Synthesis and regulation of sex hormone-binding globulin in obesity.

Sex hormone-binding globulin (SHBG) is a plasma glycoprotein with high binding affinity for testosterone and dihydrotestosterone and lower affinity for estradiol. SHBG is synthesized in the liver, and its plasma level is important in the regulation of plasma free and albumin-bound androgens and estrogens. Obesity and particularly excess visceral fat, known risk factors for cardiovascular and metabolic diseases, are associated with decreased testosterone levels in males and SHBG levels in both sexes. SHBG is usually positively correlated with high-density lipoprotein cholesterol and negatively correlated with triglyceride and insulin concentrations. A positive association between SHBG and various measures of insulin sensitivity has been demonstrated in both sexes, suggesting that decreased SHBG levels may be one of the components of the metabolic syndrome. We have examined pituitary-adrenocortical function, glucose tolerance, and lipoprotein and hormone levels in a large cohort of Finnish males. Abdominal obesity appears to be associated with slight hypocortisolemia and increased sensitivity to exogenous adrenocorticotropin stimulation, which may contribute to the hyperinsulinemia and related metabolic changes including decreased SHBG levels in males.

Baroreflex sensitivity and variants of the renin angiotensin system genes.

OBJECTIVES: Because the renin-angiotensin-aldosterone system (RAS) modifies cardiovascular autonomic regulation, we studied the possible associations between baroreflex sensitivity (BRS) and polymorphism in the RAS genes. BACKGROUND: Wide intersubject variability in BRS is not well explained by cardiovascular risk factors or life style, suggesting a genetic component responsible for the variation of BRS. METHODS: Baroreflex sensitivity as measured from the overshoot phase of the Valsalva maneuver and genetic polymorphisms were examined in a random sample of 161 women and 154 men aged 41 to 61 years and then in an independent random cohort of 29 men and 37 women aged 36 to 37 years. An insertion/deletion (I/D) polymorphism of angiotensin-converting enzyme (ACE), M235T variants of angiotensinogen (AGT) and two diallelic polymorphisms in the gene encoding aldosterone synthase (CYP11B2), one in the promoter (-344C/T) and the other in the second intron, were identified by polymerase chain reaction. RESULTS: In the older population, BRS differed significantly across CYP11B2 genotype groups in women (10.1 +/- 4.5, 8.7 +/- 3.8 and 7.1 +/- 3.2 ms x mm Hg(-1) in genotypes -344TT, CT and CC, respectively, p = 0.003 and 11.1 +/- 4.4, 8.9 +/- 4.1 and 7.5 +/- 3.4 ms x mm Hg(-1) in intron 2 genotypes 1/1, 1/2 and 2/2, respectively, p = 0.002), but not in men. No comparable associations were found for BRS with the I/D polymorphism of ACE or the M235T variant of AGT. In the younger population, BRS was even more strongly related to the CYP11B2 promoter genotype (p = 0.0003). The association was statistically significant both in men (p = 0.015) and in women (p = 0.03). CONCLUSIONS: Common genetic polymorphisms in the aldosterone synthase (CYP11B2) gene is associated with interindividual variation in BRS.

Joint effects of an aldosterone synthase (CYP11B2) gene polymorphism and classic risk factors on risk of myocardial infarction.

BACKGROUND: The -344C allele of a 2-allele (C or T) polymorphism in the promoter of the gene encoding aldosterone synthase (CYP11B2) is associated with increased left ventricular size and mass and with decreased baroreflex sensitivity, known risk factors for morbidity and mortality associated with myocardial infarction (MI). We hypothesized that this polymorphism was a risk factor for MI. METHODS AND RESULTS: We used a nested case-control design to investigate the relationships between this polymorphism and the risk of nonfatal MI in 141 cases and 270 matched controls from the Helsinki Heart Study, a coronary primary prevention trial in dyslipidemic, middle-aged men. There was a nonsignificant trend of increasing risk of MI with number of copies of the -344C allele. However, this allele was associated in a gene dosage-dependent manner with markedly increased MI risk conferred by classic risk factors. Whereas smoking conferred a relative risk of MI of 2.50 (P=0.0001) compared with nonsmokers in the entire study population, the relative risk increased to 4.67 in -344CC homozygous smokers (relative to nonsmokers with the same genotype, P=0.003) and decreased to 1.09 in -344TT homozygotes relative to nonsmokers with this genotype. Similar joint effects were noted with genotype and decreased HDL cholesterol level as combined risk factors. CONCLUSIONS: Smoking and dyslipidemia are more potent risk factors for nonfatal MI in males who have the -344C allele of CYP11B2.

