A highly sensitive high-throughput luminescence assayfor malonyl-CoA decarboxylase
Mei-Chu Lo
a,*
, Minghan Wang
b
, Ki Won Kim
b
, James Busby
b
, Harvey Yamane
c
,James Zondlo
c
, Chester Yuan
d
, Stephen W. Young
a
, Shou-Hua Xiao
a,*
a
Chemistry Research and Discovery, Amgen, South San Francisco, CA 94080, U.S.A.
b
Metabolic Disorders, Amgen, Thousand Oaks, CA 91320, U.S.A.
c
Protein Science, Amgen, Thousand Oaks, CA 91320, U.S.A.
d
Chemistry Research and Discovery, Amgen, Thousand Oaks, CA 91320, U.S.A.
Received 17 December 2007Available online 2 February 2008
Abstract
Malonyl-CoA decarboxylase (MCD) catalyzes the conversion of malonyl-CoA to acetyl-CoA and thereby regulates malonyl-CoAlevels in cells. Malonyl-CoA is a potent inhibitor of mitochondrial carnitine palmitoyltransferase-1, a key enzyme involved in the mito-chondrial uptake of fatty acids for oxidation. Abnormally high rates of fatty acid oxidation contribute to ischemic damage. Inhibition of MCD leads to increased malonyl-CoA and therefore decreases fatty acid oxidation, representing a novel approach for the treatment of ischemic heart injury. The commonly used MCD assay monitors the production of NADH fluorometrically, which is not ideal for libraryscreening due to potential fluorescent interference by certain compounds. Here we report a luminescence assay for MCD activity. Thisassay is less susceptible to fluorescent interference by compounds. Furthermore, it is 150-fold more sensitive, with a detection limit of 20nM acetyl-CoA, compared to 3
l
M in the fluorescence assay. This assay is also amenable to automation for high-throughput screeningand yields excellent assay statistics (
Z
0
> 0.8). In addition, it can be applied to the screening for inhibitors of any other enzymes thatgenerate acetyl-CoA.
Ó
2008 Elsevier Inc. All rights reserved.
Keywords:
Malonyl-CoA decarboxylase; High-throughput screening; Luminescence assay; Coupled assay; IC
50
; Inhibitor;
K
m
;
k
cat
Thehearthasahighenergy demand thatismetprimarilybytheoxidationoffattyacidsandsecondarilybythemetab-olismofglucoseandlactate[1].Fattyacidoxidationishighlyregulated to ensure that energy supply closely matchesdemand. Carnitine palmitoyltransferase-I is a key enzymeinvolvedinmitochondrialuptakeoffattyacidsandisinhib-ited by malonyl-CoA[2]. As a result, a rise in cardiac malo-nyl-CoA levels suppresses mitochondrial fatty acidoxidation by decreasing fatty acid transport, whereas adecreaseinmalonyl-CoAleadstoincreasedfattyacidoxida-tion[3]. The steady state level of malonyl-CoA is controlledby its synthesis and degradation. It is synthesized from car-boxylation of acetyl-CoA by acetyl-CoA carboxylase(ACC)
anddegradedtoacetyl-CoAbymalonyl-CoAdecar-boxylase (MCD). Under ischemic conditions, limited oxy-gen supply reduces both fatty acid and glucose oxidationand thereby decreases subsequent ATP production. TomaintainATPlevelsduringischemia,glycolysisisstimulatedto drive anaerobic ATP production. Whereas overall mito-chondrial oxidative metabolism decreases during ischemia,two factors cause excessively high rates of fatty acid oxida-tionattheexpense ofglucoseoxidation: (1)circulatingfattyacidlevelsincreaseduringischemia[4]and(2)malonyl-CoA
0003-2697/$ - see front matter
Ó
2008 Elsevier Inc. All rights reserved.doi:10.1016/j.ab.2008.01.033
*
Corresponding authors.
