Background
One of the most common cancers worldwide is breast cancer, and it is the second leading cause of mortality in women from the United States [
1,
2]. Although conventional therapies and surgical approaches have been developed, none are completely effective in removing and annihilating the cancer.
Cancer cells alter their metabolic machinery to maintain a high level of metabolism and prevent the depletion of a host of substrates, such as glucose and amino acids that are used for energy. A significant aspect of this reprogramming involves changes in the glycolytic pathway, referred to as the Warburg effect [
3,
4]. These changes include an increase in pyruvate that is generated via the glycolytic pathway. Pyruvate is converted to lactic acid instead of acetyl-CoA which enters the TCA cycle and ultimately forms citrate. In addition to metabolic changes from Warburg effect, some cancers depend on glutamate metabolism for fixing ammonia to acquire the nitrogen required for cellular growth [
5]. Further, glutamine synthetase is a target for activated β-catenin and is regulated by the oncogene Myc [
5], which connects the metabolic regulation of cancer cells to several important growth and developmental signaling pathways.
Dietary nutritional control may be a feasible option for supplementing cancer treatment. The use of caloric restriction (CR) is effective in inhibiting cancer development in non-human primates and rodents [
6,
7] and the onset of age-related diseases [
6]. However, recent evidence suggests that CR can reduce function of the immune system [
8,
9], which may not be ideal for people already fighting cancer. Further, the general population is not in favor of reducing food consumption as required by CR. A viable alternative to CR may be to restrict the intake of dietary methionine.
Methionine is an essential amino acid with a multitude of functions. It is prominent in protein translation, since it is the N-terminal amino acid of most mammalian proteins. Methionine is a precursor of glutathione, a tripeptide that reduces reactive oxygen species (ROS) and protects cells from oxidative stress. Methionine is needed for polyamine synthesis, in which polyamines function during nuclear and cell division. Moreover, methionine is a precursor of
S-adenosyl methionine the major source of methyl groups needed for methylation of DNA, proteins and low
M
r biomolecules/metabolites. In rodents, a diet low in methionine (20-35 % of regular chow) reduced adiposity in the fat depots and reduced blood levels of lipids, glucose, IGF-1, and leptin, while elevating levels of FGF21 and adiponectin [
10‐
15]. Reduction in mitochondrial free radical production and oxidative stress also occurs during MR in organs, such as, liver, heart, and brain [
16‐
18]. Finally, MR in rodents promotes longevity and delays onset of age-related impairments and chronic diseases [
10,
19‐
21].
In intestinal, colon, and prostate cancer models, methionine levels were restricted or excluded from diets to examine the effects on tumor progression. In an intestinal tumorigenesis model with Apc(Min/+) mice, folate deficiency in combination with depletion of choline, methionine, and vitamin B12 resulted in reduced tumor size in mice treated by 5 weeks of age, but produced no difference when started at 10 weeks of age [
22]. Azoxymethane-treated rats when fed MR diet had a significant reduction in formation of aberrant crypt foci in colon, suggesting an inhibition of cell proliferation [
21]. More recently, dietary MR in TRAMP mouse model demonstrated a decrease in prostatic intraepithelial neoplasia with a concomitant drop in plasma IGF-1 levels and reduced proliferation in prostate lobe-specific manner [
23]. In human prostate cancer cell lines grown in the absence of methionine, the expression of cell cycle inhibitors P21 and P27 increased and were identified as a possible mechanism for halting the cell cycle and increasing apoptosis [
24]. These studies suggest that methionine depletion in human prostate cells can inhibit proliferation either by halting the cell cycle at the G1/S checkpoint or by directing cells to go through apoptosis. However, little is known about how MR alters the cell cycle in other cancers, such as breast cancer. The present study uses a xenograft model for breast cancer, an immortalized human breast cell line, and an invasive breast cancer cell line to examine whether MR alters cell cycle inhibitors that could inhibit tumor progression.
Methods
Xenograft model for testing efficacy of methionine restriction
MCF10AT1, human transformed breast cells, were a gift obtained from Steven Santner at Barbara Ann Karmanos Cancer Institute, Detroit, MI. The cells were maintained in DMEM/F12 (Invitrogen, Carlsbad, CA) with 5 % horse serum, 0.029 M sodium bicarbonate, and 10 mM HEPES, and supplemented with insulin (10 μg/ml), EGF (20 ng/ml), hydrocortisone (0.5 μg/ml), cholera toxin (100 ng/ml), and 1 % penicillin-streptomycin solution. Cells were routinely passaged weekly and maintained in 5 % CO2 at 37 °C.
