Malic enzyme 1 may be a novel target of breast cancer metastasis

Background Malic enzyme 1 (ME1) catalyzes malate to pyruvate and thus promotes glycolysis, playing a part in the Warburg effect. It also has a potential role in tumor progression, but its function in breast cancer remains to be fully clarified. This work aimed to investigate the prognostic value of ME1 and its possible mechanism in breast cancer. Methods We evaluated ME1 expression in 220 early breast cancer patients with tissue microarray-based immunohistochemistry and explored the relationships between ME1 expression and clinicopathological features. Survival analyses were further performed to determine its prognostic value. The public database was used to confirm tissue microarray results. Moreover, we profiled ME1 expression in breast cancer cell lines via western blotting, and then assessed it in cell viability and motility via Cell counting kit-8 (CCK-8), colony formation, transwell migration and invasion assays. Reactive oxygen species (ROS) was detected by dihydroethidium (DHE) and 2’,7’-Dichlorodihydrofluorescein diacetate (DCFH-DA). Results high ME1 expression was significantly associated with larger tumor size, more lymph node metastasis and more extensive lymph-vascular Survival analysis showed high ME1 was significantly correlated with recurrence free survival (RFS). Multivariate analysis further identified high ME1 expression as an independent prognostic factor for RFS, which was confirmed by mRNA results factor receptor-2 and triple cancer cell higher of ME1, lower ME1. MCF-7 P 0.05 is considered statistically significant. Significant P -values are in bold. HR status was evaluated using immunohistochemistry. HER2 status was evaluated using both immunohistochemistry and fluorescence in situ hybridization. P 0.05 is considered statistically significant. Significant P -values are in bold. HR status was evaluated using immunohistochemistry. HER2 status was evaluated using both immunohistochemistry and fluorescence in situ hybridization.

important, when pretreated with hydrogen peroxide, an oxidizing agent, MCF-7 cells overexpressing ME1 lost its motility, whereas MDA-MB-468 cells with knock-down of ME1 restored its motility when pretreated with N-acetyl cysteine, an antioxidant.
Conclusions To our knowledge, these clinical and experiment work first suggested that ME1 may be a potential therapeutic target for breast cancer metastasis, and its biological effect is mainly controlled by manipulating ROS.

Background
Breast cancer is the most common malignancy in women and one of the three most common malignant tumors worldwide (1). Surgery, radiation, chemotherapy, endocrine therapy and anti-HER2 targeting therapy are applied for early stage patients depending on clinical features and specific molecular subtype (2). Nonetheless, even in patients with no lymph node metastasis, nearly 20% of them suffer from breast cancer recurrence (3). In addition, breast cancer is estimated to rank second only to lung cancer as the leading cause of cancer death in women (15% of all cancer deaths) in America (4). Therefore, it is necessary to further characterize patients with high recurrence risk and explore the molecular mechanism of breast cancer progression to help develop personalized treatment.
Reprogramming energy metabolism is one of the hallmarks of cancer, which is known as Warburg effect: cancer cells prefer glycolysis even in the presence of oxygen (5). It has been well accepted that cancer cells reprogram the metabolic patterns to satisfy their rapid proliferation and metastasis (6). Malic enzyme 1 (ME1) is a cytosolic metabolic enzyme that catalyzes the citric acid cycle intermediate malate to pyruvate, the substrate of glycolysis, and converts Nicotinamide adenine dinucleotide phosphate (NADP) to Nicotinamide adenine dinucleotide phosphate hydrate (NADPH), the necessities of various biosynthetic reactions. Besides, ME1 can form physiological complexes with 6-Phosphogluconate dehydrogenase and further increase the pentose phosphate pathway flux and NADPH generation (7). Hence, ME1 promotes glycolysis and is involved in reactive oxygen species (ROS) homeostasis, glutamine metabolism and lipogenesis (8), playing an important role in Warburg effect.
Recently, the potential function of ME1 has been revealed in tumor progression. It has been reported that knockdown of ME1 leads to increased ROS level, upregulated cell apoptosis, slower cell growth and reductive metastasis in gastric cancer (9). ME1 induces epithelial-mesenchymal transition and is associated with a worse prognosis in hepatocellular carcinoma (10) and oral squamous cell carcinoma (11). In addition, downregulation of ME1 activates the tumor suppressor gene p53 which further inhibits ME1 in a positive-feedback, modulating metabolism and senescence (8). Since there is a lacking of research on the role of ME1 in breast cancer, we performed data mining from the public Oncomine database and found that ME1 was upregulated in various cancer types including breast cancer. These findings prompted us to assess the value of ME1 in breast cancer.
In the present study, we used tissue microarray to determine ME1 expression and then analyzed the relationships between ME1 expression and clinicopathological parameters as well as its prognostic value in breast cancer patients. Next, we profiled the expression of ME1 in diverse breast cancer cells and investigated its role in viability and motility besides its impact on ROS level. Our clinical and experimental work identified ME1 as a novel prognostic indicator, highlighted its potential role in the development of breast cancer and found its effect could be mainly blocked by manipulating ROS.

