- Email: [email protected]
“Emerging Alzheimer’s disease therapies: focusing on the future”
Neurobiology of Aging 23 (2002) 985–990
Brief overview of the symposium
“Emerging Alzheimer’s disease therapies: focusing on the future” John Q. Trojanowski∗ Department of Pathology and Laboratory Medicine, The Center for Neurodegenerative Disease Research, University of Pennsylvania School of Medicine, HUP-Maloney Bldg., 3rd Floor, 36th and Spruce Streets, Philadelphia, PA 19104-4283, USA Received 1 April 2002; accepted 2 April 2002
Abstract The Center for Neurodegenerative Disease Research (CNDR) organized a 1 day symposium entitled “Emerging Alzheimer’s disease Therapies: Focusing On The Future” on November 7th, 2001 at the University of Pennsylvania in Philadelphia, PA. The agenda (Fig. 1) focused on novel therapies for Alzheimer’s disease (AD) designed to prevent/eliminate A deposits in the brains of AD patients. While fibrillar A deposits known as senile plaques (SPs) and intraneuronal tau fibrils known as neurofibrillary tangles (NFTs) are diagnostic of AD, >50% of patients with familial or sporadic AD as well as elderly Down’s syndrome patients with AD harbor a third type of brain amyloid known as Lewy bodies formed by intraneuronal alpha-synuclein fibrils [1,6]. Thus, AD is a “triple brain amyloidosis” since three different proteins (tau, alpha-synuclein) or peptide fragments (A) of a larger A precursor protein (APP) fibrillize and aggregate into pathological deposits of amyloid within (NFTs, LBs) and outside (SPs) neurons in AD brains. The symposium is summarized here followed by reviews from symposium speakers who describe potential anti-A therapies some of which are in clinical trials [2,3]. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Neurodegenerative diseases; A; Brain amyloidosis; Drug discovery
Immune Therapy For Alzheimer’s Disease Ivan Lieberburg, Ph.D., M.D., Élan Corporation “Systemic Immunotherapy for Alzheimer’s Disease” Dr. Lieberburg summarized evidence from a number of groups that has been developed over the past 15 years supporting the role of fibrillar A amyloid deposits or various forms of the A peptide as being causal in the onset and/or progression of AD. Briefly, various A peptides, but in particular the presumably more toxic 42 amino acid long peptide (A1–42), are derived from APPs, and they are released from APP with and without further modifications following beta and gamma-secretase cleavage. Significantly, both of these enzymes are aspartyl proteases that also are targets for AD drug discovery, independent of anti-A amyloid drugs, as discussed in subsequent presentations in this symposium. Using the first transgenic mouse model of ∗
Tel.: +1-215-662-6399; fax: +1-215-349-5909. E-mail address: [email protected] (J.Q. Trojanowski).
AD-like A amyloidosis (e.g. the PDAPP line of mice) that was reported in 1995, Elan scientists discovered that immunization of young PDAPP mice with human A peptides prevented the immunized mice from developing AD-like A amyloid neuropathology at 6 months of age as well as beyond this time point throughout the life span of these mice which is normal. Moreover, similar immunization of 1 year old PDAPP mice, which have substantial A amyloid neuropathology by this age, stabilized, prevented further A deposition, or even reversed this pathology with advancing age for six months beyond the time of initial immunization. These results have been repeated and extended by several other groups using other lines of transgenic mice that develop AD-like A amyloidosis (e.g. the Tg2576 line), and Elan scientist have gone on to show that passive immunization with some, but not all, anti-A antibodies yield similar therapeutic effects in these PDAPP mice. Current thinking by Elan scientists is that the anti-A antibodies cross the blood brain barrier (BBB) into the central nervous system (CNS), bind to A fibrils in amyloid deposits and promote clearance of these plaques by brain microglia. Further, other groups have reported that human A peptide immunized
0197-4580/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 7 - 4 5 8 0 ( 0 2 ) 0 0 1 2 3 - 9
986
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Fig. 1. Emerging Alzheimer’s disease therapies: focusing on the future.
