Targeting And Metabolites In Human Eccrine Sweat As Biomarkers For Mental Health

Item Type text; Electronic Thesis

Authors Kubacki, Marisa Taylor

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 01/10/2021 18:08:16

Item License http://rightsstatements.org/vocab/InC/1.0/

Link to Item http://hdl.handle.net/10150/632840 TARGETING CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT AS

BIOMARKERS FOR MENTAL HEALTH

By

MARISA TAYLOR KUBACKI

______

A Thesis Submitted to the Honors College

In Partial Fulfillment of the Bachelor’s Degree With Honors in

Psychology

THE UNIVERSITY OF ARIZONA

MAY 2019

Approved by:

______

Dr. Esther M. Sternberg Director of Research, UA Center for Integrative Medicine CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 1

Abstract Studies have found that the use of human eccrine sweat for tracking human biomarkers, such as cortisol and downstream cortisol metabolites, was a viable, non-invasive option that yielded similar results to plasma cortisol (Jia et al., 2016, p. 2053). These biomarkers were found to be stable in human eccrine sweat and that their production was due, in part, to stress-related physiological events (Runyon, et al., 2019, p. 1). Also, patients with a current anxiety disorder displayed significantly higher awakening cortisol levels (p = 0.02) than those without a diagnosed anxiety disorder (Vreeburg, et al., 2010, p. 340). These findings indicate that cortisol levels found in human eccrine sweat can be used as a potential quantitative indicator of a mental health disorder with cortisol and metabolite concentrations in sweat being used to aid in the research and advancement of the understanding of underlying mechanisms. This is a preliminary study that will further the research currently being done on sweat collection systems and research the connection between cortisol and anxiety by identifying any correlations in selected target biomarkers found in eccrine sweat (such as cortisol, , and cortisol metabolites) and differing anxiety levels as indicated through the State-Trait Anxiety Index (STAI).

CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 2

Targeting Cortisol and Metabolites in Human Eccrine Sweat as Biomarkers for Mental Health

Scientists and psychologists have long been fascinated about how the brain works, how mental disorders develop, and how these illnesses can have differential effects on individuals.

New technology and medical advancements have made it possible to more thoroughly explore what it means to be diagnosed with a mental disorder. Currently, a person can be diagnosed with an anxiety disorder if they experience excessive anxiety in certain aspects of their life for an average of six months (“Anxiety Disorders,” 2018). This excessive worry must cause disturbances in a person’s personal, work, or school life in order to be considered an anxiety disorder and should be subsequently diagnosed and treated. Symptoms can include feeling restless, fatigue, difficulty concentrating, irritability, difficulty controlling anxious thoughts, or having difficulty sleeping (“Anxiety Disorders,” 2018). Anxiety disorders, such as Generalized

Anxiety Disorder (GAD), have a very high prevalence rate with a lifetime prevalence rate of almost 30% and despite the high prevalence, anxiety disorders remain under- or mis-diagnosed, which can lead to inappropriate treatment for these patients (Kasper, 2006, p. 3). Under- diagnosis of anxiety can occur due to high comorbidity rates, in which 75% of those with anxiety also meet the criteria for another psychiatric illness (Zanni, 2012). In addition, mis-diagnosis can be due to the fact that certain illnesses may be more “visible” than certain anxiety disorders, such as physical illness or other psychiatric illnesses like panic disorder (Kasper, 2006, p. 3).

Salivary daytime cortisol levels correlate with symptom severity of those with anxiety, and that those with anxiety had a higher overall level of plasma cortisol as compared to healthy participants (Elnazer & Baldwin, 2014, p. 195). Participants with a current anxiety disorder displayed significantly higher awakening cortisol levels (p = 0.02) than those without a diagnosed anxiety disorder (Vreeburg, et al., 2010, p. 340). This correlation of cortisol and CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 3 anxiety can be used as a quantitative indicator of the presence of anxiety with cortisol levels being used as a quantitative measure to aid in the research and advancement of the understanding of underlying mechanisms involved with certain anxiety disorders, as outlined by the criteria for anxiety in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).

