Amyloid β and Alzheimer disease therapeutics: the devil may be in the details

Abstract

Alzheimer disease (AD) is characterized by the progressive accumulation of amyloid β protein (Aβ) in areas of the brain serving cognitive functions such as memory and language. The first of two separate reports (see the related articles beginning on pages 415 and 440) reveals that intrinsic T cell reactivity to the self-antigen Aβ exists in many humans and increases with age. This finding has implications for the design of Aβ vaccines. The second report demonstrates that a number of FDA- approved nonsteroidal anti-inflammatory drugs are capable of lowering Aβ levels in mice. The work suggests that further testing of the therapeutic utility of these types of compounds for the potential treatment of AD is warranted.

Alzheimer disease (AD) has received a lot of recent attention, particularly in areas related to novel treatments. Recently, the potential therapeutic usefulness of the immune system has become apparent, leading to the question of whether it can be used to directly or indirectly influence AD-related pathology in beneficial ways. Active immunization with amyloid β (Aβ) peptides takes advantage of the immune system to generate antibodies that can somehow decrease Aβ-related pathology in mouse models of AD (1). Similarly, passive immunization involves direct administration of anti-Aβ antibodies, bypassing the need for an active immune response (2, 3). Since genetic, pathologic, and animal studies suggest that the buildup of Aβ in the brain leads directly or indirectly to cell dysfunction, cell death, and cognitive impairment, increased generation of anti-Aβ antibodies has the potential to prevent or treat AD by decreasing amyloid burden and its consequences in the brain. Though the first clinical trials for Aβ vaccination were halted due to CNS inflammation in a small subset of subjects, active and passive immunization strategies remain a viable potential therapy worth continued exploration. If positive effects can be seen in future trials, it will be important to minimize unwanted toxicity. In this issue of the JCI, Monsonego and colleagues (4) further characterize the innate immune response to Aβ in humans, thus revealing important details about how the elderly body reacts to Aβ, and opening new avenues to modify existing vaccination protocols. Also in this issue, Eriksen and colleagues (5) studied traditional NSAIDs that appear to have a nontraditional, COX-independent effect on decreasing Aβ42 production. While these drugs are often used to treat inflammation, they appear to have a novel effect on amyloid precursor protein (APP) cleavage, which is only now becoming apparent and which may be useful in the future as a therapeutic.

Aβ-reactive T cells increase with age

Monsonego et al. (4) found that some healthy, elderly individuals, as well as individuals with AD, contain elevated baseline levels of Aβ-reactive T cells. While the general trend is toward a diminished immune response with aging, this demonstrates a selective increase in Aβ-reactive T cells in older individuals with and without dementia. The reason for this selective expansion of Aβ-reactive T cells in elderly individuals remains unclear. It is often presumed that cognitively normal middle-aged and elderly individuals are similar in that they lack AD pathology; however, Aβ deposition in plaques appears to begin about 10–20 years prior to the onset of even the earliest symptoms suggestive of dementia due to AD (6). This means that some cognitively normal elderly subjects in this study likely possessed aggregated Aβ deposits in the brain, while it is also likely that most middle-aged individuals (younger than age 50) did not have AD pathology. One interesting possibility is that this change in T cell population is a response to the presence of Aβ aggregates even in the absence of dementia. The conformation of aggregated Aβ in AD is predominantly as β-sheets, whereas the soluble Aβ present in blood and cerebral spinal fluid has little or no β-sheet structure. Perhaps, this conformational change in endogenous Aβ stimulates a T cell response. Future studies will be necessary to determine if the peripheral T cell population correlates to CNS pathology or future AD symptoms (i.e., an antecedent bio-marker).

T cells, Aβ, and CNS inflammation

While speculative, individuals with elevated Aβ-reactive T cells may host a greater immune response to an active immunization with Aβ than someone who lacks this T cell change. The positive effects of Aβ immunization in mouse models (e.g., decreased plaque burden, behavioral improvement) appear to be mediated by antibodies, not the cellular response (7–10). Thus, augmentation of the production of anti-Aβ antibodies is likely to be beneficial. However, in the first trial of active Aβ immunization in AD patients, about 5% of individuals developed a side effect of CNS inflammation. There is evidence that this complication following active Aβ immunization is due to a T cell response (11). It therefore seems logical that minimizing certain aspects of T cell activation would decrease the likelihood of CNS inflammation. Consequently, it may be useful in future vaccination strategies to either exclude subjects that have already demonstrated a substantial T cell reaction to Aβ or to consider these subjects only for passive immunization. Monsonego and colleagues (4) found that the epitopes for Aβ-reactive T cells in humans are primarily amino acids 16–42. Interestingly, however, in studies of active immunization of humans and of mouse models of AD, the primary epitope to which antibodies are generated are amino acids 1–12 (12, 13). Because the cellular and humoral immune responses appear to have distinct, dominant epitopes, perhaps an antigen and adjuvant combination can be designed that favor a humoral immune response over a T cell response.

