Co-author Jing-He Tan is a PLOS ONE Editorial Board member, but this does not alter the authors' adherence to PLOS ONE Editorial policies and criteria.
Conceived and designed the experiments: JHT. Performed the experiments: SG YLM GZJ MJS HL JL MJL. Analyzed the data: SG YLM. Wrote the paper: JHT.
‡ These authors contributed equally to this work.
Although plasma corticosterone is considered the main glucocorticoid involved in regulation of stress responses in rodents, the presence of plasma cortisol and whether its level can be used as an indicator for rodent activation of stress remain to be determined. In this study, effects of estrous cycle stage, circadian rhythm, and acute and chronic (repeated or unpredictable) stressors of various severities on dynamics and correlation of serum cortisol and corticosterone were examined in mice. A strong (r = 0.6–0.85) correlation between serum cortisol and corticosterone was observed throughout the estrous cycle, all day long, and during acute or repeated restraints, chronic unpredictable stress and acute forced swimming or heat stress. Both hormones increased to the highest level on day 1 of repeated-restraint or unpredictable stresses, but after that, whereas the concentration of cortisol did not change, that of corticosterone showed different dynamics. Thus, whereas corticosterone declined dramatically during repeated restraints, it remained at the high level during unpredictable stress. During forced swimming or heat stress, whereas cortisol increased to the highest level within 3 min., corticosterone did not reach maximum until 40 min. of stress. Analysis with HPLC and HPLC-MS further confirmed the presence of cortisol in mouse serum. Taken together, results (i) confirmed the presence of cortisol in mouse serum and (ii) suggested that mouse serum cortisol and corticosterone are closely correlated in dynamics under different physiological or stressful conditions, but, whereas corticosterone was a more adaptation-related biomarker than cortisol during chronic stress, cortisol was a quicker responder than corticosterone during severe acute stress.
It is known that stress enhances the activity of the hypothalamus-pituitary-adrenal (HPA) axis and results in increased secretion of corticosteroids from the adrenal cortex. Cortisol and corticosterone are thus often used as biomarkers for stress and depressive disorders. Although corticosterone is considered the main glucocorticoid involved in regulation of stress responses in rodents, researchers often choose to detect cortisol for stress indicators in consideration of convenience and kits availability. In fact, several studies have observed increased cortisol in plasma and adrenal glands of mice following stress [
Studies in rabbits indicated that the ratio of cortisol to corticosterone might be influenced by some physiological conditions. For example, representation of cortisol was observed to be favored in fetal stages of development [
Taken together, the above review suggests that the representativeness of cortisol and corticosterone in rodents may be different under different physiological or stressful conditions. Thus, there is an urgent need for research on the dynamics and correlation of cortisol and corticosterone under different physiological or stressful conditions. In this study, effects of estrous cycle stage, circadian rhythm, and acute and chronic (repeated or unpredictable) stressors of various severities on dynamics and correlation of plasma cortisol and corticosterone were examined in mice.
Mouse care and use were conducted exactly in accordance with the guidelines and approved by the Animal Research Committee of the Shandong Agricultural University, P. R. China (Permit number: 20010510). According to the guidelines of the committee, the animal handling staff (including each post-doc, doctoral or masters student) must be trained before using animals. Mice must be housed in a temperature-controlled room with proper darkness-light cycles, fed with a regular diet, and maintained under the care of the Experimental Animal Center, Shandong Agricultural University College of Animal Science and Vet Medicine. In the present study, mice were sacrificed by decapitation. The only procedure performed on the dead animals was the collection of trunk blood.
Unless otherwise specified, all chemicals and reagents used in the present study were purchased from Sigma Chemical Co. (St. Louis, MO, USA).
