Cited2 is required in trophoblasts for correct placental capillary patterning.

Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Developmental Biology journal homepage: www.elsevier.com/locate/developmentalbiology

Cited2 is required in trophoblasts for correct placental capillary patterning Julie L.M. Moreau a,1, Stanley T. Artap b,1,2, Hongjun Shi a, Gavin Chapman a,b, Gustavo Leone c, Duncan B. Sparrow a,b,n,3, Sally L. Dunwoodie a,b,d,n,3 a

Developmental and Stem Cell Biology Division, Victor Chang Cardiac Research Institute, Sydney, Australia St. Vincent's Clinical School, Faculty of Medicine, University of New South Wales, Sydney, Australia c Human Cancer Genetics Program, The Ohio State University, Colombus, USA d School of Biotechnology and Biomolecular Sciences, Faculty of Science, University of New South Wales, Sydney, Australia b

art ic l e i nf o

a b s t r a c t

Article history: Received 23 December 2013 Received in revised form 21 April 2014 Accepted 23 April 2014

CITED2 is a transcriptional co-factor with important roles in many organs of the developing mammalian embryo. Complete deletion of this gene causes severe malformation of the placenta, and results in significantly reduced embryonic growth and death from E14.5. The placenta is a complex organ originating from cells derived from three lineages: the maternal decidua, the trophectoderm, and the extra-embryonic mesoderm. Cited2 is expressed in many of these cell types, but its exact role in the formation of the placenta is unknown. Here we use a conditional deletion approach to remove Cited2 from overlapping subsets of trophectoderm and extra-embryonic mesoderm. We find that Cited2 in sinusoidal trophoblast giant cells and syncytiotrophoblasts is likely to have a non-cell autonomous role in patterning of the pericytes associated with the embryonic capillaries. This function is likely to be mediated by PDGF signaling. Furthermore, we also identify that loss of Cited2 in syncytiotrophoblasts results in the subcellular mislocalization of one of the major lactate transporters in the placenta, SLC16A3 (MCT4). We hypothesize that the embryonic growth retardation observed in Cited2 null embryos is due in part to a disorganized embryonic capillary network, and in part due to abnormalities of the nutrient transport functions of the feto-maternal interface. & 2014 Elsevier Inc. All rights reserved.

Keywords: Cited2 Placenta Mouse Vasculature Trophoblast PDGF signaling

Introduction The placenta is a complex organ that is essential for embryonic development (reviewed in John and Hemberger, 2012). It provides a link between embryonic and maternal blood circulatory systems, and has an essential role in nutrient, gas and waste exchange. In addition, it has important functions as a hormone signaling center, as well as a source of hematopoeitic stem cells and immunomodulatory molecules. It is comprised of cells derived from three separate lineages: the maternal decidua, the trophectoderm, and the extra-embryonic mesoderm. These cells are structured into

n

Correspondence to: 405 Liverpool Street, Darlinghurst, NSW, 2010, Australia. E-mail addresses: [email protected] (J.L.M. Moreau), [email protected] (S.T. Artap), [email protected] (H. Shi), [email protected] (G. Chapman), [email protected] (G. Leone), [email protected] (D.B. Sparrow), [email protected] (S.L. Dunwoodie). 1 These authors contributed equally to this work. 2 Present address: Oettgen Laboratory, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA. 3 These authors contributed equally to this work.

four layers. The decidual tissue is most proximal to the uterine implantation site, and is separated from the main body of the placenta by a layer of parietal trophoblast giant cells (TGC) derived from the trophectoderm. Next lies the junctional zone that also consists solely of trophectoderm-derived cells. Finally there is the labyrinthine layer, which consists of a mixture of trophectodermderived trophoblasts, and embryonic vessels derived from the extra-embryonic mesoderm. It is within this layer that nutrient, gas and waste exchange between the maternal and embryonic circulatory systems occurs. In mice, the feto-maternal interface of the labyrinthine layer consists of four layers of tightly-associated cells (Enders, 1965). The embryonic vasculature is lined with vascular endothelial cells (VECs), and the maternal sinuses with mononuclear sinusoidal trophoblast giant cells (S-TGC). In-between these cells are two layers of multi-nucleated syncytiotrophoblasts. The elaborate branched structure of the embryonic vessels and maternal sinuses in the labyrinthine layer forms after the extra-embryonic allantois fuses with the trophectodermderived chorion (reviewed in Arora and Papaioannou, 2012). Prior to chorio-allantoic fusion, cells in the distal allantois differentiate into VECs that coalesce to form endothelial tubes, and then undergo remodeling to form the umbilical vessels. These vessels

http://dx.doi.org/10.1016/j.ydbio.2014.04.023 0012-1606/& 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Moreau, J.L.M., et al., Cited2 is required in trophoblasts for correct placental capillary patterning. Dev. Biol. (2014), http://dx.doi.org/10.1016/j.ydbio.2014.04.023i

2

J.L.M. Moreau et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

then invade the chorion, and undergo branching morphogenesis (angiogenesis). When this process is complete, the ingressing arterioles traverse the entire labyrinthine layer, before forming an elaborate capillary mass extending back toward the base of the placenta (Adamson et al., 2002; Rodriguez et al., 2004; Withington et al., 2006). By contrast, the venules only penetrate a short way into the labyrinthine layer, reaching up to meet the arteriole capillaries. At the same time, ACTA2-positive vascular smooth muscle cells also migrate from the allantois, and coat the larger vessels. ACTA2-positive pericytes penetrate deeper into the placenta, and form cores around which the arteriole capillary loops are organized (Ohlsson et al., 1999). In the placenta, angiogenesis is regulated in much the same manner as elsewhere in the developing embryo (reviewed in Chung and Ferrara, 2011). The most important growth factors involved in this process are the platelet-derived growth factor (PDGF)/vascular endothelial growth factor (VEGF) family. These are secreted disulfide-linked proteins that bind to receptors of the receptor tyrosine kinase (RTK) superfamily of proteins. There are 4 different PDGF proteins with two PDGF-specific receptors, and 5 VEGF proteins with three VEGF-specific receptors (Flt-1, KDR and Flt-4) and an additional soluble form of Flt-1 (sFlt-1 or sVEGFR1). These proteins have multiple roles in embryonic development, but in placental angiogenesis VEGF secreted by the allantois and labyrinthine stromal cells stimulates the migration and proliferation of VECs that express VEGF receptors (Arora and Papaioannou, 2012). By contrast, PDGF is expressed by VECs and PDGF receptors by perivascular mesenchymal cells, and PDGF signaling is required for pericyte recruitment, proliferation and differentiation (Jain, 2003; Ribatti et al., 2011). Genetic deletion of VEGF or VEGF receptors results in loss of VECs in the allantois and/or placenta (Ferrara et al., 1996; Shalaby et al.,1995), whereas loss of PDGF signaling results in reduced numbers of placental pericytes, and dilated embryonic capillaries (Bjarnegard et al., 2004; Ohlsson et al., 1999). CITED2 (CBP/p300-interacting transactivator, with Glu/Asp-rich carboxy-terminal domain 2) is a multi-functional protein that is widely expressed in embryonic and extra-embryonic tissues during mammalian development, and in the adult (Barbera et al., 2002; Dunwoodie et al., 1998; Weninger et al., 2005; Withington et al., 2006). It has been the subject of intensive investigation over the past few years, uncovering important roles in embryonic development, embryonic and adult stem cells and the regulation of transcription and signal transduction pathways. We and others have shown that homozygous Cited2 null (Cited2ΔlacZ/ΔlacZ) embryos die around embryonic day (E)14.5 (Bamforth et al., 2001; Barbera et al., 2002; Yin et al., 2002). These embryos have a wide range of defects including disruptions of left-right patterning, neural tube closure and heart formation (Bamforth et al., 2001, 2004; Barbera et al., 2002; Lopes Floro et al., 2011; Weninger et al., 2005; Yin et al., 2002). Conditional knockout embryos that survive past E14.5 have revealed that Cited2 also plays a role in the formation of a wide variety of tissues in more mature embryos such as the liver, lung and eye (Chen et al., 2008, 2009; Qu et al., 2007; Xu et al., 2008). Cited2 is also required for the maintenance of quiescence of hematopoietic stem cells (Chen et al., 2007; Du et al., 2012, 2014; Kranc et al., 2009) and for regulation of pluripotency and differentiation in mouse embryonic stem cells (Li et al., 2012; Pritsker et al., 2006). Despite these findings, there is no unifying model of CITED2's mode of action, as it appears to act variously as: a transcriptional activator (Bamforth et al., 2004; Glenn and Maurer, 1999; Tien et al., 2004); a competitive inhibitor of transcription (Bhattacharya et al., 1999; Yokota et al., 2003); or a modulator of protein acetylation (Sakai et al., 2012). Therefore its requirement in each tissue or cell type needs to be investigated individually. The placentas of complete Cited2 null (Cited2ΔlacZ/ΔlacZ) conceptuses are significantly smaller than Cited2 þ / þ placentas from E12.5, and Cited2ΔlacZ/ΔlacZ embryos are

significantly smaller two days later (E14.5) and die before birth (Barbera et al., 2002). At E14.5, Cited2ΔlacZ/ΔlacZ placentas have fewer differentiated trophoblast cell types, and their fetal vasculature is disorganized (Withington et al., 2006). In particular, the capillaries at the tips of the arterioles and venules are severely stunted, disorganized and irregular in shape and thickness. By contrast, the maternal blood spaces in Cited2ΔlacZ/ΔlacZ placentas only show minor defects, with a slight increase in the size of sinusoids in some placentas. Here we have used a conditional deletion approach to investigate the role of Cited2 in embryonic capillary formation in the developing placenta. We find that Cited2 in S-TGC and syncytiotrophoblasts is likely to have a non-cell autonomous role in patterning of the pericytes associated with the embryonic capillaries. This function is likely to be mediated by PDGF signaling. Furthermore, we also identify that loss of Cited2 in syncytiotrophoblasts results in the subcellular mislocalization of one of the major lactate transporters in the placenta, SLC16A3 (MCT4). We hypothesize that the embryonic growth retardation observed in Cited2ΔlacZ/ΔlacZ embryos is due in part to a disorganized embryonic capillary network, and in part due to abnormalities of the nutrient transport functions of the feto-maternal interface.

