Expression of unphosphorylated form of human double-stranded RNA-activated protein kinase in Escherichia coli.

Interferon (IFN)-inducible, double-stranded (dsRNA)-activated protein kinase (PKR) is a key mediator of the antiviral and antiproliferative effects of IFN. PKR is present within cells in a latent state. In response to binding dsRNA, the enzyme becomes activated, causing autophosphorylation and an increase in specific kinase activity. In order to study PKR and its inhibitors, a large amount of the enzyme in its latent, unphosphorylated state is required. When PKR is fused to glutathione S-transferase (GST-PKR) and the fusion protein is expressed in Escherichia coli, the PKR obtained is fully activated by autophosphorylation. Therefore, we have developed an expression plasmid in which both GST-PKR and bacteriophage lambda protein phosphatase (lambda-PPase) genes were placed downstream of a T7 promoter. After induction of expression, unphosphorylated GST-PKR was obtained in good yield, and purified to near homogeneity. The purified enzyme has dsRNA-dependent activation and phosphorylates the translation initiation factor eIF2 alpha. Using the recombinant protein, we analyzed the inhibition mechanisms of two viral inhibitors, vaccinia virus K3L protein and adenovirus virus-associated RNA I (VAI RNA). K3L inhibited both autophosphorylation of PKR and phosphorylation of eIF2 alpha, whereas VAI RNA inhibited only autophosphorylation. The separation of autophosphorylation and catalytic activity shows that the recombinant PKR is useful in analyzing the functions of PKR, its inhibitors, and its regulatory molecules. The coexpression system of protein kinase with lambda-PPase described here will be applicable to obtaining unphosphorylated and unactivated forms of other protein kinases.

cDNA cloning and chromosomal mapping of genes encoding novel protein kinases termed PKU-alpha and PKU-beta, which have nuclear localization signal.

We have cloned cDNAs for novel serine/threonine protein kinases (PK), termed PKU-alpha and PKU-beta, by screening a bacteriophage expression library for kinase activity. Sequence analysis of PKU-alpha and PKU-beta genes revealed that their open reading frames (ORF) were 2151 and 2361 nucleotides (nt) encoding polypeptides of 717 and 787 amino acid (aa) residues, respectively. The deduced aa sequences of PKU-alpha and PKU-beta contained typical serine/threonine PK domains at the C-terminal region and were 86% identical to each other, indicating that they belong to the same PK family. Northern analysis reveals that they are expressed in nearly all human tissues and in cultured cells. The genes for PKU-alpha and PKU-beta were mapped to chromosome 17q23 and 8p12-p22, respectively, by fluorescence in situ hybridization. The proteins encoded by both cDNAs contain a putative nuclear localization signal (NLS) in their N-terminal region. These signals are likely to function in nuclear localization. Glutathione S-transferase (GST)-fusions to regions of PKU-alpha and beta containing the NLS were efficiently localized to the nucleus. In addition, PKU-beta transiently expressed in COS-1 cells was predominantly nuclear. PKU-alpha and PKU-beta differ: a consensus sequence for a nt binding motif is present near the NLS of PKU-beta. These results suggest that PKU-alpha and beta may phosphorylate serine and/or threonine residues on similar proteins, but their activities are regulated through distinct interactions with a nuclear component.

Genetic organization and diversity of the 3' noncoding region of the hepatitis C virus genome.

The 3' noncoding region (3' NCR) of the hepatitis C virus (HCV) genome contained in viral particles was analyzed by an RNA linker ligation followed by reverse transcription-polymerase chain reaction. Sequence analysis of the amplified fragment from four strains, including different genotypes 1b, 2b, 3a, and 3b indicated that the 3' NCR is composed of between 200 and 235 nts. The sequence of the 3' NCR consists of a type-specific region (immediately following the termination codon), a poly(U) stretch, a C(U)n-repeat, and highly conserved region termed the core element. The poly(U) stretch and C(U)n-repeat regions varied in length and in sequence among different genotypes. Core elements having putative secondary structure consisted of 98 or 100 nts and were highly conserved in all genotypes. Most of the nt changes found in different genotypes did not affect the secondary structure of the core elements, suggesting that this region may play an important role in replication, stabilization of the HCV RNA, and/or packaging of the genome. Most of the HCV-1b strains carried two U residues at the 3' end of the core element, while the minor HCV-1b strains had no U residues, demonstrating that there are two variants in type 1b strains. Amplification of the core element using linker-primed cDNA was comparable with that using the 3' proximal core element-primed cDNA, indicating that the 3' end of HCV genome was terminated by an OH group.

