Hypoxia-induced AFAP1L1 regulates pathological neovascularization via the YAP-DLL4-NOTCH axis

Abstract

Background: Pathological neovascularization plays a pivotal role in the onset and progression of tumors and neovascular eye diseases. Despite notable advancements in the development of anti-angiogenic medications that target vascular endothelial growth factor (VEGF) and its receptors (VEGFRs), the occurrence of adverse reactions and drug resistance has somewhat impeded the widespread application of these drugs. Therefore, additional investigations are warranted to explore alternative therapeutic targets. In recent years, owing to the swift advancement of high-throughput sequencing technology, pan-cancer analysis and single-cell sequencing analysis have emerged as pivotal methodologies and focal areas within the domain of omics research, which is of great significance for us to find potential targets related to the regulation of pathological neovascularization.

Methods: Pan-cancer analysis and scRNA-seq data analysis were employed to forecast the association between Actin filament-associated protein 1 like 1 (AFAP1L1) and the development of tumors and endothelial cells. Tumor xenograft model and ocular pathological neovascularization model were constructed as well as Isolectin B4 (IsoB4) staining and immunofluorescence staining were used to assess the effects of AFAP1L1 on the progression of neoplasms and neovascular eye diseases in vivo. Transwell assay, wound scratch assay, tube forming assay, three-dimensional germination assay, and rhodamine-phalloidin staining were used to evaluate the impact of AFAP1L1 on human umbilical vein endothelial cells (HUVECs) function in vitro; Dual luciferase reporting, qRT-PCR and western blot were used to investigate the upstream and downstream mechanisms of pathological neovascularization mediated by AFAP1L1.

Results: Our investigation revealed that AFAP1L1 plays a crucial role in promoting the development of various tumors and demonstrates a strong correlation with endothelial cells. Targeted suppression of AFAP1L1 specifically in endothelial cells in vivo proves effective in inhibiting tumor formation and ocular pathological neovascularization. Mechanistically, AFAP1L1 functions as a hypoxia-related regulatory protein that can be activated by HIF-1α. In vitro experiments demonstrated that reducing AFAP1L1 levels can reverse hypoxia-induced excessive angiogenic capacity in HUVECs. The principal mechanism of angiogenesis inhibition entails the regulation of tip cell behavior through the YAP-DLL4-NOTCH axis.

Conclusion: In conclusion, AFAP1L1, a newly identified hypoxia-related regulatory protein, can be activated by HIF-1α. Inhibiting AFAP1L1 results in the inhibition of angiogenesis by suppressing the germination of endothelial tip cells through the YAP-DLL4-NOTCH axis. This presents a promising therapeutic target to halt the progression of tumors and neovascular eye disease.

Keywords: AFAP1L1; HIF-1α; Hypoxia; Ocular pathologic neovascularization; Tumor angiogenesis; Vascular tip cell.

Fig. 1

The pan-cancer landscape of AFAP1L1 expression. A The mRNA expression of AFAP1L1 in different human tissues from the HPA dataset. B The mRNA expression of AFAP1L1 in different human tissues from the GTEx dataset. C The mRNA expression of AFAP1L1 in various cancers and normal tissues from the TIMER2 database. Columns labelled with colors indicate statistically significant differences between tumor tissues and normal tissues (upregulation is indicated in red; downregulation is indicated in blue; *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001)

Fig. 2

Association of AFAP1L1 mRNA expression with prognosis of various tumors. A The TCGA survival data for AFAP1L1 from the TISCH2 database. B, C The survival curves for OS and RFS manifested that the association between the expression of AFAP1L1 and survival prognosis of pan-cancer patients from the GSCA database. D The mRNA expression of AFAP1L1 in pathological stages across human cancers, including COAD, READ and STAD

Fig. 3

Analysis of DNA methylation of AFAP1L1 in human cancers. A The DNA Methylation of AFAP1L1 in various cancers and normal tissues from the GSCA database. B Correlation between AFAP1L1 DNA methylation and mRNA expression levels in human cancers from the GSCA database. Several representative correlations were shown as scatter plots, such as C CESC, D PRAD, E TGCT, F SKCM, G KIRP, H LGG, I STAD and J BLCA. K Survival difference between high and low methylation in each cancer from the GSCA database. Several representative correlations were labelled with blue boxes and shown as the survival curves for OS and RFS, including L, P KIRP, M, Q ACC, O, S STAD and N, R BLCA

