Share this article on:

Rat liver lysosomal and mitochondrial activities are modified by anabolic-androgenic steroids


Medicine & Science in Sports & Exercise: February 1999 - Volume 31 - Issue 2 - p 243-250
Basic Sciences: Original Investigations

Rat liver lysosomal and mitochondrial activities are modified by anabolic-androgenic steroids. Med. Sci. Sports Exerc., Vol. 31, No. 2, pp. 243-250, 1999.

Purpose: The aim of this study was to examine the separate and combined effects of an 8-wk treatment with high doses of 17α-alkylated anabolic-androgenic steroids (AAS) and exercise training on selected lysosomal and mitochondrial enzyme activities in rat liver.

Methods: Sedentary and treadmill-trained (25 m·min−1, 45 min·d−1, 5 d·wk−1) male rats were treated with fluoxymesterone, methylandrostanolone, or stanozolol (2 mg·kg body weight−1, 5 d·wk−1) for 8 wk.

Results: Acid phosphatase, arylsulfatase, β-glucuronidase, and β-galactosidase activities were increased in liver homogenates of sedentary and trained AAS-treated rats. The mitochondrial respiratory chain activities rotenone-sensitive NADH-cytochrome c reductase (NCCR), succinate cytochrome c reductase (SCCR), and cytochrome oxidase (COX) showed a significant decrease in steroid-administered rats, whereas citrate synthase (CS), a matrix enzyme, exhibited no changes in activity, pointing to a selective effect of AAS on mitochondrial membrane complexes. In vitro studies in mitochondrial fractions isolated from the liver of control rats showed that COX and CS activities were insensitive to the AAS, whereas NCCR and SCCR activities were partly inhibited. On the other hand, the mean values of serum parameters related to hepatic function were within normal ranges in all the experimental groups of animals.

Conclusions: The present data show that 8-wk ingestion of three different anabolic-androgenic steroids, either with or without concurrent exercise training, affects lysosomal hydrolases and mitochondrial respiratory chain electron transport in rat liver without modifying classical serum indicators of hepatic function.

Department of Biochemistry and Molecular Biology I, Faculties of Chemistry and Biology, Complutense University, Madrid, SPAIN

Submitted for publication July 1997.

Accepted for publication May 1998.

We thank Zambon S. A. (Barcelona, Spain) for the gift of stanozolol. This work was supported by grants SAF93-0287 from the CICYT and PM95-0190 from the DIGICYT (Spain).

Address for correspondence: Dra. Alicia Megías, Dpto. Bioquímica y Biología Molecular I, Facultad de Biología, Universidad Complutense, 28040-Madrid, Spain. E-mail:

At present, the abuse of anabolic-androgenic steroids (AAS) to enhance physical performance and appearance is widespread in competitive and noncompetitive sports communities (21,40). AAS users self-administer high doses, for long periods and often combining two or more AAS. However, long-term consequences of these patterns of abuse are not known (40).

Among the many reported adverse effects caused by AAS are various types of liver disorders, all with a low incidence (15). Hepatic complications, including cholestasis, peliosis hepatis, hyperplasia, and tumors, have been mainly associated with the use of the orally active 17α-alkylated anabolic-androgenic steroids, synthetic derivatives of testosterone developed to slow degradation and to prolong their action (16,32). Although most reports of serious liver disturbances are connected with therapeutic use of AAS, there are several clinical case reports of disturbances among athletes and body-builders abusing AAS (5,10,15). Of primary concern is the fact that when hepatic function is monitored in healthy athletes, usually only serum indicators are investigated, and liver disorders do not always result in blood test abnormalities (2). In studies performed with AAS-users, liver serum parameters showed no changes or slight elevations that often revert to normal levels after discontinuing the drug (1,14,15).

It is well-known that androgens modulate several hepatic functions, such as metabolism of lipoproteins, steroids, and drugs (23). Anabolic-androgenic steroids exhibit some characteristics and actions of endogenous androgens, but, in the high doses taken by athletes, they may be regarded as a xenobiotic load for the liver. Unlike testosterone and testosterone esters, 17α-alkylated AAS have been reported to interfere with the excretory function of the liver in human patients and in animal models (16,26,32) and to be directly toxic to isolated hepatocytes (38). After prolonged administration of AAS, in both clinical and experimental studies, electron microscopic examination of liver tissue reveals ultrastructural alterations in the canaliculi and degenerative changes in mitochondria and lysosomes (11,16). Furthermore, in eugonadal male rats, 8-wk ingestion of high doses of AAS was shown to cause an increase in the activity of the outer carnitine palmitoyltransferase of hepatic mitochondria (12) and a decrease in some components of the liver microsomal drug-metabolizing system (29).

