DNA shuffling of uricase gene leads to a more “human like” chimeric uricase with increased uricolytic activity

G Model

ARTICLE IN PRESS

BIOMAC-5467; No. of Pages 8

International Journal of Biological Macromolecules xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac

DNA shuffling of uricase gene leads to a more “human like” chimeric uricase with increased uricolytic activity Jing Chen a,1 , Nan Jiang a,1 , Tao Wang b,1 , Guangrong Xie a , Zhilai Zhang a , Hui Li a , Jing Yuan c , Zengxian Sun d,∗∗ , Jianhua Chen a,∗ a

School of Life Science and Technology, China Pharmaceutical University, Nanjing 210009, China Department of Neurosurgery, Shanghai 5th People’s Hospital, Shanghai Medical College, Fudan University, Shanghai 200240, China c School of life science, Faculty of Health and Life science, University of Liverpool, Liverpool, L69 3BX, UK d Department of Pharmacy, The First People’s Hospital of Lianyungang, Lianyungang 222002, China b

a r t i c l e

i n f o

Article history: Received 15 July 2015 Received in revised form 17 October 2015 Accepted 19 October 2015 Available online xxx Keywords: Urate oxidase DNA shuffling High-throughput screening

a b s t r a c t Urate oxidase (Uox) is the enzyme involved in purine metabolism. Pseudogenization of Uox gene is the underlying mechanism of hyperuricemia and gout in human. Although Uox from various microorganisms has been used in clinical practice for many years, its application is limited by potential immunogenicity. In order to develop a more “human like” uricase, DNA shuffling was used to create chimeric uricase with both improved enzymatic activity and increased homology with deduced human uricase (dHU) gene. By using wild porcine uricase (wPU) gene and dhu as parental genes, a diverse chimeric library was generated. After preliminary screening by a “homebrew” high throughput protocol, approximately 100 chimeras with relatively high enzymatic activity were obtained. By further activity comparison of the purified enzymes, chimera-62 with increase in both activity and homology with dHU compared with wPU was selected. Its Km and catalytic efficiency were determined as 9.43 ± 0.04 ␮M and 2.67 s−1 ␮M−1 respectively. There were 33 amino acid substitutions in chimera-62 when compared with dHU and 5 substitutions when compared with wPU. By homology modeling and 3-D structure analysis, it was speculated that mutations G248S and L266F contributed to the increased activity of chimera-62 by increasing the stability of ␣-helix and surface polarity respectively. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Urate oxidase, or uricase (EC 1.7.3.3; Uox), is a peroxisomal enzyme that is active in most non-human primates and other mammals [1–3]. It converts urate to allantoin which is more soluble than uric acid and excreted easily by the kidney [4–6]. It has been proved that a progressive accumulation of slightly deleterious amino acid replacements lead to inactivation of human uricase (HU) gene [7]. As a consequence, uric acid levels are 3 to 10 times higher in human than in other mammals possessing a functional uricase, leading to hyperuricemia [8]. Overtime, chronic hyperuricemia can result in

∗ Corresponding author. School of Life Science and Technology, China Pharmaceutical University, No.24,Tong Jia Xiang, Nanjing 210009, PR China. Tel.: +86 2583271265; fax: +86 2583271383. ∗∗ Corresponding author. Department of Pharmacy, The First People’s Hospital of Lianyungang, No.182, Tongguan North Road, Lianyungang 222002, PR China. E-mail addresses: [email protected] (Z. Sun), [email protected], [email protected] (J. Chen). 1 These authors contributed equally to this paper.

destructive crystalline urate deposits around joints, in soft tissues and in some organs, causing the clinical symptoms of gout. Conventional treatment for hyperuricemia and goat includes uricosuric agents that promote uric acid excretion and inhibitors of xanthine oxidase that block urate formation [9,10]. However, these agents induce very slow reduction in uriate deposits and also lead to severe complications [11]. It was more than 20 years ago when the first uricase product was marketed under the brand name of Uricozyme for treatment of severe hyperuricemia. Despite its various advantages over conventional therapies such as more rapid onset of action, serious allergic reactions caused by the microorganism origin (Aspergillus flavus) prove to be a limiting factor for the therapeutic use of Uricozyme [12]. Human uricase gene is disabled by accumulation of nonsense and missense mutations during evolution, it is therefore difficult to recover HU activity by amino acid mutations. As such, the medical community has a strong interest in developing a recombinant “human-like” uricase to treat gout. Canine-human chimeric uricases were constructed and studied by Chun Zhang et.al. A canine uricase with amino acid residues 241-304 of HU introduced at the

http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053 0141-8130/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model

