- Email: [email protected]
Disproportional Body Growth in Female Estrogen Receptor-α-Inactivated Mice
Biochemical and Biophysical Research Communications 265, 569 –571 (1999) Article ID bbrc.1999.1711, available online at http://www.idealibrary.com on
Disproportional Body Growth in Female Estrogen Receptor-a-Inactivated Mice O. Vidal,* M. Lindberg,* L. Sa¨vendahl,† D. B. Lubahn,‡ E. M. Ritzen,† J. Å. Gustafsson,§ and C. Ohlsson* *Department of Internal Medicine, Division of Endocrinology, Sahlgrenska University Hospital, S-41345 Go¨teborg, Sweden; †Department of Woman and Child Health, Karolinska Institutet, Stockholm, Sweden; ‡Department of Biochemistry and Child Health, University of Missouri–Columbia, Columbia, Missouri; and §Department of Medical Nutrition, Karolinska Institute, Novum, Huddinge, S-14157, Sweden
Received October 13, 1999
Estrogens play an important role in the regulation of longitudinal bone growth in man, as demonstrated by recent descriptions of individuals with estrogen insensitivity or aromatase deficiency. Two estrogen receptors, ERa and ERb, have been cloned. The aim of the present study was to investigate the function of ERa in the regulation of body growth and skeletal growth. Adult female mice with inactivated ERa (ERa2/2) demonstrated an increased body weight compared with wild-type mice (114% of control). However, the length of the appendicular skeleton was decreased in adult ERa2/2 mice (femur 93% of control). In contrast, the axial skeleton was normal (crown–rump length 98% of control). The decreased growth of the appendicular skeleton was associated with decreased serum levels of IGF-I (77% of control), indicating that the GH/IGF-I axis may be involved in the decreased longitudinal bone growth seen in female ERa2/2 mice. © 1999 Academic Press
Estrogens play an important role in the regulation of longitudinal bone growth (1, 2). Cellular responses to estrogens are mediated by estrogen receptors (ER) which belong to the nuclear receptor super family. Until recently estrogens were believed to act via a single nuclear receptor, now denoted Estrogen Receptor-a (ERa). However, a second estrogen receptor (ERb) was cloned (3) and this finding has raised the question of the relative importance of estrogen receptor-a and -b in different target tissues. We and others have demonstrated that ERa as well as ERb is expressed in growth plate chondrocytes and osteoblasts, indicating a possible role for both receptor subtypes in the regulation of skeletal growth (4 –9). It has been reported that a man who was homozygous for an inactivating point mutation in the ERa gene failed to ossify his growth plates (2). Based on this
clinical finding it was assumed that ERa, at least in males, is critical for the regulation of longitudinal bone growth. No female patient with an inactivating ERa gene mutation has yet been reported. We have used female mice lacking functional ERa (ERa2/2) to study the effects of ERa on skeletal growth. MATERIALS AND METHODS Mouse strains. Adult wild-type (1/1) or homozygous (2/2) ERa2/2 sibling mice of the same genetic background (129/J/C57BL/ 6J) were used in the present study (10). PCR genotyping was performed as described earlier (10). The following primers were used: CTACGGCCAGTCGGGCAT and AGACCTGTAGAAGGCGGGAG for the wild-type gene, producing a 241 bp fragment. The primers TTCCACACACTTCATTCTCA and CTCCACTGGCCTCAAACACCTG were used for the knockout gene resulting in a 650 bp fragment. Mice were fed with standard pellets (Cat. No. 901) obtained from B&K Universal AB, Sollentuna, Sweden. X-ray and weight measurements. Measurements of lengths of femur, tibia, crown–rump, and diameter of the skull were performed at 31, 65, and 118 days of age. Body weights were measured at 17, 23, 31, 46, 53, 60, 65, 81, 101, and 118 days of age. Organ weights (heart, lung, liver, kidney, ovary, and uterus) were measured when the animals were killed at 118 days of age. IGF-I RIA. The IGF-I concentration in serum was determined by using RIA after acid ethanol extraction according to the manufacturer’s protocol (Nicols Institute Diagnostics, San Juan Capistrano, CA) in a single assay.