The DSM-5 is currently used to diagnose anxiety disorders, but other anxiety measures have been widely used in research to measure a participant’s risk for anxiety. Some anxiety measures include the State-Trait Anxiety Index (STAI), Beck Anxiety Inventory (BAI), and

Hospital Anxiety and Depression Scale-Anxiety (HADS-A), but these measures are limited in construct validity due to high correlations being found between anxiety and depression measures, which may coincide with the overlapping symptoms of depression and anxiety (Julian, 2011).

For this study, the STAI will be used, which is a 40-item measure that assesses the severity of current anxiety symptoms with personality traits associated with anxiety, which has been used to compare healthy and psychiatric populations. There 20 items for state anxiety that measures how anxious a person has been feeling within the past seven days, which helps to account for environmental factors in a person’s life that may be causing a significant amount of stress. Trait anxiety makes up the other 20 items, which accounts for certain personality traits that are associated with high levels of anxiety. Both subscales provide a specific insight into an individual’s anxiety levels, with the state subscale used to detect a longitudinal change in a person, while the trait subscale is used to identify anxiety as a fixed characteristic (Julian, 2011).

This measure will be used to understand how a person’s cortisol levels correlate to their current stress state, as well as identify any general trends for those who are considered an “anxious person” in regard to their concentration of target biomarkers. CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 4

Cortisol, commonly referred to as the “stress hormone,” is a hormone that is produced in the adrenal glands and is then released into the blood. In response to stress, the hypothalamic-pituitary-adrenal (HPA) axis releases an excess amount of cortisol into the blood

(“Cortisol,” 2017). When discussing anxiety disorders, it is more important to look at changes of cortisol throughout the day, as opposed to only looking at cortisol levels at a single point in time, due to changes in cortisol quantities being indicative of increased stress levels (Vedhara, et al.,

2003, p. 96). Different diseases are associated with constant heightened or diminished cortisol levels, such as diseases associated with the HPA-axis, as opposed to a disruption in cortisol regulation throughout the day – a factor that needs to be considered before determining if cortisol levels can be used to help indicate the presence of anxiety. For instance, Cushing’s syndrome can develop from too much cortisol production, while adrenal insufficiency (AI) is due to a constant low level of cortisol in the body. Even though there is a significant correlation between those with Cushing’s syndrome and personality traits associated with high anxiety (Dimopoulou, et al.,

2013, p.144), for simplicity, the focus will remain on anxiety with no history of comorbidity in diseases that affect the HPA-axis.

To track changes in cortisol levels, a method must be put in place to provide accurate data on cortisol levels throughout the day. A previous study found that serum and salivary cortisol levels show a significant correlation of 0.728, even if salivary responses could not track serum cortisol as closely at higher concentrations with peak serum levels occurring immediately after exercise and peak salivary levels occurring 30 minutes into the recovery period post- exercise (VanBruggen, et al., 2011, p. 401). However, it was discovered that the use of human eccrine sweat for measuring the concentration of human biomarkers, such as cortisol and its downstream metabolites, was a viable non-invasive option that yielded similar results to plasma CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 5 cortisol (Jia, et al., 2016, p. 2053). These biomarkers were found to be stable in human eccrine sweat and that their production was due to stress-related physiological events (Runyon, et al.,

2019, p. 1). A method used to quantify target biomarkers in human sweat is liquid chromatography-tandem mass spectrometry (LC-MS), which has been shown to be both selective and sensitive specifically to cortisol and its downstream metabolites (Runyon, et al.,

2019, p. 8). This method works by separating cortisol, its metabolites, and cortisone from other compounds found in sweat through liquid chromatography. Then the mass spectrometer is used to differentiate cortisol, its metabolites, and cortisone from each other based on their fragmentation patterns to generate concentration levels of each targeted biomarker. LC-MS methodology has been shown to be highly effective and has been used in a study that found that the conversion of cortisol to its downstream metabolites occurred in the body, either the skin or sweat glands, as opposed to in the sweat itself or during the sampling process (Runyon, et al.,

2019, p. 7). This discovery indicates that the conversion from cortisol to its metabolites is enzyme-mediated by the body, which strengthens the case for using cortisol in sweat as an accurate and non-invasive collection strategy.

