Journal ofNeurochemislry Raven Press, Ltd., New York 0 1990 International Society for Neurochemistry
Presence of 3-Hydroxyanthranilic Acid in Rat Tissues and Evidence for Its Production from Anthranilic Acid in the Brain Halina Baran and Robert Schwarcz Maryland Psychiatric Research Center, University of Maryland School of Medicine, Baltimore, Maryland, U S . A
Abstract: As assessed by HPLC with electrochemical detection, 3-hydroxyanthranilic acid (3-HANA) was found to be present in the rat brain and peripheral organs. The highest concentrations were measured in the kidney (86 fmol/mg of tissue) and spleen (56 fmol/mg of tissue), whereas the adrenal gland, liver, heart, and several forebrain areas (hippocampus, striatum, parietal cortex, thalamus, amygdala/pyriform cortex, and frontal cortex) contained less 3-HANA (between 15 and 22 fmol/mg of tissue). Slightly lower concentrations of 3-HANA were found in the brainstem and the cerebellum. The metabolic disposition of 3-HANA was examined in tissue slices which were incubated in Krebs-Ringer buffer at 37°C in vitro. Incubation for up to 2 h did not affect 3-HANA concentration in brain tissue. However, inhibition of 3HANA degradation by the specific 3-hydroxyanthranilic acid oxygenax blocker 4-chloro-3-hydroxyanthranilic acid (4-C13-HANA; 10 r M ) resulted in a rapid (within 2.5 min) doubling of 3-HANA levels in slices from cerebral cortex. No further increases were observed after incubations of up to 120 min. Exposure of cortical dices to 3-HANA’s putative bioprecursors, 3-h ydroxykynurenine (3-HK) and anthranilic
acid (ANA), in the absence of4-Cl-3-HANA resulted in rapid, transient increases in 3-HANA production. Maximal 3HANA synthesis from ANA exceeded the maximal effect of 3-HK by approximately 1 1-fold. In the presence of 4-CI-3HANA, 1 mM 3-HK and 1 mM ANA produced 9.0 ? 0.3 and 89.0 ? 9.3 (5 min) or 51.6 k 7.9 and 187.5 k 11.2 (120 min) fmol of newly synthesized 3-HANA/mg of brain tissue, respectively. In the brain, but not in the spleen, ANA proved to be a superior 3-HANA precursor at lower concentrations as well. Time and dose relationships for the de novo production of 3-HANA from ANA in brain slices were established in the presence of 10 pM 4-Cl-3-HANA. Biosynthesis of 3-HANA from ANA in the brain may be critically involved in the function or dysfunction of 3-HANA’s principal catabolic product, the endogenous excitotoxin quinolinic acid. Key Words: Anthranilic acid-Excitotoxins-3-Hydroxyanthranilic acid-Kynurenines-Neurodegenerative disorders. Baran H. and Schwarcz R. Presence of 3-hydroxyanthranilic acid in rat tissues and evidence for its production from anthranilic acid in the brain. J. Neurochem. 55, 738744 ( 1990).
Quinolinic acid (QUIN), a brain metabolite with convulsive and neurodegenerative properties, has recently attracted considerableattention as a hypothetical pathogen in neurological diseases (Schwarcz et al., 1987). In addition to its possible role in the etiology of temporal lobe epilepsy (Lapin, 1981; Schwarcz et al., 1984a), Huntington’s disease (Schwarcz et al., 1984b; Beal et al., 1986), and other neurodegenerative conditions, QUIN may also serve physiological functions in the brain as a selective agonist at the N-methyl-Daspartate NMDA receptor (Stone and Connick, 1985). In spite of the well-documented effects of QUIN at its receptor, however, relatively little is known about
its metabolism in the brain. The enzyme 3-hydroxyanthranilic acid oxygenase (3-HAO), which converts 3hydroxyanthranilic acid (3-HANA) to QUIN (Nishizuka and Hayaishi, 1963), has been identified and characterized in both rodent and human brain and found to be localized in astrocytes (Foster et al., 1986; Okuno et al., 1987; Schwarcz et al., 1988), but the origin of 3-HANA remains obscure. In the periphery, 3HANA derives from tryptophan via a series of enzymatic reactions along the so-called kynurenine pathway (Schlossberger et al., 1984). However, the two enzymes preceding 3-HA0 in this established biosynthetic cascade, i.e., kynurenine hydroxylase and kynureninase,
Received September 15, 1989; revised manuscript received January 15, 1990; accepted January 22, 1990.
Abbreviations used: ANA, anthranilic acid; 4-C1-3-HANA, 4chloro-3-hydroxyanthranilic acid; 3-HANA, 3-hydroxyanthranilic acid; 3-HAO, 3-hydroxyanthranilic acid oxygenase; 3-HK, 3-hydroxykynurenine; PCA, perchloric acid; QUIN, quinolinic acid.
Address correspondence and reprint requests to Dr. R. Schwarcz at Maryland Psychiatric Research Center, P.O. Box 21247, Baltimore, MD 2 1228, USA.
738
3-HYDROXYANTHRANILIC ACID IN RAT BRAIN
show very low activity in the rat brain (Battie and Verity, 1981; Kawai et al., 1988), and attempts to incorporate kynurenine into QUIN in vivo (Speciale et al., 19896) and in vitro (unpublished observations) have failed to reveal activity of the brain kynurenine pathway in rats. In our quest to examine the biology of the brain’s kynurenine system, it therefore appeared prudent to explore alternative routes of 3-HANA biosynthesis, in particular the possibility of a direct 3-hydroxylation of anthranilic acid (ANA), which has been reported to occur in the mammalian liver (Ueda et al., 1978), in plants (Nair and Vaidyanathan, 1965), and in microorganisms (Schott et al., 1982). I n the present article, we report first results of our search for a new avenue of 3-HANA production in the brain. Thus, using a novel, sensitive analytical method for 3-HANA determination, we have examined the presence of 3-HANA in rat tissues and followed its de novo synthesis from 3-hydroxykynurenine (3-HK) and ANA, respectively, in tissue slices in vitro. A preliminary account of this work has been published in abstract form (Baran and Schwarcz, 1989).
MATERIALS AND METHODS Materials ANA, 3-HANA, and 3-HK were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). 4-Chloro-3-hydroxyanthranilic acid (4-Cl-3-HANA) was prepared as described by Todd et al. (1989) and was kindly provided by Dr. Barry K. Carpenter (Cornell University, Ithaca, NY, U.S.A.). Rat liver 3-HA0 was purified in our laboratory as described by Okuno et al. (1987). All other materials and reagents were of the highest quality commercially available.
Animals Male Sprague-Dawley rats (240-280 g), kept under standard laboratory conditions on a 12-h light/dark cycle with free access to food and water, were used in all experiments.
Measurement of 3-HANA Tissue extracts or standard 3-HANA (see below) were applied to a Dowex formate anion-exchange resin (Bio-Rad AG l-X8,200-400 mesh; column size 0.5 cm X 0.7 cm) prewashed with 2 ml ofdistilled water. Subsequently, the column was washed with 2 ml of distilled water and 0.4 ml of 1 M formate. 3-HANA was eluted with 0.4 ml of 2 M and 0.4 ml of 3 M formate. The combined formate fractions were mixed thoroughly with 0.8 ml of ethyl acetate to extract 3-HANA into the organic phase. After centrifugation at 3,500 g for 5 min, 650 pl of the 3-HANA-containing phase were lyophilized. At this stage, samples could be frozen and kept at -20°C until analysis. The samples were then resuspended in 60 p1 of 0.05 Mperchloric acid (PCA), and an appropriate aliquot (routinely 20 p l ) was subjected to HPLC using a system comprising an 8-cm CISreversed-phase HR-80 column (3 pm spherical octadecylsilane particles; ESA, Bedford, MA, U.S.A.) and a coulometric detector (ESA model 5 IOOA). The mobile phase consisted of 0.1 M sodium phosphate, 0.75 Woctanesulfonic acid, and 7% methanol, pH 3.0. 3-HANA was eluted at a flow rate of 0.5 ml/min and detected electrochemically at +0.40 V with the aid of a C-RSA integrator (Shimadzu, Columbia, MD, U.S.A.).
739
Enzymatic identification of 3-HANA In pilot experiments, the identity of 3-HANA was ascertained by its degradation using a purified preparation of the specific 3-HANA-catabolizing enzyme 3-HA0 (Okuno et al., 1987). For this purpose, Dowex formate eluates containing 3-HANA standards or putative 3-HANA from rat tissue were extracted with ethyl acetate, and the organic phase was dried down. After resuspension in 60 pl of 30 mM 2-(N-morpho1ino)ethanesulfonic acid buffer, pH 6.5, several identical samples were combined. The resulting sample was then split into equal parts, which were mixed with 2 p1 of active or heat-inactivated (control) purified 3-HA0 (sp. act. 7.5 pmol/ min/mg of protein). The samples were then incubated for 60 min at 37°C in the presence of 300 pMFeS04 (Foster et al., 1986), and 10 pl of 1 M PCA were added to stop the reaction. Subsequently, the mixture was centrifuged (10 min, 28,000 g), and an aliquot of the supernatant subjected to HPLC analysis as described above.
