The 869 nt 3' untranslated region (3' UTR) of Barley yellow dwarf luteovirus (BYDV) RNA facilitates cap-independent and poly(A) tail-independent translation. A 105 nt cap-independent translation element (BTE, or BYDV TE) within the 3' UTR was previously defined using wheat germ translation extracts (Guo et al., 2000), but additional 3' UTR sequence is required for translation in plant cells. One domain, nt 4918--5008, confers cap-independent translation in vivo in a discrete and strong manner. Some of this extra sequence places the BTE on a phylogenetically conserved, long double-stranded "stalk", making it more accessible to cellular translational machinery. Another sequence in the 3' UTR, nt 5089--5142, downstream of the cap-mimic is functionally replaceable by a poly(A) tail on the 3' end of the mRNA. The influence of this sequence on translation is somewhat weaker than that of the cap-mimic sequence, and may be correlated with length. The requirement for these elements suggests that at least 250 nt of BYDV 3' UTR are necessary for the virus to compete effectively with host cell mRNAs in recruiting the translation apparatus in vivo.;The BTE recruits wheat translation initiation factors, including eukaryotic initiation factor 4E (eIF4E), the cap-binding protein, and its isoform eIFiso4E. Together with eIF4G and eIFiso4G respectively, these proteins complex to form eIF4F and eIFiso4F. Pulldown experiments using BTE RNA as bait showed that both eIF4E and eIF4G are associated with the BTE RNA. UV-crosslinking followed by immunoprecipitation with antisera to eIF4F and eIFiso4F shows that wild type BTE RNA associates with these factors while a nonfunctional mutant BTE does not. Depletion of cap-binding factors resulted in a substantial loss of translation of viral RNA in vitro which was recovered by adding back recombinant eIF4F and eIFiso4F. Thus, cap-binding proteins facilitate cap-independent translation of BYDV RNA. Finally, m7GTP added in trans inhibits translation in vitro of both uncapped and capped full-length infectious RNA, suggesting that the cap-binding site of eIF4E or eIFiso4E enables recruitment of the factor for BYDV translation.