J. Dairy Sci. 86:2864-2874
© American Dairy Science Association, 2003.
Mammary Localization and Abundance of Laminin, Fibronectin, and Collagen IV Proteins in Prepubertal Heifers
S. D. K. Berry*,
R. D. Howard
and
R. M. Akers*
* Department of Dairy Science, Virginia Tech, Blacksburg, 24061
Virginia Maryland Regional College of Veterinary Medicine, Blacksburg, VA 24061
1 Corresponding author: R. M. Akers; e-mail: rma{at}vt.edu.
 |
ABSTRACT
|
|---|
The objective was to determine localization and abundance of extracellular matrix proteins fibronectin, laminin, and collagen in mammary tissues from ovariectomized or intact prepubertal heifers. Mammary parenchyma and fat pad tissues were collected from 14 6-mo-old heifers: eight were ovariectomized between 1 to 3 mo of age, and six were used as intact controls. Distribution of total collagen was assessed by Sirius Red staining of tissue sections. Fibronectin, laminin, and type IV collagen were assessed by immunohistochemistry. Abundance of fibronectin and laminin was also analyzed by western blotting. Total mammary mass was much less in ovariectomized animals (130 ± 21 vs. 304 ± 25 g). Histological structure differed as parenchyma from intact animals contained abundant, complex branching epithelial terminal ductular units, whereas terminal ductular units from ovariectomized animals were mostly major ductal structures with little or no branching. Collagen fibers were abundant and densely packed throughout interlobular stroma and were less abundant and more diffuse within intralobular stroma. Type IV collagen was primarily in basal lamina of mature ducts, whereas fibronectin and laminin staining were present throughout parenchymal stroma, in both intact and ovariectomized animals. Using western blotting, fibronectin was more abundant within parenchyma than in the fat pad and significantly higher in parenchyma from ovariectomized heifers. Laminin was more abundant in parenchyma from intact than ovariectomized animals (30 vs. 17 densitometric units/mg of tissue), but laminin was similar between parenchyma and fat pad. These results provide initial evidence that fibronectin, laminin, and collagen participate in regulation of heifer prepubertal mammary development.
Key Words: extracellular matrix • mammary • prepubertal
Abbreviation key: ECM = extracellular matrix, HRP = horseradish peroxidase, MFP = mammary fat pad, MMP = matrix metalloproteinases, PAR = parenchyma, OVX = ovariectomized, TDU = terminal ductular unit(s)
|
INTRODUCTION
|
|---|
Proper heifer mammary development is essential for achieving maximal milk yield in future lactation (Sejrsen et al., 2000). Consequently, detailed understanding of the mechanisms controlling proliferation and morphogenesis of mammary epithelial tissue is important. At birth, the heifer mammary gland consists of a rudimentary ductular structure, which is supported by the mammary fat pad. During the peripubertal period of allometric mammary growth (approximately 3 to 9 mo of age), the ductular system rapidly expands into the mammary fat pad via ductal branching and elongation. As the mammary epithelium advances into the fat pad, it is surrounded by stromal tissue, which consists of fibroblasts, adipocytes, capillaries, and extracellular matrix (ECM) components. Consequently, stromal tissue must be remodeled or degraded to allow penetration of mammary epithelial structures. One aspect of heifer mammary gland development, which has not been studied, is the role of the ECM in directing ductal branching and morphogenesis. However, previous animal and cell culture models suggest that ECM proteins play a distinct role in mammary development and epithelial cell proliferation. In particular, fibronectin, laminin, and collagen may be important for development of the heifer mammary gland. Laminin and collagen IV are major components of the basement membrane, an essential structural requirement for proper mammary function (Lelievre et al., 1996; Klinowska et al., 1999), whereas fibronectin is an adhesive protein and aids attachment of the extracellular matrix to the basement membrane. The ECM proteins may also modulate mammary epithelial responsiveness to hormonal stimulation (for review, see Woodward et al., 1998; Hansen and Bissell, 2000). In rodents, mRNA expression of collagen I, collagen IV, and laminin were highest during periods of rapid mammary growth (Keely et al., 1995b). Furthermore, the spatial distribution of each protein indicated distinct roles in mammary development: whereas collagen I was abundant around larger ducts, laminin was abundant at terminal end buds and alveoli. Extracellular matrix proteins may also mediate the effects of systemic hormones and growth factors on mammary development. For instance, epithelial expression of fibronectin in mouse mammary tissue was dramatically reduced by ovariectomy, and restored by treatment with estrogen (Woodward et al., 2001). In a separate study, laminin inhibited estrogen-induced proliferation of MCF-7 cells, but not IGF-I- or EGF-induced proliferation (Woodward et al., 2000b).
