Kinetics and mechanism of allantoin racemization

Article abstract of DOI:10.1006/bioo.2000.1162

The kinetics and mechanism of racemization of allantoin have been examined; racemization proceeds via two independent pathways that can be separately monitored. One pathway involves proton exchange at C5 with solvent, The other pathway occurs via intramolecular attack of N8 on C4 to form a symmetrical bicyclic intermediate, which can decompose to form either enantiomer of allantoin. The intramolecular pathway proceeds more rapidly from the allantoin anion than from neutral allantoin. This result is explained by conformational analyses based on experimental NMR data and computational results, which suggest that the ureido arm of anionic allantoin adopts a cis-conformation, allowing intramolecular attack. Neutral allantoin adopts a trans-conformation. The proton exchange pathway is buffer-catalyzed and also proceeds more rapidly at basic pH, although it is suggested that the reaction occurs from neutral allantoin. The relatively slow rate of racemization, particularly at physiological pH, suggests that nonenzymatic racemization of allantoin is not a viable mechanism for the in vivo generation of (S)-allantoin. (C) 2000 Academic Press.

Full text of DOI:10.1006/bioo.2000.1162

Bioorganic Chemistry 27, 62–721 (2000)
doi:10.1006/bioo.2000.1162, available online at http://www.idealibrary.com on
Kinetics and Mechanism of Allantoin Racemization
Kalju Kahn and Peter A. Tipton1
Department of Biochemistry, University of Missouri-Columbia, Columbia, Missouri 65211
Received November 16, 1999
The kinetics and mechanism of racemization of allantoin have been examined; racemiza-
tion proceeds via two independent pathways that can be separately monitored. One pathway
involves proton exchange at C5 with solvent. The other pathway occurs via intramolecular
attack of N8 on C4 to form a symmetrical bicyclic intermediate, which can decompose to
form either enantiomer of allantoin. The intramolecular pathway proceeds more rapidly
from the allantoin anion than from neutral allantoin. This result is explained by conforma-
tional analyses based on experimental NMR data and computational results, which suggest
that the ureido arm of anionic allantoin adopts a cis-conformation, allowing intramolecular
attack. Neutral allantoin adopts a trans-conformation. The proton exchange pathway is
buffer-catalyzed and also proceeds more rapidly at basic pH, although it is suggested that
the reaction occurs from neutral allantoin. The relatively slow rate of racemization, par-
ticlarly at physiological pH, suggests that nonenzymatic racemization of allantoin is not a
viable mechanism for the in vivo generation of (S)-allantoin.
© 2000 Academic Press
INTRODUCTION
Allantoin (5-ureidohydantoin) serves as an important carrier of metabolic nitrogen
in tropical leguminous plants such as soybeans. In these plants nitrogen is fixed from
the atmosphere through the action of nitrogenase, which reduces gaseous dinitrogen
to ammonia. The ammonia is used to synthesize purines which are converted to urate
(1) and further oxidized through the ureide pathway to produce allantoin (3) and allan-
toate (4), the so-called ureides (Scheme 1).
The chemistry of the ureide pathway is surprisingly rich. Although it is widely
believed that allantoin is the product of the urate oxidase reaction (1,2) it has recent-
ly been demonstrated that it is not the immediate product of the enzymatic reaction,
and the details of the biogenesis of allantoin remain unknown. The true product of
the urate oxidase reaction is 5-hydroxyisourate (2), which quantitatively decompos-
es to allantoin under neutral aqueous conditions in vitro (3,4). The conversion of
5-hydroxyisourate to allantoin involves hydrolysis of the purine ring and an unusu-
al 1,2 carboxylate shift (3). It is unclear whether those transformations which are
observed in vitro are a part of the ureide pathway. Nonenzymatic decomposition
of optically active 5-hydroxyisourate which is produced by urate oxidase generates
1 To whom correspondence and reprint requests should be addressed. Fax: (573) 884-4812. E-mail:
tiptonp@missouri.edu.
62
0045-2068/00 $35.00
Copyright © 2000 by Academic Press
All rights of reproduction in any form reserved.