The procedures and treatments used on the athymic nude mice were reviewed and approved by Penn State College of Medicine Institutional Animal Care and Use Committee (IACUC#2012-082), and followed standard procedures described in
Guide for the Care and Use of Laboratory Animals, 8
th edition from the National Research Council. A completed ARRIVE guidelines checklist is included in Additional file
1. A total of ten million MCF10AT1 cells with matrigel (1:1 (v/v), BD Biosciences, San Jose, CA) were injected subcutaneously into the left flank region near the mammary fat pad of each of 40 female athymic nude mice (Strain-088 homozygous; 8 weeks of age; Charles River, Wilmington, MA). The mice were fed sterile standard rodent diet for 1 week after cell implantation. These mice were divided into 2 feeding groups: (a) 20 mice on a sterile, a chemically-defined AIN-76-based diet containing 0.86 % methionine (control-fed diet; CF); and (b) 20 mice on a sterile chemically-defined AIN-76-based diet containing 0.12 % methionine (methionine-restricted diet; MR). The mice were housed in groups of five and maintained on a 12 h light-dark cycle and fed these diets ad libitum for 12 weeks. Body weights and average diet intake for each group were measured weekly.
At the end of the study, unfasted mice were euthanized using CO2; whole blood was collected by cardiac puncture, centrifuged, and plasma was collected and stored at −80 °C for analysis. Plasma samples were analyzed for amino acid content, insulin, IGF-1, FGF21, leptin, adiponectin, cholesterol, triglycerides, and glucose.
Tumors were excised, weighed, and measured prior to fixing in 10 % neutral buffered formalin (NBF). Tumor volumes were determined using the formula: 0.523 x a
2 x b, where a is the smallest diameter and b is the largest. The fixed tumors were paraffin-embedded, sectioned and stained for hematoxylin & eosin. Histopathology analysis was performed by Dr Timothy Cooper (Penn State College of Medicine), and the percent of lesions morphologically resembling mild, moderate, or florid intraductal papillomas (IDPs); ductal carcinoma in situ (DCIS); or invasive carcinoma were calculated. The tumor sections also were used for immunohistochemical analysis of proliferation and apoptosis for all groups. Additionally, the mammary glands from 10 mice of each group were fixed in NBF for histology. At sacrifice, organ weights for liver, kidneys, spleen, mammary fat pads, and left gastrocnemius muscle were recorded.
Tumor cell proliferation and apoptosis
The percentages of Ki67 and cleaved caspase-3 positive cells counted in triplicate under a high power field (400X) were used as a measure of proliferation and apoptosis, respectively. Briefly, formalin-fixed paraffin-embedded mammary and MCF10AT1 tumor sections were hydrated, subjected to antigen retrieval, and incubated at room temperature with 1:100 anti-Ki67 antibody (M7240, Dako North America Inc, Carpinteria, CA) for 30 min and 1:100 cleaved-caspase-3 antibody (9661, Cell Signaling, Danvers, MA) for 1 h. Slides were developed using the Dako Envision™ +/HRP Polymer detection system (K4001, Dako North America Inc, Carpinteria, CA) and visualized with 3,3'-Diaminobenzidine (DAB) chromagen followed by hematoxylin counterstain. Staining was completed using a Dako Autostainer Plus.
Amino acid analysis
Plasma amino acid concentrations were measured using the Acquity UPLC (Waters Corporation, Milford, MA) with the AccQ.Tag Ultra derivatization Kit (Waters Corporation, Milford, MA). Plasma samples were deproteinized with a solution of 10 % sulfosalicylic acid and 250 pmol/μl of norvaline was used as the internal standard.
Assays on plasma parameters
Enzyme-linked immunosorbent assay (ELISA) kits were used to measure insulin (ALPCO Diagnostics, Salem, NH), leptin, IGF-1, adiponectin (R&D Systems, Minneapolis, MN), and FGF21 (Millipore Corp., Billerica, MA). Colorimetric assays were used to determine plasma triglycerides (TG) and total cholesterol (TC) (Thermo Electron Corp. Beverly, MA). Blood glucose was measured using an AbbottH Freestyle glucometer and glucose test strips.
Methionine restriction in cell culture
MDA-MB-231 and MCF10A cells were obtained from American Type Culture Collection (Manassas, VA). MDA-MB-231 cells were maintained in DMEM/F12 media with 10 % fetal bovine serum, and MCF10A cells were maintained in DMEM/F12 media containing 5 % horse serum with the addition of EGF (20 ng/ml), hydrocortisone (0.5 μg/ml), cholera toxin (100 ng/ml), and insulin (10 μg/ml). Cells were routinely passaged weekly and maintained in 5 % CO2 at 37 °C. Methionine restricted media were specially formulated without methionine, cysteine, and cystine (Life Technologies, Grand Island, NY), and the FBS or horse serum was dialyzed in PBS pH 7.2 three times over a period of 24 h at 4 °C using a Slide-A-Lyzer™ dialysis flask with a 3.5 K MWCO (Thermo Fisher Scientific, Waltham, MA) to remove all amino acids. Media were specially formulated to be 80 % reduced in methionine from the regular DMEM/F12 media (17.24 mg Met/L), with no cysteine or cystine, but with all amino acids and non-essential amino acids (methionine-restricted, cysteine-depleted (MRCD) media). Cells were plated and grown to 80 % confluence, washed once with PBS, and then given either MRCD or regular DMEM/F12 media. Twenty-four hours later, RNA was harvested from cells using Tri-Reagent (Molecular Research Center, Cincinnati, OH), according to the manufacturer’s instructions.
