Isolation of an Organic Solvent-TolerantLipolytic Enzyme from Uncultivated Microorganism

Changhyun Roh & Rolf D. Schmid

Received: 31 May 2013 /Accepted: 22 August 2013# Springer Science+Business Media New York 2013

Abstract Although the use of lipases as biocatalysts has frequently been proposed, it is yetscarcely being implemented in industrial processes. This is mainly due to the difficultiesassociated with the discovery and engineering of new enzymes and the lack of versatile screeningmethods. In this study, we screened the available strategy from a metagenomic pool for theorganic solvent-tolerant lipase with enhanced performance for industrial processes. A novellipase was identified through functional screening from a metagenomic library of activatedsludge in an Escherichia coli system. The gene encoding the lipase from the metagenomic pool,metalip1, was sequenced and cloned by PCR. Metalip1 encoding a polypeptide of 316 aminoacids had typical residues essential for lipase such as pentapeptide (GXSXGG) and catalytic triadsequences (Ser160, Asp260, and His291). The deduced amino acid sequence of metalip1 showedhigh similarity to a putative lipase from Pseudomonas sp. CL-61 (80 %, ABC25547). Metalip1was expressed in E. coli BL21 (DE3) with a his-tag and purified using a Ni-NTA chelatingcolumn and characterized. This enzyme showed high expression level and solubility in theheterologous E. coli host. This enzyme was active over broad organic solvents. Among organicsolvents examined, dimethyl formamide was the best organic solvent for metalip1. We showedthat function-based strategy is an effective method for fishing out an organic solvent-tolerantlipase from the metagenomic library. Also, it revealed high catalytic turnover rates, which makethem a very interesting candidate for industrial application.

Keywords Lipase . Screening . Purification . Organic-tolerant enzyme . Uncultivatedmicroorganism

Introduction

Screening for novel enzymes from cultivated microorganisms using conventional biomo-lecular techniques has limits in exploring the tremendous genetic diversity of environmental

Appl Biochem BiotechnolDOI 10.1007/s12010-013-0464-z

C. Roh : R. D. SchmidInstitute of Technical Biochemistry, University of Stuttgart, Allmandring 31, 70569 Stuttgart, Germany

C. Roh (*)Division of Biotechnology, Advanced Radiation Technology Institute, Korea Atomic Energy ResearchInstitute, 1266 Shinjeong-dong, Jeongeup-si, Jeolabuk-do, Republic of Koreae-mail: chroh@kaeri.re.kr

microorganisms because less than 1 % of microbes present in natural environments could becultivated [1, 4, 5, 15, 20, 21, 23]. Understanding the genome reservoirs of uncultivatedmicroorganisms should be of major interest for the screening of novel enzymes. Thecontinuous attempts to find novel enzymes have approached through activity-based strategyfrom a metagenomic pool constructed by extraction of genomic DNA directly from envi-ronmental sources [9–12, 14, 18]. Accordingly, the metagenomic approach for independentcultivation of microorganisms allows to fish for various biocatalysts such as esterases andlipases from the metagenomic pool [6, 9–12, 16, 19, 22].

Lipolytic enzymes such as esterases (EC 3.1.1.1) and lipases (EC 3.1.1.3) biotransformboth the fat hydrolysis and the synthesis of fatty acid esters including triglycerides andinteresterification of triglycerides as biocatalysts in synthetic organic chemistry of regio-specificity and enantio-selectivity [3, 7]. Lipolytic enzymes are pivotal biocatalysts forindustrial interest for production of fatty acids due to their useful properties such asremarkable stability in organic solvents and broad substrate specificity [8]. Lipases haveversatile potential applications in the detergents, food, textiles, cosmetics, pharmaceuticals,and pulp and paper industries [2, 17]. Because of industrial and scientific interest in lipolyticenzymes, many studies have been performed on lipolytic enzyme-catalyzed esterification inorganic media until now. With potential reason, the methodology for screening enzymestable in organic solvents has been actively investigated. A novel screening technology forthe industrial use of lipase may be overcome by metagenomics.

The goal of this study is to obtain a new lipolytic gene which can be identified by directdetection of fat hydrolysis from a metagenomic library. We showed that a library constructedfrom microorganisms of an uncultivable process can be screened for the isolation of fishingout an organic solvent-tolerant lipase.

Materials and Methods

Sampling

Activated sludge was collected from a sewage center at the Institute for SanitaryEngineering, Water Quality and Solid Waste Management of the University of Stuttgart,Germany.

