hththfh

were highly similar to C.reinhardtii HSP70A contained only

small inserts that corresponded to a small part of the probe.

Clones L1and L30both encoded proteins that were much more

similar to members of the luminal binding protein (BiP)family

of Hsp70proteins than they were to any cytoplasmic Hsp70

protein in the database.For instance,the sequenced portion of

L30(183bp)encodes a polypeptide that is 82%identical to part

of an Arabidopsis thaliana BiP (locus BAA12348,GenBank

accession number D84414.1),but is only 54%identical to the

corresponding region of C.reinhardtii HSP70A .Thus clones L1and L30likely correspond to a gene or genes encoding ER-specific Hsp70s of V .carteri .None of the hsp70cDNAs isolated was long enough to encode a full-length Hsp70,so we used sequence from the genomic clone and from the longest cDNA (D10,2094bp)to design primers for RACE to obtain a 501-bp cDNA fragment that overlaps with cDNA D10and encodes the N-terminal part of the protein.The complete cDNA encodes a deduced protein that contains a 395-aa ATPase domain,which is highly con-served in all members of the Hsp70family (Bukau

and Fig.2.Putative regulatory sequences within the vc-hsp70A gene.Sequence of part of the 6.5-kb Ssp I –Ssp I restriction fragment (Fig.1)that spans the entire vc-hsp70A gene is represented,beginning at bp −530with respect to the start codon,through the end of the fragment.The dots between the start codon and stop codon represent the coding sequence,which is not shown.Two imperfect HSE-like sequences are underlined and in bold,the predicted TATA box is in bold,alternative polyadenylation sites are indicated by arrowheads,and typical algal polyadenylation signals (TGTAA,TGAAA)that precede three of the polyadenylation sites are marked by dashed underlines.The predicted transcription start site is marked by a solid diamond below the sequence.115

Q.Cheng et al./Gene 371(2006)112–120