Construction of a mutant library of horseradish peroxidase gene by directed evolution and development of an in situ screening method

Abstract

A process of directed evolution applied to obtain a library of mutants of horseradish peroxidase (HRP) enzyme is described. We have introduced slight variations into the original DNA shuffling protocol. A DNA template was prepared by PCR amplification and digested with Dnase I during 1 hour. Dnase I products were concentrated by precipitation with isopropanol. Gel electrophoresis showed fragments of the desired size range (20-600 pb) without a full-length template remaining in the reaction mixture. A high concentration of fragments was crucial for performing PCR without primers. In this case, a template concentration of 32.5 ng/mu l was appropriate. Amplification of recombinant genes in a standard PCR reaction (template dilution 1:100) produced a smear with a low yield for the full-length sequence. A single product of the correct size was obtained by PCR with nested primers separated from the previously used primers by 40 pb. In our laboratory, native HRP has been functionally expressed in a baculovirus expression vector system. The purpose is to develop the screening of the first generation of random mutants in this system. To facilitate detection of those clones that have high peroxidase activity, we developed a rapid method: after five days postinfection agarose plates with six wells were covered with DAB (3,3´-diaminobenzidine) and H2O2. The appearance of brown-black stain allows visualization of up to 100 active clones/well in only 1 min.

directed evolution; mutant library; horseradish peroxidase

Construction of a mutant library of horseradish peroxidase gene by directed evolution and development of an in situ screening method

F.M.Mendive^II; M.M.Segura^I; H.M.Targovnik^II; O.Cascone^I; M.V.Miranda^I

ICátedra de Microbiología Industrial y Biotecnología

^IICátedra de Genética y Biología Molecular. Facultad de Farmacia y Bioquímica. Universidad de Buenos Aires, Junín 956, (1113), Buenos Aires, Argentina

^{Address to correspondence}

ABSTRACT

A process of directed evolution applied to obtain a library of mutants of horseradish peroxidase (HRP) enzyme is described. We have introduced slight variations into the original DNA shuffling protocol. A DNA template was prepared by PCR amplification and digested with Dnase I during 1 hour. Dnase I products were concentrated by precipitation with isopropanol. Gel electrophoresis showed fragments of the desired size range (20-600 pb) without a full-length template remaining in the reaction mixture. A high concentration of fragments was crucial for performing PCR without primers. In this case, a template concentration of 32.5 ng/m l was appropriate. Amplification of recombinant genes in a standard PCR reaction (template dilution 1:100) produced a smear with a low yield for the full-length sequence. A single product of the correct size was obtained by PCR with nested primers separated from the previously used primers by 40 pb.

In our laboratory, native HRP has been functionally expressed in a baculovirus expression vector system. The purpose is to develop the screening of the first generation of random mutants in this system. To facilitate detection of those clones that have high peroxidase activity, we developed a rapid method: after five days postinfection agarose plates with six wells were covered with DAB (3,3´-diaminobenzidine) and H₂O₂. The appearance of brown-black stain allows visualization of up to 100 active clones/well in only 1 min.

Keywords: directed evolution, mutant library, horseradish peroxidase.

INTRODUCTION

The use of biocatalysts (both enzymes and whole cells) as benign alternatives to chemical catalysts for producing renewable chemicals, pharmaceuticals, polymers, and fuels is integral to realizing current visions of sustainable development (Nedwin, 1997). But there are discrete performance limitations that have impeded realization of the industrial potential of biocatalysts. These limitations originate in the physiological role that biocatalysts play. Most enzymes have evolved to function optimally on a narrow set of substrates and under the precise conditions (temperature, ionic strength, pH, etc.) of their natural niche, and these characteristics are often suboptimal for biotechnological applications.

Directed evolution is a strategy for optimizing proteins. The strength of the combinatorial nature of DNA shuffling was first demonstrated using the TEM-1 b-lactamase gene. An evolved bacterial clone had shown a 32,000-fold increase in drug resistance compared with the parental gene (Stemmer, 1994a). By employing random mutagenesis and gene recombination followed by screening or selection, directed evolution has been successfully applied to improve a variety of enzyme properties, such as substrate specificity (Zhang et al., 1997), protein folding (Crameri et al., 1996), enhanced activity (Stemmer, 1994b), activity in organic solvents (Moore et al., 1997), stability at high temperature (Giver et al., 1998) and resistance to chemical modification (Matsumura et al., 1999), which are often critical to industrial applications.

