Current Biology 19, R421–R425, May 26, 2009 ª2009 Elsevier Ltd All rights reserved DOI 10.1016/j.cub.2009.04.022

MinireviewThe DeMSTification of MammalianSte20 Kinases

Maria Radu and Jonathan Chernoff*

When first reported in 1995, the mammalian Ste20-likekinases (Mst) 1 and 2 were so named both for their simi-larity to the yeast kinase Ste20 and for the fact that theirfunction was, to us, a deep mystery. While much remainsto be explained about the regulation and role of thesekinases, the veil has been at least partly raised on theMsts, revealing unexpected modes of activation and func-tion. Work in model organisms suggests a central growth-suppressive role for Mst orthologs, with intriguingpossible links to other established tumor suppressors.This minireview underlines our current understanding ofhow Mst1 and Mst2 are regulated, and how activation ofthese proteins influences cell survival and proliferation.

The mammalian Ste20-like kinases Mst1 and Mst2 areclosely related, evolutionarily conserved serine/threonineprotein kinases that play an important role in cell prolifera-tion, survival, morphology, and motility. In Drosophila, theMst1/2 ortholog Hippo clearly has tumor suppressor activity,and it is suspected, but not proven, that the mammalian Mstsmay share this property. As such, the Msts have recentlybecome the subjects of intense scrutiny, yet we know rela-tively little about how these kinases are regulated or the iden-tity of their key substrates.

Mst: A Short HistoryMst1 (also known as Stk4 and Krs2) and Mst2 (also known asStk3 and Krs1) were initially cloned from a lymphoid cDNAlibrary by PCR during a search for human relatives ofSaccharomyces cerevisiae Ste20 [1–3]. Subsequently, theseproteins were rediscovered by protein purification askinases that respond to extreme cellular stress [4]. Mst1and 2 are closely related to one another and broadly ex-pressed, but are structurally and functionally distinct fromMst3 and Mst4, two other more distantly related kinaseswith similar names, and from Ysk1/Sok1. Early biochemicalexperiments revealed that the carboxyl terminus of Mst1contains a coiled-coil dimerization motif, subsequentlytermed the SARAH (Salvador, RASSF1, Hippo) domain,which mediates self-association as well as association withother SARAH-domain-containing proteins [5].

Physiological stimuli for the Msts, and the signaling path-ways affected by these kinases, remained obscure until itwas realized that a w36 kDa protein kinase observed inapoptotic cells was in fact an amino-terminal, caspase-mediated cleavage product of Mst1 [6]. This finding providedan important clue into both the regulation and the cellularfunction of Mst1, linking it for the first time to a physiologicalsignaling process, namely apoptosis. Soon thereafter, cas-pase-cleaved Mst1 was shown to shuttle from the cytosol

Fox Chase Cancer Center, 333 Cottman Ave, Philadelphia, PA 19111,

USA.

*E-mail: J_Chernoff@fccc.edu

to the nucleus where it was found to play a key role in chro-matin condensation during apoptosis [7]. However, it was asa result of discoveries made in Drosophila that a coresignaling activity of these kinases was first elucidated.Here, genetic screens for flies with defects in organ size iden-tified Hippo, the Mst1/2 ortholog, as a key regulator of cellgrowth and survival [8,9]. Subsequent epistasis experimentsplaced Hippo atop a tumor suppressor signaling pathwayinvolving the adaptor proteins Salvador (WW45, in man)and MATS (Mob1), the protein kinase WARTS (Lats), andthe transcriptional coactivator Yorkie (Yap), which regulatesthe expression of a number of cell proliferation and survivalgenes. This model, well-described in a number of recentreviews [10–12], forms the basis for our current under-standing of Hippo function in flies and may well apply toMst function in man as well.

