Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The kynurenine pathway (KP) is a major route for the conversion of tryptophan, yielding l-kynurenine (l-KYN) and ultimately several other metabolites, called kynurenines, which are derived directly or indirectly from l-KYN (Figure 5-1 ). The metabolic cascade was originally known to be a source of nicotinamide adenine dinucleotide (NAD) and nicotinamide adenine dinucleotide phosphate (NADP), two coenzymes of basic cellular processes, and only later it was discovered that some of its metabolites could exhibit neuromodulatory actions. Interest in the research of KP emerged as it turned out that two metabolites of the pathway, quinolinic acid (QUIN) and kynurenic acid (KYNA) exhibit activity at glutamate receptors (Stone and Perkins, 1981; Perkins and Stone, 1982). QUIN was shown to inhibit N-methyl-d-aspartate (NMDA) receptors, whereas KYNA proved to be a broad-spectrum antagonist of excitatory amino acid (EAA) receptors with particular high affinity to the glycine-coagonist site of the NMDA receptor (Kessler et al., 1989). Subsequently, studies from several laboratories have clarified the role of kynurenines in the brain function under physiological and pathological conditions. This has led to the identification of potential new drug targets for various neurological disorders.

Figure 5-1
figure 1_5

Chemical structure of l-kynurenine and neuroactive kynurenines

Full size image

2 l-Kynurenine and Neuroactive Metabolites of the Kynurenine Pathway

2.1 l-Kynurenine

l-KYN (Figure 5-1 ) is a major compound of the KP, serving as a source for the synthesis of all the other metabolites of the pathway. l-KYN is present in the blood, brain, and peripheral organs in low micromolar concentrations, and gets transported through the blood–brain barrier by the neutral amino acid carrier (Fukui et al., 1991). Although l-KYN does not directly influence neuronal function, systemic or intracerebral administration of it decreases blood pressure (Lapin, 1976) and evokes convulsions (Lapin, 1978, 1981, 1982; Pinelli et al., 1984), probably by getting converted to its neuroactive metabolites.

2.2 Kynurenic Acid

KYNA (Figure 5-1 ) was long known to be a side product of tryptophan metabolism but no particular biological function was assigned to it until more recently when in neurophysiological experiments it was shown to inhibit neurons (Perkins and Stone, 1982). Subsequently, KYNA was recognized as a broad-spectrum antagonist of ionotropic EAA receptors. Since EAA receptor activation takes place in a variety of pathological states, KYNA has been tested as a neuroprotective agent. Indeed, at high concentrations, KYNA was shown to block excitotoxic damage and seizures induced by QUIN in rats (Foster et al., 1984) and to protect against various conditions like ischemia, traumatic brain injury (Germano et al., 1987; Andiné et al., 1988; Hicks et al., 1994; Salvati et al., 1999). Intracerebroventricular administration of KYNA was shown to dose-dependently evoke characteristic behavior in the rat: increased stereotypy and ataxia (Vécsei and Beal, 1990a, 1991). At lower concentrations, KYNA was recognized to act as a competitive antagonist at the glycine site of the NMDA receptor (IC50 ≅ 8 μM) (Kessler et al., 1989) and as a noncompetive blocker at the α7-nicotonic receptor (IC50 ≅ 7 μM) (Hilmas et al., 2001). These receptors could be the major sites of action of KYNA in the brain and thus endogenous KYNA could modify glutamatergic and cholinergic neurotransmission. Indeed, the level of KYNA seems to determine the vulnerability of the brain against excitotoxins, because intraperitoneal (i.p.) injection of amphetamine – which leads to reduction in the brain concentration of KYNA-potentiated quinolinate but not kainate excitotoxicity, and readjusting the level of KYNA to control levels by pharmacological manipulation of the KP, restored the vulnerability (Poeggeler et al., 1998; Rassoulpour et al., 2002). Furthermore, endogenous KYNA might influence glutamatergic neurotransmission presynaptically because modest elevations of KYNA in the rat striatum in vivo and in synaptosomes in vitro are able to inhibit glutamate release (Carpenedo et al., 2001).