Tissue preparation Animals were killed by decapitation, and their organs rapidly removed and dissected out at 4°C. For the determination of 3-HANA concentration, tissues were obtained and frozen on dry ice within 2 rnin after killing. The samples were kept frozen for up to 2 days at -20°C. In separate experiments, slices of cerebral cortex and spleen (base 1 X I mm; approximate weight 1.5 mg/slice) were obtained using a McIlwain chopper (Mickle Laboratory Engineering Co., Gomshall, U.K.). Tissue was cut in one direction, subsequently chopped perpendicular to the original orientation, and immediately used for the experiment.
3-HANA determination in tissue homogenates Frozen tissues were sonicated (1: 10, wt/vol) in 0.1 MPCA, and the resulting homogenate was centrifuged for 15 min at 28,000 g. In every experiment, selected duplicate samples received 1-4 pmol of 3-HANA as an internal standard prior to sonication. One milliliter of the resulting supernatant was applied to the Dowex-formate column, and 3-HANA was eluted, processed, and measured as described above.
Experiment with tissue slices Slices were separated on ice, transferred to culture wells (10 slices/well) containing 1 ml of oxygenated Krebs-Ringer buffer (NaCI, 118.5 mM; KCI, 4.75 mM; MgS04, 1.18 mM, NaH2P04, 12.9 mM, Na2HP04,3 mM; CaCI2, 1.77 mM, Dglucose, 5 mM, pH 7.4), and preincubated for 10 min at 37°C. The tissue was then incubated further in a shaking water bath at 37°C for varying lengths of time (2.5-120 min) in the presence or absence of 3-HK and ANA, two putative bioprecursors of 3-HANA. In some experiments, the selective 3-HA0 blocker 4-CI-3-HANA (Parli et al., 1980; Heyes et al., 1988; Todd et al., 1989) was added at the beginning of the preincubation period, routinely at a final concentration of 10 pM. Following incubation, the culture wells were immediately placed on ice and the slices rapidly separated from the incubation medium. Slices from two wells were pooled and their total weight determined. The slices were then homogenized by sonication ( 1 :10, wt/vol) as described for brain homogenates and the concentration of 3-HANA determined as detailed above.
RESULTS 3-HANA assay and identification When applied directly to HPLC, 3-HANA was eluted with a retention time of between 9.5 and 11 J. Neurochem., Vol. S5, No. 3, 1990
H. BARAN AND R. SCHWARCZ
740
min. Linearity of the assay was ascertained up to 100 pmol of 3-HANA. The sensitivity limit was found to be 25 fmol of 3-HANA (Fig. 1). Processing of standard or tissue 3-HANA through the entire ion-exchange/ extraction/HPLC procedure described in Materials and Methods yielded a recovery of 60 f 2% (mean k SEM of all experiments). Incubation of standard 3-HANA, which had been subjected to the ion exchange/ethyl acetate extraction procedure, with 3-HA0 resulted in the abolishment of the 3-HANA response, thus confirming the identity of 3-HANA. No degradation of 3-HANA was observed when heat-inactivated 3-HA0 was used in the experiment (Fig. 2).
3-HANA content of rat tissues 3-HANA could be measured in all organs examined. As shown in Table 1, the highest tissue concentrations were found in the kidney (86.0 f 10.9 fmol/mg of tissue) and spleen (56.1 f 8.2 fmol/mg of tissue), and approximately equal quantities of 3-HANA (between 10 and 20 fmol/mg of tissue) were detected in the adrenal gland, liver, heart, and forebrain (brain minus cerebellum; Fig. 3b). In separate experiments, conducted with renal, splenic, hepatic, and cerebral tissue, 3-HANA was positively identified using 3-HA0 digestion as described in Fig. 2 (data not shown). Assessment of the regional distribution of 3-HANA in the rat brain revealed only small differencesbetween forebrain areas. However, relatively low concentrations of 3-HANA were found in brainstem and cerebellum (Table 2). Production of 3-HANA in slices Incubation of cerebral cortical slices in Krebs-Ringer buffer for up to 2 h yielded tissue 3-HANA concentrations which were virtually indistinguishable from those measured in brain homogenates (Table 3), indicating the preservation of steady-state levels over an extended period of time ex vivo. Inclusion of the selective 3H A 0 inhibitor 4-C1-3-HANA (10 p M ) in the incubation medium resulted in a rapid (within 2.5 min) doubling of 3-HANA concentrations (Table 3). No
c
r
0
/
6
p' 2 0
1,f 0
/ ,
loo
200 300 4 0 0 SHANA ( fmol)
I
SO0
FIG. 1. 3-HANA standard curve obtained by HPLC (peak height refers to original HPLC records).
J. Neurochem.. Vol. 55, No. 3. 1990
a
1
J JU-
",
-
FIG. 2. HPLC profilesof 700 fmol of standard 3-HANA (arrowhead) subjected to the ion-exchange/extraction/HPLC procedure d e scribed in Materials and Methods. Samples were incubated with heat-inactivated3-HA0 (a) or active 3-HA0 (b)after ethyl acetate extraction as detailed in the text.
further increases were noted upon prolonged incubation in the presence of the drug for up to 120 min. Lower (3 p M ) and higher (30 p M ) concentrations of 4-C1-3-HANA had virtually identical effects as 10 pM 4-C1-3-HANA (data not shown). Incubation of cortical slices in the presence of 3HANA's putative bioprecursors 3-HK and ANA ( 1 m M each), in the absence of 4-C1-3-HANA, resulted in rapid and transient increases in 3-HANA concentrations. Peak effects for both substances were noted within 15 min of incubation. In quantitative terms, maximal 3-HANA production from ANA exceeded maximal 3-HANA synthesis from 3-HK by approximately 1 1-fold (Fig. 4). The synthesis of 3-HANA from ANA was verified by 3-HAO-induced degradation of the metabolic product (cf. Fig. 3). To this end, cerebral cortical slices were incubated for 5 min with 1 mMANA, the samples TABLE 1. Distribution of 3-HANA in rat tissues Organ
fmol/mg of tissue
Kidney Spleen Adrenal gland Liver Heart Forebrain
86.0 k 10.9 56.1 f 8.2 19.7 t 5.1 18.8 f 2.6 17.5 f 4.2 13.2 ? 1.7
Data are the means i SEM of tissues obtained from five animals.
741
3-HYDROXYANTHRANILIC ACID IN R A T BRAIN N OD t 0)
TABLE 3. 3-HANA in cortical slices in the absence or presence of 4-Cl-3-HANA Incubation time (min)
4-CI-3-HANA (10 p M )
2.5
5
120
12.7 t 0.1 25.8 t 1.5'
11.9 t 0.6 28.5 f 1.9"
13.6 t 1.9 31.5 k 2.7"
Tissue slices were incubated for various time periods as detailed in Materials and Methods. Data are expressed in fmol of 3-HANA/ mg of tissue and represent the means t SEM of three samples, each the average of duplicate incubations. " p < 0.05, as compared to slices incubated in the absence of 4CI-3-HANA (Student's t test).
FIG. 3. HPLC profiles of 500 fmol of standard 3-HANA subjected directly to HPLC (a) and a characteristic chromatogram of a rat forebrain extract (b).
processed through the ion-exchange/extraction procedure, and 3-HANA identification carried out as described in Materials and Methods. As shown in Fig. 5, the newly synthesized compound was thus identified as authentic 3-HANA. The next set of experiments was designed to compare the effects of 3-HK and ANA (100 pM and 1 m M concentrations of each compound) on 3-HANA production in the presence of 10 pM 4-Cl-3-HANA. Bioprecursor effects were studied at two time points (5 min and 120 min), and the analysis was performed in parallel in slices from cerebral cortex and spleen. 3H A 0 inhibition prevented the breakdown of newly synthesized 3-HANA regardless of the nature of the bioprecursor. Thus, in the presence of 4-C1-3-HANA,
the 3-HANA content of slices kept increasing when either 3-HK or ANA was used (Table 4). In the brain, ANA was superior to 3-HK as a precursor of 3-HANA under all experimental conditions tested. The most dramatic differences between the two compounds were apparent at a concentration of 100 pMand an incubation time of 5 min. In absolute terms, 3-HANA synthesis from ANA in the brain was comparable to that in the spleen. In contrast, 3-HK was far more active in the spleen, where substantial increases in 3-HANA production were measured when 3-HK concentrations were raised from 100 pM to 1 m M or when the incubation time was extended from 5 to 120 min. Thus, in the spleen 3-HK clearly served as the preferential source of 3-HANA under all conditions tested (Table 4). Time and dose relationships for the de novo production of 3-HANA from ANA in the presence of 10 pM 4-Cl-3-HANA were examined in greater detail in slices from cerebral cortex. Using 100 pM ANA, 3160-
80
TABLE 2. Regional distribution 0f3-HANZ4
a a
in the rat bruin Brain area
fmol/mg of tissue
Hippocampus Striatum Parietal cortex Thalamus Amygdala/pyriform cortex Frontal cortex Brainstem Cerebellum
2 1.9 & 4.4 (5) 21.0 f 2.3 (5) 20.8 f 2.5 (7) 19.9 k 2.7 (6) 19.8 t 3.8 (7) 15.5 t 1.9 (7) 12.2 t 1.8 (6) 8.7 f 1.1 (7)
Data are the means parentheses.
