HSF- Thesis

Method Development Towards Quantifying Marijuana
Consumption Using Sewage Based Drug Epidemiology
By
Heather Fryhle
A dissertation submitted in partial fulfillment
of the degree requirements for the degree of
Bachelor of Science
University of Puget Sound
Department of Chemistry
Advisor: Dr. Daniel A. Burgard
2015

Approved by____________________________________________________________
Fryhle 2

Table of Contents
1. INTRODUCTION 4
1.1. MARIJUANA USAGE 4
1.2. WASTEWATER TESTING 5
1.2.1. Solid Phase Extraction of 11-nor-9-carboxy- 6
tetrahydrocannabinol
1.2.2. Liquid Chromatography Tandem Mass Spectrometry 6
1.3. RESEARCH METHODS 7
2. MATERIALS AND METHODS 8
2.1. STANDARDS AND REAGENTS 8
2.1.1 Standard Calibration Curves 10
2.2. GENERAL METHODS 10
2.2.1. Wastewater Samples 10
2.2.2. Lab Maintenance 10
2.3. EXTRACTION AND QUANTIFICATION PROCEDURES 10
2.3.1. Testing of HLB, MCX, and MAX SPE Cartridges 10
2.3.2. Testing of Strata-XC SPE Cartridge 11
2.3.3. SPE vs. Lyophilizing 12
2.3.4. Filtration 13
2.3.5. Testing of LC-MS/MS Optimal Ion Mode 13
2.3.6. Testing of Optimal Eluents 14
2.3.7. Testing of D3 vs. D9 Internal Standards 14
2.3.8. Final Extraction Procedure 14
3. RESULTS AND DISCUSSION 16
3.1. HLB, MCX, AND MAX SPE CARTRIDGE RESULTS 16
3.2. STRATA-XC INITIAL CARTRIDGE RESULTS 17
3.3. SPE VS. LYOPHILING RESULTS 18
3.4. FILTRATION RESULTS 19
3.5. TESTING OF LC-MS/MS OPTIMAL ION MODE RESULTS 22
3.6. TESTING OF OPTIMAL ELUENT RESULTS 23
3.7. D3 VS. D9 INTERNAL STANDARD RESULTS 26
3.8. FINAL METHODS PRELIMINARY RESULTS 27
4. CONCLUSIONS 27
5. ACKNOWLEDGEMENTS 29
6. REFERENCES 30
7. APPENDICES 33
7.1. APPENDIX A 33
7.2. APPENDIX B 36
Fryhle 3

1. INTRODUCTION
1.1. MARIJUANA USAGE
Marijuana (or cannabis) is currently the most abused illicit drug in the world.1-5
Since it was first banned in California 100 years ago, the last decade has seen a rise in
recreational marijuana use in the United States.6 Use is becoming increasingly prevalent
within the youth population of the U.S. and a huge part of the youth culture in the
Netherlands, Portugal, and Uruguay. For example, cannabis usage in Amsterdam in
adults ranging from ages 18-20 rose from 34% to 44% between 1987 and 1997.7 In 2001,
Portugal decriminalized all drugs including marijuana (though it is still not legalized).8,9
Though it has been decriminalized, both the Netherlands and Portugal do not yet have the
means to sell or grow cannabis legally.5 In addition, the implementation of medical
marijuana in over 20 states has become a pressing issue.4 Washington and Colorado were
the first two states in the U.S. to legalize adult recreational marijuana in 2012 (by
Initiative-502 and the Amendment 64) with hopes that they could reap the benefits of
marijuana sales from the black market, such as revenue from tax income as well as
reduction in the costs of incarceration. Washington, Colorado, and Uruguay, who began
selling cannabis starting April 2014, will be the first to implement legal growing and
sales.10 However, since medical marijuana will likely not be taxed, it is possible that
users will still turn to “green cards” and the black market for recreational pot instead of
legal dispensaries (which will be taxed at a total of 75% from growing, producing, and
retail).5
There is no current scientific research regarding the effects of these new laws and
it is difficult to know what the implications of legalization will be. Since dispensaries
opened in Colorado, they have noticed an increase in accidents involving driving under
the influence of marijuana.11 Many studies have shown negative effects from marijuana
in coordination, thinking, problem solving, learning, and memory.4 According to a study
at Columbia University, the percentage of drivers killed in car accidents who tested
positive for marijuana nearly tripled between 1999 and 2010.11 With legalization, it is
likely that these numbers will continue to rise. Determining usage patterns associated
with marijuana legalization may come too late for Washington, but it will help weigh the
pros and cons for other states considering legalization. This crucial transition period at
Fryhle 4

the beginning of legalized recreational marijuana allows for a unique opportunity to
gather data concerning trends in marijuana usage due to its legalization.
Typically, studies concerning illicit drug use come from self-report surveys,
police reports, and autopsies. However, in the last 10 years a promising new type of
research has evolved for the quantification of illicit drugs. Research teams in Western
Europe have used a new method called sewer-based drug epidemiology to quantify
several illicit compounds in wastewater. A 2005 study in Milan, Italy specifically
assessed the concentration of cocaine’s metabolite, benzoylecgonine, discovering that it
was present in untreated urban wastewater and a major river.12 The significance of this
research prompted teams in Switzerland, Spain, England, parts of Italy, and several other
countries (mostly European) to perform similar experiments on several other substances
(and/or their metabolites) including heroine, ecstasy, ephedrine, tetrahydrocannabinol
(THC), and the main metabolite of THC, 11-nor-9-carboxy-tetrahydrocannabinol (THC-
COOH). 1,-3,13-15All of these were abundantly detected in raw sewage samples except for
THC, due to its hydrophobic properties.1,-3,14-15 Analysis of marijuana by sewer based
drug epidemiology is much more effective if THC-COOH is quantified because it is more
hydrophilic and thus more likely to be found in the aqueous phase. In addition, analyzing
the metabolite gives information about the quantities of marijuana that were actually
consumed. This research has sought to apply sewer-based drug epidemiology to THC-
COOH in influent sewage samples in South Tacoma, WA pre- and post-legalization, as
well as focus on method development for optimal quantification.
1.2. WASTEWATER TESTING
Analysis of drugs and their metabolites in sewage or wastewater typically relies
on two methods, solid phase extraction (SPE) and liquid chromatography tandem mass
spectrometry (LC-MS/MS) for isolation and quantification of drug analytes. 1-3,12-15Using
this method of wastewater testing, I attempted to test levels of THC-COOH in wastewater
samples collected both before and after legalization. To aid in this comparison, a human
population biomarker compound was identified and a method was developed to account
for wastewater dilution and excretion rates. The biomarker for this project is 5-
hydroxyindoleacetic acid (5-HIAA), a break-down product of serotonin that is regularly
Fryhle 5

