Biosynthesis of riboflavin. Single turnover kinetic analysis of 6,7-dimethyl-8-ribityllumazine synthase

Article abstract of DOI:10.1021/ja028226k

6,7-Dimethyl-8-ribityllumazine synthase (lumazine synthase) catalyzes the condensation of 5-amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione with 3,4-dihydroxy-2-butanone 4-phosphate, affording the riboflavin precursor, 6,7-dimethyl-8-ribityllumazine. Single turnover experiments monitored by multiwavelength photometry were performed with the recombinant lumazine synthase of Bacillus subtilis. Mixing of the enzyme with the pyrimidine type substrate is conducive to a hypsochromic shift as well as a decrease in absorbance of the heterocyclic substrate; the rate constant for that reaction is 0.02 s^-1 μM^-1. Rapid mixing of the complex between enzyme and pyrimidine type substrate with the second substrate, 3,4-dihydroxy-2-butanone 4-phosphate, is followed by the appearance of an early optical transient characterized by an absorption maxima at 330 nm of low intensity which was tentatively assigned as a Schiff base intermediate. The subsequent elimination of phosphate affords a transient with intense absorption maxima at 455 and 282 nm, suggesting an intermediate with an extended system of conjugated double bonds. The subsequent formation of the enzyme product, 6,7-dimethyl-8-ribityllumazine, is the rate-determining step.

Full text of DOI:10.1021/ja028226k

Published on Web 03/20/2003
Biosynthesis of Riboflavin. Single Turnover Kinetic Analysis
of 6,7-Dimethyl-8-ribityllumazine Synthase*
Nicholas Schramek, Ilka Haase, Markus Fischer, and Adelbert Bacher*
Contribution from the Lehrstuhl f u¨ r Organische Chemie und Biochemie,
Technische UniVersit a¨ t M u¨ nchen, Lichtenbergstr. 4, D-85747 Garching, Germany
Received August 21, 2002; E-mail: adelbert.bacher@ch.tum.de
Abstract: 6,7-Dimethyl-8-ribityllumazine synthase (lumazine synthase) catalyzes the condensation of
5-amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione with 3,4-dihydroxy-2-butanone 4-phosphate, affording
the riboflavin precursor, 6,7-dimethyl-8-ribityllumazine. Single turnover experiments monitored by multi-
wavelength photometry were performed with the recombinant lumazine synthase of Bacillus subtilis. Mixing
of the enzyme with the pyrimidine type substrate is conducive to a hypsochromic shift as well as a decrease
in absorbance of the heterocyclic substrate; the rate constant for that reaction is 0.02 s µM-1. Rapid
mixing of the complex between enzyme and pyrimidine type substrate with the second substrate, 3,4-
dihydroxy-2-butanone 4-phosphate, is followed by the appearance of an early optical transient characterized
by an absorption maxima at 330 nm of low intensity which was tentatively assigned as a Schiff base
intermediate. The subsequent elimination of phosphate affords a transient with intense absorption maxima
at 455 and 282 nm, suggesting an intermediate with an extended system of conjugated double bonds. The
subsequent formation of the enzyme product, 6,7-dimethyl-8-ribityllumazine, is the rate-determining step.
-1
Introduction
Flavocoenzymes derived from vitamin B2 (riboflavin) are
essential for the metabolism of all organisms. The vitamin is
biosynthesized in plants and numerous microorganisms, whereas
animals are absolutely dependent on nutritional sources.
The biosynthetic pathway conducive to riboflavin has been
1-3
studied in some detail. Briefly, GTP is converted to 5-amino-
-ribitylamino-2,4-(1H,3H)-pyrimidinedione (compound 2, Fig-
6
ure 1) which is then condensed with the carbohydrate 3,4-
dihydroxy-2-butanone 4-phosphate (compound 1, Figure 1),
affording 6,7-dimethyl-8-ribityllumazine (compound 8) under
the catalytic action of 6,7-dimethyl-8-ribityllumazine synthase
4
(
lumazine synthase). The pteridine derivative undergoes an
unusual dismutation, affording riboflavin and 5-amino-6-ribit-
ylamino-2,4-(1H,3H)-pyrimidinedione (compound 2) which is
catalyzed by riboflavin synthase. The pyrimidine derivative 2
is reutilized as substrate by lumazine synthase. In summary, 1
equiv. of riboflavin is formed from 1 equiv. of the pyrimidine
derivative 2 and 2 equiv. of the carbohydrate 1.5
Figure 1. Hypothetical reaction mechanism for 6,7-dimethyl-8-ribityllu-
mazine synthase.4
bacteria and from plants are icosahedral capsids of 60 identical
subunitswhicharebestdescribedasdodecamersofpentamers.6
Structures of lumazine synthases from microorganisms and
plants have been studied by X-ray crystallography. The enzymes
from fungi and from the bacterium Brucella abortus are
,10-12
(
6) Persson, K.; Schneider, G.; Jordan, D. B.; Viitanen, P. V.; Sandalova, T.
