Abstract
The tryptophan photoproduct 6-formylindolo[3,2-b]carbazole (FICZ) exhibits the highest aryl hydrocarbon receptor (AhR) binding affinity reported so far. In different cells, in vitro, both extracts of UV-irradiated tryptophan and the synthesized pure compound FICZ induce a rapid and transient expression of AhR-regulated genes. The transient induction suggests that the biotransformation gene battery induced by AhR activation takes part in a metabolic degradation of the ligand, whereby a low steady-state level is regained. The down-regulation of AhR-regulated gene expression was previously shown to be dependent on cytochrome P450 1A1 (CYP1A1). Metabolism of FICZ generates five major metabolites, which appeared as three peaks (M1-M3) in the high performance liquid chromatography. The aim of the present study was to use rat liver S9 from Aroclor-pretreated rats to produce large enough quantities of FICZ metabolites for structure characterization and to determine their product precursor relationship. NMR analysis of large combined fractions of the metabolites indicated that M3 and M2 contained 2 isomers, respectively. By means of liquid chromatography-mass spectrometry (negative ion electrospray mode) and NMR spectroscopy (by 1H-NMR, correlation spectroscopy, and nuclear Overhauser effect spectroscopy techniques) five metabolites of FICZ were identified, and their structures were elucidated. The molecular weights of the two M3 isomers were 300 and both M2 and M1 compounds demonstrated molecular weights of 316, corresponding to addition of one (M3) and of two oxygen (M2 and M1), respectively. The structures were assigned as 2- and 8-hydroxy (M3), 2,10- and 4,8-dihydroxy (M2) and 2,8-dihydroxy derivatives of indolo[3,2-b]carbazole-6-carboxaldehyde (6-formylindolo[3,2-b]carbazole).
Tryptophan, one of the least abundant essential amino acids, is the precursor of several biological signal transducers such as serotonin, melatonin (Hayaishi, 1993), tryptamine, and the plant hormone, indole-3-acetic acid (auxin). Furthermore, the formation of the pyridine nucleotide coenzymes NAD and NADPH is another important tryptophan-dependent process (Hayaishi, 1993).
The amino acids histidine and tryptophan can act as chromophores, and tryptophan is the most strongly near-UV absorbing amino acid. Consequently, both histidine and tryptophan are largely responsible for the UV absorbance of proteins. Upon UV irradiation of tryptophan a variety of photoproducts are formed and some of these have been identified (Creed, 1984; Borkman et al., 1995; Rannug et al., 1995). One group of photoproducts formed consists of substituted indolo[3,2-b]carbazoles that exhibit high affinity for the aryl hydrocarbon receptor (AhR1) (Rannug et al., 1987). The monosubstituted 6-formylindolo[3,2-b]carbazole (FICZ) or indolo[3,2-b]carbazole-6-carboxaldehyde (Rannug et al., 1995) has recently been synthesized (Tholander and Bergman, 1999).
