Research Article

Received: 27 October 2008 Accepted: 9 July 2009 Published online in Wiley Interscience:

(www.interscience.com) DOI 10.1002/jms.1624

Studies on the metabolism of the�9-tetrahydrocannabinol precursor�9-tetrahydrocannabinolic acid A(�9-THCA-A) in rat using LC-MS/MS,LC-QTOF MS and GC-MS techniques

Julia Jung,a Markus R. Meyer,b Hans H. Maurer,b Christian Neusuß,c

Wolfgang Weinmanna and Volker Auwartera∗

In Cannabis sativa, �9-Tetrahydrocannabinolic acid-A (�9-THCA-A) is the non-psychoactive precursor of�9-tetrahydrocannabinol (�9-THC). In fresh plant material, about 90% of the total �9-THC is available as �9-THCA-A.When heated (smoked or baked), �9-THCA-A is only partially converted to �9-THC and therefore, �9-THCA-A can be detectedin serum and urine of cannabis consumers. The aim of the presented study was to identify the metabolites of �9-THCA-A and toexamine particularly whether oral intake of �9-THCA-A leads to in vivo formation of �9-THC in a rat model. After oral applicationof pure �9-THCA-A to rats (15 mg/kg body mass), urine samples were collected and metabolites were isolated and identifiedby liquid chromatography-mass spectrometry (LC-MS), liquid chromatography-tandem mass spectrometry (LC-MS/MS) andhigh resolution LC-MS using time of flight-mass spectrometry (TOF-MS) for accurate mass measurement. For detection of�9-THC and its metabolites, urine extracts were analyzed by gas chromatography-mass spectrometry (GC-MS). The identi-fied metabolites show that �9-THCA-A undergoes a hydroxylation in position 11 to 11-hydroxy-�9-tetrahydrocannabinolicacid-A (11-OH-�9-THCA-A), which is further oxidized via the intermediate aldehyde 11-oxo-�9-THCA-A to 11-nor-9-carboxy-�9-tetrahydrocannabinolic acid-A (�9-THCA-A-COOH). Glucuronides of the parent compound and both main metabolites wereidentified in the rat urine as well. Furthermore, �9-THCA-A undergoes hydroxylation in position 8 to 8-alpha- and 8-beta-hydroxy-�9-tetrahydrocannabinolic acid-A, respectively, (8α-Hydroxy-�9-THCA-A and 8β-Hydroxy-�9-THCA-A, respectively)followed by dehydration. Both monohydroxylated metabolites were further oxidized to their bishydroxylated forms. Severalglucuronidation conjugates of these metabolites were identified. In vivo conversion of �9-THCA-A to �9-THC was not observed.Copyright c© 2009 John Wiley & Sons, Ltd.

Keywords: �9-Tetrahydrocannabinolic acid-A (�9-THCA-A); metabolism; LC-MS/MS; LC-QTOF MS; GC-MS

Introduction

More than 400 compounds have been identified in Cannabis

sativa, more than 60 belonging to the class of cannabinoids.[1,2]

In the plant, cannabinoids are biosynthesized and accumulated as

cannabinoid acids and non-enzymatically decarboxylated into

their neutral forms followed by degradation into secondary

products by temperature, auto-oxidation and light.[3 – 6] Decar-

boxylation occurs slowly during storage and fermentation, and

promptly during heating or smoking. However, when smoked,

�9-tetrahydrocannabinolic acid-A (�9-THCA-A) is only partially

converted to �9-THC. Dussy et al. investigated the temperature

dependence of the decarboxylation of �9-THCA-A under various

analytical and smoking conditions.[7] The highest conversion rate

resulting in about 70% �9-THC was observed under optimized

analytical conditions (temperatures higher than 140 ◦C), whereas

in a simulated smoking process only about 30% of the spiked

�9-THCA-A was recovered as �9-THC.

�9-THCA-A was detected in serum and urine of cannabis con-

sumers using liquid chromatography-tandem mass spectrometry

(LC-MS/MS) by our research group.[8] The highest �9-THCA-A

concentration was found, when blood sampling occurred shortly

after consumption of cannabis resulting in a high molar ratio of

�9-THCA-A/�9-THC. The inclusion of �9-THCA-A into our stan-

dard procedure for the quantification of �9-THC and its metabo-

lites by gas chromatography–mass spectrometry (GC-MS) re-

vealed that more than 80% of the �9-THC positive serum samples

of cases from driving under the influence of drugs (DUID) were

also positive for �9-THCA-A.[9] Furthermore, �9-THCA-A was de-

∗ Correspondence to: Volker Auwarter, Institute of Forensic Medicine, Forensic

Toxicology, University Medical Centre Freiburg, Albertstraße 9, D-79104

Freiburg, Germany. E-mail: volker.auwaerter@uniklinik-freiburg.de

a Institute of Forensic Medicine, Forensic Toxicology, University Medical Centre

Freiburg, D-79104 Freiburg, Germany

b Department of Experimental and Clinical Toxicology, Institute of Experimental

and Clinical Pharmacology and Toxicology, Saarland University, D-66421

Homburg (Saar), Germany

c Department of Chemistry, University of Applied Sciences, D-73430 Aalen,

Germany

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd.