Protein Sci. 1999, 8, 2355-2365.
homopentamers.6 Lumazine synthases from a variety of other
-9
(
7) Braden, B. C.; Velikovsky, C. A.; Cauerhff, A. A.; Polikarpov, I.;
Goldbaum, F. A. J. Mol. Biol. 2000, 297, 1031-1036.
*
Corresponding author.
(8) Meining, W.; M o¨ rtl, S.; Fischer, M.; Cushman, M.; Bacher, A.; Ladenstein,
R. J. Mol. Biol. 2000, 299, 181-197.
(
1) Bacher, A.; Eberhardt, S.; Eisenreich, W.; Fischer, M.; Herz, S.; Illarionov,
B.; Kis, K.; Richter, G. Vitam. Horm. 2001, 61, 1-49.
(9) Gerhardt, S.; Haase, I.; Steinbacher, S.; Kaiser, J. T.; Cushman, M.; Bacher,
A.; Huber, R.; Fischer, M. J. Mol. Biol. 2002, 318, 1317-1329.
(10) Ritsert, K.; Huber, R.; Turk, D.; Ladenstein, R.; Schmidt-B a¨ se, K.; Bacher,
A. J. Mol. Biol. 1995, 253, 151-167.
(
(
2) Young, D. W. Nat. Prod. Rep. 1986, 3, 395-419.
3) Bacher, A.; Eberhardt, S.; Richter, G. In Escherichia coli and Salmonella
typhimurium, 2nd ed.; Neidhardt, F. C., Ingraham, J. L., Low, K. B.,
Magasanik, B., Schaechter, M., Umbarger, H. E., Eds.; American Society
for Microbiology: Washington, D.C., 1996; Vol. 1, pp 657-664.
4) Kis, K.; Volk, R.; Bacher, A. Biochemistry 1995, 34, 2883-2892.
5) Plaut, G. W. J. Biol. Chem. 1960, 235, PC41-42.
(11) Zhang, X.; Meining, W.; Fischer, M.; Bacher, A.; Ladenstein, R. J. Mol.
Biol. 2001, 306, 1099-1114.
(
(
(12) Ladenstein, R.; Schneider, M.; Huber, R.; Bartunik, H. D.; Wilson, K.;
Schott, K.; Bacher, A. J. Mol. Biol. 1988, 203, 1045-1070.
4460
9
J. AM. CHEM. SOC. 2003, 125, 4460-4466
10.1021/ja028226k CCC: $25.00 © 2003 American Chemical Society

Analysis of 6,7-Dimethyl-8-ribityllumazine Synthase
A R T I C L E S
The peptide folds of pentameric and icosahedral lumazine
synthase are closely similar. The active sites of all lumazine
synthases are invariably located at subunit interfaces of the
pentamer assemblies. The lumazine synthases of Bacillaceae
Equilibrium Dialysis. Equilibrium dialysis experiments were per-
formed as published elsewhere.23
Single Turnover Experiments. Stopped flow experiments were
performed with an SFM4/QS apparatus from Bio-Logic (Claix, France)
equipped with a linear array of three mixers and four independent
syringes. The content of a 1.5 mm light path quartz cuvette behind the
last mixer was monitored with a Tidas diode array spectrophotometer
contain a riboflavin synthase trimer in the core of the icosahedral
capsid.1
3-15
This complex topology is conducive to kinetic
16
anomalies best described as substrate channeling.
