cAMP-dependent protein kinases in the rat testis: regulatory and catalytic subunit associations

Quantitative proteomic analysis of gene regulation by miR-34a and miR-34c

microRNAs (miRNAs) repress target genes by destabilizing mRNAs and/or by inhibiting translation. The best known factor for target recognition is the so called seed--a short continuous region of Watson-Crick base pairing between nucleotides 2-7 of the miRNA and complementary sequences in 3' untranslated regions of target mRNAs. The miR-34 family consists of three conserved members with important tumor suppressor functions linked to the p53 pathway. The family members share the same seed, raising the question if they also have the same targets. Here, we analyse the effect of miR-34a and miR-34c on protein synthesis by pSILAC. Despite significant overlap, we observe that the impact of both family members on protein synthesis differs. The ability to identify specific targets of a family member is complicated by the occurrence of * strand mediated repression. Transfection of miR-34 chimeras indicates that the 3'end of the miRNA might be responsible for differential regulation in case of targets without a perfect seed site. Pathway analysis of regulated proteins indicates overlapping functions related to cell cycle and the p53 pathway and preferential targeting of several anti-apoptotic proteins by miR-34a. We used luciferase assays to confirm that Vcl and Fkbp8, an important anti-apoptotic protein, are specifically repressed by miR-34a. In summary, we find that miR-34a and miR-34c down-regulate distinct subsets of targets which might mediate different cellular outcomes. Our data provides a rich resource of miR-34 targets that might be relevant for clinical trials that want to implement the miR-34 family in cancer therapy.

Cyclic AMP-dependent protein kinase isoenzymes in human myeloid leukemia (HL60) and breast tumor (MCF-7) cells

Combinations of retinoic acid (RA) and cAMP mediate many biological responses in a large variety of cell types. While the basis for the apparent synergistic effects of RA and cAMP are not clearly defined, it is likely that activation of PKA by cAMP is involved. However, literature reports concerning the identity of PKA isoforms in HL60 and MCF-7 cells are conflicting. The purpose of the present investigation is to identify PKA isoforms in HL60 and MCF-7 cells. Utilization of high-performance anion-exchange liquid chromatography, immunoblotting, and 8-azido-cAMP photoaffinity binding resulted in the finding that HL60 cells contain PKA types I alpha and II alpha, while MCF-7 cells contain PKA types I alpha, II alpha, and II beta. PKA type I alpha in both HL60 and MCF-7 cells eluted from columns as two well-separated peaks. One peak eluted at a low salt concentration in agreement with previous reports. The second HL60 PKA type I alpha peak eluted at a salt concentration intermediate between that eluting the first peak and that eluting PKA type II alpha and contained approximately 62% of the total RI alpha protein. However, the second MCF-7 PKA type I alpha peak contained approximately 66% of the total RI alpha protein and co-eluted with PKA types II alpha and II beta. This "contamination" of PKA type II fractions with PKA type I has led, in some cases, to interpretations that may need reevaluation.

Characterization of recombinant RI beta and evaluation of the presence of RI beta protein in rat brain and testicular extracts

Based upon recent reports that the mRNA from the regulatory (R) RI beta subunit of cAMP-dependent protein kinase (PKA) was expressed in testicular extracts, we determined whether testicular extracts exhibited RI beta protein. To accomplish this goal, we initially determined the fundamental labeling and ionic characteristics of recombinant RI beta. Recombinant RI beta eluted from DEAE-cellulose with a salt concentration (of 0.075 M) equivalent to its elution position from soluble mouse brain extracts with catalytic subunit-free RI alpha. As predicted by its amino acid sequence homology to RI alpha, recombinant RI beta was not phosphorylated by PKA but was labeled specifically with 8-azido-adenosine 3':5'-[32P]monophosphate (8-N3[32P]cAMP). Additionally, RI antisera reacted equally with RI alpha (47 kDa) and recombinant RI beta (53 kDa). However, recombinant RI beta exhibited an unexpectedly basic pI of 6.65-6.85. By using a pH gradient for isoelectric focussing that allowed for clear focussing of 8-N3[32P]cAMP-labeled recombinant RI beta, 8-N3[32P]cAMP-labeled RI beta was readily detected by two-dimensional gel electrophoresis in rat brain particulate extracts and exhibited a pI equivalent to that of recombinant RI beta. The 53-kDa RI beta was undetectable either by its immunoreactivity or upon photoaffinity labeling with 8-N3[32P]cAMP by one or two-dimensional gel electrophoresis in soluble or particulate extracts of testes of 14-day-old, 45-day-old, or adult rats or in epididymal sperm. However, 8-N3[32P]cAMP-labeled RI beta was detected, albeit in very small levels, by two-dimensional electrophoresis upon separation of PKAs in testes of 14-day-old rats by DEAE-cellulose chromatography but was absent in equivalent extracts from adult rat testes. These results demonstrate that the unexpectedly basic pI of RI beta allows for its clear separation by two-dimensional electrophoresis from the RII proteins and therefore allows for its unambiguous identification. Further studies, however, are required to resolve the basis for the apparent disparity in testis RI beta mRNA and protein.