Báo cáo khoa học: pH dependence, substrate specificity and inhibition of human kynurenine aminotransferase I ppt

pH dependence, substrate specificity and inhibition of human
kynurenine aminotransferase I
Qian Han, Junsuo Li and Jianyong Li
Department of Pathobiology, University of Illinois at Urbana-Champaign, Urbana, IL, USA
Human kynurenine aminotransferase I/glutamine trans-
aminase K (hKAT-I) is an important multifunctional
enzyme. T his study systematically studies the s ubstrates of
hKAT-I and reassesses t he effects of pH, Tris, amino acids
and a-keto acids on the a ctivity of t he enzyme. T he experi-
ments were comprised of functional expression o f the
hKAT-I in an insect cell/baculovirus expression system,
purification of its recombinant protein, and functional
characterization of the purified enzyme. This study demon-
strates that hKAT-I can catalyze kynurenine to kynurenic
acid under physiological pH conditions, indicates indo-3-
pyruvate and cysteine a s efficient inhibitors for hKAT-I, and
also provides biochemical information about the s ubstrate
specificity and cosubstrate inhibition of the enzyme. hKAT-I
is inhibited by Tris under physiological pH conditions, w hich
explains why it has been concluded that t he enzyme could
not efficiently catalyze kynurenine transamination. Our
findings provide a biochemical basis towards understanding
the o verall physiological role of hKAT-I in vivo and insight
into controlling the levels of endogenous kynu renic acid
through m odulation o f the enzyme in the human brain.
Keywords: cysteine; indo-3-pyruvate; kynurenic acid;
kynurenine a minotransferase; p H effect.
In mammals, kynurenine aminotransferase I/glutamine
transaminase K (EC 2.6.1.64; KAT-I) is a multifunctional
enzyme. In vitro, the enzyme catalyzes the t ransamination o f
several amino acids (e.g. glutamine, methionine, aromatic

amino acids including kynurenin e) and a lso possesses
cysteine S-conjugate b-lyase activity (EC 4.4.1.13) [1].
Kynurenic acid (KYNA), the stable product derived from
the kynurenine transamination pathway [2–4], is involved in
several physiological aspects of the central nervous system
(CNS) by acting as an antagonist at both the glutamate-
binding site and the allosteric glycine site of the N-methyl-
D
-aspartate receptor a nd possibly by b locking the 7-nicotinic
acetylcholine receptor [5–8]. Low KYNA levels in the central
nervous system are correlated to cerebral diseases such as
schizophrenia and H untington’s disease [9–13]. Only two
pyridoxal 5¢-phosphate (PLP)-dependent aminotransferases
that are able to catalyze the transamination of kynurenine to
KYNA, arbitrarily termed KAT- I and II , have b een
described in r at and human brains [14–16].
In addition, KYNA is involved in maintaining physio-
logical arterial blood pressure. In rats, the region of the
rostral and caudal medulla in the CNS plays an important
role in regulating cardiovascular function [17–20]. Sponta-
neously hypertensive rats that have higher arterial blood
pressure were found to h ave significantly l ower KAT
activity and KYNA content in their rostral and caudal
medulla than the control rats [20]. Injection o f KYNA into
the rostral ventrolateral medulla of these rats significantly
decreased their arterial pressure [21], which suggests that
KYNA is involved in maintaining physiological arterial
blood pressure. Recently, the mutant KAT-I from all the
strains o f s pontaneously hypertensive rats dis played a ltered
kinetics; l ower initial velocity and K

m
for both kynurenine
and pyruvate [22]. This mutation m ay explain the enhanced
sensitivity to glutamate and nicotine seen in spontaneously
hypertensive rats, suggesting it m ay be related t o a n
underlying mechanism of hypertension and increased sen-
sitivity to stroke [22]. However, Cooper suggested that
another mechanism, i.e. the involvement of altered gluta-
mine transamination and sulfur and aromatic amino acid
metabolism should also be considered [1].
Although a number o f studies described t he characters of
the enzyme ([1] and references therein), only an insect
homologue, Aedes aegypti kynurenine aminotransferase [23]
and a bacterium homologue [24] were systematically
characterized using purified recombinant proteins. To
compare the characteristics of KAT-Is in different living
organisms, determine s ubstrate s pecificity, and evaluate the
possible effect of other amino acids and keto a cids on
hKAT-I, we expressed the enzyme in a baculovirus/insect
cell protein expression system. Our large scale hKAT-I
expression and subsequent purification enabled us to obtain
a large amount (mg range) of pure hKAT-I f or extensive
biochemical characterization. Our results revealed some
Correspondence to J. L i , Department of Pathobiology, University of
Illinois at Urbana-Champaign, 2001 South Lincoln Avenue, Urbana,
IL 61802, USA. Fax: +1 217 2447421, Tel.: + 1 217 2443913,
E-mail:
Abbreviations:AeKAT,Aedes aegypti kynurenine aminotransferase;
CNS, central nervous system; hKAT-I, human kynurenine amino-
transferase I; HTS, high-titre viral stocks; KYNA, kynurenic acid;

aKMB, a-keto-methylthiobutyric acid; PLP, pyridoxal 5¢-phosphate;
Sf9, Spodoptera frugiperda insect cells.
Enzymes: kynurenine a mino transferase I /glutamin e transaminase K
(EC 2.6.1.64); cysteine 5 -conjugat e b-lyase (EC 4.4.1.13).
(Received 2 3 August 2 004, revised 13 October 2004,
accepted 21 October 2004)
Eur. J. Biochem. 271, 4804–4814 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04446.x
interesting biochemical characteristics of hKAT-I, which
have not been systematically addressed before. For example,
the pH p rofile of hKAT-I exhibits high activity under
neutral conditions, which contrasts its reported pH profile
showing that the enzyme exhibited high activity only at
basic pH values [15,25]. Tris buffer significantly inhibited
enzyme activity at neutral c onditions, but showed no
inhibition under basic conditions, which might explain
why previous studies have reported that hKAT-I had
limited activity at physiological pH conditions and
displayed optimum activity at fairly basic conditions.
Moreover, for the first time, we found that cysteine and
indo-3-pyruvate are effective inhibitors of the enzyme
in vitro. Our data provide a better overall picture of the
enzyme and should be helpful in a comprehensive under-
standing of the role of hKAT-I, especially in KYNA
biosynthesis in the human brain.
Experimental procedures
Enzyme expression and purification
Construction of recombinant transfer vectors. The codin g
sequence of hKAT-I was am plified from first s trand human
liver cDNA (Clontech, Palo Alto, CA, USA) using a
specific forward (5¢-

