Letter | Published:

Histidine catabolism is a major determinant of methotrexate sensitivity

Naturevolume 559pages632636 (2018) | Download Citation

Abstract

The chemotherapeutic drug methotrexate inhibits the enzyme dihydrofolate reductase1, which generates tetrahydrofolate, an essential cofactor in nucleotide synthesis2. Depletion of tetrahydrofolate causes cell death by suppressing DNA and RNA production3. Although methotrexate is widely used as an anticancer agent and is the subject of over a thousand ongoing clinical trials4, its high toxicity often leads to the premature termination of its use, which reduces its potential efficacy5. To identify genes that modulate the response of cancer cells to methotrexate, we performed a CRISPR–Cas9-based screen6,7. This screen yielded FTCD, which encodes an enzyme—formimidoyltransferase cyclodeaminase—that is required for the catabolism of the amino acid histidine8, a process that has not previously been linked to methotrexate sensitivity. In cultured cancer cells, depletion of several genes in the histidine degradation pathway markedly decreased sensitivity to methotrexate. Mechanistically, histidine catabolism drains the cellular pool of tetrahydrofolate, which is particularly detrimental to methotrexate-treated cells. Moreover, expression of the rate-limiting enzyme in histidine catabolism is associated with methotrexate sensitivity in cancer cell lines and with survival rate in patients. In vivo dietary supplementation of histidine increased flux through the histidine degradation pathway and enhanced the sensitivity of leukaemia xenografts to methotrexate. The histidine degradation pathway markedly influences the sensitivity of cancer cells to methotrexate and may be exploited to improve methotrexate efficacy through a simple dietary intervention.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Additional information

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

References

  1. 1.

    Farber, S. & Diamond, L. K. Temporary remissions in acute leukemia in children produced by folic acid antagonist, 4-aminopteroyl-glutamic acid. N. Engl. J. Med. 238, 787–793 (1948).

  2. 2.

    Lieberman, I. & Ove, P. Control of growth of mammalian cells in culture with folic acid, thymidine, and purines. J. Biol. Chem. 235, 1119–1123 (1960).

  3. 3.

    Hitchings, G. H. & Burchall, J. J. Inhibition of folate biosynthesis and function as a basis for chemotherapy. Adv. Enzymol. 27, 417–468 (1965).

  4. 4.

    ClinicalTrials.gov https://clinicaltrials.gov/ct2/results?cond=&term=methotrexate&cntry1=&state1=&recrs= (NIH, 2017).

  5. 5.

    Howard, S. C., McCormick, J., Pui, C. H., Buddington, R. K. & Harvey, R. D. Preventing and managing toxicities of high-dose methotrexate. Oncologist 21, 1471–1482 (2016).

  6. 6.

    Shalem, O. et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84–87 (2014).

  7. 7.

    Wang, T., Wei, J. J., Sabatini, D. M. & Lander, E. S. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80–84 (2014).

  8. 8.

    Solans, A., Estivill, X. & de la Luna, S. Cloning and characterization of human FTCD on 21q22.3, a candidate gene for glutamate formiminotransferase deficiency. Cytogenet. Cell Genet. 88, 43–49 (2000).

  9. 9.

    Wilson, P. M., Danenberg, P. V., Johnston, P. G., Lenz, H. J. & Ladner, R. D. Standing the test of time: targeting thymidylate biosynthesis in cancer therapy. Nat. Rev. Clin. Oncol. 11, 282–298 (2014).

  10. 10.

    Wang, T. et al. Identification and characterization of essential genes in the human genome. Science 350, 1096–1101 (2015).

  11. 11.

    Assaraf, Y. G. & Schimke, R. T. Identification of methotrexate transport deficiency in mammalian cells using fluoresceinated methotrexate and flow cytometry. Proc. Natl Acad. Sci. USA 84, 7154–7158 (1987).

  12. 12.

    Matherly, L. H. & Taub, J. W. Methotrexate pharmacology and resistance in childhood acute lymphoblastic leukemia. Leuk. Lymphoma 21, 359–368 (1996).

  13. 13.

