Targeting Her2-Positive Breast Cancer Metabolism Through Dietary and Pharmacological Interventions to Improve Therapeutic Efficacy

TARGETING HER2-POSITIVE BREAST CANCER METABOLISM THROUGH DIETARY AND PHARMACOLOGICAL INTERVENTIONS TO IMPROVE THERAPEUTIC EFFICACY

Magdalena A. Rainey

A thesis submitted to the faculty at the University of North Carolina at Chapel Hill in partial fulfillment of the requirements for the degree of Master of Science in the Department of Nutrition in the Gillings School of Global Public Health.

Chapel Hill 2019

Approved by:

Stephen D. Hursting

John E. French

Alyssa J. Cozzo

ABSTRACT

Magdalena A. Rainey: Targeting HER2-Positive Breast Cancer Metabolism Through Dietary and Pharmacological Interventions to Improve Therapeutic Efficacy (Under the direction of Stephen D. Hursting)

HER2-positive breast cancers (HER2+ BC) are an aggressive subset characterized by therapeutic resistance in ~40-60% of patients. We hypothesized that modulating energy- responsive pathways through either dietary restriction or pharmacological metabolic reprogramming interventions (MRI) would increase the efficacy of HER2-directed therapies trastuzumab and lapatinib. Dietary restriction was modeled in vitro and the efficacy of lapatinib was increased under serum-restricted conditions. This effect was recapitulated with

MRIs and we determined that inhibition of IGF1R/insulin receptor and mTORC1 increased the antiproliferative effects of lapatinib. Ongoing work is investigating the impact of intermittent fasting regimens on therapeutic efficacy in vivo by restricting the dietary intake of mice with HER2+ BC for 48 hours coinciding with the administration of trastuzumab or lapatinib. We conclude that inhibition of metabolic pathways implicated in HER2+ BC drug resistance through intermittent fasting regimens and pharmacological MRIs has the potential to increase the efficacy of HER2-directed therapies.

iii

ACKNOWLEDGEMENTS

I would like to thank Dr. Hursting for his unwavering encouragement throughout my time in his laboratory. His excitement for pursuing research with the goal of improving human health has inspired me to continue cancer research in my medical career. I am extraordinarily grateful to have had a patient and kind mentor in Laura Smith, who taught me all of my laboratory skills and worked with me to develop this project in its entirety. Laura’s perseverance through the inevitable highs and lows in basic science research have allowed me to understand the stamina that this career requires and to enjoy pursuing the work along the way. The time that Alyssa Cozzo has dedicated to not only developing this project with me, but also teaching me how to rigorously evaluate the scientific process in each step, have given me lifelong skills that I will use in every research endeavor moving forward. I am thankful that Alyssa pushed me to ask the right questions, select better analytical methods, and write more effectively. I hope that I find another mentor that is as encouraging and dedicated to my development as a scientist as Alyssa has been. Further, I owe many thanks to Michael Coleman for his brilliance in designing studies and troubleshooting laboratory techniques. Mike has been a fantastic mentor, and I am inspired by his love for science. My close friend and colleague, Jane Pearce has been instrumental in carrying out this project day to day. I am immensely grateful for her dedication to the Hursting laboratory and for her grit while working long hours in the lab besides me. Finally, I would like to thank our laboratory manager, Erika Rezeli, for her kindness in assisting me with the day to day aspects of my project. Being part of the Hursting laboratory has been a tremendously rewarding experience, and I am thankful for the opportunity to contribute to this team during my three years.

TABLE OF CONTENTS

LIST OF FIGURES ...... vii

LIST OF ABBREVIATIONS ...... viii

CHAPTER 1: INTRODUCTION ...... 1

HER2-Positive Breast Cancer Overview ...... 1

Signaling Mechanisms...... 1

Therapeutic Strategies ...... 3

Drug Resistance ...... 5

Project Goals ...... 7

CHAPTER 2: NUTRIENT STRESS ALTERS CELL GROWTH AND PROLIFERATION OF HER2-POSITIVE BREAST CANCER CELLS IN VITRO ...... 8

Background ...... 8

Methods ...... 9

Results ...... 12

Nutrient stress altered cellular viability and growth...... 12

Serum restriction promoted cellular apoptosis ...... 14

Serum restriction induced a G0/G1 arrest ...... 15

Serum restriction reduced cellular viability in human and murine-derived HER2+ BC cell lines ...... 17

Discussion ...... 18

CHAPTER 3: METABOLIC REPROGRAMMING INTERVENTIONS IMPROVE RESPONSE TO LAPATINIB AND TRASTUZUMAB IN VITRO ...... 21

Background ...... 21

Metabolic Reprogramming in Breast Cancer ...... 21

Perturbing Glutamine Metabolism ...... 24

Perturbing Growth Factor Signaling ...... 24

Autophagy in HER2-Positive Breast Cancer ...... 25

Methods ...... 27

Results ...... 29

Serum restriction decreased cellular viability in combination with HER2-directed therapies ...... 29

Metabolic reprogramming interventions modified activation of downstream targets ...... 32

Inhibition of IGF1R and mTOR increased efficacy of lapatinib...... 34

Dual inhibition of mTORC1 and autophagy increased efficacy of trastuzumab ...... 36

Discussion ...... 38

CHAPTER 4: ONGOING AND PLANNED STUDIES – IMPACT OF INTERMITTENT FASTING REGIMENS ON TUMOR PROGRESSION AND RESPONSE TO HER2-DIRECTED THEAPIES IN VIVO ...... 41

Background ...... 41

Metabolic Impact of Dietary Restriction ...... 41

Intermittent Fasting ...... 42

Fasting and Anti-Cancer Therapy ...... 44

Animal Design Protocol ...... 45

Planned Analysis ...... 49

Anticipated Results ...... 50

CHAPTER 5: CONCLUSIONS ...... 52

REFERENCES ...... 53

LIST OF FIGURES

Figure 1. HER2 breast cancer signaling mechanisms ...... 3

Figure 2. HER2-directed therapeutic agents ...... 5

Figure 3. Nutrient stress altered HER2+ breast cancer cellular viability and proliferation .... 13

Figure 4. Serum restriction promoted cellular apoptosis ...... 15

Figure 5. Serum restriction induced G0/G1 arrest ...... 16

Figure 6. Serum restriction reduced cellular viability ...... 18

Figure 7. Metabolic reprogramming interventions ...... 23

Figure 8. Serum restriction increased growth inhibitory effect of lapatinib ...... 30

Figure 9. Efficacy of trastuzumab was not improved by serum restriction ...... 31

Figure 10. Target inhibition was achieved with MRIs...... 34

Figure 11. MRIs in combination with lapatinib...... 35

Figure 12. MRIs in combination with trastuzumab ...... 37

Figure 13. MMTV/neu mouse study pilot and design...... 48

Figure 14. BT-474 mouse study design ...... 48

vii

LIST OF ABBREVIATIONS

ADCC Antibody-dependent cellular cytotoxicity

AKT RAC-alpha serine/threonine-protein kinase

AMP Adenosine monophosphate

ANOVA Analysis of variance

ASCT2 Glutamine transporter

ATG5, 7, 12 Autophagy-related proteins

BC Breast cancer

BCS Bovine calf serum

Beclin1 Autophagy-related protein

BMS IGF1R/IR inhibitor

BPTES Bis-2-(5-pheylacetamido-1,2,4-thiadiazol-2-yl)ethyl

CD3, CD8 T cells cMET Tyrosine-protein kinase Met

CQ Chloroquine

CR Caloric or calorie restriction

DMEM Dulbecco's Modified Eagle Medium

DMSO Dimethyl sulfoxide

ER Estrogen receptor

ERBB2 Erb-b2 receptor tyrosine kinase 2

ERK 1/2 Extracellular signals regulated kinases 1 and 2

ERR1 Steroid hormone receptor ERR1

F4/80 Macrophages marker

FBS Fetal bovine serum

Glc Glucose

GLS Glutaminase

GLUT 1, 4 Glucose transporters

viii

GPNA L-Glutamic acid γ-(p-nitroanilide)

GSEA Gene set enrichment analysis

HER1, 2, 3, 4 Human epidermal growth factor receptor subtypes

IGF1 Insulin-like growth factor 1

IGF1R Insulin-like growth factor 1 receptor

IR Insulin receptor

LC3B Microtubule-associated protein 1A/1B-light chain 3

MAPK Mitogen-activated protein kinase

MMTV/neu FVB/N-Tg(MMTVneu)202Mul/J mTOR Mammalian target of rapamycin

MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

PBS Phosphate-buffered saline pCR Pathologic complete response

PFKFB2 Fructose-2,6-biphosphatase 2

PI Propidium Iodide

PI3K Phosphatidylinositol 3-kinase

PIP2, PIP3 Phosphorylate plasma membrane intrinsic protein

PTEN Phosphatidylinositol 3,4,5-trisphosphate 3-phosphatase

RTK Receptor tyrosine kinase

S6K1 Ribosomal protein S6 kinase beta-1

SEM Standard error of mean

T-DM1 Trastuzumab emtansine

TCA Tricarboxylic Acid Cycle

UFO Tyrosine-protein kinase receptor UFO

UV Ultraviolet

CHAPTER 1: INTRODUCTION

HER2-Positive Breast Cancer Overview

In the United States, breast cancer (BC) is the second leading cause of cancer death among women, with an estimated 252,000 new invasive breast cancer cases diagnosed annually (1). Human epidermal growth factor receptor (HER2)-positive breast cancers are defined by overexpression of the HER2 protein or amplification of the HER2 gene copy number (1). This molecular subtype accounts for 15-20% of early breast cancers and 35% of locally advanced and metastatic breast cancers (2).

Compared to HER2-negative breast cancers, at time of diagnosis HER2-positive breast cancers are typically larger in size, more likely to be node-positive, and classified as a higher grade (3). Importantly, the growth and proliferation of these tumor cells is primarily driven by HER2, a phenomenon termed oncogene addiction.

For this reason, the advent of HER2-directed therapeutics, including antibodies, antibody-drug conjugates, and tyrosine kinase inhibitors has demonstrated long-term clinical benefits and improved patient outcomes. However, new strategies to combat relapse, metastasis, and intrinsic and acquired resistance to therapies are required (4).

Signaling Mechanisms

The gene encoding HER2 (ERBB2) is a member of the ErbB receptor tyrosine kinase family. Homo- or heterodimerization of HER2 with one of multiple ligand-bound epidermal growth factor receptors (HER1, HER3, HER4) results in phosphorylation of the intracellular domains of the receptor tyrosine kinases (RTK) and recruitment of molecules to initiate several signaling cascades (5). The most common HER2 signaling pathways include the phosphatidylinositol 3-kinase /RAC-alpha serine/threonine-protein

kinase (PI3K/AKT) and mitogen-activated protein kinase (MAPK) cascades (6) (see

Figure 1). In the MAPK cascade, adaptor proteins bind phosphorylated HER2 and activate RAS, a protein kinase, that initiates a cascade to activate extracellular signals- regulated kinases 1 and 2 (ERK1/2). ERK interacts with kinases and transcription factors to promote cell growth, proliferation, and survival (7). The PI3K/AKT pathway is activated by HER2/RTK dimerization and intracellular phosphorylation. The heterodimer PI3K is recruited and activated to phosphorylate plasma membrane intrinsic protein (PIP2) to PIP3. PIP3 induces translocation of AKT to the plasma membrane where it regulates proteins involved in cell survival. Through these interactions, the PI3K/AKT activation is involved with inhibition of apoptosis and activation of survival pathways (8). These pathways can be inappropriately activated in cells that overexpress HER2 via HER2-induced changes in HER1 autophosphorylation of specific tyrosine residues, resulting in stabilized signaling through HER1 (9). This change in tyrosine phosphorylation patterns bypasses regulatory mechanisms, increasing the oncogenic potential of HER1, and inducing cell transformation and tumorigenesis (9).

Figure 1. HER2 breast cancer signaling mechanisms

Therapeutic Strategies

The aberrant growth signals promoted by HER2 dimerization are effectively blocked with a regimen of HER2-directed therapies used in combination with chemotherapy. In HER2-positive early breast cancer, the standard of care is a combination of neoadjuvant chemotherapy, breast surgery, and 12 months of trastuzumab, a humanized monoclonal antibody directed against the extracellular

domain of HER2 that blocks signaling independent of ligand activity (10-12). For the treatment of metastatic HER2-positive breast cancer, trastuzumab combined with paclitaxel or docetaxel is the standard first-line treatment (13). In advanced stages of this disease, lapatinib, a small-molecule tyrosine kinase inhibitor, is inducated (14,

15).

Lapatinib is a reversible inhibitor of HER1 and HER2 tyrosine kinases which prevents autophosphorylation by blocking the cytoplasmic ATP-binding domain (16).

In clinical trials, weekly paclitaxel chemotherapy with trastuzumab alone, lapatinib alone, or trastuzumab combined with lapatinib significantly increased the rate of pathologic complete response (pCR) compared to trastuzumab alone (46.8% in combination and 27.6% in trastuzumab alone); however, in other clinical trials, there were no significant differences pCR when trastuzumab was combined with lapatinib following treatment with doxorubicin, cyclophosphamide and paclitaxel (12). Other tyrosine kinase inhibitors being explored for the treatment of HER2-positive breast cancer, including afatinib and neratinib, irreversible inhibitors of HER1/2/4 have yet to be proven more effective in combination with trastuzumab than the existing standard of care (12).

In contrast, there are several proposed mechanisms for the actions of trastuzumab including HER2 internalization and degradation, induction of antibody- dependent cellular cytotoxicity (ADCC), and interference with the MAPK and

PI3K/AKT signaling cascades (17). More recently, the addition of pertuzumab, a humanized monoclonal antibody that works complementary with trastuzumab by blocking HER2/HER3 dimerization with HER family receptors, has been approved for use in the metastatic and neoadjuvant settings in combination with trastuzumab

and docetaxel (18). Contributing to the arsenal of anti-HER2 therapies, trastuzumab emtansine (T-DM1) has also been approved. T-DM1 is a conjugate of trastuzumab and the cytotoxic agent DM1, which inhibits microtubule polymerization. T-DM1 is currently approved for metastatic HER2-positive breast cancer. Unlike the combination of trastuzumab and standard chemotherapy, T-DM1 reduces systemic cytotoxic exposure by directing its effects to HER2-overexpressing cells alone; however, the combination of T-DM1 plus pertuzumab has not demonstrated clinically meaningful improvement in progression-free survival compared to trastuzumab plus taxane regimens (19).

