Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by McClay, K. Articles by Steffan, R. J. Search for Related Content PubMed PubMed Citation Articles by McClay, K. Articles by Steffan, R. J. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB INDIGO INDOLE Agricola Articles by McClay, K. Articles by Steffan, R. J.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, September 2005, p. 5476-5483, Vol. 71, No. 9
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.9.5476-5483.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Mutations of Toluene-4-Monooxygenase That Alter Regiospecificity of Indole Oxidation and Lead to Production of Novel Indigoid Pigments

Kevin McClay,¹ Corinne Boss,¹ Ivan Keresztes,² and Robert J. Steffan^1*

Shaw Environmental, Inc., Lawrenceville, New Jersey 08648,¹ Cornell University, Ithaca, New York 14853²

Received 3 November 2004/ Accepted 1 April 2005

Top Abstract Introduction Materials and Methods Results Discussion References

 
Broad-substrate-range monooygenase enzymes, including toluene-4-monooxygenase (T4MO), can catalyze the oxidation of indole. The indole oxidation products can then condense to form the industrially important dye indigo. Site-directed mutagenesis of T4MO resulted in the creation of T4MO isoforms with altered pigment production phenotypes. High-pressure liquid chromatography, thin-layer chromatography, and nuclear magnetic resonance analysis of the indole oxidation products generated by the mutant T4MO isoforms revealed that the phenotypic differences were primarily due to changes in the regiospecificity of indole oxidation. Most of the mutations described in this study changed the ratio of the primary indole oxidation products formed (indoxyl, 2-oxindole, and isatin), but some mutations, particularly those involving amino acid G103 of tmoA, allowed for the formation of additional products, including 7-hydroxyindole and novel indigoid pigments. For example, mutant G103L converted 17% of added indole to 7-hydroxyindole and 29% to indigoid pigments including indigo and indirubin and two other structurally related pigments. The double mutant G103L:A107G converted 47% of indole to 7-hydroxyindole, but no detectable indigoid pigments were formed, similar to the product distribution observed with the toluene-2-monooxygenase (T2MO) of Burkholderia cepacia G4. These results demonstrate that modification of the tmoA active site can change the products produced by the enzyme and lead to the production of novel pigments and other indole oxidation products with potential commercial and medicinal utility.

Top Abstract Introduction Materials and Methods Results Discussion References

 
A number of oxygenase enzymes have been shown to oxidize indole, leading ultimately to the production of the industrially useful dye indigo via a combination of enzymatic and nonenzymatic reactions (1, 2, 3, 7, 8, 9, 10, 11, 20, 30, 34, 40, 43). The generally accepted pathway for production of indigo and the related indigoid pigment indirubin involves an initial oxidation of indole at the 3 position, producing indoxyl and isatin (Fig. 1A). In the presence of oxygen, indoxyl undergoes a dimerization, leading to the formation of indigo. Alternatively, indoxyl and isatin can condense to form indirubin. It has been proposed that the use of various indigo-forming oxygenases paired with the inherent indole production of Escherichia coli (1) could serve as an economically viable green chemistry process for the production of this commodity scale textile dye from simple carbon sources (1, 3, 13, 30). To this end, research has been conducted to identify key pathways, enzymes, and fermentation conditions that can be controlled to increase the efficiency of the glucose to pigment conversion and limit the production of the undesired by-product indirubin by limiting the production and accumulation of isatin (1, 30).

View larger version (48K): [in this window] [in a new window]

FIG. 1. Products of enzyme-catalyzed oxidation of indole. (A) Indole oxidation products leading to the formation of indigo and indirubin. (B) Indole oxidation products not known to contribute to the formation of indigoid pigments.

In addition to serving as dye stuffs, indigoid pigments have been shown to be of therapeutic value for a number of diseases, including Alzheimer's disease (24), certain forms of cancer (16, 19, 41), and delayed hypersensitivity (23). The therapeutic action is believed to be due their interaction with the human aromatic hydrocarbon receptor (14) and a number of kinases, including glycogen synthase kinase 3, cyclin-dependent kinase 5, and cyclin-dependent kinase 1 (21, 23, 24, 27, 37). Unlike the textile industry, where indirubin is considered a contaminant of indigo preparations, indirubin is preferred for therapeutic applications, in part because it is more polar than indigo and therefore more bioavailable, allowing for a higher effective dose. Recently, indirubin with various substitutions at the 4-, 5-, 6-, or 7- position was shown to be more inhibitory toward kinases than indirubin (14).

While much work has been done to improve and optimize the biological production of indigo, relatively few studies have investigated the nature or potential application of biologically generated indole oxidation products other than indigo and indirubin, even though some of the products may serve as dyes and pharmaceutical precursors (5, 6, 33, 41, 42). Recent work with engineered variants of human P450 2A6 set a precedent, showing that other indole oxidation products, such as 2-oxindole and 6-hydroxyindole, are possible products of enzymatic indole oxidation (Fig. 1A) (13, 31). Similarly, recent studies with the toluene-2-monooxygenase (T2MO) of Burkholderia cepacia G4 indicate that protein engineering of the diiron-containing monooxygenase can alter the ratio of the indole-derived products produced (4-hydroxyindole, isatin, indigo, indirubin, and isoindigo), which in part accounts for the range of colors seen in the chloroform extracts of cultures expressing the modified isoforms of T2MO (38).

