PI3K/AKT/mTOR pathway and its related molecules participate in PROK1 silence-induced anti-tumor effects on pancreatic cancer

2.1 Animals

Male BALB/c nude mice (Inbred strain of mouse, 6 weeks old) were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. (China). The mice were kept under specific-pathogen-free conditions and provided with enough food, water, and space.

1. Ethical approval: The research related to animal use has been complied with all the relevant national regulations and institutional policies for the care and use of animals, and was approved by the Ethics Committee of the General Hospital of Ningxia Medical University (approval number: KYLL-2022-0628).

2.2 shRNA-mediated knockdown of PROK1

The steps for the shRNA-mediated knockdown of PROK1 were the same as those followed in our previous study [8]. ShPROK1 was used to transfect PANC-1 cells to knock down PROK1 expression. The shRNA sequences for shPROK1 and the negative control (shCon) were designed and synthesized for the vector pYr-1.1-hU6-EGFP-Neo (Catalog No. #VRY0359, YRbio; Changsha, China). The shPROK1 sequence is as follows: 5′-CACCCTCCATGGACTTGAAGAACATCTCGAGATGTTCTTCAAGTCCATGGAGTTTTTTG-3′ (underlined: sequence target at PROK1). The negative control sequence is as follows: 5′-CACCCCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACCTTAGGTTTTTTG-3′ (underlined: sequence target at no gene). Before lentivirus packaging, the BsaI restriction enzyme (Dalian TAKARA Co. Ltd., China) was used to digest the shuttle plasmid, and then, T4 ligase was used to connect the pYr-1.1 fragments and anneal the double-stranded shRNA. The recombined lentivirus was used to infect the cells when cell confluence reached 80–90%. After 72 h, the supernatant was collected, centrifuged at 4°C, filtered with a filter (0.45 µm), and finally, the product was collected. Then, the cells were transferred to six-well plates and incubated at 37°C with 5% CO₂ for 24 h. In the presence of polybrene (6 µg/mL), the cells were incubated with shPROK1. A scramble sequence, designated as shCon, was ligated into the vector and served as a negative control. After 24 h, the infected cells were screened out using puromycin, and the cells were harvested to assess the knockdown of PROK1.

2.3 Establishing a nude mouse xenograft model bearing human PC cells

Male BALB/c nude mice (6 weeks old; n = 10) were randomly divided into two groups. The mice in one group were injected with 2 × 10⁶ control PANC-1 cells (shCon), and those in the other group were injected with PANC-1 cells transfected with sh-PROK1–1. The cell suspension (2 × 10⁶ cells/mL, 200 µL) containing shPROK1 in PBS was injected into the right armpits, and the PC tumors were allowed to grow in the mice for 49 days. The tumor volume was measured using calipers, and the weight of all animals was recorded throughout the experiment. The tumor volume was calculated using the formula: tumor volume = length (cm) × width (cm²) × 0.5236. On Day 49, the mice were euthanized. The tumor tissue was macroscopically observed after execution and prepared for histopathology and biochemical analysis.

2.4 Immunohistochemical (IHC) staining

Subcutaneous PC tumor masses were used for the IHC analysis to detect the expression of PROK1 in human PC xenograft in BALB/c nude mice. The prepared sections (4 µm) were dry-roasted and dehydrated in an oven at 65°C for 30 min and then were dewaxed in xylene, dehydrated in alcohol, and finally immersed in a solution tank containing double distilled water for 10 min. The sections were immersed in 3% H₂O₂ for 10 min, washed in distilled water thrice for 2 min, and immersed in 0.01 M tannic acid repair solution. Finally, they were heated in a microwave oven for 10 min at 92–98°C. The sections were placed in a glass jar containing 3% hydrogen peroxide and incubated at room temperature for 15 min to eliminate endogenous peroxidase activity. Normal sheep serum blocking solution was added dropwise to the sections, which were then incubated at 37°C for 30 min. Then, the excess liquid was removed from the sections. The primary antibody against PROK1 (Santa Cruz Biotechnology; sc-30343; dilution, 1:50) or PCNA (Cell Signaling Technology; #13110; dilution, 1:10,000) was added to the sections, which were then placed in a humidified space at 4°C overnight. The sections were taken out of the refrigerator, rewarmed at 37°C for 30 min, and washed thrice with PBS buffer for 5 min. The biotin-labeled secondary antibody was then added dropwise to the sections and incubated at 37°C for 30 min. The sections were stained using the Histostain-Plus Generation 3 IHC Assay Kit (Invitrogen, USA) following the manufacturer’s protocol. The positively stained cells were counted (for PCNA) or densitometrically analyzed (for PROK1) in five views under a microscope.

