Loss of Alx4, a stromally-restricted homeodomain protein, impairs mammary epithelial morphogenesis

Loss of Alx4, a stromally-restricted homeodomain protein, impairs mammary epithelial morphogenesis

Developmental Biology 297 (2006) 284 – 294 www.elsevier.com/locate/ydbio Loss of Alx4, a stromally-restricted homeodomain protein, impairs mammary ep...

984KB Sizes 0 Downloads 6 Views

Developmental Biology 297 (2006) 284 – 294 www.elsevier.com/locate/ydbio

Loss of Alx4, a stromally-restricted homeodomain protein, impairs mammary epithelial morphogenesis Purna A. Joshi, Hong Chang, Paul A. Hamel ⁎ Department of Laboratory Medicine and Pathobiology, Room 6318, Medical Sciences Building, 1 King’s College Circle, University of Toronto, Toronto, Ontario, Canada M5S 1A8 Received for publication 8 November 2005; revised 22 May 2006; accepted 24 May 2006 Available online 2 June 2006

Abstract Postnatal development of the mammary gland is determined by reciprocal interactions between the ductal epithelia and adjacent stroma. Alx4 is a mesenchymally restricted homeodomain transcription factor expressed in a number of developing tissues, including skin appendages such as hair follicles, whiskers and teeth. We show here that Alx4 is expressed in a subset of ERα-expressing mammary stromal cells adjacent to terminal end buds and alveoli during puberty and pregnancy, respectively. Alx4 expression is induced in mammary stromal cells at the onset of puberty and can be induced in prepubescent mice by administration of 17β-estradiol. In order to determine the role of Alx4 during mammary gland development, we characterized mammary gland development of mice homozygous for the null allele of Alx4, lstD. Mammary glands from animals lacking Alx4 activity exhibit profound alterations in ductal morphogenesis. Overall development is delayed, ducts being grossly distorted in size and structure. Terminal end buds are also disoriented, displaying aberrant architecture during bifurcation. Despite the developmental delay, the ductal network typically reaches the limits of the fat pad. However, during puberty and in the adult virgin mice, the frequency and density of branch points is significantly reduced. We show further that the defective ductal morphogenesis is due to defects in stromal cells. Specifically, when injected into the cleared fat pad of wild-type recipients, mixed populations of wild-type epithelial cells and Alx4-deficient stromal cells give rise to retarded ductal morphogenesis. Wild-type stromal cells mixed with Alx4-deficient epithelial cells result in normal progression of ductal development. Defective branching morphogenesis in Alx4-deficient females is not due to a loss in expression of HGF, since the level of HGF message in mammary stromal cells is similar in mutant and wild-type littermates. MMP3 is similarly expressed while a 40% increase in MMP2 and a 50% decrease in MMP9 message levels in Alx4-deficient mice relative to their wild-type littermates is observed. Thus, the activity of the stromally restricted homeodomain factor, Alx4, is required for normal branching morphogenesis of the ductal epithelia during pubescent mammary gland development. © 2006 Elsevier Inc. All rights reserved. Keywords: Mammary gland; Strong’s luxoid; Stroma; Estrogen receptor; Alx4; Terminal end bud

Introduction The mammary gland is a specialized skin appendage which, like tooth buds and hair follicles, develops under the influence of inductive, reciprocal interactions between the embryonic ectoderm and underlying mesenchyme (van Genderen et al., 1994; Veltmaat et al., 2003). While the principle architectural components of the mammary gland are established prior to birth, significant postnatal development occurs for this tissue. The rudimentary ductal tree of the mammary gland established during embryogenesis is hormonally stimulated at puberty to ⁎ Corresponding author. Fax: +1 416 978 5959. E-mail address: [email protected] (P.A. Hamel). 0012-1606/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2006.05.032

expand into the fat pad (Silberstein, 2001; Silberstein et al., 1994). Branching morphogenesis terminates at the end of puberty by which time the ductal structures have reached the limits of the fat pad. While undergoing limited estrousdependent cyclical changes in the adult virgin, a second phase of postnatal development arises with the onset of pregnancy, initially through side-branching morphogenesis followed by the appearance of alveoli (Nandi, 1958; Richert et al., 2000; Silberstein, 2001). The alveoli, which produce milk proteins during lactation, then regress during involution through programmed cell death, returning the mammary gland to a similar, though phenotypically distinct state relative to the mammary gland of the adult virgin (Richert et al., 2000; Master et al., 2005).