Construction of Lipase Gene from Metagenomic Library

The procedure was carried out with a 0.5-g sample as wet weight and stored at −20 °C before use.DNA isolation method was based on a method described previously [18]. DNA samples fromactivated sludge were used to construct a genomic DNA library following partial digestion withSau3AI, optimized to maximize fragments in the 100-bp to 5-kb size range. Fragments of100 bp–5 kb by agarose gel were excised and concentrated. The Bam HI-digested pUC 18treated with phosphatase was overhung by using T4 DNA ligase (Roche Diagnostics, Germany)to the genomic DNA library digested with Sau3AI. Five microliters of vector DNA (10 μg) wasmixed with 80 μl of genomic DNA (1 μg); then, 5 μl (5U) of T4 DNA ligase and 10 μl of 10×ligase buffer were added, and the solution was incubated for 18 h at 16 °C. The DNA librarysolution was used to electrotransform Escherichia coli DH 5α (Novagen). The functional lipasescreening was performed by plating the transformed cells onto Luria–Bertani (LB) agar platescontaining 50 μg of ampicillin and 1 % tricaprylin emulsified, and was incubated at 37 °C.

Appl Biochem Biotechnol

Subcloning, Expression, and Purification of Recombinant Lipase

The gene was amplified by PCR with the primer set, sense: 5′-catcatatgatgaaccgcatccaggtcttg-3′, antisense: 5′-gctgaattctcaccatgcgggattgtgggctac-3′, containing restriction enzymesites of Nde I/Eco RI. PCR was run with following conditions on a Thermal Cycler:denaturation at 94 °C for 1 min, annealing at 62 °C for 30 s, and an extension step at72 °C for 1 min 30 s. The sequence was repeated 35 times followed by a 7-min finalextension step at 72 °C. The PCR product was digested with Nde I/Eco RI and then ligatedinto Nde I/Eco RI-digested expression vector pET 28a+ (Novagen, Madison, WI) andtransformed into E. coli BL21 (DE3) (Stratagene, La Jolla, CA). The E. coli transformedwith this plasmid was plated on LB agar containing 50 μg/ml kanamycin and 1 %tricaprylin. The transformant was grown in a 250-ml flask containing 50 ml LB mediumsupplemented with 50 μg/ml of kanamycin at 37 °C until the cell concentration reached anOD600nm of 0.6 and isopropyl-thio-β-D-galactopyranoside (IPTG) of a final concentrationof 0.1 mM, followed by additional growing overnight at 25 °C with shaking at 180×g. Cellswere harvested by centrifugation at 3,000×g for 30 min at 4 °C and resuspended in100 mmol/l potassium phosphate buffer (pH 7.5) containing 1 mmol/l phenylmethylsulfonylfluoride (PMSF). Cells were lysed by Sonicator (W250 Sonifier, Branson, Dietzenbach,Germany). The cell debris was removed by centrifugation at 8,000×g for 30 min. Thesupernatant was collected, and the recombinant metalip1 was purified with a Ni-nitrilotriacetic acid (Ni-NTA) affinity chromatography column (Qiagen, Germany).The supernatant was equilibrated with buffer A (10 mmol/l Tris–HCl, 500 mmol/lNaCl, 5 mmol/l imidazole, pH 8.0). The bound protein was eluted with buffer B(10 mmol/l Tris–HCl, 500 mmol/l NaCl, 500 mmol/l imidazole, pH 8.0) at 4 °C. Thepurity of the purified protein was estimated by sodium dodecyl sulfate polyacrylamidegel electrophoresis (SDS-PAGE) in the eluted fractions, using 12 % polyacrylamiderunning gels [13]. The concentration of the purified protein was determined using theBradford assay kit (Bio-Rad, Hercules, CA).

DNA Sequencing

The DNA sequencing reaction was carried out on both strands of double-stranded templatesusing the BigDye Terminator Cycle Sequencing Kit RR-100 (Applied Biosystems, Weiterstadt,Germany). The sequencing product was analyzed on a ABI PrismTM 377 DNA Sequencer(PerkinElmer, Shelton, USA).

