Amyotrophic lateral sclerosis (ALS) is the most commonfatal, adult onset motor neuron degenerative disease.Approximately 90% of ALS cases are sporadic and of the10% familial cases, 20% are attributed to mutations in thesuperoxide dismutase gene (Bruijn et al. 2004). ALSselectively targets motor neurons in the motor cortex, brainstem and spinal cord, leading to loss of motor units, generalmuscle weakness and atrophy, paralysis and ultimately, death(Bruijn et al. 2004). The neuropathology of ALS isassociated with multiple features, including excitotoxicity,oxidative stress and protein aggregation (Shaw 2005).Neuroinflammation is also a typical hallmark of ALS(McGeer and McGeer 2002). However, the precise diseasemechanism is presently still unknown.

In physiological conditions, the kynurenine pathway (KP)(Fig. 1) is involved in the catabolism of tryptophan to yield

NAD+. During neuroinflammation, the KP is activated in braincells and lead to the production of both neurotoxic andneuroprotective molecules. Intermediates kynurenic acid

Received July 22, 2010; revised manuscript received December 7, 2010;accepted December 17, 2010.Address correspondence and reprint requests to Dr Gilles Guillemin,

Department of Pharmacology, School of Medical Sciences, University ofNew South Wales, NSW 2052, Australia.E-mails: g.guillemin@unsw.edu.au; g.guillemin@cfi.unsw.edu.auAbbreviations used: 1-MT, 1-methyl tryptophan; ALS, amyotrophic

lateral sclerosis; APV, 2-amino-5-phosphonopentanoic acid; FBS, foetalbovine serum; IDO-1, indoleamine-2, 3 dioxygenase; IFN-c, interferongamma; KAT, kynurenine amino transferase; KP, kynurenine pathway;KYNA, kynurenic acid; KYNU, kynureninase; LDH, lactate dehy-drogenase; MIP, macrophage inflammatory protein; PIC, picolinic acid;PS, penicillin-streptomycin; QPRT, quinolinate phosphoribosyl trans-ferase; QUIN, quinolinic acid; TDO, tryptophan 2,3-dioxygenase.

, ,

*Department of Pharmacology, School of Medical Sciences, University of New South Wales, Sydney,

New South Wales, Australia

�St Vincent’s Centre for Applied Medical Research, Darlinghurst, New South Wales, Australia

�Department of Neurology, St. Vincent’s Hospital, Darlinghurst, New South Wales, Australia

Abstract

Amyotrophic lateral sclerosis (ALS) is the most common type

of motor neuron degenerative disease for which the aetiology

is still unknown. The kynurenine pathway (KP) is a major

degradative pathway of tryptophan ultimately leading to the

production of NAD+ and is also one of the major regulatory

mechanisms of the immune response. The KP is known to be

involved in several neuroinflammatory disorders. Among the

KP intermediates, quinolinic acid (QUIN) is a potent ex-

citotoxin, while kynurenic acid and picolinic acid are both

neuroprotectant. This study aimed to (i) characterize the

components of the KP in NSC-34 cells (a rodent motor neuron

cell line) and (ii) assess the effects of QUIN on the same cells.

RT-PCR and immunocytochemistry were used to characterize

the KP enzymes, and lactate dehydrogenase (LDH) test was

used to assess the effect of QUIN in the absence and pre-

sence of NMDA receptor antagonists, kynurenines and 1-

methyl tryptophan. Our data demonstrate that a functional KP

is present in NSC-34 cells. LDH tests showed that (i) QUIN

toxicity on NSC-34 cells increases with time and concentra-

tion; (ii) NMDA antagonists, 2-amino-5-phosphonopentanoic

acid, MK-801 and memantine, can partially decrease QUIN

toxicity; (iii) kynurenic acid can decrease LDH release in a

linear manner, whereas picolinic acid does the same but non-

linearly; and (iv) 1-methyl tryptophan is effective in decreasing

QUIN release by the rodent microglial cell line BV-2 and thus

protects NSC-34 from cell death. There is currently a lack of

effective treatment for ALS and our in vitro results provide a

novel therapeutic strategy for ALS patients.

Keywords: amyotrophic lateral sclerosis, kynurenine path-

way, motor neuron disease, neuroprotection, quinolinic acid.

J. Neurochem. (2011) 118, 816–825.

JOURNAL OF NEUROCHEMISTRY | 2011 | 118 | 816–825 doi: 10.1111/j.1471-4159.2010.07159.x

816 Journal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825� 2011 The Authors

(KYNA) (Perkins and Stone 1982; Hilmas et al. 2001) andpicolinic acid (PIC) (Kalisch et al. 1994; Jhamandas et al.2000) are considered as neuroprotectant whereas quinolinicacid (QUIN) is a potent excitotoxin (Stone and Perkins 1981).Within the CNS, indoleamine-2, 3 dioxygenase (IDO-1) is theenzyme initiating the KP (Takikawa et al. 1986). IDO-1 isexpressed in most of the human brain cells (Guillemin et al.2000, 2001, 2005c, 2007; Owe-Young et al. 2008) and isparticularly sensitive to activation by interferon gamma (IFN-c) (Yoshida and Hayaishi 1978). The KP is known to beimplicated in a number of neurodegenerative diseases, such asAlzheimer’s disease (Guillemin et al. 2005a), multiplesclerosis (Lim et al. 2010) and is likely to play a role in ALS(Guillemin et al. 2005b; Chen et al. 2010).

The NSC-34 mouse motor neuron cell line was developedthrough the fusion of mouse neuroblastoma cells and motorneuron enriched spinal cord cells (Cashman et al. 1992),creating an easily accessible, immortal and clonally uniformcell line that overcomes the problems associated with theculturing of primary motor neurons, such as low yields andlimited purity. The NSC-34 model exhibits many of theunique morphological and physiological characteristics ofmotor neurons, for instance, the ability to produce andaccumulate acetylcholine, generate action potentials, extendneuritic processes, express neurofilament proteins, formsynapses with cultured myotubes and induce myotubetwitching (Cashman et al. 1992). Since its creation, theNSC-34 cell line has become one of the established in vitromodels for ALS research.