ARTICLE IN PRESS

BIOMAC-5467; No. of Pages 8

J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

2

C-terminus exhibited lower immunogenicity while retaining the enzymatic activity [13]. A porcine-human chimeric uricase consisting of 1-222 amino acids of wild porcine uricase (wPU) and 223-304 amino acid residues of HU also retained enzymatic activity and demonstrated lower immunogenicity after further PEGylation [14]. In addition, investigations on pig-baboon chimeric uricase were conducted by Duke University and Savient Corp, resulting in the approval of Pegloticase (Krystexxa) by FDA in 2010 for refractory gout [15,16]. The pig portion provides a high catalytic efficiency and the substantially less active baboon portion may help to avoid an immune response in humans. The development of therapeutic uricase for human use is an intractable challenge that activity, stability and immunoreactivity should all taken into consideration. Until now, construction of chimeric enzymes with higher homology to hypothetic human uricase proves to be the most promising strategy. DNA shuffling developed by Stemmer [17] refers to the recombination of equivalent genes from natural homologous families rather than random mutagenesis of a single gene [18]. It is a powerful technique for molecular directed evolution in vitro, which generates a library of chimeras by recombination of homologous sequences and combines useful mutation for individual genes [19]. Briefly, two or more parental genes are fragmented with DNase I and the fragments with various lengths are then reassembled in a primerless PCR. The fragments anneal where there is sufficient sequence identity, resulting in full-length variants of the original gene that have inherited mutations from multiple templates [20]. By utilizing mutations that have been proven to be functional in nature, DNA shuffling can significantly accelerate the accumulation of beneficial mutations. DNA shuffling has been successfully used to create thermostable and alkaline stable xylanase variants [21], to improve transport activity of Na+ /H+ antiporters [19], to improve thermal stability of GH43 ␤-xylosidase XylBH43 [22] and to alter substrate specificity of avidin [23]. Due to the instability and poor solubility of mammalian uricase below pH 9.0 [24], it is difficult to obtain 3-D structure of mammalian uricase by X-ray crystallography. As a result, rational approach of protein evolution is limited by the inadequate elucidation of structure-activity relationship. However, DNA shuffling is especially applicable when the structural determinants behind molecular activity are difficult to determine. In this study, a deduced human uricase (dHU) with restoration of two premature stop codons at amino acid position 33 and 187 and wPU gene were used as parental genes in a two-round DNA shuffling to generate mutants with increase in both activity and homology with dHU. The resulting mutants were constructed into B. subtilis WB800 for secretory expression. A high throughput screening method was developed to select mutants with enzymatic activity. Finally, one chimera with increased activity and higher homology with dHU was screened out. Amino acid sequence alignment and homology modeling revealed certain clues to the increased activity of the chimera.

2. Materials and methods 2.1. Microorganisms, vectors and materials Plasmid vectors pET-22b(+)/wPU and pET-22b(+)/dHU were constructed for amplification of the parental genes used in DNA shuffling. pP43NMK [25], a generous gift from Dr. Xiao-Zhou Zhang (Virginia Polytechnic Institute and State University, Blacksburg, USA), was used to amplify nprB signal peptide-encoding sequence. pP43X [26] was reconstructed from pP43NMK and used as a shuttle vector to achieve secretory production of uricase in B. subtilis

WB800. E. coli DH5␣ was used for gene amplification of recombinant plasmid. Restriction endonucleases, T4 DNA ligase and Agarose Gel DNA Purification Kit Ver.2.0. were purchased from TaKaRa (TaKaRa Ltd., Dalian, China). DNase I and Taq DNA polymerase were from Fermantas. 200 bp DNA ladder marker and protein marker were from Transgen (TransGen Biotech, Beijing, China). 2.2. Medium composition and culture conditions Inocula of E. coli and Bacillus strains were cultured at 37 ◦ C and 200 rpm in Luria-Bertani (LB) medium containing 0.5% (w/v) yeast extract, 1% (w/v) Tryptone and 1% (w/v) NaCl. To prepare the solid medium, 2% agar was added. When appropriate, ampicillin (Amp; 100 ␮g/ml for E. coli) or kanamycin (Kan; 25 ␮g/ml for Bacillus) was added to the LB medium. The screening medium was supplemented with 0.1% uric acid. 2.3. Construction of pET-22b(+)/wPU Reverse transcription (RT)-PCR was performed on total RNA extracted from porcine liver cells using High Fidelity Prime Script® RT-PCR Kit according to the manufacturer’s protocol. PCR was carried out with the generated cDNA as a template to amplify ORF of wPU gene. Primer 1 and 2 with introduction of Nde I and Hind III site respectively were designed and used in PCR amplification. PCR conditions were as following: one cycle of initial denaturation at 94 ◦ C for 2 min, followed by 35 cycles at 94 ◦ C for 30 s, 55 ◦ C for 30 s, and 72 ◦ C for 1 min, followed by a final extension at 72 ◦ C for 10 min. The PCR products were separated by agarose gel electrophoresis and the resulting fragment was inserted into pET-22b(+) vector to obtain recombinant plasmid pET-22b(+)/wPU. After amplification in E. coli DH5␣, the target gene was confirmed by DNA sequencing. 2.4. Construction of pET-22b(+)/dHU A codon-optimized full length dHU gene with two premature mutations at codon 33 and 187 replaced by CGA was designed and synthesized based on the protein sequence of HU. Restriction sites Nde I and Hind III were introduced at the 5 -and 3 -terminus respectively. After digestion and ligation, dhu was inserted into the multiple cloning site of pET-22b(+) vector. After amplification in E. coli DH5␣, dhu was used as one of the parental gene in DNA shuffling. 2.5. Construction of chimeric library wPU gene and dHU gene were amplified from pET-22b(+)/wPU and pET-22b(+)/dHU using primers 3/4 and 5/6, respectively. The reaction mixture contained 10ng plasmid, 60 pmol of each primer, 0.2 mM dNTP mix and 2.5 U Pfu DNA polymerase in 1 × Pfu buffer containing 2 mM MgCl2 . Fragments about 900 bp were obtained and purified by Agarose Gel DNA Purification Kit. Equal amount of the two gene preparations were mixed and digested with DNase I. The digestion mixture contained 0.5 U DNaseI, 0.05 ␮g of each parent DNA, 4.25 ␮l of 0.2 M Tris-HCl, pH 7.5 buffer, 0.25 ␮l of 10 mg/ml BSA, and 0.25 ␮l of 0.1 M MnCl2 . After digestion at 18 ◦ C for 25 min and subsequent heat inactivation at 90 ◦ C for 10 min, DNA fragments with random lengths were obtained and subjected to electrophoresis on 2% low melting point agarose gels. The desired DNA fragments between 50 and 100 bp were isolated and purified. Primerless PCR was conducted in a Bio-Rad Cycler as follows: 94 ◦ C for 1 min, 40 cycles of 94 ◦ C for 30 s, 50 ◦ C for 30 s and 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min. The product of primerless PCR was used as a template to amplify the full-length genes using primer 3 and 8 (with introduction of Xba I). The PCR program was

