Last July, Dale Schenk and colleagues from Elan Pharmaceuticals reported the remarkable observation that simple immunization of PDAPP transgenic mice with amyloid-forming peptide sequences both prevents plaque formation and ameliorates existent plaques in brain (Abstract). PDAPP transgenic mice express a human amyloid precursor protein (APP) containing the FAD-associated V717F mutation under the control of the platelet-derived growth factor promoter. These mice show an age-dependent accumulation of extracellular amyloid plaques and an increase in astrocytosis. According to the amyloid hypothesis, amyloid buildup in brain is the primary cause of cognitive dysfunction in Alzheimer's disease (AD). Consequently, removal of such amyloid plaques should have beneficial effects in patients with AD. Schenk's results, therefore, suggested that it might be possible to remove amyloid plaques from human brain if similar immunization worked in humans.
At the meeting, Shenk reported on safety issues relating to the use of Aβ vaccination in animals, the mechanism by which vaccination removes plaques, and his group's progress towards developing a human Aβ vaccine. The safety of Aβ was studied in four different animal models: Swiss Webster mice, guinea pigs, rabbits, and monkeys. Schenk reported no adverse clinical manifestations related to Aβ immunization (including reactions at injection sites, overall clinical status, histopathology and effects on body weight). Moreover, in PDAPP transgenic mice, Aβ immunization reduced extracellular buildup of plaques but did not affect APP protein expression levels. Their results suggest Aβ immunization should not affect the normal physiological functions of APP, and that the immunization is selective for extracellular Aβ.
Schenk next showed that humoral-mediated (antibody-mediated) response, rather than T cell mediated response, is the most likely mechanism by which Aβ immunization removes plaques from brain. To prove this, they purified IgG antibodies from Aβ-injected animals and reinjected the purified antibodies into PDAPP transgenic animals at one-week intervals for six months. Amyloid burden was significantly reduced in the animals injected with IgG obtained from the mice injected with Aβ compared to animals injected with IgG purified from mice injected with an irrelevant antigen. Shenk subsequently showed that monoclonal antibodies specific for Aβ could also reduce plaque burden when injected into PDAPP transgenic animals.
The mechanism by which these anti-Aβ antibodies function appears to involve direct binding to antibody to amyloid within brain. The data suggest that a small amount of antibody must pass through the blood-brain barrier and enter the central nervous system. Schenk and colleagues used in vivo and cell culture systems to further show that the binding of antibody to amyloid deposits correlates with removal of those deposits by phagocytosis involving microglia. These results suggest that it may be possible to remove existent amyloid plaques in human brain using anti-Aβ antibody injections. The use of antibody injection therapy to remove amyloid deposits in human brain may be important since it is well-established that the immune response and antibody repertoire are both reduced during aging. Thus, for elderly AD patients who do not elicit an effective immune response, the use of antibody injections may circumvent potential problems relating to this inability of patients to elicit an Aβ immune response.
Finally, Schenk reported that his group had begun Phase I clinical trials, in both in the USA and the United Kingdom, to test Aβ immunization in patients suffering from mild cognitive impairment. So far, they report that vaccination of such patients with the Aβ formulation into humans is well tolerated and appears to be safe. The AD research community and world audiences now wait in anticipation for outcome of these Aβ immunization trials. It would be remarkable if immunization, which has been so effective in eradicating a number of human diseases such as smallpox and polio, may yet again prove to be a simple and effective way to prevent one of the most common and devastating human neurodegenerative disorders.
Several other groups have now replicated these Aβ immunization studies and have found that the procedure is effective in reducing amyloid plaques in different transgenic mouse models. At the special "Hot Topics" session held on July 13, David Morgan of the University of South Florida reported similar findings and, more remarkably, that Aβ vaccination also improved short-term learning behavior of APP transgenic mice despite little alteration of Aβ-plaque load in brain. In their studies, they used mice that were doubly transgenic for APP and PS1 (containing familial AD-linked mutations in both genes). These mice develop abundant Aβ plaques in an age-dependent fashion. Morgan and colleagues devised a modified Morris maze test, which typically involves mice learning to swim to one of several platforms on which food can be found. Instead, in their modified test they constructed six swimming alleys, akin to a maze, of which one ended at a hidden platform. The number of errors made by the mice in locating the hidden platform was used to determine if the mice had intact or defective memory. In essence their test measures short term spatial memory. After a learning period, normal non-transgenic mice are able to find the platform with less than one error in the fifth trial. In contrast, APP-PS1 transgenic mice at 15 months of age, which by this time have numerous plaques in brain, perform poorly in this test.
Morgan's group then studied the effects of Aβ immunization on this short-term spatial memory task. Transgenic APP mice immunized with Aβ for several months showed no deficits in learning, and were able to efficiently find the platform. In contrast, age-matched APP transgenic mice injected with control carrier keyhole limpet hemocyanin polypeptide, performed poorly in the test. Immunization with Aβ improved maze test performance in both APP and PS1 doubly transgenic mice as well as APP single transgenic mice. Surprisingly, even APP transgenic mice immunized with Aβ for a very short period also showed improved memory despite no noticeable change in Aβ accumulation within their brains. Their results suggest that, at least by this one memory test based on performance, Aβ immunization improves memory and that this can occur in the absence of any change in Aβ plaque deposition. Additional studies are needed to determine if indeed Aβ immunization improves memory assayed by other criteria, and whether Aβ plaque removal from brain is essential for improving memory.
Robert Malinow from Cold Spring Harbor Laboratories presented evidence (345) that β-amyloid is secreted during neuronal activity and may function in a negative feedback loop to block further synaptic transmission. Malinow showed that treatment of organotypic hippocampal slice cultures with agents that block neuronal activity, such as tetrodotoxin and high concentrations of magnesium, cause a 50%-70% reduction in secretion of Aβ into the medium. In contrast, agents that increased neuronal activity also increased Aβ secretion. He also demonstrated that wild-type and FAD mutant APP genes cause a 30% reduction in synaptic transmission when transduced into neurons. In contrast to these genes, both of which can be processed to release Aβ into the medium, transduction of a mutant APP gene, which cannot be processed into Aβ, has no effect on synaptic transmission. These results suggest that there is a complex interplay of APP processing and indirectly suggests that APP cleavage products have direct effects in modulating synaptic activity. The diverse roles of APP, including Aβ and its other cleavage products have been well-established, especially of their roles in growth control, calcium regulation, and apoptosis. The success of Aβ therapies will depend in large part on whether Aβ can be successfully eliminated from brain without compromising its other functions.—Mervyn J. Monteiro
Note: by Dominic Walsh
The aim of these studies is primarily to determine the safety of the vaccine in humans and to optimize the formulation so as to induce a good immune response. However, it is the outcome with respect to changes in mental status that will be awaited with great interest.
Although already in clinical trials there are a number of burning questions, which remain unanswered with regard to the usefulness of Aβ immunization. The PD-APP mouse is a good model for amyloid deposition; however, it is not a particularly good model for AD. While immunization is capable of removing amyloid plaques, what effect if any will it have on prefibrillar species? (For a discussion of the role of prefibrillar species in neurodegeneration, see the commentary on Peter Lansbury’s presentation.) It is possible (although I hope unlikely) that immunization by removing fibrils may effectively elevate the level of prefibrillar species and thus accentuate rather than alleviate Aβ-induced toxicity. Another, but less worrisome, possibility is that, if intracellular Aβ either directly or indirectly induces toxicity, then vaccination will not be able to alter that process. Over the coming months these important issues should be addressed in experimental models.—Dominic Walsh
No Available References
No Available Further Reading