The protocol of this retrospective study was approved by the Institutional Review Board of China-Japan Union Hospital, Jilin University. Specimens were collected from 81 consecutive patients who were diagnosed with skin and mucosal melanoma at the Department of Pathology (China-Japan Union Hospital) between September 2016 and December 2018. Informed consent was obtained from patients at the time of sample collection. The diagnosis had been established based on pathological examination following complete surgical excision of the lesion. The eligibility criteria included i) age 18-80 years and ii) primary skin lesion without a history of radiotherapy or chemotherapy. The exclusion criteria included overweight status (body mass index >25 kg/m²), diabetes, and concomitant tumors, as elevated UCP2 expression may be associated with these conditions. Based on these criteria, the present study included 65 patients (33 men and 32 women) with cutaneous (n=52) and mucosal (n=13) melanoma. For comparison, control skin tissues were obtained from 49 healthy individuals who had undergone cosmetic surgery (control group) and surgical specimens were also collected from 51 healthy individuals who had undergone excision of compound nevus (compound nevus group). The clinical characteristics of the patients are summarized in Tables I and II.

Human melanoma A375 cells were purchased from the American Type Culture Collection (CRL-1619). Human melanoma SK-Mel-28 cells were kindly provided by Dr Stephan Witt from our institution (originally purchased from the American Type Culture Collection). A375 cells are more aggressive compared with SK-Mel-28 cells (17). The cells were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (Atlanta Biologicals, Inc.) and 1 mM sodium pyruvate, which was maintained at 37̊C in a humidified incubator (95% air and 5% CO₂). Mycoplasma testing was routinely performed for the cell lines.

UCP2 shRNA lentivirus (LVPi026640) and scramble shRNA lentivirus (LVP015G) were purchased from Applied Biological Materials. Both vectors contained a green fluorescent protein (GFP) tag. The target sequences were as follows: 5′-CGG TTA CAG ATC CAA GGA GAA-3′, 5′-GGC CTG TA TGA TTC TGT CA-3′, 5′-GCA CCG TCA ATG CCT ACA A-3′ and 5′-CGT GGT CAA GAC GAG ATA CAT GAA CTC TG-3′. JC-1 dye was purchased from Cayman Chemicals (cat. no. 15003).

All primary antibodies were diluted at a ratio of 1:1,000. Antibodies to β-actin (cat. no. sc-47778), ERK (cat. no. sc-94), and phosphorylated ERK (p-ERK; cat. no. sc-7383) were purchased from Santa Cruz Biotechnology, Inc. Antibodies to p-4E-BP1 (cat. no. 13396), 4E-BP1 (cat. no. 9452), p-Akt (cat. no. 9275), Akt (cat. no. 2920), p-p70S6K (cat. no. 9205) and p70S6K (cat. no. 9202) were purchased from Cell Signaling Technologies, Inc.

A375 and SK-Mel-28 cells were seeded in 24-well plates (20,000 cells/well). On the next day, a cell infection mixture was prepared: 10 MOI viruses per 1 ml of culture medium plus 2 µl of polybrene (4 µg/µl). The cell culture medium was then removed and replaced with 500 µl of the cell infection mixture. After 24 h, the infection mixture was replaced with fresh culture medium and the cells were incubated at 37˚C for another 24 h before the addition of medium with puromycin (1 µg/ml). Clonal selection lasted for 12 days with the puro-mycin-containing medium replaced once every 3 days. The scramble shRNA-infected clones were enriched via sorting of GFP‑positive cells using flow cytometry and collected in bulk, whereas the UCP2 knockdown (KD) clones were collected as single cells via GFP sorting using flow cytometry. The UCP2 KD clones were expanded and western blot analysis was performed for selection.