Previously, it was noted that some mutations that impacted the enantiospecificity of butadiene epoxidation catalyzed by the toluene-4-monooxygenase (T4MO) of Pseudomonas mendocina strain KR1 (K. McClay and R. J. Steffan application no. 60/136,602, U.S. Patent Office) also impacted the indole oxidation phenotype of the mutants, varying the intensity and hue of the indigoid pigments formed (26). To determine if such mutations could allow for the production of indole-derived products other than indigo and indirubin, we evaluated the effects of a series of T4MO mutations at positions I100, G103, A107, and F196 on the product distribution resulting from indole oxidation. We found that, in addition to the indigoid pigment precursors (indoxyl and isatin), 2-oxindole and 7-hydroxyindole could be produced in varying quantities by some toluene monooxygenase isoforms. The ratio of these primary oxidation products was altered by single amino acid substitutions, particularly when the amino acids that were altered were adjacent to ligands of the diiron center and also comprised part of the substrate-binding pocket. However, to create T4MO isoforms that produced the pharmaceutical intermediate 7-hydroxyindole in significant quantities, multiple mutations were required. Furthermore, a number of pigments that, to the best of our knowledge, have not been previously described were produced by the G103L variant of T4MO.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Chemicals and reagents.
Indole was obtained from Mallinckrodt Specialty Chemicals (Paris, KY). 2-Oxindole (97%), isatin (98%), 5-hydroxyindole (97%), 4-hydroxyindole (99%), and indigo (95%) were purchased from Aldrich (Milwaukee, WI). Radiolabeled [¹⁴C]indole (labeled at C2 [99%, 50 mCi/mmol]) was obtained from American Radiolabeled Chemicals Inc. (St. Louis, MO). 6-Hydroxyindole (98%) and 7-hydroxyindole were purchased from Matrix Scientific (Columbia, SC) and Synchem Inc. (Des Plaines, IL), respectively.

Bacterial strains, plasmids, and media.
E. coli strain DH10B (Gibco, Rockville, MD) was used as the default cloning and expression host for indole oxidation studies. The initial cloning and site-directed mutagenesis of tmoA was performed via overlap extension PCR essentially as described by Pikus et al. (36) using the primers listed in Table 1. In all cases, the products of the secondary PCRs were digested with EcoRI and BglII and ligated to the plasmid pRS184f (35) that had been similarly digested. The resulting plasmids were then transformed into E. coli DH10B. The DNA encompassing the portion of tmoA that was subject to PCR amplification during the mutagenesis was sequenced to verify that the intended mutations were present and to identify any unplanned mutations that had occurred. A catechol 2,3-dioxygenase-negative variant of the toluene-2-monooxygenase (T2MO)-bearing plasmid pMS64 (39) was constructed by removing the internal KpnI fragment of the dioxygenase gene. The pMS64 plasmid and the T4MO double mutant G103L::A107S were generous gifts from Malcolm Shields (Idaho State University) and Brian Fox (University of Wisconsin, Madison), respectively. E. coli DH10B cultures containing the cloned genes of the various oxygenase isoforms were grown overnight at 30°C in LB broth with 100 µg/ml ampicillin as a selective pressure, and the cloned genes were induced with 0.5 mM isopropyl-ß-D-thiogalactopyranoside (IPTG) for 2 h prior to being harvested.

View this table: [in this window] [in a new window]

TABLE 1. PCR primers used for the creation of T4MO mutants via overlap extension

Indole oxidation and product formation.
Indole and its oxidation products were analyzed using high-pressure liquid chromatography (HPLC), UV spectrophotometry, and thin-layer chromatography (TLC). In general, following growth and induction of T4MO expression, the E. coli cultures were harvested by centrifugation, washed once with Luria-Bertani (LB) broth, and resuspended in a 1:10 dilution of LB broth and basal salts medium (15) containing ampicillin and 0.05 mM IPTG, to a final optical density at 550 nm of 2. The LB broth was included to sustain metabolic activity throughout the assay. Ten milliliters of the resuspended cultures was added to 50-ml serum vials to which indole dissolved in ethanol was added to a final concentration of 0.5 to 5 mM. For samples that were to be analyzed by UV spectroscopy, triplicate vials were prepared for each time point because spectroscopic analysis required the sacrifice of entire samples for each time point. Spectroscopic determination of the relative quantities of indigoid pigments was performed by extracting the entire volume of the serum vials with chloroform, evaporating the extract to dryness under a vacuum, dissolving the pellet in dimethylformamide, and measuring the absorbance at 597 nm as described previously (34).

For HPLC analysis, 500 µl of the culture supernatant was withdrawn through the septa of the serum vials at selected time points and placed in microcentrifuge tubes. The tubes were centrifuged for 10 min at 10,000 x g. The cleared supernatants were transferred into HPLC autosampler vials and analyzed on a Hewlett-Packard 1050 series HPLC device. Indole and the non-cell-associated metabolites thereof were resolved on a Hewlett-Packard Zorbax Eclipse XDB C₈ column (4.6 by 150 mm) with an isocratic mix of methanol and water (50:50) at a flow rate of 0.85 ml/min. Elution of the indole-derived products from the column was monitored at a wavelength of 270 nm. Under these conditions, 4- and 5-hydroxyindole did not fully separate and eluted at 2.7 min. Isatin eluted at 3.1 min, followed by 6-hydroxyindole at 3.4 min, 2-oxindole at 3.8 min, 7-hydroxyindole at 4.5 min, and indole at 7.1 min. Indoxyl was not stable enough to be analyzed by this method.

Analytical TLC separation of the indole-derived products was performed by twice extracting 0.5 to 5 ml of the liquid cultures with equal volumes of chloroform, which was then removed with a stream of nitrogen or under vacuum. The resulting dried pigment pellet was dissolved in chloroform to a volume of 100 µl. Twenty microliters of the extract was then applied to polyester-backed silica gel thin-layer chromatography plates (250-µm-thick layer; Whatman Ltd., Maidstone, Kent, England). The TLC plates were then resolved with a solvent mixture of 50% toluene, 25% acetone, and 25% chloroform. Preparative TLC plates of the same composition were handled in a similar manner to prepare samples for nuclear magnetic resonance (NMR) analysis.

