MARCH5 Overexpression Contributes to Tumor Growth and Metastasis and Associates with Poor Survival in Breast Cancer

Journal name: Cancer Management and Research Article Designation: Original Research Year: 2019 Volume: 11 Cancer Management and Research Dovepress Running head verso: Tang et al Running head recto: MARCH5 in tumor growth and metastasis open access to scientific and medical research DOI: http://dx.doi.org/10.2147/CMAR.S190694

Open Access Full Text Article Original Research MARCH5 overexpression contributes to tumor growth and metastasis and associates with poor survival in breast cancer

Haili Tang1,* Background: Human MARCH5 is a mitochondrial localized E3 ubiquitin-protein ligase that Shujia Peng1,* is critical for the regulation of mitochondrial dynamics. A body of evidence has indicated the Yanming Dong1,* close links between unbalanced mitochondrial dynamics and cancers. However, the expression, Xiaojun Yang1 biological functions, and prognostic significance of MARCH5 in breast cancer (BC) have not Ping Yang1 been determined. The mRNA and protein expressions of MARCH5 were evaluated Lin Yang1 Materials and methods: by quantitative real-time PCR and Western blot analysis in BC cell lines and tumor tissues. Bing Yang2 Clinical prognostic significance of MARCH5 was assessed in 65 patients with BC. The bio- 1 Guoqiang Bao logical functions of MARCH5 were determined by in vitro cell proliferation, apoptosis, cell 1Department of General Surgery, cycle, migration and invasion assays, and in vivo tumor growth and metastasis assays through Tangdu Hospital, The Fourth Military knockdown or overexpression of MARCH5 in BC cells. In addition, the underlying mechanisms Medical University, Xi’an, Shaanxi 710038, China; 2Department of by which MARCH5 regulated BC cell growth and metastasis were explored. General Surgery, JingYang Country Results: MARCH5 was substantially upregulated in BC cells mainly due to the downregulation Hospital, XianYang, Shaanxi 713700, China of miR-30a, which contributed to the poor survival of BC patients. MARCH5 promoted the growth and metastasis of BC cells both in vitro and in vivo by inducing G1–S cell cycle arrest *These authors contributed equally to this work and epithelial–mesenchymal transition. Mechanistic investigations revealed that the oncogenic effect of MARCH5 was mainly mediated by increased mitochondrial fission and subsequent ROS production in BC cells. Conclusion: Our findings demonstrate that MARCH5 plays a critical oncogenic role in BC cells, which provides experimental evidence supporting MARCH5 as a potential therapeutic target in BC therapy. Keywords: MARCH5, proliferation, invasion, prognosis, BC

Introduction Breast cancer (BC) is the most common malignancy and the second leading cause of death from cancer among women.1 Although significant progress has been made in early BC treatment, the prognosis of patients with metastasis continues to be poor. Thus, identification of novel molecular mechanisms leading to metastatic growth of GC cells is critical for the development of new therapeutic targets for treatment of Correspondence: Guoqiang Bao this malignancy. Department of General Surgery, Tangdu Mitochondria are double-membrane-bound organelle that are vital for ATP Hospital, The Fourth Military Medical University, 1 Xinsi Road, Xi’an 710038, production and macromolecule biosynthesis in most eukaryotic organisms. A Shaanxi, China significant association has been observed between the dysfunction of mitochondrial Tel/fax +86 29 8477 7731 Email [email protected] and many types of cancer.2–4 Investigation of novel oncogenic regulators involved in

submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 201–215 201 Dovepress © 2019 Tang et al. This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at https://www.dovepress.com/terms. php and incorporate the Creative Commons Attribution – Non Commercial (unported, v3.0) License (http://creativecommons.org/licenses/by-nc/3.0/). By accessing the work http://dx.doi.org/10.2147/CMAR.S190694 you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms (https://www.dovepress.com/terms.php). Tang et al Dovepress the dysfunction of mitochondria become imperative and Fisher Scientific, Waltham, MA, USA) following the manu- urgent, which will provide insights into the molecular basis facturer’s protocols. The following sequences were targeted of cancer and therapeutic targets for treatment of this disease. for human MARCH5 siRNA: MARCH5#1: 5′-GCA- MARCH5, also known as membrane-associated ring CUUGGGAGUAAUUUGA-3′; MARCH5#2: 5′-GCACAC- finger (C3HC4) 5, is an E3 ubiquitin-protein ligase that is GUGUCCGAUUUAU-3′. In addition, synthetic miR-30a localized in the mitochondrial outer membrane, which is precursor (Thermo Fisher Scientific) was used. For generation involved in the regulation of mitochondrial morphology of MARCH5 overexpression vectors, the coding sequences through ubiquitination of dynamin-related protein (DMN1L), of MARCH5 were amplified from the cDNA of MCF-7 cells mitofusin-1 (MFN1), and mitofusin 2 (MFN2) that are pivotal and cloned into the pcDNA™3.1(C) vector (Thermo Fisher regulators of mitochondrial fission and fusion.5,6 Recently, Scientific). For generation of shRNA expression vectors a body of evidence has indicated the close link between targeting MARCH5, the pSilencer™ 3.1-H1 puro vector unbalanced mitochondrial fission/fusion and cancer.7,8 The (Ambion, Austin, TX, USA) was used. Transfection was aberrant expressions of DRP1, MFN1, and MFN2 have been applied using Lipofectamine 2000 (Thermo Fisher Scientific) demonstrated in human cancers of lung,9 liver,10 and blad- according to the manufacturer’s instruction. der.11 In addition, a few recent studies have implicated the involvement of disruption of mitochondrial fission in BC Real-time PCR analysis progression.12–14 However, the expression, biological effects, Total RNA was extracted from BC cell lines and tissue and then and prognostic significance of MARCH5, a critical regulator reverse transcribed into cDNA using random primers. Real- of mitochondrial fission and fusion, in tumorigenesis and time PCR (RT-PCR) was performed by SYBR Green master development are unknown, especially in BC. mixture (Thermo Fisher Scientific) on a Corbett 6200. For the In this study, we will first investigate the expression pat- detection of miR-30a, the U6 snRNA was used as endogenous terns and prognostic significance of MARCH5 in BC. More control. Primer sequences are listed in Table S1. importantly, the biological functions and their underlying molecular mechanisms will be explored in depth in this Western blot analysis malignancy. RIPA buffer was used for whole cell extracts. Equal amount of protein was separated on SDS-PAGE and transferred onto Materials and methods polyvinylidene fluoride membranes. After blocking with 5% BC cell lines and tissue sample collection milk, membranes were immunostained with specific primary All human BC cell lines of MDA-MB-231, MDA-MB-468, antibody and secondary antibody. The reactions were finally MDA-MB-435, MCF-7, T47D, and nonmalignant human breast detected by an enhanced chemiluminescence system. Primary cell lines MCF 10A and MCF 12A were purchased from the antibodies and their dilutions are listed in Table S2. ATCC. Cells were cultured in L-15, DMEM, or RPMI-1640 medium supplemented with 10% fetal bovine serum (HyClone). Immunohistochemistry analysis In addition, 65 primary BC tumor tissue samples were Immunohistochemistry (IHC) analysis was performed using collected during surgical resection at the Tangdu Hospital of an IHC detection kit (Thermo Fisher Scientific). All sections the Fourth Military Medical University in Xi’an, China. All were treated with boiling citrate buffer (pH=6) under pressure samples were histologically confirmed as BC, and written to unmask epitopes, followed by incubation with primary informed consent was obtained from each patient. The study antibody. Color was developed using DAB for 3 minutes, protocol was approved by the Research Ethics Committee of and then hematoxylin was used for counterstaining. Finally, Tangdu Hospital of the Fourth Military Medical University the expression of MARCH5 was evaluated by assigning the and performed in compliance with the Declaration of staining intensity and percentage of positive cells. Helsinki. IHC staining intensities were scored by two investiga- tors. The staining of MARCH5 was evaluated under a light Knockdown and forced expression of microscope at a magnification of× 100. The staining intensity target genes was graded on a scale of 0–3 (0 for no staining, 1 for weak Cells were transfected with siRNA (synthesized by GenePh- staining, 2 for moderate staining, and 3 for strong staining). arma, Shanghai, China) against MARCH5 or expression Percentages of the stained area were also scored (0 if 0%, vectors for MARCH5 using Lipofectamine 2000 (Thermo 1 if <25%, 2 if 25%–50%, 3 if 51%–75%, and 4 if >75%).