For rapid sample collection, new technology is being developed to create wearable sweat biosensors that can monitor various target biomarkers found in sweat. These biosensors will show the concentration levels of these target biomarkers, electrolytes, and skin temperature to give an accurate overview of a person’s general health in a non-invasive way (Gao, et al., 2016, p. 509). This technology can also be used to target certain biomarkers that are associated with anxiety. Developing a cortisol pathway monitor that tracks a person’s cortisol, cortisone, and cortisol metabolite concentration through sweat from a biosensor will provide a non-invasive way to quantify a person’s stress levels in real-time. However, this type of technology is CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 6 currently in the developmental stage, with electrochemical biosensors being designed for various body fluids that include not only sweat but also tears, saliva, and interstitial fluid of the skin

(Heikenfeld, et al., 2019). While many biosensors are being developed for general health that can be used to monitor a range of physiological signals, other biosensors are being engineered with high selectivity for a specific analyte, or biomarker. These high selectivity biosensors target a specific analyte, such as cortisol, by using a receptor, which is a biological component that selects for cortisol, that is attached to a transducer and converts the signal from the interaction of receptor-analyte into a quantitative value (Bandodkar & Wang, 2014, p. 363). Another factor in developing a biosensor is the question of rapid, repetitive and accurate detection of cortisol. A study on the development of a wearable cortisol biosensor determined that the biosensor prototype was able to distinguish miniscule changes in cortisol concentration with the same accuracy of lab equipment when the two were plotted against each other with an R2 value of

0.999 (Kinnamon, et al., 2017). While current development shows biosensors that give accurate readings of biomarkers, there is still a question of battery life in these devices, though now some biosensors are being made with the use of biofuel cells that do not require a battery (Bandodkar et al., 2019). Still, a limitation of the use of biosensors for anxiety is that cortisol is not the sole factor when determining if a disorder is present. Many factors are in play including the environment, hormone levels, personality (state versus trait anxiety), as well as genetics.

However, there is evidence indicating that biosensors used to monitor and provide feedback to the user will significantly impact disease prevention, diagnosis, and treatment (Ajami &

Teimouri, 2015, p. 1208). Even if cortisol pathway monitors cannot definitively diagnosis someone with anxiety, it can still be used to quantify cortisol levels and used as a potential diagnostic indicator of the presence of anxiety, as well as help determine if treatment is CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 7 necessary (Rashkova, et al., 2010, p. 67). By monitoring cortisol levels in eccrine sweat, a wearable biosensor can be used to identify if treatment is being used effectively or if the treatment plan needs to be altered.

The overarching goal of this research is to be able to identify any correlations in selected target biomarkers found in eccrine sweat (such as cortisol, cortisone, and cortisol metabolites) and differing anxiety levels as indicated through the STAI. If there is a positive general trend in the data, then a wearable cortisol pathway biosensor may be useful in providing a non-invasive way to quantify stress levels in real-time. These stress levels can be used as a possible indicator of an anxiety disorder, as well as provide insight into what treatment may be necessary. Use of a cortisol pathway monitor for anxiety should be used in conjunction with approved diagnostic criteria from the DSM-5 in order to help lower the cases of under- and mis-diagnosed anxiety disorders. This is a preliminary study that will further the research currently being done on developing cortisol pathway biosensors as well as to further explore the connection between cortisol and anxiety.

Methods

Participants

Participants will be selected on a volunteer basis from the University of Arizona population with the minimum age requirement being 18 years old. As for strategies to recruit participants, it was found that approximately 68% of US adults (aged 18-24) are currently on

Facebook, with about three-quarters of these users frequenting the website daily (Smith &

Anderson, 2018). Since this study is targeting college students on the University of Arizona campus, Facebook is a viable option to recruit participants by posting on a University of Arizona group page. Interested participants will be asked to fill out a survey on Qualtrics to determine CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 8 eligibility. Exclusions will include those currently taking anxiolytic or anxiogenic medication, , cortisol pathway creams and medications, or thyroid medications to reduce any confounds from the results.

