Abstract

It is now well established that tumours undergo changes in cellular metabolism1. As this can reveal tumour cell vulnerabilities and because many tumours exhibit enhanced glucose uptake2, we have been interested in how tumour cells respond to different forms of sugar. Here we report that the monosaccharide mannose causes growth retardation in several tumour types in vitro, and enhances cell death in response to major forms of chemotherapy. We then show that these effects also occur in vivo in mice following the oral administration of mannose, without significantly affecting the weight and health of the animals. Mechanistically, mannose is taken up by the same transporter(s) as glucose3 but accumulates as mannose-6-phosphate in cells, and this impairs the further metabolism of glucose in glycolysis, the tricarboxylic acid cycle, the pentose phosphate pathway and glycan synthesis. As a result, the administration of mannose in combination with conventional chemotherapy affects levels of anti-apoptotic proteins of the Bcl-2 family, leading to sensitization to cell death. Finally we show that susceptibility to mannose is dependent on the levels of phosphomannose isomerase (PMI). Cells with low levels of PMI are sensitive to mannose, whereas cells with high levels are resistant, but can be made sensitive by RNA-interference-mediated depletion of the enzyme. In addition, we use tissue microarrays to show that PMI levels also vary greatly between different patients and different tumour types, indicating that PMI levels could be used as a biomarker to direct the successful administration of mannose. We consider that the administration of mannose could be a simple, safe and selective therapy in the treatment of cancer, and could be applicable to multiple tumour types.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Data availability

The data supporting the findings of this study are available within the paper and its Supplementary Information. Source Data for Figs. 1–4 and Extended Data Figs. 1–10 are available with the online version of the paper. Data are available from the corresponding author upon reasonable request.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011).

  2. 2.

    Pavlova, N. N. & Thompson, C. B. The emerging hallmarks of cancer metabolism. Cell Metab. 23, 27–47 (2016).

  3. 3.

    Thorens, B. & Mueckler, M. Glucose transporters in the 21st century. Am. J. Physiol. Endocrinol. Metab. 298, E141–E145 (2010).

  4. 4.

    Cairns, R. A., Harris, I. S. & Mak, T. W. Regulation of cancer cell metabolism. Nat. Rev. Cancer 11, 85–95 (2011).

  5. 5.

    Chaube, B. & Bhat, M. K. AMPK, a key regulator of metabolic/energy homeostasis and mitochondrial biogenesis in cancer cells. Cell Death Dis. 7, e2044 (2016).

  6. 6.

    Zhang, C. S. et al. Fructose-1,6-bisphosphate and aldolase mediate glucose sensing by AMPK. Nature 548, 112–116 (2017).

  7. 7.

    DeRossi, C. et al. Ablation of mouse phosphomannose isomerase (Mpi) causes mannose 6-phosphate accumulation, toxicity, and embryonic lethality. J. Biol. Chem. 281, 5916–5927 (2006).

  8. 8.

    Fischer, U., Jänicke, R. U. & Schulze-Osthoff, K. Many cuts to ruin: a comprehensive update of caspase substrates. Cell Death Differ. 10, 76–100 (2003).

  9. 9.

    Elmore, S. Apoptosis: a review of programmed cell death. Toxicol. Pathol. 35, 495–516 (2007).

  10. 10.

    Westphal, D., Dewson, G., Czabotar, P. E. & Kluck, R. M. Molecular biology of Bax and Bak activation and action. Biochim. Biophys. Acta 1813, 521–531 (2011).

  11. 11.

    Tait, S. W. & Green, D. R. Mitochondria and cell death: outer membrane permeabilization and beyond. Nat. Rev. Mol. Cell Biol. 11, 621–632 (2010).

  12. 12.

    Pradelli, L. A. et al. Glycolysis inhibition sensitizes tumor cells to death receptors-induced apoptosis by AMP kinase activation leading to Mcl-1 block in translation. Oncogene 29, 1641–1652 (2010).

  13. 13.

    Alves, N. L. et al. The Noxa/Mcl-1 axis regulates susceptibility to apoptosis under glucose limitation in dividing T cells. Immunity 24, 703–716 (2006).

  14. 14.

    Coloff, J. L. et al. Akt-dependent glucose metabolism promotes Mcl-1 synthesis to maintain cell survival and resistance to Bcl-2 inhibition. Cancer Res. 71, 5204–5213 (2011).

  15. 15.

