Abstract

This is a study on the role of tuberous sclerosis complex1 (TSC1) mutation and mTOR activation in endothelial cells during angiogenic and embryonic development. Past studies had shown that Tsc1/Tsc2 mutant genes lead to overactivation of mTOR in the regulating pathways in developing fetus. We used conditional Cre-loxp gene knockout approach to delete Tsc1 in mice's endothelial cells in our experimental models. Similarly, activation of mTOR signaling in endothelial cells of these embryos (Tie2-Cre/Tsc1−/−) was found. Majority of Tie2-Cre/Tsc1−/− embryos died at embryonic day 14.5 in utero. Cardiovascular defects, subcutaneous edema and hemorrhage were present among them. Whole-mount immunostaining in these embryos revealed a disorganized vascular network, defective sprouting of vessels in yolk sac and thickening of the labyrinth layer in the placenta. A thinner ventricular wall with disorganized trabeculae was present in the hearts of Tie2-Cre/Tsc1−/− embryos. Endothelial cells in Tsc1-deficient mice showed defective mitochondrial and endoplasmic reticular morphology, but no significant change was observed in cell junctions. The mutant embryos displayed significantly reduced cell proliferation, increased apoptosis and disturbed expression of angiogenic factors. A cohort of mice was treated prenatally with mTOR inhibitor rapamycin. The offspring of these mutant mice survived up to 22 days after birth. It was concluded that physiological TSC1-mTOR signaling in endothelial cells is crucial for vascular development and embryogenesis. We postulated that disruption of normal angiogenic pathways through hyperactive mTOR signaling maybe the mechanism that lead to deranged vascular pathogenesis in the tuberous sclerosis complex.

INTRODUCTION

Tuberous sclerosis complex (TSC) is an autosomal-dominant hamartoma disorder (1). The clinical phenotypes of TSC are characterized by abnormal vascularization in multiple organs, including skin angiofibroma, retinal hamartomas, cardiac rhabdomyomas, pulmonary lymphangioleiomyomatosis (LAM), renal angiomyolipoma (AML) and liver hemangiomas (2). The underlying genetic etiology of TSC involves mutations in one of two tumor suppressor genes, TSC1 and TSC2, which encode the hamartin and tuberin proteins, respectively (3,4). The hamartin–tuberin complex functions as an inhibitor of a key regulator of cell growth and proliferation, mechanistic target of rapamycin (mTOR) (5–9). mTOR modulates anabolic processes that promote cell growth and is exquisitely sensitive to cellular growth conditions. TSC1-TSC2 complex turns off mTOR under perturbed cellular growth conditions. Potentiated by various growth stimuli, AKT inactivates TSC1-TSC2 complex and consequently the activated mTOR promotes a subset of protein synthesis (10).

Animal models have been established to investigate the pathogenic role of mTOR signaling in TSC (11). A homozygous Tsc1 knockout (KO) mouse model was found to be lethal at embryonic (E) days 10.5–11.5 (7). And a conditional KO model of the Tsc1 gene in neuroprogenitor cells in the brain based on a nestin promoter-driven Cre allele was shown to be lethal within 24 h after birth, and reduced pup–maternal interaction was implicated as the major cause of death (12). Although ventricular myocyte-specific deletion of Tsc1 does not result in embryonic lethality, these mice die at approximately postnatal month 6, showing severe myocyte hypertrophy, with morphology similar to that observed in human rhabdomyoma which is the first sign of fetal TSC (13). Thus, the expression of Tsc1 is essential for proper embryogenesis.

Aneurysms in the brain, chest and abdomen have been reported in TSC patients and are associated with increased mobility and mortality (14–16). AMLs are composed of anomalous vessels, immature smooth muscle cells and adipocytes (17). They are associated with a high risk of spontaneous hemorrhage, arterial hypertension and kidney failure. Currently, little is understood about the underlying vascular pathogenesis of TSC. To investigate whether malfunction of Tsc1 contributes to TSC pathogenesis through its influence on vascularization, possibly involving the PI3 K/Akt/mTOR signaling pathway, a vascular endothelial cell (EC)-specific Tsc1 KO mouse model was generated and analyzed.

RESULTS

Generation of endothelial Tsc1 homozygous mice

To address the potential role of Tsc1 in the homeostasis of vascular ECs, mice containing a floxed Tsc1 allele were crossed with transgenic mice expressing Cre under the control of the endothelial-specific Tie2 promoter (18). The resultant Cre-mediated deletion of Tsc1 exons 17 and 18 leads to a frame-shift mutation that produces a non-functional protein (Fig. 1A). To verify the efficiency and specificity of Tie2-Cre-mediated recombination in ECs, we bred Tie2-Cre mice with mT/mG mice, a double fluorescent reporter mice, which express membrane-targeted red fluorescent tdTomato (mT) prior to Cre excision and membrane-targeted EGFP (mG) following Cre excision in non-recombined and recombined cells, respectively (19,20). The pattern of mT and mG fluorescence was observed in the yolk sacs of Tie2-Cre;mT/mG and mT/mG embryos yolk sacs by a fluorescent microscope. The results revealed only strong red fluorescence in mT/mG yolk sac vessels. In contrast, Tie2-Cre-induced green fluorescent protein (GFP) expression was observed in vessels, as demonstrated by a red-to-green fluorescence color switch following crossing with mT/mG mice (Fig. 1B). When PCR analysis of the yolk sacs was performed with primers targeting the wildtype (WT), floxed and mutant Tsc1 alleles, the various genotypes were observed in PCR gels as expected (Fig. 1C). We localized the vascular ECs with the CD31 marker and then found that the expression of Cre was colocalized with CD31 in vascular ECs in mutant embryos tissues (mT/mG+/−; Tie2-Cre+/−) (Supplementary Material, Fig. S1). Immunofluorescent staining indicated that Tsc1 (hamartin) protein expression was decreased in the mutant embryo endothelia, labeled by endothelial marker, endomucin (Fig. 1D) (21,22). Furthermore, western blotting analysis demonstrated that Tsc1 was dramatically decreased in yolk sacs of the EC Tsc1 KO (Tie2-Cre/Tsc1−/−) mice compared with the controls (Fig. 1E). Taken together, the results showed that Tsc1 deficiency was mediated by Cre-recombinase in ECs.

Figure 1.

Generation of endothelial cell-specific Tsc1 knockout mice. (A) Illustration of the Tsc1 exons of mutant and floxed alleles. (B) Tie2 promotor-driven Cre-recombination activity was assessed by crossing with a double fluorescence protein reporter mouse mT/mG. Images of yolk sacs were visualized by fluorescence microscope at E16.5. Yolk sac of mT/mG; Tie2-Cre−/− showed red fluorescence (a) and GFP fluorescence was illuminated in mT/mG; Tie2-Cre+/- embryos (b). Fixed tissue section of mT/mG; Tie2-Cre yolk sacs showed mG labeling in ECs, a switch from red fluorescence mT (c) to green fluorescence mT protein (d). (C) Genotypes of yolk sac samples were identified by PCR. Lanes: Tsc1fx/+ (fx/+), Tsc1fx/fx (fx/fx), Tie2-Cre(+)/Tsc1+/ −(+/−), Tie2-Cre(+)/Tsc1−/−(−/−). Similar results were from samples of mouse tails. (D) Immunofluorescence detection of Tsc1 (green) in vessels (endomucin staining; red) of WT littermates and mutant embryos at E13.5. KO vessels show reduced Tsc1 staining (c and d) compared with WT (a and b). Nuclei are shown in blue (DAPI staining). (E) Western blotting detection of Tsc1 in the yolk sacs of E11.5, Tsc1 was reduced in mutant (KO) yolk sacs compared with control littermates (WT).

Figure 1.

Generation of endothelial cell-specific Tsc1 knockout mice. (A) Illustration of the Tsc1 exons of mutant and floxed alleles. (B) Tie2 promotor-driven Cre-recombination activity was assessed by crossing with a double fluorescence protein reporter mouse mT/mG. Images of yolk sacs were visualized by fluorescence microscope at E16.5. Yolk sac of mT/mG; Tie2-Cre−/− showed red fluorescence (a) and GFP fluorescence was illuminated in mT/mG; Tie2-Cre+/- embryos (b). Fixed tissue section of mT/mG; Tie2-Cre yolk sacs showed mG labeling in ECs, a switch from red fluorescence mT (c) to green fluorescence mT protein (d). (C) Genotypes of yolk sac samples were identified by PCR. Lanes: Tsc1fx/+ (fx/+), Tsc1fx/fx (fx/fx), Tie2-Cre(+)/Tsc1+/ −(+/−), Tie2-Cre(+)/Tsc1−/−(−/−). Similar results were from samples of mouse tails. (D) Immunofluorescence detection of Tsc1 (green) in vessels (endomucin staining; red) of WT littermates and mutant embryos at E13.5. KO vessels show reduced Tsc1 staining (c and d) compared with WT (a and b). Nuclei are shown in blue (DAPI staining). (E) Western blotting detection of Tsc1 in the yolk sacs of E11.5, Tsc1 was reduced in mutant (KO) yolk sacs compared with control littermates (WT).