    Guo, W. et al. Mechanisms of methotrexate resistance in osteosarcoma. Clin. Cancer Res. 5, 621–627 (1999).

  14. 14.

    Mao, Y. et al. Structure of the bifunctional and Golgi-associated formiminotransferase cyclodeaminase octamer. EMBO J. 23, 2963–2971 (2004).

  15. 15.

    Hatefi, Y., Osborn, M. J., Kay, L. D. & Huennekens, F. M. Hydroxymethyl tetrahydrofolic dehydrogenase. J. Biol. Chem. 227, 637–647 (1957).

  16. 16.

    Field, M. S., Szebenyi, D. M. & Stover, P. J. Regulation of de novo purine biosynthesis by methenyltetrahydrofolate synthetase in neuroblastoma. J. Biol. Chem. 281, 4215–4221 (2006).

  17. 17.

    La Du, B. N., Howell, R. R., Jacoby, G. A., Seegmiller, J. E. & Zannoni, V. G. The enzymatic defect in histidinemia. Biochem. Biophys. Res. Commun. 7, 398–402 (1962).

  18. 18.

    HumanCyc https://humancyc.org (2017).

  19. 19.

    Yu, C. et al. The development of PIPA: an integrated and automated pipeline for genome-wide protein function annotation. BMC Bioinformatics 9, 52 (2008).

  20. 20.

    Brand, L. M. & Harper, A. E. Studies on the production and assessment of experimental histidinemia in the rat. Biochim. Biophys. Acta 444, 294–306 (1976).

  21. 21.

    Cancer Cell Line Encyclopedia https://broadinstitute.org/ccle

  22. 22.

    Cooper, S. L. & Brown, P. A. Treatment of pediatric acute lymphoblastic leukemia. Pediatr. Clin. North Am. 62, 61–73 (2015).

  23. 23.

    Kodidela, S., Suresh Chandra, P. & Dubashi, B. Pharmacogenetics of methotrexate in acute lymphoblastic leukaemia: why still at the bench level? Eur. J. Clin. Pharmacol. 70, 253–260 (2014).

  24. 24.

    Roberts, K. G. et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N. Engl. J. Med. 371, 1005–1015 (2014).

  25. 25.

    Liem, N. L. et al. Characterization of childhood acute lymphoblastic leukemia xenograft models for the preclinical evaluation of new therapies. Blood 103, 3905–3914 (2004).

  26. 26.

    Holmes, W. B. & Appling, D. R. Cloning and characterization of methenyltetrahydrofolate synthetase from Saccharomyces cerevisiae. J. Biol. Chem. 277, 20205–20213 (2002).

  27. 27.

    Stover, P. & Schirch, V. Serine hydroxymethyltransferase catalyzes the hydrolysis of 5,10-methenyltetrahydrofolate to 5-formyltetrahydrofolate. J. Biol. Chem. 265, 14227–14233 (1990).

  28. 28.

    Stover, P., Kruschwitz, H. & Schirch, V. Evidence that 5-formyltetrahydropteroylglutamate has a metabolic role in one-carbon metabolism. Adv. Exp. Med. Biol. 338, 679–685 (1993).

  29. 29.

    Birsoy, K. et al. Metabolic determinants of cancer cell sensitivity to glucose limitation and biguanides. Nature 508, 108–112 (2014).

  30. 30.

    Wang, T., Lander, E. S. & Sabatini, D. M. Single guide RNA library design and construction. Cold Spring Harb. Protoc. 2016, pdb.prot090803 (2016).

  31. 31.

    Wang, T., Lander, E. S. & Sabatini, D. M. Viral packaging and cell culture for CRISPR-based screens. Cold Spring Harb. Protoc. 2016, pdb.prot090811, (2016).

  32. 32.

    Wang, T., Lander, E. S. & Sabatini, D. M. Large-scale single guide RNA library construction and use for CRISPR–Cas9-based genetic screens. Cold Spring Harb. Protoc. 2016, pdb.top086892 (2016).

  33. 33.