Generation of kynurenine aminotransferase knockout (KAT II KO) mice – which have reduced brain KYNA levels early in development (<28 days) – has provided a useful tool for studying the role of KYNA in brain function and underlined the importance of endogenous KYNA in modulating glutamatergic and cholinergic neurotransmission. Intrastriatal injection of EAA receptor agonist QUIN induced significantly larger lesion in KAT II KO mice compared with wild-type mice (Sapko et al., 2003). Furthermore, the observation that KAT II KO mice have increased activity of α7-nicotonic receptor has proved the previous hypothesis that endogenous KYNA is an important regulator of α7-nicotonic receptor activity (Alkondon et al., 2003). The changes observed in the KAT II KO mice are evident early in development (<28 days), when cerebral KYNA levels are significanly decreased but tend to be reverted as the KYNA levels return to normal levels. These observations suggest that changes in endogenous KYNA levels have profound effects on the function of glutamatergic and cholineric receptors.

2.3 Quinolinic Acid

QUIN (Figure 5-1 ) is present in the brain in concentrations similar to that of KYNA (50–100 nmol). It has pronounced effects on neuronal activity being an agonist at the NMDA receptor (Stone and Perkins, 1981), preferentially activating the NR2A and NR2B NMDA receptor subtypes (de Carvalho et al., 1996). Besides acting at the NMDA receptor, QUIN produces toxic free radicals (Rios and Santamaria, 1991). Due to the compound's excitotoxic and free-radical-generating property, injection of QUIN into the rat striatum leads to excitotoxic damage (Schwarcz et al., 1983) duplicating the neurochemical features of Huntington's disease (HD) (Beal et al., 1986). In addition, long-term exposure to submicromolar concentrations of QUIN results in neuronal death in vitro (Whetsell and Schwarcz, 1989). In pathological states, like neuroinflammatory diseases, dramatic increases of its level in the brain could occur which could result in neuronal damage.

2.4 3-Hydroxykynurenine

No important physiological function in the brain has so far been assigned to 3-hydroxykynurenine (3-HK) (Figure 5-1 ), but in the primate lenses it could act as major UV filter together with its glucoside derivative, l-KYN and 4-(2-amino-3-hydroxyphenyl)-4-oxobutanoic acid O-β-d-glucoside and may be useful in protecting the retina from UV radiation (Vazquez et al., 2002). However, the autooxidation of these metabolites has been proposed in the processes leading to opacification of the lens and cataract formation (Chiarugi et al., 1999). In the brain, 3-HK can cause neuronal death in cell cultures by generating toxic free radicals. 3-HK is present in nanomolar concentrations in the mammalian brain, but under pathological conditions similarly to QUIN its level could increase dramatically reaching the micromolar range (0.8–1.2 μM) (Eastman and Guilarte, 1989). Chronic exposure of neuronal cultures to these levels of 3-HK could cause neuronal death (Okuda et al., 1996). 3-HK should get into neurons to induce toxicity, because the inhibition of its uptake by large neutral amino acid prevents from neuronal death (Okuda et al., 1998). 3-HK potentiates QUIN toxicity in the rat striatum, which suggests that these metabolites may act in concert to induce neuronal damage (Guidetti and Schwarcz, 1999).

3 Enzymes of the Kynurenine Pathway

Figure 5-2 summarizes the enzymes and metabolites of the KP. The first and rate-limiting step in the formation of l-KYN is the conversion of l-tryptophan to formyl kynurenine, which is catalyzed by two distinct heme-containing enzymes. In peripheral tissues and particularly in the liver, tryptophan dioxygenase (TDO; tryptophan pyrrolase; EC 1.13.1.2.) is mainly responsible for yielding l-KYN but in most mammalian organs including intestine, lung, epididymis, placenta, central nervous system, reticuloendothelial system, indoleamine 2,3-dioxygenase (IDO; EC 1.13.11.42) is able to convert tryptophan to formyl kynurenine. In the second step, formyl kynurenine is rapidly metabolized into l-KYN by formamidase, an enzyme abundant in most mammalian organs (Mehler and Knox, 1950).

