Triphosphohydrolase Activity of SAMHD1
O
ployed IEX-HPLC (see Fig. S1 in the supplemental material), re-
verse-phase HPLC, or TLC (13, 15, 16, 18) to measure the product
or the substrate. A comparison of the steady-state parameters for
GTP-activated SAMHD1 hydrolysis of TTP, dATP, and clofara-
bine-TP (Fig. 5B) revealed only small differences in the rate of
hydrolysis, with the kcat for clofarabine-TP reduced from that for
TTP but similar to that of the parent purine dATP, although with
a significantly sigmoidal concentration dependency (Table 1).
SAMHD1 activation by GTP, acyclovir-TP, and ganciclovir-TP
was also analyzed by measuring the TTP hydrolysis rate as a func-
tion of the activator concentration (Fig. 5C). These data yielded an
Acyclovir Triphosphate
N
N
NH
NH
O
O
O
HO
P
O
P
O
P
O
N
O
NH
2
O
O
OH
OH
OH
Ganciclovir Triphosphate
N
N
O
O
O
HO
P
O
P
O
P
O
N
NH
2
OH
OH
OH
apparent affinity for an activator molecule (K ), defined as the
concentration of activator required to achieve 50% of the maximal
A
OH
NH
N
2
Clofarabine Triphosphate
rate of TTP hydrolysis. The K values determined in this way were
A
N
N
N
similar for acyclovir-TP and GTP, so even without a deoxyribose
moiety but with the guanine, acyclovir-TP still activated
SAMHD1 efficiently. In contrast, the structurally related ganciclo-
vir-TP, which also contained a guanine base but maintained a 3=
carbon and hydroxyl, was unable to activate SAMHD1.
O
O
O
HO
P
O
P
O
P
O
Cl
O
F
OH
OH
OH
OH
Endpoint assays also revealed an inhibitory effect on SAMHD1
catalysis by dApNHpp, the modified dATP. To quantify this inhi-
bition and to determine if ganciclovir-TP might also have an in-
hibitory effect, the steady-state hydrolysis of TTP by GTP-acti-
vated SAMHD1 was measured in reaction mixtures containing
increasing amounts of dApNHpp or ganciclovir-TP. These data
NH
N
dApNHpp
2
2
’-deoxy-adenosine-5’-[(α,β)-imido]TP
N
N
N
O
O
O
H
N
HO
P
O
P
P
O
O
OH
OH
OH
(
Fig. 5D) clearly showed potent inhibition by dApNHpp but no
OH
effect from ganciclovir-TP. Fitting the dApNHpp data with a
FIG 3 Chemical structures of nucleoside-triphosphate analogues. Chemical
structures for the antiviral agents acyclovir-TP and ganciclovir-TP, the anti-
leukemia drug clofarabine-TP, and the dATP analogue dApNHpp are shown
competitive inhibition model yielded a mean (ϮSEM) inhibition
constant, K , of 1.13 Ϯ 0.07 M, which is 50- to 100-fold lower
i
than the K measured for the dNTP substrates.
(from top to bottom, respectively).
m
DISCUSSION
activation at the SAMHD1 allosteric site; as expected, no activa- SAMHD1 is expressed in many immune cells and is a dNTP
tion was observed with either clofarabine-TP or dApNHpp. How- triphosphohydrolase that is also allosterically activated by nucle-
ever, whereas acyclovir-TP activated SAMHD1 with around 80% oside triphosphates. This has significant implications for the effi-
efficiency compared to that of GTP, ganciclovir-TP showed no cacy and off-target effects of Nt-AIs, either by their turnover or
activation of SAMHD1-TTP hydrolysis, despite being closely re- through misregulated activation/inhibition of SAMHD1. There-
lated in structure (Fig. 3).
fore, to assess Nt-AI turnover and the modulation of SAMHD1
Finally, to assess the inhibition of SAMHD1, each compound activity, we developed a highly reproducible assay to screen com-
was added to a standard reaction of GTP-activated SAMHD1 at a pound libraries for the effects on SAMHD1 triphosphohydrolase
concentration equimolar to the allosteric activator (200 M), activity that can also be utilized to determine precise K and k
m
cat
with 1 mM TTP as the substrate (Fig. 4D). Under these conditions, values and the inhibition and activation constants for the com-
TTP hydrolysis was not significantly inhibited by the addition of pounds of interest.
acyclovir-TP or ganciclovir-TP and was diminished only slightly
In the endpoint mode, the assay employed large amounts of
by clofarabine-TP. In contrast, the addition of dApNHpp resulted enzyme combined with malachite green for the detection of the
in a substantial reduction in TTP hydrolysis to near background phosphate. This made the assay less sensitive but highly suited to
levels, revealing it as a potential inhibitor of SAMHD1 activity.
application in compound library screening to assess the capacities
Continuous assays. In order to confirm the findings of end- of compounds as the substrates, inhibitors, or activators. In the
point assays and obtain quantitative time-resolved information continuous mode, the assay was more time demanding. However,
about the compound hydrolysis, activation, and inhibition of the high sensitivity of the PBP-based detection allowed for an
SAMHD1, MDCC-PBP was used to measure phosphate release in assessment of turnover in the early course of the reaction and so
real time. The kinetic parameters derived from these assays are was more amenable to obtaining fully quantitative kinetic param-
summarized in Table 1. A Michaelis-Menten analysis of the eters once a compound of interest was identified.
steady-state kinetics of TTP hydrolysis by GTP-activated
SAMHD1 is shown in Fig. 5A. These continuous measurements vealed that the anti-leukemia agent clofarabine-TP is hydrolyzed
Fig. 5A, inset) were highly reproducible. Moreover, analysis of by SAMHD1 at a rate comparable to that of natural dNTP sub-
the data yielded an apparent mean (Ϯ standard error of the mean strates. The anti-herpes agent acyclovir-TP, while not a substrate,
Of the sample compounds tested in this study, this assay re-
(
[
0
SEM]) K of 96.3 Ϯ 1.7 M and a mean (ϮSEM) k of 1.32 Ϯ is able to activate SAMHD1 to hydrolyze other dNTPs with
.07 s for TTP hydrolysis, comparable to the quantitative ki- efficiency similar to that of GTP. The dATP analogue
S
c
a
t
Ϫ1
netic parameters reported by noncontinuous assays that em- dApNHpp is a strong competitive inhibitor of SAMHD1 activ-
January 2015 Volume 59 Number 1
Antimicrobial Agents and Chemotherapy
aac.asm.org 189