Design, Synthesis, and Evaluation of 9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones Bearing Alkyl Phosphate and α,α-Difluorophosphonate Substituents As Inhibitors of Riboflavin Synthase and Lumazine Synthase

Article abstract of DOI:10.1021/jo030278k

Lumazine synthase and riboflavin synthase catalyze the last two steps in the biosynthesis of riboflavin, an essential metabolite that is involved in electron transport processes. To obtain structural probes of these two enzymes, as well as inhibitors of p

Full text of DOI:10.1021/jo030278k

Design , Syn th esis, a n d Eva lu a tion of
9-D-Ribityla m in o-1,3,7,9-tetr a h yd r o-2,6,8-p u r in etr ion es Bea r in g
Alk yl P h osp h a te a n d r,r-Diflu or op h osp h on a te Su bstitu en ts a s
In h ibitor s of Ribofla vin Syn th a se a n d Lu m a zin e Syn th a se
Mark Cushman,*,† Thota Sambaiah,† Guangyi J in,† Boris Illarionov,‡ Markus Fischer,‡ and
Adelbert Bacher‡
Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and
Pharmacal Sciences, Purdue University, West Lafayette, Indiana 47907, and Lehrstuhl fu¨r Organische
Chemie und Biochemie, Technische Universita¨t Mu¨nchen, D-85747 Garching, Germany
cushman@pharmacy.purdue.edu
Received September 2, 2003
Lumazine synthase and riboflavin synthase catalyze the last two steps in the biosynthesis of
riboflavin, an essential metabolite that is involved in electron transport processes. To obtain
structural probes of these two enzymes, as well as inhibitors of potential value as antibiotics, a
series of ribitylpurinetriones bearing alkyl phosphate and R,R-difluorophosphonate substituents
were synthesized. Since the purinetrione ring system and the ribityl hydroxyl groups can be
alkylated, the synthesis required the generation of these two moieties in protected form before the
desired alkylation reaction could be carried out. These substances were designed as intermediate
analogue inhibitors of lumazine synthase that would bind to its phosphate-binding site. All of the
compounds were found to be effective inhibitors of both Bacillus subtilis lumazine synthase as
well as Escherichia coli riboflavin synthase. Molecular modeling of the binding of 3-(1,3,7,9-
tetrahydro-9-D-ribityl-2,6,8-trioxopurin-7-yl)propane 1-phosphate provided a structural explanation
for how these compounds are able to effectively inhibit both enzymes. Interestingly, the enzyme
kinetics of these new compounds in comparison with the parent purinetrione demonstrated
unexpectedly that the phosphate and phosphonate substituents contributed negatively to the
binding. A possible explanation for these effects on lumazine synthase would be that the inorganic
phosphate in the assay buffer competes with the substituted purinetriones for binding to the enzyme.
This would be consistent with the observed increase in Km of the 3,4-dihydroxybutanone-4-phosphate
substrate from 5.2 µM in Tris buffer or from 6.7 µM in MOPS buffer to 50 µM in phosphate buffer
when tested on Bacillus subtilis lumazine synthase. However, when tested in Tris buffer vs
Mycobacterium tuberculosis lumazine synthase, three of the phosphate inhibitors displayed
inhibition constants in the 4-5 nM range, indicating that they are much more potent than the
parent purinetrione. Under these conditions, the phosphate moieties of the inhibitors do contribute
positively to their binding. The R,R-difluorophosphonate analogue, which is expected to have
enhanced metabolic stability relative to the phosphates, was also found to be an inhibitor of
Mycobacterium tuberculosis lumazine synthase with a Ki of 60 nM.
In tr od u ction
Although the intricate details remain to be established,
the lumazine synthase-catalyzed reaction most likely
proceeds along the mechanistic pathway outlined in
Scheme 2.6 Condensation of the primary amino group of
the substituted pyrimidinedione 1 with the ketone 2 is
proposed to result in the formation of a Schiff base 5,
The riboflavin synthase-catalyzed dismutation of two
molecules of 6,7-dimethyl-8-D-ribityllumazine (3) (Scheme
1) results in the formation of one molecule of riboflavin
(4) and one molecule of the substituted pyrimidinedione
(1). In a wonderful example of efficient recycling cata-
lyzed by lumazine synthase, the product 1 of the ribo-
flavin synthase-catalyzed reaction is condensed with the
four-carbon phosphate 2 to form the riboflavin synthase
substrate 3.1-5
(2) Plaut, G. W. E. In Comprehensive Biochemistry; Florkin, M.,
Stotz, E. H., Eds.; Elsevier: Amsterdam, 1971; Vol. 21, pp 11-45.
(3) Beach, R. L.; Plaut, G. W. E. J . Am. Chem. Soc. 1970, 92, 2913-
2916.
(4) Bacher, A.; Eberhardt, S.; Richter, G. In Escherichia coli and
Salmonella: Cellular and Molecular Biology, 2nd ed.; Neidhardt, F.
C., Ed.; ASM Press: Washington, DC, 1996; pp 657-664.
(5) Bacher, A.; Fischer, M.; Kis, K.; Kugelbrey, K.; Mo¨rtl, S.;
Scheuring, J .; Weinkauf, S.; Eberhardt, S.; Schmidt-Ba¨se, K.; Huber,
R.; Ritsert, K.; Cushman, M.; Ladenstein, R. Biochem. Soc. Trans.
1996, 24, 89-94.
* To whom correspondence should be addressed. Tel: 765-494-1465.
Fax: 765-494-6790.
† Purdue University.
‡ Technische Universita¨t Mu¨nchen.
(1) Plaut, G. W. E.; Smith, C. M.; Alworth, W. L. Annu. Rev.
Biochem. 1974, 43, 899-922.
(6) Volk, R.; Bacher, A. J . Am. Chem. Soc. 1988, 110, 3651-3653.
10.1021/jo030278k CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/03/2004
J . Org. Chem. 2004, 69, 601-612
601

Cushman et al.
SCHEME 1
SCHEME 2
which eliminates phosphate to yield the enol 6. Tau-
tomerization of the enol 6 to the ketone 7, ring closure,
and dehydration of the covalent hydrate 8 would provide
the product 3.
X-ray crystallography of a complex of the phosphonate
analogue 97 with lumazine synthase has established that
the phosphate of the hypothetical Schiff base binds far
away from the ribityl side chain as depicted roughly in
structure 5.8 As shown in Figure 1, the observed confor-
mation of the phosphonate side chain of 9 implies that
the phosphate of 5 would be stabilized by hydrogen
bonding to the Arg136 residue and to the neighboring
Thr95 (not shown). However, the details of the isomer-
ization of the trans Schiff base to a cis Schiff base, or
possible initial formation of a thermodynamically less
stable cis Schiff base, have not been established. In
addition, the timing of the phosphate release relative to
the conformational change of the side chain to form 7 is
still an open question. For example, a case has recently
been made for the movement of the whole phosphate side
chain of 5 toward a cyclic conformation before phosphate
elimination occurs.9
inhibitor of Bacillus subtilis lumazine synthase and
displayed a Ki of 46 µM. In addition, it was a competitive
inhibitor of Escherichia coli riboflavin synthase with a
Ki of 0.61 µM. These Ki values indicated that the
purinetrione 10 was among the most potent inhibitors
of these two enzymes.10 The present research project was
undertaken to test the hypothesis that the attachment
of alkyl phosphate side chains to the N-7 nitrogen of the
purinetrione 10 would position the phosphate near the
phosphate-binding site of the enzyme and thus result in
an increase in potency for inhibition of lumazine syn-
thase. Since the exact linker chain length that would be
most favorable was an open question, a homologous series
of four phosphates were planned in which the linker
length was varied from three to six carbons. The resulting
inhibitors might also be of potential therapeutic value
as antibiotics because certain Gram-negative pathogenic
bacteria and yeasts lack an efficient riboflavin uptake
The purinetrione 10 was recently synthesized as part
of a program to obtain lumazine synthase inhibitors that
might have value as structural probes of the active site
of lumazine synthase.10 Compound 10 acted as a partial
(7) Cushman, M.; Mihalic, J . T.; Kis, K.; Bacher, A. J . Org. Chem.
1999, 64, 3838-3845.
(8) Meining, W.; Mo¨rtl, S.; Fischer, M.; Cushman, M.; Bacher, A.;
Ladenstein, R. J . Mol. Biol. 2000, 299, 181-197.
(9) Zhang, X.; Meining, W.; Cushman, M.; Haase, I.; Fischer, M.;
Bacher, A.; Ladenstein, R. J . Mol. Biol. 2003, 328, 167-182.
(10) Cushman, M.; Yang, D.; Kis, K.; Bacher, A. J . Org. Chem. 2001,
66, 8320-8327.
602 J . Org. Chem., Vol. 69, No. 3, 2004