Mst Activation: Mysteries Still AboundWhat activates the Msts? In flies, it has been suggested thatthe ezrin–radixin–moesin proteins Merlin (a known tumorsuppressor in man) and Expanded lie upstream of Hippo.Additional evidence that Merlin and Mst signaling are relatedcan be inferred from a recent report that shows increasedYap expression and nuclear localization in Merlin-deficienthuman meningioma cells, as well as rescue of cell-cycledefects in such cells by siRNA-mediated silencing of Yapexpression [13–15]. What is missing, however, is any directevidence that Merlin associates with or affects the activityof Mst. For these reasons it remains unclear whether Merlinand Mst operate in series or in parallel signaling modulesthat converge on a final common pathway. Clarifying therelationship between Merlin and Mst is an important goal,as Merlin remains one of the least understood of the classicaltumor suppressors and any success in tying the pathwaystogether could have important therapeutic implications.

What we do know about Mst activation can be summa-rized in three words: cleavage, binding, and phosphorylation(Figure 1). These three processes, alone or in combination,play an important role in regulating Mst activity in a varietyof settings. At present, cleavage by caspase-3 remains thebest known and only mechanistically understood stimulusfor Mst. In this case, cleavage of the regulatory carboxylterminus liberates a fragment comprising the amino-terminalkinase domain, which is nearly ten times more active than theparent enzyme [16]. This cleavage also separates the amino-terminal kinase domain from nuclear exclusion sequenceslocated in the carboxyl terminus, thus allowing the transloca-tion of the kinase domain from the cytosol to the nucleusand, presumably, offering it a different menu of substratesto phosphorylate.

Proteolytic cleavage is not the sole mechanism of Mst acti-vation, however. Full-length Mst is activated by staurospor-ine (which is curious, as this is a general and potent kinaseinhibitor in most circumstances) as well as by otherapoptotic or stress stimuli, including UV radiation, tumornecrosis factor a (TNFa), retinoic acid, serum starvation,heat shock at 55�C, hydrogen peroxide, reactive oxygenspecies [16], and various anti-cancer drugs [16–18]. When

Current Biology Vol 19 No 10R422

activated by such stimuli, Mst autophosphorylates atmultiple sites, including key threonine residues in the activa-tion loop, which are likely to play a critical role in activatingthe enzyme [19,20]. Mst is also negatively regulated byphosphorylation. For example, Mst1 phosphorylation afterepidermal growth factor (EGF) stimulation results indecreased kinase activity, while dephosphorylation in vitroby protein phosphatase 2A (PP2A) restores kinase activity[1]. In addition to activating and inhibiting Mst, phosphoryla-tion can also regulate the sensitivity to cleavage as well asthe substrate specificity of Mst. In HEK293 cells, Jang et al.[21] have shown that Akt-mediated phosphorylation ofMst1 on Thr387 renders it resistant to cleavage by caspases,less able to promote apoptosis, and less active towards thepro-apoptotic transcription factor FOXO3, yet more activetowards histone H2B. These results are puzzling, as intactMst1 is mostly cytosolic, and therefore the cleavage-resis-tant phosphorylated or phosphomimetic form would not beexpected to have access to nuclear H2B [22]. These resultscould be interpreted to mean that the effects of Mst1 onFOXO3 outweigh its effects on H2B, at least in terms of pro-grammed cell death.

Finally, it is clear that Mst activity and/or function is regu-lated by association with other proteins. In addition to form-ing SARAH-domain-mediated homodimers, the Mst proteinsalso heterodimerize with other SARAH-domain-containingproteins. For example, Mst1 and Mst2 heterodimerize withWW45 and with Nore1A/Nore1B, members of the RASSFfamily of linker proteins that bind activated Ras and RapGTPases [23]. It appears that Mst–WW45 and Mst–Nore1complexes are mutually exclusive, as no evidence fora ternary Mst–WW45–Nore1 complex has been found(although the same may not hold true for the related proteinRassf1A, which is reported to co-precipitate with bothendogenous Mst2 and WW45 from HeLa cells) [23]. WW45promotes Mst kinase activity and is itself a target of thiskinase [23]. The association of Nore1 with Mst, which isthought to be constitutive, is required for Rap-mediated‘inside-out’ integrin signaling in T cells. Here, activation of

Figure 1. Apoptotic signals lead to Mst acti-vation by phosphorylation and proteolyticcleavage.

