1. Cancer Biology

Three-dimensional growth of breast cancer cells potentiates the anti-tumor effects of unacylated ghrelin and AZP-531

  1. CheukMan C Au
  2. John B Furness
  3. Kara Britt
  4. Sofya Oshchepkova
  5. Heta Ladumor
  6. Kai Ying Soo
  7. Brid Callaghan
  8. Celine Gerard
  9. Giorgio Inghirami
  10. Vivek Mittal
  11. Yufeng Wang
  12. Xin Yun Huang
  13. Jason A Spector
  14. Eleni Andreopoulou
  15. Paul Zumbo
  16. Doron Betel
  17. Lukas Dow
  18. Kristy A Brown  Is a corresponding author
  1. Department of Medicine, Weill Cornell Medicine, United States
  2. Centre for Cancer Research, Hudson Institute for Medical Research, Australia
  3. Department of Molecular and Translational Sciences, Monash University, Australia
  4. Department of Anatomy and Neuroscience, University of Melbourne, Australia
  5. Peter MacCallum Cancer Centre, Australia
  6. Sir Peter MacCallum Department of Oncology, University of Melbourne, Australia
  7. Weill Cornell Medicine - Qatar, Qatar
  8. Department of Pathology, Weill Cornell Medical College, United States
  9. Department of Cardiothoracic Surgery, Department of Cell and Developmental Biology, Neuberger Berman Lung Cancer Center, Weill Cornell Medicine, United States
  10. Department of Physiology and Biophysics, Weill Cornell Medical College of Cornell University, United States
  11. Department of Surgery, Weill Cornell Medicine, United States
  12. Applied Bioinformatics Core, Weill Cornell Medical College, United States
  13. Institute for Computational Biomedicine, Weill Cornell Medical College, United States
https://doi.org/10.7554/eLife.56913
Share this article
Cite this article
  1. CheukMan C Au
  2. John B Furness
  3. Kara Britt
  4. Sofya Oshchepkova
  5. Heta Ladumor
  6. Kai Ying Soo
  7. Brid Callaghan
  8. Celine Gerard
  9. Giorgio Inghirami
  10. Vivek Mittal
  11. Yufeng Wang
  12. Xin Yun Huang
  13. Jason A Spector
  14. Eleni Andreopoulou
  15. Paul Zumbo
  16. Doron Betel
  17. Lukas Dow
  18. Kristy A Brown
(2020)
Three-dimensional growth of breast cancer cells potentiates the anti-tumor effects of unacylated ghrelin and AZP-531
eLife 9:e56913.
https://doi.org/10.7554/eLife.56913
  2. Download BibTeX
  3. Download .RIS

Abstract

Breast cancer is the most common type of cancer in women and notwithstanding important therapeutic advances, remains the second leading cause of cancer-related death. Despite extensive research relating to the hormone ghrelin, responsible for the stimulation of growth hormone release and appetite, little is known of the effects of its unacylated form, especially in cancer. The present study aimed to characterize effects of unacylated ghrelin on breast cancer cells, define its mechanism of action, and explore the therapeutic potential of unacylated ghrelin or analog AZP-531. We report potent anti-tumor effects of unacylated ghrelin, dependent on cells being cultured in 3D in a biologically-relevant extracellular matrix. The mechanism of unacylated ghrelin-mediated growth inhibition involves activation of Gαi and suppression of MAPK signaling. AZP-531 also suppresses the growth of breast cancer cells in vitro and in xenografts, and may be a novel approach for the safe and effective treatment of breast cancer.

Introduction

Breast cancer is the most commonly diagnosed cancer in women and second only to lung cancer in terms of mortality (Noone et al., 2018). The majority of breast cancers occur after menopause and are hormone receptor positive, and in these women, first-line therapy usually involves endocrine therapy, for example aromatase inhibitors or tamoxifen (Reinert and Barrios, 2015). Despite the efficacy of endocrine therapy, a number of women experience severe and debilitating side effects due to the global inhibition of estrogen biosynthesis or action, and some will cease the use of their potentially life-saving treatment (Lønning and Eikesdal, 2013). A proportion of women will also be resistant to treatment or develop resistance over time, and some will have tumors that cannot be treated with targeted therapies, that is triple negative breast cancers (TNBCs). Aggressive breast cancers are often associated with activation of RAS/MAPK signaling (Giltnane and Balko, 2014; Santen et al., 2002), despite only a minority carrying a mutation in these genes (Tilch et al., 2014). The prognosis for these patients is poor and hence, there is a need to identify alternative treatments that are safe and effective.

The unacylated form of ghrelin is closely related to the appetite-stimulating hormone ghrelin. However, because it lacks octanoylation, it does not bind to the cognate ghrelin receptor, GHSR1a. Initially, unacylated ghrelin was believed to be produced as a by-product of ghrelin gene expression. However, recent studies have established an important role for this peptide hormone in regulating energy homeostasis, including reducing fat mass, improving insulin sensitivity and decreasing fasting glucose levels (Benso et al., 2012; Zhang et al., 2008). An unacylated ghrelin analog, AZP-531 (levolitide), is currently in clinical trials for the treatment of Prader-Willi Syndrome and type II diabetes (Allas et al., 2016). AZP-531 has an established safety profile and better pharmacokinetic properties than unacylated ghrelin (Allas et al., 2016). The receptor for unacylated ghrelin is currently unknown (Au et al., 2016) and consequently, there is an important gap in our understanding of the mechanisms mediating its effects.

We have previously demonstrated that unacylated ghrelin has activity in non-cancer tissue, including adipose stromal cells, where it suppresses the expression of the estrogen-biosynthetic enzyme aromatase, and breast adipose tissue macrophages, where it inhibits the production of inflammatory mediators (Au et al., 2017; Docanto et al., 2014). Here, we show that the unacylated form of ghrelin is a potent suppressor of breast cancer cell growth, independent of effects on the stroma, and provide a novel mechanism of action via activation of Gαi, suppression of cAMP production, and inhibition of MAPK and Akt signaling. Importantly, the potent effects of unacylated ghrelin are dependent on growth of cells in 3D within a relevant extracellular matrix (ECM). Our findings, and that of others, suggest that tumor cell culture context affects response to therapy and that many translational failures may have resulted from the inappropriate model systems used to date (Tian et al., 2015). It also suggests that others may have overlooked many effective therapies due to a lack of response in 2D cultures. We therefore believe that unacylated ghrelin is a prototypic 3D-specific breast cancer cell therapeutic and that characterizing its mechanism of action in a biologically relevant ECM will lead to a better understanding of how the tumor microenvironment affects response to therapy. We also demonstrate consistent effects in patient-derived breast cancer cells and breast cancer xenografts in preclinical models, where both unacylated ghrelin and AZP-531 are effective at causing growth inhibition.

Results

Unacylated ghrelin inhibits the 3D growth of breast cancer cells

Previous studies examining the effect of ghrelin and unacylated ghrelin on the growth of breast cancer cells showed little activity at doses below 1 µM (Cassoni et al., 2001). Considering the increasingly acknowledged role of the ECM in dictating the biology of tumors in vivo, including aggressiveness and response to treatment, we sought to examine the effect of ghrelin and more importantly, unacylated ghrelin, in 3D cultures of breast cancer cells. Although both are hypothesized to bind alternate ghrelin receptors, beneficial effects on energy homeostasis have only been reported for unacylated ghrelin – likely due to ghrelin stimulating insulin resistance via its cognate receptor. The dependence on 3D culture was examined for unacylated ghrelin, where treatment of MCF7 and MDA-MB-468 cells with 100 pM resulted in suppression of cell growth in matrigel and collagen, but not in 2D (Figure 1—figure supplement 1a and b). At this dose, no effects on MDA-MB-231 cell growth were observed in either 2D or 3D cultures (Figure 1—figure supplement 1c). Inhibition of MCF7 and MDA-MB-468 cell growth was observed at 10-18-10-10 M, with maximal effects observed at picomolar doses (Figure 1—figure supplement 1d and e). The degree of inhibition is directly related to the degree of growth stimulation induced by serum or estradiol. Again, no significant effect was observed in MDA-MB-231 cells (Figure 1—figure supplement 1f). The effects of unacylated ghrelin were mimicked by ghrelin, where 100 pM of either ghrelin and unacylated ghrelin were found to suppress the growth of MCF7 breast cancer cells when grown in 3D in matrigel (Figure 1—figure supplement 1g). To gain insights into the breadth of effects of unacylated and potential predictors of response, the effect of 100 pM unacylated ghrelin was tested in a panel of breast cancer cell lines grown in 3D, including ER+/PR+/HER2- (MCF7, T47D), ER+/PR+/HER2+ (ZR-75), HER2+ (SKBR3), TNBC (MDA-MB-468, DU4475, MDA-MB-157, Hs578T, MDA-MB-231), and tamoxifen-resistant (LCC2) cells (Figure 1a–c). In the presence of serum, unacylated ghrelin significantly inhibited the growth of all breast cancer cells examined except for three of the TNBC cell lines (Figure 1c; DU4475, Hs578T and MDA-MB-231). Unacylated ghrelin also suppressed the estradiol-stimulated growth of ER+ breast cancer cells (Figure 1b). Responsive cells included those having mutations in PTEN and PIK3CA, while resistant cell lines carried mutations in BRAF, HRAS and/or KRAS (Figure 1c; Table 1). To test whether these mutations confer resistance to treatment, the effects of unacylated ghrelin were examined in BRAFV600E-transfected MCF7 cells or colon cancer cells (RKO, HCT116) that carry mutations in BRAF or KRAS, respectively. Transient transfection of MCF7 cells with mutant BRAF led to resistance to unacylated ghrelin treatment in estradiol- and serum-stimulated cells (Figure 1d). Unacylated ghrelin had no effect on the growth of RKO and HCT116 cells (Figure 1e and f). However, loss of the mutant alleles for BRAF or KRAS led to sensitization of RKO and HCT116 cells, respectively. Loss of the wild-type allele had no significant effect. Binding of Cy3-labeled unacylated ghrelin to responsive and non-responsive cells (Figure 1—figure supplement 1h) suggests that resistance is not due to lack of binding or receptor expression.

Figure 1 with 1 supplement see all
Download asset Open asset
Unacylated ghrelin inhibits the 3D growth of breast cancer cells.

(a, c) Unacylated ghrelin (UAG; 100 pM) inhibits the growth of a panel of breast cancer cell lines under serum-stimulated conditions (six replicates/group) or (b) ER+ breast cancer cell lines in the presence of estradiol (10 nM; six replicates/group). (c) UAG (100 pM) suppresses cell growth of basal-like and mesenchymal-like TNBC breast cancer cell lines that are WT for BRAF and KRAS (6–9 replicates/group). Effects of UAG are abrogated in (d) BRAF-transfected MCF7 cells, and (e) BRAF- and (f) KRAS-mutated colon cancer cells (6–12 replicates/group). Loss of mutated alleles of BRAF or KRAS sensitizes cells to the effect of UAG. Data represent mean ± SEM. Experiments were repeated at least twice. UAG: unacylated ghrelin; VC: vehicle control; FBS: fetal bovine serum; E2: estradiol.

Figure 1—source data 1

Unacylated ghrelin inhibits the 3D growth of breast cancer cells.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig1-data1-v1.xlsx
Table 1
Characteristics of breast cancer cell lines and patient-derived breast cancer cells, and responsiveness to unacylated ghrelin.
Breast cancer cell line/Patient sampleKnown mutationsIntrinsic subtypeReceptor statusResponsive to
Unacylated Ghrelin
Cell line
MCF7CDKN2A, PIK3CALuminal AER+/PR+/HER2-Yes
LCC2N/ALuminal AER+/PR+/HER2-Yes
T47DPIK3CA, TP53Luminal AER+/PR+/HER2-Yes
ZR75PTENLuminal BER+/PR+/HER2+Yes
SKBR3TP53HER2+HER2+Yes
MDA-MB-468PTEN, RB1, SMAD4,
TP53		Basal-likeTNBCYes
MDA-MB-157NF1, TP53Mesenchymal-likeTNBCYes
MDA-MB-231BRAF, KRAS, TP53,
CDKN2A, NF2		Mesenchymal-likeTNBCNo
HS578THRAS, TP53Mesenchymal-likeTNBCNo
DU4475BRAF,APC, MAP2K4,
RB1		Basal-likeTNBCNo
Patient samples
ER+ Case 1N/ALuminal AER+Yes
ER+ Case 2N/ALuminal AER+Yes
2147-TG5N/ABasal-likeTNBCYes
4013-TG3N/ABasal-likeTNBCYes
3887-TG7N/AMesenchymal-likeTNBCNo
3204-TG6N/AMesenchymal-likeTNBCNo

  1. Abbreviations: ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TNBC, triple negative breast cancer; N/A: not available.