Methods
Patients, tissue specimens and clinical data. The study cohort consisted of 220 female breast cancer patients who underwent radical surgery between 11 August 2015 and 17 May 2016 in Fudan University Shanghai Cancer Center (FUSCC, Shanghai, China). Eligible patients were women who had histologically confirmed invasive breast cancer; had no evidence of distant metastasis; and provided sufficient tissues for further research.
Patients were ineligible if they had received neoadjuvant chemotherapy or radiation therapy. Clinical data of the patients was retrieved from the Outcome unit. We completed the follow-up on 26 June 2018 and the medium follow-up period was 29.2 months (range 0.50-34.25). Recurrence free survival (RFS) time was calculated from the date of radical surgery to the date of breast cancer recurrence (ipsilateral breast, local-regional, or distant), death or the last time of follow-up.
After deparaffinized and rehydrated, the sections were heated in an autoclave at 120℃ in sodium citrate buffer (pH 6.0) for 10 min for antigen retrieval. The sections were then Immunohistochemical staining score. The standards for IHC staining scoring was previously described (12). Herein the intensity range was 0 = negative; 1 = low; 2 = medium and 3 = high. The quantity 0 = no positivity; 1 = positivity in 0-10%; 2 = positivity in 11-50%; 3 = positivity in 51-80%; 4 = positivity in > 80%. The final immunoreactive score (IRS, ranging from 0-12) was obtained by multiplying the intensity score and the quantity score. Two pathologists blinded to the patients' information scored the immunohistochemical staining. In discrepant cases, they further reviewed the cases and reached a consensus. For ME1 low and high expression was defined as IRS ≤ 6 and IRS > 6, respectively.

ME1 expression in breast cancer tissues
To confirm the role of ME1 in breast cancer, we first performed IHC staining of ME1 in a tissue microarray containing 220 early stage breast cancer patients. We confirmed the specificity of the antibody via ME1 knockdown (see Figure 4F). Interstitial tissue was taken as negative control. Several representative cases are shown in Figure 1A-1D. We evaluated ME1 protein expression semi-quantitatively. As defined above, 51.8% (114/220) and 48.2% (106/220) of the patients were categorized as ME1-high and ME1-low cases, respectively.

Associations between ME1 expression and clinicopathological parameters
Next, we explored the correlations between ME1 expression by IHC and clinicopathological parameters in our cohort (Table 1). Comparing to ME1-low cases, ME1-high cases were significantly associated with larger tumor size (P = 0.036), more lymph node metastasis (P < 0.001) and more extensive lymph-vascular invasion (P = 0.001). However, we did not find any specific relations of ME1 expression with age, histopathologic type, histologic grade, ki67 index or molecular subtype, partially due to the limited sample size of our cohort. Moreover, we investigated the correlation between ME1 mRNA level and clinicopathological parameters in Curtis Breast dataset via Oncomine database (13). High ME1 mRNA level correlated to higher histologic grade, later TNM stage, TNBC and HER2 positive breast cancer (see Figure 2A). In short, high expression of ME1 in both the mRNA and protein levels was associated with risk factors of breast cancer recurrence.

Associations between ME1 expression and clinical outcomes
Since ME1 expression correlated with risk factors of breast cancer recurrence, we further explored its relationship with clinical outcomes. Herein, we defined RFS as date from radical surgery to first recurrence of breast cancer (ipsilateral breast, local-regional, or distant), death or the last time of follow-up and evaluated the association between ME1 expression and RFS in our cohort. Survival analysis by the log-rank test showed high ME1 expression was significantly correlated with worse RFS (P < 0.01) (see Figure 2B). In addition, univariable cox regression showed pathologic N stage ≥ 2, lymph-vascular invasion and ME1 high expression were associated with worse RFS (P < 0.1). Further multivariable cox regression identified ME1 high expression as an independent negative prognostic factor for RFS (HR = 5.343, 95% CI = 1.191-23.971, P = 0.029) ( Table 2).
Collectively, high expression of ME1 indicated worse prognosis of breast cancer patients. Overexpression of ME1 in MCF-7 cells enhanced cell motility in both transwell migration and invasion assays (see Figure 4B and 4C), accompanied by the development of epithelial-mesenchymal transition (EMT) (see Figure 4F). In contrast, knockdown of ME1 in

MDA-MB-468 cells had a profound inhibitory effect on cell migration, invasion and EMT
(see Figure 4D, 4E and 4F).

