Simultaneous determination of 2-arachidonoylglycerol, 1- arachidonoylglycerol and arachidonic acid in mouse brain tissue using liquid chromatography/tandem mass spectrometry

Article abstract of DOI:10.1002/jms.1701

Endocannabinoids (ECs), such as anandamide (AEA) and 2-arachidonoylglycerol (2-AG), modulate a number of physiological processes, including pain, appetite and emotional state. Levels of ECs are tightly controlled by enzymatic biosynthesis and degradation in vivo. However, there is limited knowledge about the enzymes that terminate signaling of the major brain EC, 2-AG. Identification and quantification of 2-AG, 1-AG and arachidonic acid (AA) is important for studying the enzymatic hydrolysis of 2-AG. We have developed a sensitive and specific quantification method for simultaneous determination of 2-AG, 1-AG and AA from mouse brain and adipose tissues by liquid chromatography/tandem mass spectrometry (LC/MS/MS) using a simple brain sample preparationmethod. The separations were carried out based on reversed phase chromatography. Optimization of electrospray ionization conditions established the limits of detection (S/N=3) at 50, 25 and 65 fmol for 2-AG, 1-AG and AA, respectively. The methodswere selective, precise (%R.S.D.<10%) and sensitive over a range of 0.02-20, 0.01-10 and 0.05-50 ng/mg tissue for 2-AG, 1-AG and AA, respectively. The quantificationmethod was validated with consideration of thematrix effects and the mass spectrometry (MS) responses of the analytes and the deuterium labeled internal standard (IS). The developed methods were applied to study the hydrolysis of 2-AG from mouse brain extracts containingmembrane bound monoacylglycerol lipase (MAGL), and to measure the basal levels of 2-AG, 1-AG and AA in mouse brain and adipose tissues. Copyright

Full text of DOI:10.1002/jms.1701

Research Article
Received: 28 July 2009
Accepted: 30 October 2009
Published online in Wiley Interscience: 30 November 2009
(www.interscience.com) DOI 10.1002/jms.1701
Simultaneous determination of 2-
arachidonoylglycerol, 1-arachidonoylglycerol
and arachidonic acid in mouse brain tissue
using liquid chromatography/tandem mass
spectrometry
Mei-Yi Zhang,a∗ Ying Gao,b Joan Btesh,b Natasha Kagan,a Edward Kerns,a
Tarek A. Samadb and Pranab K. Chandab
Endocannabinoids (ECs), such as anandamide (AEA) and 2-arachidonoylglycerol (2-AG), modulate a number of physiological
processes, including pain, appetite and emotional state. Levels of ECs are tightly controlled by enzymatic biosynthesis and
degradation in vivo. However, there is limited knowledge about the enzymes that terminate signaling of the major brain
EC, 2-AG. Identification and quantification of 2-AG, 1-AG and arachidonic acid (AA) is important for studying the enzymatic
hydrolysis of 2-AG. We have developed a sensitive and specific quantification method for simultaneous determination of
2-AG, 1-AG and AA from mouse brain and adipose tissues by liquid chromatography/tandem mass spectrometry (LC/MS/MS)
using a simple brain sample preparation method. The separations were carried out based on reversed phase chromatography.
Optimization of electrospray ionization conditions established the limits of detection (S/N=3) at 50, 25 and 65 fmol for 2-AG,
1-AG and AA, respectively. The methods were selective, precise (%R.S.D. <10%) and sensitive over a range of 0.02–20, 0.01–10
and 0.05–50 ng/mg tissue for 2-AG, 1-AG and AA, respectively. The quantification method was validated with consideration of
the matrix effects and the mass spectrometry (MS) responses of the analytes and the deuterium labeled internal standard (IS).
The developed methods were applied to study the hydrolysis of 2-AG from mouse brain extracts containing membrane bound
monoacylglycerol lipase (MAGL), and to measure the basal levels of 2-AG, 1-AG and AA in mouse brain and adipose tissues.
c
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
Keywords: 2-arachidonoylglycerol; 2-arachidonoylglycerol hydrolysis; quantification; liquid chromatography; tandem mass spectrom-
etry; brain sample preparation
Introduction
tributor in this process. Using a functional proteomic approach,
Blankman et al.[24] showed that ∼85% of brain 2-AG hydrolase ac-
tivity can be attributed to MAGL, and the remaining 15% is mostly
catalyzedbyABHD6andABHD12.Accumulatedevidencesuggests
that inhibition of specific EC hydrolyzing enzymes might be useful
as a promising therapeutic approach to relieve discrete symptoms
withoutproducing sideeffects.[25] Functionalstudiesinvolving the
physiological roles of ECs have been mainly focused on AEA and
2-AG.[18,26,27] Development of reliable extraction and quantifi-
cation methods to accurately measure the levels of ECs from
biological tissue samples has been a growing interest in many
areas of neuroscience research to facilitate studies of the signally
mechanism and to evaluate the potential inhibitors of specific EC
biosynthesizing or hydrolyzing enzyme.
Endocannabinoids (ECs), such as anandamide (AEA) and
2-arachidonoylglycerol (2-AG), modulate a number of physio-
logical and pathological processes, including pain, schizophrenia,
stroke, obesity, Alzheimer’s disease, multiple sclerosis, cancer, ap-
petite, neural development, immune cell activation, retrograde
neural signaling and emotional state.[1–11] 2-AG acts as a full
agonist for both CB1 and CB2 cannabinoid receptors.[12–16] In
contrast, AEA acts as a partial agonist at the CB receptors.[15–17]
The endogenous levels of 2-AG present in rat brain were found
at more than 170- to 1000-fold higher than AEA.[17,18] The ex-
tent of EC accumulation in tissue and corresponding receptor
activation is tightly controlled by enzymatic biosynthesis and
degradation in vivo. The hydrolysis of AEA leading to the produc-
tion of arachidonic acid (AA) and ethanolamine is primarily carried
out by a single enzyme fatty acid amide hydrolase (FAAH) and
is well characterized.[19,20] Multiple enzymes, including FAAH,[21]
neuropathy target esterase (NTE),[22] and monoacylglycerol lipase
(MAGL),[23] have been shown to degrade 2-AG in vitro to AA and
glycerol. However, there is limited knowledge about the enzymes
that terminate signaling of the major brain EC, 2-AG, in the brain,
although MAGL has generally been assumed to be the main con-
∗
Correspondenceto:Mei-Yi Zhang,ChemicalSciences,WyethResearch,CN8000,
Princeton, NJ 08543, USA. E-mail: zhangm@wyeth.com
a
Chemical Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543, USA
Discovery Neurosciences, Wyeth Research, CN 8000, Princeton, NJ 08543, USA
b
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.

M.-Y. Zhang et al.