Aldosterone synthase (CYP11B2) polymorphisms and cardiovascular function.

In addition to regulating renal sodium resorption and, thus, intravascular volume, aldosterone may have direct effects on the cardiovascular system. We previously identified a polymorphism (-344C/T) in the promoter of the aldosterone synthase (CYP11B2) gene that affects binding of the SF-1 transcription factor and thus might influence gene expression. We found that, whereas this polymorphism has inconsistent associations with levels of aldosterone secretion and blood pressure, the -344C allele is strongly associated with increased left ventricular size and decreased baroreflex sensitivity in healthy individuals. These physiological parameters are cardiovascular risk factors. Indeed, preliminary studies suggest that the -344C allele is also associated with increased risk of myocardial infarction in high risk dyslipidemic males.

Relationships between the pituitary-adrenal hormones, insulin, and glucose in middle-aged men: moderating influence of psychosocial stress.

We examined whether the relationships between the pituitary-adrenal hormones (corticotropin [ACTH) and cortisol), insulin, and glucose differ as a function of psychosocial stress defined in terms of vital exhaustion (VE) and depressive behavior (DB). The participants were 69 normotensive and 21 unmedicated borderline hypertensive (BH) middle-aged men whose work is stressful. Hormonal and metabolic variables were measured during an oral glucose tolerance test (OGTT), and the cortisol response to dexamethasone (DXM) suppression and intravenous ACTH stimulation was also measured. We found that the basal ACTH level during the OGTT was positively associated with the cortisol response to ACTH at 60 minutes, the fasting insulin level, and the insulin to glucose ratio among exhausted and high DB men, while the reverse was true for nonexhausted and low DB men. Also, a high cortisol response to ACTH, a low cortisol level during the OGTT, and a high ratio of these cortisol determinations (cortisol ratio) were associated with high fasting insulin and glucose levels, the summed insulin values, and the insulin to glucose ratio only among nonexhausted and low DB men; among exhausted and high DB men, these associations were less pronounced, absent, or in the opposite direction. The findings suggest that VE and DB have a moderating influence on the relationships among the hormonal and metabolic parameters studied. Psychosocial stress may affect the pituitary-adrenocortical system in complex ways, contributing thereby to insulin resistance, hyperinsulinemia, and coronary heart disease (CHD) risk.

Associations between human aldosterone synthase (CYP11B2) gene polymorphisms and left ventricular size, mass, and function.

BACKGROUND: Aldosterone has direct and indirect effects on the heart, and genetic variations in aldosterone synthesis could therefore influence cardiac structure and function. Such variations might be associated with polymorphisms in the gene encoding aldosterone synthase (CYP11B2), the enzyme catalyzing the last steps of aldosterone biosynthesis. METHODS AND RESULTS: A Finnish population sample of 84 persons (44 women) aged 36 to 37 years was studied by M-mode and Doppler echocardiography to assess left ventricular size, mass, and function. Subjects were genotyped through the use of the polymerase chain reaction for two diallelic polymorphisms in CYP11B2: one in the transcriptional regulatory region (promoter) and the other in the second intron. In multiple regression analyses, the CYP11B2 promoter genotype predicted statistically significant variations in left ventricular end-diastolic diameter (beta=.40, P<.0001), end-systolic diameter (beta=.33, P=.0009), and mass (beta=.17, P=.023). These effects were independent of potentially confounding factors, including sex, body size, blood pressure, physical activity, smoking, and ethanol consumption. Genotype groups also differed in a measure of left ventricular diastolic function, the heart rate-adjusted atrial filling fraction (P=.018). Increased dietary salt, which is known to predict increased left ventricular mass, had this effect only in association with certain CYP11B2 genotypes (P<.001). CONCLUSIONS: Genetic variations in or near the aldosterone synthase (CYP11B2) gene strongly affect left ventricular size and mass in young adults free of clinical heart disease. These polymorphisms may also influence the response of the left ventricle to increases in dietary salt.

