Maize ribosome-inactivating protein (b-32). Homologs in related species, effects on maize ribosomes, and modulation of activity by pro-peptide deletions.

The ribosome-inactivating protein (RIP) from maize (Zea mays L.) is unusual in that it is produced in the endosperm as an inactive pro-form, also known as b-32, which can be converted by limited proteolysis to a two-chain active form, alpha beta RIP. Immunological analysis of seed extracts from a variety of species related to maize showed that pro/alpha beta forms of RIP are not unique to maize but are also found in other members of the Panicoideae, including Tripsacum and sorghum. Ribosomes isolated from maize were quite resistant to both purified pro- and alpha beta maize RIPs, whereas they were highly susceptible to the RIP from pokeweed. This suggests that the production of an inactive pro-RIP is not a mechanism to protect the plant's own ribosomes from deleterious action of the alpha beta RIP. RIP derivatives with various pro-segments removed were expressed at high levels in Escherichia coli. Measurement of their activity before and after treatment with subtilisin Carlsberg clearly identified the 25-amino acid intradomain insertion, rather than the N- or C-terminal extensions, as the major element responsible for suppression of enzymatic activity. A RIP with all three processed regions deleted had activity close to that of the native alpha beta form.

Characterization and molecular cloning of a proenzyme form of a ribosome-inactivating protein from maize. Novel mechanism of proenzyme activation by proteolytic removal of a 2.8-kilodalton internal peptide segment.

Ribosome-inactivating proteins (RIPs) are a widely distributed family of plant enzymes that are remarkably potent catalytic inactivators of eukaryotic protein synthesis. All RIPs described to date, including the A-chain of the plant cytotoxin ricin, are polypeptides of 25-32 kDa and share significant amino acid sequence homologies. We have characterized and cloned an RIP from maize (Zea mays). In contrast to previously described RIPs, we have found that maize RIP is synthesized and stored in the kernel as a 34-kDa inactive precursor (isoelectric point = 6.5). During germination, this neutral precursor is converted into a basic, active form (isoelectric point greater than 9) by limited proteolysis, which removes 25 amino acids (2.8 kDa) of net charge -6 from the center of the polypeptide chain. Additional processing also occurs at the amino and carboxyl termini of the polypeptide. The sequence of the internal processed region is unique and it is equivalent to an insertion centered around Thr-156 in the amino acid sequence of ricin toxin A-chain, i.e. in the center of the enzymatically active domain. The generation of an active enzyme by removal of a large amino acid segment from the middle of a precursor polypeptide chain represents a novel mechanism of proenzyme activation that is distinct from more conventional activation mechanisms involving NH2-terminal proteolytic processing. A two-chain active RIP (comprised of 16.5- and 8.5-kDa fragments that remain tightly associated) is produced from this processing event.

Poliovirus snapback double-stranded RNA isolated from infected HeLa cells is deficient in poly(A).

A portion of poliovirus double-stranded RNA (25 to 50%) isolated from infected HeLa cells contains hairpin loops at one end of the duplex structure. These structures rapidly reformed double-stranded molecules after denaturation and appeared as molecules of up to two times genome length upon electrophoresis in denaturing agarose gels. A second form of poliovirus double-stranded RNA was readily denaturable into genome length strands. When the hairpin RNA was treated with S1 nuclease, subsequent denaturation resulted in formation of strands of up to genome length. Hairpin molecules contained very little, if any, poly(A) sequences, suggesting that the hairpin forms after nucleolytic removal of the 3' end of plus-strand templates. We conclude that the hairpin double-stranded RNA found in infected cells is likely generated by intracellular nicking and self-priming and that it does not represent an intermediate in the process of RNA replication.

Host factor-induced template modification during synthesis of poliovirus RNA in vitro.

Poliovirus RNA polymerase, 3Dpol, transcribes poliovirus RNA in vitro in the presence of a host factor (HF) preparation from uninfected HeLa cells to yield heterogeneous-size product RNAs. The products include some molecules larger than the template that represent self-primed elongations of the template from a 3'-terminal hairpin. We showed that transcription proceeded through the formation of a modified RNA intermediate that was generated by nucleolytic cleavage of the template by HF in the absence of nucleoside triphosphates. Cleavage resulted in the loss of the original poly(A) 3' end and the generation of new, heterogeneous 3' ends that formed self-priming structures that could then be elongated by 3Dpol or reverse transcriptase. The two stages of the reaction, (i) cleavage to yield self-priming templates and (ii) subsequent chain elongation to yield heterogeneous-size products up to nearly dimer length, could be separated. RNAs whose original 3' ends were chemically oxidized so as to prevent chain elongation showed no reduction in template activity after preincubation with HF. We conclude that an HF preparation that contains a low level of nuclease activity is sufficient to activate RNA templates for transcription by 3Dpol to generate up to apparent dimer-length products. This reaction likely has little relevance to the mechanism of poliovirus RNA replication in vivo. It is likely that numerous other factors or activities, in addition to 3Dpol, would also result in transcription of poliovirus RNA in vitro.

Synthesis of plus- and minus-strand RNA from poliovirion RNA template in vitro.

The poliovirus RNA polymerase, 3Dpol, was used to synthesize RNA in vitro in the presence of a host factor preparation from uninfected HeLa cells and poliovirion RNA as the template. The transcription products included molecules approximately twice the length of the template, apparently resulting from hairpin formation and template-directed elongation, as previously reported (D. C. Young, D. M. Tuschall, and J. B. Flanegan, J. Virol. 54:256-264, 1985). Other polyadenylated template RNAs also yielded products that were twice the length of the template. The polarity of the products synthesized from plus-strand poliovirus RNA template was analyzed by Southern blotting using labeled product RNA to probe single-stranded poliovirus DNAs cloned into M13 vectors. The results demonstrated that host factor-mediated polymerase products contain newly synthesized plus-strand sequences as well as the expected minus-strand sequences. Polymerase products primed with oligo(U) were all of minus-strand polarity.

Two forms of VPg on poliovirus RNAs.

The protein (VPg) linked to the 5' termini of poliovirus RNAs resolved into two species when subjected to non-equilibrium electrofocusing. The differently charged forms of VPg were not due to protein phosphorylation nor to variability of the number of phosphate residues associated with the nucleotide moiety remaining after RNase digestion of the nucleoprotein. Single-stranded viral RNA isolated from mature virions contained predominantly the more basic form of VPg, whereas unpackaged single-stranded RNa remaining in cells at the end of the virus replication cycle contained predominantly the more acidic form of VPg. Replicative-form (RF) molecules also contained both species of VPg, with the more acidic form representing the major species. Both plus and minus RNA strands in RF had similar VPg compositions; however, there appeared to be a strongly selective loss of VPg from only the minus strand in RF, particularly at late times postinfection.