The objectives of this study were to determine the localization and abundance for collagen, laminin, and fibronectin proteins in ovariectomized and intact prepubertal heifers to provide a foundation for future studies of the mechanisms for ECM regulation of bovine mammary development. To enable comparison of different physiological states, we also examined the localization of each protein in mammary tissue from postpubertal heifers and lactating cows.
|
MATERIALS AND METHODS
|
|---|
Animals, Surgery, and Experimental Design
All experiments were conducted with the approval of the Virginia Tech Animal Care committee (approval number 98-036-DASC). To test the effect of ovariectomy on mammary epithelial proliferation and ECM expression, 14 newborn Holstein heifer calves were assigned to one of two treatments: intact (control, n = 6) or ovariectomized (OVX; n = 8), using a completely randomized design. Between 1 and 3 mo of age, surgery was performed on each calf to remove the ovaries from animals in the OVX group (n = 8), as previously described (Berry et al., 2003). Heifers were sacrificed at 6 mo of age. Samples were taken from the parenchymal: stromal interface and prepared for histological analysis as described below. Samples of mammary parenchyma (PAR), and mammary fat pad (MFP) were obtained for Western blotting analysis of fibronectin and laminin. Mammary tissues were also obtained from previous experiments (Berry et al., 2001) to provide different developmental stages (postpubertal heifer and lactation) for comparison with the prepubertal heifer tissue obtained from the current experiment.
General Histological Procedures
Mammary tissue samples were fixed in phosphate-buffered formalin (4%, pH 7.4) overnight before being transferred to 70% ethanol. Subsequently, tissues were dehydrated through a graded series (70, 95, and 100%) of ethanol and embedded in paraffin. Sections were cut at 5 µm and placed on positively charged microscope slides (Fisher Scientific, Pittsburgh, PA). One slide from each block was stained with hematoxylin and eosin as follows to allow analysis of general histological structures (example panel in Figure 1F
). Slides were first deparaffinized in two changes of xylene (5 min each) and subsequently rehydrated through a graded series (100, 95, and 70%) of ethanol followed by incubation in water. Sections were then stained by incubation in Ehrlich’s Hematoxylin and Eosin (Sigma, St. Louis, MO) for 10 min. Following staining, slides were washed in water for 5 to 10 min, dehydrated (2 min each in 70, 95, and 100% ethanol), incubated in two changes of xylene (5 min each), and cover slips were added using Permount mounting media (Fisher Scientific). To determine abundance and distribution of collagen localization, sections were stained for 1 h in 0.1% Sirius red/0.1% fast green (counter-stain) in saturated picric acid (Sigma).
View larger version (149K):
[in this window]
[in a new window]
|
Figure 1. Examples of collagen IV staining in heifer mammary tissue. A) Epithelial structure containing large duct (large arrow) and smaller terminal ductular units (TDU) (small arrow). Note the presence of a prominent basement membrane on the large duct and its absence on TDU. B) Higher magnification of the edge of the large duct shown in A. C) Higher magnification of the TDU shown in A. D) Parenchyma from a lactating cow. Note presence of basement membrane around individual alveoli. E) Negative control for immunostaining showed little to no background. F) Hematoxylin and Eosin staining of heifer mammary tissue. Magnification bars represent 10 µm.
|
|
Immunohistochemistry: Collagen IV, Fibronectin, and Laminin
Sections were deparaffinized in two changes of xylene (5 min each) and rehydrated in a graded series (100, 95, and 70%) of ethanol finishing in 100% water. Following rehydration, endogenous peroxidases were quenched by incubating in 3% H2O2 (15 min, room temperature). Antigen sites were retrieved by heating the slides in 400 ml of 10 mM citrate buffer, pH 6.0, in a microwave, for three periods of 5 min each, with a 5-min cooling between each period. Following the final heating period, slides were allowed to cool for 30 min. The slides were then washed 3 x 2 min in PBS and incubated in 5% nonimmune goat serum for 30 min to block nonspecific binding of antibody. Sections were incubated with 100 µl of the primary antibody overnight at 4°C. Primary antibodies were mouse monoclonal anti-fibronectin, 1:500 (clone FBN11, NeoMarkers, Fremont, CA); rabbit polyclonal anti-laminin, 1:100 (NeoMarkers, Fremont, CA); and mouse monoclonal anti-collagen IV, 1:10 (clone CIV 22; Oncogene, Boston, MA). Subsequently, slides were washed in PBS (3 x 2 min), and detection of the primary antibody was performed using the Broad Spectrum Histostain Kit (Zymed Laboratories Inc., San Francisco, CA). Sections were incubated with biotinylated secondary antibody (an equimolar mixture of goat anti-mouse, rat, rabbit, and guinea pig IgG) for 30 min, washed (3 x 2 min in PBS), and incubated with streptavidin-peroxidase (HRP) conjugate for 10 min. The sections were again washed (3 x 2 min in PBS) before the antibody-HRP complex was visualized by incubation with diaminobenzidine (DAB, Zymed Laboratories Inc.) for 5 min. Slides were briefly counterstained in hematoxylin and dehydrated, and cover slips added. Omitting the primary antibody and incubating slides with nonimmune serum served as negative controls. Little to no background staining was observed in all negative control sections (Figure 1E
