EDITORIAL Reddy HDAC inhibition begets more MDSCs phages after LPS and IL-4 stimulation, REFERENCES macrophage apoptosis requires activation of 1. Mosser, D. M., Edwards, J. P. (2008) Explor- ornithine decarboxylase by c-Myc. J. Biol. respectively, were identical in WT and Chem. 280, 22492–22496. ⌬ ing the full spectrum of macrophage activa- Arg1 macrophages. Therefore, the au- tion. Nat. Rev. Immunol. 8, 958–969. 9. Zhang, M., Caragine, T., Wang, H., Cohen, thors conclude that “-1-indepen- 2. Van den Bossche, J., Lamers, W. H., Koehler, P. S., Botchkina, G., Soda, K., Bianchi, M., E. S., Geuns, J. M. C., Alhonen, L., Uimari, Ulrich, P., Cerami, A., Sherry, B., Tracey, dent polyamine production stimulates A., Pirnes-Kahru, S., Van Overmeire, E., Mo- K. J. (1997) Spermine inhibits proinflamma- the expression of alternatively activated rias, Y., Brys, L., Vereecke, L., De Baetselier, tory cytokine synthesis in human mononu- P., Van Ginderachter, J. A. (2012) Arginase-1- clear cells: a counterregulatory mechanism macrophage markers”. This point puts independent polyamine production stimu- that restrains the immune response. J. Exp. into question the use of Arg1 as a lates the expression of IL-4-induced alterna- Med. 185, 1759–1768. tively activated macrophage markers while 10. Hasko, G., Kuhel, D. G., Marton, A., Nemeth, marker of alternatively activated macro- inhibiting LPS-induced expression of inflam- Z. H., Deitch, E. A., Szabo, C. (2000) Sperm- phages. The authors imply that there is matory . J. Leukoc. Biol. 91, 685–699. ine differentially regulates the production of 3. Ruan, H., Hill, J. R., Fatemie-Nainie, S., Mor- interleukin-12 p40 and interleukin-10 and a mechanism whereby polyamine syn- suppresses the release of the T helper 1 cyto- ris, D. R. (1994) Cell-specific translational ␥ thesis occurs through a metabolic path- regulation of S-adenosylmethionine decarbox- kine interferon- . Shock 14, 144–149. 11. Kepka-Lenhart, D., Mistry, S. K., Wu, G., Mor- way that does not involve arginase activ- ylase mRNA. Influence of the structure of the 5Ј transcript leader on regulation by the up- ris Jr., S. M. (2000) Arginase I: a limiting fac- ity, but this theoretical pathway remains stream open reading frame. J. Biol. Chem. 269, tor for nitric oxide and polyamine synthesis by activated macrophages? Am. J. Physiol. 279, to be elucidated. Regardless of which 17905–17910. 4. Bussiere, F. I., Chaturvedi, R., Cheng, Y., R2237–R2242. pathway(s) may be used by macro- Gobert, A. P., Asim, M., Blumberg, D. R., Xu, 12. Louis, C. A., Mody, V., Henry Jr., W. L., Reichner, J. S., Albina, J. E. (1999) Regula- phages for polyamine biosynthesis, the H., Kim, P. Y., Hacker, A., Casero Jr., R. A., Wilson, K. T. (2005) Spermine causes loss of tion of arginase isoforms I and II by IL-4 in data in this manuscript demonstrate innate immune response to Helicobacter pylori cultured murine peritoneal macrophages. Am. J. Physiol. 276, R237–R242. that endogenous polyamines play a ma- by inhibition of inducible nitric-oxide syn- thase translation. J. Biol. Chem. 280, 2409– 13. Pesce, J. T., Ramalingam, T. R., Mentink- jor function in regulation of macro- 2412. Kane, M. M., Wilson, M. S., El Kasmi, K. C., phage activation. 5. Flamigni, F., Stanic, I., Facchini, A., Cetrullo, Smith, A. M., Thompson, R. W., Cheever, S., Tantini, B., Borzi, R. M., Guarnieri, C., A. W., Murray, P. J., Wynn, T. A. (2009) Argi- Caldarera, C. M. (2007) Polyamine biosynthe- nase-1-expressing macrophages suppress Th2 sis as a target to inhibit apoptosis of non-tu- cytokine-driven inflammation and fibrosis. PLoS Pathog. 