Pigment characterization.
Following TLC separation of the indigoid pigments described above, the separated pigment-containing bands were isolated from the TLC plate and the pigments were extracted from the silica matrix with chloroform (indigo and indirubin) or acetone (pigments 1 and 2). The solvents were evaporated, and the pigments were suspended in dimethylformamide for spectral analysis or dissolved in acetone-d₆ for NMR analysis. A spectral scan (280 nm to 1,100 nm) of the individual pigments was performed using a Thermo Spectronic GENESYS 2 UV/Vis spectrophotometer (Rochester, NY). NMR analyses were performed on a Varian INOVA-600 spectrometer (Varian Inc., Palo Alto, CA) operating at 599.867 MHz and 150.869 MHz for ¹H observation and ¹³C decoupling, respectively. Homonuclear chemical shift correlations were acquired with the standard Varian gradient-selected correlated spectroscopy (gCOSY) pulse sequence. ¹³C chemical shifts were determined indirectly using the standard Varian heteronuclear single quantum correlation and heteronuclear multiple band correlation pulse sequences.

Determining the distribution of indole oxidation products using radiolabeled indole.
To supplement the HPLC data concerning the fate of the indole oxidized by the T4MO isoforms, an indole oxidation experiment was performed as described above, except 0.07 mM (1.47 µCi) [¹⁴C]indole was added along with 0.92 mM unlabeled indole per vial. Following overnight incubation, the cultures were exposed to UV light to ensure that the characteristic green fluorescence demonstrating the presence of indoxyl was no longer visible (40), indicating that the active transformation of indole had ceased. The cultures were then prepared for HPLC as described above, except that a 100-µl subsample was removed prior to centrifugation and placed directly into 5 ml of Optiphase Hi Safe 3 liquid scintillation cocktail. The level of radioactivity in this sample was compared to the same volume of liquid following centrifugation by using liquid scintillation counting in a Wallac 1209 Rackbeta scintillation counter (Pharmacia LKB, Gaithersburg, MD). This allowed us to determine the percentage of the total radioactivity that was precipitated with the cells during centrifugation. This fraction was presumed to be comprised mostly of the hydrophobic indigoid pigments that bound to the cells and were removed by centrifugation. The supernatant was analyzed via HPLC as before, except that the waste stream was collected with a fraction collector that captured 0.425-ml aliquots (30 s of column flow). Then, 400 µl of each collected fraction was added to 5 ml of liquid scintillation cocktail to determine the level of radioactivity in each fraction. The level of radioactivity in each fraction was converted to a percentage of the total radioactivity present in the entire sample that was analyzed.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Differences in the indole oxidation characteristics of the oxygenase isoforms.
The observed differences in the indigo-forming phenotypes of the wild-type T4MO and the mutant isoforms G103L and F196L suggested that these mutations alter the indole oxidation capability of T4MO (see Fig. 3A). To determine if the altered pigmentation of the colonies expressing the T4MO isoforms G103L and F196L was caused by a difference in the quantity of indole oxidized or differences in the composition of the products resulting from indole oxidation, assays were conducted using wild-type T4MO, G103L, and F196L. Initial experiments tracked the degradation of indole via HPLC analysis and the simultaneous formation of pigments by measuring the absorbance of chloroform-extractable pigments at a wavelength of 597 nm. Wild-type T4MO, G103L and F196L all oxidized indole to completion within 5 h, yet the accumulation of indigoid products was not equivalent with the various T4MO isoforms. Pigment accumulation with wild-type T4MO and the F196L mutant reached its maximum at the 5-h time point and decreased slightly over the next 17 h. Ultimately, the F196L clone produced only 42% as much pigment as the wild-type clone. With the G103L isoform, pigment formation occurred rapidly during the first 5 h but, instead of decreasing during the next 17 h, the pigments continued to accumulate, reaching a final level that was threefold greater than that achieved with the wild type.

View larger version (13K): [in this window] [in a new window]

FIG. 3. Identification of indole oxidation products produced by selected T4MO isoforms. (A) LB agar plate displaying selected T4MO isoforms with distinct indole oxidation phenotypes. (B) TLC-based separation of indigoid pigments produced by wild-type T4MO and the G103L isoform. The identity of indigo was determined by comparison with authentic standard (not shown), while the identity of indirubin was determined by comparison to literature values. The max of the selected isolated pigments is also shown. (C) TLC-based separation and identification of commercially obtained hydroxyindoles and 7-hydroxyindole produced by the T4MO isoform G103L::A107G. Color development occurred postseparation during 24 h of exposure to air.

TLC analysis conducted following spectraphotometric analysis of the extracts revealed that all three of the T4MO isoforms produce multiple indole-derived products, including indigo, indirubin, and isatin. The G103L mutant produced a particularly complex mixture of products, including two colored metabolites not produced by the other two isoforms (Fig. 3B). Because the absorbance peaks of the pigments overlapped, the relative contribution of indigo (_max = 602), indirubin (_max = 536), and pigment 1 (_max = 566) to the total absorbance of the extracts as measured at 597 nm could not be determined. The spectrophotometric method of quantifying the production of indigoid pigments was therefore deemed unreliable. Furthermore, the presence of the unidentified pigments created by G103L, and the relative absence of pigments produced by F196L, could not be explained with the traditional pathway proposed for the formation of indigo and indirubin (Fig. 1A), indicating that indole metabolites other than indoxyl and isatin were being created by both G103L and F196L.

Standards of commercially available isatin, 2-oxindole, 4-hydroxyindole, 5-hydroxyindole, 6-hydroxyindole, and 7-hydroxyindole were obtained and used with HPLC analysis to identify and quantify the soluble metabolites of indole oxidation. While most of the standards could be prepared and analyzed as a mixture, fresh stocks of isatin as a single compound were prepared daily because isatin decomposes in aqueous media to form isatic acid, which further degrades to form anthranilate (Fig. 1B) under acidic conditions (18). Furthermore, we found that isatin reacts rapidly with 2-oxindole in a concentration-dependent manner. The product of this reaction was not identified but is likely to be isoindigo (16).

