- Email: [email protected]
A unique and essential role for insulin in the phenotypic expression of rat mammary epithelial cells unrelated to its function in cell maintenance
Biochimica et Biophysica Acta, 763 (1983) 309-314
309
Elsevier BBA 11216
A U N I Q U E AND ESSENTIAL ROLE FOR INSULIN IN T H E P H E N O T Y P I C E X P R E S S I O N OF RAT M A M M A R Y E P I T H E L I A L C E L L S U N R E L A T E D T O I T S F U N C T I O N IN CELL M A I N T E N A N C E KEVIN R. NICHOLAS, LAKSHMANAN SANKARAN and YALE J. TOPPER Laboratory of Biochemistry and Metabolism, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20205 (U.S.A.)
(Received February 25th, 1983) (Revised manuscript received June 10th, 1983)
Key words: Insulin," Growth factor," Phenotypic expression," (Rat epithelial cell)
Mammary explants from pregnant rats can be induced in regard to casein synthesis and a-lactalbumin activity when cultured in the presence of hydrocortisone, prolactin and levels of insulin approaching physiological concentrations. No detectable induction occurs in the absence of insulin. Although epidermal growth factor and multiplication stimulating activity, in the presence of hydrocortisone, can maintain the initial level of NADH-cytochrome c reductase as well as insulin, neither can substitute effectively for insulin in the induction of the milk proteins. Proinsulin, nerve growth factor, platelet-derived growth factor and fibroblast growth factor are also ineffective substitutes for insulin in this regard. Whereas prolonged tissue exposure to multiplication stimulating activity, hydrocortisone and prolactin does not result in induction of a-lactalbumin activity, subsequent addition of insulin leads to prompt response. The results suggest that the ability of insulin to function as a unique, essential factor in the induction of rat milk proteins is independent of its ceil-maintenance activity. Thus, in addition to its well established functions in metabolic processes, insulin appears to play a vital role in certain developmental processes.
Introduction It has been shown that insulin, glucocorticoid and prolactin are essential for the induction of the synthesis of phosphorylated casein by m a m m a r y explants from intact mice [1,2]. It was conceivable that fewer hormones might be required for the accumulation of mouse casein m R N A . However, it is now agreed that glucocorticoid, in addition to prolactin, is necessary for the accumulation of casein m R N A in mouse m a m m a r y explants [3,4]. Although insulin was routinely added to the cultures in these studies, a possible intimate role for
Abbreviation: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. 0167-4889/83/$03.00 © 1983 Elsevier Science Publishers B.V.
this hormone in the phenotypic expression of m a m m a r y cells has not been considered [5]. Rather, insulin has been regarded only as an agent involved in cell maintenance [6,7]. Recently, it was reported [8] that, although insulin, epidermal growth factor and somatomedin c can function equally well as cell maintenance agents, only insulin, together with cortisol and prolactin, can induce casein m R N A accumulation in mouse m a m m a r y tissue in vitro. On this basis, it was suggested that insulin may be essential for the expression of certain genes related to cell differentiation, in addition to serving as a metabolic and growth factor. Since a number of species differences relating to the hormonal requirements for m a m m a r y development are recognized [9], it was of interest to
310 investigate the role of insulin in the phenotypic expression of rat mammary epithelial cells in vitro. In this study, a number of additional growth factors have been compared with insulin in relation to the induction of both casein synthesis and alactalbumin activity. Moreover, the previous evidence [8] which indicated that insulin is not a mere cell-maintenance factor in the mammary explant system has been extended. Evidence that insulin may be vital in certain other developmental systems, as well, will be discussed. Materials and Methods
Materials Ovine prolactin (NIH-PRL-14 and 15) was a gift from the Hormone Distribution Program, NIADDK, NIH, Bethesda, MD, and crystalline porcine insulin was a gift from Eli Lilly Co. (Indianapolis, IN). Multiplication stimulating activity, epidermal growth factor (receptor grade), nerve growth factor (2.5 S), fibroblast growth factor and platelet-derived growth factor were purchased from Collaborative Research, Inc. (Waltham, MA). Proinsulin, Lot 605-70N-279-1, was a gift kindly provided by Dr. F. DePablo, NIADDK. Hydrocortisone and cytochrome c (equine heart, salt-free) were purchased from Calbiochem (La JoUa, CA). Medium 199 containing Hanks' salts was from Grand Island Biological Company (Grand Island, NY). Hepes, NADH, bovine serum albumin (crystallized and lyophilized), uridine-5'-diphosphogalactose, uridine-5'-triphosphate, bovine milk galactosyltransferase (4.5 units/mg), penicillin G and bovine a-lactalbumin (Grade II) were from Sigma Chemical Co. (St. Louis, MO). Type I collagenase (204 units/mg) was purchased from Millipore Corporation (Freehold, N J). Uridine diphospho-D-[6-3H]galactose (7 Ci/mmol)was from Amersham Corporation (Arlington Heights, IL) and 3H-labeled L-amino acid mixture and Triton X-100 were from New England Nuclear (Boston, MA). Rabbit anti-sheep IgG was obtained from Miles Biochemical (Elkhart; IN). Tissue culture 3-4-month-old pregnant (13-14 days of gestation) Sprague-Dawley rats (Taconic Farms, Germantown, NY) were used. Explants were prepared
