Biotechnol Lett (2008) 30:869875 DOI 10.

1007/s10529-007-9627-8

ORIGINAL RESEARCH PAPER

Site-directed mutagenesis of substrate binding sites of azoreductase from Rhodobacter sphaeroides

Guangfei Liu Jiti Zhou Jing Wang Bin Yan Jingmei Li Hong Lu Yuanyuan Qu Ruofei Jin

Received: 15 October 2007 / Revised: 11 December 2007 / Accepted: 11 December 2007 / Published online: 29 December 2007 Springer Science+Business Media B.V. 2007

Abstract Comparison of three-dimensional structures of avin-dependent azoreductases revealed two conserved loops around the avin mononucleotide (FMN) cofactor. Tyr74, His75 and Lys109 in the two loops of azoreductase AZR from Rhodobacter sphaeroides were replaced with Trp, Asn and Ala/His by site-directed mutagenesis, respectively. The optimal pH values of K109H and H75N were pH 6, and those of K109A and Y74W were pH 9. The optimal temperature (30C) was not affected by mutation. Positively charged residues at position 109 is critical for the binding of methyl red. K109 might only be involved in the binding of the 20 -phosphate group of NADPH and have no effect on the binding of NADH. Y74W and H75N mutations decreased the binding of methyl red/nitrofurazone and had no affect on the binding of NADPH. Keywords Azoreductase Flavodoxin Nitroreductase Rhodobacter sphaeroides Site-directed mutagenesis

Introduction Azo dyes characterized by the presence of one or more azo groups are widely used in textile, printing, cosmetics, pharmaceuticals, food and many other industries because of their ease of synthesis and chemical stability (Raf et al. 1997). However, the release of these compounds into the environment is undesirable, not only because of their color, but also because many azo dyes and their breakdown products are toxic and/or mutagenic to life (Shore 1996; Chung et al. 1983). Biological treatment of azo dyes using bacteria has been widely studied recently (Stolz 2001; Dos Santos et al. 2007). Azoreductase is the key enzyme expressed in azodye-degrading bacteria. Utilizing NAD(P)H as electron donor, it catalyzes the reductive cleavage of the azo bond. Besides bioremediation, azoreductase is also involved in the site-specic delivery of azo prodrugs (Hanauer 1996). There are at least two different types of azoreductases in bacteria: those requiring avin and those that do not (Chen 2006). Flavindependent azoreductases can be classied into two families according to their amino acid sequences (Chen et al. 2005; Liu et al. 2007b). Azoreductases from Escherichia coli and Bacillus sp. OY1-2 are representative of the two families, respectively (Nakanishi et al. 2001; Suzuki et al. 2001). Many studies have demonstrated that the reduction catalyzed by azoreductase follows the ping-pong mechanism (Nakanish et al. 2001; Yan et al. 2004).

G. Liu J. Zhou J. Wang J. Li H. Lu Y. Qu R. Jin (&) School of Environmental and Biological Science and Technology, Dalian University of Technology, 116024 Dalian, P.R. China e-mail: ruofeijin@yahoo.cn B. Yan Department of Environmental Engineering, Xiamen University of Technology, 361024 Xiamen, P.R. China

123

870

Biotechnol Lett (2008) 30:869875 Table 1 Primers used Primer Oligonucleotide sequence

Structure determination of azoreductase facilitates the elucidation of its molecular mechanism of function. For avin-dependent azoreductases, residues in the vicinity of the isoalloxazine ring are suggested to be involved in the binding of substrates (Ito et al. 2006; Liu et al. 2007b). However, there is still no site-directed mutagenesis study on the active site involved in the binding of azo dye and/or NAD(P)H. The 3D structure model of azoreductases AZR from Rhodobacter sphaeroides was constructed in our previous work. It displayed a avodoxin-like fold with a three layer a/b/a structure (Liu et al. 2007a). In this study, conserved residues at positions 74, 75 and 109 were targeted for site-directed mutagenesis and their effects on activities of AZR were examined.