produced by the human body at a fairly constant rate. THC-COOH concentrations can be
normalized against 5-HIAA concentrations to better compare daily THC-COOH levels.
THC-COOH concentrations in wastewater are very low, in the parts per trillion (ppt)
range. For this reason, raw influent sewage samples must be highly concentrated for the
metabolite to be detected by LC-MS/MS. This leads us to the need for SPE.
1.2.1. Solid Phase Extraction of 11-nor-9-carboxy-tetrahydrocannabinol
Solid Phase Extraction (SPE) is a method of sample preparation used to isolate
specific analytes from a solution. SPE consists of three main procedural components-
first, extraction of analytes from solution (typically larger volume) by sorption onto a
material (in this case an SPE cartridge), then washing the cartridge to purify the sample
and remove interfering compounds, and then release of analytes by elution with an
appropriate solvent (typically smaller volume) which concentrates the resulting
sample.17,18 SPE uses the same chromatographic principles as high-performance liquid
chromatography (HPLC), but in this case it is used to prepare a liquid sample before
analysis by another method (i.e. MS/MS).19 The main benefits of Solid Phase Extraction
include sample purification, simplification of matrix effects (the detection of non-analyte
sample components), fractionation, and enrichment of low-level compounds.19
As mentioned above, THC-COOH is found in wastewater in the ppt range, so pre-
concentration is needed to analyze the metabolite by MS/MS. For example, 50 mL of
sewage sample could be run through an SPE cartridge and then reconstituted in just 1 mL
of solvent, creating a 50-fold concentration step. This is an extremely powerful step for
sample preparation. In order to successfully retain and elute the analyte, three months
was spent testing solutions with varying pH and various brands of SPE cartridges. SPE is
a vital step in wastewater analysis because it is efficient and allows easy fractionation of
different compounds. It is also good for a wide range of volumes, which is beneficial for
this project because we will be using small sample sizes.17
1.2.2. Liquid Chromatography Tandem Mass Spectrometry
LC-MS/MS is the analytical instrument required to quantify THC-COOH in
prepared samples. Liquid chromatography (LC) involves the separation of components
Fryhle 6

in a liquid sample between a mobile phase and a stationary phase in a packed column.
Mass spectrometry can be coupled with LC by ionizing the sample components when
they elute from the chromatography column and then quantifying them based on
characteristic mass-to-charge ratios. LC-MS/MS is the best method for this procedure
because it combines the benefits of chemical isolation by LC and mass-to-charge
separation and quantification by MS/MS.20 In addition, MS/MS gives the most accurate
results because it induces fragmentation of selected precursor ions allowing subsequent
analysis of specific product ions.2,13,14 Our research group has access to an Agilent 6460
triple quadrupole mass spectrometer at the UW-Tacoma Center for Urban Waters.
1.3. RESEARCH METHODS
This research sought to quantify the amount of marijuana consumed and trends in
marijuana use in Tacoma during the 6 months leading up to the legal dispensaries (which
opened in July 2014) as well as the initial month or two after legalization. Professor
Burgard has been collecting and freezing samples from the Central Wastewater
Treatment Plant in Tacoma, WA since November 2013. In addition to sampling, the main
part of this project was significant method development for the extraction process. Since
research began in May 2014 and dispensaries didn’t open until June 2014, our goal was
to quantify THC-COOH concentrations in wastewater prior to the opening of retail stores
to estimate THC consumption. Our hypothesis was that if we found significantly greater
amounts of THC-COOH in the first few months after dispensaries opened, we could
assume that total consumption increased, instead of just a shift in source (from illegal to
legal). It would also be possible to know the degree of increased use by comparing our
findings to previous marijuana usage levels. In addition, the nature of this project
allowed our research group to look at trends in the amount of THC-COOH found in two
distinct regions of Tacoma. My project involved analysis of samples from the Central
Wastewater treatment plant (WWTP #1), whereas my lab partner Melissa Pellman
worked with the North WWTP (#2). WWTP #1 serves approximately 150,000 people,
where as WWTP #2 serves approximately 50,000 people. The boundary between these
two plants is shown in Figure 1. As this figure shows, there are different demographics
between the populations served by the two plants. The demographics of the populations
Fryhle 7

servicing the two treatment plants can be determined from U.S. census data because the
treatment boundaries fall nearly on census tract boundaries.16 These data could be useful
to other states or municipalities considering legalization.
Figure 1. Average Tacoma incomes for regions of the Central (#1) and North (#2)
Wastewater Treatment Plants (WWTP) (Darker colors mean higher incomes).
Wastewater testing is a fast, reliable method for drug epidemiology. By testing
influent samples from the Central WWTP, we have attempted to obtain valuable data that
could directly relate to the implementation of I-502. Though most of this research was
spent on method development, continuation of this project will allow for trends in THC-
COOH levels to be observed and marijuana use estimates to be estimated pre- and post-
legalization. Since there is no current scientific data in recreational marijuana use
changes after legalization, this project provides cutting edge research that could be
applied globally.16 In a time where marijuana legalization is sprouting all around the
world (it was even added to the ballot in Alaska for this fall), countries are struggling to
find a way to regulate and control marijuana usage. This method may provide insight to
many pressing issues surrounding I-502 including consumption patterns, demographics,
and economics.
2. MATERIALS AND METHODS
2.1. STANDARDS AND REAGENTS
(+/-)-11-nor-delta9-Carboxy-tetrahydrocannabinol (THC-COOH), THC-COOH-
D3 and THC-COOH-D9 were all purchased from Cerilliant (Round Rock, TX). All
Fryhle 8

