An RB insensitive to CDK regulation

ABSTRACT

The N-term phosphorylation of Retinoblastoma (RB) by the p38 stress-activated protein kinase (SAPK) makes RB insensitive to cyclin-dependent kinase (CDK)-Cyclin inhibition, which enhances the transcriptional repression of E2F-driven promoters and delays tumor cell growth. This novel mechanism of RB regulation opens up a window for developing new cancer drug treatments for tumors harboring high CDK-Cyclin activity and a wild-type RB gene.

Upon stress, cells deploy a plethora of mechanisms in order to adapt and maximize their chances of survival.¹ Among these mechanisms, the control of cell division is essential. Although it is well established that stress-activated protein kinases (SAPKs) such as p38 play a key role in cell cycle regulation,^2,3 the molecular mechanisms employed to achieve such control are not yet fully understood. In the G1-S transition, p38 regulates core components of the cell cycle machinery such as the cyclin-dependent kinase (CDK) inhibitors p21^Cip1, p27^Kip1 and p57^Kip2.⁴ p57^Kip2 phosphorylation by p38 enhances its association to CDK2-Cyclin complexes, leading to reduced CDK2 activity and a transient G1 cell cycle arrest. Moreover, cells lacking p^57Kip2 fail to properly arrest at G1 and show reduced cell viabiligy in response to stress.^4,5 However, stressed p57^Kip2 knockout cells are still able, albeit to a lesser extent, to delay G1, which is a clear indication that other mechanisms that delay cell cycle progression might exist.

Remarkably, upon stress, p38 also inhibits the transcription of E2F-dependent genes, which are necessary for cell cycle progression. Since E2F-mediated gene transcription is repressed by the binding of the tumor suppressor Retinoblastoma (RB) to E2F transcription factors,⁶ it was likely that p38 was somehow controlling RB function. Indeed, activated p38 specifically phosphorylates two residues at the N-terminus of RB, the serine 249 (S249), and the threonine 252 (T252). Upon phosphorylation at these two sites, RB is recruited longer to E2F-driven gene promoters where it represses E2F-mediated gene transcription. This event leads to a delayed G1 cell cycle progression and a slower cell division rate (Fig. 1A). Most notably, mutation of S249 and T252 to nonphosphorylatable residues such as alanine makes RB insensitive to p38 regulation, whereas the mutation of these two sites to phosphomimetic residues such as glutamic acid makes RB to behave as a constitutive transcriptional repressor regardless of p38 function and cell stress.⁷

Figure 1.

N-terminal phosphorylation renders Retinoblastoma (RB) insensitive to regulation by cyclin-dependent kinases (CDKs). (A) Unphosphorylated RB associates to E2F-driven gene promoters inhibiting transcription by binding to E2F transcription factors through the RB pocket and C-term domains. When CDK-Cyclin complexes phosphorylate such domains of RB, the repressor dissociates from E2F, eliciting transcription and cellular progression through G1-S. Upon stress, N-terminal phosphorylation of RB by p38 restricts E2F transcriptional activity, delaying cell cycle progression and cell proliferation, despite the phosphorylation by CDKs. (B) Tumor progression can be caused by high CDK activity. The phosphorylation by p38, mutations that mimic it, or a drug that promotes a conformational change that resembles the phosphorylation by p38 could override RB inactivation by CDKs, leading to tumor growth arrest.

It is well established that RB controls the restriction point that licenses cell transition from G1 to S, thus allowing cell division and proliferation.⁸ During G1, RB remains mostly in its unphosphorylated form, and it is found on E2F-driven gene promoters inhibiting transcription by interacting with, and repressing, E2F transcription factors as well as recruiting chromatin remodelers and transcriptional co-repressors. RB binding to E2F is mediated through the RB pocket and C-term domains. Upon mitogenic cues, the coordinated activation of G1 and S phase CDK-Cyclin complexes phosphorylate multiple sites along the pocket and C-term domains of RB and consequently, RB dissociates from E2F.⁹ Once relieved from its inhibitor, E2F can start transcribing genes. Thus, there is an opposite effect of the phosphorylation of RB by CDKs to those by p38. Of note, the phosphomimetic mutant of S249 and T252 represses transcription even in the presence of high CDK activity, suggesting a dominant effect for these N-term phosphorylations over the inactivation of RB by CDK phosphorylations.⁷ These results pointed out that a noncharacterized domain on the RB N-term is responsible for regulating RB activity. In silico modeling of the N-term of RB showed that the residues between 244 and 269 had a strong resemblance to the C-term motif involved in E2F binding (residues 829 to 852) and that the N-term of RB might serve as a novel docking platform for these transcription factors binding. In vitro binding assays demonstrated that the N-terminus of RB could associate to E2F when both S249 and T252 were phosphorylated. Importantly, RB remained bound to E2F even in the presence of a CDK-mediated phosphorylated pocket and C-term domains. This unprecedented observation indicates that N-term phosphorylation could override RB inactivation by CDKs, thus leading to cell growth arrest. Indeed, the RB phosphomimetic mutant reduces tumor cell growth in an orthotopic mouse model.⁷

The discovery of RB over two decades ago, the first gene shown to have a tumor suppressor activity, raised the expectations for drug intervention. Additionally, the INK4/RB/E2F/CDK-Cyclin pathway plays an essential role in controlling cell division fitness, prevents abnormal cell proliferation, and is altered in a large proportion of human cancers.¹⁰ However, despite its relevance in human tumor development, the finding of specific drugs targeting this pathway has been proven difficult. Early this year, the US Food and Drug Administration approved the use of palbociclib (Ibrance™), a small molecule compound developed by Pfizer for the treatment of ER-positive and HER2-negative breast cancer. This is up to date the first commercially available drug for cancer treatment that targets the RB pathway by specifically inhibiting CDK4-Cyclin activity. We believe that the discovery of a novel mechanism of RB activity opens the possibility for the development of a new kind of cancer drugs by developing compounds capable of promoting the association of E2F transcription factors to the N-term of RB. In this scenario, RB would become insensitive to CDK-Cyclin inactivation, and its activity as a tumor suppressor would increase (Fig. 1B). Many human tumors with high CDK-Cyclin levels still carry a functional RB gene, and therefore, they might benefit of such compounds.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

The FP and EdN's laboratory is supported by grants from the Spanish Ministry of Economy and Competitiveness (BFU2015–64437-P and FEDER, BFU2014-52125-REDT, and BFU2014-51672-REDC to FP; BFU2014–52333-P and FEDER to EdN), the Catalan Government (2014 SGR 599) and the Fundación Botín, and the Banco Santander through its Santander Universities Global Division to FP. FP and EdN are the recipients of an ICREA Acadèmia (Generalitat de Catalunya).