Fig. 4

AFAP1L1 is associated with vascular endothelial features in cancer and retina. A The heatmap represents the correlation of AFAP1L1 expression with immune infiltration level in diverse cancer types. B The heatmap represents the correlation between AFAP1L1 expression level and immune infiltration of endothelial cells and several known vascular endothelial cell markers in diverse cancer types using the TIMER2 database based on EPIC, MCPCOUNTER and XCELL algorithms. C–D Clustering is performed by the cell type (C), the patient types or the tumor types D and visualized using Tsne algorithm from the publicly available scRNA-seq data of lung tumor endothelial cell through the Seruat package. E Cross validation of the identified EC marker genes (PECAM1), tip cell marker genes (DLL4 and CXCR4) and AFAP1L1 from the publicly available scRNA-seq data of lung tumor endothelial cell through the Seruat package. F The grouped bar charts show the expression of AFAP1L1 in total and different types of lung cancer. G The stacked bar charts show the expression of AFAP1L1 in different lung cancer types and different cell types. H Colocalization of AFAP1L1 and tip cell marker (DLL4) (DLL4: blue; AFAP1L1: green; IsoB4: red) in the anterior end of the retinal vascular. Scale bar: 25 μm. (n = 4 independent experiments)

Fig. 5

AFAP1L1-related gene enrichment analysis. A STRING protein network map of top 100 AFAP1L1-related genes from the GEPIA2 database. B GO analysis based on the top 100 AFAP1L1-related genes. C According to the differentially expressed genes between the high and low expression groups of AFAP1L1, GSEA analysis revealed the top 25 up-regulated and the top 25 down-regulated enrichment biological processes. Enrichment biological processes ranked based on the normalized enrichment scores (NES) are shown in the heat map. D Average correlations between AFAP1L1 and 14 functional states in different cancers from the CancerSEA database and the bar charts show the number of datasets in which AFAP1L1 is significantly related to the corresponding state

Fig. 6

AFAP1L1 knockdown promotes tumor growth and angiogenesis in vivo. A–C Xenografts were established by injecting A549 cells subcutaneously on the dorsal flank of BALB/c nude mice, and intratumoral administration of NC-ECKD, AFAP1L1-ECKD #1 or AFAP1L1-ECKD #2 was performed once the tumor length reached 100 mm.³ (n = 4 mice for each group). Representative images were shown (A); quantification of tumor growth curves of xenograft in nude mice (Day 0 represents the time of the initial AAV injection) (B); quantification of tumor weights of xenograft in nude mice (C). D–E Xenografts at the end points were sliced into sections for vascular visualization (n = 4 mice for each group); the sections were labeled with IsoB4 (red) and the nuclei stained with DAPI (blue) (D); quantification of IsoB4 stained vessels in the sections of xenografts (E). P values were calculated by one-way ANOVA with Bonferroni’s post hoc test (B, C, E). Error bars represent the mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001

Fig. 7

AFAP1L1 knockdown promotes ocular neovascularization in vivo. A–D IsoB4 staining of whole-mount retinas from OIR mice at P17 intravitreously injected with NC-ECKD, AFAP1L1-ECKD #1 or AFAP1L1-ECKD #2 (Scale bar: 1 mm) (n = 4 mice for each group); the purple area indicates avascular area, and the white area indicates NVTs (A). Quantification of neovascular tuft area and avascular area in OIR mice (B). The red areas highlighted vascular area especially filopodia in the anterior retina of the nonperfusion area (C). Quantification of the number and length of filopodia (green asterisk) respectively. (Scale bar: 40 μm) (n = 4 mice for each group) (D). E–F IsoB4 staining of whole-mount retinas from L-CNV mice intravitreously injected with NC-ECKD, AFAP1L1-ECKD #1 or AFAP1L1-ECKD #2. (n = 4 mice for each group) (E). Quantification of size of CNV lesions in L-CNV mice (Scale bar: 200 μm) (F). G–H All the cornea samples of S-CNV mice subconjunctivally injected with NC-ECKD, AFAP1L1-ECKD #1 or AFAP1L1-ECKD #2 were collected for vascular visualization by slit lamp. (n = 4 mice for each group) (G). Quantification of size of CNV lesions in S-CNV mice H (Scale bar: 400 μm). P values were calculated by one-way ANOVA with Bonferroni's post hoc test B, D, F, H. Error bars represent the mean ± SD. *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001; n.s.: no significance

