Exosomal NNMT from peritoneum lavage fluid promotes peritoneal metastasis in gastric cancer

Abstract

Peritoneal metastasis (PM) is the major cause of recurrence in patients with gastric cancer (GC) and is associated with poor prognosis. The oncogenic role of Nicotinamide N‐methyltransferase (NNMT) in GC has been reported, but the role of secreted NNMT that is transported by exosomes remains unknown. In this study, exosomes were isolated from GC patients with or without PM and from GC cell line, including GC‐114, GC‐026, MKN45, and SNU‐16 cells. The contents of NNMT were significantly enhanced in exosomes isolated from GC patients with PM compared with those from GC patients without PM. Furthermore, the levels of NNMT were significantly enhanced in exosomes from GC cell lines relative to those from normal human gastric epithelial cell line GES‐1 cells. These data indicate that NNMT may be involved in intercellular communication for peritoneal dissemination. Moreover, colocalization of GC‐derived exosomal NNMT was found in human peritoneal mesothelial cell line HMrSV5 cells. Additionally, relative to GES‐1 exosomes, SNU‐16 exosomes significantly activated TGF‐β/smad2 signaling in HMrSV5 cells. However, when NNMT was silenced, the activation of TGF‐β/smad2 by SNU‐16 exosomes was abolished in HMrSV5 cells. We propose that NNMT‐containing exosomes derived from GC cells could promote peritoneal metastasis via TGF‐β/smad2 signaling.

1. INTRODUCTION

Peritoneal carcinomatosis is a major form of metastatic spread in patients with gastric cancer (GC). ¹ , ² Due to this high resistance, therapy of disseminated peritoneal lesions is especially difficult. ³ The process of peritoneal metastasis (PM) is complex, and knowledge of the underlying mechanisms remains lacking. ⁴ , ⁵ The intraperitoneal release of cancer cells plays a key role in metastatic formation from primary tumors. ³ However, intraperitoneal free cancer cells are not necessary for the occurrence of peritoneal dissemination. ⁶

Exosomes are small membrane vesicles measuring 50 to 100 nm in diameter that are secreted from cells, and they are key regulators of intercellular communication. ⁷ Tumor cells can secrete large amounts of exosomes that carry mRNA, microRNA, and proteins to communicate signals to local and remote cells and tissues. ⁸ Accumulating evidence has shown that exosomes carrying mRNA and microRNA play key roles in the progression of tumors, but the specific functions of proteins in exosomes are largely unknown. ⁷ , ⁹ , ¹⁰

Nicotinamide N‐methyltransferase (NNMT), an S‐adenosyl‐L‐methionine‐dependent cytoplasmic enzyme, mainly catalyzes the methylation of nicotinamide and other pyridines into pyridinium ions. ¹¹ Overexpression of NNMT in different tumors, including liver cancer, ovarian cancer, and breast cancer, has been widely reported. ¹² , ¹³ , ¹⁴ Liang et al. report that the upregulation of NNMT in GC cells may enhance the expression of TGF‐β1 and then activate epithelial‐mesenchymal transition. ¹⁵ Chen et al. suggests that NNMT is a potential biomarker of prognosis among GC patients. ¹⁶ However, little is known about the involvement of NNMT in PM in GC patients.

In this study, exosomes were isolated from intraoperative peritoneal lavage fluid (PLF) samples from GC patients. We also investigated the exosomal NNMT levels and the molecular mechanism underlying the development of PM induced by NNMT in GC patients.