On the other hand, the liver is the major organ responsible for the metabolism of drugs and steroids in mammals and, because of its central position in the energy metabolism, it must also cope with the increased demands of the working muscle during exercise. In addition, exercise influences a large number of physiological factors, such as splanchnic blood flow and hormonal balance, that may affect the pharmacokinetics of drugs and, thus, alter their efficiency and/or their toxicity (33). Investigations of effects of chronic exercise on the biotransformation and disposition of xenobiotics, in humans and rodents, are scarce as well as inconclusive, probably as a consequence of differences in exercise protocols or in the chemical structures of the drugs (33,37). Therefore, it is possible that the concurrence of exercise training and AAS ingestion may modify the potential adverse hepatic effects of these compounds.

Considering the limited information available on the consequences of anabolic-androgenic steroid abuse on liver function and the reported alterations in hepatocyte organelles, we designed a study to investigate the separate and combined effects of an 8-wk treatment with high doses of 17α-alkylated AAS and an exercise program on rat liver lysosomes and mitochondria. More specifically, serum parameters related to hepatic function were assessed and enzyme activities representative of lysosomes, and mitochondrial oxidative metabolism and respiratory chain were determined in liver homogenates.

Back to Top | Article Outline


Anabolic-Androgenic Steroids

Fluoxymesterone (9α-fluoro-11β,17β-dihydroxy-17α-methylandrost-4-en-3-one) and methylandrostanolone (17β-hydroxy-17α-methyl-5α-androstan-3-one) were obtained from Sigma Chemical Co. (St Louis, MO). Stanozolol (17β-hydroxy-17α-methyl-5α-androst-2-eno[3,2-c]-pyrazole) was provided by Zambon S. A. (Barcelona, Spain). Methylandrostanolone, a derivative of 5α-dihydrotestosterone, is more androgenic than classic anabolic steroids, such as fluoxymesterone and stanozolol (27).

Back to Top | Article Outline

Training Program and Steroid Treatment

Male Wistar rats (initial body weight: 115 ± 5 g) were obtained from Charles River Laboratories (Barcelona, Spain). Animals had free access to laboratory chow and tap water. They were maintained on a 12:12-h light-dark cycle and housed in an animal room where temperature (22-24°C) and humidity levels were controlled. Rat care and handling and all experimental procedures employed were in accordance with the policy statement of the American College of Sports Medicine on research with experimental animals.

Forty-eight rats were randomly assigned to one of eight experimental groups, resultant from a two factor (2 × 4) design (factor 1: exercise training; factor 2: AAS treatment). Initially, animals were equally divided between a sedentary group (cage-confined for 12 wk) and an exercise-trained group. The rats of the trained group were exercised by running on a motor-driven treadmill 5 d·wk−1 for 12 wk. Exercise was always performed midway through the dark period of the light-dark cycle. During the first 4 wk, the speed and duration of the daily exercise sessions were progressively increased until the rats were capable of running continuously for 45 min at 25 m·min−1, treadmill speed corresponding to a relative workload of ≈65% O2max (19). At the beginning of the 5th training week, when maximal exercise intensity was reached, the sedentary and trained groups were arbitrarily subdivided into four groups: control, fluoxymesterone-treated, methylandrostanolone-treated, and stanozolol-treated. The animals selected for anabolic-androgenic steroid treatment received by gastric intubation 2 mg steroid·kg body weight−1, as a homogeneous suspension in 1 mL of water, 5 d·wk−1 for 8 wk. The high dosage of anabolizing androgens was chosen in an attempt to simulate the massive doses of AAS used in athletics.

After completion of the 12-wk exercise program, rats were not exercised for 36-44 h and received the last steroid dose 14-18 h before they were sacrificed (between 8:00 and 12:00 a.m.). Animals were fasted overnight, weighed, and under ether anesthesia, killed by decapitation. Blood was collected and the liver was rapidly exposed, excised, washed with cold saline, and immediately frozen in liquid nitrogen and stored at −80°C until examination.

Back to Top | Article Outline

Serum Analyses

Serum was obtained by centrifugation at 3000 × g for 15 min and stored at −40°C. Total and direct bilirubin were determined in fresh serum aliquots by a colorimetric procedure (Ames, Miles Italiana S.p.A., Milano, Italy). The activities of the serum enzymes: aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), and γ-glutamyltransferase (γ-GT) were analyzed by kinetic assays (Boehringer Biochemica GmbH, Mannheim, Germany). Cholesterol was measured by an enzymatic method (CHOD-PAD) using a commercial kit from Boehringer (Boehringer Biochemica GmbH, Mannheim, Germany). High density lipoprotein (HDL)-cholesterol (HDL-C) concentration was determined after precipitation of other lipoproteins with magnesium chloride-phosphotungstic acid (Boehringer reagent).

Back to Top | Article Outline

Preparation of Liver Homogenates and Mitochondrial Fractions

Procedures were carried out at 0-4°C. Livers were finely chopped with scissors and homogenized with a Potter-Elvehjem loose-fitting glass-Teflon homogenizer in 5 volumes of 10 mM Tris-HCl, pH 7.4, containing 0.25 M sucrose, 1 mM EDTA, and 0.3 mM phenylmethylsulfonyl fluoride. The homogenate was filtered through four layers of cheesecloth and aliquots were stored at −80°C until assay.