202 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

Points for intensity and percentage were calculated to obtain cal dish. After an overnight culture, cells were fixed in 4% a final semiquantitative IHC score. Median values of the IHC paraformaldehyde (Sigma-Aldrich Co., St Louis, MO, USA) score were chosen as the cutoff point for distinguishing the and incubated with a MitoTracker Green probe (Molecular low and high MARCH5 expression levels. Probes) for 2 hours. The immunofluorescence images were obtained by a confocal microscope (Olympus Corporation, Cell viability and colony formation assays Tokyo, Japan). Mitochondrial length was determined with the MTS assay was used for determination of cell viability ImageJ software (NIH, Bethesda, MD, USA). (Promega Corporation, Fitchburg, WI, USA, G3581) based on the manufacturer’s instructions. Briefly, BC cells were seeded Mouse xenograft model onto 6-well plates at a density of 1,000 cells/well. Cell viability The expression vector of shMARCH5 was stably trans- was determined based on the absorbance of 490 nm at 0, 24, fected into MDA-MB-231 cells for establishment of stable 48, 72, 96, and 120 hours, respectively, after cells were seeded. MARCH5 knockdown BC cells. Then, cells were subcuta- For the detection of colony formation, 1,000 BC cells neously injected into the dorsal flank of Balb/c nude mice were seeded onto 6-well plates and cultured for approximately (six/group). Tumor volume was monitored every 3 days for 5 2 weeks. After that, colonies were fixed with 70% ethanol and weeks. Mice were sacrificed, and the volume and wet weight stained with crystal violet. Finally, the colonies were counted. of tumors were measured. All animal studies were approved by the Institutional Animal Care and Use Committee of the Apoptosis and cell cycle assays Fourth Military Medical University. Animal experiments Flow cytometry was used for cell cycle and apoptosis analyses. were carried out in accordance with the UK Animals (Scien- For cell cycle assay, BC cells were fixed in 70% cold ethanol tific Procedures) Act, 1986 and associated guidelines. at –20°C for 12 hours and incubated in 1 mL staining solution with RNase (50 U/mL) and propidium iodide (50 µg/mL) for Tail vein metastatic assay 30 minutes at 4°C. Cell cycle profiles were analyzed by a flow A total of 1×106 BC cells with different MARCH5 expression cytometry (Beckman, Fullerton, CA, USA). For apoptosis levels were injected into the Balb/c nude mice (six/group) assay, an Annexin V (FITC-conjugated) apoptosis kit (F-6012, through the tail vein. The mice were sacrificed 8 weeks after US Everbright Inc , SuZhou, China) was used following the injection, and the lungs were dissected and stained with H&E. manufacturer’s instructions. Briefly, binding buffer, Annexin V-FITC, and propidium iodide solution were added to BC cells Statistical analysis and incubated for 20 minutes at room temperature. Cell apop- SPSS version 17.0 software was used for statistical analysis. totic profiles were analyzed by a flow cytometry (Beckman). The results are shown as the mean ± SEM, and P-value <0.05 was considered statistically significant. Student’st -test was Wound healing and Matrigel invasion used for comparisons between two groups. Pearson correla- assays tion analysis was used for correlations between measured Cell migration was assessed by the wound healing assay. variables. The overall survival (OS) and recurrence-free Briefly, BC cells were collected and seeded into 6-well plates. survival were determined by the Kaplan–Meier method. When cells grow to 90% confluence, a scratch with a plastic pipette tip was placed in the middle of the wells, followed Results by washing with PBS. The scratched BC cells were photo- MARCH5 is upregulated in BC cells and graphed after 0 and 24 hours through a light microscope. contributes to worse prognosis For Matrigel invasion assay, extracellular matrix was First, the expression of MARCH5 was detected in five BC coated onto the upper surface of 24-well transwell chamber cell lines (MCF-7, T47D, MDA-MB-231, MDA-MB-468, (BD Companion Plate). Then, 1×105 BC cells were seeded and MDA-MB-435) and two nonmalignant human breast and incubated for 48 hours. Penetrated cells were fixed with cell lines (MCF-10A and MCF-12A) by quantitative 4% paraformaldehyde and stained by crystal violet. real-time PCR (qRT-PCR) and Western blot analysis. MARCH5 expression was significantly higher in five BC Mitochondrial morphology analysis cell lines when compared with nonmalignant breast cell lines MitoTracker Green staining was used for mitochondrial mor- (Figure 1A, B). To validate the results from BC cell lines, we phology analysis. Briefly, BC cells were seeded into a confo- determined the expression level of MARCH5 in 18 paired

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 203 Dovepress Tang et al Dovepress

levels 5 2.0 4 1.5 3 A A 1.0 2 7 1 0.5 of MARCH5 0 MCF-10 MCF-12 MCF- MDA-MB-231MDA-MB-468MDA-MB-435T47D 0.0

7 MARCH5/β-Actin 7

Relative mRNA F- MARCH5 F- T47D T47D MC β-Actin MC MCF-10AMCF-12A MCF-10AMCF-12A DA-MB-231DA-MB-468DA-MB-435 DA-MB-231DA-MB-468DA-MB-435 M M M M M M

C D T1 N1 T2 N2 T3 N3 T4 N4 T5 N5 T6 N6

4 MARCH5 3 β-Actin 2 2 1 1 expression T7 N7 T8 N8 T9 N9 T10 N10 T1 N1 T12 N12

T/P ) 1 T/P )

2 MARCH5 2

(Log 0 0 β-Actin (Log –1 MARCH5 protein levels

MARCH5 mRNA T13 N13 T14 N14 T15 N15 T16 N16 T17 N17 T18 N18 –2 –2 036912 15 18 MARCH5 BC patients (n=18) BC patients (n=18) β-Actin

E Peritumor Tumor F

1.0

y Low MARCH5 15 0.8 expression (n=27) s * 0.6 10 0.4 High MARCH5 of MARCH5

5 Survival probabilit 0.2 expression (n=38) IHC staining score 0.0 P=0.029 0 Peritumor Tumor 050 100 150 200 Time (months) G

1.00

y Low/medium-expression 0.75 (n=812)

0.50 High expression 0.25 (n=269) Survival probabilit P=0.048 0.00 0 2,000 4,000 6,000 8,000 Time (days)

Figure 1 MARCH5 is upregulated in BC cells and contributes to worse prognosis. Notes: (A and B) Quantitative real-time PCR and Western blot assays for expressions of MARCH5 in five BC cell lines (MDA-MB-231, MDA-MB-468, MDA-MB-435, MCF- 7, and T47D) and two nonmalignant human breast cell lines (MCF 10A and MCF 12A). (C and D) Quantitative real-time PCR and Western blot assays for expressions of MARCH5 in 18 paired BC tumor tissues and peritumor tissues (T, tumor; P, peritumor). Scale bars, 50 µm. The relative MARCH5 expression ratio of tumor to peritumor was log2-transformed. (E) Representative immunohistochemical (IHC) staining images (left panel, scale bar, 50 µm) and staining intensities (right panel) of MARCH5 in tumor and peritumor tissues from 65 BC patients. (F) Kaplan–Meier curve for survival analysis of 65 BC patients according to their MARCH5 expression levels. (G) Prognostic significance analysis for MARCH5 in BC using the online web portal UALCAN http://ualcan.path.uab.edu( ). *P<0.05. Abbreviation: BC, breast cancer.

