Mammalian fatty acid synthase activity from crude tissue lysates tracing 13C-labeled substrates using gas chromatography-mass spectrometry

Article abstract of DOI:10.1016/j.ab.2012.06.013

Fatty acid synthase (FASN or FAS, EC 2.3.1.85) is the sole mammalian enzyme to synthesize fatty acids de novo from acetyl- and malonyl-coenzyme A (CoA) esters. This article describes a new method that directly quantifies uniformly labeled ¹³C₁₆-labeled palmitate ([¹³C ₁₆]palmitate) by tracing [¹³C₂]acetyl-CoA and [¹³C₃]malonyl-CoA using an in vitro FASN assay. This method used gas chromatography-mass spectrometry (GC-MS) to detect [ ¹³C₁₆]palmitate carboxylate anions (m/z 271) of pentafluorobenzyl (PFB) derivatives and was highly sensitive at femtomole quantities. Uniformly incorporated [¹³C₁₆]palmitate was the primary product of both recombinant and crude tissue lysate FASN. Quantification of FASN protein within crude tissue lysates ensured equal FASN amounts, preserved steady-state kinetics, and enabled calculation of FASN-specific activity. FASN activity determined by [¹³C ₁₆]palmitate synthesis was consistent with values obtained from β-nicotinamide adenine dinucleotide 2′-phosphate (NADPH) oxidation assays. Analysis of FASN activity from tissue extracts was not hampered by contaminating enzymes or preexisting fatty acids. Crude mammary gland and liver lysates had significantly different activities at 82 and 65 nmol min ^-1 mg^-1, respectively, suggesting that tissue-specific activity levels differ in a manner unrelated to FASN amount. GC-MS quantification of [¹³C₁₆]palmitate synthesis permits sensitive evaluation of FASN activity from tissues of varied physiological states and of purified FASN activity in the presence of modifying proteins, enzymes, or drugs.

Full text of DOI:10.1016/j.ab.2012.06.013

Analytical Biochemistry 428 (2012) 158–166
Contents lists available at SciVerse ScienceDirect
Analytical Biochemistry
journal homepage: www.elsevier.com/locate/yabio
Mammalian fatty acid synthase activity from crude tissue lysates tracing 13C-labeled
substrates using gas chromatography–mass spectrometry
b
a
c
c
Michael C. Rudolph a, , N. Karl Maluf , Elizabeth A. Wellberg , Chris A. Johnson , Robert C. Murphy ,
⇑
Steve M. Anderson a
a Department of Pathology, University of Colorado, Anschutz Medical Campus, Aurora, CO 80045, USA
b School of Pharmacy, University of Colorado, Anschutz Medical Campus, Aurora, CO 80045, USA
c Department of Pharmacology, University of Colorado, Anschutz Medical Campus, Aurora, CO 80045, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 May 2012
Received in revised form 11 June 2012
Accepted 13 June 2012
Available online 20 June 2012
Fatty acid synthase (FASN or FAS, EC 2.3.1.85) is the sole mammalian enzyme to synthesize fatty acids de
novo from acetyl- and malonyl-coenzyme A (CoA) esters. This article describes a new method that
13
directly quantifies uniformly labeled
C
-labeled palmitate ([13C16]palmitate) by tracing [13C2]acetyl-
16
CoA and [13C3]malonyl-CoA using an in vitro FASN assay. This method used gas chromatography–mass
13
spectrometry (GC–MS) to detect [
C ]palmitate carboxylate anions (m/z 271) of pentafluorobenzyl
16
(PFB) derivatives and was highly sensitive at femtomole quantities. Uniformly incorporated [13C16]palmi-
tate was the primary product of both recombinant and crude tissue lysate FASN. Quantification of FASN
Keywords:
Fatty acid synthase
protein within crude tissue lysates ensured equal FASN amounts, preserved steady-state kinetics, and
Gas chromatography–mass spectrometry
Mammary gland/liver lysates
13
enabled calculation of FASN-specific activity. FASN activity determined by [
C ]palmitate synthesis
16
was consistent with values obtained from b-nicotinamide adenine dinucleotide 20-phosphate (NADPH)
oxidation assays. Analysis of FASN activity from tissue extracts was not hampered by contaminating
enzymes or preexisting fatty acids. Crude mammary gland and liver lysates had significantly different
activities at 82 and 65 nmol minꢀ1 mgꢀ1, respectively, suggesting that tissue-specific activity levels differ
in a manner unrelated to FASN amount. GC–MS quantification of [13C16]palmitate synthesis permits sen-
sitive evaluation of FASN activity from tissues of varied physiological states and of purified FASN activity
in the presence of modifying proteins, enzymes, or drugs.
13C-labeled substrate incorporation
13
[
C ]palmitic acid
16
Ó 2012 Elsevier Inc. All rights reserved.
Fatty acid synthase (FASN1 or FAS, EC 2.3.1.85) is expressed in
and posttranslational modification of proteins [4–8]. FASN abun-
dance and de novo fatty acid synthesis activity are commonly dereg-
ulated in a wide variety of human cancers and in metabolic diseases,
underscoring the importance for the ability to study the enzyme
[9–11]. Importantly, the enzymatic conversion of acetyl-CoA and
malonyl-CoA substrates into palmitic acid can be recapitulated in
the test tube, which has fostered elegant dissection of the mecha-
nisms underlying FASN activity [3].
The majority of in vitro FASN activity assays can be grouped
into two main categories: (i) assays that measure consumption/
production of reaction components and (ii) assays that measure
incorporation of heavy atom labeled substrates into the fatty acid
products. Assays in the first group commonly quantify b-nicotin-
amide adenine dinucleotide 20-phosphate (NADPH) oxidation into
NADP+ by monitoring changes in ultraviolet (UV) absorbance at
340 nm. Meticulous biochemistry during the 1990s determined
the stereochemistry for the hydrogen atoms derived from NADPH,
water, or malonyl-CoA during elongation of the fatty acyl chain to
confirm that 14 molecules of NADPH are consumed for each palmi-
tate synthesized [12,13]. Researchers have used this phenomenon
most human tissues and is indispensable in the mouse because the
FASN null mice are embryonic lethal [1,2]. FASN is a complex en-
zyme that is absolutely essential for the cellular synthesis of short
and medium chain fatty acids in mammals [3]. Mammalian FASN
is an approximately 270-kDa enzyme that contains seven enzymatic
domains and harbors the complete pathway to synthesize palmitic
acid de novo from acetyl and malonyl esters of coenzyme A (CoA)
as substrates (Fig. 1). The products of FASN, free fatty acids, have
multiple biological functions, including lipid storage, phospholipid
biosynthesis, both endocrine and nuclear hormone signaling ligands,
⇑
Corresponding author. Fax: +1 303 724 3712.
E-mail address: michael.rudolph@ucdenver.edu (M.C. Rudolph).
Abbreviations used: FASN (or FAS), fatty acid synthase; CoA, coenzyme A; PFB,
1
pentafluorobenzyl; NADPH, b-nicotinamide adenine dinucleotide 2’-phosphate; CPM,
7-diethylamino-3-(4’-maleimidylphenyl)-4-methylcoumarin; GC–MS, gas
chromatography–mass spectrometry; NICI, negative ion chemical ionization; DIEA,
N,N-diisopropylethylamine; DTT, dithiothreitol; EDTA, ethylenediaminetetraacetic
acid; DMSO, dimethyl sulfoxide; SDS, sodium dodecyl sulfate; PVDF, polyvinylidene
fluoride; PFB-Br, pentafluorobenzyl bromine; MEC, mammary gland epithelial cell.