from the abdominal mammary glands and cultured in Medium 199 as described previously [10]. Hormones were added to the media at the concentrations indicated. Growth factors, obtained as sterile, lyophilized preparations, were dissolved in sterile media and added aseptically. Tissue analysis Casein synthesis was measured in homogenates of explants cultured for three days by immunoprecipitation [11] with sheep anti-rat casein following a 16-h pulse with 3H-labeled amino acids (20 /~Ci/ml). The immunoprecipitation method can detect both unphosphorylated and phosphorylated caseins, a-Lactalbumin activity was measured in homogenates by determining lactose formation in the presence of galactosyltransferase, as described previously [ 10]. NADH-cytochrome c reductase was measured [12] in epithelial cell-enriched fractions following collagenase digestion of mammary gland explants [13]. Results
An insulin dose-response curve in relation to the induction of a-lactalbumin activity is shown in Fig. 1. No response occurs in the absence of insulin. However, concentrations of the hormone approaching the physiological level (a few ng/ml), particularly in the presence of bovine serum albumin, do evoke response. Some loss of insulin by adsorption is presumably prevented by bovine serum albumin. Similar results were obtained with casein (data not shown). A comparison of six factors for their ability to support the induction of casein synthesis is depicted in Table I. Each was tested at 0.25/~g/ml, a concentration considerably above their physiological level [14-18]; the blood level of fibroblast growth factor has not been reported. Appreciable induction occurs only in the presence of insulin. Proinsulin and multiplication stimulating activity are less than 10% as active as insulin. Multiplication stimulating activity and epidermal growth factor had no more activity when tested at a concentration of 1 ~g/ml (data not shown). Similar results were obtained in terms of a-lactalbumin; in this case, nerve growth factor was also tested, and had virtually no activity (data not shown).
311
12
==
z
-.J
0
0
10
20
30
40
,~0
INSULIN CONCENTRATION (ng/ml) Fig. 1. Effect of insulin concentration on a-lactalbumin activity in pregnant rat mammary gland explants. Tissue was cultured for 96 h in media containing hydrocortisone (0.05 /zg/ml), prolactin (1 /~g/ml) and insulin (concentrations indicated) in the presence (1 m g / m l ) (O) or absence (O) of bovine serum albumin, a-Lactalbumin was measured in tissue extracts. A indicates the a-lactalbumin activity in the mammary gland prior to culture. Each value represents mean-FS.E, for three rats.
In an effort to determine whether the unique ability of insulin to support the induction of alactalbumin activity and casein synthesis is related to a unique ability of the hormone to maintain the mammary epithelium in a viable state, two types of experiment were performed. In the first (Table II), insulin, multiplication stimulating activity and epidermal growth factor were compared in relation to NADH-cytochrome c reductase. Hydrocortisone alone does not maintain the initial activity of the membrane-associated enzyme. However, the activity is maintained when any one of the factors is added together with hydrocortisone. The second experimental approach was to determine whether multiplication stimulating activity, a factor which has very little ability to support induction of casein synthesis or a-lactalbumin activity, can nevertheless sustain the cells' potential for response to insulin. The time-course for induction of a-lactalbumin in the presence of insulin, hydrocortisone and prolactin is shown in Fig. 2A. After an initial 24-h lag period, induction begins during the second 24 h. The increment produced during the second day is shown in Fig. 2B, which also demonstrates that although multiplication-
TABLE I
TABLE II
COMPARISON OF INSULIN, PROINSULIN AND SEVERAL GROWTH FACTORS FOR THEIR ABILITY TO I N D U C E CASEIN SYNTHESIS IN P R E G N A N T RAT M A M M A R Y G L A N D EXPLANTS
CAPACITY OF INSULIN, MULTIPLICATION STIMULATING ACTIVITY AND EPIDERMAL GROWTH FACTOR TO MAINTAIN N A D H - C Y T O C H R O M E c REDUCATSE ACTIVITY IN P R E G N A N T RAT M A M M A R Y G L A N D EXPLANTS
Explants from 14-day-pregnant rats were cultured for 72 h in medium containing hydrocortisone and prolactin (both at 1 /.tg/ml), and either insulin, proinsulin, multiplication stimulating activity (MSA), epidermal growth factor (EGF), fibroblast growth factor (FGF) or platelet-derived growth factor (PDGF), all at 0.25/~g/rnl. The explants were pulsed with 20 # C i / m l of 3H-labeled L-amino acid mixture during the last 16 h of incubation and the amount of casein formed was measured by immunoprecipitation using sheep anti-rat casein serum. Values for casein synthesis are cpm ( × 1 0 - 2 ) / h per mg tissue and are + S.E. for the number of rats shown, each analyzed separately.
