The EBER genes are unusual in that they are transcribed by RNA polymerase III and yet they utilize promoter elements normally associated with protein-coding (class II) genes (16). As with most RNA polymerase III transcripts, the 3' ends of the EBERs (and those of the HVPs) are composed of a short stretch of uridine residues. The La protein, a 50-kDa phosphoprotein first identified as an autoantigen, binds this poly(U) tail on >90% of the EBER and HVP RNAs (15). We previously identified a second protein, EAP (EBER-associated protein), which also binds to these viral small RNAs (32). EAP is an abundant, 15-kDa cellular protein which appears to be the mammalian homolog of a previously identified sea urchin protein. A cDNA encoding the sea urchin protein, which was called 217, was identified in a screen for genes expressed in a developmentally specific manner (8,12). EAP and the 217 protein are 77% identical. Unfortunately, the functions of both remain a mystery.We have now raised anti-EAP antibodies and have used them to characterize the interaction between EAP and the * Corresponding author.EBERs. Whereas previous experiments had suggested that EAP bound both EBER 1 and EBER 2 (32), the binding studies presented here indicate that EAP's association with EBER 1 is by far the more significant. In contrast to that of many other RNA-binding proteins, EAP's ability to bind the third stem-loop of EBER 1 is reduced by many mutations in double-stranded regions, even when secondary structure is maintained. Characterization of this binding site should aid in identifying the cellular RNAs associated with EAP.
MATERIALS AND METHODSGlutathione transferase fusion protein and antibody production. A glutathione S-transferase (GST)-EAP fusion vector was prepared by inserting an EAP-encoding DNA fragment containing a 5'-terminal BamHI site into an expression vector (pGEX) (2) encoding 27 kDa of the GST protein with a BamHI site at its 3' end. The EAP-encoding insert contained an open reading frame with a BamHI site followed by a factor X cleavage site and the entire EAP sequence, excluding the initiating methionine. A stop codon immediately after the EAP C terminus was followed by a short polylinker including a BamHI and a Sall site. This insert was generated by the polymerase chain reaction from an EAP cDNA template. The terminal restriction sites and factor X cleavage site were included in the deoxyoligonucleotides: 5', CGGATCCTGATCGAGGGCAGGGCCCCCGTGAAGAA GCTGGTGGTGAAGGGCGGAAAGAAGAAGAAGCA, and 3', GGGATCCAGCTGTCGACTJAATCCTCGTCTT CCTCCTCTTCTTCGTCCTGG. The polymerase chain reaction was carried out as described by Toczyski and Steitz (32). The EAP insert was cloned into pGEX-3X after digestion of both the insert and the vector with BamHI