Recent studies have demonstrated that hypertension correlates with tumorigenesis and prognosis of clear-cell renal cell carcinoma (ccRCC); however, the underlying molecular mechanisms remain unclear. By analyzing bulk and single-cell RNA sequencing data and experimental examining of surgical excised ccRCC samples, we found that tissue inhibitors of metalloproteinases 3 (TIMP3), a pivotal paracrine factor in suppressing tumor progression, was significantly reduced in the tumor endothelial cells of patients with hypertensive ccRCC. Besides, in tumor xenograft of NCG mouse model, compared with saline normotensive group the expression of TIMP3 was significantly decreased in the angiotensin II–induced hypertension group. Treating human umbilical vein endothelial cells (HUVEC) with the plasma of patients with hypertensive ccRCC and miR-21–5p, elevated in the plasma of patients with hypertensive ccRCC, reduced the expression of TIMP3 compared with normotensive and control littermates. We also found that the inhibition of TIMP3 expression by miR-21–5p was not through directly targeting at 3′UTR of TIMP3 but through suppressing the expression of TGFβ receptor 2 (TGFBR2). In addition, the knockout of TGFBR2 reduced TIMP3 expression in HUVECs through P38/EGR1 (early growth response protein 1) signaling axis. Moreover, via coculture of ccRCC cell lines with HUVECs and mouse tumor xenograft model, we discovered that the TIMP3 could suppress the proliferation and migration of ccRCC.

Implications:

Overall, our findings shed new light on the role of hypertension in promoting the progression of ccRCC and provide a potential therapeutic target for patients with ccRCC with hypertension.

Hypertension is a notorious risk factor of heart disease, cerebrovascular disease, and hypertensive renal disease, the leading causes of death (1). About 45% of adults in the United States suffer hypertension, suggesting elevated blood pressure is a persistent threat to the health of a large population (2).

Mounting studies have demonstrated that hypertension is associated with higher cancer risk, including renal cell carcinoma (RCC), endometrial carcinoma, colorectal cancer, breast cancer, esophageal carcinoma, and liver cancer (3–6). The association between hypertension and RCC risk is the most significant (5, 6). Besides, studies have shown that compared with normal blood pressure, hypertension is a significant risk factor for metastasis-free survival [HR, 1.71; 95% confidence interval (CI), 1.20–2.44], overall survival (HR, 1.40; 95% CI, 1.13–1.73), and cancer-specific survival (HR, 1.31; 95% CI, 1.00–1.72) in patients with RCC (7). However, the precise molecular mechanism of hypertension on the development and progression of RCC remains unclear. Emerging data implicates that hypertension could impair the endothelial function by several aspects (8, 9). Thus, we hypothesized that through affecting tumor endothelial cells (TEC), hypertension promoted the progression of clear-cell renal cell carcinoma (ccRCC). To assess the impact of hypertension on TECs of ccRCC, single-cell RNA sequencing (scRNA-seq) of ccRCC samples was performed and analyzed. We found that tissue inhibitors of metalloproteinases 3 (TIMP3) is mainly secreted by TECs in ccRCC microenvironment and significantly downregulated in tumor tissues of patients with hypertensive ccRCC.

Increasing studies revealed the critical role of the extracellular matrix (ECM) on the initiation, progression, and metastasis of cancers. Maintenance of the ECM is primarily achieved by balancing the activities of several extracellular proteinases, especially matrix metalloproteinases (MMP) and TIMPs (10, 11). Past studies have reported that TIMP3 could suppress cancer cell proliferation and metastasis in multiple cancer types (12–15). In addition, TIMP3 is also involved in multiple cardiovascular diseases, such as hypertension, myocardial disease, aortic aneurysm, and atherosclerosis (16). Therefore, TIMP3 is a critical extracellular proteinases in pathology of both cardiovascular and cancer. Importantly, our study revealed that vasculature which is vulnerable to hypertension is a major “donor” of TIMP3 in ccRCC and that the expression of TIMP3 is significantly decreased in both endothelial cluster and whole ccRCC tissue of hypertensive patients, which indicates TIMP3 could be a candidate through which hypertension exacerbates the progression of ccRCC.

Several studies reported on the critical role of the miRNAs and TGFβ pathway in regulating TIMP3 expression (17–21). In our study, TGFβ receptor 2 (TGFBR2) was downregulated by hypertension via circulating miR-21–5p, which reduced TIMP3 expression in endothelial cells (EC) through the p38/EGR1 signaling axis. Our findings provide insights into the mechanism of hypertension on the progression of ccRCC and offer a new treatment option by targeting tumor endothelial TIMP3 for patients with ccRCC with hypertension.

Cell culture

Human ccRCC cancer cell lines 786O (RRID: CVCL_1051) and Caki-1 (RRID: CVCL_0234) were purchased from Cell Bank, Chinese Academy of Sciences (Shanghai, China). Human umbilical vein endothelial cells (HUVEC) were obtained from ATCC and passage before 10 were used. Human cell lines have been authenticated using short tandem repeat profiling by a commercial testing facility (Genetic Testing Biotechnology Corporation) and screened for Mycoplasma contamination by a modified PCR method. Cells were cultured at 37°C in humidified air with 5% CO2.

Human ccRCC samples

This study was approved by the Ethics Committee of Shenzhen Second People's Hospital (20211011008), and this study was conducted in accordance with the Declaration of Helsinki. All samples (n = 63) are confirmed by pathologic examination. The patients with systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg in admission check-up or previously diagnosed with hypertension were defined as hypertension. The clinical details of these samples were shown in Supplementary Table S1.

Patient-derived ccRCC organoid

The detail of how to establish and maintain the patient-derived ccRCC organoid is in our recent paper (22).

All animal experiments

All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Shenzhen Institutes of Advanced Technology. (i) To investigate the impact of TIMP3 on ccRCC tumor xenograft. 786O (1.8 × 106) combined with TIMP3-overexpressing or control HUVECs (2 × 105) were suspended in Matrigel (100 μL) and slowly injected into the right flank of 7-week-old BALB/c nude mice. (ii) In the angiotensin II (Ang II) induced hypertension mouse model, hypertension was induced by subcutaneous infusion of 400 ng/kg/min Ang II from implanted osmotic pumps as described previously (23). Saline or Ang II osmotic pumps were planted on the 8th day after 786O (1.8×106) combined with HUVECs (2×105) injected into the right flank of 6-week-old NCG mice. To test the therapeutic effect of TIMP3 on ccRCC tumor xenograft in hypertension mice, TIMP3 protein purification from the medium of TIMP3-overexpressing HUVEC was injected into Ang II–induced hypertension mice, at 20, 24, 28, and 32 days after tumor inoculation. Tumor volumes were measured weekly by measuring the length and width of tumor diameters; tumor volumes (V) were calculated using the formula V = (L × W2)/2. At the end of the study, tumor weights were recorded.

Plasmids construction

Single-guide RNA (sgRNA) sequences targeting TIMP3 were as follows: sgRNA 1 5′-GCCTGTAGGTCGCGTCTATGAG-3′; sgRNA 2 5′ -GTCATAGACGCGACCTACAGGG-3′. The Sequences of TIMP3 sgRNAs were cloned to plenti-hU6-sgRNA-EFS-hCas9-IRES- puromycin vectors (Ubigene Technology). sgRNA sequences targeting TGFBR2 were as follows: sgRNA 1 5′-TGCTGGCGATACGCGTCCACAGG-3′; sgRNA 2 5′-GGACGATGTGCAGCGGCCACAGG-3′. The Sequences of TGFBR2 sgRNAs are cloned to plenti-hU6-sgRNA-hPGK-EGFP-IRES-Hygro.

miRNA analysis

The serum miRNA of patients with ccRCC was isolated by miRNeasy Serum/Plasma Advanced Kit (Qiagen 217204). miRNAs poly-A tailing and cDNA were synthesized by using miDETECT A Track miRNA qPCR Kit (Ribobio), The miDETECT primers for miRNA targets were purchased from Ribobio company. RT-PCR is performed on the QuantStudioTM 3 Real-Time PCR System (Thermo Fisher) using SYBR Green Mix in miDETECT A Track miRNA qPCR Kit (Ribobio). U6 was used as an internal control. Results were transformed to relative fold changes by calculation of the comparative threshold cycle (CT) method.

RT-PCR analysis

Total RNA was extracted using TRIzol reagent (Invitrogen). cDNA was synthesized using a cDNA Synthesis Kit (Yeasen). RT-PCR is performed on the QuantStudioTM 3 Real-Time PCR System (Thermo Fisher) using a PowerUpTM SYBRTM Green Master Mix (Applied Biosystems) following the manufacturer's instructions. GAPDH was used as an internal control. Results were transformed to relative fold changes by calculation of the comparative threshold cycle (CT) method. Sequences of all indicated primers were listed in Supplementary Table S2.