The mitochondrial fraction was obtained from the liver of control rats by the differential centrifugation procedure reported by Guzman et al. (12). After homogenization of the minced tissue as described above, the homogenate was centrifuged at 3000 × g for 1 min. The supernatant was decanted, filtered through cheesecloth, and centrifuged at 20000 × g for 1 min. The resultant pellet was washed with homogenization buffer and recentrifuged at 20000 × g for 1 min. The pelleted mitochondrial fraction was resuspended in 10 mM Tris-HCl, pH 7.4, containing 0.3 M sucrose and 1 mM EDTA. Protein concentrations were determined by the method of Lowry et al. (20), using bovine serum albumin as a standard.

Back to Top | Article Outline

Liver Enzyme Assays

Lysosomal hydrolases. Enzymes activities were determined fluorometrically using the appropriate 4-methylumbelliferyl (4-MU) substrate (Sigma), as described by Morgan et al. (24) with modifications. All the assays were carried out at 37°C in a final volume of 1 mL of 0.2 M sodium acetate buffer, pH 4.0 or 5.0, containing 0.33 M sucrose, 0.1% (w/v) Triton X-100 and a suitable amount of liver homogenate. The reaction was initiated as the substrate was added and was stopped after 10 min by the addition of 0.5 mL of 5% (w/v) trichloroacetic acid. Then 4 mL of 0.5 M glycine-sodium hydroxide buffer, pH 10.5, were added, and the mixture was centrifuged for 10 min. The fluorescence of 4-MU released during the incubation and present in the supernatant was measured with a Perkin-Elmer MPF-44 spectrofluorometer (excitation and emission wavelengths: 360 and 440 nm, respectively). For each assay, 4-MU in the appropriate buffer was used as standard.

The activity of the following lysosomal enzymes was measured: acid phosphatase (EC activity was determined using 0.2 mM 4-MU-phosphate in pH 5.0 buffer; arylsulfatase (EC activity was quantified using 0.5 mM 4-MU-sulfate in pH 5.0 buffer; β-glucuronidase (EC activity was determined using 0.5 mM 4-MU-β-D-glucuronide in pH 4.0 buffer; and β-galactosidase (EC activity was measured using 0.5 mM 4-MU-β-D-galactopyranoside in pH 5.0 buffer.

Mitochondrial enzymes. Citrate synthase (CS, EC activity was measured at 37°C by the method of Srere (34). The assay system contained in a total volume of 1 mL: 100 mM Tris-HCl, pH 8.1, 1 mM 5,5′-dithiobis-(2-nitrobenzoate), 0.3 mM acetyl-CoA, 0.5 mM oxalacetate and a suitable amount of protein. Additionally, 0.2% (w/v) Triton X-100 for homogenates or 0.02% (w/v) Triton X-100 for mitochondrial fractions was present in the assay mixture to permeabilize membranes. The reaction was started by addition of oxalacetate and the increase in absorbance was recorded at 412 nm (∊ = 13.6 mM−1·cm−1).

NADH-cytochrome c reductase (NCCR), succinate cytochrome c reductase (SCCR), and cytochrome c oxidase (COX, EC activities were measured by following at 550 nm either the reduction or the oxidation of cytochrome c (ε = 21.1 mM−1·cm−1).

NCCR activity was assayed, according to Sherrat et al. (31) with some modifications, at 25°C in 1 mL of 0.1 M potassium phosphate, pH 7.5, containing 0.1 mM NADH, 100 μM cytochrome c (III), 1 mM KCN, 2.5 mg·mL−1 defatted bovine serum albumin, and the adequate amount of protein. For mitochondrial membrane disruption, preparations were diluted 17.5-fold in hypotonic medium (10 mM potassium phosphate, pH 7.4) and subjected to three cycles of freeze-thawing. The reaction was started by addition of NADH. Rotenone-sensitive NCCR was the activity inhibited by 5 μM rotenone and was considered as due to Complex I plus III of the respiratory chain.

SCCR activity was assayed at 37°C by the method of Cooperstein et al. (6) with some modifications. The reaction mixture contained 40 mM potassium phosphate, pH 7.5, 17 mM sodium succinate, 100 μM cytochrome c (III), 1 mM KCN, 5 μM rotenone, and either homogenate or mitochondrial protein in a final volume of 1 mL. Because a variable proportion of complex II is deactivated because of the tight binding of the competitive inhibitor oxalacetate, to ensure that complex II was fully activated before assay, it was necessary to preincubate the protein with succinate for 5 min at 25°C before the addition of the remaining components for the assay. The reaction was initiated as cytochrome c was added. The activity was considered as due to complex II plus III of the respiratory chain.

