A New Alpha in Line Between KRAS and NF-Kb Activation?

VIEWS

IN THE SPOTLIGHT A New Alpha in Line Between KRAS and NF-kB Activation? Chorom Pak 1 , 2 , 3 and Shigeki Miyamoto 1 , 2 Summary: Bang and colleagues report a novel role for GSK-3α, rather than the well-studied GSK-3β, as the link between oncogenic KRAS and the canonical and noncanonical activation pathways of NF-κB in pancreatic cancer. Although the mechanism through which it promotes noncanonical activation remains unclear, the authors show that GSK-3α binds and stabilizes TAK1-TAB complexes to constitutively activate canonical NF-κB signaling. Consequently, the inhibition of GSK-3α retards pancreatic cancer growth in vitro and in vivo , thereby revealing this relatively less-studied kinase as a potential therapeutic target for treatment of KRAS-positive pancreatic cancer. Cancer Discov; 3(6); 613–5. ©2013 AACR. See related article by Bang et al., p. 690 (2).

KRAS-activating mutations have long been implicated in dif- metabolism, Wnt and Hedgehog signal transduction, protein ferent human cancers, particularly a form of pancreatic cancer synthesis, mitosis, and apoptosis ( 5 ). There are 2 isoforms of called pancreatic ductal adenocarcinoma (PDAC). However, GSK-3, α and β; they have masses of 51 and 47 kDa, respec- attempts to target KRAS therapeutically have proven to be a tively, and are encoded by different genes. GSK-3α contains a challenge. Therefore, understanding downstream effectors may glycine-rich region at the N-terminus that GSK-3β does not. illuminate new targets and strategies. Previous studies have The 2 isoforms share 98% identity within their kinase domains; shown that activated KRAS turns on many downstream sign- however, they share only 36% identity in the 76 C-terminal resi- aling pathways, including phosphoinositide 3-kinase (PI3K)/ dues. GSK-3β also has a minor splice variant, GSK-3β2, which AKT, mitogen-activated protein kinase (MAPK), and NF-κB ( 1 ). contains a 13-residue insert within the kinase domain. Most Here, Bang and colleagues ( 2 ) report a new mediator of onco- of the literature on GSK-3 focuses on the GSK-3β isoform, genic KRAS in PDAC, GSK-3α, that activates not only canoni- rather than GSK-3α. Gsk3b- null mice are embryonic lethal due cal but also noncanonical NF-κB signaling pathways, thereby to liver degeneration caused by TNF-α toxicity, whereas Gsk3a- potentially revealing a new therapeutic target for PDAC. null mice are viable but display enhanced glucose and insulin It has previously been shown that KRAS activates NF-κB, sensitivity, as well as reduced fat mass ( 6 ). These fi ndings which is crucial for cell survival and tumor transformation, clearly highlight different physiologic roles that these proteins along with concomitant p53 loss ( 3 ). Others have attempted to play. GSK-3β has been reported to mediate NF-κB activation understand the intermediate players between KRAS-activating and gene transcription to promote pancreatic cancer, oste- mutations and NF-κB signaling. It has also been shown that osarcoma, oral cancer, certain leukemias, and gliomas, as well the signaling adaptor p62 is required for both RAS-induced as other cancers ( 7, 8 ). The phenotype of Gsk3b- null mice is transformation and NF-κB activation in a lung tumor model remarkably similar to that of p65-defi cient and IKKβ-defi cient ( 1 ). The induction of p62 occurred through an ERK- and AKT- mice, and fi broblasts from these mice also showed reduced dependent pathway, which then activated TRAF6 E3-ubiquitin NF-κB function, further establishing the genetic link between ligase activity as well as NF-κB via phosphorylation of the GSK-3β and NF-κB signaling. However, the role of GSK-3α in upstream kinase IκB kinase (IKKα/β). TBK1, a serine-threonine pro-cancer effects is thus far limited to the context of acute kinase that can activate NF-κB, was also found to be a syntheti- myeloid leukemia ( 9 ), and no clear link to NF-κB signaling has cally lethal partner with mutant KRAS ( 4 ). Bang and colleagues been revealed. By using siRNA specifi c to each GSK-3 isoform, ( 2 ) now introduce GSK-3α into the fray. Bang and colleagues ( 2 ) found that growth of pancreatic cancer The study of glycogen synthase kinase-3 (GSK-3) spans over cell lines harboring a KRAS mutation (Panc-1 and MiaPaCa-2) 30 years since its discovery as one of several protein kinases that was reduced by inhibition of GSK-3α, whereas inhibition of can phosphorylate glycogen synthase. GSK-3 is now known to GSK-3β had much less effect. Thus, the authors expanded the be involved in myriad biologic functions, including glycogen limited knowledge of GSK-3α by implicating its striking pro- tumorigenic role through NF-κB signaling in pancreatic cancer cells harboring oncogenic KRAS mutation ( Fig. 1 ). Authors’ Affi liations: 1 Department of Oncology, McArdle Laboratory for How does GSK-3α activate NF-κB and promote growth of 2 3 Cancer Research, Carbone Cancer Center, and Molecular Cellular Phar- KRAS-mutant pancreatic cancer cells? Knockdown of GSK-3α, macology Graduate Program, University of Wisconsin-Madison, Madison, β Wisconsin but not GSK-3 , or treatment with the GSK-3 inhibitor AR– A014418, showed a decrease in TGF-β–activated kinase 1 Corresponding Author: Shigeki Miyamoto, University of Wisconsin- Madison, 6190 Wisconsin Institute for Medical Research, 1111 Highland (TAK1) levels, hinting at a novel mechanism at play. TAK1 Avenue, Madison, WI 53705. Phone: 608-262-9281; Fax: 608-262-1257; initiates downstream NF-κB and MAPK signaling in asso- E-Mail: [email protected] ciation with its TAK1-binding partners (TAB1–4) in response doi: 10.1158/2159-8290.CD-13-0193 to an array of extracellular signals and intracellular stress ©2013 American Association for Cancer Research. conditions. TAK1 leads to canonical activation of NF-κB by

