We used the Gene Expression Omnibus (GEO), a publicly accessible genomics database maintained by the The National Center for Biotechnology Information (NCBI), and the The Cancer Genome Atlas (TCGA) data portal (https://tcga-data.nci.nih.gov/tcga/) to investigate the expression and prognostic significance of VGF in PCa.

We used the UCSC Xena platform (http://xena.ucsc.edu/) and the Gliovis database (http://gliovis.bioinfo.cnio.es/) to investigate the expression and prognostic importance of FTL in glioma. The Cancer Genome Atlas (TCGA), Rembrandt, and the Ivy Prostate Cancer Atlas Project (IVY) datasets normalized RSEM gene-level RNA-seq and related clinical data were retrieved from Gliovis. We obtained specific data on postoperative chemotherapy and radiation therapies received by glioma patients from the UCSC Xena platform. Clinical information and normalized mRNA expression [mRNA-array_693, (batch 1)] were also collected from the Chinese Glioma Genome Atlas. The 2021 World Health Organization (WHO) classification of malignancies of the central nervous system classified low-grade glioma as WHO grade I–II and high-grade glioma as WHO grade III-IV [31]. We used JASPAR (https://jaspar.elixir.no/) site predictions to explore the specific mechanisms of action of VGF and HIF-1$\alpha$.

Between 2022 and 2024, tissue samples were taken from 42 patients with PCa and 21 individuals who had benign prostatic hyperplasia (BPH) at the First Affiliated Hospital of Chongqing Medical University (China). Identities and conditions were anonymized to maintain the integrity of the double-blind review process. In the inpatient department, 37 patients had BPH, and 53 patients had PCa at the same time as plasma samples were taken. Prior to the experiments, specimens were stored in liquid nitrogen to preserve their integrity. All clinical samples were ethically approved by the First Affiliated Hospital of Chongqing Medical University (Approval No. 2022-K275) and conducted in accordance with the Declaration of Helsinki. The patients or legal guardians gave written approval for the biological studies. Every specimen underwent a histological examination and was categorized based on WHO standards and UICC guidelines [32].

IHC labeling was used to determine the VGF expression levels in tissue samples from 42 PCa and 21 BPH patients. After being deparaffinized in xylene and rehydrated in distilled water and graded alcohol, tissue pieces were pressure-cooked to extract antigens. Then, they were blocked with endogenous peroxidase for 5 min and then incubated at room temperature for 30 min with goat serum. Sections were then incubated with primary antibody for 14 h at 4 °C. VGF nerve growth factor inducible (VGF) (1:300, NBP2-31596, Novus, St. Louis, MO, USA) antibody. Vimentin (1:100, WL01960, Wanlei Bio, Shenyang, Liaoning Province, China) and E-cadherin (1:100, WL00941, Wanlei Bio, Shenyang, Liaoning Province, China). HIF-1$\alpha$ (1:50, A11945, ABclonal, Wuhan, China). The SP Immunohistochemistry Commercial Detection Kit (SP-9000, Beijing Zhongshan Jinqiao Biotechnology Co., Ltd., Beijing, China) was used to perform immunostaining in accordance with the instructions. Following that, they were incubated for 15 min in a working solution of streptavidin/peroxidase complex and then kept for 30 minutes at 37 °C with the kit’s biotinylated secondary antibody (1:200, SP-9000; Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.). A diaminobenzidine (DAB) kit (ZLI9018, Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) was used for the peroxidase staining process. The specimens were stained with hematoxylin staining solution (G1080, Solarbio, Beijing, China) at room temperature for 1 min. Semiquantitative scores were used to calculate staining intensity and immunoreactivity rates. Those classified as negative/weak were regarded as staining negatively for statistical reasons, whereas those with moderate or strong staining were regarded as positive.