204 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

BC tumor tissues and peritumor tissues. In concordance with promoted the growth of BC cells mainly through induction the results from BC cell lines, a significant upregulation of of G1–S cell cycle transition. MARCH5 was also observed in tumor tissues compared with peritumor tissues (Figure 1C, D). Knockdown of MARCH5 inhibited the To evaluate the clinical significance of MARCH5 in BC, migration and invasion abilities of BC MARCH5 expression was further evaluated by IHC analysis cells through inhibition of epithelial– in another 65 paired tumor and peritumor tissues. As shown mesenchymal transition (EMT) in Figure 1E, IHC score of MARCH5 was remarkably We also investigated the role of MARCH5 in the migration increased in tumor tissues of BC when compared with paired and invasion of BC cells by scratch wound healing and peritumor tissues. Kaplan–Meier survival curves showed that Matrigel invasion assays. As shown in Figure 3A, B, BC patients with high expression levels of MARCH5 had knockdown of MARCH5 significantly reduced both the significantly poorer survival than those with low MARCH5 wound closure and invasion of BC cells. EMT is a process (Figure 1F), which was further confirmed by prognostic by which epithelial cancer cells acquire more migratory significance analysis using the online web portal UALCAN15 mesenchymal phenotype characterized by reduced cell–cell (http://ualcan.path.uab.edu) (Figure 1G). These findings col- contact and increased cell motility.16,17 We thus explored lectively indicate that MARCH5 is upregulated in BC cells, whether EMT was involved in the suppression of BC which predicts a poor prognosis of patients with BC. migration and invasion when MARCH5 was knocked down. qRT-PCR and Western blot analysis indicated that Knockdown of MARCH5 suppressed BC knockdown of MARCH5 significantly increased the levels cell growth by inducing G1–S cell cycle of epithelial markers of E-cadherin and ZO-1 but reduced the arrest levels of mesenchymal markers of N-cadherin and vimentin To investigate the biological functions of MARCH in BC in BC cells (Figure 3C, D). Except for EMT, increased cells, MARCH5 was knocked down by siRNA in MDA- matrix metalloproteinase (MMP) expression also have MB-231 and MDA-MB-468 cells, which have relative high been implicated in tumor metastasis.18 To further determine MARCH5 expression (shown in Figure 1A, B). As shown in whether MMPs were also involved in MARCH5-mediated Figure S1A, B, a significant decrease in MARCH5 at both BC cell invasion, the expression levels of four most common mRNA and protein levels was observed in MDA-MB-231 MMPs were analyzed by qRT-PCR and Western blot analysis and MDA-MB-468 cells transfected with siMARCH5 when in BC cells with MARCH5 knockdown. As shown in Figure compared with scramble siRNA groups. MARCH5 knock- 3E, F, no significant change of the expressions of MMPs down significantly suppressed BC cell growth, as evidenced was observed in BC cells when MARCH5 was knocked by MTS cell viability and colony formation assays (Figure down, suggesting that MMP production were not involved 2A, B). Considering that growth promotive effect could be in MARCH5-mediated invasion of BC cells. These findings caused by induction of cell cycle progression or inhibition collectively suggest that MARCH5 promotes the migration of apoptosis, or both. To further explore the mechanism by and invasion abilities of BC cells through induction of EMT which MARCH5 promotes the growth of GC cells, we deter- transition. mined whether MARCH5 regulated cell cycle distribution and apoptosis of GC cells by flow cytometry. A significant Knockdown of MARCH5 suppressed BC increase of cells in G1 phase and decrease of cells in S phase growth and metastasis in vivo were found in BC cells when MARCH5 was knocked down To study the role of MARCH5 in BC growth in vivo, sub- (Figure 2C). Consistently, fewer proliferating cells were cutaneous xenograft models were established by injecting detected by EdU (5-ethynyl-2′-deoxyuridine) incorporation MDA-MB-231 cells with stable MARCH5 knockdown in assay in BC cells with MARCH5 knocked down compared 4-week-old male nude mice. As shown in Figure 4A, tumors with their control cells (Figure 2D). However, no significant developed from MDA-MB-231 cells with stable MARCH5 changes of cell apoptosis were observed in BC cells when knockdown (shMARCH5) grew slower than those in con- MARCH5 was knocked down, as evidenced by both the flow trols (shCtrl). Consistently, the volume and net wet weight cytometry (Figure 2E) and Western blot analysis (Figure of tumors from shMARCH5 group were also significantly 2F, G). Collectively, these results suggest that MARCH5 decreased compared with those in control group (Figure 4B).

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 205 Dovepress Tang et al Dovepress

MDA-MB-231 MDA-MB-468 A MDA-MB-231 MDA-MB-468 B

r 300 *

r * 2.0 siCtrl siCtrl 300 * siCtrl 2.0 siCtrl siMARCH5#1 * siMARCH5#1 y 200 y 1.5 1.5 siMARCH5#2 200 siMARCH5#2 ** ** 1.0 1.0 100 siMARCH5#1 100 siMARCH5#1 Cell viabilit 0.5 0.5 Colony numbe Cell viabilit 0 Colony numbe 0.0 0.0 0 01234 56 01234 56 2 siCtrl Time (days) Time (days) siMARCH5#2 siCtrl siMARCH5#2 siMARCH5#1siMARCH5#2 siMARCH5#1siMARCH5# C MDA-MB-231 MDA-MB-468

siCtrl siMARCH5#1 siMARCH5#2 siCtrl siMARCH5#1 siMARCH5#2 90 * siCtrl 90 * siCtrl * siMARCH5#1 75 * siMARCH5#1 G1:71.59% G1:70.82% 75 siMARCH5#2 G1:61.85% G1:70.73% G1:72.61% siMARCH5#2 G1:62.19% 60 60 S:14.64% S:18.67% S:21.15% S:17.18% S:13.76% S:22.67% G2:13.76% 45 45 G2:10.51% G2:17.00% G2:12.09% G2:13.63% G2:15.14% 30 30 Number 15 Number 15 Cell percentage Cell percentage 0 0 G1 S G2/M G1 S G2/M Channels (FL3 Lin -FL3 LIN) Channels (FL3 Lin-FL3 LIN) D MDA-MB-231 MDA-MB-468 siCtrl siMARCH5#1 siMARCH5#2 siCtrl siMARCH5#1 siMARCH5#2 60 * 80 * * 60 * 40 40 20 20 Percentage of Percentage of EdU-positive cells EdU-positive cells 0 0

siCtrl siCtrl

siMARCH5#1siMARCH5#2 siMARCH5#1siMARCH5#2 E MDA-MB-231 MDA-MB-468 ) siCtrl siMARCH5#1 siMARCH5#2 15 siCtrl siMARCH5#1 siMARCH5#2 ) 20 3 ns 3 3 ns 103 103 10 10 10 103 B1 B2 B1 B2 B1 B2 B1 B2 B1 B2 B1 B2 3.1% 4.3% 3.6% 4.2% 4.1% 4.3% ns 3.9% 5.9% 6.0% 5.4% 3.1% 5.5% ns 2 2 2 15 102 102 10 10 10 10 102

