Trigonopsis variabilis D-Amino Acid Oxidase: Control of Protein Quality and Opportunities for Biocatalysis through Production in Escherichia coli

Share Embed


Descrição do Produto

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 2007, p. 331–333 0099-2240/07/$08.00⫹0 doi:10.1128/AEM.01569-06 Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Vol. 73, No. 1

Trigonopsis variabilis D-Amino Acid Oxidase: Control of Protein Quality and Opportunities for Biocatalysis through Production in Escherichia coli䌤† Iskandar Dib, Damir Stanzer, and Bernd Nidetzky* Research Centre Applied Biocatalysis, Petersgasse 14, A-8010 Graz, Austria, c/o Institute of Biotechnology and Biochemical Engineering, Graz University of Technology, Petersgasse 12, A-8010 Graz, Austria Received 7 July 2006/Accepted 10 October 2006

established (17) and was therefore reexamined in this study, in which we focused on strain physiology and protein quality, as well as on the development of new oxidase preparations based on the recombinant enzyme. Expression vectors bearing the wild-type TvDAO gene and two novel genes encoding chimeric enzymes in which a 12amino-acid peptide (Strep-Tag II) was fused in frame to the N terminus and the C terminus of TvDAO were constructed using standard methods (15). Further details and information about the oligonucleotide primers used are summarized in Table S.1 in the supplemental material. After sequencing, the final constructs pTvDAO, pTvDAOstrepN, and pTvDAOstrepC were transformed into E. coli BL21(DE3). Selected positive clones were grown at 37°C in Luria-Bertani medium containing 50 mg/liter kanamycin and, after switching to 25°C at an optical density of about 0.8, were induced with 0.5 mM isopropyl-␤-D-thiogalactopyranoside (IPTG) for 18 h (see Fig. S.1 in the supplemental material for details concerning enzyme production). For the strain bearing the pTvDAOstrepC construct no significant levels of activity were observed; analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that this enzyme variant (TvDAOstrepC) was almost completely associated with the insoluble cell fraction (data not shown). Interestingly, the volumetric activity observed for the N-terminally tagged enzyme (TvDAOstrepN) was up to 1.5fold greater than that of the wild type. We also examined how different process parameters affect the production of TvDAOstrepN. An IPTG concentration in the range from 20 to 500 ␮M was not a significant factor. Increasing the temperature to 30°C did not affect the specific activity but enhanced the volumetric enzyme activity by 22% due to its effect on cell growth. A possible dependence of oxidase production on the oxygen transfer rate (OTR) (1) was investigated by varying the working volume in the 300-ml shaken flask from 50 to 200 ml, because OTR decreases as the volume of the medium increases. Figure 1 shows that a decrease in the OTR resulted in a significant increase (almost 50%) in specific oxidase activity. By contrast, the volumetric activity of the enzyme decreased with relatively low levels of

D-Amino acid oxidases (DAOs) (EC 1.4.3.3.) are well-studied and technologically useful flavoenzymes that catalyze the O2-dependent transformation of an ␣-amino acid substrate into the corresponding ␣-keto acid, H2O2, and NH3 (for reviews, see references 11, 12, and 17). Economically, the most important use of DAO is the industrial conversion of cephalosporin C into glutaryl-7-aminocephalosporanic acid (2, 10, 14). DAO from the yeast Trigonopsis variabilis (TvDAO) has been a prime candidate for use in commercial applications because it displays good activity with cephalosporin C and is reasonably resistant to the oxidants (O2, H2O2) occurring in the process (13). Current industrial enzyme production integrates relatively slow, multistage bioreactor cultivation of T. variabilis and several steps of downstream processing to obtain an immobilized biocatalyst deficient in catalase and esterase activities (8, 14). The technical-grade oxidase obtained in this way is structurally microheterogeneous due to the partial oxidation of Cys108 into a cysteine sulfinic acid (16). Oxidatively modified TvDAO is catalytically only about one-quarter as efficient as the native enzyme (16). Production of a recombinant TvDAO potentially could eliminate many shortcomings of the established process, especially in a prokaryotic host with a reducing cellular environment that should prevent cysteine oxidation in TvDAO. Escherichia coli thus seems to be a logical candidate, but toxicity of the enzyme for this bacterium may have hampered efficient heterologous expression so far. TvDAO produced in E. coli has often been precipitated in inclusion bodies (3, 6, 7, 17), or the soluble portion of it has displayed much lower specific activity than the natural enzyme, likely because a large fraction of the protein was recovered in the inactive apo form (1, 9). Thus, recombinant production of this important industrial biocatalyst in E. coli has not been