). Positive control slides of mouse kidney sections supplied by the manufacturer were also used. In addition, we purchased purified bovine collagen IV and I (nos. 1200-02S, and 1260-02S, respectively, SouthernBiotech, Birmingham, AL) for use in competition assays. Addition of excess collagen IV completely blocked staining with antibody against type IV collagen. Furthermore, addition of excess collagen I, mouse laminin (no. 354232, BD Biosciences, Bedford, MA) or bovine fibronectin (Biomedical Technologies, Inc., Stoughton, MA) had no effect on staining intensity. Similarly, addition of excess laminin blocked staining with the anti-laminin antibody but addition of collagen (I or IV), or fibronectin had no effect. Responses were similar with the anti-fibronectin antibody, i.e., excess fibronectin decreased staining but the other proteins did not. We were able to secure bovine collagen I and IV as well as bovine fibronectin for these control assays, but mouse laminin was used for competition assays, i.e., we could not find a source of purified bovine laminin.
To analyze distribution of each protein, between 5 and 10 photomicrographs at various magnifications (objective lenses 20x, 40x, and 100x) were made of each section. The photomicrographs were then compared for differences in localization patterns and staining intensity.
Western Blotting: Fibronectin and Laminin
To determine the relative amounts of fibronectin and laminin in mammary tissue from intact and ovariectomized calves, mammary extracts were prepared from each tissue as previously described (Berry et al., 2001). Fifty micrograms of protein was applied to each lane, electrophoresed through a 7.5% SDS-PAGE gel, and transferred to nitrocellulose. Membranes were blocked in 5% BSA and subsequently incubated in either mouse monoclonal anti-fibronectin, 1:400 (clone FBN11, NeoMarkers, Fremont, CA) or rabbit polyclonal anti-laminin, 1:400 (NeoMarkers, Fremont, CA) for 2 h at room temperature. Membranes were then washed (3 x 15 min), in TBS (100 mM NaCl; 5 mM Tris-HCl containing 0.02% Igepal CA-630NP40, Sigma). Sections were then incubated with goat anti-mouse IgG (Sigma) or donkey anti-rabbit IgG (Amersham, Piscataway, NJ) at 1:1000 for 1.5 h at room temperature. Detection was by ECL, using the Supersignal Chemiluminescent Substrate Kit (Pierce, Rockford, IL) for 5 min followed by exposure to Kodak Biomax ML film (Sigma) for 1 to 5 min. The laminin antibody resulted in a doublet of approximately 200 kDa and the fibronectin a single band at approximately 200 kDa. Negative controls were performed by omitting the primary antibody and resulted in no signal for either of the secondary antibodies. Western blots were quantified using scanning densitometry. Subsequent electrophoresis of purified proteins confirmed molecular weights of labeled proteins.
Statistics
Western blot data were analyzed using the GLM procedure in the PC-SAS statistical package version 8.0 (SAS Inc., Cary, NC,). The model tested for main effects of treatment (ovariectomized or control), tissue (PAR or MFP), and the treatment x tissue interaction. When the treatment x tissue interaction was significant, t-tests were used to make comparisons between individual means. Differences of P < 0.05 were considered significant. Data are presented as least squares means ± standard error of the mean.
|
RESULTS
|
|---|
Tissue Structure and Collagen Distribution
Representative images of sections stained with Sirius red/fast green are shown in Figure 2A to D. Within mammary PAR, collagen staining was most intense in the interlobular stroma. Within intralobular stroma (interspersed throughout groups of terminal ductular units [TDU]), collagen staining was diffuse and less intense (Figure 2A). All epithelial structures were surrounded with a distinct, continuous structure characteristic of a basement membrane (Figure 2B). Larger, more mature ductular structures were surrounded with densely packed collagen fibers, whereas the smaller TDU were surrounded with loosely arranged, pale staining collagen (Figure 2B and C). Ovariectomized and control animals could be distinguished on the basis of Sirius red staining, because OVX animals had abundant areas of interlobular stroma with intensely stained, densely packed collagen and few TDU surrounded with pale staining intralobular stroma (Figure 2A). On the other hand, control animals had abundant TDU and intralobular stroma and fewer areas of interlobular stroma (Figure 2B). Parenchyma from postpubertal heifers consisted of TDU that ranged from immature ductular structures to the rounded structures similar to immature alveolar buds. Secretion from epithelial cells was evident in lumenal spaces of some sections. Sirius red staining of sections from postpubertal heifers revealed a similar pattern of collagen distribution as for control prepubertal heifers (Figure 2D). Terminal ductular units were surrounded with a continuous collagen structure, and there was little collagen staining within intralobular stroma, whereas interlobular stroma exhibited intensely stained, tightly packed collagen fibers.