J. Jung et al.

Table 1. Transitions obtained from the EPI spectra of pooled rat urine samples, which were used for MRM method set up

Parent compound/metabolite Transitions

DeclusteringPotential

[eV]

CollisionEnergy

[eV]

Retentiontime[min]

�9-THCA-A 357.2 → 313.2 −30 −35 15.84

357.2 → 245.2 −30 −50

357.2 → 191.1 −30 −50

�9-THCA-A glucuronide (ester) 533.2 → 357.2 −30 −35 14.74

533.2 → 313.2 −30 −50

533.2 → 245.2∗ −30 −50

�9-THCA-A glucuronide (ether) 533.2 → 357.2 −30 −35 12.40

533.2 → 313.2 −30 −50

533.2 → 245.2∗ −30 −50

11-OH-�9-THCA-A 373.2 → 311.2 −30 −35 13.54

373.2 → 268.2 −30 −50

373.2 → 173.1 −30 −50

11-OH-�9-THCA-A glucuronide 549.2 → 373.2 −30 −35 12.63

549.2 → 311.2 −30 −50

549.2 → 268.2∗ −30 −50

8β ,11-Bis-OH-�9-THCA-A 389.2 → 327.2 −30 −35 11.94

389.2 → 309.2 −30 −50

389.2 → 269.2 −30 −50

8β ,11-Bis-OH-�9-THCA-A glucuronide 565.2 → 389.2 −30 −35 11.05

565.2 → 327.2∗ −30 −50

565.2 → 309.2∗ −30 −50

8α,11-Bis-OH-�9-THCA-A 389.2 → 327.2 −30 −35 11.56

389.2 → 309.2 −30 −50

389.2 → 269.2 −30 −50

8α,11-Bis-OH-�9-THCA-A glucuronide 565.2 → 389.2 −30 −35 10.60

565.2 → 327.2 −30 −50

565.2 → 309.2∗ −30 −50

�9-THCA-A-8-one 371.2 → 327.2 −30 −35 13.79

371.2 → 284.2 −30 −35

371.2 → 189.1 −30 −50

�9-THCA-A-8-one glucuronide 547.2 → 371.2 −30 −35 12.79

547.2 → 327.2 −30 −50

547.2 → 284.2 −30 −50

�9-THCA-A-COOH 387.2 → 299.2 −30 −35 12.01

387.2 → 245.2 −30 −50

387.2 → 191.1 −30 −50

�9-THCA-A-COOH glucuronide 563.2 → 387.2 −30 −35 10.83

563.2 → 299.2 −30 −50

563.2 → 245.2 −30 −50

∗ calculated m/z values from known fragmentation patterns of the aglycon

tected in oral fluid up to 8 h after marijuana smoking by Moore

et al.[10]

�9-THC is the well-known psychoactive component in

Cannabis sativa and the most widely used illegal drug. In addition,

�9-THC shows a variety of therapeutic properties such as the relief

of nausea caused by cancer chemotherapy, the compensation

of anorexia and nausea under HIV therapy and the suppression

of spasticity and neuropathic pain associated with multiple

sclerosis.[11 – 16]

In the liver, �9-THC is subject to metabolic degradation.

Thus, 11-hydroxy-�9-tetrahydrocannbinol (11-OH-�9-THC) is

formed by the cytochrome P450 isoenzyme CYP2C9.[17 – 21] In

rat liver microsomes, cytochrome P450 isoenzyme CYP2C11

is the major enzyme responsible for the 11-hydroxylation in

microsomes from male rats, whereas CYP2C6 plays an impor-

tant role in the 11-hydroxylation in microsomes from female

rats.[22,23] Further oxidation of 11-OH-�9-THC via the intermediate

aldehyde 11-oxo-�9-tetrahydrocannbinol to 11-nor-9-carboxy-

�9-tetrahydrocannbinol (�9-THC-COOH) is catalyzed by a mi-

crosomal oxygenase.[19,24] To a small extent, �9-THC is also

eliminated as acetalic (ether) glucuronide.[25] After enzymatic

glucuronidation of the carboxy function, 11-nor-9-carboxy-

www.interscience.wiley.com/journal/jms Copyright c© 2009 John Wiley & Sons, Ltd. J. Mass. Spectrom. (2009)

�9-THCA-A metabolism in rats

Figure 1a. Negative LC-ESI-MS/MS EPI spectra (CE −20, −35, −50eV, summed) and reconstructed ion chromatograms (RIC), respectively, retentiontimes, structures and predominant fragmentation patterns of �9-THCA-A and its metabolites. The numbers of the spectra correspond to those of thestructures shown in Fig. 3.

�9-tetrahydrocannbinol glucuronide (�9-THC-COOH glu-

curonide) is excreted in urine.[26] Several minor metabolites

carrying hydroxyl functions at C1′ to C5′ of the side chain

have been reported, which are transformed to the corre-

sponding carboxylic acids by beta-oxidation.[21] Furthermore,

�9-THC undergoes hydroxylation in position eight to two

diastereomeric 8-hydroxy metabolites followed by dehydra-

tion as well as formation of an epoxide at C9–10 fol-

lowed by hydrolysis or glutathione conjugation.[20,21] Both

metabolic steps are catalyzed by CYP3A4.[20] In addition,

several combinations of the metabolic steps have been

described.[21]

Several models for estimation of the time of last cannabis

consumption and of the mental impact of �9-THC in relation to

the concentrations of �9-THC and one or two of its metabolites

have been established.[27 – 31] Two mathematic models for the

prediction of time of last cannabis use from the analysis of a

single plasma specimen were developed by Huestis et al.[29,30]

Model I is based on the �9-THC concentrations and model II is

based on the ratio of �9-THC-COOH and �9-THC concentrations.

Applying these models to clinical studies, they correctly predicted

the time of cannabis use for more than 90% of the specimens

evaluated, but tended to underestimate this interval at later

times. Therefore, today, the combination of model I and II is

recommended, which increases the accuracy of the predictions.

Furthermore, in clinical studies the test persons normally are

cannabinoid free and cannabis use is defined regarding time

and dose. However, most cannabis consumers suspected for

DUID are chronic users and take several doses within a few

hours.

The mental impact of �9-THC is estimated by the cannabis

influence factor (CIF) introduced by Daldrup et al.[31] The CIF

is based on the ratio of the sum of the molar concentra-

tions of �9-THC and 11-OH-�9-THC divided by the molar

concentration of �9-THC-COOH. The calculated CIF is lower

in frequent consumers due to accumulation of �9-THC-COOH.

Improved prediction and analytical evidence of the time of

last cannabis consumption would be helpful for the interpre-

tation of �9-THC concentrations. Possibly, �9-THCA-A and/or

its metabolite/s can provide valuable information regarding

the time between last cannabis consumption and blood sam-

pling.

