Urate oxidase (uricase; EC 220.127.116.11; Uox) is responsible for the first step of degradation of uric acid into more water-soluble allantoin that can be more readily excreted through the kidneys [1
]. Uox is found in all three domains of life, but not in all five genera of hominoids (humans, chimpanzees, orangutans, gorillas, and gibbons). As a result, uric acid is the end product of purine metabolism in hominoids, resulting in a 10-fold increase of serum uric acid level compared to non-primate mammals [4
]. Uric acid production and excretion are normally balanced in humans. Hyperuricemia results from an imbalance between the rates of production and excretion of uric acid [6
]. Due to the low solubility (6 mg/dL, 20 °C) of uric acid, long-term hyperuricemia leads to destructive crystalline urate deposits around joints, in soft tissues, and in some organs, causing a number of disorders, including gout and urate nephropathy associated with tumor lysis syndrome [7
]. Studies have also shown that hyperuricemia in humans can increase the risk of cardiovascular diseases, chronic nephropathy, impaired renal function, hypertension, and stroke [9
Due to those severe complications of hyperuricemia, its therapeutic countermeasures have attracted great attention in the medical community. There are currently three major classes of drugs available for reducing uric acid level, namely, uricosuric drugs, uric acid synthesis inhibitors, and recombinant Uox preparations [11
]. Uricosuric agents promote uric acid excretion, but are ineffective if renal function is impaired. Allopurinol, a strong inhibitor of xanthine oxidase, is the mainstay of therapy in patients with tophaceous gout, renal insufficiency, leukemia, and some inherited disorders. However, some patients are refractory to the therapy and ~2% of patients receiving allopurinol develop allergic reactions, even severe hypersensitivity syndrome (~0.4%) [13
), a recombinant form of Uox from Aspergillus flavus
, is the first marketed Uox preparation for the treatment and prophylaxis of acute hyperuricemia resulting from tumor lysis syndrome in children with cancer [15
]. Its outstanding ability to decrease uric acid level and dissolve tophi makes Uox-based therapy a more promising strategy for the treatment of hyperuricemia [17
]. However, the clinical use of rasburicase is limited by its immunogenicity and short half-life [19
]. In 2010, pegloticase (Krystexxa), a PEGylated chimeric porcine-baboon Uox, was approved by the USA Food and Drug Administration (FDA) for treatment of chronic gout in adult patients refractory to conventional therapy [20
]. With a higher sequence homology to hypothetic human uricase and PEGylation, the immunogenicity of pegloticase is significantly reduced and half-life prolonged. In addition to the FDA approved porcine–baboon chimera (PBC), investigations on several other chimeric uricases have also been reported, such as canine–human chimeric uricase [22
] and porcine–human chimeric uricase [23
]. The development of therapeutic uricase for human use is an intractable challenge as activity, stability, and immunoreactivity should all be taken into consideration. As such, the medical community has a strong interest in developing a recombinant “human-like” uricase to treat hyperuricemia and gout [24
The pseudogene of human uricase was mapped to chromosome 1p22 [25
].The eight exons in dhu
collectively give a nucleotide sequence of 915 bp and code for 304 amino acids. Although this enzyme is lost in hominoid primates, it is crucial in controlling uric acid levels in other mammals. Porcine uricase, with eight exons and a high homology (87.5%) to human uricase, was detected to be the most active among mammalian uricases. In this report, exon replacement was performed on wild porcine uricase (wPU) gene with corresponding exon from dhu
to investigate the overall detrimental effect of each exon on the enzyme activity. Subsequently, exon restoration was performed on dhu
by replacing the three most deleterious exons with corresponding exons from wpu
. After further site-directed mutations at amino acid residues 24 and 83, chimeric uricase H1-2
(E24D & E83G) with increased activity and higher homology with dHU was obtained.
The buildup of serum urate in the blood (hyperuricemia) can have severe consequences for human health. Although Uox preparations are very promising therapies for hyperuricemia and gout, their clinical use has been restricted by immunogenicity, the same as other non-human proteins with useful therapeutic properties. In the development of therapeutic uricase for human use, there should be a balance between activity, stability, and immunoreactivity. Molecular engineering has been frequently applied to construct enzymes with desired properties [26
]. Due to the lack of detailed structure–activity correlation of uricase, rational design is not feasible for engineering the enzyme. Previously, we obtained a more human-like chimeric uricase by DNA shuffling, which was an irrational method of molecular engineering [29
were used as parental genes to generate a diverse chimeric library from which mutant chimeras with desired properties were selected. However, this molecular engineering strategy was relatively tedious and relied greatly on the availability of high-throughput screening methodology.