The AhR is a cytoplasmic ligand binding regulatory protein that mediates transcriptional responses to ligands such as the environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin. It is a prototypical member of the basic helix-loop-helix/Per-Arnt-Sim (bHLH/PAS) class of transcriptional factors (Gu et al., 2000). Proteins carrying PAS domains are involved in developmental processes and in adaptation to environmental changes such as in the monitoring of changes in light, redox potential, and oxygen tension as well as changes in the concentration of small ligands (Gonzalez and Fernandez-Salguero, 1998;Taylor and Zhulin, 1999; Gu et al., 2000). A heterodimer consisting of the two PAS proteins, AhR and the AhR nuclear translocator, plays a central role in the induction of a battery of AhR-regulated genes. This battery includes drug-metabolizing enzymes such as the cytochrome P450 family members CYP1A1/1A2 and 1B1, UDP-glucuronosyltransferase, NADPH quinone oxidoreductase, aldehyde dehydrogenase-3, glutathione transferase, as well as a large number of genes involved in growth, differentiation, and cellular homeostasis (reviewed by Nebert et al., 2000). FICZ, the monosubstituted formyl derivative of indolo[3,2-b]carbazole (ICZ) displays an extremely strong affinity toward the AhR with a Kdvalue of 0.07 nM as compared with 0.48 nM for 2,3,7,8-tetrachlorodibenzo-p-dioxin under the same conditions (Rannug et al., 1987). ICZ (Gillner et al., 1985; Bjeldanes et al., 1991) as well as other ICZ derivatives (Gillner et al., 1993) and tryptophan metabolites (Heath-Pagliuso et al., 1998) also appear to bind to the AhR, although with lower affinities. FICZ exhibits the highest binding affinity reported so far, and we have earlier suggested FICZ as a potential endogenous ligand for the AhR. Studies by Wei et al. (1998) have demonstrated that FICZ is an efficient and rapid inducer of CYP1A1 gene expression at very low concentrations (0.1 nM or less) in human keratinocytes and blood mononuclear cells. The induction is transient when low concentrations of FICZ are used (Wei et al., 1998). Previous studies by our group using a liver S9 mix from Aroclor-pretreated rats indicated that FICZ is rapidly metabolized (Wei et al., 1998; 2000). Three major metabolites were detected, and their formation was time-dependent. Studies using mouse Hepa-1 cells of wild type and c37, a cell line defective in CYP1A1 activity, established that the CYP1A1 enzyme is of importance for the metabolism of FICZ in vitro. Metabolites could only be observed in the cell line with a functional CYP1A1 enzyme. Moreover, the results also showed that the basal CYP1A1 mRNA level in c37 cells was dependent on the concentration of tryptophan in the cell culture media (Wei et al., 2000).
It needs to be verified whether FICZ is one of possibly several endogenous AhR ligands. Due to the rapid metabolism of FICZ, its detection in biological systems will be challenging. Therefore, it is essential to identify possible stable metabolites of this substance. Since conversion of tryptophan by enteric bacteria into AhR ligands has been reported (Perdew and Babbs, 1991), as well as conversion of indole-3-carbinol into ICZ after oral intubation in mice (Bjeldanes et al., 1991), it is necessary to unequivocally identify specific metabolites.
In the present study we have characterized the chemical structure of the substances in three major metabolic fractions initially formed from FICZ by an Aroclor-induced rat liver S9-mix and determined their product precursor relationship.
Materials and Methods
Chemicals.
FICZ was synthesized as described earlier (Tholander and Bergman, 1999). Glucose 6-phosphate, NADP, and NADPH were from Sigma-Aldrich (Stockholm, Sweden), dimethyl sulfoxide (DMSO) was from Merck (Darmstadt, Germany). Methanol, acetonitrile, ethyl acetate, and acetone, all of HPLC grade, were also purchased from Merck.
Microsomes/Cytosol Preparations.
Liver S9-fraction from Aroclor 1254-treated Sprauge-Dawely rats was purchased from MOLTOX (Molecular Toxicology Inc., Boone, NC). Microsomes were prepared from the S9 homogenate by a 60-min centrifugation at 105,000g at 4°C. The pellet was homogenized in a small amount of 0.25 M sucrose 50 mM Tris buffer, centrifuged once more and dissolved in 0.15 M KCl. Microsome and cytosol preparations were stored at −70°C. Protein determinations were carried out according to Lowry et al. (1951).
Incubations with Rat Liver S9 Fraction, Microsomes, and Cytosol.