COX activity was measured at 25°C in 1 mL of reaction buffer containing 20 mM potassium buffer, pH 7.0, 80 μM cytochrome c (II), and a suitable amount of protein. The reaction was initiated as cytochrome c was added. Cytochrome c (II) was prepared by addition of severalfold molar excess ascorbate to cytochrome c (III), and separation of reduced cytochrome c was performed by Sephadex G-25 chromatography. The cytochrome c (II) was stored in aliquots at −80°C, and its concentration was determined before each batch of assays.

All the enzymatic assays were demonstrated to be linear with time and with protein under the employed conditions.

Back to Top | Article Outline

Statistical Analysis

All samples were individually processed and measured in duplicate in the same assay. Results for each experimental group are reported as means ± SD. Data were analyzed by a two-way ANOVA to test for the two main effects (exercise training and AAS administration) and for the interaction between them, employing a standard computerized statistical program (Sigma, Horus Hardware S. A., Madrid, Spain). When a significant F-value was obtained, a Scheffé post-hoc analysis was performed to determined specific differences. A level of P < 0.05 was selected to indicate statistical significance.

Back to Top | Article Outline


Rat Body Weight, Liver Weight, and Protein Content in the Liver

As shown in Table 1, after 12 wk of experimental period, final body and liver weights were significantly lower in the trained groups in comparison with the sedentary groups, whereas no significant differences associated with anabolic-androgenic steroid treatment were detected. Because all animals gained weight steadily during the 12 wk, the endurance training program resulted in a lower gain in body weight for exercising rats, probably because of the inadequate compensation by food intake in male rats of the increased energy expenditure (13). As previously reported for some treadmill running programs (13), the observed decrease in liver weight of trained animals was not accompanied by a significant variation in liver weight to body weight ratio (data not shown). Likewise, the amount of protein in liver homogenate per gram wet tissue was not influenced by either exercise training or anabolic steroid administration (Table 1).



Back to Top | Article Outline

Serum Parameters

The determination of metabolites and enzyme activities in the serum can be of great value for the detection of liver alterations. In this respect, mean values of serum total and direct bilirubin, total- and HDL-cholesterol, as well as transaminases, alkaline phosphatase, and γ-glutamyltransferase activities were within the rat normal range (3) in all of the studied groups (Table 2). Neither the endurance training nor the administration of anabolic-androgenic steroids induced significant differences in the evaluated serum parameters. These results are consistent with those reported for rats treated with steroids (3) and for power athletes who self-administered very high doses of anabolic steroids (1,18). Nevertheless, it must be pointed out that among the most significant adverse effects of AAS abuse in athletes is a disruption in the balance of serum cholesterol fractions with a striking reduction of HDL-cholesterol levels (21). The lack of significant changes in HDL-cholesterol levels of anabolic androgenic steroid-treated rats may be a consequence of the well-known differences in serum lipid metabolism between humans and rats or might be attributable to an insufficient period of treatment or to the use of insufficiently high AAS doses.



Back to Top | Article Outline

Liver Lysosomal Hydrolase Activities

Lysosomes are cytoplasmic organelles delimited by a single membrane. Therefore, lysosomal enzymes show structure-linked latency, and their activity is only fully expressed when the lysosomal membrane is disrupted. Comparison among several established methods of membrane disruption, such as several cycles of freeze-thawing in isotonic media or addition of detergent (Triton X-100, Lubrol PX or CHAPS), allowed us to select the addition of the nonionic detergent Triton X-100, as the most reproducible and easiest method to obtain maximum activities with liver homogenates (not shown).

The measured activities of two esterases, acid phosphatase and arylsulfatase, and of two glycosidases, β-glucuronidase and β-galactosidase, in hepatic homogenates from the animals of the different experimental groups are presented in Figure 1. The four lysosomal hydrolase activities exhibited a very similar behavior. In no case was a main effect of exercise training observed. However, each one of the activities determined showed a significant main effect for the factor AAS administration (two-way ANOVA, P < 0.01), and the activity measured in the control group was significantly lower than that determined in the AAS-treated groups (P < 0.01). Although both in sedentary and in trained rats there was a trend toward lysosomal enzyme activities being greater in the 17α-alkylated AAS-receiving animals than in the control rats, the effect was more marked in the exercising animals.

Figure 1

Figure 1

Back to Top | Article Outline

Mitochondrial Enzyme Activities

Four enzyme activities were used as markers of the liver mitochondrial oxidative capacity. CS, located in the matrix space, is a key enzyme in the tricarboxylic acid cycle, whereas rotenone-sensitive NCCR, SCCR, and COX, embedded in the inner mitochondrial membrane, measure segments of the mitochondrial electron transport chain. In view of the subcellular location of the selected enzyme activities, disruption of mitochondrial membranes is necessary to ensure free access of substrates and inhibitors to the enzymes. Therefore, we compared the effects on the mitochondrial activities of different standard procedures to disrupt membranes, such as freeze-thawing in hypotonic or isotonic media, hypotonic shock, or addition of detergents (data not shown). For rotenone-sensitive NCCR, optimal activity was obtained by freeze-thawing the liver preparations in hypotonic potassium phosphate buffer; under these experimental conditions, 20-30% of total NADH-cytochrome c reductase activity was inhibited by 5 μM rotenone, because of the high content of liver in NADH-cytochrome b5 reductase, an enzyme that contributes to rotenone-insensitive NCCR activity. Maximum citrate synthase activity was obtained by addition of Triton X-100 to the assay medium, whereas for measurement of SCCR and COX activities, no permeabilizing treatment was necessary.

