Arxula adeninivorans xanthine oxidoreductase and its application in the production of food with low purine content

Authors


Correspondence

Gotthard Kunze, Leibniz Institute of Plant Genetics and Crop Plant Research (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany. E-mail: kunzeg@ipk-gatersleben.de

Abstract

Aims

Isolation and characterization of xanthine oxidoreductase and its application in the production of food with low purine content.

Methods and Results

The A. adeninivorans xanthine oxidoreductase is an inducible enzyme. The best inducers were identified by enzyme activity tests and real-time PCR and used to produce large amounts of the protein. Xanthine oxidoreductase was partially purified and biochemically characterized, showing pH and temperature optimum of 8·5 and 43°C, respectively. The enzyme decreased xanthine and hypoxanthine concentrations in yeast extract and was active simultaneously with other purine-degrading enzymes so that all of the substrates for uric acid production were reduced in a single step.

Conclusions

It was shown that induced A. adeninivorans can produce sufficient amount of xanthine dehydrogenase and that the enzyme is able to reduce xanthine and hypoxanthine content in food, and when used in conjunction with other enzymes of the pathway, uric acid concentration is significantly reduced.

Significance and Impact of the Study

Reduction in dietary purines is recommended to people suffering from hyperuricemia. Elimination of most purine-rich foods may affect balanced nutrition. Food with lowered purine concentration will assist in controlling the disease. This study is a continuation of previous studies that characterized and overexpressed other enzymes of the purine degradation pathway.

Introduction

Purine degradation occurs in all living organisms and follows the route from purines to CO2 and ammonia or to intermediates (Werner and Witte 2011). In reptiles, birds and primates, the degradation leads to uric acid, which is the final excreted product. Some lower mammals excrete allantoin, which is the product of uric acid oxidation by the enzyme urate oxidase (Moriwaki et al. 1999). Uric acid is 10 times less soluble than allantoin, and as a result, it may accumulate in the blood causing hyperuricemia that can finally lead to a painful arthritis – gout. The increase in uric acid level can be due to a purine metabolism disorder, kidney malfunction or intake of high-purine-content food. Gout becomes a serious problem in ageing populations; people suffering from gout must deal with recurrent joint pain, inflammation and swelling (Bieber and Terkeltaub 2004; Choi et al. 2005; Yamamoto 2008).

Pharmaceutical companies have developed novel medications against gout such as recombinant urate oxidase that, when injected to a human body, leads to a decrease in serum uric acid concentration (Navolanic et al. 2003). This treatment has proven to be very effective; however, there are reports of antigenicity and other side effects associated with its use (Pui et al. 1997; Eggebeen 2007). Also, the cost of this treatment is very high. Other treatments, such as nonsteroidal anti-inflammatory drugs, steroids or inhibitors, may result in risks that can outweigh the benefits (Masseoud et al. 2005). These factors suggest that a prophylactic long-term therapy that is much safer and less expensive for patients may be useful. Our strategy is reduction in purine content of food by enzymatic hydrolysis during food production or preparation processes. This study focuses on characteristics of xanthine oxidoreductase.

Xanthine oxidoreductase (XOR) belongs to the group of molybdenum-containing hydroxylases. It catalyses the oxidation of hypoxanthine to xanthine and xanthine to uric acid using NAD+ (XDH; xanthine dehydrogenase – EC 1·17·1·4) or O2 (XO; xanthine oxidase – EC 1·17·3·2) as an electron acceptor. This enzyme is present in different eukaryotic organisms such as Drosophila melanogaster (Smith et al. 1963), Arabidopsis thaliana (Hesberg et al. 2004) and Homo sapiens (Ichida et al. 1993) and also in various prokaryotes such as Escherichia coli (Xi et al. 2000), Streptomyces cyanogenus (Ohe and Watanabe 1979), etc. In this study, we used xanthine oxidoreductase isolated from an ascomycetous, nonpathogenic yeast – Arxula adeninivorans – which has unusual biochemical features (Van der Walt et al. 1990; Wartmann et al. 1995). It can assimilate a broad-spectrum of substances, including purines, n-alkanes and starch as carbon and/or nitrogen sources (Middelhoven et al. 1984; Gienow et al. 1990).

We have previously overexpressed the other enzymes involved in purine degradation, and the overall project goal is to use the complete set of these enzymes to simultaneously degrade purines in food. Here, we show that the combination of all four enzymes is able to dramatically reduce the purine content of a yeast extract, suggesting that this is a potential method to reduce uric acid content in processed foods. This result should decrease the risk of gout.

