Signal Transduction: How Rad53 Kinase Is Activated

Dispatch R769

Signal Transduction: How Rad53 Kinase Is Activated Recent studies have elucidated the activation mechanism of the Rad53 checkpoint kinase and the role of Rad9-like adaptor proteins in mediating signal transduction from PIKK sensor kinases that detect DNA damage to the effector kinases that play a part in mending that damage. Achille Pellicioli and Marco Foiani Are you lost in the intricate signaling networks that mediate ‘checkpoint’ responses which act to preserve genome integrity in the face of DNA damage? It would not be surprising if you were, given the complexity of these networks. But help should come from recent studies in budding yeast which have shed some new light on the mechanism of activation of the Rad53 kinase. Using powerful proteomic approaches, Smolka et al. [1] and Sweeney et al. [2] have succeeded in mapping and characterizing the phosphorylation residues of the Rad53 kinase (Figure 1), revealing how the sensor PIKK — PI (3) kinase-like kinase — checkpoint kinases activate and modulate the Rad53-like effector kinases. Rad53 was initially identified in 1991 [3] in a biochemical screen as a serine/threonine/tyrosine kinase, but became soon known as a key guardian of genome integrity [4]. We now know that the Rad53 family members, which include the Chk2 protein in human cells, play a central role in the signal transduction pathway that is activated in response to DNA lesions and helps to prevent genome rearrangements and cancer [5]. At present, the mechanisms leading to activation of the socalled ‘genome integrity checkpoints’ are best understood in the model yeast species Saccharomyces cerevisiae and Schizosaccharomyces pombe. Briefly, the DNA damage-induced formation of single-stranded DNA regions coated by RP-A trigger the recruitment of checkpoint

sensors such as the PIKK Mec1 in budding yeast (or ATR in human cells) [6]. In addition, a number of checkpoint factors, including Rad9, associate with the sites of damaged DNA and are phosphorylated by Mec1 [7]. However, an important issue still awaits elucidation, and that is how the initial checkpoint response to DNA damage is finally amplified to a global cellular response which ultimately results in cell cycle delay, DNA repair and, in higher eukaryotes, in the apoptotic program [5], or even in the activation of the innate immune response that leads to the elimination of damaged cells [8]. From the observation that overexpression of Rad53 kinase in bacteria results in its autophosphorylation and activation in the absence of DNA damage, Lowndes and collaegues [9] proposed that Rad53 might be

activated simply by processes that facilitate a local increase in its concentration. This was confirmed by hydrodynamic analysis of protein complexes isolated from UV-treated cells, which showed that the Rad9 protein acts as a scaffold for concentrating Rad53 (Figure 2A). Following in trans autophosphorylation, the Rad53 kinase is released from the Rad9 complex, thus amplifying the alarm signal throughout the cell [9]. The solid-state catalyst model proposed for Rad53 activation [9] does not, however, explain the role of PIKKs in Rad53 signaling, whereas the observation that Mec1 mediates the in trans phosphorylation of a kinase defective Rad53 protein clearly indicates PIKKs are involved [10]. Further, although Rad53 undergoes autophosphorylation events even in in situ renaturation kinase assays — ruling out the involvement of putative associated limiting factors required for the catalytic reaction — this reaction still requires the presence of functional Mec1 and Rad9 proteins [10]. Using mass spectrometry based methods, Smolka et al. [1] and Sweeney et al. [2] have determined the phosphorylation sites of Rad53 in yeast cells

Phospho-residues according to Sweeney et al. [2] S795 S793 S791 S789 S750 S748 S746 S745

T543 S547 S171 T170

S184 S185

S165

1

T354

S375

S198 S350

FHA1 S24 S49

S373 S411

S489 S485

S560 T563 S568 S569

Kinase S175

S350

T372 S373

S747

S424 S489

S375

821

FHA2

S560 S547 T543

S748 S774 S789 S793

Phospho-residues according to Smolka et al. [1] Current Biology

Figure 1. Rad53 phosphorylation sites. Relevant Rad53 domains are indicated as follows: the gray boxes indicate the S/T-Q clusters; the green box at the carboxyl-end of the protein represents a bipartite nuclear localization signal; the red box inside the kinase domain (brown) represents the activation segment. Proline-directed sites are underlined and bold.

