The Involvement of Kynurenine Pathway in Neurodegenerative Diseases

Abstract

Background: A growing body of evidence has shown the involvement of the kynurenine pathway (KP), the primary route of tryptophan (TRP) catabolism, in the pathophysiology of neuropsychiatric disorders.

Objective: The study aims to provide a comprehensive and critical overview of the clinical evidence on the KP involvement in the pathophysiology of Alzheimer’s disease (AD) and Parkinson's disease (PD), discussing therapeutic opportunities.

Methods: We searched for studies investigating KP metabolites in human subjects with AD and/or PD.

Results: Postmortem studies showed altered levels of KP metabolites in the brain of AD and PD patients compared with controls. Cross-sectional studies have reported associations between peripheral levels (serum or plasma) of KP metabolites and cognitive function in these patients, but the results are not always concordant.

Conclusion: Given the emerging evidence of the involvement of KP in the pathophysiology of neuropsychiatric/neurodegenerative diseases and promising results from preclinical pharmacological studies, a better understanding of the KP involvement in AD and PD is warranted. Future longitudinal studies are needed to define the direction of the observed associations and specific therapeutic targets within the KP.

1. INTRODUCTION

One of the consequences of the increased life expectancy worldwide is the growing incidence of age-related diseases, mainly neurodegenerative diseases [1], such as Alzheimer’s disease (AD) and Parkinson's disease (PD) [2, 3]. Alzheimer’s disease (AD) is the most frequent cause of dementia, corresponding to 60-70% of the cases [4, 5]. The number of people with dementia is expected to triple by 2050 (from approximately 57 million to 152 million people worldwide), mainly due to population growth and aging, varying according to the world region [6]. A similar scenario is projected for PD, with its incidence expected to double until 2030 [7].

Because of the limited efficacy of the current treatments, neurodegenerative diseases have a major public health impact. Therefore, there is a critical need to identify and validate newtherapeutic targets for these conditions. However, the development of pharmacological strategies targeting the central nervous system (CNS) has one of the lowest success rates in the pharmaceutical industry [8].

A growing body of evidence has shown the involvement of the kynurenine pathway (KP), the primary route of tryptophan (TRP) catabolism, in the pathophysiology of neuropsychiatric disorders [9, 10]. Previous studies have demonstrated, for example, changes in the central and peripheral levels of KP metabolites in AD and PD [10-13]. The KP is regulated by immune/inflammatory mechanisms, and, as expected, its metabolites change with aging [14]. Inflammatory mediators upregulate the expression of the enzyme indoleamine 2, 3-dioxygenase (IDO) that catalyzes a rate-limiting step of KP [15]. Immune and/or inflammatory mechanisms have also been implicated in the pathophysiology of major psychiatric disorders, including mood disorders and neurodegenerative diseases [16-18]. The aging process itself is associated with enhanced inflammatory response, a phenomenon commonly referred to as ‘inflammaging’ [14]. Indeed, inflammation is a common factor in aging and neurodegenerative diseases. Several endogenous signals, such as extracellular ATP, β-amyloid, tau, and α-synuclein, among others originating from dysfunctional cells, act as damage-associated molecular patterns (DAMPs), inducing constant activation of the immune system and, as a consequence, a chronic low-grade inflammation. Besides an increase in the production of these DAMPs or stress-related molecules, the mechanisms (e.g., autophagy) to control or eliminate them decline with aging [14, 19, 20].

In this review, we will provide a comprehensive and critical overview of the clinical evidence on KP involvement in the pathophysiology of the most common neurodegenerative diseases (AD and PD), discussing therapeutic opportunities. We used the keywords ‘Alzheimer’s disease’ or ‘Parkinson's disease' in combination with ‘Kynurenine Pathway’ to select studies that evaluated the levels of KP metabolites in patients with AD and/or PD in comparison with controls. These studies are shown in Tables 1 and 2.

Table 1.

Clinical studies that measured kynurenine pathway metabolites in patients with Alzheimer’s disease.

Study

Groups

Sample

KP Metabolites and Enzymes

TRP

KYN

KYNA

AA

3-HK

3-HAA

XA

QUIN

TDO

IDO

KATs

Beal et al., 1992 [97]

AD cases (n=9-17) Controls (n=14-23) (Sample size varies according to the brain area)

Post-mortem brain tissue (Precentral gyrus, frontal cortex, inferior temporal middle temporal, caudate)

-

=

=

-

=

-

-

-

-

-

-

Heyes et al., 1992 [98]

AD cases (n=39) Controls (n=30)

CSF

=

=

↓

-

-

-

-

=

-

-

-

Pearson & Reynolds, 1992 [99]

AD cases (n=12) Controls (n=12)

Post-mortem brain tissue (Temporal cortex)

-

-

-

-

=

-

-

-

-

-

-

Baran et al., 1999 [100]