Full-length sequence of the genome of hepatitis C virus type 3a: comparative study with different genotypes.

Hepatitis C virus (HCV) type K3a (type 3a), which represents a minor genotype in Europe, the U.S.A. and Asia, appears to be significantly distributed throughout Australia and Brazil. We amplified the HCV-K3a/650 genome by reverse transcription polymerase chain reaction in ten overlapping fragments and determined the nucleotide sequences. The total sequence was 9454 bases in length and contained an open reading frame of 3021 amino acids, which is 10 or 11 amino acids longer than in HCV type 1 and 12 amino acids shorter than the sequence of type 2. These differences were due to the different lengths of both the putative envelope protein E2 and the NS5A regions, whose nucleotide lengths differ between types 1 and 2 also. Phylogenetic analysis of the putative core region and a portion of NS5B encoding the Gly-Asp-Asp motif indicated that HCV-K3a closely matched the corresponding type 3a group. The deletion and addition of amino acids in both E2 and NS5A may be associated with their pathobiological features.

Rat guanidinoacetate methyltransferase: mutation of amino acids within a common sequence motif of mammalian methyltransferase does not affect catalytic activity but alters proteolytic susceptibility.

1. Manual alignment of amino acid sequences of mammalian S-adenosylmethionine-dependent methyltransferases of known sequence revealed the presence of 2 homologous regions. 2. The sequence of the region at the C-terminal side is unique to mammalian methyltransferases, and in guanidinoacetate methyltransferase this sequence occurs at residues 159-165. 3. Mutagenesis of 5 conserved residues in this sequence did not affect the catalytic activity but altered tryptic susceptibility at Arg20.

Two regions in human DNA polymerase beta mRNA suppress translation in Escherichia coli.

Although human DNA polymerase beta (DNA pol beta) shows 96% identity with rat DNA pol beta at the amino acid level, it is weakly expressed in Escherichia (E.) coli relative to the rat enzyme. The mechanism of this suppression was investigated. Pulse-chase protein labeling and steady state mRNA analysis showed that mature human DNA pol beta protein is relatively stable in E. coli and the levels of human and rat DNA pol beta mRNA were comparable indicating that the human DNA pol beta expression is suppressed at the translational level. By systematic expression analysis of a number of chimeric genes composed of human and rat cDNAs, two strong translational suppression regions were mapped in the human DNA pol beta mRNA; one was named TSR-1, corresponding to CGG encoding arginine (arg) at position 4 and the other, termed TSR-2, is located between codons 153 and 199. Since substitution of the rat Arg-4 codon with synonymous codons showed strong effects upon the expression level, we propose that the arg codon at the N-terminal coding region plays a role in modulating expression.

Aspartic acid residues at positions 190 and 192 of rat DNA polymerase beta are involved in primer binding.

The sequence Gly-Asp-Met-Asp, spanning positions 189-192 of rat DNA polymerase beta, is similar to the sequence motif Gly-Asp-Thr-Asp that is highly conserved in a number of replicative DNA polymerases from eukaryotic cells, viruses, and phages. The role of this sequence in the catalytic function of rat DNA polymerase beta was investigated by individually changing each amino acid in this region by site-directed mutagenesis. The mutant enzymes DE190 and DE192, in which aspartic acid residues at positions 190 and 192, respectively, were replaced by glutamic acid, showed about 0.1% activity of the wild-type enzyme. On the other hand, the replacement of Gly-189 by alanine or Met-191 by isoleucine or threonine only slightly affected the enzyme activity. A gel mobility shift assay showed that DNA complexes with enzyme DE190 and especially with DE192 were less stable than the corresponding complex with the wild-type enzyme. Kinetic analysis with these mutant enzymes indicate that their Km's for primer DNA were about 10-fold higher than that of the wild type, while Km's for deoxyribonucleoside triphosphate were not changed. Since neither DE190 nor DE192 had any significant alteration in secondary structure, our results suggest that both Asp-190 and Asp-192 are located in the active site and are involved in the interaction of DNA polymerase beta with primer.

Site-directed mutagenesis of recombinant rat DNA polymerase beta: involvement of arginine-183 in primer recognition.

