Requirements for guinea pig cytomegalovirus tropism and antibody neutralization on placental amniotic sac cells

Abstract

Congenital cytomegalovirus (cCMV) is a leading cause of birth defects. The guinea pig is the only small cCMV animal model. Guinea pig cytomegalovirus (GPCMV) encodes similar glycoprotein complexes to human CMV (HCMV) including gB and the gH-based pentamer complex (PC). In HCMV, both gB and PC are neutralizing antibody antigens. The relevance of GPCMV PC for virus tropism and vaccine target remains controversial. A novel guinea pig placental amniotic sac epithelial (GPASE) cell-line did not express viral cell receptor platelet derived growth factor receptor alpha (PDGFRA) and resulted in requirement for the PC for GPCMV infection unless PDGFRA was ectopically expressed. High titer anti-gB sera from a GPCMV gB vaccine study was evaluated for GPCMV neutralizing capability on GPASE cells in comparison to convalescent sera from GPCMV(PC+) or GPCMV(PC-) infected animals. Anti-gB sera neutralized fibroblast infection but was less effective compared to anti-GPCMV(PC-), which had antibodies to gH/gL. However, both anti-GPCMV(PC-) and anti-gB sera similarly had reduced neutralizing capability on GPASE and renal epithelial cells in comparison to anti-GPCMV(PC+) sera, which had additional antibodies to PC. Overall, results demonstrate the importance of the PC for GPCMV tropism to various cell types that lack PDGFRA expression and the limited ability of anti-gB sera to neutralize GPCMV on non-fibroblast cells despite the essential nature of gB glycoprotein.

Keywords: CMV; Guinea pig cytomegalovirus; amniotic sac; congenital infection; glycoprotein gB; pentamer complex; placenta.

Fig. 1.

Characterization of GPASE cells by immunostaining, Western blot and RT-PCR assays. (i) Guinea pig amniotic sac (GPASE) cell line was verified as epithelial by immunohistochemical staining for cytokeratin marker [22] compared to fibroblasts . Comparative cytokeratin staining was carried out in six well plates on GPASE cells (top, panels A and B) and fibroblasts (bottom, panels C and D). Individual bright field images of cells stained with mouse anti-pankeratin + anti-mouse IgG-HRP (Sigma) (A, C) or secondary only negative control (B, D). Magnification ×20. (ii) Western blot analysis of total cell lysate for cytokeratin. (a) Pan-keratin expression in various guinea pig cell lines. Lanes, cell lysates:(1) fibroblast (GPL); (2) renal epithelial (REPI); (3) trophoblast (TEPI); (4) GPASE. (b) β-actin Western blot cell lysate lane loading control for (a). (iii) RT-PCR for vimentin (a) or GAPDH expression (b) from RNA extracted from GPASE (lane 1); trophoblasts (lane 2); or no template control (lane 3). (iv) Detection of GPCMV IE2 protein by immunostaining. IE2 protein staining of amniotic sac cells (A-D) compared to fibroblast cells (E-H) were carried out in six well plates. Monolayers infected with wild-type GPCMV (moi 1 pfu cell^–1) (A, B, E, F) and mock infected monolayers (C, D, G, H) were immunostained with custom anti-IE2 (GPCMV) rabbit polyclonal antibody (Genscript) +anti-rabbit IgG-HRP (Sigma) or secondary antibody only (B, D, F, H) at 48 hpi. Individual bright field images are representative of multiple experiments (six per panel). Images taken at 20× magnification show nuclear IE2 protein staining in virus infected cells [38].

Fig. 2.

Comparative growth kinetics for GPCMV(PC+) and GPCMV(PC-) on amniotic sac (GPASE) and fibroblast (GPL) cells. (i) GPCMV(PC+) virus growth curve on GPL (blue square) or GPASE cells (purple circle). (ii) GPCMV(PC-) virus growth curve on GPL (blue diamond) or GPASE cells (purple triangle). All infections performed at moi=1pfu cell^–1. Infections were carried out in duplicate six well plates. Virus titer at specific days post infection (DPI) 1–6 determined by titration of time points on GPL cells and represented as pfu ml^–1 in log scale in the vertical axis. (iii)-(v) Evaluation of cell associated (CA), cell released (CR) virus and total virus at 3 days post infection for: (iii) GPCMV(PC+) on GPASE; (iv) GPCMV(PC+) GPL; (v) GPCMV(PC-) GPL.