purchased THC-COOH standards were purchased as 100 μg/mL standards in methanol.
Both of the deuterated standards were used as internal standards during method
development, and were sometimes used as surrogates to account for recovery. Analyte
structures and deuterated compounds can be seen in Scheme 1. Mobile phase solvents
were made from LC-MS grade Formic Acid obtained from Sigma Aldrich (St. Louis,
MO) and LC-MS-grade Methanol obtained from J.T. Baker Avantor Performance
Materials Inc. (Center Valley, PA). SPE wash solutions used ACS grade OMNISolve
Acetonitrile from EMD Chemicals (Gibbstown, NJ) and Milli-Q purified water (Milford,
MA). SPE eluent solutions used ethylacetate (Sigma Aldrich, St. Louis, MO) and
Isopropanol (EMD Chemicals, Gibbstown, NJ). Hydrochloric acid (EMD Chemicals,
Gibbstown, NJ) was also used for acidifying samples.
Scheme 1. Structures of analytes (non-deuterated) and internal standards (deuterated) and
their mass to charge (m/z) ratios of precursor ions into product ion fragments for the final
methods.
11-nor-9-carboxy-
tetrahydrocannabinol (THC-
COOH)
m/z: 343 --> 299
11-nor-9-carboxy-
tetrahydrocannabinol-D3 (THC-
COOH-D3)
m/z: 346 --> 302
11-nor-9-carboxy-
tetrahydrocannabinol-D9 (THC-
COOH-D9)
m/z: 352 --> 308
Fryhle 9

2.1.1 Standard Calibration Curves
Calibration standards of the analytes were prepared at 0.1, 0.2, 0.5, 1, 2.5, 5. 10,
25, 50, and 100 ng/mL. Internal standard calibrants were used at 10, 25, and 100 ng/ml
throughout the study but currently the method calls for 25 ng/ml of THC-COOH-D3.
2.2. GENERAL METHODS
2.2.1. Wastewater Samples
24-hour composite wastewater samples were obtained from August 2013 to the
present from the Central Wastewater Treatment Plant (WWTP #1) in Tacoma, WA by
volume-proportional sampling and 1 L aliquots were immediately frozen. The region
utilizing the Central WWTP is shown in Figure 1. Sample collection day was determined
through a stratified random approach where within each annual quarter each day was
sampled twice for 14 samples per quarter, 56 annually (beginning December 1st 2015).
Daily samples were also obtained during the months of January and July.
2.2.2. Lab Maintenance
All glassware was washed with soap, bleach, and MeOH and baked at 450oC prior
to use (Appendix A). Samples that had to be stored overnight prior to LC-MS/MS
analysis were stored at 4oC for analysis.
2.3. EXTRACTION AND QUANTIFICATION PROCEDURES
2.3.1. Testing of HLB, MCX, and MAX SPE Cartridges
Two beakers of deionized (DI) water (100 g) were acidified to pH 2 using
concentrated HCl, and one beaker of DI water (100 g) was brought to pH 6 using 0.5%
acetic acid and NaOH. Ten milliliters of each solution were placed in separate beakers,
and were each spiked at a final concentration of 20 ng/mL 11-nor-9-carboxy-
tetrahydrocannabinol (THC-COOH) internal standard and 20 ng/mL 11-nor-9-carboxy-
tetrahydrocannabinol-D9 (THC-COOH-D9) as a surrogate. Three different SPE cartridge
types were tested including Oasis HLB 3cc (60 mg) extraction cartridges, Oasis MXC
3cc (60 mg) extraction cartridges, and Oasis MAX 3cc (60 mg) extraction cartridges
(structures shown in Figure 2).
Fryhle 10

HLB and MCX. One HLB cartridge and one MCX cartridge were conditioned
with 1 mL of HPLC-grade MeOH and 2 mL of pH 2 DI water. The acidic spike solutions
were loaded in 2 mL aliquots onto the separate cartridges at 2mL/min. The cartridges
were then washed with 2 mL of pH 2 water, eluted with 2 mL of MeOH, and eluted with
2 mL of MeOH with 5% NH4OH. All fractions following each step (load, wash, and
elute) were collected for analysis. The eight resulting fractions were dried down under
Nitrogen (N2) and reconstituted in 200 μL 0.5% Acetic Acid containing 200 ng/mL THC-
COOH-D3 internal standard.
MAX. One MAX cartridge was conditioned with 1 mL HPLC-grade MeOH and 2
mL pH 6 DI water. The pH 6 spike solution was in a 2 mL aliquot onto the cartridge at
2mL/min, and the fraction was collected for analysis. The cartridge was then washed with
2 mL of pH 6 water, eluted with 2 mL of MeOH, and eluted with 2 mL of 5% Acetic
Acid. All fractions following each step were collected for analysis. The eight resulting
fractions were dried down under Nitrogen gas (N2) and reconstituted in 200 μL 0.5%
Acetic Acid containing 200 ng/mL THC-COOH-D3 internal standard.
Figure 222. MCX, HLB, and MAX cartridge structures.
2.3.2. Testing of Strata-XC SPE Cartridge
DI water (pH 2) was spiked at 1 μg/mL THC-COOH and 1 μg/mL THC-COOH-
D9 as a surrogate. The acidified spike solution (2 mL) was loaded onto a 200 mg/3mL
Fryhle 11

Phenomenex Strata-X-C 33u Polymer Strong Cation Exchange Cartridge at 5 mL/min
(structure shown in Figure 3). The cartridge was washed with 2 mL 15:85
acetonitrile:water, dried under vacuum for five minutes, and eluted with 2 mL 85:15
ehtylacetate:isopropanol. All fractions from each step were dried down under Nitrogen
gas (N2) and reconstituted in 2 mL of a 1:1 mobile phase mixture of mobile phase A and
mobile phase B, where mobile phase A is 1 mM ammonium formate with 0.1% formic
acid and mobile phase B is 0.1% formic acid in MeOH:acetonitrile (1:1). The
reconstitution solution had a final THC-COOH-D3 internal standard concentration of 0.2
μg/mL.
Figure 3. Strata-XC Cation Exchange Cartridge Structure.
2.3.3. SPE vs. Lyophilizing
SPE. A 2 mL aliquot of wastewater spiked at a final concentration of 1 ng/mL
THC-COOH and 1 ng/mL THC-COOH-D9 was loaded onto a 200 mg/3mL Phenomenex
Strata-X-C 33u Polymer Strong Cation Exchange Cartridge at 5 mL/min. The same was
done with 2 mL of pH 2 ultrapure milli-Q (MQ) water. Both cartridges were washed with
2 mL 15:85 acetonitrile:water, dried for five minutes, and eluted with 2 mL 85:15
ethylacetate:isopropanol. The eluents were dried down under Nitrogen gas (N2) and
reconstituted in 2 mL of a 1:1 mobile phase mixture of 1 mM ammonium formate with
0.1% formic acid and 0.1% formic acid in MeOH:acetonitrile (1:1), with a THC-COOH-
D3 concentration of 200 ng/mL.
Lyophilizing. 40 g of wastewater was spiked at 1 ng/mL THC-COOH-D9 in a 100
mL round bottom flask. The wastewater was cooled with liquid Nitrogen and placed on
the lyophilizer overnight. The residue remaining in the round bottom flask was
reconstituted in 1 mL of a 1:1 mobile phase mixture of 1 mM ammonium formate with
Fryhle 12