Theextractionof2-AGfrombraintissuereportedintheliterature
was primarily based on solvent-based extraction, which requires
further time consuming sample preparation procedures, including
chemical derivatization for gas chromatography-mass spectrom-
etry (GC/MS),[7,28,29] and solid phase extraction (SPE) for liquid
chromatography-massspectrometry(LC/MSorLC/MS/MS).[5,30–32]
Selected ion monitoring (SIM) mode was performed in the GC/MS
or LC/MS methods to measure 2-AG, identified based on a deriva-
tive fragment ion or on the molecular ion. Kingsley et al. were the
first to describe the use of a LC/MS/MS method, based on the
high specificity of multiple reaction monitoring (MRM) mode, for
analyzing 2-AG via a silver adduct.[31] Many reported methods in
the literature for quantification of 2-AG did not include valida-
tion data. Richardson et al. described validation of a LC/MS/MS
method for the simultaneous measurement of AEA, 2-AG and
other related compounds.[32] However, the ion suppression and
the MS response of the endogenous analytes and their internal
standards (ISs) in biological matrixes were not discussed. In ad-
dition, the reported measurements of 2-AG were based on the
combined concentrations of 2-AG and 1-AG, in which they were
not chromatographically separated.
An analytical technique was previously reported for simulta-
neous measurement of 2-AG, 1-AG and AA, based on the use
of high-performance liquid chromatography (HPLC) coupled with
ultraviolet (UV) detection.[33] The detection of these analytes re-
lies only on their chromatographic retention times (RTs), without
method validation.
Coupling MS with chromatographic separation techniques
provides the advantage of detecting these analytes by both
their chromatographic RTs and the corresponding molecular ions.
The MRM scan mode on a triple quadrupole mass spectrometer
monitors specific product ions from dissociation of their parent
molecular ions. This approach, combined with the specific LC RT
of analytes, provides a sensitive and selective analysis, which is
unique for individual analytes.
Preparation of membrane extracts from mouse brain
The brain tissue was harvested from male C57BL/6J mice (Charles
River Laboratories International, Inc., Wilmington, MA, USA), and
immediately kept frozen in dry ice. Frozen tissues were stored at
−80 ◦C. For membrane-bound preparation of MAGL, a Dounce
homogenization was performed in Tris buffer (50 mM Tris-HCl, pH
7.5), containing protease inhibitors and 150 mM NaCl. Following
brief sonication, a low-speed centrifugation was performed at 4 ◦C
to remove cell debris. The supernatant was centrifuged at high
speed for 40 min at 4 ◦C in order to isolate membrane-bound
MAGL. This high-speed centrifugation was repeated twice. The
final pellet was re-suspended in Tris buffer, sonicated and kept at
−80 ◦C until use.
Enzymatic activity assays
2-AG (50 µM) was incubated in a buffer consisting of 54 mM
HCl, 1.1 mM ethylenediaminetetraacetic acid (EDTA), 100 mM
NaCl, 5 mM MgCl2, 1 mM dithiothreitol (DTT) and 0.5% bovine
serum albumin (BSA) (pH 7.4) using either recombinant MAGL
(rMAGL) protein (expressed and purified from Escherichia coli)
(unpublished) or membrane-bound MAGL (5 µg membrane
protein) as discussed above. At time points of 0, 10, 30, 60, 90 and
120 min, 100 µl of sample was removed from the incubation and
200 µl of ACN (pH 3.0) was added to stop the enzymatic reaction.
The pH of the sample was adjusted to 3.0 using formic acid to
stabilize 2-AG against a possible post-incubation chemical acyl
migration reaction to 1-AG. Samples were centrifuged. Twenty
micro-litter of 2-AG-d8 (IS) solution was added into an aliquot
of 80 µl of supernatants. The samples were kept at −80 ◦C
until LC/MS/MS analysis. The degradation (both enzymatic and
chemical) of 2-AG was determined by simultaneously monitoring
2-AG, 1-AG and AA using LC/MS/MS.
Mouse brain and adipose tissue sample preparation
Herewereportthedevelopmentandvalidationofasimplebrain
andadiposetissuesamplepreparationmethodinconjunctionwith
a sensitive and selective quantitative method for simultaneous
determination of 2-AG, 1-AG and AA from mouse brain tissues
using LC/MS/MS. This rapid and cost-effective sample preparation
method consisted of a simple protein precipitation step by using
acetonitrile (ACN). The HPLC separation was carried out using
reversed phase chromatography. The developed and validated
methods were applied to study the hydrolysis of 2-AG from
mouse brain extracts containing membrane-bound MAGL, and to
measure the basal levels of 2-AG, 1-AG and AA in mouse brain and
adipose tissues.
Male C57BL/6J mice (Charles River Laboratories International, Inc.,
Wilmington, MA, USA) between 8 and 12 weeks old and weighing
20–30 g were housed in groups of 4 at a temperature of 20 ◦C
+/− 5 ◦C and were kept on a 12 : 12-h light : dark cycle with free
access to food and water for at least 1 week prior to testing. Animal
protocols were performed in accordance with NIH guideline for
the Care and Use of Laboratory Animals and approved by Wyeth’s
IACUC.
Mouse brain and adipose samples (∼100 mg) were homoge-
nized in 100 µl of 0.02% trifluoroacetic acid (TFA, pH 3.0) (1000 mg
tissue/ml). The brain homogenate, 20 µl of IS and 4 ml of ACN
were mixed in a silanized glass tube. The mixture was centrifuged
(3000 rpm, 4 ◦C) for 15 min. The supernatant was transferred into
a clean silanized glass tube and evaporated to dryness under ni-
trogen using a Zymark TurboVap evaporator at 35 ◦C. The residue
was reconstituted into 100 µl of ACN and vortexed, followed by a
brief centrifugation to remove any precipitates. The samples were
transferred into HPLC vials for storage at −80 ◦C before analysis
by LC/MS/MS.
Experimental
Reagents and chemicals
2-AG, 1-AG, AA and 2-AG-d8 were purchased from Cayman
Chemical (Ann Arbor, MI, USA). Formic acid, LC grade water,
methanol (MEOH) and ACN were obtained from EM Science
(Gibbstown, NJ, USA). Reagentsusedtopreparethebuffersolution
used for the in vitro activity assay were purchased from Sigma-
Aldrich (St. Louis, MO, USA). Acetic acid was from J. T. Baker
(Phillipsburg, NJ, USA). Mouse MAGL cDNA was purchased from
Open Biosystems (Huntsville, AL, USA).
To determine the extraction efficiency of 2-AG, 1-AG, AA and
2-AG-d8 from brain homogenate, analyte-spiked brain samples
and analyte-spiked control brain samples were prepared using
the sample preparation procedure discussed above. The analyte-
spiked brain samples were prepared by spiking homogenate from
100 mg brain tissue with 20 µl of each stock solution of 2-AG, 1-AG,
AA and 2-AG-d8, followed by adding 4 ml of ACN. The samples
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2010, 45, 167–177

Determination of 2-AG, 1-AG, AA using LC-MS/MS
were vortexed, and then centrifuged. The supernatant was dried
down under N2 and reconstituted into 100 µl of ACN. The analyte-
spiked control brain samples were prepared by aliquoting 100 mg
of brain tissue, followed by homogenization, protein precipitation
by ACN and centrifugation. The supernatant was dried under
N2 and reconstituted into 20 µl of ACN, plus 20 µl of each stock
solution of 2-AG, 1-AG, AA and 2-AG-d8 to make a final volume of
100 µl.
analysis. The MRM analyses were preformed by passing molecular
ions through the first quadrupole (Q1) followed by collisional
dissociatingthemolecularionsinthesecondquadrupole(collision
cell – Q2). A selected product ion, based on intensity and structure
characteristics, was isolated by the third quadrupole (Q3) and
detected with the photomultiplier set at 650. The MRM transitions
of m/z 379 → 287 for 2-AG and 1-AG, 305 → 93 for AA and 387 →
295 for 2-AG-d8 were simultaneously monitored. This approach
provides a sensitive and selective analysis that is unique for non-
isomeric analytes. The concentrations of 2-AG, 1-AG and AA were
determined by calculating their corresponding peak area ratio to
that of the IS using a linear fit weighting to the calibration curve.