E-mail addresses:
1
Abbreviations used:
ACC, acetyl-CoA carboxylase; BSA, bovine serumalbumin; CoA, coenzyme A; DMSO, dimethyl sulfoxide; DTT, dithio-threitol; HTS, high-throughput screening; MCD, malonyl-CoA decarbox-ylase; PP
i
, pyrophosphate.
www.elsevier.com/locate/yabio
Available online at www.sciencedirect.com
Analytical Biochemistry 376 (2008) 122–130
ANALYTICALBIOCHEMISTRY
levels are reduced by ischemia-induced activation of AMP-activated protein kinase, which phosphorylates and inhibitsACC[5]. Fatty acid oxidation leads to elevated ratios of NADH/NAD
+
and acetyl-CoA/free CoA that further inhi-bit glucose oxidation through inhibition of pyruvate dehy-drogenase[6]. As a result, the pyruvate produced fromglycolysis is not readily oxidized in the mitochondria and isinsteadreduced tolactate, resultinginaccumulationofmet-abolic by-products (lactate and protons) that can lead toimpaired cardiac function, decreased cardiac efficiency,and increased myocardial tissue injury[7].Becausetheexcessiveuseoffattyacidsleadstocontractiledysfunction and ischemic injury, one approach to the treat-ment of ischemia is to decrease fatty acid oxidation byincreasing malonyl-CoA through inhibition of MCD[8].Potent MCD inhibitors have been developed[9–11]andtested in a number of ischemia/reperfusion animal models[12]. These MCD inhibitors significantly increased glucoseoxidation rates and reduced lactate production comparedwith vehicle-treated hearts and showed a significant accom-panying increase in cardiac work compared with controls.However, no efficacious and safe MCD inhibitor has beendemonstrated clinically and approved for use in humansyet.ThereforeweareinterestedinthediscoveryofadditionalMCD inhibitors, especially of different pharmacophores.Currently, the commonly used MCD assay worksthrough coupling to malate dehydrogenase/citrate synthaseand measuring the production of NADH spectrophotomet-rically[13,14]. Because this assay monitors the fluorescenceemission of NADH, interference from fluorescent com-pounds may be an issue when screening compounds usingthis assay. Here we report the development of a lumines-cence-based MCD assay that is more sensitive and less sus-ceptible to compound interference. We have characterizedthe steady state kinetics of purified human MCD using thisassay and compared the results with previously reporteddata from the literature. In addition, the assay was adaptedto a high-throughput screening (HTS) format with a robust
Z
0
factor.
Materials and methods
General
AMP, acetyl-CoA, malonyl-CoA,
L
-malate, NAD
+
, andPP
i
were purchased from Sigma–Aldrich (Milwaukee, WI).Acetyl-CoA synthetase,
L
-malate dehydrogenase, and cit-rate synthase were obtained from Roche Applied Science(Indianapolis, IN). Kinase-Glo Luminescent Kinase Assaywas from Promega (Madison, WI). Compounds A and Bwere synthesized in-house.
