SEARCH

SEARCH BY CITATION

Keywords:

  • GTP cyclohydrolase;
  • biosynthesis;
  • pteridines;
  • folate;
  • riboflavin;
  • reaction mechanism

Abstract

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

GTP cyclohydrolases generate the first committed intermediates for the biosynthesis of certain vitamins/cofactors (folic acid, riboflavin, deazaflavin, and tetrahydrobiopterin), deazapurine antibiotics, some t-RNA bases (queuosine, archaeosine), and the phytotoxin, toxoflavin. They depend on divalent cations for hydrolytic opening of the imidazole ring of the substrate, guanosine triphosphate (GTP). Surprisingly, the ring opening reaction is not the rate-limiting step for GTP cyclohydrolases I and II whose mechanism have been studied in some detail. GTP cyclohydrolase I, Ib, and II are potential targets for novel anti-infectives. Genetic factors modulating the activity of human GTP cyclohydrolase are highly pleiotropic, since the signal transponders whose biosyntheses require their participation (nitric oxide, catecholamines) impact a very wide range of physiological phenomena. Recent studies suggest that human GTP cyclohydrolase may become an oncology target. © 2013 IUBMB Life 65(4):310–322, 2013.

The chromophores of several universal or wide-spread coenzymes including the flavin cofactors, FMN and FAD, the deazaflavins derived from 7,8-didemethyl-8-hydroxy-5-deazariboflavin, the one-carbon transponders tetrahydrofolate (1) and tetrahydromethanopterin (4), and the redox cofactor tetrahydrobiopterin (2) are all derived from GTP (Scheme 1). The cofactor chromophores retain the pyrimidine ring carbon atoms, the nitrogen atoms 1, 3, and 9, and the position 4 carbonyl oxygen of the GTP precursor. Some coenzymes also retain N-7, the position 2 amino group and three carbon atoms of the ribosyl side chain, respectively. The chromophores are the business ends of the cofactors serving as redox transponders (flavins, deazaflavins, and tetrahydrobiopterin), one-carbon transponders (folates and methaonopterin), or photoreptors (flavins, deazaflavins, and folate derivatives).

thumbnail image

Scheme 1. Metabolites whose biosynthesis involves a GTP cyclohydrolase.

Download figure to PowerPoint

Work started in the 1950s had established that the chromophores of flavins and of folate are derived from a guanine nucleotide (for review see refs.1, 2). The biosynthesis of the GTP-derived cofactors in Scheme 2 starts with the opening of the imidazole ring of the nucleotide catalyzed by metal-dependent GTP cyclohydrolases (specifically, GTP cyclohydrolases I, Ia, II, III, and MptA) affording four different biosynthesis intermediates (dihydroneopterin 3′-triphosphate (10), dihydroneopterin 2′,3′-cyclophosphate (11), 2,5-diamino-6-ribosylamino-4(3H)-pyrimidinone 5′-phosphate (12), and 2-amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5′-phosphate (13) (Scheme 2).

thumbnail image

Scheme 2. GTP cyclohydrolases and their products.

Download figure to PowerPoint

The “classical” GTP cyclohydrolases, that is GTP cyclohydrolases I catalyzing the first committed step in the biosynthetic pathways of tetrahydrofolate and tetrahydrobiopterin, and GTP cyclohydrolase II catalyzing the first committed step in the biosynthesis of riboflavin were discovered in the 1970s. Both enzymes have been studied in considerable detail over the following decades.

During the last decade, whole genome sequencing and comparative genomics have been conducive to the discovery of several new GTP cyclohydrolase types, and the number of pathways known to involve these enzymes has increased. Not surprisingly, these novel additions are known in less detail than the classical enzymes. However, they open important novel avenues for future research. This review will be focused on structural and mechanistic details of the classical enzymes and on the discovery and function of the more recently found family members. No attempt will be made to provide a comprehensive history for the early decades of GTP cyclohydrolase research.

GTP Cyclohydrolase I

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

A database search for GTP cyclohydrolase I retrieves more than 1,400 entries, with a majority of those papers dealing with medical aspects. Early work on the enzymes of tetrahydrofolate biosynthesis was done in the 1960s in the labs of G. M. Brown, T. Shiota, and G. Guroff. Working with [8-14C]GTP as substrate, these pioneers showed the presence of an enzyme activity in microbial cell extracts that produced [8-14C]formate from GTP (for review see ref.2). Purification of GTP cyclohydrolase I of Escherichia coli to apparent homogeneity was reported by Yim and Brown in 1976 (3). The enzyme was shown to catalyze a ring expansion conducive to the formation of dihydroneopterin 3′-triphosphate (10, Scheme 2). The carbon atoms required for the formation of the pyrazine moiety as well as the position 6 polyol side chain are derived from the ribose moiety of the substrate, GTP.

A gene specifying GTP cyclohydrolase I was first cloned from E. coli, and its recombinant expression enabled the preparation of highly purified enzyme as a prerequisite for crystallization and X-ray structure determination. The structure of the enzyme from E. coli was reported in 1995 (4), but the metal requirement of the enzyme was not immediately recognized.

The X-ray crystal structure of human GTP cyclohydrolase I revealed, for the first time, the presence of essential zinc ions at the active sites (5, 6). The essential metal ions had escaped detection in the earlier crystallographic study of the E. coli enzyme, due to the use of EDTA-containing buffers for purification.