ECs-specific deletion of Tsc1 activates mTOR signaling

We examined whether loss of the Tsc1 allele led to hyperactivation of mTOR. Phosphorylation of S6 (pS6) or 4EBP-1 (p-4EBP-1) reflects mTOR activity. The phosphorylation levels of both S6 and 4EBP-1 were elevated in the mutant yolk sacs (Tie2-Cre/Tsc1−/−) compared with control littermates (Fig. 2A). Phospho-Akt was decreased in the mutant yolk sacs, indicating hyperactive mTOR-mediated negative feedback regulation of Akt signaling (Fig. 2A) (7,23,24). Additionally, the Tsc1 mutant embryos displayed enhanced pS6 expression in vascular ECs compared with control littermates in immunostaining analysis (Fig. 2B). Collectively, the results revealed excessive mTOR activity in yolk sacs of Tie2-Cre/Tsc1−/− embryos.

Figure 2.

Loss of Tsc1 activates mTOR in endothelial cells. (A) Western blotting detection of mTOR signaling pathway in E11.5 control littermates (WT) and endothelial Tsc1 knockout yolk sac (KO). KO yolk sacs showed up-regulation of mTOR, characterized with enhanced phospho-S6 and phospho-4EBP-1 and reduced phospho-Akt. (B) Immunofluorescence staining of pS6 (green) in vessels (endomucin staining; red) of WT littermates (a–c) and mutant embryos (d–f) at E13.5. Nuclei are shown in blue (DAPI staining).

Figure 2.

Loss of Tsc1 activates mTOR in endothelial cells. (A) Western blotting detection of mTOR signaling pathway in E11.5 control littermates (WT) and endothelial Tsc1 knockout yolk sac (KO). KO yolk sacs showed up-regulation of mTOR, characterized with enhanced phospho-S6 and phospho-4EBP-1 and reduced phospho-Akt. (B) Immunofluorescence staining of pS6 (green) in vessels (endomucin staining; red) of WT littermates (a–c) and mutant embryos (d–f) at E13.5. Nuclei are shown in blue (DAPI staining).

The endothelial Tsc1 homozygous mutation is embryonic lethal

Genotyping of tail DNA revealed that no endothelial Tsc1 KO (Tie2-cre/Tsc1−/−) offsprings were obtained from the mating of Tie2-Tsc1+/− and Tsc1fx/fx mice, indicating in utero lethality due to loss of Tsc1 expression (Table 1). To determine the developmental stage at which lethality occurred, embryos were collected at different gestational times and analyzed. No obvious differences in morphology could be detected between WT and mutant embryos from E9.5 to 11.5 (Supplementary Material, Fig. S2). At E12.5, the blood vessels in the heads were very pale in 60% of the Tie2-Cre/Tsc1−/− embryos, indicating reduced amounts of red blood cells or a lack of head vasculatures (Fig. 3Aa and b). Furthermore, the majority of Tsc1 mutant embryos began to die at E13.5, as demonstrated by the lack of a heartbeat or the presence of necrotic tissue. Hemorrhage was observed at this stage, and the yolk sacs of mutant embryos were pale, whereas those of WT embryos displayed well-formed vessels filled with red blood cells. Up to E14.5, the Tsc1 mutant embryos showed various and more severe vascular defects, including edema and subcutaneous hemorrhage at multiple sites (Fig. 3Ac and d). After E14.5, no living Tsc1 mutant embryos were observed, and all homozygous mutant embryos began to degenerate. Intriguingly, there was no obvious difference in the size of the mutant embryos compared with WT littermates, indicating that the mutation did not cause obvious developmental delay in utero. The data suggest that the Tsc1-deficient embryos died during mid-gestational development.

Table 1.

Distribution of mouse embryos with various genotypes from the mating between Tie2-Cre/Tsc1+/− and Tsc1fx/fx mice

Age Tie2-Cre/Tsc1−/− Tie2-Cre/Tsc1+/− Tsc1fx/fx Tsc1fx/+ Uncleara 
E11.5 34 31 30 24 
E12.5 30(2)b 18 35 32 10 
E13.5 39(8)b 39 33 35 
E14.5 18(16)b 17 20 23 10 
E16.5 21 18 15 
P0 16 21 17 
Age Tie2-Cre/Tsc1−/− Tie2-Cre/Tsc1+/− Tsc1fx/fx Tsc1fx/+ Uncleara 
E11.5 34 31 30 24 
E12.5 30(2)b 18 35 32 10 
E13.5 39(8)b 39 33 35 
E14.5 18(16)b 17 20 23 10 
E16.5 21 18 15 
P0 16 21 17 

E, embryonic day; P, postnatal day.

aGenotyping was precluded by resorption.

bThe values in parentheses indicate the numbers of embryos that died.

Figure 3.

Endothelial disruption of Tsc1 causes defects in angiogenesis of embryos. (A) Gross phenotypes of ECs Tsc1-deficient embryos. The mutant embryo (b) was paler, and the vasculature of the head region was irregularly shaped compared with wildtype embryo (a) at E12.5. Edema (white arrow) and hemorrhage (white arrowhead) were apparent in E14.5 KO embryo (d) compared with WT (c). (B) Whole-mount CD31 immunostaining of embryos. KO embryos showed decreased vessel formation and a disorganized vascular network in Tsc1−/ embryo at E9.5 (b) and E10.5 (d). (C) In the primary head veins, the E11.5 KO embryos (b) showed less developed vessels, with a reduced caliber and number of branches compared to WT (a). (D) Whole-mount immunostaining of endomucin in the yolk sac showed impaired angiogenesis and reduced numbers of branched blood vessels from vitelline vessels (b) and smaller vessels (d) and capillaries (f) in KO embryos (white arrow) compared with WT (a, c and e). (E) Histological sections of placental tissues from E14.5 embryos. H&E-stained sections of mutant embryo placentas presented structural abnormalities. The Tsc1−/− labyrinth layer (La) of KO embryo (b) was much thicker than that of WT (a). There was no significant difference in the size of the spongiotrophoblast (Sp) between mutant placentas versus controls. (F) The relative width of the La layer in KO placentas was significantly increased compared to that of WT embryos, though the difference of Sp between WT and KO embryos was not significant. *P < 0.05, n = 5. Results are expressed as the means ± SEM.

Figure 3.

Endothelial disruption of Tsc1 causes defects in angiogenesis of embryos. (A) Gross phenotypes of ECs Tsc1-deficient embryos. The mutant embryo (b) was paler, and the vasculature of the head region was irregularly shaped compared with wildtype embryo (a) at E12.5. Edema (white arrow) and hemorrhage (white arrowhead) were apparent in E14.5 KO embryo (d) compared with WT (c). (B) Whole-mount CD31 immunostaining of embryos. KO embryos showed decreased vessel formation and a disorganized vascular network in Tsc1−/ embryo at E9.5 (b) and E10.5 (d). (C) In the primary head veins, the E11.5 KO embryos (b) showed less developed vessels, with a reduced caliber and number of branches compared to WT (a). (D) Whole-mount immunostaining of endomucin in the yolk sac showed impaired angiogenesis and reduced numbers of branched blood vessels from vitelline vessels (b) and smaller vessels (d) and capillaries (f) in KO embryos (white arrow) compared with WT (a, c and e). (E) Histological sections of placental tissues from E14.5 embryos. H&E-stained sections of mutant embryo placentas presented structural abnormalities. The Tsc1−/− labyrinth layer (La) of KO embryo (b) was much thicker than that of WT (a). There was no significant difference in the size of the spongiotrophoblast (Sp) between mutant placentas versus controls. (F) The relative width of the La layer in KO placentas was significantly increased compared to that of WT embryos, though the difference of Sp between WT and KO embryos was not significant. *P < 0.05, n = 5. Results are expressed as the means ± SEM.