Cloning, expression, and purification of His-tagged humanMCD
A cDNA clone containing the coding region of human MCD (residues 41–493) was obtained fromhuman liver cDNA using high-fidelity
Pfu
Ultra Poly-merase (Stratagene, La Jolla, CA). The 5
0
and 3
0
PCRprimers used in the reaction were 5
0
-AAC ATA TGCATC ATC ACC ATC ACC ATG ACG AGC TGCTGC GCC GCG-3
0
and 5
0
-AAC TCG AGT CAGAGC TTG CTG TTC TTT TGA AAC-3
0
, respectively.Two restriction digestion sites,
Nde
I and
Xho
I, wereintroduced into the 5
0
and 3
0
primers outside the openreading frame, respectively. In addition, an in-frame6X His tag was introduced at the 5
0
terminus so thatthe expressed protein started with the sequenceMHHHHHH followed by the residue 41 onward of the native protein. The PCR product was subclonedinto a sequencing vector (PCR2.1) and verified bysequencing. The insert in the PCR2.1 vector wasdigested with restriction enzymes
Nde
I and
Xho
I andsubcloned into an Amgen proprietary plasmid,pAMG21, for bacterial expression of the MCD protein.This plasmid was derived from Amgen plasmidpCFM1656 (ATCC 69576) to allow for chemical induc-tion of recombinant proteins.The recombinant pAMG21 His6-MCD plasmid wastransformed into
Escherichia coli
°
C. A fermentorbatched with Terrific Broth at 25
°
C was inoculated 1%v/v from the overnight culture and growth continued at25
°
C until an
A
600
of approximately 1.5 was reached.Temperature was then dropped to 20
°
C and expressionchemically induced. Cell paste was harvested by centrifu-gation 16 h postinduction. SDS–PAGE analysis of cellpaste fractions indicated that approximately 50% of thetotal recombinant protein appears in the solublefraction.To purify the His-tagged human MCD, the cellpaste was resuspended in a buffer containing 20 mMTris–HCl, pH 7.5, 150 mM NaCl, 5 mM imidazole,10% glycerol, and 20 mM
b
-mercaptoethanol. Cellswere lysed by mechanical shearing in a Microfluidizer(Microfluidics, Newton, MA). The resulting lysate wasclarified by centrifugation at 20,000
g
for 2 h at 4
°
C,followed by filtration through a 0.45-
l
m membrane.The clarified lysate was mixed with Talon metal affinityresin (Clontech, Mountain View, CA) for 1 h at 4
°
C.After washing with a buffer consisting of 20 mM Tris– HCl, pH 7.5, 500 mM NaCl, 5 mM imidazole, 10%glycerol, and 20 mM
b
-mercaptoethanol, bound pro-teins were eluted with 20 mM Tris–HCl, pH 7.5, 150mM NaCl, 100 mM imidazole, 10% glycerol, and 20mM
b
-mercaptoethanol. The Talon pool was dilutedwith four volumes of 20 mM Tris–HCl, pH 7.0, 10%glycerol, and 5 mM DTT and applied to a columnof SP Sepharose HP (GE Healthcare, Piscataway, NJ)equilibrated in the same buffer. Elution was performedwith a linear gradient of 0–0.5 M NaCl in the equili-bration buffer. Peak fractions containing MCD werepooled and aliquots were stored at
À
80
°
C prior touse.
A luminescence HTS assay for malonyl-CoA decarboxylase/M.-C. Lo et al./Anal. Biochem. 376 (2008) 122–130
123
Luminescence-based MCD assay
In the luminescence-based MCD assay, MCD activitywas measured by coupling to the acetyl-CoA synthetaseand luciferase reactions as shown in Eqs.(1)–(3):Malonyl-CoA
!
MCD
acetyl-CoA
þ
CO
2
;
ð
1
Þ
Acetyl-CoA
þ
AMP
þ
PP
i
(
+
acetyl-CoA Synthetase
acetate
þ
ATP
þ
CoA and
ð
2
Þ
Beetle luciferin
þ
ATP
þ
1
=
2 O
2
!
luciferaseMg
2
þ
oxyluciferin
þ
AMP
þ
PP
i
þ
CO
2
þ
light
:
ð
3
Þ
Acetyl-CoA synthetase catalyzes the formation of ATP inthe presence of AMP, PP
i
, and the MCD product, acetyl-CoA. ATP was subsequently quantified by measuring lumi-nescence intensity generated by monooxygenation of beetleluciferin catalyzed by luciferase in the presence of Mg
2+
,ATP, and molecular oxygen.The MCD reactions were performed in a white 384-well Opti-plate (PerkinElmer, Shelton, CT) with a totalvolume of 20
l
L. To determine the
K
m
of malonyl-CoA,reactions were conducted with 0.1 nM MCD and variousmalonyl-CoA concentrations at 1.9, 3.8, 7.5, 15, 30, 60,and 90
l
M in buffer containing 40 mM Tris, pH 8.0,0.02% Tween 20, 0.1% BSA, and 1 mM DTT. Reactionswere stopped at different time points by adding 5
l
l
M (1000-fold above IC
50
); 25
l
L of theluminescence detection reagent containing 0.9X Kinase-Glo Reagent, 20
l
M AMP, 20
l
M PP
i
, and 0.003 U/
l
Lacetyl-CoA synthetase was then added and 1X Kinase-Glo Reagent was prepared by adding 10 mL of Kinase-Glo Buffer to one vial of Kinase-Glo Substrate followingthe manufacture’s instructions. After adding the lumines-cence detection reagent, the plate was incubated at roomtemperature for at least 5 min, and luminescence intensitywas measured using an Envision microplate reader (Perk-inElmer) with 0.2 s per well integration time. The initialvelocities were determined from the slopes of the progresscurves in the linear range and the luminescence units wereconverted to the acetyl-CoA concentrations based on thestandard curve of acetyl-CoA.
