Direct immunization with aggregated amyloid β protein (Aβ) and passive immunization with anti-Aβ antibodies (Abs) reduce plaque burden in Alzheimer's disease (AD) mouse models and improve cognitive deficits present in those models 12345. Although no adverse effects of immunization were noted in earlier studies, more recent data in mice indicate that there is the potential of exacerbation of cerebral-amyloid angiopathy (CAA) associated microhemmorhages in certain mouse strains following passive immunization with certain anti-Aβ antibodies 678. An active immunization trial in humans was initiated using fibrillar Aβ42+QS-21 adjuvant (AN-1792) but was halted due to a meningio-encephalitic presentation in ~6% of individuals 91011. Reports of individuals enrolled in the trial suggest that those subjects who developed modest anti-plaque antibody (Ab) titers did show some clinical benefit relative to subjects that did not develop detectable titers 91112. A small phase II study of AD patients administered human IVIG containing anti-Aβ Abs showed slight improvement in ADAScog following administration; however the clinical effect was modest and only a few subjects were evaluated 13.

Given the pre-clinical data, hints of efficacy in humans, and the lack of disease-modifying therapies for AD, Aβ immunotherapy or derivative approaches are still worthy of pursuing. However, the mechanism or mechanisms through which Aβ immunotherapy works remain enigmatic 1415. The amount of Aβ deposited when immunization is initiated, the AD mouse model used, and the properties of the anti-Aβ antibodies used, all affect the outcome 12161718. One of the debates with respect to mechanism centers on peripheral versus a central action of the antibody 31920. There is evidence to support both mechanisms, and it will be a very difficult issue to definitively address this through additional experimentation. Another debate is in regard to the role of microglia activation. Several groups report transient or stable enhancements of microglia activation associated with Aβ removal; others do not 1212223. In postmortem human tissue from AD patients who had received the AN-1792 vaccine, Aβ-laden microglia were noted in areas where Aβ clearance is hypothesized to have occurred 24. Thus, microglial activation has been proposed to facilitate removal of Aβ from the brain following vaccination.

The IL-1 superfamily (including IL-1β, IL-1α and IL-18) is a group of cytokines that exhibit a large number of biological responses 25. Interleukin-1β is a key mediator of host response to infections and a primary cause of inflammation 25. In vivo, IL-1β is elevated during infections and in several chronic inflammatory diseases such as arthritis, scleroderma, systemic lupus erythematosus, vasculitis, sepsis, septic shock, and atherosclerotic lesions as well as in brains of AD patients 25. As least two IL-1 receptors (IL-1R) have been identified: type I and type II receptors (IL-RI and IL-RII) 26. IL-1β binds IL-1RI and upon IL-1 binding, IL-1RI recruits the accessory protein IL-1R-AcP, and initiates a stimulatory signal transduction cascade 26. IL-1RII acts as a decoy receptor and competes with IL-1RI to down-modulate IL-1 activity 27. In AD and Down's syndrome, IL-1β production is increased in microglial cells in the vicinity of amyloid plaques 2829. Initial studies examining the association of polymorphisms in the IL-1 and IL-1 receptor genes showed positive association of certain alleles with AD risk 3031323334. However, like many AD genetic association studies, subsequent studies failed to confirm the initial association. Meta-analyses of all studies on IL-1α and β linkage show no evidence for association of these loci with AD http://www.alzforum.org/res/com/gen/alzgene/. A recent report shows that activation of microglia with secreted APP (sAPPα) results in a dose-dependent increase in secreted IL-1β 35. Similarly, cortical neurons treated with IL-1β showed a dose-dependent increase in sAPPα secretion, elevated levels of α-synuclein and phosphorylated tau 35. In APP transgenic mice, IL-1 reactivity and other inflammatory markers are increased in microglial cells surrounding amyloid deposits during various stages of amyloid deposition in these mice 3637. Another member of the IL-1 superfamily, IL-1 receptor antagonist (IL-1Ra) 38, is also synthesized and released in parallel to IL-1β, IL-1α, and IL-18. IL-1Ra binds to IL-1RI and blocks IL-1 dependent signal transduction, thus functioning as an endogenous, IL-1 selective inhibitor of inflammation 38. Interestingly, IL-1Ra knockout mice show enhanced microglial activation and neuronal damage following intracerebroventricular infusion of human Aβ 39. Collectively, these data suggest that IL-1 is a key mediator of microgliosis and subsequent inflammatory responses following Aβ deposition as well as in the production of substrates necessary for neuropathological changes seen in AD.

