- Email: [email protected]
Crim1 has an essential role in glycogen trophoblast cell and sinusoidal-trophoblast giant cell development in the placenta
Placenta 33 (2012) 175e182
Contents lists available at SciVerse ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
Crim1 has an essential role in glycogen trophoblast cell and sinusoidaltrophoblast giant cell development in the placenta D.J. Pennisi a, b, *, G. Kinna a, c, H.S. Chiu a, b, D.G. Simmons b, L. Wilkinson a, M.H. Little a a
Institute for Molecular Bioscience, The University of Queensland, Brisbane 4072, Australia School of Biomedical Sciences, The University of Queensland, Chancellor’s Place, Brisbane 4072, Australia c Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane 4072, Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 16 December 2011
Normal placental development and function is essential for fetal growth of eutherian mammals. Mutational studies have shown that numerous growth factors are required for placental development and differentiation of placental lineages. Here, using a gene-trap mutant mouse line, Crim1KST264, we show that Crim1 is essential for murine placental development. Crim1 is a developmentally expressed, trans-membrane regulator of growth factor activity. Crim1KST264/KST264 mutant placentae displayed hypoplasia from 13.5 dpc, and altered structure from 15.5 dpc, including alterations in cell number in both the junctional and labyrinth zones. Using the reporter gene from the Crim1KST264 allele, we found that Crim1 is expressed in multiple cell types of the placenta, including strong expression in the spongiotrophoblast cells of the junctional zone. In the junctional zone of Crim1KST264/KST264 placentae, there was an increase in the glycogen trophoblast cells adjacent to the spongiotrophoblast cells. In the labyrinth zone, we found a decrease in the density of sinusoidal-trophoblast giant cells. Our findings show that Crim1 is required for placental development, and is necessary for the proper differentiation of sinusoidal-trophoblast giant cells and glycogen trophoblast cells. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Crim1 Spongiotrophoblast Glycogen trophoblast cells Lineage development
1. Introduction The placenta is an organ essential for normal embryonic development and survival in utero. In the mouse, placental development begins soon after implantation at around 4.5 days post coitum (dpc) when a population of trophoblast stem cells emerge from the polar trophectoderm overlying the inner cell mass (ICM) of the blastocyst. Concurrently, the trophectoderm not in close contact with the ICM, the mural trophectoderm, undergoes differentiation into the first wave of trophoblast giant cells (TGC); a large polyploid cell type involved in early invasion, endocrine signalling and formation of the yolk sac [1]. The continued growth of the polar trophectoderm produces two important structures of the early placenta, the ectoplacental cone (EPC) and the chorion. The EPC contains progenitors for three trophoblast lineages: a second wave of TGCs that form a layer adjacent to the maternal decidua, and spongiotrophoblast and glycogen trophoblast cells of the middle
layer of the placenta; together these three cell types comprise the junctional zone (JZ). In contrast, the chorion contains progenitors for the syncytiotrophoblast cells of the labyrinth zone (LZ), the site of foetal-maternal exchange. Soon after chorio-allantoic fusion at around 8.0e8.5 dpc, structural changes appear in the chorion trophoblast cells, and the chorion begins to develop folds that extend into primary branches in the nascent LZ [2]. Within the LZ, the allantoic mesoderm gives rise to mesenchyme and blood vessels, whereas the trophoblast cells give rise to several different epithelial derivatives. By early 9.0 dpc, primary villous branches forming within the developing labyrinth are apparent and are lined by thin, elongated trophoblast cells [2]. These primary branches then undergo secondary branching and elongation to form the mature LZ. Ultimately, the labyrinthine trophoblast cells form three layers that make up the placental membrane separating maternal blood from the foetal vasculature [3]. 1.1. Trophoblast giant cells
* Corresponding author. School of Biomedical Sciences, The University of Queensland, Chancellor’s Place, Brisbane 4072, Australia. Tel.: þ61 7 3365 4656; fax: þ61 7 3365 1766. E-mail address: [email protected] (D.J. Pennisi). 0143-4004/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2011.12.014
While all TGCs are large polyploid cells with an established endocrine activity, the population consists of several unique subtypes that differ somewhat in the extent of their ploidy, gene expression profile, location, timing of appearance and
176
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
Fig. 1. The gross morphology and histology of Crim1KST264/KST264 placentae. AeC: Micrographs of 15.5 dpc placentae and embryos with intact yolk sac viewed in whole-mount. Arrowheads indicate the extent of the junctional zone (AeC). Note the exencephaly in the Crim1KST264/KST264 embryo (arrow, C). DeF: Micrographs of 15.5 dpc placentae after dissection of the embryo and yolk sac, viewed in whole-mount. Note the diminished size of the Crim1KST264/KST264 placenta relative to littermate controls. GeI: Micrographs of 17.5 dpc placentae and embryos viewed in whole-mount. Arrowheads indicate the extent of the junctional zone (GeI). Note, that in the Crim1KST264/KST264 placenta (I), the junctional zone is less prominent in proximal view, consistent with a loss of cells in this region. JeL: Micrographs of representative mid-sagittal, haematoxylin and eosin-stained sections of
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
developmental origins [4]. Parietal TGCs (P-TGCs) are the TGCs that line the implantation chamber and separate underlying placental layers from the maternal decidua; this population is comprised of both primary TGCs, the first differentiated cell type derived from the mural trophectoderm of the blastocyst that contribute largely to the yolk sac, and secondary TGCs derived from the EPC. TGCs that migrate into the decidua usually invade only a short distance, however, a subgroup invades deeper into the spiral arteries to establish the connection between the endothelial cell-lined arteries to the hemochorial blood spaces found in rodent and primate placentas [5]. These endovascular TGCs (spiral artery-associated TGCs, or SpA-TGCs) express the proliferin gene, but not the placental lactogens commonly expressed by the other TGC subtypes [4]. When the spiral arteries of the rodent placenta reach the level of the JZ, they coalesce into several large canals that bring blood to the base of the placenta where it empties into smaller capillary sized sinusoids in the labyrinth, bringing maternal blood into close association with foetal capillaries. Unique TGC subtypes also line these canals (C-TGCs) and sinusoids (S-TGCs) [4]. Previously, all secondary TGCs were thought to pass through a Tpbpa-expressing, intermediate stage in the EPC upon differentiation, however, more recent cell lineage tracing studies have indicated that P-TGCs and CTGCs can be derived from both Tpbpa-positive and Tpbpa-negative EPC progenitors [4]. Conversely, SpA-TGCs appear to be derived from Tpbpa-positive cells while S-TGCs arise exclusively from Tpbpa-negative progenitors. Despite these different developmental origins, TGCs do have universal aspects to their differentiation; all TGCs appear to require the transcription factor Hand1 for proper differentiation [6]. 1.2. Spongiotrophoblast and glycogen trophoblast cells The middle layer of the rodent placenta contains spongiotrophoblast (SpT) intermingled with islets of glycogen trophoblast cells (GCs). In the second half of gestation, some GCs leave the confines of this layer and invade interstitially into the decidua, congregating around the area of spiral arteries [7]; spongiotrophoblast do not appear to be motile and remain within the JZ of the placenta. The function of this cell layer remains incompletely understood, although it is known to secrete considerable amounts of peptide hormones and proteases [1] and is critically dependent on the transcription factor Mash2 for its formation [8]. Expression of Tpbpa is characteristic of both SpT and GCs [5,9] and is often used as a marker for these cell populations. The early identification of GCs has traditionally been done using PAS staining to localize glycogen content, prompting the initial idea that GCs were derived from SpT sometime after E11.5 [5,10], coinciding with the detection of significant PAS staining in the developing placenta. However, recent studies have suggested that GCs may be derived from specific, Pcdh12-expressing progenitors present within the EPC as early as E7.5, although this notion awaits confirmation from cell lineage tracing studies [11]. 1.3. Crim1
177