This study aims to investigate the enzymatic and proteincomponents of the KP in NSC-34 cells and its implication onALS will be discussed.

Materials and methods

NSC-34 and BV-2 cell culturesAll cell culture reagents were purchased from Invitrogen (Carlsbad,

CA, USA) unless otherwise stated.

NSC-34 cells were maintained in Dulbecco’s modified eagle

medium supplemented with 10% foetal bovine serum (FBS)

(Bovogen, Essendon, Victoria, Australia) and 1% penicillin-

streptomycin (PS) (Cashman et al. 1992) and subcultured every

3–4 days. To slow cell proliferation and enhance cell differentia-

tion, the medium was exchanged for 1 : 1 Dulbecco’s modified

eagle medium plus Ham’s 12 with 1% FBS, 1% PS and 1%

non-essential amino acids (Sigma-Aldrich, St Louis, MO, USA)

(Eggett et al. 2000). The medium was changed every 2 days

and the cells grown for up to 7 days. The differentiated

cells were maintained in reduced-serum medium and passaged

twice without loss of viability (Eggett et al. 2000). Only cells that

had undergone differentiation were used in the following

experiments.

The rodent microglial BV-2 cells were maintained in Roswell

Park Memorial Institute formula 1640 supplemented with 1% FBS,

1% PS and 1% Glutamax. The medium was changed every 2–

3 days and passaged biweekly by shaking the culture vigorously for

cell detachment.

Characterization of KP enzymes by RT-PCRThe RT-PCR protocol used has been previously described

(Guillemin et al. 2001, 2007). The primer sequences used to detect

murine KP enzymes are summarized in Table 1. The positive

control used was IFN-c treated BV-2 cells, while the negative

control was the omission of cDNA from the PCR process. The PCR

product was then semi-quantified using Adobe Photoshop (version

CS2; Adobe systems, San Jose, CA, USA). The experiment was

performed in triplicate and 10 density values were taken and

Tryptophan

Quinolinic acid

Picolinicacid

Kynurenic acid

Formylkynurenine

Kynurenine

Kynurenine formylase

Kynurenine amino-

Aminocarboxymuconatesemialdehyde decarboxylase(ACMSD)

transferases (KATs)

Kynurenine mono-oxygenase

Kynureninase (KYNU)

(KMO)

Tryptophan or indoleamine 2,3-dioxygenase (TDO/IDO1)

3-hydroxykynurenine

3-hydroxyanthranalic acid

3-hydroxyanthranilate dioxygenase

Quinolinate phosphoribosyl transferase (QPRT)

(3HAO)

NAD

Fig. 1 Schematic diagram of the kynur-

enine pathway (KP). The KP converts

tryptophan to NAD+ and generates various

neuroactive intermediates, such as quino-

linic acid (QUIN) and picolinic acid (PIC).

� 2011 The AuthorsJournal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825

Kynurenine pathway expression in motor neuron | 817

averaged for each gel. The values were calculated as a ratio of

housekeeping gene, GAPDH. The results were compared using

Mann–Whitney test, with a p < 0.05 taken to be statistically

significant.

ImmunocytochemistryAll antibodies were purchased from Abnova (Taipei City, TW, USA)

unless otherwise stated. NSC-34 cells were plated onto slide flasks

(Nunc, Rochester, NY, USA) and when 70% confluence was

reached, they were either left untreated or treated with 100 IU/mL of

IFN-c for 24 h. The cells were then stained for IDO-1, tryptophan

2,3-dioxygenase (TDO2), kynureninase (KYNU), QUIN (Millipore,

Bedford, MA, USA), PIC and SMI-32 (Covance, Emeryville, NJ,

USA), a motor neuron marker, according to a previously described

protocol (Guillemin et al. 2001, 2007).

QUIN binding and uptake time-course study usingimmunocytochemistryNSC-34 cells were grown in slide flasks to 70% confluence.

Subsequently, they were treated with 1 lM QUIN (sub-neurotoxic

concentration for NSC-34 – see Fig. 5) and the treatment stopped at

0 h, 30 min, 60 min and 90 min whereby the media were removed,

the cells washed with phosphate-buffered saline and fixed (Rahman

et al. 2009). Immunocytochemistry was then carried out to observe

the movement of QUIN in NSC-34 cells.

Quantification of QUIN toxicity on NSC-34 cells using lactatedehydrogenase (LDH) testAll reagents were purchased from Sigma-Aldrich unless otherwise

stated. For all LDH testing with NSC-34 cells, cells were grown to

confluence in 24-well plates. All experiments were done in

triplicates.

First, the toxicity of varying concentrations of QUIN at 24, 48

and 72 h was investigated (standard curve).

The LDH results and the cell conditions after treatment were then

analysed to obtain a suitable QUIN concentration and time period

for subsequent LDH experiments involving the presence of

antagonists. QUIN 2 lM for 48 h was chosen as it gave a strong

LDH detection level while still maintaining viable cells after

treatment.

The effect of QUIN was then assessed in the presence of NMDA

antagonists 2-amino-5-phosphonopentanoic acid (APV), MK-801

and memantine; and neuroprotective kynurenines (KYNA and PIC)

on NSC-34.

BV-2 is a microglia cell line that produces QUIN following IFN-cstimulation. To assess the effect of the IDO-1 inhibitor, 1-methyl-DL-

tryptophan (1-MT), BV-2 cells were pre-treated with different

concentrations of 1-MT for 30 min followed by 100 IU/lL of IFN-cfor 48 h. The supernatant was then transferred onto NSC-34 cells

and incubated for 48 h.

For the LDH assay, the NSC-34 supernatants were collected and

diluted 1 : 100 in deionized water. In a 96-well plate, sodium

pyruvate (11.5 mM) and NAD (700 lM in 0.1 M potassium

phosphate buffer) were added to the diluted sample in each well.