9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
F IGURE 1. Crystal structure of phosphonate 9 bound in the active site of Saccharomyces cerevisiae riboflavin synthase. The
figure is programmed for walleyed viewing.
system and are therefore absolutely dependent on en-
dogenous riboflavin biosynthesis.11-14
and involves nucleophilic displacement of one of the
iodides of 1,3-diiodopropane with silver dibenzyl phos-
phate, followed by the displacement of the second iodide
with intermediate 15 as the nucleophile to afford 19. This
route was not tried out of necessity because the other
sequence failed. It was simply tried in order to see if the
reversed sequence of steps would work.
Selective deprotection of the ribityl side chains of 19-
22 with hydrogen fluoride-pyridine in THF afforded the
four tetraols 23-26.16-20 Interestingly, the potential
conversion of the secondary alcohols to alkyl fluorides,
which is reported to occur with hydrogen fluoride-
pyridine in n-hexane, was not observed in the present
case.21 Cleavage of the two methyl groups and the two
benzyl groups in the penultimate intermediates 23-26
with HCl in refluxing methanol afforded the four desired
target compounds 27-30.
Although the four phosphates 27-30 are interesting
as potential lumazine synthase and riboflavin synthase
inhibitors and as structural probes of the active sites of
these enzymes, their potential use as antibiotics could
be limited by the fact that phosphates are in general
readily hydrolyzed by phosphatases. One approach to
overcome this problem would be to replace the phosphate
group in these compounds with an analogue that mimics
the steric and electronic character of a phosphate, but is
metabolically more stable. An R,R-difluorophosphonate
seems like a good choice in the present case, since it
would more closely mimic the acidity of a phosphate than
a simple unfluorinated phosphonate would.22,23 To test
whether the enzyme inhibitory activity would be retained
as a result of this replacement, the R,R-difluorophospho-
nate analogue of the least active riboflavin synthase
inhibitor 28 was synthesized. The starting material 33
(Scheme 5) was prepared as described in the literature
by reaction of the anion derived from diethyl difluo-
romethanephosphonate with 1,4-diiodobutane.24 Reaction
of the protected purinetrione 15 with the iodide 33 in the
Resu lts a n d Discu ssion
Since the ribityl hydroxyl groups and the pyrim-
idinedione ring of 10 have the potential to be alkylated,
the instillation of alkyl phosphate side chains on N-7 of
the purinetrione system 10 requires that the ribityl
substituent and the pyrimidinedione ring be generated
in protected forms. Consequently, commercially available
6-chloro-2,4-dimethoxypyrimidine (11) was chosen as the
starting material (Scheme 3). Our plan was to carry the
dimethoxypyrimidine moiety through to a relatively late
stage in the synthesis and then unmask the pyrim-
idinedione system under acidic conditions. This strategy
would also allow the instillation and removal of tert-
butyldimethylsilyl protecting groups for the alcohols.
Nitration of the starting material was performed as
described in the literature with a mixture of fuming nitric
acid and sulfuric acid to afford the nitrated intermediate
12.15 A nucleophilic aromatic substitution reaction with
ribitylamine yielded the expected product 13. The four
ribityl hydroxyl groups were then protected as their tert-
butyldimethylsilyl ethers to provide compound 14. The
pyrimidine system was converted to the substituted
purine 15 through catalytic reduction of the nitro group
to the corresponding amine over palladium on carbon.
Reaction of the resulting primary amine with ethyl
chloroformate, followed by heating the intermediate ethyl
carbamate with sodium ethoxide in refluxing ethanol,
provided the protected purine system 15. Reaction of 15
with 1,4-diiodobutane, 1,5-diiodopentane, and 1,6-di-
iodohexane in DMF, with K2CO3 as the base, provided
the corresponding monoalkylated products 16, 17, and
18 in good yields. Nucleophilic displacement reactions of
the three iodides with silver dibenzyl phosphate resulted
in the phosphates 20, 21, and 22, having linker chain
lengths of four, five, and six carbons.
The dibenzyl phosphate 19, having a linker chain
length of three carbons, was obtained during the explora-
tion of a slightly different route outlined in Scheme 4.
This route is simply a reversal in the sequence of steps
(16) Kendall, P. M.; J ohnson, J . V.; Cook, C. E. J . Org. Chem. 1979,
44, 1421-1424.
(17) Newton, R. F.; Reynolds, D. P.; Webb, C. F.; Roberts, S. M. J .
Chem. Soc., Perkin Trans. 1 1981, 2055-2058.
(18) Ogawa, Y.; Nunomoto, M.; Shibasaki, M. J . Org. Chem. 1986,
51, 1625-1627.
(11) Wang, A. Yichuan Xuebao 1992, 19, 362-368.
(12) Oltmanns, O.; Lingens, F. Z. Naturforsch. 1967, 22, 751-754.
(13) Logvinenko, E. M.; Shavlovsky, G. M. Mikrobiologiya 1967, 41,
978-979.
(14) Neuberger, G.; Bacher, A. Biochem. Biophys. Res. Commun.
1985, 127, 175-181.
(15) Nutiu, R.; Boulton, A. J . J . Chem. Soc., Perkin Trans. 1 1976,
1327-1331.
(19) Snider, B. B.; Shi, Z. J . Am. Chem. Soc. 1994, 116, 549-557.
(20) Merino, P.; Franco, S.; Merchan, F. L.; Tejero, T. Tetrahedron
Lett. 1998, 39, 6411-6414.
(21) Olah, G. A.; Nojima, M.; Kerekes, I. Synthesis 1973, 786-787.
(22) Kabachnik, M. I.; Mastrukova, T. A.; Shipov, A. E.; Melentyeva,
T. A. Tetrahedron 1960, 9, 10-28.
(23) Smyth, M. S.; Ford, H., J r.; Burke, T. R., J r. Tetrahedron Lett.
1992, 33, 4137-4140.
J . Org. Chem, Vol. 69, No. 3, 2004 603

Cushman et al.
SCHEME 3a
SCHEME 4a
a
Reagents and conditions: (a) AgOPO(OBn)2, CH3CN, reflux
(10 h); (b) 15, K2CO3, DMF, 23 °C (24 h).
SCHEME 5a
a
Reagents and conditions: (a) K2CO3, DMF, 23 °C (24 h); (b)
HF-Py, THF, 0-23 °C (19 h); (c) TMSI, CH2Cl2, 23 °C (24 h).
presence of potassium carbonate in DMF afforded the
alkylation product 34. The TBDMS groups were removed
from 34 with hydrogen fluoride-pyridine in THF, result-
ing in intermediate 35. Finally, the methyl ethers and
ethyl esters were cleaved in the presence of trimethylsilyl
iodide to yield the desired R,R-difluorophosphonate ana-
logue 36. This transformation proceeded in higher overall
yield than the conversions of 23-26 to 27-30 with HCl-
MeOH. However, when TMSI was tried for the conver-
sions of 23-26 to 27-30, the undesired cleavage of the
phosphate groups to the corresponding alkyl iodides was
observed.
The phosphates 27-30, as well as the R,R-difluoro-
phosphonate analogue 36, were tested as inhibitors of
recombinant B. subtilis lumazine synthase â60 capsids
and recombinant riboflavin synthase from E. coli. The
inhibition constants and types of inhibition displayed by
these compounds are listed in Table 1, along with the
previously published results of the structurally related
phosphonates 37-40 and 9, the phosphate 41, and the
purinetrione 10. The new phosphates 27-30 and the R,R-
difluorophosphonate 36 are reaction intermediate ana-
logues that contain structural elements corresponding to
both of the substrates 1 and 2. Consequently, these
a
Reagents and conditions: (a) fuming HNO3, H2SO4, 80-85 °C
(2.5 h); (b) ribitylamine, DMF, 23 °C (48 h); (c) TBDMSCl,
imidazole, 23 °C (2 days); (d) (1) H2, Pd/C, MeOH, 23 °C (3 days),
(2) EtOCOCl, Et3N, CH2Cl2, 0 °C (6 h), (3) NaOEt, EtOH, reflux
(24 h); (e) diiodoalkane, K2CO3, DMF, 23 °C (24 h); (f) AgOPO-
(OBn)2, CH3CN, reflux (12 h); (g) HF-Py, THF, 0-23 °C (19 h);
(h) HCl, MeOH, reflux (3 h).
(24) Halazy, S.; Ehrhard, A.; Danzin, C. J . Am. Chem. Soc. 1991,
113, 315-317.
604 J . Org. Chem., Vol. 69, No. 3, 2004

9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
TABLE 1. In h ibition Con sta n ts vs Ba cillu s su btilis Lu m a zin e Syn th a se a n d Esch er ich ia coli Ribofla vin Syn th a se
lumazine synthasea Ki (µM)
compd parameter
variable 1 concnc
variable 2 concnd
riboflavin synthaseb Ki (µM)
e
27
Ks
3.82 ( 0.403
2.26 ( 0.113
39.6 ( 4.24
1.94 ( 0.270
16.66 ( 0.487
3.63 ( 0.431
f
kcat
3.62 ( 0.145
Kig
41.4 ( 18.1
h
Kis
98.3 ( 45.8
1.15 ( 0.265
211 ( 18.0
kcat
′
mechanism: partial inhibition
3.18 ( 0.397
mechanism: uncompetitive inhibition mechanism: competitive inhibition
28
29
30
36
Ks
25.7 ( 1.57
5.38 ( 0.111
2.09 ( 0.222
16.67 ( 0.357
332 ( 83.0
kcat
Ki
Kis
3.13 ( 0.101
168 ( 25.6
852 ( 103
mechanism: competitive inhibition mechanism: uncompetitive inhibition mechanism: competitive inhibition
Ks
3.82 ( 0.403
40.0 ( 3.92
1.96 ( 0.214
16.67 ( 0.387
2.44 ( 0.230
kcat
Ki
Kis
3.05 ( 0.094
3.65 ( 0.131
271 ( 84.6
852 ( 388
653 ( 163
817 ( 212
mechanism: mixed inhibition
3.00 ( 0.496
mechanism: mixed inhibition
34.7 ( 4.00
mechanism: competitive inhibition
2.26 ( 0.267
Ks
kcat
Ki
Kis
1.88 ( 0.081
3.85 ( 0.153
175 ( 25.3
17.3 ( 0.462
17.3 ( 1.72
78.3 ( 20.4
518 ( 178
mechanism: mixed inhibition
3.7 ( 0.78
mechanism: competitive inhibition
mechanism: competitive inhibition
2.13 ( 0.257
Ks
kcat
Ki
Kis
kcat
2.7 ( 0.092
12.53 ( 0.304
39.36 ( 9.91
132 ( 45
543 ( 169
′
1.1 ( 0.22
mechanism: partial inhibition
mechanism: competitive inhibition
37
9
Ki
Kis
440 ( 200
640 ( 300
>1000
mechanism: mixed inhibition
180 ( 88
Ki
Kis
>1000
>1000
350 ( 22
mechanism: mixed inhibition
130 ( 33
38
10
Ki
Kis
140 ( 15
mechanism: mixed inhibition
Ks
Ki
Kis
kcat
9.6 ( 0.9
0.61 ( 0.05
46 ( 5
250 ( 42
39.9 ( 0.7
mechanism: competitive inhibition
7.7 ( 2.3
mechanism: partial inhibition
39
40
41
Ks
Ki
Kis
> 860
160 ( 87
390 ( 63
mechanism: mixed inhibition
6.8 ( 2.0
Ks
Ki
Kis
>1000
170 ( 79
590 ( 90
mechanism: mixed inhibition
5.7 ( 0.9
Ks
Ki
Kis
33 ( 3.2
830 ( 170
2000 ( 360
190 ( 39
8200 ( 3200
mechanism: mixed inhibition
a
b
Recombinant â60 capsids from B. subtilis. Recombinant riboflavin synthase from E. coli. c The concentration of the dihydroxybutanone
phosphate substrate 2 was held constant at 100 µM during the assay, while the concentration of the pyrimidinedione substrate 1 was
d
varied. The concentration of the pyrimidinedione substrate 1 was held constant at 170 µM during the assay, while the concentration of
the dihydroxybutanone phosphate 2 was varied. e Ks is the substrate dissociation constant for the equilibrium E + S f P. kcat is the rate
f
constant for the process ES f E + P. Ki is the inhibitor dissociation constant for the process E + I f EI. ηKis is the inhibitor dissociation
g
constant for the process ES + I f ESI.
compounds can be classified as “bisubstrate analogue
inhibitors” of lumazine synthase that could potentially
achieve increased potency through an entropic advantage
resulting from the linking of structural moieties that
mimic each substrate. Since these compounds can po-
tentially inhibit the binding of both substrates to lu-
mazine synthase, it is possible that apparent differences
in the inhibition mechanism and the inhibition constants
could result from which substrate concentration is varied
and which is held constant during the course of the
enzyme inhibition experiments. To explore this question
in more detail, two sets of inhibition experiments were
conducted with lumazine synthase, one with variable
pyrimidinedione 1 concentration and one with variable
dihydroxybutanone phosphate 2 concentration. The re-
sults listed in Table 1 document an apparent change in
inhibitory mechanism from partial inhibition with 27, or
from competitive inhibition with 28, to uncompetitive
inhibition as the substrate with the variable substrate
concentration is switched from 1 to 2. Moreover, the
J . Org. Chem, Vol. 69, No. 3, 2004 605

