Báo cáo Y học: Biosynthesis of vitamin B2 An essential zinc ion at the catalytic site of GTP cyclohydrolase II ppt

Biosynthesis of vitamin B2An essential zinc ion at the catalytic site of GTP cyclohydrolase II Johannes Kaiser1, Nicholas Schramek1, Sabine Eberhardt1, Stefanie Pu¨ttmer2, Michael Schust

Biosynthesis of vitamin B2

An essential zinc ion at the catalytic site of GTP cyclohydrolase II

Johannes Kaiser1, Nicholas Schramek1, Sabine Eberhardt1, Stefanie Pu¨ttmer2, Michael Schuster2

and Adelbert Bacher1

1

Lehrstuhl fu¨r Organische Chemie und Biochemie, Technische Universita¨t Mu¨nchen, Garching, Germany;2Lehrstuhl fu¨r

Anorganische und Analytische Chemie, Technische Universita¨t Mu¨nchen, Garching, Germany

GTP cyclohydrolase II catalyzes the hydrolytic release

of formate and pyrophosphate from GTP producing

2,5-diamino-6-ribosylamino-4(3H)-pyrimidinone

5¢-phos-phate, the first committed intermediate in the biosynthesis of

riboflavin The enzyme was shown to contain one zinc ion

per subunit Replacement of cysteine residue 54, 65 or 67

with serine resulted in proteins devoid of bound zinc and

unable to release formate from the imidazole ring of GTP or

from the intermediate analog,

2-amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5¢-triphosphate

How-ever, the mutant proteins retained the capacity to release

pyrophosphate from GTP and from the formamide-type intermediate analog The data suggest that the enzyme catalyzes an ordered reaction in which the hydrolytic release

of pyrophosphate precedes the hydrolytic attack of the imidazole ring Ring opening and formate release are both dependent on a zinc ion acting as a Lewis acid, which acti-vates the two water molecules involved in the sequential hydrolysis of two carbon–nitrogen bonds

Keywords: formate; GTP cyclohydrolase; imidazole ring; pyrophosphate; zinc ion

GTP cyclohydrolases catalyze the first steps in the

biosyn-thetic pathways of riboflavin, tetrahydrofolate and

tetra-hydrobiopterin More specifically, GTP cyclohydrolase I

catalyzes the release of C8 of GTP (Compound 1, Fig 1)

followed by the formation of a novel pyrazine ring with

inclusion of carbon atoms 1¢ and 2¢ of the ribose side chain

[1,2] The reaction product, dihydroneopterin triphosphate

(Compound 3, Fig 1), is the first precursor in the

biosyn-thetic pathways of tetrahydrofolate and tetrahydrobiopterin

[3,4] GTP cyclohydrolase II catalyzes the hydrolytic release

of C8 of GTP accompanied by the release of pyrophosphate

from the carbohydrate side chain of GTP The enzyme

product, 2,5-diamino-6-ribosylamino-4(3H)-pyrimidinone

5¢-phosphate (Compound 4, Fig 1), is the first committed

precursor in the biosynthesis of riboflavin (vitamin B2) [5]

Recently, GTP cyclohydrolase II was also shown to catalyze

the formation of GMP from GTP at 10% the rate of

formation of the main product, Compound 4 [6] It was also

shown that the 5¢-triphosphates of

8-oxo-7,8-dihydro-2¢-deoxyguanosine and 8-oxo-7,8-dihydroguanosine can be

converted into the respective monophosphates, although the

enzyme is unable to open the imidazole ring of the

structurally modified guanine residues of these nucleotides

[7] Despite certain similarities in their reaction mechanisms,

GTP cyclohydrolases I and II have no detectable sequence

similarity

Recently, we found that GTP cyclohydrolase I contains

an essential zinc ion at each active site [8] Mutant proteins

unable to bind zinc are totally unable to catalyze the opening of the imidazole ring of GTP