Tissue was cultured for 96 h in media containing hydrocortisone (0.05 /tg/ml) alone, or a combination of hydrocortisone and either insulin, multiplication stimulating activity (MSA) or epidermal growth factor (EGF) (each at a concentration of 0.25 /~g/ml). NADH-cytochrome c reductase was measured [12] in epithelial cell-enriched fractions following collagenase digestion of the explants [13]. Values represent mean-I-S.E, with the number of observations shown in parentheses. For uncultured and epidermal growth factor+hydrocortisone systems, actual values from two observations are given.
Addition
Casein synthesis
Number of rats
Culture conditions
Insulin Proinsulin MSA EGF FGF PDGF
12.7±0.9 1.2±0.5 0.8±0.1 0.3±0.1 0.3±0.1 0.1±0.05
12 12 6 3 3 3
NADH-cytochrome c reductase ( A A s s 0 / m i n per g tissue)
Uncultured Hydrocortisone Insulin + hydrocortisone MSA + hydrocortisone E G F + hydrocortisone
0.297, 0.457 0.119 + 0.01 (5) 0.400 +_0.07 (4) 0.373 + 0.03 (5) 0.324, 0.364
312
30
A
2O ".= 10
>I.,> I.-
0
24
48
72
96
HOURS IN CULTURE 7
z m .J
5
~- 4
lIIl
l
1ii
MSAFP2MSAFP ~, 72MSAFP ~ 72 MSAFP
24
IFP2,
MSAFP,
24
IFP24 '~'
24
CULTURE CONDITIONS
Fig. 2. Capacity of multiplication stimulating activity to maintain hormone-responsive mammary epithelial cells. (A) Explants from pregnant rat mammary glands were cultured in media containing insulin (0.25 #g/ml), hydrocortisone (0.05 /~g/ml) and prolactin (1/~g/ml), and the a-lactalbumin activity in tissue extracts determined at daily intervals. Each value represents mean±S.E, for three groups of rats (two rats per group). (B) Explants were cultured in media containing 0.05 #g/m1 hydrocortisone (F), 1 #g/ml prolactin (P) and 0.25 #g/ml multiplication stimulating activity (MSA) for 72 h. Fresh medium with or without 0.25 #g/ml insulin (I) was added and cultures incubated for a further 24 h. a-Lactalbumin activity was determined in tissue extracts. The values shown for the system without insulin correspond to those in A. Each value is the mean ± S.E. for three groups of rats (two rats per group).
stimulating activity, together with hydrocortisone and prolactin, elicits almost no response during 4 days of culture, the tissue does respond when insulin is added on the 4th day of this culture. Discussion
The intent of this study is to determine whether or not insulin, in addition to its well established
general metabolic roles, m a y also serve as a specific, essential factor in the phenotypic expression of certain cell types. The h o r m o n a l induction, in vitro, of casein synthesis and a-lactalbumin activity in the rat m a m m a r y epithelial cell is the model selected for this inquiry. Although insulin, in addition to glucocorticoid and prolactin, has been f o u n d to be a necessary c o m p o n e n t of the culture media used in studies of mouse m a m m a r y develo p m e n t [ 1], a possible specific role for the hormone in the terminal differentiation of these cells has not been considered [5]. Rather, insulin has been thought to serve only a cell-maintenance function in this system [6,7]. Establishment of a more direct, unique role for insulin in the expression of the m a m m a r y phenotype would require that three criteria be met. (1) The h o r m o n e should be effective at a level approaching the physiological concentration. (2) Other factors should be incapable of supplanting insulin in regard to the induction of milk proteins. (3) At least one of these factors should be able to maintain the cells in the presence of glucocorticoid and prolactin. The first criterion is satisfied by the dose-response curves relating insulin to the induction of a-lactalbumin activity and casein synthesis. N o response occurs in either instance unless insulin is present, in addition to glucocorticoid and prolactin. Increments are manifested in both cases when insulin is added at a level which approaches the physiological concentration. N o n e of the other factors tested appears capable of substituting for insulin in the induction of casein synthesis or a-lactalbumin activity. It is conceivable that multiplication stimulating activity at higher levels might exert more activity via the insulin receptor. The relative inactivity of proinsulin suggests that the effects elicited by insulin involve interaction between the h o r m o n e and its own receptor. The results satisfy the second criterion stated above. The design of the experiments represented in Table II addresses the third criterion, i.e., do factors which cannot p r o m o t e the induction of alactalbumin activity and casein synthesis nervetheless maintain the cells as well as insulin? It is clear that glucocorticoid alone cannot maintain the initial level of N A D H - c y t o c h r o m e c reductase, but
313