We quantified plasma VGF levels in patients with prostate disease using an ELISA kit (JM5286H1, Jiangsu Jingmei Biotechnology Co., Ltd., Yancheng, China). Briefly, peripheral blood samples (EDTA-K2 anticoagulant) were collected from all patients and then stored at –80 °C until use. In line with the advice provided by the manufacturer, we used a custom-made ELISA kit (JM5286H1, Jiangsu Jingmei Biotechnology Co., Ltd., Yancheng, China) to quantify plasma VGF. The kit was equilibrated to room temperature; the solution was mixed thoroughly to avoid foaming, and the standard was reconstituted with the standard diluent and mixed thoroughly before use. The sample was then diluted 5-fold with the diluent. Next, 50 µL of standard or diluted sample was added to each well and mixed gently by shaking. The plate was covered with a membrane (microplate sealant) and incubated at 37 °C for 30 min. The liquid was discarded, the plate was washed, and the procedure was repeated for six washes. Next, 50 µL of Horseradish peroxidase (HRP) coupling reagent (JM5286H1, Jiangsu Jingmei Biotechnology Co., Ltd.) was added to each well and mixed gently by shaking. The plate was covered with a membrane and incubated at 37 °C for 30 min. After discarding the liquid, the plate was cleaned, and the process was carried out six times. Subsequently, each well was filled with 50 µL of chromogenic solution A and 50 µL of chromogenic solution B and shaken slightly to combine, then allowed to sit at 37 °C (in the dark) for 10 min. After stopping the reaction with 50 µL of stop solution in each well, the optical density was measured at 450 nm for the detection wavelength and 620 nm for the reference wavelength, using a microplate reader (Thermo Varioskan Flash, Thermo Fisher Scientific (China) Co., Ltd., Shanghai, China). A standard curve of optical density measurements was used to determine the amount of VGF protein in each sample.

American Type Culture Collection (ATCC, Manassas, VA, USA) provided the human normal prostate epithelial cells RWPE-1 and the human PCa cells PC3, 22Rv1, and C4-2. All cells were short tandem repeat (STR)-validated to maintain cell line stability. Mycoplasma-free cells were used for all studies. Thermo Fisher Scientific (Boston, Mass., USA) provided 10% FBS, and Beyotime Institute of Biotechnology (Shanghai, China) provided 100 U/mL penicillin/streptomycin as supplements for the RPMI-1640 (Gibco, Thermo Fisher Scientific, MA, USA) medium in which PCa cells were cultivated. The specified keratinocyte-SFM (1$\times$) fluid (Invitrogen, Carlsbad, CA, USA) was used to sustain RWPE-1 cells. Every cell line was cultivated in a humidified environment with 5% CO₂ at 37 °C. Cells were cultivated under constant 1% O hypoxia conditions for in vitro hypoxia investigations. The lentiviral VGF vector and lentiviral HIF-1$\alpha$ vector were constructed by Beijing Qingke Biotechnology Co., Ltd. The target sequence of VGF short hairpin RNA (shRNA) is 5^′-CTTCTGCCTTCTGCTGATCAA-3^′, and the target sequence of HIF-1$\alpha$ shRNA is 5^′-GCCGCTCAATTTATGAATATT-3^′. Lentivirus-infected 22Rv1 cells were selected with 1 µg/mL puromycin to establish control sh-NC 22Rv1, sh-VGF 22Rv1, and sh-HIF-1$\alpha$ 22Rv1 cells. Overexpression plasmids for VGF and HIF-1$\alpha$ were constructed by Hanheng Biotechnology Co., Ltd., and according to the manufacturer’s instructions, cells were cultured in 6-well plates and transfected using lip2000 (Invitrogen, Carlsbad, CA, USA). For the subsequent investigation, cells were used 48 h after transfection. Akt activator SC79 (HY-18749, MedChemExpress (MCE), Monmouth Junction, NJ, USA) and docetaxel (HY-B0011, MCE) (Transwell: add to serum medium when inoculating cells; wound healing assays: add to serum-free medium after the cells were injured with a 200 µL pipette tip; WB: serum medium was added after cell attachment).