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model

ARTICLE IN PRESS

BIOMAC-5467; No. of Pages 8

J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx Table 1 Oligonucleotide Primers for gene cloning. Primer

Sequence(5 -3 )

P1 P2 P3 P4 P5 P6 P7 P8 P9

GCCAATTCCATATGGCTCATTACCGTAATGAC CGATAAGCTTTCACAGCCTTGAAGTCAGC ATGGCTCATTACCGTAATGAC TCACAGCCTTGAAGTCAGC ATGGCCCACTACCATAACAAC TCACAGTCTTGAAGACAACTTCC AAACTGCAGATGCGCAACTTGACCAAGACA GCTCTAGATCACAGTCTTGAAGACAACTTCC TCATTACGGTAATGAGCCATTGCAGCTGAGGCAT

94 ◦ C for 1 min, 40 cycles of 94 ◦ C for 30 s, 50 ◦ C for 30 s and 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min. Primer 7(with introduction of PstI) and 9 were designed for the cloning of nprB signal peptide encoding sequence from plasmid pP43NMK. Fragments from DNA shuffling and DNA sequence of the signal peptide were spliced by OE-PCR [27] with the following twostep program: 94 ◦ C for 3 min, 10 cycles of 94 ◦ C for 30 s, 55 ◦ C for 1 min, 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min; addition of primer 7 and 8, followed by 30 cycles of 94 ◦ C for 30 s, 65 ◦ C for 1 min, 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min. To generate mutant library, the OE-PCR products were digested with Pst I and Xba I before purification and ligation to vector pP43X which was digested by the same enzyme. The ligation product was transformed to protease-deficient strain B. subtilis WB800 by electroporation (Table 1). 2.6. Screening for active chimeras Transformed B. subtilis WB800 were plated on a selective medium (solid LB medium supplemented with 25 ␮g/ml kanamycin and 0.1% uric acid) and incubated for 24 h at 37 ◦ C and further incubated for 5–7 days at 4 ◦ C. If active uricase is secreted into the medium, a transparent halo will be produced around the producing colony. A small number of active chimeras from the first round of screening were picked out and heterogeneously amplified in the same flask. The plasmid DNA was extracted and used in the second round of shuffling. 2.7. Expression and activity assay of wPU, dHU and chimeric uricase wPU gene and dHU gene were amplified from PET-22b(+)/wPU and PET-22b(+)/dHU respectively with introduction of restrictive site Xba I at the 3 -terminus, followed by ligation with DNA sequence of the signal peptide from pP43NMK by OE-PCR. After double digestion and ligation, the target gene was inserted into pP43X. The ligation product was transformed to protease-deficient strain B. subtilis WB800 for secretary production of uricase. Transformed B. subtilis WB800 was first inoculated in liquid LB medium and allowed to grow overnight at 37 ◦ C and 200 rpm. One milliliter of the inocula was then transferred into 50 ml of the growth medium consisting of 50 g L−1 glucose, 20 g L−1 tryptone, 50 g L−1 oil cake, 5 g L−1 (NH4 )2 SO4 , 8 g L−1 Na2 HPO4 , 2 g L−1 KCl and 1 g L−1 MgSO4 in 500 ml shake flask. Cultivation was conducted at 37 ◦ C for 24 h. Chimeric mutants displaying relatively larger halos on the selective plate were also picked up for uricase fermentation. After fermentation, supernatant was harvested by centrifugation at 10,000 × g for 10 min at 4 ◦ C. Solid ammonium sulfate was added to the recovered supernatant to 10% saturation at 4 ◦ C. The precipitate was collected by centrifugation for 20 min at 10,000 ×g. The precipitate was washed by double distilled water and then dissolved in Na2 CO3 -NaHCO3 buffer (0.1 M, pH 10.3). The dissolved