0003-2697/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ab.2012.06.013

Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
159
13
[
1,2- C2] Acetyl-CoA x 1
14 NADPH
14 H+
[1,2,3-13C3] Malonyl-CoA x 7
[13C16] Palmitic acid (16:0)
FASN
14 NADP+ + 7 13CO2 + 8 CoA + 6 H2O
(dimer)
Fig.1. Reaction equation for the activity of fatty acid synthase showing conversion of [13C2]acetyl-CoA and [13C3]malonyl-CoA substrates, NADPH, and protons into 16-carbon
saturated [13C16]palmitic acid plus by-products of enzyme catalysis.
to measure loss of absorption at 340 nm due to NADPH oxidation
as a surrogate for FASN activity [14]. An alternative to evaluating
NADPH consumption is to monitor the production of free CoA mea-
sured using fluorescent molecules that react with the free thiol of
CoA such as 7-diethylamino-3-(40-maleimidylphenyl)-4-methyl-
coumarin (CPM) [15,16]. However, free CoA plays a critical role
in the initial substrate sorting steps of the FASN reaction, and as
a result, scavenging of CoA using covalent dyes such as CPM could
adversely affect FASN catalysis [17]. In addition to CoA scavenging,
thiol-reactive dyes have the potential to bind to FASN, and this
interaction could alter catalysis. Moreover, the use of thiol-reactive
compounds to measure FASN activity in a crude tissue or cell lysate
is likely to have decreased sensitivity due to dye interaction with
molecules in the lysates. Despite these pitfalls, indirect assays for
FASN activity remain popular and are used successfully largely be-
cause they use basic laboratory equipment, are easily scalable, and
therefore are well suited for high-throughput screens.
standard [13C4]palmitate ion (m/z 259). Peak areas of the analyte or
of the standard were measured, and the ratio of the area from the
analyte-derived ion to that from the internal standard was calcu-
lated. The ratios were then compared with the calibration curve
(see below) for the analyte prepared from commercially available
standards to quantify [13C16]palmitate.
Palmitic acid calibration curve
13
[
C
]palmitate, [1,2,3,4-13C4]palmitate, and [D31]palmitate
16
were purchased from Sigma–Aldrich (St. Louis, MO, USA) with 99
atom% 13C and 98 atom% D respectfully. Stock solutions for each
compound were 1 mg/ml in 90% methanol. [13C16]palmitate was
the analyte and [13C4]palmitate was the internal standard. The
standard curve was prepared as a 5-fold serial dilution, and the
analyte-to-standard ratios were measured in triplicate at seven dif-
ferent analyte amounts ranging from 460 pmol to 150 fmol, with
the internal standard held constant at 8 pmol. Analyte-to-standard
dilutions for generation of the standard curve made in 90% meth-
anol were acidified to pH 6 4.0, isooctane extracted, derivatized,
and transferred to vials as described in the next section, with the
exception that the derivatized pellet for the 460-pmol point was
resuspended in 1.0 ml of isooctane and that for the 150-fmol point
FASN activity assays in the second category measure the end
product of the FASN—de novo synthesized, nonesterified fatty
acids. The majority of these assays rely on incorporation of a radio-
active precursor into synthesized fatty acids and quantification of
reaction products by liquid scintillation counting [18]. Others have
used nonradioactive tracers, including D2O and 13C1-labeled malo-
nyl-CoA ([13C1]malonyl-CoA) incorporation, to measure palmitate
synthesis with success [19,20]. NADPH oxidation, radioactive trac-
ers, and nonradioactive tracer methods have been problematic for
a variety of reasons, including the indirect measurement of palmi-
tate synthesis, loss of information about fatty acid chain length,
hazardous radioactive waste, isotope purity of substrates, deute-
rium isotope effects, and decarboxylation of malonyl-CoA from
crude tissue extracts. In this study, we describe a novel gas
chromatography–mass spectrometry (GC–MS)-based FASN activity
was resuspended in 25 ll of isooctane.
Sample preparation for GC–MS
High-performance liquid chromatography (HPLC)-grade re-
agents were purchased from Sigma–Aldrich. Following the acidifi-
cation step, 8 pmol of [13C4]palmitate was added as the internal
reference standard, which locked the analyte-to-standard ratio
prior to fatty acid extraction. Fatty acids were extracted with
1.0 ml of isooctane; samples were vortexed vigorously for 10 s and
allowed to stand for 5 min at room temperature, after which they
assay that uses high isotopic purity
[
[
13C2]acetyl-CoA and
13C3]malonyl-CoA reaction substrates to directly measure FASN
product, de novo synthesized, uniformly incorporated [13C16]pal-
mitate. This method is well suited for small-scale in vitro reactions
with sensitivity at the femtomole level, uses commercially avail-
able 99% isotope purity substrates and standards, and calculates
FASN-specific activity directly from crude extracts, obviating the
need for complicated FASN purifications.
were centrifuged for 1 min at 300g. An 800-
ganic phase was transferred to new glass tubes and taken to dryness
using a Savant SpeedVac. Pellets were resuspended in 30 l of 1%
PFB bromide in acetonitrile, and 30 l of 1% N,N-diisopropylethyl-
ll volume of the top or-
l
l
amine (DIEA) dissolved in acetonitrile was added to initiate deriva-
tization [21]. The derivatization reaction was allowed to proceed at
room temperature for 30 min, after which the samples were dried
using a Savant SpeedVac. The resulting pellets were resuspended
Materials and methods
in 100 ll (unless otherwise noted) of isooctane and transferred into
vials for GC–MS.
GC–MS fatty acid quantification
Methods used in this study were adapted from Zarini and
coworkers [21]. Sample analysis was performed by negative ion
chemical ionization (NICI) GC–MS using the Finnigan DSQ GC–MS
system (Thermo Finnigan, Thousand Oaks, CA, USA) with a ZB-l col-
umn (15 m–0.25 mm inner diameter, 0.10 mm film thickness, Phe-
nomenex). The GC was programmed to increase the temperature
from 125 to 230 °C at 20 °C/min, 230 to 245 °C at 5 °C/min, and fi-
nally 245 to 300 °C at 30 °C/min before being held at 310 °C for
1 min. The mass spectrometer was operated in the NICI mode using
methane as reagent gas. Data were acquired in full scan mode to
identify palmitic acid, [13C16]palmitate (m/z 271), and the internal
FASN activity assay
Reactions were performed in a total volume of 200 ll, and FASN as-
say buffer consisted of 0.1 M potassium phosphate (pH 6.8), 1.0 mM
dithiothreitol (DTT), and 1.0 mM ethylenediaminetetraacetic acid
(EDTA). [1,2-13C2]acetyl-CoA at 99 atomic% 13C, [1,2,3-13C3]malonyl-
CoA at 99 atomic% 13C, and NADPH reduced tetrasodium salt were pur-
chased from Sigma–Aldrich. Reactions contained 2 lg (representing
36.7 nm dimer concentration) of recombinant FASN or FASN isolated
from tissues (see below). Recombinant FASN or FASN isolated from tis-
sues was diluted to 0.2 lg/ll in 250 mM potassium phosphate (pH

160
Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
7.0), 1.0 mM DTT, and 1.0 mM EDTA and then equilibrated to room
temperature for 1.0 h to restore FASN dimers. FASN assay buffer,
and 3ꢁ Roche complete EDTA-free protease inhibitors). Homoge-
nized samples were centrifuged at 3500g for 10 min, and superna-
tant was collected and cleared at 100,000g for 30 min by
ultracentrifugation. The protein concentrations present in the
resulting cytosolic fraction were estimated using Pierce 660-nM
40
(2
l
l
M [13C2]acetyl-CoA, 180
l
M NADPH, and 10
g total) were preincubated at 37 °C for 2 min. The reactions were
M [13C3]malonyl-CoA and mixed. Fol-
ll of diluted FASN
initiated by the addition of 110
lowing a 60-s reaction time, the samples were quenched with 2 vol-
umes (500 l) of ice-cold methanol, acidified with 40 l of 1.0 M HCl
l
Protein Assay (Thermo Fisher). A 60-lg amount of total cytosolic
l
l
protein was loaded to denaturing SDS–PAGE (polyacrylamide gel
electrophoresis) and transferred to low-fluorescence PVDF mem-
brane (Millipore). Immunoblotting of FAS was performed as above
using anti-FAS primary antibody and the Licor goat anti-rabbit IR-
Dye 800cw secondary and was scanned using the LiCor Odyssey
system. The amount of FASN present in tissue lysates was quanti-
fied by comparison with a standard curve using purified recombi-
to approximately pH 3.5, and then gently vortexed. Cerulenin was pur-
chased from Sigma–Aldrich and dissolved to 5 mM in dimethyl sulfox-
ide (DMSO). In experiments with cerulenin, samples were incubated in
the FASN assay buffer without substrates for 30 min at room tempera-
ture, alone, with 6
ll of DMSO (2%, v/v) or with 1.0, 2.0, 10, 50, or 100
lM cerulenin (each in 6
ll of DMSO). Following cerulenin incubation,
assay was conducted as before. Samples were then prepared for GC–
nant FASN. The volume of cytosolic extract representing 2 lg of
MS as described above.