63
KAHN AND TIPTON
SCHEME 1
racemic allantoin (5), and it has been reported that allantoinase, which catalyzes the con-
version of allantoin the allantoate, is specific for the S-enantiomer of allantoin (6). Given
the metabolic importance of the ureide pathway, and the presumption that flux through
it should be rapid and efficient, two scenarios for the metabolic conversion of
5-hydroxyisourate to allantoate can be envisioned: that there exist hitherto undescribed
enzymes that mediate the stereospecific conversion of 5-hydroxyisourate to S-allantoin
or that allantoinase draws off the S-enantiomer from the pool of racemic allantoin, which
reequilibrates through nonenzymatic or enzymatic chemistry. There have been no
reports of an allantoin racemase occurring in plants, to our knowledge. In order to eval-
uate whether nonenzymatic racemization of allantoin could constitute a viable bio-
chemical pathway, we have examined the kinetics and mechanism of that reaction.
Like other 5-substituted hydantoins, allantoin could racemize via an SE2-type reac-
tion or an SE1 mechanism involving a carbanion intermediate that is stabilized by tau-
tomerization (Scheme 2) (7). Allantoin also has the unique possibility of racemizing
via formation of a symmetrical bicyclic intermeidate formed by attack of the primary
amide nitrogen at C4 (8). The former mechanisms involve cleavage of the C5-H bond
and can be monitored by measuring the rate at which the C5 proton exchanges with
solvent D2O. In the latter mechanism the C5-H bond is not cleaved, but if this mech-
anism is operative the rate of racemization can be determined by measuring the rate at
which C2 and C7 of allantoin interconvert. We have determined the pH-dependence
of proton exchange and of C2–C7 interconversion, as well as the pH-dependence of
allantoin racemization. In order to rationalize the kinetics data, we have examined the
conformation of neutral allantoin and its anion by NMR and computational methods.
EXPERIMENTAL SECTION
Allantoin was purchased from Sigma Chemical Co. and used without further purifi-
cation. [2-13C]Allantoin and [7-13C]allantoin were generated by the urate oxidase-
catalyzed oxidation of [8-13C]urate and [2-13C]urate, respectively. The syntheses of
regiospecifically-labeled [13C]urates have been described (3). Recombinant thioredoxin-
urate oxidase fusion protein was purified as described (18). Catalase was purchased
from Sigma Chemical Co. Enzyme-catalyzed urate oxidations were typically performed
by dissolving 0.03 g (0.17 mol) [13C]urate in 8 ml 100 mM phosphate buffer in D2O

64
KAHN AND TIPTON
SCHEME 2
at pD 7.2; approximately 5 units of urate oxidase and 1200 units of catalase were
added, and the solution was gently bubbled with O2 at room temperature for 5 h. The
enzymes were removed by ultrafiltration, and the filtrate containing [13C]allantoin was
evaporated to dryness under reduced pressure. (-)Allantoin was purified from a culture
of Pseudomonas aeruginosa, strain PAO1, grown on racemic allantoin as the sole
carbon source as described (19).
NMR spectroscopy. 1H NMR spectra of allantoin were acquired in a 50:50 (v/v) mixture
of d6-DMSO and CDCl3 with 0.12 Hz spectral resolution using a Bruker AVANCE DRX500
spectrometer. A selective 1D saturation experiment was performed by irradiating either the
N1-H resonance or the N8-H resonance and observing the coupling pattern of the C5 proton.
The NOESY spectrum of allantoin was recorded in d6-DMSO using a mixing time of 0.8 s
and a repetition time of 1.24 s. The three-bond proton–proton coupling constants were
obtained from the 1D spectrum of allantoin. The values of 3JCH (C4-C5-N1-H1) and 3JCH (C4-
C5-N6-H6) were obtained from the analysis of the 1H NMR spectrum of [4-13C]allantoin in
d6-DMSO. Values of 3JC(O)NCH were obtained from the analysis of 1H and 13C NMR spectra
of [2-13C]allantoin and [7-13C]allantoin in D2O at pD 7.0 and 10.9. The 13C NMR spectra that
provided the heteronuclear coupling information were recorded with a spectral resolution of
0.075 Hz using a gated decoupling pulse program. NOEs detected between pairs of protons
in allantoin were classified as strong, medium, or weak; the NOEs were compared to a ref-
erence NOE observed between C5-H and N1-H. In all conformers of allantoin the distance
between these two protons is 2.6 ± 0.1 Å, and the observed NOE was classified as strong.
The dihedral angles in allantoin and the corresponding three-bond coupling con-
stants were evaluated using Karplus-type relationships (20). The expected coupling
constants between C5-H and N1-H and between C5-H and N6-H in each conformer
were calculated using Eq. [1] (21). The expected value of the heteronuclear coupling
between C4 and N1-H and between C4 and N6-H was calculated using Eq. [2] (22).

