•

Analytical balance

The animals are male rats ( Rattus norvegicus) obtained from Faculty of Veterinary, Universitas Syiah Kuala, Banda Aceh, Indonesia. The animals must be three–four months old, weighed 160–250 g, and not showing any abnormalities.

To ensure the statistical power will be achieved, the sample size is determined by Federer’s equation ([n − 1][t − 1] ≥15); where n is the number of sample per group and t is the number of groups. In this study, the number of groups is four, so that six animals are at least required for each group.

Anechoic chamber has been built at Acoustic, Vibration, and Thermal Laboratory, Faculty of Engineering, Universitas Syiah Kuala, Indonesia. Installation of sound-insulating foam panels on the wall results in a chamber with field-free and sound propagation aligned with ‘inverse square law’. Specific construction design, including the access door and ventilation, allows high sound attenuation and free from structure-borne noise. The chamber has 7.8 m length, 6.3 m width, and 5 m height. It is built with two layers of wall and three layers of iron door to minimize the entrance of signal from the outside. Its interior has a dimension of 6 m × 4.5 m × 3.1 m (length, width, and height, respectively), where the airflow is maintained by a controllable ventilation. The chamber has been installed with a lighting system (6 × 80 Watt lamps) with brightness level of 500 lux. Appearance of the anechoic chamber has been presented in Figure 2.

As many as three modules of sensor dht-22 are installed surrounding the chamber interior. The sensor is connected to a microcontroller using cable to record the temperature and humidity. Data obtained in the microcontroller are then recorded and displayed through a computer.

Signal strengths from the Global System for Mobile communication (GSM) in the anechoic chamber are measured with spectrum analyzer (max. 3000 MHz). The schematic diagram for signal strength measurement using spectrum analyzer has been presented in Figure 3.

The signal of radio frequency (RF) in the anechoic chamber has a frequency of 891—893 MHz (GSM and UMTS signals) with Gaussian Minimum-Shift Keying (GMSK) modulation scheme for GSM, Quadrature Phase Shift Keying (QPSK) – UMTS, and 8PSK – EDGE. In the case of GMSK, the signal is smoothed using low-pass Gaussian filter prior to its modulation into signal carrier and proceeding to frequency modulator. As for QPSK and 8PSK, variants of Phase Shift Keying (PSK), the difference is only located at the code rate capacity. The design and photographed images of the GSM module used in this research have been presented in Figure 4.

The signal noise in the anechoic chamber is very low, ranged from -79 dBM to -74 dBM (1.2589 × 10 ^-8 mW). The number was obtained from the determination using spectrum analyzer. Several measurement results of the RF/GSM signal frequencies at distance of 10–50 cm has been presented in Table 1.

RF applicator, in a form of a cage, has a dimensional size of 33 × 20 × 7 cm ³ with pentagonal shape divided into five chambers. The cage is made from glass where wire mesh is used to cover the top for air circulation during experiment. The distance between the RF signal and the cage is maintained at 10–30 cm.

The level of electromagnetic wave radiation is represented by Specific Absorption Rate (SAR), an expression to explain the absorbance rate of the electromagnetic radiation (RF) per time unit per kilogram (W/kg). SAR is varied in different parts of the body. The mathematical expression of SAR has been presented in Equation 1. SAR = σ E 2 ρ (1)

Where σ is electrical conductivity in the observed tissue (S/m), E is electrical field (V/m), and ρ is the density of body tissue (kg/m ³).

In the real world, the value of ρ is varied, considering each rat has differences in their skin, fat, bones, and so on. Herein, the body density of each rat is assumed to be 1000 kg/m ³. Since the value of σ in thyroid gland is unknown, we assume the value is equal to that in the tissue nearby, namely trachea (σ = 1 S/m for 2 GHz frequency). Since the value is applicable for rat with the size of around 20 cm × 4.5 cm × 4.5 cm with an averaged body weight of around 200 g.