The aim of the presented study was to identify the metabo-

lites of �9-THCA-A in rat urine, which could be used as

markers of cannabis consumption and to examine particularly

whether oral intake of �9-THCA-A leads to in vivo formation of

�9-THC.

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/jms

J. Jung et al.

Figure 1b. (Continued).

Experimental

Chemicals and materials

�9-THCA-A was isolated from Marijuana by an in-house

procedure with a purity of more than 95%.[32] The methano-

lic solutions (100 µg/ml) of (−)-�9-tetrahydrocannabinol-D3

(�9-THC-D3), (±)-11-hydroxy-�9-tetrahydrocannabinol-D3

(11-OH-�9-THC-D3) and (±)-11-nor-9-carboxy-

�9-tetrahydrocannabinol-D3 (�9-THC-COOH-D3) were purchased

from Promochem (Wesel, Germany). N-methyl-N-(trimethylsilyl)-

trifluoracetamide (MSTFA) was obtained from Sigma-Aldrich

(Steinheim, Germany). All other chemicals and biochemicals

were of the highest analytical grade and were provided by

Merck (Darmstadt, Germany). Deionized water was prepared on a

cartridge deionizer from Memtech (Moorenweis, Germany). Solid

phase extraction (SPE) columns (Chromabond C18, 3 ml, 500 mg)

were supplied by Macherey-Nagel (Duren, Germany).

Urine samples

Investigations were performed using the urine of male Wistar

rats (Charles River, Sulzfleck, Germany) for toxicological diagnostic

purposes according to the corresponding German law. Two rats

were administered a single 15 mg/kg body mass (BM) dose

of �9-THCA-A (10.1 mg �9-THCA-A/0.5 ml ethanol) by gastric

www.interscience.wiley.com/journal/jms Copyright c© 2009 John Wiley & Sons, Ltd. J. Mass. Spectrom. (2009)

�9-THCA-A metabolism in rats

Figure 1c. (Continued).

intubation. During the study, the rats were housed in metabolism

cages. The urine was separated from the faeces by a glass

construction designed in-house and collected over a 24 h period

in a fluoridated Erlenmeyer flask (2.5 mg potassium fluoride/1 ml

urine). The urine samples were pooled and aliquots of 200 µl were

stored at −20 ◦C prior to analysis. Blank rat urine samples had

been collected before drug administration to check whether they

were free of interfering compounds.

Sample preparation for identification of metabolites byLC-MS/MS and LC-QTOF MS

A 200 µl aliquot of urine was adjusted to pH 5.2 with acetic acid

(1 mol/l) and incubated at 50 ◦C for 1.5 h with 20 µl of a mixture of

β-D-glucuronidase (40 U/ml) and arylsulfatase (20 U/ml) from

Helix pomatia. After addition of 500 µl refrigerated acetonitrile for

protein precipitation and centrifugation at 4000 rpm for 10 min,

540 µl of the supernatant was transferred to an autosampler vial.

After adding 50 µl internal standard (IS) (50 ng �9-THC-COOH-D3)

to the samples, they were evaporated to dryness at 40 ◦C under

a gentle stream of nitrogen. For LC-MS/MS analysis, the residue

was reconstituted in 100 µl of the LC mobile phase (solvent

A : B, 90 : 10, v : v) and 20 µl were injected into the LC-MS/MS

system.

Another 100 µl aliquot of urine was prepared as described

above, but without enzymatic cleavage of conjugates prior to

protein precipitation.

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/jms

J. Jung et al.

Figure 1d. (Continued).

Figure 2. Reconstructed LC-MRM ion chromatogram of a pooled rat urinesample collected 24 h after oral intake of 15 mg/kg BM �9-THCA-A. Thenumbers correspond to those of the spectra and structures shown in Fig. 1.

Sample preparation for identification of �9-THC and itsmetabolites by GC-MS

After addition of 25 µl IS (5 ng �9-THC-D3, 5 ng 11-OH-

�9-THC-D3, 25 ng �9-THC-COOH-D3) to a 1 ml aliquot

of urine, the urine was adjusted to pH 5.2 with

acetic acid (1 mol/l) and incubated at 50 ◦C for

1.5 h with 100 µl of a mixture of β-D-glucuronidase

(40 U/ml) and arylsulfatase (20 U/ml) from Helix pomatia, then

diluted with 2 ml 0.1 M acetic acid, shortly mixed and transferred

to a SPE column preconditioned with 2 ml of methanol and of

0.1 M acetic acid each at a flow rate of 2 ml/min. After sample

loading at a flow rate of 1 ml/min, the SPE columns were washed

with 1 ml of 0.1 M acetic acid and of acetonitrile/water (70 : 30,

v : v) (flow rate: 1 ml/min) and dried under a gentle stream of

nitrogen for 2 min. Elution was performed with 1.5 ml of acetoni-

trile. The extracts were evaporated to dryness at 60 ◦C under a

gentle stream of nitrogen. After addition of 25 µl ethyl acetate and

25 µl MSTFA, derivatization was carried out at 90 ◦C for 45 min.

1 µl of the derivatized extracts were injected into the GC-MS

system.

Another aliquot of urine (1 ml) was prepared as described

above, but without enzymatic cleavage of conjugates, and one

with alkaline hydrolysis for cleavage of conjugates by adding

250 µl of 10 M sodium hydroxide to 1 ml of urine and incubation

at 60 ◦C for 45 min followed by neutralization with 2 ml of 0.1 M

acetic acid.

LC-MS/MS and LC-TOF-MS apparatus for identificationof metabolites

The LC-MS/MS system consisted of a SIL-20AC Prominence au-

tosampler, a CTO-20AC Prominence column oven, a CBM-20A

communications bus module, a DGM-20A3 Prominence degasser,

two LC-20AD SP Prominence liquid chromatography pumps

(Shimadzu GmbH, Duisburg, Germany) combined with a QTrap

3200 linear ion trap triple-quadrupole mass spectrometer (MS)

equipped with a TurboIonSpray interface and Analyst software

version 1.4.2 (Applied Biosystems/Sciex, Darmstadt, Germany).