2. MATERIALS AND METHODS

2.1. Patient samples

Sixty paired samples of GC and adjacent noncancer tissues from patients who underwent surgery at Affiliated Hangzhou First People Hospital were obtained between January and December 2018. At the time of tissue sampling, peritoneal lavage fluid was obtained. Under general anesthesia, laparoscopic observation (P1) or positive cytology of peritoneal lavage fluids (CY1) was used for the diagnosis of peritoneal metastasis. Surgically resected samples, including those of the primary tumors, paired adjacent noncancerous tissues, peritoneal metastatic lesions, and normal peritoneal tissue adjacent to the corresponding possible peritoneal metastasis, were collected. Details of the patients were presented in Table 1. The present study was approved by the Institutional Review Board (IRB) of Affiliated Hangzhou First People Hospital (Hangzhou, China, AHFPH‐20180736) and all the participants have provided written informed consent for this study.

TABLE 1.

Characteristics of patients with gastric cancer

Variable	Patients without PM	Patients with PM
Age (mean and range; yr)	69 (36–81)	67 (38–81)
Gender
M	20	8
F	20	12
Clinical T stage
T1	9	4
T2	10	5
T3	11	6
T4	10	5
Initial or recurrence
Initial	28	11
Recurrence	12	9
Differentiation (Lauren classification)
Intestinal	11	7
Diffuse	29	13
Clinical N stage
N0	8	5
N1	11	4
N2	9	6
N3	12	5
Ascites
None	20	18
Mild	7	2
Moderate	8	0
Severe	5	0
Other distant metastasis
Negative	31	17
Positive	9	3

2.2. Cell culture

The human GC cell lines GC‐114, GC‐026, MKN45, and SNU‐16, the normal human gastric epithelial cell line GES‐1 and the human peritoneal mesothelial cell line HMrSV5 were purchased from the Cell Center of Shanghai Institutes for Biological Sciences (Shanghai, China). All cell lines were cultured in RPMI‐1640 (GE Healthcare Life Sciences, Logan, UT) supplemented with 10% fetal bovine serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA), streptomycin (100 mg/ml), and penicillin (100 U/ml) at 37°C in a humidified atmosphere containing 5% CO₂.

2.3. Isolation of exosomes

To remove floating cells, the peritoneal fluid samples were centrifuged at 2000 g for 10 min. Then, the supernatants were filtered through an 800 nm filter (Millipore, Burlington, MA) to remove cell debris and ultracentrifuged at 150,000 g for 70 min at 4°C. In addition, GC cells were cultured in RPMI‐1640 supplemented with 10% exosome‐free FBS for 48 hour. Next, the medium was collected, and the GET™ Exosome Isolation Kit (GET301‐10, Genexosome Technologies, New Jersey, Ohio) was used to isolate exosomes according to the manufacturer's instructions. The presence of isolated extracellular vesicles (EV) was validated using an HT‐7700 transmission electron microscope (Hitachi High‐Technologies, Tokyo, Japan) (bar = 50 nm). Purified exosomes were labeled with the PKH‐26 red fluorescent linker Mini Kit (Sigma) according to the manufacturer's instructions. For the treatment of cells, 40 μg exosomes were used for each sample.

2.4. RNA isolation

To extract RNA from the exosomes or cells, an RNeasy Micro Kit (Qiagen, Hilden, Germany) was used following the manufacturer's instructions. The concentration and the purity of the extracted RNA were determined by measuring the ratio optical density (OD) 260/OD280.

2.5. qPCR

RNA reverse transcription was performed according to the instructions of the QuantiTect Reverse Transcription Kit (Thermo Fisher Scientific, Inc., Waltham, MA). SYBR Green Super Mix (Bio Rad Laboratories, Inc., Hercules, CA) was used for real‐time quantitative PCR according to the manufacturer's instructions. The PCR protocol was as follows: 95°C for 30 second, followed by 45 cycles of 5 second at 95°C and 30 second at 60°C. Relative mRNA expression was normalized to GAPDH expression using the 2^‐∆∆Cq method. ¹⁷

2.6. Transwell migration assay

Cell migration assays were performed using Boyden chambers (8‐μm pore filter; Corning Inc., Corning, NY). Cells (1 × 10⁵/well) were plated into the top chamber, and medium containing 10% FBS was placed into the bottom chamber. After incubation at 37°C in 5% CO₂ for 12 hour, the cells remaining at the upper surface of the membrane were removed with a cotton swab. The cells that had migrated through the 8 μm pores and adhered to the lower surface of the membrane were fixed with 4% paraformaldehyde, stained with crystal violet, and photographed.

