Steroid hormones are involved in the regulation of several important bodily functions (e.g., inducing male secondary characteristics, regulating steroidogenesis, retaining body nitrogen, and muscle growth) through both systemic delivery and local production. Systemic steroids permeate plasma membranes and bind to their receptor, and these complexes are subsequently translocated to the nucleus, where they bind to DNA and perform their main biological functions (1–3). In addition, steroid hormones can be locally produced (intracrine) via steroidogenesis in specific cells, including skeletal muscle cells. For instance, the steroid hormone testosterone is primarily produced in Leydig cells in the testicles of men, and cortisol is produced in the cells of the zona fasciculata in the adrenal cortex. However, skeletal muscle cells also have steroidogenesis potential (4, 5), as they express the enzymes that convert cholesterol to a range of steroids, including testosterone, dehydroepiandrosterone (DHEA), dihydrotestosterone (DHT), and cortisol (5–10). Although serum steroid hormones can permeate the sarcolemma and perform their biological function within the myonuclei, evidence supporting that their role, at least partially, in stress-induced responses in skeletal muscles is scarce.

An interesting model to test stress-induced responses in skeletal muscle is the use of resistance exercise (RE), as it can significantly increase serum steroid concentrations (e.g., DHEA, DHT, testosterone, and cortisol) for 15 minutes or more after an exercise bout (11–15). This higher availability of serum steroid hormones could increase intramuscular steroid hormone concentrations through increased permeation. Additionally, exercise-induced increases in serum steroids after an RE bout can increase the expression of steroid receptors over time in skeletal muscle cells (2, 16, 17), further increasing the potential biological effects of elevated hormonal concentrations in these cells. Mechanical stretch (e.g., muscle contractions) per se appears to increase the mRNA levels of steroidogenic enzymes in muscle cells (18), which could allow for additional synthesis of steroid hormones intramuscularly. Some authors showed that steroid concentrations and steroidogenic enzyme expression in human skeletal muscle increased after several RE bouts in older individuals (5). Nevertheless, RE-induced muscle steroidogenesis data are equivocal in human trials (16, 19). Thus, studies are needed to determine whether RE-induced increases in serum steroid hormone concentrations or intramuscular steroidogenesis per se affect intramuscular steroid concentrations.

Based on the aforementioned literature, the purpose of the current study is to investigate whether RE-induced increases in serum steroid hormone concentrations (total and free testosterone, dehydroepiandrosterone sulfate, and cortisol) are accompanied by elevated skeletal muscle steroid concentrations (total testosterone, dehydroepiandrosterone sulfate, dihydrotestosterone, and cortisol), or whether muscle contractions per se induced by RE can increase intramuscular steroid concentrations. Our hypotheses are (1) the RE-induced increase in serum steroid hormone concentrations is accompanied by elevated skeletal muscle steroid concentrations, and (2) muscle contractions per se induced by RE can induce increases in intramuscular steroid concentrations.

Serum cortisol significantly increased at 15 and 30 min after the HH protocol (P = 0.0006; P = 0.0020), with greater concentrations for the HH protocol than the LH protocol at the corresponding postexercise time points (P = 0.0217; P = 0.0335) ( Figure 1 ). There were no significant changes in other serum steroid concentrations ( Figure 1 ) or in muscle steroid concentrations after either protocol ( Figure 2 ).

The intrasubject CVs (between baseline 1 and baseline 2) for serum concentrations of total and free testosterone, DHEA, and cortisol were 17% and 28%, 29%, and 63%, respectively. The muscle concentrations of testosterone, DHEA, DHT and cortisol intrasubject CVs (baseline 1 and baseline 2) were 27%, 17%, 31% and 28%, respectively. To verify whether the delta changes in serum and muscle steroid concentrations of each participant were higher than the measurement error (1xCV and 2xCV), we plotted the delta changes ( Figures 3 , 4 ). For serum cortisol, 83% of subjects presented an increase two times higher than the CV at 15 and 30 min after the HH protocol. It is noteworthy that except for serum cortisol, all other changes in serum and muscular steroid hormone concentrations were within 2x CV.

We aimed to determine whether the RE-induced increase in serum steroids was accompanied by elevated steroid concentrations in skeletal muscle tissue. Our main finding was that the increase in cortisol serum concentration did not align with the changes in the muscle steroid concentrations. In the present study, only serum DHEA and cortisol changed above 1xCV (29%) and 2xCV (126%) after the HH protocol, respectively, with no changes in any muscle steroid concentrations. Additionally, RE-related contractions of the deltoid muscle were not sufficient to change steroid concentrations in the muscle, although the volume and intensity of the LH protocol were high for the mentioned muscle (10 sets of 8 to 12 RM - 3 min rest interval between sets). The lack of change in muscle steroid after the LH and HH protocols suggests that individuals with previous experience in RE were desensitized to the exercise stimuli (serum and local) (11); it is also possible that the single postexercise timepoint investigated in this study (45 min after RE) to assess muscle steroid hormones might be too early or too late to observe any such changes.