Fig. 8

HIF-1α directly suppresses AFAP1L1 transcription under hypoxia. A AFAP1L1mRNA levels in HUVEC cells exposed to hypoxia (1% O2) for the indicated times. (n = 4 independent experiments). B Volcano plots showed the correlation between AFAP1L1 and HIF-1α or HIF-2α. C Western blot analysis of AFAP1L1, HIF-1α, HIF-2α protein levels in HUVEC cells exposed to hypoxia (1% O2) for the indicated times. Densitometric quantitation of western blot band intensity shown in C. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). D Western blot analysis of AFAP1L1, HIF-1α, HIF-2α and β-actin protein levels in HUVEC cells left untreated or exposed to hypoxia (1% O2) for 24 h as well as transfected with siNC, siHIF-1α or siHIF-2α. Densitometric quantitation of western blot band intensity shown in D. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). E Schematic diagram depicting the human AFAP1L1 promoter with the presence of hypoxia response element (HRE) sites from the JASPAR database and constructed mutant HRE sites (left panel) (TSS: transcription start site; ATG: initiating methionine codon). Luciferase reporter assay for AFAP1L1 promoter activity in HEK293T cells transfected with control vector, HIF-1α overexpression or HIF-2α overexpression plasmids (n = 4 independent experiments) (middle panel). Luciferase reporter assay for AFAP1L1 promoter activity in HEK293T cells following transfection of wHRE or mHRE luciferase reporter plasmids under HIF-1α overexpression (right panel). (n = 4 independent experiments). P values were calculated by one-way ANOVA with Bonferroni's post hoc test (A, C, F). Error bars represent the mean ± SD. HUVECs are transfected with shNC, shAFAP1L1 #1 or shAFAP1L1 #2 and then exposed to hypoxia (1% O2) for 24 h or left untreated. F Transwell assays of HUVECs under hypoxia (1% O2) or simultaneously transfected with shRNA. Scale bar: 50 µm. G Wound scratching assays of HUVECs under hypoxia (1% O2) or simultaneously transfected with shRNA. Scale bar: 200 µm. [rhodamine-conjugated phalloidin: red (to visualize the actin cytoskeleton); DAPI: blue (to visualize nuclei)]. H Tube formation Assay of HUVECs under hypoxia (1% O2) or simultaneously transfected with shRNA. Scale bar: 200 µm. I Rhodamine-phalloidin staining reveals the actin cytoskeleton and filopodia of HUVECs under hypoxia (1% O2) or simultaneously transfected with shRNA (rhodamine-conjugated phalloidin: red; DAPI: blue). Scale bar: 25 µm. J Quantification of migrated cells. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 per group, data pooled from 4 independent experiments). K Quantification of wound area. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 per group, data pooled from 4 independent experiments). L Quantification of tube formation length. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). M, N Quantification of length and number of filopodia per cell. Results are presented as mean ± SEM, statistical analyses were performed using Kruskal–Wallis with Bonferroni’s post hoc test. (n = 4 independent experiments). *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001; n.s.: no significance

Fig. 9

AFAP1L1 regulates the angiogenic activity of ECs via the YAP-DLL4-NOTCH axis. A HUVECs were transfected with shNC, shAFAP1L1 #1, shYAP or simultaneously added with DAPT for 24 h. Three-dimensional (3D) Bead Sprouting Assay reveals the in vitro sprouting capabilities of HUVECs under different treatments. Scale bar: 100 µm. Quantification of length and number of sprouts per bead. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). B Western blot analysis of DLL4, NICD, HES1, HES1 and β-actin protein levels in HUVEC cells transfected with shNC, shAFAP1L1 #1, shYAP or simultaneously added with DAPT for 24 h. Densitometric quantitation of western blot band intensity shown in B. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). C HUVECs were transfected with Vector, OE AFAP1L1 #1, shYAP or added with VP for 24 h. Three-dimensional (3D) Bead Sprouting Assay reveals the in vitro sprouting capabilities of HUVECs under different treatments. Scale bar: 100 µm. Quantification of length and number of sprouts per bead. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). D Western blot analysis of p-YAP(Ser127), YAP, DLL4, NICD and β-actin protein levels in HUVEC cells transfected with Vector, OE AFAP1L1, shYAP or simultaneously added with VP for 24 h. Densitometric quantitation of western blot band intensity shown in D. Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). E HUVECs were transfected with Vector, OE AFAP1L1 or simultaneously transfected with shAFAP1L1 #1 or shAFAP1L1 #2. Total, cytoplasmic, nuclear extracts from the resulting cells are analyzed by western blot for YAP expression. Densitometric quantitation of western blot band intensity shown in E Results are presented as mean ± SEM, statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc test. (n = 4 independent experiments). F Localization of YAP is demonstrated by immunofluorescence. Scale bar, 25 μm. Quantification of the rate of Nuclear/Total YAP fluorescence. Results are presented as mean ± SEM, statistical analyses were performed using Kruskal–Wallis with Bonferroni's post hoc test. (n = 4 independent experiments). *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001; n.s.: no significance