Regional Adipose Tissue Metabolism in Postmenopausal Women After Treatment with Exogenous Sex Steroids Vila-Beth Lindberg, N. Crona, G. Silfverstolpe, P. Bjomtorp and M. Rebuffe-Scrive Department of Obstetrics and Gynecology, Medicine I, Sahlgrenska Hospital, University of Goteborg, Goteborg, Sweden
In order to evaluate the effect of exogenous sex steroids on adipose tissue metabolism, two groups of postmenopausal women were studied. In one of the groups, the effect of 50 |ig ethinyl estradiol (EE) was investigated given orally alone and in combination with 10 mg norethisterone acetate (NET). This combination is reminiscent of an old high dose oral contraceptive. In the other group, the effect of 3 mg 17p-estradiol was evaluated when administered percutaneously alone and in combination with 300 mg micronized progesterone given orally. These substances and doses were chosen to provide a "physiological" hormonal influence. In the femoral region 50 u.g EE induced an increase in LPL activity. This elevated LPL value was reversed with the addition of 10 mg NET. Moreover, during treatment with 50 ug EE, a decrease in norepinephrine stimulated lipolysis was seen in the abdominal region. The percutaneous administration of 17p-estradiol with or without micronized progesterone, however, was inert as regards subcutaneous adipose tissue metabolism. Our findings indicate, therefore, that EE in doses used in oral contraception might promote lipid accumulation in the femoral adipose tissue depot. Key words Estrogens — Progestogens — Femoral and Abdominal Adipocytes — Lipolysis — Lipoprotein Lipase
Introduction There are well-known differences between men and women in the regional distribution of the subcutaneous layer of adipose tissue. Young women tend to accumulate adipose tissue in the femoral region (Rebuffe-Scrive,
Horm. meta. Res. 22 (1990) 345 - 3 51 © Georg Thieme Verlag Stuttgart • New York
Enk, Crona, Lonnroth, Abrahamsson, Smith and Bjomtorp 1985), while men accumulate in the abdominal region instead {Krotkiewski, Bjomtorp, Sjostrom and Smith 1983). The profile of fat distribution undergoes modifications with aging. In postmenopausal women the pattern of adipose tissue distribution seems to become more like that of men (Rebuffe-Scrive, Lonnroth, Marin, Wesslau, Bjomtorp and Smith 1987). These age and sex related differences might at least partly reflect the activity of endogenous sex steroids. As far as oral contraceptives (OCs) currently in use are concerned, there is no evidence of any weight-increasing effects (Carpenter and Neinstein 1986). However, it is a common clinical experience that the earlier "high-dose" OCs caused an increase in weight in a substantial sub-group of treated women. Also, during the treatment of endometriosis with progestogens, as well as during treatment of tall girls prepubertally with high-dose estrogenic therapy, weight-increases can be a problem (Atares, Zachmann and Prader 1986). During hormonal replacement therapy (HRT) at ordinary doses, most women seem to be weight-stable. But, even when the body-weight goes unchanged, changes might occur in body composition. Signs of increasing muscle mass have been observed during high-dose HRT, although the bodyweight remained unchanged (Jensen, Christiansen and Redbro 1986). It has been speculated that an increase in weight during the administration of exogenous sex steroids might be mediated by their influence on the central nervous system. They might also act directly on peripheral adipose tissue. Little information is available on such effects and no longitudinal study concerning this problem has been performed. In order to investigate the effects of estrogens alone or in combination with progestogens on adipose tissue metabolism, two groups of postmenopausal women were studied during treatment with two different hormonal sequences. In one of the groups, the effect of 50 (ig ethinyl estradiol (EE) was studied when administered orally alone and in combination with 10 mg norethisterone (NET). In the other group the effect of 3 mg 17P-estradiol was evaluated when administered percutaneously alone and in combination with 300 mg micronized progesterone given orally.
Received: 18 Apr. 1989
Accepted: 17 Nov. 1989
Downloaded by: Universite Laval. Copyrighted material.
