Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Oct 6;538(7623):109-113.
doi: 10.1038/nature19788. Epub 2016 Sep 28.

The lipolysis pathway sustains normal and transformed stem cells in adult Drosophila

Affiliations

The lipolysis pathway sustains normal and transformed stem cells in adult Drosophila

Shree Ram Singh et al. Nature. .

Abstract

Cancer stem cells (CSCs) may be responsible for tumour dormancy, relapse and the eventual death of most cancer patients. In addition, these cells are usually resistant to cytotoxic conditions. However, very little is known about the biology behind this resistance to therapeutics. Here we investigated stem-cell death in the digestive system of adult Drosophila melanogaster. We found that knockdown of the coat protein complex I (COPI)-Arf79F (also known as Arf1) complex selectively killed normal and transformed stem cells through necrosis, by attenuating the lipolysis pathway, but spared differentiated cells. The dying stem cells were engulfed by neighbouring differentiated cells through a draper-myoblast city-Rac1-basket (also known as JNK)-dependent autophagy pathway. Furthermore, Arf1 inhibitors reduced CSCs in human cancer cell lines. Thus, normal or cancer stem cells may rely primarily on lipid reserves for energy, in such a way that blocking lipolysis starves them to death. This finding may lead to new therapies that could help to eliminate CSCs in human cancers.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing financial interests. Readers are welcome to comment on the online version of the paper.

Figures

Extended Data Figure 1 |
Extended Data Figure 1 |. Stem cells are resistant to apoptosis.
a, Stem cells in the adult Drosophila digestive system. In this system, three organs, the posterior midgut, the hindgut and the Malpighian tubules, meet and join at the junction of the posterior midgut and hindgut. Stem cells in these organs exhibit different degrees of quiescence. The intestinal stem cells (ISCs), located in the posterior midgut, divide once every 24 h,; the renal and nephric stem cells (RNSCs), located in the Malpighian tubules, divide about once a week; and the quiescent hindgut intestinal stem cells (HISCs), found at the midgut–hindgut junction, divide only during stress-induced tissue repair,. GaSCs are gastric stem cells at the foregut–midgut junction. GSSCs are gastric stem cells in the middle of the midgut. The colours just make the cell types or organs more visible and do not exactly reflect different regions in the digestive system. b–n Stem cells are resistant to apoptosis. b, NP1ts > rpr, 18 °C, 24 h (n = 37). c, NP1ts > rpr, 29 °C, 24 h (n = 29). d, e, esgts > lacZRNAi, 29 °C, 7 d (n = 32). f, g, esgts > rpr, 29 °C, 7 d (n = 35). h, wgts > lacZRNAi, 29°C, 7 d (n = 27). i, wgts > rpr, 29°C, 7 d (n = 24). j, NP1ts > p53, 29 °C, 5 d (n = 31). k, esgts > p53, 29 °C, 7 d (n = 38). l, Quantification of Apoptag+ cells in the indicated panels. m, Quantification of GFP+ cells in the indicated panels. n, Quantification of Pros+ cells in the indicated panels. Data are represented as mean ± s.d.. Statistical significance determined by Student’s t-test, ***P < 0.0001. NS, not significant (P > 0.05). As reported previously, 24-h induction of rpr in enterocytes resulted in widespread apoptosis (compare c with b and see the quantitative comparison in l). The induction of rpr by esg-Gal4 (f, g) or wg-Gal4 (i) had little effect on the progenitor or stem cells (that is, enteroblasts, ISCs, RNSCs and HISCs) at one week, compared to wild-type controls (compare f, g with d, e; i with h, and see the quantitative comparison in m). We also found that the overexpression of Drosophila p53 could effectively ablate the enterocytes in five days (compare j with b and see the quantitative comparison in l) but had little effect on stem cells even after one week, compared to controls (compare k with e and see the quantitative comparison in m). Because NP1–Gal4 and esg–Gal4 are not expressed in enteroendocrine cells, as expected, we did not find significant changes in enteroendocrine cells in these experiments (n). o–u, Activation of proliferation accelerates apoptotic cell death of hyperplastic stem cells but fails to completely eliminate neoplastic stem cells. o, Quantification of Apoptag+ cells in the indicated panels. Data are represented as mean ± s.d. Statistical significance was determined by Student’s t-test, ***P < 0.0001. p, esgts > upd, 29 °C, 4 d (n = 28). q, esgts > upd + rpr, 29°C, 4 d (n = 33). r, s, esgts > NDN, 29 °C, 7 d (n = 25). t, u, esgts > NDN + rpr, 29 °C, 7 d (n = 32). White arrows point to the hindgut–midgut junction in h, i, p, q; yellow arrows point to Pros+ enteroendocrine cells in r and t; green arrows point to Dl+ ISCs in r and t. White dotted lines outline GFP+ stem cell clusters in r and t. Yellow dotted lines outline enteroendocrine cell clusters in r and t. Expression of rpr or Arf79FRNAi in ISCs did not kill differentiated cells. The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies and analysed by confocal microscopy. Scale bars in b–k and p–u, 10 μm.
