Tpl-2 induces apoptosis by promoting the assembly of protein complexes that contain caspase-9, the adapter protein Tvl-1, and procaspase-3.

The Tpl-2 proto-oncoprotein promotes cellular proliferation when overexpressed in a variety of tumor cell lines. Here, we present evidence that when overexpressed in immortalized non-transformed cells, Tpl-2 induces apoptosis by promoting the activation of caspase-3 via a caspase-9-dependent mechanism, and that apoptosis is enhanced when Tpl-2 is co-expressed with the newly identified ankyrin repeat protein Tvl-1. The activation of caspase-3 by caspase-9 is known to depend on the assembly of a multimolecular complex that includes Apaf-1 and caspase-9. Data presented here show that co-expression of Tpl-2 with Tvl-1 promotes the assembly of a complex that involves several proteins that bind Apaf-1 including Tvl-1, itself, Tpl-2 and phosphorylated procaspase-9. More important, procaspase-3, which under normal growth conditions is not associated with the complex, binds Tvl-1 conditionally in response to Tpl-2-generated apoptotic signals. The conditional association of procaspase-3 with Tvl-1 promotes the in vivo proteolytic maturation of procaspase-3 by caspase-9, a process casually linked to apoptosis.

Isolation of chromatin fragments released by apoptotic cells.

A two-step procedure for isolation of chromatin fragments released from apoptotic cells is described. The first step includes extraction of the chromatin fragments by a hypotonic treatment of isolated nuclei. The second step provides stabilization of the extracted chromatin fragments and their collection by a high-speed centrifugation. The procedure ensures not only a complete extraction and collection of the chromatin fragments but also intactness of their components. A number of control experiments confirmed this statement: they showed that the percentage of recovered fragmented DNA does not exceed 10% from the total DNA of apoptotic cells and that it is specific in both nature and chromosomal localization; the experiments reveal, moreover, that the protein components of the fragments including the well known histone and non-histone species possess also ability to support in in vitro conditions their specific phosphorylation. Developed for recovery of chromatin fragments from mouse erythroleukemia cells, spontaneously entering apoptosis, the procedure is practically applicable for all eukaryotic cell systems sharing programmable death.

Spontaneous apoptosis in mouse F4N-S erythroleukemia cells induces a nonrandom fragmentation of DNA.

This study characterizes the fragmented DNA of mouse erythroleukemia (MEL) cells spontaneously entering apoptosis. Fragmented DNA was isolated by introducing a novel procedure that ensured a complete extraction of the characteristic oligonucleosomal ladder. As the results show, less than 10% of DNA of apoptotic cells is fragmented in this form. The main conclusion from experiments to characterize the nature of fragmented DNA is that spontaneous apoptosis induces a nonrandom cleavage of genomic DNA. The Southern analysis performed with total apoptotic DNA revealed that it is strongly enriched in interspersed repetitive sequences. In situ hybridizations with such DNA showed further than in interphase nuclei these sequences flock together and form clusters spread throughout the whole nuclear area whereas in mitotic chromosomes they are located predominantly at their pericentromeric/peritelomeric ends. Partial cloning and sequencing reinforces the notion that the apoptotic DNA is representative for a heterochromatinic portion of the mouse genome. Support for such an unexpected conclusion is coming also from experiments indicating that this DNA is heavily methylated and poorly transcribed.

Spatial and structural segregation of the transcribed and nontranscribed alleles of c-myc in Namalva-S cells.

By using various approaches we received evidence that, in Namalva-S cells carrying a t(8;14) translocation and highly expressing c-myc, the two alleles of the gene are spatially and structurally segregated. Spatial segregation of the alleles was observed in all nuclei analyzed by in situ hybridization technique. Their structural segregation, i.e., association with different intranuclear structures, was confirmed in a number of experiments. When high-salt extracted nuclei were digested with EcoRI, which is known to produce fragments containing the entire c-myc locus, the sequences of the gene were found separated between the pellet, containing sequences firmly associated with the heavier matrix structures, and the supernatant, containing sequences from the free length of the DNA loops. Southern hybridization performed with a probe representative for the constant region of the human IgH locus revealed that this fractionation in fact segregates the reorganized from the normal allele of c-myc. Run-on experiments carried out with two fractions, topologically equivalent to the above P and S but isolated as intact chromatin structures, indicated that the allele associated with nuclear matrix is actively transcribed, while that located in the free length of the chromatin loops is practically nontranscribed. Studies on the chromatin organization of transcribed and nontranscribed alleles revealed the existence in them of two alternative chromatin structures. Control experiments with beta-globin gene, performed with cells constitutively nontranscribing or actively transcribing this gene, confirmed our conclusions about the spatial segregation of the two alleles and clarified that their structural segregation occurs when the gene is activated for transcription.