Quinazoline-based tricyclic compounds that regulate programmed cell death, induce neuronal differentiation, and are curative in animal models for excitotoxicity and hereditary brain disease

Expanding on a quinazoline scaffold, we developed tricyclic compounds with biological activity. These compounds bind to the 18 kDa translocator protein (TSPO) and protect U118MG (glioblastoma cell line of glial origin) cells from glutamate-induced cell death. Fascinating, they can induce neuronal differentiation of PC12 cells (cell line of pheochromocytoma origin with neuronal characteristics) known to display neuronal characteristics, including outgrowth of neurites, tubulin expression, and NeuN (antigen known as 'neuronal nuclei', also known as Rbfox3) expression. As part of the neurodifferentiation process, they can amplify cell death induced by glutamate. Interestingly, the compound 2-phenylquinazolin-4-yl dimethylcarbamate (MGV-1) can induce expansive neurite sprouting on its own and also in synergy with nerve growth factor and with glutamate. Glycine is not required, indicating that N-methyl-D-aspartate receptors are not involved in this activity. These diverse effects on cells of glial origin and on cells with neuronal characteristics induced in culture by this one compound, MGV-1, as reported in this article, mimic the diverse events that take place during embryonic development of the brain (maintenance of glial integrity, differentiation of progenitor cells to mature neurons, and weeding out of non-differentiating progenitor cells). Such mechanisms are also important for protective, curative, and restorative processes that occur during and after brain injury and brain disease. Indeed, we found in a rat model of systemic kainic acid injection that MGV-1 can prevent seizures, counteract the process of ongoing brain damage, including edema, and restore behavior defects to normal patterns. Furthermore, in the R6-2 (transgenic mouse model for Huntington disease; Strain name: B6CBA-Tg(HDexon1)62Gpb/3J) transgenic mouse model for Huntington disease, derivatives of MGV-1 can increase lifespan by >20% and reduce incidence of abnormal movements. Also in vitro, these derivatives were more effective than MGV-1.

The role of 18 kDa mitochondrial translocator protein (TSPO) in programmed cell death, and effects of steroids on TSPO expression

The mitochondrial 18 kDa Translocator Protein (TSPO) was first detected by its capability to bind benzodiazepines in peripheral tissues and later also in glial cells in the brain, hence its previous most common name peripheral benzodiazepine receptor (PBR). TSPO has been implicated in various functions, including apoptosis and steroidogenesis, among others. Various endogenous TSPO ligands have been proposed, for example: Diazepam Binding Inhibitor (DBI), triakontatetraneuropeptide (TTN), phospholipase A2 (PLA2), and protoporphyrin IX. However, the functional implications of interactions between the TSPO and its putative endogenous ligands still have to be firmly established. The TSPO has been suggested to interact with a mitochondrial protein complex, summarized as mitochondrial membrane permeability transition pore (MPTP), which is considered to regulate the mitochondrial membrane potential (ΔΨm). In addition, the TSPO is associated with several other proteins. The associations of the TSPO with these various proteins at the mitochondrial membranes have been attributed to functions such as apoptosis, steroidogenesis, phosphorylation, reactive oxygen species (ROS) generation, ATP production, and collapse of the ΔΨm. Interestingly, while TSPO is known to play a role in the modulation of steroid production, in turn, steroids are also known to affect TSPO expression. As with the putative endogenous TSPO ligands, the effects of steroids on TSPO functions still have to be established. In any case, steroid-TSPO interactions occur in organs and tissues as diverse as the reproductive system, kidney, and brain. In general, the steroid-TSPO interactions are thought to be part of stress responses, but may also be essential for reproductive events, embryonic development, and responses to injury, including brain injury. The present review focuses on the role of TSPO in cell death i.e. the notion that enhanced expression and/or activation of the TSPO leads to cell death, and the potential of steroids to regulate TSPO expression and activation.