ALLANTOIN RACEMIZATION
65
The expected values of the coupling constants between C5-H and C2 and between C5-
H and C7 were calculated with Eq. [3] (23).
3JHCNH = 6.4 cos2α − 1.4 cos α + 1.9
3JC(O)CNH = 9.5 cos2β − 4.4cos β - 0.8
3JC(O)NCH = 3.96 cos2θ − 1.83 cos θ + 0.81
[1]
[2]
[3]
Proton exchange assay. The exchange of the proton at C5 of allantoin with solvent
deuterium was studied at 22°C over the pH range 5.7 to 10.0 by integrating the allantoin pro-
ton resonance at 5.40 ppm relative to the signal from the internal standard 3-(trimethylsilyl)
propionate-2,2,3,3,−d4 in 99.9% D2O. Under these experimental conditions proton loss
from C5 is essentially irreversible. The pH of the solution was maintained at the desired
value using sodium phosphate. It was determined that phosphate catalyzed the allantoin
proton exchange with solvent; therefore, the reaction was conducted at several different
phosphate concentrations between 10 and 125 mM at each pH value. The decrease in
intensity of the C5 proton signal followed first order kinetics; the observed first order rate
constant was linearly dependent on phosphate concentration, so the rate constant for the
uncatalyzed exchange at each pH value was obtained by linear extrapolation to zero phos-
phate concentration. The pH dependence of the rate constants for the uncatalyzed reaction
was fitted to Eq. [4], where kmax is the plateau value of the rate constant Ka is the acid dis-
sociation constant governing the observed process.
logk = log [kmax/(1 + [H+]/Ka)]
[4]
13C Exchange assay. The exchange of 13C between C2 and C7 of allantoin was monitored
by 13C NMR at 22°C. Solutions containing 25 mM [7-13C]allantoin were prepared in phos-
phate buffer in 90% D2O; control experiments established that phosphate does not catalyze
the interconversion of C2 and C7. The rate of exchange was determined in the pH range 6.9
to 9.6. The exchange process followed first order kinetics. The observed rate constant for
exchange from C2 to C7 was the same as for exchange from C7 to C2 (data not shown). The
microscopic rate constant for conversion of [7-13C]allantoin to [2-13C]allantoin is identical to
the microscopic rate constant for the reverse process, and the microscopic rate constants are
equal to half of the observed rate constant. The pH dependence of the observed rate constants
was fitted to Eq. [4].
Circular dichroism assay. The loss of optical activity from solutions of (−)-allantoin was
monitored by circular dichroism spectropolarimetry. Solutions of 6 mM (−)-allantoin were
prepared in sodium phosphate buffers in the pH range 6.5 to 10.0. The phosphate buffer con-
centrations were 10, 100, and 200 mM; however, it was determined that the rate of loss of
optical activity was independent of the phosphate concentration. CD spectra were obtained
at 25°C in 1-mm pathlength quartz cuvettes. The maximum Cotton effect for allantoin was
observed to vary from 210 nm at pH 6.5 to 232 nm at pH 10, and the rate of loss of optical
activity at each pH value was measured by monitoring the CD signal at the signal maximum.
The observed CD intensities decreased exponentially over time, and the first order rate con-
stants determined at each pH value were fitted to Eq. [4].
Computational methods. Conformations of allantoin were evaluated computationally as
follows. Allantoin consists of a relatively rigid imidazoline ring and a more flexible ureido