To gain additional insight into the role of IL-1 signaling on microglial activation, on IL-1-mediated inflammatory responses following Aβ vaccination, and on Aβ deposition during normal aging, we used interleukin-1 receptor 1-knockout (IL-1 R1-/-) mice 404142 that were crossed to APP Tg2576 transgenic mice (APP/IL-1 R1-/-). The IL-1 R1-/- mice lack the type 1 interleukin-1 receptor, but develop normally. Moreover, with a few exceptions, these mice are normal, showing alterations in IL-1-mediated immune response to certain stimuli. Following penetrating brain injury in IL1-R1-/- mice, fewer amoeboid microglia/macrophages are present near the sites of injury, astrogliosis is mildly abrogated and cyclooxygenase-2 (Cox-2) and IL-6 expression are reduced 42. In another report, IL-1 R1-/- mice failed to respond to IL-1 in several assays, including IL-1-induced IL-6 and E-selectin expression, and IL-1-induced fever and acute-phase responses to turpentine 41. These data in IL-1 R1-/- mice demonstrate that IL-1 R1 is critical for most IL-1-mediated signaling events tested. We performed passive immunization with an anti-Aβ mAb in Tg2576 mice crossed into the IL-1 R1-/- background (APP/IL-1 R1-/-), and determined whether microglial activation and consequent inflammatory responses are necessary for Aβ reduction. These studies show that passive immunization with anti-Aβ mAb is effective in reducing plaque load in APP/IL-1 R1-/- mice when the immunization is started prior to significant plaque deposition and thus support our general hypothesis that microglial activation may not be required for efficacy of immunization in Tg2576 mice.

Despite multiple studies of anti-Aβ immunotherapy in mice, there is still no consensus on how anti-Aβ immunotherapy works 1415, particularly as it relates to the role of microglial activation. It was originally proposed that Aβ immunization triggers phagocytosis of antibody-bound Aβ immune complexes via microglial FcR. After immunization, increased number of microglial cells stained with anti-Aβ antibodies were observed 1. Indeed, using an ex vivo strategy, it was shown that anti-Aβ antibodies induce phagocytosis of Aβ plaques 2. Importantly, Fab fragments of these antibodies fail to induce Aβ phagocytosis, suggesting that the enhanced uptake is attributable to FcR 2. Subsequent studies have shown that at least in Tg2576 APP mice, a role for enhanced phagocytosis of mAb:Aβ complexes via the FcR can largely be ruled out, since Aβ1-42 immunization in Tg2576 × FcRγ-/- crossed mice was effective in reducing Aβ loads 23. Additional studies now show that an intact mAb (and therefore FCR interactions) is not required for efficacy; since Fab fragments 46 and scFv fragments (Levites and Golde, unpublished observation) are efficacious in immunotherapy. Several groups have reported that following Aβ immunotherapy, there are transient or stable enhancements of microglial activation associated with Aβ removal; whereas others do not find this 1212223. Furthermore, in humans receiving the AN-1792 vaccine, Aβ-laden microglia have been noted in postmortem studies 24. Although antibody and microglial Fc receptor-mediated interactions have been suggested to activate microglia following vaccinations, other inflammatory consequences may play a role in this paradigm. Based on published reports, it has been suggested that clearance of amyloid deposits in patients enrolled in the AN-1792 trial may have been due to an adverse inflammatory response to the vaccine rather than due to the anti-Aβ antibodies 47. This proposition may be supported by some recent reports, wherein induction of experimental autoimmune encephalitis (EAE) and nasal vaccination with glatiramer acetate reportedly decrease amyloid plaques in APP transgenic mice 48. Another report by the same group shows that, in mice over expressing IFN-gamma in the CNS, amyloid vaccination lead to meningoencephalitis and T cell-dependent clearance of amyloid plaques from the brain 49. Both of these reports provide evidence that peripheral inflammatory responses and CNS autoreactive T cells may play a role in vaccination-induced clearance of plaques. Furthermore, some recent reports have indicated that inflammatory insults, either by injecting LPS directly into the brain 4450 or overexpression of TGF-β in the CNS 51, can result in reductions of amyloid deposits. Enhanced microglial activation was noted in both of these reports and is suggested to contribute to the clearance of amyloid deposits.

