Skip main navigation

The Noonan Syndrome Gene Lztr1 Controls Cardiovascular Function by Regulating Vesicular Trafficking

Originally publishedhttps://doi.org/10.1161/CIRCRESAHA.119.315730Circulation Research. 2020;126:1379–1393

Abstract

Rationale:

Noonan syndrome (NS) is one of the most frequent genetic disorders. Bleeding problems are among the most common, yet poorly defined complications associated with NS. A lack of consensus on the management of bleeding complications in patients with NS indicates an urgent need for new therapeutic approaches.

Objective:

Bleeding disorders have recently been described in patients with NS harboring mutations of LZTR1 (leucine zipper-like transcription regulator 1), an adaptor for CUL3 (CULLIN3) ubiquitin ligase complex. Here, we assessed the pathobiology of LZTR1-mediated bleeding disorders.

Methods and Results:

Whole-body and vascular specific knockout of Lztr1 results in perinatal lethality due to cardiovascular dysfunction. Lztr1 deletion in blood vessels of adult mice leads to abnormal vascular leakage. We found that defective adherent and tight junctions in Lztr1-depleted endothelial cells are caused by dysregulation of vesicular trafficking. LZTR1 affects the dynamics of fusion and fission of recycling endosomes by controlling ubiquitination of the ESCRT-III (endosomal sorting complex required for transport III) component CHMP1B (charged multivesicular protein 1B), whereas NS-associated LZTR1 mutations diminish CHMP1B ubiquitination. LZTR1-mediated dysregulation of CHMP1B ubiquitination triggers endosomal accumulation and subsequent activation of VEGFR2 (vascular endothelial growth factor receptor 2) and decreases blood levels of soluble VEGFR2 in Lztr1 haploinsufficient mice. Inhibition of VEGFR2 activity by cediranib rescues vascular abnormalities observed in Lztr1 knockout mice

Conclusions:

Lztr1 deletion phenotypically overlaps with bleeding diathesis observed in patients with NS. ELISA screening of soluble VEGFR2 in the blood of LZTR1-mutated patients with NS may predict both the severity of NS phenotypes and potential responders to anti-VEGF therapy. VEGFR inhibitors could be beneficial for the treatment of bleeding disorders in patients with NS.

Introduction

The incidence of Noonan syndrome (NS) is reported to be between 1 in 1000 and 1 in 2500 births.1 NS is characterized by short stature, facial dysmorphia, and congenital heart defects.2 NS is also the first cause of nuchal translucency after Down syndrome. Infants suffering from NS present varying degrees of symptoms, ranging from mild to life-threatening, including polyhydramnios, pleural effusions, and edema.3,4

Several studies have linked NS to lung lymphangiectasis, which is characterized by vessel dilatation, bronchiole collapse, and hemorrhages.5 It is usually associated with lung hypertension and can trigger pulmonary heart disease by affecting right ventricular function.6 Between 50% and 89% of patients with NS present bleeding disorders such as bruising and hemangioma,7 whereas severe hemorrhages are observed in about 3% of cases.8 Most of the studies describing the cause of bleeding disorders in patients with NS are incomplete. Coagulation defects, especially deficiencies of factors VIII, XI, XII, and various combinations of these, have been documented in about 30% of patients with NS. However, no consistent pattern of coagulation defects in NS has been established and many reports describe a poor correlation between bleeding history and clotting factor deficiencies,7 indicating that the bleeding defects in patients with NS are caused by other alterations.

An understanding of bleeding disorders in NS is clinically important because a large number of patients require surgery or take medication presenting bleeding as adverse effects, such as aspirin.9 Currently, there is no consensus on the management of bleeding complications in patients with NS. Furthermore, conventional treatment, including platelets and fresh frozen plasma, does not appear to be efficient in NS,10 demonstrating an urgent need for identification of new therapies for patients with NS.

NS is an autosomal dominant disease.11 In about half of the cases, the disease is caused by missense mutations in the PTPN11 (protein tyrosine phosphatase nonreceptor type 11) gene, resulting in a gain-of-function of the tyrosine phosphatase SHP2 (SRC homology region 2 [SH2]-containing protein tyrosine phosphatase 2),12 whereas mutations in SOS1, SOS2, RAF1, KRAS, NRAS, BRAF, SHOC2, CBL, RIT1 are found in patients with NS at low rates. LZTR1 (leucine zipper-like transcription regulator 1) has also been recently added to the list of genes causing NS and is present in up to 8% of patients with NS.13,14 Patients with NS harboring LZTR1 mutations presented the typical NS facial features, webbed neck, cardiovascular defects, and bleeding.15

LZTR1 serves as a substrate adaptor for CUL3 (CULLIN3) ubiquitin ligase complexes. In addition to NS, mutations in LZTR1 have also been associated with glioblastoma,16 hepatocarcinoma,17 familial Schwannomatosis,18 and pediatric cancers.19 In our recent study, we demonstrated that knockout of Lztr1 in mice leads to prenatal lethality, whereas Lztr1 haploinsufficiency partially recapitulates NS phenotypes including facial dimorphism and heart malformations.20 As bleeding disorders are observed in patients with NS with different genotypes, we took advantage of conventional and vascular specific Lztr1 knockout mouse models to characterize the vascular phenotypes triggered by Lztr1 loss. The results of our study provide a fundamental explanation underlying the molecular mechanisms of endothelial dysfunction and cause of the bleeding disorders in Lztr1-mutated NS as well as identify anti-VEGF (vascular endothelial growth factor) therapies as a potential treatment approach for patients with NS.

Methods

Detailed experimental procedures, mouse models, cell culture, and bioinformatics analysis are described in the Expanded Materials and Methods file in the Data Supplement. The materials used in this study are available from the corresponding author upon reasonable request. The RNAseq data generated for this study are available at GEO database with accession number GSE123044. The mass-spectrometry data have been deposited to the ProteomeXchange Consortium via the PRIDE (Proteomics Identification Database) partner repository with the data set identifier PXD011926.

Results

Lztr1 Loss Leads to Abnormal Cardiovascular Development

To elucidate the cause of bleeding disorders in LZTR1-mutated patients with NS, we analyzed the whole-body Lztr1 and the PdgfbiCreERT2, Lztr1fl/fl knockout mice. The PdgfbiCreERT2 model allows to target genes in the endothelium21,22 and is widely used to study vascular development during embryogenesis or disease progression.23,24 Vascular specific Lztr1 knockout was induced by treating PdgfbiCreERT2, Lztr1fl/fl pregnant females with tamoxifen. Survival analysis demonstrated that both the whole-body and vascular specific knockouts of Lztr1 led to embryonic lethality between E14.5 and E18.5 (Figure IA and IB in the Data Supplement). At E14.5, Lztr1−/− embryos showed growth defects, hemorrhages, or significant hydrops, whereas Lztr1+/− embryos did not show any obvious phenotypes (Figure 1A and 1B). Similarly, vascular specific Lztr1 loss led to growth retardation, superficial hemorrhages, and edema (Figure 1C and 1D). In line with these findings, yolk sac vasculature of Lztr1−/− embryos at that stage was dysfunctional as indicated by the presence of enlarged vascular structures (Figure 1E). Abnormal vascular leakage was also present in the cephalic plexus and skin vasculature (Figure 1B, 1D, and 1F).

Figure 1.

Figure 1. Loss of Lztr1 (leucine zipper-like transcription regulator) leads to vascular defects during embryogenesis.A, Representative images of Lztr1+/+, Lztr1+/−, and Lztr1−/− embryos at E14.5. The arrowhead indicates hydrops. Scale bar 1 mm. Size of E14.5 embryos is shown as mean±SEM; P values were assessed by Kruskal-Wallis test with Dunn multiple comparison test. B, Hematoxylin and eosin (H&E)-stained sections of the back skin of E14.5 Lztr1+/+, Lztr1+/−, and Lztr1−/− embryos. Arrowheads indicate hemorrhages; arrows denote dermis thickness. Scale bar 200 µm. C, Representative images of Lztr1fl/fl and PdgfbiCreERT2, Lztr1fl/fl embryos at E14.5. The arrowhead indicates nuchal translucency. Scale bar 1 mm. Size of E14.5 embryos is shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. D, H&E-stained sections of the head of Lztr1fl/fl and PdgfbiCreERT2, Lztr1fl/fl embryos at E14.5. Scale bar 100 µm (E) H&E-stained sections of Lztr1+/+, Lztr1+/− and Lztr1−/− yolk sacs. * indicates enlarged vascular structure. Scale bar 200 and 500 µm, respectively. F, Three-dimensional vasculature of brain (cephalic plexus) imaged after whole mount-staining of E14.5 embryos with Isolectin B4 (IB4), mouse iG (miG), and clearing. Scale bar 200 µm.