To calculate SARS value, it is of importance to calculate the depth of RF signal penetration using the following equation ( Equation 2): δ = 1 π μ σ f (2)

Where δ is the depth of RF penetration (m), μ is permeability of the observed tissue (H/m), and f is averaged RF signal frequencies (mW). Similar to tissue conductivity, the value of permeability for thyroid gland is currently unknown and therefore is assumed to be equal with that of trachea. In the case of RF penetration into throat tissue, the δ is 1.1 and the averaged uplink frequency for GSM is 9902.5 MHz. Therefore, by using the Equation 2 we obtain the value for δ = 0.0159 m.

In a GSM 900 system the maximum power is 2 W; hence, the maximum averaged power ( P ) is 250 mW (1/8 maximum power). If the averaged distance between the phone antenna and rat ( A ) is 10 cm, the radiation density with maximal penetration ( I ) could be calculated using Equation 3 resulting in I = 1.481 W/m ². Electrical current is expressed by Equation 4 which can be derived to Equation 5 by substituting P . With the skin impedance ( η ) value of 59.094 Ohm, the electrical field strength ( E ) could be calculated using Equation 5 resulting in E = 9.355 V/m. By inserting all the values required in Equation 1, the SAR is obtained to be 0.0962 W/kg. Equations and biological parameter constants used in calculating the SAR are derived from a previously published report. ^{26 , 27} I = P A (3) P = E 2 η (4) E = I x η (5)

Simple random sampling is employed to randomly assign the animals into four groups, where each group consists of 10 animals. All subjects are considered having the same conditions prior to the treatment. The group assignment has been presented in Table 2.

The animals, aged two–three months with body weights ranged from 160 to 200 g, are acclimated at room temperature in 12 h dark–12 h light cycle. The animals are fed ad libitum with standardized pellets and water, while being contained in a plastic cage (40 × 25 × 20 cm). The acclimation process lasts for seven days with observation on their behaviors and physical conditions.

Rats in intervention group (Group II–IV) are exposed with mobile phone-derived electromagnetic wave radiation for 120, 150, and 180 min/day, respectively. The daily exposure duration chosen in this research is based on the results from our preliminary, where changes thyroid functions and other metabolic parameters occurred from 120 to 150 min/day exposure (unpublished). In this study, 180 min/day exposure duration will be employed to observe new findings. The radiation will be exposed in dark condition using two bidirectionally interacting GSM modules, in which their set-up has been presented in Figure 5. Each cage is divided into five segments with the antenna placed in the center. Each segment will consist of a single rat, and each group will receive simultaneous exposure by employing two separated cages to avoid the disturbance of electromagnetic field. RTC is used to control the time of the interaction. The radiation is produced from the stimulation of GSM multiband 4G LTE signal with a frequency of 891–893 MHz (maximum frequency for 3G spectrum is 1800 MHz) ( Figure 6). The exposure is carried out for 12 consecutive weeks in an anechoic chamber.

Venous blood is drawn from the tail of each subject which has been mechanically constrained without anesthesia. The tail is heated for vasodilation and better venous visibility which could be performed by using heating chamber, heating pad, or heating lamp or by immersing in a warm water. Avoid excessive heating; the animal should be under the lamp no longer than three minutes with a distance of at least 30 cm. All equipment should be sterile, and the skin surface should be firstly cleaned and disinfected with alcohol 70%. Immobilize the non-dominant tail by hand, and rotate ¼ circle to access the lateral tail vein. The needle is aligned with the tail to 30° for the insertion into the distal vein. Fixation is applied slowly and evenly, and the needle is removed after finished. Observation should be carried out continuously during the study period on the subjects’ behaviour associated with the ongoing pain.

The euthanasia should be carried out by professionals. The rats are physically euthanized by cervical dislocation by applying a firm pressure using the thumb and index finger placed on the posterior skull base and spinal cord. Another hand holds the tail part for a quick backward pulling to dislocate the thoracic vertebrae. Check the respiratory arrest and the absence of heartbeat. Thereafter, the organs are harvested by dissection.