1 1 1 1 1 1 PI

10 10 10 PI 10 10 PI 10 10 PI PI PI 5 0 0 100 100 0 0 10 B3 B4 10 B3 B4 B3 B4 B3 B4 10 B3 B4 10 B3 B4 5 88.1% 4.5% 87.2% 5.0% 86.8% 4.9% 83.0% 7.2% 82.5% 6.1% 84.1% 7.3% Apoptotic cells (% Apoptotic cells (% 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 10 10 10 10 100 101 102 103 10 10 10 10 0 10 10 10 10 10 10 10 10 100 101 102 103 0 Annexin V Annexin V Annexin V 2 Annexin V Annexin V Annexin V siCtrl siCtrl

siMARCH5#1siMARCH5# siMARCH5#1siMARCH5#2 F MDA-MB-231 MDA-MB-468

siCtrl siCtrl siMARCH5#1 siMARCH5#1 siCtrl siMARCH5#1siMARCH5#2 siCtrl siMARCH5#1siMARCH5#2 siMARCH5#2 siMARCH5#2 Cyt C 15 kDa 1.5 2.0 ns Cyt C 15 kDa 1.5 ns 1.2 ns

Mito ns Mito ns ns COX IV 15 kDa 1.5 COX IV 0.9 1.0 15 kDa 1.0 1.0 0.6 Cyt C 0.5 Cyt C 0.5 15 kDa 0.5 15 kDa 0.3 Cyt C/COX IV Cyt C/COX IV Cyt C/β-Actin Cyt C/β-Actin Cyto Cyto β-Actin 43 kDa 0.0 0.0 β-Actin 43 kDa 0.0 0.0

G MDA-MB-231 MDA-MB-468 1.2 ns 2.0 ns 0.9 ns 1.5 ns siCtrl siCtrl 0.6 siMARCH5#1 (cleaved) 1.0

(cleaved) siMARCH5#1

siMARCH5#2 / β-Actin / β-Actin siMARCH5#2 RP RP 0.3 0.5 siCtrl siMARCH5#1siMARCH5#2

siCtrl siMARCH5#1siMARCH5#2 PA PA PARP 0.0 PARP 0.0 85 kDa (cleaved) (cleaved) 85 kDa 1.5 ns 1.2ns 1.5 ns 1.0 ns Caspase 9 ns Caspase 9 ns ns 37 kDa ns 0.9 37 kDa 0.8 (cleaved) 1.0 (cleaved) 1.0 0.6 Caspase 3 0.6 17 kDa Caspase 3

/ β-Actin 17 kDa / β-Actin

/ β-Actin 0.4 (cleaved) 0.5 (cleaved) / β-Actin 0.3 0.5 0.2 β-Actin 43 kDa β-Actin 43 kDa Capase 3 (cleaved) Capase 3 (cleaved) Capase 9 (cleaved) 0.0 0.0 Capase 9 (cleaved) 0.0 0.0

Figure 2 Knockdown of MARCH5 suppressed BC cell growth by inducing G1–S cell cycle arrest. Notes: (A) MTS cell viability analysis in MDA-MB-231 and MDA-MB-468 cells with transfection of siMARCH5 or siCtrl as indicated (siMARCH5, siRNA against MARCH5; siCtrl, control siRNA). (B) Colony formation assay in BC cells with treatment as indicated. (C) Flow cytometry-based cell cycle analysis in MDA-MB-231 and MDA-MB-468 cells with treatment as indicated. (D) EdU incorporation assay for determination of cell proliferation in BC cells with treatment as indicated. Scale bars, 50 µm. (E) Annexin V/PI staining for cell apoptosis analysis in BC cells with treatment as indicated. (F) Western blot analysis for cytochrome c (cyt c) levels in cytoplasm and mitochondria of BC cells with treatment as indicated. β-actin and COX IV were loading controls of cytoplasm (Cyto) and mitochondria (Mito). (G) Western blot analysis for levels of cleaved caspase 9 and caspase 3 in BC cells with treatment as indicated. *P<0.05. Abbreviations: BC, breast cancer; EdU, 5-ethynyl-2′-deoxyuridine; PI, propidium iodide; ns, not significant.

206 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

A MDA-MB-231 MDA-MB-468

siCtrl siMARCH5#1 siMARCH5#2 ) siCtrl siMARCH5#1 siMARCH5#2 ) 60 * 60 * * * 40 40

20 20

ound closure (% 0 ound closure (% 0 W W

siCtrl siCtrl

siMARCH5#1siMARCH5#2 siMARCH5#1siMARCH5#2 B MDA-MB-231 MDA-MB-468

50 50 siCtrl siMARCH5#1 siMARCH5#2 * siCtrl siMARCH5#1 siMARCH5#2 * 40 * 40 * 30 30 20 20 10 10 Number of cells Number of cells invaded per field invaded per field 0 0

siCtrl siCtrl

siMARCH5#1siMARCH5#2 siMARCH5#1siMARCH5#2 CD MDA-MB-231 MDA-MB-468 MDA-MB-231 MDA-MB-468 1 2 1 2

siCtrl siCtrl siMARCH5#1 siMARCH5#1 siCtrl siMARCH5#siMARCH5# siCtrl siMARCH5#siMARCH5# siMARCH5#2 siMARCH5#2 * E-cadherin 110 kDa E-cadherin 110 kDa 3 * 3

levels levels * * * * ZO-1 195 kDa ZO-1 195 kDa 2 * 2 * * * * * N-cadherin * * * N-cadherin 100 kDa 100 kDa 1 1 * Vimentin 54 kDa Vimentin 54 kDa 0 0 Relative mRNA Relative mRNA β-Actin 43 kDa β-Actin 43 kDa ZO-1 ZO-1 mentin Vi Vimentin E-cadherin N-cadherin E-cadherin N-cadherin

E F MDA-MB-231 MDA-MB-468 MDA-MB-231 MDA-MB-468

siCtrl siCtrl trl trl ARCH5#1 ARCH5#2 ARCH5#1 ARCH5#2 siMTP18#1 siMTP18#1 C iC i M M s iM iM s i i siMTP18#2 siMTP18#2 s s s s 1.5 1.5 MMP1 53 kDa MMP1 53 kDa levels levels

1.0 1.0 MMP2 72 kDa MMP2 72 kDa

0.5 0.5 MMP7 29 kDa MMP7 29 kDa

0.0 0.0 MMP9 95 kDa MMP9 95 kDa Relative mRNA 1 2 7 9 Relative mRNA 1 2 7 9 β-Actin 43 kDa β-Actin 43 kDa MMP MMP MMP MMP MMP MMP MMP MMP

Figure 3 Knockdown of MARCH5 inhibited the migration and invasion abilities of BC cells through inhibition of EMT. Notes: (A and B) Cell migration and invasion were evaluated by the wound healing migration and Matrigel invasion assays in MDA-MB-231 and MDA-MB-468 cells after transfection with siMARCH5 or siCtrl (siMARCH5, siRNA against MARCH5; siCtrl, control siRNA). Scale bars, 50 µm. (C and D) The expressions of EMT markers were evaluated by qRT-PCR and Western blot analysis in BC cells treated as indicated. (E and F) The mRNA levels of four matrix metalloproteinases (MMPs) were evaluated by qRT-PCR analysis in BC cells treated as indicated. *P<0.05. Abbreviations: BC, breast cancer; EMT, epithelial–mesenchymal transition; qRT-PCR, quantitative real-time PCR.