Materials and methods Mouse lines and genotyping This research was performed following the guidelines, and with the approval, of the Garvan Institute of Medical Research/St Vincent's Animal Experimentation Ethics Committee, research approvals 09/33 and 12/33. Mouse lines carrying targeted alleles and randomly inserted transgenes used in these studies are as follows: Cited2ΔlacZ [Cited2tm1Jpmb] (Barbera et al., 2002), Cited2flox [Cited2tm1.1Dunw] (Preis et al., 2006), Tpbpa-Cre [Tg(Tpbpa-cre,–EGFP)5Jcc] (Simmons et al., 2007); Tek-Cre [Tg(Tek-cre)12Flv] (Koni et al., 2001); Cyp19-Cre [Tg (Cyp19a1-cre)5912Gle] (Wenzel and Leone, 2007); R26R [Gt(ROSA) 26Sortm1Sor] (Soriano, 1999); Z/EG [Tg(CAG-Bgeo/GFP)21Lbe/J] (Novak et al., 2000). All mouse lines, with the exception of Cyp19-Cre and Z/EG, were backcrossed for a minimum of ten generations into the C57BL/6J genetic background, and were maintained on this background. The Cyp19-Cre mouse line was backcrossed for ten generations into the FVB genetic background, and was maintained on this background. The Z/EG mouse line was backcrossed for 10 generations into the Quackenbush Swiss (Q(S)) genetic background, and was maintained on this background. Mice and embryos were genotyped by PCR using the following primers: Cited2ΔLacZ 50 -GACAACCCCCCCCAAATGACTGAC-30 and 50 -GGCGATGCCTGCTTGCCGAATATC-30 ; Cited2flox 50 -GTCTCAGCGTCTGCTCGTTT-30 ; and 50 -CTGCTGCTGTTGGTGATGAT-30 ; CYP19-Cre 50 -GACCTTGCTGAGATTAGATC -30 and 50 -GAGAGAGAAGCATGTTTAGCTGGCC-30 ; Tpbpa-Cre 50 -TCCAGTGACAGTCTTGATCCTTAAT-30 and 50 -AAATTTTGGTGTACGGTCAGTAAAT-30 ; Tek-Cre 50 -CATTTGGGCCAGCTAAACAT-30 and 50 -ATTCTCCCACCGTCAGTACG-30 ; R26R 50 -CACCAACGTAACCTATTCCA-30 and 50 -TGTAGTCGGTTTATGCAGCA-30 ; Z/EG 50 -CACCAACGTAACCTATTCCA-30 and 50 -TGTAGTCGGTTTATGCAGCA-30 . Mice heterozygous for the Cited2ΔLacZ allele (Cited2ΔLacZ/ þ ) were crossed with mice carrying with various Cre alleles. For Tpbpa-Cre and Tek-Cre alleles, Cited2ΔLacZ/ þ Creþ males were mated with Cited2flox/flox females, generating three control genotypes: Cited2 þ /flox Cre-; Cited2 þ /flox Creþ ; Cited2ΔLacZ/flox Cre-; and the conditionally deleted genotype Cited2 ΔLacZ/flox Creþ. For the Cyp19-Cre allele, female double heterozygotes (Cited2ΔLacZ/ þ Creþ ) were mated with Cited2flox/flox males, generating three control genotypes: Cited2 þ /flox Cre-; Cited2 þ /flox Creþ; Cited2ΔLacZ/flox Cre-; and the conditionally deleted genotype Cited2ΔLacZ/flox Creþ . Pups, embryos



and placentas were collected at various stages of development, and were weighed and genotyped. Placental histological analysis Placentas were harvested at E14.5, fixed and sectioned in the longitudinal plane. In all cases, sections from the central region of the placenta (as judged by the umbilical attachment site) were examined since this is where the morphology was well defined and most consistent. For ß-galactosidase staining and EGFP visualization, placentas were fixed in 1% paraformaldehyde for 4 hours at 4 1C. For immunohistochemistry analysis, placentas were fixed in 4% paraformaldehyde for 5 hours at 4 1C. Placentas were processed for cryo- and paraffin sections as previously described (Rodriguez et al., 2004; Withington et al., 2006) except for the antigen retrieval step on paraffin sections which was performed by boiling sections in 10 mM Tris HCl pH 9.0, 1 mM EDTA, 0.05% Tween 20 buffer for 20 min. No antigen retrieval was used on cryosections. To reduce autofluorescence, paraffin sections were incubated in Sudan Black B as described in (Viegas et al., 2007). The following reagents were used: rat monoclonal anti-PECAM1 (#550274 BD Pharmigen, 1:200); mouse monoclonal anti-α-smooth muscle actin (#M0851 clone1A4, Dako Cytomation, 1:200); chicken polyclonal anti-ß-galactosidase (#AB9361 Abcam, 1:100); rabbit polyclonal anti-ß-galactosidase (#A-11132 Molecular probes, 1:100); mouse monoclonal anti-ATP1A1, (Naþ/Kþ transporting ATPase, alpha 1 polypeptide, clone alpha 6F, Developmental Studies Hybridoma Bank, 1:20); chicken polyclonal antiMonocarboxylate Transporter 1 (SLC16A1, #AB1286-I Merck Millipore, 1:200); rabbit polyclonal anti-Monocarboxylate Transporter 4 (SLC16A3, #AB3314P Merck Millipore, 1:200); rabbit polyclonal antiPDGFB (H-55 #7878 Santa Cruz, 1:50); rabbit polyclonal anti-PDGFRβ (968, #sc432 Santa Cruz, 1:50); rabbit polyclonal anti-GLUT1 (SLC2A1, NB110-39113 Novus Biologicals, 1:200); Isolectin GS-IB4 from Griffonia simplicifolia conjugated with Alexa Fluors 488 (#l21411 Molecular probes 1:200) or Alexa Fluors 568 (l21412 Molecular Probes 1:200); 40 ,6-Diamidino-2-phenylindole dihydrochloride (DAPI #D9542 SigmaAldrich, 1:1000); TO-PROs-3 (Invitrogen 1:1000); Hoechst 33258 (#3569 Invitrogen 1:1000). Donkey secondary antibodies raised against chicken, rabbit, rat or mouse and conjugated to fluorophores DyLight649, Alexa Fluors488, Alexa Fluors549, or DyLight 405 were purchased from Jackson ImmunoResearch. Images were captured on a Carl Zeiss LSM 7 Duo confocal microscope. Quantitative analysis Quantitative analysis was performed using the ImageJ version 1.48a software (National Institutes of Health). For capillary lumen cross-sectional area, longitudinal placental sections from three placentas were analyzed for each genotype. Individual sections were immunostained with an anti-PECAM1 antibody, and three randomly chosen fields for each section (encompassing the fetal capillaries) were imaged with a 20 objective. The lumen cross-sectional area of every capillary visible ( 60 per field) was selected using the Wand tool, and measured. Values were averaged for each placenta, the mean and standard error were calculated for each genotype and area relative to Cited2 þ /ΔlacZ was plotted. For MCT1 and MCT4 quantitation, longitudinal placental sections from three placentas were analyzed for each genotype and for GLUT1 sections, from four placentas were analyzed. Individual sections were co-stained with an anti-MCT1 antibody, and either anti-MCT4 or anti-GLUT1 antibodies, and three (two for GLUT1) randomly chosen fields for each section (encompassing the fetal capillaries) were imaged with a 20x objective. Fluorescence intensity was adjusted equally in all images using the same threshold value to remove non-specific background staining. For each protein, total staining area was measured. Values were averaged for each placenta, the mean and standard error were calculated for each

3

genotype and staining area relative to Cited2 þ /ΔlacZ was plotted. For PDGFRß quantitation, longitudinal placental sections from three placentas were analyzed for each genotype. Individual sections were immunostained with anti-PDGFRß and anti-ACTA2 antibodies, and three randomly chosen fields for each section (encompassing the fetal capillaries) were imaged with a 20 objective. Fluorescence intensity in the PDGFRß and ACTA2 channels were adjusted equally in all images using the same threshold value to remove non-specific background staining. To determine the intensity of PDGFRß staining in ACTA2-positive cells, the thresholded ACTA2 channel was binarized, the Image Calculator “multiply” function was used on the binarized ACTA2 and thresholded PDGFRß channels, and the relative PDGFRß staining per pixel measured. To determine the intensity of staining in the ACTA2-negative cells, the binarized ACTA2 channel was inverted, and the same process repeated. Values were averaged for each placenta, the mean and standard error calculated for each genotype, and mean staining intensity relative to Cited2 þ /ΔlacZ was plotted. For PDGFB quantitation, longitudinal placental sections from three placentas were analyzed for each genotype. Individual sections were immunostained with an anti-PDGFB antibody, and three randomly chosen fields for each section (encompassing the fetal capillaries) were imaged with a 20 objective. Fluorescence intensity was adjusted equally in all images using the same threshold value to remove nonspecific background staining, and then the relative PDGFB staining per pixel was measured. Values were averaged for each placenta, the mean and standard error calculated for each genotype, and staining intensity relative to Cited2 þ /ΔlacZ was plotted. For each analysis, statistical significance was calculated using one-way analysis of variance (ANOVA) with Tukey's post-hoc test. For capillary crosssectional area, prior to analysis by ANOVA the data were transformed using natural logarithms to correct for skew.

Results Cited2ΔlacZ/ΔlacZ placentas have disorganized pericytes and enlarged capillaries Previously we have shown an alteration in the protein expression pattern of the vascular smooth muscle cell and pericyte marker α smooth muscle actin (ACTA2) in Cited2ΔlacZ/ΔlacZ placentas prior to the advent of reduced weight (Withington et al., 2006). However these studies did not show precise details of the altered placental morphology. Therefore we re-evaluated the distribution of vascular endothelial cells (PECAM1-positive) and pericytes (ACTA2-positive) within the placental labyrinthine zone of Cited2 þ / þ , Cited2 þ /ΔlacZ and Cited2ΔlacZ/ΔlacZ E14.5 placentas using immunofluorescence and confocal microscopy. At this stage of placental development, ACTA2-positive pericytes are arranged in core-like structures around which the fetal capillaries are organized (Ohlsson et al., 1999). At low magnification there was no clear difference in the overall expression levels of either protein between genotypes (Fig. S1). We next examined magnified views of embryonic vessels at three different positions within the labyrinthine layer. Large vessels at the base of the placenta (Fig. S2), and medium sized vessels in the middle of the labyrinthine layer (Fig. S3), both showed no difference in PECAM1 or ACTA2 expression level or pattern between genotypes. By contrast, the ACTA2-positive pericytes adjacent to the embryonic capillaries were disorganized and had reduced ACTA2 expression in Cited2ΔlacZ/ΔlacZ placentas (Fig. 1E–H) compared to Cited2 þ / þ (not shown) and Cited2 þ /ΔlacZ placentas (Fig. 1A–D). In addition, PECAM1 staining followed by quantitation of capillary lumen cross-sectional area revealed that these capillaries were significantly enlarged in the Cited2ΔlacZ/ΔlacZ E14.5 placentas compared to those of Cited2 þ /ΔlacZ placentas (Fig. S4A), in keeping with our