2.5 TUNEL (terminal deoxynucleotidyl transferase-mediated nick end labeling)

The paraffin sections to be dyed were baked in an oven at 60°C for 3 h, dewaxed with xylene, and rehydrated using gradient alcohol. After adding proteinase K, the sections were incubated at 37°C for 8 min, followed by the addition of 0.3% hydrogen peroxide in methanol and incubation for 10 min at room temperature to block endogenous peroxide enzymes. Then, 0.1% (v/v) Triton–sodium citrate was added to the sections for 10 min for cell penetration and antigen retrieval. The TUNEL reaction mixture solution was added and incubated at 37°C in a dark box for 1.5 h. Next, the transformant-POD was added and incubated in the wet box for 30 min at 37°C. After the DAB substrate solution was added, the samples were incubated at room temperature for 1–2 min. The sections were counterstained with hematoxylin for 30 s and differentiated with 1% hydrochloric acid alcohol for 5 s. Then, the sections were dehydrated with alcohol, treated with xylene, covered with a coverslip, and sealed with neutral gum. Finally, the sections were analyzed under a confocal laser scanning microscope.

2.6 Western blot analysis

The tumor tissue collected from the nude mice was ground to a granular shape with liquid nitrogen in a mortar. The tissue was placed in a 1.5 mL centrifuge tube. To 100–200 µL of the cell lysate, a pre-warmed 1× SDS gel-loading buffer was added as soon as possible. The sample was placed in a boiling water bath for 5 min. Then, the sample was centrifuged at 10,000 rpm at room temperature. The BCA Protein Assay Reagent (Millipore, Bedford; MA, USA) was used to quantify the extracted protein. The collected proteins (30 mg) were separated by 8% SDS-PAGE and transferred onto a PVDF membrane. After blocking the membrane with non-fat milk (5%) for 2 h, the primary antibodies against the following antigens were incubated with the membrane overnight at 4°C for 24 h: PROK1 (Santa Cruz Biotechnology; sc-30343; dilution, 1:300), PCNA (CST; #13110; dilution, 1:2,000), cyclin D1 (CST; #55506; dilution, 1:2,000), Bax (CST; #2772; dilution, 1:1,000), Bcl-2 (CST; #4223; dilution, 1:1,000), cleaved caspase-3 (CST; #9654; dilution, 1:1,000), PI3K (Abcam; ab154598; dilution, 1:2,000), phospho-PI3K [Tyr458] (Abcam; ab278545; dilution, 1:2,000), AKT (Abcam; ab8805; dilution, 1:500), phospho-AKT [Ser473] (Abcam; ab81283; dilution, 1:5,000), mTOR (Abcam; ab32028; dilution, 1:2,500), and phospho-mTOR [Ser2448] (Abcam; ab109268; dilution, 1:5,000). The samples were incubated with horseradish peroxidase-conjugated secondary antibodies at room temperature for 2 h. The signals were detected using an ECL system (Amersham Biosciences, USA). Densitometric analysis of the blots was performed using the ImageJ software.

2.7 Histological analysis

The samples were fixed with neutral formalin solution (10%) for 1 day and rinsed. Increasing concentrations of ethanol were used to dehydrate the samples. The tissues were embedded in paraffin and cut into 4 µm sections using a microtome. A graded descending ethanol series, including 100, 90, and 70% ethanol solutions (5 min each) and dH₂O (10 min), were used to deparaffinize and rehydrate the sections successively. Finally, the sections were stained with hematoxylin and eosin (H&E) using the conventional method.