At least seven structures of wild type and mutant E. coli cyclohydrolase I have been reported (4, 7) (cf. Table 1). Structures have also been reported for GTP cyclohydrolase I of Yersinia pestis, Thermus thermophilus (8), Rattus norvegicus (9, 10), and Homo sapiens (5). The GTP cyclohydrolases I of eubacteria and of animals are all structurally similar, although they serve different metabolic pathways (tetrahydrofolate in bacteria, tetrahydrobiopterin in animals). They are toroid-shaped, d5-symmetric homodecamers (Figs. 1A and 1B). Their molecular symmetry implicates one 5-fold and five 2-fold symmetry axes that are orthogonal to the molecular 5-fold axis. Considering the interfaces in the complex multisubunit entity, it is appropriate to call the enzyme a pentamer of dimers, rather than a dimer of pentamers.

thumbnail image

Figure 1. GTP cyclohydrolase I of T. thermophilus (stereo pairs, RCSB accession code 1WUQ (8)). A: Viewed along the molecular c5 axis; (B) viewed along a molecular c2 axis; (C) with substrate GTP shown as sphere models; (D) active site; blue sphere, Zn2+; red spheres, fixed water molecules. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Table 1. X-ray structures of GTP cyclohydrolases
inline image

Each of the 10 topologically equivalent active sites is located at the interface of three respective subunits. These active sites are all located close to a plane that is defined by the five 2-fold axes (Fig. 1C). Each active site comprises an essential zinc ion that is coordinated by two cysteines and one histidine (Fig. 1D, Scheme 3). Moreover, the zinc ion coordinates a fixed water molecule that is believed to serve as the reactive agent for hydrolysis of the substrate's imidazole ring.

thumbnail image

Scheme 3. Active site topology of GTP cyclohydrolase I of E. coli.

Download figure to PowerPoint

The imidazole ring of the substrate, GTP (9), is located close to the zinc ion (Fig. 1D). More specifically, an aquo ligand of the zinc ion forms a bridge between the zinc ion and C-8 of the substrate's imidazole ring (7) (Scheme 3). The purine motif is involved in several hydrogen bonds. The triphosphate moiety is coordinated by several basic amino acid residues to which it is hydrogen-bonded.

The formation of dihydroneopterin 3′-triphosphate from GTP requires the hydrolytic cleavage of two carbon–nitrogen bonds (N-7/C-8 and C-8/N9), the remodeling the carbohydrate side chain by an Amadori rearrangement, and finally the closure of the dihydropyrazine ring (Scheme 4). This reaction trajectory has been studied in considerable detail by NMR spectroscopy (11), presteady state kinetics (12–14), and site-directed mutagenesis (4). The partial reaction sequence conducive to the release of formate is shown in more detail in Scheme 5.

thumbnail image

Scheme 4. Putative reaction trajectory of GTP cyclohydrolase I.

Download figure to PowerPoint

thumbnail image

Scheme 5. Formate release catalyzed by GTP cyclohydrolase I.

Download figure to PowerPoint

The GTP cyclohydrolases are all inherently slow enzymes, and that facilitates presteady state kinetic analysis using stopped-flow and quenched-flow techniques (12–14). Several optical transients could be identified and were associated with hypothetical intermediates in Schemes 4 and 5. Moreover, the consumption of substrate GTP and the transient formation of the formamide type intermediate 15 were also monitored by high-performance liquid chromatography (HPLC) analysis. These studies provided apparent rate constants for partial reactions. Surprisingly, the rate-determining step occurs late in the reaction trajectory, whereas the opening of the imidazole ring is so rapid that it could not be dissected further by presteady state kinetics (13).

Based on X-ray structure data, a comprehensive mutagenesis project was undertaken in order to modify inner shell residues of the GTP cyclohydrolase I active site (4). Importantly, the replacement of H179 of the E. coli enzyme interrupts the reaction trajectory at the level of the formamide derivate 15 (15). The mutant can be used to prepare that putative reaction intermediate, which was subsequently shown to serve as a kinetically competent substrate for GTP cyclohydrolase I. Hence, it appears safe to assume that the enzyme reaction begins with the hydrolytic cleavage of the C-8/N-9 bond (Scheme 5). Using the formamide intermediate as substrate, the equilibrium constant for the ring opening reaction was found in the range of about 0.1 (15).

The water molecule that is coordinated by the zinc ion is believed to attack C-8 of the substrate under formation of the hydrate intermediate 14 (7). In line with this interpretation, 8-oxo-GTP has been shown to be a strong inhibitor of the enzyme that may be a mimic of the formamide intermediate resulting from the cleavage of the C-8/N-9 bond of 14 (8). The zinc ion may be also involved in the hydrolysis of the formamide motif of 15. The subsequent Amadori rearrangement is stereospecific; a solvent proton has been shown to be incorporated into the pro-7R position of dihydroneopterin triphosphate (Scheme 4) (11, 16).

GTP Cyclohydrolase Feedback Regulatory Protein

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

By comparison with eubacteria, GTP cycohydrolase of animals has an extended N-terminal segment; otherwise, the sequences and structures are similar. The N-terminal segment plays an important role for the interaction of the mammalian GTP cyclohydrolase I with its feedback regulatory protein [frequently addressed in the literature as GTP cyclohydrolase feedback regulatory protein (GFRP)] that was discovered in the 1990s (17). GFRP is a c5-symmetric, ring-shaped homopentamer (Fig. 2A) (18–20). Two GFRP molecules associate with one homodecameric GTP cyclohydrolase I molecule as shown in Fig. 2B (10, 11). The entire enzyme/inhibitor complex obeys d5 symmetry.