cells cystine-knot growth factors, including BMP4, BMP7, VEGF-A, PDGF-B, Nodal, and TGFb2 [13,14]. Moreover, loss of Crim1 function can affect the availability of growth factors to neighbouring cells [14]. We have previously described the pleiotropic phenotypes associated with Crim1KST264 mouse embryos that harbour a genetrap mutation [15]. Mice homozygous for the gene-trap mutation (Crim1KST264/KST264) on an inbred, C57Bl6 genetic background die perinatally, and display phenotypes including renal dysplasia, microphthalmia, reduced lens size, digit syndactyly, dermal blebbing, and cerebral haematoma [15]. Herein, we show that Crim1 is expressed in multiple cell types of the placenta and reveal essential roles for Crim1 in murine placental development using the Crim1KST264 gene-trap mouse line. Specifically, we observed changes in the relative proportions of S-TGCs and GCs Crim1KST264/KST264 placentae, demonstrating that Crim1 is necessary for the proper differentiation of both S-TGC and GC lineages. 2. Materials and methods 2.1. Maintenance of the gene-trap mouse line The Crim1KST264 gene-trap mutation has been described [15] and was maintained on a C57Bl6 background. Use of animals in this study was done in accordance with the Animal Ethics Committee, The University of Queensland (IMB/565/07/ NHMRC, IMB/142/09/NHMRC). Genotyping was performed as described [15]. 2.2. Sample preparation and X-Gal staining Whole embryos and placentae were dissected in phosphate-buffered saline (PBS), fixed in 4% paraformaldehyde (PFA) in PBS at 4 C for 2e4 h, dehydrated in an ethanol series, cleared in xylene, and processed for paraffin infusion and embedding. Serial, 7 mm, sections were cut and mounted on Superfrost slides. For X-Gal staining on sections, tissue was fixed in 4% PFA at 4 C, washed in PBS and incubated in 30% sucrose/PBS overnight at 4 C. Samples were then embedded in OCT medium, and 10 mm sections cut on a cryostat. Slides were air dried for 10e15 mins before washing in PBS. X-Gal buffer solution was added to slides and left in the dark at room temperature to develop for 48e72 h. Slides were washed in PBS, dehydrated in an ethanol series, cleared, and mounted. Some slides were counterstained with nuclear fast red (NFR; Vector Laboratories, California). Staining was never detected in wildtype tissue (data not shown). 2.3. Histochemistry and section in situ hybridization Masson’s trichrome, periodic acid-Schiff (PAS), and haematoxylin and eosin staining were performed using standard procedures. In situ hybridization was performed on PFA-fixed, paraffin-embedded placentae according to published methods [16]. Cathepsin Q (Ctsq), Placental Lactogen II (PLII; Prl3b1), and Protocadherin12 (Pcdh12) anti-sense RNA probe production has been previously described [3]. Trophoblast-specific protein alpha (Tpbpa) anti-sense RNA probe was produced according to published methods [17]. The Tpbpa-specific primers used were; Forward Primer, 50 -TGAAGAGCTGAACCACTGGA-30 ; Reverse Primer, 50 -CGATGTTAATACGACTCACTATAGGGAGTGCAGGATCCCACTTGTC-30 . 2.4. Data documentation Digital images of tissue sections were captured using either a semi-automated “slide” System from Olympus and Soft Imaging Systems, or an Olympus BX-51 BF/ DF slide microscope with Canon digital cameras, and with “OlyVIA” software (Soft Imaging Systems, Olympus) and DP Controller software (Olympus, Japan), respectively. Whole-mount images were captured using an Olympus SZX-12 stereomicroscope with DP Controller software. Images were adjusted for colour levels, brightness and contrast, and figures compiled, using Adobe Photoshop software. 2.5. Data quantification and statistical analyses
Crim1 is a multi-domain protein that is dynamically expressed during embryogenesis and contains six von Willebrand Factor-C (vWFC)-like cysteine-rich repeat domains similar to the BMPregulating protein, Chordin [12,13]. The six vWFC domains mediate the ability of Crim1 to bind and tether to the surface of
To determine weights, between nine and fourteen placentae and embryos of each genotype, from at least four different litters, were collected and weighed. For quantification of S-TGC density, sixty high magnification (100) images from at least three biological samples were done for each genotype analysed. Quantification of area of PAS staining, and Pcdh12 expression was performed on at least four sections
15.5 dpc placentae. The junctional zone (JZ) and labyrinth zone (LZ) are indicated. MeO: Higher magnification views of the boxed areas in JeL, respectively. Glycogen trophoblast cells (GC) and spongiotrophoblast cells (SpT) are indicated. Examples of Crim1þ/þ (A, D, G, J, and M), Crim1þ/KST264 (B, E, H, K and N) and Crim1KST264/KST264 (C, F, I, L and O) embryos and placentae are shown. Scale bar; F, 1 mm; O, 100 mm.
178
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
from each sample, and at least four samples from each genotype. Total placental area, and area of staining, were quantified with ImageJ software (NIH; http://rsb. info.nih.gov/ij/). For determining the statistical significance of placental and embryonic weight differences, a one-way unpaired ANOVA with Tukey’s multiple comparison tests was used to compare genotypes. For determining the statistical significance of S-TGC density, the area of PAS staining, and the area of Pcdh12 expression, an unpaired, two-tailed student’s t-test, with unequal variance with Welch’s correction was used to compare genotypes.
3. Results The first set of experiments sought to determine whether there were placental defects in Crim1KST264/KST264 mice. At 15.5 dpc, we found that Crim1KST264/KST264 placentae, when viewed from the proximal side, exhibited a less prominent JZ relative to littermate controls (Fig. 1AeC). Moreover, Crim1KST264/KST264 placentae appeared smaller than those of littermate controls in whole-mount (Fig. 1DeF, Supplemental Fig. 1). At 17.5 dpc, the gross morphological defect in Crim1KST264/KST264 placentae became more prominent, with the junctional zone being less prominent in proximal view (Fig. 1GeI). Defects in Crim1KST264/KST264 embryos can be observed from 11.5 dpc (peridermal blebbing [15]). However, no difference was observed between Crim1KST264/KST264 placentae and those of littermate controls based on gross morphology and histological sections at 11.5 dpc (data not shown). Haematoxylin and eosin-stained sections demonstrated that at 15.5 dpc, Crim1KST264/KST264 placentae had abnormal morphology compared to littermate controls (Fig. 1JeL). Histological analyses also indicated that presumptive glycogen trophoblast cells and spongiotrophoblast cells were present in the LZ (Fig. 1MeO). As the placentae of Crim1KST264/KST264 embryos appeared smaller than those of littermate controls, we weighed placentae and embryos. We observed that of Crim1KST264/KST264 placentae were significantly lighter than those of Crim1þ/KST264 littermate controls at 13.5 dpc (P < 0.05), 15.5 dpc (P < 0.05), and 17.5 dpc (P < 0.01; Fig. 2A). As impaired placental function may result in reduced embryo growth [18,19], we quantified embryo weights. At 13.5 dpc and 15.5 dpc, there was not a significant difference between Crim1þ/ KST264 and Crim1KST264/KST264 embryos. At 17.5 dpc, however, we observed that Crim1KST264/KST264 embryos were significantly lighter than Crim1þ/KST264 littermate controls (P < 0.05; Fig. 2B). We did not find a significant difference between Crim1þ/þ and Crim1þ/KST264 in weights of placentae or embryos at the stages examined (not shown). Although there was an observable difference in the weights of Crim1þ/KST264 and Crim1KST264/KST264 placentae, examination of BrdU uptake at 15.5 dpc did not reveal a marked alteration in the pattern of BrdUþ cells in Crim1KST264/KST264 placentae (Supplemental Fig. 2).
We have previously reported that Crim1 is expressed in the parietal trophoblast giant cells (P-TGCs) of the placenta at 17.5 dpc [15]. As Crim1KST264/KST264 placentae exhibited defects, we took advantage of the b-Geo reporter construct in the gene-trap line to further characterize Crim1 expression in the developing placenta. XGal staining on cryosections revealed b-Geo expression in decidual, JZ and LZ cells of Crim1þ/KST264 and Crim1KST264/KST264 placentae. At 15.5 dpc, strong expression was observed in Crim1þ/KST264 and Crim1KST264/KST264 placentae in a discontinuous pattern in the JZ (Fig. 3A and B). Higher magnification images reveal that most of the X-Galþ cells in the JZ were presumptive SpT cells, with little staining evident in GCs (Fig. 3C and D). X-Gal staining was observed in the SpT cells from 13.5 dpc in Crim1þ/KST264 placentae (Supplemental Fig. 3). At 9.5 dpc, X-Gal staining was observed amongst the allantoic mesenchyme and at low levels in some chorionic trophoblasts of Crim1þ/KST264 placentae (Supplemental Fig. 3). At 13.5 dpc and 15.5 dpc, X-Gal staining was observed in canal trophoblasts and at lower levels in some syncytiotrophoblasts; in some perivascular cells in the LZ of Crim1þ/KST264 placentae, presumably pericytes (Supplemental Fig. 3). At 15.5 dpc, some spiral artery-associated TGCs, and a few S-TGCs were X-Galþ (Supplemental Fig. 3). As Crim1KST264/KST264 embryos had abnormal placental morphology, and Crim1 was expressed in SpT cells of the JZ, we examined the morphology of the JZ in Crim1KST264/KST264 placentae in a manner independent of the Crim1 gene-trap reporter. To that end, section in situ hybridization for Trophoblast-specific protein alpha (Tpbpa) expression was performed on 15.5 dpc placenta. Robust expression of Tpbpa was found in the JZ of control and Crim1KST264/KST264 placentae, particularly amongst spongiotrophoblast cells (Fig. 4). Based on Tpbpa expression, the JZ in Crim1KST264/ KST264 placentae is prominent, but does not extend to the proximalmost part of the placenta (Fig. 4C and F). Thus, the histological and section in situ hybridization data highlighting the JZ in Crim1KST264/ KST264 placentae corroborate morphological observations made in whole-mount views (Fig. 1). As there was an alteration in the JZ of Crim1KST264/KST264 placentae, we sought to determine whether there was a change in the number of GCs in mutant placentae. We assessed periodic acid-Schiff (PAS) staining in Crim1KST264/KST264 placentae at 15.5 dpc, to identify GCs. We did not find a significant difference in the area of PAS staining in placentae at 15.5 dpc in Crim1KST264/KST264 relative to littermate controls (Fig. 5AeC). As PAS staining will detect mucin and basement membrane components, in addition to glycogen, we utilised the expression of Protocadherin12 (Pcdh12), a recently described molecular marker of GCs [11,20]. Therefore, to more accurately quantify the amount of GCs in Crim1KST264/KST264 placentae, we
Fig. 2. Crim1KST264/KST264 placentae and embryos are smaller than littermate controls. A. The weights of Crim1KST264/KST264 placentae were significantly smaller than littermate controls at 13.5 dpc, 15.5 dpc, and 17.5 dpc. B. The weights of Crim1þ/KST264 and Crim1KST264/KST264 embryos were similar at 13.5 dpc and 15.5 dpc, but by 17.5 dpc, Crim1KST264/KST264 embryos were significantly smaller. The weights of Crim1þ/þ and Crim1þ/KST264 embryos and placentae were not significantly different at any stage examined (data not shown). Red squares, Crim1þ/KST264 values; blue triangles, Crim1KST264/KST264 values. Error bars, SDM. *, P < 0.05; **, P < 0.01.