Figure 2. HER2-directed therapeutic agents

Drug Resistance The clinical improvements that have been achieved since the introduction of

HER2-directed therapies is especially remarkable in the metastatic setting where overall survival has been extended to 4.5 years with the use of trastuzumab, pertuzumab, and chemotherapy compared to chemotherapy alone (20). However, the development of new therapeutic strategies is imperative because approximately 40 to

60 percent of patients exhibit intrinsic resistance or develop secondary resistance to

trastuzumab (21). The proposed mechanisms that cancer cells use to evade trastuzumab-induced cell death include impaired binding of the antibody due to mutations in the HER2 kinase domain, failure to trigger immune-mediated destruction, upregulation of HER2 downstream signaling pathways, and signaling through alternate pathways via activation of insulin-like growth factor 1 receptor (IGF1R), tyrosine- protein kinase Met (cMET), and HER1 (22, 23). Additional mechanisms of trastuzumab-specific resistance that have been proposed include HER2 proteolysis,

Mucin-4 overexpression, and loss of phosphatidylinositol 3,4,5-trisphosphate 3- phosphatase (PTEN) (24). Recent work has identified that high expression and activation of specific RTKs, including IGF1R, TYRO3, and PDGFR-beta, are involved in anti-HER2 drug resistance, as signaling through these receptors maintains activation of pro-growth pathways and suppress cellular senescence (25). HER2- overexpressing cells also use other RTKs to evade cell death upon treatment with lapatinib. Similar to mechanisms of trastuzumab resistance, some cancer cells are able to reactivate the PIK3/AKT and mammalian target of rapamycin (mTOR) signaling axes by overexpressing the tyrosine-protein kinase receptor UFO (26). Alterations in receptor tyrosine kinase activities in response to treatment with HER2-directed therapy modulates distinct metabolic effects that allow cancer cells to evade growth inhibition and cell death. In models of drug resistance for trastuzumab and lapatinib, alterations in glucose and glutamine utilization have been observed. For example, in HER2- overexpressing cells, treatment with lapatinib modulates phosphorylation of a residue on fructose-2,6-biphosphatase 2 (PFKFB2), an important mediator of glycolysis that reduces glucose flux (27). However, in lapatinib-resistant cells, this phosphorylation event is diminished, allowing for maintenance of flux through glycolysis and increased expression of proteins in the glucose-deprivation network (27). Glucose and glutamine utilization are also impacted by reactivation of mTOR signaling via AXL (UFO) overexpression in lapatinib-resistant cells, as previously described. Activation of

mTOR restores expression of steroid hormone receptor ERR1, a nuclear receptor that acts as a master regulator of cellular metabolism. While lapatinib-sensitive cells have decreased flux in glucose and glutamine utilization, re-activation of ERR1 in resistant cells mediates transcriptional reprogramming changes to re-establish glutamine metabolism in the presence of the drug (28). These mechanisms of drug resistance establish an important association between cellular energetics and response to anti-

HER2 therapies. Together, these studies argue for approaches which induce broad changes in cellular metabolic pathways in HER2-positive breast cancer cells, targeting flux through glucose and glutamine metabolism through regimens of dietary restriction or pharmacologic interventions that reprogram specific metabolic pathways.

Project Goals The goals of the project are as follows: 1) Characterize the phenotypic and mechanistic impact of nutrient restriction on HER2-amplified breast cancer cells. 2)

Determine the impact of short-term fasting regimens on tumor progression and response to

HER2-targeting therapies in vivo. 3) Identify pharmacologic interventions that target pathways associated with nutrient stress to attenuate growth and increase therapeutic response in HER2-positive breast cancer cells.

CHAPTER 2: NUTRIENT STRESS ALTERS CELL GROWTH AND PROLIFERATION OF HER2-POSITIVE BREAST CANCER CELLS IN VITRO

Background

The Hursting lab has identified an association between dietary energy balance and breast cancer progression across multiple BC subtypes (29). Through preclinical studies on MMTV-HER2/neu transgenic mice, which hyperactively express the HER2 oncogene, the impact of dietary energy balance on HER2-amplified breast cancer has been examined. Chronic caloric restriction (CR) was modeled by placing female MMTV-

HER2/neu transgenic and wild type FVB mice on either a low-fat control regimen (10 kcal% fat), a 30% CR regimen relative to control, or a high-fat diet induced obesity (DIO,

60 kcal% fat) regimen (30). Compared to the control and DIO groups, calorie restricted mice had decreased adiposity and obesity-related serum markers including IGF1, insulin, leptin, as well as preserved ER-alpha and ER-beta expression, increased disease-free survival rates, and reduced tumor burden (30). These novel findings highlight the need for further investigation of the mechanisms underlying these diet- gene interactions in order develop strategies to exploit its metabolic targets without prolonged dietary restriction.

Relative to untransformed cells, cancer cells have increased reliance on utilization of glucose for proliferative activities and growth factor signaling in serum for mitogenic activities. We developed an in vitro model to test whether breast cancer cells would show reduced cellular viability under low glucose and serum conditions, to test the hypothesis that this dependence may underlie the in vivo sensitivity of breast cancer cells to dietary interventions such as caloric restriction or fasting.

Methods

Cell Lines Used

Murine E18-9A-42 cells that were isolated from a spontaneous mammary tumor generated in transgenic FVB/N-Tg(MMTVneu)202Mul/J mice, referred to as MMTV/neu breast cancer cells, were maintained in Dulbecco's Modified Eagle Medium (DMEM) containing 4.5g/L glucose supplemented with 10% bovine calf serum (BCS) and 1% penicillin/streptomycin. The human breast cancer cell lines SKBR3 (ATCC HTB-30) and BT-

474 (ATCC HTB-20) derived from breast adenocarcinoma and ductal carcinoma, respectively, were cultured in RPMI 1640 (phenol free) supplemented with 10% fetal bovine serum (FBS), glucose (4.5g/L), HEPES buffer (10mM), sodium pyruvate (1mM), and 1% penicillin/streptomycin. Cells were maintained at 37C in a 5% humidified chamber. Nutrient stress conditions were modeled in vitro through manipulation of media glucose and serum concentrations as indicated.

Nutrient Stress Conditions in vitro

Murine E18-9A-42 cells that were isolated from a spontaneous mammary tumor generated in transgenic FVB/N-Tg(MMTVneu)202Mul/J mice, referred to as MMTV/neu breast cancer cells, were maintained in Dulbecco's Modified Eagle Medium (DMEM) containing 4.5g/L glucose supplemented with 10% bovine calf serum (BCS) and 1% penicillin/streptomycin. To initially investigate the impact of nutrient stress in vitro we cultured cells in severe glucose- and serum-restricted conditions that are beyond physiologic parameters. Cells were treated with medias containing reduced serum (1%

BCS), reduced glucose (1mM Glc), or reduced serum and reduced glucose (1% BCS/1mM

Glc) compared to control media (10% BCS/25mM Glc) (31). Following these initial experiments, nutrient stressed conditions were expanded to include more physiologically relevant conditions in vitro. MMTV/neu cells were cultured in medias with lower glucose concentrations as the reference (10mM) and intermediate levels of reduced serum (5% and

1% BCS) and reduced glucose (5mM and 2.5mM Glc) were compared to control (10%

BCS/10mM Glc). Cells were maintained at 37C in a 5% humidified chamber.

The human breast cancer cell lines SKBR3 (ATCC HTB-30) and BT-474 (ATCC

HTB-20) derived from breast adenocarcinoma and ductal carcinoma, respectively, were cultured in RPMI 1640 (phenol free) supplemented with 10% fetal bovine serum (FBS), glucose (4.5g/L), HEPES buffer (10mM), sodium pyruvate (1mM), and 1% penicillin/streptomycin. To model nutrient restriction in vitro, serum levels were modified (5% and 1% FBS) and glucose concentration was reduced (5mM and 2.5mM Glc). Control media

(10% FBS/ 10mM Glc) was used as the comparator.

MTT assay

MMTV/neu cells were seeded (2x103 cells/well) in a 96-well plate overnight, serum and/or glucose restricted for 24 hours, and then treated with one of the nutrient-restricted medias for an additional 24 hours. Media was aspirated and cells were incubated in a solution containing 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (0.5 mg/mL) in PBS. After three hours, MTT reagent was aspirated and cells and precipitate were solubilized with dimethyl sulfoxide (DMSO) and agitation on a plate shaker for 5 minutes. UV spectroscopy was used to measure absorbance at 570 and 690 nm via a

Cytation 3 Cell Imaging Reader (BioTek).

Cellular Proliferation

MMTV/neu cells were seeded (2.5x104 cells/well) in a 6-well plate overnight and treated with nutrient restricted-medias as indicated. At treatment time intervals of 24, 48, 72, and 96 hours, cells were trypsinized (Trypsin-EDTA 0.05%, Gibco) and counted with a

Countess II FL Automated Cell Counter (Thermo Fisher Scientific).

Apoptosis Assay

MMTV/neu cells were seeded (2.5x104 cells/well) in a 6-well plate overnight, serum and glucose starved for 24 hours, then treated with nutrient-restricted medias for an additional 24 hours. Cells were trypsinized and stained using the Annexin V-FITC Dead Cell

Apoptosis Kit for Flow Cytometry (ThermoFisher) according to the manufacturer’s protocol.

Data was acquired using a BD Accuri C6 cytometer (BD Biosciences). Laser excitation at

488nm was used to capture 30,000 events for each sample. PI fluorescence intensity was plotted against FITC fluorescence intensity. Live cells were categorized as PI and Annexin

V-FITC negative, early apoptotic cells as Annexin V-FITC positive, and late apoptotic or dead cells as both PI and Annexin-V FITC positive.

Cell Cycle Analysis

MMTV/neu cells were seeded (2.5x104 cells/well) in a 6-well plate overnight and treated with either complete or nutrient-restricted medias for 24 hours. Cells were fixed, permeabilized, and stained with propidium iodide (PI) for flow cytometric analysis of cell cycle according to manufacturer’s instructions (Abcam). Single cells were analyzed using a

CyAn ADP High-Performance Flow Cytometer (Beckman Coulter) and PI emissions were recorded at 488nm. Forward scatter and side scatter were assessed for 10,000 events and plots were gated to exclude cellular debris.

Calcein-AM Live Cell Assay

MMTV/neu (5x102 cells/well), SKBR3 (2x103 cells/well), and BT-474 (5x103cells/well) cells were seeded into black 96-well, clear bottom plates in complete media overnight. Cells were cultured in control or nutrient-restricted media as indicated for 48 hours (MMTV/neu) or

72 hours (SKBR3 and BT-474). Following incubation, plates were centrifuged at 1,500 rpm for 5 minutes, media was aspirated, cells washed with PBS, and plates spun a second time.

PBS was removed and cells were incubated in labeling solution containing the lipophilic reagent Calcein-AM (1M) dissolved in PBS for 40 minutes at room temperature. The fluorescent intensity of calcein was captured at an excitation/emission spectrum of

490/520nm using a Cytation 3 Cell Imaging Reader (BioTek). The relative intensities captured in each well were normalized to the average of the control.

Statistical Analysis

Data were analyzed using GraphPad Prism 8 Software (GraphPad Software, Inc.).

Differences between means were assessed using one-way analysis of variance (ANOVA) with post-hoc Tukey’s multiple comparison test for statistical differences. Data are graphically presented as mean ± standard error of mean (SEM) unless otherwise noted.

Results

Nutrient stress altered cellular viability and growth

Examination of the phenotypic changes that occur in HER2-overexpressing breast cancer under conditions of nutrient stress was completed by comparing murine MMTV/neu cells cultured under control conditions (10% serum, 25 mM glucose) to those grown in serum (1% BCS)- or glucose-restricted (1mM Glc) media, or a combination of serum and glucose restriction (1% BCS and 1mM Glc, respectively), for 24 hours. Cellular viability was assessed via MTT viability assay, which measures the concentration of metabolically active cells in a given well. Following 24 hours of nutrient restriction, MMTV/neu cells cultured in reduced serum media had significantly decreased relative abundance of viable, metabolically active cells compared to cells cultured in complete media (Fig. 2A, 1% BCS p=0.02; 1% BCS and 1mM Glc, p=0.03). However, glucose restriction alone had a moderate non-statistically significant reduction in relative cellular viability during a 24-hour treatment period.

Next, changes in cell number were examined under the same nutrient stress conditions by measuring the number of cells under various culture conditions over 24-hour intervals (Fig. 2B). The greatest differences in cell number compared to control were measured at 96 hours, as serum restriction (p=0.0002), glucose restriction (p=0.0013), and the combination of both (p=0.0003) significantly reduced cellular proliferation in vitro (Fig.

2C) at that time point. Taking the findings from Figs 2A-2C together, the data suggest that reduced serum availability and (to a lesser extent) glucose restriction, induce a reduction in metabolic activity and cell number in MMTV/neu HER2+ BC cells. Moreover, no additive or

synergistic effect of combining serum and glucose restriction on viability or number of

MMTV/neu HER2+ BC cells was observed.

1.5 e

g a,b n

a a

h C 1.0 d b

l b

l i

b 0.5

l e

C 0.0 l o S lc lc tr C G G n B M M o % C 1 m m 1 /1 S C B % 1

B C Nutrient Restriction - 96 Hours 30 30

) a

)

1 1

20 x

(

1 20

(

r e

b b

m u

10 u N

l b l

e b

C C

0 0 l 0 24 48 72 96 o S lc lc tr C G G n B o M M Time (hours) C % m m 1 1 1 / Control 1% 1mM 1%/1mM S C B % 1

Figure 3. Nutrient stress altered HER2+ breast cancer cellular viability and proliferation MMTV/neu cells were cultured under control conditions (10% BCS and 25 mM glucose), serum restriction (1% BCS), glucose restriction (1mM glucose), or a combination of both (1% BCS and 1mM glucose). (A) Cellular viability following 24 hours of nutrient restriction was measured using an MTT viability assay (1% p=0.02; 1%/1mM p=0.03 vs. control). Data from n=4 experimental replicates is presented as mean ± SEM. (B) Number of cells in culture over 24-hour intervals in nutrient-restricted conditions. (C) Cell number following 96 hours of culture. Serum restriction (p=0.0002), glucose restriction (p=0.0013), or the combination of both (p=0.0003) significantly reduced cellular proliferation in vitro relative to control conditions. Data from n=3 experimental replicates is presented as mean ± SEM. (B- C) Analyzed via one-way ANOVA with post-hoc Tukey’s multiple comparisons.

Serum restriction promoted cellular apoptosis

To identify potential mechanisms through which cellular viability and proliferation are altered by nutrient stress, we next determined whether the reduced cell number in response to nutrient restriction observed in Figure 2 B-C was primarily due to cell death or to a reduction in the number of cells actively progressing through the cell cycle. We measured the proportion of apoptotic and dead cells relative to control in each of the treatment conditions previously described. Annexin V is a protein which efficiently binds phosphatidylserine that has translocated to the outer surface of the cell membrane, indicating the induction of apoptotic signaling, while propidium iodine (PI) is a nuclear dead cell stain that emits a red signal upon laser excitation. In this assay, annexin V was labeled with the fluorochrome FITC and Annexin V- and/or PI-positive cells were detected via flow cytometry (Fig. 3A). Quantification of the number of dead and apoptotic cells relative to control demonstrated that under nutrient-stressed conditions, serum restriction increased cellular apoptosis compared to control (Fig. 3B). Cells cultured in glucose-restricted media alone demonstrated a modest, non-significant increase in apoptosis compared to control.

Figure 4. Serum restriction promoted cellular apoptosis MMTV/neu cells were cultured in nutrient-restricted medias (control, 1% BCS, 1mM glucose, 1% BCS and 1mM glucose) for 48 hours, and apoptotic cell death was measured using Annexin V/PI staining via flow cytometry. (A) Representative image of flow cytometry gating. (B) Percentage of dead and apoptotic cells following nutrient restricted conditions. Data from n=2 experimental replicates is presented with means.

Serum restriction induced a G0/G1 arrest

The impact of nutrient stress on cell cycle regulation was further explored by utilizing propidium iodine staining to quantify the proportion of cells in each phase of the cell cycle. In this assay, the proportion of propidium iodine staining correlates with the amount of DNA present in the cell, where two copies of each chromosome (2N) characterizes non-dividing cells, either senescent cells lacking replicative potential or cells in the G1 or G0 phases.

Cells actively replicating DNA (2N-4N) are in the S phase. Following DNA replication, cells enter G2 or M phase (4N). Sharp peaks furthest left on histograms of forward scatter versus side scatter following propidium iodide staining are characteristic of the G1/G0 phase, and peaks shifting to the right correspond to the S and G2/M phases, respectively (Fig. 4A).

Cells cultured in serum-restricted media experienced G0/G1 arrest compared to control (Fig.

3B 1% p=0.0031; 1%/1mM p=0.0039). However, none of the nutrient restriction groups had significantly different portions of cells in the S or G2/M phases (Fig. 3B).

Figure 5. Serum restriction induced G0/G1 arrest Cell cycle analysis of MMTV/neu cells cultured in serum- or glucose-restricted media. (A) Representative image of cell cycle flow cytometric analysis. (B) Frequency of cells in each phase of the cell cycle. Serum restriction induced G0/G1 arrest compared to control (1% p=0.0031; 1%/1mM p=0.0039). Data from n=3 experimental replicates presented as mean ± SEM.