4


Fig. 1. Comparison of vascular endothelial cell and smooth muscle cell markers in fetal capillaries of Cited2ΔlacZ/ þ , Cited2ΔlacZ/ΔlacZ and conditionally deleted placentas. Representative longitudinal cryosections of (A–D) Cited2ΔlacZ/ þ (Het), (E–H) Cited2ΔlacZ/ΔlacZ (Null), (I–L) Cited2ΔlacZ/flox; Tek-Creþ, (M–P) Cited2ΔlacZ/flox; Tpbpa-Creþ and (Q–T) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas stained with PECAM1 (A,E,I,M,Q) and ACTA2 (B,F,J,N,R). (C,G,K,O,S) Merged images with nuclei stained with TO-PRO-3 (blue). (D,H, L,P,T) Magnified views of the boxed areas. Total number of placentas examined: Cited2 þ / þ n¼ 4 (not shown), Cited2ΔlacZ/ þ n ¼7, Cited2ΔlacZ/ΔlacZ n¼ 9, Tek-Cre n¼ 9, Tpbpa-Cre n¼ 5 and Cyp19-Cre n¼ 5. The scale bar represents 50 μm in panels A–O, and 13 μm in the magnified views.

previous observations using fetal vascular resin casts of E14.5 Cited2ΔlacZ/ΔlacZ placentas (Withington et al., 2006). Conditional deletion of Cited2 in subsets of placenta cells results in decreased placenta and embryo weight, but does not affect embryo survival Cited2 is expressed in many different types of cell within the placenta, including both mesoderm- and trophectoderm-derived cells. We therefore used a conditional deletion strategy to determine in

which cell type(s) Cited2 was required for patterning of the embryonic capillaries. The Cited2ΔlacZ allele is expressed in endothelial cells of the fetal vasculature (Withington et al., 2006). We therefore used the TekCre transgene to determine if Cited2 was required within these cells for placental patterning. This allele drives expression of Cre recombinase in all vascular endothelial cells (VECs) from E7.5 in both embryonic and extra-embryonic vessels, including those of the yolk sac and placenta (Constien et al., 2001; Koni et al., 2001; Tang et al., 2004). It is also expressed in hematopoietic cells. We crossed Cited2ΔlacZ/ þ Tek-Creþ males with Cited2flox/flox females and collected, weighed



5

Fig. 2. Conditional deletion of Cited2 affects placental and embryo weight. Graphs showing normalized weights of (A) Tek-Cre E16.5 placentas, (B) Tek-Cre E18.5 embryos, (C) Tpbpa-Cre E16.5 placentas, (D) Tpbpa-Cre P0 pups, (E) Cyp19-Cre E14.5 placentas and (F) Cyp19-Cre E16.5 embryos. Error bars indicate standard deviations. Student's t-test was used to compare the weights of conditionally deleted embryos and placentas to those of the three control genotypes. n p o 0.05, nn p o0.01, nnn p o 0.001. C2f/þ Cre-, C2f/ þ Creþ and C2f/C2Δ Cre- represent the three control genotypes; and C2f/C2Δ Creþ represents conditionally deleted embryos.

and genotyped mice and placentas at various developmental stages from E14.5 to birth. In contrast to Cited2ΔlacZ/ΔlacZ (Withington et al., 2006), deletion of Cited2 with the Tek-Cre allele had no effect on embryo or pup survival (Table S1). However conditionally deleted placentas were significantly smaller than controls at E16.5 (Fig. 2A) and conditionally deleted embryos were smaller than controls at E18.5 (Fig. 2B). Cited2 is broadly expressed in many trophoblast cell types. We next used the Tpbpa-Cre transgene (Simmons et al., 2007) to delete Cited2 from a subset of these cells. This transgene is active from E8.5, with recombinase activity evident in spongiotrophoblasts, some trophoblast giant cells (TGCs: spiral artery-TGCs, canal-TGCs and parietal-TGCs), but not in sinusoidal-TGCs or labyrinthine syncytiotrophoblasts (Hu and Cross, 2011; Simmons et al., 2007). We crossed Cited2ΔlacZ/þ Tpbpa-Creþ males with Cited2flox/flox females and

collected mice and placentas at various developmental stages from E14.5 to birth. Deletion of Cited2 from Tpbpa-Cre expressing trophoblasts had no effect on embryo or pup survival (Table S1), however conditionally deleted placentas were significantly smaller than controls from E16.5 (Fig. 2C) and conditionally deleted pups were smaller than controls at P0 (Fig. 2D). We next conditionally deleted Cited2 from a broader range of derivatives of trophoblast stem cells, including spongiotrophoblasts, labyrinthine trophoblasts and some secondary TGC, using the Cyp19-Cre transgene (Wenzel and Leone, 2007). The Cyp19-Cre transgene shows the strongest and most stable expression when inherited through the maternal line (Wenzel and Leone, 2007), thus for experiments using this transgene we crossed Cited2ΔlacZ/ þ Cyp19-Creþ females with Cited2flox/flox males and collected mice and placentas at various developmental stages from E14.5 to birth.



6

Deletion of Cited2 from Cyp19-Cre expressing trophoblasts had no effect on embryo or pup survival (Table S1), however conditionally deleted placentas were significantly smaller than controls at E14.5 (Fig. 2E) and conditionally deleted embryos were significantly smaller than controls at E16.5 (Fig. 2F). These results suggest that Cited2 may be required in multiple cell types for correct formation of the placenta. However since the time of onset of placental and embryonic growth restriction is earliest in the Cyp19-Cre conditional deletion, then the most important function of Cited2 is most likely to be in trophoblasts. Conditional deletion of Cited2 using Cyp19-Cre and Tek-Cre partially recapitulates the placental phenotype of the Cited2ΔlacZ/ΔlacZ We next examined the placental phenotypes present in conditionally deleted placentas at the molecular level. We compared the distribution of vascular endothelial cells (PECAM1) and pericytes (ACTA2) within the placental labyrinthine zone of conditionally deleted E14.5 placentas to those of Cited2 þ / þ (not shown), Cited2ΔlacZ/ þ (Fig. 1A–D) and Cited2ΔlacZ/ΔlacZ (Fig. 1E–H) placentas. In Tpbpa-Cre deleted placentas the capillary lumen cross-sectional area was similar in size to those in Cited2 þ / þ or Cited2 þ /ΔlacZ placentas (Fig. S4A), and the arrangement of ACTA2-positive pericytes was essentially the same (Fig. 1M–P). By contrast, in Cyp19-Cre deleted placentas the capillary lumen cross-sectional area was significantly increased (Fig. S4A), and the ACTA2-positive pericytes were more sparse and less well organized (Fig. 1Q–T). Overall the Cyp19-Cre deleted placentas appeared very similar to Cited2ΔlacZ/ΔlacZ placentas. Interestingly, the Tek-Cre deleted placentas showed an intermediate phenotype, with considerable variation between individual placentas: 4/9 Tek-Cre placentas were similar to Cited2ΔlacZ/ΔlacZ (Fig. 1I–L), 2/9 had an intermediate phenotype (not shown) and 3/9 were similar to Cited2ΔlacZ/ þ placentas (Fig. 1A–D). Due to the great variation in the phenotype of individual placentas, capillary lumen cross-sectional area was not quantified for this line. These results confirm that the most important placental function of Cited2 at E14.5 is in Cyp19-Cre expressing trophoblasts. The Cited2flox allele Is more broadly deleted than expected Since deletion of Cited2 by each transgene gave a different phenotype, and each of the three Cre transgenes is expressed in a partially overlapping subset of placental labyrinthine cells, we next sought to use this information to determine exactly which cells in the labyrinthine layer were responsible for the phenotype of Cited2ΔlacZ/ΔlacZ. Previously the expression of each of the Cre transgenes has been described by use of the R26R reporter mouse line. This line only produces cytoplasmic ß-galactosidase protein in cells in which Cre recombinase activity is present, or in cells that have been derived from these (Soriano, 1999). However it is well known

that the efficiency of deletion of different floxed loci can vary (Vooijs et al., 2001). Therefore we made use of the LacZ gene in the Cited2flox allele to confirm that all three Cre lines were deleting Cited2 in the previously reported patterns. We designed the Cited2flox allele such that when the Cited2 coding region is deleted by Cre recombinase, it is replaced by an in-frame nuclear LacZ coding region (nLacZ) under the transcriptional control of the Cited2 promoter (Preis et al., 2006). Thus, with this reporter ß-galactosidase will only be expressed in the nucleus of cells where: (1) Cre has been expressed; and (2) the Cited2 locus is currently transcriptionally active. To test the Tek-Cre and Tpbpa-Cre alleles, we crossed Creþ males with either R26R or Cited2flox/flox females, and for the Cyp19-Cre allele, we crossed Creþ females with R26R or Cited2flox/flox males. Creþ; Cited2flox/ þ and Creþ; R26R þ /- placentas were collected at E14.5, fixed, cryosectioned and stained for ßgalactosidase activity. For each of the three Cre alleles, there was clearly more ß-galactosidase activity detectable from the Cited2flox allele than from the R26R locus (Fig. S5). Surprisingly, in each case ß-galactosidase activity from the Cited2flox allele was also present in a much broader domain than that expressed from the R26R locus. For example, deletion of Cited2 by the Tek-Cre allele occurred in the junctional zone (Fig. S5A, arrow) and in non-VECs in the labyrinth (Fig. S5B, arrowheads), and Tpbpa-Cre deleted Cited2 in the labyrinthine layer (Fig. 5H) in addition to the junctional zone (Fig. S5G, arrow). This could indicate that either the R26R locus is not well expressed in the placenta, or that the Cited2flox allele is ectopically expressing ß-galactosidase. To distinguish between these possibilities, we repeated our deletion analysis using a third reporter allele Z/EG (Novak et al., 2000). This reporter allele ubiquitously expresses lacZ but Cre excision removes the lacZ gene, activating expression of a second reporter, EGFP. This locus is integrated just upstream of the Rasa4 gene on mouse chromosome 5 (Colombo et al., 2010), and its expression is driven by the chicken ß-actin promoter with upstream (pCAGGS) enhancer elements. Thus it is completely independent of the R26R promoter. To test the Tek-Cre and Tpbpa-Cre alleles with this reporter, we crossed Creþ males with Z/EG þ females, and for the Cyp19-Cre allele, we crossed Creþ females with Z/EG þ males. Creþ ; Z/EG þ placentas were collected at E14.5, fixed, cryosectioned and imaged for EGFP expression. For each of the three Cre alleles, the pattern of EGFP reporter gene expression was similar to the pattern of ß-galactosidase activity detectable from the Cited2flox allele, rather than that from the R26R locus (Fig. S5C,D,I,J,O,P). This supports the conclusion that the R26R locus is not expressed in all placental cell types. Confirmation of the identity of Cited2 deleted cells for each Cre allele We next used a combination of gross morphology and immunohistochemistry to precisely determine in which cells Cited2flox was deleted by each Cre allele (results summarized in Table 1).