ME1 promotes breast cancer motility via ROS regulation
It is acknowledged that ME1 catalyzes malate to pyruvate, accompanied by NADP+ converting to NADPH, which plays crucial role on ROS homeostasis. Thus, we assessed ROS level to explore the possible mechanism by which ME1 could promote breast cancer progression. In accordance with our speculation, DHE fluorescence revealed that ME1

Discussion
In the present study, we demonstrated that ME1 high expression was associated with risk factors of breast cancer recurrence and indicated worse prognosis. Thus, we found out the important role of ME1 in breast cancer. In addition to regular risk factors, clinicians can further distinguish high recurrence-risk patients after radical surgery according to ME1 expression and possibly develop suitable treatment for these patients. Further work showed ME1 promoted cell viability, motility and EMT in breast cancer cells, probably by decreasing intracellular ROS level, which probably reflects ROS level in the tumor microenvironment.
Previous study of ME1 in breast cancer only focusing on TNBC subtype drew the same conclusion as ours that ME1 was high expressed in TNBC cells (15), but its upstream upregulators have not been reported yet. We also found the relationship between ME1 expression and HER2 positive in Curtis Breast Dataset. This phenomenon was supported by the evidence that NR1D1, the positive transcription factor of ME1, resides on ERBB2containing 17q12-21 amplicons and is part of the ERBB2 expression signature (16).
The role of ME1 in breast cancer has not been fully elucidated. Although ME1 is reported to enhance cell proliferation and metastasis capacity in multiple cancer types (10,11,15,17), the underlying mechanism of which still remains inexplicit. On one hand, ME1 catalyzes malate to pyruvate, inducing cell glucose uptake and lactate production and thus promote Warburg effect, which is favorable for tumor (15). On the other hand, On the other hand, NADPH generated from ME1-catalyzed reaction reduces ROS accumulation, which is the common byproduct during tumor progression. Otherwise, excessive ROS can result in macromolecules damage including lipids, proteins, and nucleic acids, which is unfavorable for tumor (18). Moreover, ROS released to micro-environment may lead to ROS activation in other cell types including fibroblasts, vascular endothelial cells, immune cells and etc., which can in turn have a crosstalk with tumor cells. Our study implied that ME1 could help maintain the homeostasis of ROS and partially explained the function of ME1, in accordance with the results in gastric cancer cells (9).
As far as we know, there is still no anti-tumor drugs targeting ME1. Considering the important role of ME1 in tumor progression, we also reflect on whether ME1 is a potential therapeutic target. Since ME1 is widely expressed in human tissues and it proves essential for the survival of normal fibroblasts (19), ME1 inhibition might do harm to normal tissues severely and may not be a suitable therapeutic strategy. As demonstrated above, ME1 is highly expressed in HER2 positive cancer, hence ME1 inhibition conjugated to trastuzumab could be a possible treatment for HER2 positive patients. Alternatively, targeting glycolysis, NADPH-producing enzymes, redox-regulating enzymes (20) or potential downstream molecules of ME1 may be a better approach.
There are some limitations in our study. The follow-up period of our cohort was still short and the recurrence rate was only 7.7% (17/220) by the end-up point of our study, but our results were supported by data from KM plotter with a longer follow-up period. In order to identify whether ME1 influence the efficacy of treatment, information of adjuvant therapy including chemotherapy, anti-HER2 therapy and radiotherapy requires to be updated, which is ongoing in our center. Besides, we did not find the exact molecular mechanisms by which ME1 promoted tumor progression, which is now in active research in our laboratory. Despite these limitations, this study can clearly indicate the important role of ME1 in breast cancer and give emphasis on the relationship between Warburg effect and breast cancer.

Conclusion
This study suggested that ME1 correlated with worse prognosis and enhanced cell progression via manipulating micro-environment ROS in breast cancer. More research is guaranteed to prove that ME1 is a potential therapeutic target for breast cancer.

Consent for publication Not applicable.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests The authors declare that they have no competing interests
Funding Not applicable.
Authors' contributions Hu XC conceived, designed the study and revised the manuscript.
Liu C performed the experiments in laboratory and wrote the manuscript. Lin SC and Zhao YN did data analyses. Cao J and Tao ZH managed the clinical data and follow-ups of the patients. All authors read and approved the final manuscript for publication.