thumbnail image

Figure 2. A: Rat GTP cyclohydrolase I feedback regulatory protein (GFRP) viewed along the molecular c5 axis (stereo pair, RCSB accession code 1JG5 (20)); (B) complex of rat GTP cyclohydrolase I in complex with GFRP, viewed along a molecular c2 axis (stereo pair, RCSB accession code 1IS7 (9)); (C) complex of rat GTP cyclohydrolase I in complex with GFRP, viewed along a molecular c2 axis (stereo pair); bound phenylalanine shown as sphere model. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Whereas the isolated homodecameric enzyme, without the inhibitor protein, is not subject to feedback inhibition, the enzyme/inhibitor complex follows a complex regulatory pattern. Tetrahydrobiopterin (2) acts as an inhibitor (Scheme 6). On the other hand, phenylalanine is an activator (the catabolism of phenylalanine requires tetrahydrobiopterin as cofactor, and the synthesis of that cofactor is enhanced by an excess of phenylalanine that needs to be catabolized).

thumbnail image

Scheme 6. Regulatory properties of mammalian GTP cyclohydrolase I in complex with GTP cyclohydrolase I feedback regulatory protein (GFRP). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

GFRP can bind to its target enzyme in the absence of agonists (tetrahydrobiopterin, phenylalanine), conducive to an increase of vmax and of the Hill coefficient and a decrease of Km (21).

Phenylalanine has been shown to bind at the enzyme/GFRP interfaces with a total of 10 topologically equivalent binding sites (Fig. 2C) (10) The N-terminal segment of the enzyme is disordered in crystals; it is believed to have significant mobility that is relevant for the allosteric features of the enzyme/GFRP complex.

2,4-Diamino-6(1H)-pyrimidinone (also known as 2,4-diamino-6-hydroxypyrimidine, DAHP) has played an important role in GTP cyclohydrolase I research as an inhibitor. Notably, its inhibitory activity depends on the presence of GFRP (22, 23).

GTP Cylohydrolase I of Plants

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

Animals are unable to synthesize folate type cofactors, for which they depend on dietary sources. Even though folates can be handed down in the food chain, their primary sources are plants and bacteria. Surprisingly, plant genomes do not specify homologs of the decameric tetrahydrofolate synthase found in animals and microorganisms. However, a bioinformatics approach identified a tomato gene specifying a protein comprising two fused domains, each with significant similarity to GTP cycohydrolase I of microbial or animal origin (24). The recombinant expression of the tomato open reading frame afforded a homodimer with GTP cyclohydrolase I activity, with subunits of about 50 kDa, about twice the size of microbial and animal GTP cyclohydrolase I. The sequence similarity between the two plant domains is lower than the similarity of each respective plant domain with the decameric enzymes of microbes and animals. Hence, it appears likely that the bimodal architecture of the plant gene has a long evolutionary history.

The well-known structure of the bacterial and animal orthologs implicates a set of crucial active site amino acid residues, at the interface of three adjacent subunits. Neither of the putative domains of the plant enzyme has a conserved set of all canonical residues, but each domain has a subset of conserved residues. Thus, the two domains could jointly provide all necessary components of an active site, which could then be formed at the interface between N-terminal and C-terminal domains.

Since the gene coding for plant GTP cyclohydrolase I does not specify an N-terminal targeting sequence, the protein should be located in the cytoplasmic compartment.

GTP Cyclohydrolase IB

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

More than thousand microbial genomes have been sequenced in recent years. Systematic bioinformatics searches failed to identify GTP cyclohydrolase I orthologs in archaea and in about one-fifth of sequenced eubacteria. In eubacteria carrying out the biosynthesis of folate cofactors in the apparent absence of GTP cyclohydrolase I, the bioinformatics search was conducive to the discovery of GTP cyclohydrolase Ib which is devoid of sequence similarity with the classical, homodecameric enzyme (25).

GTP cyclohydrolase Ib occurs in archaea and in about 20% of eubacteria. Notably, however, whereas the Ib type enzymes from eubacteria produce dihydroneopterin triphosphate (10), the archaeal sequence homologs are really GTP cycloyhdrolase Ib paralogs which produce dihydroneopterin 2′,3′-cyclophosphate (11) and have been renamed MptA since they catalyze the first committed step in the biosynthesis of tetrahydromethanopterin (4) (26). In contrast to the classical GTP cyclohydrolase I, the eubacterial type Ib enzyme is not specifically zinc-dependent and can use various divalent cations (Mn2+, Fe2+, Mg2+, Co2+, Zn2+, Ni2+) as cofactors, with Mn2+ supporting the highest level of catalytic activity.

Certain eubacteria carry genes for decameric GTP cyclohydrolase I as well as GTP cyclohydrolase Ib; whereas the decameric enzyme is constitutivey expressed, the type Ib enzyme is specifically expressed under conditions of zinc starvation.

GTP cyclohydrolase Ib is a homotetramer (Fig. 3) (27). The large subunits fold into two domains that are both members of the tunnel (T-fold) superfamily, which also includes the decameric GTP cyclohydrolase I. The active sites of the type Ib enzymes are located at the interfaces of three respective subunit domains, reminiscent of the type I enzyme's active site topology. The catalytic metal ion of the type Ib enzyme is coordinated by a histidine and a glutamate residue. Crystal structures have been determined of the Zn+2 and the Mn2+ form. The enzyme could possibly serve as target for novel drugs directed against methicillin-resistant Staphyolococcus aureus.

thumbnail image

Figure 3. GTP cyclohydrolase Ib of N. gonorrhoeae viewed along the molecular c2 axis (stereo pair, RCSB accession code 3D2O (27)); blue sphere, Mn2+. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Archaeal GTP Cyclohydrolase MPTA

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

As reported above, archaea and a significant subgroup of eubacteria are devoid of GTP cyclohydrolase I genes which are replaced by members of the type Ib family, with the qualification that the similarity of the type Ib genes of archaea with those of eubacterial origin is relatively low. The recombinant expression of the type Ib gene of Methanocaldococcus jannaschii afforded a protein that required Fe2+ for activity. However, whereas the type Ib GTP cyclohydrolases and the type Ib enzymes of eubacterial origin produce dihydroneopterin 3′-triphosphate, the archaeal enzyme was shown to produce dihydroneopterin 2′,3′-cyclophosphate (11) (26). The reaction product is believed to serve as the first committed intermediate in the biosynthesis of tetrahydromethanopterin (4), an essential cofactor for methane generation in methanogenic archaea.