In this paper, we show that a zinc ion complexed to three cysteine residues is absolutely required for the release of formate from GTP by GTP cyclohydrolase II, whereas the metal ion is not required for the enzyme-catalyzed release of pyrophosphate from the substrate

E X P E R I M E N T A L P R O C E D U R E S

Materials Oligonucleotides were custom-synthesized by MWG Bio-tech, Ebersberg, Germany Nucleotide triphosphates were purchased from Sigma-Aldrich Fine Chemicals, Munich, Germany 2-Amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5¢-triphosphate was prepared as described previously using the H179A mutant of GTP cyclohydrolase

I [9] DNA sequencing was performed by MWG Biotech Micro-organisms and plasmids

Bacterial strains and plasmids used in this study are summarized in Table 1

Site-directed mutagenesis Site-directed mutagenesis was performed by PCR using the overlap extension technique [10] PCR was performed with Pfx polymerase (Gibco BRL, Karlsruhe, Germany) to minimize the error rate The internal mismatch primers are shown in Table 1

The general scheme of mutagenetic PCR involved three rounds of amplification cycles using two mismatch and two flanking primers (primers MF and BamH1rev, Table 1) During the first round, 20 amplification cycles were carried

Correspondence to A Bacher, Lehrstuhl fu¨r Organische Chemie und

Biochemie, Technische Universita¨t Mu¨nchen, Lichtenbergstr 4,

D-85747 Garching, Germany Fax: + 49 89 289 13363,

Tel.: + 49 89 289 13360, E-mail: adelbert.bacher@ch.tum.de

(Received 12 June 2002, revised 5 September 2002,

accepted 9 September 2002)

out with one of the flanking primers and the corresponding

mismatch primer The plasmid pECH2 [11] was used as

template During the second amplification cycle, 20

ampli-fication cycles were carried out using the second flanking

primer and the corresponding mismatch primer The

plasmid pECH2 was used as template Compounds from

both amplification rounds were purified by agarose gel

electrophoresis During the third round, the products of both round one and two were used as templates, and 20 amplification cycles were carried out using the two flanking primers The resulting compound was subjected to agarose gel electrophoresis, digested with BamHI and EcoRI, purified using the QIAquick PCR purification kit, and ligated into plasmid pNCO-113 which had been digested with the same restriction enzymes The ligation mixture was transformed into Escherichia coli XL1-blue cells (Strata-gene, Heidelberg, Germany) All gene constructs were verified by DNA sequencing Protein expression was performed in E coli M15 (Table 2)

Enzyme purification Recombinant GTP cyclohydrolase II of E coli was purified

by published procedures [12]

Enzyme assays Enzyme-catalyzed formation of Compound 4 by GTP cyclohydrolase II was monitored by a published procedure [6]

Assay of pyrophosphate release Reaction mixtures containing 100 mM Tris/HCl, pH 8.0,

5 mMMgCl2, 2 mMsubstrate (GTP or Compound 2) and protein were incubated at 37C Aliquots of 100 lL were retrieved at intervals The reaction was stopped by removal

of the enzyme by ultrafiltration (Nanosep 10K Omega; Pall Life Sciences, Ann Arbor, MI, USA), and 30 lL aliquots of the filtrates were applied to an HPLC anion-exchange column (Gromsil SAX; 200· 4 mm; 5 lm; Grom, Herren-berg, Germany) The column was washed with 30 mL 5 mM ammonium phosphate, pH 2.7, and was developed with a linear gradient of 5–530 mM ammonium phosphate,

pH 3.8 The flow rate was 1 mLÆmin)1 The effluent was monitored photometrically (Knauer Wellchrom K-2600; Knauer, Berlin, Germany) at 254 nm, 272 nm, 293 nm and

330 nm

Zinc determination

A solution containing 2M HCl and 400 lgÆmL)1protein was incubated at 90C for 5 h and then analyzed using

a Unicam 919 flame atomic absorption spectrometer (Unicam, Cambridge, UK)