(200-610 nm) equipped with a 15 W deuterium lamp as light source
Certain pathogenic microorganisms such as Enterobacteri-
aceae are unable to absorb riboflavin or its derivatives from
the environment and are therefore absolutely dependent on the
(J&M Analytische Mess- und Regeltechnik, Aalen, Germany). The
reaction buffer contained 100 mM potassium phosphate, pH 6.9, and
5 mM dithiotreithol. The enzyme and all substrates were diluted in
reaction buffer. Solutions were mixed at 25 °C with a total flow rate
endogenous synthesis of the vitamin.1
7,18
Inhibitors of the
-
1
of 4 mL s . Optical spectra integrated over 96 ms were recorded at
intervals of 100 ms in the wavelength range of 240-500 nm.
Data Analysis. Optical spectra were corrected for absorbance of
buffer and enzyme by subtraction of a blank data set obtained without
addition of substrate. Data reduction and stronger weighting of early
spectra were achieved by extracting 300 spectra on a pseudo-logarithmic
time base from the difference data sets. These data sets were then
analyzed using the program SPECFIT/32 3.0.30 (Spectrum Software
Associates, Marlborough, MA).
riboflavin pathway could therefore serve as antiinfective agents.
In light of the rapid resistance development, the exploration of
novel antiinfective targets appears urgent, and information on
the structure and mechanism of riboflavin biosynthetic enzymes
could serve as a basis for their development. To elucidate the
reaction mechanism of lumazine synthase, we have performed
presteady-state kinetic experiments reported in this paper.
Experimental Procedures
Determination of Reaction Order. Experiments were performedd
by the initial rate method using the SFM4/QS apparatus described
previously. The reaction buffer contained 100 mM potassium phosphate,
pH 6.9, and 5 mM dithiotreithol. The enzyme and all substrates were
diluted in reaction buffer. Solutions were mixed at 20 °C with a total
Materials. 3,4-Dihydroxy-2-butanone 4-phosphate was prepared
from ribose 5-phosphate using recombinant 3,4-dihydroxy-2-butanone
1
9
20
synthase. 5-Nitro-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione and
,7-dimethyl-8-ribityllumazine13 were synthesized according to pub-
lished procedures. 5-Amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedi-
6
-
1
flow rate of 4 mL s . Absorbance at 284 nm was recorded over a
period of 5 s at intervals of 10 ms. Enzyme and compound 2 were
used in the concentration range of 20-200 µM.
2
1
one was freshly prepared as described earlier.
Purification of Recombinant B. subtilis Lumazine Synthase. The
Analytical Ultracentrifugation. Sedimentation velocity experiments
were performed with an analytical ultracentrifuge (Optima XL-A,
Beckman Instruments) equipped with absorbance and interference
optics. Experiments were performed in double sector cells with
aluminum centerpieces and sapphire windows. Protein concentration
was monitored photometrically at 280 nm. Protein solutions were
dialyzed against 50 mM potassium phosphate, pH 7.0.
22
recombinant B. subtilis strain BR151[pBL1]-p602-BS-ribH was grown
aerobically as described in LB medium containing 15 mg of erythro-
mycin and 20 mg of kanamycin per liter at 32 °C. At an optical density
of 0.7 (600 nm), isopropylthiogalactoside was added to a concentration
of 2 mM. The suspension was incubated at 32 °C for 12 h, and the
cells were harvested by centrifugation. Wet cell mass from 1 L of
bacterial culture was suspended in 30 mL of 50 mM potassium
phosphate, pH 7.0, containing 0.5 mM sodium sulfite, 0.5 mM EDTA,
and 3 mg of lysozyme. The suspension was incubated for 20 min at 37
Results
A hypothetical reaction mechanism of lumazine synthase is
°
C and was subsequently ultrasonically treated. The cells were
1
6
centrifuged, and the supernatant was applied to a column of Q-
Sepharose FF (3 × 25 cm) that had been equilibrated with 20 mM
potassium phosphate, pH 7.0. The column was washed with 100 mL
of 20 mM potassium phosphate, pH 7.0, and was subsequently
developed with a gradient of 20-1000 mM potassium phosphate, pH
summarized in Figure 1. Based on the known regiochemistry
4,24
of the enzyme, it has been suggested that the carbonyl group
of 3,4-dihydroxy-2-butanone 4-phosphate (compound 1) reacts
with the position 5 amino group of the pyrimidine derivative 2
under formation of a Schiff base (compound 3) which is
assumed to eliminate inorganic phosphate under formation of
-
1
7
.0 (total volume, 800 mL), at a flow rate of 2 mL min . Fractions
were combined and concentrated by ultracentrifugation (32 000 rpm,
6 h, 4 °C). The solution was applied to a Sephacryl 400 column (2.6
60 cm) that had been equilibrated with 100 mM potassium phosphate,
pH 7.0. The column was developed with 500 mL of 100 mM potassium
4
,25
compound 4.
