*2.3. EC-Specific Deletion of Jag1 in Early Pregnancy*

*2.3. EC-Specific Deletion of Jag1 in Early Pregnancy*  To determine the role of endothelial Jag1 in the decidual vasculature in early pregnancy, we used a tamoxifen inducible driver, *Cdh5-CreERT2*, to express Cre recombinase in the endothelium and delete *Jag1* during decidual angiogenesis. Tamoxifen was administered at embryo implantation, E4.5, and pregnancies and uterine phenotypes were assessed at E7.5, the end of the active period of decidual angiogenesis (Figure 4A). Notch proteins and ligands are expressed in the ovarian endothelium, including in ECs of corpora lutea [45] (Supplementary Figure S1). Given the requirement for progesterone secretion from ovarian corpora lutea to maintain early pregnancy, progesterone supplementation was initiated at E4.5 to overcome potential defects in *Jag1∆EC* mutant ovaries. To determine the recombination efficiency of the *Cdh5-CreERT2* line using this tamoxifen regimen, *Cdh5-CreERT2* mice were crossed to the *ROSA26 tdTomato* reporter strain (Figure 4B). Expression of fluorescent tdTomato protein throughout the implantation site is observed in a pattern similar to CD31 in the mesometrial region and anti-mesometrial region (Figure 1C). Implantation To determine the role of endothelial Jag1 in the decidual vasculature in early pregnancy, we used a tamoxifen inducible driver, *Cdh5-CreERT2*, to express Cre recombinase in the endothelium and delete *Jag1* during decidual angiogenesis. Tamoxifen was administered at embryo implantation, E4.5, and pregnancies and uterine phenotypes were assessed at E7.5, the end of the active period of decidual angiogenesis (Figure 4A). Notch proteins and ligands are expressed in the ovarian endothelium, including in ECs of corpora lutea [45] (Supplementary Figure S1). Given the requirement for progesterone secretion from ovarian corpora lutea to maintain early pregnancy, progesterone supplementation was initiated at E4.5 to overcome potential defects in *Jag1*∆*EC* mutant ovaries. To determine the recombination efficiency of the *Cdh5-CreERT2* line using this tamoxifen regimen, *Cdh5-CreERT2* mice were crossed to the *ROSA26 tdTomato* reporter strain (Figure 4B). Expression of fluorescent tdTomato protein throughout the implantation site is observed in a pattern similar to CD31 in the mesometrial region and anti-mesometrial region (Figure 1C). Implantation sites were stained with endothelial cell marker, CD31, or mural cell marker, NG2, to assess tdTomato expression

with respect to ECs and mural cells. Co-expression is observed with CD31<sup>+</sup> ECs, but not with NG2<sup>+</sup> mural cells (Figure S2). tdTomato is expressed in 94–98% of CD31<sup>+</sup> ECs in the mesometrial and anti-mesometrial regions and in 90% of NG2<sup>+</sup> SpAs in the mesometrial region (Figure S2). These data demonstrate efficient recombination in decidual ECs of capillaries and SpAs at E7.5. sites were stained with endothelial cell marker, CD31, or mural cell marker, NG2, to assess tdTomato expression with respect to ECs and mural cells. Co-expression is observed with CD31+ ECs, but not with NG2+ mural cells (Figure S2). tdTomato is expressed in 94–98% of CD31+ ECs in the mesometrial and anti-mesometrial regions and in 90% of NG2+ SpAs in the mesometrial region (Figure S2). These data demonstrate efficient recombination in decidual ECs of capillaries and SpAs at E7.5.

*Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 7 of 19

**Figure 4.** Tamoxifen-induced Cre recombination decreases expression of Jag1 in *Cdh5-CreERT2;Jag1flox/flox* (*Jag1ΔEC*) pregnancy. (**A**) Schematic of the mouse gestation timeline highlighting key processes involved, and experimental timeline (created with BioRender.com). At E4.5, tamoxifen is administered to induce Cre recombination and progesterone was administered to minimize secondary angiogenic defects in the uterus due to vascular defects in the ovaries. Pregnancies and uterine phenotype were assessed at E7.5. (**B**) Representative image of an implantation site at E7.5 from the *Cdh5-CreERT2*; *ROSA26 tdTomato* reporter. The decidua is within the dashed white. tdTomato expression, indicating Cre-induced recombination, is detected in a vascular pattern throughout the decidua. Dashed line denotes MR and AMR. (**C**,**D**) Representative images of implantation sites of *Cdh5-CreERT2* control and *Jag1ΔEC* pregnancies stained for Jag1 and CD31. Expression of Jag1 is reduced in the decidua (outlined with dashed white lines), but not the embryo of *Jag1ΔEC* mutants. (**E**) qRT-PCR determination of *Jag1* expression in implantation sites from control (*n* = 6) and *Jag1ΔEC*  (*n* = 5) pregnancies. The relative expression level of *Jag1* was compared to β-actin. *Jag1* was significantly decreased in *Jag1ΔEC* mutants as compared to *Cdh5-CreERT2* controls. (**F**) Representative images of the MP of control and *Jag1ΔEC* implantation sites double stained for EC marker CD31 and Jag1. At the MP, expression of Jag1 in all CD31+ ECs is reduced in *Jag1ΔEC* mutants (*n* = 6) as compared **Figure 4.** Tamoxifen-induced Cre recombination decreases expression of Jag1 in *Cdh5-CreERT2;Jag1flox*/*flox* (*Jag1*∆*EC*) pregnancy. (**A**) Schematic of the mouse gestation timeline highlighting key processes involved, and experimental timeline (created with BioRender.com). At E4.5, tamoxifen is administered to induce Cre recombination and progesterone was administered to minimize secondary angiogenic defects in the uterus due to vascular defects in the ovaries. Pregnancies and uterine phenotype were assessed at E7.5. (**B**) Representative image of an implantation site at E7.5 from the *Cdh5-CreERT2*; *ROSA26 tdTomato* reporter. The decidua is within the dashed white. tdTomato expression, indicating Cre-induced recombination, is detected in a vascular pattern throughout the decidua. Dashed line denotes MR and AMR. (**C**,**D**) Representative images of implantation sites of *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies stained for Jag1 and CD31. Expression of Jag1 is reduced in the decidua (outlined with dashed white lines), but not the embryo of *Jag1*∆*EC* mutants. (**E**) qRT-PCR determination of *Jag1* expression in implantation sites from control (*n* = 6) and *Jag1*∆*EC* (*n* = 5) pregnancies. The relative expression level of *Jag1* was compared to β-actin. *Jag1* was significantly decreased in *Jag1*∆*EC* mutants as compared to *Cdh5-CreERT2* controls. (**F**) Representative images of the MP of control and *Jag1*∆*EC* implantation sites double stained for EC marker CD31 and Jag1. At the MP, expression of Jag1 in all CD31<sup>+</sup> ECs is reduced in *Jag1*∆*EC* mutants (*n* = 6) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**G**) Representative images of the MP of control and *Jag1*∆*EC* implantation sites double stained for Jag1

and NG2. Expression of Jag1 in the ECs of the NG2<sup>+</sup> surrounded SpAs (white arrowheads) is reduced in *Jag1*∆*EC* mutants (*n* = 5) as compared to *Cdh5-CreERT2* controls (*n* = 6). AMR= anti-mesometrial region; CTR = central region; e = embryo; MP = mesometrial pole. Scale bars = 100 µm in (**B**–**D**,**F**,**G**). Data shown as median + IQR; \*\* *p* < 0.01. to *Cdh5-CreERT2* controls (*n* = 5). (**G**) Representative images of the MP of control and *Jag1ΔEC* implantation sites double stained for Jag1 and NG2. Expression of Jag1 in the ECs of the NG2+ surrounded SpAs (white arrowheads) is reduced in *Jag1ΔEC* mutants (*n* = 5) as compared to *Cdh5- CreERT2* controls (*n* = 6). AMR= anti-mesometrial region; CTR = central region; e = embryo; MP =

*Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 8 of 19

To evaluate the role of endothelial Jag1 in post-implantation, pre-placentation pregnancy, uteri from pregnant *Jag1*∆*EC* and *Cdh5-CreERT*<sup>2</sup> control females were isolated at E7.5. A decrease in Jag1 is observed in the decidua ECs of *Jag1*∆*EC* uteri as compared to controls (Figure 4C,D). In contrast, analysis of the uteri from *Jag1*∆*EC* pregnancies demonstrates similar expression of Jag1 in embryos of control and *Jag1*∆*EC* mutants. qRT-PCR performed on whole implantation sites shows a significant decrease in *Jag1* expression in *Jag1*∆*EC* mutants as compared to controls (Figure 4E). To quantify Jag1 expression in ECs and SpAs at the mesometrial pole, implantation sites were stained to detect Jag1 in CD31<sup>+</sup> ECs and Jag1 expressed in NG2<sup>+</sup> SpAs (Figure 4F,G). Expression of Jag1 is significantly lower in all CD31<sup>+</sup> ECs (Figure 4F,G) in *Jag1*∆*EC* pregnancies when compared to controls, demonstrating similar efficacy of *Cdh5-CreERT2 -*mediated deletion of *Jag1* in capillary and SpA ECs, as seen for the *ROSA26 tdTomato* reporter, and confirming that our model systems works. mesometrial pole. Scale bars = 100 μm in (**B**–**D**,**F**,**G**). Data shown as median + IQR; \*\* *p* < 0.01. To evaluate the role of endothelial Jag1 in post-implantation, pre-placentation pregnancy, uteri from pregnant *Jag1∆EC* and *Cdh5-CreERT*2 control females were isolated at E7.5. A decrease in Jag1 is observed in the decidua ECs of *Jag1ΔEC* uteri as compared to controls (Figure 4C,D). In contrast, analysis of the uteri from *Jag1*Δ*EC* pregnancies demonstrates similar expression of Jag1 in embryos of control and *Jag1*Δ*EC* mutants. qRT-PCR performed on whole implantation sites shows a significant decrease in *Jag1* expression in *Jag1*Δ*EC* mutants as compared to controls (Figure 4E). To quantify Jag1 expression in ECs and SpAs at the mesometrial pole, implantation sites were stained to detect Jag1 in CD31+ ECs and Jag1 expressed in NG2+ SpAs (Figures 4F,G). Expression of Jag1 is significantly lower in all CD31+ ECs (Figure 4F,G) in *Jag1ΔEC* pregnancies when compared to controls, demonstrating similar efficacy of *Cdh5-CreERT2-*mediated deletion of *Jag1* in capillary and SpA ECs, as

#### *2.4. Loss of Endothelial Jag1 Does Not Impact Pregnancy at E7.5* seen for the *ROSA26 tdTomato* reporter, and confirming that our model systems works.

progressed to a similar stage of development (Figure 5D).

To determine if EC-specific loss of *Jag1* impacts pregnancy and embryo development at E7.5, implantation sites were collected and counted, and embryo morphology was assessed in H&E stained sections. Litter size for *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies is similar (Figure 5A). Embryo development involves the formation and disappearance of morphologic structures during gestation [46]. To determine embryo morphology with respect to embryonic age, all tissue sections containing embryos were scored for the presence or absence of the following morphologic structures: primitive streak, allantois, cranial neural fold and somites (Figure 5B,C). These morphologic "landmarks" are common to gastrulating wild-type embryos between E6.5 and E8.5, and the timing of their appearance is used to compare stage of embryo development. Comparison of *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies revealed that embryos from *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies progressed to a similar stage of development (Figure 5D). *2.4. Loss of Endothelial Jag1 Does Not Impact Pregnancy at E7.5*  To determine if EC-specific loss of *Jag1* impacts pregnancy and embryo development at E7.5, implantation sites were collected and counted, and embryo morphology was assessed in H&E stained sections. Litter size for *Cdh5-CreERT2* control and *Jag1*Δ*EC* pregnancies is similar (Figure 5A). Embryo development involves the formation and disappearance of morphologic structures during gestation [46]. To determine embryo morphology with respect to embryonic age, all tissue sections containing embryos were scored for the presence or absence of the following morphologic structures: primitive streak, allantois, cranial neural fold and somites (Figures 5B,C). These morphologic "landmarks" are common to gastrulating wild-type embryos between E6.5 and E8.5, and the timing of their appearance is used to compare stage of embryo development. Comparison of *Cdh5-CreERT2* control and *Jag1*Δ*EC* pregnancies revealed that embryos from *Cdh5-CreERT2* control and *Jag1*Δ*EC* pregnancies