Aldosterone synthase (CYP11B2) polymorphisms and cardiovascular function.

In addition to regulating renal sodium resorption and, thus, intravascular volume, aldosterone may have direct effects on the cardiovascular system. We previously identified a polymorphism (-344C/T) in the promoter of the aldosterone synthase (CYP11B2) gene that affects binding of the SF-1 transcription factor and thus might influence gene expression. We found that, whereas this polymorphism has inconsistent associations with levels of aldosterone secretion and blood pressure, the -344C allele is strongly associated with increased left ventricular size and decreased baroreflex sensitivity in healthy individuals. These physiological parameters are cardiovascular risk factors. Indeed, preliminary studies suggest that the -344C allele is also associated with increased risk of myocardial infarction in high risk dyslipidemic males.

Associations between pituitary-adrenocortical function and abdominal obesity, hyperinsulinaemia and dyslipidaemia in normotensive males.

OBJECTIVES: To examine the relationships between pituitary-adrenal cortical (PA) function, abdominal obesity, hyperinsulinaemia, and dyslipidaemia. DESIGN: A prospective study. SETTING: Helsinki University Central Hospital, Finland. SUBJECTS: Seventy-one healthy males aged 30-55 years. MAIN OUTCOME MEASURES: Insulin sensitivity was assessed by the oral glucose tolerance test (OGTT). Basal PA activity was examined by measuring urinary and serum concentrations of hormones, followed by dexamethasone suppression and corticotrophin (ACTH) stimulation tests to determine functional PA activity. RESULTS: The means of waist-to-hip ratio (WHR), body-mass index (BMI), HDL-cholesterol and triglyceride levels, and insulin and C-peptide measurements during the OGTT were significantly different across the tertiles for insulin:glucose ratio. The ratio of 12-h urinary cortisol excretion to BMI, preceding the OGTT, and the mean basal cortisol level during the OGTT were decreased, while the net increments of cortisol and 17-hydroxyprogesterone (17-OHP) from 0 to 60 min, as well as the ratio of net 17-OHP to 11-deoxycortisol increments, after ACTH, were elevated in the upper compared with the lower tertile. The mean cortisol during the OGTT, and the ratio of urinary cortisol to BMI were negatively related, while absolute DHEA and cortisol responses to ACTH were positively related to fasting and mean insulin levels. Hormonal variables, WHR, insulin, and triglycerides were successfully integrated into a tentative mathematical model by the use of covariance structure (path) analyses. CONCLUSIONS: Several alterations in the PA function, suggestive of decreased 21-hydroxylase activity, mild cortisol deficiency and slight adrenal hyperplasia, are associated with abdominal obesity which, in turn, appears to be an important prelude to insulin resistance and dyslipidaemia.

Psychosocial stress and the insulin resistance syndrome.

We examined the association between psychosocial stress-related variables and insulin resistance syndrome (IRS) risk-factor clustering. In 90 middle-aged male volunteers, psychosocial stress-related variables, defined as feelings of excessive tiredness and as personality and behavioral factors reflecting a stress-inducing life-style (type A behavior, hostility, and anger), were significantly correlated with the hyperinsulinemia, hyperglycemia, dyslipidemia, hypertension, increased abdominal obesity, and increased plasminogen activator inhibitor-1 (PAI-1) antigen comprising the IRS. The correlations remained significant after adjusting for body mass index (BMI), age, educational level, smoking status, alcohol consumption, and physical activity. However, the different stress-related factors reflected different risk-factor clustering profiles. Type A behavior was associated with normotension and a normal metabolic profile (canonical r = .50, chi2(36) = 59.1, P = .008). Hostility was related to elevated systolic blood pressure (SBP) and elevated triglycerides (TGs) (canonical r = .38, chi2(14) = 23.2, P = .052), whereas feelings of excessive tiredness were related to abdominal obesity, augmented glycemic responses to glucose ingestion, dyslipidemia, and increased PAI-1 antigen (canonical r = .39, chi2(24) = 36.8, P = .046). Although hostility and feelings of excessive tiredness have partly overlapping but clearly different clinical and metabolic correlates, their combination represents a full-blown IRS. Thus, even though insulin resistance is presumably to some extent genetically determined, these results suggest that considering psychosocial stress may be beneficial in understanding IRS risk-factor clustering.

