Immune modulating effect by a phosphoprotein-deleted rabies virus vaccine vector expressing two copies of the rabies virus glycoprotein gene

Abstract

The type of immune response induced by a vaccine is a critical factor that determines its effectiveness in preventing infection or disease. Inactivated and live rabies virus (RV) vaccine strains elicit an IgG1-biased and IgG1/IgG2a-balanced antibody response, respectively. However, IgG2a antibodies are potent inducers of anti-viral effector functions, and therefore, a viral vaccine vector that can elicit an IgG2a-biased antibody response may be more effective against RV infection. Here we describe the humoral immune response of a live replication-deficient phosphoprotein (P)-deleted RV vector (SPBN-DeltaP), or a recombinant P-deleted virus that expresses two copies of the RV glycoprotein (G) gene (SPBN-DeltaP-RVG), and compare it to a UV-inactivated RV. Mice inoculated with UV-inactivated RV induced predominantly an IgG1-specific antibody response, while live recombinant SPBN-DeltaP exhibited a mixed IgG1/IgG2a antibody response, which is consistent with the isotype profiles from the replication-competent parental viruses. Survivorship in mice after pathogenic RV challenge indicates a 10-fold higher efficiency of live SPBN-DeltaP compared to UV-inactivated SPBN-DeltaP. In addition, SPBN-DeltaP-RVG induced a more rapid and robust IgG2a response that protected mice more effectively than SPBN-DeltaP. Of note, 10(3)ffu of SPBN-DeltaP-RVG-induced anti-RV antibodies that were 100% protective in mice against pathogenic RV challenge. The increased immune response was directed not only against RV G but also against the ribonucleoprotein (RNP), indicating that the expression of two RV G genes from SPBN-DeltaP-RVG enhances the immune response to other RV antigens as well. In addition, Rag2 mice inoculated intramuscularly with 10(5)ffu/mouse of SPBN-DeltaP showed no clinical signs of rabies, and no viral RNA was detected in the spinal cord or brain of inoculated mice. Therefore, the safety of the P-deleted vectors along with the onset and magnitude of the IgG2a-induced immune response by SPBN-DeltaP-RVG indicate that this vector holds great promise as either a therapeutic or preventative vaccine against RV or other infectious diseases.

Figure 1

Construction of SPBN-ΔP. The P gene was deleted from SPBN (28) (top), which is based on the attenuated SAD-B19 RV vaccine strain, using a PCR strategy and standard cloning techniques. Infectious virus was recovered and propagated on BSR cells stably expressing RV P, as described in Materials and Methods

Figure 2

Live SPBN-ΔP induces high levels of anti-RV G and anti-RNP antibodies, and protects mice more effectively against pathogenic RV challenge than UV-inactivated RV. Groups of 10 Balb/c mice were immunized intramuscularly (i.m.) with the indicated doses of live or UV-inactivated SPBN-ΔP and sera from mice 14 days post-immunization were tested for total IgG antibodies against RV G or RNP by ELISA at a 1:50 sera dilution (A), or virus neutralizing antibodies (VNA) expressed in international units/ml (IU/ml) (B). The data represents the average and standard error of the mean (SEM) of n=10 completed in duplicate and analyzed using the unpaired t-test, 95% confidence (see the text for p values). Four weeks post-immunization, the mice were challenged i.m. with 100 LD₅₀ of pathogenic Challenge Virus Strain (CVS)-N2c and observed for three weeks and their survivorship recorded (C). The ED₅₀ was calculated using the survivorship rates between the two groups of immunized animals (D) and indicates a 10-fold higher efficiency of live SPBN-ΔP versus UV-inactivated SPBN-ΔP. (***p<0.001; **p=0.001-0.01; *p=0.01-0.05)

Figure 3

Live SPBN-ΔP induces a mixed IgG1/IgG2a antibody titer compared with an IgG1-biased response from UV-inactivated RV. Fourteen days post-immunization, sera from mice that received either 10⁴ (A) or 10⁶ ffu (B) of live SPBN-ΔP or UV-inactivated RV, as described in Figure 2, were tested by ELISA for anti-RV G (left) or anti-RNP (right) antibodies at a 1:50 dilution. The data shows an increase in IgG2a-specific antibody titers against both RV G and RNP at both doses used of live SPBN-ΔP compared with UV-inactivated RV. The data represents the average and SEM of n=10 completed in duplicate and analyzed using the unpaired t-test, 95% confidence (p values provided in the text; ***p<0.001; **p=0.001-0.01; *p=0.01-0.05).