2.8 Public data mining

We further determined the related molecules that regulate PROK1 or are regulated or affected by PROK1 using GEPIA 2 (http://gepia2.cancer-pku.cn/) in the next-generation sequencing databases TCGA (https://portal.gdc.cancer.gov/) and GTEx (https://www.gtexportal.orghome/). The expression level of PROK1 and its possible related genes (von Willebrand factor [VWF], GNG11, FGF7, TNXB, IL3RA, and LPAR1) in PAAD (pancreatic adenocarcinoma) and the data from paired adjacent tissue or normal pancreas tissue served as the control. The association between the expression level of the genes and the survival of the patients was analyzed. Initially, the possible related genes were screened out (criteria: Pearson’s correlation coefficient >0.3 and top 500 genes) from those that were significantly correlated with PROK1 in the PAAD dataset from the TCGA database using the tools “Similar Genes Detection” on GEPIA 2 and then were assessed by the KEGG enrichment analysis (WebGestalt, http://www.webgestalt.org/). We selected the genes enriched in the PI3K/AKT signaling pathway and overlapped them with all differentially expressed genes (DEGs, cutoff: |Log₂FC| = 1, p < 0.01) in PAAD. The protein levels of VWF, GNG11, FGF7, and IL3RA were obtained from THPA (https://www.proteinatlas.org/), and survival analysis was performed using the Kaplan–Meier plotter (https://kmplot.com/analysis/) or ToPP (http://www.biostatistics.online/topp/index.php) for univariate or multivariate factors.

2.9 Statistical analysis

Statistical analyses were performed using the GraphPad Prism software (version 8.0). All data are presented as mean ± SD (standard deviation) from at least three replicates. The differences among groups were determined by performing a one-way analysis of variance (ANOVA) followed by the least significant difference test. All differences among and between groups were considered to be statistically significant at P < 0.05.

3.1 PROK1 knockdown decreased the volume and weight of human PC xenografts in BALB/c nude mice

Tumor volume was measured using calipers every 6 days, after 35 days post-injection. The differences in the tumor volume between the groups increased with time (P < 0.01, Figure 1a). The size of the tumor in the sh-PROK1–1 group was smaller than that in the shCon group (Figure 1b). Tumor weight was recorded after harvesting the tissue. The tumor weight of the sh-PROK1–1 group was lesser than that of the shCon group (P < 0.01, Figure 1c). These results indicated that PROK1 knockdown significantly inhibited the growth of PC in BALB/c nude mice. Because PROK1 is an angiogenic inducer, vascular characteristic changes could be determined from the results of H&E staining. After histological evaluation of the tissue, we found no other noticeable morphological changes in the blood vessels, but the formation of blood vessels decreased, and more fluid and crowded cells were present in the shPROK1 group than in the shCon group (Figure S1).

Figure 1

PROK1 knockdown decreased the volume and weight of human PC xenograft in BALB/c nude mice. (a) Tumor volume was measured using calipers every week after injecting cancer cells. Ten mice in the shCon group were injected with 2 × 10⁶ control PANC-1 cells, and the mice in the other group were injected with PANC-1 cells transfected with sh-PROK1–1. (b and c) We measured the size of tumor tissue using calipers and the tumor weight; *P < 0.05 and **P < 0.01, compared to the shCon group.

3.2 PROK1 knockdown inhibited the PI3K/AKT/mTOR pathway and proliferation markers and increased apoptosis

The TUNEL micrographs of the tumor sliced from the sh-PROK1–1 group showed multiple brown-stained (TUNEL-positive, indicating apoptosis) cancer cells (Figure 2a). After PROK1 knockdown with shRNA, the proliferating marker PCNA-positive stained cells decreased significantly (Figure 2b). The results of the IHC analysis showed that PROK1 was weakly expressed in the shPROK1 group but strongly expressed in the shCon group (Figure 2c). This indicated that treatment with sh-PROK1–1 helped to maintain the reduced expression of PROK1 during tumor growth.

Figure 2

PROK1 silencing increased apoptosis and decreased the expression of PROK1. (a) The TUNEL micrograph of the sh-PROK1–1 group. Multiple brown-stained cancer cells were found in the sh-PROK1–1-treated cells, indicating apoptosis. (b) The cell proliferation marker PCNA was detected by the IHC analysis. (c) The PROK1 protein level in PC tissues was detected by the IHC analysis. Scale bar: 100 µm; *P < 0.05 and **P < 0.01, versus the shCon group.