Tautomerization of compound 5 could afford
1
×
the conjugated iminoketone 6. Cyclization by addition of the
position 8 ribitylamino group of compound 6 to the carbonyl
group could yield a hydrated pteridin species, and dehydration
could terminate the reaction.
Certain reactants involved in that hypothetical reaction
sequence have characteristic visible and/or ultraviolet absorption.
The substrate, 2, has an absorption maximum at 284 nm at pH
-
1
phosphate, pH 7.0, at a flow rate of 3 mL min . Fractions were
combined and concentrated by ultracentrifugation. The enzyme was
stored at 4 °C in 100 mM potassium phosphate, pH 7.0.
(
13) Bacher, A. Methods Enzymol. 1986, 122, 192-199.
(
14) Bacher, A.; Schnepple, H.; Mail a¨ nder, B.; Otto, M. K.; Ben-Shaul, Y.
FlaVins FlaVoproteins, Proc. Int. Symp., 6th 1980, 579-586.
7
.0. Lumazine synthase binds compound 2 with an apparent
(
15) Ludwig, H. C.; Lottspeich, F.; Henschen, A.; Ladenstein, R.; Bacher, A.
KD of about 9 µM. Notably, in a mixture with lumazine synthase,
the absorbance of compound 2 is significantly lowered and
shifted to 278 nm (Figure 2A).
The enzyme product, 6,7-dimethyl-8-ribityllumazine, has a
highly characteristic spectrum with maxima at 408 and 255 nm
FlaVins FlaVoproteins, Proc. Int. Symp., 8th 1984, 379-382.
(
16) Kis, K.; Bacher, A. J. Biol. Chem. 1995, 270, 16788-16795.
17) Lingens, F.; Oltmanns, O.; Bacher, A. Z. Naturforsch., B: Chem. Sci. 1967,
(
2
2, 755-758.
(
18) Logvinenko, E. M.; Trach, V. M.; Koltun, L. V.; Shavlovskii, G. M. Tr.
S’ezda Mikrobiol. Ukr. 4th 1975, 8-9.
(
19) Richter, G.; Krieger, C.; Volk, R.; Kis, K.; Ritz, H.; G o¨ tze, E.; Bacher, A.
Methods Enzymol. 1997, 280, 374-382.
(
(
20) Cresswell, R. M.; Wood, H. C. S. J. Chem. Soc. 1960, 4768-4775.
21) Bacher, A.; Baur, R.; Eggers, U.; Harders, H. D.; Otto, M. K.; Schnepple,
H. J. Biol. Chem. 1980, 255, 632-637.
(23) Fischer, M.; Haase, I.; Feicht, R.; Richter, G.; Gerhardt, S.; Changeux, J.
P.; Huber, R.; Bacher, A. Eur. J. Biochem. 2002, 269, 519-526.
(24) Nielsen, P.; Neuberger, G.; Fujii, I.; Bown, D. H.; Keller, P. J.; Floss, H.
G.; Bacher, A. J. Biol. Chem. 1986, 261, 3661-3669.
(
22) Braun, N.; Tack, J.; Fischer, M.; Bacher, A.; Bachmann, L.; Weinkauf, S.
J. Cryst. Growth 2000, 212, 270-282.
(25) Volk, R.; Bacher, A. J. Am. Chem. Soc. 1988, 110, 3651-3653.
J. AM. CHEM. SOC.
9
VOL. 125, NO. 15, 2003 4461

A R T I C L E S
Schramek et al.
Figure 3. Binding of 5-amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione
(compound 2) to lumazine synthase. Optical spectra from a stopped flow
experiment at 25 °C in 100 mM potassium phosphate, pH 6.9. After rapid
mixing, the optical cuvette contained 93 µM 5-amino-6-ribitylamino-2,4-
(1H,3H)-pyrimidinedione (compound 2) and 119 µM enzyme subunits.