Apoptotic stimuli lead to Mst autophosphory-lation and activation of caspases. Autophos-phorylated Mst is active and phosphorylatessubstrates that amplify the apoptotic signal,further increasing caspase activation. In addi-tion, active caspases cleave off the Mst car-boxy-terminal regulatory domain. This eventliberates the active amino-terminal kinasedomain, which translocates into the nucleuswhere it phosphorylates transcription factorsand histones.

the small GTPase Rap1 results in therecruitment of Nore1B (also known asRapL), presumably as part of a Mst–Nore1B complex. Mst1 is activated inthese circumstances, but not inT cells that lack Nore1B or in cells ex-pressing mutant forms of Nore1B thatcannot bind Mst1 [24]. Mst1 activationplays a key role in T-cell development,

as Mst1-null mice show decreased numbers of naı̈veT cells, most likely due to the elimination of prematurelyactivated T cells [24]. This function is consonant with thegeneral theme of Msts as negative regulators of cell growthand survival. Mature T cells from Mst1-null mice alsoshowed defects in adhesion and motility [25]. Mst1 can inaddition bind to other members of the RASSF family,including the aforementioned protein Rassf1A, which liber-ates Mst from a complex with Raf (in which Mst is inactive)[26] and brings Mst in proximity to active Ras at the plasmamembrane, where Mst becomes active [27]. In this case, themeaning of the association between Mst and the adaptorRassf1A is less clear, as overexpression of Rassf1A hasbeen reported to have both stimulatory and inhibitoryeffects on Mst activity [28]. Such apparently discordantbehavior is not unusual for adaptor proteins: too muchRassf1A might sequester Mst and prevent its activation,whereas lower amounts might assist in activation. Whateverthe exact nature of this interaction may be, both the Mst–Nore1B and the Mst–Rassf1A associations link Msts withupstream small GTPases, a theme common to severalmembers of the Ste20 superfamily of protein kinases,such as the p21-activated kinases [29].

Mst SubstratesDespite the flurry of interest in the Msts that followed thediscovery of the Hippo tumor suppressor pathway in flies,we know very little about the cellular substrates of Mst. Infact, the total sum of known substrates is still a single digitnumber (Figure 2). Nevertheless, even this small list revealsseveral interesting facets of Mst function, especially with re-gard to the role of Mst in promoting apoptosis and inhibitingcell-cycle progression. For example, one of the best-charac-terized mammalian substrates of Mst1 is H2B, which is phos-phorylated at serine 14 in vitro and in apoptotic cells [30]. Thephosphorylation is catalyzed by the amino-terminal proteo-lytic fragment of Mst1 that translocates into the nucleus.Upon phosphorylation by Mst1, the Ser14-phosphorylatedform of H2B contributes to chromatin condensation and

ReviewR423

Mst1/2

H2B

FOXO

FOXOP

H2B

P

Akt

FOXO

P

Apoptosis

JNK

Nore1

Mst1/2

Yap

hSav

RASSF

Lats

Nore1Ras

Lats

P

P

Mob

Current Biology

Figure 2. Multiple apoptotic signaling path-ways are mediated by Mst targets.

Akt inhibits Mst kinase activity towards FOXOtranscription factors, but increases Mstkinase activity towards H2B. In addition, Aktcan directly phosphorylate FOXO at a distinctsite (shown by the red circle), which results ininhibition of apoptotic signaling. JNK isa central player in Mst-mediated chromatincondensation during apoptosis. Binding ofNore1 to Mst brings Mst into proximity withRas. The Ras–Nore1–Mst complex is knownto be involved in mediating Ras-inducedapoptosis. Mst phosphorylates Mob1 andpromotes its binding to Lats, thereby stimu-lating Lats phosphorylation. Mst also phos-phorylates Lats, which ultimately results inthe phosphorylation of cytoplasmic Yap andits translocation into the nucleus, where itinduces activation of pro-apoptotic transcrip-tion factors.