AD cases (n=11) Controls (n=13)

Post-mortem brain tissue (Frontal cortex, hippocampus, putamen*, caudate nucleus*, and cerebellum)

-

=

↑*

-

=

-

-

-

-

-

↑*

Widner et al., 2000 [51]

AD cases (n=21) Controls (n=20)

Serum

=

=

-

-

-

-

-

-

-

-

-

Guillemin et al., 2005 [40]

AD cases (n=6) Controls (n=4)

Post-mortem brain tissue (hippocampus)

-

-

-

-

-

-

↑

-

↑

-

Bonda et al., 2010 [42]

AD cases (n=12) Young controls (n=3) Age-matched controls (n=7)

Post-mortem brain tissue (hippocampus)

-

-

-

-

↑

-

-

-

-

↑

-

Gulaj et al., 2010 [50]

AD cases (n=34) Controls (n=18)

Plasma

↓

=

↓

=

=

-

-

↑

-

-

-

Schwarz et al., 2013 [101]

AD cases (n=20) Controls (n=19)

Serum

=

=

=

-

↑

-

-

=

-

-

-

Wu et al., 2013 [41]

AD cases (n=4) Controls (n=4)

Post-mortem brain tissue (hippocampus)

-

-

-

-

-

-

-

-

↑

↑

-

Giil et al., 2017 [49]

AD cases (n=65) Controls (n=65)

Plasma

↓

↓

-

-

-

↓

↓

↓

-

-

-

Oxenkrug et al., 2017 [60]

AD cases (n=20) Controls (n=24)

Serum

=

=

=

↓

↑

-

=

-

-

-

-

Muguruma et al., 2018 [102]

AD cases (n=10) Controls (n=10)

Post-mortem CSF

↓

↓

-

↓

↓

-

-

-

-

-

-

Jacobs et al., 2019 [45]

AD cases (n=20) Controls (n=18)

Plasma CSF

= =

= =

= ↑

= =

= =

↓ =

-

= =

-

-

-

Van der Velpen et al., 2019 [43]

AD cases (n=40) Controls (n=34)

Plasma CSF

↓ =

= ND

= ↑

ND =

= ND

-

-

= ↑

-

-

-

Sorgdrager et al., 2019 [53]

AD cases (n=33) Controls (n=39)

Serum CSF

= =

= =

= ↓

-

= =

-

↓ =

= =

-

-

-

González-Sánchez et al., 2020 [44]

AD cases (n=85) Controls (n=23)

Plasma CSF

= =

-

= ↑

-

-

-

-

-

-

-

-

Whiley et al., 2021 [103]

AD cases (n=103 – serum; n=176 – urine) Controls (n=86 – serum; n=171 – urine)

Serum Urine

↓ ↓

↓ =

= ↓

-

= =

= =

↓ ↓

=

-

-

-

Willette et al., 2021 [104]

AD cases (n=112) Controls (n=58)

Serum CSF

↑ ↓

= =

-

-

-

-

-

-

-

-

-

Symbols: (↓) decreased, (↑) increased, (=) no changes. Abbreviations: 3-HAA, 3-hydroxyanthranilic acid; 3-HK, 3-hydroxykynurenine; Alzheimer disease, AD; AA, anthranilic acid; CSF, cerebrospinal fluid; IDO, indoleamine 2, 3-dioxygenase; KATs, kynurenine aminotransferases; KYN, kynurenine; kynurenine pathway, KP; KYNA, kynurenic acid; ND, not detected or below the limit of quantification; QUIN, quinolinic acid; TDO, tryptophan 2,3-dioxygenase; TRP, tryptophan; XA, xanthurenic acid.

Table 2.

Clinical studies that measured kynurenine pathway metabolites in individuals with Parkinson's disease.

Study

Groups

Sample

KP Metabolites and Enzymes

TRP

KYN

KYNA

AA

3-HK

3-HAA

XA

QUIN

TDO

IDO

KATs

Beal et al., 1992 [97]

PD cases (n=9-11) Controls (n=14-18) (Sample size varies according to the brain area)

Post-mortem brain tissue (Precentral gyrus and caudate)

-

=

=

-

=

-

-

-

-

-

-

Tohgi et al., 1992 [65]

PD cases (n=16) Controls (n=16)

CSF

=

↓

-

-

↓

-

-

-

-

-

-

Tohgi et al., 1993 [66]

PD untreated (n=16) PD treated (n=13) Controls (n=10)

CSF

=

↓

-

-

↓

-

-

-

-

-

-

Widner et al., 2002 [61]

22 PD cases (n=22) Controls (n=11)

Serum CSF

↓* =

= ↑*

-

-

-

-

-

-

-

-

-

Hartai et al., 2005 [63]

PD cases (n=19) Controls (n=17)