pGEX50 50 -GGG CTG GCA AGC CAC GTT TGG TG-30 pGEX30 50 -CCG GGA GCT GCA TGT GTC AGA GG-30 K109A K109H Y74W 50 -TGG TGG CGG TGC AGG TGG AAT AAA TG-30 50 -TTC CAC CTG CAC CGC CAC CAG CAA C-30 50 -TGG TGG CGG TCA CGG TGG AAT AAA TG-30 50 -TTC CAC CGT GAC CGC CAC CAG CAA C-30 50 -TAC ACC AGA ATG GCA TAA TGC GAT GAG-30 50 -CAT TAT GCC ATT CTG GTG TAC ATA AT-30 H75N 50 -ACC AGA ATA TAA TAA TGC GAT GAG CG-30 50 -CG CAT TAT TAT ATT CTG GTG TAC AT-30 Nucleotides underlined are the positions chosen for mutation

Materials and methods Chemicals Methyl Red and Acid Red 14 were kindly provided by Dye Synthetic Laboratory, Dalian University of Technology. Nitrofurazone, NAD(P)H and BSA were purchased from Sigma-Aldrich. IPTG, restriction and modication enzymes were obtained from TaKaRa. All other reagents used were of analytical grade.

sequencing before being subcloned into the expression vector.

Expression and purication of wild-type and mutant AZR E. coli JM109 strains transformed with original wildtype or mutated pGEX-AZR were grown aerobically at 37C in LB medium supplemented with ampicillin (1 mM). After reaching an OD660 of 0.7, cells were induced with 1 mM IPTG and incubation continued for another 6 h. The purication of wild-type and mutant fused enzymes were performed as described previously (Yan et al. 2004). Non-fusion enzymes were obtained by digestion with thrombin (Liu et al. 2007a). Protein concentration was measured according to the Bradford procedure, using BSA as a standard.

Bacterial strains and plasmids E. coli JM109 (TaKaRa, Dalian) was used for expression studies. The expression vector, plasmid pGEX 4T-1 (Amersham Pharmacia) was used for cloning azoreductase gene.

Site-directed mutagenesis The AZR expression vector pGEX-AZR was previously constructed by inserting the azr gene into pGEX 4T-1 under the control of a lac promoter. The azr gene was fused to a glutathione S-transferase gene (Yan et al. 2004). Oligonucleotide-mediated mutagenesis of the azr gene was performed by a PCRbased procedure using pGEX-AZR as the template. The mutagenic primers used in the study were shown in Table 1. The mutations were conrmed by

Enzyme assays The standard reaction system containing 20 mM phosphate buffer (pH 7.0), 0.3 mM NAD(P)H, 0.1 mM of a given electron acceptor, and suitable amount of enzyme in 2 ml of reaction mixture, was incubated at 30C. The reaction was initiated by the addition of NAD(P)H. One unit of enzyme activity was dened as the amount of enzyme required to reduce 1 lmol electron acceptor per min under the assay conditions.

123

Biotechnol Lett (2008) 30:869875

871

Azoreductase activity was detected by following the disappearance of Methyl Red (e = 23.4 mM-1 cm-1) or acid red 14 (e = 17.1 mM-1 cm-1) at their respective maximum wavelength (Yan et al. 2004). Nitroreductase activity was monitored based on the reduction of nitrofurazone at 400 nm (e = 12.96 mM-1 cm-1) (Liu et al. 2007a). All the experiments were carried out at least three times.

Kinetic characterization To determine the kinetic parameters of the enzyme, initial rates of the enzymatic reaction were measured by varying the concentration of NAD(P)H while the concentration of the electron acceptor was kept constant. Values of Km and Vmax were obtained from LineweaverBurk doublereciprocal plots.