0.1% formic acid and 0.1% formic acid in MeOH:acetonitrile (1:1), with a THC-COOH-
D3 concentration of 1 ng/mL.
2.3.4. Filtration
Acidic Polymer filters. Four 50 g aliquots of pH 2 MQ water were spiked at 20
ng/mL THC-COOH and 20 ng/mL THC-COOH-D3. Each spiked aliquot was filtered
through one of four filter types: Filter A = 0.45 μm polyethersulfone, Filter B = 0.45 μm
polytetrafluoroethylene (PTFE) Teflon, Filter C = 0.2 μm regenerated cellulose (RC)
membrane, or Filter 0 = no filter. The filtered aliquots were analyzed by LC-MS/MS.
Basic/Acidic Polymer filters. Four 10 g aliquots of basic pH 9 MQ water were
spiked at 20 ng/mL THC-COOH and 20 ng/mL THC-COOH-D3. Each spiked aliquot
was filtered through either Filter A (as above), Filter B (as above), Filter C (as above), or
F0 (no filter). Each filtered aliquot was then acidified to pH 2 and added to autosampler
vials for LC-MS/MS analysis.
Regenerated Cellulose (RC) filters. 100 g of wastewater was GF-A filtered. A 10
g aliquot was weighed out and spiked at a final concentration of 2 ng/mL THC-COOH-
D3. The spiked aliquot was then adjusted to pH 9 using 6M NaOH and filtered by RC
membrane filtration. The RC filtered aliquot was then adjusted to pH 2 and transferred to
an autosampler vial for analysis. The same procedure was repeated for 100 g of MQ
water at 50 ng/mL THC-COOH, 100 g of MQ water at 0.2. ng/mL THC-COOH, and 100
g of MQ water at no THC-COOH concentration.
2.3.5. Testing of LC-MS/MS Optimal Ion Mode
A standard curve was made in MQ water with concentrations ranging from 1 to
1000 ng/mL THC-COOH and with internal standard and surrogate concentrations of 100
ng/mL THC-COOH-D3 and 100 ng/mL THC-COOH-D9, respectively. The curves were
run under optimal LC-MS/MS parameters (Section 2.3.8. Final Extraction Procedure,
LC-MS/MS).
Fryhle 13

2.3.6. Testing of Optimal Eluents
Triplicate 20 g GF-A filtered wastewater samples were spiked at 100 ng/mL
THC-COOH-D9 and 25 ng/mL 5-HIAA-D5. Each sample was loaded onto a
Phenomenex, Strata-X-C 33u Polymer Strong Cation Cartridge, 200 mg/3mL at
approximately 5 mL/min, washed with 2 mL Acetonitrile:Milli-Q Water (15:85), and
dried under vacuum for 5 minutes. This procedure was repeated four times up to this step.
The first three cartridges (group A) were eluted with 4 mL HPLC-grade MeOH. The
fractions were dried down under Nitrogen gas and reconstituted in 1 mL 0.1% Formic
Acid (plain mobile phase instead of 1:1 mixture of mobile phases, as in section 2.3.2).
The same three cartridges were then eluted with 2 mL Ethylacetate:Isopropanol (85:15).
These fractions were dried down under Nitrogen gas and reconstituted in 1 mL 5 mM
Ammonium Formate. The next three cartridges (group B) were eluted first with 2 mL
MeOH, and then eluted with 2 mL Ethylacetate:Isopropanol (85:15). The combined
fractions were dried down and reconstituted in 1 mL 5 mM Ammonium Formate. The
third set of triplicate cartridges (group C) were first eluted with 2 mL
ethylacetate:isopropanol (85:15). These fractions were dried down under Nitrogen gas
and reconstituted in 1 mL 5 mM Ammonium Formate. The Group C cartridgs were then
eluted again with 2 mL MeOH. These fractions were dried down under Nitrogen gas and
reconstituted in 1 mL 0.1% Formic Acid. The last set of triplicate cartridges (group D)
were only eluted with 2 mL Ethylacetate:Isopropanol (85:15). The fractions were dried
down under Nitrogen gas and reconstituted in 5 mM ammonium formate.
2.3.7. Testing of D3 vs. D9 Internal Standards
THC-COOH-D3 and THC-COOH-D9 were both tested in varying concentrations
as internal standards, both in wastewater and in MQ water samples.
2.3.8. Final Extraction Procedure
SPE. Wastewater samples were thawed at room temperature for use. Wastewater
(300 g) was weighed and spiked at a final concentration of 25 μg/mL 5-
hydroxyindoleacetic acid-D5 (5-HIAA-D5) and 25 μg/mL 11-nor-9-carboxy-
tetrahydrocannabinol-D3 (THC-COOH-D3) internal standards. The spiked 300 g was
Fryhle 14