Preparation of standard and quality control samples
Serials stock solutions of 2-AG, 1-AG and AA were prepared by
dilution of 1 mg/ml of ACN solutions of each with ACN, as 2-AG
is known to be more stable in ACN.[32] A stock solution of 2-AG-
d8, used as IS, was prepared at a concentration of 20 µg/ml in
ACN. The stock solutions were stored at −80 ◦C. Standard and
quality control (QC) samples were prepared by spiking 20 µl of
each corresponding stock solution of 2-AG, 1-AG and AA and 20 µl
of IS into 100 µl of water. The spiked solution and 4 ml of ACN
were mixed in a silanized glass tube and transferred into a clean
silanized glass tube and evaporated to dryness under nitrogen at
35 ◦C. The residuals were reconstituted into 100 µl of ACN. Seven-
point calibration curves were constructed for 2-AG, 1-AG and AA in
the concentration range of 0.02–20, 0.01–10 and 0.05–50 ng/mg
tissue for 2-AG, 1-AG and AA, respectively.
Validation was carried out by performing calibration experi-
ments over three non-consecutive days. For each day, calibration
samples were prepared in duplicate at seven concentration lev-
els according to the procedure described above. QC samples at
three different concentration levels were prepared separately in
duplicate for method validation.
Results and Discussion
Optimization of brain sample preparation method
Current literature suggests that conversion of 2-AG to 1-AG occurs
through a chemical acyl migration reaction under particular
experimental conditions.[33,34] Various solvent systems, such
as chloroform, chloroform/methanol and ethyl acetate/hexane,
were used in the literature for extraction of 2-AG from brain
homogenates.[31,32,34–37] The stability of 2-AG during brain and
adipose tissue preparation using these solvents including ACN
were investigated. Our studies showed that more than ∼40%
of 2-AG was converted to 1-AG when chloroform was used as
the extraction solvent (data are not shown). The acyl migration
converting 2-AG to 1-AG (or 2-AG-d8 to 1-AG-d8) was minimized
when ACN or ethyl acetate/hexane was used. Figure 1 shows a
comparison of the MRM traces for 2-AG-d8, 1-AG-d8 and AA-d8
from the mouse brain tissue extracts spiked with 2-AG-d8 (Fig. 1(a)
and (c)), and from the ACN solutions of 2-AG-d8 (Fig. 1(b)) or
AA-d8 standards (Fig. 1(d)). The brain extracts were prepared by
spiking with 2-AG-d8 immediately after the brain tissues were
homogenized, acidified and protein precipitated with ACN. As
shown in Fig. 1(d), AA-d8 eluted at the RT of 6.79 min from the
ACN solution of AA-d8, but it was not observed from the brain
extracts spiked with 2-AG-d8 ((Fig. 1(b)). 1-AG-d8, eluted at the
RT of 6.23 min, was originally observed in the ACN solution of
2-AG-d8 (Fig. 1(b)) [a fresh solution from Cayman Chemical (Ann
Arbor, MI, USA). The percent ratios of [2-AG-d8] : [1-AG-d8] was
94 0.46 : 6 0.4 (n = 5) in both the brain extracts spiked with
2-AG-d8 (Fig. 1(a)) and the ACN solution of 2-AG-d8 (Fig. 1(b)).
These observations showed that the enzymatic degradation of
2-AG-d8 to AA-d8 or the acyl migration converting 2-AG-d8 to 1-
AG-d8 did not occur during the brain sample preparation process.
Extractionof 2-AG frombraintissuesusingethylacetate/hexane
displayed high extraction recovery for 2-AG that required
additional purification steps using SPE[31,32] to remove some
extracted proteins or lipids, which plugged the column during the
analysis. However, SPE sample preparation methods are complex
and time consuming, which may limit the overall throughput. A
modified approach described in this report using 40-fold excess of
cold ACN compared to the sample volumes to extract 2-AG from
mouse brain tissue homogenates provided optimal recovery of
2-AG. Moreover, ACN effectively removed the proteins and lipids
in the brain and adipose tissue homogenates without the need for
further sample clean up.
Liquid chromatography
An Agilent 1100 series HPLC system (Hewlett-Packard GmbH,
Waldbronn, Germany) was used in this study. The system consists
of two quaternary pumps, a vacuum degasser, a temperature
controlledautosamplerandathermostatedcolumncompartment.
The chromatographic separation was carried out using a Chro-
molith RP-18E (100 × 3.0 mm i.d) (Merck KGaA, Darmstadt, Ger-
many), maintained at 40 ◦C. The mobile phase consisted of solvent
A: 0.2% acetic acid in water–methanol (H2O : MEOH=95 : 5, v/v)
and B: 0.2% acetic acid in water–methanol (H2O : MEOH=5 : 95,
v/v). The HPLC analysis started with 40% B for 0.5 min, then fol-
lowed by a gradient from 40% to 95% B in 1 min and subsequently
hold at 95% B for 6.5 min. The flow rate was 0.5 ml/min. The
HPLC flow was split before the MS and ∼0.2 ml/min effluent was
directed into the electrospray ionization (ESI) source of the mass
spectrometer. The injection volume was 20 µl.
Mass spectrometry
On-line LC/MS/MS analyses were performed using a Micromass
Quatro Micro tandem quadrupole mass spectrometer (Waters,
Beverly, MA, USA) operated in positive ESI mode with the ion
source temperature of 125 ◦C. The positive ESI ((+)ESI) conditions
for 2-AG, 1-AG, AA and 2-AG-d8 were optimized to a desolvation
temperature of 350 ◦C, a spray voltage of 3.5 kV and a cone
voltage of 20 V. Nitrogen was used as both desolvation (1000
l/h) and nebulizer gas (fully open). The pressure of the argon
collision gas was set at 5 psi and adjusted to an analyzer
pressure of 2.0–3.0 × 10−4 mbar. MRM modes were used for
There has been a technical challenge to study the recovery of
ECs from a biological matrix owing their endogenous nature.
Richardson et al. reported two approaches to determine the
recoveries of ECs with the consideration of the endogenous
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M.-Y. Zhang et al.