transgenic mouse models of AD-like A amyloidosis that show evidence of clearance of their age related accumulations of A amyloid also are able to learn more effectively than non-immunized transgenic mice. In their initial studies, Elan scientists conducted extensive toxicological testing of this immunotherapy in several mammalian species, and they did not detect evidence of toxicity, but clinical trial studies in human AD and control subjects have not demonstrated any correlation between the presence or absence of anti-A antibodies or levels thereof with the disease state. Elan scientists completed phase 1 clinical trials of this immunotherapy in the USA and the UK that did not show any deleterious effects of the A peptide vaccine, and they then initiated a multi-site phase 2 clinical trial study in >350 mild to moderate AD patients. However, recent reports on this clinical trial show that some of the patients have developed cerebri-
tis or meningoencephalitis and Elan recently terminated this clinical trial (for recent updates see http://www.elan.com/, and http://www.alzforum.org/). The implications of these perplexing and unanticipated complications are incompletely understood at this time, but it is important to emphasize that the cause of these complications have not yet been elucidated, and vaccine therapy using other types of A peptides (see below) or passive immune therapeutic strategies should still be considered a viable and potentially effective novel approach to treating AD A amyloidosis. However, since there is no evidence that these approaches will alter the existing tangle burden in the brains of AD patients, immune mediated removal of A deposits may leave these patients with impairments similar to those found in tangle only tauopathies such as Pick’s disease or Guam dementia.
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Cynthia Lemere, Ph.D., Harvard Medical School “Intra-nasal Immunotherapy for Alzheimer’s Disease” Dr. Lemere’s presentation summarized the efforts and progress of her group to develop intra-nasal immunotherapy for AD in transgenic mouse models of AD-like amyloidosis by delivering a human A peptide vaccine through the nasal route. She reviewed the components of the nasal epithelium immune system and then reported results similar to those of the Elan group with respect to the prevention and/or reversal of brain deposits of A amyloid in transgenic mice. However, she also reviewed data suggesting an alternative mechanism for removal of A amyloid deposit clearance involving the peripheral elimination of A from brain leading to increased levels of plasma A levels, and a redistribution of brain A through a “sink”-like effect to the plasma for subsequent clearance. Differences in the mechanisms of A clearance between Elan scientists and other groups notwithstanding, it is increasingly evident that A immunotherapy is solidly grounded in a wealth of preclinical data, but, as noted above, the outcome of ongoing clinical trials of this therapy in AD patients will determine the extent to which it will prove to be an efficacious intervention for the treatment or prevention of AD. Blas Frangione, M.D., Ph.D., New York University School of Medicine “Modification of Soluble Aβ Derivatives: An Approach to Alzheimer’s Disease Therapy” Dr. Frangione’s colleague, Dr. E.M. Sigurdsson, gave this presentation on another alternative for immunotherapy of AD. Dr. Sigurdsson reviewed recent studies of transgenic mice with AD-like brain amyloidosis that were immunized with aggregated human A1–42 derivatives showing that these mice have a reduced cerebral A amyloid burden. However, Dr. Sigurdsson noted that the use of native or unmodified human A1–42 peptides in patients may have drawbacks because, if they cross the BBB into the CNS and form toxic fibrils, they could promote or seed A fibril formation in the brain, even if they are present in very small quantities. Indeed, these investigators suggest that these safety issues are of particular concern in the elderly because older individuals do not always mount an adequate immune response to vaccines. Thus, Dr. Frangione, Dr. Sigurdsson and colleagues developed an A homolog, A1–30NH2 with polylysine attached to its N-terminus (K6A1–30) that is highly soluble, non-amyloidogenic and non-toxic in human neuronal cell culture systems. Moreover, they recently reported that immunization with K6A1–30 in 11–12 month old transgenic mice that model AD-like brain amyloidosis (Tg2576 line of mice) for 7 months reduced the brain A amyloid burden by >80%, while brain levels of soluble A1–42 were reduced by 57%. Further, rami-
987
