| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Mental Health and Alcohol Research, National Public Health Institute (T.S., B.v.d.P., C.J.P.E.), FIN-00101 Helsinki, Finland; and Folkhälsan Research Center and Division of Clinical Chemistry, University of Helsinki (H.A., S.H.), FIN-00014 Helsinki, Finland
Address all correspondence and requests for reprints to: C. J. Peter Eriksson, Ph.D., Department of Mental Health and Alcohol Research, National Public Health Institute, POB 719, FIN-00101 Helsinki, Finland. E-mail: peter.eriksson{at}ktl.fi
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
After intake of alcohol, an acute elevation in plasma testosterone, a decline in androstenedione levels, and an elevation in the ratio of testosterone to androstenedione were observed in both groups. The effects lasted throughout the period of ethanol elimination and were abolished during pretreatment with 4-methylpyrazole (10–15 mg/kg, orally). The acute effects were higher in the group using oral contraceptives than in the nonusers. The testosterone effect in plasma was reflected in the free testosterone fraction. A decline in urinary androsterone and etiocholanolone levels, the principal catabolic products of androgens, was observed during alcohol intoxication.
In conclusion, the present acute effects on plasma and urinary steroid hormones seem to be explained by an inhibited catabolism mediated by the alcohol-induced change in the redox state in the liver. Our results suggests that the liver should be included as a major site in the acute endocrinological effects of alcohol on steroid hormones in women.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Heavy acute alcohol intake leads to decreased testosterone levels in healthy men (4, 5, 6). This effect has been attributed to inhibition at different sites along the hypothalamic-pituitary-gonadal axis (7, 8, 9), the pathway that is generally regarded as the biological regulator of sex steroid levels. In premenopausal women, however, the alcohol effect was surprisingly found to be the opposite of that in men, with an elevation in plasma testosterone as well as a decrease in plasma androstenedione levels, but no effect on dehydroepiandrosterone and dihydrotestosterone, during alcohol intake (10, 11). Several cross-sectional studies including moderate to heavy drinking women report a positive association between alcohol intake and plasma androgen levels (12, 13, 14, 15, 16). Elevations in estradiol and decreases in estrone levels during alcohol intake have been found among both premenopausal women (17, 18) as well as postmenopausal women receiving estrogen replacement therapy (19), suggesting a common mechanism behind the alcohol-related acute sex steroid changes.
The objective of the study was to characterize the acute androgen effects in premenopausal women and further elucidate the mechanism behind the effects using 4-methylpyrazole, an inhibitor of alcohol metabolism.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Substudy A included nine healthy Caucasian premenopausal female students (age, 23 ± 1 yr; body mass index, 21.2 ± 0.5 kg/m2) who regularly used oral contraceptives. Substudy B included 12 healthy Caucasian premenopausal female students (age, 25 ± 1 yr; body mass index, 21.9 ± 0.5 kg/m2) who regularly used oral contraceptives and 10 healthy Caucasian premenopausal female students who did not use contraceptives (nonusers; age, 21 ± 1 yr; body mass index, 21.5 ± 0.9 kg/m2). For subjects taking oral contraceptives, all reported regular use of oral contraceptives containing ethinyl estradiol and a progestin component, and they had a history of regular menstrual cycles. None used any medication other than the oral contraceptive. The menstrual cycles were synchronized, and they were provided with the same brand of an oral contraceptive containing 30 µg ethinyl estradiol and 75 µg gestoden. The experimental sessions were scheduled on day 14 of the menstrual cycle, i.e. the seventh day of taking the pill, during consecutive menstrual cycles. For subjects not using oral contraceptives, all had a history of regular menstrual cycles. The first experimental session was scheduled as close as possible to the midcycle phase (menstrual cycle phase reported at the first session was 16 ± 1 days), and the following 3 sessions were arranged every 4 weeks during 3 consecutive menstrual cycles (menstrual cycle phases reported at sessions were 16 ± 1, 15 ± 2 and 14 ± 3 days, respectively). None of the participants reported a typical alcohol consumption of more than 14 standard drinks of alcohol (1 drink = 12 g ethanol) per week, and they were all classified as light drinkers. No alcohol was allowed for 1 week preceding the experimental sessions.
Both substudies A and B were conducted in accordance with the guidelines proposed in the Declaration of Helsinki. They were both approved by an ethical committee, and substudy B, in addition, was approved by the Finnish National Agency for Medicines. Participation was confirmed by obtaining a signed informed consent.
