key: cord-1042527-wt4y96xu
authors: Yu, Mathilde W. N.; Lemieux, Suzanne; Talbot, Pierre J.
title: Genetic control of anti‐idiotypic vaccination against coronavirus infection
date: 2005-11-17
journal: Eur J Immunol
DOI: 10.1002/eji.1830261258
sha: e6016892312aa847c3fb3a6945823ce583944aaf
doc_id: 1042527
cord_uid: wt4y96xu

The idiotypic network can be experimentally altered to induce protective immune responses against microbial pathogens. Both internal image and noninternal image anti‐idiotypic (anti‐Id) antibodies have been shown to trigger antigen (Ag)‐specific immune responses. Therefore, mechanisms of anti‐Id vaccination appear to go beyond structural mimicry of Ag, but remain undefined. Using the neurotropic murine coronavirus animal model, we have previously shown that a polyclonal noninternal image anti‐Id (Ab2) could vaccinate BALB/c mice. To characterize its mode of action, we have examined the immune modulating capability of this Ab2 in vivo in strains of mice with different H‐2 haplotypes. Even though only internal image anti‐Id are expected to induce non‐genetically restricted immunity, this noninternal image Ab2 induced protective immunity in four of eight genetically different strains of mice susceptible to coronavirus infection. These were BALB/c (H‐2(d)), DBA/1 (H‐2(d)), DBA/1 (H‐2(q)), and SWR (H‐2(q)) mice. Protection was generally correlated with the induction of specific antiviral Ab (Ab3) that showed biological properties, such as virus neutralization in vitro, similar to the initial Ab1. To evaluate the genetic implication of the H‐2 haplotypes in protection, congenic mice were also tested. Vaccination profiles suggest that cooperation between background gene(s) of the BALB/c mouse with H‐2(d) and H‐2(q) loci is necessary for an optimal protective immune response, although the main genetic element(s) regulating the antiviral response to Ab2 inoculation appeared to be located outside the major histocompatibility complex. These results are consistent with the ability of Ab2 to induce protective antiviral antibodies in genetically different animals by biological mimicry.

The use of anti-idiotypic antibodies (anti-Id Ab) as potential vaccines against infectious agents has been of interest because of the capacity of these Ab to act as surrogate antigens (Ag). Anti-Id are classified into three major categories: Ab2,, Ab2, and Ab2.,. The Ab2, and Ab2, recognize framework idiotopes either near or distant from the paratope of the Abl, respectively, whereas Ab2,, also called internal image anti-Id, recognize paratope-associated idiotopes. By their potential to molecularly mimic the Ag, Ab2, have been shown to induce neutralizing Ab, protective Ab, or both against various infectious agents [l-31 and to activate T lymphocyte [4] . A true internal image would not be genetically restricted in its ability to induce an immune response, and would be efficient across species barriers [5, 61. [I 160401

Theoretically, only internal image anti-Id were expected to act as surrogate Ag. It was therefore surprising that noninternal image anti-Id induced neutralizing, protective, or both immune responses against a pathogen [7-91. However, in contrast to immunization with internal image Ab2,, non-internal image anti-Id may be genetically restricted by their inability to induce an immune response in different species [lo] . Recent studies have also shown that a T cell response could be elicited by noninternal image anti-Id [ll, 121. These observations emphasize the complexity of the idiotypic network and the need to characterize the mechanisms of action of Ab2.

Neurotropic murine coronaviruses such as MHV-A59 induce multiple sclerosis-like diseases in rodents and provide an excellent animal model for studies of the idiotypic network in the context of a virus infection. We have previously reported the induction of a protective anti-coronavirus immune response in BALB/c mice by polyclonal Ab2 produced in rabbits immunized with a neutralizing and protective mAbl designated 7-10A [20].

Since the Ab2 preparation did not contain detectable internal image activity and inhibited the interaction between Abl and virus, this Ab2 preparation was identified as Ab2,

To characterize further the protective effect of these Ab2, we asked whether their vaccinating potential could be 

Strain A59 of mouse hepatitis virus (MHV-A59) was originally obtained from the American Type Culture Collection (Rockville, MD), plaque-purified twice, and passaged four times at a multiplicity of infection of 0.01 on DBT cells as described [21].

The immunization protocol for generating anti-Id anti-7-10A, their purification and characterization, as well as the Ab2 immunization protocol, the protection assays, the ELISA and virus neutralization assays for detection of Ab3 against MHV-A59 were performed as described [20].

Results of protection assays in vivo were analyzed with the Kaplan-Meier survival curve procedure [22] . Cox's proportional hazards model was used to analyze the reduction in mortality associated with the administration of the anti-Id compared to NRIg [22] . Antiviral Ab3 antibody responses were evaluated and analyzed with the Mann-Whitney test [22] . In these tests, a p value of cO.05 was accepted as statistically significant.