The replacement of any of three conserved histidine residues of the archaeal enzyme that are believed to coordinate a catalytic Fe2+ ion was conducive to lowered activity, although significant residual activity remained.

An Fe2+-dependent hydrolase from M. jannaschii has been reported to open the pentacyclic cyclophosphate motif of dihydroneopterin 2′,3′-cyclophosphate, affording a mixture of the 2′ and 3′ monophosphates of dihydroneopterin (28). The hydrolase has been designated MptB and is believed to catalyze the second step of methanopterin biosynthesis.

GTP Cyclohydrolase II

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

In the course of studies on GTP cyclohydrolase I, Foor and Brown (29) reported a novel enzyme from E. coli that catalyzed the release of [14C]formate from [8-14C]GTP but was rapidly and completely inactivated by EDTA, due to a requirement for Mg2+ ions. Surprisingly, the new enzyme, designated GTP cyclohydrolase II, was found to not only catalyze the release of formate from the pyrimidine ring of GTP, but also the release of inorganic pyrophosphate. Two different molecular motifs, that is the imidazole ring of the guanine moiety and the remote triphosphate moiety at the 5′ end of the ribose moiety appeared to be attacked in virtual synchrony.

When highly purified protein GTP cyclohydrolase II became available by way of recombinant expression, GMP (18) was identified as a second reaction product of GTP cyclohydrolase. More specifically, the ratio of the main product, 2,5-diamino-6-ribosylamino-pyrimidine-4 one (12) and GMP (18) were found to be produced at an approximate ratio of 10:1 (30). In other words, GTP cyclohydrolase II is a slow GTPase, if only in terms of a side reaction.

With Hmath imageO as solvent, GTP cyclohydrolase II incorporates 18O from solvent into the organic product 12 rather than into inorganic pyrophosphate (30). This suggests that an early reaction step affords a covalently protein-bound guanylate intermediate, under release of pyrophosphate. A reaction sequence of imidazole ring opening, followed by formate release and phosphodiester hydrolysis, would then yield [18O]-12, whereas phosphodiester hydrolysis without prior ring opening affords [18O]GMP. The formation of a covalently bound diester intermediate is well in line with crystallographic evidence (31). Also in line with these findings, GTP cyclohydrolase II has been shown to convert 8-oxo-GTP into 8-oxo-GMP, although it does not catalyze the hydrolytic opening of the oxidized imidazole ring of 8-oxo-GTP (32). 8-Oxo-dGTP can also serve as substrate, this explains the action of GTP cyclohydrolase as a suppressor of mutT mutations.

Surprisingly, presteady state kinetic analysis revealed the formation, early in the reaction trajectory, of the phosphodiester adduct 16 as the rate-determining step. 2-Amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5′-triphosphate (15) can serve as substrate but does not fulfill the criteria for a kinetically competent intermediate and is thus not located on the reaction trajectory from GTP to 12 (33).

Streptomyces coelicolor codes for three proteins that can be assigned to the GTP cyclohydrolase II family. One of these proteins appears to be a GTP cyclohydrolase II paralog producing the formamide 18 rather than the 5-aminopyrimidine 12 (34).

Like GTP cyclohydrolase I, the cyclohydrolase II depends on Zn2+ ions for catalytic activity (Scheme 8) (35). The replacement of one of the cysteine residues 54, 65, or 67 of the E. coli enzyme abrogates the binding of zinc ions, and the mutant proteins are unable to hydrolytically open the imidazole ring. However, they can still catalyze the removal of pyrophosphate from GTP and the formamide derivative 15, respectively, thus supporting the view that the formation of a covalent phosphodiester intermediate and its hydrolytic cleavage can occur independently of ring opening.7

thumbnail image

Scheme 7. Putative reaction trajectory of GTP cyclohydrolase II. Redrawn from ref. 30.

Download figure to PowerPoint

thumbnail image

Scheme 8. Formate release catalyzed by GTP cyclohydrolase II.

Download figure to PowerPoint

The results of the mutation studies are all supported by the crystal structure of GTP cyclohydrolase II (Fig. 4) (31). The zinc ion at the active site of the E. coli enzyme is indeed coordinated by three cysteines of a CX2GX7CXC motif (Scheme 9). Moreover, in parallel to GTP cyclohydrolase I, a water molecule serves as an additional ligand to the zinc ion of GTP cyclohydrolase II and is strategically positioned to attack C-8 of the bound GMP. In the complex of GTP cyclohydrolase II with a hydrolysis-resistant GTP analog, a magnesium ion is straddling the α and β phosphate moieties of the triphosphate motif in a strategic position to interact with the phosphoanhydride oxygen that connects these moieties, which are simultaneously hydrogen-bonded to Arg128, the amino acid residue that has been proposed to attack the triphosphate motif under formation of a phosphoamide bond (31). All available evidence supports the concept that ring hydrolysis by GTP cyclohydrolyse II must be preceded by formation of the covalent guanyl adduct. This explains the early finding (29) that formate is not produced in the presence of EDTA, which traps the magnesium ions required for covalent guanylation of Arg128. It should be noted that, in contrast to GTP cyclohydrolase I, the type II enzyme operates by way of covalent catalysis.

