Recombinant expression, purification, and characterization of XorKII: A restriction endonuclease from Xanthomonas oryzae pv. oryzae

Protein Expression and Purification 62 (2008) 230–234

Contents lists available at ScienceDirect

Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep

Recombinant expression, purification, and characterization of XorKII: A restriction endonuclease from Xanthomonas oryzae pv. oryzae Won Jae Moon a, Jae-Yong Cho b, Young Kee Chae a,* a b

Department of Chemistry, 98 Gunja-Dong, Gwangjin-Gu, Seoul 143-747, Republic of Korea Department of Bioindustry and Technology, Sangji University, 660 Woosan-Dong, Wonju-Si, Gangwon-Do, 220-702, Republic of Korea

a r t i c l e

i n f o

Article history: Received 9 July 2008 and in revised form 20 August 2008 Available online 31 August 2008 Keywords: XorKII Restriction endonuclease Xanthomonas Recombinant

a b s t r a c t An endonuclease from Xanthomonas oryzae pathovar oryzae (Xoo) KACC10331, XorKII, was recombinantly produced in Escherichia coli by applying the stationary state induction method, which was necessary to prevent the unwanted lysis of E. coli cells. XorKII was purified by immobilized metal affinity chromatography on an FPLC system. The yield was 3.5 mg of XorKII per liter of LB medium. The purified recombinant XorKII showed that it recognized and cleaved to the same site as PstI. It behaved as a dimer as evidenced by the size exclusion chromatography. The specific activity of the purified XorKII was determined to be 31,300 U/mg. The enzyme activity was monitored by cleaving lambda DNA or YEp24 plasmid as substrates. The enzyme was the most active at 10 mM Tris–HCl pH 7.0, 10 mM MgCl2, 1 mM dithiothreitol at 37 °C. XorKII was easily inactivated by heating at 65 °C for 5 min, but retained most of the original activity after incubation at 37 °C for 24 h. Ó 2008 Elsevier Inc. All rights reserved.

Introduction Restriction endonucleases play an important role in defending bacteria against invading viruses. They cooperate with methyl transferases so that unmethylated foreign DNA at N4 or C5 at cytosine or N6 at adenine within the recognition sequence gets cleaved while the properly methylated host DNA is protected [1]. These restriction endonucleases can be classified into four different types: Type I, II, III, and IV [2]. Among these four types, Type II restriction endonucleases are known to be homodimeric or tetrameric, cleave DNA within or close to the specific recognition sites, and require divalent ions such as Mg2+ for catalysis. Xanthomonas oryzae pv. oryzae (Xoo)1 is a member of the c-subdivision of the Proteobacteria and causes bacterial blight on rice [3]. Bacterial blight is a vascular disease resulting in white lesions along the leaf veins. Severe infestation can cause yield losses as high as 50%. With a goal to better understand the pathogen and thus minimize the possible losses by this pathogen, the Xoo genome project was completed [4]. However, the restriction/modification system of the KACC10331 strain was highly active, and genetic analysis of this strain has failed. To overcome this predicament, it was thought to be essential to first understand the restriction/ modification system.

* Corresponding author. Fax: +82 2 462 9954. E-mail address: [email protected] (Y.K. Chae). 1 Abbreviations used: Xoo, Xanthomonas oryzae pv. oryzae; ORFs, open reading frames. 1046-5928/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2008.08.007

In this work, we focused on XorKII, one of the two restriction endonucleases from Xoo. The other endonuclease, XorKI, was reported elsewhere [7,8]. XorKII was known to be an isoschizomer of PstI [5]. However, in the previous report [5], XorKII was directly produced from Xoo and purified only to a minor degree, so a detailed analysis was impossible. The previous result was reported in 1980, and there has been little further work done except some on methyl transferases [10]. This was rather curious, and is likely due to the fact the production of necessary amount of pure enzyme was very difficult. Here we report the optimized production and purification of recombinant XorKII and its biophysical and biochemical properties.