R E S U L T S Earlier studies on GTP cyclohydrolase II suggested the complex reaction pathway shown in Fig 1 [6] Briefly, it was proposed that the reaction is initiated by the hydrolytic release of pyrophosphate, possibly involving the formation

of a covalent phosphoguanosyl derivative of the enzyme Carbon 8 of the guanine moiety is then assumed to be released as formate by two consecutive hydrolytic reactions The reaction is believed to be terminated by hydrolysis of the presumed phosphodiester or phosphoamide bond between enzyme and substrate This reaction sequence can explain the formation of GMP as a side product accounting for about 10% of the enzyme activity [6]

Fig 1 Reactions catalyzed by GTP cyclohydrolases (A) GTP

cyclo-hydrolase I; (B) GTP cyclocyclo-hydrolase II [6].

The recent finding that GTP cyclohydrolase I requires

zinc for the hydrolytic opening of the imidazole ring of GTP

[8] prompted us to analyze GTP cyclohydrolase II from

E colifor the presence of zinc ions by atomic absorption

spectrometry As shown in Table 3, the wild-type protein

from E coli prepared by a published procedure [12] was

found to contain 0.71 zinc ions per subunit

As structural information on GTP cyclohydrolase II is

not available, we decided to screen for amino acids involved

in zinc chelation by site-directed mutagenesis Sequence

comparison revealed a CX2GX7CXC motif which occurs in

all cyclohydrolase II sequences (Fig 2) Each of the three

cysteine residues was replaced with serine by PCR-assisted

mutagenesis The mutations were confirmed by DNA

sequencing The mutant genes could be expressed to high

levels, and the proteins could be purified by the protocol

reported for the wild-type enzyme Replacement of any of

the three cysteine residues in GTP cyclohydrolase II resulted

in proteins that were devoid of zinc within the limit of

experimental accuracy (Table 3)

The conversion of GTP into the product, Compound 4,

can be monitored photometrically The series of spectra

shown in Fig 3 shows isosbestic points at 231 and 271 nm

Product formation can best be monitored photometrically

at 300 nm at which the substrate, GTP, shows only very low

absorbance

Each of the mutant proteins failed to convert GTP into

2,5-diamino-6-ribosylamino-4(3H)-pyrimidinone

5¢-phos-phate (Compound 4, Table 4) The catalytic rates were less

than 1 nmolÆmg)1Æmin)1 This translates into less than one product molecule formed per subunit and per hour

We have previously shown that 2-amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5¢-triphosphate (Com-pound 2, Table 4) can serve as substrate for GTP cyclohydrolase II, although it does not qualify as a kinetically competent intermediate [13] The compound is

Fig 2 Sequence comparison ofGTP cyclohydrolase II Aae, Aquifex aeolicus; Aac, Actinobacillus actinomycetemcomitans; Apl, Actinoba-cillus pleuropneumoniae; Ath, Arabidopsis thaliana; Bsu, BaActinoba-cillus sub-tilis; Cmu, Chlamydia muridarum; Cpn, Chlamydophila pneumoniae; Ctr, Chlamydia trachomatis; Eco, Escherichia coli; Hin, Haemophilus influenzae; Hpy, Helicobacter pylori; Mtu, Mycobacterium tuberculosis; Nme, Neisseria meningitides; Pgu, Pichia guilliermondii; Sce, Sac-charomyces cerevisiae; Ssp, Synechocystis species; Tma, Thermotoga maritima.

Table 1 Oligonucleotides used in this study Oligonucleotides were designed to hybridize to the sense (–) and antisense (+) strand of the ribA.

Designation

Primer

Table 3 Zinc content ofGTP cyclohydrolase II ofE coli.

Mutant Zn2+per subunit (mol/mol)

Table 2 Micro-organisms and plasmids used in this study.