Separation was performed at 40 ◦C with a Luna phenylhexyl

column (50 × 2 mm, 5 µm) fitted with a phenyl propyl guard

cartridge (4 × 2 mm) (Phenomenex, Torrance, CA, USA) using

gradient elution with 5 mM ammonium acetate pH 6.5 (sol-

vent A), and acetonitrile (solvent B) with a total flow rate of

0.2 ml/min and the following solvent gradient: 0–3 min, 10%

B; 3–20 min linear from 10 to 70% B; 20–21 min linear from

70 to 95% B; 21–23 min, 95% B; 23–25 min linear from 95 to

10% B; 25–30 min, 10% B. With a switching valve (Rheodyne,

Bensheim, Germany), the LC-effluent was admitted to the MS

only between 5 min and 21 min of the chromatographic reten-

tion time (RT). Negative electrospray ionization (ESI) was used

and the ion source was operated at 400 ◦C with a needle volt-

age of −4500 V and with nitrogen as curtain gas and nebulizer

gas. For the detection of the deprotonated molecules, a single-

quadrupole mass spectrum was acquired in the Q1-scan mode

(m/z 50–600 amu) and the enhanced mass spectrometer (EMS)

scan mode (m/z 50–600 amu) using three different decluster-

ing potentials (−30, −50 and −130 eV). For confirmation of the

www.interscience.wiley.com/journal/jms Copyright c© 2009 John Wiley & Sons, Ltd. J. Mass. Spectrom. (2009)

�9-THCA-A metabolism in rats

A

Figure 3a. Proposed metabolic pathway of �9-THCA-A in rats: phase I metabolism (A) and phase II metabolism (B). The numbers correspond to thoseof the spectra, structures and peaks shown in Fig. 1 and 2. The compounds in brackets are assumed intermediates.

glucuronides, precursor ion scan mode was used (�9-THCA-A glu-

curonide: precursor of m/z 357, scanned from m/z 345 to 745 amu;

11-OH-�9-THCA-A glucuronide: precursor of m/z 373, scanned

from m/z 345 to 745 amu; �9-THCA-A-COOH glucuronide: precur-

sor of m/z 387, scanned from m/z 345 to 745 amu; �9-THCA-A-8-

one glucuronide: precursor of m/z 371, scanned from m/z 345 to

745 amu; 8,11-Bis-OH-�9-THCA-A glucuronide: precursor of m/z

389, scanned from m/z 345 to 745 amu). Enhanced product ion

(EPI) spectra were recorded at three different collision energies

(−20, −35, and −50 eV) using the molecular ions found in the

Q1-scan mode and in the EMS scan mode as precursor ions. The

multiple reaction monitoring (MRM) method was set up using the

transitions obtained from the EPI spectra (Table 1).

For further confirmation the LC-Quadrupole Time-of-Flight MS

(LC-QTOF MS) system consisted of a UltiMate 3000 thermostatted

Analytical Sampler, a UltiMate 3000 thermostatted Column

Compartment, two UltiMate 3000 analytical pumps (Dionex

Corporation, Sunnyvale, USA) combined with a micrOTOFQ

and Bruker Daltonics DataAnalysis software version 3.4 (Bruker

Daltonics, Bremen, Germany). Separation was performed at 40 ◦C

with a Luna phenylhexyl column (50 × 2 mm, 5 µm) fitted with

a phenyl propyl guard cartridge (4 × 2 mm) (Phenomenex,

Torrance, CA, USA) using gradient elution with 5 mM ammonium

acetate pH 6.5 (solvent A) and acetonitrile (solvent B) with

a total flow rate of 0.2 ml/min and the following solvent

gradient: 0–3 min, 10% B; 3–20 min linear from 10 to 70%

B; 20–21 min linear from 70 to 95% B; 21–23 min, 95% B;

23–25 min linear from 95 to 10% B; 25–30 min, 10% B. The

micrOTOFQ was primarily operated in ‘RF-only’ modus, i.e.

MS/MS spectra were not acquired. An electrospray voltage of

4500 V was applied at the inlet of the MS (negative mode).

The ion transfer was optimized by means of sodium acetate

cluster (5 mM NaOH in 0.2% acetic acid in isopropanol : water,

1 : 1, v : v).

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/jms

J. Jung et al.

B

Figure 3b. (Continued).

GC-MS apparatus for identification of �9-THC and itsmetabolites

The samples were analyzed using a 6890 Series GC sys-

tem combined with a 5973 Series mass selective detector, a

7683 B Series injector and a Chem Station G1701GA version

D.03.00.611 (Agilent, Waldbronn, Germany). The GC conditions

were as follows: splitless injection mode; column, Optima-5-

MS capillary (30 m × 0.25 mm I.D., 0.25 µm film thickness)

(Macherey-Nagel, Duren, Germany); injection port temperature,

250 ◦C; carrier gas, helium; flow rate, 1.5 ml/min; oven tempera-

ture, initially 140 ◦C for 2 min, increased to 200 ◦C at 60 ◦C/min,

to 230 ◦C at 2.5 ◦C/min, to 310 ◦C at 60 ◦C/min, 310 ◦C for

3 min.