CTCGAGATGGCCAAACAGCTG
CAG) and reverse primer (5-
AAGCTTAGAGTTCCAC
CTTCCACTT) containing a XhoIandaHindIII restriction
site (underlined sequence), respectively. The P CR products
were cloned into a TOPO TA cloning vector and then
subcloned into a baculovirus transfer vector pBlueBac4.5
(Invitrogen, Carlsbad, CA, USA). Recombinant transfer
vectors were sequenced and confirmed to ensure that the
inserted DNA sequences were in frame.
Production of recombinant baculoviruses. Recombinant
pBlueBac4.5 transfer vectors were cotransfected with
linearized Bac-N-Blue
TM
Autographa californica multiple
nuclear polyhedrosis virus DNA in the presence of
InsectinPlus
TM
insect cell-specific liposomes to Spodoptera
frugiperda (Sf9) insect cells (Invitrogen). The recombinant
baculoviruses were purified through the plaque assay
procedure. Blue putative recombinant plaques were trans-
ferred to 12-well microtitre plates and amplified in Sf9 cells.
Viral DNA was isolated for PCR analysis to d etermine the
purity of the recombinant viruses. High-titre viral stocks
(HTS) for individual recombinant viruses were generated by
amplification in Sf9 cell suspension culture.
Recombinant protein expression. Sf9 insect cells were used
for protein expression. The cells were cultured at 27 °Cinan
Ultimate Insect

TM
serum-free medium (Invitrogen) supple-
mented with 10 unitsÆmL
)1
heparin (Sigma, St. Louis, MO,
USA) in culture spinner flasks and constant stirring at
80 r.p.m. W hen the cell d ensity reached 2 · 10
6
cellsÆmL
)1
,
they were inoculated with the HTS of recombinant
baculoviruses at a multiplicity of infection of six viral
particles per cell.
Purification of recombinant hKAT-I. Sf9 cells in 2 L of
cell culture were harvested on the fourth day after
hKAT-I recombinant virus inoculation by centrifugation
(800 g for 15 min at 4 °C) and the ce ll pe llets were
dissolved in a lysis buffer containing 25 m
M
phosphate,
0.1 m
M
pyridoxal 5¢-phosphate (PLP), 2 m
M
dithiothre-
itol, 2 m
M
EDTA, 1 m
M

phenylmethylsulfonyl fluoride
and 150 m
M
NaCl with a final pH of 7.4. After
incubation on ice for 30 min, cell lysates were centrifuged
at 18 000 g fo r 20 min at 4 °C and the supernatant was
collected and a ssayed for KAT-I activity. Soluble
proteins in the supernatant were precipitated with 65%
saturation of ammonium sulfate. Protein precipitate was
then redissolved in 10 % saturated ammonium sulfat e
and applied to a phenyl sepharose column. A linear
gradient of ammonium sulfate from 10% to 0% in
10 m
M
sodium phosphate buffer (pH 7.0) was used for
protein elution. The hKAT-I active fractions were
pooled, dialyzed and then separated by DEAE Sepharose
chromatography with a linear NaCl gradient (0–500 m
M
)
in the same phosphate buffer. The active fractions were
collected, concentrated and then separated by a Super-
dex
TM
200 gel filtration column. The hKAT-I active
fractions were collected and concentrated. The purity of
the enzyme was assessed by SDS/PAGE a nalysis.
hKAT-I content was determined by a Bio-Rad (Hercules,
CA, USA) protein assay kit using bovine serum albumin
as a standard. The concentration of hKAT-I stock

solution was adjusted to 20 mgÆmL
)1
in 20 m
M
phos-
phate buffer (pH 7.5), aliquoted into 200 lLmicrocen-
trifuge tubes with 10 lLineachandfrozenat)80 °C.
Biochemical characterization
hKAT-I activity assay. All chemicals were purch ased from
Sigma Chemical Company unless otherwise specified.
hKAT-I activity assay was based on methods described in
previous reports [26,27]. Briefly, a typical r eaction m ixture
of 50 lL containing 5 m
M
kynurenine, 2 m
M
a-ketobu ty-
rate, 40 l
M
PLP and 2 lg hKAT-I was prepared using a
200 m
M
phosphate buffer, pH 7.5. The reaction mixture
was incubated for 10 min at 45 °C, and the reaction was
stopped by adding an equal volume of 0.8
M
formic acid.
Supernatant w as obtained by centrifugation o f t he reaction
mixture at 1 5 000 g at 4 °C for 10 min and analyzed by
HPLC-UV at 330 nm for KYNA. The amount of KYNA

formed in the reaction mixture was calculated based on a
standard curve generated using authentic KYNA and
the specific activity of the enzyme was expressed as
lmolÆmin
)1
Æmg
)1
.
Effect of buffer and pH on hKAT-I. To determine the
effect of pH on hKAT-I activity, a buffer mixture consisting
of 100 m
M
phosphate and 100 m
M
boric acid was prepared
and the pH of the buffer was adju sted to 6.0, 6.5, 7.0, 7 .5,
8.0, 8.5, 9.0, 9.5, respectively. The buffer mixture was s elec-
ted to maintain a relatively consistent buffer composition
and i onic strength, yet have sufficient buffering capacity for
relatively broad pH range. A typical reaction mixture
containing 5 m
M
kynurenine, 2 m
M
a-ketobutyrate, and
2 lg hKAT-I was prepared using the buffer mixture at
different pHs. The reaction mixture was incubated a nd
analyzed as described in the hKAT-I activity assay. Initial
results showed that the specific activity of hKAT-I in the
reaction mixture prepared in Tris buffer was much lower

Ó FEBS 2004 Characterization of human kynurenine aminotransferase I (Eur. J. Biochem. 271) 4805
than in the above buffer mixture. To compare hKAT-I
activity in different buffers, 200 m
M
phosphate alone with
pH 6.0–8.0, and 200 m
M
Tris alone with pH 7.5–8.5 were
prepared and used for hKAT-I activity assays under the
same conditions.
Spectral analysis. PLP has an absorption peak in the
visible region under neutral or weak basic conditions. Its
visible peak shifts to wards longer wavelengths when
associated with transaminase and diminishes upon forma-
tion of pyridoxamine after reacting with amino group
donors (the half reaction of the overall transaminase-
mediated reactions). The spectrum of hKAT-I i n phosphate
buffer, pH 7.5, was analyzed using a Hitachi U2001 double-
beam spectrophotometer and compared to t he spectrum o f
free PLP i n the same buffer. The s ame spectral a nalysis was
also used to evaluate the potential interaction o f P LP and
Tris through comparison of the spectral characteristics of
PLP in Tris buffer at different pHs with those of PLP in
phosphate buffer at the same pH conditions.
Substrate specificity. To determine the possible transam-
ination activity of hKAT-I to o ther amino a cids, a different
amino acid at varying concentrations (0.1–32 m
M
)wasused
to replace kynuren ine and 16 m

