Oncogene (2000) 19, 6594 ± 6599 ã 2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00 www.nature.com/onc Development of anticancer drugs targeting the MAP kinase pathway Judith S Sebolt-Leopold*,1 1P®zer Global Research and Development, Ann Arbor Laboratories, Ann Arbor, Michigan, MI 48105, USA Since the discovery of the role of ras oncogenes in of low molecular weight G-proteins, ERK has been the tumorigenesis, we have witnessed an explosion of best characterized and is more pertinent to aberrant research in the signal transduction area. In the quest to signaling in human cancer. For some cancers, espe- understand how Ras transmits extracellular growth cially those of hematopoietic origin, the p38 and jun signals, the MAP kinase (MAPK) pathway has emerged kinase pathways may in fact yield targets exploitable as the crucial route between membrane-bound Ras and for anticancer drug development. However, a broad the nucleus. The MAPK pathway encompasses a cascade array of solid tumors is known to express constitutive of phosphorylation events involving three key kinases, levels of phosphorylated ERK1 and ERK2. Activation namely Raf, MEK (MAP kinase kinase) and ERK of ERK is critical for a large number of Ras-induced (MAP kinase). This kinase cascade presents novel cellular responses. Included among these responses is opportunities for the development of new cancer transcriptional activation of multiple genes (Hill and therapies designed to be less toxic than conventional Treisman, 1995). The best-characterized physiological chemotherapeutic drugs. Furthermore, as a signal substrates of ERK are ternary complex factors (TCFs), transduction-based approach to cancer treatment, inhibi- which are directly phosphorylated by ERK to activate tion of any one of these targets has the potential for their transcription activation potential (Gille et al., translational pharmacodynamic evaluation of target 1992; Janknect et al., 1993; Marais et al., 1993). TCFs, suppression. The rationale for targeting the MAP kinase in association with serum response factor, is thought to pathway will be reviewed here along with a discussion of be critical for the activation of numerous mitogen- various pharmacological approaches and the promise inducible genes (Hill and Treisman, 1995). they hold for a new generation of anticancer drugs. Many molecules ultimately contribute to activation Oncogene (2000) 19, 6594 ± 6599. of the Ras-ERK pathway, including a number that are involved in protein-protein interactions. With respect Keywords: mitogen-activated protein kinase (MAPK); to pharmacological intervention, it is generally dicult extracellular signal-regulated kinase (ERK); MAP to selectively target the binding site shared by two kinase kinase (MEK); raf proteins. It is therefore not coincidental that the development of agents targeting the Ras-MAPK path- way has largely focused on the design of small Introduction molecule inhibitors of enzyme function. As will be explored in more detail below, four proteins have Many receptor tyrosine kinases and cytokine receptors in emerged as key players in the quest to intervene in this association with heterotrimeric G proteins are known to pathway: Ras, Raf, MEK (MAP kinase kinase), and activate intracellular protein serine/threonine kinases ERK. Ras is the subject of a paper that appears termed mitogen-activated protein kinases (MAPKs). Of elsewhere in this review issue and therefore will not be the various families of MAPKs, which are also referred covered further here. to as extracellular signal-regulated kinases (ERKs), the ®rst to be characterized were ERK1 and ERK2. Both of Rationale for targeting the MAP kinase pathway these ERKs are activated in response to diverse extracellular stimuli and by protooncogene-encoded Figure 1 provides a simpli®ed schematic representation proteins that induce proliferation. A cascade of of the signaling events leading to activation of the MAP phosphorylation events downstream from Ras activates kinase pathway. Initially, Ras interacts with and these kinases. Upstream regulation of the MAP kinase activates the serine/threonine protein kinase Raf1 in a pathway is complex as evidenced by the number of GTP-dependent manner (Daum et al., 1994; Stokoe et functions ful®lled by its activation. Processes impacted al., 1994). A family of Raf protein kinases has been by MAPK activation encompass the cytoplasm, nucleus, identi®ed and is comprised of A-Raf, B-Raf, and c- cytoskeleton, and the membrane. The reader is referred Raf1. It has been suggested that this family of kinases, elsewhere for comprehensive reviews on the subject of which is known to regulate proliferation, dierentia- regulation through MAPK cascades (Cobb, 1999; Lewis tion, and apoptosis, have both overlapping and unique et al, 1998; Kolch, 2000). regulatory functions (Hagemann and Rapp, 1999). For The Raf-MEK-ERK pathway represents one of the example, transfection of oncogenic H-ras led to a best characterized Ras signaling pathways. Raf