Several of the other T4MO isoforms of interest, and T2MO from B. cepacia G4, were used in additional product distribution studies that included the addition of [¹⁴C]indole as a substrate. Results of these analyses are shown in Table 2. The only water-soluble products of the initial oxidation of indole detected via HPLC with these clones were 2-oxindole and 7-hydroxyindole. Indoxyl and isatin also were formed, as indicated by TLC and fluorescent analysis, but they were too unstable to be detected using the described HPLC sampling and analysis protocols.

View this table: [in this window] [in a new window]

TABLE 2. Product distribution of indole oxidation metabolites with T4MO mutantsa

The percentage of indole remaining in the assay media and the percentage of indole converted to oxindole, 7-hydroxyindole, and unknown products were quantified using HPLC and radiometric analysis. The percentage of indole converted to cell-associated pigments was determined by measuring the difference in the amount of radioactivity in solution before and after removal of the cells via centrifugation. In the case of the negative control (plasmid-free E. coli DH10B), a discrepancy existed between the amount of indole added and the amount of indole detected via HPLC (approximately 0.05 mM under these conditions). This presumably was caused by the conversion of tryptophan to indole, catalyzed by the tryptophanase enzyme produced by E. coli (1), allowing for more than 100% of the added indole to be detected. While the uncertainty introduced by the endogenous indole production was undesirable, the use of an E. coli host was preferable to other expression hosts such as Pseudomonas putida strain PPO200 (22) that metabolized indole independently of the cloned T4MO genes (data not presented). Cultures of a plasmid-free E. coli DH10B removed only 3% of the ¹⁴C-labeled indole from the media in the centrifugation step.

With cultures expressing the various oxygenase isoforms, there was wide variation in the product distribution resulting from indole oxidation (Table 2). For example, 8% of the oxidized indole was converted to oxindole by the mutants A107T and G103L::A107G, whereas F196L converted 74% of the oxidized indole to oxindole; nearly double the oxindole production of the wild-type T4MO. The single mutations G103L, G103V, A107C, and A107T converted up to 17% of the oxidized indole to 7-hydroxyindole, whereas the double mutation G103L::A107G directed 47% of the oxidized indole toward the production of 7-hydroxyindole. Wild-type T2MO converted 38% of the oxidized indole to 7-hydroxyindole. Although more stable than isatin, 7-hydroxyindole and the other hydroxyindoles are also somewhat unstable. In the presence of oxygen, 7-hydroxyindole undergoes a reaction that converts it from a colorless compound into a hydrophilic red/brown pigment that cannot be extracted with chloroform (see Fig. 3A and C). This decay process is likely to lead to an underestimation of the yield of 7-hydroxyindole.

The T4MO mutant that produced the most cell-associated product was the I100V isoform that converted 46% of the oxidized indole to the cell-associated pigments indigo and indirubin, while a low of 11% conversion was achieved with the double mutant G103L::A107G. The fact that both T2MO and G103L::A107G contained levels of cell-associated pigments that were above that of the control cells cannot be explained by the production of indigo or indirubin. These pigments were never observed during TLC analysis with these isoforms, and there was no evidence of the production of the indigoid pigment precursors (indoxyl and isatin) during the active oxidation of indole. However, the accumulation of the nearly colorless compound isoindigo, purportedly formed via the dimerization of 2-oxindole (38), could account for the elevated levels of cell-associated radioactivity. Another potential cause of elevated levels of cell-associated radioactivity is the covalent linkage of reactive indole oxidation products, such as indoxyl and epoxides, to cellular material (1, 32).

Structure of novel pigments.
Even though pigments 1 and 2 (Fig. 1) were isolated from the same TLC plates, pigment 1 was obtained in a higher yield. We therefore focused on determining the structure of pigment 1. Comparison of the chemical shifts obtained for pigment 1 with that of indigo and indirubin (Table 3) suggested that it is structurally related to indigo with some significant differences, though some ambiguity remains regarding the proposed structure of the molecule. This is because not all ¹³C chemical shifts have been identified; notably, vinylic C-2 and C-2', carbonyl C-3', and enolic C-7'. Two-dimensional homonuclear chemical shift correlation spectroscopy (COSY) experiments establish H-4' and H-5' as vicinal aromatic hydrogens, consistent with the proposed 1,2,3,4-tetrasubstituted benzene ring. H-6' and H-8' are coupled to each other with a small coupling constant characteristic of long-range coupling (Fig. 2). No long-range coupling is observed between H-5' and H-6', but their spatial proximity is confirmed by nOe crosspeaks in two-dimensional nuclear Overhauser effect spectroscopy (NOESY) experiments. A key feature of pigment 1 is the surprisingly downfield chemical shift (176.72 ppm) assigned to the C-8b' carbon. The carbonyl-like downfield shift is consistent with the unexpected double bond between N-1' and C-8b', which may account for the unique coloring of this molecule (Fig. 3B).

View this table: [in this window] [in a new window]

TABLE 3. NMR chemical shift data for indigo, indirubin, and pigment 1

View larger version (12K): [in this window] [in a new window]

FIG. 2. Proposed structure of pigment 1. The numbering scheme corresponds to the coordinates reported in Table 3 describing the NMR chemical shifts for pigment 1.

Top Abstract Introduction Materials and Methods Results Discussion References

 
During indole oxidation, the wild-type T4MO produced a mixture of primary oxidation products including indoxyl, isatin, and 2-oxindole. Single amino acid substitutions at the amino acid positions I100, G103, A107, and F196 impacted the regiospecificity of indole oxidation, altering the ratio of the primary oxidation products produced and also allowing for the production of an additional primary oxidation product, 7-hydroxyindole. Because indoxyl production is required for the formation of indigoid pigments (Fig. 1A), shifting the product distribution away from indoxyl in favor of 2-oxindole or 7-hydroxyindole explains the observed decrease in pigment formation with some clones but does not alone account for the presence of the novel indigoid pigments created in the G103L isoform (Fig. 2 and 3B).