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
179
Fig. 3. The Crim1 gene-trap reporter is strongly expressed in the junctional zone of the developing placenta. A, B: Micrographs of representative mid-sagittal, X-Gal-stained sections of a 15.5dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae sections. C, D: Higher magnification images of representative mid-sagittal, X-Gal-stained sections that have been counterstained with nuclear fast red of 15.5 dpc Crim1þ/KST264 (C) and Crim1KST264/KST264 (D) placentae. Note the strong reporter gene expression in the spongiotrophoblasts. Note also the punctate, sub-cellular localization of X-Gal staining (C, D), reflecting the localization of the b-Geo reporter of this gene-trap construct [30]. Glycogen trophoblast cells (GC), spongiotrophoblast cells (SpT), junctional zone (JZ), and labyrinth zone (LZ) are indicated. Scale bars; B, 1 mm; D, 100 mm.
Fig. 4. The junctional zone has an altered morphology in Crim1KST264/KST264 placentae at 15.5 dpc. AeF: Micrographs of section in situ hybridization for Tpbpa expression in 15.5 dpc Crim1þ/KST264 (A, C, and D) and Crim1KST264/KST264 (B, E, and F) placentae. Representative mid-sagittal sections are shown. C, D and E, F: Higher magnification views of the indicated boxed areas in A and B, respectively. Note the strong Tpbpa expression in the spongiotrophoblast cells and lower levels of expression in glycogen trophoblast cells in Crim1þ/KST264 and Crim1KST264/KST264 placentae. Glycogen trophoblast cells (GC), spongiotrophoblast cells (SpT), junctional zone (JZ), and labyrinth zone (LZ) are indicated. Scale bars; A and B, 1.2 mm; C and E, 500 mm; D and F, 200 mm.
180
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
Fig. 5. There are more glycogen trophoblast cells (GCs) in Crim1KST264/KST264 placentae at 15.5 dpc. A, B: Micrographs of representative mid-sagittal, PAS-stained sections of 15.5 dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae. A0 , B0 : Higher magnification views of the boxed areas in A and B, respectively. C. Quantification of PAS staining in Crim1þ/KST264 and Crim1KST264/KST264 placentae. The difference in the fraction area stained PAS-positive was not significant (ns). D, E: Micrographs of section in situ hybridization for Pcdh12 expression in 15.5 dpc Crim1þ/KST264 (D) and Crim1KST264/KST264 (E) placentae. Representative mid-sagittal sections are shown. D0 , E0 : Higher magnification views of the boxed areas in D and E, respectively. F. Quantification of Pcdh12 expression in Crim1þ/KST264 and Crim1KST264/KST264 placentae. There was a significant increase in the fraction area stained positive for Pcdh12 expression in Crim1KST264/KST264 placentae (P < 0.005). Scale bars; B, E, 500 mm.
performed in situ hybridization to detect Pcdh12 transcripts. In littermate control 15.5 dpc placentae, we found that 4.4% (SDM, 1.9%) of placenta area stained positively for Pcdh12 expression. In contrast, we observed 7.6% (SDM, 0.9%) in Crim1KST264/KST264 placentae, a significant increase relative to littermate controls (P < 0.005; Fig. 5DeF). Note that these values for Pcdh12 expression (ranging from 4.4% to 7.6%) are lower than those obtained for PAS staining (ranging from 14.6% to 19.1%) and, presumably, reflect the more selective nature of the Pcdh12 expression. Although the strongest expression of the Crim1 gene-trap reporter was in the SpT cells of the JZ, lesser expression was also observed in some cells of the LZ. We therefore examined smooth muscle alpha actin (SMaA) protein, CD31 protein and Mest mRNA, markers of some LZ cell types. At 15.5 dpc, we did not observe a change in SMaA pattern or density in the LZ of Crim1KST264/KST264 placentae (Supplemental Fig. 4). We did, however, observe an increase in the fetal endothelial expression of Mest mRNA and a decrease in CD31 protein in Crim1KST264/KST264 placentae (Supplemental Fig. 5). As Crim1 was expressed in a dynamic manner in the LZ, we also sought to determine whether a lineage of the LZ, the S-TGCs, was affected in Crim1KST264/KST264 placentae. Therefore, we quantified the number of S-TGCs after Masson’s Trichrome staining on 15.5 dpc placentae (Fig. 6AeC). S-TGCs have large round nuclei and
are easily distinguished from other cell types such the fetal endothelium (FE) or the syncytium. We found 14.5 105 S-TGC/mm2 (SDM, 0.6 105) in control placentae. However, in Crim1KST264/ KST264 placentae, there was a significant reduction in the number, with 11.2 105 S-TGC/mm2 (SDM, 0.7 105) (P < 0.0005; Fig. 6). To confirm there was a reduction in the number of S-TGCs in Crim1KST264/KST264 placentae, we examined the expression of the S-TGC marker genes, Cathepsin Q (Ctsq) and Placental Lactogen II (PLII) [21e23] by in situ hybridisation. Relative to littermate control placentae, the LZ of Crim1KST264/KST264 placentae showed fewer cells that expressed Ctsq or PLII (Fig. 6DeG), providing qualitative data supporting the quantification of S-TGCs in Crim1KST264/KST264 placentae. The data demonstrate that in Crim1KST264/KST264 placentae, there is a reduction in the number of S-TGCs and a concurrent increase in GCs. 4. Discussion Numerous growth factors have been implicated in placental development and function, including members of the VEGF/PDGF family, TGFb superfamily, Wnts, and FGFs [24e27]. Using a genetrap mutant mouse line, we have found that Crim1, a transmembrane regulator of growth factor activity, is essential for normal placental development. The placental defects included
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
181
Fig. 6. There are less sinusoidal-trophoblast giant cells (S-TGCs) in Crim1KST264/KST264 placentae at 15.5 dpc. A, B: Micrographs of representative mid-sagittal, Masson’s trichrome-stained sections of the labyrinth zone of 15.5 dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae. S-TGCs are characterized by large nuclei (open arrowheads). Fetal blood space (FBS), fetal endothelium (FE; arrows), and maternal blood space (MBS; arrowheads) are indicated. C. Quantification of the density of S-TGCs in Crim1þ/KST264 and Crim1KST264/KST264 placentae. D, E: Micrographs of section in situ hybridization for Ctsq expression in the labyrinth zone of 15.5 dpc Crim1þ/KST264 (D) and Crim1KST264/KST264 (E) placentae. F, G: Micrographs of section in situ hybridization for PLII expression in the labyrinth zone of 15.5 dpc Crim1þ/KST264 (F) and Crim1KST264/KST264 (G) placentae. Note the reduction in the number of cells positive for Ctsq or PLII expression in the S-TGCs in Crim1KST264/KST264 LZ. S-TGCs are indicated (open arrowheads). Representative mid-sagittal sections are shown. Scale bars, 50 mm.