The rate of reduction in absorbance was measured at 340 nm using a

microplate reader (Bio-Rad, Hercules, CA, USA), the reading

correlating with the level of NAD oxidation because of the amount

of LDH in the sample. The amount of LDH produced was then

calculated as a ratio of the amount of protein, determined using the

Bradford assay. The cells that remained in the 24-well plates, after

the removal of media, were filled with phosphate-buffered saline

and sonicated at 20 kHz for 10 s per well to lyse the cells

completely. In a 96-well plate, c-globulin was used as the protein

standard and Bradford reagent (Bio-Rad) was added to both

standards and samples. The protein levels were determined by

measuring the absorbance at 595 nm.

The results were compared using linear and multiple regression,

with a p < 0.05 taken to be statistically significant.

Results

End-point RT-PCR on NSC-34 cellsThe results from RT-PCR showed that all the KP enzymestested, except kynurenine amino transferase (KAT)-II, wereexpressed in both IFN-c treated BV-2 (positive control forKP activation) and NSC-34 cells, although to varyingdegrees (Fig. 2). For IFN-c treated NSC-34 cells, theexpression of KAT-I was visibly reduced to the point ofbeing undetectable.

The semi-quantification of the electrophoresis gel bandswas expressed relative to GAPDH (Fig. 2b). The overalltrend appeared to be a general increase in enzyme expressionin IFN-c treated NSC-34 cells compared with untreated cells.Using Mann–Whitney test, it was revealed that only TDO,KYNU, aminocarboxymuconate semialdehyde decarboxy-lase and quinolinate phosphoribosyl transferase (QPRT) weresignificantly increased (p < 0.05) in treated cells, and thatKAT-I was significantly decreased (p < 0.05).

Table 1 Sequences of the primers for murine KP

Gene Primer Sequence (5¢–3¢)

Mouse IDO (NM_008324) Forward cgagtgtgtgaatggtctgg

Reverse ggtgttttctgtgccctgat

Mouse TDO (NM_019911) Forward tgtgagcgacaggtacaagg

Reverse tttccaggattggaccaaaa

Mouse KYNU (NM_027552) Forward ctcttgtgaggccagaggtc

Reverse cgacctgggttcaattccta

Mouse KAT-I (NM_172404) Forward ggcagctacttcctcattgc

Reverse cagcgccatctctgtgataa

Mouse KAT-II (NM_011834) Forward agagtggcatgttcccaaag

Reverse tggatccatcctgtcagtca

Mouse KMO (NM_133809) Forward ccgcttctgtcttctccaag

Reverse ccagggtcctatgcaggtta

Mouse 3HAO (NM_025325) Forward ctctgtggccttgtctgtga

Reverse ccagtgacagctctcaacca

Mouse ACMSD

(NM_001033041)

Forward ctagaaggctcggggtctct

Reverse gcctcaaacacagacccatt

Mouse QPRT (NM_133686) Forward ctggacaacctcacccagtt

Reverse aggcactggggtatctcctt

Mouse GAPDH (NM_008084) Forward ggcattgctctcaatgacaa

Reverse tgtgagggagatgctcagtg

Journal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825� 2011 The Authors

818 | Y. Chen et al.

Immunocytochemical characterization of the KPcomponents in NSC-34 cellsNSC-34 cells were stained for IDO-1, TDO, KYNU, PIC,QUIN and SMI-32. Both treated and untreated cells showedpositive staining for TDO, KYNU, PIC and SMI-32, andnegative staining for QUIN (Fig. 3a and b). IDO-1immunoreactivity, evident in the treated cells, was the onlydifference observed between the two groups (Fig. 3b).

QUIN binding and uptake time-course studyNSC-34 cells were treated with 1 lM QUIN for different timeperiods: 0 h, 30 min, 60 min and 90 min (Fig. 4). At 0 h, noQUIN was detected in the cells. Starting from 30 min,positive QUIN staining was observed. The intensity of thestaining gradually increased at each subsequent time point,indicative of the uptake and accumulation of QUIN in thecytoplasm of the cells over time.

QUIN toxicity and LDH testsNSC-34 cells were incubated with varying concentrations ofQUIN (0–4 lM) over three time periods: 24, 48 and 72 h.From the standard curve (Fig. 5), it can be seen thatincreased cell death was dependent on both QUINconcentration and incubation time, confirmed using linearregression analysis (p < 0.01).

In the presence of 2 lM of QUIN for 48 h, linearregression showed that NMDA antagonists, either individu-ally or in combination, were effective in significantlylowering LDH release with increasing concentrations(p < 0.05) (Fig. 6). To compare if the effects of theantagonists were significantly different from each other,linear regression analysis with dummy variables was used.All four groups were found to be statistically different fromeach other (p < 0.016). Therefore, the effectiveness inconferring neuroprotection among the four groups is:

APV < memantine < MK-801 < APV + memantine + MK-801.

In the presence of 2 lM of QUIN for 48 h, linearregression showed that KYNA was effective in loweringLDH release (p < 0.05) (Fig. 7). PIC, however, assumed anon-linear relationship. From 0 to 0.5 lM PIC protectedNSC-34 cells from QUIN-induced toxicity; beyond 1 lM itexerted an additive toxic effect on NSC-34 cells.

In the presence of increasing concentrations of 1-MT onIFN-c treated BV-2 cells, linear regression showed the partialinhibition of LDH release by NSC-34 cells (p < 0.01)(Fig. 8).

Discussion

The characterization of the KP in NSC-34In the CNS, the KP is fully expressed in microglia,macrophages and neurons, but only partially in astrocytesand oligodendrocytes (Guillemin et al. 2001, 2003, 2007;Lim et al. 2007). Astrocytes lack the enzyme, kynureninemono-oxygenase, and are thus unable to produce theexcitotoxin, QUIN (Guillemin et al. 2001). On the otherhand, oligodendrocytes lack the first KP enzyme, IDO-1strongly limiting their ability to down-regulate the immuneresponse (Lim et al. 2007). This study describes thecomplete expression of the KP enzymes in the mouse motorneuron cell line, NSC-34. The application of IFN-c led tothe activation of IDO-1. IFN-c may also be involved in thepost-translational regulation of the enzyme as IDO-1 mRNAwas present in untreated cells but the protein was absent(Fig. 2).

