Assignment of downfield proton resonances in purine nucleoside phosphorylase immucillin-H complex by saturation-transferred NOEs.
ABSTRACT Purine nucleoside phosphorylase (PNP) catalyzes N-ribosidic bond phosphorolysis in 6-oxypurine nucleosides and deoxynucleosides to form purine and alpha-D-phosphorylated ribosyl products. The transition state has oxacarbenium ion character with partial positive charge near C-1', ionic stabilization from the nearby phosphate anion, and protonation at N-7 of the purine. Immucillin-H (ImmH) has a protonated N-7 and resembles the transition-state charge distribution when N-4' is protonated to the cation. It binds tightly to the PNPs with a K(d) value 56 pM for human PNP. Previous NMR studies of PNP. ImmH.PO(4) have shown that the N-4' of bound ImmH is a cation and is postulated to have a significant contribution to its tight binding. Several unassigned downfield proton resonances (>11 ppm) are specific to the PNP. ImmH.PO(4) complex, suggesting the existence of strong hydrogen bonds. In this study, two of the proton resonances in this downfield region have been assigned. Using (15)N-7-labeled ImmH, a resonance at 12.5 ppm has been assigned to N-7H. The N-7H resonance is shifted downfield by only approximately 1 ppm from its position for ImmH free in aqueous solution, consistent with only a small change in the hydrogen bonding on N-7H upon binding of ImmH to PNP. In contrast, the downfield resonance at 14.9 ppm in the PNP. ImmH.PO(4) complex is assigned to N-1H of ImmH by using saturation-transferred NOE measurements on the PNP. ImmH complex. The approximately 4 ppm downfield shift of the N-1H resonance from its position for ImmH free in solution suggests that the hydrogen bonding to the N-1H in the complex has a significant contribution to the binding of ImmH to PNP. The crystal structure shows Glu201 is in a direct hydrogen bond with N-1H and to O-6 through a water bridge. In the complex with 6-thio-ImmH, the N-1H resonance is shifted further downfield by an additional 1.5 ppm to 16.4 ppm, but the relative shift from the value for 6-thio-ImmH free in solution is the same as in the ImmH complex. Since the binding affinity to hPNP for 6-thio-ImmH is decreased 440-fold relative to that for ImmH, the loss in binding energy is primarily due to the hydrogen bond energy loss at the 6-thiol.
- [show abstract] [hide abstract]
ABSTRACT: Genetic defects in human purine nucleoside phosphorylase cause T-cell deficiency as the major phenotype. It has been proposed that efficient inhibitors of the enzyme might intervene in disorders of T-cell function. Compounds with features of the transition-state structure of purine nucleoside phosphorylase were synthesized and tested as inhibitors. The transition-state structure for purine nucleoside phosphorylase is characterized by (1) an elevated pKa at N7 of the purine ring for protonation or favorable H-bond interaction with the enzyme and (2) oxocarbenium ion formation in the ribosyl ring (Kline, P. C., and Schramm, V. L. (1995) Biochemistry 34, 1153-1162). Both features have been incorporated into the stable transition-state analogues, (1S)-1-(9-deazahypoxanthin-9-yl)-1,4-dideoxy-1,4-imino-D-ribitol (immucillin-H) and (1S)-1-(9-deazaguanin-9-yl)-1,4-dideoxy-1, 4-imino-D-ribitol (immucillin-G). Both inhibitors exhibit slow-onset tight-binding inhibition of calf spleen and human erythrocyte purine nucleoside phosphorylase. The inhibitors exhibit equilibrium dissociation constants (Ki) from 23 to 72 pM and are the most powerful inhibitors reported for the enzyme. Complete inhibition of the homotrimeric enzyme occurs at one mole of inhibitor per mole of enzymic trimer. Binding of the transition-state inhibitor at one site per trimer prevents inhibitor binding at the remaining two sites of the homotrimer. A mechanism of sequential catalysis at each subunit, similar to that of F1 ATPase, is supported by these results. Slow inhibitor dissociation (e.g., t1/2 of 4.8 h) suggests that these compounds will have favorable pharmacologic properties. Interaction of transition-state inhibitors with purine nucleoside phosphorylase is different from reactant-state (substrate and product analogue) inhibitors of the enzyme which bind equally to all subunits of the homotrimer.Biochemistry 07/1998; 37(24):8615-21. · 3.38 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Plasmodium falciparum is responsible for the majority of life-threatening cases of malaria. Plasmodia species cannot synthesize purines de novo, whereas mammalian cells obtain purines from de novo synthesis or by purine salvage. Hypoxanthine is proposed to be the major source of purines for P. falciparum growth. It is produced from inosine phosphorolysis by purine nucleoside phosphorylase (PNP). Immucillins are powerful transition state analogue inhibitors of mammalian PNP and also inhibit P. falciparum PNP as illustrated in the accompanying article (Kicska, G. A., Tyler, P. C., Evans, G. B., Furneaux, R. H., Kim, K., and Schramm, V. L. (2002) J. Biol. Chem. 277, 3219-3225). This work tests the hypothesis that erythrocyte and P. falciparum PNP are essential elements for growth and survival of the parasite in culture. Immucillin-H reduces the incorporation of inosine but not hypoxanthine into nucleic acids of P. falciparum and kills P. falciparum cultured in human erythrocytes with an IC(50) of 35 nm. Growth inhibition by Imm-H is reversed by the addition of hypoxanthine but not inosine, demonstrating the metabolic block at PNP. The concentration of Imm-H required for inhibition of parasite growth varies as a function of culture hematocrit, reflecting stoichiometric titration of human erythrocyte PNP by the inhibitor. Human and P. falciparum PNPs demonstrate different specificity for inhibition by immucillins, with the 2'-deoxy analogues showing marked preference for the human enzyme. The IC(50) values for immucillin analogue toxicity to P. falciparum cultures indicate that inhibition of PNP in both the erythrocytes and the parasite is necessary to induce a purine-less death.Journal of Biological Chemistry 03/2002; 277(5):3226-31. · 4.65 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Immucillins are logically designed transition-state analogue inhibitors of mammalian purine nucleoside phosphorylase (PNP) that induce purine-less death of Plasmodium falciparum in cultured erythrocytes (Kicska, G. A., Tyler, P. C., Evans, G. B., Furneaux, R. H., Schramm, V. L., and Kim, K. (2002) J. Biol. Chem. 277, 3226-3231). PNP is present at high levels in human erythrocytes and in P. falciparum, but the Plasmodium enzyme has not been characterized. A search of the P. falciparum genome data base yielded an open reading frame similar to the PNP from Escherichia coli. PNP from P. falciparum (P. falciparum PNP) was cloned, overexpressed in E. coli, purified, and characterized. The primary amino acid sequence has 26% identity with E. coli PNP, has 20% identity with human PNP, and is phylogenetically unique among known PNPs with equal genetic distance between PNPs and uridine phosphorylases. Recombinant P. falciparum PNP is catalytically active for inosine and guanosine but is less active for uridine. The immucillins are powerful inhibitors of P. falciparum PNP. Immucillin-H is a slow onset tight binding inhibitor with a K(i)* value of 0.6 nm. Eight related immucillins are also powerful inhibitors with dissociation constants from 0.9 to 20 nm. The K(m)/K(i)* value for immucillin-H is 9000, making this inhibitor the most powerful yet reported for P. falciparum PNP. The PNP from P. falciparum differs from the human enzyme by a lower K(m) for inosine, decreased preference for deoxyguanosine, and reduced affinity for the immucillins, with the exception of 5'-deoxy-immucillin-H. These properties of P. falciparum PNP are consistent with a metabolic role in purine salvage and provide an explanation for the antibiotic effect of the immucillins on P. falciparum cultured in human erythrocytes.Journal of Biological Chemistry 03/2002; 277(5):3219-25. · 4.65 Impact Factor
Assignment of Downfield Proton Resonances in Purine Nucleoside
Phosphorylase‚Immucillin-H Complex by Saturation-Transferred NOEs†
Hua Deng,*,‡Andrzej Lewandowicz,‡Sean M. Cahill,‡Richard H. Furneaux,§Peter C. Tyler,§Mark E. Girvin,‡
Robert H. Callender,‡and Vern L. Schramm‡
Department of Biochemistry, Albert Einstein College of Medicine, 1300 Morris Park AVenue, Bronx, New York 10461, and
Carbohydrate Chemistry Team, Industrial Research Limited, Lower Hutt, New Zealand
ReceiVed October 7, 2003; ReVised Manuscript ReceiVed NoVember 21, 2003
ABSTRACT: Purine nucleoside phosphorylase (PNP) catalyzes N-ribosidic bond phosphorolysis in 6-oxy-
purine nucleosides and deoxynucleosides to form purine and R-D-phosphorylated ribosyl products. The
transition state has oxacarbenium ion character with partial positive charge near C-1′, ionic stabilization
from the nearby phosphate anion, and protonation at N-7 of the purine. Immucillin-H (ImmH) has a
protonated N-7 and resembles the transition-state charge distribution when N-4′ is protonated to the cation.