Lztr1 loss also caused pulmonary dysplasia. Histological analysis showed that Lztr1−/− lungs presented narrower airspaces and discontinuous lung vessels (Figure IC and ID in the Data Supplement). SMA (alpha-smooth muscle actin) and CD31 immunostaining revealed disruption of the lung structure and vasculature in Lztr1−/− embryos at E14.5 (Figure IE in the Data Supplement). Vascular dysfunction might be partially responsible for the dysmorphic lung phenotype as it was also observed in PdgfbiCreERT2, Lztr1fl/fl embryos (Online Figure IF and IG in the Data Supplement).

Furthermore, we observed defective cardiac chamber maturation in E14.5 Lztr1−/− embryos as indicated by a moderate decrease in ventricular wall thickness (Figure IIA in the Data Supplement), suggesting that cardiac failure could also be a possible cause of the Lztr1−/− embryonic lethality. The defective cardiac maturation could also explain the cardiomyopathy phenotype previously described in adult Lztr1+/− mice.20 Ventricular trabeculation and myocardial defects have been associated with endothelial dysfunction in other NS mice models.25 Concordantly, the heart of PdgfbiCreERT2, Lztr1fl/fl embryos displayed a striking noncompaction phenotype, as indicated by the hyper-trabeculation and decreased ventricular wall thickness (Figure IIB in the Data Supplement). These data indicate that Lztr1 is essential for normal vascular function and subsequent coordination of lung alveolar and heart development.

Lztr1 Loss in Adult Mice Leads to Cardiovascular Dysfunction

To examine the vascular phenotypes caused by Lztr1 loss, we focused on the analysis of the adult Lztr1+/− mice. We observed higher vessel density in the cardiac ventricles and relatively higher levels of phosphorylated VEGFR2 (VEGF receptor 2) in cardiac ventricle vessels of adult Lztr1+/− mice (Figure IIC and IID in the Data Supplement). Adult PdgfbiCreERT2, Lztr1fl/fl mice injected with tamoxifen at weaning presented severe fibrosis in the right ventricle (Figure IIE in the Data Supplement), demonstrating the endothelial contribution to the severe cardiac dysfunction previously described in Lztr1+/− mice.20

Histological analyses also demonstrated that Lztr1+/− animals presented multiple lung defects including vessel dilatation, pulmonary hemorrhages, and lung fibrosis (Figure 2A through 2D). Collectively, these results indicate that Lztr1 inactivation partially recapitulates pulmonary and cardiovascular phenotypes commonly observed in patients with NS presenting LZTR1 mutations,5,26 indicating that the Lztr1 knockout mice represent a clinically relevant model.

Figure 2.

Figure 2. Loss of Lztr1 (leucine zipper-like transcription regulator) leads to pulmonary vascular disease.A, Gross anatomy and hematoxylin and eosin (H&E) staining of lungs of 52-week old Lztr1+/+ and Lztr1+/− mice. Scale bar 500 µm. * indicate bronchioles; pulmonary vessels are delineated in red. B, Masson Trichrome–stained sections of Lztr1+/+ and Lztr1+/− lungs. Fibrosis is colored in blue. Scale bar 500 μm. C, Fibrotic area in lungs of 52-week old Lztr1+/+ and Lztr1+/− mice. Data are shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. D, H&E-staining of lungs of 10-week old Lztr1+/+ and Lztr1+/− mice. Scale bar 500 µm. Hemorrhagic area is shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. E, Lungs of 52-week old Lztr1+/+ and Lztr1+/− mice immunostained with anti-SMA and anti-CD31 antibodies. Data are shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. Scale bar 20 µm. F, ECs isolated from Lztr1+/+ and Lztr1+/− mice immunostained with antibodies against VE-cadherin and ZO1 (zonula occludens-1). Blue rectangles delineate the magnified areas (below). Data are shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. Scale bar 20 µm. G, Evans blue extravasation from lungs of Lztr1+/+ and Lztr1+/− mice. Lines shows mean; P values were assessed by Wilcoxon-Mann-Whitney test. H, Evans blue extravasation from Lztr1fl/fl and PdgfbiCre ERT2, Lztr1fl/fl tissues. Lines shows mean; P values were assessed by Wilcoxon-Mann-Whitney test. I, Paracellular permeability of Lztr1+/+ and Lztr1+/− ECs as measured by quantifying FITC (fluorescein isothiocyanate) Dextran fluorescence in a trans-well assay. P values were assessed by 2-way ANOVA with Bonferroni post-tests.

CD31 immunostaining showed that lung vessels in Lztr1+/− mice were thinner and discontinuous (Figure 2E). Immunostaining of isolated endothelial cells (ECs) demonstrated that Lztr1 haploinsufficiency led to abnormal adherent and tight junctions (Figure 2F; Figure IIF in the Data Supplement). The Evans blue extravasation assay revealed increased vascular permeability in the lungs of Lztr1+/− mice (Figure 2G). We observed an even more striking increase in vascular permeability in the lungs and skin of PdgfbiCreERT2, Lztr1fl/fl mice injected with tamoxifen at weaning (Figure 2H). In line with these findings, Lztr1+/− sorted ECs were more permeable to 40 kDa Dextran when compared with Lztr1+/+ ECs (Figure 2I).

The observed phenotypes suggest a potential role of LZTR1 in the control of vascular function. Concordantly, aortic ring assay27 demonstrated that Lztr1+/− ECs were more neo-angiogenic as indicated by an increase in the sprouting area (Figure IIIA through IIIC in the Data Supplement). To further assess the angiogenic potential of Lztr1+/− ECs in vivo, we implanted Lewis lung carcinoma cells28 in the flanks of Lztr1+/+ and Lztr1+/− mice. Lung carcinoma tumors appeared earlier in Lztr1+/− mice and displayed a higher weight at euthanization (Figure IIID and IIIE in the Data Supplement). Histological analyses showed that lung carcinoma tumors in Lztr1+/− mice were more vascularized and presented increased vascular leakage (Figure IIIF through IIII in the Data Supplement). Aortic ring assay using aortic rings isolated from PdgfbiCreERT2, Lztr1fl/fl mice injected with tamoxifen at weaning also demonstrated that vascular specific loss of Lztr1 was sufficient to increase neo-angiogenesis (Figure IIIJ in the Data Supplement). These data indicate that Lztr1 deletion disrupts the balance between vascular integrity and angiogenesis, leading to a more proangiogenic profile associated with less stable junctions.

LZTR1 Controls Endothelial Function by Regulating Vesicular Trafficking

We next explored a potential mechanism by which Lztr1 loss leads to a proangiogenic phenotype. In our recent study, we found that LZTR1 controls the activity of the RAS pathway.20 Indeed, we observed increased phosphorylation of ERK1/2 (extracellular signal-regulated kinase 1/2) and AKT1 (AKT serine/threonine kinase 1) when we deleted Lztr1 in Lztr1fl/fl ECs derived from different mice by overexpressing CreERT2 recombinase (Figure IVA and IVB in the Data Supplement). However, the MEK1/2 (mitogen-activated protein kinase kinases 1 and 2) inhibitor pimasertib, the AKT inhibitor ipatasertib, or the combination of both inhibitors only partially rescued the embryonic lethality phenotype, as upon either of the treatments we recovered Lztr1−/− embryos at E18.5, but no pups at birth (Figure IVC in the Data Supplement). Treatment with pimasertib or ipatasertib did not rescue lung dysplasia in E18.5 Lztr1−/− embryos. E18.5 Lztr1−/− embryos treated with the combination of pimasertib and ipatasertib showed more moderate lung dysplasia, but still displayed massive lung hemorrhages, indicating that concurrent inhibition of MEK1/2 and AKT1 in Lztr1−/− embryos could not restore normal lung vascular function (Figure IVD and IVE in the Data Supplement). Moreover, pimasertib, ipatasertib, or the combination of both inhibitors did not rescue the formation of abnormal adherent and tight junctions in Lztr1fl/fl-CreERT2 ECs (Figure IVF in the Data Supplement). This suggests that Lztr1 loss may contribute to human disease by additional mechanisms rather than the sole activation of the RAS pathway.

Using mass-spectrometry based approaches, we explored how LZTR1 loss alters the proteome and ubiquitome landscapes. The Ingenuity Pathway Analysis of either differentially expressed or differentially ubiquitinated proteins in HeLa cells harboring wt-LZTR1 or LZTR1-indels demonstrated that endocytosis was one of the top canonical pathways altered by LZTR1 loss (Figure 3A and 3B). Several putative LZTR1 interactors identified by the Virotrap approach20 were also associated with vesicular trafficking (Figure 3C). Collectively, these data suggest a potential role of LZTR1 in the regulation of vesicular trafficking.

Figure 3.