E coli XL1-Blue recA1, endA1, gyrA96, thi–1, hsdR17, supE44,

relA1, lac [F¢, proAB, lacl q

ZDM15, Tn10(tetr)]

Stratagene [28]

E coli M15 [pREP4] lac, ara, gal, mtl, recA + , uvr + , [pREP4: lacl, kana r ] [29]

deformylated by the wild-type enzyme at a rate of

122 nmolÆmg)1Æmin)1 and has been interpreted as an

intermediate analog that can be converted into the enzyme

product, Compound 4, but does not occur as an

interme-diate in the physiological reaction starting with GTP as

substrate All mutants shown in Table 4 have lost the ability

to catalyze the release of formate from the formamide-type

compound It follows that a zinc ion is absolutely required

for the opening of the imidazole ring of GTP as well as for

the subsequent hydrolysis of the resulting formamide motif

of Compound 4

Studies with the H179A mutant of GTP cyclohydrolase I

had shown the formation of the formamide-type

Com-pound 2 (Fig 1) from GTP to be a reversible reaction with

an equilibrium constant of 0.1 at 30 C and pH 7.0 [9] It

was therefore in order to check whether the proteins under

study can catalyze ring closure of Compound 2 with

formation of a guanosine nucleotide Attempts to detect

GMP or GTP in reaction mixtures containing one of the

proteins in Table 3 and Compound 2 as substrate were

unsuccessful

We previously showed that GTP cyclohydrolase II

catalyzes the formation of GMP as a minor product by

the release of pyrophosphate from GTP [6] Specifically,

GMP was formed at 10% of the rate of formation of the

enzyme product, Compound 4 All mutants shown in

Table 3 can catalyze the formation of GMP from GTP,

albeit at a reduced velocity Specifically, the rate for the

C54S mutant was  60% of that of the wild-type, and

the relative rates of the C65S and C67S mutants were in the

range 10–20% It follows that the zinc ion is not required for

the release of pyrophosphate from GTP However, the

release of phosphate from GTP by the wild-type and mutant

proteins requires magnesium ions

The wild-type enzyme has been shown to catalyze the

release of pyrophosphate from the intermediate analog,

Compound 2 We have now found that the mutants in

Table 3 retain the ability to catalyze that reaction with for-mation of 2-amino-5-formylamino-6-ribosylamino-4(3H)-pyrimidinone 5¢-monophosphate, which was identified by ion-exchange HPLC (Table 4)

GMP and GDP do not serve as substrates for formation

of Compound 2 by wild-type GTP cyclohydrolase II, as shown already by Foor & Brown [5] These authors also reported that GTP cyclohydrolase II is unable to use nucleotide triphosphates other than GTP as substrate A reinvestigation using the recombinant E coli enzyme showed that pyrophosphate can be catalytically released from deoxyGTP by the wild-type enzyme as well as by the zinc-deficient mutants obtained by replacement of cysteine residues Moreover, the wild-type enzyme was able to catalyze ring-opening reactions with deoxyGTP as substrate

as shown by photometric analysis (Fig 4) UV absorbance changes observed with GTP and deoxyGTP were similar (data not shown) The rate of the ring-opening reaction catalyzed by wild-type GTP cyclohydrolase II was

182 nmolÆmin)1Æmg)1 for GTP and 38 nmolÆmin)1Æmg)1 for deoxyGTP as substrate

D I S C U S S I O N Flavin coenzymes are indispensable in all cellular organisms because of their involvement in redox processes of central metabolic pathways that are crucial for energy transduction The precursor of flavocoenzymes, riboflavin (vitamin B2), is

Fig 3 Ultraviolet spectra A reaction mixture containing 100 m M

Tris/HCl, pH 8.0, 10 m M MgCl 2 , 90 l M GTP, and 0.25 mg protein

was incubated at 30 C Spectra were recorded at intervals of 75 s.

Fig 4 Formation ofproducts from deoxyGTP by GTP cyclohydrolase II.

Table 4 Catalytic activity ofGTP cyclohydrolase II mutants The activity with different substrates (first row) and products (second row) is shown.