RESULTS Body Weight Both wt and ERa2/2 mice demonstrated a high weight gain during puberty (day 23– 44, Figs. 1 and 2). The prepubertal (day 17–23) and pubertal (day 23– 44) weight gains were similar in ERa2/2 and wt female mice. However, the adult (day 44 –118) weight gain was increased in the ERa2/2 compared with wt mice (Fig. 2). This affected final weight, which was greater in ERa2/2 mice (Fig. 1).
569
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 265, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Body weight in wild-type (WT, square) and ERa2/2 mice (circle) at different ages. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. The body weights in WT and ERa2/2 mice at different ages were first analysed by a 2-way analysis of variance (ANOVA). Thereafter, unpaired t-tests was used to determine differences between wt and ERa2/2 mice at each time point. *P , 0.05 ERa2/2 versus WT.
Dimensions of Bones and Serum IGF-I The growth of the appendicular as well as the axial skeleton was followed using repeated X-ray measurements. The lengths of femur and tibia were chosen as measurements of the appendicular skeleton while the crown–rump (CR) length was used as a measurement of the axial skeleton. The lengths of both femur and tibia were decreased in 118 days old ERa2/2 mice compared with wt mice (tibia 93 6 1%, femur 93 6 1% of control), while no significant effect was seen at 31 or 65 days of age (Fig. 3A). The CR length did not differ between ERa2/2 and wt mice (Fig. 3B). The ratio between the appendicular and axial skeleton (femur/CR length) was decreased in 118 days ERa2/2 mice com-
FIG. 3. Length of femur (A) and crown–rump (B) and femur/CR ratio (C) in wild-type mice (square) and ERa2/2 mice (circle) at 31, 65 and 118 days of age. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, **P , 0.01 versus WT, unpaired t-test.
FIG. 2. Weight gain/day in wild-type (black) and ERa2/2 mice (white) during prepubertal growth (17–23 days of age), pubertal growth (23– 44 days of age) and adult growth (44 –115 days of age). n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, versus WT, unpaired t-test.
pared with wt mice (Fig. 3C). The skull diameter was similar in ERa2/2 and wt mice (data not shown). Serum IGF-I levels were measured as a screening procedure to investigate whether or not the GH/IGF-I axis might be involved in the decreased growth of femur and tibia in ERa2/2 mice. Serum IGF-I levels
570
Vol. 265, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Organ Weights
Heart Lung Liver Kidney Ovary Uterus
WT
ERa2/2
0.509 6 0.017 0.806 6 0.114 4.380 6 0.066 1.218 6 0.039 0.054 6 0.010 0.360 6 0.044
0.433 6 0.041 0.882 6 0.074 4.374 6 0.163 1.253 6 0.057 0.059 6 0.014 0.148 6 0.011**
Note. Organ weights expressed as % of total body weight at 115 days of age. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, **P , 0.01 versus wild-type (WT), unpaired t-test.
were decreased by 23% in 118 days old ERa2/2 mice compared with wt mice (ERa2/2 186 6 18, wt 242 6 5 ng/ml, P , 0.05, t-test). Organ Weight The weights of several other organs were measured to see if the effect of ERa inactivation on the appendicular skeleton was organ specific (Table 1). The weights of lung, liver, kidneys and ovaries were unchanged in 118 days old ERa2/2 compared with wt mice (Table 1). A nonsignificant tendency to decrease in the weight of the heart was seen (215%, P 5 0.12, t-test). The weight of the uterus was much lower in ERa2/2 mice (Table 1).