Table 1 A summary of Cited2flox and R26R ß-galactosidase reporter gene expression after Cre-mediated deletion. Cre line

Reporter

Trophoblasts

Vasculature

TGC

Tek-Cre Tpbpa-Cre Cyp19-Cre

R26R Cited2flox R26R Cited2flox R26R Cited2flox

P

Spa

C

S

– þþ 7 þþþ þþ þþþ

– – þþ þ þ þ

– þþ 7 þþ þþ þþ

– þþþ – 7 þþ þþ

SpT

GlyT

SynT-I

SynT-II

EC

PC

– þþ þþ þþ þþ þþþ

– þþ þþþ þþ þþ þþþ

– þ – þ þþ þþ

– 7 – – þþ þþ

þþþ – – – – –

– – – þþþ – 7

TGC ¼trophoblast giant cell; P ¼parietal TGC; Spa ¼spiral artery TGC; C¼ canal TGC; S¼ sinusoidal TGC; SpT ¼spongiotrophoblast cells; GlyT¼ glycogen trophoblast cells; SynT-I¼syncytiotrophoblast layer I; SynT-II¼ syncytiotrophoblast layer II; EC¼ endothelial cell; PC ¼pericyte.



7

Fig. 3. Comparison of ß-galactosidase expressing cells with vascular endothelial cells in Cited2 conditionally deleted placentas. Representative longitudinal paraffin sections of (A–D) Cited2flox/ þ ; Tek-Creþ , (E–H) Cited2flox/ þ ; Tpbpa-Creþand (I–L) Cited2flox/ þ ; Cyp19-Creþ E14.5 placentas. (A,E,I) ß-galactosidase (red) and TO-PRO-3 (blue) staining. (B,F,J) The same sections stained with ß-galactosidase (red) and isolectin B4 (IB4, green). (C,G,K) Merged images. (D,H,L) Schematic representation of the boxed areas in panels D,H,L showing the organization of fetal blood vessels (fbv), sinusoidal-TGC (asterisks) and syncytiotrophoblast layers (arrows). Total number of placentas examined: Tek-Cre n¼ 4, Tpbpa-Cre n ¼4 and Cyp19-Cre n¼ 4. The scale bar represents 50 μm in immunofluorescence panels and 13 μm in schematic panels.

Given the major phenotypic effect of Cited2 deletion in the placenta was a disruption of embryonic capillary formation, we focused our attention on the labyrinthine layer. To test the Tek-Cre and Tpbpa-Cre alleles, we crossed Creþ males with Cited2flox/flox females, and for the Cyp19-Cre allele, we crossed Creþ females with Cited2flox/flox males. Placentas were collected at E14.5, fixed and paraffin sectioned. Immunostaining of ß-galactosidase protein in paraffin sections (Fig. 3) was consistent with ß-galactosidase activity in cryosections (Fig. S5). Sinusoidal TGC line the maternal blood spaces, and are easily identified by their large nuclei (Fig. 3, indicated by asterisks). In both Tek-Cre (Fig. 3A–D) and Cyp19-Cre (Fig. 3I–L), all S-TGC were positive for ß-galactosidase, although staining intensity varied considerably between individual S-TGC in the Cyp19-Cre line. By contrast, only half of the S-TGC in Tpbpa-Cre placentas were positive for ß-galactosidase (Fig. 3E–H). Staining of placental sections with isolectin GS-IB4 from Griffonia simplicifolia (IB4) to mark vascular endothelial cells lining the embryonic vasculature revealed that high levels of ß-galactosidase were also present in some of the small cells situated between the maternal blood spaces and the embryonic vasculature (Fig. 3, indicated by arrows). These cells are syncytiotrophoblasts that are organized into two layers, with SynT-I cells adjacent to the S-TGC lining the maternal blood spaces and SynT-II cells adjacent to the VEC layer lining the embryonic vasculature (Enders, 1965; HernandezVerdun, 1974). These cells can be distinguished by immunostaining for SLC16A1 (also called MCT1, expressed at the apical membrane of SynT-I) and SLC16A3 (also called MCT4, expressed on the basal membrane of SynT-II; (Nagai et al., 2010)). Placentas for all three Cre crosses were co-stained for ß-galactosidase, MCT1 and a membrane marker (Naþ/K þ transporting ATPase,

(Mobasheri et al., 2001); Fig. 4). In all three cases, ß-galactosidase was clearly present in SynT-I cells (Fig. 4, SynT-I nuclei indicated by arrows). By contrast, co-staining with ß-galactosidase, MCT4 and ATPase revealed that ß-galactosidase was only expressed in SynT-II cells in placentas from the Tek-Cre and Cyp19-Cre crosses (Fig. 5, SynT-II nuclei indicated by arrows). Next we co-stained placentas for all three Cre crosses for ß-galactosidase and the pericyte marker ACTA2 (Fig. 6). Surprisingly, ß-galactosidase was clearly apparent in all pericytes in placentas from the Tpbpa-Cre cross (Fig. 6E–H), and 50% of those from the Cyp19-Cre cross (Fig. 6I–L), but completely absent from those from the Tek-Cre cross (Fig. 6A–D). Finally we also examined ß-galactosidase expression in the junctional and decidual layers of the placenta and found that ß-galactosidase was also expressed in spongiotrophoblasts, glycogen trophoblasts and parietal TGC in all three Cre crosses; and in spiral artery trophoblasts in Tpbpa-Cre and Cyp19Cre crosses (Table 1; data not shown). Deletion of Cited2 in syncytiotrophoblasts affects MCT4 protein expression The capillary phenotype was present to varying degrees in both Cyp19-Cre and Tek-Cre deleted placentas, but not in Tpbpa-Cre deleted placentas. The only cells in which Cited2 is expressed in Cited2 þ / þ and Tpbpa-Cre placentas, but is deleted in Cyp19-Cre and Tek-Cre, are SynT-II and some S-TGC. Thus we hypothesized that the expression of Cited2 in these cells might be required for correct capillary patterning. We therefore compared the expression of molecular markers in syncytiotrophoblasts between Cited2 þ / þ , Cited2ΔlacZ/ þ , Cited2ΔlacZ/ΔlacZ and conditionally deleted


8


Fig. 4. Comparison of ß-galactosidase expressing cells with SynT-I cells in Cited2 conditionally deleted placentas. Representative longitudinal paraffin sections of (A–D) Cited2flox/ þ ; Tek-Creþ , (E-H) Cited2 Cited2flox/ þ ; Tpbpa-Creþ and (I–L) Cited2flox/ þ ; Cyp19-Creþ E14.5 placentas. (A,E,I) ß-galactosidase (white) and MCT1 (green) staining. (B,F, J) The same sections stained with MCT1 (green), ATPase (red) and Hoechst 33258 (blue). (C,G,K) Merged images. (D,H,L) Schematic representation of the boxed areas in panels D,H,L showing the apical membrane of the SynT-I layer (green) facing the maternal sinusoids (ms), sinusoidal TGC (asterisks), and SynT-I nuclei (arrows). Total number of placentas examined: Tek-Cre n¼4, Tpbpa-Cre n¼ 4 and Cyp19-Cre n ¼4. The scale bar represents 50 μm in immunofluorescence panels and 25 μm in schematic panels.

E14.5 placentas for each of the Cre alleles. We intercrossed Cited2ΔlacZ/ þ males and females to produce Cited2 þ / þ , Cited2ΔlacZ/ þ and Cited2ΔlacZ/ΔlacZ placentas; for the Tek-Cre and Tpbpa-Cre lines we crossed Cited2ΔlacZ/ þ Creþ males with Cited2flox/flox females; and for the Cyp19-Cre line we crossed Cited2ΔlacZ/ þ Creþ females with Cited2flox/flox males. E14.5 placentas were collected, fixed and paraffin sections immunostained using antiMCT1 and anti-MCT4 antibodies (Fig. 7). There were no obvious differences in the expression levels or pattern of MCT1 between any of the samples, suggesting that SynT-I cells were unaffected by deletion of Cited2. This was confirmed by quantification of the staining area of MCT1 in Cited2ΔlacZ/ þ , Cited2ΔlacZ/ΔlacZ, Tpbpa-Cre and Cyp19-Cre lines (Fig. S4B). Tek-Cre placentas were not quantified due to their potentially variable phenotype (see below). By contrast, the subcellular localization of MCT4 was clearly altered in the Cited2ΔlacZ/ΔlacZ (Fig. 7E–H) and Cyp19-Cre conditionally deleted placentas (Fig. 7Q–T), compared to Cited2 þ / þ (not shown), Cited2ΔlacZ/ þ (Fig. 7A–D) and Tpbpa-Cre conditionally deleted placentas (Fig. 7M–P). Tek-Cre placentas had an intermediate phenotype with 3/4 placentas similar to Cited2ΔlacZ/ΔlacZ (Fig. 7I–L) and 1/4 similar to Cited2 þ / þ (not shown). Instead of the normal sharp cellsurface expression of MCT4 (Fig. 7D and P, indicated by arrows), SynT-II cells in Cited2ΔlacZ/ΔlacZ and Cyp19-Cre and Tek-Cre conditionally deleted placentas showed broad MCT4 expression throughout the cytoplasm (Fig. 7H,L,T, indicated by arrowheads). Quantification confirmed a significant increase of the MCT4 staining area in Cited2ΔlacZ/ΔlacZ and Cyp19-Cre placentas compared to Cited2ΔlacZ/ þ (Fig. S4C). Tek-Cre placentas were not quantified due

to their variable phenotype. Thus altered MCT4 expression was only detected in placentas with the capillary phenotype described above. We next sought to determine if the altered localization of MCT4 was limited to this protein, or reflected a more general perturbation of protein sorting to the membrane in SynT-II cells. MCT1 and MCT4 are the major lactate transporters expressed in the murine placenta (Nagai et al., 2010). Another transporter protein expressed in SynT-II cells is SLC2A1 (also known as GLUT1, Nagai et al., 2010). In contrast to MCT4, this protein is symmetrically expressed, being present on both apical and basal membranes of SynT-II. We immunostained cryosections from the same placentas described above with anti-GLUT1 and anti-MCT1 antibodies (Fig. S6). There were no obvious differences in the expression levels or pattern of GLUT1 between any of the samples. This was confirmed by quantification of the staining area of GLUT1 in Cited2ΔlacZ/ þ , Cited2ΔlacZ/ΔlacZ, Tpbpa-Cre and Cyp19-Cre lines (Fig. S4D). Tek-Cre placentas were not quantified due to their potentially variable phenotype. This suggests that deletion of Cited2 did not broadly affect protein sorting to the membrane of SynT-II cells. Deletion of Cited2 results in altered expression of PDGFRß and PDGFB proteins Loss of either PDGFB ligand or the PDGFRß receptor in mouse embryos results in a similar fetal capillary phenotype to that of Cited2ΔlacZ/ΔlacZ (Bjarnegard et al., 2004; Ohlsson et al., 1999). We therefore compared the expression of these proteins between Cited2 þ / þ , Cited2ΔlacZ/ þ , Cited2ΔlacZ/ΔlacZ and conditionally deleted