divided into two 150 g sections. Section A was acidified to pH 2 and filtered by glass
microfiber (GF-A) filters. Section B was basified to pH 8 with NaOH and GF-A filtered.
Each 150 g section was divided into triplicate 50 g sections and loaded onto mixed-mode
ion exchange SPE cartridges (Phenomenex, Strata-X-C 33u Polymer Strong Cation
Cartridge, 200 mg/3mL) equipped with a 50 ml glass short-stem funnel with parafilm
wax seal at approximately 5 mL/min (Figure 4) (cartridges were first conditioned with 1
mL MeOH and 1 mL DI water). Cartridges were then washed with 2 mL
Acetonitrile:Milli-Q Water (15:85), dried under vacuum for 5 minutes, eluted with 2 mL
MeOH, and eluted again with 2 mL ethylacetate:isopropanol (85:15). Eluents were dried
under Nitrogen gas (N2) at 40º C and reconstituted in 1 mL 5 mM ammonium formate.
Figure 4. Solid Phase Extraction (SPE) manifold with 50 mL funnels for sample loading.
LC-MS/MS. 100 μL injections were performed with a 20 s MeOH needle wash on
an Agilent 1290 UHPLC with a 2.6 x 2.1 μm 100 mm Kinetex C18 column by
Phenomenex with a 6460 triple quadrupole mass spectrometer (LC-MS/MS) located at
the Center for Urban Waters, Tacoma, WA. The instrument had a gradient flow of 0.35
mL/min with mobile phase A as 5 mM ammonium formate and mobile phase B as
MeOH. Table 1 shows the mobile phase gradient program. The instrument was run in
negative ion mode ([M-H]-). The MS QQQ parameters were as follows: gas temperature
320 oC, gas flow at 11 L/min, nebulizer pressure at 18 psi, sheath gas temperature 320 oC,
sheath gas flow at 12 L/min, capillary voltage +2500 V and -4000, and charging voltage
+1000 V and -0 V.
Fryhle 15

Table 1. Solvent composition for LC-MS/MS final methods.
Time
(min)
Solvent Composition
1 1.00 A: 95% B: 5%
2 3.00 A: 5% B: 95%
3 5.00 A: 5% B: 95%
4 5.50 A: 95% B: 5%
5 6.50 A: 95% B: 5%
3. RESULTS AND DISCUSSION
3.1. HLB, MCX AND MAX SPE CARTRIDGE RESULTS
The HLB, MCX, and MAX cartridges were ineffective with the conditions that
we tried and results were inconclusive in all three cases. All qualifier to quantifier ratios
for all samples were outside of the allowed range. This can be seen in figure 5, an
example of an MCX cartridge. There are no sample peaks found in any of these fractions,
at the desired retention time (tR) of tR = 8.4 min. However, the caveat is that these
cartridges were all being evaluated before we had optimized the LC-MS/MS method.
Fryhle 16

Figure 5. LC-MS/MS chromatograms of DI water samples spiked at 20 ng/mL THC-
COOH and 20 ng/mL THC-COOH-D9, processed by SPE with an Oasis MCX cartridge.
The chromatograms, in descending order from top, are as follows: load, wash (pH2
water), first elute (MeOH), second elute (MeOH with 2% NH3). See section 2.3.1. HLB
and MCX for methods. tR = 8.4 min.
3.2. STRATA-XC INITIAL CARTRIDGE RESULTS
Due to inconclusive results from the HLB, MCX, and MAX SPE cartridges, a
Strata-XC cation exchange cartridge was tested. Results were also inconclusive for this
experiment. No analyte peaks were found by LC-MS/MS analysis, and the internal
standard peak was too small to be considered a significant peak. This is shown in Figure
6.
Fryhle 17

Figure 6. LC-MS/MS chromatograms of a wastewater sample from August 6, 2013,
spiked at 1 ng/mL THC-COOH and 1 ng/mL THC-COOH-D9. Analyzed by SPE using
Strata-XC 33u Polymeric 200 mg/3mL SPE cartridge. The top chromatogram is the
analyte, and the bottom chromatogram is the internal standard. See section 2.3.3. for
methods. tR = 7.5 min.
Due to inconclusive results from the HLB, MCX, MAX and Strata-XC extraction
cartridge methods, lyophilizing was tested as an alternative form of analyte isolation.
This method is described in section 3.3.
3.3. SPE VS. LYOPHILIZING RESULTS
A comparison was made between SPE and lyophilizing to see if lyophilizing
could better isolate the analyte. The lyophilizing results were inconclusive and there were
no resolved peaks after LC-MS/MS analysis. The compared SPE results from this test
were also inconclusive, and there were no significant analyte peaks. These results can be
seen in Figure 7.
Fryhle 18

Figure 7. LC-MS/MS analyte chromatograms of wastewater spiked at 1 ng/mL THC-
COOH-D9 internal standard. The top chromatogram is a lyophilized sample, and the
bottom chromatogram is a sample analyzed by Strata-XC 33u Polymeric 200 mg/3mL
SPE cartridge. See section 2.3.4. for methods. tR = 7.5 min.
As shown above, neither lyophilizing nor SPE provided adequate sample peaks at
the desired retention time. Since both SPE and lyophilizing did not produce results up to
this point, filtration was tested as an alternative method of analyte isolation.
3.4. FILTRATION RESULTS
Three forms of filtration were tested using MQ water samples (see section 2.3.5
for methods).
Acidic Polymer filters. The first form of filtration involved four different polymer
filters, used for spiked acidic MQ water samples. The filter types are as follows: F0 (no
filter), FA (0.45 μm polyethersulfone), FB (0.45 μm polytetrafluoroethylene (PTFE)
Teflon), and FC (0.2 μm regenerated cellulose (RC) membrane). The results for this
method are shown in Figure 8.
Fryhle 19

Figure 8. LC-MS/MS analyte chromatograms of acidic MQ water spiked at 20 ng/mL
THC-COOH and 20 ng/mL THC-COOH-D3. The chromatograms, in descending order
from top, are as follows: F0 (no filter), FA (0.45 μm polyethersulfone), FB (0.45 μm
polytetrafluoroethylene (PTFE) Teflon), and FC (0.2 μm regenerated cellulose (RC)
membrane). See section 2.3.5. Acidic Polymer filters for methods. tR = 7.3 min.
As seen in the above chromatograms, peaks were seen around the desired
retention time of 7.3 minutes for no filter, and for the 0.45 μm PTFE filter. The no filter
sample produced over an order of magnitude more signal than the filtered samples. The
0.45 μm PTFE filter was ruled out because the wastewater samples were too dificult to
push through it. After further deliberation, it was decided that the same procedure should
be repeated but with basic samples. Since THC-COOH has a pKa of approximately 4, we
hoped that by making the wastewater basic the analyte would stay negatively charged,
and therefore slip through the filters into the filtrate more easily. This would allow for a
greater quantity of analyte in our samples and greater signal overall.
Basic/Acidic Polymer filters. The second form of filtration involved the same
filters as with the first form, only the wastewater was basified prior to filtration to ensure
that the THC-COOH remained an ion, as described above. This produced much better
results, as shown in Figure 9.
Fryhle 20