DNA fragmentation, and thus to theoverall cell death program. However,the importance of Mst-catalyzed, H2BSer14 phosphorylation to chromatincondensation has been challenged byreports that Mst1 also activates JNK,and that activation of Mst1 does notcause chromatin condensation orapoptosis in embryonic stem cellslacking the JNK activators Mkk4 andMkk7 [31]. These findings suggestthat it is caspase-mediated cleavage of Mst1, followed byMst1-mediated activation of the JNK pathway, rather thandirect H2B phosphorylation, that is the mechanism mostlikely to be responsible for Mst-mediated induction of chro-matin condensation during apoptosis. Alternatively, theseresults may arise from cell-type specific differences.

The FOXO transcription factors represent a second classof Mst substrates that is relevant to its pro-apoptotic effects.When phosphorylated within the forkhead domain by Mst1,FOXO3 can no longer bind 14-3-3. In the absence of 14-3-3binding, FOXO3 is free to translocate to the nucleus, whereit activates a number of pro-apoptotic genes [32]. Interest-ingly, the anti-apoptotic kinase Akt1 has been shown tophosphorylate Mst1, preventing caspase cleavage of Mst1and markedly decreasing Mst-mediated FOXO3 phosphory-lation [21]. The relationship between Mst1 and Akt appearsto be mutually antagonistic, as might be expected fora pro- and an anti-apoptotic kinase, respectively: Akt bindsto, phosphorylates, and inhibits Mst1, and Mst1 (and Mst2)inhibits Akt, though it appears that this inhibition is due tothe association of these proteins rather than by phosphory-lation of Akt by Mst. Recent experiments carried out inneuronal cultures suggest that another member of FOXOfamily, specifically FOXO1, is a substrate for Mst1 and thatthis phosphorylation event facilitates FOXO1 translocationinto the nucleus by hindering the binding of 14-3-3 toFOXO1. In such cells, Mst1–FOXO1 signaling appears to beimportant in mediating neuronal death upon deprivation ofserum and of neuronal activity [33].

The AGC-family kinases Ndr1/2 and Lats1/2 are additionalsubstrates of Mst that contribute to apoptosis as well as cell-

cycle arrest. Both Fas-stimulated and okadaic-acid stimu-lated phosphorylation and activation of Ndr1/2 are mediatedby Mst, which is present in a complex with Ndr and its co-activator, Mob1 [34–36]. Similarly, Lats–Mob complexes,key elements in the conserved Mst/Hippo pathway, arephosphorylated and activated by Mst [37]. WW45 is requiredfor Mst binding to Lats and facilitates the phosphorylation ofLats by Mst. Lats, in turn, phosphorylates and sequestersYap in the cytoplasm. Importantly, the related kinases Ndr1and Ndr2 are unable to bind Yap or to phosphorylate Yapin vitro. In mammals, cytoplasmic retention of the Yorkieortholog Yap leads to p73 stabilization and induction ofexpression of the p53-inducible protein Puma, thuspromoting apoptosis and cell-cycle arrest [26]. It should benoted that Mst achieves the same ends in Drosophila, butby different means. In Drosophila, Hippo signaling throughWarts also leads to the phosphorylation of Yorkie and theinduction of apoptosis, but the genes regulated by Yorkieare almost certainly different from those regulated by Yapin mammalian cells. That is, active (unphosphorylated)Yorkie upregulates the expression of the apoptosis inhibitordIAP, the cell-cycle protein cyclin E, and the microRNAbantam [38,39]. With the possible exception of cyclin E[13], the equivalents of these transcriptional targets are notknown to be affected by Yap in mammalian cells. Lats canalso regulate mitotic exit [40] by binding and inhibiting theactin regulator LIM kinase [41], resulting in defective actinpolymerization and inhibition of cytokinesis.

DAP4 is a constitutively nuclear protein that homodimer-izes through its amino terminus and binds Mst1 throughits carboxy-terminal region. DAP4 is not significantly

Current Biology Vol 19 No 10R424

phosphorylated by Mst1 and it does not modulate Mst1kinase activity in vivo or in vitro. However, whereas overex-pression of DAP4 does not cause apoptosis, coexpressionof DAP4 with Mst1 enhances Mst1-induced apoptosisthrough a p53-dependent pathway. Mst1 is unable to directlyphosphorylate p53. DAP4 binds p53 and possibly bringsMst1 in contact with p53, promoting apoptosis [42]. Interest-ingly, DAP4 was initially identified as a possible mediator ofinterferon g (IFNg)-induced apoptosis [43], but Mst proteinsfail to be activated in response to IFNg treatment.