~tSEM
e
of the number of animals given in
40
Minutes
FIG. 4. Time dependency of d e novo 3-HANA production from cortical slices incubated with 1 mM 3-HK (A)or 1 mM ANA (0)at 37°C in the absence of 3-HA0 inhibition. Data are the means k SEM of four to six samples, each the average of duplicate incubations.
J. Neurochem.. Vol. 55. No. 3, 1990
H . BARAN AND R. SCHWARCZ
742
1
n
i
b
0
1
5
10
60
Minutes FIG. 6. Synthesis of 3-HANA in cortical slices exposed to 100 p M ANA and 10 phf 4-Cl-3-HANA for various time periods at 37°C. Data represent newly produced 3-HANA in excess of the effect of 4-CI-3-HANA alone and are the means f SEM of three samples, each the average of duplicate incubations.
FIG.5. HPLC profiles of the 3-HANA-like compound (arrowhead) isolated from cortical slices incubated for 5 min with 1 mM ANA. Samples were incubated with heat-inactivated3-HA0 (a)or active 3-HA0 (b) after ethyl acetate extraction as described in Materials and Methods.
HANA synthesis proceeded rapidly in a linear fashion for 2.5 min, but slowed thereafter, reaching a production rate of approximately 2 fmol/lO min/mg of tissue after 5-60 min of incubation (Fig. 6). Variation of ANA concentration between 10 and 1,000 pM, using an assessment of 3-HANA production after a 5-min incubation period, yielded a curve which approached saturation above 300 pM ANA (Fig. 7).
DISCUSSION 3-HANA has long been known as an integral link in the kynurenine pathway, the primary route of tryptophan degradation in mammalian cells. Because of its possible role in the pathogenesis of bladder cancer
(Boyland and Williams, 1956; Benassi et al., 1963; Teulings et al., 1973), which has been speculatively related to its ability to interfere with mitochondria1 oxidative phosphorylation (Quagliariello et al., 1964), 3-HANA received substantial attention from researchers in the 1950sand 1960s(Gorrod et al., 1967). Studies focussed on the disposition of the substance in urine where 3-HANA occurs in both conjugated and free forms (Brown and Price, 1956; Price et al., 1965), and on the enzymatic mechanisms which are responsible for the hydrolysis of conjugated 3-HANA (Watanabe and Minegishi, 1972; Watanabe et al., 19723). Until recently, only sporadic attempts had been made to examine the presence and function of the kynurenine pathway in the brain. Because of the possible involvement of two kynurenines, QUIN and kynurenic acid, in the pathogenesis of neurodegenerative diseases (Schwarcz et al., 1987), the cerebral disposition of these two compounds has become the subject of increasingly extensive scrutiny. However, prior to the present work, the presence or biosynthesis of 3-HANA in the brain had not been documented (cf. GB1 and Sherman, 1980), although exogenously applied 3-
TABLE 4. De novo production of 3-HANA in cerebral cortical and splenic slices ~
Brain
Spleen
Bioprecursor
Concentration
5 min
120 min
5 min
120 min
3-HK 3-HK
100 pcM
1.0 t 0.1 9.0 0.3
2.1 f 0.2 51.6 ? 7.9
51.4 k 2.1 489.3 f 72.9
2,141.7 k 276.3 4,618.7 k 380.8
ANA ANA
100 pM 1 mM
82.7 f 0.8 187.5 k 11.2
23.9 f 3.2 48.0 f 4.6
44.8 k 1.7 121.7 k 7.2
ImM
* 53.5 f 6.2 89.0 * 9.3
Slices were exposed to 3-HK or ANA at 37°C for 5 or 120 rnin in the presence of 10 pM 4-CI-3-HANA as described in Materials and Methods. Data represent newly produced 3-HANA in excess of the effects of 4-C1-3-HANA alone. They are expressed as fmol of 3-HANA/mg of tissue and are the means k SEM of three samples, each the average of duplicate incubations.
J. Norrochem., Vol. SS, No. 3. 1990
3-HYDROXYANTHRANILIC ACID IN RAT BRAIN
_., I00
.
.
.
I
.
x
I
1
500
I
.
1000
Anthrontlic octd [ pM I
FIG. 7. 3-HANA production in cortical slices incubated for 5 min at 37°C with different concentrations of ANA in the presence of 10 N4-Cl-3-HANA. Data represent newly produced 3-HANA in excess of the effect of 4-CI-3-HANA alone and are the means rt SEM of three samples, each the average of duplicate incubations.
HANA was shown to serve as an effectivebioprecursor of brain QUIN both in vitro and in vivo (Speciale et al., 1988, 1989b). Using a novel tissue extraction/HPLC method with electrochemical 3-HANA detection, and taking advantage of the availability of a highly purified 3-HA0 preparation for its unequivocal identification,we found very small concentrations of 3-HANA in the rat brain. The steady-state brain content of 3-HANA was substantially lower than that in serum (unpublished obYeh and servations)and urine (Watanabe et al., 1972~; Brown, 1979, and also severalfold lower than renal or splenic 3-HANA. Notably, the cerebral concentration of 3-HANA, distributed rather uniformly throughout the rat brain, was found to be approximately half that of 3-HK (Heyes and Quearry, 1988) and one order of magnitude lower than that of its immediate catabolic product QUIN (Wolfensberger et al., 1983; Moroni et al., 1984). Because 3-HANA does not enter the brain from the periphery under physiological conditions (Q. Smith, personal communication), subsequent studies were designed to assess its metabolism in brain slices. Incubation of the tissue in a physiological environment for up to 120 min did not affect 3-HANA content. Moreover, the data obtained from slices were virtually identical to the 3-HANA levels measured in brain homogenates, suggesting a delicate balance of 3-HANA’s anabolic and catabolic enzymes in the intact cell. Indeed, selective blockade of 3-HAO, presumably the major degradative enzyme for 3-HANA in mammals, by 4-C1-3-HANA resulted in a rapid increase in 3HANA tissue concentration. Notably, the 3-HANA surge following 3-HA0 inhibition was limited in scope, most probably because of 3-HANA efflux or the exhaustion of endogenous bioprecursor(s).
743
The next series of experiments was designed to investigate 3-HANA production from exogenously supplied putative bioprecursors. In accordance with the low activity of kynureninase in rat brain (Kawai et al., 1988), 3-HK was found to be only modestly active in effecting the de novo synthesis of 3-HANA in brain slices, while giving rise to much larger quantities of 3HANA in spleen slices under identical experimental conditions. In the absence of 4-C1-3-HANA, 3-HANA levels decreased after reaching an early maximum, suggesting that brain 3-HK, which can enter the cell through the transport site for large neutral amino acids (Speciale et al., 1989a), is converted rather rapidly to 3-HANA, which in turn undergoes fast degradation. In contrast to the spleen, where 3-HK was confirmed as 3-HANA’s preferred bioprecursor, ANA was found to be surprisingly active and, in fact, superior to 3-HK in producing 3-HANA under our experimental conditions. In spite of the fact that ANA penetrates the cell membrane quite poorly (unpublished observations), sufficient ANA must have entered the cell for further sequestration by enzymatic hydroxylation (Schwarcz et al., 1989). De novo synthesized 3-HANA peaked very rapidly after ANA administration; notably, the ANA effect was similar in speed to the 3-HANA increase observed following 3-HA0 inhibition with 4Cl-3-HANA (cf. Table 3 and Fig. 4). Speed and degree of the effect, therefore, indicate that ANA can serve as an efficient bioprecursor of 3-HANA in the brain. The present study did not reveal the existence of separate intracellular pools in the brain which could conceivably be in place to produce 3-HANA from 3HK and ANA. Thus, 3-HA0 inhibition by 4-Cl-3HANA caused rather modest increases in 3-HANA concentrations after 5- 120 min of incubation, regardless of the nature of the bioprecursor. Feedback inhibition of 3-HANA on both of its biosynthetic enzymes or 3-HANA release may afford self-regulation and a substantial slowing of intracellular 3-HANA accumulation after a sudden surge in 3-HANA production. In conclusion, the data presented here suggest that inlhe rat brain 3-hydroxylatinn of ANA may provide a major source of 3-HANA (and hence of QUIN). The presence of ANA in cerebral tissue has not been ascertained as yet, but its synthesis from kynurenine has been reported to occur in brain homogenates in vitro (Kawai et al., 1988). Studies currently in progress in our laboratory, therefore, are designed to examine several elementary questions regarding ANA neurobiology. In turn, the new information can be expected to reveal valuable clues concerning the function and possible dysfunction of QUIN in the brain.