The 18 kDa mitochondrial translocator protein (TSPO) prevents accumulation of protoporphyrin IX. Involvement of reactive oxygen species (ROS)

By exposing cells of the U118MG glioblastoma cell line to protoporphyrin IX (PPIX) in culture, we found that the 18 kDa mitochondrial translocator protein (TSPO) prevents intracellular accumulation of PPIX. In particular, TSPO knockdown by stable transfection of TSPO silencing siRNA vectors into U118MG cells leads to mitochondrial PPIX accumulation. In combination with light exposure, the PPIX accumulation led to cell death of the TSPO knockdown cells. In the sham control cells (stable transfection of scrambled siRNA vectors), TSPO expression remained high and no PPIX accumulation was observed. The prevention of PPIX accumulation by TSPO was not due to conversion of PPIX to heme in the sham control cells. Similar to TSPO knockdown, the reactive oxygen species (ROS) scavenger glutathione (GSH) also enhanced PPIX accumulation. This suggests that that ROS generation as modulated by TSPO activation may present a mechanism to prevent accumulation of PPIX.

Cigarette Smoke Decreases Salivary 18 kDa Translocator Protein Binding Affinity – in Association with Oxidative Stress

Reactive oxygen species (ROS) generated by cigarette smoke may contribute to lung and oral cancer. The 18 kDa Translocator protein (TSPO) has been reported to be affected by ROS as well as to participate in ROS generation at mitochondrial levels, and has been implicated in pro-apoptotic and anti-carcinogenic functions. The present study reports the presence of TSPO in the cellular fraction of human saliva. In cells collected from untreated saliva, the specific TSPO ligand [(3)H]PK 11195 showed saturable binding with high affinity, with mean B(max) and K(d) values of 6,471 +/- 501 fmol/mg protein and 6.2 +/- 0.5 nM, respectively. Our study further indicates that the cellular fraction of human saliva possesses TSPO with binding characteristics similar to that of cells from other tissues of human origin. Following exposure of saliva to cigarette smoke a three-fold decrease in the affinity of salivary TSPO to its specific ligand, [(3)H]PK 11195 (p < 0.01) occurred in the cellular fraction of the saliva, in comparison to sham treated control, without significant accompanying changes in TSPO B(max), TSPO protein levels, or general protein levels. The changes in affinity of TSPO from the cellular fraction of saliva exposed to cigarette smoke were accompanied by changes in the mean levels of protein oxidation products (carbonyls) and lipid peroxides, which were three-fold higher (p < 0.01) and two-fold higher (p < 0.01), respectively, compared to those of sham treated controls. Thus, our study shows that TSPO is present in the cellular component of saliva. Interestingly, in vitro this cellular TSPO is affected by exposure of the whole saliva to cigarette smoke, in negative correlation with oxidative stress.

The effect of oxygenation level on cerebral post-traumatic apoptotsis is modulated by the 18-kDa translocator protein (also known as peripheral-type benzodiazepine receptor) in a rat model of cortical contusion

AIMS

Hyperbaric hyperoxia has been shown to reduce apoptosis in brain injury. As the 18-kDa translocator protein (TSPO), also known as peripheral-type benzodiazepine receptor, is closely associated with the mitochondrial transition pore and because of its role in mitochondrial respiration and apoptosis, we hypothesized that reduction of apoptosis by hyperoxia may involve the TSPO.

METHODS

TSPO and transferase-mediated dUTP nick end labelling (TUNEL) immunopositivity was first assessed in cortical contusion, created by dynamic cortical deformation, by immunohistochemistry in rats exposed to normoxia [(dynamic cortical deformation (DCD)], normobaric hyperoxia or hyperbaric hyperoxia [hyperbaric oxygen therapy (HBO)]. In a second step, transmembrane mitochondrial potential (Deltapsi(M)) and caspase 9 activity were assessed in the injured area in comparison with the noninjured hemisphere. Measurements were performed in DCD and HBO groups. A third group receiving both HBO and the TSPO ligand PK11195 was investigated as well.