Feelings of exhaustion, emotional distress, and pituitary and adrenocortical hormones in borderline hypertension.

OBJECTIVE: To examine whether feelings of exhaustion and emotional distress reflecting chronic perceived stress contribute to a pattern of pituitary and adrenocortical responses that would in turn be able to distinguish borderline hypertensives from normotensive controls. DESIGN: A cross-sectional study. SETTING: Helsinki University Central Hospital. PARTICIPANTS: Twenty-one middle-aged, male borderline hypertensives (140/90 mmHg < or = blood pressure < 160/95 mmHg) and 69 healthy normotensive controls (blood pressure < 140/90 mmHg). Main outcome measures Basal pituitary-adrenocortical activity was assessed by measurements of plasma cortisol and adrenocorticotrophin concentrations during the oral glucose tolerance test. Cortisol responses to dexamethasone suppression and adrenocorticotrophin stimulation tests were measured to determine the functional pituitary-adrenocortical activity. Feelings of exhaustion, namely, feelings of excess fatigue, loss of energy, increased irritability, demoralization and emotional distress were measured using a questionnaire. RESULTS: As has previously been shown, feelings of exhaustion and emotional distress are associated with a hormonal pattern consisting primarily of an elevation in cortisol response to adrenocorticotrophin stimulation and secondarily of dominance of cortisol in the ratio of mean basal cortisol level to mean basal adrenocorticotrophin level. This particular neuroendocrine pattern, denoting a defeat type of reaction to stress, was in turn able to distinguish borderline hypertensives from normotensive controls significantly. Adjustment for age and health-related lifestyle factors including smoking, alcohol consumption and physical activity did not alter the difference found between the groups. CONCLUSIONS: The results suggest that the variance shared by feelings of exhaustion, emotional distress and pituitary-adrenocortical hormones could elucidate the mechanisms by which stress exerts its influence towards an increased risk for hypertension.

Pituitary-adrenocortical function in abdominal obesity of males: evidence for decreased 21-hydroxylase activity.

Certain differences in regional fat distribution might be explicable by subtle hyperactivity of the hypothalamic-pituitary-adrenal (HPA) axis. We examined prospectively PA function relative to abdominal obesity defined by waist-to-hip circumference ratio (WHR) in 71 normotensive men aged 30-55 years. Basal PA activity was assessed by measurements of serum cortisol and plasma corticotropin (ACTH) concentrations during the oral glucose tolerance test (OGTT). Functional activity was examined by dexamethasone suppression and ACTH stimulation tests; responses of 17-hydroxyprogesterone (17-OHP), 11-deoxycortisol (S), cortisol, dehydroepiandrosterone (DHEA), and androstenedione were determined. When the subjects were divided into tertiles for the WHR, the ratio of mean ACTH to mean cortisol during the OGTT was increased (p < 0.05), and the ratio of urinary cortisol to body-mass index was decreased (p < 0.01), whilst the net increments of cortisol (p < 0.05) and 17-OHP (p < 0.05) from 0 to 60 min, as well as the ratio of 17-OHP to S increments (p < 0.05) after ACTH were elevated in the highest vs lowest WHR tertile. The ratio of mean ACTH to mean cortisol (r = 0.495; p < 0.001) during the OGTT, the ratio of net 17-OHP to S increments (r = 0.404; p < 0.001), and the net DHEA (r = 0.276; p = 0.020) and 17-OHP (r = 0.336; p = 0.005) responses to ACTH at 60 min correlated with WHR. In multivariate analyses the ratio of mean ACTH to cortisol, cortisol response to ACTH, and the ratio of net 17-OHP to S increments were all significant predictors of WHR independent of smoking, physical activity, and BMI explaining 49.0% of the variance in WHR. Thus, abdominal obesity may be associated with decreased activity of adrenal 21-hydroxylase. Either obesity-related functional alteration of 21-hydroxylase activity or the high carrier prevalence of genetic defects of this enzyme may explain these findings.

