Newly developed smartwatch measures key stress hormone


The human body responds to stress, from the everyday to the extreme, by producing a hormone called cortisol.

To date, it has been impractical to measure cortisol as a way to potentially identify conditions such as depression and post-traumatic stress, in which levels of the hormone are elevated. Cortisol levels traditionally have been evaluated through blood samples by professional labs, and while those measurements can be useful for diagnosing certain diseases, they fail to capture changes in cortisol levels over time.

Now, a UCLA research team has developed a device that could be a major step forward: A smartwatch that assesses cortisol levels found in sweat—accurately, noninvasively and in real time. Described in a study published in Science Advances, the technology could offer wearers the ability to read and react to an essential biochemical indicator of stress.

“I anticipate that the ability to monitor variations in cortisol closely across time will be very instructive for people with psychiatric disorders,” said co-corresponding author Anne Andrews, a UCLA professor of psychiatry and biobehavioral sciences, member of the California NanoSystems Institute at UCLA and member of the Semel Institute for Neuroscience and Human Behavior. “They may be able to see something coming or monitor changes in their own personal patterns.”

Cortisol is well-suited for measurement through sweat, according to co-corresponding author Sam Emaminejad, an associate professor of electrical and computer engineering at the UCLA Samueli School of Engineering, and a member of CNSI.

“We determined that by tracking cortisol in sweat, we would be able to monitor such changes in a wearable format, as we have shown before for other small molecules such as metabolites and pharmaceuticals,” he said. “Because of its small molecular size, cortisol diffuses in sweat with concentration levels that closely reflect its circulating levels.”

The technology capitalizes on previous advances in wearable bioelectronics and biosensing transistors made by Emaminejad, Andrews and their research teams.

In the new smartwatch, a strip of specialized thin adhesive film collects tiny volumes of sweat, measurable in millionths of a liter. An attached sensor detects cortisol using engineered strands of DNA, called aptamers, which are designed so that a cortisol molecule will fit into each aptamer like a key fits a lock. When cortisol attaches, the aptamer changes shape in a way that alters electric fields at the surface of a transistor.

The invention—along with a 2021 study that demonstrated the ability to measure key chemicals in the brain using probes—is the culmination of a long scientific quest for Andrews. Over more than 20 years, she has spearheaded efforts to monitor molecules such as serotonin, a chemical messenger in the brain tied to mood regulation, in living things, despite transistors’ vulnerability to wet, salty biological environments.

Sweating the small stuff: Smartwatch developed at UCLA measures key stress hormone
The technology capitalizes on previous work by Sam Emaminejad, Anne Andrews and their UCLA research teams. Credit: Emaminejad Lab and Andrews Lab/UCLA

In 1999, she proposed using nucleic acids—rather than proteins, the standard mechanism—to recognize specific molecules.

“That strategy led us to crack a fundamental physics problem: how to make transistors work for electronic measurements in biological fluids,” said Andrews, who is also a professor of chemistry and biochemistry.

Meanwhile, Emaminejad has had a vision of ubiquitous personal health monitoring. His lab is pioneering wearable devices with biosensors that track the levels of certain molecules that are related to specific health measures.

“We’re entering the era of point-of-person monitoring, where instead of going to a doctor to get checked out, the doctor is basically always with us,” he said. “The data are collected, analyzed and provided right on the body, giving us real-time feedback to improve our health and well-being.”

Emaminejad’s lab had previously demonstrated that a disposable version of the specialized adhesive film enables smartwatches to analyze chemicals from sweat, as well as a technology that prompts small amounts of sweat even when the wearer is still. Earlier studies showed that sensors developed by Emaminejad’s group could be useful for diagnosing diseases such as cystic fibrosis and for personalizing drug dosages.

FIG. 1. Noninvasive cortisol biomarker monitoring using a wearable aptamer-field-effect transistor sensing system.
(A) The hypothalamus-pituitary-adrenal (HPA) axis controls cortisol levels in response to circadian rhythm and stress. ACTH, adrenocorticotropic hormone; CRH, corticotropin-releasing hormone. (B) The fraction of circulating cortisol not bound to blood plasma proteins is available for excretion by salivary and sweat glands. (C) Saliva and sweat samples can be analyzed by an aptamer-field-effect transistor (FET) sensing system. Top: Photograph of an aptamer-FET-enabled biosensing smartwatch. Bottom: Schematic illustration of cortisol sensing by an aptamer-FET sensor. VG, gate voltage; VS, source voltage; VD, drain voltage; ADC, analog-digital converter. (D) Photograph of a FET sensor array with In2O3 semiconductor channels fabricated on a flexible polyimide substrate. Schematic layers not to scale. (E) Expanded view of the key components of an aptamer-FET biosensing smartwatch. Liquid crystal display (LCD). (F) Overview of FET-array signal acquisition via a multichannel on-board source measurement unit (SMU). Data processing is via a microcontroller unit (MCU), display, and transmission. IDS, source-drain current; VGS, gate voltage. Photo credit: Zhaoqing Wang, Yichao Zhao, UCLA.