Fig. 1 Comparison of structures of avin-dependent azoreductases of two families. The gure was prepared with program PyMol (DeLano 2003). Ribbon representation of AZR from R. sphaeroides and AzoR from E. coli are shown in red and green, respectively. The FMN cofactors are shown as stick models

Results and discussion Analysis of sequence and structure similarity Crystal structures of Family I avin-dependent azoreductases from E. coli (PDB 1V4B), Enterococcus faecalis (PDB 2HPV) and Pseudomonas aeruginosa (PDB 2V9C) have been determined (Ito et al. 2006; Liu et al. 2007b; Wang et al. 2007). We constructed a structure model of azoreductase AZR from R. sphaeroides, a member of Family II avin-dependent azoreductase, with homology modeling method (Liu et al. 2007a). Although there is no sequence similarity between azoreductases of the two families, they all share a common overall topology (rmsd of 9.136 A over 735 atoms). The avin mononucleotide (FMN) cofactor is also bound in a similar manner, with the re-face buried and the si-face largely exposed to the solvent (Fig. 1). Multiple sequence alignment of amino acid sequence of AZR with those of other Family II avindependent azoreductases reveals high sequence similarities from Pro72 to Gly111 (Fig. 2). The signature sequence (P-X-Y-H/N-6X-L-K-N-S/A-L/I-D) of a new avodoxin family (Liu et al. 2007a) and the glycinerich region (Gly106Gly111) are well conserved among all these azoreductases. According to the analysis of 3D structures, the two conserved regions

correspond to two loops around the FMN group, respectively. The signature sequence (loop 1) is reported to be involved in the binding of FMN. The glycine-rich region (loop 2) located at the edge of the pyrimidine ring moiety is suggested to be putative nucleotide binding motif, which may be used by NAD(P)H to dock in direct vicinity to FMN (Liger et al. 2004; Chen et al. 2005). In addition, similar loops were also found in the other avin-dependent azoreductase family (Fig. 3), further revealing the structural similarity of the two families. All avin-dependent azoreductases have a tyrosine residue at the re-face of the FMN ring except for E. faecalis azoreductase, where the tyrosine residue is replaced by a tryptophan residue. On the other hand, most crystallized avodoxins have a tryptophan residue at the re-face (Freigang et al. 2002). Based on sequence and structure analysis, Tyr74, His75 and Lys109 in the two loops of wild-type AZR were changed to Trp, Asn and Ala/His, respectively.

Temperature and pH proles of AZR As shown in Fig. 4a, the maximum activity of wildtype AZR was observed from pH 78. The H75N and K109H mutants had optimal activities at pH 6, while Y74W and K109A mutants had optimal activities at pH 9. The maximum activities of Y74W, H75N,

123

872

Biotechnol Lett (2008) 30:869875

Fig. 2 Multiple sequence alignment of amino acid sequences of Family II avin-dependent azoreductases from R. sphaeroides AS1.1737 (Rsph, GenBank accession number AAN17400), Bacillus sp. OY1-2 (BOY1-2, GenBank accession number BAB13746), Bacillus cereus (Bcer, GenBank accession number AE016877), Bacillus anthracis (Bant, GenBank accession number AE017334), Geobacillus stearothermophilus (Gste, GenBank accession number AB071367), Fig. 3 Comparison of conserved loops around FMN cofactors of avindependent azoreductases from (a) R. sphaeroides, (b) S. cerevisiae, (c) E. coli and (d) E. faecalis. The gures were generated using program PyMol (DeLano 2003). Helices, strands and conserved loops are red, yellow and blue, respectively. Bound FMN is represented by stick

Saccharomyces cerevisiae (Scer, GenBank accession number 1T0IA), Schizosaccharomyces pombe (Spom, GenBank accession number NP588084) and Staphylococcus aureus (Saur, GenBank accession number AY545994). Two putative conserved loops (loop 1 of Pro72-Ser79 and loop 2 of Gly106 Gly111) are labeled. The alignment was generated using program GeneDoc (Nicholas et al. 1997)

K109H and K109A mutants decreased by 35%, 24%, 31% and 90%, respectively. The effects of temperature on the activity of AZR were studied at pH 7.0. As shown in Fig. 4b, wild-type and mutant AZR all have their maximum activities at 30C. The maximum activities of Y74W, H75N, K109H and K109A mutants decreased by 48%, 40%, 46% and 91%, respectively.

