Activated Pancreatic Stellate Cells Enhance the Warburg Effect to Cause the Malignant Development in Chronic Pancreatitis


 Background: Chronic pancreatitis (CP) is a precancerous condition associated with pancreatic ductal adenocarcinoma (PDAC), but its evolutionary mechanism is unclear. pancreatic stellate cells (PSCs) are closely related to the occurrence and development of CP and PDAC. We aimed to find out whether PSCs play a key role in this " inflammationcancer transition ". Methods: To evaluate the effect of activated pancreatic stellate cells on normal pancreatic duct epithelial cells and pancreatic cancer cells, pancreatic stellate cells isolated from human tissues were co-cultured with these two cells, respectively. Functional assays assessed the proliferation, migration, and invasion of these two cells. RT-qPCR and western blotting were used to detect the mRNA and protein expressions of glycolytic enzymes in these two cells. Lactate production and glucose utilization assays assessed the aerobic glycolysis level of these two cells. Immunohistochemistry was used to detect the expression of glycolytic enzymes and α-SMA, and the correlation between the two was analyzed in human tissues. Results: Our research found that co-culture with activated PSCs promoted the proliferation, migration and invasion of normal pancreatic duct epithelial cells and pancreatic cancer cells. At the same time, activated PSCs had a significant effect on the expression of the glycolytic enzymes PKM2 and LDHA in normal pancreatic duct epithelial cells and pancreatic cancer cells and increased lactic acid production and glucose consumption in these two cells. In vivo experiments showed that the expression of the glycolytic enzymes PKM2 and LDHA in pancreatic duct epithelial cells and the marker protein (α-SMA) of activated PSCs in the pancreatic duct peripancreatic interstitium were higher in pancreatic cancer tissues and chronic pancreatitis tissues than in normal pancreatic tissues in both animals and humans. In addition, analysis of human tissue specimens showed that there is a correlation between the expression of PKM2/LDHA and α-SMA. Conclusion: These findings indicate that activated PSCs play an important role in the development and progression of chronic pancreatitis into pancreatic cancer by regulating and promoting aerobic glycolysis. Our research provides a new theoretical basis for further understanding the mechanism of CP malignancy and the selection of targets for reversing CP malignancy.