Study design
The study design was a controlled interventional study with a
cross-over design, and the main outcome was the steroid level. In
substudy B, each subject participated in four different experimental
sessions (placebo plus alcohol, 4-methylpyrazole plus alcohol,
4-methylpyrazole plus placebo, and placebo plus placebo). Subjects
received the different treatments in random order, and they were
allocated to groups to keep the different treatments equally
represented at each experimental session. 4-Methylpyrazole (free base
liquid from Sigma-Aldrich Corp., St. Louis, MO; the active
solution contained 0.80 g 4-methylpyrazole, 12.0 g liquorice
extract, 7.0 g distilled water, and 4 drops of anise extract;
final dose, 10–15 mg/kg) or placebo (13 g liquorice extract, 7.0
g distilled water, and 1 drop of anise extract) was given orally in a
double blind manner. The success of blinding the subject was confirmed
by asking "did you get 4-methylpyrazole or placebo?" 1 h after
intake at each session. The proportions of correct answers were 0.46,
0.66, 0.63, and 0.60 during the four different sessions and did not
statistically differ from 0.5 (P > 0.05 for each, by
2 test).
In substudy A, the alcohol dose was 0.4 g/kg, orally (8% wt/vol in lingonberry juice), and in substudy B, the alcohol dose was 0.5 g/kg, orally (10% wt/vol), corresponding to two or three standard drinks. The placebo drink contained an equal amount of juice only. Drinking time was 15 min, and subjects remained seated throughout the experiment. A blood sample (10 mL), collected into tubes containing 22.5 mg sodium fluoride and 22.5 mg potassium oxalate as anticoagulants, from an iv catheter in the cubital vein was taken without tourniquet before and at different time points after the various treatments. Subjects were asked to abstain from heavy physical exercise and from intake of food for 4 h before blood sampling. In substudy A, a spot urine sample was collected in vials containing sodium azide (final concentration, 1.0 g/L) before intake of the drink and 4 h later. All experimental sessions were performed at the laboratory starting at 1600 h.
Analytical procedures
Measurements of hormone and ethanol were made in plasma and urine samples stored at -70 C until determinations. Ethanol levels were determined by headspace gas chromatography (Sigma 2000, Perkin-Elmer Corp., Norwalk, CT). The intra- and interassay coefficients of variation were 4.0% and 5.1% at the level of 1.5 mmol/L (n = 10), and the detection limit was 0.01 mmol/L. Testosterone (within-assay variability, 6.6%; between-assay variability, 7.0% at the level of 0.96 nmol/L; n = 10; detection limit, 0.1 nmol/L), free testosterone (within-assay variability, 4.3%; between-assay variability, 5.5% at the level of 4.6 pmol/L (n = 10), detection limit 0.5 pmol/L), and androstenedione (within-assay variability, 8.5%; between-assay variability, 9.8% at the level of 5.3 nmol/L; n = 10; detection limit, 0.14 nmol/L) levels were determined by standard RIA reagent sets [Orion Diagnostica (Helsinki, Finland) for testosterone; Diagnostic Products (Los Angeles, CA) for free testosterone and androstenedione]. To check for possible changes in hormone levels in vitro caused by alcohol as well as for possible interactions of alcohol with hormone assays, ethanol was added to fresh blood from 10 nonusers and 6 oral contraceptive users 18- to 24-yr-old healthy female subjects to a final concentration of 10 mmol/L. No significant effects on androstenedione and total testosterone were observed.
Androsterone (5
-androstan-3-ol-17-one) and etiocholanolone
(5ß-androstan-3-ol-17-one) in urine were determined by gas
chromatography [GLC 8000 Top, CE Instruments (Milan, Italy);
column SGE 25QC2/BP1 0.25, flame ionization detector] as previously
described (20). The specificity of the peaks in one sample
from each subject was confirmed with gas chromatography-mass
spectrometry (GLC 8000, column 12QC2/BP1 0.25, detector, Fisons MD
1000; Milan, Italy). The intraassay coefficients of variation
were 6.8% and 6.8%, and the interassay coefficients of variation were
6.5% and 7.6% for androsterone and etiocholanolone,
respectively.
The urinary creatinine levels were determined with the VITROS 250 Chemistry System (Johnson & Johnson). The intra- and interassay coefficients of variation were less than 5.0% at the level of 440 µmol/L (n = 7), and the detection limit in the sample was 84 µmol/L.
Statistical methods
Results are reported as the mean ± SEM if not otherwise specified. The urinary steroid to creatinine ratio was used to correct for dilution. Statistical significance was tested using ANOVA for repeated measures [drug (4-methylpyrazole/placebo), drink (alcohol/placebo), time, and experiment as within-group factors; oral contraceptive status as a between-group factor], followed by paired t test or Wilcoxon matched pairs test in the case of nonnormal distribution of data. Data were analyzed using SPSS (version 10.0, SPSS, Inc., Chicago, IL) and Prism (version 2.0, GraphPad Software, Inc., San Diego, CA) statistical software.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). The elevation in total testosterone
was reflected in the free testosterone fraction as well (F = 5.2;
P = 0.013; drink by time interaction; levels increased
from 0.4 ± 0.1 to 1.4 ± 0.3 pmol/L during alcohol
intoxication). A concomitant decrease in plasma androstenedione levels
(F = 10.0; P < 0.001; drink by time interaction;
Fig. 1) and, as a result, an increase in the testosterone to
androstenedione ratio (F = 34.7; P < 0.001; drink
by time interaction; the ratio increased from 0.14 ± 0.01 to
0.89 ± 0.11 during alcohol intoxication) was found. Decreases in
urinary androsterone (F = 10.2; P = 0.013; drink
by time interaction) and etiocholanolone (F = 13.7;
P = 0.006; drink by time interaction) levels were
observed during alcohol intoxication (Fig. 2).
|
|
).
