|Home | About | Journals | Submit | Contact Us | Français|
Xanthine dehydrogenase (EC 188.8.131.52, XDH) is a typical and complex molybdenum-containing flavoprotein which has been extensively studied for over 110 years. This enzyme catalyzes the oxidation of purines, pterin and aldehydes with NAD+ or NADP+ as electron acceptor, and sometimes can be transformed to xanthine oxidase (EC 184.108.40.206, XOD) capable of utilizing oxygen as the electron acceptor. XDHs are widely distributed in all eukarya, bacteria and archaea domains, and are proposed to play significant roles in various cellular processes, including purine catabolism and production of reactive oxygen species (ROS) and nitric oxide (NO) in both physiological and pathological contexts. The recent applications of XDHs include clinical detections of xanthine and hypoxanthine content in body fluidics, and other diagnostic biomarkers like inorganic phosphorus, 5′-nucleotidase and adenosine deaminase. XDHs can also find applications in environmental degradation of pollutants like aldehydes and industrial application in nucleoside drugs like ribavirin. In this commentary, we would outline the latest knowledge on occurrence, structure, biosynthesis, and recent advances of production and applications of XDH, and highlighted the need to develop XDHs with improved performances by gene prospecting and protein engineering, and protocols for efficient production of active XDHs in response to the increasing demands.
Xanthine dehydrogenase (EC 220.127.116.11, XDH) is an old typical and complex molybdenum-containing flavoenzyme, which has been extensively studied for its biochemical and structural properties ever since 1902.1 It is well known that this oxidoreductase catalyzes the successive oxidation of hypoxanthine to xanthine and xanthine to uric acid with the concomitant reduction of NAD+ to NADH. In addition, XDH catalyzes the hydroxylation of sp2-hybridized carbon atoms of heterocyclic compounds including purines, pterins and aldehydes.2 Some mammalian XDHs, like bovine milk XDH, can be converted from the dehydrogenase form into the xanthine oxidase (EC 18.104.22.168, XOD) form, a more convenient form capable of utilizing molecular oxygen in the air as the electron receptor.3 Thus, XDHs have attracted wide attentions in physiological, clinical, environmental and industrial fields, such as clinical disease diagnosis, synthesis of nucleoside drugs, and detection and bioremediation of organic pollutants.4,5
Since the first demonstration of the oxidation of formaldehyde by fresh milk XDH by Schardinger in 1902, dozens of XDHs have been characterized from eukaryotes and bacteria. Eukaryotic XDHs are originated from animals, including cow, chicken and rat, and from plants like Arabidopsis thaliana, and usually exist as α2 form.6-9 Bacterial XDHs, including Rhodobacter capsulatus, Pseudomonas putida and Streptomyces cyanogenus, are found in the α2, α4, (αβ)2, (αβ)4 and αβγ forms.10-14 XDH consists of 3 redox center domains, one of which contains 2 distinct iron-sulfur clusters ([2Fe-2S]), another that includes a flavin adenine dinucleotide (FAD), and a third that incorporates a sulfurated molybdenum cofactor (Moco). XDHs arrange in a highly similar way despite the differences in subunit composition, as disclosed by the structural analysis.15,16 A few XDH genes have also been cloned and expressed in different hosts.6,17,18 The bacterial XDHs usually show better catalytic activity and thermal stability than that of eukaryotic XDHs, however, both of them display optimum catalytic activity at nearly neutral pH and relatively low temperatures, and very limit pH-activity range, which limit their biotechnological applications (see the review19).
In this addendum, we would describe the latest developments of occurrence, biosynthesis, structure, production, and the potential applications of the XDHs. We also have highlighted the need to develop XDHs with improved properties by gene prospecting and protein engineering, and protocols for efficient production of active XDH in response to the increasing demands.
Bovine milk XDH is undoubtedly the first and best-characterized XDH so far, which has been reported as mostly pure form as early as 1939.20 Since then, several eukaryotic XDHs have been purified and characterized from animals like human, rat, sheep, chicken and fruit fly.21,22 In 1960, Bradshaw and Barker have reported the purification and characterization of the first bacterial XDHs from Clostridium cylindrosporum. Other bacterial XDHs, such as Micrococcus, Rhodobacter, Streptomyces genus, have also been purified and characterized.23,24 With the progress of genome sequencing techniques, tens of thousands of new XDH genes have been disclosed. As of April 9, 2016, a total of 104, 303 XDH gene sequences have been deposited in the GenBank database. They are 96,077 bacterial XDHs (93.25%), 7513 eukaryotes XDHs (7.2%), 643 archaea XDHs (0.62%), 61 metagenomic ones (0.058%), 6 virus XDHs and 3 synthetic ones (Fig. 1). The top 3 subclasses (phyla) are the gamma-Proteobacteria (40045), β-Proteobacteria (17183) and high GC Gram+ Actinobacteria (16928). Up to now, only the bacteria and eukaryote XDHs are characterized experimentally, none of archaea and metagenomic XDHs have been identified experimentally. The only well-annotated virus XDH comes from Acinetobacter phage Ab105-3phi, and a homology search in GenBank database reveals 100% identity to the genomic XDH genes of Acinetobacter baumannii. It seems plausible that the similarity is a result of horizontal gene transfer. Further studies need to clarify the inheritance relationship between the XDHs of Acinetobacter genus and its virus in the future.
