Can salivary lactate be used as an anaerobic biomarker?

Samples

The participants were 20 professional sprinters, including eleven females and nine males. Six of them were 100-meter runners, three were 200-meter runners and eleven were 400-meter runners. They had been engaged in training for three years or longer at a training center where the study was conducted. Verbal informed consent from the athletes and their coaches were obtained prior to sampling. All the participants fulfilled the following criteria: (1) they were in good health, without sport injuries or bleeding wounds in the mouths, nor underwent dental surgery in the last three months; (2) they did not smoke, drink alcohol, or participate in strenuous physical activities 24 h prior to the study; and (3) they did not eat food or chew gum two hours prior to sampling. The study was conformed to the Declaration of Helsinki and approved by the ethics committee of the Stomatological Hospital of Chongqing Medical University, with the approval number of 2021(061).

To obtain a saliva sample, a medical cotton ball weighing approximately 0.3 g was chewed for one minute, and spitted into a 10 ml centrifuge tube. The tube was immediately sealed and temporarily stored in an ice box for no more than 2 h before transported to the laboratory and stored at −20 °C.

The evaluation was consisted of five parts: (1) validation of method accuracy and precision, (2) the influence of mouth rinses, (3) consistency at resting state, (4) variation during training, and (5) correlation with blood lactate. Accordingly, the participants were asked to provide saliva samples as follows: (1) at resting state, before and after one, two and three times of mouth rinses with 10 ml of drinking water (four females and six males); (2) at resting state, before breakfast (approximately 7 am) in Monday mornings of three consecutive weeks and after three times of mouth rinses (11 females and nine males); and (3) five minutes before, immediately after and five minutes after acute high-intensity treadmill training with increasing load or cycle ergometer trainings (eight females and eight males) after three times of mouth rinses.

As many professional sprinters in the training center as possible were recruited into this study. Sample sizes required were calculated based on the effect sizes of previous similar studies (Bardon, Ceder & Kollberg, 1983; Ohkuwa et al., 1995; Segura et al., 1996; Calvo et al., 1997; Bocanegra et al., 2012; Tekus et al., 2012), the significant level of 0.05 and the power of 0.8, which were 6 (95% CI: 0–13) for paired group comparisons and 11 (95% CI: 3–20) for Pearson correlation analysis, respectively. Hence, the planned sample sizes (10, 20 and 16 for paired group comparisons and 32 for Pearson correlation analysis of lactate concentrations, respectively) were larger than the required sample sizes (6 for paired group comparisons and 11 for Pearson correlation analysis of lactate concentrations, respectively).

The treadmill and cycle ergometer trainings started at approximately 9 am on two Thursday mornings with a one-month interval. The participants provided their information including sex, age, height and body weight, and sat quietly for 20 min prior to the trainings. The acute high-intensity treadmill training with increasing load was conducted with a Metalyzer 3B (Cortex, Leipzig, Germany) spiroergometry system and a RUN 7410 treadmill (RUNNER, Modena, Italy) preheated for 30 min. The training included five-minute warm-up at 8 km/h and 0° grade, 8 min load-increasing stage with increasing speed and slope every minute based on the participants’ individual capacities, and three-minute recovery stage at 5–7 km/h, 0° grade. The participants were equipped with heart rate monitors and accompanied by a team doctor. They were encouraged to complete the training but allowed to stop anytime when exhausted. The exhaustion criteria were set as the heart rate remaining at no less than 180 beats/min for two minutes, and the rate of perceived exertion (RPE) scale (Borg, 1982) reaching 18–20. The cycle ergometer training was similar to the Wingate test, which was completed on a Monark 874E weight cycle ergometer. Each participant was asked to ride on the cycle ergometer for 3 × 30 s with 3 min intervals at the load of 0.075 kg/kg body weight. Blood lactate was measured before and immediately after the cycle ergometer training using a Lactate Scout Lactate analyzer (EKF Diagnostics, Cardiff, UK), and transformed from mmol/L to µg/ml. Since the actual sample size of blood lactate was more than double of the required sample size, no blood lactate test was conducted during the treadmill training to reduce the participants’ stress.