Unacylated ghrelin suppresses breast cancer cell growth via Gαi-dependent inhibition of cAMP formation

Unacylated ghrelin does not bind to the cognate ghrelin receptor GHSR1a and the receptor for unacylated ghrelin, believed to be an alternate ghrelin receptor, is currently unknown, but hypothesized to be a GPCR. Effects of unacylated ghrelin on second messenger systems were assessed by measuring the formation of cAMP and the release of intracellular calcium in MCF7 cells (Figure 2a and Figure 2—figure supplement 1a). Unacylated ghrelin significantly suppressed the formation of cAMP, but had no effect on intracellular calcium release. Effects on cAMP suggest a Gαi-dependent mechanism. Activation of Gαi by unacylated ghrelin was observed after 2 hr treatment, as measured by pull-down of GTP-bound Gαi (Figure 2b). To test dependence of growth inhibitory effects of unacylated ghrelin on Gαi, 3D growth assays were performed in cells where the Gαi-encoding gene, guanine nucleotide-binding protein, alpha subunit (GNAI), was knocked out (CRISPR) or in the presence of Gαi inhibitor pertussis toxin. There are three Gαi encoding genes in mammalian cells (GNAI1, GNAI2, and GNAI3). Unacylated ghrelin had no effect on MCF7 cells lacking Gαi subunit 2 (GNAI2), whereas the growth of cells was significantly suppressed where Gαi subunit 1 (GNAI1) and Gαi subunit 3 (GNAI3) were targeted (Figure 2c). GNAI2 was also found to be required for unacylated ghrelin activity in MDA-MB-468 cells (Figure 2—figure supplement 1b). Effects of unacylated ghrelin were also attenuated in MCF7, MDA-MB-468 and ZR-75 cells treated with pertussis toxin, in the presence of estradiol or serum (Figure 2d and e and Figure 2—figure supplement 1c). To determine whether inhibition of cAMP is sufficient to suppress the serum-stimulated growth of breast cancer cells, MCF7 and MDA-MB-468 cells were treated with adenylyl cyclase and PKA inhibitors, SQ22536 and KT5720, and cAMP antagonist, cAMPS-RP (Figure 2f; Figure 2—figure supplement 1d). Inhibition of cAMP formation and PKA led to a significant reduction in cell number. Effects of cAMP on cell growth were then examined in cells treated with adenylyl cyclase stimulator, forskolin (Figure 2g and Figure 2—figure supplement 1e). Forskolin stimulated the growth of MCF7, ZR75, MDA-MB-468 and MDA-MB-231 cells, and similar to effects of unacylated ghrelin in serum- or estradiol-stimulated conditions, unacylated ghrelin suppressed the forskolin-mediated induction of MCF7, ZR75 and MDA-MB-468 cell growth, but not MDA-MB-231. As resistance to unacylated ghrelin was observed in cells that carry mutations in RAS and RAF, the link between cAMP and MAPK signaling was next examined in MCF7, MDA-MB-468 and MDA-MB-231 cells in 3D (Figure 2h). Inhibition of MEK activity using U0126 led to a significant reduction in the forskolin-stimulated growth of all cell lines.

Figure 2 with 1 supplement see all
Download asset Open asset
Unacylated ghrelin suppresses breast cancer cell growth via Gαi-dependent inhibition of cAMP formation.

(a) Unacylated ghrelin (UAG; 10–1000 pM) inhibits the forskolin-stimulated production of cAMP in MCF7 cells (3–4 replicates/group). (b) UAG (100 pM) stimulates activation of Gαi (three replicates/group). (c) UAG (100 pM) suppresses the growth of CRISPR GNAI1 and GNAI3 KO cells, but not GNAI2 KO MCF7 cells, suggesting Gαi2-coupled GPCR-mediated effects. Suppression of (d) estradiol- or (e) serum-stimulated breast cancer cell growth with UAG (100 pM) is prevented in the presence of Gαi inhibitor, pertussis toxin (20 ng/ml, 200 ng/ml; three replicates/group). (f) PKA inhibitor (KT5720), adenylyl cyclase inhibitor (SQ22536) or cAMP antagonist (cAMPS-RP) suppress the serum-stimulated growth of MCF7 cells (three replicates/group). (g) UAG (100 pM) inhibits the forskolin-stimulated growth of MCF7 and MDA-MB-468 cells, but not MDA-MB-231 (three replicates/group). (h) U0126 (MEK inhibitor) inhibits the forskolin-stimulated growth of MCF7, MDA-MB-468 and MDA-MB-231 cells (three replicates/group). (i) A model summarizing the putative mechanism of action of UAG in breast cancer cells and compounds used to dissect mechanism of action. Data represent mean ± SEM. Experiments were repeated at least twice. UAG: unacylated ghrelin; VC: vehicle control; FBS: fetal bovine serum; E2: estradiol; PT: pertussis toxin; M: melatonin; FSK: forskolin.

Figure 2—source data 1

Unacylated ghrelin suppresses breast cancer cell growth via Gαi-dependent inhibition of cAMP formation.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig2-data1-v1.xlsx
Figure 2—source data 2

Unacylated ghrelin suppresses breast cancer cell growth via Gαi-dependent inhibition of cAMP formation.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig2-data2-v1.pdf

Unacylated ghrelin inhibits MAPK and Akt signaling

Since unacylated ghrelin suppresses cAMP, which we found to stimulate cell growth via MAPK-dependent mechanisms, we next sought to determine whether unacylated ghrelin affected MAPK signaling in 3D cultures, both acutely and chronically. Effects of unacylated ghrelin on ERK activity were examined in real-time using time-lapse confocal microscopy of the FRET-based extracellular signal-regulated kinase activity reporter (EKAR) in MCF7 cells (Figure 3a). Serum stimulated EKAR activity, whereas unacylated ghrelin and the MEK inhibitor, U0126, inhibited this effect. Effects of unacylated ghrelin on MAPK signaling were then examined. Unacylated ghrelin caused a decrease in the phosphorylation of ERK and downstream target, p90RSK (Figure 3b). The effect on p90RSK was sustained for 24 hr in MCF7 and MDA-MB-468 cells, but this suppression was not observed in MDA-MB-231 cells, at any time point (not shown). The levels of MAPK target cMyc, induced by serum, were also suppressed in cells treated with unacylated ghrelin (Figure 3c). Effects on Akt phosphorylation and activity were then examined by immunoblotting and by quantifying levels of FoxO3 nuclear localization in 3D in real time (Figure 3d and e). Unacylated ghrelin inhibited the serum-stimulated phosphorylation of Akt (Ser473) and downstream target p70S6K (Thr389). Inhibition of Akt phosphorylation was noticeable at 5 and 15 min, and sustained for 24 hr. Effects on Akt activity, measured by examining the degree of serum-stimulated FoxO3 nuclear exclusion, were observed in real time, with unacylated ghrelin and PI3K inhibitor (LY294002) causing a significant reduction in FoxO3 nuclear exclusion after 1 hr treatment. No effect of unacylated ghrelin on Akt phosphorylation or activity was observed in MDA-MB-231 cells (not shown). These data suggest that unacylated ghrelin inhibits cell proliferation via effects on MAPK and Akt signaling.

Figure 3
Download asset Open asset
Unacylated ghrelin (UAG) inhibits MAPK and Akt signaling.

(a) Unacylated ghrelin (UAG; 100 pM) inhibits ERK activity (EYFP FRET) in EKAR-transfected MCF7 cells (10 replicates/group). Data were normalized to vector ECFP signal. Scale bar represent 50 μm. Western blotting demonstrates that UAG causes a decrease in the (b) phosphorylation of ERK1/2 and downstream MAPK target p90RSK and (c) expression of cMYC in MCF7 cells. UAG also causes a decrease in (d) the phosphorylation of Akt and its downstream target, p70S6K, as well as (e) FoxO3a nuclear localization FoxO3a-RFP-transfected cells, an effect that is attenuated in cells treated with PI3K inhibitor LY294002 (five replicates/group). Data represent mean ± SEM. Experiment were repeated at least twice. UAG: unacylated ghrelin; FBS: fetal bovine serum.

Figure 3—source data 1

Unacylated ghrelin suppresses EKAR and FoxO3 nuclear localization.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig3-data1-v1.xlsx
Figure 3—source data 2

Unacylated ghrelin inhibits MAPK and Akt signaling.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig3-data2-v1.pdf

Unacylated ghrelin causes cell cycle arrest and apoptosis

Additional studies were then undertaken to determine whether effects of unacylated ghrelin on cell number were due to stimulation of cell cycle arrest or apoptosis, or both. The effects of unacylated ghrelin on cell proliferation were examined by assessing EdU incorporation in breast cancer cell lines grown in 3D. Estradiol and serum caused a significant increase in EdU incorporation, and unacylated ghrelin suppressed DNA synthesis at 10, 100 and 1000 pM in MCF7, ZR75 and tamoxifen-resistant LCC2 cells (Figure 4a–c, Figure 4—figure supplement 1a). The effect of unacylated ghrelin on phases of the cell cycle was assessed in live cells in 3D using the Premo FUCCI Cell Cycle Sensor, which consists of RFP-tagged cdt1 and GFP-tagged geminin (Figure 4d–e, and Figure 4—figure supplement 1b). In G1, geminin is degraded, leaving only the RFP signal. In S, G2 and M phases, cdt1 is degraded and hence, only the GFP signal is detected. During the G1/S transition, cells express both proteins with overlapping fluorescence appearing as yellow. Treatment of MCF7 and MDA-MB-468 cells with unacylated ghrelin was associated with an increase in cells arrested at the G1 phase of the cell cycle, with no significant effect in MDA-MB-231 cells. Unacylated ghrelin also caused apoptosis in serum-stimulated MCF7 and MDA-MB-468 cells, measured in 3D using membrane-impermeant nucleic acid dye ethidium homodimer III and using FACS analysis of Annexin V-stained cells (Figure 4f; Figure 4—figure supplement 1). Consistently, unacylated ghrelin suppressed the serum-stimulated expression of CDK4 and cyclin D3, important for cell cycle G1 phase progression, and decreased phosphorylation of Rb at Ser795, known to prevent the degradation of this cell cycle arrest protein. Treatment with unacylated ghrelin was also associated with a decreased expression of anti-apoptotic protein BCL2, while stimulating the expression of pro-apoptotic factor BAX (Figure 4g–h). This effect was not observed in MDA-MB-231 cells (not shown). These data therefore suggest that unacylated ghrelin decreases cell number by stimulating cells to arrest in G1 and stimulating apoptosis.

Figure 4 with 1 supplement see all
Download asset Open asset
Unacylated ghrelin causes cell cycle arrest and apoptosis.