    Quirce, R. et al. New insight of functional molecular imaging into the atheroma biology: 18F-NaF and 18F-FDG in symptomatic and asymptomatic carotid plaques after recent CVA. Preliminary results. Clin. Physiol. Funct. Imaging 36, 499–503 (2016).

  16. 16.

    Sharma, V., Ichikawa, M. & Freeze, H. H. Mannose metabolism: more than meets the eye. Biochem. Biophys. Res. Commun. 453, 220–228 (2014).

  17. 17.

    Chattopadhyay, U. & Bhattacharyya, S. Inhibition by monosaccharides of tumor associated macrophages mediated antibody dependent cell cytotoxicity to autologous tumor cells. Neoplasma 34, 295–303 (1987).

  18. 18.

    Gartner, S. L. & Williams, T. J. Modulation of interleukin-1 induced thymocyte proliferation by d-mannose. Thymus 19, 117–126 (1992).

  19. 19.

    Alton, G. et al. Direct utilization of mannose for mammalian glycoprotein biosynthesis. Glycobiology 8, 285–295 (1998).

  20. 20.

    De Robertis, M. et al. The AOM/DSS murine model for the study of colon carcinogenesis: from pathways to diagnosis and therapy studies. J. Carcinog. 10, 9 (2011).

  21. 21.

    Long, J. S. et al. Extracellular adenosine sensing—a metabolic cell death priming mechanism downstream of p53. Mol. Cell 50, 394–406 (2013).

  22. 22.

    Riley, J. S. et al. Mitochondrial inner membrane permeabilisation enables mtDNA release during apoptosis. EMBO J. 37, e99238 (2018).

  23. 23.

    Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).

  24. 24.

    Sakamaki, J. I. et al. Bromodomain protein BRD4 is a transcriptional repressor of autophagy and lysosomal function. Mol. Cell 66, 517–532 (2017).

  25. 25.

    Mrschtik, M. et al. DRAM-3 modulates autophagy and promotes cell survival in the absence of glucose. Cell Death Differ. 22, 1714–1726 (2015).

  26. 26.

    Mackay, G. M., Zeng, L., van den Broek, N. J. F. & Gottlieb, E. Analysis of cell metabolism using LC–MS and isotope tracers. Methods Enzymol. 561, 171–196 (2015).

  27. 27.

    Karvela, M. et al. ATG7 regulates energy metabolism, differentiation and survival of Philadelphia-chromosome-positive cells. Autophagy 12, 936–948 (2016).

  28. 28.

    Rosenfeldt, M. T. et al. E2F1 drives chemotherapeutic drug resistance via ABCG2. Oncogene 33, 4164–4172 (2014).

  29. 29.

    Rosenfeldt, M. T. et al. p53 status determines the role of autophagy in pancreatic tumour development. Nature 504, 296–300 (2013).

  30. 30.

    Cammareri, P. et al. TGFβ pathway limits dedifferentiation following WNT and MAPK pathway activation to suppress intestinal tumourigenesis. Cell Death Differ. 24, 1681–1693 (2017).

Download references

Acknowledgements

We thank J. Knight, D. Lewis and E. Johnson for help and advice. This work was supported by Worldwide Cancer Research (16-1194) and Cancer Research UK (A15816 and A17196).

Reviewer information

Nature thanks H. Christofk, J. Cleveland and A. Villunger for their contribution to the peer review of this work.

Author information

Author notes

    • Darren Ennis
    •  & Iain A. McNeish

    Present address: Department of Surgery and Cancer, Imperial College London, London, UK

  1. These authors contributed equally: Simone Cardaci, Valentin J. A. Barthet

Affiliations

  1. Cancer Research UK Beatson Institute, Glasgow, UK

    • Pablo Sierra Gonzalez
    • , James O’Prey
    • , Simone Cardaci
    • , Valentin J. A. Barthet
    • , Jun-ichi Sakamaki
    • , Florian Beaumatin
    • , David M. Gay
    • , Gillian Mackay
    • , Gaurav Malviya
    • , Elżbieta Kania
    • , Shona Ritchie
    • , Alice D. Baudot
    • , Barbara Zunino
    • , Agata Mrowinska
    • , Colin Nixon
    • , Owen J. Sansom
    •  & Kevin M. Ryan
  2. Division of Genetics and Cell Biology, San Raffaele Scientific Institute, Milan, Italy