Recently, Sun and colleagues [44] showed that cardiactroponin is a substrate for Mst1. Phosphorylation by Mst1resulted in conformational changes of cardiac troponin andaltered its interactions with other components of thetroponin complex. In the heart, activation of Mst causes car-diomyocyte death and dilated cardiomyopathy, suggestingan important role for Mst1 in regulation of cardiac musclecontractility, probably by modulating signaling pathwaysthat control the myofilament function.

Concluding RemarksThough we have learned a lot about the Msts, much aboutthese kinases remains obscure and, given the potential roleof Msts as tumor suppressors, these mysteries are too impor-tant to leave unsolved. The major challenges are to uncoverthe molecular links between cell-surface receptors and Mstactivation, to get a more complete picture of Mst substrates,and to determine the functions of the two Mst isoforms inmammalian organisms. Do Merlin and/or the humanExpanded homolog activate Mst, and, if so, how? Is thispathway related to or independent of small GTPase/RASSF1inputs? Regarding the issue of Mst substrates, we need notwait for the next genetic screen in flies; several new bioinfor-matic and biochemical tools are available that might speedup the hunt in mammalian cells. For example, phosphopro-teomic methods have revealed the optimal (and highlyunusual) phosphorylation consensus site preferred byMst1/2 — WYNTMKRR [45] — and this information, encodedin matrix form, could be used to search databases of knownphosphopeptides to predict new Mst substrates. In addition,new unbiased schemes to detect direct substrates for proteinkinases [46,47] could and should be brought to bear on thisproblem. At the level of biochemistry, we know little aboutthe interplay, if any, between Mst1 and Mst2. Do the twoMsts form heterodimers, and, if so, is the regulation, location,function, and/or substrate specificity of such heterodimersdifferent from Mst homodimers? In addition, much remainsto be learned about Mst genetic functions in mammals. AreMst1 and Mst2 redundant in mammals, or do they serveunique functions? Are either of these kinases tumor suppres-sors? Hypermethylation of the Mst1 and Mst2 promotershave been shown to inhibit the WW45–Rassf1–Mst pathwayin human sarcomas [48] and loss of cytoplasmic Mst1 hasbeen reported to be a marker for tumor progression in colo-rectal cancer, suggesting that Mst1 can function as a tumorsuppressor [49]. However, the Mst1 knockout mouse isreported to have a surprisingly mild phenotype and doesnot appear, on its own, to have tumor suppressor function[25,50]. Whether Mst2, which can replace Hippo function inflies, will have such activity is not known. These and relatedquestions will keep us busy for some time, but it is to behoped that solving the mystery of the Msts will have a bearingon our evolving understanding of how cells regulate theirproliferation and survival.

References1. Creasy, C.L., and Chernoff, J. (1995). Cloning and characterization of

a member of the MST subfamily of Ste20-like kinases. Gene 167, 303–306.

2. Creasy, C.L., Ambrose, D.M., and Chernoff, J. (1996). The Ste20-like proteinkinase, Mst1, dimerizes and contains an inhibitory domain. J. Biol. Chem.271, 21049–21053.

3. Creasy, C.L., and Chernoff, J. (1995). Cloning and characterization ofa human protein kinase with homology to Ste20. J. Biol. Chem. 270,21695–21700.

4. Taylor, L.K., Wang, H.C., and Erikson, R.L. (1996). Newly identified stress-responsive protein kinases, Krs-1 and Krs-2. Proc. Natl. Acad. Sci. USA93, 10099–10104.

5. Scheel, H., and Hofmann, K. (2003). A novel interaction motif, SARAH,connects three classes of tumor suppressor. Curr. Biol. 13, R899–R900.

