Alternative splicing and alternative initiation of translation explain the four forms of the Ia antigen-associated invariant chain.

The Ia antigen-associated invariant chain (In) exists in humans as four related polypeptides, p33, p35, p41 and p43, all associated with HLA-class II antigens. As described previously, two of these forms of In chain, p33 and p35, result from the use of two in-phase initiation AUG codons on the unique In p33 mRNA. In addition to cDNA clones derived from In p33 mRNA, we have isolated a new cDNA clone, called p41-1, which differs from p33-1 by an additional segment in the coding region. The DNA sequence encoding the segment unique to p41-1 was identified in the genomic sequence in the intron between exon 6 and 7, and we refer to it as exon 6b. Cells transfected with a full length p41 cDNA clone in an expression vector synthesize the two larger forms of the In chain, p41 and p43. We propose that the larger mRNA, encoding p41, results from alternative splicing of exon 6b, and that p41 and p43 result from the use of the two functional initiation AUG codons identified in p33 mRNA. Alternative splicing, together with alternative initiation of translation, allows therefore the synthesis of four related In chain polypeptides from a single gene.     

The upstream open reading frame mediates constitutive effects on translation of cytochrome p-450c27 from the seventh in-frame AUG codon in rat liver

The 2.3-kb mRNA that codes for cytochrome P-450c27 (CYP27) has an unexpectedly long 5'-untranslated region (UTR) that holds six AUGs, leading to several upstream open reading frames (uORFs). The initiation of translation from the seventh AUG forms a putative 55-kDa precursor, which is processed in mitochondria to form a 52-kDa mature protein. The first three AUGs form fully overlapping uORF1, uORF2, and uORF3 that are in-frame with the seventh AUG and next two form fully overlapping uORF4 and uORF5 that are out-of-frame with the seventh AUG. Although not recognized by the scanning ribosomes under normal conditions, the sixth in-frame AUG forms a putative 57-kDa extension of the main open reading frame. The purpose of this study was to identify the elements in the 5'-UTR that direct CYP27 mRNA translation exclusively from the seventh AUG. Expression of 5' deletion mutants in COS cells reveal that the intact 5'-UTR not only directs the initiation of translation from the seventh AUG but also acts as a negative regulator. A 2-kb deletion mutant that lacks uORF1 initiates translation equally from the sixth and the seventh AUGs, forming both 57- and 55-kDa precursor proteins with a 2-fold increase in rate of translation. However, induction in translation does not affect the levels of the mature 52-kDa form in mitochondria but causes accumulation of the precursor form in cytosol not seen in COS cells transfected with wild-type cDNA. Mutation of the stop codon that terminates uORF1 completely shifts the initiation of translation from the seventh to the first AUG, forming a 67-kDa precursor that is processed into a 52-kDa mature protein in mitochondria. Confirmation of the bicistronic nature of CYP27 mRNA by epitope mapping of uORF1 suggests that translation of CYP27 mRNA from the seventh AUG is directed and regulated by uORF1 expression.     

A single species of pX mRNA of human T-cell leukemia virus type I encodes trans-activator p40x and two other phosphoproteins.

Human T-cell leukemia virus type I (HTLV-I) contains the pX sequence which codes for the trans-activator of the long terminal repeat (LTR) and is thus postulated to be associated with leukemogenesis in adult T-cell leukemia. Overlapping open reading frames (ORF) in the pX sequence were recently found to code for p27x-III and p21x-III by ORF III, in addition to p40x coded for by ORF IV. The mechanism of expression of these newly identified proteins and their possible association with trans-activation were studied. On transfection of an expression plasmid that contains a cDNA sequence of the pX mRNA, products from both ORFs III and IV were detected in the cells. The RNA was synthesized in vitro from the cDNA clone by SP6 RNA polymerase and translated in a rabbit reticulocyte lysate. As translation products, two proteins, p27x-III and p21x-III, were detected in addition to p40x. Elimination of the first and second ATG codons in ORF III resulted in loss of the ability to code for p27x-III and p21x-III, respectively, which indicated that the translations from these two ATG codons were independent. A mutant that lacked both ATG codons was fully active in trans-activation of chloramphenicol acetyltransferase gene expression directed by the LTR. These results indicate that a 2.1-kilobase pX mRNA of HTLV-I independently encodes three proteins, p40x, p27x-III, and p21x-III, by different ORFs and that the last two proteins are not involved in trans-activation of the unintegrated LTR.     

