Intron-specific RNA binding proteins in the chloroplast of the green alga Chlamydomonas reinhardtii.

Mitochondria and chloroplasts both contain group II introns which are believed to be the ancestors of nuclear spliceosomal introns. We used the mitochondrial group II intron rI1 from the green alga Scenedesmus obliquus for biochemical characterization of intron-specific RNA binding proteins. rI1 is correctly spliced from a chloroplast precursor RNA when integrated into the chloroplast genome of Chlamydomonas reinhardtii. Glycerol gradients revealed the sedimentation profile of transcripts containing intron rI1 in native C. reinhardtii extracts and in deproteinized RNA preparations, thus indicating the association of rI1 containing transcripts with high molecular weight ribonucleoprotein complexes in vivo. Furthermore, the specific binding of a 61 kDa protein and a 31 kDa protein with the conserved domain IV was demonstrated using a set of intron derivatives for in vitro RNA binding experiments. We propose that we have biochemically characterized 'general splicing factors', which enable the successful splicing even of mitochondrial introns in chloroplasts.

The Nac2 gene of Chlamydomonas encodes a chloroplast TPR-like protein involved in psbD mRNA stability.

The psbD mRNA, which encodes the D2 reaction center polypeptide of photosystem II, is one of the most abundant chloroplast mRNAs. We have used genomic complementation to isolate the nuclear Nac2 gene, which is required for the stable accumulation of the psbD mRNA in Chlamydomonas reinhardtii. Nac2 encodes a hydrophilic polypeptide of 1385 amino acids with nine tetratricopeptide-like repeats (TPRs) in its C-terminal half. Cell fractionation studies indicate that the Nac2 protein is localized in the stromal compartment of the chloroplast. It is part of a high molecular weight complex that is associated with non-polysomal RNA. Change of a conserved alanine residue of the fourth TPR motif by site-directed mutagenesis leads to aggregation of Nac2 protein and completely abrogates its function, indicating that this TPR is important for proper folding of the protein and for psbD mRNA stability, processing and/or translation.

Identification of cis-acting RNA leader elements required for chloroplast psbD gene expression in Chlamydomonas.

The psbD mRNA of Chlamydomonas reinhardtii is one of the most abundant chloroplast transcripts and encodes the photosystem II reaction center polypeptide D2. This RNA exists in two forms with 5' untranslated regions of 74 and 47 nucleotides. The shorter form, which is associated with polysomes, is likely to result from processing of the larger RNA. Using site-directed mutagenesis and biolistic transformation, we have identified two major RNA stability determinants within the first 12 nucleotides at the 5' end and near position -30 relative to the AUG initiation codon of psbD. Insertion of a polyguanosine tract at position -60 did not appreciably interfere with translation of psbD mRNA. The same poly(G) insertion in the nac2-26 mutant, which is known to be deficient in psbD mRNA accumulation, stabilized the psbD RNA. However, the shorter psbD RNA did not accumulate, and the other psbD RNAs were not translated. Two other elements were found to affect translation but not RNA stability. The first comprises a highly U-rich sequence (positions -20 to -15), and the second, called PRB1 (positions -14 to -11), is complementary to the 3' end of the 16S rRNA. Changing the PRB1 sequence from GGAG to AAAG had no detectable effect on psbD mRNA translation. However, changing this sequence to CCUC led to a fourfold diminished rate of D2 synthesis and accumulation. When the psbD initiation codon was changed to AUA or AUU, D2 synthesis was no longer detected, and psbD RNA accumulated to wild-type levels. The singular organization of the psbD 5' untranslated region could play an important role in the control of initiation of psbD mRNA translation.

The 54 kDa RNA-binding protein from mustard chloroplasts mediates endonucleolytic transcript 3' end formation in vitro.

A 54 kDa protein from mustard chloroplasts was previously shown to interact specifically with a conserved U-rich sequence element in RNA derived from the 3' flanking regions of the plastid trnK and rps16 genes, which code for tRNA(Lys) and ribosomal protein CS19, respectively (Nickelsen and Link, 1991). This RNA-binding protein has now been purified by affinity chromatography on heparin Sepharose and poly(U) Sepharose. In vitro processing experiments and nuclease S1 analyses of the processing products revealed that the 54 kDa polypeptide is an endonuclease. The in vitro cleavage sites are consistent with the positions of corresponding transcript in vivo 3' ends downstream of trnK and rps16, suggesting that RNA 3' end formation takes place endonucleolytically also in vivo.

RNA-protein interactions at transcript 3' ends and evidence for trnK-psbA cotranscription in mustard chloroplasts.

In vitro transcripts from the 3' flanking regions of mustard chloroplast genes were tested for protein binding in a chloroplast extract. Efficient and sequence-specific RNA-protein interaction was detected with transcripts of the genes trnK, rps16 and trnH, but not with the 3' terminal region of trnQ RNA. The transacting component required for specific complex formation is probably a single 54 kDa polypeptide. The protein-binding region of the rps16 3' terminal region was mapped and compared with that of the trnK transcript determined previously. Both regions reveal a conserved 7-mer UUUAUCU followed by a stretch of U residues. Deletion of the trnK 3' U cluster resulted in more than 80% reduction in the binding activity, and after deletion of both the U stretch and the 7-mer motif no binding at all was detectable. RNase protection experiments indicate that the protein-binding regions of both the rps16 and trnK transcripts correlate with the positions of in vivo 3' ends, suggesting an essential role for the 54 kDa binding protein in RNA 3' end formation. In the case of the trnK gene, evidence was obtained for read-through transcripts that extend into the psbA coding region, thus pointing to the possibility of trnK-psbA cotranscription.