weights were increased and no effect was seen on the relative weights of heart, lung, liver or kidneys, while the uterus weights were decreased. The cellular mechanism of action for estrogen in the skeleton is unclear. Both estrogen receptor-a and estrogen receptor-b are expressed in skeletal tissues (4–9), indicating direct effects of estrogen. However, it is well known that estrogens regulate the GH/IGF-I axis (11), suggesting indirect effects of estrogens on skeletal tissues via a regulation of the GH/IGF-I axis. We found that the decreased appendicular skeletal growth in the ERa2/2 mice was associated with decreased serum levels of IGF-I. It could therefore be speculated that some of the decreased appendicular skeletal growth observed in ERa2/2 mice is mediated via an inhibition of the GH/IGF-I axis. In conclusion, female ERa2/2 mice demonstrate a disproportional growth phenotype including increased body weight and decreased appendicular skeletal growth while axial skeletal growth is unchanged. The decreased appendicular growth was associated with reduced serum IGF-I levels, indicating that the GH/ IGF-I axis may be involved in the mechanism behind the growth phenotype in female ERa2/2 mice. ACKNOWLEDGMENTS This study was supported by the Swedish Medical Research Council, the Bergvall Foundation, the Lundberg Foundation, and the Swedish Association Against Rheumatic Disease, the Novo Nordisk Foundation.
DISCUSSION
REFERENCES
The endochondral bone formation in mice is different from man in that the growth plates do not fuse completely during sexual maturation. Therefore, it can be argued that estrogen receptors do not mediate growth plate fusion in mice, which seems to be the case in humans (2). However, after puberty, the mouse growth plate diminishes considerably in thickness and the chondrocytes synthesise less protein, indicating that sex steroids might be of importance for the regulation of the growth cartilage in mice as well. Our present results demonstrate that estrogen receptor-a is predominantly involved in the regulation of appendicular but not axial skeletal growth in female mice. The earlier report of a man with an inactivating point mutation in the ERa gene had delayed epiphyseal fusion, resulting in tall stature, predominantly due to increased appendicular growth (eunuchoid habitus) (2) while the female ERa2/2 mice demonstrated decreased appendicular skeletal growth. Thus, the direction of the effect of ERa on the regulation of the growth of the appendicular skeleton seems to be species specific and this difference might be explained by the fact that growth plates in humans but not mice fuse after puberty. The inhibitory effect on the appendicular skeletal growth in female ERa2/2 mice was specific as their body
1. Turner, R. T., Riggs, B. L., and Spelsberg, T. C. (1994) Endocr. Rev. 15, 275–300. 2. Smith, E. P., Boyd, J., Frank, G. R., Takahashi, H., Cohen, R. M., Specker, B., Williams, T. C., Lubahn, D. B., and Korach, K. S. (1994) N. Engl. J. Med. 331, 1056 –1061. 3. Kuiper, G. G., Enmark, E., Pelto-Huikko, M., Nilsson, S., and Gustafsson, J. A. (1996) Proc. Natl. Acad. Sci. USA 93, 5925–5930. 4. Ben-Hur, H., Mor, G., Blickstein, I., Likhman, I., Kohen, F., Dgani, R., Insler, V., Yaffe, P., and Ornoy, A. (1993) Calcif. Tissue Int. 53, 91–96. 5. Ben-Hur, H., Thole, H. H., Mashiah, A., Insler, V., Berman, V., Shezen, E., Elias, D., Zuckerman, A., and Ornoy, A. (1997) Calcif. Tissue Int. 60, 520 –526. 6. Onoe, Y., Miyaura, C., Ohta, H., Nozawa, S., and Suda, T. (1997) Endocrinology 138, 4509 – 4512. 7. Arts, J., Kuiper, G. G., Janssen, J. M., Gustafsson, J. A., Lowik, C. W., Pols, H. A., and Van Leeuwen, J. P. (1997) Endocrinology 138, 5067–5070. 8. Nilsson, L. O., Boman, A., Savendahl, L., Grigelioniene, G., Ohlsson, C., Ritzen, E. M., and Wroblewski, J. (1999) J. Clin. Endocrinol. Metab. 84, 370 –373. 9. Vidal, O., Kindblom, L. G., and Ohlsson, C. (1999) J. Bone Miner. Res. 14, 923–929. 10. Lubahn, D. B., Moyer, J. S., Golding, T. S., Couse, J. F., Korach, K. S., and Smithies, O. (1993) Proc. Natl. Acad. Sci. USA 90, 11162–11166. 11. Jansson, J. O., Eden, S., and Isaksson, O. (1985) Endocr. Rev. 6, 128 –150.