5, e1000371. ACKNOWLEDGMENTS moral cells. Amino Acids 33, 197–202. 6. Pello, O. M., De Pizzol, M., Mirolo, M., 14. Chaturvedi, R., Asim, M., Hoge, S., Lewis, Soucek, L., Zammataro, L., Amabile, A., N. D., Singh, K., Barry, D. P., de Sablet, T., This work was supported by NIH grants Doni, A., Nebuloni, M., Swigart, L. B., Evan, Piazuelo, M. B., Sarvaria, A. R., Cheng, Y., R01DK053620, R01AT004821, G. I., Mantovani, A., Locati, M. (2012) Role Closs, E. I., Casero, R. A., Jr., Gobert, A. P., of c-MYC in alternative activation of human Wilson, K. T. (2010) Polyamines impair im- P01CA116087, and P01CA028842; by macrophages and tumor-associated macro- munity to Helicobacter pylori by inhibiting L- the Vanderbilt Digestive Disease Re- phage biology. Blood 119, 411–421. arginine uptake required for nitric oxide pro- 7. Lewis, N. D., Asim, M., Barry, D. P., de Sab- duction. Gastroenterology 139, 1686–1698. search Center grant P30DK058404; and let, T., Singh, K., Piazuelo, M. B., Gobert, by a Merit Review grant from the Office A. P., Chaturvedi, R., Wilson, K. T. (2011) of Medical Research, Department of Immune evasion by Helicobacter pylori is medi- ated by induction of macrophage arginase II. KEY WORDS: Veterans Affairs (all to K.T.W.). A.P.G. J. Immunol. 186, 3632–3641. ⅐ ⅐ 8. Cheng, Y., Chaturvedi, R., Asim, M., Bussiere, arginase ornithine decarboxylase classi- is supported in part by the Philippe F. I., Scholz, A., Xu, H., Casero Jr., R. A., Wil- cal activation ⅐ alternative activation ⅐ IL-4 Foundation. son, K. T. (2005) Helicobacter pylori-induced ⅐ LPS

Editorial: HDAC inhibition begets more MDSCs Pavan Reddy1 Department of Internal Medicine, University of Michigan Comprehensive Cancer Center, Ann Arbor, Michigan, USA RECEIVED NOVEMBER 3, 2011; REVISED NOVEMBER 22, 2011; ACCEPTED DECEMBER 6, 2011. DOI: 10.1189/jlb.1111541 ‹ SEE CORRESPONDING ARTICLE ON PAGE 701

merging data demonstrate that cytotoxic doses [1, 2]. The mecha- HDACs remove acetyl groups from HDACi, initially developed as nisms and promise of HDACi-medi- ␧-N-acetyl lysine amino acids on his- E anticancer agents, are also po- ated immune modulation are tone tails and regulate chromatin tent anti-inflammatory agents at non- increasingly understood [1]. In this structure and dynamics [5–7]. Emerg- issue, Rosborough and colleagues [3] ing data also show that in addition to report a novel role for HDACi in regu- regulating of lysines within

Abbreviations: BMϭbone marrow, GVHDϭgraft- lating immune responses (Fig. 1). versus-host disease, HDACiϭhistone deacety- They demonstrate that HDACi en- 1. Correspondence: Department of Internal lase inhibitors, HSCϭhematopoietic stem cell, hance the generation and expansion Medicine, University of Michigan Comprehen- MDSCϭmyeloid-derived suppressor cell, of MDSCs, a key subset of regulatory sive Cancer Center, 3312 CCC, 1500 East Med- SAHAϭsuberoylinalide hydroxamic acid, ical Center Dr., Ann Arbor, MI 48109-0942, Tregϭregulatory T cell, TSAϭtrichostatin-A APCs [4]. USA. E-mail: [email protected]