9

Fig. 5. Comparison of ß-galactosidase expressing cells with SynT-II cells in Cited2 conditionally deleted placentas. Representative longitudinal paraffin sections of (A–D) Cited2flox/ þ ; Tek-Creþ , (E–H) Cited2flox/ þ ; Tpbpa-Creþ and (I–L) Cited2flox/ þ ; Cyp19-Creþ E14.5 placentas. (A,E,I) ß-galactosidase (white) and MCT4 (green) staining. (B,F,J) The same sections stained with MCT4 (green), ATPase (red) and Hoechst 33258 (blue). (C,G,K) Merged images. (D,H,L) Schematic representation of the boxed areas in panels D,H,L showing the basal membrane of the SynT-II layer (green) facing the fetal blood vessels (fbv), sinusoidal TGC (asterisks) and SynT-II nuclei (arrows). Total number of placentas examined: Tek-Cre n¼4, Tpbpa-Cre n¼ 4 and Cyp19-Cre n¼ 4. The scale bar represents 50 μm in immunofluorescence panels and 25 μm in schematic panels.

E14.5 placentas for each of the Cre alleles. Previous reports suggest that in the placenta PDGFRß is expressed in pericytes and some trophoblasts (probably S-TGC), whereas PDGFB is expressed in VECs (Chhabra et al., 2012; Ohlsson et al., 1999). We sought to more precisely identify which placental cells expressed PDGFRß and PDGFB proteins in the placental labyrinth using immunofluorescence (Fig. S7). Firstly, cryosections of E14.5 Cited2 þ / þ and Cited2ΔlacZ/ þ placentas were stained for PDGFRß, PECAM1 and MCT1 (Fig. S7A–E). PDGFRß was not expressed in either VECs (Fig. S7C0 , arrow) or SynT-I (Fig. S7E0 , arrowhead). However expression was clearly present in both S-TGC (Fig. S7E0 , asterisk) and pericytes (Fig. 8C, arrow). Next, cryosections of E14.5 Cited2 þ / þ and Cited2ΔlacZ/ þ placentas were stained for PDGFB, PECAM1 and MCT1 (Fig. S7F–J). PDGFB was clearly expressed in VECs (Fig. S7H0 , arrow), but was also present between VECs and SynT-I cells (Fig. S7J0 , circle), most likely in SynT-II cells. High levels of PDGFB were also apparent in S-TGC (Fig. S7J0 , asterisk), but not pericytes (not shown). To compare the expression levels of PDGFRß between E14.5 placentas of all genotypes, cryosections were stained for PDGFRß, ACTA2 and MCT1 (Fig. S8, magnified views in Figs. 8 and 9). PDGFRß appeared to be expressed at equivalent levels in ACTA2-positive pericytes in all placentas regardless of their genotype. This was despite the reduced number and poor organization of these cells in Cited2ΔlacZ/ΔlacZ and Cyp19-Cre and Tek-Cre conditionally deleted placentas (Fig. 8). This was confirmed by quantitation of the PDGFRß staining intensity in ACTA2-positive cells in sections from three placentas from each genotype (Fig. S4E). The Tek-Cre placentas were not quantitated due to

potential phenotype variation. By contrast, the predominantly cellsurface expression of PDGFRß in S-TGC in Cited2 þ / þ (not shown), Cited2ΔlacZ/ þ (Fig. 9A–C) and Tpbpa-Cre conditionally deleted (Fig. 9J–L) placentas was replaced by a low level cytoplasmic expression in Cited2ΔlacZ/ΔlacZ (Fig. 9D–F) and Cyp19-Cre (Fig. 9M– O) conditionally deleted placentas. Tek-Cre conditionally deleted placentas had an intermediate phenotype with 2/4 placentas similar to Cited2ΔlacZ/ΔlacZ (not shown) and 2/4 the same as Cited2 þ / þ (Fig. 9G–I). To confirm our observations, we quantified PDGFRß staining intensity in S-TGC (defined as non-ACTA2expressing cells) in sections from three placentas from each genotype except Tek-Cre (Fig. S4F). The Tek-Cre placentas were not quantitated due to variation in phenotype of individual placentas. These data confirm a slight but significant reduction in PDGFRß protein levels in non-ACTA2-expressing cells from Cited2ΔlacZ/ΔlacZ and Cyp19-Cre placentas. To compare the expression levels of PDGFB between E14.5 placentas of all genotypes, cryosections were stained for PDGFB, PECAM1 and MCT1 (Fig. S9, magnified views in Figs. 10 and 11). PDGFB expression in Cited2ΔlacZ/ΔlacZ placentas was reduced in VECs, SynT-II cells and S-TGC (Figs. 10 and 11D–F) compared to Cited2 þ / þ (not shown), Cited2ΔlacZ/ þ (Figs. 10 and 11A–C) and Tek-Cre (Figs. 10 and 11G–I), Tpbpa-Cre (Figs. 10 and 11J–L) and Cyp19-Cre (Figs. 10 and 11M–O) conditionally deleted placentas To analyze further, we quantified PDGFB staining intensity in sections from three placentas from each genotype (Fig. S4G). These data confirm a reduction in mean PDGFB protein levels in Cited2ΔlacZ/ΔlacZ placentas, however statistical significance was not achieved, possibly due to small sample


10


Fig. 6. Comparison of ß-galactosidase expressing cells with pericytes in Cited2 conditionally deleted placentas. Representative longitudinal paraffin sections of (A–D) Cited2flox/ þ ; Tek-Creþ, (E–H) Cited2flox/ þ ; Tpbpa-Creþ and (I–L) Cited2flox/ þ ; Cyp19-Creþ E14.5 placentas. (A,E,I) ß-galactosidase (red) and DAPI (blue) staining. (B,F,J) The same sections stained with ACTA2. (C,G,K) Merged images. (D,H,L) Schematic representation of the boxed areas in panels C,G,K with pericytes indicated with arrows. Total number of placentas examined: Tek-Cre n¼ 4, Tpbpa-Cre n ¼4 and Cyp19-Cre n¼ 4. The scale bar represents 50 μm in immunofluorescence panels and 13 μm in schematic panels.

size. Thus in Cited2ΔlacZ/ΔlacZ and conditionally deleted placentas with a capillary phenotype, PDGFRß receptor expression was altered in trophoblasts, and in Cited2ΔlacZ/ΔlacZ placentas PDGFB expression was slightly reduced.

Discussion Complete deletion of Cited2 from the placenta results in abnormal placental formation and reduced embryonic growth. The labyrinthine layer of affected placentas has two major phenotypes. Firstly the capillaries of the embryonic vasculature are abnormally patterned. Secondly there are fewer capillary-associated pericytes, and those that remain have disrupted organization. Here we have shown that conditional deletion of Cited2 from overlapping subsets of cells in the labyrinthine layer of the placenta using Cyp19-Cre or Tek-Cre (but not Tpbpa-Cre) can recapitulate the placental phenotype of Cited2ΔlacZ/ΔlacZ (summarized in Fig. 12). We deduce that Cited2 is required in the trophectoderm-derived S-TGC and/or syncytiotrophoblasts cells for correct patterning of the adjacent embryonic vasculature. We also provide evidence that this phenotype may arise via an alteration of PDGF signaling. Cited2 expression is required in sinusoidal trophoblast giant cells or syncytiotrophoblasts Cited2 is expressed in many different types of cell within the placenta, including both extra-embryonic mesoderm- and trophectoderm-derived cells and is required for normal placental

development (Withington et al., 2006). We used a conditional deletion approach to more specifically define in which cells Cited2 is required for patterning of the embryonic vasculature and associated pericytes. We used three Cre transgenes with overlapping expression in mesodermal and trophectodermal components of the placental labyrinth. The capillary and pericyte phenotype was present in Cyp19-Cre and to a variable extent in Tek-Cre deleted placentas, but was absent from Tpbpa-Cre deleted placentas. In the labyrinthine layer, Tpbpa-Cre deleted the Cited2 locus in ACTA2positive pericytes, C-TGC and 50% of S-TGC. Since there was no capillary or pericyte phenotype detected in these placentas, we conclude that Cited2 is not required in pericytes or C-TGC for the formation of the embryonic vasculature or pericyte patterning, migration or proliferation. Secondly, in the labyrinthine layer TekCre deleted the Cited2 locus from S-TGC, C-TGC, SynT-I and SynT-II cells. However no ß-galactosidase activity was detectable in the major site of Tek-Cre expression, VECs. This suggests that Cited2 is not actively transcribed in VECs. This is contrary to our previous suggestion based on ß-galactosidase activity from the Cited2ΔlacZ allele (Withington et al., 2006). However our latest results are more convincing because we have used co-immunofluorescence and confocal microscopy to simultaneously visualize ß-galactosidasefrom the Cited2flox allele and IB4 labeled VECs in the same section (Fig. 3A–D). Therefore we conclude that Cited2 is not required in VECs for formation of the embryonic vasculature or in pericytes for pericyte patterning, migration or proliferation. Lastly, Cyp19-Cre deleted the Cited2 locus from S-TGC, C-TGC, SynT-I and SynT-II cells, and some pericytes. Taken together with the results from the other two Cre alleles, we conclude that Cited2 expression in S-TGC or



11

Fig. 7. Conditional deletion of Cited2 in the placenta affects MCT4 expression. Representative longitudinal paraffin sections of (A–D) CitedΔlacZ/ þ (Het), (E–H) Cited2ΔlacZ/ΔlacZ (Null), (I–L) Cited2ΔlacZ/flox; Tek-Cre þ , (M–P) Cited2ΔlacZ/flox; Tpbpa-Creþ and (Q–T) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas. A,E,I,M,Q) MCT4 (red) and DAPI (blue) staining. (B,F,J,N,R) The same sections stained with MCT1 (green) and DAPI (blue) staining. (C,G,K,O,S) Merged images of MCT1, MCT4 and DAPI staining, with magnified views of the areas boxed in white (D,H,L,P,T respectively). Arrows indicate regions of normal MCT4 staining, and arrowheads indicate regions of altered MCT4 staining. Total number of placentas examined: Cited2 þ / þ (not shown) n¼ 2, Cited2ΔlacZ/ þ n ¼3, Cited2ΔlacZ/ΔlacZ n¼ 4, Tek-Cre n¼4, Tpbpa-Cre n¼ 4 and Cyp19-Cre n¼4. The scale bar represents 50 μm in panels A–C, E–G, I–K, M–O, Q–S, and 13 μm in the magnified views (D,H,L,P,T).