The results of the Western blot analysis showed that the protein level of PCNA, cyclin D1, and Bcl-2 was lower in the shPROK1 group than that in the shCon group, while the protein level of Bax and cleaved caspase-3 in the shPROK1 group was higher than that in the other groups (Figure 3a and b). To determine whether the apoptosis of the human PC xenograft cells induced by PROK1 knockdown was related to the inactivation of the PI3K/AKT/mTOR signal, the level of the proteins associated with the pathway was measured. The results showed that the levels of p-PI3K, p-AKT, and p-mTOR were significantly reduced by the knockdown of PROK1 in the BALB/c mouse PC xenograft (P < 0.05 and P < 0.01 versus the shCon group), suggesting that the inhibition of the PI3K/AKT/mTOR pathway was induced by PROK1 knockdown (Figure 3c and d). To summarize, our results revealed that the inhibition of the phosphorylation of the PI3K/AKT/mTOR mediated by PROK1 knockdown induced cell apoptosis, which might be why PROK1 silencing promoted the human PC xenograft tissues to shrink and the cells in BALB/c nude mice to undergo apoptosis.

Figure 3

Effect of PROK1 knockdown on proteins involved in the PI3K/AKT/mTOR pathway, proliferation, and apoptosis. (a and b) The protein levels of p-PI3K, PI3K, p-AKT, AKT, mTOR, and p-mTOR in the BALB/c mouse PC xenograft were measured by performing Western blotting, and densitometric analysis was performed using the ImageJ software. (c) Western blotting was performed to measure the relative expression of the PCNA and cyclin D1 proteins (markers expressed only in proliferating cells), Bcl-2 (a marker of anti-apoptosis), and Bax and cleaved caspase-3 (markers of pro-apoptosis). (d) The optical density of each protein was analyzed; *P < 0.05 and **P < 0.01, compared to the shCon group.

3.3 The role of PROK1 in PC patient survival

The difference in the transcription of PROK1 in PAAD tissues and non-PAAD tissues was not significant (Figure 4a and b). Moreover, the expression level significantly increased with the increase in the tumor stage, suggesting a possible role in the progression of PAAD. The level of PROK1 and the overall survival (OS) or relapse-free survival (RFS) of PAAD patients were not associated (Figure 4c and d). Considering that the mutation of some genes like KRAS usually occurs in PAAD, we subgrouped the data with mutated or non-mutated KRAS and reanalyzed the relationship. In patients with mutated KRAS, higher PROK1 had longer RFS (Figure 4e). We also analyzed the relationship in stage 2 patients and found that survival probability (including OS and RFS) increased with the increase in the PROK1 level (Figure 4f and g), which was unexpected.

Figure 4

The possible role of PROK1 in the survival of PAAD patients. (a and b) Using public sequencing data in the TCGA and GTEx databases and the online tools GEPIA 2, ToPP, and Kaplan–Meier plotter, the expression level of PROK1 in PAAD was determined. (c–g) The relationship between the expression level of PROK1 and the overall or progression-free survival of the patients was determined.

3.4 The possible related molecules of PROK1 and their roles in regulating tumor growth in PC

The 500 genes correlated with PROK1 in PAAD (Pearson’s correlation, R > 0.3) were obtained from the TCGA database using GEPIA 2 and then submitted to WebGestalt for the KEGG pathway enrichment analysis. The results showed that they were mainly enriched in ovarian steroidogenesis, tryptophan metabolism, and malaria. However, several genes (including 11 genes: IGF1, NGFR, VWF, GNG11, NGF, FGF7, TNXB, IL3RA, LPAR1, TCL1B, and GH1) were associated with the activation of PI3K/AKT, which was interesting. These genes were overlapped with 9,170 DEGs in PAAD, which were obtained by comparing the mRNA expression profile between pancreatic cancer in TCGA and the normal pancreas in GTEx; six specific genes were found, including VWF, GNG11, FGF7, TNXB, IL3RA, and LPAR1 (Figure 5a and b). The expression of all six genes was significantly upregulated in PAAD compared to that in non-PAAD (Figure 5c). According to the data on THPA, most of these genes (VWF, GNG11, FGF7, and IL3RA) were also higher in PAAD tissues at the protein level (compared to non-tumor tissues, Figure 5d). The results of the univariate analysis for the association between the interested gene and the survival of the patients showed no notable relationship between them irrespective of whether KRAS was mutated (Figure S2). However, the results of the multivariate analysis showed that the increase in the transcription of these genes was associated with a decrease in the OS of the PAAD patients (Figure 5e).

Figure 5

Screening of genes related to PROK1 and their expression and association with survival. (a) The correlated genes from GEPIA 2 were submitted to WebGestalt to perform gene enrichment. (b) The genes involved in PI3K/AKT signaling were overlapped with DEGs. After the intended genes were identified, their mRNA level, protein level, and possible association with survival were analyzed using (c) GEPIA 2, (d) THPA, and (e) ToPP.