Search

Expression of Pim-1, an oncogenic serine/threonine kinase, is highly regulated at the transcriptional, posttranscriptional, and posttranslational levels. Here, we report that expression of Pim-1 kinase is additionally regulated at the translational level. Pim-1 protein expression did not increase in Hut-78 lymphocytes in response to PMA1/ionomycin stimulation despite approximately 20-fold increases in mRNA levels, suggesting that translation was repressed. Sequence analysis of the 5'-untranslated region (UTR) indicated a long (400 nucleotide), 76% G + C-rich region, characteristics known to inhibit translation. Deletion of the 5'-UTR of pim-1 increased translation of the Pim-1 protein approximately 10-fold in vitro in reticulocyte lysates and approximately 1.6-fold in vivo in NIH-3T3 cells. When full-length 5'-UTR-containing pim-1 cDNA constructs were transfected into NIH-3T3 cells overexpressing eukaryotic translation initiation factor 4E (eIF-4E), approximately 6-fold higher levels of Pim-1 protein were produced, as compared to that produced in control NIH-3T3 cells. Moreover, eIF-4E overexpression had little effect in the absence of the 5'-UTR, suggesting that it relieved 5'-UTR-mediated inhibition of Pim-1 expression.

The human pim-1 proto-oncogene was expressed in Escherichia coli as a glutathione-S-transferase (GST)-fusion protein and the enzymatic properties of its kinase activity were characterized. Likewise, a Pim-1 mutant lacking intrinsic kinase activity was constructed by site-directed mutagenesis (Lys67 to Met) and expressed in E. coli. In vitro assays with the mutant Pim-1 kinase showed no contaminating kinase activity. The wild-type Pim-1 kinase-GST fusion protein showed a pH optimum of 7 to 7.5 and optimal activity was observed at either 10 mM MgCl2 or 5 mM MnCl2. Higher cation concentrations were inhibitory, as was the addition of NaCl to the assays. Previous work by this laboratory assaying several proteins and peptides showed histone H1 and the peptide Kemptide to be efficiently phosphorylated by recombinant Pim-1 kinase. Here we examine the substrate sequence specificity of Pim-1 kinase in detail. Comparison of different synthetic peptide substrates showed Pim-1 to have a strong substrate preference for the peptide Lys-Arg-Arg-Ala-Ser*-Gly-Pro with an almost sixfold higher specificity constant kcat/Km over that of the substrate Kemptide (Leu-Arg-Arg-Ala-Ser*-Leu-Gly). The presence of basic amino acid residues on the amino terminal side of the target Ser/Thr was shown to be essential for peptide substrate recognition. Furthermore, phosphopeptide analysis of calf thymus histone H1 phosphorylated in vitro by Pim-1 kinase resulted in fragments containing sequences similar to that of the preferred synthetic substrate peptide shown above. Therefore, under optimized in vitro conditions, the substrate recognition sequence for Pim-1 kinase is (Arg/Lys)3-X-Ser/Thr*-X', where X' is likely neither a basic nor a large hydrophobic residue.