thumbnail image

Figure 4. GTP cyclohydrolase II of Escherichia coli (RCSB accession code 2BZ0 (31)). A: Viewed along the molecular c2 axis, bound substrate analog (phosphomethylphosphonic acid guanyl ester, GMPCPP) shown as sphere model (stereo pair). B: Active site with bound substrate analog (phosphomethylphosphonic acid guanyl ester, GMPCPP); blue sphere, Zn2+; green sphere, Mg2+; red spheres, water (stereo pair). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

thumbnail image

Scheme 9. Active site topology of GTP cyclohydrolase II of E. coli.

Download figure to PowerPoint

Bifunctional GTP Cyclohydrolase II/3,4-Dihydroxy-2-butanone 4-phosphate Synthase

Certain eubacteria have homodimeric, monofunctional GTP cyclohydrolase II as described above. In other eubacteria and in plants, a GTP cyclohydrolase II domain is fused with a 3,4-dihydroxy-2-butanone 4-phosphate synthase domain (36). Thus, the bifunctional enzymes can catalyze both starting reactions of the convergent riboflavin pathway (Scheme 10). Structurally, the bifunctional enzymes are homodimers. The bifunctional plant enzymes are expressed with an N-terminal plastid-targeting sequence which is believed to be removed during import into chloroplasts.

thumbnail image

Scheme 10. Biosynthesis of riboflavin. Reactions catalyzed by bifunctional GTP cyclohydrolase II/3,4-dihydroxy-2-butanone 4-phosphate synthase of plants and certain eubacteria are shown in color. For details see ref.1. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

GTP cyclohydrolase II is believed to catalyze the rate-limiting step in the biosynthesis of riboflavin (37). Since riboflavin is a bulk commodity produced by fermentation, attempts have been made to increase its turnover by in vitro evolution. However, although vmax of the enzyme is really quite low, in the nmol mg−1 min−1 range, only modest rate enhancements were achieved by mutagenesis (38).

GTP Cyclohydrolase III

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

Archaea have no apparent orthologs of GTP cyclohydrolase II, but use instead a recently discovered enzyme type designated GTP cyclohydrolase III (39). The protein has no sequence similarity with GTP cyclohydrolase II of bacteria and plants. Its products have been reported to be 2-amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5′-phosphate (13) and two equivalents of orthophosphate, whereas the products of GTP cyclohydrolase II had been shown earlier to be 12 and inorganic pyrophosphate. Notably, however, earlier work had also shown that an enzyme-bound formamide 17 serves as an intermediate in GTP cyclohydrolase II catalysis. Also of note, the S. coelicolor genome specifies three GTP cyclohydrolase II homologs, and one of them has been found to produce the formamide 13 that is also produced by GTP cyclohydrolase III (34).

An archaeal Fe(II) dependent hydrolase has been shown to convert the formamide intermediate 13 into 2,5-diamino-6-ribitylamino-4(3H)-pyrimidinone 5′-phosphate (12) (40).

GTP cyclohydrolase III is a homotetramer (41). Each subunit folds into two domains with closely similar folding patterns (Fig. 5). Hence, it appears likely that the cognate gene originated by gene duplication. The subunit fold has similarity with certain phosphodiesterase domains.

thumbnail image

Figure 5. GTP cyclohydrolase III of M. jannaschii (RCSB accession code 2QV6 (41)). A: Viewed along the molecular c2 axis; bound GTP shown as sphere model (stereo pair). B: Active site with bound GTP; blue spheres, divalent metal ions; red spheres, water (stereo pair). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

The enzyme requires Mg2+ ions for catalytic activity and can be activated by monovalent cations (K+, NHmath image). The crystal structure shows three metal ion binding sites. Two aspartate residues have been shown to be involved in binding of divalent cations. The reaction mechanism of GTP cyclohydrolase III has not been studied in closer detail.

Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

Starting with the genome of an important human pathogen, Haemophilus influenza, whole genome sequencing has provided virtually complete coverage of human pathogens during the past two decades. Around the turn of the century, hopes were high that the mining of this treasure trove by the combined use of combinatorial chemistry and computational tools should afford convenient access to novel drugs in order to fight tropical diseases and to counteract the spreading of pathogen resistance to virtually all current anti-infective drugs. These high hopes have been rather incompletely fulfilled; in fact, all large drug companies have meanwhile scaled down or outsourced their antibiotics research activities. More often than not, even the scarce novel anti-infective drugs that have been reaching the market in recent years have been follow-up derivatives of existing drugs rather than innovative chemical entities. Economic considerations are likely to have been important factors. Notably, however, the expected bonanza of novel druggable anti-infective targets has hitherto failed to materialize.

Many human pathogens are critically dependent on endogenous biosynthesis of riboflavin and tetrahydrofolate; some, possibly including Mycobacterium tuberculosis, may also be critically dependent on biosynthesis of deazaflavins. Inhibitors of folate biosynthesis and turnover (sulfonamides, pyrimethamin) have a long and highly successful history. In fact, the sulfonamides, which were introduced in the 1930s as the first synthetic agents with a broad spectrum of target pathogens, are still playing an important role, for example in the therapy of malaria.

