Sequence and phylogenetic analysis of lambdaA and lambdaC protein encoding genes of 12 avian reoviruses is described. The sequence of lambdaA possesses a variable region (residues 19-51) located within a conserved hydrophilic region (residues 1-110) and a C(2)H(2) zinc-binding motif (residues 182-202). lambdaC shows the two conserved K residues at positions 169 and 188 indicative of guanylyltransferase activity, an ATP/GTP-binding site motif A (residues 379-386), and a conserved S-adenosyl-l-methionine-binding motif (residues 822-830). Pairwise sequence comparisons show that the mean sequence identities of lambdaA encoding genes and lambdaA proteins are 92% and 98%, respectively, and those of lambdaC encoding genes and lambdaC proteins are 91% and 95%, respectively. Phylogenetic analysis of lambdaA and lambdaC encoding genes reveals that both encoding genes have diverged into three distinct lineages, respectively, and that there is no correlation between lineages and viral serotypes or pathotypes. Also, reassortment of gene segments L1 and L3 has been observed between viruses.

Analysis of the amino acid sequence of core protein muA of avian reovirus has indicated that it may share similar functions to protein mu2 of mammalian reovirus. Since mu2 displayed both nucleotide triphosphatase (NTPase) and RNA triphosphatase (RTPase) activities, the purified recombinant muA ( muA) was designed and used to test these activities. muA was thus expressed in bacteria with a 4.5 kDa fusion peptide and six His tags at its N terminus. Results indicated that muA possessed NTPase activity that enabled the protein to hydrolyse the beta-gamma phosphoanhydride bond of all four NTPs, since NDPs were the only radiolabelled products observed. The substrate preference was ATP>CTP>GTP>UTP, based on the estimated k(cat) values. Alanine substitutions for lysines 408 and 412 (K408A/K412A) in a putative nucleotide-binding site of muA abolished NTPase activity, further suggesting that NTPase activity is attributable to protein muA. The activity of muA is dependent on the divalent cations Mg(2+) or Mn(2+), but not Ca(2+) or Zn(2+). Optimal NTPase activity of muA was achieved between pH 5.5 and 6.0. In addition, muA enzymic activity increased with temperature up to 40 degrees C and was almost totally inhibited at temperatures higher than 55 degrees C. Tests of phosphate release from RNA substrates with muA or K408A/K412A muA indicated that muA, but not K408A/K412A muA, displayed RTPase activity. The results suggested that both NTPase and RTPase activities of muA might be carried out at the same active site, and that protein muA could play important roles during viral RNA synthesis.

Sequence and phylogenetic analysis of λA and λC protein encoding genes of 12 avian reoviruses is described. The sequence of λA possesses a variable region (residues 19–51) located within a conserved hydrophilic region (residues 1–110) and a C2H2 zinc-binding motif (residues 182–202). λC shows the two conserved K residues at positions 169 and 188 indicative of guanylyltransferase activity, an ATP/GTP-binding site motif A (residues 379–386), and a conserved S-adenosyl-l-methionine-binding motif (residues 822–830). Pairwise sequence comparisons show that the mean sequence identities of λA encoding genes and λA proteins are 92% and 98%, respectively, and those of λC encoding genes and λC proteins are 91% and 95%, respectively. Phylogenetic analysis of λA and λC encoding genes reveals that both encoding genes have diverged into three distinct lineages, respectively, and that there is no correlation between lineages and viral serotypes or pathotypes. Also, reassortment of gene segments L1 and L3 has been observed between viruses.

The sequences and phylogenetic analyses of the M-class genome segments of 12 avian reovirus strains are described. The S1133 M1 genome segment is 2283 base pairs long, encoding a protein μA consisted of 732 amino acids. Each M2 or M3 genome segment of 12 avian reovirus strains is 2158 or 1996 base pairs long, respectively, encoding a protein μB or μNS consisted of 676 and 635 amino acids, respectively. The S1133 genome segment has the 5′ GCUUUU terminal motif, but each M2 and M3 genome segment displays the 5′ GCUUUUU terminal motif which is common to other known avian reovirus genome segments. The UCAUC 3′-terminal sequences of the M-class genome segments are shared by both avian and mammalian reoviruses. Noncoding regions of both 5′- and 3′-termini of the S1133 M1 genome segment consist of 12 and 72 nucleotides, respectively, those of each M2 genome segment consist of 29 and 98 nucleotides, respectively, and those of each M3 genome segment are 24 and 64 nucleotides, respectively. Analysis of the average degree of the M-class gene and the deduced μ-class protein sequence identities indicated that the M2 genes and the μB proteins have the greatest level of sequence divergence. Computer searches revealed that the μA possesses a sequence motif (NH2-Leu-Ala-Leu-Asp-Pro-Pro-Phe-COOH) (residues 458–464) indicative of N-6 adenine-specific DNA methylase. Examination of the μB amino acid sequences indicated that the cleavage site of μB into μBN and μBC is between positions 42 and 43 near the N-terminus of the protein, and this site is conserved for each protein. During in vitro treatment of virions with trypsin to yield infectious subviral particles, both the N-terminal fragment δ and the C-terminal fragment φ were shown to be generated. The site of trypsin cleavage was identified in the deduced amino acid sequence of μB by determining the amino-terminal sequences of φ proteins: between arginine 582 and glycine 583. The predicted length of δ generated from μBC is very similar to that of δ generated from mammalian reovirus μ1C. Taken together, protein μB is structurally, and probably functionally, similar to its mammalian homolog, μ1. In addition, two regions near the C-terminal and with a propensity to form α-helical coiled-coil structures as previously indicated are observed for each protein μB. Phylogenetic analysis of the M-class genes revealed that the predicted phylograms delineated 3 M1, 5 M2, and 2 M3 lineages, no correlation with serotype or pathotype of the viruses. The results also showed that M2 lineages I–V consist of a mixture of viruses from the M1 and M3 genes of lineages I–III, reflecting frequent reassortment of these genes among virus strains.

We have previously shown that avian reovirus (ARV) σA and σNS proteins possess dsRNA and ssRNA binding activity and suggested that there are two epitopes on σA (I and II) and three epitopes (A, B, and C) on σNS. To further define the location of epitopes on σA and σNS proteins and to further elucidate the biological functions of these epitopes by using monoclonal antibodies (MAbs) 62, 1F9, H1E1, and 4A123 against the ARV S1133 strain, the full-length and deletion fragments of S2 and S4 genes of ARV generated by polymerase chain reaction (PCR) were cloned into pET32 expression vectors and the fusion proteins were overexpressed in Escherichia coli BL21 strain. Epitope mapping using MAbs and E. coli-expressed deletion fragments of σA and σNS of the ARV S1133 strain, synthetic peptides, and the cross reactivity of MAbs to heterologous ARV strains demonstrated that epitope II on σA was located at amino acid residues 340QWVMAGLVSAA350 and epitope B on σNS at amino acid residues 180MLDMVDGRP188. The MAbs (62, 1F9, and H1E1) directed against epitopes II and B did not require the native conformation of σA and σNS, suggesting that their binding activities were conformation-independent. On the other hand, MAb 4A123 only reacted with complete σNS but not with truncated σNS fusion proteins in Western blot, suggesting that the binding activity of MAb to epitope A on σNS was conformation-dependent. Amino acid sequence analysis and the binding assays of MAb 62 to heterologous ARV strains suggested that epitope II on σA was highly conserved among ARV strains and that this epitope is suitable as a serological marker for the detection of ARV antibodies following natural infection in chickens. On the contrary, an amino acid substitution at position 183 (M to V) in epitope B of ARV could hinder the reactivity of the σNS with MAb 1F9. The σNS of ARV with ssRNA-binding activity could be blocked by monoclonal antibody 1F9. The epitope B on σNS is required for ssRNA binding because its deletion fully abolished the ssRNA binding activity of σNS.