Figure 4

Construction of SPBN-ΔP-RVG. Two copies of the RV G gene were digested from SPBN-RVG (top) using the two unique restriction sites, XmaI and NheI. The two RV G genes were inserted into SPBN-ΔP (middle) also digested with XmaI and NheI, resulting in SPBN-ΔP-RVG (bottom).

Figure 5

SPBN-ΔP-RVG induces rapid and potent anti-RV G and anti-RNP antibodies. Groups of five Balb/c mice were immunized i.m. with 10⁵, 10^4,, or 10³ ffu of SPBN-ΔP or SPBN-ΔP-RVG and sera was collected on days 5, 7, and 14 post-immunization (Experiment #1). The experiment was repeated with an additional five mice, except in this experiment, sera were collected on days 5, 7, 14 and 28 post-immunization. At each time point, sera were tested for total IgG by ELISA for anti-RV G (A) or anti-RV RNP (B) antibodies. The data for days 5, 7, and 14 represent the average and SEM for both Experiments #1 and #2 (n=10) completed in duplicate and for day 28 from Experiment #2 (n=5) also completed in duplicate. The data were analyzed using the unpaired t-test, 95% confidence (see the text for p values; ***p<0.001; **p=0.001-0.01; *p=0.01-0.05).

Figure 6

SPBN-ΔP-RVG induces strong virus neutralization antibody titers and provides protection against pathogenic RV challenge with as little as 10³ ffu/mouse. The mice described in Figure 5 were challenged i.m. 4 (Experiment #1) or 6 (Experiment #2) weeks post-immunization with 100 LD₅₀ pathogenic CVS-N2c. Pre-challenge virus neutralizing antibody titers, expressed in international units/ml, are shown (A), and survivorship was recorded (B). Virus neutralizing antibodies were also determined from all mice on day 5 post-challenge and from mice that survived at day 35 post-challenge.

Figure 7

SPBN-ΔP-RVG induces an IgG2a-biased antibody response, which is amplified post-challenge. Sera from mice collected 14 days post-immunization (A) or 5 days post-challenge (B) were analyzed by ELISA for IgG1- or IgG2a-specific antibodies against RV G at a sera dilution of 1:50. The average and SEM antibody results from each group of mice (n=5) (A and B) or from individual mice (C and D) are shown. Data were analyzed using the unpaired t-test, 95% confidence (see the text for p values; (***p<0.001; **p=0.001-0.01; *p=0.01-0.05).

Figure 8

Input RV P is sufficient for cell surface expression of RV G. NA cells were infected with SPBN (dotted line), SPBN-ΔP (solid line) or SPBN-ΔP-RVG (dashed line) at an moi of 3, or left uninfected (solid histogram). At 24 (top) and 48 h p.i. (bottom), cells were incubated with a polyclonal rabbit anti-RV G antiserum, followed by a Cy-2-conjugated anti-rabbit antibody. Surface expression was determined by flow cytometry, and fluorescence intensity was plotted.

Figure 9

SPBN-ΔP does not spread to the spinal cord or brain of infected RAG2 KO mice. Groups of 5 RAG2 KO mice were inoculated i.m. with 10⁵ ffu of either SPBN (A), SPBN-333 (B) or SPBN-ΔP (C) and their weights were recorded (left panel). Once one or more animal(s) in a group became moribund due to rabies infection, all mice in that group were euthanized. Mice inoculated with the highly attenuated SPBN-333 vector served as a guide for the time point to collect samples from mice inoculated with SPBN-ΔP since it was not known whether SPBN-ΔP would induce visible signs of rabies infection. Total RNA was isolated from muscle, brain and spinal cord and viral RT-PCR and quantitative real-time PCR was used to amplify and quantify viral genomic (middle panels) or messenger (right panels) RNA from individual samples.