Plasma Erythrocytes

-

-

= ↑

-

-

-

-

-

-

-

↓ KAT I and II ↑ KAT II

Lewitt et al., 2013 [105]

PD cases (n=48) Controls (n=57)

CSF

-

-

-

↑

-

-

-

-

-

-

Luan et al., 2015 [64]

PD cases (n=106) Controls (n=104)

Urine

-

↑

=

-

-

↑

↑

-

-

-

-

Oxenkrug et al., 2017 [60]

PD patients (n=18) Controls (n=24)

Serum

↓

↑

↑

↑

ND

-

=

-

-

-

-

Chang et al., 2018 [59]

Two-stage design: -Discovery cohort: PD cases (n=82) Controls (n=82)

Plasma

↓

↓

↓

=

=

=

-

=

-

-

-

-Independent cohort: PD cases (n=118) Controls (n=47)

Plasma

=

=

↓

=

=

=

-

↑

-

-

-

Sorgdrager et al., 2019 [53]

PD patients (n=33) Controls (n=39)

Serum CSF

↓ =

= =

↓ ↓

-

= =

-

↓ =

= =

-

-

-

Bai et al., 2020 [62]

PD cases (n=41) Controls (n=41)

Urine

↑

-

-

-

-

-

-

-

-

-

Heilman et al., 2020 [58]

PD cases (n=97) Controls (n=89)

Plasma CSF

= =

= =

= ↓

-

↑ =

↓ =

-

= =

-

-

-

Iwaoka et al., 2020 [106]

PD cases (n=20) Controls (n=13)

CSF

=

↑

=

-

↑

-

-

=

-

-

-

Symbols: (↓) decreased, (↑) increased, (=) no changes, (*) Only at advanced disease. Abbreviations: 3-HAA, 3-hydroxyanthranilic acid; 3-HK, 3-hydroxykynurenine; AA, anthranilic acid; CSF, cerebrospinal fluid; IDO, indoleamine 2, 3-dioxygenase; KATs, kynurenine aminotransferases; KYN, kynurenine; kynurenine pathway, KP; KYNA, kynurenic acid; ND, not detected or below the limit of quantification; PD, Parkinson disease; QUIN, quinolinic acid; TDO, tryptophan 2,3-dioxygenase; TRP, tryptophan.

2. OVERVIEW OF TRYPTOPHAN METABOLISM AND KYNURENINE PATHWAY

TRP is an essential amino acid (i.e., only obtained through diet) that plays several physiological roles, including modulation of neuronal activity and immune regulation [10]. TRP is used in the synthesis of proteins, biogenic amines [e.g., tryptamine, serotonin (5-hydroxytryptamine, 5-HT), melatonin (N-acetyl-5-methoxytryptamine)], while is degraded into kynurenines [15]. Fig. (1) summarizes TRP metabolism with a focus on the KP.

Fig. (1).

Tryptophan (TRP), an essential amino acid obtained from the diet, plays many roles in the body, including protein and neurotransmitter synthesis (serotonin: 5-hydroxytryptamine, 5-HT). TRP is metabolized mainly through the kynurenine (KYN) pathway. In the first and rate-limiting step of the KP, TRP is catalyzed by TDO or IDO originating KYN. Inflammatory molecules (e.g., IFN-γ, LPS) and infection (virus, bacteria) upregulate IDO. KP segregates into two major branches: (i) ‘a neurotoxic’ due to the production of neurotoxic molecules (3-HK, 3-HAA, QUIN) through the action of the enzyme KMO; and (i) a ‘neuroprotective’ with the synthesis of KYNA by the action of KATs. Enzymes are in light gray in the chart. Abbreviations: 3-HAA, 3-hydroxyanthranilic acid; 3-HK, 3-hydroxykynurenine; AA, anthranilic acid; CA, cinnabarinic acid; IDO, indoleamine 2, 3-dioxygenase; INF-γ, interferon-gamma; KATs, kynurenine aminotransferases; KMO, kynurenine 3-monooxygenase; KYN, kynurenine; KYNA, kynurenic acid; KYNU, kynureninase; LPS, lipopolysaccharide; NAD, nicotinamide adenine dinucleotide; QUIN, quinolinic acid; TDO, tryptophan 2,3-dioxygenase; TPH, tryptophan hydroxylase; TRP, tryptophan; XA, xanthurenic acid.

Around 95% of free TRP is metabolized through the KP. In the first and rate-limiting step of this pathway, TRP is metabolized by the enzymes IDO or tryptophan 2,3-dioxygenase (TDO) into N-formylkynurenine that is converted into kynurenine (KYN) by kynurenine formamidase (AFMID) [15]. TDO is expressed mainly in the cytosol of liver cells, being responsible for 90% of TRP degradation. TDO is upregulated by TRP and corticosteroids. Conversely, IDO is expressed in different cell types, including immune cells (macrophages and dendritic cells) and central nervous system cells, such as astrocytes and microglia (the resident immune cells in the brain). IDO expression and activity are induced by pro-inflammatory mediators, such as tumor necrosis factor (TNF), interferons (IFN)-α, β, and γ (mainly the latter), and immune stimuli, like lipopolysaccharide (LPS), a molecule present in the membrane of gram-negative bacteria [15].