that both multiplication stimulating activity and epidermal growth factor, in the presence of hydrocortisone, are as effective as insulin in this respect. The results also indicate that the two factors are as effective as insulin in maintaining cellular responsiveness to hydrocortisone [12]. A much more direct experimental approach toward satisfying the third criterion is represented in Fig. 2B. It is apparent that although multiplication-stimulating activity, in the presence of glucocorticoid and prolactin, has little ability to promote the induction of a-lactalbumin activity, it maintains, during 3 days of culture, the cells' potential for prompt response to the delayed addition of insulin. More specifically, the presence of insulin for 1 day following 3 days with multiplication stimulating activity, glucocorticoid and prolactin, resulted in approximately as much induction as that which occurred during a 1-day period in a system containing insulin from the start. The emergence of ct-lactalbumin activity upon the delayed addition of insulin, observed here, is similar phenomenologically to the emergence of casein synthesis upon the delayed addition of prolactin to a system previously cultured with insulin and glucocorticoid [19]. In one instance, it might be tempting to regard insulin as 'the' lactogenic hormone, while in the other, prolactin might be so regarded. Actually, both are required, and it is not possible at this time to assign greater importance to one or the other. It is clear that while insulin is not unique in terms of cell maintenance in the explant system, it does appear to have an essential and unique role in the phenotypic expression of the rat mammary epithelial cell. In this respect, the rat and mouse [8] are similar, while the rabbit mammary epithelial cell does not appear to require insulin for its phenotypic expression [9]. Insulin may have a role in the phenotypic expression of other cell types also. The pancreas of diabetic rats manifests a selective reduction in the level of m R N A for amylase [20] and the liver of mildly diabetic rats shows a selective reduction in the level of mRNA for t~2~ globulin [21]. It is not clear, however, whether the defects are the result of deprivation of insulin per se, or of the systemic derangements of diabetes. Lipoprotein lipase activity in 3T3-L1 fatty fibroblasts is stimulated to
maximum levels after 4 days of culture in the presence of insulin [22]. In addition, insulin can effect the determination of pre-adipocytes [23]. It will be of interest to learn whether the hormone is effective at physiological levels, and whether other growth factors that can function in cell maintenance are active in the determination of preadipocytes. These considerations would be pertinent to any system which appears to require insulin for development in vitro. The mammary epithelial cells in mature, nonpregnant animals are actually in an immature functional state. Unlike most other cells in the mammal, they achieve their potential only during pregnancy and lactation. Insulin, as we have seen, is one of the hormones required for this maturation of the murine mammary cell. The foregoing discussion raises the possibility that, in addition to its metabolic functions, insulin may also serve as a 'developmental' hormone which is essential for selective gene expression in certain cells at particular stages of ontogeny. References 1 Juergens, W.G., Stockdale, F.E., Topper, Y.J. and Elias, J.J. (1965) Proc. Natl. Acad. Sci. U.S.A. 54, 629-634 2 Turkington, R.W., Juergens, W.G. and Topper, Y.J. (1965) Biochim. Biophys. Acta 111,573-576 3 Nagaiah, K., Bolander, F.F., Nicholas, K.R., Takemoto, T. and Topper, Y.J. (1981) Biochem. Biophys. Res. Commun. 98, 380-387 4 Ganguly, R., Ganguly, N., Mehta, N.M. and Banerjee, M.R. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 6003-6006 5 Banerjee, M.R., Ganguly, R., Mehta, N.M. and Ganguly, N. (1982) in Hormone Regulation of Experimental Mammary Tumors (Leung, B., ed.), Eden, Los Angeles, in the press 6 Barnawell, E.B. (1965) J. Exp. Zool. 160, 189-206 7 Rosen, J.M., Matusik, R.J., Richards, D.A., Gupta, P. and Rodgers, J,R. (1980) Rec. Prog. Horm. Res. 36, 157-193 8 Bolander, F.F., Nicholas, K.R., Van Wyk, J.J. and Topper, Y.J. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 5682-5684 9 Topper, Y.J. and Freeman, C.S. (1980) Physiol. Rev. 60, 1049-1106 10 Nicholas, K.R., Sankaran, L. and Topper, Y.J. (1981) Endocrinology, 109, 978-980 11 Ono, M. and Oka, T. (1980) Cell 19, 473-480 12 Oka, T. and Topper, Y.J. (1971) J. Biol. Chem. 246, 7701-7707 13 Freeman, C.S. and Topper, Y.J. (1978) Endocrinology 103, 186-192 14 Moses, A.C., Nissley, S.P., Short, P.A., Rechler, M.M., White, R.M., Knight, A.B. and Higa, O.Z. (1980) Proc.
314 Natl. Acad. Sci. U.S.A. 77, 3649-3653 15 Antoniades, H.N. and Scher, D.C. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 1973-1977 16 Cohen, S. and Savage, C.R. (1974) Rec. Prog. Horm. Res. 30, 551-574 17 Jaffe, B.M. and Behrman, H.R. (1979) Methods of Hormone Radioimmunoassay Appendix 3, 1005-1014, Academic Press, New York 18 Wilde, C.J. and Kuhn, N.J. (1979) Biochem. J. 182, 287-294 19 Nicholas, K.R. and Topper, Y.J. (1980) Biochem. Biophys. Res. Commun. 94, 1424-1431
20 Korc, M., Owerbach, D., Quinto, C. and Rutter, W.J. (1981) Science 213, 351-353 21 Roy, A.K., Chatterjee, B., Prasad, M.S.K. and Unaker, N.J. (1980) J. Biol. Chem. 255, 11614-11618 22 Van Obberghen, E., Spooner, P.M., Kahn, C.R., Chernick, S.S., Garrison, M.M., Karlsson, F.A. and Grunfeld, C. (1979) Nature 280, 500-502 23 Sager, R. and Kovac, P. (1982) J. Cell Biol. 79. 480-484
309
Elsevier BBA 11216
A U N I Q U E AND ESSENTIAL ROLE FOR INSULIN IN T H E P H E N O T Y P I C E X P R E S S I O N OF RAT M A M M A R Y E P I T H E L I A L C E L L S U N R E L A T E D T O I T S F U N C T I O N IN CELL M A I N T E N A N C E KEVIN R. NICHOLAS, LAKSHMANAN SANKARAN and YALE J. TOPPER Laboratory of Biochemistry and Metabolism, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20205 (U.S.A.)