Unacylated ghrelin (UAG) significantly inhibits the proliferation of (a) MCF7, (b) ZR75 and (c) MDA-MB-468 in the presence of estradiol (10 nM) or serum (3–6 replicates/group). Representative images showing EdU incorporation (green). Hoechst nuclear stain; blue. Scale bar represent 100 μm. Effects are mediated via induction of G1-phase cell cycle arrest (RFP+) and a reduction in the number of cells in S/G2/M-phase (GFP+) and G1/S transition (YFP+) in (d) MCF7 and (e) MDA-MB-468 cells (four replicates/group). (Hoechst nuclear stain; blue). Scale bar represents 100 μm. (f) UAG stimulates cell death in MCF7 and MDA-MB-468 cells, but not MDA-MB-231 cells (three replicates/group). Western blot results demonstrating that UAG inhibits CDK4/cyclin D3, pRB (Ser 795) and BCL2, and stimulates BAX in (g) MCF7 and (h) MDA-MB-468 cells. Data represent mean ± SEM. Experiments were repeated at least twice. UAG: unacylated ghrelin; VC: vehicle control; FBS: fetal bovine serum; E2: estradiol.

Figure 4—source data 1

Unacylated ghrelin causes cell cycle arrest and apoptosis.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig4-data1-v1.xlsx
Figure 4—source data 2

Unacylated ghrelin causes cell cycle arrest and apoptosis.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig4-data2-v1.pdf

Unacylated ghrelin inhibits tumor growth in xenograft models and patient-derived tumor cells

The effect of unacylated ghrelin on tumor growth in vivo was examined in orthotopic xenograft and allograft mouse models. In MCF7 and ZR75 xenografts, daily s.c. injection of 50 µg/kg and 100 µg/kg unacylated ghrelin led to a significant reduction in tumor volume (Figure 5a and b). Treatment with unacylated ghrelin had no detrimental effect on weight and no abnormalities associated with treatment were observed with histopathology assessment of a subset of mice. Effects were then assessed in a syngeneic model of mammary cancer. In the J110 allograft model, unacylated ghrelin caused a significant reduction in tumor growth at 100 µg/kg and 200 µg/kg (Figure 5c). Effects of unacylated ghrelin to suppress the growth of J110 cells was not dependent on the host immune context, as similar results were obtained when cells were xenografted in Balb/c nude immunocompromised mice (Figure 5—figure supplement 1d). The degree of apoptosis in tumors at endpoint was quantified by counting the percentage of cells with pyknotic nuclei in tumors. Unacylated ghrelin significantly increased the number of apoptotic cells in MCF7, ZR75 and J110 xenografts/allografts (Figure 5d–f). The effects of unacylated ghrelin were also examined in patient-derived breast cancer cells. Unacylated ghrelin at 100pM caused a significant reduction in the serum-induced growth of ER+ breast cancer cells, while inhibiting or having no effect in the TNBC patient samples examined (Figure 5g, Figure 5—figure supplement 1e; Table 1). BRAF and KRAS mutation status has not been characterized in these patient samples. Interestingly, Ingenuity Pathway Analysis (causal analysis) of baseline gene expression indicated that MAPK-target genes were differentially expressed between responsive and non-responsive TNBC patient-derived samples, such that non-responsive cells had gene expression consistent with activation of MAPK signaling. This gene signature may allow prediction of responders in a patient population.

Figure 5 with 2 supplements see all
Download asset Open asset
Unacylated ghrelin inhibits tumor growth in xenograft models and patient-derived tumor cells.

Tumor volume in response to treatment with 50 μg/kg (blue), 100 μg/kg (red) or 200 μg/kg (purple) UAG in mice xenografted with (a) MCF7 (six replicates/group), (b) ZR75 (five replicates/group), or allografted with (c) J110 (five replicates/group) cells. Representative tumor (below) with scale bar representing 10 mm. UAG significantly increases the number of cells with apoptotic nuclei in (d) MCF7, (e) ZR75 and (f) J110 xenografts. (g) UAG (100 pM) significantly inhibits the growth of patient-derived ER+ breast cancer cells and 4013-TG3 TNBC cells, but not 3204-TG6 TNBC cells. (h) Heatmap representing baseline differential expression of MAPK-target genes in responsive vs. non-responsive patient-derived cells. Data represent mean ± SEM. UAG: unacylated ghrelin; VC: vehicle control; FBS: fetal bovine serum.

Figure 5—source data 1

Unacylated ghrelin and cyclic analog AZP-531 inhibit tumor growth in xenograft models and patient-derived tumor cells.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig5-data1-v1.xlsx

Unacylated ghrelin analog, AZP-531, inhibits breast cancer cell growth in vitro, ex vivo and in vivo

AZP-531 is a cyclic analog of unacylated ghrelin comprised of amino acids Ser6-Gln12 of the unacylated ghrelin peptide. It is a well-tolerated drug that is more stable than unacylated ghrelin. In order to determine whether this clinically available treatment has similar effects to parent peptide unacylated ghrelin, the effect of AZP-531 on breast cancer cell growth was examined in MCF7, MDA-MB-468, and patient-derived TNBC cells, and compared to chemotherapeutic doxorubicin (Figure 6a–c). AZP-531 caused the dose-dependent inhibition of the serum-stimulated growth of cells with greater potency than doxorubicin, but unlike doxorubicin, did not reduce cell number beyond that which was stimulated by serum. In order to confirm a similar mechanism of action, the effects of unacylated ghrelin and AZP-531 on cAMP levels within MCF7 cells were compared (Figure 6d). Both unacylated ghrelin and AZP-531 caused a dose-dependent inhibition of the forskolin-mediated production of cAMP at 100 and 1000 pM, with no significant differences observed when both treatments were compared. Importantly, AZP-531 also caused the suppression of cell growth in preclinical models, including TNBC MDA-MB-468 xenografts and ER+ J110 allografts, at 200 µg/kg (Figure 6e and f, respectively, and Figure 5—figure supplement 1). These data suggest that AZP-531 may be efficacious at suppressing the growth of ER+ tumors and TNBC in women.

Figure 6
Download asset Open asset
Unacylated ghrelin analog, AZP-531, inhibits breast cancer cell growth in vitro, ex vivo and in vivo.

AZP-531 causes the dose-dependent inhibition of (a) MCF7 and (b) MDA-MB-468 and (c) patient-derived TNBC breast cancer cell growth in 3D, compared with chemotherapeutic agent doxorubicin (three replicates/group). (d) Unacylated ghrelin (UAG; 10–1000 pM) and AZP-531 (AZP; 10–1000 pM) inhibits the forskolin-stimulated production of cAMP in MCF7 cells (3–6 replicates/group). Data represent mean ± SEM. Experiments were repeated at least twice. Tumor volume in response to treatment with 200 μg/kg AZP-531 (purple) in mice xenografted with (e) MDA-MB-468 (8–9 replicates/group) or allografted with (f) J110 (five replicates/group) cells. Representative images (below) with scale bars representing 10 mm.

Figure 6—source data 1

Unacylated ghrelin analog, AZP-531, inhibits breast cancer cell growth in vitro, ex vivo and in vivo.

https://cdn.elifesciences.org/articles/56913/elife-56913-fig6-data1-v1.xlsx

Discussion

Our work provides evidence that unacylated ghrelin is a potent inhibitor of breast cancer cell growth and provides mechanistic insights not previously described for this peptide hormone. Little is known of the relationship between ghrelin, unacylated ghrelin and effects on breast cancer risk and progression. One report recently demonstrated that tumor ghrelin expression is associated with a favorable outcome in invasive breast cancer (Grönberg et al., 2012). More specifically, ghrelin immunoreactivity is significantly correlated with low histological grade, estrogen receptor positivity, small tumor size, low proliferation, as well as better recurrence-free and breast cancer-specific survival. The polyclonal antibody used was generated against full-length human ghrelin and hence potentially cross-reacts with unacylated ghrelin. In the present manuscript, both ghrelin and unacylated ghrelin were found to cause potent inhibition of breast cancer cell growth at picomolar doses when cells are grown in a biologically relevant extracellular matrix (ECM). Previous in vitro work examining the effects of ghrelin and unacylated ghrelin in breast cancer saw very little effects at sub-micromolar doses (Cassoni et al., 2001). This is likely due to prior studies having been undertaken using traditional culture methods where cells were grown in 2D on plastic and highlights important mechanistic differences when cells are cultured in 3D surrounded by an ECM. The importance of modeling breast and other cancers using 3D systems has been emphasized in a number of recent studies. Importantly, differences in signaling pathway activation are noted and 3D systems are hypothesized to allow for better prediction of in vivo effects (Gangadhara et al., 2016; Weigelt et al., 2014). Our findings, and that of others, suggest that tumor cell culture context affects response to therapy and that many translational failures may have resulted from the inappropriate model systems used to date. It also suggests that others may have overlooked many effective therapies due to a lack of response in 2D cultures.

The cognate ghrelin receptor, GHSR1a, does not bind to unacylated ghrelin and is undetectable in most breast cancer cell lines (Callaghan and Furness, 2014; Cassoni et al., 2001), suggesting that both ghrelin and unacylated ghrelin are acting at an alternate ghrelin receptor. In order to start dissecting the mechanism of action of unacylated ghrelin, cues were taken from our previous studies in adipose stromal cells where treatment results in inhibition of cAMP (Docanto et al., 2014). Results in the present study are consistent with these previous findings, and further demonstrate that unacylated ghrelin treatment is associated with activation of Gαi. The evidence for the importance of cAMP in breast cancer biology has been inconsistent. A number of studies have demonstrated pro-proliferative effects (Aronica et al., 1994; Küng et al., 1983), while others show that cAMP and activation of PKA inhibit cell growth (Chen et al., 1998; Cho-Chung et al., 1981; Fontana et al., 1987). Effects appear to be cell and context dependent. Directionality also seems more robust for studies performed using clinical samples. For example, cAMP levels have been shown to be 15 times higher in human breast cancer compared to normal breast tissue (Minton et al., 1974) and the overall ability to hydrolyze cyclic nucleotides has been shown to be decreased in faster growing and more invasive mammary cancers (Fajardo et al., 2014). PKA regulatory subunit expression and catalytic activity have also been shown to be increased in malignant compared to normal breast tissue (Gordge et al., 1996), suggesting that cAMP and PKA promote breast cancer growth. Here, we demonstrate that inhibition of cAMP formation or action mimics the effects of unacylated ghrelin to inhibit breast cancer cell growth in 3D. Our findings demonstrating that cells are insensitive to the effects of unacylated ghrelin in the presence of a BRAF or KRAS mutation also suggests that unacylated ghrelin acts upstream of these signaling proteins. Consistently, the growth-stimulatory effects of cAMP are dependent on MAPK signaling, known to be tightly regulated by BRAF/KRAS signaling.

Effects of unacylated ghrelin to inhibit cell cycle progression and stimulate apoptosis are also consistent with inhibition of MAPK and Akt signaling. Growth factor signaling has previously been shown to cause the accumulation of myc and cyclin D proteins, key drivers of the G1-to-S phase transition, as well as lead to the activation of CDK4 and the hyperphosphorylation of Rb (Hipfner and Cohen, 2004; O'Leary et al., 2016). Because cell number is not suppressed beyond what is stimulated by mitogens, effects are likely dependent on signaling downstream of the growth stimulus. The activity of unacylated ghrelin and unacylated ghrelin analogs to suppress MAPK and Akt signaling, myc, cyclin D3 and CDK4 expression, could therefore be leveraged in a therapeutic setting.

The unacylated ghrelin analog, AZP-531, was shown to be well tolerated in a phase I clinical trial performed in overweight and obese subjects with type II diabetes, and significant improvements in glucose variables were observed (Allas et al., 2016). A phase II study was also undertaken in individuals with Prader-Willi Syndrome, where daily subcutaneous injections of AZP-531 for 14 days caused a significant improvement in scores on the hyperphagia questionnaire, and a reduction in waist circumference, fat mass and post-prandial glucose levels (Allas et al., 2018). We have demonstrated efficacy of AZP-531 in reducing the growth of breast cancer cell lines and patient-derived cancer cells, consistent with effects of unacylated ghrelin. Based on the data obtained using a panel of breast cancer cell lines and effects seen in patient-derived tumor cells, we also identify the subset of breast cancers that will be resistant to treatment, that is TNBCs with KRAS or BRAF mutations, or TNBCs with high MAPK activity. KRAS has been shown to maintain mesenchymal features of TNBCs (Kim et al., 2015). As less than 1% of breast tumors carry these mutations (Tilch et al., 2014), it is likely that unacylated ghrelin and AZP-531 will be effective in the majority of breast cancers. Since unacylated ghrelin has also been shown to prevent chemotherapy-induced muscle cell death (Nonaka et al., 2017), there is potential to combine unacylated ghrelin or AZP-531 with chemotherapy, while also reducing cardiotoxicity.

