Proteasome inhibitors induce apoptosis in glucocorticoid-resistant chronic lymphocytic leukemic lymphocytes.

Our previous work showed that the nuclear scaffold (NS) protease is required for apoptosis of both thymocytes and chronic lymphocytic leukemic (CLL) lymphocytes. Because partial sequencing of one of the subunits of the NS protease revealed homology to the proteasome, we tested the effects of classical proteasome inhibitors on apoptosis in CLL cells. Here we report that proteasome inhibition caused high levels of DNA fragmentation in all patients analyzed, including those resistant to glucocorticoids or nucleoside analogs, in vitro. Proteasome inhibitor-induced DNA fragmentation was associated with activation of caspase/ICE family cysteine protease(s) and was blocked by the caspase antagonist, zVADfmk. Analysis of the biochemical mechanisms involved showed that proteasome inhibition resulted in mitochondrial dysregulation leading to the release of cytochrome c and a drop in mitochondrial transmembrane potential (triangle upPsi). These changes were associated with inhibition of NFkappaB, a proteasome-regulated transcription factor that has been implicated in the suppression of apoptosis in other systems. Together, our results suggest that drugs that target the proteasome might be capable of bypassing resistance to conventional chemotherapy in CLL.

Bleomycin hydrolase (Blh1p), a multi-sited thiol protease in search of a distinct physiological role.

Bleomycin hydrolase, Blh1p, from yeast was co-purified with Gce1p, a cAMP-binding ectoprotein, anchored to the plasma membrane by a glycosyl-phosphatidylinositol (GPI) anchor. Blh1p is a hydrophilic thiol protease lacking transmembrane domains. We have used polyclonal antibodies to study the topology of the over-expressed protein in yeast and have found that it is amphitropic. Part of Blh1p is associated with plasma membranes, and most of the rest occurs in the cytosol. Both the growth conditions and calcium were found to have minor influences on the topology of Blh1p, in that glucose and the earth-alkali ion slightly enhanced recruitment to the membrane. We have examined the possibility that co-purification of Blh1p with Gce1p has a functional basis, and have observed that over-expression of BLH1 in yeast leads to an acceleration of the glucose-induced amphiphilic to hydrophilic conversion of Gce1p, wherein Blh1p could either directly catalyse the proteolytic removal of the polar head-group of the GPI anchor subsequent to an initial lipolytic cleavage by a GPI-specific phospholipase C or indirectly modulate the reaction. The data show that a thiol protease is involved, but point to an indirect role of Blh1p in GPI processing. Proteases with similar or overlapping substrate specificity are likely to exist, since deletion of BLH1 neither entails a growth-defect on any carbon source tested, nor the loss of proteolytic processing of the GPI anchor of Gce1p. Reduced proteolytic GPI processing is, however, observed in the blh1 mutant and the corresponding acceleration in the respective BLH1 multi-copy transformant.

Overexpression of DEAD box protein pMSS116 promotes ATP-dependent splicing of a yeast group II intron in vitro.

The group II intron bl1, the first intron of the mitochondrial cytochrome b gene in yeast is self-splicing in vitro. Genetic evidence suggests that trans-acting factors are required for in vivo splicing of this intron. In accordance with these findings, we present in vitro data showing that splicing of bl1 under physiological conditions depends upon the presence of proteins of a mitochondrial lysate. ATP is an essential component in this reaction. Overexpression of the nuclear-encoded DEAD box protein pMSS116 results in a marked increase in the ATP-dependent splicing activity of the extract, suggesting that pMSS116 may play an important role in splicing of bl1.

Overexpression of DEAD box protein pMSS116 promotes ATP-dependent splicing of a yeast group II intron in vitro.

The group II intron bI1, the first intron of the mitochondrial cytochrome b gene in yeast is self-splicing in vitro. Genetic evidence suggests that trans-acting factors are required for in vivo splicing of this intron. In accordance with these findings, we present in vitro data showing that splicing of bI1 under physiological conditions depends upon the presence of proteins of a mitochondrial lysate. ATP is an essential component is this reaction. Overexpression of the nuclear-encoded DEAD box protein pMSS-116 results in a marked increase in the ATP-dependent splicing activity of the extract, suggesting that pMSS116 may play an important role in splicing of bI1.

New reactions catalyzed by a group II intron ribozyme with RNA and DNA substrates.

Here we describe three novel reactions of the self-splicing group II intron bI1 (the first intron of the COB gene of yeast mitochondria) demonstrating its catalytic versatility: reversal of the first step of the self-splicing reaction catalyzed by a linear form of the intron utilizing the energy of a phosphoanhydride bond for transesterification, ligation of a single-stranded DNA to an RNA, and cleavage of a single-stranded DNA substrate. These results have the following evolutionary implications: use of the alpha-beta bond of a terminal triphosphate for transesterification suggests that an RNA RNA replicase could use mononucleotide triphosphates as precursors, and cleavage of single-stranded DNA and DNA-RNA ligation suggests that excised group II introns might integrate directly into DNA without prior reverse transcription.

Structural requirements for selection of 5'- and 3' splice sites of group II introns.

The group II intron bl1 in the gene for apocytochrome b in yeast mitochondrial DNA (COB) is self-splicing in vitro. It could recently be shown that self-splicing of this intron is fully reversible in vitro. In addition, intron integration is not restricted to parental exons, since the intron can also integrate into a foreign RNA. The position of insertion seems to be immediately 3' to a cryptic intron binding site 1 (IBS1). We confirmed and extended these results by sequencing 26 individual RNAs with transposed introns after reverse transcription and PCR amplification. Results show that intron integration into authentic exons is generally correct, but that integration into a foreign RNA is often inaccurate, i.e. insertion is one nt downstream or upstream of the 3' end of IBS1. This leads to the generation of 5' splice junctions of the new intron-harbouring 'preRNAs' with addition (or deletion) of a single A residue at the 3' end of IBS1. To investigate which structures help to define the position of 5'- and 3' cleavage, preRNAs of i) these clones with aberrant 5' splice junctions and ii) preRNAs with artificial hairpins between domains 5 and 6 of the intron were spliced under different reaction conditions. Results obtained let us conclude that i) branchpoint dependent 5' cleavage is directed by the 5' terminal G residue of the intron and, ii) the first nucleotide(s) of the 3' exon play an important role in defining the 3' splice site.