fied microglia expressing interleukin-1 that are commonly associated with the A plaque deposits were absent in the immunized mice, and this is consistent with reduced inflammation following this vaccination protocol. While additional studies of this method of immunotherapy are in progress in these transgenic mice, based on their data, these investigators suggest that immunization with soluble A derivatives represents a potentially safer therapeutic approach to reduce A amyloid burden in AD patients. Significantly, the complications that arose in the Elan vaccine clinical trial were reported several months after this symposium, and it remains to be determined if the concerns mentioned by Sigurdsson and colleagues account for any of these adverse events. Hormonal Modulation Of A Levels As Therapy For Alzheimer’s Disease Samuel Gandy, M.D., Ph.D., Thomas Jefferson Medical School “Gonadal Hormones Control Alzheimer’s β-Amyloid Levels In Vivo” Dr. Gandy began his presentation by pointing out that 14 of 15 epidemiological studies reported since 1990 have shown that hormone replacement therapy (HRT) reduced the relative risk of AD in postmenopausal women by about 50%. Thus, Dr. Gandy and colleagues have hypothesized that HRT might modulate AD risk through hormonal control of A metabolism, and they have tested this hypothesis by examining the effect of ovariectomy on A levels in the brains of experimental animals and in the circulation of elderly men with prostate cancer who were treated with anti-gonadal therapy. Notably, compared to controls, brain levels of A40 and A42 were increased by about 150% in guinea pigs subjected to ovariectomy, but brain levels of A returned nearly to control values when the ovariectomized guinea pigs were treated orally with 1 or 5 mg/kg per day or 17-estradiol as HRT. Dr. Gandy and colleagues also have extended these studies to a model of the human clinical situation by examining six men who were undergoing therapeutic androgen suppression therapy for treatment of prostate cancer, and Gandy’s team reported that plasma levels of A doubled in association with declining levels of circulating testosterone and estrogen. Further, the elevated plasma A levels persisted for at least the first 6 months of therapy. Thus, these data suggest that gonadal hormones play a key role in controlling A levels in vivo in experimental animals as well as in human patients. However, other studies report a lack of efficacy in treating advanced AD with HRT in recent clinical trials using estrogens, but it remains possible that earlier HRT intervention in individuals with mild cognitive impairment, a prodromal phase of AD, or even in early AD, could show some therapeutic benefit by slowing disease progression.
988
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Amyloid Plaque And A Fibril Disrupting Therapy For Alzheimer’s Disease Ashley Bush, M.D., Ph.D., Harvard Medical School “Metal Complexing Agents as Therapies for Alzheimer’s Disease” According to Dr. Bush, A is rapidly precipitated by Zn2+ at low physiological concentrations, and Cu2+ and Fe3+ also induce A aggregation. Further, Dr. Bush noted that Zn/Cu-selective chelators markedly enhance the solubilization of A deposits from post-mortem AD brains, while he and his colleagues also have purified A from human brain to demonstrate that it is metallated with Zn and Cu. Hence, Dr. Bush asserted that A should be regarded as a metalloprotein. Moreover, zinc, copper and iron are markedly elevated in the AD brain, and they are especially highly concentrated in A deposits in humans and in APP transgenic mice that model AD-like A amyloidosis. Significantly, when it binds Cu2+ or Fe3+ , A reduces the metal ions and produces H2 O2 by electron transfer to O2 . This is important because of growing evidence for oxidation injury in AD, and Dr. Bush speculated that A cytotoxicity could be mediated by Cu2+ :A interactions that generate H2 O2 , but this process might be blocked or attenuated by chelators of zinc and/or copper. However, since complications from the Elan vaccine clinical trial were reported after this symposium, it is not yet clear if the mechanisms alluded to here by Dr. Bush account for any of these adverse events. For the reasons summarized above, Dr. Bush and his colleagues have pursued metal-A interactions as a target for therapeutics, and they recently reported that A1–42 possesses an attomolar affinity Cu2+ binding site that is among the strongest known binding affinities, and this is similar to the affinity of Cu2+ for the antioxidant enzyme superoxide dismutase 1. Dr. Bush and his team also found that the Cu2+ -mediated H2 O2 production by A is suppressed by co-binding of Zn2+ , and indeed, despite precipitating A into amyloid, Zn2+ rescues A neurotoxicity in cell culture systems. These and other studies have led Dr. Bush and his colleagues to conclude that AD amyloid plaques may represent the