The inability of eukaryotic proteins to fold properly in E. coli represents one of the bottlenecks in directed evolution (Lin et al., 1999). Another challenge of protein design by molecular breeding is the formulation of a screen that precisely emulates the final process conditions. In general, assays that quickly analyze an enormous population of variants tend to place in jeopardy the characteristics for which the variants are screened, sacrificing veracity and approppriateness for increased throughput. Consequently, while a high capacity assay may be an efficient way to reduce the number of candidates, it is essential to progress to a more accurate screen before doing additional cycles of recombination (Ness et al., 2000).

There is growing interest in the use of eukaryotic peroxidases as industrial biocatalysts. This enzyme is a hemoprotein stabilized by four disulfide bonds that is highly glycosylated (~21% of its molecular weight). These two properties make its recombinant expression in bacteria difficult because of the inability of procaryotes to carry out posttranslational modifications. It´s expression in E. coli has yielded inclusion bodies with no functional expression (Ortlepp et al., 1989), but in one case, the expressed protein was obtained in its active form (Smith et al., 1990). After refolding, the active protein yield was 100 mg/L. To solve this problem, Arnold and coworkers reported on a variant of HRP obtained by directed evolution. In this case, HRP expresses in active form in E. coli to a level of about 110 mg/L, which is similar to that reported above after a laborious in vitro refolding (Lin et. al., 1999).

On the other hand, the baculovirus insect cell system is a popular choice for expressing foreign genes. Eukaryotic insect cells provide a means for overexpressing soluble proteins that are posttranslationally modified. Native horseradish peroxidase has been successfully expressed in the baculovirus expression system (Hartmann and Ortiz de Montellano, 1992).

Here we report on a process of directed evolution applied to obtain a library of mutants of HRP. For its screening in the baculovirus expression system, we developed a methodology suitable for detecting HRP activity on lisis plaque assay within a short time.

MATERIALS AND METHODS

Materials

Plasmid pAcGP67-B (Pharmingen, San Diego, CA, USA) containing the gene for horseradish peroxidase isozyme c was kindly provided by Dr. P.E. Ortiz de Montellano (University of California, San Francisco). The Baculogoldâ transfection kit was from Pharmingen, San Diego, CA, USA. Peroxidase substrate DAB (3,3´diaminobenzidine) was from Sigma (St. Louis, MO). Taq polimerase, Grace´s medium, penicillin-streptomycin solution and Dnase I were from Gibco BRL (Gaithersburg, MD, USA). Fetal calf serum (FCS) was from Nutrientes Celulares S.A. (Buenos Aires, Argentina).

METHODS

Substrate Preparation

The substrates for the shuffling reaction were 1.0 kb double-stranded DNA. PCR products were amplified from plasmid pAcGP67-B containing the HRP isozyme c gene (pGP6HHRP) with the primer sequences CACACAAGCAAGATGGTAAGCGC (forward) and GCTTCATCGTGTCGGGTTTAACATTACGG (reverse). The PCR conditions (100 ml final volume) were 0.8 mM each primer, 1x Taq buffer, 0.4 mM each dNTP and 2 U Taq polymerase. The PCR program was 95°C for 6 min (denaturalization time) and 30 cycles, 95°C for 30 sec, 52°C for 30 sec and 72°C for 1 min. Free primers from the PCR product were removed using the Concert^TM PCR purification system (Gibco, BRL, Gaithersburg, MD, USA). The DNA concentration was 0.4 mg/ml.

Dnase I Digestion

The mixture containing 60 ml of PCR product was diluted to 100 ml with 25 mM Tris-HCl, pH 7.4, 0.5 mM MgCl₂ and 0.15 U Dnase I were added. The mixture was digested at room temperature for 1 hour.

Recovery and Concentration of Fragments

Fragments were concentrated by precipitation with isopropanol. The pellet was resuspended in 25 ml of sterile bidistilled water. The size of the resulting fragments was determined on 1.5% agarose gel without another purification step. Fragments obtained were around 20-600 bp and had a concentration of 160 ng/ml.

Fragment Reassembly

Ten ml of the fragment mixture was resuspended in PCR reaction solution containing 0.4 mM each dNTP, 2.5 mM MgCl₂, 1x Taq buffer, and 2.5 U of Taq polymerase in a 100 ml volume. The PCR program was 95°C for 1 min (denaturing time) and 60 cycles, 95 °C for 30 sec, 50°C for 30 sec and 72 °C for 30 sec, followed by an extension step at 72°C for 10 min.

PCR with Primers

Primerless PCR product was diluted 1:100. PCR reactions containing 0.8 mM each original primer in combination with different template concentrations were carried out. The PCR program was 95°C for 6 min, and 15 cycles, 95°C for 30 seg, 52°C for 30 sec, 72°C for 1 min.

