The caspase-9 derived C-terminal fragment of cytokeratin 18 modulates topoisomerase action.

During early apoptosis the 33 amino acid C-terminal cytokeratin 18 (CK18) fragment is released by caspase-9 cleavage at the 393DALD/S site. This basic peptide relocates from the cytoskeleton to the nucleoplasm as shown by confocal laser scanning. It is shown that the C-terminal peptide modulates topoisomerase activity as measured by relaxation of plasmid DNA. In an in vitro assay recombinant caspase-induced chromatin condensation is inhibited by the peptide and at the electron microscopical level a clear inhibition of nucleolar breakdown was observed in its presence. We hypothesize that the C-terminal CK18 fragment exerts an effect in the nucleolus by stimulating rRNA transcription and processing via modulation of enzymatic activity of topoisomerase I. This leads to preservation of general transcriptional activity required to exert active steps during early stages of programmed cell death.

Proteasomes act in the pre-mitochondrial signal transduction route towards roscovitine-induced apoptosis.

The role of the ubiquitin-proteasome pathway during roscovitine induced apoptosis was evaluated in the non-small cell lung carcinoma cell line MR65. To this end specific inhibitors of proteasome activity, MG132 and lactacystin were used. Addition of MG132 or lactacystin, 1 h prior to the addition of the CDK-inhibitor roscovitine to the cell cultures inhibited apoptosis significantly, as measured by PS exposure, cytokeratin 18 cleavage and caspase-3 activation. Furthermore, we show that inhibition of proteasome activation prior to induction of apoptosis by roscovitine prevents loss of mitochondrial inner transmembrane potential (DeltaPsim). In addition we found that MG132 and lactacystin prevent release of cytochrome c from the mitochondrion. In contrast to the above findings we see no effect of proteasome inhibition in Fas-mediated apoptosis. Taken together our data suggest a specific role for proteasomes very early in roscovitine-induced apoptosis, upstream from the caspase cascade and mitochondrion.

Transient expression of phosphatidylserine at cell-cell contact areas is required for myotube formation.

Cell surface exposure of phosphatidylserine (PS) is shown to be part of normal physiology of skeletal muscle development and to mediate myotube formation. A transient exposure of PS was observed on mouse embryonic myotubes at E13, at a stage of development when primary myotubes are formed. The study of this process in cell cultures of differentiating C2C12 and H9C2 myoblasts also reveals a transient expression of PS at the cell surface. This exposure of PS locates mainly at cell-cell contact areas and takes place at a stage when the structural organization of the sarcomeric protein titin is initiated, prior to actual fusion of individual myoblast into multinucleated myotubes. Myotube formation in vitro can be inhibited by the PS binding protein annexin V, in contrast to its mutant M1234, which lacks the ability to bind to PS. Although apoptotic myoblasts also expose PS, differentiating muscle cells show neither loss of mitochondrial membrane potential nor detectable levels of active caspase-3 protein. Moreover, myotube formation and exposure of PS cannot be blocked by the caspase inhibitor zVAD(OMe)-fmk. Our findings indicate that different mechanisms regulate PS exposure during apoptosis and muscle cell differentiation, and that surface exposed PS plays a crucial role in the process of myotube formation.

The effect of the cyclin-dependent kinase inhibitor olomoucine on cell cycle kinetics.

The effect of the cyclin-dependent (CDK) inhibitors olomoucine and roscovitine on cell kinetics was studied. To this end, nonsmall cell lung cancer (NSCLC) cell line MR65 and neuroblastoma cell line CHP-212 were pulse labeled with bromodeoxyuridine (BrdUrd) and chased in culture medium, to which various concentrations of olomoucine or roscovitine were added. A dose-dependent inhibition of the G1/S-phase and G2/ M-/G1 transitions was observed. Furthermore, S-phase progression was also inhibited in a dose-dependent manner. Similarly, roscovitine, another CDK inhibitor with a 10-fold higher efficiency for both CDK1 and CDK2 as compared to olomoucine, showed the same effects at a 10-fold lower concentration. At the highest tested doses both olomoucine (200 microM) and roscovitine (40 microM) induced a complete cell cycle block in both cell lines, paralleled by the appearance of apoptotic figures. In these cultures a decrease in CDK1 protein level was found as shown by Western blotting. Bivariate CDK1/DNA analysis confirmed these observations and showed that a subpopulation of cells with characteristics of apoptosis became CDK1 negative. The presented data suggest that cyclins and CDKs are involved at an important nodal point shared by pathways regulating cellular proliferation and apoptosis.

Detailed analysis of cell cycle kinetics upon proteasome inhibition.

