Nucleotide sequence analysis of soybean small heat shock protein genes belonging to two different multigene families.

In soybean, the small heat shock proteins of 15 to 18 kDa are encoded in the nucleus by at least two different multigene families, designated class I and class VI. Genomic DNA sequences of two new heat shock genes and flanking regions were determined: Gmhsp18.5-C, a class I gene, and Gmhsp17.9-D, the first known class VI gene. Comparison of both genes revealed a moderate homology (approx. 38%) mainly within the 3' ends of their coding regions. Hydropathic characterizations and secondary-structure predictions of the deduced amino acid sequences revealed two conserved domains within the C-terminal halves of the polypeptides that are also present in related proteins of other organisms. The transcription of both genes is heat shock dependent and the mRNA start sites, as determined by S1 nuclease mapping, are located downstream from typical TATA box sequences and multiple heat shock promoter elements such as 5' CT-GAA--TTC-AG. The putative promoter regions of the genes are preceded by long tracts of repetitive sequences with a high A + T content of 79 to 89%, which are bordered by runs of "simple sequences" such as (A) 12/13, (T)10 and (TA)10. Similar characteristic features are present in the promoter and 5'-flanking regions of other soybean heat shock genes. The possible function of these distinct sequences is discussed.