a reduction in size relative to wild-type controls, morphological changes, as well as changes in the proportion of the placenta’s constituent cell types. The earliest detected placental phenotype was the reduction in size at 13.5 dpc. This was followed by a reduction in the size of Crim1KST264/KST264 embryos at 17.5 dpc. It remains to be determined whether any potential dysgenesis or insufficiency in Crim1KST264/KST264 placentae contributed to the reduced size observed in Crim1KST264/KST264 embryos late in gestation. In this study, we used two markers to identify GCs in histological sections, PAS staining and Pcdh12 in situ hybridization. In addition to the substantial glycogen in GCs, PAS staining will detect mucin and basement membrane components. This could lead to an overestimation of GC content in sections of placentae using unbiased staining and morphometry techniques. Using Pcdh12 in situ hybridization to quantify GC content gave lower values than did the use of PAS staining in all genotypes. Furthermore, high magnification images of Pcdh12 in situ hybridization revealed strong signal in GCs with no background staining in adjacent SpTs. Together, these observations further support the use of Pcdh12 as a useful marker of GCs as suggested by Bouillot et al. [11]. Loss of Crim1 function resulted in an expansion of GCs in the JZ, and a decrease in S-TGCs of the LZ of the placenta. Although the syncytiotrophoblast and S-TGC lineage are derived from Tpbpanegative cells of the chorion and inner ectoplacental cone, our knowledge of the various stages of the development of the S-TGC lineage has been lacking [4]. Our data implicate Crim1 as an important regulator of the development of this lineage. Recent lineage tracing experiments based on TpbpaeCre activity, indicate that S-TGCs and GCs do not share an immediate, common precursor [4]. Therefore, it seems unlikely that the increase in GCs in Crim1KST264/KST264 placentae was directly related to the concurrent decrease in S-TGCs. Crim1 has been shown to bind cystine-knot growth factors, and tether them to the cell surface when co-expressed in the same cell
[13,14]. Furthermore, loss of Crim1 function in the podocytes of renal glomeruli results in dysregulation of VEGF and overactivation of VEGFR-2 in adjacent capillary endothelial cells [14]. Numerous cystine-knot growth factors are expressed in the developing placenta and have essential roles in the development or function of the placenta, including VEGF, PlGF, PDGF, BMPs and TGFbs [24e26]. Interestingly, Nodal, a TGFb superfamily member, is expressed in SpTs, and placentae from Nodal mutant mice displayed an expansion of giant cells and an overall expansion of the JZ, with a reduction in the LZ [28]. It may be possible that Nodal is acting upon SpTs in a cell-autonomous manner (in addition to other possible effects). Furthermore, the phenotypes of Nodal mutant placentae were evident from 10.5 dpc, earlier than those in Crim1KST264/KST264 placentae. We utilized the Crim1 reporter gene to monitor expression in this study. It remains a formal possibility, however, that the reporter may not mark all Crim1-expressing cells. There is evidence in the developing kidney there may be sites of Crim1 mRNA expression that are not detected by the b-galactosidase reporter: Crim1 mRNA but not X-Gal staining is detected in condensates of the nephrogenic zone and in distal comma-shaped bodies of the embryonic kidney. In other sites, such as the glomerular podocytes, Crim1 mRNA and b-galactosidase reporter are detected [15,29]. Importantly, no spurious b-galactosidase reporter activity has been reported. As strong Crim1 reporter gene expression was observed in the spongiotrophoblasts (SpT) rather than the GCs, we hypothesise that Crim1 is regulating a growth factor(s) produced by SpT required for GC growth, differentiation or survival. In the LZ, the deficiency in S-TGCs may be due to a change in growth factor regulation from adjacent syncytiotrophoblasts in an analogous manner. It is also possible that there is an autocrine or cellautonomous defect as some S-TGCs were observed to be X-Galþ in Crim1þ/KST264 13.5 dpc placenta. A further possibility is that, since S-TGC most likely arise from the base of the EPC, below the developing SpT layer, that Crim1 may be regulating growth factor
182
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
availability in the vicinity of S-TGC progenitors in the EPC, thereby affecting the differentiation of S-TGCs. As Crim1 is capable of binding numerous cystine-knot growth factors, the placental defects observed in Crim1KST264 mice may be due to perturbation of multiple growth factor signalling pathways. Development of a conditional mutant allele for Crim1 (and of suitable Creexpressing lines) will help elucidate the tissue-specific roles for Crim1 in placental development. In summary, the key findings of this study are: 1. Crim1KST264/KST264 mutant placentae display aberrant development with altered structure and reduced size. 2. Crim1 is strongly expressed in the spongiotrophoblast cells in the junctional zone. 3. Crim1KST264/KST264 placentae show an increase in the number of glycogen trophoblast cells in the junctional zone, and a decrease in the density of sinusoidal-trophoblast giant cells of the labyrinth zone. Acknowledgements We thank the staff of The University of Queensland Biological Resources animal facilities for support. This study conformed to the Institute’s Animal Ethics Committee guidelines for animal use in research. This work was funded by Project Grants from the National Health and Medical Research Council of Australia to DJP (grant number 631658) and MHL (grant numbers 301056 and 455972). Appendix. Supplementary material Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.placenta.2011.12.014. References [1] Cross JC. How to make a placenta: mechanisms of trophoblast cell differentiation in miceea review. Placenta 2005 Apr;26(Suppl. A):S3e9. [2] Cross JC, Nakano H, Natale DR, Simmons DG, Watson ED. Branching morphogenesis during development of placental villi. Differentiation 2006 Sep;74(7):393e401. [3] Simmons DG, Rawn S, Davies A, Hughes M, Cross JC. Spatial and temporal expression of the 23 murine prolactin/placental lactogen-related genes is not associated with their position in the locus. BMC Genomics 2008;9:352. [4] Simmons DG, Fortier AL, Cross JC. Diverse subtypes and developmental origins of trophoblast giant cells in the mouse placenta. Dev Biol 2007 Apr 15;304(2): 567e78. [5] Adamson SL, Lu Y, Whiteley KJ, Holmyard D, Hemberger M, Pfarrer C, et al. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev Biol 2002 Oct 15;250(2):358e73. [6] Riley P, Anson-Cartwright L, Cross JC. The Hand1 bHLH transcription factor is essential for placentation and cardiac morphogenesis. Nat Genet 1998 Mar; 18(3):271e5. [7] Coan PM, Conroy N, Burton GJ, Ferguson-Smith AC. Origin and characteristics of glycogen cells in the developing murine placenta. Dev Dyn 2006 Dec; 235(12):3280e94. [8] Guillemot F, Caspary T, Tilghman SM, Copeland NG, Gilbert DJ, Jenkins NA, et al. Genomic imprinting of Mash2, a mouse gene required for trophoblast development. Nat Genet 1995 Mar;9(3):235e42.
[9] Lescisin KR, Varmuza S, Rossant J. Isolation and characterization of a novel trophoblast-specific cDNA in the mouse. Genes Dev 1988 Dec;2(12A): 1639e46. [10] Georgiades P, Ferguson-Smith AC, Burton GJ. Comparative developmental anatomy of the murine and human definitive placentae. Placenta 2002 Jan; 23(1):3e19. [11] Bouillot S, Rampon C, Tillet E, Huber P. Tracing the glycogen cells with protocadherin 12 during mouse placenta development. Placenta 2006 Aug;27(8): 882e8. [12] Kolle G, Georgas K, Holmes GP, Little MH, Yamada T. CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis. Mech Dev 2000 Feb;90(2):181e93. [13] Wilkinson L, Kolle G, Wen D, Piper M, Scott J, Little M. CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. J Biol Chem 2003 Sep 5;278(36):34181e8. [14] Wilkinson L, Gilbert T, Kinna G, Ruta LA, Pennisi D, Kett M, et al. Crim1KST264/ KST264 mice implicate Crim1 in the regulation of vascular endothelial growth factor-a activity during glomerular vascular development. J Am Soc Nephrol 2007 Jun;18(6):1697e708. [15] Pennisi DJ, Wilkinson L, Kolle G, Sohaskey ML, Gillinder K, Piper MJ, et al. Crim1KST264/KST264 mice display a disruption of the Crim1 gene resulting in perinatal lethality with defects in multiple organ systems. Dev Dyn 2007 Feb; 236(2):502e11. [16] Wilhelm D, Hiramatsu R, Mizusaki H, Widjaja L, Combes AN, Kanai Y, et al. SOX9 regulates prostaglandin D synthase gene transcription in vivo to ensure testis development. J Biol Chem 2007 Apr 6;282(14):10553e60. [17] Chiu HS, Szucsik JC, Georgas KM, Jones JL, Rumballe BA, Tang D, et al. Comparative gene expression analysis of genital tubercle development reveals a putative appendicular Wnt7 network for the epidermal differentiation. Dev Biol 2010 Aug 15;344(2):1071e87. [18] Constância M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, et al. Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature 2002 Jun 27;417(6892):945e8. [19] Wu L, de Bruin A, Saavedra HI, Starovic M, Trimboli A, Yang Y, et al. Extraembryonic function of Rb is essential for embryonic development and viability. Nature 2003 Feb 27;421(6926):942e7. [20] Rampon C, Prandini MH, Bouillot S, Pointu H, Tillet E, Frank R, et al. Protocadherin 12 (VE-cadherin 2) is expressed in endothelial, trophoblast, and mesangial cells. Exp Cell Res 2005 Jan 1;302(1):48e60. [21] Ishida M, Ono K, Taguchi S, Ohashi S, Naito J, Horiguchi K, et al. Cathepsin gene expression in mouse placenta during the latter half of pregnancy. J Reprod Dev 2004 Oct;50(5):515e23. [22] Lee SJ, Talamantes F, Wilder E, Linzer DI, Nathans D. Trophoblastic giant cells of the mouse placenta as the site of proliferin synthesis. Endocrinology 1988 May;122(5):1761e8. [23] Duckworth ML, Schroedter IC, Friesen HG. Cellular localization of rat placental lactogen II and rat prolactin-like proteins A and B by in situ hybridization. Placenta 1990 MareApr;11(2):143e55. [24] Jones RL, Stoikos C, Findlay JK, Salamonsen LA. TGF-beta superfamily expression and actions in the endometrium and placenta. Reproduction 2006 Aug;132(2):217e32. [25] Ohlsson R, Falck P, Hellström M, Lindahl P, Boström H, Franklin G, et al. PDGFB regulates the development of the labyrinthine layer of the mouse fetal placenta. Dev Biol 1999 Aug 1;212(1):124e36. [26] Reynolds LP, Borowicz PP, Vonnahme KA, Johnson ML, Grazul-Bilska AT, Wallace JM, et al. Animal models of placental angiogenesis. Placenta 2005 Nov;26(10):689e708. [27] Cross JC, Baczyk D, Dobric N, Hemberger M, Hughes M, Simmons DG, et al. Genes, development and evolution of the placenta. Placenta 2003 FebeMar; 24(2e3):123e30. [28] Ma GT, Soloveva V, Tzeng SJ, Lowe LA, Pfendler KC, Iannaccone PM, et al. Nodal regulates trophoblast differentiation and placental development. Dev Biol 2001 Aug 1;236(1):124e35. [29] Georgas K, Bowles J, Yamada T, Koopman P, Little MH. Characterisation of Crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern. Dev Dyn 2000 Dec;219(4):582e7. [30] Leighton PA, Mitchell KJ, Goodrich LV, Lu X, Pinson K, Scherz P, et al. Defining brain wiring patterns and mechanisms through gene trapping in mice. Nature 2001 Mar 8;410(6825):174e9.
