Reagents were purchased from the following sources: The primary rabbit antibodies against Arg-II (#55003), Sirt3 (#5490) and eNOS (#9571S) were purchased from Cell Signaling Technology (Allschwil, Swizerland); rabbit anti-iNOS (ab178945), nNOS (ab76067) and p-γH2A.X (phos-S139) antibody (ab81299) was purchased from Abcam (Cambridge, UK); The primary mouse antibody against Sirt1 (ab110304) were purchased from Abcam (Cambridge, UK), the mouse antibody against Arg-I (sc-18351) was obtained from Santa-Cruz (Schaffhausen, Switzerland), the mouse anti-tubulin (Art.T5168) was from Sigma-Aldrich (Schaffhausen, Switzerland); Alexa Fluor® 594 conjugated-goat anti-rabbit IgG (H + L) secondary antibody (A-11012) and goat anti-mouse IgG (H + L) (A-21057) as well as MitoSOX (M36008) and Wheat Germ Agglutinin (WGA) (W11261) were purchased from ThermoFisher Scientific (Basel, Switzerland). Mito-TEMPO (SML0737) and crystal violet (C0775) were purchased from Sigma-Aldrich (Schaffhausen, Switzerland). Sirt3 siRNA (SC61555), and control siRNA (SC36869) were purchased from Santa Cruz (Heidelberg, Germany). Sirt3 Flag plasmid (pCDNA3.1+/CMV-Sirt3-Flag) was purchased from Addgene (13,814). Trans-well Insert chambers with 8.0 µm Coloured PET Membrane (351,152) were bought from Corning (Arizona, United States).

The Me276 cells (a human melanoma cell line derived from lymph node of a patient with melanoma metastasis (Caballero et al., 2010)) and A549 cells [adenocarcinoma human alveolar basal epithelial cells (Giard et al., 1973)] was purchased from ATCC and were maintained in RPMI1640 (PAN-Biotech, Aidenbach, Germany), supplemented with 5% heat-inactivated foetal bovine serum (FBS) (Gibco), 100 U/mL penicillin and 100 μg/mL streptomycin (Gibco, Zug, Switzerland). Cells were cultured at 37°C in a humidified incubator containing 5% CO₂ and the culture media was changed every 2 days. Hypoxic conditions were achieved by placing the cultures dishes or plates in a Coy In Vitro Hypoxic Cabinet System (The Coy Laboratory Products, Grass Lake, MI USA) at 1% O₂ with 5% CO₂ and N₂ as balance (Liang et al., 2019).

arg-ii ^−/− A549 cells were generated using CRISP/Cas9 technologies as described (Zhu et al., 2023). arg-ii ^−/− Me276 cells were generated using the same method. Briefly, sgRNA targeting human arg-ii (the top strand of the sgRNA that recognizes the target DNA region of human arg-ii: GGGACTAACCTATCGAGA) was cloned into pSpCas9(BB)-2A-Puro (PX459) V2.0 (Plasmid #62988, Addgene) to generate pSpCas9(BB)-2A-Puro (PX459)-U6/sgRNA-harg-ii. Transfection of pSpCas9(BB)-2A-Puro (PX459)-U6/sgRNA-harg-ii was performed using Lipofectamine™ 3,000 Transfection Reagent (L3000008, Invitrogen™) according to the manufacturer’s protocol. Briefly, Me276 cells were plated in 6-cm dish at a density of 1 × 10⁶ cells 24 h before transfection. Per 1 × 10⁶ cells, diluted plasmid DNA (5 μg, diluted with P3000™ Reagent) and diluted Lipofectamine™ 3,000 Transfection Reagent were mixed at a 1:1 ratio and incubated at room temperature for 15 min. The DNA-lipid complex was then added to the cells. To select the sgRNA-positive cells, 48 h post transfection, cells were treated with puromycin (1 μg/mL) for 72 h until all non-transfected cells died. Puromycin-resistant cells were then allowed to recover in medium without puromycin for 1 week before seeding single cells into 96-well plate by dilution method. Single clones were then expanded and screened for arg-ii ^−/− by immunoblotting.