Like IDO-1, TDO is also one of the first enzymes in theKP (see Fig. 1). It is the predominant enzyme in hepatictissue and may also be present in the CNS, but to a muchlesser extent than IDO-1 (Miller et al. 2004; Guillemin et al.

+ve

cont

rol

–ve

cont

rol

NSC

34N

SC34

+ IF

N-γ

TDO

120.000

(b)(a)

NSC34

NSC34 + IFN-γ

*

*

**100.000

80.000

60.000

% o

f G

AP

DH

40.000

20.000

0.000TDO IDO KATI KMO KYNU 3HAO ACMSD QPRT KATII

IDO

KATI

KMO

KYNU

3HAO

ACMSD

QPRT

GAPDH

Fig. 2 Characterization and semi-quantification of the kynurenine

pathway (KP) enzymes in NSC-34 cells. (a) From left to right: inter-

feron gamma (IFN-c) treated BV-2 cells (positive control), NSC-34

cells, IFN-c treated NSC-34 cells, and PCR without cDNA (negative

control). In untreated cells, except for kynurenine amino transferase

(KAT)-II, all of the nine enzymes tested were present. In treated cells,

both KAT-I and -II were absent. (b) Semi-quantification of PCR pro-

ducts indicated a significant increase (p < 0.05) in tryptophan 2,3-di-

oxygenase (TDO), kynureninase (KYNU) and aminocarboxymuconate

semialdehyde decarboxylase (ACMSD) and quinolinate phosphor-

ibosyl transferase (QPRT) and a significant decrease (p < 0.05) in

KAT-1 in treated cells.

� 2011 The AuthorsJournal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825

Kynurenine pathway expression in motor neuron | 819

2007). In the presence of IFN-c, there is a concomitant rise inboth TDO and IDO-1 mRNA expression in NSC-34 cells.This is in contrast to the inverse relationship reported inprimary human neurons and human SK-N-SH neuroblastomacell line, whereby the increase in IDO-1 was accompanied bya decrease in TDO (Guillemin et al. 2007), reflecting thedissimilar enzyme expression pattern between (i) species(Heyes et al. 1997) and (ii) primary cells versus cell lines(Guillemin et al. 2007; Lim et al. 2007).

The enzymes KAT-I, -II, -III and -IV are responsible forsynthesising the neuroprotectant KYNA (Han et al. 2010).Compared with the other KP enzymes, the mRNA expressionof KAT-1 is much lower. In the presence of IFN-c, itsexpression became undetectable, suggesting a switch in thepathway towards either PIC or NAD+. Further analysis of thehistogram reveals a significant rise in aminocarboxymuco-nate semialdehyde decarboxylase expression, the enzymeleading to PIC synthesis. PIC is the main endogenous metalchelator within the CNS and is also able to antagonize QUINactions (Jhamandas et al. 1990; Beninger et al. 1994). It alsohas significant effects on the immune response as it actssynergistically with IFN-c to amplify the inhibitory effect of

neutrophils, nitric oxide synthase and tumour necrosis factor-a expression (Melillo et al. 1994; Pais and Appelberg 2000;Abe et al. 2004).

Lastly, immunocytochemistry in NSC-34 cells shows thelack of QUIN staining (Fig. 3). QUIN, in concentrationsabove 150 nM, is an excitotoxin for human corticalneurons and astrocytes (Braidy et al. 2009a). The lack ofQUIN synthesis by motor neurons is supported by thecurrent literature showing that astrocytes and neuronscatabolize rather than produce QUIN (Guillemin et al.2001, 2007; Rahman et al. 2009). Hence, under normalconditions, QUIN is only transiently present in motorneurons and is metabolized almost immediately to produceNAD+.

Neurotoxicity of QUIN in NSC-34 cellsOver the last few years, there has been a growing interest inthe role of non-neuronal cells, such as astrocytes andmicroglia, in the neuropathogenesis of ALS (Clement et al.2003). In sporadic ALS, the involvement of non-neuronalcells may be through the inflammatory response, which ischaracteristic of both familial and sporadic ALS. In both

IDO(a)

(b)

TDO KYNU

PIC QUIN SMI32

IDO TDO KYNU

PIC QUIN SMI32

Fig. 3 Characterization of the kynurenine

pathway (KP) in NSC-34 cells by im-

munocytochemistry. (a) In untreated cells,

positive staining (in red) were found in tryp-

tophan 2,3-dioxygenase (TDO), kynur-

eninase (KYNU), picolinic acid (PIC) and

SMI-32; (b) in interferon gamma (IFN-c)

treated cells, in addition to the positive

staining seen in (a), indoleamine-2, 3 dioxy-

genase (IDO-1) immunoreactivity was also

detected. Quinolinic acid (QUIN) was not

detected in either (a) or (b). Nuclei are in blue

(DAPI staining), while SMI-32 was used as a

motor neuron marker.

Journal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825� 2011 The Authors

820 | Y. Chen et al.

instances, histological evidence shows reactive microglia,astrocytes and leucocytes surrounding and accumulating inthe degenerating areas, including the primary motor cortex,the motor nuclei of the brain stem and the corticospinal tract(Kawamata et al. 1992; Sekizawa et al. 1998; Alexianuet al. 2001; Almer et al. 2002).

One of the chief hypotheses explaining the aetiology ofALS is glutamatergic excitotoxicity (Spreux-Varoquauxet al. 2002) and the only drug approved for ALS treatment– riluzole – is an anti-glutamatergic drug. Under inflamma-tory conditions, activated microglia produce excitotoxicamounts of QUIN that can act on the glutamate site of theNMDA receptor and affect both the receptor and glutamate

levels (Stone and Perkins 1981; Guillemin et al. 2004).Being closely associated with motor neurons in ALS, therelease of QUIN into the microenvironment would result in:(i) the stimulation of synaptosomal glutamate release, andinhibition of astrocytic and synaptic vesicular glutamateuptake (Tavares et al. 2000, 2002); (ii) the overstimulation ofNMDA receptors and excitotoxicity on motor neurons; (iii)the decrease in glutamine synthetase activity, thus reducingglutamate-glutamine recycling (Baverel et al. 1990); and (iv)the uptake of QUIN by human motor neurons (Chen et al.2010).