Fig. 3.

Both PC and gB are required for GPCMV infection of GPASE cells. GFP tagged PC+/gO+/ gB negative GPCMV BAC mutant (GP129FRT/GP55Km) [22] transfected onto GPASE cells failed to produce infectious virus. (a) Single mutant GPCMV BAC DNA transfected cells identified by GFP reporter gene expression. (b) gB rescue (GP129FRT/GP55Km rescue). Mutant gB BAC (GP129FRT/GP55km) was co-transfected onto GPL cells with a GP55 rescue fragment to restore GPASE tropism. Virus spread detected by GFP reporter gene expression on GPASE cells. Images taken at day 14 days post transfection ×10 magnification.

Fig. 4.

Ectopic expression of guinea pig PDGFRA on amniotic cells enables PC negative virus infection (i) Western blot analysis of total cell lysate of GPL and GPASE for endogenous PDGFRA (a) or Beta-actin (b) expression. (ii) Western blot for transient plasmid expression of pgpPDGFRA in GPASE transfected cells with FLAG-tagged pgpPDGFRA transient expression plasmid. PDGFRA expression detected with mouse anti-FLAG primary antibody and secondary anti-mouse IgG HRP (a) or ß actin (b). (iii) Ectopic expression of PDGFRA and GPCMV (PC-) infection of GPASE cells. GPL (1) and GPASE (2) transfected with pgpPDGFRA expression plasmid or control vector (pmcherryC1 (Clontech) were infected with GPCMV(PC-) virus (moi 1 pfu cell^–1) 24 h post transfection. A control experiment was also carried out for GPCMV(PC+) virus on GPL and GPASE cells. At 4 dpi, all wells were harvested and the viral titers determined by titration of samples on GPL cells. Results shown are for GPCMV(PC-) viral titers for different cells and +/-transient PDGFRA expression plasmid and control GPCMV(PC+) infection. (iv) Co-localization of PDGFRA+(mCherry+) transfected cells and PC negative virus (GFP+) infected cells. a-c: Colocalization of cells pretransfected with pLVXPDGFRAIRESmCherry and subsequently infected with GFP tagged PC negative virus in six well plates (moi 1pfu cell^–1). d-e: Control monolayers pretransfected with pLVXmCherry and infected with GFP tagged virus. Images taken at 3 days post infection at 10× maginification.

Fig. 5.

Immune response to GPCMV(PC+), GPCMV(PC-) and gB determined by ELISA and neutralization assays. Anti-GPCMV antibody titers (i) and anti-gB antibody titers (ii) of hyperimmune sera pooled from animals inoculated with GPCMV(PC+) (black) or GPCMV(PC-) (blue) virus determine by ELISA [27] compared to guinea pig serum negative for GPCMV (green). (iii) Western blot detection of AdgB transduced cell lysate (a, Lane 1) and mock infected (lane 2) with GPCMV gB specific antibody [36]. Antibody response to AdgB detected by ELISA (b). Anti-gB ELISA of pooled sera from animals inoculated with AdgB virus (red) compared to guinea pig serum negative for GPCMV (green). Black dash line represents ELISA negative cut-off. (iv) Pooled sera from GPCMV(PC+) (black circle); GPCMV(PC-) (blue square); gB (red triangle) animals; or guinea pig negative for GPCMV (green upside down triangle) were tested for NA₅₀ neutralization against wild-type GPCMV on GPL, REPI and GPASE. *P<0.001, Student t-test.

Fig. 6.

Antibodies to gB can neutralize virus infection of GPASE cells ectopically expressing PDGFRA. GPASE cells in six well plates pre-transfected with PDGFRA expression plasmid pLVXPDGFRAIRESmCherry (a-f) or control mCherry vector (pLVXIRESmCherry) were infected at 24 h post transfection with GFP tagged PC negative virus (10⁵ pfu) pretreated with anti-gB sera (1:40 dilution) or control GPCMV seronegative sera (1:40 dilution). Fluorescent images of GPASE imaged at 3 days post infection for mCherry (a, d, g, j) or GFP (b, e, h, k). Merged images: c (a+b); f (d+e); i (g+h); l (j+k). Images at 4× magnification. Individual panels represent average field for approximately 20 fields. Control experiments also carried out on GPL fibroblasts (data not shown).