Cushman et al.
change in the substrate whose concentration is varied
was found to have significant effects on the apparent Ki
values. For example, for the inhibition of lumazine
synthase by compound 29, the Ki value changed from 271
to 852 µM as the substrate with variable concentration
changed from 1 to 2.
recombinant E. coli riboflavin synthase. However, they
are all less potent inhibitors than the purinetrione 10
(Ki 0.61 µM). On the other hand, the previously reported
phosphonates 37, 9, and 38 were all inactive as riboflavin
synthase inhibitors, so the present phosphate and R,R-
difluorophosphonate replacements for the phosphonates
of these compounds, when combined with attachment to
a purinetrione system, do make a positive contribution
to the enzyme inhibitory activity.
The R,R-difluorophosphonate analogue 36 (Ki 132 µM)
is similar to the corresponding phosphate 28 (Ki 168 µM)
in potency as an inhibitor of lumazine synthase. The
phosphate 28 is an uncompetitive inhibitor of lumazine
synthase, while the R,R-difluorophosphonate 36 is a
partial inhibitor of lumazine synthase. In addition, it
should be noted that the R,R-difluorophosphonate 36 was
a significantly more potent competitive inhibitor of
riboflavin synthase (Ki 39 µM) than the corresponding
phosphate 28 (Ki 332 µM). The R,R-difluorophosphonate
modification to increase metabolic stability therefore
appears to offer the additional advantage of conferring
greater activity in some of the present testing systems.
When the potencies of the phosphates 27-30 are
compared using variable pyrimidinedione substrate 1
concentration, the most potent compound proved to be
27 (Ki 41 µM), in which the phosphate is separated from
the purinetrione ring system by a three-carbon linker.
That was followed by 30 (Ki 78 µM), 28 (Ki 168 µM), and
29 (Ki 271 µM). In comparison with the other phosphates,
the three-carbon linker of 27 makes it the closest
structural analogue of the reaction intermediate 5, and
this may explain why it is the most potent lumazine
synthase inhibitor.
With variable dihydroxybutanone phosphate substrate
2 concentration, the potencies of 29 (Ki 852 µM) and 30
(Ki 175 µΜ) can be compared with that of the parent
purinetrione 10 (Ki 46 µM), whose enzyme inhibition
studies were performed previously with variable sub-
strate 2 concentration.10 It is interesting to note that the
potency of the parent compound 10 (Ki 46 µM) is greater
than that of either of the phosphates 29 or 30, indicating
an unexpected opposite of the expected increase in
potency anticipated through the incorporation of the
phosphate side chains that could bind to the phosphate-
binding site of lumazine synthase. One possible explana-
tion for this effect would be that the enzyme kinetics
experiments were carried out, as usual, in phosphate
buffer, and the inorganic phosphate could conceivably
bind to the phosphate-binding site of the enzyme and
thus inhibit the binding of 29 and 30. To examine this
hypothesis in more detail, the Km values for the binding
of the dihydroxybutanone phosphate substrate were
determined in 50 mM Tris [tris(C-hydroxymethyl)ami-
nomethane] buffer, 50 mM MOPS [3-(N-morpholino)-
propanesulfonic acid] buffer, and 50 mM phosphate
buffer, all at pH 7. The Km values for the substrate 2 were
5.2 ( 0.5 µM in Tris buffer, 6.7 ( 0.4 µM in MOPS buffer,
and 50 ( 10 µM in phosphate buffer. Inorganic phosphate
therefore does appear to inhibit the binding of the
substrate 2, and most likely the inhibitors 29 and 30, to
lumazine synthase. This could possibly explain why the
purinetrione 10 appears to be a better inhibitor than 29
or 30 in phosphate buffer, since the purinetrione does
not bind to the phosphate-binding site of the enzyme.
Like the purinetrione 10, all of the new compounds 27
(Ki 3.63 µM), 28 (Ki 332 µM), 29 (Ki 2.44 µM), 30 (Ki 17.3
µM), and 36 (Ki 39.4 µM) were competitive inhibitors of
Finally, a comparison can be made between the present
compounds and the previously reported pyrimidinedione
inhibitors 39, 40, and 41 having an amide chain con-
nected to phosphonate or phosphate moieties. In contrast
to the present series, these amides were inactive or
displayed very weak activity vs lumazine synthase, but
were riboflavin synthase inhibitors in the Ki 160-190 µM
range.25 All of these compounds are less potent than 27,
29, 30, and 36 vs both lumazine synthase and riboflavin
synthase, indicating that the extension of the phosphate
or phosphonate side chains from a purinetrione ring
system is generally more favorable than connecting them
to a pyrimidinedione system through an amide linkage.
The one exception is the low activity (Ki 332 µM)
displayed by 28 vs riboflavin synthase.
Crystal structures are available of the complexes
formed between the substrate analogue 42 and the
lumazine synthases of B. subtilis and Schizosaccharo-
myces pombe.26,27 X-ray structures have also been deter-
mined of complexes of the intermediate analogue 9 bound
(25) Cushman, M.; Yang, D.; Mihalic, J . T.; Chen, J .; Gerhardt, S.;
Huber, R.; Fischer, M.; Kis, K.; Bacher, A. J . Org. Chem. 2002, 67,
6871-6877.
(26) Ritsert, K.; Huber, R.; Turk, D.; Ladenstein, R.; Schmidt-Ba¨se,
K.; Bacher, A. J . Mol. Biol. 1995, 253, 151-167.
(27) Gerhardt, S.; Haase, I.; Steinbacher, S.; Kaiser, J . T.; Cushman,
M.; Bacher, A.; Huber, R.; Fischer, M. J . Mol. Biol. 2002, 318, 1317-
1329.
606 J . Org. Chem., Vol. 69, No. 3, 2004

9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
F IGURE 2. Hypothetical model for the binding of compound 27 to Bacillus subtilis lumazine synthase. The figure is programmed
for walleyed viewing.
to Saccharomyces cerevisiae lumazine synthase (Figure
1) and the product analogue 43 bound to S. pombe
lumazine synthase.8,27 These structures allow the rational
docking and energy minimization of additional lumazine
synthase inhibitors. In the present case, a hypothetical
model was constructed of the binding of the lumazine
synthase/riboflavin synthase inhibitor 27 to B. subtilis
lumazine synthase. This model was produced by overlap-
ping the structure of 27 with that of 42 in a 15 Å radius
spherical fragment surrounding the ligand in one of the
sixty equivalent active sites of B. subtilis lumazine
synthase.26 The structure of 42 was then removed and
the energy of the complex minimized using the MMFF94
force field while allowing the ligand and the protein
structure contained within a 6 Å sphere surrounding the
ligand to remain flexible with the remainder of the
F IGURE 3. Hydrogen bonds and distances in the calculated
protein structure frozen. The resulting Figure 2 was
model of the inhibitor 27 bound in the active site of Bacillus
subtilis lumazine synthase.
constructed by displaying the amino acid residues cal-
culated to be involved in bonding of the protein with
ligand 27, using a maximum distance of 3.5 Å between
the donor and acceptor atoms to be considered a hydrogen
bond. As expected, the calculated structure shows the
purinetrione ring of the ligand 27 stacked with the phenyl
ring of Phe22. The phosphate of the ligand is extensively
hydrogen bonded with the side chain nitrogens of Arg127,
as well as the backbone nitrogens of Ala85 and Thr86
and the side-chain hydroxyl of Thr86. The ribityl hy-
droxyl groups are hydrogen bonded to a water molecule,
the side chain amino group of Lys135, the backbone
carbonyl of Phe113, and the side chain carboxylate of
between monomeric S. pombe lumazine synthase and
Glu58. The purinetrione ring of the ligand is hydrogen
bonded to the side-chain amino group of Lys135, the
backbone nitrogen of Ala56, the backbone carbonyl of
Thr80, and the side-chain nitrogen of Asn23. These
contacts are summarized in Figure 3. It should be noted
that these contacts are either similar or identical to those
seen in the crystal structure of 42 bound to B. subtilis
lumazine synthase.26
6-carboxyethyl-7-oxo-8-D-ribityllumazine (43).28 Accord-
ing to these studies, as well as NMR investigations of
the structure of the N-terminal domain of riboflavin
synthase,29 the active site of the enzyme is derived from
the intermolecular juxtaposition of residues belonging to
both of the barrels in the C-terminal and N-terminal
domains of the two monomer units, with one substrate
An X-ray structure of E. coli riboflavin synthase, which
contains three identical subunits, has been published
along with a hypothetical model of the bound substrate
6,7-dimethyl-8-D-ribityllumazine (3). In addition, a crys-
tal structure has been determined of a complex formed
(28) Gerhardt, S.; Schott, A.-K.; Kairies, N.; Cushman, M.; Illarionov,
B.; Eisenreich, W.; Bacher, A.; Huber, R.; Steinbacher, S.; Fischer, M.
Structure 2002, 10, 1371-1381.
(29) Truffault, V.; Coles, M.; Diericks, T.; Abelmann, K.; Eberhardt,
S.; Lu¨ttgen, H.; Bacher, A.; Kessler, H. J . Mol. Biol. 2001, 309, 949-
960.
J . Org. Chem, Vol. 69, No. 3, 2004 607