Protein

Activity (nmolÆmg)1Æmin)1) GTP

4

dGTP 12

2 4

GTP GMP

dGTP dGMP

2 9

biosynthesized by plants and many micro-organisms,

whereas animals depend on nutritional sources

For numerous pathogenic micro-organisms, the enzymes

of the riboflavin biosynthetic pathway are essential proteins

Specifically, Enterobacteriaceae are virtually unable to

absorb flavins from the environment and are therefore

absolutely dependent on their endogenous production [14]

The same has been shown for several yeast species including

Candida guilliermondii[15,16]

Mycobacterium tuberculosis and Mycobacterium leprae

both have complete sets of riboflavin biosynthesis genes As

these genes have apparently survived the extensive

frag-mentation of genes in M leprae [17], they are likely to be

essential for the intracellular lifestyle of Mycobacteria The

genes of riboflavin biosynthesis are therefore putative

targets for the treatment of infections caused by

Gram-negative bacteria and possibly by Mycobacteria and

pathogenic yeasts The exploration of novel anti-infective

targets is of supreme importance in the light of the rapid

progression of resistance development in all microbial

pathogens

In contrast with the riboflavin biosynthetic pathway, the

dihydrofolate pathway was already validated as an

anti-infective target in the first half of the last century (for reviews

see references [18,19]) In fact, sulfonamides inhibiting

dihydropteroate synthase were the first chemotherapeutic

agents with a broad antimicrobial and antiprotozoal

spectrum of activity Later, trimethoprim, an inhibitor of

dihydrofolate reductase, was introduced for the treatment

of bacterial infections, often in combination with

sulfona-mides

Both the riboflavin and tetrahydrofolate pathway start

from GTP (Fig 1) The first step of each pathway involves

the hydrolytic opening of the imidazole ring of the substrate

with formation of formate as a byproduct However, the

enzyme products are different in structure In the case of

GTP cyclohydrolase I, the ring-opening step is followed by

a complex series of reactions leading to formation of a

dihydropterin [1,2,20–24] Another difference is the release

of pyrophosphate by GTP cyclohydrolase II but not by

GTP cyclohydrolase I

Although GTP cyclohydrolase I has been known for

more than three decades, it was only recently shown that the

hydrolytic opening of the imidazole ring and the subsequent

release of formate requires a zinc ion acting as a Lewis acid,

which sequentially activates the two water molecules that

serve as nucleophiles in the two consecutive hydrolytic

reactions [8] Because of the mechanistic similarities of the

two different GTP cyclohydrolases, we investigated the type

II enzyme for the presence of zinc The data are consistent

with the presence of one zinc ion per subunit of the

homodimeric enzyme of E coli The comparison of

numer-ous putative GTP cyclohydrolase II sequences indicated a

pattern of three absolutely conserved thiols (Fig 2) The

sequence motif fits well with the typical short spacer/long

spacer motif found in a number of catalytic zinc-binding

sites [25]

The replacement of any of the three conserved cysteine

residues produced mutants with zinc levels below the level of

detection This suggests that the catalytic zinc ion is chelated

by cysteine residues 54, 65 and 67 of the E coli enzyme, and

that loss of any one of the three thiol groups is sufficient to

abolish the zinc-binding capacity of the protein For

comparison, the catalytic zinc of GTP cyclohydrolase I is chelated by two cysteine and one histidine residues, whereas

a second histidine residue contacts the metal via an interpolated water molecule (J Rebelo, G Auerbach, A Bracher, G Bader, H Nar, M Fischer, C Ho¨sl, N Schramek, J Kaiser, R Huber, and A Bacher, unpublished work) The replacement of any of the four amino-acid residues involved in zinc chelation is sufficient to abolish the zinc-binding capacity as well as the catalytic activity of the enzyme