Acknowledgment: We gratefully acknowledge the help of Mr. Peter Krivanek during the early phases of the project. We thank Joyce Burgess for excellent secretarial assistance. This work was supported by USPHS grants NS 16102 and NS 20509 and by a Max Kade Fellowship to H.B. J. Neurochem.. Vol. 55, No. 3, 1990
H . BARAN AND R . SCHWARCZ
744 REFERENCES
Baran H. and Schwarcz R. (1989) Preferential production of 3-hydroxyanthranilic acid from anthranilic acid in rat brain slices. SOC.Neurosci. Abstr. 15, 328.5. Battie C. and Verity M. A. (1981) Presence of kynurenine hydroxylase in developing rat brain. J. Neurochem. 36, 1308-1310. Beal M. F., Kowall N. W., Ellison D. W., Mazurek M. F., Swartz K. T., and Martin J. B. (1986) Replication of the neurochemical characteristicsof Huntington’s disease by quinolinic acid. Nature 321, 168-171. Benassi C. A,, Perissinotto B., and Allegri G. (1963) The metabolism of tryptophan in patients with bladder cancer and other urological diseases. Clin. Chim. Acta 8, 822-831. Boyland E. and Williams D. C. (1956) Metabolism of tryptophan. 2. The metabolism of tryptophan in patients suffering from cancer of the bladder. Biochem. J. 64,578-582. Brown R, R. and Price J. M. (1956) Quantitative studies on metab olites of tryptophan in the urine of the dog, cat, and man. J. Biol. Chem. 219,985-997. Foster A. C., White R. J., and Schwarcz R. (1986) Synthesis of quinolinic acid by 3-hydroxyanthranilicacid oxygenase in rat brain tissue in vitro. J. Neurochem. 41,23-30. Gil E. M. and Sherman A. D. (1980) L-Kynurenine: its synthesis and possible regulatory function in brain. Neurochem. Res. 5, 223-239. Gorrod J. W., Alifano A., Papa S., and Quagliariello E. (1967) Effect of aromatic amines on isolated rat-liver mitochondria. Cancer Res. 21, 668-673. Heyes M. P. and Queany B. J. (1988) Quantification of 3-hydroxykynurenine in brain by high-performanceliquid chromatography and electrochemical detection. J. Chromatogr.428, 340-344. Heyes M. P., Hullo B., and Markey S. P. (1988) 4-Chloro-3-hydroxyanthranilate inhibits brain 3-hydroxyanthranilateoxidase. Neurochem. Int. 13,405-408. Kawai J., Okuno E., and Kido R. (1988) Organ distribution of rat kynureninase and changes of its activity during development. Enzyme 39, 181- 189. Lapin I. L. (1981) Kynurenines and seizures. Epilepsia 22,257-265. Moroni I., Lombardi G., Carl2 V., and Moneti G. (1984) The excitotoxin quinolinic acid is present and unevenly distributed in the rat brain. Brain Res. 295, 352-355. Nair P. M. and Vaidyanathan C. S. (1965) Anthranilic acid hydroxylase from Tecoma stuns. Biochim. Biophys. Acta 110, 521531. Nishizuka Y. and Hayaishi 0. (1963) Studies on the biosynthesis of nicotinamide adenine dinucleotide. I. Enzymatic synthesis of niacin ribonucleotidesfrom 3-hydroxyanthranilicacid in mammalian tissues. J. Biol. Chem. 238, 3369-3377. Okuno E., Kohler C., and Schwarcz R. (1987) Rat 3-hydroxyanthranilic acid oxygenase: purification from the liver and immunocytochemical localization in the brain. J. Neurochem. 49, 771-780. Parli C. T., Krieter P., and Schmidt B. (1980) Metabolism of 6chlorotryptophanto 4-chloro-3-hydroxyanthranilicacid a potent inhibitor of 3-hydroxyanthranilicacid oxidase. Arch. Biochem. Biophys. 203, 161-166. Price J. M., Brown R. R., and Yess N. (1965) Testing the functional capacity of the tryptophan niacin pathway in man by analysis of urinary metabolites. Adv. Metab. Disord. 2, 159-225. Quagliariello E., Papa S., Saccone C., and Alifano A. (1964) Effect of 3-hydroxyanthranilic acid on the mitochondrial respiratory system. Biochem. J. 91, 137-146. SchlossbergerH. G., Kochen W., Linzen B., and Steinhart H., eds ( 1984) Progress in Tryptophan and Serotonin Research. de Gruyter, Berlin.
J. Neurochem.. Vol. 55, No. 3, 1990
Schott H. H., Langenbach-Schmidt D., and Zinke K. (1982) Determination of anthranilate, 3-hydroxyanthranilateand 2,3dihydroxybenzoate in the enzyme assay of anthranilate hydroxylase by thin-layer and high-performance liquid chromatography. J. Chromatogr. 245,85-9 1. Schwarcz R., Brush G. S., Foster A. C., and French E. D. (1984a) Seizure activity and lesions following intrahippocampal injection of quinolinic acid. Exp. Neurol. 84, 1-17. Schwarcz R., Foster A. C., French E. D., Whetsell W. 0. Jr., and Kohler C. (19846) Excitotoxic models for neurodegenerative disorders. Life Sci. 35, 19-32. Schwarcz R., Okuno E., Speciale C., Kohler C., and Whetsell W. 0. Jr. (1987) Neuronal degeneration in animals and man: the quinolinic acid connection, in Neurotoxins and Their Pharmacological Implications (Jenner P., ed), pp. 19-32. Raven Press, New York. Schwarcz R., Okuno E., White R. J., Bird E. D., and Whetsell W. 0. Jr. (1988) 3-Hydroxyanthranilate oxygenase activity is increased in the brains of Huntington disease victims. Proc. Nafl. Acad. Sci. USA 85,4079408 1. Schwarcz R., Baran H., and Okuno E. (1989) Anthranilate hydroxylase in rat brain. Soc. Neurosci. Abstr. 15, 328.6. Speciale C., Ungerstedt U., and Schwarcz R. (1988) Quinolinic acid formation in rat brain in vivo and in vitro: effects of striatal lesions. SOC.Neurosci. Abstr. 14, 479.9. Speciale C., Hares K., Schwarcz R., and Brookes N. (1989a) Highaffinity uptake of L-kynurenine by a Na*-independenttransport of neutral amino acids in astrocytes. J. Neurosci. 9,2066-2072. Speciale C., Ungerstedt U., and Schwarcz R. (1989b) Production of extracellular quinolinic acid in the striatum studied by microdialysis in unanesthetized rats. Neurosci. Left. 104, 345-350. Stone T. W. and Connick T. H. (1985) Quinolinic acid and other kynurenines in the central nervous system. Neuroscience 15, 597-617. Teulings F. A. G., Fokkens W., Kaalen J. G. A. H., and Van der Werf-MessingB. (1973)The concentrationoffree and conjugated 3-hydroxyanthranilicacid in the urine of bladder tumour patients before and after therapy, measured with an enzymatic method. Br. J. Cancer 21,3 16-322. Todd W. P., Carpenter B. K., and Schwarcz R. (1989) Preparation of 4-halo-3-hydroxyanthrananilicacids and demonstration of their inhibition of 3-hydroxyanthranilateoxygenase activity in rat and human brain tissue. Prep. Biochem. 19, 155-165. Ueda T., Otsuka H., Goda K., Ishiguro I., Naito J., and Kotake Y. (1978) The metabolism of [carbo~yl-’~C]anthranilic acid. I. The incorporation of radioactivity into NAD’ and NADP’. J. Biochem. 84,687-696. Watanabe M. and Minegishi K. ( 1972)Studies on carcinogenictryp tophan metabolites. 11. Enzymatic formation and hydrolysis of sulfuric ester of 3-hydroxyanthranilicacid. Biochem. Pharmacol. 21, 1347-1356. Watanabe M., Minegishi K., and Tsutsui Y .( 19720) Urinary excretion of free and conjugated forms of 3-hydroxyanthranilicacid. Cancer Res. 32, 2049-2053. Watanabe M., Ohkudo K., and Tamura Z. (1972b) Studies on carcinogenic tryptophan metabolites. I. Enzymatic formation and hydrolysis of glucuronide of 3-hydroxyanthranilicacid. Biochem. Pharmacol. 21, 1337-1346. Wolfensberger M., Amsler U., CuCnod M., Foster A. C., Whetsell W. 0. Jr., and Schwarcz R. (1983) Identification of quinolinic acid in rat and human brain tissue. Neurosci. Lett. 41,247-252. Yeh J. K. and Brown R. R. (1975) Urinary tryptophan metabolites: assay of free and acid hydrolyzable conjugates of 3-hydroxykynurenine and 3-hydroxyanthranilicacid. Biochem. Med. 14, 12-23.
