Induction of type I interferon secretion through recombinant Newcastle disease virus expressing measles virus hemagglutinin stimulates antibody secretion in the presence of maternal antibodies

Abstract

Measles virus (MV) vaccine effectively protects seronegative individuals against infection. However, inhibition of vaccine-induced seroconversion by maternal antibodies after vaccination remains a problem, as it leaves infants susceptible to MV infection. In cotton rats, passive transfer of MV-specific IgG mimics maternal antibodies and inhibits vaccine-induced seroconversion. Here, we report that immunization in the presence of passively transferred IgG inhibits the secretion of neutralizing antibodies but not the generation of MV-specific B cells. This finding suggested that MV-specific B cells require an additional stimulus to mature into antibody-secreting plasma cells. In order to provide such a stimulus, we generated a recombinant Newcastle disease virus (NDV) expressing the MV hemagglutinin (NDV-H). In contrast to MV, NDV-H induced high levels of type I interferon in plasmacytoid dendritic cells and in lung tissue. In cotton rats immunized with NDV-H, neutralizing antibodies were also generated in the presence of passively transferred antibodies. In the latter case, however, the level and kinetics of antibody generation were reduced. In vitro, alpha interferon stimulated the activation of MV-specific B cells from MV-immune spleen cells. NDV infection (which induces alpha interferon) had the same effect, and stimulation could be abrogated by antibodies neutralizing alpha interferon, but not interleukin 6 (IL-6). In vivo, coapplication of UV-inactivated MV with NDV led to increased MV-specific antibody production in the presence and absence of passively transferred antibodies. These data indicate that MV-specific B cells are being generated after immunization in the presence of maternal antibodies and that the provision of alpha interferon as an additional signal leads to antibody secretion.

FIG. 1.

Construction and characterization of a recombinant NDV vector expressing MV H. (A) Schematic representation of the NDV-H cDNA construct. The hemagglutinin gene from the MV Schwarz vaccine strain (MV H) was inserted between the P and M genes of the molecular clone of rNDV/F3aa as described in Materials and Methods. (B) Expression of MV H protein in infected cells. Vero cells mock infected or infected with NDV-H were harvested 48 h postinfection, and the cell lysates were analyzed for MV H protein expression by Western blotting. (C) MV H expression by immunostaining of infected Vero cells. Vero cells were infected with NDV-H or measles virus; 24 h after infection, the cells were fixed and stained with anti-NDV polyclonal antibodies (red) and MV H monoclonal antibodies (green). DAPI staining revealed nuclear chromatin (blue). The arrow indicates a measles virus-infected syncytium. (D) Measles virus hemagglutinin is incorporated into the NDV virion. Wild-type NDV or recombinant viruses expressing RFP, MV H, or influenza A virus hemagglutinin were purified once or twice over a sucrose gradient. Virus proteins were separated by SDS-PAGE and stained with NDV- and MV-specific antisera (not shown) or Coomassie blue. The positions of the NDV hemagglutinin/neuraminidase (HN), nucleoprotein (NP), fusion protein (F1), polymerase (P), and matrix protein (M) are indicated on the right. Measles virus hemagglutinin is indicated by an asterisk. (E) MV H-specific antibodies block NDV-H expression. NDV-RFP and NDV-H were preincubated with neutralizing antibodies specific for MV H at different neutralizing titers and used to infect Vero cells. Virus-infected cells were visualized 48 h later by immunofluorescence.

FIG. 2.

NDV-H is a potent inducer of type I interferon in the cotton rat lung. An IFN-α/β bioassay was performed with BAL fluid to determine the levels of induction at days 1, 2, 3, 4, and 7 after intranasal inoculations with 10⁵ PFU NDV-H or MV Schwarz. Differences at days 1, 2, and 3 were statistically significant (analysis of variance [ANOVA]). The results reflect the means of four animals per group plus standard deviations (SD).

FIG. 3.

Comparison of the generation of MV-specific B cells and the induction of neutralizing antibodies. Cotton rats were immunized i.n. with 5 × 10⁵ PFU of MV Schwarz (A and B) or NDV-H (C and D) in the presence or absence of human MV-specific IgG (NT, 320). From serum, neutralizing antibody titers were determined (A and C), and from spleen cells, B-cell antibody-forming cells (AFC) were determined (B and D) in cotton rats immunized with MV Schwarz (A and B) and NDV-H (C and D). The results reflect the means of four animals per group plus SD. The difference in induction of neutralizing antibodies in the presence of human MV-specific IgG between NDV-H and MV is statistically significant (P < 0.002).

FIG. 4.

Antibody generation is delayed in cotton rats immunized in the presence of MV-specific IgG. Cotton rats were immunized i.n. with 5 × 10⁵ PFU NDV-H in the presence and absence of human MV-specific IgG (NT, 320). Human antibodies declined over time (as measured by ELISA) (open triangles). In cotton rats immunized with NDV-H in the absence of human MV-specific IgG, MV-specific cotton rat antibodies were found as early as 1 week after infection, with an increase in antibody titers over time (open diamonds). In animals immunized in the presence of passively transferred MV-specific antibody, the development of MV-specific cotton rat antibodies was delayed by 2 weeks (closed squares). All time points reflect the means of four animals per group ± SD. The dashed line indicates the threshold of the assay (optical density [O.D.] = 0.14).

FIG. 5.

IFN-α induced by NDV infection stimulates MV-specific B-cell responses. Bone marrow cells from MV-immune cotton rats were stimulated with gradient-purified, UV-inactivated MV antigen. In addition, cotton rat IFN-α, NDV-GFP, and neutralizing sera against IFN-α or IL-6 were added, and 24 h later, the plates were developed and the MV-specific B cells were counted. After the addition of neutralizing serum against IFN-α to NDV-GFP-infected bone marrow cells, the number of MV-specific B cells was significantly reduced compared to NDV-GFP infection only and was not statistically significantly different from stimulation with MV antigen alone. The error bars indicate SD.

FIG. 6.

Coinfection with NDV increases neutralizing antibody response against MV antigen. Cotton rats were inoculated intranasally with gradient-purified, UV-inactivated MV antigen alone or together with NDV-GFP in the absence or presence of human MV-specific IgG (NT, 100). After 2, 3, and 4 weeks, serum was obtained and tested for the presence of neutralizing antibodies. The difference in neutralizing antibody titers after immunization in the presence of passively transferred MV-specific IgG was statistically significant (P < 0.003; t test). The error bars indicate SD.