Incubations of FICZ (dissolved in DMSO) with the Aroclor-induced rat liver S9, sample extraction and HPLC analysis were performed according to Wei et al. (2000) with a few modifications. For the metabolite isolation prior to NMR analysis and LC-MS analysis, the FICZ concentration used was either 15 or 20 μM in a reaction volume of 10 ml of S9-mix with a protein concentration of 3.3 mg/ml. The extract was either analyzed by LC-MS or NMR. Structural elucidation of the major metabolites was obtained by 1H-NMR, which required purification of large amounts by multiple HPLC runs. Each crude extract was fractionated and evaporated until dryness using a speed-vac centrifuge. Before the NMR analysis, the fractions were dissolved in methanol, and those containing the same metabolite were combined, evaporated, and stored under Ar (gas) at −20°C.
To study the metabolic pathway of each metabolite, the concentrated metabolite fractions (dissolved in DMSO) were incubated with an S9-mix (1.6 mg of protein/ml), with or without the addition of NADP, for 2.5 min. The extracts obtained were stored at −20°C and analyzed within 48 h by HPLC with UV detection.
The incubation with liver microsomes or cytosol mixtures with or without the addition of 1 mM NADPH contained 5 mM glucose 6-phosphate, 8 mM MgSO4, 45 mM KCl in 100 mM sodium phosphate buffer (pH 7.4). The protein concentration was between 1.6 and 3.6 mg/ml. The reactions were initiated by the addition of a concentrated FICZ metabolite fraction (dissolved in DMSO) and incubated as above.
HPLC Fractionation.
HPLC fractionation and analysis was performed with a Merck Hitachi LaCrom instrument equipped with a l-7100 pump, al-7455 diode array detector and a Shimadzu RF-535 fluorescence HPLC monitor (excitation 390 nm, emission 525 nm) (Shimadzu, Kyoto, Japan). A reverse phase Kromasil 100–5C18 column (250 × 4.6 mm) from Scantec (Partille, Sweden) was employed for the separation. The same mobile phase and gradient was applied as described by Wei et al. (2000) but without the addition of 0.1% trifluoroacetic acid.
Liquid Chromatography Mass Spectrometry.
The HPLC pump employed was a Waters model 2690 separations module (Waters Chromatography, Milford, MA). The built-in autosampler was used for all injections. A polyetheretherketone T-piece splitter was fitted into the polyetheretherketone transfer line connecting the column outlet with the electrospray probe, thus splitting the eluent in a 1:3 ratio between the mass spectrometer and a Waters 996 diode array detector (DAD; Waters Chromatography), respectively. The column used was a Hypersil HyPurity Elite C18 (50 × 2.1 mm, 5-μm particle diameter) from Hypersil (Cheshire, UK). The mobile phase system consisted of 20% acetonitrile in water, to which formic acid had been added to obtain a pH of 3 (eluent A), and 100% acetonitrile (eluent B). Following a 6.5-min isocratic period, a 10-min linear gradient was applied from 0 to 50% B, followed by a 15 min isocratic period, after which the composition was changed to 100% B for 1 min to rinse the column. The flow rate was 0.1 ml/min, and the injection volume was 10 μl. The DAD scan range was 220 to 550 nm.
The mass spectrometer was a Quattro-II triple quadrupole mass spectrometer from Micromass (Manchester, UK). This instrument was fitted with an atmospheric pressure ionization source, with a so-called pepper-pot counter electrode, which throughout this work was used in the negative ion electrospray mode. Nitrogen from a nitrogen generator (1660-NG-MINI-5 Unit, from Aquilo Gas Separation B.V., Etten-Leur, The Netherlands) was used as both the drying gas and nebulizing gas at flow rates of 300 and 20 l/h, respectively. Argon from Air Liquide (Malmö, Sweden) was used as collision gas at a pressure of 1 × 10−3 mBar. The ion source temperature of the mass spectrometer was 80°C, and the capillary voltage was 4.0 kV. The mass spectrometer was used in full scan mode, selected ion recording and product ion scan (daughter ion scan) mode. Mass spectral data were acquired in centroid mode. Acquisition and processing of data from both the mass spectrometer and the DAD were performed using the MassLynx software version 3.2 from Micromass.
NMR Analysis.