M
a-ketobutyrate used a s an
amino group acceptor in the t ypical reaction mixture (50 lL
total volume prepared in 200 m
M
phosphate buffer, pH 7.5)
specified in the hKAT-I activity assay. The mixture was
incubated for 10 min a t 4 5 °C. The product was quan tified
based on the detection o f o-phthaldialdehyde thiol (OPT)-
amino acid produ ct conjugate by HPLC with fluorescent
detection ( excitation: 325 nm; emission: 465 nm) after their
corresponding reaction mixtures were derived by OPT
reagent [28]. To determine the substrate specificity for
a-keto acids, 16 individual a-keto acids were tested for t heir
ability to f unction as the amino group acce ptor for hKAT-I.
In the assays, a different a-keto acid at varying concentra-
tions (0.25–16 m
M
) was used to replace a-ketobutyrate in
the presence of 15 m
M
kynurenine in the typical reaction
mixture a nd the r ate of KYNA production was determined
as described in the hKAT-I activity assay.
Presence of other amino acids or other keto acids on
hKAT-I catalyzed KYNA formation. Analysis of substrate
specificity revealed t hat a number of amino acids a nd a-keto
acids can serve as the amino group donor and acceptor,
respectively, for hKAT-I. To determine the effect of a
competing a mino acid or keto acids on hKAT-I catalyzed

KYNA production from kynurenine, a different amino acid
(with a final concen tration of either 2 m
M
or 32 m
M
)ora
different a-keto acid (with a final concentration of 2 m
M
)
was incorporated into the typical re action mixture c ontain-
ing 5 m
M
(for testing amino acids) or 15 m
M
(for testing
a-keto acids) kynurenine, 2 m
M
a-ketobutyrate, and 2 lg
hKAT-I in a total volume of 50 lL and the enzyme
activitywasassayedinthesamemannerasdescribedfor
the hKAT-I activity assay.
All assays were performed in at least triplicate. The results
for the e ffects o f ke to acids and amino acids w ere analyzed
by using the Student’s t-test. The kinetic parameters of the
recombinant enzyme towards different amino acids or
a-ke to acids were calculated by fitting the experimental
data to the Michaelis–Menten equation using the
ENZYME
KINETICS MODULE
(SPSS Science, Chicago, IL, USA).

Results
Spectral characteristics of the recombinant hKAT-I
Purified hKAT-I showed a single protein band on SDS/
PAGE with a relative molecular mass of 46 kDa (Fig. 1A,
insert), which closely matches its calculated mass
(47 875 Da). Spectral analysis of the purified enzyme in
phosphate buffer (pH 7.5) revealed an absorption peak in
the visible region with a k
max
at 422 nm, which corresponds
to protein associated PLP. The protein associated PLP w as
easily distinguished from free PLP that had its visible
absorption peak with a k
max
at 388 nm (Fig. 1A). Based on
protein concentration in comparison with the absorbance
of protein associated PLP, an approximate 1 : 1 ratio of
the protein against PLP was established. The 4 22 nm
A
B
Fig. 1. Spectral characteristics of hKAT-I. (A) The absorption peak of
protein a ssociated PLP in the visible region in phosphate buffer,
pH 7.5, has a k
max
at 422 nm, which can be distinguished from the
corresponding absorption peak of free PLP that has a k
max
at 388 n m.
Insert is SDS/PAGE gel that illustrates s oluble protein f rom hKAT-I
recombinant baculovirus in fected insect cells (lane 1), purified hKAT-I

(lane 2) and protein standard (lane 3), respectively. (B) Spectral
changesofhKAT-Ifollowingtheadditionof10m
M
glutamine into
hKAT-I solution. The concomitant decrease of the 422 nm peak an d
increase of the 335 nm peak indicate the c onversio n of the enzyme
associated PLP to pyridoxamine. The reaction mixture wa s scanned a t
1 min intervals f ollowing glutamine a ddition.
4806 Q. Han et al.(Eur. J. Biochem. 271) Ó FEBS 2004
absorption peak was rapidly diminished upon incorporation
of glutamine into the reaction mixture with the conco mitant
formationofanewabsorptionpeakwithak
max
at 33 5 nm
(Fig. 1B), indicating the formation of pyridoxamine f rom
the e nzyme associated PLP. Addition of kynurenine dimin-
ished the 422 nm peak, but the absorption peak of
kynurenine overlapped with the pyridoxamine peak and
partially overlapped with the 422 nm peak (not shown).
Incorporation of a-ketobutyrate or other a-keto acid
concomitantly decreased and increased the 335 nm and
422 nm peak, respectively, and the relative dimensions of
the pyridoxamine peak (335 nm) and enzyme-PLP peak
(422 nm) were dependent on the molar ratio between the
amino grou p donor and the acceptor in t he mixture. These
results established that the expressed recombinant protein is
folded properly with its PLP p rosthetic group and r etains its
biochemical activities.
Effect of buffer and pH on hKAT-I activity
When the phosphate and borate buffer mixture, adjusted

to pH 6.0–9.5, was used to prepare hKAT-I/kynurenine/
a-ketobutyrate reactio n mixtures, hKAT-I showed little
activity with pH 6.0, beca me fairly active at pH 6.5–7.0, and
displayed high activity at pH 7.5–9.0 (Fig. 2A). The high
activity of hKAT-I in catalyzing the kynurenine t o K YNA
Fig. 2. Effect of pH a n d Tr is buffer on hKAT-I a ctivity. hKAT-I was
incubated in the presence of 5 m
M
kynurenine and 2 m
M
a-ketobutrate
as described in Experimental procedures. (A) hKAT-I activity profiles
at diffe rent pHs in a bu ffer mixture contain ing 100 m
M
phosphate and
100 m
M
boric a cid. The pH ranged from 6.0 to 9.5. (B) Inhibition of
hKAT-I activity by Tris at ne utral and weak basic condition: ))),
hKAT-I activity in a r eaction mixture prepared using 200 m
M
Tris
buffer; ÆÆÆÆ, hKAT-I a ctivity in a reac tion mixture prepared using
200 m
M
phosphate buffer.
Fig. 3. Interaction of free PLP with Tris amine under different pH
conditions. One h undred microliters of either 20 m
M
phosphate b uffer