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

Development of the three-dimensional ductal architecture of the mammary gland depends critically on the formation of a unique structure at the ends of ducts known as the terminal end bud (TEB) (Silberstein, 2001; Silberstein et al., 1994). The TEB, which harbors mammary gland stem cells capable of reconstituting the entire epithelial aspects of the mammary gland, is characterized by multilayered body cells surrounded by a single layer of caps cells. Body cells appear to give rise to the luminal epithelial cells while cap cells are progenitors of the myoepithelial cell layer. During puberty, branching morphogenesis producing the ductal network of the mammary gland occurs principally through bifurcation of the TEB. Terminal end buds persist throughout puberty until the ducts reach the limits of the fat pad. In the adult virgin, the ends of the ducts retain a bilayered structure once the TEBs have regressed (Richert et al., 2000). Mammary epithelial morphogenesis depends on both instructive and permissive signals emanating from the supporting mammary stroma (Wiseman and Werb, 2002). The stroma includes fibroblasts, the extracellular matrix (ECM), adipocytes, preadipocytes,blood vessels, as well asinflammatory cells such as macrophages and eosinophils (Gouon-Evans et al., 2000; Pollard and Hennighausen, 1994). The importance of stromal signals for morphogenetic behavior of the epithelial compartment has been demonstrated in a number of systems. So, for example, soluble growth factors expressed from the fibroblasts surrounding the epithelial ducts, including hepatocyte growth factor (HGF) and insulin-like growth factor-1 (IGF-1), are responsible for both morphogenic and mitogenic processes as well as for cell survival in the adjacent epithelial compartments (Kleinberg et al., 2000; Niranjan et al., 1995; Soriano et al., 1998; Walden et al., 1998; Yang et al., 1995). Indeed, estrogen-dependent mammary epithelial morphogenesis appears to be mediated, at least in part, through estrogen receptor alpha (ERα)-dependent transcriptional activation of HGF (Zhang et al., 2002). The activation of stromal Epidermal Growth Factor receptor (EGFR) in a paracrine mechanism by ADAM17 mediated release of epithelial amphiregulin has also recently been shown to be required for mammary epithelial development (Sternlicht et al., 2005). Additional cell types, such as macrophages and eosinophils, are recruited to the TEBs and are also required for normal branching morphogenesis of the postnatal mammary gland. In animals deficient for these myeloid cells, delayed and aberrant ductal maturation ensues, resulting in defective TEB formation, impaired ductal branching and disoriented penetration of the fat pad (Gouon-Evans et al., 2000, 2002). Aristaless-like 4 (Alx4) is a paired-like homeodomain factor expressed in the mesenchymal compartment of a number of specific developing tissues, including hair follicles, teeth, craniofacial skeletal and skull precursors (Hudson et al., 1998; Qu et al., 1997). Expression of Alx4 in the anterior aspect of the developing limb bud confines the Sonic HedgeHog-expressing cells of the Zone of Polarizing Activity (ZPA) to the posterior aspect of this tissue (Chan et al., 1995; Qu et al., 1998). Mice lacking Alx4 activity exhibit a duplication of the ZPA resulting in preaxial polydactyly. Alx4-deficient mice also exhibit dorsal alopecia, defective craniofacial structures, hemimelia (absence

285

of the tibia) and body wall closure defects. In humans, both vascular and cortical abnormalities have been attributed to a DNA-binding defective mutant of ALX4 in parietal foramina type 2 (Mavrogiannis et al., 2001; Valente et al., 2004; Wu et al., 2000; Wuyts et al., 2000). We demonstrated previously that Alx4 is expressed in a number of tissues whose development depends on expression of the mediator of Wnt signaling, lymphoid enhancer factor 1 (Lef1) (Hudson et al., 1998). Mice deficient for Lef1 lack teeth, hair, whiskers and interestingly, embryonic mammary glands also fail to develop in these mice (van Genderen et al., 1994). Indeed, we demonstrated previously that Alx4 and Lef1 form co-complexes, regulate the promoter activity of at least one candidate gene (Boras and Hamel, 2002) and, when simultaneously absent, result in early embryonic lethality (9.0 dpc) due to defective vasculogenesis (BorasGranic et al., 2006). Given the role of Alx4 activity in the mesenchymal compartment of skin appendages whose development also depends on Lef1 activity (Kratochwil et al., 1996; van Genderen et al., 1994), we hypothesized that Alx4 activity would also play a role in the development of the mammary gland. We demonstrate here that Alx4 is expressed in a subset of stromal cells adjacent to TEBs during puberty and alveoli during pregnancy. We show further that in the absence of Alx4 activity specifically in mammary stromal cells, impairment during puberty of mammary epithelial ductal elongation and branching morphogenesis ensues. Materials and methods Mice and antibodies Strong’s luxoid(lstD), an inbred strain of mice harboring a point mutation (R206Q) in the homeodomain of Alx4, thereby inhibiting its DNA-binding activity (Forsthoefel, 1962, 1963; Forsthoefel et al., 1966; Qu et al., 1998) was obtained from Dr. R. Wisdom (St. Jude's Childrens Hospital) and backcrossed onto the C57Bl/6 background (Charles River) for >12 generations. The antiAlx4 monoclonal antibody, KAb07, was developed in our lab, the rabbit antiERα antibody (sc-542) was purchased from Santa Cruz and anti-BrdU antibody (347580) from BD Pharmingen.

Immunohistochemistry, immunofluorescence and whole-mount analysis Paraffin-embedded tissue sections were deparaffinized in xylene and rehydrated gradually through 100%, 95%, 70% ethanol and water. Antigen retrieval was performed by boiling sections for 10 min in 10 mM sodium citrate preheated in a microwave oven for 3 min before immersing slides. Immunohistochemistry was performed according to the HRP-AEC system from R&D (CTS003 and CTS006). Briefly, after the solution cooled, sections were blocked for endogenous peroxidase activity with 3% H2O2 for 15 min. Sections were further blocked in serum for 1 h, avidin and biotin blocking reagents and the primary antibody of interest diluted in PBS was incubated on slides at 4°C overnight. Samples were washed and incubated with biotinylated secondary antibody for 1 h. Following washes, streptavidin conjugated to HRP was incubated on slides for 30 min. Sections were incubated with AEC chromogen in the supplied buffer and staining stopped by a further washing step. For immunofluorescence, sections were permeabilized with Triton X-100 in PBS for 5 min following antigen retrieval. Sections were blocked with 5% serum in PBS with saponin (10 μl/ml) for 1 h. Primary antibody was applied for 1 h, washed in PBS and a fluorescently labeled secondary antibody added to samples for 45 min. Whole-mount analysis of mammary glands was performed as previously described (Rasmussen et al., 2000).

286

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

Hormone injection experiments 17β-Estradiol (Sigma), prepared from 1 mg/ml stock solution in ethanol, was administered subcutaneously at 0.1 mg in 0.1 ml corn oil per 10 g mouse. Prepubescent mice at 2 weeks of age were primed with an injection of 17βestradiol on day 1, followed by a second injection on day 2 and sacrificed by cervical dislocation 24 or 48 h postpriming. Controls were left untreated or treated with corn oil alone. BrdU (BD Pharmingen) was injected intraperitoneally into mice (50 mg/kg in PBS) 2 h prior to sacrifice.