Cushman et al.
F IGURE 4. Hypothetical model of the binding of two molecules of inhibitor 27 to Escherichia coli riboflavin synthase. The C-barrel
is orange and the N-barrel is green. Hydrogen atoms have been omitted for clarity. The model is programmed for walleyed viewing.
SCHEME 6a
molecule bound to each barrel. These structures allow
molecular modeling of the complex formed between 27
and E. coli riboflavin synthase to proceed in a rational
way. The structure displayed in Figure 4 was arrived at
after docking two molecules of the ligand 27 in the active
site and energy minimization using a frozen protein
structure and two mobile ligand molecules. The N-barrel
residues in Figure 4 are colored orange, while the
C-barrel residues are green. As predicted from the known
regiochemistry of the riboflavin synthase-catalyzed reac-
tion and the available crystal structures, two molecules
of the ligand are stacked in the active site with their
ribityl chains pointing in opposite directions.3 The two
phosphate side chains are therefore also pointing in
opposite directions, and the model suggests the possibility
of hydrogen bonding of each phosphate with some of the
ribityl hydroxyl groups of the neighboring ligand mol-
ecule.
Human infections with drug-resistant Mycobacterium
tuberculosis have been an ever-increasing worldwide
health problem. New antibiotics that are effective against
these resistant strains are therefore needed urgently.
Consequently, the phosphates prepared in the present
a
Reagents and conditions: (a) 1,3-diiodopropane, K2CO3, DMF,
study were also tested vs M. tuberculosis lumazine
synthase. Also, for the sake of comparison, the parent
purinetrione 10 was examined, along with the three-
carbon alcohol 46 corresponding to the most potent
phosphate 27. The synthesis of 46 is described in Scheme
6. Alkylation of the starting material 15 with 1,3-
diiodopropane in DMF in the presence of potassium
carbonate as the base afforded the iodide 44, which was
hydrolyzed to the alcohol 45 with sodium hydroxide in
refluxing methanol. Removal of the tert-butyldimethyl-
silyl protecting groups and cleavage of the two methyl
ethers present in 45 with methanolic hydrochloric acid
yielded the desired product 46.
As a first step to the determination of the activities of
the phosphates 27-30 vs M. tuberculosis lumazine
synthase, an attempt was made to determine the Km of
the phosphate substrate 2. This proved to be impossible
in 100 mM phosphate buffer, because even at a substrate
concentration of 1000 µM, the initial reaction velocity still
23 °C (24 h); (b) NaOH, MeOH, reflux (12 h); (c) HCl, MeOH, reflux
(3 h).
increased linearly with concentration, indicating that the
apparent Km for the substrate in 100 mM phosphate
buffer is much greater than 1000 µM. In contrast, in
MOPS buffer (50 mM MOPS pH 7.0, 100 mM NaCl, 5
mM EDTA, 5 mM DTT), maximal reaction velocity was
reached by 400 µM phosphate substrate 2 concentration,
and in Tris buffer (50 mM Tris-HCl pH 7.0, 100 mM
NaCl, 5 mM EDTA, 5 mM DTT), maximal reaction
velocity was reached by 250 µM phosphate substrate 2
concentration. Evidently, the M. tuberculosis lumazine
synthase reaction rate is much more sensitive to inor-
ganic phosphate than the B. subtilis enzyme is. For this
reason, the subsequent studies on enzyme inhibition by
the phosphates 27-30 were performed in Tris buffer,
using a variable pyrimidinedione 1 concentration and a
constant phosphate substrate 2 concentration. As shown
608 J . Org. Chem., Vol. 69, No. 3, 2004

9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
TABLE 2. In h ibition Con sta n ts vs Mycoba cter iu m
tu ber cu losis Lu m a zin e Syn th a sea
filtered, washed three times with ice-water, and dried under
vacuum to give pure 12 as a yellow solid (4.00 g, 64%): mp
1
66-67 °C (lit.15 mp 67-69 °C); H NMR (300 MHz, CDCl3) δ
compd
Ki
mechanism
4.07 (s, 3 H), 4.11 (s, 3 H).
10
46
27
28
29
30
36
9.1 ( 0.61 µM
2.6 ( 0.31 µM
4.5 ( 3.3 nM
4.1 ( 2.3 nM
4.7 ( 1.9 nM
170 ( 1.8 nM
60 ( 6.8 nM
competitive
mixed
2,4-Dim eth oxy-5-n itr o-6-D-r ibitylam in opyr im idin e (13).
6-Chloro-2,4-dimethoxy-5-nitropyrimidine (12) (5.84 g, 26.85
mmol) was added to a solution of D-ribitylamine (8.83 g, 58.55
mmol) in dry DMF (200 mL), and the mixture was stirred at
room temperature for 48 h under nitrogen atmosphere. The
solvent was removed using reduced pressure, and the resultant
residue was purified using flash chromatography (30 g, 2 ×
30 cm column of SiO2), eluting with chloroform-methanol (10:
1), to afford the desired product 13 (7.46 g, 75%) as a light
yellow solid: mp 70-71 °C; IR (KBr) 3346, 1598 cm-1; 1H NMR
(300 MHz, CD3OD) δ 4.02 (s, 3 H), 3.99 (s, 3 H), 4.01-3.92
(m, 2 H), 3.80-3.57 (m, 5 H); 13C NMR (75 MHz, CD3OD) δ
168.4, 164.9, 160.3, 113.3, 79.4, 74.5, 74.2, 71.7, 64.5, 55.9, 55.8,
45.2; EIMS (MH+) m/z 335. Anal. Calcd for C11H18N4O8: C,
39.52; H, 5.43; N, 16.76. Found: C, 39.75; H, 5.32; N, 17.02.
competitive
competitive
competitive
competitive
competitive
a
The experiments were conducted at
a
fixed substrate 2
concentration of 150 µM and a variable substrate 1 concentration
of 1-400 µM.
in Table 2, the resulting Ki values of the phosphates 27-
29 were all between 4 and 5 nM! The phosphate 30,
having the longest phosphate side chain, was less active,
displaying a Ki of 0.17 µM. In contrast, the parent
purinetrione 10 displayed a Ki of 9.1 µM, the three-carbon
alcohol had a Ki of 2.6 µM. It is obvious that the
phosphate moieties of the inhibitors 27-30 play an
important role in increasing the affinity of these enzyme
inhibitors for the M. tuberculosis lumazine synthase. It
is also gratifying to see inhibition constants for the
phosphates 27-29 in the low nanomolar range vs M.
tuberculosis lumazine synthase. The crystal structure of
M. tuberculosis lumazine synthase has not yet been
determined, so the structural basis for the differences
seen with this enzyme vs the B. subtilis enzyme remain
to be elucidated.
As mentioned above, a potential limitation of the
phosphates 27-30 as therapeutically useful antibiotics
results from their expected metabolic instability toward
phosphatases. Accordingly, the R,R-difluorophosphonate
36 was also tested as an inhibitor of M. tuberculosis
lumazine synthase. As shown in Table 2, compound 36
displayed a Ki of 60 nM, which is clearly in the range of
interest for possible therapeutic use.
In conclusion, the results presented here demonstrate
that certain purinetriones bearing phosphate side chains
can inhibit both lumazine synthase as well as riboflavin
synthase, and molecular modeling with compound 27
suggests possible binding modes to each enzyme. Anti-
biotics that would inhibit both lumazine synthase and
riboflavin synthase would be less likely to suffer from
the development of antibiotic resistance by the organisms
that they are supposed to treat, since pathogenic micro-
organisms would have to simultaneously select for muta-
tions in both enzymes in order to escape the cytotoxic
effects of the antibiotics.
6-(2′,3′,4′,5′-Tet r a -O-ter t-b u t yld im et h ylsilyl-D-r ib it yl-
a m in o)-2,4-d im eth oxy-5-n itr op yr im id in e (14). Compound
13 (11.5 g, 3.44 mmol), tert-butyldimethylsilyl chloride (36.0
g, 24.1 mmol), and imidazole (16.3 g, 24.1 mmol) were dissolved
in dry DMF (250 mL), and the mixture was stirred at room
temperature for 2 days. DMF was then removed under reduced
pressure, and the residue was stirred in ethyl acetate (200 mL)
for 2 h. The precipitated salt was filtered, and the filtrate was
concentrated to form a yellow oil, which was separated by
using flash chromatography (SiO2), eluting with hexanes-
EtOAc (95:5), to afford pure 14 (20.3 g, 75%) as a viscous oil:
IR (CDCl3, film) 3356, 2954, 2930, 1594, 1538, 1472, 1361,
1
1257, 1093 cm-1; H NMR (300 MHz, CDCl3) δ 9.00 (s, 1 H),
4.12 (m, 1 H), 4.09 (s, 3 H), 3.95 (s, 3 H), 3.88-3.93 (m, 2 H),
3.82 (dd, J ) 2.66 Hz, J ) 5.9 Hz, 1 H), 3.69 (dd, J ) 5.2 Hz,
J ) 10.7 Hz, 1 H), 3.50-3.57 (m, 2 H), 0.87 (s, 9 H), 0.86 (s, 9
H), 0.84 (s, 9 H), 0.83 (s, 9 H), 0.10 (s, 3 H), 0.08 (s, 3 H), 0.07
(s. 6 H), 0.03 (s, 6 H), -0.02 (s, 3 H), -0.04 (s, 3 H); EIMS m/z
791. Anal. Calcd for C35H74Si4N4O8: C, 53.12; H, 9.43; N, 7.08.
Found: C, 53.22; H, 9.58; N, 6.99.
9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsilyl-D-r ibityl)-
7,9-d ih yd r o-2,6-d im eth oxyp u r in -8-on e (15). Compound 14
(3.80 g, 4.80 mmol) was dissolved in methanol (100 mL), and
palladium on charcoal (10%, 0.5 g) was added. The reaction
mixture was hydrogenated at room temperature and 1.4 psi
for 3 days. The catalyst was filtered off, and the filtrate was
concentrated under reduced pressure to remove the methanol
completely. The residue was dissolved in dichloromethane (40
mL), and triethylamine (4.00 mL, 28.8 mmol) was added. The
mixture was cooled to 0 °C, and a solution of ethyl chlorofor-
mate (0.51 mL, 5.28 mmol) in CH2Cl2 (10 mL) was added
dropwise. The reaction mixture was stirred at room temper-
ature for 6 h, washed with water (2 × 50 mL), dried with
anhydrous Na2SO4, and concentrated under reduced pressure,
and the residue was dried well under vacuum. The residue
and NaOEt (820 mg, 12.0 mmol) in absolute EtOH (100 mL)
were heated under reflux for 24 h. The solvent was removed
under reduced pressure, and compound 15 (1.76 g, 47.0%) was
isolated as a solid by flash chromatography (SiO2, eluting with
hexanes-EtOAc 4:1): mp 57-59 °C; IR (CDCl3, film) 3149,
Exp er im en ta l Section
2953, 2928, 2851, 1709, 1635, 1614, 1462, 1375, 1253 cm-1
;
Gen er a l Meth od s. Melting points were determined in
capillary tubes and are uncorrected. Proton nuclear magnetic
resonance spectra (1H NMR) were determined at 300 or 500
MHz as noted. Phosphorus nuclear magnetic resonance (31P
NMR) chemical shifts are reported relative to 85% H3PO4 as
external standard. Silica gel used for column chromatography
was 230-400 mesh.
6-Ch lor o-2,4-d im eth oxy-5-n itr op yr im id in e (12). Sulfu-
ric acid (12.5 mL) was added dropwise to fuming nitric acid
(7.5 mL) at 0 °C, and 6-chloro-2,4-dimethoxypyrimidine (11)
(5.00 g, 28.7 mmol) was added. The reaction mixture was
stirred at 80-85 °C for 2.5 h, cooled to room temperature,
poured into ice-water, and stirred for 20 min. The solid was
1H NMR (500 MHz, CDCl3) δ 9.10 (s, 1 H), 4.48 (d, J ) 9.76
Hz, 1 H), 4.30 (dd, J ) 10.1, 13.94 Hz, 1 H), 4.06 (dd, J )
1.62, 13.97 Hz, 1 H), 4.01 (s, 4 H), 3.94 (s, 3 H), 3.78 (t, J )
7.14 Hz, 1 H), 3.71 (d, J ) 7.15, 10.36 Hz, 1 H), 3.58 (dd, J )
5.60, 10.45 Hz, 1 H), 0.95 (s, 9 H), 0.88 (s, 9 H), 0.87 (s, 9 H),
0.65 (s, 9 H), 0.23 (s, 3 H), 0.13 (s, 3 H), 0.11 (s, 3 H), 0.10 (s,
3 H), 0.049 (s, 3 H), 0.045 (s, 3 H), -0.049 (s, 3 H), -0.47 (s,
3 H); EIMS (MH+) m/z 787. Anal. Calcd for C36H74N4O7Si4: C,
54.96; H, 9.41; N, 7.12. Found: C, 55.10; H, 9.36; N, 7.06.
Gen er a l P r oced u r e for th e Syn th esis of th e Iod id es
16-18. Compound 15 (394 mg, 0.50 mmol) was dissolved in
dry DMF (15 mL). Anhydrous K2CO3 (138 mg, 1.00 mmol) and
J . Org. Chem, Vol. 69, No. 3, 2004 609