Summary
Horm. meta. Res. 22 (1990)
Ulla-Beth Lindberg, N. Crona, G. Silfverstolpe, P. Bjomtorp and M. Rebuffe-Scrive
Subjects—StudyDesign Two groups of women were engaged for the study. Group A consisted initially of 14 postmenopausal women but one woman was excluded because of liver disease. The remaining definitive group included 13 women had a mean age of 56.4 (range 53-59) years. Group B included from the beginning 9 postmenopausal and 3 perimenopausal women. However, one woman was excluded from this group also since the hormone assays indicated noncompliance. The definitive group had a mean age of 56.8 (range 50-62) years. There were 4 smokers in group A and 2 in group B. None were taking any drugs with known effects on the parameters studied, nor did any of the women change her diet during the study. The study was carried out with a longitudinal design. All women were without hormonal treatment for at least 4 weeks before the start of the study. Each woman was examined 3 times, before start of the hormonal treatment and after each treatment period. Blood samples were obtained and adipose tissue biopsies were taken after an overnight fast. The heights and weights of the women were recorded and their body mass index (BMI) was calculated as weight (kg)/height 2(m ). The waist/hip circumference ratio (WHR) was also calculated according to a standardized procedure. A waist circumference was measured between the costal arch and the iliac crest in a normal exspiratory position and the hip circumference over the widest part with the woman standing.
Group A (n = 13) Initially, 50 u.g ethinyl estradiol (EE)/day (Linoral, Organon, Oss, The Netherlands) was given orally for 3 weeks. Then for a further 3-week period, 10 mg norethisterone acetate (NET) (Primolut-Nor, Schering, Berlin, West-Germany) orally was included each day.
Group B(n = 11) These women were treated initially percutaneously with 5 g of an alcohol-water gel containing 3 mg of 17f)-estradiol/day (Oestrogel, Laboratoire Besins-Iscovesco, Paris, France) for 3 weeks. The gel was applied over the upper region of the body (e. g. shoulders, overarms, upper part of back). Then, for a further 3-week period, 300 mg of micronized progesterone (Utrogestan, Laboratoire Besins-Iscovesco, Paris, France) was given orally each day. The study was approved by the Ethical Committee of the University of Goteborg. The women were informed orally and in writing, and they gave their informed consent to participate.
Methods Blood samples were drawn from an antecubital vein in the overnight fasted state. Blood-glucose was measured with a glucose-oxidase method (GLOX, Kabi, Stockholm, Sweden) and plasma insulin by a radioimmunoassay (Phadebas, Pharmacia, Uppsala, Sweden). Serum triglycerides were determined using a commercial kit (TC 297771, Boehringer-Mannheim, Germany). Follicle stimulating hormone (FSH) was determined by a double-antibody polyethyleneglycol radioimmunoassay using the 2nd IRP HMG as calibrator (Diagnostic Products Corp., Los Angeles, CA, USA); sex-hormone binding globulin (SHBG) by immunoradiometric assay calibrated against a (3H)dihydrotestosterone-binding ammonium-sulphate precipitation assay (Farmos Diagnostica, Turku, Finland); progesterone by a nonextraction coated-tube radioimmunoassay with a 125-labeled ligand (Diagnostic Products Corp.) and 17P-estradiol by a nonextraction solid-phase (particles) radioimmunoassay with a ' 5I-labeled ligand (Baxter Merz-Dade AG, Dudingen, Switzerland). Needle-biopsies (about 500 mg) of adipose tissue were obtained from the subcutaneous depot in the femoral (just below
the trochanter) and the abdominal (lateral to the umbilicus) regions. These biopsies were taken under local anaesthesia of the skin with 2% lidocain (Astra, Sodertalje, Sweden). The fat cell size and the number of isolated cells were measured, as described by Smith, Sjostrom and Bjomtorp (1972). A mean fat cell surface area (SA) was calculated according to the formula (SA = n(S + SD 2), where d =- mean cell diameter and SD = standard deviation. Part of the adipose tissue was used for lipoprotein lipase (LPL) determinations following homogenization and acetoneether extraction according to the method described by Nilsson-Ehle, Tornqvist and Belfrage (1972). Lipolytic measurements were made on adipocytes isolated using collagenase. The suspension of adipocytes had a low lipocrit (1 —2 %), and was incubated for 2 hours at 37 °C in Parker medium 199 containing 4% bovine serum albumin with or without catecholamines or insulin added at the indicated concentrations. Cells and medium were separated through silicone oil {Gammeltoft and Gliemann 1973). The glycerol content of the medium was measured enzymatically (Laurell and Tibbling 1966) and taken as an index of lipolysis. Preliminary experiments where adenosine deaminase (0.5 U/l) was added to the incubation medium showed no changes either in basal or norepinephrine-stimulated lipolysis. All metabolic results are expressed per unit cell surface area to correct for differences in fat cell size.