Extended Data Figure 2 |
Extended Data Figure 2 |. The COPI–Arf1 complex regulates stem but not differentiated cell survival.
a, The COPI–lipolysis–β-oxidation pathway. The COPI–Arf1 complex controls lipid homeostasis by regulating adipocyte differentiation-related protein (ADRP), tailinteracting protein of 47kDa (Tip47) and adipocyte triglyceride lipase (ATGL). Triglycerides (TG), diglyceride (DG), fatty acid (FA), Acyl-CoA synthetase (ACS). b–j, The COPI–Arf1 complex regulates stem cell survival. The genotypes of the flies in each panel were: b, esgts wgts > lacZRNAi, 29 °C, 7 d (n = 38). c, esgts wgts > β-CopRNAi, 29 °C, 7 d (n = 23). d, esgts wgts > δ-CopRNAi, 29 °C, 7 d (n = 32). e, esgts wgts > γ-CopRNAi, 29 °C, 7 d (n = 27). f, esgts wgts > ζ-CopRNAi, 29 °C, 7 d (n = 31). g, esgts wgts > β’-CopRNAi, 29 °C, 7 d (n = 29). h, esgts wgts > garzRNAi, 29 °C, 7 d (n = 27). i, esgts wgts > AcslRNAi, 29 °C, 7 d (n = 32). j, esgts wgts >bgmRNAi, 29 °C, 7 d (n = 25). k, Quantification of GFP+ cells in the indicated panels. Data show mean ± s.e.m. Statistical significance was determined by Student’s t-test, ***P < 0.0001. NS, not significant (P >0.05). l–q Knockdown of the COPI–Arf79F complex did not kill differentiated cells. The genotypes of the flies in each panel were: l, n, NP1ts > lacZRNAi, 29 °C, 7 d (n = 25). m, o, NP1ts > Arf79FRNAi, 29 °C, 7 d (n = 32). p, tshts >lacZRNAi, 29 °C, 7 d (n = 22). q, tshts > Arf79FRNAi, 29 °C, 7 d (n = 25). r, esgts wgts > RasV12 + ζ-CopRNAi, 29 °C, 7 d (n = 27). s, esgts wgts > NDN + δ-CopRNAi, 29 °C, 7 d (n = 30). t, u, esgts > NDN + Arf79FRNAi, 29 °C, 7 d (n = 40). v, esgts wgts > RasV12 + AcslRNAi, 29 °C, 7 d. (w) esgts wgts > β-CopRNAi + Hnf4, 29 °C, 7 d. x, esgts wgts > ζ-CopRNAi + scu, 29 °C, 7 d. White arrows in b–j and r, s, w, x, point to the hindgut–midgut junction. Yellow arrows point to Pros+ enteroendocrine cells in t; green arrows point to Dl+ ISCs in t, and a white arrow points to a remaining GFP+ stem cell in u. Scale bars in b–j and l–x, 10 μm.
Extended Data Figure 3 |
Extended Data Figure 3 |. The COPI–Arf79F complex regulates stem cell survival through lipolysis and knockdown of these genes blocks lipolysis, but promotes lipid storage.
