revealed that the astrocytes migrated rather than proliferated across the scratch. Time course immunocytochemical studies showed that scratching human astrocytes induced the activation (phosphorylation) of ERK 1/2 at 10 min after scratch. The MEK 1/2 inhibitor U0126 inhibited both the ERK 1/2 phosphorylation and the migration of the astrocytes across the wound after scratch. Thus, the migration of human astrocytes after injury is partly initiated by activation of the MEK-ERK signalling pathway.

M-CSF increases proliferation and phagocytosis while modulating receptor and transcription factor expression in adult human microglia

BackgroundMicroglia are the primary immune cells of the brain whose phenotype largely depends on their surrounding micro-environment. Microglia respond to a multitude of soluble molecules produced by a variety of brain cells. Macrophage colony-stimulating factor (M-CSF) is a cytokine found in the brain whose receptor is expressed by microglia. Previous studies suggest a critical role for M-CSF in brain development and normal functioning as well as in several disease processes involving neuroinflammation.MethodsUsing biopsy tissue from patients with intractable temporal epilepsy and autopsy tissue, we cultured primary adult human microglia to investigate their response to M-CSF. Mixed glial cultures were treated with 25 ng/ml M-CSF for 96 hours. Proliferation and phagocytosis assays, and high through-put immunocytochemistry, microscopy and image analysis were performed to investigate microglial phenotype and function.ResultsWe found that the phenotype of primary adult human microglia was markedly changed following exposure to M-CSF. A greater number of microglia were present in the M-CSF- treated cultures as the percentage of proliferating (BrdU and Ki67-positive) microglia was greatly increased. A number of changes in protein expression occurred following M-CSF treatment, including increased transcription factors PU.1 and C/EBPβ, increased DAP12 adaptor protein, increased M-CSF receptor (CSF-1R) and IGF-1 receptor, and reduced HLA-DP, DQ, DR antigen presentation protein. Furthermore, a distinct morphological change was observed with elongation of microglial processes. These changes in phenotype were accompanied by a functional increase in phagocytosis of Aβ1-42 peptide.ConclusionsWe show here that the cytokine M-CSF dramatically influences the phenotype of adult human microglia. These results pave the way for future investigation of M-CSF-related targets for human therapeutic benefit.

Cellular composition of human glial cultures from adult biopsy brain tissue

Microglia and astrocytes play vital roles in normal human brain function and in neurological disorders. To study their physiological and pathological roles it is desirable to establish in vitro systems that are derived from the adult human brain. Although several groups have successfully cultured cells from the human brain, the composition of these cultures remains controversial. Using morphological criteria, immunocytochemical analysis and a BrdU incorporation assay we demonstrate the presence of poorly proliferative microglia and astrocytes in cultures derived from epilepsy biopsy tissue. In addition, we characterized a third cell type as fibronectin and prolyl 4-hydroxylase immunopositive fibroblast-like cells, which are highly proliferative and become the predominant cell type after successive sub-culturing. Therefore, although cultures from adult human brain tissue provide an excellent resource for studying human glial cells, careful consideration must be given to their cellular composition when performing studies using these methods.