In addition, IHC staining showed significantly lower levels of of metastatic nodules in the lungs in shMARCH5 group than MARCH5 in tumors obtained from the nude mice injected those in the shCtrl group (Figure 4E). with shMARCH5-transfected cells than in the controls, suggesting that the tumor growth inhibition effect was exerted Forced expression of MARCH5 by MARCH5 knockdown (Figure 4C). As expected, Ki-67 promoted growth and metastasis of BC staining showed significantly fewer proliferating cells in cells shMARCH5 xenografts when compared with control (Figure To further confirm the promotive effects of MARCH5 4D). Moreover, in vivo metastatic assay via tail vein injection on the growth and metastasis of BC cell, MARCH5 was of GC cells further showed significantly decreased formation upregulated by transfecting the MARCH5 expression vector

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 207 Dovepress Tang et al Dovepress

AB )

3 1,500 shCtrl

) 1.6 shMARCH5 * 1.2 1,000 * shCtrl 0.8 500

mor weight (g 0.4 shMARCH5 mor volume (mm Tu

Tu 0.0 0 5 024 6 shCtrl Weeks after injection shMARCH

C D shCtrl shMARCH5 shCtrl shMARCH5 100 * 10 80 8 * 60 6 40 4 2 20 0 Ki67 positive cells 0

IHC scores of MTP1 8 5 5

shCtrl shCtrl

shMARCH shMARCH

E shCtrl shMARCH5

6 *

2 Number of lung

metastasis nodules 0 5

shCtrl

shMARCH

Figure 4 Knockdown of MARCH5 suppressed breast cancer growth and metastasis in vivo. Notes: (A) Growth of subcutaneous xenografts established from MDA-MB-231 cells with stable MARCH5 knockdown or control cells. shMARCH5, shRNA expression vector against MARCH5; shCtrl, control shRNA. (B) The weight of tumors dissected from mice was compared between the shMARCH5 and shCtrl groups. (C and D) The expression levels of MARCH5 and Ki-67 were detected by IHC analysis in the two groups of subcutaneous xenografts tumors. Scale bars, 10 µm. (E) Representative images and numbers of lung metastatic foci in each group are shown. Scale bars, 10 µm. *P<0.05.

or empty vector into MCF-7 and T47D cells, which have that MARCH5 upregulation significantly increased both relative low MARCH5 expression (shown in Figure 1A, B). migration and invasion abilities of MCF-7 and T47D cells Upregulation of MARCH5 was evidenced by qRT-PCR and (Figure 5F, G). Collectively, these results suggest a crucial Western blot analysis (Figure S2A, B). As shown in Figure oncogenic role for MARCH5 in the regulation of BC growth 5A, B, upregulation of MARCH5 significantly enhanced and metastasis. cell viability and clonogenicity of MCF-7 and T47D cells. As expected, MARCH5 upregulation also induced G1–S cell Overexpression of MARCH5 is mainly cycle progression and EdU incorporation (Figure 5C–E), mediated by downregulation of miR-30a which further supported the tumor growth promotive effect MicroRNAs are crucial components in the network of of MARCH5 by causing cell cycle progression in BC. More- gene expression regulation. A previous study has reported over, wound healing and Matrigel invasion assays indicated MARCH5 as a novel target of miR-30a.19 Considering that

208 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

A B MCF-7 T47D MCF-7 T47D 1.5 1.5 EV EV 600 600

y y * 1.2 MARCH5 1.2 MARCH5 * EV EV 0.9 * 0.9 * 400 400 0.6 0.6 5

Cell viabilit Cell viabilit 200 200 0.3 0.3 Colony number Colony number 0.0 0.0 0 0 MARCH 5 01234 56 01234 56 MARCH EV MARCH5 EV MARCH5 Time (days) Time (days) C MCF-7 T47D

EV MARCH5 EV MARCH5 90 * EV 80 * 75 G1:73.47% EV G1:81.26% G1:65.33% MARCH5 G1:63.89% MARCH5 S:9.62% S:21.90% 60 S:15.86% S:17.81% 60 45 G2:9.12% G2:12.77% G2:10.67% G2:18.29% 40

Number 30 15 Number 20 Cell percentage 0 Cell percentage G1 S G2/M 0 G1 S G2/M Channels (FL3 Lin-FL3 LIN) Channels (FL3 Lin-FL3 LIN)

D MCF-7 T47D EV MARCH5 EV MARCH5

60 60 * * 40 40

20 20 Percentage of Percentage of EdU-positive cells EdU-positive cells 0 0 EV MARCH5 EV MARCH5

E MCF-7 T47D

EV MARCH5 ) EV MARCH5

20 ) 103 103 103 103 25 B1 B2 B1 B2 ns B1 B2 B1 B2 ns 8.0% 5.6% 3.0% 4.1% 4.5% 8.7% 5.3% 8.2% 102 102 15 102 102 20

1 1 1 15 10 10 10 101 P I P I 10 P I P I 10 100 100 100 0 B3 B4 B3 B4 5 B3 B4 10 B3 B4 77.6% 8.8% 84.5% 8.4% 77.5% 9.3% 77.4% 9.1% 5 Apoptotic cells (% 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 10 10 10 10 10 10 10 10 0 10 10 10 10 10 10 10 10 Apoptotic cells (% Annexin V Annexin V Annexin V 0 EV MARCH5 Annexin V EV MARCH5

F MCF-7 T47D EV MARCH5 EV MARCH5

) 60 ) 60 * * 40 40

20 20 ound closure (% ound closure (%

W 0 W 0 EV MARCH5 EV MARCH5

MCF-7 T47D G EV MARCH5 EV MARCH5 40 40 * * 30 30

20 20

10 10 Number of cells Number of cells invaded per field invaded per field 0 0 EV MARCH5 EV MARCH5

Figure 5 Forced expression of MARCH5 promoted growth and metastasis of BC cells. Notes: (A and B) MTS cell viability and colony formation were analyzed in MCF-7 and T47D cells after transfection with MARCH5 (MARCH5) expression vector or empty vector (EV). (C) Flow cytometry-based cell cycle analysis in BC cells with treatment as indicated. (D) Cell proliferation was analyzed by the EdU incorporation assay in BC cells with treatment as indicated. Scale bar, 20 µm. (E) Cell apoptosis was analyzed by the flow cytometry in BC cells treated as indicated. F( and G) Wound healing migration and transwell Matrigel invasion assays in BC cells treated as indicated. Scale bars, 50 µm. *P<0.05. Abbreviations: BC, breast cancer; EdU, 5-ethynyl-2′-deoxyuridine; EV, empty vector; ns, not significant.

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 209 Dovepress Tang et al Dovepress miR-30a has been proved as a frequently downregulated and protein levels in BC cells (Figure 6A, B). To provide tumor suppressor in various types of cancer,20 including further supports, we examined the levels miR-30a in 18 BC BC.21 We hypothesized that miR-30a may be involved in the tissue samples. We found a significant negative correlation upregulation of MARCH5 in BC cells. To test this possibility, between the levels of miR-30a and MARCH5 (r=–0.545, synthetic precursor of miR-30a was transfected into MCF-7 P=0.019) (Figure 6C). Moreover, forced expression of and T47D cells. Forced expression of miR-30a significantly miR-30a significantly attenuated both the tumor growth and downregulated the expression of MARCH5 at both mRNA metastasis of BC cells promoted by MARCH5 (Figure 6D–G).