Figure 3. Lztr1 (leucine zipper-like transcription regulator) loss affects vesicular trafficking.A, Ingenuity Pathway Analysis (IPA) analysis of differentially expressed proteins in HeLa cells harboring wt-LZTR1 or LZTR1-indels. Differentially expressed proteins were identified by MS-based shot gun analysis using the FDR approach (N=3). B, IPA analysis of differentially ubiquitinated peptides in HeLa cells harboring wt-LZTR1 or LZTR1-indels. GG-modified peptides were captured by anti-K-ε-GG antibody and quantified by MS. Differentially ubiquitinated proteins were identified by the FDR approach (N=3). C, Putative LZTR1 interactors identified by the Virotrap approach as described in Steklov et al,20 2018. Endocytosis-related proteins are shown in red (N=3). D and E, Lztr1fl/fl ECs expressing an empty vector (EV) or Cre-recombinase were immunostained with anti-SNX1 antibody. Scale bar 10 µm. Each dot shows mean value for 25 ECs isolated from one mouse. Lines show mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. Sphericity ranges from 0 to 1 indicating if the structures are spheric (close to 1) or more tubular (close to 0.5). F and G, Lztr1fl/fl ECs transduced with an EV or Cre-recombinase coding virus (Cre) were treated with DMSO (dimethyl sulfoxide), Dynasore (40 µmol/L), or TyrA23 (10 µmol/L) for 12 h and immunostained with anti-ZO1 (zonula occludens-1) and anti-VE-cadherin antibody. Scale bar 20 µm. Quantification of the number of junctional gaps per cells. N=4; 25 cells per group were analyzed; P values were assessed by Kruskal Wallis test with Dunn multiple comparison test (H) progeny from Lztr1+/− mattings at E18.5 and after birth in a mock group or a Dynasore-treated group (1 mg/kg; starting from E2.5). P values were calculated using Fisher Exact test. AKT indicates serine/threonine-protein kinase; CAND1, cullin-associated and neddylation dissociated 1; CHMP1B, charged multivesicular protein 1B; CLIC1, chloride intracellular channel 1; CKB, creatine kinase B; CUL3, CULLIN3; DDX6, DEAD-box helicase 6; EIF2, eukaryotic initiation factor 2; HRAS, HRas proto-oncogene; ILK, integrin-linked kinase; LIMS1, LIM zinc finger domain containing 1; mTOR, mammalian target of rapamycin; NRAS, NRAS proto-oncogene; PI3K, phosphoinositide 3-kinase; PLSCR1, phospholipid scramblase 1; SCYL2, SCY1-like pseudokinase 2; SLC44A1, solute carrier family 44 member 1; TXN, thioredoxin; VPS4A, vacuolar protein sorting-associated protein 4A; and VTA1, vacuolar protein sorting-associated protein VTA1.

To test this idea, we performed immunostaining of LZTR1-deleted cells for SNX1 (sorting nexin-1), a protein essential for early to late endosome trafficking. Lztr1 knockout in Lztr1fl/fl ECs infected with a CreERT2 recombinase led to abnormal endosomal tubulation, a phenotype associated with vesicular trafficking impairment (Figure 3D and 3E). Concordantly, SNX1-positive structures appeared spherical in wt-LZTR1 HeLa cells, whereas LZTR1 loss led to the accumulation of endosomal tubules (Figure VA in the Data Supplement). These results implicate LZTR1 in the control of endosomal maturation and recycling.29,30

To ascertain the role of LZTR1 in vesicular trafficking, we performed a set of rescue experiments using the inhibitors of vesicular trafficking, Dynasore,31,32 and TyrA23 (tyrphostin A23).33,34 Both treatments abolished the difference in the appearance of endosomal tubules between wt-LZTR1 and LZTR1-deleted cells (Figure 3D and 3E; Figure VA in the Data Supplement). Furthermore, both inhibitors normalized adherent and tight junctions in Lztr1fl/fl-CreERT2 ECs (Figure 3F and 3G). Depletion of Dynamin I with a siRNA (Figure IVG in the Data Supplement) also restored normal junctions in Lztr1fl/fl-CreERT2 ECs (Figure IVG through IVI in the Data Supplement). Prenatal treatment with Dynasore partially rescued the embryonic lethality phenotype caused by Lztr1 knockout, as we observed a nearly normal ratio of Lztr1−/− embryo at E18.5, but not at birth (Figure 3H). In contrast to the treatment with either pimasertib or ipatasertib, Dynasore recovered lung dysplasia in E18.5 Lztr1−/− embryos (Figure IVD and IVE in the Data Supplement). Together, these results indicate that LZTR1-mediated vascular dysfunction could be explained by defects in vesicular trafficking.

LZTR1 Controls Ubiquitination of the ESCRT-III Component CHMP1B

We next assessed the molecular mechanism by which LZTR1 is implicated in vesicular trafficking. The component of ESCRT-III (endosomal sorting complex required for transport III) trafficking machinery, CHMP1B (charged multivesicular protein 1B), was present among both the putative LZTR1 interactors identified by the Virotrap screen.20 CHMP1B also showed one of the highest scores in the fold change ranking when we compared ubiquitination landscapes of wt-LZTR1 and LZTR1-indel cells. Apart from the lysine 6 site, all CHMP1B ubiquitination sites (lysines 42, 59, 87, 90, 104, and 110) showed at least a 2-fold decrease in LZTR1-indel cells (Figure 4A). This suggests that LZTR1 might regulate vesicle trafficking by controlling CHMP1B ubiquitination.

Figure 4.

Figure 4. LZTR1 (leucine zipper-like transcription regulator 1) regulates CHMP1B (charged multivesicular protein 1B) ubiquitination.A, Ubiquitinome analysis of wt-LZTR1 or LZTR1-indel HeLa cells, N=3. For each protein, log2 (fold change between both conditions) is plotted vs Fold Change Rank. CHMP1B ubiquitination sites are shown in red. B and C, An empty vector (EV), HA-tagged wt-LZTR1, or LZTR1 mutants were overexpressed in HeLa cells. Endogenous CHMP1B was immunoprecipitated using anti-CHMP1B antibody conjugated to agarose. The amount of pulled-down LZTR1 was detected by immunoblotting with anti-HA antibody. D, CHMP1B was pulled down from ECs expressing HA-wt-LZTR1 using an anti-CHMP1B antibody. The amount of pulled-down HA-wt-LZTR1 was detected by immunoblotting. E, Ubiquitinated proteins were immunoprecipitated from Lztr1+/+ and Lztr1+/− ECs isolated using anti-ubiquitin FK2 antibody. The amount of pulled-down CHMP1B was detected by immunoblotting. F, Ubiquitinated proteins were immunoprecipitated from ECs treated with VEGF (vascular endothelial growth factor; 100 ng/ml, 30 min) or MLN4924 (1 µmol/L, 12 h) using anti-Ubiquitin FK2 antibody. The amount of pulled-down CHMP1B was detected by immunoblotting. G, Ubiquitinated proteins were immunoprecipitated from HeLa cells with Lztr1-indel overexpressing EV, wt-LZTR1, or NS-associated LZTR1 mutants using anti-ubiquitin FK2 antibody conjugated to agarose. The amount of pulled-down CHMP1B was detected by immunoblotting. * indicates ubiquitinated CHMP1B; and WCL, whole cell lysate.

We confirmed the interaction of endogenous CHMP1B and HA-tagged LZTR1 using co-immunoprecipitation (Figure 4B and 4C). Of note, CHMP1B forms homo- and heteromers and is detected in cell lysates as a monomer (28kDa), a dimer (55kDa), and a multimer even under denaturating conditions, which is consistent with previously published results.35 Moreover, a recent report demonstrated that stimulation with growth factors or cytokines promotes CHMP1B ubiquitination.35 Concordantly to this observation, we found increased binding of LZTR1 to CHMP1B in ECs upon stimulation with VEGF (Figure 4D). On the contrary, treatment with the CULLIN neddylation inhibitor MLN4924, which blocks the activity of the LZTR1/ CUL3 complex, did not affect the interaction between LZTR1 and CHMP1B, indicating that their binding does not depend on the ubiquitination status of CHMP1B (Figure 4D).

We then assessed LZTR1 contribution to the control of CHMP1B ubiquitination. CHMP1B was present in mono- or multi-ubiquitinated forms. This is consistent with the mass-spectrometry results showing ubiquitination of CHMP1B at 7 different lysines. Substitution of 7 lysines to arginines abolished CHMP1B ubiquitination (Figure VB in the Data Supplement). We found that LZTR1-indel in HeLa cells also led to decreased ubiquitination of V5-tagged wild-type and endogenous CHMP1B (Figure VC in the Data Supplement). Moreover, we observed lower levels of CHMP1B ubiquitination upon Lztr1 knockout in mice when comparing Lztr1+/+ and Lztr1+/− yolk sacs; Lztr1+/+ and Lztr1+/− sorted ECs; and Lztr1fl/fl ECs expressing an empty vector or CreERT2 recombinase (Figure 4E; Figure VD and VE in the Data Supplement). Increased binding of LZTR1 to CHMP1B upon VEGF stimulation in ECs was associated with increased levels of ubiquitinated CHMP1B, whereas MLN4924 treatment abolished VEGF-induced CHMP1B ubiquitination (Figure 4F), further confirming that LZTR1 controls CHMP1B ubiquitination in a Cullin-dependent manner.