a–h, The COPI complex autonomously regulates stem cell survival. The three- or four-day-old adult female flies were heat-shocked twice with an interval of 8–12 h, at 37 °C, for 60 min to induce MARCM clones. In wild-type clones, small GFP+ cell clusters were detected 2 d ACI (a, h; n = 33), which grew into large clusters that contained both ISCs and their differentiated progenies by 7 d ACI (b, h; n = 37). In the δ-COP mutant clones, only a few GFP+ cells were identified 2 d ACI (c, h; n = 34), and none were seen at 7 d ACI (f, h; n = 31). Similarly, only a few GFP+ cells were identified at 2 d ACI in γ-COP (e, h; n = 27) mutant clones, and none were seen at 7 d ACI (f; h; n = 34). Expressing UAS–γ-COP–GFP in γ-COP-mutant MARCM clones (g and h; n = 31) completely rescued the stem cell death phenotype. These results suggest that the COPI complex cell-autonomously regulates stem cell survival. Dotted lines in a and b outline GFP+ clones. White arrows in c and e point to individual GFP+ cells. h, Quantification of GFP+ cells in the indicated panels. Data show the mean ± s.e.m. Statistical significance was determined by Student’s t-test, ***P < 0.0001. The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies and analysed by confocal microscopy. i–p, The lipolysis pathway is active in stem cells. To further investigate the function of lipolysis in stem cells, we investigated the expression of a lipolysis reporter (GAL4–dHFN4; UAS–nlacZ). In our system, this reporter showed strong β-galactosidase expression in mira-GFP-positive ISCs and RNSCs (i–k, n = 15), but not in enterocytes, enteroendocrine cells, and the quiescent HISCs of mature adult flies (i, white arrows, 3–5 days old) or in the quiescent ISCs of freshly emerged young adult flies (less than 3 days old; l and m, n = 17) at 29 °C culture conditions. Expressing δ-COPRNAi (esgts > δ-COPRNAi + GAL4–dHFN4; UAS–nlacZ) almost completely eliminated the reporter expression (n, n = 24), suggesting that the reporter is specifically regulated by the COPI complex. We also expressed a constitutively active form of JAK (hopTum-l) with GAL4–dHFN4; UAS–nlacZ and found that the reporter was expressed in hopTum-l-activated HISCs (o, white arrows, n = 20). The GAL4 in the reporter system is under the control of an hsp70 promoter; we heat-shocked the flies for 30 min at 37 °C 12 h before dissection and found that the reporter was strongly expressed in ISCs, RNSCs and HISCs (particularly strong in HISCs), but not in enteroendocrine cells and enterocytes (p, white arrows, n = 17). Arrows in i, n, o and p point to HISCs at the hindgut–midgut junction. q–v, Arf79F knockdown promotes lipid storage in stem cells. The genotypes of the flies in each panel were: q–s, esgts > lacZRNAi, 29 °C, 4 d (n = 30). t–v, esgts > Arf79FRNAi, 29 °C, 4 d (n = 37). The posterior midguts of flies with the indicated genotypes were dissected, stained with Oil Red O (red), anti-GFP (green) and DAPI (blue), and analysed by confocal microscopy. Dotted lines outline stem cells and white arrows point to lipid droplets in stem cells. Scale bars in a–g and i–v, 10 μm.
Extended Data Figure 4 |
Extended Data Figure 4 |. The lipolysis-β-oxidation pathway regulates survival of transformed stem cells.
a–l, Arft inhibitors kill RasV12-transformed RNSCs through the ROS–Rac–JNK pathway. The GFP-labelled RNSC tumour clusters were induced by expressing RasV in RNSC clones, using the positively marked mosaic lineage (PMML) labelling technique, in adult Drosophila. The flies with RasV12-PMML clones were cultured for 4 d at room temperature on normal food to let the tumour grow and then switched to food with indicated drugs for another 4 d. Flies with RasV12-tumours were given normal food with DMSO (a), 50 ng ml−1 BFA (b), 5 μM GCA (c), 50 μM LM11 (d), 100 μM LG8 (e), 50 μM secin H3 (f), 50 μM secin H3 + 50 μM JNK inhibitor Sp600125 (g), 50 μM secin H3 + 100 μM Rac1 inhibitor (h) or 50 μM secin H3 + 10 mM NAC (i). j,k, esgts flies were fed with normal food with either DMSO (j, n = 20) or 50 μM secin H3 (k, n = 22). n = number of tissues observed. l, Quantification analysis