Isolation of primary mouse mammary fibroblasts Primary mammary fibroblasts were isolated from pubescent mouse mammary glands as described previously (Niranjan et al., 1995). In brief, dissected mammary glands were minced and homogenized in 4 ml/g of a digestion cocktail of Ham's F12 media containing 3 mg/ml Collagenase A (Boerhinger Mannheim #103586) followed by an incubation for 1 h at 37°C in a shaker. Contents were then centrifuged for 1 min at 150 rpm, and the supernatant containing loosely associated fibroblast cells was decanted into another tube and centrifuged further at 1500 rpm for 10 min. The pellet was washed in DMEM + 10% FBS, centrifuged, resuspended in Ham's F12 plus 10% FBS:DMEM plus 10% FBS (1:1) and supplemented with transferrin, insulin, EGF, gentamycin and plated. The pellet containing epithelial cells and tightly associated fibroblasts from the first digestion cocktail was subjected to another round of digestion for 30 min at 37°C. The cocktail was centrifuged for 5 min at 1500 rpm, the pellet washed and plated after resuspension in the media described above. After the tightly associated fibroblast cells settled, the media containing epithelial organoids were removed and fibroblast cells replenished with fresh media. Detection of message for Alx4, HGF and IGF-1 was performed using Superscript One-Step RT-PCR kit (Invitrogen) according to the manufacturer's instructions. Primers for PCR were as follows: IGF-1: Forward: 5′-GCTCTTC AGTTCGTGTGTGG-3′, Reverse: 5′-TTGGGCATGTCAGTGTGG-3′; HGF: Forward: 5′-TTGGCCATGAATTTGACCTC-3′ Reverse: 5′-ACATCAGTCTCATTCACAGC-3′; Alx4: Forward: 5′-CAAAGGCAAGAAGCGGCGGAATC-3′, Reverse: 5′-GGTACATTGAGTTGTGCTGTCC-3′.

Transplantation experiments and quantitative PCR Fractions of purified mammary epithelial cells and stromal cells from wildtype and Alx4lstD/lstD mice were isolated as described above and previously (for review, see Ip and Darcy, 1996). Epithelial cells from wild-type mice and stromal cells from Alx4lstD/lstD mice as well as the reciprocal recombination were then injected into the cleared fat pad of anesthetized prepubescent (3 weeks) C57Bl/6 female recipients. In each case, the reciprocal recombinations (wildtype epithelia/Alx4lstD/lstD stroma and Alx4lstD/lstD epithelia/wild-type stroma) were injected into contralateral #4 mammary glands of the same recipient. Recipients were allowed to recover and reconstituted mammary glands harvested for whole-mount analysis 5 weeks postimplantation. For quantitative RT-PCR, RNA from purified fractions of epithelial and stromal cells from wild-type and Alx4lstD/lstD mice was isolated as described above. Eight nanograms were used for each Q-PCR reaction. Each sample was performed in triplicate, standardized against genomic DNA and corrected for differences using β-actin as a control. Primers pairs were as follows: HGF: Forward: 5′ aggttacaggggaaccagca 3′, Reverse: 5′ gcttgtgagggtactgcgaatcc 3′; MMP2: Forward: 5′-tgcgggttctctgcgtcctgtg 3′, Reverse: 5′-acttgatgatgggcgatggtgc 3′; MMP3: Forward: 5′-tgaagggtcttccggtcctg-3′, Reverse: 5′-ccttgcactgtcatgcaatgggt-3′; MMP9: Forward: 5′-tgaacaaggtggaccatgaggtg 3′, Reverse: 5′-tcaagggcactgcaggaggtcg 3′.

Results Stromal-specific expression of Alx4 during postnatal mammary gland development We determined previously that Alx4 transcript is expressed in stromal cells adjacent to TEBs in pubescent female mice

(Hudson et al., 1998). As Fig. 1 depicts, Alx4 expression is altered during distinct phases of postnatal mammary gland development. Specifically, Alx4 is expressed rarely in stromal cells surrounding the primitive ductal structure prior to puberty (Figs. 1A–B). As mice enter puberty, the appearance of TEBs coincides with an expansion of cells expressing Alx4 (Figs. 1C–D). Alx4-expressing cells are observed in the stroma directly adjacent to the TEB, an apparent concentration of these cells present along the flanks of the TEB where it narrows to form luminal and myoepithelial cell layers of the subtending duct. However, not all of the cells adjacent to the TEB express Alx4. In contrast to the TEB, a small proportion of stromal cells adjacent to the ductal structures express Alx4 (arrowhead, Fig. 1E). By the end of puberty and in the adult virgin (Figs. 1F–G), Alx4 expression is generally reduced although it can still be detected in stromal cells associated with the end of ducts and occasionally adjacent to the ducts (arrowheads, Fig. 1H). Strong expression of Alx4 protein is also apparent during pregnancy where many Alx4-expressing cells are observed associated with alveoli (Figs. 1I–J). Fig. 2 illustrates that Alx4-expressing stromal cells also express estrogen receptor-alpha (ERα). Specifically, while ERα is highly expressed in the mammary luminal epithelial cells, its expression is also observed in the adjacent stromal cells coincident with cells which express Alx4. We isolated loose stromal cells as well as stromal cells tightly associated with the epithelial component of the pubescent mammary gland. Immunofluorescence confirms the presence of Alx4 protein in the nuclei of most fibroblasts 48 h postplating (Fig. 3A). Alx4 expression is also detected by RT-PCR in cells plated for 4 h and 3 days in tissue culture (Fig. 3B). Furthermore, this fraction of stromal fibroblasts expresses HGF and IGF-1 (Fig. 3C), as has been reported previously (Kamalati et al., 1999; Kleinberg et al., 2000; Niranjan et al., 1995; Ruan and Kleinberg, 1999). Hormonal regulation of Alx4 expression We determined next that Alx4 expression could be induced following administration of 17β-estradiol in prepubescent mice. Specifically, corn oil alone or corn oil with 17βestradiol was injected subcutaneously into 2-week-old, prepubescent mice. Two hours prior to harvesting the mammary glands, animals were also injected with BrdU in order to determine the proportion of cells in S phase. As Fig. 4 illustrates, induction of Alx4 expression in stromal cells is evident in mammary glands in prepubescent mice 48 h postpriming with 17β-estradiol (Fig. 4C). Untreated animals (Fig. 4A) or animals injected with corn oil alone (Fig. 4B) did not express Alx4. Fig. 4G shows further that an increased number of stromal, and epithelial cells were proliferating under the influence of 17β-estradiol as evidenced by the increase in BrdU containing cells relative to the control animals (Figs. 4E–F). The induction of Alx4 is not synchronous with 17βestradiol-dependent induction of cell cycle progression, however. Twenty-four hours after 17β-estradiol administration, an increase in BrdU-positive cells (Fig. 4H), but not Alx4expressing cells (Fig. 4D), is observed.