Statistical methods Data were evaluated by Student's paired t-test. Site differences at each time-point were determined. As regards fat cell size, lipoprotein lipase activity and lipolysis, measurements were done at pretreatment, after estrogenic, and after combined estrogen-progestogen treatment. Comparisons were done within the femoral and within the abdominal regions. The differences between the abdominal and femoral values at each examination were also compared. Results Clinical GroupA
observations
Body-weight increased after the treatments with 50 |Xg EE and with 50 \ig EE + 10 mg N E T when compared with pretreatment values (P < 0.001). There was also a significant weight increase when comparing the effects of 50 jig EE with those of 50 u.g EE + 10 mg NET (P < 0.01) (Table 1). Group B Three mg of 17P-estradiol induced a significant weight increase compared with the pretreatment values (P < 0.05). During treatment with 50 ng EE + 10 mg NET, a decrease in blood-pressure was observed (P < 0.01); otherwise no other changes in blood-pressure were seen (not shown). Only minor side effects, such as light nausea or breast tension, were experienced in both groups. Serum levels of triglycerides, glucose and insulin (Table 2) GroupA Serum triglyceride concentration increased after both 50 ug EE and 50 ug EE + 10 mg NET compared
Downloaded by: Universite Laval. Copyrighted material.
346
Sex Steroids and Adipose Tissue Metabolism in Postmenopausal Women characteristics Group A (n = 13)
Group B (n = 11)
Age (years)
56.4 ±0.5
56.8 ±1.2
Height (cm)
166.3 ±1.5
165.7 ±1.9
WHR
0.90 ±0.02
0.92 ±0.02
BMI (kg/m2) Pretreatment First treatment period Second treatment period
26.1 ±0.8 26.6 ±0.8*** 26.7 ±0.8***
27.3 ±0.9 27.6 ±0.9* 27.6 ±0.9
Weight (kg) Pretreatment First treatment period Second treatment period Duration of menopause (years)
72.1 ±2.1 73.3 ±2.1*** 73.6 ±2.0*** 3.9 ±0.5
Group B No significant changes were recorded in these variables. Serum levels of I7$-estradiol, progesterone, SHBG and FSH (Table 3)
74.9 ±2.2 75.5 ±2.4* 75.5 ±2.4 5.5 ±0.6
+
Means ±SEM + This figure concerns 8 postmenopausal women - 3 perimenopausal women not included. Asterisks denote significant difference from pretreatment values. *P
In order to evaluate the effect of exogenous sex steroids on adipose tissue metabolism, two groups of postmenopausal women were studied. In one of the groups, the effect of 50 |ig ethinyl estradiol (EE) was investigated given orally alone and in combination with 10 mg norethisterone acetate (NET). This combination is reminiscent of an old high dose oral contraceptive. In the other group, the effect of 3 mg 17p-estradiol was evaluated when administered percutaneously alone and in combination with 300 mg micronized progesterone given orally. These substances and doses were chosen to provide a "physiological" hormonal influence. In the femoral region 50 u.g EE induced an increase in LPL activity. This elevated LPL value was reversed with the addition of 10 mg NET. Moreover, during treatment with 50 ug EE, a decrease in norepinephrine stimulated lipolysis was seen in the abdominal region. The percutaneous administration of 17p-estradiol with or without micronized progesterone, however, was inert as regards subcutaneous adipose tissue metabolism. Our findings indicate, therefore, that EE in doses used in oral contraception might promote lipid accumulation in the femoral adipose tissue depot. Key words Estrogens — Progestogens — Femoral and Abdominal Adipocytes — Lipolysis — Lipoprotein Lipase
Introduction There are well-known differences between men and women in the regional distribution of the subcutaneous layer of adipose tissue. Young women tend to accumulate adipose tissue in the femoral region (Rebuffe-Scrive,
Horm. meta. Res. 22 (1990) 345 - 3 51 © Georg Thieme Verlag Stuttgart • New York