The activity of human uricase was gradually lost during hominoids’ evolution as a result of the accumulation of both nonsense mutations and missense mutations [30
]. It has been proved that human uricase cannot be resurrected simply by replacing the two premature stop codons 33 and 187 present in the pesudogene, leading to the speculation that elimination of other missense mutations are needed to restore the activity. Kratzer et al.
used site-directed mutagenesis to determine the amino acid substitutions responsible for the decreases in human uricase activity [31
]. From the ancestral uricase An19/22 to nonfunctional human uricase, there are 22 amino acids replacement (including codons 33 and 187); nearly all of them are deleterious to enzymatic activity. However, permutation and combination of the remaining 20 missense mutations will result in numerous scenarios, and the construction of these mutants and screening for active ones will become an insuperable task. In this study, we demonstrated successful development of a chimeric uricase with increased homology with dHU and activity through a combination of exon replacement/restoration and site-directed mutagenesis, which could be considered a semi-rational approach.
Silencing or pseudogenization of the human uricase gene is a result of multiple, independent evolutionary events. Although not protein-coding, a pseudogene represents a record of once-functional genetic characteristics. Despite increased sequencing and annotation of pseudogenes, they tend to be ignored. In our study, the sequence of human uricase pseudogene was used for the development of a drug candidate that is more human-like than the FDA-approved PBC, setting a good example for exploring the full potential of pseudogenes. With a high homology with dHU, wPU is also the most active among other mammalian uricases [32
]. Both wPU and dHU are composed of 304 amino acids, which are distributed into exons with exactly the same length. Each exon in wpu
can be considered as the functional counterpart of dhu
. Therefore, we decided to construct porcine–human chimeric uricase by exon replacement, which allowed for investigation of these mutations collectively within an exon. In the subsequent exon restoration, the activity of dHU was recovered by replacing those significantly deleterious exons. This semi-rational approach circumvents the tedious screening of the irrational approach and allows for the study of exon-based structure–activity correlation.
An exon replacement study showed that exon 6 had the greatest detrimental effect, leading to a total loss of activity. This was consistent with a previous study by Kratzer et al.
]. The three mutations during hominoids’ evolution with the largest deleterious effect on the catalytic activity of uricase are all distributed in exon 6—S232L, Y240C, and F222S. A single mutation of S232L nearly abolished the catalytic activity of uricase An19/22, the ancestral uricase with much higher activity than wPU. The other two mutations prevented uricase solubility and thus the determination of activity [30
]. It is very reasonable to conclude that mutations in exon 6 are the predominant reason for the inactivation of HU. From the most deteriorating to the least destructive, exons in dhu
follow this sequence: exon 6, exon 5, exon 3, exons 1–2, exon 4, and exons 7–8. It is interesting to note that exon 5 in dhu
was only two amino acids (positions 202 and 208) different from that of wPU; however, a much larger detrimental effect on enzymatic activity was produced than the collective effect of all the eight mutations existing in exon 3. Exons with a significant detrimental effect on human uricase were chosen to be replaced by the corresponding exons in wpu
to restore dHU activity. After replacement of three exons, moderate activity was detected for chimera H1-2
. However, this activity was too low to have any clinical significance.
In order to further increase the activity of porcine–human uricase to an acceptable level and at the same time retain a high degree of homology to dHU, exon restoration was not continued for exons 1–2, 4, and 7–8. An exon replacement study suggested that mutations in these five exons were generally not so detrimental to the enzyme activity. However, some of them may be more liablethan others. Our strategy was to identify the more liablemutations in these five exons by multiple sequence alignment and homology modeling. Sequence alignment of functional uricase and non-functional uricase revealed that 24E and 83E were shared by nonfunctional dHU and H1-2P3H4P5-6H7-8 with poor activity, while 24D and 83G were found in all other uricases in the alignment with relatively high activity. We hypothesized that 24D and 83G might be critical for enzyme activity, which was further predicted by molecular modeling. In our analysis of the modeled 3D structures, 24E in H1-2P3H4P5-6H7-8 results in increased distance between the two dimers of uricase homotetramer and is thus disadvantageous to enzyme activity. With a larger side chain than glycine, glutamic acid at position 83 increases the steric hindrance at the cornering between the two α-helixes of uricase monomer and thus adversely affects the accurate positioning of active residues of the enzyme. Based on these predictions, site-directed mutagenesis was performed to obtain chimera H1-2P3H4P5-6H7-8 (E24D & E83G), which was more homologous to dHU and more active than PBC.