View larger version (148K):
[in this window]
[in a new window]
|
Figure 2. Localization of collagen within the heifer mammary gland. A) Parenchyma from ovariectomized heifer. Note large quantities of intensely stained interlobular stroma (asterisks), compared with paler-staining intralobular stroma (arrow). B) Higher magnification of terminal ductular units and surrounding intralobular stroma from a control heifer. Note distinct basement membrane and with little intralobular staining. C) Large duct with abundant, densely arranged, collagen fibrils adjacent to the epithelium. D) Parenchyma from postpubertal heifer. Magnification bars represent 10 µm.
|
|
Localization of Fibronectin
Western analysis of fibronectin revealed a band at approximately 200 kDa (Figure 3A). Fibronectin protein was significantly more abundant in PAR than MFP (478 vs. 92 relative densitometric units/mg tissue, respectively; P < 0.001; Figure 3B). A significant treatment x tissue interaction revealed that fibronectin was also more abundant in parenchyma from OVX than control heifers (572 vs. 384 relative densitometric units/mg tissue; P < 0.05, Figure 3B). Immunohistochemistry showed that in both OVX and control heifers, fibronectin was loosely arranged around parenchymal TDU (Figure 4A) and more densely arranged immediately next to larger ducts (Figure 4C) and within interlobular stroma. In prepubertal heifers, there was substantial nuclear staining of both epithelial and stromal cells. Interestingly, fibronectin was arranged differently in postpubertal heifers (Figure 4B). In the older heifers, fibronectin formed an apparent continuous structure around individual TDU. Over all, fibronectin fibers within the intralobular stroma of postpubertal heifers took on a more organized appearance. Nuclear staining was decreased in epithelial and stromal cells in comparison to tissues from prepubertal heifers. Fibronectin arrangement next to large ducts and in interlobular stroma was similar in older heifers and prepubertal heifers. Parenchyma from a lactating animal showed that fibronectin formed a distinct, continuous layer around the basal side of alveoli (Figure 4D).
View larger version (33K):
[in this window]
[in a new window]
|
Figure 3. Western blot analysis of fibronectin in prepubertal heifer mammary tissue. A) Example western of parenchyma (PAR) and mammary fat pad (MFP) tissue from a control and an ovariectomized heifer. B) Quantification of western blots, open bars (CON) and shaded bars (OVX). There was a significant main effect of tissue (P < 0.001; described within text). *P < 0.05 compared to CON PAR.
|
|
View larger version (159K):
[in this window]
[in a new window]
|
Figure 4. Fibronectin staining in heifer mammary parenchyma. A) Example of fibronectin staining in prepubertal parenchyma. B) Postpubertal parenchyma, note appearance of distinct basement membrane around epithelial structures. C) Example of a large duct with a dense arrangement of fibronectin adjacent to it. D) Parenchyma from a lactating cow. Note distinct basement membrane and minimal intralobular stromal staining. Magnification bars represent 10 µm.
|
|
Localization of Laminin
Western analysis of laminin revealed a doublet of approximately 200 kDa (Figure 5A). A significant treatment x tissue interaction revealed that the lower molecular weight band was significantly more abundant (per mg of tissue) in control vs. OVX parenchyma (30 vs. 17 densitometric units/mg tissue; P < 0.05; Figure 5C) but did not differ between parenchyma and mammary fat pad. There were no significant main effects of treatment or tissue for the higher molecular weight band (Figure 5B). The observation that laminin was less abundant in parenchyma from OVX animals was supported by immunohistochemistry (Figure 6A to D). In control animals, laminin staining in parenchymal stroma was abundant and consisted of many laminin "fibers" that were most plentiful adjacent to TDU and were arranged to follow the shape of the epithelial structures (Figure 6A). Laminin was also present in the basement membrane of capillaries. In OVX animals, laminin staining was less prominent and most evident around capillaries (Figure 6B). Within the epithelial structures, laminin staining was observed within the cytoplasm (but not nuclei) of epithelial cells, as well as along the basement membrane of epithelial structures (Figure 6C). Laminin staining was similar in postpubertal heifers and control prepubertal heifers. Parenchyma from lactating animals showed that laminin formed a continuous basement membrane around each alveolar structure (Figure 6D).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 5. Western analysis of laminin in prepubertal heifer mammary tissue. A) Example western of parenchyma (PAR) and mammary fat pad (MFP) from a control and ovariectomized heifer. B) Quantification of western blots, open bars (CON) and shaded bars (OVX) for higher molecular weight laminin band. There were no significant main effects of tissue. C) Quantification of western blots, open bars (CON) and shaded bars (OVX) for lower molecular weight band. *P < 0.05 compared to control PAR.
