Mammalian reovirus L2 gene and lambda2 core spike protein sequences and whole-genome comparisons of reoviruses type 1 Lang, type 2 Jones, and type 3 Dearing.

The reovirus L2 genome segment encodes the core spike protein lambda2, which mediates enzymatic reactions in 5' capping of the viral plus-strand transcripts. Complete nucleotide-sequence determinations were made for the L2 genome segments of eight mammalian reoviruses, including the prototype isolates of serotypes 1 and 2: Lang (T1L) and Jones (T2J), respectively. Each L2 segment was found to be 3912 or 3915 bases in length. Partial nucleotide-sequence determinations were also made for the 3916-base L2 segment of reovirus type 3 Dearing (T3D), the prototype isolate of serotype 3. The whole-genome sequence of reovirus T3D was reported previously. The T1L L2 analysis represents completion of the whole-genome sequence of that isolate as well. The T2J L2 analysis leaves only the sequence of the M1 segment yet to be reported from the genome of that isolate. The T2J M1 sequence made available from analysis in another lab was used for initiating whole-genome comparisons of reoviruses T1L, T2J, and T3D in this report. The nine L2 gene sequences and deduced lambda2 protein sequences were used to gain further insights into the biological variability, structure, and functions of lambda2 through comparisons of the sequences and reference to the crystal structure of core-bound lambda2. Phylogenetic comparisons suggest the presence of three evolutionary lines of divergent L2 alleles among the nine isolates. Localized regions of conserved amino acids in the lambda2 crystal structure include active-site clefts of the RNA capping enzyme domains, sites of interactions between lambda2 domains within the pentameric spike structure, and sites of interaction between lambda2 subunits and other proteins in viral particles.

Identification of the guanylyltransferase region and active site in reovirus mRNA capping protein lambda2.

The 144-kDa lambda2 protein of mammalian reovirus catalyzes a number of enzymatic activities in the capping of reovirus mRNA, including the transfer of GMP from GTP to the 5' end of the 5'-diphosphorylated nascent transcript. This reaction proceeds through a covalently autoguanylylated lambda2-GMP intermediate. The smaller size of RNA capping guanylyltransferases from other organisms suggested that the lambda2-associated guanylyltransferase would be only a part of this protein. Limited proteinase K digestion of baculovirus-expressed lambda2 was used to generate an amino-terminal M(r) 42,000 fragment that appears to be both necessary and sufficient for guanylyltransferase activity. Although lysine 226 was identified by previous biochemical studies as the active-site residue that forms a phosphoamide bond with GMP in autoguanylylated lambda2, mutation of lysine 226 to alanine caused only a partial reduction in guanylyltransferase activity at the autoguanylylation step. Alanine substitution for other lysines within the amino-terminal region of lambda2 identified lysine 190 as necessary for autoguanylylation and lysine 171 as an important contributor to autoguanylylation. A novel active-site motif is proposed for the RNA guanylyltransferases of mammalian reoviruses and other Reoviridae members.

Rotavirus open cores catalyze 5'-capping and methylation of exogenous RNA: evidence that VP3 is a methyltransferase.

Rotavirus open cores prepared from purified virions consist of three proteins: the RNA-dependent RNA polymerase, VP1; the core shell protein, VP2; and the guanylyltransferase, VP3. In addition to RNA polymerase activity, open cores have been shown to contain a nonspecific guanylyltransferase activity that caps viral and nonviral RNAs in vitro. In this study, we examined the structure of RNA caps made by open cores and have analyzed open cores for other capping-related enzymatic activities. Utilizing RNase digestion and thin-layer chromatography, we found that the majority ( approximately 70%) of caps made by open cores contain the tetraphosphate linkage, GppppG, rather than the triphosphate linkage, GpppG, found on mRNAs made by rotavirus double-layered particles. Enzymatic analysis indicated that the GppppG caps resulted from the lack of a functional RNA 5'-triphosphatase in open cores, to remove the gamma-phosphate from the RNA prior to capping. RNA 5'-triphosphatases commonly exhibit an associated nucleoside triphosphatase activity, and this too was not detected in open cores. Caps of some RNAs contained an extra GMP moiety (underlined) and had the structure 3'-GpGp(p)ppGpGpC-RNA-3'. The origin of the extra GMP is not known but may reflect the cap serving as a primer for RNA synthesis. Methylated caps were produced in the presence of the substrate, S-adenosyl-l-methionine (SAM), indicating that open cores contain methyltransferase activity. UV cross-linking showed that VP3 specifically binds SAM. Combined with the results of earlier studies, our results suggest that the viral guanylyltransferase and methyltransferase are both components of VP3 and, therefore, that VP3 is a multifunctional capping enzyme.