Serum restriction reduced cellular viability in human and murine-derived HER2+ BC cell lines

To increase the translational potential of these findings, the culture conditions were expanded to include human HER-positive cell lines (BT-474 and SKBR3), as well as more physiologically relevant glucose levels (5mM and 2.5mM Glc) and serum restriction concentrations (5% and 1%) than used in our previous experiments. The control group used for comparison was 10% serum/10mM glucose, as compared to the 10% serum/ 25mM glucose control used in the previously described experiments. In agreement with our initial, proof-of-concept findings, serum restriction to 1% reduced the number of live cells relative to control across the three breast cancer cell lines, as assessed using a calcein AM viability assay (Fig. 5A MMTV/neu p<0.0001; Fig. 5B BT-474 p=0.0002; Fig. 5C SKBR3 p<0.0001).

Interestingly, an intermediate level of serum restriction (5% serum) decreased the live cell number only in BT-474 cells (Fig. 5B; p=0.0092). Surprisingly, limiting the glucose concentration in MMTV/neu cells to 2.5mM resulted in a greater number of live cells relative to control (Fig. 5D; p=0.0001). However, consistent with our initial findings using the

MMTV/neu mouse cell line, human HER+ breast cancer cells did not demonstrate sensitivity to decreased glucose availability (Figure 5E-F). Together these results indicate that manipulation of glucose availability is less impactful on HER2+ BC survival and/or proliferation than limitation of other components found in serum, such as hormones, growth factors or other signaling proteins, or unidentified essential nutrients.

A B C

MMTV/neu BT-474 SKBR3

r r

r 1.5 1.5 1.5

e e

**** %

b 0 0 *** ****

1 **

l l

1.0 l 1.0 1.0

i i

0.5 0.5 i 0.5

R R 0.0 0.0 R 0.0 10 5 1 10 5 1 10 5 1 [Serum] (%) [Serum] (%) [Serum] (%)

D E F

MMTV/neu BT-474 SKBR3

r r

r 4 *** 1.5 1.5

M M

M 3

l l

l 1.0 1.0

t t

t 2

i i

i 0.5 0.5

v v

v 1

R R R 0 0.0 0.0 10 5 2.5 10 5 2.5 10 5 2.5 [Glucose] (mM) [Glucose] (mM) [Glucose] (mM)

Figure 6. Serum restriction reduced cellular viability Murine- and human-derived HER2-positive breast cancer cells were treated with nutrient- restricted media for 48 and 72 hours, respectively, and cell viability was assessed via calcein AM viability assay. Low-serum conditions reduced the number of live cells relative to control conditions (A) p<0.0001 (B) p=0.0092 (5%), p=0.0002 (10%) (C) p<0.0001. (D) Glucose restriction increased cellular viability in MMTV/neu cells (p=0.0001 2.5 mM vs 10 mM). Technical replicates (n=3) graphically represented with mean ± SEM.

Discussion In this study, an in vitro model of dietary restriction was developed to test the hypothesis that low-energy and/or nutrient availability underlies the in vivo sensitivity of breast cancer cells to dietary restriction. Murine MMTV/neu and human SKBR3 and BT-474 cell lines were cultured under low glucose and serum conditions. It was determined that restriction of serum, but not restriction of glucose, results in reduced cellular metabolic activities and cellular viability. These changes were mediated through reduced cellular proliferation, increased cellular apoptosis, and dysregulation of cell cycle maintenance

patterns. While these mechanisms were not confirmed in the human cell lines used in this study, the BT-474 and SKBR3 HER2-positive cell lines demonstrated similar patterns of reduced cellular viability in response to serum restriction, but not glucose restriction.

Considering that the major components present in serum include proteins, vitamins, minerals, insulin and related growth factors, and major hormones for growth promotion, there are many different mechanisms through which reducing serum concentration in the media may have facilitated these changes. Of particular interest is reduced signaling through IGF1R and mTOR which occur in response to caloric restriction regimens. The nutrient-sensing IGF1R/mTOR pathway is stimulated by insulin and amino acids and plays an important role in stimulating growth and proliferation. Lamming, et al. found that mice on a protein restricted diet bearing ER+/PR-/HER2- breast tumor xenografts had decreased circulating amino acid levels and reduced phosphorylation of mTORC1 (32). Further, human studies have demonstrated the importance of modest protein restriction in a chronic CR regimen in order to modulate anti-cancer effects associated with decreased IGF1 levels

(33). This highlights the importance of examining the impact of caloric restriction with respect to macronutrient composition of dietary regimen in vivo or corresponding mimetic in vitro. To test the hypothesis that reduced IGF1R activation under serum-restricted conditions is the primary driver of reduced cellular viability, future experiments will be conducted by 1) supplementing exogenous IGF1 to serum-restricted media and 2) using an anti-IGF1R antibody in complete serum conditions. We predict that HER2-positive breast cancer cells will not have reduced viability in the presence of exogenous IGF1 in serum-restricted media compared to control and that cells treated with an anti-IGF1R antibody will have reduced cellular viability, similar to cells treated with serum-restricted medias.

In this study, we also found that serum restriction resulted in increased cellular apoptosis and dysregulation of cell cycle maintenance patterns. This agrees with the work of

Lee, et al. which found that 4T1 breast cancer cells cultured under starvation conditions (1%

FBS, 0.5g/L glucose) had increased markers of DNA damage and cellular apoptosis (34).

Similar results were found in vivo by Saleh, et al. in a model of triple-negative breast cancer which found that mice on a CR diet regimen had increased markers of apoptosis in tumor tissue, which was linked to downregulation of the IGF1 signaling pathway (35). These findings are consistent with previous in vivo studies suggesting dietary restriction achieved through chronic CR, intermittent fasting, or protein restriction rely in part on downregulation of IGF1 signaling to reduce cancer cell growth and proliferation.

Further, in this study we found that alterations in cell viability and growth can be attributed to differences in cell cycle regulation impacted by nutrient stress. In cells cultured in serum-restricted medias, there was an accumulation of cells in the G0/G1 phase compared to control. Serum restriction may result in cell cycle arrest through induction of a stress response prompted by the lack of external growth factors that halts cellular proliferation. Serum restriction may also induce stress-induced senescence, a non- proliferative, viable state. However, additional studies will be needed to establish a role for cellular senescence, including examination of morphological changes, activation of tumor suppressor pathways, and increased lysosomal B-D-galactosidase activity.

In conclusion, our in vitro findings indicate that the sensitivity of HER2-positive breast cancer cells to dietary restriction may occur through reduction in the components present in serum, particularly IGF1. Further examination of pharmacological interventions that target metabolic pathways involved in growth factor signaling may provide a promising method of disrupting breast cancer metabolism and rendering cells sensitive to anti-cancer therapy.

CHAPTER 3: METABOLIC REPROGRAMMING INTERVENTIONS IMPROVE RESPONSE TO LAPATINIB AND TRASTUZUMAB IN VITRO

Background While dietary interventions in the form of fasting and fasting-mimicking regimens have demonstrated safety in clinical trials, there remains uncertainty regarding the feasibility of achieving compliance due to the potential discomfort and weight loss that accompanies dietary restriction. Therefore, a group of compounds that tolerably provides similar metabolic benefits without the challenges of strict dietary restriction and without causing potentially harmful adverse effects demonstrates an appealing clinical option. Combinations of pharmacological agents that target specific metabolic pathways that are altered in cancer cells , and that are reversed by dietary restriction, may provide a strategy to improve the sensitivity of cancer cells to therapies and overcome intrinsic and acquired drug resistance.

Metabolic Reprogramming in Breast Cancer

Cancer cells experience unique and dynamic shifts in their metabolic function in order to survive, proliferate, and evade growth inhibition in a resource-scarce environment.

These metabolic changes can vary by, and are highly dependent on, the driver genetic mutations underlying specific cancer subtypes. However, most cancer cells increase their glucose uptake, lipid synthesis, and glutaminolysis and switch glucose metabolism away from oxidative phosphorylation towards glycolysis when compared with normal cells. This allows for cancer cells to increase the biosynthesis of essential building blocks and manipulate substrates available in surrounding environment in order to increase tumor vascularization and survive (36). Breast cancers specifically reprogram their metabolic function to increase uptake and utilization of glucose, glutamine, acetate, and folate to

stimulate biosynthesis. While global patterns of metabolic aberrations are similar across different breast cancer subtypes, recent metabolomic analysis of primary breast tumors have demonstrated that each subtype preferentially drives metabolic fluxes through distinct pathways, which creates unique metabolic profiles (37). Of these distinctions, mechanisms that drive increased reliance on glucose and glutamine metabolism in HER2-overexpressing breast cancers are of particular interest.

Breast cancer cells rely on high levels of flux through glycolysis accompanied by lactate secretion under aerobic conditions, known as the Warburg effect. Partially catabolizing glucose allows for cells to produce sufficient ATP while also reserving carbon intermediates that are shunted into anabolic pathways in order to sustain growth (38). This is achieved by increasing expression of the glucose transporters GLUT1 and GLUT4 to promote glucose uptake as well as overexpressing the glycolytic enzyme hexokinase 2 (39).

Further, the SCF-Skp2 ubiquitin ligase increases activation of AKT which subsequently promotes activity of hexokinase and phosphofructokinase (40). In HER2-overexpressing breast cancer cells, metabolomic analyses have confirmed this reliance on glucose metabolism by the presence of high levels of fructose-1,6-bisphosphate, glucose-6- phosphate, and pentose phosphate metabolites accumulated within cells (41). Signaling through the PI3K/AKT axis in breast cancer cells also promotes overexpression of monocarboxylate transporter 4 (MCT4), which is especially high in HER2-positive cell lines.

MCT4 transports lactate across the plasma membrane, to maintain pH balance and regenerate cytosolic NAD to support proliferative activities and support survival in a hypoxic environment (42).

In addition to high levels of glucose metabolism, HER2-positive breast cancers are also glutamine addicted. Compared to breast cancers of the luminal type, HER2- overxpressing and triple negative breast cancers have the highest expression of proteins involved in glutamine metabolism (41). Breast cancer cells are able to increase utilization of glutamine through overexpression of the transcription factor Myc which promotes

transcription of glutamine transporters and glutaminase (GLS), which converts glutamate to glutamine. Mitochondrial glutaminolysis provides carbon substrates for entry into the tricarboxylic acid cycle (TCA) and nitrogen for amino acid synthesis (43). By recognizing the aberrant metabolic activities that breast cancer cells rely on to grow uncontrollably and resist apoptosis, therapeutic strategies can be developed with the goal of normalizing cellular metabolism and sensitizing cells to therapeutic interventions.

Figure 7. Metabolic reprogramming interventions

Perturbing Glutamine Metabolism

Given the dependency of HER2-positive breast cancer cells on glutamine metabolism for sustained growth, inhibitors of glutaminase and glutamine transporters have been developed.

Bis-2-(5-pheylacetamido-1,2,4-thiadiazol-2-yl)ethyl (BPTES) is a small molecule glutaminase inhibitor that disrupts mitochondrial glutaminolysis (44). Preclinical evaluation of BPTES in the luminal-like MCF7 and basal-like MDA-MB231 human breast cancer cell lines has revealed widespread metabolic alterations with its administration in vitro. Analysis of changes in metabolites demonstrated that BPTES alters flux through glycolysis, TCA cycle, and nucleotide biosynthesis pathways, which are modulated in hypoxic conditions (45). Another method of reducing glutamine utilization can be achieved by blocking mitochondrial uptake of glutamine. L-Glutamic acid γ-(p-nitroanilide) (GPNA) inhibits the glutamine transporter ASCT2 and through off-target effects, competes with system L transporters to decrease uptake of essential amino acids, which in turn reduces the activity of mTORC1 (46).

Perturbing Growth Factor Signaling

IGF1 is a nutrient-sensitive endocrine hormone that is primarily secreted by the liver. Upon binding to its receptor (IGF1R), phosphorylation events lead to the activation of two signaling axes - MAPK and PI3K/AKT. This promotes increased cell proliferation and evasion of cell death (47, 48). Decreased levels of IGF1 in the fasted state are postulated to stimulate the potent anticancer effects of dietary restriction.

In HER2-overexpressing breast cancers, one of the potential mechanisms of resistance to trastuzumab occurs through upregulation of IGF1R and subsequent cross- phosphorylation and activation of HER2 (49). Therefore, blocking the activity of this receptor provides a useful strategy to improve drug response. BMS-754807 is a reversible inhibitor of insulin-like growth factor receptor 1 (IGF1R) and insulin receptor

(IR). It has demonstrated effectiveness in decreasing signaling through MAPK and

PI3K/AKT by competing with ATP for binding to the catalytic domain of IGF1R. BMS-

754807 may mitigate the potential for drug resistance by working in combination with

anticancer therapies to reduce cellular proliferation and induce apoptosis, as this IGF1R inhibitor has demonstrate effectiveness in vitro in combination with anti-cancer therapies for the treatments of breast, pancreatic, colon, lung, and gastric cancers (50).

Further, the combination of an mTORC1 inhibitor, specifically the FDA-approved everolimus (RAD001), with HER2-directed therapies is also a promising approach to combat drug resistance. In drug resistant HER2-positive breast tumors, trastuzumab treatment increases phosphorylation of PDK1 and mTOR, which activates ribosomal protein S6 kinase beta-1 (S6K1) and promotes anabolic activities (51). Blocking this escape pathway with everolimus, a derivative of rapamycin that inhibits activation of mTORC1 by binding to FKBP12, is one approach to improve the efficacy of trastuzumab.

Treatment with everolimus downregulates the nutrient-sensing effects of mTOR and results in reduced protein synthesis, cellular proliferation, and glucose uptake, as well as increased autophagic flux (52). In HER2-positive breast cancer patients treated with trastuzumab plus everolimus, serum metabolomic analysis revealed that this combination modulated a physiological state similar to that which occurs during fasting where lipolysis and autophagy are upregulated and gluconeogenesis and glycogenolysis are decreased (52).

Autophagy in HER2-Positive Breast Cancer

While the utilization of single-agent metabolic inhibitors in combination with anti- cancer therapies has demonstrated promising preclinical benefits, the metabolic flexibility of cancer cells may provide an additional escape pathway through the induction of autophagy.

In HER2-positive breast cancer, the precise role of autophagy in tumorigenesis and tumor progression is actively under investigation. Recent work has demonstrated that HER2- positive breast cancer cells utilize lower levels of basal autophagy compared to HER2- negative breast cancers under normal conditions, but under stressed conditions, induce autophagy to a greater extent. This is mediated in part through activation of ATG4B, a protease that cleaves pro-LC3B to form LC3-I during autophagosome formation (53). This

highlights the context-dependent manner through which this breast cancer selectively utilizes autophagy. Further, increased autophagy induction in response to treatment with the

HER2-directed therapies trastuzumab and lapatinib has been implicated as a mechanism of drug resistance. Compared to trastuzumab-sensitive SKBR3 breast cancer cells, trastuzumab-resistant JIMT-1 cells constitutively utilize higher levels of autophagy in order to sustain proliferative activities (54). Similarly, treatment of HER2-positive cells with lapatinib has also been show to increase autophagy induction, which if sustained, can allow cells to survive and develop drug-resistance (55-57). However, the role of autophagy in promoting cell survival in the presence of HER2-directed therapies remains controversial. A recent study by Vega-Rubin-de-Celis, et al. demonstrated a novel mechanism of autophagy suppression via HER2 interaction with Beclin1 (BECN1) a key autophagy protein that promotes autophagosome initiation under starvation conditions. This is significant because

HER2-positive breast cancer patients with allelic loss of BECN1 have worse clinical prognosis, implicating that suppression of autophagy through this interaction may have protumorigenic effects (58). Further, disruption of this interaction using a small molecule, Tat-

Beclin 1, in mice bearing BT-474-VH2 xenografts resulted in increased autophagy induction and reduced tumor progression as effectively as treatment with lapatinib (58). These studies highlight the context-dependent role of autophagy and the importance of conducting autophagic flux analysis with interventions that target HER2-positive breast cancer.