FASN protein levels was evaluated using the FASN activity assay
as above.
Recombinant FAS
Results
Recombinant rat FASN enzyme was kindly provided by S. Smith
and A. Witkowski (Children’s Hospital Oakland Research Institute,
Oakland, CA, USA). Enzyme concentration was measured using a
Nanodrop spectrophotometer (Thermo Scientific), and concentra-
tion was estimated using a 280-nm extinction coefficient of
482,200 Mꢀ1 cmꢀ1 (dimer without disulfides) according to Protein
Calculator version 3.3 (http://www.scripps.edu/~cdputnam/prot-
calc.html) using rat FASN primary sequence at http://www.Uni-
Prot.org (entry P12785). Details of the preparation of C-
terminally FLAG-tagged recombinant FASN can be found elsewhere
[22]. Recombinant FASN protein was used to generate a standard
curve to quantify the amount of FASN present in biological samples
isolated from tissues. Triplicate dilutions ranging from 10 to 900 ng
of FASN protein were run on sodium dodecyl sulfate (SDS) poly-
acrylamide gels, transferred to low-fluorescence polyvinylidene
fluoride (PVDF) membrane (Millipore, Billerica, MA, USA), and
incubated with 1:1000 FASN primary antibody from Santa Cruz
Biotechnology (rabbit anti-human FASN, sc-20140, Santa Cruz,
CA, USA), which was generated against a 299-amino acid region
in the C terminus of human FASN (aa 2205–2504). The immuno-
blots were incubated in 1:15,000 goat anti-rabbit IRDye 800CW
secondary (LI-COR Biosciences, Omaha, NE, USA). Infrared immu-
noblots were scanned using the LiCor Odyssey system, and data
were analyzed using LiCor Image Studio 2.0 software (LI-COR Bio-
sciences). Amounts of endogenous FASN were calculated according
to the following equation:
GC–MS chromatograms and mass spectra of fatty acids
GC resolved individual PFB fatty acid esters of discrete chain
lengths based on chromatography retention times. Pentafluorob-
enzyl bromine (PFB-Br) and DIEA were used as the derivatization
agents that generated fatty acid carboxylate anions under NICI.
The abundance of fatty acid carboxylate anions eluted from the
column was detected by the mass spectrometer, which provided
far superior signal-to-noise than other methods such as analysis
of fatty acid methyl esters. Four palmitate isotopomers were pur-
chased from a commercial vendor—unlabeled palmitate, [13C4]pal-
mitate,
13
[ C ]palmitate, and [D31]palmitate—and an optimal
16
internal reference standard was determined. These isotopomers
of palmitic acid were mixed together, extracted, and derivatized,
and then the mixture was subjected to GC–MS analysis. Identical
retention times for unlabeled and 13C-labeled palmitate are ob-
served in the extracted ion chromatogram (Fig. 2A), but the reten-
tion time of the [D31]palmitate (m/z 286) standard was shifted 8.4 s
earlier. The unique m/z value of each palmitic acid isotopomer was
readily distinguished by the mass spectrometer for the unlabeled
palmitate (m/z 255), [13C4]palmitate (m/z 259), [13C16]palmitate
(m/z 271), and [D31]palmitate (m/z 286) carboxylate anions
(Fig. 2B). Also detected were the carboxylate anions of the natural
1.1% abundance of 13C of the isotopomers for unlabeled palmitate
(m/z 256), [13C4]palmitate (m/z 260), and [D31]palmitate (m/z
287). No natural 1.1% abundance of 13C ion (e.g., m/z 272) was de-
tected for [13C16]palmitate because all 16 carbons of palmitate
were already 13C labeled. Instead, a small fraction of [12C]palmitate
was detected (m/z 270) due to the 99 atom% 13C isotopic purity
from the commercial vendor. In addition, a large percentage of
the [D31]palmitate produced m/z 285, 284, and 283 ions, indicating
that the 98 atom% D for this labeled palmitate was less than desir-
able. [13C4]palmitate was used as the internal standard and not
ꢀ
ꢁ
ng FASN
Regression Slope in
Signal Intensity
ꢀ
ꢁ
Signal Intensity
g Cytosolic Protien
ng FASN
ꢁ
¼
lg Cytosolic Protein
60
l
Isolation of FASN from mouse mammary gland and liver
Animals were maintained in the Center for Comparative Medi-
cine under protocols approved by the institutional animal care
and use committee of the University of Colorado–Denver. C57Bl/6
mice were purchased from Jackson Laboratories (Bar Harbor, ME,
USA) and maintained on standard laboratory chow and a standard
day/night cycle. Mammary gland epithelial cells were prepared as
described previously with the exception that phenylmethanesulfo-
nyl fluoride (PMSF) was omitted from the process [23]. Liver tissues
from the identical mice used for preparation of mammary epithelial
cells were collected and immediately snap-frozen on dry ice.
Approximately 150 to 250 mg of tissue was homogenized in 1 ml
of FASN lysis buffer (250 mM sucrose, 20 mM Hepes [pH 7.6],
2 mM MgCl2, 1 mM DTT, 1 mM EDTA, 50 mM NaF, 5 mM NaVO4,
[D31]palmitate because the retention time was identical to
13
[
C
]palmitate and [13C4]palmitate (m/z 259) differed by 12 Da
16
from the [13C16]palmitate (m/z 271) synthesized in vitro by FASN.
A standard curve for quantification was established using
13
[
C
]palmitate as the analyte and [13C4]palmitate as the internal
16
standard. Six different amounts of [13C16]palmitate that spanned
nearly four orders of magnitude (0.15–460 pmol) were subjected
to GC–MS analysis. The signal ratio of the analyte to the standard rel-
ative to the amount of analyte generated a standard regression curve
(Fig. 3A). The dilution series was conducted in triplicate and had an
R
2 correlation value of 0.9991 with a slope of 0.104 0.00007, and
the inset of Fig. 3A shows linearity of the low end of the curve
(0.15–3.7 pmol of analyte). This method was highly sensitive and

Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
161
A
B
[13C4]-16:0
RT: 6.54
259.2
RT: 6.54
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
[13C16]-16:0
[D31]-16:0
271.3
RT: 6.54
100
286.4
Unlabeled
RT: 6.42
90
80
70
60
50
40
30
20
10
0
16:0
255.2
Unlabeled 16:0
13C4]-16:0
285.4
[
13
[
C ]-16:0
16
284.4
287.4
260.3
260
270.3
256.2
[D31]-16:0
6.7 6.8
6.2
6.3
6.4
6.5
6.6
6.9
0
250
255
265
270
275
280
285
290
Time (min)
m/z
Fig.2. Extracted ion chromatogram and spectrum of commercially available isotopomers of palmitate (16:0) mixed together, extracted, derivatized, and injected into the GC–
MS. (A) Resolution of PFB fatty acid esters by chromatographic retention time. [13C4]palmitate and [13C16]palmitate coeluted at 6.54 min with the unlabeled palmitate, but
[D31]palmitate eluted from the column 0.14 min (8.4 s) earlier than the [13C]palmitate molecules. (B) Extracted ion spectrum from retention time 6.40 to 6.7 for carboxylate
anions of palmitate isotopomers at m/z 255 (unlabeled), m/z 259 (13C4), m/z 271 (13C16), and m/z 286 (D31).