Downstream of KYN, the KP segregates into two major branches. One is catalyzed by kynurenine aminotransferases (KATs) that convert KYN into kynurenic acid (KYNA). KYNA is an antagonist of the ionotropic excitatory amino acid receptors [N-methyl D-aspartate (NMDA), kainate, and AMPA receptors] and plays a role as a free radical scavenger, being implicated in neuroprotection [9]. In the CNS, KATs are localized mainly in astrocytes. The second branch is catalyzed by kynurenine-3-monooxygenase (KMO), an enzyme mainly expressed in microglia that converts KYN into 3-hydroxykynurenine (3-HK) and other metabolites, including 3-hydroxyanthranilic acid (3-HAA) and quinolinic acid (QUIN). 3-HK and 3-HAA can display pro-oxidant effects by the generation of superoxide anion and highly reactive hydroxyl radicals [9, 21]. 3-HK also affects synaptic functioning in the hippocampus, at least in part by the production of xanthurenic acid (XA) [22]. QUIN can cause neuronal dysfunction through different mechanisms, including its role as an NMDA receptor agonist, increased production of free radicals, exacerbated lipid peroxidation, disruption of the blood-brain barrier integrity, and induction of astrocyte apoptosis [23]. QUIN is also used in the production of nicotinamide adenine dinucleotide (NAD+), a coenzyme involved in energy metabolism [10, 24]. Two other metabolites from the KYN pathway, cinnabarinic acids (CA) and xanthurenic acid (XA), are produced, respectively, by the spontaneous condensation of two 3-HAA and transamination of 3-HK catalyzed by KAT. These metabolites interact with glutamate receptors and have displayed antipsychotic-like effects in pre-clinical studies [25-27]. CA is an orthosteric agonist of mGlu4 receptors, while XA is an agonist of Group II (mGlu2 and mGlu3) metabotropic glutamate receptor and a vesicular glutamate transporter (VGLUT) inhibitor [28]. It has also been suggested that XA can affect synaptic function via inhibition of VGLUT [22, 29]. Finally, there is a third pathway mediated by kynureninase (KYNU) in which KYN is used for the synthesis of anthranilic acid (AA) [10].

In addition to the KP, TRP can be metabolized into serotonin, a neurotransmitter involved in several physiological roles (e.g., regulation of mood, cognition, sleep, and appetite) and implicated in the pathophysiology of neuropsychiatric disorders, especially mood disorders [15]. A small percentage of the ingested tryptophan (less than 1%) is used in the synthesis of serotonin [15], and the bulk of serotonin in the body is produced in the gut [30] by intestinal enterochromaffin cells [10]. In the brain, serotonin is synthesized in specific brainstem nuclei, modulating several neurophysiological processes. Serotonin is also used in the synthesis of melatonin in the pineal gland, a neurohormone involved in the regulation of circadian rhythm [31]. With aging, there is a progressive decrease in serotonin and melatonin levels, probably contributing to the deterioration of cognitive and emotional processes associated with neurodegenerative diseases [32, 33].

IDO and KMO are upregulated in response to pro-inflammatory or immune stimuli, which can deviate TRP metabolism from the production of serotonin into the production of neurotoxic molecules [34]. As mentioned before, aging has been associated with chronic low-grade inflammation or ‘inflammaging’ [14, 19, 20]. Corroborating the view that inflammation activates IDO and, as a consequence, the KP, previous studies have demonstrated an increase in KYN levels and KYN/TRP ratio with age [11, 12]. Furthermore, a large observational study found a positive association between plasma level of KYN/TRP ratio, 3-HK, and AA with risk of mortality [13].

The activation of immune cells and neuroinflammation plays a role in the pathophysiology of neurodegenerative diseases [35, 36]. Given this and the regulatory effect of inflammation on the KP, we will address in the following sessions the current evidence implicating KP's involvement in the two main human neurodegenerative diseases, AD and PD.

3. IMPLICATION OF KP IN ALZHEIMER’S DISEASE

The neuropathological hallmark of AD is the accumulation of β-amyloid plaques and neurofibrillary tangles (composed of phosphorylated forms of the protein tau), phenomena that start decades before the onset of cognitive and behavioral symptoms [37, 38].