Recombinant adenovirus (rAd) expressing arg-ii driven by CMV promoter (rAd/CMV-arg-ii) and the relevant control vector (rAd/CMV-lacz) were produced as previously described (Yepuri et al., 2012). To overexpress arg-ii, cells were seeded in 6-well or 12-well plate at a density of 3*10⁵/mL or 5*10⁴/mL, respectively. After overnight culture, cells were transduced with rAd/CMV-lacz as control or rAd/CMV- arg-ii at the titers of 100 MOI (multiplicity of infection) and cultured in complete medium for 48 h before the experiments were performed.

Melanoma cells were seeded in 6-well or 12-well plate at a density of 3*10⁵ cells or 5*10⁴ cells, respectively. After overnight culture, siRNA targeting Sirt3 or scrambled control siRNA, pcDNA3.1+/CMV-Sirt3-Flag or empty control plasmid pcDNA3.1+/CMV were transfected by using Lipofectamine 3,000 (Thermo Fisher, Rockford, IL, United States) according to the manufacturer’s instructions. 48 h post transfection, experiments were performed. Gene silencing efficacy as well as overexpression was assessed by immunoblotting analysis.

The wild type (wt) and arg-ii ^−/− Me276 or A549 cells were seeded at 500 cells/well in 6-well culture plate in duplicate. For arg-ii-overexpression experiments, 200 cells/well (6-well) were seeded 48 h post transduction. After seeding, the cells were kept at 37°C in 5% CO₂ incubator for 12 days and culture media were changed every 3 days. After 12 days, cells were washed with 1x PBS and then fixed with 10% Neutral Buffered Formalin (NBF) buffer for 15 min, followed by staining with 0.5% crystal violet buffer for 15 min. The images of total colonies in each well were taken by Canon G16 camera. The number of visible colonies after staining with crystal violet was counted manually after circling individual colonies.

Since some very big colonies may be resulted from a merge of more than one colony, the total area of colonies in a plate was also quantified by ImageJ software and presented along with the colony number in the same graphics.

The cell number counting was performed 48 h post transduction. For experiment with TEMPO, cells were treated with TEMPO (10 μmol/L) 48 h post transduction for 1 h and further cultivated for 24 h in the absence of TEMPO and then subjected to collection of cell suspension by trypsinization and cell counting. Cell number was counted with the Neubauer cell counting chamber (Blaubrand) and expressed as cell number/well (Park et al., 2024).

In some figures (Figure 3C; Supplementary Figure S4B), the number of cells/mm² was obtained by first counting cell number in the DAPI-stained confocal image manually, and then quantified by ImageJ software (US National Institutes of Health) (Venter and Niesler, 2019).

Except for colony-formation assay, cells were treated with TEMPO (10 μmol/L) for 1 h. After removal of TEMPO, the cells were further cultivated for 24 h in the absence of TEMPO and then subjected to the relevant assays as described in individual figures. For colony-formation assay, TEMPO was added to the medium when cells were attached after seeding. Fresh media containing TEMPO (10 μmol/L) was changed every 3 days for a total period of 12 days.

Cell lysate preparation, SDS-PAGE and immunoblotting, antibody incubation, and signal detection were performed as described previously (Ming et al., 2012). Briefly, cells were lysed in a lysis buffer containing the protease inhibitors cocktail (B14002, B15002, Biotool). After three cycles of freeze-thaw in liquid nitrogen, cell debris and nuclei were removed by centrifugation (Sorvall Legend Micro 17R) at 12,000 × g for 15 min at 4°C. The protein concentration of the supernatant was determined by a Lowry protein assay (500-0116, Bio-Rad). Proteins from the samples were loaded on the sodium dodecyl sulphate-polyacrylamide gel (SDS-PAGE), separated by electrophoresis, and transferred to PVDF membranes (Millipore). Ponceau staining and tubulin were used as the loading control. The resultant membranes were first blocked with skimmed milk (5%) for 1 h, incubated with the designated primary antibody (in PBS-Tween supplemented with 5% milk) overnight at 4°C. Membranes were washed with 5% milk buffer three times and further incubated with anti-mouse (Alexa fluor 680 conjugated) or anti-rabbit (IRDye 800 conjugated) secondary antibody for 1.5 h. Dilutions of each primary antibody are presented in Supplementary Table S1. The signals were detected using Odyssey Infrared Imaging System (LI-COR Biosciences). Quantification of the signals was performed using Image Studio Lite version 5.2.