A
B
0.5
0.4
0.3
0.2
0.1
0.0
460 pmol Analyte : 8 pmol Standard
0.15 pmol Analyte : 8 pmol Standard
50
40
30
20
10
0
13
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
RT: 6.53
[
13C4]-16:0
100
RT: 6.53
[
C
]-16:0
16
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
1
2
3
4
RT:6.54
[
13C4]-16:0
13
RT: 6.52
6.3
[
C ]-16:0
16
0
100
200
300
400
500
0
0
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.2
6.4
6.5
6.6
6.7
6.8
6.9
Picomoles of analyte
Time (min)
Time (min)
Fig.3. (A) Serial dilution regression curve of the [13C16]palmitate analyte-to-[13C4]palmitate standard ratio relative to the amount of analyte. The analyte amount varied from
0.15 to 460 pmol, and the standard was held at 8 pmol. The dilution series was performed in triplicate, with a slope of 0.104 and a correlation value of 0.9991. The inset shows
the range from 0.15 to 3.7 pmol of analyte. B) Extracted ion chromatograms for the 460:8 pmol point (left) and the 0.15:8 pmol point (right) for the [13C16]palmitate analyte
(front peaks) and the [13C4]palmitate standard (back peaks). The sensitivity of this assay was 0.15 pmol of [13C16]palmitate analyte.
routinely capable of detecting [13C16]palmitate fatty acid carboxyl-
ate anions at and below 150 fmol. The relative abundance of the
460 pmol and 150 fmol [13C16]palmitate analyte (front peaks) to
the internal [13C4]palmitate 8-pmol standard (back peaks) is dem-
onstrated in the extracted ion chromatograms (Figs. 3A and B). The
strong linear correlation value of the standard curve, combined with
the sensitivity of the mass spectrometer using PFB fatty acid deriva-
tives, enabled measurements with a broad dynamic range down to
femtomole amounts of [13C16]palmitate.
with the full-length FASN-encoding baculovirus expression vector.
FASN activity assays have been well defined in the literature
[18,24] and were conducted with minor modifications according
to Materials and Methods. FASN dimerization is required for com-
plete enzyme activation and subsequent synthesis of palmitic acid,
but dimer assembly is sensitive to temperatures below 20 °C and
solutions of low ionic strength (<50 mM potassium phosphate buf-
fer), such that FASN dimers slowly dissociate under either condition
[25]. A critical aspect of either the [13C16]palmitate synthesis or
NADPH oxidation method for FASN activity was fully assembled
FASN dimers. Recombinant FASN or tissue-derived FASN was rou-
tinely incubated in 0.25 M potassium phosphate (pH 7.0), 1 mM
EDTA, 1 mM DTT, and 10% glycerol for 1 h at room temperature to
facilitate FASN dimer assembly. Without this process, only 20% of
FASN activity assay
Conditions of the in vitro assay were established using affinity-
purified recombinant rat FASN isolated from Sf9 insect cells infected

162
Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
A
B
10 s
60 s
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
RT: 6.54
271.3
RT: 6.53
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
[13C16]-palmitate
analyte
140
120
100
80
RT: 6.54
RT: 6.54
0
0
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
Time (min)
Time (min)
Unlabeled
palmitate
[13C4]-palmitate
standard
60
255.2
40
259.2
20
270.3
y = 1.6806x + 9.2054
R² = 0.9967
256.2
0
0
0
10
20
30
40
50
60
250 252 254 256 258 260 262 264 266 268 270 272 274
Time (s)
m/z
Fig.4. (A) Extracted ion spectrum for [13C16]palmitate analyte (m/z 271), [13C4]palmitate (m/z 259) standard, and unlabeled palmitate (m/z 255) from the recombinant FASN
assay. The primary product is [13C16]palmitate when supplying [13C2]acetyl-CoA and [13C3]malonyl-CoA substrates in the reaction. (B) Recombinant FASN activity over 1 min
monitoring the [13C16]palmitate carboxylate anion (m/z 271). The reaction proceeds linearly, demonstrating steady-state kinetics and that recombinant FASN synthesized
101 nmol [13C16]palmitate minꢀ1 mgꢀ1. The abundance of [13C16]palmitate (front peak) relative to the [13C4]palmitate standard (back peak) in extracted ion chromatograms
for the 10- and 60-s points increased over time (insets).
the fully reactivated FASN activity was observed (data not shown).
The need to reactivate recombinant, purified, or crude tissue lysate
FASN was underscored by the observation that FASN reactivation
(dimer assembly) occurred during the reaction at 37 °C because
the rate of [13C16]palmitate synthesis was parabolic up to 2 min into
the reaction when the rate then became linear (not shown).
The primary product of recombinant FASN was [13C16]palmitate
(m/z 271) when the reaction was supplied with [13C2]acetyl-CoA
concentrations that yielded the most optimal slope and linearity
(R2 correlation value) were selected. Later time points, such as 90
and 120 s, using 40 l l
M [13C2]acetyl-CoA and 110 M [13C3]malo-
nyl-CoA tended to plateau, indicating that the reaction had deviated
from steady-state kinetics, presumably due to high concentrations
of free CoA that inhibits FASN catalysis [17]. The amount of
13
[
C ]palmitate synthesized in the reaction was quantified relative
16
to the analyte-to-standard ratio regression curve (Fig. 3). FASN syn-
thesized [13C16]palmitate at 101 nmol minꢀ1 mgꢀ1 (Fig. 4B), and the
increased m/z 271 [13C16]palmitate (front peak) relative to m/z 259
and [13C3]malonyl-CoA substrates (Fig. 4A). The ion m/z 270
13
[
C ]palmitate was also routinely detected, which was due to the
16
99 atomic% 13C of the two substrates. Also detected in the FASN
activity assay was unlabeled palmitate (m/z 255) that originated
from the stock of purified recombinant enzyme. The presence of
m/z 255 unlabeled palmitate ion was verified by extracting fatty
acids from the recombinant enzyme stock alone (data not shown).
[
13C4]palmitate standard (back peak) from 10 to 60 s (left and right
insets of Fig. 4B, respectively) is represented in extracted ion chro-
matograms. The NADPH oxidation method was then compared with
the direct quantification of [13C16]palmitate (Table 1). NADPH oxida-
tion was monitored continuously via the spectrophotometric meth-
od and FASN oxidized overall 1544 nmol NADPH minꢀ1 mgꢀ1 ( 137)
compared with the [13C16]palmitate at 102 nmol minꢀ1 mgꢀ1 ( 4.9)
synthesized by FASN in that set of experiments. Because 14 mole-
cules of NADPH are oxidized per molecule of [13C16]palmitate syn-
Here, 50-ll volumes were sampled every 10 s for 60 s, and the reac-
tion proceeded linearly and under steady-state kinetics (Fig. 4B).
Working reaction conditions were determined for optimal steady-
state kinetics of FASN by varying substrate concentrations over a
longer time series, and the best time frame and substrate
thesized, 102 ꢁ 14 gives
a converted NADPH oxidation of
Table 1
Comparison of NADPH oxidation with direct [13C]16-palmitate quantification.
Sample 1
Sample 2
Sample 3
Sample 4
Average
NADPH run 1
NADPH run 2
Overall average
Converted
13
1267.2
1632.0
1544 ( 137)
110.0
1459.2
1536.0
1546.0
1728.0
1612.8
1584.0
1469 ( 77)
1616 ( 100)
[
[
C
C
]palmitate 1
]palmitate 2
98.4
108.8
101.4
98.5
96.8
105.2
109.4
100.2
101 ( 5.6)
103 ( 5.22)
16
13
16
Overall average
102 ( 4.9)
Note. The [13C16]palmitate quantification method was compared with the NADPH oxidation method using recombinant FASN and identical reaction conditions outlined in
Materials and Methods. Oxidation of NADPH was continuously monitored on two different runs, and numbers are expressed as nmoles NADPH oxidized minꢀ1 mgꢀ1. In two
separate experiments, the [13C16]palmitate carboxylate anion (m/z 271) was monitored and quantified relative to the regression curve in Fig. 3. Specific activity of recom-
binant FASN using the NADPH method was 1544 nmol minꢀ1 mgꢀ1, and FASN synthesized [13C16]palmitate at 102 nmol minꢀ1 mgꢀ1. NADPH oxidation was converted to
nmoles palmitate by dividing 1544 by 14, which gave 110 nmol minꢀ1 mgꢀ1, because 14 molecules of NADPH are oxidized during the synthesis of one palmitate molecule.

Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
163
120
100
80
60
40
20
0
cell or tissue extracts is the amount of ‘‘contaminating enzymes’’
within the cytosolic fraction [28]. Contaminating enzyme activity
was generally avoided using complicated FASN purification meth-
ods such as anion exchange chromatography [29]. Added to the
difficulty of evaluating FASN activity from crude lysates was that
FASN protein amounts could not be quantified. Calculation of FASN
amount from crude lysates is necessary for several reasons: to en-
sure that the amount of enzyme from each sample is equivalent, to
be certain that the levels of FASN in the in vitro assay preserve
steady-state kinetics, and to calculate the specific activity based
on milligrams of FASN. Crude cytosolic lysate FASN from mammary
gland and liver tissues was quantified using an immunoblot ap-
proach relative to a standard regression curve generated with re-
combinant rat FASN (Fig. 6A). The anti-FASN antibody is a rabbit
polyclonal generated against a 299-amino acid region (aa 2205–
2504) at the C terminus of human FASN. There is 95% sequence
identity (283/299 amino acids) between human and mouse FASN
in this region, 94% identity (280/299 amino acids) between human
and rat FASN in this region, and 97% identity (296/304 amino acids)
between rat and mouse FASN, suggesting that the cross-reactivity
between rat and mouse epitopes is robust. Immunoblots were
quantified using the Licor Odyssey instrument, and each standard
No
Cerulenin
2% v/v No NADPH 1.0 μM
2.0 μM
10 μM
50 μM
100 μM
DMSO
Control
Cerulenin Treatment
Fig.5. Recombinant FASN activity assay performed according to conditions in
Materials and Methods. Samples were incubated with and without increasing
concentrations of cerulenin. The [13C16]palmitate carboxylate anion m/z 271 was
monitored and quantified relative to the regression curve in Fig. 3A. Specific activity
13
of recombinant FASN without cerulenin synthesized
[ C ]palmitate at
16
104 nmol minꢀ1 mgꢀ1 and [13C16]palmitate synthesis was significantly inhibited
by 20% (P = 0.0018) with 1 M cerulenin up to 98% with 100 M cerulenin (32–98%
samples, P 6 0.0005). DMSO (2%, v/v) had no effect on the reaction, and the
l
l
13
[
C ]palmitate carboxylate anion m/z 271 was not detected in the absence of
16
NADPH.
0
0.01 0.03
0.1
0.3
0.9
L
A
|
|
|
|
|
Recombinant
|
FASN in (μg)
1
1428 nmol minꢀ1 mgꢀ1. Thus, the two methods differ by 7.5% and
are in reasonable agreement when measuring the activity of recom-
binant FASN.
y = 0.0001x
R² = 0.99885
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
PFB derivatives of [13C16]palmitate from commercial sources
generated a carboxylate anion at m/z 271. Spectra from commer-
cial [13C16]palmitate compared with spectra from recombinant
FASN validated that [13C16]palmitate was a direct product of FASN
catalysis. In addition, exclusion of NADPH from the reaction re-
sulted in no m/z 271 carboxylate anion from [13C16]palmitate;
therefore, no FASN catalysis occurred (Fig. 5). [13C16]palmitate syn-
thesis was evaluated in the presence of the ‘‘suicide’’ FASN inhibi-
tor cerulenin. Cerulenin is known to irreversibly inhibit the b-
ketoacyl synthase condensing enzyme domain of type I fatty acid
synthases, and inhibitory concentrations are known to be directly
proportional to FAS inhibition [26,27]. The addition of DMSO (2%,
0
1000
2000
3000
4000
5000
6000
7000
8000
Background adjusted intensity / 1000
Tissue Type
Mouse
Mammary Gland
Liver
B
v/v), the solvent for cerulenin, did not alter the synthesis of
13
[
C ]palmitate, and nearly the same reaction rate was observed
16
L
A
B
C
D
a
b
c
d
compared with no cerulenin (Fig. 5). Cerulenin was added at vari-
ous concentrations to reactions, and reactions were carried out
according to Materials and Methods. [13C16]palmitate synthesis
FASN
was inhibited by 20% (P = 0.0018) with 1
pressed 88% by the addition of 50
( 1.5) nmol minꢀ1 mgꢀ1 relative to no cerulenin conditions at
104 nmol minꢀ1 mgꢀ1 (Fig. 5). The addition of 100
M cerulenin
l
M cerulenin and sup-
β-Tubulin
lM cerulenin at 12.8
Mouse
FAS in ng/μg cytosolic extract
l
resulted in near complete (98%) inhibition of [13C16]palmitate syn-
thesis (32–98% samples, P 6 0.0005). By using 13C-labeled acetyl-
CoA and malonyl-CoA incorporation to directly measure [13C16]pal-
mitate synthesis, this method was capable of investigating enzyme
inhibition in vitro with enough sensitivity to detect very subtle
changes in catalysis at lower than reported drug concentrations.
A
37.7
41.4
13.6
21.8
B
C
D
37.2
32.9
38.3
11.3
Mean (SEM)
38.6 (+/- 1.9)
19.8 (+/- 9.7)
A significant 20% inhibition was observed (P = 0.0018) at 1.0 lM
Fig.6. (A) Regression curve for the serial dilution series using known amounts of
recombinant FASN protein (in g). The dilution was performed in triplicate with a
correlation value of 0.99885. The slope of the dilution series was used to calculate
cerulenin. These data demonstrate a specific and sensitive method
l
to directly quantify FASN catalyzed synthesis of [13C16]palmitate.
the amount of FASN isolated from tissue extracts. (B) Calculation of FASN quantity
(in ng/lg) of cytosolic extract from mammary gland epithelium and from liver
FASN activity assay from tissue extracts
tissues. Values in the table are calculated from the signal intensity values quantified
in the immunoblot for all immunoreactive bands shown. These signal intensity
values are relative to the recombinant FASN regression curve in panel A and are
calculated according to the equation in Materials and Methods. L denotes the
marker ladder for the immunoblot.
Of great interest is to evaluate the FASN activity directly from
crude lysates of a wide variety of tissues and cell lines under nor-
mal and pathological conditions. A critical aspect when using crude

164
Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
curve dilution point was evaluated in triplicate. A regression line
was generated with a slope of 0.0001 and an R2 correlation value
of 0.99885. Next, crude cytosolic fractions were prepared from
mammary gland epithelial cells (MECs), and matched livers of
the same animals as described in Materials and Methods and
amounts of crude cytosolic lysate FASN were calculated using the
(Fig. 7). A large quantity of preexisting unlabeled palmitate derived
from the cytosolic fractions of the lysates was detected (m/z 255
ion, shown), as were unsaturated fatty acids 18:1 m/z 281 and
18:2 m/z 279 that did not originate from FASN catalysis in the assay
(data not shown). Detection of preexisting fatty acids in the lysates
emphasized the importance of supplying uniformly 13C-labeled
substrates in the crude FASN assay reaction because variable levels
of unlabeled palmitate would confound palmitate detection using
unlabeled substrates. The ion m/z 270 resulted from the 99 atom%
13C isotopic purity of the substrates, as was observed from the re-
combinant FASN assays. The specific activity of FASN from the
crude cytosolic MECs and the liver lysates was calculated
recombinant FASN regression curve. When standardized to 60 lg
of total cytosolic fraction protein, crude lysate FASN was quantified
using the signal intensity values from all immunoreactive bands
shown (Fig. 6B). FASN amounts were consistent from the MECs
but variable from the liver. b-Tubulin indicated cytosolic protein
in each well but was not used for any quantification or normaliza-
tion purposes. Crude lysate FASN from MECs and liver FASN was
(Fig. 7B). Equalized amounts (2 lg) of MECs and liver FASN synthe-
estimated at 38.6 ( 1.9) and 19.8 ( 9.7) ng/
l
g cytosolic protein,
sized [13C16]palmitate at 82 and 65 nmol minꢀ1 mgꢀ1 FASN, respec-
tively. The specific activity of FASN from in vivo samples was 20 to
35% less than that observed for recombinant FASN. Although the
amount of FASN from the crude cytosolic extracts was equal, the
mammary gland FASN had significantly higher specific activity
(1.26-fold, P = 0.02) compared with liver FASN, a trend that was ob-
served in three independent experiments. This observation sug-
gested that the 1.26-fold difference might be due to factors other
respectively. Using this strategy to quantify endogenous FASN in
each sample (table in Fig. 6B), a known FASN amount was added
to the following activity assays.