or Tris buffer at pH 7 .0, 7.5, 8.0, o r 8.5 w as mixed with 400 lLof
0.25 m
M
PLP prepared in distilled water and the spectra of the
mixtures were recorded using a Hitachi U2001 double-beam spectro-
photometer 2.0 m in after b uffer add ition. (A) Sp ectra of PL P in the
presence of phosphate, pH 8.0 (trace 1) a nd Tr is, pH 8.0 (trace 2).
(B) Spectra of PLP in Tris p H 7.0 (trace 1), 7.5 (trace 2), 8.0 (t race 3)or
8.5 (trace 4). (C) Spectra of PLP in phosphate pH 8.0 before (trace 1)
and after incorporation of 4 m
M
glutamine (trace 2),whichservesasa
control for suggesting a similar interaction of Tris amine with PLP a s
that of amino acid s ubstrate.
Ó FEBS 2004 Characterization of human kynurenine aminotransferase I (Eur. J. Biochem. 271) 4807
Table 1. Kinetic parameters of hKAT-I towards different amino acids and comparison with the rate of purified brain hKAT-I and the k
cat
/K
m
of other
aminotransferases. The activities w ere measured as d escribe d in E xperimental procedures. T he a-keto butyrate concentrations were 16 m
M
in the
presence of different concentrations of amino acids. The kinetic p arameters of the enzyme to amino acids were calculated by fitting the experimental
data to th e M ichaelis–Menten e quation u sing the
ENZYME KINETICS MODULE
. BGlnAT, bacterium homologue, glutamine:phenylpyruvate
aminotransferase from Thermus thermophilus HB8 [24]; AeKAT, insect homologue, from Aedes aegypti [23]; RGlnAT, rat glutamine transaminase
K/KAT-I[30];BhKAT-I,hKAT-I,purifiedfromhuman brain [ 29]; 3-HK, 3-hydroxykynurenine.
Amino acid

hKAT-I BGlnAT AeKAT RGlnAT BhKAT-I
K
m
m
M
k
cat
min
)1
k
cat
/K
m
min
)1
Æm
M
)1
k
cat
/K
m
min
)1
Æm
M
)1
k
cat
/K

m
min
)1
Æm
M
)1
k
cat
/K
m
min
)1
Æm
M
)1
rate
lmolÆmin
)1
Æmg
)1
Glutamine 2.8 ± 0.5 440.5 ± 28.7 157.3 4 147.8 0.04 1.8
Phenylalanine 1.7 ± 0.3 91.0 ± 4.8 53.5 13 80.9 0.17 0.37
Leucine 7.6 ± 3 339.9 ± 43.1 44.7 1.3 · 10
)4
22 < 0.05
Kynurenine 4.7 ± 0.4 201.1 ± 19.2 42.8 5.7 40.2 2.1
Tryptophan 1.2 ± 0.3 43.1 ± 3.4 35.9 3 22 0.17
Methionine 6.4 ± 0.9 215.4 ± 14.4 33.7 6.3 116.6 0.012 < 0.05
Tyrosine 3.2 ± 0.4 91.0 ± 4.8 28.4 200 154.7 0.0031 < 0.05
Histidine 5.4 ± 1 143.6 ± 14.4 26.6 1.2 · 10

)3
112 < 0.05
Cysteine 0.7 ± 0.1 9.6 ± 0.48 13.7 158.8 < 0.05
Amino-butyrate 21.3 ± 4.7 38.3 ± 4.8 1.8 3.4
Asparagine 23.1 ± 5.7 14.4 ± 0.14 0.6 2.2 · 10
)3
37.8
Glycine
a
Alanine
a
4.8 · 10
)5
6.3
Arginine
a
1.5 · 10
)5
< 0.05
Serine
a
5.3 · 10
)4
10.6
Lysine
a
Threonine
b
4 · 10
)6

< 0.05
Isoleucine
b
Aspartate
b
1.4 · 10
)5
< 0.05
Glutamate
b
2.3 · 10
)5
< 0.05
Valine
b
Aminoadipate
b
< 0.05
3-HK
b
< 0.05
a
The enzyme activity towards these amino acids at 32 m
M
are 0.06–0.4 lmolÆmin
)1
Æmg
)1
, K
m

>32m
M
.
b
The enzyme activity towards
these amino acids at 32 m
M
was < 0.05 lmolÆmin
)1
Æmg
)1
, and the enzyme shows no detectable activity towards 3-hydroxykynurenine.
Fig. 4. Transamination activity o f hKAT-I towards different amino acids with a-ketobutyrate a s an a mino acceptor. Pur ified recombinant hK AT-I
was incubated with each of the 24 ami no acids at 20 m
M
(A), except 3-hydroxykynurenine, which was not tested at 20 m
M
due to its low solubility in
aqueous solutio n, an d 2 m
M
(B) in the presen ce of 16 m
M
a-ketobutyrate or oxaloacetate (for activity towards aminobutyrate), respectively as
described in Experimental procedures. The activity w as quantified by t he amount of am inobut yrate or aspartate pro duced in the r eaction mixture.
4808 Q. Han et al.(Eur. J. Biochem. 271) Ó FEBS 2004
pathway under physiological conditions contrasted with
previous reports that mammalian KAT-I had extremely
limited activity in catalyzing the production of KYNA
from kynurenine under neutral conditions [14,15,25,29].
However, when the same reaction mixtures were prepared

in Tris buffer alone, as described in some previous reports,
hKAT-I showed extremely low activity at pH 7.5, and
essentially no activity at pH 8.0, but became active at
pH 8.5 (Fig. 2B, dashed line), which is quite similar to
earlier studies [14,15,25,29]. It is intriguing that the pH
profiles of enzyme activity were so different between Tris
and phosphate buffer (Fig. 2B, dotted line).
The aldehyde group of PLP can react with a primary
amine to form a fairly stable Schiff base, so the inhibition of
Tris on hKAT-I might be due to competition of the amino
group on Tris molecules with enzyme associated PLP.
When 400 lL of a PLP solution, prepared in distilled water
at 0.25 m
M
, was mixed with 100 lLof20m
M
phosphate
bufferor20 m
M
Tris b uffer at pH 8.0, the visible absorption
peak of PLP was shifted towards longer wavelengths a fter
the addition of Tris buffer, as compared to that of PLP a fter
the addition of phosphate buffer (Fig. 3A). Other than a
slight change in peak dimension, pH changes of phosphate
buffer ( pH 6.5–8.0) did not lead to a noticeable spectral shift
of the PLP absorption peak (not shown), but apparent
spectral shift towards longer wavelengths was observed in
the PLP solution after the addition of Tris at pH 7.5, 8.0,
and 8 .5, respectively, as compare d to the addition of Tris at
pH 7.0 (Fig. 3B). The same s pectral shift was also observed

when glu tamine was added to the phosphate prepared PLP
Fig. 5. Cosubs trate specificity of h KAT-I . hKAT-I was incubated in t he presence of k y nurenine at 1 5 m
M
and a different am ino group acceptor
(a-keto acid) at concentrations ranging from 0.1 to 16 m
M
as described in Experimental procedures. The activity was quantified by the amount of
KYNA produced in the reaction mixture.
Ó FEBS 2004 Characterization of human kynurenine aminotransferase I (Eur. J. Biochem. 271) 4809
solution (Fig. 3C). These results suggest that the amino
group of the Tris molecule probably interacts with enzyme
associated PLP under basic conditions, thereby decreasing
its transaminase activity. As a similar spectral shift of PLP
towards a longer wavelength was observed in Tris buffer
at pH 8.5, other competing mechanisms between the Tris
amine and the amino acid substrates may be involved as
well, which requires further elucidation.
Substrate study of hKAT-I
hKAT-I was tested for aminotransferase activity towards
different amino acids using a-ketobutyrate as a primary
amino group acceptor. The selection of a-ketobutyrate as
the amino group acceptor was based on initial r esults that
this keto acid showed no substrate inhibition at saturating
concentrations (discussed below). hKAT-I showed detect-
able activity towards aromatic amino acids (including,
kynurenine, phenylalanine, tryptophan and tyrosine); sulfur
containing amino a cids (including, methionine an d cysteine)
and other aliphatic a mino acids (including, glutamine,
leucine, histidine, and aminobutyrate). This is different from
a p revious study reporting that hKAT-I exhibited relatively