The mutant proteins resulting from the replacement of any of the conserved cysteine residues in GTP cyclohydro-lase II with serine failed to catalyze the formation of the enzyme product, Compound 2, from GTP at a detectable rate Moreover, these mutants failed to release formate from the formamide-type Compound 4, which is a substrate of wild-type GTP cyclohydrolase II, although it lacks the characteristics of a kinetically competent intermediate [13]

We conclude that the catalytic action of zinc is required for the hydrolytic opening of the imidazole ring as well as for the subsequent hydrolysis of the formamide-type product 7 with formation of formate

These findings suggest the hypothetical mechanism shown in Fig 5 The formation of a covalent linkage between the substrate and the enzyme with formation of pyrophosphate is the first and rate-determining step [13] A nucleophilic attack by a zinc-activated water molecule leads

to the formation of a GTP hydrate, Compound 14 Cleavage of the C8–N9 bond leads to the formamide intermediate, Compound 15 In analogy with the mechan-ism of zinc proteases (Fig 6) [26], the co-ordination number

of zinc could then be increased to five through complexation

of an additional water molecule, which attacks the zinc-complexed formyl group of the intermediate 16 The resulting tetrahedral intermediate could lose formate, and product 4 could be released by hydrolysis of the covalent bond between enzyme and substrate In a final hydrolytic step, the product is released from the enzyme

It was recently shown that the 5¢-triphosphates of 8-oxo-7,8-dihydro-2¢-deoxyguanosine and 8-oxo-7,8-di-hydroguanosine can be converted into the respective monophosphates by GTP cyclohydrolase II, although the enzyme is unable to open the imidazole ring of the structurally modified guanine residues of these nucleotides [7] GTP cyclohydrolase II has also been shown to catalyze the conversion of GTP into GMP This side reaction occurs

at a rate of about 10% compared with the formation of the product, Compound 4, in the case of the wild-type enzyme

of E coli Mutants obtained by replacement of cysteine 54,

65 or 67 retain the capacity to produce GMP from GTP by hydrolytic release of pyrophosphate, although at a reduced rate It follows that zinc is not required for the hydrolytic release of pyrophosphate On the other hand, magnesium ions are required for pyrophosphate release In fact, none of the partial reactions specified in Fig 1 can be observed in the absence of magnesium ions

These observations are all consistent with the hypothesis

of an ordered mechanism in which the release of pyrophos-phate depending on the co-operation of magnesium ions must precede all other reaction steps (Fig 7) In parallel to many other reactions involving nucleoside triphosphates, magnesium may be required for complexation of the triphosphate motif before substrate binding The formation

of a covalent linkage between the substrate and GTP cyclohydrolase via a phosphodiester or phosphoamide motif is likely to be the rate-limiting step [27] After the hydrolytic release of formate, the covalent linkage between enzyme and reaction intermediates can be cleaved hydro-lytically Cleavage of the phosphodiester or phosphoamide bond can also occur without preliminary ring opening, thus affording GMP from GTP However, it should be noted that the covalent binding of the intermediate to the enzyme has not yet been documented directly

A C K N O W L E D G E M E N T S This work was supported by the Deutsche Forschungsgemeinschaft, by European Community Grant ERB FMRX CT98-0204, the Fonds der Chemischen Industrie and the Hans Fischer-Gesellschaft We thank Angelika Werner for expert help with the preparation of the manuscript.

R E F E R E N C E S

1 Burg, A.W & Brown, G.M (1968) The biosynthesis of folic acid VIII Purification and properties of the enzyme that catalyzes the production of formate from carbon atom 8 of guanosine tripho-sphate J Biol Chem 243, 2349–2358.

2 Shiota, T & Palumbo, M.P (1965) Enzymatic synthesis of the pteridine moiety of dihydrofolate from guanine nucleotides.

J Biol Chem 240, 4449–4453.

3 Brown, G.M & Williamson, J.M (1987) Biosynthesis of folic acid, riboflavin, thiamine, and pantothenic acid In Escherichia coli and Salmonella typhimurium (Neidhardt, F.C., Ingraham, J.L., Low, K.B., Magasanik, B., Schaechter, M & Umbarger, H.E., eds), pp 521–538 American Society for Microbiology, Wash-ington, DC.