One challenge in using cortisol levels to diagnose depression and other disorders is that levels of the hormone can vary widely from person to person—so doctors can’t learn very much from any single measurement. But the authors foresee that tracking individual cortisol levels over time using the smartwatch may alert wearers, and their physicians, to changes that could be clinically significant for diagnosis or monitoring the effects of treatment.

Among the study’s other authors is Janet Tomiyama, a UCLA associate professor of psychology, who has collaborated with Emaminejad’s lab over the years to test his wearable devices in clinical settings.

“This work turned into an important paper by drawing together disparate parts of UCLA,” said Paul Weiss, a UCLA distinguished professor of chemistry and biochemistry and of materials science and engineering, a member of CNSI, and a co-author of the paper. “It comes from us being close in proximity, not having ego problems and being excited about working together. We can solve each other’s problems and take this technology in new directions.”

The paper’s co-first authors are UCLA postdoctoral scholar Bo Wang and Chuanzhen Zhao, a former UCLA graduate student. Other co-authors are Zhaoqing Wang, Xuanbing Cheng, Wenfei Liu, Wenzhuo Yu, Shuyu Lin, Yichao Zhao, Kevin Cheung and Haisong Lin, all of UCLA; and Milan Stojanović and Kyung-Ae Yang of Columbia University.

Among the many hormones in circulation throughout the body, cortisol (C21H30O5) is one of the most influential hormones affecting the physiological processes that alter the human body’s homeostasis. Cortisol is classified as a steroid hormone that is synthesized from cholesterol in the zona fasciculata of the kidneys’ adrenal complexes, and it is key to the body’s fight-or-flight state when a stressor occurs [1].

Hence, cortisol has a reputation of being the stress hormone [2]. Under ideal homeostasis, cortisol levels will fluctuate in a day-long cycle, peaking in the morning, and the hormone is released from an unexpected change experienced [3]. As such, cortisol can reach concentrations in the body that are too large or small, resulting in unforeseen effects that may indicate that the glucocorticoid feedback inhibition cycle is impaired [4].

In other studies, cortisol has also been associated with several common stress-based diseases and other disorders. A review conducted by Kiesner and Granger attempted to see if cortisol dysfunction correlated with the onset of premenstrual syndrome and premenstrual dysphoric disorder (PMS/PMDD), but further study was warranted to obtain more conclusive findings [5]. Wei et al. found that cortisol levels in hair samples increased in patients with first-episodic depression, which indicates that cortisol may be a biomarker for depression [6]. Furthermore, a study conducted by Ettman et al. found that depression rates tripled since the onset of the COVID-19 pandemic, making it much more relevant now [7].

Yang et al. found that individuals with Autism Spectrum Disorder (ASD) had higher cortisol and serotonin levels while having lower oxytocin [8]. With increased awareness and criteria changes for ASD diagnoses, cortisol has become more relevant in analyzing ASD. In addition, an influx of evidence suggests that cortisol may be a contributing factor to coronary heart disease (CHD) if it is present in large volumes, especially if maternal cortisol is present during pregnancy [9]. Relatively minor symptoms can also originate from higher and lower cortisol levels affecting (and undermining) the immune system, including insomnia, fatigue, and headaches [10].

Thus, it is necessary to accurately measure cortisol levels at any time of the day while reducing the difficulty in finding cortisol levels.
It is vital to have an accurate measurement at the point of care (POC) to ensure the tested patient’s proper diagnosis and prognosis. Current methods of cortisol testing, such as blood, urine, and saliva tests, are used to detect at least the 10% of cortisol present freely in the blood with methods including immunoassays and liquid chromatography-tandem mass spectrometry (LC-MS/MS) [2,11].

However, it can be a relatively time-consuming process, and some methods must be conducted away from home at an expense to the patient. Furthermore, these tests must be completed at specific times because cortisol levels are ideally higher in the morning and lower in the evening [3]. Apart from collecting a blood sample at a particular time of day, medical professionals may need the patient to collect all their urine in a 24-h period.