NMR spectra were obtained on a Bruker Avance 300 DPX spectrometer (Bruker, Newark, DE) operating at 300 MHz and a Jeol Eclipse + 500 FT NMR spectrometer (Jeol Ltd., Tokyo, Japan) operating at 500 MHz. Spectra of the different metabolite fractions were as recorded in acetone-d6 at 294 K using1H-NMR, homodecoupling, COSY, NOESY, ROESY and1H-13C-HMQC techniques.
Results
Characterization of Metabolite Structures.
Previous studies (Wei et al., 2000) have revealed that three major metabolites of FICZ, 6-formylindolo[3,2-b]carbazole (1, Fig. 1), are formed by an Aroclor induced rat liver S9 mix. These metabolites have been designated metabolite 1 (M1), 2 (M2) and 3 (M3), respectively, where M1 corresponds to the most polar substance. In the present study, the raw extract as well as fractions from incubations of FICZ in the presence of an S9 mix collected from HPLC runs were subjected to LC-MS analyses. Large amounts of combined metabolite fractions were analyzed by NMR for structure determination.
HPLC-MS Analysis.
Two HPLC systems were applied in this investigation, one for fractionation and the other for the LC-MS analysis. Since different columns were used in the two HPLC systems, it was necessary to calibrate them relative to each other. This was done by comparing chromatograms and spectra obtained from the diode array detectors. LC-MS analysis was performed on crude extracts, as well as on pooled fractions, prior to, as well as after, NMR analysis.
Thus, it was possible to tentatively assign molecular weights of the metabolites, which were 316 for M1 and M2 and 300 for M3, respectively (data not shown). These molecular weights correspond to the addition of one to two oxygen atoms to FICZ (molecular weight 284), e.g., resulting from hydroxylation or epoxidation.
HPLC/MS-MS experiments using collision-induced dissociation exhibited a common fragmentation pattern (Fig. 2, a-c) for metabolites of assigned molecular weights of 300 and 316. All three metabolite spectra showed neutral losses ofm/z = 28, 29 and 30, most likely from loss of the formyl group as CO, CHO·, and CH2O, respectively. These spectra were recorded at high cone voltage (70 V) and collision energy (30 eV), thus demonstrating a relatively high stability of the metabolites.
NMR Analysis.
The structures of the five FICZ metabolites constituting fractions M1-M3 (corresponding to metabolites 1–3 in Wei et al., 2000) were elucidated by LC-MS analysis combined with different NMR techniques of individual fractions containing the metabolites. The NMR data of the metabolites are summarized in Table 1. Proton assignments in the NMR spectra were based on a calculated molecular structure of the parent molecule FICZ. A simple PM3 energy optimization indicated that the aldehyde oxygen and the hydrogen atom on nitrogen in position 5 form a planar conformation, which may be ascribed to a hydrogen bond. This hydrogen bond will restrain the aldehyde proton (H6) so that the distance in space to H7 is short, giving rise to a nuclear Overhauser effect (i.e., a cross peak in the NOESY spectrum).
Metabolite fraction 3 (M3).