We have previously shown that rabbit polyclonal anti-Id anti-7-10A (Ab2) were able to induce a protective immune response in BALB/c mice against MHV-A59 [20]. To examine whether the vaccinating property of these Ab2 was genetically controlled, mice of eight different inbred strains were immunized with Ab2 or control normal rabbit immunoglobulins (NRIg) and challenged intracerebrally with ten LD50 of MHV-A59. The LD50 values ranged between 1.75 x lo4 and 8 x lo4 PFU, except for B6 and SWR strains, where it was around 6 x lo3, and B10.D2 and BIO.G mice where it was 1 x 10'-1.3 x 10' (data not shown). In agreement with our previous observations [20] , about 75% of AbZimmunized BALBk (H-2d) mice survived over 60 days after virus challenge (Fig. 1) . However, marked variations in the survival rates were detected from one mouse strain to another. An almost complete protection was obtained in SWR and DBA/l (H-27 mice, whereas only 25-30% of immunized DBA/2 (H-2d) and CBA (H-2k) mice survived virus challenge. However, the protection observed in CBA mice was not statistically significant. Moreover, Ab2 had no vaccinating capacity in B6 and B10 (H-2') mice, and an apparent reduced survival was even seen in C3H (H-2k) mice in comparison with control NRIg-treated mice. Thus, AbZmediated rotection was seen in H-2q and H-2d mice, but not in H-2 and H-2' mice. These results indicate that the efficacy of Ab2 treatment is genetically controlled and suggest that it may be H-2dependent. haplotypes in mice of the B10 background may reverse their resistance to Ab2-induced protection against MHV-A59 infection. Results of these experiments are reported in Fig. 2 . As expected, reduction in survival was observed in AbZimmunized BALB/c congenic mice expressing the H-2b or H-2k haplotype. However, the introduction of H-2q or H-2d haplotype in mice of B10 background was not sufficient to override the failure of Ab2 treatment to protect B10 mice from virus challenge (Fig. 2) . These results suggest that the capacity of Ab2 to protect mice against MHV-A59 infection is mainly dependent on genetic elements located outside the MHC complex. However, given an appropriate genetic background, the amplitude of the Ab2-induced antiviral response would probably be regulated by H-2 or H-2-linked gene(s). This conclusion is supported by the observed variations in survival in MHCcongenic mice on the BALB/c background. It remains to be elucidated whether a unique background gene shared by BALB/c, DBA/l, DBA/2, and SWR mice is responsible for the capacity of these mice to respond to Ab2 anti-7-10A. This is, however, unlikely since mice of these four strains were not protected to the same level following Ab2 vaccination. Further experiments are needed before the genes involved are identified. Testing the AbZmediated anti-coronavirus protection in CXB recombinant inbred mice may help to characterize the BALB/c-associated gene.

To evaluate whether variations in AbZmediated protection of mice belonging to diffeent strains correlated with the presence of antiviral antibodies (Ab3), the sera of immunized mice were tested by ELISA and virus neutralization assays for the presence of MHV-A59-specific Ab3. The results presented in Table 1 indicate that mice of any strain in which protection was observed after Ab2 treatment developed significant levels of MHV-A59-reactive and neutralizing Ab3. This suggests an overall link between AbZinduced antiviral Ab3 and the observed protection of mice from viral challenge. This was confirmed by the Cox's proportional hazards statistical model, where the high serological ELISA response of each mouse was defined as a response higher than the median serological scores specific for responding mice of a given strain. For Ab2-protected mouse strains, the. reduction in mortality correlated with a high serological response. Indeed, we observed a 78 % reduction in mortality at a p value of less than 0.001 for responder strains, whereas no significant reduction of mortality was seen in nonresponder strains (data not shown).

The ability of Ab3 to neutralize virus infectivity suggests that protection against MHV-A59 observed in Ab2immunized mice may be mostly due to this property of Ab3. However, careful examination of the data allowed us to identify some differences in titers of neutralizing antibodies in the sera of BALB.B and BALB.K mice that showed the same level of AbZinduced protection (data not shown). These differences could indicate the possible involvement of other components of the immune response in protection. Such studies are in progress.

Due to their capacity to induce immune response across species barriers, Ab2, remains undoubtedly the anti-Id of interest for vaccination against microbial pathogens. However, the present demonstration that rabbit polyclonal Ab2, anti-Id can induce a protective immunity against coronavirus infection in mice of unrelated inbred strains supports the idea that noninternal image anti-Id may also be considered as good candidates for vaccination against pathogens.

We have reported elsewhere that unlike the Ab2, used in the present study, another rabbit Ab2, was inefficient in vaccinating BALBk mice against lethal MHV-A59 infection [23] . The reasons why a noninternal image Ab2, induces a protective immunity, whereas another of the same type raised against a mAbl with related properties does not, are intriguing. The relative contribution of the different components of the immune network in AbZinduced vaccination is not yet fully elucidated. It should also be kept in mind that, under certain circumstances, immunization with anti-Id may even be deleterious. Indeed, a significant increase in mortality, rather than protection, was observed in one of the mouse strains (C3H) used in this study. Similarly, Kennedy et al. [24] reported an increase in the pathogenicity of herpes simplex virus following immunization of BALB/c mice with a polyclonal Ab2. Therefore, a better understanding of the interactions between the components of the idiotypic network is certainly needed before anti-Id can be exploited as successful, reproducible and safe alternative vaccines against microbial pathogens.

Proc. Natl. Acad. Sci

Proc. Natl. Acad. Sci

Proc. Natl. Acad. Sci. USA 1992. 89: 2546. Kraaijeveld

Statistical methods in medical resenrch