Background
Pancreatic cancer is a highly malignant tumour with a poor prognosis. Its mortality ranked seventh among all malignant tumours in 2020, and its incidence is increasing annually. [1] Early pancreatic cancer lacks specific clinical symptoms. Nearly 85% of patients have lost the chance for radical surgical resection by the time they consult a doctor, and their 5-year survival rate is less than 6%. Even after radical surgery, the 5-year survival rate of patients is less than 25%. [2] Understanding the mechanism of pancreatic cancer will help to guide research on the prevention, early diagnosis and treatment of pancreatic cancer. The literature points out that pancreatic cancer is related to smoking, drinking, obesity, age, chronic pancreatitis, etc. [3] Among these factors, chronic pancreatitis is considered a precancerous condition associated with pancreatic cancer. Chronic pancreatitis is an inflammatory process in which the endocrine and exocrine glands of the pancreas are progressively and irreversibly damaged and the pancreatic parenchyma is replaced by fibrous tissue, which ultimately leads to various morphological and functional changes in the pancreas and triggers a series of clinical signs of inflammation. [4] Clinical studies have shown that patients with chronic pancreatitis have a much higher risk of pancreatic cancer than normal people. [5,6] Therefore, exploring the mechanism of pancreatic cancer transformation from chronic pancreatitis and reversing the progression of chronic pancreatitis are important issues that urgently need to be solved.
PSCs are pancreatic-specific stromal cells that are considered to be the targets of various cytokines in chronic pancreatitis and the core of the progression of chronic pancreatitis. Various physical and chemical factors stimulate the conversion of quiescent PSCs into a myofibroblast phenotype characterized by the expression of α-smooth muscle actin (α-SMA), by which they become activated PSCs, which synthesize a large amount of extracellular matrix (ECM) and secrete various cytokines to promote irreversible pancreatic fibrosis in chronic pancreatitis. [7,8] Not only are PSCs the initiating factors of chronic pancreatitis, a large number of reports also indicate that they play an important role in the occurrence, development and metastasis of pancreatic cancer. [9][10][11] However, whether PSCs play an important role in the malignant transformation of chronic pancreatitis to PDAC has not been reported. To maintain their abnormal proliferation ability and enhanced invasion characteristics, tumour cells reprogram nutrient transporters and glucose metabolism enzymes to maintain a high level of metabolism.
This metabolic characteristic of cancer cells is manifested by a greatly increased rate of glucose uptake and very active glycolysis even in the case of sufficient oxygen supply. This phenomenon is called aerobic glycolysis (the Warburg effect). [12] Currently, some studies have confirmed that PSCs promote glycolytic metabolism in pancreatic cancer, [13,14] but whether PSCs can promote the malignant transformation of CP by affecting the Warburg effect has not been reported.
In this study, we first isolated and cultured PSCs from human pancreatic cancer and normal pancreatic tissues and then performed phenotypic identification. We successfully obtained activated PSCs (derived from PDAC tissue, CaPSCs) and quiescent PSCs (derived from normal pancreas tissue, NaPSCs). Subsequently, we co-cultured PSCs from different sources with pancreatic cancer cells/normal pancreatic duct epithelial cells and found that co-culture promoted the proliferation, migration and invasion of these two cell types. Finally, we tested the expression of glycoproteins and glucose metabolism levels related to the Warburg effect in pancreatic cancer cells/normal pancreatic duct epithelial cells and further confirmed these experimental results in vivo. Our findings suggest that activated PSCs can enhance the Warburg effect to promote the malignant transformation of CP. The specific mechanism needs further research.

Patients and tissue samples
Primary pancreatic ductal adenocarcinoma tissue, chronic pancreatitis tissue and normal pancreatic tissue were obtained from 40 patients with pancreatic head cancer, 19 patients with chronic pancreatitis and 20 patients with duodenal benign tumours between January 2012 and January 2018 at the First Affiliated Hospital of University of Science and Technology of China.
All tissue specimens were fixed in formalin for 24 h and then embedded in paraffin. The study was approved by the Ethics Committee of the Anhui Provincial Hospital of the First Affiliated Hospital of the University of Science and Technology of China, and informed consent was obtained from each participant.

Cell lines and cell culture
PANC-1 cells were obtained from Shanghai Cell Bank. hTERT-HPNE cells (a human normal pancreatic ductal epithelial cell line) were obtained from the American Type Culture Collection.
PANC-1 cells were cultured in DMEM (HyClone Laboratories Inc, USA) containing 10% foetal bovine serum (Biological Industries, ISR). hTERT-HPNE cell lines were cultured in the recommended complete growth medium, which included 5% foetal bovine serum, 75% DMEM without glucose, 25% Medium M3 Base, 10 ng/mL human recombinant EGF, 5.5 mmol/L Dglucose (1 g/L) and 750 ng/mL puromycin. We collected surgical resection of PDAC tissues from 5 patients with pancreatic head cancer and normal pancreatic tissues from 5 patients with benign space-occupying duodenum between January 2019 and December 2019 in the Department of Biliary and Pancreatic Surgery of the First Affiliated Hospital of University of Science and Technology of China and then used a previous method to isolate and culture PSCs. [15] Then, CaPSCs derived from pancreatic cancer and NaPSCs derived from normal pancreas were obtained. All cell lines were cultured in a 37°C, 5% CO2 incubator. Transwell co-culture system (6-well plate, 0.4μm, 3412#, Corning, USA) was used. The upper chamber was seeded with 3×10 5 PANC-1 or hTERT-HPNE, and the lower plate was seeded with 3×10 5 human pancreatic stellate cells from different sources in a 5% incubator at 37 ℃ for 24 h.