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

287

Fig. 1. Alx4 is expressed during postnatal mammary gland development. Whole-mount analysis of mammary glands in prepubescent (A), pubescent (C), adult virgin (F) and day 9 pregnant mice (I). Immunohistoichemical detection of Alx4 expression in prepubescent (B), pubescent (D–E), adult virgin (G–H) and day 9 pregnant mice (J). Arrows in panels E and H denote Alx4 expressing cells. Scale bars in A–200 μm; scale bars in panels C, F and I–1000 μm; scale bars in panels B, D, E, G, H and J–15 μm.

Branching morphogenesis is impaired in the mammary glands D D of Alx4lst /lst mice Normal mammary gland development during puberty requires signaling between TEBs, which give rise to the luminal and myoepithelial cells, and the adjacent stromal cells. We hypothesized that loss of Alx4 activity would impair signaling from Alx4-expressing stromal cells to the epithelial compart-

ment, resulting in defective ductal morphogenesis. In order to test this hypothesis, we employed mice lacking functional Alx4. Strong’s luxoid (lstD ) is an inbred strain of mice harboring a point mutation (R206Q) in the DNA-binding homeodomain of Alx4, rendering the protein inactive (Qu et al., 1998). Mice homozygous for the lstD allele (Alx4lstD/lstD ) phenocopy mice with a complete deletion of Alx4 (Qu et al., 1998). We backcrossed the lstD mutation onto C57Bl/6 (>12 generations)

Fig. 2. Colocalization of Alx4 and ERα expression in mammary gland stromal cells. Immunofluorescent labeling reveals Alx4 expression (A) in mammary stroma colocalizes with stromal cells expressing ERα (B). Overlapping expression seen in merged image (C). Scale bar in A–15 μm.

288

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

Fig. 3. Detection of Alx4 in primary fibroblast populations isolated from the pubescent mouse mammary gland. (A) Immunofluorescent detection of Alx4 protein in the nuclei of primary mammary fibroblast cells 48 h after plating in tissue culture dishes. (B) Confirmation by RT-PCR of expression of the Alx4 transcript in isolated stromal cells 4 h and 3 days after plating stromal fibroblasts. (C) RT-PCR on fraction of isolated fibroblasts expressing Alx4 also express message for HGF and IGF. Scale bars in panel A–20 μm.

and determined the effect of loss of Alx4 activity on postnatal mammary gland development. Lack of Alx4 activity results in profound impairment in ductal morphogenesis. By whole-mount analysis, several striking features are evident in the developing ductal epithelia in Alx4lstD/lstD animals during puberty (Fig. 5). Progression of ductal invasion into the fat pad during puberty is greatly delayed in the animals lacking Alx4 (Fig. 5A versus B). It is also evident that aberrant morphogenesis arises at this stage. Specifically, ducts between the nipple region and the lymph node are very large in diameter, often appearing disoriented and having highly distorted and variable shapes. For example, ducts display

sudden alterations in diameter (Fig. 5E, arrow) where an unusually large duct will give rise to a smaller, distal structure. Bifurcation of the terminal end buds is also often impaired with the angle of bifurcation considerably reduced relative to wildtype TEBs (compare Figs. 5C–D with F and I, double arrow head). Indeed, TEB are occasionally observed to overlap (Figs. 5I and F, double arrow head) rather than extend away from each other (Figs. 5C–D, double arrow head). The TEB themselves are typically misshaped, the region containing body cells often extending backwards from the TEB (Fig. 5H, solid arrow heads). Interestingly, ducts often appear to have nodes of cells in their lumens (Figs. 5F and J, single arrows), a structure not seen in wild-type ducts. We also examined the structure of the mammary glands of adult virgin Alx4lstD/lstD animals (Fig. 6). The ductal network in the mammary glands of Alx4lstD/lstD females typically reach the limits of the fat pad. It is evident, however, that the defects observed in the pubescent mice persist in adult virgin animals. So, for example, ducts proximal to the nipple region are distorted, exhibiting twisted and kinked shapes and overall being considerably larger in diameter (compare Figs. 6E and C). In order to confirm that branching morphogenesis was impaired, we quantified the number of branch points in pubescent and adult animals. As Fig. 7 reveals, a significant reduction in total number of branches in mammary glands of Alx4lstD/lstD mice versus wild-type litter mates, respectively, is observed in both pubescent (45.33 ± 14.77 versus 164 ± 17.93; P = 0.0342) and in adult virgin (78.33 ± 8.01 versus 230.67 ± 39.60; P = 0.0413) animals. The reduction of branch points during puberty is reflected in the significant reduction in TEBs (7.67 ± 1.76 versus 21.33 ± 1.2; P = 0.0111) in Alx4lstD/lstD mice compared to wild-type mammary glands, respectively. When corrected for differences in the size of the adult mammary glands, differences in the density of branching is also evident. Thus, during puberty (45.33 ±14.77 versus 95.67 ± 14.84, P = 0.0016) and in adult virgins (78.33 ± 8.01 versus

Fig. 4. 17β-Estradiol-induced expression of Alx4 in the prepubescent mammary gland. (A–D) Alx4 expression in prepubescent mammary glands from animals that were untreated (A), treated with corn oil for 48 h (B), treated with corn oil plus 17β-estradiol for 48 h (C) or treated with corn oil plus 17β-estradiol for 24 h (D). (E–H) Detection of BrdU incorporation in the same mammary glands as panels A–D. Arrows in panels B and C–Alx4-expressing cells. Scale bars in panels A–H–15 μm.