Different methodological approaches have described KP changes in AD. An in vitro study showed that β-amyloid 1-42 (Aβ1-42) induces IDO expression leading to a significant increase in QUIN production in human fetal microglia and adult macrophages [39]. Pathoanatomical studies showed increased levels of QUIN [40] and IDO [40, 41] in the hippocampus of AD patients when compared with controls. Another study described a significant increase in KYNA in the putamen and caudate nuclei, which positively correlated with increased KAT I activity in the same brain areas [42].

Most studies have also shown an increase in the cerebrospinal fluid (CSF) levels of KYNA in AD patients when compared with controls, while the plasma level of this marker did not differ between groups [43-45]. Given the potential neuroprotective role of KYNA, some authors speculated that this might represent a compensatory mechanism to reduce, via competitive inhibition, the effects of chronic excitotoxicity mediated by the interaction between QUIN and the NMDA receptor [44, 45]. Conversely, van der Velpen et al. (2019) reported a higher CSF level of QUIN in AD patients when compared with cognitively intact controls and its positive correlation with the CSF levels of tau and phosphorylated tau 181 (pTau-181) [43], markers of neurodegeneration [46]. The same authors showed that CSF Aβ1-42 level was positively correlated with CSF KYNA level independent of age and sex [43]. Together, these studies suggest that KP metabolites change in association with AD-related pathological processes. The direction of these associations remains to be determined, i.e., whether these changes are secondary to AD pathology and/or contribute to it.

Despite conflicting results, clinical studies have also shown alterations in the peripheral levels of KP metabolites in patients with AD and their association with cognitive performance. It is worth noticing that KYNA and QUIN cross the blood-brain barrier very poorly (dependent on passive diffusion), while TRP, KYN, and 3-HK can actively cross the blood brain barrier through the large neutral amino acid transport system [10, 47]. A recent systematic review of human studies corroborated this, as there is a correlation between peripheral and central levels of KYN and 3-HK in subjects with neuropsychiatric disorders [48].

Previous studies have found lower plasma levels of TRP [43, 49, 50] and higher plasma KYN/TRP and QUIN/3-HK ratios [50] in patients with AD when compared with controls, suggesting activation of IDO, KP and its ‘neurotoxic’ branch/KMO. Cognitive performance of AD patients correlated negatively with serum KYN/TRP ratio [51] and plasma QUIN levels [49, 50] but correlated positively with plasma KYNA levels [50], suggesting that activation of KP and its neurotoxic branch is associated with more severe disease and greater cognitive impairment. Interestingly, alterations in KP metabolites seem to start before cognitive impairment parallel with the neuropathological changes. Chatterjee et al. (2018) observed higher serum levels of KYN, AA, and 3-HK in cognitively normal women with preclinical AD, as defined by the high neocortical amyloid-β load on positron emission tomography (PET) using 18F-florbetaben, when compared with people without AD [52]. In addition, two studies have shown a reduction in the serum levels of XA in patients with AD compared to controls [49, 53], while no changes in the CSF levels [53]. Serum levels of XA were inversely correlated with age, suggesting an effect of aging on this metabolite [49]. Blood levels of XA and CA are reduced in schizophrenia [27, 54], but no clinical studies have measured CA in patients with AD.

Table 1 summarizes the studies measuring KP metabolites in patients with AD and controls.

4. IMPLICATION OF KP IN PARKINSON'S DISEASE

PD is a progressive neurodegenerative disease characterized by motor symptoms consisting of bradykinesia, resting tremor, rigidity and postural instability. As in AD, the neurodegeneration in PD starts before the onset of motor symptoms. The neuropathological hallmark of PD is the loss of dopaminergic neurons in the substantia nigra pars compacta and intra-neuronal aggregates of α-synuclein in the form of Lewy bodies and Lewy neurites [55, 56].

Ogawa et al. (1992) observed a decrease in KYN and KYNA concentrations in the post-mortem frontal cortex of patients with PD who received levodopa therapy. The levels of 3-HK, on the other hand, were increased only in the putamen of PD patients who did not receive the therapy, highlighting that both disease and treatment can affect KP metabolites [57].

Studies have also assessed peripheral levels of KP metabolites in PD. Despite conflicting results [58], most studies have shown lower peripheral TRP levels in patients with PD when compared with controls [53, 59-61]. Some studies found no difference in the circulating levels of KYN [53, 58], while others observed an increase [59] or decreased [60] in PD compared with controls. Recently, Bai et al. (2021) showed an increase in the urinary concentration of KYN in patients with PD. KYN urinary level positively correlated with severity and duration of the disease but negatively correlated with cognition, as assessed by the Mini-Mental State Examination (MMSE) [62]. Regarding other KP metabolites downstream of KYN, Hartai et al. (2005) did not find changes in the plasma levels of KYNA, but the levels of KYNA and the activity of KAT II were higher in the erythrocytes of PD patients compared with controls [63]. More recently, Chang et al. (2018) showed that patients with PD had higher plasma levels of QUIN and lower plasma levels of KYNA, indicating activation of the ‘neurotoxic’ branch. These peripheral changes seem to be progressive since patients at the advanced stages of the disease were those who had lower KYNA and KYNA/KYN ratios while higher QUIN and QUIN/KYNA ratios [59].