FASN from crude cytosolic extracts of both tissues synthesized
13
[
C
16
]palmitate (m/z 271) as the primary reaction product
[13C16]-palmitate
255.2
100
A
analyte
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
than overall enzyme abundance. Finally, the addition of 100
lM
cerulenin inhibited FASN catalysis by more than 93% (Fig. 7B),
which validated that the presence of the [13C16]palmitate m/z 271
carboxylate anion in reaction using crude lysates was due to FASN.
Unlabeled
palmitate
271.3
Discussion
FASN has been studied by biochemists for decades to elucidate
enzyme mechanism and reaction kinetics and to determine the
requirements for enzyme catalysis. Here, a new method to directly
quantify [13C16]palmitate synthesis that uses the incorporation of
[13C4]-palmitate
standard
high-purity, nonradioactive heavy atom
[
13C2]acetyl-CoA and
[
13C3]malonyl-CoA substrates has been described (Fig. 1). This
256.2
method detected carboxylate anions of [13C16]palmitate PFB deriv-
270.3
atives using GC–MS. GC–MS simultaneously measured both the
259.2
13
0
[ C
]palmitate analyte and the [13C4]palmitate internal standard
16
250 252 254 256 258 260 262 264 266 268 270 272 274
m/z
at the same retention time to provide greater precision than nonin-
ternal standards [30]. The finding that the primary product of FASN
is saturated palmitate was validated using this method by compar-
ing the m/z 271 ion of [13C16]palmitate with purchased [13C16]pal-
mitate. GC–MS directly quantifies de novo fatty acid synthesis
in vitro, and the specific activity for both recombinant FASN and tis-
sue-derived FASN was determined. This new method integrates the
use of [13C2]acetyl-CoA and [13C3]malonyl-CoA heavy atom tracers
with the power of GC–MS to provide a very sensitive assay that de-
tects FASN catalyzed de novo synthesized fatty acids at the femto-
mole level. Most important, this method is not confounded by the
presence of unlabeled palmitate that often contaminates solvents
and blanks because it measures [13C16]palmitate synthesis from
uniformly 13C-labeled acetyl- and malonyl-CoA esters.
100
B
P = 0.02
MECs
Liver
80
60
40
20
0
The direct quantification of [13C16]palmitate was compared with
the NADPH oxidation method using recombinant FASN and identical
reaction conditions according to Materials and Methods. The values
observed in Table 1 are consistent with previous reports of FASN activ-
ity calculated using the NADPH oxidation method that indicated a
broad range of values between 1500 and 2000 nmol minꢀ1 mgꢀ1
[24,29,31,32]. In our hands, the NADPH oxidation method gave
1544 nmol minꢀ1 mgꢀ1. The direct [13C16]palmitate quantification
method was consistent with, but slightly lower than, reported results
at approximately 1428 nmol NADPH oxidized minꢀ1 mgꢀ1 (converted
by multiplying 102 nmol minꢀ1 mgꢀ1 times 14). The [13C16]palmitate
quantification value is slightly lower, presumably because we moni-
tored the abundance of only the [13C16]palmitate carboxylate anion
at m/z 271. To that end, other FASN products are synthesized in the
> 93% inhibition
14:0
16:0
18:0
14:0
16:0
18:0
No Cerulenin
100 μM Cerulenin
Fig.7. (A) Extracted ion spectrum of the FASN activity assay using tissue-derived
FASN. The primary reaction product is uniformly incorporated [13C16]palmitate (m/z
271). Unlabeled palmitate (m/z 255) was detected in the spectrum from the crude
cytosolic extract, and [13C4]palmitate (m/z 259) is the internal standard. (B) FASN
activity assay using the volume of crude cytosolic extracts that corresponded to
2
l
g of total FASN from mammary epithelial cells and livers of the same mice. A
1.26-fold (P = 0.02) significant decrease in specific activity was observed by liver
13
FASN for
[ C ]palmitate synthesis. Reactions were conducted according to
16
Materials and Methods without and with 100
lM cerulenin, inhibiting FASN
activity greater than 93%. Data plotted were quantified by monitoring [13C14]myr-
istate (m/z 241), [13C16]palmitate (m/z 271), and [13C18]stearate (m/z 301) ions.

Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
165
reaction, including myristic and stearic acids and short acyl ‘‘by-prod-
ucts’’ of FASN such as butyryl (4:0) and crotonyl (4:1) CoA esters [33].
With respect to palmitate synthesis, these additional FASN products all
consume NADPH over the course of the reaction. Because oxidation of
NADPH was used to calculate FASN catalysis, the production of these
minor products alters calculations of FASN activity from NADPH oxida-
tion during continuous 340-nM monitoring of the aqueous reaction
mixture using the spectrophotometric method. This observation sug-
gested that the difference in calculation of specific activity could be a
result of the methods used rather than due to variations in catalysis
of recombinant FASN itself.
Previous methods to quantify fatty acids typically used radio-
active materials and liquid scintillation counting to measure fatty
acid synthesis. Although this procedure successfully measures the
products of FASN catalysis, critical information regarding fatty
acid chain length cannot be determined using this approach. Con-
ventional GC is useful to discriminate among fatty acids of varied
chain length [4]; however, heavy atom tracer information would
be lost using standard flame ionization. Radiolabeled GC was used
to quantify fatty acids of various chain lengths by retention time,
and it retains radioactive tracer content; however, quantification
of each fatty acid is accomplished using reference standards that
are unrelated to the fatty acid of interest [34,35]. Radiolabeled
HPLC methods also exist to resolve and detect discrete fatty acid
chain lengths using flow-through scintillation detectors [33,36].
Although the sensitivity of radioassays is quite good, the accuracy
is often poor [20]. Both liquid scintillation counting and radio
chromatography techniques use radioactive materials that are of-
ten expensive, require special handling, and cause difficulty for
waste disposal due to the generation of hazardous mixed wastes.
This method directly quantified FASN activity using [13C2]acetyl-
CoA and [13C3]malonyl-CoA incorporation to avoid the undesir-
able issues with radioisotope methods.
Researchers have routinely isolated FASN from mammary tis-
sues and liver, and some have purified FASN from tumors and tu-
mor cell lines to investigate enzyme activity [18,37,38]. However,
calculating FASN concentrations from crude lysates of biological
samples has been difficult. The addition of known FASN amounts
to the reaction is essential to ensure that the analysis is conducted
under steady-state conditions. Here, we have described an immu-
noblotting approach to calculate the amount of FASN, eliminating
the need for enzyme purification that typically abolishes protein–
protein interactions that may be important in understanding the
cell-type specific biology. To that end, crude tissue lysates are
known to contain ‘‘interfering enzymes’’ that could consume
NADPH or decarboxylate malonyl-CoA or could alter acetyl-CoA
levels [20,28,39]. For example, the presence of interfering enzymes
might affect calculations of FASN activity via NADPH oxidation if
NADPH is oxidized by factors other than FASN. In addition, decar-
boxylation of malonyl-CoA by malonyl-CoA decarboxylase in the
crude lysate results in acetyl-CoA, which would alter the levels of
malonyl-CoA that are needed for acyl elongation by FASN. Con-
versely, this method supplied micromolar concentrations of
[
13C2]acetyl-CoA, [13C3]malonyl-CoA, and NADPH, which insulated
FASN catalysis against interfering enzymes present in crude ly-
sates. Therefore, the method outlined is suitable to conduct FASN
assays using tissue-derived crude cytosolic fractions.