syncytiotrophoblasts is required for correct patterning of the embryonic vasculature and/or pericyte recruitment, proliferation or differentiation. CITED2 may regulate components of PDGF signaling PDGF signaling is required for the correct development and patterning of the vasculature in both embryo and placenta (Leveen et al., 1994; Ohlsson et al., 1999; Soriano, 1994). Total loss of either

PDGFB ligand or the PDGFRß receptor in mouse results in a similar capillary and pericyte phenotype in the placenta to that observed in our study (Ohlsson et al., 1999). PDGF signaling also has a role in regulating the hematopoietic stem cell niche within the placenta (Chhabra et al., 2012). Here PDGFB secreted by VECs signals through PDGFRß in S-TGC to suppress secretion of erythropoeitin (Epo). Previously, the placental expression pattern of PDGFRß protein in the mouse has been described in pericytes and S-TGC (Chhabra et al., 2012; Ohlsson et al., 1999). We confirmed this in


12


Fig. 8. Conditional deletion of Cited2 in the placenta does not affect PDGFRß expression in pericytes. Longitudinal cryosections of (A–C) Cited2ΔlacZ/ þ (Het), (D–F) Cited2ΔlacZ/ΔlacZ (Null), (G–I) Cited2ΔlacZ/flox; Tek-Cre þ , (J–L) Cited2ΔlacZ/flox; Tpbpa-Creþ and (M–O) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas. (A,D,G.J,M) PDGFRß immunostaining. (B,E,H,K,N) The same sections stained with ACTA2. (C,F,I,L,O) Merged images showing PDGFRß (green), ACTA2 (red) and DAPI (blue) staining. Pericytes are indicated with arrows and S-TGC with asterisks. Total number of placentas examined: Cited2 þ / þ (not shown) n¼ 2, Cited2ΔlacZ/ þ n ¼3, Cited2ΔlacZ/ΔlacZ n¼ 5, Tek-Cre n¼4, Tpbpa-Cre n¼ 5 and Cyp19-Cre n¼ 5. The scale bar represents 13 μm.

normal placentas (Fig. S7C0 and E0 ). However, despite dramatic changes in the number and organization of pericytes in affected placentas, we did not observe any alteration in the expression of PDGFRß in these cells (Fig. 8D–F). By contrast, PDGFRß expression

in affected placentas was downregulated in S-TGC (Fig. 9D–F). Previously it has been reported that PDGFB expression in the placenta is limited to VECs and some hematopoietic cells (Chhabra et al., 2012; Ohlsson et al., 1999). We confirmed high levels of



13

Fig. 9. Conditional deletion of Cited2 in the placenta affects PDGFRß expression in S-TGC and SynT-II. Representative longitudinal cryosections of (A–C) Cited2ΔlacZ/ þ (Het), (D–F) Cited2ΔlacZ/ΔlacZ (Null), (G–I) Cited2ΔlacZ/flox; Tek-Creþ, (J–L) Cited2ΔlacZ/flox; Tpbpa-Creþ and (M–O) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas. (A,D,G.J,M) PDGFRß immunostaining. (B,E,H,K,N) The same sections stained with MCT1. (C,F,I,L,O) Merged images showing PDGFRß (green), MCT1 (red) and DAPI (blue) staining. SynT-I cells are indicated with arrowheads and S-TGC with asterisks. Total number of placentas examined Cited2 þ / þ (not shown) n¼ 2, Cited2ΔlacZ/ þ n¼3, Cited2ΔlacZ/ΔlacZ n¼ 5, Tek-Cre n ¼4, Tpbpa-Cre n ¼5 and Cyp19-Cre n¼ 5. The scale bar represents 13 μM.

expression of PDGFB in VECs in normal placentas (Fig. S7H0 ), but we also detected significant expression of PDGFB in SynT-II and S-TGC (Fig. S7J0 ). In Cited2ΔlacZ/ΔlacZ we found slightly reduced

levels of PDGFB in S-TGC, VECs and SynT-II cells (Fig. 10D–F). Taken together, our results suggest placental pericytes expressing PDGFRß respond to PDGFB ligand secreted by VECs as well as from


14


Fig. 10. Conditional deletion of Cited2 in the placenta affects PDGFB expression in VECs. Representative longitudinal cryosections of (A–C) Cited2ΔlacZ/ þ (Het), (D–F) Cited2ΔlacZ/ΔlacZ (Null), (G–I) Cited2ΔlacZ/flox; Tek-Creþ , (J–L) Cited2ΔlacZ/flox; Tpbpa-Creþ and (M-O) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas. (A,D,G,J,M) PDGFB immunostaining. (B,E,H,K,N) The same sections stained with PECAM1. (C,F,I,L,O) Merged images showing PDGFB (green), PECAM1 (red) and DAPI (blue) staining. VECs indicated by arrows and S-TGC with asterisks. Total number of placentas examined: Cited2 þ / þ (not shown) n¼ 2, Cited2ΔlacZ/ þ n ¼3, Cited2ΔlacZ/ΔlacZ n¼ 5, Tek-Cre n¼ 5, TpbpaCre n ¼5 and Cyp19-Cre n¼ 5. The scale bar represents 13 μm.

the trophectoderm-derived S-TGC and syncytiotrophoblasts. This results in the migration, proliferation and differentiation of pericytes to form the characteristic core-like structures around

which the fetal capillaries are organized. Thus the placental defect in Cited2ΔlacZ/ΔlacZ and conditionally deleted placentas is non-cell autonomous, such that loss of Cited2 in S-TGC and/or



15

Fig. 11. Conditional deletion of Cited2 in the placenta affects PDGFB expression in S-TGC and SynT-II. Representative longitudinal cryosections of (A–C) Cited2ΔlacZ/ þ (Het), (D–F) Cited2ΔlacZ/ΔlacZ (Null), (G–I) Cited2ΔlacZ/flox; Tek-Cre þ , (J–L) Cited2ΔlacZ/flox; Tpbpa-Creþ and (M–O) Cited2ΔlacZ/flox; Cyp19-Creþ E14.5 placentas. (A,D,G.J,M) PDGFB immunostaining. (B,E,H,K,N) The same sections stained with MCT1. (C,F,I,L,O) Merged images showing PDGFB (green), MCT1 (red) and DAPI (blue) staining. SynT-I cells are indicated by arrows and S-TGC with asterisks. Total number of placentas examined: Cited2 þ / þ (not shown) n¼ 2, Cited2ΔlacZ/ þ n ¼3, Cited2ΔlacZ/ΔlacZ n¼ 5, Tek-Cre n¼ 5, TpbpaCre n ¼5 and Cyp19-Cre n¼ 5. The scale bar represents 13 μm.

syncytiotrophoblasts is likely to result in the observed reduction in pericyte formation through reduced PDGF signaling. In support of this model, mosaic deletion of PDGFB from 70% of VECs in the

placenta using the Tie1-Cre transgene results in a similar phenotype to complete PDGFB deletion (Bjarnegard et al., 2004), with enlarged maternal and embryonic vessel lumens, and reduced


16


Fig. 12. Summary of results of conditional deletion of Cited2 in the placenta. Schematic representation of the cellular organization in the labyrinthine layer of the placenta. The fetal blood vessels (FBV) contain nucleated embryonic blood cells (ebc; gray), are lined with vascular endothelial cells (VEC; gray) and are organized around cores of pericytes (PC, pink). The maternal sinusoids (MS) contain enucleated maternal blood cells (mbc; red) and are lined with sinusoidal trophoblast giant cells (S-TGC; yellow). Inbetween the VECs and S-TGC are two layers of syncytiotrophoblasts, with the SynT-I layer (blue) immediately adjacent to the S-TGC layer, and the SynT-II layer (green) immediately adjacent to the VEC layer. (B) Summary of the cells in which Cited2, MCT1, MCT4, PDGFRß and PDGFB are expressed; the cells in which the Cited2flox allele is deleted for each Cre allele; and the placental phenotype of each deletion allele.

numbers of pericytes. However it should be noted that the efficacy of the Tie1-Cre transgene was only tested using the R26R reporter line. As discussed below, the R26R reporter gene may be poorly expressed in the placenta, and thus in Bjarnegard et al. (2004) Tie1Cre may have also deleted PDGFB expression in other cells such as S-TGC or syncytiotrophoblasts. CITED2 is a transcriptional modulator, acting by direct binding to other proteins to positively (Braganca et al., 2003; Chou et al., 2006; Glenn and Maurer, 1999; Qu et al., 2007; Tien et al., 2004) or negatively (Bhattacharya et al., 1999; Lou et al., 2011; Wu et al., 2011) alter their transcriptional activity. Therefore it is possible that the CITED2 protein is acting as a transcriptional co-activator to regulate Pdgfb and Pdgfrß expression in S-TGC and syncytiotrophoblasts. Furthermore, it is possible that CITED2 is part of an autoregulatory loop in S-TGCs. Chhabra et al. (2012) showed that PDGF signaling suppresses Epo expression in S-TGC. In the absence of PDGF signaling, Epo levels in S-TGC increase, triggering ectopic hematopoiesis. Cited2 transcription can be activated by Epo signaling (Bakker et al., 2007), therefore conditions of reduced PDGF signaling in S-TGC might be expected to result in increased CITED2 protein expression. This could suggest that Cited2 has an additional role downstream of Epo in triggering hematopoiesis. Indeed it has been shown in mouse that Cited2 is required for hematopoietic stem cell (HSC) maintenance in both fetal liver and adult bone marrow (Chen et al., 2007; Du et al., 2012). This downstream role could also explain why we do not see an obvious increase in numbers of hematopoietic stem cells in placentas lacking Cited2, despite their reduced PDGFRß protein expression. Finally, it is interesting to note that S-TGCs express high levels of both PDGFRß and PDGFB. S-TGCs are secretory cells that produce other growth factors including placental lactogen II (Simmons et al., 2008). This might suggest there is an as yet unidentified cell autonomous role of PDGF signaling in S-TGC proliferation, migration or patterning. For example, PDGF signaling might be required for the recruitment of S-TGC to line the maternal sinuses, in a manner analogous to the recruitment of pericytes to the embryonic vasculature. Loss of Cited2 may cause reduced embryonic lactate levels and reduced embryonic growth Embryo weight was significantly reduced in Cited2ΔlacZ/ΔlacZ and conditionally deleted conceptuses. This suggests that the observed