After being lysed for around 30 min on ice in RIPA Lysis Buffer (P0013, Beyotime Biotechnology, Shanghai, China), cell samples were centrifuged for 10 min at 12,000 rpm. For each 100 mg tissue, we added 1 mL of RIPA lysate. Using the BCA Assay Kit (Cat. No. P0010, Beyotime Biotechnology), the concentration of proteins was measured. Before loading, protein samples were added to the loading buffer (P0015, Beyotime Biotechnology) and cooked for 4 min in boiling water. In short, the protein was loaded in equal proportions onto 8–12% SDS-PAGE and then transferred to PVDF membranes (Millipore, MA, USA). Then, it was blocked in 5% skim milk for 1 h and incubated with the primary antibody overnight at 4 °C. Membranes were incubated for 1 h at room temperature using a secondary antibody. Protein bands were identified by using an improved chemiluminescence reagent (Cat. No. P10100, Suzhou New Saimei Biotechnology Co., Ltd.). The primary antibodies were as follows: VGF (1:1000; NBP2-31596; Novus); HIF-1$\alpha$ (1:1000; 36169; CST); AKT (1:1000; 9272; CST); p-AKT (1:2000; 4060; CST); PI3K (1:1000; 4292; CST); p-PI3K (1: 1000; 4228; CST); E-cadherin (1:2000; 20874; Proteintech); vimentin (1:1000; 10366; Proteintech); snail 1 (1:3000; WL01863; Wanleibio); and $\beta$-actin (1:2000; 20536; Proteintech); $\beta$-tubulin (1:2000; 11224; Proteintech). The secondary antibodies are as follows: HRP-conjugated Goat Anti-Mouse IgG (H+L) (1:5000; SA00001-1; Proteintech); HRP-conjugated Goat Anti-Rabbit IgG (H+L) (1:5000; SA00001-2; Proteintech). ImageJ software (1.53j, LOCI, University of Wisconsin, Madison, WI, USA) was used to analyze protein expression.

TRIzol (TaKaRa, Tokyo, Japan) was used to extract total RNA from cell samples, and the Prime Script RT kit (TaKaRa) was used to reverse-transcribe the RNA according to the manufacturer’s protocol. PCR was performed using the SYBR Premix Ex Taq™ II Kit (TaKaRa). The 2^{-Δ⁢Δ⁢CT} technique was used to quantify gene expression. Three replicates of the real-time PCR were run, and the outcomes were adjusted for $\beta$-actin. The following are the primer sequences: VGF: (sense: 5^′-GGAACTGCGAGATTTCAGTCC-3^′, antisense: 5^′-GTGCGGGTTCCGTCTCTG-3^′); HIF-1$\alpha$: (sense: 5^′-TGATGACCAGCAACTTGAGG-3^′, antisense: 5^′-CTGGGGCATGGTAAAAGAAA-3^′); $\beta$-actin: (sense: 5^′-GGGACCTGACTGACTACCTC-3^′, antisense: 5^′-ACGAGACCACCTTCAACTCCAC-3^′).

Cells (2 $\times$ 10⁴) were seeded with mixed Matrigel-coated membranes in the upper chamber for Transwell invasion experiments, and cultured in media devoid of fetal bovine serum (FBS). Medium supplemented with 15% FBS was added to the bottom chamber. After a 24-h incubation period, invasive cells were stained with crystal violet, and five visible regions were systematically counted under a microscope at a magnification of 100$\times$ (Nikon, Tokyo, Japan). In 6-well plates, cells were sown for wound healing experiments. Once the cells reached 80% confluence, they were injured with a 200 µL pipette tip and continued incubating for 24–48 h. At designated intervals, the wound healing process was examined under a microscope.