3

solution was loaded onto an anion exchanger (Q-Sepharose Fast Flow) pre-equilibrated with Na2 CO3 -NaHCO3 buffer (0.1 M, pH 10.3). The bound protein fractions were eluted using the same buffer containing 0.5 M NaCl. Fractions showing uricolytic activities were pooled from Q-Sepharose Fast Flow and subjected to gel filtration on a Sephadex S-300 column pre-equilibrated with Na2 CO3 -NaHCO3 buffer (0.1 M, pH 10.3). After elution with the same buffer, fractions showing uricolytic activity were collected and dialyzed against de-ionized water, freeze-fried and stored at 4◦ C for further analysis. The uricolytic activity of Uox was measured by the decrease in absorbance at 293 nm due to enzymatic oxidation of uric acid as described previously [28]. Enzyme activity was carried in 0.1 M sodium borate buffer (pH8.6) at 37 ◦ C. An extinction coefficient of 12,300 M−1 cm−1 for uric acid was used [29]. One unit (U) of enzymatic activity is defined as the amount of enzyme that catalyzes the oxidation of 1 ␮mol of uric acid per minute. Total protein was measured by the Bradford method [30]. Specific activity was calculated for comparison. Kinetic parameters Km and Vmax were determined using different concentrations of uric acid according to LineweaverBurk plot. The turn over number (Kcat ) was calculated based on the value of Vmax , concentration of the purified enzyme as well as the molecular weight. The catalytic efficiency was calculated from Kcat /Km . 2.8. Sequence analysis and homology modeling The chimeras with improved uricolytic activity and increased homology with human uricase after the second round of shuffling was aligned with dHU and wPU for nucleotide sequence and amino acid sequence by CLUSTAL X program to identify substituted residues. In order to investigate how the altered amino acid affected the function of uricase, the structure of Uox from Aspergillus flavus [31] was used as a template to model the 3-D structures of dHU, wPU and porcine-human chimeric uricase by using MOE 2010.10 (Chemical Computing Group Inc., Montreal, Canada). 3. Results 3.1. Construction of pET-22b(+)/PU and pET-22b(+)/dHU cDNA of wPU was obtained by RT-PCR on total RNA obtained from porcine liver cells. Primer 1and 2 were used to amplify wPU gene by PCR, resulting in an approx.900 bp DNA fragment. After digestion and ligation, the DNA fragment was inserted into pET22b(+). DNA sequencing confirmed that wPU gene was successfully cloned. dhu was synthesized and inserted into pET-22b (+) vector. The recombinant plasmids were transferred into E. coli DH5␣ for amplification. 3.2. DNA shuffling and screening of the library wpu and dhu were amplified from pET-22b(+)/PU and pET 22b(+)/dHU respectively. After digestion with DNase I, fragments of 50–100 bp were obtained and reassembled by primerless PCR. The reassembled products were used as templates to amplify the fulllength gene in the presence of primer 3 and 8(with introduction of Xba I). As a result, a clear gene band on agarose gel corresponding to the expected size was obtained. The detection of products in the process of DNA shuffling is shown in Fig. 1. DNA sequence of the signal peptide from pP43NMK was introduced at the 5’-terminus of the full length uricase gene by overlap extension PCR. After double digestion and ligation into vector pP43X, the chimeric library was constructed into B. subtilis WB800. As an irrational way of molecular engineering, DNA shuffling does not require an in-depth knowledge of the structural features

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model BIOMAC-5467; No. of Pages 8 4

ARTICLE IN PRESS J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

Fig. 1. Construction of chimeric library a. Fragments encoding wPU and dHU were amplified by PCR. b. 50–100 bp fragments obtained by digesting the parental genes with 0.5U DNase I for 25 min at 18 ◦ C. c. Purified fragments were reassembled by primerless PCR. d. A single band of PCR product with correct size was obtained after PCR amplification with primers.

of the enzyme but usually suffers from higher cost and longer time in generating and screening mutant libraries. Because large numbers of mutants must be generally screened to obtain a significant, desired effect on enzyme activity, the development of a high-throughput screening methodology that allows identification of desired property is of the utmost importance. Not all enzyme activities are readily amenable to developing a high-throughput screening method, nor are all screening methodologies easy to implement at the required scale [32]. Until now, efficient, validated screening methods for uricase mutants are unavailable [33]. In this study, a high throughput method was developed for screening active uricase mutants. The water solubility of uric acid is rather low (6 mg/dl, 20 ◦ C). When a mass of uric acid above the saturation point is added into the culture medium to prepare agar plate, crystals of uric acid will suspend and make the plate opaque. If active uricase is secreted by colonies growing on this plate, the enzyme will convert uric acid into water-soluble allantoin. Consumption of uric acid crystals in the agar plate leads to the formation of transparent halos around the colonies harboring active chimeras. One plate can be used to screen approx.100 colonies. Since larger transparent halos generally indicate higher enzymatic activity, the selective plate can also be used for a rapid estimation of enzymatic activity. Transformed B. subtilis WB800 was plated on the selective medium supplemented with 25 ␮g/ml kanamycin and 0.1% uric acid. After incubation at 30 ◦ C for 1 day and further incubation at 4 ◦ C for 5–7days, halos of variable sizes were observed around the single colony producing active mutant uricase (Fig. 2). Incubation at 4 ◦ C was strategically designed for two purposes. Firstly, low temperature will further decrease the solubility of uric acid in the agar plate so that the opaque layer formed by the crystals will be very obvious. Secondly, low temperature can slow down the growth of bacteria. It takes time for the secreted uricase to diffuse through the solid medium and transform uric acid to produce transparent halos. If the bacteria are allowed to grow under 30◦ C, fast–growing colonies may cover the transparent halos and thus interfere with identification of active chimeras and estimation of uricolytic

Fig. 2. Halos formed around single colonies producing active mutant uricase. The chimeric uricase library was plated onto the selective plate. After incubation at 37 ◦ C for 24 h and 4 ◦ C for additional 5–7 days, transparent halos (indicated by arrow) formed around the single colonies producing active mutant uricase.

activity. Among the 2047 colonies grown on the selective plate in the second round of shuffling, about 30% formed clear halos, indicating the secretary expression of active uricase. Approximately 100 colonies with relatively larger halos were selected. 3.3. Expression and activity assay of mutant uricase The 100 selected colonies were inoculated into LB medium for growth at 37 ◦ C for 10 h. One milliliter of the inocula was then transferred into 50 ml of the growth medium in 500 ml shake flask. Cultivation was conducted at 37 ◦ C for 30 h. As a preliminary screening, activity of each mutant was roughly estimated by performing enzyme assay on supernatant of the fermentation broth. More accurate assay of the top 20 mutants from the preliminary screening was further conducted after purification(See supplementary data). Because a signal peptide was constructed at the N-terminus of uricase, secretory production of uricase was achieved, which was advantageous for downstream purification. After precipitation with ammonium sulfate, anion exchange and gel filtration, a single band of uricase was obtained by electrophoresis (Fig. 3), suggesting that the enzyme was purified to homogeneity. For parallel comparison of enzyme activity, wpu and dhu were also ligated with DNA sequence encoding the signal

Fig. 3. Homogeneity of purified wPU and chimera-62 by SDS-PAGE analysis. Lane 1: purified wPU; lane 2: purified chimera-62; lane M: molecular weight markers.