Mitochondrial superoxide generation was assessed by MitoSOX™ Red Mitochondrial Superoxide Indicator (Thermo Fisher) according to the manufacturers’ instructions. Briefly, live cells were incubated with medium containing MitoSox (5 μmol/L) for 10 min in cell culture incubator. After gentle washing 3 times with PBS (5 min each time), the cells were fixed with 10% NBF for 10 min, permeabilized with 0.1% Triton X-100 for 15 min, washed with PBS 3 times followed by counterstaining of the nuclei with DAPI for 5 min. Images were taken under the Leica TCS SP5 confocal laser microscope with ×63 objectives. The intensity of the MitoSox fluorescence signals per cell was quantified by Leica X Life Science Microscope software (Leica Microsystems).

Staining of mammalian cell plasma membranes was performed using WGA Alexa Fluor™ 488 Conjugate according to the instructions from the manufacturer (ThermoFisher). In brief, cells were incubated with diluted WGA Alexa Fluor™ 488 Conjugate (10 μg/mL) medium for 15 min in the incubator and then fixed with 10% NBF buffer for 10 min, followed by permeabilization with 0.1% triton X-100 for 15 min and counterstaining of the nuclei with DAPI for 5 min. Signals were detected and photographed by the Leica TCS SP5 confocal laser microscope with ×63 objectives.

Nuclei were stained with DAPI and images were captured under a Leica fluorescence microscope. Nuclear deformation was assessed by observing the morphology of the DAPI-stained nuclei. To better observe the shape of the nucleus, blue colour of DAPI was converted to white on dark background using Leica software. The percentage of normal ovoid or kidney-shaped nuclei (1 lobe) and deformed multilobed nuclei with ≥2-lobes (equal to or more than 2 lobes) in the field of view for each image was counted and calculated. Five fields of each image were analysed and the average number of cells with deformed multilobed nuclei was presented as percentage of the total cell number.

Cells were grown on sterile glass slides in Petri dishes. After gently washing with PBS, cells were fixed with 10% NBF for 10 min and permeabilized with 0.1% Triton X-100 solution for 15 min. Cells were then incubated in 1% blocking solution (10% goat serum in 1% BSA solution) for 1 h at room temperature followed by incubation with anti-p-γH2AX antibody (1:100) at 4°C overnight. After three times of washing with PBS, the secondary antibody was added and incubated at room temperature for 1.5 h. After washing with PBS three times, cells were then counter-stained with DAPI for 5 min. Immunofluorescence signals were visualized under Leica TCS SP5 confocal laser microscope.

The comet assay was performed according to the manufacturer’s instructions (ab238544, Abcam Plc., Cambridge, United Kingdom). Briefly, wild-type (wt) and arg-ii ^−/− Me276 melanoma cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. 7,000 Me276 melanoma cells were collected for the comet assay. The cells were washed once with phosphate-buffered saline (PBS) and then mixed with Comet Agarose at a 1:10 ratio (v/v). A total of 75 μL of the cell-agarose mixture was immediately transferred onto the top of the Comet Agarose Base Layer. The cells were lysed at 4 °C, followed by electrophoresis in a cold alkaline electrophoresis buffer for 15 min at 1 V/cm. The comet slides were rinsed twice with double-distilled H₂O for 5 min each and fixed with 70% cold ethanol for 5 min. After air-drying, the slides were stained with Vista Green DNA Dye and imaged using a fluorescence microscope. Cells positive for comet tails were quantified as DNA-damaged cells.

