Synthesis and SAR of 4-aryl-2-hydroxy-4-oxobut-2-enoic acids and esters and 2-amino-4-aryl-4-oxobut-2-enoic acids and esters: potent inhibitors of kynurenine-3-hydroxylase as potential neuroprotective agents.

The synthesis and structure-activity relationship of a series of 4-aryl-2-hydroxy-4-oxobut-2-enoic acids and esters and 2-amino-4-aryl-4-oxobut-2-enoic acids and esters as potent inhibitors of kynurenine-3-hydroxylase are described. These compounds are the most potent inhibitors of the kynurenine-3-hydroxylase enzyme so far disclosed. Additionally methyl 4-(3-chlorophenyl)-2-hydroxy-4-oxobut-2-enoate (2d), 4-(3-chlorophenyl)-2-hydroxy-4-oxobut-2-enoic acid (3d), methyl 4-(3-fluorophenyl)-2-hydroxy-4-oxobut-2-enoate (2f), and 4-(3-fluorophenyl)-2-hydroxy-4-oxobut-2-enoic acid (3f) prevent the increase in the interferon-gamma-induced synthesis of quinolinic acid in primary cultures of cultured human peripheral blood monocyte-derived macrophages.

Interferon response heterogeneity: activation of a pro-inflammatory response by interferon alpha and beta. A possible basis for diverse responses to interferon beta in MS.

Interferon gamma (IFN-gamma) stimulates the (pro-inflammatory) type II interferon receptor and is known to exacerbate multiple sclerosis (MS). In contrast, IFN-alpha and IFN-beta are ligands for the (anti-inflammatory) type I interferon receptor and are beneficial in some (but not all) patients with MS. Should IFN-beta elicit a type-II-like pro-inflammatory response, the beneficial effects might be attenuated. These studies were undertaken to test this possibility with the use of quinolinic acid (QUIN) formation as a measure of type II receptor activation. In normal human macrophage cultures, IFN-gamma was the most potent stimulus for QUIN formation. Generally, IFN-beta and IFN-alpha were less potent. However, an unexpected inter-patient variability was observed. In some subjects, IFN-alpha was more potent than IFN-beta. In other subjects, IFN-beta was more potent than IFN-alpha. The present data demonstrate an inter-subject variability for QUIN production following exposure to the interferons. MS patients who demonstrate a pro-inflammatory response to IFN-beta (e.g., increased QUIN) may be less likely to benefit from this therapy.

Altered tryptophan metabolism in mice with herpes simplex virus encephalitis: increases in spinal cord quinolinic acid.

Mice infected with the herpes simplex virus, type-1, developed a paralysis which was associated with increased levels of the neurotoxin quinolinic acid (QUIN). The largest increases in QUIN were observed in the spinal cord with much smaller changes in the rostral forebrain or serum. The time course for the paralysis coincided with the increase in spinal cord QUIN, a maximal 40-fold elevation, at 7-10 days post infection. The time course suggested that the increases in QUIN were due to its local synthesis. Consistent with this possibility, herpes virus-infected mice had increased activities of indoleamine 2,3-dioxygenase and kynurenine hydroxylase (two key enzymes in QUIN formation), when compared to non-infected controls. Since QUIN is formed by activated macrophages, these new data are consistent with QUIN formation as part of the host response to a pathogen whose importance is discussed.

HLA-DR4 influences glial activity in Alzheimer's disease hippocampus.

The importance of inflammatory/immune mechanisms in Alzheimer's disease is supported by evidence that the human leukocyte antigen (HLA)-DR genotype influences risk of the disease, with a protective effect associated with the HLA-DR4 allele. We investigated the influence of the HLA-DR4 allele on glial activity, assessed by quantification of glial fibrillary acidic protein (GFAP), in hippocampal tissue from subjects with Alzheimer's disease. The mean GFAP level was significantly higher in Alzheimer's disease hippocampal specimens lacking the HLA-DR4 allele compared to specimens with similar neuropathological findings that were HLA-DR4 positive. Apolipoprotein E genotype had no influence on GFAP levels. These results indicate that HLA-DR4 may exert a protective influence on Alzheimer's disease via modulation of glial activity.