6. Graves, J.D., Gotoh, Y., Draves, K.E., Ambrose, D., Han, D.K., Wright, M.,Chernoff, J., Clark, E.A., and Krebs, E.G. (1998). Caspase-mediated activa-tion and induction of apoptosis by the mammalian Ste20-like kinase Mst1.EMBO J. 17, 2224–2234.

7. Ura, S., Masuyama, N., Graves, J.D., and Gotoh, Y. (2001). Caspase cleavageof MST1 promotes nuclear translocation and chromatin condensation. Proc.Natl. Acad. Sci. USA 98, 10148–10153.

8. Pantalacci, S., Tapon, N., and Leopold, P. (2003). The Salvador partner Hippopromotes apoptosis and cell-cycle exit in Drosophila. Nat. Cell Biol. 5,921–927.

9. Udan, R.S., Kango-Singh, M., Nolo, R., Tao, C., and Halder, G. (2003). Hippopromotes proliferation arrest and apoptosis in the Salvador/Warts pathway.Nat. Cell Biol. 5, 914–920.

10. Harvey, K., and Tapon, N. (2007). The Salvador-Warts-Hippo pathway - anemerging tumour-suppressor network. Nat. Rev. Cancer 7, 182–191.

11. Pan, D. (2007). Hippo signaling in organ size control. Genes Dev. 21, 886–897.

12. Saucedo, L.J., and Edgar, B.A. (2007). Filling out the Hippo pathway.Nat. Rev. Mol. Cell. Biol. 8, 613–621.

13. Striedinger,K.,VandenBerg,S.R.,Baia,G.S.,McDermott,M.W.,Gutmann,D.H.,and Lal, A. (2008). The neurofibromatosis 2 tumor suppressor geneproduct, merlin, regulates human meningioma cell growth by signalingthrough YAP. Neoplasia 10, 1204–1212.

14. Lau, Y.K., Murray, L.B., Houshmandi, S.S., Xu, Y., Gutmann, D.H., and Yu, Q.(2008). Merlin is a potent inhibitor of glioma growth. Cancer Res. 68,5733–5742.

15. Yokoyama,T.,Osada,H.,Murakami,H.,Tatematsu,Y.,Taniguchi,T.,Kondo,Y.,Yatabe,Y.,Hasegawa,Y.,Shimokata,K.,Horio,Y.,etal. (2008).YAP1 is involvedin mesothelioma development and negatively regulated by Merlin throughphosphorylation. Carcinogenesis 29, 2139–2146.

16. Kakeya, H., Onose, R., and Osada, H. (1998). Caspase-mediated activationof a 36-kDa myelin basic protein kinase during anticancer drug-inducedapoptosis. Cancer Res. 58, 4888–4894.

17. Watabe, M., Kakeya, H., Onose, R., and Osada, H. (2000). Activation of MST/Krs and c-Jun N-terminal kinases by different signaling pathways duringcytotrienin A-induced apoptosis. J. Biol. Chem. 275, 8766–8771.

18. Watabe, M., Kakeya, H., and Osada, H. (1999). Requirement of protein kinase(Krs/MST) activation for MT-21-induced apoptosis. Oncogene 18,5211–5220.

19. Glantschnig, H., Rodan, G.A., and Reszka, A.A. (2002). Mapping of MST1kinase sites of phosphorylation. Activation and autophosphorylation.J. Biol. Chem. 277, 42987–42996.

20. Graves, J.D., Draves, K.E., Gotoh, Y., Krebs, E.G., and Clark, E.A. (2001).Both phosphorylation and caspase-mediated cleavage contribute to regula-tion of the Ste20-like protein kinase Mst1 during CD95/Fas-inducedapoptosis. J. Biol. Chem. 276, 14909–14915.

21. Jang, S.W., Yang, S.J., Srinivasan, S., and Ye, K. (2007). Akt phosphorylatesMstI and prevents its proteolytic activation, blocking FOXO3 phosphoryla-tion and nuclear translocation. J. Biol. Chem. 282, 30836–30844.