5 × 10⁴ cells per well were plated in the trans-well insert (Corning® FluoroBlok™ 24-Multiwell Insert with 8.0 µm Colored PET Membrane) with complete medium with 5% FBS in both lower and upper chamber. 2 h post plating, 10% FBS was added into the lower chamber. After 48 h, inserts were transferred to a clean 24-well plate, washed with 1x PBS for 2 min and cells were fixed with 10% NBF buffer for 10 min. After two times washing with PBS (5 min per wash), cells were permeabilized with 0.1% Triton X-100 for 15 min followed by staining of the nuclei with DAPI for 5 min in darkness. The DAPI signals from cells that have migrated to the bottom of the trans-well insert were then detected by a bottom-reading fluorescence plate reader (BioTek Cytation 5). The results were analysed by Gen5 software and presented as migrated cells/mm².

Gene Expression Analysis: the expression of arg-ii was analyzed using the open-source database GEPIA (Gene Expression Profiling Interactive Analysis), an online platform that processes RNA sequencing expression data from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases. Box plots were generated using the platform’s differential expression analysis and profiling features. In this study, the GEPIA database (http://gepia.cancer-pku.cn/index.html) was utilized to investigate arg-ii gene expression across multiple cancer types (Tang et al., 2017).

Survival Analysis: to evaluate the prognostic significance of arg-ii expression, Kaplan-Meier survival plots were generated using data from all stage lung cancer patient samples. The analysis was performed using the KMplot online tool (https://kmplot.com/analysis/index.php?p=home), which integrates survival data from multiple cancer studies (Gyorffy, 2024a; Gyorffy, 2024b).

In all experiments, n indicates the number of individual experiments. The Shapiro-Wilk test (that determines normality when n ≥ 3) was used first to determine whether the data deviated from Gaussian distributions. Since all data are normally distributed, parametric statistical analysis was performed. Student’s t-test for unpaired observations was applied to compare the averages of two groups and one-way analysis of variance (ANOVA) was used to analyze the difference between the means of more than two groups. All data are expressed as standard error of the mean (SEM). GraphPad Prism software version 9.0 was used for the statistical analysis. Differences were considered statistically significant at p < 0.05.

To determine the effects of Arg-II on tumour cell colony formation which is an important process for cancer progression, human melanoma cell line Me276 lacking arg-ii ^−/− (arg-ii ^−/− ) was generated by CRISPR/Cas9 technology. Immunoblotting confirmed the Arg-II deficiency in the cells (Figure 1A). Twelve days after cell seeding in culture, arg-ii ^−/− Me276 cells showed fewer colonies and smaller colony area as compared to the wt control cells (Figure 1B). This result is consistent with previous data showing a reduced cell number in cells transduced with rAd/U6-arg-ii (to silence arg-ii) (Yu et al., 2020). Moreover, Me276 cell migration activity as examined by trans-well migration assay was significantly reduced by arg-ii knockout (arg-ii ^−/− ) (Figure 1C). To further confirm the role of Arg-II in colony formation and migration of the melanoma cells, arg-ii gene was overexpressed in the Me276 cells by adenoviral transduction, which was verified by immunoblotting 48 h post transduction (Figure 1D). In agreement with arg-ii knockout experiments, arg-ii overexpressing Me276 cells formed more colonies and bigger colony area (Figure 1E) and exhibited enhanced migration activity (Figure 1F) as compared to the wt control cells. These results demonstrate an important role of Arg-II in promoting melanoma cell growth and migration. Similar results were also demonstrated with another cancer cell line, i.e., the lung carcinoma cell A549 (Supplementary Figure S1).