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

289

Fig. 5. Impairment of ductal morphogenesis during puberty in Alx4lstD/lstD mice. Whole-mount analysis of wild-type (A, C, D) and Alx4lstD/lstD (B, E–J) mammary glands. Invasion of the fat pad during puberty is delayed in Alx4lstD/lstD mutant mammary glands. TEBs often bifurcate at very close angles in Alx4lstD/lstD mice compared to the significant spacing seen in wild-type TEBs (compare double arrows heads in panels I and F versus C and D, respectively). Lumens of ducts in Alx4lstD/lstD mammary glands frequently exhibit nodes of cells (closed arrows in panels F and J). Body cells in TEBs in Alx4lstD/lstD mammary glands often extend back towards the subtending ducts (arrowheads in panel H). Radical changes in the size of ducts at bifurcations are also seen in Alx4lstD/lstD mice where large, distorted ducts give rise to a duct with very small diameter (open arrows in panels E and G). Scale bars in panels A–B 1000 μm; scale bars in panels C, E, G and I–250 μm; scale bars in panel D–125 μm; scale bars in panels F, H and J–100 μm.

122.67 ± 15.90, P = 0.0387), the density of branches in Alx4lstD/lstD mice versus wild-type littermates, respectively, is significantly reduced. Impairment of branching morphogenesis is due to defects in stromal cells In order to discount the possibility that the observed alterations were due to systemic effects in Alx4lstD/lstD mice and to prove that they arose due to defects in the stromal compartment, we performed transplantation experiments using recombined fractions of stromal and epithelial cells. Specifically, separate populations of stromal cells and epithelial cells were isolated from 6-week-old Alx4lstD/lstD mice and wild-type littermates. Stromal cells from Alx4lstD/lstD mice were then mixed with wild-type epithelial cells and vice versa and these reciprocally recombined populations injected into the cleared fat pads of prepubescent wild-type recipients. As Fig. 8 demonstrates, retarded ductal morphogenesis is seen five weeks after

transplantation in the mammary glands of recipients receiving wild-type epithelial cells but stromal cells from Alx4lstD/lstD mice. In contrast, development of mammary glands of recipients receiving wild-type stromal cells but epithelial cells from Alx4deficient mice progressed normally. Thus, these data reveal that impaired ductal morphogenesis in Alx4lstD/lstD mice arises specifically due to loss of Alx4 activity in mammary stromal cells. Despite alterations in the gross morphology of the mammary glands in Alx4-deficient mice, consistent differences in the fine structure of ducts and surrounding stroma are not observed (Fig. 9). For example, mutant Alx4 protein is expressed in Alx4lstD/lstD mice. Fig. 9B illustrates that the cells which express mutant Alx4 are present in the region surrounding the terminal end buds, similar to Alx4-expressing cells in wild-type littermates (Fig. 9A). Likewise, expression of p63 in myoepithelial cells and in the cap cell layer of TEBs in Alx4-deficient animals is indistinguishable from wild-type littermates (Figs. 9E–F). Picro-sirius red staining for extracellular matrix

290

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

Fig. 6. Defective ductal morphogenesis in adult virgin Alx4lstD/lstD animals. Whole-mount analysis of wild-type (A, C and D) and Alx4lstD/lstD (B, E, F) mammary glands. In adult virgin Alx4lstD/lstD mammary glands, ductal growth has reached the limits of the fat pad (B) as observed in wild types (A). However, ducts in Alx4lstD/lstD animals retain many aspects of abnormal morphology found in pubescent mutants. Ducts proximal to the nipple region in Alx4lstD/lstD animals are distorted and much larger (E) than those in comparable locations in wild types (C), while ducts distal to the nipple are much thinner (F) than wild-type ducts (D).

Fig. 7. Defective branching in Alx4lstD/lstD mammary glands. Branch points in mammary glands from Alx4lstD/lstD and wild-type littermates were quantified from whole-mount preparations of mammary gland #4. The total number of branch points during puberty and in adult virgins was significantly reduced in Alx4lstD/lstD compared to wild-type littermates. When corrected for differences in the size of the mammary glands, a significant reduction in the density of branches in Alx4lstD/lstD females was apparent at both stages. Comparison of the number of TEBs during puberty also showed a significant reduction in Alx4lstD/lstD females relative to wild-type littermates.

proteins also failed to reveal differences between mutant and wild-type animals (Figs. 9C–D). Normal expression patterns for laminin, keratin 5, keratin 14, smooth muscle actin, E-cadherin and β-catenin are also observed in Alx4lstD/lstD mammary glands (see Supplementary data). Furthermore, the apparent rate of proliferation in terminal end buds, as determined by BrdU incorporation, is similar between Alx4lstD/lstD and wild-type littermates (Figs. 9G–H). Targets for Alx4-dependent transcription are not known. However, a number of factors expressed in the stromal compartment of the mammary gland are known to be required for normal morphogenesis of the mammary gland during puberty, such as HGF (Woodward et al., 1998; Zhang et al., 2002) and matrix metalloproteinases (Wiseman et al., 2003). Thus, we quantified the levels of expression of factors expressed in stromal cells of 5-week-old wild-type and Alx4lstD/lstD pubescent littermates. As illustrated in Fig. 10, when normalized against the β-actin control, no significant differences in the levels of HGF or MMP3 message were evident in stromal cells isolated from wild type compared to Alx4lstD/lstD littermates. In contrast, we observed a 40% increase in MMP2 and 50% reduction in MMP9 mRNA expression. Thus, the defects associated with the loss of Alx4 activity in the pubescent mammary gland are associated with altered expression of two factors known to play a role in mammary gland development.