Figure 9. LC-MS/MS analyte chromatograms of basic MQ water spiked at 20 ng/mL
THC-COOH and 20 ng/mL THC-COOH-D3. The chromatograms, in descending order
from top, are as follows: F0 (no filter), FA (0.45 μm polyethersulfone), FB (0.45 μm
polytetrafluoroethylene (PTFE) Teflon), and FC (0.2 μm regenerated cellulose (RC)
membrane). See section 2.3.5. Basic/Acidic Polymer filters for methods. tR = 7.3 min.
As shown in Figure 9, making the samples basic prior to filtration produced much better
results than the acidic samples in Figure 8. Based on the amount of signal shown by each
chromatogram, Filter C (0.2 μm RC membrane filter) was chosen for wastewater analysis
due to its larger signal.
Regenerated Cellulose (RC) filters. After choosing the RC membrane filter as the
optimal filter type, August, 2013 wastewater samples were tested using the method
described in section 2.3.5. Regenerated Cellulose (RC) filters. Following RC filtration
and LC-MS/MS analysis, THC-COOH peaks appeared in our samples. The results of this
method are shown below in Figure 10. Note that these results were obtained when we
were still using positive ion mode in the mass spectrometer.
Fryhle 21

Figure 10. LC-MS/MS chromatogram of an RC membrane filtered, basic wastewater
sample from August 9, 2013, spiked at 2 ng/mL THC-COOH-D3 internal standard. See
section 2.3.5. Regenerated Cellulose (RC) membrane filters for methods. tR = 7.1 min.
This method produced nicely resolved peaks. Many additional August, 2013
wastewater samples were analyzed by this method. However, after further analysis, it was
concluded that, while peaks were detected, they were not large enough in signal for the
purpose of this project.
3.5. TESTING OF LC-MS/MS OPTIMAL ION MODE RESULTS
After SPE, lyophilizing, and filtration proved to be unsuccessful methods, we
reverted back to SPE and decided to instead tweak the parameters of the LC-MS/MS
instrument. Up until this point, our methods closely followed those listed in the literature.
All other research teams have used LC-MS/MS in positive ion mode to analyze THC-
COOH. Due to lack of ideas for the next step, we tested our method in negative ion
mode. As a result, resolved peaks with acceptable ion counts were produced. As shown in
Figure 11, analyte peaks were shown with ion counts in the 106 ranges.
Fryhle 22

Figure 11. LC-MS/MS chromatogram of a MQ water sample in negative ion mode with
a retention time of 4.7 minutes. The top chromatogram represents the analyte and the
bottom chromatogram represents the THC-COOH-D9 internal standard. tR = 4.7 min
This was a pivotal point in this research and allowed for a revisit to solid phase
extraction. From this point on, the instrument was run in negative ion mode.
3.6. TESTING OF OPTIMAL ELUENT RESULTS
Following the switch from positive to negative ion mode, it was necessary to
further optimize our SPE method. From experimentation with Strata-XC cartridges, it
was concluded that THC-COOH was seen in the eluent fractions, after the cartridges
were eluted with 85:15 ethylacetate:isopropanol (EtOAc/IPA). It is important to note that,
while the Strata-XC cartridges are cations, better results were produced when THC-
COOH was also made a cation. This could be explained by the aromatic nature of the
cartridge. It is possible that the structural similarities (pi bonding and hydrophobic
regions) between the cartridge (Figure 3) and the analyte (Scheme 1) caused the analyte
to stick even though they are both negatively charged. After concluding that the analyte
came off in the EtOAc/IPA elute, different combinations of eluents were tested to ensure
maximum elution of THC-COOH. MeOH was added as an eluent due to its strong
attraction of polar organic compounds. In order to make sure our method was optimized,
we tested four different combinations of the two eluents listed above, which are referred
to as Group A, Group B, Group C, and Group D (methods described in section 2.3.6.).
Fryhle 23

Group A. This group involved separate elution steps- first with MeOH, and
second with 85:15 ethylacetate:isopropanol. The fractions were kept separate and were
analyzed separately by LC-MS/MS.
Group B. This group involved combined elution steps of both MeOH and 85:15
ethylacetate:isopropanol in the same test tube.
Group C. This group involved separate elution steps- first with 85:15
ethylacetate:isopropanol, and second with MeOH. The fractions were kept separate and
were analyzed separately by LC-MS/MS.
Group D. This group involved only one elution step with 85:15
ethylacetate:isopropanol.
The results from these trials are shown below in Figure 12.
Fryhle 24

Figure 12. LC-MS/MS chromatograms of Groups A-D (as described above). The first
two chromatograms represent Group A, the third represents Group B, the next two
represent Group C, and the last represents Group D. See section 2.3.6. for methods. tR =
4.4 min.
As shown in Figure 12, peaks were only seen for the fractions that included 85:15
ethylacetate:isopropanol. Because the fraction consisting of combined methanol and
GroupAGroupBGroupCGroupD
Fryhle 25

85:15 ethylacetate:isopropanol (Group B) produced the same amount of signal (104) as
the fractions with just 85:15 ethylacetate:isopropanol, it was decided that the method
used for Group B would be the final eluting method because of the need to
simultaneously elute the 5-HIAA analyte in the sample.
3.7. D3 VS. D9 INTERNAL STANDARD RESULTS
Both THC-COOH-D3 and THC-COOH-D9 internal standards were tested at
varying concentrations. Initially, THC-COOH-D9 was chosen over THC-COOH-D3 as
the best IS. At this time, we were using an IS concentration of 100 ng/mL and were
concerned that D3 might have a greater contribution to the THC-COOH signal (which we
were trying to quantify at 0.1 ng/mL) than the D9 standard. The D9 standard reported
0.00% D0 by LC-MS analysis whereas the D3 standard reported 0.02% D0.25, 26 However,
after further consideration, we switched back to D3 as the best IS because D9 had too
much interference, as seen in Figure 13.
Figure 13. LC-MS/MS chromatograms of THC-COOH-D3 and THC-COOH-D9 internal
standards. Quantitative (quant) and qualitative (qual) chromatograms are shown for both
internal standards.
THC-COOH
analyte (45
ng/ml)
THC-COOH-
D3 IS (quant)
THC-COOH-
D9 IS (quant)
THC-COOH-
D3 IS (qual)
THC-COOH-
D9 IS (qual)
Fryhle 26