22. Anand, R., Kim, A.Y., Brent, M., and Marmorstein, R. (2008). Biochemicalanalysis of MST1 kinase: Elucidation of a C-terminal regulatory region.Biochemistry, epub ahead of print.

23. Guo, C., Tommasi, S., Liu, L., Yee, J.K., Dammann, R., and Pfeifer, G.P.(2007). RASSF1A is part of a complex similar to the Drosophila Hippo/Salva-dor/Lats tumor-suppressor network. Curr. Biol. 17, 700–705.

24. Zhou, D., Medoff, B.D., Chen, L., Li, L., Zhang, X.F., Praskova, M., Liu, M.,Landry, A., Blumberg, R.S., Boussiotis, V.A., et al. (2008). The Nore1B/Mst1 complex restrains antigen receptor-induced proliferation of naiveT cells. Proc. Natl. Acad. Sci. USA 105, 20321–20326.

25. Katagiri, K., Katakai, T., Ebisuno, Y., Ueda, Y., Okada, T., and Kinashi, T.(2009). Mst1 controls lymphocyte trafficking and interstitial motility withinlymph nodes. EMBO J., epub ahead of print.

26. Matallanas, D., Romano, D., Yee, K., Meissl, K., Kucerova, L., Piazzolla, D.,Baccarini, M., Vass, J.K., Kolch, W., and O’Neill, E. (2007). RASSF1A elicitsapoptosis through an MST2 pathway directing proapoptotic transcriptionby the p73 tumor suppressor protein. Mol. Cell 27, 962–975.

27. Praskova, M., Khoklatchev, A., Ortiz-Vega, S., and Avruch, J. (2004).Regulation of the MST1 kinase by autophosphorylation, by the growth inhib-itory proteins, RASSF1 and NORE1, and by Ras. Biochem. J. 381, 453–462.

ReviewR425

28. Oh, H.J., Lee, K.K., Song, S.J., Jin, M.S., Song, M.S., Lee, J.H., Im, C.R.,Lee, J.O., Yonehara, S., and Lim, D.S. (2006). Role of the tumor suppres-sor RASSF1A in Mst1-mediated apoptosis. Cancer Res. 66, 2562–2569.

29. Sells, M.A., and Chernoff, J. (1997). Emerging from the Pak: the p21-activated protein kinase family. Trends Cell Biol. 7, 162–167.

30. Cheung, W.L., Ajiro, K., Samejima, K., Kloc, M., Cheung, P., Mizzen, C.A.,Beeser, A., Etkin, L.D., Chernoff, J., Earnshaw, W.C., et al. (2003). Apoptoticphosphorylation of histone H2B is mediated by mammalian sterile twentykinase. Cell 113, 507–517.

31. Ura, S., Nishina, H., Gotoh, Y., and Katada, T. (2007). Activation of the c-JunN-terminal kinase pathway by MST1 is essential and sufficient for theinduction of chromatin condensation during apoptosis. Mol. Cell Biol. 27,5514–5522.

32. Lehtinen, M.K., Yuan, Z., Boag, P.R., Yang, Y., Villen, J., Becker, E.B.,DiBacco, S., de la Iglesia, N., Gygi, S., Blackwell, T.K., et al. (2006). Aconserved MST-FOXO signaling pathway mediates oxidative-stressresponses and extends life span. Cell 125, 987–1001.

33. Yuan, Z., Lehtinen, M., Merlo, P., and Bonni, A. (2009). Regulation of neuronalcell death by MST1-FOXO1 signaling. J. Biol. Chem. 284, 11285–11292.

34. Chan, E.H., Nousiainen, M., Chalamalasetty, R.B., Schafer, A., Nigg, E.A.,and Sillje, H.H. (2005). The Ste20-like kinase Mst2 activates the human largetumor suppressor kinase Lats1. Oncogene 24, 2076–2086.

35. Hirabayashi, S., Nakagawa, K., Sumita, K., Hidaka, S., Kawai, T., Ikeda, M.,Kawata, A., Ohno, K., and Hata, Y. (2008). Threonine 74 of MOB1 is a putativekey phosphorylation site by MST2 to form the scaffold to activate nuclearDbf2-related kinase 1. Oncogene 27, 4281–4292.