Materials and methods Plasmid construction The genomic DNA of Xanthomonas oryzae pv. oryzae (Xoo) KACC10331 was prepared as described in [4]. The gene coding for XorKII was amplified by PCR. The primers were designed based on the Xoo genome sequence (GenBank Accession No. NC_006834). The forward primer was 50 -GGG CCC GGA TTC GTG AGC TTG CCT CCC TAC GTC-30 , and the reverse primer, 50 -GGG CCC CTC GAG TTA AAC GCC GTG CAT CAA CGT-30 . The PCR product was purified, cut by BamHI and XhoI, purified again, and ligated with the pET28a vector (Novagen, Madison, WI, USA), which was previously digested with the same enzymes. The resulting plasmid was named pET-28a/XorKII. The plasmid was brought into Rosetta2(DE3)/

W.J. Moon et al. / Protein Expression and Purification 62 (2008) 230–234

pLysS (Novagen, Madison, WI, USA) for recombinant expression. This strain did not have the cognate methyl transferase gene for protection against XorKII. Small scale expression test A single colony was used to inoculate a 3 ml LB medium supplemented with 50 lg/ml kanamycin and 34 lg/ml chloramphenicol. When fully-grown, 10 ll of the culture was used as an inoculum to 3 ml each of the following media: 50%, 60%, 70%, 80%, 90%, and 100% LB. The cultures were grown overnight at 37 °C in a shaking incubator. The next morning, a proper volume of 10 LB medium was added to the culture so that the final concentration of the nutrient in the culture became equal to LB medium. IPTG was also added at the same time to the final concentration of 0.5 mM to induce protein production. The culture was further grown for another 3 h, and 50 ll were taken from each culture and harvested by centrifugation. The cells were resuspended in 50 ll of 10 mM Tris–HCl pH 7.5 containing 8 M urea and mixed with 50 ll of 2 SDS sample buffer for analysis. Protein production A single colony was used to inoculate a 1000 ml medium containing 7 g of bactotrypton (BD, Franklin Lakes, NJ, USA), 3.5 g of yeast extract (BD, Franklin Lakes, NJ, USA), and 7 g of NaCl, supplemented with 50 lg/ml kanamycin and 34 lg/ml chloramphenicol. The culture was grown overnight at 37 °C in a shaking incubator. The next morning, 3 g of bactotrypton, 1.5 g of yeast extract and 3 g of NaCl were directly mixed with the fully-grown culture. After the nutrients were completely dissolved, IPTG was added to the final concentration of 0.5 mM to induce protein production. The culture was further grown for another 3 h, and harvested by centrifugation at 5000 rpm for 10 min at 4 °C. The harvested cells were resuspended in 50 ml of 10 mM Tris–HCl pH 8.0 and frozen at 20 °C.

231

at 37 °C. Buffers used to find the optimal cleavage condition were as follows: Buffer 1, 10 mM Tris–HCl pH 7.0, 10 mM MgCl2, 1 mM dithiothreitol; Buffer 2, 50 mM NaCl, 10 mM Tris–HCl pH 7.9, 10 mM MgCl2, 1 mM dithiothreitol; Buffer 3, 100 mM NaCl, 50 mM Tris–HCl pH 7.9, 10 mM MgCl2, 1 mM dithiothreitol; Buffer 4, 50 mM potassium acetate, 20 mM Tris-acetate pH 7.9, and 10 mM Magnesium Acetate, 1 mM dithiothreitol. For the inactivation test, XorKII was preincubated at 65 °C for 5, 10, 15, 20, 25, 30 min before mixing with the substrate. For the stability test, XorKII was preincubated at 37 °C for 4, 8, 12, 18, and 24 h before mixing with the substrate. Specific activity determination The concentration of XorKII was determined by using Protein Assay kit (Bio-Rad, Hercules, CA, USA). Bovine serum albumin was used to set up the standard curve. One unit was defined as the amount of XorKII to cleave 1 lg of lambda DNA at least once in an hour. The reaction mixtures contained 5 ll of 10 reaction buffer, 1 lg of DNA substrate, and the total volume was set to 50 ll. A serial dilution of the purified XorKII was made, and 1 ll from each dilution was added to the reaction mixture. The mixture was incubated at 37 °C for an hour, and the reaction was stopped by incubation at 80 °C for 15 min. The degree of cleavage was visualized by agarose gel electrophoresis. The highest dilution showing at least one cleavage was used to determine 1 U of XorKII.

Results and discussion XorKII gene cloning and plasmid construction

The cells were lysed by freeze-and-thaw, and the DNA was fragmented by ultrasonication. The soluble fraction was retained after centrifugation at 15,000 rpm for 20 min at 4 °C and then loaded onto the HiPrep Chelate column (5 ml) charged with Ni2+ (GE Healthcare, Piscataway, NJ, USA). Imidazole gradient of 10 mM to 300 mM was applied to the column on ÄKTA Basic system (GE Healthcare, Piscataway, NJ, USA). The fractions containing XorKII were pooled, concentrated, and buffer exchanged with 100 mM Tris–HCl pH 8.0 containing 200 mM NaCl, 2 mM DTT, 0.2 mM EDTA, and 0.1% Triton X-100 by Amicon Ultra (Millipore, Billerica, MA, USA). The resulting concentrated XorKII (4 ml) was mixed with the same volume of 100% glycerol and kept at 20 °C.

