| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Studies |
,17ß-Diol Glucuronide as a Steroid Correlate of Visceral Obesity in Men1
Lipid Research Center (A.T., D.P., J.-P.D.), the Medical Research Council Group in Molecular Endocrinology (F.L., A.B.), and the Diabetes Research Unit (A.N.), Laval University Medical Research Center, and the Physical Activity Sciences Laboratory (C.B., A.T.), Laval University, Ste-Foy, Canada
Address all correspondence and requests for reprints to: Jean-Pierre Després, Ph.D., Lipid Research Center, Laval University Medical Research Center, 2705 Laurier boulevard (TR-93), Ste-Foy, Quebec, Canada G1V 4G2.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
,17ß-diol glucuronide (3-DIOL-G)
and androsterone glucuronide (ADT-G) as well as testosterone and
adrenal C19 steroid concentrations were measured in a
sample of 80 men in whom visceral adipose tissue (AT) accumulation was
also determined by computed tomography. Plasma 3-DIOL-G
concentrations showed significant positive correlations with total body
fat mass (r = 0.31; P < 0.05) and percent
body fat (r = 0.28; P < 0.05). Furthermore,
plasma 3-DIOL-G levels were significantly associated with visceral
adipose tissue accumulation (r = 0.41; P <
0.0005) as well as fasting plasma insulin (r = 0.35;
P < 0.005) and glycemic and insulinemic responses
to an oral glucose load (r = 0.39; P < 0.0005
and r = 0.32; P < 0.005, respectively).
However, associations between 3-DIOL-G and plasma glucose-insulin
homeostasis indexes were no longer significant after adjustment for
visceral AT area. ADT-G levels were not significantly associated with
any of the adiposity variables. Subjects matched for abdominal sc AT
area but with either low or high levels of visceral AT area showed
significant differences in 3-DIOL-G concentrations
(P < 0.05), whereas subjects with low or high
levels of abdominal sc AT but similar levels of visceral AT had similar
3-DIOL-G concentrations. Among men with high testosterone levels,
subjects with reduced 3-DIOL-G concentrations had lower visceral
adipose tissue accumulation than subjects with increased 3-DIOL-G
levels. The present results indicate that plasma 3-DIOL-G, but not
ADT-G, is a steroid correlate of visceral obesity. Excess visceral
adipose tissue and/or concomitant alterations in insulin levels or
in vivo insulin action could be responsible for the
increased 3-DIOL-G formation observed in this condition. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Glucuronidation is a metabolic pathway by which lipophilic steroids are
transformed into more water-soluble molecules, thus enhancing their
rate of excretion (6). Although there is evidence that the liver
markedly contributes to steroid glucuronide derivatives in the
circulation, recent data clearly indicate the presence of steroid
glucuronosyl transferase in extrahepatic tissues, namely androgen
target tissues (7). Conjugation of steroids with glucuronic acid has,
therefore, been suggested to play a major role in the regulation of the
intracellular levels of unconjugated steroids as well as their
biological activities in tissues (8, 9). Several groups have suggested
that in men, circulating levels of androstane-3,17ß-diol
glucuronide (3-DIOL-G) and androsterone glucuronide (ADT-G), the two
major 5-reduced androgen metabolites, originate from both the
testicular androgen testosterone and the adrenal C19
steroids (8, 9, 10). Thus, the study of glucuronide derivatives in
visceral obesity could contribute some insights on factors modulating
steroid metabolism and their plasma levels in this condition.