of tumour sizes in Malpighian tubules of indicated panels. We classified all tumours into four categories based on the total number of GFP+ cells in each tumour clone (<10 cells, 10–20 cells, 20–50 cells and 50–100 cells). Total number of tumours investigated for each treatment: DMSO (466 tumours, n = 27 Malpighian tubules), BFA (63 tumours, n = 30 Malpighian tubules), GCA (73 tumours, n = 32 Malpighian tubules), LM11 (94 tumours, n = 35 Malpighian tubules), LG8 (86 tumours, n = 27 Malpighian tubules), secin H3 (64 tumours, n = 25 Malpighian tubules), secin H3 + JNK inhibitor (220 tumours, n = 30 Malpighian tubules), Secin H3 + Rac1 inhibitor (211 tumours, n = 27 Malpighian tubules), and Secin H3 + NAC (297 tumours, n = 35 Malpighian tubules). Arrows point to GFP+ RNSC tumour clusters in a–i. m–r, The lipolysis pathway regulates survival of transformed stem cells. The genotypes of the flies in each panel were: m, m′, esgts > NDN, 29 °C, 4 d (m, n = 30; m′, n = 35). n, n′, esgts > NDN + Hnf4, 29 °C, 4 d (n, n = 25; n′, n = 27). o, o ′, esgts > RasV12, 29 °C, 4 d (o, n = 25; o′, n = 32). p, p′, esgts > RasV12 + Hnf4, 29 °C, 4 d (p, n = 25; p′, n = 32). q, q′, esgts > RasV12 + scu, 29 °C, 4 d (q, n = 23; q′, n = 30). The flies were fed with normal food with either DMSO (m–q) or BFA (m′ and n′, 200 ng ml−1; o′–q′, 50 ng ml−1) for 4 d. Expressing Hnf4 or scu partially blocked the effect of BFA on transformed stem cells. r, Quantification of esg > GFP+ tumour cells in 5 × 103 μm2 per treatment in indicated panels. Data show the mean ± s.e.m. Statistical significance was determined by Student’s t-test, ***P < 0.0001. NS, not significant (P > 0.05). Arrows point to GFP+ RNSC tumour clusters in o–q′. s–y, FAO inhibitors, but not 2-DG, kill RasV12-transformed RNSCs. The GFP-labelled RNSC tumour clusters were induced by expressing RasV12 in RNSC clones using the PMML technique in adult Drosophila. The flies with RasV12-PMML clones were cultured for 4 d at room temperature on normal food to let the tumour grow and then switched to food with indicated drugs for another 4 d. Flies with RasV12-tumours were given normal food with DMSO (s), 5 μM triacsin C (t), 100 μM mildronate (u, n = 27), 100 μM etomoxir (v), 100 μM enoximone (w, n = 37) or 50 mM 2-deoxyglucose (2-DG) (x, n = 32). y, Quantification analysis of tumour sizes in Malpighian tubules of indicated panels. Total number of tumours investigated for each treatment: DMSO (474 tumours, n = 30 Malpighian tubules), triacsin C (47 tumours, n = 32 Malpighian tubules), mildronate (69 tumours, n = 27 Malpighian tubules), etomoxir (73 tumours, n = 35 Malpighian tubules), enoximone (86 tumours, n = 27 Malpighian tubules) and 2-DG (264 tumours, n = 32 Malpighian tubules). Arrows point to GFP+ RNSC tumour clusters. The gut of flies with the indicated genotypes was dissected after cultured, stained with the indicated antibodies and analysed by confocal microscopy. Scale bars in a–k, m–q’ and s–x, 10 μm.
Extended Data Figure 5 |
Extended Data Figure 5 |. Knockdown of components of the COPI–Arf79F–Acsl pathway kill normal and transformed stem cells through necrosis.
a–i, The genotypes of the flies in each panel were: a–c and f–i, esgts > Arf79FRNAi, 29 °C, 7 d (n = 27). d, e, esgts > lacZRNAi, 29 °C, 7 d (n = 20). In d–i, a dye (CellMask) marks plasma membranes. In a-c and g-i, a dying ISC is engulfed by a neighbouring enterocytes. j, esgts wgts > lacZRNAi, 29 °C, 4 d (n = 30). k, esgts wgts > ζ-copRNAi, 29 °C, 4 d (n = 36). l, esgts wgts > β-CopRNAi > 29 °C, 4 d (n = 34). m, esgts wgts > garzRNAi, 29 °C, 4 d (n = 32). n, o, esgts wgts > AcslRNAi, 29 °C, 4 d (n = 32). p, Quantification of propidium-iodide-positive cells in the indicated panels. Data show the mean ± s.d. Statistical significance was determined by Student’s t-test, ***P < 0.0001 (compared to control). Yellow arrows point to hindgut–midgut junctions in j and n, white arrows point to GFP- and propidium-iodide-positive stem cells in k and o. The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies or reagents and analysed by confocal microscopy. Scale bars in a–o: 10 μm.
