Plasma Exosome-Derived SENP1 May Be a Potential Prognostic Predictor for Melanoma

Objective To evaluate plasma exosome-derived SUMO-specific protease (SENP)1 levels and assess their prognostic value in melanoma. Patients and Methods We extracted exosomes from the plasma of 126 melanoma patients, and identified them with transmission electron microscopy, nanoparticle tracking analysis and western blotting. The plasma exosome-derived SENP1 levels of melanoma patients and healthy controls were detected with ELISA. Results Plasma exosome-derived SENP1 levels in melanoma patients were significantly upregulated than in healthy controls (P < 0.001). Plasma exosome-derived SENP1 levels in melanoma patients with tumor size >10 cm, located in the mucosa or viscera, with Clark level IV/V, with lymph node metastasis, and TNM stages IIb–IV were significantly higher than in patients in with tumor size <10 cm, located in the skin, with Clark level I–III, without lymph node metastasis, and TNM stages IIb–IV (all P < 0.05). Disease-free survival (DFS) and overall survival (OS) were worse in melanoma patients who had higher plasma exosome-derived SENP1 levels than lower plasma exosome-derived SENP1 levels (both P < 0.001). Area under the receiver operating characteristic curve (AUROC) of plasma exosome-derived SENP1 for predicting 3-year DFS of melanoma patients was 0.82 [95% confidence interval (CI): 0.74–0.88], with a sensitivity of 81.2% (95% CI: 69.9–89.6%) and specificity of 75.4% (95% CI: 62.2–85.9%). The AUROC of plasma exosome-derived SENP1 for predicting 3-year OS of melanoma patients was 0.76 (95% CI: 0.67–0.83), with a sensitivity of 95.7% (95% CI: 85.5–99.5%) and specificity of 62.0% (95% CI: 50.4–72.7%). Conclusions Melanoma patients with higher plasma exosome-derived SENP1 levels had worse DFS and OS. The plasma exosome-derived SENP1 levels may be a potential prognostic predictor for 3-year DFS and 3-year OS of melanoma.


INTRODUCTION
Melanoma is a malignant tumor caused by abnormal differentiation of melanocytes (1). Melanocytes exist in the neural crest of the embryo and migrate to many parts of the body during fetal development; mainly in the basal epidermis, hair follicles, mucosal surfaces, meninges and the choroidal layers of the eyes (2,3). Melanoma is most common in the skin, but the rectum, eyes, anus and vulva are also frequent sites. Malignant melanoma of eyelid skin is rare, accounting for only 1% of eyelid skin malignant tumors (4,5). Melanoma is difficult to detect during onset and has a high degree of malignancy. At present, its incidence is continuing to increase considerably at a rate of 3-8% per year, even as that of many other tumors declines (6). A century ago, melanoma was still a rare cancer, but the average incidence in western caucasian people reached 1 in 50 by the beginning of the 21st century (7).
According to data from Surveillance, Epidemiology, and End Results Program (SEER) in 2007, 1 in 63 Americans had melanoma (8). The incidence of melanoma was 27.5 per 100,000 in caucasian and 1.1 per 100,000 in black people (9). In the global statistics for malignant melanoma in 2012, 232,000 new cases and 55,000 deaths were reported, with a high fatality rate of 23.7% (10). Melanoma is considered a highly aggressive and metastatic disease with a survival rate of <10% for ≥5 years (11). After melanoma metastasizes, the life expectancy of patients generally does not exceed 1 year (12). In addition, approximately one-third of melanoma patients relapse. Although almost all organs can be affected, the most common target sites are the liver, bone and brain (13). Advances in the primary pathogenesis and therapeutic intervention have allowed melanoma patients to receive better treatment, including targeted therapies such as BRAF and MEK inhibitors, new immunotherapies such as anti-CTLA4 or anti-PD1 therapy, radiation therapy, biochemical therapy, and gene therapy (14). These treatments have improved the progression-free survival and overall survival (OS) rates of patients with advanced unresectable melanoma, but melanoma is still a deadly cancer, especially when patients are diagnosed at an advanced stage (15).
The post-translational modifications of proteins include methylation, phosphorylation and threacylation. Small ubiquitinrelated modifier (SUMO) modification is one of the posttranslational modifications of proteins (16). By covalently modifying several amino acid residues on the substrate protein, it can change its activity, stability and intracellular localization, and participate in the regulation of various cellular processes (17). Once the dynamic balance between SUMO and de-SUMO is broken, it can lead to tumor occurrence. Sumoylation is catalyzed by E1, E2 and E3 enzymes, while desumoylation is performed by SUMOspecific proteases (SENPs) (18). Human desumoylation proteases SENPs, including SENP1, SENP 2, SENP 3, SENP 5, SENP 6 and SENP 7, play a key role in this pathway (19). One is to catalyze the transformation of SUMO precursor into its active form; the other is to cut off the isopeptide bond between SUMO and the target protein to realize desumoylation (20). SENP is involved in the regulation of the cell cycle, cell proliferation, oxidative stress signaling or tumor gene fusion, leading to tumor formation. SENP family members play different roles (21,22). SENP1 is a nuclear protein, which can catalyze the desumoylation of SUMO1, SUMO2 and SUMO3 modified target proteins (23). In recent years, several studies have reported that SENP1 is highly expressed in prostatic and pancreatic cancer and osteosarcoma, and knockout of SENP1 affects the biological function of these tumors (24,25).
Exosomes are small membranous vesicles ranging from 40 to 100 nm (26). They can be used as functional mediators in cell interactions leading to cancer metastasis (27). Metastasis is a complex multistep process of cancer cell invasion, vascular survival, adhesion and host organ colonization. Exosomes affect every step of this cascade, and play an important role in cell-to-cell communication, which can be targeted by tumor therapy (28). A large number of studies have found that exosomes contain many important proteins, which can be used for early diagnosis of tumor, prognostic analysis of patients and targeting by tumor therapy (29,30). In this study, we aimed to investigate plasma exosome-derived SENP1 levels and determine their prognostic value in melanoma patients.