Study Design and Procedure

• IRB approval was gained to perform experiment IRB Approval

• Qualtrics platform • Participants must complete it before their lab session STAI

• Room heated to 89OF with 40% relative humidity • Recumbent bike was ridden at 70% of age-determined CATS Lab maximum heart rate for one hour Beta Testing • ECG, blood pressure, and respiratory rate are recorded

• LC-MS used to select and differentiate cortisol and metabolites Data • Positive polarity used for all biomarkers Analysis • APCI used for ionization

Figure 1. Procedure flowchart of what has been accomplished in this study.

This is an observational study that will use the data collected from the STAI as well as the data found through the analytical methodology of the LC-MS to quantify the target biomarkers to find general trends and correlations between a participant’s reported levels of state and trait anxiety and their concentration of target biomarkers in their eccrine sweat. This study will first exclude anyone currently taking anxiolytic (anti-anxiety) or anxiogenic (Ritalin, antipsychotics or dopamine antagonists, caffeine, or GABA antagonists) medication, steroids CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 9

(including asthma inhalers), , cortisol pathway creams and medications, or thyroid medications. Also, participants will be asked to refrain from ingesting any form of caffeine the day of their scheduled session. To further reduce any possible confounds, the session times will all be scheduled around the same time of day as cortisol concentration levels decline naturally throughout the day (Vedhara, et al., 2003, p. 96). Once participants are deemed eligible, they will be asked to complete the 40-item STAI online, through Qualtrics, at least 48 hours before their scheduled session time. Participants will then be asked to come into the lab to ride a recumbent bike at 70% of age-determined maximum heart rate for approximately one hour in a room heated to 89oF with 40% relative humidity. Sweat will be collected periodically, with about 10 samples being collected per participant. Sweat collection will be done by attaching a small tygon tube to a suction device that feeds directly into an Eppendorf low protein bind polypropylene vial. Once the participant’s sweat samples are collected, they will be run through the analytical methodology to be used for data analysis. Using the LC-MS can help isolate cortisol from its isomers for accurate analysis of cortisol levels (Jia, et al., 2016). Liquid chromatography helps separate cortisol, its metabolites, and cortisone from other compounds found in sweat. Then the mass spectrometer is used to differentiate cortisol, its metabolites, and cortisone from each other based on their fragmentation patterns to generate concentration levels of each biomarker.

Measures and Physiological Testing

This study will take place in the Clinical and Translational Science Research Center

(CATS) at Banner UMC for the physical exercise portion, while the STAI will be administered online through the Qualtrics platform. While riding the recumbent exercise bike in CATS, the participants’ physiological responses will be monitored. For physiological monitoring, BIOPAC technology will be used to collect data from an ECG (electrocardiogram) to measure electrical CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 10 activity of the heart, respiratory bands on their chest and abdomen to measure pulmonary ventilation, and an upper arm blood pressure cuff will be used to measure systolic and diastolic blood pressure. This data will be compiled into each participant’s file upon completion of the session.

Data Analysis

For STAI analysis, the Qualtrics account will be used to generate a numerical value of both state and trait anxiety for each participant. This information will be placed in an excel spreadsheet, along with each participant’s average biomarker concentration as collected through the LC-MS. To make sure that cortisol levels are associated with varying levels of anxiety, the

STAI will be used to place participants on a continuum of low scores (indicating little to no anxiety) to high scores (indicating high anxiety) to give a clear image of where cut-offs should be placed for those with low, intermediate, and high anxiety levels in relation to their eccrine sweat cortisol levels. This measure, in conjunction with the LC-MS data, will be used to show how a person’s cortisol levels are affected by their current stress state (their state anxiety score), as well as if there are any general trends for those who are considered an “anxious person” (their trait anxiety score) in regard to their average concentration of target biomarkers.