Extended Data Figure 6 |
Extended Data Figure 6 |. Arf79F knockdown kills transformed and normal stem cells through necrosis.
(a–d) The genotypes of the flies in each panel were: a, esgts > NDN, 29 °C, 4 d (n = 25). b, c, esgts > NDN + Arf79FRNAi, 29°C, 4 d (n = 25). d, Quantification of PI+ cells in the indicated panels. Data show the mean ± s.d. Statistical significance was determined by Student’s t-test, ***P < 0.0001. NS, not significant (P > 0.05). White arrows point to GFP- and propidiumiodide-positive stem cells in c. e–l, The genotypes of the flies in each panel were: e, f, i, j, esgts > lacZRNAi, 29 °C, 7 d (e, f, n = 27; i, j, n = 30). g, h, k, l, esgts > Arf79FRNAi, 29 °C,7 d (g, h, n = 45; k, l, n = 30). White arrows point to GFP+ stem cells. No DHE or LysoTracker signals were detected in the wild-type midgut, but these signals were intense in the esgts > Arf79FRNAi flies, indicating that the dying ISCs had accumulated ROS and were intracellularly acidified. The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies or dyes and analysed by confocal microscopy. Scale bars in a–c and e–l: 10 μm.
Extended Data Figure 7 |
Extended Data Figure 7 |. Overexpressing Cat rescues the ISC death induced by Arf79FRNAi but not CycTRNAi expression and dying ISCs activate JNK signalling and autophagy in enterocytes.
a–h, The genotypes of the flies in each panel were: a, esgts > Cat, 29 °C, 7 d (n = 25). b, esgts > Arf79FRNAi + Cat, 29 °C, 7 d (n = 32). c, esgts > Sod, 29 °C, 7 d (n = 24). d, esgts > Arf79FRNAi + Sod, 29 °C, 7 d (n = 30). e, esgts > Sod2, 29 °C, 7 d (n = 22). f, esgts > Arf79FRNAi + Sod2, 29 °C, 7 d (n = 32). g, esgts > CycTRNAi, 29 °C, 7 d (n = 35). h, esgts > CycTRNAi + Cat, 29 °C, 7 d (n = 37). Overexpressing Cat, but not sod or sod2, in stem cells (esgts > Arf79FRNAi + Cat) rescued the stem-cell death induced by Arf79F knockdown but not that induced by CycT knockdown (esgts>CycTRNAi+Cat). i–n, The genotypes of the flies in each panel were: i, puc–lacZ, 29 °C, 7 d (n = 17). j, esgts > Arf79FRNAi + puc–lacZ, 29 °C, 7 d (n = 20). Yellow arrows point to GFP+ cells. k, esg > GFP, 29 °C, 4 d (n = 12). l, FRT82B–γ-COP10 MARCM clones, 4 d (n = 17). Yellow arrows point to GFP+ clones. m, esgts > lacZRNAi + pmCherry–Atg8a, 29 °C, 7 d (n = 22). n, esgts > Arf79FRNAi + pmCherry–Atg8a, 29 °C, 7 d (n = 25). Arf79F knockdown in ISCs induced Puc–lacZ (compare i with j) and Cherry–Atg8a (compare m with n) expression in enterocytes, p-JNK was induced in enterocytes in γ-COP mutant MARCM clones, compare k with l). The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies and analysed by confocal microscopy. Scale bars in a–n: 10 μm.
Extended Data Figure 8 |
Extended Data Figure 8 |. Knockdown of components of the JNK pathway or engulfment genes in ISCs did not block the ISC death induced by Arf79FRNAi or δ-CopRNAi expression.
