Introduction

Tumor growth, similar to that of normal organs, depends greatly on the outgrowth of blood vessels to form neovasculatures that supply raw materials for metabolism. Neovascular structures originate from existing capillaries via angiogenesis1, which is a robust process that involves the establishment of a dynamic balance of pro- and anti-angiogenic factors suspended in the tissue niche. VEGF-A is a critical proangiogenic factor that renders the development of anti-VEGF therapies a promising avenue of research, particularly in the treatment of cancers2. Inhibitors targeting the VEGF pathway have exhibited significant anti-tumor activities in multiple mouse models; however, anti-VEGF therapies have not yet been successfully clinically applied to most cancers in humans3,4. In the absence of VEGF, other proangiogenic factors, including FGF, and ephrin, regulate angiogenesis in anti-VEGF-treated tumors and it has been found that anti-VEGF therapy induces unexpected vessel normalization and extracellular matrix remodeling, with improved vessel perfusion5,6. Vascular abnormalities in cancers include insufficient pericyte coverage, hyperpermeability, and immunosurveillance escape7,8. The challenges faced by novel anti-angiogenesis therapy in cancers have led to rigorous investigations, particularly in the elucidation of the underlying mechanisms.

The protein tyrosine phosphatase SHP2, encoded by PTPN11, is expressed ubiquitously, including in blood vessel cells. As an important participant in growth factor and cytokine signaling, SHP2 is frequently upregulated or mutated in tumors, and its oncogenic behavior has been attributed to abnormal SHP2 expression9,10,11. Structurally, there are two SH2 domains, a protein tyrosine phosphatase (PTP) domain, and a C-terminal tail containing phosphorylatable tyrosine residues. The SH2 domains interact with the PTP domain to form a self-inhibitory intramolecular interaction, and have recently been used to develop allosteric inhibitors12,13. New SHP2 inhibitors have been demonstrated to possess tremendous anti-tumor activity, particularly in combination with other conventional drugs14,15,16,17,18,19,20. Owing to the ubiquitous expression of druggable SHP2, its function in tumor microenvironments has been studied thoroughly. Our group has demonstrated that SHP2 suppressed CXCL9 production in macrophages, which prevented T cell infiltration and promoted tumor growth in miceCells

HUVECs were isolated from umbilical cord vein by lavaging with 0.2% (w/v) collagenase solution at 37 °C for 15 min, the cells were collected and suspended in complete M199 medium. HUVECs between passages 4 and 6 were cultured in a complete medium containing 53% M199 (Corning Incorporated), 37% human endothelial serum-free medium (Thermo Fisher Scientific), and 15 μg/mL of endothelial cell growth supplement (Sigma-Aldrich). Human cerebral microvessel endothelial cells (hCMECs) were purchased from Zhejiang Meisen Cell Technology Co., Ltd. Primary mouse lung endothelial cells (MLECs) were isolated from 6- to 8-week-old mice. Briefly, 1 month after tamoxifen administration, mice were anesthetized with 80 mg/kg ketamine and 12 mg/kg xylazine, and then subjected to perfusion by intracardiac injection of PBS (0.1% BSA) for blood removal. Lung tissue was then dissected and cut into 1–2-mm pieces, and digested with a solution containing type-I collagenase (2 mg/mL, Biosharp), dispase (1 mg/mL, Roche), and DNase (10 μg/mL, Roche) in DPBS for 45 min at 37 °C in a rotatory shaker (at 80 rpm). After digestion, the enzymes were neutralized by DMEM plus 20% FBS. A single-cell suspension was prepared using a 40-μm cell strainer. Finally, the cells were further purified using anti-mouse CD31-conjugated magnetic beads (Invitrogen) and maintained in a complete medium containing DMEM with 10% FBS, endothelial cell growth supplement (15 μg/mL, Sigma-Aldrich), 1% nonessential amino acids (Gibco), and heparin (0.1 mg/mL, Solarbio).

Mouse Lewis lung cancer (LLC), B16 melanoma cells, and E0771 purchased from ATCC were cultured in DMEM. Non-small cell lung cancer (NSCLC) (H460, A549, H1299) and hepatocellular carcinoma LM3 cells purchased from ATCC were grown in RPMI-1640 (Thermo Fisher Scientific). The media were supplemented with 10% FBS, 100 IU/mL penicillin, and 100 μg/mL streptomycin, and the cells were incubated at 37 °C under 5% CO2.