Enk, Crona, Lonnroth, Abrahamsson, Smith and Bjomtorp 1985), while men accumulate in the abdominal region instead {Krotkiewski, Bjomtorp, Sjostrom and Smith 1983). The profile of fat distribution undergoes modifications with aging. In postmenopausal women the pattern of adipose tissue distribution seems to become more like that of men (Rebuffe-Scrive, Lonnroth, Marin, Wesslau, Bjomtorp and Smith 1987). These age and sex related differences might at least partly reflect the activity of endogenous sex steroids. As far as oral contraceptives (OCs) currently in use are concerned, there is no evidence of any weight-increasing effects (Carpenter and Neinstein 1986). However, it is a common clinical experience that the earlier "high-dose" OCs caused an increase in weight in a substantial sub-group of treated women. Also, during the treatment of endometriosis with progestogens, as well as during treatment of tall girls prepubertally with high-dose estrogenic therapy, weight-increases can be a problem (Atares, Zachmann and Prader 1986). During hormonal replacement therapy (HRT) at ordinary doses, most women seem to be weight-stable. But, even when the body-weight goes unchanged, changes might occur in body composition. Signs of increasing muscle mass have been observed during high-dose HRT, although the bodyweight remained unchanged (Jensen, Christiansen and Redbro 1986). It has been speculated that an increase in weight during the administration of exogenous sex steroids might be mediated by their influence on the central nervous system. They might also act directly on peripheral adipose tissue. Little information is available on such effects and no longitudinal study concerning this problem has been performed. In order to investigate the effects of estrogens alone or in combination with progestogens on adipose tissue metabolism, two groups of postmenopausal women were studied during treatment with two different hormonal sequences. In one of the groups, the effect of 50 (ig ethinyl estradiol (EE) was studied when administered orally alone and in combination with 10 mg norethisterone (NET). In the other group the effect of 3 mg 17P-estradiol was evaluated when administered percutaneously alone and in combination with 300 mg micronized progesterone given orally.
Received: 18 Apr. 1989
Accepted: 17 Nov. 1989
Downloaded by: Universite Laval. Copyrighted material.
Summary
Horm. meta. Res. 22 (1990)
Ulla-Beth Lindberg, N. Crona, G. Silfverstolpe, P. Bjomtorp and M. Rebuffe-Scrive
Subjects—StudyDesign Two groups of women were engaged for the study. Group A consisted initially of 14 postmenopausal women but one woman was excluded because of liver disease. The remaining definitive group included 13 women had a mean age of 56.4 (range 53-59) years. Group B included from the beginning 9 postmenopausal and 3 perimenopausal women. However, one woman was excluded from this group also since the hormone assays indicated noncompliance. The definitive group had a mean age of 56.8 (range 50-62) years. There were 4 smokers in group A and 2 in group B. None were taking any drugs with known effects on the parameters studied, nor did any of the women change her diet during the study. The study was carried out with a longitudinal design. All women were without hormonal treatment for at least 4 weeks before the start of the study. Each woman was examined 3 times, before start of the hormonal treatment and after each treatment period. Blood samples were obtained and adipose tissue biopsies were taken after an overnight fast. The heights and weights of the women were recorded and their body mass index (BMI) was calculated as weight (kg)/height 2(m ). The waist/hip circumference ratio (WHR) was also calculated according to a standardized procedure. A waist circumference was measured between the costal arch and the iliac crest in a normal exspiratory position and the hip circumference over the widest part with the woman standing.
Group A (n = 13) Initially, 50 u.g ethinyl estradiol (EE)/day (Linoral, Organon, Oss, The Netherlands) was given orally for 3 weeks. Then for a further 3-week period, 10 mg norethisterone acetate (NET) (Primolut-Nor, Schering, Berlin, West-Germany) orally was included each day.
Group B(n = 11) These women were treated initially percutaneously with 5 g of an alcohol-water gel containing 3 mg of 17f)-estradiol/day (Oestrogel, Laboratoire Besins-Iscovesco, Paris, France) for 3 weeks. The gel was applied over the upper region of the body (e. g. shoulders, overarms, upper part of back). Then, for a further 3-week period, 300 mg of micronized progesterone (Utrogestan, Laboratoire Besins-Iscovesco, Paris, France) was given orally each day. The study was approved by the Ethical Committee of the University of Goteborg. The women were informed orally and in writing, and they gave their informed consent to participate.