    • Simone Cardaci
  3. Institute of Cancer Sciences, University of Glasgow, Glasgow, UK

    • Antonia Roseweir
    • , Darren Ennis
    • , Iain A. McNeish
    • , Owen J. Sansom
    • , Joanne Edwards
    •  & Kevin M. Ryan
  4. Department of Pathology, Queen Elizabeth University Hospital, Glasgow, UK

    • Aoisha Hoyle
    •  & David Millan

Authors

  1. Search for Pablo Sierra Gonzalez in:

  2. Search for James O’Prey in:

  3. Search for Simone Cardaci in:

  4. Search for Valentin J. A. Barthet in:

  5. Search for Jun-ichi Sakamaki in:

  6. Search for Florian Beaumatin in:

  7. Search for Antonia Roseweir in:

  8. Search for David M. Gay in:

  9. Search for Gillian Mackay in:

  10. Search for Gaurav Malviya in:

  11. Search for Elżbieta Kania in:

  12. Search for Shona Ritchie in:

  13. Search for Alice D. Baudot in:

  14. Search for Barbara Zunino in:

  15. Search for Agata Mrowinska in:

  16. Search for Colin Nixon in:

  17. Search for Darren Ennis in:

  18. Search for Aoisha Hoyle in:

  19. Search for David Millan in:

  20. Search for Iain A. McNeish in:

  21. Search for Owen J. Sansom in:

  22. Search for Joanne Edwards in:

  23. Search for Kevin M. Ryan in:

Contributions

K.M.R., P.S.G. and A.D.B. conceived the study. P.S.G., J.O., J.-i.S., F.B., E.K. and S.R. conducted and analysed cell growth, cell death and enzyme assays and western blotting. P.S.G., S.C. and G. Mackay conducted and analysed metabolic experiments. V.J.A.B., D.M.G. and B.Z. performed animal experiments. G. Malviya and A.M. performed and analysed PET and MRI scans. C.N. performed immunohistochemistry. A.R., D.E., A.H. and D.M. generated and analysed tissue microarrays. P.S.G. and K.M.R. wrote the manuscript. I.A.M., O.J.S., J.E. and K.M.R. supervised the study.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Kevin M. Ryan.

Extended data figures and tables

  1. Extended Data Fig. 1 Mannose affects cell growth and metabolism.

    a, Growth curves of Saos-2 cells supplemented without (control) or with 25 mM of hexoses (man, mannose; gal, galactose; fru, fructose; fuc, fucose; glc, glucose). bd, Growth curves of PA-TU-8902 (b) and A549 cells (d) in DMEM alone or with an additional 25 mM mannose; K562 cells in 10% FBS RPMI-1640 medium with or without 11.1 mM mannose (c). e, Western blots showing the levels of phospho-AMPKα (T172) and total AMPKα after 5, 15, 30 and 45-min incubation of U2OS-E1a with standard medium or medium supplemented with 25 mM mannose. f, Growth curves of SKOV3 and RKO cells in DMEM alone or with an additional 25 mM mannose. g, Levels (expressed as peak area per microgram of protein) of 2-deoxyglucose-phosphate (2-DG-P) in RKO, SKOV3, SAOS-2 and U2OS-E1a (U2OS) cells incubated with 10 mM 2-deoxyglucose for 6 h in the presence of 25 mM mannose in the culture medium (DMEM). Data are the average of three technical replicates and are representative of two independent experiments. h, Levels (measured as the percentage of peak area per microgram of protein, on a log10 scale) of hexoses-6-phosphate (hexoses-6P) in U2OS-E1a cells after 6 h incubation in 10% dialysed FBS with 5 mM glucose in DMEM either with or without 5 mM mannose. i, Peak area per microgram of protein of hexoses-6-phosphate in U2OS-E1a cells incubated for 6 h in DMEM, with or without an additional 25 mM of the indicated sugars. j, k, Peak area per microgram of protein of intracellular non-phosphorylated mannose (j) or relative amounts of glucose (k) after a 6 h incubation of U2OS-E1a cells in 5 mM glucose in DMEM, with or without 5 mM mannose. l, Relative amount per microgram of protein of non-phosphorylated glucose m + 2 after 6 h incubation of U2OS-E1a cells in glucose-free DMEM either with 5 mM 1,2-13C2-d-glucose alone or with 5 mM 1,2-13C2-d-glucose and 5 mM 13C6-d-mannose. mp, Peak area per microgram of protein of glyceraldehyde-3-phosphate (GA3P) (m), phosphoenolpyruvate (PEP) (n), lactate (o) and UDP-GlcNAc (p) in U2OS-E1a cells after 6 h incubation in DMEM, with or without an additional 25 mM of the indicated sugars. n = 3 independent experiments (ad, f, hp). Data are representative of two independent experiments (e). All data are mean ± s.e.m. and were analysed by one-way ANOVA (i, mp), two-way ANOVA followed by Tukey's multiple comparisons (a) or paired two-tailed Student’s t-test (h, jl). *P < 0.05, **P < 0.01, ***P < 0.001. Source data