The melanoma cells were seeded in 96-well plates (6,000 cells/well) and cultured at 37˚C overnight. The cells were then treated using various cisplatin concentrations (15, 5, 1.67, 0.56, 0.19 and 0 µM). Cisplatin has been approved for the treatment of metastatic melanoma in the U.S. (18). Cisplatin (cat. no. 1134357, Sigma-Aldrich; Merck KGaA) was dissolved in phosphate‑buffered saline (PBS). After incubation at 37˚C for 48 h, cell viability was determined using the MTT assay (M5655; Sigma-Aldrich; Merck KGaA), with 10% MTT diluted in serum-free medium added to each well before a 4-h incubation at 37˚C. The MTT solutions were then replaced with dimethyl sulfoxide and the plates were shaken for 15 min at room temperature. The absorbance at 595 nm was then measured using a 96-well plater reader (Bio-Rad Laboratories, Inc.). All experiments were repeated at least three times.

A total of 10,000 melanoma cells were seeded in 96-well plates. On the next day, the culture medium was replaced with fresh medium containing the JC-1 dye (2 µg/ml) and incubated at 37˚C for 30 min. The medium was then removed and the cells were washed once using PBS. The fluorescence intensities were measured immediately using a fluorescence spectrophotometer (JC‑1 green: Ex=485 nm, Em=525 nm; JC-1 red: Ex=535 nm, Em=590 nm), and the ratio of JC-1 red vs. JC-1 green was used to evaluate the mitochondrial membrane potential. All experiments were repeated at least three times.

The Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (A22188, Molecular Probes; Thermo Fisher Scientific, Inc.) was used to measure intracellular hydrogen peroxide (H₂O₂) levels based on the manufacturer's instructions. The melanoma cells were grown in p100 dishes and collected via centrifugation (600 x g for 5 min) at room temperature. The cell pellets were then suspended in 400 µl PBS containing proteinase inhibitors (cat. no. sc-29130, Santa Cruz Biotechnology, Inc.). The cells were then sonicated and the lysates collected via centrifugation at 4˚C (12,800 x g for 30 min). Freshly prepared cell lysates were filtered through 10K cut‑off columns (82031‑348, VWR) and the product's absorbance was measured at 560 nm using a 96‑well plate reader. Background fluorescence was corrected by subtracting the value derived from the no-H₂O₂ control. All experiments were repeated at least three times.

Freshly prepared cell lysates were prepared as for the H₂O₂ assay and filtered through 10K cut-off columns before being used for both assays. The Lactate Colorimetric/Fluorometric Assay Kit (cat. no. K607-100, BioVision) was used to evaluate intracellular lactate levels, based on the absorbance at 570 nm measured using a 96-well plate reader. The ATP Luminescence Detection Assay Kit (cat. no. 700410, Cayman Chemicals) was used to evaluate intracellular ATP levels, based on the luminescence measured using a BioTek Multi-Mode microplate reader. All experiments were repeated at least three times.

Invasion ability was evaluated using Transwell inserts (cat. no. 3422, Corning, Inc.) that were coated with 50 µg/ml of Matrigel. A total of 50,000 melanoma cells were suspended in serum-free medium and 100 µl of the cell suspension was transferred into the inserts, whereas complete growth medium was added to the bottom wells. The cells were then cultured for 12 h at 37˚C before the inserts were removed, fixed in 10% neutral‑buffered formalin, and stained at room temperature using 0.5% crystal violet solution. All experiments were repeated at least three times.

The liquid overlay technique was used to grow melanoma spheroids (19). First, 96-well plates were coated with 50 µl of agar (1.25%) and 30 min later 25,000 melanoma cells (200 µl) were added to the wells. Spheroids were allowed to form and images were captured on the following day. The volume of the spheroids was calculated as follows: Volume=(4/3)π x b² x c, where b is the longest (semi-major) axis and c is the shortest (semi-minor) axis. All experiments were repeated at least three times.