a–i, The genotypes of the flies in each panel were: a, esgts > Arf79FRNAi, 29 °C, 7 d (n = 27). b, esgts > δ-CopRNAi + Rac1DN, 29 °C, 7 d (n = 32). c, esgts > δ-CopRNAi + mbcRNAi, 29 °C, 7 d (n = 25). d, esgts > Arf79FRNAi + bskDN, 29 °C, 7 d (n = 30). e, esgts > Arf79FRNAi + drprRNAi, 29 °C, 7 d (n = 28). f, esgts > δ-CopRNAi + Atg5RNAi, 29 °C, 7 d (n = 32). g, esgts > Arf79FRNAi + p35, 29 °C, 7 d (n = 22). h, esgts > δ-CopRNAi + PSRRNAi, 29 °C, 7 d (n = 30). i, esgts > Arf79FRNAi + mysRNAi, 29 °C, 7 d (n = 28). bskDN is a dominant-negative form of Drosophila JNK (ref. 51), draper (drpr) encodes a homologue of the C. elegans transmembrane phagocytic receptor (ref. 52), Rac1 encodes a small GTPase that is a homologue of the C. elegans engulfment gene ced-10 (ref. 53), myoblast city (mbc)/Crk/dCed-12 encodes a Rac1 guanine nucleotide exchange factor (GEF) (ref. 53), PSR encodes a phosphatidylserine receptor (ref. 54) and mys encodes the β-subunit of integrin, which is involved in mammalian cell engulfment (ref. 55). Light chain 3 (LC3) in autophagosomes is involved in the rapid degradation of the internalized cargo (reviewed in Han and Ravichandran in ref. 27). j–l, Activation of hep or Rac1 genes in enterocytes induced the ISC death. The genotypes of the flies in each panel were: j, mira–GFP, 29 °C, 7 d (n = 17). k, mira–GFP + NP1ts (–UASGFP) > hepCA (a constitutively activate form of hep), 29 °C, 7 d (n = 15). l, mira–GFP + NP1ts (–UASGFP) > Rac1V12 (a constitutively activate form of Rac1), 29 °C, 3 d (n = 12). m–n, Overexpression of drpr in enterocytes did not induce EC death. m, NP1ts > lacZRNAi, 29 °C, 5 d (n = 15). n, NP1ts > drpr, 29 °C, 5 d (n = 20). The posterior midguts of flies with the indicated genotypes were dissected, stained with the indicated antibodies and analysed by confocal microscopy. Scale bars in a–n: 10 μm.
Extended Data Figure 9 |
Extended Data Figure 9 |. Knockdown of components of the JNK pathway or engulfment genes in enterocytes blocks the ISC death induced by Arf79FRNAi or δ-CopRNAi expression.
The genotypes of the flies in each panel were: a, NP1ts esgts > lacZRNAi, 29 °C, 7 d (n = 20). b, NP1ts esgts > Arf79FRNAi, 29 °C, 7 d (n = 32). c, NP1ts esgts > δ-CopRNAi, 29 °C, 7 d (n = 30). d, NP1ts esgts > Arf79FRNAi + bskDN, 29 °C, 7 d (n = 30). NP1ts esgts > δ-CopRNAi + Rac1DN, 29 °C, 7 d (n = 202). NP1ts esgts > δ-CopRNAi + mbcRNAi, 29 °C, 7 d (n = 18). NP1ts esgts > Arf79FRNAi + drprRNAi, 29 °C, 7 d (n = 32). NP1ts esgts > δ-CopRNAi + Atg5RNAi, 29 °C, 7 d (n = 17). i NP1ts esgts > δ-CopRNAi + Atg12RNAi, 29 °C, 7 d (n = 25). j, NP1ts esgts > δ-CopRNAi + PSRRNAi, 29 °C, 7 d (n = 22). k, NP1ts esgts > Arf79FRNAi + mysRNAi, 29 °C, 7 d (n = 35). NP1ts esgts > Arf79FRNAi + p35, 29 °C, 7 d (n = 27). m, NP1ts > JraAsp (a constitutively activate form of Jun), 29 °C, 7 d (n = 25). n, NP1ts > drpr, 29 °C, 7 d (n = 20). Expresslon of JraAsp and drpr in enterocytes ehmmates Dl+ ISCs. The posterior midguts of files with the indicated genotypes were dissected, stamed with the indicated antibodies and analysed by confocal microscopy. White arrows point to Dl+ ISCs, yellow arrowheads point to Pros+ enteroendocrine cells and green arrows point to enterocytes. Scale bars m a–n: 10 μm.
Extended Data Figure 10 |
Extended Data Figure 10 |. Arf1 and FAO inhibitors suppress CSCs in human cancer cell lines.