G103 forms part of the hydrophobic substrate-binding pocket and is adjacent to the iron ligand E104 (36). It has been proposed that residues homologous to G103 dictate the degree of enantiospecificity observed during the alkene epoxidation catalyzed by different diiron-containing monooxygenases (12, 26, 28, 29). This hypothesis was supported by the observation that replacing G103 in T4MO with the progressively larger residues valine and leucine increased the production of the S-enantiomer of butadiene monoepoxide from 67% to 77% and 90%, respectively (unpublished results; application no. 60/136,602, U.S. Patent Office), approaching the enantiospecificity of T2MO (26). Consequently, it was proposed that the shift in the regiospecificity seen with the G103L isoform is caused by an alteration in how toluene is positioned within the active site, essentially as depicted in Fig. 4A. The repositioning of toluene is likely due to steric interactions between the methyl group of toluene and the larger R group of the leucine at position 103. In this study, we found that converting G103 to the progressively larger amino acids valine and leucine increased the amount of oxidized indole directed toward 7-hydroxyindole production from 0% to 10 and 17%, respectively. Thus, the change in the regiospecificity of indole oxidation catalyzed by the G103L mutant may be satisfactorily explained by extending the altered binding hypothesis described for toluene oxidation to indole (Fig. 4). The progressively larger side chains of the G103V and G103L mutants result in increased levels of steric interaction between the aromatic ring of indole and the R groups of the amino acids located at position 103. Larger side chains appear to favor a positioning of indole that places the aromatic ring near the catalytic diiron center, thereby promoting oxidation of the aromatic ring and the formation of 7-hydroxyindole as indicated in Fig. 4C and D. Furthermore, empirical evidence suggests that the repositioning of indole, as shown in Fig. 4D, is hindered by the steric interactions with the A107 side chain. This hypothesis is supported by the fact that the elimination of the A107 side chain, as in the G103L::A107G mutant, causes a threefold increase in the amount of 7-hydroxyindole that is formed with the single mutation G103L.

View larger version (27K): [in this window] [in a new window]

FIG. 4. Toluene- and indole-binding models describing the formation of the products shown. Panels A and B were adapted from Mitchell et al. (28). The relative positions, as determined by sequence alignments and the crystal structures of methane monooxygenase (37), of two key active-site amino acids that influence the regiospecificity of toluene and indole are shown.

In addition to acquiring the ability to hydroxylate the 7-position of indole, the G103L mutant formed novel indigoid pigments. The structural characterization of these pigments has not been completed, but proposed structures are supported by the combined evidence obtained through NMR, optical, and TLC analysis. The formation of the double bond between N-1' and C-8b', as indicated by the high, carbonyl-like chemical shift for C-8b', may account for the unique color of pigment 1. The color of indigo is affected by the hydrogen bonding that occurs between the amine hydrogen attached to N-1 and the carbonyl oxygen attached to C-3. When intramolecular hydrogen bonds predominate, indigo appears violet in color. Indigo appears blue, however, when intermolecular hydrogen bonds between these atoms and the solvent predominate (44). The proposed elimination of the H-1' hydrogen by formation of the N-1'C-8b' double bond would be expected to alter the color of pigment 1 by altering the hydrogen bonding of the molecule. Unlike pigment 1, the H-1' amino hydrogen of pigment 2 (which is believed to be a structural analog of indirubin) is spatially excluded from forming a hydrogen bond with the oxygen attached to C-3; thus, its elimination would not be expected to vary the color of this pigment, leaving it the same color as indirubin. The additional bulk and hydrophilic nature of these pigments, caused by the addition of the second fused five-member ring (C-6', C-7', and C-8'), would account for the decreased mobility of the pigments on the TLC plates with the described solvent system.

The pathway leading to formation of these pigments has not been completely elucidated, but their formation appears to be dependant on the ability of the oxygenase isoforms to form 7-hydroxyindole. To date, the only enzymes we have found that produce these pigments in appreciable quantities are the G103L isoform and an unidentified enzyme cloned from chromosomal DNA of Pseudomonas sp. strain ENVPC5 (25) that also makes 7-hydroxyindole (data not presented). However, it does not appear that 7-hydroxyindole is converted to the novel pigments directly, because adding 7-hydroxyindole as an extracellular substrate did not increase the level of novel pigments produced (data not presented). Because the novel pigments are modified at positions 6 and 7, corresponding to the two carbons that would be involved in forming the epoxide precursor to 7-hydroxyindole, it is likely that enzymatic or chemical modification of the reactive epoxide precursor of 7-hydroxyindole is involved in the formation of these pigments.

Based on the positioning of the T4MO active-site mutations that altered product distribution of the indole oxidation, it is expected that the analogous active-site mutation of T2MO would yield isoforms that produce indigo. Canada et al. (2) identified an indigo-forming isoform of T2MO that had been altered at a position homologous to T4MO residue I100 (V106), and Rui et al. (38) found that altering the amino acids homologous to T4MO residues I100 and A107 resulted in changes in the amount of isatin, indigo, indirubin, and isoindigo produced. These results are similar to ours, except that we found that wild-type T2MO and the T4MO double mutant G103L::A107G converted 38 and 47% of added indole to 7-hydroxyindole (as determined by comparison to authentic standards via HPLC and TLC separation and color formation following air oxidation [Fig. 3C]), respectively, whereas Rui et al. reported that wild-type T2MO produces an undefined quantity of 4-hydroxyindole. Furthermore, the most abundant product of indole oxidation by wild-type T2MO reported by Rui et al. was isoindigo (86%), hypothesized to be the dimerization product of 2-oxindole (even though it has been shown that isoindigo is formed chemically by the reaction of 2-oxindole and isatin under acidic conditions [16, 38]). If 38% of the added indole is converted to 7-hydroxyindole by T2MO as we have demonstrated, only 60% of the added indole would be available for isoindigo formation. While some of the discrepancies between the product distributions might be explained by culture conditions and sampling techniques, these factors cannot explain the discrepancy between the production of 4- and 7-hydroxyindole. However, Rui et al. did not analyze for the later product and may have misidentified the 7-hydroxyindole as 4-hydroxyindole.