S-aryl cysteine S,S-dioxides as inhibitors of mammalian kynureninase.

A series of 2-amino-S-aryl cysteine S,S-dioxides have been synthesised and shown to inhibit kynureninase an important enzyme in the biosynthesis of the known excitotoxic moiety quinolinic acid. The most potent of these, 2-amino-5-methyl-S-phenyl cysteine S,S-dioxide 6d, inhibits interferon-gamma induced synthesis of quinolinic acid in human macrophages.

Quinolinic acid and lymphocyte subsets in the intrathecal compartment as biomarkers of SIV infection and simian AIDS.

Cerebrospinal fluid (CSF) samples were collected from monkeys infected with SIVmac251 (SIV) or HIV-1/SIVmac chimeric viruses (SHIV(HXBc2) and SHIV(89.6P)) to investigate quinolinic acid (QUIN) levels in the intrathecal compartment. CSF levels of QUIN were elevated in the SIV-infected monkeys, especially in animals with end-stage disease, and in those infected with pathogenic SHIV(89.6P), but not after infection with the nonpathogenic construct SHIV(HXBc2). QUIN elevations occurred in association with reduced CD4+ and increased CD8+ lymphocytes, cellular alterations that were more pronounced in CSF than in the blood. These findings support the view that the intrathecal compartment provides a unique window on viral infection, and are in keeping with the a priori prediction that QUIN increases primarily in response to more pathogenic viral strains.

Activated human microglia produce the excitotoxin quinolinic acid.

We aimed to determine the relative role of quinolinic acid synthesis in purified human microglia, monocyte-derived macrophages and astrocytes in the human brain following immune stimulation. Microglia and macrophages significantly increased quinolinic acid synthesis from tryptophan following activation by either lipopolysaccharide or interferon-gamma. Quinolinic acid synthesis by individual microglia was heterogeneous, and its production by activated macrophages was approximately 32-fold greater than its microglial synthesis. Quinolinic acid synthesis by astrocytes was undetectable. Microglia may, therefore, be the primary endogenous cell type responsible for quinolinic acid synthesis in the brain parenchyma. However, under pathological conditions which precipitate blood-brain barrier compromise and/or leukocytic infiltration, intracerebral quinolinic acid may be derived chiefly from cells of the peripheral immune system such as activated macrophages.

The regulation of quinolinic acid in human immunodeficiency virus-infected monocytes.

Quinolinic acid (Quin) is thought to underlie cognitive and motor dysfunctions for a variety of neurological disorders. Specifically, in human immunodeficiency virus (HIV)-associated dementia, Quin levels correlate with the degree of neurological dysfunction observed in affected individuals. Since recent data from our laboratories suggest that both HIV-1 infection and activation of brain macrophages are required for the development of neurotoxicity we examined Quin production during virus infection and immune activation. HIV-1 infection of monocytes induced low levels of Quin while lipopolysaccharide (LPS) or interferon-gamma (IFN-gamma) activation of the virus-infected cells elicited 10-fold higher levels. The combined effects of LPS and IFN-gamma for Quin production in HIV-infected monocytes was identical to each factor added alone. Little or no Quin was detected in unstimulated uninfected monocytes. LPS or IFN-gamma activation of uninfected monocytes produced substantially higher levels of Quin than found in similarly stimulated HIV-1-infected monocytes. These results were at variance to the production of tumor necrosis factor-alpha (TNF-alpha). Here, a 2-to 5-fold increase in TNF-alpha levels were observed in culture fluids of LPS-activated HIV-infected cells when compared to similarly stimulated uninfected monocytes. The effect of LPS-induced Quin production by HIV-infected monocytes was not altered by primary human astrocytes. These data suggest that Quin levels seen in HIV dementia are a reflection of macrophage/ microglial activation seen during advanced clinical disease. These findings could help explain, in part, why few HIV-1-infected brain macrophages can give rise to significant neurological impairments.