It binds tightly to the PNPs with a Kd value 56 pM for human PNP. Previous NMR studies of PNP‚
ImmH‚PO4have shown that the N-4′ of bound ImmH is a cation and is postulated to have a significant
contribution to its tight binding. Several unassigned downfield proton resonances (>11 ppm) are specific
to the PNP‚ImmH‚PO4complex, suggesting the existence of strong hydrogen bonds. In this study, two of
the proton resonances in this downfield region have been assigned. Using15N-7-labeled ImmH, a resonance
at 12.5 ppm has been assigned to N-7H. The N-7H resonance is shifted downfield by only ∼1 ppm from
its position for ImmH free in aqueous solution, consistent with only a small change in the hydrogen
bonding on N-7H upon binding of ImmH to PNP. In contrast, the downfield resonance at 14.9 ppm in the
PNP‚ImmH‚PO4complex is assigned to N-1H of ImmH by using saturation-transferred NOE measurements
on the PNP‚ImmH complex. The ∼4 ppm downfield shift of the N-1H resonance from its position for
ImmH free in solution suggests that the hydrogen bonding to the N-1H in the complex has a significant
contribution to the binding of ImmH to PNP. The crystal structure shows Glu201 is in a direct hydrogen
bond with N-1H and to O-6 through a water bridge. In the complex with 6-thio-ImmH, the N-1H resonance
is shifted further downfield by an additional 1.5 ppm to 16.4 ppm, but the relative shift from the value
for 6-thio-ImmH free in solution is the same as in the ImmH complex. Since the binding affinity to hPNP
for 6-thio-ImmH is decreased 440-fold relative to that for ImmH, the loss in binding energy is primarily
due to the hydrogen bond energy loss at the 6-thiol.
Purine nucleoside phosphorylase (PNP)1catalyzes the
reversible phosphorolysis of the C-1′ to N-9 bond of
6-oxypurine nucleosides and deoxynucleosides (Scheme 1).
In humans, the metabolic role is to remove deoxyguanosine
that accumulates from DNA turnover and the genetic
deficiency of PNP causes a T-cell immunodeficiency due to
dGTP accumulation in dividing T-cells (1). Inhibition of PNP
inhibits the growth of activated T-cells, providing a clinical
means to ameliorate T-cell proliferative disorders (2). The
catalytic acceleration of PNP and phosphoribosyltransferases
is achieved through stabilization of an enzyme-bound oxa-
carbenium ion precursor and strong leaving group interac-
tions to the purine. These features facilitate ribosyl electro-
phile migration from the purine leaving group to a phosphorus
nucleophile immobilized at the catalytic site (3). Immucillins
incorporate features of the proposed transition states and are
potent inhibitors of PNPs from bovine, human, microbial,
and malarial sources (4-6), as well as for human and
malarial HGPRTases (7-9). ImmH (Chart 1) binds with a
Kdof 23 pM to bovine PNP and 56 pM to the human enzyme,
making it a potent inhibitor of these enzymes.
Features that make ImmH similar to the transition state
include an N-4′ substitution for O-4′ in the sugar ring and
protonation at N-7 of the 9-deaza ring. Previous NMR studies
have shown that ImmH bound to Mycobacterium tuberculosis
and human PNP is doubly protonated to form the cation at
N-4′ and thus mimic the oxocarbenium ion transition state
(10). Protonation at N-7 results from replacement of nitrogen
by carbon in the 9-deazapurine and from the formation of a
chemically stable carbon-carbon bond between the C-1′ of
†Supported by Research Grants EB001958 (R.H.C.) and GM41916
(V.L.S.) from the NIH and an award from the New Zealand Office of
Science and Technology.
* To whom correspondence should be addressed. Phone: (718) 430-
2437. Fax: (718) 430-8565. E-mail: email@example.com.
‡Albert Einstein College of Medicine.
§Industrial Research Limited.
1Abbreviations: 1 hPNP, human purine nucleoside phosphorylase;
hgPRTase, hypoxanthine-guanine pyrophosphoribosyltransferase; ImmH,
immucillin-H; ImmHP, immucillin-H 5′-phosphate; S-ImmH, 6-thio-
immucillin-H; 9dHX, 9-deazahypoxanthine; STNOE, saturation-
transferred nuclear Overhauser effect; STD, saturation transfer differ-
ence; HMQC, heteronuclear multiple quantum coherence, HSQC,
heteronuclear single quantum coherence: HMBC, heteronuclear mul-
tiple bond correlation; DPFGSE, double pulsed field gradient spin-
echo; TSP, perdeuterated 3-(trimethylsilyl)propionate sodium salt.
Biochemistry 2004, 43, 1980-1987
10.1021/bi0358115 CCC: $27.50© 2004 American Chemical Society
Published on Web 01/29/2004
the sugar analogue and the 9-deazapurine moiety. It has been
suggested that the elevated pKaof >10 at N-7 may capture
a strong hydrogen bond interaction between the purine and
the enzyme that is a feature of the transition state but not
the Michaelis complex. This interaction has been identified
by NMR and X-ray crystallography in complexes of HGPRT‚
ImmHP‚P2O7(7-9) and by X-ray crystallography in PNP‚
The NH resonances for PNP are overlapped in the proton
NMR spectrum except for a few unassigned resonances that
are in the downfield region of >11 ppm (11). However, these
downfield resonances are of interest in NMR studies since
the corresponding protons are often involved in strong
hydrogen-bonding interactions in enzymes (12-16) and are
often at or near the active site. Assignment of these
resonances enables one to characterize these hydrogen bonds
in the enzyme and enhance our understanding of important
enzyme-ligand interactions. As reviewed by Mildvan et al.