Materials and methods

Strains, media and growth conditions

Escherichia coli TOP 10 [F- mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacΧ74 recA1 araD139 Δ(araleu) 7697 galU galK rpsL (StrR) endA1 nupG] from Invitrogen (United States) was used as the host for DNA manipulation and plasmid isolation.

Arxula adeninivorans LS3 – wild-type strain, initially isolated from wood hydrolysates in Siberia (Russia) and deposited as A. adeninivorans SBUG 724 in the strain collection of the University of Greifswald, and its auxotrophic mutant strain, G1212 [aleu2 atrp1::ALEU2], were used in this study (Kunze and Kunze 1994; Samsonova et al. 1996; Steinborn et al. 2007).

Arxula adeninivorans LS3 strain was cultivated under selective conditions in yeast minimal medium with glucose (55·5 mmol l−1) as the carbon source (YMM glucose) (Tanaka et al. 1967; Rose et al. 1990). NH4H2PO4 or NaNO3 (both 43·5 mmol l−1) or different purines (2·5 and 5 mmol l−1) served as nitrogen source. Cultivation of yeast cells was carried out in two steps: a preculture with an initial OD600 = 0·1 in YMM glucose was cultured at 30°C, for 24 h (growth phase). The cells were then shifted to YMM glucose–purines to induce gene expression and grown at 30°C for up to 72 h (production phase).

The biomass (dry cell weight, dcw) of yeast samples was determined from 1 ml cell culture that was washed with H2OMQ and freeze-dried.

Note: The water used in this work, defined here as H2OMQ, was purified from tap water using Milli-Q Advantage A10 Ultrapure Water Purification System (Merck Millipore, Billerica, MA, USA).

Transformation techniques and isolation of nucleic acids

Transformation of E. coli was performed according to Hanahan (1983), and yeast cells were transformed according to Rösel and Kunze (1998).

Plasmid DNA was purified from 2 ml of overnight culture using buffers P1, P2 and P3 (Qiagen, Germany). After precipitation and washing steps, the isolated plasmid DNA was resuspended in H2OMQ. DNA bands were purified from agarose gels using NucleoSpin® Extract II (Macherey-Nagel, Düren, Germany). The genomic yeast DNA was isolated from 2 ml cultures based on modified method of Hoffman and Winston (1987). The cells were washed with H2OMQ, resuspended in Hoffman's extraction buffer and vortexed 1 min in the presence of 0·1 ml of silica beads (0·5-mm silica beads, BioSpec Products, Bartlesville, OK, USA) and 200 μl phenol–chloroform. After the addition of TE buffer (10 mmol l−1 Tris–HCl + 1 mmol l−1 EDTA, pH 8·0), the samples were vortexed and centrifuged. The upper phase was removed and incubated with Qiagen buffer P1 for 10 min at 25°C. The genomic DNA was precipitated by the addition of isopropanol and then washed with 80% ethanol, air-dried and redissolved in 20 μl TE buffer. Total RNA was prepared from the cell pellet using Z6 extraction buffer according to Logemann et al. (1987). cDNA was synthesized according to the manual provided with RevertAid™ H Minus First Strand cDNA Synthesis Kit (Fermentas, St Leon-Rot, Germany) using oligo(dT) primers. To quantify the AXOR transcript accumulation, the appropriate cDNA fragments were amplified by real-time PCR (ABI PRISM 7700 Sequence Detection System, Applied Biosystems, Foster City, CA, USA). The Arxula genes AHSB4, TEF1, AALG9 and ATFIID were used as endogenous reference genes (Rösel and Kunze 1995; Wartmann et al. 2003). The relative expression level of AXOR was calculated using Pfaffl (2001) method modified by Hellemans et al. (2007). Primers used in this study are listed in Table 1.

Table 1. Oligonucleotides used in this study
NameNucleotide sequenceFeature
  1. Bold type letters – restriction enzyme sites.