RNA-seq and analysis

Clear-cell renal carcinoma tissue from 63 patients was used for RNA isolation and sequencing libraries preparation. Paired-end 150 bp reads were sequenced using Illumina NovaSeq 6000 platform. After quality control, the sequence reads were aligned to the hg19 using STAR (v2.2.0) and the TPM values of each gene were calculated by RSEM (v1.2.22). Limma package was used for differential expression analysis, and GSVA packages were used for pathway enrichment analysis. For The Cancer Genome Atlas (TCGA) cohort analysis, the expression matrix and clinical information were downloaded using TCGAbiolinks (2.20.1). The Kaplan–Meier survival analyses were performed using “survival” package in R based on the median cutoff or the optimal cutoff which was detected by the ‘surv_cutpoint’ function.

scRNA-seq and data analysis

ccRCC tumor tissues from 4 patients were collected and gently dissociated into single-cell suspension. After removing dead cells, 10X Genomics Single Cell platform was used for scRNA-seq library preparation. Libraries were sequenced on the NovaSeq 6000 platform and raw fastq files were generated followed 10X Genomics guidance. CellRanger v3.0 was used for generating the raw expression matrix. Seurat3 was used for scRNA-seq data processing. Single-cell level gene set enrichment analysis (GSEA) was done using Variance-adjusted Mahalanobis in R. Differential expressed genes and differential enriched pathways were detected using FindMarker module in Seurat package.

Western blot

The ccRCC tissues and cells were homogenized and lysed in RIPA buffer (Beyotime). Equal amounts of proteins were loaded, separated on a 10% SDS-PAGE gel, electrophoretically transferred onto polyvinylidene difluoride membranes (Millipore), blocked, and incubated with the antibodies. Chemiluminescent substrates (Thermo) were added to membranes, and chemiluminescent signals were captured by Amersham Imager 680 Blot and Gel Imagers. The sources of antibodies were: TIMP3 (MAB973; R&D), TIMP3 (MAB3318; Millipore), TGFBR2 (A11765; ABclonal), CD31 (ab28364; Abcam), [CD31(3528), p-Akt (4060), Akt (4685), Erk1/2 (4695), p38 (8690), p-p38 (4511), p-Erk1/2 (4370), SAPK/JNK (9252), p-SAPK/JNK (4668), Smad2/3 (3102), p-Smad2 (18338), and GAPDH (8884; Cell Signaling Technology)].

Dual-luciferase reporter assay

2000bp upstream of the TIMP3 transcription start site and 3000 bp 3′ UTR region of TIMP3 were cloned to pGL3-basic and PGL3-CMV-LUC-MCS, respectively (Genomeditech). The pRL-SV40 plasmid was transfected, and the renilla luciferase activity was used for normalization (Promega). Following the manufacturer's protocols, plasmids were transfected into HUVECs using a NeonTM transfection system (Thermo Fisher). Luciferase activity was assessed through a Dual-Luciferase Reporter Assay System (Yeasen) by the microplate reader (PerkinElmer). All assays were performed in duplicate.

Trans-well assay

2 × 105 ccRCC cells (786O and Caki-1) combined with 2 × 105 HUVECs were suspended in serum-free medium and inoculated into the trans-well chambers with inserts of 8-μm pore size (Corning, 3422). Five hundred microliter of 10% FBS medium were added to the lower chamber. After 24 hours, trans-well chambers were taken out and fixed with 4% paraformaldehyde. The cells on the membrane of chambers were stained with 0.2% crystal violet, and cotton swabs were used to remove the cells on the top surface of the membrane. The cell numbers were counted under a light microscope.

Wound healing assay

In the coculture system, the 786O and Caki-1 were labeled with red while HUVECs were labeled with green by lentivirus. Then 1 × 105 ccRCC cells (786O and Caki-1) combined with 1 × 105 HUVECs were mixed and seeded into the 12-well plates. After growing up to 95% confluence, cells were scratched by 200-μL plastic pipette tips to form a cross in the center of the well. Then, floating cells were washed out by PBS. Finally, the wound distance was measured to calculate the migration index at 0 hour, 24 hours. The migration index was defined as follows:

Migration index = (1-wound width at 24 h/wound width at 0 h) × 100%

Cell proliferation assay

Cell proliferation was examined by Cell-Light EdU Apollo567 In Vitro Kit (Ribobio) following the manufacturer's instructions. In brief, cells and organoids were incubated with Edu for 2 hours and 8 hours, respectively. Afterward, tumor cells were digested, separated from ECs by magnetic beads cell isolation, fixed, permeabilized, and. The percentage of positive cells was detected by T cytoflex flow cytometer (Beckman) and analyzed by Flowjo software. 2.15.

Statistical analysis

All values were presented as the mean ± SD from at least three independent experiments. Differences between the two groups were analyzed using the Student t test. Pearson correlation was used to identify associations between TGFBR2 expression and TIMP3 expression. Two-way ANOVA test was used for statistical analysis of tumor volume. P < 0.05 was considered statistically significant. Statistical analysis was performed using GraphPad Prism 8 (Graphpad Software, USA).

Data availability

The raw sequence data reported in this paper were deposited in the National Genomics Data Center Genome Sequence Archive in the BIG Data Center, Chinese Academy of Sciences, under accession number HRA001508 (bulk RNA-seq) and HRA001507 (scRNA-seq). Other data that support the findings of this study are available from the corresponding author upon request.

RNA-seq data revealed the changes in patients with hypertensive ccRCC

To assess the impact of hypertension on ccRCC, we sequenced the ccRCC tissue using bulk RNA-seq (n = 63). Results of differential expression analysis showed genes were dynamically expressed between patients with hypertensive and normotensive ccRCC, and most of those genes were downregulated in patients with hypertensive ccRCC (883 downregulated vs. 68 upregulated, Fig. 1A). Using ssGSEA method, we found numerous pathways were also differentially enriched between patients with hypertensive and normotensive ccRCC (Fig. 1B). We found several pathways which associated with function of vascular endothelial, blood pressure regulation, TGFβ, and ECM cell signaling were lowly enriched in hypertension, which were confirmed by GSEA method (Fig. 1C). Those data suggested under hypertension status, substantial pathways, especially related to endothelial functions, was altered in ccRCC.

Figure 1.

Bulk RNA-seq and scRNA-seq analysis revealed the molecular difference between patients with hypertensive and normotensive ccRCC. A, Volcano plot showing the differential expressed genes between patients with different blood pressure statuses. Significantly DEGs were labeled in red (highly expressed in normal blood pressure patients) or blue (highly expressed in hypertensive patients) if genes with |log2(Fold Change)| ≧ 0.5 and P value ≦ 0.01. B, Heatmap showing the differential enriched pathways (P value ≦ 0.01 which were tested by Wilcoxon test) between patients with different blood pressure statuses. C, GSEA plot showing the enrichment status of three specified pathways between patients with different blood pressure status. D, tSNE plot showing the seven main identified cell clusters in patients with ccRCC. E, Heatmap showing the DEGs (|log2(Fold Change) | ≧ 0.5 and P value≦ 0.01 which were tested by Wilcoxon test using FindMarker module) between patients with different blood pressure statuses. F, Bubble diagram showing the enrichment of several pathways in hypertensive ccRCC ECs compared with normotensive ccRCC ECs. Bubbles with red color means the associated pathways were enriched in hypertensive ccRCC endothelium.

Figure 1.

Bulk RNA-seq and scRNA-seq analysis revealed the molecular difference between patients with hypertensive and normotensive ccRCC. A, Volcano plot showing the differential expressed genes between patients with different blood pressure statuses. Significantly DEGs were labeled in red (highly expressed in normal blood pressure patients) or blue (highly expressed in hypertensive patients) if genes with |log2(Fold Change)| ≧ 0.5 and P value ≦ 0.01. B, Heatmap showing the differential enriched pathways (P value ≦ 0.01 which were tested by Wilcoxon test) between patients with different blood pressure statuses. C, GSEA plot showing the enrichment status of three specified pathways between patients with different blood pressure status. D, tSNE plot showing the seven main identified cell clusters in patients with ccRCC. E, Heatmap showing the DEGs (|log2(Fold Change) | ≧ 0.5 and P value≦ 0.01 which were tested by Wilcoxon test using FindMarker module) between patients with different blood pressure statuses. F, Bubble diagram showing the enrichment of several pathways in hypertensive ccRCC ECs compared with normotensive ccRCC ECs. Bubbles with red color means the associated pathways were enriched in hypertensive ccRCC endothelium.

Close modal

To further investigate the differences between the TECs in patients with hypertensive ccRCC and those in patients with normotensive ccRCC, we performed scRNA-seq on 4 patients with ccRCC with or without hypertension (2 patients for each group). Seven main clusters were found, including the ECs (Fig. 1D). Differential analysis showed the molecular expression profiles of endothelial cluster between hypertension and normotension were different (Fig. 1E), in which most of the genes were downregulated in the EC from patients with hypertensive ccRCC. Then, we performed pathway enrichment analysis, the results showed vasculature development, mechanical/cytokine stimulus, cell adhesion and metalloendopeptidase activity were lowly enriched in ECs from patients with hypertensive ccRCC compared with normotensive littermates (Fig. 1F). All the above data suggested the endothelium acquired different ability in educating the microenvironment under hypertension status.

