Functional Characterization of Glycoprotein H Chimeras Composed of Conserved Domains of the Pseudorabies Virus and Herpes Simplex Virus 1 Homologs

Abstract

Membrane fusion is indispensable for entry of enveloped viruses into host cells. The conserved core fusion machinery of the Herpesviridae consists of glycoprotein B (gB) and the gH/gL complex. Recently, crystal structures of gH/gL of herpes simplex virus 2 (HSV-2) and Epstein-Barr virus and of a core fragment of pseudorabies virus (PrV) gH identified four structurally conserved gH domains. To investigate functional conservation, chimeric genes encoding combinations of individual domains of PrV and herpes simplex virus 1 (HSV-1) gH were expressed in rabbit kidney cells, and their processing and transport to the cell surface, as well as activity in fusion assays including gB, gD, and gL of PrV or HSV-1, were analyzed. Chimeric gH containing domain I of HSV-1 and domains II to IV of PrV exhibited limited fusion activity in the presence of PrV gB and gD and HSV-1 gL, but not of PrV gL. More strikingly, chimeric gH consisting of PrV domains I to III and HSV-1 domain IV exhibited considerable fusion activity together with PrV gB, gD, and gL. Replacing PrV gB with the HSV-1 protein significantly enhanced this activity. A cell line stably expressing this chimeric gH supported replication of gH-deleted PrV. Our results confirm the specificity of domain I for gL binding, demonstrate functional conservation of domain IV in two alphaherpesviruses from different genera, and indicate species-specific interactions of this domain with gB. They also suggest that gH domains II and III might form a structural and functional unit which does not tolerate major substitutions.

Importance: Envelope glycoprotein H (gH) is essential for herpesvirus-induced membrane fusion, which is required for host cell entry and viral spread. Although gH is structurally conserved within the Herpesviridae, its precise role and its interactions with other components of the viral fusion machinery are not fully understood. Chimeric proteins containing domains of gH proteins from different herpesviruses can serve as tools to elucidate the molecular basis of gH function. The present study shows that the C-terminal part of human herpesvirus 1 (herpes simplex virus 1) gH can functionally substitute for the corresponding part of suid herpesvirus 1 (pseudorabies virus) gH, whereas other tested combinations proved to be nonfunctional. Interestingly, the exchangeable fragment included the membrane-proximal end of the gH ectodomain (domain IV), which is most conserved in sequence and structure and might be capable of transient membrane interaction during fusion.

FIG 1

Structures of HSV-2 gH/gL (PDB accession number 3M1C) (A, C, and E), and of PrV gH^c (PDB accession number 2XQY) (B, D, and F). Amino and carboxyl termini of gH are indicated (N and C, respectively). gL is shown in pink in the HSV-2 structure, and gH domains are labeled with roman numerals and highlighted in gray (I), blue (II), yellow (III), and green (IV). Domain I is absent from the PrV gH^c fragment. The complete structures (A and B) and enlargements of domain II (C and D) are shown in the same orientation, whereas the enlargements of domains III and IV (E and F) were rotated ≈90° counterclockwise. The conserved syntaxin-like bundle of α-helices (SLB) and the β-sheet fence in domain II, as well as the flap in domain IV, are marked by gray arrows. The strands of the fence are numbered to serve as landmarks. Conserved disulfide bonds (SS) in domain IV and at the end of domain II are labeled. Red spheres marked by black arrows and numbers represent individual atoms of the indicated junction residues used for generation of the chimeric proteins (Fig. 2).

FIG 2

Diagram of the generated gH chimeras. The ectodomains of PrV and HSV-1 gH are shown in different shades of green and red, respectively. SP, signal peptide; C-tail, cytoplasmic tail; TM, transmembrane domain. Structurally conserved subdomains are indicated by roman numerals (I to IV). The last amino acids of the determined or predicted (Geneious) structural elements of PrV gH (GenBank accession number AAA47466) and HSV-1 gH (GenBank accession number AJE60179), as well as the last and first amino acids of the two components of chimeric proteins, are given above the respective bars. For clarity, the proteins were not drawn to scale.

FIG 3

Western blot analyses of transfected RK13 cells. At 48 h after transfection with expression plasmids for the indicated gH variants, cell lysates were prepared and separated by SDS-PAGE. Blots were incubated with monospecific rabbit antiserum against PrV gH or the N-terminal and C-terminal parts of HSV-1 gH. The probable primary translation products (asterisks), as well as immature (black arrows) and mature (gray arrows) glycosylated forms of native or chimeric gH, are labeled. Molecular masses of marker proteins are indicated on the left.

FIG 4

IIF analyses of transfected RK13 cells. At 48 h after transfection with expression plasmids for the indicated native or chimeric gH and, optionally, gL of PrV (P) or HSV-1 (H), the cells were fixed with 3% paraformaldehyde for detection of surface proteins or permeabilized with 0.5% Triton X-100 for determination of total protein. In panel A surface-located gH was detected using a monospecific antiserum against the PrV protein (left panels) or the N-terminal part of HSV-1 gH (right panels) and Alexa Fluor 488-conjugated secondary antibodies. Chromatin was counterstained with propidium iodide. Scale bar, 100 μm. In panel B gH-specific fluorescence intensities were compared between nonpermeabilized and permeabilized cells to evaluate the proportions of surface-exposed gH. Shown are the mean percentages and negative standard deviations from at least three independent experiments. Statistically significant (P < 0.05) differences to surface detection of PrV gH in the presence (a) and HSV-1 gH in the absence (b) of matching gL are indicated.

FIG 5

In vitro fusion assays. RK13 cells were cotransfected with different combinations of expression plasmids for GFP, C-terminally truncated gB, gD and gL of PrV (P) or HSV-1 (H), and native or chimeric gH as indicated. If expression plasmids for individual glycoproteins were omitted (X), they were replaced by the same DNA amount of the empty expression vector. After different times syncytia were detected and measured by fluorescence microscopy. (A) For all tested protein combinations, statistical significance (P < 0.01) of size differences to a negative control (PrV glycoprotein set without gB) was determined, and positive results were roughly quantified (+ to ++++). (B and C) Mean areas of syncytia induced by the fusogenic protein combinations or in the absence of gH were compared after 24 h (B) or 48 h (C) to the value obtained with the complete PrV set of proteins, which was set to 100%. All shown combinations formed significantly (P < 0.001) larger (a) or smaller (b) syncytia than the PrV positive control. Negative standard deviation is also indicated.

FIG 6

trans-Complementation of gH-deleted PrV. RK13, RK13-PgH, and RK13-P^I–III/H^IVgH cells were infected with phenotypically complemented gH-negative pPrV-ΔgHABF (ΔgH) or the parental gH-positive virus pPrV-ΔgGG (wild type, WT). (A) After 2 days (pPrV-ΔgGG on RK13) or 3 days (others) at 37°C under plaque assay conditions, autofluorescence of virus-expressed GFP was analyzed. Scale bar, 500 μm. (B) Mean areas of 50 plaques per virus and cell were calculated. To facilitate comparison, plaque sizes of pPrV-ΔgGG were set to 100% on each cell line. Negative standard deviations are indicated. (C) Mean progeny virus titers and standard deviations were determined 4 days after infection at an MOI of 0.01 in three independent experiments. Replication properties significantly (P < 0.05) different from those of pPrV-ΔgHABF on RK13 cells (a), pPrV-ΔgHABF on RK13-PgH cells (b), or wild-type PrV on the same cell line (c) are indicated.