DOI: 10.1126/science.1226603, 222 (2013);339 Science

et al.Ng Shyh-ChangHistone Methylation

-Adenosylmethionine andSInfluence of Threonine Metabolism on

This copy is for your personal, non-commercial use only.

clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

here.following the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles

): August 18, 2013 www.sciencemag.org (this information is current as ofThe following resources related to this article are available online at

http://www.sciencemag.org/content/339/6116/222.full.htmlversion of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/content/suppl/2012/10/31/science.1226603.DC1.html can be found at: Supporting Online Material

http://www.sciencemag.org/content/339/6116/222.full.html#relatedfound at:

can berelated to this article A list of selected additional articles on the Science Web sites

http://www.sciencemag.org/content/339/6116/222.full.html#ref-list-1, 10 of which can be accessed free:cites 24 articlesThis article

http://www.sciencemag.org/content/339/6116/222.full.html#related-urls4 articles hosted by HighWire Press; see:cited by This article has been

http://www.sciencemag.org/cgi/collection/physiologyPhysiology

subject collections:This article appears in the following

registered trademark of AAAS. is aScience2013 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on A

ugus

t 18,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

largely suppressed in HFD-fed FKO mice (figs.S10A, S11, and S12). Moreover, JNK deficiencyreduced chemokine expression by macrophagesin vitro under a variety of stimulation conditions(figs. S10B and S13 to S15). Because tissuechemokine expression is mediated by both pa-renchymal cells and macrophages, it is likely thatthe reduced chemokine expression detected inHFD-fed FKO mice is a consequence of both aprimary macrophage defect and a secondary re-sponse to improved glycemia. Collectively, thesedata indicate that reduced chemokine signalingmay contribute to the decreased accumulation ofATMs inFKO mice. However, no selective defectin the expression of CCR2 or CCR5 receptors orligands was detected inFKOmice compared withFWTmice. The loss ofM1ATMs inFKOmicemaytherefore represent an intrinsic defect in M1 polar-ization rather than inefficient M1 ATM recruitment.

To test the requirement of JNK forM1macro-phage function in vitro, we isolated bonemarrowderived macrophages (BMDMs) from FWT andFKO mice (Fig. 4). Treatment with IFN-g causesM1 differentiation (3) and expression of the cell-surface marker CD11c (20). Macrophage-specificJNK deficiency caused reducedM1 differentiationas compared with that in control mice (Fig. 4A).Furthermore, JNK deficiency caused decreasedexpression ofM1marker genes (Ccr7 andCd11c),chemokines (Ccl2 and Ccl5), and cytokine genes(Il1b, Il6, and Tnfa) in IFN-gstimulated macro-phages (Fig. 4B and fig. S14). Similar defects weredetected in lipopolysaccharide (LPS)stimulatedmacrophages (fig. S15). Moreover, the blood con-centration of M1-associated cytokines and che-mokines was significantly reduced in LPS-treatedFKO mice as compared with FWT mice in vivo

(Fig. 4C). Together, these data demonstrate thatJNK deficiency suppresses M1 polarization. Incontrast, no significant difference in M2 differ-entiation betweenBMDMs isolated fromFWTandFKO mice was detected (figs. S16 to S18).

The observation that JNK is required for thedifferentiation of pro-inflammatory macrophagesprovides an explanation, in part, for previous find-ings that have implicated JNK in inflammatoryresponses (17), including the promotion of TH1immune responses (2123). JNK-dependent inflam-matory macrophages in the liver may account forthe requirement of JNK for the development offulminant hepatitis in mouse models (24). JNK-dependent tissue macrophages may also contributeto the inflammation associated with insulin resist-ance causedbyobesity. Indeed, JNK inmacrophagesis required for the accumulation of inflammatorytissue macrophages and insulin resistance causedby diet-induced obesity. Drug-mediated targetingof macrophage-expressed JNK therefore repre-sents a potential therapeutic approach to suppressinflammation that may be applicable to the treat-ment of inflammatory disorders.

References and Notes1. K. M. Flegal, B. I. Graubard, D. F. Williamson, M. H. Gail,

JAMA 298, 2028 (2007).2. M. Qatanani, M. A. Lazar, Genes Dev. 21, 1443 (2007).3. F. O. Martinez, A. Sica, A. Mantovani, M. Locati,

Front. Biosci. 13, 453 (2008).4. M. E. Shaul, G. Bennett, K. J. Strissel, A. S. Greenberg,

M. S. Obin, Diabetes 59, 1171 (2010).5. S. P. Weisberg et al., J. Clin. Invest. 112, 1796 (2003).6. C. N. Lumeng, J. L. Bodzin, A. R. Saltiel, J. Clin. Invest.