Materials and methods

Experimental model and subject details

Human breast cancer tissues and preclinical patient derived xenograft models

Request a detailed protocol

Patient-derived breast tumors (PDX [Zhang et al., 2013]) were obtained from Dr. Giorgio Inghirami and estrogen receptor positive (ER+) breast tumors were obtained from Dr. Eleni Andreopoulou at Weill Cornell Medicine under IRB-approved protocols (WCM 1410015560 and 1603017108). All patients provided written informed consent. Cells were isolated and maintained in DMEM, supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin/streptomycin and 1% sodium pyruvate (ThermoFisher Scientific). Cells were grown at 37°C in a humidified atmosphere with 5% CO2. See Materials and method Details for cell growth and proliferation work specific procedures.

Human breast cancer cells

Request a detailed protocol

Human breast cancer cell lines, MCF7, T47D, ZR75, MDA-MB-231, MDA-MB-468, SKBR3, Hs578T and HEK293T, DU4475 and MDA-MB-157 were purchased from ATCC, USA. Human 4-OH tamoxifen resistant breast cancer cells (LCC2) were obtained from Prof. Robert Clarke. RKO, RKO-T29, RKO-A19, HCT116, HCT116-HWT, HCT116-HMUT were obtained from Dr. Jihye Yun and Dr. Bert Vogelstein and culture methods were performed as described previously (Yun et al., 2009). MCF7, MDA-MB-231, MDA-MB-468, MDA-MB-157, Hs578T, HEK293T were grown in DMEM (Invitrogen), supplemented with 10% fetal calf serum (FCS) (Invitrogen), 100 U/ml penicillin/streptomycin and 1% sodium pyruvate. ZR75, T47D, SKBR3 and DU4475 were grown in RPMI, supplemented with 10% FCS, 100 U/ml penicillin/streptomycin and 2 mM L-Glutamine (ThermoFisher Scientific). LCC2 were grown in phenol red-free DMEM, supplemented with 10% charcoal stripped serum (CCS; ThermoFisher Scientific), 100 U/ml penicillin/streptomycin and 1% sodium pyruvate. All cell lines were grown at 37°C with 5% CO2 in a humidified environment. Mycoplasma detection of all cell lines were tested and negative results were observed. Authenticated cell lines were used within six months of purchase from the ATCC.

Animal experimental model

Request a detailed protocol

All animal experiments were performed in Monash Animal Research Platform at Hudson Institute for Medical Research and Research Animal Resource Center at Weill Cornell Medicine. Experimental procedures were in accordance with guidelines for Animal Care and Use, approved by Monash University Animal Ethics Committee (MMCA2014/20) and Weill Cornell Medicine IACUC Protocol (REQ00016929). Athymic Balb/c nude, FVB/N and NOD Scid gamma (NSG) mice were used for this study. All mouse procedures were performed with 6–8 weeks old female mice and treatments administered via subcutaneous injection. See Materials and Methods details for in vivo work specific procedures.

Methods details

Cell growth and proliferation assays

Request a detailed protocol

3D culture of a panel of breast cancer and colon cancer cells were seeded at a density of 3,000 cells per well in media containing 30% growth-factor-reduced matrigel or 5 mg/ml collagen in optical-bottom 96-well plates. Patient-derived breast tumors were first dissociated into single cell suspensions and seeded at a density of 32,000 cells per well in media containing 30% matrigel. Cells were serum-starved overnight and cultured under different experimental conditions: 10–1000 pM unacylated ghrelin, 100 pM ghrelin, 1.5 × 10–13- 1.5 × 10–6 M doxorubicin, 1 × 10–18–1 × 10–7 M AZP-531, 20–200 ng/ml pertussis toxin, 10 μM U0126 (MEK inhibitor), 10 μM LY294002 (PI3K inhibitor), 60 nM KT5720, 1.4 μM SQ22536, 15 μM cAMPs-RP, 10 nM melatonin and the combination of doxorubicin/unacylated ghrelin or doxorubicin/AZP-531, with or without 10 nM estradiol or 10% serum or 1 μM forskolin and medium was replaced every 2 days. At the end of the experimental time point, cells were fixed in 100% methanol. The total number of cells, or the number of dead or proliferating cells, was assessed using Hoechst 33342, EarlyTox Dead Assay Kit, or Click-IT EdU Kit, respectively, and analyzed according to the manufacturer’s protocols.

Microscope image acquisition

Request a detailed protocol

Imaging of wells with greater than 95% coverage was acquired using a 10X/NA 0.8 objective by tiling 3 × 3 using confocal microscopy (Zeiss LSM880) with Axiocam. The fixed cells were imaged at room temperature, while the live cell images were acquired at 37°C placed inside a temperature and CO2 regulated chamber. The number of nuclei/Ethidium Homodimer-III+/EdU+ cells were measured using the Zeiss Enhanced Navigation (Zen) and counted using the Imaris software.

Binding assays

Request a detailed protocol

Cy3-tagged UAG binding assays were performed in 3D cultures of breast cancer cells. MCF7, MDA-MB-468 and MDA-MB-231 cells were seeded at a density of 3,000 cells per well in Matrigel in optical-bottom 96-well plates. Cells were serum-starved overnight. Hoechst nuclear stain was added prior to the live cell imaging. 1 μM Cy3-tagged UAG with 10% serum was then added and time-lapse confocal imaging was performed to examine the localization of peptide binding.

BRAF transient transfections

Request a detailed protocol

Two million cells were harvested and transfected with or without 1.5 µg of BRAFV600E plasmid (obtained from Dr. Dan Gough) using AMAXA Nucleofector (Lonza), according to supplier’s instructions. Transfected cells were then cultured according to above sections (cell growth and proliferation assay). After 5 days, cell number per field was assessed using Hoechst nuclear stain, confocal microscopy and Image J.

GNAI CRISPR KO generation

Request a detailed protocol

The lentiviral construct, lentiCRISPRv2 (containing hSpCas9 and the chimeric guide RNA cassettes), was digested using BsmBI. Prior to ligation, each pair of oligos (100 μM) were annealed. The oligos (GNAI sequence guide strands) designed were based on the target site sequence (20 bp) and were flanked on the 3’ end by a 3 bp NGG PAM sequence. See Supplementary file 1 - Oligonucleotide Sequences for GNAI sequence guide strands used in this study.

The gel-purified, BsmBl digested plasmid was ligated to the diluted (1:200) annealed oligo. The ligation product was transformed into competent Escherichia coli Stbl3TM cells. Ampicillin resistance colonies were selected for miniprep/maxiprep.

For virus production, HEK293T cells were plated in a 10 cm dish and transfected 24 hr later (80% confluence) with a prepared mix in DMEM media (no supplements) containing 5 μg of empty vector or gRNA plasmid of GNAI1, GNAI2 or GNAI3, 2.5 μg of psPAX2, 1.25 μg of VSV.G, and 15 μl of polyethylenimine (PEI; 1 μg/ml). 24 hr following transfection, media was replaced and supernatants (GNAI1, GNAI2 or GNAI3) were harvested and collected every 24 hr up to 72 hr post-transfection.

To generate CRISPR/Cas9 sgRNA stable breast cancer cell lines, breast cancer cells were plated in a 6-well plate. 24 hr following plating, cells were transfected with CRISPR/Cas9 sgRNA lentiviral vector and 8 μg/ml of polybrene. 24 hr after transfection, media was replaced. Cells were then selected in puromycin (2 μg/ml) for 5 days. CRISPR/Cas9 sgRNA stable breast cancer cell lines were then transduced with viral supernatants (GNAI1, GNAI2 or GNAI3 virus) in the presence of polybrene (8 μg/ml). 24 hr after transduction cells were selected in Blasticidin S (2 μg/ml) for 5 days. Selected cells were used to perform cell growth assays.

ERK and Akt activity assays

Request a detailed protocol

ERK activity assay: The EKAR FRET-based system was used to monitor ERK activity. Briefly, breast cancer cells were co-transfected with pPBJ—puro-FRET3-EKAR-nls and the pCMV- hyPBase transposase vector (obtained from Dr. John Albeck), and stably transfected cells selected using FACS for EYFP and ECFP-positive cells. Cells were then cultured in 3D and serum-starved overnight. Prior live cell time-lapse confocal imaging, medium was replaced with 100 pM unacylated ghrelin and 10% serum. Images were acquired following FRET and analyzed using Imaris software. Data were normalized to ECFP signal.AKT activity assay: pMSCV-puro-Foxo3a-H212R-N400-mCherry (obtained from Dr. John Albeck) was transfected into breast cancer cells using Amaxa Nucleofector. After 3 days of puromycin (2 µg/ml) selection, transfected cells were cultured in 3D. Cells were serum-starved overnight. Prior to live cell imaging, medium was replaced with 100 pM unacylated ghrelin and 10% serum. Time-lapse imaging was performed to examine mCherry-tagged FoxO3 localization. Data were analyzed using Hoechst nuclear stain to mask nuclei using Imaris software.

cAMP assay

Request a detailed protocol

The Lance Ultra cAMP kit was used to measure the effect of unacylated ghrelin on cAMP production. All kit components were prepared according to manufacturer’s specifications. Briefly, MCF7 (475 cells/well) were incubated at room temperature for 60 min with unacylated ghrelin (10–1000 pM), in the absence or presence of forskolin (0.5 μM). The Eu-cAMP tracer and Ulight-anti cAMP reagents were then added for 1 hr at room temperature. The plate was then read using the Envision TRF capable reader (Perkin Elmer) and fluorescence was measured with excitation wavelengths of 340 or 340 nm and emission of 665 nm according to the manufacturer’s instructions.

Intracellular calcium release assay

Request a detailed protocol

Intracellular Ca2+ levels were measured in MCF7, ZR75 and J110 cells by fluorescence using the Flexstation (Molecular Devices, Sunnyvale, CA, USA) as previously described (Callaghan et al., 2012). Briefly, cells were plated, allowed to reach 50–70% confluency, and loaded with 2 μM fura 2-AM for 1 hr in the presence of 2.5 mM probenecid and 0.01% pluronic F-127 at 37°C. Cells were then washed twice with assay buffer, and changes in fluorescence in response to drug addition were measured over 100 s using excitation wavelengths of 340 and 380 nm and emission of 520 nm.

Gαi activation assay

Request a detailed protocol

Gαi activation assay was performed on MCF7 cells according to the manufacturer’s protocol. Briefly, cell lysates from 3D cell culture were incubated with an anti-active Gαi antibody (1:1000). The precipitated active Gαi was immunoblotted with an anti- Gαi antibody (1:1000). Bound antibodies were revealed with HRP-conjugated secondary antibodies (1:2000) using SuperSignal West pico chemiluminescent solution (Pierce, Rockford, IL). Protein amount was normalized to β–tubulin. Membranes were scanned and the densitometric analysis of the bands was performed using the ChemiDoc MP imaging system (BioRad).

Western blot analysis

Request a detailed protocol

Western blotting was performed as described previously (Brown et al., 2009). Briefly, cells isolated from or 5 mg/ml Collagen (obtained from Dr. Jason Spector) were lysed in RIPA lysis buffer (Sigma-Aldrich) supplemented with 100x Protease/Phosphatase inhibitor cocktail (Cell Signaling Technology Inc). Cell extracts (20 μg per lane) were separated by NuPAGE 4–12% Bis-Tris protein gels (ThermoFisher Scientific) and transferred to nitrocellulose membranes. The following primary antibodies were used: CDK4, Cyclin D3, phospho-Rb (Ser795), Rb, Bcl-2, Bax, phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204), p44/42 MAPK (Erk1/2), phospho-p90RSK (Ser380), RSK1/RSK2/RSK3, c-Myc, phospho-Akt (Ser473), Akt, phospho-p70 S6 Kinase (Thr389), p70 S6 Kinase. Bound antibodies were revealed with HRP conjugated secondary antibodies. The following secondary antibodies were used: Goat anti-mouse IgG H and L (HRP), Donkey anti- Rabbit IgG H and L (HRP). Detected bands were consistent with expected molecular weights based on antibody datasheets (Key Resource Table). β–actin was used as a loading control. See Key Resources Table for antibody details used in this study. Membranes were scanned using the western lightning plus-ECL (Thermo Fisher Scientific). Signal intensities were quantified using ImageLab software (BioRad).