|
|
View larger version (160K):
[in this window]
[in a new window]
|
Figure 6. Examples of laminin staining in heifer mammary tissue. A) Parenchyma from a control heifer. Note abundant stromal staining near terminal ductular unit (TDU) structures. B) Parenchyma from an OVX heifer with less apparent staining in surrounding stroma. C) High magnification of an epithelial structure to show presence of laminin in the basement membrane. D) Parenchyma from a lactating cow. Note distinct basement membrane around individual alveoli. Magnification bars represent 10 µm.
|
|
Expression of Collagen IV
Type IV collagen was present in the basement membrane of capillaries and along the basal surface of ductal epithelial structures. Collagen IV was more abundant in the basement membrane around larger ducts (Figure 1A to C) than TDU structures. In the structure shown in Figure 1A to C, collagen IV is prominent along the edge of the larger duct (large arrow in Figure 1A, enlarged in Figure 1B) but disappears along the edge of the smaller TDU (small arrow in Figure 1A, enlarged in Figure 1C). Collagen IV localization did not differ between OVX and control heifers. In lactating mammary tissue, collagen IV formed a continuous membrane around the base of each alveolus and of capillaries in between alveoli (Figure 1D). Figure 1E depicts a negative control (nonimmune serum) showing little background staining. Figure 1F is a section of prepubertal parenchymal tissue stained in hematoxylin and eosin to show the histological structure.
|
DISCUSSION
|
|---|
Despite detailed knowledge of extracellular matrix biochemistry, surprisingly little information is available regarding the in vivo functions of ECM proteins in regulating mammary epithelial cell proliferation, survival, morphogenesis, or differentiation. Consequently, the aim of our experiment was to describe the localization of collagen, laminin, and fibronectin with respect to developing epithelial structures in actively growing (control) and growth impaired (OVX) prepubertal heifers mammary tissue. Ovariectomy provided an excellent model of impaired mammary development for two reasons. First, the treatment was applied for an extended period (approximately 4 mo), potentially allowing significant remodeling of the mammary fat pad and intralobular stroma. Secondly, the treatment was very effective, allowing for a dramatic contrast between the two groups. In OVX controls, total udder weight was reduced by more than 50% and epithelial proliferation was reduced by tenfold (Berry et al., 2003). The results described here provide initial evidence for the involvement of laminin, fibronectin, and collagen IV in regulating heifer mammary development.
Our observations of collagen localization are in general agreement with the findings of Keely et al. (1995a, 1995b). Using Sirius red, a stain with affinity for all collagen proteins, we found that collagen was densely arranged around large ducts and in interlobular stroma within the bovine mammary gland. Conversely, groups of TDU (which represent areas of proliferation and ductular branching; Capuco et al., 2002) were surrounded with intralobular stroma, which contained much less collagen than what was observed adjacent to major ducts. Collagen I is the most ubiquitous fibrillar collagen of connective tissue (Linsenmayer, 1981), so it is likely that much of the staining we observed in the connective tissue stroma is due primarily to collagen I. In the previously reported experiment (Keely et al., 1995b), collagen I was most abundantly expressed around major ductular structures in the virgin mouse mammary gland, but was sparse around actively growing end buds, implying a role for collagen I in regulating morphogenesis of ductular structures. Our observations suggest a comparable role for collagen proteins in prepubertal heifer mammary development. Similarly, local implants of TGFß inhibited epithelial development in rodent mammary glands and induced deposition of collagen around terminal end buds (Silberstein et al., 1990). Conversely, in control animals, collagen was abundant around ductal structures, but terminal end buds were relatively free of collagen fibers. That collagen I may act to direct branching morphogenesis or ductal development rather than epithelial cell proliferation within the mammary gland is supported by several in vitro observations. Human-derived mammary epithelial cells (T47D) formed duct-like structures on collagen I but not on matrigel (Keely et al., 1995a). Growth of mammary epithelial cells on collagen I does not promote cell survival, but mammary epithelial cells cultured on laminin or collagen IV are protected from apoptosis (Farrelly et al., 1999). Ovine granulosa cells also grew rapidly on laminin or fibronectin but not on collagen I (Huet et al., 2001). In ruminant mammary development, the epithelial structure advances from the margin of the gland cistern into the fat pad as a compact but branched mass of tissue. As the tissue advances, the mammary fat pad must be remodeled to allow epithelial penetration. This is in contrast to rodent mammary development in which the ductal system penetrates the entire mammary fat pad before significant branching occurs. In comparison to interlobular stroma, intralobular stroma was relatively devoid of collagen. It is tempting to speculate that, as mammary ducts branch into smaller TDU, collagen is reorganized to allow penetration of epithelial structures into the mammary fat pad. Several distinct matrix metalloproteinases (MMP) are secreted by mammary fibroblasts. For example, in rat primary mammary epithelial cells, MMP-2 and MMP-9 had maximum activity during periods of rapid proliferation and ductal branching (Lee et al., 2001). Whether MMP specific for collagen are developmentally regulated in the heifer mammary gland is not known. In OVX heifers, mammary epithelium was characterized by major ductal structures with little to no branching and very few TDU. Consistent with a role in duct formation but not epithelial proliferation, parenchyma from OVX heifers had very little of the pale-staining intralobular stroma. In contrast, mammary tissue from control heifers had abundant TDU and, consequently, abundant areas of the pale-staining intralobular stroma. In other words, actively growing mammary epithelium was associated with weak collagen staining, whereas inactively growing mammary tissue was associated with intense collagen staining.