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model

ARTICLE IN PRESS

BIOMAC-5467; No. of Pages 8

J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

5

Table 2 Characteristics of wHU, dHU and chimera-62. Uricase

Km (␮M)

Kcat (s−1 )

Kcat /Km (s−1 ␮M−1 )

Specific activity (U/mg)

Homology with dHU(%)

dHU wPU Chimera-62

ND 10.12 ± 0.10 9.43 ± 0.04

ND 23.54 ± 0.03 25.17 ± 0.08

ND 2.33 2.67

ND 4.32 ± 0.07 4.71 ± 0.03

100 87.5 89.1

Values are means ± standard deviation of three replicates; ND, not determined.

peptide from pP43NMK, followed by transformation into proteasedeficient strain B. subtilis WB800 for production and enzyme assay. All purified enzymes were subjected to activity assay. As shown in Table 2, it is consistent with previous report that dHU did not exhibit any catalytic activity, confirming that human uricase cannot be resurrected simply by replacing the two premature stop codons 33 and 187 present in the gene. For the mutants with relatively high enzymatic activity, their encoding genes were sequenced and deduced amino acid sequences were obtained. Based on amino acid sequence alignment, their homology with dHU was calculated. The most satisfactory result was observed with chimera-62 which was more active than wPU, with a smaller Km of 9.43 ± 0.04 ␮M and a higher catalytic efficiency of 2.67 s−1 ␮M−1 . At the same time, it exhibited a higher homology (89.1%) with dHU than wPU(87.5%). Coincidently, chimera-62 was nearly the same as the FDA approved drug Pegloticase with only one amino acid difference at position 249. Apart from mammals, uricase has also been found in plant [34] and microbial cells including Candida utilis [35], Bacillus sp.TB90 [36], Arthrobacter globiformis [37] and Aspergillus flavus[38]. Although uricases of microbial origin exhibit much higher activity than those of mammalian origin, they were not selected to serve as the parental gene in DNA shuffling due to their low level of sequence identity with dHU. In the ancestral uricase enzymes resurrected by Kratzer et al., notable, but surprisingly gradual, decreases in catalytic efficiency were observed along with the evolution of hominoid primates until activity was completely abolished in the ancestor of the great apes [39]. Among those active mammalian uricases, enzymes from crab-eating macaque, rhesus macaque and baboon are more homologous to dHU than wPU. However, their activities are much lower than that of wPU which was determined to be the most active among mammalian uricases [40]. Since a more active enzyme is preferred in terms of therapeutic efficacy, wild-type porcine uricase was selected among other mammalian uricases as one of the parental gene for DNA shuffling. The selection of dhu as the parental gene aimed to obtain chimeras with high homology with HU. This is also the first report of DNA shuffling based on a pseudogene. Since dHU is inactive, a second round of shuffling with inclusion of active chimeras obtained in the first round as parental genes was performed in order to increase the percentage of active“hits”. Compared with wPU, chimera-62 was more“human-like”. Although the increase in activity and catalytic efficacy were not so dramatic, the simultaneous increase in the homology made any marginal increase in activity especially important. All these results suggested that chimera-62 was a very promising drug candidate for hyperuricemia and gout.

the identities of deduced amino acid sequence, and the exon in which the mutation occurs when compared with dHU and wPU were respectively presented in Table 3 and Table 4. There are 33 amino acid substitutions in chimera-62 when compared with dHU, while only 5 substitutions when compared with wPU. All the mutations are derived from either of both parents, indicating chimera-62 was well shuffled (Fig. 4). It is interesting to note that amino acid substitutions compared with wPU were concentrated in the last three exons 6-8 while amino acid substitutions compared with dHU distributed across exon1-6. It was well established that a progressive accumulation of slightly deleterious amino acid replacement led to the inactivation and pseudogenization of human uricase. From the most ancient of the ancestral uricase with highest enzymatic activity to human uricase, there are 22 amino acid replacements. Nearly all Table 3 DNA and amino acid substitutions in chimera-62 compared with dhu. DNA substitution

Amino acid substitution

Property

Exon

CAT→CGT AAC→GAC GAA→GAT GTA→ATA ATC→GTC CAT→AAT GAA→GGC GGC→ACT GGT→GCT AAT→ACT AAC→AAG ATG→GTG ATC→CTT CAT→CGT CTT→TTT GGA→GAA CAC→ACT CTG→ATA AGT→AAT CAA→CCA TGC→GGC AAG→GAG ATT→GTT GAC→AGC CTT→ATT ATG→CTG GAG→CAG TCT→TTT TTG→TCG ACC→CCC TGT→TAT CGA→CAG GCG→GAG

5H→5R 7N→7D 24E→24D 26V→26I 65I→65V 76H→76N 83E→83G 89A→89T 91G→91A 93N→93T 103N→103K 112M→112V 115I→115V 119H→119R 120L→120F 121G→121E 133H→133Y 146L→146I 148S→148N 151Q→151P 202C→202G 208K→208E 214I→214V 216D→216S 217L→217I 219M→219L 220E→220Q 222S→222F 232L→232S 233T→233P 240C→240Y 249R→249Q 252A→252E

Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination Recombination

1 1 2 2 2 2 2 3 3 3 3 3 3 3 3 3 4 4 4 5 5 5 6 6 6 6 6 6 6 6 6 6 7

3.4. Sequence analysis of shuffled chimera-62 Chimera-62 with increased uricolytic activity and higher homology with dHU was compared to the parental genes in terms of both nucleotide and amino acid sequence. Numerous nucleotide substitutions were identified by nucleotide sequence alignment, although many of them resulted in silent mutations. However, a number of amino acid residue substitutions were observed especially in the comparison with dHU. The mutant DNA sequences,

Table 4 DNA and amino acid substitutions in chimera-62 compared with wPU. DNA substitution

Amino acid substitution

Property

Exon

ACC→TCC GGC→AGC TTA→TTC AGG→AAA ACT→TCT

246T→246S 248G→248S 266L→266F 291R→291K 301T→301S

Recombination Recombination Recombination Recombination Recombination

6 6 7 8 8

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model BIOMAC-5467; No. of Pages 8 6

ARTICLE IN PRESS J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

Fig. 4. Protein sequence alignment of the shuffled chimera-62 with the two parents. Conserved residues are indicated in white on black background. Amino acid residues that are identical with only one parent are indicated in black on gray background.

these mutations are located across exon1-6 and cause a substantial decrease in catalytic efficiency of uricase, some of them nearly inactivate the enzyme(S232L, Y240C and F222S) [41]. As a result, shuffling in exon1-6 was likely to cause deteriorating effect on the enzyme activity and chimeras containing amino acid replacement in these exons were less likely to show increased activity compared with wPU. This provides a reasonable explanation to the result that chimera-62 with high uricolytic activity sparsely contains amino acid replacement in exon 1–5. 3.5. Homology modeling and analysis of underling mechanisms of increased activity Homology modeling of chimera-62 showed very similar results with previous studies. Uricase is a homotetrameric enzyme with the existence of a small architectural domain that is called tunnelling-fold (T-fold). The T-fold assembles to form a perfect unusual dimeric alpha 8 beta 16 barrel [41]. The functional enzyme has four active sites located at the interface within the same dimer, which is divided into two parts: the uric acid-binding site and the oxygen-binding site. The uric acid-binding site includes

Arg187, Val235 and Gln236 from the first monomer and Thr68 and Asp69 from the other monomer. The oxygen-binding site is located above the uric acid-binding site and contains Asn262 and His264 from the first monomer and Lys23 and Thr68 from the other monomer [42]. However, none of the amino acid replacements taking place in chimera-62 compared with wPU is related to these active sites. The amino acid replacements in Table 4 were individually analyzed by replacing the original amino acid residue in wPU and predicting their potential relation with the increased activity. As shown in Fig. 5a, amino acid residue 248(glycine) of wPU is in the secondary structure of ␣-helix. It was shuffled to serine in chimera-62 (Fig. 5b). Different amino acids have different propensities for forming ␣-helix. With a helical propensity of 1.00 Kcal/mol (alanine arbitrarily set as zero), glycine is one the amino acids that tend to disrupt helices because its high conformational flexibility makes it entropically expensive to adopt the relatively constrained ␣-helical structure [43]. In shuffled chimera-62, 248G→248S is considered as a favorable mutation since serine has a lower helical propensity (0.5 Kcal/mol), which is favorable to the stability of the ␣-helix and thus to the enzymatic activity.

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model BIOMAC-5467; No. of Pages 8

ARTICLE IN PRESS J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

7

Fig. 5. Part of the tertiary structure of wPU (a) and the shuffled uricase chimera-62 (b) constructed by Swiss-model server. Amino acid residue 248 is particularly indicated. ␣-helix is shown in red, ␤-sheet is shown in yellow.

Fig. 6. Tertiary structure of wPU (left) and the shuffled chimera-62 (right) constructed by Swiss-model server. Hydrogen binding area (in pink), hydrophobic area (in green) and mild polar area (in blue) are indicated.

Another amino acid replacement that is speculated to confer increased activity of chimera-62 is L266F. As shown in Fig. 6, amino acid residue 266 is located on the surface of uricase. As a highly hydrophobic protein, the activity of uricase is largely dependent on its stability and solubility in aqueous solution. Active uricases of different origins are all homotetramers associated by non-covalent binding. Any factor causing the disassociation of the tetramer such as extreme pHs will lead to exposure of inner hydrophobic amino acid residues and final precipitation of the enzyme, significantly deteriorating the enzymatic activity. Decreasing surface hydrophobicity is considered one of the strategies to improve the stability of hydrophobic protein in aqueous solution. In the shuffled chimera62, amino acid 266 was shuffled from leucine to phenylalanine. Due to the conjugated effect of benzene ring, an increased polarity is observed with phenylalanine compared with leucine. Accordingly, mild polar area (indicated in blue in Fig. 6) increases around Phe266 in chimera-62, which is favorable to the solubility and stability of the enzyme and thus to its activity. 4. Conclusions wPU and dHU were cloned and used as parental genes to perform two-round DNA shuffling in order to obtain chimeric uricase with improved enzymatic activity and increased homology with dHU. All the coding sequences were ligated with a signal peptide from pP43NMK and transformed into B. subtilis WB800 for secretory production. A high throughput screening method was developed by using LB medium supplemented with 25 ␮g/ml kanamycin and 0.1% uric acid. Chimeras with productivity of active uricase created transparent halos around the producing colony. After shuffling and screening, about 30% of the 2047 colonies grown on the selective plate formed clear halos.

