Infrared Thermography for the Detection of Changes in Brown Adipose Tissue Activity

Summary

Here, we present a protocol for measuring brown adipose tissue activity after a meal in humans and laboratory animals.

Abstract

Measuring brown adipose tissue (BAT) activity by positron emission tomography computed tomography (PET-CT) via the accumulation of ¹⁸F-fluorodeoxyglucose (FDG) after a meal or in obese or diabetic patients fails as the method of choice. The main reason is that ¹⁸F-FDG competes with the postprandial high glucose plasma concentration for the same glucose transporter on the membrane of BAT cells. In addition, BAT uses fatty acids as a source of energy as well, which is not visible with PET-CT and could be changed along with glucose concentration in obese and diabetic patients. Therefore, to estimate the physiological importance of BAT in animals and humans, a new infrared thermography method used in recent publications is applied.

After overnight fasting, BAT activity was measured by infrared thermography before and after a meal in human volunteers and female wild-type mice. The camera software calculates the object's temperature using distance from the object, skin emissivity, reflected room temperature, air temperature, and relative humidity. In mice, the shaved area above the BAT was a region of interest for which average and maximal temperatures were measured. The phase of the estrous cycle in female mice was determined after an experiment by vaginal smears stained with cresyl violet (0.1%) stain solution. In healthy volunteers, two skin areas of the neck were selected: the supraclavicular area (above the collarbone, where BAT cells are present) and the interclavicular area (between the collarbones, where there is no BAT tissue detected). BAT activity is determined by the subtraction of those two values. Also, the average and maximal temperatures of skin areas could be determined in animals and human participants.

Changes in BAT activity after a meal measured by infrared thermography, a non-invasive and more sensitive method, were shown to be sex, age, and phase of the estrous cycle dependent in laboratory animals. As part of diet-induced thermogenesis, BAT activation in humans was also proven to be sex, age, and body mass index dependent. Further determining the pathophysiological changes in BAT activity after a meal will be of great importance for participants with high glucose plasma concentrations (obesity and diabetes mellitus type 2), as well as in different laboratory animals (knock-out mice). This method is also a variable tool for determining possible activating drugs that could rejuvenate BAT activity.

Introduction

Brown adipose tissue (BAT), in contrast to white adipose tissue (WAT), does not store but rather spends energy. Upon sympathetic stimulation, BAT utilizes fatty acids and glucose and produces heat by the activation of uncoupling protein 1 (UCP1). The function of UCP1 is to use an H⁺ gradient between two mitochondrial membranes to produce heat instead of ATP. The function of BAT is to increase heat production under cold conditions, which leads to an increase in energy expenditure¹. After cold exposure, sensory inputs from the skin inhibit warm-sensitive neurons in the median preoptic (MnPO) nucleus of the hypothalamic preoptic area (POA), which diminishes the inhibitory effect of POA neurons on the rostral raphe pallidus (rRPa). The activation of rRPa neurons increases sympathetic activity, which is followed by an increase in BAT activity^2,3. Cold-induced BAT activation improves insulin sensitivity in humans⁴, and this activity is decreased in humans with increased body mass index (BMI) and age^1,5,6,7.

Apart from its role in cold-induced thermogenesis, after a meal, glucose uptake in the BAT increases in the lean male population, contributing to diet-induced thermogenesis (DIT), which is higher in BAT-positive male subjects^8,9. The state-of-the-art technique used for measuring BAT activity is positron emission tomography computed tomography, known as PET-CT. This method determines BAT activity by measuring the accumulation of the radiotracer fluorodeoxyglucose (¹⁸F-FDG). However, PET-CT fails as the method of choice for detecting the activation of BAT after a meal. One of the reasons is that, after a meal, ¹⁸F-FDG competes with postprandial hyperglycemia for the same glucose transporter, which makes it unsuitable for determining BAT activation after a meal, especially when comparing BAT activity in healthy and diabetic participants with possible differences in blood glucose concentrations. Furthermore, BAT uses fatty acids as a source of energy for heat production which is not visible with PET-CT. ¹⁸F-FDG accumulation in BAT after a meal is barely visible¹⁰ and is, therefore, interpreted as a negative result in most cases. Unsurprisingly, recently, it was suggested that the activation of BAT is more pronounced in the human population than we had previously thought; therefore, a new approach to detect BAT activity and its involvement in metabolic disorders is necessary⁷. An attempt to solve this problem is to measure the volume of BAT with magnetic resonance imaging (MRI) in prediabetic patients and patients with diabetes mellitus type 2 (T2DM) with insulin resistance¹¹. However, BAT volume measured by MRI is not a sufficient indicator for estimating the everyday function and usage of glucose and fatty acids by BAT. Therefore, to estimate real differences in BAT activity in healthy versus T2DM patients, a new approach is needed that offers a possibility to find out the pathological mechanism of BAT malfunction in T2DM patients.

