CONTAMINANT TESTING IN MARIJUANA: PESTICIDES, MYCOTOXINS AND RESIDUAL SOLVENTS Katherine K. Stenerson, Olga I. Shimelis, Michael Ye, Michael Halpenny
MilliporeSigma is a business of Merck KGaA, Darmstadt, Germany
Cannabis in the United States The Current State of Things... Recreational Use Legal in 4 US states (Colorado, Washington, Oregon, Alaska) Medical Use Legal in 25 states Testing Contaminants Pesticides Mycotoxins Residual solvents Heavy metals Profiling and content in plant material Cannabinoids
Terpenes No standardized methods currently exist
Agenda
1
Pesticide Residues
QuEChERS extraction and cleanup
2
Residual Solvents
Solid Phase Microextraction (SPME)
GC/MS/MS analysis
Quick and simple method
Evaluated two different cleanups
Quantitative analysis using hemp extract
3
Mycotoxins
Extract cleanup using SupelTox SPE Alternative to immunoaffinity Simpler and easier method
Pesticide Residues
Analyte list chosen based on current testing; includes different classes of pesticides: • • • •
Triazole fungicides Organophosphorus Organochlorine Pyrethroid
Marijuana sample provided courtesy of Dr. Hari H Singh*.
*Program Director at the Chemistry & Physiological Systems Research Branch of the National Institute on Drug Abuse at the National Institute of Health.
Analysis of Pesticide Residues in Marijuana
• QuEChERS extraction & cleanup • Analysis by GC/MS/MS • Analyzed samples unspiked and spiked (50 ng/g) QuEChERS procedure used
1 g ground cannabis + 10 mL water
5
Add 10 mL of acetonitrile, shake for 10 min at 2500 rpm
Add citrate salts and shake for 1 min
Centrifuge, 5000 rpm/5 min
Transfer 1 mL of supernatant to 2 mL tube containing cleanup sorbent. Shake for 2 min
Centrifuge, 5000 rpm/5 min
Draw off supernatant for GC/MS/MS analysis
Evaluated two sorbents for cleanup 1. PSA/C18/GCB 2. Supel QuE Verde
What is Supel QuE Verde ??? Supelclean™ ENVI-Carb™ Y
Supel QuE Z-Sep+
Supelclean PSA
•Z-Sep+ Zr & C18 functionalized silica ZrO2 C18 ZrO2 C18 ZrO2 C18 ZrO2 C18 6
Specialized GCB that balances chlorophyll removal and analyte recovery Removes unwanted pigment (color) and lipid (fat) interferences Removes unwanted fatty acids, organic acids, polar pigments, sugars ENVI-Carb Y Specially engineered graphitized carbon black
0.00E+002.00E+074.00E+07
GC/MS Scan analysis of extract background in marijuana extracts no cleanup Sum of all peak areas = 540,827,547
10
20
30
PSA/C18/ENVI-Carb Sum of all peak areas = 449,294,987
10
20
Supel QuE Verde no cleanup
30 T ime (min)
0.00E+00 2.00E+07 4.00E+07
0.00E+002.00E+074.00E+07
Time (min)
Supel QuE Verde Sum of all peak areas = 281,530,256
10
20
30 T ime (min)
cannabinoids
PSA/C18/ENVI-Carb
Analyte Recovery Avg. overall % recovery PSA/C18/ENVI-Carb: 71% Supel QuE Verde: 75%
Avg. overall % RSD PSA/C18/ENVI-Carb: 8% Supel QuE Verde: 5%
120 110 100
Supel QuE PSA/C18/ENVI-Carb Supel QuE Verde
Avg. % Recovery
90 80 70 60 50 40 30
20 10 0
8
Pesticides
Marijuana oil produced by extraction of cannabis flower buds
Extraction often uses organic solvents
Residual Solvents
Some solvent can remain behind in the final extract Testing can be done by headspace GC Traditional headspace requires a separate analyzer connected to the GC SPME can be used as an alternative; and can be automated with an X-Y-Z autosampler
What is “SPME”? Solid phase microextraction
10
•