    The RNAi Consortium. The RNAi Consortium shRNA Library https://www.broadinstitute.org/rnai-consortium/rnai-consortium-shrna-library (2017).

  34. 34.

    Chen, L., Ducker, G. S., Lu, W., Teng, X. & Rabinowitz, J. D. An LC-MS chemical derivatization method for the measurement of five different one-carbon states of cellular tetrahydrofolate. Anal. Bioanal. Chem. 409, 5955–5964 (2017).

  35. 35.

    Allen, F., Pon, A., Wilson, M., Greiner, R. & Wishart, D. CFM-ID: a web server for annotation, spectrum prediction and metabolite identification from tandem mass spectra. Nucleic Acids Res. 42, W94–W99 (2014).

  36. 36.

    Böcker, S. & Dührkop, K. Fragmentation trees reloaded. J. Cheminform. 8, 5 (2016).

  37. 37.

    Dührkop, K., Shen, H., Meusel, M., Rousu, J. & Böcker, S. Searching molecular structure databases with tandem mass spectra using CSI:FingerID. Proc. Natl Acad. Sci. USA 112, 12580–12585 (2015).

  38. 38.

    Fernandez, C. A., Des Rosiers, C., Previs, S. F., David, F. & Brunengraber, H. Correction of 13C mass isotopomer distributions for natural stable isotope abundance. J. Mass Spectrom. 31, 255–262 (1996).

  39. 39.

    Lewis, C. A. et al. Tracing compartmentalized NADPH metabolism in the cytosol and mitochondria of mammalian cells. Mol. Cell 55, 253–263 (2014).

Download references

Acknowledgements

We thank all members of the Sabatini laboratory for their advice and help; W. Chen and L. Shinefeld for helpful comments on the manuscript; M. Pacold for advice; M. Cohen for the R-luciferase expression vector; and M. Lazzara, D. Lauffenburger, M. Zaitseva, A. Al-Katib and R. Jensen for providing cell lines. We thank J. Selhub for advice in the folate field. This work was supported by grants from National Institutes of Health/National Cancer Institute (R01 CA129105) to D.M.S. and the Department of Defense (W81XWH-15-1-0337) to E.F. Fellowship support was provided by the European Molecular Biology Organization (EMBO) (Long-Term Fellowship ALTF 350-2012) and the American Association for Cancer Research (16-40-38-KANA) to N.K., by the American Cancer Society (PF-12-099-01-TBG) and the Koch Institute (Ludwig Postdoctoral Fellowship) to J.R.C., and by the EMBO (Long-Term Fellowship ALTF 1-2014) to M.A.-R. Additional support to N.K. was provided by the Women In Science/Revson Foundation Award (Weizmann Institute) and The Advancement of Women in Science Award (The Hebrew University). D.M.S. is an investigator of the Howard Hughes Medical Institute and an American Cancer Society Research Professor.

Reviewer information

Nature thanks C. Frezza and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Affiliations

  1. Whitehead Institute for Biomedical Research and Massachusetts Institute of Technology, Department of Biology, Cambridge, MA, USA

    • Naama Kanarek
    • , Heather R. Keys
    • , Jason R. Cantor
    • , Caroline A. Lewis
    • , Sze Ham Chan
    • , Tenzin Kunchok
    • , Monther Abu-Remaileh
    • , Elizaveta Freinkman
    •  & David M. Sabatini
  2. Howard Hughes Medical Institute, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA, USA

    • Naama Kanarek
    • , Jason R. Cantor
    • , Monther Abu-Remaileh
    •  & David M. Sabatini
  3. Koch Institute for Integrative Cancer Research and Massachusetts Institute of Technology, Department of Biology, Cambridge, MA, USA

    • Naama Kanarek
    • , Jason R. Cantor
    • , Monther Abu-Remaileh
    •  & David M. Sabatini
  4. Broad Institute of Harvard and Massachusetts Institute of Technology, Cambridge, MA, USA

    • Naama Kanarek
    • , Jason R. Cantor
    • , Monther Abu-Remaileh
    • , Lawrence D. Schweitzer
    •  & David M. Sabatini