TIMP3 was reduced in patients with hypertensive ccRCC and xenograft of NCG mice with Ang II–induced hypertension

By analyzing the bulk RNA-seq and scRNA-seq of ccRCC samples, we revealed that expression levels of TIMP3, TGFBR2, GRB10, and TMEM150C were markedly higher in the endothelial cluster than in other clusters of ccRCC and their expression levels in either endothelial cluster or whole tumor tissue between patients with normotensive and hypertensive ccRCC were also significantly different (Fig. 2AC; Supplementary Table S3). Among these four genes, TIMP3 encodes a secretory protease with an important juxtacrine and paracrine functions, which enable TIMP3 to mediate the interaction between ECs and tumors. Therefore, we examined the expression level of TIMP3 in ccRCC tissues. Consistent with the bulk RNA-seq results, a marked reduction in mRNA levels of TIMP3 in patients with hypertensive ccRCC was observed (Fig. 2D). Decreased TIMP3 expression was further validated at the protein level by analyzing 48 patients with ccRCC with or without hypertension (Fig. 2E).

Figure 2.

TIMP3 was reduced in patients with hypertensive ccRCC and xenograft of NCG mice with Ang II–induced hypertension, and the EC is the major donor of TIMP3 in ccRCC. A, Venn diagram showing the intersection of DEGs between patients with ccRCC with normal blood pressure and hypertension (P value ≦ 0.01 and |log2FC| ≧ 0.5), DEGs in ECs between patients with ccRCC with normal blood pressure and hypertension (P value ≦ 0.01 and |log2FC| ≧ 0.5) and genes with higher expression levels in ECs than other cells. B and C, The Violin plot showing the expression level of TIMP3 in patients with ccRCC with normal blood pressure and hypertension from the scRNA-seq (B) and bulk RNA-seq (C) analyses. D and E, The mRNA (D) and protein (E) level of TIMP3 in ccRCC tissues from patients with normal blood pressure and hypertension [n = 33 in each group for mRNA study and n = 24 (12 samples showed) for each group in protein study]. F, The mRNA level of TIMP3 in the whole single-cell suspension of ccRCC, EC-depleted single-cell suspension of ccRCC and EC of ccRCC (n = 3; *, P < 0.05, paired t test). G, The mRNA level of TIMP3 in xenograft of NCG mice from saline control group or Ang II–induced hypertension group (n = 4 in saline group and n = 5 for Ang II group; **, P < 0.01, t test). H, Representative immunofluorescent staining images of TIMP3 (green) and DAPI (blue) in xenograft of NCG mice from saline control group or Ang II–induced hypertension group, Scale bars: 50 μm. I, Representative immunofluorescent staining images of CD31 (red), TIMP3 (green), DAPI (blue), and merged pictures in ccRCC tissues. Scale bars: 100 μm. J, Kaplan–Meier survival analysis showing the correlation between overall survival and TIMP3 expression in TGCA patients with ccRCC by using median cutoff.

Figure 2.

TIMP3 was reduced in patients with hypertensive ccRCC and xenograft of NCG mice with Ang II–induced hypertension, and the EC is the major donor of TIMP3 in ccRCC. A, Venn diagram showing the intersection of DEGs between patients with ccRCC with normal blood pressure and hypertension (P value ≦ 0.01 and |log2FC| ≧ 0.5), DEGs in ECs between patients with ccRCC with normal blood pressure and hypertension (P value ≦ 0.01 and |log2FC| ≧ 0.5) and genes with higher expression levels in ECs than other cells. B and C, The Violin plot showing the expression level of TIMP3 in patients with ccRCC with normal blood pressure and hypertension from the scRNA-seq (B) and bulk RNA-seq (C) analyses. D and E, The mRNA (D) and protein (E) level of TIMP3 in ccRCC tissues from patients with normal blood pressure and hypertension [n = 33 in each group for mRNA study and n = 24 (12 samples showed) for each group in protein study]. F, The mRNA level of TIMP3 in the whole single-cell suspension of ccRCC, EC-depleted single-cell suspension of ccRCC and EC of ccRCC (n = 3; *, P < 0.05, paired t test). G, The mRNA level of TIMP3 in xenograft of NCG mice from saline control group or Ang II–induced hypertension group (n = 4 in saline group and n = 5 for Ang II group; **, P < 0.01, t test). H, Representative immunofluorescent staining images of TIMP3 (green) and DAPI (blue) in xenograft of NCG mice from saline control group or Ang II–induced hypertension group, Scale bars: 50 μm. I, Representative immunofluorescent staining images of CD31 (red), TIMP3 (green), DAPI (blue), and merged pictures in ccRCC tissues. Scale bars: 100 μm. J, Kaplan–Meier survival analysis showing the correlation between overall survival and TIMP3 expression in TGCA patients with ccRCC by using median cutoff.

Close modal

In our RNA-seq data, hypertensive patients have higher average age and more advanced stage than normotensive patients at the time of surgery. To investigate whether hypertension per se accounts for the decreased TIMP3 level in ccRCC rather than age or stage, we firstly found out the expression level of TIMP3 did not significantly associate with age in both patients with hypertension and normal blood pressure ccRCC (Supplementary Fig. S1j). Then, we found TIMP3 were significantly decreasingly expressed in patients with hypertension ccRCC with age larger than 55 or lesser than 55 (Supplementary Figures S1k–S1l). Besides, we found the expression level of TIMP3 was significantly decreased in hypertensive patients compared with normotensive patients for each independent stage (Supplementary Fig. S1m). The above data suggested age and stage were not the main factors that account for differences in TIMP3 expression levels. Importantly, in our Ang II–induced hypertension moused model, compared with saline normotensive group, the expression of TIMP3 in xenograft (mixture of 786O and HUVEC) of NCG mice was significantly decreased in Ang II–induced hypertension group, which further supported the expression of TIMP3 was affected by blood pressure (Fig. 2G and H).

On the basis of TCGA database analysis, we found that among all cancer types, TIMP3 was a favorable prognostic marker in ccRCC, uveal melanoma, low-grade glioma, glioblastoma multiforme, and sarcoma based on Kaplan–Meier survival analysis using either a median cutoff or an optimal cutoff, and TIMP3 showed the most significant favorable prognostic marker in ccRCC using both types of cutoff (Fig. 2J; Supplementary Figures S1a–S1i). Considering the role of TIMP3 in the proliferation, invasion, and metastasis of cancer cells, we thought TIMP3 could exert an important role in inhibiting ccRCC progression.

The EC is the major donor of TIMP3 in ccRCC

An increasing body of evidence suggests TIMP3 is predominantly detected in stromal cells adjacent to cancer cells in several carcinomas, and it has also been confirmed that TIMP3 is mainly present around tumor blood vessels in colorectal cancer (24–26). Consistent with previous findings, immunofluorescence and IHC co-staining of CD31, a cell marker for ECs, with TIMP3 showed co-localization of CD31 with TIMP3 in ccRCC (Fig. 2I). By using immunomagnetic cell separation method, we depleted the ECs from the single-cell suspension of the ccRCC sample. The results showed the mRNA level of TIMP3 was extremely high in endothelial population and reduced by about 80% due to depletion of ECs in the whole single-cell suspension of ccRCC samples, indicating ECs make more significant contributions to total TIMP3 mRNA than other cell types in ccRCC (Fig. 2F). The efficiency of EC depletion was tested (Supplementary Figures S2a–S2b).

Overexpression of TIMP3 in HUVECs suppressed the proliferation and metastasis of ccRCC in a coculture system

To investigate the effect of endothelial TIMP3 on the proliferation and metastasis of ccRCC, we cocultured 786O and Caki-1, two widely used ccRCC cell lines, with HUVECs. A significant reduction in the proliferation of 786O and Caki-1 after coculture with TIMP3-overexpressing HUVECs, compared with control HUVECs, was detected by flow cytometry-based proliferation assay with 5-Ethynyl-2′-deoxyuridine (EdU) incorporation (Fig. 3A). Moreover, inhibition on the proliferation of ccRCC organoid was also observed by culturing ccRCC organoid with conditioned medium from TIMP3-overexpressing HUVECs or conditioned medium from control HUVECs with the addition of 2-μg/mL recombinant TIMP3 protein compared with conditioned medium from control HUVECs (Fig. 3B). Previous studies revealed the strong expression of TIMP3 in the invasive tumor edge, suggesting a potential role of TIMP3 in tumor invasiveness (24–26). Thereby, we examined the migration of 786O and Caki-1 by wound healing and trans-well assays using a coculture system of ccRCC and HUVECs with normal or high expression of TIMP3. In this coculture system, the 786O and Caki-1 were labeled with red, whereas HUVECs were labeled with green by lentivirus. Overexpression of TIMP3 in HUVECs results in less 786O and Caki-1 translocation on the membrane's lower surface in the trans-well assay (Fig. 3FH). In addition, there was a reduction in the wound healing capacity of 786O and Caki-1 after coculture with TIMP3-overexpressing HUVECs compared with control HUVECs (Fig. 3I and J). It has been widely reported that epithelial–mesenchymal transition (EMT) contributes to the migration of cancer cells (27). To investigate whether TIMP3 suppresses the migration of ccRCC through inhibition of EMT, we examined several EMT markers in ccRCC organoids treated with 2-μg/mL recombinant TIMP3 protein. We found TIMP3 showed no effects on the EMT of ccRCC organoids (Fig. 3K).