Figure 2 shows the measured mitochondrial enzyme activities for all the study groups. No differences were detected between sedentary and trained animals for any enzyme activity. A significant main effect was observed for anabolic-androgenic steroid treatment (two-way ANOVA, P < 0.01) on the three activities involved in the respiratory chain, but citrate synthase activity was not significantly affected by the ingestion of steroids. The oral administration of stanozolol for 8 wk induced significant decreases (P < 0.05) in NCCR, SCCR, and COX activities of rat liver homogenates, whereas the treatment with methylandrostanolone elicited decreases in NCCR (P < 0.05) and COX (P < 0.01) activities, and the ingestion of fluoxymesterone only caused a significant decrease in NCCR activity (P < 0.05). The combination of exercise training with anabolic-androgenic steroid treatment elicited no greater changes.

Figure 2

Figure 2

To test the possibility that the observed decreases in enzyme activities were caused by a direct action of anabolic-androgenic steroids on mitochondria, the effects of 17α-alkylated AAS were investigated in vitro, so that the problems associated with drug administration and metabolism and with the potential modification of other organic, hepatic or subcellular systems, could be avoided. The limited solubility of methylandrostanolone in aqueous medium prevented its in vitro use. The effects of increasing concentrations of stanozolol and fluoxymesterone on enzyme activities of liver mitochondria isolated from control rats are shown in Figure 3. The two 17α-alkylated AAS inhibited NCCR and SCCR activities, but they did not affect COX and CS activities. NCCR activity was more sensitive to steroids than SCCR activity and for both activities the inhibitory effect of stanozolol was more marked than that of fluoxymesterone. It should be noted that although the employed steroid concentrations were in the micromolar range, the maximum inhibition observed was 35% for NCCR in the presence of 100-μM stanozolol.

Figure 3

Figure 3

Back to Top | Article Outline


This investigation was designed to study the effects of high doses of anabolic-androgenic steroids on liver lysosomal and mitochondrial activities, controlling the potential influence of simultaneous exercise training. The 12 wk of moderate-endurance training induced the well-characterized increase in skeletal muscle oxidative capacity, assessed by the increase in the activities of mitochondrial enzymes, such as carnitine palmitoyltransferase I (73% in soleus and 48% in extensor digitorum longus) and succinate dehydrogenase (38% and 48%, respectively) (12,28). In the present study, the training program was shown to cause a decrease in rat body weight and liver weight, but no effects were detected on any other of the studied parameters. Analysis of our results indicated that exercise training played no main role, either by itself or by influencing the observed anabolic-androgenic effects, on the determined hepatic enzyme activities.

Liver enzyme activities were measured in total homogenates rather than in isolated lysosomal or mitochondrial fractions, because the introduction of purification steps could yield preparations likely representative of only part of these systems due to the intrinsic heterogeneity of lysosomal and mitochondrial populations. In addition, microscopic studies of the liver from humans and animals after anabolic-androgenic steroid treatment consistently show structural and ultrastructural changes (11,16,32) that may cause modifications in the liver subcellular fractionation pattern, leading to an erroneous determination of the total enzyme activities present in the liver.

To our knowledge, except for ultrastructural studies (11,16), there have been no previous attempts to explore the response of liver lysosomes to anabolic-androgenic steroids administration. Nevertheless, testosterone has been shown to increase the specific and total activities of lysosomal hydrolases in several rodent tissues (17,39), and methandrostenolone, in combination with exercise training, induced an increase in lysosomal enzyme activities in the right ventricular wall of dog heart (36).

The finding that the activity of lysosomal enzymes was increased in homogenates of liver from rats treated with anabolic steroids compared with those from control animals is consistent with our previous ultrastructural study that showed an increase in the number of lysosomes dispersed throughout the cytoplasm of hepatocytes from the treated rats (11). However, taking into account the prominent changes in the number, electron density, morphology, and size of lysosomes observed by microscopic examination of the hepatic tissue (11), a more dramatic alteration in the activity of the lysosomal hydrolases might have been expected. This apparent inconsistency can be explained by several factors. First, in the ultrastructural study, only hepatocytes were analyzed, and nonparenchymal cells, such as Kupffer or endothelial cells, containing numerous lysosomes are known to contribute up to 20% to the lysosomal enzyme activities measured in liver homogenates (9). Second, release of hepatocyte lysosomal content into bile by exocytosis is an important pathway for the elimination of foreign materials from the liver and for the turnover of lysosomal enzymes (25). Thus, the modest augment of lysosomal hydrolase activities detected in liver homogenates could be accompanied by a substantial increase in biliary lysosomal enzyme excretion, as observed for the lysosomotropic agent chloroquine (30). An increased activity of lysosomal hydrolases in liver (and plasma) appears to be more related to cholestasis than to other liver pathologies, even as serious as cirrhosis (9). In our experimental model, conventional serum markers of cholestasis such as bilirubin levels and alkaline phosphatase and γ-glutamyltransferase activities were not affected either by anabolic-androgenic steroid administration or by exercise training, pointing to the absence of biliary obstruction. However, the possibility that higher doses of 17α-alkylated AAS or longer periods of treatment may induce a more noticeable increase in liver lysosomal activities should not be dismissed, because the association between anabolic-androgenic steroids with an alkyl group in the C-17 position and development of a cholestatic pattern of hepatic injury has been convincingly proved (16).