Another amino acid that, like G103, is part of the hydrophobic substrate-binding pocket and adjacent to an iron ligand is F196 (35). Analogs of this residue have been hypothesized to be important for determining the enantiospecificity of alkene monooxygenases (12). Unlike G103L, however, F196L has no appreciable impact on the enantiospecificity of T4MO during the epoxidation of butadiene (unpublished results; application no. 60/136,602, U.S. Patent Office). The F196 position, however, did have a definite impact on the indole oxidation product distribution, directing 74% of the oxidized indole toward 2-oxindole formation, as opposed to only 38% with the wild type. The fact that such a large percentage of the oxidized indole is directed toward oxindole presumably accounts for the decreased production of indigo by the F196L isoform. The mechanism underlying the shift toward 2-oxindole production is difficult to ascertain. Extension of the epoxidation model of catalysis to describe the oxidation of indole by wild-type T4MO predicts that indole should proceed through a 2,3-epoxy-indole intermediate, yielding an indole derivative oxidized at the 2 position. However, only 38% of the indole oxidized is converted to 2-oxindole with wild-type T4MO. This indicates that either the chemistry of the heterocyclic ring affects the resolution of the 2,3-epoxy-indole intermediate or, alternately, the binding of indole in the wild-type T4MO active site might render the 2,3 double bond of indole somewhat inaccessible, favoring the direct hydroxylation of indole at the 3 position to form indoxyl. A similar mechanism has been proposed for the hydroxylation of the methyl groups of p-xylene by T4MO (35). Thus, a subtle shift in the binding position of indole caused by the F196L mutation could make the double bond more accessible, favoring epoxide formation, ultimately leading to increased 2-oxindole production.

As discussed earlier, a great amount of research has been devoted to optimizing the conversion of glucose to indigo by using naphthalene dioxygenase as the catalyst for the oxidation of indole (1, 10). Substitution of the naphthalene dioxygenase with selected toluene monooxygenase isoforms could provide immediate access to a green chemistry route for the synthesis of other valuable indole-derived products, including novel indigoid pigments, 7-hydroxyindole, and 2-oxindole. Although the purity of the novel pigments produced by the G103L isoform may currently be insufficient for use in the textile dye industry, the increased polarity of the indigoid pigments, as demonstrated via TLC, means that they would have a greater bioavailability than indigo and indirubin if they were administered as therapeutic agents. For therapeutic applications, the cost of product separation would be less of a concern than for textile applications. The 7-hydroxyindole produced by the G103L::A107G mutant, and the derivatives thereof, can be used as a keratin fiber dye (6), as antagonists of the insulin-stimulating hormone calmodulin (4, 41), or as a stimulants of dopamine production (32). Similarly, the F196L mutant is potentially useful as a catalyst for the production of 2-oxindole or for converting indoles with aromatic ring substitutions into derivatives of 2-oxindole such as 5-chloro-2-oxindole, a precursor to the antirheumatic drug Tenidap (5). Future research will be directed toward exploring the active-site flexibility of T4MO to improve the yield of desired indole-derived products and toward a better understanding of the pathway leading to the formation of the novel pigments formed by the G103L isoform of T4MO.

 
This work was supported in part by an NSF Small Business Innovative Research Grant (no. DMI-9460076) to R.J.S.

The generous gift of bacterial strains and plasmids from Malcolm Shields and Brian G. Fox is gratefully acknowledged.