The MS conditions were as follows: transfer line heater, 280 ◦C;

ion source temperature, 230 ◦C; electron impact ionization (EI)

mode; ionization energy, 70 eV; electron multiplier voltage

(EMV), 400 V. Analysis was performed in selected-ion moni-

toring (SIM) mode using a fully validated standard procedure

for the detection of the trimethylsilylated (TMS) derivatives

of �9-THC, 11-OH-�9-THC and �9-THC-COOH with the

following program: solvent delay, 12 min; time window I,

12–16 min, m/z 306.2, 374.3, 389.3 (target ion, t) for

�9-THC-D3 and m/z 303.2, 371.3, 386.3 (t) for �9-THC;

time window II, 16–17 min, m/z, 374.3 (t), 462.4, 477.4 for 11-OH-

�9-THC-D3 and m/z, 371.3 (t), 459.4, 474.4 for 11-OH-�9-THC;

time window III, 17–18 min, m/z, 374.3 (t), 476.3, 491.3 for

�9-THC-COOH-D3 and m/z, 371.3 (t), 473.3,

488.3 for �9-THC-COOH. The limits of determina-

tion were 0.21 ng/ml for �9-THC, 0.08 ng/ml for

11-OH-�9-THC and 2.0 ng/ml for �9-THC-COOH, respec-

tively and the limits of quantification were 0.21 ng/ml for �9-THC,

0.28 ng/ml for 11-OH-�9-THC and 5–7 ng/ml for �9-THC-COOH,

respectively.

www.interscience.wiley.com/journal/jms Copyright c© 2009 John Wiley & Sons, Ltd. J. Mass. Spectrom. (2009)

�9-THCA-A metabolism in rats

Results and Discussion

Identification of the metabolites by LC-MS/MS andLC-QTOF MS

The urinary metabolites of �9-THCA-A were separated by LC and

identified by ESI-MS after protein precipitation with and without

enzymatic cleavage of conjugates. The analytical strategy for

identification of the metabolites was as follows: For the detection

of the deprotonated metabolites single-quadrupole mass spectra

were acquired in the Q1-scan mode and the EMS mode with the

linear ion trap applying three different declustering potentials

(−30, −90, −130 V). The existence of glucuronides was confirmed

using the precursor ion scan mode. EPI spectra were recorded with

collision energies of −20, −35, and −50 eV for each metabolite.

Molecular masses of the metabolites were verified by LC-QTOF

MS in the MS mode (quadrupole in ‘RF-only’ modus) with a mass

accuracy of better than ±10 ppm and the attribution of a correct

isotopic pattern (sigma fit).

The postulated structures of the metabolites were deduced

from the fragments detected in the EPI scan mode, which were in-

terpreted in correlation with those of the parent compound (Fig. 1).

The mass spectra, the structures and the predominant fragmen-

tation patterns of �9-THCA-A and its main metabolites are shown

in Fig. 1. The numbers of the respective mass spectra in Fig. 1 are

given in brackets. In the rat urine samples the following metabolites

of �9-THCA-A (retention time, RT: 15.84 min) (1) could be identi-

fied: �9-tetrahydrocannabinolic acid-A glucuronide (�9-THCA-A

glucuronide) (RT: 14.74 min and 12.40 min, respectively) (2),

11-hydroxy-�9-tetrahydrocannabinolic acid-A (11-OH-�9-THCA-

A) (RT: 13.54 min) (3), 11-hydroxy-�9-tetrahydrocannabinolic acid-

A glucuronide (11-OH-�9-THCA-A glucuronide) (RT: 12.63 min)

(4), 8β ,11-Bis-hydroxy-�9-THCA-A (8β ,11-Bis-OH-�9-THCA-A)

(RT: 11.56 min) (5), 8β ,11-Bis-hydroxy-�9-THCA-A glucuronide

(8β ,11-Bis-OH-�9-THCA-A glucuronide) (RT: 10.60 min) (6),

8α,11-Bis-hydroxy-�9-THCA-A (8α,11-Bis-OH-�9-THCA-A) (RT:

11.94 min) (7), 8α,11-Bis-hydroxy-�9-THCA-A glucuronide

(8α,11-Bis-OH-�9-THCA-A glucuronide) (RT: 11.05 min) (8),

�9-THCA-A-8-one (RT: 13.79 min) (9), �9-THCA-

A-8-one glucuronide (RT: 12.79 min) (10), 11-

nor-9-carboxy-�9-tetrahydrocannabinolic acid-A

(�9-THCA-A-COOH) (RT: 12.01 min) (11), and

11-nor-9-carboxy-�9-tetrahydrocannabinolic acid-A glucuronide

(�9-THCA-A-COOH glucuronide) (RT: 10.83 min) (12).

In the following paragraph, possible fragmentation patterns

of �9-THCA-A and its postulated metabolites are discussed.

�9-THCA-A (1) showed a molecular ion of m/z 357. Decarboxyla-

tion of the carboxy group in position two may lead to a fragment

ion of m/z 313. Neutral loss of C5H8 may lead to a fragment ion

of m/z 245 and an additional α-cleavage of the pentyl side chain

could produce a fragment ion of m/z 191. �9-THCA-A glucuronide

(2) showed a molecular ion of m/z 533. Loss of glucuronic acid

(�m = 176) would result in a fragment ion of m/z 357. Further

fragmentation was observed according to that of the aglycon

�9-THCA-A.

A molecular ion of m/z 373 was shown by 11-OH-�9-THCA-A

(3). Decarboxylation of the carboxy group in position two and

additional water elimination in position 11 may lead to a fragment

ion of m/z 311. The fragment ions of m/z 268 and of m/z 173 could

not be related to structures. A molecular ion of m/z 549 was shown

by 11-OH-�9-THCA-A glucuronide (4). Loss of glucuronic acid

(�m = 176) would result in a fragment ion of m/z 373. Further

fragmentation was observed according to that of the aglycon

11-OH-�9-THCA-A.

Also 8β ,11-Bis-OH-�9-THCA-A (5) and 8α,11-Bis-OH-

�9-THCA-A (7) showed molecular ions of m/z 389. Water elimina-

tion could produce a fragment ion of m/z 371 and additional

decarboxylation of the carboxy group in position two may

lead to a fragment ion of m/z 327. Further water elimination

would result in a fragment ion of m/z 309. The fragment ion of

m/z 312 could not be related to a structure. Molecular ions of

m/z 565 were shown by 8β ,11-Bis-OH-�9-THCA-A glucuronide (6)

and 8α,11-Bis-OH-�9-THCA-A glucuronide (8). Loss of glucuronic

acid (�m = 176) may lead to the fragment ion of m/z 389. Water

elimination and additional decarboxylation of the carboxy group

in position two would result in a fragment ion of m/z 327.