Characterization of the Recombinant Enzyme

Enzymatic activity for p-nitrophenyl esters (pNPEs) as substrates was examined by measuringthe amount of p-nitrophenyl released by enzyme-based hydrolysis. The production ofp-nitrophenyl was continuously monitored at 405 nm by use of an Ultraspec 3000 UV/visspectrometer (Amersham Biosciences, Sweden). The reaction mixture consisted of 10 μl of10 mM substrate in acetonitrile, 40 μl of ethanol, and 950 μl of 50 mmol/l Tris–HCl buffer(pH 7.5) containing quantified amount of the enzyme. The enzyme reaction was performed at28 °C. The amount of p-nitrophenol liberated during the reaction was measured by absorbanceat 405 nm. One unit of enzyme activity was defined as the amount of activity required to release1 μmol of p-nitrophenyl per minute at 28 °C from p-nitrophenyl ester. Under the conditionsdescribed, the extinction coefficient of p-nitrophenol was ε=1.85 mmol−1 mm−1.

Appl Biochem Biotechnol

Effects of pH, Temperature, and Organic Solvents

For optimization of pH of the enzyme, lipase activity was measured for a pH range of 4.5 to11. The used buffers were 50 mmol/l sodium acetate (pH 4.5, 5.0, and 5.5), 50 mmol/l MES(pH 6.5), 50 mmol/l Tris–HCl (pH 7.5 and 8.5), 50 mmol/l glycine–NaOH (pH 9.5 and10.0), and 50 mmol/l Na2HPO4–NaOH (pH 11.0). The enzymatic activity is represented as apercentage of the maximum activity. The optimum temperature for enzyme activity wasdetermined by assaying lipase activities toward p-nitrophenyl butyrate in a Tris–HCl buffer(pH 7.5) for a temperature range of 10 to 100 °C. The effect of detergents on the metalip1was analyzed by addition of 1 % (w/v) detergent in Tris–HCl buffer (pH 7.5) to the enzymesolution. The measurements of enzymatic activity were carried out immediately after 1 h ofincubation at 28 °C. Enzymatic activity in organic solvents was measured by addition of 1 %(v/v) organic solvents in 50 mmol/l Tris–HCl buffer (pH 7.5), containing acetone, butanol,chloroform, cyclohexane, diethylether, dimethyl formamide, dimethyl sulfoxide, 1,2-dichloroethane, ethylacetate, n-hexane, isooctane, methylene chloride, methanol,propanol, and toluene. The enzymatic activity for organic solvents was examined usingp-nitrophenyl butyrate as a substrate at 28 °C.

Results and Discussion

Construction of Metagenomic Library and Characterization of Recombinant Lipase

A novel lipase of metalip1 on activity-based strategy was screened. The genomic DNA forconstruction of the metagenomic library of activated sludge was extracted and purified. Thefunctional lipase gene was cloned into standard cloning vector such as pUC. Thetransformant with a clear halo around an individual colony was chosen as a possible clone,which indicated hydrolysis of the tricaprylin. The positive lipase clone was detected with anincidence of 1/130,000 on the 1 % tricaprylin plate assay (Fig. 1). The lipase gene consistedof 948 bp by encoding a polypeptide of 316 amino acids. Amino acid sequence analysis toidentify open reading frames (ORFs) and subsequent BLASTP analysis for lipase gene ORFrevealed one potential lipase gene containing the typical conserved sequence motif. Basedon amino acid sequence analysis of lipase, most of the lipases known in microbial genomedatabases have an oxyanion hole sequence (HGG) and a consensus sequence of a catalytictriad (Ser, Asp, His). Most of the lipases could be divided into two evolutionarily distinctclasses on the basis of the codon required for the conserved site, serine. This active serineresidue is located within a semiconserved pentapeptide, either Gly(Ala)-X-Ser-X-Gly-Gly(where X means any amino acid). The identification of metalip1 from the metagenomiclibrary as a lipase is provided by the presence of the pentapeptide, Gly-Val-Ser-Val-Gly-Gly.The deduced amino acid sequence of metalip1 was used to perform a BLAST search underthe GenBank database of the National Center for Biotechnology Information (NCBI). Thisstudy showed relatively high identity with other known lipases at the amino acid level. Thesequences were between 64 and 80 % identical to previously known lipases: 80 % identity atlipase (accession no. ABC25547) from Pseudomonas sp. CL-61, 69 % identity atesterase/lipase/thioesterase (accession no. YP_349726) from Pseudomonas fluorescensPf0-1, 69 % identity at esterase/lipase (accession no. ZP_04946738) from Burkholderiadolosa AUO158, 67 % identity at alpha/beta hydrolase domain-containing protein (acces-sion no. YP_001861722) from Burkholderia phymatum STM815, 66 % identity atalpha/beta hydrolase domain-containing protein (accession no. NP_745942) from