5′-SacII-FwTTATCCGCGGGTGGCCCGACATGTGATAC Mutant generation
5′-SpeI-RvTTATACTAGTATGGATTTCTCGTAGCTTTCTGMutant generation
3′-SalI-FwTTATGTCGACGTCAATAATGAGCTGGTACGGMutant generation
3′-ApaI-RvTTATGGGCCCTGGGTCACTACCGTAATTCCMutant generation
AXOR-FwATAGAATTCATGGTTGACCTGCAACTCCATCMutant control
AXOR-RvAATATAATATGCGGCCGCCATCATCTTTGCGCTTAGCTTGCMutant control
AXOR-qPCR-FwTTGTAGCACCCATGACGACqPCR
AXOR-qPCR-RvCCATTAGGCATTGACAAAGAqPCR
AHSB4-qPCR-FwCTCATTTATGAGGAGACCCGqPCR
AHSB4-qPCR-RvATAAAGAGTTCGTCCCTGTCqPCR
ALG9-qPCR-FwCATGGGCCAAGGTATACTGqPCR
ALG9-qPCR-RvAGAATGCGACCGATACCAqPCR
TEF-qPCR-FwCTCTTGGACGATTCGCCqPCR
TEF-qPCR-RvCCGTTACCGACAATCTATTTqPCR
TFIID-qPCR-FwAGTTTGTGTCTGATATTGCCTCqPCR
TFIID-qPCR-RvGAAGAGCTTGCTACCGAACTqPCR

Sequencing of all nucleic acids was performed using an automatic laser fluorescence DNA sequencer (Pharmacia, Sweden). The nucleotide sequence of isolated gene has been entered into the GenBank/EMBL data libraries (HF935016).

Enzyme assays and protein analysis

Crude cell extracts were made from 2 ml of yeast culture, which was harvested, washed with H2OMQ and suspended in 500 μl 0·1 mol l−1 glycine–NaOH buffer, pH 8·5. Cells were homogenized using 500 μl of 0·5-mm silica beads (BioSpec Products) in a Mixer Mill MM 400 (Retsch, Haan, Germany) for 3 min at 30 Hz, 4°C. The protein extract was separated from cell debris and silica beads by centrifugation at 17 000 g for 5 min, 4°C.

Xanthine oxidoreductase activity in the extracts was measured spectrophotometrically in the 96-well microtiter plates. The reaction mixture consisted of 1·33 mmol l−1 xanthine, 1 mmol l−1 NAD+, 0·001 mmol l−1 FAD, 33mmol l−1 glycine–NaOH buffer, pH 8·5 (for characterization), or 33 mmol l−1 potassium phosphate buffer, pH 8·0 (for purification), and enzyme in a total volume of 300 μl. Xanthine stock solution was prepared by dissolving xanthine in 0·1 mol l−1 NaOH to a concentration of 8 mmol l−1. The working solution was made by mixing 100 parts of stock solutions with 85 parts of water and 15 parts of 0·5 mol l−1 HCl, resulting in 4 mmol l−1 xanthine solution of pH 9 to 10. The measurement started after addition of NAD+, and the progress of reaction at 43°C was monitored at 340 nm for 30-50 min. One unit of enzymatic activity is defined as the amount of protein that catalyses the formation of 1 μmol of NADH per minute at 43°C.

Urate oxidase was assayed in 0·1 mol l−1 potassium phosphate buffer, pH 7·8, with 0·2 mmol l−1 uric acid in a total volume of 200 μl. The reaction was run at 37°C and followed at 293 nm for up to 30 min.

α-d-mannosidase (EC 3·2·1·24) was used as a marker for vacuolar enzymes and assayed according to Opheim (1978). Glucose-6-phosphate-dehydrogenase (EC 1·1·1·49) served as a cytosolic marker and was assayed as described by Gibon et al. (2002, 2004).

The protein content of samples was measured by the method of Bradford (1976) using purified bovine serum albumin as standard.

Purified wild-type xanthine oxidoreductase was prepared from cells grown on 2·5 mmol l−1 hypoxanthine as the nitrogen source as described in 'Methods and Results'. Cells from a 1 L culture were harvested after 6 h of cultivation, resuspended in 5 ml of potassium phosphate buffer (50 mmol l−1, pH 8·0), frozen with liquid nitrogen and ground with mortar and pestle. After centrifugation, ammonium sulfate was added to the crude extract to 40% saturation, stirred for 30 min and centrifuged. The ammonium sulfate concentration was then raised to 60%. The resulting precipitate was redissolved in potassium phosphate buffer and dialysed against potassium phosphate buffer (50 mmol l−1, pH 8·0) for 12 h. The xanthine oxidoreductase was then purified using a DEAE-Sepharose ion exchange column equilibrated with 50 mmol l−1 potassium phosphate buffer (pH 8·0). The enzyme was eluted with phosphate buffer containing a linear gradient of KCl. Active fractions were then loaded onto Superdex™ 200 gel filtration column that had been equilibrated with 0·05 mol l−1 sodium phosphate buffer (pH 7·4) containing 0·15 mol l−1 sodium chloride. The gel filtration column was calibrated with dextran blue (Mr 2 000 000), ferritin (Mr 440 000), catalase (Mr 232 000), alcohol dehydrogenase (Mr 150 000), bovine serum albumin (Mr 68 000) and ovalbumin (Mr 45 000).