Figure 2. Ultraviolet spectra at pH 6.9. (A) (s) free 5-amino-6-
ribitylamino-2,4-(1H,3H)-pyrimidinedione (compound 2), (- - - -) enzyme
bound 5-amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione (compound 2);
(B) (s) free 6,7-dimethyl-8-ribityllumazine (compound 8), (- - - -) enzyme
bound 6,7-dimethyl-8-ribityllumazine (compound 8).
Figure 4. van’t Hoff plot of the initial rates (20 °C, 100 mM potassium
phosphate, pH 6.9) as a function of the concentration of compound 2.
Experimental points represent the average of four runs with the standard
deviation of the mean represented by the vertical bars. Linear regression
resulted in a slope of 1.93 ( 0.04 which represents the overall reaction
order.
and a shoulder at 276 nm at neutral pH (Figure 2B). The
compound has a pK of about 8.3,26 and deprotonation affords
an equilibrium mixture of various anionic forms arising by
intramolecular addition of aldityl side chain hydroxy groups to
ring carbon atom 7 under formation of various tricyclic
structures with absorbance maxima at 313, 279, and 227 nm;
an exomethylene type anion species present in low amounts in
the equilibrium mixture may absorb in the long wavelength
range around 366 nm.28
Lumazine synthase binds 6,7-dimethyl-8-ribityllumazine with
an apparent KD of about 160 µM, as shown by equilibrium
dialysis experiments. Optical spectra of mixtures containing
enzyme and 6,7-dimethyl-8-ribityllumazine at pH 7.0 showed
a weak band at about 320 nm, in addition to more intense bands
centered at 256, 274, and 408 nm (Figure 2B). 6,7-Dimethyl-
27,28
Lumazine synthase of B. subtilis was purified using anionic
exchange chromatography and gel filtration as described under
Experimental Procedures. The protein showed a single 16 kDa
band when analyzed by SDS polyacrylamide gel electrophoresis.
Analytical gel permeation chromatography of the native protein
afforded a single peak with the retention volume characteristic
of the 60-mer. Analytical ultracentrifugation afforded a single,
symmetrical boundary indicating an apparent sedimentation
8
-ribityllumazine without protein displays substantially lower
absorption around 320 nm. Binding to the enzyme may increase
the acidity; 6,7-dimethyl-8-ribityllumazine absorbance at 320
nm may indicate an anionic, enzyme bound species at neutral
pH.
Figure 3 shows optical spectra obtained after rapid mixing
of lumazine synthase with the pyrimidine derivative 2. After
rapid mixing, the sample solution contained 93 µM compound
2 and 119 µM lumazine synthase subunits. The dead time was
approximately 8 ms, and spectra were acquired at intervals of
100 ms using a diode array detector. We note a decrease of
absorbance over a period of about 1 s accompanied by an 8 nm
blue shift of the absorbance band of the pyrimidine derivative.
The final spectra in Figure 3 agree with the steady-state spectrum
of enzyme-bound compound 2 in Figure 2A. Numerical decon-
2
1,29
coefficient of 25 S in agreement with earlier studies.
(
26) Pfleiderer, W.; Bunting, J. W.; Perrin, D. D.; N u¨ bel, G. Chem. Ber. 1966,
9
9, 3503-3523.
(
27) Bown, D. H.; Keller, P. J.; Floss, H. G.; Sedlmaier, H.; Bacher, A. J. Org.
Chem. 1986, 51, 2461-2467.
(
28) Beach, R. L.; Plaut, G. W. E. J. Org. Chem. 1971, 36, 3937-3943.
29) Bacher, A.; Ludwig, H. C.; Schnepple, H.; Ben Shaul, Y. J. Mol. Biol.
(
1
986, 187, 75-86.
4462 J. AM. CHEM. SOC.
9
VOL. 125, NO. 15, 2003

Analysis of 6,7-Dimethyl-8-ribityllumazine Synthase
A R T I C L E S
volution of the spectra using the program SPECFIT/32 indicates
a second-order process characterized by a rate constant of 0.017
-
1
-1
s
µM .
The overall reaction order of the process was determined by
initial rate measurements performed under stopped flow condi-
tions. The reaction scheme in eq 1 affords eq 2 describing a
second-order reaction.
k
compound 2 + enzyme
d[compound 2]
9
8 enzyme-compound 2 (1)
a
b
-(
) V ) k[compound 2] [enzyme]
0
)t)0
dt
(2)
When the reactants are used at a stoichiometric ratio of 1:1
mol of substrate per mol of protein subunits), eq 2 can be
(
rewritten as
d[compound 2]
(a+b)
-(
) V ) k(a+b)[compound 2]
0
)t)0
dt
(3)
In logarithmic form, eq 3 can be rewritten as
ln(V ) ) (a + b) ln(compound 2) + ln(k(a+b))
(4)
0
0
Figure 5. Optical spectra from single turnover stopped flow experiments.
