Stereospecificity, substrate, and inhibitory properties of nucleoside diphosphate analogs for creatine and pyruvate kinases

Article abstract of DOI:10.1016/j.bioorg.2008.03.001

Antiviral α-P-borano substituted NTPs are promising chain terminators targeting HIV reverse transcriptase (RT). Activation of antiviral nucleoside diphosphates (NDPs) to NTPs may be carried out by pyruvate kinase (PK) and creatine kinase (CK). Herein, are presented the effects of nucleobase, ribose, and α-phosphate substitutions on substrate specificities of CK and PK. Both enzymes showed two binding modes and negative cooperativity with respect to substrate binding. The stereospecificity and inhibition of ADP phosphorylation by α-P-borano substituted NDP (NDPαB) stereoisomers were also investigated. The Sp-ADPαB isomer was a 70-fold better substrate for CK than the Rp isomer, whereas PK preferred the Rp isomer of NDPαBs. For CK, the Sp-ADPαB isomer was a competitive inhibitor; for PK, the Rp-ADPαB isomer was a poor competitive inhibitor and the Sp-ADPαB isomer was a poor non-competitive inhibitor. Taken together, these data suggest that, although the Rp-NDPαB isomer would be minimally phosphorylated by CK or PK, it should not inhibit either enzyme.

Full text of DOI:10.1016/j.bioorg.2008.03.001

Bioorganic Chemistry 36 (2008) 169–177
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Stereospecificity, substrate, and inhibitory properties of nucleoside diphosphate
analogs for creatine and pyruvate kinases
1
*
Charlotta K. Wennefors, Mikhail I. Dobrikov, Zhihong Xu, Ping Li , Barbara Ramsay Shaw
Department of Chemistry, French Family Science Center, Duke University, 124 Science Drive, Durham, NC 27708-0354, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 January 2008
Available online 22 April 2008
Antiviral a-P-borano substituted NTPs are promising chain terminators targeting HIV reverse transcrip-
tase (RT). Activation of antiviral nucleoside diphosphates (NDPs) to NTPs may be carried out by pyruvate
kinase (PK) and creatine kinase (CK). Herein, are presented the effects of nucleobase, ribose, and a-phos-
phate substitutions on substrate specificities of CK and PK. Both enzymes showed two binding modes and
negative cooperativity with respect to substrate binding. The stereospecificity and inhibition of ADP
phosphorylation by a-P-borano substituted NDP (NDPaB) stereoisomers were also investigated. The Sp-
ADPaB isomer was a 70-fold better substrate for CK than the Rp isomer, whereas PK preferred the Rp iso-
mer of NDPaBs. For CK, the Sp-ADPaB isomer was a competitive inhibitor; for PK, the Rp-ADPaB isomer
was a poor competitive inhibitor and the Sp-ADPaB isomer was a poor non-competitive inhibitor. Taken
together, these data suggest that, although the Rp-NDPaB isomer would be minimally phosphorylated by
CK or PK, it should not inhibit either enzyme.
Keywords:
Creatine kinase
Pyruvate kinase
a-P-borano phosphate
Phosphonate
Negative cooperativity
TSAC
Kinetics
Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction
viral DNA, the Rp isomers of a-P-borano substituted AZT-TP and
d4T-TP have demonstrated increased stability toward the ATP-
Nucleoside analogs targeting retroviral reverse transcriptases
(RTs) are an important class of antiviral agents [1–4]. The need
for these analogs to undergo metabolic conversion to be biologi-
cally active [1,4,5] has lead to problems such as poor phosphoryla-
tion by intracellular kinases [3,5,6] and undesired metabolic effects
[7,8]. Another disadvantage of existing nucleoside analogs is the
eventual emergence of drug resistance [4,9–12]. It has been dem-
onstrated that a-P-borano modifications in clinically relevant dide-
oxy NTPs (ddNTPs) such as AZT, d4T [2], ddA [4,10], and acycloT
[13] improve their incorporation into viral DNA by wildtype HIV-
1 RT [2], and even more so by mutant drug resistant HIV-1 RTs
[4,10] and MMLV RT [13,14]. Furthermore, after incorporation into
dependent repair mechanism that contributes to drug resistance
[2,4,10]. Thus a-P-borano nucleotide analogs are promising antivi-
ral candidates for selectively targeting mutant RTs.
In boranophosphates analogs, one of the non-bridging oxygens
on the a-phosphate of nucleoside mono- (NMP), di- (NDP), or tri-
ꢀ
phosphates (NTPs) is replaced by a borane ꢀBH3 group (Fig. 1)
[15–26].
The borane group is isoelectronic with oxygen in normal phos-
phate, isolobal with sulfur in phosphorothioates, and isosteric with
the methyl group in methylphosphonates [15–24]. Whereas the
boranophosphates share the same net charge and geometry about
the phosphorus as unmodified phosphates, the boranophosphates
have a longer P–B bond than the P–O bond in normal phosphate
(1.91 versus 1.51 Å), a reduced tendency to coordinate metal ions
or form H bonds, and an altered polarity [27]. The lower electro-
negativity of boron (2.04) than the oxygen (3.44) in a phosphodies-
ter bond [27,28] gives reason to the observation that a-P-borano
phosphate groups accelerate the incorporation of chain termina-
tors into viral DNA [2,4,10,14,26,29], resulting in the increased po-
tency of these drugs. Ultimately, however, the potency of these
nucleotide analogs as antiviral drugs is highly dependent on their
phosphorylation by host cellular kinases. To circumvent one or
two steps involved in the activation of ddNTPaBs, prodrugs of
ddNMPaB and ddNDPaB were recently developed [13,30,31]. De-
spite these efforts, the last step in the phosphorylation of
Abbreviations: CK, creatine kinase; PCr, creatine phosphate; PK, pyruvate kinase;
PeP, phosphoenolpyruvate; NDPK, nucleoside diphosphate kinase; TSAC, transition
state analog complex; TS, transition state; N, nucleoside; NDP, nucleoside diphos-
phate; NTP, nucleoside triphosphate; dNTP, deoxynucleoside triphosphate; ddNTP,
20,30-dideoxynucleoside triphosphate; ddNTPaB, a-P-borano substituted 20,30-dide-
oxynucleoside triphosphate; ATPaB, a-P-borano substituted ATP; ATPaS, a-P-thio
substituted ATP; AZT, 30-deoxy-30-azidothymidine; d4T, 20,30-dideoxythymidine;
ddA, 20,30-dideoxyadenosine, acycloT, 1-(2-hydroxyethoxymethyl)-thymine; HIV,
human immunodeficiency virus; NRTI, nucleoside reverse transcriptase inhibitor;
HPLC, high performance liquid chromatography.
* Corresponding author. Fax: +1 919 660 1605.
E-mail address: barbara.r.shaw@duke.edu (B.R. Shaw).