(16), three methods are normally used to assign the downfield
proton resonances in the proton NMR spectra of protein-
ligand complexes. The most conclusive assignments of
downfield resonances have been aided by15N labeling of
the inhibitor using the15N HMQC type of measurements
(see, e.g., refs 7, 17, and 18). Two other methods include
mutagenesis of protein (19) or alteration of ligand (20) and
transient or truncated driven NOEs (12, 14, 18, 21). Here,
all three techniques have been used to assign the downfield
resonances in the proton spectrum of the hPNP‚ImmH‚PO4
complex. However, it has been found in our enzyme system
that the transient and the truncated driven NOE experiments
could not yield positive results. The transient NOEs failed
because the rapid incoherent exchange process between NH
protons of bound and free ligands and the solvent is faster
than the coherent transfer process under our experimental
conditions. The truncated driven NOEs failed because of the
difficulty in assigning the C-2H/C-8H resonances of the
bound ligand. Thus, we have developed a variation of these
NOE methods, the saturation-transferred NOE (STNOE)
method, to aid the assignment of N-1H and N-7H proton
resonances found in the downfield NMR spectrum of PNP
complexed with immucillin inhibitors.
In this study, both the N-1H and N-7H proton resonances
in the human PNP‚ImmH‚PO4complex have been assigned
and compared with their values free in solution. To our
surprise, the N-7H proton resonance shifts by only ∼1 ppm
downfield upon ImmH binding to the complex compared to
its value in aqueous solution, suggesting that the hydrogen-
bonding change on N-7H is significant but not large. On
the other hand, a ∼4 ppm downfield shift of the N-1H proton
upon ImmH binding to the complex is observed, consistent
with formation of a short, strong hydrogen bond for N-1H
in the complex. The results suggest that the major contribu-
tion to the binding of the base portion of ImmH is from N-1H
rather than from N-7H. Substitution of 6-thio-ImmH for
ImmH does not noticeably affect the hydrogen bonding on
N-1H, suggesting the 440-fold binding affinity decrease for
S-ImmH compared with ImmH in the hPNP complex is
primarily due to the loss of hydrogen bonding at C-6dS.
Leaving group activation is a major catalytic force for PNP,
and the results presented here establish the dominant interac-
tions at C-6dO and N-1 of the purine ring. The relative H
bond lengths explain site-directed mutagenesis and inhibitor
specificity of mammalian PNPs (2, 22, 23).
MATERIALS AND METHODS
The compounds ImmH, 6-thio-ImmH (S-ImmH), and [15N-
7]ImmH were synthesized as reported (24, 25). Human PNP
is a homotrimer of molecular weight 96000 (108000 with
the N-terminal extension used in this study). Human purine
nucleoside phosphorylase was produced in recombinant form
in E. coli BL21(DE3) containing the human PNP cDNA in
a T7/NT TOPO vector and purified to homogeneity on an
affinity resin. Samples of hPNP for NMR were prepared by
dialysis of the sample with 0.5 M NaCl at pH 6.0 or 7.0
without buffer, followed by sample concentration with a
Centricon-100 to a final concentration of 2 mM monomer,
as determined by UV using an extinction coefficient of 30
mM-1cm-1at 280 nm. The concentration of ImmH was
determined by UV using an extinction coefficient of 9.5
mM-1cm-1at 261 nm (4). The hPNP‚ImmH complex was
prepared by adjusting the pH of ImmH solutions to 6.0 or
7.0, and lyophilizing appropriate quantities to powder. The
proper amount of dry ImmH was added to the concentrated
hPNP. The concentration ratio of hPNP to ImmH was 2 mM:
10 mM. The ternary complexes of hPNP‚(ImmH or S-ImmH)‚
PO4were prepared by adding ImmH and phosphate to the
concentrated hPNP solution in 0.5 M NaCl to a final
concentration ratio of 2 mM:4 mM:10 mM. The ternary
complexes of hPNP‚[15N-7]ImmH‚PO4 were prepared by
mixing hPNP, inhibitor, and phosphate with a concentration
ratio of 1:1.5:5, and the complex was washed three times in
a Centricon-100 with 10 volumes of 0.5 M NaCl at pH 6.0,
so that the final free inhibitor concentration was less than
0.1% of the protein concentration. D2O (5%) was added to
all NMR samples in aqueous solution to provide deuterium
Scheme 1: Reactions Catalyzed by PNP (X ) H or OH, Y ) H or NH2)
Structure of ImmH, Transition-State Analogue for
Downfield Proton Resonances in PNP‚Immucillin-H
Biochemistry, Vol. 43, No. 7, 2004 1981
Solution NMR measurements for proteins, inhibitors, and
protein-ligand complexes were obtained on a DRX300 MHz
Bruker instrument. 1D proton spectra for samples in aqueous
solution were typically collected using 1-1 water suppres-
sion using 512 scans, a sweep width of 30 ppm sampled
with 16K points, and a recycle delay of 1.5 s. The 1-1 delay
was adjusted so that the maximum excitation was at ∼13
ppm, and a 5 Hz line broadening was applied on all presented
spectra. Chemical shifts in most 1D spectra were referenced
to water resonance (4.7 ppm) unless indicated otherwise. The
15N edited 1D spectra of [15N-7]ImmH were acquired using
a 1D version of the HMQC experiment with 7K scans, where
all proton pulses were replaced by their 1-1 counterparts
(7). 2D15N HMQC spectra for the hPNP‚[15N-7]ImmH‚PO4
complex were acquired with the 1-1 version of the HMQC
experiments. They were collected with 2K and 16 complex
points in t2(1H) and t1(15N), respectively, with 128 scans
per t1point and a recycle delay of 1.5 s. The value of t1,max
was set to 0.88 ms, and t2,max ) 120 ms. An exponential
line broadening (10 Hz) was applied to the proton dimension
in data processing, and in the15N dimension, the data were
linearly predicted to 64 points and zero filled to 128 points
before processing with a cosine bell window function. A
sensitivity-enhanced HSQC experiment was used for 2D
15N-1H single or multiple bond correlation spectra of [15N-
7]ImmH in DMSO-d6or in aqueous solution (with 5% D2O).
In these cases, the t1,maxin the15N dimension was increased
to 6.5 ms with 64 data points and 8 scans for each point.
The data in the15N dimension were linearly predicted to
128 points and zero filled to 256 points before processing
with a cosine bell window function. The chemical shifts in
the proton dimension in the 2D spectra are referenced to
that of TSP at 0.0 ppm.