Our immunocytochemical study shows that the uptake ofQUIN by NSC-34 cells is evident as early as 30 min after

0

20

40

60

80

100

120

0 0.05 0.1 0.2 0.5 1 2 4

Concentration of antagonists (µM)

LD

H (

%)

APVMemantineMK801APV + Memantine + MK-801

Fig. 6 The effect of NMDA antagonists on quinolinic acid (QUIN)

toxicity on NSC-34 cells. Increasing concentrations of selective NMDA

antagonists with QUIN resulted in varying decrease in levels of lactate

dehydrogenase (LDH) released by NSC-34 cells after 48 h. The ef-

fectiveness of the NMDA antagonists in conferring neuroprotection

was: APV < memantine < MK-801 < NMDA antagonists combined.

0

20

40

60

80

100

120

0 0.05 0.1 0.2 0.5 1 2 4

Concentration (µM)

LD

H (

%)

24 h48 h72 h

Fig. 5 Neurotoxicity of quinolinic acid (QUIN) on NSC-34 cells. Linear

regression analysis indicated that increasing concentrations of QUIN

resulted in increasing lactate dehydrogenase (LDH) release over time

(24, 48 and 72 h).

0 min 30 min

60 min 90 min

Fig. 4 Time-course study of quinolinic acid

(QUIN) uptake in NSC-34 cells. Im-

munocytochemistry showed positive cyto-

plasmic QUIN staining, increasing in

intensity, over a 90-min period.

� 2011 The AuthorsJournal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825

Kynurenine pathway expression in motor neuron | 821

QUIN incubation (Fig. 4). The gradual increase in intensityof staining also indicates that QUIN is not rapidlymetabolized by QPRT and accumulates over time. Undernormal circumstances, the uptake of small amounts of QUINmay be beneficial if converted quickly by QPRT to NAD+

(Braidy et al. 2009b), a cofactor in adenosine triphosphatesynthesis. Moreover, QPRT is easily saturable, at concentra-tions > 300 QUIN nM in human primary cortical neurons(Rahman et al. 2009). Hence, any large increase in QUINuptake would accumulate to the detriment of the motorneurons. The toxic accumulation of QUIN has beendemonstrated in Alzheimer’s disease brain (Guillemin et al.2005a) and in motor neurons in ALS (Chen et al. 2010). Inaddition, QUIN has also been reported to induce tauphosphorylation in primary human neurons (Guillemin et al.2005a; Rahman et al. 2009), and may be involved in theformation of neurofibrillary tangles. More recently, Pierozanet al. (2010) showed that QUIN increases phosphorylation ofneurofilament and glial fibrillary acid protein in rats.Therefore, it could be speculated that in ALS, QUIN mightalso lead to cytoskeletal disorganization within the motorneurons. We found that the survival of NSC-34 cells stronglydecreased with increasing concentrations of QUIN (Fig. 5). It

should be noted that the QUIN concentrations used in thisstudy are significantly higher than the pathophysiologicallevels found in serum and CSF of ALS patients (Chen andGuillemin 2009; Chen et al. 2010). However, if humanprimary motor neurons were used instead, QUIN concentra-tions would have been significantly lower.

NMDA antagonistsAmong the NMDA antagonists, APV had the lowestneuroprotective effect whereas the APV + MK-801 + mem-antine combination had the highest effect (Fig. 6). For thekynurenines, PIC had a non-linear relationship, with anincreased neurotoxicity at higher concentrations. APV, MK-801 and memantine are all NMDA antagonists but withdifferent properties. APV is a low affinity competitiveantagonist; MK-801 is a non-competitive channel blockerand memantine is a low affinity, uncompetitive, voltage-dependent open-channel blocker. In vitro, they all displayedpartial protection against QUIN but they conferred almosttotal protection in combination (Fig. 6).

The overactivity of the excitatory pathway has beenobserved in many CNS disorders, ranging from acute andchronic neurodegeneration to various neurological illnesses.As such, the modulation of the NMDA receptors is a goal forrational drug design. Nonetheless, it has been a greatchallenge as the NMDA receptor plays a crucial role innormal physiological functioning and many new agents failbecause of the numerous side effects they elicit. Memantineis one of the few successes, having been used clinically forover two decades, primarily in the treatment of dementia. Invitro, memantine protects against glutamate and NMDAmediated cell death (Chen et al. 1992; Krieglstein et al.1996), and we show here that it is also effective againstQUIN. This protection against QUIN has been previouslydemonstrated in vivo as well, preventing QUIN-inducedconvulsions, hippocampal damage and death (Keilhoff andWolf 1992).

Inhibition of QUIN production by microgliaIndoleamine-2, 3 dioxygenase is associated with immuno-tolerance and immuno-suppression. IDO-1 activity is ele-vated in many CNS and peripheral disorders (Chen andGuillemin 2009). 1-MT is an IDO-1 inhibitor able tosuppress KP activation and QUIN production (Lob et al.2008).

Kynurenic acid and PIC are both considered as neuropro-tective intermediates of the KP. KYNA is an endogenousNMDA and 2-amino-3-(5-methyl-3-oxo-1,2-oxazol-4-yl)-propanoic acid (AMPA) antagonist (Stone 1993; Stone andAddae 2002), while PIC is an endogenous iron and zincchelator (Jhamandas et al. 1990) that can be produced byhuman neurons (Guillemin et al. 2007). We found thatKYNA protects NSC-34 cells from QUIN toxicity in a dose-dependent manner (Fig. 7). However, under normal physio-

0

20

40

60

80

100

120

140

0 0.05 0.1 0.2 0.5 1 2 4

Concentration (µM)

LD

H (

%)

KYNAPIC

Fig. 7 Effects of neuroprotective kynurenine pathway (KP) metabo-

lites on quinolinic acid (QUIN) toxicity on NSC-34 cells. In the pre-

sence of kynurenic acid (KYNA), lactate dehydrogenase (LDH)

release by QUIN-treated NSC-34 cells followed a linear fashion de-

crease showing a dose-dependent neuroprotective effect, whereas

picolinic acid (PIC) was non-linear, exerting being neuroprotective a

low concentration but having an additive toxic effect on NSC-34 cells

when concentrations exceeded 1 lM.