Cushman et al.
diiodoalkane (2.25 mmol) were added, and the mixture was
stirred under argon at room temperature for 24 h. The DMF
was removed under reduced pressure and the residue purified
by flash chromatography (SiO2), eluting with hexanes-EtOAc
(95:5), to afford the pure iodides 16-18 in 66-94% yields as
colorless oils.
4-[9-(2′,3′,4′,5′-Tet r a -O-ter t-b u t yld im et h ylsilyl-D-r ib i-
t yl)-7,9-d ih yd r o-2,6-d im et h oxy-8-oxop u r in -7-yl]-1-iod o-
bu ta n e (16). This intermediate was obtained by the procedure
above in 80% yield as a colorless oil: 1H NMR (500 MHz,
CDCl3) δ 4.48 (d, J ) 9.67 Hz, 1 H), 4.30 (dd, J ) 10.08, 13.90
Hz, 1 H), 4.06 (d, J ) 1.78 Hz, 2 H), 4.03 (s, 3 H), 3.93 (s, 3 H),
3.90 (m, 2 H), 3.77 (m, 1 H), 3.67 (dd, J ) 7.60, 10.35 Hz, 1
H), 3.57 (dd, J ) 5.60, 10.45 Hz, 1 H), 3.19 (t, J ) 6.41 Hz, 2
H), 1.55 (m, 4 H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.65 (s, 9 H),
0.20 (s, 3 H), 0.11 (s, 6 H), 0.087 (s, 3 H), 0.042 (s, 3 H), 0.036
(s, 3 H), -0.058 (s, 3 H), -0.48 (s, 3 H); EIMS (MH+) m/z 969.
Anal. Calcd for C40H81IN4O7Si4: C, 49.56; H, 8.42; N, 5.78.
Found: C, 49.33; H, 7.93; N, 5.24.
trated under reduced pressure. The resultant oils were purified
by flash chromatography (28 g, 5 × 40 cm column of SiO2),
eluting with hexanes-ethyl acetate (7:3), to furnish the
products as colorless oils.
Diben zyl 4-[9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsi-
lyl-D-r ibityl)-7,9-d ih yd r o-2,6-d im eth oxy-8-oxop u r in -5-yl]-
bu ta n e 1-P h osp h a te (20). Compound 20 was obtained by the
procedure above in 83% yield as a colorless oil: 1H NMR (300
MHz, CDCl3) δ 7.31 (s, 10 H), 5.01 (m, 4 H), 4.47 (d, J ) 9.9
Hz, 1 H), 4.30 (dd, J ) 10.08, 13.90 Hz, 1 H), 4.06 (d, J ) 1.78
Hz, 1 H), 4.02 (d, J ) 2.1 Hz, 2 H), 3.99 (s, 1 H), 3.96 (s, 3 H),
3.96 (s, 3 H), 3.84 (m, 2 H), 3.77 (m, 1 H), 3.66 (dd, J ) 5.60,
10.45 Hz, 1 H), 3.56 (dd, J ) 5.8, 10.35 Hz, 1 H), 1.72 (m, 4
H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.63 (s, 9 H), 0.20 (s, 3 H),
0.11 (s, 6 H), 0.09 (s, 3 H), 0.043 (s, 3 H), 0.041 (s, 3 H), -0.071
(s, 3 H), -0.52 (s, 3 H); EIMS (MH+) m/z 1119. Anal. Calcd
for C54H94N4O11PSi4: C, 57.93; H, 8.55; N, 5.00. Found: C,
58.00; H, 8.65; N, 5.08.
Diben zyl 5-[9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsi-
lyl-D-r ibityl)-7,9-d ih yd r o-2,6-d im eth oxy-8-oxop u r in -7-yl]-
p en ta n e 1-P h osp h a te (21). Intermediate 21 was obtained
by the procedure above in 64% yield as a colorless oil: 1H NMR
(300 MHz, CDCl3) δ 7.32 (m, 10 H), 5.01 (m, 4 H), 4.48 (d, J )
9.63 Hz, 1 H), 4.47 (dd, J ) 10.14, 13.92 Hz, 1 H), 4.07 (d,
J ) 2.06 Hz, 1 H), 4.029 (br s, 1 H), 3.98 (s, 3 H), 3.96 (m,
1 H), 3.94 (m, 1 H), 3.93 (s, 3 H), 3.81 (m, 3 H), 3.67 (dd, J )
5.60, 10.45 Hz, 1 H), 3.56 (dd, J ) 5.8, 10.35 Hz, 1 H), 1.62
(m, 4 H), 1.33 (m, 2 H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.64 (s,
9 H), 0.20 (s, 3 H), 0.107 (s, 6 H), 0.085 (s, 3 H), 0.042 (s, 3 H),
0.036 (s, 3 H), -0.064 (s, 3 H), -0.507 (s, 3 H); EIMS (M+) m/z
1132. Anal. Calcd for C55H97PN4O11Si4: C, 58.30; H, 8.57; N,
4.95. Found: C, 58.28; H, 8.40; N, 4.89.
5-[9-(2′,3′,4′,5′-Tetr a-O-ter t-bu tyldim eth ylsilyl-D-r ibityl)-
7,9-d ih yd r o-2,6-d im et h oxy-8-oxop u r in -7-yl]-1-iod op en -
ta n e (17). This compound was obtained by the procedure
above in 86% yield as a colorless oil: 1H NMR (300 MHz,
CDCl3) δ 4.48 (d, J ) 9.75 Hz, 1 H), 4.30 (dd, J ) 10.14, 13.92
Hz, 1 H), 4.16 (t, J ) 6.53 Hz, 2 H), 4.029 (s, 3 H), 3.93 (s, 3
H), 3.88 (m, 2 H), 3.76 (m, 1 H), 3.67 (dd, J ) 7.60, 10.35 Hz,
1 H), 3.56 (dd, J ) 5.60, 10.45 Hz, 1 H), 3.16 (t, J ) 6.53 Hz,
2 H), 1.84 (m, 2 H), 1.69 (m, 2 H), 1.50-1.40 (m, 2 H), 0.94 (s,
9 H), 0.87 (s, 18 H), 0.64 (s, 9 H), 0.20 (s, 3 H), 0.107 (s, 6 H),
0.087 (s, 3 H), 0.042 (s, 3 H), 0.036 (s, 3 H), -0.060 (s, 3 H),
-0.49 (s, 3 H); EIMS (MH+) m/z 983. Anal. Calcd for C41H83
-
IN4O7Si4‚C6H14: C, 52.74; H, 9.07; N, 5.23. Found: C, 52.92;
H, 8.71; N, 4.96.
Diben zyl 6-[9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsi-
lyl-D-r ibityl)-7,9-d ih yd r o-2,6-d im eth oxy-8-oxop u r in -5-yl]-
h exa n e 1-P h osp h a te (22). The phosphate 22 was obtained
by the procedure above in 58% yield as a colorless oil: IR (neat)
6-[9-(2′,3′,4′,5′-Tetr a-O-ter t-bu tyldim eth ylsilyl-D-r ibityl)-
7,9-d ih yd r o-2,6-d im et h oxy-8-oxop u r in -7-yl]-1-iod oh ex-
a n e (18). This compound was obtained by the procedure above