www.jleukbio.org Volume 91, May 2012 Journal of Leukocyte Biology 679 SUMMARY: mechanisms of suppression of T cell re- • HDAC inhibition regulates innate responses and DC functions sponses by MDSCs include a high level of • Inhibition of HDAC in GMCSF treated BM progenitors impairs DC but promotes arginase activity, NO, ROS production, MDSC differentiation induction of Tregs, secretion of TGF-␤, • The MDSCs generated by HDAC inhibition demonstrate the expected in vitro suppressive depletion of cysteine, and up-regulation functions that are dependent on iNOS and HO-1 of PGE2. The distinct subsets of MDSCs • HDAC inhibition augments MDSCs numbers in vivo use differential pathways for regulating immune responses [4, 22, 23]. For exam- IMPLICATIONS: ple, granulocytic MDSCs are reported to • Insight into HDAC inhibition mediated immune regulation be dependent on ROS, whereas mono-

• Provide a platform for ex vivo expansion of MDSCs that may be exploited therapeutically cytic MDSCs are more dependent on argi- nase and NO for regulating T cell re- • The stability, and functional relevance of the in vivo expansion of MDSCs remain to be analyzed sponses [4]. The powerful immune-suppressive • The specific HDACs and HATs involved in MDSC differentiation need to be determined features of MDSC make them attrac- • The critical epigenetic and the non- that must be acetylated need to be understood tive candidates for use in cell therapy to reduce unwanted and exuberant Figure 1. Summary and implications. immune responses, such as in autoim- munity, graft rejection, and GVHD [23, 24]. To facilitate such studies, it , HDAC enzymes regulate the activity [7]. They have been shown to would be essential to generate, ex acetylation of several nonhistone pro- regulate the function of various im- vivo, relatively large and stable im- teins [8–11]. They are classified into mune cells and modulate in vivo dis- mune-suppressive MDSCs. Murine four main classes: class I HDACs in- ease states in experimental models [1, studies have demonstrated that G-CSF clude HDAC1, -2, -3, and -8; class II 17]. Specifically, their impact on Tregs expands MDSCs in vitro and in vivo HDACs are HDAC4, -5, -7, and -9 and T cell responses is being appreci- [23]. Rosborough et al. [3] analyzed (class IIa) and HDAC6 and -10 (class ated increasingly [13, 18]. Recent data the impact of HDAC inhibition on the IIb); class III HDACs are homologs of have demonstrated the ability of these generation and function of MDSCs. yeast silent information regulator 2 agents to inhibit the production of They demonstrate that HDAC inhibi- proteins; and class IV HDAC is multiple proinflammatory cytokines tion of BM GM-CSF-treated cultures HDAC11 [11]. Class I HDAC enzymes from various APCs [1]. HDAC inhibi- with TSA or SAHA expanded the HSC are expressed in most cells and are tion has been shown to enhance IDO and progenitor compartments, skewed largely, but not exclusively, restricted expression in DCs, promote conver- myeloid differentiation, impaired the to the nucleus. By contrast, class II sion of inflammatory macrophages development of DCs, and enhanced HDAC enzymes demonstrate a more into a tolerogenic phenotype, decrease the generation of MDSCs. Although restricted, tissue-specific expression TLR signaling, reduce costimulatory addition of rIL-4 to the cultures ex- and shuttle between the nucleus and molecule expression, and increase panded the progenitor cells and dem- cytoplasm [7, 12]. Most HDACi inhibit IL-10 expression [19–22]. But, the onstrated skewed myeloid differentia- class I and II enzymes with varying ef- mechanisms of action of HDACi on tion, it did not enhance the genera- ficiency [11]. The catalytic activities of different APC subsets and their gener- tion of MDSCs. Upon further the class I and II HDACs differ [7]. ation and function are not understood characterization, they found that these ϩ Class I HDAC enzymes exhibit strong completely. MDSCs expressed