Present address: Department of Chemistry, Massachusetts Institute of Technol-
1
ogy, 77 Massachusetts Avenue, Cambridge, MA 02139, USA.
0045-2068/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2008.03.001

170
C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
Fig. 1. Nucleoside diphosphate (NDP) ribose and a-P modifications. NDP: R = OH, R0 = OH; 20-dNDP: R = H, R0 = OH; 30-dNDP: R = OH, R0 = H; 20,30-ddNDP: R = H, R0 = H. a-P
modifications: a,b-AMP(CH2)P (bottom left). The Rp-ADPaB isomer (top right), and the Sp-ADPaB isomer (bottom right).
nucleoside analog diphosphates to their respective triphosphates
remains largely unexplored [5].
complex (TSAC) was evaluated to determine the affinity of the
NDP substrate for the transition state (Fig. 2). For both CK and
PK, the TSAC is formed when a nitrate ion, NO3 occupies the posi-
ꢀ
Previous studies revealed that nucleoside diphosphate kinase
(NDPK), a major enzyme for NDP phosphorylation in cells, phos-
phorylates AZT-DP, d4T-DP, and ddC-DP poorly [1,3,5,32]. Krishnan
and co-workers evaluated the roles of creatine kinase (CK), 3-phos-
phoglycerate kinase (PGK), and pyruvate kinase (PK) in NDP analog
phosphorylation and proposed that the specificity of the kinases
toward the antiviral ddNDPs is dependent on both the configura-
tion of the analog (L or D) and the presence of a 30-hydroxyl group
in the sugar moiety [5]. Furthermore, the researchers suggest that
CK and PK may be responsible for the last phosphorylation step of
20,30-dideoxy- and acyclo-nucleoside diphosphates in vivo [5].
Previous studies by Meyer and co-workers indicated that a-P-
borano substitutions on AZT-DP and d4T-DP enhance efficiency
of phosphorylation by NDPK by 10-fold [2]. Although the catalytic
efficiency was enhanced by the borano, it was still 1000-fold less
than for the parent NDPs [2], suggesting that other kinases, per-
haps CK or PK, might be more efficient enzymes for the phosphor-
ylation of ddNDP to ddNTP. CK and PK have opposite
stereospecificity for binding and phosphorylation of the Rp and
Sp isomers of a-P-thio substituted NDPs [33]. It is therefore likely
that the Rp- and Sp-stereoisomers of a-P-borano substituted NDPs
will be recognized distinctly by these enzymes. Thus, it is of inter-
est to study the effects of the NDPaB stereoisomers on phosphory-
lation and substrate specificity of CK and PK.
tion of the transferable phosphate during the transition state of the
enzyme catalyzed reaction as demonstrated previously for ADP
[37–42].
Herein, we report the specificity of CK and PK toward nine dif-
ferent nucleobase, ribose, and a-phosphate substituted NDPs and
the effect of the a-P-borano modification on the phosphorylation
of NDPs (refer to Fig. 1). Substrate properties, inhibitory properties,
and stereochemical effects of a-P-borano nucleoside diphosphates
for CK and PK are also presented.
2. Materials and methods
2.1. Materials
Creatine, NaNO3, Hepes, Bicine, KCl, sodium oxalate, glycerol,
tryptophan methyl ester hydrochloride, creatine phosphate and
nucleoside diphosphates (including AMP(CH2)P) and triphosphates
were obtained from Sigma–Aldrich. KOH was obtained from Fisher.
MgCl2 was obtained from Ambion. Rabbit muscle creatine kinase
and rabbit muscle pyruvate kinase were purchased as desiccated
powders from Roche. ADPaB and GDPaB were synthesized, puri-
fied, and the isomers were separated by HPLC as published previ-
ously [43].
CK is endogenous to muscle tissues that require large energy
fluxes [34,35] as well as to non-muscle cells [36]. The dimeric en-
zyme catalyzes the reversible reaction between ADP and phospho-
creatine or ATP and creatine. PK is a tetrameric enzyme involved in
the glycolytic pathway by catalyzing the transfer of a phosphate
group from phosphoenolpyruvate to ADP, to yield pyruvate and
ATP. The kinetic schemes for both CK and PK are initiated by direct
binding of ADP and creatine phosphate (for CK) or phosphoenol-
pyruvate (for PK) to the enzyme, subsequent formation of the tran-
sition state involving a conformational change of the enzyme,
followed by a rate limiting phosphate transfer reaction, and finally
product release (Scheme 1).
2.2. Stock solutions
Concentrations of NDPs in 70 mM Hepes (pH 7) were deter-
mined by Cary UV–vis spectrophotometer and the purity was
determined by Varian HPLC. For CK studies the direct binding
(DB) buffer [38] was adjusted to pH 8.3 and contained 50 mM Bi-
cine, 50 mM KCl, and 5 mM MgCl2; the transition state analog com-
plex (TSAC) binding buffer [38] was adjusted to pH 8.3 and
contained 50 mM Bicine, 5 mM MgCl2, 50 mM NaNO3, and
20 mM creatine. For PK studies the direct binding buffer was ad-
justed to pH 7.5 and contained 50 mM Hepes, 5% glycerol, 5 mM
MgCl2, 2 mM sodium oxalate, and 100 mM KCl. The TSAC binding
buffer was adjusted to pH 7.5 and contained 50 mM Hepes, 5%
glycerol, 5 mM MgCl2, 2 mM sodium oxalate, 50 mM KCl, and
In the present study, the binding affinities of the substrate ana-
logs used in this study were determined by a fluorescence quench-
ing assay using equilibrium titration. A transition state analog
Scheme 1. CK and PK kinetic scheme. ADP and phosphate donor binds to the enzyme. A conformational change of the enzyme forming the transition state is followed by the
rate limiting phosphate transfer, followed by product release. E = CK or PK, PX = creatine phosphate (for CK) or phosphoenolpyruvate (for PK), X = creatine (for CK) or pyruvate
(for PK), ADP = adenosine diphosphate, and ATP = adenosine triphosphate.

C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
171
DDGb ¼ ꢀRT lnðKd1 ADP=Kd1 NDP
Þ
ð3Þ
where R is 8.314 J/mol K, T is 298 K, and Kd1 ADP and Kd1 NDP are the
dissociation constants associated with the first binding mode of
ADP and the corresponding NDP analog, respectively.