RESULTS

TSPO correlated quantitatively and regionally with TUNEL immunopositivity in the perilesional area. Hyperoxia reduced both the number of TSPO expressing and TUNEL positive cells in the perilesional area, and this effect proved to be pressure dependent. After contusion, we demonstrated a dissipation of Deltapsi(M) in isolated mitochondria and an elevation of caspase 9 activity in tissue homogenates from the contused area, both of which could be substantially reversed by hyperbaric hyperoxia. This protective effect of hyperoxia was reversed by PK11195.

CONCLUSIONS

The present findings suggest that the protective effect of hyperoxia may be due to a negative regulation of the proapoptotic function of mitochondrial TSPO, including conservation of the mitochondrial membrane potential.

PK 11195 attenuates kainic acid-induced seizures and alterations in peripheral-type benzodiazepine receptor (PBR) protein components in the rat brain

Peripheral-type benzodiazepine receptors (PBR) are located in glial cells in the brain and in peripheral tissues. Mitochondria form the primary location for PBR. Functional PBR appear to require at least three components: an isoquinoline binding protein, a voltage-dependent anion channel, and an adenine nucleotide carrier. In the present study, rats received intraperitoneal kainic acid injections, which are known to cause seizures, neurodegeneration, hyperactivity, gliosis, and a fivefold increase in PBR ligand binding density in the hippocampus. In the forebrain of control rats, hippocampal voltage-dependent anion channel and adenine nucleotide carrier abundance was relatively low, while isoquinoline binding protein abundance did not differ between hippocampus and the rest of the forebrain. One week after kainic acid injection, isoquinoline binding protein abundance was increased more than 20-fold in the hippocampal mitochondrial fraction. No significant changes were detected regarding hippocampal voltage-dependent anion channel and adenine nucleotide carrier abundance. Pre-treatment with the isoquinoline PK11195, a specific PBR ligand, attenuated the occurrence of seizures, hyperactivity, and increases in isoquinoline binding protein levels in the hippocampus, which usually follow kainic acid application. These data suggest that isoquinoline binding protein may be involved in these effects of kainic acid injections.

Acute and repeated swim stress effects on peripheral benzodiazepine receptors in the rat hippocampus, adrenal, and kidney

Peripheral benzodiazepine receptor (PBR) density has been found to be sensitive to stress. We set out to compare the influences of acute and repeated swim stress on behavior and PBR density. Following acute and repeated swim stress, rats were tested in an elevated plus-maze and an open-field test for anxiety levels, and tissues were collected from the adrenal gland, kidney, and hippocampus for measurements of PBR density. The acute rather than the repeated stress led to robust alterations in PBR density. The largest reduction in hippocampal and adrenal gland PBR density was found one hour after acute stress. In the hippocampus, acute stress caused a biphasic change in PBR density: a robust reduction in PBR density one hour after the acute stress and a distinct elevation in PBR density at 24 hours, while 72 hours after stress the elevation in PBR density appeared to be reduced.

Tissue-specific regulation of the peripheral benzodiazepine receptor by antidepressants and lithium

Peripheral benzodiazepine receptors (PBR) have been identified in peripheral organs as well as in brain glial cells. PBR differ from central benzodiazepine receptors (CBR) in their lack of coupling to the gamma-aminobutyric acid receptors and the chloride ion channels. We investigated the effect of 21 days administration, followed by 7 days withdrawal, of fluvoxamine (10 mg/kg), desipramine (10 mg/kg) and lithium carbonate (25 mg/kg) on PBR and CBR binding characteristics in male Sprague-Dawley rats. All three agents significantly increased PBR density in the testes and adrenals. All tested drugs induced a significant decrease in PBR density in the kidney and liver. After withdrawal, PBR density remained decreased in the liver in all three groups and in the kidneys of the desipramine- and lithium-treated animals. In the cerebral cortex, CBR density increased in response to all three agents, whereas PBR density decreased significantly in response to desipramine and lithium carbonate. Chronic treatment with fluvoxamine, desipramine and lithium carbonate is apparently associated with a modulation in PBR expression in the testes, adrenals, kidneys, liver and brain, and in CBR expression in brain. The relevance of these tissue-selective alterations to the antidepressive and/or anxiolytic effects of these agents, or their adverse effects, still needs to be determined.