A B MCF-7 T47D C MCF-7 T47D 1.2r= –0.545 P=0.019 1.5 levels 0.9 1.5 a 0a * * 3 l l 0.6 1.0 1.0 tr C miR-30 Ctr miR- 0.3 0.5 0.5 MARCH5 34 kDa MARCH5 34 kDa expression expression miR-30a mRNA 0.0 0.0 0.0 β-Actin 43 kDa β-Actin 43 kDa 0.00.3 0.6 0.91.2

Relative MARCH5 mRNA a Relative MARCH5 mRNA a MARCH5 mRNA levels Ctrl Ctrl

miR-30 miR-30

D E MCF-7 T47D

MCF-7 T47D 300 400 * EV * * EV * EV MARCH5 EV MARCH5 300 1.5 200 1.2 MARCH5+miR-30a MARCH5+miR-30a 200

y 1.2 y * 100 0.8 0.9 * 100

* * Colony number Colony number

0.6 MARCH5 MARCH5 0 0.4 0 Cell viabilit Cell viabilit 0.3 EV EV 0.0 0.0 MARCH5 MARCH5 01234 56 0 1234 56 Time (days) Time (days) MARCH5 MARCH5 +miR-30a MARCH5-miR-30a +miR-30a MARCH5-miR-30a

F MCF-7 T47D

EV MARCH5 MARCH5+miR-30a EV MARCH5 MARCH5+miR-30a

) 80 * ) 80 * * 60 60 *

40 40

und closure (% 20 und closure (% 20 Wo Wo 0 0

EV EV

MARCH5 MARCH5

MARCH5-miR-30a MARCH5-miR-30a

G MCF-7 T47D

EV MARCH5 MARCH5+miR-30a EV MARCH5 MARCH5+miR-30a 50 * 60 * * 40 * 40 30 20 20 Number of cells Number of cells

invaded per field 10 invaded per field 0 0

EV EV

MARCH5 MARCH5

MARCH5-miR-30a MARCH5-miR-30a

Figure 6 Overexpression of MARCH5 is mainly mediated by downregulation of miR-30a. Notes: (A and B) MARCH5 expression was evaluated by quantitative real-time PCR and Western blot in MCF-7 and T47D cells after transfection with the synthetic miR- 30a precursor. (C) The correlation between mRNA levels of MARCH5 and miR-30a was determined in 18 tumor tissues of BC. (D and E) MTS cell viability and colony formation were analyzed in MCF-7 and T47D cells treated as indicated. (F and G) Wound healing migration and transwell Matrigel invasion assays in MCF-7 and T47D cells treated as indicated. Scale bars, 50 µm. *P<0.05. Abbreviations: BC, breast cancer; EV, empty vector.

210 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

MARCH5 promotes tumor growth which contribute to poor survival of patients with HCC.10 In and metastasis through elevation of addition, Rehman et al reported that DRP1 is upregulated 9 mitochondrial fission-mediated ROS and MFN2 is downregulated in lung cancer cells. Moreover, in BC cells, Zhao et al found that mitochondria were more production fragmented in metastatic BC cells with higher levels of DRP1 Previous studies have demonstrated MARCH5 as a critical and lower levels of MFN1 compared with nonmetastatic BC regulator of mitochondrial fission and fusion.5 To explore the cells.12 These observations support the interpretation that molecular basis for the oncogenic properties of MARCH5 dysfunction of mitochondrial dynamics plays a critical role in BC, we first assessed the morphological changes of in the progression of most tumor types. mitochondria in BC cells after MARCH5 was knocked Deregulation of miR-30a has been reported in various down or upregulated. As shown in Figure 7A, MARCH5 malignances, including lung, thyroid, gastric, and colon knockdown promoted extensive elongation of mitochondria cancers.20,24–26 Moreover, a frequently downregulated miR- in MDA-MB-231 cells, while MARCH5 overexpression 30a has also been observed in BC cells.21 This study demon- resulted in a remarkable fragmentation of mitochondria in strated that MARCH5 is a novel target of miR-30a in BC and MCF-7 cells, as evidenced by a MitoTracker Green staining. downregulated miR-30a was involved in the overexpression Mitochondria is a major source of ROS, which play an of MARCH5 in BC, suggesting that miR-30a is a critical important role in the regulation of tumor growth and metas- tumor suppressor microRNA in BC. However, considering tasis by activating several oncogenic signaling pathways.22 that a wide range of mechanisms are used by eukaryotes for Thus, we speculated that mitochondrial fission-regulated the regulation of gene expression, we still cannot rule out the ROS production may be involved in the regulation of tumor possibility that other molecular mechanisms may be involved growth and metastasis by MARCH5 in BC. As shown in in the overexpression of MARCH5 in BC, which still needs Figure 7B, ROS production was significantly reduced when further clarification. MARCH5 was knockdown in MDA-MB-231 cells, while Overexpression of MARCH5 in BC suggested that significantly elevated upon overexpression of MARCH5 in MARCH5 could function as a potential oncogene. We MCF-7 cells. To test whether elevated production of ROS is therefore investigated the biological functions of MARCH5 involved in the oncogenic properties of MARCH5, BC cells by gain- and loss-of-function assays in human BC cell were treated with H O or NAC (a ROS scavenger).23 Our 2 2 lines. Knockdown of MARCH5 significantly inhibited cell results showed that H O treatment significantly increased 2 2 growth and metastasis in BC cell lines MDA-MB-231 and the growth and metastasis of MDA-MB-231 cells suppressed MDA-MB-468. Conversely, forced expression of MARCH5 by MARCH5 knockdown (Figure 7C–F). In contrast, NAC significantly enhanced cell growth and metastasis in BC cell treatment significantly decreased the growth and metastasis lines MCF-7 and T47D. Consistently, a previous study in of MCF-7 cells induced by MARCH5 overexpression. Col- ovarian cancer also demonstrated that MARCH5 promoted lectively, these results suggest that the oncogenic effects of the migration and invasion of SKOV3 cells both in vitro and in MARCH5 in BC cells are mainly mediated by increased vivo.19 MARCH5 has also been shown to act as a prosurvival mitochondrial fission and subsequently ROS production. factor under stress conditions in HeLa cells.27 Moreover, our study further indicated that MARCH5 promoted BC cell Discussion growth and metastasis mainly through induction of G1–S cell In this study, we observed that MARCH5, a critical regula- cycle and EMT, but not through suppression of apoptosis and tor of mitochondria dynamics, is commonly upregulated induction of MMP production. These results together support in BC cell lines and tumor tissues mainly due to the down- a pro-oncogenic role for MARCH5 in tumor progression. regulation of miR-30a. In addition, our results revealed that The close link between mitochondrial fragmentation high expression levels of MARCH5 predict poorer overall and cancer has been well established in many different survival and recurrence-free survival of patients with BC. types of cancer, including BC. Zhao et al showed that MFN Consistently, abnormal expressions of other mitochondrial silencing-mediated mitochondrial fragmentation significantly dynamic regulators have also been demonstrated in differ- enhanced metastatic abilities of BC cells.12 Our results ent types of cancer. For example, Huang et al have shown showed that MARCH5 overexpression induced a remarkable that mitochondrial DRP1 is upregulated and MFN1 is fragmentation of mitochondria, indicating the involvement of downregulated in hepatocellular carcinoma (HCC), both of mitochondrial fragmentation in BC progression. Mitochondria