Binding site for S-adenosyl-L-methionine in a central region of mammalian reovirus lambda2 protein. Evidence for activities in mRNA cap methylation.

One or more proteins in mammalian reovirus core particles mediate two RNA methylation activities, (guanosine-7-N)-methyltransferase and (guanosine-2'-O)-methyltransferase, that contribute to forming the 5' cap 1 structure on viral mRNA. We used UV irradiation to identify core proteins that bind S-adenosyl-L-methionine (SAM), the methyl-group donor for both methyltransferases. A [methyl-3H]SAM-binding site was observed among the reovirus lambda proteins; was shown to be specific by competition with low levels of S-adenosyl-L-homocysteine, the product of methyl-group transfer from SAM; and was subsequently localized to protein lambda2. lambda2 mediates the guanylyltransferase reaction in cap formation and was previously proposed to mediate one or both methylation reactions as well. SAM binding was demonstrated for both lambda2 in cores and lambda2 expressed in insect cells from a recombinant baculovirus. Using three different methods to cleave lambda2, a binding site for SAM was tentatively localized to a central region of lambda2, between residues 792 and 1100, which includes a smaller region with sequence similarity to the SAM-binding pocket of other methyltransferases. Alanine substitutions at positions 827 and 829 within this predicted binding region greatly reduced the capacity of baculovirus-expressed lambda2 protein to undergo UV cross-linking to SAM but had no effects on either the guanylyltransferase activity of this protein or its conformation as judged by partial proteolysis, suggesting that one or both of these residues is essential for SAM binding. Based on these findings, we propose that the two methyltransferase activities involved in mRNA capping by reovirus cores utilize a single SAM-binding pocket within a central region of lambda2.

Localization of a C-terminal region of lambda2 protein in reovirus cores.

The 144-kDa lambda2 protein is a structural component of mammalian reovirus particles and contains the guanylyltransferase activity involved in adding 5' caps to reovirus mRNAs. After incubation of reovirus T3D core particles at 52 degrees C, the lambda2 protein became sensitive to partial protease degradation. Sequential treatments with heat and chymotrypsin caused degradation of a C-terminal portion of lambda2, leaving a 120K core-associated fragment. The four other proteins in cores--lambda1, lambda3, mu2, and sigma2--were not affected by the treatment. Purified cores with cleaved lambda2 were subjected to transmission cryoelectron microscopy and image reconstruction. Reconstruction analysis demonstrated that a distinctive outer region of lambda2 was missing from the modified cores. The degraded region of lambda2 corresponded to the one that contacts the base of the sigma1 protein fiber in reovirus virions and infectious subvirion particles, suggesting that the sigma1-binding region of lambda2 is near its C terminus. Cores with cleaved lambda2 were shown to retain all activities required to transcribe and cap reovirus mRNAs, indicating that the C-terminal region of lambda2 is dispensable for those functions.

The effect of high pH upon diphtheria toxin conformation and model membrane association: role of partial unfolding.

Penetration of membranes by diphtheria toxin in vivo is at least partially triggered by a low pH-induced conformational change occurring within the lumen of an acidic organelle. In order to gain insight into the nature of this change the behavior of the toxin at high pH was characterized and compared to that previously determined at low pH. We find that near pH 10.5 a major conformational change occurs. This change is accompanied by a marked decrease in fluorescence intensity, a red shift in fluorescence emission maximum, and increased susceptibility of protein fluorescence to acrylamide quenching. Differential scanning calorimetry shows that the high pH conformational change involves a cooperative endothermic unfolding transition. These changes at high pH are very similar to those induced by low pH, supporting the conclusion that the changes at low pH also involve a denaturation-like process. In addition, at high pH the toxin gains the ability to bind to model membranes, again similar to its behavior at low pH. On the basis of these studies we conclude that exposure of hydrophobic sequences due to partial unfolding is one dominating component in inducing hydrophobic behavior at both high and low pH, but that at low pH Asp/Glu protonation also contributes to hydrophobicity.

Domain-specific bias in arginine/lysine usage by protein toxins.

The content of lysine and arginine residues in a number of A-B type protein toxins has been examined. It is found that the A subunit, or its equivalent, often shows a strong bias in the type of basic amino acid residue used tending towards nearly exclusive use of either arginine or lysine rather than use of both, whereas the B subunit or its equivalent shows no such bias. Although arginine codons are GC-rich and lysine codons are AT-rich, the content of GC and AT in the genes coding for the toxins does not adequately explain this bias. Other explanations are discussed, including the possibility that the bias is linked to catalytic function or membrane interaction. Understanding this bias may yield valuable insights into toxin structure and function. Furthermore, identification of bias in sequences may be a useful tool for identifying new toxins and their domains.