Heilman et al. (2020) observed increased plasma levels of 3-HK and decreased plasma levels of 3-HAA in patients with PD compared with controls. In addition, Sorgdrager et al. (2019) observed reduced serum levels of XA in patients with PD and no difference in the CFS levels compared with controls [53], while Luan et al. (2015) found increased urinary levels of XA in patients with PD [64]. Heilman et al. (2020) also measured the levels of KP metabolites in the CSF, showing that only KYN was lower in PD patients compared with controls [58], in agreement with previous studies [65, 66]. Widner et al. (2002) observed that patients in advanced stages of PD have higher CSF levels of KYN than those in earlier stages, suggesting increased activation of the KP with disease progression [61], akin to studies using blood samples [59]. Furthermore, the severity of the disease, as assessed by the Unified Parkinson’s Disease Rating Scale (UPDRS), correlated positively with CFS levels of QUIN and plasma levels of 3-HK [58]. Interestingly, CSF levels of KYNA correlated positively with the University of Pennsylvania Smell Identification Test (UPSIT) scores, with higher scores indicating better olfaction [58]. As loss of olfaction is a common non-motor symptom in early PD, this finding suggests that activation of the ‘neuroprotective’ branch is associated with less impact on the disease [55, 56]. Table 2 summarizes the studies measuring KP metabolites in patients with PD and controls [95-104].

5. CHALLENGES AND OPPORTUNITIES

Despite the advances in the understanding of the pathophysiology of neurodegenerative diseases, no available treatment can prevent or delay the progression of these conditions and, as a consequence, the negative clinical outcomes that characterize them [5]. Therefore, there is a huge need to develop effective strategies against these conditions.

The studies of the KP in neurodegenerative diseases are still in their early stages, involving mainly preclinical models. For example, studies using transgenic mice and fly models of neurodegenerative diseases demonstrated the effect of KMO inhibitors in decreasing the production of downstream neurotoxic KP metabolites and protecting against neurodegeneration [67]. In the early 2000s, Giorgini et al. (2005) showed that the suppression of the KMO enzyme protects against cellular toxicity caused by mutant huntingtin fragment in a yeast model for Huntington’s disease (HD), a monogenic neurodegenerative disease that shares pathological and clinical similarities with AD and PD [68, 69]. A subsequent study showed the promising effect of genetic and pharmacological inhibition of KMO on increasing the levels of the neuroprotective KYNA relative to the neurotoxic 3-HK and ameliorating neurodegeneration in the fruit fly model (Drosophila melanogaster) of HD. Similar results were obtained by inhibiting the TDO enzyme [70]. The inhibition of both enzymes, KMO and TDO, also ameliorated disease-related phenotypes in fly models of AD and PD, including neurodegeneration, locomotor abnormalities, and shortened lifespan [70]. In addition, the administration of JM6 [2-(3,4-Dimethoxybenzenesulfonylamino)-4-(3-Nitrophenyl)-5-(Pipe- ridin-1-yl)Methylthiazole], a prodrug inhibitor of KMO, prevented spatial memory loss in a transgenic mouse model of AD [71]. Of note, the role of JM6 as a KMO inhibitor has been debatable [72]. Despite very promising results, the poor brain permeability of KMO inhibitors has been a limitation for using these drugs, as suggested by Zhang et al. (2021) [67]. Based on the finding that peripheral KMO inhibition increases central KYNA levels [71, 73], some authors have suggested that the peripheral increase of KYN levels – a metabolite that actively crosses the blood-brain barrier – can lead to increased KYNA levels in the brain [67, 71].

IDO inhibitors have also led to interesting results in pre-clinical models of neurodegenerative diseases. Mice models of PD treated with 1-methyltryptophan (1-MT), an IDO-1 inhibitor, improved motor impairment and biochemical-relevant parameters, such as markers of oxidative stress, inflammation, mitochondrial dysfunction, and neuronal apoptosis [74]. 1-MT also restored striatum dopamine and brain-derived neurotrophic factor levels [74]. Another study also showed that 1-MT prevented neuronal death caused by viral infection [75]. Of note, IDO inhibitors have been tested in clinical trials for cancer immunotherapy [76] but not for neuropsychiatric disorders.