Current Biology Vol 15 No 18 R770

A

Pre-active Rad53-Rad9 complex Active Rad53 Rad53 release

Inactive Rad53

B Pre-active Mec1-Rad9-Rad53 complex

Rad9 1

Active Rad53

Inactive Rad53 3 Rad53 release

Mec1 2

C

Pre-active Mec1-Rad9-Rad53 complex 1

Rad9

Active Rad53

Inactive Rad53

Rad53 release

Mec1

2 2

3 Current Biology

Figure 2. Models for Rad53 activation. (A) The solid-state catalyst model proposed by Lowndes and colleagues [9]. Hyperphosphorylation of Rad9 molecules (dark blue) results in the recruitment of Rad53 protein kinase (yellow). Rad53 molecules are thus concentrated, facilitating in trans autophosphorylation reactions. The small black round shapes indicate the phospho-residues. Autophosphorylation of Rad53 results in its release and transition into the active state (green). Inactive Rad53 molecules (red) are then re-cycled to form pre-active Rad9-Rad53 complexes. (B) The adaptor-based model proposed by Durocher and colleagues [2]. Mec1-dependent hyperphosphorylation of Rad9 protein (step 1) promotes the recruitment of inactive Rad53 protein. Mec1 then phosphorylates Rad53 (step 2) which undergoes through autophosphorylation (step 3) and release from Rad9. (C) The adaptor-catalyst model. Mec1 phosphorylates Rad9 protein (step 1) at multiple sites. A threshold of Rad9 phosphorylation is required to recruit and phosphorylate Rad53 at multiple sites (step 2). Rad53 then undergoes autophosphorylation: Rad9 oligomerizes and facilitates Rad53 in trans autophosphorylation (step 3) by increasing the local concentration of Rad53 molecules. Active Rad53 kinases are then released from the complex.

experiencing DNA damage after treatment with 4-nitroquinoline oxide (4-NQO, a UV light mimetic compound) or methyl metanesulfonate (MMS), respectively (Figure 1). Moreover, by an in vitro kinase assay with the purified proteins, Sweeney et al. [2] were able to reconstitute Mec1-dependent Rad53 signaling, demonstrating unambiguously that Mec1 phosphorylates Rad53 even in the absence of the catalytic activity of the effector kinase. Intriguingly, the hyperphosphorylated form of Rad9 represents a limiting and essential element for the Mec1mediated Rad53 signaling, supporting the view that Rad9 is an adaptor required by Mec1 to

target Rad53 molecules (Figure 2B). Indeed, the Rad9 protein is a perfect molecular bridge for connecting PIKKs and Rad53-like kinases, as it is itself phosphorylated by Mec1 at multiple sites [11,12], and the Rad9 hyperphosphorylated segments interact with the two Rad53 forkhead-associated (FHA) domains, specialized in targeting phospho-peptides [13,14]. Considering that the Rad9 protein is dispensable for Mec1dependent-Rad53 signaling in response to replication blocks induced by hydroxyurea treatment [10], it is likely that the Mrc1 protein carries out the adaptor function under these conditions

[15]. So, it will be of great interest to confirm the existence of other Rad9-like PIKK-adaptor factors, not only in yeast cells, but also in higher eukaryotes, where the PIKK-dependent networks are certainly much more complex than those of yeast cells [16]. Impressively, the Rad53 protein turns out to be phosphorylated at more than twenty residues (Figure 1) and, whereas a number of them are generated by autophosphorylation events, as demonstrated by Sweeney et al. [2], many others result from in trans phosphorylation mediated by apical kinases. As previously anticipated [17], most of the in trans phosphorylated sites are located in S/T-Q clusters, the

Dispatch R771

typical consensus for PIKKs. Further, both studies [1,2] found proline-directed phosphorylated sites in Rad53, suggesting that the cell cycle kinase CDK1 might directly modulate Rad53 activity, as also supported by recent observations implicating CDK1 in checkpoint activation [18]. We note that the DNA damagedependent phospho-residues found by the two studies [1,2] do not always overlap (Figure 1). In particular, Smolka et al. [1] found phospho-residues in the S/T-Q cluster at the amino-terminal part of Rad53 in cells damaged with MMS, whereas the same sites were not identified by Sweeney et al. [2] in cells treated with 4NQO (Figure 1). An attractive hypothesis is that, depending on the nature of the primary DNA lesions, the in trans phosphorylation by PIKKs is required not only to activate Rad53, but also to direct the active form of the kinase to the appropriate signaling pathway, by creating specialized phosphointerfaces in the Rad53 protein. Indeed, the genetic requirements influencing Rad53 activation in response to MMS or 4-NQO are different [10,19]. Considering that Rad53 and Rad9 are phosphorylated at multiple sites, the two models for Rad53 activation may not be necessarily exclusive (Figure 2A,B) and could be reconciled in a stepwise model in which Rad9 acts both as an adaptor mediating the interaction between Mec1 and Rad53 and as a scaffold protein facilitating the concentration of Rad53 molecules (Figure 2C). Once again, phosphorylation at multiple sites emerges as a common strategy in kinase-based signal transduction processes. This mechanism of activation, in theory, would be ideal to avoid futile activation of the pathway when the signal is below a certain threshold. Further, the redundancy of phospho-sites represents a warranty for the cell against mutations on those residues, as the inactivation of one of these sites should not necessarily impair the transduction cascade.