To determine the activation of BAT, we performed measurements of BAT heat production before and after a meal using infrared (IR) thermography (Figure 1)^12,13. Establishing IR thermography as a method of choice for measuring BAT activity after a meal in healthy and obese individuals or patients with diabetes mellitus will have a huge impact on the field. To this day, IR thermography is used for the determination of cold-induced activation of BAT^13,14,15. In recent human history, cold-induced BAT activity is not very pronounced anymore (due to proper heating of habitats, proper clothing), while BAT activation after a meal occurs every day. Furthermore, the physiological regulation of these two BAT functions via the hypothalamus is completely different. After a meal, the activation of proopiomelanocortin (POMC)-expressing neurons in the hypothalamic arcuate nucleus (Arc) leads to an increase in sympathetic nerve activity via rRPa¹⁶. Cold-induced activation of BAT measured by IR thermography or PET-CT is improper when used as a measure for everyday BAT activity. Increased BAT activity after a meal is followed by glucose utilization, which is ultimately important for maintaining glucose homeostasis, insulin sensitivity, and the daily regulation of glucose concentration. Postprandial BAT activation leads to an increase in postprandial glucose consumption, followed by an increase in heat production and body temperature (DIT). This was shown to be gender, age, and BMI dependent¹². Similar gender differences in BAT activation after a meal are observed in male and female laboratory mice¹⁷. These findings correspond to recently discovered gender differences in the regulation of BAT by Burke et al., who showed that the hypothalamic regulation of BAT browning via a subpopulation of POMC neurons differs in male and female mice¹⁸. The postprandial activation of BAT is smaller in women, older populations, and obese people. The lack of BAT activation after a meal (decreased glucose utilization) could lead to a higher prevalence of impaired glucose tolerance in women^19,20,21,22. Unfortunately, the majority of studies on BAT activation were done only on men. By activating BAT after a meal, glucose uptake increases in the lean male population. It is not surprising that, after BAT activation, DIT is higher in BAT-positive male subjects^8,9. Furthermore, BAT transplantation in male mice improves glucose tolerance, increases insulin sensitivity, and decreases body weight and fat mass²³.

PET-CT fails as a method of choice for measuring BAT activity, especially after a meal. Therefore, a non-invasive and more sensitive method was developed. IR thermography enables the estimation of BAT activity in different laboratory animals (knock-out mice), as well as human participants, regardless of gender, age, or the effects of different pathological conditions on BAT activity. An additional benefit of this method is the simplicity for participants and laboratory animals, which allows us to estimate the potential benefits of BAT booster therapy. The recent studies using IR thermography for determining the physiological behavior of BAT after cold exposure or a meal are described in the recent publication of Brasil et al.²⁴.

Protocol

All experimental procedures on laboratory animals were approved by the National Ethical Committee and the Ministry of Agriculture (EP 185/2018). The experiments were conducted in accordance with the Ethical Codex of The Croatian Society for Laboratory Animal Science and ARRIVE guidelines. All procedures performed in studies involving human participants were in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the University of Zagreb, School of Medicine (UP/I-322-01/18-01/56). In this study, we present the results from three female participants (BMI: 29 kg/m² ± 5 kg/m²). Informed consent was obtained from all human volunteers for their participation in the study and for presenting the data.

1. Measuring the activation of brown adipose tissue after a meal in humans

NOTE: Perform the experiments during the summertime when the daily temperature is not below 22 °C to keep basal BAT activity as low as possible.