Figure 5-2
figure 2_5

The kynurenine pathway

Full size image

TDO is able to metabolize l- but not d-tryptophan (Schutz et al., 1972), whereas IDO not only acts on l-tryptophan, but it can also cleave the indole ring of the d-tryptophan, l- or d-5-hydroxytryptophan, indoleamins like 5-HT, tryptamine, and melatonine (Hirata and Hayaishi, 1975; Yoshida and Hayaishi, 1987). The gene encoding TDO contains glucocorticoid-responsive elements (Danesh et al., 1983, 1987) which explains that glucocorticoid administration increases TDO formation in the liver, which causes a decrease in blood tryptophan levels with l-KYN accumulation. The transcription of the gene of IDO is under tight immunological control, and it contains two interferon-stimulated response elements (ISRE) and at least one gamma-interferon-activated sequence (GAS) (Dai and Gupta, 1990; Tone et al., 1990). Interferons and proinflammatory cytokines stimulate (Carlin et al., 1989; Taylor and Feng, 1991), whereas other cytokines and growth factors, like interleukin-4 or TGF-beta, inhibit IDO expression and activity (Musso et al., 1994; Yuan et al., 1998).

l-KYN is the key player of the pathway serving as a substrate of several enzymes: kynurenine 3-hydroxylase (EC 1.14.13.9; yielding 3-HK), kynureninase (EC 3.7.1.3; yielding anthranilic acid), and kynurenine aminotransferases (KATs; yielding KYNA). Kynurenine 3-hydroxylase is present in the liver, placenta, spleen, kidney, and brain (Erickson et al., 1992) and requires NADPH and molecular oxygen for the conversion of l-KYN to 3-HK. The enzyme has a high affinity for its substrate (K m ≅ 1 μM), implicating that it converts most of the available l-KYN under physiological conditions.

Kynureninase has been characterized from rodent liver, kidney, and spleen (Kawai et al., 1988). It is a pyridoxal phosphate-dependent enzyme that converts l-KYN and 3-HK into l-alanine and anthranilic or 3-hydroxyanthranilic acid, respectively. It has been shown that the affinity of the enzyme is tenfold higher for 3-HK than for l-KYN (K m for l-KYN: 250 μM, K m for 3-HK: 25 μM; Alberati-Giani et al., 1996; Toma et al., 1997).

3-Hydroxyanthranilate oxygenase (3-HAO, EC 1.13.11.6) has been purified from mammalian liver or kidney and it requires ferrous ions (Fe2+) and sulfhydryl groups for its activity (Long et al., 1954; Decker et al., 1961; Koontz and Shiman, 1976). 3-HAO cleaves the benzene ring of 3-hydroxyanthranilate into α-amino-β-carboxymuconate-ε-semialdehyde, an unstable compound, which nonenzymatically gets transformed into QUIN. A portion of α-amino-β-carboxymuconate-ε-semialdehyde is metabolized by α-amino-β-carboxymuconate-ε-semialdehyde decarboxylase (EC 4.1.1.45) to produce picolinic acid or aminomuconic acids.

3-HAO has been shown to be present in the brain (Foster et al., 1986), and increased activity of the enzyme has been detected in various pathological states, like in the striatum of HD patients (Schwarcz et al., 1988a), in gerbil hippocampus after global ischemia (Saito et al., 1993), and in epileptic rats (Du et al., 1993). QUIN is metabolized by quinolinic acid phosphoribosyltransferase (QPRT; EC 2.4.2.19) yielding nicotinic acid mononucleotide and subsequent degradation products, including NAD+. Since QPRT has a very low activity, the levels of QUIN in the extracellular space are essentially determined by its rate of synthesis.

KYNA is converted from l-KYN in the brain by two distinct KATs (Okuno et al., 1991; Guidetti et al., 1997). Arbitrarily termed KAT I and KAT II, these enzymes have substantially different pH optimum and substrate specificity. KAT I (EC 2.6.1.14; also called glutamine transaminase K) has an optimal pH of 9.5–10, whereas KAT II has a maximal activity in the neutral pH. KAT I prefers pyruvate as a cosubstrate and it is potently inhibited by glutamine. KAT II shows no preference for pyruvate, and it is not sensitive to inhibition by glutamine. These differences suggest that KAT II might be responsible for most part of the KYNA synthesis in the brain. Lesion and pharmacological studies have also confirmed that in most brain regions KYNA derives primarily from KAT II activity (Guidetti et al., 1997). Under pathological conditions, however, such as after exposure to mitochondrial toxins, a massive increase in KAT I immunostaining has been observed in the neurons in the affected brain areas (Csillik et al., 2002; Knyihár-Csillik et al., 1999). Both KATs have high affinity for their substrate (K m in the millimolar range), suggesting that l-KYN bioavailability determines the rate for KYNA biosynthesis. Indeed, in experiments, when rat brain slices or human astrocyte cultures are exposed to l-KYN, the levels of KYNA increase linearly with l-KYN availability (Turski et al., 1989; Kiss et al., 2003).