Appl Biochem Biotechnol

Fig. 2 a Comparison of the deduced amino acid sequences from multiple sequence alignment of lipolyticproteins containing the conserved motifs of HGGG, GXSXGG, and putative catalytic triad residues of thelipolytic enzymes from the metagenomic library (metalip1), Pseudomonas sp. CL-61 (80 %, ABC25547.1), P.fluorescens Pf0-1 (69 %, YP_349726), B. dolosa AUO158 (69 %, ZP_04946738), B. phymatum STM815(67 %, YP_001861722), P. putida KT2440 (66 %, NP_745942), and S. coelicolor A3(2) (64 %, NP_631192).The conserved motifs of HGG, GXSXGG, and the putative triad catalytic residues composed of Ser160,Asp260, and His291 of lipases/esterases are indicated with a box. The sequence data reported here have beendeposited in the GenBank database and assigned the accession no. ADG03645. The multiple sequencealignments were performed in ClustalX1.83. The identical and conserved residues are marked by asterisks,dots, and colons, respectively

Fig. 1 Tricaprylin-emulsifyingplate (TCN plate) screening forlipase activity. A lipase screenedin 1 % tricaprylin agar plate

Appl Biochem Biotechnol

Pseudomonas putida F1, and 64 % identity at lipase (accession no. NP_631192) fromStreptomyces coelicolor A3(2). The primary sequence analysis of metalip1 indicates highsimilarity to lipase protein known on databases; metalip1 showed typical lipase conservedmotif sequence rule (Fig. 2). The amino acid sequence of lipase from the metagenomiclibrary shows a redundant identity with that of Pseudomonas sp., which means metalip1 is amember of the Pseudomonas sp. group.

To characterize the lipase protein of metalip1, we performed cloning of the gene bysubcloning with Nde I and Eco RI of restriction enzyme site into a T7 RNA polymerase-derived E. coli expression vector of pET28a+ (Novagen, Madison, WI, USA). We inducedthe protein expression level of the metalip1 with 0.5 mmol/l IPTG in N-terminal his-tagsystem and purified the protein by Ni-NTA chromatography. Its protein size was observed atapproximately 35 kDa in a Coomassie-stained SDS-PAGE (Fig. 3). This lipase showed highexpression level and solubility in heterologous E. coli host.

Substrate Specificity of Recombinant Lipase

The lipase obtained from the metagenomic library hydrolyzed broad substrates containingthe acyl group with chain lengths between C2 and C16. p-Nitrophenyl butyrate as substrate

Table 1 Substrate specificity and kinetic parameters of the purified metalip1 enzyme

Substrate Specificactivitya

(U/mg)

Km(mmol)

Kcat (s−1) Kcat/Km(s−1 mmol−1)

Relativeactivity(%)

p-Nitrophenyl acetate (C2) 98.5 0.24 27.5 114.6 66

p-Nitrophenyl propionate (C3) 102.7 0.21 25.4 121.0 69

p-Nitrophenyl butyrate (C4) 148.5 0.18 30.5 169.4 100

p-Nitrophenyl hexanoate (C6) 143.2 0.22 34.8 158.2 96

p-Nitrophenyl caprylate (C8) 92.3 0.27 28.7 106.3 62

p-Nitrophenyl caprate (C10) 81.7 0.34 27.8 81.8 55

p-Nitrophenyl laurate (C12) 99.4 0.22 25.7 116.8 67

p-Nitrophenyl pentadecanoate (C15) 105.5 0.23 28.6 124.3 71

p-Nitrophenyl palmitate (C16) 91.7 0.25 25.7 102.8 62

a Specific activity (in micromoles per minute per milligram) was measured using 1 μg of the purified metalip1

Fig. 3 SDS-PAGE gel (12 %)showing the expression and puri-fication of recombinant metalip1protein. M protein marker, 1 be-fore induction of metalip1, 2 totalform of metalip1, 3 soluble formof metalip1, 4 purified form ofmetalip1

Appl Biochem Biotechnol

showed the highest activity (C4 acyl group) among the p-nitrophenyl esters examined. InTable 1, the substrate specificity and kinetic parameters of recombinant lipase were repre-sented. The relative activity of lipase for the C2 to C6 groups was between 66 and 100 %,and for C8 to C16, between 55 and 71 %. Interestingly, lipase activity was 71 and 62 % forthe longer chain substrates, C15 and C16.