(Received February 25th, 1983) (Revised manuscript received June 10th, 1983)
Key words: Insulin," Growth factor," Phenotypic expression," (Rat epithelial cell)
Mammary explants from pregnant rats can be induced in regard to casein synthesis and a-lactalbumin activity when cultured in the presence of hydrocortisone, prolactin and levels of insulin approaching physiological concentrations. No detectable induction occurs in the absence of insulin. Although epidermal growth factor and multiplication stimulating activity, in the presence of hydrocortisone, can maintain the initial level of NADH-cytochrome c reductase as well as insulin, neither can substitute effectively for insulin in the induction of the milk proteins. Proinsulin, nerve growth factor, platelet-derived growth factor and fibroblast growth factor are also ineffective substitutes for insulin in this regard. Whereas prolonged tissue exposure to multiplication stimulating activity, hydrocortisone and prolactin does not result in induction of a-lactalbumin activity, subsequent addition of insulin leads to prompt response. The results suggest that the ability of insulin to function as a unique, essential factor in the induction of rat milk proteins is independent of its ceil-maintenance activity. Thus, in addition to its well established functions in metabolic processes, insulin appears to play a vital role in certain developmental processes.
Introduction It has been shown that insulin, glucocorticoid and prolactin are essential for the induction of the synthesis of phosphorylated casein by m a m m a r y explants from intact mice [1,2]. It was conceivable that fewer hormones might be required for the accumulation of mouse casein m R N A . However, it is now agreed that glucocorticoid, in addition to prolactin, is necessary for the accumulation of casein m R N A in mouse m a m m a r y explants [3,4]. Although insulin was routinely added to the cultures in these studies, a possible intimate role for
Abbreviation: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. 0167-4889/83/$03.00 © 1983 Elsevier Science Publishers B.V.
this hormone in the phenotypic expression of m a m m a r y cells has not been considered [5]. Rather, insulin has been regarded only as an agent involved in cell maintenance [6,7]. Recently, it was reported [8] that, although insulin, epidermal growth factor and somatomedin c can function equally well as cell maintenance agents, only insulin, together with cortisol and prolactin, can induce casein m R N A accumulation in mouse m a m m a r y tissue in vitro. On this basis, it was suggested that insulin may be essential for the expression of certain genes related to cell differentiation, in addition to serving as a metabolic and growth factor. Since a number of species differences relating to the hormonal requirements for m a m m a r y development are recognized [9], it was of interest to
310 investigate the role of insulin in the phenotypic expression of rat mammary epithelial cells in vitro. In this study, a number of additional growth factors have been compared with insulin in relation to the induction of both casein synthesis and alactalbumin activity. Moreover, the previous evidence [8] which indicated that insulin is not a mere cell-maintenance factor in the mammary explant system has been extended. Evidence that insulin may be vital in certain other developmental systems, as well, will be discussed. Materials and Methods
Materials Ovine prolactin (NIH-PRL-14 and 15) was a gift from the Hormone Distribution Program, NIADDK, NIH, Bethesda, MD, and crystalline porcine insulin was a gift from Eli Lilly Co. (Indianapolis, IN). Multiplication stimulating activity, epidermal growth factor (receptor grade), nerve growth factor (2.5 S), fibroblast growth factor and platelet-derived growth factor were purchased from Collaborative Research, Inc. (Waltham, MA). Proinsulin, Lot 605-70N-279-1, was a gift kindly provided by Dr. F. DePablo, NIADDK. Hydrocortisone and cytochrome c (equine heart, salt-free) were purchased from Calbiochem (La JoUa, CA). Medium 199 containing Hanks' salts was from Grand Island Biological Company (Grand Island, NY). Hepes, NADH, bovine serum albumin (crystallized and lyophilized), uridine-5'-diphosphogalactose, uridine-5'-triphosphate, bovine milk galactosyltransferase (4.5 units/mg), penicillin G and bovine a-lactalbumin (Grade II) were from Sigma Chemical Co. (St. Louis, MO). Type I collagenase (204 units/mg) was purchased from Millipore Corporation (Freehold, N J). Uridine diphospho-D-[6-3H]galactose (7 Ci/mmol)was from Amersham Corporation (Arlington Heights, IL) and 3H-labeled L-amino acid mixture and Triton X-100 were from New England Nuclear (Boston, MA). Rabbit anti-sheep IgG was obtained from Miles Biochemical (Elkhart; IN). Tissue culture 3-4-month-old pregnant (13-14 days of gestation) Sprague-Dawley rats (Taconic Farms, Germantown, NY) were used. Explants were prepared