117, 175 (2007).7. C. N. Lumeng, J. B. DelProposto, D. J. Westcott,

A. R. Saltiel, Diabetes 57, 3239 (2008).8. D. Patsouris et al., Cell Metab. 8, 301 (2008).

9. K. Kang et al., Cell Metab. 7, 485 (2008).10. J. I. Odegaard et al., Cell Metab. 7, 496 (2008).11. J. I. Odegaard et al., Nature 447, 1116 (2007).12. G. Sabio, R. J. Davis, Trends Biochem. Sci. 35, 490 (2010).13. G. Solinas et al., Cell Metab. 6, 386 (2007).14. G. Sabio et al., Science 322, 1539 (2008).15. S. N. Vallerie, M. Furuhashi, R. Fucho, G. S. Hotamisligil,

PLoS ONE 3, e3151 (2008).16. J. Hirosumi et al., Nature 420, 333 (2002).17. R. J. Davis, Cell 103, 239 (2000).18. A. Mantovani et al., Trends Immunol. 25, 677 (2004).19. H. Kitade et al., Diabetes 61, 1680 (2012).20. J. Wang, H. Beekhuizen, R. van Furth, Clin. Exp. Immunol.

95, 263 (1994).21. D. D. Yang et al., Immunity 9, 575 (1998).22. C. Dong et al., Science 282, 2092 (1998).23. C. Dong et al., Nature 405, 91 (2000).24. M. Das et al., Cell 136, 249 (2009).

Acknowledgments: We thank T. Barrett, V. Benoit, Y. Lee,and J.-H. Liu for technical assistance; S. Vernia for designingTaqman probes; and K. Gemme for administrative assistance.These studies were supported by grants from the NationalInstitutes of Health (DK090963, DK080756, and CA065861).The University of Massachussetts Mouse Metabolic PhenotypingCenter is supported by grant DK093000. M.S.H. wassupported by a postdoctoral fellowship (7-10-BETA-02) fromthe American Diabetes Association. R.J.D. and J.K.K. aremembers of the University of Massachussetts Diabetes andEndocrinology Research Center (DK032520). R.J.D. andR.A.F. are Investigators of the Howard Hughes MedicalInstitute. Mouse strains described are available to qualifiedinvestigators upon completion of a standard institutionalmaterials transfer agreement. The data presented in thismanuscript are tabulated in the main paper and in thesupplementary materials.

Supplementary Materialswww.sciencemag.org/cgi/content/full/science.1227568/DC1Materials and MethodsFigs. S1 to S19References (2530)

17 July 2012; accepted 6 November 2012Published online 6 December 2012;10.1126/science.1227568

Influence of Threonine Metabolismon S-Adenosylmethionine andHistone MethylationNg Shyh-Chang,1,2,3,4,5,6 Jason W. Locasale,5,6* Costas A. Lyssiotis,5,6 Yuxiang Zheng,5,6 Ren Yi Teo,1

Sutheera Ratanasirintrawoot,1,2,3 Jin Zhang,1,2,3 Tamer Onder,1,2,3 Juli J. Unternaehrer,1,2,3

Hao Zhu,1,2,3 John M. Asara,5 George Q. Daley,1,2,3,4 Lewis C. Cantley5,6

Threonine is the only amino acid critically required for the pluripotency of mouse embryonicstem cells (mESCs), but the detailed mechanism remains unclear. We found that threonine andS-adenosylmethionine (SAM) metabolism are coupled in pluripotent stem cells, resulting inregulation of histone methylation. Isotope labeling of mESCs revealed that threonine provides asubstantial fraction of both the cellular glycine and the acetylcoenzyme A (CoA) needed forSAM synthesis. Depletion of threonine from the culture medium or threonine dehydrogenase (Tdh)from mESCs decreased accumulation of SAM and decreased trimethylation of histone H3 lysine4 (H3K4me3), leading to slowed growth and increased differentiation. Thus, abundance of SAMappears to influence H3K4me3, providing a possible mechanism by which modulation of ametabolic pathway might influence stem cell fate.

Connections between pluripotency and themetabolic state of mouse embryonic stemcells (mESCs) are only beginning to be described (1). Metabolically, mESCs are charac-terized by a distinct mode of amino acid catabo-lism driven by the enzyme that catalyzes threonine

(Thr) oxidation, threonine dehydrogenase (Tdh)(1). The abundance of Tdh in mESCs is morethan 200 times that in mouse embryonic fibroblasts(MEFs), and Thr restriction or Tdh inhibitionabolishes mESC growth (13). Thus, Thr oxida-tion appears to be critical for mESCs, but therelationship between downstream metabolitesand the pluripotent state remains unclear.

To investigate how metabolism is alteredupon reprogramming to the pluripotent state,

1Stem Cell Transplantation Program, Division of PediatricHematology and Oncology, Manton Center for Orphan DiseaseResearch, Boston Childrens Hospital and Dana-Farber CancerInstitute, Boston, MA 02115, USA. 2Department of BiologicalChemistry and Molecular Pharmacology, Harvard MedicalSchool, Boston, MA 02115, USA. 3Harvard Stem Cell Institute,Cambridge, MA 02138, USA. 4Howard Hughes Medical Insti-tute, Harvard Medical School, Boston, MA 02115, USA. 5De-partment of Medicine, Division of Signal Transduction, BethIsrael Deaconess Medical Center, Boston, MA 02215, USA.6Department of Systems Biology, Harvard Medical School,Boston, MA 02115, USA.