As shown in Table 1, pairwise alignments of primary sequences of XDH homologues reveal that: 1) eukaryotic XDHs share higher similarity and identity than those of bacteria, eubacteria and metagenomes; and 2) the [2Fe-2S] domainis are more conserved than the Moco domain, and FAD domain is the least conservative one. For example, the similarity and identity of [2Fe-2S] domain between the 9 XDH homologues representing the animal, plant, bacteria, archea and uncultured microorganisms distribute in a range of 70.1%–32.2% and 56.9%–22.4%, the Moco domain are 47.6%–34.7% and 47.6%–20.6%, and the FAD domain are only 41.3%–21.7% and 41.3%–9.9%, respectively (Table 1). Recently, we have firstly demonstrated a unique industrially applicable A.baumannii XDH, which shows only modest similarity to all the previous already-characterized XDHs.19 This work may persuade the further exploitation of other low homologous XDHs, especially the never reported thermophiles, archaea and metagenomic ones.
As of April 9, 2016, only the crystal structure data of the bovine, rat, human and R. capsulatus XDHs are available from the PDB database.25 The most-intensively studied structures are the bovine milk XDH and R. capsulatus XDH, which are the first and best resolved mammalian XDH and the only prokaryotic one, respectively. Compared to the bovine α2XDH, which consists of single polypeptide incorporating all the 3 redox center domains, the R. capsulatus α2β2XDH arranges the FAD and [2Fe-2S] domains and the Moco domain into 2 separate subunits. However, both the XDHs fold in a highly similar way, which lays the structural foundation for the same catalytic mechanism.15,16
As shown in Figure 2, multiple sequence alignment (see Supplementary Figs. S1, S2, and S3) of 9 representative XDH homologues reveals the conserved functional and structural features of primary sequences. First, [2Fe-2S] domain is the most conserved domain, and then comes the Moco domain, and the last the FAD domain. The two [2Fe-2S] cluster binding motifs: C-X4-C-X2-C-Xn-C and C-X2-C-Xn-C-X1-C, where X indicates any amino acid, and n means any number of amino acids, are completely identical (Fig. 2a). This absolute conservation may provide the good explanation to the requirement of [2Fe-2S] cluster to transfer the electrons from the Moco domain to the FAD domain.
Second, as for the Moco domain, where the catalytic activity takes place, catalytic amino acid residues and the substrate-binding pocket region are very conserved (Fig. 2b). Using bovine XDH numbering, Glu802 binds the substrate and stabilizes the transition state, Glu1261 is the catalytic base, Arg880 and Thr1010 bind the substrate and decrease the reaction activation energy, Phe914 and Phe1009 orientate the substrate via π–π stacking, Val1011 is the key residue channeling the substrate, and Gln758 is responsible for releasing the product. The results are in strong consistent with structural superimposition results revealed by a previous research.19 There is an obvious variation of key residues channeling the substrate and binding pocket, which affect the substrate entry and product release, resulting in different catalytic activity and enzymatic properties. Surprisingly, the 2 pairs of cysteines, C535 and C992, and C1316 and C1324 numbering in bovine XDH, which are proposed to control the reversible post-translational conversion from XDH to XOD by forming 2 cysteine disulfide bonds, are totally absent in all other 8 XDHs, even the eukaryotic A. thaliana XDH1. This absence may be the reasons for the true XDH, like R. capsulatus and chicken XDH, that can't be converted to oxidase form by neither proteolytic cleavage nor oxidation of specific cysteine residues.17
Third, as for the FAD domain, the FAD cofactor binding motifs are conserved in 4 regions each with an identical amino acid residue (Fig. 2c). There is also a good conservation in the loop423–433 residues (numbering in bovine XDH), which dominates the electrostatic potential difference between the XDH and XO.26 The conservations probably lay the foundation for the reduction of NAD by transferring the electron received from the Fe/S cluster to NAD.15 The conservative substitutions in the above motifs except the identical sites, together with the variations in the loop423–433 determine the electron transfer efficiency from Fe/S cluster to NAD, bringing up the different enzymatic properties.15,19