Methods Blood samples were drawn from an antecubital vein in the overnight fasted state. Blood-glucose was measured with a glucose-oxidase method (GLOX, Kabi, Stockholm, Sweden) and plasma insulin by a radioimmunoassay (Phadebas, Pharmacia, Uppsala, Sweden). Serum triglycerides were determined using a commercial kit (TC 297771, Boehringer-Mannheim, Germany). Follicle stimulating hormone (FSH) was determined by a double-antibody polyethyleneglycol radioimmunoassay using the 2nd IRP HMG as calibrator (Diagnostic Products Corp., Los Angeles, CA, USA); sex-hormone binding globulin (SHBG) by immunoradiometric assay calibrated against a (3H)dihydrotestosterone-binding ammonium-sulphate precipitation assay (Farmos Diagnostica, Turku, Finland); progesterone by a nonextraction coated-tube radioimmunoassay with a 125-labeled ligand (Diagnostic Products Corp.) and 17P-estradiol by a nonextraction solid-phase (particles) radioimmunoassay with a ' 5I-labeled ligand (Baxter Merz-Dade AG, Dudingen, Switzerland). Needle-biopsies (about 500 mg) of adipose tissue were obtained from the subcutaneous depot in the femoral (just below
the trochanter) and the abdominal (lateral to the umbilicus) regions. These biopsies were taken under local anaesthesia of the skin with 2% lidocain (Astra, Sodertalje, Sweden). The fat cell size and the number of isolated cells were measured, as described by Smith, Sjostrom and Bjomtorp (1972). A mean fat cell surface area (SA) was calculated according to the formula (SA = n(S + SD 2), where d =- mean cell diameter and SD = standard deviation. Part of the adipose tissue was used for lipoprotein lipase (LPL) determinations following homogenization and acetoneether extraction according to the method described by Nilsson-Ehle, Tornqvist and Belfrage (1972). Lipolytic measurements were made on adipocytes isolated using collagenase. The suspension of adipocytes had a low lipocrit (1 —2 %), and was incubated for 2 hours at 37 °C in Parker medium 199 containing 4% bovine serum albumin with or without catecholamines or insulin added at the indicated concentrations. Cells and medium were separated through silicone oil {Gammeltoft and Gliemann 1973). The glycerol content of the medium was measured enzymatically (Laurell and Tibbling 1966) and taken as an index of lipolysis. Preliminary experiments where adenosine deaminase (0.5 U/l) was added to the incubation medium showed no changes either in basal or norepinephrine-stimulated lipolysis. All metabolic results are expressed per unit cell surface area to correct for differences in fat cell size.
Statistical methods Data were evaluated by Student's paired t-test. Site differences at each time-point were determined. As regards fat cell size, lipoprotein lipase activity and lipolysis, measurements were done at pretreatment, after estrogenic, and after combined estrogen-progestogen treatment. Comparisons were done within the femoral and within the abdominal regions. The differences between the abdominal and femoral values at each examination were also compared. Results Clinical GroupA
observations
Body-weight increased after the treatments with 50 |Xg EE and with 50 \ig EE + 10 mg N E T when compared with pretreatment values (P < 0.001). There was also a significant weight increase when comparing the effects of 50 jig EE with those of 50 u.g EE + 10 mg NET (P < 0.01) (Table 1). Group B Three mg of 17P-estradiol induced a significant weight increase compared with the pretreatment values (P < 0.05). During treatment with 50 ng EE + 10 mg NET, a decrease in blood-pressure was observed (P < 0.01); otherwise no other changes in blood-pressure were seen (not shown). Only minor side effects, such as light nausea or breast tension, were experienced in both groups. Serum levels of triglycerides, glucose and insulin (Table 2) GroupA Serum triglyceride concentration increased after both 50 ug EE and 50 ug EE + 10 mg NET compared
Downloaded by: Universite Laval. Copyrighted material.
346
Sex Steroids and Adipose Tissue Metabolism in Postmenopausal Women characteristics Group A (n = 13)
Group B (n = 11)
Age (years)
56.4 ±0.5
56.8 ±1.2
Height (cm)
166.3 ±1.5
165.7 ±1.9
WHR
0.90 ±0.02
0.92 ±0.02
BMI (kg/m2) Pretreatment First treatment period Second treatment period
26.1 ±0.8 26.6 ±0.8*** 26.7 ±0.8***
27.3 ±0.9 27.6 ±0.9* 27.6 ±0.9
Weight (kg) Pretreatment First treatment period Second treatment period Duration of menopause (years)
72.1 ±2.1 73.3 ±2.1*** 73.6 ±2.0*** 3.9 ±0.5
Group B No significant changes were recorded in these variables. Serum levels of I7$-estradiol, progesterone, SHBG and FSH (Table 3)
74.9 ±2.2 75.5 ±2.4* 75.5 ±2.4 5.5 ±0.6
+
Means ±SEM + This figure concerns 8 postmenopausal women - 3 perimenopausal women not included. Asterisks denote significant difference from pretreatment values. *P