Mice of the Kunming strain were used at the age of 6–8 weeks. The mice were kept in a room with 14h/10h light-dark cycles, the dark starting from 20:00 pm. The stage of the estrous cycle in female mice was determined by observing vaginal appearance [
Restraint of small animals is an experimental procedure developed for studying psychogenic stress [
For forced swimming testing, animals were forced to swim for different times in a rectangular plastic tank (45×35×18 cm) containing 15 cm deep water. The water temperature was maintained at approximately 23°C. After swimming, the mice were toweled dry before being sacrificed for blood collection.
For unpredictable stress treatment, mice were exposed to different stressors for two rounds of 4 days. Stressors used on different days were as follows: Day 1, 2-h heat stress in oven at 42°C; Day 2, 2-h shaker stress (160 rpm); Day 3, 2-h forced swimming at 23°C; and Day 4, 8-h restraint stress. Whereas the stressors lasting for 2 h were administered from 14:00 to 16:00 pm, the 8-h restraint took place from 8:00 am to 16:00 pm as described above.
In the present study, mice were always sacrificed at 15:00–16:00 pm for blood collection except for the experiment on the effect of the estrous cycle stage in which mice were always killed at 19:30–20:00 pm and for the experiment on the effect of circadian rhythm in which animals were killed at different times of day. Mice were decapitated rapidly at the end of the stress period, and trunk blood (about 1 ml) was collected into ice-cooled centrifugal tubes and centrifuged (1700 ×g, 10 min, 4°C) to separate serum. The serum collected was divided into two parts; one for assay of cortisol and the other for assay of corticosterone. The serum samples were stored at −80°C until hormone assay.
Radioimmunoassay for cortisol was conducted by the Central Hospital of Tai-An City using commercial kits from 3V Biomedical Techniques Co. Ltd., Weifang, China. The kit measures total cortisol in serum including the cortisol combined with corticosteroid-binding globulin (CBG). The minimum level of detection for assays of cortisol was 0.15 ng/ml. The intra- and inter-assay coefficients of variation were <10% and <10%. The cross reactivity of the cortisol RIA kit for corticosterone is 0.11% (tested at the 50% binding). To further evaluate the cross-reactivity of the cortisol RIA with corticosterone, corticosterone and cortisol standards or their mixtures at different concentrations were subjected to the radioimmunoassay kit.
Corticosterone concentrations were measured with an ELISA kit purchased from Arbor Assays Company (Catalog Number K014-H1). The minimum level of detection of the kit was 16.9 pg/ml corticosterone. The kit measures total corticosterone in serum including the corticosterone combined with corticosteroid-binding globulin (CBG). The cross reactivity of the corticosterone kit for cortisol is 0.38% (tested at the 50% binding). Briefly, 50 μl of standards or samples were added in duplicate to wells of the micro-titer plate. 75-μl of assay buffer was added to the non-specific binding (NSB) wells and 50 μl of assay buffer was added to wells to act as maximum binding wells. Then, 25 μl of the DetectX Corticosterone Conjugate and 25 μl of the DetectX Corticosterone Antibody (except the NSB wells) were added to each well and the titer plate was shaken for 1 h at room temperature. After the plate was washed using the wash solution and blot dried by hitting plate onto paper towels, 100 μl of the TMB Substrate were added to each well and incubated for 30 min at room temperature. The optical density (O.D.) of corticosterone was read at 450 nm wavelength using a plate reader within 15 min after the reaction was terminated by adding 50 μl of the Stop Solution. The concentration of corticosterone was calculated according to the standard curves.