As shown in Figure 4, a van’t Hoff plot of the initial rates of
ligand binding afforded a straight line with a slope of 1.93. Thus,
ligand binding must be described as a second-order process.
Calculation of kon using the intercept of the y-axis (-11.1 (
(
(
A) Enzyme presaturated with compound 2 was mixed with compound 1.
B) Enzyme presaturated with compound 1 was mixed with compound 2.
The concentrations after rapid mixing were 97 µM 5-amino-6-ribitylamino-
,4-(1H,3H)-pyrimidinedione (compound 2), 383 µM 3,4-dihydroxy-2-
butanone 4-phosphate (compound 1), and 115 µM enzyme subunits.
2
Figure 6. Ultraviolet spectra from the early phase (A, C) and the later phase (B, D) of the single turnover experiments shown in Figure 5.
J. AM. CHEM. SOC. VOL. 125, NO. 15, 2003 4463
9

A R T I C L E S
Schramek et al.
maximum at about 330 nm (Figure 6A). The subsequent reaction
phase is dominated by absorbance in the long wavelength range
with maxima around 455 and 408 nm. The absorbance above
4
50 nm reaches a maximum value at 17 s and subsequently
decreases to low values over a period of about 2 min.
Absorbance at 408 nm, the λmax of the enzyme product 8
begins to increase from a very low value at approximately 1 s
after mixing and reaches a plateau after about 120 s. At the
end of the single turnover reaction, the optical spectrum of the
reaction mixture is similar to that of a steady state mixture of
enzyme and 6,7-dimethyl-8-ribityllumazine, as shown in Figure
2
B. The maxima at 406, 280, and 255 nm clearly represent the
neutral form of 6,7-dimethyl-8-ribityllumazine. The additional
maxima at 307 and 317 nm may represent a protein-bound anion
form of 6,7-dimethyl-8-ribityllumazine.
In a similar experiment, a solution containing compound 2
was rapidly mixed with a solution containing compound 1 and
protein (Figure 5B). The final concentrations in the optical
cuvette were identical with to those in Figure 5A (115 µM
enzyme subunits, 383 µM compound 1, and 97 µM compound
2
), although they were obtained by a different mixing protocol.
In excellent agreement with expectations based in the foregoing
experiments, Figures 5B, 6C and D, 7B, and 8C and D show a
rapid decrease at 284 nm in close similarity with the experiment
in Figure 2A (notably, that rapid decrease of absorbance is not
observed in Figure 5A where the reaction mixture was obtained
by a different mixing protocol). Later, Figure 7A shows a rapid
appearance of the low absorbance transient at 330 nm which is
observed earlier in Figure 6A. The spectra obtained in the late
reaction phase (later than 5 s) are closely similar in Figure 5A
and B; the optical spectra in Figure 6B and D become fairly
similar.
Numerical deconvolution of the data in Figure 5A indicated
two linearly independent processes (Figure 8A,B). A transient
species with absorbance at 455 and 282 nm reached a maximum
concentration at about 17 s and was subsequently consumed
under formation of 6,7-dimethyl-8-ribityllumazine. This reac-
tionsequence can be described by the following equation:
Figure 7. Absorbance changes observed during the single turnover stopped
flow experiments shown in Figure 6. Symbols represent experimental data
from Figure 6. Solid lines represent the numerical simulation using the
kinetic constants in Table 1. For other details, see legend to Figure 6.
.2; ꢀ for compound 2 at 284 nm ) 15 000 M-1 cm ; path
-1
0
-
1
-1
length, 1.5 mm) results in a value of 0.0067 µM
s
at 20
°
C. With an apparent KD of about 9 µM, koff can be calculated
-
1
as 0.06 s .
k2
k3
Stopped flow experiments involving both enzyme substrates
A
98 C
98 D
(5)
(i.e., compounds 1 and 2) are shown in Figures 5-8. In these
experiments, 3,4-dihydroxy-2-butanone 4-phosphate 1 was
proffered at a 3.3-fold excess over the concentration of enzyme
subunits in order to ensure saturation of the enzyme and to
minimize problems involved in the kinetic analysis of a two-
substrate reaction. This could be done without compromising
the optical measurements, since the carbohydrate phosphate is
transparent in the spectral range of the stopped flow experi-
ment.
