THE JOURNAI. OF Bvx.oc~cn~ CHEMISTRY Vol. 254, No. 20, Issue of October 25, pp. 10466-10475, 1979 Prmtedm U.SA

Dispersed Mammary Epithelial Cells RECEPTORS OF LACTOGENIC HORMONES IN VIRGIN, PREGNANT, AND LACTATING RABBITS*

(Received for publication, November 15, 1978, and in revised form, June 4, 1979)

Y. M. L. Suard and J.-P. Kraehenhuhl From the Znstitute of Biochemistry, University of Lausanne, Lausanne, Switzerland

M. L. Aubert From the Department of Pediatrics, University of Geneva, Geneva, Switzerland

The number and affinity of binding sites for lacto- genie hormones have been determined in dispersed mammary cells from virgin, pregnant, and lactating rabbits. Dispersed epithelial cells, prepared from mam- mary glands by enzyme digestion, calcium chelation, and gentle shearing, were separated from nonepithelial cells by density centrifugation. lZ51-labeled ovine pro- lactin (oPRL) and lz51-labeled human growth hormone (hGH) were used as tracers.

Association and dissociation of lZ51-oPRL or L251-hGH were time- and temperature-dependent. The rate of association followed a second order reversible reaction with a rate constant of -0.5 at 4”C, -2.0 at 23”C, and -9 x 10’ M-’ min-’ at 37°C. Maximum binding was achieved after 120 h at 4”C, 48 h at 23”C, and 2 to 4 h at 37°C. Dissociation of ‘251-oPRL or hGH from cells by unlabeled oPRL was complete at 4°C after 160 h, follow- ing a first order reaction (km1 = 9.9 x 10m5 min) and incomplete at 23°C and 37°C even after prolonged time. Internalization of receptor-bound ‘251-oPRL was stud- ied by quantitative electron microscope autoradiogra- phy. Grain distribution over- and volume densities of cellular organelles was analyzed as a function of time and temperature. At 37”C, there was a rapid and spe- cific translocation of lactogenic hormones to intracel- lular organelles. Autoradiographic grains were found associated with vesicles, Golgi elements, lysosome-like structures, and the nucleus. One class of high affinity binding sites was estimated from Scatchard plot and direct kinetic analyses at 4°C. Whereas the apparent affinity constant (-lOlo M-‘) did not change signifi- cantly throughout pregnancy and early lactation, the number of receptors extrapolated from Scatchard plots at 4°C varied in an inverse relation to serum proges- terone concentration. Thus, -1900 sites were detected in virgin rabbits (progesterone, -200 pg/ml), -850 at midpregnancy (progesterone, -15,000 pg/ml), and -1800 during early lactation (progesterone, -500 pg/ ml). The binding properties of lactogenic hormones to dispersed cells was compared with those to Triton X- 100 solubilized microsomal membrane preparations. Good correlation between the two systems was found indicating that cell dispersion did not alter binding properties.

Qur results indicate that dispersed mammary cells

* This work was supported by Grants 3.496-0.75, 3.514-0.75, and 3.122-0.77 from the Swiss National Foundation for Scientific Re- search. Part of this work was presented at the symposium on Prolactin Physiology, Nice, France, 1977 (2). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact.

bind lactogenic hormones in a saturable and reversible process, that the number of exposed receptors varies throughout gestation and lactation, and finally that lactogenic hormones are internalized following inter- action with their membrane receptors.

The differentiation of the mammary gland during gestation, characterized by intense cell proliferation and initiation of lactation at parturition, is modulated by variations in the blood concentration of several hormones (48). In the rabbit, as in other species, it has been established that lactogenic hormones regulate gene expression of milk specific proteins such as casein with a rapid increase of the level of mRNA (35, 37,47). Progesterone limits this induction (23,24,36), whereas glucocorticoids, inactive alone, potentiate prolactin effects (10). The molecular mechanisms whereby lactogenic hor- mones regulate the expression of specific genes is currently under extensive investigation (10,23,24,35-37,47). Very little is known, however, of the cellular events which allow lacto- genie hormones to exert their effect on gene expression. Bind- ing sites for lactogenic hormones have been identified in mice by Turkington et al. (49) and visualized by light microscope autoradiography by Birkinshaw and Falconer (3). It has been shown that binding of lactogenic hormones to membrane receptors is a necessary step for the subsequent action of the hormones (43-46), but interaction with the receptor may not be sufficient.

Understanding of the relationship between hormone bind- ing to target cell receptors and the triggering of events leading to expression of biological activity requires intact and viable cells. Studies with fibroblasts (22), dispersed adipocytes (B), Leydig cells (30), and lymphocytes (15) have been definitive on such problems as binding kinetics, receptor metabolism, and transmission of the hormonal signal from the receptor into the cell and its relationship to biological response. A procedure for the preparation of viable dispersed mammary cells, free of interstitial cells, has been developed (27). In this study we have investigated the interaction of lactogenic hor- mones with available binding sites exposed at the surface of dispersed cells from virgin, pregnant, and lactating rabbits and analyzed the variation in number of sites through preg- nancy and lactation. Finally, the fate of labeled lactogenic hormones upon interaction with binding sites was examined morphologically.

MATERIALS AND METHODS

Animals-Virgin 4- to 5-month-old white New Zealand rabbits were used in all experiments. Females bred for the fist time were killed on Days 15, 18, 21, and 28 of gestation 01 2 and 3 days after

10466

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells 10467

parturition. Some lactating rabbits were subcutaneously injected with 2 mg of bromocryptine 36, 24, and 12 h prior to removal of the mammary glands. Prior to killing, a blood sample was taken from the ear marginal vein to measure progesterone. Following induction of barbiturate anesthesia (pentobarbital: 50 mg per kg weight), mam- mary glands were removed under sterile conditions. For each experi- ment, one gland was used to prepare an enriched plasma membrane fraction according to the procedure of Shiu and Friesen (43).

lo&nation-oPRL’ and hGH were iodinated with iodine 125 ac- cording to the lactoperoxidase procedure, as previously described (1) (Fig. 1, Table I).

Dispersed Mammary Cells-Glands of virgin, midpregnant, and lactating females were dissociated according to the procedure of Kraehenbuhl (27) with the following modifications: a-chymotrypsin was omitted and commercially available collagenase was replaced by collagenase purified by the procedure of Gunther et al. (20). The specific activity of purified collagenase was 16.8 units/mg at 37°C with N-carboxy-Gly-L-Pro-Gly-L-Pro-r-Ala as a synthetic substrate. The proteolytic activity of the enzyme mixture was carefully con- trolled in order to minimize proteolytic degradation of PRL receptors exposed at the surface of dispersed mammary cells. The following activities were recorded per milliliter of enzyme mix: 3 units of collagenolytic activity, 10 units of elastolytic activity, and -650 units of hyaluronidase activity. Clostripain activity was less than 0.005 units/ml and trypsin-like activity was completely abrogated by the addition of 0.1 mg/ml of soybean trypsin inhibitor. No phospholipase C activity was detected.

Epithelial cells were separated from interstitial cells by isopycnic centrifugation on a 35 to 0.5% bovine serum albumin gradient as previously described.(l’i). The cells were loaded on top of the gradient, centrifuged 30 min at 10,000 x g,,, at room temperature. Finally, after washing, the cells were suspended in Medium Mlw complemented with 2%~ bovine serum albumin and 100 pg/ml of gentamycin.

Membrane Preparation-In each experiment, one of six glands from a rabbit was used to prepare a plasma membrane-enriched fraction according to the technique of Shiu and Friesen (43).

Binding Assays-Miss containing 2% bovine serum albumin and gentamycin (100 pg/ml) was used as diluent for all binding assays. All media were filtered through a Millipore filter, 0.22 pm. Media were equilibrated with 5% CO?, 95% air and the pH was maintained at 7.4 during all assays. Labeled hormones (5 to 25 PM) and dispersed cells (2 x 10” cells) were incubated in sterile plastic tubes (12 X 75 mm) for various times at 4,23, or 37°C in the presence or absence of unlabeled hormone in a total incubation volume of 0.5 ml. Cells were shaken periodically. The incubation was terminated by addition of 2 ml of ice-cold diluent to each tube followed by immediate centrifugation at 1500 x g for 30 min. The supernatants were decanted, the tubes drained, and the precipitates counted in a Packard y-counter for 10 min. For each assay condition, nonspecific binding was assessed in parallel by addition of about 2 )rg of excess hormone to the medium and specific binding was the difference between total and nonspecific binding. For the estimation of the rate of dissociation, a set of tubes was incubated until equilibrium; then excess hormone (2 pg) in 2 ml of medium was added to each tube and, at the appropriate time, the tubes were processed as above.

Kinetic and Statistic Analysis-Analysis of the data was per- formed with the help of a CDC Cyber 170 computer, using a radioim- munoassay data analysis program, kindly provided to us by Dr. D. Rodbard (34). When suitable, Scatchard plot analysis was performed. Correction for nonspecific binding in Scatchard plots was made as suggested by Chamness and McGuire (7). Rate of association was calculated using the equation for a reversible second order reaction, as proposed by Shiu and Friesen (42).

Electron Microscopy-Dissociated cells were fixed in 2% formal- dehyde, 2% glutaraldehyde in 0.1 M sodium cacodylate, pH 7.4, for 2 to 4 h at 4°C by mixing 1 vol of cell suspension with 2 vol of the above fixative. After fixing the cells, they were transferred to a polyethylene microfuge tube and centrifuged at -10,000 X g.,, for 2 min (Microfuge 152: Beckman Instruments Inc., Fullerton, Calif.). The resulting pellet was cut into discs, postfixed in 1% OsOl for 1 h at 4”C, stained in block for 1 h at room temperature with 0.5% uranyl

’ The abbreviations used are: bGH, bovine growth hormone; bPRL, bovine prolactin; DNP, dinitrophenol; hCG, human chorionic gonad- otropin; hCS, human chorionic somatomammotropin (placental lac- togen); hGH, human growth hormone; oPRL, ovine prolactin; PRL, prolactin; rPRL, rat prolactin, and QO,, oxygen consumption.

acetate in veronal/acetate buffer, pH 7.4, dehydrated in ethanol and propylene oxide, and embedded in Epon. Thin sections were examined in a Philips 400 EM and micrographs were taken at 60 kV.

Autoradiography-Cells from three rabbits (15th to 18th day of gestation) were also processed for light and EM autoradiography using Ilford Lq emulsion (39). For each time point, two separate samples of cells and three grids per sample were prepared for EM. Thus, six grids from each time point were examined and 150 micro- graphs were analyzed (-180 autoradiographic grains). The grain number (per cent of total) was plotted as a function of the distance of the grain center from the plasma membrane and histograms con- structed according to Salpeter et al. (40). The final magnification of all micrographs was 20,000. The percentage of autoradiographic grains associated with intracellular organelles was evaluated by superimpos- ing on aach grain a circle with a diameter of the grain (Figs. 14 and 15). The circle center was determined and the center points accu- mulated by a given organelle were estimated and expressed as percent of total number of grains counted.

Morphometry-The volume density of organelles from the same cells used for grain counts was estimated by point counting with the “multipurpose test system” (51).

RESULTS

Dispersed Cells: Homogeneity and Morphological Features

Mammary glands from virgin, pregnant, and lactating rab-

bits were dissociated by a combination of enzyme digestion, calcium chelation, and mechanical disruption. Freshly dis- persed cells from glands at different stages of differentiation were enriched in epithelial cells by equilibrium density cen- trifugation in bovine serum albumin solutions. Cells from virgin rabbits were roughly separated into a heavy fraction (p = 1.075) which contained mainly epithelial cells (Fig. 2a) as identified at the EM level by the presence of membrane specializations (remnants of desmosomes), and a pellet which was enriched in erythrocytes, plasma cells, fibroblasts, and macrophages. Epithelial cells were characterized by a small cytoplasm to nucleus ratio, numerous free ribosomes situated in the narrow rim of cytoplasm, and a poorly developed Golgi complex (Fig. 2b). Cells from pregnant rabbits were recovered in three fractions (27). The light fraction (5 y 1.045) consisted of more than 90% epithelial cells containing large fat droplets (27). Cells from lactating rabbits were recovered in three fractions. The lightest fraction (p = 1.055) contained cells sharing all the structural features of secreting cells, i.e. an abundant rough endoplasmic reticulum, a well developed Golgi complex, and numerous secretory vacuoles (Fig. 2, c and d). Cells from pseudopregnant rabbits were also recovered in three fractions. The light fraction (;a = 1.045) contained cells similar to those of the light fraction from pregnant animals. The heavy fraction (5 = 1.073) was enriched in undifferen- tiated cells similar to those obtained from truly virgin adult animals but was contaminated by nonepithelial cells (-20%). Criteria of cell viability are given in the Miniprint.’

Binding Kinetics

The binding of ““I-oPRL and ‘“‘I-hGH to receptors exposed on mammary cells from virgin, pregnant, and lactating rabbits was found to be a time- and temperature-dependent, satura- ble, and reversible process at all stages of differentiation. At 37”C, binding of radioiodinated lactogenic hormone stopped after about 2- to 4-h incubation and cells reaggregated forming visible clumps which could not be dispersed by shaking. The

’ Portions of this paper (including Figs. 1, 4, 5, 6, 7, 9, and 12 and Tables I, III, V, and VI) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biolog- ical Chemistry, 9650 Rockville Pike, Bethesda, Md. 20014. Request Document No. 78M-2039, cite author(s), and include a check or money order for $1.80 per set of photocopies.

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells

FIG. 2. Electron micrographs of dispersed mammary epithe- lial cells from virgin (a and b) and lactating (c and d) rabbits. Staining with uranyl acetate and lead citrate. a, several small cells with a low cytoplasm to nucleus ratio. Some cells are aggregated with desmosomes and tight junctions present between cells. x 1690. b, typical undifferentiated cell with a poorly developed secretory appa- ratus, rough endoplasmic reticulum restricted to the perinuclear space, numerous free ribosomes, a small Golgi complex (G), and no

binding capacity was lower for ““I-oPRL than for ‘%I-hGH (Fig. 4), although the concentration of both radioiodinated hormones in the incubation medium measured by radioim-

munoassay or radioreceptor assay did not vary significantly for a period up to 24 h. At 23”C, equilibrium was reached by 48 h (Fig. 3) and the cells remained well dispersed. By 24 h, about 80% of maximum binding was reached. At 4”C, a much slower binding rate was observed and equilibrium was reached only after -120-h incubation. No difference in binding capac- ity for ““I-oPRL versus ‘S”I-hGH was detected at 4 and 23°C. Nonspecific binding remained low and steady at 4”C, with no significant difference between 12 and 120 h. At 23”C, a slight increase was observed with -10% increase after 72 h, whereas at 37°C the increase reached -25% after 24 h (Fig. 4).

Rate of Association-Association rates of ‘251-oPRL and ‘251-hGH for dispersed cells from pregnant and lactating ani- mals followed a second order reaction of a reversible type (Table II). Using the irreversible second order reaction equa-

secretory granules. X 7500. c, several cells from a lactating animal (2nd day), which exhibit a well developed secretory apparatus, better visualized in d. x 1650. d, a typical lactating cell with an abundant rough endoplasmic reticulum (RER). a developed Go@ complex (G), and numerous secretory vacuoles (u). The polarity of the cell is maintained as evidenced by the presence of microvilli (mu) restricted to one pole of the cell. x 5000.

tion, nonlinearity was observed. In our cell system, the asso- ciation rate of ?-insulin to dispersed cells from midpregnant rabbits was rapid and equilibrium was reached after 2 h at 4°C (data not shown). Similar rates were obtained when incubating insulin with mammary membrane preparations. Since cells were incubated for a long period of time, it was essential to determine whether binding capacity of cells was altered with time. In a typical assay illustrated in Fig. 5, cells preincubated for 48 h at 23’C without tracer exhibited the same association rate (k+l = 0.78 X 10’ M-’ min-‘) and similar maximum binding capacity as freshly dispersed cells (k+l = 0.81 X 10’ M-l min-‘).

Rate of Dissociation-At 4”C, bound 12”1-oPRL was com- pletely displaced by unlabeled hormone following fast order reaction kinetics (Fig. 6). After 7-day dissociation (13-day total incubation), binding reached the level of nonspecific binding measured in control tubes incubated for 13 days. The dissociation rate constant was 9.9 x 10e5 min-‘. At 23”C, it

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells 10469

0 24 40 72 96 120 2u

Time of incubation(h)

FIG. 3. Comparison of rate of association of ?-oPRL (9.2 PM) to 2 X lo6 dispersed mammary cells from midpregnant rabbits at three different temperatures. All three assays started simultaneously with freshly dispersed cells. At 37”C, cells rapidly re- established junctions and binding kinetics was obscured. Equilibrium was reached after 48-h incubation at 23”C, after 5 days at 4°C. Binding values were corrected for nonspecific binding.

TABLE II

Rate constant of association of ‘25Z-oPRL to dispersed mammary cells

At 4 and 23”C, the rate constant was calculated over the entire association period, whereas at 37°C the constant was estimated during the fist half hour of association.

‘hmerature TWX of cells Rate constant 10’ M-I mX*

4°C Midpregnant 0.37 f 0.03 (n = 3)

23°C Midpregnant 2.01 -e 0.76 (n = 4)” Lactating 1.69 + 0.15 (n = 3)

37°C Midpregnant 8.9

n In one experiment, the association rate constant was calculated for ‘*‘I-oPRL (2.0 x lo7 Mm’ mm’) and “?-hGH (1.8 X 10’ Mm’ min-‘) indicating a striking similarity of binding kinetics between the two lactogenic hormones.

was found repeatedly that only 50% of the total amount of bound hormone could be displaced (Fig. 7). After about a 3- day incubation, no more decrease of bound oPRL was ob- served. Thus, no dissociation rate constant could be calcu- lated.

Affinity Constant-The affinity constant K, was calculated directly from analysis of the kinetic data (K, = f~+~/h-,) or determined from Scatchard analysis at 4°C performed at equilibrium after 120 h. A typical Scatchard plot is shown in Fig. 8. When experimental points were analyzed by the com- puter program, a best fit for one class of binding sites was established with an asymptote corresponding to nonspecific binding. Nevertheless, in some assays, including that of Fig. 8, a reasonably good fit (similar residual variance) was obtained for two classes of binding sites, a high affinity (K, = 2.3 x 10”’ M-l) low capacity (350 sites) site similar to the one class binding site, and a low affinity (5.2 x 10’ Mm’) high capacity (2970 sites) site. Apparent affinity constants and number of high affinity binding sites per cell at different stages of differ- entiation are summarized in Table III, upper part. No signif- icant change in apparent K, occurred during gestation and lactation. At 4’C, apparent K, determined from equilibrium kinetics were in good agreement with K, calculated from rate constants, 2.4 X 10” M-’ versus 3.5 X 10” M-I, respectively,

At 23”C, the K, could not be determined by analysis of the kinetic data since dissociation of PRL was incomplete (Fig. 7). Although binding equilibrium was achieved after 48 h at 23”C, dissociation of bound hormone is a complex reaction in relation to internalization of bound PRL and some of the criteria for performing a Scatchard analysis are not fulfilled.

0.05

O.OL

0.03

IL . m

0.02

0.01

0.00

\=

\

\ l

‘,: 2 -* = . \ \ l

\ \ l \ \ \ : \ \ - l \ \ \

\

l

\ \

l 0

---. \ l

“ -------- l

i --w* -

\ .

0 2.5

fmol I 2 1 lo6 ceils

5

FIG. 8. Scatchard plot for the binding of ““I-oPRL (10 f~) to dispersed mammary cells from a midpregnant rabbit at 4°C. Fit for one class of binding sites with an asymptote corresponding to nonspecific binding (p). Fit for two binding sites (- - -), a high affinity low capacity site (k, = 2.3 X IO”’ M. ‘, 350 sites), and a low affinity high capacity site (k, = 5.2 x lOa M-‘, 2970 sites).

N -- I

FIG. 10. Variation of PRL receptors on dispersed mammary cells throughout gestation and lactation in relation to proges- terone and prolactin serum concentrations. Number of binding sites were estimated from Scatchard plots established at 4°C. All pregnant and lactating rabbits were treated with bromocryptine to reduce endogenous cccupancy. Progesterone and rabbit prolactin concentrations were measured by radioimmunoassays and each point represents determinations from four different rabbits. The blood was collected at 10 a.m. Results were expressed as mean values + S.E. Number of experiments for receptor determinations are given in parentheses.

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

10470 Receptors of Lactogenic Hormones on Dispersed Mammary Cells

Thus, affinity constants derived from Scatchard analysis per- formed at 23°C (Table III, Zowerpczrt) are only indicative and comparison with data obtained at 4°C is then not possible.

Number of Binding Sites-The number of PRL receptors per cell was estimated by maximum binding capacity derived from Scatchard plots at 4°C (Fig. 8) and from the cell concen- tration present in the incubation medium (Fig. 9). Receptor site concentration was high on cells of virgin animals, de- creased 2-fold during gestation, and increased again during lactation in an inverse relation to blood concentration of progesterone (Fig. 10). There were important variations in early lactation (Table III). It should be stressed that only free receptor sites are measured by the present technique and, therefore, there could be an underestimate of the total number of binding sites in the situation of high occupancy of sites by endogenous hormones. Endogenous occupancy was investi- gated in pairs of pregnant or lactating rabbits (sisters bred on the same day), of which one animal was pretreated with bromocryptine as indicated under “Materials and Methods,” the other animal serving as control. At midpregnancy, the number of binding sites was 70% higher in the treated animal. In lactating rabbits, the treatment generated milk depletion and absence of nursing. The number of sites was 450% higher in the treated lactating animal. Similar results were obtained on corresponding membrane preparations, with a 50% increase in midpregnant animals and a 110% .increase in lactating rabbits. In the early part of this study, at a time when internalization of PRL had not yet been recognized, binding studies were performed at 23”C, and the number of binding sites derived from Scatchard analysis (Table III, lower part). As expected, the values were underestimated when compared to data at 4°C (Table III, upper part) and thus were not computed in Fig. 10.

Specificity-The specificity of the PRL binding sites of dispersed mammary cells was assessed using lactogenic or nonlactogenic hormones. As shown in Fig. 11, rat FSH and LH, porcine insulin, and hCG were ineffective in displacing bound ‘“51-oPRL; the activity of bGH, a hormone without lactogenic activity, may be intrinsic or due to a small bPRL contamination. The receptor of dispersed cells showed a sim- ilar competition for oPRL and hGH, whereas hCS exhibited 25% of the activity of oPRL in agreement with its known lactogenic activity. Cell specificity was determined using a human mammary tumor cell time (BT 20, a gift from Dr. Ozello, Department of Pathology, University of Lauyanne, Lausanne, Switzerland), devoided of hormone sensitivity; no oPRL receptor could be found on these cells.

Internalization of Lactogenic Hormones

The increase with time of nondisplaceable ‘““I-oPRL when cells were incubated at 23 and 37°C with ““I-hormone and excess unlabeled hormone and the incomplete dissociation of bound ‘““I-oPRL by unlabeled hormone suggested internali- zation of hormone. This was further analyzed by EM autora- diography. Cells, incubated with lZ51-oPRL (5 ng/106 cells) for 160 h at 4°C and washed once with oPRL-free medium, were incubated 5, 15, and 30 min at 37°C and fixed and processed for light and EM autoradiography. Number of labeled cells were obtained from light autoradiograms, whereas grain dis- tribution profdes at different time points were established from EM autoradiograms. Volume density of intracellular

1

FIG. 11. Displacement curves of lz51-oPRL (0.1 ng) from lac- togenic receptors (lo6 mammary cells from a pregnant rabbit) by increasing concentration of lactogenic hormones (bGH, oPRL, rPRL, and bCS) and nonlactogenic hormones (rFSH, rLH, bovine insulin, and bCG). Displacement by bGH, a hormone without lactogenic activity, may be intrinsic or due to a bPRL contamination. Similar results were obtained with soluble membranes using ““I-hGH as tracer.

TABLE IV Internalization of ““I-oPRL: grain distribution analysis related to volume density

Autoradiograms were analyzed with light microscopy to determine membrane and the perinuclear space; CY, cytoplasm including rough per cent labeled cells and with the electron microscope to determine endoplasmic reticulum; F, fat droplet. Volume densities of V and GO intracellular distribution (40) and volume densities (51). Abbrevia- were determined by considering both their membranes and cisternal tions: PM, plasma membrane; V, vacuoles and vesicles; GO, Golgi space. In italic: ratio of % grains to volume density. complexes; MI, mitochondria; NU, nucleus including the nuclear

Time point 5% labeled cells

Grains counted PM

Cellular organ&s

V GO NU MI CY F

0

5

15

30

75

73

78

78

189

160

202

216

% total grains

volume density W total grains

volume density % total grains

volume density % total grains

volume densitv

65.0 17.0 1.0 0 2.0 15.0 0 26.0 1.7 0.6 0 0.4 0.4 0

2.5 10.0 1.8 32.0 5.7 34.0 14.0 37.0 18.0 8.0 6.0 4.0 27.0 0 12.0 1.7 4.0 0.2 1.0 0.7 0

3.0 10.5 2.0 33.5 4.0 40.5 6.5 23.0 14.0 17.0 17.0 5.0 24.0 0

8.0 1.9 7.0 0.5 0.9 0.6 0 3 7.5 2.5 34 5.5 40 7.5

18.0 14.0 20.0 15.0 6.0 27.0 0 5.0 2.0 13.0 0.5 0.9 0.7 0 3.5 7.0 1.5 32.0 7.0 39.5 9.5

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells

FIGS. 13-16. Thin sections of dispersed mammary cells from midpregnant rabbits (18th day of gestation). Autoradiographic grains were developed after 3 months. Stained with uranyl acetate and lead citrate.

FIG. 13. Low magnification of cells incubated 160 h at 4°C with ““I-oPRL. AU the a-rains (arrows) are associated with the cell

- surface. X 7500. FIG. 14. The cells were incubated 160 h at 4°C with “*I-

oPRL, washed free of unbound labeled hormone, and warmed 5 min at 37°C. Autoradiographic grains (cwcles) are associated with the saccules of the Golgi apparatus (G) which surrounds the centriole

organelles at different time points were estimated by mor- phometry (51). Results, which are summarized in Table IV and Fig. 12, and illustrated in Figs. 13 to 16, clearly indicate that translocation of ‘%I-oPRL into the cell occurred at 37°C. Grain distribution profiles (Fig. 12) indicate a rapid and sig- nificant increase in plasma membrane distant grains and a concomitant decrease in plasma membrane associated grains as a function of time. At 4”C, the majority of grains was associated with the plasma membrane, the adjacent cyto-

(C) on the side away from the nucleus (N). M, mitochondria. x 16,000.

FIG. 15. Cells were treated as in Pig. 14, but incubated 30 min at 37°C. Two autographic grains (circles) are associated with a lysosome-like structure (L) and one grain with the cell surface. F, fat droplet; M, mitochondria. x 20,000.

FIG. 16. Cells treated as in Pig. 14, but incubated 15 min at 37°C. Three grains overlay the nucleus (arrows) and two grains remain associated with the cell surface (arrowheads). F, fat droplet. x 11,ooo.

plasm, and nearby vesicles, which could represent plasma membrane invagination (Fig. 13). After 5 min at 37”C, almost half of the grains were internalized (Table IV) and, as time proceeds, grains were found associated with vesicles and vac- uoles more distant from the plasma membrane (Fig. 12) with Golgi elements (Fig. 14), lysosome-like structures (Fig. 15), and the nucleus (Fig. 16). At 15 min, 17% of the grams were detected over the nucleus of which approximately one-fourth was overlying the nuclear membrane and three-fourths the

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

10472 Receptors of Lactogenic Hormones on Dispersed Mammary Cells

chromatin, especially euchromatin (Fig. 16). To take into account changes in volume density which could result from the temperature switch, the percentage of grains associated with a given cell organelle was expressed as a function of volume density (Table IV). In 30 min at 37”C, the ratio (per cent grain/volume density) decreased from 26 to 5 for the plasma membrane and increased from 0.6 to 13 for the Golgi apparatus, including both saccules and vesicles, and from 0 to 0.5 for the nucleus. On mitochondria and the cytoplasm which includes the rough endoplasmic reticulum, the ratio increased immediately following the temperature switch and remained steady. Although fat droplets occupy -10% of the total cell volume, less than 1% of the grains was detected over these structures. When cells were incubated at 4 or 37°C with iz51- oPRL (5 ng) and excess unlabeled hormone (2 pg), very few grains were observed (less than one grain per 1000 pm’). Finally, displacement of ““I-oPRL with unlabeled hormone was effective at 4°C since no grains were observed over cells.

DISCUSSION

Dispersed mammary cells from virgin, pregnant, and lactat- ing rabbits exhibit specific binding sites for lactogenic hor- mones. The cell-lactogenic hormone interaction was shown to be a time-dependent, saturable, and reversible process. The specificity of binding of lactogenic hormones to dispersed mammary cells was demonstrated by the fact that binding of “‘I-oPRL or ‘““I-hGH could be displaced exclusively by lac- togenic hormones, including oPRL, hGH, and hCS but not by hormones without lactogenic properties, such as rat LH or FSH, bovine insulin, and hCG. The lactogenic properties of hGH and hCS have been well demonstrated (14) and the use of these hormones as tracers or standard in in vitro assays as an alternate for PRL has been well documented (43, 46). Saturation analysis of membrane binding sites with either lz51- oPRL or ‘““I-hGH showed strict parallelism with displacement using oPRL as a standard, thus indicating that iodination at the degree of specific activities used (40 to 200 pCi/pg) did not affect binding properties (Table I).

Temperature greatly influenced the rate of association of radiolabeled lactogenic hormones to dispersed mammary cells with maximum binding around 120 h at 4°C and 2 to 4 h at 37°C. The rate of association followed a second order reaction, which was constant over the entire experimental period at 4 and 23”C, but only during the first half hour at 37°C. The rate of association at all temperatures was unexpectedly slow for both cell and membrane preparations. Low rates are probably an inherent property of lactogenic hormone recep- tors and not a consequence of cell dissociation or membrane disruption since insulin binds rapidly at 4°C to both cells and membranes as seen in this report and by others in murine gland (32). In dissociation rates, striking differences were found between cells and membranes. At 4°C labeled hormone bound to cells could be completely displaced by unlabeled hormone with a half displacement time of about 90 h, which is in constrast with the remarkable slowness of dissociation in membranes (12, 16, 21, 42). The irreversibility of hormone receptor interaction in membranes explains the discrepancy observed between the association constant estimated from Scatchard plots and from direct kinetic analysis. In cells both K, were in good agreement with a value of 2.4 X 10” M-l from Scatchard plots uersus 3.5 x 10” Me’ from kinetic data. An explanation for the discrepancy in membrane preparations has been proposed based on multiple step hormone receptor binding kinetics which tend to tighten bonds between inter- acting molecules as a function of time (13). Since good agree- ment has been observed with cells, receptor disorganization may follow membrane disruption and explain altered binding

properties of lactogenic hormones. In most assays, a best fit for one class of binding sites (Kc1

E 10”’ Me’) was given by the computer program (34). In several experiments, however, a fit for two classes of binding sites was given, the second with a K, between lo7 and 10’ Mm’. This second binding site may be of biological relevance. Preliminary results indicate that the KI, (l/K,) of the biolog- ical response of dispersed cells to prolactin, i.e. secretion of casein as measured by a radioimmunoassay, is in the range of 5 X lo-’ to lo-’ M. The apparent K,, 1.1 X lo” M PRL for dispersed murine mammary cells estimated from equilibrium kinetics, has been reported recently (38) which is 1 order of magnitude lower than values given in the present study. This could be species-related but more probably reflects method- ological differences. Thus, in the murine system, competitive studies at 23°C were performed under nonequilibrium condi- tions and commercially available collagenase with trypsin-like and clostripain activities was used which may alter the integ- rity of lactogenic receptors.

In our cell system, both ligands and receptors were stable. At 4°C both hormones remained intact over a 6-day period. The stability of the receptor was determined by comparing early association of ‘““I-oPRL to dispersed cells with late association after a 48-h preincubation in absence of labeled hormone. Despite a small decrease in maximum binding (-10%) at 23”C, the receptor remained relatively stable; no change at all was found at 4°C. At 37”C, however, substan- tially more binding was obtained with ““I-hGH than with ‘251- oPRL (Fig. 5). This difference cannot be explained by inacti- vation of oPRL in the incubation medium since the hormone concentration remained constant over 24 h as judged by radioreceptor and radioimmunoassays. Partial degradation of bound labeled oPRL could result in loss of radioiodine without a concomitant loss of binding property of the ligand. Finally, cells incubated at 37°C exhibited a sharp and rapid arrest of binding after 2 to 4 h which could be the result of receptor inacessibility. Aggregation of cells which occurred rapidly at 37°C accompanied by formation of junctional complexes could obstruct the access of hormones to most binding sites.

The number of available binding sites on dispersed mam- mary cells were estimated at 4°C from Scatchard plot analysis, and the number of cells was determined by direct cell count or by DNA measurement assuming a correlation factor of 7 pg of DNA per cell (27). Table III compares numbers of available binding sites as calculated from assays performed both at 4 and 23°C. Only assays performed at 4°C fulfill all criteria for a Scatchard analysis (binding at equilibrium, sec- ond order association reaction, first order dissociation reac- tion). At 23°C dissociation kinetics is complex since part of the radioactive ligand is internalized; therefore, interpretation of Scatchard analysis has to be made with caution. In fact, as seen from Table III, the numbers of binding sites observed in assays performed at 23°C compare well with values estab- lished at 4°C although there is a slight underestimation (-10%) at the higher temperature. At 4°C the number of high affinity receptors varied thoughout gestation and lactation, with about 1900 sites in virgin animals, 850 sites at midpreg- nancy, and 1800 sites in early lactation (Table III, upperpart; Fig. 10). The number of receptors was relatively constant at midpregnancy, but great heterogeneity was found both in virgin and lactating animals. Djiane et al. (12) compared the binding of lactogenic hormones to membranes from bromo- cryptine-treated and control rabbits and found at least 50% occupancy by endogenous prolactin from day 14 of gestation through lactation. We confirmed Djiane’s observations in cells obtained from lactating rabbits (2nd day) with a 5-fold in- crease in binding capacity, and from midpregnant animals

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells 10473

with a 2-fold increase. These findings of receptor occupancy by endogenous prolactin agree with the serum profile of prolactin (29) characterized by low levels from day 4 to 26 of gestation, followed by a sharp increase from day 26 (10 ng/ ml) to parturition (200 rig/ml). In addition, no placental lactogen has been clearly identified in rabbits (29). In virgin animals over 5 months of age, pseudopregnancy sometimes occurred spontaneously as indicated by the high level of serum progesterone. Epithelial cells from pseudopregnant animals were separated into differentiated cells with a low number of sites, and undifferentiated cells which exhibited the same receptor concentration as cells from “truly” virgin rabbits. The low number of binding sites in some virgin animals was the result of cell heterogeneity induced by spontaneous pseu- dopregnancy (Table III). The concentration of PRL receptors in the mammary gland during pregnancy and lactation is very sensitive to hormonal environment and might be one of the essential parameters modulating the intensity of prolactin activity. It has been shown that prolactin amplifies its own receptor level (33) and that progesterone counteracts the self- regulated increase of prolactin receptors in the mammary gland (11). Our data on dispersed cells extend these observa- tions, showing an inverse relation of the number of PRL receptors to serum concentration of progesterone. Such an inverse relation has also been reported for murine mammary cells (38). The mechanism by which progesterone regulates PRL receptors remains unknown. The effect could be direct on PRL receptor metabolism or indirect by receptor dilution consecutive to cell division. It is well established that proges- terone in association with estrogens stimulates DNA synthesis and cell proliferation (48). We observed in dispersed mam- mary cells a dose-dependent stimulation of labeled thymidine incorporation into DNA mediated by progesterone, with an optimum response at -10 rig/ml which is the serum concen- tration at midpregnancy. Thus, it should be possible now to examine in vitro the effect of progesterone on PRL receptor metabolism of dispersed cells.

This study presents evidence that prolactin can enter mam- mary cells upon interaction with surface receptors. Such in- ternalization has been well documented for several peptide hormones (4-6,9,17-19,41,50). Endogenous immunoreactive prolactin has been detected in rat mammary epithelial cells with some nuclear localization (31), and recently uptake of prolactin in the Golgi apparatus of rat hepatocytes has been reported (25). In mammary cells, our data indicate rapid internalization of ‘*‘I-oPRL with three-fourths of surface- bound hormone translocated into the cell in 15 min. In mam- mary cells, autoradiographic grains are found associated with endocytic vesicles, the Golgi elements (Fig. 14), and the nu- cleus (Fig. 16). In our morphometric analysis, lysosomes were counted as vacuoles since no cytochemical markers were used to clearly identify these organelles. In some micrographs (Fig. 15), grains could be related to lysosome-like structures. In mammary cells, lysosomal degradation does not seem to play an important role in prolactin metabolism at least during the first half hour of incubation at 37°C since internalized ‘251- prolactin can be quantitatively recovered from the cells, co- migrates on sodium dodecyl sulfate-polyacrylamide gel elec- trophoretograms with native oPRL, and remains active by a radioreceptor assay (data not shown). Prolactin rapidly con- centrates in the Golgi apparatus, as shown by the high per- centage of autoradiographic grains associated with this organ- elle (Fig. 14), thus confming the study of Josefsberg et al. (25) in rat hepatocytes. As suggested by these authors, the Golgi apparatus could mediate hormone action inside the cell and/or be involved in the process of hormone degradation. Nuclear localization of radiolabeled hormone has been re-

ported (6, 18), but interpretation of such results remains controversial. In this study, the percentage of grains was related to volume density in order to take into account volume changes which could result from the temperature switch. A significant increase of the ratio occurred with time over the nucleus, whereas it remained steady over mitochondria after the initial increase following incubation at 37°C. Since the nucleus represents one third of the cell volume, this ratio remains low; thus, if all internalized grains would overlay the nucleus, the ratio would not exceed 2.25, whereas it could reach 14 for mitochondria (-5.5% of the total celI volume). In 15 min, the ratio over the nucleus reaches 22% of the maximum value and only 7% over mitochondria. This suggests that some radiolabeled hormone translocates to the nucleus. We realize, however, that our kinetic and statistical approach does not provide direct evidence for interaction of PRL with nuclear constituents and that binding of PRL to chromatin or to the nuclear membrane will have to be analyzed as recently re- ported for nerve growth factor (52). Finally, it is of interest that the kinetic of uptake of prolactin by mammary cells parallels the rapid increase in rate of casein gene transcription by lactogenic hormone.”

In conclusion, lactogenic hormones bind to dispersed mam- mary cells as a saturable, reversible, temperature-dependent process. Variation of receptor concentration occurs through- out gestation and lactation and could represent an essential parameter modulating prolactin activity. Internalization of lactogenic hormone, which has been visualized morphologi- cally in dispersed mammary cells, could represent a means by which the hormone regulates gene expression in a manner similar to steroid hormones.

Acknowledgments-We wish to acknowledge Dr. C. H. Li, Univer- sity of California, San Francisco, for providing us with oPRL, hGH, and hCS; Dr. A. F. Parlow and the National Institute of Arthritis, Metabolism, and Digestive Diseases for rat hormones; and Dr. E. de1 Pozo, Sandoz Ltd., Basel, for bromocryptine. We are specially in- debted to Liliane Racine and Barbara M. Miller who collaborated in the experiments described in this paper, and to Alec J. S. Robert and Dr. P. C. Sizonenko who measured progesterone in the blood samples. We are grateful to Dr. Alan S. McNetiy who measured rabbit prolac- tin nlasma levels in this studv. We thank Dr. H. R. MacDonald for critically revising the manuschpt and Margot Korn for her secretarial assistance.

1.

2.

3.

4. 5.

6.

7. 8.

9.

10.

11. 12.

REFERENCES

Aubert, M. L., Grumbach, M. M., and Kaplan, S. L. (1974) Acta Endocrinol. 77,460-476

Aubert, M. L., Suard, Y., Sizonenko, P. C., and Kraehenbuhl, J. P. (1978) in Progress in Prolactin Physiology and Pathology (Robvn, C., and Harter, M., eds) pp. 45-57, Elsevier, Amster- dam

Birkinshaw, M., and Falconer, I. R. (1972) J. Endocrinol. 55, 323-334

Carpenter, G., and Cohen, S. (1976) J. Cell Biol. 71, 159-171 Carpenter, G.. Lembach, K. J.. Morrison. M. M.. and Cohen, S.

(i975) j. Biol. Chem. i50,4297-4304 Carpentier, J. L., Gorden, P., Amhardt, M., Van Obberchen, E.,

Kuhn, L. R., and Orci, L. (1978) J. Clin. Invest. 61, 1057-1070 Chamness, G. C., and McGuire, W. L. (1975) Steroids 26,538-542 Cuatrecasas, P., and Parikh, I. (1974) in Enzymology (Jacoby, W.

B., and Wilche, M. K., eds) Vol. 34, Part B, pp. 670-688, Academic Press, New York

Das, M., and Fox, F. (1978) Proc. Natl. Acad. Sci. U. S. A. 75, 2644-2648

Devinoy, E., and Houdebine, L.-M. (1977) Eur. J. Biochem. 75, 411-416

Djiane, J., and Durand, P. (1977) Nature (London) 266,641~643 Djiane, J., Durand, P., and Kelly, P. A. (1977) Endocrinology

100, 1348-1356

3 W. A. Guyette, R. J. Matasik, and J. M. Rosen (1979) Cell, in press.

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

10474 Receptors of Lactogenic Hormones on Dispersed Mammary Cells

13. Donner, D. M., Martin, D. W., and Sonenberg, M. (1978) Proc. 34. Rodbard, D. (1974) Clin. Chem. 20, 1255-1270 Natl. Acad. Sci. U. S. A. V&672-676 35. Rosen, J. M., and Barker, S. W. (1976) Biochemistry 15, 5272-

14. Frantz, A. G., Kleinberg, D. L., and Noel, G. L. (1972) Recent 5280 Prog. Horm. Res. 28, 527-590 36. Rosen, J. M., O’Neal, D. L., McHugh, J. E., and Comstock, J. P.

15. Gavin, J. R., III, Gorden, P., Roth, J., Archer, J. A., and BueII, D. N. (1973) J. Biol. Chem. 248, 2202-2207

(1978) Biochemistry 17, 290-297 37. Rosen, J. M., Woo, S. L. C., and Comstock, J. P. (1975) Biochem-

16. Gerasino, P., Djiane, J., and Kelly, P. A. (1979) Mol. Cell. Endo- istry 14,2895-2903 crinol. 13, II-23 38. Sakai, S., Enami, J., Nandi, S., and Banarjee, M. R. (1978) Mol.

17. Goldline, I. D., Smith, G., Wong, K. Y., and Jones, A. L. (1977) Cell. Endocrinol. 12, 285-298 Proc. Natl. Acad. Sci. U. S. A. 74, 1368-1372 39. Salpeter, M. M., and Bachmann, L. (1972) in Principles and

18. Goldfine, I. D., Jones, A. L., Hradek, G. T., Wony, K. Y., and Techniques of Electron Microscopy (Hayat, M. A., ed) Vol. 2, Mooney, I. S. (1978) Science 202, 760-763 pp. 220-278, Van Nostrand Reinhold Co., New York

19. Goldstein, J. L., and Brown, M. S. (1977) Anna. Reu. Biochem. 46,897-930

40. Salpeter, M. M., Fertuck, H. C., and Salpeter, E. E. (1977) J. Cell Biol. 72, 161-173

20. Gunther, G. R., Schultz, G. S., Hull, B. E., Alicea, H. A., and 41. Schlessinger, J., Shechter, Y., Willingham, M. C., and Pastan, I. Jamieson, J. D. (1977) J. Cell. Biol. 75,368a (ST 803) (1978) Proc. Natl. Acad. Sci. U. S. A. 75, 2659-2663

21. Herington, A. C., Veith, N. M., and Burger, H. G. (1976) Biochem. 42. Shiu, R. P. C., and Friesen, H. G. (1974) Biochem. J. 140, 301- J. 158,61-69 311

22. HoIIenberg, M. D., and Cuatrecasas, P. (1975) J. Biol. Chem. 250, 43. Shiu, R. P. C., and Friesen, H. G. (1974) J. Biol. Chem. 249,7902- 3845-3853 7911

23. Houdebine, L. M. (1976) Eur. J. Biochem. 3,615-630 44. Shiu, R. P. C., and Friesen, H. G. (1976) Biochem. J. 157, 619- 24. Houdebine, L. M., and Gaye, P. (1976) Mol. Cell. Endocrinol. 3, 626

37-55 45. Shiu, R. P. C., and Friesen, H. G. (1976) Science 192, 259-261 25. Josefsberg, Z., Posner, B. I., Patel, B., Bergeron, J. J. M. (1979) J. 46. Shiu, R. P. C., Kelly, P. A., and Friesen, H. G. (1973) Science 180,

Biol. Chem. 254,209-214 968-971 26. Kissane, J. M., and Robins, E. (1958) J. Biol. Chem. 233,184-188 47. Shuster, R. T., Houdebine, L. M., and Gaye, P. (1976) Eur. J. 27. Kraehenbuhl, J. P. (1977) J. Cell Biol. 72, 390-405 Biochem. 71, 193-199 28. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. 48. Topper, Y. J. (1970) Recent Prog. Harm. Res. 26, 287-308

(1951) J. Biol. Chem. 193,265-275 49. Turkington, R. W., Majumder, G. C., Kadohama, N., McIndoe, J. 29. McNeiIIy, A. S., and Friesen, H. G. (1978) Endocrinology 102, H., and Frantz, W. L. (1973) Recent Prog. Horm. Res. 29,417-

1548-1554 455 30. Mendelson, C., Dufau, M., and Catt, K. (1975) J. Biol. Chem. 250, 50. Vigneri, R., Goldfine, I. D., Wong, K. Y., Smith, G. J., and Pezzino,

8818-8823 V. (1978) J. Biol. Chem. 253,2098-2103 31. Nolin, J. M., and Witorsch, R. J. (1976) Endocrinology 99, 949- 51. Weibel, E. R., and Bolander, R. P. (1973) in Principles and

958 Techniques of Electron Microscopy (Hayat, M. A., ed) p. 241, 32. O’Keefe, E., and Cuatrecasas, P. (1974) Biochim. Biophys. Acta Van Nostrand Reinhold, New York

343,64-77 52. Yankner, B. A., and Shooter, E. M. (1979) Proc. Natl. Acad. Sci. 33. Posner, B. I., Kelly. P. A.. and Friesen, H. G. (1975) Science 188, ZJ. S. A. 76, 1269-1273

57-59

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Receptors of Lactogenic Hormones on Dispersed Mammary Cells 10475

+ “,.-.- .I y %,

/ O<,

. I

NSE -,--..-- /-•--*

l --. e----e

02

P ::

-: n

= 0.1 E

0.0

0 2L L8 72 96 120 Time cd inc”batim,h,

I’ I

n o.2- p-- o--

; L % - 0.1 - N $

I / i-

o” - .

0.0 - l ----..--. p----~/o-”

0- 0 26 (8 72 96 120

Time of rxubation ,I,,

Day of dissociation

2 L 6 6

0 2 L 6 6 10 12 1‘ Day of incubumn

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Y M Suard, J P Kraehenbuhl and M L Aubertvirgin, pregnant, and lactating rabbits.

Dispersed mammary epithelial cells. Receptors of lactogenic hormones in

1979, 254:10466-10475.J. Biol. Chem. 

  http://www.jbc.org/content/254/20/10466Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/254/20/10466.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on July 2, 2020http://w

ww

.jbc.org/D

ownloaded from