To examine this issue, we measured unconjugated and
glucuronidated steroid concentrations (namely 3-DIOL-G and ADT-G),
body fatness, and body fat distribution variables as well as glucose
tolerance and plasma insulin levels in a sample of 80 men with varying
levels of total body fat.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Plasma steroid measurements
Fasting plasma was obtained after centrifugation (2000 x g for 15 min) of blood samples obtained in the morning after a 12-h fast and was frozen at -80 C until steroids were assayed. Steroids were extracted from plasma with ethanol before centrifugation at 2200 x g for 15 min, as previously described (11). The resulting pellet was resuspended in ethanol before recentrifugation. The two extracts were then combined, and the ethanol was evaporated under nitrogen. The residue obtained was suspended in water-methanol (95:5) and chromatographed on C18 columns (Amersham, Oakville, Canada). Unconjugated steroids were isolated by elution with water-methanol (15:85), whereas glucuronide derivatives were eluted with water-methanol (60:40). The fraction containing glucuronide conjugates was solubilized with 1 mL 0.1 mol/L phosphate buffer (pH 6.5), and the commercial preparation of ß-glucuronidase was dialyzed in 0.1 mol/L phosphate buffer before use. Steroid glucuronides were then hydrolyzed with ß-glucuronidase for 72 h at 37 C with two daily additions of 30 U ß-glucuronidase solution. The steroids released were extracted twice with ethyl ether. The organic solvent was evaporated from the fractions containing the unconjugated and the hydrolyzed steroids with a Speed-Vac rotary concentrator (Savant Instruments, Armingdale, NY).
To separate steroids, chromatography on LH-20 columns was performed
(12). After solubilization in 1 mL isooctane-toluene-methanol (90:5:5),
the steroids were deposited on Sephadex LH-20 columns (Pharmacia,
Uppsala, Sweden). Elution was performed by increasing the polarity of
the organic solvent mixture, and four fractions were collected. After
deposition of steroids, 15 mL isooctane-toluene-methanol (90:5:5) were
passed through the column and discarded. After the addition of 20 mL
isooctane-toluene-methanol (90:5:5), 
4-DIONE,
androsterone, and dehydroepiandrosterone (DHEA) were collected in the
eluant. Elution of 3-DIOL and testosterone was then achieved by the
addition of another 20 mL of the same solvent. Charcoal-treated plasma
was used as a blank for the elution on LH-20 columns, and the
background levels obtained for these blanks were generally found to be
lower than the limit of detection. Steroids were measured by RIA, as
previously described (11, 12). Intra- and interassay coefficients of
variation for the various steroid measurements were always below 9%
and 15%, respectively.
Measurement of body fatness and adipose tissue distribution
Body density was determined by hydrostatic weighing (13), with measurement of the pulmonary residual volume by the helium dilution method (14) before immersion in a hydrostatic tank. The equation of Siri (15) was used to derive percent body fat mass values from the mean of six body density measurements.
Measurements of cross-sectional abdominal sc and visceral adipose tissue areas were performed by computed tomography, as previously described (16, 17), with a Siemens Somatom DHR scanner (Erlangen, Germany). Subjects were examined in the supine position. The abdominal measurement was made between the L4 and L5 vertebrae, and the visceral AT area was obtained by drawing a line within the muscle wall surrounding the abdominal cavity using a graph pen and an attenuation range of -190 to -30 Hounsfield units (16, 17).
Oral glucose tolerance test
An oral glucose tolerance test was performed in the morning after an overnight fast. Blood samples were collected through a venous catheter from an antecubital vein in Vacutainer tubes (Becton-Dickinson, Franklin Lake, NJ) containing Trasylol (Miles, Rexdale, Ontario, Canada) and ethylenediamine tetraacetate. Sampling was performed 15 min before and 0, 15, 30, 45, 60, 90, 120, and 180 min after the ingestion of 75 g glucose. Insulin concentrations were determined using polyethylene glycol separation (18), and glucose levels were measured enzymatically (19). The glucose and insulin areas under the curves were determined with the trapezoid method.