We further investigated the effects of arg-ii-knockout on nuclear morphology and DNA damage on the tumour cells in view that these are indicative of tumour metastasis and malignancy. A decrease in number of cells with multi-lobed nuclei was observed in arg-ii ^−/− melanoma cells when compared with the wt cells (Figure 2A). To further confirm that Arg-II induces nuclear deformation and to better demonstrate whether the heteromorphic nuclei are in individual cells, cells were stained with DAPI and WGA to visualize nuclei and cell border, respectively (Figure 2B). For evaluation, the percentage of ovoid or kidney-shaped (1-lobed) nuclei and heteromorphic multi-lobed nuclei in the field of view for each image was counted and calculated. Arg-ii ^−/− cells revealed significantly lower percentage of cells with heteromorphic multi-lobed nuclei (Figures 2A, B). In line with these results, arg-ii-overexpressing cells showed more multi-lobed nuclei (Figures 2E, F). Of note, arg-ii-overexpression caused more severe nuclear deformation as evidenced by more cells with 3- and 4-lobed nuclei (Figure 2F), while wt cells show mostly only 2-lobed nuclei (Figure 2B).

Along with these results, phosphorylated histone γ-H2AX on serine 139 (p-γH2AX-S139), a sensitive molecular marker of DNA damage, was detected at baseline levels in wt cells and decreased in arg-ii ^−/− melanoma cells (Figures 2C, D). Moreover, the decreased DNA damage in arg-ii ^−/− melanoma cells was also evident by comet assay revealing a decresed percentage of comet tail-positive cells (Supplementary Figure S2). In line with these data, the levels of p-γH2AX-S139 were enhanced in melanoma cells overexpressing arg-ii (Figures 2G, H). Similar results were also obtained in A549 cells (Supplementary Figure S3). These results demonstrate that the Arg-II-induced nuclear deformation is linked to increase in migration and DNA damage, which increases the cancer cell malignancy.

Next, we wish to investigate how Arg-II affects tumour growth and malignancy. Since one of the best characterized biological functions of Arg-II is attributable to its regulation of nitric oxide synthase (NOS)-NO pathway, we examined the expression of eNOS, iNOS and nNOS as well as another isoform of arginase (Arg-I) in melanoma and lung cancer cells. Neither eNOS, iNOS, nNOS nor Arg-I could be detected by immunoblotting in the human melanoma or A549 lung cancer cells (data not shown). This result implies that the effects of Arg-II on melanoma or lung cancer cell growth and malignancy are independent of Arg-I or NOS-NO pathway.

Since previous studies provide evidence that mtROS generation is enhanced by Arg-II in various cell types (Ming et al., 2012; Ren et al., 2022), the role of mtROS in Arg-II-mediated effects on the cancer cells were then investigated. Indeed, the production of mtROS (mitochondrial superoxide) in the melanoma cells as demonstrated by MitoSOX staining, was significantly reduced in arg-ii ^−/− cells as compared to the wt melanoma cells (Figures 3A, B), which was accompanied by decrease in cell number (Figure 3C). Conversely, overexpression of arg-ii enhanced mtROS levels and the cell number, which was inhibited by MitoTEMPO (10 μmol/L), a mitochondria-targeted antioxidant and a specific scavenger of mitochondrial superoxide (Figures 3D–G). Moreover, the multi-lobed nuclei formation (Figures 3D, F) and DNA damage (Figures 3H, J) as well as colony formation and migration of the cells overexpressing arg-ii (Figure 4) all were inhibited by MitoTEMPO. Of note that MitoTEMPO did not affect Arg-II levels (Figure 3I). These results were also observed in the lung cancer cells (Supplementary Figures S4, S5). Together, these results demonstrate that Arg-II promotes melanoma and lung cancer cell growth and malignancy through enhanced mtROS production.

Sirt3, a mitochondrial sirtuin, has been reported to exert antioxidant and anti-cancer effects (Bell et al., 2011). In accordance with its anti-cancer effects, Sirt3 was enhanced in arg-ii ^−/− melanoma cells (Figures 5A, B). Conversely, overexpression of arg-ii decreased Sirt3 levels (Figures 5D, E). In contrast to Sirt3, Sirt1 was not affected either by arg-ii knockout or by arg-ii overexpression (Figures 5A, C, D, F). It’s to note that silencing sirt3 (Figures 5G, H) did not significantly affect Arg-II protein levels (Figures 5G, I), suggesting Arg-II is the upstream of Sirt3. Similar results were obtained in A549 lung cancer cells (Supplementary Figure S6). These results demonstrate that Arg-II is the upstream of Sir3 and suppresses Sirt3 expression.