Noonan-Associated LZTR1 Mutations Diminish the Ability of LZTR1 to Ubiquitinate CHMP1B

We examined whether NS-associated phenotypes could also be associated with the impaired ability of mutated LZTR1 to control CHMP1B ubiquitination. Both truncated and missense mutations of LZTR1 have been reported in NS. Missense mutations of LZTR1 are spread throughout the whole gene.13,36,37 In our recent study, we found that LZTR1 mutations within the BTB (broad-complex, tramtrack and bric-a-brac)-BACK domains abolish the formation of the LZTR1/CUL3 complex, suggesting that the BTB-BACK LZTR1 mutations abolish ubiquitination of all LZTR1 substrates.20 On the contrary, LZTR1 mutations within the Kelch domain, which is responsible for the substrate binding, might diminish the ability of LZTR1 to ubiquitinate CHMP1B by affecting the interaction between LZTR1 and CHMP1B. Therefore, we tested the ability of the Kelch domain LZTR1 mutants to bind and ubiquitinate CHMP1B. We found that all Kelch domain LZTR1 mutations, except for M202R, showed decreased ability to interact with CHMP1B (Figure 4B and 4C). Importantly, 3 LZTR1 mutations, R97L, N145L, and R284C, observed in patients with NS showing bleeding phenotype,38,39 caused the inability of LZTR1 to ubiquitinate CHMP1B (Figure 4G). This suggests that dysregulation of CHMP1B ubiquitination could be associated with the bleeding disorders observed in LZTR1-mutated patients with NS.

Dysregulation of LZTR1-Mediated Ubiquitination of CHMP1B Impedes Vesicle Trafficking

Our next question was how dysregulation of LZTR1-mediated ubiquitination of CHMP1B affects vesicular trafficking. The most well-described function of CHMP1B is to promote the formation of recycling tubules from endosomes40 by co-assembling with the other ESCRT-III component IST1 (increased sodium tolerance 1).41,42 Indeed, Chmp1b depletion led to the dysregulation of recycling trafficking as we detected an accumulation of RAB11 (Ras-related protein RAB11)-positive recycling endosomes (Figure 5A and 5B). Lztr1 knockout in ECs presented a similar accumulation of RAB11-positive endosomes. We also found an accumulation of CHMP1B/RAB11-positive recycling endosomes in Lztr1+/− ECs and Lztr1fl/fl ECs infected with a CreERT2 recombinase (Figure 5C through 5E). In contrast, we did not observe an increase of CHMP1B in early EAA1-positive or late LAMP1-positive endosomes (Figure 5D and 5E). On the contrary, suppression of Chmp1b in Lztr1−/− ECs did not affect the localization of RAB11 (Figure 5A and 5B), suggesting that the effect of CHMP1B on trafficking is LZTR1 dependent.

Figure 5.

Figure 5. LZTR1 (leucine zipper-like transcription regulator 1)-mediated CHMP1B (charged multivesicular proteins 1B) ubiquitination modulates vesicle trafficking.A, Immunostaining of Lztr1+/+ and Lztr1+/− ECs transfected with a Scrambled siRNA or siRNA targeting Chmp1b with antibodies against RAB11, CHMP1B, and VEGFR2 (vascular endothelial growth factor receptor 2). Scale bar 20 µm. B, Quantification of endosomal RAB11 levels in Lztr1+/+ and Lztr1+/− ECs transfected with a Scrambled siRNA or siRNA targeting Chmp1b. Data are shown as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test. More than 25 cells per mouse were analyzed. C, Lztr1fl/fl ECs expressing an empty vector (EV) or Cre-recombinase were immunostained with the indicated antibodies. Scale bar 10 µm. D, ECs isolated from Lztr1+/+ and Lztr1+/− mice immunostained with the indicated antibodies. Scale bar 10µm. E, Quantification of co-localization of CHMP1B and endosomal markers, EAA1 (early endosome antigen 1), RAB11, and LAMP1 (lysosomal-associated membrane protein 1), in Lztr1+/+ and Lztr1+/− ECs is shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test. Twenty-five cells per mice were analyzed. F, An EV, V5-tagged wt-CHMP1B, or CHMP1B-7K>R mutant was overexpressed in wt-LZTR1 or LZTR1-indels HeLa cells expressing shCHMP1B. Quantification of endosomal RAB11 is shown as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test, N=4.

We further confirmed the effect of LZTR1-mediated CHMP1B ubiquitination on vesicle trafficking by rescue experiments. Whereas suppression of CHMP1B led to the accumulation of RAB11-positive vesicles, overexpression of wild-type CHMP1B restored the trafficking of recycling endosomes. On the contrary, ubiquitination-deficient CHMP1B-7K>R mutant did not rescue the accumulation of RAB11-positive endosomes triggered by CHMP1B depletion. Finally, modulation of CHMP1B expression in LZTR1-indels cells showed no effect on recycling endosome trafficking (Figure 5F, Figure VF and VG in the Data Supplement), indicating that LZTR1-mediated ubiquitination of CHMP1B plays a crucial role in vesicular trafficking.

We also assessed the mechanism by which LZTR1-mediated ubiquitination of CHMP1B affects its functioning. Consistently to a recent study,35 we did not observe any degradative poly-ubiquitination of CHMP1B (Figure 4E and 4F; Figure VB through VE in the Data Supplement). On the contrary, it was reported that USP8 (ubiquitin-specific peptidase 8)-mediated deubiquitination of CHMP1B promotes its co-assembly with IST1.35 In line with this observation, LZTR1 loss increased the interaction between CHMP1B and IST1 (Figure VH in the Data Supplement). Collectively, these results indicate that LZTR1 loss affects recycling trafficking by abolishing CHMP1B ubiquitination and dysregulating the assembly of the CHMP1B/IST1 complex, a key component of the ESCRT-III complex necessary for endosomal trafficking.

Lztr1 Loss Activates the VEGF Pathway by Dysregulating VEGFR2 Trafficking

As a component of the ESCRT-III complex, CHMP1B is implicated in the trafficking of stimulated growth factor receptors, suggesting that LZTR1-mediated dysregulation of CHMP1B could affect receptor signaling. Consistent with this idea, the timing of VEGF-induced CHMP1B ubiquitination coincided with the onset of VEGFR2 internalization and activation (Figure 4D and 4F), suggesting that CHMP1B ubiquitination might regulate VEGFR2 trafficking. Moreover, we observed higher levels of phosphorylated VEGFR2 in Lztr1−/− heart vessels, suggesting that LZTR1 might contribute to the regulation of VEGFR2 activity (Figure IID in the Data Supplement). The Ingenuity Pathway Analysis Upstream Regulator Analysis of genes differentially expressed in Lztr1+/+ and Lztr1+/− mouse embryonic fibroblasts also predicted VEGF activation in Lztr1 knockout cells (Figure VIA in the Data Supplement). Besides, the VEGF/VEGFR2 pathway plays a central role in endothelial cell function; and among the angiogenic drugs, the VEGFR2 specific inhibitors have produced notable results in different diseases.43–45 Therefore, we decided to specifically focus on VEGF signaling.

We assessed the effect of Lztr1 on the intracellular localization of VEGFR2 by performing immunocytochemistry analysis of VEGFR2 in ECs with or without permeabilization. Whereas VEGFR2 was strongly expressed on the cell surface of Lztr1+/+ ECs, we detected less VEGFR2 signal on the surface of Lztr1+/− ECs (Figure VIB and VIC in the Data Supplement). On the contrary, we observed an accumulation of VEGFR2 on RAB11/CHMP1B-positive vesicles in Lztr1-deleted ECs (Figure 6A through 6C). A similar accumulation of VEGFR2 on endosomes was observed in Chmp1b-depleted cells, indicating abnormal trafficking of the receptor (Figures 5A and 6C). It has been recently reported that the constitutive VEGFR2 internalization protects the receptor against ectodomain cleavage,46 suggesting that Lztr1 loss might result in decreased VEGFR2 shedding. Indeed, ELISA analysis revealed lower levels of soluble VEGFR2 in the blood serum of Lztr1+/− mice (Figure 6D). Treatment with Dynasore or TyrA23 abolished endosomal accumulation of VEGFR2 in Lztr1+/− ECs (Figure 6A and 6B). Dynamin I suppression also reduced internalization of the receptor in Lztr1fl/fl ECs expressing CreERT2 (Figure VID and VIE in the Data Supplement). Altogether, these results indicate that dysregulation of LZTR1-mediated ubiquitination of CHMP1B could play a critical role in the regulation of VEGFR2 trafficking.

Figure 6.