Tsc1 deficiency in endothelium causes severe defects in angiogenesis

To discern whether the lethality of ECs Tsc1-null embryos was attributable to defects in angiogenesis, we analyzed the vascular morphology of the embryo proper as well as the yolk sac. Vascular remodeling into small and large vessels was significantly reduced in the head region of mutant embryos compared with WT littermates (Fig. 3A). Whole-mount immunostaining of CD31 revealed obvious vascular remodeling defects in the E9.5 and 10.5 mutant embryos (Fig. 3B). Vascular remodeling into small and large vessels was significantly reduced in the head region of mutant embryos compared with WT littermates. The mutant embryos exhibited large branching vessels, but the formation of the capillary network was perturbed. At E11.5, the primary head veins were less elaborated, with thinner vessel branches (Fig. 3C). The yolk sac vasculature was also affected by EC-specific Tsc1 deletion, as revealed through visualization of the embryonic vasculature by staining with an antibody against endomucin. The WT yolk sacs showed a well-organized hierarchical structure of large and small vessels. In contrast, the mutant yolk sacs displayed decreased sprouting angiogenesis and a much smaller number of branched vessels. Furthermore, although the primary capillary plexuses formed in the mutant yolk sacs, the major vitelline vessels were irregularly shaped (Fig. 3D; Supplementary Material, Fig. S3). We also observed abnormalities in the placenta when Tsc1 was deleted. At E14.5, the placentas of the mutant embryos exhibited a dilated labyrinth (La), while the spongiotrophobast (Sp) layer appeared to be unaffected (Fig. 3E). The thickness of the placental La was increased compared with littermates, whereas there was no significant difference in the Sp layer detected between the placentas of Tie2-Cre/Tsc1−/−mice and their littermates (Fig. 3F). No difference was observed in vasculogenesis between mutant and WT embryos. Blood smears from the mutant embryos and their liver tissues showed no difference in the proportion of hemangioblasts compared with WT mice (Supplementary Material, Fig. S4). These data suggested that disruption of Tsc1 was not required for vasculogenesis but prevented later angiogenesis.

EC-specific deletion of Tsc1 induces cell apoptosis, decreases cell proliferation and disrupts the expression of angiogenic factors

The expression of alpha-smooth muscle actin (α-SMA) was significantly reduced in mutant vessels. Consistent with the observed vessel formation defects, the expression of the CD31 was also markedly decreased (Fig. 4A). Quantitative detection of CD31 and α-SMA revealed an obvious reduction in the number of arteries in the Tie2-cre/Tsc1−/−embryos (Fig. 4B and C). To test whether this vessel remodeling was attributable to a difference in cell growth between WT and mutant mice, we next examined apoptosis in the vessels using terminal deoxynucleotidyl transferase 2-deoxyuridine, 5-triphosphate nick end-labeling (TUNEL) assays. Apoptosis in ECs was increased in mutant compared with WT embryos. Moreover, the proliferation of ECs was significantly lower in mutant embryos, as determined by Ki67 immunostaining (Fig. 4D). The results demonstrated that cell growth was abnormal during vessel development in the mutant embryos. Finally, the mRNA levels of angiogenic factors were evaluated by qPCR analysis of the yolk sacs in mutant embryos and their WT littermates. The yolk sacs of mutant embryos showed significantly elevated levels of transforming growth factor-β1 (Tgf-β1), platelet-derived growth factor-b (Pdgf-b), Tek and Egfr mRNA, but significantly decreased levels of vascular cell adhesion molecule-1 (Vcam-1), basic fibroblast growth factor (bfgf) and Angiopoietin-1 (Angpt-1) (Fig. 4E). No differences in the expression levels of vascular endothelial growth factor-α (Vegf-a) and hypoxia-induced factor-1α (Hif-1α) were detected.

Figure 4.

Abnormal endothelia in Tie2-cre/Tsc1−/− embryos. (A) Immunohistochemical detection of α-SMA and CD31 in the embryonic vasculature revealed decreased expression of both α-SMA (b) and CD31 (d) in E13.5 KO embryos than WT (a and c). (B) Quantitative data on α-SMA expression indicated a 3-fold decrease in mutant embryonic sections. (C) Analysis of CD31 expression demonstrated a 5-fold decrease in KO embryos. (D) Few TUNEL-positive cells were observed in WT embryos (a), while many TUNEL-positive cells were present in the KO vessels (b). Immunohistochemical staining of Ki67 revealed a reduced number of Ki67-positive cells in the mutant ECs (d) than WT (c). These sections were counterstained with hematoxylin. (E) qPCR analysis of angiogenic factors in WT and KO (Tie2-cre/Tsc1−/−) yolk sacs at E11.5. Tgf-β1, Pdgf-b, Tek, and Egfr were up-regulated in KO mice, while Vcam-1, bFgf, and Angpt-1 were down-regulated in KO mice. *P < 0.05; **P < 0.01.

Figure 4.

Abnormal endothelia in Tie2-cre/Tsc1−/− embryos. (A) Immunohistochemical detection of α-SMA and CD31 in the embryonic vasculature revealed decreased expression of both α-SMA (b) and CD31 (d) in E13.5 KO embryos than WT (a and c). (B) Quantitative data on α-SMA expression indicated a 3-fold decrease in mutant embryonic sections. (C) Analysis of CD31 expression demonstrated a 5-fold decrease in KO embryos. (D) Few TUNEL-positive cells were observed in WT embryos (a), while many TUNEL-positive cells were present in the KO vessels (b). Immunohistochemical staining of Ki67 revealed a reduced number of Ki67-positive cells in the mutant ECs (d) than WT (c). These sections were counterstained with hematoxylin. (E) qPCR analysis of angiogenic factors in WT and KO (Tie2-cre/Tsc1−/−) yolk sacs at E11.5. Tgf-β1, Pdgf-b, Tek, and Egfr were up-regulated in KO mice, while Vcam-1, bFgf, and Angpt-1 were down-regulated in KO mice. *P < 0.05; **P < 0.01.

EC-specific deletion of Tsc1 damages mitochondria and endoplasmic reticulum

We further investigated whether Tsc1 deficiency in ECs affected the ultrastructure of subcellular organelles associated with cellular proliferation and apoptosis via transmission electron microscopy (TEM). Significant ultrastructural changes were detected in mutant ECs from E11.5 yolk sacs. It was demonstrated that the mitochondria of the mutants were generally swollen, edema and pale. In contrast, the mitochondria in the ECs of control mice showed a normal morphology, with intact cristae and electron-dense bodies in the mitochondrial matrix (Fig. 5A). The numbers of abnormal mitochondria per cell were increased in sections from the mutant embryos (Fig. 5B). Similarly, abnormally enlarged endoplasmic reticulum (ER) was increased in the mutant ECs compared with fine structure of WT mice (Fig. 5C). We observed a 1.9-fold increase in the number of abnormal ER-positive cells in ECs (Fig. 5D). Additionally, the integrity of tight junctional contacts related to cell permeability between ECs was intact in both WT and KO mice (Fig. 5E). These data indicated that ECs harboring a Tsc1 deficiency showed a defective mitochondrial and endoplasmic reticular morphology, but no significant change was observed in cell junctions.

Figure 5.

Abnormal mitochondria and endoplasmic reticulum in Tsc1−/− ECs of yolk sacs. (A) Normal mitochondrial morphology was observed in wildtype mice (a), whereas Tsc1−/− ECs showed an abnormal mitochondrial structure, with fewer electron-dense bodies and deficient cristae, and in some regions, cristae were difficult to be recognized at E11.5 (b) under transmission electron microscope (TEM). (B) The relative contents of abnormal mitochondria were increased in mutant ECs, n≥60 images, * P < 0.05. (C) Dilated ER was visible in mutant yolk sacs (b) in comparison with WT yolk sacs (a) (white arrow). (D) Quantitative analysis revealed a dramatic increase of the numbers of ECs exhibiting abnormally expanded ER in KO mice. (E) No significant differences in endothelial cell tight junctions were detected between WT (a) and KO mice (b) (black arrow). Mi, mitochondria; ER, endoplasmic reticulum; N, nucleus.

Figure 5.

Abnormal mitochondria and endoplasmic reticulum in Tsc1−/− ECs of yolk sacs. (A) Normal mitochondrial morphology was observed in wildtype mice (a), whereas Tsc1−/− ECs showed an abnormal mitochondrial structure, with fewer electron-dense bodies and deficient cristae, and in some regions, cristae were difficult to be recognized at E11.5 (b) under transmission electron microscope (TEM). (B) The relative contents of abnormal mitochondria were increased in mutant ECs, n≥60 images, * P < 0.05. (C) Dilated ER was visible in mutant yolk sacs (b) in comparison with WT yolk sacs (a) (white arrow). (D) Quantitative analysis revealed a dramatic increase of the numbers of ECs exhibiting abnormally expanded ER in KO mice. (E) No significant differences in endothelial cell tight junctions were detected between WT (a) and KO mice (b) (black arrow). Mi, mitochondria; ER, endoplasmic reticulum; N, nucleus.