**Figure 5.** EC-specific loss of *Jag1* does not disrupt pregnancy progression or litter size at E7.5. (**A**) Number of implantation sites at E7.5 is similar between controls (*n* = 8) and *Jag1*∆*EC* pregnancies

(*n* = 7). (**B**,**C**) Representative H&E stained embryos, showing morphologic structures including the primitive streak (arrowheads), cranial neural fold (red asterisk), allantois (yellow asterisk) and somites (white arrows), observed in embryos at E7.5 (**B**) and E8.5 (**C**). (**D**) The presence of the primitive streak, allantois and cranial neural fold structures is similar in control and *Jag1*∆*EC* pregnancies. Data were analyzed using Fisher's exact test, comparing morphologic structures in control and *Jag1*∆*EC* pregnancies to those expected in pregnancies at E7.5 [46]. Number of implantation sites at E7.5 is similar between controls (*n* = 8) and *Jag1ΔEC* pregnancies (*n*  = 7). (**B**,**C**) Representative H&E stained embryos, showing morphologic structures including the primitive streak (arrowheads), cranial neural fold (red asterisk), allantois (yellow asterisk) and somites (white arrows), observed in embryos at E7.5 (**B**) and E8.5 (**C**). (**D**) The presence of the primitive streak, allantois and cranial neural fold structures is similar in control and *Jag1ΔEC* pregnancies. Data were analyzed using Fisher's exact test, comparing morphologic structures in control and *Jag1ΔEC* pregnancies to those expected in pregnancies at E7.5 [46].

**Figure 5.** EC-specific loss of *Jag1* does not disrupt pregnancy progression or litter size at E7.5. (**A**)

#### *2.5. Expression of Notch Ligand, Dll4, and Notch E*ff*ectors Is Increased in Jag1*∆*EC Pregnancies at E7.5 2.5. Expression of Notch ligand, Dll4, and Notch Effectors Is Increased in Jag1∆EC Pregnancies at E7.5*

To assess the impact of loss of endothelial *Jag1* on expression of Dll4, *Jag1*∆*EC* and *Cdh5-CreERT2* uteri were double stained for CD31 or NG2, and Dll4. The percentage of CD31<sup>+</sup> ECs expressing Dll4 was quantified in the mesometrial and anti-mesometrial regions. At the mesometrial pole and in the anti-mesometrial region, Dll4 expression is significantly increased in the CD31<sup>+</sup> capillary ECs in *Jag1*∆*EC* pregnancies compared to control pregnancies (Figure 6A,B). Dll4 expression is increased but not significantly, in the ECs of the NG2<sup>+</sup> cell covered SpAs (Figure 6C). These data show that loss of *Jag1* in ECs leads to increased EC Dll4 expression. To understand the impact of EC-specific loss of *Jag1* and increased EC Dll4 expression on Notch signaling, we assessed the expression of downstream effectors of Notch signaling. Total RNA was isolated from *Jag1*∆*EC* mutant and control whole implantation sites, from which myometrium was removed. Expression of direct Notch target genes was determined by qRT-PCR. Expression of Notch effectors, *Hey2* and *Nrarp*, is significantly increased in *Jag1*∆*EC* mutants relative to control pregnancies. These data suggest increased Dll4/Notch signaling in the implantation sites of *Jag1*∆*EC* mutants (Figure 6D). To assess the impact of loss of endothelial *Jag1* on expression of Dll4, *Jag1*Δ*EC* and *Cdh5-CreERT2* uteri were double stained for CD31 or NG2, and Dll4. The percentage of CD31+ ECs expressing Dll4 was quantified in the mesometrial and anti-mesometrial regions. At the mesometrial pole and in the anti-mesometrial region, Dll4 expression is significantly increased in the CD31+ capillary ECs in *Jag1∆EC* pregnancies compared to control pregnancies (Figure 6A,B). Dll4 expression is increased but not significantly, in the ECs of the NG2+ cell covered SpAs (Figure 6C). These data show that loss of *Jag1* in ECs leads to increased EC Dll4 expression. To understand the impact of EC-specific loss of *Jag1* and increased EC Dll4 expression on Notch signaling, we assessed the expression of downstream effectors of Notch signaling. Total RNA was isolated from *Jag1*Δ*EC* mutant and control whole implantation sites, from which myometrium was removed. Expression of direct Notch target genes was determined by qRT-PCR. Expression of Notch effectors, *Hey2* and *Nrarp*, is significantly increased in *Jag1∆EC* mutants relative to control pregnancies. These data suggest increased Dll4/Notch signaling in the implantation sites of *Jag1∆EC* mutants (Figure 6D).