For the double luciferase reporter assay, the promoter fragment containing multiple predicted c-Jun binding sites (nucleotides -2000 to 100 of human SOX7 gene loci) was cloned into a pGL3-basic vector. In 12-well plates, HEK293 cells were transfected with vector or c-Jun for 24 h. The double luciferase reporter assay was conducted following the manual for the Dual-Luciferase reporter gene assay kit (Promega). The reporter gene activity was detected by GloMax® 20/20 Luminometer (Promega) and normalized to the activity of Renilla luciferase.

To construct SHP2WT and inhibitor-resistant mutant SHP2 (SHP2T253M/Q257L) LM3 cell lines, SHP2WT and SHP2T253M/Q257L cDNA were constructed into the PLVX-NEO vector and packaged into lentiviruses. LM3 cells were infected with shSHP2-3′utr lentivirus and screened by puromycin. The SHP2 knockdown LM3 cells were then infected with SHP2WT or SHP2T253M/Q257L lentivirus, and neomycin was further used to screen stable transgenic strains.

Mouse tumor models

To establish syngeneic mouse tumor models, suspensions of LLC (4 × 105 cells in 100 μL) or B16 (8 × 105 cells in 100 μL) cells were subcutaneously implanted in the dorsal flank of 8- to 10-week-old male mice. For the orthotopic mouse breast tumor model, E0771 cells (8 × 105) were suspended in 100 μL of Matrigel (50% v/v; Corning) in RPMI1640 medium and injected into the mammary fat pads of 8-week-old female mice. The tumor volumes were measured using a digital caliper every 2 days and calculated according to the following formula: V = 0.52 × L × W2 (V, tumor volume; L, longest diameter of the tumor; W, perpendicular diameter of L)69,70. Sixteen or eighteen days after cell injection, tumors were harvested and fixed with 4% paraformaldehyde (PFA) for further histological analysis.

Matrigel plug assay

Matrigel (500 μL), supplemented with the recombinant mouse FGF2 and VEGF-A (400 ng/mL each) and heparin (50 units/mL) was injected subcutaneously into the flanks of 8- to 10-week-old mice, and they were sacrificed 7 days after injection. The Matrigel plugs were removed, photographed, and fixed in 4% PFA for further analysis.

Mouse aortic ring assay

The thoracic aortas were removed from the mice under anesthesia and cut into 1-mm rings. The aortic rings were then placed between two layers of 100 μL growth factor-reduced Matrigel (Corning) supplemented with heparin (20 U/mL, Solarbio), VEGF-A (100 ng/mL, Novoprotein), and FGF2 (100 ng/mL, Novoprotein), and incubated in culture medium (10% FBS, 100 IU/mL penicillin and 100 μg/mL streptomycin). The culture medium was changed every other day. The sprouting area was recorded under a phase-contrast microscope on day 7 and measured using Image J 1.49 v.

Co-immunoprecipitation, immunoblotting, and immunofluorescence staining

For western blotting, the total protein concentrations (20 μg) from MLECs or HUVECs were separated using SDS-PAGE gels and transferred onto nitrocellulose membranes (Pall, Port Washington). The primary antibodies used are shown in Supplemental Table 1.

For co-immunoprecipitation, cells were harvested and lysed in a cell lysis buffer used for western blotting and IP (Beyotime), protease inhibitor cocktail (Roche), and PhosSTOP (Roche) on ice. Protein G Dynabeads were pretreated with the primary antibody for 10 min at 25 °C and then washed with PBST three times. The cell lysates were then incubated overnight with antibody-conjugated beads at 4 °C. The immunoprecipitates were then used for western blotting by odyssey (version 3.0.29) or LI-COR image studio (version 5.2).