3.8. FINAL METHODS PRELIMINARY RESULTS
The final method is described in section 2.3.8. This method was applied to four
days of wastewater analysis of December, 2013 wastewater samples. An example of one
day of wastewater analysis can be seen below in Figure 14.
Figure 14. LC-MS/MS chromatograms of a December 12, 2013 wastewater sample
spiked at 25 ng/mL THC-COOH-D3 internal standard. The chromatograms on the left are
quantifiers and the chromatograms on the right are qualifiers. The top two are analyte
chromatograms and the bottom two are internal standard chromatograms. tR = 4.4 min.
This method produces nicely resolved peaks with ion counts up to 104, as seen in
Figure 14. THC-COOH-D3 was chosen as the optimal internal standard for this method.
The calibration curve used for these samples can be seen in Appendix B.
4. CONCLUSIONS
After extensive method development, a final method was decided upon (See
section 2.3.8.). Strata-XC was found to be the optimum SPE cartridge, with THC-COOH-
D3 as an internal standard. Negative ion mode was found to be the best mode for the LC-
MS/MS instrument. In addition, it was found that wastewater must be made basic for
solid phase extraction, and that THC-COOH is best extracted and retained by the column
under basic conditions.
This method produced good results for four days of wastewater samples from
December, 2013. These data can be seen in Table 2, below.
IS
Analyte
Fryhle 27

Table 2. Average calculated THC-COOH concentration and %RSD, Influent THC-
COOH, THC-COOH Total mass, and Total Marijuana Consumed for four days of
December, 2013 influent wastewater samples from the Central Wastewater Treatment
Plant in Tacoma, WA.
Date
Avg. Calculated
THC-COOH
(ng/mL)
%RSD THC-
COOH
Influent
THC-COOH
(ng/mL)
THC-COOH
Total Mass
(g/day)
Marijuana
Consumeda
(kg/day)
12-Dec-13 73.9 3.6 1.48 85.2 118
16-Dec-13 71.2 4.63 1.42 89.8 125
17-Dec-13 52.5 8.79 1.05 68.3 95
22-Dec-13 81.5 4.18 1.63 103.1 143
aAssuming THC excreted at 0.6%23 as THC-COOH and 12% potency
Using the influent wastewater flow rate, the percent THC-COOH excreted, and an
estimated potency percentage, total marijuana consumed in kg/day was calculated.
According to a report done by RAND corporation in 2013, Pierce County consumed 16
metric tons of marijuana in the year 201324. In kg/day, this amounts to 43.84 kg/day.
According to our study, we are finding values in the upper nineties and hundreds of
kg/day in the Central Tacoma area (Table 2). This is extremely significant because it
shows that survey data could be significantly underestimating use. While our data is not
necessarily accurate due to uncertainties in THC-COOH excretion and marijuana potency
in Washington (as well as uncertainty in other variables), it is greater than twice what
survey data estimate, showing that more research is necessary to provide answers to the
question regarding post-legal marijuana consumption.
Our method proved to be successful for these four days of December, 2013
wastewater samples. However, it has not been successful for all days of wastewater
samples. Therefore, more work is needed to further optimize the method so that it may be
applied across the board to all wastewater samples.
Fryhle 28

5. Acknowledgements
Thanks to Dan Burgard for all of his help and guidance with this project, and
thanks to Melissa Pellman for working with me in the lab. Thank you to the staff at the
Center for Urban Waters for allowing us to use their LC-MS/MS instrument. In addition,
thanks to the University of Puget Sound Chemistry Department for funding this project.
Fryhle 29

6. References
1. Berset, J.D.; Brenneisen R.; Matheiu, C. Analysis of llicit and illicit drugs in waste,
surface, and lake water samples using large volume direct injection high
performance liquid chromatography-Electrospray tandem mass spectrometry
(HPLC-MS/MS). Chemosphere. 2010, 81, 859-866.
2. Lubertus, B.; Sancho, J.V.; Pitarch, E.; Ibanez, M.; and Hernandez, F. Simultaneous
ultra-high-pressure liquid chromatography-tandem mass spectrometry
determination of amphetamine and amphetamine-like stimulants, cocaine and its
metabolites, and a cannabis metabolite in surface water and urban wastewater. J.
Chromatog. A. 2009, 3078-3089.
3. Ettore, Z.; Chiabrando, C.; Castiglioni, S.; Bagnati, R.; Fanelli, R. Estimating
Community Drug Abuse by Wastewater Analysis. Environ. Health. Perspect.
2008, 116,1027-1032.
4. Drug Facts: Marijuana, Nation Institute on Drug Abuse. <http://www.drugabuse.
gov/publications/drugfacts/marijuana>. (accessed February 4, 2014).
5. Patrick Radden, K. Buzzkill: Washington discovers that it's not so easy to create a
legal marijuana economy. New Yorker. 2013.
6. Christina, V. Marijuana History in United States Sees Milestone, Pot First Banned in
California 100 Years Ago. Huffington Post. 2013.
7. Beukenhorst, D.J. Comparative Cannabis Use Data. Br. J. Psychiatry. 2007, 179, 175-
177.
8. Maia, S. Drugs in Portugal: Did Decriminalization Work? Time: Science and Space.
2009.
9. Glenn, G. Drug Decriminalization in Portugal: Lessons for Creating Fair and
Successful Drug Policies. Cato Institute. 2009.
Fryhle 30