high activity to only four a mino acid substrates [29]
(Table 1). h KAT-I also exhibited low activity to other
tested amino acids at high concentrations (Fig. 4). Kine tic
results provided a better view regarding the efficiency of
hKAT-I towards the individual amino acids (Table 1).
Basedontheparameterofk
cat
/K
m
, it is apparent that
hKAT-I is efficient in catalyzing the transamination of a
number of amino acids, including glutamine, phenylalanine,
leucine, kynurenine, methionine, tyrosine, histidine, cys-
teine, and aminobutyrate (Table 1). Although KAT-I
enzymes from different species, including insect kynurenine
aminotransferase from Aedes aegypti (AeKAT) [23], rat
glutamine transaminase K/KAT-I [30] and hKAT-I,
behave in a similar manner, they display apparent differ-
ences in substrate preference. For example, hKAT-I is most
efficient in catalyzing the transamination of glutamine,
AeKAT is most efficient toward cysteine and tyrosine,
bacterium e nzyme is most efficient toward tyrosine, and rat
KAT-I is most efficient to phenylalanine (Table 1). These
differences suggest that the enzyme may have different
functional priorities in different species.
Sixteen a-keto acids wer e tested for their potential as the
amino g roup acceptor for hKAT-I with 15 m
M
kynurenine
as the amino group donor. Among them, 12 a-keto acids

displayed detectable activity after 10 min of incubation
(Fig. 5) and five (a-ketoglutarate, a-ketoiosleucine, indo-
3-pyruvate, a-ketoadipate an d a-ketovaline) showed detect-
able activity only when i ncubation time lasted for an hour.
Among the 12 a-keto acids capable of functioning as amino
group acceptors for hKAT-I, a-ketobutyrate, mecapto-
pyruvate and oxaloacetate showed no substrate inhibition
at saturating concentrations, but the others, especially
p-hydroxy-phenylpyruvate, aKMB and a-ketovalerate,
showed substrate inhibition at relatively low concentrations
(Fig. 5). Although pyruvate has been the most commonly
used amino group acceptor for KAT-I activity assays, it
was much less efficient as the amino group acceptor for
hKAT-I than a number o f other a-keto acids listed in
Table 2 . Due to substrate inhibition, the K
m
and k
cat
/K
m
could not be determined for p-hydroxy-phenylpyruvate,
aKMB, a-ketovalerate, and a-ketocaproic acid (Table 2
and Fig. 5), but based on reaction r ates, they should also be
more efficient than pyruvate as the amino group acceptor
for hKAT-I.
Effects of other amino acids on hKAT-I catalyzed
kynurenine transamination
BasedontheK
m
of hKAT-I towards different amino a cids,

tryptophan, glutamine, phenylalanine, methionine, histi-
dine, tyrosine, cysteine and leucine have either similar
affinity or better affinity to hKAT-I than kynurenine;
accordingly, the presence o f any of these amino acid s i n the
kynurenine/hKAT-I/a-keto acid mixture should lead to the
competitive inhibition of hKAT-I activity towards kynur-
enine. When 32 m
M
of tryptophan, glutamine, phenylalan-
ine, cysteine, methionine, histidine, tyrosine or leucine was
incorporated into the kynurenine/hKAT-I/a-ke to acid
reaction mixture with 5 m
M
kynurenine, the r ate of K YNA
formation was significantly decreased (Fig. 6B). When
2m
M
of a d ifferent amino a cid was incorporated into the
kynurenine/hKAT-I/a-keto a cid reaction mixture, the rate
of KYNA production was decreased only by tryptophan,
glutamine, phenylalanine, and cysteine at 70%, 60%, 60%,
and 30%, respectively (Fig. 6A). Ap parently, the decrease
in the rate of KYNA production in the hKAT-I/kynure-
nine/a-keto acid reaction mixture in the presence of a
different amino acid was due to competitive inhibition,
because t he extent of inhibition approximately matched the
K
m
value of th e corresponding competing a mino acid in the
reaction mixture. Cysteine derivatives were proposed to be

modulators for KYNA production in the mammalian brain
Table 2. Kinetic parameters of hKAT-I towards a-keto acids. The
activities were measured as described in Experimental procedures. The
K
m
and k
cat
for amino acceptors were derived by using varying con-
centrations of individual a mino acc eptors in t he presen ce of 15 m
M
of
kynurenine. The parameters were calculated by fitting the experimental
data to the Michaelis–Menten e quat ion using the
ENZYME KINETICS
MODULE
(Fig. 5).
a-Keto acids
K
m
m
M
k
cat
(min
)1
)
k
cat
/K
m

(min
)1
Æm
M
)1
)
a-Ketoleucine 1 ± 0.3 296.8 ± 28.7 247.4
Glyoxylate 1.5 ± 0.5 263.3 ± 28.7 175.5
Phenylpyruvate 0.8 ± 0.4 110.1 ± 19.2 137.6
Mercaptopyruvate 2.5 ± 0.4 234.6 ± 14.4 93.8
a-Ketobutyrate 3 ± 0.4 234.6 ± 9.6 78.2
Oxaloacetate 4.2 ± 0.4 143.6 ± 9.6 34.2
Pyruvate 12.1 ± 4.9 28.7 ± 4.8 2.4
a-Ketocaproic acid
a-Ketovalerate
aKMB
High activity, but unable to calculate kinetic
parameters because of their substrate
inhibition at low concentration
p-Hydroxy-phenylpyruvate
a-Ketoglutarate
a-Ketoisoleucine
a-Ketoadipate
Activity was detectable only when incuba-
tion time lasted at least an hour, ranking
as listed.
a-Ketovaline
Indo-3-pyruvate
4810 Q. Han et al.(Eur. J. Biochem. 271) Ó FEBS 2004
[31,32] and cysteine had an intriguing effect on AeKAT