Axorp was visualized on 7·5% SDS-PAGE and blotted onto a polyvinylidene difluoride (PVDF) membrane as described by Kunze et al. (1998).

The kinetic parameters of the Axorp were determined at its optimum pH and temperature. The substrate was added in twelve increasing concentrations ranging from 1·7 to 1,000 μmol l−1 for NAD, from 3·3 to 1,333 μmol l−1 for xanthine and hypoxanthine and from 3·3 to 2,000 μmol l−1 for purine. The measurements were taken at 340 nm at intervals of 1 min. The kinetic constants were determined by nonlinear regression using Microsoft Excel add-in, Excel Solver.

Subcellular fractionation

The preparation of protoplasts was performed as described previously by Jankowska et al. (2013) with small modifications in the cultivation process: A. adeninivorans LS3 cells were inoculated into 50 ml YMM glucose–adenine (2·5 mmol l−1) at OD600 ~0·2 and grown to OD600 1·0. Additionally, the protoplast fractionation spins were for 3·5 h at 150 000 g at 4°C. 23 fractions were collected and stored in 1·5-ml tubes on ice for up to 1 week.

HPLC analysis of xanthine oxidoreductase substrates and products

The xanthine oxidoreductase Axorp was used to hydrolyse xanthine and hypoxanthine in yeast extract. The progress of reaction was followed using HPLC (TSK gel Amide-80 analytical column; 4·6 mm ID × 25 cm, Tosoh Bioscience, Japan). An enzyme mix containing 1 U of Axorp and 0·125 U each of urate oxidase, adenine deaminase and guanine deaminase was added to 0·2 g of yeast extract (Ohly® KAT GMP, Ohly, Germany) dissolved in 0·3 ml 0·5 mol l−1 NaOH and 1·2 ml 20 mmol l−1 potassium phosphate buffer, pH 3·0, and incubated at 40°C for up to 240 min in the presence of 4·5 mmol l−1 NAD+. After the reaction, the samples were heated at 100°C for 5 min and centrifuged, and 20 μl was loaded on an HPLC column. The conditions of the HPLC analysis were as previously described by Jankowska et al. (2013).

Construction of A. adeninivorans Δaxor gene disruption mutant

To assemble the AXOR gene replacement vector, 1-kb lengths of a 5′ fragment and a 3′ fragment were amplified by PCR. These sequences were located at the 5′ and 3′ ends of the open reading frame (ORF). PCR of the chromosomal DNA as the template incorporated SacII, SpeI, SalI and ApaI restriction sites of the 5′ and 3′ fragments, respectively. The isolated sequences were inserted into an Xplor2 plasmid between its SacII-SpeI and SalI-ApaI restriction sites. The selection marker, ALEU2 promoter, ATRP1 m gene and ATRP1 terminator (Steinborn et al. 2007) present in this plasmid, allowed for mutant selection. The resulting plasmid Xplor2·2 5′ fragment–3′ fragment was used as template to amplify the final cassettes (Fig. 1) using primers 5′-SacII-Fw and 3′-ApaI-Rv prior to transformation of A. adeninivorans G1212.

Figure 1.

Physical map of the AXOR gene disruption cassette, which was used for homologous recombination. The upstream and downstream flanking DNA sequences (5′ and 3′ fragments) are identical to those, which flank the targeted locus. The construct contains a selection marker ATRP1 m with the ALEU2 promoter.

Results

Isolation of the AXOR gene

A 4218-bp sequence containing the complete AXOR gene was identified by comparing annotated A. adeninivorans genomic sequences (manuscript in progress) with fungal xanthine oxidoreductase sequences using BLAST. The derived amino acid sequence exhibits a high degree of similarity to xanthine oxidoreductases from Pichia pastoris (58%; XP_002492094), Trichophyton equinum (54%; EGE04942), Gromerella graminicola (53%; EFQ34799), Neurospora crassa (53%; XP_956459·2), Aspergillus nidulans (53%; XP_663217) and Aspergillus niger (53%; XP_001401908).