The majority of the characterized XDHs display optimum catalytic activity at neutral pH values (from 6.5 to 9.0) and relatively low temperatures (25°C–55°C), and narrow pH-activity and temperature-activity profiles, which can not meet the diversified requirements, such as very harsh operational conditions in biotechnological applications (see a review in the Supplementary Table 1 in reference19). Comparatively speaking, eukaryotic XDHs have preferable substrate affinity (lower Kmvalues), but the bacterial ones usually perform better activity, stability and broader ranges of pH-activity and temperature-activity relationships. The newly characterized A. baumannii XDH extends the pH tolerance to no less than pH 11.0, and the catalytic activity to the turnover number of 25 s−1 and catalytic efficiency of 2.74 μM−1.s−119. A. baumannii XDH exhibits very promising applicable properties, suggesting that bacteria could be used as a source for obtaining novel and useful XDHs. To satisfy the various specific biotechnological applications, like extreme pH values, it is urgent to develop more excellent XDHs with better catalytic activity, pH tolerance, thermo-tolerant performances. Further investigation of the newly annotated XDH genes in GenBank, especially those derived from extremophiles utilizing substrates including purines, pterines, and aldehydes, may potentially hit the right nail on the head. Meanwhile, it is also very promising to tailor the already existing efficient bacterial XDHs for specific properties, such as better substrate affinity, by using the protein engineering and directed evolution approaches, which have not been applied in XDHs except some mutational research for identifying the functional roles of specific active site residues.
The biosynthesis of active XDH is a multistep process, which requires the synthesis and formation of multimeric apoprotein, and ordered incorporation of redox centers into the apoprotein.17,27 Leimkuhler group has proposed a pioneering mechanism for the assembly and cofactor insertion of R. capsulatus (αβ)2XDH.28 According to the mechanism, the biosynthesis of active XDH is proposed to involve 5 processes: 1) synthesis of apoproteins consisting of the 3 domains, depicted as S, M and L; 2) formation of heteromultimer apoprotein in the case of heteromultimer multi-subunit XDH; 3) insertion of 2 [2Fe-2S] clusters and a FAD into the S and M domains respectively; 4) dimerization of apoproteins assembled with [2Fe-2S] and FAD cofactors via the L domains; 5) insertion of sulfurated Moco into the L domains with the help of chaperone protein XdhC (Fig. 3 part II). The Moco insertion is the last step to form active XDH and occurs after the formation of dimerization of apoproteins assembled with [2Fe-2S] and FAD cofactors. The sulfurated Moco is synthesized from molybdenum by a series of helper proteins, including molybdenum insertase, chaperone XdhC and Moco sulfurase (NifS4).29 XdhC binds stoichiometric amount of Moco, interacts with NifS4 for the sulfuration of Moco, protects sulfurated Moco from oxidation, and further transfers to XDH29 (Fig. 3 part I). The efficient synthesis and insertion of sulfurated Moco is the bottleneck to produce the active XDHs. Some mammalian XDHs including bovine milk and rat liver can be converted reversibly into active XOD form by forming disulfide bond or irreversibly by limited proteolysis3,30 (Fig. 3 part III). These post-translational conversion steps become another hint for efficient production of active XOD.
The first commercial XDH product is the bovine milk XDH, which can be tracked to the 1970s and was extracted from bovine milk, and it is still the most important market product. Other available commercial XDHs include the E. cloacae XOD and A. luteus XOD, produced by Kikkoman Company and Toyobo Company, respectively (Table 2). However, all these commercial products are extracted from native host and usually mixed with a large quantity of inactive form XDHs, including desulfo-, and demolybdo-XORs,21 because of the complex biosynthesis and multiple-step production processes. Several XDH genes have been cloned and expressed in exogenous hosts, which includes the rat liver XDH expression in insect cell system, A. thalianas XDH in Pichia pastoris, and R. capsulates and A. baumannii XDHs in E. coli.6,17,18 A recombinant E. coli XDH can even be obtained from Sigma Company in the market. However, all the commercial XDHs are suffered in low catalytic activity and efficiency, in comparison to other industrial enzymes with hundreds of thousands of turnover numbers. There is an urgent need to improve the productivity of active XDH and XOD, it is worth trying to overexpress the helper proteins to assist the synthesis and assembly processes of active XDH and then its posttranslational modification to XOD.