Both avian reovirus core protein sigma A purified from virus-infected cell extracts and the purified bacterially expressed protein sigma A (e sigma A) were characterized for their nucleoside triphosphate (NTP) hydrolysis activity by thin-layer chromotography. Protein sigma A from both preparations has a nonspecific nucleotidyl phosphatase activity that hydrolyzes four types of NTP to their corresponding nucleoside di- and monophosphates and free phosphate. The divalent cation requirement for this activity of e sigma A was further examined by the addition of Mn(2+), Mg(2+), Ca(2+), and Zn(2+) ions. NTP hydrolysis by e sigma A was maximal when Mn(2+), Mg(2+), or Ca(2+) concentrations were 5, 4, or 1 mM, respectively. Addition of Mn(2+) or Mg(2+) stimulated the reactions up to 4- or 3-fold, respectively, higher than Ca(2+) (2.2-fold). However, Zn(2+) ion inhibited this activity of e sigma A. The results suggest that nucleotidyl phosphatase activity of e sigma A is absolutely dependent on the divalent cations Mn(2+), Mg(2+), or Ca(2+), but not Zn(2+). Similar results were obtained from the analysis of divalent cation requirements for the protein sigma A nucleotidyl phosphatase activity. Optimal pH for nucleotidyl phosphatase activity of protein sigma A from both preparations was determined using reaction mixtures buffered at different pH. The results show that the optimal activities of both proteins were similar and were achieved between pH 7.5 and 8.5.

Both avian reovirus core protein σA purified from virus-infected cell extracts and the purified bacterially expressed protein σA (eσA) were characterized for their nucleoside triphosphate (NTP) hydrolysis activity by thin-layer chromotography. Protein σA from both preparations has a nonspecific nucleotidyl phosphatase activity that hydrolyzes four types of NTP to their corresponding nucleoside di- and monophosphates and free phosphate. The divalent cation requirement for this activity of eσA was further examined by the addition of Mn2+, Mg2+, Ca2+, and Zn2+ ions. NTP hydrolysis by eσA was maximal when Mn2+, Mg2+, or Ca2+ concentrations were 5, 4, or 1 mM, respectively. Addition of Mn2+ or Mg2+ stimulated the reactions up to 4- or 3-fold, respectively, higher than Ca2+ (2.2-fold). However, Zn2+ ion inhibited this activity of eσA. The results suggest that nucleotidyl phosphatase activity of eσA is absolutely dependent on the divalent cations Mn2+, Mg2+, or Ca2+, but not Zn2+. Similar results were obtained from the analysis of divalent cation requirements for the protein σA nucleotidyl phosphatase activity. Optimal pH for nucleotidyl phosphatase activity of protein σA from both preparations was determined using reaction mixtures buffered at different pH. The results show that the optimal activities of both proteins were similar and were achieved between pH 7.5 and 8.5.

Ten monoclonal antibodies (MAbs) were prepared against the nonstructural protein sigmaNS of avian reovirus S1133. Eight of them were selected for two-way competitive binding assay after coupling with horseradish peroxidase. The results allowed the definition of three epitopes, designated A, B, and C. Blocking assay of poly(A)-Sepharose binding activity of sigmaNS with MAbs indicated that MAb recognizing epitope B was able to block poly(A) oligomer binding, suggesting that epitope B is involved in ssRNA binding of sigmaNS. An immuno-dot binding assay was used to analyze the effect of denaturation on antibody recognition of the epitopes. All MAbs bound to protein sigmaNS in its native form. After denaturation by boiling in SDS and 2-mercaptoethanol, the binding of MAbs recognizing epitopes B and C was not affected. The reactivity of MAbs recognizing epitope A was fully abolished by denaturation. These results suggest that the binding of MAbs directed against epitope A is conformation-dependent; however, the recognition by MAbs of epitopes B and C is not conformation-dependent. In addition, the results from the cross-reactivity of MAbs to heterologous avian reovirus strains suggest that the three epitopes are highly conserved among these virus strains.