Statistical analyses
Pearson correlation coefficients were computed to quantify the
relationships among total body fatness, body fat distribution indexes,
glucose-insulin homeostasis variables, plasma concentrations of free
steroids, and glucuronide conjugates. The comparison of steroid levels
between overweight vs. nonobese men was performed using
Student’s t test. Comparison of subjects with low
vs. high levels of visceral AT but matched for levels of
abdominal sc AT or of subjects with low vs. high abdominal
sc AT area but matched for levels of visceral AT was performed using
paired t tests. The total sample was also subdivided into
subgroups with low vs. high testosterone levels according to
the 50th percentile of the distribution of plasma testosterone (11.52
nmol/L) and into subjects with low vs. high 3-DIOL-G
concentrations according to the 50th percentile of 3-DIOL-G
distribution (9.40 nmol/L). Differences among these four subgroups were
examined with Duncan’s multiple range test. All statistical analyses
were performed with the SAS package (SAS Institute, Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. The subjects’
average plasma testosterone level reached 12.14 ± 3.17 nmol/L,
whereas overweight subjects had reduced testosterone concentrations
compared to nonobese subjects (obese, 11.29 ± 3.02 nmol/L; lean,
13.64 ± 2.90 nmol/L; P < 0.001). Adrenal
C19 steroid levels were also found to be lower in
overweight compared to nonobese men (35% lower DHEA levels and 23%
lower androstenedione levels) (1). Overweight men had significantly
higher plasma concentrations of 3-DIOL-G (36% higher in obese) than
nonobese controls, whereas ADT-G levels only tended to be higher in
obese men. When subjects were arbitrarily subdivided on the basis of
percent body fat values derived from underwater weighing, men with
values above 25% body fat had significantly higher plasma 3-DIOL-G
levels compared to leaner subjects (<25% body fat), 10.62 ±
3.60 vs. 8.27 ± 3.72 nmol/L; P <
0.006, respectively. However, plasma ADT-G levels were not
significantly different in these two subgroups.
|
. Although ADT-G
levels were not significantly correlated to adiposity and body fat
distribution indexes, 3-DIOL-G concentrations showed significant
positive correlations with total body fat mass (r = 0.31;
P < 0.05). Furthermore, 3-DIOL-G levels were also
positively associated with abdominal sc AT area (r = 0.33;
P < 0.005) as well as with visceral adipose tissue
accumulation measured by computed tomography (r = 0.41;
P < 0.0005). Statistical control for the concomitant
variation in total body fat mass failed to eliminate the association
between plasma 3-DIOL-G levels and visceral AT area (r = 0.30;
P < 0.05). Furthermore, multivariate analyses revealed
that 3-DIOL-G explained a significant portion of the variance in
visceral AT accumulation (r2 = 4.3%; P <
0.01) even after total body fat mass had been included in the model
(not shown).
|
-DIOL-G concentration was noted as a function of increasing levels
of visceral AT (Fig. 2). Androstenedione levels showed a
similar, although less pronounced, decline with increasing visceral AT
accumulation, whereas ADT-G levels were not significantly different
among these subgroups (data not shown).
|
-DIOL-G levels were associated with both abdominal sc and
visceral AT accumulation, it was of interest to examine the independent
contribution of 3-DIOL-G as a correlate of the visceral
vs. sc abdominal AT. For that purpose, two subgroups of men
with low vs. high levels of visceral AT but who were
individually matched for levels of abdominal sc AT were compared. The
results shown in Fig. 3A indicate that the two subgroups
showed a 2.5-fold difference in visceral AT area but had similar levels
of abdominal sc AT. However, 3-DIOL-G levels were significantly
higher in men with high levels of visceral AT (12.4 ± 4.7
vs. 8.2 ± 3.9 nmol/L; P < 0.05). On
the other hand, when subjects with low vs. high abdominal sc
AT areas but similar levels of visceral AT were compared, no difference
in 3-DIOL-G levels was noted despite the fact that men with high
levels of sc abdominal AT also had greater levels of total body fat
than subjects with low levels of sc AT (26.0 vs. 15.3 kg;
P < 0.0005; Fig. 3B). Plasma 3-DIOL-G levels,
therefore, appear to be more closely associated with visceral AT
accumulation; the association noted between this glucuronide conjugate
and abdominal sc AT area probably reflects the positive correlation
between 3-DIOL-G and total body fat mass. Similar results were
obtained when comparing subjects with low or high visceral AT area but
matched for levels of total body fat. Indeed, it was found that
regardless of total body fat mass, individuals with high 3-DIOL-G
levels were characterized by larger accumulations of visceral AT (not
shown).