Our observations that laminin was down regulated in parenchyma of OVX heifers suggests an important role for this protein in regulating ovarian stimulated mammary development. A role for laminin in active mammary growth was also implied in a previous rodent experiment (Keely et al., 1995b), who reported that laminin was localized primarily around actively growing end buds. Previous in vitro studies also suggest that laminin is a survival factor for mammary epithelial cells (Farrelly et al., 1999). However, the exact role of laminin in mammary epithelial development is undefined. The MCF-7 cells cultured on laminin prevented estrogen-induced cell proliferation via reduced activity of the estrogen response element (Woodward et al., 2000a). However, laminin did not impair the proliferative response of MCF-7 cells to other mitogens such as IGF-I or EGF. In another study, laminin and estrogen synergized to increase tissue plasminogen activator activity of MCF-7 cells, potentially enhancing ECM remodeling (Sonohara et al., 1998). Clearly, the mechanisms of laminin action within the mammary gland are complex and involve regulation at multiple levels.
Consistent with previous immunohistochemistry studies (Warburton et al., 1982; Ferguson et al., 1992), fibronectin was abundant throughout mammary stroma. Around TDU, fibronectin formed fibrous structures that followed the shape of the epithelial structure, and fibronectin was densely arranged adjacent to major ducts. It was interesting to note that fibronectin appeared to become more organized with each developmental state (prepubertal, postpubertal, and lactating animals). In lactating mammary tissue, fibronectin was present primarily as a distinct basement membrane around individual alveoli. This is consistent with previous observations in which fibronectin was observed to be in the basement membrane of lactating, but not resting, rat mammary tissue (Warburton et al., 1982). In contrast to previous observations (Woodward et al., 2001) and to our hypothesis that fibronectin would be reduced in OVX heifers, we found that fibronectin was more abundant (per mg of tissue) in parenchyma of OVX compared with control heifers. The changes in fibronectin localization through developmental stages imply a role for this protein in mammary development.
The localization of collagen IV also suggested a role for this protein in heifer mammary development. Interestingly, collagen IV was more prominent in the basement membrane of major ducts compared with TDU, perhaps implying a role for collagen IV in duct formation or branching morphogenesis.
Taken together, the results presented here provide initial evidence that laminin, fibronectin, and collagen are involved in regulation of prepubertal heifer mammary development. There are a number of possible ways in which ECM proteins could influence development of mammary epithelium. First, ECM proteins may provide physical attachments for epithelial and stromal cells, as well as physical barriers that prevent invasion of the mammary fat pad. This is supported by evidence that MMP are differently regulated throughout mammary development (Witty et al., 1995). The MMP cleave ECM proteins allowing remodeling or degradation of the ECM to allow epithelial penetration of surrounding stroma. In fact, previous studies have shown that the presence of ECM alone is not sufficient for epithelial development and branching morphogenesis. Local remodeling of ECM by MMP action is also required (Witty et al., 1995; Lee et al., 2001; Simian et al., 2001). Possibly, cleavage of ECM proteins may also enhance mammary development through release of stored growth factors or through the direct actions of proteolytic peptides produced from MMP enzymatic reactions (Schedin et al., 2000). Second, ECM proteins may mediate the effects of systemic hormones such as GH or E. This hypothesis is supported by observations that fibronectin was reduced in mammary tissue of ovariectomized mice (Woodward et al., 2001), suggesting that this protein may act to mediate the effects of estrogen within the mammary gland. Moreover, the epithelium was required for this effect because no change was observed within the cleared fat pad. Expression of ECM within human breast tissue is also regulated throughout the menstrual cycle, with periods of low proliferation corresponding to increased expression of ECM components (Ferguson et al., 1992).
