Characterization of cDNA sequences for LHCI apoproteins in Euglena gracilis: the mRNA encodes a large precursor containing several consecutive divergent polypeptides

Abstract

Screening of a lambda gt11 cDNA expression library with antibodies directed against LHCII allowed the detection of several clones which differ markedly from the previously described cDNAs encoding LHCII apoproteins. A Northern analysis revealed a transcript size of 4.2 kb, whereas clones encoding LHCII hybridize with mRNAs in the range of 7.5 kb. Nucleotide sequencing of 2 clones showed open reading frames of 530 and 331 codons, respectively. Within these reading frames, 5 analogous motifs can be delimited corresponding to coding regions for around 180 amino acids (molecular weight, 18-19 kDa). The 5 segments share between 50% and 80% homology. Comparison with a tomato LHCI sequence indicates conserved regions at the two ends, and a central part highly divergent and containing a large deletion. By hybrid-selected translation, followed by immunoprecipitation with a monoclonal antibody directed against LHCI apoprotein, a protein of around 100 kDa is obtained. In vivo, the same antibody recognizes peptides of around 20 kDa. These results, coupled with our previous observations concerning LHCII, confirm that in Euglena at least, some chloroplast proteins encoded in the nucleus are synthesized by large mRNAs containing the information for several consecutive divergent peptides. Implications for processing and chloroplast import are discussed.