(a–f) Arft inhibitors suppress proliferation and sphere formation in DU145 cells. a, b, Crystal violet staining was used to detect cell survival after 2 days of treatment with BFA or GCA at the indicated concentrations in DU145 cells. The growth of DU145 cells was strongly inhibited by 30 ng ml−1 BFA (a, b) and 2.5 μM GCA (b). We also tested the two inhibitors in tumour sphere formation by cancer cells, a widely used in vitro technique for assessing CSC self-renewal capacity. Spheres were cultured with or without BFA or GCA. The two inhibitors also inhibited tumour sphere formation (c, d). GCA was a weak inhibitor of growth (b), but a strong inhibitor of tumour sphere formation (d) in DU145 cells, indicating that these inhibitors may specifically target CSCs. The mRNA level of CD44 and E-Cadherin, two potent prostate cancer tumour-initiating cell markers, were reduced by BFA treatment (e, f). Data show the mean ± s.e.m. Statistical significance was determined by Student’s t-test, *P < 0.05; **P < 0.01 (compared to DMSO). g–l, Arf1 inhibitors suppress proliferation and sphere formation in HT29 and MCF7 cells. Crystal violet staining was used to detect cell survival after 2 days of treatment with BFA or GCA at the indicated concentrations in HT29 and MCF7 cells. The inhibitors reduced the cell survival rate (g–j). Spheres were cultured with or without BFA or GCA. The inhibitors inhibited sphere formation dramatically (k, l). Data show the mean ± s.e.m. Statistical significance was determined by Student’s t-test, * P < 0.05; **P < 0.01 (compared to DMSO). (m, n) BFA and FAO inhibitors reduce CSC in DU145, HT29, MCF7 and MDA-MB-231 cells. m, Flow cytometry was used to detect cancer stem cell surface makers in DU145, HT29 and MDA-MB-231 cells after 2 days of treatment with BFA. Cell subpopulations enriched with cancer stem cells (CD44+ and CD24 for DU145 and MDA-MB-231, CD44+ for HT29) were marked with red line. n, Crystal violet staining was used to detect cell survival after 2 days of treatment with triascin C or etomoxir using indicated concentrations in DU145, HT29 and MCF7 cells (left). Spheres were cultured with or without triascin C or etomoxir. The inhibitors markedly inhibited sphere formation in DU145 and MCF7 cells (right). Data show the mean ± s.e.m. Statistical significance was determined by Student’s t-test, **, P < 0.01 (compared to DMSO).
Figure 1 |
Figure 1 |. Activation of proliferation accelerates apoptotic cell death of hyperplastic stem cells but fails to completely eliminate neoplastic stem cells.
a, Diagram of three types of stem cells near the hindgut–midgut junction, and the cells in which esg–Gal4 (esg) and wg–Gal4 (wg) were expressed. b–i, Representative images of the posterior midguts of flies with the indicated phenotypes. b, esgts > upd, 29 °C, 4 d (n = 33). c, esgts > upd + rpr, 29 °C, 4 d (n = 35). d, esgts > RasV12, 29 °C, 7 d (n = 34). e, esgts > RasV12 + rpr, 29°C, 7 d (n = 29). f, esgts > RasV12 + Arf79FRNAi, 29 °C, 7 d (n = 35). g, esgts > NDN, 29 °C, 7 d (n = 24). h, esgts > NDN + rpr, 29 °C, 7 d (n = 29). i, esgts > NDN + Arf79FRNAi, 29 °C,7 d (n = 38). j, Quantification of GFP+ cells from midguts isolated from flies with the indicated genotypes. Data are represented as mean ± s.e.m. Statistical significance determined by Student’s t-test, ***P < 0.0001. The posterior midguts of flies with the indicated genotypes were dissected, stained with the GFP and Prospero (Pros) antibodies and analysed by confocal microscopy. White arrows in b and c point to the hindgut–midgut junction. g, h, Red arrows with white dotted lines point to clusters of ISCs enteroblasts and yellow arrows with yellow dotted lines point to clusters of enteroendocrine cells. i, Red and yellow arrows point to remaining ISCs/enteroblasts, and enteroendocrine cells, respectively. Scale bars in b–i, 10 μm.
Figure 2 |
Figure 2 |. The COPI–Arf79F complex regulates stem cell survival through a lipolysis pathway.