Fluorescence activated cell sorting (FACS)

Request a detailed protocol

FACS analysis was performed to characterize effects of unacylated ghrelin on cell cycle and apoptosis. Cells were plated at a density of 500,000 cells in a 10 cm2 petri dish. Cells were serum- starved overnight with phenol red-free media and treated with different concentrations (0–100 pM) of unacylated ghrelin, with or without 10 nM estradiol. In order to determine effects on cell cycle, cells were treated for 5 days, with media being changed every 2 days. After 5 days, cells were harvested and fixed with ice cold 70% ethanol, stored overnight at −20°C, and stained with propidium iodine (PI) staining buffer (1 mg/ml RNase A, 0.1% Triton X-100, 100 μg/ml PI in PBS). To evaluate effects on cell apoptosis, cells were treated for 6 hr, harvested and then stained with Annexin V-FITC for 15 min at room temperature in PBS. Cells were analysed with a FACSCANTO II flow cytometer (BD Biosciences, USA). For FACS data analysis by FlowJo software, forward scatter (FS) vs. side scatter (SS) plots were used for gating cells and to identify any changes in the scatter properties of the cells. Annexin V FITC-A vs Propidium Iodide-A plots from the gated cells show the populations corresponding to viable and non-apoptotic (Annexin V–PI–), early (Annexin V+PI–), and late (Annexin V+PI+) apoptotic cells.

Analysis of cell cycle progression using the fluorescence ubiquitination cell cycle indicator (FUCCI)

Request a detailed protocol

To investigate cell cycle progression and division in live cells, we used the fluorescent ubiquitination-based cell cycle indicator (FUCCI) to track the G1/G0 phase and S/G2/M phases. Breast cancer cells (15 × 104 cells/well) in six wells plate were seeded in 2D and transduced with the Premo FUCCI Cell Cycle Sensor according to the manufacturer’s protocol. After optimal expression of the FUCCI sensor was achieved (16 hr), cells were detached by TrypLE Express (Thermo Fisher Scientific), counted using a hemocytometer and seeded in media containing 30% matrigel. Cells were serum-starved overnight, treated with 100 pM unacylated ghrelin for 5 days with 10% serum and medium was replaced every 2 days. At endpoint, fluorescence was analyzed using confocal microscopy and cell counts obtained. Data are presented as a percentage of the total number of fluorescent cells examined.

Xenograft and allograft studies

Request a detailed protocol

One million MCF7 or ZR75 cells were injected into the mammary fat pad of 6-week-old ovary-intact female athymic Balb/c nude (immunodeficient) mice. Estrogen pellets were prepared in the laboratory using 17β-estradiol (estrogen) powder and silicone according to previous publications (Laidlaw et al., 1995) and were implanted (0.8 mg/pellet, 60-day release) subcutaneously between the shoulder blades at the time of ER+ breast cancer cell injection. After injection, the mice were monitored daily for well-being, pain and distress, and for tumor growth by palpation. Tumors typically appeared within 14 days, with an engraftment rate of approximately 75%. Once palpable, the tumors were measured daily in the long (L) and short (W) axes with digital callipers, and tumor volume estimated using the standard formula (LxW2)/2. Once the tumor reached a volume of 200 mm3, the animal was randomized to receive saline (vehicle control) or unacylated ghrelin) via subcutaneous injection for a maximum of 28 days or until tumor size reached 10 mm in any axis, at which point the mice were humanely sacrificed by cervical dislocation.

A syngeneic breast cancer mouse model was created in female FVB/N mice. 1.25 × 105 J110 cells (obtained from Dr. Myles Brown) were injected into the mammary fat pad of 6-week-old ovary-intact FVB (Immunocompetent) mice. Once the tumor reached a volume of 70 mm3, the mice were randomized to receive saline (vehicle control) or treatment (unacylated ghrelin or AZP-531) via subcutaneous injection. After 7 days of treatment, the mice were humanely sacrificed by cervical dislocation. Additional studies were also performed in NSG mice. MDA-MB-468 (1 × 106) were injected into the mammary fat pad of 6 week old ovary-intact NSG (immunodeficient) mice. Once the tumor reached a volume of 150 mm3, the mice were randomized to receive saline (vehicle control) or AZP-531 via subcutaneous injection. After 28 days of treatment, the mice were humanely sacrificed by cervical dislocation.

Measurement of apoptotic cells from xenograft/allograft studies

Request a detailed protocol

MCF7, ZR75 and J110 tumor sections post treatment with 100 μg/kg or 200 μg/kg of unacylated ghrelin were assessed using Hoechst nuclear stain and confocal microscopy. Quantification of cells with apoptotic nuclei was performed using Image J (Fiji).

Microarray datasets

Request a detailed protocol

Gene expression profiles from GSE34412 were retrieved via GEO2R (https://www.ncbi.nlm.nih.gov/geo/geo2r/). Expression profiles were based on a Custom Human Agilent array (GPL8269). GEO2R was used to perform differential gene expression analysis between responders (GSM847888, GSM847905) and non-responders (GSM847901, GSM847893). Ingenuity Pathway Analysis (IPA) was used to determine regulatory pathways that distinguished between responders and non-responders. Differences in gene expression were consistent with activation of upstream regulator MAPK1 in non-responders vs. responders (Z-score 3.888; overlap p-value 3.07E-15), with 86 genes of 144 having measurements consistent with MAPK1 activation. Log expression values corresponding to the MAPK target genes were extracted from GEO2R for each sample. If a gene was represented by more than one probe, the expression value corresponding to the probe with the largest absolute fold-change between responders and non-responders was selected. Genes predicted by IPA to be activated were visualized in a heatmap (expression values were centered and scaled).

Statistical analysis

Request a detailed protocol

All experiments were performed at least twice with n ≥ 3 per experiment and all data are expressed as the mean ± SEM. Statistical analysis was carried out with software Graph Pad Prism 7. For experiments with multiple comparisons, statistical analysis was performed using one-way ANOVA followed by Dunnett multiple comparison, where means of each column were compared to the mean of a control column. For experiments with comparison of two independent groups, statistical analysis was performed using two-way ANOVA followed by Dunnett multiple comparison, where means of each experimental condition were compared to the control. A p-value was reported and significance was classified as p<0.05(*), p≤0.005(**), p≤0.0005(***), p≤0.0001(****), n.s.: not significant.