Effects of 540C91 [(E)-3-[2-(4'-pyridyl)-vinyl]-1H-indole], an inhibitor of hepatic tryptophan dioxygenase, on brain quinolinic acid in mice.

Studies were undertaken to assess the role of the liver in the formation of the neurotoxin quinolinic acid in the brain. A selective and potent inhibitor of hepatic tryptophan 2,3-dioxygenase, 540C91 [(E)-3-[2-(4'-pyridyl)-vinyl]-1H-indole], largely prevented the elevation in mouse brain quinolinic acid resulting from parenteral injection of tryptophan (TRP). In contrast, 540C91 did not affect basal levels of the neurotoxin. Following induction of indoleamine dioxygenase with bacterial lipopolysaccharide, 540C91 was less effective in preventing the TRP-induced elevations in quinolinic acid. The data suggest that kynurenines, formed from tryptophan, by the liver and other extrahepatic organs can give rise to brain quinolinic acid.

The neurotoxin quinolinic acid is increased in spinal cords of mice with herpes simplex virus encephalitis.

Following retroperitoneal, intradermal inoculation of mice with HSV-1, signs of encephalomyelitis (hind-limb paralysis, flaccid tail and loss of bladder control) appeared 6-7 days later. Levels of quinolinic acid (QUIN; determined by gas chromatography with mass-spectrometry), rose approximately 40-fold in mice with encephalomyelitis, primarily in the spinal cord. Live virus could also be grown from homogenates of the affected spinal cords. Time-course studies, demonstrated that the increase in QUIN coincided with the appearance of encephalomyelitis. Large increases in indoleamine dioxygenase activity were observed in spinal cords from the affected mice, suggesting that the QUIN was synthesized within the spinal cord. It is, therefore, possible that QUIN may contribute to the pathology of HSV-1 encephalomyelitis.

Neurotoxin quinolinic acid is selectively elevated in spinal cords of rats with experimental allergic encephalomyelitis.

Experimental allergic encephalomyelitis (EAE) is an autoimmune, animal model of multiple sclerosis (MS) in which demyelination and paralysis are evident. Quinolinic acid (QUIN) is a neurotoxin and endogenous N-methyl-D-aspartate receptor agonist formed from tryptophan. The role of neurotoxins in general and QUIN in particular in EAE or MS is unknown. Lewis rats inoculated with myelin basic protein developed signs of EAE by day 12, were killed, and their tissues assayed for QUIN by gas chromatography with mass spectrometry. QUIN levels were significantly elevated in the more caudal regions of the spinal cords of animals with EAE. Brain, serum, and liver levels of QUIN were not altered. In a similar manner, QUIN in mylin basic protein-injected, asymptomatic animals was not different from control animals. The time course for QUIN was similar to the neurological signs of the disorder; however, the initial elevation in QUIN occurred before the appearance of behavioral signs. Last, treatment with the glucocorticoid dexamethasone prevented both the signs of EAE and the elevation in spinal cord QUIN. It is not known whether QUIN contributes to the paralysis in EAE. However, if QUIN is pathogenic in EAE this finding could have therapeutic implications for MS.

Differences in the reserpine-sensitive storage in vivo of 1-methyl-4-phenylpyridinium in rats and mice may explain differences in catecholamine toxicity to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine.

Administration of reserpine, an inhibitor of vesicular catecholamine storage, differentially reduced the accumulation of MPP+ formed from MPTP in rats and mice. The effects were most pronounced in the adrenal gland for either species. In rats, reserpine decreased striatal and hippocampal MPP+ levels while in mice reserpine did not affect the disposition of MPP+ in the striatum but decreased hippocampal MPP+. The data suggest that mice may be more sensitive to the toxicant because less striatal MPP+ appears to be stored in the reserpine-sensitive storage vesicle.

Elevation of the neurotoxin quinolinic acid occurs following spinal cord trauma.