2.7. Lentivirus vectors silencing or overexpressing NNMT

Lentivirus vectors inhibiting NNMT expression (Len‐shNNMT) or overexpressing NNMT (Len‐NNMT) and control vector (Len‐NC) were purchased from Genechem (Shanghai, China).

2.8. Western blot

Protein was isolated with radioimmunoprecipitation assay buffer (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China). A bicinchoninic protein assay kit (Pierce; Thermo Fisher Scientific, Inc.) was used to determine the protein concentration. Equal amounts of protein (15 μg/lane) were separated using 10% SDS‐PAGE at 300 mA for 2 hour and then transferred onto a polyvinylidene fluoride membrane. Then, primary antibodies, including NNMT (ab119758), Alix (ab275377), TSG101 (ab125011), CD63 (ab134045), TGF‐β1 (ab215715), p‐smad2 (ab188334), GAPDH (ab8245) (1:1000, all from Abcam, Cambridge, UK) were incubated with the membranes at room temperature for 2 hour. Following several washes with TBST, the membranes were incubated with HRP‐conjugated goat anti‐rabbit IgG (1:5000; ZB‐2306, Zhongshan Gold Bridge Biological Technology Co., Beijing, China) for 2 hour at room temperature and then washed. The proteins were detected using an enhanced chemiluminescence detection kit according to the manufacturer's protocol (Merck KGaA, Darmstadt, Germany). ImageJ 1.8.0 (National Institutes of Health, Bethesda, MD) was applied to quantify the relative protein levels. GAPDH was used as an internal control.

2.9. ELISA

Cell lysates were centrifuged at 16,000× g for 15 min at 4°C and levels of TGF‐β1 (R&D Systems, Inc., Minneapolis, MN) according to the instructions.

2.10. Immunohistochemical (IHC) staining

Peritoneal samples were fixed in 4% phosphate‐buffered neutral formalin at room temperature for 20 min, embedded in paraffin and cut into 5‐μm thick sections. The sections were incubated with TGF‐β1 (ab215715) Abcam, Cambridge, UK) at a 1:50 dilution and 4°C overnight. Detection of the primary antibody was performed via incubation with a horseradish peroxidase‐conjugated goat anti‐rabbit secondary antibody (ZDR‐5036, Zhongshan Gold Bridge, Beijing, China) for 1 hour at room temperature and visualized using light microscopy (magnification, 40×, Olympus CK40, Olympus Corporation, Japan).

2.11. Statistical analysis

Data were expressed as the mean ± standard error. Each experiment was carried out with three replicates. Multiple comparisons were performed using one‐way analysis of variance followed by Tukey's multiple comparison test. P < 0.05 was considered to indicate a statistically significant difference. The data were analyzed using the SPSS software, version 20.0 (SPSS, Inc., Chicago, IL).

3. RESULTS

3.1. NNMT is present in the peritoneal fluid exosomes of GC

First, we explored whether NNMT was present in the exosomes isolated from the peritoneal fluid samples of GC patients. As shown in Figure 1(A), exosomes were successfully isolated from the peritoneal fluid samples of GC patients, having a diameter of approximately 100 nm. Western blot assay showed that the contents of NNMT were significantly enhanced in exosomes isolated from GC patients with PM compared with those isolated from GC patients without PM (Figure 1(B)).

FIGURE 1.