020406080

100120

0 0.1 0.2 0.5 1 2Concentration (µM)

LD

H (

%)

Fig. 8 The effect of 1-methyl tryptophan (1-MT) on interferon gamma

(IFN-c)-treated BV-2 supernatant on NSC-34 cells. In IFN-c-treated

BV-2 supernatant, increasing 1-MT concentrations resulted in de-

creased lactate dehydrogenase (LDH) release by NSC-34 cells.

Journal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825� 2011 The Authors

822 | Y. Chen et al.

logical conditions, KYNA levels are too low to exert asubstantial antagonistic effect on QUIN (Heyes et al. 1991).

Like KYNA, PIC is also able to limit QUIN-inducedneurotoxicity, but not the neuroexcitatory component(Jhamandas et al. 1990; Beninger et al. 1994). PIC is lesseffective in comparison with KYNA and acts by attenuatingcalcium-dependent glutamate release and/or chelating en-dogenous zinc (Cockhill et al. 1992; Vrooman et al. 1993;Jhamandas et al. 1998). In high amounts, PIC metalchelating property may explain the additive toxic effect withQUIN on NSC-34 cells (Fig. 7) as it deprives the cells ofzinc and iron, essential for cellular functioning. Wepreviously published that PIC is cytotoxic on primary humanneurons and astrocytes at concentrations over 1 lM (Braidyet al. 2009b). PIC can act synergistically with IFN-c, nitricoxide synthase and tumour necrosis factor-a (Melillo et al.1994; Pais and Appelberg 2000; Abe et al. 2004) and is alsoassociated with the up-regulation in macrophage inflamma-tory protein (MIP)-1a and MIP-1b mRNA in its inhibition ofhuman immunodeficiency virus-1 infection (Cocchi et al.1995; Alkhatib et al. 1996; Bosco et al. 2000). Itsstimulatory effect on MIP secretion is antagonized by IFN-c (Bosco et al. 2000), highlighting the complex interplaybetween PIC and IFN-c and the importance of thesemolecules on the inflammatory response (Bosco et al. 2000).

In conclusion, the NSC-34 cell displays a complete KP,similarly to primary human cortical neurons (Guillemin et al.2007). QUIN is neurotoxic for NSC-34 cells in a dose-dependent manner and we demonstrated that pre-treatmentwith NMDA antagonists (APV, MK-801 and memantine)and compounds targeting QUIN synthesis and/or toxicity (1-MT, KYNA and PIC) had protective effects on NSC-34 cells.As there is currently no effective treatment for ALS, acombination drug therapy involving agents targeting KP mayprovide a novel treatment strategy (Chen et al. 2009).

Acknowledgements

This work was funded by the University of New South Wales,

l’Association pour la recherche sur la sclerose laterale amyotrophi-

que – ARS (France), the Rebecca Cooper Foundation (Australia)

and the Deb Bailey Foundation (Australia).

References

Abe S., Hu W., Ishibashi H., Hasumi K. and Yamaguchi H. (2004)Augmented inhibition of Candida albicans growth by murineneutrophils in the presence of a tryptophan metabolite, picolinicacid. J. Infect. Chemother. 10, 181–184.

Alexianu M. E., Kozovska M. and Appel S. H. (2001) Immune reactivityin a mouse model of familial ALS correlates with disease pro-gression. Neurology 57, 1282–1289.

Alkhatib G., Combadiere C., Broder C. C., Feng Y., Kennedy P. E.,Murphy P. M. and Berger E. A. (1996) CC CKR5: a RANTES,MIP-1alpha, MIP-1beta receptor as a fusion cofactor for macro-phage-tropic HIV-1. Science 272, 1955–1958.

Almer G., Teismann P., Stevic Z., Halaschek-Wiener J., Deecke L.,Kostic V. and Przedborski S. (2002) Increased levels of the pro-inflammatory prostaglandin PGE2 in CSF from ALS patients.Neurology 58, 1277–1279.

Baverel G., Martin G. and Michoudet C. (1990) Glutamine synthesisfrom aspartate in guinea-pig renal cortex. Biochem. J. 268, 437–442.

Beninger R. J., Colton A. M., Ingles J. L., Jhamandas K. and BoegmanR. J. (1994) Picolinic acid blocks the neurotoxic but not the neu-roexcitant properties of quinolinic acid in the rat brain: evidencefrom turning behaviour and tyrosine hydroxylase im-munohistochemistry. Neuroscience 61, 603–612.

Bosco M. C., Rapisarda A., Massazza S., Melillo G., Young H. andVaresio L. (2000) The tryptophan catabolite picolinic acid selec-tively induces the chemokines macrophage inflammatory protein-1alpha and -1 beta in macrophages. J. Immunol. 164, 3283–3291.

Braidy N., Grant R., Adams S., Brew B. J. and Guillemin G. J. (2009a)Mechanism for quinolinic acid cytotoxicity in human astrocytesand neurons. Neurotox. Res. 16, 77–86.

Braidy N., Grant R., Brew B. J., Adams S., Jayasena T. and Guillemin G.J. (2009b) Effects of kynurenine pathway metabolites on in-tracellular NAD+ synthesis and cell death in human primary as-trocytes and neurons. Int. J. Trypto. Res. 2, 61–69.

Bruijn L. I., Miller T. M. and Cleveland D. W. (2004) Unraveling themechanisms involved in motor neuron degeneration in ALS. Annu.Rev. Neurosci. 27, 723–749.

Cashman N. R., Durham H. D., Blusztajn J. K., Oda K., Tabira T., ShawI. T., Dahrouge S. and Antel J. P. (1992) Neuroblastoma x spinalcord (NSC) hybrid cell lines resemble developing motor neurons.Dev. Dyn. 194, 209–221.

Chen Y. and Guillemin G. J. (2009) Kynurenine pathway metabolites inhumans: disease and Healthy states. Int. J. Trypto. Res. 1, 1–19.