Instruments and procedures

Prior to analysis, the cotton ball soaked with saliva was defrosted at room temperature and transferred into a 4 ml Eppendorf tube with a small hole drilled at the bottom. The 4 ml tube was then put into a clean 10 ml Eppendorf tube whose cap was cut off. The inner diameter of the 10 ml Eppendorf tube was just between the outer diameters of the body and edge of the 4 ml Eppendorf tube, enabling the 4 ml tube to be stuck at the top of the 10 ml tube. After that, the 10 ml tube was centrifuged with an Eppendorf Centrifuge 5920 R (Hamburg, Germany) at 4 °C and 3,000 rpm for 3 min, so that the saliva in the cotton ball was collected in the 10 ml Eppendorf tube. 100 µl of the saliva sample was transfer into a 1.5 ml Eppendorf tube, and vortex-mixed with 20 µl of 1 µg/ml internal standard 2-Cl-phenylalanine. Then, 1 ml of acetonitrile was added and the mixture vortexed again. After that, the 1.5 ml centrifuge tube was centrifuged with an Eppendorf Centrifuge 5427 R (Hamburg, Germany) at 4 °C and 10,000 rpm for 10 min. The supernatant was transferred to a new 1.5 ml Eppendorf tube, dried with a Labconco CentriVap vacuum dryer (MO, USA) at 40 °C, and then reconstituted in 100 µl of 5% methanol. The solution was centrifuged again at 4 °C, 10,000 rpm for 5 min and 80 µl of the supernatant was transfer to an LC-MS vial for injection. Calibration samples were prepared in the same way (starting from internal standard addition) as the saliva samples. A pool sample was prepared by mixing 100 µl aliquots of all the test samples and used in method validation.

The LC-MS analysis was performed using an Agilent 1290-6495C Ultra High Performance Liquid Chromatography-Triple Quadrupole Tandem Mass Spectrometry (CA, USA). LC separation was performed using a Waters HSS T3 (100 ×2.1 mm, 1.7 µm) column with a Waters ACQUITY UPLC HSS T3 VanGuard (2.1 × 5 mm, 1.8 µm) pre-column. Mobile phase A was aqueous solution of 0.1% formic acid, and mobile phase B was acetonitrile. The gradients were: 0–0.5 min: 5% B, 0.5–1 min: 5–50% B, 1–7.5 min: 50% B, 7.5–8 min: 50–5% B, 8–10 min: 5% B. The flow rate was 0.3 ml/min, the injection volume was 2 µl, and the column temperature was 40 °C. The ion source parameters for the mass spectrometry were: Gas Temp: 200 °C, Gas Flow: 14 L/min, Nebulizer: 35 psi, Sheath Gas Temp: 300 °C, Sheath Gas flow: 11 L/min, Capillary: −3000 V, Nozzle Voltage: 500 V, iFunnel H: −150 V, L: −60 V. Multiple reaction monitoring (MRM) mode was used to analyze lactate and the internal standard 2-Cl-phenylalanine. Q1(m/z), Q3(m/z) and CV(eV) were set as 89.0, 43.2 and 13, respectively, for lactate; and 198.0, 181.0 and 9, respectively, for 2-Cl-phenylalanine. Lactate standard was purchased from Bidepharm (Shanghai, China), and 2-Cl-phenylalanine was from MedChemExpress (Shanghai, China).

The method was validated for accuracy and precision. Accuracy was determined by adding 10 µl/ml of lactate into a pool sample and calculating the recovery, which was the ratio of measured concentration increase to the theoretical concentration increase (10 µl/ml). Precision was determined by analyzing seven 100 µl sub-aliquots of the pool sample twice, and calculating the relative standard deviation (RSD) of the fourteen measured concentrations.

Each set of data is expressed as mean ± standard deviation (SD). Paired one-way analysis of variance (ANOVA) and Tukey’ post-hoc multiple comparisons test were used to compare data among different time points of the same group of participants, and multiplicity adjusted P value for each comparison with family-wise significance and confidence level set at 0.05 was reported. Intra-class correlation coefficients (ICC) and RSD of multiple sampling times were used to evaluate data variances of each participant. Pearson correlation coefficient (r) was calculated for salivary lactate and blood lactate of the same participants at the same time points. Effect sizes and power were calculated following ANOVA and correlation analyses. ANOVA, RSD and Pearson correlation were calculated using GraphPad Prism 8 for mac (GraphPad Software, Boston, MA, USA), while ICC was calculated with IBM SPSS Statistics software, version 20.0.0 (IBM Corporation, Armonk, NY, USA). Effect sizes of ANOVA (η²) were calculated from ANOVA F values, using the equation η² = F × (k−1) / [F × (k−1) + (n−k)] (Cohen, 1988), while those of Tukey’s post-hoc tests (Cohen’s d) and t-tests were calculated with R for Mac OS X GUI 1.70 (R Core Team, 2020) and package “esc” (Lüdecke, 2019). Power (1–β) and sample sizes of all analyses were calculated with R software and package “pwr” (Champely et al., 2020).