A possible role for ECM proteins in mediating the effects of mammogenic hormones in heifer mammary development is suggested by our observations of fibronectin and laminin abundance in OVX heifers. Third, ECM proteins may influence cell proliferation by interacting with growth factor signaling pathways. Extracellular matrix proteins promoted synergy between IGF-I and EGF in MCF-7 cells via increased binding to IGF-I and EGF receptors as well as decreased expression of IGF binding proteins (Woodward et al., 2000b). The cell survival effects of laminin appear to be mediated through the 
6ß1 integrin and through enhanced phosphorylation of IRS-I as well as (in the presence of insulin) PI-3-kinase association with IRS-I (Farrelly et al., 1999). Another possible role of ECM proteins within the mammary gland could be truncation of IGF-I resulting in release of IGF-I from inhibitory binding proteins.
|
CONCLUSIONS
|
|---|
Clearly, modifications in ECM proteins are an important aspect of mammary morphogenesis. The results presented here describe the localization of collagen, fibronectin, and laminin proteins within the prepubertal heifer mammary gland. That each of these proteins was localized to distinct portions of the mammary epithelium and stroma suggests specific roles in regulating development of mammary epithelium. Furthermore, laminin was significantly less abundant and fibronectin significantly more abundant in mammary parenchyma from OVX heifers, implying regulation by, or interactions with, ovarian secretion(s).
|
IMPLICATIONS
|
|---|
Because mammary ductular development provides the tissue foundation for appearance of alveoli associated with rapid mammary growth during gestation, factors that modulate proliferation of ductular epithelial cells are fundamentally important to overall udder development. Our data are among the first to provide direct evidence that alterations in extracellular matrix proteins in the prepubertal bovine mammary gland are correlated with ductal morphogenesis and parenchymal growth. Development of techniques to modify extracellular matrix protein expression may lead to enhanced parenchymal growth and therefore increased milk production.
|
ACKNOWLEDGEMENTS
|
|---|
Technical assistance from Pat Boyle was gratefully received. The authors are also grateful for financial assistance from USDA grant (98-03664) to RMA and from the Frank Sydenham and C. Alma Baker scholarships for graduate study in agriculture (to SDKB).
Received for publication December 6, 2002.
Accepted for publication April 23, 2003.
|
REFERENCES
|
|---|
Berry, S. D., P. M. Jobst, R. D. Howard, S. E. Ellis, A. V. Capuco, and R. M. Akers. 2003. Estrogen receptor and epithelial proliferation in prepubertal heifers: Effects of ovariectomy and growth hormone. J. Dairy Sci. 86:2098–2105.[Abstract/Free Full Text]
Berry, S. D., T. B. McFadden, R. E. Pearson, and R. M. Akers. 2001. A local increase in the mammary IGF-1: IGFBP-3 ratio mediates the mammogenic effects of estrogen and growth hormone. Domest. Anim. Endocrinol. 21:39–53.[Medline]
Capuco, A. V., S. Ellis, D. L. Wood, R. M. Akers, and W. Garrett. 2002. Postnatal mammary ductal growth: Three-dimensional imaging of cell proliferation, effects of estrogen treatment and expression of steroid receptors in prepubertal calves. Tissue Cell 34:9–20.
Farrelly, N., Y. J. Lee, J. Oliver, C. Dive, and C. H. Streuli. 1999. Extracellular matrix regulates apoptosis in mammary epithelium through a control on insulin signaling. J. Cell Biol. 144:1337–1348.[Abstract/Free Full Text]
Ferguson J. E., A. M. Schor, A. Howell, and M. W. Ferguson. 1992. Changes in the extracellular matrix of the normal human breast during the menstrual cycle. Cell Tissue Res. 268:167–177.[Medline]
Hansen, R. K, and M. J. Bissell. 2000. Tissue architecture and breast cancer: The role of extracellular matrix and steroid hormones. Endocr.-Related Cancer 7:95–113.