redox-silencing and entombment of A by zinc. Hence, they speculate that SPs may not be the pathogenic culprit in AD, and there is evidence to suggest that the toxic forms of A may be the soluble species or the diffused non-fibrillar A collections in the brains of cognitively impaired patients. On the basis of growing information about Zn2+ and Cu2+ interactions with A mediating the biochemistry of A, Dr. Bush and colleagues recently embarked on a trial of copper-chelators to attempt to inhibit A accumulation in Tg2576 transgenic mice that model AD-like A amyloidosis. Treatment with clioquinol (CQ), a retired antibiotic and a bioavailable Cu/Zn chelator, induced a 49% decrease in brain A deposition in a blinded study of Tg2576 mice treated orally with CQ for 9 weeks. Further, there was no ev-
idence of neurotoxicity or increased non-amyloid pathology. This drug may work by a combined action that facilitates disaggregation of the A collections, while also inhibiting H2 O2 production. Importantly, CQ treatment did not induce a loss in metal levels systemically, probably because it is a relatively weak chelator, and these metals may be redistributed rather than excreted following CQ administration. Thus, CQ treatment appears to be a potent inhibitor of A accumulation. CQ may be the first credible drug candidate based on the amyloid hypothesis of AD, and Dr. Bush reported that a phase II double-blind clinical trial on the effects of CQ in AD patients is currently in progress. Virginia M.-Y. Lee, Ph.D., University of Pennsylvania School of Medicine “Amyloid Binding Ligands as Alzheimer’s Disease Therapies” Dr. Lee summarized studies conducted by investigators in CNDR in collaboration with the laboratory of Dr. H. Kung and Dr. M.-P. Kung at Penn that seek to identify new therapies for AD. The therapeutic targets of the studies they have conducted and propose for the future are the extracellular SPs formed by fibrillar A. Dr. Lee noted that, although, A vaccines and inhibitors of amyloidogenic secretases are potential AD therapies, multifaceted strategies may be needed to effectively interrupt A amyloidosis and prevent/arrest AD. Thus, she and her colleagues have specifically targeted the inhibition of A fibrillization as a potential therapy for AD. Accordingly, Dr. Lee noted that certain amyloid-binding molecules, such as Congo red (CR) and chrysamine G (CG), can indeed arrest the formation of A fibrils; however, CR and CG are unsuitable for AD therapy because they do not cross the BBB. Therefore, working with the Kung lab at Penn, Dr. Lee and colleagues in CNDR have generated novel CG derivatives and other small molecules that specifically recognize fibrillar A in vitro, arrest the formation of A fibrils, and cross the BBB of transgenic mice that model AD amyloidosis. As proof of their ability to cross the BBB and of their high specificity for A fibrils in vivo, Dr. Lee and colleagues showed that following intravenous injection in transgenic mice these compounds specifically label AD-like brain deposits of fibrillar A. Building on these preliminary data, they propose to test the hypothesis that small amyloid-binding compounds that inhibit A fibrillization in vitro and cross the BBB are potential AD therapeutic agents. To accomplish this, Dr. Lee described a multi-disciplinary research program, including the design and synthesis of novel amyloid-binding compounds, biophysical, biochemical, molecular biological and ultrastructural studies using cell culture and animal models of AD-like A amyloidosis to develop such novel amyloid-binding compounds and evaluate their potential as therapeutic agents for the treatment of AD. While these studies are still at the preclinical stage in model systems, amyloid disrupting compounds hold promise not only for targeting A amyloidosis, but also
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
other types of brain amyloid including those due to accumulations of tau and alpha-synuclein fibrils. Secretase Inhibition To Reduce A Levels As Therapy For Alzheimer’s Disease Martin Citron, Ph.D., Amgen Incorporated “Beta-secretase Inhibitor Therapy for Alzheimer’s Disease” Dr. Citron began his presentation by emphasizing that the cerebral deposition of A is an early and critical feature of AD, and that A is released from APP by the sequential action of two proteases, beta-secretase and gamma-secretase. Thus, Dr. Citron noted that these proteases are prime targets for therapeutic intervention, and for this reason he and his colleagues recently cloned a novel aspartic protease, BACE1, with all the known properties of the beta-secretase that is responsible for the release of A from APP. Dr. Citron then briefly summarized how he and his colleagues conducted studies that led to the initial identification