Presence of 3-Hydroxyanthranilic Acid in Rat Tissues and Evidence for Its Production from Anthranilic Acid in the Brain Halina Baran and Robert Schwarcz Maryland Psychiatric Research Center, University of Maryland School of Medicine, Baltimore, Maryland, U S . A
Abstract: As assessed by HPLC with electrochemical detection, 3-hydroxyanthranilic acid (3-HANA) was found to be present in the rat brain and peripheral organs. The highest concentrations were measured in the kidney (86 fmol/mg of tissue) and spleen (56 fmol/mg of tissue), whereas the adrenal gland, liver, heart, and several forebrain areas (hippocampus, striatum, parietal cortex, thalamus, amygdala/pyriform cortex, and frontal cortex) contained less 3-HANA (between 15 and 22 fmol/mg of tissue). Slightly lower concentrations of 3-HANA were found in the brainstem and the cerebellum. The metabolic disposition of 3-HANA was examined in tissue slices which were incubated in Krebs-Ringer buffer at 37°C in vitro. Incubation for up to 2 h did not affect 3-HANA concentration in brain tissue. However, inhibition of 3HANA degradation by the specific 3-hydroxyanthranilic acid oxygenax blocker 4-chloro-3-hydroxyanthranilic acid (4-C13-HANA; 10 r M ) resulted in a rapid (within 2.5 min) doubling of 3-HANA levels in slices from cerebral cortex. No further increases were observed after incubations of up to 120 min. Exposure of cortical dices to 3-HANA’s putative bioprecursors, 3-h ydroxykynurenine (3-HK) and anthranilic
acid (ANA), in the absence of4-Cl-3-HANA resulted in rapid, transient increases in 3-HANA production. Maximal 3HANA synthesis from ANA exceeded the maximal effect of 3-HK by approximately 1 1-fold. In the presence of 4-CI-3HANA, 1 mM 3-HK and 1 mM ANA produced 9.0 ? 0.3 and 89.0 ? 9.3 (5 min) or 51.6 k 7.9 and 187.5 k 11.2 (120 min) fmol of newly synthesized 3-HANA/mg of brain tissue, respectively. In the brain, but not in the spleen, ANA proved to be a superior 3-HANA precursor at lower concentrations as well. Time and dose relationships for the de novo production of 3-HANA from ANA in brain slices were established in the presence of 10 pM 4-Cl-3-HANA. Biosynthesis of 3-HANA from ANA in the brain may be critically involved in the function or dysfunction of 3-HANA’s principal catabolic product, the endogenous excitotoxin quinolinic acid. Key Words: Anthranilic acid-Excitotoxins-3-Hydroxyanthranilic acid-Kynurenines-Neurodegenerative disorders. Baran H. and Schwarcz R. Presence of 3-hydroxyanthranilic acid in rat tissues and evidence for its production from anthranilic acid in the brain. J. Neurochem. 55, 738744 ( 1990).
Quinolinic acid (QUIN), a brain metabolite with convulsive and neurodegenerative properties, has recently attracted considerableattention as a hypothetical pathogen in neurological diseases (Schwarcz et al., 1987). In addition to its possible role in the etiology of temporal lobe epilepsy (Lapin, 1981; Schwarcz et al., 1984a), Huntington’s disease (Schwarcz et al., 1984b; Beal et al., 1986), and other neurodegenerative conditions, QUIN may also serve physiological functions in the brain as a selective agonist at the N-methyl-Daspartate NMDA receptor (Stone and Connick, 1985). In spite of the well-documented effects of QUIN at its receptor, however, relatively little is known about
its metabolism in the brain. The enzyme 3-hydroxyanthranilic acid oxygenase (3-HAO), which converts 3hydroxyanthranilic acid (3-HANA) to QUIN (Nishizuka and Hayaishi, 1963), has been identified and characterized in both rodent and human brain and found to be localized in astrocytes (Foster et al., 1986; Okuno et al., 1987; Schwarcz et al., 1988), but the origin of 3-HANA remains obscure. In the periphery, 3HANA derives from tryptophan via a series of enzymatic reactions along the so-called kynurenine pathway (Schlossberger et al., 1984). However, the two enzymes preceding 3-HA0 in this established biosynthetic cascade, i.e., kynurenine hydroxylase and kynureninase,
Received September 15, 1989; revised manuscript received January 15, 1990; accepted January 22, 1990.
Abbreviations used: ANA, anthranilic acid; 4-C1-3-HANA, 4chloro-3-hydroxyanthranilic acid; 3-HANA, 3-hydroxyanthranilic acid; 3-HAO, 3-hydroxyanthranilic acid oxygenase; 3-HK, 3-hydroxykynurenine; PCA, perchloric acid; QUIN, quinolinic acid.
Address correspondence and reprint requests to Dr. R. Schwarcz at Maryland Psychiatric Research Center, P.O. Box 21247, Baltimore, MD 2 1228, USA.
738
3-HYDROXYANTHRANILIC ACID IN RAT BRAIN
show very low activity in the rat brain (Battie and Verity, 1981; Kawai et al., 1988), and attempts to incorporate kynurenine into QUIN in vivo (Speciale et al., 19896) and in vitro (unpublished observations) have failed to reveal activity of the brain kynurenine pathway in rats. In our quest to examine the biology of the brain’s kynurenine system, it therefore appeared prudent to explore alternative routes of 3-HANA biosynthesis, in particular the possibility of a direct 3-hydroxylation of anthranilic acid (ANA), which has been reported to occur in the mammalian liver (Ueda et al., 1978), in plants (Nair and Vaidyanathan, 1965), and in microorganisms (Schott et al., 1982). I n the present article, we report first results of our search for a new avenue of 3-HANA production in the brain. Thus, using a novel, sensitive analytical method for 3-HANA determination, we have examined the presence of 3-HANA in rat tissues and followed its de novo synthesis from 3-hydroxykynurenine (3-HK) and ANA, respectively, in tissue slices in vitro. A preliminary account of this work has been published in abstract form (Baran and Schwarcz, 1989).
MATERIALS AND METHODS Materials ANA, 3-HANA, and 3-HK were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). 4-Chloro-3-hydroxyanthranilic acid (4-Cl-3-HANA) was prepared as described by Todd et al. (1989) and was kindly provided by Dr. Barry K. Carpenter (Cornell University, Ithaca, NY, U.S.A.). Rat liver 3-HA0 was purified in our laboratory as described by Okuno et al. (1987). All other materials and reagents were of the highest quality commercially available.
Animals Male Sprague-Dawley rats (240-280 g), kept under standard laboratory conditions on a 12-h light/dark cycle with free access to food and water, were used in all experiments.
Measurement of 3-HANA Tissue extracts or standard 3-HANA (see below) were applied to a Dowex formate anion-exchange resin (Bio-Rad AG l-X8,200-400 mesh; column size 0.5 cm X 0.7 cm) prewashed with 2 ml ofdistilled water. Subsequently, the column was washed with 2 ml of distilled water and 0.4 ml of 1 M formate. 3-HANA was eluted with 0.4 ml of 2 M and 0.4 ml of 3 M formate. The combined formate fractions were mixed thoroughly with 0.8 ml of ethyl acetate to extract 3-HANA into the organic phase. After centrifugation at 3,500 g for 5 min, 650 pl of the 3-HANA-containing phase were lyophilized. At this stage, samples could be frozen and kept at -20°C until analysis. The samples were then resuspended in 60 p1 of 0.05 Mperchloric acid (PCA), and an appropriate aliquot (routinely 20 p l ) was subjected to HPLC using a system comprising an 8-cm CISreversed-phase HR-80 column (3 pm spherical octadecylsilane particles; ESA, Bedford, MA, U.S.A.) and a coulometric detector (ESA model 5 IOOA). The mobile phase consisted of 0.1 M sodium phosphate, 0.75 Woctanesulfonic acid, and 7% methanol, pH 3.0. 3-HANA was eluted at a flow rate of 0.5 ml/min and detected electrochemically at +0.40 V with the aid of a C-RSA integrator (Shimadzu, Columbia, MD, U.S.A.).
739
Enzymatic identification of 3-HANA In pilot experiments, the identity of 3-HANA was ascertained by its degradation using a purified preparation of the specific 3-HANA-catabolizing enzyme 3-HA0 (Okuno et al., 1987). For this purpose, Dowex formate eluates containing 3-HANA standards or putative 3-HANA from rat tissue were extracted with ethyl acetate, and the organic phase was dried down. After resuspension in 60 pl of 30 mM 2-(N-morpho1ino)ethanesulfonic acid buffer, pH 6.5, several identical samples were combined. The resulting sample was then split into equal parts, which were mixed with 2 p1 of active or heat-inactivated (control) purified 3-HA0 (sp. act. 7.5 pmol/ min/mg of protein). The samples were then incubated for 60 min at 37°C in the presence of 300 pMFeS04 (Foster et al., 1986), and 10 pl of 1 M PCA were added to stop the reaction. Subsequently, the mixture was centrifuged (10 min, 28,000 g), and an aliquot of the supernatant subjected to HPLC analysis as described above.