Some saliva samples may need to be collected at multiple periods throughout the day. These tests are prone to inaccuracies due to the design of the interface between cortisol and quantitative scales for analysis, such as Cohen’s perceived stress scale (PSS) [12]. The result is an overall cost of time and convenience to the patient and healthcare system. However, E. Russel et al. found that cortisol levels in bodily sweat are comparable to those in saliva samples, which indicates that sweat and hair samples could accurately reflect the concentration of cortisol in the body [13].

In recent years, additional research in bio-interfaces has increased for biosensor applications for several different material types. Among these materials, graphene has been a material of interest to researchers and government entities for over four decades [14]. Significant resources and funding have been invested, where the British government alone has invested over 20 million GBP in developing graphene products [15]. The fabled material possesses unique mechanical properties, such as a considerably high mechanical strength of 1 TPa, and it has already seen research and applications for solar cells and nanoelectronic devices [16].

Other mechanical properties associated with graphene include its thermo-conductivity and charge carrier-mobility, measured to be 3000 W/mK and 10,000 cm2/V∙s, respectively [16]. One specific variation of graphene comes in the form of graphene oxide (GO), a graphene-based material with oxygen functional groups covalently attached. GO has garnered international interest since the mid-2000s due to the material’s mechanical, electrical, and thermal applications, and it, along with reduced graphene oxide (rGO), has been tested for electrochemical sensors [17].

GO is now a commonplace material for self-assembling monolayers (SAMs), and its properties allow for easy functionalization with other chemicals and biomolecules [18]. This is further justified with GO’s high surface area to volume ratio—specifically a surface area of approximately 2630 m2/g—and its ability to function well in aqueous environments [19].

The larger surface area makes it possible for more biomolecules to be functionalized more efficiently. GO is also ideal for medical devices if the sample used on the said device is a liquid (i.e., sweat). The use of GO has been associated with increased specificity in what electrodes and other biological molecules are detected in a sensor, and that application has been exploited in several past studies focusing on cortisol detection and analysis.

Several vital proteins and carbohydrates found in the body can be used as a base to analyze a patient’s physiology. These sensors can detect minute variations in concentrations of various hormones, proteins, and chemicals in real-time with high sensitivity and specificity to their respective applications. An electrode can be designed by coating it in GO and adding specific antibodies to the surface, providing the specificity required by a biosensor [16,20].

Glucose sensors have been designed for research using graphene as done by S. Chaiyo et al., where they developed a paper-based biosensor to detect glucose levels in serum samples [21]. S. Cinti et al. developed a biosensor that detects chloride ions (Cl−) using screen-printed filter paper with hydrophilic and phobic sites coated in a sulfuric acid solution with Cl− ions [22].

A lactate biosensor from K. Lin et al. used GO nanosheets coated in a dimethyl-sulfoxide (DMSO) and 1-pyrenebutyric acid–N-hydroxysuccinimide ester (PANHS) to detect lactate concentrations in sweat samples [23]. Concerning cortisol, one application explored by S. Tuteja et al. included a Bluetooth-based means to obtain data from an electro-reduced graphene oxide (e-RGO) sensor using an anti-cortisol antibody CORT-2 and a lactate antibody to detect and isolate cortisol in sweat samples [20].

In addition, more cortisol biosensors were designed by M. Sekar et al. using a conductive carbon fiber material to detect cortisol levels in sweat, and their findings indicated that the sensor was sensitive and specific enough to be properly used specifically for near-complete cortisol detection [24]. A GO biosensor using π-stacked rabbit anti-cortisol antibodies and denatured bovine serum albumin (d-BSA) was created by K. Kim et al. to detect cortisol in saliva samples by utilizing the antibodies’ specificity to detect the hormone [10].

The previously mentioned GO-biosensors maintained their capacities to detect cortisol in serum, sweat, and saliva samples to inform users of potential physiological changes and abnormalities affecting them. To provide a convenient, non-invasive, and swift cortisol test for diagnostic and personalized care applications, research in a sweat-based cortisol biosensor was conducted.

With the varying cortisol concentrations throughout the body in mind, this research intended to determine whether a GO interface’s electrochemical responses were affected by those concentrations. These differences in the electrochemical responses based on cortisol concentration may be a determining factor in developing a user-friendly cortisol biosensor for applications at the POC.

reference link :

More information: Bo Wang et al, Wearable aptamer-field-effect transistor sensing system for noninvasive cortisol monitoring, Science Advances (2022). DOI: 10.1126/sciadv.abk0967


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.