EC-specific Tsc1 deletion compromises cardiac development

To determine whether Tsc1 KO in ECs affected cardiac development, serial heart sections were stained with H&E from E12.5 to E14.5. It was observed that loss of Tsc1 in ECs led to poorly developed trabeculae and a decreased ventricular wall thickness compared with WT littermates (Fig. 6A). However, the cardiac phenotype of ∼20% of the Tie2-Cre/Tsc1−/− embryos resembled hypertrophic cardiomyopathy (Supplementary Material, Fig. S5), similar to smooth muscle protein-22-mediated deletion of Tsc1 (13), that shows increased thickness of the ventricular wall. TUNEL assays showed that the mutant embryos exhibited significantly more cardiac myocyte apoptosis than the controls (Fig. 6B). To determine whether the observed abnormalities in heart development were caused by Cre-mediated deletion of Tsc1 in cardiomyocytes, we crossed Tie2-Cre mice with mT/mG reporter mice. Fluorescent labeling showed that the endocardium expressed GFP, whereas cardiomyocytes did not. Examination of heart sections under higher magnification indicated that Cre-mediated recombination occurred only in ECs. Red fluorescence was localized to cardiomyocytes (Supplementary Material, Fig. S6). Overall, histological analysis revealed a decreased thickness of the ventricular wall associated with the defect in cardiomyocyte differentiation.

Figure 6.

Cardiac defects in Tsc1−/− mutant embryos. (A) Histological analysis of serial sections of cardiac tissues from E12.5 to E14.5 embryos. The thickness of the ventricular wall was significantly thinner in the E12.5 and E14.5 mutant hearts, and the trabeculae were disorganized (b and d) compared with littermate controls (a and c). (B) Detection of apoptosis in the ventricular myocardium via TUNEL assays. The results showed a significantly increased number of apoptotic cells (green) in the heart sections from mutant embryos (b) compared with their WT counterparts (a); DAPI staining marks nuclei (blue).

Figure 6.

Cardiac defects in Tsc1−/− mutant embryos. (A) Histological analysis of serial sections of cardiac tissues from E12.5 to E14.5 embryos. The thickness of the ventricular wall was significantly thinner in the E12.5 and E14.5 mutant hearts, and the trabeculae were disorganized (b and d) compared with littermate controls (a and c). (B) Detection of apoptosis in the ventricular myocardium via TUNEL assays. The results showed a significantly increased number of apoptotic cells (green) in the heart sections from mutant embryos (b) compared with their WT counterparts (a); DAPI staining marks nuclei (blue).

Prenatal mTOR inhibition suppresses the embryonic lethality of Tie2-Cre/Tsc1−/−mice

To investigate whether the embryonic lethality observed in Tie2-cre/Tsc1−/− embryos could be alleviated, embryos were prenatally treated with mTOR inhibitor rapamycin between E12 and E13. Genotyping of the offspring showed that Tsc1 mutant mice survived to birth (Table 2). This finding indicated that rapamycin was able to rescue in utero embryonic lethality. We next examined the gross morphology of embryos isolated at E14.5 to E18.5 obtained from crosses between Tsc1fx/fx and Tie2-Cre/Tsc1+/− mice. The data showed that the level of phospho-S6 detected in the yolk sacs of mutant embryos was markedly decreased by rapamycin treatment (Fig. 7A). It was also found that rapamycin-treated mutant embryos were relatively morphologically normal compared with WT mice (Fig. 7B). Thus, rapamycin was sufficient to partially rescue the embryonic lethality of Tsc1-deficient mice. However, a few heterozygous and WT embryos also died in utero following rapamycin treatment (Table 2). As the WT embryos grew more rapidly, the differences between the mutants and WT embryos gradually became more significant from post-natal day (P) 7. All of the surviving rescued mutant mice showed severe growth retardation compared with their littermates (Fig. 7C). This growth retardation contributed to remarkably poor weight gain in the surviving mutant mice compared with their littermate controls (mutants 3.1 ± 0.15 g, controls 8.4 ± 0.2 g, P = 0.0001) (Fig. 7D). Most of mutant mice died within 24 h after birth, showing systemic cyanosis (Supplementary Material, Fig. S7), while a few of the mice survived to P22 (Fig. 7E). These data indicated that prenatal inhibition of mTOR hyperactivity could rescue embryonic lethality but was insufficient to prevent developmental defects after birth.

Table 2.

Redistribution of mouse embryos with various genotypes after treatment with rapamycin at E12-E13

Age Tie2-Cre/ Tsc1−/−a Tie2-Cre/ Tsc1+/− Tsc1fx/fx Tsc1fx/+ Unclearb 
E14.5 8(2) 
E16.5 19(4)a 17 16 15 
E18.5 20(6)a 13 10 
Age Tie2-Cre/ Tsc1−/−a Tie2-Cre/ Tsc1+/− Tsc1fx/fx Tsc1fx/+ Unclearb 
E14.5 8(2) 
E16.5 19(4)a 17 16 15 
E18.5 20(6)a 13 10 

E, embryonic day; P, postnatal day.

aThe values in parentheses indicate numbers of embryos that died.

bGenotyping was precluded by resorption.

Figure 7.

Prenatal rapamycin treatment reduces the embryonic lethality of endothelial cell-specific Tsc1 knockout mice. Rapamycin was delivered as a single dose (1 mg/kg) to the pregnant mice between E12 and 13. (A) Immunoblot analysis of phospho-S6 in E12.5 (WT, KO) and E18.5 (KO+Rapamycin) from WT and KO (Tie2-cre/Tsc1−/−) yolk sacs. (B) No phenotypic differences were observed at E16.5 between WT (a) and KO (b) embryos treated with rapamycin. (C) The survived KO mice showed growth retardation at P17. (D) Body weights of the KO and WT mice after birth, following prenatal treatment with rapamycin. The weight of the KO mice was markedly reduced compared with that of the WT mice. (E) Kaplan–Meier curves showing the survival of Tie2-cre/Tsc1−/− mice and littermate controls. Prenatal rapamycin treatment could extend the survival of Tie2-cre/Tsc1−/− mice. Most of the KO mice pre-treated with rapamycin died within 24 h after birth. The longest survival recorded for a KO mouse was 22 days after birth.

Figure 7.

Prenatal rapamycin treatment reduces the embryonic lethality of endothelial cell-specific Tsc1 knockout mice. Rapamycin was delivered as a single dose (1 mg/kg) to the pregnant mice between E12 and 13. (A) Immunoblot analysis of phospho-S6 in E12.5 (WT, KO) and E18.5 (KO+Rapamycin) from WT and KO (Tie2-cre/Tsc1−/−) yolk sacs. (B) No phenotypic differences were observed at E16.5 between WT (a) and KO (b) embryos treated with rapamycin. (C) The survived KO mice showed growth retardation at P17. (D) Body weights of the KO and WT mice after birth, following prenatal treatment with rapamycin. The weight of the KO mice was markedly reduced compared with that of the WT mice. (E) Kaplan–Meier curves showing the survival of Tie2-cre/Tsc1−/− mice and littermate controls. Prenatal rapamycin treatment could extend the survival of Tie2-cre/Tsc1−/− mice. Most of the KO mice pre-treated with rapamycin died within 24 h after birth. The longest survival recorded for a KO mouse was 22 days after birth.

DISCUSSION

Our experiments clearly demonstrated that mTOR signaling in the endothelium is essential for angiogenic and embryonic development. Tsc1 deficiency in ECs caused embryonic lethality. Whether this lethality was attributable to defects in angiogenesis or cardiac dysfunction, or to other causes, remains unclear. A combination effect is likely, though. mTOR activity is required for cell growth, proliferation, protein translation and other cellular metabolic processes (25). Hyperactive mTOR signaling is an underlying mechanism for the abnormal angiogenesis associated with tumorigenesis (26).

During embryonic development, ECs form blood islands, which are the primary capillary plexuses in the yolk sac. This process is referred to as vasculogenesis. Hemangioblasts are the precursors of ECs and crucial for vasculogenesis. The primitive capillary plexuses subsequently undergo angiogenic process of growth and remodeling to shape the mature vascular network (27,28). The primary vasculature is established at E7.5, but vascular morphological defects were apparent after E9.5 in this study. Therefore, we concluded that Tsc1 plays a critical role in embryonic angiogenesis, but not in vasculogenesis. This conclusion was further supported by the finding that hemangioblast differentiation was no difference in the blood and liver between WT and mutant embryos.