The experiment in Figure 5A involved the rapid mixing of
compound 1 with a mixture containing compound 2 and
lumazine synthase subunits. After rapid mixing, the solution
contained 115 µM enzyme subunits, 383 µM compound 1, and
The numerical deconvolution failed to pick up the weak and
early optical transient characterized by an absorption of 330
nm. This failure is not surprising in light of (i) the weak
absorbance of the transient species and (ii) the rapid appearance
of absorbance contribution by the transient species B and of
6
,7-dimethyl-8-ribityllumazine in the enzyme bound anionic
form.
A similar deconvolution analysis of the experiment in Figure
B is shown in Figure 8C,D. The analysis supports three linearly
5
independent processes. This reaction sequence can be described
by the following equation:
9
7 µM compound 2. Optical spectra were then acquired at
k1[2]
k2
k3
intervals of 100 ms over a spectral range of 240-500 nm.
Selected spectra are replotted for clarity in Figure 6A,B.
Absorption at selected wavelengths as a function of reaction
time is shown in Figure 7A.
The initial reaction phase is characterized by the rapid
appearance of a very low intensity absorption band with a
A
8 B
9
8 C
9
8 D
(6)
The first optical transient virtually duplicates the second-order
process observed in Figure 3 (mixing of enzyme with compound
2). The apparent rate constant k1 is 0.017 s µM , in close
-
1
-1
similarity with the rate constant for the experiment in Figure 3.
4464 J. AM. CHEM. SOC.
9
VOL. 125, NO. 15, 2003

Analysis of 6,7-Dimethyl-8-ribityllumazine Synthase
A R T I C L E S
Figure 8. Numerical deconvolution of the stopped flow data shown in Figure 5. (A, C) Reconstructed absorbance spectra of transient chromophors. (B, D)
Concentration of transient species.
The numerical values for k2 and k3 are closely similar in the
experiments shown in Figure 5A and B.
synthase capsid with a core of riboflavin synthase (heavy
riboflavin synthase).
1
2,16
The early transient absorbing weakly around 330 nm can be
attributed tentatively to the Schiff base intermediate 3 in Figure
Discussion
1
. The π electron system of the hypothetical Schiff base
Binding of compound 2 to lumazine synthase proceeds as a
intermediate 3 is similar to that of dihydropterins such as
dihydroneopterin with absorbance maxima at 330 nm (ꢀ ) 6300
M
-
1
-1
second-order reaction with a rate around 0.02 s µM as
shown by initial rate measurements under stopped flow condi-
tions (Figure 4). Thus, kon is low by comparison with those of
most enzymes, but the same holds true for the overall catalytic
rate of lumazine synthase.
-
1
-1
31
cm ) and 280 nm.
The transient C with an absorption maximum at 455 nm
suggests an extended π electron system. The hypothetical
intermediates 4, 5, and 6 have extended π electron systems
which appear sufficient for the observed long wavelength
absorption, but an unequivocal structural assignment of the 455
nm transient is not yet possible.
The rate constant for binding of the substrate 2 is not
significantly affected by the presence or absence of the second
substrate. Hence, the loading of the enzyme with the substrates
1
and 2 lacks the characteristics of an ordered process.
Stopped flow experiments with lumazine synthase of Mag-
The low kon rate of lumazine synthase may be related to
32
naporthe grisea have been reported by Zheng et al. The authors
used single wavelength detection at 408 nm and interpreted the
absorbance at that wavelength as a measure for product
formation. This interpretation ignores the absorbance contribu-
tion of the optical transient C with an absorption maximum at
the unusual structure of the protein. The catalytic sites
are located close to the inner surface of the icosahedral
1
2,30
capsid.
The equilibrium X-ray structure fails to explain
how substrates and products penetrate the densely packed
capsids wall, and major dynamic motions of the capsid struc-
ture have been proposed as a possible mode for substrate and
product exchange with the bulk solvent. A low rate for the
penetration of substrate and products through the capsid wall
is well in line with the observation of intermediate channeling
in the enzyme complex comprising an icosahedral lumazine
4
55 nm which has a significant absorbance at 408 nm (ꢀ455/ꢀ408
0.5). Due to the failure to detect that highly absorbing species,
)
the kinetic model of Zheng et. al. is bound to become incorrect.