(a)
100
6.08
387 > 295
2.05e4
2-AG-d8
1-AG-d8
6.23
7.56
0
2.00
2.00
2.00
4.00
6.00
6.08
8.00
10.00
10.00
10.00
(b)
100
387 > 295
2.15e4
2-AG-d8
1-AG-d8
6.23
6.00
0
4.00
8.00
(c)
6.09
100
312 > 96
406
0
4.00
6.00
8.00
(d)
100
312 > 96
2.04e4
6.79
AA-d8
0
2.00
44.00
6.00
8.00
10.00
12.00
Time (min)
Figure 1. The MRM ion chromatograms of: a) 2-AG-d8 and 1-AG-d8 from the mouse brain tissue extracts spiked with 2-AG-d8, (b) 2-AG-d8 and 1-AG-d8
from the ACN solution of 2-AG-d8, c) AA-d8 from the mouse brain tissue extracts spiked with 2-AG-d8 and d) AA-d8 from the ACN solution of AA-d8.
levels of ECs.[32] A modified approach was used in our study
to consider both the endogenous levels of 2-AG, 1-AG and AA
and the matrix effects during ESI ionization. In our approach, the
recovery of analytes was evaluated by comparing the area ratios of
an analyte peak in analyte-spiked brain samples to that in analyte-
spiked control brain samples. These samples were prepared by
using the same sample preparation procedure as described in
the experimental section. However, for the analyte-spiked brain
samples,20 µlofcorrespondingstocksolutionsofeachanalyteand
IS were spiked before adding ACN for sample preparation. For the
analyte-spiked control brain samples, 20 µl of corresponding stock
solutions of each analyte and IS were spiked in the reconstitution
step after ACN protein precipitation. This approach minimized the
difference of the matrix effect between the analyte-spiked brain
samples and analyte-spiked control brain samples under (+) ESI
ionization, which might cause false positive or negative results.
The recoveries of spiked 2-AG, 1-AG and AA were evaluated after
subtracting the basal levels of 2-AG, 1-AG and AA measured from
the blank brain samples. Recoveries of >80% were achieved for
2-AG, 1-AG, AA and 2-AG-d8 in these studies (data are not shown).
The developed method also provided high recovery of AEA from
the brain tissue (unpublished).
and the same MRM transition (379 → 287). 1-AG could be
formed by an aryl migration reaction of 2-AG, which could occur
under some particular experimental conditions. The combined
measurement of 1-AG and 2-AG without HPLC separation was
used in most quantification methods for 2-AG reported in the
literature.[31,32,38,39] However, our study showed that the MS
response of 1-AG under (+) ESI MRM was 2 times higher than
that of 2-AG. The combined measurement of 1-AG and 2-AG in
LC/MS/MS analysis could significantly affect the measurement
accuracy of 2-AG, especially in enzymatic activity assays, in which
the concentrations of 2-AG and 1-AG were comparable.
Mobile phase systems containing 28% phosphate buffer was
used by Saario et al. to separate 2-AG and 1-AG.[33] However, this
mobile phase system is not ideal for ESI, since it could significantly
reduce the ionization efficiency due to ion suppression. A number
of reversed phase methods were evaluated. We found that using
a Chromolith RP-18E column in combination with a mobile
phase system consisting of H2O and MEOH with 0.2% acetic
acid provided the optimal separation of 1-AG and 2-AG. It is a
common knowledge that use of MEOH as mobile phase solvent
generates higher column back pressure compared to ACN. The
low back pressure on Chromolith columns compared to other
reversed phase columns provided an advantage in our use of a
methanol-containing solvent system. Furthermore, the ionization
efficiency of 1-AG and 2-AG was 2 times higher when MEOH was
used as a mobile phase solvent instead of ACN. As a result, our
method provided the optimal HPLC conditions and the optimal
MS responses.
Optimization of LC/MS/MS conditions
The HPLC method was optimized to ensure the separation of the
analytes. It was particularly important with regard to separation
of 2-AG and 1-AG, because they share the same molecular weight
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2010, 45, 167–177

Determination of 2-AG, 1-AG, AA using LC-MS/MS
(a)
Daughters of 379ES+
379.0
4.79e5
100
287.1
CID=25 eV
203.2
%
269.2
245.1
57.2
133.2
189.2
305.0
361.1
0
(b)
Daughters of 379ES+
1.09e6
379.0
100
CID=20 eV
%
287.0
269.0
203.2
95.2
245.2
66.9
50
361.1
350
305.0
300
m/z
0
100
150
200
250
400
450
500
Figure 2. ESI product ion spectra of [M+H]+ at m/z 379 for 2-AG using a collision energy of: a) 25 eV and b) 20 eV.
The ESI product ion spectra of the protonated molecular ion of
O+
OH+
2-AG ([M+H]+ atm/z 379)byusingcollisionenergies(CE)of20and
OH
O
25 eV are shown in Fig. 2. Dissociation of the molecular ion of 2-AG
OH
produced a most abundant product ion at m/z 287 by using a CE
of 20 eV. Numbers of low-mass fragment ions, which may not be
structural specific product ions, were produced when the CE was
[M+H]+ @ m/z 379
2-Arachidonoylglycerol (2-AG)
m/z 287
increased to 25 eV, while the abundance of the ion at m/z 287 was
decreased. The fragment ion at m/z 287 produced at a CE energy
of 20 eV was the most abundance ion among all fragment ions
observed in a CE energy range of 15–35 eV. In addition, the ion of
m/z 287wasformedbyaneutrallossofglycerolfromthemolecular
ion, which was a structural characteristic ion for 2-AG (Scheme 1).
Dissociationoftheprotonatedmolecularionsof1-AGand2-AG-d8
generated the fragment ions of m/z 287 and 295, respectively, by
a neutral loss of glycerol from the corresponding molecular ions
as well (Scheme 1). Dissociation of the protonated molecular ion
of AA ([M+H]+ @ m/z 305) produced mainly low-mass fragment
ions (Fig. 3). The ion at m/z 93 was the most abundant fragment
ion from the dissociation of the molecular ion of AA (Scheme 1).
Therefore, the MRM transitions of m/z 379 → 287 for 2-AG and 1-
AG, 305 → 93 for AA and 387 → 295 for 2-AG-d8 were chosen for
quantitative analysis. 2-AG-d8 was used as the IS for quantification
of all three analytes. AA-d8 was not selected as an IS for AA, since
AA-d8 formed mainly [M+Na]+ ions instead of [M+H]+ ions in
our system. These structurally specific MRM transitions associated
with the specific HPLC retention times were used to confirm the
identity of 2-AG, 1-AG, AA and 2-AG-d8 in the samples.
O+
OH+
O
OH
OH
[M+H]+ @ m/z 379
m/z 287
O+
1-Arachidonoylglycerol (1-AG)
OH+
OH
[M+H]+ @ m/z 305
Arachidonic acid (AA)
m/z 93
O+
OH+
D
D
D
D
D
D
D
D
D
D
OH
OH
O
D
D
D
D
D
D
[M+H]+ @ m/z 387
2-Arachidonoylglycerol-d8 (2-AG-d8)( IS)
m/z 295
Scheme 1. Dissociation pathways of: a) 2-AG, b) 1-AG, c) AA and d) 2-AG
−d8.
mode provided a sensitive and selective analysis that was unique
for individual compounds. The limits of detection (LOD) were 50,
25 and 65 fmol (S : N = 3 : 1) for 2-AG, 1-AG and AA, respectively.
The lowest quantification levels (LOQ) were 100, 50 and 130 fmol
Sensitivity and linearity
The selected mobile phase systems allowed these analytes to be
ionized with high ionization efficiency under (+) ESI. The MRM
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M.-Y. Zhang et al.
Daughters of 305ES+
93.1
100
305.1
119.2
%
81.3
133.1
161.5
175.2
207.1
m/z
0
50
100
150
200
250
300
350
400
Figure 3. ESI product ion spectrum of [M+H]+ at m/z 305 for AA (CE=30 eV).