In order to study the effect of the interaction between HIF1A and FTL, plasmids of wild-types WT1 and WT2, and mutant types Mut1 and Mut2 containing VGF promoter sequences, were constructed (GS1-21120205; Wuhan Jinkairui Bioengineering Co; WT1: GGTACCGAGCTGATGGGCTTTCTTCTGGGAAAGTCGAGCCACTGATGGAAGCGAGAAGCCACTGCTGGTTATAGAGAGAAAGCACGTGAGTGTGTGTGTAGGGAGGGGGAGGTTAGAAGGAGGGTCAGTGCCAGGAAGAGGTGAGGAGGGGGGCGACTCGAG; WT2: GGTACCCCCCTGTCAGGGGGCTGCCACCCGCACTGCCGATTCGCGGACAGCGCCCGCAGGCGTGCAGATCTGTCCCTCTGCACTCAGGTTCACGCCGTCCTTGGGCGCGTGGTCTCGGGGTGGGGAACCCGGCCCCCTGGTCGGCTCTTGAATCTTCTCGAG; Mut1: GGTACCGAGCTGATGGGCTTTCTTCTGGGAAAGTCGAGCCACTGATGGAAGCGAGAAGCCACTGCTGGTTATAGAGAGAAAACTCACTCGCGTGTGTGTAGGGAGGGGGAGGTTAGAAGGAGGGTCAGTGCCAGGAAGAGGTGAGGAGGGGGGCGACTCGAG; Mut2: GGTACCCCCCTGTCAGGGGGCTGCCACCCGCACTGCCGATTCGCGGACAGCGCCCGCGCACGCCTAGATCTGTCCCTCTGCACTCAGGTTCACGCCGTCCTTGGGCGCGTGGTCTCGGGGTGGGGAACCCGGCCCCCTGGTCGGCTCTTGAATCTTCTCGAG). HEK293T cells were cotransfected with sh-HIF-1$\alpha$ plasmid (or sh-NC plasmid) and WT (or Mut) plasmid and a plasmid containing Renilla luciferase gene. Lipofectamine 2000 (Invitrogen) was used for the transfection process. After 48 h, the luciferase activity was measured using a dual-luciferase reporter system (Promega, Madison, WI). The ratio of firefly to Renilla luciferase activities is known as the relative luciferase activity.

ChIP was carried out to check for possible binding between HIF1A and the VGF promoter region. For 24 h, 22Rv1 cells were cultivated under hypoxia. Cell Signaling Technologies provided the HIF1A antibody (Cat. No. 36169). The VGF fragment-containing precipitated DNA was subsequently amplified by quantitative PCR. The following primer sequences were used to find the HREs in the VGF promoter: sense, 5^′-ACTGATGGAAGCGAGAAG-3^′, and antisense, 5^′-TCCTGGCACTGACCCT-3^′.

The supplier of the nude mice was Chongqing Ensiweier Biotechnology Co., Ltd. The experiment Animal Ethics Committee’s requirements (Approval No. 2022-K275; the First Affiliated Hospital of Chongqing Medical University) were followed during the care of the mice and the experimental methods. Log phase-grown, stable transfected 22Rv1 cells were produced. The cells were reconstituted in PBS at a density of 5 $\times$ 10⁶ cells/100 µL. For the in vivo administration of docetaxel therapy, twenty 5-week-old male nude mice, each weighing approximately 15 grams, were randomly divided into four groups, with five mice per group. In the first and second groups, the sh-NC stabilized cell line was injected into the tail vein of the nude mice. Groups III and IV were injected with the sh-VGF-stabilized cell lines into the tail vein of the nude mice. Live imaging for metastasis was conducted four weeks post injection. Nude mice in groups I and III were administered saline (10 mg/kg, once per week) via tail vein injection for three consecutive weeks; nude mice in groups II and IV were injected with docetaxel (10 mg/kg, 1/week) via tail vein for three consecutive weeks. Following tail vein injection, the health and condition of the mice were observed daily [33, 34]. After the third week of docetaxel treatment, all nude mice were imaged in vivo (IVIS Lumina LT) and then euthanized with 2% pentobarbital (100 mg/kg; IV) plus a high concentration of CO₂. Liver and lung tissue was obtained for sectioning and analysis.

Each experiment in this investigation was carried out independently at least three times, and SPSS17.0 (Version 17.0.1, SPSS Software, IBM., Armonk, NY, USA) was used to evaluate the data statistically. All data were given as means $\pm$ standard deviation (SD). T-tests or ANOVA were utilized to conduct statistical comparisons between groups; Bonferroni correction was used for post hoc comparisons, and the hypothesis of normality was tested separately for each group of samples using the Kolmogorov–Smirnov (KS test). The correlation of VGF expression and clinicopathologic factors in PCa patients was assessed using Fisher precision testing and the paired Chi-square test. The fold enrichment method was used for statistical data analysis of ChIP. p 0.05 was deemed to be a significant difference.

While specific human cancer tissues have been found to overexpress VGF [16, 17], the functional involvement of VGF in PCa remains poorly understood. Using the TCGA database and bioinformatic analysis, we examined the expression of VGF and its potential prognostic significance in PCa.