571
Disproportional Body Growth in Female Estrogen Receptor-a-Inactivated Mice O. Vidal,* M. Lindberg,* L. Sa¨vendahl,† D. B. Lubahn,‡ E. M. Ritzen,† J. Å. Gustafsson,§ and C. Ohlsson* *Department of Internal Medicine, Division of Endocrinology, Sahlgrenska University Hospital, S-41345 Go¨teborg, Sweden; †Department of Woman and Child Health, Karolinska Institutet, Stockholm, Sweden; ‡Department of Biochemistry and Child Health, University of Missouri–Columbia, Columbia, Missouri; and §Department of Medical Nutrition, Karolinska Institute, Novum, Huddinge, S-14157, Sweden
Received October 13, 1999
Estrogens play an important role in the regulation of longitudinal bone growth in man, as demonstrated by recent descriptions of individuals with estrogen insensitivity or aromatase deficiency. Two estrogen receptors, ERa and ERb, have been cloned. The aim of the present study was to investigate the function of ERa in the regulation of body growth and skeletal growth. Adult female mice with inactivated ERa (ERa2/2) demonstrated an increased body weight compared with wild-type mice (114% of control). However, the length of the appendicular skeleton was decreased in adult ERa2/2 mice (femur 93% of control). In contrast, the axial skeleton was normal (crown–rump length 98% of control). The decreased growth of the appendicular skeleton was associated with decreased serum levels of IGF-I (77% of control), indicating that the GH/IGF-I axis may be involved in the decreased longitudinal bone growth seen in female ERa2/2 mice. © 1999 Academic Press
Estrogens play an important role in the regulation of longitudinal bone growth (1, 2). Cellular responses to estrogens are mediated by estrogen receptors (ER) which belong to the nuclear receptor super family. Until recently estrogens were believed to act via a single nuclear receptor, now denoted Estrogen Receptor-a (ERa). However, a second estrogen receptor (ERb) was cloned (3) and this finding has raised the question of the relative importance of estrogen receptor-a and -b in different target tissues. We and others have demonstrated that ERa as well as ERb is expressed in growth plate chondrocytes and osteoblasts, indicating a possible role for both receptor subtypes in the regulation of skeletal growth (4 –9). It has been reported that a man who was homozygous for an inactivating point mutation in the ERa gene failed to ossify his growth plates (2). Based on this
clinical finding it was assumed that ERa, at least in males, is critical for the regulation of longitudinal bone growth. No female patient with an inactivating ERa gene mutation has yet been reported. We have used female mice lacking functional ERa (ERa2/2) to study the effects of ERa on skeletal growth. MATERIALS AND METHODS Mouse strains. Adult wild-type (1/1) or homozygous (2/2) ERa2/2 sibling mice of the same genetic background (129/J/C57BL/ 6J) were used in the present study (10). PCR genotyping was performed as described earlier (10). The following primers were used: CTACGGCCAGTCGGGCAT and AGACCTGTAGAAGGCGGGAG for the wild-type gene, producing a 241 bp fragment. The primers TTCCACACACTTCATTCTCA and CTCCACTGGCCTCAAACACCTG were used for the knockout gene resulting in a 650 bp fragment. Mice were fed with standard pellets (Cat. No. 901) obtained from B&K Universal AB, Sollentuna, Sweden. X-ray and weight measurements. Measurements of lengths of femur, tibia, crown–rump, and diameter of the skull were performed at 31, 65, and 118 days of age. Body weights were measured at 17, 23, 31, 46, 53, 60, 65, 81, 101, and 118 days of age. Organ weights (heart, lung, liver, kidney, ovary, and uterus) were measured when the animals were killed at 118 days of age. IGF-I RIA. The IGF-I concentration in serum was determined by using RIA after acid ethanol extraction according to the manufacturer’s protocol (Nicols Institute Diagnostics, San Juan Capistrano, CA) in a single assay.