4 Nichol, C.A., Smith, G.K & Duch, D.S (1985) Biosynthesis and metabolism of tetrahydrobiopterin and molybdopterin Annu Rev Biochem 54, 729–764.

5 Foor, F & Brown, G.M (1975) Purification and properties of guanosine triphosphate cyclohydrolase II from Escherichia coli.

J Biol Chem 250, 3545–3551.

6 Ritz, H., Schramek, N., Bracher, A., Herz, S., Eisenreich, W., Richter, G & Bacher, A (2001) Biosynthesis of riboflavin Studies

on the mechanism of GTP cyclohydrolase II J Biol Chem 276, 22273–22277.

7 Kobayashi, M., Ohara-Nemoto, Y., Kaneko, M., Hayakawa, H., Sekiguchi, M & Yamamoto, K (1998) Potential of Escherichia coli GTP cyclohydrolase II for hydrolyzing 8-oxo-dGTP, a mutagenic substrate for DNA synthesis J Biol Chem 273, 26394–26399.

8 Auerbach, G., Herrmann, A., Bracher, A., Bader, G., Gu¨tlich, M., Fischer, M., Neukamm, M., Garrido-Franco, M., Richardson, J., Nar, H., Huber, R & Bacher, A ( 2000) Zinc plays a key role in human and bacterial GTP cyclohydrolase I Proc Natl Acad Sci USA 97, 13567–13572.

9 Bracher, A., Fischer, M., Eisenreich, W., Ritz, H., Schramek, N., Boyle, P., Gentili, P., Huber, R., Nar, H., Auerbach, G & Bacher,

A (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, 16727– 16735.

10 Horton, R.M & Pease, L.R (1991) Recombination and muta-genesis of DNA sequences using PCR In Directed Mutamuta-genesis (McPherson, M.J., ed.), pp 217–246 Oxford University Press, New York.

11 Eberhardt, S., Richter, G., Ritz, H., Brandt, J & Bacher, A (1994) Biosynthesis of riboflavin Cloning, sequencing, mapping, and

Fig 5 Hypothetical mechanism for release of formate by GTP

cyclo-hydrolase II.

Fig 6 Reaction mechanism ofzinc proteases [26].

Fig 7 Cleland notation for the GTP cyclohydrolase II mechanism.

hyperexpression of the genes ribA coding for GTP cyclohydrolase

II and ribC coding for riboflavin synthase of Escherichia coli In

Flavins Flavoproteins 1993: Proceedings of the 11th International

Symposium pp 63–66 De Gruyter, Berlin, Germany.

12 Bacher, A., Richter, G., Ritz, H., Eberhardt, S., Fischer, M &

Krieger, C (1997) Biosynthesis of riboflavin: GTP cyclohydrolase

II, deaminase, and reductase Methods Enzymol 280, 382–389.

13 Schramek, N., Bracher, A & Bacher, A (2001) Biosynthesis of

riboflavin Single turnover kinetic analysis of GTP cyclohydrolase

II J Biol Chem 276, 44157–44162.

14 Shavlovskii, G.M., Tesliar, G.E & Strugovshchikova, L.P (1982)

Flavinogenesis regulation in riboflavin-dependent Escherichia coli

mutants Mikrobiologiia 51, 986–992.

15 Shavlovskii, G.M., Logvinenko, E.M., Kashchenko, V.E.,

Kol-tun, L.V & Zakal’skii, A.E (1976) Detection in Pichia

guillier-mondii of GTP-cyclohydrolase, an enzyme involved in the 1st stage

of flavinogenesis Dokl Akad Nauk SSSR 230, 1485–1487.

16 Oltmanns, O., Bacher, A & Lingens, F (1968) Growth of Candida

guilliermondii deficiency mutants with ethyl methanesulfonate Z.

Naturforsch B 23, 1556.