a–g, Representative images are shown. The genotypes of the flies in each panel were: a, esgts wgts > lacZRNAi, 29 °C, 7 d (n = 34). b, esgts wgts > Arf79FRNAi, 29 °C, 7 d (n = 38). c, esgts wgts > Arf79FRNAi + bmm, 29 °C, 7 d (n = 31). d, esgts wgts > Arf79FRNAi + Hnf4, 29 °C, 7 d (n = 29). e, esgts wgts > Arf79FRNAi + scu, 29 °C, 7 d (n = 32). f, esgts wgts > Arf79FRNAi + p35, 29 °C, 7 d (n = 37). g, esgts wgts > Arf79FRNAi + Atg12RNAi, 29 °C, 7 d (n = 32). h, Quantification of GFP+ cells from midguts isolated from flies with the indicated genotypes. Data are represented as mean ± s.d. Statistical significance determined by Student’s t-test, ***P < 0.0001. NS, not significant (P > 0.05). i–k, MARCM clones of flies with the following genotypes: i, UAS–Hnf4; FRT82B–γ-Cop, 7 d (n = 30) after clonal induction (ACI). j, UAS-scu; FRT82B-γ-Cop, 7 d ACI (n = 32). k, UAS–cat; FRT82B–γ-Cop, 7 d ACI (n = 35). The posterior midguts of flies with the indicated genotypes were dissected, stained with the GFP, Prospero (Pros) and Armadillo (Arm) antibodies and analysed by confocal microscopy. White arrows in a–g point to the hindgut-midgut junction. White dotted lines in i–k outline GFP+ clones. Scale bars in a–g and i–k, 10 μm.
Figure 3 |
Figure 3 |. Knockdown of components of the COPI–Arf79F–β-oxidation pathway kill stem cells through necrosis.
The genotypes of the flies in each panel were: a, b, e, f, esgts wgts > lacZRNAi, 29 °C, 4 d (n = 32). c, d, g, h, esgts wgts > Arf79FRNAi, 29 °C, 4 d (n = 28). i, esgts wgts > δ-CopRNAi, 29 °C, 4 d (n = 30). j, esgts wgts > Arf79FRNAi + scu, 29 °C, 4 d (n = 28). k, esgts wgts > Arf79FRNAi + Hnf4, 29 °C, 4 d (n = 30). l, Quantification of PI+ cells from midguts isolated from flies with the indicated genotypes. Data are represented as mean ± s.d. Statistical significance determined by Student’s t-test, ***P < 0.0001; NS, not significant (P > 0.05). The posterior midguts of flies with the indicated genotypes were dissected, stained with GFP and Armadillo (Arm) antibodies or indicated reagents, and analysed by confocal microscopy. White arrows in a–i point to esg-GFP+ cells and in j–k point to the hindgut–midgut junction. White dotted lines in d outline ISCs/enteroblasts, yellow dotted lines outline enterocytes. Scale bars in a–k, 10 μm.
Figure 4 |
Figure 4 |. Dying ISCs are engulfed by neighbouring enterocytes through the draper–Rac–JNK (Bsk) pathway.
a, Representative images from esgts > Arf79FRNAi cells, 29°C, 7 d (n = 32). A dying ISC is engulfed by a neighbouring enterocyte. The posterior midguts of flies were dissected, stained with the GFP and Armadillo (Arm) antibodies, and analysed by confocal microscopy. Arrow points to GFP+ stem cell/progenitors and the dotted line outlines a nucleus and cell membrane of an enterocyte. Scale bar, 10 μ m. b, Quantification of Dl+ (Delta+) ISCs cells in flies with the NP1ts esgts driver and indicated genotypes (See images in Extended Data Fig. 9 for details). Data are represented as mean ± s.e.m. Statistical significance determined by Student’s t-test, *P < 0.05, ***P < 0.0001. NS, not significant (P > 0.05). c, Model of ISC death induced by knockdown of the COPI–Arf1 complex. Details are described in the text. The autophagosome is involved in the last step of phagocytosis (degradation of internalized cargo), and autophagy may both function downstream of and be regulated by the drpr–Mbc–Rac1–JNK pathway.

References

    1. Kleffel S & Schatton T Tumor dormancy and cancer stem cells: two sides of the same coin? Adv. Exp. Med. Biol 734, 145–179 (2013). - PubMed
    1. Micchelli CA & Perrimon N Evidence that stem cells reside in the adult Drosophila midgut epithelium. Nature 439, 475–479 (2006). - PubMed
    1. Ohlstein B & Spradling A The adult Drosophila posterior midgut is maintained by pluripotent stem cells. Nature 439, 470–474 (2006). - PubMed
    1. Singh SR, Liu W & Hou SX The adult Drosophila malpighian tubules are maintained by multipotent stem cells. Cell Stem Cell 1, 191–203 (2007). - PMC - PubMed
    1. Fox DT & Spradling AC The Drosophila hindgut lacks constitutively active adult stem cells but proliferates in response to tissue damage. Cell Stem Cell 5, 290–297 (2009). - PMC - PubMed

MeSH terms

LinkOut - more resources