Perturbing cancer cell metabolism by targeting growth factor signaling and glutamine metabolism is capable of stimulating autophagy, therefore, the FDA approved autophagy inhibitor, chloroquine (CQ), is used in combination with the aforementioned agents.

Chloroquine blocks endosomal acidification and disrupts the maturation of the autophagolysosome (59). Autophagy is primarily regulated by PI3K/mTOR, therefore under low-energy conditions modulated by nutrient stress of pharmacologic inhibition of energy- responsive pathways, it is upregulated (60). We postulated that target inhibition modulated

by MRIs would stimulate autophagy, therefore combining them with chloroquine could potentially block this escape pathway.

Methods

In Vitro Dose Selection and Combination Assays using Metabolic Reprogramming Interventions and HER2-directed Therapies

For lapatinib/serum restriction combination assays, MMTV/neu cells (1x103 cells/well) were seeded into black 96-well, clear bottom plates in complete media overnight.

Cells then cultured in control (10%BCS/ 10mM Glc) media and reduced serum (5% and 1%

BCS/10mM Glc) medias with lapatinib (EC50 = 10M) or DMSO vehicle for 48 hours.

Following incubation, calcein emission was measured (see chapter 2 methods).

For metabolic reprogramming intervention assays, MMTV/neu cells (5x102 cells/well) were seeded into black 96-well, clear bottom plates in complete DMEM overnight. Dose selection curves were completed using the following metabolic reprogramming interventions

(MRI) in low-glucose DMEM (5.5mM) for 48 hours: lapatinib (Selleckchem), l-γ-glutamyl-p- nitroanilide (GPNA; Sigma-Aldrich), Bis-2-(5-phenylacetamido-1,2,4-thiadiazol-2-yl)ethyl sulfide 3 (BPTES; MedChemExpress), sodium oxamate (Sigma-Aldrich), 3-bromopyruvate

(Sigma-Aldrich), chloroquine (Sigma-Aldrich), everolimus (MedChemExpress), BMS-754807

(Selleckchem). Drug dosages that resulted in approximately 30 percent growth inhibition were selected and used for treatments in combination with lapatinib (EC30 = 5M) and chloroquine (EC30 = 2M).

BT-474 cells (5x103 cells/well) were seeded into black 96-well, clear bottom plates in complete RPMI media overnight. Dose selection curves were completed on the following compounds in low-glucose RPMI (5.5mM) for 72 hours: BPTES (MedChemExpress), chloroquine (Sigma-Aldrich), everolimus (MedChemExpress), BMS-754807 (Selleckchem).

Calcein AM Live cell staining (see Methods, Chapter 2) was used to assess the relative number of live cells compared to control. Drug dosages that resulted in ~30 percent growth

inhibition were selected and used for treatments in combination with Trastuzumab

(10ng/mL) and chloroquine (EC30 = 10M).

Western Blotting to Confirm Target Inhibition

To confirm target inhibition following treatment with trastuzumab, BT-474 cells (5x105 cells/well) were seeded into 6-well plates overnight to reach 70% confluence. Cells were gently washed with PBS and treated with complete (10% BCS/ 5.5mM Glc) or serum free media (5.5mM Glc) containing Trastuzumab (10ng/mL, Selleckchem) or vehicle (PBS) for 4 hours. Whole cell lysates were generated using RIPA lysis buffer (Sigma Aldrich) supplemented with protease inhibitor cocktail (Sigma Aldrich). Cells scraped from dish and incubated on ice for 10 minutes. Cellular debris separated via centrifugation at 14000 RPM for 15 minutes at 4C and the protein-containing supernatant was collected. Protein concentration was determined via Bradford protein quantification assay. Protein lysates denatured by boiling for 5 minutes at 95 °C in solution containing Loading Buffer (Biorad) and 5% -mercaptoethanol (BME). Protein separated via SDS-PAGE and transferred to

PVD membrane for western immunoblotting. Membranes blocked at room temperature for 1 hour in Tris-buffered saline containing 0.1% Tween (TBST) and 5% Bovine Serum Albumin.

Membranes were incubated with primary antibody overnight at 4C, washed with TBST, and blocked in secondary antibody for 1 hour at room temperature. Chemiluminescent detection utilized with SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Fisher).

Images generated with the ChemiDoc XRS+ System (Biorad). Images analyzed using

ImageJ software according to NIH gel analysis procedures. BT-474 (5x105 cells/well) and

MMTV/neu cells (2.5x105/well) were seeded into 6-well plates overnight to reach 70% confluence. Cells were gently washed with PBS and treated for 4 hours with complete or serum-free media controls and a single dose of chloroquine, BMS-754807, everolimus, or

BPTES corresponding to the EC30 for each cell line as described above (doses indicated in corresponding figures). The following primary antibodies were used: p-Erk Tyr 202/204 (CST

#9101, 1:1000), p-IGF1R/IR Tyr 1135/1136 (CST #3024, 1:1000), p-S6 Ser 235/236 (CST

#2211, 1:1000), LC3B (CST #2775, 1:1000), -Tubulin (CST #2146, 1:2000), -Actin (CST

#4970, 1:2000).

Statistical Analysis

Data were analyzed using GraphPad Prism 8 Software (GraphPad Software, Inc.).

Differences between means were assessed using one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons for statistical differences. Data graphically presented as mean ± standard error of mean (SEM) unless otherwise noted.

Results

Serum restriction decreased cellular viability in combination with HER2-directed therapies

Having determined that restricting serum availability in vitro decreases cellular viability and growth in HER2-positive breast cancer cells, it was posited that serum restriction in combination with HER2-directed therapies would enhance growth inhibition to a greater extent than the therapy alone. Since trastuzumab is a humanized monoclonal antibody, this drug is only effective against the human HER2 receptor, not the rat homolog, neu. Therefore, lapatinib was selected as the HER-2 directed therapy for the murine-derived

MMTV/neu cell line and trastuzumab with the BT-474 cell line.

MMTV/neu cells were treated with serum restricted medias plus lapatinib (10μM) for

48 hours. Compared to the lapatinib treated cells cultured in complete media (10mm/10%

BCS) that experienced ~50% growth inhibition, cells treated with lapatinib in serum restricted media had significantly increased growth inhibition. This effect was strengthened in a concentration-dependent manner, where greater reductions in the proportion of serum available in media increased the growth inhibitory effect of lapatinib (Fig. 7B p=0.0005 10% vs. 5%; p<0.0001 10% vs. 1%; p=0.0080 5% vs. 1%). However, when BT-474 cells were cultured in serum restricted medias with trastuzumab (10ng/mL) at a dose selected through

confirmation of target inhibition (Fig. 8A, B), there were no statistically significant differences in growth inhibition under serum restricted conditions relative to control (Fig. 8C).

A B l

l d o

o 100 100

n n

o c

n b

b i

50 i 50

o r

r a

0 0

% % 0.0 0.5 1.0 1.5 2.0 10 10 5 1 Lapatinib Log10[Lapatinib] (mM) - + + + (10mM) [Serum] (%)

Figure 8. Serum restriction increased growth inhibitory effect of lapatinib (A) MMTV/neu cells were treated with lapatinib (1.5-50μM) for 24 hours in complete low glucose DMEM (5.5mM Glc/10% BCS). Treated cells were normalized to vehicle control.

Data from n=3 experiments is presented as mean ± SEM (EC50 = 17.16μM, EC30 ~ 10μM). (B) MMTV/neu cells treated with vehicle or 10μM lapatinib in complete or serum-restricted media for 48 hours. Treated cells were normalized to vehicle control (10mM/10%BCS + 0.1% DMSO). Data from n=3 experiments is presented as mean ± SEM. One-way ANOVA with Tukey’s multiple comparisons. Serum restriction increased the growth inhibitory effect of lapatinib in a concentration-dependent manner (p=0.0005 10% vs. 5%; p<0.0001 10% vs. 1%; p=0.0080 5% vs. 1%).

Figure 9. Efficacy of trastuzumab was not improved by serum restriction (A) BT-474 cells treated with complete media (CON) or serum starved (SS) with or without trastuzumab (TZB; 10ng/mL) for 4 hours. Protein lysates collected and relative abundance of phosphorylated Erk (p-Erk) normalized to tubulin loading control measured via western blotting. (B) Fold change in p-Erk in each treatment group compared to control plotted. Biological replicates plotted mean ± SEM (n=3) (SS p=0.0429; CON+TZB p=0.0005; SS+TZB p=0.0016). (C) BT-474 cells treated with serum restricted medias plus 10ng/mL trastuzumab for 72 hours. Technical replicates normalized to control in complete media (10mM/10%BCS) with mean ± SEM.

Metabolic reprogramming interventions modified activation of downstream targets

In order to confirm that the metabolic reprogramming interventions selected effectively modulated anticipated protein targets at the EC30 dose in each cell line, western blotting was used to examine the abundance of downstream targets following short-term incubation with each compound. Treatment with the dual IGF1R/IR inhibitor BMS-754807 decreased phosphorylation of IGF1R in MMTV/neu (Fig.9A; p=0.0238) and in BT-474 cells

(Fig. 9B; n=1). To determine the specificity of the mTORC1 inhibitor everolimus, phosphorylation of ribosomal protein S6 (p-S6) was measured. When mTOR is active, signaling occurs through ribosomal protein S6 kinase beta-1 (S6K1), which phosphorylates

S6. In the HER2-positive cell lines treated with everolimus, phosphorylated S6 was undetectable compared to control, indicating that the selected dose of this drug effectively impairs mTORC1 downstream signaling (Fig. 9C p<0.0001; Fig. 9D p<0.0001).

It was postulated that selectively targeting growth factor-associated signaling pathways with these interventions would stimulate a cellular stress response and subsequently induce increased autophagic flux (61). LC3II abundance was measured, which is a proxy for the number of autophagosomes present in the cells under different treatment conditions. However, none of the MRI treatment conditions resulted in increased LC3II accumulation in MMTV/neu or BT-474 cells (Fig. 9E; Fig. 9F). The lack of LC3II accumulation may be due to a lack of autophagic response or, alternatively, unchanged steady-state levels of LC3II despite increased flux through the autophagic pathway.

A B MMTV/neu: BMS-754807 BT-474: BMS-754807

1.5 1.1

)

6 3

3 **

1 g

g *** 1.0

3 h

1.0 h

1 C

C 0.9

(

0.8

n i

0.5 i

F A

A 0.7

p p 0.0 0.6 N S S N S S O S M O S M C B C B

C D MMTV/neu: Everolimus BT-474: Everolimus

1.5 1.5

)

/ h

1.0 h 1.0

(

n i

0.5 i 0.5

/ /

0.0 0.0 N S s N S s O S u O S u m m C li C li ro ro e e v v E E

E MMTV/neu F BT-474 e

6 e 4 g

** g

d l

4 l

n i

i 2

b u

2 u

3 C

0 C 0

L L N S S s S Q N S S s S Q O S M u E C O S M u E C m T m T C B li P C B li P ro B ro B e e v v E E

Figure 10. Target inhibition was achieved with MRIs Modulation of protein targets in MMTV/neu and BT-474 cells treated with BMS-754807, everolimus, and BPTES for 4 hours with the EC30 dose confirmed via western blotting. Complete media (CON) and serum starvation (SS) serve as controls for all experiments. Protein abundance was normalized to loading control and represented as fold change relative to control. BMS-754807 (100nM) inhibited phosphorylation of IGF1R in (A) MMTV/neu cells (p=0.0238) and (B) BT-474 cells. Everolimus (100nM) blocked activation of downstream mTOR protein S6 ribosomal protein (p-S6) in (C) MMTV/neu cells (p<0.0001) and (D) BT-474 cells (p<0.0001). LC3II was increased under serum-starved conditions in (E) MMTV/neu cells (p=0.0058), but no significant changes were detected in (F) BT-474 cells. Representative western blots from n=1 experiment shown for (G) MMTV/neu cells and (H) BT-474 cells. Data shown from n=3 experimental replicates for MMTV/neu cell line, n= 1-3 experimental replicates for BT-474 cell line. Data presented as mean ± SEM.

Inhibition of IGF1R and mTOR increased efficacy of lapatinib

In response to the HER2-directed therapy lapatinib, HER2-positive BC cells upregulate several pathways those involved in the glucose deprivation network and the mTOR signaling axis. It was postulated that enhanced growth inhibition could be achieved with the combination of MRIs that target these pathways. Since these drugs have the potential to induce autophagy given their mechanistic targets, triple combinations of these compounds with chloroquine were also examined. Of the compounds tested, inhibition of

IGF1R/IR with BMS and mTORC1 with everolimus demonstrated significantly increased growth inhibitory effects in combination with lapatinib relative to lapatinib treatment alone

(Fig. 10A, p<0.0001; Fig. 10B, p=0.0192). However, addition of chloroquine to these regimens did not meaningfully increase growth inhibition beyond that achieved by the dual combination IGF1R and mTORC1 inhibitors plus lapatinib. The addition of the glutaminase

inhibitor, BPTES, to lapatinib treatment also did not significantly improve growth inhibition, even with the addition of chloroquine (Figure 10C).

A B

l o

o 60

60 r

r t

t c,e e n

n d

o o

C c,d c,d

n 40 40

i t

t b,c b b i b i b

b b

b i

i b b b

20 20

w o

o a r

r a

0 0

% % Lapatinib (5mM) - + - - - + + + Lapatinib (5mM) - + - - - + + + BMS-754807 (100nM) - - + - + + - + Everolimus (100nM) - - + - + + - + Chloroquine (2mM) - - - + + - + + Chloroquine (2mM) - - - + + - + +

C l

o 50 r b t b

n b o b C 40 b / b

n o

i 30

i h

n 20

h t

w 10 a

r G

0 % Lapatinib (5mM) - + - - - + + + BPTES (500nM) - - + - + + - + Chloroquine (2mM) - - - + + - + +

Figure 11. MRIs in combination with lapatinib MMTV/neu cells were treated with EC30 dose of lapatinib, chloroquine, or metabolic reprogramming interventions (BMS-754807 (BMS), everolimus, and BPTES, singly or in dual or triple combinations for 48 hours. Data from n=3 experimental replicates is presented as mean ± SEM. (A) The addition of BMS to lapatinib treatment significantly reduced cellular viability compared to lapatinib alone (p<0.0001). (B) Addition of everolimus to lapatinib treatment reduced viability compared to lapatinib alone (p=0.0192). (C) Addition of BPTES to lapatinib treatment did not meaningfully increase growth inhibition in a dual combination or in a triple combination with chloroquine compared to lapatinib alone.

Dual inhibition of mTORC1 and autophagy increased efficacy of trastuzumab

The metabolic reprogramming agents aforementioned were next combined with trastuzumab in BT-474 cells to determine if enhanced growth inhibition could be achieved with these agents. Contrary to the increased growth inhibition that occurred with dual treatment of BMS and lapatinib in MMTV/neu cells, the combination of BMS and trastuzumab did not significantly increase growth inhibition compared to either agent alone, even with the addition of chloroquine (Fig. 11A). Of the combinations explored with trastuzumab in BT-474 cells, only the addition of everolimus and chloroquine to trastuzumab treatment was more effective than trastuzumab alone (Fig. 11B p<0.0001). However, this likely occurred independently of trastuzumab as the dual combination of everolimus and chloroquine inhibited growth to a degree similar to the triple combination and was more effective than any single agent (Fig. 11B, p<0.0001 vs. trastuzumab; p<0.0001 vs. everolimus; p=0.0317 vs. chloroquine). Similar to the results in MMTV/neu cells, the addition of BPTES to trastuzumab treatment in BT-474 cells did not meaningfully increase growth inhibition, nor did the triple combination of BPTES plus chloroquine with trastuzumab (Fig.