Tissue preparation Animals were killed by decapitation, and their organs rapidly removed and dissected out at 4°C. For the determination of 3-HANA concentration, tissues were obtained and frozen on dry ice within 2 rnin after killing. The samples were kept frozen for up to 2 days at -20°C. In separate experiments, slices of cerebral cortex and spleen (base 1 X I mm; approximate weight 1.5 mg/slice) were obtained using a McIlwain chopper (Mickle Laboratory Engineering Co., Gomshall, U.K.). Tissue was cut in one direction, subsequently chopped perpendicular to the original orientation, and immediately used for the experiment.
3-HANA determination in tissue homogenates Frozen tissues were sonicated (1: 10, wt/vol) in 0.1 MPCA, and the resulting homogenate was centrifuged for 15 min at 28,000 g. In every experiment, selected duplicate samples received 1-4 pmol of 3-HANA as an internal standard prior to sonication. One milliliter of the resulting supernatant was applied to the Dowex-formate column, and 3-HANA was eluted, processed, and measured as described above.
Experiment with tissue slices Slices were separated on ice, transferred to culture wells (10 slices/well) containing 1 ml of oxygenated Krebs-Ringer buffer (NaCI, 118.5 mM; KCI, 4.75 mM; MgS04, 1.18 mM, NaH2P04, 12.9 mM, Na2HP04,3 mM; CaCI2, 1.77 mM, Dglucose, 5 mM, pH 7.4), and preincubated for 10 min at 37°C. The tissue was then incubated further in a shaking water bath at 37°C for varying lengths of time (2.5-120 min) in the presence or absence of 3-HK and ANA, two putative bioprecursors of 3-HANA. In some experiments, the selective 3-HA0 blocker 4-CI-3-HANA (Parli et al., 1980; Heyes et al., 1988; Todd et al., 1989) was added at the beginning of the preincubation period, routinely at a final concentration of 10 pM. Following incubation, the culture wells were immediately placed on ice and the slices rapidly separated from the incubation medium. Slices from two wells were pooled and their total weight determined. The slices were then homogenized by sonication ( 1 :10, wt/vol) as described for brain homogenates and the concentration of 3-HANA determined as detailed above.
RESULTS 3-HANA assay and identification When applied directly to HPLC, 3-HANA was eluted with a retention time of between 9.5 and 11 J. Neurochem., Vol. S5, No. 3, 1990
H. BARAN AND R. SCHWARCZ
740
min. Linearity of the assay was ascertained up to 100 pmol of 3-HANA. The sensitivity limit was found to be 25 fmol of 3-HANA (Fig. 1). Processing of standard or tissue 3-HANA through the entire ion-exchange/ extraction/HPLC procedure described in Materials and Methods yielded a recovery of 60 f 2% (mean k SEM of all experiments). Incubation of standard 3-HANA, which had been subjected to the ion exchange/ethyl acetate extraction procedure, with 3-HA0 resulted in the abolishment of the 3-HANA response, thus confirming the identity of 3-HANA. No degradation of 3-HANA was observed when heat-inactivated 3-HA0 was used in the experiment (Fig. 2).
3-HANA content of rat tissues 3-HANA could be measured in all organs examined. As shown in Table 1, the highest tissue concentrations were found in the kidney (86.0 f 10.9 fmol/mg of tissue) and spleen (56.1 f 8.2 fmol/mg of tissue), and approximately equal quantities of 3-HANA (between 10 and 20 fmol/mg of tissue) were detected in the adrenal gland, liver, heart, and forebrain (brain minus cerebellum; Fig. 3b). In separate experiments, conducted with renal, splenic, hepatic, and cerebral tissue, 3-HANA was positively identified using 3-HA0 digestion as described in Fig. 2 (data not shown). Assessment of the regional distribution of 3-HANA in the rat brain revealed only small differencesbetween forebrain areas. However, relatively low concentrations of 3-HANA were found in brainstem and cerebellum (Table 2). Production of 3-HANA in slices Incubation of cerebral cortical slices in Krebs-Ringer buffer for up to 2 h yielded tissue 3-HANA concentrations which were virtually indistinguishable from those measured in brain homogenates (Table 3), indicating the preservation of steady-state levels over an extended period of time ex vivo. Inclusion of the selective 3H A 0 inhibitor 4-C1-3-HANA (10 p M ) in the incubation medium resulted in a rapid (within 2.5 min) doubling of 3-HANA concentrations (Table 3). No
c
r
0
/
6
p' 2 0
1,f 0
/ ,
loo
200 300 4 0 0 SHANA ( fmol)
I
SO0
FIG. 1. 3-HANA standard curve obtained by HPLC (peak height refers to original HPLC records).
J. Neurochem.. Vol. 55, No. 3. 1990
a
1
J JU-
",
-
FIG. 2. HPLC profilesof 700 fmol of standard 3-HANA (arrowhead) subjected to the ion-exchange/extraction/HPLC procedure d e scribed in Materials and Methods. Samples were incubated with heat-inactivated3-HA0 (a) or active 3-HA0 (b)after ethyl acetate extraction as detailed in the text.
further increases were noted upon prolonged incubation in the presence of the drug for up to 120 min. Lower (3 p M ) and higher (30 p M ) concentrations of 4-C1-3-HANA had virtually identical effects as 10 pM 4-C1-3-HANA (data not shown). Incubation of cortical slices in the presence of 3HANA's putative bioprecursors 3-HK and ANA ( 1 m M each), in the absence of 4-C1-3-HANA, resulted in rapid and transient increases in 3-HANA concentrations. Peak effects for both substances were noted within 15 min of incubation. In quantitative terms, maximal 3-HANA production from ANA exceeded maximal 3-HANA synthesis from 3-HK by approximately 1 1-fold (Fig. 4). The synthesis of 3-HANA from ANA was verified by 3-HAO-induced degradation of the metabolic product (cf. Fig. 3). To this end, cerebral cortical slices were incubated for 5 min with 1 mMANA, the samples TABLE 1. Distribution of 3-HANA in rat tissues Organ
fmol/mg of tissue
Kidney Spleen Adrenal gland Liver Heart Forebrain
86.0 k 10.9 56.1 f 8.2 19.7 t 5.1 18.8 f 2.6 17.5 f 4.2 13.2 ? 1.7
Data are the means i SEM of tissues obtained from five animals.
741
3-HYDROXYANTHRANILIC ACID IN R A T BRAIN N OD t 0)
TABLE 3. 3-HANA in cortical slices in the absence or presence of 4-Cl-3-HANA Incubation time (min)
4-CI-3-HANA (10 p M )
2.5
5
120
12.7 t 0.1 25.8 t 1.5'
11.9 t 0.6 28.5 f 1.9"
13.6 t 1.9 31.5 k 2.7"
Tissue slices were incubated for various time periods as detailed in Materials and Methods. Data are expressed in fmol of 3-HANA/ mg of tissue and represent the means t SEM of three samples, each the average of duplicate incubations. " p < 0.05, as compared to slices incubated in the absence of 4CI-3-HANA (Student's t test).
FIG. 3. HPLC profiles of 500 fmol of standard 3-HANA subjected directly to HPLC (a) and a characteristic chromatogram of a rat forebrain extract (b).
processed through the ion-exchange/extraction procedure, and 3-HANA identification carried out as described in Materials and Methods. As shown in Fig. 5, the newly synthesized compound was thus identified as authentic 3-HANA. The next set of experiments was designed to compare the effects of 3-HK and ANA (100 pM and 1 m M concentrations of each compound) on 3-HANA production in the presence of 10 pM 4-Cl-3-HANA. Bioprecursor effects were studied at two time points (5 min and 120 min), and the analysis was performed in parallel in slices from cerebral cortex and spleen. 3H A 0 inhibition prevented the breakdown of newly synthesized 3-HANA regardless of the nature of the bioprecursor. Thus, in the presence of 4-C1-3-HANA,
the 3-HANA content of slices kept increasing when either 3-HK or ANA was used (Table 4). In the brain, ANA was superior to 3-HK as a precursor of 3-HANA under all experimental conditions tested. The most dramatic differences between the two compounds were apparent at a concentration of 100 pMand an incubation time of 5 min. In absolute terms, 3-HANA synthesis from ANA in the brain was comparable to that in the spleen. In contrast, 3-HK was far more active in the spleen, where substantial increases in 3-HANA production were measured when 3-HK concentrations were raised from 100 pM to 1 m M or when the incubation time was extended from 5 to 120 min. Thus, in the spleen 3-HK clearly served as the preferential source of 3-HANA under all conditions tested (Table 4). Time and dose relationships for the de novo production of 3-HANA from ANA in the presence of 10 pM 4-Cl-3-HANA were examined in greater detail in slices from cerebral cortex. Using 100 pM ANA, 3160-
80
TABLE 2. Regional distribution 0f3-HANZ4
a a
in the rat bruin Brain area
fmol/mg of tissue
Hippocampus Striatum Parietal cortex Thalamus Amygdala/pyriform cortex Frontal cortex Brainstem Cerebellum
2 1.9 & 4.4 (5) 21.0 f 2.3 (5) 20.8 f 2.5 (7) 19.9 k 2.7 (6) 19.8 t 3.8 (7) 15.5 t 1.9 (7) 12.2 t 1.8 (6) 8.7 f 1.1 (7)
Data are the means parentheses.