in 94% yield as a colorless oil: IR (neat) 2935, 2864, 1717,
1628 cm-1; 1H NMR (300 MHz, CDCl3) δ 4.47 (d, J ) 9.54 Hz,
1 H), 4.39 (dd, J ) 9.98, 13.81 Hz, 1 H), 4.06 (d, J ) 1.96 Hz,
2 H), 4.02 (s, 3 H), 3.92 (s, 3 H), 3.86 (m, 2 H), 3.77 (m, 1 H),
3.67 (dd, J ) 7.58, 10.31 Hz, 1 H), 3.56 (dd, J ) 5.62, 10.40
Hz, 1 H), 3.15 (t, J ) 6.96 Hz, 2 H), 1.79 (qn, 2 H), 1.67 (qn,
2 H), 1.38 (m, 4 H), 0.93 (s, 9 H), 0.87 (s, 18 H), 0.64 (s, 9 H),
0.20 (s, 3 H), 0.10 (s, 6 H), 0.080 (s, 3 H), 0.035 (s, 3 H), 0.029
(s, 3 H), -0.066 (s, 3 H), -0.49 (s, 3 H); EIMS (MH+) m/z 997.
Anal. Calcd for C42H85IN4O7Si4: C, 50.60; H, 8.53; N, 5.62.
Found: C, 50.22; H, 8.71; N, 5.45.
Diben zyl 3-[9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsi-
lyl-D-r ibityl)-7,9-d ih yd r o-2,6-d im eth oxy-8-oxop u r in -7-yl]-
p r op a n e 1-P h osp h a te (19). Compound 15 (0.551 g, 0.70
mmol) was dissolved in dry DMF (20 mL) under an atmosphere
of argon. Anhydrous K2CO3 (0.495 g, 3.58 mmol) and compound
32 (0.313 g, 0.70 mmol) were added, and the mixture was
stirred at room temperature for 24 h. Excess solvent was
removed using reduced pressure, and the residue was purified
by using flash chromatography (28 g, 5 × 40 cm column of
SiO2), eluting with hexanes-ethyl acetate (80:20), to furnish
19 (0. 512 g, 66%) as colorless oil: IR (neat) 2954, 1716, 1627
cm-1; 1H NMR (300 MHz, CDCl3) δ 7.33-7.29 (m, 10 H), 5.01
(m, 4 H), 4.46 (d, J ) 9.58 Hz, 1 H), 4.29 (dd, J ) 10.04, 13.95
Hz, 1 H), 4.09-3.96 (m, 5 H), 3.94 (s, 4 H), 3.92 (s, 3 H), 3.76
(m, 1 H), 3.66 (dd, J ) 7.60, 10.35 Hz, 1 H), 3.55 (dd, J ) 5.60,
10.45 Hz, 1 H), 2.01 (m, 2 H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.63
(s, 9 H), 0.20 (s, 3 H), 0.10 (s, 6 H), 0.08 (s, 3 H), 0.039 (s, 3 H),
0.033 (s, 3 H), -0.068 (s, 3 H), -0.52 (s, 3 H); EIMS (MH+)
m/z 1105. Anal. Calcd for C53H93PN4O11Si4: C, 57.61; H, 8.42;
N, 5.07. Found: C, 57.29; H, 8.23; N, 4.83.
2943, 2864, 1717, 1628 cm-1 1H NMR (300 MHz, CDCl3) δ
;
7.31 (s, 10 H), 5.01 (m, 4 H), 4.47 (d, J ) 9.31 Hz, 1 H), 4.30
(dd, J ) 9.99, 13.86 Hz, 1 H), 4.08 (d, J ) 1.78 Hz, 1 H), 4.02
(m, 1 H), 3.99 (s, 3 H), 3.96 (s, 1 H), 3.96 (s, 3 H), 3.93 (t, J )
2.40 Hz, 1 H), 3.82 (m, 3 H), 3.68 (dd, J ) 7.54, 10.34 Hz, 1
H), 3.56 (dd, J ) 5.8, 10.35 Hz, 1 H), 1.61 (m, 4 H), 1.30 (m, 4
H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.64 (s, 9 H), 0.20 (s, 3 H),
0.11 (s, 6 H), 0.095 (s, 3 H), 0.041 (s, 3 H), 0.034 (s, 3 H), -0.064
(s, 3 H), -0.50 (s, 3 H); 13C NMR (75 MHz, CDCl3) δ 159.8,
153.5, 153.0, 151.4, 135.9, 135.8, 128.5, 128.4, 127.8, 102.0,
77.6, 69.1, 69.0, 67.8, 67.7, 64.6, 60.3, 54.8, 53.8, 43.67, 42.3,
30.0, 29.6, 26.0, 25.9, 25.6, 18.3, 18.1, 17.5, -4.2, -4.3, -4.4,
-4.6, -5.3, -5.4, -5.8; 31P NMR (121 MHz, CDCl3) δ -0.28;
EIMS (MH+) m/z 1147. Anal. Calcd for C56H99PN4O11Si4: C,
58.63; H, 8.63; N, 4.88. Found: C, 58.52; H, 8.56; N, 4.86.
Gen er a l P r oced u r e for th e Syn th esis of In ter m ed ia tes
23-26. Compounds 19-22 (0.46 mmol) were dissolved in dry
THF (12 mL) and transferred to dry polyethylene vials
equipped with rubber septa and a magnetic stirring bars. The
cold (0 °C), magnetically stirred solutions were treated under
argon with HF-Py (2 mL) and stirred 1 h, and then the ice
bath was removed. Stirring was continued for another 6 h,
HF-Py (2 mL) was added again, and the mixtures were stirred
for 12 h. The cold (0 °C) solutions were diluted with ether (50
mL) and neutralized with saturated NaHCO3 solution (10 mL),
and then solid NaHCO3 was added slowly until the mixture
reached pH 7. The solutions were filtered, the filtrates were
extracted with chloroform (3 × 15 mL), and the extracts were
combined and concentrated. The residues were purified by
silica gel column chromatography (15 g, 3 × 30 cm column of
SiO2), eluting with CHCl3-MeOH (9:1), to afford the desired
compounds 23-26 as colorless oils.
Gen er a l P r oced u r e for th e Syn th esis of Diben zyl
P h osp h a tes 20-22. Compounds 16, 17, or 18 (0.95 mmol)
and silver dibenzyl phosphate (0.366 g, 0.95 mmol) were heated
at reflux in dry acetonitrile (20 mL) for 8 h under an
atmosphere of argon. The solutions were filtered and concen-
Diben zyl 3-(7,9-Dih yd r o-2,6-d im eth oxy-8-oxo-9-D-r ibi-
tylp u r in -7-yl)p r op a n e 1-P h osp h a te (23). This product was
obtained by the above procedure in 58% yield: IR (neat) 3384,
2953, 1696 cm-1; 1H NMR (500 MHz, CDCl3) δ 7.31-7.28 (m,
610 J . Org. Chem., Vol. 69, No. 3, 2004