11C).

A B l

l e o

o 60 80

r t

t e n

n b b

/ 60

b c,d n

n 40 b b d d

o i

i d

t i

i a,b

a,b

b i

i 40 h

h b,d

h t

t 20

o r

r a a

0 0

% % Trastuzumab (10ng/mL) - + - - - + + + Trastuzumab (10ng/mL) - + - - - + + + BMS-754807 (1mM) - - + - + + - + Everolimus (1nM) - - + - + + - + Chloroquine (10mM) - - - + + - + + Chloroquine (10mM) - - - + + - + +

C l

o 60 r

t b n b

o b

a,b

n 40 a,b o

i a,b

i b

i a,b

n I

w o

r a G

0 % Trastuzumab (10ng/mL) - + - - - + + + BPTES (15mM) - - + - + + - + Chloroquine (10mM) - - - + + - + +

Figure 12. MRIs in combination with trastuzumab BT-474 cells were treated with EC30 dose of trastuzumab, chloroquine, or metabolic reprogramming interventions (BMS-754807 (BMS), everolimus, and BPTES), singly or in dual or triple combinations for 72 hours. Data from n=3 experimental replicates is presented as mean ± SEM. (A) Addition of BMS-754807 (BMS) to treatment with trastuzumab did not significantly increase growth inhibition compared to trastuzumab alone, even with the addition of chloroquine. (B) Compared to trastuzumab treatment alone, addition of everolimus did not significantly increase growth inhibition (p=0.0991). The combination of everolimus and chloroquine resulted in significantly increased growth inhibition compared to any single agent (p<0.0001 v. trastuzumab; p<0.0001 v. everolimus; p=0.0317 v. chloroquine). The triple combination of trastuzumab with chloroquine plus everolimus significantly increased growth inhibition compared to trastuzumab treatment alone (p<0.0001), but was not more effective than chloroquine plus everolimus without trastuzumab (p=0.8407). (C) The addition of BPTES to treatment with trastuzumab did not meaningfully increase growth inhibition compared to trastuzumab alone (p=0.9477), nor did the triple combination of BPTES plus chloroquine with trastuzumab (p=0.6549).

Discussion Here we demonstrated that modulation of energy-responsive pathways through either serum restriction, which models caloric restriction in vitro, or metabolic reprogramming interventions, has the potential to increase the antiproliferative effect of lapatinib. Our finding that lapatinib treatment in reduced serum conditions decreases cellular viability to a greater extent than with treatment in complete media agrees with the work of Caffa et. al. which demonstrated that SKBR3 and BT474 cells cultured in starvation conditions (0.5g/L glucose,

1% serum) had increased antiproliferative effects of lapatinib compared to control (62). It is plausible that inducing a low-energy state with serum restriction decreases signaling through

PI3K/AKT, and prevents compensatory mechanisms that are activated in response to lapatinib treatment, thereby more effectively reducing cellular viability. This finding was used to inform our current in vivo model which aims to increase the efficacy of lapatinib and trastuzumab by restricting food adjuvant to therapy administration in 48-hour intermittent fasting cycles. While the preliminary data did not support that serum restriction increases the antiproliferative effect of trastuzumab in vitro, an animal model with an intact immune system is necessary to fully address this hypothesis with an intact immune system. Short-term fasting has been demonstrated to enhance tumor immunosurveillance, which is crucial for the antibody-dependent cellular cytotoxicity function of trastuzumab (63).

Further, targeting metabolic pathways implicated in HER2-directed therapeutic resistance through metabolic reprogramming interventions, specifically inhibition of

IGF1R/IR and mTOR demonstrated effectiveness in combination with lapatinib. This is in agreement with studies that have treated BT-474 and ZR-75-30 cell lines (HER2+/ER+), with a combination of trastuzumab, BMS-754807, and neratinib (a TKI of pan-HER family receptors), and demonstrated that increased cell death occurs compared to trastuzumab or neratinib alone (49). Similarly, inhibiting IGF1R with the agent AEW541 elicited potent antitumor effects and reduced tumor volume in combination with lapatinib in a mouse model

with BT-474 tumors (25). However, our approach was novel in combining inhibition of HER2 and IGF1R/IR with a transient autophagy inhibitor. We did not find that the addition of chloroquine increased the growth-inhibitory effect of BMS-754807 with either lapatinib or trastuzumab. Similarly, treatment with everolimus increased the growth-inhibitory effect of lapatinib, but this effect was not strengthened with the addition of chloroquine. In order to explore this mechanism further, it should be determined if the combination of IGF1R/IR or mTOR and HER2 inhibition increases autophagic flux. This could be done by transducing

HER2-positive cell lines with a lentivirus that permits visualization of autophagosomes and autolysomes in order to measure autophagic flux. Further, if autophagy is upregulated from baseline levels with BMS or everolimus treatment, experiments should also be conducted with higher doses of chloroquine in combination with these MRIs and HER-directed drugs to determine if a synergistic effect can be achieved with these compounds.

Another major limitation in the MRI experiments was the use of BPTES as the glutaminase inhibitor. Relatively high doses of this drug in the micromolar range were required for an effect to occur, raising the concern of potential off-target effects. Future experiments should be conducted using a glutaminase inhibitor with higher specificity such as the agent CB-839 which is already being used in clinical trials for different cancer subtypes. In triple negative breast cancers, which are also glutamine-addicted, combination of CB-839 with the chemotherapies cisplatin and etoposide reduced cellular proliferation and increased apoptosis (64). This indicates that our findings with BPTES may not capture the full potential of combining glutaminase inhibition with HER2-directed therapies.

However, this project is effective in informing future experiments that will determine if combinations of everolimus and BMS are able to re-sensitize drug-resistant HER2-positive cell lines to therapy. Lapatinib-resistant SKBR3 and BT-474 cell lines will be treated with combinations of everolimus, BMS, lapatinib, and chloroquine to determine if these agents are capable of re-sensitizing resistant cells to therapy. Changes in cellular viability will be

determined using live cell assays and clonogenicity will be measured using colony formation assays. It is postulated that these MRIs will be effective in achieving this goal as combination of everolimus with trastuzumab in HER-positive, drug-resistant advanced breast cancer improved progression free survival to 7.00 months vs. 5.78 months in trastuzumab in a phase III clinical trial (65). However, it has not yet been determined if the addition of chloroquine to this regimen has the potential to increase drug efficacy. We hypothesize that chloroquine will improve drug response because previous studies in lapatinib-resistant BT-474 cells have demonstrated that transient inhibition of autophagy with chloroquine re-sensitized cells to lapatinib by synergistically promoting apoptosis and reducing cell proliferation (57). Together, these studies highlight the potential of combining

MRIs with chloroquine in lapatinib-resistant cells to improve drug response.

Together, this study identified BMS and everolimus as efficacious metabolic reprogramming interventions that improve therapeutic response to lapatinib in HER2- positive breast cancer cells.

CHAPTER 4: ONGOING AND PLANNED STUDIES – IMPACT OF INTERMITTENT FASTING REGIMENS ON TUMOR PROGRESSION AND RESPONSE TO HER2- DIRECTED THEAPIES IN VIVO Background

Metabolic Impact of Dietary Restriction

Dietary restriction (DR), which broadly encompasses a variety of caloric restriction regimens, has been demonstrated to promote longevity and reduce tumor growth across multiple breast cancer molecular subtypes (66). For example, chronic calorie restriction

(CR), or dietary energy restriction by 20-40%, has demonstrated tumor suppressive effects in breast, colon, and pancreatic (30, 63, 67). Proposed mechanisms include enhanced apoptosis within tumors, reduced angiogenesis, and alterations in key metabolites and systemic signaling pathways involving IGF1, mTOR, and AMP (48). While its beneficial effects have been well-defined in the context of continuous energy restriction, a more translational approach may be achieved through cycles of acute dietary restriction centered around the administration of anti-cancer therapy.

Fasting protocols indicate that subjects voluntarily abstain from solid food for brief intervals of time while consuming water alone or, in the case of partial fasting regimens, consume vegetable broths and fruit juices (250-500 kcal/day) (68). The safety of completely abstaining from food for periods of 2 or more days has been demonstrated in a medically- supervised setting and indicates that the majority of healthy patients experience minimal adverse reactions (69).

Broadly, fasting is postulated to promote the fitness of model organisms by improving adaptation to cellular stress and by reducing exposure to oxidative stress which in turn improves mitochondrial health. By modulating the activities of the nutrient-sensing metabolic regulators, fasting is also implicated to expand longevity in different species (70).

During periods of fasting, serum glucose levels decrease, and hepatic glycogen stores diminish within 24 hours (66). Alternative metabolic pathways are upregulated to provide substrates to energy utilization. Gluconeogenesis is engaged to provide glucose to specific tissues, primarily the brain, and β-oxidation is upregulated to harness energy from free-fatty acids released from adipose tissue. The ketone bodies β-hydroxybutyrate and acetoacetate, released as a byproduct of β-oxidation and from the conversion of ketogenic amino acids, are utilized in the process of ketolysis (66).

In rodents, fasting and caloric restriction modulate similar metabolic targets, but elicit distinct physiological responses (71). During a 48 hour fast, rodents experience weight loss of approximately 20 percent, compared to a 4 day fast in humans which causes less than 2 percent weight loss (72). Within a 48-hour fasting period, rodents experience a decrease in blood glucose by ~50 percent (73). While blood glucose levels in humans decrease after two days in the fasted state, clinically acceptable glucose levels are maintained within this period

(74). Additionally, in humans and mice, IGF1 levels decrease by 40 and 70 percent respectively during periods of fasting ranging from 24 to 72 hours. Alternatively, IGF1 decreases by 15 to 25 percent with continuous caloric restriction (71). Another key mechanism through which dietary restriction may have its anticancer effects it’s through autophagy induction. During fasting periods of 12 to 24 hours, mice have increased induction of autophagy which may be mediated through sirtuin-1 mediated deacetylation of key autophagy proteins including ATG5, ATG7, ATG12, and LC3, as well as decreased signaling through the PI3K/mTOR axis (75-77).

Intermittent Fasting

A variety of short-term fasting protocols have been developed to augment chemotherapy cycles including intermittent fasting, periodic fasting, and fasting-mimicking diets (78). Unlike intermittent fasting, periodic fasting regimens last for 3 days or longer and are repeated every 2 or more weeks and fasting-mimicking regimens use a plant-based, calorie-restricted, low carbohydrate and low-protein diet that is indicated for use every 3 to 4 weeks (66) (67).

Intermittent fasting regimens involve cycles of short-term partial or complete caloric restriction in intervals ranging from 1 to 3 days per week. This method encompasses protocols for whole-day fasting, time-restricted feeding, and alternate-day fasting (79).

Whole day fasting indicates total caloric restriction for periods ranging for 24 to 48 hours per week, either consecutively or nonconsecutively with ad libitum feeding on remaining days

(79). Time-restricted feeding regimens define consecutive periods of ad libitum feeding that range from 3 to 12 hours per day followed by complete dietary restriction. Alternatively, intermittent fasting can be achieved through alternate day fasting or by following the 5:2 diet.

In protocols for the 5:2 diet, caloric consumption is restricted to approximately 25% of energetic needs on scheduled non-consecutive fasting days with ad libitum feeding on the remaining days of the feeding cycle (80). While these protocols vary in the indicated feeding schedule, human studies have demonstrated that adherence promotes metabolic enhancements including decreased serum concentration of triglycerides, glucose, and low- density lipoprotein cholesterol (81).

In mice, intermittent fasting regimens are modeled by completely restricting food for approximately 24 hours every 5 to 7 days (66). In an animal model, this intervention is demonstrated to be as effective as continuous caloric restriction in decreasing fasting insulin and glucose concentrations, as well as reducing total plasma cholesterol and triglycerides

(82). In mice, examination of a regimen that restricted calories by 85% on alternate fasting days found decreases in IGF1, leptin, and visceral fat, and increased levels of adiponectin

(83). Human studies have demonstrated that adherence to these protocols promotes modest weight loss and reductions in total plasma cholesterol and triglyceride concentrations (84, 85). However, in human studies conducted on individuals that were overweight or obese for periods of 8 weeks to 6 months, intermittent fasting regimens have not consistently been demonstrated to improve insulin and glucose control (80). However, one studied that compared of the metabolic impact of intermittent caloric restriction (2 days per week) with continuous caloric restriction (7 days per week) in overweight,

premenopausal women demonstrated that intermittent caloric restriction resulted in a greater reduction in fasting insulin levels and insulin resistance compared to continuous energy restriction. In both interventions, similar decreases in leptin, C-reactive protein, LDL cholesterol, and triglycerides were achieved, but, the differences in glycemic control that followed adherence to each respective regimen indicates that different mechanisms may be driving the metabolic alterations (86).

Fasting and Anti-Cancer Therapy

It is postulated that fasting and fasting-mimicking regimens target distinct metabolic pathways in cancer cells versus non-transformed cells, eliciting differential response to treatment with chemotherapy. Compared to oncogene-driven cells, non-transformed cells exposed to short-term starvation enter a chemo-protective state characterized by decreased cellular division and increased utilization of repair pathways, termed the differential stress resistance (87, 88). This has significant implications for reducing the cytotoxicity associated with chemotherapy. Conversely, malignant cells are selectively sensitized to chemotherapeutics by bouts of short-term fasting, termed differential stress-sensitization where cancer cells are preferentially destroyed by chemotherapy (89). In response to fasting and fast-mimicking diets, normal cells down regulate protumorigenic pathways (mTOR,

PKA, AKT, IGF1), which reduces growth and induces protection against the cytotoxic effects of chemotherapies. Cancer cells, which constitutively upregulate these protumorigenic pathways due to oncogene activation, are unable to decrease flux these pathways rendering them sensitive to chemotherapy.

Animal Design Protocol

MMTV/neu Pilot Study

To determine the growth trajectory of MMTV/neu breast cancer cells in a tumor transplant model, 1x106 cells/50uL were orthotopically injected into the 4th mammary fat pad of female wildtype FVB mice (n=6) or transgenic FVB/N-Tg(MMTVneu)202Mul/J mice

(Jackson Laboratory) (n=12). FVB mice received cells in a 1:1 ratio PBS and Matrigel.

Transgenic FVB/N-Tg(MMTVneu)202Mul/J mice received either cells in a solution of PBS

(n=6) or a 1:1 ratio of PBS with Matrigel (Corning). Tumor palpations were conducted biweekly for 16 weeks post-injection until mice developed tumors that were 1.5 cm in any direction.

Lentiviral Transduction

The MMTV/neu transgenic mouse spontaneously generates tumors at 6 months of age, therefore orthotopically injected tumor cells were tagged with a construct that expresses luciferase in order to distinguish the transplanted tumors from any potential spontaneously generated tumors. MMTV/neu cells were transduced with a firefly luciferase lentivirus (KeraFAST) according to manufacturer’s protocol. Transduction results in intracellular production of firefly luciferase, an enzymatic product capable of oxidizing luciferin and generating a bioluminescence reaction for detection with IVIS imaging.

Transduced cells were selected with puromycin (5g/mL) and maintained in DMEM supplemented with 10% BCS and puromycin (2g/mL).

MMTV/neu mouse model`

Seven to eight-week-old FVB/N-Tg(MMTVneu)202Mul/J mice (Jackson Laboratory) were acclimated for two weeks on a low-fat control diet (10 percent kcal/fat; Research Diets

D12450J). Mice received an orthotopic injection of the luciferase-expressing MMTV/neu breast cancer cell line (2x106 cells/50uL) into the fourth mammary fat pad. Biweekly palpations were conducted to measure tumor size and the formula V = (W(2) × L)/2 was

used to calculate approximate tumor volume (90). Upon development of tumors of appropriate size (100mm3), mice were randomized to one of four treatment groups

(n=18/group): 1) Lapatinib, 2) Vehicle, 3) Lapatinib + 48-hour fast, or 4) Vehicle + 48-hour fast. In the non-fasting groups, Lapatinib (100mg/kg) or Vehicle was administered 6 days/week for three weeks(91, 92). In the fasting groups, mice were fasted for 48-hours total, beginning 24-hours prior to initiation of treatment with Lapatinib or vehicle and ending

24-hours following initiation, with a total of 6 days of Lapatinib or Vehicle treatment/week.