We have studied specific effects of proteasome inhibition on cell cycle progression. To this end, the protease inhibitors MG115, calpain inhibitor I, and calpain inhibitor II, which display differential inhibitory effects on proteasomes, were used. Cell kinetic studies using bromodeoxyuridine pulse labeling revealed a complete block of G1/S and metaphase transitions and a delayed progression through S phase in cell cultures treated with 54 microM of MG115. Calpain inhibitor I in similar concentrations displayed a fivefold lower effect on cell cycle kinetics. Calpain inhibitor II and MG2M, which is a structural analogue of MG115, had no effect on the cell cycle. The inhibitory effect of MG115 treatment was reversible, because the cell cycle was immediately resumed when the MG115-containing culture medium was replaced by fresh culture medium. Because ubiquitinated proteins accumulated after MG115 treatment, it was confirmed that ubiquitin-dependent protein degradation, and thus proteasomal activity were blocked. By comparison of biochemical and in vitro proteasome inhibition experiments, it was hypothesized that chymotrypsin-like activity of proteasomes may play an important role in cell cycle kinetics.

Subcellular localization of proteasomes in apoptotic lung tumor cells and persistence as compared to intermediate filaments.

We have studied the subcellular localization and expression levels of proteasomes during apoptosis in a lung cancer cell line. Apoptosis was induced by exposing the cells to 200 microM olomoucine, a specific cyclin-dependent kinase inhibitor. The morphological changes characteristic for apoptotic cells were visible: the cells reduced in size, the chromatin condensed and the membranes became convoluted. As the process continued, the nuclei became fragmented, and the cells broke up into cytoplasmic vesicles and apoptotic bodies. Immunocytochemically, apoptotic cells were detected by the ability to bind annexin V at their surface. During the initial stages of apoptosis, proteasomes were present in the nucleus as well as in the cytoplasm. Upon increased chromatin condensation, nuclear proteasomes were found predominantly surrounding the chromatin, while the chromatin itself remained devoid of staining. That the proteasomes persisted relatively long in the apoptotic cells was shown by immunoblotting of non-denaturing gels, which indicated that both 20S and 26S proteasomes were present in apoptotic cells. In immunofluoresence microscopy the proteasome fluorescence intensity of apoptotic cells seemed higher than that of non-apoptotic cells. These differences in intensity were even more pronounced after Triton X-100 extraction. Flow cytometry revealed that the absolute levels of proteasome staining in cells were decreased after Triton X-100 extraction. However, no differences in staining levels were detected between apoptotic and non-apoptotic cells. A relative increase of proteasome concentration through cell shrinkage or a concentration in certain cell compartments may be the origin of the apparently increased signal that was seen in immunofluorescence microscopy. Furthermore, proteasomes were clearly detectable in the apoptotic bodies and cytoplasmic vesicles at the time immunocytochemical reactivity for cytokeratins and lamins had diminished to a large extent. Immunoblotting of denaturing polyacrylamide gels confirmed the results obtained by flow cytometry. The proteasome content was retained only partially in the cells after Triton X-100 extraction, while the intermediate filaments were not detectable anymore in the apoptotic cells.

Changes in immunocytochemical detectability of proteasome epitopes depending on cell growth and fixation conditions of lung cancer cell lines.

The localization of proteasome epitopes in the lung cancer cell lines NCI-H82, derived from a small cell lung cancer, and MR65, derived from a squamous cell lung carcinoma, was studied in relation to cell growth conditions. For this purpose the proteasome monoclonal antibodies MCP34 and MCP20 were applied to the cells growing under different nutritional conditions, resulting in different proliferative states. Using indirect immunofluorescence microscopy with brief fixation in methanol (5 sec, -20 degrees C) followed by three dips in acetone (5 sec at room temperature), it became obvious that the intracellular detectability of the proteasomes changes depending on the nutritional and proliferative status of the tumor cells. Two types of experiments were carried out: (1) cells were grown for two days at different cell densities, with an excess of culture medium, and (2) cells were seeded in a low cell density and monitored for 6 days without change of medium. In cells grown at low density, the proteasomes can be detected mainly in the nuclei, while the nucleoli are almost devoid of staining, and the cytoplasm is only slightly stained. In cells grown at high density, the staining pattern changes with a much less pronounced nuclear staining than in the cells at low density, while the cytoplasm remains slightly stained. In the nutrient depletion experiment similar changes were seen. In cells growing under favorable conditions (1 or 2 days in fresh medium) proteasomes are detected mainly in the nuclei, whereas when the medium becomes depleted of nutrients (4 or 5-day-old medium) the staining pattern changes to one with a much less pronounced nuclear staining. However, in immunofluorescence studies on cells grown under similar conditions but fixed in ethanol (-20 degrees C) for 15 min, the changes in proteasome localization pattern were not detected during medium depletion. Using this fixation protocol the proteasomes are detected mainly in the nuclei at all stages of the medium exhaustion experiment. These apparently contrasting results suggest that upon nutrient depletion the proteasome epitopes become less accessible to the antibodies used. Apparently, the epitopes can regain accessibility if an extended ethanol fixation is used. This hypothesis was confirmed by flow cytometry and immunoblotting experiments. In flow cytometry of ethanol-fixed cells the fluorescence intensity of only a minor part of the cell population decreases to some extent with medium depletion, but in the majority of the cells fluorescence remains at its initial level. The immunoblotting experiments show no quantitative changes in proteasome content of the tumor cells at the different growth conditions.