DNA epitope vaccine containing complement component C3d enhances anti-amyloid-β antibody production and polarizes the immune response towards a Th2 phenotype

Abstract

We have engineered a DNA epitope vaccine that expresses 3 self-B cell epitopes of Aβ₄₂ (3Aβ_1–11), a non-self T helper (Th) cell epitope (PADRE), and 3 copies of C3d (3C3d), a component of complement as a molecular adjuvant, designed to safely reduce CNS Aβ. Immunization of mice with 3Aβ_1–11-PADRE epitope vaccine alone generated only moderate levels of anti-Aβ antibodies and a pro-inflammatory T helper (Th1 phenotype) cellular immune response. However, the addition of 3C3d to the vaccine construct significantly augmented the anti-Aβ humoral immune response and, importantly, shifted the cellular immune response towards the potentially safer anti-inflammatory Th2 phenotype.

Introduction

One of the major health concerns for elderly people is Alzheimer's disease (AD), which is the most common form of dementia with progressive loss of memory and general cognitive decline. The first immunotherapy clinical trial on Alzheimer's patients, AN-1792, was halted when a subset of those immunized with the vaccine containing the amyloid-beta peptide (Aβ₄₂) developed adverse events (aseptic meningoencephalitis) in the central nervous system (CNS) (Orgogozo et al., 2003, Schenk, 2002, Steinberg, 2002). While the actual cause of the adverse events is unknown, speculation has centered on autoreactive T cells specific for the T cell epitope in Aβ, the conventional adjuvant (QS21), and the reformulation of the vaccine with polysorbate 80 during the phase IIa portion of the trial (Ferrer et al., 2004, Nicoll et al., 2003, Schenk, 2002). Subsequent analysis of postmortem brain tissue from patients that received the AN-1792 vaccine showed an overall reduction of Aβ burden in the CNS (Boche et al., 2007, Ferrer et al., 2004, Holmes et al., 2008, Masliah et al., 2005, Nicoll et al., 2006, Nicoll et al., 2003, Nitsch and Hock, 2007, Patton et al., 2006), and some suggestion of diminished progressive cognitive decline associated with the disease (Gilman et al., 2005, Hock et al., 2003), although this observation was not universal (Holmes et al., 2008). However, there was also evidence of increased CNS Aβ by ELISA (Patton et al., 2006), and increased incidence of cerebral vascular Aβ deposition (Holmes et al., 2008, Masliah et al., 2005, Nicoll et al., 2006, Patton et al., 2006). In the elderly AD patients, there were a low percentage of responders and the generally low titers in response to a self-Aβ antigen in the AN-1792 vaccine, even in the presence of a very potent adjuvant (Gilman et al., 2005, Patton et al., 2006). These results emphasize the difficulty facing active immunization approaches in elderly AD patients because the elderly generally develop functional deficits in their immune system or immunosenescence (Grubeck-Loebenstein and Wick, 2002). Accordingly, to avoid the problems associated with active immunization of elderly AD patients, recent clinical trials based on passive vaccination (AAB-001) were initiated. In these studies, different concentrations of humanized monoclonal anti-Aβ antibody are passively transferred to AD patients. However, passive immunotherapy requires the repeated administration of high doses of expensive humanized monoclonal anti-Aβ antibody. More importantly, this strategy is not likely to be useful for protective vaccination due to the substantial cost, invasive nature of the treatment, and the recurrent clinical visits necessary for effective delivery of the immunotherapy. Thus, there is growing consensus among some researchers based on both analysis of pre-clinical studies (Mamikonyan et al., 2007, Nickolic et al., 2007, Petrushina et al., 2007), as well as from the AN-1792 trial (Holmes et al., 2008, Patton et al., 2006, St George-Hyslop and Morris, 2008) that early preventive immunization prior to substantial neuropathology, neuronal loss, and cognitive deficits have become firmly established may be more effective and safer for future patients receiving immunotherapy. Especially if patients can be identified in a pre-clinical stage by validation of AD biomarkers (de Jong et al., 2006, Fagan et al., 2007a, Fagan et al., 2007b, Klunk et al., 2004).

Based on the hypothesis that early treatment is better, we previously proposed an active vaccination strategy based on an epitope vaccine composed of the immunodominant self-B cell epitope of Aβ₄₂ and a non-self T helper (Th) cell epitope. We demonstrated the feasibility of this strategy in wild-type (Agadjanyan et al., 2005) mice and then showed the efficacy and safety of epitope peptide vaccine in two different strains of APP/Tg mice (Mamikonyan et al., 2007, Petrushina et al., 2007). However, there are several problems associated with development of an epitope peptide vaccine for human clinical trials. First, there are technical limitations that are difficult to overcome due to the problems encountered in synthesizing large quantities of highly purified peptide epitope vaccines, which are also potent antigens. Secondly, there is the requirement for a potent conventional adjuvant and the only currently approved adjuvant for use in humans is Alum that does not induce robust anti-Aβ immune responses in APP/Tg mice (Ghochikyan et al., 2006). Our stratagem to overcome these problems was to design a DNA vaccine construct that includes the epitope vaccine concept linked to a potent molecular adjuvant, macrophage-derived chemokine (MDC/CCL21), to induce a robust anti-Aβ antibody response (Movsesyan et al., 2008). In this report, we have investigated another potential molecular adjuvant, complement (C′) component C3d, which has previously been shown to enhance antibody responses to several antigens when three copies of C3d (3C3d) were combined with the antigen as a fusion protein (Dempsey et al., 1996, Green et al., 2003, Ross et al., 2000). Our DNA epitope vaccine was re-engineered to include 3C3d as the molecular adjuvant and we found that the 3Aβ_1–11-PADRE-3C3d construct dramatically enhanced the anti-Aβ antibody response in mice compared to the immune response generated with the 3Aβ_1–11-PADRE DNA construct.