Pin-Pointing a New DAP Kinase Function: The Peptidyl-Proly Isomerase Pin1 Is Negatively Regulated by DAP Kinase-Mediated Phosphorylation

Molecular Cell

Previews Pin-Pointing a New DAP Kinase Function: The Peptidyl-Proly Isomerase Pin1 Is Negatively Regulated by DAP Kinase-Mediated Phosphorylation Shani Bialik1 and Adi Kimchi1,* 1Department of Molecular Genetics, Weizmann Institute of Science, Rehovot 76100, Israel *Correspondence: [email protected] DOI 10.1016/j.molcel.2011.04.002

In this issue of Molecular Cell, Lee et al. (2011) identify the peptidyl-prolyl isomerase Pin1 as a substrate of DAP kinase, simultaneously providing a critical regulatory mechanism for Pin1 inhibition and a potential mechanism that accounts for DAPK’s tumor-suppressive activities. The Pin1 peptidyl-prolyl isomerase catalyzes the conversion between the cis/ trans conformations of proline, within specific phospho-Ser/Thr-Pro sequences (Lu and Zhou, 2007). Pin1-induced conformational changes have been proposed to control many protein functions, including their catalytic activity, interactions with other proteins, subcellular localization, protein stability, and phosphorylation status. Thus, isomerization of phospho-Ser/Thr-Pro by Pin1 is important for signal transduction and, moreover, provides a molecular timer controlling the activity of many phosphoproteins. In fact, Pin1 has been implicated in the regulation of a wide range of cellular activities such as mitosis, cell cycle, tumorigenesis, immune responses, neuroprotection, and apoptosis (Lu and Zhou, 2007). While regulation of Pin1 is presumably critical for its functions, little is known in this regard. Expression of Pin1 is tightly controlled in normal cells and correlates with the cell’s proliferative capacity, increasing dramatically in cancer cells. Pin1 activity can be regulated by phosphorylation of Ser16, which disrupts its interactions with its target proteins and is stabilized by a second phosphorylation event on Ser65 (Lu and Zhou, 2007). The accompanying paper by Lee et al. (2011) presents another critical mechanism of regulation involving phosphorylation by the tumor suppressor DAP kinase (DAPK). The authors identify a phospho site, Ser71, within the catalytic active site, phosphorylation of which inhibits isomerization activity in vitro. Molecular modeling, based on the known crystal structure of Pin1 in complex with a phospho-Thr-

containing peptide inhibitor, predicts that phospho-Ser71 forms Hydrogen bonds with the side chain of Arg69. This Arg is the critical determinant of substrate specificity as it binds the phosphate group of the phospho-Ser/Thr-Pro, and disruption of these interactions is predicted to prevent recognition and binding of a phosphorylated substrate. Most importantly, DAPK was identified as the kinase that physically associates with Pin1 and is responsible for Ser71 phosphorylation both in vitro and in cells. Moreover, DAPK blocks Pin1’s nuclear accumulation. This was also attributed to the predicted interactions between phosphoSer71 and Arg69, which lies within Pin1’s nuclear localization domain. The authors then show that phosphorylation on Ser71, achieved by either ectopic expression of DAPK or mutation of Ser71 to phospho-mimicking Asp or Glu, abrogates Pin1’s ability to induce transcriptional activity of cyclin D1, b-catenin, or NF-kB promoters and to increase the stability of endogenous cyclin D1 protein. These experiments indicate that phosphorylation of Pin1 by DAPK has the potential to regulate a wide range of Pin1 activities (Figure 1B). DAPK is a Ser/Thr kinase with multiple functions involving phosphorylation of and/or interaction with a variety of cellular targets (Figure 1A). DAPK is best known for its ability to mediate cell death, including apoptosis and autophagic cell death (Bialik and Kimchi, 2006, 2010). Furthermore, it regulates acto-myosin contractility, stress fiber formation, and cell motility and adhesion through phosphorylation of myosin regulatory light

chain and inhibition of integrin signaling (Bialik and Kimchi, 2006). Its ability to sensitize tumor cells to death stimuli and block metastasis through cytoskeleton modulation contribute to its tumor-suppressive functions. The findings reported by Lee et al. (2011) indicate that negative regulation of Pin1 is an additional critical pathway by which DAPK can suppress tumorigenesis. Expression of WT Pin1, or nonphosphorylatible Ser71Ala mutant, leads to centrosome amplification, abnormal chromosome separation, mitotic spindle defects, and cell transformation. Significantly, these phenotypes are all suppressed by expression of the phospho-mimetic Pin1 mutants. Furthermore, the authors observed a significant correlation between DAPK expression levels and Pin1 Ser71 phosphorylation in human breast tumor samples, and an inverse relationship between centrosome amplification and Ser71 phosphorylation. The authors further show that knockdown of Pin1 attenuates the increased cell migration that results from DAPK depletion in breast cancer cells, suggesting that phosphorylation of Pin1 is critical for DAPK’s anti-metastatic/anti-cell migratory function. Still to be addressed is how Pin1 inhibition is related to DAPK’s other tumorsuppressive functions and whether loss of DAPK and Pin1 overexpression are redundant or mutually exclusive events occurring in tumor samples. An important avenue for future research is to identify the cellular stimuli that activate the DAPK-Pin1 pathway and to determine in which cellular settings this pathway is relevant. Pin1 differs from other DAPK substrates in that its