The AXOR gene is present in a single copy in the A. adeninivorans LS3 genome and is located on the largest of the four chromosomes, Arad1D. It encodes a protein of 1406 amino acids with theoretical molecular mass of 153 kDa and isoelectric point of 6·04.

Disruption of the AXOR gene

To confirm the function of the AXOR gene, mutant strains lacking AXOR sequence were constructed (Fig. 1). The gene was replaced by the ATRP1 m selection marker as described in 'Methods and Results'. Because the G1212 strain is auxotrophic for tryptophan, the transformants could be selected by complementation with an ATRP1 m gene carried in the replacement vector. 96 colonies were analysed for the presence of the Δaxor mutation using PCR with two primer pairs: 5′-SacII-Fw/3′-ApaI-Rv and gene-specific AXOR-Fw/AXOR-Rv, and the positives were confirmed using an enzyme test for xanthine oxidoreductase activity. The selected mutant G1224 [aleu2::ALEU2 atrp1::ATRP1 axor] exhibits less than 1% of the wild-type strain activity towards xanthine and hypoxanthine under inducible conditions (2·5 mmol l−1 adenine), and both strains show no activity after cultivation without inducer.

Urate oxidase activity test was used to complete the mutant analysis (Table 2). It is known (Scazzocchio et al. 1973; Reinert and Marzluf 1975) that transcription of the urate oxidase gene is induced mainly by uric acid and not by the other purines, and our results are in agreement with those. These results also demonstrate that the Δaxor mutant was obtained.

Table 2. Urate oxidase activity test results for Δaxor mutant Arxula adeninivorans G1224 in comparison with the wild-type strain LS3
StrainInducer/N-sourceActivity (U mg−1)Relative activity (%)
  1. All cultures were grown in the presence of 2·5 mmol l−1 inducer/N-source.

G1224Adenine0·11 23
Hypoxanthine0·13 28
Xanthine0·13 27
Uric acid0·46100
NaNO30·12 26
LS3Adenine0·32116
Hypoxanthine0·30112
Xanthine0·16 61
Uric acid0·27100
NaNO30·05 20

Purine-dependent expression of the AXOR gene

Arxula adeninivorans is known to grow on different nitrogen and carbon sources (Middelhoven et al. 1984). The enzymes of its purine degradation pathway are subjected to induction by their substrates and/or products, as shown for adenine deaminase (Jankowska et al. 2013) and urate oxidase (Trautwein-Schult et al. 2013). We tested the inducibility of the next enzyme of this pathway – xanthine oxidoreductase. Figures 2a,b show the transcript accumulation level measured by quantitative real-time PCR in two experiments. The most efficient inducer, hypoxanthine, increased gene transcription to twice the level of the transcription increase due to adenine. Transcription reached maximum level after 2-4 h, but then decreased to the basal level. Enzyme assay tests (Fig. 2c,d) showed that the enzyme production peaked at 6 h; however, the level of enzyme produced was only slightly less than that with hypoxanthine. Both approaches demonstrate limited influence of uric acid on AXOR gene induction.

Figure 2.

Purine-dependent AXOR transcription measured using quantitative real-time PCR (a,b) and xanthine oxidoreductase activities (c,d) after cultivation in YMM glucose with different concentrations of (□) adenine, (▲) hypoxanthine, (○) uric acid and (●) NH4Cl.

More information on the effect of different inducers is presented in Fig. 3. A. adeninivorans cells showed low levels of xanthine oxidoreductase activity when grown on ammonium and nitrate. In addition, cells grown in medium containing 43 mmol l−1 ammonium/nitrate and 2·5 mmol l−1 adenine showed enzyme levels only 2- to 3-fold higher than the noninduced level. This is in contrast to cells grown only on adenine where the increase was more than 10-fold. The addition of another N-source to the adenine-induced culture, however, appears to stop the induction. Obviously, ammonium and nitrate do not have any influence on the amount of enzyme already produced, but they play a role in the gene induction.

Figure 3.