Authors

  1. Search for Naama Kanarek in:

  2. Search for Heather R. Keys in:

  3. Search for Jason R. Cantor in:

  4. Search for Caroline A. Lewis in:

  5. Search for Sze Ham Chan in:

  6. Search for Tenzin Kunchok in:

  7. Search for Monther Abu-Remaileh in:

  8. Search for Elizaveta Freinkman in:

  9. Search for Lawrence D. Schweitzer in:

  10. Search for David M. Sabatini in:

Contributions

N.K. and D.M.S. formulated the research plan and interpreted experimental results. N.K. designed and performed all experiments with assistance from J.R.C. and M.A.-R. H.R.K. performed data analysis for the CRISPR–Cas9-based screen, Cancer Cell Line Encyclopedia (CCLE) gene expression data and data from patients with ALL. J.R.C. carried out the genomic barcoding of 42 haematopoietic cell lines. E.F. and C.A.L. ran the metabolite-profiling core facility and E.F., C.A.L., S.H.C. and T.K. advised and helped to plan experiments for metabolite detection by LC–MS. S.H.C. and T.K. assisted with LC–MS sample preparation and running of the method. L.D.S. designed and prepared some figures for the manuscript, N.K. and D.M.S. wrote the manuscript, and all authors edited it.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to David M. Sabatini.

Extended data figures and tables

  1. Extended Data Fig. 1 Loss of FTCD decreases the sensitivity of cancer cells to methotrexate.

    a, Relative daily cell counts of the mixed culture of 42 genomically barcoded haematopoietic cancer cell lines. We focused our efforts on haematopoietic cell lines because methotrexate is most commonly used to treat haematopoietic malignancies9. Cells were co-cultured and treated with three concentrations of methotrexate (0.1, 0.5, or 5 μM). Cell counts are presented relative to the vehicle-treated co-culture. n = 3, biological replicates. Error bars indicate s.d. b, Sequencing results of the competitive co-culture experiment. Heat map of relative barcode abundance for each indicated cell line after treatment with 5 μM methotrexate or vehicle after 2, 4, or 6 days compared to those in the initial cultures. c, Validation of the results of the competitive co-culture experiment as shown by individual dose–response curves of seven relatively sensitive cell lines (MOLM13, EOL-1, SUPM2, DEL, HEL, NALM6 and SUDHL1), and six relatively resistant cell lines (HL-60, Ramos, HBL-1, Raji, SUDHL8 and Jeko-1). n = 2, biological replicates. Each biological replicate included technical triplicate. Error bars indicate s.d. d, Top 40 genes that scored in the genome-wide CRISPR–Cas9-based positive selection screen. These genes had the largest differential CRISPR scores between vehicle- and methotrexate-treated samples. The heat map represents the CRISPR score of two biological replicates of each screen (vehicle- and methotrexate-treated). e, Abundance of each of the individual sgRNAs targeting FTCD in the screen. Read counts for each sgRNA is presented for each of the screen biological replicates of the vehicle-treated and methotrexate-treated samples. f, Fold change in the EC90 values of methotrexate and doxorubicin in HEL cells stably expressing shRNA targeting either SLC19A1 or FTCD compared to the average of three non-targeting shRNAs (shGFP, shRFP and shLacZ). n = 2, biological replicates. Each biological replicate included technical triplicate. Error bars indicate s.e.m. g, Expression levels of SLC19A1 (left) and FTCD (right) in HEL cells stably expressing non-targeting shRNAs (shLacZ and shGFP) or targeting shRNAs (shSLC19A1 and shFTCD). Expression levels were measured by qPCR and normalized to the average of two control genes, UBC and HPRT1. n = 3, technical replicates. Error bars indicate s.e.m. h, Validation of genetic depletion of FTCD by the CRISPR–Cas9 system in HEL, Ramos and LAMA84 cell lines. Expression of FTCD was measured by qPCR and normalized to the average of two control genes, UBC and HPRT1. n = 3 (except for HEL cells expressing sgFTCD_2 + mFTCD, then n = 2 owing to loss of RNA from one sample), biological replicates. Error bars indicate s.d. Source data