To measure serum glucocorticoids, a liquid-liquid extraction was performed by adding 1.6 ml ether to 0.8 ml of serum and vortexing for 2 min. After 10 min of rotating at 1200×g, the sample was centrifuged at 3800×g for 10 min at 4°C. The organic layer was collected and transferred to a 15 ml tube. This liquid-liquid extraction was repeated by adding another 1.6 ml ether to the remaining serum material. The organic phase was dried under a gentle stream of nitrogen at a temperature of 40°C and reconstituted in 2 ml methanol for HPLC analyses. Qualitative detection of the glucocorticoids in serum of mice was performed by reversed phase high performance liquid chromatography (HPLC) and electrospray ionization mass spectrometry (ESIMS). The HPLC was conducted on an Agilent 1260 series instrument (Agilent Technologies, Waldbron, Germany) equipped with an in-line degasser, quaternary pump, autosampler and a diode array and multiple wavelength detector, using a 5-μm YMC-Pack Pro C18 (250 mm×4.6 mm i.d.) column for separation, with acquisition set at 242 nm for reference substances and serum samples of mice. The ESIMS was carried out on a Thermo Finnigan MSQ 10275. Mass and MS/MS spectra were achieved by ESI in positive ion mode. The electrospray voltage was set at 4.2 kV, the capillary temperature at 300°C and vaporizer temperature at 350°C. The sheath and auxiliary gas are nitrogen and their pressures were set to 40 and 5 arbitrary units. The mobile phases for HPLC consisted of water: methyl alcohol (40: 60% v/v) at a flow rate 1 ml min-1. The mobile phases for ESIMS consisted of 2 mM ammonium acetate: methyl alcohol (40: 60% v/v) at a flow rate 0.7 ml min-1. The injection volume was 20 μl and the column temperature was set at 35°C. Peaks were identified based on the retention time of the standards and confirmed by comparison of the wavelength scan spectra (set between 210 nm and 400 nm).
At least three replicates were performed for each treatment. Percentage data were arc sine transformed and analyzed with ANOVA; a Duncan multiple comparison test was used to locate differences. The Statistical Package for Social Science software (version 11.5; SPSS, Inc.) was used. Data are expressed as the mean ± SEM, with P<0.05 considered to be statistically significant. Pearson correlation coefficient test was performed to determine the correlation between cortisol and corticosterone under different circumstances, with P<0.05 considered to be statistically significant. All values were checked for Gaussian distribution by K-S (Kolmogorov-Smirnov) in SPSS before further analysis.
Between 19:30 and 20:00 on each experimental day, female mice were examined for the stage of the estrous cycle by observing vaginal appearance. Mice at different stages of the estrous cycle were then sacrificed to collect blood. Immediately following blood collection, the mice were examined again by observing vaginal lavage smears to confirm their estrous cycle stages. No significant changes were observed in either cortisol or corticosterone concentration throughout the estrous cycle (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
Male mice were sacrificed at different times of day to collect blood for assay of cortisol and corticosterone. Concentrations of both cortisol and corticosterone remained constant from 3:00 in the early morning to 15:00 in the afternoon, but they went up significantly at 20:00 in the evening (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
Female mice were restrained for different times before being sacrificed to collect blood for hormone assays. Concentrations of both cortisol and corticosterone went up to the highest level within 1 h of restraint stress (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
Female mice were repeatedly restrained for different days before being sacrificed to collect blood for hormone assays. Concentrations of both cortisol and corticosterone went up to the highest level on day 1 of restraint (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
To verify our above conclusion that the dynamics of corticosterone was more adaptation-reflective than that of cortisol, cortisol and corticosterone dynamics was compared between the repeated restraint stress that is known to cause animal adaptations and the unpredictable stress that is supposed not to induce animal adaptations. Male mice were exposed to repeated restraints or unpredictable stressors for different days before being sacrificed to collect blood for hormone assays. Both hormones went up to the highest level on day 1 of both repeated restraint and unpredictable stresses (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
To study the rapid effects of severe stresses on dynamics and correlation of serum cortisol and corticosterone, male mice were exposed to forced swimming or heat stress for different times before being sacrificed to collect blood for hormone assays. During both forced swimming and heat stresses, whereas the concentration of cortisol increased to the highest level within 3 min, the concentration of corticosterone did not reach the highest level until 40 min of the stresses (
Values without a common letter differ (P < 0.05) within cortisol or corticosterone groups.