1. Carefully choose the control healthy participants (if BAT activity is to be estimated under pathological conditions) since BAT activity is gender, age, BMI, and even the phase of estrous cycle dependent.
   1. To estimate the phase of the menstrual cycle of female participants, ask them questions about how long their average menstrual cycle is and the date of the first day of their last menstruation. Do not forget to mark the date of the experiments.
      NOTE: Proper selection of matching control subjects is the hardest part of clinical studies since healthy controls and participants with pathological conditions should be as similar as possible and differ only in the disease investigated.
2. Ask the participants to rest well, not to have breakfast (fasting-no intake of calories), to gather in the morning hours for the experiments, and to rest for at least 30 min to avoid possible BAT activation during muscle activity via sympathetic activation.
3. Ask the participants to remove their upper clothing 15 min before the measurements to avoid the possible effects of warming the skin surface (thermal effects of clothes) while determining the baseline BAT activity. Perform measurements at appropriate room temperature (22-27 °C).
4. Perform infrared measurements.
   1. While the participants are resting, mount the thermal camera (detector type: uncooled microbolometer; detector pitch: 17 µm; camera spectral range: 7.5-14.0 µm; thermal sensitivity: 20 mK at 30 °C; lenses: 36 mm; resolution: 1024 pixels x 768 pixels; instantaneous field of view [IFOV]: 0.47 mRad) on the tripod and position it 1 m away from the spot where the participant will be seated.
      NOTE: If the measurements are performed in colder weather (outside air temperatures below 15 °C at 50% humidity), put the camera at room temperature and turn it on for at least 1 h before performing the measurements. Cold instruments can give different results after warming up to room temperature due to autocalibration.
   2. Connect the thermal camera to a computer and software as instructed by the manufacturer. Record aluminum foil (crumpled then stretched aluminum foil) at a focal distance of 1 m to determine the reflected temperature of the room, presented as measured temperature. In the camera software, input the distance of 0 m and emissivity of 1.
      NOTE: The reflected apparent temperature is a parameter obtained when the emissivity of the camera is set to 1.0 and the distance to 0 m and the measurements are taken on crumpled then stretched aluminum foil. Reflected apparent temperature represents an approximation of the total incident infrared radiation on the detector from the environment.
   3. Just before starting the measurements, determine the room air temperature and air humidity (necessary for later analysis). Instead of taking one thermal image, record a movie. From the movie, later choose the best possible image frame for analysis to reduce the possibility of losing valuable data.
   4. Before starting the recording, set the following parameters: the duration of video recording at 10-15 s (or any other desired value), the frame rate at 5 fps (frames per second) or any other value (in our hands, 5 fps is the maximum needed), and a location on the disk where the movie will be saved, as described below.
      1. In the software above the main camera window, choose the third icon from the left. In the pop-up menu, choose Edit Record Settings, after which a new window will open.
      2. In record mode, select Record to Disk, and below that, set the Record for This Duration at the desired time. In the record options of the same window, limit record rate to 5 (Hz) and choose the location where the recordings will be saved.
      3. To set the frame rate, close the existing window, open Edit in the main menu, and select Preferences. In the right part of the opened window, enter 5 in Target Frame Rate. In the same window below, select Hotkey/Remote Start Can Stop Record, and from the drop-down menu, select In Start/Stop mode.
         NOTE: Try to make the shortest possible movies with the lowest possible frame rate since it is memory-consuming. At these settings, one record will have approximately 100 Mb.
5. Position the participant so that the supraclavicular area of the neck, above the collarbone where BAT is located (Figure 1), is at a 1 m focal distance and record a short movie (10-15 s) at a frame rate of 5 fps by pressing the F5 key. The recording will stop at the designated time.
6. In the room at the time of the measurements, ensure only the participant and the person who is performing the measurements are present. Avoid air movement or draft (e.g., from air conditioning). Ensure the participants are away from cold draft, sunlight (direct or indirect), or any source of heat like light bulbs.
7. If suitable, measure the blood glucose concentrations in capillary blood from the fingertip with a standard glucometer and the body temperature using an axillary thermometer.
8. Ensure all the participants eat the same meal. Pay attention to the food restrictions and requirements of the tested subjects (for example, the meal for diabetic patients). All participants, including control (healthy) people and participants with metabolic disorders, should eat the same meal.
   NOTE: For more details about meals that diabetic patients can consume, contact a local endocrinologist or discuss it with the participants suffering from diabetes mellitus.
9. At the desired time after a meal, make the new recording by pressing F5 using the same setting values. Do not repeat the set protocol for recordings. Repeat the measurements at 30 min, 1 h, 2 h, and 3 h after a meal¹². For your specific study design, the time after a meal could be shorter or longer, but we recommend at least the first three time points.
   ​NOTE: The limitation of the number of participants is four to six, even though the measurements are done fast. With a higher number of participants, the delay time for some will be too long.

2. Measuring the activation of brown adipose tissue after a meal in laboratory animals

NOTE: Since the animals are housed in an animal facility with regulated room temperature and a day/night cycle of 12 h/12 h the experiments could be performed during any season. The room temperature during experiments should be between 22 °C and 27 °C. In this study, six female animals in diestrus and six male wild-type (WT) C57Bl/6NCrl animals were examined.