Nested PCR

1 ml of the last PCR product was amplified using nested primers. Primer sequences were TTTGCGGCGGATCTTGGAT (forward) and CCAGGAAAGGATCAGATCTGC (reverse). PCR conditions were 0.5 pmol primers, 0.4 mM dNTPs, 0.6 M MgCl₂, and 2 U of Taq polymerase.

Tissue Culture

Spodoptera frugiperda Sf9 cells were maintained in monolayers at 27°C in Grace´s medium containing 10% FCS and supplemented with 50U/ml penicillin and 50mg/ml streptomycin. Cells were routinely subcultured every 2 to 3 days.

Vector

Plasmid pGP6HHRP (native HRP containing a 6xHis tag) was amplified by transformation of competent E. coli strain DH5a and selection of colonies positive for ampicillin resistance. This transfer vector was purified using Wizard PCR Preps (Promega, Madison, Wi, USA).

Production of Recombinant Virus

Sf9 cells were coinfected with purified vector and wild-type Autographa californica nuclear polyhedrosis virus (AcMNPV) in the presence of calcium phosphate (Burand et. al, 1980). After five days, the medium was removed and centrifuged. Following one round of plaque purification, viral clone was isolated and amplified on monolayers in 5 ml of Grace´s medium. After three days, the cell medium was removed and titered by plaque assay.

Screening for HRP Activity

Healthy cells in log-growth phase with at least 90% viability were allowed to attach firmly to a six-well microplate (30 min, 27°C). Then serial dilutions of the viral stock were made in Grace´s medium containing 10% FCS and 900 ml of each dilution was added per plate. A negative control was performed by adding 900 ml of culture medium without virus. The plaque was mixed gently by rocking at 27°C for 1 hour to allow virus particles to infect the cells. Then an overlay of agarose was poured on the monolayer and incubated for five days at 27 °C until visible plaques developed. Lastly, a second overlay of agarose containing 0.29 mg/ml DAB (3,3´- diaminobenzidine) and 0.8 mg/ml H₂O₂ in 25mM Tris-HCl buffer was poured.

RESULTS AND DISCUSSION

DNA Shuffling

The fidelity of Taq is the lowest of several DNA polymerases, so we chose this enzyme for amplification of the original gene by PCR reaction. Stemmer´s protocol probably offers the best cross-over frequency, that meaning the average number of recombination events per gene that can be achieved. This is an important parameter for obtaining a large number of mutants. There are two protocols that cut DNA fragments within the desired size range from the gel, the first involving <100 bp fragments and the second, 100-200 bp. In our case, we did not collect fragments from the agarose gel, so DNA ranged from 20 to 600 bp, the product of DNase digestion in the presence of Mg²⁺ during 1 hour at room temperature (Fig.1A). We demonstrated that in this range it is possible to obtain an amplification reaction, but as screening was not carried out, the amount of mutation was not clear. To ensure that no template remained in this mixture, a PCR reaction with primers was held by using 20-600 bp fragments as the template, and no product was obtained.

Fragments were reassembled by overlap extension PCR. In this step we found that a high concentration of fragments was needed. The template concentration was 32.5 ng/ml. The number of cycles depends on fragment size. In our case, 60 cycles were enough to perform this step. Annealing temperature (50°C) was lower than that of the first PCR reaction, and an extension time of 10 min allowed an increase in DNA fragment size (Fig. 1A).

The next PCR with primers was identical to the original in obtaining the template to shuffle, but the number of cycles was 15. In this step a lot of reactions were held at different template-to- primer concentration ratios (Fig. 1B). In all cases amplification produced a smear with a very low yield for the full-length sequence. Therefore, we selected the reaction with a high yield and a low amount of nonspecific products (Fig. 1B, lane 2) and reamplified it with nested primers separated by 40 bp of the previously used primers (Fig. 1C). In this way, nonspecific products disappeared and material was ready for purification, digestion with appropriate restriction endonucleases, and ligation onto the cloning vector (pAcGP67B).

In Situ HRP Activity Detection

A plaque assay was developed for detection of HRP activity. In this assay cell monolayers (3.1.10⁶ total cells/well) were infected with a low virus ratio so only isolated cells became infected. Viral stock dilution was adjusted for inoculum to obtain about 100 viral plaques per well. The first overlay of agarose keeps the cells stable and limits the spread of virus. Each group of infected cells is referred to as a plaque. After several infection cycles, the infected cells in the center of the plaques began to lyse. The plaques can be visualized either by the naked eye or by light microscopy and represent a single virus. The second overlay of agarose transported peroxidase substrate (DAB and hydrogen peroxide) without disturbing the monolayer and viral plaques. Clonal virus populations that show positive reaction can be purified by isolating individual plaques. Positive plaques are visualized as brown-black cells by microscopy or the naked eye (Figure 2).