Kinetic characterization of AZR The catalytic activities of the wild-type and mutant AZR were measured using NADPH as electron donor and Methyl Red as electron acceptor. As shown in Table 2, whereas its Km (Methyl Red) value remained unchanged, the K109H substitution caused a 7.5-fold increase in Km (NADPH) as well as an almost 5-fold

123

Biotechnol Lett (2008) 30:869875

873

A
Relative Activity (%)

120 100 80 60 40 20 0 4

Wild Type K109H K109A Y74W H75N

pH

10

100

Relative Activity (%)

80

Wild Type K109H K109A Y74W H75N

60

40

20

0 20 30

Temperature (C)

40

50

60

70

Fig. 4 Effects of pH (a) and temperature (b) on activities of wild-type and mutant AZR. The activity was measured using Acid Red 14 as a substrate at 30C in 20 mM sodium acetate buffer (pH 46), 20 mM sodium phosphate buffer (pH 57) and 20 mM Tris/HCl buffer (pH 710), respectively. The azoreductase activities were also determined from 20 to 70C by assay in 20 mM sodium phosphate buffer, pH 7. Relative activity is expressed as a percentage of the maximum activity (3.97 U mg-1) of wild-type AZR at 30C, pH 7

decrease in Vmax, thus resulting in an almost 5-fold decrease in Vmax/Km (Methyl Red) value. In addition, the K109A substitution caused over 3-fold increase in values of Km (NADPH) and Km (Methyl Red), and resulted in almost complete loss of activity. Thus, a positively charged residue at position 109 is critical for an effective binding of Methyl Red. However, K109H is not a conserved mutation for the binding of NADPH. The Km (NADPH) values of Y74W and H75N were close to that of the wild-type enzyme. However, Y74W and H75N respectively exhibited 4.5- and 3-fold increase in Km (Methyl Red) values along with over 5-and 2.8-fold decrease in Vmax values, thus resulting in about 22- and 8-fold decrease in their Vmax/Km (Methyl Red) values, respectively. It seems that Y74W and H75N mutations have no effect on the binding of NADPH, while decreasing the binding of Methyl Red. Using NADH as electron donor, the azoreductase activities of K109H and K109A mutants were studied. Compared with those of wild-type AZR, the Vmax and Km (Methyl Red) values of K109H were not affected, thus remaining an almost unchanged catalytic efciency (Vmax/Km (Methyl Red)). Whereas the Km (Methyl Red) value of K109A increased 2.5-fold, its Vmax value decreased 7-fold, thus resulting in about 17-fold decrease in Vmax/Km (Methyl Red) value (Table 3). The results also demonstrated the importance of a positively charged residue at position 109 for the binding of Methyl Red. However, in contrast to the results obtained with NADPH, the Km (NADH) values of K109H and K109A are comparable to that of wildtype AZR. The nucleotide binding motif (GXGXXG) is conserved in many NAD(P)H-dependent enzymes (Jornvall et al. 1995; Bellamacina 1996; Bottoms et al. 2002). The positively charged lysine is the only non-glycine residue in the motif sequence of AZR.

Table 2 Kinetic parameters of wild-type and mutant NADPH-dependent Methyl Red reductase

Enzyme

Vmax (U mg-1) 68.38 14.55 1.13 13.63 24.27

Km (Methyl (mM) 0.27 0.27 0.66 1.21 0.80

Red)

Km (NADPH) (mM) 0.09 0.68 0.31 0.11 0.12

Vmax/Km (Methyl Red) (U mg-1 mM-1) 253.26 53.89 1.71 11.26 30.34

Wild-type K109H K109A Y74W H75N

123

874 Table 3 Kinetic parameters of wild-type and mutant NADH-dependent Methyl Red reductase

Biotechnol Lett (2008) 30:869875

Enzyme

Vmax (U mg-1) 0.91 0.87 0.13

Km (Methyl (mM) 0.06 0.05 0.15

Red)