Mice euthanized following 3 weeks of therapy or when a tumor endpoint of 1.5 cm in diameter was reached.

Body Composition Measurements

Body composition measurements will be taken at baseline, when MMTV/neu mice reach a tumor volume of 100mm3, and again at study endpoint via Magnetic Resonance

Imaging (MRI) conducted by the UNC Animal Metabolism Phenotyping Core.

In Vivo Bioluminescent Imaging

Once mice reach an appropriate tumor volume to initiate therapy (100mm3), In Vivo

Imaging System (IVIS) will be used to confirm the presence of the transplanted tumors in the mammary fat pad rather than spontaneously developed tumors. Ten minutes prior to imaging, mice will be anesthetized with isoflurane and the luciferin substrate (150mg/kg,

Sigma-Aldrich) will be administered. Mice imaged using IVIS Lumina Imaging System

(Perkin Elmer).

BT-474 Mouse Model

Eight-week old athymic nude (nu/nu) mice acclimated for 1 week on a low-fat control diet (Control, 10 percent kcal from fat, Research Diets D12450J). Mice were anesthetized with isoflurane for implantation of 17-b-estradiol (0.36 mg, 90-day release) pellets three days prior to tumor cell inoculation. BT-474 (ATCC HTB-20) cells derived from human ductal carcinoma were orthotopically injected into the 4th mammary fat pad (5x106 cells/50uL) in solution containing 1:1 PBS and Matrigel (Corning). Biweekly palpations were conducted to

measure tumor size and the formula V = (W(2) × L)/2 was used to calculate approximate tumor volume (90). Upon development of tumors of appropriate size (100mm3), mice were randomized to one of four treatment groups (n=21 /group): 1) Trastuzumab, 2) Vehicle, 3)

Trastuzumab + 48-hour fast, or 4) Vehicle + 48-hour fast. Tumors palpated twice weekly.

Mice were given Trastuzumab (5 mg/kg) or vehicle via intraperitoneal (IP) injection 1 day/week for three weeks(92-94). In the fasting groups, mice were fasted for 48-hours total, starting 24-hours prior to initiation of Trastuzumab or vehicle treatment and ending 24-hours after initiation. Mice euthanized following 3 weeks of therapy or when a tumor endpoint of

1.5 cm in diameter was reached.

A MMTV/neu Tumor Growth

) 3

m 2000

(

l o

V 1000

u T 0 5 10 15 Weeks Post-injection

Therapy Administration Round 1 Round 2 Round 3

Inject 2x106 Endpoint MRI MMTV-HER2/neu cells into MFP Baseline MRI Control + Vehicle (n=18)

Control + Lapatinib (n=18) 6 week-old female MMTV/neu mice Aclimation Tumor growth (100 mm3)

48h Fast + Vehicle (n=18)

48h Fast + Lapatinib (n=18)

Weeks 0 1 7 8 9 10

Figure 13. MMTV/neu mouse study pilot and design

Therapy Administration Round 1 Round 2 Round 3

Implant 17-b-estradiol Submandibular bleeds pellet

Inject 5x106 Control + Vehicle (n=21) BT-474 cells

Control + Trastuzumab (n=21) 6 week-old Tumor female growth nu/nu mice Aclimation (100 mm3)

48h Fast + Vehicle (n=21)

48h Fast + Trastuzumab (n=21)

Weeks 0 1 3 6 7 8

Figure 14. BT-474 mouse study design

Planned Analysis Upon completion of this study, several key analyses will be completed. Weights of mice will be collected throughout the fasting period in order to determine the amount of weight lost during fasting and the time required for mice to regain weight to baseline levels.

In the MMTV/neu mouse model, body composition measurements will be taken via magnetic resonance imaging prior to therapy administration and at study endpoint. Together, data on body weight maintenance patterns and body composition will allow us to determine whether fat mass and/or lean mass is modulated by fasting. In the nu/nu mouse model, circulating estrogen levels will be measured via Elisa assay from samples taken in a cohort of mice post-implantation of estradiol pellets. This will allow us to confirm that the appropriate estrogen levels were established in order to maintain growth of BT-474 tumors. Prior to each interim sacrifice and at study endpoint, blood glucose measurements will also be collected in order to compare changes in glucose levels both during a fasting period and upon refeeding.

Serum collected during the interim sacrifice and at study endpoint will be used to measure levels of insulin, leptin, major cytokines and inflammatory markers. Tumors will be excised and weighed during interim sacrifice and at study endpoint. Following collection of tumor tissue, we will utilize Mouse Clariom S Gene Expression Microarray (Affymetrix) and Gene

Set Enrichment Analysis (GSEA; Broad Institute) to analyze expression of gene sets.

Pathways involved in inflammation, tumor proliferation, apoptosis and drug resistance will be examined. This will allow us to identify common biological functions that are altered in the fasted state and in response to therapy. Within the tumor tissue, we will also confirm target inhibition of the HER2-directed drugs by extracting protein and performing western blotting to measure changes in phosphorylation of HER2 and downstream targets. Additionally, alterations in immune function will be assessed in the tumor and surrounding microenvironment. Immune infiltration in the tumor will be assessed by staining for T cell infiltrates (CD3 and CD8), and for a macrophage marker F4/80.

Anticipated Results Several rodent studies have demonstrated that short-term fasting sensitizes cancer cells to chemotherapies in mouse models of lung, breast, and pancreatic cancer (34, 63, 88).

However, this relationship has only been explored for HER2-positive breast cancer in vitro.

We hypothesize that the combination of short-term fasting cycles with HER2-directed therapy administration will improve drug response and reduce tumor growth rates more than in mice that are fed continuously. Preliminary analyses of a subset of mice in the BT-474 model after one round of therapy have indicated that intermittent fasting regimens adjuvant to trastuzumab reduce tumor growth more than in the mice fed without restriction. We predict that this decrease may occur through blockade of potential drug resistance pathways including reduced signaling through IGF1R and mTOR, similar to the mechanisms that enhanced the antiproliferative effect of the drugs in vitro in combination with MRIs. In vivo, we also predict that fasting will enhance the immune-mediated response within tumors and promote CD8+ tumor cytotoxicity. Work by Di Biase, et al. in a triple negative breast cancer model demonstrated that in mice on a fasting-mimicking diet timed around chemotherapy administration there were increased tumor infiltrating immune cells present (63). We predict that this will hold true in our model with 48-hour intermittent fasting because this regimen has been demonstrated to be safe for multiple mouse strains and more effective than 24- hour fasting in reducing chemotherapy-induced toxicities and immunosuppression (95, 96).

One important consideration with fasting mice for 48 hours is the potential adverse effects that could occur upon rapid re-feeding. Studies have demonstrated that mice lose ~20% of their body weight during a 48-hour fast and regain this weight to baseline within approximately 5 days, demonstrating that there is a significant compensatory feeding effect that occurs (34). One potential limitation of rapid refeeding is induction of the Crabtree effect, a short-term metabolic adaptation that cancer cells use to reversibly switch between oxidative and glycolytic metabolism in response to changes in their environment (97). In a mouse model of colon cancer, this mechanism was used to explain how intermittent caloric

restriction paradoxically increased cancer stemness and promoted the acquisition of a metastatic phenotype (98). While the models used in this current study are not prone to developing metastases, it is plausible that the metabolic flexibility of tumor cells could allow for increased growth in the refeeding period in the absence of anti-cancer therapy. However, since fasting combined with chemotherapy in models of breast cancer have demonstrated increased therapeutic efficacy in previous studies, we hypothesize that tumor growth rates will be reduced combining intermittent fasting with HER2-directed therapy.

CHAPTER 5: CONCLUSIONS

Study aims were designed to determine a method of translating the potent cancer- preventative effects conferred by caloric restriction into the treatment setting for HER2- positive breast cancer. It was hypothesized that modulating energy-responsive pathways implicated in drug resistance through either short-term fasting regimens or pharmacological

MRIs could mimic the beneficial effects conferred by caloric restriction and increase efficacy of HER2-directed therapies. We determined that inhibition of IGF1R/IR and mTOR activation has the potential to enhance the antiproliferative effects of lapatinib; however, further investigation is necessary to determine if this effect can be strengthened with the addition of an autophagy inhibitor. Importantly, future experiments will also determine if combinations of

MRIs can re-sensitize resistant HER2-positive breast cancer cell lines to lapatinib, which has translational potential. Ongoing work with dietary restriction will determine the impact of a

48-hour intermittent fasting regimen on HER-2 directed therapeutic efficacy in vivo, which could have important population health implications if demonstrated to be tolerable and efficacious. Altogether, we conclude that modulation of metabolic pathways implicated in

HER2-positive breast cancer drug resistance through intermittent fasting regimens and pharmacological MRIs has the potential to increase the efficacy of HER2-directed therapies.

REFERENCES

1. Society AC. Breast Cancer Facts & Figures 2017-2018. Atlanta: American Cancer Society, Inc., 2017.

2. Sevcikova K, Vertakova-Krakovska B, Spanik S. Neoadjuvant Treatment in Patients with HER2-Positive Breast Cancer. ISRN Oncol. 2013;2013:362467. Epub 2013/06/14. doi: 10.1155/2013/362467. PubMed PMID: 23762609; PMCID: PMC3676960.

3. Killelea BK, Chagpar AB, Horowitz NR, Lannin DR. Characteristics and treatment of human epidermal growth factor receptor 2 positive breast cancer: 43,485 cases from the National Cancer Database treated in 2010 and 2011. American journal of surgery. 2017;213(2):426- 32. Epub 2016/10/23. doi: 10.1016/j.amjsurg.2016.05.018. PubMed PMID: 27769548.

4. Meisel JL, Venur VA, Gnant M, Carey L. Evolution of Targeted Therapy in Breast Cancer: Where Precision Medicine Began. American Society of Clinical Oncology educational book American Society of Clinical Oncology Annual Meeting. 2018(38):78-86. Epub 2018/09/21. doi: 10.1200/edbk_201037. PubMed PMID: 30231395.

5. Dittrich A, Gautrey H, Browell D, Tyson-Capper A. The HER2 Signaling Network in Breast Cancer--Like a Spider in its Web. Journal of mammary gland biology and neoplasia. 2014;19(3-4):253-70. Epub 2014/12/30. doi: 10.1007/s10911-014-9329-5. PubMed PMID: 25544707.

6. Paplomata E, O'Regan R. The PI3K/AKT/mTOR pathway in breast cancer: targets, trials and biomarkers. Ther Adv Med Oncol. 2014;6(4):154-66. Epub 2014/07/25. doi: 10.1177/1758834014530023. PubMed PMID: 25057302; PMCID: PMC4107712.

7. Roskoski R, Jr. The ErbB/HER family of protein-tyrosine kinases and cancer. Pharmacological research. 2014;79:34-74. Epub 2013/11/26. doi: 10.1016/j.phrs.2013.11.002. PubMed PMID: 24269963.

8. Fresno Vara JA, Casado E, de Castro J, Cejas P, Belda-Iniesta C, Gonzalez-Baron M. PI3K/AKT signalling pathway and cancer. Cancer treatment reviews. 2004;30(2):193-204. Epub 2004/03/17. doi: 10.1016/j.ctrv.2003.07.007. PubMed PMID: 15023437.

9. Hartman Z, Zhao H, Agazie YM. HER2 stabilizes HER1 and itself by altering autophosphorylation patterns in a manner that overcomes regulatory mechanisms and promotes proliferative and transformation signaling. Oncogene. 2013;32(35):4169-80. Epub 2012/10/03. doi: 10.1038/onc.2012.418. PubMed PMID: 23027125; PMCID: PMC3538112.

10. Ahmed S, Sami A, Xiang J. HER2-directed therapy: current treatment options for HER2- positive breast cancer. Breast cancer (Tokyo, Japan). 2015;22(2):101-16. Epub 2015/01/31. doi: 10.1007/s12282-015-0587-x. PubMed PMID: 25634227.

11. Carter P, Presta L, Gorman CM, Ridgway JB, Henner D, Wong WL, Rowland AM, Kotts C, Carver ME, Shepard HM. Humanization of an anti-p185HER2 antibody for human cancer therapy. Proceedings of the National Academy of Sciences of the United States of America. 1992;89(10):4285-9. Epub 1992/05/15. PubMed PMID: 1350088; PMCID: PMC49066.

12. Wuerstlein R, Harbeck N. Neoadjuvant Therapy for HER2-positive Breast Cancer. Reviews on recent clinical trials. 2017;12(2):81-92. Epub 2017/02/07. doi: 10.2174/1574887112666170202165049. PubMed PMID: 28164759.

13. Loibl S, Gianni L. HER2-positive breast cancer. Lancet (London, England). 2017;389(10087):2415-29. Epub 2016/12/13. doi: 10.1016/s0140-6736(16)32417-5. PubMed PMID: 27939064.

14. Xia W, Mullin RJ, Keith BR, Liu LH, Ma H, Rusnak DW, Owens G, Alligood KJ, Spector NL. Anti-tumor activity of GW572016: a dual tyrosine kinase inhibitor blocks EGF activation of HER1/erbB2 and downstream Erk1/2 and AKT pathways. Oncogene. 2002;21(41):6255-63. Epub 2002/09/06. doi: 10.1038/sj.onc.1205794. PubMed PMID: 12214266.

15. Moasser MM, Krop IE. The Evolving Landscape of HER2 Targeting in Breast Cancer. JAMA oncology. 2015;1(8):1154-61. Epub 2015/07/24. doi: 10.1001/jamaoncol.2015.2286. PubMed PMID: 26204261.

16. Kostyal D, Welt RS, Danko J, Shay T, Lanning C, Horton K, Welt S. Trastuzumab and lapatinib modulation of HER2 tyrosine/threonine phosphorylation and cell signaling. Medical oncology (Northwood, London, England). 2012;29(3):1486-94. Epub 2011/07/20. doi: 10.1007/s12032-011-0025-7. PubMed PMID: 21769502.

17. Vu T, Claret FX. Trastuzumab: updated mechanisms of action and resistance in breast cancer. Frontiers in oncology. 2012;2:62. Epub 2012/06/22. doi: 10.3389/fonc.2012.00062. PubMed PMID: 22720269; PMCID: PMC3376449.

18. Jhaveri K, Esteva FJ. Pertuzumab in the treatment of HER2+ breast cancer. J Natl Compr Canc Netw. 2014;12(4):591-8. Epub 2014/04/11. PubMed PMID: 24717573.

19. Perez EA, Barrios C, Eiermann W, Toi M, Im YH, Conte P, Martin M, Pienkowski T, Pivot X, Burris H, 3rd, Petersen JA, Stanzel S, Strasak A, Patre M, Ellis P. Trastuzumab Emtansine With or Without Pertuzumab Versus Trastuzumab Plus Taxane for Human Epidermal Growth Factor Receptor 2-Positive, Advanced Breast Cancer: Primary Results From the Phase III MARIANNE Study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2017;35(2):141-8. Epub 2017/01/06. doi: 10.1200/jco.2016.67.4887. PubMed PMID: 28056202; PMCID: PMC5455677.

20. Mendes D, Alves C, Afonso N, Cardoso F, Passos-Coelho JL, Costa L, Andrade S, Batel- Marques F. The benefit of HER2-targeted therapies on overall survival of patients with metastatic HER2-positive breast cancer--a systematic review. Breast cancer research : BCR. 2015;17:140. Epub 2015/11/19. doi: 10.1186/s13058-015-0648-2. PubMed PMID: 26578067; PMCID: PMC4650834.