Concerning mitochondrial enzymes, a decrease in the activities of the respiratory chain complexes was detected in the liver of the treated animals irrespective of their being exercised or not. In contrast, the activity of citrate synthase, a matrix enzyme, was not affected. This points to a selective effect of the anabolic-androgenic steroids on the inner membrane enzymes rather than a generalized effect on the mitochondrial enzymes.

Changes in the level of expression of the respiratory chain protein complexes could account for the observed effect. In this regard, it has been demonstrated that steroid hormones, as well as thyroid hormones, regulate the expression of mitochondrial genes encoding for subunits of the respiratory enzymes in several tissues, including liver (7,8). Because anabolic-androgenic steroids have been reported to bind to the androgen receptors (27), it would be conceivable that the action of these compounds on the mitochondrial complexes were mediated by a mechanism similar to that of endogenous hormones. However, the possibility of down-regulation by AAS appears to be unlikely because reduction in expression of mitochondrial genes, including cytochrome c oxidase subunits I, II and III and NADH dehydrogenase subunit 5, has been observed only after 4-wk castration, with partial to full recovery to precastration levels upon testosterone replacement (7). Although we have found that AAS treatment reduced circulating testosterone levels in rats to about 50% of control values (data not shown), it can be expected that the remaining testosterone along with the administrated AAS, would be able to maintain the androgenic status. In addition, mitochondrial gene expression in liver has been shown to exhibit scarce androgen responsiveness as compared with other tissues (7). Thus, the effects on liver mitochondrial membrane activities observed in our experimental model are unlikely to be related to changes in mitochondrial gene expression.

Another factor that may induce the decrease in respiratory chain enzyme activities could be the unfavorable modification of the lipid environment of these complexes. In this respect, changes in the phospholipid composition, both in classes and fatty acid moieties, and induction of lipid peroxidation have been reported in liver mitochondrial membranes from rats treated with the adrenal androgen dehydroepiandrosterone (22,35). On the other hand, anabolic-androgenic steroids themselves could affect the mitochondrial membrane due to their hydrophobic nature. Results from in vitro studies appear to support this idea. The fact that COX activity was not modified, whereas NCCR and SCCR activities were inhibited, indicates that electron transport is disturbed by stanozolol and fluoxymesterone before complex IV. Because NCCR and SCCR activities measure complexes I plus III and complexes II plus III, respectively, and the degree of inhibition is similar in both cases, it seems likely that anabolic-androgenic steroids inhibit either complex III or the interaction between complexes I-III and II-III, probably involving ubiquinone. The extent of inhibition of NCCR and SCCR activities by 17α-alkylated AAS appears to be related to steroid structure, the more hydrophobic stanozolol being more potent than fluoxymesterone.

Although in vitro studies point to a direct effect of anabolic-androgenic steroids on mitochondrial membranes, it is unclear whether a similar mechanism is occurring in vivo. The effective concentration of steroids in hepatic cells in vivo is not known, and it could be lower than those used in vitro. In addition, hepatic metabolism of AAS must be taken into account because of the possibility that some metabolite(s) of the administered steroids is actually responsible for the observed effects. On the other hand, it should be noted that COX activity, which was unaffected by steroids in vitro, was decreased in liver homogenates, suggesting a more profound alteration of mitochondrial membrane.

In our experimental model, ultrastructural abnormalities of hepatocyte mitochondria and lysosomes were detected (11) in addition to the above-described changes in lysosomal and mitochondrial inner membrane activities. These alterations suggest an acceleration of lysosome-mediated degradation and a perturbation of energetic metabolism. The in vivo consequences of these changes are difficult to ascertain but they could affect adversely the functional capacity of the liver. On the other hand, AAS treatment has been shown to induce modifications of the rat drug-metabolizing system (29) and of the activity of the outer carnitine palmitoyltransferase, the enzyme controlling the rate of long-chain fatty acid oxidation (12). Likewise, 17α-alkylated steroids have been shown to be directly toxic to rat hepatocytes in culture, whereas nonalkylated steroids had no effects at similar doses (38). Therefore, the available experimental evidences obtained in controlled studies on rodents indicate that 17α-alkylated AAS can induce liver injury, although the underlying molecular and cellular mechanisms are largely unknown. Furthermore, a 6-month treatment of male mice with a combination of testosterone, testosterone cypionate and two 17α-alkylated steroids resulted in serious liver damage, along with a dramatic decrease of the life span of the treated animals (4).