General considerations in performing saturation-transferred
NOE (STNOE) experiments, such as determination of the
effects of spin diffusion and optimizing experimental condi-
tions, will be discussed in theoretical and experimental detail
with relaxation and exchange matrix calculations (Deng et
al., submitted for publication). Experimentally, the STNOE
measurements were performed on a protein-ligand mixture
with excess ligand. This method is based on the principles
of truncated driven NOE (26) and exchange-transferred NOE
(27). It is expected to work with protein-ligand systems with
the binding constant in the millimolar to submicromolar
range, a condition which is similar to that of the exchange-
transferred NOE. The implementation of the STNOE mea-
surement is similar to that of the recently published STD
measurement (28, 29) except that the transfer period is
limited to the initial NOE buildup of the proton resonances
of the free ligand and just one protein resonance is selected
for saturation. In our implementation, a selective Gaussian-
shaped inversion pulse was applied repeatedly on (or off)
the proton resonance of interest for a desired time before
the observation pulse. A DPFGSE pulse sequence was used
for water suppression (30). The saturation time was con-
trolled by a loop counter as well as the length of the selective
inversion pulse; in a typical measurement, the length of the
selective pulse was set to 25 or 30 ms. The acquisition time
was set to 0.92 s and recycle delay to 2.5 s. The data were
sampled with 16K points with 512 to 1K scans for each
The downfield region of the proton NMR spectrum of
hPNP in aqueous solution (with 5% D2O) at pH 7.0 and at
40 °C obtained with a 1-1 pulse sequence contains four
resonances at 15.7, 13.7, 12.7, and 11.6 ppm (Figure 1A).
The intensities of these resonances decrease substantially with
decreasing temperature. In the spectrum of hPNP in a binary
complex with the inhibitor ImmH under the same conditions,
the intensities of the two resonances at 15.7 and 13.7 ppm
are greatly reduced and new resonances appear at 14.7 ppm
and at ∼12.5 ppm, as a shoulder to the resonance at 12.7
ppm (Figure 1B). The binding of the inhibitor results in the
narrowing of the enzyme proton resonances. Bovine PNP is
known from H/D exchange to tighten its protein structure
in response to ImmH binding (31). The proton NMR
spectrum of the hPNP‚ImmH‚PO4ternary complex under the
same conditions shows the most downfield proton resonance
at 14.9 ppm, slightly shifted downfield from that in the binary
complex (compare parts B and C of Figure 1). It is reasonable
to assume that they are due to the same proton. In addition,
approximately three overlapping resonances between 12 and
13 ppm are observed, and several unresolved new resonances
between 11 and 12 ppm are also observed. The dissociation
constant of ImmH to PNP in the absence of phosphate is
approximately 0.44 µM, orders of magnitude looser than that
in the presence of the phosphate because of an ionic
interaction between bound PO4and the iminoribitol cation
of bound ImmH. This geometry suggests that the resonances
at 14.9 ppm and between 12 and 13 ppm may be related to
the interaction between the base portion of bound ImmH
while other resonances may be related to the binding of
phosphate and/or the iminoribitol group of ImmH. Figure
1D shows the spectrum of the hPNP‚S-ImmH‚PO4complex
under the same conditions. The most noticeable change
FIGURE 1: Proton spectra of (A) hPNP at 2 mM, (B) hPNP‚ImmH
complex (2 mM:10 mM), (C) hPNP‚ImmH‚PO4complex (2 mM:4
mM:10 mM), (D) hPNP‚S-ImmH‚PO4complex (2 mM:4 mM:10
mM), and (E) 10 mM ImmH in aqueous solution at pH 5 and 25
°C. All samples were prepared in 95% H2O + 5% D2O with 0.5
M NaCl. The pH was 7.0 and the temperature was 40 °C for all
protein samples. Other experimental details are described in the
Material and Methods.
1982 Biochemistry, Vol. 43, No. 7, 2004
Deng et al.
compared to the hPNP‚ImmH‚PO4 spectrum is the most
downfield shifted resonance at 16.4 ppm, suggesting that this
resonance is due to a proton near C-6 of (S-)ImmH in the
complex. No significant changes were observed in these
complexes when the pH was adjusted between 6 and 8.
Figure 1E shows the spectrum of ImmH alone in aqueous
solution (5% D2O) at pH 5 and at 25 °C. At this pH, the
N-4′ (pKa6.9) in ImmH is a doubly protonated cation (10).
A single resonance at 11.5 ppm is observed in this spectral
region. The line width of this resonance increases with
increasing temperature and/or increasing pH.
Assignments of the ImmH N-7H resonances in the enzyme
complexes and in aqueous solution were achieved using [15N-
7]ImmH. The proton spectrum of hPNP‚[15N-7]ImmH‚PO4
complex in aqueous solution (5% D2O), pH 6.0 at 35 °C,
with15N decoupling during acquisition (Figure 2A) was
similar to that of the unlabeled ImmH complex obtained
under somewhat different conditions (compare Figure 1C).
Using a 1D version of the
resonance at 12.5 ppm in the hPNP‚ImmH‚PO4complex was
identified as the N-7H proton (Figure 2B). The environmental
change of the N-7H moiety upon ImmH binding to the
enzyme complex was established by comparison of the
proton NMR spectrum for the [15N-7]ImmH in aqueous
solution with that in the complex. Proton spectra in 95%
H2O + 5% D2O, pH 6.0 at 4 °C, with and without15N
decoupling during acquisition was used to assign the signal
at 11.4 ppm to the N7-H proton (Figure 2C,D). Thus, an
about 1 ppm downfield shift of the N-7H proton resonance
occurred upon binding of ImmH to hPNP‚ImmH‚PO4.
The effect of solvent on the chemical shift of N-7 and
N-7H was measured using 2D15N-1H correlation measure-
ments with [15N-7]ImmH in DMSO, H2O + 5% D2O at pH
15N HMQC experiment, the
6, and with the hPNP‚[15N-7]ImmH‚PO4 ternary complex
at pH 6.0 (Figure 3). Due to the rapid chemical exchange of
the N-7H proton with water, the direct measurement in
aqueous solution did not yield a 2D spectrum with good
signal-to-noise ratio within a reasonable time for ImmH at
this pH. Thus, the15N chemical shift of the N-7 of ImmH in
aqueous solution was obtained from a
spectrum in which the N-7 shows a correlation to the C-8H
proton. The result of this measurement was combined with
the result shown in Figure 2C to yield the predicted resonance
(represented by a square) in Figure 3. It can be seen that
while the changes of the N-7 and N-7H chemical shifts of
ImmH upon binding to the hPNP complex were significant,
they do not suggest a strong hydrogen bonding on N-7H in
Measurements by the STNOE method were used to map
other downfield proton resonances from the enzyme-bound
inhibitors. By selectively saturating one of the downfield
proton resonances observed in the protein-ligand complex,
an NOE may be observed on one or more of the free ligand
protons through one- or multiple-step spin-spin relaxation
(spin diffusion) and then through the exchange between
bound and free ligands. Such NOEs should occur when the
proton of the bound ligand is spatially close to the proton
being saturated, or when the saturation time is long enough.
The NOE buildup could be due to either direct spin-spin
relaxation or spin diffusion, followed by the exchange
between free and bound ligands. The closer this proton is to
the one being saturated, the shorter the saturation time needed
to observe the NOE on this proton in the free ligand. Thus,
by monitoring the NOE buildup of the proton resonance of
the free ligands versus the saturation time, one can determine
the proximity of this proton to that being saturated. Under
favorable conditions, for example, when an internal reference
is available, the downfield proton resonance may be assigned.