[23], so its effect on hKAT-I mediated kynurenine
transamination was tested more thoroughly. Cysteine
showed apparent inhibition of hKAT-I-catalyzed KYNA
production from kynurenine a t concentrations of 2 m
M
(Fig. 6C), but did not stimulate hKAT-I acitivity at low
concentration as seen in the Ae KAT catalyzed reaction [23].
For most of the other tested amino acids, no inhibition on
KYNA production was observed (Fig. 6A,B), which c ould
be explained by their rather low affinity to the enzyme.
Effects of other keto acids on hKAT-I activity
Ability to function as the amino group acceptor for a
number of biologically relevant keto acids and their
substrate inhibition above certain concentrations (Fig. 5)
indicated that the presence of two keto acids might have
either a positive or negative impact on hKAT-I mediated
KYNA production, as compared to the presence o f a single
amino group acceptor. To test this hypothesis, a different
keto acid was added to the kynurenine/hKAT-I/a-ketobu-
tyrate mixture and the rate of KYNA production in the
reaction mixture was compared with that of the control
reaction mixture with a-ketobutyrate alone. A number of
a-ke to acids significantly increased the rate of KYNA
production, but indo-3-pyruvate showed significant inhibi-
tion on the enzyme, and a-ket oglutarate, a-ketoiosleucine,
Fig. 6. Effect of other amino acids on hKAT-I activity. Assay condi-
tions were sim ilar to those described in E xperimental procedures,
except kynurenine and a-ketob utyrate concentrations were 5 m
M
and

2m
M
, respectively. The concentrations of amino acids tested were
2m
M
(A) and 32 m
M
(B), respectively. Cysteine was tested from
0.3 m
M
to 16 m
M
(C). Th e a ctivity w as quantified by the amo unt o f
KYNA produced in the reaction mixture. *P <0.5, **P <0.01
significant difference from the control; 3-HK, 3-hydroxykynurenine.
Fig. 7. Eff ect of the multip le a-keto acids on hKAT-I-catalyzed KYNA
production. The activities were quantified by the amount of KYNA
produced in the reaction mixtures. The vo lume of the reaction m ixture
was 50 lLconsistingof2lgofhKAT-I,15m
M
kynurenine a nd two
different a-keto acids. O ther conditions were sim ilar to t hose described
in Experimen tal procedures. (A) Rate of KYNA production in the
hKAT-I and kynurenine mixture in the presence of 2 m
M
of
a-ketobutyrate and 2 m
M
of a diff erent a- keto acid. *P <0.5,
**P < 0.01, significant difference from the control. (B) Effect of

different concen trations of indo -3-pyruvate on hKAT-I activity.
p-HPP, p-hydroxyphenylpyruvate.
Ó FEBS 2004 Characterization of human kynurenine aminotransferase I (Eur. J. Biochem. 271) 4811
a-ketoadipate, a-ketovaline, pyruvate, and phenylpyruvate
had no significant effect on the rate of KYNA production
by hKAT-I (Fig. 7A). Because indo -3-pyruvate has not
been reported as an inhibitor to hKAT-I in other studies, its
inhibition of hKAT-I at a broad range of concentrations
wasfurthertested.AnalysisofhKAT-Iactivityinthe
presence of varying concentrations of indo-3-pyruvate
showed that the compound inhibited hKAT-I a ctivity at a
very low concentration of 0.08 m
M
, and completely abol-
ished the enzyme activity at 5.0 m
M
(Fig. 7B).
Discussion
Analysis of substrate specificity confirmed that h KAT-I is a
multifunctional aminotransferase. Kinetic analysis of the
enzyme to wards different amino acids showed that the
enzyme is efficient in catalyzing the transamination of
glutamine, phenylalanine, leucine, kynurenine, tryptophan,
methionine, tyrosine, histidine, cysteine and amino-
butyrate, which contrasts an earlier report showing that
purified hKAT-I exhibited high activity to only four
individual amino acids (kynurenine, glutamine, phenyl-
alanine, and t ryptophan) [29] (Table 1). The large spectrum
of amino acid substrates of KAT-I supports the proposed
role of the enzyme in sparing the essential amino acids

methionine, histidine, phenylalanine and tyrosine [1] and
providing a mechanism t o maintain a continual equilibrium
among the amino acids [33]. Moreover, the high activity
of hKAT-I towards kynurenine under physiological pH
conditions provides the basis for suggesting its function in
brain KYNA synthesis.
By studying the pH profile of the enzyme, we demon-
strated, for the first time, that h KAT-I has a broad optimal
pH range, and is c apable of efficiently catalyzing the
kynurenine to KYNA pathway under physiological condi-
tions, which contrasts the published pH profile of hKAT-I
[15,25]. The inability of hKAT-I to catalyze efficient
transamination reactions in previous studies was probably
caused by Tris inhibition of the enzyme. In transaminases,
PLP is bound in a Schiff base linkage with the e-NH
2
group
of an active site lysine residue [34,35]. The mechanisms
regarding PLP-mediated transamination reactions (i.e. the
protonation of the Schiff base, removal of the Ca proton
from the amino acid substrate, electron relocalization and
rearrangement of the Schiff base double bond from the
pyridoxal aldehyde carbon to the Ca of the amino acid
substrate, etc.) have been discussed in numerous reports
[36,37]. Spectral shift of free PLP towards longer wave-
lengths in the presence of Tris at weak basic conditions
suggests that the amine can interact with the enzyme
associated PLP, which may lead to the formation of a S chiff
base through an initial nucleophilic addition to the carbonyl
group of P LP, followed by rapid proton transfer, leading to

water elimination a nd the formation of an imine. The
apparent spectral shift of PLP in Tris buffer at weak basic
pH, the absence of such a spectral shift of PLP in a
phosphate buffer of pH 7.0–9.0, in conjunction with the
same spectral shift of PLP in phosphate buffer upon
incorporation of glutamine and the high activity of hKAT-I
in phosphate buffer at both neutral and w eak basic
conditions (pHs 7.0–8.0), provides a reasonable basis for
suggesting that the extremely low activity o f KAT-I under
physiological pH is due to KAT-I inhibition by Tris amine.
Our data established that hKAT-I is quite capable of
catalyzing the kynurenine to KYNA pathway at physiolo-
gical conditions.
Data concerning the effect of other amino acids on
hKAT-I catalyzed KYNA production determined that
KYNA production is not seriously affected by most amino
acids, except for tryptophan, glutamine, phenylalanine and
cysteine, which decreased KYNA production by 30–70%,
which is consistent with a previous report, i.e. tryptophan,
glutamine and phenylalanine are inhibitors of the enzyme
[29]. Cysteine was reported to be a good substrate for
glutamine transaminase K [38], but its effect on hKAT-I
activity has not been tested. In mammals, endogenous
cysteine displays neuroexcitatory a ctions similar to those o f
glutamate [39,40]. Cysteine derivatives, homocysteine, cys-
teine sulfinate, homocysteine s ulfinate and cysteate, were
able to reduce t he production of KYNA in cortical slices in
rats, due presumably to their interaction with KATs, and
they were considered endogenous modulators of KYNA
formation in the brain [31,32]. However, because there are