Whole-cell lysates were prepared using the RIPA lysis buffer (cat. no. sc-24948A, Santa Cruz Biotechnology, Inc.) and collected via centrifugation at 12,800 x g at 4˚C for 30 min. Protein concentrations were determined using the Bradford method and 50 µg of the samples were denatured and loaded onto a 10% polyacryl-amide gel. After separation, the proteins were transferred onto a polyvinylidene fluoride membrane, which was blocked with 5% non-fat milk for 1 h followed by overnight incubation with the primary antibody at 4˚C on a shaker. The membrane was then washed three times using PBS/0.05% Tween 20 and incubated with a horseradish peroxidase (HRP)-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, cat. no. 111-035-003, dilution 1:2,500) for 1 h. All bands were detected using an ECL Western blot kit (Genesee, 20-302B). All experiments were repeated at least three times.

Paraffin-embedded sections of normal skin, nevus, or melanoma were dewaxed and gradually rehydrated before being immersed in an EDTA solution (pH 8.0) and heated using a pressure cooker for 2 min. After cooling to room temperature, the tissue sections were rinsed with PBS and normal goat serum was used to block non‑specific binding. The tissue sections were then incubated with a mouse anti-human UCP2 monoclonal antibody (1:200 dilution) overnight at 4̊C. After washing with PBS, the tissue sections were incubated with an HRP-conjugated secondary antibody for 30 min at room temperature. The Ultraview Red agent was added to distinguish positive staining from skin pigmentation, and red particles deposited in the cytoplasm were considered as a positive result. Hematoxylin and eosin staining was performed separately to evaluate the pathological characteristics.

The stained sections were evaluated via double-blind scoring based on a slightly modified version of the procedure described by Bosman et al (20). Five different fields were randomly selected, and histochemical scores were calculated according to the positive rate of tumor cells [P(i)] and the staining intensity [S(i)] as follows:

In this equation, the P(i) is scored as 0 (no positive cells), 1 (<10% positive cells), 2 (10-50% positive cells), or 3 (>50% positive cells). The S(i) was scored as 0 (no staining), 1 (light yellow staining), 2 (brownish yellow staining), or 3 (red staining). The mean score for all 5 fields was calculated, and the result was graded as negative (-, score 0-1), mild (+, score 2‑3), or strong (++, score ≥4).

The statistical analyses were performed using the χ² test or analysis of variance as appropriate. One-way analysis of variance followed by Tukey-Kramer adjustment was used to examine differences between multiple groups. All statistical analyses were performed using SPSS software, version 13.0 (SPSS Inc.), and the results were considered statistically significant at P‑values of <0.05.

Immunohistochemistry was used to evaluate UCP2 expression in the tissue specimens from the melanoma group (n=65), the compound nevus group (n=51) and the control group (n=49). As shown in Fig. 1, the red particles deposited in the cytoplasm indicate a positive result and the quantified results are presented in Table I. The UCP2 positivity rates were low in normal skin tissue (2.0%) and compound nevus tissue (5.9%), but relatively high in melanoma tissue (76.9%).

It was also evaluated whether UCP2 expression was correlated with Clark level and Breslow thickness (depth of invasion), lymph node infiltration and presence of ulceration. As summarized in Table II, UCP2 expression increased with the Clark level (P<0.001) and was significantly correlated with Breslow thickness (P=0.0235), but was not significantly correlated with lymph node infiltration status or the presence of ulceration.

As UCP2 was more highly expressed in melanoma tissues compared with nevus tissues, it was further evaluated whether inhibiting UCP2 expression could suppress melanoma cell growth and sensitize these cells to cisplatin. Two widely used melanoma cell lines, A375 and SK-Mel-28, were infected using the control or UCP2 shRNA-containing lentivirus. After antibiotic selection, control lentivirus-infected cells (LC) and two stable UCP2 KD clones were established in each cell line (Fig. 2A). The UCP2 protein levels were reduced by 40-50% in the A375 KD clones and by 50-70% in the Mel-28 KD clones. As shown in Fig. 2B, after growing for 60 h, the proportion of cell growth was 72.4 and 79.3% for the A375 clones, relative to the control cells. For the SK-Mel-28 cells, the proportion of cell growth was 81.2 and 73.8%, relative to the control cells.