The LC system achieved separation of cortisol and cortisol metabolites in standards and sweat samples with a binary gradient elution. Mobile phase A consisted of 0.1% formic acid in

LCMS water and mobile phase B consisted of 0.1% of formic acid in LCMS acetonitrile. Total run time was 15 minutes with mobile phase B holding at 0% for the first minute, then ramping up from 0%-20% between 1-2 minutes, then ramping up again from 20%-25% from 2-10 minutes, then increasing from 25%-100% between 10-11 minutes, then decreasing from 100%-

0% from 12-12.5 minutes, then holding at 0% from 12.5-15 minutes until the run was completed. CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 11

All cortisol and metabolites targeted had a positive polarity and the ionization was achieved using atmospheric pressure chemical ionization (APCI). Identification of each target compound was achieved using LC retention times and selected reaction monitoring (SRM) transition

(Runyon, 2019, p. 4). Table 1 shows the target compounds SRM transitions as well as their respective collision energies as found using the MS system.

Compound SN R2 Precursor Product Collision Ratio3 Values (m/z) (m/z) Energy (V)

Cortisol 605 0.997 363.2 121.1 23 Cortisone 701 0.9918 361.2 163.1 22.438

120α-DHCN 477 0.9865 363.2 163.1 22 120β-DHCN 621 0.9947 363.2 163.1 22

220α-DHCL 243 0.9955 365 269.2 15.5 2 20β-DHCL 1284 0.9966 365 269.2 15.5 Table 1. Cortisol and Cortisol Metabolite SRM Transitions and Collision Energy. 120α/β-dihydrocortisone 220α/β-dihydrocortisol 3 this is an example of the signal-to-noise (SN) ratio from sweat sample 3, as seen in the chromatogram in figure 4

Results

This study has been beta-tested with one participant and the preliminary results indicated that the LC-MS methodology developed can be used to specifically identify different biomarkers found in human eccrine sweat. Also, all peaks analyzed using the chromatograms in Figures 1-3 were found to be significant as the signal-to-noise (SN) ratios were all above 10 (Table 1). The chromatograms developed from the LC-MS methodology in Figures 1-3 was used to develop concentrations for each metabolite based on the peak area found from each peak. Linear least- squared regression was used to fit the curve of the calibration as seen in Figures 5-10 with excellent linearity (R2 of 0.98 for 20α-DHCN and 0.99 or greater for the rest of the metabolites). CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 12

cortisol

cortisone L ine 1

20α/β-DHCN

20α/β-DHCL

Figure 2. Chromatogram of sweat sample 1. Retention times: cortisol 9.80 minutes, cortisone 11.07 minutes, 20α/β DHCN 8.48 and 8.95 respectively, and 20α/β-DHCL 7.44 minutes and 7.93 minutes respectively.

cortisol cortisone

cortisone cortisol

20α/β-DHCN

20α/β-DHCN 20α/β-DHCL

20α/β-DHCL

Figure 3. Chromatogram of sweat sample 2. Retention times: cortisol 9.66 minutes, cortisone 10.93 minutes, 20α/β- DHCN 8.35 and 8.82 minutes respectively, and 20α/β-DHCL 7.286 and 7.78 minutes respectively.

CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 13

cortisol

cortisone

20α/β-DHCN L ine 1

20α/β-DHCL

Figure 4. Chromatogram of sweat sample 3. Retention times: cortisol 9.64 minutes, cortisone 10.92 minutes, 20α/β- DHCN 8.31 and 8.79 respectively, and 20α/β-DHCL 7.26 and 7.75 minutes respectively.

Cortisol Calibration Curve Cortisone Calibration Curve 140000 140000 y = 2393.9x - 2227.9 y = 2568.3x - 3353.5 R² = 0.997 R² = 0.9918 120000 120000

100000 100000

80000 80000 Calibration calibration 60000 60000

Peak Area Sample Peak Area sample 40000 40000 Validation Validation 20000 20000 sample sample 0 0 0 5 10 15 20 25 30 35 40 45 50 55 0 5 10 15 20 25 30 35 40 45 50 55 Concentration (ng/mL) Concentration (ng/mL) Figure 5. Cortisol calibration curve. Concentrations: sweat Figure 6. Cortisone calibration curve. Concentrations: sweat sample 1 had 27.90 ng/mL, sweat sample 2 had 39.47 ng/mL, sample 1 had 26.93 ng/mL, sweat sample 2 had 23.62 ng/mL, and sweat sample 3 had 9.15 ng/mL. and sweat sample 3 had 7.55 ng/mL.

CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 14

20α-dihydrocortisone Calibration Curve 20β-dihydrocortisone Calibration Curve 450000 700000 y = 8492.6x - 13682 y = 13369x - 16227 400000 R² = 0.9865 600000 R² = 0.9947 350000 500000 300000 250000 400000 Calibration Calibration

200000 300000 Peak Area

Peak Area Sample 150000 Sample 200000 100000 Validation Validation 100000 50000 sample sample 0 0 0 5 10 15 20 25 30 35 40 45 50 55 0 5 10 15 20 25 30 35 40 45 50 55 Concentration (ng/mL) Concentration (ng/mL) Figure 7. 20α-DHCN calibration curve. Concentrations: sweat Figure 8. 20β-DHCN calibration curve. Concentrations: sweat sample 1 had 7.59 ng/mL, sweat sample 2 had 6.12 ng/mL, sample 1 had 6.03 ng/mL, sweat sample 2 had 4.89 ng/mL, and and sweat sample 3 had 2.86 ng/mL. sweat sample 3 had 2.22 ng/mL.

20α-dihydrocortisol Calibration Curve 20β-dihydrocortisol Calibration Curve 700000 800000 y = 12099x - 12519 y = 14945x - 14860 600000 R² = 0.9955 700000 R² = 0.9966

500000 600000 500000 400000 Calibration Calibration 400000

300000 Peak Area Peak Area 300000 Sample Sample 200000 200000 Validation 100000 Validation 100000 sample sample 0 0 0 5 10 15 20 25 30 35 40 45 50 55 0 5 10 15 20 25 30 35 40 45 50 55 Concentration (ng/mL) Concentration (ng/mL) Figure 9. 20α-DHCL calibration curve. Concentrations: sweat Figure 10. 20β-DHCL calibration curve. Concentrations: sweat sample 1 had 2.28 ng/mL, sweat sample 2 had 1.93 ng/mL, and sample 1 had 6.50 ng/mL, sweat sample 2 had 4.97 ng/mL, and sweat sample 3 had 1.27 ng/mL. sweat sample 3 had 2.15 ng/mL.

CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 15

Compound Sweat Sample 1 Sweat Sample 2 Sweat Sample 3 Concentrations Concentrations Concentrations (ng/mL) (ng/mL) (ng/mL) Cortisol 27.8996 39.4678 9.1507 Cortisone 26.9289 23.6193 7.5523 120α-DHCN 7.5944 6.1157 2.862 120β-DHCN 6.0264 4.8921 2.2223 220α-DHCL 2.2765 1.9266 1.2751 220β-DHCL 6.5007 4.970 2.1517

Table 2. Concentrations of all target metabolites found in the sweat samples. 120α/β-dihydrocortisone 220α/β-dihydrocortisol

Compound Validation Validation Validation Sample 1 Sample 2 Sample 3 Concentrations Concentrations Concentrations (ng/mL) (ng/mL) (ng/mL) Cortisol 0.39 12.5 4.9112 Cortisone 0.39 12.5 4.3708 120α-DHCN 0.39 12.5 3.6907 120β-DHCN 0.39 12.5 4.1504 220α-DHCL 0.39 12.5 3.896 220β-DHCL 0.39 12.5 4.1873 Table 3. Concentrations of all target metabolites found in the validation samples. 120α/β-dihydrocortisone 220α/β-dihydrocortisol

Equations were made based off of calibration curves (Figures 5-10) for each target metabolite across a range of concentrations. Once a calibration curve was created, a validation sample was run for each metabolite to test the reliability of the equation with two known mixtures of the metabolites, one mixture had a concentration of 0.39 ppb and the other mixture had a concentration of 12.5 ppb. These two mixtures with differing concentrations were then mixed together in equal parts and run through the LC-MS, and all of the metabolites in the unknown mixture were found to have a concentration between 3.96 ppb and 4.91 ppb, which was the expected concentration range. This shows that the equations used for each metabolite can be used reliably for sweat samples as the unknown mixtures all showed concentration levels that were between the known mixtures. For the three sweat samples that were collected in the beta- test, all of the target metabolites, except cortisol, showed a decreasing pattern over time. The CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 16 highest concentrations were found in the first sample that was collected. However, as seen in