from the abdominal mammary glands and cultured in Medium 199 as described previously [10]. Hormones were added to the media at the concentrations indicated. Growth factors, obtained as sterile, lyophilized preparations, were dissolved in sterile media and added aseptically. Tissue analysis Casein synthesis was measured in homogenates of explants cultured for three days by immunoprecipitation [11] with sheep anti-rat casein following a 16-h pulse with 3H-labeled amino acids (20 /~Ci/ml). The immunoprecipitation method can detect both unphosphorylated and phosphorylated caseins, a-Lactalbumin activity was measured in homogenates by determining lactose formation in the presence of galactosyltransferase, as described previously [ 10]. NADH-cytochrome c reductase was measured [12] in epithelial cell-enriched fractions following collagenase digestion of mammary gland explants [13]. Results
An insulin dose-response curve in relation to the induction of a-lactalbumin activity is shown in Fig. 1. No response occurs in the absence of insulin. However, concentrations of the hormone approaching the physiological level (a few ng/ml), particularly in the presence of bovine serum albumin, do evoke response. Some loss of insulin by adsorption is presumably prevented by bovine serum albumin. Similar results were obtained with casein (data not shown). A comparison of six factors for their ability to support the induction of casein synthesis is depicted in Table I. Each was tested at 0.25/~g/ml, a concentration considerably above their physiological level [14-18]; the blood level of fibroblast growth factor has not been reported. Appreciable induction occurs only in the presence of insulin. Proinsulin and multiplication stimulating activity are less than 10% as active as insulin. Multiplication stimulating activity and epidermal growth factor had no more activity when tested at a concentration of 1 ~g/ml (data not shown). Similar results were obtained in terms of a-lactalbumin; in this case, nerve growth factor was also tested, and had virtually no activity (data not shown).
311
12
==
z
-.J
0
0
10
20
30
40
,~0
INSULIN CONCENTRATION (ng/ml) Fig. 1. Effect of insulin concentration on a-lactalbumin activity in pregnant rat mammary gland explants. Tissue was cultured for 96 h in media containing hydrocortisone (0.05 /zg/ml), prolactin (1 /~g/ml) and insulin (concentrations indicated) in the presence (1 m g / m l ) (O) or absence (O) of bovine serum albumin, a-Lactalbumin was measured in tissue extracts. A indicates the a-lactalbumin activity in the mammary gland prior to culture. Each value represents mean-FS.E, for three rats.
In an effort to determine whether the unique ability of insulin to support the induction of alactalbumin activity and casein synthesis is related to a unique ability of the hormone to maintain the mammary epithelium in a viable state, two types of experiment were performed. In the first (Table II), insulin, multiplication stimulating activity and epidermal growth factor were compared in relation to NADH-cytochrome c reductase. Hydrocortisone alone does not maintain the initial activity of the membrane-associated enzyme. However, the activity is maintained when any one of the factors is added together with hydrocortisone. The second experimental approach was to determine whether multiplication stimulating activity, a factor which has very little ability to support induction of casein synthesis or a-lactalbumin activity, can nevertheless sustain the cells' potential for response to insulin. The time-course for induction of a-lactalbumin in the presence of insulin, hydrocortisone and prolactin is shown in Fig. 2A. After an initial 24-h lag period, induction begins during the second 24 h. The increment produced during the second day is shown in Fig. 2B, which also demonstrates that although multiplication-
TABLE I
TABLE II
COMPARISON OF INSULIN, PROINSULIN AND SEVERAL GROWTH FACTORS FOR THEIR ABILITY TO I N D U C E CASEIN SYNTHESIS IN P R E G N A N T RAT M A M M A R Y G L A N D EXPLANTS
CAPACITY OF INSULIN, MULTIPLICATION STIMULATING ACTIVITY AND EPIDERMAL GROWTH FACTOR TO MAINTAIN N A D H - C Y T O C H R O M E c REDUCATSE ACTIVITY IN P R E G N A N T RAT M A M M A R Y G L A N D EXPLANTS
Explants from 14-day-pregnant rats were cultured for 72 h in medium containing hydrocortisone and prolactin (both at 1 /.tg/ml), and either insulin, proinsulin, multiplication stimulating activity (MSA), epidermal growth factor (EGF), fibroblast growth factor (FGF) or platelet-derived growth factor (PDGF), all at 0.25/~g/rnl. The explants were pulsed with 20 # C i / m l of 3H-labeled L-amino acid mixture during the last 16 h of incubation and the amount of casein formed was measured by immunoprecipitation using sheep anti-rat casein serum. Values for casein synthesis are cpm ( × 1 0 - 2 ) / h per mg tissue and are + S.E. for the number of rats shown, each analyzed separately.
Tissue was cultured for 96 h in media containing hydrocortisone (0.05 /tg/ml) alone, or a combination of hydrocortisone and either insulin, multiplication stimulating activity (MSA) or epidermal growth factor (EGF) (each at a concentration of 0.25 /~g/ml). NADH-cytochrome c reductase was measured [12] in epithelial cell-enriched fractions following collagenase digestion of the explants [13]. Values represent mean-I-S.E, with the number of observations shown in parentheses. For uncultured and epidermal growth factor+hydrocortisone systems, actual values from two observations are given.