Dissociating context and space within the hippocampus: effects of complete, dorsal, and ventral excitotoxic hippocampal lesions on conditioned freezing and spatial learning

Rats with complete excitotoxic hippocampal lesions or selective damage to the dorsal or ventral hippocampus were compared with controls on measures of contextually conditioned freezing in a signaled shock procedure and on a spatial water-maze task. Complete and ventral lesions produced equivalent, significant anterograde deficits in conditioned freezing relative to both dorsal lesions and controls. Complete hippocampal lesions impaired water-maze performance; in contrast, ventral lesions improved performance relative to the dorsal group, which was itself unexpectedly unimpaired relative to controls. Thus, the partial lesion effects seen in the 2 tasks never resembled each other. Anterograde impairments in contextual freezing and spatial learning do not share a common underlying neural basis; complete and ventral lesions may induce anterograde contextual freezing impairments by enhancing locomotor activity under conditions of mild stress.

Evidence for the preferential localization of glutamate receptor-1 subunits of AMPA receptors to the dendritic spines of medium spiny neurons in rat striatum

Although immunohistochemical studies have typically found the perikarya of striatal projection neurons to be devoid of immunohistochemical labelling for the GluR1 AMPA type glutamate receptor subunit, the striatal neuropil is rich in GluR1 immunolabelling and in situ hybridization histochemistry has indicated the presence of GluR1 message in many striatal neurons. To explore the possibility that GluR1 subunits may be synthesized by many striatal projection neurons, but selectively localized to their dendrites, we have used light-microscopic and electron-microscopic immunohistochemistry in combination with single-cell reverse transcription-polymerase chain reaction. Light-microscopic immunohistochemical studies confirmed the presence of abundant GluR1 immunoreactivity in the striatal neuropil in rats. Perikaryal labelling was restricted to neurons previously identified as parvalbuminergic neurons. Single-cell reverse transcription-polymerase chain reaction for individual striatal neurons in rats confirmed that most striatal projection neurons (i.e. containing either or both substance P message or enkephalin message) make GluR1 message. For example, 94% of enkephalin-containing neurons, 75% of substance P-containing neurons, and 87% of enkephalin and substance P co-containing neurons expressed GluR1 messenger RNA. Electron-microscopic immunohistochemistry revealed that GluR1 immunolabelling was prominent in 61% of dendritic spines and 53% of dendritic shafts. While prominent perikaryal GluR1 immunolabelling was observed only in a small population of interneurons, sparse perikaryal GluR1 immunolabelling was found associated with the rough endoplasmic reticulum, the Golgi apparatus, the outer membranes of the mitochondria, and the outer envelope of the nucleus of about 30% of striatal projection neurons (identified by their non-indented nuclei). These results indicate that striatal projection neurons selectively target GluR1 subunits to their spines and dendritic shafts. Our finding has implications for the functioning of striatal projection neurons and for the general issue of whether neurons can control the subcellular localization of glutamate receptors.

Electrolytic lesion of globus pallidus ameliorates the behavioral and neurodegenerative effects of quinolinic acid lesion of the striatum: a potential novel treatment in a rat model of Huntington's disease

Bilateral electrolytic pallidal lesion ameliorated the deleterious effects of bilateral quinolinic acid (QA) lesion to the striatum on post-surgery weight, activity level, and performance in a water maze task, and reduced the extent of striatal damage. Given that the neurodegenerative and behavioral effects of QA striatal lesion are thought to mimic those seen in Huntington's disease, these results may point to a potential novel treatment for this disease.