Figure 3.

Overexpression of TIMP3 suppressed the proliferation and migration of ccRCC in a coculture system and in xenograft tumors of hypertensive NCG mice induced by Ang II. A, Flow cytometry analysis of EdU staining in 786O and Caki-1 cocultured with control or TIMP3-overexpressing HUVECs (n = 3 for each group in 786O study and n = 4 for each group in Caki-1 study; **, P < 0.01 and ***, P < 0.001, t test). B, Flow cytometry analysis of EdU staining in ccRCC organoid incubated with conditioned medium (n = 4 for each group; **, P < 0.01, t test). C, Picture of xenograft tumors in NCG mice treated with saline (n = 4), Ang II (n = 5) and Ang II+TIMP3 (n = 5). D, Statistical analysis of tumor weight (*, P < 0.05, t test). E, Statistical analysis of tumor volume (**, P < 0.01 and ****, P < 0.0001, Two-way ANOVA test was used for statistical analysis of tumor volume). F, Trans-well assay of 786O and Caki-1 cocultured with control or TIMP3-overexpressing HUVECs. G and H, Statistical analysis of trans-well assay (four biological replicates in each group for both 786O and Caki-1 study; **, P < 0.01 and ***, P < 0.001, t test). I, Wound-healing experiment. HUVECs are labeled with green in left column, whereas 786O and Caki-1 are labeled with red in the middle column. The right column is the merged image from the left and middle column. J, Statistical analysis of wound healing experiment (eight biological replicates in each group for both 786O and Caki-1 study; *, P < 0.05 and **, P < 0.01, t test). K, The mRNA level of EMT markers in ccRCC organoid (n = 3, no significant, t test). The values are expressed as the means ± SD.

Figure 3.

Overexpression of TIMP3 suppressed the proliferation and migration of ccRCC in a coculture system and in xenograft tumors of hypertensive NCG mice induced by Ang II. A, Flow cytometry analysis of EdU staining in 786O and Caki-1 cocultured with control or TIMP3-overexpressing HUVECs (n = 3 for each group in 786O study and n = 4 for each group in Caki-1 study; **, P < 0.01 and ***, P < 0.001, t test). B, Flow cytometry analysis of EdU staining in ccRCC organoid incubated with conditioned medium (n = 4 for each group; **, P < 0.01, t test). C, Picture of xenograft tumors in NCG mice treated with saline (n = 4), Ang II (n = 5) and Ang II+TIMP3 (n = 5). D, Statistical analysis of tumor weight (*, P < 0.05, t test). E, Statistical analysis of tumor volume (**, P < 0.01 and ****, P < 0.0001, Two-way ANOVA test was used for statistical analysis of tumor volume). F, Trans-well assay of 786O and Caki-1 cocultured with control or TIMP3-overexpressing HUVECs. G and H, Statistical analysis of trans-well assay (four biological replicates in each group for both 786O and Caki-1 study; **, P < 0.01 and ***, P < 0.001, t test). I, Wound-healing experiment. HUVECs are labeled with green in left column, whereas 786O and Caki-1 are labeled with red in the middle column. The right column is the merged image from the left and middle column. J, Statistical analysis of wound healing experiment (eight biological replicates in each group for both 786O and Caki-1 study; *, P < 0.05 and **, P < 0.01, t test). K, The mRNA level of EMT markers in ccRCC organoid (n = 3, no significant, t test). The values are expressed as the means ± SD.

Close modal

TIMP3 decreased weight and volume of tumor xenograft in nude mice and hypertensive NCG mice induced by Ang II

To investigate the impact of TIMP3 on ccRCC tumor xenograft, we firstly seeded 786O combined with TIMP3-overexpressing or control HUVECs into the subcutaneous tissue of nude mice. We found that compared with control littermates, TIMP3-overexpression in HUVECs suppressed the growth of 786O cancer cells implanted in nude mice (Supplementary Figs. S2c–S2e). Then, we establish another ccRCC xenograft model by seeding 786O cells and HUVECs (9:1) into the right flank of NGC mice. On the basis of that model, we further applied subcutaneous infusion of Ang II to induced hypertension on the NCG xenograft mice followed by treatment with TIMP3 or not. The systolic and diastolic pressure of mice were measured to validate the hypertension model at 1, 3, 5, 7, 14, 21, and 28 days after implantation of osmotic pumps (Supplementary Fig. S2f). Compared with the saline normotensive group the weight and volume of tumor xenograft were significantly increased in Ang II–induced hypertension group, but the increase was suppressed by TIMP3 injection (Fig. 3CE).

Circulating miR-21–5p was increased and miR-21–5p mimic reduced TIMP3 expression in ECs

Substantial evidence indicated circulating secretory factors were altered in patients with hypertension (28–30). To determine whether certain secretory factors in hypertensive patients’ blood attenuated TIMP3 expression in ECs, we examined TIMP3 expression in HUVECs after incubation with the plasma of patients with ccRCC. We found that the expression of TIMP3 HUVECs was downregulated by plasma of patients with ccRCC with hypertension compared with normal blood pressure (Fig. 4A). Previous studies have revealed plasma miRNA profile was altered in hypertension (31–33). We selected five miRNAs: Let-7e-5p, miR-21–5p, miR-122–5p, miR-221–3p, and miR222–3p, which are abundant in blood, significantly upregulated in blood of hypertensive group compared with the normotensive groups in other studies and potentially regulation of TIMP3 expression. Then we validated the level of indicated miRNAs in plasma of patients with ccRCC and found out that Let-7e-5p, miR-21–5p, and miR-122–5p was increased in hypertensive group compared with the normotensive group (Fig. 4B). In addition, we detected the expression of TIMP3 in HUVECS after treatment with miRNA negative control, Let-7e-5p mimic, miR-21–5p mimic, and miR-122–5p mimic. We found that miR-21–5p mimic significantly reduced, while Let-7e-5p mimic and miR122–5p mimic slightly increase, the expression of TIMP3 in HUVECs (Fig. 4C).

Figure 4.

Circulating miR-21–5p was increased and miR-21–5p mimic reduced TIMP3 expression in ECs. A, The mRNA level of TIMP3 in HUVECs after 12 and 24 hours of incubation with the plasma of patients with ccRCC with normal blood pressure and hypertension. (n = 19 for normal blood pressure group and n = 16 for hypertension group; ***, P < 0.001, ns: not significant; Two-way ANOVA test was used for statistical analysis of the effects of treatment time and hypertension status on TIMP3 expression). B, The concentration of indicated miRNA in the plasma of patients with ccRCC with normal blood pressure and hypertension (n = 10 for each group; **, P < 0.01 and ****, P < 0.0001, t test within each group). C, The mRNA level of TIMP3 in HUVECs after treatment of miRNAs negative control and indicated miRNA mimic (n = 4 for each group; **, P < 0.01, t test). D and E, The relative luciferase activity of reporters carrying 3′ UTR and promoter region of TIMP3 in HUVEC after treatment of miRNAs negative control and miR-21–5p mimic (four biological replicates with two technical replicates in each group for 3′ UTR and promoter studies; *, P < 0.05, ns: no significant, t test). The values are expressed as the means ± SD.

Figure 4.

Circulating miR-21–5p was increased and miR-21–5p mimic reduced TIMP3 expression in ECs. A, The mRNA level of TIMP3 in HUVECs after 12 and 24 hours of incubation with the plasma of patients with ccRCC with normal blood pressure and hypertension. (n = 19 for normal blood pressure group and n = 16 for hypertension group; ***, P < 0.001, ns: not significant; Two-way ANOVA test was used for statistical analysis of the effects of treatment time and hypertension status on TIMP3 expression). B, The concentration of indicated miRNA in the plasma of patients with ccRCC with normal blood pressure and hypertension (n = 10 for each group; **, P < 0.01 and ****, P < 0.0001, t test within each group). C, The mRNA level of TIMP3 in HUVECs after treatment of miRNAs negative control and indicated miRNA mimic (n = 4 for each group; **, P < 0.01, t test). D and E, The relative luciferase activity of reporters carrying 3′ UTR and promoter region of TIMP3 in HUVEC after treatment of miRNAs negative control and miR-21–5p mimic (four biological replicates with two technical replicates in each group for 3′ UTR and promoter studies; *, P < 0.05, ns: no significant, t test). The values are expressed as the means ± SD.