JUNE 2013CANCER DISCOVERY | 613

NF-κB Oral cancer β-Adrenergic Liver cancer receptor NF-κB p53 Renal cell carcinoma β-Catenin/Wnt GSK-3α GSK-3β

β -Catenin/Wnt Melanoma N- gly rich Kinase -C vs. N- Kinase -C

CML β Pancreatic A plaques Alzheimer’s cancer AML Osteosarcoma

Gastrointestinal Pancreatic Glioblastoma cancer cancer

Figure 1. The roles of GSK-3α vs. GSK-3β in cancer. Much of the analysis of GSK-3 isoforms in the literature has been focused on GSK-3β, which is implicated in many cancer and disease types. In contrast, the analysis of its relative, GSK-3α, has been rather limited. The molecular targets of GSK-3 are depicted in yellow, and disease types are depicted in green. Bang and colleagues (2) introduced a new role for GSK-3α in the context of NF-κB signaling in pancreatic cancer harboring mutant KRAS oncogenes (left). directly phosphorylating IKKβ and activating the IKK com- How does GSK-3α association with TAK1 promote stabiliza- plex, which then phosphorylates the inhibitor of κB alpha tion and activation of the TAK1–TAB complex? Because the (IκBα). IκBα is then degraded, and active NF-κB is liberated. kinase activity of GSK-3α is important based on the chemical TAK1 has previously been shown to be involved in the survival inhibitor analysis, GSK-3α presumably phosphorylates target of colorectal cancers as well as pancreatic cancers (10, 11), but protein(s). As there is no consensus substrate motif in TAK1 the mechanistic details have not been well elucidated. Con- (S/T-XXX-S/T), the authors hypothesize that GSK-3α may sistent with GSK-3α being a downstream effector of KRAS, phosphorylate TAB1 or TAB2, each of which harbors the knockdown of KRAS in HPDE6KR+ cells reduced TAK1 pro- consensus site. Most of the time, GSK-3 requires a primed tein levels. GSK-3α was found in a TAK1–TAB1 complex, and substrate or a substrate that has already been phosphorylated silencing of either GSK-3α or KRAS led to reduced TAK1– at the fourth amino acid downstream of the target phospho- TAB1 interaction in HPDE6KR+ cells. As predicted, TAK1 rylation serine or threonine site; however, in some cases, such inhibition by RNAi resulted in diminished phosphorylation as β-catenin, this is not necessary. Other questions include: of IκBα and p65/RelA. TAK1 inhibition with 5Z-7-oxozaenol How does KRAS regulate the engagement of GSK-3α in the led to decreased proliferation in pancreatic cancer cells via TAK1–TAB complex? How does GSK-3α mediate the non- κ G2 –M arrest with only a modest increase in apoptosis. These canonical NF- B signaling pathway? The authors provided fi ndings led the authors to propose that GSK-3α activates evidence that this might occur within the nucleus. Why does TAK1–TAB complexes by directly binding and stabilizing the GSK-3β fail to contribute in a similar way to KRAS-positive complex to chronically stimulate canonical NF-κB signaling. pancreatic cancer? Is there any critical amino acid motif Surprisingly, the role of GSK-3α was not limited only to present in GSK-3α but missing in GSK-3β which controls canonical NF-κB signaling. Bang and colleagues ( 2 ) found that canonical and noncanonical NF-κB signaling? As mentioned the effect of TAK1 inhibition was consistently less than that above, there are signifi cant differences in primary sequences of GSK-3 inhibition, suggesting the presence of additional between GSK-3α and β. Is the target of the TAK1/TAB downstream target(s) of GSK-3α. Therefore, they looked to complex limited only to the canonical NF-κB pathway, or the noncanonical activation pathway of NF-κB, which has also is it phosphorylating other target substrates as well, such as been implicated in pancreatic cancer. The noncanonical activa- extracellular signal-related kinase (ERK) or c-jun -NH 2-kinase tion pathway involves NF-κB–inducing kinase activation of the (JNK)? Finally, is the new role of GSK-3α revealed by Bang IKK complex, which is composed of an IKKα dimer. The IKK and colleagues (2 ) limited only to KRAS-positive pancreatic complex phosphorylates p100, causing it to be processed to an cancer or is this also relevant to other cancer types? While active p52 subunit of NF-κB. Bang and colleagues ( 2 ) found that illuminating new therapeutic possibilities to target in pancre- depletion of GSK-3α, but again not GSK-3β, led to a decrease in atic cancer, the study by Bang and colleagues (2 ) also opened p100 to p52 processing in oncogenic KRAS-positive PDAC cells. up a new world of investigation into the less understood α Thus, the authors identifi ed 2 new modes of GSK-3α action in isoform of GSK-3 and its role in different signaling pathways pancreatic cancer cells, KRAS–GSK-3α–TAK1–IKKβ–canonical and perhaps other cancer types. and KRAS–GSK-3α–IKKα–noncanonical NF-κB pathways. As with many studies that initiate a new road, the study by Disclosure of Potential Confl icts of Interest Bang and colleagues (2 ) raises more questions than answers. No potential confl icts of interest were disclosed.