The rate-limiting step in our experiments is the formation of
,7-dimethyl-8-ribityllumazine 8 but cannot be unequivocally
6
(
31) Pfleiderer, W. In Folates and Pterins; Blakely, R. L., Benkovic, S. J., Eds.;
(
30) Ladenstein, R.; Meyer, B.; Huber, R.; Labischinski, H.; Bartels, K.; Bartunik,
John Wiley & Son: New York, 1985; Vol. 2, pp 43-114.
H. D.; Bachmann, L.; Ludwig, H. C.; Bacher, A. J. Mol. Biol. 1986, 187,
(32) Zheng, Y.-J.; Viitanen, P. V.; Jordan, D. B. Bioorg. Chem. 2000, 28, 89-
8
7-100.
97.
J. AM. CHEM. SOC.
9
VOL. 125, NO. 15, 2003 4465

A R T I C L E S
Schramek et al.
Table 1. Kinetic Constants Obtained from Single Turnover
Stopped Flow Experiments with Recombinant Lumazine Synthase
from B. subtilisa
NMR experiments. The observation of transfer NOE cross peaks
between the signals of the free and the enzyme bound ligand
requires that the dissociation of the enzyme product complex
occurs on the time scale of the NMR experiment. Although koff
for the bound product has not been measured, the dissociation
of the complex as observed qualitatively by NMR is definitely
too rapid to come even close to being the rate-limiting step of
experiment
1
2
3
k1 (s µM-1
-
1
)
0.020 ( 0.0007
0.017 ( 0.01
0.18 ( 0.003
0.024 ( 0.0002
-
-
1
1
k2 (s
k3 (s
)
)
0.17 ( 0.001
0.022 ( 0.0002
a
32
Experiment 1, enzyme mixed with compound 2 without the second
substrate compound 1; experiment 2, enzyme preincubated with compound
the overall enzyme reaction, as suggested by Zheng et al.
The detailed kinetic analysis of complex enzyme mechanisms
by time-resolved optical spectroscopy is only possible if at least
certain reaction components have characteristic spectra in the
visible and/or near-ultraviolet range. Certain enzymes involved
in the biosynthesis of bicyclic heterocycles such as GTP
cyclohydrolase I catalyzing the formation of dihydroneopterin
triphosphate and lumazine synthase are particularly favorable
for that type of analysis due to the sequence of “colorful”
reaction intermediates and their relatively low reaction rates.3
2
3
5
and mixed with compound 1; experiment 3, enzyme preincubated with
,4-dihydroxy-2-butanone 4-phosphate (compound 1) and mixed with
-amino-6-ribitylamino-2,4-(1H,3H)-pyrimidinedione (compound 2).
attributed to a specific reaction step in Figure 1, since the 455
nm transient may represent any of the hypothetical intermediates
4
-6. More specifically, the rate constant k3 (Table 1) may
comprise any of the following reactions: (i) the enolization of
, (ii) the rotation of the position 5 side chain from the extended
4,35
5
Acknowledgment. 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.
conformation observed in the crystallographic studies with the
intermediate analogue 5-(6-D-ribitylamino-2,4-dihydroxypyri-
midine-5-yl)-1-pentenyl phosphonic acid into a conformation
where the position 8 amino group can attack the carbonyl group
of the position 7 side chain, (iii) the ring closure reaction, and
JA028226K
(
iv) the elimination of water from the covalent hydrate 7.
Oschkinat et al. studied the conformation of 6,7-dimethyl-
-ribityllumazine bound to lumazine synthase by transfer NOE
(33) Oschkinat, H.; Schott, K.; Bacher, A. J. Biomol. NMR 1992, 2, 19-32.
(34) Schramek, N.; Bracher, A.; Fischer, M.; Auerbach, G.; Nar, H.; Huber,
R.; Bacher, A. J. Mol. Biol. 2001, 316, 827-835.
3
3
8
(35) Bracher, A.; Schramek, N.; Bacher, A. Biochemistry 2001, 40, 7896-7902.
4466 J. AM. CHEM. SOC.
9
VOL. 125, NO. 15, 2003