Addition
Casein synthesis
Number of rats
Culture conditions
Insulin Proinsulin MSA EGF FGF PDGF
12.7±0.9 1.2±0.5 0.8±0.1 0.3±0.1 0.3±0.1 0.1±0.05
12 12 6 3 3 3
NADH-cytochrome c reductase ( A A s s 0 / m i n per g tissue)
Uncultured Hydrocortisone Insulin + hydrocortisone MSA + hydrocortisone E G F + hydrocortisone
0.297, 0.457 0.119 + 0.01 (5) 0.400 +_0.07 (4) 0.373 + 0.03 (5) 0.324, 0.364
312
30
A
2O ".= 10
>I.,> I.-
0
24
48
72
96
HOURS IN CULTURE 7
z m .J
5
~- 4
lIIl
l
1ii
MSAFP2MSAFP ~, 72MSAFP ~ 72 MSAFP
24
IFP2,
MSAFP,
24
IFP24 '~'
24
CULTURE CONDITIONS
Fig. 2. Capacity of multiplication stimulating activity to maintain hormone-responsive mammary epithelial cells. (A) Explants from pregnant rat mammary glands were cultured in media containing insulin (0.25 #g/ml), hydrocortisone (0.05 /~g/ml) and prolactin (1/~g/ml), and the a-lactalbumin activity in tissue extracts determined at daily intervals. Each value represents mean±S.E, for three groups of rats (two rats per group). (B) Explants were cultured in media containing 0.05 #g/m1 hydrocortisone (F), 1 #g/ml prolactin (P) and 0.25 #g/ml multiplication stimulating activity (MSA) for 72 h. Fresh medium with or without 0.25 #g/ml insulin (I) was added and cultures incubated for a further 24 h. a-Lactalbumin activity was determined in tissue extracts. The values shown for the system without insulin correspond to those in A. Each value is the mean ± S.E. for three groups of rats (two rats per group).
stimulating activity, together with hydrocortisone and prolactin, elicits almost no response during 4 days of culture, the tissue does respond when insulin is added on the 4th day of this culture. Discussion
The intent of this study is to determine whether or not insulin, in addition to its well established
general metabolic roles, m a y also serve as a specific, essential factor in the phenotypic expression of certain cell types. The h o r m o n a l induction, in vitro, of casein synthesis and a-lactalbumin activity in the rat m a m m a r y epithelial cell is the model selected for this inquiry. Although insulin, in addition to glucocorticoid and prolactin, has been f o u n d to be a necessary c o m p o n e n t of the culture media used in studies of mouse m a m m a r y develo p m e n t [ 1], a possible specific role for the hormone in the terminal differentiation of these cells has not been considered [5]. Rather, insulin has been thought to serve only a cell-maintenance function in this system [6,7]. Establishment of a more direct, unique role for insulin in the expression of the m a m m a r y phenotype would require that three criteria be met. (1) The h o r m o n e should be effective at a level approaching the physiological concentration. (2) Other factors should be incapable of supplanting insulin in regard to the induction of milk proteins. (3) At least one of these factors should be able to maintain the cells in the presence of glucocorticoid and prolactin. The first criterion is satisfied by the dose-response curves relating insulin to the induction of a-lactalbumin activity and casein synthesis. N o response occurs in either instance unless insulin is present, in addition to glucocorticoid and prolactin. Increments are manifested in both cases when insulin is added at a level which approaches the physiological concentration. N o n e of the other factors tested appears capable of substituting for insulin in the induction of casein synthesis or a-lactalbumin activity. It is conceivable that multiplication stimulating activity at higher levels might exert more activity via the insulin receptor. The relative inactivity of proinsulin suggests that the effects elicited by insulin involve interaction between the h o r m o n e and its own receptor. The results satisfy the second criterion stated above. The design of the experiments represented in Table II addresses the third criterion, i.e., do factors which cannot p r o m o t e the induction of alactalbumin activity and casein synthesis nervetheless maintain the cells as well as insulin? It is clear that glucocorticoid alone cannot maintain the initial level of N A D H - c y t o c h r o m e c reductase, but
313