Effects of Organic Solvents on a Metalip1 Lipase Activity

Metalip1 was examined for activity on various organic solvents including acetone, buta-nol, chloroform, cyclohexane, diethylether, dimethyl formamide, dimethyl sulfoxide, 1,2-dichloroethane, ethylacetate, n-hexane, isooctane, methylene chloride, methanol,propanol, and toluene. The effects of various organic solvents on metalip1 are summarizedin Table 2. In general, we observed a significant increase in metalip1 activity with the

Table 2 The effects of organicsolvents on metalip1 enzymeactivity

aActivities of metalip1 (finalconc., 0.1 mg ml−1) were mea-sured spectrophotometrically in20 mmol/l Tris–HCl (pH 7.5)buffer containing each organicsolvent at 28 °C for 1 h: Resid-ual activity was determined with0.5 mmol/l p-nitrophenyl buty-rate in 20 mmol/l Tris–HCl (pH7.5) buffer at 28 °C andexpressed as the percent of thereference value which meanswith no addition of organic sol-vent. The average values oftriplicate measurements werecarried out

Organic solvent Relative activitya at a concentrationof 20 % (80:20, v/v)

Acetone 148±5

Butanol 185±3

Chloroform 138±7

Cyclohexane 185±2

Diethylether 232±8

Dimethyl formamide 258±5

Dimethyl sulfoxide 210±3

1,2-Dichloroethane 165±7

Ethylacetate 135±6

n-Hexane 130±3

Isooctane 159±4

Methylene chloride 130±2

Methanol 195±4

Propanol 184±6

Toluene 140±5

Fig. 4 Effect of pH on the purified metalip1 activity. a The enzymatic assay was performed in various pHbuffers at 28 °C for 1 h using 10 mM p-nitrophenyl butyrate as a substrate (closed circles). b For the residualstability activity, the purified metalip1 was assayed after incubation at various pH levels for 24 h (open circles)

Appl Biochem Biotechnol

addition of organic solvents. The presence of dimethyl formamide increased the highestactivity among organic solvents examined, which means approximately twofold increasesin activity. Dimethyl formamide and dimethyl sulfoxide showed the relatively highactivity of 258 and 210 %. To examine the effects of organic solvents on metalip1 activity,enzymatic activity was performed at 1-h intervals in various organic solutions. Theenzyme was found to be most active in dimethyl formamide and to have the lowestactivity in n-hexane. With this result, we could postulate that metalip1 is a powerfulcandidate for organic solvent-tolerant lipase in industrial application.

Effects of pH and Temperature on a Metalip1 Lipase

In Fig. 4a, we examined the activity of metalip1 at different pH conditions (pH4.5–11.0)using p-nitrophenyl butyrate as a substrate. The optimal range of pH for lipase activity ofmetalip1 was at pH 6.5–9.5. The metalip1 enzyme showed a high level of activity at pH 8.5.But, this enzyme showed low activity at low range than pH 5.5 or at pH 10.0 above.Specifically, metalip1 showed over 50 % of its maximal value in the pH range of 5.0–10.0.For residual stability of pH, metalip1 showed relative activity over 60 % in the pH range of6.5–10.0 (Fig. 4b). In Fig. 5a, the optimal range of temperature for lipase activity of metalip1was at 10–60 °C. Above 60 °C, its activity was decreased. The metalip1 enzyme showed thehighest activity at 40 °C. The effect of residual thermostability analysis exhibited thatmetalip1 was stable at 40 °C and inactivated relatively at 10 °C (Fig. 5b).

Conclusions

In this study, we showed that activity-based strategy is one approach method for fishing out afunctional lipase from the metagenomic library. The metalip1 lipase from the metagenomiclibrary was able to hydrolyze tricaprylin and p-nitrophenol esters with chain length speci-ficity, and the optimal activity of this enzyme is at 40 °C and pH 8.5. Moreover, metalip1enzyme was an organic solvent-tolerant lipase derived from the metagenomic library ofactivated sludge microorganisms. The result suggests that isolation of novel genes from

Fig. 5 Effect of temperature and thermostability of the purified metalip1. aOptimized temperature of the purifiedmetalip1. Relative activity of p-nitrophenyl butyrate hydrolysis at different temperatures by purified metalip1. bThermostability of the purified metalip1. Residual activity of the metalip1 after incubation at different tempera-tures. 10 °C (closed circles), 20 °C (open circles), 30 °C (closed inverted triangles), 40 °C (open invertedtriangles), 50 °C (closed diamonds), and 60 °C (open diamonds) toward p-nitrophenyl butyrate at 28 °C