66
KAHN AND TIPTON
sidechain attached to C5. A grid search with 20 degree spacing over the conformational
space of the ureido sidechanin was performed with a Tripos force field (24), using the
program SYBYL (Tripos, Inc.). The energies of 5832 conformations were evaluated and
six local minima were identified. Full molecular mechanics optimizations starting from
these minima converged to six conformations that were further studied with the AM1 semi-
empirical quantum mechanical model (25), using the program MOPAC (Version 6.0, U.S.
Air Force Academy). The AM1 geometries were used as starting points for optimization at
the HF/6-311+G(d,p) level with Gaussian 94 (26). Initial geometries for the optimization of
the allantoin anions were obtained by removing a proton from N3 of the neutral conformer.
RESULTS AND DISCUSSION
It seems clear that racemization of allantoin can occur via two distinct pathways
which can be monitored independently. Figure 1 illustrates that C2 and C7 of allantoin
interconvert over time under physiologically relevant conditions. The most likely mech-
anism for this interconversion is via the intermediacy of the bicyclic species shown in
FIG. 1. Isotopic exchange between C2 and C7 of allantoin at pD 7.4. Starting material was [7-
13C]allantoin. The lines through the experimental points were obtained by fitting the data to the equation
describing a first order process.

ALLANTOIN RACEMIZATION
67
Scheme 2, which was suggested long ago (9). In early studies of urate oxidase, it was
observed that allantoin arising from the oxidation of [2-14C]urate contained 14C in both
the ureido portion of the molecule and in the hydantoin ring (10); these results were
interpreted to be an indication that the urate oxidase reaction proceeded through a sym-
metrical intermediate. It is now apparent that the 14C labels equilibrated after formation
of allantoin. The observation that [14C]allantoin prepared by heating 5-chlorohydantoin
with [14C]urea contained the label distributed between the ureido sidechain and the
hydantoin ring was also explained in terms of the bicyclic intermediate (11).
Scheme 2 illustrates the C2–C7 scrambling process occuring by attack of N8 on the
C4 carbonyl carbon; an alternative possibility is that N8 attacks C4 in the enol(ate) tau-
tomer shown in the SE1 pathway. However, the finding that the rate constant for C2–C7
scrambling is much larger than for proton exchange indicates that proton exchange is not
a prerequisite for C2–C7 scrambling. A mechanism in which proton exchange occurs
from the bicyclic intermediate via dehydration and rehydration can be excluded, because
it would require the formation of an intermediate in which C4 and C5 are equivalent,
and no scrambling between those two positions was observed (data not shown).
Formation of the bicyclic intermediate and its conversion back to allantoin does not
involve cleavage of the C5–H bond, so the observation that C5–H exchanges with
solvent suggests the existence of another pathway for racemization of allantoin. Two
mechanisms are possible and are illustrated in Scheme 2. The SE1 mechanism involves
enolization of allantoin and reprotonation; in the SE2 (back) mechanism incorporation
of solvent deuterium at C5 occurs concomitant with the proton leaving from the oppo-
site face. In general, the SE1 and SE2 (back) mechanisms can be distinguished by com-
parisons of the rate constants for racemization and deuteration. In the SE1 mechanism
the rate constants for those two processes should be equal, whereas in the SE2 (back)
mechanism, since each deuteriation event results in inversion of stereochemistry and
the reduction of the enantiomeric excess in the solution by two molecules, the rate con-
stant for racemization is twice that for deuteration (12). In the case of allantoin, the
major pathway for racemization is through the formation of the bicyclic intermedi-
ate, and against that background our kinetic data are not precise enough to determine
whether the SE1 or SE2 (back) mechanism is operative.
The pH dependence of the observed rate constants for proton exchange from C5 of
allantoin, interconversion of C2 and C7, and the loss of optical activity from a solution
of (–)-allantoin are shown in Fig. 2. All three curves are similar, showing a plateau at
high pH and rates that decrease with decreasing pH. These results are consistent with
earlier measurements of the pH dependence of allantoin racemization and hydrolysis,
although in that study the separate pathways for racemization were not distinguished
(13). The pK defined by monitoring the loss of optical activity is 8.43 ± 0.07, in good
13
agreement with the value of 8.7 determined by C NMR spectroscopy (3). The pKs
13
defined by the proton exchange experiment and the C scrambling experiment are
8.92 ± 0.09 and 9.2 ± 0.2, respectively. These values were determined in D2O and cor-
respond to pKs in H2O of 8.5 and 8.8, also in good agreement with the values determined
by other methods.
The pH dependence of the rate constants for both pathways was similar; both rate
constants decreased by a factor of 10 per pH unit below the pK of allantoin. 13C NMR
titrations of allantoin indicate that the site of ionization in the anion is N3 (3). In order