Certain NSAIDs decrease Aβ production

Many pathological studies have shown evidence of an inflammatory response (gliosis, increased cytokines) surrounding Aβ deposits in the AD brain. It is thought that this response may result in increased neuronal injury, which suggests the possibility that decreasing this response may be beneficial. In light of this, it is of interest that retrospective, epidemiological studies show that NSAID use is associated with a decreased risk of developing AD. Herein, Eriksen and colleagues (5) further define a different molecular mechanism that may be relevant to this relationship. It appears that certain NSAIDs, potentially in a novel, direct interaction with the γ-secretase complex, can alter APP cleavage and the subsequent species of Aβ produced. Eriksen et al. screened 18 NSAID compounds, including several enantiomers that do not inhibit COX. Interestingly, though structurally similar, these compounds can have different effects on what species of Aβ is produced; some decrease Aβ42, while others decrease Aβ40. The mechanism may be via a direct effect on the γ-secretase complex, which presumably causes a subtle conformational change and alters APP cleavage (Figure 1). In future studies, it will be important to investigate the molecular details of the NSAID/γ-secretase complex interaction. In addition, the drugs most effective in decreasing Aβ levels in humans will need to be determined.

Figure 1

Model of how certain NSAIDs decrease Aβ42 production. NSAIDs may directly bind to the γ-secretase complex and alter APP processing to decrease Aβ42 production and also change production of other Aβ species.

Eriksen and colleagues (5) have focused on decreasing the more aggregation-prone Aβ42 species in order to potentially treat AD. Another potential treatment avenue, however, is to decrease both Aβ42, as well as other species such as Aβ40, the peptide that builds up extensively in cerebral amyloid angiopathy (CAA). If an NSAID compound, or derivative, could be designed to decrease both pathological species of Aβ, it may benefit both diseases. While an important aim is to find a drug to decrease Aβ42, it will be important not to increase Aβ40 levels as a consequence. This could potentially lead to increased risk for developing CAA and its consequences such as hemorrhage.

These studies provide exciting new insights and avenues for AD treatment by suggesting improvements in current vaccination strategies or by furthering our understanding of how NSAIDs alter Aβ42 production. While it is not going to be easy, there remains much hope that the amyloid hypothesis of AD will be tested and that truly effective therapies for AD can be developed.

Footnotes

Conflict of interest: The authors have declared that no conflict of interest exists.

Nonstandard abbreviations used: Alzheimer disease (AD); amyloid β protein (Aβ); amyloid precursor protein (APP); cerebral amyloid angiopathy (CAA).

References

1. Schenk, D, et al. Immunization with amyloid-β attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature. 1999. 400:173-177.
   View this article via: PubMed CrossRef Google Scholar
2. DeMattos, RB, et al. Peripheral anti-A beta antibody alters CNS and plasma A beta clearance and decreases brain A beta burden in a mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A. 2001. 98:8850-8855.
   View this article via: PubMed CrossRef Google Scholar
3. Bard, F, et al. Peripherally administered antibodies against amyloid β-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nat. Med. 2000. 6:916-919.
   View this article via: PubMed CrossRef Google Scholar
4. Monsonego, A, et al. Increased T cell reactivity to amyloid β protein in older humans and patients with Alzheimer disease. J. Clin. Invest. 2003. 112:415-422. doi:10.1172/JCI200318104.
   View this article via: JCI PubMed Google Scholar
5. Eriksen, JL, et al. NSAIDs and enantiomers of flurbiprofen target γ-secretase and lower Aβ42 in vivo. J. Clin. Invest. 2003. 112:440-449. doi:10.1172/JCI200318162.
   View this article via: JCI PubMed Google Scholar
6. Morris, JC, Price, AL. Pathologic correlates of nondemented aging, mild cognitive impairment, and early-stage Alzheimer’s disease. J. Mol. Neurosci. 2001. 17:101-118.
   View this article via: PubMed CrossRef Google Scholar
7. Morgan, D, et al. A beta peptide vaccination prevents memory loss in an animal model of Alzheimer’s disease. Nature. 2000. 408:982-985.
   View this article via: PubMed CrossRef Google Scholar
8. Janus, C, et al. A beta peptide immunization reduces behavioural impairment and plaques in a model of Alzheimer’s disease. Nature. 2000. 408:979-982.
   View this article via: PubMed CrossRef Google Scholar
9. Dodart, JC, et al. Immunization reverses memory deficits without reducing brain Abeta burden in Alzheimer’s disease model. Nat. Neurosci. 2002. 5:452-457.
   View this article via: PubMed Google Scholar
10. Kotilinek, LA, et al. Reversible memory loss in a mouse transgenic model of Alzheimer’s disease. J. Neurosci. 2002. 22:6331-6335.
    View this article via: PubMed Google Scholar
11. Nicoll, JA, et al. Neuropathology of human Alzheimer disease after immunization with amyloid-β peptide: a case report. Nat. Med. 2003. 9:448-452.
    View this article via: PubMed CrossRef Google Scholar
12. Town, T, et al. Characterization of murine immunoglobulin G antibodies against human amyloid-β1-42. Neurosci. Lett. 2001. 307:101-104.
    View this article via: PubMed CrossRef Google Scholar
13. Hock, C, et al. Generation of antibodies specific for β-amyloid by vaccination of patients with Alzheimer disease. Nat. Med. 2002. 8:1270-1275.
    View this article via: PubMed CrossRef Google Scholar