Close modal

miR-21–5p suppressed the TIMP3 expression via reduction of TGFBR2 in ECs

Several studies have also demonstrated miR-21–5p could downregulate TIMP3 (34–36). To investigate whether miR-21–5p downregulated TIMP3 expression through the posttranscriptional level by directly targeting the long 3′ UTR of TIMP3 or through transcriptional level, we constructed recombinant reporter plasmid in which the luciferase gene was placed under the control of the promoter and 3′ UTR region of TIMP3, respectively. We found out that luciferase activity driven by 3′ UTR of TIMP3 remained unchanged but driven by the promoter of TIMP3 was reduced by miR-21–5p mimic, which indicated certain factors targeted by miR-21–5p are involved in downregulation of TIMP3 (Fig. 4D and E). Therefore, we searched for that factor from the top 100 genes that have similar expression patterns with TIMP3 in ccRCC by GEPIA2 (Supplementary Table S4) and the Differentially Expressed Genes (DEG) in the endothelial cluster of our single-cell sequencing data between patients with ccRCC with hypertension and normal blood pressure (Supplementary Table S3). Moreover, highly ranked intersected genes were further narrowed down through whether the 3′ UTR region of candidates could be targeted by miR-21–5p by software estimate. After analysis, TGFBR2 was singled out.

TCGA database analysis and our sequencing data consistently showed a strong correlation between TGFBR2 and TIMP3 in ccRCC (Fig. 5A). Similar to TIMP3, TGFBR2 is mainly expressed in the endothelial cluster in single-cell sequencing, and ECs contribute about 80% of total TGFBR2 mRNA in ccRCC (Fig. 5D). Both bulk RNA-seq data and scRNA-seq data demonstrated the decreased expression of TGFBR2 in patients with hypertensive ccRCC compared with normotensive ones (Fig. 5B and C). This marked reduction of TGFBR2 expression in ccRCC caused by hypertension was further validated by quantitative PCR, and Western blot analysis (Fig. 5E and F). Moreover, in our Ang II–induced hypertension moused model, compared with saline normotensive group, the expression of TIMP3 in xenograft (mixture of 786O and HUVEC) of NCG mice was significantly decreased in Ang II–induced hypertension group (Fig. 5G).

Figure 5.

miR-21–5p suppressed the TIMP3 expression via reduction of TGFBR2 in ECs. A, Correlation between TIMP3 and TGFBR2 in patients with ccRCC from TCGA database and our sequenced data (n = 535 for TCGA and n = 63 for our data, tested by pearson correlation). B and C, The Violin plot showing the expression level of TGFBR2 in patients with ccRCC with normal blood pressure and hypertension from the scRNA-seq (B) and bulk RNA-seq (C) analysis. D, The mRNA level of TGFBR2 in the whole single-cell suspension of ccRCC, EC-depleted single-cell suspension of ccRCC and EC of ccRCC (n = 3; *, P < 0.05, paired t test). E and F, The mRNA (E) and protein (F) level of TGFBR2 in ccRCC tissues from patients with normal blood pressure and hypertension (n = 33 in each group for mRNA study and n = 24 (12 samples showed) for each group in protein study; *, P < 0.05, t test). G, The mRNA level of TGFBR2 in xenograft of NCG mice from saline control group or Ang II–induced hypertension group (n = 4 in saline group and n = 5 for Ang II group; **, P < 0.01, t test). H, The mRNA level of TGFBR2 in HUVECs after 12 and 24 hours of incubation with the plasma of patients with ccRCC with normal blood pressure and hypertension. (n = 19 for normal blood pressure group and n = 16 for hypertension group; ****, P < 0.0001; Two-way ANOVA test was used for statistical analysis of the effects of treatment time and hypertension status on TIMP3 expression). I, The mRNA level of TGFBR2 in HUVECs after treatment of miRNAs negative control and miR21–5p mimic (n = 4 for each group; ***, P < 0.001, t test). J, The mRNA level of TIMP3 in HUVECs after treatment of miRNAs negative control plus GFP lenti-virus, miR21–5p mimic plus GFP lenti-virus, and miR21–5p mimic plus TGFBR2-overexpressing lenti-virus (n = 4 for each group; *, P < 0.05, t test). K, The mRNA level of TIMP3 and TGFBR2 after knockout of TGFBR2 and TIMP3 in HUVECs (three biological replicates in each group for both TIMP3 and TGFBR2 knockout study, ns: no significant and ***, P < 0.001, t test). The values are expressed as the means ± SD.

Figure 5.

miR-21–5p suppressed the TIMP3 expression via reduction of TGFBR2 in ECs. A, Correlation between TIMP3 and TGFBR2 in patients with ccRCC from TCGA database and our sequenced data (n = 535 for TCGA and n = 63 for our data, tested by pearson correlation). B and C, The Violin plot showing the expression level of TGFBR2 in patients with ccRCC with normal blood pressure and hypertension from the scRNA-seq (B) and bulk RNA-seq (C) analysis. D, The mRNA level of TGFBR2 in the whole single-cell suspension of ccRCC, EC-depleted single-cell suspension of ccRCC and EC of ccRCC (n = 3; *, P < 0.05, paired t test). E and F, The mRNA (E) and protein (F) level of TGFBR2 in ccRCC tissues from patients with normal blood pressure and hypertension (n = 33 in each group for mRNA study and n = 24 (12 samples showed) for each group in protein study; *, P < 0.05, t test). G, The mRNA level of TGFBR2 in xenograft of NCG mice from saline control group or Ang II–induced hypertension group (n = 4 in saline group and n = 5 for Ang II group; **, P < 0.01, t test). H, The mRNA level of TGFBR2 in HUVECs after 12 and 24 hours of incubation with the plasma of patients with ccRCC with normal blood pressure and hypertension. (n = 19 for normal blood pressure group and n = 16 for hypertension group; ****, P < 0.0001; Two-way ANOVA test was used for statistical analysis of the effects of treatment time and hypertension status on TIMP3 expression). I, The mRNA level of TGFBR2 in HUVECs after treatment of miRNAs negative control and miR21–5p mimic (n = 4 for each group; ***, P < 0.001, t test). J, The mRNA level of TIMP3 in HUVECs after treatment of miRNAs negative control plus GFP lenti-virus, miR21–5p mimic plus GFP lenti-virus, and miR21–5p mimic plus TGFBR2-overexpressing lenti-virus (n = 4 for each group; *, P < 0.05, t test). K, The mRNA level of TIMP3 and TGFBR2 after knockout of TGFBR2 and TIMP3 in HUVECs (three biological replicates in each group for both TIMP3 and TGFBR2 knockout study, ns: no significant and ***, P < 0.001, t test). The values are expressed as the means ± SD.

Close modal

To determine whether s miR-21–5p suppressed the TIMP3 expression via the reduction of TGFBR2 in ECs, we examined TGFBR2 expression in HUVECs after incubation with miR-21–5p mimic and the plasma of patients with hypertensive ccRCC. We found that the expression of TGFBR2 in HUVECs was downregulated by miR-21–5p mimic (Fig. 5I) and the plasma of patients with ccRCC with hypertension compared with control and normotensive littermates (Fig. 5H). Moreover, the reduced TIMP3 expression induced by miR-21–5p mimic was restored by overexpression of TGFBR2 (Fig. 5J), and knockout of TGFBR2 significantly reduced TIMP3 expression while knockout of TIMP3 did not significantly change TGFBR2 expression in HUVECs (Fig. 5K).

TGFBR2 regulated TIMP3 expression via the p38 signaling pathway in ECs

There are two pathways for TGFBR2 to initiate cellular signaling: (i) the canonical SMAD-dependent signaling pathway; (ii) noncanonical SMAD-independent pathways, including PI3K/AKT pathways and several branches of MAPK pathways, such as ERKs, p38, and JNKs (21). Accordingly, we assessed the effect of TGFBR2 knockout on various downstream signaling pathways in HUVECs. Knockout of TGFBR2 decreased the phosphorylation of Smad2 and p38 but enhanced the phosphorylation of AKT and ERK1/2 while the phosphorylation of JNK remained unchanged (Fig. 6A). TGFβ1, TGFβ receptor agonist and A83–01 (an inhibitor of ALK5 (type I TGFβ receptor)) did not alter TIMP3 expression in ECs, suggesting the regulatory function of TGFBR2 on TIMP3 expression in HUVECs does not involve the TGFβ/SMAD signaling axis (Fig. 6B). Efficacy of TGFβ1 and A83–01was validated by detecting Smad2 phosphorylation (Supplementary Figs. S3c and S3d). Furthermore, analysis of TCGA database also revealed no significant correlation between the expression of TGFB1 and TIMP3 in ccRCC, indicating TGFβ agonist is dispensable for regulating TIMP3 expression by TGFBR2 in ccRCC (Supplementary Figures S3a and S3ab). However, SB202190, a potent p38 MAPK inhibitor, suppressed the expression of TIMP3 in HUVECs. Given that knockout of TGFBR2 led to decreased phosphorylation of p38, the p38 signaling pathway potentially contributed to the expression of TIMP3 regulated by TGFBR2. Then we abolished the activation of AKT and ERK1/2 signaling pathway induced by the knockout of TGFBR2 by their antagonists (MK2206 and SCH772984) in TGFBR2 Knockout HUVECs. However, the expression of TIMP3 was not restored or decreased. Likewise, treatment of MK2206 and SCH772984 caused a reduction of TIMP3 expression in control HUVECs (Fig. 6C). Overall, our results reveal that the p38 pathway, rather than the TGFβ/SMAD, Akt, and Erk1/2 pathways, contribute to TIMP3 downregulation induced by TGFBR2 knockout in HUVECs.

