The RNA helicase and nucleotide triphosphatase activities of the bovine viral diarrhea virus NS3 protein are essential for viral replication.

Helicase/nucleoside triphosphatase (NTPase) motifs have been identified in many RNA virus genomes. Similarly, all the members of the Flaviviridae family contain conserved helicase/NTPase motifs in their homologous NS3 proteins. Although this suggests that this activity plays a critical role in the viral life cycle, the precise role of the helicase/NTPase in virus replication or whether it is essential for virus replication is still unknown. To determine the role of the NS3 helicase/NTPase in the viral life cycle, deletion and point mutations in the helicase/NTPase motifs of the bovine viral diarrhea virus (BVDV) (NADL strain) NS3 protein designed to abolish either helicase activity alone (motif II, DEYH to DEYA) or both NTPase and helicase activity (motif I, GKT to GAT and deletion of motif VI) were generated. The C-terminal domain of NS3 (BVDV amino acids 1854 to 2362) of these mutants and wild type was expressed in bacteria, purified, and assayed for RNA helicase and ATPase activity. These mutations behaved as predicted with respect to RNA helicase and NTPase activities in vitro. When engineered back into an infectious cDNA for BVDV (NADL strain), point mutations in either the GKT or DEYH motif or deletion of motif VI yielded RNA transcripts that no longer produced infectious virus upon transfection of EBTr cells. Further analysis indicated that these mutants did not synthesize minus-strand RNA. These findings represent the first report unequivocably demonstrating that helicase activity is essential for minus-strand synthesis.

Systematic analysis of hMSH2 and hMLH1 in young colon cancer patients and controls.

Germ-line mutations in DNA mismatch-repair genes impart a markedly elevated cancer risk, often presenting as autosomal dominant hereditary nonpolyposis colorectal cancer (HNPCC). However, there are no pathognomonic features of HNPCC, not all gene carriers have a family history of the disease, and families fulfilling the Amsterdam criteria are relatively uncommon. Genetic testing of probands with early-onset colorectal cancer, irrespective of family history, is one approach that would allow predictive genetic testing of at-risk relatives. We cloned and sequenced hMSH2 and hMLH1 introns, to optimize genomic sequencing. We then systematically analyzed the entire hMSH2 and hMLH1 genes, by genomic sequencing and in vitro synthesized-protein-truncation assay (IVSP), in 50 colorectal cancer patients <30 years of age at diagnosis. To determine polymorphic variants, 26 anonymous donors also were sequenced. All subjects analyzed had at least 1 of 37 different polymorphic or pathogenic variants. IVSP complemented genomic sequencing, by detection of mutations not identified by genomic analysis. Fourteen cancer patients (28%) had pathogenic mutations, and a number of other variants also may have had a pathogenic significance that remains to be elucidated. Tumor replication-error status was useful in targeting sequencing efforts for this cohort of young patients: sensitivity was 86%, specificity 73%, and positive and negative predictive values 63% and 90%, respectively. These data indicate that an appreciable proportion of young colon cancer probands carry a germ-line mutation in a DNA mismatch-repair gene.

Evolution of host cell RNA into efficient template RNA by Q beta replicase: the origin of RNA in untemplated reactions.

Q beta replicase can replicate a single molecule of certain species of RNA to 10(14) copies in minutes. This replication ability has been used for in vitro studies of molecular evolution and is currently being utilized as a method of amplifying RNAs that contain probe sequences. It has been observed that Q beta replicase can produce replicatable RNA even in the absence of exogenously added template RNA. The origin of this RNA has been ascribed either to contamination with replicatable RNA or to an ability of Q beta replicase to synthesize RNA de novo from the nucleotides present in the reaction. Technologies that employ Q beta replicase require a thorough understanding of the conditions that lead to this so-called spontaneous RNA production. We have created an expression system and purification method with which we produce gram quantities of highly purified Q beta replicase, and we have identified reaction conditions that prevent the amplification of RNA in assays that do not contain added RNA. However, when these reaction conditions are relaxed, spontaneous RNA replication is seen in up to 100% of the assays. To understand the origin of this RNA, we have cloned several spontaneously produced RNAs. Sequence analysis of one of these RNAs shows that it arose by the evolution of Escherichia coli tRNA into a replicatable template and not by de novo synthesis from nucleoside triphosphates in the reaction.

Structural domains of the extracellular domain of human nerve growth factor receptor detected by partial proteolysis.

Using partial proteolytic cleavage, the nerve growth factor (NGF) binding site and the epitopes for two anti-NGF receptor (NGFR) monoclonal antibodies were localized on the recombinant extracellular domain (RED) of the NGFR. The RED was prepared in the baculovirus-insect cell system and was purified by immunoaffinity and ion-exchange chromatography. The four cysteine-rich repeat domains and some additional C-terminal sequences were resistant to proteolysis with papain or proteinase K. The Mr 32,000 papain-resistant fragment (P32) and the Mr 30,000 proteinase K-resistant fragment (K30) share the same N terminus as the intact RED and have C termini in the vicinity of residue 170. Even though P32 and K30 have the same N terminus and probably differ by only a small number of amino acids at the C terminus, P32, but not K30, binds 125I-NGF. As judged by Western blot analysis, two anti-NGFR antibodies (ME20.4 and NGFR5) bind to P32 but have a lesser affinity for K30. Since antibody ME20.4 inhibits NGF binding but antibody NGFR5 does not, these antibodies bind to distinct epitopes. However, these epitopes apparently are closely spaced since these antibodies compete with each other for binding to biotinylated RED. NGF, but not the control protein cytochrome c, protects RED from papain digestion. Therefore, the P32 C terminus is important for the expression of the NGF binding site and the antibody-defined epitopes, even though the NGF binding site and antibody-defined epitopes probably are not encoded by the P32 C terminus. These data suggest that complex interactions occur between different regions of the RED, and that optimum NGF binding requires the integrity of multiple RED domains, including a short sequence to the C terminus of residue 170.