FASN protein from crude in vivo extracts was quantified to en-
able calculation of the specific activity based on milligrams of FASN
added to the reaction. The mammary gland samples contained sev-
eral minor bands detected in the immunoblot but not present in
the liver samples. It is not clear whether these minor products rep-
resent a catalytically functional FASN or if they are breakdown
products of nonfunctional FASN. The specific activity of FASN from
the crude extracts was 20% to 35% less than that observed for re-
combinant FASN. It is possible that the immunoblotting approach
used to quantify mouse FASN in crude cytosolic extracts, relative
to a rat FASN regression curve, introduced error into the calcula-
tion of specific activity for the mouse enzyme. However, the epi-
tope of the anti-FASN antibody for mouse and rat was highly
similar (97% identical) and is unlikely to account for the 20 to
35% difference in activity. This means that the trend observed be-
tween mouse tissue extracts would not be altered, even though the
‘‘absolute’’ amount of total FASN could be variable. The difference
between mouse crude extract FASN and the affinity-purified rat
FASN could be species specific; however, column-purified FASN
from species as divergent as rat and chicken had nearly identical
activities at 1500 to 2000 and 1600 nmol NADPH oxidized
minꢀ1 mgꢀ1, respectively [24,29]. On the other hand, the 20% to
35% difference was consistent with the observations that com-
pared recombinant FASN with tissue-derived FASN, where the tis-
sue-derived FASN had lower activity than the purified recombinant
FASN [29]. We speculate that the difference in activity between the
mouse and the rat synthases is most likely due to the type of FASN
preparation used, crude cytosolic extract versus affinity-purified
recombinant FASN.
Importantly, calculating the specific activity of FASN in mouse
tissues based on the immunoblotting quantification of FASN made
it possible to discern whether factors other than FASN amount
influence the enzyme’s activity under steady-state conditions.
The novel observation that the specific activity of FASN from lactat-
ing mammary epithelial cells is significantly higher (1.26-fold,
P = 0.02) when compared with FASN from liver suggested that
FASN activity may be modified in a tissue-specific manner. It is
tempting to speculate that the observed increase between the
MECs and liver-specific activity of FASN resulted from the presence
of mammary-specific factors that complex with FASN. Alterna-
tively, this increase may be due to the presence of tissue-specific
fatty acid binding proteins. Binding of de novo synthesized fatty
Two methods were reported for either deuterium or [13C1]mal-
onyl-CoA incorporation into labeled palmitate using a similar
in vitro FASN activity assay and GC–MS [19,20]. Monitoring deute-
rium incorporation was not desirable due to isotope effects of deu-
terium in the FASN assay [13,19]. Alternatively, Ohashi and
coworkers compared the NADPH method with [13C1]malonyl-CoA
incorporation into 13C-labeled de novo fatty acids synthesized by
column-purified FASN and FASN from crude cytosolic extracts
[20]. A major issue with this study was the isotopic purity of the
[
13C1]malonyl-CoA used in the FASN assay; [13C1]malonyl-CoA
was synthesized and purified in the lab at 92 atom% 13C isotopic
purity with the label at the central methylene carbon of malo-
nyl-CoA. Using lower purity
[
13C1]malonyl-CoA, Ohashi and
coworkers detected 13C6-, 13C7-, and 13C8-incorporated palmitate
requiring summation of these ions [20]. Conversely, commercial
sources currently exist to provide 99 atom% 13C isotopic purity
of [13C2]acetyl-CoA and [13C3]malonyl-CoA substrates, which en-
ables synthesis of completely incorporated [13C16]palmitate (m/z
271) and only minor amounts of the [13C16]palmitate carboxylate
anion (m/z 270) (610%). Another improvement from the method
reported by Ohashi and coworkers, who measured fatty acid
methyl esters and McLafferty rearrangement ion m/z 74, is the
quantitation of carboxylate anions of nonfragmented fatty acids
[20]. PFB-Br/DIEA derivatization of [13C16]palmitate generated car-
boxylate anions of fatty acids that were robustly detected by the
mass spectrometer, resulting in an excellent signal-to-noise ratio.
The derivatization strategy for this method was sensitive in the
low femtomole range, which was far superior to either the NADPH
oxidation method or the quantitation of fatty acid methyl esters.
Finally, the previous [13C1]malonyl-CoA used nanomolar concen-
trations of substrates over a 30-min reaction, which did not mea-
sure fatty acid synthesis reaction under steady-state conditions
from either column-purified FASN or crude extracts.

166
Fatty acid synthase activity from tissue lysates / M.C. Rudolph et al. / Anal. Biochem. 428 (2012) 158–166
[9] J.A. Menendez, R. Lupu, Fatty acid synthase and the lipogenic phenotype in
cancer pathogenesis, Nat. Rev. Cancer 7 (2007) 763–777.
acids to binding proteins could enhance catalysis by relieving
product inhibition of FASN. Alternatively, protein activators of
FASN, such as 40-phosphopantetheine transferase [40], might exist
at different amounts in the two tissues, thereby increasing pantet-
heinyl activation of the mammary gland FASN. In the end, it is clear
that this method can be directly applied to assays with recombi-
nant FASN or column-purified FASN or to partially enriched FASN
present in a cell or tissue cytosolic lysate.
[10] J.A. Menendez, Fine-tuning the lipogenic/lipolytic balance to optimize the
metabolic requirements of cancer cell growth: molecular mechanisms and
therapeutic perspectives, Biochim. Biophys. Acta 2010 (1801) 381–391.
[11] F.P. Kuhajda, Fatty-acid synthase and human cancer: new perspectives on its
role in tumor biology, Nutrition 16 (2000) 202–208.
[12] V.E. Anderson, G.G. Hammes, Stereochemistry of the reactions catalyzed by
chicken liver fatty acid synthase, Biochemistry 23 (1984) 2088–2094.
[13] Y. Seyama, T. Kasama, T. Yamakawa, A. Kawaguchi, K. Saito, Origin of hydrogen
atoms in the fatty acids synthesized with yeast fatty acid synthetase, J.
Biochem. 82 (1977) 1325–1329.
In conclusion, we have described a new method that evaluates
FASN activity based on the detection and quantitative measure-
[14] B.L. Horecker, A. Kornberg, The extinction coefficients of the reduced band of
pyridine nucleotides, J. Biol. Chem. 175 (1948) 385–390.
ment of enzyme product formation—de novo synthesized
[15] C.C. Chung, K. Ohwaki, J.E. Schneeweis, E. Stec, J.P. Varnerin, P.N. Goudreau, A.
Chang, J. Cassaday, L. Yang, T. Yamakawa, O. Kornienko, P. Hodder, J. Inglese, M.
Ferrer, B. Strulovici, J. Kusunoki, M.R. Tota, T. Takagi, A fluorescence-based
thiol quantification assay for ultra-high-throughput screening for inhibitors of
coenzyme A production, Assay Drug Dev. Technol. 6 (2008) 361–374.
[16] D. Cerne, I.P. Zitnik, M. Sok, Increased fatty acid synthase activity in non-small
cell lung cancer tissue is a weaker predictor of shorter patient survival than
increased lipoprotein lipase activity, Arch. Med. Res. 41 (2010) 405–409.
[17] J.M. Soulie, G.J. Sheplock, W.X. Tian, R.Y. Hsu, Transient kinetic studies of fatty
acid synthetase: a kinetic self-editing mechanism for the loading of acetyl and
malonyl residues and the role of coenzyme A, J. Biol. Chem. 259 (1984) 134–140.
[18] D.A. Roncari, Fatty acid synthase from human liver, Methods Enzymol. 71
(1981) 73–79.
[19] Y. Seyama, A. Kawaguchi, S. Okuda, T. Yamakawa, New assay method for fatty
acid synthetase with mass fragmentography, J. Biochem. 84 (1978) 1309–1314.
[20] K. Ohashi, H. Otsuka, Y. Seyama, Assay of fatty acid synthetase by mass
fragmentography using [13C]malonyl-CoA, J. Biochem. 97 (1985) 867–875.
[21] S. Zarini, M.A. Gijon, G. Folco, R.C. Murphy, Effect of arachidonic acid
reacylation on leukotriene biosynthesis in human neutrophils stimulated
with granulocyte-macrophage colony-stimulating factor and formyl-
methionyl-leucyl-phenylalanine, J. Biol. Chem. 281 (2006) 10134–10142.
[22] A.K. Joshi, V.S. Rangan, S. Smith, Differential affinity labeling of the two
subunits of the homodimeric animal fatty acid synthase allows isolation of
heterodimers consisting of subunits that have been independently modified, J.
Biol. Chem. 273 (1998) 4937–4943.