**Figure 6.** Expression of Notch ligand, Dll4, and Notch effectors are increased in ECs of *Jag1ΔEC* decidua. (**A**–**C**) Expression of Dll4 in capillaries and SpAs in *Cdh5-CreERT2* control and *Jag1∆EC* pregnancies was determined by double staining IF for Dll4 and CD31 or NG2. Representative images of the MP (**A**) and AMR (**B**) of control and *Jag1ΔEC* implantation sites. (**A**) Dll4 expression is increased in CD31+ ECs at the MP of *Jag1ΔEC* mutants (*n* = 4) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**B**) **Figure 6.** Expression of Notch ligand, Dll4, and Notch effectors are increased in ECs of *Jag1*∆*EC* decidua. (**A**–**C**) Expression of Dll4 in capillaries and SpAs in *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies was determined by double staining IF for Dll4 and CD31 or NG2. Representative images of the MP (**A**) andAMR (**B**) of control and *Jag1*∆*EC* implantation sites. (**A**) Dll4 expression is increased in CD31<sup>+</sup> ECs at the MP of *Jag1*∆*EC* mutants (*n* = 4) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**B**) Dll4 expression is

Dll4 expression is increased in CD31+ capillary ECs in the AMR of *Jag1ΔEC* mutants (*n* = 6) as compared to *Cdh5-CreERT2* controls (*n* = 6). (**C**) Representative images of the MP of control and *Jag1ΔEC* implantation sites double stained for Dll4 and NG2. Expression of Dll4 in the ECs of the NG2+ surrounded SpAs is unchanged in *Jag1ΔEC* mutants (*n* = 6) as compared to *Cdh5-CreERT2* controls (*n* =

increased in CD31<sup>+</sup> capillary ECs in the AMR of *Jag1*∆*EC* mutants (*n* = 6) as compared to *Cdh5-CreERT2* controls (*n* = 6). (**C**) Representative images of the MP of control and *Jag1*∆*EC* implantation sites double stained for Dll4 and NG2. Expression of Dll4 in the ECs of the NG2<sup>+</sup> surrounded SpAs is unchanged in *Jag1*∆*EC* mutants (*n* = 6) as compared to *Cdh5-CreERT2* controls (*n* = 6). (**D**) qRT-PCR determination of Notch effector gene expression in implantation sites from control (*n* = 5) and *Jag1*∆*EC* (*n* = 5) pregnancies. The relative expression level of each gene was compared to β-actin. *Nrarp* and *Hey2* are significantly increased in *Jag1*∆*EC* mutants as compared to *Cdh5-CreERT2* controls. AMR = anti-mesometrial region; MP = mesometrial pole. Scale bars = 50 µm. Data shown are median +IQR. \* *p* < 0.05, \*\* *p* < 0.01. *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 10 of 19 6). (**D**) qRT-PCR determination of Notch effector gene expression in implantation sites from control (*n* = 5) and *Jag1ΔEC* (*n* = 5) pregnancies. The relative expression level of each gene was compared to β-actin. *Nrarp* and *Hey2* are significantly increased in *Jag1ΔEC* mutants as compared to *Cdh5-CreERT2*  controls. AMR = anti-mesometrial region; MP = mesometrial pole. Scale bars = 50 μm. Data shown are median +IQR. \* *p* < 0.05, \*\* *p* < 0.01.