~tSEM
e
of the number of animals given in
40
Minutes
FIG. 4. Time dependency of d e novo 3-HANA production from cortical slices incubated with 1 mM 3-HK (A)or 1 mM ANA (0)at 37°C in the absence of 3-HA0 inhibition. Data are the means k SEM of four to six samples, each the average of duplicate incubations.
J. Neurochem.. Vol. 55. No. 3, 1990
H . BARAN AND R. SCHWARCZ
742
1
n
i
b
0
1
5
10
60
Minutes FIG. 6. Synthesis of 3-HANA in cortical slices exposed to 100 p M ANA and 10 phf 4-Cl-3-HANA for various time periods at 37°C. Data represent newly produced 3-HANA in excess of the effect of 4-CI-3-HANA alone and are the means f SEM of three samples, each the average of duplicate incubations.
FIG.5. HPLC profiles of the 3-HANA-like compound (arrowhead) isolated from cortical slices incubated for 5 min with 1 mM ANA. Samples were incubated with heat-inactivated3-HA0 (a)or active 3-HA0 (b) after ethyl acetate extraction as described in Materials and Methods.
HANA synthesis proceeded rapidly in a linear fashion for 2.5 min, but slowed thereafter, reaching a production rate of approximately 2 fmol/lO min/mg of tissue after 5-60 min of incubation (Fig. 6). Variation of ANA concentration between 10 and 1,000 pM, using an assessment of 3-HANA production after a 5-min incubation period, yielded a curve which approached saturation above 300 pM ANA (Fig. 7).
DISCUSSION 3-HANA has long been known as an integral link in the kynurenine pathway, the primary route of tryptophan degradation in mammalian cells. Because of its possible role in the pathogenesis of bladder cancer
(Boyland and Williams, 1956; Benassi et al., 1963; Teulings et al., 1973), which has been speculatively related to its ability to interfere with mitochondria1 oxidative phosphorylation (Quagliariello et al., 1964), 3-HANA received substantial attention from researchers in the 1950sand 1960s(Gorrod et al., 1967). Studies focussed on the disposition of the substance in urine where 3-HANA occurs in both conjugated and free forms (Brown and Price, 1956; Price et al., 1965), and on the enzymatic mechanisms which are responsible for the hydrolysis of conjugated 3-HANA (Watanabe and Minegishi, 1972; Watanabe et al., 19723). Until recently, only sporadic attempts had been made to examine the presence and function of the kynurenine pathway in the brain. Because of the possible involvement of two kynurenines, QUIN and kynurenic acid, in the pathogenesis of neurodegenerative diseases (Schwarcz et al., 1987), the cerebral disposition of these two compounds has become the subject of increasingly extensive scrutiny. However, prior to the present work, the presence or biosynthesis of 3-HANA in the brain had not been documented (cf. GB1 and Sherman, 1980), although exogenously applied 3-
TABLE 4. De novo production of 3-HANA in cerebral cortical and splenic slices ~
Brain
Spleen
Bioprecursor
Concentration
5 min
120 min
5 min
120 min
3-HK 3-HK
100 pcM
1.0 t 0.1 9.0 0.3
2.1 f 0.2 51.6 ? 7.9
51.4 k 2.1 489.3 f 72.9
2,141.7 k 276.3 4,618.7 k 380.8
ANA ANA
100 pM 1 mM
82.7 f 0.8 187.5 k 11.2
23.9 f 3.2 48.0 f 4.6
44.8 k 1.7 121.7 k 7.2
ImM
* 53.5 f 6.2 89.0 * 9.3
Slices were exposed to 3-HK or ANA at 37°C for 5 or 120 rnin in the presence of 10 pM 4-CI-3-HANA as described in Materials and Methods. Data represent newly produced 3-HANA in excess of the effects of 4-C1-3-HANA alone. They are expressed as fmol of 3-HANA/mg of tissue and are the means k SEM of three samples, each the average of duplicate incubations.
J. Norrochem., Vol. SS, No. 3. 1990
3-HYDROXYANTHRANILIC ACID IN RAT BRAIN
_., I00
.
.
.
I
.
x
I
1
500
I
.
1000
Anthrontlic octd [ pM I
FIG. 7. 3-HANA production in cortical slices incubated for 5 min at 37°C with different concentrations of ANA in the presence of 10 N4-Cl-3-HANA. Data represent newly produced 3-HANA in excess of the effect of 4-CI-3-HANA alone and are the means rt SEM of three samples, each the average of duplicate incubations.
HANA was shown to serve as an effectivebioprecursor of brain QUIN both in vitro and in vivo (Speciale et al., 1988, 1989b). Using a novel tissue extraction/HPLC method with electrochemical 3-HANA detection, and taking advantage of the availability of a highly purified 3-HA0 preparation for its unequivocal identification,we found very small concentrations of 3-HANA in the rat brain. The steady-state brain content of 3-HANA was substantially lower than that in serum (unpublished obYeh and servations)and urine (Watanabe et al., 1972~; Brown, 1979, and also severalfold lower than renal or splenic 3-HANA. Notably, the cerebral concentration of 3-HANA, distributed rather uniformly throughout the rat brain, was found to be approximately half that of 3-HK (Heyes and Quearry, 1988) and one order of magnitude lower than that of its immediate catabolic product QUIN (Wolfensberger et al., 1983; Moroni et al., 1984). Because 3-HANA does not enter the brain from the periphery under physiological conditions (Q. Smith, personal communication), subsequent studies were designed to assess its metabolism in brain slices. Incubation of the tissue in a physiological environment for up to 120 min did not affect 3-HANA content. Moreover, the data obtained from slices were virtually identical to the 3-HANA levels measured in brain homogenates, suggesting a delicate balance of 3-HANA’s anabolic and catabolic enzymes in the intact cell. Indeed, selective blockade of 3-HAO, presumably the major degradative enzyme for 3-HANA in mammals, by 4-C1-3-HANA resulted in a rapid increase in 3HANA tissue concentration. Notably, the 3-HANA surge following 3-HA0 inhibition was limited in scope, most probably because of 3-HANA efflux or the exhaustion of endogenous bioprecursor(s).
743
The next series of experiments was designed to investigate 3-HANA production from exogenously supplied putative bioprecursors. In accordance with the low activity of kynureninase in rat brain (Kawai et al., 1988), 3-HK was found to be only modestly active in effecting the de novo synthesis of 3-HANA in brain slices, while giving rise to much larger quantities of 3HANA in spleen slices under identical experimental conditions. In the absence of 4-C1-3-HANA, 3-HANA levels decreased after reaching an early maximum, suggesting that brain 3-HK, which can enter the cell through the transport site for large neutral amino acids (Speciale et al., 1989a), is converted rather rapidly to 3-HANA, which in turn undergoes fast degradation. In contrast to the spleen, where 3-HK was confirmed as 3-HANA’s preferred bioprecursor, ANA was found to be surprisingly active and, in fact, superior to 3-HK in producing 3-HANA under our experimental conditions. In spite of the fact that ANA penetrates the cell membrane quite poorly (unpublished observations), sufficient ANA must have entered the cell for further sequestration by enzymatic hydroxylation (Schwarcz et al., 1989). De novo synthesized 3-HANA peaked very rapidly after ANA administration; notably, the ANA effect was similar in speed to the 3-HANA increase observed following 3-HA0 inhibition with 4Cl-3-HANA (cf. Table 3 and Fig. 4). Speed and degree of the effect, therefore, indicate that ANA can serve as an efficient bioprecursor of 3-HANA in the brain. The present study did not reveal the existence of separate intracellular pools in the brain which could conceivably be in place to produce 3-HANA from 3HK and ANA. Thus, 3-HA0 inhibition by 4-Cl-3HANA caused rather modest increases in 3-HANA concentrations after 5- 120 min of incubation, regardless of the nature of the bioprecursor. Feedback inhibition of 3-HANA on both of its biosynthetic enzymes or 3-HANA release may afford self-regulation and a substantial slowing of intracellular 3-HANA accumulation after a sudden surge in 3-HANA production. In conclusion, the data presented here suggest that inlhe rat brain 3-hydroxylatinn of ANA may provide a major source of 3-HANA (and hence of QUIN). The presence of ANA in cerebral tissue has not been ascertained as yet, but its synthesis from kynurenine has been reported to occur in brain homogenates in vitro (Kawai et al., 1988). Studies currently in progress in our laboratory, therefore, are designed to examine several elementary questions regarding ANA neurobiology. In turn, the new information can be expected to reveal valuable clues concerning the function and possible dysfunction of QUIN in the brain.