Besides enzymatic inhibitors, there are other potential pharmacological strategies targeting TRP metabolism, as recently reviewed by Modoux et al. (2021) [77]. Greater attention has been given to the aryl hydrocarbon receptor (AhR), a transcription factor that plays a role in the link between aging/neurodegenerative diseases and gut microbiota [78]. AhR acts as an environmental sensor regulating immune response at barrier sites and contributing to intestinal homeostasis [77]. After activation, AhR induces the transcription of cytokines such as interleukin (IL)-22 and IL-17, which are pivotal for host defense against bacterial and fungal infections [79]. Unabsorbed TRP is metabolized by gut microbiota, and the resulting metabolites, such as indole derivatives, tryptamine, and skatole, are ligands for AhR [77]. Because of the variety of ligands, the participation of AhR in innate and adaptive immune responses depends on the nature of the stimulus [79]. Despite the potential relevance of TRP-AhR pathway activation in the gut, the role of its clinical manipulation remains to be established. And for that, the researchers are focusing on studying the potential anti-inflammatory effect of the AhR using its agonists [79].

The bidirectional communication between microbiota/gut and brain, the so-called microbiota-gut-brain axis, relies on multiple pathways, including signals carried by the vagus nerve and enteric nervous system, neuroendocrine pathways, microbial metabolites, including TRP-related metabolites [80]. Microbiota stimulates IDO1, which leads to the production of KYN and other KP metabolites. The concentration of KYNA increases along the gastrointestinal tract and may play a role in mucosal protection and immunoregulation by its interaction with G protein-coupled receptor, GPR35, expressed mainly in epithelial and immune cells [81]. Gut microbiota is comprised of over 1,000 distinct species [82-85]. Over the lifespan, the composition of the microbiota passes throughout remodeling and there is a reduction in the diversity and abundance [14, 86]. However, these changes seem to be compensated in subjects with extreme longevity (>105 y). In this population, there is an increase in some subdominant species (e.g., Akkermansia spp., Bifidobacterium spp., and Christensenellaceae) in the microbiota compared with younger people [86]. Other than age, some factors contribute to microbiota modulation, such as lifestyle, nutrition, and the presence of inflammatory diseases. The process of ‘inflammaging’, for instance, may induce unfavorable variations in the microbiota [82, 87], a phenomenon called dysbiosis. The increase in the aerobiosis and the production of Reactive oxygen species (ROS) during inflammation is harmful to the strictly anaerobic bacteria (e.g., Firmicutes) but favorable to amplify the facultative aerobes, mainly pathobionts, microorganisms usually associated with disease development. Pathobionts contribute to maintaining an inflammatory milieu in the gut, favoring an inflammatory circle and possibly the progression of diseases [82].

People with AD and PD present altered microbiota, a condition associated with increased gut permeability, also aggravated by aging. Indeed, molecules from the intestinal lumen, such as bacterial particles (e.g., LPS), have been found in the brain of AD patients [87]. In parallel, a growing body of evidence suggests that the gut is the initiating site of PD since aggregates of α-synuclein have been found in the enteric nerves of patients [88]. Altogether, these results highlight that the microbiota modulation would be interesting in AD and PD patients since both microbiota and TRP metabolites are altered in these individuals. Nowadays, the most studied interventions to modulate microbiota are the use of prebiotics (e.g., fructooligosaccharides), probiotics (e.g., Lactobacillus), or symbiotics (prebiotics plus probiotics), fecal microbiota transplantation, and specific dietary patterns that are composed of a large amount of microbiota-modulating substances, such as dietary fiber, polyphenols, and polyunsaturated fatty acids (e.g., Mediterranean diet) [88, 89]. It remains to be established whether the potential positive effects of these strategies are modulated through KP changes. In this regard, a recent exploratory intervention showed that AD patients using probiotics (a mix of Lactobacillus and Bifidobacterium) for four weeks changed their microbiota composition and KYN levels but without clinical effects [90].

Beyond microbiota, vitamin B6 seems to be crucial for the proper functioning of several enzymes. KYNU and KAT enzymes are involved in KP downstream of KYN production and both are dependent on vitamin B6 [13, 91]. Therefore, the production of the neuroprotective metabolite KYNA is dependent on the supply of vitamin B6. Midttun et al. (2011) found an inverse correlation between plasma levels of B-6 form pyridoxal 5’-phosphate (PLP) and 3-HK in patients with cardiovascular diseases [92]. There was also a negative correlation between the level of PLP and the 3-HK: KYNA ratio, suggesting an increase in the production of KP neurotoxic metabolites at lower concentrations of vitamin B6. Interestingly, the oral supplementation of 40mg daily of vitamin B6 (pyridoxine hydrochloride) for 28 days reduced the 3-HK: KYNA ratios by up to 60%, particularly in subjects with low plasma PLP at baseline [91]. Lower vitamin B6 status (PLP <43 nmol/L) and B6 dietary intake were associated with a greater rate of cognitive decline in older adults [93], while a higher intake of vitamin B6 was linked to a decreased risk of AD [94]. Furthermore, elevated plasma levels of homocysteine, which reflect the status of vitamins B6, B12, and folate, constitute a risk factor for AD [95]. Therefore, the connection between vitamin B6, KP metabolites, and cognition needs to be better investigated [96].