10. Uruguay’s President Nominated for Nobel Peace Prize for Legalizing Marijuana.
<http://rt.com/news/uruguay-nobel-mujica-marijuana-849/>. (accessed February
23, 2014).
11. Ingold, J. Coloardo marijuana legalization's impact on stoned driving unknown.
http://www.denverpost.com/marijuana/ci_25101187/colorado-marijuana-
legalizations-impact-stoned-driving-unknown%3E.?source=infinite. (accessed
February 2, 2014).
12. Ettore, A.; Chiabrando, C.; Castiglioni, S.; Calamari, D.; Bagnati, R.; Schiarea, S.;
Fanelli, R. Cocaine in surface waters: a new evidence-based tool to monitor
community drug abuse. Environ. Health J.: A Global Access Science Source.
2005, 4.
13. Sara, C.; Zuccato, E.; Crisci, E.; Chiabrando, C.; Fanelli, R.; Bagnati. R.
Identification and Measurement of Illicit Drugs and Their Metabolites in Urban
Wastewater by Liquid Chromatography-Tandem Mass Spectrometry. J. Anal.
Chem. 2006, 78.24 8421-8429.
14. Postigo, C.; Lopez de Alda M.J.; Barcelo. D. Fully Automated Determination in the
Low Nanogram per Liter Level of Different Classes of Drugs of Abuse in Sewage
Water by On-Line Solid-Phase Extraction-Liquid Chromatography-Electrospray-
Tandem Mass Spectrometry. Anal. Chem. 2008, 80, 3123-3134.
15. Postigo, C.; Lopez de Alda, M.J.; Barcelo. D. Drugs of abuse and their metabolites in
the Ebro River Basin: Occurrence in sewage and surface water, sewage treatment
plants removal efficiency, and collective drug usage estimation. Environ. Int.
2010, 36, 75-84.
16. Burgard, D.A.; Banta-Green, C.; Field, J.A. Working Upstream: How far can you go
with sewage-based drug epidemiology? Environ. Sci. Technol. 2014. http://dx.doi.
org/10.1021/es4044648.
17. Johnson, R. "Solid Phase Extraction 101." Webinar. C&EN Webinars. 04 02 2014.
Lecture.
18. Strata-X-C General Method for the Extraction of Weak Bases (TN-0026).
<https://phenomenex.blob.core.windows. net/documents/eef91af2-37d7-4b54-
a188-4946ec9be169.pdf>. (accessed February 26, 2014).
Fryhle 31

19. Beginners Guide to SPE (Solid Phase Extraction). Waters: The Science of What's
Possible. <http://www.waters.com/waters/en_US/Beginner's-Guide-to-SPE-
[Solid-Phase-Extraction] /nav.htm?locale=en_US&cid=13472147 6>. (accessed
February 23, 2014).
20. " Liquid Chromatography-Tandem Mass Spectrometry. <http://www.perkinelmer.
com/Catalog/Category/ID/Liquid ChromatographyTandem Mass Spectrometry>.
(accessed February 23, 2014).
21. United States. goveME: government Made Easy. http://www.govme.org/govME
/Admin/Inter/StartPage/default.aspx (accessed 2014)
22. Waters Corporation: The Science of What’s Possible. http://www.waters.com/
waters/home.htm (accessed May 25, 2015).
23. Khan, U.; Nicell, J. A. Sewer epidemiology mass balances for assessing the illicit use
of methamphetamine, amphetamine and tetrahydrocannabinol. Sci. Total Environ.
2012, 421–422, 144–162.
24. Rand Corporation. http://www.rand.org/content/dam/rand/pubs/research_reports/
RR400/RR466/RAND_RR466.pdf (accessed April 1, 2015)
25. Cerilliant Analytical Reference Standards. (+/-)-11-nor-9-carboxy-Δ9-THC-D3,
Primary Standard. https://www.cerilliant.com/shoponline/COA.aspx?itemno
=3021ccb3-bd2b-481e-bb9b-9b0a03292be4&lotno=FE11141402. (accessed May
2015).
26. Cerilliant Analytical Reference Standards. (+/-)-11-nor-9-carboxy-Δ9-THC-D9,
Primary Standard. https://www.cerilliant.com/shoponline/COA.aspx?itemno
=782fd462-7c21-403a-b15f-07b7cb00e9a2&lotno=FE01131505. (accessed May
2015).
Fryhle 32

7. APPENDICES
7. 1. APPENDIX A
Glassware was baked using a Barnstead Thermolyne industrial benchtop furnace
(Dubuque, Iowa).
Instrument: Model No: 56028C-80
Serial No: 1060010655006
Volts: 208
Amps: 19.2
Watts: 4000
Hz: 50/60
Instructions for use of this furnace are as follows.
1. Power: ON
2. Hold down green “SEL” button until it reads:
PV: “LoCE” (parameter lock level)
SV: 0002
For the rest of the instructions, make sure the instrument reads whatever is
instructed for the “PV” and “SV” settings. The instructions will describe
what each of the readings means.
3. Click “SEL” button to switch to the next parameter setting.
PV: Nod
SV: Auto
PV: At (autotuning feature)
SV: oFF (Normal operation mode, automatically sets optimum values.)
PV: Stat (status)
SV: 1-rP (Cannot be changed.)
PV: Tine = “time” (time remaining in current program)
SV: 12.58
PV: ProG = “program” (program options)
SV: run
PV: Sü-1 = “Set Point 1”
SV: 450 (Temperature at which 1st ramp is aiming and at which 1st dwell
will end.)
PV: tN1r = “Ramp Time 1”
Fryhle 33

SV: 1.00 (Time in hours that the controller will take to charge T from its
present T to value of Set Point 2.)
PV: tN1S = “Dwell Time 2”
SV: 4.00 (Time in hours that the controller will hold the furnace at Set
Point 1 Temperature.)
SV: 10
SV: 8.00
SV: 0.00
SV: 0.00
SV: 0.00
SV: 0.00
SV: 0.00
PV: tN4r “Ramp Time 4”
SV: 0.00
SV: 0.00
PV: P-on
SV: YES (Furnace will immediately begin running the set program.)
PV: P (proportional band)
SV: 1.7
PV: i (integral)
SV: 449
PV: d (derivative)
SV: 69.1
Fryhle 34

PV: AL1T (Alarm type)
SV: 0031 (Should be set to this. If not, change it to this.)
PV: AL11 (Furnace over Temperature alarm)
SV: 1200 (Determines absolute upper limit on furnace chamber
temperature.)
PV: AL12 (High deviation alarm)
SV: 20
PV: LooP (thermocouple break alarm detection time)
SV: 0.00
PV: CTrL (control type)
SV: FUZY
29. Let the program run, and turn the machine off when finished.
Fryhle 35

7.2. APPENDIX B
Calibration curve of analyte, prepared at 0.1, 0.2, 0.5, 1, 2.5, 5. 10, 25, 50, and 100 ng/m.
Internal standard concentration of 25 ng/mL THC-COOH-D3. R2 = 0.9945.
Fryhle 36

HSF- Thesis

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to HSF- Thesis

Similar to HSF- Thesis (20)

HSF- Thesis