Figure 6. LZTR1 (leucine zipper-like transcription regulator 1)-mediated VEGFR2 (vascular endothelial growth factor receptor 2) internalization affects blood vessel integrity.A and B, Lztr1fl/fl ECs were infected with an empty vector (EV) or Cre-recombinase (Cre) and treated with DMSO, Dynasore (40 µmol/L), or TyrA23(10 µmol/L) for 12 h. After permeabilization, the cells were immunostained with anti-VEGFR2 antibody. Scale bar 20 µm. Data shown as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test. C, Levels of internalized VEGFR2 were assessed by immunostaining of Lztr1+/+ and Lztr1+/− ECs transfected with Scrambled siRNA or siRNA targeting Chmp1b. Data are presented as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test. D, ELISA analysis of soluble VEGFR2 in the plasma of Lztr1+/+ and Lztr1+/− mice. Data are present as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test. E and F, Immunoblotting analysis of phosphorylated and total VEGFR2 in Lztr1fl/fl ECs infected with an EV or Cre-recombinase (Cre) and treated with DMSO, MLN4924 (1 µmol/L), or Tyrphostin A23(10 µmol/L) for 12 h. Data shown as mean±SEM; P values were assessed by Mann-Whitney test. G and H, Lztr1fl/fl ECs transduced with EV or Cre-recombinase (Cre), treated for 12 h with DMSO or cediranib (10 µmol/L), and immunostained with antibodies against ZO1 (zonula occludens-1) and VE-cadherin. Scale bar 20 µm. Quantification of the number of cells with junctional gaps is shown as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test to compare efficiency of each specific treatment within each group. I, Evans blue extravasation from PdgfbiCreERT2, Lztr1fl/fl and Lztr1fl/fl mice skin after treatment with phosphate-buffered saline or VEGF. Miles assay performed on the ears, N=4 mice per group. Data are normalized to tissue weight and shown as mean±SEM; P values were assessed by Wilcoxon Mann-Whitney test to compare efficiency of treatment within group. J, Evans blue extravasation from PdgfiCreERT2, Lztr1fl/fl and Lztrfl/fl mice skin after vehicle or cediranib treatment. Miles assay, N=4 mice per group. Data are normalized to tissue weight and shown as mean±SEM; P values were assessed by Wilcoxon-Mann-Whitney test to compare efficiency of treatment within group.

Because internalization and phosphorylation of the VEGFR2 complex are hallmarks of VEGF signaling activation,47,48 we hypothesized that LZTR1-mediated dysregulation of VEGFR2 trafficking might lead to its activation. Concordantly, we detected a higher level of phosphorylated VEGFR2 in Lztr1+/− ECs and Lztr1+/− yolk sacs (Figure VIF, VIG, and VII in the Data Supplement). Lztr1 knockout in Lztr1fl/fl ECs by overexpressing CreERT2 recombinase also led to increased levels of VEGFR2 phosphorylation (Figure VIH and VII in the Data Supplement). Suppression of Dynamin I expression partially rescued increased levels of VEGFR2 phosphorylation in Lztr1-depleted cells (Figure VIK in the Data Supplement). VEGFR2 activation was CULLIN-dependent as MLN4924 abolished the accumulation of phosphorylated VEGFR2 (Figure 6D and 6E). TyrA23 treatment partially rescued VEGFR2 activation in Lztr1 knockout ECs, indicating that increased VEGFR2 activity is due to dysregulated endosomal trafficking (Figure 6E and 6F).

To assess whether enhanced VEGFR signaling is responsible for the vascular phenotypes in Lztr1 knockout mice, we blocked VEGFR2 activity in Lztr1fl-fl-CreERT2 ECs with cediranib.49,50 Cediranib treatment normalized adherent and tight junctions (Figure 6G and 6H), confirming that Lztr1 deletion affects junctional integrity through VEGFR activation. Furthermore, VEGF treatment significantly increased Evans blue extravasation in the skin of Lztr1fl/fl mice, but only moderately in PdgfbiCreERT2, Lztr1fl/fl mice, confirming that the pathway is already activated in Lztr1-deleted cells (Figure 6I). Importantly, blocking of VEGFR2 activity with cediranib normalized tight and adherent junctions and rescued vascular leakage in the skin of PdgfbiCreERT2, Lztr1fl/fl mice (Figure 6J). Altogether, these data indicate that loss of Lztr1 leads to the upregulation of VEGFR2 signaling by dysregulating vesicular trafficking, thus implying that blocking the catalytic activity of VEGFR could be advantageous for patients with NS.

Discussion

Nearly two-thirds of patients with NS exhibit clinical signs of a bleeding diathesis that may manifest with easy bruising, postoperative bleeding complications, and rarely spontaneous hemorrhages.26 Here we explored the role of LZTR1, a gene recently linked to NS development, in the regulation of vascular function. We found that Lztr1 whole-body knockout mice died in utero, suffering from severe hemorrhages and in some cases, hydrops. This is in line with previous observations from other NS models.25,51 Similarly, Ptpn11D61G and Sos1E846K embryos show hemorrhages at E13.5.51,52Lztr1 haploinsufficiency in adult mice also partially recapitulates bleeding phenotypes observed in patients with NS, especially in the context of novel studies making the link between LZTR1 mutations and coagulation defects.38,39

Patients with PTPN11 mutations show an increased prevalence of bleeding disorders compared with those having other mutations.12 On the contrary, bleeding disorders in NS appear to be least correlated with KRAS and RAF1 mutations.1,53 This suggests that bleeding phenotypes might be associated with RAS-independent functions of the NS genes. Concordantly to this idea, PTPN11, which encodes the protein tyrosine phosphatase SHP2, protects endothelial barrier through VE-cadherin stabilization.54 Here, we also found that MEK1/2 inhibition does not rescue vascular dysfunction induced by Lztr1 loss. Unbiased proteome and ubiquitome analyses revealed that endocytosis is one of the top canonical pathways altered by Lztr1 loss. Importantly, treatment with either Dynasore, TyrA23, or dynamin I siRNA rescues endothelial dysfunction mediated by Lztr1 loss, confirming that LZTR1-mediated vascular phenotypes are caused by dysregulation of vesicular trafficking.

Unbiased interaction screen and ubiquitome analysis indicate that the key component of the ESCRT-III complex CHMP1B could be a direct substrate of the LZTR1/CUL3 complex. A previous study demonstrated that decreased ubiquitination of CHMP1B appears to block membrane fission by affecting its interaction with IST1.55 Concordantly, LZTR1 loss appears to reduce CHMP1B ubiquitination on lysine residues important for its interaction with IST1.41,42 CHMP1B and IST1 form a 3-dimensional structure that gives the plasticity to membrane invagination required for the fusion and fission of vesicles. In fact, LZTR1 loss strengthens the binding of CHMP1B to IST1, resulting in inhibition of membrane fission. Therefore, we could speculate that LZTR1-triggered defects in vesicular trafficking might be due to the dysregulation of the interaction between CHMP1B and IST1 that may affect the disassembly of the ESCRT-III complex. Importantly, all LZTR1 mutations found in patients with NS with bleeding disorders led to a decreased ability to interact with CHMP1B. Nonetheless, we could not exclude the possibility that LZTR1 might modulate other regulators of vesicular trafficking.

Dysregulation of vesicle trafficking might lead to the activation of several receptor pathways.56 Here, we demonstrated that Lztr1 loss activates VEGFR2 signaling, which is the primary tyrosine kinase receptor transmitting VEGF signals to control the balance between endothelial plasticity and junctional stability. Increased internalization and phosphorylation of VEGFR2 in Clathrin-dependent vesicles led to the activation of angiogenesis and increased endothelial permeability.32,57 Therefore, the Lztr1-mediated bleeding phenotypes could be at least partially explained by the activation of VEGF signaling. Consistent with this idea, overexpression of VEGFA results in severe endothelial dysfunction, such as abnormal angiogenesis and vascular leakage during development,58,59 leading to the accumulation of endocytic vesicles containing active VEGFR2.

Even though inhibition of the MAPK (mitogen-activated protein kinase) and PI3K (phosphoinositide 3-kinase)/AKT pathways did not rescue endothelial dysfunction observed in Lztr1 knockout mice, VEGFR2 led to the activation of multiple signaling pathways.60 As an example, VEGFR2 recruits the adaptor protein NCK1 (Nck adaptor protein 1) and the SRC family tyrosine kinase FYN (SRC family proto-oncogene kinase Fyn). NCK1/FYN complex formation regulates phosphorylation of PAK2 (p21-activated kinase 2), which in turn activates CDC42 and p38 MAPK, leading to actin remodeling. Recruitment of PLCγ1 (1-phosphatidylinositol 4,5-bisphosphate phosphodiesterase gamma-1) and SHB (SH2-domain-containing adaptor protein B) by VEGFR2 facilitates the interaction with FAK (focal adhesion kinase) and contributes to endothelial cell migration and attachment. SHB activation of PI3K induces eNOS (endothelial nitric oxide synthase) that promotes cell survival and nitric oxide-induced vascular permeability, respectively.

