AKT/FOXO1 axis links cross-talking of endothelial cell and pericyte in TIE2-mutated venous malformations

Abstract

Background: Venous malformations (VMs), most of which associated with activating mutations in the endothelial cells (ECs) tyrosine kinase receptor TIE2, are characterized by dilated and immature veins with scarce smooth muscle cells (SMCs) coverage. However, the underlying mechanism of interaction between ECs and SMCs responsible for VMs has not been fully understood.

Methods: Here, we screened 5 patients with TIE2-L914F mutation who were diagnosed with VMs by SNP sequencing, and we compared the expression of platelet-derived growth factor beta (PDGFB) and α-SMA in TIE2 mutant veins and normal veins by immunohistochemistry. In vitro, we generated TIE2-L914F-expressing human umbilical vein endothelial cells (HUVECs) and performed BrdU, CCK-8, transwell and tube formation experiments on none-transfected and transfected ECs. Then we investigated the effects of rapamycin (RAPA) on cellular characteristics. Next we established a co-culture system and investigated the role of AKT/FOXO1/PDGFB in regulating cross-talking of mutant ECs and SMCs.

Results: VMs with TIE2-L914F mutation showed lower expression of PDGFB and α-SMA than normal veins. TIE2 mutant ECs revealed enhanced cell viability and motility, and decreased tube formation, whereas these phenotypes could be reversed by rapamycin. Mechanically, RAPA ameliorated the physiological function of mutant ECs by inhibiting AKT-mTOR pathway, but also facilitated the nuclear location of FOXO1 and the expression of PDGFB in mutant ECs, and then improved paracrine interactions between ECs and SMCs. Moreover, TIE2 mutant ECs strongly accelerated the transition of SMCs from contractile phenotype to synthetic phenotype, whereas RAPA could prevent the phenotype transition of SMCs.

Conclusions: Our data demonstrate a previously unknown mechanistic linkage of AKT-mTOR/FOXO1 pathway between mutant ECs and SMCs in modulating venous dysmorphogenesis, and AKT/FOXO1 axis might be a potential therapeutic target for the recovery of TIE2-mutation causing VMs. Video Abstract.

Keywords: AKT; Endothelial cells; FOXO1; Smooth muscle cells; TIE2-L914F; Venous malformation.

Fig. 1

Reduced PDGFB and α-SMA expression in patients with TIE2-L914F mutant venous malformation. a Representative DNA sequences of VM patients (below) and healthy controls (up) detected by SNP site sequencing. b Mutation rate of TIE2-L914F in VM patients and healthy controls detected by pyrosequencing. c Representative immunohistochemical images of arteries (A), normal veins (V), VM veins (VM). Immunohistochemistry staining images showing expression levels of PDGFB (up) and α-SAM (below) in A, V and VM. On the far right are the magnifications of the VM images. d α-SAM positive cells per μm of lumen length and PDGFB positive area per pixels of lumen length in the V and VM groups were calculated. Scale bar, 50 μm. The data are means ± SEM (n = 6). **P < 0.01

Fig. 2

TIE2-L914F mutation activates AKT/FOXO1 signaling. a ECs were transfected with lentivirus expressing GFP, TIE2-WT, TIE2-L914F respectively. Red arrows: typical elongated cell morphology of ECs. b ECs expressing lentivirus vectors under fluorescence microscopy. c Relative mRNA level of TIE-2. d, e 48 h after transfecting, the expression levels of AKT, p-AKT (ser473), FOXO1 and p-FOXO1 in ECs were then measured by Western blot. Flag tags were used to verify effectiveness of transfection. Scale bar, 200 μm. The data are means ± SEM (n = 3). ns: no significant, **P < 0.01

Fig. 3

Effects of TIE2-L914F mutation on the biological characteristics of ECs. a, b 48 h after transfecting, ECs underwent BrdU assay. The ratio of positive cells to total cells in each visual field was analyzed by ImageJ software. c Different treated ECs underwent CCK-8 experiment, the OD values of 450 nm at day 1, 2, 3, 4 were tested. **P < 0.01 TIE-L914F versus NC, GFP and TIE2-WT. d, e Transwell assay was used to study the migration ability of NC-ECs, GFP-ECs, TIE2-WT-ECs and TIE2-L914F-ECs over a period of 24 h. f Tube formation assay of different groups of ECs. (up: inverted microscope; below: fluorescence microscope). g Branch points (left) and total length (right) of each group were measured. Scale bars, 100 μm (a), 200 μm (d), 500 μm (f). The data are means ± SEM (n = 3). ns: no significant, **P < 0.01

Fig. 4

AKT/FOXO1 pathway inhibition amends the abnormal characteristics of TIE2-L914F mutant ECs. a, b BrdU assay was performed to study the proliferation of NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs treated with DMSO or rapamycin. c CCK-8 experiment was performed to study the effects of rapamycin on the cell activity of NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs. **P < 0.01 TIE-L914F versus TIE2-L914F + RAPA d, e Transwell assay was used to study the effects of rapamycin on the migration ability of NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs over a period of 24 h. f, g Apoptotic assay showed the apoptosis of NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs, and the effect of rapamycin on cellular apoptosis. h, i Tube formation assay was carried on NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs treated with DMSO of rapamycin. The total length and the number of branch points were used as indicators. j, k Expression of autophagy related indexes (p62, LC3I, LC3II) in different treated ECs was detected by western blot. Scale bars, 100 μm (a, d), 500 μm (h). The data are means ± SEM (n = 3). ns: no significant, *P < 0.05, **P < 0.01

Fig. 5

RAPA promotes nuclear location of FOXO1 and PDGFB expression. a, b Western blot analysis of AKT, p-AKT, FOXO1, p-FOXO1, mTOR and p-mTOR on different groups of ECs treated with DMSO or rapamycin. c, d Expression of FOXO1 in the nucleus was detected by western blot. e NC-ECs, TIE2-WT-ECs and TIE2-L914F-ECs were subjected to IF staining with anti-FOXO1 antibody after the RAPA (15 nM) treatments for 48 h. f Average fluorescence intensity of each field was statistically analyzed by ImageJ software. g Expression levels of PDGFB were measured using real-time quantitative PCR. h Production of PDGFB released from the ECs was determined using ELISA. Scale bar, 50 μm. The data are means ± SEM (n = 3). ns: no significant, *P < 0.05, **P < 0.01

Fig. 6

RAPA promotes the interaction of ECs and SMCs. a, b Transwell assay was used to study the migration ability of SMCs after co-culturing with NC, TIE2-WT and TIE2-L914F mutant ECs treated with DMSO or rapamycin over a period of 24 h. c, d Different groups of ECs were treated with DMSO or rapamycin (RAPA), and then co-cultured with SMCs. The expression of OPN and α-SMA were measured in SMCs by western blot. Scale bar, 100 μm. The data are means ± SEM (n = 3). ns: no significant, *P < 0.05, **P < 0.01