potentially two K ATs in t he brain, which one is inhibited by
cysteine derivatives has not been fully understood, altho ugh
the possible inhibition of KAT-II has been p roposed [31].
The inhibition of hKAT-I by cysteine in vitro suggests that
the reduction of KYNA production in cortical slices in the
presence of cysteine or cysteine derivatives could be due to
their inhibition to KAT-I.
Most naturally occ urring a-keto acids can serve as the
amino group acceptor for hKAT-I. Pyruvate has been the
most commonly used a mino group acceptor for character-
izing hKAT-I [14,15,25,29]. Through kinetic analysis, it is
clear t hat pyruvate i s not an efficient amino group accepto r
due to its high K
m
(Fig. 5). The results dealing with
cosubstrate profiles in a previous s tudy [29] could have been
different if cosubstrate inhibition had been taken into
consideration. Our data confirmed that glyoxylate, aKMB,
p-hydroxy-phenylpyruvate, a-ketovalerate, a-ketocaproic
acid, a-ketoleucine, mercaptopyruvate, a-ketobutyrate,
pyenylpyruvate and oxaloacetate are more efficient amino
group acceptors than pyruvate for hKAT-I (Fig. 5
and Table 2), which is similar to the previously given
information about glutamine transaminase K , i.e. wide
a-k eto acid specificity, but high activity with a-KMB and
glyoxylate, strong substrate i nhibition with ph enylpyruvate
and poor affinity toward alanine and pyruvate [30,41].
In summary, by systematically studying the potential
substrates, amino acids and a-keto acids for hKAT-I, a new
substrate map for hKAT-I is obtained. This study con-

firmed that hKAT-I is a multifunctional enzyme. New pH
profiles of hKAT-I were described a nd the reasons why it is
different from the reported pH p rofile were discussed. Indo-
3-pyruvate and cysteine were found to be efficient inhibitors
for h KAT-I. Based on our data, i t i s reasonable to p ropose
that hKAT-I might be an important player in KYNA
synthesis under physiological conditions in the human
brain. However, much more research is needed to fully
understand its overall physiological role in vivo. Neverthe-
less, this study provides rather comprehensive bioch emical
characteristics of this important enzyme, which should be
highly useful towards elucidating the accurate r ole h KAT-I
plays in brain KYNA synth esis and towards controlling the
4812 Q. Han et al.(Eur. J. Biochem. 271) Ó FEBS 2004
levels of endogenous kynurenic acid in the human brain
through modulating hKAT-I activity.
Acknowledgements
We are grateful to Prof. Bruce M. Christensen, D epartment of
Animal Health and Biomedical Sciences, University of Wisconsin
(Madison, WI, USA) and Dr Menico Rizzi (De partment of G enetics,
University of Pavia, Italy) f or their crit ical reading of the manus cript.
This study w as support ed b y the National Institutes of H ealth G rant
AI 44399.
References
1. Cooper, A.J. ( 2004) T he role of glutamine transaminase K (GTK)
in sulfur and alpha-ke to acid metabolism in t he brain, and i n the
possible bioactivation of neurotoxicants. Neurochem. Int. 44,557–
577.
2. Mosca, M., Cozzi, L., Breton, J., Speciale, C ., Okuno, E.,
Schwarcz, R. & Benatti, L. (1994) Molecular cloning of rat

kynurenine aminotransferase: identity with glutamine transami-
nase K. FE BS Lett. 353, 21–24.
3. Malherbe, P., Alberati-Giani,D.,Kohler,C.&Cesura,A.M.
(1995) Identification of a mitochondrial form of kynurenine
aminotransferase/glutamine transaminase K f rom rat brain.
FEBS Lett. 367 , 141–144.
4. Abraham, D.G. & Cooper, A.J. (1996) Cloning and e xpression of
a rat kidney cytosolic glutamine t ransaminase K that has strong
sequence homo logy to kyn urenine pyruvate aminotransferase.
Arch. B iochem. Biophys. 335, 311–320.
5. Moroni, F. (1999) Tryptophan metabolism and brain function.
Focus on k ynurenine and other indole metabolites. Eur. J. Phar-
macol. 375, 87–100.
6. Schwarcz, R. & Pellicciari, R. (2002) Manipulation of brain
kynurenines: glial targets, n euronal e ffects, and c linical op portu-
nities. J. Pharmacol. Exp. Ther. 303, 1–10.
7. Stone, T.W. & Darlin gton, L.G. (2002) Endogenous kynurenines
as targets for dru g discovery and development. Nat. Rev. Drug
Discov. 1, 609–620.
8. Alkondon, M . & Pereira, E.F., Yu, P., Arruda, E.Z., Almeida,
L.E.,Guidetti,P.,Fawcett,W.P.,Sapko,M.T.,Randall,W.R.,
Schwarcz, R., Tag le, D.A. & Albuquerque, E.X. ( 2004) Targeted
deletion o f the kynurenine aminotransferase ii gene reveals a
critical role of endo genous kynurenic acid in the regulation of
synaptic transmission via alpha7 nicotinic receptors in the hip-
pocampus. J. Neurosci. 24, 463 5–4648.
9. Beal,M.F.,Matson,W.R.,Storey,E.,Milbury,P.,Ryan,E.A.,
Ogawa, T. & Bird, E.D. (1992) Kynurenic acid c oncentrations a re
reduced in Huntington’s disease cerebral cortex. J. Ne urol. Sci.
108, 80–87.

10. Beal, M.F., Matson, W.R., Swartz, K.J., Gamache, P.H. & Bird,
E.D. (1 990) Ky nurenine pathway m easurements in H untington’s
disease striatum: evid ence for reduced formation of kyn urenic
acid. J. Neurochem. 55, 1 327–1339.
11. Jauch, D., Urbanska, E.M., Guidetti, P., Bird, E.D., Vo nsattel,
J.P., Whetsell, W.O. Jr & Schwarcz, R. (1995) Dysfunction of
brain kynurenic acid metabolism in Huntington’s disease: focus on
kynurenine aminotra nsferases. J. Ne urol. Sci. 130, 39–47.
12. Schwarcz,R.,Rassoulpour,A.,Wu,H.Q.,Medoff,D.,Tam-
minga, C.A. & Roberts, R.C. (2001) Increased c ortical kynurenic
acid co ntent in schizophrenia. Biol. Psychiatry 50, 521 –530.
13. Erhardt, S., Blennow, K ., Nordin, C ., Skogh, E., Lindstrom, L .H.
& Engberg, G. (2001) K ynurenic acid levels are elevated in the
cerebrospinal fluid of p atient s with schizophrenia. Neurosci. L ett.
313, 96–98.
14. Okuno, E., Nakamura, M. & Schwarcz, R. (1991) Two kynur-
enine aminotransferases in human brain. Brain Res. 542, 3 07–312.
15. Guidetti, P., Okuno, E. & S chwarcz, R. (1997) Characterization of
rat brain kynurenine aminotransferases I and II. J. Neurosci. Res.
50, 4 57–465.
16. Gramsbergen,J.B.,Schmidt,W.,Turski,W.A.&Schwarcz,R.
(1992) Age-related changes in kynurenic acid production in rat
brain. Brain R es. 588, 1 –5.
17. Ito, S., Komatsu, K., Tsukamoto, K. & Sved, A.F. (2000)
Excitatory amino acids in the rostral ventro lateral medulla
support blood pressure in spontaneously hypertensive rats.
Hypertension 35 , 413–417.
18. Kiely, J.M. & Gordon, F.J. (1994) Role of rostral ventrolateral
medulla in c entrally me diated pre sso r respon ses. Am . J. Physiol.
267, H 1549–H1556.