In supporting the recognized anti-oxidant and anti-cancer effects of Sirt3, silencing sirt3 in wt melanoma cells enhanced mtROS production along with increased melanoma cell growth, which was prevented by inhibition of mtROS with MitoTEMPO (Figures 6A–C). These results confirm that Sirt3 inhibits cell growth by preventing mtROS generation. In parallel, silencing of Sirt3 in the melanoma cells resulted in a significant nuclear deformation (Figures 6D, E), an increase in migration (Figures 6F, G) and DNA damage (Figures 6H–J). Of note, mtROS didn’t affect Arg-II-induced suppression of Sirt3 (Supplementary Figure S7).

To demonstrate that Arg-II promotes mtROS production resulting in increased cancer cell malignancy indeed through suppression of Sirt3, sirt3 was co-overexpressed with arg-ii in melanoma cells. sirt3 overexpression prevented arg-ii overexpression-indued increase in mtROS generation (Figures 7A, B), cell number (Figures 7A, C), the percentage of cells with multi-lobed nuclei (Figures 7A, D). Moreover, Arg-II-induced DNA damage as monitored by p-γH2AX-S139 (Figures 7E, F), the increase in colony formation (Figure 8A) and migration (Figure 8B) induced by arg-ii overexpression were prevented by sirt3 co-overexpression. Similar results were obtained in A549 lung cancer cells (Supplementary Figure S8). These data provide evidence that Arg-II suppresses Sirt3 leading to enhanced mtROS production, which in turn promotes melanoma growth and malignancy.

Hypoxia is a key feature of many solid tumours and is associated with increased metastasis and poor survival in patients (Hockel and Vaupel, 2001). Given that hypoxia is a strong stimulus upregulating Arg-II in various cells (Ren et al., 2022; Liang et al., 2019; Krotova et al., 2010; Prieto et al., 2011; Chen et al., 2014; Liang et al., 2021), we wish to examine whether hypoxia enhances melanoma cell malignancy by upregulating Arg-II. As observed in the other cell types, hypoxia also enhanced Arg-II in wt Me276 melanoma cells, which was accompanied by a decrease in Sirt3 (Figures 9A–C), increase in nuclear deformation (Figure 9E) and DNA damage (Figure 9G). Importantly, knockout of arg-ii (arg-ii ^−/− ) not only enhanced Sirt3 under normoxic condition (Figure 9D), but also reduced hypoxia-induced decrease in Sirt3 (Figure 9D), increase in cells with multi-lobed nuclei (Figure 9F), DNA damage (Figures 9G–I). The results implicate that hypoxia suppresses Sirt3 and induces nuclear deformation and DNA damage through Arg-II-Sirt3 cascade.

To strengthen translational implications, we analysed public datasets. Analysis of the open-source database GEPIA (Gene Expression Profiling Interactive Analysis) (http://gepia.cancer-pku.cn/index.html) reveals that arg-ii mRNA levels are elevated in various tumor samples compared to normal counterparts, including skin cutaneous melanoma (SKCM), lung adenocarcinoma, and lung squamous cell carcinoma (Supplementary Figure S9A). Moreover, Kaplan-Meier survival curves for SKCM patients shows that higher arg-ii levels are associated with lower survival rate (please see the figure on website https://www.proteinatlas.org/ENSG00000081181-ARG2/cancer/melanoma#SKCM_TCGA). Similar results were observed with Kaplan-Meier survival curves for lung cancer patients of all stages obtained from the open-source website (https://kmplot.com/analysis/) (Supplementary Figure S9B). These clinical data support our findings and further imply that targeting Arg-II would be a potential therapeutic strategy for melanoma and lung cancer treatment.