Contents lists available at SciVerse ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
Crim1 has an essential role in glycogen trophoblast cell and sinusoidaltrophoblast giant cell development in the placenta D.J. Pennisi a, b, *, G. Kinna a, c, H.S. Chiu a, b, D.G. Simmons b, L. Wilkinson a, M.H. Little a a
Institute for Molecular Bioscience, The University of Queensland, Brisbane 4072, Australia School of Biomedical Sciences, The University of Queensland, Chancellor’s Place, Brisbane 4072, Australia c Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane 4072, Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 16 December 2011
Normal placental development and function is essential for fetal growth of eutherian mammals. Mutational studies have shown that numerous growth factors are required for placental development and differentiation of placental lineages. Here, using a gene-trap mutant mouse line, Crim1KST264, we show that Crim1 is essential for murine placental development. Crim1 is a developmentally expressed, trans-membrane regulator of growth factor activity. Crim1KST264/KST264 mutant placentae displayed hypoplasia from 13.5 dpc, and altered structure from 15.5 dpc, including alterations in cell number in both the junctional and labyrinth zones. Using the reporter gene from the Crim1KST264 allele, we found that Crim1 is expressed in multiple cell types of the placenta, including strong expression in the spongiotrophoblast cells of the junctional zone. In the junctional zone of Crim1KST264/KST264 placentae, there was an increase in the glycogen trophoblast cells adjacent to the spongiotrophoblast cells. In the labyrinth zone, we found a decrease in the density of sinusoidal-trophoblast giant cells. Our findings show that Crim1 is required for placental development, and is necessary for the proper differentiation of sinusoidal-trophoblast giant cells and glycogen trophoblast cells. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Crim1 Spongiotrophoblast Glycogen trophoblast cells Lineage development
1. Introduction The placenta is an organ essential for normal embryonic development and survival in utero. In the mouse, placental development begins soon after implantation at around 4.5 days post coitum (dpc) when a population of trophoblast stem cells emerge from the polar trophectoderm overlying the inner cell mass (ICM) of the blastocyst. Concurrently, the trophectoderm not in close contact with the ICM, the mural trophectoderm, undergoes differentiation into the first wave of trophoblast giant cells (TGC); a large polyploid cell type involved in early invasion, endocrine signalling and formation of the yolk sac [1]. The continued growth of the polar trophectoderm produces two important structures of the early placenta, the ectoplacental cone (EPC) and the chorion. The EPC contains progenitors for three trophoblast lineages: a second wave of TGCs that form a layer adjacent to the maternal decidua, and spongiotrophoblast and glycogen trophoblast cells of the middle
layer of the placenta; together these three cell types comprise the junctional zone (JZ). In contrast, the chorion contains progenitors for the syncytiotrophoblast cells of the labyrinth zone (LZ), the site of foetal-maternal exchange. Soon after chorio-allantoic fusion at around 8.0e8.5 dpc, structural changes appear in the chorion trophoblast cells, and the chorion begins to develop folds that extend into primary branches in the nascent LZ [2]. Within the LZ, the allantoic mesoderm gives rise to mesenchyme and blood vessels, whereas the trophoblast cells give rise to several different epithelial derivatives. By early 9.0 dpc, primary villous branches forming within the developing labyrinth are apparent and are lined by thin, elongated trophoblast cells [2]. These primary branches then undergo secondary branching and elongation to form the mature LZ. Ultimately, the labyrinthine trophoblast cells form three layers that make up the placental membrane separating maternal blood from the foetal vasculature [3]. 1.1. Trophoblast giant cells
* Corresponding author. School of Biomedical Sciences, The University of Queensland, Chancellor’s Place, Brisbane 4072, Australia. Tel.: þ61 7 3365 4656; fax: þ61 7 3365 1766. E-mail address: [email protected] (D.J. Pennisi). 0143-4004/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2011.12.014
While all TGCs are large polyploid cells with an established endocrine activity, the population consists of several unique subtypes that differ somewhat in the extent of their ploidy, gene expression profile, location, timing of appearance and
176
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
Fig. 1. The gross morphology and histology of Crim1KST264/KST264 placentae. AeC: Micrographs of 15.5 dpc placentae and embryos with intact yolk sac viewed in whole-mount. Arrowheads indicate the extent of the junctional zone (AeC). Note the exencephaly in the Crim1KST264/KST264 embryo (arrow, C). DeF: Micrographs of 15.5 dpc placentae after dissection of the embryo and yolk sac, viewed in whole-mount. Note the diminished size of the Crim1KST264/KST264 placenta relative to littermate controls. GeI: Micrographs of 17.5 dpc placentae and embryos viewed in whole-mount. Arrowheads indicate the extent of the junctional zone (GeI). Note, that in the Crim1KST264/KST264 placenta (I), the junctional zone is less prominent in proximal view, consistent with a loss of cells in this region. JeL: Micrographs of representative mid-sagittal, haematoxylin and eosin-stained sections of
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
developmental origins [4]. Parietal TGCs (P-TGCs) are the TGCs that line the implantation chamber and separate underlying placental layers from the maternal decidua; this population is comprised of both primary TGCs, the first differentiated cell type derived from the mural trophectoderm of the blastocyst that contribute largely to the yolk sac, and secondary TGCs derived from the EPC. TGCs that migrate into the decidua usually invade only a short distance, however, a subgroup invades deeper into the spiral arteries to establish the connection between the endothelial cell-lined arteries to the hemochorial blood spaces found in rodent and primate placentas [5]. These endovascular TGCs (spiral artery-associated TGCs, or SpA-TGCs) express the proliferin gene, but not the placental lactogens commonly expressed by the other TGC subtypes [4]. When the spiral arteries of the rodent placenta reach the level of the JZ, they coalesce into several large canals that bring blood to the base of the placenta where it empties into smaller capillary sized sinusoids in the labyrinth, bringing maternal blood into close association with foetal capillaries. Unique TGC subtypes also line these canals (C-TGCs) and sinusoids (S-TGCs) [4]. Previously, all secondary TGCs were thought to pass through a Tpbpa-expressing, intermediate stage in the EPC upon differentiation, however, more recent cell lineage tracing studies have indicated that P-TGCs and CTGCs can be derived from both Tpbpa-positive and Tpbpa-negative EPC progenitors [4]. Conversely, SpA-TGCs appear to be derived from Tpbpa-positive cells while S-TGCs arise exclusively from Tpbpa-negative progenitors. Despite these different developmental origins, TGCs do have universal aspects to their differentiation; all TGCs appear to require the transcription factor Hand1 for proper differentiation [6]. 1.2. Spongiotrophoblast and glycogen trophoblast cells The middle layer of the rodent placenta contains spongiotrophoblast (SpT) intermingled with islets of glycogen trophoblast cells (GCs). In the second half of gestation, some GCs leave the confines of this layer and invade interstitially into the decidua, congregating around the area of spiral arteries [7]; spongiotrophoblast do not appear to be motile and remain within the JZ of the placenta. The function of this cell layer remains incompletely understood, although it is known to secrete considerable amounts of peptide hormones and proteases [1] and is critically dependent on the transcription factor Mash2 for its formation [8]. Expression of Tpbpa is characteristic of both SpT and GCs [5,9] and is often used as a marker for these cell populations. The early identification of GCs has traditionally been done using PAS staining to localize glycogen content, prompting the initial idea that GCs were derived from SpT sometime after E11.5 [5,10], coinciding with the detection of significant PAS staining in the developing placenta. However, recent studies have suggested that GCs may be derived from specific, Pcdh12-expressing progenitors present within the EPC as early as E7.5, although this notion awaits confirmation from cell lineage tracing studies [11]. 1.3. Crim1
177