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 211 Dovepress Tang et al Dovepress

A MDA-MB-231 MCF-7

siCtrl si-MARCH5#1 si-MARCH5#2 Fragmented EV MARCH5 Fragmented Intermediate Intermediate Tubulated Tubulated ) 80 * ) 80 * * 60 60 40 40 20 20 Cell population (% 0 Cell population (% 0

siCtrl EV

MARCH5 si-MARCH5#1si-MARCH5#2 MDA-MB-231 B CDMDA-MB-231

y MDA-MB-231 siMARCH5 300 siCtrl si-MARCH5 * siCtrl 30 * EV +H O * * 1.8 2 2 siMARCH5 200 20 y siMARCH5+H2O2 1.2 * * 100

#1 10 Colony number 0.6 Cell viabilit

si-MARCH5 0 0 5 O 2 siCtrl H 2 Mean fluorescence intensit siCtrl 0.0 0123456 siMARCH

#2 Time (days) si-MARCH5#1si-MARCH5#2 siMARCH5+ si-MARCH5 MCF-7 MCF-7 MCF-7 EV 300 * 1.5 MARCH5 * MARCH5 MARCH5 EV +NAC 30 y MARCH5+NAC 200 EV 1.0 * 20 * * 100

0.5 Colony number 10 Cell viabilit 0 C 0.0 EV 0 0123456 MARCH5

MARCH5 EV

Mean fluorescence intensity Time (days) MARCH5+NA MARCH5

E MDA-MB-231 MCF-7 ) siCtrl siMARCH5 siMARCH5+H O ) 2 2 80 * EV MARCH5 MARCH5+NAC 100 * * 60 80 * 60 40 40 20 20 ound closure (% ound closure (% W 0 W 0 5 C O 2 EV siCtrl H 2 MARCH5 siMARCH MARCH5+NA siMARCH5+ FGMDA-MB-231 miR-30a MARCH5 60 siMARCH5 * siCtrl siMARCH5 * +H2O2 40 Fission 20 Number of cells invaded per field 0 5 O 2 siCtrl H 2 siMARCH ROS

MCF-7 siMARCH5+ 80 MARCH5 * EV MARCH5 * Cell cycle Migration and invasion +NAC 60 40

20 Number of cells invaded per field 0 BC growth and metastasis C EV

MARCH5

MARCH5+NA

Figure 7 MARCH5 promotes tumor growth and metastasis through elevation of mitochondrial fission-mediated ROS production. Notes: (A) Left panel: Representative mitochondrial morphology of MDA-MB-231 cells with MARCH5 knockdown and MCF-7 cells with MARCH5 overexpression. Scale bars, 1 µm. Right panel: The proportions of BC cells (30 cells for each sample) with fragmented, intermediate, and tubulated were quantified. B( ) The intracellular ROS levels were determined by flow cytometry in BC cells treated as indicated. C( and D) MTS cell viability and colony formation abilities were analyzed in BC cells treated with H2O2

(90 µM, 12 hours) or NAC (30 µM, 12 hours) as indicated. (E and F) Wound healing and transwell Matrigel invasion assays in BC cells treated with H2O2 (90 µM, 12 hours) or NAC (30 µM, 12 hours) as indicated. Scale bars, 50 µm. (G) Schematic depicting the oncogenic role of MARCH5 upregulation in tumor growth and metastasis of BC and their underlying mechanism. *P<0.05. Abbreviations: BC, breast cancer; EV, empty vector.

212 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis is the main source of ROS, which is critical for tumor 9. Rehman J, Zhang HJ, Toth PT, et al. Inhibition of mitochondrial progression.22 A previous study in HCC has demonstrated fission prevents cell cycle progression in lung cancer. FASEB J. 2012;26(5):2175–2186. a promoting effect of mitochondrial fragmentation on ROS 10. Huang Q, Zhan L, Cao H, et al. Increased mitochondrial fission promotes production.10 Our study showed a similar effect of MARCH5- autophagy and hepatocellular carcinoma cell survival through the ROS- modulated coordinated regulation of the NFKB and TP53 pathways. mediated mitochondrial fragmentation on the production of Autophagy. 2016;12(6):999–1014. ROS in BC cells. Moreover, we demonstrated that elevated 11. Jin B, Fu G, Pan H, et al. Anti-tumour efficacy of mitofusin-2 in urinary ROS production was involved in the promotion of BC growth bladder carcinoma. Med Oncol. 2011;28(Suppl 1):S373–S380. 12. Zhao J, Zhang J, Yu M, et al. Mitochondrial dynamics regulates migration and metastasis by MARCH5. and invasion of breast cancer cells. Oncogene. 2013;32(40):4814–4824. In summary, our results show that MARCH5 is frequently 13. von Eyss B, Jaenicke LA, Kortlever RM, et al. A MYC-driven change in mitochondrial dynamics limits YAP/TAZ function in mammary upregulated in BC, and upregulation of MARCH5 plays a epithelial cells and breast cancer. Cancer Cell. 2015;28(6):743–757. critical oncogenic role in breast carcinogenesis by promotion 14. Han XJ, Yang ZJ, Jiang LP, et al. Mitochondrial dynamics regulates of both BC growth and metastasis (Figure 7G), which hypoxia-induced migration and antineoplastic activity of cisplatin in breast cancer cells. Int J Oncol. 2015;46(2):691–700. provides experimental evidence supporting MARCH5 as a 15. Chandrashekar DS, Bashel B, Balasubramanya SAH, et al. UALCAN: potential therapeutic target in BC therapy. a portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia. 2017;19(8):649–658. 16. Przybyla L, Muncie JM, Weaver VM. Mechanical control of epithelial- Acknowledgment to-mesenchymal transitions in development and cancer. Annu Rev Cell This work was supported by the National Natural Science Dev Biol. 2016;32:527–554. 17. Ye X, Weinberg RA. Epithelial-mesenchymal plasticity: a central regula- Foundation of China (grant no. 81272201). tor of cancer progression. Trends Cell Biol. 2015;25(11):675–686. 18. Kessenbrock K, Wang CY, Werb Z. Matrix metalloproteinases in stem cell regulation and cancer. Matrix Biol. 2015;44–46:184–190. Disclosure 19. Hu J, Meng Y, Zhang Z, et al. MARCH5 RNA promotes autoph- The authors report no conflicts of interest in this work. agy, migration, and invasion of ovarian cancer cells. Autophagy. 2017;13(2):333–344. 20. Yang X, Chen Y, Chen L. The versatile role of microRNA-30a in human References cancer. Cell Physiol Biochem. 2017;41(4):1616–1632. 1. Wallace MC, Preen D, Jeffrey GP, Adams LA. The evolving epidemi- 21. Fu J, Xu X, Kang L, et al. miR-30a suppresses breast cancer cell ology of hepatocellular carcinoma: a global perspective. Expert Rev proliferation and migration by targeting Eya2. Biochem Biophys Res Gastroenterol Hepatol. 2015;9(6):765–779. Commun. 2014;445(2):314–319. 2. Wallace DC. Mitochondria and cancer. Nat Rev Cancer. 2012;12(10): 22. Sabharwal SS, Schumacker PT. Mitochondrial ROS in cancer: initiators, 685–698. amplifiers or an Achilles’ heel? Nat Rev Cancer. 2014;14(11):709–721. 3. Boland ML, Chourasia AH, Macleod KF. Mitochondrial dysfunction 23. Downs I, Liu J, Aw TY, Adegboyega PA, Ajuebor MN. The ROS in cancer. Front Oncol. 2013;3:292. scavenger, NAC, regulates hepatic Vα14iNKT cells signaling during 4. Guerra F, Guaragnella N, Arbini AA, Bucci C, Giannattasio S, Moro Fas mAb-dependent fulminant liver failure. PLoS One. 2012;7(6): L. Mitochondrial dysfunction: a novel potential driver of epithelial-to- e38051. mesenchymal transition in cancer. Front Oncol. 2017;7:295. 24. Kumarswamy R, Mudduluru G, Ceppi P, et al. MicroRNA-30a 5. Nagashima S, Tokuyama T, Yonashiro R, Inatome R, Yanagi S. Roles inhibits epithelial-to-mesenchymal transition by targeting Snai1 of mitochondrial ubiquitin ligase MITOL/MARCH5 in mitochondrial and is downregulated in non-small cell lung cancer. Int J Cancer. dynamics and diseases. J Biochem. 2014;155(5):273–279. 2012;130(9):2044–2053. 6. Xu S, Cherok E, Das S, et al. Mitochondrial E3 ubiquitin ligase 25. Li X, Zhang Y, Zhang Y, Ding J, Wu K, Fan D. Survival prediction of gas- MARCH5 controls mitochondrial fission and cell sensitivity to stress- tric cancer by a seven-microRNA signature. Gut. 2010;59(5):579–585. induced apoptosis through regulation of MiD49 protein. Mol Biol Cell. 26. Zhang Q, Tang Q, Qin D, et al. Role of microRNA 30a targeting 2016;27(2):349–359. insulin receptor substrate 2 in colorectal tumorigenesis. Mol Cell Biol. 7. Srinivasan S, Guha M, Kashina A, Avadhani NG. Mitochondrial dys- 2015;35(6):988–1000. function and mitochondrial dynamics – the cancer connection. Biochim 27. Park YY, Nguyen OT, Kang H, Cho H. MARCH5-mediated quality Biophys Acta Bioenerg. 2017;1858(8):602–614. control on acetylated Mfn1 facilitates mitochondrial homeostasis and 8. Simula L, Nazio F, Campello S. The mitochondrial dynamics in cancer cell survival. Cell Death Dis. 2014;5:e1172. and immune-surveillance. Semin Cancer Biol. 2017;47:29–42.

