New tetrameric forms of the rotavirus NSP4 with antiparallel helices.

Rotavirus nonstructural protein 4, the first viral enterotoxin to be identified, is a multidomain, multifunctional glycoprotein. Earlier, we reported a Ca2+-bound coiled-coil tetrameric structure of the diarrhea-inducing region of NSP4 from the rotavirus strains SA11 and I321 and a Ca2+-free pentameric structure from the rotavirus strain ST3, all with a parallel arrangement of α-helices. pH was found to determine the oligomeric state: a basic pH favoured a tetramer, whereas an acidic pH favoured a pentamer. Here, we report two novel forms of the coiled-coil region of NSP4 from the bovine rotavirus strains MF66 and NCDV. These crystallized at acidic pH, forming antiparallel coiled-coil tetrameric structures without any bound Ca2+ ion. Structural and mutational studies of the coiled-coil regions of NSP4 revealed that the nature of the residue at position 131 (Tyr/His) plays an important role in the observed structural diversity.

Severe diffraction anisotropy, rotational pseudosymmetry and twinning complicate the refinement of a pentameric coiled-coil structure of NSP4 of rotavirus.

The crystal structure of the region spanning residues 95-146 of the rotavirus nonstructural protein NSP4 from the asymptomatic human strain ST3 was determined at a resolution of 2.5 Å. Severe diffraction anisotropy, rotational pseudosymmetry and twinning complicated the refinement of this structure. A systematic explanation confirming the crystal pathologies and describing how the structure was successfully refined is given in this report.

Sequence analysis of VP4 and VP7 genes of nontypeable strains identifies a new pair of outer capsid proteins representing novel P and G genotypes in bovine rotaviruses.

During a limited epidemiological study, the serotype specificities of several isolates of bovine rotavirus, exhibiting identical electropherotypes, from a single cattle farm near Bangalore, India, could not be determined using a panel of serotyping monoclonal antibodies (MAbs) specific for G serotypes 1-6 and 10. To determine the genotypes of these isolates, the nucleotide sequences of the genes encoding the outer capsid proteins VP4 and VP7 of two representative isolates, Hg18 and Hg23, were determined. The corresponding gene sequences from the two isolates were identical, indicating that these isolates represented a single strain of bovine rotavirus. Comparison of the VP4 nucleotide (nt) and the deduced amino acid (aa) sequences with those of several human and animal rotavirus strains representing all of the currently recognized 20 different VP4 (P) genotypes revealed low nt and aa sequence identities of 61.0 to 74.2% and 57.9 to 78.2% for VP4. The percentages of amino acid homology for the VP8* and VP5* regions of VP4 were 37.7 to 67.9 and 68.1 to 84.2%, respectively. The nt and aa sequences of the VP7 gene were also distinct from those of human and animal strains belonging to the previously established 14 VP7(G) serotypes (65.9 to 75.5% nt and 59.5 to 77.6% aa identities). These findings suggest the classification of the VP4 and VP7 genes of the bovine isolates represented by Hg18 as new P and G genotypes and provide further evidence for the vast genetic/antigenic diversity of group A rotaviruses.

Characterization of human symptomatic rotavirus isolates MP409 and MP480 having 'long' RNA electropherotype and subgroup I specificity, highly related to the P6[1],G8 type bovine rotavirus A5, from Mysore, India.

In an epidemiological study of symptomatic human rotaviruses in Mysore, India during 1993 and 1994, isolates MP409 and MP480 were isolated from two children suffering from severe, acute dehydrating diarrhea. Both isolates exhibited 'long' RNA pattern and subgroup I specificity suggesting the likelihood of their animal origin. Both isolates did not react with monoclonal antibodies (MAbs) specific for serotypes G1 to G6 as well as G10. To determine the genetic origin of these isolates, complete nucleotide sequences of genes encoding the outer capsid proteins VP4 and VP7, nonstructural proteins NSP1 and NSP3 and viral enterotoxin protein NSP4 from MP409 and partial sequences of genes from MP480 were determined. Comparison of the 5' and 3' terminal sequences of 250 nucleotides revealed complete identity of the gene sequences in both strains suggesting that MP409 and MP480 are two different isolates of a single strain. Comparison of the nucleotide and deduced amino acid sequences of VP4, VP7, NSP1 and NSP3 of MP409 with published sequences of strains belonging to different serotypes revealed that both outer capsid proteins VP4 and VP7 and NSPI are highly related to the respective proteins from the P6[1], G8 type bovine rotavirus A5 isolated from a calf with diarrhoea in Thailand and that the NSP3 is highly homologous to that of bovine rotaviruses. The NSP4 protein showed greatest sequence identity with NSP4s belonging to the KUN genetic group to which NSP4s from human G2 type strains and bovine rotaviruses belong. MP409 and MP480 likely signify interspecies transmission of P6[1], G8 type strains from cattle to humans and represent the first P6[1] type rotaviruses isolated in humans. These and our previous studies on the asymptomatic neonatal strain 1321 are of evolutionary and epidemiological significance in the context of close association of majority of the Indian population with cattle.