Chen H. S., Pellegrini J. W., Aggarwal S. K., Lei S. Z., Warach S.,Jensen F. E. and Lipton S. A. (1992) Open-channel block of N-methyl-D-aspartate (NMDA) responses by memantine: therapeuticadvantage against NMDA receptor-mediated neurotoxicity. J.Neurosci. 12, 4427–4436.

Chen Y., Meininger V. and Guillemin G. J. (2009) Recent advances inthe treatment of amyotrophic lateral sclerosis. Emphasis on ky-nurenine pathway inhibitors. Cent. Nerv. Syst. Agents Med. Chem.9, 32–39.

Chen Y., Stankovic R., Cullen K. M., Meininger V., Garner B., CogganS., Grant R., Brew B. J. and Guillemin G. J. (2010) The kynureninepathway and inflammation in amyotrophic lateral sclerosis. Neu-rotox. Res. 18, 132–142.

Clement A. M., Nguyen M. D. and Roberts E. A. et al. (2003) Wild-typenonneuronal cells extend survival of SOD1 mutant motor neuronsin ALS mice. Science 302, 113–117.

Cocchi F., DeVico A. L., Garzino-Demo A., Arya S. K., Gallo R. C. andLusso P. (1995) Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-suppressive factors produced by CD8+ Tcells. Science 270, 1811–1815.

Cockhill J., Jhamandas K., Boegman R. J. and Beninger R. J. (1992)Action of picolinic acid and structurally related pyridine carboxylicacids on quinolinic acid-induced cortical cholinergic damage.Brain Res. 599, 57–63.

Eggett C. J., Crosier S., Manning P., Cookson M. R., Menzies F. M.,McNeil C. J. and Shaw P. J. (2000) Development and character-isation of a glutamate-sensitive motor neurone cell line. J. Neu-rochem. 74, 1895–1902.

Guillemin G. J., Smith D. G., Kerr S. J., Smythe G. A., Kapoor V.,Armati P. J. and Brew B. J. (2000) Characterisation of kynureninepathway metabolism in human astrocytes and implications inneuropathogenesis. Redox Rep. 5, 108–111.

� 2011 The AuthorsJournal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825

Kynurenine pathway expression in motor neuron | 823

Guillemin G. J., Kerr S. J., Smythe G. A., Smith D. G., Kapoor V.,Armati P. J., Croitoru J. and Brew B. J. (2001) Kynurenine path-way metabolism in human astrocytes: a paradox for neuronalprotection. J. Neurochem. 78, 1–13.

Guillemin G. J., Smith D. G., Smythe G. A., Armati P. J. and Brew B. J.(2003) Expression of the kynurenine pathway enzymes inhuman microglia and macrophages. Adv. Exp. Med. Biol. 527,105–112.

Guillemin G. J., Kerr S. J. and Brew B. J. (2004) Involvement of qui-nolinic acid in AIDS dementia complex. Neurotox. Res. 7, 103–124.

Guillemin G. J., Brew B. J., Noonan C. E., Takikawa O. and Cullen K.M. (2005a) Indoleamine 2, 3 dioxygenase and quinolinic acidimmunoreactivity in Alzheimer’s disease hippocampus. Neuro-pathol. Appl. Neurobiol. 31, 395–404.

Guillemin G. J., Meininger V. and Brew B. J. (2005b) Implications forthe kynurenine pathway and quinolinic acid in amyotrophic lateralsclerosis. Neurodegener. Dis. 2, 166–176.

Guillemin G. J., Smythe G., Takikawa O. and Brew B. J. (2005c) Ex-pression of indoleamine 2, 3-dioxygenase and production of qui-nolinic acid by human microglia, astrocytes, and neurons. Glia 49,15–23.

Guillemin G. J., Cullen K. M., Lim C. K., Smythe G. A., Garner B.,Kapoor V., Takikawa O. and Brew B. J. (2007) Characterization ofthe kynurenine pathway in human neurons. J. Neurosci. 27,12884–12892.

Han Q., Cai T., Tagle D. A. and Li J. (2010) Structure, expression, andfunction of kynurenine aminotransferases in human and rodentbrains. Cell. Mol. Life Sci. 67, 353–368.

Heyes M. P., Brew B. J. and Martin A. et al. (1991) Quinolinicacid in cerebrospinal fluid and serum in HIV-1 infection: re-lationship to clinical and neurological status. Ann. Neurol. 29,202–209.

Heyes M. P., Chen C. Y., Major E. O. and Saito K. (1997) Differentkynurenine pathway enzymes limit quinolinic acid formation byvarious human cell types. Biochem. J. 326(Pt 2), 351–356.

Hilmas C., Pereira E. F., Alkondon M., Rassoulpour A., Schwarcz R.and Albuquerque E. X. (2001) The brain metabolite kynurenic acidinhibits alpha7 nicotinic receptor activity and increases non-alpha7nicotinic receptor expression: physiopathological implications.J. Neurosci. 21, 7463–7473.

Jhamandas K., Boegman R. J., Beninger R. J. and Bialik M. (1990)Quinolinate-induced cortical cholinergic damage: modulation bytryptophan metabolites. Brain Res. 529, 185–191.

Jhamandas K. H., Boegman R. J., Beninger R. J. and Flesher S. (1998)Role of zinc in blockade of excitotoxic action of quinolinic acid bypicolinic acid. Amino Acids 14, 257–261.

Jhamandas K. H., Boegman R. J., Beninger R. J., Miranda A. F. andLipic K. A. (2000) Excitotoxicity of quinolinic acid: modulation byendogenous antagonists. Neurotox. Res. 2, 139–155.

Kalisch B. E., Jhamandas K., Boegman R. J. and Beninger R. J. (1994)Picolinic acid protects against quinolinic acid-induced depletion ofNADPH diaphorase containing neurons in the rat striatum. BrainRes. 668, 1–8.

Kawamata T., Akiyama H., Yamada T. and McGeer P. L. (1992) Im-munologic reactions in amyotrophic lateral sclerosis brain andspinal cord tissue. Am. J. Pathol. 140, 691–707.

