Search

Proteins involved in class I MHC (MHC-I) Ag processing, such as the TAP, are deficient in some human tumor cells. This suggests that antitumor responses by CD8 T cells provide selection pressure to favor outgrowth of cells with defective processing of tumor Ags. Nonetheless, this evidence is only correlative, and controlled in vivo experiments have been lacking to demonstrate that TAP deficiency promotes survival of tumor cells. To explore the role of Ag processing defects in tumor progression, matched panels of TAP1-positive and TAP1-negative tumor cell lines were generated from a parental transformed murine fibroblast line. Inoculation of C57BL/6 mice with TAP1-negative cells produced large and persistent tumors. In contrast, TAP1-positive cells did not generate lasting tumors, although small tumors were detected transiently and regressed spontaneously. Both TAP1-positive and TAP1-negative cells produced tumors in athymic mice, confirming that TAP-dependent differences in tumorigenicity were due to T cell-dependent immune responses. Inoculation of C57BL/6 mice with mixtures of TAP1-positive and TAP1-negative cells produced tumors composed exclusively of TAP1-negative cells, indicating in vivo selection for cells with TAP deficiency. Thus, loss of TAP function allows some tumor cells to avoid T cell-dependent elimination, resulting in selection for tumor cells with deficient Ag processing.

To elucidate signal transduction pathways leading to neuronal differentiation, we have investigated a conditionally immortalized cell line from rat hippocampal neurons (H19-7) that express a temperature sensitive simian virus 40 large T antigen. Treatment of H19-7 cells with the differentiating agent basic fibroblast growth factor at 39 degrees C, the nonpermissive temperature for T function, resulted in the activation of c-Raf-1, MEK, and mitogen-activated protein (MAP) kinases (ERK1 and -2). To evaluate the role of Raf-1 in neuronal cell differentiation, we stably transfected H19-7 cells with v-raf or an oncogenic human Raf-1-estrogen receptor fusion gene (deltaRaf-1:ER). deltaRaf-1:ER transfectants in the presence of estradiol for 1 to 2 days expressed a differentiation phenotype only at the nonpermissive temperature. However, extended exposure of the deltaRaf-1:ER transfectants to estradiol or stable expression of the v-raf construct yielded cells that extended processes at the permissive as well as the nonpermissive temperature, suggesting that cells expressing the large T antigen are capable of responding to the Raf differentiation signal. deltaRaf-1:ER, MEK, and MAP kinase activities in the deltaRaf-1:ER cells were elevated constitutively for up to 36 h of estradiol treatment at the permissive temperature. At the nonpermissive temperature, MEK and ERKs were activated to a significantly lesser extent, suggesting that prolonged MAP kinase activation may not be sufficient for differentiation. To test this possibility, H19-7 cells were transfected or microinjected with constitutively activated MEK. The results indicate that prolonged activation of MEK or MAP kinases (ERK1 and -2) is not sufficient for differentiation of H19-7 neuronal cells and raise the possibility that an alternative signaling pathway is required for differentiation of H19-7 cells by Raf.

Proteins involved in class I MHC (MHC-I) Ag processing, such as the TAP, are deficient in some human tumor cells. This suggests that antitumor responses by CD8 T cells provide selection pressure to favor outgrowth of cells with defective processing of tumor Ags. Nonetheless, this evidence is only correlative, and controlled in vivo experiments have been lacking to demonstrate that TAP deficiency promotes survival of tumor cells. To explore the role of Ag processing defects in tumor progression, matched panels of TAP1-positive and TAP1-negative tumor cell lines were generated from a parental transformed murine fibroblast line. Inoculation of C57BL/6 mice with TAP1-negative cells produced large and persistent tumors. In contrast, TAP1-positive cells did not generate lasting tumors, although small tumors were detected transiently and regressed spontaneously. Both TAP1-positive and TAP1-negative cells produced tumors in athymic mice, confirming that TAP-dependent differences in tumorigenicity were due to T cell-dependent immune responses. Inoculation of C57BL/6 mice with mixtures of TAP1-positive and TAP1-negative cells produced tumors composed exclusively of TAP1-negative cells, indicating in vivo selection for cells with TAP deficiency. Thus, loss of TAP function allows some tumor cells to avoid T cell-dependent elimination, resulting in selection for tumor cells with deficient Ag processing.