In this report, we sought to determine the role of IL-1-mediated microglial activation on IL-1-mediated inflammatory responses following Aβ vaccination and on Aβ deposition during normal aging using interleukin-1 receptor 1-knockout (IL-1 R1-/-) mice 404142 that were crossed to APP Tg2576 transgenic mice (APP/IL-1 R1-/-). We first tested the efficacy of Aβ immunization in APP/IL-1 R1-/- mice. Our results show that passive immunization with an anti-Aβ mAb is effective in reducing plaque loads both in APP/IL-1 R1-/- mice and APP/IL-1 R1+/- littermates, when immunization is started prior to significant plaque deposition. However, as we have seen previously, immunization was not efficacious in mice that have pre-existing Aβ loads 171852. Thus, these results support our general hypothesis that microglial activation may not be required for efficacy of immunization in Tg2576 mice. The lack of IL-1 R1 (in -/- mice) did not significantly alter Aβ deposition in untreated mice. There were no significant differences in total extractable Aβ levels or overall histochemical loads, at any time, between the APP/IL-1 R1-/- mice and APP/IL-1 R1+/- littermates compared to wild type Tg2576 mice (IL-1 R1+/+). Curiously, in 2 of 7 15-month-old APP/IL-1 R1-/- mice examined, an unusual pattern of Aβ plaque staining was noted, with an abundance of diffuse immuno-reactive Aβ plaques in the neocortex of these mice. It is not clear at this time whether this unusual pattern of diffuse Aβ deposits is due to the IL-1 R1-/- phenotype or some mouse background effect. We then examined the effects of IL-1 R1 knockout on the state of microglial activation and astrogliosis surrounding amyloid plaques deposits. For microglial staining, we used two well characterized markers for microglial activation, anti-mouse CD45 and Iba1, and for activated astrocytes we used anti-GFAP staining. Our results show that there were abundant numbers of CD45 and Iba1 immuno-reactive microglia present, surrounding Aβ plaques in APP/IL-1 R1-/-, APP/IL-1 R1+/- and wild type Tg2576 mice (IL-1 R1+/+), with no significant differences in the immuno-reactivity of staining using these markers. Similarly, robust GFAP staining was seen in all three groups of mice analyzed, with no significant differences seen in the GFAP immuno-reactivity comparing all three groups of mice.

Based on our immuno-staining analysis, we were not able to ascertain whether abrogated IL-1 signaling in the IL-1 R1-/- mice blunted the inflammatory microglial response or astrogliosis in the region of deposited Aβ plaques. Previous experiments in IL-1 R1-/- mice have shown abrogated IL-1-mediated responses following acute inflammatory stimuli. In a stab wound model of injury in the brain, IL1-R1-/- mice had fewer amoeboid microglia/macrophages near the sites of injury, mildly abrogated astrogliosis and reduced expression of cytokines induced by IL-1 expression 42. In another report, IL-1 R1-/- mice had reduced IL-6 and E-selectin expression, and reduced IL-1-induced fever and acute phase responses to turpentine 41. However, IL-1 R1-/- mice do not differ from control mice in their responses to either a lethal challenge with D-galactosamine plus LPS or high dose LPS 40, indicating that IL-1 R1 signaling functions may not be necessary for the response to LPS. Thus, it is possible that the chronic nature of the microglial response during the course of amyloid deposition may abrogate any acute or subtle signaling events mediated through the IL-1 R1 receptor. Certainly, it is possible that other receptors for IL-1 may compensate for the lack for IL-1 R1 in this situation. Besides the IL-1 R1 and IL-1 RII receptors, the recently reported P2X7 receptor has also been implicated to be a key player in IL-1 signaling 53 and could compensate for the lack of IL-1R1 in IL-1 mediated signaling events. Alternatively, the microglial response to deposited Aβ may not require signaling through the IL-1 R1 receptor. The LPS receptor (CD14) 5455, the scavenger receptor complex (CD36) 56 and toll-like receptors (TLR-2, TLR-4) 57 can directly activate microglia in response to amyloid deposition, possibly circumventing any IL-1 R1-mediated signaling events in the IL-1 R1-/- mice.

Like our previous studies, these studies suggest that microglial activation is not required for immunization to work in Tg2576 mice, although this should not be viewed as definitive. As indicated above, in the IL-1 R1-/- mice, microgliosis and astrogliosis are mildly abrogated at best and do not result in microglial paralysis. Thus experiments using recently developed CD11b-HSVTK mice. developed by Aguzzi and colleagues 58. that enable selective killing of microglia cells may provide more definitive results.

The author(s) declare that they have no competing interests.

PD conceived the design of the study, performed experimental analysis and data interpretation and prepared the manuscript. LAS bred and maintained the IL-1 R1-/- mice, performed immunizations, harvested tissues, performed CD45 staining. RWP performed the Aβ ELISA. VMH performed the image quantification and immunohistology. YL performed Aβ plasma ELISA and aided in the preparation of the manuscript. PC performed Iba1 immunostaining and APP western blotting. TEG conceived the design of the study, aided in the preparation of the manuscript, and provided critical analysis of the manuscript.