Km (NADH) (mM) 0.38 0.34 0.36

Vmax/Km (Methyl Red) (U mg-1 mM-1) 15.17 17.40 0.87

Wild-type K109H K109A

Table 4 Kinetic parameters of wild-type and mutant NADPH-dependent nitrofurazone reductase

Enzyme

Vmax (U mg-1) 13.78 3.71 0.80 5.48 ND

Km (nitrofurazone) (mM) 0.13 0.12 0.10 0.65 ND

Km (NADPH) (mM) 0.06 0.20 0.37 0.05 ND

Vmax/Km (nitrofurazone) (U mg-1 mM-1) 106 30.92 8 8.43 ND

Wild-type K109H K109A Y74W

ND, activity that cannot be detected

H75N

Positively charged residues, such as arginine, lysine and histidine, are usually involved in the recognition of the 20 -phosphate group of NADPH (Wilson et al. 1992; Reyes et al. 1995; Kobori et al. 2001). Compared with results obtained using NADPH as electron donor, the unaffected Km (NADH) values of K109H and K109A suggested that K109 might only be involved in interaction with (the 20 -phosphate group of) NADPH but not NADH. Our previous studies have demonstrated that AZR also functions as nitroreductase (Liu et al. 2007a). Nitroreductase activities of wild-type and mutant AZR were investigated using NADPH as electron donor and nitrofurazone as electron acceptor. As shown in Table 4, compared with the wild-type AZR, the Vmax values of K109H and K109A decreased about 4- and 17-fold, respectively. The Km (NADPH) values of K109H and K109A increased over 3- and 6-fold, respectively, which also indicates the involvement of K109 in NADPH binding. Surprisingly, K109H and K109A mutations had almost no effect on Km (nitrofurazone) values, indicating that K109 might not be involved in the binding of nitrofurazone. It seems that residues involved in the binding of azo and nitro substrates are not identical. The Vmax/Km (nitrofurazone) values of K109H and K109A decreased over 3- and 13-fold, respectively. Whereas the Km (nitrofurazone) value of Y74W increased 5-fold, its Km (NADPH) value remained unaffected and Vmax value decreased 2.5-fold, thus resulting in over 12-fold decrease in its Vmax/Km (Methyl Red) value. Again, Y74W is not involved in the binding of NADPH while decreasing the binding of the electron

acceptor. H75N showed no detectable nitroreductase activity, suggesting that H75 is essential for AZRs nitroreductase activity. Asparagine residues were found in corresponding positions of amino acid sequences of azoreductases from S. cerevisiae and S. pombe (Fig. 2), however, H75N was not a conserved mutation for AZRs activity. In conclusion, we report here for the rst time that avin-dependent azoreductases with no sequence homology demonstrated similar three dimensional structures. Y74, H75 and K109 in the two conserved loops around the FMN cofactor of azoreductase AZR from R. sphaeroides are involved in the binding of substrates and crucial for enzyme activities.

References
Bellamacina CR (1996) The nicotinamide dinucleotide binding motif: a comparison of nucleotide binding proteins. FASEB J 10:12571269 Bottoms CA, Smith PE, Tanner JJ (2002) A structurally conserved water molecule in Rossmann fold dinucleotidebinding domains. Protein Sci 11:21252137 Chen H (2006) Recent advances in azo dye degrading enzyme research. Curr Protein Pept Sci 7:101111 Chen H, Hopper SL, Cerniglia CE (2005) Biochemical and molecular characterization of an azoreductase from Staphylococcus aureus, a tetrameric NADPH-dependent avoprotein. Microbiology 151:14331441 Chung KT (1983) The signicance of azo-reducion in the mutagenesis and carcinogenesis of azo dyes. Mutat Res 114:269281 DeLano WL (2003) PyMOL reference manual. DeLano Scientic LLC, San Carlos