Inactivation of the retinoblastoma gene product (pRb) occurs concomitant with the appearance of its hyperphosphorylated form in mid to late G1. Multiple cyclin/CDK complexes are implicated in the cell cycle phosphorylation of pRb. Using in vivo expression systems, we show that cyclins A, E, D1, D2, and D3 each function to phosphorylate and inactivate pRb. In vivo, G1 cyclin/kinase complexes enhance the phosphorylation of pRb, and these effects of cyclin/kinases on pRb can be overcome by the addition of p21, a wide spectrum inhibitor of G1 kinases. Kinases associated with cyclins A, E, and D1 phosporylate pRb indistinguishably in vivo, according to proteolytic maps. Although cyclin D1 has been reported to bind to pRb directly, requiring the pRb-binding motif LXCXE, a mutant D1 lacking the pRb-binding motif remains able to phosphorylate pRb in vivo and in vitro and is also able to reverse the growth-inhibitory properties of pRb in intact cells. Finally, coexpression of G1 cyclins and kinases represses pRb-mediated growth inhibition in Saos-2 cells. The multiplicity of mechanisms for pRb phosphorylation and inactivation suggests that several pathways exist for the regulation of pRb by phosphorylation.

The signal transduction kinase MEK (mitogen-activated protein (MAP) or extracellular signal-regulated (Erk) kinase)-1 is activated via phosphorylation by MEKK (MEK kinase) and raf kinases. We show here that these two kinases phosphorylate rat MEK-1 exclusively on two serine codons, Ser218 and Ser222. Phosphorylation of MEK-1 on serines 218 and 222 is both necessary and sufficient for MEK-1 to be activated and able to phosphorylate MAP kinase. A mutant form of MEK-1 that replaces these two codons with alanine cannot be activated, and one that substitutes glutamic acid residues in place of these 2 serines is active independent of activation by phosphorylation. These sites of activation occur in a region of MEK-1 that is similar to sites of activating phosphorylation in several other serine/threonine kinases, suggesting that this region may represent a conserved "activating domain" of many kinases. MEKK and raf display differences in site preference between these two codons, with MEKK showing preference for the amino acid at codon 218 and raf phosphorylating each residue approximately equally. This site preference might result in differences in the temporal or subsequent substrate patterns of MEK activation that result from these two activation pathways.

The product of the retinoblastoma susceptibility gene is a 105-kDa protein that has properties of a cell cycle regulatory factor. Previous reports indicated that two distinct DNA-binding factors, RBF-1 and ATF, play an important part in the transcription of the human retinoblastoma gene (Rb). Recently, we demonstrated that pRb activates expression of the human transforming growth factor-beta 2 gene through ATF-2. Since the human Rb gene promoter also contains an ATF-2-like binding site, we examined whether pRb can regulate its own expression through ATF-2. Here we report that overexpression of Rb stimulates Rb promoter activity through the ATF binding site in a variety of different cell types. Mutation of the ATF binding site of the Rb promoter abolishes the Rb autoinduction. We have also determined that the carboxyl-terminal domain of pRb is responsible for the Rb autoinduction through ATF-2. Rb autoinduction may be important for maintaining the action of pRb during cell growth, and loss of autoinductibility may contribute to retinoblastoma.

We have examined the functional consequences of mutations present in defective alleles of the retinoblastoma susceptibility gene (RB1) isolated from two spontaneously arising tumors. Unlike cDNA clones expressing the wild-type protein p110Rb, those encoding the two mutant proteins failed to induce the appearance of senescent cells in transfected Saos-2 human osteosarcoma cells. The mutant proteins were also defective in binding to the E1A oncoprotein, were unable to become hyperphosphorylated, and failed to become tightly associated with nuclear structures. We conclude that mutations in two distinct regions of the protein concomitantly affect these four aspects of p110Rb function.