|
-DIOL-G or ADT-G levels were associated with indexes of
plasma glucose-insulin homeostasis. The results shown in Table 2 indicate that 3-DIOL-G, but not ADT-G,
concentrations were significantly and positively associated with
fasting glucose (r = 0.28; P < 0.005) and insulin
levels (r = 0.35; P < 0.005) as well as with the
areas under the curve of glucose (r = 0.39; P <
0.0005) and insulin (r = 0.32; P < 0.005)
concentrations measured during the oral glucose challenge. However,
covariance analysis revealed that controlling for the concomitant
variation in body fat mass and visceral AT accumulation eliminated
the associations between 3-DIOL-G levels and indexes of plasma
glucose-insulin homeostasis. Thus, it appears that the relationship
between visceral AT and 3-DIOL-G concentrations is independent
of the concomitant variations in glucose tolerance and plasma
insulin levels.
|
-DIOL-G was negative but rather weak, and it barely reached
significance (r = -0.23; P = 0.05; not shown).
Thus, in the present sample, it was possible to identify subjects with
high testosterone and low 3-DIOL-G levels or with low testosterone
and high 3-DIOL-G concentrations. On the other hand, plasma
testosterone levels were lower in overweight men of the present sample
and were also significant correlates of visceral adiposity (1). To
determine whether plasma testosterone levels were related to the
increased 3-DIOL-G concentrations found in individuals with high
levels of visceral adipose tissue, the sample was divided into two
groups on the basis of plasma testosterone concentrations (low
vs. high), using the 50th percentile of the distribution of
testosterone as a cut-off point (11.52 nmol/L). We then further
subdivided each of these subgroups into subjects with low
vs. high 3-DIOL-G levels [according to the 50th
percentile (9.40 mmol/L) of the distribution of 3-DIOL-G values].
As shown in the upper panel of Fig. 4, there
was a marked difference in 3-DIOL-G levels between subjects with low
vs. high 3-DIOL-G, although there was no difference
between either groups 1 vs. 3 or groups 2 vs. 4.
As shown in the middle panel of Fig. 4, subjects with low
testosterone levels (groups 1 and 2) had testosterone concentrations of
approximately 10 nmol/L, whereas subgroups with high testosterone (3, 4) had testosterone values above 14 nmol/L. When visceral AT
accumulation was compared among these subgroups (Fig. 4, lower
panel), it was found that subjects with both high testosterone
levels and low 3-DIOL-G were characterized by the lowest visceral AT
accumulation (84.1 ± 9.9 cm2). Furthermore, among
subjects with high testosterone concentrations (groups 3 and 4),
visceral AT accumulation was significantly higher in subjects with high
3-DIOL-G levels. As adrenal C19 steroid levels were also
lower in these obese men compared to those in the lean subjects, and
these steroids were good correlates of visceral AT accumulation,
similar analyses were performed with DHEA and androstenedione. Results
essentially similar to those obtained with testosterone were noted, as
subjects with low 3-DIOL-G levels and high DHEA or androstenedione
had the lowest visceral AT accumulation (results not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-DIOL-G concentrations were
significantly higher in overweight men, particularly among those with
excess levels of visceral adipose tissue. One previous study on
glucuronide conjugates in obesity (20) had shown that plasma
3-DIOL-G levels were not increased in obese men, suggesting that
5-reductase and glucuronosyl transferase activities were not related
to obesity. The present results are obviously at variance with these
previous observations, as we report that 3-DIOL-G levels are
increased in visceral obesity. It is important to point out, however,
that the subjects in the present sample only included overweight or
moderately obese men, as opposed to the study by Giagulli and
colleagues (20), in which moderate and massively obese subjects were
studied. Another possible explanation for this discrepancy could be
differences in the pattern of body fat distribution among study groups,
which had not been extensively characterized previously (20). As
3-DIOL-G appears to be a good correlate of visceral AT accumulation
measured by computed tomography, higher levels or a greater range of
variation in visceral adipose tissue accumulation could account for the
increased 3-DIOL-G levels or the significant association between
visceral AT and 3-DIOL-G concentrations reported in the present
study.