Molecular Cell 42, April 22, 2011 ª2011 Elsevier Inc. 139

Molecular Cell

Previews A

DAP kinase P PKD p19ARF P NF-κB

MARK1/2

Integrin

Tropomyosin

P

MLC

P mdm2

Beclin 1

p53

Acto-myosin contractility

Tau

T cell activation blocked

P

Cell detachment

Autophagy

Programmed necrosis

Apoptosis

Stress fiber formation Membrane blebbing

MT instability

JNK

B DAP kinase

P Pin1

P Tau

cyclin D1

Tau MT

NF-κB

(Nuclear)

Chromosomal instability

cyclin D1

NF-κB NF-κB

Centrosome amplification

β-catenin

cyclin D1

Mitotic spindle deformities

Cell transformation

Figure 1. Molecular Pathways Triggered by DAPK (A) DAPK signaling network. DAPK has several distinct target proteins, each of which mediates the various functional arms of the kinase. Direct targets are depicted by polygons, indirect targets by ovals. Known substrates are indicated with a P for phosphorylation. Note that DAPK activates MARK1/2 independently of phosphorylation. (B) The DAPK/Pin1 connection. DAPK phosphorylates Pin1, thereby inhibiting its isomerase activity. Solid arrows lead to the functions and cellular outcomes of Pin1 that have been shown to be affected by DAPK, while dashed arrows lead to those that are postulated to be relevant to DAPK signaling, but not yet proven. MT, microtubules.

modification is predicted to affect a wide range of cellular functions, reflecting Pin1’s various roles as a master regulator of phosphorylation events. It is therefore important to assess whether DAPK can control other functions of Pin1 not related to oncogenesis. Intriguingly, the DAPK and Pin1 signaling networks intersect at several points. Perhaps the most exciting overlap is in the pathogenesis of Alzheimer’s disease (AD), in which both have been implicated by modulation of the phosphorylation of the microtubule binding protein Tau. Pin1 promotes Tau dephosphorylation, and Pin1 knockout mice develop neuropathologies that resemble AD (Balastik et al., 2007). Conversely, DAPK activates the Tau kinases MARK1/2, leading to enhanced phosphorylation of Tau, and promotes taupathy in Drosophila via activation of the MARK1/2 ortholog Par-1 (Wu et al., 2011). A genetic link between DAPK vari-

ants and AD has been suggested (Li et al., 2006) but not yet confirmed (Minster et al., 2009). Inhibition of Pin1 by DAPK may provide a second mechanism to promote Tau phosphorylation (Figure 1B). Another example relates to NF-kB signaling, where Pin1-activates NF-kB (Ryo et al., 2003), while DAPK has been shown to inhibit NF-kB activation specifically downstream of T Cell Receptor signaling (Chuang et al., 2008). It is possible that inhibition of Pin1 by DAPK contributes to this effect (Figure 1B). Another intriguing question is whether Pin1 can regulate DAPK. DAPK is phosphorylated by the Pro-directed ERK on Ser735 (Chen et al., 2005), which lies within the same domain shown to be critical for the Pin1/ DAPK interaction, creating a potential phospho-Ser-Pro target site for Pin1. The connection between two important signaling molecules with wide ranges of functional activities has provided signifi-

140 Molecular Cell 42, April 22, 2011 ª2011 Elsevier Inc.

cant insight into the mechanisms of action and regulation of DAPK and Pin1, respectively. Moreover, many exciting new questions can now be asked and addressed regarding these two master regulators, in respect to cancer, neurodegeneration, and more.

REFERENCES Balastik, M., Lim, J., Pastorino, L., and Lu, K.P. (2007). Biochim. Biophys. Acta 1772, 422–429. Bialik, S., and Kimchi, A. (2006). Annu. Rev. Biochem. 75, 189–210. Bialik, S., and Kimchi, A. (2010). Curr. Opin. Cell Biol. 22, 199–205. Chen, C.H., Wang, W.J., Kuo, J.C., Tsai, H.C., Lin, J.R., Chang, Z.F., and Chen, R.H. (2005). EMBO J. 24, 294–304. Chuang, Y.T., Fang, L.W., Lin-Feng, M.H., Chen, R.H., and Lai, M.Z. (2008). J. Immunol. 180, 3238–3249.

Molecular Cell

Previews Lee, T.H., Chen, C.-H., Suizu, F., Huang, P., Schiene-Fischer, C., Daum, S., Zhang, Y.J., Goate, A., Chen, R.-H., Zhou, X.Z., and Lu, K.P. (2011). Mol. Cell 42, this issue, 147–159.

