Kinetics of the Development of Protective Immunity in Mice Vaccinated with a Live Attenuated Retrovirus

Abstract

Vaccination of mice with a live attenuated vaccine virus induces potent protection against subsequent challenge with pathogenic Friend retroviral complex. The kinetic studies presented here demonstrate protection from acute splenomegaly as early as 1 week postvaccination. At this time point virus-specific cytotoxic T lymphocytes (CTL) were demonstrable in direct chromium release assays. However, during the first 2 weeks after vaccination protection was incomplete since the mice were not protected against establishment of low-level persistent infections in the spleen. By 3 weeks postvaccination the animals were protected against the establishment of persistent virus as well as acute splenomegaly. The timing of this complete protection correlated with the presence of both virus-neutralizing antibodies and primed CTL in the immunized mice. Within 3 days of virus challenge, vaccinated mice showed high levels of activated B cells and CD4⁺ and CD8⁺ T cells, indicating an efficient priming of all lymphocyte subsets. Despite very limited replication of the vaccine virus, the protective effect was long lived and was still present 6 months after immunization.

The use of live attenuated viruses as vaccines against viral diseases dates as far back as 1796, when Edward Jenner successfully used inoculations of cowpox virus to prevent deadly disease from the related smallpox virus. Live attenuated viruses still comprise the bulk of modern day viral vaccines, and they are regarded as the most effective experimental vaccines in the simian immunodeficiency virus (SIV) model for AIDS (23). However, there are continuing worries and mounting evidence indicating that live attenuated retroviruses are unsafe for use in humans (1, 12). While live attenuated retroviruses may never be used as vaccines in humans, it is important to study the attributes which contribute to their efficacy. Understanding the protective mechanisms of vaccine viruses in animal models can be very important for the rational design and testing of new vaccine strategies.

We have used the Friend virus (FV) model in mice to study the protection induced by vaccination with a live attenuated retrovirus (7, 9). FV was the first model in which vaccine protection by infection with a live attenuated retrovirus was described (27). It is an interesting and useful model because it is one of the few immunosuppressive retroviruses that causes disease in immunocompetent adult mice (for reviews, see references 3 and 19). Pathogenic FV is a retroviral complex comprised of two components. The first is a nonpathogenic replication-competent helper virus, Friend murine leukemia virus (F-MuLV), which contains the immunological determinants necessary for immunization (7). The second component is a replication-defective spleen focus-forming virus (SFFV), which is required for pathogenicity in adult mice (for a review, see reference 24). In susceptible strains of mice, FV induces rapid splenomegaly because SFFV defective envelope proteins bind to erythropoietin receptors on erythroid precursor cells, causing false proliferation signals (21, 26). Proliferation of these precursors expands the population of target cells for retroviral infection, and lethal erythroleukemia ensues within several weeks of initial infection (29, 31, 38).

In recent experiments we used the F-MuLV helper component as an attenuated vaccine because it replicates poorly in the absence of SFFV-induced proliferation (7). To further attenuate replication, we have used N-tropic F-MuLV in Fv-1^b/b genetically resistant mice. In our previous experiments this vaccine induced strong protection against acute disease (7) and also protected against the establishment of persistent FV infections (8). The protective effect of the vaccine virus was mediated by immune cells rather than by interference mechanisms (7, 9). Adoptive transfer experiments with lymphocyte subsets from F-MuLV-vaccinated mice showed that complex immune responses, including CD4⁺ and CD8⁺ T cells and B cells, were required for protection against pathogenic FV challenge (9). However, no prechallenge analysis of immunological responses has been done in previous experiments, and it was not known how the kinetics of vaccine virus replication were associated with the onset or duration of protective immunity. In the present studies we addressed these issues and also analyzed the immune activation of primed lymphocytes after FV challenge. The results indicate that broad priming of immunological memory rather than induction of persistent immunological effectors is key to long-lasting protection against retroviral infection.

MATERIALS AND METHODS

Mice.

(B10.A × A.BY)F₁ female mice of 3 to 6 months of age at experimental onset were used for all experiments except the long-term protection experiments where the challenge was done at 3 and 6 months postvaccination. In those experiments the more susceptible strain, (B10.A × A/Wy)F₁, was used. Parental mouse strains for breeding the F₁ mice were obtained from the Jackson Laboratories. Breeding of F₁ strains was done at Rocky Mountain Laboratories. All mice were treated in accordance with National Institutes of Health regulations and the guidelines of the Animal Care and Use Committee of Rocky Mountain Laboratories.