NNMT was present in the peritoneal fluid exosomes of gastric cancer (GC). (A) Exosomes were successfully isolated from the peritoneal fluid samples of GC patients, having a diameter of approximately 100 nm. (B) the contents of NNMT were significantly enhanced in exosomes isolated from GC patients with PM compared with those isolated from GC patients without PM

3.2. Exosome‐derived NNMT is present in peritoneum cells

We then analyzed whether exosome‐derived NNMT could be transferred from gastric cancer cells to peritoneum cells. Here, exosomes from the GC cell lines GC‐114, GC‐026, MKN45, and SNU‐16 and the normal human gastric epithelial cell line GES‐1 were isolated. Western blot assay showed that the levels of NNMT were significantly enhanced in exosomes from GC‐114, GC‐026, MKN45, and SNU‐16 cells relative to exosomes from GES‐1 cells (Figure 2(A)). Among the GC cells, the SNU‐16 cells had the highest level of NNMT (Figure 2(A)). Hence, exosomes isolated from SNU‐16 cells were used for coculture with the human peritoneal mesothelial cell line HMrSV5. Subsequently, we transfected Len‐NNMT‐shRNA into SNU‐16 cells and isolated the corresponding exosomes. As shown in Figure 2(B), compared with Len‐NC transfection, transfection with Len‐NNMT‐shRNA significantly decreased the expression of NNMT in SNU‐16 cells. Furthermore, exosomal NNMT was found to be reduced in SNU‐16 cells transfected with Len‐NNMT‐shRNA compared with Len‐NC‐transfected cells (Figure 2(C)). We then analyzed whether exosomal NNMT could be transferred into recipient cells. As shown in Figure 2(D), GFP‐tagged NNMT was found in the exosomes, exhibiting PKH‐26 red fluorescence. These exosomes were found to quickly enter HMrSV5 cells (Figure 2(D)).

FIGURE 2.

Exosome‐derived NNMT is present in peritoneum cells. (A) Western blot assay showed that the levels of NNMT were significantly enhanced in gastric cancer (GC)‐114, GC‐026, MKN45, and SNU‐16 cells relative to GES‐1 cells. (B) Compared with Len‐NC transfection, transfection with Len‐NNMT‐shRNA significantly decreased the expression of NNMT in SNU‐16 cells. (C) Exosomal NNMT was found to be reduced in SNU‐16 cells transfected with Len‐NNMT‐shRNA relative to those transfected with Len‐NC. (D) The exosomes were found to quickly enter HMrSV5 cells. *P < 0.05, **P < 0.01, ***P < 0.001 vs. control

3.3. Exosome NNMT activates TGF‐β/smad2 signaling in HMrSV5 cells

A previous study showed that NNMT activates TGF‐β/smad2 signaling in gastric cancer. ¹⁶ Hence, we investigated the effects of exosome‐derived NNMT on the activation of TGF‐β/smad2 signaling in HMrSV5 cells. 40 micrograms of SNU‐16 exosomes were added to medium containing 10⁶ HMrSV5 cells in a six‐well plate. As shown in Figure 3(A), compared with GES‐1 exosomes, SNU‐16 exosomes significantly activated TGF‐β/smad2 signaling. However, when SNU‐16 cells by transfection with Len‐NNMT‐shRNA, the activation of TGF‐β/smad2 signaling by SNU‐16 exosomes was abolished (Figure 3(A)).

FIGURE 3.

Exosomal NNMT derived from SNU‐16 cells activates TGF‐β/smad2 signaling in HMrSV5 cells. (A) Compared with GES‐1 exosomes, SNU‐16 exosomes significantly activated TGF‐β/smad2 signaling. (B) RT‐PCR analysis showed that the overexpression of NNMT significantly enhanced the mRNA levels of TGF‐β/smad2 in HMrSV5 cells. (C) The expression of TGF‐β/smad2 was also increased in HMrSV5 cells transfected with Len‐NNMT. (D) ELISA assay showed that TGFβR was inhibited by Ki26894, the level of TGF‐β1 was decreased even when NNMT was overexpressed in SNU‐16 cells. (E) Inhibition of TGFβR by Ki26894 obviously reversed NNMT‐induced SNU‐16 cell migration. *P < 0.05, **P < 0.01, ***P < 0.001 vs. control