K., Toombs, T.A., Kwok, S., et al. (2006). Hum. Mol. Genet. 15, 2560–2568.

Li, Y., Grupe, A., Rowland, C., Nowotny, P., Kauwe, J.S.K., Smemo, S., Hinrichs, A., Tacey,

Minster, R.L., DeKosky, S.T., and Kamboh, M.I. (2009). Neurobiol. Aging 30, 1890–1891.

Lu, K.P., and Zhou, X.Z. (2007). Nat. Rev. Mol. Cell Biol. 8, 904–916.

Ryo, A., Suizu, F., Yoshida, Y., Perrem, K., Liou, Y.-C., Wulf, G., Rottapel, R., Yamaoka, S., and Lu, K.P. (2003). Mol. Cell 12, 1413–1426. Wu, P.R., Tsai, P.I., Chen, G.C., Chou, H.J., Huang, Y.P., Chen, Y.H., Lin, M.Y., Kimchi, A., Chien, C.T., and Chen, R.H. (2011). Cell Death Differ., in press. Published online February 11, 2011.

DNA Damage Discrimination at Stalled Replication Forks by the Rad5 Homologs HLTF and SHPRH George-Lucian Moldovan1 and Alan D. D’Andrea1,* 1Department of Radiation Oncology, Dana-Farber Cancer Institute, Harvard Medical School, 44 Binney Street, Boston, MA 02215, USA *Correspondence: [email protected] DOI 10.1016/j.molcel.2011.03.018

In this issue of Molecular Cell, Lin et al. (2011) describe how HLTF and SHPRH, the human homologs of yeast Rad5, can discriminate between MMS-induced versus UV-induced DNA damage. The results have important implications for the suppression of damage-specific mutagenesis and for the maintenance of genomic stability. Essential for allowing cells to cope with DNA damage, the postreplication DNA repair (PRR) pathway is nevertheless one of the least understood genome preservation mechanisms. This pathway, also referred to as the RAD6 pathway, includes genes encoding specialized translesion synthesis (TLS) polymerases, able to replicate through DNA damage. Other PRR pathway genes encode proteins of the ubiquitin conjugating system. Rad6, for example, is a ubiquitin-conjugating enzyme (Jentsch et al., 1987), yet the relevant ubiquitinated substrate(s) of Rad6 and the molecular mechanism(s) of the PRR pathway remained unknown for many years. This situation dramatically changed when Jentsch and colleagues demonstrated that Rad6, along with its partner E3 ligase Rad18, ubiquitinates PCNA, thereby allowing replication bypass of DNA lesions (Hoege et al., 2002). An elegant model of how PCNA ubiquitination controls lesion bypass emerged. Rad6 and Rad18 monoubiquitinate PCNA at K164, inducing a polymerase switch between the replicative polymerase and a mutagenic TLS polymerase. Many TLS polymerases have ubiquitin-binding

domains in addition to their PCNA-interacting motifs, providing them with a higher affinity for ubiquitinated PCNA (Bienko et al., 2005; Kannouche et al., 2004). Furthermore, a Rad6-Rad18Rad5-Mms2-Ubc13 complex multiubiquitinates PCNA in yeast, directing a poorly understood error-free lesion bypass involving the sister chromatid (known as template switching). Mammalian cells have two Rad5 homologs, SHPRH and HLTF, which promote PCNA multiubiquitination and genomic stability (Motegi et al., 2006, 2008). In this issue of Molecular Cell, Cimprich and coworkers now show that, besides their function in template switching, HLTF and SHPRH also employ distinct mechanisms to control the recruitment of an appropriate polymerase for TLS bypass (Lin et al., 2011). The identification of PCNA posttranslational modifications as decision makers in the PRR pathway answered many questions; however, several aspects of the PRR pathway remained puzzling. In yeast, rad5 mutants are more sensitive to DNA damage than mms2 or ubc13 mutants, suggesting that Rad5 has additional functions beyond template-switching

bypass. Moreover, the conventional model of TLS does not explain how particular polymerases are recruited to specific DNA lesions. This is an important aspect of the pathway, since in vitro biochemical experiments demonstrated that some TLS polymerases are more efficient than others at bypassing particular DNA lesions without the generation of errors. Using an elegant experimental design, the authors found that, unexpectedly, HLTF and SHPRH contribute in differential ways to specify DNA damageinduced mutagenesis: HLTF is required for correct bypass of UV lesions, while SHPRH is required for MMS-induced lesions. The authors recognized that this pattern corresponded to the differential participation of TLS polymerases h and k in DNA damage tolerance. Specifically, Polh can accurately insert the correct base pairs across UV lesions, while Polk can bypass alkylated bases (i.e., the kinds of lesions typically induced by MMS) 10-fold more accurately than other polymerases. Indeed, the authors show that MMS induces the formation of Rad18-SHPRHPolk complexes, allowing correct bypass

Molecular Cell 42, April 22, 2011 ª2011 Elsevier Inc. 141