2.4. Steady-state kinetics
ꢀ
Initial NDP phosphorylation reactions were performed to deter-
mine enzyme concentrations and reaction times to be used in the
steady-state kinetic experiments. Enzyme concentrations and reac-
tion times were varied while keeping all other components con-
stant. Substrate concentrations were kept at 200 lM in a reaction
mixture containing 20 mM creatine phosphate, 50 mM Bicine,
5 mM MgCl2, and 50 mM KCl, buffered to pH 8.3. The concentration
of enzyme was varied from 0 to 200 nM. The enzyme was added to
the reaction mixture to start the reaction. The reaction was
quenched in a 100 °C water bath with subsequent filtering of the
reaction mixture using Microcon centrifugal filter devices (NMWL:
30,000). The amount of conversion of NDP to NTP was determined
using HPLC. Complete separation of NDPs and NTPs was observed
on a C18 column using a 3–15% methanol gradient in 50 mM TEAA
(triethyl ammonium acetate buffer, buffered to pH 6.8). Next, the
mixture was allowed to react for different time points while keep-
ing [enzyme] and [NDP] constant. The [enzyme] and reaction time
was chosen (below 20% substrate conversion) where the product
formation was linearly dependent so the product amount directly
correlates with the steady-state rate of product formation.
Fig. 2. The nitrate transition state analog complex (TSAC) is formed when NO3
occupies the position of the transferable phosphate, during the transition state of
the enzyme catalyzed reaction. Three species; X (creatine or pyruvate), nitrate (r-
ight), and ADP are interacting through hydrogen bonding, mimicking the phos-
phoryl transition state (one complex) (left).
50 mM KNO3. Fresh CK and PK stock solutions were prepared
immediately prior to each kinetic experiment and kept on ice
throughout the experiment. The enzyme concentrations were
determined using the absorbance at 280 nm with extinction coef-
ficients of 0.88 for 1 mg/mL CK (MW = 81,000 g/mol) [44], and
0.54 for 1 mg/mL PK (MW = 237,000 g/mol) [45]. The active enzyme
concentration was determined using the specific activity of
200 lmol of ATP converted to ADP per min per mg of CK (forward
reaction) at pH 9.0 and 30 °C [37,44], and 260 lmol pyruvate per
min per mg of PK at pH 7.0 and 25 °C [46,47].
2.3. Affinity study
The affinities of the substrate analogs for direct and TSAC bind-
ing were determined with a fluorescence quenching assay using
equilibrium titration. The solution of 1 lM CK, or 0.2 lM PK (for
guanosine derivatives) or 0.7 lM PK (for adenosine derivatives)
in ‘‘direct binding buffer” or ‘‘TSAC binding buffer” was titrated
with NDP as in [38]. After each subsequent aliquot (0.25–10 lL)
of stock NDP (or NTP) was added to the 1-mL buffer solution, the
solution was allowed to equilibrate for approximately 1 min before
fluorescence was measured in triplicate. All fluorescence measure-
ments were made at 25 1 °C. Excitation of tryptophan residues in
CK or PK was carried out at 295 2 nm and emission was mea-
sured at 340 3 nm. A 4 lM solution of tryptophan methyl ester
hydrochloride in deionized water was titrated with NDP for correc-
tion of collisional quenching and dilution.
For kinetic analysis the substrate concentration was then varied
between 0 and 3 mM while keeping [CK] and reaction time con-
stant. The steady-state kinetic data were fit to the Lineweaver–
Burk equation (Eq. (4)), using SIGMAPLOT 9.0, to determine values
for kcat and Km which then were used to determine the efficiency of
phosphorylation for each of the substrates,
1=kcat ðappÞ ¼ ðKm=kcatÞð1=½SꢁÞ þ 1=kcat
ð4Þ
where kcat is the rate constant for the phosphorylation, [S] is the
substrate concentration, and Km is the Michaelis–Menten dissocia-
tion constant. Control experiments were performed to ensure
NDPaB stability throughout enzyme quenching.
The accumulated fluorescence values were converted to %
quenching according to Q = 100(F0 ꢀ F/F0), where F0 is the initial
fluorescence measured, and F is the fluorescence measured at a
specific point. The readings were corrected (Q0) for inner filter ef-
fects and Q0max was determined by fitting the highest concentration
data to the following Lineweaver–Burk equation,
2.5. Inhibition study
For CK, 0–600 lM of Rp- or Sp-ADPaB isomer were added to the
reaction mixture used for ADP (0–2 mM) phosphorylation (50 mM
Bicine, 5 mM MgCl2, 50 mM KCl, and 20 mM creatine phosphate, at
pH 8.3). The reactions were quenched after 2 min in a 100 °C water
bath and the reaction mixture was subsequently filtered using
Microcon centrifugal filter devices (NMWL: 30,000). The product
formation was determined with HPLC as described above.
For PK, 0–2 mM of the Rp- or Sp-ADPaB isomers were added to
the reaction mixture used for ADP (0–3 mM) phosphorylation
(100 mM Hepes, 125 mM KCl, 3 mM MnCl2, 3% glycerol, and
2.5 mM phosphoenolpyruvate at pH 7.5). Here, PK was added at
room temperature to initiate the reaction and quenched after
30 s by adding 10 lL of 10 mM 20,40-dinitrophenylhydrazine
(DNPH) in 3 M HCl. After incubating for 10 min at 37 °C, 40 lL of
0.1 M EDTA in 3 M KOH was added. The UV–vis absorbance at
520 nm was determined after 10 min incubation at room
temperature.
1=Q0 ¼ ðKd=Qm0 axÞð1=½SꢁÞ þ 1=Qm0 ax
ð1Þ
where Q0 is the % quenching (corrected for inner filter effects), Q0max
is the maximum quenching, [S] is the substrate concentration, and
Kd is the equilibrium dissociation constant for the substrate.
Each set of data for the fluorescence quenching experiment was
then analyzed using the following Hill equation,
log½Q0=ðQm0 ax ꢀ Q0Þꢁ ¼ n log½Sꢁfree ꢀ log Kd
ð2Þ
where Q0 is the % quenching (corrected for inner filter effects) at a
specific point, Q0max is the maximum quenching, [S]free is the free
*
substrate concentration ([S]free = [S]total ꢀ [CK]total Q0/Q0max), n is
the Hill coefficient, and Kd is the equilibrium dissociation constant
for the substrate. By fitting the data to Eq. (2), the equilibrium dis-
sociation constants, Kd1 and Kd2, and the Hill coefficients, n1 and n2,
were determined.
The data was analyzed on double reciprocal plots using SIGMA-
PLOT 9.0. The inhibitory constant, Ki, was calculated using the
Lineweaver–Burk equation for competitive inhibition
The difference in the binding free energy change (DDGb) be-
tween the different nucleoside diphosphate analogs was calculated
using the following equation,
1=kcat ¼ Km=kcat½ð1 þ ½Iꢁ=KiÞð1=½SꢁÞꢁ þ 1=kcat
ð5Þ
and non-competitive inhibition,

172
C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
1=kcat ¼ Km=kcat½ð1 þ ½Iꢁ=KiÞð1=½SꢁÞꢁ þ ½ð1 þ ½Iꢁ=KiÞð1=kcatÞꢁ
ð6Þ
where kcat is the rate constant for the phosphorylation, [S] is the
substrate (ADP) concentration, [I] is the inhibitor concentration
(Rp-ADPaB or Sp-ADPaB), Km is the Michaelis–Menten dissociation
constant, and Ki is the inhibitory constant.