21. D'Amato V, Raimondo L, Formisano L, Giuliano M, De Placido S, Rosa R, Bianco R. Mechanisms of lapatinib resistance in HER2-driven breast cancer. Cancer treatment reviews. 2015;41(10):877-83. Epub 2015/08/16. doi: 10.1016/j.ctrv.2015.08.001. PubMed PMID: 26276735.

22. Pohlmann PR, Mayer IA, Mernaugh R. Resistance to Trastuzumab in Breast Cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2009;15(24):7479-91. Epub 2009/12/17. doi: 10.1158/1078-0432.CCR-09-0636. PubMed PMID: 20008848; PMCID: PMC3471537.

23. Hirotsu Y, Nakagomi H, Amemiya K, Oyama T, Inoue M, Mochizuki H, Omata M. Intrinsic HER2 V777L mutation mediates resistance to trastuzumab in a breast cancer patient. Medical oncology (Northwood, London, England). 2017;34(1):3. Epub 2016/12/03. doi: 10.1007/s12032-016-0857-2. PubMed PMID: 27900589.

24. Nahta R, Takahashi T, Ueno NT, Hung MC, Esteva FJ. P27(kip1) down-regulation is associated with trastuzumab resistance in breast cancer cells. Cancer Res. 2004;64(11):3981-6. Epub 2004/06/03. doi: 10.1158/0008-5472.CAN-03-3900. PubMed PMID: 15173011.

25. Alexander PB, Chen R, Gong C, Yuan L, Jasper JS, Ding Y, Markowitz GJ, Yang P, Xu X, McDonnell DP, Song E, Wang XF. Distinct Receptor Tyrosine Kinase Subsets Mediate Anti- HER2 Drug Resistance in Breast Cancer. The Journal of biological chemistry. 2017;292(2):748-59. Epub 2016/12/03. doi: 10.1074/jbc.M116.754960. PubMed PMID: 27903634; PMCID: PMC5241747.

26. Liu L, Greger J, Shi H, Liu Y, Greshock J, Annan R, Halsey W, Sathe GM, Martin AM, Gilmer TM. Novel mechanism of lapatinib resistance in HER2-positive breast tumor cells: activation of AXL. Cancer Res. 2009;69(17):6871-8. Epub 2009/08/13. doi: 10.1158/0008- 5472.CAN-08-4490. PubMed PMID: 19671800.

27. Ruprecht B, Zaal EA, Zecha J, Wu W, Berkers CR, Kuster B, Lemeer S. Lapatinib Resistance in Breast Cancer Cells Is Accompanied by Phosphorylation-Mediated Reprogramming of Glycolysis. Cancer Res. 2017;77(8):1842-53. Epub 2017/02/18. doi: 10.1158/0008-5472.Can-16-2976. PubMed PMID: 28209619.

28. Deblois G, Smith HW, Tam IS, Gravel SP, Caron M, Savage P, Labbe DP, Begin LR, Tremblay ML, Park M, Bourque G, St-Pierre J, Muller WJ, Giguere V. ERRalpha mediates metabolic adaptations driving lapatinib resistance in breast cancer. Nature communications. 2016;7:12156. Epub 2016/07/13. doi: 10.1038/ncomms12156. PubMed PMID: 27402251; PMCID: PMC4945959.

29. Allott EH, Hursting SD. Obesity and cancer: mechanistic insights from transdisciplinary studies. Endocrine-related cancer. 2015;22(6):R365-86. Epub 2015/09/17. doi: 10.1530/erc- 15-0400. PubMed PMID: 26373570; PMCID: PMC4631382.

30. Rossi EL, Dunlap SM, Bowers LW, Khatib SA, Doerstling SS, Smith LA, Ford NA, Holley D, Brown PH, Estecio MR, Kusewitt DF, deGraffenried LA, Bultman SJ, Hursting SD. Energy Balance Modulation Impacts Epigenetic Reprogramming, ERalpha and ERbeta Expression, and Mammary Tumor Development in MMTV-neu Transgenic Mice. Cancer Res. 2017;77(9):2500-11. Epub 2017/04/05. doi: 10.1158/0008-5472.Can-16-2795. PubMed PMID: 28373182.

31. Harvey AE, Lashinger LM, Hays D, Harrison LM, Lewis K, Fischer SM, Hursting SD. Calorie restriction decreases murine and human pancreatic tumor cell growth, nuclear factor- kappaB activation, and inflammation-related gene expression in an insulin-like growth factor- 1-dependent manner. PloS one. 2014;9(5):e94151. Epub 2014/05/09. doi: 10.1371/journal.pone.0094151. PubMed PMID: 24804677; PMCID: PMC4013119.

32. Lamming DW, Cummings NE, Rastelli AL, Gao F, Cava E, Bertozzi B, Spelta F, Pili R, Fontana L. Restriction of dietary protein decreases mTORC1 in tumors and somatic tissues of a tumor-bearing mouse xenograft model. Oncotarget. 2015;6(31):31233-40. Epub 2015/09/18. doi: 10.18632/oncotarget.5180. PubMed PMID: 26378060; PMCID: PMC4741600.

33. Fontana L, Weiss EP, Villareal DT, Klein S, Holloszy JO. Long-term effects of calorie or protein restriction on serum IGF1 and IGFBP-3 concentration in humans. Aging Cell. 2008;7(5):681-7. Epub 2008/10/10. PubMed PMID: 18843793; PMCID: PMC2673798.

34. Lee C, Raffaghello L, Brandhorst S, Safdie FM, Bianchi G, Martin-Montalvo A, Pistoia V, Wei M, Hwang S, Merlino A, Emionite L, de Cabo R, Longo VD. Fasting cycles retard growth of tumors and sensitize a range of cancer cell types to chemotherapy. Sci Transl Med. 2012;4(124):124ra27. Epub 2012/02/11. doi: 10.1126/scitranslmed.3003293. PubMed PMID: 22323820; PMCID: PMC3608686.

35. Saleh AD, Simone BA, Palazzo J, Savage JE, Sano Y, Dan T, Jin L, Champ CE, Zhao S, Lim M, Sotgia F, Camphausen K, Pestell RG, Mitchell JB, Lisanti MP, Simone NL. Caloric restriction augments radiation efficacy in breast cancer. Cell cycle (Georgetown, Tex). 2013;12(12):1955-63. Epub 2013/05/28. doi: 10.4161/cc.25016. PubMed PMID: 23708519; PMCID: PMC3735710.

36. Schulze A, Harris AL. How cancer metabolism is tuned for proliferation and vulnerable to disruption. Nature. 2012;491(7424):364-73. Epub 2012/11/16. doi: 10.1038/nature11706. PubMed PMID: 23151579.

37. Dubuis S, Baenke F, Scherbichler N, Alexander LT, Schulze A, Zamboni N. Metabotypes of breast cancer cell lines revealed by non-targeted metabolomics. Metabolic engineering. 2017;43(Pt B):173-86. Epub 2017/01/01. doi: 10.1016/j.ymben.2016.12.009. PubMed PMID: 28038952.

38. Smith B, Schafer XL, Ambeskovic A, Spencer CM, Land H, Munger J. Addiction to Coupling of the Warburg Effect with Glutamine Catabolism in Cancer Cells. Cell reports. 2016;17(3):821-36. Epub 2016/10/13. doi: 10.1016/j.celrep.2016.09.045. PubMed PMID: 27732857; PMCID: PMC5108179.

39. Garrido P, Osorio FG, Moran J, Cabello E, Alonso A, Freije JM, Gonzalez C. Loss of GLUT4 induces metabolic reprogramming and impairs viability of breast cancer cells. Journal of cellular physiology. 2015;230(1):191-8. Epub 2014/06/17. doi: 10.1002/jcp.24698. PubMed PMID: 24931902.

40. Wang YP, Lei QY. Perspectives of Reprogramming Breast Cancer Metabolism. Advances in experimental medicine and biology. 2017;1026:217-32. Epub 2017/12/29. doi: 10.1007/978- 981-10-6020-5_10. PubMed PMID: 29282686.

41. Cappelletti V, Iorio E, Miodini P, Silvestri M, Dugo M, Daidone MG. Metabolic Footprints and Molecular Subtypes in Breast Cancer. Disease markers. 2017;2017:7687851. Epub 2018/02/13. doi: 10.1155/2017/7687851. PubMed PMID: 29434411; PMCID: PMC5757146.

42. Baenke F, Dubuis S, Brault C, Weigelt B, Dankworth B, Griffiths B, Jiang M, Mackay A, Saunders B, Spencer-Dene B, Ros S, Stamp G, Reis-Filho JS, Howell M, Zamboni N, Schulze A. Functional screening identifies MCT4 as a key regulator of breast cancer cell metabolism and survival. The Journal of pathology. 2015;237(2):152-65. Epub 2015/05/13. doi: 10.1002/path.4562. PubMed PMID: 25965974.

43. Li C, Zhang G, Zhao L, Ma Z, Chen H. Metabolic reprogramming in cancer cells: glycolysis, glutaminolysis, and Bcl-2 proteins as novel therapeutic targets for cancer. World journal of surgical oncology. 2016;14(1):15. Epub 2016/01/23. doi: 10.1186/s12957-016-0769-9. PubMed PMID: 26791262; PMCID: PMC4721116.

44. Huang Q, Stalnecker C, Zhang C, McDermott LA, Iyer P, O'Neill J, Reimer S, Cerione RA, Katt WP. Characterization of the interactions of potent allosteric inhibitors with glutaminase C, a key enzyme in cancer cell glutamine metabolism. The Journal of biological chemistry. 2018;293(10):3535-45. Epub 2018/01/11. doi: 10.1074/jbc.M117.810101. PubMed PMID: 29317493; PMCID: PMC5846160.

45. Nagana Gowda GA, Barding GA, Jr., Dai J, Gu H, Margineantu DH, Hockenbery DM, Raftery D. A Metabolomics Study of BPTES Altered Metabolism in Human Breast Cancer Cell Lines. Frontiers in molecular biosciences. 2018;5:49. Epub 2018/06/06. doi: 10.3389/fmolb.2018.00049. PubMed PMID: 29868609; PMCID: PMC5962734.

46. Chiu M, Sabino C, Taurino G, Bianchi MG, Andreoli R, Giuliani N, Bussolati O. GPNA inhibits the sodium-independent transport system L for neutral amino acids. Amino acids. 2017;49(8):1365-72. Epub 2017/05/19. doi: 10.1007/s00726-017-2436-z. PubMed PMID: 28516268.

47. Meynet O, Ricci JE. Caloric restriction and cancer: molecular mechanisms and clinical implications. Trends Mol Med. 2014;20(8):419-27. Epub 2014/06/12. doi: 10.1016/j.molmed.2014.05.001. PubMed PMID: 24916302.

48. O'Flanagan CH, Smith LA, McDonell SB, Hursting SD. When less may be more: calorie restriction and response to cancer therapy. BMC medicine. 2017;15(1):106. Epub 2017/05/26. doi: 10.1186/s12916-017-0873-x. PubMed PMID: 28539118; PMCID: PMC5442682.

49. Chakraborty A, Hatzis C, DiGiovanna MP. Co-targeting the HER and IGF/insulin receptor axis in breast cancer, with triple targeting with endocrine therapy for hormone-sensitive disease. Breast cancer research and treatment. 2017;163(1):37-50. Epub 2017/02/27. doi: 10.1007/s10549-017-4169-9. PubMed PMID: 28236033.

50. Carboni JM, Wittman M, Yang Z, Lee F, Greer A, Hurlburt W, Hillerman S, Cao C, Cantor GH, Dell-John J, Chen C, Discenza L, Menard K, Li A, Trainor G, Vyas D, Kramer R, Attar RM, Gottardis MM. BMS-754807, a small molecule inhibitor of insulin-like growth factor- 1R/IR. Molecular cancer therapeutics. 2009;8(12):3341-9. Epub 2009/12/10. doi: 10.1158/1535-7163.Mct-09-0499. PubMed PMID: 19996272.

51. Huynh FC, Nguyen D, Jones FE. Trastuzumab stimulation of ribosomal protein S6 kinase 1 (S6K1) predicts de novo trastuzumab resistance. Biochemical and biophysical research communications. 2017;483(1):739-44. Epub 2016/12/21. doi: 10.1016/j.bbrc.2016.12.072. PubMed PMID: 27993682.

52. Jobard E, Tredan O, Bachelot T, Vigneron AM, Ait-Oukhatar CM, Arnedos M, Rios M, Bonneterre J, Dieras V, Jimenez M, Merlin JL, Campone M, Elena-Herrmann B. Longitudinal serum metabolomics evaluation of trastuzumab and everolimus combination as pre- operative treatment for HER-2 positive breast cancer patients. Oncotarget. 2017;8(48):83570-84. Epub 2017/11/16. doi: 10.18632/oncotarget.18784. PubMed PMID: 29137365; PMCID: PMC5663537.

53. Bortnik S, Choutka C, Horlings HM, Leung S, Baker JH, Lebovitz C, Dragowska WH, Go NE, Bally MB, Minchinton AI, Gelmon KA, Gorski SM. Identification of breast cancer cell subtypes sensitive to ATG4B inhibition. Oncotarget. 2016;7(41):66970-88. Epub 2016/08/25. doi: 10.18632/oncotarget.11408. PubMed PMID: 27556700; PMCID: PMC5341851.

54. Cufi S, Vazquez-Martin A, Oliveras-Ferraros C, Corominas-Faja B, Cuyas E, Lopez-Bonet E, Martin-Castillo B, Joven J, Menendez JA. The anti-malarial chloroquine overcomes primary resistance and restores sensitivity to trastuzumab in HER2-positive breast cancer. Scientific reports. 2013;3:2469. Epub 2013/08/24. doi: 10.1038/srep02469. PubMed PMID: 23965851; PMCID: PMC3749547.

55. Tang Y, Hamed HA, Cruickshanks N, Fisher PB, Grant S, Dent P. Obatoclax and lapatinib interact to induce toxic autophagy through NOXA. Mol Pharmacol. 2012;81(4):527-40. Epub 2012/01/06. doi: 10.1124/mol.111.076851. PubMed PMID: 22219388; PMCID: PMC3310419.

56. Lozy F, Cai-McRae X, Teplova I, Price S, Reddy A, Bhanot G, Ganesan S, Vazquez A, Karantza V. ERBB2 overexpression suppresses stress-induced autophagy and renders ERBB2-induced mammary tumorigenesis independent of monoallelic Becn1 loss. Autophagy. 2014;10(4):662-76. Epub 2014/02/05. doi: 10.4161/auto.27867. PubMed PMID: 24492513; PMCID: PMC4091153.

57. Chen S, Zhu X, Qiao H, Ye M, Lai X, Yu S, Ding L, Wen A, Zhang J. Protective autophagy promotes the resistance of HER2-positive breast cancer cells to lapatinib. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine. 2016;37(2):2321-31. Epub 2015/09/16. doi: 10.1007/s13277-015-3800-9. PubMed PMID: 26369543.

58. Vega-Rubin-de-Celis S, Zou Z, Fernandez AF, Ci B, Kim M, Xiao G, Xie Y, Levine B. Increased autophagy blocks HER2-mediated breast tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America. 2018;115(16):4176-81. Epub 2018/04/04. doi: 10.1073/pnas.1717800115. PubMed PMID: 29610308; PMCID: PMC5910832.

59. White E. Deconvoluting the context-dependent role for autophagy in cancer. Nat Rev Cancer. 2012;12(6):401-10. Epub 2012/04/27. doi: 10.1038/nrc3262. PubMed PMID: 22534666; PMCID: PMC3664381.

60. Saxton RA, Sabatini DM. mTOR Signaling in Growth, Metabolism, and Disease. Cell. 2017;168(6):960-76. Epub 2017/03/12. doi: 10.1016/j.cell.2017.02.004. PubMed PMID: 28283069; PMCID: PMC5394987.