Current treatments for patients with NS suffering from bleeding including platelets and fresh frozen plasma do not appear to be efficient in NS.10 Several anti-VEGF approaches have been recently validated to limit blood loss, including angiogenesis inhibitors such as bevacizumab, an antibody targeting VEGF, cediranib,61,62 or thalidomide.63 Our results strongly indicate that inhibiting the catalytic activity of VEGF receptors could be beneficial to reduce bleeding in patients with NS. Moreover, we found that Lztr1 loss leads to increased levels of soluble VEGFR2 in blood serum, suggesting that ELISA-based screening of patients with NS for the blood levels of soluble VEGFR2 might allow predicting potential responders to anti-VEGF therapy. Collectively, our results provide a potential explanation for the role of LZTR1 in vascular function and suggest novel therapeutic approaches for patients with NS.

Nonstandard Abbreviations and Acronyms

BTB

broad-complex, tramtrack and bric-a-brac

CHMP1B

charged multivesicular protein 1B

CUL3

CULLIN3

EC

endothelial cell

eNOS

endothelial nitric oxide synthase

ESCRT-III

endosomal sorting complex required for transport III

FAK

focal adhesion kinase

IST1

increased sodium tolerance 1

NCK1

Nck adaptor protein 1

NS

Noonan syndrome

SHB

SH2-domain-containing adaptor protein B

SNX1

sorting nexin-1

TyrA23

tyrphostin A23

VEGF

vascular endothelial growth factor

VEGFR2

VEGF receptor 2

Acknowledgments

R.N. Sewduth, S. Pandolfi, and B. Lechat performed mice experiments; R.N. Sewduth performed biochemical, cellular, and immunostaining experiments; A.A. Sheryazdanova and R. Quarck generated cell lines; N. Criem performed the molecular cloning; B. Lechat performed the mouse genotyping; F. Impens and M. Steklov performed the proteomics experiments; F. Impens and P. Zhao performed the bioinformatics analysis; R.N. Sewduth and A.A. Sablina designed the study; R.N. Sewduth, N. Criem, M.F. Baietti, and A.A. Sablina analyzed the data and wrote the article. All authors discussed the results and commented on the article.

Supplemental Materials

Expanded Materials & Methods

Major Resources Table

Online Figures I–VI

Uncropped Blots

Footnotes

For Sources of Funding and Disclosures, see page 1391.

The Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/CIRCRESAHA.119.315730.

Correspondence to: Anna Sablina, PhD, VIB-KU Leuven Center for Cancer Biology, VIB/KULeuven, Herestraat 49, 3000 Leuven, Belgium. Email