3. Results
3.1. Affinity of NDPs and NTPs for CK and PK
Equilibrium dissociation constants of direct binding and TSAC
binding were determined for a wide range of substrates by moni-
toring the fluorescence quenching of the tryptophan residue near
the active site (Trp227 for CK, Trp157 for PK) upon substrate bind-
ing. The Hill plots show that at very low and very high substrate
concentrations the slopes (n1, n2) approach 1.0. Furthermore, all
NDP and NTP substrates for both CK and PK (including PK-NTP
binding results published previously [48]) have a slope corre-
sponding to the Hill coefficient, nH, of <1.0 (0.5–0.8) between the
points corresponding to 0:3Q0max and 0:7Qm0 ax. By fitting each linear
part of the Hill plot to the Hill equation (2), values for Hill coeffi-
cients (n1 and n2) and dissociation constants (Kd1 and Kd2) were
calculated.
In the sample graph shown (Fig. 3A), Kd1 and Kd2 for the Sp-AD-
PaB isomer were calculated as 35 lM and 525 lM for direct bind-
ing, and 275 lM and 1380 lM for binding to the TSAC, where
ꢀ
ꢀ
NO3 is replacing and mimicking the PO3 in the TS. A collection
of the first binding modes for NDPs bound to CK in the TSAC are
compiled in Fig. 3B and all NDP substrate analog binding constants
in Table 1.
Similar binding behavior was observed with PK. PK is a tetra-
meric enzyme that potentially could have four binding modes.
Here, we observed the two first binding modes (Fig. 4A). Kd1 and
Kd2 for the Rp-GDPaB isomer (Fig 4A) were calculated as 135 lM
and 660 lM for direct binding, and 107 lM and 380 lM for binding
to the PK-TSAC. The NDP substrate analog first binding modes are
compiled in Fig. 4B and all binding constants are summarized in
Table 2.
All dissociation constants (Kd1, Kd2) and differences in free en-
ergy change values, DDGb, can be found in Tables 1–3. Complete ta-
bles including maximum quenching values (Qm0 ax
coefficients (n1, n2) can be found in Supplementary data.
Þ
and Hill
3.2. Steady-state kinetics of phosphorylation of ADP and ADPaB
isomers by CK and PK
Fig. 3. (A) Hill plot of the Sp-ADPaB isomer for (ꢂ) direct binding and (s) TSAC
binding with CK. (B) First binding mode of NDPs ((ꢂ) ADP, (D) a,b-AMP(CH2)P, (N)
GDP, (s) Sp-ADPaB, (j) CDP, (h) Rp-ADPaB) bound to CK in TSAC. One micromolar
CK in direct binding buffer (50 mM Bicine, 5 mM MgCl2, and 50 mM KCl at pH 8.3)
or TSAC binding buffer (50 mM Bicine, 5 mM MgCl2, 50 mM NaNO3, and 20 mM
creatine at pH 8.3) was titrated with 0–2 mM NDP at 25 °C. Experiments were
performed in triplicate.
In an effort to better understand the mechanism of phosphory-
lation of ADP and the effect of stereoisomers of ADPaB, the phos-
phorylation of NDPs was studied using steady-state kinetics.
From the Lineweaver–Burk plots of the phosphorylation data ob-
tained for ADP and the stereoisomers of ADPaB (Fig. 5), values
for kcat and Km, were obtained and then used to determine the effi-
ciency of phosphorylation for each of the substrates (Table 4).
The Sp-ADPaB isomer has a lower Km than ADP (KmADPaB = 8 lM,
Km ADP = 67 lM) and a lower rate of phosphorylation (kcat Sp-AD-
PaB = 0.03 sꢀ1, kcat ADP = 370 sꢀ1) (Table 4). The Rp-ADPaB isomer
has a larger Km (1000 lM) and a lower kcat (0.06 sꢀ1) than ADP.
While a similar study was carried out using PK, no measurable
phosphorylation was observed for either isomer of ADPaB using
PK.
an inhibition assay was performed to determine if either stereoiso-
mer has an inhibitory effect on the phosphorylation of normal ADP
(Fig. 6).
For CK, the Sp-ADPaB isomer is a strong competitive inhibitor of
the ADP reaction (Km ADP = 67 lM), as all extrapolations of these
plots intersected on the y-axis with a Ki of 49 lM. However, for
the Rp-ADPaB isomer, no CK inhibition is observed. In contrast,
for PK, the Rp-ADPaB isomer is a weak competitive inhibitor
(Ki = 450 lM) of the ADP phosphorylation, whereas the Sp-ADPaB
isomer is a weak non-competitive inhibitor (Ki = 2.7 mM). The effi-
ciency of ADP phosphorylation by PK, here in the presence of Mn2+
(Km ADP = 240 lM, kcat = 150 sꢀ1), is comparable to that reported in
3.3. ADPaB inhibition of ADP phosphorylation by CK and PK
the literature in the presence of Mg2+ (Km ADP = 390 lM, kcat
=
Since the initial experiments showed that the Rp and the Sp
isomer of ADPaB are only minimally phosphorylated by CK or PK,
288 sꢀ1) [49].

C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
173
Table 1
Creatine kinase—NDP equilibrium dissociation constants (Kd1, Kd2), and difference in free energy levels (DDGb) for directa and TSACb binding
Direct bindinga
TSAC bindingb
Kd1 (lM)
Kd2 (lM)
DDGb (kJ/mol)
Kd1 (lM)
Kd2 (lM)
DDGb (kJ/mol)
ADP
89 22
31
181 16
331
0
11
38
4
51 10
100
0
AMP(CH2)Pc
Sp-ADPaB
Rp-ADPaB
20-dADP
GDP
ꢀ2.6
ꢀ2.4
+2.9
+1.3
+1.5
+1.4
+1.3
+2.7
+3.1
+8.3
+10.7
+5.7
+6.8
+7.8
+7.6
+5.8
34
2
417 152
2928 436
288 61
571 179
615 110
1137 389
1800 424
315 56
816 148
111 13
170 20
258 62
233 83
115 20
1500 170
5250 1280
145 53
488 45
500 88
1601 682
1220 139
291 50
149 19
163 53
20-dGDP
CDP
20-dCDP
155
5
150 53
261 38
All experiments were performed in triplicate and reported as averages and standard deviations.
a
One micromolar CK in 50 mM Bicine, 5 mM MgCl2, and 50 mM KCl at pH 8.3 was titrated with 0–2 mM NDP at 25 °C.
One micromolar CK in 50 mM Bicine, 5 mM MgCl2, 50 mM NaNO3, and 20 mM creatine at pH 8.3 was titrated with 0–2 mM NDP at 25 °C.