The most recognized roles of XDH/XOD are key enzymes in catabolizing purines and other heterocyclic compounds with sp2-hybridized carbon atoms like aldehydes. The former directly involves the formation of uric acid and its abnormity becomes the main cause of gout, and the latter takes part in detoxification of drugs and harmful intermediate metabolites. In addition, XOD has been implicated as a key oxidative enzyme to produce superoxide radicals and hydrogen peroxide molecules, of which the increase could lead to oxidative stress damage, metabolic syndrome, cardiovascular diseases such as hypertension and ischemia reperfusion.31 In recent years, the generation of reactive oxygen species (ROS) and nitric oxide (NO) in both physiological and pathological contexts like in ischemia-reperfusion injury has also been proposed.21 However, the structural complexity and specialized tissue distribution and highly regulation of XDH/XOD suggest other not fully identified functions, which require additional work to elucidate. The extensively distributed bacterial XDHs in different habitants and the largely different activity and substrate specificity may also relate to some physiological roles in comparison to mammal XDH/XODs.
XDH finds use in diverse applications, which include: drug treatment of heterocyclic compounds such as caffeine and hypoxanthine, nucleoside analogs production by eliminating by-products, diagnosis of purines and purine-relating enzymes in clinical, foods and environment, and biological remediation (Fig. 4). Of these, an important commercialized industrial application of XDH/XOD is the production of ribavirin, which is an anti-viral drug retrieved on the World Health Organization's List of Essential Medicines, a list of the most important medication needed in a basic health system.32 The timely removal of by-product of hypoxanthine by application of XDH/XOD shifts the reaction equilibrium to the direction of synthesising ribavirin, resulting in increased productivity.33 Another inspiring applications are the clinical detections of xanthine and hypoxanthine content in body fluidics, and other disease biomarkers like inorganic phosphorus, 5′-nucleotidase and adenosine deaminase by coupling with purine-nucleoside phosphorylase and uricase. There must be promising prospects in the widely adopted XOD-based test kits markets for the true XDHs, like RcXDH and AbXDH, which selectively utilize NAD but not molecular oxygen as electron receptor and can eliminate the requirement of unmanageable oxygen in test samples in comparison to XOD.
Although the biosynthesis of active XDH and its conversion into XOD form are complex multistep processes and require several helper proteins, the XDHs exist ubiquitously in all organisms of 3 domains of life. The XDHs show great variability in gene sequence, protein sequence, and the subunit composition, but they fold in a highly similar way, which lays the structural foundation for the same catalytic mechanism. The wide range of enzymatic activities and properties of XDHs can be a great advantage of survival in many different niches including extreme habitats and an efficient adaptive mechanism, although their specific physiological and pathological roles require further investigation. During the past more than one hundred years of research history, the XDHs are proposed to play significant roles in various cellular processes, including purine catabolism and production of reactive oxygen species (ROS) and nitric oxide (NO) in both physiological and pathological contexts and show great applicable prospects in drug metabolism, industrial synthesis of nucleoside analogs, clinical diagnosis and biological remediation. However, there is still a great demand for better catalytic activity, stability and productivity of XDHs from the view-point of practical applications.
As stated above, systematic research capable of linking discovery-engineering-production-application will promote the progress of XDHs. Some promising research aspects in the future include: 1) prospecting novel XDHs from extremophiles and metagenomes from various natural environments; 2) improving the existing enzymes by protein engineering, fusion protein or evolution approaches by combining the advanced high throughput technology; 3) designing and constructing engineered microbial cell factories for efficient XDH production with high activity and thermostability by overexpressing potential helper genes based on the complicated biosynthesis mechanism; 4) developing practical applications in diagnosis, detection, industrial biocatalysis and environmental remediation. The first future study will contribute to exploration of novel excellent enzymes more suitable for harsher operating environments with higher pH-stability, thermal stability and catalytic activity. This investigation will also help to understand the mechanisms responsible for the stability and activity of valuable XDHs in comparison with the 3 already well-characterized XDHs from bovine milk, R. capsulates and A. baumannii. The second research direction will evolve the already existing XDHs for improved stability, substrate affinity, activity and specificity to cater for the challenges to be solved by the first research direction. The commercially available XDHs and newly reported highly active AbXDH and RcXDH can be good starting materials. The third investigation may contribute to more efficient industrialized production of XDHs, which is an urgent task because the current inefficient extraction methods impede a variety of applications like industrial biocatalysis and environmental remediation due to the low efficiency and high cost. With these future researches on XDHs, it will be promising to expand their applications to in-situ enzymatic diagnosis and detection of target molecules such as purines and aldehydes in harsh environments using more sensitive but robust XDHs, and to detoxifying environmental pollutants like heterocyclic compounds including purines and aldehydes and synthesising nucleoside drugs like ribavirin in large scale.
No potential conflicts of interest are disclosed.
This work was supported by grants from the National Natural Science Foundation of China (Grant Number: 21406132), and the China Postdoctoral Science Foundation (Grant Numbers: 2014M550743 and 2015T80094).