RESULTS Body Weight Both wt and ERa2/2 mice demonstrated a high weight gain during puberty (day 23– 44, Figs. 1 and 2). The prepubertal (day 17–23) and pubertal (day 23– 44) weight gains were similar in ERa2/2 and wt female mice. However, the adult (day 44 –118) weight gain was increased in the ERa2/2 compared with wt mice (Fig. 2). This affected final weight, which was greater in ERa2/2 mice (Fig. 1).
569
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 265, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Body weight in wild-type (WT, square) and ERa2/2 mice (circle) at different ages. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. The body weights in WT and ERa2/2 mice at different ages were first analysed by a 2-way analysis of variance (ANOVA). Thereafter, unpaired t-tests was used to determine differences between wt and ERa2/2 mice at each time point. *P , 0.05 ERa2/2 versus WT.
Dimensions of Bones and Serum IGF-I The growth of the appendicular as well as the axial skeleton was followed using repeated X-ray measurements. The lengths of femur and tibia were chosen as measurements of the appendicular skeleton while the crown–rump (CR) length was used as a measurement of the axial skeleton. The lengths of both femur and tibia were decreased in 118 days old ERa2/2 mice compared with wt mice (tibia 93 6 1%, femur 93 6 1% of control), while no significant effect was seen at 31 or 65 days of age (Fig. 3A). The CR length did not differ between ERa2/2 and wt mice (Fig. 3B). The ratio between the appendicular and axial skeleton (femur/CR length) was decreased in 118 days ERa2/2 mice com-
FIG. 3. Length of femur (A) and crown–rump (B) and femur/CR ratio (C) in wild-type mice (square) and ERa2/2 mice (circle) at 31, 65 and 118 days of age. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, **P , 0.01 versus WT, unpaired t-test.
FIG. 2. Weight gain/day in wild-type (black) and ERa2/2 mice (white) during prepubertal growth (17–23 days of age), pubertal growth (23– 44 days of age) and adult growth (44 –115 days of age). n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, versus WT, unpaired t-test.
pared with wt mice (Fig. 3C). The skull diameter was similar in ERa2/2 and wt mice (data not shown). Serum IGF-I levels were measured as a screening procedure to investigate whether or not the GH/IGF-I axis might be involved in the decreased growth of femur and tibia in ERa2/2 mice. Serum IGF-I levels
570
Vol. 265, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Organ Weights
Heart Lung Liver Kidney Ovary Uterus
WT
ERa2/2
0.509 6 0.017 0.806 6 0.114 4.380 6 0.066 1.218 6 0.039 0.054 6 0.010 0.360 6 0.044
0.433 6 0.041 0.882 6 0.074 4.374 6 0.163 1.253 6 0.057 0.059 6 0.014 0.148 6 0.011**
Note. Organ weights expressed as % of total body weight at 115 days of age. n 5 5 for WT and n 5 6 for ERa2/2 mice. Values are given as means 6 SEM. *P , 0.05, **P , 0.01 versus wild-type (WT), unpaired t-test.
were decreased by 23% in 118 days old ERa2/2 mice compared with wt mice (ERa2/2 186 6 18, wt 242 6 5 ng/ml, P , 0.05, t-test). Organ Weight The weights of several other organs were measured to see if the effect of ERa inactivation on the appendicular skeleton was organ specific (Table 1). The weights of lung, liver, kidneys and ovaries were unchanged in 118 days old ERa2/2 compared with wt mice (Table 1). A nonsignificant tendency to decrease in the weight of the heart was seen (215%, P 5 0.12, t-test). The weight of the uterus was much lower in ERa2/2 mice (Table 1).