References

  • 1. Roberts AE, Allanson JE, Tartaglia M, Gelb BD. Noonan syndrome.Lancet. 2013; 381:333–342. doi: 10.1016/S0140-6736(12)61023-XCrossrefMedlineGoogle Scholar
  • 2. Sznajer Y, Keren B, Baumann C, Pereira S, Alberti C, Elion J, Cavé H, Verloes A. The spectrum of cardiac anomalies in Noonan syndrome as a result of mutations in the PTPN11 gene.Pediatrics. 2007; 119:e1325–e1331. doi: 10.1542/peds.2006-0211CrossrefMedlineGoogle Scholar
  • 3. Bakker M, Pajkrt E, Mathijssen IB, Bilardo CM. Targeted ultrasound examination and DNA testing for Noonan syndrome, in fetuses with increased nuchal translucency and normal karyotype.Prenat Diagn. 2011; 31:833–840. doi: 10.1002/pd.2782CrossrefMedlineGoogle Scholar
  • 4. Nisbet DL, Griffin DR, Chitty LS. Prenatal features of Noonan syndrome.Prenat Diagn. 1999; 19:642–647. doi: 10.1002/(sici)1097-0223(199907)19:7<642::aid-pd610>3.0.co;2-1CrossrefMedlineGoogle Scholar
  • 5. Hernandez RJ, Stern AM, Rosenthal A. Pulmonary lymphangiectasis in Noonan syndrome.AJR Am J Roentgenol. 1980; 134:75–80. doi: 10.2214/ajr.134.1.75CrossrefMedlineGoogle Scholar
  • 6. Witt DR, Hoyme HE, Zonana J, Manchester DK, Fryns JP, Stevenson JG, Curry CJ, Hall JG. Lymphedema in Noonan syndrome: clues to pathogenesis and prenatal diagnosis and review of the literature.Am J Med Genet. 1987; 27:841–856. doi: 10.1002/ajmg.1320270412CrossrefMedlineGoogle Scholar
  • 7. Nugent DJ, Romano AA, Sabharwal S, Cooper DL. Evaluation of bleeding disorders in patients with Noonan syndrome: a systematic review.J Blood Med. 2018; 9:185–192. doi: 10.2147/JBM.S164474CrossrefMedlineGoogle Scholar
  • 8. Massarano A, Wood A, Tait R, Stevens R, Super M. Noonan syndrome: coagulation and clinical aspects.Acta Paediatr. 2010; 85:1181–1185. doi:10.1111/j.1651-2227.1996.tb18225.xCrossrefGoogle Scholar
  • 9. Sharland M, Patton MA, Talbot S, Chitolie A, Bevan DH. Coagulation-factor deficiencies and abnormal bleeding in Noonan’s syndrome.Lancet. 1992; 339:19–21. doi: 10.1016/0140-6736(92)90141-oCrossrefMedlineGoogle Scholar
  • 10. Tofil NM, Winkler MK, Watts RG, Noonan J. The use of recombinant factor VIIa in a patient with Noonan syndrome and life-threatening bleeding.Pediatr Crit Care Med. 2005; 6:352–354. doi: 10.1097/01.PCC.0000160656.71424.D1CrossrefMedlineGoogle Scholar
  • 11. Tartaglia M, Gelb BD, Zenker M. Noonan syndrome and clinically related disorders.Best Pract Res Clin Endocrinol Metab. 2011; 25:161–179. doi: 10.1016/j.beem.2010.09.002CrossrefMedlineGoogle Scholar
  • 12. Tartaglia M, Mehler EL, Goldberg R, Zampino G, Brunner HG, Kremer H, van der Burgt I, Crosby AH, Ion A, Jeffery S, et al. Mutations in PTPN11, encoding the protein tyrosine phosphatase SHP-2, cause Noonan syndrome.Nat Genet. 2001; 29:465–468. doi: 10.1038/ng772CrossrefMedlineGoogle Scholar
  • 13. Johnston JJ, van der Smagt JJ, Rosenfeld JA, Pagnamenta AT, Alswaid A, Baker EH, Blair E, Borck G, Brinkmann J, Craigen W, et al; Members of the Undiagnosed Diseases Network. Autosomal recessive Noonan syndrome associated with biallelic LZTR1 variants.Genet Med. 2018; 20:1175–1185. doi: 10.1038/gim.2017.249CrossrefMedlineGoogle Scholar
  • 14. Nakaguma M, Jorge AAL, Arnhold IJP. Noonan syndrome associated with growth hormone deficiency with biallelic LZTR1 variants.Genet Med. 2019; 21:260. doi: 10.1038/s41436-018-0041-5CrossrefMedlineGoogle Scholar
  • 15. Aoki Y, Niihori T, Inoue S, Matsubara Y. Recent advances in RASopathies.J Hum Genet. 2016; 61:33–39. doi: 10.1038/jhg.2015.114CrossrefMedlineGoogle Scholar
  • 16. Frattini V, Trifonov V, Chan JM, Castano A, Lia M, Abate F, Keir ST, Ji AX, Zoppoli P, Niola F, et al. The integrated landscape of driver genomic alterations in glioblastoma.Nat Genet. 2013; 45:1141–1149. doi: 10.1038/ng.2734CrossrefMedlineGoogle Scholar
  • 17. Cancer Genome Atlas Research Network. Electronic address: ; Cancer Genome Atlas Research Network. Comprehensive and integrative genomic characterization of hepatocellular carcinoma.Cell. 2017; 169:1327.e23–1341.e23. doi:10.1016/j.cell.2017.05.046Google Scholar
  • 18. Paganini I, Chang VY, Capone GL, Vitte J, Benelli M, Barbetti L, Sestini R, Trevisson E, Hulsebos TJ, Giovannini M, et al. Expanding the mutational spectrum of LZTR1 in schwannomatosis.Eur J Hum Genet. 2015; 23:963–968. doi: 10.1038/ejhg.2014.220CrossrefMedlineGoogle Scholar
  • 19. Ruggieri M, Praticò AD, Evans DG. Diagnosis, management, and new therapeutic options in childhood neurofibromatosis type 2 and related forms.Semin Pediatr Neurol. 2015; 22:240–258. doi: 10.1016/j.spen.2015.10.008CrossrefMedlineGoogle Scholar
  • 20. Steklov M, Pandolfi S, Baietti MF, Batiuk A, Carai P, Najm P, Zhang M, Jang H, Renzi F, Cai Y, et al. Mutations in LZTR1 drive human disease by dysregulating RAS ubiquitination.Science. 2018; 362:1177–1182. doi: 10.1126/science.aap7607CrossrefMedlineGoogle Scholar
  • 21. Claxton S, Kostourou V, Jadeja S, Chambon P, Hodivala-Dilke K, Fruttiger M. Efficient, inducible Cre-recombinase activation in vascular endothelium.Genesis. 2008; 46:74–80. doi: 10.1002/dvg.20367CrossrefMedlineGoogle Scholar
  • 22. Hellström M, Gerhardt H, Kalén M, Li X, Eriksson U, Wolburg H, Betsholtz C. Lack of pericytes leads to endothelial hyperplasia and abnormal vascular morphogenesis.J Cell Biol. 2001; 152:543–553. doi:10.1083/jcb.153.3.543CrossrefGoogle Scholar
  • 23. Sewduth RN, Jaspard-Vinassa B, Peghaire C, Guillabert A, Franzl N, Larrieu-Lahargue F, Moreau C, Fruttiger M, Dufourcq P, Couffinhal T, et al. The ubiquitin ligase PDZRN3 is required for vascular morphogenesis through Wnt/planar cell polarity signalling.Nat Commun. 2014; 5:4832. doi: 10.1038/ncomms5832CrossrefMedlineGoogle Scholar
  • 24. Sewduth RN, Kovacic H, Jaspard-Vinassa B, Jecko V, Wavasseur T, Fritsch N, Pernot M, Jeaningros S, Roux E, Dufourcq P, et al. PDZRN3 destabilizes endothelial cell-cell junctions through a PKCζ-containing polarity complex to increase vascular permeability.Sci Signal. 2017; 10:eaag3209. doi:10.1126/scisignal.aag3209CrossrefMedlineGoogle Scholar
  • 25. Lauriol J, Cabrera JR, Roy A, Keith K, Hough SM, Damilano F, Wang B, Segarra GC, Flessa ME, Miller LE, et al. Developmental SHP2 dysfunction underlies cardiac hypertrophy in Noonan syndrome with multiple lentigines.J Clin Invest. 2016; 126:2989–3005. doi: 10.1172/JCI80396CrossrefMedlineGoogle Scholar
  • 26. Artoni A, Selicorni A, Passamonti SM, Lecchi A, Bucciarelli P, Cerutti M, Cianci P, Gianniello F, Martinelli I. Hemostatic abnormalities in Noonan syndrome.Pediatrics. 2014; 133:e1299–e1304. doi: 10.1542/peds.2013-3251CrossrefMedlineGoogle Scholar
  • 27. Baker M, Robinson SD, Lechertier T, Barber PR, Tavora B, D'Amico G, Jones DT, Vojnovic B, Hodivala-Dilke K. Use of the mouse aortic ring assay to study angiogenesis.Nat Protoc. 2012; 7:89–104. doi:10.1038/nprot.2011.435CrossrefGoogle Scholar
  • 28. Futakuchi M. Animal model of lung metastasis of hepatocellular carcinoma: a tool for the development of anti-metastatic therapeutics.J Cancer Ther. 2013; 4:420–425. doi:10.4236/jct.2013.42A051CrossrefGoogle Scholar
  • 29. Newton TM, Reid E. An automated image analysis system to quantify endosomal tubulation.PLoS One. 2016; 11:e0168294. doi: 10.1371/journal.pone.0168294CrossrefMedlineGoogle Scholar
  • 30. Bryant DM, Kerr MC, Hammond LA, Joseph SR, Mostov KE, Teasdale RD, Stow JL. EGF induces macropinocytosis and SNX1-modulated recycling of E-cadherin.J Cell Sci. 2007; 120(pt 10):1818–1828. doi: 10.1242/jcs.000653CrossrefMedlineGoogle Scholar
  • 31. Preta G, Cronin JG, Sheldon IM. Dynasore - not just a dynamin inhibitor.Cell Commun Signal. 2015; 13:24. doi: 10.1186/s12964-015-0102-1CrossrefMedlineGoogle Scholar
  • 32. Basagiannis D, Zografou S, Galanopoulou K, Christoforidis S. Dynasore impairs VEGFR2 signalling in an endocytosis-independent manner.Sci Rep. 2017; 7:45035. doi: 10.1038/srep45035CrossrefMedlineGoogle Scholar
  • 33. Stahlschmidt W, Robertson MJ, Robinson PJ, McCluskey A, Haucke V. Clathrin terminal domain-ligand interactions regulate sorting of mannose 6-phosphate receptors mediated by AP-1 and GGA adaptors.J Biol Chem. 2014; 289:4906–4918. doi: 10.1074/jbc.M113.535211CrossrefMedlineGoogle Scholar
  • 34. Robertson MJ, Deane FM, Stahlschmidt W, von Kleist L, Haucke V, Robinson PJ, McCluskey A. Synthesis of the Pitstop family of clathrin inhibitors.Nat Protoc. 2014; 9:1592–1606. doi: 10.1038/nprot.2014.106CrossrefMedlineGoogle Scholar
  • 35. Crespo-Yàñez X, Aguilar-Gurrieri C, Jacomin AC, Journet A, Mortier M, Taillebourg E, Soleilhac E, Weissenhorn W, Fauvarque MO. CHMP1B is a target of USP8/UBPY regulated by ubiquitin during endocytosis.PLoS Genet. 2018; 14:e1007456. doi: 10.1371/journal.pgen.1007456CrossrefMedlineGoogle Scholar
  • 36. Ghedira N, Kraoua L, Lagarde A, et al. Further evidence for the implication of LZTR1, a gene not associated with the ras-mapk pathway, in the pathogenesis of noonan syndrome.Biol Med. 2017; 9:4–7. doi:10.4172/0974-8369.1000414CrossrefGoogle Scholar
  • 37. Umeki I, Niihori T, Abe T, Kanno SI, Okamoto N, Mizuno S, Kurosawa K, Nagasaki K, Yoshida M, Ohashi H, et al. Delineation of LZTR1 mutation-positive patients with Noonan syndrome and identification of LZTR1 binding to RAF1-PPP1CB complexes.Hum Genet. 2019; 138:21–35. doi: 10.1007/s00439-018-1951-7CrossrefMedlineGoogle Scholar
  • 38. Pagnamenta AT, Kaisaki PJ, Bennett F, Burkitt-Wright E, Martin HC, Ferla MP, Taylor JM, Gompertz L, Lahiri N, Tatton-Brown K, et al; DDD Study. Delineation of dominant and recessive forms of LZTR1-associated Noonan syndrome.Clin Genet. 2019; 95:693–703. doi: 10.1111/cge.13533CrossrefMedlineGoogle Scholar
  • 39. Jacquinet A, Bonnard A, Capri Y, Martin D, Sadzot B, Bianchi E, Servais L, Sacré JP, Cavé H, Verloes A. Oligo-astrocytoma in LZTR1-related Noonan syndrome.Eur J Med Genet. 2020; 63:103617. doi: 10.1016/j.ejmg.2019.01.007CrossrefMedlineGoogle Scholar
  • 40. Allison R, Lumb JH, Fassier C, Connell JW, Ten Martin D, Seaman MN, Hazan J, Reid E. An ESCRT-spastin interaction promotes fission of recycling tubules from the endosome.J Cell Biol. 2013; 202:527–543. doi: 10.1083/jcb.201211045CrossrefMedlineGoogle Scholar
  • 41. McCullough J, Clippinger AK, Talledge N, Skowyra ML, Saunders MG, Naismith TV, Colf LA, Afonine P, Arthur C, Sundquist WI, et al. Structure and membrane remodeling activity of ESCRT-III helical polymers.Science. 2015; 350:1548–1551. doi: 10.1126/science.aad8305CrossrefMedlineGoogle Scholar
  • 42. Bajorek M, Schubert HL, McCullough J, Langelier C, Eckert DM, Stubblefield WM, Uter NT, Myszka DG, Hill CP, Sundquist WI. Structural basis for ESCRT-III protein autoinhibition.Nat Struct Mol Biol. 2009; 16:754–762. doi: 10.1038/nsmb.1621CrossrefMedlineGoogle Scholar
  • 43. Park MS, Ravi V, Araujo DM. Inhibiting the VEGF-VEGFR pathway in angiosarcoma, epithelioid hemangioendothelioma, and hemangiopericytoma/solitary fibrous tumor.Curr Opin Oncol. 2010; 22:351–355. doi: 10.1097/CCO.0b013e32833aaad4CrossrefMedlineGoogle Scholar
  • 44. Kodack DP, Chung E, Yamashita H, Incio J, Duyverman AM, Song Y, Farrar CT, Huang Y, Ager E, Kamoun W, et al. Combined targeting of HER2 and VEGFR2 for effective treatmentof HER2-amplified breast cancer brain metastases.Proc Natl Acad Sci USA. 2012; 109: E3119–E3127. doi:10.1073/pnas.1216078109CrossrefMedlineGoogle Scholar
  • 45. Pandey AK, Singhi EK, Arroyo JP, Ikizler TA, Gould ER, Brown J, Beckman JA, Harrison DG, Moslehi J. Mechanisms of VEGF (Vascular Endothelial Growth Factor) inhibitor-associated hypertension and vascular disease.Hypertension. 2018; 71:e1–e8. doi: 10.1161/HYPERTENSIONAHA.117.10271LinkGoogle Scholar
  • 46. Basagiannis D, Christoforidis S. Constitutive endocytosis of VEGFR2 protects the receptor against shedding.J Biol Chem. 2016; 291:16892–16903. doi: 10.1074/jbc.M116.730309CrossrefMedlineGoogle Scholar
  • 47. Sawamiphak S, Seidel S, Essmann CL, Wilkinson GA, Pitulescu ME, Acker T, Acker-Palmer A. Ephrin-B2 regulates VEGFR2 function in developmental and tumour angiogenesis.Nature. 2010; 465:487–491. doi: 10.1038/nature08995CrossrefMedlineGoogle Scholar
  • 48. Genet G, Boyé K, Mathivet T, Ola R, Zhang F, Dubrac A, Li J, Genet N, Henrique Geraldo L, Benedetti L, et al. Endophilin-A2 dependent VEGFR2 endocytosis promotes sprouting angiogenesis.Nat Commun. 2019; 10:2350. doi: 10.1038/s41467-019-10359-xCrossrefMedlineGoogle Scholar
  • 49. Medinger M, Esser N, Zirrgiebel U, Ryan A, Jürgensmeier JM, Drevs J. Antitumor and antiangiogenic activity of cediranib in a preclinical model of renal cell carcinoma.Anticancer Res. 2009; 29:5065–5076.MedlineGoogle Scholar
  • 50. Brave SR, Ratcliffe K, Wilson Z, James NH, Ashton S, Wainwright A, Kendrew J, Dudley P, Broadbent N, Sproat G, et al. Assessing the activity of cediranib, a VEGFR-2/3 tyrosine kinase inhibitor, against VEGFR-1 and members of the structurally related PDGFR family.Mol Cancer Ther. 2011; 10:861–873. doi: 10.1158/1535-7163.MCT-10-0976CrossrefMedlineGoogle Scholar
  • 51. Araki T, Mohi MG, Ismat FA, Bronson RT, Williams IR, Kutok JL, Yang W, Pao LI, Gilliland DG, Epstein JA, et al. Mouse model of Noonan syndrome reveals cell type- and gene dosage-dependent effects of Ptpn11 mutation.Nat Med. 2004; 10:849–857. doi: 10.1038/nm1084CrossrefMedlineGoogle Scholar
  • 52. Uhlén P, Burch PM, Zito CI, Estrada M, Ehrlich BE, Bennett AM. Gain-of-function/Noonan syndrome SHP-2/Ptpn11 mutants enhance calcium oscillations and impair NFAT signaling.Proc Natl Acad Sci USA. 2006; 103:2160–2165. doi: 10.1073/pnas.0510876103CrossrefMedlineGoogle Scholar
  • 53. Kroisel PM, Häusler M, Klaritsch P, Karpf E, Zebedin D, Tiran B, Pertl B, Wagner K. Targeted enrichment sequencing in two midterm pregnancies with severe abnormalities on ultrasound.Lancet. 2017; 389:1857–1858. doi: 10.1016/S0140-6736(17)31049-8CrossrefMedlineGoogle Scholar
  • 54. Zhang J, Huang J, Qi T, Huang Y, Lu Y, Zhan T, Gong H, Zhu Z, Shi Y, Zhou J, et al. SHP2 protects endothelial cell barrier through suppressing VE-cadherin internalization regulated by MET-ARF1.FASEB J. 2019; 33:1124–1137. doi: 10.1096/fj.201800284RCrossrefMedlineGoogle Scholar
  • 55. Derivery E, Sousa C, Gautier JJ, Lombard B, Loew D, Gautreau A. The Arp2/3 activator WASH controls the fission of endosomes through a large multiprotein complex.Dev Cell. 2009; 17:712–723. doi: 10.1016/j.devcel.2009.09.010CrossrefMedlineGoogle Scholar
  • 56. Sadowski L, Pilecka I, Miaczynska M. Signaling from endosomes: location makes a difference.Exp Cell Res. 2009; 315:1601–1609. doi: 10.1016/j.yexcr.2008.09.021CrossrefMedlineGoogle Scholar
  • 57. Tian Y, Gawlak G, O’Donnell JJ, Birukova AA, Birukov KG. Activation of Vascular Endothelial Growth Factor (VEGF) receptor 2 mediates endothelial permeability caused by cyclic stretch.J Biol Chem. 2016; 291:10032–10045. doi:10.1074/jbc.M115.690487CrossrefMedlineGoogle Scholar
  • 58. Miquerol L, Langille BL, Nagy A. Embryonic development is disrupted by modest increases in vascular endothelial growth factor gene expression.Development. 2000; 127:3941–3946.CrossrefMedlineGoogle Scholar
  • 59. Le Cras TD, Spitzmiller RE, Albertine KH, Greenberg JM, Whitsett JA, Akeson AL. VEGF causes pulmonary hemorrhage, hemosiderosis, and air space enlargement in neonatal mice.Am J Physiol Lung Cell Mol Physiol. 2004; 287:L134–L142. doi: 10.1152/ajplung.00050.2004CrossrefMedlineGoogle Scholar
  • 60. Smith GA, Fearnley GW, Tomlinson DC, Harrison MA, Ponnambalam S. The cellular response to vascular endothelial growth factors requires co-ordinated signal transduction, trafficking and proteolysis.Biosci Rep. 2015; 35:e00253. doi:10.1042/BSR20150171CrossrefMedlineGoogle Scholar
  • 61. Peterson TE, Kirkpatrick ND, Huang Y, Farrar CT, Marijt KA, Kloepper J, Datta M, Amoozgar Z, Seano G, Jung K, et al. Dual inhibition of Ang-2 and VEGF receptors normalizes tumor vasculature and prolongs survival in glioblastoma by altering macrophages.Proc Natl Acad Sci USA. 2016; 113:4470–4475. doi: 10.1073/pnas.1525349113CrossrefMedlineGoogle Scholar
  • 62. Kang S, Park KC, Yang KJ, Choi HS, Kim SH, Roh YJ. Effect of cediranib, an inhibitor of vascular endothelial growth factor receptor tyrosine kinase, in a mouse model of choroidal neovascularization.Clin Exp Ophthalmol. 2013; 41:63–72. doi: 10.1111/j.1442-9071.2012.02813.xCrossrefMedlineGoogle Scholar
  • 63. Bauditz J, Lochs H. Angiogenesis and vascular malformations: antiangiogenic drugs for treatment of gastrointestinal bleeding.World J Gastroenterol. 2007; 13:5979–5984. doi: 10.3748/wjg.v13.45.5979CrossrefMedlineGoogle Scholar
  • 64. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA-seq data with or without a reference genome.Bioinforma Impact Accurate Quantif Proteomic Genet Anal Res. 2014:41–74. doi: 10.1201/b16589CrossrefGoogle Scholar