19. Kapoor, V., Kapoor, R. & Ch almers, J. (1994) Kynu renic acid, an
endogenous glutamate antagonist, in SHR an d WKY rats:
possible role in central blood pressure regulation. Clin. Exp.
Pharmacol. Physiol. 21, 891–896.
20. Kapoor, V., Thuruthyil, S.J. & Human, B. (1998) Reduced
kynurenine aminotransferase-I activity in SHR rats m ay be du e to
lack of KAT-Ib activity. Ne uroreport 9, 1431–1434.
21. Mizutani, K., Sugimoto, K ., Okuda, T., Katsuya, T., Miyata, T.,
Tanabe, T., Higaki, J., Ogihara, T., Yamori, Y., Tsujita, Y., Tago,
N. & Iwai, N. (2002) Kynureninase i s a novel candidate gene for
hypertension in spontaneously hypertensive rats. Hype rtens. Res.
25, 1 35–140.
22. Kwok, J.B., Kapoor, R., Gotoda, T., Iwamoto, Y., Iizuka, Y.,
Yamada, N., Isaacs, K.E., Kushwaha, V.V., Church, W.B.,
Schofield, P.R. & Kapoor, V. (2002) A missense mutation in
kynurenine aminotran sferase-1 in spontaneously hypertensive
rats. J. Biol. Chem. 277 , 35779–35782.
23. .Han, Q. & Li, J. (2004) C ysteine a nd ke to ac ids m odulate m os-
quito kynurenine aminotransferase catalyzed kynurenic acid
production. FEBS Lett. 577, 381–385.
24. Hosono, A., Mizuguchi, H., Hayashi, H., Goto, M., Miyahara, I.,
Hirotsu, K. & Kagamiyama, H. (2003) Glutamine: phenylpyruvate
Aminotransferase from an Extremely Thermophilic Bacterium,
Thermus t hermophilus HB8. J. Biochem. (Tokyo) 134, 843–851.
25. Schmidt, W., Guidetti, P., Okuno, E. & Schwarcz, R. (1993)
Characterization of human b rain k ynurenine aminotransferases
using [
3
H]kynurenine as a substrate. Neuro scienc e 55, 177–184.
26. Okuno, E.F ., Ishikawa, T., Tsujimoto, M., Nakamura, M.,

Schwarcz, R. & Kido, R. (1990) Purification and characterization
of kynurenine-pyruvate aminotransferase from rat kidney and
brain. Brain R es. 534, 3 7–44.
27. Li, J. & Li, G. (1997) Transamination of 3-hy droxykynurenine to
produce x anthu renic ac id: a major branch pathway o f tryptophan
metabolism in the mosquito, Aedes aegypti,duringlarvaldevel-
opment. Insect Biochem. Mol. Bi ol. 27, 8 59–867.
28. Han, Q. & Li, J. (2002) Comparative characterization of Aedes
3-hydroxykynurenine transaminase/alanine glyoxylate transami-
nase and Drosophila se rine pyruvate aminotransferase. FEB S Lett.
527, 1 99–204.
29. Baran, H., O kuno, E ., Kido, R. & Sc hwarcz, R . ( 1994) Purifica-
tion and c haracterization of kyn urenine aminotransferase I from
human brain. J. Neurochem. 62, 730–738.
30. Cooper, A.J. & Meister, A. (1981) Com parative studies of gluta-
mine transaminases from rat tissues. Comp. Biochem. P hys iol. B
69B, 137 –145.
31. Kocki, T., Luchowski, P., Luchowska, E., Wielosz, M., Turski,
W.A. & U rbanska, E.M. (2003)
L
-cysteine sulphinate, endogenous
sulphur-containing amino a cid, inh ibits rat brain kynu renic a cid
production via selective interference with kynurenine amino-
transferase II. Neurosci. Lett. 346, 9 7–100.
Ó FEBS 2004 Characterization of human kynurenine aminotransferase I (Eur. J. Biochem. 271) 4813
32. Wejksza, K., Rzeski, W ., Parada-Turska, J., Zdzisinska, B.,
Rejdak, R., Kocki, T., Okuno, E., Kandefer-Szerszen, M.,
Zrenner, E. & Turski, W.A. (2004) Kynurenic acid p roductio n
in cultured bovine aortic endothelial cells. Homocysteine is
a pot ent i nhibitor. N aunyn-Schmiedebergs Arch. Pharmacol. 36 9,

300–304.
33. Han, Q., Fang, J. & Li, J. (2001) Kynurenine aminotransferase
and glutamine transaminase K of Escherichia coli: identity with
aspartate a minotransferase. Bioc hem. J. 360 , 617–623.
34. Garrett, R. & Grisham, C.M. (1999) Biochemistry, 2nd edn.
Saunders College Publications, Fort W orth.
35. Rossi, F., Han, Q., Li, J., Li, J. & Rizzi, M. (2004) Crystal
structure o f h uman kynurenine aminotransferase I. J. Biol. Chem.
279, D OI: 10.1074/jbc.M409291200
36. John, R.A. (1995) Pyridoxal phosphate-dependent enzymes.
Biochim. Bi ophys. Acta. 1248, 81–96.
37. Jansonius, J.N . (1998) Structur e, evolution and action of vitamin
B6-dependent enzymes. Curr. Opin. Stru ct. Biol. 8, 759– 769.
38. Cooper, A.J., Haber, M.T. & Meister, A. (1982 ) On the chemistry
and b iochemistry of 3-mercaptop yruvic acid, t he alpha-keto a cid
analog of cysteine. J. Bi ol. Chem. 257, 816–826.
39.Cuenod,M.,Grandes,P.,Zangerle,L.,Streit,P.&Do,K.Q.
(1993) Sulphur-containing excitatory amino acids in intercellular
communic atio n. Biochem. Soc . Trans. 21 , 72–77.
40. Turski, W.A., Herrling, P.L. & Do, K.Q. (1987) Effects of
L
-cysteine-sulphinate and
L
-aspartate, mixed excitatory amino
acid agonists, on the membrane potential of cat caudate neurons.
Brain Res. 414 , 330–338.
41. Cooper, A.J. & Meister, A. (1974) Isolation and properties of a
new glutamine transaminase from rat kidney. J. Biol. Chem. 249,
2554–2561.
4814 Q. Han et al.(Eur. J. Biochem. 271) Ó FEBS 2004