weights were increased and no effect was seen on the relative weights of heart, lung, liver or kidneys, while the uterus weights were decreased. The cellular mechanism of action for estrogen in the skeleton is unclear. Both estrogen receptor-a and estrogen receptor-b are expressed in skeletal tissues (4–9), indicating direct effects of estrogen. However, it is well known that estrogens regulate the GH/IGF-I axis (11), suggesting indirect effects of estrogens on skeletal tissues via a regulation of the GH/IGF-I axis. We found that the decreased appendicular skeletal growth in the ERa2/2 mice was associated with decreased serum levels of IGF-I. It could therefore be speculated that some of the decreased appendicular skeletal growth observed in ERa2/2 mice is mediated via an inhibition of the GH/IGF-I axis. In conclusion, female ERa2/2 mice demonstrate a disproportional growth phenotype including increased body weight and decreased appendicular skeletal growth while axial skeletal growth is unchanged. The decreased appendicular growth was associated with reduced serum IGF-I levels, indicating that the GH/ IGF-I axis may be involved in the mechanism behind the growth phenotype in female ERa2/2 mice. ACKNOWLEDGMENTS This study was supported by the Swedish Medical Research Council, the Bergvall Foundation, the Lundberg Foundation, and the Swedish Association Against Rheumatic Disease, the Novo Nordisk Foundation.
DISCUSSION
REFERENCES
The endochondral bone formation in mice is different from man in that the growth plates do not fuse completely during sexual maturation. Therefore, it can be argued that estrogen receptors do not mediate growth plate fusion in mice, which seems to be the case in humans (2). However, after puberty, the mouse growth plate diminishes considerably in thickness and the chondrocytes synthesise less protein, indicating that sex steroids might be of importance for the regulation of the growth cartilage in mice as well. Our present results demonstrate that estrogen receptor-a is predominantly involved in the regulation of appendicular but not axial skeletal growth in female mice. The earlier report of a man with an inactivating point mutation in the ERa gene had delayed epiphyseal fusion, resulting in tall stature, predominantly due to increased appendicular growth (eunuchoid habitus) (2) while the female ERa2/2 mice demonstrated decreased appendicular skeletal growth. Thus, the direction of the effect of ERa on the regulation of the growth of the appendicular skeleton seems to be species specific and this difference might be explained by the fact that growth plates in humans but not mice fuse after puberty. The inhibitory effect on the appendicular skeletal growth in female ERa2/2 mice was specific as their body
1. Turner, R. T., Riggs, B. L., and Spelsberg, T. C. (1994) Endocr. Rev. 15, 275–300. 2. Smith, E. P., Boyd, J., Frank, G. R., Takahashi, H., Cohen, R. M., Specker, B., Williams, T. C., Lubahn, D. B., and Korach, K. S. (1994) N. Engl. J. Med. 331, 1056 –1061. 3. Kuiper, G. G., Enmark, E., Pelto-Huikko, M., Nilsson, S., and Gustafsson, J. A. (1996) Proc. Natl. Acad. Sci. USA 93, 5925–5930. 4. Ben-Hur, H., Mor, G., Blickstein, I., Likhman, I., Kohen, F., Dgani, R., Insler, V., Yaffe, P., and Ornoy, A. (1993) Calcif. Tissue Int. 53, 91–96. 5. Ben-Hur, H., Thole, H. H., Mashiah, A., Insler, V., Berman, V., Shezen, E., Elias, D., Zuckerman, A., and Ornoy, A. (1997) Calcif. Tissue Int. 60, 520 –526. 6. Onoe, Y., Miyaura, C., Ohta, H., Nozawa, S., and Suda, T. (1997) Endocrinology 138, 4509 – 4512. 7. Arts, J., Kuiper, G. G., Janssen, J. M., Gustafsson, J. A., Lowik, C. W., Pols, H. A., and Van Leeuwen, J. P. (1997) Endocrinology 138, 5067–5070. 8. Nilsson, L. O., Boman, A., Savendahl, L., Grigelioniene, G., Ohlsson, C., Ritzen, E. M., and Wroblewski, J. (1999) J. Clin. Endocrinol. Metab. 84, 370 –373. 9. Vidal, O., Kindblom, L. G., and Ohlsson, C. (1999) J. Bone Miner. Res. 14, 923–929. 10. Lubahn, D. B., Moyer, J. S., Golding, T. S., Couse, J. F., Korach, K. S., and Smithies, O. (1993) Proc. Natl. Acad. Sci. USA 90, 11162–11166. 11. Jansson, J. O., Eden, S., and Isaksson, O. (1985) Endocr. Rev. 6, 128 –150.
571