Epidemiology of symptomatic human rotaviruses in Bangalore and Mysore, India, from 1988 to 1994 as determined by electropherotype, subgroup and serotype analysis.

Epidemiology of symptomatic rotaviruses from Bangalore and Mysore in Southern India was investigated. While serotype G3 predominated throughout the 7-year study period from 1988 to 1994 in Bangalore, serotype G1 was more predominant than serotype G3 in Mysore during 1993 and 1994. Serotype G2 strains were either not detected or infrequently observed in both the cities. However, several strains with subgroup I and 'short' RNA pattern that exhibited high reactivity with typing MAbs specific for serotype 2 as well as other serotypes were detected throughout the period. Among the nonserotypeable strains from both cities, several exhibited dual subgroup (SGI + II) or subgroup I specificity and 'long' RNA pattern indicating their probable animal origin. Notably, a gradual, yet highly significant reduction in rotavirus gastroenteritis, from 45.3% in 1988 to 1.8% during 1994, was observed in Bangalore in stark contrast to the consistently high (about 34%) incidence of asymptomatic infections among neonates by I321-like G10P11 type strains during the same period. Moreover, I321-like asymptomatic strains were not detected in children with diarrhea.

Nucleotide sequence and expression in E. coli of the complete P4 type VP4 from a G2 serotype human rotavirus.

The complete sequence of a P4 type VP4 gene from a G2 serotype human rotavirus, IS2, isolated in India has been determined. Although the IS2 VP4 is highly homologous to the other P4 type alleles, it contained acidic amino acid substitutions at several positions that make it acidic among the P4 type alleles that are basic. Moreover, comparative sequence analysis revealed unusual polymorphism in members of the P4 type at amino acid position 393 which is highly conserved in members of other VP4 types. To date, expression of complete VP4 in E. colic has not been achieved. In this study we present successful expression in E. coli of the complete VP4 as well as VP8* and VP5* cleavage subunits in soluble form as fusion proteins of the maltose-binding protein (MBP) and their purification by single-step affinity chromatography. The hemagglutinating activity exhibited by the recombinant protein was specifically inhibited by the antiserum raised against it. Availability of pure VP4 proteins should facilitate development of polyclonal and monoclonal antibodies (MAbs) for P serotyping of rotaviruses.

High frequency adventitious shoot regeneration from excised leaves ofPaulownia spp. cultured in vitro.

High frequency, direct regeneration of shoots was induced in leaf cultures ofPaulownia tomentosa, P. fortunei x P. tomentosa andP. kawakamii. The optimum culture medium for the leaf explants derived from shoot cultures was Murashige-Skoog (MS) medium supplemented with 10 µM indole-3-acetic acid and 50 µM benzyladenine. Up to 40 shoots were obtained over a 4 month culture period from each leaf explant. Rooting occurred spontaneously in the shoots that were about 1 cm tall when subcultured on phytohormone-free MS medium. The plantlets could be transplanted successfully. Some of the transplantedP. tomentosa plantlets flowered in the greenhouse one year after transplanting. The protocol is suitable not only for rapid multiplication of the various species ofPaulownia, but also for analytical studies associated with adventitious shoot regeneration.

Comparative nucleotide and amino acid sequence analysis of the sequence-specific RNA-binding rotavirus nonstructural protein NSP3.

NSP3, an acidic nonstructural protein, encoded by gene 7 has been implicated as the key player in the assembly of the 11 viral plus-strand RNAs into the early replication intermediates during rotavirus morphogenesis. To date, the sequence of NSP3 from only three animal rotaviruses (SA11, SA114F, and bovine UK) has been determined and that from a human strain has not been reported. To determine the genetic diversity among gene 7 alleles from group A rotaviruses, the nucleotide sequence of the NSP3 gene from 13 strains belonging to nine different G serotypes, from both humans and animals, has been determined. Based on the amino acid sequence identity as well as phylogenetic analysis, NSP3 from group A rotaviruses falls into three evolutionarily related groups, i.e., the SA11 group, the Wa group, and the S2 group. The SA11/SA114F gene appears to have a distant ancestral origin from that of the others and codes for a polypeptide of 315 amino acids (aa) in length. NSP3 from all other group A rotaviruses is only 313 aa in length because of a 2-amino-acid deletion near the carboxy-terminus. While the SA114F gene has the longest 3' untranslated region (UTR) of 132 nucleotides, that from other strains suffered deletions of varying lengths at two positions downstream of the translational termination codon. In spite of the divergence of the nucleotide (nt) sequence in the protein coding region, a stretch of about 80 nt in the 3' UTR is highly conserved in the NSP3 gene from all the strains. This conserved sequence in the 3' UTR might play an important role in the regulation of expression of the NSP3 gene.

Both surface proteins (VP4 and VP7) of an asymptomatic neonatal rotavirus strain (I321) have high levels of sequence identity with the homologous proteins of a serotype 10 bovine rotavirus.

The nucleotide sequence of genes 4 and 9, encoding the outer capsid proteins VP4 and VP7 of a serotype 10 tissue culture-adapted strain, I321, representative of asymptomatic neonatal rotaviruses isolated from neonates in Bangalore, India, were determined. Comparison of nucleotide and deduced amino acid sequences of I321 VP4 and VP7 with previously published sequences of various serotypes revealed that both genes were highly homologous to the respective genes of serotype 10 bovine rotavirus, B223. The VP4 of I321 represents a new human P serotype and the I321 and related strains represent the first description of neonatal rotaviruses that appear to derive both surface proteins from an animal rotavirus.

Serotypic and genotypic characterization of human serotype 10 rotaviruses from asymptomatic neonates.

Human rotaviruses were isolated from asymptomatic neonates at various hospitals and clinics in the city of Bangalore, India, and were found to be subgroup I specific and possess long RNA patterns (M. Sukumaran, K. Gowda, P. P. Maiya, T. P. Srinivas, M. S. Kumar, S. Aijaz, R. R. Reddy, L. Padilla, H. B. Greenberg, and C. D. Rao, Arch. Virol. 126:239-251, 1992). Three of these strains were adapted to tissue culture and found by serotype analysis and neutralization assays to be of serotype 10, a serotype commonly found in cattle but infrequently found in humans and not previously identified in neonates. By RNA-RNA hybridization, a high level of relatedness to a serotype 10 bovine rotavirus strain and a low-to-medium level of relatedness to a human rotavirus strain were observed. Since this human isolate shares a genogroup with bovine rotavirus, it is likely that it originated by interspecies transmission. A human rotavirus strain isolated from asymptomatic neonates and similar to bovine rotavirus might represent a good vaccine candidate.

Exclusive asymptomatic neonatal infections by human rotavirus strains having subgroup I specificity and "long" RNA electropherotype.

A large number of stool specimens, of healthy newborn infants, collected from various hospitals and clinics in Bangalore City, India, have been examined for the presence of asymptomatic rotaviral excretion. Out of 370 samples analysed during a three year period from 1988 to 1991, 133 specimens (36%) were positive for rotavirus RNA. All these asymptomatic neonatal strains, without exception, showed "long" RNA pattern, but subgroup I specificity. Serotype analysis by ELISA or by hybridization with serotype-specific probes indicated that these strains probably represent a new serotype in newborn children. We find an exclusive association of human rotaviruses having "long" RNA pattern and subgroup I specificity with asymptomatic neonatal infections in contrast to the earlier observations of association of such unusual strains with acute gastroenteritis in young children.

Functional identification of regulatory elements within the promoter region of platelet-derived growth factor 2.

Human platelet-derived growth factor (PDGF) is composed of two polypeptide chains, PDGF-1 and PDGF-2, the human homolog of the v-sis oncogene. Deregulation of PDGF-2 expression can confer a growth advantage to cells possessing the cognate receptor and, thus, may contribute to the malignant phenotype. We investigated the regulation of PDGF-2 mRNA expression during megakaryocytic differentiation of K562 cells. Induction by 12-O-tetradecanoylphorbol-13-acetate (TPA) led to a greater than 200-fold increase in PDGF-2 transcript levels in these cells. Induction was dependent on protein synthesis and was not enhanced by cycloheximide exposure. In our initial investigation of the PDGF-2 promoter, a minimal promoter region, which included sequences extending only 42 base pairs upstream of the TATA signal, was found to be as efficient as 4 kilobase pairs upstream of the TATA signal in driving expression of a reporter gene in uninduced K562 cells. We also functionally identified different regulatory sequence elements of the PDGF-2 promoter in TPA-induced K562 cells. One region acted as a transcriptional silencer, while another region was necessary for maximal activity of the promoter in megakaryoblasts. This region was shown to bind nuclear factors and was the target for trans-activation in normal and tumor cells. In one tumor cell line, which expressed high PDGF-2 mRNA levels, the presence of the positive regulatory region resulted in a 30-fold increase in promoter activity. However, the ability of the minimal PDGF-2 promoter to drive reporter gene expression in uninduced K562 cells and normal fibroblasts, which contained no detectable PDGF-2 transcripts, implies the existence of other negative control mechanisms beyond the regulation of promoter activity.

The predicted DBL oncogene product defines a distinct class of transforming proteins.

The DBL transforming gene was originally identified by transfection of NIH 3T3 cells with DNA from a human B-cell lymphoma. This gene was found to have arisen as a result of recombination of the 3' portion of the DBL protooncogene coding sequences with an unrelated segment of human DNA. It encodes a cytoplasmic protein that is equally distributed between cytosol and crude membrane fractions. To further characterize this transforming gene, a biologically active cDNA clone of the DBL transforming gene mRNA was isolated. Analysis of the sequence of the DBL oncogene cDNA revealed a long open reading frame that encodes a hybrid protein whose first 50 amino acids (at least) derive from a complete exon of a different locus. No significant homology with known oncogenes or any known protein sequences was demonstrated. The computer analysis of the predicted DBL protein indicated it is highly hydrophilic with no hydrophobic domains characteristic of a membrane-spanning region or signal peptide. Thus, the DBL oncoprotein is distinct among known transforming gene products.

Primary structure of the human fgr proto-oncogene product p55c-fgr.

Normal human c-fgr cDNA clones were constructed by using normal peripheral blood mononuclear cell mRNA as a template. Nucleotide sequence analysis of two such clones revealed a 1,587-base-pair-long open reading frame which predicted the primary amino acid sequence of the c-fgr translational product. Homology of this protein with the v-fgr translational product stretched from codons 128 to 516, where 32 differences among 388 codons were observed. Sequence similarity with human c-src, c-yes, and fyn translational products began at amino acid position 76 of the predicted c-fgr protein and extended nearly to its C-terminus. In contrast, the stretch of 75 amino acids at the N-terminus demonstrated a greatly reduced degree of relatedness to these same proteins. To verify the deduced amino acid sequence, antibodies were prepared against peptides representing amino- and carboxy-terminal regions of the predicted c-fgr translational product. Both antibodies specifically recognized a 55-kilodalton protein expressed in COS-1 cells transfected with a c-fgr cDNA expression plasmid. Moreover, the same protein was immunoprecipitated from an Epstein-Barr virus-infected Burkitt's lymphoma cell line which expressed c-fgr mRNA but not in its uninfected fgr mRNA-negative counterpart. These findings identified the 55-kilodalton protein as the product of the human fgr protooncogene.

The 5' untranslated sequence of the c-sis/platelet-derived growth factor 2 transcript is a potent translational inhibitor.

c-sis/platelet-derived growth factor 2 (PDGF-2) is a prototype growth factor with transforming potential. The c-sis/PDGF-2 transcript contains a long 5' untranslated sequence (UTS) that is highly G.C rich. To examine the influence of this sequence on sis/PDGF-2 expression, we localized the c-sis/PDGF-2 promoter and used this promoter or the simian virus 40 early promoter to drive expression of the bacterial chloramphenicol acetyltransferase or sis/PDGF-2 gene. The 5' UTS of c-sis/PDGF-2 mRNA had no effect on RNA expression but was shown to exert a potent inhibitory effect on translation. By deletion analysis, we demonstrated that the 5' UTS inhibited protein expression by as much as 40-fold. The inhibitory effect was independent of reporter gene, cell type, or promoter used. A highly G.C-rich 140-base-pair sequence immediately preceding the c-sis/PDGF-2 initiation codon was shown to be nearly as effective as the entire 5' UTS in translational inhibition. Transfection analysis demonstrated that the 5' UTS significantly reduced the transforming efficiency of the sis/PDGF-2 gene as well. Thus, our findings raise the possibility that changes in regulation at the level of sis/PDGF-2 translation may play a role in development of the neoplastic phenotype.

Detection of PDGF-2 homodimers in human tumor cells.

The v-sis oncogene encodes a protein structurally and functionally related to human platelet-derived growth factor (PDGF). In the present studies, we show that the primary translational product of the human sis proto-oncogene is a 26-kd protein, p26c-sis. This product is processed to yield a disulfide-linked homodimer, p56c-sis, which is further processed to a 35,000-dalton dimer, p35c-sis. Like the v-sis gene product, the PDGF-2 precursor undergoes N-linked glycosylation, implying its processing through the endoplasmic reticulum. The PDGF-2 product was shown to possess functional properties of PDGF. Whereas lysates of control COS-1 cells lacked mitogenic activity, lysates of COS-1 cells transfected with a c-sis/PDGF-2 expression vector specifically stimulated DNA synthesis of quiescent fibroblasts. Moreover, this activity was completely inhibitable by PDGF antibody. Identical forms of the sis/PDGF-2 product were identified in human tumor cells that expressed c-sis/PDGF-2 transcripts. These proteins were shown to be specifically associated with the membrane component of the tumor cells and were not detectably secreted into the culture medium. These findings support the concept that expression of the sis/PDGF-2 product in human cells responsive to its proliferative actions can be an important step in the processes leading to malignancy.

Structure and sequence of the human c-sis/platelet-derived growth factor 2 (SIS/PDGF2) transcriptional unit.

The structure of the normal human c-sis/platelet-derived growth factor 2 (SIS/PDGF2) transcript was determined by a combination of cDNA cloning, nuclease S1 mapping, and primer extension. Nucleotide sequence analysis revealed that the 3373-nucleotide SIS/PDGF2 mRNA contained only a 723-base-pair (bp) coding sequence for the PDGF2 precursor polypeptide. The coding sequence was flanked by long 5' (1022 bp) and 3' (1625 bp) untranslated regions. The 5' noncoding region, as well as upstream flanking genomic sequences, contained clusters of specific short repeat sequences. A consensus transcriptional promoter sequence, TATAAA, was identified 24 bp upstream of the mRNA start site and an enhancer-like "TG element" was detected about 180 bp downstream from the site of polyadenylylation. These findings identify putative regulatory elements of the SIS/PDGF2 gene.

Role of genes for normal growth factors in human malignancy.

The human homologue of the v-sis oncogene encodes one chain of human platelet-derived growth factor (PDGF-2). Previous studies have shown that expression of the coding sequence for this growth factor induces malignant transformation of NIH/3T3 cells. We demonstrate the detection of sis/PDGF-2 products indistinguishable from functional PDGF-2 homodimers in human tumor cells. These findings support the concept that expression of the sis/PDGF-2 product in human cells responsive to its proliferative actions can be an important step in the processes leading to malignancy. Unlike sis/PDGF-2, which remains tightly cell-associated, another growth factor, termed transforming growth factor alpha (TGF alpha), is actively secreted. Expression vectors for the TGF alpha coding sequence failed to induce primary transformed foci upon transfection of NIH/3T3 cells despite high levels of TGF alpha synthesized by these cells. Moreover, transfectants selected for secretion of high levels of TGF alpha failed to form colonies on contact inhibited NIH/3T3 monolayers or to induce tumors upon inoculation of nude mice. Thus, the ability of a growth factor to induce autonomous in vitro or in vivo growth is dependent upon more than expression of cognate receptors by the cell in which it is synthesized.

Complete sequence of bluetongue virus L2 RNA that codes for the antigen recognized by neutralizing antibodies.

The complete sequence of the RNA which encodes the major outer-shell-neutralizing antigen (VP2) of bluetongue virus serotype 10 was determined from overlapping cDNA clones inserted into pBR322. The segment L2 RNA was 2,926 base pairs long (1.87 X 10(6) daltons) and had, in one strand, an open reading frame capable of coding for a protein that had a calculated size of 111,122 daltons (956 amino acids) and a +11.5 net charge. The coding strands of both the L2 gene and the group-specific L3 gene of bluetongue virus serotype 17 (M. Purdy, J. Petre, and P. Roy, J. Virol. 51:754-759, 1984) had common sequences of some six nucleotides at their 5' termini (namely, GUUAAA...) and eight nucleotides (namely, ...ACACUUAC) at their 3' termini. Both had short 5' noncoding regions with AUG codons at residues 20 to 22 (L2) and 18 to 20 (L3). The sequences flanking these AUG codons were similar (A/GCCAUGG). The 3' noncoding regions were longer (36 nucleotides for L2, 49 nucleotides for L3). The predicted amino acid sequence of the L2, compared with the similarly sized L3 gene product, was rich in cysteine residues and charged amino acids.