* Corresponding author. Mailing address: Institute of Biotechnology and Biochemical Engineering, Graz University of Technology, Petersgasse 12/I, A-8010 Graz, Austria. Phone: 43-316-873-8400. Fax: 43316-873-8434. E-mail: [email protected]. † Supplemental material for this article may be found at http://aem .asm.org/. 䌤 Published ahead of print on 20 October 2006. 331

Downloaded from http://aem.asm.org/ on July 12, 2015 by guest

Trigonopsis variabilis D-amino acid oxidase accounts for 35% of Escherichia coli protein when added Dmethionine suppresses the toxic activity of the recombinant product. Permeabilized E. coli cells are reusable and stabilized enzyme preparations. The purified oxidase lacks the microheterogeneity of the natural enzyme. Oriented immobilization of a chimeric oxidase maintains 80% of the original activity in microparticle-bound enzymes.

332

DIB ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 2. Comparison of protein elution profiles in MonoQ chromatography for purified TvDAOstrepN (solid line) and a technical-grade preparation of TvDAO (dashed line). Samples (0.4 and 3 mg of protein, respectively) were applied and eluted as described previously (16). The identities of TvDAO peaks are as follows: peak 1, aggregated; peak 2, native (Cys108-SH); peak 3, oxidized (Cys108-SO2⫺); and peak 4, fragmented enzyme form. The dotted line indicates elution by isocratic flow steps. (Inset) Chromatography of soluble and insoluble cell fractions during production of TvDAOstrepN (lanes a and b, respectively) and purified enzyme (lane c). Lane d contained the molecular mass standard (molecular masses [in kDa] are indicated on the right). The probable position of the TvDAO protein band (39 kDa) is indicated by arrows. AU, absorbance units.

aeration, reflecting the reduced cell growth under these conditions. In an ideal case, however, production of recombinant TvDAO would combine high specific and volumetric yields of the enzyme. Interestingly, therefore, supplementation of the medium with D-Ala or D-Met markedly enhanced the specific and volumetric activities at the same time (Fig. 1). This result supports the notion that DAO activity is toxic in E. coli, likely because it depletes D-amino acids that are essential components of cell wall biosynthesis. D-Met was more effective than D-Ala in promoting enzyme overproduction. Induced cells grown in the presence of 10 mM D-Met contained TvDAOstrepN having a specific activity of 58 U/mg. A comparison of specific activities before and after purification revealed that functional recombinant oxidase accounted for approximately 35% of total soluble E. coli protein under these conditions. TvDAOstrepN was purified to apparent homogeneity (Fig. 2, inset) by affinity chromatography using Strep-Tactin Superflow cartridges (IBA GmbH, Go ¨ttingen, Germany) according to the protocol of the manufacturer. Oxidase activity was recovered with a 70% yield. The specific activity of TvDAOstrepN was 160 U/mg, as assayed by measurement of enzymatic oxygen consumption rates with a fiber optic microsensor (with D-Met as the substrate [18]). Native oxidase isolated from T. variabilis (16) had the same specific activity. Exogenous flavin adenine dinucleotide (FAD) (10 to 1,000 ␮M) added to assay or protein storage buffers did not enhance the catalytic activity of native and recombinant TvDAO. The absorbance spectra of native and recombinant enzymes were superimposable in the range from 250 to 700 nm, and the ratios of the intensities of the absorbance bands at 280 nm (protein) and 450 nm (FAD) were also the same (⬃9.5). These results suggest that the FAD