Keilhoff G. and Wolf G. (1992) Memantine prevents quinolinicacid-induced hippocampal damage. Eur. J. Pharmacol. 219, 451–454.

Krieglstein J., Lippert K. and Poch G. (1996) Apparent independentaction of nimodipine and glutamate antagonists to protect culturedneurons against glutamate-induced damage. Neuropharmacology35, 1737–1742.

Lim C. K., Smythe G. A., Stocker R., Brew B. J. and Guillemin G. J.(2007) Characterization of the kynurenine pathway in human oli-godendrocytes. Int. Congr. Ser. 1304, 213–217.

Lim C. K., Brew B. J. and Guillemin G. J. (2010) Understanding theroles of the kynurenine pathway in multiple sclerosis progression.Int. J. Tryptophan. Res. 3, 1–11.

Lob S., Konigsrainer A., Schafer R., Rammensee H. G., Opelz G. andTerness P. (2008) Levo- but not dextro-1-methyl tryptophan ab-rogates the IDO activity of human dendritic cells. Blood 111,2152–2154.

McGeer P. L. and McGeer E. G. (2002) Inflammatory processes inamyotrophic lateral sclerosis. Muscle Nerve 26, 459–470.

Melillo G., Cox G. W., Biragyn A., Sheffler L. A. and Varesio L. (1994)Regulation of nitric-oxide synthase mRNA expression by inter-feron-gamma and picolinic acid. J. Biol. Chem. 269, 8128–8133.

Miller C. L., Llenos I. C., Dulay J. R., Barillo M. M., Yolken R. H. andWeis S. (2004) Expression of the kynurenine pathway enzymetryptophan 2, 3-dioxygenase is increased in the frontal cortex ofindividuals with schizophrenia. Neurobiol. Dis. 15, 618–629.

Owe-Young R., Webster N. L., Mukhtar M., Pomerantz R. J., Smythe G.,Walker D., Armati P. J., Crowe S. M. and Brew B. J. (2008)Kynurenine pathway metabolism in human blood-brain-barriercells: implications for immune tolerance and neurotoxicity. J.Neurochem. 105, 1346–1357.

Pais T. F. and Appelberg R. (2000) Macrophage control of mycobacterialgrowth induced by picolinic acid is dependent on host cell apop-tosis. J. Immunol. 164, 389–397.

Perkins M. N. and Stone T. W. (1982) An iontophoretic investigation ofthe actions of convulsant kynurenines and their interaction with theendogenous excitant quinolinic acid. Brain Res. 247, 184–187.

Pierozan P., Zamoner A., Krombauer Soska A., Bristot Silvestrin R.,Oliveira Loureiro S., Heimfarth L., Mello E. S. T., Wajner M. andPessoa-Pureur R. (2010) Acute intrastriatal administration of qui-nolinic acid provokes hyperphosphorylation of cytoskeletal inter-mediate filament proteins in astrocytes and neurons of rats. Exp.Neurol. 224, 188–196.

Rahman A., Ting K., Cullen K. M., Braidy N., Brew B. J. and GuilleminG. J. (2009) The excitotoxin quinolinic acid induces tau phos-phorylation in human neurons. PLoS ONE 4, e6344.

Sekizawa T., Openshaw H., Ohbo K., Sugamura K., Itoyama Y. andNiland J. C. (1998) Cerebrospinal fluid interleukin 6 in amyo-trophic lateral sclerosis: immunological parameter and comparisonwith inflammatory and non-inflammatory central nervous systemdiseases. J. Neurol. Sci. 154, 194–199.

Shaw P. J. (2005) Molecular and cellular pathways of neurodegenerationin motor neurone disease. J. Neurol. Neurosurg. Psychiatry 76,1046–1057.

Spreux-Varoquaux O., Bensimon G., Lacomblez L., Salachas F., PradatP. F., Le Forestier N., Marouan A., Dib M. and Meininger V.(2002) Glutamate levels in cerebrospinal fluid in amyotrophiclateral sclerosis: a reappraisal using a new HPLC method withcoulometric detection in a large cohort of patients. J. Neurol. Sci.193, 73–78.

Stone T. W. (1993) Neuropharmacology of quinolinic and kynurenicacids. Pharmacol. Rev. 45, 309–379.

Stone T. W. and Addae J. I. (2002) The pharmacological manipulation ofglutamate receptors and neuroprotection. Eur. J. Pharmacol. 447,285–296.

Stone T. W. and Perkins M. N. (1981) Quinolinic acid: a potent en-dogenous excitant at amino acid receptors in CNS. Eur. J. Phar-macol. 72, 411–412.

Takikawa O., Yoshida R., Kido R. and Hayaishi O. (1986) Tryptophandegradation in mice initiated by indoleamine 2, 3-dioxygenase. J.Biol. Chem. 261, 3648–3653.

Journal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825� 2011 The Authors

824 | Y. Chen et al.

Tavares R.G., Tasca C. I., SantosC. E.,WajnerM., SouzaD.O. andDutra-Filho C. S. (2000) Quinolinic acid inhibits glutamate uptake intosynaptic vesicles from rat brain. Neuroreport 11, 249–253.

Tavares R. G., Tasca C. I., Santos C. E., Alves L. B., Porciuncula L. O.,Emanuelli T. and Souza D. O. (2002) Quinolinic acid stimulatessynaptosomal glutamate release and inhibits glutamate uptake intoastrocytes. Neurochem. Int. 40, 621–627.

Vrooman L., Jhamandas K., Boegman R. J. and Beninger R. J. (1993)Picolinic acid modulates kainic acid-evoked glutamate release fromthe striatum in vitro. Brain Res. 627, 193–198.

Yoshida R. and Hayaishi O. (1978) Induction of pulmonary in-doleamine 2, 3-dioxygenase by intraperitoneal injection of bac-terial lipopolysaccharide. Proc. Natl Acad. Sci. USA 75, 3998–4000.

� 2011 The AuthorsJournal of Neurochemistry � 2011 International Society for Neurochemistry, J. Neurochem. (2011) 118, 816–825

Kynurenine pathway expression in motor neuron | 825