In sum, non-pharmacological therapies, such as manipulation of the gut microbiota and adequate nutrient supply (in particular those that are co-factors for KP enzymes), also contribute to KP modulation, but the resulting impact on neurodegenerative disease remains to be determined (Table 3) [105, 106].

Table 3.

Potential targets for disease-modifying therapies.

Enzymatic Inhibitors	Described Neuroprotective Effects	Molecule Examples	Refs.
KMO inhibitors	- ↓3-HK and ROS (in vitro); - ↓3-HK/ KYNA; - Peripheral KMO inhibition increases central KYNA levels; - Ameliorated disease-related phenotypes in fly models of AD and PD, including neurodegeneration, locomotor abnormalities, and shortened lifespan.	Ro 61-8048 JM6	[69, 70, 72, 73]
IDO/TDO inhibitors	- ↑ TRP; - ↓3-HK/ KYNA; - Improvement in motor impairment, ↓ oxidative stress, ↓ cytokines (TNF-α, IFN-γ, IL-6), and ↓ neuronal apoptosis in the striatum of PD animal model; - Restoration of striatum neurotransmitter (dopamine and homovanillic acid) and neurotrophic factor (brain-derived neurotrophic factor) levels; - Prevent neuronal death caused by viral infection; - Ameliorated disease-related phenotypes in fly models of AD and PD	1-MT 680C91 Epacadostat Navoximod Indoximod Linrodostat	[70, 74-77]

Abbreviations: AD, Alzheimer’s disease; IDO, indoleamine 2, 3-dioxygenase; KMO, kynurenine-3-monooxygenase; KP, kynurenine pathway; KYNA, kynurenic acid; PD, Parkinson's disease; TDO, tryptophan 2,3-dioxygenase, 1-MT, 1-methyltryptophan; 3-HK, 3-hydroxykynurenine.

CONCLUSION

Given the emerging evidence of the involvement of KP in the pathophysiology of neurodegenerative diseases and the fact that this pathway has multiple potential pharmacological targets, a better understanding of the KP role in neurodegeneration is definitely warranted. This will involve defining changes of KP metabolites across AD and PD courses, their relationship with specific disease markers (e.g., beta-amyloid for AD; alpha synuclein for PD) and markers of degeneration (e.g., neurofilament light chain), clinical (cognitive, motor, and behavioral) measures, and in different biological matrices (e.g., plasma vs. CSF vs. urine).

LIST OF ABBREVIATIONS

AA

Anthranilic Acid

AD

Alzheimer’s Disease

AhR

Aryl Hydrocarbon Receptor

Aβ1-42

β-amyloid 1-42

CA

Cinnabarinic Acid

CNS

Central Nervous System

CSF

Cerebrospinal Fluid

DAMPs

Damage-Associated Molecular Patterns

3-HAA

3-Hydroxyanthranilic Acid

HD

Huntington’s Disease

3-HK

3-Hydroxykynurenine

5-HT

5-Hydroxytryptamine

IDO

Indoleamine 2, 3-Dioxygenase

IL

Interleukin

INF-γ

Interferon-Gamma

JM6

2-(3,4-Dimethoxybenzenesulfonylamino)-4-(3-Nitrophenyl)-5-(Piperidin-1-yl)Methylthiazole

KATs

Kynurenine Aminotransferases

KMO

Kynurenine 3-Monooxygenase

KP

Kynurenine Pathway

KYN

Kynurenine

KYNA

Kynurenic Acid

KYNU

Kynureninase

LPS

Lipopolysaccharide

MMSE

Mini-Mental State Examination

1-MT

1-Methyltryptophan

NAD

Nicotinamide Adenine Dinucleotide

NMDA

N-methyl D-aspartate

PD

Parkinson's Disease

PET

Positron Emission Tomography

PLP

Pyridoxal 5’-phosphate

pTau-181

Phosphorylated Tau 181

QUIN

Quinolinic Acid

TDO

Tryptophan 2,3-dioxygenase

TNF

Tumor Necrosis Factor

TPH

Tryptophan Hydroxylase

TRP

Tryptophan

UPDRS

Unified Parkinson’s Disease Rating Scale

UPSIT

University of Pennsylvania Smell Identification Test

XA

Xanthurenic Acid

CONSENT FOR PUBLICATION

Not applicable.

FUNDING

AL Teixeira is a CNPq fellowship recipient. The Neuropsychiatry Program is funded by the UTHealth Department of Psychiatry and Behavioral Sciences; National Institute of Aging (Grant no. AG072380-01A1), Texas Alzheimer’s Research and Care Consortium (Grant no. TARCC 2020-23-93-II).

CONFLICT OF INTEREST

The authors declare no conflict of interest, financial or otherwise.