CD11b F4/80int deacetylase activity, whereas most class MDSCs are now being appreciated in- and Ly-6Chigh, suggesting generation II HDACs are enzymatically less active creasingly as key APC subsets that are re- of a monocytic MDSC phenotype. and act primarily as scaffolding pro- sponsible for regulating immune re- These cells demonstrated equivalent teins within large multimolecular com- sponses. They potently suppress T effector suppression of allogeneic T cell re- plexes. cell responses while enhancing Tregs sponses in vitro. However, in contrast HDACi belong to different classes of [23]. They are a heterogeneous popula- to control MDSCs, those derived fol- drugs with distinct chemical structures tion of immature myeloid cells that con- lowing HDAC inhibition showed re- and abilities to inhibit HDAC enzymes sists of myeloid progenitors and precur- duced expression of arginase, iNOS, [11, 13–15]. The two HDACi, used by sors [4]. MDSCs are identified in murine and HO-1. Furthermore, arginase was Rosborough et al. [3]—TSA and studies as cells that are positive for CD11b not required, whereas iNOS and HO-1 SAHA—are nonselective, pan-HDACi and Gr-1. Based on expression of Ly-6C activity was required for the suppres- that inhibit class I and II [11, 16]. or Ly-6G, they can be characterized fur- sive effects on allogeneic T cells. Im- HDACi can also modulate the acetyla- ther as monocytic MDSCs (CD11bϩ Ly- portantly, HDAC inhibition also aug- tion of HDAC proteins themselves, 6GϪ Ly-6Chigh) or granulocytic MDSCs mented in vivo expansion of MDSCs causing alterations in their stability or (CD11bϩ Ly-6Gϩ Ly-6Clow) [23]. The in BM and spleen by GM-CSF, al-

680 Journal of Leukocyte Biology Volume 91, May 2012 www.jleukbio.org EDITORIAL Reddy HDAC inhibition begets more MDSCs though the functional impact of this ACKNOWLEDGMENTS deacetylase homologue bound to the TSA and SAHA inhibitors. Nature 401, 188–193. in vivo expansion was not addressed. 17. Reddy, P., Maeda, Y., Hotary, K., Liu, C., This study, like all interesting and P.R. was supported by NIH grants AI- Reznikov, L. L., Dinarello, C. A., Ferrara, J. L. 075284, HL-090775, and CA-143379. (2004) inhibitor suberoylani- seminal observations, while illuminat- lide hydroxamic acid reduces acute graft-versus- ing a role for HDAC inhibition in the host disease and preserves graft-versus-leukemia generation of MDSCs, also raises sev- effect. Proc. Natl. Acad. Sci. USA 101, 3921–3926. 18. Tao, R., de Zoeten, E. F., Ozkaynak, E., Chen, C., eral additional questions. Why did the REFERENCES Wang, L., Porrett, P. M., Li, B., Turka, L. A., Ol- addition of IL-4 prevent the increase 1. Dinarello, C. A., Fossati, G., Mascagni, P. son, E. N., Greene, M. I., Wells, A. D., Hancock, (2011) Histone deacetylase inhibitors for W. W. (2007) Deacetylase inhibition promotes the in MDSC generation despite the in- treating a spectrum of diseases not related to generation and function of regulatory T cells. Nat. crease in HSC and progenitors? Is the cancer. Mol. Med. 17, 333–352. Med. 13, 1299–1307. 