#### *2.6. Jag1*/*Notch Signaling Regulates Angiogenic Gene Expression and Endothelial Proliferation in the Anti-Mesometrial Decidua 2.6. Jag1/Notch Signaling Regulates Angiogenic Gene Expression and Endothelial Proliferation in the Anti-Mesometrial Decidua*

Dll4/Notch signaling has been shown in the retina to suppress VEGFR2 expression leading to reduced EC proliferation [17]. To evaluate the impact of EC-specific loss of *Jag1* on endothelial Dll4/Notch signaling, we determined the expression of Notch1 with an antibody against the cytoplasmic domain, the Notch1 ICD (N1ICD). We also determined expression of VEGFR2 and EC proliferation. *Jag1* inactivation increases nuclear expression of N1ICD in anti-mesometrial capillary ECs. This increase in nuclear Notch1 expression is consistent with an increase in expression of Notch1 protein and in EC Notch1 signaling in *Jag1*∆*EC* mutants (Figure 7A). We found N1ICD expression in cells adjacent to Dll4<sup>+</sup> cells in the capillaries of *Jag1*∆*EC* mutants, suggesting *trans*-activation of Notch1 by Dll4 (Figure 7B). The percentage of CD31<sup>+</sup> ECs in the anti-mesometrial decidua expressing Ki67 (Figure 7C) and VEGFR2 (Figure 7D) is significantly decreased in *Jag1*∆*EC* mutants relative to controls. In contrast, expression of VEGFR2 in the myometrium (Figure 7D, black arrowheads) is similar in *Jag1*∆*EC* mutants and controls. Together, the data show that increased Notch signaling is associated with decreased VEGFR2 expression and EC proliferation in the anti-mesometrial decidua. Dll4/Notch signaling has been shown in the retina to suppress VEGFR2 expression leading to reduced EC proliferation [17]. To evaluate the impact of EC-specific loss of *Jag1* on endothelial Dll4/Notch signaling, we determined the expression of Notch1 with an antibody against the cytoplasmic domain, the Notch1 ICD (N1ICD). We also determined expression of VEGFR2 and EC proliferation. *Jag1* inactivation increases nuclear expression of N1ICD in anti-mesometrial capillary ECs. This increase in nuclear Notch1 expression is consistent with an increase in expression of Notch1 protein and in EC Notch1 signaling in *Jag1∆EC* mutants (Figure 7A). We found N1ICD expression in cells adjacent to Dll4+ cells in the capillaries of *Jag1∆EC* mutants, suggesting *trans*-activation of Notch1 by Dll4 (Figure 7B). The percentage of CD31+ ECs in the anti-mesometrial decidua expressing Ki67 (Figure 7C) and VEGFR2 (Figure 7D) is significantly decreased in *Jag1∆EC* mutants relative to controls. In contrast, expression of VEGFR2 in the myometrium (Figure 7D, black arrowheads) is similar in *Jag1∆EC* mutants and controls. Together, the data show that increased Notch signaling is associated with decreased VEGFR2 expression and EC proliferation in the anti-mesometrial decidua.