Figure 6.

TGFBR2 regulated the TIMP3 expression through P38/EGR1 axis in ECs. A, The expression level of total and phosphorylated Smad2, Jnk, Erk1/2, p38, and Akt in control and TGFBR2 knockout HUVECs (four biological replicates in each group). B, The mRNA level of TIMP3 in HUVECs after treatment with 5 μmol/L A83–01, 5 ng/mL TGFβ1, or 10 μmol/L SB202190 for 24 hours (n = 4 in each group; *, P < 0.05, t test). C, The mRNA expression level of TIMP3 in control HUVECs and TGFBR2 knockout HUVECs after treatment with DMSO, 10 μmol/L SCH772984, or 10 μmol/L MK2206 for 24 hours (n = 3 in each group). D, The relative luciferase activity of reporters carrying promoter region of TIMP3 in control HUVECs and TGFBR2 knockout HUVECs (four biological replicates with two technical replicates in each group; *, P < 0.05, t test). E, The relative luciferase activity of reporters carrying promoter region of TIMP3 in HUVECs treated with DMSO or SB202190 (four biological replicates with two technical replicates in each group; *, P < 0.05, t test). F, The mRNA level of indicated genes in control HUVECs, TGFBR2 knockout HUVECs, and TGFBR2-overexpressing HUVECs determined by RT-PCR (n = 3 in control and TGFBR2-overexpressing group, n = 4 in TGFBR2 knockout group; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; Ordinary one-way ANOVA). G, The mRNA level of indicated genes in HUVECs treated with DMSO or SB202190 was determined by RT-PCR (n = 3 for each group; *, P < 0.05, t test). H and I, The relative mRNA expression of TIMP3 (H) and the relative luciferase activity of TIMP3 promoter–reporter (I) in HUVECs transfected with EGR1 siRNA or transfected with EGR1-overexpressing plasmid compared with the HUVECs transfected with siRNA NC or empty vector (three biological replicates for each group in EGR1 siRNA study and four biological replicates for each group in EGR1-overexpressing study; ***, P < 0.001, t test).

Figure 6.

TGFBR2 regulated the TIMP3 expression through P38/EGR1 axis in ECs. A, The expression level of total and phosphorylated Smad2, Jnk, Erk1/2, p38, and Akt in control and TGFBR2 knockout HUVECs (four biological replicates in each group). B, The mRNA level of TIMP3 in HUVECs after treatment with 5 μmol/L A83–01, 5 ng/mL TGFβ1, or 10 μmol/L SB202190 for 24 hours (n = 4 in each group; *, P < 0.05, t test). C, The mRNA expression level of TIMP3 in control HUVECs and TGFBR2 knockout HUVECs after treatment with DMSO, 10 μmol/L SCH772984, or 10 μmol/L MK2206 for 24 hours (n = 3 in each group). D, The relative luciferase activity of reporters carrying promoter region of TIMP3 in control HUVECs and TGFBR2 knockout HUVECs (four biological replicates with two technical replicates in each group; *, P < 0.05, t test). E, The relative luciferase activity of reporters carrying promoter region of TIMP3 in HUVECs treated with DMSO or SB202190 (four biological replicates with two technical replicates in each group; *, P < 0.05, t test). F, The mRNA level of indicated genes in control HUVECs, TGFBR2 knockout HUVECs, and TGFBR2-overexpressing HUVECs determined by RT-PCR (n = 3 in control and TGFBR2-overexpressing group, n = 4 in TGFBR2 knockout group; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; Ordinary one-way ANOVA). G, The mRNA level of indicated genes in HUVECs treated with DMSO or SB202190 was determined by RT-PCR (n = 3 for each group; *, P < 0.05, t test). H and I, The relative mRNA expression of TIMP3 (H) and the relative luciferase activity of TIMP3 promoter–reporter (I) in HUVECs transfected with EGR1 siRNA or transfected with EGR1-overexpressing plasmid compared with the HUVECs transfected with siRNA NC or empty vector (three biological replicates for each group in EGR1 siRNA study and four biological replicates for each group in EGR1-overexpressing study; ***, P < 0.001, t test).

Close modal

TGFBR2 mediated the TIMP3 expression at the transcriptional level by regulation of EGR1 in HUVECs

Knockout of TGFBR2 and suppression of the p38 signaling pathway by SB202190 in HUVECs reduced TIMP3 promoter–reporter gene activity or left the TIMP3 3′ UTR reporter gene activity unchanged, indicating TGFBR2 mediates the expression of TIMP3 by influencing transcription factors which target the promoter region of TIMP3 (Fig. 6D and E; Supplementary Figs. S3e and S3f). Hence, we searched for transcription factors that can potentially bind to the promoter region of TIMP3, and high-score transcription factors with an essential function in ECs were selected (Supplementary Table S5). The expression of selected transcription factors was assessed after TGFBR2 overexpression, knockout of TGFBR2, and treatment with SB202190 in HUVECs, respectively. EGR1, a transcription factor rapid induced and highly involved in several vascular diseases, was enhanced after knockout of TGFBR2 or treatment of SB202190 and suppressed by overexpression of TGFBR2 in HUVECs (Fig. 6F and G). Other candidate genes failed to show significant opposite changes induced by overexpression and knockout of TGFBR2. Furthermore, the TIMP3 promoter–reporter gene activity and the expression of TIMP3 were inhibited by EGR1 overexpression and increased by knockdown of EGR1 (Fig. 6H and I; Supplementary Fig. S3g).

Although it has been widely reported that the loss of TIMP3 promotes cancer progression, few studies have mentioned the contribution of ECs and other stromal cells to TIMP3 expression in cancers (12–15). Low expression of TIMP3 in tumor tissues has always been attributed to the loss of TIMP3 expression in cancer cells and several mechanisms related to cancer cells. To the best of our knowledge, this is the first study to demonstrate that ECs play a significant role in the regulation of TIMP3 expression in ccRCC and investigate the role of TIMP3 on proliferation and migration by coculture of ccRCC and ECs with overexpressed TIMP3.

Previous studies indicated that hypertension could affect TECs through circulating secretory factors and mechanical stress (28–30, 37, 38). On the one hand, dysregulation of important secreted vasoactive factors in the circulation system of hypertensive patients results in a broad spectrum of biochemical events in TECs (28–30). In our in vitro study, the plasma of patients with hypertensive ccRCC and miR-21–5p which is upregulated in hypertensive patients’ plasma reduced the expression of TGFBR2 and TIMP3 in HUVECs compared with the control and normotensive littermates, indicating hypertension could affect TECs through circulating factors. On the other hand, ECs layered on the intima of blood vessels are under constant mechanical stimuli, including the shear stress and circumferential stretch generated by blood flow and pressure (37, 38). An elevation in blood pressure increases the circumferential stretch on the vessel wall, which is the culprit of ECs dysfunction and vascular remodeling in patients with hypertension (37, 38). In addition to TIMP3, our study also reveals TMEM150C which encodes Tentonin 3, a mechanosensitve ion channel that is activated by mechanical stimuli in various cell types, is significantly reduced in the TECs of patients with hypertensive ccRCC. Emerging data indicates Tentonin 3/TMEM150C is critical for blood pressure regulation and genetic ablation of Tentonin 3 induces hypertension in mice (39, 40). We will further investigate the effects of Tentonin 3 deficiency and mechanical stress on TECs in future study.

Increasing reports support the TECs play a crucial role in the tumor metabolism, immune infiltration and metastasis (41, 42). By affecting TEC, hypertension could play an important role in tumor progression. Capillaries, mainly composed of ECs, supply tumor cells with oxygen, nutrients, and chemical mediators and drain carbon dioxide and metabolic waste out, which is pivotal for the proliferation and metabolism of tumors (43). Our sequencing results demonstrate the signaling related to the metabolism of lipid and triglyceride are different between patients with ccRCC with or without hypertension. Emerging data indicate that secretion of chemokines, cytokines, and extracellular enzymes from TECs could function as paracrine or endocrine regulators for immune infiltration and tumor microenvironment (41). In our study, sequencing analysis shows leukocyte transendothelial migration and endothelial chemotaxis signaling are altered by hypertension in ccRCC microenvironment. Moreover, a growing body of evidence suggests that TECs are vital for tumor metastasis due to the physical contact with tumor cells and their interaction by juxtacrine and paracrine signaling (42). Importantly, our sequencing data shows pathway associated with the establishment of endothelial barrier is changed by hypertension in ccRCC TECs.