Table 1. Intra-day and inter-day accuracy and precision of 2-AG, 1-AG and AA in QC samples
Intra-day (N = 6)
Inter-day (N = 18)
Accuracy
Nominal Conc (ng/mg
tissue) or (ng/µl)
Accuracy
(%)
Analyte
2-AG
Mean
SD
%RSD
Mean
(%)
SD
%R.S.D.
0.1
1
0.10
1.0
104
102
102
0.01
0.06
0.4
6.1
5.5
3.7
0.10
1.1
104
106
102
0.01
0.05
0.4
5.4
4.7
3.8
10
10.2
10.2
1-AG
AA
0.05
0.5
5
0.055
0.51
4.5
110
102
90
0.003
0.02
0.3
4.5
4.3
5.4
0.055
0.50
4.6
110
99
0.003
0.02
0.2
5.3
4.6
3.6
91
0.2
2
0.19
1.8
95
92
97
0.02
0.1
10.6
6.4
0.20
1.9
99
95
95
0.02
0.2
8.5
8.3
6.6
20
19.4
0.8
7.7
18.9
0.6
(S : N = 10 : 1) for 2-AG, 1-AG and AA, respectively. The LOD (50
fmol) for 2-AG achieved in this method was more sensitive than
the previously reported methods using GC/MS (1 pmol),[40,41]
LC/MS (∼5 pmol)[38] and LC/MS/MS (250 fmol).[32] The LOD at 13
fmol for 2-AG was reported by Kingsley et al. using silver adduct
coordination tandem MS.[31] The LOD values for 1-AG and AA were
not previously reported in the literature.
Seven-point standard curves were prepared to establish the
calibration ranges for 2-AG, 1-AG and AA. The linear regression
analysis with 1/x2 weighting was applied. Calibration curves of
2-AG, 1-AG and AA were linear in the concentration range of
0.02–20, 0.01–10 and 0.05–50 ng/mg tissue, respectively, with
the correlation coefficients greater than 0.997. The quantification
ranges were selected based on concentration ranges of these
analytes present in the mouse brain studied here and also in our
enzymatic activity studies.
of samples, analyzed on three different days. The intra-day and in
inter-day precision and accuracy are summarized in Table 1. The
intra-day precision (%RSD) was better than 6.1%, 5.4% and 10.6%
for 2-AG, 1-AG and AA, respectively. The value of the intra-day
accuracy was in a range of 90–110%. The inter-day precision and
accuracy were determined by pooling all validation data from
all QC samples at each concentration. The inter-day precision
was better than 5.4%, 5.3% and 8.5% for 2-AG, 1-AG and AA,
respectively. The value of the inter-day accuracy was in a range of
91–110%. These results indicate good precision and accuracy.
MS response, matrix effect, selectivity and stability
Accuratequantificationofendogenouschemicalshasalwaysbeen
a technical challenge. The isotopic dilution methods using a
deuterium labeled analog of an analyte have been traditionally
employed for determination of ECs. However, there were few
discussions regarding the MS response of the analyte and its
deuterium labeled analog in the literature. Kingsley et al.[31]
reported that the silver adducts of 2-AG and 2-AG-d8 showed
a similar MS response at the same concentrations. Therefore, the
Precision and accuracy
Precision and accuracy measurements were acquired for the
QC samples at three different concentrations from three sets
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2010, 45, 167–177

Determination of 2-AG, 1-AG, AA using LC-MS/MS
preparation procedure described in the experimental section and
using 2-AG-d8 as IS was evaluated. In the study, both MS responses
and matrix effects were evaluated by comparing the peak area
ratios of each analyte to 2-AG-d8 in the standard solutions to
that in the analyte-spiked brain samples at three different spiked
analyte concentration levels. The basal levels of 2-AG, 1-AG and AA
were measured using the aliquots of blank brain samples. It should
be noted that due to the high basal level of 2-AG in brain tissue,
the lowest spiked level of 2-AG at 1 ng/mg tissue was selected. The
peak area ratios of spiked 2-AG, 1-AG and AA to 2-AG-d8 in brain
samples were calculated after subtracting the basal levels of 2-AG,
1-AG and AA. As presented in Table 2, the peak area ratios of 2-AG,
1-AG and AA to 2-AG-d8 in ACN solutions and in brain samples
were very similar. The accuracy determined by the percent ratios
of [analyte] : [IS] in brain samples to that in ACN solutions were
in the range of 91–111%. These observations indicated that even
though there was a large difference between the MS responses
of 2-AG and 2-AG-d8, the matrix effects on 2-AG and 2-AG-d8 in
the same matrix were very similar. Our data suggest that use of
calibration curves prepared in pure solvents and use of 2-AG-d8
as the IS provided reliable quantification for 2-AG, 1-AG and AA in
brain tissue samples.
The MRM ion chromatograms of 2-AG (100 fmol), 1-AG (50
fmol), AA (130 fmol) and 2-AG-d8 (IS, 200 pmol) in a standard
solution are presented in Fig. 4. 1-AG-d8 was observed with a
relative amount of ∼6% of 2-AG-d8 in the solution. Utilization of
structurally specific MRM transitions associated with the specific
HPLC retention times were used to confirm the identity of these
analytes and IS.
The stabilities of 2-AG, 1-AG and AA spiked in control brain
homogenates were evaluated at both 4 and 25 ◦C for 24 h. Our
observations showed that all analytes were stable under these
experimental conditions.
Table 2. Matrix effects on ratios of 2-AG, 1-AG and AA to 2-AG-d8
Area ratio [analyte]/[2-AG-d8]
Nominal Conc of analyte
Analyte (ng/mg tissue) or (ng/µl) In brain In solvent
∗Accuracy
(%)
2-AG
1-AG
AA
1
10
20
0.89
7.99
0.81
8.39
110
95
14.7
16.0
92
0.1
0.18
2.53
0.17
2.64
106
96
1
10
31.4
33.7
93
0.5
2.5
25
0.33
1.67
0.30
1.72
110
97
22.1
19.8
111
∗ Accuracy = [ratio]brain/[ratio]solvent
responseof2-AGsilveradductand2-AG-d8silveradductwastaken
as equal for quantification calculations using the isotopic dilution
method.[31] In our study, the response ratio of 2-AG to 2-AG-d8 was
determined by using 2-AG : 2-AG-d8 standard solutions with molar
ratios of 1 : 10, 1 : 1 and 10 : 1. The 2-AG : 2-AG-d8 area ratios were
0.75 0.015, 6.53 0.131 and 64.4 2.72, respectively (n = 3,
mean SD). Despite the structural similarity of 2-AG and 2-AG-d8,
the MS response of 2-AG was found to be about sevenfold higher
than 2-AG-d8 using (+) ESI MRM scan mode. This large difference
in the MS response between the analyte and its deuterium labeled
analog could significantly affect the measurement accuracy using
the traditional isotopic dilution methods.[42]
In the present study, a different approach of using calibration
standards prepared in pure solvents following the sample
379 > 287
6.22
6.10
100
%
2-AG
1-AG
0
305 >93
6.79
100
AA
%
0
387 > 295
6.07
100
%
2-AG-d8
1-AG-d8
6.19
0
2.00
4.00
6.00
Time (min)
8.00
12.00
10.00
Figure 4. The MRM ion chromatograms of 2-AG (100 fmol), 1-AG (50 fmol), AA (130 fmol) and 2-AG-d8 (200 pmol) in an ACN solution.