GTP cyclohydrolase I, Ia, and II catalyze the first committed steps in the respective biosynthetic pathway conducive to flavocoenzymes, deazaflavin coenzymes, and the tetrahydrofolate coenzyme family. The enzymes are likely to be rate-limiting for the respective pathways. The flavin and deazaflavin pathways are absent in animals, and their putative inhibitors should be exempt from target-related toxicity.

Whereas animals are unable to biosynthesize folates, they do use GTP cyclohydrolase I (but not GTP cyclohydrolase Ib) for the biosynthesis of tetrahydrobiopterin. GTP cyclohydrolase Ib could therefore be a target for the development of non-broadband anti-infectives. On the other hand, for putative drugs directed against bacterial cyclohydrolase I, at least some selectivity with regard to the human enzyme would be required.

Medical Aspects of Human GTP Cyclohydrolase I

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References

Inherited GTP Cyclohydrolase I Deficiency

Mutations of enzymes of the human tetrahydrobiopterin biosynthesis enzymes including GTP cyclohydrolase I can cause atypical phenylketonuria with massive developmental and neurological deficits. Less deleterious mutations of GTP cyclohydrolase cause DYT5 Dopa-responsive dystonia, one out of more than a dozen hereditary dystonia forms. The abundant literature on these pathogenic entities has been covered in recent reviews (42, 43).

A Pain-Protective Haplotype of the Human GTP Cyclohydrolase I Gene

Pioneering work reported in 2006 established a correlation of human GTP cyclohydrolase I polymorphism with pain sensitivity (44, 45). Specifically, a haplotype occurring with a frequency of about 15% showed reduced experimental pain sensitivity and reduced pain sensitivity after surgical procedures. The pain-protective haplotype comprises specific nucleotides at 15 positions, all of them in non-coding gene segments (46) and is believed to prevent excessive injury-evoked de novo tetrahydrobiopterin synthesis via modulation of GTP cyclohydrolase I activity. In line with that hypothesis, leukocytes of carriers show reduced upregulation of GTP cyclohydrolase I activity after stimulation.

These seminal findings have already generated considerable research activity and the GTP cyclohydrolase I/GFRP complex now appears as a promising new drug target (47). Importantly, recent research also shows that inhibition of GTP cyclohydrolase I reduces tumor proliferation (48). Thus, a putative single drug targeted at GTP cyclohydrolase I might have the potential to cope simultaneously with tumor growth and with tumor-induced pain. Also of note, however, therapeutic reduction of GTP cyclohydrolase I activity might enhance cardiovascular risks (49).