that both multiplication stimulating activity and epidermal growth factor, in the presence of hydrocortisone, are as effective as insulin in this respect. The results also indicate that the two factors are as effective as insulin in maintaining cellular responsiveness to hydrocortisone [12]. A much more direct experimental approach toward satisfying the third criterion is represented in Fig. 2B. It is apparent that although multiplication-stimulating activity, in the presence of glucocorticoid and prolactin, has little ability to promote the induction of a-lactalbumin activity, it maintains, during 3 days of culture, the cells' potential for prompt response to the delayed addition of insulin. More specifically, the presence of insulin for 1 day following 3 days with multiplication stimulating activity, glucocorticoid and prolactin, resulted in approximately as much induction as that which occurred during a 1-day period in a system containing insulin from the start. The emergence of ct-lactalbumin activity upon the delayed addition of insulin, observed here, is similar phenomenologically to the emergence of casein synthesis upon the delayed addition of prolactin to a system previously cultured with insulin and glucocorticoid [19]. In one instance, it might be tempting to regard insulin as 'the' lactogenic hormone, while in the other, prolactin might be so regarded. Actually, both are required, and it is not possible at this time to assign greater importance to one or the other. It is clear that while insulin is not unique in terms of cell maintenance in the explant system, it does appear to have an essential and unique role in the phenotypic expression of the rat mammary epithelial cell. In this respect, the rat and mouse [8] are similar, while the rabbit mammary epithelial cell does not appear to require insulin for its phenotypic expression [9]. Insulin may have a role in the phenotypic expression of other cell types also. The pancreas of diabetic rats manifests a selective reduction in the level of m R N A for amylase [20] and the liver of mildly diabetic rats shows a selective reduction in the level of mRNA for t~2~ globulin [21]. It is not clear, however, whether the defects are the result of deprivation of insulin per se, or of the systemic derangements of diabetes. Lipoprotein lipase activity in 3T3-L1 fatty fibroblasts is stimulated to
maximum levels after 4 days of culture in the presence of insulin [22]. In addition, insulin can effect the determination of pre-adipocytes [23]. It will be of interest to learn whether the hormone is effective at physiological levels, and whether other growth factors that can function in cell maintenance are active in the determination of preadipocytes. These considerations would be pertinent to any system which appears to require insulin for development in vitro. The mammary epithelial cells in mature, nonpregnant animals are actually in an immature functional state. Unlike most other cells in the mammal, they achieve their potential only during pregnancy and lactation. Insulin, as we have seen, is one of the hormones required for this maturation of the murine mammary cell. The foregoing discussion raises the possibility that, in addition to its metabolic functions, insulin may also serve as a 'developmental' hormone which is essential for selective gene expression in certain cells at particular stages of ontogeny. References 1 Juergens, W.G., Stockdale, F.E., Topper, Y.J. and Elias, J.J. (1965) Proc. Natl. Acad. Sci. U.S.A. 54, 629-634 2 Turkington, R.W., Juergens, W.G. and Topper, Y.J. (1965) Biochim. Biophys. Acta 111,573-576 3 Nagaiah, K., Bolander, F.F., Nicholas, K.R., Takemoto, T. and Topper, Y.J. (1981) Biochem. Biophys. Res. Commun. 98, 380-387 4 Ganguly, R., Ganguly, N., Mehta, N.M. and Banerjee, M.R. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 6003-6006 5 Banerjee, M.R., Ganguly, R., Mehta, N.M. and Ganguly, N. (1982) in Hormone Regulation of Experimental Mammary Tumors (Leung, B., ed.), Eden, Los Angeles, in the press 6 Barnawell, E.B. (1965) J. Exp. Zool. 160, 189-206 7 Rosen, J.M., Matusik, R.J., Richards, D.A., Gupta, P. and Rodgers, J,R. (1980) Rec. Prog. Horm. Res. 36, 157-193 8 Bolander, F.F., Nicholas, K.R., Van Wyk, J.J. and Topper, Y.J. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 5682-5684 9 Topper, Y.J. and Freeman, C.S. (1980) Physiol. Rev. 60, 1049-1106 10 Nicholas, K.R., Sankaran, L. and Topper, Y.J. (1981) Endocrinology, 109, 978-980 11 Ono, M. and Oka, T. (1980) Cell 19, 473-480 12 Oka, T. and Topper, Y.J. (1971) J. Biol. Chem. 246, 7701-7707 13 Freeman, C.S. and Topper, Y.J. (1978) Endocrinology 103, 186-192 14 Moses, A.C., Nissley, S.P., Short, P.A., Rechler, M.M., White, R.M., Knight, A.B. and Higa, O.Z. (1980) Proc.
314 Natl. Acad. Sci. U.S.A. 77, 3649-3653 15 Antoniades, H.N. and Scher, D.C. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 1973-1977 16 Cohen, S. and Savage, C.R. (1974) Rec. Prog. Horm. Res. 30, 551-574 17 Jaffe, B.M. and Behrman, H.R. (1979) Methods of Hormone Radioimmunoassay Appendix 3, 1005-1014, Academic Press, New York 18 Wilde, C.J. and Kuhn, N.J. (1979) Biochem. J. 182, 287-294 19 Nicholas, K.R. and Topper, Y.J. (1980) Biochem. Biophys. Res. Commun. 94, 1424-1431
20 Korc, M., Owerbach, D., Quinto, C. and Rutter, W.J. (1981) Science 213, 351-353 21 Roy, A.K., Chatterjee, B., Prasad, M.S.K. and Unaker, N.J. (1980) J. Biol. Chem. 255, 11614-11618 22 Van Obberghen, E., Spooner, P.M., Kahn, C.R., Chernick, S.S., Garrison, M.M., Karlsson, F.A. and Grunfeld, C. (1979) Nature 280, 500-502 23 Sager, R. and Kovac, P. (1982) J. Cell Biol. 79. 480-484