content of TvDAOstrepN was identical to that of the native oxidase. Kinetic evaluation of TvDAOstrepN revealed that the apparent catalytic efficiencies for oxidation of D-Ala (3.77 mM⫺1 s⫺1; measured using the coupled enzymatic assay with dioxygen as a cosubstrate [18]) and reduction of dioxygen (152 mM⫺1 s⫺1; measured with 20 mM D-Met as the substrate [16]) were identical within the limits of error (⫾15%) to the corresponding kinetic constants of the native enzyme. Likewise, the thermal stabilities of the two oxidases at 50°C, determined as described previously (4), were indistinguishable one from another. Purified TvDAOstrepN eluted from a MonoQ column as a single protein peak containing all of the applied protein and enzyme activity, as shown in Fig. 2, which compares the protein elution profile of the recombinant enzyme with the corresponding protein elution profile of a microheterogeneous preparation of natural TvDAO (16). Colorimetric titration with 5,5⬘-dithiobis(2-nitrobenzoic acid) (performed as described in reference 16) revealed that all six cysteines in the subunit of the recombinant enzyme were present in a reduced state (data not shown). In summary, these results demonstrate for the first time that TvDAO can be produced in E. coli as a fully functional recombinant protein. They also show that microheterogeneity due to modification of Cys108 was avoided completely during production and purification. Two novel enzyme preparations were derived from strain E. coli BL21(DE3)(pTvDAOstrepN). Mildly permeabilized cells in which TvDAOstrepN was physically entrapped were employed for the batchwise conversion of D-Ala (which was used as a model reaction) in an aerated enzyme reactor (see the supplemental material for details). The cells could be reused for at least three rounds of reaction with only small losses of

Downloaded from http://aem.asm.org/ on July 12, 2015 by guest

FIG. 1. Effect of culture conditions on production of recombinant TvDAOstrepN in E. coli. Expression was obtained in 300-ml flasks containing 50, 100, and 200 ml of broth. D-Alanine and D-methionine were added to a 100-ml culture at two concentrations, as indicated. The results are expressed relative to the results for a 100-ml culture (“standard culture”) without additives. The reference values for specific (solid bars) and volumetric (open bars) activities obtained for the standard culture were 14.5 U/mg and 12,340 U/liter of culture. In all experiments we used 0.5 mM IPTG and 18 h of induction. Note that growth kinetics and biomass yield were not affected by the addition of the amino acids, and the wet cell weight was 9 to 10 g/liter after 18 h for all experiments using a 100-ml culture.

VOL. 73, 2007

RECOMBINANT

REFERENCES 1. Alonso, J., J. L. Barredo, P. Armise´n, B. Dı´ez, F. Salto, J. M. Guisa ´n, J. L. Garcı´a, and E. Corte´s. 1999. Engineering the D-amino acid oxidase from

2.

3.

4.

5.

6.

7.

8. 9.

10.

11. 12. 13.

14. 15.

16.

17.

18.