CONCLUSIONS

A method based on directed evolution to obtain a library of mutant HRP genes was described. As the original strategy has a low fragment yield and PCR without primers is a critical step that requires a high concentration of template, we adjusted the DNase digestion reaction at a selected time to obtain fragments of the desired size that were concentrated by precipitation.

On the other hand, with the colorimetric method it will be possible to screen a large number of recombinant viruses to select the best candidates. Then, after viral clone purification, a new assay in solution under the final process conditions must be developed. In addition, the method described here can be applied to distinguish recombinant viruses that express peroxidase from the wild type when necessary.

ACKNOWLEDGMENTS

We thank Dr. Frances Arnold for her helpful advice.

M.V.M., O.C. and H.M.T. are career researchers at the CONICET.

Received: March 5, 2002

Accepted: September 12, 2002

• Burand, J.P., Summers, M.D., and Smith, G.E. (1980).Transfection with Baculovirus DNA. Virology 101, 286-290.
• Crameri, A., Whitehorn, E.A., Tate, E., and Stemmer, W.P.C. (1996). Improved Green Fluorescent Protein by Molecular Evolution Using DNA Shuffling. Nature Biotechnology, 14, 315-319.
• Giver, L., Gershenson, A., Freskgard, P.O., and Arnold, F.H. (1998). Directed Evolution of a Thermostable Esterase. Proc. Natl. Acad. Sci. USA, 95, 12809-12813.
• Hartmann, C. and Ortiz de Montellano, P.R. (1992). Baculovirus Expression and Characterization of Catalytically Active Horseradish Peroxidase. Arch. Biochem. Biophys., 297, 61-72.
• Lin, Z., Thorsen, T., and Arnold, F.H. (1999). Functional Expression of Horseradish Peroxidase in E. coli by Directed Evolution. Biotechnol. Prog., 15, 467-471.
• Matsumura, I., Wallingford, J.B., Surana, N.K., Vize, P.D., and Ellington, A.D. (1999). Directed Evolution of the Surface Chemistry of the Reporter Enzyme b-glucuronidase. Nature Biotechnology, 17, 696-701.
• Moore, J.C., Jin, H.M., Kuchner, O., and Arnold, F.H. (1997). Strategies for the in vitro Evolution of Protein Function: Enzyme Evolution by Random Recombination of Improved Sequences. J. Mol. Biol., 272, 336-347.
• Nedwin, G. (1997). Using Enzymes as Benign Substitutes for Synthetic Chemicals and Harsch Conditions in Industrial Processes. Biotechnology in the Sustainable Environment, G. Sayler, J. Sanseverino and K. Davis, eds. Plenum Press, New York, 13-32.
• Ness, J.E., Del Cardayré, S.B., Minshull, J., and Stemmer, W.P.C. (2000). Molecular Breeding: The Natural Approach to Protein Design. Advances in Protein Chemistry. Vol 55: Evolutionary Protein Design. Arnold, F.H., ed., Academic Press, San Diego, London, 261- 292.
• Ortlepp, A.S., Pollard-Knight, D., Chiswell, D.J. (1989). Expression and Characterization of a Protein Specified by a Synthetic Horseradish Peroxidase Gene in Escherichia coli. J. Biotechnol. 11, 353-364.
• Smith, A.T., Santama, N., Dacey, S., Edwards, M., Bray, R.C., Thorneley, R.N. and Burke, J.F. (1990). Expression of a Synthetic Gene for Horseradish Peroxidase c in Escherichia coli and Folding and Activation of the Recombinant Enzyme with Ca²⁺ and Heme. J. Biol. Chem., 265, 13335-13343.
• Stemmer, W.P.C. (1994a). DNA Shuffling by Random Fragmentation and Reassembly: In vitro Recombination for Molecular Evolution. Proc. Natl. Acad. Sci. USA., 91, 10747.
• Stemmer, W.P.C. (1994b). Rapid Evolution of a Protein in vitro by DNA Shuffling. Nature, 370, 389-391.
• Zhang, J.-H., Dawes, G., and Stemmer, W.P.C. (1997). Directed Evolution of a Fucosidase from a Galactosidase by DNA Shuffling and Screening. Proc. Natl. Acad. Sci. USA, 94, 4505-4509.

Publication Dates

• Publication in this collection
  19 Mar 2003
• Date of issue
  Mar 2003

History

• Received
  05 Mar 2002
• Accepted
  12 Sept 2002