Immunofluorescence
The specific operation process was the same as before. [15] Isolated PSCs from different sources were inoculated into a six-well culture plate equipped with a cover slip, cultured under standard conditions for 48 hours, and then incubated with primary antibody (rabbit monoclonal anti-α-SMA, 1:100, Abcam, USA). The cells were detected and imaged with a fluorescence microscope, and the purity of PSCs was analysed.

Cell proliferation assay
PANC-1 and HPNE cells co-cultured with PSCs from different sources were routinely digested and seeded in a 96-well plate at 3000 cells per well. The Cell Counting Kit-8 (CCK-8, Beyotime, Shanghai, China) method was used to detect the proliferation of pancreatic cancer cells, and cells that were not co-cultured were used as controls. Briefly, premixed medium (10 μL of CCK-8 solution, 100 μL of medium) was added to each well, and after incubation in the dark at 37°C for 3 h, a microplate reader was used to measure the light absorption value at 450 nm to evaluate cell viability.

Migration and invasion assays
The effect of activated PSCs on the migration and invasion of pancreatic ductal cells was evaluated using Transwell chambers. CaPSCs and NaPSCs were inoculated in a 24-well plate The migrated cells on the lower surface were counted to assess migration and invasion.

RNA isolation and quantitative real-time PCR
Cellular RNA was extracted by an RNA extraction kit (Takara, Japan), and cDNA was synthesized by a reverse transcription kit (Takara, Japan). qPCR was performed on a real-time PCR system (Roche, Switzerland) using a SYBR Green PCR kit(Takara, Japan). All primers (Table 1) were synthesized by Sangon Biotech (Shanghai, China). The RNA levels were normalized to human ACTB(β-actin) levels.

Lactate production and glucose utilization assays
Human pancreatic stellate cells (3×10 5 cells/well) from different sources were seeded into 6well plates for later use. Transwell co-culture system (6-well plate, 0.4μm, 3412#, Corning, USA) was used. The upper chamber was seeded with 3×10 5 PANC-1 or HPNE, while the lower chamber was not seeded with cells. After 12h culture with 10%FBS medium, the cells were The animal models were confirmed by pathological diagnosis.

Immunohistochemistry
The paraffin-embedded sections were dewaxed with xylene, hydrated with gradient ethanol and boiled in citrate buffer for 20 min. Endogenous peroxidase activity was blocked by 3% H2O2.
Rabbit polyclonal antibodies (Abcam, USA) against PKM2, LDHA and α-SMA were added at a concentration of 1:100 and incubated overnight at 4 ℃. After incubation with the secondary antibody for 1 hour, DAB was used for colour development. Finally, the sections were re-stained with haematoxylin and fixed by dehydration. For the negative control group, the main antibodies were replaced with PBS under the same conditions. Semi-quantitative method was used to score the results: staining area: 0, no staining; 1 point, 10%; 2 points, 10%-30%; 3, 30% staining; Dyeing intensity: 0, none; 1 points, light yellow; 2 points, yellow; 3 points, brown and yellow.
The total score was the product of the staining area and staining intensity score. Low expression was defined as a total score ≤3, and high expression was defined as a total score > 3. All immunohistochemical slides were scored independently by two experienced pathologists, and differences were adjusted for by a joint assessment.

Statistical analysis
Statistical data are presented as the means ± SDs (n = 3). Student's t test, chi-square test and Spearman correlation analysis were used for statistical analysis. P values of <0.05 were accepted as statistically significant.