68
KAHN AND TIPTON
FIG. 2. Observed pH dependence of the rate constants for loss of optical activity in (−)-allantoin (᭺);
C2–C7 interconversion (᭹); and proton exchange with solvent (᭿). Loss of optical activity was determined
in H2O, but C2–C7 interconversion and proton exchange experiments were conducted in D2O.
to gain insight into the origin of the kinetic behavior, the conformations of neutral
allantoin and the allantoin anion were examined by NMR and computational methods.
Geometry optimization of neutral allantoin at the HF/6-31 + G(d,p) level yielded
four stable conformers in the gas phase (Table 1). The two most stable conformers
were characterized by a trans-amide bond in the ureido side-chain and are designated
trans-1 and trans-2. Two stable conformers with cis-amide bonds were found, and are
designated cis-1 and cis-2; the cis-2 conformer adopted a scorpion-like conformation
with the ureido group above the imidazoline ring and a torsion angle a of 21°. The rel-
ative energies of the stable conformers of neutral allantoin are reported in Table 2.
The experimental NMR data are consistent with the assignment of neutral allantoin
to an extended trans configuration. Observed coupling constants are reported in Table
2 and NOE data are in Table 3. Only a very weak NOE was observed between the pro-
ton at N1 and the proton at N6, consistent with the calculated distances between them
of 3.8 and 3.5 Å, respectively. In constrast, in the cis-1 conformation, the calculated
distance between N1-H and N6-H is 2.8 Å, and a strong NOE would have been
expected between C5-H and the protons on N8, but no NOE was observed between
those atoms. The cis-2 conformation predicts a medium-strength NOE between N1-H
and the protons on N8, but only a very weak NOE was observed.
Table 2 compares the observed and calculated coupling constants for the neutral
allantoin conformers. Good agreement is evident for the trans-2 conformer. The cis-2
conformer shows a reasonably close match as well, but, as described above, it is not
consistent with the NOE data. (R,S)-Allantoin in the crystalline form adopts a con-
formation that is very similar to the trans-2 conformer (14). The energy barrier for

ALLANTOIN RACEMIZATION
69
TABLE 1
Conformers of Neutral Allantoin
interconversion of the two trans conformers of allantoin is small, and in solution it is
likely that an ensemble of conformers exists in an equilibrium mixture.
NOEs were not obtained for the allantoin anion because it is soluble only in aque-
ous solutions where the rapid exchange of protons from nitrogens precluded NOE
measurements. Two heteronuclear coupling constants were measured in the allantoin
3
3
anion; JCH (C2–N1–C5–H5) was 2.89 Hz and JCH (C7–N6–C5–H5) was 4.54 Hz.
These data alone do not allow the solution conformation of the allantoin anion to be
determined unambiguously, but the computational results gave a clear indication of
a preference for a single conformer. The cis-1 conformer was the most stable con-
former (Table 4); the trans-1 conformer was 4.0 kcal/mol higher in energy, and the
cis-2 and trans-2 conformers were not stable. A distinct cis-conformer, designated
cis-3, was stable only as the anion, but was 5.2 kcal/mol higher in energy than the cis-1
conformer.