The 1H-NMR spectrum (Fig.3) showed that the HPLC-fraction of metabolite 3 in fact contained two coeluting substances since the spectrum showed a relative integral relationship of 10:8 between the two substances. In addition, two spin coupling systems, one smaller coupling of 2.3 Hz and one larger coupling of 8.7 Hz were observed, something that can be attributed to a substituent in the A or E ring (1, Fig. 1) meta or para to the indole nitrogen. In the NOESY spectrum, two cross peaks were observed. The first cross peak correlated H6 (the formyl group) of the major component at 11.36 ppm to a doublet at 7.88 ppm (J = 2.3 Hz) that was attributed to H7. In the COSY spectrum (Fig. 4) H7 coupled to the doublet of doublet (dd) at 7.11 ppm (J = 8.7, 2.3) represented by proton H9. The doublet at 7.49 ppm (J = 8.7 Hz) corresponds to H10 that coupled with the neighboring proton H9. On the basis of COSY and NOESY spectrum, the protons in the E ring, of the major component, could be assigned to be 8-hydroxyindolo[3,2-b]carbazole-6-carboxaldehyde (2, Fig. 1). An independent synthesis of this substance confirmed the structure, and the chemical shifts were identical with those assigned. The second cross peak in the NOESY spectrum connected the singlet at 11.44 ppm, of the minor component, to the doublet at 8.48 ppm (J = 7.8 Hz) i.e., H7. From the COSY spectrum, it was obvious that the two different substances in this fraction had the hydroxyl group in different rings since H7-H10 of the minor substance could be correlated to each other in contrast to the COSY spectrum of the major component. We were unable to correlate H4 with the H5 of the minor substance with NOESY spectroscopy, something that would have established the relative position of the hydroxyl group in the A ring. A doublet at 7.65 ppm (J = 2.3 Hz) coupled to the dd at 7.06 ppm (J = 8.2, 2.3 Hz) that consecutively coupled to a doublet at 7.63 (J = 8.2 Hz). These were believed to correspond to H1, H3 and H4, respectively. An independent synthesis of both 2-hydroxyindolo[3,2-b] carbazole-6-carboxaldehyde and 3-hydroxyindolo[3,2-b]carbazole-6-carboxaldehyde, two alternatives of the minor component, established the structure unequivocally as 2-hydroxyindolo[3,2-b]carbazole-6-carboxaldehyde (3, Fig. 1), which could be obtained by demethylation (BBr3) of the correspondingO-methylated derivative, which in turn was prepared by cyclization of 2,2-dichloro-1-[2-(5-methoxy-1H-indol-3-ylmethyl)-1H-indol-3-yl]-ethanone with hydrochloric acid in ethanol (Tholander and Bergman, 1999).
Metabolite fraction 2 (M2).
From the 1H-NMR spectrum, it was obvious that this HPLC fraction, as well, contained two coeluting substances since the relative integral showed a relationship of 10:4. A cross peak, in the NOESY spectrum belonging to the more abundant substance in this fraction was recorded between the singlet at 11.39 (H6) and the doublet at 7.95 (J = 7.8 Hz) i.e., H7. Together with COSY (Fig.5) and homodecoupling spectra, the spin system at position 7–9, could be established. H7coupled with both NOESY and COSY to H8 at 7.06 ppm that in turn coupled to a doublet at 6.97 ppm (J = 7.8 Hz) representing H9. Consequently, the hydrogen at position 10 is substituted with a hydroxyl group. As for metabolite fraction 3, no connection from the hydrogen at position 5 to the hydrogen at position 4 could be detected in the NOESY spectrum, resulting in uncertainty about the relative position, 2 or 3, of the remaining hydroxyl group of the major component (see discussion). However, hydroxylation in position 3 is highly unlikely due to the strong directing effect of the indole nitrogen. For the minor substance in metabolite fraction 2, the spin system in the E ring could be established with 1H-NMR, NOESY, and COSY spectrum. In the NOESY spectrum a cross-peak between H6 at 11.36 ppm and a small doublet at 7.89 ppm (J = 1.8 Hz) was observed. This peak, representing H7, was further correlated by COSY to H9 at 7.12 ppm, which in turn coupled to a doublet at 7.49 ppm (J = 8.7 Hz) thus establishing that a hydroxyl group was situated in position 8 in the E ring. In the A ring in this minor substance a similar spin system as in ring system E of the major substance was observed. The peak at 6.99 ppm coupled to the peak at 7.11 ppm, which in turn correlated to a doublet at 7.75 ppm. Unfortunately no cross-peak between position 4 and 5 could be observed in the NOESY spectrum.
Metabolite fraction 1 (M1).