17 Cole, S.T., Eiglmeier, K., Parkhill, J., James, K.D., Thomson,

N.R., Wheeler, P.R., Honore, N., Garnier, T., Churcher, C.,

Harris, D., Mungall, K., Basham, D., Brown, D., Chillingworth,

T., Connor, R., Davies, R.M., Devlin, K., Duthoy, S., Feltwell, T.,

Fraser, A., Hamlin, N., Holroyd, S., Hornsby, T., Jagels, K.,

Lacroix, C., Maclean, J., Moule, S., Murphy, L., Oliver, K., Quail,

M.A., Rajandream, M.A., Rutherford, K.M., Rutter, S., Seeger,

K., Simon, S., Simmonds, M., Skelton, J., Squares, R., Squares,

S., Stevens, K., Taylor, K., Whitehead, S., Woodward, J.R &

Barrell, B.G (2001) Massive gene decay in the leprosy bacillus.

Nature (London) 409, 1007–1011.

18 Schweitzer, B.I., Dicker, A.P & Bertino, J.R (1990) Dihydrofolate

reductase as a therapeutic target FASEB J 4, 2441–2452.

19 Hitchings, G.H (1971) Folate antagonists as antibacterial and

antiprotozoal agents Ann NY Acad Sci 186, 444–451.

20 Wolf, W.A & Brown, G.M (1969) The biosynthesis of folic acid.

X Evidence for an Amadori rearrangement in the enzymatic

formation of dihydroneopterin triphosphate from GTP Biochim Biophys Acta 192, 468–478.

21 Shiota, T., Baugh, C.M & Myrick, J (1969) The assignment of structure to the formamidopyrimidine nucleoside triphosphate precursor of pteridines Biochim Biophys Acta 192, 205–210.

22 Bracher, A., Eisenreich, W., Schramek, N., Ritz, H., Go¨tze, E., Herrmann, A., Gu¨tlich, M & Bacher, A (1998) Biosynthesis of pteridines NMR studies on the reaction mechanisms of GTP cyclohydrolase I, pyruvoyltetrahydropterin synthase, and sepiapterin reductase J Biol Chem 273, 28132–28141.

23 Schramek, N., Bracher, A & Bacher, A (2001) Ring opening is not rate-limiting in the GTP cyclohydrolase I reaction J Biol Chem 276, 2622–2626.

24 Schramek, N., Bracher, A., Fischer, M., Auerbach, G., Nar, H., Huber, R & Bacher, A (2002) Reaction mechanisms of GTP cyclohydrolase I Single turnover experiments using a kinetically competent reaction intermediate J Mol Biol 316, 829–837.

25 Vallee, B.L & Auld, D.S (1989) Short and long spacer sequences and other structural features of zinc binding sites in zinc enzymes FEBS Lett 257, 138–140.

26 Christianson, D.W., David, P.R & Lipscomb, W.N (1987) Mechanism of carboxypeptidase A: hydration of a ketonic sub-strate analogue Proc Natl Acad Sci USA 84, 1512–1515.

27 Bacher, A., Eisenreich, W., Kis, K., Ladenstein, R., Richter, G., Scheuring, J & Weinkauf, S (1993) Biosynthesis of flavins In Bioorganic Chemistry Frontiers (Dugas, H & Schmidtchen, F.P., eds), pp 147–192 Springer, Berlin.

28 Bullock, W.O., Fernandez, J.M & Short, J.M (1987) XL1-Blue: a high efficiency plasmid transforming recA Escherichia coli with beta galactosidase selection Bio Techniques 5, 376–380.

29 Stu¨ber, D., Matile, H & Garotta, G ( 1990) System for high-level production in Escherichia coli and rapid purification of recombinant proteins: application to epitope mapping, prepara-tion of antibodies, and structure-funcprepara-tion analysis In Immuno-logical Methods (Lefkovits, I & Pernis, P., eds), pp 121–152 Academic Press, Orlando, FL.