This method relies on the exchange between bound and
free ligands. Since the binding of ImmH in the hPNP‚ImmH‚
PO4ternary complex is extremely tight (56 pM), we applied
this method to the hPNP‚ImmH complex with a concentra-
tion ratio of 2 mM:10 mM. The STNOE buildup of the
aromatic proton resonances in free ImmH was readily
FIGURE 2: Proton spectra of (A) hPNP‚[15N-7]ImmH‚PO4complex
(2 mM:2 mM:2 mM) with15N decoupling during acquisition, (B)
the same protein complex with15N filtering (HMQC), (C) [15N-7]-
ImmH in aqueous solution with15N decoupling during acquisition,
and (D) [15N-7]ImmH in aqueous solution without15N decoupling
during acquisition. All samples were prepared in 95% H2O + 5%
D2O with 0.5 M NaCl. The pH was 6.0 and the temperature was
35 °C for all protein samples. For ImmH solution samples, the pH
was 6.0 and temperature was 4 °C.
complex at pH 6.0 compared to [15N-7]ImmH in DMSO and in
aqueous solution at pH 6.0. The chemical shifts are referenced to
that of TSP. Other experimental details are described in the Material
15N-1H correlation spectra of hPNP‚[15N-7]ImmH‚PO4
Downfield Proton Resonances in PNP‚Immucillin-H
Biochemistry, Vol. 43, No. 7, 2004 1983
apparent upon saturation of the resonance at 14.7 ppm in
the spectrum of the hPNP‚ImmH complex (Figure 4). The
top spectrum in Figure 4 is the 1-1 proton spectrum of the
hPNP‚ImmH complex, and the arrow points to the resonance
being saturated. It is obvious that the STNOE on the C-2H
resonance increases as the time of saturation on the 14.7 ppm
resonance increases, establishing that the 14.7 ppm resonance
belongs to a proton that is close to the C-2H of ImmH in
the complex but is not close to C-8H.
The broad band between 12 and 13 ppm consists of two
nearly overlapping resonances which can be clearly resolved
on a 600 MHz spectrometer (data not shown). The STNOE
buildup of the free ImmH upon saturation of the resonance
at 12.5 ppm (the upfield side of the broad band, known to
be from N-7H as shown by the15N HMQC measurements
using [15N-7]ImmH, Figures 2 and 3) is shown in Figure 5.
The top proton spectrum is from the hPNP‚ImmH complex.
The arrow points to the 12.5 ppm resonance being saturated.
In this case, both the C-2H and C-8H protons of ImmH
showed STNOE buildup with increasing saturation of the
N-7H resonance at 12.5 ppm. The STNOE intensity buildup
on the C-2H resonance started ∼50 ms later than that on
the C-8H and increased at a significantly slower rate. These
results indicate qualitatively that the 12.5 ppm resonance
belongs to a proton that is closer to the C-8H of ImmH in
the complex than to C-2H. Comparing Figures 4 and 5, one
can see that the STNOE on C-2H and C-8H appears at about
the same saturation time and increases at the same rate with
increasing saturation time upon saturation of the resonances
at 14.7 and 12.5 ppm, respectively. Thus, the distance
between the proton with the resonance at 14.7 ppm and the
C2-H should be the same as the distance between the proton
with the 12.5 ppm resonance and the C-8H. Since it is known
that the 12.5 ppm resonance in the complex is due to N-7H,
we conclude that the 14.7 ppm resonance must be due to
Our measurement of S-ImmH in DMSO shows that the
chemical shift of N-1H in DMSO is ∼1.5 ppm downfield
shifted (data not shown) relative to that of N-1H in ImmH
in DMSO (Figure 3), which is the same relationship between
the 16.4 ppm resonance in the hPNP‚S-ImmH‚PO4complex
(Figure 1D) and the 14.9 ppm resonance in the hPNP‚ImmH‚
PO4 complex (Figure 1C). Therefore, it is reasonable to
conclude that the 16.4 and 14.9 ppm resonances are from
N-1H of S-ImmH and ImmH, respectively, in the complex.
The observation of an STNOE buildup on C-2H upon
saturation of the resonance at 12.5 ppm was puzzling. While
it could be argued that N-7H is close enough to C-2H so
that an indirect NOE can occur through spin diffusion, it
cannot explain why a similar STNOE effect was not observed
on C-8H when the resonance at 14.7 ppm was saturated
(Figure 4). Several possibilities for this observation are
explored in the Discussion.
As a control, similar measurements were made using
selective pulses on several other resonances, including the
resonances at 12.8 and 11.5 ppm of the hPNP‚ImmH
complex. No STNOE effect was observed in these cases until
the saturation time exceeded ∼350 ms. One of these control
measurements, with the selective pulse on the 11.5 ppm
resonance (indicated with an arrow), is shown in Figure 6.
An STNOE effect on both the C-2H and C-8H resonances
started to appear at about the same time when the saturation
time reached 360 ms. Spin diffusion pathways within the
protein are likely responsible for their appearance in the
STNOE experiment. Such results indicate that the resonances
at 12.8 and 11.5 ppm belong to protons that are relatively
far away from the binding site of the ImmH base. Since
resonances at the same chemical shifts are also observed in
the apo-hPNP spectrum (Figure 1A), these protons may have
The immucillins are transition-state analogues of N-
ribosyltransferases, and chemical synthesis of these analogues
with isotopic labels provides novel spectroscopic probes.
FIGURE 4: STNOE buildup experiments with the hPNP‚ImmH
complex (2 mM:10mM, pH 7 at 35 °C). The selective pulse was
set on resonance at 14.7 ppm (arrow), and the saturation times are
indicated next to each curve. Experimental details are described in
the Material and Methods.
FIGURE 5: STNOE buildup experiments with the hPNP‚ImmH
complex (2 mM:10mM, pH 7 at 35 °C). The selective pulse was
set on resonance at 12.5 ppm (arrow), and the saturation times are
indicated next to each curve. Experimental details are described in
the Material and Methods.
1984 Biochemistry, Vol. 43, No. 7, 2004
Deng et al.
Transition-state structures for N-ribosyltransferases have been
obtained from kinetic isotope effects (KIEs) and computa-
tional chemistry (32, 33). The N-ribosyltransferases cleave
the C-N bond of nucleosides and nucleotides in dissociative
mechanisms that form partially developed oxacarbenium ions
at the anomeric carbon of ribose in the transition state. The
bond order to the nucleophile at the transition state is
minimal, implying that transition-state formation is domi-
nated by protonation of the purine ring to make it a better
leaving group. Electrostatic stabilization of the oxacarbenium
ion by the nearby phosphate anion at the enzyme active site
also assists in forming the transition state. The transition state
for bovine PNP is formed when the N-ribosidic bond length
has increased by only 0.4 Å, followed by a >1.0 Å
translocation of the ribosyl C-1′ toward the enzymatically
fixed nucleophile (33). This mechanism has been proposed
to occur in most N-ribosyltransferases (34).