cells cystine-knot growth factors, including BMP4, BMP7, VEGF-A, PDGF-B, Nodal, and TGFb2 [13,14]. Moreover, loss of Crim1 function can affect the availability of growth factors to neighbouring cells [14]. We have previously described the pleiotropic phenotypes associated with Crim1KST264 mouse embryos that harbour a genetrap mutation [15]. Mice homozygous for the gene-trap mutation (Crim1KST264/KST264) on an inbred, C57Bl6 genetic background die perinatally, and display phenotypes including renal dysplasia, microphthalmia, reduced lens size, digit syndactyly, dermal blebbing, and cerebral haematoma [15]. Herein, we show that Crim1 is expressed in multiple cell types of the placenta and reveal essential roles for Crim1 in murine placental development using the Crim1KST264 gene-trap mouse line. Specifically, we observed changes in the relative proportions of S-TGCs and GCs Crim1KST264/KST264 placentae, demonstrating that Crim1 is necessary for the proper differentiation of both S-TGC and GC lineages. 2. Materials and methods 2.1. Maintenance of the gene-trap mouse line The Crim1KST264 gene-trap mutation has been described [15] and was maintained on a C57Bl6 background. Use of animals in this study was done in accordance with the Animal Ethics Committee, The University of Queensland (IMB/565/07/ NHMRC, IMB/142/09/NHMRC). Genotyping was performed as described [15]. 2.2. Sample preparation and X-Gal staining Whole embryos and placentae were dissected in phosphate-buffered saline (PBS), fixed in 4% paraformaldehyde (PFA) in PBS at 4 C for 2e4 h, dehydrated in an ethanol series, cleared in xylene, and processed for paraffin infusion and embedding. Serial, 7 mm, sections were cut and mounted on Superfrost slides. For X-Gal staining on sections, tissue was fixed in 4% PFA at 4 C, washed in PBS and incubated in 30% sucrose/PBS overnight at 4 C. Samples were then embedded in OCT medium, and 10 mm sections cut on a cryostat. Slides were air dried for 10e15 mins before washing in PBS. X-Gal buffer solution was added to slides and left in the dark at room temperature to develop for 48e72 h. Slides were washed in PBS, dehydrated in an ethanol series, cleared, and mounted. Some slides were counterstained with nuclear fast red (NFR; Vector Laboratories, California). Staining was never detected in wildtype tissue (data not shown). 2.3. Histochemistry and section in situ hybridization Masson’s trichrome, periodic acid-Schiff (PAS), and haematoxylin and eosin staining were performed using standard procedures. In situ hybridization was performed on PFA-fixed, paraffin-embedded placentae according to published methods [16]. Cathepsin Q (Ctsq), Placental Lactogen II (PLII; Prl3b1), and Protocadherin12 (Pcdh12) anti-sense RNA probe production has been previously described [3]. Trophoblast-specific protein alpha (Tpbpa) anti-sense RNA probe was produced according to published methods [17]. The Tpbpa-specific primers used were; Forward Primer, 50 -TGAAGAGCTGAACCACTGGA-30 ; Reverse Primer, 50 -CGATGTTAATACGACTCACTATAGGGAGTGCAGGATCCCACTTGTC-30 . 2.4. Data documentation Digital images of tissue sections were captured using either a semi-automated “slide” System from Olympus and Soft Imaging Systems, or an Olympus BX-51 BF/ DF slide microscope with Canon digital cameras, and with “OlyVIA” software (Soft Imaging Systems, Olympus) and DP Controller software (Olympus, Japan), respectively. Whole-mount images were captured using an Olympus SZX-12 stereomicroscope with DP Controller software. Images were adjusted for colour levels, brightness and contrast, and figures compiled, using Adobe Photoshop software. 2.5. Data quantification and statistical analyses
Crim1 is a multi-domain protein that is dynamically expressed during embryogenesis and contains six von Willebrand Factor-C (vWFC)-like cysteine-rich repeat domains similar to the BMPregulating protein, Chordin [12,13]. The six vWFC domains mediate the ability of Crim1 to bind and tether to the surface of
To determine weights, between nine and fourteen placentae and embryos of each genotype, from at least four different litters, were collected and weighed. For quantification of S-TGC density, sixty high magnification (100) images from at least three biological samples were done for each genotype analysed. Quantification of area of PAS staining, and Pcdh12 expression was performed on at least four sections
15.5 dpc placentae. The junctional zone (JZ) and labyrinth zone (LZ) are indicated. MeO: Higher magnification views of the boxed areas in JeL, respectively. Glycogen trophoblast cells (GC) and spongiotrophoblast cells (SpT) are indicated. Examples of Crim1þ/þ (A, D, G, J, and M), Crim1þ/KST264 (B, E, H, K and N) and Crim1KST264/KST264 (C, F, I, L and O) embryos and placentae are shown. Scale bar; F, 1 mm; O, 100 mm.
178
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
from each sample, and at least four samples from each genotype. Total placental area, and area of staining, were quantified with ImageJ software (NIH; http://rsb. info.nih.gov/ij/). For determining the statistical significance of placental and embryonic weight differences, a one-way unpaired ANOVA with Tukey’s multiple comparison tests was used to compare genotypes. For determining the statistical significance of S-TGC density, the area of PAS staining, and the area of Pcdh12 expression, an unpaired, two-tailed student’s t-test, with unequal variance with Welch’s correction was used to compare genotypes.
3. Results The first set of experiments sought to determine whether there were placental defects in Crim1KST264/KST264 mice. At 15.5 dpc, we found that Crim1KST264/KST264 placentae, when viewed from the proximal side, exhibited a less prominent JZ relative to littermate controls (Fig. 1AeC). Moreover, Crim1KST264/KST264 placentae appeared smaller than those of littermate controls in whole-mount (Fig. 1DeF, Supplemental Fig. 1). At 17.5 dpc, the gross morphological defect in Crim1KST264/KST264 placentae became more prominent, with the junctional zone being less prominent in proximal view (Fig. 1GeI). Defects in Crim1KST264/KST264 embryos can be observed from 11.5 dpc (peridermal blebbing [15]). However, no difference was observed between Crim1KST264/KST264 placentae and those of littermate controls based on gross morphology and histological sections at 11.5 dpc (data not shown). Haematoxylin and eosin-stained sections demonstrated that at 15.5 dpc, Crim1KST264/KST264 placentae had abnormal morphology compared to littermate controls (Fig. 1JeL). Histological analyses also indicated that presumptive glycogen trophoblast cells and spongiotrophoblast cells were present in the LZ (Fig. 1MeO). As the placentae of Crim1KST264/KST264 embryos appeared smaller than those of littermate controls, we weighed placentae and embryos. We observed that of Crim1KST264/KST264 placentae were significantly lighter than those of Crim1þ/KST264 littermate controls at 13.5 dpc (P < 0.05), 15.5 dpc (P < 0.05), and 17.5 dpc (P < 0.01; Fig. 2A). As impaired placental function may result in reduced embryo growth [18,19], we quantified embryo weights. At 13.5 dpc and 15.5 dpc, there was not a significant difference between Crim1þ/ KST264 and Crim1KST264/KST264 embryos. At 17.5 dpc, however, we observed that Crim1KST264/KST264 embryos were significantly lighter than Crim1þ/KST264 littermate controls (P < 0.05; Fig. 2B). We did not find a significant difference between Crim1þ/þ and Crim1þ/KST264 in weights of placentae or embryos at the stages examined (not shown). Although there was an observable difference in the weights of Crim1þ/KST264 and Crim1KST264/KST264 placentae, examination of BrdU uptake at 15.5 dpc did not reveal a marked alteration in the pattern of BrdUþ cells in Crim1KST264/KST264 placentae (Supplemental Fig. 2).
We have previously reported that Crim1 is expressed in the parietal trophoblast giant cells (P-TGCs) of the placenta at 17.5 dpc [15]. As Crim1KST264/KST264 placentae exhibited defects, we took advantage of the b-Geo reporter construct in the gene-trap line to further characterize Crim1 expression in the developing placenta. XGal staining on cryosections revealed b-Geo expression in decidual, JZ and LZ cells of Crim1þ/KST264 and Crim1KST264/KST264 placentae. At 15.5 dpc, strong expression was observed in Crim1þ/KST264 and Crim1KST264/KST264 placentae in a discontinuous pattern in the JZ (Fig. 3A and B). Higher magnification images reveal that most of the X-Galþ cells in the JZ were presumptive SpT cells, with little staining evident in GCs (Fig. 3C and D). X-Gal staining was observed in the SpT cells from 13.5 dpc in Crim1þ/KST264 placentae (Supplemental Fig. 3). At 9.5 dpc, X-Gal staining was observed amongst the allantoic mesenchyme and at low levels in some chorionic trophoblasts of Crim1þ/KST264 placentae (Supplemental Fig. 3). At 13.5 dpc and 15.5 dpc, X-Gal staining was observed in canal trophoblasts and at lower levels in some syncytiotrophoblasts; in some perivascular cells in the LZ of Crim1þ/KST264 placentae, presumably pericytes (Supplemental Fig. 3). At 15.5 dpc, some spiral artery-associated TGCs, and a few S-TGCs were X-Galþ (Supplemental Fig. 3). As Crim1KST264/KST264 embryos had abnormal placental morphology, and Crim1 was expressed in SpT cells of the JZ, we examined the morphology of the JZ in Crim1KST264/KST264 placentae in a manner independent of the Crim1 gene-trap reporter. To that end, section in situ hybridization for Trophoblast-specific protein alpha (Tpbpa) expression was performed on 15.5 dpc placenta. Robust expression of Tpbpa was found in the JZ of control and Crim1KST264/KST264 placentae, particularly amongst spongiotrophoblast cells (Fig. 4). Based on Tpbpa expression, the JZ in Crim1KST264/ KST264 placentae is prominent, but does not extend to the proximalmost part of the placenta (Fig. 4C and F). Thus, the histological and section in situ hybridization data highlighting the JZ in Crim1KST264/ KST264 placentae corroborate morphological observations made in whole-mount views (Fig. 1). As there was an alteration in the JZ of Crim1KST264/KST264 placentae, we sought to determine whether there was a change in the number of GCs in mutant placentae. We assessed periodic acid-Schiff (PAS) staining in Crim1KST264/KST264 placentae at 15.5 dpc, to identify GCs. We did not find a significant difference in the area of PAS staining in placentae at 15.5 dpc in Crim1KST264/KST264 relative to littermate controls (Fig. 5AeC). As PAS staining will detect mucin and basement membrane components, in addition to glycogen, we utilised the expression of Protocadherin12 (Pcdh12), a recently described molecular marker of GCs [11,20]. Therefore, to more accurately quantify the amount of GCs in Crim1KST264/KST264 placentae, we
Fig. 2. Crim1KST264/KST264 placentae and embryos are smaller than littermate controls. A. The weights of Crim1KST264/KST264 placentae were significantly smaller than littermate controls at 13.5 dpc, 15.5 dpc, and 17.5 dpc. B. The weights of Crim1þ/KST264 and Crim1KST264/KST264 embryos were similar at 13.5 dpc and 15.5 dpc, but by 17.5 dpc, Crim1KST264/KST264 embryos were significantly smaller. The weights of Crim1þ/þ and Crim1þ/KST264 embryos and placentae were not significantly different at any stage examined (data not shown). Red squares, Crim1þ/KST264 values; blue triangles, Crim1KST264/KST264 values. Error bars, SDM. *, P < 0.05; **, P < 0.01.