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 213 Dovepress Tang et al Dovepress

Supplementary materials

Table S1 Sequence of primers used for quantitative real-time PCR (qRT-PCR) analysis 1. Primers used in qRT-PCR analysis MARCH5 Forward primer GTCCAGTGGTTTACGTCTTGG Reverse primer CCGACCATTATTCCTGCTGC E-cadherin Forward primer ATTTTTCCCTCGACACCCGAT Reverse primer TCCCAGGCGTAGACCAAGA Z0-1 Forward primer CGACCAGATCCTCAGGGTAA Reverse primer TCCATAGGGAGATTCCTTCTCA N-cadherin Forward primer AGCTCCATTCCGACTTAGACA Reverse primer CAGCCTGAGCACGAAGAGTG Vimentin Forward primer GACGCCATCAACACCGAGTT Reverse primer CTTTGTCGTTGGTTAGCTGGT MMP1 Forward primer CACAGCTTTCCTCCACTGCTGCT Reverse primer GGCATGGTCCACATCTGCTCTTG MMP2 Forward primer ACCTGGATGCCGTCGTGGAC Reverse primer TGTGGCAGCACCAGGGCA MMP7 Forward primer TGAATTTGGCCACTCTCTGGGTCT Reverse primer TCTGAATGCCTGCAATGTCGTCCT MMP9 Forward primer TGTACCGCTATGGTTACACTCG Reverse primer GGCAGGGACAGTTGCTTCT GAPDH Forward primer ACAACTTTGGTATCGTGGAAGG Reverse primer GCCATCACGCCACAGTTTC miR-30a Forward primer GCGTGTAAACATCCTCGAC Reverse primer GTGCAGGGTCCGAGGT U6 Forward primer CTCGCTTCGGCAGCACA Reverse primer AACGCTTCACGAATTTGCGT

Table S2 Primary antibodies used for Western blot and immunohistochemistry analysis Antibody Company (Cat. No.) Working dilutions MARCH5 Novus Biologicals (NBP2-21583) WB: 1/1000 IHC: 1/300 Cytochrome c Abcam (ab90529) WB: 1/1,000 COX IV Abcam (ab14744) WB: 1/800 Caspase 3 Abcam (ab13847) WB: 1/1,000 Caspase 9 Abcam (ab32539) WB: 1/1,000 E-cadherin Abcam (ab1416) WB: 1/1,000 Z0-1 Abcam (ab190085) WB: 1/800 N-cadherin Abcam (ab98952) WB: 1/1,000 Vimentin Abcam (ab8978) WB: 1/1,000 MMP1 Abcam (ab137332) IHC:1/1,000 MMP2 Abcam (ab37150) IHC:1/800 MMP7 Abcam (ab5706) IHC:1/800 MMP9 Abcam (ab38898) WB: 1/1,000 Ki-67 Abcam (ab15580) WB: 1/1,000 IHC: 1/600 β-actin Proteintech (20536-1-AP) WB: 1/1,000

214 submit your manuscript | www.dovepress.com Cancer Management and Research 2019:11 Dovepress Dovepress MARCH5 in tumor growth and metastasis

AB MDA-MB-231 MDA-MB-468 MDA-MB-231 MDA-MB-468

1.5 1.5 * * * * 1.0 1.0

0.5 siCtrl 0.5 siCtrl siMARCH5#1siMARCH5#2 siMARCH5#1siMARCH5#2 expression expression

0.0 0.0 MARCH5 MARCH5 1 2

1 2 Relative MARCH5 mRNA Relative MARCH5 mRNA siCtrl siCtrl β-Actin β-Actin

siMARCH5#siMARCH5# siMARCH5#siMARCH5#

Figure S1 Knockdown of MARCH5 in MDA-MB-231 and MDA-MB-468 cells was confirmed by quantitative real-time PCR (qRT-PCR) A( ) and Western blot (B) analysis at mRNA and protein levels after transfection with siRNAs against MARCH5 (siMARCH5) or control siRNA (siCtrl). Note: Data were presented as the mean ± SEM from three independent repeats, *P<0.05.

AB MCF-7 T47D MCF-7T47D

6 6 * * CH5 CH5 4 4 EV EV MAR MAR

2 2 MARCH5 expression MARCH5 expression

0 0 β-Actin β-Actin Relative MARCH5 mRNA EV MARCH5 Relative MARCH5 mRNA EV MARCH5

Figure S2 Forced expression of MARCH5 in MCF-7 and T47D cells was confirmed by quantitative real-time PCR (qRT-PCR) A( ) and Western blot (B) analysis at mRNA and protein levels after transfection with expression vector encoding MARCH5 (MARCH5) or empty vector (EV). Note: Data were presented as the mean ± SEM from three independent repeats, *P<0.05.

Cancer Management and Research Dovepress Publish your work in this journal Cancer Management and Research is an international, peer-reviewed a very quick and fair peer-review system, which is all easy to use. Visit open access journal focusing on cancer research and the optimal use of http://www.dovepress.com/testimonials.php to read real quotes from preventative and integrated treatment interventions to achieve improved published authors. outcomes, enhanced survival and quality of life for the cancer patient. The manuscript management system is completely online and includes

Submit your manuscript here: https://www.dovepress.com/cancer-management-and-research-journal

Cancer Management and Research 2019:11 submit your manuscript | www.dovepress.com 215 Dovepress