No catabolic enzyme or cellular re-uptake system of KYNA exists in mammals (Turski et al., 1989); so KYNA is extruded from the brain by a probenecid-sensitive carrier system (Moroni et al., 1988). More recently, QUIN has been also shown to exit the brain by the same carrier system (Morrison et al., 1999).

All of the enzymes of the KP are detectable in the brain, but their activities are much lower than in the peripheral organs (Stone, 1993). KP enzymes are predominantly localized in glial cells according to immunocytochemical, lesion, and molecular biological studies (Guidetti et al., 1995; Schwarcz et al., 1996; Guillamin et al., 2001).

Astrocytes seem to produce most part of KYNA in the brain (Guillamin et al., 2001), whereas microglial cells are the primary source of metabolites of the QUIN branch of the pathway (Lehrmann et al., 2001).

4 Alterations of the KP under Pathological Conditions

Elucidation of the importance of the KP in brain function has facilitated extensive research, investigating the alterations of the pathway in various neurological disorders. In the literature, a recent extensive review is available on this topic (Stone, 2001). In a great number of diseases – especially in neuroimmunological disorders – dramatic increases in the level of QUIN have been detected, suggesting its role in the pathological processes. Whereas, in diseases with cognitive alterations, elevated levels of cerebral KYNA have been shown which could contribute to cognitive defects by inhibiting NMDA receptor function.

4.1 Diseases with Increased Cerebral QUIN Levels

Infection of the CNS with viruses, bacteria, parazites: Significant increases in QUIN levels in the CNS have been described in patients with AIDS–dementia complex (Heyes et al., 1991, 1998), Lyme borelliosis (Halperin and Heyes, 1992), in mice infected with Herpes simlex virus type 1, cerebral malaria, and Toxoplasma gondii (Reinhard, 1998; Sanni et al., 1998; Fujigaki et al., 2002), in macaques with poliovirus infection and septicaemia (Heyes and Lackner, 1990; Heyes et al., 1992). Reinhard et al. (1994) listed many other infectious diseases in which alterations of the KP have been described.

Multiple sclerosis (MS): Experimental allergic encephalomyelitis is an autoimmune inflammatory disorder and serves as an animal model of human MS. The levels of neurotoxic kynurenine metabolites, 3-HK and QUIN, are elevated in the spinal cord of affected animals (Flanagan et al., 1995; Chiarugi et al., 2001). Since QUIN has been shown to be toxic to oligodendrocytes (Cammer, 2001), elevated levels of QUIN might contribute to the pathology of the disease.

Huntington's disease: QUIN has long been hypothesized to play an important role in HD, because intrastriatal injection of QUIN duplicates many of the distinct neuropathological features of the striatum of patients with HD (Schwarcz et al., 1983; Beal et al., 1996). However, no changes in QUIN levels in tissue samples have been detected in HD patients dying after a prolonged illness (Reynolds et al., 1998; Schwarcz et al., 1988b). Of potential interest is the finding that elevated cortical and striatal content of 3-HK and QUIN has been shown in early grade HD, suggesting a causal role of QUIN in nerve cell loss in HD (Guidetti et al., 2000, 2003).

Parkinson's disease (PD): 3-HK has been shown to be significantly increased in the putamen and substantia nigra in patients with PD, which could play a causal role in the neuronal loss in PD (Ogawa et al., 1992).

Ischaemia: Dramatic increases in cerebral QUIN levels have been observed several days after global ischemia in a gerbil model of the disease (Heyes and Nowak, 1990; Saito et al., 1993). The changes have been observed in brain areas exposed to ishemia, but not in areas with an uninterrupted blood supply (Saito et al., 1992).

Traumatic brain/spinal cord injury: Both in humans and in the animal model of the disease, massive increases (50-fold) in the level of QUIN in the CSF (Sinz et al., 1998) and in affected areas have been shown (Blight et al., 1995, 1997).