*Present address: Division of Nutritional Sciences, CornellUniversity, Ithaca, NY 14850, USA.To whom correspondence should be addressed. E-mail:george.daley@childrens.harvard.edu (G.Q.D.); lewis_cantley@hms.harvard.edu (L.C.C.)

11 JANUARY 2013 VOL 339 SCIENCE www.sciencemag.org222

REPORTS

on A

ugus

t 18,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

we used liquid chromatographybased tandemmass spectrometry (LC-MS/MS) (4) to profilemetabolomic changes during reprogramminginto induced pluripotent stem cells (iPSCs). WeusedMEFs transfectedwith doxycycline-activatedtransgenes encoding Oct4, Sox2, Klf4, and Myc(iOSKM) (2, 5) and profiled metabolism duringreprogramming. Hierarchical clustering basedon the abundance of each metabolite revealedthat iPSCs were similar to mESCs and were dis-tinct from MEFs (Fig. 1, A and B). Inspection ofthe top metabolites regulated by OSKM-inducedreprogramming revealed a large number of gly-colytic intermediates (fig. S1A and table S1),many of which were significantly more abundantwithin just 4 days of reprogramming (Fig. 1C),when proliferation was accelerated but the cellswere not yet pluripotent (2). Thus, increases inglycolytic intermediates occurred before theacquisition of pluripotency (Fig. 1C). The othertop metabolites regulated during reprogrammingare involved in Thr and S-adenosylmethionine(SAM) metabolism. In contrast to glycolytic in-termediates, Thr- and SAM-related metaboliteschanged late in reprogramming, which suggeststhat enhanced Thr and SAM metabolism ac-

companied acquisition of the pluripotent state(Fig. 1D). Upon reprogramming, some of thelargest decreases were in Thr, Cys, and folate,and the largest increases were in SAM andcystathionine.

Metabolism is also rewired during mESCdifferentiation. In mESCs, the let-7 microRNApromotes differentiation by suppressing the plu-ripotency network, whereas Lin28a promotes plu-ripotency by repressing let-7 (6, 7). Hence, weprofiled metabolic changes in mESCs acutely af-ter dox-induced activation of a transgene encod-ing let-7 or Lin28a (iLet-7 or iLin28a) while thecell cycle remained unperturbed (fig. S1B). Lin28aand let-7 regulated 38metabolites in a reciprocalmanner (fig. S1C), among which were a largenumber of Thr and SAMmetabolites (Fig. 1E);these findings provide further support for the ideathat Thr and SAM metabolism are coupled topluripotency.

We integrated our metabolomics data onmESCs (table S1) with cDNA microarray dataon mESCs (2) using the KEGG database ofmetabolic networks. This analysis showed thatmany of the metabolic enzymes that channel Thrmetabolism into SAM metabolism (e.g., Tdh,

Gcat, Gldc, Dhfr, Fol1r, Mat2a/b, and Ahcy) aremore abundant inmESCs than inMEFs (Fig. 2A).Consistent with the enzyme expression pat-terns, several Thr-SAM pathway inputs suchas Thr, Cys, and folate were less abundant inmESCs than in MEFs, whereas downstreamoutputs such as SAM and cystathionine weremore abundant inmESCs than inMEFs (Fig. 1D).Thus, a Thr-SAM pathway appears to be acti-vated in mESCs.

To test this pathway, we traced the metabolicfate of [U-14C]Thr inmESCswith high-performanceliquid chromatography (HPLC). 14C was incor-porated into Gly and Glu, indicating that Thr wasused to synthesize these amino acids (Fig. 2B).In contrast, MEFs incubated with [U-14C]Thrdid not exhibit Thr catabolism (fig. S2A). Wealso traced the fate of [U-13C]Thr in mESCswith LC-MS/MS metabolomics (fig. S2, B to F,and table S1). mESCs used Thr to synthesizeacetyl-CoAderived tricarboxylic acid (TCA)cycle intermediates (Fig. 2, C and D). At steadystate, [U-13C]Thr contributed ~20% of the citratevia acetyl-CoA, whereas [U-13C]glucose contrib-uted ~35% via acetyl-CoA (+2 isotopomer). Thus,Thr contributes significantly to the acetyl-CoA

Fig. 1. The Thr-SAM pathway is ac-tivated by pluripotency factors. (A)Heat map showing relative abun-dance ofmetabolites in iOSKM-MEFsin the absence of doxycycline treat-ment (no dox); 4 days after dox (dox4D); 18 days after dox, whereuponiPSC clones are fully reprogrammed(dox 18D); andmESCs, normalized tototal biomass (n = 3), and measuredusing selected reaction monitoring(SRM) and LC-MS/MS. (B) Scatterplotof metabolite intensities in mESCs(x axis) versus iPSCs (y axis). Metab-olite intensities were obtained fromthe integrated total ion current froma single SRM transition. (C) SRManalysis of abundance in glycolyticintermediates in iOSKM-MEFs dur-ing reprogramming (n = 3). *P