ImmH incorporates distinct features of the transition state
including positive charge on the iminoribitol and an elevated
pKa at N-7 of the 9dHX ring. The nitrogen at N-7 is
protonated as a consequence of the 9-deaza modification,
which also provides chemical stability in the C-1′ to C-9
bond replacing the N-ribosidic bond. These features are
missing in the reactant state but are found in the transition
state for PNP. The affinity of ImmH for PNP exceeds that
of normal substrates by approximately 106in the hPNP‚
ImmH‚PO4 complex. Complexes of the inhibitors with
several PNPs have been characterized by X-ray crystal-
lography. Recent NMR studies have shown that the N-4′ of
ImmH is cationic when bound to PNP (10), and our current
NMR experiments contribute information about the environ-
ment of the NH groups of the bound ImmH base.
Hydrogen-Bonding Interaction on N-1H in hPNP‚ImmH‚
PO4. Our STNOE studies establish that the most downfield
resonance of this complex at 14.9 ppm is due to N-1H
(Figure 4). Compared with its value in DMSO, the N-1H
resonance is downfield shifted by ∼4 ppm, suggesting a
relatively short hydrogen bond to N-1H in the complex
according to an empirically established correlation between
hydrogen-bonding distances and NH/OH proton chemical
shift values (16, 35). Several factors that cause a ligand N-H
proton chemical shift change in proteins include ring current,
local hydrogen bonding and/or long-rang electrostatic inter-
actions, and distortion of the ligand (for a recent treatment
of the problem, see Neal et al. (36)). The X-ray crystal
structure of this complex with bovine PNP reveals that the
nearby Tyr188 ring is perpendicular to the 9-deaza ring, with
its C-OH bond parallel to the C-2H bond of ImmH. Both
N-7H and N-1H of ImmH are in approximately the same
plane as this Tyr ring about 5 Å away from the center of the
ring (3). Thus, the ring current from this Tyr may contribute
to the observed downfield shift of the N-1H and N-7H proton
resonances. However, the estimated downfield shift of a
proton resonance caused by an aromatic ring current more
than 4 Å away is expected to be quite small, <0.2 ppm (37).
Thus, it is reasonable to suggest that the major contribution
to the downfield chemical shift change of N-1H/N-7H upon
ImmH binding to PNP is due to hydrogen bonding/
electrostatic interactions. The hydrogen-bonding distances
can also be determined by measurement of H/D fractionation
factors (12, 17, 19, 38, 39). Currently, such measurements
are under way.
The N-1H resonance is only slightly affected by the
binding of phosphate to the enzyme complex even though
the binding of ImmH to hPNP is ∼5 orders of magnitude
tighter in the presence of phosphate. The N-7H resonance is
also insensitive to the binding of phosphate, suggesting that
the binding energies from the 9dHX portion remain constant
and those from the iminoribitol portion of ImmH increase
the total binding energy of ImmH in the PNP‚ImmH‚PO4
complex. Previous studies based on the chemical substitutions
of ImmH on bovine PNP are consistent with this proposal
(11). For hPNP, a 6-thio substitution causes a 440-fold
decrease of ImmH binding affinity in the ternary complex
from 56 pM to 25 nM (our unpublished observation). This
is in clear contrast with the bovine PNP, in which the binding
affinity of S-ImmH is decreased 82600-fold relative to that
of ImmH (11). Although the N-1H resonance in the hPNP‚
S-ImmH‚PO4complex is 1.5 ppm downfield from that found
in the corresponding ImmH complex, the N-1H chemical
shift of S-ImmH is also 1.5 ppm downshifted compared with
that of ImmH in DMSO (data not shown). Thus, the ∼4 ppm
downfield shift of the N-1H proton relative to the values
free in solution is induced by the hydrogen bonding in their
respective hPNP complexes for both ImmH and S-ImmH.
Such results suggest that the N-1H chemical shift difference
is due to the difference in the molecular structures of the
two inhibitors rather than due to the hydrogen bonding to
N-1H in the ImmH and S-ImmH PNP complexes. Thus, the
loss of the binding energy in the S-ImmH complex is
primarily due to the loss of bonding on C-6dS.
Hydrogen-Bonding Interaction on N-7H in hPNP‚ImmH‚
PO4. On the basis of the transition-state structure of PNP, a
strong hydrogen bond between N-7H and the carbonyl from
the carboxamide of Asn243 might be expected. However,
the N-7 and N-7H resonances shift downfield by only ∼2
and ∼1 ppm, respectively, compared with their values in
aqueous solution (Figure 3). These results establish that the
hydrogen bond at N-7H becomes only modestly shorter in
FIGURE 6: STNOE buildup experiments with the hPNP‚ImmH
complex (2 mM:10mM, pH 7 at 35 °C). The selective pulse was
set on resonance at 11.5 ppm (arrow), and the saturation times are
indicated next to each curve. Experimental details are described in
the Material and Methods.
Downfield Proton Resonances in PNP‚Immucillin-H
Biochemistry, Vol. 43, No. 7, 2004 1985
the hPNP‚ImmH‚PO4 complex than in unliganded ImmH.
The N-7H resonance is known to be sensitive to hydrogen
bonding since in structures involving HGPRTase complexed
with immucillins, the downfield shift of the N-7H resonances
is correlated with the increased binding affinity of the
inhibitor, with the most downfield shift found at 14.3 ppm
(7). The X-ray structure of bovine PNP‚ImmH‚PO4indicates
that the hydrogen-bonding distance between N-7 and the side
chain carboxamide oxygen of Asn243 is 2.8 Å ((0.2 Å),
the same distance as that found in the structure of a similar
complex involving HGPRTase (8, 9). However, the proton
acceptor in HGPRTase is a carboxyl oxygen of Asp137. The
carboxamide oxygen is a weaker hydrogen bond partner with
the N-7H in PNP and may explain the reduced hydrogen-
bonding energy in hPNP. Transition-state stabilization
compares Michaelis and transition states. The Michaelis
complex of PNP with inosine shows a distance of 3.3 Å
between unprotonated N-7 of inosine and a group that could
be the carbonyl or the amino group of Asn243 (1, 20). The
weak interaction for substrate and the change of the H bond
acceptor/donor nature of N-7 on conversion of substrate to
transition state results in a favorable interaction between
N-7H of ImmH and the carboxamide oxygen of Asn243 at
2.8 Å. This interaction provides a powerful differential force
for transition-state stabilization.
Possibility of a Third Proton Resonance near 12.5 ppm
in hPNP‚ImmH. There are two resonances between 12 and
13 ppm in the spectrum of the hPNP‚ImmH complex (Figure
1B; these become resolved in the spectrum taken with a 600
MHz spectrometer, data not shown). When the downfield
side of the 12.5 ppm resonance is being saturated, no STNOE
is observed (the same as shown in Figure 6). However, when
the upfield side of the 12.5 ppm resonance is being saturated,
an STNOE effect is observed not only on C-8H but also on
C-2H of ImmH (Figure 5). The STNOE effect on C-2H starts
to appear 50 ms later and with a slower rate of increase than
that on C-8H. This STNOE effect on C-2H is not likely due
to direct spin-spin relaxation between N-7H and C-2H since
they are about 6 Å apart. It cannot be due to spin diffusion
through N-1H since if that is the case, a similar STNOE
effect would be observed on C-8H when N-1H is being
saturated, which is not observed (Figure 4). The possibility
that the base of ImmH flips during binding in the hPNP‚
ImmH complex can also be excluded on the basis of the
same argument. One possibility is that the NOE from N-7H
to C-2H is mediated by protons located adjacent to the 9dHX
ring. For example, the NH from Gly122 may be part of this
pathway. Another possibility is that a proton whose resonance
is near 12.5 ppm is responsible for this observation. Clearly,
more studies are required to investigate these possibilities.