36. Vichalkovski, A., Gresko, E., Cornils, H., Hergovich, A., Schmitz, D., andHemmings, B.A. (2008). NDR kinase is activated by RASSF1A/MST1 inresponse to Fas receptor stimulation and promotes apoptosis. Curr. Biol.18, 1889–1895.

37. Praskova, M., Xia, F., and Avruch, J. (2008). MOBKL1A/MOBKL1B phos-phorylation by MST1 and MST2 inhibits cell proliferation. Curr. Biol. 18,311–321.

38. Huang, J., Wu, S., Barrera, J., Matthews, K., and Pan, D. (2005). The Hipposignaling pathway coordinately regulates cell proliferation and apoptosisby inactivating Yorkie, the Drosophila Homolog of YAP. Cell 122, 421–434.

39. Thompson, B.J., and Cohen, S.M. (2006). The Hippo pathway regulates thebantam microRNA to control cell proliferation and apoptosis in Drosophila.Cell 126, 767–774.

40. Yabuta, N., Okada, N., Ito, A., Hosomi, T., Nishihara, S., Sasayama, Y.,Fujimori, A., Okuzaki, D., Zhao, H., Ikawa, M., et al. (2007). Lats2 is an essen-tial mitotic regulator required for the coordination of cell division. J. Biol.Chem. 282, 19259–19271.

41. Yang, X., Yu, K., Hao, Y., Li, D.M., Stewart, R., Insogna, K.L., and Xu, T.(2004). LATS1 tumour suppressor affects cytokinesis by inhibiting LIMK1.Nat. Cell Biol. 6, 609–617.

42. Lin, Y., Khokhlatchev, A., Figeys, D., and Avruch, J. (2002). Death-associatedprotein 4 binds MST1 and augments MST1-induced apoptosis. J. Biol.Chem. 277, 47991–48001.

43. Kimchi, A. (1998). DAP genes: novel apoptotic genes isolated by a functionalapproach to gene cloning. Biochim. Biophys. Acta. 1377, F13–F33.

44. You, B., Yan, G., Zhang, Z., Yan, L., Li, J., Ge, Q., Jin, J.P., and Sun, J. (2009).Phosphorylation of cardiac troponin I by mammalian sterile 20-like kinase 1.Biochem. J. 418, 93–101.

45. Miller, M.L., Jensen, L.J., Diella, F., Jorgensen, C., Tinti, M., Li, L., Hsiung, M.,Parker, S.A., Bordeaux, J., Sicheritz-Ponten, T., et al. (2008). Linear motifatlas for phosphorylation-dependent signaling. Sci. Signal 1, ra2.

46. Blethrow, J.D., Glavy, J.S., Morgan, D.O., and Shokat, K.M. (2008). Covalentcapture of kinase-specific phosphopeptides reveals Cdk1-cyclin Bsubstrates. Proc. Natl. Acad. Sci. USA 105, 1442–1447.

47. Chi, Y.,Welcker,M.,Hizli, A.A.,Posakony,J.J.,Aebersold,R., andClurman,B.E.(2008). Identification of CDK2 substrates in human cell lysates. Genome Biol. 9,R149.

48. Seidel, C., Schagdarsurengin, U., Blumke, K., Wurl, P., Pfeifer, G.P.,Hauptmann, S., Taubert, H., and Dammann, R. (2007). Frequent hypermethy-lation of MST1 and MST2 in soft tissue sarcoma. Mol. Carcinog. 46, 865–871.

49. Minoo, P., Zlobec, I., Baker, K., Tornillo, L., Terracciano, L., Jass, J.R., andLugli, A. (2007). Prognostic significance of mammalian sterile20-like kinase1 in colorectal cancer. Mod. Pathol. 20, 331–338.

50. Anguera, M.C., Liu, M., Avruch, J., and Lee, J.T. (2008). Characterization oftwo Mst1-deficient mouse models. Dev. Dyn. 237, 3424–3434.