For tissue immunofluorescence staining, samples were fixed in 4% PFA, dehydrated in a 30% sucrose solution for 24 h, and embedded using the Tissue-Tek OCT compound. Frozen blocks were cut into 10-μm-thick sections. For cell immunofluorescence staining, endothelial cells were fixed with 4% PFA and permeabilized with 0.5% Triton for 15 min. The samples were blocked with 5% goat or donkey serum in PBST and incubated overnight at 4 °C with primary antibodies (Supplemental Table 1). After performing multiple washes, the samples were incubated for 1 h at room temperature with the secondary antibodies (Supplemental Table 1). The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Beyotime). The samples were then mounted with fluoromount-G (Southern Biotech), and immunofluorescence images were acquired using a confocal microscope (FV3000, Olympus). The images were further processed using ImageJ 1.49v or OlyVIA VS200.

Quantitative PCR

The total RNA was isolated from HUVECs and MLECs by TRIzol and reverse-transcribed to cDNA using the ReverTraAce qPCR RT kit (Toyobo Inc.). Quantitative RT-PCR was conducted using SYBR green dye by CFX96 Touch Real-Time PCR Detection System (Bio-Rad) and the primers listed in Supplemental Table 2 and Supplementary Table 3. Gene expression was calculated using the equation RQ = 2−△△Ct and normalized to GAPDH or 18S RNA values.

Pimonidazole staining, vascular leakage, and perfusion assay

To measure tumor hypoxia, the mice were injected with Hypoxyprobe-1 (60 mg/kg, Hypoxyprobe) i.p. 1 h before they were sacrificed. Tumors were then removed and embedded in Tissue-Tek OCT. FITC-conjugated mouse anti-pimonidazole monoclonal antibody was applied following the manufacturer’s protocol. To evaluate vascular permeability, 100 μL of FITC-conjugated dextran (25 mg/mL, 70 kDa, Sigma-Aldrich) was intravenously injected and the mice were perfused with 1% PFA to remove circulating dextran after 30 min. Tumors were then dissected and embedded in the Tissue-Tek OCT compound for further analysis. To measure vessel perfusion, DyLight 488-conjugated Tomato lectin (1 mg/mL, 100 μL, Vector Laboratories) was i.v. injected 30 min before sacrifice, and the tumors were harvested for further analysis.

EdU incorporation assay

Cells were seeded in the 24-well plates at a density of 4 × 104 cells per well. After overnight incubation, the cells were incubated with EdU (Beyotime) for 5 h at 37 °C and then fixed in 4% PFA. After washing with PBS, the cells were treated with 0.5% Triton X-100 for 15 min at room temperature for permeabilization. The cells were incubated and protected from light using a click additive solution and stained with DAPI. Images were obtained using a confocal microscope (FV3000, Olympus).

Transwell cell migration assay

HUVECs or MLECs suspended in the complete medium were seeded in the upper chamber. Four hours later, the medium in the upper chamber was changed to 0.1% FBS in M199, and the lower chamber was filled with 600 μL of complete medium. After incubation for 24 h at 37 °C, non-migrating cells were removed and cells that migrated through the membrane were fixed in 4% PFA and stained with crystal violet solution (Beyotime), after photographed, the crystal violet-stained cells were incubated with 33% acetic acid, and absorbance was detected at 570 nm (SynergyMx M5, Molecular Devices).

Endothelial cell tube formation

Cells were seeded in Matrigel-coated u-slides at a density of 1 × 104 cells per well (Ibidi, Germany) and incubated at 37 °C for 4 h and for a further 15 min with calcein (Yeasen). Images were acquired using a fluorescence microscope (IX70, Olympus). Five different fields for each condition were quantified by counting the number of junctions, total tube length, and total branching length using ImageJ 1.49v.

NSCLC tissues

NSCLC and matched adjacent normal tissue specimens were collected from the First Affiliated Hospital, College of Medicine, Zhejiang University (Hangzhou, China), and confirmed by pathological diagnosis. This study was approved by the Research Ethics Committee of the First Affiliated Hospital, College of Medicine, Zhejiang University and abided by the Declaration of Helsinki principles. Written informed consent was obtained from all patients prior to the study.

Quantification and statistical analysis

Statistical analysis was performed using Prism software (GraphPad Inc, version 6.0 and 8.0). All quantitative data are presented as the mean ± SEM. Unpaired Student’s t-tests were used for comparisons between two groups. Multiple group comparisons were conducted by one-way or two-way ANOVA followed by Tukey’s post hoc tests or multi-comparisons. p < 0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.