9-D-Ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
10 H), 5.02-4.94 (m, 4 H), 4.18 (br s, 2 H), 4.09-4.05 (m,
2 H), 4.02-3.93 (m, 4 H), 3.95 (s, 3 H), 3.89 (s, 3 H), 3.79 (m,
2 H), 3.73 (m, 1 H), 3.48 (q, J ) 6.31 Hz, 1 H), 1.99 (m, 2 H);
31P NMR (121 MHz, CDCl3) δ -0.25; EIMS (MH+) m/z 649.
Anal. Calcd for C29H37N4O11P‚0.7H2O: C, 52.63; H, 5.81; N,
8.47. Found: C, 52.58; H, 5.91; N, 8.28.
Diben zyl 4-(7,9-Dih yd r o-2,6-d im eth oxy-8-oxo-9-D-r ibi-
tylp u r in -7-yl)bu ta n e 1-P h osp h a te (24). This intermediate
was obtained by the above procedure in 43% yield: 1H NMR
(300 MHz, CDCl3) δ 7.27 (s, 10 H), 4.94 (dd, J ) 12.5 and 20.7
Hz, 4 H), 4.78 (br s, 1 H), 4.17 (m, 2 H), 4.07 (m, 2 H), 3.96 (s,
3 H), 3.94 (m, 3 H), 3.87 (s, 3 H), 3.85 (m, 3 H), 3.75 (br s, 2
H), 3.47 (t, J ) 6.7 Hz, 1 H), 1.68 (m, 2 H), 1.59 (m, 2 H); 13C
NMR (75 MHz, CDCl3) δ 159.8, 153.8, 153.7, 151.1, 135.6,
135.5, 128.4, 127.8, 102.0, 73.3, 72.3, 71.7, 69.2, 69.1, 67.2, 67.1,
63.5, 54.9, 54.1, 43.4, 42.0, 26.9, 25.6; EIMS (MNa+) m/z 685.
Anal. Calcd for C30H39N4O11P‚0.5H2O: C, 53.60; H, 5.95; N,
8.34. Found: C, 53.50; H, 5.76; N, 8.30.
product was concentrated under reduced pressure to a very
small volume, and methanol (3 × 20 mL) was added and
distilled off. The residue was dissolved in deionized water. The
clear aqueous solution was lyophilized to result in a white
solid, which was dried under vacuum to afford 28 as a white
hygroscopic solid (60 mg, 63%): 1H NMR (300 MHz, D2O) δ
4.07-3.54 (m, 11 H), 1.75-1.57 (m, 4 H); 13C NMR (75 MHz,
D2O) δ 155.5, 152.9, 151.9, 139.3, 98.7, 73.0, 72.4, 69.8, 67.2,
62.8, 44.5, 42.2, 26.7, 25.3; ESIMS m/z 455 (MH+); HRMS calcd
for C14H23N4O11P 454.1179, found 454.1172. Anal. Calcd for
C14H23N4O11P‚0.3H2O: C, 36.54; H, 5.13; N, 12.18. Found: C,
36.54; H, 5.12; N, 11.81.
Diben zyl 3-Iod op r op yl 1-P h osp h a te (32). A mixture of
1,3-diiodopropane (1.20 mL, 10.4 mmol) and silver dibenzyl
phosphate (1.20 g, 3.13 mmol) in dry acetonitrile (30 mL) was
heated at reflux under argon for 10 h. The reaction mixture
was filtered, the filtrate was concentrated under reduced
pressure, and the residue was separated with flash chroma-
tography (SiO2, 230-400 mesh) (elution: hexanes-EtOAc 4:1)
to afford pure 32 (1.05 g, 86.0%) as colorless oil: 1H NMR (300
MHz, CDCl3) δ 7.34 (s, 10 H), 5.03 (m, 4 H), 4.02 (q, J ) 6.14
Hz, 2 H), 3.12 (t, J ) 6.75 Hz, 2 H), 2.03 (m, 2 H).
Diben zyl 5-(7,9-Dih yd r o-2,6-d im eth oxy-8-oxo-9-D-r ibi-
tylp u r in -7-yl)p en ta n e 1-P h osp h a te (25). This intermediate
was obtained by the above procedure in 80% yield: IR (neat)
1
3382, 2952, 1697 cm-1; H NMR (300 MHz, CDCl3) δ 7.26 (s,
10 H), 4.93 (m, 4 H), 4.81 (br s, 1 H), 4.42 (br s, 1 H), 4.23-
4.08 (m, 4 H), 3.97 (s, 3 H), 3.95-3.81 (m, 6 H), 3.87 (s, 3 H),
Dieth yl 5-[9-(2′,3′,4′,5′-Tetr a -O-ter t-bu tyld im eth ylsilyl-
D-r ibityl)-7,9-d ih yd r o-2,6-d im eth oxy-8-oxop u r in -7-yl]-1,1-
d iflu or op en ta n e 1-P h osp h on a te (34). Compound 15 (1.00
g, 1.27 mmol) was dissolved in dry DMF (30 mL) under an
atmosphere of argon. Anhydrous K2CO3 (0.898 g) and com-
pound 3324 (0.469 g, 1.27 mmol) were added, and the mixture
was stirred at room temperature for 24 h. Excess solvent was
removed using reduced pressure, and the residue was purified
by using flash chromatography (28 g, 5 × 40 cm column of
SiO2), eluting with hexanes-ethyl acetate (80:20), to furnish
34 (1.24 g, 95%) as colorless oil: 1H NMR (500 MHz, CDCl3)
δ 4.48 (d, J ) 9.70 Hz, 1 H), 4.29 (dd, J ) 10.17, 13.95 Hz, 1
H), 4.23 (q, J ) 7.31 Hz, 4 H), 4.09-4.02 (m, 2 H), 4.02 (s, 3
H), 3.93 (s, 3 H), 3.89 (m, 2 H), 3.77 (m, 1 H), 3.68 (dd, J )
7.78, 10.35 Hz, 1 H), 3.56 (dd, J ) 5.62, 10.5 Hz, 1 H), 2.07
(m, 2 H), 1.74 (m, 2 H), 1.63 (m, 2 H), 1.35 (t, J ) 7.05 Hz, 6
H), 0.94 (s, 9 H), 0.87 (s, 18 H), 0.64 (s, 9 H), 0.20 (s, 3 H),
0.11 (s, 6 H), 0.087 (s, 3 H), 0.043 (s, 3 H), 0.036 (s, 3 H), -0.061
(s, 3 H), -0.49 (s, 3 H); EIMS (MH+) m/z 1029. Anal. Calcd
for C45H91F2N4O10PSi4: C, 52.50; H, 8.91; N, 5.44. Found: C,
52.11; H, 8.72, N, 5.33.
3.75 (s, 2 H), 3.49 (br s, 1 H), 1.60 (m, 4 H), 1.30 (m, 2 H); 13
C
NMR (75 MHz, CDCl3) δ 159.6, 153.6, 153.5, 151.0, 135.5,
135.4, 128.3, 127.6, 101.9, 73.2, 72.1, 71.7, 69.0, 69.0, 67.5, 67.4,
63.4, 43.2, 42.2, 29.3, 29.2, 28.9, 22.0; EIMS (MH+) m/z 677.
Anal. Calcd for C31H41N4O11P‚0.2H2O: C, 54.68; H, 6.08; N,
8.23. Found: C, 54.67; H, 6.27; N, 7.93.
Diben zyl 6-(7,9-Dih yd r o-2,6-d im eth oxy-8-oxo-9-D-r ibi-
tylp u r in -7-yl)h exa n e 1-P h osp h a te (26). This product was
obtained by the above procedure in 58% yield: 1H NMR (300
MHz, CDCl3) δ 7.30 (s, 10 H), 4.98 (dd, J ) 12.97 and 20.89
Hz, 4 H), 4.24 (br s, 2 H), 4.04 (m, 2 H), 4.03 (s, 3 H), 3.96-
3.82 (m, 6 H), 3.93 (s, 3 H), 3.77 (d, J ) 3.98 Hz, 2 H), 3.38 (t,
J ) 6.44 Hz, 2 H), 1.66 (qn, J ) 6.36 Hz, 2 H), 1.57 (qn, J )
6.69 Hz, 2 H), 1.29 (m, 4 H); 13C NMR (75 MHz, CDCl3) δ 159.7,
153.7, 153.6, 151.0, 135.6, 135.5, 128.3, 127.7, 102.0, 73.1, 71.9,
71.8, 69.1, 69.0, 67.2, 67.1, 63.4, 54.8, 53.9, 43.2, 42.4, 29.8,
29.7, 29.4, 25.1, 24.7; 31P NMR (121 MHz, CDCl3) δ -0.26;
EIMS (MNa+) m/z 713. Anal. Calcd for C32H43N4O11P‚0.5
H2O: C, 54.93; H, 6.34; N, 8.00. Found: C, 55.00; H, 6.37; N,
7.96.
Diet h yl 5-(5,7-Dih yd r o-2,4-d im et h oxy-8-oxo-9-D-r ib i-
tylp u r in -5-yl)-1,1-d iflu or op en ta n e 1-P h osp h on a te (35).
Compound 34 (1.23 g, 1.19 mmol) was dissolved in dry THF
(20 mL) and the solution transferred to a dry polyethylene vial
equipped with a rubber septum and a magnetic stirring bar.
The cold (0 °C), magnetically stirred solution was treated
under argon with HF-Py (5.2 mL) and stirred for 1 h, and
then the ice bath was removed. Stirring was continued for
another 6 h, HF-Py (5.2 mL) was added again, and the
mixture was stirred overnight. The cold solution (0 °C) was
diluted with ether (50 mL) and neutralized with saturated
NaHCO3 solution. The solution was filtered, and the filtrate
was lyophilized. The residue was purified by silica gel column
chromatography (15 g, 3 × 30 cm column of SiO2), eluting with
CHCl3-MeOH (9:1) to afford desired compound 35 (0.309 g,
46%): 1H NMR (300 MHz, CDCl3) δ 4.79 (br s, 1 H), 4.69 (br
s, 1 H), 4.25-4.13 (m, 7 H), 4.00 (s, 4 H), 3.88 (s, 4 H), 3.79 (br
s, 1 H), 3.72 (br s, 2 H), 3.42 (br s, 1 H), 2.11-1.92 (m, 2 H),
1.72 (m, 2 H), 1.55 (m, 2 H), 1.29 (t, J ) 7.06 Hz, 6 H); 13C
NMR (75 MHz, CDCl3) δ 159.8, 153.8, 153.7, 151.1, 73.3, 72.4,
71.6, 64.5, 64.4, 63.5, 54.9, 54.1, 43.4, 42.2, 33.2 (q), 29.1, 17.6,
16.2, 16.2; EIMS (MH+) m/z 573. Anal. Calcd for C21H35F2N4-
PO10: C, 44.06; H, 6.16. Found: C, 44.02; H, 6.18.
3-(1,3,7,9-Tet r a h yd r o-9-D-r ib it yl-2,6,8-t r ioxop u r in -7-
yl)p r op a n e 1-P h osp h a te (27). Compound 23 (310 mg, 0.48
mmol) was dissolved in a mixture of concd HCl (5 mL) and
methanol (5 mL), and the mixture was heated at reflux for 3
h. The solvent was removed by vacuum distillation to result
in a very small volume. Methanol (3 × 20 mL) was added and
distilled off. The residue was dissolved in 2-propanol-
methanol (10:1) at reflux, and the solution was then cooled to
room temperature and the precipitate was collected by filtra-
tion. The solid was dissolved in deionized water, and the clear
water solution was lyophilized to result in a white solid, which
was dried under vacuum to afford 27 (132 mg, 60%) as a white
hygroscopic solid: 1H NMR (300 MHz, D2O) δ 4.11-3.66 (m,
11H), 2.07 (m, 2 H); 13C NMR (75 MHz, D2O) δ 155.2, 152.4,
151.5, 139.0, 98.5, 72.2, 69.5, 64.3, 62.6, 44.5, 40.1, 30.0;
ESIMS m/z 441 (MH+); HRMS calcd for C13H21N4O11P 440.1023,
found 440.1017. Anal. Calcd for C13H21N4O11P‚0.2 (CH3)2-
CHOH: C, 36.11; H, 5.04; N, 12.38. Found: C, 35.98; H, 5.10;
N, 12.14.
4-(1,3,7,9-Tet r a h yd r o-9-D-r ib it yl-2,6,8-t r ioxop u r in -7-
yl)bu ta n e 1-P h osp h a te (28). Compound 24 (240 mg, 0.21
mmol) was dissolved in a mixture of concd HCl (3 mL) and
methanol (3 mL), and the mixture was heated at reflux for
3 h. The solvent was removed by vacuum distillation, and the
residue was first separated by flash chromatography (SiO2),
eluting with MeOH-CH2Cl2-H2O-AcOH (7:4:2:1), and then
further purified with Dowex 50 1 × 8 anion-exchange resin,
eluting with 1.5% aq HCl. The fraction which contained the
5-(1,3,7,9-Tet r a h yd r o-9-D-r ib it yl-2,6,8-t r ioxop u r in -7-
yl)-1,1-d iflu or op en ta n e 1-P h osp h on a te (36). To a solution
of compound 35 (0.309 g, 0.54 mmol) in dry CH2Cl2 (10 mL)
was added trimethylsilyl iodide (0.23 mL), and the reaction
mixture was stirred at room temperature for 24 h under argon
atmosphere. Water (15 mL) was added and the mixture stirred
J . Org. Chem, Vol. 69, No. 3, 2004 611