61. Li L, Meng Y, Li Z, Dai W, Xu X, Bi X, Bian J. Discovery and development of small molecule modulators targeting glutamine metabolism. European journal of medicinal chemistry. 2019;163:215-42. Epub 2018/12/07. doi: 10.1016/j.ejmech.2018.11.066. PubMed PMID: 30522056.

62. Caffa I, D'Agostino V, Damonte P, Soncini D, Cea M, Monacelli F, Odetti P, Ballestrero A, Provenzani A, Longo VD, Nencioni A. Fasting potentiates the anticancer activity of tyrosine kinase inhibitors by strengthening MAPK signaling inhibition. Oncotarget. 2015;6(14):11820- 32. Epub 2015/04/25. doi: 10.18632/oncotarget.3689. PubMed PMID: 25909220; PMCID: PMC4494907.

63. Di Biase S, Lee C, Brandhorst S, Manes B, Buono R, Cheng CW, Cacciottolo M, Martin- Montalvo A, de Cabo R, Wei M, Morgan TE, Longo VD. Fasting-Mimicking Diet Reduces HO-1 to Promote T Cell-Mediated Tumor Cytotoxicity. Cancer Cell. 2016;30(1):136-46. Epub 2016/07/15. doi: 10.1016/j.ccell.2016.06.005. PubMed PMID: 27411588; PMCID: PMC5388544.

64. Chen L, Cui H, Fang J, Deng H, Kuang P, Guo H, Wang X, Zhao L. Glutamine deprivation plus BPTES alters etoposide- and cisplatin-induced apoptosis in triple negative breast cancer cells. Oncotarget. 2016;7(34):54691-701. Epub 2016/07/16. doi: 10.18632/oncotarget.10579. PubMed PMID: 27419628; PMCID: PMC5342373.

65. Andre F, O'Regan R, Ozguroglu M, Toi M, Xu B, Jerusalem G, Masuda N, Wilks S, Arena F, Isaacs C, Yap YS, Papai Z, Lang I, Armstrong A, Lerzo G, White M, Shen K, Litton J, Chen D, Zhang Y, Ali S, Taran T, Gianni L. Everolimus for women with trastuzumab-resistant, HER2-positive, advanced breast cancer (BOLERO-3): a randomised, double-blind, placebo- controlled phase 3 trial. The Lancet Oncology. 2014;15(6):580-91. Epub 2014/04/20. doi: 10.1016/s1470-2045(14)70138-x. PubMed PMID: 24742739.

66. Longo VD, Mattson MP. Fasting: molecular mechanisms and clinical applications. Cell metabolism. 2014;19(2):181-92. Epub 2014/01/21. doi: 10.1016/j.cmet.2013.12.008. PubMed PMID: 24440038; PMCID: PMC3946160.

67. Brandhorst S, Choi IY, Wei M, Cheng CW, Sedrakyan S, Navarrete G, Dubeau L, Yap LP, Park R, Vinciguerra M, Di Biase S, Mirzaei H, Mirisola MG, Childress P, Ji L, Groshen S, Penna F, Odetti P, Perin L, Conti PS, Ikeno Y, Kennedy BK, Cohen P, Morgan TE, Dorff TB, Longo VD. A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive Performance, and Healthspan. Cell metabolism. 2015;22(1):86-99. Epub 2015/06/23. doi: 10.1016/j.cmet.2015.05.012. PubMed PMID: 26094889; PMCID: PMC4509734.

68. Wilhelmi de Toledo F, Buchinger A, Burggrabe H, Holz G, Kuhn C, Lischka E, Lischka N, Lutzner H, May W, Ritzmann-Widderich M, Stange R, Wessel A, Boschmann M, Peper E, Michalsen A, Medical Association for F, Nutrition. Fasting therapy - an expert panel update of the 2002 consensus guidelines. Forschende Komplementarmedizin (2006). 2013;20(6):434-43. Epub 2014/01/18. doi: 10.1159/000357602. PubMed PMID: 24434758.

69. Finnell JS, Saul BC, Goldhamer AC, Myers TR. Is fasting safe? A chart review of adverse events during medically supervised, water-only fasting. BMC Complement Altern Med. 2018;18(1):67. Epub 2018/02/21. doi: 10.1186/s12906-018-2136-6. PubMed PMID: 29458369; PMCID: PMC5819235.

70. Michalsen A, Li C. Fasting therapy for treating and preventing disease - current state of evidence. Forschende Komplementarmedizin (2006). 2013;20(6):444-53. Epub 2014/01/18. doi: 10.1159/000357765. PubMed PMID: 24434759.

71. Lee C, Longo VD. Fasting vs dietary restriction in cellular protection and cancer treatment: from model organisms to patients. Oncogene. 2011;30(30):3305-16. Epub 2011/04/26. doi: 10.1038/onc.2011.91. PubMed PMID: 21516129.

72. Pietrocola F, Demont Y, Castoldi F, Enot D, Durand S, Semeraro M, Baracco EE, Pol J, Bravo-San Pedro JM, Bordenave C, Levesque S, Humeau J, Chery A, Metivier D, Madeo F, Maiuri MC, Kroemer G. Metabolic effects of fasting on human and mouse blood in vivo. Autophagy. 2017;13(3):567-78. Epub 2017/01/07. doi: 10.1080/15548627.2016.1271513. PubMed PMID: 28059587; PMCID: PMC5361613.

73. Jensen TL, Kiersgaard MK, Sorensen DB, Mikkelsen LF. Fasting of mice: a review. Lab Anim. 2013;47(4):225-40. Epub 2013/09/13. doi: 10.1177/0023677213501659. PubMed PMID: 24025567.

74. Lieberman HR, Caruso CM, Niro PJ, Adam GE, Kellogg MD, Nindl BC, Kramer FM. A double-blind, placebo-controlled test of 2 d of calorie deprivation: effects on cognition, activity, sleep, and interstitial glucose concentrations. Am J Clin Nutr. 2008;88(3):667-76. Epub 2008/09/10. doi: 10.1093/ajcn/88.3.667. PubMed PMID: 18779282.

75. Kroemer G, Marino G, Levine B. Autophagy and the integrated stress response. Mol Cell. 2010;40(2):280-93. Epub 2010/10/23. doi: 10.1016/j.molcel.2010.09.023. PubMed PMID: 20965422; PMCID: PMC3127250.

76. Marino G, Pietrocola F, Madeo F, Kroemer G. Caloric restriction mimetics: natural/physiological pharmacological autophagy inducers. Autophagy. 2014;10(11):1879- 82. Epub 2014/12/09. doi: 10.4161/auto.36413. PubMed PMID: 25484097; PMCID: PMC4502795.

77. Lee IH, Finkel T. Regulation of autophagy by the p300 acetyltransferase. The Journal of biological chemistry. 2009;284(10):6322-8. Epub 2009/01/07. doi: 10.1074/jbc.M807135200. PubMed PMID: 19124466; PMCID: PMC5405322.

78. Brandhorst S, Longo VD. Fasting and Caloric Restriction in Cancer Prevention and Treatment. Recent results in cancer research Fortschritte der Krebsforschung Progres dans les recherches sur le cancer. 2016;207:241-66. Epub 2016/08/26. doi: 10.1007/978-3-319- 42118-6_12. PubMed PMID: 27557543.

79. Tinsley GM, La Bounty PM. Effects of intermittent fasting on body composition and clinical health markers in humans. Nutrition reviews. 2015;73(10):661-74. Epub 2015/09/17. doi: 10.1093/nutrit/nuv041. PubMed PMID: 26374764.

80. Patterson RE, Laughlin GA, LaCroix AZ, Hartman SJ, Natarajan L, Senger CM, Martinez ME, Villasenor A, Sears DD, Marinac CR, Gallo LC. Intermittent Fasting and Human Metabolic Health. J Acad Nutr Diet. 2015;115(8):1203-12. Epub 2015/04/11. doi: 10.1016/j.jand.2015.02.018. PubMed PMID: 25857868; PMCID: PMC4516560.

81. Rothschild J, Hoddy KK, Jambazian P, Varady KA. Time-restricted feeding and risk of metabolic disease: a review of human and animal studies. Nutrition reviews. 2014;72(5):308-18. Epub 2014/04/18. doi: 10.1111/nure.12104. PubMed PMID: 24739093.

82. Varady KA, Roohk DJ, Loe YC, McEvoy-Hein BK, Hellerstein MK. Effects of modified alternate-day fasting regimens on adipocyte size, triglyceride metabolism, and plasma adiponectin levels in mice. J Lipid Res. 2007;48(10):2212-9. Epub 2007/07/04. doi: 10.1194/jlr.M700223-JLR200. PubMed PMID: 17607017.

83. Varady KA, Hudak CS, Hellerstein MK. Modified alternate-day fasting and cardioprotection: relation to adipose tissue dynamics and dietary fat intake. Metabolism. 2009;58(6):803-11. Epub 2009/04/21. doi: 10.1016/j.metabol.2009.01.018. PubMed PMID: 19375762.

84. Klempel MC, Kroeger CM, Varady KA. Alternate day fasting (ADF) with a high-fat diet produces similar weight loss and cardio-protection as ADF with a low-fat diet. Metabolism. 2013;62(1):137-43. Epub 2012/08/15. doi: 10.1016/j.metabol.2012.07.002. PubMed PMID: 22889512.

85. Johnson JB, Summer W, Cutler RG, Martin B, Hyun DH, Dixit VD, Pearson M, Nassar M, Telljohann R, Maudsley S, Carlson O, John S, Laub DR, Mattson MP. Alternate day calorie restriction improves clinical findings and reduces markers of oxidative stress and inflammation in overweight adults with moderate asthma. Free Radic Biol Med. 2007;42(5):665-74. Epub 2007/02/13. doi: 10.1016/j.freeradbiomed.2006.12.005. PubMed PMID: 17291990; PMCID: PMC1859864.

86. Harvie MN, Pegington M, Mattson MP, Frystyk J, Dillon B, Evans G, Cuzick J, Jebb SA, Martin B, Cutler RG, Son TG, Maudsley S, Carlson OD, Egan JM, Flyvbjerg A, Howell A. The effects of intermittent or continuous energy restriction on weight loss and metabolic disease risk markers: a randomized trial in young overweight women. International journal of obesity (2005). 2011;35(5):714-27. Epub 2010/10/06. doi: 10.1038/ijo.2010.171. PubMed PMID: 20921964; PMCID: PMC3017674.

87. Lee C, Safdie FM, Raffaghello L, Wei M, Madia F, Parrella E, Hwang D, Cohen P, Bianchi G, Longo VD. Reduced levels of IGF-I mediate differential protection of normal and cancer cells in response to fasting and improve chemotherapeutic index. Cancer Res. 2010;70(4):1564-72. Epub 2010/02/11. doi: 10.1158/0008-5472.CAN-09-3228. PubMed PMID: 20145127; PMCID: PMC2836202.

88. Raffaghello L, Lee C, Safdie FM, Wei M, Madia F, Bianchi G, Longo VD. Starvation- dependent differential stress resistance protects normal but not cancer cells against high- dose chemotherapy. Proceedings of the National Academy of Sciences of the United States of America. 2008;105(24):8215-20. Epub 2008/04/02. doi: 10.1073/pnas.0708100105. PubMed PMID: 18378900; PMCID: PMC2448817.

89. Buono R, Longo VD. Starvation, Stress Resistance, and Cancer. Trends in endocrinology and metabolism: TEM. 2018. Epub 2018/02/22. doi: 10.1016/j.tem.2018.01.008. PubMed PMID: 29463451.

90. Faustino-Rocha A, Oliveira PA, Pinho-Oliveira J, Teixeira-Guedes C, Soares-Maia R, da Costa RG, Colaco B, Pires MJ, Colaco J, Ferreira R, Ginja M. Estimation of rat mammary tumor volume using caliper and ultrasonography measurements. Lab Anim (NY). 2013;42(6):217-24. Epub 2013/05/22. doi: 10.1038/laban.254. PubMed PMID: 23689461.

91. Chakrabarty A, Sanchez V, Kuba MG, Rinehart C, Arteaga CL. Feedback upregulation of HER3 (ErbB3) expression and activity attenuates antitumor effect of PI3K inhibitors. Proceedings of the National Academy of Sciences of the United States of America. 2012;109(8):2718-23. Epub 2011/03/04. doi: 10.1073/pnas.1018001108. PubMed PMID: 21368164; PMCID: PMC3286932.

92. Gee MS, Upadhyay R, Bergquist H, Alencar H, Reynolds F, Maricevich M, Weissleder R, Josephson L, Mahmood U. Human breast cancer tumor models: molecular imaging of drug susceptibility and dosing during HER2/neu-targeted therapy. Radiology. 2008;248(3):925- 35. Epub 2008/07/24. doi: 10.1148/radiol.2482071496. PubMed PMID: 18647846; PMCID: PMC2798096.

93. Canonici A, Gijsen M, Mullooly M, Bennett R, Bouguern N, Pedersen K, O'Brien NA, Roxanis I, Li JL, Bridge E, Finn R, Siamon D, McGowan P, Duffy MJ, O'Donovan N, Crown J, Kong A. Neratinib overcomes trastuzumab resistance in HER2 amplified breast cancer. Oncotarget. 2013;4(10):1592-605. Epub 2013/09/07. doi: 10.18632/oncotarget.1148. PubMed PMID: 24009064; PMCID: PMC3858548.

94. Ithimakin S, Day KC, Malik F, Zen Q, Dawsey SJ, Bersano-Begey TF, Quraishi AA, Ignatoski KW, Daignault S, Davis A, Hall CL, Palanisamy N, Heath AN, Tawakkol N, Luther TK, Clouthier SG, Chadwick WA, Day ML, Kleer CG, Thomas DG, Hayes DF, Korkaya H, Wicha MS. HER2 drives luminal breast cancer stem cells in the absence of HER2 amplification: implications for efficacy of adjuvant trastuzumab. Cancer Res. 2013;73(5):1635-46. Epub 2013/02/28. doi: 10.1158/0008-5472.CAN-12-3349. PubMed PMID: 23442322; PMCID: PMC3600586.

95. Dorff TB, Groshen S, Garcia A, Shah M, Tsao-Wei D, Pham H, Cheng CW, Brandhorst S, Cohen P, Wei M, Longo V, Quinn DI. Safety and feasibility of fasting in combination with platinum-based chemotherapy. BMC cancer. 2016;16:360. Epub 2016/06/11. doi: 10.1186/s12885-016-2370-6. PubMed PMID: 27282289; PMCID: PMC4901417.

96. Chen Y, Ling L, Su G, Han M, Fan X, Xun P, Xu G. Effect of Intermittent versus Chronic Calorie Restriction on Tumor Incidence: A Systematic Review and Meta-Analysis of Animal Studies. Scientific reports. 2016;6:33739. Epub 2016/09/23. doi: 10.1038/srep33739. PubMed PMID: 27653140; PMCID: PMC5031958.

97. Diaz-Ruiz R, Rigoulet M, Devin A. The Warburg and Crabtree effects: On the origin of cancer cell energy metabolism and of yeast glucose repression. Biochimica et biophysica acta. 2011;1807(6):568-76. Epub 2010/09/02. doi: 10.1016/j.bbabio.2010.08.010. PubMed PMID: 20804724.

98. Kusuoka O, Fujiwara-Tani R, Nakashima C, Fujii K, Ohmori H, Mori T, Kishi S, Miyagawa Y, Goto K, Kawahara I, Kuniyasu H. Intermittent calorie restriction enhances epithelial- mesenchymal transition through the alteration of energy metabolism in a mouse tumor model. International journal of oncology. 2018;52(2):413-23. Epub 2018/01/19. doi: 10.3892/ijo.2017.4229. PubMed PMID: 29345287; PMCID: PMC5741368.