Solvent-free extraction technique for nearly any sample or matrix
•
Alternative to headspace GC and solid phase extraction (SPE) techniques
•
Directly interfaced with GC analysis
•
Quantitative
•
Non-destructive to sample
•
Reusable (50-100+ times)
•
Inexpensive
•
Fast
•
Easy to automate SPME process
SPME Fiber Coating: The Business End • Not an exhaustive extraction technique • An equilibrium is set up between analytes dissolved in the sample (solution or gas phase) and in the liquid coating on the fiber. • The fiber coatings consist of: • Polymer films (e.g. PDMS)
• Particles + binder (e.g. carbons or DVB in PDMS)
Enlargement of the SPME fiber coating 11
Equilibrium of analyte conc. in fiber and sample
Details of Analysis
Residual Solvents Tested
Samples: Pure hemp oil, spiked at 10 µg/g (triplicate analyses)
Peak #
Solvent
Class
4
Acetone
III
3
Acetonitrile
II
8
Benzene
I
9
Cyclohexane
II
2
Ethanol
III
GC/MS, full scan
10
Heptane
III
7
Hexane
II
Supel-Q™ PLOT, 30 m x 0.32 mm I.D. capillary column
5
Isopropanol
III
1
Methanol
II
6 11 12&13
Tetrahydrofuran
II
Toluene
II
Xylene (o,m,p)
II
Soybean oil blanks Quantitation: external standard 6-point calibration curve (6-100 µg/g) in soybean oil Analysis:
12
3
1
2 4
Class per ICH guidelines
7
4
6
8
10 5
10
11
8,9
13
6
12 14 Time (min)
16
18
Oven: 50°C (5 min), 10°C/min to 230°C (5 min) Carrier: He, 2 mL/min constant flow Splitter open during injection/desorption (10:1)
20
22
Headspace SPME Method for Residual Solvents Sample/matrix:
SPME Fiber:
Extraction:
Desorption:
Fiber Postbake:
5 g hemp oil in 10 mL vial
Carboxen®/PDMS, 75µm (CAR/PDMS) Strong adsorbent fiber; provides retention of light compounds- down to C3. 5 min, headspace, 40°C At 40°C, only a short extraction time is needed.
3 min, 320°C; split 10:1 High temp. used to efficiently and completely desorb analytes. High sensitivity of SPME requires split of 10:1 to prevent overload 2 min, 320°C Cleans fiber & prevents carryover
Method Calibration For Residual Solvents; HS SPME using CAR/PDMS Fiber 3000000 R² = 0.9858
Response (absolute)
2500000
R² = 0.9869
2000000 methanol THF
1500000
heptane R² = 0.9936
1000000
o xylene R² = 0.9806
R² = 0.9806
isopropanol
R² = 0.9864
500000
0 0
20
40
60
80
100
120
Conc. (ug/g)
Standards made using soybean oil Overload starting at 70 ug/g for some compounds 14
HS SPME Method; Measurement Accuracy & Reproducibility 10 ug/g spiking level in hemp oil n=3 % RSD 140%
% Accuracy
100%
80%
8%
7%
120% 6% 5% 3%
8%
7%
9%
6%
9%
60% 40% 20%
0%
Detected in unspiked hemp oil at 58.5 ug/g
7%
6%
Method accuracy 80% for all compounds Good reproducibility: RSDs < 10% High level of hexane detected in unspiked hemp oil
Produced by fungi present on cannabis plants They present acute and chronic toxicity risks to humans - including suspected carcinogenicity
Mycotoxins
Quantitative analysis involves extraction and cleanup followed by chromatographic separation (GC and HPLC) Complex matrix of cannabis necessitates the need for extract cleanup Extract Cleanup usually done using immunoaffinity; however SPE offers an easier alternative
Methods for Sample Cleanup for Mycotoxin Analysis Method Immunoaffinity (IAC)
Bind & elute
SPE
Interference removal or bind & elute
Issues with IAC
# of Steps (minimum)
Storage
Type of cartridges
3
Refrigeration
Mycotoxin specific
1 or 3
Room temp.