  2. Extended Data Fig. 2 Mannose sensitivity is not associated with mannose uptake but with changes in cellular metabolism.

    a, Levels (expressed as peak area per microgram of proteins) of 2-deoxy-2-fluoro-d-mannose-phosphate (2-DFM-P) in RKO, SKOV3, SAOS-2 and U2OS-E1a (U2OS) cells after 6 h treatment with or without 2 mM 2-deoxy-2 fluoro-d-mannose. bh, Relative levels of hexoses-6-phosphate (b), lactate (c), glyceraldehyde-3-phosphate (glyceraldehyde-3P) (d), phosphoenolpyruvate (e), malate (f), ribose-5-phosphate (ribose-5P) (g) and UDP-GlcNac (h) in SKOV3 and U2OS-E1a (U2OS) cells after 6 h treatment with or without 25 mM mannose in the culture medium (DMEM). Data in a are the average of three technical replicates and are representative of two independent experiments. In bh, n = 2 independent experiments, each involving technical triplicates. All data are mean. Source data

  3. Extended Data Fig. 3 Mannose has a rapid effect on cellular metabolism.

    Metabolite content (expressed as peak area per microgram of proteins) of hexoses-6-phosphate (a), ATP (b), AMP (c), fructose-1,6-bisphosphate (F1,6-BP) (d), ribose-5-phosphate (e) and UDP-GlcNac (f) in U2OS-E1a (U2OS) cells after 5 min treatment with or without 25 mM mannose in the culture medium (DMEM). n = 3 independent experiments, presented as mean ± s.e.m. and analysed by a two-tailed unpaired t-test. *P < 0.05, ***P < 0.001. Source data

  4. Extended Data Fig. 4 Mannose sensitizes cells to chemotherapy-induced cell death via the intrinsic apoptotic pathway.

    a, b, Percentage of propidium-iodide-positive KP-4 cells after 24 h treatment with 40 µM cisplatin (a) or 1 µg ml−1 doxorubicin (b) in the presence or absence of 25 mM mannose. c, Percentage of U2OS-E1a propidium-iodide-positive cells after 24 h treatment with or without 10 µM cisplatin in DMEM with or without 25 mM of the indicated additional sugars. d, Percentage of U2OS-E1a propidium-iodide-positive cells after 24 h treatment with or without 1 µg ml−1 doxorubicin, with or without 25 mM mannose and with or without 50 µM zVAD-FMK. e, f, Fold increase of the percentage of propidium-iodide-positive Saos-2 (NTC, caspase-8 and FADD CRISPR) cells upon 24 h treatment in DMEM with or without 1 µg ml−1 doxorubicin (e) or 10 μM cisplatin (f) and with or without 25 mM mannose. g, Western blots showing the levels of caspase-8, FADD and ERK2 in NTC, caspase-8 and FADD CRISPR Saos-2 cells. h, Percentage of empty and Bax/Bak CRISPR U2OS-E1a propidium-iodide-positive cells with or without 1 µg ml−1 doxorubicin and with or without 25 mM mannose. i, Western blots showing the levels of Bax, Bak and HSP90 in empty and Bax/Bak CRISPR U2OS-E1a cells. j, Western blots showing the levels of Bim, Noxa, Bad, Bax, Actin and HSP-90 in U2OS-E1a cells after 24 h treatment with or without 10 µM cisplatin, with or without 25 mM mannose and with or without 50 µM zVAD-FMK. The HSP-90 blot is identical to the one shown in Fig. 2f as the blots are from the same experiment. k, Western blots showing the levels of Mcl-1 and Bcl-XL in U2OS-E1a cells after 48 h with or without 10 µM cisplatin, with or without 25 mM mannose and in the absence or presence of 10 μM MG132 as indicated. l, m, PCR with reverse transcription (RT–PCR) showing the levels of BCL2L1 (Bcl-XL) (l) and MCL1 (m) mRNAs in U2OS-E1a cells after 48 h treatment with 10 µM cisplatin alone, 25 mM mannose alone, or both 10 µM cisplatin and 25 mM mannose, in the presence of 50 μM zVAD-FMK. n, o, Percentage of U2OS-E1a (empty), Mcl-1 and Bcl-XL overexpressing propidium-iodide-positive cells after 24 h treatment with or without 1 µg ml−1 doxorubicin and with or without 25 mM mannose. p, q, Western blot showing the levels of Mcl-1, Bcl-XL and HSP-90 in U2OS-E1a (empty), Mcl-1 and Bcl-XL overexpressing cells. n = 3 independent experiments (ac, e, f, h, lo). Data are representative of two independent experiments (g, j, k) or one experiment (i, p, q). In d, n = 3 independent experiments (each involving technical triplicates) (−zVAD-FMK) and n = 2 independent experiments (each involving technical triplicates) (+zVAD-FMK). All data are mean ± s.e.m. and were analysed by two-way ANOVA with Bonferroni correction (af, h, n, o). *P < 0.05, ***P < 0.001. Source data