Appendix 1

Appendix 1—key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Strain, strain background (Escherichia coli)One Shot Stbl3
Chemically Competent E. coli		ThermoFisher ScientificCat# C737303
Strain, strain background (Mus-musculus)BALB/c-Foxn1nu/Arc (BALB/c nude) miceAnimal Resources Centre, AustraliaCat# BCNU; RRID:MGI:2161064Female, 6 week old
Strain, strain background (Mus-musculus)FVB/NJArc (FVB/N) miceAnimal Resources Centre, AustraliaCat# 001800; RRID:IMSR_JAX:001800Female, 6 week old
Strain, strain background (Mus-musculus)NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) miceNOD Scid gamma, Jackson laboratoryCat# 005557; RRID:IMSR_JAX:005557Female, 6 week old
Cell line (Homo-sapiens)MCF7ATCCATCC Cat# HTB-22; RRID:CVCL_0031
Cell line (Homo-sapiens)T47DATCCATCC Cat# HTB-133; RRID:CVCL_0553
Cell line (Homo-sapiens)ZR75-1ATCCATCC Cat# CRL-1500; RRID:CVCL_0588
Cell line (Homo-sapiens)MDA-MB-231ATCCATCC Cat# HTB-26; RRID:CVCL_0062
Cell line (Homo-sapiens)MDA-MB-468ATCCATCC Cat# HTB-132; RRID:CVCL_0419
Cell line (Homo-sapiens)SKBR3ATCCATCC Cat# HTB-30; RRID:CVCL_0033
Cell line (Homo-sapiens)Hs578TATCCATCC Cat# HTB-126; RRID:CVCL_0332
Cell line (Homo-sapiens)HEK293TATCCATCC Cat# CRL-1573; RRID:CVCL_0045
Cell line (Homo-sapiens)DU4475ATCCATCC Cat# HTB-123; RRID:CVCL_1183
Cell line (Homo-sapiens)MDA-MB-157ATCCATCC Cat# HTB-24; RRID:CVCL_0618
Cell line (Homo-sapiens)LCC2obtained from Prof. Robert Clarke
Cell line (Homo-sapiens)RKOYun et al., 2009
Cell line (Homo-sapiens)RKO-T29Yun et al., 2009
Cell line (Homo-sapiens)RKO-A19Yun et al., 2009
Cell line (Homo-sapiens)HCT116Yun et al., 2009
Cell line (Homo-sapiens)HCT116-HWTYun et al., 2009
Cell line (Homo-sapiens)HCT116-HMUTYun et al., 2009
Cell line (Homo-sapiens)J110obtained from Dr. Myles Brown
Biological sample (human)Patient-derived breast tumorsobtained from Dr. Giorgio InghiramiSee Materials and methods section
Biological sample (human)Estrogen receptor positive (ER+) breast tumorsobtained from Dr. Eleni AndreopoulouSee Materials and methods section
AntibodyGoat anti-mouse IgG H and L (HRP)AbcamCat# ab205719; RRID:AB_27550491:5000
AntibodyDonkey anti-Rabbit IgG H and L (HRP)AbcamCat# ab7083; RRID:AB_9554161:5000
AntibodyMouse Monoclonal anti-Tubulin, beta, (KMX-1)MilliporeSigmaCat# MAB3408; RRID:AB_946501:10000
AntibodyMouse Monoclonal anti-β-Actin−PeroxidaseSigma-AldrichCat#A3854; RRID:AB_2620111:5000
AntibodyRabbit Monoclonal anti-CDK4 (D9G3E)Cell SignalingCat# 12790; RRID:AB_26311661:1000
AntibodyMouse Monoclonal anti-Cyclin D3 (DCS22)Cell SignalingCat# 2936; RRID:AB_20708011:2000
AntibodyRabbit Polyclonal anti-phospho-Rb (Ser795)Cell SignalingCat# 9301; RRID:AB_330013)1:1000
AntibodyMouse Monoclonal anti-Rb (4H1)Cell SignalingCat# 9309; RRID:AB_8236291:2000
AntibodyRabbit Monoclonal anti-Bcl-2 (D55G8) (Human Specific)Cell SignalingCat# 4223; RRID:AB_19039091:1000
AntibodyRabbit Monoclonal anti-BaxCell SignalingCat# 2772; RRID:AB_106958701:1000
AntibodyRabbit Polyclonal anti-phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204)Cell SignalingCat# 9101;
RRID:AB_331646		1:1000
AntibodyRabbit Monoclonal anti-p44/42 MAPK (Erk1/2) (137F5)Cell SignalingCat# 4695;
RRID:AB_390779		1:1000
AntibodyRabbit Monoclonal anti-phospho-p90RSK
(Ser380) (D3H11)		Cell SignalingCat# 11989; RRID:AB_26876131:1000
AntibodyRabbit Monoclonal anti-RSK1/RSK2/RSK3 (32D7)Cell SignalingCat# 9355; RRID:AB_6599001:1000
AntibodyRabbit Monoclonal
anti-c-Myc (D84C12)		Cell SignalingCat# 5605; RRID:AB_19039381:1000
AntibodyRabbit Polyclonal anti-phospho-Akt (Ser473)Cell SignalingCat# 9271; RRID:AB_3298251:1000
AntibodyRabbit Polyclonal anti-AktCell SignalingCat# 9272; RRID:AB_3298271:1000
AntibodyRabbit Monoclonal anti-phospho-p70 S6 Kinase (Thr389) (108D2)Cell SignalingCat# 9234; RRID:AB_22698031:1000
AntibodyRabbit Monoclonal anti-p70 S6 Kinase (49D7)Cell SignalingCat# 2708; RRID:AB_3907221:1000
Recombinant DNA reagentpPBJ—puro-FRET3-EKAR-nlsobtained from Dr. John Albeck
Recombinant DNA reagentpCMV-hyPBase
transposase vector		obtained from
Dr. John Albeck
Recombinant DNA reagentpMSCV-puro-Foxo3a-H212R-N400-mCherryobtained from Dr. John Albeck
Recombinant DNA reagentlentiCRISPR v2Sanjana et al., 2014Addgene Cat# 52961; RRID:Addgene_52961
Recombinant DNA reagentBRAFV600E plasmidobtained from Dr. Dan Gough
Recombinant DNA reagentpsPAX2This paperAddgene Cat# 12260; RRID:Addgene_12260
Recombinant DNA reagentVSV-GReya et al., 2003Addgene Cat# 14888;
RRID:Addgene_14888
Sequence-based reagentSee Supplementary file 1 for GNAI sequence guide strands used in this study
Peptide, recombinant proteinRat des-octanoyl ghrelinChina PeptidesCat# Rat des-octanoyl ghrelin
Peptide, recombinant protein[Des-octanoyl]-Ghrelin (rat)Tocris BioscienceCat# 2951
Peptide, recombinant proteinGhrelin (rat)Tocris BioscienceCat# 1465
Peptide, recombinant proteinAZP531MedChem ExpressCat# HY-P0231
Peptide, recombinant proteinCy3-tagged UAGPepmic Co, LtDCat# Cy3-GR-28
Commercial
assay or kit		EarlyTox Live/Dead Assay KitMolecular DevicesCat# P/N R8340
Commercial
assay or kit		Click-iT Plus EdU Alexa Fluor 488 Imaging KitThermoFisher ScientificCat# C10637
Commercial
assay or kit		Cell Line NucleofectorKit VLonzaCat# VCA-1003
Commercial
assay or kit		Lance Ultra cAMP Detection KitPerkin ElmerCat# TRF0262
Commercial
assay or kit		Gαi Activation Assay KitNew East BiosciencesCat# 80301
Commercial
assay or kit		Premo FUCCI Cell Cycle Sensor (BacMam 2.0)ThermoFisher ScientificCat# P36237
Commercial
assay or kit		PureYield Plasmid
Miniprep System		PromegaCat# A1223
Commercial
assay or kit		PureYield Plasmid Maxiprep SystemPromegaCat# A2392
Chemical compound, drugPertussis Toxin from B. pertussis, Lyophilized (Salt-Free)List Biological LaboratoriesCat# 181
Chemical compound, drugU0126Cell SignalingCat# 9903S
Chemical compound, drugLY294002Cell SignalingCat# 9901
Chemical compound, drugKT 5720Tocris BioscienceCat# 1288
Chemical compound, drugSQ 22536Tocris BioscienceCat# 1435
Chemical compound, drugcAMPS-Rp, triethylammonium saltTocris BioscienceCat# 1337
Chemical compound, drugMelatoninTocris BioscienceCat# 3550
Chemical compound, drugDoxorubicin hydrochlorideSigma-AldrichCat# D1515
Chemical compound, drugForskolinSigma-AldrichCat# F3917
Chemical compound, drugβ-EstradiolSigma-AldrichCat# E8875-1G
Software, algorithmPrismGraphPadRRID:SCR_005375http://www.graphpad.com/scientific-software/prism/
Software, algorithmImarisBitplaneRRID:SCR_007370https://imaris.oxinst.com/
Software, algorithmFijiImageJRRID:SCR_002285https://imagej.net/Fiji
Software, algorithmImage LabBIO-RADRRID:SCR_014210http://www.biorad.com/en-us/product/image-lab-software?ID=KRE6P5E8Z;
Software, algorithmZEN softwareZEISSRRID:SCR_013672https://www.zeiss.com/
microscopy/
us/products/microscope-software/zen-lite.html
Software, algorithmIngenuity Pathway AnalysisQiagenRRID:SCR_008653https://www.qiagenbioinformatics.
com/products/ingenuity-pathway-analysis
Software, algorithmFlowJo softwareFlowJoRRID:SCR_008520https://www.flowjo.com/solutions/flowjo
OtherHoechst 33342, Trihydrochloride,
Trihydrate		ThermoFisher ScientificCat# H3570
OtherFITC Annexin VBD PharmingenCat# 556420
OtherPropidium IodineThermoFisher ScientificCat# P1304MP
OtherFura-2, AM, cell permeantThermoFisher ScientificCat# F1221
OtherProbenecid, Water SolubleThermoFisher ScientificCat# P36400
OtherPluronic F-127ThermoFisher ScientificCat# P6867
OtherBsmBINew England BioLabsCat# R0580S
OtherPolyethylenimine,
linear (PEI)		Polysciences, IncCat# 23966–1
OtherPolybreneSanta Cruz BiotechnologyCat# sc-134220
OtherPuromycin
Dihydrochloride		ThermoFisher ScientificCat# A1113803
OtherBlasticidin S HCl (10 mg/mL)ThermoFisher
Scientific		Cat# A1113903
OtherCorning Matrigel Growth Factor Reduced (GFR) Basement Membrane Matrix, Phenol Red-free, *LDEV-free,CorningCat# 356231
OtherCollagenObtained from Dr. Jason Spector

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for all figures.

The following previously published data sets were used
    1. Zhang X
    2. Prat A
    3. Perou CM
    (2013) NCBI Gene Expression Omnibus
    ID GSE34412. A Renewable Tissue Resource of Phenotypically Stable, Biologically and Ethnically Diverse, Patient-derived Human Breast Cancer Xenografts.
    https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE34412

References

    1. Giltnane JM
    2. Balko JM
    (2014)
    Rationale for targeting the ras/MAPK pathway in triple-negative breast Cancer
    Discovery Medicine 17:275–283.

Decision letter

  1. Richard M White
    Senior Editor; Memorial Sloan Kettering Cancer Center, United States
  2. Yi Arial Zeng
    Reviewing Editor; Chinese Academy of Sciences, China
  3. Yi Arial Zeng
    Reviewer; Chinese Academy of Sciences, China
  4. Jason Carroll
    Reviewer; Cancer Research UK, United Kingdom
  5. Jeremy Borniger
    Reviewer; Cold Spring Harbor Laboratory, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The hormone ghrelin is responsible for the stimulation of growth hormone release and appetite, yet little is known of the effects of its unacylated form. The present study uncovers an inhibitory role of unacylated ghrelin on breast cancer cells, and explores in vivo and ex vivo efficacy of a therapeutic analogue AZP-531. Targeting unacylated ghrelin may provide a new approach for the breast cancer treatment.

Decision letter after peer review:

Thank you for submitting your article "Three-dimensional growth of breast cancer cells potentiates the anti-tumor effects of unacylated ghrelin and AZP-531" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Yi Arial Zeng as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen Richard White as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Jason Carroll (Reviewer #2); Jeremy Borniger (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

As the editors have judged that your manuscript is of interest, but as described below that additional experiments are required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

In particular, please pay attention to the comments 1-2 by reviewer 1 and comments 1-3 by reviewer 2. While some of these will require additional data, others may be addressed be a careful editing of the conclusions made in the manuscript.

Reviewer #1:

The manuscript by Au et al., by the Brown lab uncovers a new role of unacylated ghrelin (UAG) for breast cancer cell inhibition. This inhibition effect is obvious in 3D culture, but not in 2D culture that was used in previous studies. Au and co-authors discover that UAG leads to cell cycle arrest and apoptosis, with a mechanism involves activation of Gαi and suppression of MAPK signaling. This mechanism is reflected by the lack of response of UAG when treating cancer cells with BRAF or HRAS mutation. A cyclic analog of UAG, AZP-531 also inhibits breast cancer cell growth in vitro and in xenografts.

Overall, this study is nicely designed and executed with a battery of in vitro and in vivo analyses. This study has a clinical implication for breast cancer treatment.

My only main concern is the clarity in writing and statistical analysis. They are specified below. Fortunately, those are issues that can be addressed without further experiments.

1) The statements regarding the effects on TNBC are confusing. In Figure 1C, the two UAG-responsive TNBC lines were MDA-MB-468 (basal) and MDA-MB-157 (mesenchymal), the three not-responsive TNBC lines were DU4474 (BRAF*), Hs578T (HRAS*) and MDA-MB-231 (BRAF* HRAS*). It is clear that the response is dependent on the mutation state rather than the basal/mesenchymal subtype.

In Figure 5G, UAG was tested in two TNBC-PDX lines. One responded, one didn't. One happened to be basal subtype, and the other is mesenchymal. BRAF and KRAS mutation had not been characterized in these patient samples, but MAPK-target genes were differentially expressed between responsive and no-responsive TNBC samples (subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibit tumor growth in xenograft models and patient-derived tumor cells”). The statement that "UAG caused a significant reduction in the basal-like subset of TNBC cells, but not mesenchymal TNBCs" (subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibit tumor growth in xenograft models and patient-derived tumor cells”) is misleading, so does the labeling of basal and mesenchymal in the figure. These should be fixed, probably by directly using the name of the PDX line.

2) The statistic analyses need to be clarified. In almost all panels, the "not significant (n.s)" has been missed. These should all be fixed. (Figure 1C-F, 2D, 2G, Figure 1—figure supplement 1A-C etc.)

There are several places where the n.s is especially important. For example, in Figure 1C, the n.s should be indicated for the three UAG-not-responsive TNBC cell lines. This is essential as "not-responsive" is one of the main conclusions in this panel. Similarly, in Figure 4F and Figure 5G, the n.s should be labelled for the UAG-not-responsive TNBC lines, MB-231 (4F) and PDX line 3204-TG6 (5G).

3) In many places of the figure legend, it went "…n=3. Experiments were repeated at least twice." Is n=3 here a technical repeat? If it is experimental repeat, it should be written as n>=2. Please modify or clarify.

Reviewer #2:

This manuscript investigates the role of the hormone ghrelin and the unacylated form of ghrelin, which has a distinct functional profile. The authors assess the growth inhibitory effects of unacylated ghrelin and ghrelin in a panel of breast cancer cells grown in 3D or 2D, revealing a distinction in the anti-tumor activity in specific models. Unacylated ghrelin was shown to only elicit growth suppressive activity in full media and some work is conducted to show that MCF7 response is blunted in the presence of B-Raf mutants. Some mechanistic work is provided to show that G-alpha-i is a potential mediator of unacylated ghrelin activity in a process that involves cAMP. Additional work is provided to link unacylated ghrelin with the MAPK pathway and some downstream endpoints are explored, with follow-up work showing that unacylated ghrelin mediates its effects via both cell cycle arrest and induced apoptosis. Finally, in vivo and ex vivo efficacy of unacylated ghrelin and a therapeutic analogue called AZP-531 is explored.

This is an interesting paper, on a very niche and underexplored hormonal pathway. The paper is well written and the figures are clear and convincing. Overall, the work is well conducted and the conclusions are valid, but a few issues need to be addressed.

- I understand the rationale behind expressing B-Raf mutants in MCF7 cells to assess the altered response to unacylated ghrelin, but what are the basal physiological effects of this on the MCF7 model? Does it change estrogen dependence or basal cell viability/phenotype? Given that B-Raf mutations are not associated with ER+ disease, forcing a non-relevant mutant into a cell line model, likely has profound basal changes that will confound the experimental questions.

- In the in vivo experiments, there was no mention of weight loss or potential physiological consequences of giving unacylated ghrelin (or analogues) on normal tissue functioning. Did the authors assess some of the standard 'health' indicators in this experiment?