From the 1H-NMR, COSY (Fig.6) and homodecoupling spectra, two spin systems of ring A and E could be established. In the ROESY spectrum, one cross-peak was found between H6 at 11.32 ppm and a doublet at 7.86 ppm (J = 2.3 Hz), which represents H7. This proton coupled to the dd at 7.10 ppm (J = 2.3, 8.7 Hz), representing H9, which further coupled by ROESY to the neighboring proton H10, a doublet at 7.47 ppm (J = 8.7 Hz), establishing the relative position of one hydroxyl group at position 8. As for the minor metabolite in fraction 3, we were unable to correlate H4 to H5, consequently the relative position of the second hydroxyl group in the E ring was uncertain. A doublet at 7.63 ppm (J = 2.3 Hz) coupled to the dd at 7.04 ppm (J = 2.3, 8.7 Hz), which coupled to a doublet at 7.61 (J = 8.7 Hz). These peaks were believed to correspond to H1, H3, and H4, respectively.1H-NMR homodecoupling and ROESY spectrum were used to establish the position of the hydroxyl group in the E ring to position 8 and in the A ring most likely to position 2 (seeDiscussion). An1H-13C-heteronuclear multiple-bond correlation spectroscopy experiment confirmed that the singlet at 8.45 ppm emanated from a CH signal and not from a hydroxyl group.
Metabolic Pathways and Elucidation of Product-Precursor Relationships.
The three different metabolite fractions were further metabolized by an S9 mix to verify whether there is a product-precursor relationship between the mono-hydroxylated FICZ and the di-hydroxylated derivatives of FICZ and whether NADPH is a crucial cofactor. Metabolism of the most nonpolar metabolite fraction M3 resulted in the formation of both M2 and M1 (Fig. 7) in an NADPH-generating system. The di-hydroxylated metabolites M2 and M1 decreased in the presence of an NADPH-generating system (data not shown). The involvement of microsomal versus cytosolic enzymes in the further metabolism of the most polar metabolite fraction M1 was also investigated. A substantial loss of M1 was only detected in the microsomal system. Only 4% of M1 (2,8-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde) remained after 2.5 min of incubation in the presence of NADPH compared with the control incubation without NADPH. In the cytosolic incubation, 88% remained in the presence of NADPH eliminating the significance of cytosolic enzymes such as aldehyde dehydrogenase etc. in the further metabolism of metabolite 1.
Discussion
The present study on the metabolism in vitro of the Ah-receptor ligand FICZ (1, Fig. 1) has been focused on the structural elucidation of the major metabolites observed in our previous study (Wei et al., 2000). By means of LC-MS and NMR spectroscopy, five metabolites (fractions M1-M3) of FICZ have been identified and their structures elucidated. The metabolism of FICZ is NADPH-dependent indicating a cytochrome P450-mediated metabolic pathway and hence hydroxylation reactions in the aromatic rings. Based on the NMR spectra, it was difficult to unequivocally determine the relative position of the hydroxyl groups for three of the metabolites, one monohydroxylated and two dihydroxylated. However, by studying the product precursor relationship of the individual metabolites, one could determine the positions of most of the hydroxyl groups.
As described under Results, it was problematic to establish a correlation between the proton at position 4 with the hydrogen in the 5-position with NOESY spectroscopy both for the minor metabolite in fraction 3 and for the major metabolite in fraction 2, as well as for metabolite 1. Such a correlation would have firmly established the relative position of the hydroxyl group in these substances. However, an independent synthesis of both the 2 and 3-hydroxy isomers of FICZ defines the structure of the minor metabolite to 3 (Fig. 1). The relative position of the hydroxyl group in ring A in the major substance of metabolite fraction 2 was also difficult to determine by NMR spectroscopy. From the 1H NMR spectrum and COSY spectrum, a spin system attributed to the A ring was seen. Comparison of the chemical shifts and coupling constants for the spin system in ring A for this major substance of metabolite fraction 2, with the peaks and coupling constants for the spin system in ring A in the minor metabolite of fraction 3 (3, Fig. 1) demonstrated a close relationship with almost identical chemical shifts and coupling constants. Due to this close relationship, it seems likely that the major substance in metabolite fraction 2 is 2,10-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde (5, Fig. 1). The minor substance of metabolite fraction 2 has a hydroxyl group situated at position 8 in the E ring. There are two possible structures in which the second hydroxyl group could be situated in the A ring, at position 1 or 4. Based on the polarity of the smaller component of metabolite fraction 2 and the fact that it coeluate with compound 5 (Fig. 1), we conclude that substitution in position 4 seems most likely. This gives the structure 4,8-dihydroxylindolo[3,2-b]carbazole-6-carboxaldehyde (4, Fig. 1).