2. Leoni, F., Zaliani, A., Bertolini, G., Porro, 19. Reddy, P., Sun, Y., Toubai, T., Duran-Struuck, increase in HSCs and progenitors criti- G., Pagani, P., Pozzi, P., Dona`, G., Fossati, R., Clouthier, S. G., Weisiger, E., Maeda, Y., cal and relevant? Or is it merely an G., Sozzani, S., Azam, T., Bufler, P., Fan- Tawara, I., Krijanovski, O., Gatza, E., Liu, C., tuzzi, G., Goncharov, I., Kim, S. H., Pomer- Malter, C., Mascagni, P., Dinarello, C. A., Fer- epiphenomenon? The developmental antz, B. J., Reznikov, L. L., Siegmund, B., rara, J. L. (2008) Histone deacetylase inhibi- pathways that are critical for MDSC Dinarello, C. A., Mascagni, P. (2002) The tion modulates indoleamine 2,3-dioxygenase- antitumor histone deacetylase inhibitor sub- dependent DC functions and regulates exper- generation remain largely unknown, eroylanilide hydroxamic acid exhibits anti- imental graft-versus-host disease in mice. and how would those pathways af- inflammatory properties via suppression of J. Clin. Invest. 118, 2562–2573. cytokines. Proc. Natl. Acad. Sci. USA 99, 20. Roger, T,. Lugrin, J., Le Roy, D., Goy, G., Mom- fected by HDAC inhibition remain to 2995–3000. belli, M., Koessler, T., Ding, X. C., Chanson, A. L., be deciphered? What would be the 3. Rosborough, B. R., Castellaneta, A., Natara- Reymond, M. K., Miconnet, I., Schrenzel, J., Fran- impact on the phenotype of jan, S., Thomson, A. W., Turnquist, H. R. c¸ois, P., Calandra, T. (2011) Histone deacetylase (2011) Histone deacetylase inhibition facili- inhibitors impair innate immune responses to MDSCs, especially in vivo? Is that func- tates GM-CSF-mediated expansion of myeloid- Toll-like receptor agonists and to infection. Blood tionally relevant? An intriguing obser- derived suppressor cells in vitro and in vivo. 117, 1205–1217. J. Leukoc. Biol., 91, 701–709. 21. Brogdon, J. L., Xu, Y., Szabo, S. J., An, S., Buxton, vation is that the differential mecha- 4. Condamine T, Gabrilovich, D. I. (2011) Mo- F., Cohen, D., Huang, Q. (2007) Histone deacety- nisms that might be used by the lecular mechanisms regulating myeloid-de- lase activities are required for innate immune cell rived suppressor cell differentiation and func- control of Th1 but not Th2 effector cell function. HDACi induced MDSCs. The role of tion. Trends Immunol. 32, 19–25. Blood 109, 1123–1130. arginase, iNOS, and HO-1, in addition 5. Narlikar, G. J., Fan, H. Y., Kingston, R. E. 22. Umemura, N., Saio, M., Suwa, T., Kitoh, Y., (2002) Cooperation between complexes that Bai, J., Nonaka, K., Ouyang, G. F., Okada, M., to the other pathways, such as induc- regulate chromatin structure and transcrip- Balazs, M., Adany, R., Shibata, T., Takami, T. tion of Tregs, relevance of cysteine tion. Cell 108, 475–487. (2008) Tumor-infiltrating myeloid-derived 6. Kouzarides, T. (2007) Chromatin modifica- suppressor cells are pleiotropic-inflamed depletion, and PGE2, remains to be tions and their function. Cell 128, 693–705. monocytes/macrophages that bear M1- and understood as well [23, 25–27]. Fur- 7. Yang, X-J., Seto, E. (2008) Lysine acetylation: M2-type characteristics. J. Leukoc. Biol. 83, codified crosstalk with other posttranslational 1136–1144. thermore, the key molecular mecha- modifications. Mol. Cell 31, 449–461. 23. Gabrilovich, D. I., Nagaraj, S. (2009) My- nisms remain to be explored. In addi- 8. Kouzarides, T. (2000) Acetylation: a regula- eloid-derived suppressor cells as regulators tory modification to rival phosphorylation? tion to histone deacetylation and epi- of the immune system. Nat. Rev. Immunol. 9, EMBO J. 19, 1176–1179. 