One trial only.
b
c
4. Discussion
Structure–activity relationships for creatine kinase and pyru-
vate kinase (Fig. 7) were established based on the equilibrium dis-
sociation constants of direct binding and nitrate TSAC (Fig. 2)
binding for a wide range of nucleobase, ribose, and a-phosphate
modified NDPs and NTPs. All substrate analogs have two binding
modes (Figs. 3 and 4), with Hill coefficients approaching 1.0 at very
high and very low substrate concentrations, and an intermediate
negatively cooperative step with Hill coefficients ranging from
0.5 to 0.8. The distinct dissociation constants for each binding
mode indicates allosteric behavior of the enzyme [50–52]. Both
CK and PK exhibit a 2- to 10-fold lower binding affinity of the sec-
ond substrate molecule, particularly for unnatural substrates, as
compared to the binding affinity of the first substrate molecule
(Kd1 < Kd2). This observation is consistent with the proposition that
CK and PK both display negative cooperativity with respect to sub-
strate binding [50–52]. It is hypothesized that the binding of the
first substrate molecule to the CK dimer, or PK tetramer, changes
the conformation of one subunit of the enzyme which in turn af-
fects the stability of the neighboring subunit conformation and de-
creases the affinity of the second substrate molecule for the vacant
substrate binding site.
Whereas several studies have reported allosteric behavior of the
subunits in CK, with results showing negative cooperativity or non-
identical active sites [39,53–57], other studies suggest the contrary
and support a non-cooperative binding behavior [40]. The fluores-
cence quenching approach used in our study to determine the affin-
ity for a wide range of NDPs was previously used by Borders et al. to
determine the affinity for each of the components involved in the
binding of ADP to the nitrate TSAC [38]. The intrinsic dissociation
constants determined by Borders et al. for direct binding
(Kd = 135 lM) and TSAC binding (Kd = 14 lM) of ADP to CK correlate
wellwiththe dissociation constants for ADPdetermined in ourstudy
forthefirstbindingmode(Kd1 DB = 89 lM, Kd1 TSAC = 11 lM)(Table1).
Our studies also show that PK demonstrates a negatively coop-
erative binding behavior. Although this is a tetrameric enzyme and
thus has four potential binding modes, only the first two binding
modes were identified for PK.
Fig. 7A and B display the structure–activity relationships for CK
and PK, respectively, and the binding interactions between the en-
zymes and ADP in their corresponding TSACs. After determining
dissociation constants for a wide range of substrates a quantitative
characterization of each enzyme–substrate interaction and an
evaluation of its importance for binding affinity could be made.
The differences in free energy change (DDGb) upon substitutions
on base, sugar, and a-phosphate from the natural substrates
(ADP or ATP) are noted in Fig. 7.
Fig. 4. (A) Hill plot of the Rp-GDPaB isomer for (r) direct binding and (e) TSAC
binding with PK. (B) First binding mode of NDPs ((D) a,b-AMP(CH2)P, (.) 30-dADP,
(r) GDP, (N) 30-dGDP, (e) Rp-GDPaB, (s) Sp-GDPaB) bound to PK in TSAC. PK (0.2–
0.7 lM) in direct binding buffer (50 mM Hepes, 5% glycerol, 5 mM MgCl2, 2 mM
sodium oxalate, and 100 mM KCl at pH 7.5) or TSAC binding buffer (50 mM Hepes,
5% glycerol, 5 mM MgCl2, 2 mM sodium oxalate, and 50 mM KNO3 at pH 7.5) was
titrated with 0–2 mM NDP at 25 °C. Experiments were performed in triplicate.

174
C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
Table 2
Pyruvate kinase—NDP equilibrium dissociation constants (Kd1, Kd2), and difference in free energy levels (DDGb) for directa and TSACb binding
Direct bindinga
TSAC bindingb
Kd1 (lM)
Kd2 (lM)
DDGb (kJ/mol)
Kd1 (lM)
Kd2 (lM)
DDGb (kJ/mol)
ADP
5.7 1.2
5.6 1.4
68 6.5
46 11
37 4.5
789 105
871 132
555 93
635 102
611 81
151 18
604 59
633 85
403 58
315 46
217 39
0
2.3 0.4
0.4 0.1
632 49
182 33
721 75
737 68
198 35
141 41
427 74
361 65
363 47
471 13
381 44
0
AMP(CH2)P
Rp-ADPaB
Sp-ADPaB
20-dADP
30-dADP
GDP
Rp-GDPaB
Sp-GDPaB
20-dGDP
30-dGDP
ꢀ0.04
+6.1
+5.2
+4.6
+4.4
+3.3
+7.3
+7.8
+3.9
+3.6
ꢀ4.3
+6.8
+8.5
+6.3
+5.5
+3.9
+9.4
+10.8
+7.4
+8.0
36
6
79 18
29
21
11
4
3
3
6
33
22
5
4
108 12
132 15
102
182 42
45 10
27
24
5
4
57
8
All experiments were performed in triplicate and reported as averages and standard deviations.
a
0.2 (G derivatives) or 0.7 lM (A derivatives) PK in 50 mM Hepes, 5% glycerol, 5 mM MgCl2, 2 mM sodium oxalate, and 100 mM KCl, at pH 7.5 was titrated with 0–2 mM
NDP at 25 °C.
b
0.2 (G derivatives) or 0.7 lM (A derivatives) PK in 50 mM Hepes, 5% glycerol, 5 mM MgCl2, 2 mM sodium oxalate, and 50 mM KNO3, at pH 7.5) was titrated with 0–2 mM
NDP at 25 °C.
Table 3
Table 4
Creatine kinase—NTP direct binding equilibrium dissociation constants (Kd1, Kd2), and
Steady-state kinetic and inhibitory constant values for the phosphorylation of NDPs
difference in free energy levels (DDGb)a
by CK and PKa
Direct bindinga
Km (lM)
kcat (sꢀ1
)
kcat/Km (Mꢀ1 sꢀ1
)
Ki (lM)
Kd1 (lM)
Kd2 (lM)
DD Gb (kJ/mol)
Creatine kinase
ADP
Sp-ADPaB
Rp-ADPaB
67 10
370 15
0.03 0.002
0.06 0.02
5 ꢃ 106
4 ꢃ 103
6 ꢃ 101
—
ATP
77 20
125 38
299 91
58 25
249 28
285 10
98 20
260 47
281 56
159 51
1900 173
379 39
1600 424
482 243
446 129
562 18
508 155
1252 357
930 396
694 67
0
8
3
49 15
20-dATP
30-dATP
2030-ddATP
GTP
+1.2
+3.4
ꢀ0.7
+2.9
+3.2
+0.6
+3.0
+3.2
+1.8
b
1000 350
Pyruvate kinase
ADP
Sp-ADPaB
Rp-ADPaB
240 85
150 31
6 ꢃ 105
—
c
c
c
20-dGTP
2030-ddGTP
CTP
20-dCTP
2030-ddCTP
2700 305
450 67
c
c
c
Each experiment was performed in triplicate. See materials and methods for
experimental conditions.
a
Km is the Michaelis–Menten dissociation constant, kcat is mol of substrate
a
turned over per mol of enzyme per second, the ratio of kcat/Km is the enzyme effi-
ciency, and Ki is the inhibitory constant of ADPaB for the natural ADP reaction.