4. Materials and Methods
4.1. Microorganisms, Vectors, and Materials
Host strain E. coli BL 21 StarTM (DE3) and vector pET-22b(+) were from Invitrogen (Carlsbad, CA, USA). DNA polymerase, DNA marker, T4 DNA ligase, and restriction endonucleases Nde I and Hind III were from Fermentas (Waltham, MA, USA). A polymerase chain reaction (PCR) amplification kit (including PCR buffer and dNTP mix) was obtained from Takara (Shiga, Japan). The plasmid mini kit I and PCR product recovery kit were purchased from Omega Bio-Tek (Norcross, GA, USA). Uric acid standard was from Sigma-Aldrich (St. Louis, MO, USA). All other reagents were of analytical grade.
4.2. Construction of dHU, wPU, and Porcine-Human Chimeras
A codon-optimized full length dHU gene with two premature mutations at codon 33 and 187 replaced by CGA (arginine) was designed and synthesized based on the hypothetic 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. The wPU gene was also chemically synthesized in a similar manner and constructed into a pET-22b(+) vector. A PBC gene was accidentally obtained in our previous study on DNA shuffling between wpu and dhu.
In the exon replacement study, each exon in wpu
was replaced by the corresponding exon in dhu
to investigate the combined effect of all mutations accumulated during evolution in one exon on the activity of uricase. The plasmid and strains in this study are shown in Table 2
. The procedures are briefly described by taking exon 3 as an example. The nucleotide sequence of exon 3 in dhu
was amplified using primers 15 and 16 in Table S1
. In order to substitute the corresponding exon in wpu
, two bulky fragments containing exons 1–2 and exons 4–8 were amplified from wpu
using primers 1 & 2 and 5 & 12. With overlapping sequences, the two bulky fragments were spliced, with exon 3 amplified from dhu
by overlap extension PCR (SOE-PCR). Replacement of other exons was carried out in a similar manner. In the subsequent exon restoration study, the three most deleterious exons in dhu
were replaced by the corresponding exons in wpu
one after another to gradually restore the activity of dHU. The primers used for splicing were designed according to the nucleotide sequence of each exon in dhu
by Primer Premier 5.0 and are shown in Table S1
. All the chimeric genes were confirmed by DNA sequencing. All the recombinant plasmids were transformed into E. coli
BL 21. Positive clones were screened on an LB plate supplemented with 100 μg/mL ampicillin.
4.3. Expression and Purification of Porcine–Human Chimeras (PHC), wPU, and PBC
For expression of each chimeric enzyme, a seed culture was prepared by overnight cultivation in an LB medium containing 100 μg/mL ampicillin and then inoculated into a fresh fermentation medium. After cultivation at 37 °C for 4 h, protein expression was induced by the addition of isopropyl-β-d-thiogalactoside (IPTG; 0.2 mM final concentration), followed by further cultivation for 6 h. Cells were harvested by centrifugation at 10,000× g for 10 min at 4 °C. The cells’ pellets were re-suspended in a lysis buffer and homogenized by sonication. The cell lysate was centrifuged at 10,000× g for 20 min at 4 °C to completely remove the cell debris. Solid ammonium sulfate was added to the recovered supernatant to 10% saturation at 4 °C. The precipitate was re-dissolved in an Na2CO3–NaHCO3 buffer (0.1 M, pH 10.3), loaded onto an anion exchanger (Q-Sepharose Fast Flow), and eluted using 0.5 M NaCl. Fractions showing uricolytic activities were pooled and loaded onto a Sephacryl S-300 column (GE Healthcare, Chicago, IL, USA). After elution with an Na2CO3–NaHCO3 buffer (0.1 M, pH 10.3), target fractions were collected and stored at 4 °C for further analysis.
4.4. Protein Analysis and Enzymatic Assay
SDS-PAGE was carried out to determine the homogeneity of purification and the molecular mass of the chimeras [33
]. Protein content was measured by the Bradford method with bovine serum albumin as a standard [34
]. The enzymatic activity of purified uricase was determined spectrophotometrically by monitoring the decrease of uric acid in absorbance at 293 nm as described previously [35
]. Solutions of uric acid were prepared in 50 mM sodium borate buffer pH 8.5 to a final concentration of 100 μM. Purified uricase was dissolved in the same buffer and mixed with the substrate. The enzymatic reaction was carried out at 37 °C for 3 min with monitoring of absorbance at 293 nm every 4 s. The maximum rate of decrease in the absorbance per minute was calculated. An extinction coefficient of 12,300 M−1
for uric acid was used [36
]. One unit (U) of enzymatic activity was defined as the amount of enzyme that consumes 1 μmol of uric acid per minute. Specific activity was expressed as U/mg. The kinetic parameters Km
of dHU, wPU, H1-2
(E24D & E83G), and PBC were estimated by the double reciprocal plot method. Using different concentrations of uric acid (0.001–0.144 mM), the enzyme activity was assayed as described above. The turnover number (kcat
) was calculated based on the value of Vmax
, the concentration of the purified enzyme, and the molecular weight. Km
was calculated by the Lineweaver–Burk plotting.