Virus and virus infections.

The B-tropic FV complex used in these experiments was from an uncloned virus stock obtained from 10% spleen cell homogenates from BALB/c mice infected 9 days previously with polycythemia-inducing FV stocks (4, 13). The N-tropic F-MuLV helper virus stock (2) was a 24-h supernatant from infected Mus dunni cells (25). F-MuLV vaccinations were done by intravenous injections of 4,000 focus-forming units (FFU) of virus in 0.5 ml of phosphate-buffered, balanced salt solution (PBBS) containing 2% fetal bovine serum. In virus challenge experiments, mice were injected intravenously with 0.5 ml of PBBS containing 2% fetal bovine serum and 10,000 spleen focus-forming units (SFFU) of Friend virus complex. Disease was monitored by palpation for splenomegaly in a blinded fashion as described elsewhere (18).

Virus-neutralizing antibody assays.

To test plasma samples for virus-neutralizing antibodies, heat-inactivated (56°C, 10 min) samples at titrated dilutions were incubated with an aliquot of F-MuLV virus stock in the presence of complement at 37°C as previously described (30). The samples were then analyzed by focal infectivity assays (36) on susceptible M. dunni cells pretreated with 4 μg of Polybrene per ml. The cultures were incubated for 4 days, fixed with ethanol, labeled first with F-MuLV envelope-specific monoclonal antibody (MAb) 720 (33), and then labeled with goat anti-mouse peroxidase-conjugated antiserum (Cappel, West Chester, Pa.). The titer was defined as the plasma dilution at which greater than 75% of the input virus was neutralized.

Infectious center assays.

Titrations of single cell suspensions from persistently infected mouse spleens were plated onto susceptible M. dunni cells, cocultivated for 5 days, fixed with ethanol, stained with F-MuLV envelope-specific MAb 720 (33), and developed with peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG) and substrate (Cappel) to detect foci.

T-cell depletions.

T-cell depletions were performed as described previously (5, 15, 32). Briefly, mice were inoculated intraperitoneally with 0.5 ml each of MAb 191.1 (anti-CD4) and MAb 169.4 (anti-CD8) supernatant fluid obtained from artificial capillary cultures (Cellco, Germantown, Md.). Mice were inoculated five times prior to vaccination. Both MAbs belonged to the rat anti-mouse IgG2b isotype. Blood samples from all mice were checked for T-cell depletion levels by flow cytometry at 7 to 10 days after the last injection of antibody. T-cell levels in mononuclear blood cells from depleted mice ranged from <1 to 3% of the nucleated peripheral blood cells.

CTL assays.

Cytotoxic T-lymphocyte (CTL) assays were performed as described previously (17). Briefly, spleen cells from vaccinated mice were mixed with radioactive chromium-labeled FBL-3 Friend virus-induced tumor cells at a ratio of 200:1 (2 × 10⁶ spleen cells per 10⁴ tumor cells in 200 μl of Iscove’s medium with penicillin and 10% fetal bovine serum in 96-well tissue culture plates). The cells were incubated at 37°C for 6 h, and then 100 μl of supernatant was sampled and counted with a gamma counter for ⁵¹Cr release.

Flow cytometric analyses.

At 3 days postchallenge with FV, nucleated spleen cells suspensions were made from two vaccinated and two unvaccinated mice. The cells were analyzed by flow cytometry with a FACStar I flow cytometer modified for five-parameter analysis. A total of 10,000 cells were analyzed for CD69 (37) coexpression on CD4, CD8, and CD19⁺ cells. Labeled antibodies were obtained from Pharmingen (San Diego, Calif.). The cells were gated for lymphocytes by forward and side scatter, and dead cells were gated out by propidium iodide staining.

RESULTS

Kinetics of protection induced by live attenuated F-MuLV.