One micromolar CK in 50 mM Bicine, 5 mM MgCl2, and 50 mM KCl at pH 8.3
was titrated with 0–2 mM NDP at 25 °C. Experiments were performed in triplicate
and reported as averages and standard deviations.
b
No inhibition observed.
Phosphorylation was too low to be measured.
c
4.1. Effect of ribose and nucleobase substitution
We found that a single substitution on the ribose (20 or 30) or on
the nucleobase (C or G) in ADP/ATP decreases the binding affinity
for both CK and PK (Tables 1–3). However, for CK, NTPs bind with
affinities in the order of 20,30-ddATP P ATP > 20-dATP > 30-dATP
(Table 3). The removal of the ribose 20 hydroxyl group decreases
the binding affinity, as does the removal of the 30 hydroxyl group.
However, substitution of both 20 and 30 hydroxyl groups for hydro-
gens (resulting in dideoxy NTP) increases the affinity of the sub-
strate for CK (Table 3). These results indicate that 20,30-ddATP
may be a preferred substrate for CK. Guanosine and cytidine deriv-
atives show similar trends where the NTPs bind with affinities in
the
order
of
20,30-ddGTP > GTP > 20-dGTP
and
20,30-
ddCTP > CTP > 20-dCTP (Table 3). These results are in agreement
with previously published data which show that the pyrimidine
ddNDPs are substrates for CK [5]. The finding that the lack of both
20 and 30 ribose hydroxyl groups enhances CK binding requires fur-
ther investigation. For PK, the 20,30-ddNTPs did not show increased
affinity compared to the parent NTP.
Fig. 5. Lineweaver–Burk plot for phosphorylation of ( ) ADP and the (s) Sp-and (h)
ꢂ
4.2. Effect of bridging oxygen substitution
Rp-stereoisomers of ADPaB by creatine kinase. Reaction conditions: 0–2 mM NDP in
50 mM Bicine, 5 mM MgCl2, 50 mM KCl, and 20 mM creatine phosphate at pH 8.3 and
25 °C. Optimized enzyme concentrations [CK] and reaction times (t) were as follows
for ADP: [CK] = 6 nM, t = 2 min; Sp-ADPaB: [CK] = 600 nM, t = 15 min; and Rp-ADPaB:
[CK] = 2 lM, t = 70 min. Each experiment was done in triplicate.
The substitution of the bridging oxygen in ADP by a methylene
group, as in AMP(CH2)P (Fig. 1), results in a greater affinity for both
CK and PK (Tables 1 and 2), perhaps because the methylene group

C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
175
ꢀ
NO3 and creatine), these binding interactions seem less important
and the a-P-borano substitution increases the affinity relative to
the natural ADP substrate. Conversely, the Rp isomer is a poor sub-
strate in both direct binding and TSAC formation with CK, possibly
due to a lack of interaction of the borane with the Mg2+ ion. CK
clearly shows stereospecificity toward the Sp isomer of ADPaB,
where the phosphoryl oxygen can interact with Mg2+
.
PK shows opposite stereospecificity. For all of the studied
NDPaB substrate analogs, a greater TSAC affinity is observed for
the Rp isomer over the Sp isomer (Table 2). The opposite stereo-
specificity of CK and PK may be explained by the interaction (or
lack thereof) with Mg2+. Divalent metal ions have been shown to
be essential in phosphoryl transfer reactions [60]. Mg2+ plays an
important role in the substrate binding process by decreasing elec-
trostatic repulsion and stabilizing the transition state. When the
ꢀ
Rp-ADPaB isomer binds to CK, the BH3 group replaces the oxygen
on the a-phosphorus that interacts with Mg2+. The BH3 group
ꢀ
most likely does not interact with the Mg2+ ion [27] and hence
the affinity of the Rp isomer for CK significantly decreases. The
Mg2+ ion plays a similar role in PK substrate binding, with the dif-
ference that the chelating Mg2+ is placed on the opposite side at the
ꢀ
active site (Fig. 7B). Therefore, the Sp isomer BH3 group replaces
the oxygen (on the a-phosphorus) that interacts with Mg2+ and
the resultant limited interaction may cause
affinity.
a decrease in
These results are in accordance with previous studies that re-
ported CK stereospecificity toward the Rp-ATPaS isomer, which
has an Sꢀ substitution in place of the oxygen interacting with
Mg2+ [33] and has the same configuration as the Sp-ATPaB isomer.
In an effort to better understand the mechanism of phosphory-
lation of ADP and the stereoisomers of ADPaB, the phosphorylation
reactions were studied using steady-state kinetics. Both isomers of
ADPaB are minimally phosphorylated by CK. Notably, the natural
substrate (ADP) of CK is phosphorylated 1000 times more effi-
ciently than the Sp-ADPaB isomer. However, significant stereo-
specificity is observed, as the Sp-ADPaB isomer is phosphorylated
70 times more efficiently than the Rp-ADPaB isomer. By contrast,
for PK, no measurable phosphorylation of either ADPaB stereoiso-
mer is observed.
Fig. 6. Lineweaver–Burk plot for (A) Sp-ADPaB ((s) 100 lM, (.) 200 lM, (D) 40-
0 lM, and (j) 600 lM) competitive inhibition of ( ) ADP phosphorylation by CK.
ꢂ
4.4. Stereospecific inhibition of CK and PK
No inhibition was observed with (h) Rp-ADPaB. Reaction conditions: 0–2 mM ADP,
[I] = Sp-ADPaB: 0–600 lM in 50 mM Bicine, 5 mM MgCl2, 50 mM KCl, and 20 mM
creatine phosphate at 25 °C at pH 8.3. [CK] = 6 nM, t = 2 min. (B) (h) Rp-ADPaB
The Sp-ADPaB isomer is a strong competitive inhibitor of the
ADP reaction catalyzed by CK (Table 4). The Sp-ADPaB isomer
seems to bind to CK at the active site but the enzyme may not
adapt to the configuration necessary for efficient catalysis. How-
ever, with the Rp-ADPaB isomer, no inhibition is observed. The
Rp isomer does not bind or adapt to the transition state efficiently
probably due to the a-P-borano substitution eliminating the ability
to chelate the Mg2+ ion. For PK, the Rp-ADPaB isomer is a weak
competitive inhibitor of the ADP phosphorylation, whereas the
Sp-ADPaB isomer is a weak non-competitive inhibitor. Interest-
ingly, PK exhibits opposite stereospecificity than CK. Here, the Rp
isomer of ADPaB seems to bind, as a weak competitive inhibitor,
to the active site of PK but is not phosphorylated with any measur-
able efficiency. The Sp isomer, however, likely binds as a weak non-
competitive inhibitor at a different location than the active site on
PK and hence is not phosphorylated.