614 | CANCER DISCOVERYJUNE 2013 www.aacrjournals.org

Grant Support 5. Forde JE , Dale TC . Glycogen synthase kinase 3: a key regulator of cellular fate. Cell Mol Life Sci 2007 ; 64 : 1930 – 44 . This work was funded by NIH T32 GM008688, T32 CA009135, 6. MacAulay K , Doble BW , Patel S , Hansotia T , Sinclair EM , Drucke r D J, and R01 CA155192. et al. Glycogen synthase kinase 3alpha-speciﬁ c regulation of murine hepatic glycogen metabolism. Cell Metab 2007 ; 6 : 329 – 37 . Published online June 7, 2013. 7. Ougolkov AV , Fernandez-Zapico ME , Savoy DN , Urrutia RA , Billadeau DD . Glycogen synthase kinase-3beta participates in nuclear factor kappaB-mediated gene transcription and cell survival in pan- REFERENCES creatic cancer cells. Cancer Res 2005 ; 65 : 2076 – 81 . 8. Tang QL , Xie XB , Wang J , Chen Q , Han AJ , Zou CY , et al. Glycogen 1. Duran A , Linares JF , Galvez AS , Wikenheiser K , Flores JM , Dia z-Meco synthase kinase-3beta, NF-kappaB signaling, and tumorigenesis of MT , et al. The signaling adaptor p62 is an important NF-kappaB human osteosarcoma. J Natl Cancer Inst 2012 ; 104 : 749 – 63 . mediator in tumorigenesis. Cancer Cell 2008 ; 13 : 343 – 54 . 9. Banerji V , Frumm SM , Ross KN , Li LS , Schinzel AC , Hahn CK , et al. 2. Bang D , Wilson W , Ryan M , Yeh JJ , Baldwin AS . GSK-3alpha pro- The intersection of genetic and chemical genomic screens identiﬁ es motes oncogenic KRAS function in pancreatic cancer via TAK1-TAB GSK-3alpha as a target in human acute myeloid leukemia. J Clin stabilization and regulation of noncanonical NF-kappaB . Cancer Invest 2012 ; 122 : 935 – 47 . Discov 2013;3:690–703 . 10. Singh A , Sweeney MF , Yu M , Burger A , Greninger P , Benes C , et al. 3. Chenette EJ . Cancer: a Ras and NF-kappaB pas de deux . Nat Rev Drug TAK1 inhibition promotes apoptosis in KRAS-dependent colon can- Discov 2009 ; 8 : 932 . cers. Cell 2012 ; 148 : 639 – 50 . 4. Barbie DA , Tamayo P , Boehm JS , Kim SY , Moody SE , Dunn IF , et al. 11. Melisi D , Xia Q , Paradiso G , Ling J , Moccia T , Carbone C , et al. Modu- Systematic RNA interference reveals that oncogenic KRAS-driven lation of pancreatic cancer chemoresistance by inhibition of TAK1. cancers require TBK1. Nature 2009 ; 462 : 108 – 12 . J Natl Cancer Inst 2011 ; 103 : 1190 – 204 .

JUNE 2013CANCER DISCOVERY | 615

Chorom Pak and Shigeki Miyamoto

Cancer Discovery 2013;3:613-615.

Updated version Access the most recent version of this article at: http://cancerdiscovery.aacrjournals.org/content/3/6/613

Cited articles This article cites 11 articles, 2 of which you can access for free at: http://cancerdiscovery.aacrjournals.org/content/3/6/613.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://cancerdiscovery.aacrjournals.org/content/3/6/613.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerdiscovery.aacrjournals.org/content/3/6/613. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.