Immunofluorescence shows that CaPSCs had the characteristics of activated PSCs
We used previous methods to obtain CaPSCs and NaPSCs from pancreatic cancer and normal pancreatic tissues, respectively. 15 Phenotypic identification of CaPSCs and NaPSCs was performed by immunofluorescence labelling of α-SMA, a marker protein of activated pancreatic stellate cells. Although quiescent PSCs will gradually activate and express α-SMA in vitro, immunofluorescence can distinguish them from activated PSCs. The expression of α-SMA in CaPSCs and their phenotypic characteristics were significantly different from NaPSCs, indicating that CaPSCs had the characteristics of activated PSCs. (Figure 1)

Co-culture with CaPSCs promotes the proliferation, migration and invasion of benign and malignant pancreatic duct epithelial cells
We examined the percentage changes in the proliferation, migration and invasion abilities of PANC-1 and HPNE cells in each co-culture system. The CCK-8 proliferation assay showed that the proliferation ability of HPNE and PANC-1 cells in the CaPSC co-culture group was significantly improved compared with that in the blank groups. (Figure 2a, b). Transwell experiments showed that the migration and invasion abilities of HPNE and PANC-1 cells in the CaPSC co-culture group were significantly improved compared with those in the NaPSC and blank groups. (Figure 2c, d)

Co-culture with CaPSCs promotes the Warburg effect in pancreatic duct epithelial cells (increased expression of PKM2 and LDHA; increased lactic acid production and glucose consumption)
To explore why CaPSCs promote the proliferation, migration and invasion of pancreatic cancer cells and normal pancreatic duct epithelial cells, we examined the effect of co-culture on aerobic glycolysis (Warburg effect) in HPNE and PANC-1 cells. First, we detected the changes in the lactic acid production and glucose consumption of HPNE and PANC-1 cells after co-culture, and the results showed that the lactic acid production and glucose consumption of HPNE and PANC-1 cells in the CaPSC group were significantly higher than those in the NaPSC group and blank group. (Figure 3a After finding that CaPSCs can promote glycolysis in pancreatic cancer and normal pancreatic duct epithelial cells, we further investigated whether the two are related in animal and human tissues. We successfully obtained animal models of CP and PDAC in rats by means of drug induction. (Figure 4) The statistical results of animal model establishment are briefly listed in Table 2 and Table 3. The CP rat model was established by DBTC caudal intravenous injection, and the incidence of CP in rats was 35% after 6 weeks; the PDAC rat model was established using DMBA surgical implantation into the pancreas, and the incidence of PDAC in rats was 30% after 16 weeks. Subsequently, paraffin sections of CP and PDAC pancreatic tissues were stained with α-SMA and PKM2/LDHA immunohistochemistry, and the staining results were scored. The results showed that PKM2 and LDHA were not expressed or weakly expressed in pancreatic duct epithelial cells and that α-SMA was weakly expressed in peri-pancreatic duct stroma in normal pancreatic tissue of rats. (Figure 5a-d) PKM2/LDHA in pancreatic duct epithelial cells and α-SMA in peripancreatic duct stroma were significantly higher in the CP group than in the normal pancreas group. (Figure 5e-h) This phenomenon has also been observed in rat pancreatic cancer tissues, and protein expression at these sites was higher than that in the CP group. (Figure 5i-j)

Immunohistochemical analysis of human CP and PDAC tissues shows that the activation of PSCs is correlated with the high expression of glycolytic proteins.
In immunohistochemical sections of human tissues, we observed findings similar to those in the animal model, namely, activation of PSCs and enhancement of glycolytic metabolism.
( Figure 6) Considering the heterogeneity of human pathological sections, we analysed the correlation between α-SMA in the peripancreatic duct interstitium and glycoprotein expression in pancreatic duct epithelial cells in the CP and PDAC groups. We first confirmed that the expression of α-SMA in the CP group and PDAC group was significantly different from that in the normal pancreas group. (Table 4) Correlation analysis (Table 5) showed that there was a correlation between PKM2/LDHA expression in pancreatic duct epithelial cells and α-SMA expression in the peri-pancreatic duct interstitium in both the CP and PDAC groups and that the correlation between α-SMA and PKM2/LDHA expression was higher in the PDAC group than in the CP group.   [22,23] In this study, we successfully obtained activated and quiescent PSCs from pancreatic cancer patients and individuals with a normal pancreas using a method described previously, and phenotypic identification was performed by immunofluorescence. In our previous work, we found that activated PSCs promoted the proliferation, migration and invasion of pancreatic cancer cells. [24] We replicated the results using activated cells isolated in this study.
Subsequently, we co-cultured activated PSCs with HPNE cells, a normal pancreatic duct epithelial cell line that is thought to have the potential to differentiate into cancer cells, [25] and found that activated PSCs also promoted the proliferation, invasion, and migration of HPNE