As for the major metabolite of fraction 2 (5, Fig. 1), the chemical shifts in the A ring of metabolite 1 were very much the same as those in the A ring of 2-hydroxylindolo[3,2-b]carbazole-6-carboxaldehyde. In this context we suggest that metabolite 1 is 2,8-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde (6, Fig. 1). These observations are also reinforced by the fact that 2,8-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde is a metabolic product (Fig. 7) of either the 2 or 8-substituted metabolite of FICZ, 3 and 2, respectively in Fig. 1 (i.e., metabolite fraction 3). In our previous report (Wei et al., 2000), we postulated a product-precursor relationship, which could be confirmed by demonstrating the formation of both M2 and M1 from M3 when incubated in the presence of a rat liver S9 system with NADPH.
An autoregulatory feedback loop has been proposed to exist for controlling transcriptional activation of the AhR-regulated genes through the functional CYP1A1 enzyme (Hankinson et al., 1985;RayChaudhuri et al., 1990; Chang and Puga, 1998). Our experimental observations presented in this paper together with our earlier data (Wei et al., 1998; 2000) further support the idea of FICZ as an endogenous AhR ligand and a CYP1A1 substrate, thereby explaining the observed feedback regulation of AhR activation. The mechanism of formation of FICZ from tryptophan is still not known in detail. It has been reported that irradiation of tryptophan leads to the formation of indole-3-acetaldehyde (Saito et al., 1984), which is a possible precursor of FICZ and also a precursor of the plant hormone auxin mentioned above. It has been suggested that the oxidation of indol-3-acetaldehyde would lead to the formation of a α-hydroxyacetaldehyde (Narumiya et al., 1979), which readily could undergo acid-catalyzed condensation to yield FICZ (Rannug et al., 1995). Studies exploring the involvement of other AhR-regulated metabolizing enzymes in the metabolism of FICZ are under way.
Footnotes
- Abbreviations used are::
- AhR
- aryl hydrocarbon receptor
- FICZ
- 6-formylindolo[3,2-b]carbazole
- bHLH/PAS
- helix-loop-helix/Per-Arnt-Sim
- ICZ
- indolo[3,2-b]carbazole
- DMSO
- dimethyl sulfoxide
- HPLC
- high performance liquid chromatography
- LC-MS
- liquid chromatography-mass spectrometry
- DAD
- diode array detector
- COSY
- correlation spectroscopy
- NOESY
- nuclear Overhauser effect spectroscopy
- ROESY
- rotating frame nuclear Overhauser effect spectroscopy
- Mx
- metabolite fractions 1, 2, or 3
- 1
- 6-formylindolo[3,2-b]carbazole
- 2
- 8-hydroxyindolo[3,2-b]carbazole-6-carboxaldehyde
- 3
- 2-hydroxyindolo[3,2-b]carbazole-6-carboxaldehyde
- 4
- 4,8-dihydroxylindolo[3,2-b]carbazole-6-carboxaldehyde
- 5
- 2,10-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde
- 6
- 2,8-dihydroxyindolo[3,2-b]carbazole-6-carboxaldehyde
- Received April 16, 2002.
- Accepted October 29, 2002.
- The American Society for Pharmacology and Experimental Therapeutics