To determine how rapidly protection developed after vaccination with live attenuated virus (N-tropic F-MuLV), susceptible mice were challenged with a high dose of pathogenic, B-tropic FV complex at different time points after immunization. All unvaccinated mice and those challenged at 1 day postvaccination developed fulminant infections characterized by severe FV-induced splenomegaly sustained over a 6-week time period (Fig. 1). We have previously shown that such splenomegaly not resolved by 6 weeks postchallenge is indicative of FV-induced erythroleukemia with greater than 95% fatality (18). In contrast to unvaccinated controls, mice challenged as early as 1 week postvaccination were fully protected against FV-induced splenomegaly (Fig. 1). To determine whether this rapid protection was immunological in nature, mice were depleted of CD4⁺ and CD8⁺ T cells prior to vaccination (10, 15–17). At 1 week after vaccination, the animals were challenged with FV. After challenge all of the T-cell-depleted animals developed rapid and lethal disease, demonstrating a requirement for T cells in protection (Fig. 1).

FIG. 1.

Kinetics of protection from FV-induced splenomegaly in mice vaccinated with F-MuLV. Age-matched (B10.A × A.BY)F₁ mice were vaccinated with live attenuated F-MuLV. The mice were then challenged at different time points postvaccination as indicated and were monitored for the induction and progression of splenomegaly. Symbols: ▵, unvaccinated controls (n = 8); □, challenge at 1 day postvaccination (n = 6); ▴, challenge at 1 week postvaccination (n = 6); ■, challenge at 2 weeks postvaccination (n = 5); ●, challenge at 3 weeks postvaccination (n = 6); ⧫, challenge at 4 weeks postvaccination (n = 6). Six mice were T-cell depleted prior to F-MuLV vaccination and were challenged 1 week later (○).

In addition to measuring acute FV-induced splenomegaly, we also determined the ability of vaccination to prevent establishment of low-level persistent infections since previous results showed that it was more difficult to protect against the establishment of persistent infections than to protect against acute disease (8). Clearance of persistent virus was tested by assaying spleens for infectious centers at 6 weeks postchallenge. Protection from low-level persistent spleen virus was not achieved in most mice until 3 weeks postvaccination (Table 1). Thus, protection, as measured by clearance of the challenge virus, developed more slowly than protection against acute splenomegaly.

TABLE 1.

Persistent FV challenge virus in mice vaccinated with F-MuLV

Animal no.	No. of infectious centers ata:
	1 wk	2 wks	3 wks	4 wks
1	2	12	0	0
2	1	9	0	0
3	5	5	0	0
4	3	1	0	0
5	1	0	0	0
6	1		0	0

^aNumber of infectious centers per 2 × 10⁷ spleen cells from individual mice. Mice were tested at 6 weeks postchallenge. The time indicates the number of weeks after vaccination when mice were challenged with FV. 

Development of immune responses after vaccination.

Since the T-cell depletion experiment indicated that T cells were important for protection against challenge at 1 week postvaccination, it was of interest to determine whether T-cell effectors could be detected at that early time point. We were especially interested in CTL, since earlier experiments showed vaccine-induced CD8⁺ T cells to have potent antiviral effects in adoptive transfer experiments at 1 month postvaccination (9). Direct CTL assays with spleen cells from vaccinated mice revealed lysis of Friend virus-induced FBL-3 tumor cells in four of four mice at both 1 and 2 weeks postvaccination (Fig. 2). Thus, CTL activity correlated with protection from splenomegaly. Interestingly, the CTL activity was transient and no longer detectable at 3 weeks postvaccination unless the spleen cells were restimulated in vitro (data not shown). The diminished CTL activity at 3 weeks postvaccination coincided with complete loss of detectable vaccine virus in the spleens of the vaccinated mice (Table 2).

FIG. 2.

Kinetics of virus-specific CTL responses in F-MuLV-vaccinated mice. After vaccination with F-MuLV, spleen cells were analyzed at weekly intervals for direct CTL killing as described in Materials and Methods. Effector cells from individual mice were not stimulated in vitro before the assay. The target cells were ⁵¹Cr-labelled Friend virus-transformed FBL-3 tumor cells, and the effector/target ratio was 200:1. Each bar represents the percentage of virus-specific lysis of individual mice against FBL-3 tumors cells. There was no significant reactivity against YAC-1 natural killer cell targets or uninfected EL-4 cells (data not shown).

TABLE 2.