As EC-specific deletion of Tsc1 led to progressive systemic severe edema and hemorrhage, both of which are indicators of EC insufficiency, we speculate that the observed embryonic lethal phenotypes resulted from circulatory failure due to defective angiogenesis. Indeed, the mutant embryos failed to form branched vascular networks in the head region, and the carotid arteries were less elaborated in mutant embryos. The perturbation of the circulatory network structure was confirmed by endomucin staining in yolk sacs. Vitelline vessels were narrowed and disorganized. Placental failure was a major contributing factor to the embryonic lethal phenotype. Three placental layers (the labyrinth, spongiotrophoblast and decidua) were readily discernible. The labyrinth layer is the main site of contact between the fetal and maternal circulation, consisting of a highly branched fetal vascular network (29). The EC-specific deletion of Tsc1 affected placental vascularization as well, causing a significant increase in the thickness of the labyrinthine layer. This type of alteration can compromise embryonic growth and viability because the highly vascular labyrinth is the site of maternal-fetal oxygen and nutrient exchange (30). The enlarged labyrinthine layer phenotype may result from hyperactivity of mTOR in Tsc1-deleted ECs. The hypoplastic Tsc1 mutant embryos may have been compensated by the enlarged placental labyrinth, which would have facilitated the development of normal fetal size. Angiogenesis insufficiency may have been the major cause of the mid-gestation lethal phenotype of Tsc1 KO mice. Collectively, our results showed that the mutant embryos have exhibited defective angiogenesis in the yolk sac, embryo body and placenta. The edema and hemorrhaging observed in the head and dorsal regions of embryos harboring an EC-specific deletion of Tsc1 may have been related to defects in EC adhesion properties. Although the cell junctions of the mutant ECs appeared normal under electron microscopy, the mRNA expression of Vcam-1, a marker of cell adhesion, was significantly decreased in yolk sacs. Vcam-1 is expressed at the cell surface, where it mediates interactions with extracellular matrix components and regulates vascular permeability (31). In addition, the mutant ECs showed higher levels of apoptosis, but lower levels of proliferation, which may be explained by the markedly reduced expression of the bFgf anti-apoptosis factor in the mutant embryos' yolk sacs (32). Taken together, these data indicated the angiogenesis defect associated with Tsc1 deletion is mediated by altered cell proliferation and differentiation of ECs.

To gain further insight into the potential mechanisms underlying the defective vessel formation observed in Tsc1−/−embryos, we analyzed the endothelial ultrastructure. The ER is a specialized organelle that plays crucial roles in cell homeostasis and survival through its functions in protein folding, lipid biosynthesis and calcium and redox homeostatic processes (33). The mTOR signaling factor mediates cellular protein synthesis through phosphorylation of S6 K and 4EBP-1 and subsequent activation of cellular metabolism. We hypothesized that the enlarged ER might be related to up-regulated mTOR activity. Severe mitochondrial dysfunction was associated with swollen mitochondria, abnormal cristae and matrix structure, contributing to insufficient energy production and apoptosis. Thus, defective ECs were responsible for the abnormal angiogenesis observed during embryogenesis.

Future studies should address the mechanisms associated with a hyperactive mTOR signaling pathway in angiogenesis. Angiogenesis is a complex process involving finely balanced interactions between signaling factors, such as Vegf-a, Tgf-β1, Pdgf-b, Angpt-1, Vcam-1 and Hif-1α (34). Hif-1α and Vegf are also known to control EC proliferation and migration, both of which are fundamental for proper angiogenesis (35). It has been reported that activation of phosphoinositide 3-kinase/protein kinase B (PI3K/Akt) signaling promotes both mTOR-dependent and mTOR-independent HIF-1α/VEGF transcription during tumor angiogenesis (36,37). Homozygous KO of the Hif-1α or Vegf gene results in embryonic lethality in early stages of gestation due to impaired blood vessel development (38,39). However, in the EC-specific Tsc1 KO mice, there was no obvious effect on the transcription of either Hif-1α or Vegf-a. Angpt/Tek signaling is also involved in the establishment of vessel structures during embryonic development (40). Mouse KO models of Angpt-1 and Tek both show prominent endocardial and myocardial defects (41). The interaction of Angpt1 and Tek in ECs is essential for vasculature maturation and normal remodeling processes (42). However, the EC-specific Tsc1 KO mice showed reduced levels of Angpt-1, but increased levels of Tek. It has been suggested that disrupted Angpt/Tek signaling is associated with defective angiogenesis during the late gestational stage. Under normal physiological conditions, Pdgf-b promotes the proliferation and migration of pericyte precursor cells (43). Mice deficient in Pdgf-b or its receptor (Pdgfr-b) die in utero due to microvascular defects, which include vessel dilatation and microaneurysms (44). Similarly, activation of the EGFR signaling pathway mediates cell proliferation and angiogenesis, and overactivation of this pathway in various cancers has been shown to promote metastasis and inhibit apoptosis (45). Other factors that are necessary for the proper formation of blood vessels are α-SMA (46) and TGF-β1, which is a potent inhibitor of vascular smooth muscle cell proliferation (47). As our EC-specific Tsc1 KO embryos overexpressed Pdgf-b, Tgf-β1 and Egfr, it is possible that the observed vascular defects are related to perturbed activation of pericytes, decreased stability of blood vessel walls, an impaired EC proliferation ability and inhibited neovasculization. The abnormal expression of angiogenic factors caused by up-regulated mTOR activity in the mutant ECs may manifest as an interruption of angiogenesis homeostasis.

During embryonic heart development, ECs and cardiomyocytes arise from the same cardiac mesodermal precursors (48). The endocardial monolayer adjacent to the myocardial trabeculation releases growth factors that regulate myocyte development and proper cardiac function (49). Our results indicated that the hearts of the EC-specific Tsc1 KO embryos had a disorganized architecture and a thinner ventricular wall. Importantly, the Tsc1 mutant hearts exhibited a dramatic increase in the number of apoptotic cardiomyocytes. We suggested that inactivation of Tsc1 in ECs could affect heart development, resulting in a disturbed balance between proliferation and apoptosis during the process of cardiomyocyte differentiation. Myocardial ECs are found in the endocardium as the main component of cardiac capillaries, which supply nutrients and oxygen for cardiomyocyte growth (50). While it remains unknown whether vascular or cardiac defects were the primary cause of the observed embryonic lethal phenotype, we believe that cardiac failure could be secondary to the EC abnormalities.

Furthermore, to confirm that the lethality observed in homozygous Tsc1 KO embryos was due to the hyperactivation of mTOR, we attempted to rescue the lethal phenotypes of mutant embryos via rapamycin treatment. Prenatal treatment with rapalogues in TSC mutant mouse models has previously been described (12,51,52). Here, a single dose of rapamycin significantly improved embryonic viability. Nevertheless, most of the mutant pups died within 24 h after birth, showing systemic cyanosis. Only a few of the mutant pups survived up to 22 days, showing poor weight gain and significant developmental retardation. Survival of the mutant mice might be improved by rapamycin treatment after birth.

In conclusion, reduction of Tsc1 gene expression in the endothelium leads to embryonic death due to overactivation of mTOR, which can be rescued by prenatal rapamycin treatment. Hence, Tsc1 plays a crucial role in cardiovascular integrity and embryonic development through mTOR. Aberrant mTOR signaling may contribute to the vascular pathogenesis observed in TSC.

MATERIALS AND METHODS

Mice

All animal protocols adhered to the Animal Care and Use Committee Guidelines of Peking Union Medical College. Tie2-Cre (C57BL/6J) mice and Tsc1-floxed mice (Tsc1fx/fx) (129S4/SvJae) were obtained from Jackson Laboratory (Bar Harbor, ME, USA) (53,54). The Tie2-Cre mice were first crossed with Tsc1fx/fx mice to obtain Tie2-Cre/Tsc1+/− mice, which were then back-crossed with Tsc1fx/fx mice to obtain conditional KO Tie2-Cre/Tsc1−/− mice. The Gt(ROSA)26Sortm4(ACTB-tdTomato,-EGFR)Luo/J mice (mT/mG, C57BL/6) were procured from Jackson Laboratory. Tie2-Cre mice were crossed with mT/mG to get mT/mG+/−; Tie2-Cre+/− mice for analysis.

Isolation of mouse embryos

Timed matings were performed using the presence of a vaginal plug to assess fertilization between Tie2-Cre/Tsc1+/− male and Tsc1fx/fx female mice. The morning vaginal plug was designated E0.5. The impregnated females were sacrificed at several time points after conception. The embryos were isolated by removing the muscular wall of the uterus and yolk sacs in ice-cold PBS under a stereomicroscope (Nikon, Tokyo, Japan). The embryos were fixed overnight in 4% paraformaldehyde (PFA) at 4 or −80°C.

Genotyping

Tail or yolk sac samples were prepared as previously described (7). The published primers used in these assays allow detection of the WT Tsc1 allele (295 bp amplicon), the floxed allele (235 bp amplicon) and the mutant allele (370 bp amplicon) (7,55). The floxed allele contains loxP sites flanking the sequence, to be excised upon Cre recombination. The mutant allele lacks the sequence flanked by the loxP sites and produces a non-functional hamartin protein.