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

291

Fig. 8. Altered branching morphogenesis in Alx4lstD/lstD mammary glands is due to defects in stromal cells. Mammary epithelial cells and stromal cells were individually isolated from 5-week-old wild-type and Alx4lstD/lstD females. Mutant stromal cells and wild-type epithelia as well as wild-type stroma and mutant epithelia were recombined and transplanted into contralateral cleared fat pads of wild-type pubescent recipients. Mammary glands from recipients were harvested 5 weeks later and whole mounts of mammary glands reconstituted with Alx4lstD/lstD epithelia plus wild-type stroma (top panel) or wild-type epithelia plus Alx4lstD/lstD stroma (bottom panel) prepared for analysis.

Taken together, these data define a subset of stromal cells in the murine mammary gland which express the homeodomain factor Alx4. Significantly, in the absence of Alx4 activity in these stromal cells, mammary epithelial ductal morphogenesis is impaired giving rise to aberrant ductal structures and a reduction in the number of branches in the principal ductal network. Discussion Branching morphogenesis of the postnatal mammary gland requires continuous reciprocal signaling between the growing epithelia and its surrounding stromal microenvironment (Parmar and Cunha, 2004; Robinson et al., 1999; Silberstein, 2001; Wiseman and Werb, 2002). In this study, we identified a subset of stromal cells associated primarily with the TEB which express the homeodomain factor, Alx4. We showed previously that Alx4 is expressed in the mesenchymal aspect of another skin appendage, the hair follicle, during both embryonic and postnatal development (Hudson et al., 1998 and P. Joshi, unpublished observation). In the case of the mammary gland, Alx4 expression is not observed during embryonic mammary gland development (data not shown) or during quiescent stages of postnatal mammary gland development. However, induction of Alx4 expression is observed as the mammary ductal structures

penetrate the fat pad during puberty and during alveolar formation at pregnancy. Consistent with the appearance of its expression during ovarian hormone-dependent stages of mammary gland morphogenesis, Alx4 expression increases following administration of 17β-estradiol in prepubescent females. We showed further a role for Alx4 in mammary gland development where, in its absence, impairment of ductal and branching morphogenesis arises. This effect is due to loss of Alx4 activity specifically in the stromal compartment since transplantation of wild-type epithelia and Alx4-deficient stromal cells into the cleared fat pad of a wild-type recipient resulted in impaired ductal morphogenesis. Thus, we have identified a homeodomain factor defining a subset of stromal cells adjacent to TEBs important for normal mammary epithelial morphogenesis. The principal hormone regulating ductal growth during puberty is estrogen secreted by the ovaries. One mediator of estrogen signaling, ERα, is expressed in both epithelial and stromal cells of the mouse mammary gland (Fendrick et al., 1998; Silberstein et al., 1994). While ERα is expressed in both mammary stromal cells as well as in epithelial cells, recent evidence revealed that expression of ERα in mammary epithelial cells, but not in mammary stromal cells, was required for morphogenesis during puberty (Mallepell et al., 2006). Previous studies indicated that ERα may mediate some of its morphogenic effects through transcriptional control of HGF

292

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

Fig. 9. Alx4lstD/lstD mammary glands are similar to wild types at the histological level. Expression of the Alx4 functionally null protein is detected in stromal cells surrounding TEBs in Alx4lstD/lstD mammary glands (B), comparable to the expression pattern along wild-type TEBs (A). Picro-sirius red staining of collagen and the basement membrane did not reveal any defects in the ECM of Alx4lstD/lstD mammary glands (D) compared to the wild type (C). Organization of cells in TEBs was intact in Alx4lstD/lstD animals as evidenced by p63 staining for the cap cell layer (compare F with E). Cell proliferation was mostly localized to the cap cell layer in Alx4lstD/lstD TEBs (H) and was similar to the wild type (G). Scale bars in panels A–B–15 μm; scale bars in panels C–D–50 μm; scale bars in panels E–H–25 μm.

expression in mammary fibroblasts (Zhang et al., 2002), suggesting that ERα activity in the stromal compartment plays a role in normal mammary gland development. Given the induction of Alx4 at the onset of puberty and in prepubescent mice following administration of 17-β-estradiol, we predict that Alx4 may mediate, at least in part, some of these estrogendependent effect during mammary gland morphogenesis at puberty. The mechanisms determining TEB bifurcation is not well understood (for example see (Silberstein, 2001). We observed that at all stages of postnatal mammary gland development, the relative number of branch points in animals lacking Alx4 was significantly reduced. This reduction was also observed for adult virgin Alx4lstD/lstD mice where, despite reaching the limits of the

fat pad, the absolute and relative number of branch points remained reduced compared to the wild type. Failure to restore the relative number of branch points in adult Alx4lstD/lstD mice is similar to Csfmop mice (Gouon-Evans et al., 2000). In these latter animals, loss of macrophage-colony stimulating factor-1 (CSF1) expression from epithelial cells results in a failure to recruit macrophages to the TEB during puberty. Like Alx4-deficient animals, Csfmop mice exhibit retarded ductal morphogenesis and a decrease in the number of branches. Defective ductal morphogenesis in Csfmop and Alx4lstD/lstD mice have a number of additional overlapping characteristics. Significantly, as was described for Csfmop mice and as we showed for Alx4lstD/lstD mice, terminal end buds are often disoriented. This disorientation presumably gives rise to aberrant ductal elongation, ducts

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

293

Fig. 10. Altered expression of factors in mammary stromal cells from Alx4lstD/lstD mice. Expression levels of HGF, MMP2, MMP3 and MMP9 in stromal cells isolated from wildtype (WT) and Alx4lstD/lstD (lstD) mice were determined by quantitative RT-PCR (qPCR). No differences in the levels of HGF and MMP3 were observed between pairs of WT and lstD mice. In contrast, MMP2 was expressed at levels 40% higher and MMP9 at levels 50% lower in stromal cells from lstD mice relative to their WT littermates.