4.5. Multiple Protein Sequence Alignment and Homology Modeling
Amino acid sequences of three functional mammal uricases (pig, baboon, and dog) and three functional chimeric uricases (canine–human, porcine–human, and pig–baboon) were retrieved from GenBank or obtained from related patents. Multiple protein sequence alignment of those functional Uox with H1-2P3H4P5-6H7-8 and dHU was generated using Clustal X version 2.0 (Conway Institute, UCD, Dublin, Ireland) to analyze the inconsistent residues across these sequences.
In order to predict the influence of specific amino acid residues (E24 and E83) on the activity of uricase, the 3D structure of mammalian uricase deposited by Ortlund and colleagues in the PDB database (PDI ID: 4MB8) was used as a template to model the 3D structures of mutated chimeric uricases using MOE 2010.10 (Chemical Computing Group Inc., Montreal, QC, Canada) [30
4.6. Site-Directed Mutagenesis of E24D and E83G
Site-directed mutagenesis was implemented to obtain E24D and E24D & E83G mutants of H1-2
. Two pairs of primers (P13 & 26 and P24 & 25) with introduction of mutation E24D were designed against the gene fragments coding for amino acid residues 1–36 and 22–304 of H1-2
(see Table S1
) using Primer Premier 5. The two gene fragments were amplified and spliced by SOE-PCR, resulting in mutant H1-2
(E24D). Mutation E83G was further introduced into the E24D mutant by similar procedures, using primer pairs P27 & 24 and P13 & 28. After transformation into E. coli
BL 21, positive clones harboring the recombinant plasmid were obtained and mutations were confirmed by DNA sequencing. Chimeric enzymes were expressed and purified according to the same procedures described above.
4.7. Simulation of Interaction between Uricase H1-2P3H4P5-6H7-8 (E24D & E83G) and Uric Acid
In order to predict the interaction between uricase H1-2P3H4P5-6H7-8 (E24D & E83G) and uric acid, structural modeling of uric acid to the binding site of uricase H1-2P3H4P5-6H7-8 (E24D & E83G) was performed in silico by employing MOE 2010.10. 4MB8 was superposed to Arthrobacter globiformis uricase (PDI ID: 2YZB), whose crystal structure was complexed with uric acid. With three conserved residues in the active site (Q236, R187, and F170), a conserved water molecule involved in catalytic activity and the two residues (N262 and T68) hydrogen-bonded to it restrained, the 4MB8 structure complexed with uric acid was modeled and further used as a template to model the 3D structures of uricase H1-2P3H4P5-6H7-8 (E24D & E83G). Interaction between uric acid and the binding site of uricase H1-2P3H4P5-6H7-8 (E24D & E83G) was simulated in 2D view and 3D view by employing MOE 2010.10 and Visualizer module of Discovery studio (DS) 3.0 package (Accelrys Software, Inc., San Diego, CA, USA), respectively.
4.8. Effect of pH and Temperature on the Activity of H1-2P3H4P5-6H7-8 (E24D & E83G)
The effect of pH and temperature on the enzyme activity of H1-2P3H4P5-6H7-8 (E24D & E83G) was investigated by determining its uricolytic activity at a pH range of 6.0–11.0 using different buffer systems (NaH2PO3–Na2HPO3 6.0–7.0, borate buffer pH 7.0–9.0, Na2CO3–NaHCO3 buffer pH 9.0–11.0) as well as a temperature range of 20–70 °C. The results were expressed as the percentage of activity obtained at either the optimum pH or the optimum temperature. wPU and PBC were also included for comparison. pH stability of H1-2P3H4P5-6H7-8 (E24D & E83G) was measured by pre-incubating (in a ratio of 1:1) the enzyme solution in the abovementioned buffer systems for 30 min at room temperature and subsequently measuring its activity at 35 °C. The percentage of residual enzyme activity was calculated by assuming enzyme activity at the beginning of the reaction as 100%. Thermal stability of the enzyme was determined by pre-incubating the enzyme at different temperatures for 30 min. The percentage of residual enzyme activity was determined after cooling or warming the sample to 35 °C.