Journal of General Virology (1996), 77, 889-897. Printed in Great Britain 889 Beet necrotic yellow vein virus 42 kDa triple gene block protein binds nucleic acid in vitro Claudine Bleykasten,* D. Gilmer, H. Guilley, K. E. Richards and G. Jonard Institut de Biologie Mol6culaire des Plantes du CNRS et de l'Universit~ Louis Pasteur, 67084 Strasbourg Cedex, France The triple gene block (TGB) of beet necrotic yellow vein virus RNA 2 is required for cell-to-cell movement of the virus RNA. The protein P42 encoded by the 5'-proximal gene of the TGB has consensus sequence motifs characteristic of an ATP/GTP-dependent helicase. P42 was over-expressed in Escherichia coli and shown to bind both single- and double-stranded RNA and DNA by Northwestern blotting. Site-directed mutagenesis located the nucleic acid-binding domain to the N- terminal 24 amino acids of the protein and a point mutation or deletions in the region of P42 containing the helicase consensus sequences did not affect nucleic acid- binding activity of the immobilized protein. Electro- phoretic mobility-shift assays revealed that P42 also binds nucleic acids in solution and that deletion of the N-terminal region inhibits this binding. Mutations in both the N-terminal nucleic acid-binding domain and the helicase domain blocked infection of leaves, indica- ting that both regions of P42 are important for its activity in vivo. Introduction Beet necrotic yellow vein virus (BNYVV) is responsible for the rhizomania disease of sugarbeet. The genome of the virus consists of four plus-sense ssRNA molecules ranging from 6746 (RNA 1) to 1467 (RNA 4) nucleotides in length (for review see Richards & Tamada, 1992). RNA 1 encodes the virus replicase and is necessary and sufficient for infection of protoplasts. Moreover, RNA 1 combined with RNA 2 are necessary and sufficient for infection of leaves. RNAs 3 and 4 encode accessory functions involved in infection of roots and virus transmission by the fungal vector, Polymyxa betae. The genetic organization of BNYVV RNA 2 is shown in Fig. 1. The 5'-proximal pair of genes encodes the major and minor virus capsid proteins (Schmitt et al., 1992; Haeberle et al., 1994) and the 3'-proximal gene encodes a protein which regulates RNA 2 replication and coat protein expression (Hehn et al., 1995). The central portion of RNA 2 contains a cluster of three slightly overlapping genes known as the triple gene block (TGB), encoding, in order, the proteins P42, P13 and a putative protein P15. Mutations to knock out each of these proteins separately did not interfere with infection of protoplasts but abolished lesion formation * Author for correspondence. Fax +33 88 61 44 42. e-mail bleykasten@medoc.u-st rasbg.fr on leaves, demonstrating that the TGB plays a role in cell-to-cell movement of the virus RNA (Gilrner et al., 1992). TGB homologues (Petty & Jackson, 1990; Beck et al., 1991 ; Koonin & Dolja, 1993; Herzog et al., 1994) are present in the genomes of a number of other plant viruses, including the potex- and carlaviruses, barley stripe mosaic hordeivirus (BSMV) and peanut clump furovirus (PCV). Little information is available concerning the mech- anism by which the TGB proteins might facilitate cell-to- cell movement. It is known, however, that defective coat protein mutants of BNYVV and BSMV are infectious to leaves, indicating that virion formation is not required for passage from one cell to another (Petty & Jackson, 1990; Schmitt et al., 1992). This observation suggests that TGB-mediated cell-to-cell movement involves a virus protein capable of interacting with virus RNA. The first TGB protein of BNYVV, P42, has sequence motifs characteristic of a superfamily 1 DNA or RNA helicase (Koonin & Dolja, 1993), including a 'P-Loop' ATP/GTP-binding domain (Fig. 1). These motifs are also present in the corresponding TGB proteins of the other viruses referred to above. In this paper we have expressed BNYVV P42 in bacteria and showed that it possesses nucleic acid-binding activity. Site-directed mutagenesis revealed that the binding site is situated near the N terminus of the protein. The P-Loop motif is not required for nucleic acid binding. Deletions within both the N-terminal nucleic acid-binding domain and the 0001-3735 © 1996 SGM