**Figure 7.** Increased Notch signaling in the anti-mesometrial decidua decreases EC proliferation and expression of VEGFR2. (**A**–**C**) High magnification images of sections double stained to detect expression of N1ICD and CD31, Dll4 or Ki67. (**A**) Expression of the N1ICD in CD31+ ECs in the AMR of *Cdh5-CreERT2* control and *Jag1∆EC* pregnancies was determined by double staining IF for N1ICD and CD31. Representative images of control and *Jag1ΔEC* implantation sites are shown. Nuclear expression of N1ICD (white arrowheads) is increased in CD31+ ECs in the AMR of *Jag1ΔEC* mutants **Figure 7.** Increased Notch signaling in the anti-mesometrial decidua decreases EC proliferation and expression of VEGFR2. (**A**–**C**) High magnification images of sections double stained to detect expression of N1ICD and CD31, Dll4 or Ki67. (**A**) Expression of the N1ICD in CD31<sup>+</sup> ECs in the AMR of *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies was determined by double staining IF for N1ICD and CD31. Representative images of control and *Jag1*∆*EC* implantation sites are shown. Nuclear expression of N1ICD (white arrowheads) is increased in CD31<sup>+</sup> ECs in the AMR of *Jag1*∆*EC* mutants (*n* = 5) as

Ki67 expression (white arrowheads) is decreased in CD31+ ECs in the AMR of *Jag1ΔEC* mutants (*n* =

(*n* = 5) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**B**) N1ICD and Dll4 are expressed in adjacent cells

compared to *Cdh5-CreERT2* controls (*n* = 5). (**B**) N1ICD and Dll4 are expressed in adjacent cells in the AMR (white arrowheads) of *Jag1*∆*EC* pregnancies. (**C**) EC proliferation in the AMR of *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies was determined by double staining IF for EC proliferation marker, Ki67 and CD31. Representative images of control and *Jag1*∆*EC* implantation sites are shown. Ki67 expression (white arrowheads) is decreased in CD31<sup>+</sup> ECs in the AMR of *Jag1*∆*EC* mutants (*n* = 5) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**D**) Expression of VEGFR2 with respect to CD31<sup>+</sup> ECs in the AMR was determined by comparison of expression of VEGFR2 and CD31 in adjacent sections, in like-regions, of implantation sites from *Cdh5-CreERT2* control and *Jag1*∆*EC* pregnancies. Representative images of the AMR highlighting areas (decidua within dashed red lines) used to measure signal density from control and *Jag1*∆*EC* implantation sites are shown (i–iv). VEGFR2 expression is decreased in CD31<sup>+</sup> capillary ECs in the anti-mesometrial decidua of *Jag1*∆*EC* mutants (*n* = 5) as compared to *Cdh5-CreERT2* controls (*n* = 5). Insets highlight VEGFR2 expression in decidual vessels. Black arrowheads highlight VEGFR2 expression in the myometrial vessels, which is similar in *Jag1*∆*EC* mutants and controls. AMR = anti-mesometrial region; N1ICD = Notch1 intracellular domain. Scale bars = 50 µm. Data shown as median + IQR; \* *p* < 0.05, \*\* *p* < 0.01. *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 11 of 19 5) as compared to *Cdh5-CreERT2* controls (*n* = 5). (**D**) Expression of VEGFR2 with respect to CD31+ ECs in the AMR was determined by comparison of expression of VEGFR2 and CD31 in adjacent sections, in like-regions, of implantation sites from *Cdh5-CreERT2* control and *Jag1∆EC* pregnancies. Representative images of the AMR highlighting areas (decidua within dashed red lines) used to measure signal density from control and *Jag1ΔEC* implantation sites are shown (i–iv). VEGFR2 expression is decreased in CD31+ capillary ECs in the anti-mesometrial decidua of *Jag1ΔEC* mutants (*n* = 5) as compared to *Cdh5-CreERT2* controls (*n* = 5). Insets highlight VEGFR2 expression in decidual vessels. Black arrowheads highlight VEGFR2 expression in the myometrial vessels, which is similar in *Jag1ΔEC* mutants and controls. AMR = anti-mesometrial region; N1ICD = Notch1 intracellular domain. Scale bars = 50 μm. Data shown as median + IQR; \* *p* < 0.05, \*\* *p* < 0.01.