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M.-Y. Zhang et al.
O
110
90
O
CH3
OH
OH
1-AG
Acyl migration
Hydrolysis
2-AG
1-AG
AA
O
OH
OH
70
O
CH3
2-AG
50
Hydrolysis
O
OH
OH
OH
CH3
HO
+
30
AA
glycerol
Scheme 2. Hydrolysis and acyl migration reactions of 2-AG in mouse brain
10
extracts containing membrane-bound MAGL.
-10
0
20
40
60
80
100
120
Time (min)
Table 3. Acryl migration and hydrolysis of 2-AG (mean +SE) in buffer,
mouse brain extract and purified rMAGL at 30-min incubation
Figure 5. Acyl migration and hydrolysis of 2-AG in purified rMAGL.
% 2-AG, 1-AG and AA to total species
Compound
In buffer
In mouse brain extract In purified rMAGL
100.0
80.0
2-AG
1-AG
AA
76 6.8
24 3.0
26 6.3
5.9 0.6
68 7.2
8.2 2.1
6.5 0.4
85 4.5
Not detected
60.0
40.0
20.0
0.0
2-AG
1-AG
AA
Application
There is limited knowledge about the enzymes that terminate
signaling of the major brain EC, 2-AG, in the brain, although
MAGL has generally been assumed to be the main contributor
in this process. The presented methods have been applied to
understand EC signaling through their production and enzymatic
degradation. The developed methods have been applied to study
2-AG hydrolysis in vitro and was also used to measure basal levels
of 2-AG in mouse brain and adipose tissues.
0
20
40
60
80
100
120
Time (min)
Figure 6. Acyl migration and hydrolysis of 2-AG in membranes of mouse
brain extracts.
Enzymatic activity for hydrolysis of 2-AG in mouse cerebellar
membranes
species of 2-AG, 1-AG and AA. The time course of degradation of
2-AG when incubated either with rMAGL protein or membrane-
bound MAGL, prepared from the whole mouse brain, are shown
in Figs 5 and 6, respectively. The relative concentration of
2-AG was 26 6.3% and 8.2 2.1% after a 30-min incubation
with brain extract and rMAGL, respectively (Table 3). The relative
concentration of 1-AG was 5.9 0.6% and 6.5 0.4% after a
30-min incubation with brain extract and rMAGL, respectively.
The relative concentration of AA, formed by the hydrolysis of
2-AG and 1-AG, was 68 7.2% and 85 4.5% by brain extract
and rMAGL, respectively. These observations suggested that the
majority of 2-AG was hydrolyzed by MAGL. In comparison, Saario
et al. reported the relative concentration of AA was ∼20% after a
30-min incubation with rat cerebellar membranes.[33]
As presented in Scheme 2, 2-AG could be enzymatically degraded
to AA and glycerol by a hydrolysis reaction, where 1-AG, formed
by acyl migration of 2-AG, could also be degraded to AA.[33] Saario
et al. previously reported the investigation of monoglyceride
lipase-like enzymatic activity for hydrolysis of 2-AG using an HPLC
method.[33] In their method, the relative concentrations of 2-AG,
1-AG and AA were estimated on the basis of corresponding UV
peak areas. In the present work, the degradation of 2-AG was
investigated by incubating 2-AG in membrane-free buffer, rMAGL
proteinormembrane-boundMAGLpreparedfromthewholebrain
following the published procedure.[15] 2-AG, 1-AG and AA were
monitored simultaneously using our validated LC/MS/MS method,
which provides higher accuracy and sensitivity than the HPLC/UV
method.
As illustrated in Table 3, 24% of 2-AG was converted to 1-AG
after a 30-min incubation in membrane-free buffer, suggesting
the formation of 1-AG by a chemical acyl migration reaction. AA
was not observed when 2-AG was incubated with buffer alone,
indicating that the formation of AA was enzyme dependent. These
observations agreed with Saario’s finding.[33]
The basal levels of 2-AG, 1-AG and AA from mouse brain
and adipose tissues
The sample preparation and LC-MS/MS methods reported here
were applied to measure the basal levels of 2-AG, 1-AG and AA
from mouse brain tissues. Due to the complex nature of the brain
samples, two MRM transitions for each analyte and the IS were
monitored simultaneously for the analysis of the brain extracts.
The degradation of 2-AG was presented as the relative percent
concentrations (µM) of each analyte compared to the total three
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2010, 45, 167–177

Determination of 2-AG, 1-AG, AA using LC-MS/MS
Table 4. Comparison of literature reports of basal levels of 2-AG in mouse and rat extracts
2-AG basal level
(pmol/mg)
Reference
Species/region
(ng/mg)
Detection method
Bisogno et al. (1999)[29]
Fezza et al. (2002)[5]
Valenti et al. (2004)[30]
Kingsley et al. (2003)[31]
Richardson et al. (2006)[32]
Current work
Different rat brain regions
Whole rat brain
1.5–5.3
1.7
4.0–14.0
4.5
GC/EI/MS
LC/APCI/MS
LC/APCI/MS
LC/ESI/MS/MS
LC/ESI/MS/MS
LC/ESI/MS/MS
Different rat brain regions
Whole mouse brain
5–15
5.0
13.2–39.7
13.2
Different rat brain regions
Whole mouse brain
4.1–11.3
3.6
10.8–29.9
9.5
6.82
6.82
AA
305 > 119
100
0
2.00
4.00
4.00
4.00
4.00
4.00
4.00
6.00
8.00
10.00
10.00
10.00
10.00
10.00
10.00
12.00
305 > 93
100
0
AA
2.00
2.00
2.00
2.00
2.00
6.00
6.12
8.00
8.00
8.00
8.00
8.00
12.00
379 > 287
100
0
2-AG
1-AG
6.26
6.00
6.12
12.00
2-AG
1-AG
100
0
379 > 203
6.26
6.00
6.08
12.00
2-AG-d8
2-AG-d8
1-AG-d8
100
0
387 > 295
6.21
6.21
7.56
6.00
6.08
12.00
387 > 84
1-AG-d8
100
0
6.00
12.00
Time (min)
Figure 7. The MRM ion chromatograms of the basal 2-AG, 1-AG, AA and the spiked 2-AG-d8 (IS) from the mouse brain tissue extracts.
4.5
The MRM transitions were m/z 379 → 287 and 379 → 203 for
2-AG and 1-AG; m/z 305 → 93 and 305 → 119 for AA; and
4.0
m/z 387 → 295 and 387 → 84 for 2-AG-d8. The identities of
2-AG, 1-AG, AA and 2-AG-d8 were confirmed by comparing their
HPLC retention times and the MRM transitions with the standard
compounds. The ion chromatograms of the 2-AG, 1-AG and AA
from the mouse brain tissue extract and the spiked 2-AG-d8 are
presented in Fig. 7. The basal levels of 2-AG, 1-AG and AA in
mouse tissue were determined to be 3.6 0.41, 0.34 0.03
and 18.0 2.1 ng/mg tissue, respectively. The basal levels of
2-AG, measured using GC/EI/MS, LC/APCI/MS and LC/ESI/MS/MS
reported in the literature are summarized in Table 4. The basal
level of 2-AG in the whole mouse brain described in this report is
compatible with that reported by Kingsley et al. using the silver
adduct LC/MS/MS method.[31]
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Brain tissue
Adipose tissue
The basal levels of 2-AG were also determined in mouse adipose
tissue using our methods. We observed ten fold higher levels of
2-AG in the brain compared to the adipose tissue (Fig. 8) (N = 3,
P < 0.005), which was consistent with the reported finding that
MAGL activity was higher in adipose tissue compared to the brain
tissue.