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Testosterone levels can be affected by changes in its synthesis in the adrenals and ovaries, by peripheral conversion from blood androstenedione and dehydroepiandrosterone, and through changed catabolism in the liver (21). As the testosterone elevation was accompanied by a decrease in plasma androstenedione as well as a decrease in urinary androsterone and etiocholanolone, it seems that the testosterone elevation is not caused by an acute increase in adrenal, gonadal, or peripheral androgen synthesis but, rather, by an effect on its metabolism in the liver. We previously found that the magnitude of the testosterone elevation is similar after rapid intake of different amounts of alcohol (11). This suggested that the androgen effect could be related to the zero order mechanism of ethanol elimination and could be mediated by the change in the redox state, i.e. the NADH to NAD+ ratio, as this phenomenon is rather constant during different dose and time conditions (22). Therefore, we extended our earlier studies and examined the plasma steroid effects during pretreatment with 4-methylpyrazole, a compound recently approved by the FDA for the treatment of detected and suspected ethylene glycol poisoning in humans (23). 4-Methylpyrazole inhibits the alcohol dehydrogenase class I enzymes that account for the major part of ethanol elimination in the human liver (24, 25). Hence, this compound was used to differentiate the metabolic derangements of ethanol elimination in the liver from other effects of ethanol. To our knowledge, this is the first study including the use of 4-methylpyrazole in humans in vivo to elucidate the role of ethanol metabolism in the actions of alcohol on steroids.
The catabolism of testosterone involve a series of reactions in the liver to transform the steroid to water-soluble conjugated forms, mainly androsterone and etiocholanolone, that are consequently excreted in the urine (26). These reactions include oxidation of the 17-hydroxyl group of testosterone. This reaction, the oxidation of testosterone to androstenedione, is catalyzed by the 17ß-hydroxysteroid dehydrogenase type 2 enzyme (17ßHSD2) found in the human liver (27, 28). In the reaction NAD+ is reduced to NADH. The competitive situation seems to be in favor of alcohol oxidation, the major part of which takes place in the liver (29), as the alcohol-mediated elevation in the NADH to NAD+ ratio has been shown to lead to a secondary shift in the ratio of 17-hydroxy- to 17-ketosteroids, including certain conjugated steroid pairs (30), and in the ratio of estradiol to estrone (18). That alcohol oxidation is coupled to the reduction of conjugated 17- hydroxy- to 17-ketosteroids in humans has been confirmed using labeled ethanol (30). The finding that the effect in the present study was pronounced among women using oral contraceptives may at least in part be explained by a progestin-induced increase in the expression of 17ßHSD2, as has been shown to occur in the human endometrium (28, 31). The mechanism also provides an explanation for the 3-fold acute transient elevation in estradiol caused by alcohol among postmenopausal women receiving estrogen replacement therapy (19) as well as the lack of an acute effect on ethinyl estradiol protected by the ethinyl group at the 17 position (32).
In conclusion, the testosterone elevation seems to be the result of an
inhibited catabolism in the liver, i.e. a decreased overall
oxidation of testosterone due to a secondary shift in the equilibrium
between androstenedione to testosterone. This is mediated by the
alcohol-induced elevation in the ratio of NADH to
NAD+ (Fig. 5).
These results imply that the liver, in addition to the
hypothalamic-pituitary-gonadal and -adrenal axes, should be seen as a
major site for the acute actions of alcohol on steroid levels in
women.
|
Received November 9, 2000.
Revised January 18, 2001.
Accepted January 18, 2001.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-hydroxysteroid dehydrogenase activity. J Biol
Chem. 268:12964–12969.-hydroxysteroid dehydrogenase activities by progestins
in human endometrium. Endocrinology. 104:1745–1748.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  N. C. Onland-Moret, P. H. M. Peeters, Y. T. van der Schouw, D. E. Grobbee, and C. H. van Gils Alcohol and Endogenous Sex Steroid Levels in Postmenopausal Women: A Cross-Sectional Study J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1414 - 1419. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) on plasma testosterone (A) and androstenedione (B)
levels in nine premenopausal women using oral contraceptives. 
,
Plasma ethanol levels. *, P < 0.05; **,
P < 0.01 (compared with placebo).