placental phenotype culminated in placental insufficiency. Deletion of Cited2 in the placenta lead to a loss of MCT4 cell-surface expression in SynT-II cells, whereas GLUT1 expression in SynT-II cells and MCT1 expression in SynT-I cells was unaffected. MCT1 and MCT4 are the major lactate transporters expressed in the murine placenta while GLUT1 is a glucose transporter (Nagai et al., 2010). In the mouse, the embryonic circulation is high in lactate and maternal circulation is low in lactate, therefore these proteins may facilitate transport from mother to embryo (Nagai et al., 2010). Lactate in the embryo can be used for both fatty acid synthesis and as a metabolic energy source. Therefore the loss of cell-surface expression of MCT4 may well result in reduced embryonic blood lactate, and be at least partially responsible for the observed reduction in embryonic growth. However it would be technically challenging to accurately measure blood lactate levels in mid-gestation mouse embryos to prove this hypothesis. Alternatively, this issue could be addressed by investigation of the placental phenotype of embryos with targeted deletion of MCT4. The observed differences in severity of placental phenotype (i.e. poor vascular patterning and fewer and disorganized pericytes) at E14.5 are likely to be reflected in the differences in the time of onset of reduced placental weight (Fig. 2). Cited2ΔlacZ/ΔlacZ has significantly lighter placental weights from E12.5; the Cyp19-Cre deleted placentas from E14.5; but the Tek-Cre and Tpbpa-Cre deleted placentas are not significantly smaller until E16.5. The Cited2flox allele is more broadly deleted than the R26R allele using the same Cre transgene In the course of these studies, we determined unexpected differences between the extent of Cited2flox and Z/EG reporter gene deletion and that previously published using the same Cre transgenes but a different reporter gene (R26R). There are two possible explanations for these observations. Firstly, the Cited2flox and Z/EG alleles may be more readily deleted than the R26R allele. Distinct deletion efficiencies between different floxed alleles have been previously reported (Vooijs et al., 2001), and may reflect differences in chromatin state between loci that might impede access of Cre protein to the loxP sites. Alternatively, the Cited2 and pCAGGS promoters may activate transcription more strongly and/or in a greater proportion of placental cells than the R26R promoter. Our findings are an important lesson for all conditional deletion studies. It emphasizes



the need for careful confirmation of the extent of deletion using any particular combination of floxed and Cre alleles, rather than relying on published studies of different tissues to those being investigated. In addition, these results may suggest a re-evaluation of the conclusions of other studies where deletion of a gene in the placenta has been performed using one or more of these Cre transgenes. In conclusion, we believe that Cited2 has an important role in the patterning of the developing embryonic vasculature of the placenta. This involves a non-cell autonomous action on pericyte recruitment, proliferation and/or differentiation, and is most likely to be mediated by regulation of PDGF signaling. Interestingly, in this and previous (Withington et al., 2006) studies we have shown that only the embryonic capillaries of the placenta, and not the large- or medium-sized placental vessels, are affected by total or partial loss of placental Cited2 expression. A similar result has recently been reported for the embryonic vasculature: deletion of a different Cited2flox allele from all cardiac progenitors using a heart-specific Nkx2-5-Cre allele disrupted the patterning of coronary vessels (MacDonald et al., 2013). Here too the number of large vessels was unchanged, but there were almost half the number of capillaries present. This suggests that Cited2 may have a broader role in regulating capillary patterning throughout the developing conceptus.

Acknowledgments We wish to thank James C. Cross for providing mouse lines; Joelene Major and Kavitha Iyer for genotyping; BioCORE staff; and Herbert Smith for generously donating the confocal microscope used in this study. The work was supported by Australian National Health and Medical Research Council Senior Research Fellowships 514900 and 1042002 and the Australian Research Council Discovery Grants DP303705 and DP0346729.

Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ydbio.2014.04.023. References Adamson, S.L., Lu, Y., Whiteley, K.J., Holmyard, D., Hemberger, M., Pfarrer, C., Cross, J.C., 2002. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev. Biol. 250, 358–373. Arora, R., Papaioannou, V.E., 2012. The murine allantois: a model system for the study of blood vessel formation. Blood 120, 2562–2572. Bakker, W.J., Harris, I.S., Mak, T.W., 2007. FOXO3a is activated in response to hypoxic stress and inhibits HIF1-induced apoptosis via regulation of CITED2. Mol. Cell 28, 941–953. Bamforth, S.D., Braganca, J., Eloranta, J.J., Murdoch, J.N., Marques, F.I., Kranc, K.R., Farza, H., Henderson, D.J., Hurst, H.C., Bhattacharya, S., 2001. Cardiac malformations, adrenal agenesis, neural crest defects and exencephaly in mice lacking Cited2, a new Tfap2 co-activator. Nat. Genet. 29, 469–474. Bamforth, S.D., Braganca, J., Farthing, C.R., Schneider, J.E., Broadbent, C., Michell, A. C., Clarke, K., Neubauer, S., Norris, D., Brown, N.A., Anderson, R.H., Bhattacharya, S., 2004. Cited2 controls left-right patterning and heart development through a Nodal-Pitx2c pathway. Nat. Genet. 36, 1189–1196. Barbera, J.P., Rodriguez, T.A., Greene, N.D., Weninger, W.J., Simeone, A., Copp, A.J., Beddington, R.S., Dunwoodie, S., 2002. Folic acid prevents exencephaly in Cited2 deficient mice. Hum. Mol. Genet. 11, 283–293. Bhattacharya, S., Michels, C.L., Leung, M.K., Arany, Z.P., Kung, A.L., Livingston, D.M., 1999. Functional role of p35srj, a novel p300/CBP binding protein, during transactivation by HIF-1. Genes Dev. 13, 64–75. Bjarnegard, M., Enge, M., Norlin, J., Gustafsdottir, S., Fredriksson, S., Abramsson, A., Takemoto, M., Gustafsson, E., Fassler, R., Betsholtz, C., 2004. Endotheliumspecific ablation of PDGFB leads to pericyte loss and glomerular, cardiac and placental abnormalities. Development 131, 1847–1857. Braganca, J., Eloranta, J.J., Bamforth, S.D., Ibbitt, J.C., Hurst, H.C., Bhattacharya, S., 2003. Physical and functional interactions among AP-2 transcription factors, p300/CREB-binding protein, and CITED2. J. Biol. Chem. 278, 16021–16029.

17

Chen, Y., Carlson, E.C., Chen, Z.Y., Hamik, A., Jain, M.K., Dunwoodie, S.L., Yang, Y.C., 2009. Conditional deletion of Cited2 results in defective corneal epithelial morphogenesis and maintenance. Dev. Biol. 334, 243–252. Chen, Y., Doughman, Y.Q., Gu, S., Jarrell, A., Aota, S., Cvekl, A., Watanabe, M., Dunwoodie, S.L., Johnson, R.S., van Heyningen, V., Kleinjan, D.A., Beebe, D.C., Yang, Y.C., 2008. Cited2 is required for the proper formation of the hyaloid vasculature and for lens morphogenesis. Development 135, 2939–2948. Chen, Y., Haviernik, P., Bunting, K.D., Yang, Y.C., 2007. Cited2 is required for normal hematopoiesis in the murine fetal liver. Blood 110, 2889–2898. Chhabra, A., Lechner, A.J., Ueno, M., Acharya, A., Van Handel, B., Wang, Y., IruelaArispe, M.L., Tallquist, M.D., Mikkola, H.K., 2012. Trophoblasts regulate the placental hematopoietic niche through PDGF-B signaling. Dev. Cell 22, 651–659. Chou, Y.T., Wang, H., Chen, Y., Danielpour, D., Yang, Y.C., 2006. Cited2 modulates TGF-beta-mediated upregulation of MMP9. Oncogene 25, 5547–5560. Chung, A.S., Ferrara, N., 2011. Developmental and pathological angiogenesis. Annu. Rev. Cell Dev. Biol. 27, 563–584. Colombo, S., Kumasaka, M., Lobe, C., Larue, L., 2010. Genomic localization of the Z/EG transgene in the mouse genome. Genesis 48, 96–100. Constien, R., Forde, A., Liliensiek, B., Grone, H.J., Nawroth, P., Hammerling, G., Arnold, B., 2001. Characterization of a novel EGFP reporter mouse to monitor Cre recombination as demonstrated by a Tie2 Cre mouse line. Genesis 30, 36–44. Du, J., Chen, Y., Li, Q., Han, X., Cheng, C., Wang, Z., Danielpour, D., Dunwoodie, S.L., Bunting, K.D., Yang, Y.C., 2012. HIF-1alpha deletion partially rescues defects of hematopoietic stem cell quiescence caused by Cited2 deficiency. Blood 119, 2789–2798. Du, J., Li, Q., Tang, F., Puchowitz, M.A., Fujioka, H., Dunwoodie, S.L., Danielpour, D., Yang, Y.C., 2014. Cited2 is required for the maintenance of glycolytic metabolism in adult hematopoietic stem cells. Stem Cells Dev. 23, 83–94. Dunwoodie, S.L., Rodriguez, T.A., Beddington, R.S., 1998. Msg1 and Mrg1, founding members of a gene family, show distinct patterns of gene expression during mouse embryogenesis. Mech. Dev. 72, 27–40. Enders, A.C., 1965. A comparative study of the fine structure of the trophoblast in several hemochorial placentas. Am. J. Anat. 116, 29–67. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O'Shea, K.S., PowellBraxton, L., Hillan, K.J., Moore, M.W., 1996. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380, 439–442. Glenn, D.J., Maurer, R.A., 1999. MRG1 binds to the LIM domain of Lhx2 and may function as a coactivator to stimulate glycoprotein hormone alpha-subunit gene expression. J. Biol. Chem. 274, 36159–36167. Hernandez-Verdun, D., 1974. Morphogenesis of the syncytium in the mouse placenta. Ultrastructural study. Cell Tissue Res. 148, 381–396. Hu, D., Cross, J.C., 2011. Ablation of Tpbpa-positive trophoblast precursors leads to defects in maternal spiral artery remodeling in the mouse placenta. Dev. Biol. 358, 231–239. Jain, R.K., 2003. Molecular regulation of vessel maturation. Nat. Med. 9, 685–693. John, R., Hemberger, M., 2012. A placenta for life. Reprod. Biomed. Online 25, 5–11. Koni, P.A., Joshi, S.K., Temann, U.A., Olson, D., Burkly, L., Flavell, R.A., 2001. Conditional vascular cell adhesion molecule 1 deletion in mice: impaired lymphocyte migration to bone marrow. J. Exp. Med. 193, 741–754. Kranc, K.R., Schepers, H., Rodrigues, N.P., Bamforth, S., Villadsen, E., Ferry, H., Bouriez-Jones, T., Sigvardsson, M., Bhattacharya, S., Jacobsen, S.E., Enver, T., 2009. Cited2 is an essential regulator of adult hematopoietic stem cells. Cell Stem Cell 5, 659–665. Leveen, P., Pekny, M., Gebre-Medhin, S., Swolin, B., Larsson, E., Betsholtz, C., 1994. Mice deficient for PDGF B show renal, cardiovascular, and hematological abnormalities. Genes Dev. 8, 1875–1887. Li, Q., Ramirez-Bergeron, D.L., Dunwoodie, S.L., Yang, Y.C., 2012. Cited2 gene controls pluripotency and cardiomyocyte differentiation of murine embryonic stem cells through Oct4 gene. J. Biol. Chem. 287, 29088–29100. Lopes Floro, K., Artap, S.T., Preis, J.I., Fatkin, D., Chapman, G., Furtado, M.B., Harvey, R.P., Hamada, H., Sparrow, D.B., Dunwoodie, S.L., 2011. Loss of Cited2 causes congenital heart disease by perturbing left-right patterning of the body axis. Hum. Mol. Genet. 20, 1097–1110. Lou, X., Sun, S., Chen, W., Zhou, Y., Huang, Y., Liu, X., Shan, Y., Wang, C., 2011. Negative feedback regulation of NF-kappaB action by CITED2 in the nucleus. J. Immunol. 186, 539–548. MacDonald, S.T., Bamforth, S.D., Braganca, J., Chen, C.M., Broadbent, C., Schneider, J.E., Schwartz, R.J., Bhattacharya, S., 2013. A cell-autonomous role of Cited2 in controlling myocardial and coronary vascular development. Eur. Heart J. 34, 2557–2565. Mobasheri, A., Oukrif, D., Dawodu, S.P., Sinha, M., Greenwell, P., Stewart, D., Djamgoz, M.B., Foster, C.S., Martin-Vasallo, P., Mobasheri, R., 2001. Isoforms of Naþ, K þ -ATPase in human prostate; specificity of expression and apical membrane polarization. Histol. Histopathol. 16, 141–154. Nagai, A., Takebe, K., Nio-Kobayashi, J., Takahashi-Iwanaga, H., Iwanaga, T., 2010. Cellular expression of the monocarboxylate transporter (MCT) family in the placenta of mice. Placenta 31, 126–133. Novak, A., Guo, C., Yang, W., Nagy, A., Lobe, C.G., 2000. Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cremediated excision. Genesis 28, 147–155. Ohlsson, R., Falck, P., Hellstrom, M., Lindahl, P., Bostrom, H., Franklin, G., AhrlundRichter, L., Pollard, J., Soriano, P., Betsholtz, C., 1999. PDGFB regulates the development of the labyrinthine layer of the mouse fetal placenta. Dev. Biol. 212, 124–136. Preis, J.I., Wise, N., Solloway, M.J., Harvey, R.P., Sparrow, D.B., Dunwoodie, S.L., 2006. Generation of conditional Cited2 null alleles. Genesis 44, 579–583.