Also �9-THCA-A-8-one (9) showed molecular ions of m/z

371. Decarboxylation of the carboxy group in position two

may lead to a fragment ion of m/z 327. The fragment ions

of m/z 284 and m/z 189 could not be related to structures.

�9-THCA-A-8-one glucuronide (10) showed molecular ions of m/z

547. Loss of glucuronic acid (�m = 176) would result in a fragment

ion of m/z 371. Further fragmentation was observed according to

that of the aglycon �9-THCA-A-8-one.

A molecular ion of m/z 387 was shown by �9-THCA-A-COOH

(11). Decarboxylation of the carboxy group in position two would

result in a fragment ion of m/z 343 and further decarboxylation

of the carboxy group in position 11 may lead to a fragment ion

of m/z 299. Neutral loss of C4H6 could lead to a fragment ion of

m/z 245 and an additional α-cleavage of the pentyl side chain to

a fragment ion of m/z 191. The �9-THCA-A-COOH glucuronide

(12) showed a molecular ion of m/z 563. Loss of glucuronic acid

(�m = 176) may lead to the fragment ion of m/z 387. Further

fragmentation was observed according to that of the aglycon

11-nor-9-carboxy-�9-THCA-A.

A MRM method was set up using the transitions obtained

from the EPI spectra (Fig. 1). Using the MRM mode, two

�9-THCA-A glucuronides were detected (Fig. 2). Mass spectro-

metric differentiation was not possible, however differences in the

RT strongly suggest the compound eluting at 12.40 min to be the

ether glucuronide linked at the hydroxy group in position one and

the compound eluting at 14.74 min to be the ester glucuronide

linked at the carboxy group in position two. Furthermore, another

monohydroxylated �9-THCA-A metabolite was detected in the

MRM mode, but the metabolite could not be related to a structure.

It can be assumed, that the metabolite is hydroxylated in the side

chain.

The molecular formulas of the metabolites were confirmed by

accurate mass measurement using a high resolution mircrOTOF Q

with a tolerance of ±10 ppm for the major and of ±20 ppm for

the minor metabolites (Table. 2).

Identification of �9-THC and metabolites by GC-MS

For the detection of �9-THC and its metabolites, the rat

urine samples were analyzed by GC-MS in the SIM mode after

alkaline hydrolysis, enzymatic cleavage of conjugates and without

cleavage of conjugates, respectively, SPE and trimethylsilylation.

Neither in the native urine sample nor in the urine sample

incubated with β-D-glucuronidase/arylsulfatase, �9-THC or

its metabolites 11-OH-�9-THC and �9-THC-COOH could be

detected. Although β-D-glucuronidase/arylsulfatase from Helix

pomatia is not as effective as β-D-glucuronidase/arylsulfatase from

E. coli in cleavage of some glucuronides, earlier experiments in our

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/jms

J. Jung et al.

Table 2. Results of LC-QTOF MS screening of a pooled rat urine sample collected 24 h after oral intake of 15 mg/kg BM �9-THCA-A for theconfirmation of the �9-THCA-A metabolites

MetaboliteCalculated mass

[M-H]− (amu) Suggested formulaMeasured mass[M-H]− (amu) Error (ppm) Sigma fit

�9-THCA-A 357.2071 C22H29O4 357.2089 −4.9 0.0214

�9-THCA-A glucuronide 533.2390 C28H37O10 − − −

11-OH-�9-THCA-A 373.2020 C22H29O5 373.2016 1.1 0.0096

11-OH-�9-THCA-A glucuronide 549.2341 C28H37O11 549.2346 −0.9 0.0232

8β ,11-Bis-OH-�9-THCA 389.1970 C22H29O6 389.1959 2.6 0.0224

8β ,11-Bis-OH-�9-THCA glucuronide 565.2284 C28H37O12 +a +a +a

8α,11-Bis-OH-�9-THCA 389.1970 C22H29O6 389.2005b −9.8b −

8α,11-Bis-OH-�9-THCA glucuronide 565.2284 C28H37O12 +a +a +a

�9-THCA-A-8-one 371.1864 C22H27O5 371.1864 0.1 0.0433

�9-THCA-A-8-one glucuronide 547.2178 C28H35O11 +a +a +a

�9-THCA-A-COOH 387.1813 C22H27O6 387.1815 −0.4 −

�9-THCA-A-COOH glucuronide 563.2134 C28H35O12 563.2119 2.7 0.139c

a Signal with low abundance, not sufficient for unequivocal identification.b Partly resolved interference observed.c Overlapping isotopic pattern.

research group showed that glucuronides of �9-THC and

11-OH-�9-THC can be cleaved using β-D-

glucuronidase/arylsulfatase from Helix pomatia. Furthermore,

�9-THC-COOH, which is freed up from its glucuronide by β-D-

glucuronidase/arylsulfatase from Helix pomatia, would present as

the known main metabolite in urine, if �9-THCA-A was converted

to �9-THC. Therefore, in this case it would have been possible to

identify at least �9-THC-COOH after enzymatic cleavage in the

rat urine, which was not the case and proves that �9-THCA-A

and/or its major oxidative metabolites are not decarboxylated

metabolically.

However, after alkaline hydrolysis �9-THC as well as

11-OH-�9-THC and �9-THC-COOH were identified. Because

�9-THCA-A is readily converted into �9-THC

at alkaline pH values,[33] it can be con-

cluded, that �9-THCA-A and its main metabolites

11-OH-�9-THCA-A and �9-THCA-A-COOH were partly de-

carboxylated due to the high pH during the hydrolysis

step resulting in detection of �9-THC, 11-OH-�9-THC and

�9-THC-COOH.