References

  1. Top of page
  2. Abstract
  3. GTP Cyclohydrolase I
  4. GTP Cyclohydrolase Feedback Regulatory Protein
  5. GTP Cylohydrolase I of Plants
  6. GTP Cyclohydrolase IB
  7. Archaeal GTP Cyclohydrolase MPTA
  8. GTP Cyclohydrolase II
  9. GTP Cyclohydrolase III
  10. Microbial GTP Cyclohydrolases as Putative Targets for Novel Antiinfective Agents
  11. Medical Aspects of Human GTP Cyclohydrolase I
  12. References
  • 1
    Fischer, M. and Bacher, A. ( 2005) Biosynthesis of flavocoenzymes. Nat. Prod. Rep. 22, 324350.
  • 2
    Fischer, M., Thöny, B. and Leimkühler, S. ( 2010) The biosynthesis of folate and pterins and their enzymology. In Comprehensive Natural Products II. Chemistry and Biology, vol. 7. ( Mander, L., Lui, H.-W., eds.). pp. 599648, Elsevier, Oxford, England.
  • 3
    Yim, J. J. and Brown, G. M. ( 1976) Characteristics of guanosine triphosphate cyclohydrolase I purified from Escherichia coli. J. Biol. Chem. 254, 50875094.
  • 4
    Nar, H., Auerbach, G., Huber, R., Fischer, and M., Hösl, C. ( 1995) Active site topology and reaction mechanism of GTP cyclohydrolase I. J. Biol. Chem. 92, 1212012125.
  • 5
    Auerbach, G., Herrmann, A., Bracher, A., Bader, G., Gütlich, M. et al. ( 2000) Zinc plays a key role in human and bacterial GTP cyclohydrolase I. Proc. Natl. Acad. Sci. USA 97, 1356713572.
  • 6
    Suzuki, T., Kurita, H., and Ichinose, H. ( 2004) GTP cyclohydrolase I utilizes metal-free GTP as its substrate. Eur. J. Biochem. 271, 349355.
  • 7
    Rebelo, J., Auerbach, G., Bader, G., Bracher, A., Nar, H. et al. ( 2003) Biosynthesis of pteridines. Reaction mechanism of GTP cyclohydrolase I. J. Mol. Biol. 326, 503516
  • 8
    Tanaka, Y., Nakagawa, N., Kuramitsu, S., Yokoyama, S., and Masui, R. ( 2005) Novel reaction mechanism of GTP cyclohydrolase I. High-resolution X-ray crystallography of Thermus thermophilus HB8 enzyme complexed with a transition state analogue, the 8-oxoguanine derivative. J. Biochem. 138, 263275.
  • 9
    Maita, N., Okada, K., Hatakeyama, K., and Hakoshima, T. ( 2002) Crystal structure of the stimulatory complex of GTP cyclohydrolase I and its feedback regulatory protein GFRP. Proc. Natl. Acad. Sci. USA 99, 12121217.
  • 10
    Maita, N., Hatakeyama, K., Okada, K., and Hakoshima, T. ( 2004) Structural basis of biopterin-induced inhibition of GTP cyclohydrolase I by GFRP, its feedback regulatory protein. J. Biol. Chem. 279, 5153451540.
  • 11
    Bracher, A., Eisenreich, W., Schramek, N., Ritz, H., Götze, E. et al. ( 1998) Biosynthesis of pteridines. NMR studies on the reaction mechanism of GTP cyclohydrolase I, pyruvoyltetrahydropterin synthase, and sepiapterin reductase. J. Biol. Chem. 273, 2813228141.
  • 12
    Bracher, A., Schramek, N., and Bacher, A. ( 2001) Biosynthesis of Pteridines. Stopped-Flow Kinetic Analysis of GTP cyclohydrolase I. Biochemistry 40, 78967902.
  • 13
    Schramek, N., Bracher, A., and Bacher, A. ( 2001) Ring opening is not rate-limiting in the GTP cyclohydrolase I reaction. J. Biol. Chem. 276, 26222626.
  • 14
    Schramek, N., Bracher, A., Fischer, M., Auerbach, G., and Nar, H. ( 2002) Reaction Mechanism of GTP cyclohydrolase. I. Single turnover experiments using a kinetically competent reaction intermediate. J. Mol. Biol. 316, 829837.
  • 15
    Bracher, A., Fischer, M., Eisenreich, W., Ritz, H., Nar, H., et al. ( 1999) Histidine 179 mutants of GTP cyclohydrolase I catalyze the formation of 2-amino-5-formylamino-6-ribofuranosylamino-4(3H)-pyrimidinone triphosphate. J. Biol. Chem. 274, 1672716735.
  • 16
    Illarionova, V., Eisenreich, W., Fischer, M., Haußmann, and C., Römisch, W. ( 2002) Biosynthesis of Tetrahydrofolate. Stereochemistry of dihydroneopterin aldolase. J. Biol. Chem. 32, 2884128847.
  • 17
    Harada, T., Kagamiyama, H., and Hatakeyama, K. ( 1993) Feedback regulation mechanisms for the control of GTP cyclohydrolase I activity. Science 260, 15071510.
  • 18
    Yoneyama, T., Brewer J. M., and Hatakeyama, K. ( 1997) GTP Cyclohydrolase I feedback regulatory protein is a pentamer of identical subunits. Purification, cDNA cloning, and bacterial expression. J. Biol. Chem. 272, 96909696.
  • 19
    Yoneyama, T. and Hatakeyama, K. ( 1998) Decameric GTP cyclohydrolase I forms complexes with two pentameric GTP cyclohydrolase I feedback regulatory proteins in the presence of phenylalanine or of a combination of tetrahydrobiopterin and GTP. J. Biol. Chem. 273, 2010220108.
  • 20
    Bader, G., Schiffmann, S., Herrmann, A., Fischer, M., Gütlich, M., et al. ( 2001) Crystal structure of rat GTP cyclohydrolase I feedback regulatory protein, GFRP. J. Mol. Biol. 312, 10511057
  • 21
    Higgins, C. E. and Gross, S. S. ( 2011) The N-terminal peptide of mammalian GTP cyclohydrolase I is an autoinhibitory control element and contributes to binding the allosteric regulatory protein GFRP. J. Biol. Chem. 286, 1191911928.
  • 22
    Xie, L., Smith, J. A., and Gross, S. S. ( 1998) GTP cyclohydrolase I inhibition by the prototypic inhibitor 2,4-diamino-6-hydroxypyrimidine. Mechanisms and unanticipated role of GTP cyclohydrolase I feedback regulatory protein. J. Biol. Chem. 273, 2109121098.
  • 23
    Kolinsky, M. A. and Gross, S. S. ( 2004) The mechanism of potent GTP cyclohydrolase I inhibition by 2,4-diamino-6-hydroxypyrimidine: requirement of the GTP cyclohydrolase I feedback regulatory protein. J. Biol. Chem. 279, 4067740682.
  • 24
    Basset, G., Quinlivan, E. P., Ziemak, M. J., Díaz de la Garza, R., and Fischer, M. ( 2002) Folate synthesis in plants: the first step of the pterin branch is mediated by a unique bimodular GTP cyclohydrolase I. Proc. Natl. Acad. Sci. USA 99, 1248912494.