As shown in many previous reports, testosterone is negatively related
to visceral AT accumulation in men (1, 5, 21, 22, 23, 24). Indeed, studies on
the pathogenesis of the reduced plasma androgen levels in visceral
obese men have suggested that the steroid alterations found in such a
condition could result from an increased activity of the
hypothalamic-pituitary-adrenal axis, leading to elevated cortisol
secretion and reduced gonadal androgen levels (21). Intervention
studies in visceral obese patients in which testosterone levels were
restored to the normal range resulted in the mobilization of visceral
AT (21, 22, 23, 24). Indeed, testosterone treatment has been shown to stimulate
lipolysis in male rat adipocyte precursor cells and to reduce
triglyceride assimilation in adipose cells from the abdominal cavity
(25, 26, 27). Accordingly, in the present study, subjects with low
testosterone concentrations were characterized by higher levels of
visceral AT. However, as opposed to what was noted among subjects with
high plasma testosterone levels, 3-DIOL-G concentrations did not
appear to further discriminate subjects with low or high visceral AT
when testosterone concentrations were low. This lack of statistical
difference could be attributable to the fact that in the present
sample, very few subjects (8.75% of the sample) were characterized by
both reduced testosterone and a low accumulation of visceral AT
(cross-sectional area <100 cm2), as there was a
significant negative correlation between testosterone levels and
visceral AT accumulation. However, when adjusting visceral AT area for
total body fat mass in the four subgroups shown in Fig. 4
, subjects
with high testosterone and low 3-DIOL-G levels were still
characterized by the lowest visceral AT accumulation.
As stated earlier, it appears that the reduced plasma testosterone
levels observed in visceral obesity may be related to an increased
activity of the hypothalamic-pituitary-adrenal axis. However, adipose
tissue has been shown to express steroidogenic enzymes and is,
therefore, likely to be an important site of steroid formation
(28, 29, 30). The available evidence suggests that weight loss is
associated with changes in steroid hormone concentrations (31, 32),
although not all studies support this idea (33). Thus, steroid hormones
may affect fat accretion, but adipose tissue may also by itself alter
steroid hormone concentrations. The present correlational study was
obviously not designed to address this issue of causality. However, it
appears fair to suggest that visceral adipose tissue may be partly
responsible for the increased 3-DIOL-G concentrations found in
visceral obese men regardless of the origin of the low androgen
concentrations found in these men, as the correlation between
testosterone and 3-DIOL-G was weak and barely reached significance
(r = -0.23; P = 0.05). Furthermore, correlations
between plasma glucuronide conjugates and adrenal C19
steroids were not significant. Glucuronide conjugates have been
considered better markers of peripheral androgen metabolism than
circulating free steroids (8, 34). Thus, plasma total testosterone and
adrenal steroid levels, although significant correlates of obesity in
the present sample (1), may not necessarily be related to tissular
glucuronide formation. Nevertheless, the correlation between visceral
AT and plasma 3-DIOL-G levels could be due to increased androgen
formation and glucuronidation in visceral adipose tissue.
In addition to being associated with visceral adipose tissue
accumulation, 3-DIOL-G was significantly correlated with elevated
fasting insulin and glucose concentrations as well as with insulinemic
and glycemic responses to oral glucose (Table 2).