 
* Corresponding author. Mailing address: 17 Princess Road, Lawrenceville, NJ 08648. Phone: (609) 895-5350. Fax: (609) 895-1858. E-mail: rob.steffan{at}shawgrp.com.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Berry, A., T. C. Dodge, M. Pepsin, and W. Weyler. 2002. Application of metabolic engineering to improve both the production and use of biotech indigo. J. Ind. Microbiol. Biotechnol. 28:127-133.[CrossRef][Medline]
2
2. Canada, K. A., S. Iwashita, H. Shim, and T. K. Wood. 2002. Directed evolution of toluene ortho-monooxygenase for enhanced 1-naphthol synthesis and chlorinated ethene degradation. J. Bacteriol. 184:344-349.[Abstract/Free Full Text]
3
3. Choi, H. S., J. K. Kim, E. H. Cho, Y. C. Kim, J. I. Kim, and S. W. Kim. 2003. A novel flavin-containing monooxygenase from Methylophaga sp. strain SK1 and its indigo synthesis in Escherichia coli. Biochem. Biophys. Res. Commun. 306:93093-93096.
4
4. Chunmei, Y., R. T. Watson, J. S. Elmendorf, D. B. Sacks, and J. E. Pessin. 2000. Calmodulin antagonists inhibit insulin-stimulated GLUT4 (glucose transporter 4) translocation by preventing the formation of phosphatidylinositol 3,4,5-trisphosphate in 3T3L1 adipocytes. Mol. Endocrinol. 14:317-326.[Abstract/Free Full Text]
5
5. Colgan, S. T., G. R. Haggan, and R. H. Reed. 1996. Assay and purity evaluation of 5-chlorooxindole by liquid chromatography. J. Pharm. Biomed. Anal. 14:825-833.[CrossRef][Medline]
6
6. Cotteret, J., J. Alex, R. Herve, and V. J. Jacques. July 2000. Method for dyeing keratin fibers with compositions which contain 6- or 7-hydroxyindole derivatives as couplers and oxidation dye precursors. U.S. patent 6,090,160.
7
7. Doukyu, N., K. Toyoda, and R. Aono. 2003. Indigo production by Escherichia coli carrying the phenol hydroxylase gene from Acinetobacter sp. strain ST-550 in a water-organic solvent two-phase system. Appl. Microbiol. Biotechnol. 60:720-725.[Medline]
8
8. Drewlo, S., C. O. Bramer, M. Madkour, F. Mayer, and A. Steinbuchel. 2001. Cloning and expression of a Ralstonia eutropha HF39 gene mediating indigo formation in Escherichia coli. Appl. Environ. Microbiol. 67:1964-1969.[Abstract/Free Full Text]
9
9. Eaton, R. W., and P. J. Chapman. 1995. Formation of indigo and related compounds from indolecarboxylic acids by aromatic acid-degrading bacteria: chromogenic reactions for cloning genes encoding dioxygenases that act on aromatic acids. J. Bacteriol. 177:6983-6988.[Abstract/Free Full Text]
10
10. Ensley, B. D., B. J. Ratzki, T. D. Osslund, M. J. Simon, L. P. Wackett, and D. T. Gibson. 1983. Expression of naphthalene oxidation genes in Escherichia coli results in the biosynthesis of indigo. Science 222:167-169.[Abstract/Free Full Text]
11
11. Furuya, T., S. Takahashi, Y. Ishii, K. Kino, and K. Kirimura. 2004. Cloning of a gene encoding flavin reductase coupling with dibenzothiophene monooxygenase through coexpression screening using indigo production as selective indication. Biochem. Biophys. Res. Commun. 313:570-575.[CrossRef][Medline]
12
12. Gallagher, S. C., A. George, and H. Dalton. 1998. Sequence-alignment modeling and molecular docking studies of the epoxygenase component of alkene monooxygenase from Nocardia corallina B-276. Eur. J. Biochem. 254:480-489.[Medline]
13
13. Gillam, E. M., L. M. Notley, H. Cai, J. J. De Voss, and F. P. Guengerich. 2000. Oxidation of indole by cytochrome P450 enzymes. Biochemistry 39:13817-13824.[CrossRef][Medline]
14
14. Guengerich, F. P., F. Martin, M. V. McCormick, W. A. Nguyen, L. P. Glover, and E. Bradfield. 2004. Aryl hydrocarbon receptor response to indigoids in vitro and in vivo. Arch. Biochem. Biophys. 423:309-316.[CrossRef][Medline]
15
15. Hareland, W. A., R. L. Crawford, P. J. Chapman, and S. Dagley. 1975. Metabolic function and properties of 4-hydroxyphenylacetic acid 1-hydroxylase from Pseudomonas acidovorans. J. Bacteriol. 121:272-285.[Abstract/Free Full Text]
16
16. Hoessel, R., S. Leclerc, J. A. Endicott, M. E. Nobel, A. Lawrie, P. Tunnah, M. Leost, E. Damiens, D. Marie, D. Marko, E. Niederberger, W. Tang, G. Eisenbrand, and L. Meijer. 1999. Indirubin, the active constituent of a Chinese antileukaemia medicine, inhibits cyclin-dependent kinases. Nat. Cell Biol. 1:60-67.[CrossRef][Medline]
17
17. Hosoe, T., K. Nozawa, N. Kawahara, K. Fukushima, K. Nishimura, M. Miyaji, and K. Kawai. 1999. Isolation of a new potent cytotoxic pigment along with indigotin from the pathogenic basidiomycetous fungus Schizophyllum commune. Mycopathology 146:9-12.[CrossRef]
18
18. Jensen, J. B., H. Egsgaard, H. Van Onckelen, and B. U. Jochimsen. 1995. Catabolism of indole-3-acetic acid and 4- and 5-chloroindole-3-acetic acid in Bradyrhizobium japonicum. J. Bacteriol. 177:5762-5766.[Abstract/Free Full Text]
19
19. Kapadia, G. J., H. Tokuda, R. Sridhar, V. Balasubramanian, J. Takayasu, P. Bu, F. Enjo, M. Takasaki, T. Konoshima, and H. Nishino. 1998. Cancer chemopreventive activity of synthetic colorants used in foods, pharmaceuticals and cosmetic preparations. Cancer Lett. 129:87-95.[CrossRef][Medline]
20
20. Keil, H., C. M. Saint, and P. A. Williams. 1987. Gene organization of the first catabolic operon of TOL plasmid pWW53: production of indigo by the xylA gene product. J. Bacteriol. 169:764-770.[Abstract/Free Full Text]
21
21. Knockaert, M. M. Blondel, S. Bach, M. Leost, M. Elbi, C. Hager, G. L. Nagy, S. R. Han, D. Denison, M. Ffrench, M. Ryan, X. P. Magiatis, P. Polychronopoulos, P. Greengard, A. L. Skaltsounis, and L. Meijer. 2004. Independent actions on cyclin-dependent kinases and aryl hydrocarbon receptor mediate the antiproliferative effects of indirubins. Oncogene 23:4400-4412.[CrossRef][Medline]
22
22. Kukor, J. J., R. H. Olsen, and J. S. Siak. 1989. Recruitment of a chromosomally encoded maleylacetate reductase for degradation of 2,4-dichlorophenoxyacetic acid by plasmid pJP4. J. Bacteriol. 17:3385-3390.