ACID OXIDASE FROM T. VARIABILIS

333

Trigonopsis variabilis to facilitate its overproduction in Escherichia coli and its downstream processing by tailor-made metal chelate supports. Enzyme Microb. Technol. 25:88–95. Buchholz, K., V. Kasche, and U. T. Bornscheuer. 2005. Case study 1: the enzymatic production of 7-ACA from Cephalosporin C, p. 381–392. In Biocatalysts and enzyme technology. Wiley-VCH, Weinheim, Germany. Chien, L. J., J. M. Wu, I. C. Kuan, and C. K. Lee. 2004. Coexpression of Vitreoscilla hemoglobin reduces the toxic effect of expression of D-amino acid oxidase in E. coli. Biotechnol. Prog. 20:1359–1365. Dib, I., A. Slavica, W. Riethorst, and B. Nidetzky. 2006. Thermal inactivation of D-amino acid oxidase from Trigonopsis variabilis occurs via three parallel paths of irreversible denaturation. Biotechnol. Bioeng. 94:645–654. Fantinato, S., L. Pollegioni, and M. S. Pilone. 2001. Engineering, expression and purification of a His-tagged chimeric D-amino acid oxidase from Rhodotorula gracilis. Enzyme Microb. Technol. 29:407–412. Hwang, T. S., H. M. Fu, L. L. Lin, and W. H. Hsu. 2000. High-level expression of Trigonopsis variabilis D-amino acid oxidase in Escherichia coli using lactose as inducer. Biotechnol. Lett. 22:655–658. Ju, S. S., L. L. Lin, W. C. Wang, and W. H. Hsu. 1998. A conserved aspartate is essential for FAD binding and catalysis in the D-amino acid oxidase from Trigonopsis variabilis. FEBS Lett. 436:119–122. Liese, A., K. Seelbach, and C. Wandrey (ed.). 2006. Industrial biotransformations. Wiley-VCH, Weinheim, Germany. Lin, L. L., H. R. Chien, W. C. Wang, T. S. Hwang, H. M. Fu, and W. H. Hsu. 2000. Expression of Trigonopsis variabilis D-amino acid oxidase gene in Escherichia coli and characterization of its inactive mutants. Enzyme Microb. Technol. 27:482–491. Parmar, A., H. Kumar, S. S. Marwaha, and J. F. Kennedy. 1998. Recent trends in enzymatic conversion of Cephalosporin C to 7-aminocephalosporanic acid (7-ACA). Crit. Rev. Biotechnol. 18:1–12. Pilone, M. S. 2000. D-Amino acid oxidase: new findings. Cell. Mol. Life Sci. 57:1732–1747. Pilone, M. S., and L. Pollegioni. 2002. D-Amino acid oxidase as an industrial biocatalyst. Biocatal. Biotransform. 20:145–159. Pollegioni, L., L. Caldinelli, G. Molla, S. Sacchi, and M. S. Pilone. 2004. Catalytic properties of D-amino acid oxidase in Cephalosporin C bioconversion: a comparison between proteins from different sources. Biotechnol. Prog. 20:467–473. Riethorst, W., and A. Reichert. 1999. An industrial view on enzymes for the cleavage of Cephalosporin C. Chimia 53:600–607. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Slavica, A., I. Dib, and B. Nidetzky. 2005. Single-site oxidation, cysteine-108 to cysteine sulfinic acid, in D-amino acid oxidase from Trigonopsis variabilis and its structural and functional consequences. Appl. Environ. Microbiol. 71:8061–8068. Tishkov, V. I., and S. V. Khoronenkova. 2005. D-Amino acid oxidase: structure, catalytic mechanism, and practical application. Biochemistry (Moscow) 70:40–54. Trampitsch, C., A. Slavica, W. Riethorst, and B. Nidetzky. 2005. Reaction of Trigonopsis variabilis D-amino acid oxidase with 2,6-dichloroindophenol: kinetic characterisation and development of an oxygen-independent assay of the enzyme activity. J. Mol. Catal. B Enzym. 32:271–278.

Downloaded from http://aem.asm.org/ on July 12, 2015 by guest

activity (see Fig. S.2 in the supplemental material). The free enzyme, however, was rapidly inactivated under the same reaction conditions (half-life ⫽ 1.2 h). Entrapment of the recombinant oxidase in E. coli thus provided a markedly stable enzyme preparation that was obtained with only minimal downstream processing. Comparison of entrapped TvDAOstrepN with immobilized oxidase (14) in terms of operational stability was beyond the scope of this study. Using noncovalent attachment of TvDAOstrepN to StrepTactin MacroPrep particles (IBA GmbH), immobilization of the oxidase in an oriented manner was achieved for the first time. Alonso et al. (1) stated, however, without reporting the data, that N-terminally His-tagged TvDAO retained catalytic activity when it was bound to a cobalt-loaded metal-chelating resin. With a binding efficiency of 80% at low protein-to-carrier loads up to 4 mg/g, oriented immobilization of TvDAOstrepN clearly outperformed methods commonly used for random immobilization, such as covalent attachment to epoxyactivated resins (EC-EP standard grade Sepabeads; average binding efficiency, ⱕ50% [I. Dib and B. Nidetzky, unpublished results]), with regard to retention of the original enzyme activity in the immobilized oxidase. Because untagged proteins had negligible binding affinity (data not shown), TvDAOstrepN could be immobilized directly from the crude E. coli cell extract. In contrast to immobilization of His-tagged DAO from Rhodotorula gracilis on a nickel-chelate matrix (5), the noncovalent interaction between TvDAOstrepN and StrepTactin was exceptionally stable: no loss of activity to the supernatant occurred during up to 20 washes of the particles with buffer. Soluble and carrier-bound oxidases had identical activation energies (about 37.5 kJ/mol) for the O2-dependent oxidative deamination of D-Met (see Fig. S.3 in the supplemental material). The catalytic efficiency for the reaction with dioxygen (74 mM⫺1 s⫺1) was reduced only by one-half compared to that of the free enzyme and was clearly higher than that of the randomly immobilized enzyme (24 mM⫺1 s⫺1). These results especially validate oriented immobilization of NstrepTvDAO for all applications in which the native-like performance of the enzyme is crucial.

D-AMINO

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.