Association of chronic stress with plasminogen activator inhibitor-1 in healthy middle-aged men.

The effect of chronic stress on tissue-type plasminogen activator (TPA) and plasminogen activator inhibitor-1 (PAI-1) antigens was studied in 69 healthy middle-aged men. Chronic stress, defined as feelings of fatigue, lack of energy, increased irritability, and demoralization, was positively associated with plasma concentrations of PAI-1 antigen but was unrelated to TPA. The association remained unaltered after controlling for age, smoking, alcohol consumption, and physical activity but became nonsignificant after further controlling for abdominal obesity, BMI, and serum insulin and triglyceride levels. This attenuated association implies that the relationship between vital exhaustion and PAI-1 may be secondary to the effects of the metabolic variables. Thus, the present study shows that long-term stress affects the fibrinolytic system and suggests that obesity and insulin and triglyceride concentrations, which are closely correlated with the fibrinolytic parameters, may mediate the association. These findings are consistent with the hypothesis that chronic stress causes increased synthesis of PAI-1, thus promoting the risk for atherothrombotic disease by decreasing the likelihood of spontaneous fibrinolysis and increasing the likelihood of fibrin deposition.

Vital exhaustion, anger expression, and pituitary and adrenocortical hormones. Implications for the insulin resistance syndrome.

This study was undertaken to examine whether there are psychological factors that can incline an individual toward coronary heart disease and that can in turn identify a pattern of pituitary and adrenocortical responses that is associated with the Insulin Resistance Syndrome (IRS). The study was performed with 69 normotensive and 21 unmedicated borderline hypertensive men (age range, 30 to 55 years). Type A behavior, hostility (defined as cynicism, pessimism, and paranoia), vital exhaustion, and anger expressions were the behavioral variables studied. Among these, only the vital exhaustion-anger-out factor identified the neuroendocrine pattern that predicted the IRS. This neuroendocrine pattern consisted primarily of an adrenal responsiveness to ACTH and secondarily of a high mean basal cortisol-to-mean basal ACTH ratio. The contribution of this last variable was, however, slightly questionable. Instead of the traditional coronary-prone factors, ie, type A behavior and hostility, the findings emphasize the significance of vital exhaustion and emotional distress. The findings have been discussed in terms of defeat reaction, hypocortisolemia, and visceral obesity.

Type A behavior and vital exhaustion as related to the metabolic hormonal variables of the hypothalamic-pituitary-adrenal axis.

The authors examined the correlations of Type A behavior and vital exhaustion with the metabolic hormonal variables of the hypothalamic-pituitary-adrenal axis (ACTH and cortisol), as well as with the ensuing parameters (insulin, glucose, and C-peptide). The participants were 64 healthy middle-aged men with stressful work. The authors found that Type A behavior and vital exhaustion were not correlated but made independent contributions to the single metabolic hormonal variables. The central finding was that when the whole HPA axis, Type A behavior, and exhaustion were all analyzed at the same time, Type A behavior per se had different, even opposite, hormonal correlates from Type A behavior associated with vital exhaustion. Type A behavior by itself was not related to a metabolic hormonal dysfunction. However, the Type A by vital exhaustion interaction was statistically significant. This finding suggests that the quality or components of Type A behavior, particularly the presence of vital exhaustion, may be more important than the intensity of Type A behavior considered alone.

Association of stress and depression with regional fat distribution in healthy middle-aged men.

We examined whether the association of regional fat distribution with stress, defined in terms of vital exhaustion, and depression varies according to the total amount of body fat accumulation in healthy middle-aged men (n = 64). Regional fat distribution was measured using the waist-to-hip circumference ratio (WHR), and the total amount of body fat accumulation was measured using the body mass index (BMI). The results indicate that WHR in lean men was associated with characteristics contrary to those in moderately obese men. In lean men WHR tended to be associated with a high level of stress, while in moderately obese men an association was found with a low level of stress and a low level of depressive symptomatology. The present results support the suggestion that there is a difference between abdominal obesity at different degrees of generalized obesity, and they are likely to further our understanding about the differing risk for cardiovascular disorders posed by abdominal obesity in lean men compared to abdominal obesity in moderately obese men.