Conclusion
In summary, we believe that activated pancreatic stellate cells have the ability to enhance the Warburg effect in pancreatic duct epithelial cells and play a supporting and promoting role in the malignant progression of chronic pancreatitis. The idea and process of the whole experiment are shown in Figure 7. The specific influencing mechanism needs to be further studied. Our current study provides a new theoretical basis for further understanding the mechanism of CP malignancy and selecting diagnosis and treatment targets for reversing CP malignancy.

Funding
This study was supported by funds from Anhui Provincial Natural Science Foundation

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate

Consent for publication
All the authors agreed to be published

Competing interests
There is no confict of interest.

Figure 1：
The expression of α-SMA in NaPSCs and CaPSCs was detected by immunofluorescence, and CaPSCs showed a highly activated state.

Figure 2：
Co-culture with CaPSCs enhanced the proliferation, invasion and migration of HPNE and PANC-1 cells.
a: CCK-8 proliferation assay after co-culture of HPNE cells with NaPSCs or CaPSCs.
c: Transwell migration and invasion assays after co-culture of HPNE cells with NaPSCs or CaPSCs.
d: Transwell migration and invasion assays after co-culture of PANC-1 cells with NaPSCs or CaPSCs.

Figure 3
Co-culture with CaPSCs promoted lactic acid production, glucose consumption and the expression of the glycolytic proteins LDHA/PKM2 in HPNE and PANC-1 cells. a: Changes in HPNE cell lactic acid production after co-culture with NaPSCs or CaPSCs; b: Changes in HPNE cell glucose consumption after co-culture with NaPSCs or CaPSCs; c: Changes in PANC-1 cell lactic acid production after co-culture with NaPSCs or CaPSCs; d: Changes in PANC-1 cell glucose consumption after co-culture with NaPSCs or CaPSCs; e: Changes in the protein and mRNA expression of LDHA and PKM2 in HPNE cells after coculture with NaPSCs or CaPSCs; f: Changes in the protein and mRNA expression of LDHA and PKM2 in PANC-1 cells after coculture with NaPSCs or CaPSCs; *P < 0.05, **P < 0.01, ***P < 0.001

Figure 4
Establishment and identi cation of CP and PDAC model rats a: The rat CP models were established by injecting DBTC solution through the tail vein. B-C: After 6 weeks of modelling, the pancreas of rats in the CP group and the pancreas of rats in the control group (injection of saline) were collected. Severe brosis was observed in the pancreas -yellow arrow of the CP group. b: Rats with orthotopic pancreatic cancer induced by surgical implantation of DMBA; e-f: Sixteen weeks after modelling, the pancreas of rats in situ PDAC and the pancreas of rats in the sham operation control group. Severe brosis and local mass formation were observed at the pancreatic red arrow of the rats in the in situ PDAC group. g-h: Gross pancreatic specimens and pathological HE staining of rats in the CP group and control group; i-j: Gross pancreatic specimens and pathological HE staining of rats in the in situ PDAC group and control group. Design idea and veri cation ow chart of this study. After nding that activated PSCs of pancreatic cancer origin can promote the proliferation, migration and invasion of benign and malignant pancreatic duct epithelial cells, we hypothesized that activated PSCs may promote the malignant transformation of chronic pancreatitis. In the process of searching for the mechanism, a series of experiments con rmed that the activated PSCs could promote the Warburg effect in benign pancreatic duct epithelial cells in vitro, and the correlation between the glycolytic proteins of pancreatic duct cells and the activated PSCs in the peri-pancreatic duct matrix was con rmed in vivo.