Figure 8. The basal levels of 2-AG in mouse brain and adipose tissue.
Conclusion
A brain sample preparation method and a reversed phase
LC/MS/MS method have been developed for simultaneous
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M.-Y. Zhang et al.
determination of 2-AG, 1-AG and AA in mouse brain and adipose
tissue. The current method provides higher detection sensitivity
than that reported using HPLC methods based on UV detection.
In addition, the LC/MS/MS method provides higher accuracy by
identifying the analytes using structurally specific MRM transitions
associated with the selective HPLC retention times. The use of 40-
fold excess of cold ACN for sample preparation provided optimal
recovery of 2-AG, 1-AG and AA, and at the same time, effectively
removed the proteins and lipids in the brain and adipose tissue
homogenates. Chromatographic separation of 2-AG and 1-AG
ensured the accurate quantification of 2-AG. The quantification
method was validated with consideration of matrix effects on the
MS responses of the analytes and IS.
These methods were used for both in vitro and in vivo to
simultaneously determine the concentrations of 2-AG, 1-AG
and AA. These methods appear to be sensitive and specific
and have been successfully applied to measure 2-AG, 1-AG
and AA concentrations in mouse brain and adipose tissue.
Accurate measurement of 2-AG and its enzymatically catalyzed
degradationproductswouldbeextremelyusefulinunderstanding
the EC signaling and its regulation by enzymatic biosynthesis and
degradation.
[12] N. Stella, P. Schweitzer, D. Piomelli.
A
second endogenous
cannabinoid that modulates long-term potentiation. Nature 1997,
388, 773.
[13] T. Sugiura, T. Kodaka, S. Nakane, T. Miyashita, S. Kondo, Y. Suhara,
H. Takayama, K. Waku, C. Seki, N. Baba, Y. Ishima. Evidence that the
cannabinoid CB1 receptor is a 2-arachidonoylglycerol receptor.
Structure-activity relationship of 2-arachidonoylglycerol, ether-
linked analogs, and related compounds. J. Biol. Chem.1999, 274,
2794.
[14] W. Gonsiorek, C. Lunn, X. Fan, S. Narula, D. Lundell, R. W. Hipkin.
Endocannabinoid 2-arachidonyl glycerol is a full agonist through
human type 2 cannabinoid receptor: antagonism by anandamide.
Mol. Pharmacol. 2000, 57, 1045.
[15] J. R. Savinainen, T. Jarvinen, K. Laine, J. T. Laitinen. Despite
substantial degradation, 2-arachidonoylglycerol is a potent full
efficacy agonist mediating CB(1) receptor-dependent G-protein
activation in rat cerebellar membranes. Br. J. Pharmacol. 2001, 134,
664.
[16] T. Sugiura, S. Kondo, S. Kishimoto, T. Miyashita, S. Nakane,
T. Kodaka, Y. Suhara, H. Takayama, K. Waku. Evidence that
2-arachidonoylglycerol but not N-palmitoylethanolamine or
anandamide is the physiological ligand for the cannabinoid
CB2 receptor. Comparison of the agonistic activities of various
cannabinoid receptor ligands in HL-60 cells. J.Biol.Chem. 2000, 275,
605.
[17] T. Sugiura, S. Kondo, A. Sukagawa, S. Nakane, A. Shinoda, K. Itoh,
A. Yamashita, K. Waku. 2-Arachidonoylglycerol: a possible endo-
genous cannabinoid receptor ligand in brain. Biochem. Biophys. Res.
Comm. 1995, 215, 89.
[18] N. Stella, P. Schweitzer, D. Piomelli.
A second endogenous
cannabinoid that modulates long-term potentiation. Nature 1997,
388, 773.
References
[19] B. F. Cravatt, D. K. Giang, S. P. Mayfield, D. L. Boger, R. A. Lerner,
N. B. Gilula. Molecular characterization of an enzyme that degrades
neuromodulatory fatty-acid amides. Nature 1996, 384, 83.
[20] M. K. McKinney, B. F. Cravatt. Structure and function of fatty acid
amide hydrolase. Annu. Rev. Biochem. 2005, 74, 411.
[21] S. K. Goparaju, N. Ueda, H. Yamaguchi, S. Yamamoto. Anandamide
amidohydrolase reacting with 2-arachidonoylglycerol, another
cannabinoid receptor ligand. FEBS Lett. 1998, 422, 69.
[22] M. van Tienhoven, J. Atkins, Y. Li, P. Glynn. Human neuropathy
target esterase catalyzes hydrolysis of membrane lipids. J. Biol.
Chem. 2002, 277, 20942.
[23] T. P. Dinh, T. F. Freund, D. Piomelli. A role for monoglyceride lipase
in 2-arachidonoylglycerol inactivation. Chem.Phys.Lipids 2002, 121,
149.
[24] J. L. Blankman, G. M. Simon, B. F. Cravatt. Comprehensive profile
of brain enzymes that hydrolyze the endocannabinoid
2 – arachidonoylglycerol. Chem. Biol. 2007, 14, 1347.
[25] G. G. Muccioli, C. Xu, E. Odah, E. Cudaback, J. A. Cisneros, D. M.
Lambert, M. L. L. Rodriguez, S. Bajjalieh, N. Stella. Identification of
a novel endocannabinoid – hydrolyzing enzyme expressed by
microglial cells. J. Neurosci. 2007, 27, 2883.
[26] W. A. Devane, L. Hanus, A. Breuer, R. G. Pertwee, L. A. Stevenson,
G. Griffin, D. Gibson, A. Mandelbaum, A. Etinger, R. Mechoulam.
Isolation and structure of a brain constituent that binds to the
cannabinoid receptor. Science 1992, 258, 1946.
[27] L. Hanus, S. Abu-Lafi, E. Fride, A. Breuer, Z. Vogel, D. E. Shalev,
I. Kustanovich, R. Mechoulam. 2- Arachidonyl glyceryl ether, an
endogenous agonist of the cannabinoid CB1 receptor. Proc. Nation.
Acad. Sci. USA 2001, 98, 3662.
[28] P. C. Schmid, K. D. Schwartz, C. N. Smith, R. J. Krebsbach, E. V.
Berdyshev, H. H. O. Schmid. A sensitive endocannabinoid assay.
The simultaneous analysis of N-acylethanolamines and
2-monoacylglycerols. Chem. Phys. Lipids 2000, 104, 185.
[29] T. Bisogno, F. Berrendero, G. Ambrosino, M. Cebeira, J. A. Ramos,
J. J. Fernandez-Ruiz, V. Di Marzo. Brain regional distribution
of endocannabinoids: implications for their biosynthesis and
biological function. Biochem. Biophys. Res. Commun. 1999, 256,
377.