Gemfibrozil treatment is associated with elevated adrenal androgen, androstanediol glucuronide and cortisol levels in dyslipidemic men.

We have investigated the role of steroid hormones as coronary risk factors in the Helsinki Heart Study population of dyslipidemic middle-aged men. We compare here the effects of gemfibrozil and placebo on the serum levels of dehydroepiandrosterone (DHEA), its sulfate (DHEAS), their metabolite androstanediol glucuronide (3 alpha-AdiolG), androstenedione, cortisol, testosterone, and sex-hormone binding globulin (SHBG) in non-smokers. We also examined the associations between steroid and lipoprotein levels in both treatment groups. Compared with placebo gemfibrozil treatment was associated with significant elevations of the mean levels of DHEA 10.2 vs 8.0 nmol/l; P < 0.005, of DHEAS 8.0 vs 5.8 mumol/l; P < 0.001, of 3 alpha AdiolG 18.3 vs 8.4 nmol/l; P < 0.001, of androstenedione 5.7 vs 5.1 nmol/l; P < 0.02, and of cortisol 426 vs 358 nmol/l; P < 0.001. The mean SHBG levels decreased from 46.4 to 41.7 nmol/l; P = 0.03 with gemfibrozil treatment. No difference was found in testosterone levels 17.7 vs 18.8 nmol/l; P = 0.11, or the ratio of testosterone/SHBG 0.45 vs 0.43; P = 0.23. Positive correlations were found between high density lipoprotein-cholesterol and DHEAS (r = 0.267; P < 0.01) and DHEA (r = 0.282; P < 0.01) levels and negative correlations between low density lipoprotein-cholesterol and 3 alpha-AdiolG (r = -0.400; P < 0.001) and cortisol (r = -0.281; P < 0.01) levels in the gemfibrozil group. Our results indicate that gemfibrozil treatment increases the production and turnover of adrenal androgens and cortisol, and suggest that activation of the adrenocortical function and increased metabolism of androgens are related to the improved lipoprotein pattern during gemfibrozil treatment.

The role of psychological coronary risk factors in insulin and glucose metabolism.

The association between psychological coronary risk factors and serum insulin, and C-peptide and blood glucose concentrations, [the latter measured while fasting and during the oral glucose tolerance test (OGTT)], was examined in healthy middle-aged men (n = 64). The results indicate that among the evaluated psychological risk factors, high levels of hostile paranoia and vital exhaustion were most consistently associated with an enhanced insulin/glucose ratio, and enhanced insulin, C-peptide and glucose responses during OGTT. The associations persisted after controlling for age, smoking, alcohol consumption and visceral fat distribution. Thus, in addition to age, life-style factors and obesity, psychological factors may have an effect on insulin and glucose metabolism.

Adrenal androgens and testosterone as coronary risk factors in the Helsinki Heart Study.

We investigated the role of adrenal androgens, cortisol, testosterone and sex-hormone binding globulin (SHBG) as coronary risk factors using a nested case-control design. The study population consisted of 62 cases with cardiac end-points and 97 controls on placebo during the last 4 years in the Helsinki Heart Study. Serum concentrations of dehydroepiandrosterone, dehydroepiandrosterone sulfate (DHEAS), androstenedione, androstanediol glucuronide, cortisol, testosterone, and SHBG at the first annual visit of the 5-year study period were determined by radioimmunoassays. The only significant difference was found in DHEAS, with cases having higher levels than controls (P < 0.04). DHEAS levels were positively associated with smoking (P < 0.001), alcohol consumption (P < 0.04) and triglyceride levels (P < 0.002) and with systolic (P < 0.04) and diastolic (P < 0.006) blood pressures, and negatively associated with age (P < 0.01) and HDL-cholesterol (P < 0.03). The association between DHEAS and the CHD risk was studied using logistic regression analyses with the classical risk factors--age, smoking, blood pressure, and lipid levels--as covariates in the models. Studies of the joint effects of age and DHEAS disclosed that the risk associated with elevated DHEAS was confirmed to older men (odds ratio (OR) 7.3, 95%, CI 2.3-23.3). A similar analysis with smoking revealed that the DHEAS-related risk was mainly found in smokers (OR 3.4, 95% CI 1.5-8.2). One possible explanation for these results is that some form of mild steroid biosynthetic defect of the adrenals or functional adrenal hyperplasia associated with high DHEAS levels increases the CHD risk in this population.