Patients and Clinical Samples
We collected blood samples from 126 melanoma patients from the First People's Hospital of Yancheng City, Shanghai Public Health Clinical Center and Shuyang People's Hospital from December 2015 to January 2020. The inclusion criteria were as follows: (1) first onset; (2) age ≥18 years; (3) diagnosed clinically and pathologically; (4) patients underwent surgical resection; (5) no preoperative hormone therapy, radiotherapy or chemotherapy; and (6) clinical data were complete. The exclusion criteria were as follows: (1) other types of malignant tumor; (2) congenital diseases, heart, lung and other serious organ dysfunction; (3) liver and kidney function were not complete; (4) preoperative hormone therapy, radiation therapy or chemotherapy; (5) clinical data were incomplete; and (6) lost to follow-up.
The baseline clinical data of 126 melanoma patients were collected from medical records including demographic features, tumor size, lymph node metastasis, American Joint Committee on cancer (AJCC) TNM stages, and pathological differentiation. Patients were followed up to January 2021, with a median followup duration of 42.5 months (range: 12.0-72.0 months). The survival data were collected from follow-up records, and diseasefree survival (DFS) and overall survival (OS) were calculated. DFS was defined as the duration from resection to disease recurrence, disease progression, or death. OS was defined as the time interval from resection to death. The follow-up results of the 120 patients enrolled in this study were obtained through medical records or telephone interviews.
In addition, we collected blood samples from 50 healthy people (median age 66 years, range 50-73 years) in the First People's Hospital of Yancheng City. All specimens were enrolled after

Plasma Exosome Isolation
After fasting for more than 12 hours, 5 ml of peripheral venous blood was collected by EDTA-K2 anticoagulant vacuum blood collection method. The plasma samples were centrifuged at 2500 rpm for 10 min. 5-20 ml of human plasma (1 × PBS diluted 5 times), 500 × g, 4°C centrifugation was carried out for 5 min at room temperature. Take the supernatant to a new centrifuge tube and mix it in 2000 × g. Centrifugation at 4°C for 10 min. Take the supernatant to a new centrifuge tube and add it at 10000×g. Further centrifugation at 4°C for 30 min removed large cell vesicles. 45mm filter is used to filter the supernatant to remove the large particles that may be mixed in the operation process. Take the filtrated supernatant to the ultracentrifugation tube and use 1×PBS buffer was filled up, weighed accurately and balanced, then put it on the rotor of the ultracentrifuge, 100,000×g Centrifugation at 4°C for 2 hours. The supernatant was discarded, and translucent sediments were found at the bottom of the tube. Resuspension the sediment in 1 × PBS buffer, and in 100,000×g Centrifugation at 4°C was carried out for 80 min at room temperature.

Transmission Electron Microscopy (TEM)
TEM was used to observe the morphology of exosomes. The extracted exosomes were removed from the −80°C freezer and thawed. Exosome suspension (20 ml) was added to the copper mesh. The solution was kept at room temperature for 1 min. The liquid was removed from the side with filter paper. Phosphotungstic acid solution (30 ml, pH = 6.8) was added to the copper mesh. The solution was negatively stained at room temperature for 1 min. The filter paper was used to absorb the negative dye solution, and the working voltage was adjusted to 75 kV after 10 min exposure to an incandescent lamp.

Nanoparticle Tracking Analysis (NTA)
NTA was used to observe the size of exosomes. After thawing, 500 ml exosomes was placed in the Nano Sight NS300 Instrument. We set the parameters to detect the outer body particle size. The experimental data were recorded and analyzed with NTA 3.3 software.