The collected blood sample is stored in an EDTA-containing tube at 2–8°C for maximum 5 days. Blood sample should not be hemolyzed or lipemic. The determinations of serum TSH and T4 were performed on ELISA using commercial kit specialized for Rattus novergicus. Normal ranges for serum TSH and T ₄ are 0.4–6.0 μIU/mL and 5.0–12.0 μg/dL, respectively. The level of MDA will be measured using ELISA.

Brain tissue, along with the cortex, is collected and rinsed with ice-cold sucrose solution 0.25 M. The tissue is slices and homogenized with 10× operant in a slow setting. The specimen is further treated with Dounce homogenizer as many as 3× operant in a fast speed setting. Dissolution is then carried out to obtain homogenate 6% w/v which is further centrifuged for 10 min at 1.464 × g. The resultant pellet is resuspended in isolating medium and subsequently diluted to yield suspension 6% w/v. As many as 10.4 mL of the suspension is mixed with 1.4 mL Percoll in 15 mL cortex tube and centrifuged for 30 min at 35.540 × g. Two layers are formed on the top side of the type, which are then collected and washed with 5 volumes of sucrose 0.25 M, HEPES-KOH 5 mM, pH 7.1. The obtained pellet is resuspended using the same medium before analyzed using ELISA.

The whole pituitary gland is collected and frozen in isopentane on dry ice. The sample is melted on slide superfrost (Thermoscientific) and stored at –80°C. The expression is determined based on immunohistochemical method using ELISA.

Cryosections are performed on the brain tissue, which is then melted on slide superfrost (Thermoscientific) before undergoing a quick freezing on dry ice and being stored at –80°C. Primary antibodies used for the expression determination are mouse antiβ-tubIII, mouse anticalbindin 1:2000, mouse anti –NeuN 1:500, mouse anti-PV 1:2000. Hybridization is carried out in situ at 25 μm part of the juvenile brain. Riboprobes sense and antisense are synthesized from cDNA RC3 template with spanning nucleotide of 253-486. The expression is based on relative expression in the quantitative reverse transcription polymerase chain reaction (RT-qPCR) analysis.

The cerebral cortex is dissected on an ice-cold glass plate and washed in a buffer (sucrose 0.32 M: EGTA 1 mM: Tris-HCL 10 M; pH 7.4). The sample is sliced and homogenized in the aforementioned buffer (1:20 w/v) with Potter-Elvehjem glass homogenizer. The homogenate is centrifuged 10,000 × g for 15 min at 4°C in a Sorval centrifuge. The obtained mitochondria-free supernatant is collected for the enzymatic activity investigation using ELISA.

As much as 5 mL venous blood is drawn from each sample, treated with EDTA to prevent the coagulation. DNA genome is extracted from the blood using Gentra Puregene Blood Kit following the manufacturer’s manuals. The DNA sample is stored at –20°C until further use. Indirect determination of the gene expression is carried out using RT-qPCR.

A whole thyroid gland is collected from the subject through dissection, and subsequently fixated using formalin and paraffin for 24 h. Incision is performed on the thyroid gland with a length of 5 μm and the staining is carried out using hematoxylin and eosin following the standardized procedure. Assessment on thyroid follicles is performed under a light microscope. The assessment include: •

Thyroid epithelial cells appeared in a flat shape indicates the hypofunction of TSH.

Bivariate analysis will be performed using Spearman test and Chi-square test for continuous and categorical data, respectively. The correlations between the intermediary variables and MDA level or NR3C1 expression will be established using ordinal regression and logistic regression. Statistical significance is considered achieved if p<0.05. All statistical analysis will be performed on IBM SPSS Statistics 23.0 (Armonk, NY, USA).

The animal study protocol was approved by the Institutional Ethics Committee of Faculty of Veterinary Medicine, Universitas Syiah Kuala (Ref No. 69/KEPH/XII/2020).

The study has not been completed. The exposure of the electromagnetic wave radiation has been conducted and only some of indicators that have been examined.