162–174. genetic alterations, is acetylation of 9. Yuan, Z. L., Guan, Y. J., Chatterjee, D., Chin, 24. Highfill, S. L., Rodriguez, P. C., Zhou, Q., nonhistone proteins critical? Further Y. E. (2005) Stat3 dimerization regulated by Goetz, C. A., Koehn, B. H., Veenstra, R., reversible acetylation of a single lysine resi- Taylor, P. A., Panoskaltsis-Mortari, A., Se- identification of the specific HDAC due. Science 307, 269–273. rody, J. S., Munn, D. H., Tolar, J., Ochoa, , the specific histone acetyl- 10. Sun, Y., Chin, Y. E., Weisiger, E., Malter, C., A. C., Blazar, B. R. (2010) Bone marrow Tawara, I., Toubai, T., Gatza, E., Mascagni, P., myeloid-derived suppressor cells (MDSCs) , and their putative targets Dinarello, C. A., Reddy, P. (2009) Cutting edge: inhibit graft-versus-host disease (GVHD) via in generating MDSCs will refine our negative regulation of dendritic cells through an arginase-1-dependent mechanism that is acetylation of the nonhistone STAT-3. up-regulated by interleukin-13. Blood 116, understanding of the role of protein J. Immunol. 182, 5899–5903. 5738–5747. acetylation and MDSC biology. 11. Johnstone, R. W. (2002) Histone-deacetylase 25. De Wilde, V., Van Rompaey, N., Hill, M., inhibitors: novel drugs for the treatment of Lebrun, J. F., Lemaıˆtre, P., Lhomme´, F., Kub- This study expands the scope of cancer. Nat. Rev. Drug Discov. 1, 287–299. jak, C., Vokaer, B., Oldenhove, G., Charbon- HDAC inhibition-mediated immune 12. Haberland, M., Montgomery, R. L., Olson, nier, L. M., Cuturi, M. C., Goldman, M., Le E. N. (2009) The many roles of histone regulation and provides a novel Moine, A. (2009) Endotoxin-induced my- deacetylases in development and physiology: eloid-derived suppressor cells inhibit alloim- method for generating MDSCs with implications for disease and therapy. Nat. mune responses via heme oxygenase-1. Am. J. Rev. Genet. 10, 32–42. greater efficiency, both in vitro and in Transplant. 9, 2034–2047. 13. Beier, U. H., Akimova, T., Liu, Y., Wang, L., 26. Lu, T., Ramakrishnan, R., Altiok, S., Youn, vivo. It provides texture to our current Hancock, W. W. (2011) Histone/protein J. I., Cheng, P., Celis, E., Pisarev, V., Sher- deacetylases control Foxp3 expression and man, S., Sporn, M. B., Gabrilovich, D. (2011) understanding of the role of HDAC the heat shock response of T-regulatory cells. Tumor-infiltrating myeloid cells induce tumor Curr. Opin. Immunol. 23, 670–678. inhibition in regulating immune re- cell resistance to cytotoxic T cells in mice. 14. Beier, U. H., Wang, L., Bhatti, T. R., Liu, Y., J. Clin. Invest. 121, 4015–4029. sponses. Importantly, in light of the Han, R., Ge, G., Hancock, W. W. (2011) Sir- 27. Obermajer, N., Muthuswamy, R., Lesnock, tuin-1 targeting promotes Foxp3ϩ T-regula- known immune-regulatory effects of J., Edwards, R. P., Kalinski, P. (2011) Posi- tory cell function and prolongs allograft sur- tive feedback between PGE2 and COX2 re- MDSCs, the observations by Rosbor- vival. Mol. Cell. Biol. 31, 1022–1029. ough and colleagues [3] may pave way 15. De Zoeten, E. F., Wang, L., Butler, K., Beier, directs the differentiation of human den- U. H., Akimova, T., Sai, H., Bradner, J. E., dritic cells towards stable myeloid-derived for building a platform to robustly ex- Mazitschek, R., Kozikowski, A. P., Matthias, suppressor cells. Blood 118, 5498–5505. pand MDSCs ex vivo and thus, facili- P., Hancock, W. W. (2011) Histone deacety- lase 6 and heat shock protein 90 control the tate well-designed, adoptive cell-ther- functions of Foxp3(ϩ) T-regulatory cells. apy trials to study the potential of Mol. Cell. Biol. 31, 2066–2078. 16. Finnin, M. S., Donigian, J. R., Cohen, A., Richon, KEY WORDS: MDSCs in regulating autoimmunity, V. M., Rifkind, R. A., Marks, P. A., Breslow, R., histone deacetylases ⅐ myeloid cells ⅐ im- allograft rejection, and GVHD. Pavletich, N. P. (1999) Structures of a histone mune response

www.jleukbio.org Volume 91, May 2012 Journal of Leukocyte Biology 681