often looping backwards or crossing over other ducts. Indeed, we observe in Alx4lstD/lstD mammary glands, that the architecture of bifurcation is significantly impaired. Unlike Csfmop mice, however, the defects observed in Alx4-deficient animals do not appear to be the result of impaired recruitment of eosinophils or macrophages; both of these cells types are detected in the stroma associated with TEBs in pubescent Alx4-deficient mice (data not shown). The transplantation studies confirmed that the defects in Alx4lstD/lstD mice were due specifically to stromal cells and was not a cell-autonomous effect of epithelial cells. Given that stromal cells expressing inactive Alx4 were detected around the TEBs of pubescent Alx4lstD/lstD mice, we propose that the phenotype of the mammary gland in these animals arises due to the loss of expression of factors under the transcriptional control of Alx4. While no endogenous transcriptional targets of Alx4 have been characterized, we detected altered expression of two factors, MMP2 and MMP9, known to participate in the mammary gland morphogenesis during puberty (Wiseman et al., 2003). Alterations in this class of proteins may be consistent with the fact that Alx4lstD/lstD mice exhibit altered gross morphology and impaired branching morphogenesis rather than an apparent loss of some particular cell type. Thus, changes in the expression of factors which are required for the mechanical penetration of the TEB into the fat pad might be expected, as we showed for some of the MMPs, rather than changes in factors which control morphogenesis, as was the case for HGF. Indeed, given the lack of differences in the level of expression of HGF in stromal cells from Alx4lstD/lstD versus wild-type littermates, we predict that estrogen-induced activation of Alx4 expression may be required for control of signaling pathways distinct from HGF but which may mediate HGFdependent branching morphogenesis. Microarray analysis is currently being used to identify altered gene expression in a global systematic manner. To conclude, these data demonstrate that the homeodomain factor, Alx4, is expressed in a subset of stromal cells in the

mammary gland. Furthermore, in the absence of this estrogeninducible protein, mammary epithelial morphogenesis is impaired. Acknowledgments The authors wish to thank the members of the PAH laboratory as well as Cathrin Brisken for helpful discussions during this work. This work was funded by a joint Canadian Breast Cancer Research Alliance/Canadian Institutes of Health Research (MOP-49614) grant awarded to P.A.H. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ydbio.2006.05.032. References Boras, K., Hamel, P.A., 2002. Alx4 binding to LEF-1 regulates N-CAM promoter activity. J. Biol. Chem. 277, 1120–1127. Boras-Granic, K., Grosschedl, R., Hamel, P.A., 2006. Genetic interaction between Lef1 and Alx4 is required for early embryonic development. Int. J. Dev. Biol. 50 (7), 601–610. Chan, D.C., Laufer, E., Tabin, C., Leder, P., 1995. Polydactylous limbs in Strong's Luxoid mice result from ectopic polarizing activity. Development 121, 1971–1978. Fendrick, J.L., Raafat, A.M., Haslam, S.Z., 1998. Mammary gland growth and development from the postnatal period to postmenopause: ovarian steroid receptor ontogeny and regulation in the mouse. J. Mammary Gland Biol. Neoplasia 3, 7–22. Forsthoefel, P.F., 1962. Genetics and manifold effects of Strong's luxoid gene in the mouse, including its interactions with Green's luxoid and Carter's luxate genes. J. Morphol. 110, 391–420. Forsthoefel, P.F., 1963. The embryological development of the effects of strong's luxoid gene in the mouse. J. Morphol. 113, 427–451. Forsthoefel, P.F., Fritts, M.L., Hatzenbeler, L.J., 1966. The origin and development of alopecia in mice homozygous for strong's luxoid gene. J. Morphol. 118, 565–580. Gouon-Evans, V., Rothenberg, M.E., Pollard, J.W., 2000. Postnatal mammary

294

P.A. Joshi et al. / Developmental Biology 297 (2006) 284–294

gland development requires macrophages and eosinophils. Development 127, 2269–2282. Gouon-Evans, V., Lin, E.Y., Pollard, J.W., 2002. Requirement of macrophages and eosinophils and their cytokines/chemokines for mammary gland development. Breast Cancer Res. 4, 155–164. Hudson, R., Taniguchi-Sidle, A., Boras, K., Wiggan, O., Hamel, P.A., 1998. Alx-4, a transcriptional activator whose expression is restricted to sites of epithelial-mesenchymal interactions. Dev. Dyn. 213, 159–169. Ip, M.M., Darcy, K.M., 1996. Three-dimensional mammary primary culture model systems. J. Mammary Gland Biol. Neoplasia 1, 91–110. Kamalati, T., Niranjan, B., Yant, J., Buluwela, L., 1999. HGF/SF in mammary epithelial growth and morphogenesis: in vitro and in vivo models. J. Mammary Gland Biol. Neoplasia 4, 69–77. Kleinberg, D.L., Feldman, M., Ruan, W., 2000. IGF-I: an essential factor in terminal end bud formation and ductal morphogenesis. J. Mammary Gland Biol. Neoplasia 5, 7–17. Kratochwil, K., Dull, M., Farinas, I., Galceran, J., Grosschedl, R., 1996. Lef1 expression is activated by BMP-4 and regulates inductive tissue interactions in tooth and hair development. Genes Dev. 10, 1382–1394. Mallepell, S., Krust, A., Chambon, P., Brisken, C., 2006. Paracrine signaling through the epithelial estrogen receptor alpha is required for proliferation and morphogenesis in the mammary gland. Proc. Natl. Acad. Sci. U. S. A. 103, 2196–2201. Master, S.R., Stoddard, A.J., Bailey, L.C., Pan, T.C., Dugan, K.D., Chodosh, L.A., 2005. Genomic analysis of early murine mammary gland development using novel probe-level algorithms. Genome Biol. 8, R20. Mavrogiannis, L.A., Antonopoulou, I., Baxova, A., Kutilek, S., Kim, C.A., Sugayama, S.M., Salamanca, A., Wall, S.A., Morriss-Kay, G.M., Wilkie, A.O., 2001. Haploinsufficiency of the human homeobox gene ALX4 causes skull ossification defects. Nat. Genet. 27, 17–18. Nandi, S., 1958. Endocrine control of mammarygland development and function in the C3H/He Crgl mouse. J. Natl. Cancer Inst. 21, 1039–1063. Niranjan, B., Buluwela, L., Yant, J., Perusinghe, N., Atherton, A., Phippard, D., Dale, T., Gusterson, B., Kamalati, T., 1995. HGF/SF: a potent cytokine for mammary growth, morphogenesis and development. Development 121, 2897–2908. Pollard, J.W., Hennighausen, L., 1994. Colony stimulating factor 1 is required for mammary gland development during pregnancy. Proc. Natl. Acad. Sci. U. S. A. 91, 9312–9316. Qu, S., Li, L., Wisdom, R., 1997. Alx-4: cDNA cloning and characterization of a novel paired-type homeodomain protein. Gene 203, 217–223. Qu, S., Tucker, S.C., Ehrlich, J.S., Levorse, J.M., Flaherty, L.A., Wisdom, R., Vogt, T.F., 1998. Mutations in mouse Aristaless-like4 cause Strong's luxoid polydactyly. Development 125, 2711–2721. Rasmussen, S.B., Young, L.J.T., Smith, G.H., 2000. Preparing Mammary Gland Whole Mounts from Mice. Plenum Publishers, New York. Richert, M.M., Schwertfeger, K.L., Ryder, J.W., Anderson, S.M., 2000. An atlas of mouse mammary gland development. J. Mammary Gland Biol. Neoplasia 5, 227–241. Robinson, G.W., Karpf, A.B., Kratochwil, K., 1999. Regulation of mammary gland development by tissue interaction. J. Mammary Gland Biol. Neoplasia 4, 9–19. Ruan, W., Kleinberg, D.L., 1999. Insulin-like growth factor I is essential for terminal end bud formation and ductal morphogenesis during mammary development. Endocrinology 140, 5075–5081.