Proposed metabolic pathway

On the basis of the identified metabolites, the following metabolic

pathway of �9-THCA-A, shown in Fig. 3 (A and B), can be

postulated: �9-THCA-A undergoes a single hydroxylation in

position 11 to 11-OH-�9-THCA-A. Via the intermediate alde-

hyde 11-oxo-�9-THCA-A, 11-OH-�9-THCA-A is further oxidized to

�9-THCA-A-COOH. The parent compound and both main metabo-

lites were also excreted as glucuronides, i.e. �9-THCA-A glu-

curonide, 11-OH-�9-THCA-A glucuronide and �9-THCA-A-COOH

glucuronide.

Additionally, �9-THCA-A undergoes further single hydroxy-

lation in position 8 to 8α- or 8β-OH-�9-THCA-A followed by

dehydration to �9-THCA-A-8-one. Both metabolites are further

oxidized in position 11 to their bishydroxylated forms. Also ex-

creted as glucuronides are 8α,11-Bis-OH-�9-THCA, 8β ,11-Bis-OH-

�9-THCA and �9-THCAA-8-one, respectively, i.e. 8α,11-Bis-OH-

�9-THCA glucuronide, 8β ,11-Bis-OH-�9-THCA glucuronide and

�9-THCA-A-8-one glucuronide, respectively. Although several of

the metabolites have multiple hydroxyl and carboxyl groups that

could have multiple glucuronide moieties, di-or tri-glucuronides

were not identified. There was no evidence for the transformation

of �9-THCA-A to �9-THC.

Limitations

These results provide information on the qualitative metabolism

of �9-THCA-A, but their generalizability might be limited because

they only reflect the metabolism in two rats at a single point

in time after a single dose of �9-THCA-A. The dosage was

relatively high, which could have recruited secondary rather than

primary metabolic pathways. Humans may metabolize�9-THCA-A

differently; therefore, the comparability of metabolism in humans

and in rats has to be checked with regard to the kind of metabolites

formed and the kinetics of their formation.

Conclusions

Twelve metabolites of �9-THCA-A were detected and iden-

tified in rat urine after a single oral dose. Hydroxylation of

�9-THCA-A in position 11 to 11-OH-

�9-THCA-A and further oxidation to

�9-THCA-A-COOH are the main phase I steps. The

�9-THCA-A and both main metabolites were also present

as glucuronides. Neither �9-THC nor its metabolites were

detected, which shows, that no in vivo decarboxylation of

�9-THCA-A and/or its main oxidative metabolites does occur

after oral application in rat.

The metabolism studies presented here show that the main

metabolites of �9-THCA-A are formed in close analogy to

�9-THC metabolism. It can be assumed that these metabolites

can be detected in serum and urine after cannabis consumption;

therefore, kinetic studies of �9-THCA-A and its metabolites may

lead to new markers for recent cannabis abuse. In future studies,

the duration of detectability of THCA-A and its metabolites after

single and multiple consumptions of cannabis, respectively, should

be determined and these data should be compared with those for

THC and its metabolites.

www.interscience.wiley.com/journal/jms Copyright c© 2009 John Wiley & Sons, Ltd. J. Mass. Spectrom. (2009)

�9-THCA-A metabolism in rats

Acknowledgements

The authors thank Gabriela Herzog, Jurgen Kempf, Claudia Steinert,

Ariane Wohlfarth, Gabriele Ulrich, Andreas H. Ewald, Christoph

Sauer and Svenja C. Bunz for their help.

References

[1] C. E. Turner, M. A. Elsohly, E. G. Boeren. Constituents of CannabisSativa L. XVII. A review of natural constituents. Journal of NaturalProducts 1980, 43, 169.

[2] R. Mechoulam, S. Ben-Shabat. From gan-zi-gun-nu to anandamideand 2-arachidonoylglycerol: the ongoing story of cannabis. NaturalProduct Reports 1999, 16, 131.

[3] T. Yamauchi, Y. Shoyama, H. Aramaki, T. Azuma, I. Nishioka.Tetrahydrocannabinolic acid, a genuine substance oftetrahydrocannabinol. Chemical and Pharmaceutical Bulletin 1967,15, 1075.

[4] Y. Shoyama, I. Yamauchi, V. Nishioka. Cannabis, Cannabigerolicacid monomethyl ether and cannabinolic acid. Chemical andPharmaceutical Bulletin 1970, 18, 1327.

[5] G. S. Lewis, C. E. Turner. Constituents of Cannabis sativa L. XIII-Stability of dosage form prepared by impregnating synthetic(−)-�9-trans-tetrahydrocannabinol on placebo Cannabis plantmaterial. Journal of Pharmaceutical Sciences 1978, 67, 876.

[6] R. K. Razdan, A. J. Puttick, B. A. Zitko, G. R. Handrick. HashishVI: conversion of (−)-1(6)-tetrahydrocannabinol to (−)-1(7)-tetrahydrocannabinol. Stability of (−)-1- and (−)-1(6)-tetrahydrocannabinols. Experientia 1972, 28, 121.

[7] F. E. Dussy, C. Hamberg, M. Luginbuhl, T. Schwerzmann,T. Briellmann. Isolation of �9-THCA-A from hemp and ana-lytical aspects concerning the determination of �9-THC in cannabisproducts. Forensic Science International 2005, 149, 3.

[8] J. Jung, J. Kempf, H. Mahler, W. Weinmann. Detection of �9-tetrahydrocannabinolic acid in human urine and blood serumby LC-MS/MS. Journal of Mass Spectrometry 2006, 42, 354.

[9] J. Jung, V. Auwaerter, G. Herzog, H. Mahler, J. Kempf, W. Weinmann.Determination of �9-Tetrahydrocannabinolic Acid A in SerumSamples by GC-MS. In Proceedings of the XV. GTFCh-Symposium,F. Pragst, R. Aderjan (eds). Mosbach (Baden): Bad Vilbel, 2007, 378.