Western blotting

Yolk sacs and embryos were homogenized and extracted with Tissue Protein Extract Reagent (Thermo-Fisher Scientific, Waltham, MA, USA) and a 1% protease inhibitor cocktail (Thermo-Fisher Scientific). The protein concentration was quantified using a BCA Protein Assay Kit (Thermo Fisher Scientific). The samples were next resolved in 10% Tris–HCl Ready Gels (Invitrogen, Temecula, CA, USA) and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA), then incubated in blocking buffer consisting of 5% milk in tris-buffered saline and tween 20 (20 mm Tris, 500 mm NaCl, and 0.05% Tween-20, pH 7.5) for 1 h at room temperature. The PVDF membranes were incubated overnight with the following primary antibodies: anti-TSC1 (rabbit polyclonal 1:500 dilution, Cell Signaling Technology, Beverly, MA, USA), anti-Akt and phospho-Akt (rabbit polyclonal 1:1000 dilution, Cell Signaling Technology), anti-S6 and phospho-S6 (rabbit polyclonal 1:1000 dilution, Cell Signaling Technology), 4EBP-1 and phospho-4EBP-1 (rabbit polyclonal 1:1000 dilution, Cell Signaling Technology) and β-actin (mouse monoclonal 1:2000, Santa Cruz Biotechnology), followed by incubation with the appropriate secondary antibody. Immunoreactive bands were visualized using an enhanced chemiluminescence substrate (Pierce, Thermo-Fisher).

Histology and immunohistochemistry

Embryos were placed in 4% PFA, embedded in paraffin, and sectioned at a thickness of 3 μm. Serial sections were either stained with hematoxylin and eosin (H&E) for morphological evaluation or used for immunohistochemistry analysis. The tissue sections were deparaffinized and rehydrated in xylene, heated in 10 mm citrate buffer, pre-treated with 3% H2O2 for 10 min and blocked with 5% normal goat serum for 2 h. The sections were then labeled with the following primary antibodies: α-SMA (rabbit polyclonal 1:500, Abcam, Cambridge, UK), PECAM (CD31) (goat polyclonal 1:1000, R&D Systems, Minneapolis, MN, USA) and ki67 (rabbit monoclonal 1:300, Thermo, Waltham, MA, USA) at 4°C overnight. After rinsing, the sections were incubated with the appropriate secondary antibodies for 1 h at room temperature and finally detected using diaminobenzidine (DAB) as a substrate (Vector, Burlingame, CA, USA). In addition, the morphology and staining characteristics of blood cells were investigated via the standard blood smear procedure (56). The ratio of positive cells in images was (n = 5) calculated by Carl Zeiss AxioVision Re1.4.6.

Immunofluorescence staining

Sections were dewaxed and rehydrated according to standard processing procedures. Tissue sections were blocked and incubated with anti-phospho-S6, or endomucin antibody (rat monoclonal antibody1:300, eBioscience, San Diego, CA, USA) at 4°C overnight. Cy3-conjuaged goat anti-rat and Alexa Fluor-488-conjugated goat anti-rabbit secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) were used for visualization of the resultant signals. Nuclei were labeled by DAPI staining. Yolk sacs were fixed in 4% PFA in phosphate buffered saline (PBS) overnight at 4°C. The tissues were frozen and mounted in Tissue-Tek OCT compound (Sakura, Tokyo, Japan) and sectioned at a 10 μm thickness with a cryomicrotome. The slides were then rinsed and counterstained with DAPI. Finally, the sections were observed using a confocal microscope (Olympus FV1000, Tokyo, Japan).

Whole-mount immunostaining

Embryos and yolk sacs were fixed in 4% PFA in 0.01 mm PBS overnight at 4°C, then washed twice with PBS for 10–20 min at room temperature, dehydrated via soaking in a methanol series, and bleached via incubation in 5% H2O2 for 4–5 h at room temperature. After rehydration in a reverse methanol series, the bleached embryos were blocked via incubation in 3% dry milk reconstituted in 0.1% PBS plus Triton X-100 (PBSMT) for either 1 h at room temperature or overnight at 4°C. Immunostaining to detect platelet endothelial adhesion molecule (anti-CD31) (goat polyclonal 1:300 dilution, R&D Systems, Minneapolis, MN, USA) and endomucin was carried out in 1 ml of PBSMT for 24 h at 4°C. The embryos were then washed in PBSMT five times (1 h each) with rocking, after which they were incubated with horseradish peroxidase rat anti-goat (1:300) and Cy3-conjugated AffiniPure goat anti-rat IgG (1:500) secondary antibodies (Jackson Immuno Research Laboratories, West Grove, PA, USA) at 4°C overnight with rocking. The embryos were then rinsed again in PBSMT at 4°C five times (1 h each) and washed with PBTX for 20 min. Peroxidase staining was performed with DAB at room temperature for 15 min. Finally, the embryos were post-fixed in 2% PFA with 0.1% glutaraldehyde in PBS and the stained yolk sacs were photographed using a Carl Zeiss Meditec microscope (Zeiss, Oberkochen, Germany).

TUNEL assay

Apoptotic cells were assessed via TUNEL staining using the In situ Cell Death Detection Kit (Roche Applied Science, Mannheim, Germany). Following deparaffinization and rehydration, the sections were treated with proteinase K for 10 min and then incubated with the TUNEL reaction mixture for 60 min at 37°C in the dark. The sections were next rinsed three times in PBS for 5 min each, followed by the addition of converter-POD to the samples and incubation for 30 min at 37°C. The sections were subsequently stained with DAB substrate and analyzed using a Carl Zeiss Meditec microscope. Heart sections were examined using the Dead End Fluorescence TUNEL system (Promega, Madison, WI, USA) following the manufacturer's instructions. Apoptotic cells were observed and photographed with a confocal microscope (Olympus FV1000). Nuclei were labeled by DAPI staining.

Quantitative real-time PCR analysis

RNA was extracted from complete yolk sacs using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Aliquots of 1 μg RNA were used for reverse transcription with an oligo-dT primer (Thermo-Fisher, Waltham, MA, USA). qPCR was performed using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) and gene-specific primers (sequences and thermal cycling conditions available upon request) in the StepOne and StepOnePlus Real-Time PCR Systems (Applied Biosystems). The following PCR conditions were used for the Applied Biosystems assays: 95°C for 10 min, followed by 40 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s. Quantification was performed using the ΔΔCT method. The housekeeping gene Gapdh was used for internal normalization. The following primers were used: mouse Hif-1α (accession no. NM_010431), Vcam-1 (accession no. NM_011693.3), Egfr (accession no. NM_001083119.2), bFgf (accession no. NM_008006.2), Tgf-β1 (forward, 5-GCTGCGCTTGCAGAGATTAAA; reverse, 5-TTGCTGTACTGTGTGTCCAG), Pdgf-b (forward, 5-ACTCCATCCGCTCCTTTGAT; reverse, 5-GTCTTGCACTCGGCGATTA), Tek (forward, 5-TGGAGT- CAGCTTGCTCCTTT; reverse, 5-ACCTCCAGTGGATCTTGGTG), Angpt-1 (forward, 5-GGGGGAGGTTGGACAGTAA; reverse, 5-CATCAGCTCAATCCTCAGC), Vegf-a (forward, 5-CAGGCTGCTGTAACGATGAA; reverse, 5-CTATGTGCTG- GCTTTGGTGA), Angpt1 (forward, 5-GGGGGAGGTTGGACAGTAA; reverse, 5-CATCAGCTCAATCCTCAGC) (57), Gapdh (accession no. NM_008084.2).

TEM

Yolk sacs from E10.5 embryos were fixed with 2.5% glutaraldehyde in 100 mm phosphate buffer overnight at 4°C, then post-fixed in 1% osmium tetroxide in 100 mm phosphate buffer for 2 h at 4°C, dehydrated through a graded series of ethanol, transferred to propylene oxide and embedded in eponsy resin. Ultra-thin sections were cut and visualized using a Tecnai Spirit transmission electron microscope (FEI, Hillsboro, Oregon, USA). Images were randomly selected at original magnifications of 11500-26500X. The number of mitochondria per cell was counted in at least 200 cells in each group. The proportion endoplasmic reticula showing dilation was estimated as the number of endoplasmic reticula with a width at least twice the width of a normal cell member, divided by the total number of endoplasmic reticula counted.

Rapamycin treatment

Rapamycin (North China Pharmaceutical Co., Ltd, Shijiazhuang, China) was dissolved at a concentration of 20 mg/ml in ethanol and stored at −20°C. A working solution for subcutaneous injection was prepared in 0.25% Tween-80 and 0.25% polyethylene glycol 400 diluted in PBS (12). Pregnant dams were administered a single dose of 1 mg/kg between E12 and E13.

Statistical analysis

Data are expressed as the means ± SEM. The significance of inter-group differences was assessed using a two-tailed Student's t-test or ANOVA. A P-value of <0.05 was considered statistically significant.

SUPPLEMENTARY MATERIAL

Supplementary material is available at HMG online.