Approximately 100 colonies with relatively larger halos were selected for enzyme assay. Chimeras showed high activity in preliminary assay screening as well as wPU and dHU were subjected to purification for parallel comparison of activity and catalytic efficiency. Chimera-62 with both improved activity and increased homology with dHU was finally screened out. Sequence analysis of chimera-62 and its alignment with parental genes indicated 33 amino acid substitutions (distributed across exons 1-6) when compared with dHU and 5 substitutions (concentrated in exons 6–8) when compared with wPU. This was consistent with the conclusions from previous studies that the liable mutations gradually inactivating dHU during the evolution of hominoid primates mainly accumulated in exon 1–6. Chimeras with amino acid substitutions in these exons are less likely to show high enzymatic activity. By homology modeling and 3-D structure analysis, the mechanisms of increased activity observed in chimera-62 compared with wPU were analyzed. The mutation G248S was speculated to increase the stability of the ␣-helix harboring this mutation due to a much lower helical propensity of serine than glycine. The mutation L266F tended to increase the surface polarity of the hydrophobic enzyme, which was favorable to solubility and thus activity. Acknowledgments This study was supported by National Natural Science Foundation of China (No. 308731192 and 81072560), Natural Science Foundation of Jiangsu Province of China (No.SKB2014021911) and “111 Project” from the Ministry of Education of China and State Administration of Foreign Experts Affairs of China (No. 111-207). This study was also supported by A Project Funded by the

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053

G Model BIOMAC-5467; No. of Pages 8

ARTICLE IN PRESS J. Chen et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

8

Priority Academic Program Development of Jiangsu Higher Education Institutions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ijbiomac.2015. 10.053. References [1] P.A. Simkin, Uric acid metabolism in Cebus monkeys, Am. J. Physiol. 221 (1971) 1105–1109. [2] D.C. Logan, D.E. Wilson, C.M. Flowers, P.J. Sparks, F.H. Tyler, Uric acid catabolism in the woolly monkey, Metabolism 25 (1976) 517–522. [3] P. Christen, W.C. Peacock, A.E. Christen, W.E. Wacker, Urate oxidase in primate phylogenesis, Eur. J. Biochem. 12 (1970) 3–5. [4] A.D. Sarma, P. Serfozo, K. Kahn, P.A. Tipton, Identification and purification of hydroxyisourate hydrolase, a novel ureide-metabolizing enzyme, J. Biol. Chem. 274 (1999) 33863–33865. [5] S. Hayashi, S. Fujiwara, T. Noguchi, Evolution of urate-degrading enzymes in animal peroxisomes, Cell. Biochem. Biophys. 32 (2000) 123–129. [6] I. Ramazzina, C. Folli, A. Secchi, R. Berni, R. Percudani, Completing the uric acid degradation pathway through phylogenetic comparison of whole genomes, Nat. Chem. Biol. 2 (2006) 144–148. [7] M. Oda, Y. Satta, O. Takenaka, N. Takahata, Loss of urate oxidase activity in hominoids and its evolutionary implications, Mol. Biol. Evol. 19 (2002) 640–653. [8] F. Roch-Ramel, G. Peters, Urinary excretion of uric acid in nonhuman mammalian species, Handb. Exp. Pharmacol. 51 (1978) 211–255. [9] C.A. Sarawate, K.K. Brewer, W. Yang, P.A. Patel, H.R. Schumacher, K.G. Saag, A.W. Bakst, Gout medication treatment patterns and adherence to standards of care from a managed care perspective, Mayo. Clin. Proc. 81 (2006) 925–934. [10] L. Annemans, E. Spaepen, M. Gaskin, M. Bonnemaire, V. Malier, T. Gilbert, G. Nuki, Gout in the UK and Germany: prevalence, comorbidities and management in general practice 2000–2005, Ann. Rheum. Dis. 67 (2007) 960–966. [11] R. Terkeltaub, Update on gout: new therapeutic strategies and options, Nat. Rev. Rheumatol. 6 (2010) 30–38. [12] C.M. Burns, R.L. Wortmann, Gout therapeutics: new drugs for an old disease, Lancet 377 (2011) 165–177. [13] K. Fan, C. Zhang, X. Ma, X. Mei, C. Hu, Humanized recombinant uricase and mutants thereof, 2012, US Patent 13/504,065. [14] J. Rong, H. Kuang, Z. Sun, Preparation and application method of PEG recombinant pig-human urate oxidase fusion protein, 2011, CN Patent 201110181447.X. [15] J.S. Sundy, M.A. Becker, H.S. Baraf, A. Barkhuizen, L.W. Moreland, W. Huang, R.W. Waltrip, A.N. Maroli, Z. Horowitz, Reduction of plasma urate levels following treatment with multiple doses of pegloticase (polyethylene glycol-conjugated uricase) in patients with treatment-failure gout: results of a phase II randomized study, Arthritis. Rheum. 58 (2008) 2882–2891. [16] N. Schlesinger, U. Yasothan, P. Kirkpatrick, Pegloticase, Nat. Rev. Drug. Discov. 10 (2011) 17–18. [17] W.P. Stemmer, DNA shuffling by random fragmentation and reassembly: in vitro recombination for molecular evolution, Proc. Natl. Acad. Sci. USA 91 (1994) 10747–10751. [18] J.E. Ness, M. Welch, L. Giver, M. Bueno, J.R. Cherry, T.V. Borchert, W.P. Stemmer, J. Minshull, DNA shuffling of subgenomic sequences of subtilisin, Nat. Biotechnol. 17 (1999) 893–896. [19] G. Wu, G. Wang, J. Ji, Y. Li, H. Gao, J. Wu, W. Guan, A chimeric vacuolar Na+ /H+ antiporter gene evolved by DNA family shuffling confers increased salt tolerance in yeast, J. Biotechnol. 203 (2015) 1–8. [20] J.B. Behrendorff, W.A. Johnston, E.M. Gillam, Restriction enzyme-mediated DNA family shuffling, Methods Mol. Biol. 1179 (2014) 175–187.