Generation of multiple isoforms of eukaryotic translation initiation factor 4GI by use of alternate translation initiation codons

Eukaryotic translation initiation factor 4GI (eIF4GI) is an essential protein that is the target for translational regulation in many cellular processes and viral systems. It has been shown to function in both cap-dependent and cap-independent translation initiation by recruiting the 40S ribosomal subunit to the mRNA cap structure or internal ribosome entry site (IRES) element, respectively. Interestingly eIF4GI mRNA itself has been reported to contain an IRES element in its 5' end that facilitates eIF4GI protein synthesis via a cap-independent mechanism. In HeLa cells, eIF4GI exists as several isoforms that differ in their migration in sodium dodecyl sulfate (SDS) gels; however, the nature of these isoforms was unclear. Here, we report a new cDNA clone for eIF4GI that extends the 5' sequence 340 nucleotides beyond the previously published sequence. The new extended sequence of eIF4GI is located on chromosome 3, within two additional exons immediately upstream of the previously published eIF4GI sequence. When mRNA transcribed from this cDNA clone was translated in vitro, five eIF4GI polypeptides were generated that comigrated in SDS-polyacrylamide gels with the five isoforms of native eIF4GI. Furthermore, translation of eIF4GI-enhanced green fluorescent protein fusion constructs in vitro or in vivo generated five isoforms of fusion polypeptides, suggesting that multiple isoforms of eIF4GI are generated by alternative translation initiation in vitro and in vivo. Mutation of two of the five in-frame AUG residues in the eIF4GI cDNA sequence resulted in loss of corresponding polypeptides after translation in vitro, confirming alternate use of AUGs as the source of the multiple polypeptides. The 5' untranslated region of eIF4GI mRNA also contains an out-of-frame open reading frame (ORF) that may down-regulate expression of eIF4GI. Further, data are presented to suggest that a proposed IRES embedded in the eIF4GI ORF is able to catalyze synthesis of multiple eIF4GI isoforms as well. Our data suggest that expression of the eIF4GI isoforms is partly controlled by a complex translation strategy involving both cap-dependent and cap-independent mechanisms.     

Human p53 cellular tumor antigen: cDNA sequence and expression in COS cells.

A 2.5-kb cDNA clone for human p53 tumor antigen has been isolated. This clone contains the entire coding region including 135 bp upstream of the first ATG. Comparison of the nucleotide sequence of human p53 and mouse p53 demonstrates that the first ATG in human p53 corresponds to the second ATG (codon No. 4) in mouse p53. The human p53 comprises 393 residues and is longer than the mouse p53 due to six additional codons present at the region corresponding to exon 4 of the mouse p53 gene. The DNA sequence homology between the coding regions of mouse and human p53 is 81% and the conservation of homology is not equally distributed along the molecule. When inserted into SV40-based expression vectors the human p53 cDNA successfully directs the production of a polypeptide with an apparent mol. wt. of 55 kd which can be precipitated by monoclonal antibodies to p53.     

Characterization of two mRNA species encoding human estradiol 17 beta-dehydrogenase and assignment of the gene to chromosome 17

Using two 33-mer synthetic oligonucleotides derived from the amino acid sequence of the catalytic site of estradiol 17 beta-dehydrogenase (E2DH) and polyclonal antibodies raised against the enzyme purified from human placenta, clones were isolated from a lambda gt11 human placental cDNA library. A 327-amino acid sequence was deduced from cDNA sequencing. Two mRNA species have been identified in poly(A)+ RNA from human placenta, a major species migrating at 1.3 kb while a minor one is found at approx. 2.2 kb. Primer extension and S1 nuclease analysis indicate that the major mRNA species starts 9-10 nucleotides while the minor mRNA starts 971 nucleotides upstream from the ATG initiating codon, respectively. Sequence analysis of the longest cDNA clone (2092 bp) shows that it possesses identical coding and non-coding sequences in the regions of overlap with the shorter cDNA clones. The 32P-labeled 5' non-coding fragment hybridizes only to the 2.2 kb band, thus providing evidence for the existence of two distinct mRNA species which differ in their 5' noncoding regions. Using hp E2DH-36 cDNA as a probe for in situ hybridization of translocated chromosomes, the human E2DH gene was localized to the q11-q12 region of chromosome 17.     

Characterization of human UMP-CMP kinase enzymatic activity and 5' untranslated region.

We have cloned a cDNA for human UMP-CMP kinase from a macrophage cDNA library. Sequence analysis showed that this cDNA is derived from the same gene as a previously reported EST-derived cDNA. Here we show that a conspicuous difference between these two clones, 73 additional 5' nucleotides in the EST clone, including a putative translational start site, is not functionally significant. This work shows that the additional 5'sequence in the EST clone was unnecessary for enzymatic activity and nonfunctional in the initiation of translation. Specifically, we found that protein expressed by both the macrophage-derived cDNA and the extended cDNA had the same relative molecular mass, consistent with use of an ATG internal to the macrophage-derived clone as the functional start site. In addition, this work more precisely defines the catalytic activity of UMP-CMP kinase. Here, we show a 3-fold greater substrate preference for CMP relative to UMP, identify ATP and UTP as the preferred phosphate donors for the reaction, and demonstrate that the reaction is Mg2+-dependent. In addition, investigation of UMP-CMP-kinase expression revealed two mRNA products in immune tissues and cancer cell lines. The smaller RNA product was previously undescribed.     