  2. Extended Data Fig. 2 FTCD depletion enables cancer cells to maintain THF pools and nucleotide synthesis even when treated with methotrexate (part 1).

    a, The histidine degradation pathway, as previously described14,17,18,19. Enzymes are marked in blue. b, Metabolites detected by LC–MS and their corresponding retention times. The LC column used for the detection of each metabolite is indicated. c, Greater pool of 5-methyl THF in vehicle-treated HEL cells after FTCD depletion. It is not readily clear why, in vehicle-treated HEL cells, FTCD depletion caused a reduction in THF levels (Fig. 2d). However, 5-methyl THF levels in these cells were increased significantly (see also Extended Data Fig. 3c), indicating that there was no depletion in the overall recyclable amounts of THF in these cells. 5-Methyl THF levels were measured by LC–MS in vehicle-treated HEL and Ramos cells. 5-Methyl THF levels were normalized to aminopterin as an internal standard. P values were calculated using one-way ANOVA. n = 3, biological replicates. Error bars indicate s.e.m. d, Chemical structures of the folate entities found in cells and mentioned in Fig. 2c. e, f, The labelling rate by [U-13C]serine is not different between FTCD-depleted cells and control cells. Fractional labelling of glycine (e) and serine (f) is unchanged by FTCD depletion in HEL and Ramos cells in vehicle- and methotrexate-treated cells. Glycine and serine levels were normalized to isotopically labelled glutamate as an internal standard. P values were calculated for the unlabelled fraction by one-way ANOVA. n = 3, biological replicates. Error bars indicate s.e.m. UROC1, urocanate hydratase 1. Source data

  3. Extended Data Fig. 3 FTCD depletion enables cancer cells to maintain THF pools and nucleotide synthesis even when treated with methotrexate (part 2).

    a, b, Pool sizes of glycine (a) and serine (b) are not significantly different between FTCD-depleted and control cells. Glycine and serine levels were measured in vehicle- or methotrexate-treated HEL and Ramos cells. P values were calculated by one-way ANOVA. n = 3, biological replicates. Error bars indicate s.e.m. c, d, Greater cellular pool of 5,10-methenyl THF in methotrexate-treated cells after FTCD depletion. [U-13C]serine labelling of 5,10-methenyl THF showed higher abundance of unlabelled 5,10-methenyl THF in FTCD-depleted cells, compared to control cells. This suggests higher availability of 5,10-methenyl THF in these cells at the time of labelling onset, which agrees with the higher levels of both THF and 5,10-methenyl THF in the methotrexate-treated FTCD-depleted cells compared to the methotrexate-treated control cells. c, 5,10-Methenyl THF levels were measured by LC–MS in vehicle- and methotrexate-treated HEL and Ramos cells. 5,10-Methenyl THF levels were normalized to aminopterin as an internal standard. P values were calculated by one-way ANOVA. n = 3 biological replicates. Error bars indicate s.e.m. d, HEL cells were treated with 5 μM methotrexate for 48 h, and Ramos cells were treated with 20 μM methotrexate for 72 h to decrease THF levels and nucleotide synthesis in wild-type cells. The medium was then replaced with [U-13C]serine-containing medium, and cells were incubated in [U-13C]serine plus methotrexate for an additional 24 h followed by cell collection and LC–MS analysis. The higher abundance of unlabelled 5,10-methenyl THF suggests higher availability of reduced folate at the time of labelling onset. 5,10-Methenyl THF levels were normalized to aminopterine as an internal standard. P values were calculated for the unlabelled fraction by one-way ANOVA. n = 3, biological replicates. Error bars indicate s.e.m. Source data

  4. Extended Data Fig. 4 The histidine degradation pathway affects the sensitivity of cancer cells to methotrexate and HAL expression is associated with treatment response in patients with ALL (part 1).

