Protein–protein interactions for cancer therapy

Abstract

The p53 tumor suppressor network is frequently disabled by mutation of its molecular node, the p53 gene 1–3. However, the p53 gene is wild type in ≈50% of human cancers. In the early 1990s, the p53–MDM2 autoregulatory feedback loop was discovered (reviewed in refs. 4 and 5) (Fig. 1). MDM2 (murine double minute 2; also termed HDM2 for its human equivalent) is an oncoprotein. In response to cellular stress, p53 transcriptionally transactivates the MDM2 gene, and then the MDM2 protein binds to and transports the p53 to the cytoplasm where MDM2, an E3 ubiquitin ligase, promotes p53 ubiquitination and degradation by the proteasome. Increased expression of MDM2 in human cancer involves four mechanisms: gene amplification, increased expression by activated p53, stabilization by an aberrantly spliced form of HMDX, or augmented translation (4, 6). In addition to these cancer-related mechanisms, functional single-nucleotide polymorphisms (SNP) may modulate MDM2 expression. For example, increased p53-mediated expression of MDM2 due to an SNP at nucleotide 309 (SNP309) in the MDM2 gene occurs in the germ line of the population that can increase tumor progression (7). Because MDM2 is overexpressed in certain cancers (8) and may reduce the effectiveness of p53-dependent cancer therapies, the disruption of the p53–MDM2 autoregulatory feedback loop emerged as a molecular targeting strategy. Although the interactive binding sites on p53 and MDM2 were well defined by crystallography (9), the strategy of using drug-like small molecules to block such protein–protein interactions was not considered as attractive by both academia and the pharmaceutical industry as were inhibitors of key cancer-related enzymes such as kinases. Challenging this widely held dogma, Vassilev et al. (10) developed a class of small molecules, the nutlins, that occupy the p53-binding pocket in MDM2, prevent its binding to p53, and, thus, facilitate the p53 tumor suppressor network to inhibit human cancer cell lines in vitro and as xenografts in vivo. Vassilev and coworkers (11), in this issue of PNAS, extend these initial observations by demonstrating that the nutlin-3, a tetra-substituted imidazoline, induces apoptosis most robustly in cancer cell lines with increased MDM2 expression, and this response correlates with the in vivo antitumor efficacy of nutlin-3.