Since the genome sequences of Xoo was reported, we analyzed the open reading frames (ORFs) to search for the gene coding for Type II restriction endonucleases. A candidate gene was located at sequences between 4219262 and 4218168 on the complementary strand although it was not annotated yet. This ORF had a 93% sequence identity with R.Xph I (GenBank Accession No. AF042157). From the fact that RM system is mobile, and the gene coding for endonuclease and its cognate methylase are located close to each other on the chromosome [6], we expected to find a methylase gene nearby. Fortunately, a putative methylase gene was found right next to the XorKII gene in the Xoo sequence, which increased the probability that this candidate was actually the XorKII we were searching for. The XorKII gene was amplified from the Xoo genome and inserted into pET-28a between BamHI and XhoI sites. As a consequence of using these cloning sites, the resulting protein would have 34 additional amino acid residues to the N-terminus of the wild-type protein. As will be discussed below, this stretch of additional residue did not cause any detrimental effects on the endonuclease activity.

Size exclusion chromatography

Protein production

Size exclusion chromatography was performed with a Superdex 200 10/30 column (GE Healthcare, Piscataway, NJ, USA), which was pre-equilibrated with 10 mM sodium phosphate buffer pH 7.4 containing 1 M NaCl on ÄKTA Basic system (GE Healthcare, Piscataway, NJ, USA). The proteins used for setting up a standard curve were BSA (67 kDa), ovalbumin (43 kDa), chymotrypsinogen A (25 kDa), and ribonuclease A (13.7 kDa).

The cell growth and protein induction were performed in a rather non-typical way. This was because the cells died when forced to produce XorKII at 37 °C. It was suspected that the produced endonuclease molecules were too toxic to cells in their mid-log phase. The previous version of the stationary phase induction method [9] was attempted, but the cell lysis could not be lessened. Instead of diluting the fully-grown culture as the previous method described [9], we tried growing cells in a less nutrient medium and then adding back the omitted amount of nutrients for further growth. Upon adding the nutrients, IPTG was added for the induction of protein production. As seen in Fig. 1, the best result could be obtained if 70% LB medium had been used in its ini-

Protein purification

Endonuclease activity test The YEp24 vector (GenBank Accession No. L09156) with three PstI sites was used as a substrate. Enzyme digestion was performed

232

W.J. Moon et al. / Protein Expression and Purification 62 (2008) 230–234

Fig. 2. Ten percent NuPAGE gel with MOPS running buffer. Samples were taken from various stages of the purification procedure. Lane 1, size marker; lane 2, whole cell lysate before induction; lane 3, whole cell lysate after induction; lane 4, soluble fraction of lane 3; lane 5, insoluble fraction of lane 3; lane 6, flow-through fraction of lane 4 from HiPrep Chelating column; lane 7, fractions bound to HiPrep Chelating column that corresponded to XorKII. The line indicates the XorKII band.

Fig. 1. Ten percent NuPAGE gel with MES running buffer. Samples were made of whole cells with different initial LB dilutions. Lane 1, size marker; lane 2, no induction; lane 3, 50%; lane 4, 60%; lane 5, 70%; lane 6, 80%; lane 7, 90%; lane 8, 100%. The lines beside the gel picture were drawn to differentiate two closely separated bands. The upper band corresponded to XorKII.

tial condition. As indicated with two lines at the right-hand side of the gel (Fig. 1), there is a band right beneath the XorKII, but it is not difficult to tell one from the other. Other initial conditions such as 50%, 60%, or 80% LB medium showed noticeable but weaker expression levels, while those such as 90% or 100% did not. In fact, 100% LB medium served as an internal control of this method since no additional nutrient was added upon induction, and we expected that we should get a very similar gel pattern as in the case of no induction sample. The result was surprising because the protein was successfully expressed without much cell lysis. We hope this method will serve for other proteins, especially those toxic to the host.

Fig. 3. Elution profile from the HiPrep Chelating column. An arrow was marked to indicate the XorKII peak.

(Fig. 2, lane 7) from 1 l LB medium was 3.5 mg as calculated from the standard curve. We did not cleave the His-tag because additional processes, including thrombin cleavage and at least one more purification step, could have led to the inactivation of the enzyme. We thought the speed of the purification was the top priority in purifying this not-so-stable enzyme.