Hyperinsulinaemia, dyslipidaemia and exaggerated adrenal androgen response to adrenocorticotropin in male smokers.

Insulin resistance and dyslipidaemia associated with smoking may result from increased secretion of anti-insulin hormones. We compared the pituitary-adrenocortical function using oral glucose tolerance, dexamethasone suppression and ACTH stimulation tests in smoking (n = 22) and non-smoking (n = 22) healthy males matched for age, body mass index, and waist-to-hip ratio. Smokers had lower HDL-cholesterol (p < 0.02), and higher triglyceride (p < 0.001), basal cortisol (p < 0.05), insulin (p < 0.05), and C-peptide (p < 0.02) levels, and a higher response of insulin and C-peptide to oral glucose (p < 0.005) than non-smokers, while the ACTH, cortisol, 17-hydroxyprogesterone, dehydroepiandrosterone, and androstenedione responses to oral glucose were similar in both groups. No differences were found in the response of cortisol, 17-hydroxyprogesterone, dehydroepiandrosterone, and androstenedione to dexamethasone. In contrast, the response of 17-hydroxyprogesterone (p = 0.04), dehydroepiandrosterone (p = 0.007), and androstenedione (p = 0.001) to ACTH was higher in smokers than non-smokers, while the increase in cortisol was of marginal significance (p = 0.07). In multiple regression analyses the dehydroepiandrosterone response to ACTH was a significant determinant of insulin, C-peptide, and triglyceride levels independent of physical activity, waist-to-hip ratio and HDL-cholesterol. Thus, smoking inhibits the adrenal 21-hydroxylase resulting in an increase in the production of adrenal androgens, which might contribute to the insulin resistance and dyslipidaemia in smokers.

Altered adrenocorticotropin and cortisol secretion in abdominal obesity: implications for the insulin resistance syndrome.

OBJECTIVES: To investigate the relationship between the pituitary-adrenocortical function, abdominal obesity, and insulin resistance syndrome. DESIGN: A prospective study. SETTING: Helsinki University Hospital, Finland. SUBJECTS: Sixty-six healthy males aged 30-55 years. MAIN OUTCOME MEASURES: Insulin, C-peptide, cortisol and ACTH responses during the oral glucose tolerance test (OGTT), and the cortisol response to dexamethasone suppression and intravenous adrenocorticotrophic hormone (ACTH) stimulation. RESULTS: The subjects in the highest tertile of the waist-to-hip ratio (WHR) had lower high-density lipoprotein cholesterol (HDLC) (P < 0.05), but higher triglyceride (TG), insulin, and C-peptide levels, ACTH response to glucose at 2 h, and cortisol response to ACTH (P < 0.01) than those in the lowest tertile. The cortisol response to ACTH correlated positively, but cortisol levels during the OGTT correlated negatively with WHR. The ratio of these cortisol determinations correlated positively with the body-mass index (BMI) (r = 0.554; P < 0.001), WHR (r = 0.536; P < 0.001), TG (r = 0.397; P = 0.001), fasting insulin (r = 0.534; P < 0.001) and C-peptide (r = 0.458; P < 0.001), and negatively with HDLC (r = 0.353; P = 0.004). In multiple regression analyses, BMI and the 2-h ACTH response to glucose were significant predictors of WHR and, in addition, the cortisol ratio, WHR, and BMI of insulin. CONCLUSIONS: Abdominal obesity may be associated with subtle central adrenal insufficiency, which might also affect insulin and lipoprotein metabolism.