Mechanisms of G₁ checkpoint loss in resected early stage non-small cell lung cancer

Abstract

Loss of the G₁ checkpoint appears to be extremely common among virtually all neoplasms. A variety of genetic and epigenetic mechanisms have been demonstrated to play significant roles in this process. In a consecutive series of early stage non-small cell lung cancer (NSCLC), we have established the loss of expression of the G₁ Cdk inhibitors p15^INK4b and p16^INK4a by DNA methylation is very common (37%), and methylation of p16^INK4a is tightly correlated with loss of expression of p16^INK4a protein (P=0.0018). Furthermore, methylation of p15^INK4b and p16^INK4a appear inversely correlated, although methylation of p15^INK4b is an infrequent event in this cohort (4%). Methylation was detected in all stages of NSCLC equally, and did not correlate with survival in these patients. Evidence for methylation was more frequent in squamous cell carcinomas in comparison to other tumor histologies (P=0.0156). In addition, over-expression of cyclin D1 was found to be tightly restricted (P=0.0032) to those tumors that had retained wild-type expression of pRB, and did not correlate with methylation or expression of p16^INK4a gene product. Although loss of p16^INK4a function remains tightly correlated with pRB expression, loss of other regulatory elements in NSCLC such as p53 mutation and cyclin D1 over-expression appear independent of loss of the p16^INK4a gene product.

Introduction

It is largely accepted that cell cycle regulatory genes play an important role in the development of human cancers [1]. Among the major regulatory genes of the G₁/S transition are the cyclin-dependent kinase inhibitors p15^INK4b and p16^INK4a, the retinoblastoma tumor suppressor protein (pRB), and cyclin D family members. The p16^INK4a gene, located on chromosome 9p2l, functions to maintain pRB in the hypophosphorylated status by inhibiting Cdk4/6:cyclin D complex-mediated phosphorylation of pRB [2]. The presence of hypophosphorylated pRB restrains the cell from entering S phase of the cell cycle [3]. Consequently, loss of either the p16^INK4a or Rb gene product function will result in cell cycle dysregulation. In contrast to p16^INK4a, cyclin D subtypes interact with cyclin-dependent kinases Cdk4 and Cdk6 to generate the D-type Cdk activity which directly phosphorylates pRB [4], [5]. This D-type Cdk activity is thought to be critical for cell passage through G₁ and into the S phase of the cell cycle. In the transformed cell, over-expression of cyclin D appears to be one mechanism that may drive the quiescent cell into S phase [4], [5]. Over-expression of cyclin D1 and abnormal expression of pRB have been reported in a number of human cancers, including lung cancers [6], [7]. In a similar manner, p16^INK4a has been shown to be one of the most frequently mutated or aberrantly expressed molecules in human cancer, including lung cancer, perhaps equaled or exceeded only by loss of p53 function [8]. These data imply that disruption of the Cdk4/6:cyclin D:pRB pathway may be required for the most types of human cancer development. In addition, p15^INK4b, another cyclin-dependent kinase inhibitor that plays an important role in cell cycle control [9], is also aberrant in a variety of cancers. Like its closely related chromosomal neighbor p16^INK4a, p15^INK4b may serve as a tumor suppressor gene in some hematological malignancies [10], [11], [12]. Finally, the p19ARF protein, which is encoded by the beta transcript of the p16^INK4a gene locus and plays a role in regulation of p53 mediated pathways, is also lost in a variety of tumors including lung cancer [13], [14].

Recent reports have indicated that loss of p16^lNK4a expression can be mediated by methylation of the p16^INK4a gene in non-small cell lung cancer (NSCLC) and pre-neoplastic lesions [14], [15], [16], [17], [18], [19]. In these studies, there was found a high degree of correlation between p16^INK4a gene methylation and consequent loss of protein expression. Limited data are available, however, on stage or survival differences based on the presence or absence of p16^INK4a or p15^INK4b methylation. Nonetheless, it is clear that these epigenetic mechanisms of gene inactivation are occurring in both early stage and pre-neoplastic lesions [15], [20].

In order to further address the relation of the mechanisms of loss of p16^INK4a and p15^INK4b to survival, stage, and other acquired genetic lesions, we undertook an analysis of a large cohort of consecutively resected NSCLC tumors. We have previously reported on the protein status of pRB and p16^INK4a in a large consecutive series of 100 resected early stage (50% stage I, 35% stage II) non-small cell lung cancer (NSCLC) [8]. Reflecting what was previously observed in lung cancer cell lines, there was a striking inverse correlation between pRB and p16^INK4a expression with 66% of tumors expressing exclusive loss of either pI6^INK4a or pRB. However, 35% of the tumors retained apparent expression of both of these gene products. We have sought to further characterize these tumors to determine the additional molecular events that underlie the loss of G₁ checkpoint regulation in resected early stage NSCLC, as well as to confirm recent observations on the expression patterns and implications for prognosis conferred by loss of these proteins [15], [21], [22]. In the current project, we have examined these tumors for cyclin D1 over-expression, p16^INK4a missense mutation, and p16^INK4a or p15^INK4b methylation in these NSCLC tumors.