Protein purification Oligomeric state of XorKII Although hexahistidine tagging of XorKII facilitated the purification process, the manual, stepwise elution did not yield a highly purified enzyme. The slightly smaller protein as shown in Fig. 1 acted as a major contaminant during the manual purification procedure. Removing of this contaminant was a major task in purifying XorKII. We first employed the cation exchange chromatography as a filtering step before the IMAC since not many proteins could bind to such a column. Unfortunately, the manual IMAC did not help much to remove the contaminants (data not shown). The automated imidazole gradient elution was much more efficient in terms of purity and time. As shown in Fig. 2, most of the expressed protein was in the form of inclusion bodies (lane 5). The gradient elution profile showed a small but recognizable peak at 180 mM imidazole concentration (Fig. 3). The purified XorKII

To determine the oligomeric state of XorKII, 1 mg of the purified enzyme was loaded onto Superdex 200 HR 10/30 column. We first tried running this column in a 10 mM phosphate buffer containing 150 mM NaCl, but the protein molecules inside the column unpredictably produced an elution profile of much lower intensity than expected (Fig. 4, lower trace). When the salt concentration was increased to 1 M, the elution profile changed to show a major peak around 13 ml (Fig. 4, upper trace). From the position of XorKII in the elution profile, we estimated the molecular size to be 95.3 kDa and concluded that XorKII existed as a dimer since the monomeric form should be close to 43.2 kDa. The actual size of the dimer is 86.4 kDa, and this slight overestimation could have resulted from its elongated shape (or a dumbbell shape) assuming

233

W.J. Moon et al. / Protein Expression and Purification 62 (2008) 230–234 Table 1 Specific activity of XorKII in the purification step

Supernatant Bound fraction

Fig. 4. Elution profile from Superdex 200 HR column. Upper trace, chromatography with 1 M NaCl; lower trace, with 150 mM NaCl.

the monomer was more or less globular. It is well known that the elongated molecule has a larger hydrodynamic radius than the globular one of the same molecular weight. The smaller peak around 14.5 ml happened to correspond to 46.4 kDa, a monomeric size. It was intriguing to see the corresponding peak in the lower trace. It is was tempting to say that XorKII favors a dimeric state over a monomeric one at high salt conditions and vice versa at low salt conditions, but no evidence would support that conclusion. We can at least say that the major portion of XorKII molecules behaves as a dimeric form at high salt conditions in the absence of DNA substrates. The last peak around 16 ml corresponded to 24.4 kDa, which we assigned as a contaminant. Enzyme activity tests We tested four buffers, 1–4 as described in Materials and methods, and in the presence/absence of BSA. BSA did not show any effect on the enzyme activity (data not shown). As shown in Fig. 5, XorKII showed the highest activity in Buffer 1, 10 mM Tris–HCl pH 7.0, 10 mM MgCl2, 1 mM dithiothreitol. In the previous report [5], 6 mM Tris–HCl pH 7.4, 12 mM MgCl2, 6 mM mercaptoethanol, 0.1 mg/ml BSA was used. Those two buffer conditions were

Total protein (mg)

Total activity (u)

Specific activity (u/mg)

105.9 3.58

480,000 112,000

4530 31,300

practically the same considering the low salt and near neutral pH. We tried to determine the buffer condition independently from the previous study, and had the similar result, which strengthens the validity of our result. In Buffer 3, the XorKII cleavage resulted in a smearing pattern suggesting a possible nonspecific nuclease or star activity. We speculate that XorKII may switch from the specific endonuclease to the nonspecific nuclease upon increasing the salt concentration, but this was not pursued further in this study. The contamination of any endogenous, nonspecific nuclease from the host due to co-purification was also possible but considered less likely because we think we could have observed the similar smearing pattern in at least one of the other three buffers if such nuclease had been present. Since it showed the same cleavage pattern as that of PstI, XorKII was considered to be the neoschizomer of PstI as reported in the previous study [5]. The enzyme was stable at 37 °C in the storage buffer and retained full activity up to 24 h but was easily inactivated by heating at 65 °C for as short as 5 min. (data not shown) Specific activity of XorKII XorKII cleaves k DNA 28 times, so it was not feasible to check if the digestion was complete. We defined 1 U to be the amount of XorKII to cleave 1 lg k DNA at least once at 37 °C in a total reaction volume of 50 ll in 1 h, so it was easier to tell if the cleavage had occurred. The specific activity along the purification procedure is listed in Table 1. The previous study [5] reported the specific activity of 116 U/mg, and we found 31,300 U/mg. Their definition was based upon the complete digestion of the substrate, which is 28 cleavages as mentioned above; and to roughly compare two values, we can simply multiply 28 to their value, yielding 3248 U/mg. Despite the different unit definition, the value we found was far higher than the previous one. This is because the specific activity they reported was from the partially purified enzyme. However, we think that the previous study’s enzyme seemed to have lost all activity after the further purification step since the specific activity data could not be measured. This may be due to the fact that XorKII was not so stable unless it was kept with a stabilizer like glycerol. As stated in Table 1, the total activity decreased roughly fourfold from the supernatant fraction to the bound fraction. This is partly due to the nature of the purification where the loss of the target protein is inevitable. But more importantly, the enzyme seemed to keep losing its activity along the purification steps because when we stored the supernatant or the bound fraction at 4 or 20 °C overnight and tried to purify the enzyme the next day, we found the most of the activities were lost (data not shown). Therefore, we think the speed of the purification procedure and the existence of a stabilizer are the two biggest impact factors in the case of XorKII production. Acknowledgment This work is supported by the RDA Biogreen21 Fund (grant 20050401-034-743-176-06-00).