18


Pritsker, M., Ford, N.R., Jenq, H.T., Lemischka, I.R., 2006. Genomewide gain-offunction genetic screen identifies functionally active genes in mouse embryonic stem cells. Proc. Natl. Acad. Sci. USA 103, 6946–6951. Qu, X., Lam, E., Doughman, Y.Q., Chen, Y., Chou, Y.T., Lam, M., Turakhia, M., Dunwoodie, S.L., Watanabe, M., Xu, B., Duncan, S.A., Yang, Y.C., 2007. Cited2, a coactivator of HNF4alpha, is essential for liver development. EMBO J. 26, 4445–4456. Ribatti, D., Nico, B., Crivellato, E., 2011. The role of pericytes in angiogenesis. Int. J. Dev. Biol. 55, 261–268. Rodriguez, T.A., Sparrow, D.B., Scott, A.N., Withington, S.L., Preis, J.I., Michalicek, J., Clements, M., Tsang, T.E., Shioda, T., Beddington, R.S., Dunwoodie, S.L., 2004. Cited1 is required in trophoblasts for placental development and for embryo growth and survival. Mol. Cell Biol. 24, 228–244. Sakai, M., Matsumoto, M., Tujimura, T., Yongheng, C., Noguchi, T., Inagaki, K., Inoue, H., Hosooka, T., Takazawa, K., Kido, Y., Yasuda, K., Hiramatsu, R., Matsuki, Y., Kasuga, M., 2012. CITED2 links hormonal signaling to PGC-1alpha acetylation in the regulation of gluconeogenesis. Nat. Med. 18, 612–617. Shalaby, F., Rossant, J., Yamaguchi, T.P., Gertsenstein, M., Wu, X.F., Breitman, M.L., Schuh, A.C., 1995. Failure of blood-island formation and vasculogenesis in Flk1-deficient mice. Nature 376, 62–66. Simmons, D.G., Fortier, A.L., Cross, J.C., 2007. Diverse subtypes and developmental origins of trophoblast giant cells in the mouse placenta. Dev. Biol. 304, 567–578. Simmons, D.G., Natale, D.R., Begay, V., Hughes, M., Leutz, A., Cross, J.C., 2008. Early patterning of the chorion leads to the trilaminar trophoblast cell structure in the placental labyrinth. Development 135, 2083–2091. Soriano, P., 1994. Abnormal kidney development and hematological disorders in PDGF beta-receptor mutant mice. Genes Dev. 8, 1888–1896. Soriano, P., 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71. Tang, N., Wang, L., Esko, J., Giordano, F.J., Huang, Y., Gerber, H.P., Ferrara, N., Johnson, R.S., 2004. Loss of HIF-1alpha in endothelial cells disrupts a hypoxia-driven VEGF autocrine loop necessary for tumorigenesis. Cancer Cell. 6, 485–495.

Tien, E.S., Davis, J.W., Vanden Heuvel, J.P., 2004. Identification of the CREB-binding protein/p300-interacting protein CITED2 as a peroxisome proliferator-activated receptor alpha coregulator. J. Biol. Chem. 279, 24053–24063. Viegas, M.S., Martins, T.C., Seco, F., do Carmo, A., 2007. An improved and costeffective methodology for the reduction of autofluorescence in direct immunofluorescence studies on formalin-fixed paraffin-embedded tissues. Eur. J. Histochem. 51, 59–66. Vooijs, M., Jonkers, J., Berns, A., 2001. A highly efficient ligand-regulated Cre recombinase mouse line shows that LoxP recombination is position dependent. EMBO Rep. 2, 292–297. Weninger, W.J., Lopes Floro, K., Bennett, M.B., Withington, S.L., Preis, J.I., Barbera, J. P., Mohun, T.J., Dunwoodie, S.L., 2005. Cited2 is required both for heart morphogenesis and establishment of the left-right axis in mouse development. Development 132, 1337–1348. Wenzel, P.L., Leone, G., 2007. Expression of Cre recombinase in early diploid trophoblast cells of the mouse placenta. Genesis 45, 129–134. Withington, S.L., Scott, A.N., Saunders, D.N., Lopes Floro, K., Preis, J.I., Michalicek, J., Maclean, K., Sparrow, D.B., Barbera, J.P., Dunwoodie, S.L., 2006. Loss of Cited2 affects trophoblast formation and vascularization of the mouse placenta. Dev. Biol. 294, 67–82. Wu, Z.Z., Sun, N.K., Chao, C.C., 2011. Knockdown of CITED2 using short-hairpin RNA sensitizes cancer cells to cisplatin through stabilization of p53 and enhancement of p53-dependent apoptosis. J. Cell Physiol. 226, 2415–2428. Xu, B., Qu, X., Gu, S., Doughman, Y.Q., Watanabe, M., Dunwoodie, S.L., Yang, Y.C., 2008. Cited2 is required for fetal lung maturation. Dev. Biol. 317, 95–105. Yin, Z., Haynie, J., Yang, X., Han, B., Kiatchoosakun, S., Restivo, J., Yuan, S., Prabhakar, N.R., Herrup, K., Conlon, R.A., Hoit, B.D., Watanabe, M., Yang, Y.C., 2002. The essential role of Cited2, a negative regulator for HIF-1alpha, in heart development and neurulation. Proc. Natl. Acad. Sci. USA 99, 10488–10493. Yokota, H., Goldring, M.B., Sun, H.B., 2003. CITED2-mediated regulation of MMP-1 and MMP-13 in human chondrocytes under flow shear. J. Biol. Chem. 278, 47275–47280.


Emx1 Is Required for Neocortical Area Patterning.

SUV3 helicase is required for correct processing of mitochondrial transcripts.

PAX7 is required for patterning the esophageal musculature.

A serum-free system for culturing human placental trophoblasts.

Hedgehog signaling through GLI1 and GLI2 is required for epithelial-mesenchymal transition in human trophoblasts.

The impact of ionizing radiation on placental trophoblasts.

Gtpbp2 is required for BMP signaling and mesoderm patterning in Xenopus embryos.

Amontillado is required for Drosophila Slit processing and for tendon-mediated muscle patterning.

SABRE is required for stabilization of root hair patterning in Arabidopsis thaliana.

Arabidopsis homolog of trithorax1 (ATX1) is required for cell production, patterning, and morphogenesis in root development.

Integration of hormonal signaling networks and mobile microRNAs is required for vascular patterning in Arabidopsis roots.

7 signaling is required for anterior skeletal patterning in sea urchin embryos.

Hapln1a is required for connexin43-dependent growth and patterning in the regenerating fin skeleton.

Down-regulation of PDK4 is Critical for the Switch of Carbohydrate Catabolism during Syncytialization of Human Placental Trophoblasts.

Balanced interhemispheric cortical activity is required for correct targeting of the corpus callosum.

Correct folding of alpha-lytic protease is required for its extracellular secretion from Escherichia coli.

Intestinal smooth muscle is required for patterning the enteric nervous system.

The brain is required for normal muscle and nerve patterning during early Xenopus development.

The Arabidopsis Exine Formation Defect (EFD) gene is required for primexine patterning and is critical for pollen fertility.

Reactive Oxygen Species in Planarian Regeneration: An Upstream Necessity for Correct Patterning and Brain Formation.

Primary Human Placental Trophoblasts are Permissive for Zika Virus (ZIKV) Replication.

Placental epigenetic patterning of glucocorticoid response genes is associated with infant neurodevelopment.

Sonic hedgehog through Gli2 and Gli3 is required for the proper development of placental labyrinth.

Purified first and third trimester placental trophoblasts differ in in vitro hormone secretion.