Figure 5, cortisol showed an increase from the first to second sweat sample (27.89 ppb to 39.47 ppb), and then jumped down to 9.15 ppb in the third sample. At least 10 sweat samples should be collected when running the study as the three samples that were collected may not show the complete pattern of cortisol release over time. According to the cortisol stress and recovery pattern, cortisol increases over time when stress occurs, then decreases back to baseline levels after a rest period (Elnazer & Baldwin, 2014, p. 196). While there were only three data points, it appeared that the participant’s cortisol levels followed this pattern as cortisol response increased between the first two samples, then decreased in the third sample when the participant was slowing down.

Discussion

This preliminary study has been approved by the Institutional Review Board, beta-tested, and is ready to start recruiting participants and collecting data. This study has shown that LC-MS can specifically identify and differentiate target biomarkers found in human eccrine sweat through the accuracy of the methodology used. Also, change in biomarker concentrations was seen across time, which can be used to aid future research when discussing the usefulness of tracking cortisol levels throughout the day. If there is a positive general trend in the data, meaning that there is a significant correlation between stress levels and an increase in cortisol when comparing target biomarker concentrations and the results of the STAI, then the development of a wearable cortisol pathway biosensor may be useful in providing a valid, viable, non-invasive way to quantify stress levels in real-time. These stress levels can then be used as a possible quantitative indicator of an anxiety disorder, as well as provide insight into what treatment may be necessary for a particular person. Use of a wearable cortisol pathway biosensor CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 17 for anxiety should be used in conjunction with approved diagnostic criteria from the DSM-5 and is not a replacement for diagnosis or treatment options. However, a wearable sweat collection system that tracks cortisol and metabolite levels may be useful in order to help lower the cases of under- and mis-diagnosed anxiety disorders by having quantifiable data that a professional can use to corroborate a diagnosis. While this study is focusing specifically on stress levels and the use of tracking target biomarkers in sweat to be used for anxiety disorders, a wearable sweat collection system may also be useful for mental health in general when discussing various mental health disorders. As sweat collection devices are currently being developed, it will prove useful to expand the literature and provide as much data as possible in order to further the research on how far the reach for these wearable sweat devices can have on society not only for general health, but also how they can fit into the mental health sector.

CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 18

References Ajami, S., & Teimouri, F. (2015). Features and Application of Wearable Biosensors in Medical

Care. Journal of Research in Medical Sciences, 20(12), 1208-1215. doi:10.4103/1735-

1995.172991

Anxiety Disorders. (2018, July). Retrieved February 22, 2019, from

https://www.nimh.nih.gov/health/topics/anxiety-disorders/index.shtml#part_145333.

Bandodkar, A. J., Gutruf, P., Choi, J., Lee, K., Sekine, Y., Reeder, J. T., . . . Rogers, J. A. (2019).

Battery-Free, Skin-Interfaced Microfluidic/Electronic Systems for Simultaneous

Electrochemical, Colorimetric, and Volumetric Analysis of Sweat. Science

Advances, 5(1). doi:10.1126/sciadv.aav3294

Bandodkar, A. J., & Wang, J. (2014). Non-Invasive Wearable Electrochemical Sensors: A

Review. Trends in Biotechnology, 32(7), 363-371.

doi:https://doi.org/10.1016/j.tibtech.2014.04.005

Cortisol. (2017, January). Retrieved February 22, 2019, from

http://www.yourhormones.info/hormones/cortisol/

Dimopoulou, C., Ising, M., Pfister, H., Schopohl, J., Stalla, G. K., & Sievers, C. (2013).

Increased Prevalence of Anxiety-Associated Personality Traits in Patients with Cushing's

Disease: A Cross-Sectional Study. Neuro-Endocrinology, 97, 139-145.

doi:10.1159/000338408

Elnazer, H. Y., & Baldwin, D. S. (2014). Investigation of Cortisol Levels in Patients with

Anxiety Disorders: A Structured Review. Behavioral Neurobiology of Stress-Related

Disorders, 18, 191-216. doi:10.1007/7854_2014_299

Gao, W., Emaminejad, S., Nyein, H. Y., Challa, S., Chen, K., Peck, A., . . . Javey, A. (2016). CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 19

Fully Integrated Wearable Sensor Arrays for Multiplexed in situ Perspiration

Analysis. Nature International Journal of Science, 529, 509-514.

doi:https://doi.org/10.1038/nature16521

Heikenfeld, J. et al. Accessing analytes in biofluids for peripheral biochemical monitoring.