13
[
C ]palmitate using GC–MS. This method can be applied to tis-
16
sues from all mammalian species, and it provides the opportunity
to directly compare tissue preparations because FASN amounts
are determined. FASN-specific activity is reliably measured in tis-
sues under different metabolic conditions or various physiological
and diseased states, and [13C16]palmitate synthesis can now be pre-
cisely measured from crude extracts without complication from
unlabeled fatty acids present in the lysates. FASN is generally seen
as an indicator of aggressive tumors and poor patient outcome
[9,11]; therefore, this assay could be useful to demonstrate a direct
link between FASN activity and disease prognosis. Moreover, this
method can be used to investigate the effects of FASN modifying
proteins, inhibitory peptides, and small molecule inhibitors. Finally,
important kinetic information will likely be obtained about FASN
catalysis by direct quantitative measurement enzyme product. Sen-
sitive quantification of FASN catalysis will add to the understanding
of this complex multicomponent process, which is remarkably con-
tained entirely within the single enzyme fatty acid synthase.
[23] M.C. Rudolph, E.A.Wellberg, S.M. Anderson, Adipose-depleted mammaryepithelial
cells and organoids, J. Mammary Gland Biol. Neoplasia 14 (2009) 381–386.
[24] B.G. Cox, G.G. Hammes, Steady-state kinetic study of fatty acid synthase from
chicken liver, Proc. Natl. Acad. Sci. USA 80 (1983) 4233–4237.
[25] S. Smith, S. Abraham, Fatty acid synthetase from lactating rat mammary gland:
3. Dissociation and reassociation, J. Biol. Chem. 246 (1971) 6428–6435.
[26] S. Omura, The antibiotic cerulenin, a novel tool for biochemistry as an inhibitor
of fatty acid synthesis, Bacteriol. Rev. 40 (1976) 681–697.
[27] M. Moche, G. Schneider, P. Edwards, K. Dehesh, Y. Lindqvist, Structure of the
complex between the antibiotic cerulenin and its target, b-ketoacyl-acyl
carrier protein synthase, J. Biol. Chem. 274 (1999) 6031–6034.
[28] R. Bressler, S. Wakil, The study offatty acid synthesis: IX. Theconversion ofmalonyl
coenzyme A to long chain fatty acids, J. Biol. Chem. 236 (1961) 1643–1651.
[29] A.K. Joshi, S. Smith, Construction of a cDNA encoding the multifunctional
animal fatty acid synthase and expression in Spodoptera frugiperda cells using
baculoviral vectors, Biochem. J. 296 (1993) 143–149.
Acknowledgments
The authors thank S. Smith and A. Witkowski (Children’s Hospi-
tal Oakland Research Institute) for the generosity of their recombi-
nant FASN needed to conduct this research as well as for their
thoughtful discussions regarding this work. The authors acknowl-
edge support from the U.S. Department of Defense Breast Cancer
Research Program (BC810596 to M.C.R. and BC098051 to E.A.W.),
the Colorado Clinical and Translational Sciences Institute
(5UL1RR025780 to R.C.M.), a Lipid MAPS Collaborative Grant from
the National Institutes of Health (NIH, GM069338 to R.C.M.), and
the NIH (PO1-HD38129 to S.M.A.).
[30] R.C. Murphy, S.J. Gaskell, New applications of mass spectrometry in lipid
analysis, J. Biol. Chem. 286 (2011) 25427–25433.
[31] S. Smith, S. Abraham, Fatty acid synthase from lactating rat mammary gland,
Methods Enzymol. 35 (1975) 65–74.
References
[1] S.S. Chirala, H. Chang, M. Matzuk, L. Abu-Elheiga, J. Mao, K. Mahon, M. Finegold,
S.J. Wakil, Fatty acid synthesis is essential in embryonic development: fatty
acid synthase null mutants and most of the heterozygotes die in utero, Proc.
Natl. Acad. Sci. USA 100 (2003) 6358–6363.
[2] A. Jayakumar, M.H. Tai, W.Y. Huang, W. al-Feel, M. Hsu, L. Abu-Elheiga, S.S.
Chirala, S.J. Wakil, Human fatty acid synthase: Properties and molecular
cloning, Proc. Natl. Acad. Sci. USA 92 (1995) 8695–8699.
[32] N. Singh, S.J. Wakil, J.K. Stoops, On the question of half- or full-site reactivity of
animal fatty acid synthetase, J. Biol. Chem. 259 (1984) 3605–3611.
[33] A. Witkowski, A.K. Joshi, S. Smith, Characterization of the b-carbon processing
reactions of the mammalian cytosolic fatty acid synthase: role of the central
core, Biochemistry 43 (2004) 10458–10466.
[34] E. Agradi, L. Libertini, S. Smith, Specific modification of fatty acid synthetase
from lactating rat mammary gland by chymotrypsin and trypsin, Biochem.
Biophys. Res. Commun. 68 (1976) 894–900.
[3] S. Smith, A. Witkowski, A.K. Joshi, Structural and functional organization of the
animal fatty acid synthase, Prog. Lipid Res. 42 (2003) 289–317.
[35] L.J. Libertini, S. Smith, Purification and properties of
a thioesterase from
[4] M.C. Rudolph, J. Monks, V. Burns, M. Phistry, R. Marians, M.R. Foote, D.E. Bauman,
S.M. Anderson, M.C. Neville, Sterol regulatory element binding protein (Srebf-1)
and dietary lipid regulation of fatty acid synthesis in the mammary epithelium,
Am. J. Physiol. Endocrinol. Metab. 299 (2010) E918–E927.
[5] S.A. Kliewer, S.S. Sundseth, S.A. Jones, P.J. Brown, G.B. Wisely, C.S. Koble, P.
Devchand, W. Wahli, T.M. Willson, J.M. Lenhard, J.M. Lehmann, Fatty acids and
eicosanoids regulate gene expression through direct interactions with
peroxisome proliferator-activated receptors alpha and gamma, Proc. Natl.
Acad. Sci. USA 94 (1997) 4318–4323.
lactating rat mammary gland which modifies the product specificity of fatty
acid synthetase, J. Biol. Chem. 253 (1978) 1393–1401.
[36] J. Naggert, M.L. Narasimhan, L. DeVeaux, H. Cho, Z.I. Randhawa, J.E. Cronan Jr.,
B.N. Green, S. Smith, Cloning, sequencing, and characterization of Escherichia
coli thioesterase (part II), J. Biol. Chem. 266 (1991) 11044–11050.
[37] F.P. Kuhajda, K. Jenner, F.D. Wood, R.A. Hennigar, L.B. Jacobs, J.D. Dick, G.R.
Pasternack, Fatty acid synthesis: a potential selective target for antineoplastic
therapy, Proc. Natl. Acad. Sci. USA 91 (1994) 6379–6383.
[38] P.M. Ahmad, D.S. Feltman, F. Ahmad, Studies on acetyl-CoA carboxylase and
fatty acid synthase from rat mammary gland and mammary tumours,
Biochem. J. 208 (1982) 443–452.
[6] J. Ross, A.M. Najjar, M. Sankaranarayanapillai, W.P. Tong, K. Kaluarachchi, S.M.
Ronen, Fatty acid synthase inhibition results in
a magnetic resonance-
detectable drop in phosphocholine, Mol. Cancer Ther. 7 (2008) 2556–2565.
[7] V. Rioux, P. Legrand, Saturated fatty acids: simple molecular structures with
complex cellular functions, Curr. Opin. Clin. Nutr. Metab. Care 10 (2007) 752–
758.
[8] H. Cao, K. Gerhold, J.R. Mayers, M.M. Wiest, S.M. Watkins, G.S. Hotamisligil,
Identification of a lipokine, a lipid hormone linking adipose tissue to systemic
metabolism, Cell 134 (2008) 933–944.
[39] X. Wang, W.C. Stanley, C.J. Darrow, H. Brunengraber, T. Kasumov, Assay of the
activity of malonyl-coenzyme A decarboxylase by gas chromatography–mass
spectrometry, Anal. Biochem. 363 (2007) 169–174.
[40] A.K. Joshi, L. Zhang, V.S. Rangan, S. Smith, Cloning, expression, and
characterization of a human 40-phosphopantetheinyl transferase with broad
substrate specificity, J. Biol. Chem. 278 (2003) 33142–33149.