[1] I. Matias, P. Pochard, P. Orlando, M. Salzet, J. Pestel, V. Di Marzo.
Presence and regulation of the endocannabinoid system in human
dendritic cells. Eur. J. Biochem. 2002, 269, 3771.
[2] V. Di Marzo, C. S. Breivogel, Q. Tao, D. T. Bridgen, R. K. Razdan,
A. M. Zimmer, A. Zimmer, B. R. Martin. Levels, metabolism, and
pharmacological activity of anandamide in CB1 cannabinoid
receptor knockout mice: evidence for non-CB1, non-CB2 receptor-
mediated actions of anandamide in mouse brain. J. Neurochem.
2000, 75, 2434.
[3] K. Kempe, F. F. Hsu, A. Bohrer, J. Turk. Isotope dilution mass spec-
trometric measurements indicate that arachidonylethanolamide,
the proposed endogenous ligand of the cannabinoid receptor,
accumulates in rat brain tissue post mortem but is contained at
low levels in or is absent from fresh tissue. J. Biol. Chem. 1996, 271,
17287.
[4] A. Giuffrida, de F. F. Rodriguez, D. Piomelli. Quantification of
bioactive acylethanolamides in rat plasma by electrospray mass
spectrometry. Anal. Biochem. 2000, 280, 87.
[5] F. Fezza, T. Bisogno, A. Minassi, G. Appendino, R. Mechoulam,
V. Di Marzo. Noladin ether, a putative novel endocannabinoid:
inactivation mechanisms and
a sensitive method for its
quantification in rat tissues. FEBS Lett. 2002, 513, 294.
[6] V. Di Marzo, S. K. Goparaju, L. Wang, J. Liu, S. Bitkai, Z. Jarai, F. Fezza,
G. I. Miura, R. D. Palmiter, T. Sugiura, G. Kunos. Leptin-regulated
endocannabinoids are involved in maintaining food intake. Nature
2001, 410, 822.
[7] F. Berrendero, N. Sepe, J. A. Ramos, V. DiMarzo, J. J. Fernandez-Ruiz.
Analysis of cannabinoid receptor binding and mRNA expression
and endogenous cannabinoid contents in the developing rat brain
during late gestation and early postnatal period. Synapse 1999, 33,
181.
[8] R. I. Wilson, R. A. Nicoll. Endogenous cannabinoids mediate
retrograde signalling at hippocampal synapses. Nature 2001, 410,
588.
[9] E. V. Berdyshev. Cannabinoid receptors and the regulation of
immune response. Chem. Phys. Lipids 2000, 108, 169.
[10] G. Marsicano, C. T. Wotjak, S. C. Azad, T. Bisogno, G. Rammes,
M. G. Cascio, H. Hermann, J. Tang, C. Hofmann, W. Zieglgansberger,
V. Di Marzo, B. Lutz. The endogenous cannabinoid system controls
extinction of aversive memories. Nature 2002, 418, 530.
[11] S. A. Varvel, A. H. Lichtman. Evaluation of CB1 receptor knockout
mice in the Morris water maze. J. Pharmacol. Exp. Therap. 2002, 301,
915.
[30] M. Valenti,D. Vigano,M. G. Cascio,T. Rubino,L. Steardo,D. Parolaro,
V. Di Marzo. Differential diurnal variations of anandamide and
2-arachidonoyl-glycerol levels in rat brain. Cell. Mol. Life Sci. 2004,
61, 945.
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2010, 45, 167–177

Determination of 2-AG, 1-AG, AA using LC-MS/MS
[31] P. J. Kingsley, L. J. Marnett. Analysis of endocannabinoids by Ag+
coordination tandem mass spectrometry. Anal. Biochem. 2003, 314,
8.
[37] H.-Y. T. Yang, F. Karoum, C. Felder, H. Badger, T.-C. L. Wang,
S. P. Markey.GC/MSanalysisofanandamideandquantificationofN-
arachidonoylphosphatidylethanolamides in various brain regions,
spinal cord, testis, and spleen of the rat. J. Neurochem. 1999, 72,
1959.
[38] E. J. Carrier, C. S. Kearn, A. J. Barkmeier, N. M. Breese, W. Yang,
K. Nithipatikom, S. L. Pfister, W. B. Campbell, C. J. Hillard. Cultured
rat microglial cells synthesize the endocannabinoid 2-
[32] D. Richardson, C. A. Ortori, V. Chapman, D. A. Kendall, D. A. Barrett.
Quantitativeprofilingofendocannabinoidsandrelatedcompounds
in rat brain using liquid chromatography-tandem electrospray
ionization mass spectrometry. Anal. Biochem. 2007, 360, 216.
[33] S. M. Saario, J. R. Savinainen, J. T. Laitinen, T. Jarvinen, R. Niemi.
Monoglyceride lipase-like enzymatic activity is responsible for
hydrolysis of 2-arachidonoylglycerol in rat cerebellar membranes.
Biochem. Pharmacol. 2004, 67, 1381.
[34] T. Bisogno, N. Sepe, D. Melck, S. Maurelli, L. De Petrocellis, V. Di
Marzo. Biosynthesis, release and degradation of the novel
endogenous cannabimimetic metabolite 2-arachidonoylglycerol
in mouse neuroblastoma cells. Biochem. J. 1997, 322, 671.
[35] M. Maccarrone, M. Attina, A. Cartoni, M. Bari, A. Finazzi-Agro. Gas
chromatography-mass spectrometry analysis of endogenous
cannabinoids in healthy and tumoral human brain and human
cells in culture. J. Neurochem. 2001, 76, 594.
arachidonylglycerol, which increases proliferation via
a CB2
receptor-dependent mechanism. Mol. Pharmacol. 2004, 65, 999.
[39] A. Thomas, G. Hopfgartner, C. Giroud, C. Staub. Quantitative and
qualitative profiling of endocannabinoids in human plasma using
a triple quadrupole linear ion trap mass spectrometer with liquid
chromatography. Rapid Commun. Mass Spectrom. 2009, 23, 629.
[40] L. DePetrocellis,D. Melck,T. Bisogno,A. Milone,V. DiMarzo.Finding
of the endocannabinoid signalling system in Hydra, a very primitive
organism: possible role in the feeding response. Neuroscience 1999,
92, 377.
[41] S. Gonzalez, M. Grazia Cascio, J. Fernandez-Ruiz, F. Fezza, V. Di
Marzo, J. A. Ramos. Changes in endocannabinoid contents in the
brain of rats chronically exposed to nicotine, ethanol or cocaine.
Brain Res. 2002, 954, 73.
[36] S. Ben-Shabat, E. Fride, T. Sheskin, T. Tamiri, M. H. Rhee, Z. Vogel,
T. Bisogno, L. De Petrocellis, V. Di Marzo, R. Mechoulam. An
entourage effect: inactive endogenous fatty acid glycerol esters
enhance 2-arachidonoyl-glycerol cannabinoid activity. Eur. J.
Pharmacol. 1998, 353, 23.
[42] M. Wan, S. Ding, S. Yap, J. Miller. 56th ASMS Conference, June 1–5,
2008, Denver, CO.
c
J. Mass. Spectrom. 2010, 45, 167–177
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms