| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mechanisms of Hormone Action |
Unité d'Endocrinologie Moléculaire,3 Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas Cedex, France
Gene Expression and Development,4 Roslin Institute, Midlothian EH25 9PS, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cytokines, early development, mammary glands, mechanisms of hormone action, prolactin receptor
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The involvement of PRL-R in mammary gland development has been illustrated by gene deletion experiments in mice [10, 11]. The heterozygous females show an abnormal mammogenesis characterized by a reduced lobuloalveolar development. These mice fail to produce sufficient amount of milk, and their maternal behavior is severely affected [12]. The prolactin receptor (PRL-R) is a membrane receptor with a single transmembrane domain that belongs to the class I cytokine receptor superfamily [13, 14]. These receptors are known to homo- or heterodimerize in response to ligand binding [15, 16]. A single PRL molecule induces the dimerization of two membrane-bound PRL-R molecules, resulting in the formation of an active trimeric hormone receptor complex [17]. The formation of this complex is the first step in signal transduction involving the cytoplasmic Janus tyrosine kinase 2 (Jak2) and signal transducers and activators of transcription (STAT 5a and b) pathways [15, 18].
An artificial mutant form of the rabbit (rb) PRL-R has been constructed in our laboratory [19]. This mutant has a deletion in the extracellular domain of 200 amino acids proximal to the transmembrane domain. We have shown previously that this mutant PRL-R is endowed with a constitutive activity on milk protein gene transcription. In the present study, we addressed the question of how this mutant PRL-R elicits a ligand-independent signal transduction and analyzed if the permanent signal may be activated simply by constitutive homodimerization of this receptor. The present study also reports the in vivo expression of the mutant PRL-R cDNA in transgenic mice. This was undertaken in order to establish the in vivo effects of a permanent PRL-R activation on mammogenesis, lactogenesis, galactopoiesis, and involution. To target expression to the mammary gland, we used regulatory sequences from the sheep ß-lactoglobulin (BLG) gene. Expression of the mutant PRL-R was thus specifically directed to the mammary epithelial cells [20].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents and Antibodies
Ovine PRL was kindly supplied by the National Hormone and Pituitary Program (Torrance, CA). Monoclonal M2 antibody, recognizing the Flag peptide sequence, and anti-mouse immunoglobulin G (IgG) conjugated to horseradish peroxidase were purchased from Sigma (St. Louis, MO). Monoclonal antibody 9E10, recognizing the cMyc peptide sequence, was obtained from Euromedex (Mundolsheim, France). Synthetic oligonucleotides were from Genset (Paris, France).
Construction of Plasmids
Deletion of the extracellular domain of the PRL-R generating construct p
103–203 has been previously described (Fig. 1A) [19]. The Flag tag (encoding the amino acids [aa] DYKDDDDK) and the cMyc tag (encoding the aa EQKLISEEDLLR) were appended at aa 516 of deleted rbPRL-R protein, just prior to the stop translational codon, by polymerase chain reaction (PCR) of the rbPRL-R cDNA. The 5' sense oligonucleotide primer corresponds to the sequence that extends from nucleotide (nt) 1022 to 1047 of the rbPRL-R cDNA sequence and includes a BalI site. The 3' antisense primers contained the sequence recognized by the XhoI restriction enzyme, the stop codon, the Flag or cMyc peptide, and the sequence of the PRL-R from nucleotide 1540 to 1548. Amplification products were isolated from agarose gels and ligated into the pGEM-T vector (Promega, Lyon, France). After sequencing, each BalI/XhoI DNA fragment was subcloned into the BalI/XhoI sites of the pECE expression vector [21] containing the wild-type PRL-R cDNA in which the 102–203 region has been previously deleted. These constructs are termed Flag-R and cMyc-R.
|
Transfection and Immunoblot Analysis
COS-7 monkey kidney cells were maintained in Dulbecco modified Eagle medium containing 10% (v/v) fetal calf serum. Transfections were performed by the ExGen 500 method (Euromedex) with appropriate pECE vector containing either the wild-type or the deleted PRL-R cDNA tagged with Flag or/and cMyc epitope. Prior to stimulation with PRL (400 ng/ml), transfected COS-7 cells were starved overnight in medium without serum. After washing with ice-cold phosphate-buffered saline, cells were solubilized in lysis buffer (1% Nonidet P-40 [NP-40], 20 mM Tris-HCl [pH 8], 137 mM NaCl, 2.7 mM KCl, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 2 µg/ml each of aprotinin, leupeptin, and pepstatin). Clear lysates obtained by centrifugation were incubated for 4 h at 4°C with anti-Flag or anti-cMyc antibodies. Immunocomplexes were collected using protein G-Sepharose (Sigma, Saint Quentin Fallavier, France). Samples were centrifuged and washed three times with lysis buffer, without NP-40 and glycerol, and then boiled 5 min in Laemmli loading buffer. Immunoprecipitates were separated by SDS-polyacrylamide gel electrophoresis (14% gel) and transferred onto nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany). Blots were blocked for 30 min in washing buffer (20 mM Tris-HCl [pH 7.4], 137 mM NaCl) supplemented with 5% nonfat dry milk, with (for Flag) or without (for cMyc) 0.1% Tween 20. Membranes were then incubated in the same buffer with the appropriate primary antibody (as indicated in Fig. 2), washed, and then incubated for 1 h with an anti-mouse IgG conjugated to horseradish peroxidase. Proteins bands were detected using an enhanced chemiluminescence (ECL) detection system (Amersham, Les Ulis, France). The blots were reprobed with another antibody after stripping with a specific buffer (62.5 mM Tris-HCl [pH 6.8], 2% SDS, and 100 mM ß-mercaptoethanol) at 60°C for 30 min.
|
Production of Transgenic p103–203 Mice
The 2.014-kilobase (kb) HindIII/XhoI fragment of the mutant PRL-R cDNA (p103–203) was subcloned into the unique EcoRV restriction site of the plasmid pMAD6. This plasmid is a truncated BLG gene consisting of 2.6 kb of 5'-flanking promoter sequences of the ovine BLG gene up to about 30 bases of exon 1, followed by exon 6 and intron 6, exon 7 plus 4.2 kb of 3' BLG flanking sequence (Fig. 3). A unique EcoRV site was generated between exon 1 and exon 6 deleting the intervening sequences. Thus, mice carrying the transgene express the constitutive PRL-R under the control of the BLG milk protein gene promoter [20]. For microinjection, the construct PRL-R/BLG (8.4 kb) was isolated free of vector sequences by digestion with MluI and purification by agarose gel electrophoresis. DNA was microinjected into pronuclear-stage mouse eggs obtained from (CBA x C57BL/6) F1 female mice in order to generate transgenic mice. Microinjected eggs were transferred into pseudopregnant recipient mice. Founder mice were identified by PCR analysis of tail biopsies. Five transgenic mice were identified among the 84 offspring. These five transgenic founders were bred to establish transgenic lines. The copy number of the PRL-R/BLG transgene in each line was determined by Southern blotting from animals of the F1 generation. It ranged from one to five copies of the transgene per genome. Two transgenic lines (30 and 56) were further studied.
|
Reverse Transcription-PCR Analysis of Transgene Expression
Total RNAs from mammary glands of virgin mice were extracted by the acid guanidium thiocyanate-phenol-chloroform method [22]. The reverse transcription (RT) reaction was performed using 2 µg of total RNA with 100 U Superscript II reverse transcriptase (Gibco BRL-Life Technologies, Invitrogen, Carlsbad, CA) and 650 ng random primers in a final volume of 20 µl, for 50 min at 42°C. Two microliters of each RT reaction product were amplified by PCR in a final volume of 50 µl with 200 µM of each deoxynucleotide triphosphate, 45 pmol of each primer, 2.5 U Taq DNA Polymerase (Oncor-Appligene, Illkirch, France), and 1x PCR buffer (Oncor-Appligene). The PCR mixture contained the following primers: 5'-GAGGAGCCCCAGGCTAATC-3', forward primer (nt 1093–1111) specific to the rbPRL-R cDNA sequence and 5'-GACTAGAAGGGACCAGGACTG-3', and reverse primer (nt 760–780) specific to the exon 7 ovine BLG gene. A 964-base pair (bp) amplification product was generated from the transgenic mice between these primers. Samples were amplified in a Thermocycler (Perkin-Elmer, St Quentin-en-Yvelines, France) for 30 cycles in the following sequence: denaturation at 94°C for 1 min, annealing at 60°C for 1 min 30 sec, extension at 72°C for 2 min. PCR products were separated on a 1% agarose gel and blotted onto a nylon membrane. The blot was hybridized with a 
-32P-labeled probe (3 x 106 cpm/ml; Amersham) specific to the rbPRL-R cDNA (nt 942–1940).
Morphologic Analysis
For preparing whole mount, number 4 mammary glands were surgically removed, fixed overnight in Carnoy solution, washed in 70% ethanol for 15 min, and then stained with carmine alum for 6–24 h at 4°C to visualize ductal trees. After washing in graded ethanol, tissues were stored in xylene. For histological examination, mammary tissues sampled at different stages of development were fixed in 2% paraformaldehyde in phosphate-buffered saline (PBS) for 4–12 h at 4°C. Then samples were cryoprotected in 40% sucrose in PBS, embedded in TissuTek (Sakura, Torrance, CA), and put in liquid nitrogen before storing at -80°C. Tissue sections (5 µm) were prepared and stained with Groot hematoxylin and erythrosine solution or in toluidine blue. Sections were dehydrated in graded ethanol, mounted in Eukitt (Electron Microscopy Sciences, Hatfield, PA), and examined by light microscopy. For immunohistochemistry, frozen 5-µm sections were permeabilized in 2% BSA, 0.05% saponin, and 0.05% sodium azide in PBS (PBS dilution buffer) for 1 h, followed by treatment in 50 mM ammonium chloride for 30 min. Rabbit polyclonal anti-S1-casein antibody (generous gift from Dr. Barash, ARO, Volcani Center, Bet Dragan, Israel [23]) was diluted 1:250 and then added onto the tissue sections for 1 h at room temperature. Antibody binding was visualized with fluorescein isothiocyanate (FITC)-labeled secondary antibodies (1:500; Jackson ImmunoResearch, West Grove, PA), applied onto sections for 45 min. DAPI (4,6-diamidino-2-phenylindole) was diluted 1:500 and applied for 3 min at room temperature. Between incubations, mammary sections were washed five times. After the final wash, they were mounted in Vectashield mounting medium (Vector Laboratories Inc., Burlingame, CA) and observed with a Leica DMRB microscope coupled with a DP50 Olympus Camera (Tokyo, Japan).
For whole-mount and histological studies of mammary glands, at least three mice have been analyzed.
Milk Collection and Analysis
Milk samples were collected on Day 4 of lactation from transgenic and nontransgenic control females. The mothers were intraperitoneally injected with 2 U oxytocin (Sigma) to stimulate milk ejection. The skim milk samples were diluted 1:10 in 10 mM Tris-HCl (pH 7.8), 100 mM CaCl2, and total milk protein content was estimated by the Bradford procedure. Samples (3 µg of proteins for blotting and 50 µg for Coomassie Blue staining) were then added to an equivalent volume of SDS loading buffer and resolved on a standard 12% SDS-PAGE gel. Proteins on the gel were stained with Coomassie Blue (0.1%) or blotted onto a nitrocellulose membrane (BA 85; Schleicher and Schuell, Keene, NH).
Western Blot Analysis
Mouse mammary lysates were prepared in 50 mM Tris-HCl (pH 8), 137 mM NaCl, 2.7 mM KCl, 1% NP40, 10% glycerol, containing protease and phosphatase inhibitors. Proteins were separated on 12% SDS-polyacrylamide gel and electrophoretically transferred onto nitrocellulose filters. Membranes were saturated for 30 min in washing buffer (see Transfection and Immunoblot Analysis section) and incubated for 1 h with the anti-S1-casein antibodies (1:4000) [23]. After washing, the blot was incubated for 45 min at room temperature with an anti-rabbit IgG coupled to horseradish peroxidase (1:15 000; ICN Biomedicals, Orsay, France). Immune complexes were detected using the ECL system (Amersham).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have previously constructed a mutant form of the PRL-R, p103–203, in which the second extracellular subdomain is deleted (Fig. 1A). This receptor was responsible for a constitutive activity on transactivation of the ß-lactoglobulin promoter. Moreover, in the mouse mammary epithelial cell line HC11, this deleted receptor form was able to activate the transcription of the endogenous ß-casein gene in the absence of PRL [19]. To explain this constitutive signal transduction process, it may be suggested that PRL-R mutant dimerizes spontaneously. In order to test this hypothesis, we constructed two expression plasmids containing the deleted cDNA tagged with a different epitope at its carboxyl-terminal site (see Materials and Methods). We inserted sequences encoding peptide epitopes recognized by monoclonal antibodies: anti-Flag (DYKDDDDK) or anti-cMyc (KQKLISEEDL) epitope.
Each tagged expression construct was transiently transfected into COS-7 cells. Cells were solubilized and subjected to immunoprecipitation followed by immunoblotting using either anti-Flag or anti-cMyc antibody. As illustrated in Figure 1B (top), the Flag-R (approximately 68 kDa) was efficiently expressed and specifically recognized by the anti-Flag. This antibody did not immunoprecipitate nonspecific components from COS-7 cell extracts. In addition, since the cMyc-R was not immunoprecipitated by the anti-Flag, this result confirmed the specificity of anti-Flag antibody. We then investigated whether both tagged constructs were coimmunoprecipitated when they are coexpressed and if the PRL-R forms dimers. In this case, anti-Flag or anti-cMyc antibody would immunoprecipitate a complex containing the Flag- as well as the cMyc-tagged receptor. To test this possibility, COS-7 cells were cotransfected with both tagged expression constructs. After cell solubilization, lysates underwent immunoprecipitation with anti-Flag and immunoblotting with anti-cMyc. As shown in Figure 1B (bottom), the cMyc-R was detected in the anti-Flag immunoprecipitates and anti-cMyc immunoblots. It is not the case in the absence of the Flag-R, which indicates that this immunoprecipitation may be due to the association of the cMyc-R with the Flag-R. The same blot was reprobed with the anti-Flag antibody to confirm that a comparable amount of Flag protein was present in immunoprecipitates of cells coexpressing both constructs. From cells expressing both tagged receptors, the Flag-R was also detected in the anti-cMyc immunoprecipitates (data not shown). This may suggest that the Flag-R construct forms immunoprecipitable complexes with the cMyc receptor.
These results demonstrate that the mutant PRL-R undergoes spontaneous homodimerization, which may explain its constitutive activity.
Development of Transgenic Mice Expressing Constitutively Active PRL-R Mutant
To examine the biological effects of the p103–203 PRL-R in vivo, we generated transgenic mice expressing this receptor. To direct expression to the mammary epithelium, a HindIII/XhoI (2-kb) fragment of the cDNA was cloned into the pMAD6 plasmid, which is a sheep BLG minigene, predominantly expressed in ductal and alveolar cells (Fig. 2A). This construct was used to generate transgenic mice by microinjection into fertilized oocytes from (C57Bl/6 x CBA) F1 mice. Two transgenic animals (founder mice 56 and 30) were identified as passing the transgene to female and male offspring alike. Southern blot analysis of the transgenic lines indicated that line 56 had a single-copy insert, whereas line 30 had a multiple-copy insert (five copies). Expression of the transgene mRNA was analyzed by RT-PCR of total mammary gland RNA prepared from virgin transgenic and nontransgenic control females (Fig. 2B). As soon as 8 wk old, both virgin transgenic lines showed detectable expression levels of the transgene mRNA of the expected size (1 kb) in mammary glands, confirming that the transgene was well expressed in this tissue. As shown in Figure 2B, the level of transgene expression was much higher in line 30 than in line 56. The transgene transcript was also detected in mammary glands from pregnant, lactating, and post-weaning mice (data not shown) and was not observed in wild-type mice. Tissue specificity of the transgene expression was also analyzed by RT-PCR analysis. No ectopic expression was found in brain, testis, ovaries, kidney, and liver.
Phenotypic Analysis of Mammary Gland Development in Transgenic Mice
All male and female offspring that expressed the transgene appeared normal and were fertile. To detect differences between mammary tissue from transgenic and nontransgenic mice, we used whole-mount analysis, a technique that allows visualization of the ductal and alveolar structures, and histological analysis.
Lobuloalveolar development
The results obtained from two stages of mammary development—virgin (20 wk old) and 15-day pregnant mice—are shown in Figures 3 and 4. In mature virgin wild-type mice, the mammary gland exhibited widely scattered ducts with terminal and lateral alveolar bud formation at their extremities (Fig. 3A, a and d). In contrast, in virgin transgenic mice from both lines, the mammary gland was more strongly developed (Fig. 3A, b and c). Clear demonstration was observed with line 30 mice, where mammary ducts are grossly dilated compared with those of wild-type mothers (Fig. 3A, f vs. d). In addition, for both lines, there were many more developing alveoli that are directly branched to large ductal structures (Fig. 3A, e and f). Histological analysis of thin sections from virgin mammary tissue confirmed these observations. Whereas the fat pad was preponderant in wild-type virgin mice, more alveoli were found in transgenic mice. Mammary ducts of transgenic mice were very large, and an abundant secretion was detected inside the alveolar lumina by immunohistochemical analysis (Fig. 3A, h vs. g). Moreover, virgin wild-type mammary ducts and alveoli were surrounded by a basal layer of cells (Fig. 3B). In contrast, in line 30, the lumina of the alveoli were lined with a single and not continuous layer of cells. To characterize the state of basal cell differentiation, we examined their ability to produce the S1-casein milk protein. Figure 3C shows a representative Western blot for expression levels of S1-casein in mammary gland extracts from wild-type and transgenic virgin (20 wk old) or pregnant (15 days) mice. The immunoblot revealed a protein band with a molecular mass of 42 kDa in both virgin transgenic lines, confirming the expression of the S1-casein gene. This protein band was not detected in the mammary tissue of wild-type virgin mice. We observed a clear difference in the amplitude of S1-casein expression between the two virgin transgenic lines; the expression level of S1-casein was higher in line 30. By contrast, during pregnancy, wild-type and transgenic mice lines expressed the S1-casein protein approximately at the same level. Thus, our results show that this milk protein is precociously produced in virgin transgenic mice as the consequence of a premature activation of the PRL-R signal transduction pathway.
|
On Day 15 of pregnancy, the development of normal mammary glands is characterized by the extension of the ductal structures (Fig. 4, a and c). The number and size of alveoli had increased, and they started to form larger lobules. In glands from pregnant transgenic mice, particularly in line 30, the ductal structures were again hypertrophied, and the numbers of alveolar structures significantly increased compared to those of pregnant nontransgenic animals (Fig. 4, f vs. d). When examined histologically, mammary epithelial structures were found to be more developed than in wild-type mice with an increase of the numbers and size of the alveoli and sparse duct tissue embedded in adipose tissue (data not shown). In transgenic mice, less adipose tissue was present, and casein production in alveolar structures was clearly detectable by immunohistochemistry.
Mammary gland structure during lactation and milk production Analyses of mammary whole mount after parturition (Day 3 of lactation) showed that the entire size of the mammary gland appeared to be similar between transgenic and nontransgenic animals with very dense lobuloalveolar structures (Fig. 5, a–f). However, by histological examination, epithelial development appeared to be strongly altered in transgenic animals. The lobuloalveolar structures were not well distributed, and very dilated ducts were filled with secreted proteins (Fig. 5, g–i). The nature of these secretion products is presently unclear, but they are abundant and dense and remain inside ducts and in the lumen of large alveoli (Fig. 5, i).
|
Consequently, milk production in transgenic lines proceeded with obvious problems. A marked lactation defect was apparent in transgenic females and could be illustrated by the lack of weight gain of the pups even if the offspring appeared to suckle normally (Fig. 6A). By comparison with the body weight of wild-type littermate controls, pups from lines 56 and 30 did not grow after birth and then did not survive for more than a few days. As illustrated in Figure 6B, all pups died within 6 days after birth (line 30). No milk was present in the stomach of the dead transgenic pups from line 30. The only way to maintain the newborn pups from both lines was to have them fostered by wild-type lactating mothers. In these conditions, all these animals exhibited normal suckling behavior and grew normally. To verify the presence of milk in the mammary gland of the transgenic mothers, after first suckling with their wild-type mothers, wild-type pups were given to the line 30 mothers. The majority of the wild-type pups lost weight and died a few days later, suggesting that the line 30 mothers could not produce or deliver sufficient quantities of milk to sustain normal development of the pups.
|
In conclusion, both transgenic mouse lines had strongly altered milk production, particularly line 30. This phenotype was correlated with the copy number of the transgene. The line with the lower copy number (one copy, line 56) showed less alteration in milk production than line 30, characterized by a high copy number (five copies).
The occurrence of a lactational defect might be partially explained by a change in milk composition. To investigate this point, milk samples were collected in early lactation in normal and transgenic animals after oxytocin intraperitoneal injection. Defatted and diluted milk was prepared and separated by SDS-PAGE. Coomassie Blue staining revealed a similar electrophoretic pattern of the major milk proteins in all the samples tested (data not shown). The level of S1-casein in the milk evaluated by Western blot analysis was similar in normal and transgenic mice (Fig. 7). This milk protein was detected as a major 42-kDa band and two minor ones, of higher and lower molecular masses.
|
This result indicated clearly that, during early lactation, the amplitude of milk protein gene expression was similar, though its timing may have been different, in normal and transgenic mice, suggesting that the failure to nourish the pups may be related to the drastic decrease in the ability of the gland to secrete milk consequently to the histological disorganization of the mammary gland observed in transgenic animals.
Involution of the mammary gland Three days after pup removal, there was a large difference in the development of mammary glands between wild-type and transgenic mice (Fig. 8). In wild-type mice, a marked regression of the lobuloalveolar structures was observed. In contrast, in the transgenic animals, particularly in line 30, alveolar structures remained intact. Histological analyses confirmed the delay in involution (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Immunoprecipitation and immunoblotting experiments presented in this report strongly suggest that the constitutively active PRL-R exists as permanent homodimer in the absence of PRL. This ligand-independent homodimerization process has previously been reported for cytokine receptors [24]. Previous studies have shown that PRL stimulation induces a conformational change in the extracellular domain of the receptor that is necessary for optimum alignment and bonding of residues involved in dimerization [25, 26]. This stabilizes the receptor dimer in a conformation that allows activation of cytoplasmic signaling cascades. Thus, the absence of the S2 subdomain in the mutant PRL-R can lead to the forced dimerization of the receptor. This permanent homodimerization is sufficient to transduce and trigger cytoplasmic signaling by activation of known PRL signaling molecules, supporting the view that changes in the conformation of the extracellular domain and proximity of Jak2 are the essential elements involved in the constitutive signaling. Other investigators have constructed a constitutive form of the PRL-R by deleting 178 amino acids from its extracellular ligand-binding domain [27]. These authors suggested the involvement of the WSXWS motif in the constitutive activation of the receptor, a domain that is deleted in our mutant [19]. These results may indicate that constitutive homodimerization of the PRL-R may be induced by different mechanisms. In this study, we were unable to detect the dimerization state of the native PRL-R following PRL stimulation. This dimerization process would be too transient, as shown in our previous study using Biacore analysis [17], to be easily detected by immunoprecipitation and Western blotting experiments.
The Mutant PRL-R to Analyze Mammary Gland Development
Using the mutant PRL-R transgenic mice expressing the constitutively active PRL-R, we illustrated the necessity of establishing the appropriate PRL-R expression pattern for harmonious mammary gland development.
Mammary development involves interplay between growth factors, lactogenic hormones, and steroids [28, 29]. These factors act in a precise chronology to allow the development and maintenance of structures such as ducts and alveoli. However, alterations in their expression may severely impair ductal and lobuloalveolar development [30]. The use of gene knockout mice has helped resolve the functions of various hormonal signals involved in mammary development. Thus, mice with disruption of genes encoding progesterone receptor [31], PRL-R [10], and PRL [7] do not develop lobuloalveolar structures. In contrast, the morphology of estrogen receptor (ER) knockout mammary glands indicated that estrogen/ER signaling is required at an earlier stage to induce ductal elongation [32, 33]. Nevertheless, in the studies using PRL-R knockout mice, it has not been possible to investigate mammary development during pregnancy, as homozygous females are infertile [10]. The transgenic model described in the present study made it possible to assess the precise role of PRL-R activation during mammary development since the expression of the constitutively active PRL-R exclusively in the mammary gland did not disrupt the reproductive function of the mice.
We produced transgenic mice using a transgene construct with the BLG promoter to direct expression of the p103–203 receptor specifically to the mammary gland, BLG being expressed exclusively in mammary epithelial cells. It appeared that during mammary gland development, transgene expression differed from that of the endogenous PRL-R gene. This could be explained by the fact that the BLG gene is already expressed in virgin mice. The BLG promoter is probably stimulated by Stat5, which is known to be activated in the mammary gland during specific periods of the estrous cycle [34]. Thus, the constitutively active PRL-R may be expressed in very early phases of mammary gland development. In these conditions, the precocious expression of the PRL-R may explain how PRL regulates lobuloalveolar development.
Abnormal Lobuloalveolar Development in Transgenic Mice
Our results clearly indicated that the increase in the sensitivity to PRL may determine, even before pregnancy, the magnitude of ducts and lobuloalveolar development. Indeed, the growth and development of these structures were strongly stimulated by the expression of the mutant receptor, even in early phases of development of the mammary gland (20-wk-old virgin animals), leading to an abnormal organization. Thus, the phenotype observed in these transgenic animals was likely to reflect the proliferative role of PRL on ductal cells during early mammary gland development.
At midpregnancy, histological examination of mammary glands showed that the number of alveoli was much higher in transgenic versus normal animals. The ductal network appeared strongly overdeveloped with very large structures. Moreover, at this stage, we observed a functional differentiation of the lobuloalveolar structures, which appeared dilated and contained secretory products. Unexpectedly, lactating transgenic females were unable to feed their pups, particularly in the line 30 offspring. The pups lost weight and died despite the fact that the milk protein composition did not show marked abnormalities in terms of the presence of S1-casein production. The major defect concerns the capacity of these animals to secrete a sufficient amount of milk to nourish their pups. Experiments are in progress in our laboratory to examine the ultrastructural aspect of the secretory machinery of the mammary alveolar cells in transgenic mice.
This observation that milk production was extremely low during lactation may indicate that the precocious activation of the PRL signal may lead to an abnormal ductal and lobuloalveolar development that consequently impairs harmonious histological organization for the secretion of milk during lactation. This suggestion was confirmed by a careful histological examination of these structures. In virgin wild-type mice, ducts are composed of multilayered basal cells. In contrast, in transgenic mice (line 30), alveoli were lined by a single layer of cells. Consequently, in our transgenic model, the defect in milk secretion during lactation may be explained by this aberrant mammary development during puberty and early pregnancy, which may concern the differentiation state of the myoepithelial cells involved in milk ejection. Nevertheless, the alterations in the proportion of this cell type in transgenic mice are still to be demonstrated.
This transgenic mouse model may be particularly adapted to studying the molecular events involved in the involution process of the mammary gland, normally occurring at the end of lactation in response to milk stasis. This involution process is strongly reduced in the transgenic animals, even in the absence of suckling. The present study does not bring quantitative evaluation of the involution kinetics. A more precise analysis of this phenomenon needs to be performed. In addition, it will be interesting to clarify the mechanisms involved in this dramatic curtailing of the involution process.
These observations indicate the importance of establishing the correct pattern of PRL-R expression during mammary morphogenesis in order to avoid defects in ductal and lobuloalveolar development. This is supported by the observation that, in virgin transgenic mice (20 wk of age), we observed an aberrant mammary development characterized by a direct expansion of alveoli from the ductal structures in an apparently complete absence of lateral branching. Several studies have reported the requirement of high concentration of progesterone for the growth of side branches in early pregnancy [35–37]. Our results are in agreement with the concept that PRL stimulation without high progesterone levels may lead to an abnormal development of the mammary gland, particularly during the formation of ductal side branching. Additionally, the inappropriate expression of PRL-R may alter the proliferative response of mammary epithelial cells to steroids and lactogenic hormones whose associated and coordinated actions are involved in successful establishment and maintenance of pregnancy with harmonious development of the mammary gland. It must be kept in mind that during the massive growth of the mammary gland in pregnancy, PRL-R remains at low levels [38] and that progesterone appears to have a limiting effect in the PRL induction of PRL receptor expression [39].
Changes in PRL intracellular signaling induced by the expression of the constitutively active receptor is an important feature that remains to be studied. We have previously shown that expression of this mutant receptor in transfected COS cells induced a PRL-independent stimulation of Stat5a and Stat5b [40]. The role of Stat factors in the regulation of milk protein gene expression has been widely documented. In particular, it has been shown that mammary glands from Stat5a knockout mice failed to develop during pregnancy [41, 42]. Moreover, during the involution process, Stat3 is specifically activated in parallel with the apoptosis of epithelial cells [43, 44]. Changes in the expression level and activity of Stat5 and Stat3 transcripts in our transgenic mouse model during mammary gland development are under investigation.
In addition, as recently described in the corpus luteum and decidua, PRL is able to stimulate the transcription of and ß estrogen receptors after activation of Stat5a and Stat5b [40]. It is well known that estrogens are required for inducing and differentiating mammary epithelial cells, leading to ductal and alveolar structure development [5, 45–47]. It is conceivable that the precocious mammary development observed in the virgin transgenic mice may be the consequence of overexpression of estrogen receptors induced by the expression of the constitutively active PRL-R. Thus, the inappropriate expression pattern of PRL concomitant with changes in estrogen expression may disrupt the normal hormone patterning necessary for harmonious mammary gland development.
The protein kinase Akt has been recently shown to represent an important signaling molecule involved in mammary gland development and function, and this factor is dependent on PRL stimulation [48]. In addition, expression of constitutively activated Akt in transgenic mice leads to lactational defects and delayed involution of the mammary gland [49, 50]. This phenotype is very similar to the one we observed in our study, suggesting that one of the consequences of the overexpression of the constitutively active PRL-R may be an overstimulation of the mammary cells by Akt and alteration of the secretory machinery. Indeed, expression of constitutively activated Akt leads to an excess of lipid synthesis and lactational defects. In further studies, it will be interesting to analyze Akt expression in our transgenic mice model.
Consequently, our transgenic mice may be a useful tool to study in vivo the involvement of PRL in the early phases of mammary gland development. The use of these animals may also make it possible to analyze the complex coregulation of steroids and PRL observed in mammary epithelial cells.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Isabelle Gourdou, Unité d'Endocrinologie Moléculaire, Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas Cedex, France. FAX: 0033 1 3465 2241; gourdou{at}jouy.inra.fr
Received: 10 June 2003.
First decision: 25 June 2003.
Accepted: 10 November 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and Erß transcription. Mol Endocrinol 2001 15:2172-2181This article has been cited by other articles:
|
|
 |
|
  R. L. Bogorad, C. Courtillot, C. Mestayer, S. Bernichtein, L. Harutyunyan, J.-B. Jomain, A. Bachelot, F. Kuttenn, P. A. Kelly, V. Goffin, et al. Identification of a gain-of-function mutation of the prolactin receptor in women with benign breast tumors PNAS, September 23, 2008; 105(38): 14533 - 14538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Manhes, C. Kayser, P. Bertheau, B. Kelder, J. J Kopchick, P. A Kelly, P. Touraine, and V. Goffin Local over-expression of prolactin in differentiating mouse mammary gland induces functional defects and benign lesions, but no carcinoma. J. Endocrinol., August 1, 2006; 190(2): 271 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Farmer and M.-F. Palin Exogenous prolactin stimulates mammary development and alters expression of prolactin-related genes in prepubertal gilts J Anim Sci, April 1, 2005; 83(4): 825 - 832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Li, K. G. Suresh Kumar, W. Tang, V. S. Spiegelman, and S. Y. Fuchs Negative Regulation of Prolactin Receptor Stability and Signaling Mediated by SCF{beta}-TrCP E3 Ubiquitin Ligase Mol. Cell. Biol., May 1, 2004; 24(9): 4038 - 4048. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||