The cell viability assay was used to test sensitivity to cisplatin, which is a common chemotherapeutic agent used for the treatment of melanoma (18). The UCP2 KD and control cells were treated using different concentrations of cisplatin for 48 h, and cell viability was measured using the MTT assay. As shown in Fig. 2C, the UCP2 KD cells were more sensitive to cisplatin at concentrations of 1.67 and 5 µM. For example, at a concentration of 1.67 µM, the viability of A375 UCP2 KD clones was ~30% (vs. 46% for the control cells) and the viability of SK-Mel-28 UCP2 KD cells was ~49% (vs. 64% for the control cells). These results indicate that UCP2 inhibition increased the sensitivity of melanoma cells to cisplatin.

As an uncoupling protein, UCP2 regulates the mitochondrial membrane potential, ATP synthesis and ROS generation (21). The JC-1 dye was used to examine the effects of UCP2 inhibition on the membrane potential (ΔΨm), and the UCP2 KD clones exhibited increases in the membrane potential of 40-65% for the A375 and SK-Mel-28 cells (Fig. 3A). However, the increased membrane potential did not result in increased ATP generation, with ATP levels decreasing by 30-60% in the A375 and SK-Mel-28 UCP2 KD cells (Fig. 3B). As shown in Fig. 3C, the H₂O₂ levels were also decreased by 25-40% in the A375 and SK-Mel-28 UCP2 KD cells (Fig. 3C).

Glycolysis is often enhanced in tumor cells for ATP production and anabolism, which leads to increased lactate levels (22). However, in the A375 and SK-Mel-28 UCP2 KD cells, the lactate levels were decreased by 20-30% (Fig. 3D). These results suggested that inhibiting UCP2 reduced the rate of glycolysis and ATP generation in melanoma cells.

Malignant melanoma is an invasive tumor, and the Matrigel invasion assay was used to examine whether UCP2 inhibition affected cell migration and invasion. As shown in Fig. 4A and B, after a 12-h incubation, only 60-75% of the A375 UCP2 KD cells had migrated through the Matrigel (vs. the control A375 cells) and only 65-85% of the SK-Mel-28 UCP2 KD cells had migrated (vs. the control SK-Mel-28 cells). The three-dimensional spheroid growth assay was performed to evaluate the tumorigenicity of the UCP2 KD cells. For the SK-Mel-28 cells, the UCP2 KD cells formed spheroids with ~40% of the volume of the control cell spheroids (Fig. 4C and D). For the A375 cells, one UCP2 KD clone formed smaller spheroids compared with the control, while the other clone formed larger spheroids, but with markedly lower density based on fluorescence intensity analysis. These results suggested that inhibition of UCP2 expression reduced the tumorigenicity of melanoma cells.

During melanomagenesis, Akt/mTOR signaling plays a key role in the regulation of cell proliferation, growth and apoptosis (23). Fig. 5A and B show the lower levels of phosphorylated Akt (Thr308) in both lines of UCP2 KD cells. As downstream effectors for mTOR, p70S6K and 4E-BP1 regulate cell growth and proliferation, and their activation is often associated with tumor development (24,25). Inhibition of UCP2 expression was found to be associated with significantly reduced levels of phosphorylated p70S6K (Thr389) and 4E-BP1 (Thr70) relative to the control cells (Fig. 5A, B and E). The expression of p70S6K was not significantly altered, although the expression of 4E-BP1 was increased in the A375 UCP2 KD cells, but not in the SK-Mel-28 UCP2 KD cells.

The ERK pathway also plays a crucial role in regulating melanoma cell proliferation, differentiation and apoptosis (26). The UCP2 KD clones in both cell lines exhibited significantly reduced levels of both phosphorylated ERK (Tyr204) and total ERK proteins, relative to the control cells (Fig. 5C and D); therefore, the p-ERK/ERK ratio was not decreased in the UCP2 KD clones (Fig. 5E). These results suggested that inhibition of UCP2 expression suppressed Akt/mTOR and ERK signaling in melanoma cells.