Novelty and Significance

What Is Known?

  • The Noonan syndrome (NS) gene LZTR1 (leucine zipper-like transcription regulator 1) determines substrate specificity of the CUL3 (CULLIN3) ubiquitin ligase complexes.

  • Cardiac injury and bleeding disorders have been recently described in patients with NS harboring LZTR1 mutations.

  • Currently, there are no therapeutic options for the management of bleeding complications in patients with NS.

What New Information Does This Article Contribute?

  • Loss of Lztr1 in mice leads to endothelial dysfunction that phenotypically overlaps with bleeding pathologies observed in patients with NS.

  • LZTR1 affects the dynamics of vesicle trafficking by controlling ubiquitination of the ESCRT-III (endosomal sorting complex required for transport III) component CHMP1B (charged multivesicular protein 1B).

  • Lztr1-mediated dysregulation of CHMP1B ubiquitination triggers endosomal accumulation and subsequent activation of VEGFR2 (vascular endothelial growth factor receptor 2), causing disruption of endothelial junctions and bleeding.

Our findings provide a fundamental explanation underlying the molecular mechanism of endothelial dysfunction in patients with NS harboring LZTR1 mutations. By combining comprehensive analysis of Lztr1 knockout models and unbiased proteomics and transcriptomics approaches, we demonstrated that loss of Lztr1 function leads to vascular dysfunction by dysregulating vesicular trafficking and subsequent activation of VEGFR2. Our results indicate that ELISA screening of patients with NS for the levels of soluble VEGFR2 in blood plasma may predict both the severity of NS phenotypes and potential responders to anti-VEGF therapy, whereas anti-VEGF therapies such as cediranib could be a potential treatment approach for managing bleeding disorders in LZTR1-mutated patients with NS.

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.