Effect of Phosphate Binding to the hPNP‚ImmH Complex.
Crystallographic evidence establishes that the active site in
PNP is closed during catalysis by a flap that opens to bind
substrates (2, 11). When the flap is tightly sealed, new
contacts form between the ligand and protein residues within
the catalytic site (8, 9). Structural studies of ImmH and
substrates bound to bovine PNP are particularly compelling
in that six new H bonds form when PNP‚inosine‚SO4 is
compared to PNP‚ImmH‚PO4and nine H bonds relax when
PNP‚ImmH‚PO4is compared to product complexes. The data
presented here show that the charged N-4′ of ImmH and
phosphate ion are factors that drive this hydrogen-bonding
rearrangement during the tight-binding conformational change,
as shown by the appearance of new downfield proton
resonances upon addition of phosphate to the hPNP‚ImmH
binary complex (Figure 1). Molecular dynamic simulations
of nucleoside hydrolase (a related purine N-ribosyltrans-
ferase) with an iminoribitol transition-state analogue inhibitor
bound to the active site indicate that the active site becomes
more organized and corresponding protein motions decrease
when the transition-state characteristics of the inhibitor are
introduced into the simulation (40).
Conclusions. The interaction between hPNP and 9dHX
in the PNP‚ImmH‚PO4 complex is characterized by three
hydrogen bonds. The 2.8 Å H bond from the carbonyl
carboxamide of Asn243 to N-7H (Chart 2) forms as a result
of the elevated pKaand protonation at N-7 of inosine known
to occur at the transition state. The interaction causes a
chemical shift of ∼1 ppm downfield to 12.5 ppm. Since the
interaction between N-7 and the Asn243 carboxamide is
repulsive for the inosine complex, the shift to 12.5 ppm
indicates the formation of a favorable H bond, a stabilizing
force in the complex with the transition-state analogue. The
exocyclic O-6 of ImmH accepts a H bond from a water
bridge to Glu201, and this shortens from 3.4 Å in the
Michaelis complex to 2.8 Å in the complex PNP‚ImmH‚
PO4to form a second favorable H bond interaction. Glu201
forms a 2.8 Å H bond to N-1H, giving rise to the most
downfield shifted resonance for the interaction in the complex
at 14.9 ppm. This interaction anchors the purine ring and is
equivalent in the Michaelis complex and the PNP‚ImmH‚
PO4complex. Together these interactions are estimated to
contribute 10 kcal/mol toward forward catalysis by leaving
group activation. Interactions at N-7H and O-6 are specific
to the formation of the transition state, while that at N-1H
serves to anchor the substrate.
1. Stoeckler, J. D. (1984) in DeVelopments in Cancer Chemotherapy
(Glazer, R. J., Ed.) pp 35-60, CRC Press Inc., Boca Raton, FL.
2. Bzowska, A., Kulikowska, E., and Shugar, D. (2000) Purine
nucleoside phosphorylases: properties, functions, and clinical
aspects, Pharmacology Ther. 88, 349-425.
Catalytic Site Ligands in the Michaelis Analogue Complex
of PNP‚Inosine‚SO4and the Complex of PNP‚ImmH‚PO4As
Determined in 1.5 Å X-ray Crystal Structures (3)a
Active Site Contacts between Bovine PNP and
aDistances are in angstroms (error ∼0.2 Å). Human PNP is 82%
identical with bovine PNP, and all active site residues are conserved.
1986 Biochemistry, Vol. 43, No. 7, 2004
Deng et al.
3. Fedorov, A., Shi, W., Kicska, G., Fedorov, E., Tyler, P. C.,
Furneaux, R. H., Hanson, J. C., Gainsford, G. J., Larese, J. Z.,
Schramm, V. L., and Almo, S. C. (2001) Transition state structure
of purine nucleoside phosphorylase and principles of atomic
motion in enzymatic catalysis, Biochemistry 40, 853-60.
4. Miles, R. W., Tyler, P. C., Furneaux, R. H., Bagdassarian, C. K.,
and Schramm, V. L. (1998) One-third-the-sites transition-state
inhibitors for purine nucleoside phosphorylase, Biochemistry 37,
5. Kicska, G. A., Tyler, P. C., Evans, G. B., Furneaux, R. H.,
Schramm, V. L., and Kim, K. (2002) Purine-less death in
Plasmodium falciparum induced by immucillin-H, a transition state
analogue of purine nucleoside phosphorylase, J. Biol. Chem. 277,
6. Kicska, G. A., Tyler, P. C., Evans, G. B., Furneaux, R. H., Kim,
K., and Schramm, V. L. (2002) Transition state analogue inhibitors
of purine nucleoside phosphorylase from Plasmodium falciparum,
J. Biol. Chem. 277, 3219-25.
7. Li, C. M., Tyler, P. C., Furneaux, R. H., Kicska, G., Xu, Y.,
Grubmeyer, C., Girvin, M. E., and Schramm, V. L. (1999)
Transition-state analogs as inhibitors of human and malarial
hypoxanthine-guanine phosphoribosyltransferases, Nat. Struct.
Biol. 6, 582-7.
8. Shi, W., Li, C. M., Tyler, P. C., Furneaux, R. H., Cahill, S. M.,
Girvin, M. E., Grubmeyer, C., Schramm, V. L., and Almo, S. C.
(1999) The 2.0 A structure of malarial purine phosphoribosyl-
transferase in complex with a transition-state analogue inhibitor,
Biochemistry 38, 9872-80.
9. Shi, W., Li, C. M., Tyler, P. C., Furneaux, R. H., Grubmeyer, C.,
Schramm, V. L., and Almo, S. C. (1999) The 2.0 A structure of
human hypoxanthine-guanine phosphoribosyltransferase in com-
plex with a transition-state analog inhibitor, Nat. Struct. Biol. 6,
10. Sauve, A. A., Cahill, S. M., Zech, S. G., Basso, L. A., Lewan-
dowicz, A., Santos, D. S., Grubmeyer, C., Evans, G. B., Furneaux,
R. H., Tyler, P. C., McDermott, A., Girvin, M. E., and Schramm,
V. L. (2003) Ionic states of substrates and transition state analogues
at the catalytic sites of N-ribosyltransferases, Biochemistry 42,
11. Kicska, G. A., Tyler, P. C., Evans, G. B., Furneaux, R. H., Shi,
W., Fedorov, A., Lewandowicz, A., Cahill, S. M., Almo, S. C.,
and Schramm, V. L. (2002) Atomic dissection of the hydrogen
bond network for transition-state analogue binding to purine
nucleoside phosphorylase, Biochemistry 41, 14489-98.
12. Lin, J., Westler, W. M., Cleland, W. W., Markley, J. L., and Frey,
P. A. (1998) Fractionation factors and activation energies for
exchange of the low barrier hydrogen bonding proton in peptidyl
trifluoromethyl ketone complexes of chymotrypsin, Proc. Natl.
Acad. Sci. U.S.A. 95, 14664-8.