[10] C. Moore, S. Rana, C. Coulter. Simultaneous identification of 2-carboxy-tetrahydrocannabinol, tetrahydrocannabinol, cannabinoland cannabidiol in oral fluid. Journal of Chromatography B 2007,852, 459.

[11] B. Costa. On the pharmacological properties of�9-tetrahydrocannabinol (THC). Chemistry and Biodiversity2007, 4, 1664.

[12] D. Baker, G. Pryce, G. Giovannoni, A. J. Thompson. The therapeuticpotential of cannabis. Lancet Neurology 2003, 2, 291.

[13] J. L. Croxford. Therapeutic potential of cannabinoids in CNS disease.CNS Drugs 2003, 17, 179.

[14] E. M. Williamson, F. J. Evans. Cannabinoids in clinical practice. Drugs2000, 60, 1303.

[15] M. Guzman. Cannabinoids: potential anticancer agents. Naturereviews. Cancer 2003, 3, 745.

[16] J. L. Croxford, S. D. Miller. Towards cannabis and cannabinoidtreatment of multiple sclerosis. Drugs of Today 2004, 40, 663.

[17] H. H. Maurer, C. Sauer, D. S. Theobald. Toxicokinetics of drugs ofabuse: current knowledge of the isoenzymes involved in the humanmetabolism of tetrahydrocannabinol, cocaine, heroin, morphineand codein. Therapeutic Drug Monitoring 2006, 28, 447.

[18] T. M. Bland, R. L. Haining, T. S. Tracy, P. S. Callery. CYP2C-catalyzed delta(9)-tetrahydrocannabinol metabolism: kinetics,pharmacogenetics and interaction with phenytoin. BiochemicalPharmacology 2005, 70, 1096.

[19] K. Watanabe, T. Matsunaga, I. Yamamoto, Y. Funae, H. Yoshimura.Involvement of CYP2C in the metabolism of cannabinoids byhuman hepatic microsomes from an old woman. Biological andPharmaceutical Bulletin 1995, 18, 1138.

[20] L. M. Bornheim, J. M. Lasker, J. L. Raucy. Human hepatic microsomalmetabolism of �1-tetrahydrocannabinol. Drug metabolism andDisposition: the Biological Fate of Chemicals 1992, 20, 421.

[21] M. M. Halldin, M. Widman, V. Bahr, J. E. Lindgren, B. R. Martin.Identification of in vitro metabolites of �1-tetrahydrocannabinolformed by human livers. Drug Metabolism and Disposition: theBiological Fate of Chemicals 1982, 10, 297.

[22] S. Narimatsu, K. Watanabe, T. Matsunaga, I. Yamamoto, S. Imaoka,Y. Funae, H. Yoshimura. Cytochrome P-450 isozymes in metabolicactivation of �9-tetrahydrocannabinol by rat liver microsomes.Drug Metabolism and Disposition 1990, 18(6), 943.

[23] S. Narimatsu, K. Watanabe, T. Matsunaga, I. Yamamoto, S. Imaoka,Y. Funae, H. Yoshimura. Cytochrome P-450 isozymes involved inthe oxidative metabolism of �9-tetrahydrocannabinol by livermicrosomes of adult female rats. Drug Metabolism and Disposition1992, 20(1), 79.

[24] M. E. Wall, M. Perez-Reyes. The metabolism of�9-tetrahydrocannabinol and related cannabinoids in man.Journal of Clinical Pharmacology 1981, 21, 178S.

[25] M. M. Halldin, M. Widman. Glucuronic acid conjugate of �1-tetrahydrocannabinol in the urine of man. Arzneimittelforschung1983, 33, 177.

[26] P. L. Williams, A. C. Moffat. Identification in human urine of �9-tetrahydrocannabinol-11-oic acid glucuronide: a tetrahydro-cannabinol metabolite. The Journal of Pharmacy and Pharmacology1980, 32, 445.

[27] M. A. Huestis, M. Elsohly, W. Nebro, A. Barnes, R. A. Gustafson,M. L. Smith. Estimating time of last oral ingestion of cannabisfrom plasma THC and THCCOOH concentrations. Therapeutic DrugMonitoring 2006, 28, 540.

[28] M. A. Huestis, A. Barnes, M. L. Smith. Estimating the time of lastcannabis use from plasma �9-tetrahydrocannabinol and 11-nor-9-carboxy-�9-tetrahydrocannabinol concentrations. ClinicalChemistry 2005, 51, 2289.

[29] M. A. Huestis, J. E. Henningfield, E. J. Cone. Blood cannabinoids. I.Absorption of THC and formation of 11-OH-THC and THCCOOHduring and after smoking marijuana. Journal of Analytical Toxicology1992, 16, 276.

[30] M. A. Huestis, J. E. Henningfield, E. J. Cone. Blood cannabinoids II:models for the prediction of time of marijuana exposure fromplasma concentrations of �9-tetrahydrocannabinol (THC) and 11-nor-9-carcoxy-�9-tetrahydrocannabinol (THC-COOH). Journal ofAnalytical Toxicology 1992, 16, 283.

[31] T. Daldrup, I. Meininger. Cannabis im Straßenverkehr, 1st ed. GustavFischer Verlag: Stuttgart, Jena, Lubeck, Ulm, 1998, 193.

[32] J. Jung, H. Mahler, E. Breitling, M. Mueller, F. T. Peters, H. H. Maurer,W. Weinmann, V. Auwaerter. New chromatographic method for theisolation of �9-tetrahydrocannabinolic acid A (�9-THCA-A) fromCannabis sativa. Poster at the 46th International Meeting of theTIAFT, 2–8 June 2008, Martinique, French West Indies. (Annales deToxicologie Analytique 2008, 20(S1), 75).

[33] F. Grotenhermen. Pharmacokinetics and pharmacodynamics ofcannabinoids. Clinical Pharmacokinetics 2003, 42(4), 327.

J. Mass. Spectrom. (2009) Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/jms