of BACE1, and he also described some of the distinct properties of BACE1. He then went on to review the crystal structure of BACE1 as well as data from studies conducted by his group on BACE1 knockout mice that he and his colleagues recently generated. This presentation concluded with a discussion of the pros and cons of BACE1 inhibition as a therapeutic strategy compared to other amyloid therapies. While a lively discussion of this approach to AD therapy followed, the approach is still largely theoretical since there is little published preclinical data from studies in model systems upon which to base solid predictions on the relative merits of this approach to the therapy of AD compared to the other therapies that currently target the levels of a A and the burden of A deposits in the CNS directly. Yue-Ming Li, Ph.D., Merck and Company “Gamma-secretase: Identification and Implications for Alzheimer’s Disease Therapy” Dr. Li began his presentation by reviewing the known biology of gamma-secretase, a membrane-bound protease that cleaves within the transmembrane region of APP to generate the C-termini of the two predominant forms of A known as A40 and A42, which are major constituents of SPs in AD brains. Further, Dr. Li pointed out that presenilin 1 (PS1) and presenilin 2 (PS2) are polytopic membrane spanning proteins that harbor mutations which give rise to early onset familial AD. Dr. Li then reviewed recent biochemical studies which provide compelling evidence that presenilins are novel aspartyl proteases that mediate gamma-secretase activity when engaged in the putative gamma-secretase macromolecular complex. PS1 and PS2 activities can be discriminated from one another on the basis of their susceptibility to inhibition by a potent gamma-secretase inhibitor.
989
Thus, Dr. Li concluded that presenilin/gamma-secretase or the complex that mediates this enzyme activity is a potential target for AD therapy and plays an important role in regulated intramembrane proteolysis. However, as with the efforts described above to develop beta-secretase inhibitor therapy for AD, there is little published preclinical data on gamma-secretase inhibitors upon which to base solid predictions on the merits of this approach to the therapy of AD. Summary And Conclusions On Emerging Therapies For Alzheimer’s Disease Sangram Sisodia, Ph.D., University of Chicago “Current Status Of Emerging Therapies For Alzheimer’s Disease” Dr. Sangram Sisodia provided a summary and critique of the foregoing scientific presentations wherein he outlined the strengths and weaknesses of each of the approaches discussed at the symposium. A number of these points, as well as others made by some of the >300 symposium participants are incorporated into the summaries of each of the presentations described above. Christopher Clark, M.D. and Jason Karlawish, M.D., University of Pennsylvania “What Are The Next Steps? The Challenges Of Transforming Laboratory Advances Into Treatments For Patients” Finally, the symposium was concluded by a stimulating session that was designed for attendance by scientists and clinicians as well as for the caregivers and families of AD patients. This session began with tandem presentations by Dr. Clark and Dr. Karlawish who summarized the key elements that are needed for the ethical and informative conduct of human clinical trials of new AD therapies. They noted a number of unfinished tasks for the design of such trials including issues such as informed consent for trials involving subjects who may not be able to give consent themselves, but instead require surrogates for this, and the importance of the “dyad” of patient and caregiver in judging the merits and efficacy of new drugs for AD. They also emphasized the need for better outcome measures for testing new drugs including better imaging strategies and reliable AD biomarkers that can be used in conjunction with clinical outcome measures to assess the response of the disease process to therapeutic interventions, and they discussed the types of patient populations that are most suitable for evaluating new AD therapies in a timely and informative manner. Notably, in addition to the efforts of industry to carry out clinical trials of proprietary drugs for AD, the National Institute on Aging of the National Institutes of Health has established an infrastructure of AD Centers around the USA for this and a funding mechanism through the AD Clinical Studies program to conduct clinical trials of other potential therapies for AD [4,5,7]. This
990
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
presentation stimulated robust discussion and dialogue among all the participants on the prospects of translating laboratory advances in AD research into effective therapies for AD patients.