Induction of xanthine oxidoreductase under various conditions. Wild-type cells were grown on different nitrogen sources, and samples were taken after 0 h (■), 3 h (□) and 6 h (▧). Data labels Ade 3 h+NO3 and Ade 3 h+NH4+ indicate that the cells were induced only by adenine during the first 3 h, but enrichment with ammonium or nitrate for the next 3 h resulted in no further increase. Samples named Ade+NO3 and Ade+NH4+ contain both nitrogen sources during the entire cultivation period.

The production of endogenous enzymes has been optimized by variation in glucose concentration and temperature of cultivation. Figure 4 illustrates a time course experiment of induced and noninduced yeast cells at optimal 30°C and 1% glucose. Y(P/X) represents the yield coefficient of product P (U) per biomass X (g). The highest yield appeared at 5 h induction, but it decreased thereafter.

Figure 4.

The enzyme yield from A. adeninivorans LS3 cultivated in (a) YMM glucose–adenine 2·5 mmol l−1 and (b) YMM glucose–NaNO3. The cells were grown in shake flasks up to 39 h at 30°C in 1% glucose. Dry cell weight and xanthine oxidoreductase activity were determined from 1 ml culture medium (▲), dcw (g l−1), (■) (U ml−1), which allowed the calculation of xanthine oxidoreductase output (×) Y(P/X) (U g−1 dcw).

Purification and molecular mass estimation

The wild-type enzyme was partially purified (12 fold) as described in 'Methods and Results'. It had a specific activity of 0·48 U mg−1 (Table 3), and filtration on Superdex™ 200 gave a calculated molecular mass of the native protein of approximately 360 kDa. Analysis under denatured conditions, however, revealed a strong band at ~160 kDa (Fig. 5). This band was blotted onto a PVDF membrane and subjected to N-terminal sequencing by the Metabion Company (Germany). The first six amino acids correlate with the sequence of the first six codons of the AXOR gene.

Table 3. Purification steps for endogenous xanthine oxidoreductase synthesized by Arxula adeninivorans LS3
StepTotal protein (mg)Total activity (U)Specific activity (U mg−1)Yield (%)Purification fold
Crude extract110·494·390·0401001
(NH4)2SO4 precipitation 29·072·740·09411·042·37
DEAE-Sepharose column 11·671·550·13310·093·34
Superdex™ 200 column 0·720·350·4862·6812·21
Figure 5.

Purification of endogenous xanthine oxidoreductase synthesized by A. adeninivorans LS3. M – protein marker, 1 – crude extract, 2 – 40–60% ammonium sulfate precipitation, 3 – DEAE-Sepharose chromatography column, 4 – Superdex™ 200 gel filtration column.

Properties and subcellular localization of the Axorp

Xanthine oxidoreductase had a pH optimum between 7·5 and 9·5 in potassium phosphate, Tris–HCl, Tris–acetate and glycine–NaOH buffers with the highest activity at pH 8·5 in the latter. The optimal temperature range for this enzyme was 35-50°C with the activity peak at 43°C. Xanthine oxidoreductase is quite stable; it loses 5% of activity in 1 h at 30° C, 12% at 40°C and 25% at 50°C.

Xanthine oxidoreductases transfer electrons from a substrate to an electron acceptor. The A. adeninivorans xanthine oxidoreductase uses NAD+ as its electron acceptor. NADP+ and O2 enabled less than 2% of the enzyme activity seen with NAD+.

The compounds listed in Table 4 were examined as substrates with NAD+ as an electron acceptor. Xanthine, hypoxanthine and pterine were oxidized at high rates. The enzyme can also act on purine and guanine. Table 5 presents the effect of purine analogues on rate of the reaction. Adenine slowed it significantly, whereas allopurinol, a known inhibitor, blocked it almost completely. The Michaelis constant of the xanthine oxidoreductase was determined by nonlinear regression (Excel Solver). KM values obtained were 72·8, 50·8, 162·5 and 208·0 μmol l−1 for xanthine, hypoxanthine, NAD+ and purine, respectively.

Table 4. Substrate spectrum for xanthine oxidoreductase. Final concentration of all substrates was 0·66 mmol l−1
SubstrateRelative activity (%)
Xanthine 97
Hypoxanthine100
Adenine 1
Guanine 10
Uric acid 0·1
Caffeine 0·2
Purine 30
Pterine 96
Theobromine 2
Allantoin 0·2
Allopurinol 0·6
8-azaguanine 0·0
8-azaxanthine 0·0
Table 5. Spectrum of substances with their effects on xanthine oxidoreductase activity. The substrate was 1 mmol l−1 xanthine; all the substances were 0·5 mmol l−1
Purine derivateRelative activity (%)
None100
Adenine 28
Guanine 70
Uric acid 94
Caffeine 95
Purine 85
Theobromine 99
Allantoin 97
Allopurinol 4
8-azaguanine 95
8-azaxanthine 82

The effects of various additives and divalent cations (1 mmol l−1) on the activity are presented in Table 6. The activity was inhibited by Fe2+, Cd2+ and Zn2+ and strongly inhibited by Cu2+. Conversely, Mg2+ and Mo2+ had no effect on the enzyme.