Silberstein, G.B., 2001. Postnatal mammary gland morphogenesis. Microsc. Res. Tech. 52, 155–162. Silberstein, G.B., Van Horn, K., Shyamala, G., Daniel, C.W., 1994. Essential role of endogenous estrogen in directly stimulating mammary growth demonstrated by implants containing pure antiestrogens. Endocrinology 134, 84–90. Soriano, J.V., Pepper, M.S., Orci, L., Montesano, R., 1998. Roles of hepatocyte growth factor/scatter factor and transforming growth factor-beta1 in mammary gland ductal morphogenesis. J. Mammary Gland Biol. Neoplasia 3, 133–150. Sternlicht, M.D., Sunnarborg, S.W., Kouros-Mehr, H., Yu, Y., Lee, D.C., Werb, Z., 2005. Mammary ductal morphogenesis requires paracrine activation of stromal EGFR via ADAM17-dependent shedding of epithelial amphiregulin. Development 132, 3923–3933. Valente, M., Valente, K.D., Sugayama, S.S., Kim, C.A., 2004. Malformation of cortical and vascular development in one family with parietal foramina determined by an ALX4 homeobox gene mutation. AJNR Am. J. Neuroradiol. 25, 1836–1839. van Genderen, C., Okamura, R.M., Farinas, I., Quo, R.G., Parslow, T.G., Bruhn, L., Grosschedl, R., 1994. Development of several organs that require inductive epithelial-mesenchymal interactions is impaired in LEF1-deficient mice. Genes Dev. 8, 2691–2703. Veltmaat, J.M., Mailleux, A.A., Thiery, J.P., Bellusci, S., 2003. Mouse embryonic mammogenesis as a model for the molecular regulation of pattern formation. Differentiation 71, 1–17. Walden, P.D., Ruan, W., Feldman, M., Kleinberg, D.L., 1998. Evidence that the mammary fat pad mediates the action of growth hormone in mammary gland development. Endocrinology 139, 659–662. Wiseman, B.S., Werb, Z., 2002. Stromal effects on mammary gland development and breast cancer. Science 296, 1046–1049. Wiseman, B.S., Sternlicht, M.D., Lund, L.R., Alexander, C.M., Mott, J., Bissell, M.J., Soloway, P., Itohara, S., Werb, Z., 2003. Site-specific inductive and inhibitory activities of MMP-2 and MMP-3 orchestrate mammary gland branching morphogenesis. J. Cell Biol. 162, 1123–1133. Woodward, T.L., Xie, J.W., Haslam, S.Z., 1998. The role of mammary stroma in modulating the proliferative response to ovarian hormones in the normal mammary gland. J. Mammary Gland Biol. Neoplasia 3, 117–131. Wu, Y.Q., Badano, J.L., McCaskill, C., Vogel, H., Potocki, L., Shaffer, L.G., 2000. Haploinsufficiency of ALX4 as a potential cause of parietal foramina in the 11p11.2 contiguous gene-deletion syndrome. Am. J. Hum. Genet. 67, 1327–1332. Wuyts, W., Cleiren, E., Homfray, T., Rasore-Quartino, A., Vanhoenacker, F., Van Hul, W., 2000. The ALX4 homeobox gene is mutated in patients with ossification defects of the skull (foramina parietalia permagna, OMIM 168500). J. Med. Genet. 37, 916–920. Yang, Y., Spitzer, E., Meyer, D., Sachs, M., Niemann, C., Hartmann, G., Weidner, K.M., Birchmeier, C., Birchmeier, W., 1995. Sequential requirement of hepatocyte growth factor and neuregulin in the morphogenesis and differentiation of the mammary gland. J. Cell Biol. 131, 215–226. Zhang, H.Z., Bennett, J.M., Smith, K.T., Sunil, N., Haslam, S.Z., 2002. Estrogen mediates mammary epithelial cell proliferation in serum-free culture indirectly via mammary stroma-derived hepatocyte growth factor. Endocrinology 143, 3427–3434.