Isolation and characterization of a human p53 cDNA clone: expression of the human p53 gene.

A cDNA clone for human p53 cellular tumor antigen has been isolated and characterized. This clone contains the complete 3'-untranslated region and most of the open reading frame for the protein. Nucleotide sequence analysis revealed that p53 mRNA contains an Alu repeat in the 3'-untranslated region. Hybridization selection experiments showed this clone was capable of selectively binding p53 mRNA. In vitro translation of SV80 mRNA resulted in the synthesis of two immunoreactive p53 polypeptide species. Northern blot analysis showed that human p53 mRNA was 2.8 kb in length and was present in cell lines containing high and low levels of p53 protein. There appears to be only a single p53 gene in human cells and Southern blot analysis demonstrated no major genomic rearrangements or amplification of the p53 gene in the transformed cell lines examined.     

Characterization of a cDNA clone encoding a basic protein, TP2, involved in chromatin condensation during spermiogenesis in the mouse

A cDNA clone encoding a small cysteine and serine-rich basic protein has been isolated from a mouse testis cDNA library. This cDNA clone encodes the mouse homologue of a protein involved in the initial phases of condensation of chromatin during spermiogenesis in rats, TP2, based on similarities in the sequence of the carboxyl terminus, composition, molecular weight, and electrophoretic mobility. Mouse TP2 can be divided into a highly basic domain comprising about one-third of the polypeptide chain at the carboxyl terminus and a much less basic domain comprising the remaining two-thirds at the amino terminus. The 5' end of the mouse TP2 mRNA contains two in-phase initiation codons both of which may be used generating two polypeptides which differ in length at the amino terminus. Southern blots demonstrate that there is a single copy of the TP2 gene in the mouse genome and Northern blots demonstrate that the polyadenylated TP2 mRNA is present at high and essentially equal levels in early and late haploid cells, and that it is virtually absent from meiotic cells.     

Expression of the human alpha-galactosidase A in Escherichia coli K-12

We used the prokaryotic expression vector, ptrpL1, for the expression in Escherichia coli K-12 of a cDNA clone specific for the human lysosomal hydrolase, alpha-galactosidase A. The 5' terminus of the cDNA clone was engineered so that an ATG codon precedes the first codon of the mature form of the enzyme. A clone with elevated expression of this human enzyme was constructed by increasing the distance between the Shine-Dalgarno site and the ATG start codon from 6 to 8 bp. Clones with alpha-galactosidase A specific cDNA encoding the proenzyme produce a protein of 45 kDa, the size expected for the intact proenzyme. The 45-kDa protein is specifically precipitated by antibody to alpha-galactosidase A, and its expression is repressed by tryptophan and induced by 3-beta-indoleacrylic acid as expected for this expression vector. The human enzyme is produced in E. coli in a catalytically active form at levels sufficient to support the growth of cells using alpha-galactosides as sole sources of carbon and energy. In addition, bacterial colonies that produce the human enzyme turn blue in the presence of 5-bromo-4-chloro-3-indolyl-alpha-D-galactopyranoside.     

Definition of the transcription initiation site of human plasminogen gene in liver and non hepatic cell lines

We have mapped the cap site of the human plasminogen mRNA by primer extension and PCR techniques and found that it is located at position -161 relative to the first ATG, 97 bases upstream to the 5' end of the previously isolated cDNA clone. Seven human hepatic and non hepatic cell lines and fresh liver cells were tested for human plasminogen mRNA expression: the liver and the liver derived HepG2 cell line represent the major site of plasminogen RNA synthesis while the other cell lines (Hep3B, HeLa, IMR, 293 CaCo and SW626) show much lower levels.     

Characterization of the p33 subunit of eukaryotic translation initiation factor-3 from Saccharomyces cerevisiae

Eukaryotic translation initiation factor-3 (eIF3) is a large multisubunit complex that binds to the 40 S ribosomal subunit and promotes the binding of methionyl-tRNAi and mRNA. The molecular mechanism by which eIF3 exerts these functions is incompletely understood. We report here the cloning and characterization of TIF35, the Saccharomyces cerevisiae gene encoding the p33 subunit of eIF3. p33 is an essential protein of 30,501 Da that is required in vivo for initiation of protein synthesis. Glucose repression of TIF35 expressed from a GAL1 promoter results in depletion of both the p33 and p39 subunits. Expression of histidine-tagged p33 in yeast in combination with Ni2+ affinity chromatography allows the isolation of a complex containing the p135, p110, p90, p39, and p33 subunits of eIF3. The p33 subunit binds both mRNA and rRNA fragments due to an RNA recognition motif near its C terminus. Deletion of the C-terminal 71 amino acid residues causes loss of RNA binding, but expression of the truncated form as the sole source of p33 nevertheless supports the slow growth of yeast. These results indicate that the p33 subunit of eIF3 plays an important role in the initiation phase of protein synthesis and that its RNA-binding domain is required for optimal activity.