70
KAHN AND TIPTON
TABLE 2
Observed and Calculated Coupling Constants (Hz) for Neutral Allantoin and the Four Stable
Conformers with HF/6-31+G(d,p) Optimized Geometries
Egas
C5H-N6H(α)
C5H-N1H
2.20
C4-N6H(β)
C4-N1H
7.78
C5H-C7(γ)
C5H-C2
2.61
Observed
Conformer
Trans-1
Trans-2
Cis-1
8.37
1.98
6.63
0.00
1.90
3.02
3.16
1.94
6.42
5.18
6.17
3.22
1.83
4.00
3.82
2.60
0.57
1.98
0.00
12.8
11.7
12.0
11.8
1.03
6.33
2.25
6.37
1.72
2.65
1.94
1.25
Cis-2
Note. The relative energies (Egas) were evaluated at the MP2/6-31+G(d,p)//HF/6-31+G(d,p) level and are in
kcal/mol.
The different conformations adopted by neutral allantoin and the allantoin anion
provide an explanation for the observed pH dependence of racemization. For the
C2–C7 scrambling to occur, which is the major pathway for racemization, N8 must
act as an intramolecular nucleophile toward C4. The cis conformation adopted by the
ureido arm of ionized allantoin places N8 in an appropriate position for this reaction
to occur, while the trans conformation which is the lowest energy conformation in
neutral allantoin prevents intramolecular attack by positioning N8 far from C4.
Two explanations for the observed pH dependence are formally consistent with the
observed data. The first is that the allantoin anion racemizes more rapidly than neutral
allantoin; the second is that the reaction occurs only from neutral allantoin and is
catalyzed by hydroxide. For the C2–C7 interconversion pathway it is possible to rule out
the hydroxide catalysis pathway, because studies conducted with methylallantoins indi-
cate that ionization of N3 is a prerequisite for reversible formation of the bicyclic inter-
mediate. In the last century it was shown that oxidation of either 1-methyluric acid or
7-methyluric acid yielded 3-methylallantoin as the sole product (15). This observation
can be explained by a mechanism in which the two expected isomeric methylallan-
toins form initially, but only 8-methylallantoin forms the bicyclic species, which
decomposes to reform both isomeric methylallantoins; however, 3-methylallantoin
cannot form the bicyclic species. Thus, all of the 8-methylallantoin is eventually con-
verted to 3-methylallantoin. Similarly, 6,8-dimethylallantoin isomerizes irreversibly to
1,3-dimethylallantoin (11).
For the proton exchange reaction, it is difficult to rationalize why exchange should
occur more rapidly from the anion than from neutral allantoin. Indeed, preliminary
TABLE 3
Experimentally Observed NOEs in Neutral Allantoin
δ(ppm)
N3H
C5H
N6H
N8H
N1H
N3H
C5H
N6H
8.00
10.30
5.17
None
Strong
None
Weak
None
Strong
Weak
None
None
7.02
Medium

ALLANTOIN RACEMIZATION
71
TABLE 4
Conformers of the Allantoin Anion
results suggest that C5–H proton exchange occurs in a facile manner from neutral
3-methylallantoin (16). These data suggest that the observed pH dependence arises
from buffer-catalyzed reaction with neutral allantoin. One can envision mechanisms
for general base catalysis of either the SE1 or SE2 (back) mechanism. Although kinet-
ic studies of the racemization of a number of substituted hydantoins have led to the
conclusion that those reactions proceeded via the SE2 pathway (7), our data do not
allow us to make a distinction between the two mechanistic possibilities.
CONCLUSIONS
At neutral pH the half-time for allantoin racemization is greater than 10 h. This
would seem to present a significant physiological bottleneck to allantoin metabolism
since allantoinase acts only on (S)-allantoin and suggests that plants may require a
mechanism for the stereospecific generation of allantoin in the ureide pathway. In fact,