FUNDING

This work was supported by the 973 National Basic Research Program of China (2009CB522106, 2009CB522203).

ACKNOWLEDGEMENTS

The authors thank Prof. Weiming Tong (Department of Pathology, Institute of Basic Medical Sciences and School of Basic Medicine, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China) and Prof. Zhong-Zhou Yang (Model Animal Research Center of Nanjing University, Nanjing, China) for advice of mouse study. We thank Dr Bee Hong Lo (Children's Hospital Westmead, Sydney, Australia) for her assistance in editing the abstract of this paper.

Conflict of Interest statement. None declared.

REFERENCES

1
Crino
P.B.
Nathanson
K.L.
Henske
E.P.
The tuberous sclerosis complex
N. Engl. J. Med.
 , 
2006
, vol. 
355
 (pg. 
1345
-
1356
)
2
Kwiatkowski
D.J.
Whittemore
V.H.
Thiele
E.A.
Tuberous sclerosis Complex: Genes, Clinical Features, and Therapeutics
 , 
2010
Germany
Wiley-Blackwell, Weinheim
3
The European Chromosome 16 Tuberous Sclerosis Consortium
Identification and characterization of the tuberous sclerosis gene on chromosome 16
Cell
 , 
1993
, vol. 
75
 (pg. 
1305
-
1315
)
4
Van Slegtenhorst
M.
de Hoogt
R.
Hermans
C.
Nellist
M.
Janssen
B.
Verhoef
S.
Lindhout
D.
van den Ouweland
A.
Halley
D.
Young
J.
, et al.  . 
Identification of the tuberous sclerosis gene TSC1 on chromosome 9q34
Science
 , 
1997
, vol. 
277
 (pg. 
805
-
808
)
5
Gao
X.
Zhang
Y.
Arrazola
P.
Hino
O.
Kobayashi
T.
Yeung
R.S.
Ru
B.
Pan
D.
Tsc tumour suppressor proteins antagonize amino-acid-TOR signalling
Nat. Cell. Biol.
 , 
2002
, vol. 
4
 (pg. 
699
-
704
)
6
Inoki
K.
Li
Y.
Zhu
T.
Wu
J.
Guan
K.L.
TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling
Nat. Cell. Biol.
 , 
2002
, vol. 
4
 (pg. 
648
-
657
)
7
Kwiatkowski
D.J.
Zhang
H.
Bandura
J.L.
Heiberger
K.M.
Glogauer
M.
el-Hashemite
N.
Onda
H.
A mouse model of TSC1 reveals sex-dependent lethality from liver hemangiomas, and up-regulation of p70S6 kinase activity in Tsc1 null cells
Hum. Mol. Genet.
 , 
2002
, vol. 
11
 (pg. 
525
-
534
)
8
Manning
B.D.
Tee
A.R.
Logsdon
M.N.
Blenis
J.
Cantley
L.C.
Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3-kinase/akt pathway
Mol. Cell
 , 
2002
, vol. 
10
 (pg. 
151
-
162
)
9
Potter
C.J.
Pedraza
L.G.
Xu
T.
Akt regulates growth by directly phosphorylating Tsc2
Nat. Cell. Biol.
 , 
2002
, vol. 
4
 (pg. 
658
-
665
)
10
Thoreen
C.C.
Chantranupong
L.
Keys
H.R.
Wang
T.
Gray
N.S.
Sabatini
D.M.
A unifying model for mTORC1-mediated regulation of mRNA translation
Nature
 , 
2012
, vol. 
485
 (pg. 
109
-
113
)
11
Kwiatkowski
D.J.
Animal models of lymphangioleiomyomatosis (LAM) and tuberous sclerosis complex (TSC)
Lymphat. Res. Biol.
 , 
2010
, vol. 
8
 (pg. 
51
-
57
)
12
Anderl
S.
Freeland
M.
Kwiatkowski
D.J.
Goto
J.
Therapeutic value of prenatal rapamycin treatment in a mouse brain model of tuberous sclerosis complex
Hum. Mol. Genet.
 , 
2011
, vol. 
20
 (pg. 
4597
-
4604
)
13
Malhowski
A.J.
Hira
H.
Bashiruddin
S.
Warburton
R.
Goto
J.
Robert
B.
Kwiatkowski
D.J.
Finlay
G.A.
Smooth muscle protein-22-mediated deletion of Tsc1 results in cardiac hypertrophy that is mTORC1-mediated and reversed by rapamycin
Hum. Mol. Genet.
 , 
2011
, vol. 
20
 (pg. 
1290
-
1305
)
14
Beltramello
A.
Puppini
G.
Bricolo
A.
Andreis
I.A.
el-Dalati
G.
Longa
L.
Polidoro
S.
Zavarise
G.
Marradi
P.
Does the tuberous sclerosis complex include intracranial aneurysms? A case report with a review of the literature
Pediatr. Radiol.
 , 
1999
, vol. 
29
 (pg. 
206
-
211
)
15
Jost
C.J.
Gloviczki
P.
Edwards
W.D.
Stanson
A.W.
Joyce
J.W.
Pairolero
P.C.
Aortic aneurysms in children and young adults with tuberous sclerosis: report of two cases and review of the literature
J. Vasc. Surg.
 , 
2001
, vol. 
33
 (pg. 
639
-
642
)
16
Salerno
A.E.
Marsenic
O.
Meyers
K.E.
Kaplan
B.S.
Hellinger
J.C.
Vascular involvement in tuberous sclerosis
Pediatr. Nephrol.
 , 
2010
, vol. 
25
 (pg. 
1555
-
1561
)
17
Rakowski
S.K.
Winterkorn
E.B.
Paul
E.
Steele
D.J.
Halpern
E.F.
Thiele
E.A.
Renal manifestations of tuberous sclerosis complex: Incidence, prognosis, and predictive factors
Kidney. Int.
 , 
2006
, vol. 
70
 (pg. 
1777
-
1782
)
18
Wong
A.L.
Haroon
Z.A.
Werner
S.
Dewhirst
M.W.
Greenberg
C.S.
Peters
K.G.
Tie2 expression and phosphorylation in angiogenic and quiescent adult tissues
Circ. Res.
 , 
1997
, vol. 
81
 (pg. 
567
-
574
)
19
Muzumdar
M.D.
Tasic
B.
Miyamichi
K.
Li
L.
Luo
L.
A global double-fluorescent Cre reporter mouse
Genesis
 , 
2007
, vol. 
45
 (pg. 
593
-
605
)
20
Zhang
L.
Zhang
B.
Han
S.J.
Shore
A.N.
Rosen
J.M.
Demayo
F.J.
Xin
L.
Targeting CreER(T2) expression to keratin 8-expressing murine simple epithelia using bacterial artificial chromosome transgenesis
Transgenic. Res.
 , 
2012
, vol. 
21
 (pg. 
1117
-
1123
)
21
Brachtendorf
G.
Kuhn
A.
Samulowitz
U.
Knorr
R.
Gustafsson
E.
Potocnik
A.J.
Fassler
R.
Vestweber
D.
Early expression of endomucin on endothelium of the mouse embryo and on putative hematopoietic clusters in the dorsal aorta
Dev. Dyn.
 , 
2001
, vol. 
222
 (pg. 
410
-
419
)
22
Liu
C.
Shao
Z.M.
Zhang
L.
Beatty
P.
Sartippour
M.
Lane
T.
Livingston
E.
Nguyen
M.
Human endomucin is an endothelial marker
Biochem. Biophys. Res. Commun.
 , 
2001
, vol. 
288
 (pg. 
129
-
136
)
23
Zhang
H.
Bajraszewski
N.
Wu
E.
Wang
H.
Moseman
A.P.
Dabora
S.L.
Griffin
J.D.
Kwiatkowski
D.J.
PDGFRs are critical for PI3 K/Akt activation and negatively regulated by mTOR
J. Clin. Invest.
 , 
2007
, vol. 
117
 (pg. 
730
-
738
)
24
Zhang
H.
Cicchetti
G.
Onda
H.
Koon
H.B.
Asrican
K.
Bajraszewski
N.
Vazquez
F.
Carpenter
C.L.
Kwiatkowski
D.J.
Loss of Tsc1/Tsc2 activates mTOR and disrupts PI3K-Akt signaling through downregulation of PDGFR
J. Clin. Invest.
 , 
2003
, vol. 
112
 (pg. 
1223
-
1233
)
25
Dennis
P.B.
Jaeschke
A.
Saitoh
M.
Fowler
B.
Kozma
S.C.
Thomas
G.
Mammalian TOR: a homeostatic ATP sensor
Science
 , 
2001
, vol. 
294
 (pg. 
1102
-
1105
)
26
Jiang
B.H.
Liu
L.Z.
PI3 K/PTEN signaling in angiogenesis and tumorigenesis