Kinetics of vaccine virus replication in mouse spleen cells

Animal no.	No. of infectious centers at (time after vaccination)a:
	1 wk	2 wks	3 wks	4 wks
1	3.5 × 102	0	0	0
2	1.4 × 104	0	0	0
3	1.0 × 102	2	0	0
4	2.0 × 101	0	0	0
5	2.4 × 101	0	0	0
6		1	0	0

^aNumber of infectious centers per 10⁷ spleen cells from individual mice. 

Previous studies demonstrated virus-neutralizing antibody responses at 1 month postvaccination with F-MuLV, but it was not known how the development of the antibody responses correlated with the development of protection. To address this issue, plasma samples were taken from mice at weekly intervals after vaccination. No FV-neutralizing antibodies were found during the first 2 weeks after immunization, even though those mice were protected against acute splenomegaly (Fig. 3). By 3 weeks postvaccination five of six mice had moderate virus-neutralizing antibody titers that further increased by week 4 (Fig. 3). The rise in virus-neutralizing antibody titer at 3 weeks postvaccination correlated with protection from persistent infection (Table 1).

FIG. 3.

Kinetics of virus-neutralizing antibody development in F-MuLV-vaccinated mice. After vaccination with F-MuLV, plasma samples were taken at weekly intervals to assay for the presence of F-MuLV-neutralizing antibodies as described in Materials and Methods. Each bar is from an individual mouse. The neutralizing antibody titer was considered to be the highest dilution at which greater than 75% of the input virus was neutralized.

Rapid reactivation of primed lymphocytes from vaccinated mice.

In experiments where mice were challenged with FV at 1 month postvaccination, protection occurred even though no CTL effectors were demonstrable at that time point (Fig. 2). This confirmed earlier data showing that splenic lymphocytes were not highly activated at 1 month postvaccination (9). To determine if all three major lymphocyte subsets from vaccinated mice could be quickly reactivated by challenge, we analyzed cells at 3 days postchallenge with pathogenic FV. Flow cytometric analysis of the CD69 activation marker showed high levels of activation in all three lymphocyte subsets compared to unvaccinated mice (Fig. 4). Thus, the vaccinated mice displayed a potent and broad anamnestic response which provided solid protection against virus challenge.

FIG. 4.

Reactivation of primed lymphocytes from F-MuLV-vaccinated mice at 3 days post-FV challenge. Mice vaccinated 1 month previously were challenged with FV. At 3 days after challenge, spleen cells were analyzed by flow cytometry for CD69 early activation markers on lymphocytes dually stained with antibodies specific for CD19 (B cells) or for CD4 or CD8 (T cells), as indicated. In the top panels are the activation levels of unvaccinated control mice. In the bottom panels are the results from the vaccinated mice showing high levels of activation (top right section of each panel). The number shown in each section is the percentage of cells in that section.

Long-lasting protection induced by live attenuated vaccine virus.

Since the replication of the vaccine virus was very poor and it was cleared by 3 weeks postimmunization (Table 2), it was possible that protection was short-lived. To test for long-term protection, mice were challenged after either a 3- or 6-month waiting period after immunization. All unvaccinated control mice developed severe and progressive splenomegaly indicative of FV-induced erythroleukemia (Fig. 5). In contrast, the groups vaccinated either 3 or 6 months earlier had no acute splenomegaly after FV challenge. In addition, no persistent virus was detected in the spleen cells of mice challenged at 6 months postvaccination, indicating complete protection against pathogenic FV (data not shown). Thus, despite its low-level replication and lack of persistence in adult mice, the live attenuated F-MuLV induced long-lasting immunity against FV infection.

FIG. 5.

Long-lasting protection from FV-induced splenomegaly in mice vaccinated with F-MuLV. Age-matched (B10.A × A/Wy)F₁ mice were vaccinated with live attenuated F-MuLV. The mice were challenged at 3 or 6 months postvaccination with 1,500 SFFU of Friend virus complex and monitored for the induction and progression of splenomegaly. Symbols: ▴, 3-month group, n = 10; ●, 6-month group, n = 10; ■, naive controls, n = 20.