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
179
Fig. 3. The Crim1 gene-trap reporter is strongly expressed in the junctional zone of the developing placenta. A, B: Micrographs of representative mid-sagittal, X-Gal-stained sections of a 15.5dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae sections. C, D: Higher magnification images of representative mid-sagittal, X-Gal-stained sections that have been counterstained with nuclear fast red of 15.5 dpc Crim1þ/KST264 (C) and Crim1KST264/KST264 (D) placentae. Note the strong reporter gene expression in the spongiotrophoblasts. Note also the punctate, sub-cellular localization of X-Gal staining (C, D), reflecting the localization of the b-Geo reporter of this gene-trap construct [30]. Glycogen trophoblast cells (GC), spongiotrophoblast cells (SpT), junctional zone (JZ), and labyrinth zone (LZ) are indicated. Scale bars; B, 1 mm; D, 100 mm.
Fig. 4. The junctional zone has an altered morphology in Crim1KST264/KST264 placentae at 15.5 dpc. AeF: Micrographs of section in situ hybridization for Tpbpa expression in 15.5 dpc Crim1þ/KST264 (A, C, and D) and Crim1KST264/KST264 (B, E, and F) placentae. Representative mid-sagittal sections are shown. C, D and E, F: Higher magnification views of the indicated boxed areas in A and B, respectively. Note the strong Tpbpa expression in the spongiotrophoblast cells and lower levels of expression in glycogen trophoblast cells in Crim1þ/KST264 and Crim1KST264/KST264 placentae. Glycogen trophoblast cells (GC), spongiotrophoblast cells (SpT), junctional zone (JZ), and labyrinth zone (LZ) are indicated. Scale bars; A and B, 1.2 mm; C and E, 500 mm; D and F, 200 mm.
180
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
Fig. 5. There are more glycogen trophoblast cells (GCs) in Crim1KST264/KST264 placentae at 15.5 dpc. A, B: Micrographs of representative mid-sagittal, PAS-stained sections of 15.5 dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae. A0 , B0 : Higher magnification views of the boxed areas in A and B, respectively. C. Quantification of PAS staining in Crim1þ/KST264 and Crim1KST264/KST264 placentae. The difference in the fraction area stained PAS-positive was not significant (ns). D, E: Micrographs of section in situ hybridization for Pcdh12 expression in 15.5 dpc Crim1þ/KST264 (D) and Crim1KST264/KST264 (E) placentae. Representative mid-sagittal sections are shown. D0 , E0 : Higher magnification views of the boxed areas in D and E, respectively. F. Quantification of Pcdh12 expression in Crim1þ/KST264 and Crim1KST264/KST264 placentae. There was a significant increase in the fraction area stained positive for Pcdh12 expression in Crim1KST264/KST264 placentae (P < 0.005). Scale bars; B, E, 500 mm.
performed in situ hybridization to detect Pcdh12 transcripts. In littermate control 15.5 dpc placentae, we found that 4.4% (SDM, 1.9%) of placenta area stained positively for Pcdh12 expression. In contrast, we observed 7.6% (SDM, 0.9%) in Crim1KST264/KST264 placentae, a significant increase relative to littermate controls (P < 0.005; Fig. 5DeF). Note that these values for Pcdh12 expression (ranging from 4.4% to 7.6%) are lower than those obtained for PAS staining (ranging from 14.6% to 19.1%) and, presumably, reflect the more selective nature of the Pcdh12 expression. Although the strongest expression of the Crim1 gene-trap reporter was in the SpT cells of the JZ, lesser expression was also observed in some cells of the LZ. We therefore examined smooth muscle alpha actin (SMaA) protein, CD31 protein and Mest mRNA, markers of some LZ cell types. At 15.5 dpc, we did not observe a change in SMaA pattern or density in the LZ of Crim1KST264/KST264 placentae (Supplemental Fig. 4). We did, however, observe an increase in the fetal endothelial expression of Mest mRNA and a decrease in CD31 protein in Crim1KST264/KST264 placentae (Supplemental Fig. 5). As Crim1 was expressed in a dynamic manner in the LZ, we also sought to determine whether a lineage of the LZ, the S-TGCs, was affected in Crim1KST264/KST264 placentae. Therefore, we quantified the number of S-TGCs after Masson’s Trichrome staining on 15.5 dpc placentae (Fig. 6AeC). S-TGCs have large round nuclei and
are easily distinguished from other cell types such the fetal endothelium (FE) or the syncytium. We found 14.5 105 S-TGC/mm2 (SDM, 0.6 105) in control placentae. However, in Crim1KST264/ KST264 placentae, there was a significant reduction in the number, with 11.2 105 S-TGC/mm2 (SDM, 0.7 105) (P < 0.0005; Fig. 6). To confirm there was a reduction in the number of S-TGCs in Crim1KST264/KST264 placentae, we examined the expression of the S-TGC marker genes, Cathepsin Q (Ctsq) and Placental Lactogen II (PLII) [21e23] by in situ hybridisation. Relative to littermate control placentae, the LZ of Crim1KST264/KST264 placentae showed fewer cells that expressed Ctsq or PLII (Fig. 6DeG), providing qualitative data supporting the quantification of S-TGCs in Crim1KST264/KST264 placentae. The data demonstrate that in Crim1KST264/KST264 placentae, there is a reduction in the number of S-TGCs and a concurrent increase in GCs. 4. Discussion Numerous growth factors have been implicated in placental development and function, including members of the VEGF/PDGF family, TGFb superfamily, Wnts, and FGFs [24e27]. Using a genetrap mutant mouse line, we have found that Crim1, a transmembrane regulator of growth factor activity, is essential for normal placental development. The placental defects included
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
181
Fig. 6. There are less sinusoidal-trophoblast giant cells (S-TGCs) in Crim1KST264/KST264 placentae at 15.5 dpc. A, B: Micrographs of representative mid-sagittal, Masson’s trichrome-stained sections of the labyrinth zone of 15.5 dpc Crim1þ/KST264 (A) and Crim1KST264/KST264 (B) placentae. S-TGCs are characterized by large nuclei (open arrowheads). Fetal blood space (FBS), fetal endothelium (FE; arrows), and maternal blood space (MBS; arrowheads) are indicated. C. Quantification of the density of S-TGCs in Crim1þ/KST264 and Crim1KST264/KST264 placentae. D, E: Micrographs of section in situ hybridization for Ctsq expression in the labyrinth zone of 15.5 dpc Crim1þ/KST264 (D) and Crim1KST264/KST264 (E) placentae. F, G: Micrographs of section in situ hybridization for PLII expression in the labyrinth zone of 15.5 dpc Crim1þ/KST264 (F) and Crim1KST264/KST264 (G) placentae. Note the reduction in the number of cells positive for Ctsq or PLII expression in the S-TGCs in Crim1KST264/KST264 LZ. S-TGCs are indicated (open arrowheads). Representative mid-sagittal sections are shown. Scale bars, 50 mm.