13. Frey, P. A., Whitt, S. A., and Tobin, J. B. (1994) A low-barrier
hydrogen bond in the catalytic triad of serine proteases, Science
264, 1927-30 (comment).
14. Markley, J. L., and Westler, W. M. (1996) Protonation-state
dependence of hydrogen bond strengths and exchange rates in a
serine protease catalytic triad: bovine chymotrypsinogen A,
Biochemistry 35, 11092-7.
15. Gerlt, J. A., Kreevoy, M. M., Cleland, W., and Frey, P. A. (1997)
Understanding enzymic catalysis: the importance of short, strong
hydrogen bonds, Chem. Biol. 4, 259-67.
16. Mildvan, A. S., Harris, T. K., and Abeygunawardana, C. (1999)
Nuclear magnetic resonance methods for the detection and study
of low-barrier hydrogen bonds on enzymes, Methods Enzymol.
17. Halkides, C. J., Wu, Y. Q., and Murray, C. J. (1996) A low-barrier
hydrogen bond in subtilisin: 1H and 15N NMR studies with
peptidyl trifluoromethyl ketones, Biochemistry 35, 15941-8.
18. Harris, T. K., Abeygunawardana, C., and Mildvan, A. S. (1997)
NMR studies of the role of hydrogen bonding in the mechanism
of triosephosphate isomerase, Biochemistry 36, 14661-75.
19. Zhao, Q., Abeygunawardana, C., Gittis, A. G., and Mildvan, A.
S. (1997) Hydrogen bonding at the active site of delta 5-3-
ketosteroid isomerase, Biochemistry 36, 14616-26.
20. Kintanar, A., Metzler, C. M., Metzler, D. E., and Scott, R. D.
(1991) NMR observation of exchangeable protons of pyridoxal
phosphate and histidine residues in cytosolic aspartate aminotrans-
ferase, J. Biol. Chem. 266, 17222-9.
21. Metzler, D. E., Metzler, C. M., Mollova, E. T., Scott, R. D.,
Tanase, S., Kogo, K., Higaki, T., and Morino, Y. (1994) NMR
studies of 1H resonances in the 10-18-ppm range for cytosolic
aspartate aminotransferase, J. Biol. Chem. 269, 28017-26.
22. Erion, M. D., Takabayashi, K., Smith, H. B., Kessi, J., Wagner,
S., Honger, S., Shames, S. L., and Ealick, S. E. (1997) Purine
nucleoside phosphorylase. 1. Structure-function studies, Bio-
chemistry 36, 11725-34.
23. Stoeckler, J. D., Poirot, A. F., Smith, R. M., Parks, R. E., Jr.,
Ealick, S. E., Takabayashi, K., and Erion, M. D. (1997) Purine
nucleoside phosphorylase. 3. Reversal of purine base specificity
by site-directed mutagenesis, Biochemistry 36, 11749-56.
24. Evans, G. B., Furneaux, R. H., Gainsford, G. J., Hanson, J. C.,
Kicska, G. A., Sauve, A. A., Schramm, V. L., and Tyler, P. C.
(2003) 8-Aza-immucillins as transition-state analogue inhibitors
of purine nucleoside phosphorylase and nucleoside hydrolases,
J. Med. Chem. 46, 155-60.
25. Evans, G. B., Furneaux, R. H., Hutchison, T. L., Kezar, H. S.,
Morris, P. E., Jr., Schramm, V. L., and Tyler, P. C. (2001)
Addition of lithiated 9-deazapurine derivatives to a carbohydrate
cyclic imine: convergent synthesis of the aza-C-nucleoside
immucillins, J. Org. Chem. 66, 5723-30.
26. Wagner, G., and Wu ¨thrich, K. (1979) Truncated Driven Nuclear
Overhauser Effect(TOE). A New Technique for Studies of
diffusion, J. Magn. Reson. 33, 675-680.
27. Clore, G. M., and Gronenborn, A. M. (1982) Theory and
application of the transferred nuclear overhauser effect to the study
of the conformations of small ligands bound to proteins, J. Magn.
Reson. 48, 402-417.
28. Haselhorst, T., Espinosa, J. F., Jimenez-Barbero, J., Sokolowski,
T., Kosma, P., Brade, H., Brade, L., and Peters, T. (1999) NMR
experiments reveal distinct antibody-bound conformations of a
synthetic disaccharide representing a general structural element
of bacterial lipopolysaccharide epitopes, Biochemistry 38, 6449-
29. Mayer, M., and Meyer, B. (1999) Characterization of ligand
binding by saturation transfer difference NMR spectroscopy,
Angew. Chem. Int. Ed. 38, 1784-1788.
30. Hwang, T. L., and Shaka, A. J. (1995) Water suppression that
works. Excitation sculpting using arbitrary wave-forms and pulsed-
field gradients, J. Magn. Reson., A 112, 275-279.
31. Wang, F., Miles, R. W., Kicska, G., Nieves, E., Schramm, V. L.,
and Angeletti, R. H. (2000) Immucillin-H binding to purine
nucleoside phosphorylase reduces dynamic solvent exchange,
Protein Sci. 9, 1660-8.
32. Kline, P. C., and Schramm, V. L. (1993) Purine nucleoside
phosphorylase. Catalytic mechanism and transition-state analysis
of the arsenolysis reaction, Biochemistry 32, 13212-9.
33. Kline, P. C., and Schramm, V. L. (1995) Pre-steady-state transition-
state analysis of the hydrolytic reaction catalyzed by purine
nucleoside phosphorylase, Biochemistry 34, 1153-62.
34. Schramm, V. L., and Shi, W. (2001) Atomic motion in enzymatic
reaction coordinates, Curr. Opin. Struct. Biol. 11, 657-65.
35. Wei, Y., and McDermott, A. E. (1999) in Modeling NMR Chemical
Shifts: gaining insights into structure and enVironment (Facelli,
J. C., and deDios, A. C., Eds.) pp 177-193, Oxford University
Press, Oxford, U.K..
36. Neal, S., Nip, A. M., Zhang, H., and Wishart, D. S. (2003) Rapid
and accurate calculation of protein 1H, 13C and 15N chemical
shifts, J. Biomol. NMR 26, 215-40.
37. Johnson, C. E., and Bovey, F. A. (1958) Calculation of nuclear
magnetic resonance spectra of aromatic hydrocarbon, J. Chem.
Phys. 29, 1012-1014.
38. Loh, S. N., and Markley, J. L. (1994) Hydrogen bonding in proteins
as studied by amide hydrogen D/H fractionation factors: applica-
tion to staphylococcal nuclease, Biochemistry 33, 1029-36.
39. Kreevoy, M. M., and Liang, T. M. (1980) Structures and isotopic
fractionation factors of complexes, A1HA2-, J. Am. Chem. Soc.
40. Mazumder, D., Kahn, K., and Bruice, T. C. (2002) Computer
simulations of trypanosomal nucleoside hydrolase: determination
of the protonation state of the bound transition-state analogue, J.
Am. Chem. Soc. 124, 8825-33.
1H-1H Overhauser Effects in the presence of spin
Downfield Proton Resonances in PNP‚Immucillin-H
Biochemistry, Vol. 43, No. 7, 2004 1987