  • 25
    El Yacoubi, B., Bonnett, S., Anderson, J. N., Swairjo, M. A., and Iwata-Reuyl, D. ( 2006) Discovery of a new prokaryotic type I GTP cyclohydrolase family. J. Biol. Chem. 49, 3758637593.
  • 26
    Grochowski, L. L., Xu, H., Leung, K., and White, R. H. ( 2007) Characterization of an Fe2+-dependent archaeal-specific GTP cyclohydrolase, MptA, from Methanocaldococcus jannaschii. Biochemistry 46, 66586667.
  • 27
    Sankaran, B., Bonnett, S. A., Shah, K., Gabriel, S., Reddy, R., et al. ( 2009) Zinc-independent folate biosynthesis: genetic, biochemical, and structural investigations reveal new metal dependence for GTP cyclohydrolase IB. J. Bacteriol. 191, 69366949.
  • 28
    Mashhadi, Z., Xu, H., and White, R.H. ( 2009) An Fe2+-dependent cyclic phosphodiesterase catalyzes the hydrolysis of 7,8-dihydro-D-neopterin 2′,3′-cyclic phosphate in methanopterin biosynthesis. Biochemistry 48, 93849392.
  • 29
    Foor, F. and Brown, G. M. ( 1975) Purification and properties of guanosine triphosphate cyclohydrolase II from Escherichia coli. J. Biol. Chem. 250, 35453551.
  • 30
    Ritz, H., Schramek, N., Bracher, A., Herz, S., Eisenreich, W., et al. ( 2001) Biosynthesis of riboflavin: studies on the mechanism of GTP cyclohydrolase II. J. Biol. Chem. 276, 2227322277.
  • 31
    Ren, J., Kotaka, M., Lockyer, M., Lamb, H. K., Hawkins, A. R., et. al. ( 2005) GTP cyclohydrolase II structure and mechanism. J. Biol. Chem. 280, 3691236919.
  • 32
    Kobayashi, M., Ohara-Nemoto, Y., Kaneko, M., Hayakawa, H., Sekiguchi, M., et al. ( 1998) Potential of Escherichia coli GTP cyclohydrolase II for hydrolyzing 8-oxo-dGTP, a mutagenic substrate for DNA synthesis. J. Biol. Chem. 273, 2639426399.
  • 33
    Schramek, N., Bracher, A., and Bacher, A. ( 2001) Biosynthesis of riboflavin. Single turnover kinetic analysis of GTP cyclohydrolase II. J. Biol. Chem. 276, 4415744162.
  • 34
    Spoonamore, J. E., Dahlgran, A. L., Jacobsen, N. E., and Bandarian, V. ( 2006) Evolution of new function in the GTP cyclohydrolase II proteins of Streptomyces coelicolor. Biochemistry, 45, 1214412155.
  • 35
    Kaiser, J., Schramek, N., Eberhardt, S., Püttmer, S., Schuster, M., et al. ( 2002) Biosynthesis of vitamin B2. An essential zinc ion at the catalytic site of GTP cyclohydrolase II. Eur. J. Biochem. 269, 52645270.
  • 36
    Herz, S., Eberhardt, S., and Bacher, A. ( 2000) Biosynthesis of riboflavin in plants. The ribA gene of Arabidopsis thaliana specifies a bifunctional GTP cyclohydrolase II/3,4-dihydroxy-2-butanone 4-phosphate synthase. Phytochemistry 53, 723731.
  • 37
    Hümbelin, M., Griesser, V., Keller, T., Schurter, W., Haiker, M., et al. ( 1999) GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone 4-phosphate synthase are rate-limiting enzymes in riboflavin synthesis of an industrial Bacillus subtilis strain used for riboflavin production. J. Ind. Microbiol. Biotechnol. 22, 17.
  • 38
    Lehmann, M., Degen, S., Hohmann, H. P., Wyss, M., Bacher, A., et al. ( 2009) Biosynthesis of riboflavin: screening for an improved GTP cyclohydrolase II mutant. FEBS J. 276, 41194129.
  • 39
    Graham, D. E., Xu, H., and White, R. H. ( 2002) A member of a new class of GTP cyclohydrolases produces formylaminopyrimidine nucleotide monophosphates. Biochemistry 41, 1507415084.
  • 40
    Grochowski, L. L., Xu, H., and White, R. H. ( 2009) An iron(II) dependent formamide hydrolase catalyzes the second step in the archaeal biosynthetic pathway to riboflavin and 7,8-didemethyl-8-hydroxy-5-deazariboflavin. Biochemistry 48, 41814188.
  • 41
    Morrison, S. D., Roberts, S. A., Zegeer, A. M., Montfort, W. R., and Bandarian, V. ( 2008) A new use for a familiar fold: the X-ray crystal structure of GTP-bound GTP cyclohydrolase III from Methanocaldococcus jannaschii reveals a two metal ion catalytic mechanism. Biochemistry 47, 230242.
  • 42
    Thöny, B. and Blau, N. ( 2006) Mutations in the BH4-Metabolizing genes GTP cyclohydrolase I, 6-pyruvoyl-tetrahydropterin synthase, sepiapterin reductase, carbinolamine-4a-dehydratase, and dihydropteridine reductase. Hum. Mutat. 27, 870878.
  • 43
    Blau, N. Inborn errors of H4-biopterin metabolism. IUBMB Life, in press.
  • 44
    Tegeder, I., Costigan, M., Griffin, R. S., Abele, A., Belfer, I., et al. ( 2006) GTP cyclohydrolase and tetrahydrobiopterin regulate pain sensitivity and persistence. Nat. Med. 12, 12691277.
  • 45
    Lötsch, J., Klepstad, P., Doehring, A., and Dale, O. ( 2010) A GTP cyclohydrolase 1 genetic variant delays cancer pain. Pain 148, 103106.
  • 46
    Lötsch, J., Belfer, I., Kirchhof, A., Mishra, B. K., Max M. B., et al. ( 2007) Reliable screening for a pain-protective haplotype in the GTP cyclohydrolase 1 gene (GCH1) through the use of 3 or fewer single nucleotide polymorphisms. Clin. Chem. 53, 101010155.
  • 47
    Klepstad, P., Fladvad, T., Skorpen, F., Bjordal, K., and Caraceni, A. ( 2011) Influence from genetic variability on opioid use for cancer pain: a European genetic association study of 2294 cancer pain patients. Pain 152, 11391145.
  • 48
    Pickert, G., Lim, H. Y., Weigert, A., Häussler, A., and Myrczek, T. ( 2012) Inhibition of GTP cyclohydrolase attenuates tumor growth by reducing angiogenesis and M2-like polarization of tumor associated macrophages.
  • 49
    Doehring, A., Antoniades, C., Channon, K. M., Tegeder, I., Lötsch, J., et al. ( 2008) Clinical genetics of functionally mild non-coding GTP cyclohydrolase 1 (GCH1) polymorphisms modulating pain and cardiovascular risk. Mutat Res./Rev. Mutat. Res. 659, 195201.