A spread-deficient cytomegalovirus for assessment of first-target cells in vaccination.

Human cytomegalovirus (HCMV) is a human pathogen that causes severe disease primarily in the immunocompromised or immunologically immature individual. To date, no vaccine is available. We describe use of a spread-deficient murine CMV (MCMV) as a novel approach for betaherpesvirus vaccination. To generate a spread-deficient MCMV, the conserved, essential gene M94 was deleted. Immunization with MCMV-DeltaM94 is apathogenic and protective against wild-type challenge even in highly susceptible IFNalphabetaR(-/-) mice. MCMV-DeltaM94 was able to induce a robust CD4(+) and CD8(+) T-cell response as well as a neutralizing antibody response comparable to that induced by wild-type infection. Endothelial cells were identified as activators of CD8(+) T cells in vivo. Thus, the vaccination with a spread-deficient betaherpesvirus is a safe and protective strategy and allows the linkage between cell tropism and immunogenicity. Furthermore, genomes of MCMV-DeltaM94 were present in lungs 12 months after infection, revealing first-target cells as sites of genome maintenance.

The m74 gene product of murine cytomegalovirus (MCMV) is a functional homolog of human CMV gO and determines the entry pathway of MCMV.

The glycoprotein gO (UL74) of human cytomegalovirus (HCMV) forms a complex with gH/gL. Virus mutants with a deletion of gO show a defect in secondary envelopment with the consequence that virus spread is restricted to a cell-associated pathway. Here we report that the positional homolog of HCMV gO, m74 of mouse CMV (MCMV), codes for a glycosylated protein which also forms a complex with gH (M75). m74 knockout mutants of MCMV show the same spread phenotype as gO knockout mutants of HCMV, namely, a shift from supernatant-driven to cell-associated spread. We could show that this phenotype is due to a reduction of infectious virus particles in cell culture supernatants. m74 knockout mutants enter fibroblasts via an energy-dependent and pH-sensitive pathway, whereas in the presence of an intact m74 gene product, entry is neither energy dependent nor pH sensitive. This entry phenotype is shared by HCMV expressing or lacking gO. Our data indicate that the m74 and UL74 gene products both codetermine CMV spread and CMV entry into cells. We postulate that MCMV, like HCMV, expresses alternative gH/gL complexes which govern cell-to-cell spread of the virus.

Dominant-negative proteins in herpesviruses - from assigning gene function to intracellular immunization.

Investigating and assigning gene functions of herpesviruses is a process, which profits from consistent technical innovation. Cloning of bacterial artificial chromosomes encoding herpesvirus genomes permits nearly unlimited possibilities in the construction of genetically modified viruses. Targeted or randomized screening approaches allow rapid identification of essential viral proteins. Nevertheless, mapping of essential genes reveals only limited insight into function. The usage of dominant-negative (DN) proteins has been the tool of choice to dissect functions of proteins during the viral life cycle. DN proteins also facilitate the analysis of host-virus interactions. Finally, DNs serve as starting-point for design of new antiviral strategies.

Rational engineering of a fluorescein-binding anticalin for improved ligand affinity.

The anticalin FluA is an artificial lipocalin with novelspecificity for the fluorescein group, which was engineered from an insect bilin-binding protein by targeted random mutagenesis and selection. Based on the crystal structure of FluA, an attempt was made to improve the complementarity of its ligand pocket to fluorescein by rational protein design. Several side chains participating in sub-optimal interactions with the ligand were identified and replaced by residues that promised a better steric fit. As a result, the substitution of Ala45 by Ile and of Ser114 by Thr or Arg led to a tight affinity of ca. 1 nM, which is approximately 30-fold better than that of the parental anticalin. Similar to the original FluA, the improved version shows almost complete quenching of the bound ligand fluorescence. Interestingly, the quenching effect was significantly reduced when Trp129 was replaced by Tyr, thus supporting the previously postulated role of this residue, which closely packs against the bound ligand, for efficient electron transfer to the excited fluorescein. Circular dichroism spectra revealed that all variants investigated had retained the lipocalin fold. Corresponding thermal unfolding experiments confirmed similar folding stabilities, with melting temperatures ranging from 52.9 to 60.5 degrees C (i.e., for the high-affinity variant).