- The use of AZP-531 is justified because of improved chemical properties, but the assumption is that all the previous pathway work (on cAMP, MAPK etc) are valid for AZP-531. Is this the case or is there any evidence to support the conclusions that the mechanistic insight derived from unacylated ghrelin is directly applicable to the mode of action for AZP-531? it would be useful if even just one or two of the key functional endpoints explored with unacylated ghrelin, were validated with AZP-531.

- The start of the Results section was confusing to a non-expert. The text states "The dependence on 3D culture was examined for unacylated ghrelin, where treatment with 100 pM resulted in suppression of cell growth in matrigel and collagen, but not in 2D (Figure 1—figure supplement 1A and 1B). At this dose, no effects on MDA-MB-231 cell growth were observed in either 2D or 3D cultures". It's not clear what cell model the first sentence refers to. I assumed it was the MDA-MB-231 cells, but the first line suggests there is growth inhibition in 3D, but the second sentence suggested there isn't growth inhibition in this cell line model.

Reviewer #3:

The authors examined the effects of unacylated ghrelin (UAG) on breast cancer cell growth in vitro and in xenograft models. This is an important area given the limited amount of research that has been done on the mechanisms of UAG actions on cancer cells. I applaud the authors for systematically characterizing the effects of UAG on cancer cell growth using 3D growth assays as well as many different cell lines with well characterized mutational signatures. Step-by-step inhibitor studies on different components of the cAMP, MAPK, and Akt signaling pathways strongly supports their hypothesis that UAG actions are mediated (at least in part) via a GPCR coupled to Gai. For each of the conclusions presented in the manuscript, several parallel experiments were conducted and for the most part, all of these acted synergistically to support the conclusions drawn from the studies. The paper is also well written and I did not have a problem following the narrative flow. I have no major concerns for the manuscript at this time.

https://doi.org/10.7554/eLife.56913.sa1

Author response

Reviewer #1:

The manuscript by Au et al., by the Brown lab uncovers a new role of unacylated ghrelin (UAG) for breast cancer cell inhibition. This inhibition effect is obvious in 3D culture, but not in 2D culture that was used in previous studies. Au and co-authors discover that UAG leads to cell cycle arrest and apoptosis, with a mechanism involves activation of Gαi and suppression of MAPK signaling. This mechanism is reflected by the lack of response of UAG when treating cancer cells with BRAF or HRAS mutation. A cyclic analog of UAG, AZP-531 also inhibits breast cancer cell growth in vitro and in xenografts.

Overall, this study is nicely designed and executed with a battery of in vitro and in vivo analyses. This study has a clinical implication for breast cancer treatment.

My only main concern is the clarity in writing and statistical analysis. They are specified below. Fortunately, those are issues that can be addressed without further experiments.

1) The statements regarding the effects on TNBC are confusing. In Figure 1C, the two UAG-responsive TNBC lines were MDA-MB-468 (basal) and MDA-MB-157 (mesenchymal), the three not-responsive TNBC lines were DU4474 (BRAF*), Hs578T (HRAS*) and MDA-MB-231 (BRAF* HRAS*). It is clear that the response is dependent on the mutation state rather than the basal/mesenchymal subtype.

In Figure 5G, UAG was tested in two TNBC-PDX lines. One responded, one didn't. One happened to be basal subtype, and the other is mesenchymal. BRAF and KRAS mutation had not been characterized in these patient samples, but MAPK-target genes were differentially expressed between responsive and no-responsive TNBC samples (subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibit tumor growth in xenograft models and patient-derived tumor cells”). The statement that "UAG caused a significant reduction in the basal-like subset of TNBC cells, but not mesenchymal TNBCs" (subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibit tumor growth in xenograft models and patient-derived tumor cells”) is misleading, so does the labeling of basal and mesenchymal in the figure. These should be fixed, probably by directly using the name of the PDX line.

We appreciate the reviewer’s insights and have modified the text and figures related to TNBC-PDX lines without focusing on their TNBC classification. Specifically, we have removed “basal” and “mesenchymal” labels from Figure 5G and Figure 4—figure supplement 1E, and associated figure legends. We have also modified the Results section as follows:

Subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibits tumor growth in xenograft models and patient-derived tumor cells”: “Unacylated ghrelin at 100pM caused a significant reduction in the serum-induced growth of ER+ breast cancer cells and the basal-like subset of TNBC cells, but not mesenchymal TNBCs (Figure 5G, Figure 4—figure supplement 1E; Table 1).” was replaced with “Unacylated ghrelin at 100pM caused a significant reduction in the serum-induced growth of ER+ breast cancer cells, while inhibiting or having no effect in the TNBC patient samples examined (Figure 5G, Figure 4—figure supplement 1E; Table 1).”

2) The statistic analyses need to be clarified. In almost all panels, the "not significant (n.s)" has been missed. These should all be fixed. (Figure 1C-F, 2D, 2G, Figure 1—figure supplement 1A-C etc.)

There are several places where the n.s is especially important. For example, in Figure 1C, the n.s should be indicated for the three UAG-not-responsive TNBC cell lines. This is essential as "not-responsive" is one of the main conclusions in this panel. Similarly, in Figure 4F and Figure 5G, the n.s should be labelled for the UAG-not-responsive TNBC lines, MB-231 (4F) and PDX line 3204-TG6 (5G).

We agree with the reviewer that addition of “n.s.” to bar graphs would emphasize a lack of effect of treatment and have modified Figure 1C-F, Figure 2C,D,G, Figure 4F, Figure 5D,G, Figure 1—figure supplement 1A-C, Figure 2—figure supplement 1B, Figure 3—figure supplement 1C,D,F,G,H accordingly. The definition of “n.s.” was added to the Statistical analysis section of the manuscript.

3) In many places of the figure legend, it went "…n=3. Experiments were repeated at least twice." Is n=3 here a technical repeat? If it is experimental repeat, it should be written as n>=2. Please modify or clarify.

In order to contend with routinely used in vitro experimental approaches without misleading reviewers, we replaced “n=x/group” with “x replicates/group” in all the relevant figure legends.

Reviewer #2:

This manuscript investigates the role of the hormone ghrelin and the unacylated form of ghrelin, which has a distinct functional profile. The authors assess the growth inhibitory effects of unacylated ghrelin and ghrelin in a panel of breast cancer cells grown in 3D or 2D, revealing a distinction in the anti-tumor activity in specific models. Unacylated ghrelin was shown to only elicit growth suppressive activity in full media and some work is conducted to show that MCF7 response is blunted in the presence of B-Raf mutants. Some mechanistic work is provided to show that G-alphaα-i is a potential mediator of unacylated ghrelin activity in a process that involves cAMP. Additional work is provided to link unacylated ghrelin with the MAPK pathway and some downstream endpoints are explored, with follow-up work showing that unacylated ghrelin mediates its effects via both cell cycle arrest and induced apoptosis. Finally, in vivo and ex vivo efficacy of unacylated ghrelin and a therapeutic analogue called AZP-531 is explored.

This is an interesting paper, on a very niche and underexplored hormonal pathway. The paper is well written and the figures are clear and convincing. Overall, the work is well conducted and the conclusions are valid, but a few issues need to be addressed.

- I understand the rationale behind expressing B-Raf mutants in MCF7 cells to assess the altered response to unacylated ghrelin, but what are the basal physiological effects of this on the MCF7 model? Does it change estrogen dependence or basal cell viability/phenotype? Given that B-Raf mutations are not associated with ER+ disease, forcing a non-relevant mutant into a cell line model, likely has profound basal changes that will confound the experimental questions.

This is a very interesting point. As shown below, we have reanalyzed Figure 1D of the manuscript to examine the degree of stimulation by both E2 and FBS in cells transfected with mutant BRAF vs control vector by normalizing data to vehicle control treatment. We believe that this approach provides information related cell behavior, i.e. growth in response to stimuli, while retaining the key message related to response to UAG. As such, we have replaced Figure 1D in the manuscript with the revised Figure. Interestingly, while cells maintain similar response to growth factors (FBS), their response to estradiol is diminished. That said, cells do remain responsive to estradiol and UAG is unable to suppress either the estradiol- or growth factor-stimulated proliferation of mutant BRAF-transfected cells.

As a further morphological characterization of the cells, we examined the effect of mutant BRAF transfection on nuclear shape. No obvious differences were observed, e.g. elongation of nuclei seen with EMT, when compared to vector control-transfected cells.

Author response image 1
Download asset Open asset

- In the in vivo experiments, there was no mention of weight loss or potential physiological consequences of giving unacylated ghrelin (or analogues) on normal tissue functioning. Did the authors assess some of the standard 'health' indicators in this experiment?

Mice treated with unacylated ghrelin or AZP-531 did not lose weight compared to saline control (see new Supplementary raw data). We had two unacylated ghrelin-treated mice in the MCF7 cell xenograft study assessed by Australian Phenomics Network Histopathology and Organ Pathology, The University of Melbourne, and compared to vehicle control. No abnormalities specific to treatment were identified in Adrenal glands, Bladder, Brain, Cecum, Cervix, Clitoral gland, Colon, Duodenum, Eyes, Gall bladder, Harderian glands, Head, Heart, Hind leg (Long bone, Bone marrow, Synovial joint, Skeletal muscle), Ileum, Jejunum, Kidney, Liver, Lungs, Mesenteric lymph node, Ovaries, Oviducts, Pancreas, Salivary glands and Regional lymph nodes, Skin, Spinal cord, Spleen, Stomach, Tail, Thymus, Thyroids, Trachea, Uterus, and Vagina. These data are consistent with published data showing that unacylated ghrelin and AZP-531 are safe in rodents and humans. The following sentence was added in subsection “Unacylated ghrelin and cyclic analog AZP-531 inhibits tumor growth in xenograft models and patient-derived tumor cells” “Treatment with unacylated ghrelin had no detrimental effect on weight and no abnormalities associated with treatment were observed with histopathology assessment of a subset of mice.”

- The use of AZP-531 is justified because of improved chemical properties, but the assumption is that all the previous pathway work (on cAMP, MAPK etc) are valid for AZP-531. Is this the case or is there any evidence to support the conclusions that the mechanistic insight derived from unacylated ghrelin is directly applicable to the mode of action for AZP-531? it would be useful if even just one or two of the key functional endpoints explored with unacylated ghrelin, were validated with AZP-531.

This is a very important point raised by the reviewer. Ongoing studies in the laboratory in relation to characterization of AZP-531 mechanism of action demonstrate that similar to unacylated ghrelin (UAG), AZP-531 can suppress cAMP production from forskolin-stimulated MCF7 cells. We have included these new data (right) as Figure 6D in the manuscript. We have also added the following to subsection “Unacylated ghrelin analog, AZP-531, inhibits breast cancer cell growth in vitro, ex vivo and in vivo” “In order to confirm a similar mechanism of action, the effects of unacylated ghrelin and AZP-531 on cAMP levels within MCF7 cells were compared (Figure 6D). Both unacylated ghrelin and AZP-531 caused a dose-dependent inhibition of the forskolin-mediated production of cAMP at 100 and 1000 pM, with no significant differences observed when both treatments were compared.”

- The start of the Results section was confusing to a non-expert. The text states "The dependence on 3D culture was examined for unacylated ghrelin, where treatment with 100 pM resulted in suppression of cell growth in matrigel and collagen, but not in 2D (Figure 1—figure supplement 1A and 1B). At this dose, no effects on MDA-MB-231 cell growth were observed in either 2D or 3D cultures". It's not clear what cell model the first sentence refers to. I assumed it was the MDA-MB-231 cells, but the first line suggests there is growth inhibition in 3D, but the second sentence suggested there isn't growth inhibition in this cell line model.