In vitro proliferation assay
Cells (MCF10A and MDA-MB-231) were plated with 40 μl of 1X105 cells/ml per well in a 96 well cell-culture treated dish. The next day cells were washed with PBS, and then treated with either MRCD or regular DMEM/F12 media. At 0 and 24 h cells were assayed for changes in proliferation by treating with the Cell Titer 96® Aqueous One Solution Cell Proliferation assay reagent, according to the manufacturer’s instructions (Promega, Madison, WI). Four experiments with 6 replicates per treatment were conducted. Results were analyzed using a 2-way ANOVA with Sidak’s comparison test.
Real-time PCR
RNA isolated from frozen tissue or cells from culture was converted to cDNA using the Verso cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. Primers were generated for HPRT, P21, P27, and cyclin D1 (Additional file
1: Table S1). Real-time PCR was performed using a SYBR-green GoTaq qPCR system (Promega, Madison WI), with 40 cycles at 95 °C for 15 s, and then 60 °C for 1 min using Applied Biosystems StepOne Plus Real-Time PCR System. Ct values were analyzed using the ΔΔ − Ct method [
25].
RNA isolation and RT-PCR from paraffin sections
Ten micron sections were obtained from the formalin-fixed, paraffin-embedded MCF10AT1 tumors and mammary fat pads of the athymic nude mice. Five sections were collected in a microcentrifuge tube, and RNA was isolated using miRCURY RNA isolation kit (Exiqon, Woburn, MA) according to the manufacturer’s instructions. RNA (250 ng) was converted into cDNA for real-time PCR analysis and analyzed as described above.
The Cancer Genome Atlas data analysis
The cBioPortal for Cancer Genomics (
http://www.cbioportal.org) was used to assess the changes in P21 and P27 expression in breast cancer patients from the Cancer Genome Atlas (TCGA) invasive breast carcinoma cancer study (TCGA, Nature 2012) [
26,
27]. This data set was screened for those patients containing data on mRNA levels from Agilent microarrays. Data retrieved from the TCGA controlled-access database was collected using tumors from patients who provided informed consent according to guidelines laid out by the TCGA Ethics, Law and Policy Group which is in compliance with the Helsinki Declaration (
http://www.wma.net/en/30publications/10policies/b3/index.html).
Statistical analysis and data presentation
Histological analyses of mammary and tumor sections from the nude mice were analyzed using an unpaired Mann-Whitney one-tailed t-test to determine significance. Ct values from RT-PCR results were analyzed unpaired one-tailed t-test to determine significance. In vitro proliferation assay data were analyzed using a 2-way ANOVA with Sidak’s comparison test. All analyses were performed using the statistical software package GraphPad PRISM (GraphPad Software Inc, La Jolla, CA).
Discussion
Dietary MR has been identified as a strategy for disease prevention and increased lifespan in experimental animals [
10,
19,
31,
32]. We hypothesize that MR may be used as a potential strategy for inhibiting carcinogenesis, and to test this hypothesis, we used the xenograft model for breast cancer by injecting MCF10AT1 cells into nude mice and examined the development tumors in these mice for 12 weeks. Our findings support this hypothesis and indicate that MR inhibits the growth of breast cancer tumors and induces apoptosis.
High levels of IGF1 and insulin in humans have been linked to increased risk of breast cancer [
33]. In carcinogenesis, growth factors, such as IGF1 and insulin, stimulate growth and progression of cancer [
34]. Therefore, the reduction in plasma IGF1 and insulin by MR suggest that the reduced levels in the insulin/IGF1 axis may inhibit tumor development in this xenograft model.
Methionine functions as the donor for the C2-C4 and amine nitrogen during the synthesis of the polyamines spermidine and spermine. Ornithine, the precursor of putrescine (which in turn is the precursor of spermidine and spermine), is increased in mice on the MR diet. Polyamines are involved in the regulation of proliferation, growth, and survival of cells [
35]. High levels of polyamines have been identified in several cancers [
36], and the inhibition of polyamine synthesis has been shown to have antitumor effects on skin, colon, and prostate cancers [
37,
38]. The increase in ornithine suggests an inhibition of polyamine production. The metabolism of polyamines is strictly controlled and contains two rate-limiting enzymes: ornithine decarboxylase and S-adenosyl methionine decarboxylase [
39]. Ornithine, spermidine, and spermine were previously reported to be increased in the liver of MR rats [
29]. In the Perrone et al. 2012 study [
29], gene expression of the two rate-limiting enzymes of polyamine synthesis in the liver were not inhibited. The conflicting findings regarding polyamine synthesis suggest that either MR effects on the tumor are independent of polyamine synthesis, or alternatively, polyamine function and synthesis are regulated differently in liver and mammary gland.