Huet C., C. Pisselet, B. Mandon-Pepin, P. Monget, and D. Monniaux. 2001. Extracellular matrix regulates ovine granulosa cell survival, proliferation and steroidogenesis: Relationships between cell shape and function. J. Endocrinol. 169:347–360.[Abstract]
Keely, P. J., A. M. Fong, M. M. Zutter, and S. A. Santoro. 1995a. Alteration of collagen-dependent adhesion, motility, and morphogenesis by the expression of antisense alpha(2) integrin mRNA in mammary cells. J. Cell Sci. 108:595–607.[Abstract]
Keely, P. J., J. E. Wu, and S. A. Santoro. 1995b. The spatial and temporal expression of the alpha 2 beta 1 integrin and its ligands, collagen I, collagen IV, and laminin, suggest important roles in mouse mammary morphogenesis. Differentiation 59:1–13.[Medline]
Klinowska, T. C., J. V. Soriano, G. M. Edwards, J. M. Oliver, A. J. Valentijn, R. Montesano, and C. H. Streuli. 1999. Laminin and beta1 integrins are crucial for normal mammary development in the mouse. Dev. Biol. 215:13–32.[Medline]
Lee, P. P., J. J. Hwang, L. Mead, and M. M. Ip. 2001. Functional role of matrix metalloproteinases (MMPs) in mammary epithelial cell development. J. Cell. Physiol. 188:75–88.[Medline]
Lelievre, S., V. M. Weaver, and M. J. Bissell. 1996. Extracellular matrix signaling from the cellular membrane skeleton to the nuclear skeleton: A model of gene regulation. Recent Prog. Horm. Res. 51:417–32.
Linsenmayer, T. F. 1981 Collagen. Pages 5–37 in Cell Biology of Extracellular Matrix. E. D. Hay, ed. Plenum Press, New York.
Schedin, P., R. Strange, T. Mitrenga, P. Wolfe, and M. Kaeck. 2000. Fibronectin fragments induce MMP activity in mouse mammary epithelial cells: Evidence for a role in mammary tissue remodeling. J. Cell Sci. 113:795–806.[Abstract]
Sejrsen, K., S. Purup, M. Vestergaard, and J. Foldager. 2000. High body weight gain and reduced bovine mammary growth: Physiological basis and implications for milk yield potential. Domest. Anim. Endocrinol. 19:93–104.[Medline]
Silberstein, G. B., P. Stricklan, S. Coleman, and C. W. Daniel. 1990. Epithelium-dependent extracellular matrix synthesis in transforming growth factor-B1-growth-inhibited mouse mammary gland. J. Cell Biol. 110:2209–2219.[Abstract/Free Full Text]
Simian, M., Y. Hirai, M. Navre, Z. Werb, A. Lochter, and M. J. Bissell. 2001. The interplay of matrix metalloproteinases, morphogens and growth factors is necessary for branching of mammary epithelial cells. Development 128:3117–3131.
Sonohara, S., R. Mira-Y-Lopez, and M. M. Brentani. 1998. Laminin and estradiol regulation of the plasminogen-activator system in MCF-7 breast-carcinoma cells. Int. J. Cancer 76:85.
Warburton, M. J., D. Mitchell, E. J. Ormerod, and P. Rudland. 1982. Distribution of myoepithelial cells and basement membrane proteins in the resting, pregnant, lactating and involuting rat mammary gland. J. Histochem. Cytochem. 30:667–676.[Abstract]
Witty, J. P., J. H. Wright, and L. M. Matrisian. 1995. Matrix metalloproteinases are expressed during ductal and alveolar mammary morphogenesis, and misregulation of stromelysin-1 in transgenic mice induces unscheduled alveolar development. Mol. Biol. Cell 6:1287–1303.[Abstract]
Woodward, T. L., J. W. Xie, and S. Z. Haslam. 1998. The role of mammary stroma in modulating the proliferative response to ovarian hormones in the normal mammary gland. J. Mamm. Gland Biol. Neopl. 3:117–131.[Medline]
Woodward, T. L., H. Lu, and S. Z. Haslam. 2000a. Laminin inhibits estrogen action in human breast cancer cells. Endocrinology 141:2814–2821.[Abstract/Free Full Text]
Woodward, T. L., A. S. Mienaltowski, R. R. Modi, J. M. Bennett, and S. Z. Haslam. 2001. Fibronectin and the 5ß1 integrin are under developmental and ovarian steroid regulation in the normal mouse mammary gland. Endocrinology 142:3214–3222.[Abstract/Free Full Text]
Woodward, T. L., J. Xie, J. L. Fendrick, and S. Z. Haslam. 2000b. Proliferation of mouse mammary epithelial cells in vitro: Interactions among epidermal growth factor, insulin-like growth factor I, ovarian hormones, and extracellular matrix protein. Endocrinology 141:3578–3586.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
F. Vangroenweghe, W. Van Den Broeck, A. De Ketelaere, H. van Bree, L. Duchateau, and C. Burvenich
Endoscopic Examination and Tissue Sampling of the Bovine Teat and Udder Cistern
J Dairy Sci,
May 1, 2006;
89(5):
1516 - 1524.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