VGF levels in PCa tissue were significantly higher than in normal prostate tissue (Fig. 1A, p 0.001). The analysis from GEO datasets (GSE 32571) also showed the same results (Fig. 1B, p 0.01). Moreover, in the GSE 8511 and GSE 32269 datasets, we found that PCa tissue with metastasis has a higher VGF expression than localized PCa tissues (Fig. 1C,D, p 0.001, p 0.05). Then, using open datasets, we investigated the predictive significance of VGF in PCa. We found that PCa patients in the TCGA datasets who expressed more VGF had worse disease-free survival and overall survival than those who expressed less VGF (Fig. 1E,F, p 0.01).

Fig. 1.

In prostate cancer (PCa), VGF nerve growth factor inducible (VGF) is overexpressed and related to prognosis. (A) VGF expression in the cancer genome atlas (TCGA) datasets for both PCa and normal prostate. (B) VGF expression in benign prostatic hyperplasia (BPH) and PCa in Gene Expression Omnibus (GEO) datasets. (C) VGF expression in PCa and mPCa in GEO datasets. (D) VGF expression in PCa and mCRPC in GEO datasets. (E) VGF expression survival study across all PCa patients in TCGA. (F) Study of disease-free survival for each PCa patient in TCGA according to VGF expression. (G) Magnification shows representative immunohistochemistry (IHC) and hematoxylin and eosin (H&E) staining for VGF in PCa and BPH samples. Scale bar: 100 µm. (H) The degree of molecular expression in newly obtained PCa tissues and BPH specimens. (Supplementary Material, Fig. S1A). (I) The expression of VGF in the plasma of BPH patients and PCa patients. (J) Receiver operating characteristic (ROC) curve plot for prediction of PCa by VGF. *p 0.05; **p 0.01; ***p 0.001.

The results demonstrated that VGF expression was positive in 83.33% (35/42) of PCa samples, compared to 19.05% (4/21) in the BPH samples (Fig. 1G, p 0.01).

Next, the relationship between VGF expression and the clinicopathological characteristics of PCa patients was examined. These results showed a significant relationship between the histological stage, Gleason score, and VGF expression (Table 1). The VGF nerve growth factor inducible (VGF) protein levels in 7 fresh PCa and 5 fresh BPH surgical tissue specimens were found to be considerably greater in the PCa tissue than the BPH tissues (Fig. 1H, p 0.001). The relative expression of VGF in plasma samples from 53 PCa patients and 37 BPH patients was assessed using ELISA to better assess the function of VGF in PCa. The PCa patients showed higher expression of VGF compared to the BPH patients (Fig. 1I, p 0.01). The current study further investigated the potential of VGF as a diagnostic biomarker for PCa through the analysis of the receiver operating characteristic (ROC) curve. With a 95% confidence interval [CI] = 0.812–0.969 and an area under the curve (AUC) of 0.890 (Fig. 1J, p 0.01), it appears that VGF may be a useful tool for PCa diagnosis. When combined, our results suggested that VGF may function as a prognostic biomarker and be connected to PCa development.

Consistent with the observation that VGF expression is increased in human PCa tissue, PCa cell lines (C4-2, PC3, and 22Rv1) exhibited significantly higher levels of VGF expression in both mRNA and protein levels when compared to the normal prostate epithelial cell line RWPE-1 (Fig. 2A,B, p 0.05).

Fig. 2.

VGF knockdown suppresses PCa metastasis. (A) Expression of VGF in RWPE-1, C4-2, 22Rv1, and PC3 cells was detected by quantitative real-time PCR (RT-qPCR). (B) Expression of VGF in RWPE-1, C4-2, 22Rv1, and PC3 cells was detected by western blot. (Supplementary Material, Fig. S1B). (C) Expression of VGF in blank, sh-negative control (sh-NC), and sh-VGF cells was detected by RT-qPCR. (D) Expression of VGF in blank, shNC, and sh-VGF cells was detected using western blot. (Supplementary Material, Fig. S1C). (E) Expression of VGF in blank, oe-NC, and oe-VGF cells was detected by RT-qPCR. (F) Expression of VGF in blank, oe-NC, and oe-VGF cells was detected by western blot. (Supplementary Material, Fig. S1D). (G,H) The wound-healing test and Transwell assays were used to assess the effects of sh-VGF on 22Rv1 cell migration and invasion. Scale bar: 200 µm. (I) Effects of sh-VGF on the expression of migration-related genes. (Supplementary Material, Fig. S1E). (J) Effects of oe-VGF on the expression of migration-related genes. (Supplementary Material, Fig. S1F). ns, not significant; *p 0.05; **p 0.01; ***p 0.001; ****p 0.0001.