competitive inhibition and (s) Sp-ADPaB non-competitive inhibition of ( ) ADP
ꢂ
phosphorylation by PK. Reaction conditions: 0–3 mM ADP, [I] = Rp-ADPaB: 1.6 mM,
and [I] = Sp-ADPaB: 1.9 mM in 100 mM Hepes, 125 mM KCl, 3 mM MnCl2, 3% gly-
cerol, and 2.5 mM phosphoenolpyruvate at pH 7.5. [PK] = 1.5 nM, t = 30 s.
replaces the bridging oxygen (between the a- and b-phosphate in
ADP), which does not seem to have any binding interactions with
either enzyme (Fig. 7A and B). This observation is in accordance
with previous studies that show a,b-methylene analogs of ADP
and ATP can act as substrates for CK [59].
4.3. Stereospecificity of CK and PK for a-P-borano NDP isomers
The Sp-ADPaB isomer shows an increased binding affinity for CK
for direct binding relative to ADP (DDGb = ꢀ2.4 kJ/mol) (Table 1)
but a significantly lower affinity for the TSAC than direct binding.
This may be due to a lack of H-bonding interactions between the
5. Conclusion
ꢀ
BH3 group and the enzyme which appears to be important for
ꢀ
In summary, the affinity studies performed here confirm the
observation that CK and PK play a role in the phosphorylation of
acyclo- and dideoxy-nucleoside diphosphate analogs, as a greater
affinity is observed with ddNTPs than the parent NTPs for CK
the TSAC affinity. The oxygen, replaced by BH3 in ADPaB, may
form hydrogen bonds with R320 amine and V325 amide
ꢀ
(Fig. 7A), whereas BH3 cannot form typical H bonds [27]. It is
notable that, in direct binding with CK (without the presence of

176
C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
Fig. 7. Structure–activity relationships for CK (A) and PK (B). The differences in free energy change (DDGb) upon substitutions on base (C or G), sugar (20-deoxy, 30-deoxy, or
20,30-dideoxy) and a-phosphate (a-P-CH2 or a-P-BH3 relative to the natural substrates (ADP or ATP) are listed in the figure (first value corresponds to direct binding DDGb and
ꢀ
the second value to the TSAC DDGb). The ChemDraw figures of the active site of the CK-TSAC and PK-TSAC structures were adapted from Lahiri et al. and Larsen et al. [46,58]
(distances in angstroms).
(see Fig. 7). Nucleobase or ribose substitutions from ADP are other-
wise not preferred for either CK or PK binding. Both enzymes, how-
ever, display a negatively cooperative binding behavior for all NDP
and NTP substrates and show a greater affinity for adenosine meth-
ylene diphosphonate. For CK, the a-P-borano substituted ADP is
phosphorylated 1000-fold less efficiently than ADP and the en-
zyme is significantly stereospecific for the Sp isomer of ADPaB.
PK shows opposite stereospecificity by demonstrating a preference
for the Rp isomer of NDPaBs in the TSAC. The inhibition studies re-
veal that the Sp isomer of ADPaB is a strong competitive inhibitor
for CK. For PK, the Rp-ADPaB isomer is a poor competitive inhibitor
and the Sp-ADPaB isomer is a poor non-competitive inhibitor.
Moreover, no significant inhibition was observed for the Rp isomer
of ADPaB for either enzyme. This is important as the Rp isomer of
a-P-borano substituted ddNTP analogs has shown promising anti-
viral activity and candidates for therapy should not inhibit host en-
zymes. The study described in this paper is useful as it provides
insight into the stereospecific properties of pyruvate and creatine
kinases and also their specificities for ribose, base and a-phosphate
substrate modifications.

C.K. Wennefors et al. / Bioorganic Chemistry 36 (2008) 169–177
177
[24] B.K. Krzyzanowska, K. He, A. Hasan, B.R. Shaw, Tetrahedron 54 (1998) 5119–
5128.
Acknowledgments
[25] V. Nahum, B. Fischer, Eur. J. Inorg. Chem. 20 (2004) 4124–4131.
[26] P. Li, Z.A. Sergueeva, M. Dobrikov, B.R. Shaw, Chem. Rev. 107 (2007) 4746–
4796.
[27] J.S. Summers, D. Roe, P.D. Boyle, M. Colvin, B.R. Shaw, Inorg. Chem. 37 (1998)
4158–4159.
This work was supported by NIH Grant R01 AI52061 to B.R.S.
and by Charles Bradsher and Kathleen Zielek Duke University grad-
uate fellowships to C.K.W.
[28] B.R. Shaw, M.I. Dobrikov, X. Wang, J. Wan, K. He, J.-L. Lin, P. Li, V. Rait, Z.
Sergueeva, D. Sergueev, Ann. N. Y. Acad. Sci. 1002 (2003) 12–29.
[29] M.I. Dobrikov, K.M. Grady, B.R. Shaw, Nucleosides Nucleotides Nucleic Acids
22 (2003) 275–282.
[30] P. Li, B.R. Shaw, Nucleosides Nucleotides Nucleic Acids 22 (2003) 699–701.
[31] P. Li, B.R. Shaw, Nucleosides, Nucleotides Nucleic Acids 24 (2005) 675–678.
[32] S. Gallois-Montbrun, B. Schneider, Y. Chen, V. Giacomoni-Fernandes, L. Mulard,
S. Morera, J. Janin, D. Deville-Bonne, M. Véron, J. Biol. Chem. 277 (2002)
39953–39959.
[33] F. Eckstein, R. Goody, Biochemistry 15 (1976) 1685–1691.
[34] T.D. Wallimann, M. Dolder, U. Schlattner, M. Eder, T. Hornemann, E. O’Gorman,
A. Ruck, D. Brdiczka, Biofactors 8 (1998) 229–234.
[35] V.A. Saks, Z.A. Khuchua, E.V. Vasilyeva, O.Y. Belikova, A.V. Kuznetsov, Mol. Cell.
Biochem. 133–134 (1994) 155–192.
[36] T. Walliman, W. Hemmer, Mol. Cell. Biochem. 133–134 (1994) 193–220.
[37] E.J. Milner-White, D.C. Watts, Biochem. J. 122 (1971) 727–740.
[38] C.L. Borders, M. Snider, R. Wolfenden, P. Edmiston, Biochemistry 41 (2002)
6995–7000.
[39] S.D. Lahiri, P.-F. Wang, P.C. Babbit, M.J. McLeish, G.L. Kenyon, K.N. Allen,
Biochemistry 41 (2002) 13861–13867.