4.2 Diseases with Increased Cerebral KYNA Levels

Alzheimer's disease (AD): Increased level of KYNA has been measured in the putamen and caudate nucleus of AD patients, which could be responsible for the cognitive deficits seen in AD patients (Baran et al., 1999).

Down's syndrome: The cortical levels of KYNA have been shown to be elevated in the postmortem specimens of Down's syndrome patients that could play a causal role in the impaired memory and learning defects seen in these patients (Baran et al., 1996).

Schizophrenia: Elevated levels of KYNA have been detected in the prefrontal cortex (Brodmann area 9), but not in other cortical areas of schizophrenic patients (Schwarcz and Pellicciari, 2002). Increased KYNA levels may contribute to the hypofunction of glutamatergic neurotransmission that plays a critical role in schizophrenia (Tamminga, 1998).

5 Therapeutic Approaches based on Kynurenines

After KYNA was recognized as a broad-spectrum antagonist of EAA receptors, its neuroprotective potential has been investigated in a variety of disorders. KYNA itself poorly enters the brain due to its polar structure and lack of efficient transport across the blood–brain barrier, but if systematically given in high doses, it protects against anoxia (Simon et al., 1986), ischemia (Germano et al., 1987; Andiné et al., 1988; Salvati et al., 1999), traumatic brain injury (Hicks et al., 1994), and antagonizes the toxic effects of QUIN (Foster et al., 1984). There are several strategies to generate neuroprotective drugs that have superior therapeutic potency to KYNA.

One strategy to exploit the therapeutic potential of KYNA is to develop chemically related drugs with better bioavailability and higher potency on the glycine site of the NMDA receptor.

The second approach uses prodrugs of KYNA or its analogs, which readily penetrate the blood–brain barrier, and are hydrolyzed in the CNS to form active compounds.

Third way is manipulating the KP by administering compounds that block the activity of the KP enzymes. With this strategy, one can shift l-KYN metabolism to the KYNA or QUIN branch with the aim to inhibit or excite EAA receptors, respectively.

Recently, several reviews have been published on the therapeutic alternatives based on kynureninergic manipulations (Stone, 2000, 2001; Schwarcz and Pellicciari, 2002; Stone and Darlington, 2002).

5.1 KYNA Analogs

Several attempts have been made to use the KYNA structure to develop NMDA glycine site antagonists with better bioavailability and higher potency on the glycine site of the NMDA receptor. Substitution of KYNA with halogen atoms (7-chlorokynurenic acid and 5,7-dichlorokynurenic acid; Baron et al., 1990), replacement of the 4-hydroxy group of KYNA with amido substituents (MDL 100,748, L-689,560; Baron et al., 1992; Leeson et al., 1992), substitution with a phenyl or more complex lipophil groups at position 3 of the KYNA nucleus (MDL 104,653, L-701,324; Kulagowski et al., 1994; Bristow et al., 1996), or replacment of the six-membered nitrogen containing ring by a five-carbon ring (GV150526A or gavestinel; Glaxo, 1993) provided compounds that have superior neuroprotective abilities compared with KYNA. Gavestinel has already entered clinical trials but the Glycine Antagonist in Neuroprotection (GAIN) Americas trial, a randomized, double-blind placebo-controlled phase III trial has failed to show any benefit of gavestinel treatment in acute ischemic stroke patients (Sacco et al., 2001).

5.2 Prodrugs of KYNA and of KYNA Analogs

l-KYN readily penetrates the blood–brain barrier by the neutral amino acid transporter (Fukui et al., 1991) and serves as a precursor for the neuroprotectant KYNA in the brain. This explains that systemic administration of l-KYN has been shown to protect against cerebral ischemia or local injection of NMDA in neonatal rats (Nozaki and Beal, 1992) and to counteract pentylenetetrazol- and NMDA-induced seizures in mice (Vécsei et al., 1992), but only modest effects of l-KYN treatment havebeen observed on kainate-induced seizures in rats (Vécsei et al., 1990b). However, it should be noted that l-KYN gets converted not only to KYNA but also to 3-HK and QUIN which poses considerable limitations to l-KYN therapy and explains the poor efficacy of l-KYN treatment since systemical administration of l-KYN is not capable of selectively increasing the levels of the neuroprotectant KYNA. For this reason, more potent KYNA precursors have been searched that are not getting metabolized in the QUIN branch of the pathway.