Taurine is a sulfur-containing amino acid that contributes to cell volume homeostasis and affects apoptosis mechanisms [
40]. Taurine plasma levels are decreased in mice on MR. In cancer, volume-sensitive taurine correlates with cisplatin resistance [
41]. Recently, taurine has been shown to induce apoptosis through PUMA (p53- up regulated modulator of apoptosis) in human colorectal, ovarian, and cervical cancer cells [
41‐
43]. A decrease in taurine levels resulted in reduced cell volume that induced levels of activated caspase-3, which led to apoptosis in cervical adenocarcinoma cells [
44]. Therefore, the MR effects on blood taurine levels (Table
2) could affect PUMA and lead to an increase in apoptosis in the mammary gland tissue and MCF10AT1 tumors of the MR mice, but this would need to be confirmed in future studies. Taurine has been examined as a novel tumor marker in female breast cancer patients. A decrease in serum taurine levels was observed in people with breast cancer or at high risk of breast cancer. Further, an inverse correlation between vascular endothelial growth factor (VEGF,marker for angiogenesis) and taurine concentrations has been demonstrated [
45]. Therefore, the connection between taurine, MR, and cancer is convoluted and complicated, and further research is needed to understand the implication of reduced plasma levels of taurine in MR mice.
Conclusion
In both the xenograft model and in breast cancer cell lines, the mechanism by which MR inhibits tumor progression appears to be a coordinated effort of inhibiting the cell cycle by stimulating the cell cycle inhibitors, P21 and P27. In both the MCF10AT1 tumors and surrounding mouse mammary tissue levels of P21 expression were elevated. Levels of P27 were not significantly changed in the xenograft model, suggesting that the effect of MR is more related directly to inhibiting P21 than P27. Both the metastatic breast cancer cell line (MDA-MB-231) and the immortalized breast cell line (MCF10A) responded to MR similarly to that seen in the xenograft model further supporting the relevance of P21 induction. In particular MDA-MB-231 cells had reduced cell proliferation after 24 h in MRCD media. Additionally, P27 also was elevated in the breast cancer cell lines, suggesting that it too may be involved in the suppression of tumor progression. Further evidence of P21 and P27 importance in patients with breast cancer was revealed in the analysis of the TCGA Breast Carcinoma Cancer study data set [
26,
27]. Survival analysis revealed that patients with decreased P21 or P27 expression had reduced survival. This suggests that, if MR increases P21 and/or P27 expression, there may be an increased length of survival in breast cancer patients.
The results from the present study indicate that MR may be a protective agent against cancer progression, but does not completely inhibit cancer progression. Therefore, the application of MR in a clinical setting could be both safe and a feasible option for arresting the progression of breast cancer while providing patients with more time to be treated by conventional methods to eradicate the cancer. Further studies are needed to examine the effect of MR in combination with chemotherapeutic and immunotherapeutic treatments.
Abbreviations
CC3, cleaved caspase-3; Cdkn1a, cyclin-dependent kinase inhibitor 1A (P21); Cdkn1b cyclin-dependent kinase inhibitor 1B (P27); CF, control fed; CR, caloric restriction; DAB, 3,3'-Diaminobenzidine; DCIS, ductal carcinoma in situ; ELISA, enzyme-linked immunosorbent assay; Her2, human epidermal growth factor receptor 2; IDPs, intraductal papillomas; MFP, mammary fat pad; MR, methionine restriction; MRCD, methionine-restricted, cysteine-depleted; NBF, neutral buffered formalin; PUMA, p53- up regulated modulator of apoptosis; ROS, reactive oxygen species; TC, total cholesterol; TCGA, the Cancer Genome Atlas; TG, plasma triglycerides; VEGF, vascular endothelial growth factor VEGF.
Acknowledgements
We thank Inez Augie from OFAS for analyzing the plasma amino acids. The MCF10AT1 tumors and mammary fat pads were paraffin embedded and sectioned at the Microscopy and Histology Facility, PSU College of Medicine, Hershey, Pennsylvania. Immunohistochemistry was performed by Trey Bruggeman at the Molecular and Histopathology Core Facility, PSU College of Medicine. We thank the staff at OFAS for proof-reading this manuscript. Funding for this project was from the Orentreich Foundation for the Advancement of Science, Inc.