Accumulation of p16 INK4a in mouse ®broblasts as a function of replicative senescence and not of retinoblastoma gene status Ignacio Palmero 1,3 , Beth McConnell 1 , David Parry 1,4 , Sharon Brookes 1 , Eiji Hara 1 , Stewart Bates 1,5 , Parmjit Jat 2 and Gordon Peters 1 1 Molecular Oncology Laboratory, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London WC2A 3PX and 2 Ludwig Institute for Cancer Research, Courtauld Building, 91 Riding House Street, London W1P 8BT, UK Viral transformation of mouse and human ®broblasts has very dierent eects on the composition of cyclin- dependent kinase (Cdk) complexes. In human cells transformed by the large T-antigen of simian virus 40 (SV40 T-Ag) and human tumour cell lines that lack a functional retinoblastoma gene product (pRb) no cyclin D1-Cdk4 complexes can be detected because all the available Cdk4 is associated with the Cdk-inhibitor p16 INK4a . In contrast, SV40-transformed mouse cells and ®broblasts from Rb1-nullizygous mouse embryos contain normal levels of cyclin D1-Cdk4 complexes. To investi- gate this species dierence, we have compared the biochemical properties and expression of mouse p16 INK4a with that of its human counterpart. There is a marked increase in p16 RNA and protein levels as primary embryo ®broblasts approach their ®nite lifespan in culture, but mouse p16 expression does not appear to be in¯uenced by the status of pRb. Transformed or spontaneously immortalized mouse cells therefore do not achieve the very high levels of p16 characteristic of pRb-negative human cell lines. We suggest that these dierences may be related to the dierent frequencies with which mouse and human cells can be immortalized in culture. Keywords: CDKN2a; tumor suppressor; retinoblasto- ma protein; SV40 T-antigen; replicative lifespan; Cdk inhibitor; Introduction The CDKN2a INK4a tumour suppressor gene on human chromosome 9p21 encodes a 16 kDa protein (p16 INK4a ) that speci®cally binds to and inhibits the activity of the cyclin-dependent kinases Cdk4 and Cdk6 (Serrano et al., 1993; Kamb et al., 1994; Nobori et al., 1994). These kinases, in conjunction with their regulatory subunits, cyclins D1, D2 and D3, operate in the G1 phase of the cell cycle by initiating the phosphorylation and functional inactivation of the retinoblastoma protein (reviewed in Sherr, 1994; Weinberg, 1995). The phosphorylation of pRb appears to be a critical component of the so-called restriction point in late G1 at which cells become committed to proceed with division irrespective of extracellular growth signals (Pardee, 1989; Weinberg, 1995). The functions of the p16 and pRb tumour suppressors are therefore intimately connected and the two are viewed as components of the same pathway (reviewed in Strauss et al., 1995; Hall and Peters, 1996). Loss of p16 function would in theory promote cell proliferation via the inactivation of pRb. Several observations support this contention. First, tumour-speci®c mutations in p16 and pRb are generally mutually exclusive (Okamoto et al., 1994, 1995; Otterson et al., 1994; Aagaard et al., 1995; Shapiro et al., 1995; Washimi et al., 1995). Second, the introduction of p16 by microinjection or by peptide- mediated transport causes cells to arrest in G1, provided that p16 is added prior to the restriction point (Lukas et al., 1995b; FaÊ hraeus et al., 1996). Third, the cell cycle arrest imposed by p16 is only apparent in cells that retain functional pRb (Guan et al., 1994; Okamoto et al., 1994; Koh et al., 1995; Lukas et al., 1995b; Medema et al., 1995; Serrano et al., 1995; FaÊ hraeus et al., 1996). Indeed, in cells that lack pRb, the D-cyclins and their associated Cdks are essentially redundant (Lukas et al., 1994, 1995a). An interesting manifestation of this occurs in human tumour cell lines in which the lack of pRb correlates with the complete absence of the complexes between D-cyclins and either Cdk4 or Cdk6 (Xiong et al., 1993; Bates et al., 1994b; Tam et al., 1994; Parry et al., 1995). We recently showed that the high levels of p16 in transformed or pRb-negative human cells arise through two distinct mechanisms (Hara et al., 1996). As originally postulated by Serrano et al. (1993), the expression of p16 in human cells is negatively regulated by pRb (Li et al., 1994; Hara et al., 1996). This occurs at the transcriptional level but the magnitude of the eect is not sucient to account for the approximately 50-fold dierence in steady state RNA levels observed in, for example, primary versus SV40-transformed human ®broblasts. The remaining dierence can be explained by the accumulation of p16 as cells approach their limit of population doublings in culture. Thus, senescent human ®broblasts contain substantially more p16 RNA and protein than early passage cells (Alcorta et al., 1996; Hara et al., 1996; Wong and Riabowol, 1996). Although the extent to which p16 function contributes to the senescent phenotype has yet to be established, such observations provide a plausible explanation for the high frequency of p16 alterations observed in human tumour cell lines as opposed to primary tumours (reviewed in Hall and Peters, 1996). In our initial attempts to correlate cyclin D-Cdk complexes with pRb status, we noted a striking Correspondence: G Peters Present addresses: 3 Centro Nacional de Biotecnologia, Campus Universidad Autonoma, Cantoblanco, 28049 Madrid, Spain; 4 DNAX Research Institute, 901 California Avenue, Palo Alto, California 94304-1104, USA; 5 ABL-Basic Research Program, NCI Frederick Cancer Research and Development Center, Frederick, MD 21702-1201, USA Received 16 January 1997; revised 16 April 1997; accepted 16 April 1997 Oncogene (1997) 15, 495 ± 503  1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00