l-4-Chlorokynurenine or 4,6-dichlorokynurenines are potent precursors that meet the previously mentioned criteria of not getting metabolized to QUIN. These compounds are converted to two NMDA glycine site antagonists, namely 7-chlorokynurenic acid and 5,7-dichlorokynurenic acid, respectively (Hokari et al., 1996). 4-Chlorokynurenine is also metabolized to 4-chloro-3-hydroxyanthranilic acid, which is a potent and selective inhibitor of 3-hydroxyanthranilic acid oxygenase, and thus the administration of 4-chlorokynurenine causes a reduction of QUIN synthesis besides inhibiting NMDA receptors. It penetrates into the brain easier than its metabolite7-chlorokynurenic acid, and provides several additional benefits compared with other glycine site NMDA receptor antagonists: it preferentially gets metabolized in brain areas where neurodegeneration takes place, allowing lower dosage of the drug (Lee and Schwarcz, 2001). 4-Chlorokynurenine diminishes brain damage induced by focal application of QUIN and malonate into the rat striatum and hippocampus (Guidetti et al., 2000; Wu et al., 1997) and inhibits convulsions and neurotoxicity after systemic application of kainate (Wu et al., 2002). The conversion of 4-chlorokynurenine to 7-chlorokynurenic acid has also been observed in human astrocytes, implicating that 4-chlorokynurenine therapy might be used in humans in diseases of excitotoxic origin (Kiss et al., 2003).

Another prodrug strategy uses d-glucose and d-galactose conjugates of KYNA and its analogs for allowing better penetration of these drugs into the brain, assuming that the conjugates are recognized by the glucose transporter facilitating their entry and get hydrolyzed in the brain to release the active compound (Battaglia et al., 2000a; Bonina et al., 2000). Systemic administration of 7-chlorokynurenic acid-d-glucopyranos-6′-yl ester or 7-chlorokynurenic acid-d-glucopyranos-3′-yl ester has provided protection against seizures induced by NMDA in mice.

A glucosamine–kynurenic acid conjugate has also been synthesized which induced stereotypy and ataxia in freely moving rats and reduced the evoked excitatory postsynaptic potentials in rat motor cortical slices after intracerebroventricular administration (Füvesi et al., 2004). Further studies are under way to characterize the bioavailability and therapeutic efficacy of the glucosamine–kynurenic acid conjugate after systemic administration.

5.3 KP Enzyme Inhibitors

5.3.1 Manipulation of the KP with the Aim to Elevate the Cerebral Level of KYNA

Modulation of the KP by inhibiting enzymes of QUIN synthesis is a rational approach to divert the kynurenine metabolism toward the neuroprotective KYNA. This therapy would be particularly useful in clinical situations when excessive EAA receptor activation takes place. The administration of kynurenine 3-hydroxylase inhibitors is the most rational approach to evoke marked elevation in cerebral KYNA levels with concomitant attenuation of 3-HK and QUIN formation. While application of compounds inhibiting kynureninase and 3-hydroxyanthranilic acid oxygenase – enzymes that act downstream from the 3-HK – has limited therapeutic potency, these drugs cause an increase in 3-HK levels with modest KYNA elevations.

Kynurenine 3-hydroxylase inhibitors: The first drug reported to exhibit kynurenine 3-hydroxylase inhibiting activity was nicotinylalanine (Figure 5-3 , Moroni et al., 1991), which is an analog of l-KYN. Subsequently, analogs with more potency and higher selectivity have been synthesized: meta-nitrobenzoylalanine (mNBA) is 1,000 times more active and by far more selective than nicotinylalanine (Figure 5-3 , Pellicciari et al., 1994). Modification of the aromatic ring region and the side chain yielded (R,S)-3,4-dichlorobenzoylalanine (PNU 156561, formerly known as FCE 288833A), which showed further improvements in its potency and selectivity, compared with mNBA (Figure 5-3 , Speciale et al., 1996). The screening of a library of sulfonamides led to the identification of 3,4-dimethoxy-N-[4-(3-nitrophenyl)thiazol-2-yl]benzenesulfonamide (IC50 = 37 nM, Ro-61-8048), which is currently the most potent noncompetitive inhibitor of the kynurenine 3-hydroxylase (Figure 5-3 , Röver et al., 1997).