[21] D.E. Stephens, F.I. Khan, P. Singh, K. Bisetty, S. Singh, K. Permaul, Creation of thermostable and alkaline stable xylanase variants by DNA shuffling, J. Biotechnol. 187 (2014) 139–146. [22] S.K. Singh, C. Heng, J.D. Braker, V.J. Chan, C.C. Lee, D.B. Jordan, L. Yuan, K. Wagschal, Directed evolution of GH43 ␤-xylosidase XylBH43 thermal stability and L186 saturation mutagenesis, J. Ind. Microbiol. Biotechnol. 41 (2014) 489–498. [23] B. Niederhausera, J. Siivonen, J.A. Määttäa, J. Jänis, M.S. Kulomaa, V.P. Hytönen, DNA family shuffling within the chicken avidin protein family – a shortcut to more powerful protein tools, J. Biotechnol. 157 (2012) 38–49. [24] S. Fujiwara, K. Nakashima, T. Noguchi, Insoluble uricase in liver peroxisomes of Old World monkeys, Comp. Biochem. Physiol. B 88 (1987) 467–469. [25] X.Z. Zhang, Z.L. Cui, Q. Hong, S.P. Li, High-level expression and secretion of methyl parathion hydrolase in Bacillus subtilis WB800, Appl. Environ. Microb. 71 (2005) 4101–4103. [26] J. Chen, X. Chen, J. Dai, G. Xie, L. Yan, L. Lu, J. Chen, Cloning, enhanced expression and characterization of an ␣-amylase gene from a wild strain in B. subtilis WB800, Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j. ijbiomac. [27] S.N. Ho, H.D. Hunt, R.M. Horton, J.K. Pullen, L.R. Pease, Site-directed mutagenesis by overlap extension using the polymerase chain reaction, Gene 77 (1989) 51–59. [28] T.G. Conley, D.G. Priest, Thermodynamics and stoicheiometry of the binding of substrate analogues to uricase, Biochem. J. 187 (1980) 727–732. [29] H.M. Kalckar, Differential spectrophotometry of purine compounds by means of specific enzymes; determination of hydroxypurine compounds, J. Biol. Chem. 167 (1947) 429–443. [30] M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. [31] P. Retailleau, N. Colloc’h, D. Vivarès, F. Bonneté, B. Castro, M. El Hajji, T. Prangé, Urate oxidase from Aspergillus flavus: new crystal-packing contacts in relation to the content of the active site, Crystallogr. Acta D Biol. Crystallogr. 61 (2005) 218–229. [32] R.A. Chica, N. Doucet, J.N. Pelletier, Semi-rational approaches to engineering enzyme activity: combining the benefits of directed evolution and rational design, Curr. Opin. Biotechnol. 16 (2005) 378–384. [33] X. Yang, Y. Yuan, C. Zhan, F. Liao, Uricases as therapeutic agents to treat refractory gout: current states and future directions, Drug. Dev. Res. 73 (2012) 66–72. [34] H. Suzuki, D.P. Verma, Soybean nodule-specific uricase (Nodulin-35) is expressed and assembled into a functional tetrameric holoenzyme in Escherichia coli, Plant. Physiol. 95 (1991) 384–389. [35] H. Nishimura, K. Yosida, Y. Yokota, A. Matsushima, Y. Inada, Physicochemical properties and states of sulfhydryl groups of uricase from Candida utilis, J. Biochem. 91 (1982) 41–48. [36] K. Yamamoto, Y. Kojima, T. Kikuchi, T. Shigyo, K. Sugihara, M. Takashio, S. Emi, Nucleotide sequence of the uricase gene from Bacillus sp. TB-90, J. Biochem. 119 (1996) 80–84. [37] K. Suzuki, S. Sakasegawa, H. Misaki, M. Sugiyama, Molecular cloning and expression of uricase gene from Arthrobacter globiformis in Escherichia coli and characterization of the gene product, J. Biosci. Bioeng. 98 (2004) 153–158. [38] P. Laboureur, C. Langlois, Urate oxidase of Aspergillus flavus. I. Isolation, purification, properties, Bull. SOC Chim. Biol. 50 (1968) 811–825. [39] J.T. Kratzer, M.A. Lanaspa, M.N. Murphy, C. Cicerchi, C.L. Graves, P.A. Tipton, E.A. Ortlund, R.J. Johnson, E.A. Gaucher, Evolutionary history and metabolic insights of ancient mammalian uricases, Proc. Natl. Acad. Sci. USA 111 (2014) 3763–3768. [40] B.S. Chang, Ancient insights into uric acid metabolism in primates, Proc. Natl. Acad. Sci. USA 1111 (2014) 3657–3658. [41] N. Colloc’h, M. el Hajji, B. Bachet, G. L’Hermite, M. Schiltz, T. Prangé, B. Castro, J.P. Mornon, Crystal structure of the protein drug urate oxidase-inhibitor complex at 2.05 A resolution, Nat. Struct. Biol. 4 (1997) 947–952. [42] C. Zhang, K. Fan, W. Zhang, R. Zhu, L. Zhang, D. Wei, Structure-based characterization of canine-human chimeric uricases and its evolutionary implications, Biochimie 94 (2012) 1412–1420. [43] C.N. Pace, J.M. Scholtz, A helix propensity scale based on experimental studies of peptides and proteins, Biophys. J. 75 (1998) 422–427.

Please cite this article in press as: J. Chen, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.10.053