Table 6. Effect of metal salts, DTT and chelating reagents on xanthine oxidoreductase activity at 1 mmol l−1 final concentration
SupplementRelative activity (%)
None100
MgSO4103
Ca(NO3)2 84
CuCl2 3
ZnSO4 33
Na2MoO4 98
MnSO4 65
FeSO4 25
CdSO4 33
NiSO4 49
EDTA 90
DTT 67
1,10-phenanthroline 93

Xanthine oxidoreductase amino acid sequence analysis using HMMTOP transmembrane topology prediction server (Tusnady and Simon 2001) indicated that transmembrane helices were not present. According to SOSUI algorithm (Hirokawa et al. 1998), this enzyme is soluble and free from signal peptide and should be localized in the cytoplasm.

Sucrose density gradient centrifugation provided 23 fractions, which were used for determination of marker protein and Axorp activities (Fig. 6). Cytoplasmic glucose-6-phosphate-dehydrogenase was detected in the last fractions starting with number 17. Vacuolar α-d-mannosidase migrated in the gradient similarly to Axorp (fractions 16→20-21).

Figure 6.

Subcellular localization of xanthine oxidoreductase. The cells were prepared as described in 'Methods and Results'. The cell components were separated by centrifugation and analysed for marker proteins: (●) xanthine oxidoreductase, (Δ) α-d-mannosidase, (□) glucose-6-phosphate-dehydrogenase. Sucrose refractive index is represented by (▲).

Use of xanthine oxidoreductase to reduce xanthine and hypoxanthine in food

Xanthine oxidoreductase was examined as a potential additive in food to reduce purine content. The partially purified wild-type enzyme was used together with three already overexpressed and characterized enzymes of the pathway (adenine deaminase (Jankowska et al. 2013), urate oxidase (Trautwein-Schult et al. 2013) and guanine deaminase). The test mixture was able to decrease adenine content from 198·58 to 5·0 mg l−1, xanthine from 169·24 to 39·59 mg l−1, hypoxanthine from 171·56 to 52·22 mg l−1 and uric acid from 7·04 to 0·0 mg l−1. In this trial, guanine was not detected (Fig. 7).

Figure 7.

The chromatogram shows purine content in the sample before (thick line) and after enzyme reaction (thin line). Sample containing yeast extract was treated with 1 U of Axorp and 0·125 U of enzyme mix (urate oxidase, adenine deaminase and guanine deaminase). Adenine peak (1) disappears, and both hypoxanthine (2) and xanthine (3) peaks diminished significantly. Uric acid (4) and guanine were barely detected.

Discussion

The prevalence of hyperuricemia and gout is rising in the population (Schlesinger 2005). Humans lost urate oxidase during the evolutionary process and with increasing lifespans are exposed to age-associated diseases that are risk factors for gout such as hypertension, metabolic syndrome and renal disease (Saag and Choi 2006). The commonly used drugs for treatment are nonsteroidal anti-inflammatory drugs, xanthine oxidoreductase inhibitors and uricosurics. Recombinant urate oxidase, given to the patient as an infusion during a gout attack, can rapidly reduce the symptoms (Wipfler-Freibmuth et al. 2009). However, all of these treatments have limitations (Masseoud et al. 2005). Reducing the intake of purines would help considerably in maintaining lower serum urate concentration, as one-third of body urate comes from the food (Schlesinger 2005). We propose a method to reduce or eliminate purines in food products. Our method treats purine-containing foods with four enzymes (adenine deaminase, urate oxidase, xanthine oxidoreductase and guanine deaminase) that degrade purines (Jankowska et al. 2013; Trautwein-Schult et al. 2013). In this study, xanthine oxidoreductase was characterized and tested for the first time on food.