Adv. Cancer. Res.
 , 
2009
, vol. 
102
 (pg. 
19
-
65
)
27
Adams
R.H.
Alitalo
K.
Molecular regulation of angiogenesis and lymphangiogenesis
Nat. Rev. Mol. Cell. Biol.
 , 
2007
, vol. 
8
 (pg. 
464
-
478
)
28
Beck
L.
Jr.
D'Amore
P.A.
Vascular development: cellular and molecular regulation
Faseb. J.
 , 
1997
, vol. 
11
 (pg. 
365
-
373
)
29
Watson
E.D.
Cross
J.C.
Development of structures and transport functions in the mouse placenta
Physiology (Bethesda)
 , 
2005
, vol. 
20
 (pg. 
180
-
193
)
30
Iwawaki
T.
Akai
R.
Yamanaka
S.
Kohno
K.
Function of IRE1 alpha in the placenta is essential for placental development and embryonic viability
Proc. Natl Acad. Sci. U S A.
 , 
2009
, vol. 
106
 (pg. 
16657
-
16662
)
31
Braun
M.
Pietsch
P.
Schror
K.
Baumann
G.
Felix
S.B.
Cellular adhesion molecules on vascular smooth muscle cells
Cardiovasc. Res.
 , 
1999
, vol. 
41
 (pg. 
395
-
401
)
32
Karsan
A.
Yee
E.
Poirier
G.G.
Zhou
P.
Craig
R.
Harlan
J.M.
Fibroblast growth factor-2 inhibits endothelial cell apoptosis by Bcl-2-dependent and independent mechanisms
Am. J. Pathol.
 , 
1997
, vol. 
151
 (pg. 
1775
-
1784
)
33
Rutkowski
D.T.
Kaufman
R.J.
A trip to the ER: coping with stress
Trends. Cell. Biol.
 , 
2004
, vol. 
14
 (pg. 
20
-
28
)
34
Carmeliet
P.
Angiogenesis in health and disease
Nat. Med.
 , 
2003
, vol. 
9
 (pg. 
653
-
660
)
35
Phelps
E.A.
Garcia
A.J.
Update on therapeutic vascularization strategies
Regen. Med.
 , 
2009
, vol. 
4
 (pg. 
65
-
80
)
36
Arsham
A.M.
Plas
D.R.
Thompson
C.B.
Simon
M.C.
Akt and hypoxia-inducible factor-1 independently enhance tumor growth and angiogenesis
Cancer. Res.
 , 
2004
, vol. 
64
 (pg. 
3500
-
3507
)
37
Hudson
C.C.
Liu
M.
Chiang
G.G.
Otterness
D.M.
Loomis
D.C.
Kaper
F.
Giaccia
A.J.
Abraham
R.T.
Regulation of hypoxia-inducible factor 1alpha expression and function by the mammalian target of rapamycin
Mol. Cell. Biol.
 , 
2002
, vol. 
22
 (pg. 
7004
-
7014
)
38
Ferrara
N.
Carver-Moore
K.
Chen
H.
Dowd
M.
Lu
L.
O'Shea
K.S.
Powell-Braxton
L.
Hillan
K.J.
Moore
M.W.
Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene
Nature
 , 
1996
, vol. 
380
 (pg. 
439
-
442
)
39
Ryan
H.E.
Lo
J.
Johnson
R.S.
HIF-1 alpha is required for solid tumor formation and embryonic vascularization
Embo. J.
 , 
1998
, vol. 
17
 (pg. 
3005
-
3015
)
40
Augustin
H.G.
Koh
G.Y.
Thurston
G.
Alitalo
K.
Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system
Nat. Rev. Mol. Cell. Biol.
 , 
2009
, vol. 
10
 (pg. 
165
-
177
)
41
Suri
C.
Jones
P.F.
Patan
S.
Bartunkova
S.
Maisonpierre
P.C.
Davis
S.
Sato
T.N.
Yancopoulos
G.D.
Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis
Cell
 , 
1996
, vol. 
87
 (pg. 
1171
-
1180
)
42
Jones
N.
Dumont
D.J.
Tek/Tie2 signaling: new and old partners
Cancer Metastasis Rev.
 , 
2000
, vol. 
19
 (pg. 
13
-
17
)
43
Battegay
E.J.
Rupp
J.
Iruela-Arispe
L.
Sage
E.H.
Pech
M.
PDGF-BB modulates endothelial proliferation and angiogenesis in vitro via PDGF beta-receptors
J. Cell. Biol.
 , 
1994
, vol. 
125
 (pg. 
917
-
928
)
44
Lindahl
P.
Johansson
B.R.
Leveen
P.
Betsholtz
C.
Pericyte loss and microaneurysm formation in PDGF-B-deficient mice
Science
 , 
1997
, vol. 
277
 (pg. 
242
-
245
)
45
Baselga
J.
Why the epidermal growth factor receptor? The rationale for cancer therapy
Oncologist
 , 
2002
, vol. 
7
 (pg. 
2
-
8
)
46
Munoz-Chapuli
R.
Evolution of angiogenesis
Int. J. Dev. Biol.
 , 
2011
, vol. 
55
 (pg. 
345
-
351
)
47
Dickson
M.C.
Martin
J.S.
Cousins
F.M.
Kulkarni
A.B.
Karlsson
S.
Akhurst
R.J.
Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice
Development
 , 
1995
, vol. 
121
 (pg. 
1845
-
1854
)
48
Narumiya
H.
Hidaka
K.
Shirai
M.
Terami
H.
Aburatani
H.
Morisaki
T.
Endocardiogenesis in embryoid bodies: novel markers identified by gene expression profiling
Biochem. Biophys. Res. Commun.
 , 
2007
, vol. 
357
 (pg. 
896
-
902
)
49
Shah
A.M.
Paracrine modulation of heart cell function by endothelial cells
Cardiovasc. Res.
 , 
1996
, vol. 
31
 (pg. 
847
-
867
)
50
Heineke
J.
Wag the dog: how endothelial cells regulate cardiomyocyte growth
Arterioscler. Thromb. Vasc. Biol.
 , 
2012
, vol. 
32
 (pg. 
545
-
547
)
51
Goto
J.
Talos
D.M.
Klein
P.
Qin
W.
Chekaluk
Y.I.
Anderl
S.
Malinowska
I.A.
Di Nardo
A.
Bronson
R.T.
Chan
J.A.
, et al.  . 
Regulable neural progenitor-specific Tsc1 loss yields giant cells with organellar dysfunction in a model of tuberous sclerosis complex
Proc. Natl Acad. Sci. U S A.
 , 
2011
, vol. 
108
 (pg. 
E1070
-
E1079
)
52
Way
S.W.
Rozas
N.S.
Wu
H.C.
McKenna
J.
III
Reith
R.M.
Hashmi
S.S.
Dash
P.K.
Gambello
M.J.
The differential effects of prenatal and/or postnatal rapamycin on neurodevelopmental defects and cognition in a neuroglial mouse model of tuberous sclerosis complex
Hum. Mol. Genet.
 , 
2012
, vol. 
21
 (pg. 
3226
-
3236
)
53
Kisanuki
Y.Y.
Hammer
R.E.
Miyazaki
J.
Williams
S.C.
Richardson
J.A.
Yanagisawa
M.
Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo
Dev. Biol.
 , 
2001
, vol. 
230
 (pg. 
230
-
242
)
54
Meikle
L.
Pollizzi
K.
Egnor
A.
Kramvis
I.
Lane
H.
Sahin
M.
Kwiatkowski
D.J.
Response of a neuronal model of tuberous sclerosis to mammalian target of rapamycin (mTOR) inhibitors: effects on mTORC1 and Akt signaling lead to improved survival and function
J. Neurosci.
 , 
2008
, vol. 
28
 (pg. 
5422
-
5432
)
55
Meikle
L.
McMullen
J.R.
Sherwood
M.C.
Lader
A.S.
Walker
V.
Chan
J.A.
Kwiatkowski
D.J.
A mouse model of cardiac rhabdomyoma generated by loss of Tsc1 in ventricular myocytes
Hum. Mol. Genet.
 , 
2005
, vol. 
14
 (pg. 
429
-
435
)
56
Carmeliet
P.
Angiogenesis in life, disease and medicine
Nature
 , 
2005
, vol. 
438
 (pg. 
932
-
936
)
57
Jeansson
M.
Gawlik
A.
Anderson
G.
Li
C.
Kerjaschki
D.
Henkelman
M.
Quaggin
S.E.
Angiopoietin-1 is essential in mouse vasculature during development and in response to injury
J. Clin. Invest.
 , 
2011
, vol. 
121
 (pg. 
2278
-
2289
)

Author notes

A.M. and L.W. contributed equally to this work.

Supplementary data