However, after adjustment for visceral AT accumulation, none of these
correlations remained significant, suggesting that, as for plasma free
steroid levels (35), the relationships between 3-DIOL-G and glucose
tolerance or plasma insulin levels were mediated to a large extent by
the concomitant variation in the amount of visceral adipose tissue.
The present study also reports that in contrast to 3-DIOL-G, there
was no correlation between ADT-G and obesity, adipose tissue
distribution, and the concomitant hyperinsulinemic state. Accordingly,
recent data have shown that two different enzymes are responsible for
glucuronidation of ADT and 3-DIOL (7, 36). The metabolism of ADT-G
and 3-DIOL-G may, therefore, be different in the presence of
visceral obesity and/or insulin resistance.
In summary, the present report shows that 3-DIOL-G appears to be a
good plasma steroid correlate of visceral obesity. Such results suggest
that visceral adipose tissue and/or concomitant metabolic alterations
related to an insulin-resistant hyperinsulinemic state could be
involved in 3-DIOL-G formation, although the present study was not
designed to address the issue of causality. Thus, visceral adipose
tissue accumulation appears as a condition in which steroid metabolism
is altered. Further studies are warranted to elucidate the biochemical
mechanisms responsible for these observations.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipient of a Fonds de la Recherche en Santé du
Québec-Fonds de Formation de Chercheurs et l’Aide à la
Recherche (FRSQ-FCAR Santé) fellowship.
Received August 29, 1996.
Revised January 3, 1997.
Accepted January 30, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
5–4 isomerase (3ß-HSD)
and 17ß-hydroxysteroid dehydrogenase (17ß-HSD) in adipose tissue. Int J Obes. 15:91–99.
This article has been cited by other articles:
 |
|
 |
|
  L. Vandenput, D. Mellstrom, M. Lorentzon, C. Swanson, M. K. Karlsson, J. Brandberg, L. Lonn, E. Orwoll, U. Smith, F. Labrie, et al. Androgens and Glucuronidated Androgen Metabolites Are Associated with Metabolic Risk Factors in Men J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4130 - 4137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Blouin, C. Richard, G. Brochu, F.-S. Hould, S. Lebel, S. Marceau, S. Biron, V. Luu-The, and A. Tchernof Androgen inactivation and steroid-converting enzyme expression in abdominal adipose tissue in men J. Endocrinol., December 1, 2006; 191(3): 637 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Couillard, J. Gagnon, J. Bergeron, A. S. Leon, D. C. Rao, J. S. Skinner, J. H. Wilmore, J.-P. Després, and C. Bouchard Contribution of Body Fatness and Adipose Tissue Distribution to the Age Variation in Plasma Steroid Hormone Concentrations in Men: The HERITAGE Family Study J. Clin. Endocrinol. Metab., March 1, 2000; 85(3): 1026 - 1031. [Abstract] [Full Text]
|
|
|
|
 |
|
 G. A. Bray and F. L. Greenway Current and Potential Drugs for Treatment of Obesity Endocr. Rev., December 1, 1999; 20(6): 805 - 875. [Abstract] [Full Text]
|
|
|
|
|
|
J. Pritchard, J.-P. Després, J. Gagnon, A. Tchernof, A. Nadeau, A. Tremblay, and C. Bouchard Plasma Adrenal, Gonadal, and Conjugated Steroids before and after Long Term Overfeeding in Identical Twins J. Clin. Endocrinol. Metab., September 1, 1998; 83(9): 3277 - 3284. [Abstract] [Full Text]
|
|
|
|
|
|
T. Remer, K. Pietrzik, and F. Manz Short-Term Impact of a Lactovegetarian Diet on Adrenocortical Activity and Adrenal Androgens J. Clin. Endocrinol. Metab., June 1, 1998; 83(6): 2132 - 2137. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