  5. Extended Data Fig. 5 Sensitization to cell death by mannose is connected to Bcl-2 family members.

    a, Percentage of control (NTC) and Noxa CRISPR U2OS-E1a propidium-iodide-positive cells upon 24 h treatment with or without 10 μM cisplatin and with or without 25 mM mannose. b, Percentage of U2OS-E1a propidium-iodide-positive cells upon 24 h treatment with 10 μM WEHI539 (a BclXL inhibitor) or 10 μM ABT-199 (a Bcl-2 inhibitor), with or without 25 mM mannose. n = 4 independent experiments (a); n = 3 three independent experiments (b). All data are mean ± s.e.m. and were analysed by two-way ANOVA with Bonferroni correction (a) and two-tailed unpaired t-test (b). ***P < 0.001; ****P < 0.0001; NS, not significant. Source data

  6. Extended Data Fig. 6 Mannose affects cell proliferation and the uptake/retention of 18F-FDG, but it does not affect animal weight.

    a, Mannose levels in the plasma after 60 min in mice treated with a single dose of 200 µl water or 20% mannose in water. b, c, CD1-nude mice were transplanted with KP-4 cells subcutaneously and tumours were grown for 14 days. PET and MRI scans were performed for mice treated with 200 µl of water or 20% mannose in water by oral gavage 20 min before injection of [18F]FDG into the tail vein. b, Quantification of [18F]FDG uptake by tumours represented in average percentage injected dose per ml (%ID ml−1) ± s.d. c, Volume of each tumour at the time of the PET and MRI scans. Data were analysed by unpaired two-tailed Student’s t-test. d, CD1-nude mice were injected with KP-4 cells subcutaneously and treated with normal drinking water or 20% mannose in the drinking water, plus the same treatment (either normal water or 20% mannose) by oral gavage three days a week from the third day after tumour transplantation. Shown is the quantification of [18F]FDG uptake by the tumour and different organs represented in averaged kBq ml−1 ± s.d. Data were analysed by unpaired two-tailed Student’s t-test. e, CD1-nude mice were injected with KP-4 cells subcutaneously and treated with normal drinking water or 20% mannose in the drinking water, plus the same treatment (either normal water or 20% mannose) by oral gavage three days a week from the third day after tumour transplantation. The weight of mice was recorded at the indicated times. f, g, CD1-nude mice were injected with KP-4 cells subcutaneously and treated with normal drinking water or 20% mannose in the drinking water (either normal water or 20% mannose) by oral gavage three days a week from the third day after tumour transplantation. f, Images of BrdU sections representing tumours in control (left) and mannose-treated (right) mice. g, Quantification of BrdU-positive cells per section in control tumours (n = 4) and mannose-treated (n = 4) tumours. Five representative images per tumour were analysed. h, CD1-nude mice were injected with KP-4 cells subcutaneously and tumours were grown for 10 days before treatment with mannose and/or doxorubicin (doxo) was started. Mice received normal drinking water (ctrl and doxo) or 20% mannose in the drinking water (man and doxo + man) together with the same treatment by oral gavage three times per week. Doxorubicin treatment started on day 32 and mice received 5 mg kg−1 by intraperitoneal injection once per week. The weight of mice in all groups was recorded throughout the experiment. The number of mice for each experiment is as follows: n = 3 per group (a), n = 5 (−mannose), n = 4 (+mannose) (bd); n = 10 (h). In a, c and h data are mean ± s.e.m. Data were analysed with a two-tailed unpaired t-test (c). *P < 0.05, **P < 0.01. Source data