Cushman et al.
at room temperature for 10 h and washed with CH2Cl2 (3 ×
10 mL). The water layer was lyophilized, resulting in 36 as a
colorless solid (0.235 g, 89%): mp 92-94 °C; 1H NMR (300
MHz, D2O) δ 3.85 (m, 2 H), 3.73 (m, 3 H), 3.65 (m, 1 H), 3.57
(d, J ) 2.89 Hz, 1 H), 3.53-3.42 (m, 2 H), 1.86 (m, 2 H), 1.56
(quint, J ) 6.93 Hz, 2 H), 1.35 (quint, J ) 6.29 Hz, 2 H); EIMS
(MH+) m/z 489. Anal. Calcd for C15H23N4F2O10P‚1.5H2O: C,
32.19; H, 4.26. Found: C, 32.10; H, 4.60.
3-[9-(2′,3′,4′,5′-Tetr a-O-ter t-bu tyldim eth ylsilyl-D-r ibityl)-
7,9-d ih yd r o-2,6-d im et h oxy-8-oxop u r in -7-yl]-1-iod op r o-
p a n e (44). This compound was prepared using the general
procedure described above for iodides 16-18: 1H NMR (300
MHz, CDCl3) δ 5.10 (dd, J ) 9.91, 3.09 Hz, 1 H), 4.41 (m, 4
H), 4.00 (m, 7 H), 3.80 (m, 1 H), 3.68 (m, 1 H), 3.57 (m, 3 H),
1.96 (m, 2 H), 0.86 (m, 27 H), 0.59 (s, 9 H), 0.10-0.06 (m, 21
H), -0.51 (s, 3 H). Anal. Calcd for C39H71IN4O7Si4: C, 49.03;
H, 8.34; N, 5.86. Found: C, 48.99; H, 7.92; N, 5.46.
3-[9-(2′,3′,4′,5′-Tetr a-O-ter t-bu tyldim eth ylsilyl-D-r ibityl)-
5,7-d ih yd r o-2,4-d im et h oxy-8-oxop u r in -5-yl]-1-p r op a n ol
(45). Compound 44 (238 mg, 0.25 mmol) was dissolved in a
solution of 1 N NaOH-MeOH (1:1, 10 mL), and the mixture
was heated at reflux for 12 h. The solvent was distilled off
under reduced pressure and the residue was purified by flash
chromatography (SiO2), eluting with hexanes-ethyl acetate
(5:1), to furnish compound 45 (160 mg, 75%) as a colorless oil:
1H NMR (300 MHz, CDCl3) δ 5.00 (dd, J ) 9.96, 3.11 Hz, 1
H), 4.29 (m, 4 H), 4.03 (s, 3 H), 3.91 (m, 4 H), 3.74 (m, 1 H),
3.61 (m, 1 H), 3.47 (m, 1 H), 3.13 (t, J ) 6.69 Hz, 2 H), 2.13
(m, 2 H), 0.84-0.77 (m, 27 H), 0.52 (s, 9 H), 0.06-0.05 (m, 18
H), 0.14 (s, 3 H), -0.59 (s, 3 H). Anal. Calcd for C39H80N4O8-
Si4: C, 55.41; H, 9.54; N, 6.63. Found: C, 55.12; H, 9.38; N,
6.24.
3-(1,3,7,9-Tet r a h yd r o-9-D-r ib it yl-2,6,8-t r ioxop u r in -7-
yl)-1-p r op a n ol (46). Compound 45 (150 mg, 0.18 mmol) was
dissolved in a solution of concd HCl-MeOH (1:1, 10 mL), and
the mixture was heated at reflux for 3 h. The solvent was
removed under reduced pressure, and the residue was repre-
cipitated from propanol. The solid was dried under vacuum
to afford 46 (25 mg, 35%) as a white solid: mp 187-188 °C;
1H NMR (300 MHz, D2O) δ 3.92 (m, 5 H), 3.66 (m, 6 H), 1.80
(m, 2 H); MS (ESI) 359 (M - 1). Anal. Calcd for C13H20N4O8‚
2.5H2O: C, 38.51; H, 6.17; N, 13.82. Found: C, 38.23; H, 5.72;
N, 13.44.
Molecu la r Mod elin g on Lu m a zin e Syn th a se. Using
Sybyl (Tripos, Inc., version 6.9, 2002), the X-ray crystal
structure of the complex of 5-nitro-6-ribitylamino-2,4-(1H,3H)-
pyrimidinedione (42) and the lumazine synthase of B. subtilis
(1RVV)26 was clipped to include information within a 15 Å
radius of one of the 60 equivalent ligand molecules. The
residues that were clipped in this cut complex were capped
with either neutral amino or carboxyl groups. The structure
of the inhibitor 27 was overlapped with the structure of 5-nitro-
6-ribitylamino-2,4-(1H,3H)-pyrimidinedione (42), which was
then deleted. Hydrogen atoms were added to the complex.
MMFF94 charges were loaded, and the energy of the complex
was minimized using the Powell method to a termination
gradient of 0.05 kcal/mol while employing the MMFF94 force
field. During the minimization of the complex, inhibitor 27 and
the 6 Å sphere surrounding it were allowed to remain flexible,
while the remaining portion of the complex was held ridged
using the aggregate function. Figure 2 was constructed by
displaying the amino acid residues of the enzyme that are
involved in hydrogen bonding with the inhibitor 27. The
maximum distance between donor and acceptor atoms con-
tributing to the hydrogen bonds shown in Figure 1 was set to
3.5 Å.
structure of E. coli riboflavin synthase (1I8D) was downloaded
and two molecules of the ligand 27 were docked and oriented
as suggested by the published model of the binding of two
molecules of the substrate 1 in the active site,30 as well as by
the structure of 43 bound to S. pombe riboflavin synthase.28
The C- and N-terminal groups were changed to neutral
carboxylic acid and amino groups, and hydrogens were added
to the protein structure and to the oxygens of the water
molecules. MMFF94 charges were loaded, and the energy of a
6 Å radius spherical subset including and surrounding the two
ligand molecules was minimized using the Powell method to
a termination gradient of 0.05 kcal/mol while employing the
MMFF94 force field. During energy minimization, the remain-
ing protein structure was held rigid using the aggregate
function. Figure 4 was constructed by displaying the amino
acid residues in the C- and N-barrels surrounding the two
ligand molecules.
Lu m a zin e Syn th a se Assa y.31 Reaction mixtures contained
100 mM potassium phosphate, pH 7.0, 5 mM EDTA, 5 mM
dithiothreitol, inhibitor (0-500 µM), 170 µM 5-amino-6-ribi-
tylamino-2,4(1H,3H)-pyrimidinedione (1) and B. subtilis lu-
mazine synthase (16 µg, specific activity 8.5 µmol mg-1 h-1
)
in a total volume of 1000 µL. The solution was incubated at
37 °C, and the reaction was started by the addition of a small
volume (20 µL) of L-3,4-dihydroxy-2-butanone 4-phosphate (2)
to a final concentration of 50-400 µM. The formation of 6,7-
dimethyl-8-ribityllumazine (3) was measured online with a
computer-controlled photometer at 408 nm (ꢀlumazine ) 10 200
M-1 cm-1). The velocity-substrate data were fitted for all
inhibitor concentrations with a nonlinear regression method
using the program DynaFit.32 Different inhibition models were
considered for the calculation. Ki values ( standard deviations
were obtained from the fit under consideration of the most
likely inhibition model. In a separate set of experiments, the
L-3,4-dihydroxy-2-butanone 4-phosphate concentration (2) was
kept constant at 100 µM and the concentration of the second
substrate, 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione
(1), was varied.
Ribofla vin Syn th a se Assa y.33 Reaction mixtures con-
tained buffer (100 mM potassium phosphate, 10 mM EDTA,
10 mM sodium sulfite), inhibitor (0 to 300 µM), and riboflavin
synthase (4.6 µg, specific activity 45 µmol mg-1 h-1). After
preincubation, the reactions were started by the addition of
various amounts of 6,7-dimethyl-8-ribityllumazine (3) (2.5-
200 µM) to a total volume of 1000 µL. The formation of
riboflavin (4) was measured online with a computer-controlled
photometer at 470 nm (ꢀriboflavin ) 9100 M-1 cm-1). The Ki
evaluation was performed in the same manner as described
above.
Ack n ow led gm en t. This research was made possible
by NIH Grant No. GM51469 as well as by support from
the Deutsche Forschungsgemeinschaft and Fonds der
Chemischen Industrie.
Su p p or tin g In for m a tion Ava ila ble: Procedures and
analytical data for the synthesis of compounds 29 and 30. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O030278K
(30) Liao, D.-I.; Wawrzak, Z.; Calabrese, J . C.; Viitanen, P. V.;
J ordan, D. B. Structure 2001, 9, 399-408.
(31) Kis, K.; Bacher, A. J . Biol. Chem. 1995, 270, 16788-16795.
(32) Kuzmic, P. Anal. Biochem. 1996, 237, 260-273.
(33) Eberhardt, S.; Richter, G.; Gimbel, W.; Werner, T.; Bacher, A.
Eur. J . Biochem. 1996, 242, 712-718.
Molecu la r Mod elin g on Ribofla vin Syn th a se. Using
Sybyl (Tripos, Inc., version 6.9, 2002), the X-ray crystal
612 J . Org. Chem., Vol. 69, No. 3, 2004