Mycotoxin specific
• • •
Benefits of SPE
• •
• 17
Thermally unstable Introduce many sources of variability due to the complexity of the required procedure. Require many steps: utilize “Bind and Elute” principle Do not require refrigerated storage Simple procedure; can utilize chemical filtration approach Does not require dilution of sample with water prior to cleanup
Analysis of Aflatoxins in Cannabis
1 2
3
4
Used dried cannabis for extraction obtained courtesy of Dr. H. Hari Singh at NIH
Spiked at ppb levels with Aflatoxins B1, B2, G1, G2 Spiking done after extraction and prior to cleanup to evaluate recovery performance of SPE cleanup
Cleanup done using Supel™ Tox AflaZea SPE Simple interference removal approach
Final analysis by LC/MS/MS Using matrix-matched calibration curve
Sample preparation & cleanup of aflatoxins in cannabis using SPE Extraction of homogeneous sample (0.5 g in 10 mL) Shaken 90 minutes in 86:14 ACN:H20
Filter and spike
SPE – Supel Tox AflaZea
Dilution with water – no precipitate (200 uL to FV 1 mL)
LC-MS/MS analysis 19
RESULTS AFLATOXINS IN Cannabis 200
•
180
Ascentis Express Phenyl-Hexyl 10 cm x 2.1 mm, 2.9 um particle size
160 counts
140 120
aflatoxin G2
100
aflatoxin G1
80
afaltoxin B2
60
afaltoxin B1
40
20 0 2.5
3
3.5
4
•
NH4formate/formic acid watermethanol gradient
•
10µL injection
•
MS, ESI(+), MRM 331.3/189.0, 329.1/243.0, 315.9/259.0, 313.1/241.0
4.5
Time (minutes)
Aflatoxin B1
Aflatoxin B2
Aflatoxin G1
Aflatoxin G2
24.4 ppb
6.1 ppb
24.4 ppb
6.1 ppb
Recovery (%)
102
109
108
127
RSD% (n=3)
8
12
3
9
Spiking level
Matrix matched calibration standards were used
20
Cannabis Matrix Effects
4000
Aflatoxin B1
900
3500
800
3000
700
counts
in cannabis extract
2000
in solvent
1500
counts
600
2500
500 400 300
1000
200
500
100
0 0.0000
Aflatoxin B2
0.5000
1.0000
concentration (ng/mL)
0 0.0000
in cannabis extract in solvent 0.1000
0.2000
0.3000
concentration (ng/mL)
Significant matrix effect observed Use of matrix-matched standards required for accurate quantitation Supel Tox AflaZea cleanup produced a clear extract but did not remove all interfering compounds 21
Summary & Conclusions
1
QuEChERS extraction and cleanup is a viable approach for analysis of pesticides in marijuana
2
•
Use of a cleanup sorbent containing GCB is recommended for chlorophyll removal
•
Compared to PSA/C18/GCB, Supel QuE Verde cleanup showed lower overall GC/MS background and better recovery of quinoxyfen, a planar pesticide
Residual solvents in hemp (and cannabis) extract can be quickly and quantitatively analyzed by HSSPME •
The method using the CAR/PDMS fiber could be extended to include analysis of lighter compounds such as butane
•
The SPME method offers an alternative to conventional headspace; and does not require the use of a separate headspace analyzer
3
SPE cleanup can be used in mycotoxin analysis of cannabis •
Supel Tox AflaZea offers a simpler alternative to IAC cleanup
•
Cleanup using AflaZea SPE allows for accurate detection of mycotoxins at ppb levels with the use of matrix-matched standards
Acknowledgments Many Thanks to…. Dr. Hari H. Singh, Program Director at the Chemistry & Physiological Systems Research Branch of the National Institute on Drug Abuse at the National Institute of Health for supplying the dried cannabis sample used for testing Yong Chen and Bob Shirey of MilliporeSigma for many helpful discussions on SPME
23
Thank you for your attention