  7. Extended Data Fig. 7 PMI levels dictate the response to mannose.

    a, PMI expression levels correlate with PMI activities. PMI activities were measured in eight different cell lines using coupled enzymatic reactions. Graph shows the OD340nm measured at 2 h, reflecting the amount of NADPH/H+ produced by the reactions. Results from three independent experiments were normalized relative to PMI activities measured in Saos-2 cells and represent mean ± s.e.m. bg, MPI knockdown sensitizes cells to mannose. b, Western blot of SKOV3-transfected cells showing the levels of PMI and actin after 48 h of transient transfection with siRNAs. Growth curves of SKOV3 (c), RKO (f) and IGROV1 (g) in regular DMEM supplemented with or without 25 mM mannose after transient transfection for 48 h with two NTC and two MPI-targeting siRNAs. d, e, Western blot showing the levels of PMI and ERK2 in RKO (d) and PMI and actin in IGROV1 (e) 48 h after siRNA transfection. hk, Overexpression of PMI causes resistance to the growth-suppressing and death-promoting effects of mannose. PMI expression in U2OS-E1a (h) and Saos-2 (i) cells was confirmed by western blotting. j, Saos2-PMI cells and control cells (Saos-2) were plated in the presence or the absence of 25 mM mannose and cell numbers were counted at the indicated times. k, Percentage of propidium-iodide-positive U2OS-E1a control cells (vector) or U2OS-E1a cells overexpressing PMI (PMI) after 24 h treatment with or without 1 μg ml−1 doxorubicin in the presence or absence of 25 mM mannose. lo, Percentage of lactate (l), GA3-P (m), α-KG (n) and malate (o) metabolite content (peak area) in SKOV3 cells transfected with siRNA for 48 h before 6 h incubation in complete DMEM medium with or without supplementation of 25 mM mannose. n = 3 independent experiments (a, c, f, g, j, lo). Data are representative of two independent experiments (b, d, e, k), or one experiment (h, i). Data are mean ± s.e.m. and were analysed by two-tailed unpaired t-test (a), two-way ANOVA with Tukey correction (lo) or multiple unpaired two-sided t-test with Holm–Sidak correction (j). *P < 0.05, ***P < 0.001, ****P < 0.0001. Source data

  8. Extended Data Fig. 8 PMI levels dictate mannose sensitivity, and the weight of animals containing syngeneic allografts is not affected by treatment with mannose.

    a, b, Mpi knockdown causes growth retardation upon mannose treatment in a dose-dependent manner. a, B16-F1 cells infected with NTC or shRNA against Mpi were treated with (blue columns) or without (white columns) 25 mM mannose for 72 h, after which the number of cells was determined. Data are mean ± s.e.m. n = 3 independent experiments. b, The reduction in PMI was confirmed by western blotting. c, Western blot showing a reduction in PMI after Mpi knockdown in LLC cells. The data are relative to the experiments shown in Fig. 4g, j, k. d, B16-F1 cells infected with NTC or shRNA against Mpi were treated with 25 mM mannose, 25 mM glucose, 25 mM galactose, 25 mM fructose or 25 mM fucose. At 24, 48 and 72 h after hexose treatment, the number of cells was determined. Data are mean ± s.e.m. eh, Weights of mice injected with syngeneic cell lines: B16 shNTC (e), B16 shPMI (f), LLC shNTC (g) and LLC shPMI (h); the mice were given normal drinking water or drinking water supplemented with 20% mannose (n = 10 mice per group). All data are mean ± s.d. unless otherwise stated. In d, n = 4 independent experiments were analysed by two-way ANOVA followed by Tukey’s multiple comparisons. The data in b and c were performed only once. ***P < 0.001. Source data

  9. Extended Data Fig. 9 PMI levels vary in human tumours, but do not predict overall survival and the weight of mice is not affected by mannose treatment in models of colorectal cancer.