23
23. Kunikata, T., T. Tatefuji, H. Aga, K. Iwaki, M. Ikeda, and M. Kurimoto. 2000. Indirubin inhibits inflammatory reactions in delayed-type hypersensitivity. Eur. J. Pharmacol. 410:93-100.[CrossRef][Medline]
24
24. Leclerc, S., M. Garnier, R. Hoessel, D. Marko, J. A. Bibb, G. L. Snyder, P. Greengard, J. Biernat, Y. Z. Wu, E. M. Mandelkow, G. Eisenbrand, and L. Meijer. 2001. Indirubins inhibit glycogen synthase kinase-3 beta and CDK5/p25, two protein kinases involved in abnormal tau phosphorylation in Alzheimer's disease. A property common to most cyclin-dependent kinase inhibitors? J. Biol. Chem. 276:251-260.[Abstract/Free Full Text]
25
25. Li, Q.-S., U. Schwaneberg, P. Fischer, and R. D. Schmid. 2000. Directed evolution of the fatty-acid hydroxylase P450 BM-3 into and indole-hydroxylating catalyst. Chem. Eur. J. 6:1531-1535.[CrossRef]
26
26. McClay, K., B. G. Fox, and R. J. Steffan. 2000. Toluene monooxygenase-catalyzed epoxidation of alkenes. Appl. Environ. Microbiol. 66:1877-1882.[Abstract/Free Full Text]
27
27. McGovern, S. L., and B. K. Shoichet. 2003. Kinase inhibitors: not just for kinases anymore. J. Med. Chem. 46:1478-1483.[CrossRef][Medline]
28
28. Mitchell, K. H., J. M. Studts, and B. G. Fox. 2002. Combined participation of hydroxylase active site residues and effector protein binding in a para to ortho modulation of toluene 4-monooxygenase regiospecificity. Biochemistry 41:3176-3188.[CrossRef][Medline]
29
29. Mitchell, K. H., C. E. Rogge, T. Gierahn, and B. G. Fox. 2003. Insight into the mechanism of aromatic hydroxylation by toluene 4-monooxygenase by use of specifically deuterated toluene and p-xylene. Proc. Natl. Acad. Sci. USA 100:3784-3789.[Abstract/Free Full Text]
30
30. Murdock, D., B. D. Ensley, C. Serdar, and M. Thalen. 1993. Construction of metabolic operons catalyzing the de novo biosynthesis of indigo in Escherichia coli. Bio/Technology 11:381-386.[CrossRef][Medline]
31
31. Nakamura, K., M. V. Martin, and F. P. Guengerich. 2001. Random mutagenesis of human cytochrome P450 2A6 and screening with indole oxidation products. Arch. Biochem. Biophys. 395:25-31.[CrossRef][Medline]
32
32. Niederer, C., R. Behra, A. Harder, R. P. Schwarzenbach, and B. I. Escher. 2004. Mechanistic approaches for evaluating the toxicity of reactive organochlorines and epoxides in green algae. Environ. Toxicol. Chem. 23:697-704.[CrossRef][Medline]
33
33. Nudoreku, R., B. Fuooboo, J. Amon, and K. Oberurande. September 1986. 7-Hydroxyindole derivatives, salts therof, use as drug, composition and intermediates. International patent JP61207383.
34
34. O'Connor, K. E., A. D. Dobson, and S. Hartmans. 1997. Indigo formation by microorganisms expressing styrene monooxygenase activity. Appl. Environ. Microbiol. 63:4287-4291.[Abstract]
35
35. Pikus, J. D., J. M. Studts, C. Achim, K. E. Kauffmann, E. Munck, R. J. Steffan, K. McClay, B. G. Fox. 1996. Recombinant toluene-4-monooxygenase: catalytic and Mossbauer studies of the purified diiron and rieske components of a four-protein complex. Biochemistry 35:9106-9119.[CrossRef][Medline]
36
36. Pikus, J. D., J. M. Studts, K. McClay, and R. J. Steffan, and B. G. Fox. 1997. Changes in the regiospecificity of aromatic hydroxylation produced by active site engineering in the diiron enzyme toluene-4-monooxygenase. Biochemistry 36:9283-9289.[CrossRef][Medline]
37
37. Polychronopoulos, P. P. Magiatis, A. L. Skaltsounis, V. Myrianthopoulos, E. Mikros, A. Tarricone, A. Musacchio, S. M. Roe, L. Pearl, M. Leost, P. Greengard, and L. Meijer. 2004. Structural basis for the synthesis of indirubins as potent and selective inhibitors of glycogen synthase kinase-3 and cyclin-dependent kinases. J. Med. Chem. 47:935-946.[CrossRef][Medline]
38
38. Rui, L., K. F. Reardon, and T. K. Wood. 2005. Protein engineering of toluene ortho-monooxygenase of Burkholderia cepacia G4 for regiospecific hydroxylation of indole to form various indigoid compounds. Appl. Microbiol. Biotechnol. 66:422-429.[CrossRef][Medline]
39
39. Shields, M. S., M. J. Reagin, R. R. Gerger, R. Campbell, and C. Somerville. 1995. TOM, a new aromatic degradative plasmid from Burkholderia (Pseudomonas) cepacia G4. Appl. Environ. Microbiol. 61:1352-1356.[Abstract]
40
40. Woo, H., J. Sanseverino, C. D. Cox, K. G. Robinson, and G. S. Sayler. 2000. The measurement of toluene dioxygenase activity in biofilm culture of Pseudomonas putida F1. J. Microbiol. Methods 40:181-191.[CrossRef][Medline]
41
41. Xiao, Z., Y. Hao, B. Liu, and L. Qian. 2002. Indirubin and meisoindigo in the treatment of chronic myelogenous leukemia in China. Leuk. Lymphoma 43:1763-1768.[CrossRef][Medline]
42
42. Yasokui, M. December 1989. Production of 7-hydroxyindole-2-carboxylic acid compound. International patent JP1308249.
43
43. Yen, K. M., M. R. Karl, L. M. Blatt, M. J. Simon, R. B. Winter, P. R. Fausset, H. S. Lu, A. A. Harcourt, and K. K. Chen. 1991. Cloning and characterization of a Pseudomonas mendocina KR1 gene cluster encoding toluene-4-monooxygenase. J. Bacteriol. 173:5315-5327.[Abstract/Free Full Text]
44
44. Zollinger, H. 1991. Color chemistry: syntheses, properties, and applications of organic dyes and pigments, 2nd rev. ed., p. 191-195. Wiley-VCH, Weinheim, Germany.

---

Applied and Environmental Microbiology, September 2005, p. 5476-5483, Vol. 71, No. 9
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.9.5476-5483.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by McClay, K. Articles by Steffan, R. J. Search for Related Content PubMed PubMed Citation Articles by McClay, K. Articles by Steffan, R. J. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB INDIGO INDOLE Agricola Articles by McClay, K. Articles by Steffan, R. J.