Notch signaling induces SKP2 expression and promotes reduction of p27Kip1 in T-cell acute lymphoblastic leukemia cell lines

Abstract

In T-cell acute lymphoblastic leukemia (T-ALL) NOTCH 1 receptors are frequently mutated. This leads to aberrantly high Notch signaling, but how this translates into deregulated cell cycle control and the transformed cell type is poorly understood. In this report, we analyze downstream responses resulting from the high level of NOTCH 1 signaling in T-ALL. Notch activity, measured immediately downstream of the NOTCH 1 receptor, is high, but expression of the canonical downstream Notch response genes HES 1 and HEY 2 is low both in primary cells from T-ALL patients and in T-ALL cell lines. This suggests that other immediate Notch downstream genes are activated, and we found that Notch signaling controls the levels of expression of the E3 ubiquitin ligase SKP2 and its target protein p27Kip1. We show that in T-ALL cell lines, recruitment of NOTCH 1 intracellular domain (ICD) to the SKP2 promoter was accompanied by high SKP2 and low p27Kip1 protein levels. In contrast, pharmacologically blocking Notch signaling reversed this situation and led to loss of NOTCH 1 ICD occupancy of the SKP2 promoter, decreased SKP2 and increased p27Kip1 expression. T-ALL cells show a rapid G1–S cell cycle transition, while blocked Notch signaling resulted in G0/G1 cell cycle arrest, also observed by transfection of p27Kip1 or, to a smaller extent, a dominant negative SKP2 allele. Collectively, our data suggest that the aberrantly high Notch signaling in T-ALL maintains SKP2 at a high level and reduces p27Kip1, leading to more rapid cell cycle progression.

Introduction

T-cell acute lymphoblastic leukemia (T-ALL) is a malignant disease, originating in thymocytes. T-ALL constitutes a substantial fraction of ALL tumors, both in children and in adults [1]. Current treatment is based on combination chemotherapy, but long-term survival rates are reduced, particularly in older patients, emphasizing the need for improved therapy. The molecular mechanisms underpinning T-ALL are likely to be complex, as a number of genes, including c-MYC, HOX 11, TAL1 and LMO, are involved in chromosomal translocations with the T-cell receptor locus (for review see [1]). Mutations in the Notch signaling pathway have recently emerged as an important genetic component in T-ALL. The involvement of Notch was first observed in rare t(7,9)(q34;q34.3) translocations, which brings an activated form of the NOTCH 1 receptor gene (TAN 1) under control elements from the T-cell receptor gene [2]. More recently, it was shown that more than 50% of all T-ALL patients carry NOTCH1 gain-of-function mutations that, similar to TAN 1, generate an activated form of Notch [3].

The Notch signaling pathway is important for cell fate decisions in many different cell types, including the T-cell lineage [4]. Notch signaling is initiated when the transmembrane Notch receptor interacts with DSL ligands on juxtaposed cells. This leads to two proteolytic processing events in the receptor, first at the extracellular side and subsequently in the plasma membrane. The latter cleavage is executed by the gamma-secretase complex and leads to liberation of the Notch intracellular domain (Notch ICD), which translocates to the nucleus. In the nucleus, Notch ICD binds to the DNA-binding protein CSL, and converts CSL from a repressor to a transcriptional activator [5].

T-ALL mutations lead to aberrantly high Notch signaling and affect two different regions of the NOTCH 1 receptor [3]. One class of mutations falls in the heterodimerization domain (HD), making the receptor more prone to proteolytic processing and thus less dependent on ligand-activation. The other class of mutations occurs in the PEST domain of NOTCH 1, generating truncated versions of the NOTCH 1 ICD, which increases the stability of the normally short-lived intracellular domain [6]. Further mutational analysis of the PEST domain has revealed four serine residues that serve as a negative regulatory sequence, and deletion of these residues makes the Notch ICD more active [7]. The link between deregulated Notch signaling and T-ALL receives further support from mouse models of T-ALL, where Notch 1 mutations occur as secondary hits in mice engineered for deregulation of c-Myc, TAL1, Ikaros and Pbx1 expression [8], [9], [10], [11], [12].

The downstream response of Notch signaling is only partially understood. There is a wealth of data supporting the notion that Hes and Hey genes, which encode bHLH negative transcriptional regulators [13], [14], are immediate downstream genes, but there are situations where Notch signaling exert biological effects without affecting Hes and Hey expression [15]. It is therefore likely that other immediate downstream genes exist that are directly activated by Notch ICD via CSL in specific cellular contexts, and several such genes have recently been identified, e.g. Nrarp, GFAP, GATA2, c-Myc and p21^cip1 (see [16] for review).

In T-ALL, our understanding of which downstream genes are activated, and how this affects cell cycle regulation and leads to cellular transformation, is more limited. The finding that HES 1 activation is not always required for Notch-induced growth promotion [17], may suggest that other downstream responses could be relevant for T-ALL. In keeping with this, c-MYC has recently been identified as an important new direct target gene in some T-ALL cell lines [18]. c-MYC was, however, not activated in all T-ALL cell lines [18], suggesting the existence of other immediate Notch downstream effectors.

In this report, we have addressed the downstream aspects of Notch signaling in T-ALL. We find that T-ALL tumors and cell lines have low levels of HES1 and HEY2 expression. Instead, increased Notch signaling results in elevated levels of SKP2, the F-box component in the SCF^Skp2 E3 ubiquitin ligase complex [19]. The high level of SKP2 was paralleled by low levels of the CDK inhibitor (CKI) p27Kip1, a target substrate for ubiquitinylation and thus negatively regulated by the SCF–Skp2 complex [20] (for review see [19]). Blocking Notch signaling pharmacologically reversed the situation, and this “Notch off” state was associated with reduced SKP2 and increased p27Kip1 expression and subsequent G1 cell cycle arrest. In conclusion, our data suggest that a NOTCH/SKP2/p27Kip1 axis may contribute to the development of T-ALL.