Western Blotting
Western blotting was used to confirm the exosome purification. Exosomes were isolated and added to sodium dodecyl sulfate (SDS) buffer to obtain total proteins. Total protein was separated with SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore, USA). The membranes were incubated in 5% skimmed milk for 1 h and then overnight with primary antibodies against annexin V, TSG101, CD9 and CD63 obtained from Santa Cruz Biotechnology, Inc. (TX, USA). Finally, the secondary antibody was added used to the membranes at room temperature for 1 h. Statistical Analysis SPSS 24.0 software (IBM) was performed for statistical analysis. Data were presented as mean ± standard deviation, median (range), or count (percentage). Chi square test or Wilcoxon's rank-sum test was performed to analyze the correlation. The differences in DFS and OS between two groups were assessed with log-rank test.

ELISA
Receiver operating characteristic (ROC) curve analysis was used to assess the prognostic value of plasma exosome-derived SENP1 levels in melanoma patients. Differences with P < 0.05 were considered statistically significant.

Baseline Characteristics
The baseline characteristics of 126 patients with melanoma are shown in Table 1

Characterization of Exosomes Isolated From Plasma
TEM, NTA and western blotting were used to confirm the exosome integrity and purification. The exosomes were obtained by gradient ultracentrifugation at low temperature and then fixed and stained. TEM images showed that the exosomes were clustered and connected with each other, with clear background. The diameter was between 100 and 200 nm. The exosomes had a double disc-like vesicular structure with intact lipid capsule ( Figure 1A). NTA revealed that the median value of the total particles was about 100 nm, mainly distributed between 50 and 200 nm, and the diameter of a small number of particles was between 0 and 50 nm ( Figure 1B). Annexin V, TSG101, CD9 and CD63 are common protein markers of exosomes. Western blotting showed that expression of Annexin V, Tsg101, CD9 and CD63 was positive ( Figure 1C).

Correlation Between Plasma Exosome-Derived SENP1 Levels and Clinical Pathological Parameters of Melanoma Patients
Plasma exosome-derived SENP1 levels in melanoma patients were significantly upregulated compared with those in healthy controls (P < 0.001; Figure 2A). We also investigated the correlations between plasma exosomederived SENP1 levels and tumor characteristics in melanoma patients. As to sex and age, there was no significant difference in the plasma exosome-derived SENP1 levels between melanoma patients and healthy controls (P > 0.05; Figures 2B, C). However, plasma exosome-derived SENP1 levels in melanoma patients with tumor size >10 cm, located in the mucosa or viscera, with Clark level IV/V, with lymph node metastasis, and TNM stages IIb-IV were significantly higher than in patients with tumor size <10 cm, located in the skin, with Clark level I-III, without lymph node metastasis, and TNM stages IIb-IV (all P < 0.05, Figures 2D-H).
In addition, Kaplan-Meier analysis showed that both DFS and OS were worse in melanoma patients who had higher plasma exosome-derived SENP1 levels than in those with lower plasma exosome-derived SENP1 levels (both P < 0.001; Figure 3).

Prognostic Value of Plasma Exosome-Derived SENP1 Levels in Melanoma Patients
We assessed the value of plasma exosome-derived SENP1 for predicting 3-year DFS and OS of melanoma patients with ROC curve analysis ( Table 2).
The The plasma exosome-derived SENP1 levels in melanoma patients were significantly up-regulated than in healthy controls (P < 0.001); (B) No significant differences were observed between male and female patients (P = 0.767).
(C) No significant differences were found between patients aged < 60 and >60 years (P = 0.147). (D) The plasma exosome-derived SENP1 levels in melanoma patients with tumor size more than 10 cm were significantly higher than in patients in with tumor size less than 10 cm; (E) The plasma exosome-derived SENP1 levels in melanoma patients located in the mucosa or viscera were significantly higher than in patients located in the skin (P < 0.001); (F) the plasma exosome-derived SENP1 levels in melanoma patients with Clark level IV -V were significantly higher than in patients with Clark level I -III (P < 0.001); (G) the plasma exosome-derived SENP1 levels in melanoma patients with lymph node metastasis were significantly higher than in patients without lymph node metastasis (P < 0.001); (H) the plasma exosomederived SENP1 levels in melanoma patients with TNM stages IIb-IV were significantly higher than in patients with TNM stages IIb-IV (P < 0.001).