Fig. 5. Agarose gel electrophoresis to determine the buffer condition for XorKII activity. YEp24 plasmid was used as a substrate. Lane 1: size marker; lane 2: uncut YEp24; lane 3: YEp24 cut with PstI; lanes 4 to 7, YEp24 cut with XorKII in buffers 1– 4.

References [1] T.A. Bickle, D.H. Kruger, Biology of DNA restriction, Microbiol. Rev. 57 (1993) 434–450.

234

W.J. Moon et al. / Protein Expression and Purification 62 (2008) 230–234

[2] R.J. Roberts, M. Belfort, T. Bestor, A.S. Bhagwat, T.A. Bickle, J. Bitinaite, R.M. Blumenthal, S.K. Degtyarev, D.T. Dryden, K. Dybvig, K. Firman, E.S. Gromova, R.I. Gumport, S.E. Halford, S. Hattman, J. Heitman, D.P. Hornby, A. Janulaitis, A. Jeltsch, J. Josephsen, A. Kiss, T.R. Klaenhammer, I. Kobayashi, H. Kong, D.H. Kruger, S. Lacks, M.G. Marinus, M. Miyahara, R.D. Morgan, N.E. Murray, V. Nagaraja, A. Piekarowicz, A. Pingoud, E. Raleigh, D.N. Rao, N. Reich, V.E. Repin, E.U. Selker, P.C. Shaw, D.C. Stein, B.L. Stoddard, W. Szybalski, T.A. Trautner, J.L. Van Etten, J.M. Vitor, G.G. Wilson, S.Y. Xu, A nomenclature for restriction enzymes, DNA methyl transferases, homing endonucleases and their genes, Nucleic Acids Res. 31 (2003) 1805–1812. [3] A. Ezuka, H. Kaku, A historical review of bacterial blight of rice, Bull. Natl. Inst. Agrobiol. Resour. (Japan) 15 (2000) 53–54. [4] B.M. Lee, Y.J. Park, D.S. Park, H.W. Kang, J.G. Kim, E.S. Song, I.C. Park, U.H. Yoon, J.H. Hahn, B.S. Koo, G.B. Lee, H. Kim, H.S. Park, K.O. Yoon, J.H. Kim, C.H. Jung, N.H. Koh, J.S. Seo, S.J. Go, The genome sequence of Xanthomonas oryzae pathovar oryzae KACC10331, the bacterial blight pathogen of rice, Nucleic Acids Res. 33 (2005) 577–586.

[5] R.Y. Wang, J.G. Shedlarski, M.B. Farber, D. Kuebbing, M. Ehrlich, Two sequencespecific endonucleases from Xanthomonas oryzae. Characterization and unusual properties, Biochim. Biophys. Acta. 606 (1980) 371–385. [6] I. Kobayashi, Behavior of restriction-modification systems as selfish mobile elements and their impact on genome evolution, Nucleic Acids Res. 29 (2001) 3742–3756. [7] D.K. Hwang, J.Y. Cho, Y.K. Chae, Recombinant expression and purification of functional XorII, a restriction endonuclease from Xanthomonas oryzae pv. oryzae, J. Microbiol 45 (2007) 175–178. [8] D.K. Hwang, J.Y. Cho, Y.K. Chae, Recombinant expression and purification of functional XorII, a restriction endonuclease from Xanthomonas oryzae pv. oryzae, J. Microbiol. 45 (2007) 371 (Author’s correction). [9] Y.K. Chae, K.S. Cho, W. Chun, K. Lee, Protein production by stationary phase induction (SPI), Protein Pept. Lett. 10 (2003) 369–374. [10] S.H. Choi, J.E. Leach, Identification of the XorII methyltransferase gene and a vsr homolog from Xanthomonas oryzae pv. oryzae, Mol Gen Genet. 244 (1994) 383–390.