Nature Biotechnology (2019). doi:10.1038/s41587-019-0040-3

Jia, M., Chew, W. M., Feinstein, Y., Skeath, P., & Sternberg, E. M. (2016). Quantification of

Cortisol in Human Eccrine Sweat by Liquid Chromatography - Tandem Mass

Spectrometry. Analyst,141(6), 2053-2060. doi:10.1039/c5an02387d

Julian, L. J. (2011). Measures of Anxiety: State‐Trait Anxiety Inventory (STAI), Beck Anxiety

Inventory (BAI), and Hospital Anxiety and Depression Scale‐Anxiety (HADS‐

A). Arthritis Care and Research, 63(11). doi:10.1002/acr.20561

Kasper, S. (2006). Anxiety Disorders: Under-Diagnosed and Insufficiently Treated. International

Journal of Psychiatry in Clinical Practice, 10(1), 3-9. doi:10.1080/13651500600552297

Kinnamon, D., Ghanta, R., Lin, K., Muthukumar, S., & Prasad, S. (2017). Portable Biosensor for

Monitoring Cortisol in Low-Volume Perspired Human Sweat. Scientific Reports, 7.

Retrieved March 8, 2019, from https://www.nature.com/articles/s41598-017-13684-7.

Rashkova, M., Kalchev, P., Emilova, R., Ribagin, L., Doganova, T., & Stoeva, I. (2010). Cortisol

in Saliva -- A Marker for Increased Anxiety in Children. Journal of IMAB, 16(4), 67-69.

Retrieved February 22, 2019, from https://www.journal-imab-bg.org/statii-

10/vol16_b4_p67-69.pdf.

Runyon, J. R., Jia, M., Goldstein, M. R., Skeath, P., Abrell, L., Chorover, J., & Sternberg, E. M. CORTISOL AND METABOLITES IN HUMAN ECCRINE SWEAT 20

(2019). Dynamic Behavior of Cortisol and Cortisol Metabolites in Human Eccrine

Sweat. International Journal of Prognostics and Health Management, 10, 1-11.

Retrieved March 8, 2019, from http://www.phmsociety.org/node/2560

Smith, A., & Anderson, M. (2018). Social Media Use in 2018. Pew Research Center.

Retrieved December 2, 2018, from http://www.pewinternet.org/2018/03/01/social-media-

use-in-2018/

VanBruggen, M. D., Hackney, A. C., McMurray, R. G., & Ondrak, K. S. (2011). The

Relationship Between Serum and Salivary Cortisol Levels in response to Different

Intensities of Exercise. International Journal of Sports Physiology and Performance, 6,

396-407. doi:10.1123/ijspp.6.3.396

Vedhara, K., Miles, J., Bennett, P., Plummer, S., Tallon, D., Brooks, E., . . . Farndon, J. (2003).

An Investigation into the Relationship Between Salivary Cortisol, Stress, Anxiety and

Depression. Biological Psychology, 62(2), 89-96. doi:10.1016/S0301-0511(02)00128-X

Vreeburg, S. A., Zitman, F. G., Van Pelt, J., DeRijk, R. H., Verhagen, J. C., Van Dyck, R., . . .

Penninx, B. W. (2010). Salivary Cortisol Levels in Persons With and Without Different

Anxiety Disorders. Psychosomatic Medicine, 72(4), 340-347.

doi:10.1097/PSY.0b013e3181d2f0c8

Zanni, G. R. (2012, August 16). Anxiety Disorders: Underdiagnosed and

Undertreated. Pharmacy Times. Retrieved March 8, 2019, from

https://www.pharmacytimes.com/publications/issue/2012/august2012/anxiety-disorders-

underdiagnosed-and-undertreated