Appl Biochem Biotechnol

metagenome could represent an interesting and useful reservoir for biotechnological appli-cations. However, the general limitation for fishing out a novel soluble protein with anincidence of 1/130,000 from the metagenomic library is not expressed as active form inheterologous host such as E. coli. One of such problems has usually been caused by thedifference in codon usage between the expressed gene and the expression host. To overcomethis bottleneck and extend the range of functional screening strategy, Streptomyces sp. andPseudomonas sp. could be used as alternative hosts.

Acknowledgments We are grateful to Prof. Dr. Karl-Heinrich Engesser and Dr. Manfred Roth in the Institutefor Sanitary Engineering, Water Quality and Solid Waste Management of the University of Stuttgart for thegenerous cooperation. This research was supported by the Basic Science Research Program through theNational Research Foundation of Korea (NRF), which is funded by the Ministry of Education, Science andTechnology (grant no. 2011-0010634).

References

1. Amann, R. I., Ludwig, W., & Schleifer, K. H. (1995). Microbiological Reviews, 59, 143–169.2. Bajpai, P. (1999). Biotechnology Progress, 15, 147–157.3. Drepper, T., Eggert, T., Hummel, W., Leggewie, C., Pohl, M., Rosenau, F., Wilhelm, S., & Jaeger, K. E.

(2006). Biotechnology Journal, 1, 777–786.4. Gitig, D. (2010). Biotechniques, 48, 361–365.5. Handelsman, J., Rondon, M. R., Brady, S. F., Clardy, J., & Goodman, R. M. (1998). Chemistry and

Biology, 5, R245–249.6. Hardeman, F., & Sjoling, S. (2007). FEMS Microbiology Ecology, 59, 524–34.7. Jaeger, K. E., & Eggert, T. (2002). Current Opinion in Biotechnology, 13, 390–397.8. Jaeger, K. E., Dijkstra, B. W., & Reetz, M. T. (1999). Annual Review of Microbiology, 53, 315–351.9. Jeon, J. H., Kim, J. T., Kim, Y. J., Kim, H. K., Lee, H. S., Kang, S. G., Kim, S. J., & Lee, J. H. (2009).

Applied Microbiology and Biotechnology, 81, 865–874.10. Kim, E. Y., Oh, K. H., Lee, M. H., Kang, C. H., Oh, T. K., & Yoon, J. H. (2009). Applied and

Environmental Microbiology, 75, 257–260.11. Kim, Y. H., Kwon, E. J., Kim, S. K., Jeong, Y. S., Kim, J., Yun, H. D., & Kim, H. (2010). Biochemical

and Biophysical Research Communications, 393, 45–49.12. Kim, Y. J., Choi, G. S., Kim, S. B., Yoon, G. S., Kim, Y. S., & Ryu, Y. W. (2006). Protein Expression and

Purification, 45, 315–323.13. Laemmli, U. K. (1970). Nature, 227, 680–685.14. Lee, S. W., Won, K., Lim, H. K., Kim, J. C., Choi, G. J., & Cho, K. Y. (2004). Applied Microbiology and

Biotechnology, 65, 720–726.15. Morgan, J. L., Darling, A. E., & Eisen, J. A. (2010). PLoS One, 5, e10209.16. Park, H. J., Jeon, J. H., Kang, S. G., Lee, J. H., Lee, S. A., & Kim, H. K. (2007). Protein Expression and

Purification, 52, 340–347.17. Reetz, M. T. (2002). Current Opinion Chemical Biology, 6, 145–150.18. Roh, C., Villatte, F., Kim, B. G., & Schmid, R. D. (2005). Electrophoresis, 26, 3055–3061.19. Roh, C., & Villatte, F. (2008). Journal of Applied Microbiology, 105, 116–123.20. Torsvik, V., & Ovreas, L. (2002). Current Opinion in Microbiology, 5, 240–245.21. Turnbaugh, P. J., & Gordon, J. I. (2008). Cell, 134, 708–713.22. Zhang, T., Han, W. J., & Liu, Z. P. (2009). Microbial Cell Factory, 8, 67.23. Zhao, L. (2010). Nature, 465, 879–880.

Appl Biochem Biotechnol