In humans alterations in liver are very difficult to characterize, because the pharmacological interventions and the invasive studies, for obvious ethical reasons, are limited. Therefore, most of the published investigations concerning the hepatic side-effects of AAS are case histories, and controlled studies of liver function and morphology in humans treated with AAS only exist for patients already suffering from liver disease (32). The use of an animal model avoids some of these problems, but other troubles arise, including the difficulty in providing controlled experimental conditions similar to those found in man, the necessity of extrapolation of doses, frequencies and durations, the existence of morphologic, physiologic, and metabolic species differences, etc. Thus, in the present study, a number of limitations must be taken into account: i) the more typical users of anabolic-androgenic steroids are involved in strength sports, and the rats have been subjected to an endurance training; ii) the steroids were administered to the rats over a span of 40 d (5 d a week, for 8 wk), and it is not clearly established how this period should be extrapolated to human beings; iii) the number of animals in each experimental group is small (N = 6), which could lead to the nondetection of actually existent differences (Type II statistical error); and iv) AAS abusers often take different steroids simultaneously, and each animal ingested only a single AAS. Therefore, the degree to which the above results are extrapolable to humans can be questionable, but the potential adverse consequences for the liver of exposure for long periods to high doses of 17α-alkylated anabolic-androgenic steroids must be seriously considered.