We apologize for the lack of clarity. In subsection “Unacylated ghrelin inhibits the 3D growth of breast cancer cells”, we have replaced “The dependence on 3D culture was examined for unacylated ghrelin, where treatment with 100 pM resulted in suppression of cell growth in matrigel and collagen, but not in 2D (Figure 1—figure supplement 1A and 1B).” with “The dependence on 3D culture was examined for unacylated ghrelin, where treatment of MCF7 and MDA-MB-468 cells with 100 pM resulted in suppression of cell growth in matrigel and collagen, but not in 2D (Figure 1—figure supplement 1A and 1B).”

https://doi.org/10.7554/eLife.56913.sa2

Article and author information

Author details

  1. CheukMan C Au

    1. Department of Medicine, Weill Cornell Medicine, New York, United States
    2. Centre for Cancer Research, Hudson Institute for Medical Research, Clayton, Australia
    3. Department of Molecular and Translational Sciences, Monash University, Clayton, Australia
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Validation, Investigation, Methodology, Writing - original draft, Writing - review and editing
    Competing interests
    No competing interests declared

  2. John B Furness

    Department of Anatomy and Neuroscience, University of Melbourne, Parkville, Australia
    Contribution
    Supervision
    Competing interests
    No competing interests declared

  3. Kara Britt

    1. Peter MacCallum Cancer Centre, Melbourne, Australia
    2. Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, Australia
    Contribution
    Investigation, Methodology
    Competing interests
    No competing interests declared

  4. Sofya Oshchepkova

    Department of Medicine, Weill Cornell Medicine, New York, United States
    Contribution
    Investigation
    Competing interests
    No competing interests declared

  5. Heta Ladumor

    1. Department of Medicine, Weill Cornell Medicine, New York, United States
    2. Weill Cornell Medicine - Qatar, Doha, Qatar
    Contribution
    Investigation
    Competing interests
    No competing interests declared

  6. Kai Ying Soo

    Centre for Cancer Research, Hudson Institute for Medical Research, Clayton, Australia
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared

  7. Brid Callaghan

    Department of Anatomy and Neuroscience, University of Melbourne, Parkville, Australia
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared

  8. Celine Gerard

    Centre for Cancer Research, Hudson Institute for Medical Research, Clayton, Australia
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared

  9. Giorgio Inghirami

    Department of Pathology, Weill Cornell Medical College, New York, United States
    Contribution
    Resources
    Competing interests
    No competing interests declared

  10. Vivek Mittal

    Department of Cardiothoracic Surgery, Department of Cell and Developmental Biology, Neuberger Berman Lung Cancer Center, Weill Cornell Medicine, New York, United States
    Contribution
    Resources
    Competing interests
    No competing interests declared

  11. Yufeng Wang

    Department of Physiology and Biophysics, Weill Cornell Medical College of Cornell University, New York, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared

  12. Xin Yun Huang

    Department of Physiology and Biophysics, Weill Cornell Medical College of Cornell University, New York, United States
    Contribution
    Supervision
    Competing interests
    No competing interests declared

  13. Jason A Spector

    Department of Surgery, Weill Cornell Medicine, New York, United States
    Contribution
    Resources, Methodology
    Competing interests
    No competing interests declared

  14. Eleni Andreopoulou

    Department of Medicine, Weill Cornell Medicine, New York, United States
    Contribution
    Resources
    Competing interests
    No competing interests declared

  15. Paul Zumbo

    1. Department of Physiology and Biophysics, Weill Cornell Medical College of Cornell University, New York, United States
    2. Applied Bioinformatics Core, Weill Cornell Medical College, New York, United States
    Contribution
    Formal analysis, Visualization
    Competing interests
    No competing interests declared

  16. Doron Betel

    1. Department of Medicine, Weill Cornell Medicine, New York, United States
    2. Institute for Computational Biomedicine, Weill Cornell Medical College, New York, United States
    Contribution
    Formal analysis, Supervision
    Competing interests
    No competing interests declared

  17. Lukas Dow

    Department of Medicine, Weill Cornell Medicine, New York, United States
    Contribution
    Methodology
    Competing interests
    No competing interests declared

  18. Kristy A Brown

    1. Department of Medicine, Weill Cornell Medicine, New York, United States
    2. Centre for Cancer Research, Hudson Institute for Medical Research, Clayton, Australia
    3. Department of Molecular and Translational Sciences, Monash University, Clayton, Australia
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Methodology, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    kab2060@med.cornell.edu
    Competing interests
    Cornell University has filed a patent application on the work that is described in this paper with Dr Brown as a named inventor.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-3382-5546

Funding

National Breast Cancer Foundation (NC-14-011)

  • Kristy A Brown

State Government of Victoria

  • Kristy A Brown

National Health and Medical Research Council (GNT1007714)

  • Kristy A Brown

The Endocrine Society of Australia

  • Kristy A Brown

National Breast Cancer Foundation (ECF-16-004)

  • Kristy A Brown

National Breast Cancer Foundation

  • Kara Britt

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank R Clarke, J Yun, B Vogelstein, M Brown for providing cell lines. We are grateful to D Gough (BRAF transient transfections), J Albeck (EKAR and AKT activity assay) and M Foronda (CRISPR) for provision of reagents and valuable experimental advice. We acknowledge S Yomtoubian for providing information relating to TNBC subtype and T Pham for assistance with data analysis. We are indebted J Cain, SN Jayasekara and M Docanto for assistance with in vivo work; the Monash Micro Imaging Facility, Monash Animal Research Platform, MHTP Medical Genomics Facility, MHTP Flow Cytometry Facility and Research Animal Resource Center at Weill Cornell Medicine for valuable discussions and comments.

Ethics

Human subjects: Patient-derived breast tumors (PDX (X. Zhang et al., 2013)) were obtained from Dr. Giorgio Inghirami and estrogen receptor positive (ER+) breast tumors were obtained from Dr. Eleni Andreopoulou at Weill Cornell Medicine under IRB-approved protocols (WCM 1410015560 and 1603017108). All patients provided written informed consent.

Animal experimentation: All animal experiments were performed in Monash Animal Research Platform at Hudson Institute for Medical Research and Research Animal Resource Center at Weill Cornell Medicine. Experimental procedures were in accordance with guidelines for Animal Care and Use, approved by Monash University Animal Ethics Committee (MMCA2014/20) and Weill Cornell Medicine IACUC Protocol (REQ00016929).

Senior Editor

  1. Richard M White, Memorial Sloan Kettering Cancer Center, United States

Reviewing Editor

  1. Yi Arial Zeng, Chinese Academy of Sciences, China

Reviewers

  1. Yi Arial Zeng, Chinese Academy of Sciences, China
  2. Jason Carroll, Cancer Research UK, United Kingdom
  3. Jeremy Borniger, Cold Spring Harbor Laboratory, United States

Publication history

  1. Received: March 13, 2020
  2. Accepted: June 25, 2020
  3. Version of Record published: July 15, 2020 (version 1)

Copyright

© 2020, Au et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,507
    Page views
  • 137
    Downloads
  • 4
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. CheukMan C Au
  2. John B Furness
  3. Kara Britt
  4. Sofya Oshchepkova
  5. Heta Ladumor
  6. Kai Ying Soo
  7. Brid Callaghan
  8. Celine Gerard
  9. Giorgio Inghirami
  10. Vivek Mittal
  11. Yufeng Wang
  12. Xin Yun Huang
  13. Jason A Spector
  14. Eleni Andreopoulou
  15. Paul Zumbo
  16. Doron Betel
  17. Lukas Dow
  18. Kristy A Brown
(2020)
Three-dimensional growth of breast cancer cells potentiates the anti-tumor effects of unacylated ghrelin and AZP-531
eLife 9:e56913.
https://doi.org/10.7554/eLife.56913

Categories and tags

Research organisms

Further reading

    1. Cancer Biology
    2. Computational and Systems Biology

    Comprehensive characterization of tumor microenvironment in colorectal cancer via molecular analysis

    Xiangkun Wu, Hong Yan ... Li Liang
    Research Article

    Colorectal cancer (CRC) remains a challenging and deadly disease with high tumor microenvironment (TME) heterogeneity. Using an integrative multi-omics analysis and artificial intelligence-enabled spatial analysis of whole-slide images, we performed a comprehensive characterization of TME in colorectal cancer (CCCRC). CRC samples were classified into four CCCRC subtypes with distinct TME features, namely, C1 as the proliferative subtype with low immunogenicity; C2 as the immunosuppressed subtype with the terminally exhausted immune characteristics; C3 as the immune-excluded subtype with the distinct upregulation of stromal components and a lack of T cell infiltration in the tumor core; and C4 as the immunomodulatory subtype with the remarkable upregulation of anti-tumor immune components. The four CCCRC subtypes had distinct histopathologic and molecular characteristics, therapeutic efficacy, and prognosis. We found that the C1 subtype may be suitable for chemotherapy and cetuximab, the C2 subtype may benefit from a combination of chemotherapy and bevacizumab, the C3 subtype has increased sensitivity to the WNT pathway inhibitor WIKI4, and the C4 subtype is a potential candidate for immune checkpoint blockade treatment. Importantly, we established a simple gene classifier for accurate identification of each CCCRC subtype. Collectively our integrative analysis ultimately established a holistic framework to thoroughly dissect the TME of CRC, and the CCCRC classification system with high biological interpretability may contribute to biomarker discovery and future clinical trial design.

    1. Cancer Biology

    Defining function of wild-type and three patient specific TP53 mutations in a zebrafish model of embryonal rhabdomyosarcoma

    Jiangfei Chen, Kunal Baxi ... Myron S Ignatius
    Research Article

    In embryonal rhabdomyosarcoma (ERMS) and generally in sarcomas, the role of wild-type and loss or gain-of-function TP53 mutations remains largely undefined. Eliminating mutant or restoring wild-type p53 is challenging; nevertheless, understanding p53 variant effects on tumorigenesis remains central to realizing better treatment outcomes. In ERMS, >70% of patients retain wild-type TP53, yet mutations when present are associated with worse prognosis. Employing a kRASG12D-driven ERMS tumor model and tp53 null (tp53-/-) zebrafish, we define wild-type and patient-specific TP53 mutant effects on tumorigenesis. We demonstrate that tp53 is a major suppressor of tumorigenesis, where tp53 loss expands tumor initiation from <35% to >97% of animals. Characterizing three patient-specific alleles reveals that TP53C176F partially retains wild-type p53 apoptotic activity that can be exploited, whereas TP53P153D and TP53Y220C encode two structurally related proteins with gain-of-function effects that predispose to head musculature ERMS. TP53P153D unexpectedly also predisposes to hedgehog expressing medulloblastomas in the kRASG12D-driven ERMS-model.

    1. Cancer Biology

    Proteomic characteristics reveal the signatures and the risks of T1 colorectal cancer metastasis to lymph nodes

    Aojia Zhuang, Aobo Zhuang ... Chen Ding
    Research Article Updated

    The presence of lymph node metastasis (LNM) affects treatment strategy decisions in T1NxM0 colorectal cancer (CRC), but the currently used clinicopathological-based risk stratification cannot predict LNM accurately. In this study, we detected proteins in formalin-fixed paraffin-embedded (FFPE) tumor samples from 143 LNM-negative and 78 LNM-positive patients with T1 CRC and revealed changes in molecular and biological pathways by label-free liquid chromatography tandem mass spectrometry (LC-MS/MS) and established classifiers for predicting LNM in T1 CRC. An effective 55-proteins prediction model was built by machine learning and validated in a training cohort (N=132) and two validation cohorts (VC1, N=42; VC2, N=47), achieved an impressive AUC of 1.00 in the training cohort, 0.96 in VC1 and 0.93 in VC2, respectively. We further built a simplified classifier with nine proteins, and achieved an AUC of 0.824. The simplified classifier was performed excellently in two external validation cohorts. The expression patterns of 13 proteins were confirmed by immunohistochemistry, and the IHC score of five proteins was used to build an IHC predict model with an AUC of 0.825. RHOT2 silence significantly enhanced migration and invasion of colon cancer cells. Our study explored the mechanism of metastasis in T1 CRC and can be used to facilitate the individualized prediction of LNM in patients with T1 CRC, which may provide a guidance for clinical practice in T1 CRC.