To further explore the effect of NNMT on the malignancy of peritoneal cells, we transfected HMrSV5 cells with Len‐NNMT. RT‐PCR analysis showed that the overexpression of NNMT significantly enhanced the mRNA levels of TGF‐β/smad2 in HMrSV5 cells (Figure 3(B)). Consistent with this finding, the expression of TGF‐β/smad2 was increased in HMrSV5 cells transfected with Len‐NNMT (Figure 3(C)). To clarify whether NNMT activates a TGF‐β1 autocrine loop, EILSA assay was performed. We found that overexpression of NNMT increased the level of TGF‐β1 in the supernatant of SNU‐16 cells (Figure 3(D)). However, when TGF‐β1 receptor (TGFβR) was inhibited by Ki26894, a typical TGFβR inhibitor, the level of TGF‐β1 was decreased even when NNMT was overexpressed in SNU‐16 cells (Figure 3(D)). In addition, we also explored whether NNMT enhanced cell migration via TGFβR. Our data showed that overexpression enhanced SNU‐16 cell migration, but inhibition of TGFβR by Ki26894 obviously reversed NNMT‐induced cell migration (Figure 3(E)). These data suggested that NNMT activated a TGF‐beta autocrine loop.

Moreover, we investigated the expression of NNMT and TGF‐β/smad2 in peritoneal metastatic lesions. Compared with GC patients without PM, GC patients with PM exhibited significantly enhanced expression of NNMT and TGF‐β/smad2 in peritoneal metastatic lesions (Figure 4(A)). Meanwhile, IHC staining showed that the expression of TGF‐β and Smad2 was obviously elevated in peritoneal lesions in GC patients with PM compared with that in peritoneal tissues in GC patients without PM (Figure 4(B)).

FIGURE 4.

The expression of TGF‐β/smad2 is enhanced in peritoneal metastatic lesions (A) compared with gastric cancer (GC) patients without PM, GC patients with PM exhibited significantly enhanced expression of NNMT and TGF‐β/smad2 in peritoneal metastatic lesions. (B) IHC staining showed that the expression of TGF‐β and Smad2 was obviously elevated in peritoneal lesions in GC patients with PM compared with that in peritoneal lesions in GC patients without PM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. the indicated group

3.4. TGF‐β regulates the malignancy of HMrSV5 cells

Next, we explored the effects of TGF‐β on the malignant phenotype of HMrSV5 cells. As shown in Figure 5(A) and (B), TGF‐β treatment significantly increased the mRNA and phosphorylation levels of Smad2. In contrast, shRNA targeting TGF‐β reduced the mRNA and protein levels of Smad2 in HMrSV5 cells (Figure 5(A) and (B)). We then cocultured HMrSV5 cells with SNU‐16 exosomes. Our data showed that SNU‐16 exosomes increased the expression of TGF‐β and Smad2 and that these effects could be abolished by shRNA targeting TGF‐β in HMrSV5 cells (Figure 5(C)). As expected, SNU‐16 exosomes strongly enhanced HMrSV5 cell proliferation and migration (Figure 5(D)). However, these effects were diminished in HMrSV5 cells treated with shRNA targeting TGF‐β (Figure 5(D)). These data indicated that the SNU‐16 exosomes promoted the malignancy of HMrSV5 cells via TGF‐β/Smad2 signaling.

FIGURE 5.