References 1. Smolka, M.B., Albuquerque, C.P., Chen, S.H., Schmidt, K.H., Wei, X.X., Kolodner, R.D., and Zhou, H. (2005). Dynamic changes in protein-protein interaction and protein phosphorylation probed with amine reactive isotope tag. Mol. Cell. Proteomics., in press. 2. Sweeney, F.D., Yang, F., Chi, A., Shabanowitz, J., Hunt, D.F., and Durocher, D. (2005). Saccharomyces cerevisiae Rad9 acts as a Mec1 adaptor to allow Rad53 activation. Curr. Biol. August 9 issue. 3. Stern, D.F., Zheng, P., Beidler, D.R., and Zerillo, C. (1991). Spk1, a new kinase from Saccharomyces cerevisiae, phosphorylates proteins on serine, threonine, and tyrosine. Mol. Cell. Biol. 11, 987–1001. 4. Allen, J.B., Zhou, Z., Siede, W., Friedberg, E.C., and Elledge, S.J. (1994). The SAD1/RAD53 protein kinase controls multiple checkpoints and DNA damageinduced transcription in yeast. Genes Dev. 8, 2401–2415. 5. Bartek, J., Falck, J., and Lukas, J. (2001). CHK2 kinase–a busy messenger. Nat. Rev. Mol. Cell Biol. 2, 877–886. 6. Carr, A.M. (2003). Molecular biology. Beginning at the end. Science 300, 1512–1513. 7. Muzi-Falconi, M., Liberi, G., Lucca, C., and Foiani, M. (2003). Mechanisms controlling the integrity of replicating chromosomes in budding yeast. Cell Cycle 2, 564–567. 8. Gasser, S., Orsulic, S., Brown, E.J., and Raulet, D.H. (2005). The DNA damage pathway regulates innate immune system ligands of the NKG2D receptor. Nature 436, 1186–1190 . 9. Gilbert, C.S., Green, C.M., and Lowndes, N.F. (2001). Budding yeast Rad9 is an ATP-dependent Rad53 activating machine. Mol. Cell 8, 129–136. 10. Pellicioli, A., Lucca, C., Liberi, G., Marini, F., Lopes, M., Plevani, P., Romano, A., Di Fiore, P.P., and Foiani, M. (1999). Activation of Rad53 kinase in response to DNA damage and its effect in modulating phosphorylation of the lagging strand DNA polymerase. EMBO J. 18, 6561–6572. 11. Emili, A. (1998). MEC1-dependent phosphorylation of Rad9p in response to DNA damage. Mol. Cell 2, 183–189.

12.

13.

14.

15.

16.

17.

18.

19.

Vialard, J.E., Gilbert, C.S., Green, C.M., and Lowndes, N.F. (1998). The budding yeast Rad9 checkpoint protein is subjected to Mec1/Tel1-dependent hyperphosphorylation and interacts with Rad53 after DNA damage. EMBO J. 17, 5679–5688. Sun, Z., Hsiao, J., Fay, D.S., and Stern, D.F. (1998). Rad53 FHA domain associated with phosphorylated Rad9 in the DNA damage checkpoint. Science 281, 272–274. Durocher, D., Henckel, J., Fersht, A.R., and Jackson, S.P. (1999). The FHA domain is a modular phosphopeptide recognition motif. Mol. Cell 4, 387–394. Alcasabas, A.A., Osborn, A.J., Bachant, J., Hu, F., Werler, P.J., Bousset, K., Furuya, K., Diffley, J.F., Carr, A.M., and Elledge, S.J. (2001). Mrc1 transduces signals of DNA replication stress to activate Rad53. Nat. Cell Biol. 3, 958–965. Shiloh, Y. (2003). ATM and related protein kinases: safeguarding genome integrity. Nat. Rev. Cancer 3, 155–168. Lee, S.J., Schwartz, M.F., Duong, J.K., and Stern, D.F. (2003). Rad53 phosphorylation site clusters are important for Rad53 regulation and signaling. Mol. Cell. Biol. 23, 6300–6314. Ira, G., Pellicioli, A., Balijja, A., Wang, X., Fiorani, S., Carotenuto, W., Liberi, G., Bressan, D., Wan, L., Hollingsworth, N.M., et al. (2004). DNA end resection, homologous recombination and DNA damage checkpoint activation require CDK1. Nature 431, 1011–1017. Giannattasio, M., Lazzaro, F., Longhese, M.P., Plevani, P., and Muzi-Falconi, M. (2004). Physical and functional interactions between nucleotide excision repair and DNA damage checkpoint. EMBO J. 23, 429–438.

FIRC Institute of Molecular Oncology Foundation, Via Adamello 16, 20139, Milano, Italy and Dipartimento di Scienze Biomolecolari e Biotecnologie, Università degli Studi di Milano, Via Celoria 26, 20133, Milano, Italy.

DOI: 10.1016/j.cub.2005.08.057

Potassium Channels: Complete and Undistorted The recently determined structure of a mammalian voltage-gated potassium channel has important implications for our understanding of voltage-sensing and gating mechanisms in channels. It is also the first crystal structure of an overexpressed eukaryotic membrane protein. Alessandro Grottesi, Zara A. Sands and Mark S.P. Sansom Voltage-gated potassium (Kv) channels are potassium selective ion channels which are activated by a change in transmembrane voltage. They play a key role in the physiology of excitable cells

[1], and are related to voltagegated sodium and calcium channels. Thus, an understanding of the mechanism of Kv channels will also inform our understanding of sodium and calcium channels, as illustrated by the recently reported crystal structure of a mammalian voltage-gated potassium channel [2].