[40] H. Mazon, O. Marcillat, E. Forest, C. Vial, Biochemistry 42 (2003) 13596–13604.
[41] G.H. Reed, T.S. Leyh, Biochemistry 19 (1980) 5472–5480.
[42] G.H. Reed, M. Cohn, J. Biol. Chem. 247 (1972) 3073–3081.
[43] P. Li, Z. Xu, H. Liu, C.K. Wennefors, M.I. Dobrikov, J. Ludwig, B.R. Shaw, J. Am.
Chem. Soc. 127 (2006) 16782–16783.
[44] S.A. Kuby, L. Noda, H.A. Lardy, J. Biol. Chem. 210 (1954) 65–82.
[45] A.S. Mildvan, M. Cohn, J. Biol. Chem. 240 (1965) 238–246.
[46] K.M. Plowman, A.R. Krall, Biochemistry 4 (1965) 2809–2814.
[47] A.M. Reynard, L.F. Hass, D.D. Jacobsen, P.D. Boyer, J. Biol. Chem. 236 (1961)
2277–2283.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bioorg.2008.03.001.
References
[1] J. Bourdais, R. Biondi, S. Sarfati, C. Guerreiro, I. Lascu, J. Janin, M. Véron, J. Biol.
Chem. 271 (1996) 7887–7890.
[2] P. Meyer, B. Schneider, S. Sarfati, D. Deville-Bonne, C. Guerreiro, J. Boretto, J.
Janin, M. Véron, B. Canard, EMBO J. 19 (2000) 3520–3529.
[3] B. Schneider, R. Sarfati, D. Deville-Bonne, M. Véron, J. Bioenerg. Biomembr. 32
(2000) 317–324.
[4] B. Selmi, J. Boretto, S.R. Sarfati, C. Guerreiro, B. Canard, J. Biol. Chem. 276 (2001)
48466–48472.
[5] P. Krishnan, Q. Fu, W. Lam, J. Liou, G. Dutschman, Y. Cheng, J. Biol. Chem. 277
(2002) 5453–5459.
[6] A.R. Van Rompay, M. Johansson, A. Karlsson, Pharmacol. Ther. 87 (2000) 189–
198.
[7] L. Placidi, E. Cretton-Scott, G. Gosselin, C. Pierra, R.F. Schinazi, J.-L. Imbach, M.H.
Kouni, J.-P. Sommadossi, Antimicrob. Agents Chemother. 44 (2000) 853–858.
[8] J.E. Reardon, R.C. Crouch, L.S. John-Williams, J. Biol. Chem. 269 (1994) 15999–
16008.
[9] N. Sluis-Cremer, D. Arion, M.A. Parniak, Cell. Mol. Life Sci. 57 (2000) 1408–
1422.
[10] J. Deval, B. Selmi, J. Boretto, M.P. Egloff, C. Guerreiro, S. Sarfati, B. Canard, J. Biol.
Chem. 277 (2002) 42097–42104.
[11] H. Huang, R. Chopra, G.L. Verdine, S.C. Harrison, Science 282 (1998) 1669–
1675.
[12] L. Menendez-Arias, Trends Pharmacol. Sci. 23 (2002) 381–388.
[13] P. Li, M.I. Dobrikov, H. Liu, B.R. Shaw, Org. Lett. 5 (2003) 2401–2403.
[14] M. Dobrikov, Z. Sergueeva, B.R. Shaw, Nucleosides Nucleotides Nucleic Acids
22 (2003) 1651–1655.
[15] J. Tomasz, B.R. Shaw, K. Porter, B.F. Spielvogel, A. Sood, Angew. Chem. Int. Ed.
31 (1992) 1373–1375.
[16] K. He, K.W. Porter, A. Hasan, J.D. Briley, B.R. Shaw, Nucl. Acid Res. 27 (1999)
1788–1794.
[17] B.R. Shaw, J. Wan, X. Wang, M. Dobrikov, K. He, K. Porter, J.L. Lin, V. Rait, D.
Sergueev, Z. Sergueeva, in: Collection Symposium series, 2002, pp. 169–180.
[18] J.S. Summers, B.R. Shaw, Curr. Med. Chem. 8 (2001) 1147–1155.
[19] K. He, A. Hasan, B. Krzyzanowska, B.R. Shaw, J. Org. Chem. 63 (1998) 5769–
5773.
[20] A. Sood, B.R. Shaw, B.F. Spielvogel, J. Am. Chem. Soc. 112 (1990) 9000–9001.
[21] B.R. Shaw, D. Sergueev, K. He, K. Porter, J. Summers, Z. Sergueeva, V. Rait,
Methods Enzymol.: Antisense Technol. 313 (2000) 226–257.
[22] K.W. Porter, J.D. Briley, B.R. Shaw, Nucleic Acids Res. 25 (1997) 1611–1617.
[23] B.R. Shaw, J. Madison, A. Sood, B.F. Spielvogel, in: S. Agrawal, Methods in
Molecular Biology Series 20, 1993, pp. 225–243.
[48] M.I. Dobrikov, B.R. Shaw, Nucleosides, Nucleotides Nucleic Acids 24 (2005)
983–987.
[49] R.H. Friesen, R.J. Castellani, J.C. Lee, W. Braun, Biochemistry 37 (1998) 15266–
15276.
[50] A. Ferscht, Structure and Mechanism in Protein Science—A Guide to Enzyme
Catalysis and Protein Folding, W.H. Freeman and Company, New York, 1999.
[51] D. Piszkiewicz, Kinetics of Chemical and Enzyme-Catalyzed Reactions, Oxford
University Press, New York, 1977.
[52] I.H. Segel, Enzyme Kinetics—Behavior and Analysis of Rapid Equilibrium and
Steady-State Systems, John Wiley & Sons, Inc., New York, 1975.
[53] G.F. Bickerstaff, N.C. Price, Int. J. Biochem. 9 (1978) 1–8.
[54] Y. Degani, C. Degani, Biochemistry 18 (1979) 5917–5923.
[55] T. Hornemann, D. Rutishauser, T. Walliman, Biochim. Biophys. Acta 1480
(2000) 365–373.
[56] G.A. Nevinsky, V.N. Ankilova, O.I. Lavrik, Z.S. Mkrtchyan, L.S. Nersesova, J.I.
Akopyan, FEBS Lett. 149 (1982) 36–40.
[57] N.C. Price, M.G. Hunter, Biochim. Biophys. Acta 445 (1976) 364–376.
[58] T.M. Larsen, M.M. Benning, I. Rayment, G.H. Reed, Biochemistry 37 (1998)
6247–6255.
[59] E.J. Milner-White, D.S. Rycroft, Eur. J. Biochem. 133 (1983) 169–172.
[60] I. Lascu, P.J. Gonin, Bioenerg. Biomembr. 32 (2000) 237–246.