72
KAHN AND TIPTON
we have isolated a novel enzyme from soybean root nodules which catalyzes the hydrol-
ysis of 5-hydroxyisourate (17). The nonenzymatic conversion of 5-hydroxyisourate to
allantoin begins with this reaction as well (3), so the isolation of this protein provides
the first indication of an enzymatic, potentially stereospecific pathway to (S)-allantoin.
Enzyme-dependent metabolism of 5-hydroxyisourate would be expected to avoid the
production of racemic allantoin and obviates the slow nonenzymatic equilibration of
the allantoin enantiomers, which is described in this report. Studies of the non-
enzymatic chemistry in the ureide pathway underscore the importance of the stereo-
specific enzymatic processes which are currently under investigation.
ACKNOWLEDGMENTS
This work was supported by USDA Grant 98-35305-6548. The NMR spectrometers that were used in
this work were obtained through Grants 8908304 and 9221835 from the National Science Foundation.
REFERENCES
1. Reynolds, P. H. S., Boland, M. J., Blevins, D. G., Randall, D. D., and Schubert, K. R. (1982) Trends Biol.
Sci. 366–368.
2. Voet, D., Voet, J., and Pratt, C. W. (1999) Fundamentals of Biochemistry, Wiley, New York.
3. Kahn, K., Serfozo, P., and Tipton, P. A. (1997) J. Am. Chem. Soc. 119, 5435–5442.
4. Modric, N., Derome, A. E., Ashcroft, S. J. H., and Poje, M. (1992) Tet. Lett. 33, 6691–6694.
5. Kahn, K. (1998) Kinetic and Mechanistic Characterization of the Urate Oxidase Reaction, Ph. D. thesis,
University of Missouri, Columbia.
6. Lee, K. W., and Roush, A. H. (1964) Arch. Biochem. Biophys. 108, 460–467.
7. Reist, M., Carrupt, P.-A., Testa, B., Lehmann, S., and Hansen, J. J. (1996) Helv. Chim. Acta. 79, 767–778.
8. The numbering of the carbon atoms for allantoin is shown in Scheme 2. (S)-Allantoin is shown as the reactant.
9. Titherley, A. W. (1913) J. Chem. Soc. 103, 1336–1338.
10. Canellakis, E. S., and Cohen, P. P. (1955) J. Biol. Chem. 213, 385–395.
11. Abblard, J., and Meynaud, A. (1971) Bull. Soc. Chim. Fr. 3, 942–946.
12. Cram, D. J. (1965) in Fundamentals of Carbanion Chemistry (Blomquist, A. T., Ed), pp. 85–113, Academic
Press, New York.
13. Vogels, G. D., de Windt, F. E., and Bassie, W. (1969) Rec. Trav. Chim. 88, 940–950.
14. Mootz, D. (1965) Acta Cryst. 19, 726–734.
15. Fischer, E., and Ach, F. (1899) Berichte 32, 2721–2749.
16. Horseradish peroxidase was used to oxidize 7-methylallantoin, and the resulting product, 3-methylallantoin,
was observed to undergo facile proton exchange at C5. K. Kahn, unpublished observations.
17. Sarma, A. D., Serfozo, P., Kahn, K., and Tipton, P. A. (1999) J. Biol. Chem. 274 (48), 33863–33865.
18. Kahn, K., and Tipton, P. A. (1998) Biochemistry 37, 11651–11659.
19. ’s-Gravenmade, E. J., Vogels, G. D., and van Pelt, C. (1969) Rec. Trav. Chim. 88, 929–939.
20. Eberstadt, M., Gemmecker, G., Mierke, D. F., and Kessler, H. (1995) Angew. Chem. Int. Ed Engl. 34,
1671–1695.
21. Evans, J. N. S. (1995) Biomolecular NMR Spectroscopy, Oxford Univ. Press, Oxford.
22. Hansen, P. E. (1981) Prog. Nucl. Magn. Reson. Spectrosc. 14, 175–296.
23. Kao, L.-F., and Barfield, M. (1985) J. Am. Chem. Soc. 107, 2323–2330.
24. Clark, M., Cramer III, R. D., and Van Opdenbosch, N. (1989) J. Comp. Chem. 10, 982–1012.
25. Dewar, M. J. S., Zoebisch, E. G., Healy, E. F., and Stewart, J. J. P. (1985) J. Am. Chem. Soc. 107, 3902–3909.
26. Frisch, M. J., Trucks, G. W., Schlegel, H. B., Gill, P. M. W., Johnson, B. G., Robb, M. A., Cheeseman, J. R.,
Keith, T., Petersson, G. A., Montgomery, J. A., Raghavachari, K., Al-Laham, M. A., Zakrzewski, V. G.,
Ortiz, J. V., Forseman, J. B., Cioslowski, J., Stefanov, B. B., Nanayakkara, A., Challacombe, M., Peng,
C. Y., Ayala, P. Y., Chen, W., Wong, M. W., Andres, J. L., Replogle, E. S., Gomperts, R., Martin, R. L.,
Fox, D. J., Binkley, J. S., Defrees, D. J., Baker, J., Stewart, J. P., Head-Gordon, M., Gonzalez, C., and Pople,
J. A. (1995) Gaussian 94, Revision C.3, Gaussian, Pittsburgh, PA.