By site-directed mutagenesis using synthetic oligonucleotides, amino acid residues 181Phe-Arg-Arg183 of recombinant rat DNA polymerase beta were replaced by other amino acids to clarify the roles of these residues in the DNA synthesizing reaction. Replacement of Phe-181 by alanine reduced the enzyme activity only 30%. Replacement of Arg-182 by alanine and glutamine resulted in reduction of the activity by about 67% and 95%, respectively. The Arg-182----Gln replacement increased the binding strength to single-stranded DNA but did not significantly change the Km's for the primer and dTTP, suggesting that Arg-182 is involved in modulation of binding to the template rather than to the primer or deoxyribonucleoside triphosphate. Replacement of Arg-183 by Gln resulted in reduction of the activity by about 95%, and this change, although causing little change in binding strength to single-stranded DNA, resulted in a 3-4-fold increase in the Km's for the primer and deoxyribonucleoside triphosphate. A more dramatic change was observed when Arg-183 was replaced by Ala, which resulted in a 99.98% reduction of enzyme activity. Although the Km for deoxyribonucleoside triphosphate of this mutant enzyme was hardly changed, that for the primer increased 159-fold. Therefore, it is concluded that Arg-183 occupies an important part of the primer recognition site of DNA polymerase beta.

Construction of Escherichia coli vectors for expression and mutagenesis: synthesis of human c-Myc protein that is initiated at a non-AUG codon in exon 1.

Three types of Escherichia coli vector for both gene expression and mutagenesis were constructed from a plasmid/phage chimera vector pUC118. Each vector contains the lac (pTD-lac), tac (pTD-tac), or T7 promoter (pTD-T7). Downstream from the promoter, these vectors have sequences in common, including a Shine-Dalgarno (SD), multiple cloning sequence, sequence-primer binding site, transcription termination signal, and M13 origin of replication. Using single-stranded circular DNA obtained by infection with helper phage, oligodeoxyribonucleotide (oligo)-directed mutagenesis allows the appropriate fusion between the vector SD sequence and the start codon in the inserted fragment. Since a complementary oligo representing a large deletion is generally used for this construction, the extra nucleotides in the opposing strand form a loop structure. Thus, we have designated this mutagenesis as 'loop-out mutagenesis'. Expression plasmid encoding the larger human c-Myc protein that is initiated at a non-AUG codon in exon 1 and its derivatives were constructed using a pTD-T7 vector. Expression experiments indicated that the wild-type (wt) protein was synthesized poorly after induction with isopropyl-beta-D-thiogalactopyranoside, while one of the derivatives, p62M1T, in which a threonine residue was added at the N terminus of the wt protein, was produced in a large quantity in E. coli.

Expression of active rat DNA polymerase beta in Escherichia coli.

A recombinant plasmid for expression of rat DNA polymerase beta was constructed in a plasmid/phage chimeric vector, pUC118, by an oligonucleotide-directed mutagenesis technique. The insert contained a 1005 bp coding sequence for the whole rat DNA polymerase beta. The recombinant plasmid was designed to use the regulatory sequence of Escherichia coli lac operon and the initiation ATG codon for beta-galactosidase as those for DNA polymerase beta. The recombinant clone, JMp beta 5, obtained by transfection of E. coli JM109 with the plasmid, produced high levels of DNA polymerase activity and a 40-kDa polypeptide that were not detected in JM109 cell extract. Inducing this recombinant E. coli with isopropyl beta-thiogalactopyranoside (IPTG) yielded amounts of 40-kDa polypeptide as high as 19.3% of total protein. Another recombinant clone, JMp beta 2-1, which was constructed by an oligonucleotide-directed mutagenesis to use the second ATG codon for the initiation codon, thus deleting the first 17 amino acid residues from the amino terminus, produced neither high DNA polymerase activity nor the 40-kDa polypeptide. The evidence suggests that this amino-terminal structure is important for stability of this enzyme in E. coli. The DNA polymerase was purified to homogeneity from the IPTG-induced JMp beta 5 cells by fewer steps than the procedure for purification of DNA polymerase beta from animal cells. The properties of this enzyme in activity, chromatographic behavior, size, antigenicity, and also lack of associated nuclease activity were indistinguishable from those of DNA polymerase beta purified from rat cells, indicating the identity of the overproduced DNA polymerase in the JMp beta 5 and the rat DNA polymerase beta.