    a, Genetic depletion of the enzymes HAL and AMDHD1 decreased sensitivity to methotrexate, but not to a control drug, doxorubicin. Cell viability after treatment with varying concentrations of methotrexate and doxorubicin was used to calculate the EC90 values. n = 3, biological replicates. Error bars indicate s.e.m. b, Expression levels of AMDHD1 (left) and FTCD (right) were not significantly different across methotrexate-sensitive haematopoietic cell lines compared to methotrexate-resistant cell lines. The response to methotrexate was determined in a pooled fashion using genomically barcoded cell lines (Fig. 1a and Extended Data Fig. 1a–c). Expression levels of AMDHD1 and FTCD were measured by qPCR and normalized to the average of control genes (UBC and HPRT1). P values were calculated using the Kolmogorov–Smirnov test. n = 4 for resistant cell lines and n = 6 for sensitive cell lines (biologically independent samples). Each qPCR included three technical replicates. Centre bars indicate mean, error bars indicate s.e.m. c. Fractional labelling of glycine by [U-13C]serine is unchanged by HAL depletion in HEL and Ramos cells in vehicle-and methotrexate-treated cells. d, Uptake of [U-13C]serine is higher in methotrexate-treated control cells but not in methotrexate-treated HAL-deficient cells. e, Glycine levels are not significantly different between HAL-deficient and control cells except for HEL cells treated with methotrexate. f, Serine levels are not significantly different between HAL-deficient and control cells except for HEL cells treated with methotrexate, in which HAL-deficient HEL cells have similar levels of serine to vehicle-treated cells. For cf, glycine and serine levels were normalized to isotopically labelled glutamate as an internal standard. P values were calculated for the unlabelled fraction (c, d) or for total values (e, f) by one-way ANOVA. n = 3, biological replicates. Error bars indicate s.e.m. Source data

  5. Extended Data Fig. 5 The histidine degradation pathway affects the sensitivity of cancer cells to methotrexate and HAL expression is associated with treatment response in patients with ALL (part 2).

    a, Validation of genetic depletion of HAL by the CRISPR–Cas9 system in NCIH1666 and EOL-1 cells. Expression levels of HAL were measured by qPCR and normalized to the average of two control genes, UBC and HPRT1. P values were calculated by two-way ANOVA. n = 3 technical replicates. Error bars indicate s.d. bd, Expression levels of SLC19A1 (b), AMDHD1 (c) or FTCD (d) do not predict survival of paediatric patients with ALL that are treated with a regimen that includes methotrexate. Kaplan–Meier curves of overall survival of patients with ALL with high (top quantile, coloured green) and low (bottom quantile, coloured blue)24 expression of each of the assayed genes. Patient sample size for each group is indicated. P values were calculated using the log-rank (Mantel–Cox) test. Source data

  6. Extended Data Fig. 6 In vivo histidine supplementation increases flux through the histidine degradation pathway and sensitizes tumours to methotrexate (part 1).

    a, In vivo imaging of luciferase-expressing HEL tumour xenografts. Mice were imaged before (top images) and after (bottom image) five days of methotrexate treatment alone or in combination with histidine supplementation. For a, c, HEL-cell-derived tumour-bearing mice: vehicle (n = 5), histidine supplementation (n = 4), methotrexate (n = 6), methotrexate + histidine supplementation (n = 6). Four mice per group are presented. b, In vivo imaging of luciferase-expressing SEM tumour xenografts. Mice were imaged before (top images) and after (bottom image) five days of treatment. All mice are presented. Mice are numbered in colour by their experimental group. For b, d, SEM-cell-derived tumour-bearing mice: vehicle (n = 6), histidine supplementation (n = 7), methotrexate (n = 7), methotrexate + histidine supplementation (n = 7). c. Additional images of H&E analyses of HEL-cell-derived tumour sections from methotrexate-treated and methotrexate + histidine supplementation-treated mice. d. H&E images of SEM-cell-derived tumour sections from all treatment groups. Three mice per group are presented.

