Synthetic swab

The participants were asked not to eat, drink, or use dental hygiene products at least one hour before sampling. Throat samples were collected from the tonsils with standard technique using sample collection cotton swabs. Samples were cultured immediately on agar plates (TSA with 5% sheep blood, BD, USA) and incubated 16–20 h at + 35 °C with 5% CO₂. β-haemolytic, catalase negative isolates were analysed with the Streptex agglutination test (Remel Europe Ltd, Kent, UK) to identify Lancefield group A streptococci and stored at − 80 °C. Isolates confirmed as GAS were emm typed according to the CDC standardised protocol (https://www.cdc.gov/streplab/groupa-strep/emm-typing-protocol.html).
Saliva samples were collected by chewing a synthetic swab for 1 min (Salivette, Sarstedt, Germany). To separate the saliva from the swab, the collection tubes with the swab were centrifuged (1000 × g, 2 min) and the saliva was transferred to new, sterile tubes and stored at − 80 °C until further analysis. Saliva collection and separation method used yields mainly to serous fraction of the saliva. The volume of collected saliva was recorded.

Experimental procedures and task designs are shown in Figure 1a,b. Two widely used stress paradigms were selected: a speech task to induce social stress and a math calculation task to cause mental loading stress (Figure 1b). Counting from 1 to 10 along with breathing was employed as a relaxing task (Figure 1b). Each task consisted of 4 task runs; one 15‐min resting‐state run was acquired pre‐ and post‐task runs (Figure 1a). Task details could be found in the Supporting Information.
In order to confirm that the stressful and the relaxing tasks could induce stressful and relaxation states successfully, subjective ratings on stress level and cortisol level were collected (Figure 1a). The stress level was evaluated at baseline (pre‐task), during the task (immediately after each run), and after the task (post‐task) using a visual analog scale by an integer between 1 and 10, with 10 corresponding to the most catatonic state they had experienced before. Saliva samples were collected four times: at baseline (pre‐task), during the task (after the second and fourth task run), and after the task (post‐task) by instructing subjects to chew a synthetic swab (Sarstedt) for 1 min. The saliva samples were stored at −20°C until they were assayed by Kangjiahongyuan Biotechnology Co., Ltd.