Acknowledgment: We gratefully acknowledge the help of Mr. Peter Krivanek during the early phases of the project. We thank Joyce Burgess for excellent secretarial assistance. This work was supported by USPHS grants NS 16102 and NS 20509 and by a Max Kade Fellowship to H.B. J. Neurochem.. Vol. 55, No. 3, 1990
H . BARAN AND R . SCHWARCZ
744 REFERENCES
Baran H. and Schwarcz R. (1989) Preferential production of 3-hydroxyanthranilic acid from anthranilic acid in rat brain slices. SOC.Neurosci. Abstr. 15, 328.5. Battie C. and Verity M. A. (1981) Presence of kynurenine hydroxylase in developing rat brain. J. Neurochem. 36, 1308-1310. Beal M. F., Kowall N. W., Ellison D. W., Mazurek M. F., Swartz K. T., and Martin J. B. (1986) Replication of the neurochemical characteristicsof Huntington’s disease by quinolinic acid. Nature 321, 168-171. Benassi C. A,, Perissinotto B., and Allegri G. (1963) The metabolism of tryptophan in patients with bladder cancer and other urological diseases. Clin. Chim. Acta 8, 822-831. Boyland E. and Williams D. C. (1956) Metabolism of tryptophan. 2. The metabolism of tryptophan in patients suffering from cancer of the bladder. Biochem. J. 64,578-582. Brown R, R. and Price J. M. (1956) Quantitative studies on metab olites of tryptophan in the urine of the dog, cat, and man. J. Biol. Chem. 219,985-997. Foster A. C., White R. J., and Schwarcz R. (1986) Synthesis of quinolinic acid by 3-hydroxyanthranilicacid oxygenase in rat brain tissue in vitro. J. Neurochem. 41,23-30. Gil E. M. and Sherman A. D. (1980) L-Kynurenine: its synthesis and possible regulatory function in brain. Neurochem. Res. 5, 223-239. Gorrod J. W., Alifano A., Papa S., and Quagliariello E. (1967) Effect of aromatic amines on isolated rat-liver mitochondria. Cancer Res. 21, 668-673. Heyes M. P. and Queany B. J. (1988) Quantification of 3-hydroxykynurenine in brain by high-performanceliquid chromatography and electrochemical detection. J. Chromatogr.428, 340-344. Heyes M. P., Hullo B., and Markey S. P. (1988) 4-Chloro-3-hydroxyanthranilate inhibits brain 3-hydroxyanthranilateoxidase. Neurochem. Int. 13,405-408. Kawai J., Okuno E., and Kido R. (1988) Organ distribution of rat kynureninase and changes of its activity during development. Enzyme 39, 181- 189. Lapin I. L. (1981) Kynurenines and seizures. Epilepsia 22,257-265. Moroni I., Lombardi G., Carl2 V., and Moneti G. (1984) The excitotoxin quinolinic acid is present and unevenly distributed in the rat brain. Brain Res. 295, 352-355. Nair P. M. and Vaidyanathan C. S. (1965) Anthranilic acid hydroxylase from Tecoma stuns. Biochim. Biophys. Acta 110, 521531. Nishizuka Y. and Hayaishi 0. (1963) Studies on the biosynthesis of nicotinamide adenine dinucleotide. I. Enzymatic synthesis of niacin ribonucleotidesfrom 3-hydroxyanthranilicacid in mammalian tissues. J. Biol. Chem. 238, 3369-3377. Okuno E., Kohler C., and Schwarcz R. (1987) Rat 3-hydroxyanthranilic acid oxygenase: purification from the liver and immunocytochemical localization in the brain. J. Neurochem. 49, 771-780. Parli C. T., Krieter P., and Schmidt B. (1980) Metabolism of 6chlorotryptophanto 4-chloro-3-hydroxyanthranilicacid a potent inhibitor of 3-hydroxyanthranilicacid oxidase. Arch. Biochem. Biophys. 203, 161-166. Price J. M., Brown R. R., and Yess N. (1965) Testing the functional capacity of the tryptophan niacin pathway in man by analysis of urinary metabolites. Adv. Metab. Disord. 2, 159-225. Quagliariello E., Papa S., Saccone C., and Alifano A. (1964) Effect of 3-hydroxyanthranilic acid on the mitochondrial respiratory system. Biochem. J. 91, 137-146. SchlossbergerH. G., Kochen W., Linzen B., and Steinhart H., eds ( 1984) Progress in Tryptophan and Serotonin Research. de Gruyter, Berlin.
J. Neurochem.. Vol. 55, No. 3, 1990
Schott H. H., Langenbach-Schmidt D., and Zinke K. (1982) Determination of anthranilate, 3-hydroxyanthranilateand 2,3dihydroxybenzoate in the enzyme assay of anthranilate hydroxylase by thin-layer and high-performance liquid chromatography. J. Chromatogr. 245,85-9 1. Schwarcz R., Brush G. S., Foster A. C., and French E. D. (1984a) Seizure activity and lesions following intrahippocampal injection of quinolinic acid. Exp. Neurol. 84, 1-17. Schwarcz R., Foster A. C., French E. D., Whetsell W. 0. Jr., and Kohler C. (19846) Excitotoxic models for neurodegenerative disorders. Life Sci. 35, 19-32. Schwarcz R., Okuno E., Speciale C., Kohler C., and Whetsell W. 0. Jr. (1987) Neuronal degeneration in animals and man: the quinolinic acid connection, in Neurotoxins and Their Pharmacological Implications (Jenner P., ed), pp. 19-32. Raven Press, New York. Schwarcz R., Okuno E., White R. J., Bird E. D., and Whetsell W. 0. Jr. (1988) 3-Hydroxyanthranilate oxygenase activity is increased in the brains of Huntington disease victims. Proc. Nafl. Acad. Sci. USA 85,4079408 1. Schwarcz R., Baran H., and Okuno E. (1989) Anthranilate hydroxylase in rat brain. Soc. Neurosci. Abstr. 15, 328.6. Speciale C., Ungerstedt U., and Schwarcz R. (1988) Quinolinic acid formation in rat brain in vivo and in vitro: effects of striatal lesions. SOC.Neurosci. Abstr. 14, 479.9. Speciale C., Hares K., Schwarcz R., and Brookes N. (1989a) Highaffinity uptake of L-kynurenine by a Na*-independenttransport of neutral amino acids in astrocytes. J. Neurosci. 9,2066-2072. Speciale C., Ungerstedt U., and Schwarcz R. (1989b) Production of extracellular quinolinic acid in the striatum studied by microdialysis in unanesthetized rats. Neurosci. Left. 104, 345-350. Stone T. W. and Connick T. H. (1985) Quinolinic acid and other kynurenines in the central nervous system. Neuroscience 15, 597-617. Teulings F. A. G., Fokkens W., Kaalen J. G. A. H., and Van der Werf-MessingB. (1973)The concentrationoffree and conjugated 3-hydroxyanthranilicacid in the urine of bladder tumour patients before and after therapy, measured with an enzymatic method. Br. J. Cancer 21,3 16-322. Todd W. P., Carpenter B. K., and Schwarcz R. (1989) Preparation of 4-halo-3-hydroxyanthrananilicacids and demonstration of their inhibition of 3-hydroxyanthranilateoxygenase activity in rat and human brain tissue. Prep. Biochem. 19, 155-165. Ueda T., Otsuka H., Goda K., Ishiguro I., Naito J., and Kotake Y. (1978) The metabolism of [carbo~yl-’~C]anthranilic acid. I. The incorporation of radioactivity into NAD’ and NADP’. J. Biochem. 84,687-696. Watanabe M. and Minegishi K. ( 1972)Studies on carcinogenictryp tophan metabolites. 11. Enzymatic formation and hydrolysis of sulfuric ester of 3-hydroxyanthranilicacid. Biochem. Pharmacol. 21, 1347-1356. Watanabe M., Minegishi K., and Tsutsui Y .( 19720) Urinary excretion of free and conjugated forms of 3-hydroxyanthranilicacid. Cancer Res. 32, 2049-2053. Watanabe M., Ohkudo K., and Tamura Z. (1972b) Studies on carcinogenic tryptophan metabolites. I. Enzymatic formation and hydrolysis of glucuronide of 3-hydroxyanthranilicacid. Biochem. Pharmacol. 21, 1337-1346. Wolfensberger M., Amsler U., CuCnod M., Foster A. C., Whetsell W. 0. Jr., and Schwarcz R. (1983) Identification of quinolinic acid in rat and human brain tissue. Neurosci. Lett. 41,247-252. Yeh J. K. and Brown R. R. (1975) Urinary tryptophan metabolites: assay of free and acid hydrolyzable conjugates of 3-hydroxykynurenine and 3-hydroxyanthranilicacid. Biochem. Med. 14, 12-23.