Muscle steroidogenesis capacity in resistance-trained athletes has been investigated previously. Vingren, Kraemer (19) showed no changes in the concentration of testosterone in skeletal muscle at 10 and 70 min after RE despite a robust acute RE-induced increase in serum total and free testosterone concentrations (2). However, the large interval between these two time points and the lack of data for other steroids could explain their findings. Considering that in blood, RE-induced increases in steroid concentrations subside by 30 min after an exercise bout, hormones may have been absorbed by the target tissues or metabolized in the liver (22). As a result, the points (10 and 70 min after an RE bout) used by Vingren, Kraemer (19) may not best reflect changes in muscle steroid concentrations. We therefore assessed muscle steroid concentrations (i.e., testosterone, DHEA, DHT and cortisol) 45 min after the LH and HH protocols. Although we induced an increase in serum cortisol concentration, no changes were observed in muscle cortisol concentration or any other assessed steroid hormone. Additionally, our LH group allowed us to determine if muscle contractions per se (without increases in serum steroid concentrations) increase muscle steroid concentrations, as suggested by an in vitro experiment (23). Again, no changes were observed in any steroid concentration in skeletal muscle. Thus, additional timepoints should be examined to determine if RE can indeed change muscle steroid concentrations either by skeletal muscle uptake of these hormones or the intramuscular steroidogenesis process.

Although there was a lack of change to mean values (other than serum cortisol), we further scrutinized our data using individual values, which provided interesting observations. In Figure 3 , in the LH protocol, the subject represented by the inverted triangle presents serum hormone concentrations always inside the measurement error. With close attention to the steroid concentration in skeletal muscle in Figure 4 , the same subject is two times above the measurement error for testosterone, DHEA, and DHT and close to two times above the measurement error for cortisol. For this subject, since he did not present changes in serum hormone concentrations, it could suggest that a muscle contraction per se was able to modulate steroid hormone concentrations in skeletal muscle. In a similar direction considering the HH protocol, the subject represented by the star, also in Figure 3 , presents serum hormone concentrations always inside the measurement error except for cortisol. For this subject, hormone concentrations in the skeletal muscle in Figure 4 are two times above the measurement error for testosterone, DHEA, and DHT and close two times above the measurement error for cortisol. Remarkably, this subject did not present elevated steroid hormone concentrations in skeletal muscle ( Figure 4 ) in the LH protocol (except for DHEA - > 2x CV). Thus, the changes observed in testosterone, DHEA, DHT and cortisol with the HH protocol for this subject (represented by the star in Figure 3 ) could be due to altered serum cortisol concentrations ( Figure 3 ) in addition to muscle contraction-induced DHEA concentration changes in the LH protocol ( Figure 4 ).

Finally, the present findings should be interpreted taking into consideration the following factors: a) although the sample size is small in the present study (n=6), this is the first design presenting measurement errors (coefficient of variation and typical error) for serum and muscle steroid concentrations in humans; b) the methods to determine hormone concentrations should have very high sensitivity, as the muscle concentration of steroid hormones is extremely low. Thus, liquid chromatography/electrospray tandem mass spectrometry (LC–MS/MS) analysis has been suggested for these assessments (24). c) Possibly, we have lost the RE-induced increase in serum testosterone once 15 min after exercise, which can be too late to detect this increase (2). d) Considering that the RE protocols induced serum changes at least in cortisol, we cannot rule out the hypothesis that serum changes can affect muscle steroidogenesis, such as steroidogenic enzyme concentrations and activity. Thus, future works should investigate other muscle cell steroidogenesis-related alterations induced by cortisol. Additionally, serum changes in other steroids should be ensured to investigate steroid concentration and steroidogenesis in skeletal muscle. In conjunction, it seems necessary to conduct a time-course investigation of RE bout-induced changes in the concentration of muscle steroids and steroidogenesis. In conclusion, the serum steroid concentration increases (cortisol only) do not seem to be aligned with intramuscular steroid concentrations in previously resistance-trained men. Additionally, muscle contraction per se apparently cannot modulate steroid hormone concentrations in skeletal muscle.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of human research of the School of Physical Education and Sport (University of São Paulo, Brazil). The patients/participants provided their written informed consent to participate in this study.

FCV, JV, GT, MC, CAL and CU contributed to conception and design of the study. FCV, GT, MC, ML, and FD collected the data. LR proceeded the muscle biopsies. FCV, GT, TC, and PB proceeded the biochemical analises. FCV organized the database. FCV and CU performed the statistical analysis. FCV wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.