Figure 5-3
figure 3_5

Chemical structure of selected kynurenine 3-hydroxylase inhibitors

Full size image

Systemic administration of kynurenine 3-hydroxylase inhibitors results in inhibition of maximal electroshock-induced seizures in rats and audiogenic seizures in DBA/2 mice (Russi et al., 1992; Carpenedo et al., 1994) and provides protection in a rat focal ischemia and a gerbil global ischemia model (Cozzi et al., 1999). Furthermore, these drugs significantly reduce blood and brain accumulation of QUIN in immunostimulated mice, suggesting that the application of kynurenine 3-hydroxylase inhibitors may be of benefit to the patient with neuroimmunological disorders (Chiarugi and Moroni, 1999b).

Kynureninase inhibitors: These compounds inhibit the QUIN branch of the pathway downstream of 3-HK, which explains the limited therapeutic potency of these drugs compared with the kynurenine 3-hydroxylase inhibitors. Application of these drugs also results in accumulation of 3-HK in the brain besides elevating cerebral KYNA levels. ortho-Methoxybenzoylalanine (oMBA) is a prototype inhibitor of kynurenase, which is chemically related to mNBA but preferentially inhibits kynureninase (Pellicciari et al., 1994). Administration of oMBA results in an increase in the amount of KYNA in the brain in vivo and antagonizes audiogenic convulsion in DBA2 mice (Carpenedo et al., 1994; Chiarugi et al., 1995). Although oMBA is a selective inhibitor of kynureninase in vitro, it also inhibits 3-hydroxyanthranilic acid oxygenase in vivo, making it difficult to study the metabolic effects of kynureninase inhibitors in vivo (Chiarugi and Moroni, 1999a).

3-Hydroxyanthranilic acid oxygenase inhibitors: Halogenated substrate analogs, such as 4-chloro-3-hydroxyanthranilic acid, were the first potent, selective, and competitive blockers reported to inhibit 3-hydroxyanthranilic acid oxygenase (Todd et al., 1989; Walsh et al., 1991); but more recently, dihalogenated analogs of 3-hydroxyanthranilic acid have also been described with even higher potency and selectivity (Linderberg et al., 1999). These drugs reduce functional deficits in animals exposed to experimental spinal cord injury and reduce QUIN accumulation in the brain of traumatized animals (Blight et al., 1995) and in the blood and brain of immunoactivated mice (Saito et al., 1994).

5.3.2 KP Enzyme Inhibition with the Aim to Decrease Cerebral KYNA Levels

Since certain cognition enhancers have been shown to decrease KYNA levels in the brain (Poeggeler et al., 1998; Rassoulpour et al., 1998) and others to prevent the KYNA antagonism of the NMDA receptor (Pittaluga et al., 1995), one might hypothesize that KYNA plays an important role in cognitive processes and pharmacological inhibition of cerebral KYNA levels might be a useful strategy to produce cognition enhancer drugs. Furthermore, the cognitive deficits might be attenuated with the use of these drugs in diseases with elevated cerebral KYNA levels such as schizophrenia, Alzheimer's disease, or Down's syndrome.

KAT inhibitors: KAT II is responsible for producing large part of KYNA present in the brain that makes it a prime target for influencing cerebral KYNA levels. α-Aminoadipate, quisqualate, DL-5-bromocriptine, and certain metabotropic glutamate receptor agonists, such as l-(+)-2-amino-4-phosphonobutyric acid (l-AP4), 4-carboxy-3-hydroxyphenylglycine (4C3HPG), and l-serine-o-phosphate (l-SOP) selectively block KAT II in vitro (Battaglia et al., 2000b). However, all KAT II inhibitors known so far influence other cellular processes besides acting on the KP, rendering it difficult to characterize the behavioral consequences of decreased KYNA content in the brain. Synthesis of more selective KAT II inhibitors is essential to study the importance of KYNA in the brain function.

6 Conclusion

The discovery of the importance of kynurenines in brain function under physiological and pathologic conditions has led to the identification of potential new drug targets exploiting the therapeutic potential of the pathway. Some of these compounds proved to provide neuroprotection in animal models of various human diseases, which holds promise that their effectiveness will translate to the clinic in the future.