and preferential activation of endogenous c-Raf1 as op- MEK have consequently emerged as key protein posed to A-Raf (Weber et al., 2000). Mutated Raf-1 is kinases to target for anticancer drug design. While constitutively active and possesses in vitro transforming there exist multiple MAP kinase families, e.g. jun potential (Stanton and Cooper, 1987). The potential for kinase and p38, which are also activated downstream Raf-1 to play a broad role in tumorigenesis is evidenced by its ability to become activated by either PKCa or the antiapoptotic protein Bcl-2 in a Ras-independent *Correspondence: JS Sebolt-Leopold manner (Kolch et al., 1993; Wang et al., 1996). Antitumor potential of Raf, MEK and MAPK inhibitors JS Sebolt-Leopold 6595 Theoretically, it could be argued that intervention in any of the kinase events in the MAPK cascade could represent a viable approach to crippling tumor growth. If so, then Raf-1, MEK, and ERK all emerge as reasonable anticancer drug targets. The advent of high volume screening of pharmaceutical libraries for small molecule inhibitors has most certainly produced reasonable drug candidates targeting all steps of this pathway. For example, a cascade assay has been reported that is capable of identifying inhibitors of cRaf1, MEK1, or ERK2 (McDonald et al, 1999). As we now turn to preclinical and clinical evaluation of these small molecule inhibitors, it is important to keep in mind that their ultimate promise or dierences may depend as much on their pharmacological attributes as on the merits of their targeted kinase. It should be noted that the identi®cation of pathway components in the Ras-MAP kinase pathway is likely Figure 1 Schematic representation of the Ras ± MAP kinase incomplete. For example, a Raf-1-interacting protein, pathway. The MAP kinase cascade contains three sequential RKIP, has recently been reported (Yeung et al., 1999). kinases: a MAP kinase kinase kinase (Raf), MAP kinase kinase This protein inhibits the phosphorylation and activation of (MEK), and MAP kinase (ERK) MEK by Raf-1 and has also been shown to co-localize with Raf-1. It has been proposed that RKIP binding to either Raf-1 or MEK dissociates Raf-MEK complexes, thereby Importantly, raf mutations have been identi®ed in a interrupting MEK activation and downstream signaling range of human tumors (Storm and Rapp, 1993). (Yeung et al., 2000). Discovered with the use of a yeast Independent of its mutation status, Raf is also activated two-hybrid system, the relevance of RKIP expression to in tumor cells containing enhanced growth factor signal transduction in tumor cells is unclear at the present signaling pathways, such as those induced by mutant time. Furthermore, until we learn whether RKIP expres- or constitutively expressed Ras or EGF receptor family sion is negatively regulated, it remains unclear how to members. Therefore, the collective evidence suggests pharmacologically elevate its expression to impair tumor that Raf-1 is a viable anticancer drug target. growth. Although highly speculative based on our current Alternatively, targeting the molecule immediately knowledge of the role of RKIP, it is conceivable that downstream from Raf, that is, the dual speci®city elevated expression of this protein could oer tumor cells a kinase MEK can also be envisioned as representing a mechanism of resistance to MAPK pathway inhibitors. rational approach to anticancer drug design. Subse- There exist a multitude of other newly discovered quent to its activation, Raf-1 phosphorylates and proteins that may provide insight into the design of activates both MEK1 and MEK2 (hereafter referred novel signal transduction-based cancer therapies that to as MEK) on two distinct serine residues (Dent et al., exploit the MAP kinase pathway. These include Sur-8, 1992; Crews et al., 1992; Her et al., 1993). Activated which is thought to act as a scaold to enhance Ras- MEK then phosphorylates ERK1 and ERK2 on both MAP kinase signaling by facilitating Ras-Raf interac- a tyrosine and a threonine residue (Anderson et al., tion (Li et al., 2000), as well as the kinase suppressor of 1990). No substrates for MEK have been identi®ed Ras (KSR). KSR is also thought to act as a scaolding other than ERK1 and ERK2 (Seger et al., 1992). This protein for the Ras-MAPK pathway (Stewart et al., tight selectivity in addition to the unique ability to 1999). Another interesting protein is MP-1, which has phosphorylate both tyrosine and threonine residues are been reported to enhance activation of the MAPK by consistent with this kinase playing a central role in binding MEK (Schaeer et al., 1998). Last but not integration of signals into the MAPK pathway. least, a novel ERK has recently been identi®ed, Constitutive activation of MEK has been shown to ERK1b, which is an alternatively spliced form of result in cellular transformation (Cowley et al., 1994; ERK1, that appears to be elevated in Ras-transformed Mansour et al., 1994). While MEK has not been cells (Yung et al., 2000).