  7. Extended Data Fig. 7 In vivo histidine supplementation increases flux through the histidine degradation pathway and sensitizes tumours to methotrexate (part 2).

    a, Higher magnification of H&E images of tumour sections from SEM-cell-derived tumour-bearing mice from all groups. For SEM-cell-derived tumour-bearing mice: vehicle (n = 6), histidine supplementation (n = 7), methotrexate (n = 7), methotrexate + histidine supplementation (n = 7). b, H&E analyses of kidney sections from HEL-cell-derived tumour-bearing mice from all experimental groups. For HEL-cell-derived tumour-bearing mice: vehicle (n = 5), histidine supplementation (n = 4), methotrexate (n = 6), methotrexate + histidine supplementation (n = 6).

  8. Extended Data Fig. 8 In vivo histidine supplementation increases flux through the histidine degradation pathway and sensitizes tumours to methotrexate (part 3).

    THF levels decreased after methotrexate treatment and decreased even further when methotrexate was combined with histidine supplementation. Methotrexate levels were not different in tumours from mice treated with methotrexate alone or in combination with histidine supplementation. Nucleotide abundance was significantly lower in tumours from mice treated with methotrexate and histidine supplementation compared to tumours from vehicle-treated mice. P values were calculated using non-parametric one-way ANOVA for all comparisons (Kruskal–Wallis test, Dunn’s post hoc test), except for methotrexate, for which P values were calculated by a two tailed t-test. Vehicle (n = 6), histidine (n = 6), methotrexate (n = 6), methotrexate + histidine (n = 7). Centre bars indicate mean, error bars indicate s.e.m. Source data

  9. Extended Data Fig. 9 In vivo histidine supplementation increases flux through the histidine degradation pathway and sensitizes tumours to methotrexate (part 4).

    a, H&E analyses of liver sections from HEL-cell-derived tumour-bearing mice from all experimental groups. For HEL-cell-derived tumour-bearing mice: vehicle (n = 5), histidine supplementation (n = 4), methotrexate (n = 6), methotrexate + histidine supplementation (n = 6). b, Weight loss, presented as a percentage of the weight measured on the first experimental day, before treatment. For SEM-cell-derived tumour-bearing mice: vehicle (n = 6), histidine supplementation (n = 7), methotrexate (n = 7), methotrexate + histidine supplementation (n = 7). P value was calculated using a non-parametric t-test (Mann–Whitney). c, d, Metabolites of the histidine degradation pathway were increased in HEL tumours from mice treated with histidine supplementation. Histidine (left) and FIGLU (right) levels were measured in tumours by LC–MS and normalized to isotopically labelled histidine as an internal standard. P values were calculated using a non-parametric t-test (Mann–Whitney). All metabolites measured in tumours were normalized to an average of four amino acids (phenylalanine, leucine, valine and tyrosine) as an internal loading control. For HEL-cell-derived tumour-bearing mice: vehicle (n = 5), histidine supplementation (n = 4), methotrexate (n = 5), methotrexate + histidine supplementation (n = 6). For SEM-cell-derived tumour-bearing mice: vehicle (n = 6), histidine supplementation (n = 6), methotrexate (n = 6), methotrexate + histidine supplementation (n = 7). e, f, Plasma levels of methotrexate, histidine, 5-methyl THF and folate. Methotrexate was detected in the plasma of methotrexate-treated mice only (left). No significant difference in histidine levels was detected (second from left). Levels of 5-methyl THF were significantly lower in the plasma of methotrexate-treated mice (second from right). Folate levels increased in the plasma of methotrexate-treated mice (right). Metabolite levels were measured from fresh plasma samples by LC–MS and normalized to isotopically labelled histidine or to aminopterin as an internal standard. P values were calculated using non-parametric one-way ANOVA. Group sizes are the same as in c, d. Centre bars indicate mean, error bars indicate s.e.m. Source data

  10. Extended Data Fig. 10 In vivo histidine supplementation sensitizes tumours to methotrexate without enhancement of treatment toxicity (part 1).