Because of the possibility that the radioimmunoassay used for cortisol measurement in the above experiments might have cross-reactions with corticosterone, two experiments were conducted to verify whether cortisol is indeed present in the mouse blood. First, corticosterone and cortisol standards or their mixtures at different concentrations were subjected to the radioimmunoassay. Results showed that the cortisol concentration detected in the cortisol standard sample containing 10 ng/ml cortisol did not differ from that detected in the mixture of 1000 ng/ml corticosterone and 10 ng/ml cortisol (
Each treatment was repeated 3 times using serum from 10 unstressed mice.
0 | 10 | 9.8±0.26 |
1000 | 0 | 0.6±0.09 |
1000 | 10 | 10.0±0.26 |
3000 | 0 | 1.9±0.23 |
3000 | 10 | 11.8±0.08 |
9000 | 0 | 5.0±0.11 |
9000 | 10 | 15.1±0.20 |
a-f: Values with a different letter in their superscripts differ (P<0.05). Each treatment was repeated 4 times.
The present results showed that the dynamics of serum cortisol was closely correlated with that of corticosterone under all the physiological or stressful conditions tested (r = 0.6−0.85). Both hormones increased to the highest level on day 1 of repeated-restraint or unpredictable stress, but after day 1, whereas the concentration of cortisol did not change, the level of corticosterone showed different dynamics. Thus, whereas corticosterone declined dramatically during repeated restraints, it did not change significantly during unpredictable stresses. During forced swimming or heat stress, whereas cortisol increased to the highest level within 3 min, corticosterone did not reach maximum until 40 min of stress. Furthermore, the smallest coefficients we observed during forced swimming (r = 0.594) and repeatedly restraint stresses for 8 days (r = 0.64) further verified the different dynamics between cortisol and corticosterone during these two stresses. Taken together, the results suggest that mouse serum cortisol and corticosterone are closely correlated in dynamics under different physiological or stressful conditions, suggesting that both hormones can be interchangeably used as an indicator for stress activation in mice. However, whereas corticosterone is a more adaptation-related biomarker than cortisol during chronic stress, cortisol is a quicker responder than corticosterone during severe acute stress. In guanacos (Lama guanicoe), it has been shown that cortisol and corticosterone exhibit different patterns in the field and in response to acute stressors [
Previous studies observed similar dynamics of corticosterone in rodents under the stressful conditions similar to those tested in the present study. For example, Mizobe et al. [
Previous studies also showed cortisol elevations in rodents under the stressful conditions similar to those tested in the present study. For example, a significant elevation in cortisol has been observed in mice after acute restraint stress [
Although Champlin [
The present results indicated that while concentrations of both cortisol and corticosterone remained constant from 3:00 in the early morning to 15:00 in the afternoon, they elevated significantly at 20:00 in the evening. One striking feature of the regulation of glucocorticoids is a diurnal release pattern, with peak levels linked to the start of the daily activity phase [
In summary, although plasma corticosterone is considered the main glucocorticoid involved in regulation of stress responses in rodents, the presence of plasma cortisol and whether its level can be used as an indicator of stress has not been determined in rodents before. Because previous studies indicated that the ratio of cortisol to corticosterone might be influenced by physiologic conditions, developmental stages and organ differences, we observed the effects of estrous cycle stage, circadian rhythm, and acute and chronic (repeated or unpredictable) stressors of various severities on dynamics and correlation of plasma cortisol and corticosterone in mice. Results (i) confirmed the presence of cortisol in mouse serum and (ii) suggested that mouse serum cortisol and corticosterone were closely correlated in dynamics under different physiological or stressful conditions, but whereas corticosterone was a more adaptation-related biomarker than cortisol during chronic stress, cortisol was a quicker responder than corticosterone during severe acute stress. Although the mechanisms for the different dynamics between the two hormones under different stressful conditions are not clear, a recent study by Taves et al. [