The immunogenicity of VP7, a rotavirus antigen resident in the endoplasmic reticulum, is enhanced by cell surface expression.

The glycoprotein VP7, the major serotype antigen of rotaviruses, is localized to the endoplasmic reticulum (ER) of the cell, where it is retained as a membrane-associated protein before assembly into mature virus particles. Wild-type VP7 expressed by a recombinant vaccinia virus was also located internally and was poorly antigenic. Using recombinant techniques, a correctly processed, secreted form of VP7 (S.C. Stirzaker and G.W. Both, Cell 56:741-747, 1989) was modified by addition to its C terminus of the membrane anchor and cytoplasmic domains from the influenza virus hemagglutinin. The hybrid protein was directed to the surface of cells, where it was anchored in the plasma membrane. When expressed in mice and rabbits by a recombinant vaccinia virus, the surface-anchored antigen stimulated a level of rotavirus-specific antibodies that was greater than 100-fold above the level induced by wild-type VP7. T-cell responses to the novel antigen were also elevated in comparison with the wild-type, intracellular protein. Cell surface anchoring may provide a strategy to increase the immunogenicity of intracellular antigens from other parasites and viruses.

Processing of rotavirus glycoprotein VP7: implications for the retention of the protein in the endoplasmic reticulum.

Rotaviruses are icosahedral particles that assemble in the lumen of the endoplasmic reticulum (ER). The viral glycoprotein, VP7, is also directed into this compartment and is retained for assembly onto the surface of viral cores. VP7 is therefore a resident ER glycoprotein with a luminal orientation. The VP7 gene possesses two potential in-frame initiation codons, each preceding a hydrophobic domain. Mature VP7 is derived from a precursor by cleavage but the site of cleavage has not been determined because viral VP7 has a blocked amino terminus. Using site-directed mutagenesis of the gene and in vitro transcription and translation systems, we have investigated the synthesis and processing of the primary products synthesized from each initiation codon. Proteins translated from either codon were processed in vitro to yield products indistinguishable in size. The primary translation products therefore appeared to be cleaved at the same site. The site was located empirically between Ala50 and Gln51 and mutation of the gene to convert Ala50----Val prevented processing. Amino-terminal sequence analyses of proteins synthesized in vitro, and characterization of an amino-terminal fragment of VP7 purified from virus unequivocally established Gln51 as the amino-terminal residue. Pyroglutamic acid was tentatively identified as the blocking group. Processing of VP7 therefore removes both amino-terminal hydrophobic domains from the protein. Some other mechanism not requiring the presence of these hydrophobic sequences must account for the retention of this novel glycoprotein in the ER.

Location of sequences within rotavirus SA11 glycoprotein VP7 which direct it to the endoplasmic reticulum.

The Simian 11 rotavirus glycoprotein VP7 is directed to the endoplasmic reticulum (ER) of the cell and retained as an integral membrane protein. The gene coding for VP7 predicts two potential initiation codons, each of which precedes a hydrophobic region of amino acids (H1 and H2) with the characteristics of a signal peptide. Using the techniques of gene mutagenesis and expression, we have determined that either hydrophobic domain alone can direct VP7 to the ER. A protein lacking both hydrophobic regions was not transported to the ER. Some polypeptides were directed across the ER membrane and then into the secretory pathway of the cell. For a variant retaining only the H1 domain, secretion was cleavage dependent, since an amino acid change which prevented cleavage also stopped secretion. However, secretion of two other deletion mutants lacking H1 and expressing truncated H2 domains was unaffected by this mutation, suggesting that these proteins were secreted without cleavage of their NH2-terminal hydrophobic regions or secreted after cleavage at a site(s) not predicted by current knowledge.

Vaccinia virus recombinants expressing the SA11 rotavirus VP7 glycoprotein gene induce serotype-specific neutralizing antibodies.

A cDNA copy of the gene coding for the major outer neutralizing protein (VP7) of simian 11 rotavirus was incorporated into the vaccinia virus genome under the control of the vaccinia promoter (molecular weight, 7,500). A deletion mutant of this gene which codes for a secreted form of VP7 when expressed under the control of the simian virus 40 late promoter (M. S. Poruschynsky, C. Tyndall, G. W. Both, F. Sato, A. R. Bellamy, and P. H. Atkinson, J. Cell Biol. 101:2199-2209, 1985) was also inserted. Each recombinant vaccinia virus directed the synthesis of a rotavirus protein in infected cells, and the product encoded by the mutated gene was secreted. Rabbits immunized with the two types of recombinant vaccinia virus generated antibodies that were able both to recognize simian 11 rotavirus in an enzyme-linked immunosorbent assay and to neutralize the virus in a plaque-reduction test. Antibodies induced by the recombinant vaccinia viruses expressing either form of VP7 were serotype specific.

Vasopressin-neurophysin II gene expression in the ovary: studies in Sprague-Dawley, Long-Evans and Brattleboro rats.

The neurohypophysial hormones oxytocin and arginine vasopressin (AVP) have been identified on immunological criteria in the ovary. Confirmation of extraneuronal synthesis requires the demonstration in the tissue of the specific messenger RNA (mRNA) for the preprohormone. Using a synthetic pentadecamer nucleotide probe, highly specific for the 5' region of rat neurophysin II (NP II), we have demonstrated the presence of AVP-NP II mRNA in the ovary of Sprague-Dawley, Long-Evans and Brattleboro rats, with an apparent molecular weight identical to that seen for hypothalamus. These findings, together with the presence of immunoreactive AVP in the ovaries but not hypothalami of Brattleboro rats, suggest that tissue-specific differences in AVP-NP II gene expression occur at the translational as well as transcriptional level.

Kallikrein gene expression in the rat anterior pituitary.

The report of 'kallikrein-like' activity in the rat neuro-intermediate lobe (N-IL) and its possible involvement in pro-opiomelanocortin processing led us to explore the expression of the kallikrein gene(s) in the pituitary. Using 32P-labelled rat pancreatic kallikrein cDNA, we have shown positive hybridization for rat anterior pituitary poly(A)+ RNA, of identical size on Northern blots (approximately 1.0 kb) to rat kidney poly(A)+ RNA run in parallel. Prior adrenalectomy or ovariectomy decreased the level of kallikrein mRNA seen in the anterior pituitary; total RNA from rat N-IL showed no significant hybridization. On hybridization histochemistry the anterior pituitary was strongly positive, and the neural and intermediate lobes negative. The previously reported kallikrein-like activity in the N-IL is therefore probably due to a non-kallikrein kininogenase; in the anterior pituitary, kallikrein may have a physiological role in limited precursor proteolysis, but lack kininogen activity.

The human pro-opiomelanocortin gene: organization, sequence, and interspersion with repetitive DNA.

The human pro-opiomelanocortin (POMC) gene has been characterized by molecular cloning and DNA sequence analysis. Although this gene codes for several different polypeptide hormones, only a single intron interrupts the protein coding region. The DNA in this intron, and in a second intron found in the region of the gene homologous to the mRNA 5'-untranslated sequence, contains repetitive DNA sequences. At least some of these sequences belong to the Alu family of transcribed middle repetitive DNA. The determination of the complete nucleotide sequence of the coding regions of the gene demonstrates that the pattern of homologous and variable regions seen in the POMC protein between different species is reflected at the DNA level. DNA sequences encoding the highly conserved regions of POMC are 90-95% homologous between species while the coding sequences for the variable regions of the protein are approximately 70% homologous. The very high degree of homology in the amino terminal portion of POMC is consistent with an important physiological role for this peptide.