Lentiviral vectors expressing VGF shRNA (sh-VGF) or control shRNA (sh-NC) were created to knock down the expression of VGF in 22Rv1 cells in order to ascertain if VGF plays a significant role in PCa development (Fig. 2C,D, p 0.0001, p 0.001). Simultaneously, overexpressed VGF (oe-VGF) and a control plasmid (oe-NC) were created to boost VGF expression in C4-2 cells (Fig. 2E,F, p 0.0001, p 0.001). To investigate whether sh-VGF contributes to PCa formation, 22Rv1 cell invasion and migration were investigated using Transwell and wound-healing assays. The capacity for invasion and migration of 22Rv1 cells was diminished with VGF knockdown (Fig. 2G,H, p 0.001). Consistent with the above results, western blot investigations indicated that knocking down VGF affected the amounts at which proteins linked to invasion and metastasis are expressed. In 22Rv1 cells, there was an increase in E-cadherin expression and a decrease in vimentin and snail levels (Fig. 2I, p 0.05). In addition, opposing results were observed in the C4-2 cells overexpressing VGF (Fig. 2J, p 0.01).

VGF has been reported to enhance PI3K kinase activity in human pancreatic neuroendocrine neoplasms [8]. Therefore, we tested the influence of VGF on activation of the PI3K/Akt signaling pathway by western blot. Western blot showed that phospho-PI3K p85 (Tyr458) and phospho-Akt (Ser473) were greatly inhibited in 22Rv1 cells treated with sh-VGF (Fig. 3A, p 0.001, p 0.05). In contrast, the total amount of PI3K and Akt protein remained unchanged (Fig. 3A). These results suggested that VGF affected the PI3K/Akt pathway. To further estimate the effect of the PI3K pathway on VGF knockdown cells, 22Rv1 cells were treated with the Akt activator SC79 (4 µg/mL). Western blot confirmed that phosphorylation levels of PI3K and Akt were restored by the addition of SC79 in VGF knockdown cells (Fig. 3A, p 0.01).

Fig. 3.

PI3K/Akt signaling is involved in downstream events of VGF activation. (A) Effects of SC79 and VGF together on PI3K/Akt signaling expression. (Supplementary Material, Fig. S1G). (B) Effects of SC79 and VGF together on the expression of genes linked to migration. (Supplementary Material, Fig. S1H). (C–F) Changes in 22Rv1 cell migration and invasion after SC79 sh-VGF administration were identified using Transwell assays and the wound-healing test. ns, not significant; scale bar: 200 µm; *p 0.05; **p 0.01; ***p 0.001; ns, not significant.

Next, we examined the role of the PI3K/Akt pathway in 22Rv1 cells transfected with sh-NC or sh-VGF using the wound-healing test, Transwell assay, and western blot analysis. As expected, the capacity of VGF knockdown 22Rv1 cells for invasion and migration was restored by SC79 treatment (Fig. 3B–F, p 0.01). These findings suggest that VGF influences PI3K/Akt signaling, which in turn affects PCa cell invasion and metastasis.

Docetaxel is a commonly used chemotherapy drug for mPCa, and long-term use will lead to resistance in most patients [35, 36, 37]. In clinical practice, docetaxel is often used in combination with targeted drugs [38, 39]. We assessed whether targeted knockdown of VGF strengthens docetaxel’s inhibitory effect on PCa cell metastasis. Compared with the single-treatment group, the results of wound-healing assays and Transwell assays indicated that the combined-treatment group with targeted knockdown of VGF and docetaxel showed more significant inhibition of the invasion and metastasis of PCa cells (Fig. 4A–D, p 0.01).