TGF‐β regulates the malignancy of HMrSV5 cells. TGF‐β treatment significantly increased the mRNA (A) and protein (B) levels of Smad2 in HMrSV5 cells. (C) SNU‐16 exosomes increased the expression of TGF‐β, but this effect could be abolished by shRNA targeting TGF‐β. (D) SNU‐16 exosomes strongly enhanced HMrSV5 cell migration, but the treatment of HMrSV5 cells with shRNA targeting TGF‐β could diminish this effect. *P < 0.05, **P < 0.01, ***P < 0.001 vs. as indicated

4. DISCUSSION

The mechanism by which PM is regulated remains to be fully clarified. ¹⁸ The invasion of cancer cells through the peritoneal membrane has been suggested to be a major reason for the formation of peritoneal metastasis. ¹⁹ Exosomes are small vesicles that are released by cancer cells and transfer mRNA, microRNA, and proteins from donor cells to recipient cells. ²⁰ , ²¹ In GC, NNMT has been shown to induce epithelial‐mesenchymal transition (EMT) and promote metastasis. ¹⁵ However, whether NNMT is involved in peritoneal metastasis in GC patients has not been examined.

In the present study, we showed that the contents of NNMT were significantly increased in exosomes isolated from GC patients with PM compared with those from GC patients without PM. Furthermore, the levels of NNMT were significantly enhanced in exosomes from GC cell lines relative to those from GES‐1 cells. These findings indicate that NNMT may be involved in intercellular communication for peritoneal dissemination via the alteration of recipient cells, including GC cells and normal mesothelial cells. Moreover, colocalization of exosomal NNMT was found in HMrSV5 cells, indicating that exosomal NNMT can be transferred into recipient cells. Here, we provide the first demonstration that GC cell‐secreted NNMT can be transferred to target organs, explaining how NNMT can be transported between cells.

We also explored the mechanism by which exosomal NNMT mediates the formation of PM. Interestingly, we found that both TGF‐β1 mRNA and protein was upregulated in mesothelial cells transfected with len‐NNMT. It is reported that TGF‐β1 can activate the nuclear TGF‐β1 promoter through Smads, thereby inducing the expression of endogenous TGF‐β1, upregulate thing expression of type I and II receptors, amplifying the signal through Smads pathway and forming a positive feedback loop. ²² , ²³ To clarify whether NNMT activates a TGF‐β1 autocrine loop, we applied a TGFβR inhibitor, Ki26894. Transwell assay demonstrated that NNMT enhanced cell migration via TGFβR. These data indicated that an autocrine loop existed between NNMT and TGF‐β1. Compared with GES‐1 exosomes, SNU‐16 exosomes significantly activated TGF‐β/smad2 signaling in HMrSV5 cells. However, when NNMT was silenced, the activation of TGF‐β/smad2 by SNU‐16 exosomes in HMrSV5 cells was abolished. Furthermore, we investigated the expression of NNMT and TGF‐β/smad2 in peritoneal metastatic lesions. Compared with GC patients without PM, GC patients with PM exhibited significantly enhanced expression of NNMT and TGF‐β/smad2 in peritoneal metastatic lesions. These findings suggest that exosomal NNMT derived from GC cells can facilitate PM via TGF‐β/smad2 signaling.

However, there are limitations in the current study. We did not provide in vivo data to indicate whether these exosome‐treated TGF‐β‐affected mesothelial cells had become protumor cells. In the future study, we will perform the in vivo assay to validate the conclusion that SNU‐16 exosomes promoted the malignancy of mesothelial cells via TGF‐β/Smad2 signaling.

Together, the results suggest that exosomal NNMT, secreted from GC cells, is transferred to peritoneal cells, which promotes peritoneal dissemination. Our findings indicate that exosomal NNMT may play a key role in peritoneal carcinomatosis, thereby suggesting a novel approach for the early diagnosis of peritoneal dissemination of GC.

Zhu A‐K, Shan Y‐Q, Zhang J, Liu X‐C, Ying R‐C, Kong W‐C. Exosomal NNMT from peritoneum lavage fluid promotes peritoneal metastasis in gastric cancer. Kaohsiung J Med Sci. 2021;37:305–313. 10.1002/kjm2.12334