123

Biotechnol Lett (2008) 30:869875 Dos Santos AB, Cervantes FJ, Van Lier JB (2007) Review paper on current technologies for decolourisation of textile wastewaters: perspectives for anaerobic biotechnology. Biores Technol 98:23692385 Freigang J, Diederichs K, Schafer KP, Welte W, Paul R (2002) Crystal structure of oxidized avodoxin, an essential protein in Helicobacter pylori. Protein Sci 11:253261 Hanauer SB (1996) Inammatory bowel disease. N Engl J Med 334:841848 Ito K, Nakanishi M, Lee WC, Sasaki H, Zenno S, Saigo K, Kitade Y, Tanokura M (2006) Three-dimensional structure of AzoR from Escherichia coli. An oxidereductase conserved in microorganisms. J Biol Chem 281:20567 20576 Jornvall H, Persson B, Krook M, Atrian S, Gonzalez-Duarte R, Jeffery J, Ghosh D (1995) Short-chain dehydrogenases/ reductases (SDR). Biochemistry 34:60036013 Kobori T, Sasaki H, Lee WC, Zenno S, Saigo K, Murphy ME, Tanokura M (2001) Structure and site-directed mutagenesis of a avoprotein from Escherichia coli that reduces nitrocompounds: alteration of pyridine nucleotide binding by a single amino acid substitution. J Biol Chem 276:28162823 Liger D, Graille M, Zhou CZ, Leulliot N, Quevillon-Cheruel S, Blondeau K, Janin J, van Tilbeurgh H (2004) Crystal structure and functional characterization of yeast YLR011wp, an enzyme with NAD(P)H-FMN and ferric iron reductase activities. J Biol Chem 279:3489034897 Liu G, Zhou J, Lv H, Xiang X, Wang J, Zhou M, Qv Y (2007a) Azoreductase from Rhodobacter sphaeroides AS1.1737 is a avodoxin that also functions as nitroreductase and avin mononucleotide reductase. Appl Microbiol Biotechnol 76:12711279 Liu ZJ, Chen H, Shaw N, Hopper SL, Chen L, Chen S, Cerniglia CE, Wang BC (2007b) Crystal structure of an aerobic FMN-dependent azoreductase (AzoA) from Enterococcus faecalis. Arch Biochem Biophys 463:6877

875 Nakanishi M, Yatome C, Ishida N, Kitade Y (2001) Putative ACP phosphodiesterase gene (acpD) encodes an azoreductase. J Biol Chem 276:4639446399 Nicholas KB, Nicholas HB Jr, Deereld DW II (1997) GeneDoc: analysis and visualization of genetic variation. EMBnet News 4:14 Raf F, Hall JD, Cerniglia CE (1997) Mutagenicity of azo dyes used in foods, drugs and cosmetics before and after reduction by Clostridium species from the human intestinal tract. Food Chem Toxicol 35:897901 Reyes VM, Sawaya MR, Brown KA, Kraut J (1995) Isomorphous crystal structures of Escherichia coli dihydrofolate reductase complexed with folate, 5-deazafolate, and 5,10-dideazatetrahydrofolate: mechanistic implications. Biochemistry 34:27102723 Shore J (1996) Advances in direct dyes. Ind J Fibre Text Res 21:129 Stolz A (2001) Basic and applied aspects in the microbial degradation of azo dyes. Appl Microbiol Biotechnol 56:6980 Suzuki Y, Yoda T, Ruhul A, Sugiura W (2001) Molecular cloning and characterization of the gene coding for azoreductase from Bacillus sp. OY1-2 isolated from soil. J Biol Chem 276:90599065 Wang CJ, Hagemeier C, Rahman N, Lowe E, Noble M, Coughtrie M, Sim E, Westwood I (2007) Molecular cloning, characterization and ligand-bound structure of an azoreductase from Pseudomonas aeruginosa. J Mol Biol 373:12131228 Wilson DK, Bohren KM, Gabbay KH, Quiocho FA (1992) An unlikely sugar substrate site in the 1.65 A structure of the human aldose reductase holoenzyme implicated in diabetic complications. Science 257:8184 Yan B, Zhou J, Wang J, Du C, Hou H, Song Z, Bao Y (2004) Expression and characteristics of the gene encoding azoreductase from Rhodobacter sphaeroides AS1.1737. FEMS Microbiol Lett 236:129136

123