    a, We evaluated whether methotrexate treatment combined with histidine supplementation might be more toxic than methotrexate alone by setting up a longer treatment regime of 15 days with a recovery period of two weeks. During the experiment we monitored weight loss and observed no difference between mice treated with methotrexate alone and those treated with methotrexate and histidine. NOD-SCID mice were injected with HEL cells subcutaneously on day 1, followed by weight measurement every other day, in vivo imaging of HEL-cell-derived tumours (on days 7, 15, 20 and 28), treatment on days 12 to 23 (vehicle, histidine and methotrexate injections every other day), and final termination of the experiment after 40 days, unless an early euthanization was required in accordance with the guidelines for humane experimental end-point of the Committee on Animal Care at MIT. The experiment included four experimental groups: vehicle-treated (saline) (n = 7), histidine supplementation (n = 8), methotrexate-treated (n = 8), and histidine supplementation combined with methotrexate treatment (n = 8). Serum was collected at day 0 and day 23 for metabolite profiling and liver diagnostics. The dose of methotrexate used was 25 mg kg−1 based on the weight measured at day 0. The dose of histidine was 18 mg per injection in 400 μl saline. b, A significant reduction in tumour size over time was found in mice treated with the combination of methotrexate and histidine supplementation. Tumours were imaged in vivo by luciferase expression at the indicated days. Fold changes in tumour sizes over measurements performed on day 7 are presented. P values were calculated by non-parametric one-way ANOVA. Group size changed over time owing to mouse euthanization for humane reasons. On day 15: vehicle-treated (n = 5), histidine supplementation (n = 7), methotrexate-treated (n = 7) and histidine supplementation combined with methotrexate treatment (n = 8). On day 20: vehicle-treated (n = 6), histidine supplementation (n = 6), methotrexate-treated (n = 7) and histidine supplementation combined with methotrexate treatment (n = 8). On day 28: vehicle-treated (n = 6), histidine supplementation (n = 5), methotrexate-treated (n = 7) and histidine supplementation combined with methotrexate treatment (n = 8). c, In vivo imaging of luciferase-expressing HEL-cell-derived tumours at days 7 (top) and 28 (bottom). Mice are labelled by number, with the colour of the label representing their experimental group. All mice assigned to the experiment are shown, group size is the same as in a. d, e, There was no increase in the abundance of serum markers indicative of kidney damage in mice treated with methotrexate + histidine supplementation compared to mice treated with methotrexate. Markers of kidney toxicity (urea and creatinine) were measured by LC–MS in serum samples of the tested mice. The relative abundance of urea and creatinine was normalized to isotopically labelled valine and tryptophan as an internal standard. P values were calculated using non-parametric one-way ANOVA. Group size: vehicle-treated (n = 7), histidine supplementation (n = 7), methotrexate-treated (n = 8), and histidine supplementation combined with methotrexate treatment (n = 6). f, g, Some increase in liver-toxicity markers in mice treated with the combined therapy compared to mice treated with methotrexate alone. Markers of liver toxicity (alanine aminotransferase (ALT) and aspartate aminotransferase (AST)) were measured by an external serum diagnostics laboratory (IDEXX) in serum samples of the tested mice. Measurement units are indicated. P values were calculated using non-parametric one-way ANOVA. Group size: vehicle-treated (n = 6), histidine supplementation (n = 6), methotrexate-treated (n = 6), and histidine supplementation combined with methotrexate treatment (n = 5). Centre bars indicate mean, error bars indicate s.e.m. Source data

  11. Extended Data Fig. 11 In vivo histidine supplementation sensitizes tumours to methotrexate without enhancement of treatment toxicity (part 2).

    a, b, Histological analyses indicated that the kidney and liver appeared normal at the end of the two-week recovery period. See also Supplementary Information Fig. 3a. a, H&E analyses of kidney sections from mice from all experimental groups. Tissues were collected at the conclusion of the experiment, after two weeks post-treatment recovery. Group sizes are the same as in Extended Data Fig. 9a. b, H&E analyses of liver sections from mice from all experimental groups. Group sizes are the same as in Extended Data Fig. 9a. Tissues were collected at the conclusion of the experiment, after two weeks post-treatment recovery. See also Supplementary Information Fig. 3b.

Supplementary information

  1. Supplementary Figures

    This file contains Supplementary Figures 1-3.

  2. Reporting Summary

Source Data

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/s41586-018-0316-7

Further reading

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.