K
m
and
k
cat
were obtainedby fitting the initial velocities to the Michaelis–Mentenequation (Eq.(4)) using GraphPad Prism 4.02 (Graph-Pad, San Diego, CA).
V
¼
V
max
½
S
ð
K
m
þ ½
S
Þ
;
ð
4
Þ
where [S] is the substrate concentration,
V
is the initialvelocity at a given [S],
V
max
is the maximum velocity, and
K
m
is the Michaelis–Menten constant.
k
cat
, the enzymeturnover number, was derived by dividing
V
max
by [E],the concentration of enzyme.To determine the IC
50
values of MCD inhibitors, reac-tions were conducted at 9
l
M malonyl-CoA, 0.1 nMMCD, and various concentrations of compounds. After15 min incubation at room temperature, reactions werestopped by adding compound A to a final concentrationof 10
l
M. The luminescence detection reagent was addedand luminescence intensity was measured as describedabove. To obtain IC
50
values, the luminescence intensitiesat different compound concentrations were fitted to a sig-moidal dose–response curve with a variable slope (Eq.(5)) using GraphPad Prism 4.02 (GraphPad).
V
i
¼
V
min
þð
V
max
À
V
min
Þ
1
þ ð
½
I
IC
50
Þ
Hillslope
;
ð
5
Þ
where [I] is the concentration of the inhibitor,
V
i
is the ini-tial velocity at a given [I],
V
min
is the background of the as-say, and
V
max
is the initial velocity without inhibitor.For Z
0
factor determination, 48 wells of a 384-wellplate were used as positive controls and another 48 wellswere used as negative controls. Negative controls wereuninhibited MCD reactions containing 9
l
M malonyl-CoA and 0.1 nM MCD in buffer of 40 mM Tris, pH8.0, 0.02% Tween 20, 0.1% BSA, and 1 mM DTT. Posi-tive controls were completely inhibited reactions contain-ing 10
l
M compound A (dissolved in DMSO) inaddition to the same MCD reaction components presentin negative controls. A small amount of DMSO wasadded to the negative control wells to match the DMSOconcentration in the positive control wells (final 5%DMSO). Reactions were incubated at room temperaturefor 15 min and then stopped and detected in the sameways as described above. Z
0
Z
0
¼
1
Àð
3
r
c
þ
þ
3
r
c
À
Þj
l
c
þ
À
l
c
À
j
;
ð
6
Þ
where
r
c+
and
l
c+
are the standard deviation and the meanof the positive control signal, respectively, and
r
c-
and
l
c-
are the standard deviation and the mean of the negativecontrol signal, respectively.
NOCF
3
CF
3
OHNHOOO
R /S
Compound A
OONF
3
COHCF
3
Compound B
Fig. 1. Chemical structures of MCD inhibitors, compounds A and B.124
A luminescence HTS assay for malonyl-CoA decarboxylase/M.-C. Lo et al./Anal. Biochem. 376 (2008) 122–130
Search History:
Result 00 of 00
00 results for result for
p.
A Highly Sensitive High-throughput Luminescence Assay for Malonyl-CoA Decarboxylase
2.0K
Reads
5
Readcasts
1
Embed Views
More From This User
Notes
Load more