Back to Top | Article Outline


1. Alen, M. Androgenic steroid effects on liver and red cells. Br. J. Sports Med. 19:15-20, 1985.
2. American College of Sports Medicine. Position stand on the use of anabolic-androgenic steroids in sports. Med. Sci. Sports Exerc. 19:534-539, 1987.
3. Bauman, D. H., J. T. Richerson, and A. L. Britt. A comparison of body and organ weights, physiological parameters, and pathologic changes in target organs of rats given combinations of exercise, anabolic hormone, and protein supplementation. Am. J. Sports Med. 16:397-402, 1988.
4. Bronson, F. H., and C. M. Matherne. Exposure to anabolic-androgenic steroids shortens life span of male mice. Med. Sci. Sports Exerc. 29:615-619, 1997.
5. Cabasso, A. Peliosis hepatis in a young adult bodybuilder. Med. Sci. Sports Exerc. 26:2-4, 1994.
6. Cooperstein, S. J., A. Lazarow, and N. J. Kurfess. A microspectrophotometric method for the determination of succinic dehydrogenase. J. Biol. Chem. 186:129-139, 1950.
7. Cornwall, G. A., M.-C. Orgebin-Crist, and S. R. Hann. Differential expression of the mouse mitochondrial genes and the mitochondrial RNA-processing endoribonuclease RNA by androgens. Mol. Endocrinol. 6:1032-1042, 1992.
8. Demonacos, C. V., N. Karayanni, E. Hatzogluo, C. Tsiriyiotis, D. A. Spandidos, and E. Serekis. Mitochondrial genes as sites of primary action of steroid hormones. Steroids 61:226-232, 1996.
9. Dufour, J.-F., A. Zimmermann, and J. Reichen. Increased hepatic lysosomal activity in biliary cirrhosis originates from hepatocytes rather than from macrophages. J. Hepatol. 20:524-530, 1994.
10. Forbes, G. M., B. A. Bramston, and B. J. Collins. Anabolic steroid hepatotoxicity: lessons to be learnt? Aust. N. Z. J. Med. 23:309-310, 1993.
11. Gragera, R., A. Saborido, F. Molano, L. Jimenez, E. Muñiz, and A. Megias. Ultrastructural changes induced by anabolic steroids in liver of trained rats. Histol. Histopath. 8:449-455, 1993.
12. Guzman, M., A. Saborido, J. Castro, F. Molano, and A. Megias. Treatment with anabolic steroids increases the activity of the mitochondrial outer carnitine palmitoyltransferase in rat liver and fast-twitch muscle. Biochem. Pharmacol. 41:833-835, 1991.
13. Harpur, R. P. The rat as a model for physical fitness studies. Comp. Biochem. Physiol. 66A:553-574, 1980.
14. Hartgens, F., H. Kuipers, J. A. Wijnen, and H. A. Keizer. Body composition, cardiovascular risk factors and liver function in long-term androgenic-anabolic steroids using bodybuilders three months after drug withdrawals. Int. J. Sports Med. 17:429-433, 1996.
15. Hickson, R. C., K. L. Ball and M. T. Falduto. Adverse effects of anabolic steroids. Med. Toxicol. Adverse Drug Exp. 4:254-271, 1989.
16. Ishak, K. G., and H. J. Zimmerman. Hepatotoxic effects of the anabolic/androgenic steroids. Semin. Liver Dis. 7:230-236, 1987.
17. Koenig, H., A. Goldstone, and C. Y. Lu. Testosterone-mediated sexual dimorphism of the rodent heart: Ventricular lysosomes, mitochondria, and cell growth are modulated by androgens. Circ. Res. 50:782-787, 1982.
18. Kuipers, H., J. A. G. Wijnen, F. Hartgens, and S. M. M. Willems. Influence of anabolic steroids on body composition, blood pressure, lipid profile and liver functions in body builders. Int. J. Sports Med. 12:413-418, 1991.
19. Lawler, J. M., S. K. Powers, J. Hammeren, and A. D. Martin. Oxygen cost of treadmill running in 24-month-old Fischer-344 rats. Med. Sci. Sports Exerc. 25:1259-1264, 1993.
20. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275, 1951.
21. Lukas, S. E. Current perspectives on anabolic-androgenic steroid abuse. Trends Pharmacol. Sci. 14:61-68, 1993.
22. Mohan, P. F., and M. P. Cleary. Short-term effects of dehydroepiandrosterone treatment in rats on mitochondrial respiration. J. Nutr. 121:240-250, 1991.
23. Mooradian, A. D., J. E. Morley, and S. G. Korenman. Biological actions of androgens. Endocr. Rev. 8:1-28, 1987.
24. Morgan, D. M. L., J. D. Pearson, and J. L. Gordon. Lysosomal hydrolases of human vascular cells: response to agonists of endothelial function. Biochim. Biophys. Acta 1010:184-190, 1989.
25. Nakano, A., D. L. Marks, P. S. Tietz, P. C. de Groen and N. F. Larusso. Quantitative importance of biliary excretion to the turnover of hepatic lysosomal enzymes. Hepatology 22:262-266, 1995.
26. Nishino, Y. Effects of androgens and related steroids on liver functions and enzymes. Pharmacol. Ther. B1:189-207, 1975.
27. Saartok, T., E. Dahlberg, and J.-A. Gustafsson. Relative binding affinity of anabolic-androgenic steroids: comparison of the binding to the androgen receptors in skeletal muscle and in prostate, as well as to sex hormone-binding globulin. Endocrinology 114:2100-2106, 1984.
28. Saborido, A., J. Vila, F. Molano, and A. Megias. Effect of anabolic steroids on mitochondria and sarcotubular system of skeletal muscle. J. Appl. Physiol. 70:1038-1043, 1991.
29. Saborido, A., F. Molano, and A. Megias. Effect of training and anabolic-androgenic steroids on drug metabolism in rat liver. Med. Sci. Sports Exerc. 25:815-822, 1993.
30. Sewell, R. B., S. S. Barham, and N. F. Larusso. Effect of chloroquine on the form and function of hepatocyte lysosomes. Morphological modification and physiologic alterations related to the biliary excretion of lipids and proteins. Gastroenterology 85:1146-1153, 1983.
31. Sherratt, H. S. A., N. J. Watmough, M. A. Johnson, and D. M. Turnbull. Methods for study of normal and abnormal skeletal muscle mitochondria. Methods Biochem. Anal. 33:243-335, 1988.
32. Søe, K. L., M. Søe, and C. Gluud. Liver pathology associated with the use of anabolic-androgenic steroids. Liver 12:73-79, 1992.
33. Somani, S. M., S. K. Gupta, S. Frank, and C. N. Corder. Effect of exercise on disposition and pharmacokinetics of drugs. Drug Dev. Res. 20:251-275, 1990.
34. Srere, P. A. Citrate synthase. Methods Enzymol. 13:3-5, 1969.
35. Swierczynski, J., and D. Mayer. Dehydroepiandrosterone-induced lipid peroxidation in rat liver mitochondria. J. Steroid Biochem. Mol. Biol. 58:599-603, 1996.
36. Takala, T. E. S., K. Kiviluoma, M. Kihlström, P. Rämö, and V. Vihko. Effects of physical training, methandione and their combination on the lysosomal hydrolytic activities in dog heart. Int. J. Sports Med. 13:52-55, 1992.
37. Watkins, J. B. III, S. T. Crawford, and R. A. Sanders. Chronic voluntary exercise may alter hepatobiliary clearance of endogenous and exogenous chemicals in rats. Drug Metab. Dispos. 22:537-543, 1994.
38. Welder, A. A., J. W. Robertson, and R. B. Melchert. Toxic effects of anabolic-androgenic steroids in primary rat hepatic cell cultures. J. Pharmacol. Toxicol. Methods 33:187-195, 1995.
39. Wilson, C. M., D. F. Kimberlin, J. E. Griffin, and J. D. Wilson. Specificity of androgen resistance in Mus caroli kidney. Biochem. Genet. 26:705-716, 1988.
40. Yesalis, C. E., and M. S. Bahrke. Anabolic-androgenic steroids: current issues. Sports Med. 19:326-340, 1995.


© 1999 Lippincott Williams & Wilkins, Inc.