Excitatory amino acid neurotoxicity and the inflammatory response are suspected as mediators of some of the pathological sequelae occurring as a result of spinal cord injury. Here we report temporal and regional increases of the NMDA receptor agonist, quinolinic acid (QUIN), in an experimental model of spinal contusion injury. These changes occurred at a time when the blood-brain barrier is known to be dysfunctional and the activation state and density of microglia and macrophages are increased. Thus, alterations in tissue QUIN levels may occur as a result of secondary activation of CNS inflammatory cells or from peripherally derived sources across a damaged blood-brain barrier.

A radiometric assay for kynurenine 3-hydroxylase based on the release of 3H2O during hydroxylation of L-[3,5-3H]kynurenine.

A rapid and sensitive assay for kynurenine 3-hydroxylase (KH) has been developed. This radiometric assay is based on the enzymatic synthesis of tritiated water from L-[3,5-3H]kynurenine during the hydroxylation reaction. Radiolabeled water is quantified following selective adsorption of the isotopic substrate and its metabolite with activated charcoal. The assay is suitable for detecting 0.1 pmol enzyme activity per minute per milligram protein in tissues displaying low levels of the enzyme. The amount of water produced in the reaction, as calculated from the tritium released, was stoichiometric with the 3-hydroxykynurenine product detected by HPLC. Rat liver KH was characterized by cofactor specificity and kinetic parameters. NADPH was preferred over NADH as coreductant in the reaction. Tetrahydrobiopterin was not a cofactor. The tissue distribution of KH activity in the rat suggested that the majority of active enzyme is located in liver and kidney. Detectable amounts were found in several other tissues, including brain which had low but significant levels of activity in every region assayed.

Slow changes of tyrosine hydroxylase gene expression in dopaminergic brain neurons after neurotoxin lesioning: a model for neuron aging.

Slow neuron regression develops during the adult phase of life in select brain systems of mammals. We describe a model in adult rats that resolves several phases in a slow atrophic process that differentially influences levels of mRNA and protein for tyrosine hydroxylase (TH). Responses of striatal dopaminergic markers to 6-hydroxydopamine (6-OHDA) lesions in rats indicated that the striatal terminals maintained TH protein, despite greater than 3-fold loss of TH mRNA in the substantia nigra pars compacta (SNC) cell bodies whose axons project to the striatum. The loss of TH mRNA/cell was progressive up to 9 months, whereas SNC cell body shrinkage stabilized by 3 months post-lesioning. Consideration of possible mechanisms in protein turnover motivated a search for PEST motifs in the TH of rats and other vertebrates that could be a point of regulation by altering the rate of TH protein turnover.

Striatal responses to decortication. I. Dopaminergic and astrocytic activities.

Unilateral ablation of the frontal cortex induced 30%-50% decrease of dopamine (DA) concentration in the ipsilateral striatum at 10 and 27 days after lesioning. There were increased ratios of dihydroxyphenylacetic acid (DOPAC): DA and homovanillic acid (HVA): DA by 20%-60% at 10 days post-lesioning, which suggest compensatory increases of DA metabolism. While no change in total striatal tyrosine hydroxylase (TH) polypeptide concentration was found at any post-lesion time, TH catalytic activity was decreased slightly (-25%) at 10 days. Among individual rats, at 3, 10 and 27 days post-lesioning, striatal DA concentration was inversely related to striatal glial fibrillary acidic protein (GFAP) concentration, a marker of astrocytic activity. The loss of DA was observed whether or not DA was normalized to striatal protein, which suggests that DA loss cannot be simply attributed to increased astrocytic proteins. These data suggest reciprocal relationships between the extent of astrocytic reactions after cortical deafferentation and striatal DA loss, which could involve local remodelling without primary damage to the nigro-striatal terminals.

Pargyline and gamma-butyrolactone enhance tyrosine hydroxylase immunostaining of nigrostriatal axons.

Administration of pargyline or gamma-butyrolactone enhanced immunostaining of tyrosine hydroxylase immunoreactive axons in the striatum. The qualitative characteristics of the enhancement were compound dependent and the enhancement was not associated with an increase in the amount of tyrosine hydroxylase within the striatum. These results demonstrate the pharmacological enhancement of immunostaining for neurotransmitter synthetic enzymes in the absence of increased synthesis.