Non-AUG initiation of AGAMOUS mRNA translation in Arabidopsis thaliana

The MADS box organ identity gene AGAMOUS (AG) controls several steps during Arabidopsis thaliana flower development. AG cDNA contains an open reading frame that lacks an ATG triplet to function as the translation initiation codon, and the actual amino terminus of the AG protein remains uncharacterized. We have considered the possibility that AG translation can be initiated at a non-AUG codon. Two possible non-AUG initiation codons, CUG and ACG, are present in the 5' region of AG mRNA preceding the highly conserved MADS box sequence. We prepared a series of AG genomic constructs in which these codons are mutated and assayed their activity in phenotypic rescue experiments by introducing them as transgenes into ag mutant plants. Alteration of the CTG codon to render it unsuitable for acting as a translation initiation site does not affect complementation of the ag-3 mutation in transgenic plants. However, a similar mutation of the downstream ACG codon prevents the rescue of the ag-3 mutant phenotype. Conversely, if an ATG is introduced immediately 5' to the disrupted ACG codon, the resulting construct fully complements the ag-3 mutation. The AG protein synthesized in vitro by initiating translation at the ACG position is active in DNA binding and is of the same size as the AG protein detected from floral tissues, whereas AG polypeptides with additional amino-terminal residues do not appear to bind DNA. These results indicate that translation of AG is initiated exclusively at an ACG codon and prove that non-AUG triplets may be efficiently used as the sole translation initiation site in some plant cellular mRNAs.     

Cloning of cDNAs for human phosphoribosylpyrophosphate synthetases 1 and 2 and X chromosome localization of PRPS1 and PRPS2 genes

Cloned cDNAs representing the entire, homologous (80%) translated sequences of human phosphoribosylpyrophosphate synthetase (PRS) 1 and PRS 2 cDNAs were utilized as probes to localize the corresponding human PRPS1 and PRPS2 genes, previously reported to be X chromosome linked. PRPS1 and PRPS2 loci mapped to the intervals Xq22-q24 and Xp22.2-p22.3, respectively, using a combination of in situ chromosomal hybridization and human x rodent somatic cell panel genomic DNA hybridization analyses. A PRPS1-related gene or pseudogene (PRPS1L2) was also identified using in situ chromosomal hybridization at 9q33-q34. Human HPRT and PRPS1 loci are not closely linked. Despite marked cDNA and deduced amino acid sequence homology, human PRS 1 and PRS 2 isoforms are encoded by genes widely separated on the X chromosome.     

Chloroplast and cytosolic glutamine synthetase are encoded by homologous nuclear genes which are differentially expressed in vivo

We have shown that the individual members of the plant gene family for glutamine synthetase (GS) are differentially expressed in vivo, and each encode distinct GS polypeptides which are targeted to different subcellular compartments (chloroplast or cytosol). At the polypeptide level, chloroplast GS (GS2) and cytosolic GS (GS1 and GSn) are distinct and show an organ-specific distribution. We have characterized full length cDNA clones encoding chloroplast or cytosolic GS of pea. In vitro translation products encoded by three different GS cDNA clones, correspond to the mature GS2, GS1, and GSn polypeptides present in vivo. pGS185 encodes a precursor to the chloroplast GS2 polypeptide as shown by in vitro chloroplast uptake experiments. The pGS185 translation product is imported into the chloroplast stroma and processed to a polypeptide which corresponds in size and charge to that of mature chloroplast stromal GS2 (44 kDa). The 49 amino terminal amino acids encoded by pGS185 are designated as a chloroplast transit peptide by functionality in vitro, and amino acid homology to other transit peptides. The cytosolic forms of GS (GS1 and GSn) are encoded by highly homologous but distinct mRNAs. pGS299 encodes the cytosolic GS1 polypeptide (38 kDa), while pGS341 (Tingey, S. V., Walker, E. L., and Coruzzi, G. M. (1987) EMBO. J. 6, 1-9) encodes a cytosolic GSn polypeptide (37 kDa). The homologous nuclear genes for chloroplast and cytosolic GS show different patterns of expression in vivo. GS2 expression in leaves is modulated by light, at the level of steady state mRNA and protein, while the expression of cytosolic GS is unaffected by light. The light-induced expression of GS2 is due at least in part to a phytochrome mediated response. Nucleotide sequence analysis indicates that chloroplast and cytosolic GS have evolved from a common ancestor and suggest a molecular mechanism for chloroplast evolution.     

Initiation of translation at an upstream non-AUG codon accounting for N-terminally extended minor forms of recombinant proteins expressed in insect cells

When the 34 kDa kinase domain of human spleen tyrosine kinase (Syk-KD) was expressed as a C-terminally His-tagged protein in baculovirus-infected Sf-21 insect cells, the purified protein included two forms that migrated slightly differently in SDS-polyacrylamide gel electrophoresis. Intact mass analysis and LC-MS/MS peptide mapping showed that the major and faster-migrating product had the intended amino-acid sequence and 0-6 phosphorylations. This material accounted for about 95% of the purified protein. The minor product was Syk-KD with a 26 amino-acid N-terminal extension. The result suggested the existence of an upstream alternative site for the initiation of translation, and this proved to be an ACG codon derived from the pBacPAK9 vector used to express Syk-KD. The ACG codon was preceded and followed by Kozak-type sequence elements (a purine in the -3 position and a G in the +4 position) that would have enhanced the viability of initiation at ACG. The initiating amino-acid residue was Met for both minor and major products, and both forms of the protein were α-N-acetylated. For the minor product, protein intact mass analysis and peptide mapping both gave results in agreement with the sequence predicted from the DNA. A similar result with the same underlying cause was obtained with insect cell expression of full-length Syk. It appears that similar results are possible whenever this vector is used.     