TIMP3 could exert its anti-tumor effects in various aspects. Beyond suppression of MMPs to impede tumor degradation of ECM components, TIMP3 inhibits many A Disintegrin and Metalloproteinase enzymes that regulate the shedding of surface receptors' extracellular domain, and the bioavailability of various growth factors synthesized and secreted precursors (11). Therefore, TIMP3 could mediate several cellular signals, including TNF, FAS, TGFβ, EGF, NOTCH, IL6, CXCL8, and CCL7, and exert a plethora of effects on tumor cells and the tumor microenvironment (44). Moreover, TIMP3 has been shown to inhibit angiogenesis by directly blocking VEGF from binding to VEGF receptor 2 in a manner independent of its TIMP activity (45). Therefore, through reducing TIMP3 expression in TECs, hypertension could influence the progression of ccRCC in multi-aspect.

Similar to cancers, the prevalence of hypertension increases with aging, affecting 22.4% between the ages of 18 and 39 to 74.5% among those aged over 59 (46). Consequently, due to the growing aging population, patients suffering from hypertension and cancer have rapidly increased (3). However, poor knowledge of the underlying mechanism of hypertension on the prognosis of cancers hinders the optimization of anticancer therapy in hypertensive patients. Besides, hypertension is one of the most frequently encountered cardiovascular side effects of many anticancer treatments (47, 48). The prohypertensive effect of anticancer treatment further compounds pharmacologic management of blood pressure in this population. Therefore, it is urgent to develop novel anticancer drugs with fewer cardiovascular side effects. TIMP3 is a promising therapeutic target for cancers and meanwhile with great potential in preventing blood pressure increase (10, 11, 49). Dynamic regulation of ECM by MMPs and TIMPs is highly involved in the pathogenesis of hypertension, and TIMP3 is the primary TIMP to regulate agonist-induced vascular remodeling and hypertension (49, 50). It is worthy to investigate whether prohypertensive effect of anticancer treatment could be mitigated by replenishment of TIMP3 in future study.

In sum, through downregulation of TIMP3 which is a crucial secreted factor of TECs in ccRCC, hypertension promotes the proliferation and migration of ccRCC. Our study reveals novel mechanistic insights into the role of hypertension on the progression of ccRCC and provides a new therapeutic option for patients with ccRCC with hypertension.

No disclosures were reported.

C. Wang: Conceptualization, data curation, formal analysis, validation, investigation, visualization, methodology, writing–original draft. H. Xu: Conceptualization, resources, data curation, software, formal analysis, funding acquisition, investigation, visualization, methodology, writing–original draft. X. Liao: Resources, validation, investigation, methodology. W. Wang: Resources, validation, investigation, methodology. W. Wu: Conceptualization, formal analysis, validation, investigation, visualization, methodology. W. Li: Conceptualization, formal analysis, validation, investigation, visualization, methodology. L. Niu: Conceptualization, resources, formal analysis, funding acquisition, investigation, visualization, methodology, writing–review and editing. Z. Li: Resources, investigation, visualization, methodology, writing–review and editing. A. Li: Resources, methodology, writing–review and editing. Y. Sun: Resources, supervision, funding acquisition, methodology, project administration, writing–review and editing. W. Huang: Conceptualization, resources, data curation, supervision, funding acquisition, project administration, writing–review and editing. F. Song: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing.

This work was supported by National Key R&D Program of China (2019YFA0906000, to W. Huang), the Natural Science Founding of Guangdong Province (2021A1515012488, to F. Song), Shenzhen Science and Technology Innovation Commission | Sanming Project of Medicine in Shenzhen SZSM201412018 and SZSM201512037, to W. Huang), the National Natural Science Foundation of China (81772737, to W. Huang; 31670757, to F. Song; and 82103203, to H. Xu), Shenzhen Fundamental Research Program (GJHZ20180926165202081 and JCYJ20200109120016553), Shenzhen Institute of Synthetic Biology Scientific Research Program (ZTXM20214005) and the China Postdoctoral Science Foundation Grant (2020M670051ZX, to L. Niu).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/).

