SHORT REPORT The retinoblastoma tumor suppressor inhibits cellular proliferation through two distinct mechanisms: inhibition of cell cycle progression and induction of cell death Karen E Knudsen 1,5 , Erich Weber 2,4 , Karen C Arden 1,2 , Webster K Cavenee 1,2,3,4 , James R Feramisco 2,4 and Erik S Knudsen* ,5 1 Ludwig Institute for Cancer Research, University of California at San Diego, La Jolla, California, CA 92093-0660, USA; 2 Department of Medicine, University of California at San Diego, La Jolla, California, CA 92093-0660, USA; 3 Center for Molecular Genetics, University of California at San Diego, La Jolla, California, CA 92093-0660, USA; 4 Cancer Center, University of California at San Diego, La Jolla, California, CA 92093-0660, USA; 5 Department of Cell Biology, University of Cincinnati College of Medicine, Cincinnati, Ohio, OH 45267-0521, USA Studies aimed at examining the precise function(s) of the retinoblastoma tumor suppressor protein, RB, have been hindered by the rapid phosphorylation and inactivation of ectopically expressed RB which occurs in the majority of cell types. Therefore, ectopically expressed RB is a poor inhibitor of cellular proliferation. We have designed constitutively active RB proteins, PSM-RB, that cannot be inactivated by phosphorylation. Using these proteins, we show that unlike wild-type RB, PSM-RB proteins inhibit cell cycle progression in a broad range of tumor cell types. Furthermore, unlike p16 ink4a , PSM-RB is also a potent inhibitor of cell cycle progression in RB-de®cient tumor cells. Surprisingly, we identi®ed a tumor cell line that is resistant to the cell cycle inhibitory eects of PSM-RB. This ®nding challenges the hypothesis that RB must be inactivated in all cells for cell cycle progression to occur. Further characterization of this `resistant' tumor line revealed that proliferation of these cells is still inhibited by PSM-RB. We show that this is due to PSM-RB-induced cell death. As such, these studies are the ®rst to show that RB inhibits cellular proliferation through at least two distinct mechanisms ± inhibition of cell cycle progression and induction of cell death. Keywords: apoptosis; cyclin; phosphorylation; cell death; RB The retinoblastoma tumor suppressor protein, RB, is inactivated in over 60% of studied human tumors (Bartek et al., 1997; Sherr, 1996). Inactivation of RB can be achieved through multiple distinct mechanisms, including direct loss of functional protein due to mutation, binding of oncoproteins of DNA tumor viruses, or by overt phosphorylation of RB, which inactivates its growth-suppressing function (Bartek et al., 1997; Sherr, 1996; Wang et al., 1994). It has been hypothesized that RB must be phosphorylated and inactivated by cyclin dependent kinase (CDK)/cyclin complexes in cells to allow for cell cycle progression, and in tumor cells this process is often deregulated (Bartek et al., 1997; Sherr, 1996). For example, ampli®cation of the proto-oncogenes CDK4 or Cyclin D1 results in excessive phosphorylation and inactiva- tion of RB. Likewise, loss of the tumor suppressor protein p16 ink4a which normally serves to attenuate the activity of CDK4 and cyclin D1, results in deregulated phosphorylation and inactivation of RB. It is thought that unphosphorylated (i.e. active) RB prevents cellular proliferation by sequestering a host of factors and assembling complexes which inhibit cell cycle progres- sion (De Pinho, 1998; Sidle et al., 1996; Wang et al., 1994). Phosphorylation of RB by CDK/cyclin com- plexes causes release of these proteins, and thus enables cell cycle advancement. Studies aimed at directly de®ning the role of RB in cell cycle processes have been hindered by the fact that ectopically expressed wild-type RB is a very poor inhibitor of cellular proliferation, since it is phosphory- lated and inactivated by endogenous CDK/Cyclin activity (Wang et al., 1994; Zacksenhaus et al., 1993). Because of this, the majority of studies on wild-type RB have been carried out in one cell line, the osteosarcoma-derived SAOS-2, which fails to phos- phorylate ectopically expressed wild-type RB (Qin et al., 1992; Templeton et al., 1991). We have circum- vented the problem of RB inactivation/phosphoryla- tion by constructing a series of phosphorylation- site mutated RB proteins (PSM-RB), which cannot be inactivated by phosphorylation (Knudsen et al., 1998a; Knudsen and Wang, 1997). These proteins are predicted to be constitutively active for growth suppression. In this report, we tested the ability of PSM-RB to inhibit the proliferation of a large number of tumor cell lines. Surprisingly, these studies revealed that RB inhibits cellular proliferation through at least two distinct mechanisms, one via cell cycle arrest and one via cell death. It has been proposed that phosphorylation of RB is a requirement for cell cycle progression and that a blockade of RB phosphorylation will universally inhibit cell cycle progression (Bartek et al., 1997; Sherr, 1996; Sidle et al., 1996; Weinberg, 1995). To test this hypothesis, we monitored the action of our PSM-RB proteins on cell cycle progression in seven tumor cell lines which contain functional endogenous RB. To monitor the in¯uence of PSM-RB (or other *Correspondence: ES Knudsen Received 1 February 1999; revised 14 April 1999; accepted 14 April 1999 Oncogene (1999) 18, 5239 ± 5245 ã 1999 Stockton Press All rights reserved 0950 ± 9232/99 $15.00 http://www.stockton-press.co.uk/onc