    a, Images of PMI expression for each TMA, representing one negative sample (left), one low expression (middle) and one high expression (right). One sample came from each of the ovarian, breast, prostate, colorectal or renal TMAs as indicated. b, c, Kaplan–Meier curves showing cancer-specific survival based on PMI levels in n = 698 patients with stage I–IV colorectal cancer (b) or n = 316 patients with primary operable breast cancer (c). d, Table showing the association of PMI and cancer-specific survival. Histoscores were split into high and low expression using the ROC curve analysis for each tumour type. log-rank analysis (two-sided) was used to compare PMI and cancer-specific survival using SPSS (version 22). e, Mean weight of 28 mice during treatment with AOM and DSS, with or without additional mannose treatment. f, Weight of aged VillincreERApcfl/+KrasG12D/+ mice during treatment with or without mannose. Data are mean of each group (n = 7 mice, −mannose; n = 8 mice, +mannose). Source data

  10. Extended Data Fig. 10 Mannose does not affect amino acid and fatty acid uptake, nor does it significantly affect serine and glycine levels, endoplasmic reticulum stress or proteasome activity, but it does affect autophagy, transcription and translation in a PMI-dependent manner.

    a, Exchange rates of amino acids between the indicated cells and their media measured after 48 h treatment with or without 25 mM mannose. Results are presented as peak area per microgram of protein per hour and are representative of one experiment. Shown is the mean of six technical replicates. b, Levels of 13C16-palmitate (expressed as peak area per microgram of protein) in U2OS-E1a cells incubated with 50 µM 13C16-palmitate, conjugated with 10% fatty-acid-free BSA, for 24 h in the presence or absence of 25 mM mannose in the culture medium (DMEM). n = 2 independent experiments (each involving six technical replicates). Data are mean ± s.e.m. c, d, Levels of 13C3-serine (c) and 13C2-glycine (d) (expressed as peak area per micrograms of protein) in U2OS-E1a cells incubated with 25 mM 13C6-glucose for the indicated time in the presence or absence of 25 mM mannose in the culture medium (DMEM). n = 3 independent experiments. e, f, Distribution of isotopologues of intracellular serine (e) and glycine (f) in U2OS-E1a cells incubated with 25 mM 13C6-glucose for 24 h in the presence (mannose) or absence (ctr) of 25 mM mannose in the culture medium (DMEM). n = 3 independent experiments. g, Mannose has no effect on the unfolded protein response. U2OS-E1a cells were treated with 25 mM mannose for 16 and 24 h. 2-Deoxy-d-glucose (2DG) and tunicamycin (TM) serve as positive controls. Induction of Bip (also known as GRP78) is a read-out of endoplasmic reticulum stress. The data are representative of three independent experiments. h, Proteasome activity is not affected by mannose. Mannose-sensitive U2OS-E1a cells were incubated in either DMEM or DMEM containing 25 mM mannose before measurement of proteasome activities. Cells were also treated with10 μM MG132 as a control for proteasome activity. n = 3 independent experiments. i, j, U2OS-E1a cells (i) or U2OS-E1a (U2OS) cells overexpressing PMI (j) were treated with 25 mM mannose for the indicated times in the presence or absence of 20 μM chloroquine (CQ) (4 h). Western blotting was undertaken to detect LC3-B and actin. The data are representative of three independent experiments. k, l, Relative incorporation of 32P UTP (k) or 35S methionine (l) into U2OS-E1a control cells (vector) or U2OS-E1a cells overexpressing PMI (PMI) in the presence or absence of 25 mM mannose. Where indicated, 5 μM actinomycin D was used to inhibit transcription or 100 μg ml−1 cycloheximide was used to inhibit translation. In k, n = 3 independent experiments. In l, n = 3 independent experiments (control and mannose) and n = 2 independent experiments (CHX).m, n, U2OS-E1a cells infected with lentiCRISPR-NTC 1, NTC 2, ATG5 or ATG7 were treated with 25 mM d-(+)-mannose for 72 h. m, The number of cells counted after the 72-h treatment; the results represent the mean of one independent experiment performed in triplicate. n, Western blots show loss of LC3 lipidation and p62 accumulation in ATG5 and ATG7 knockout cells. Data were analysed by two-tailed unpaired t-tests. *P < 0.05, ***P < 0.001. Unless otherwise stated, data are mean ± s.e.m. Source data

Supplementary information

  1. Supplementary Figure 1

    This file contains the uncropped gels.

  2. Reporting Summary

Source data

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/s41586-018-0729-3

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.