DISCUSSION

The present results provide new evidence that the efficacy of vaccination with live attenuated Friend virus is more related to its ability to stimulate memory in multiple arms of the immune system than to its ability to generate persistent immunological effectors. For example, vaccination elicited CTL effectors within 1 week of vaccination, but protection was better at preventing persistent infections at 3 and 4 weeks postvaccination, when the level of CTL effectors had dropped (Fig. 6). Yet we know from previous experiments that immune CD8⁺ T cells are required elements for transferring immunity to naive mice (9). The likely explanation for this paradox is the rapid reactivation of virus-specific CD8⁺ memory cells that we observed after virus challenge (Fig. 5).

FIG. 6.

Correlation between development of protection and immune responses. This figure summarizes the mean kinetic results shown in Fig. 1 to 3 and Table 1 to show kinetic correlations between vaccine virus-induced protection and immune responses. The percentages of mice with detectable CTL responses are shown as open bars, and the mean neutralizing antibody titers are shown as solid bars. The dotted line shows the percentage of mice protected against acute disease, and the dashed line shows the percentage of mice protected against persistent infection at each weekly time point.

Antibodies are important in immunity because they can neutralize free virus without the time delays that are required to reactivate cell-mediated responses. Ideally, high titers of neutralizing antibodies have the potential to completely block viral infections. However, recent studies in the macaque (22, 35) and hu-PBL-SCID mouse (14) models with chimeric simian/human immunodeficiency virus (SHIV) and human immunodeficiency virus (HIV), respectively, have shown that complete blocking of infection only occurred when virus-neutralizing antibodies were present at very high titers that are unlikely to be achieved by vaccination (28). Likewise, in the FV model, passive transfers of virus-neutralizing antibodies at titers equivalent to those induced by vaccination reduced virus loads by 100-fold but did not completely block infection (9). In fact, standing titers of virus-neutralizing antibodies were not required for protection in adoptive transfer experiments if immune B cells were transferred along with immune CD4⁺ and CD8⁺ T cells (9). Since none of the lymphocyte subsets in those experiments were activated at the time of transfer (9), the results suggested that it was the ability of these immune lymphocytes to reactivate and quickly respond to virus challenge that best correlated with protection, rather than their effector status at the time of virus challenge. In support of this hypothesis, we now show very high levels of activation in all three major types of lymphocytes at 3 days postchallenge (Fig. 4).

Protection against both acute splenomegaly and the establishment of persistent FV infection took 3 weeks to develop and correlated with the presence of virus-neutralizing antibodies in addition to primed CTL (Fig. 6). Protection against acute disease only was much quicker and was already apparent at 1 week postvaccination, when CTL but not antibodies were detectable (Fig. 6). A recent study with the simian immunodeficiency virus (SIV) model for AIDS also showed a longer lag time for the development of protection against persistent infection (15 weeks) compared to protection against acute disease (5 weeks). However, in that study the correlates of protection were not determined (6). In a recent study with bovine leukemia virus (BLV) in sheep, protection from establishment of persistent virus was achieved by vaccination with a CTL peptide only (20). It is not presently clear why protection against persistent FV requires broader immune responses than are required for BLV. However, this may relate to the relative virulence of the viruses. BLV is typically less virulent in sheep than is FV in the mouse strains used in our studies, and it is probably easier to achieve protection against a virus which replicates and spreads more slowly. Further delineation of the requirements for preventing persistent retroviral infections is important since persistent viruses can reactivate and cause disease, especially in immunosuppressed hosts (16).

In the SIV model a “threshold theory” has been proposed in which vaccine virus replication must reach a certain threshold before strong protective immunity is induced (34). The problem in the SIV model is that viruses which replicate well enough to induce strong immunity also become persistent and can revert to cause pathogenesis (1, 11). Such reversion also appears to have occurred in humans infected with an attenuated form of HIV (12). Interestingly, potent immunity in the current FV study was achieved with a weakly replicating virus that induced only a low-level, transient infection in the spleen (Table 2), with no detectable plasma viremia at any time point tested (data not shown). Our results indicate that it is possible to achieve solid and long-lasting protection with a weakly replicating retrovirus that does not become persistent. However, compared to SIV or HIV, it is much easier to remove pathogenic potential and retain immunogenicity with FV because the virus is a complex in which pathogenicity is primarily dependent on the SFFV component, whereas immunogenicity is primarily due to the F-MuLV helper component. Nevertheless, the present results suggest that it is theoretically feasible to engineer a retroviral vector that elicits protective immunological responses but does not establish persistent infections that could lead to reversion to pathogenicity.