1. Anesthetize the animals as per the institution's ethical guidelines. In this study, anesthesia was performed using i.p. injections of ketamine/xylazine (80-100 mg/kg and 6-8 mg/kg, respectively). Apply eye gel to both eyes to prevent corneal drying during anesthesia. Shave the interscapular regions of the test animals a day before the experiments (the region of the skin between the shoulder blades) using a small animal trimmer.
2. The day before the experiments, also determine the phase of the estrous cycle in female animals.
   NOTE: The phase of the estrous cycle is determined by vaginal smears.
   1. Dip a cotton tip swab in room-temperature sterile saline solution (0.9% NaCl) and insert it into the vagina. Gently scrape the vaginal wall with the swab, spread the attached cells onto a glass slide, and allow it to air dry.
   2. Put the animals back in their cages. Stain the cells with 500 µL of 0.1% cresyl violet acetate for 1 min, after which rinse them 3 times with water.
   3. View the cells under a light microscope with 100x magnification and bright-field illumination. Determine the phase of the estrous cycle based on the number of leukocytes and nucleated and cornified epithelial cells observed in the smear²⁵.
3. Remove the animals' food the evening before the experiments (fasting overnight) with water ad libitum. The best way is to transfer the animals to new clean cages to avoid possible remnants of food in the cages.
4. On the morning of the experimental day, prepare the thermal camera and recording settings as done for testing the human participants.
5. Do not disturb or cause stress to the animals before performing IR measurements. Carefully place the animal in a clean cage (ensures there are no effects of other animals' scent on the animal's sympathetic system). Place the cage under the thermal camera at a focal distance of 1 m. Record a movie by pressing F5.
6. Weigh the food pellet before giving it to each animal so that food intake can be calculated. Allow the animal to eat for 30 min in its cage and weigh the food pellet again after the meal. In this study, female animals ate 0.038 ± 0.004 g food/body weight.
   NOTE: If you decide to measure blood glucose concentrations, perform the measurements before a meal but after IR measurements to ensure that this will not lead to BAT activation by the sympathetic system.
7. Repeat IR measurements at the desired time after starting a meal (usually 30 min, 1 h, and 2 h after a meal)^17,26.
8. After all the experiments are completed, test again the phase of the estrous cycle in female animals as described above (female animals may exit the desired phase of the estrous cycle sooner than anticipated).

3. Analyzing the thermal recordings

NOTE: The thermal camera software calculates the object's temperature using five variables.

1. Set the following variables in the software before analysis: skin emissivity, e = 0.98^15,27, reflected room temperature (as calculated from the image of aluminum foil), air temperature, relative humidity, distance to the object = 1 m. Perform the analysis by using the software with these values.
   NOTE: The preferred color palette is rainbow since it uses more hues, which allows for easier detection of BAT above the collarbone.
2. For each movie, enter the listed variables into the camera software on the right side of the main window. Select the suitable frame (image) from the movie by moving the play head at the bottom of the screen or pressing the Pause button.
3. Select the region of interest (ROI) by choosing the desired shape of the area on the left side of the main window. Choose the shape that best corresponds to the area of skin above or between the collarbones.
4. When the ROI is chosen, the minimal, maximal, and average temperatures of the ROI are displayed on the right side. In the image, the red triangle represents the point of maximal recorded temperature, and the blue triangle represents the minimal recorded temperature. Repeat this step for several frames to be sure that the measured temperature is stable during a few seconds of the recording.
5. Subtract the maximal temperatures of the skin area above BAT before a meal from the maximal temperatures after a meal to determine the increase in postprandial BAT activity in laboratory animals.

Results

The easiest way to determine BAT activity is to subtract the maximal skin temperature above the BAT before and after a meal in human subjects. A better way to calculate BAT activity is to select two regions of interest: the skin area above the BAT, which is located in the supraclavicular area, and the interclavicular area of the skin where no BAT tissue is found in humans, designated as a reference area (according to PET-CT; Figure 1). BAT activity is then easily determined by the subtractio...

Discussion

Recent studies present growing evidence regarding the physiological regulation and importance of BAT activity in adult humans and animals in the development of obesity and diabetes mellitus. Furthermore, possible BAT activation by exogenous activators is becoming a target for pharmaceutical companies. To be able to estimate the physiological regulation and pathophysiological importance of BAT in very burdensome diseases, as well as discover a potential therapeutic approach, infrared thermography is becoming the method of...

Materials

Name	Company	Catalog Number	Comments
0.1% cresyl violet acetate	Commonly used chemical
Device for measuring air temperature and humidity	Kesterl	Kestrel 4200	Certificat of conformity
External data storage	Hard Drive with at least 1 TB
Glass microscopic slides	Commonly used
Small cotton tip swab	Urethral swabs
Software for analysis	FLIR Systems, Wilsonville, OR, USA	FLIR Tools
Software for meassurements	FLIR Systems, Wilsonville, OR, USA	ResearchIR software	FLIR ResearchIR Max, version 4.40.12.38 (64-bit)
Thermac Camera	FLIR Systems, Wilsonville, OR, USA	FLIR T-1020