DISCUSSION
Exosomes secreted by malignant tumor cells can carry a variety of characteristic DNA, mRNA, miRNA, long noncoding RNA and proteins into the tumor microenvironment, transfer between tumor cells and normal cells, and stimulate tumor cell proliferation, invasion and metastasis (31). As exosomes contain abundant maternal information and are widely distributed in various body fluids, they are easy to obtain, which is important for early screening, diagnosis and prognostic evaluation of tumors. SENP1 is an important member of the enzyme protein family, which regulates the reverse reaction of SUMO modification. Previous studies have shown that the overexpression of SENP1 in pancreatic cancer is related to the pathological stage and vascular invasion of patients (21). After gene knockout, the proliferation, migration and invasion of pancreatic cancer cells are inhibited. In gliomas, SENP1 expression is positively correlated with tumor grade. SENP1 is highly expressed in triple-negative breast cancer, which is associated with HER-2 loss (32). In addition, high expression of SENP1 is found in prostate cancer and thyroid oncocytoma. Luo X et al. have showed that miR-541-3p regulates proliferation, migration, and invasion of skin melanoma cell by targeting SENP1 (33). Huang et al. showed that increased SUMOylation of CD45 via loss of SENP1 suppresses CD45-mediated dephosphorylation of STAT3, which promotes myeloid-derived suppressor cells development and function, leading to tumorigenesis (34). However, the expression and prognostic value of SENP1 in melanoma patients remain unclear. In addition, it has been proved that exosomes contain many important proteins, which can be used for the early diagnosis of tumor, the prognosis analysis of patients and the targeting of tumor therapy. Hence, we aimed to investigate plasma exosome-derived SENP1 levels and determine their prognostic value in melanoma patients in this study, which is believed to be the first to assess the potential of plasma exosome-derived SENP1 in melanoma diagnosis.
We firstly extracted and characterized exosomes from the plasma of melanoma patients. TEM showed that typical exosomes had oval or bowl-shaped microvesicles. NTA showed that the peak size of plasma exosomes was 50-120 nm. Western blotting showed that the plasma exosomes were positive for Annexin V, Tsg101, CD9 and CD63. We compared the plasma exosome-derived SENP1 levels between melanoma patients and healthy controls. We showed that plasma exosome-derived SENP1 levels in melanoma patients were significantly upregulated, which is consistent with the trend in SENP1 protein expression in melanoma patients.
We also investigated the correlations between plasma exosomederived SENP1 levels and tumor characteristics in melanoma patients. The plasma exosome-derived SENP1 levels were related to tumor size, tumor location, depth of tumor invasion, lymph node metastasis, and TNM stage. The plasma exosome-derived SENP1 levels in melanoma patients with tumor size >10 cm, located in the mucosa or viscera, with Clark level IV/V, lymph node metastasis, and TNM stages IIb-IV were significantly higher than in patients in with tumor size <10 cm, located in the skin, with Clark level I-III,

A B
FIGURE 3 | Association of plasma exosome-derived SENP1 levels with DFS and OS in melanoma patients. (A) Among all melanoma patients, DFS was worse in melanoma patients who had higher plasma exosome-derived SENP1 levels than those melanoma patients with lower plasma exosome-derived SENP1 levels (P < 0.001).
(B) Among all melanoma patients, OS was worse in melanoma patients who had higher plasma exosome-derived SENP1 levels than those melanoma patients with lower plasma exosome-derived SENP1 levels (P < 0.001). without lymph node metastasis, and TNM stages IIb-IV. In addition, both DFS and OS were worse in melanoma patients who had higher plasma exosome-derived SENP1 levels than in those with lower plasma exosome-derived SENP1 levels. We showed that plasma exosome-derived SENP1 had good performance for predicting 3-year DFS and 3-year OS of melanoma patients. There were some limitations to this study. First, although this was a large study to evaluate the plasma exosome-derived SENP1 levels in melanoma patients, its prognostic value needs to be verified by multicenter and larger samples. Second, the mechanism of plasma exosome-derived SENP1 affecting the prognosis of melanoma patients needs to be studied in vivo and in vitro. Third, melanoma patients with different treatment regimens would have an impact on the prognosis. We need to expand the number of cases and group them according to different treatment regimens, so as to eliminate the impact of treatment regimens on the results. And also, we determined the expression of SENP1 in the tissues of melanoma patients by IHC, and found that the SENP1 protein level was significantly higher than that in the adjacent tissues, but its relationship with the prognosis of patients was not obvious.
In summary, our study demonstrated that melanoma patients with higher plasma exosome-derived SENP1 levels had worse DFS and OS. The plasma exosome-derived SENP1 levels may be a potential prognostic predictor for 3-year DFS and OS of melanoma patients.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
HH, BL, YS, and HW contributed to study concept and design, acquisition of data, analysis and interpretation of data, and drafting of the manuscript. BZ and YM contributed to statistical analysis. GZ contributed to study concept and design, study supervision and critical revision of the manuscript. All authors contributed to the article and approved the submitted version.