1.
Kochanek
KD
,
Murphy
SL
,
Xu
J
,
Arias
E
.
Deaths: final data for 2017
.
Natl Vital Stat Rep
2019
;
68
:
1
77
.
2.
Ritchey
MD
,
Gillespie
C
,
Wozniak
G
,
Shay
CM
,
Thompson-Paul
AM
,
Loustalot
F
, et al
.
Potential need for expanded pharmacologic treatment and lifestyle modification services under the 2017 ACC/AHA hypertension guideline
.
J Clin Hypertens
2018
;
20
:
1377
91
.
3.
Kidoguchi
S
,
Sugano
N
,
Tokudome
G
,
Yokoo
T
,
Yano
Y
,
Hatake
K
, et al
.
New concept of onco-hypertension and future perspectives
.
Hypertension
2021
;
77
:
16
27
.
4.
Kim
CS
,
Han
KD
,
Choi
HS
,
Bae
EH
,
Ma
SK
,
Kim
SW
.
Association of hypertension and blood pressure with kidney cancer risk: a nationwide population-based cohort study
.
Hypertension
2020
;
75
:
1439
46
.
5.
Christakoudi
S
,
Kakourou
A
,
Markozannes
G
,
Tzoulaki
I
,
Weiderpass
E
,
Brennan
P
, et al
.
Blood pressure and risk of cancer in the European prospective investigation into cancer and nutrition
.
Int J Cancer
2020
;
146
:
2680
93
.
6.
Seretis
A
,
Cividini
S
,
Markozannes
G
,
Tseretopoulou
X
,
Lopez
DS
,
Ntzani
EE
, et al
.
Association between blood pressure and risk of cancer development: a systematic review and meta-analysis of observational studies
.
Sci Rep
2019
;
9
:
8565
.
7.
Kim
SH
,
Park
B
,
Hwang
EC
,
Hong
SH
,
Jeong
CW
,
Kwak
C
, et al
.
Retrospective multicenter long-term follow-up analysis of prognostic risk factors for recurrence-free, metastasis-free, cancer-specific, and overall survival after curative nephrectomy in nonmetastatic renal cell carcinoma
.
Front Oncol
2019
;
9
:
859
.
8.
Dharmashankar
K
,
Widlansky
ME
.
Vascular endothelial function and hypertension: insights and directions
.
Curr Hypertens Rep
2010
;
12
:
448
55
.
9.
Konukoglu
D
,
Uzun
H
.
Endothelial dysfunction and hypertension
.
Adv Exp Med Biol
2017
;
956
:
511
40
.
10.
Su
CW
,
Lin
CW
,
Yang
WE
,
Yang
SF
.
TIMP3 as a therapeutic target for cancer
.
Ther Adv Med Oncol
2019
;
11
:
1758835919864247
.
11.
Jackson
HW
,
Defamie
V
,
Waterhouse
P
,
Khokha
R
.
TIMPs: versatile extracellular regulators in cancer
.
Nat Rev Cancer
2017
;
17
:
38
53
.
12.
Yu
J
,
Xu
QG
,
Wang
ZG
,
Yang
Y
,
Zhang
L
,
Ma
JZ
, et al
.
Circular RNA cSMARCA5 inhibits growth and metastasis in hepatocellular carcinoma
.
J Hepatol
2018
;
68
:
1214
27
.
13.
Huang
HL
,
Liu
YM
,
Sung
TY
,
Huang
TC
,
Cheng
YW
,
Liou
JP
, et al
.
Erratum: TIMP3 expression associates with prognosis in colorectal cancer and its novel arylsulfonamide inducer, MPT0B390, inhibits tumor growth, metastasis, and angiogenesis: erratum
.
Theranostics
2021
;
11
:
2079
.
14.
Su
CW
,
Chang
YC
,
Chien
MH
,
Hsieh
YH
,
Chen
MK
,
Lin
CW
, et al
.
Loss of TIMP3 by promoter methylation of Sp1 binding site promotes oral cancer metastasis
.
Cell Death Dis
2019
;
10
:
793
.
15.
Das
AM
,
Bolkestein
M
,
van der Klok
T
,
Oude Ophuis
CM
,
Vermeulen
CE
,
Rens
JA
, et al
.
Tissue inhibitor of metalloproteinase 3 (TIMP3) expression decreases during melanoma progression and inhibits melanoma cell migration
.
Eur J Cancer
2016
;
66
:
34
46
.
16.
Fan
D
,
Kassiri
Z
.
Biology of tissue inhibitor of metalloproteinase 3 (TIMP3), and its therapeutic implications in cardiovascular pathology
.
Front Physiol
2020
;
11
:
661
.
17.
Leivonen
SK
,
Lazaridis
K
,
Decock
J
,
Chantry
A
,
Edwards
DR
,
Kahari
VM
.
TGFβ-elicited induction of tissue inhibitor of metalloproteinases (TIMP) 3 expression in fibroblasts involves complex interplay between Smad3, p38alpha, and ERK1/2
.
PLoS One
2013
;
8
:
e57474
.
18.
Garcia-Alvarez
J
,
Ramirez
R
,
Checa
M
,
Nuttall
RK
,
Sampieri
CL
,
Edwards
DR
, et al
.
Tissue inhibitor of metalloproteinase 3 is upregulated by transforming growth factor-beta1 in vitro and expressed in fibroblastic foci in vivo in idiopathic pulmonary fibrosis
.
Exp Lung Res
2006
;
32
:
201
14
.
19.
Zhao
DL
,
Li
HT
,
Liu
SH
.
TIMP3/TGFβ1 axis regulates mechanical loading-induced chondrocyte degeneration and angiogenesis
.
Mol Med Rep
2020
;
22
:
2637
44
.
20.
Colak
S
,
Ten Dijke
P
.
Targeting TGFβ signaling in cancer
.
Trends Cancer
2017
;
3
:
56
71
.
21.
Goumans
MJ
,
Ten Dijke
P
.
TGFβ signaling in control of cardiovascular function
.
Cold Spring Harb Perspect Biol
2018
;
10
:
a022210
.
22.
Li
Z
,
Xu
H
,
Yu
L
,
Wang
J
,
Meng
Q
,
Mei
H
, et al
.
Patient-derived renal cell carcinoma organoids for personalized cancer therapy
.
Clin Transl Med
2022
;
12
:
e970
.
23.
Lu
H
,
Howatt
DA
,
Balakrishnan
A
,
Moorleghen
JJ
,
Rateri
DL
,
Cassis
LA
, et al
.
Subcutaneous angiotensin II infusion using osmotic pumps induces aortic aneurysms in mice
.
J Vis Exp
2015
;
103
:
53191
.
24.
Airola
K
,
Ahonen
M
,
Johansson
N
,
Heikkila
P
,
Kere
J
,
Kahari
VM
, et al
.
Human TIMP3 is expressed during fetal development, hair growth cycle, and cancer progression
.
J Histochem Cytochem
1998
;
46
:
437
47
.
25.
Byrne
JA
,
Tomasetto
C
,
Rouyer
N
,
Bellocq
JP
,
Rio
MC
,
Basset
P
.
The tissue inhibitor of metalloproteinases 3 gene in breast carcinoma: identification of multiple polyadenylation sites and a stromal pattern of expression
.
Mol Med
1995
;
1
:
418
27
.
26.
Powe
DG
,
Brough
JL
,
Carter
GI
,
Bailey
EM
,
Stetler-Stevenson
WG
,
Turner
DR
, et al
.
TIMP3 mRNA expression is regionally increased in moderately and poorly differentiated colorectal adenocarcinoma
.
Br J Cancer
1997
;
75
:
1678
83
.
27.
Dongre
A
,
Weinberg
RA
.
New insights into the mechanisms of epithelial–mesenchymal transition and implications for cancer
.
Nat Rev Mol Cell Biol
2019
;
20
:
69
84
.
28.
Onuh
JO
,
Aliani
M
.
Metabolomics profiling in hypertension and blood pressure regulation: a review
.
Clin Hypertens
2020
;
26
:
23
.
29.
Jusic
A
,
Devaux
Y
,
EU-CardioRNA COST Action (CA17129)
.
Noncoding RNAs in hypertension
.
Hypertension
2019
;
74
:
477
92
.
30.
Mikolajczyk
TP
,
Szczepaniak
P
,
Vidler
F
,
Maffia
P
,
Graham
GJ
,
Guzik
TJ
.
Role of inflammatory chemokines in hypertension
.
Pharmacol Ther
2021
;
223
:
107799
.
31.
Cengiz
M
,
Karatas
OF
,
Koparir
E
,
Yavuzer
S
,
Ali
C
,
Yavuzer
H
, et al
.
Differential expression of hypertension-associated microRNAs in the plasma of patients with white coat hypertension
.
Medicine
2015
;
94
:
e693
.
32.
Li
X
,
Wei
Y
,
Wang
Z
.
microRNA-21 and hypertension
.
Hypertens Res
2018
;
41
:
649
61
.
33.
Improta Caria
AC
,
Nonaka
CKV
,
Pereira
CS
,
Soares
MBP
,
Macambira
SG
,
Souza
BSF
.
Exercise training-induced changes in microRNAs: beneficial regulatory effects in hypertension, type 2 diabetes, and obesity
.
Int J Mol Sci
2018
;
19
:
3608
.
34.
Bo
P
,
Julang
L
.
MicroRNA-21 upregulates metalloprotease by downregulating TIMP3 during cumulus cell-oocyte complex in vitro maturation
.
Mol Cell Endocrinol
2018
;
477
:
29
38
.
35.
Hu
J
,
Ni
S
,
Cao
Y
,
Zhang
T
,
Wu
T
,
Yin
X
, et al
.
The angiogenic effect of microRNA-21 targeting TIMP3 through the regulation of MMP2 and MMP9
.
PLoS One
2016
;
11
:
e0149537
.
36.
Song
B
,
Wang
C
,
Liu
J
,
Wang
X
,
Lv
L
,
Wei
L
, et al
.
MicroRNA-21 regulates breast cancer invasion partly by targeting tissue inhibitor of metalloproteinase 3 expression
.
J Exp Clin Cancer Res
2010
;
29
:
29
.
37.
Ando
J
,
Yamamoto
K
.
Effects of shear stress and stretch on endothelial function
.
Antioxid Redox Signal
2011
;
15
:
1389
403
.
38.
Lu
D
,
Kassab
GS
.
Role of shear stress and stretch in vascular mechanobiology
.
J R Soc Interface
2011
;
8
:
1379
85
.
39.
Hong
GS
,
Lee
B
,
Oh
U
.
Evidence for mechanosensitive channel activity of tentonin 3/TMEM150C
.
Neuron
2017
;
94
:
271
3
.
40.
Lu
HJ
,
Nguyen
TL
,
Hong
GS
,
Pak
S
,
Kim
H
,
Kim
H
, et al
.
Tentonin 3/TMEM150C senses blood pressure changes in the aortic arch
.
J Clin Invest
2020
;
130
:
3671
83
.
41.
Hida
K
,
Maishi
N
,
Annan
DA
,
Hida
Y
.
Contribution of tumor endothelial cells in cancer progression
.
Int J Mol Sci
2018
;
19
:
1272
.
42.
Maishi
N
,
Hida
K
.
Tumor endothelial cells accelerate tumor metastasis
.
Cancer Sci
2017
;
108
:
1921
6
.
43.
Li
X
,
Sun
X
,
Carmeliet
P
.
Hallmarks of endothelial cell metabolism in health and disease
.
Cell Metab
2019
;
30
:
414
33
.
44.
Mochizuki
S
,
Okada
Y
.
ADAMs in cancer cell proliferation and progression
.
Cancer Sci
2007
;
98
:
621
8
.
45.
Qi
JH
,
Ebrahem
Q
,
Moore
N
,
Murphy
G
,
Claesson-Welsh
L
,
Bond
M
, et al
.
A novel function for tissue inhibitor of metalloproteinases 3 (TIMP3): inhibition of angiogenesis by blockage of VEGF binding to VEGF receptor 2
.
Nat Med
2003
;
9
:
407
15
.
46.
Ostchega
Y
,
Fryar
CD
,
Nwankwo
T
,
Nguyen
DT
.
Hypertension prevalence among adults aged 18 and over: United States, 2017–2018
.
NCHS Data Brief
2020
:
1
8
.
47.
van Dorst
DCH
,
Dobbin
SJH
,
Neves
KB
,
Herrmann
J
,
Herrmann
SM
,
Versmissen
J
, et al
.
Hypertension and prohypertensive antineoplastic therapies in cancer patients
.
Circ Res
2021
;
128
:
1040
61
.
48.
Small
HY
,
Montezano
AC
,
Rios
FJ
,
Savoia
C
,
Touyz
RM
.
Hypertension due to antiangiogenic cancer therapy with vascular endothelial growth factor inhibitors: understanding and managing a new syndrome
.
Can J Cardiol
2014
;
30
:
534
43
.
49.
Basu
R
,
Lee
J
,
Morton
JS
,
Takawale
A
,
Fan
D
,
Kandalam
V
, et al
.
TIMP3 is the primary TIMP to regulate agonist-induced vascular remodeling and hypertension
.
Cardiovasc Res
2013
;
98
:
360
71
.
50.
Climie
RE
,
van Sloten
TT
,
Bruno
RM
,
Taddei
S
,
Empana
JP
,
Stehouwer
CDA
, et al
.
Macrovasculature and microvasculature at the crossroads between type 2 diabetes mellitus and hypertension
.
Hypertension
2019
;
73
:
1138
49
.