Structure of the protein encoded by the chicken proto-oncogene c-myb.

The retroviral oncogene v-myb arose by transduction of the chicken proto-oncogene c-myb. We isolated and sequenced cDNA that represents the entire coding domain of chicken c-myb. By transcribing the cDNA into mRNA in vitro and then translating the RNA, we were able to document the integrity of the cDNA and to identify the codon responsible for initiation of translation from c-myb. Two different alleles of v-myb are extant, one in the genome of avian myeloblastosis virus (AMV) and the other in the genome of erythroblastosis virus 26 (E26V). The proteins encoded by the AMV and E26V alleles of v-myb differ from the product of c-myb in three ways: at their amino termini, they lack 71 and 80 amino acids respectively; at their carboxy termini, they are deficient in 199 and 278 residues; and 11 substitutions of amino acids are scattered throughout the product of AMV allele, whereas the product of the E26V allele contains only a single substitution. The structural origins of tumorigenicity by v-myb and the biological functions of c-myb remain enigmatic. The findings and molecular clones described here should now permit a systematic exploration of these enigmas.     

Developmentally regulated plant genes: the nucleotide sequence of a wheat gliadin genomic clone

Gliadins, the major wheat seed storage proteins, are encoded by a multigene family. Northern blot analysis shows that gliadin genes are transcribed in endosperm tissue into two classes of poly(A)+ mRNA, 1400 bases (class I) and 1600 bases (class II) in length. Using poly(A)+ RNA from developing wheat endosperm we constructed a cDNA library from which a number of clones coding for alpha/beta and gamma gliadins were identified by hybrid-selected mRNA translation and DNA sequencing. These cDNA clones were used as probes for the isolation of genomic gliadin clones from a wheat genomic library. One such genomic clone was characterized in detail and its DNA sequence determined. It contains a gene for a 33-kd alpha/beta gliadin protein (a 20 amino acid signal peptide and a 266 amino acid mature protein) which is very rich in glutamine (33.8%) and proline (15.4%). The gene sequence does not contain introns. A typical eukaryotic promoter sequence is present at -104 (relative to the translation initiation codon) and there are two normal polyadenylation signals 77 and 134 bases downstream from the translation termination codon. The coding sequence contains some internal sequence repetition, and is highly homologous to several alpha/beta gliadin cDNA clones. Homology to a gamma-gliadin cDNA clone is low, and there is no homology with known glutenin or zein cDNA sequences.     

Cloning and Sequencing of the cDNA Encoding the Rubber Elongation Factor of Hevea brasiliensis

In Hevea brasiliensis, the rubber particle in the laticiferous vessel is the site of rubber (cis-1-4-polyisoprene) biosynthesis. A 14 kilodalton protein, rubber elongation factor (REF), is associated with the rubber particle in a ratio of one REF to one rubber molecule (Dennis M, Henzel W, Bell J, Kohr W, Light D [1989] J Biol Chem 264: 18618-18628; Dennis M, Light D [1989] J Biol Chem 264: 18608-18617). To obtain more information concerning the function of REF and its synthesis and assembly in the rubber particle, we isolated cDNA clones encoding REF. We used antibodies to REF to screen a Hevea leaf γgt11 cDNA expression library and obtained several positive clones. Sequence analysis of the REF cDNA clones showed that the REF mRNA contains 121 nucleotides of 5′-nontranslated sequences and a 205 nucleotide 3′-nontranslated region. The open reading frame encodes the entire 14 kilodalton REF protein without any extra amino acids (Dennis M, Henzel W, Bell J, Kohr W, Light D [1989] J Biol Chem 264: 18618-18628). The REF cDNA was subcloned in pGEM-3Z/-4Z and expressed in vitro. The translation product is a 14 kilodalton protein that can be immunoprecipitated with antibodies to REF. Addition of microsomal membranes to the in vitro translation product did not alter the mobility of the REF protein. This, and the sequence data, indicate that REF is not made as a preprotein. Our results suggest that REF is synthesized on free polysomes in the laticifer cytoplasm and that assembly of the rubber particles is likely to occur in the cytosol.     

Efficient suppression of the amber codon in E. coli in vitro translation system

An mRNA encoding the esterase from Alicyclobacillus acidocaldarius with catalytically essential serine codon (ACG) replaced by an amber (UAG) codon was used to study the suppression in in vitro translation system. Suppression of UAG by tRNA(Ser(CUA)) was monitored by determination of the full-length and active esterase. It was shown that commonly used increase of suppressor tRNA concentration inhibits protein production and therefore limits suppression. In situ deactivation of release factor by specific antibodies leads to efficient suppression already at low suppressor tRNA concentration and allows an in vitro synthesis of fully active enzyme in high yield undistinguishable from wild-type protein.