a reduction in size relative to wild-type controls, morphological changes, as well as changes in the proportion of the placenta’s constituent cell types. The earliest detected placental phenotype was the reduction in size at 13.5 dpc. This was followed by a reduction in the size of Crim1KST264/KST264 embryos at 17.5 dpc. It remains to be determined whether any potential dysgenesis or insufficiency in Crim1KST264/KST264 placentae contributed to the reduced size observed in Crim1KST264/KST264 embryos late in gestation. In this study, we used two markers to identify GCs in histological sections, PAS staining and Pcdh12 in situ hybridization. In addition to the substantial glycogen in GCs, PAS staining will detect mucin and basement membrane components. This could lead to an overestimation of GC content in sections of placentae using unbiased staining and morphometry techniques. Using Pcdh12 in situ hybridization to quantify GC content gave lower values than did the use of PAS staining in all genotypes. Furthermore, high magnification images of Pcdh12 in situ hybridization revealed strong signal in GCs with no background staining in adjacent SpTs. Together, these observations further support the use of Pcdh12 as a useful marker of GCs as suggested by Bouillot et al. [11]. Loss of Crim1 function resulted in an expansion of GCs in the JZ, and a decrease in S-TGCs of the LZ of the placenta. Although the syncytiotrophoblast and S-TGC lineage are derived from Tpbpanegative cells of the chorion and inner ectoplacental cone, our knowledge of the various stages of the development of the S-TGC lineage has been lacking [4]. Our data implicate Crim1 as an important regulator of the development of this lineage. Recent lineage tracing experiments based on TpbpaeCre activity, indicate that S-TGCs and GCs do not share an immediate, common precursor [4]. Therefore, it seems unlikely that the increase in GCs in Crim1KST264/KST264 placentae was directly related to the concurrent decrease in S-TGCs. Crim1 has been shown to bind cystine-knot growth factors, and tether them to the cell surface when co-expressed in the same cell
[13,14]. Furthermore, loss of Crim1 function in the podocytes of renal glomeruli results in dysregulation of VEGF and overactivation of VEGFR-2 in adjacent capillary endothelial cells [14]. Numerous cystine-knot growth factors are expressed in the developing placenta and have essential roles in the development or function of the placenta, including VEGF, PlGF, PDGF, BMPs and TGFbs [24e26]. Interestingly, Nodal, a TGFb superfamily member, is expressed in SpTs, and placentae from Nodal mutant mice displayed an expansion of giant cells and an overall expansion of the JZ, with a reduction in the LZ [28]. It may be possible that Nodal is acting upon SpTs in a cell-autonomous manner (in addition to other possible effects). Furthermore, the phenotypes of Nodal mutant placentae were evident from 10.5 dpc, earlier than those in Crim1KST264/KST264 placentae. We utilized the Crim1 reporter gene to monitor expression in this study. It remains a formal possibility, however, that the reporter may not mark all Crim1-expressing cells. There is evidence in the developing kidney there may be sites of Crim1 mRNA expression that are not detected by the b-galactosidase reporter: Crim1 mRNA but not X-Gal staining is detected in condensates of the nephrogenic zone and in distal comma-shaped bodies of the embryonic kidney. In other sites, such as the glomerular podocytes, Crim1 mRNA and b-galactosidase reporter are detected [15,29]. Importantly, no spurious b-galactosidase reporter activity has been reported. As strong Crim1 reporter gene expression was observed in the spongiotrophoblasts (SpT) rather than the GCs, we hypothesise that Crim1 is regulating a growth factor(s) produced by SpT required for GC growth, differentiation or survival. In the LZ, the deficiency in S-TGCs may be due to a change in growth factor regulation from adjacent syncytiotrophoblasts in an analogous manner. It is also possible that there is an autocrine or cellautonomous defect as some S-TGCs were observed to be X-Galþ in Crim1þ/KST264 13.5 dpc placenta. A further possibility is that, since S-TGC most likely arise from the base of the EPC, below the developing SpT layer, that Crim1 may be regulating growth factor
182
D.J. Pennisi et al. / Placenta 33 (2012) 175e182
availability in the vicinity of S-TGC progenitors in the EPC, thereby affecting the differentiation of S-TGCs. As Crim1 is capable of binding numerous cystine-knot growth factors, the placental defects observed in Crim1KST264 mice may be due to perturbation of multiple growth factor signalling pathways. Development of a conditional mutant allele for Crim1 (and of suitable Creexpressing lines) will help elucidate the tissue-specific roles for Crim1 in placental development. In summary, the key findings of this study are: 1. Crim1KST264/KST264 mutant placentae display aberrant development with altered structure and reduced size. 2. Crim1 is strongly expressed in the spongiotrophoblast cells in the junctional zone. 3. Crim1KST264/KST264 placentae show an increase in the number of glycogen trophoblast cells in the junctional zone, and a decrease in the density of sinusoidal-trophoblast giant cells of the labyrinth zone. Acknowledgements We thank the staff of The University of Queensland Biological Resources animal facilities for support. This study conformed to the Institute’s Animal Ethics Committee guidelines for animal use in research. This work was funded by Project Grants from the National Health and Medical Research Council of Australia to DJP (grant number 631658) and MHL (grant numbers 301056 and 455972). Appendix. Supplementary material Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.placenta.2011.12.014. References [1] Cross JC. How to make a placenta: mechanisms of trophoblast cell differentiation in miceea review. Placenta 2005 Apr;26(Suppl. A):S3e9. [2] Cross JC, Nakano H, Natale DR, Simmons DG, Watson ED. Branching morphogenesis during development of placental villi. Differentiation 2006 Sep;74(7):393e401. [3] Simmons DG, Rawn S, Davies A, Hughes M, Cross JC. Spatial and temporal expression of the 23 murine prolactin/placental lactogen-related genes is not associated with their position in the locus. BMC Genomics 2008;9:352. [4] Simmons DG, Fortier AL, Cross JC. Diverse subtypes and developmental origins of trophoblast giant cells in the mouse placenta. Dev Biol 2007 Apr 15;304(2): 567e78. [5] Adamson SL, Lu Y, Whiteley KJ, Holmyard D, Hemberger M, Pfarrer C, et al. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev Biol 2002 Oct 15;250(2):358e73. [6] Riley P, Anson-Cartwright L, Cross JC. The Hand1 bHLH transcription factor is essential for placentation and cardiac morphogenesis. Nat Genet 1998 Mar; 18(3):271e5. [7] Coan PM, Conroy N, Burton GJ, Ferguson-Smith AC. Origin and characteristics of glycogen cells in the developing murine placenta. Dev Dyn 2006 Dec; 235(12):3280e94. [8] Guillemot F, Caspary T, Tilghman SM, Copeland NG, Gilbert DJ, Jenkins NA, et al. Genomic imprinting of Mash2, a mouse gene required for trophoblast development. Nat Genet 1995 Mar;9(3):235e42.
[9] Lescisin KR, Varmuza S, Rossant J. Isolation and characterization of a novel trophoblast-specific cDNA in the mouse. Genes Dev 1988 Dec;2(12A): 1639e46. [10] Georgiades P, Ferguson-Smith AC, Burton GJ. Comparative developmental anatomy of the murine and human definitive placentae. Placenta 2002 Jan; 23(1):3e19. [11] Bouillot S, Rampon C, Tillet E, Huber P. Tracing the glycogen cells with protocadherin 12 during mouse placenta development. Placenta 2006 Aug;27(8): 882e8. [12] Kolle G, Georgas K, Holmes GP, Little MH, Yamada T. CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis. Mech Dev 2000 Feb;90(2):181e93. [13] Wilkinson L, Kolle G, Wen D, Piper M, Scott J, Little M. CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. J Biol Chem 2003 Sep 5;278(36):34181e8. [14] Wilkinson L, Gilbert T, Kinna G, Ruta LA, Pennisi D, Kett M, et al. Crim1KST264/ KST264 mice implicate Crim1 in the regulation of vascular endothelial growth factor-a activity during glomerular vascular development. J Am Soc Nephrol 2007 Jun;18(6):1697e708. [15] Pennisi DJ, Wilkinson L, Kolle G, Sohaskey ML, Gillinder K, Piper MJ, et al. Crim1KST264/KST264 mice display a disruption of the Crim1 gene resulting in perinatal lethality with defects in multiple organ systems. Dev Dyn 2007 Feb; 236(2):502e11. [16] Wilhelm D, Hiramatsu R, Mizusaki H, Widjaja L, Combes AN, Kanai Y, et al. SOX9 regulates prostaglandin D synthase gene transcription in vivo to ensure testis development. J Biol Chem 2007 Apr 6;282(14):10553e60. [17] Chiu HS, Szucsik JC, Georgas KM, Jones JL, Rumballe BA, Tang D, et al. Comparative gene expression analysis of genital tubercle development reveals a putative appendicular Wnt7 network for the epidermal differentiation. Dev Biol 2010 Aug 15;344(2):1071e87. [18] Constância M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, et al. Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature 2002 Jun 27;417(6892):945e8. [19] Wu L, de Bruin A, Saavedra HI, Starovic M, Trimboli A, Yang Y, et al. Extraembryonic function of Rb is essential for embryonic development and viability. Nature 2003 Feb 27;421(6926):942e7. [20] Rampon C, Prandini MH, Bouillot S, Pointu H, Tillet E, Frank R, et al. Protocadherin 12 (VE-cadherin 2) is expressed in endothelial, trophoblast, and mesangial cells. Exp Cell Res 2005 Jan 1;302(1):48e60. [21] Ishida M, Ono K, Taguchi S, Ohashi S, Naito J, Horiguchi K, et al. Cathepsin gene expression in mouse placenta during the latter half of pregnancy. J Reprod Dev 2004 Oct;50(5):515e23. [22] Lee SJ, Talamantes F, Wilder E, Linzer DI, Nathans D. Trophoblastic giant cells of the mouse placenta as the site of proliferin synthesis. Endocrinology 1988 May;122(5):1761e8. [23] Duckworth ML, Schroedter IC, Friesen HG. Cellular localization of rat placental lactogen II and rat prolactin-like proteins A and B by in situ hybridization. Placenta 1990 MareApr;11(2):143e55. [24] Jones RL, Stoikos C, Findlay JK, Salamonsen LA. TGF-beta superfamily expression and actions in the endometrium and placenta. Reproduction 2006 Aug;132(2):217e32. [25] Ohlsson R, Falck P, Hellström M, Lindahl P, Boström H, Franklin G, et al. PDGFB regulates the development of the labyrinthine layer of the mouse fetal placenta. Dev Biol 1999 Aug 1;212(1):124e36. [26] Reynolds LP, Borowicz PP, Vonnahme KA, Johnson ML, Grazul-Bilska AT, Wallace JM, et al. Animal models of placental angiogenesis. Placenta 2005 Nov;26(10):689e708. [27] Cross JC, Baczyk D, Dobric N, Hemberger M, Hughes M, Simmons DG, et al. Genes, development and evolution of the placenta. Placenta 2003 FebeMar; 24(2e3):123e30. [28] Ma GT, Soloveva V, Tzeng SJ, Lowe LA, Pfendler KC, Iannaccone PM, et al. Nodal regulates trophoblast differentiation and placental development. Dev Biol 2001 Aug 1;236(1):124e35. [29] Georgas K, Bowles J, Yamada T, Koopman P, Little MH. Characterisation of Crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern. Dev Dyn 2000 Dec;219(4):582e7. [30] Leighton PA, Mitchell KJ, Goodrich LV, Lu X, Pinson K, Scherz P, et al. Defining brain wiring patterns and mechanisms through gene trapping in mice. Nature 2001 Mar 8;410(6825):174e9.