The enzyme has relatively broad substrate specificity. It can hydrolyse xanthine, hypoxanthine, guanine, purine and pterine. It is possible that Axorp may also act on other substances, as is the case of xanthine oxidoreductase from Streptomyces cyanogenus (Ohe and Watanabe 1979). Xanthine oxidoreductases from different organisms prefer different electron acceptors (Waud and Rajagopalan 1976). The A. adeninivorans enzyme is NAD+-dependent, which implies that it is a xanthine dehydrogenase. The results of kinetic studies with three substrates have revealed that the enzyme has a slightly higher affinity for hypoxanthine than for xanthine and a lower affinity for purine.

It has been reported that the optimum pH of the eukaryotic and prokaryotic xanthine oxidoreductases was 7·4 to 8·7, which puts the A. adeninivorans enzyme optimal pH of 8·5 in the upper end of the range. Axorp optimum temperature of 43°C was found to be similar to that of xanthine oxidoreductases of Pseudomonas synxantha (Sakai and Jun 1979) and Gallus gallus (Carro et al. 2009).

Electrophoresis of purified enzyme in SDS gel gave a molecular mass of about 160 000 Da. Gel filtration revealed a protein of around 360 kDa, which suggests that Axorp is a dimer. A protein of similar mass (357 kDa) was observed in Neurospora crassa (Lyon and Garrett 1978), but both enzymes are bigger than most of the characterized xanthine oxidoreductases.

Xanthine oxidoreductase localization in A. adeninivorans cells is unclear. In our experiment, three enzyme activities overlapped, revealing Axorp activity in cytosolic and vacuolar fractions. Analysis of the gene sequence indicated cytosolic localization, but there is no targeting signal, as is in the case of most xanthine oxidoreductases. The enzyme is also present in vacuole fractions, which could be due to overloading the cell with highly induced enzyme. Some XOR activities were found in peroxisomes, but this is probably associated with the next reaction in purine degradation pathway, which takes place in peroxisome (Moriwaki et al. 1999; Werner and Witte 2011).

The regulation of xanthine oxidoreductase expression in relation to nitrogen source was studied. Activity tests and qRT-PCR experiments on enzyme inducers show strong gene inducibility when cells were cultivated on hypoxanthine and adenine as sole nitrogen source. The effect of uric acid is much less significant; however, as demonstrated for Aspergillus nidulans and N. crassa, xanthine oxidoreductase is more strongly induced with uric acid than with other purines (Scazzocchio and Darlington 1968; Reinert and Marzluf 1975). Lyon and Garrett (1978) have shown that XOR of N. crassa can be induced by purines in the presence of NO3, but not with NH4+. We have observed that enzyme induction stops after supplementing the medium not only with NH4+, but also with NO3. It is known that in A. nidulans, NH4+ inactivates the GATA factor AreA, which is responsible for expression of the urate–xanthine transporter (Gournas et al. 2011). It is not clear what mechanisms trigger the repression with ammonia and nitrate in A. adeninivorans, and more experiments are needed to explain this observation.

Several attempts were made to overexpress xanthine oxidoreductase in A. adeninivorans using the well-established Xplor®2 expression system. All attempts failed leading us to suspect that we had identified only part of the full gene sequence. ORF analysis and sequence alignment, however, suggested that this was not the case. To confirm the functionality of the gene, we disrupted it and found that the Δaxor knockout mutant, A. adeninivorans G1224, lost activity with xanthine and hypoxanthine, indicating that the isolated gene is intact.

The Δaxor knockout mutant also allowed us to identify ‘true’ inducers of urate oxidase. It was shown to be induced by adenine, hypoxanthine and uric acid by Trautwein-Schult et al. (2013), but according to the reports, the first two purines are converted to uric acid before they can act as an inducer (Scazzocchio and Darlington 1968; Reinert and Marzluf 1975). Our results with the G1224 mutant agree with those findings.

Finally, this study demonstrates the potential of this approach for the production of low-purine-content food. Xanthine oxidoreductase was able to reduce hypoxanthine and xanthine concentration in the yeast extract. The use of this enzyme with the other available purine-degrading enzymes to produce a purine-reducing system is very promising (Jankowska et al. 2013; Trautwein-Schult et al. 2013). Success of this novel approach will result in better control of hyperuricemia and gout without an accompanying reduction in the quality of life.

Acknowledgements

The authors are grateful to Dr M. Giersberg and Dr S. Worch for helpful discussions and to I. Schmeling for excellent technical assistance. The research work was supported by grant from the BMWi (Project No. KF2131601MD8).

Ancillary