Fig. 4.

sh-VGF combined with docetaxel can further inhibit PCa metastasis in vitro. (A,B) Changes in 22Rv1 cell migration after docetaxel + sh-VGF administration were identified using a wound-healing test (scale bar: 200 µm). (C,D) Changes in 22Rv1 cell invasion under docetaxel + sh-VGF treatment were determined by Transwell assay (scale bar: 200 µm). *p 0.05; **p 0.01; ***p 0.001; ****p 0.0001.

To construct a tumor-metastasis model, 22Rv1 cells were injected into the tail vein of male nude mice in order to study the impact of VGF on PCa metastasis in vivo. Four weeks later, the nude mice were randomized to receive either saline or docetaxel tail vein injections (Fig. 5A) [40].

Fig. 5.

sh-VGF combined with docetaxel can further inhibit PCa metastasis in vivo. (A) Construction of tail vein transfer 324 model in nude mice. (B) In vivo tumor imaging. (C,D) H&E staining suggests the extent of cancer cell metastasis in the 325 lungs of nude mice (scale bar: 200 µm; scale bar: 50 µm). *p 0.05; **p 0.01; ***p 0.001.

As shown in Fig. 5, relative to the sh-NC group, in vivo imaging of both the sh-VGF and docetaxel groups revealed that the degree of metastasis in nude mice was less, and the lung tissue stained with H&E showed a fewer number of malignant nodules. The degree of metastases in the combination therapy group was much less than in the single-factor treatment group (Fig. 5B–D, p 0.05). The results indicated that VGF knockdown could limit PCa metastasis, and the inhibitory impact is higher when paired with docetaxel [33, 34].

A well-established feature of clinical PCa associated with poorer prognosis is hypoxia [41, 42]. To explore whether hypoxia can affect the expression of VGF, we placed 22Rv1 cells and C4-2 cells under hypoxic conditions [43]. The development of hypoxia in 22Rv1 and C4-2 cells was observed to cause an increase in the amounts of VGF mRNA and protein, as well as an increase in the protein level of hypoxia-inducible factor-1alpha (HIF-1$\alpha$), according to findings from the western blot and q-PCR (Fig. 6A–C, p 0.05).

Fig. 6.

Hypoxia-induced VGF in an HIF-1$\alpha$-dependent manner. (A) Variations in mRNA levels of VGF in 22Rv1 cells and C4-2 cells under 1% and 20% oxygen environments. (B,C) Variations in protein levels of VGF in 22Rv1 cells and C42 cells under 1% and 20% oxygen environments. (Supplementary Material, Fig. S1I-J). (D) Graphs showing the putative hypoxia response elements (HREs) and the mutant VGF promoter in the same region. (E,F) Fluorescein levels after co-transfection of WT1 or WT2 plasmids with Renilla luciferase reporter plasmids. (G) A sample of HIF1A’s ChIP binding to the VGF promoter at 1% and 20% oxygen. ns, not significant; *p 0.05; **p 0.01; ***p 0.001.

We also constructed a plasmid to knock down HIF-1$\alpha$. After the targeted suppression of HIF-1$\alpha$ in PCa cells, the amounts of VGF protein and mRNA decreased. In order to explore the specific mechanism of action of VGF and HIF-1$\alpha$, we found two hypoxia response elements on the VGF promoter using JASPAR website prediction (Fig. 6D). We used luciferase and ChIP experiments to determine whether HIF-1$\alpha$ directly binds to the VGF promoter. We found, using a dual luciferase reporter experiment, that WT1’s luciferase activity was significantly higher in hypoxia than in a normoxic environment, whereas this phenomenon was not observed in WT2 (Fig. 6E, p 0.001), indicating that the transcription factor HIF-1 binds primarily through the HRE1 site on the VGF promoter. Additionally, we found that knocking down HIF-1 decreased the luciferase activity of WT1 (Fig. 6F, p 0.001). By using a chromatin immunoprecipitation test, we demonstrated that HIF1 attaches to an area of the VGF promoter (Fig. 6G, p 0.05). Consequently, we found that HIF-1 controls VGF transcription as a transcription factor. Our findings indicate that, as a transcription factor, HIF-1 could bind to promoter regions on the VGF gene and regulate the transcription of VGF.