the research leading to the identification of better therapies for AD would not be possible without the support of the families of our AD patients. References
Acknowledgments The co-directors of CNDR who organized this symposium kindly acknowledge a grant from Nastech Pharmaceutical Co., Inc. to support this meeting, as well as generous support for the meeting from Janssen Pharmaceutica. Other support came from the Memory Disorder Clinic, the Institute on Aging and the National Institute of Aging AD Center Core at Penn as well as from all of the members of CNDR, but especially Ms. Gayle Viale, who provided key design, logistical and administrative support. In addition, the National Institute on Aging of the National Institutes of Health, and the Oxford Foundation made the symposium possible. Dr. V.M.-Y. Lee is the recipient of the John H. Ware 3rd Chair for Alzheimer’s Disease Research. Additional information on AD and related neurodegenerative diseases can be obtained at http://www.med.upenn.edu/cndr, the CNDR website. Finally, all of the speakers are thanked for their insightful presentations and discussions at the meeting, and
[1] Galvin JE, Lee VM-Y, Trojanowski JQ. Synucleinopathies: Clinical and pathological implications. Arch Neurol 2001;58:186–90. [2] Irizarry MC, Hyman BT. Alzheimer disease therapeutics. J Neuropath Exper Neurol 2001;60:923–8. [3] Mayeux R, Sano M. Treatment of Alzheimner’s disease. New Engl J Med 1999;341:1670–9. [4] Morris JC, Ernesto C, Schafer K, Coats M, Leon S, Sano M, et al. Clinical dementia rating training and reliability in multicenter studies: The Alzheimer’s disease cooperative study experience. Neurol 1997;48:1508–10. [5] Sano M, Ernesto C, Klauber MR, Schafer K, Woodbury P, Thomas R, et al. Ratinale and design of a multicenter study of selegiline and alpha-tocopherol in the treatment of Alzheimer disease using novel clinical outcomes Alzheimer’s disease cooperative study. Alzheimer Dis Assoc Disord 1996;10:132–40. [6] Trojanowski JQ, Lee VM-Y. “Fatal attractions” of proteins: A comprehensive hypothetical mechanism underlying Alzheimer’s disease and other neurodegenerative disorders. Ann New York Acad Sci 2000;924:62–7. [7] Trojanowski JQ. Alzheimer’s disease centers and the dementias of aging program of the national institute on aging: A brief overview. J Alzheimer’s Dis 2001;3:249–51.
Brief overview of the symposium
“Emerging Alzheimer’s disease therapies: focusing on the future” John Q. Trojanowski∗ Department of Pathology and Laboratory Medicine, The Center for Neurodegenerative Disease Research, University of Pennsylvania School of Medicine, HUP-Maloney Bldg., 3rd Floor, 36th and Spruce Streets, Philadelphia, PA 19104-4283, USA Received 1 April 2002; accepted 2 April 2002
Abstract The Center for Neurodegenerative Disease Research (CNDR) organized a 1 day symposium entitled “Emerging Alzheimer’s disease Therapies: Focusing On The Future” on November 7th, 2001 at the University of Pennsylvania in Philadelphia, PA. The agenda (Fig. 1) focused on novel therapies for Alzheimer’s disease (AD) designed to prevent/eliminate A deposits in the brains of AD patients. While fibrillar A deposits known as senile plaques (SPs) and intraneuronal tau fibrils known as neurofibrillary tangles (NFTs) are diagnostic of AD, >50% of patients with familial or sporadic AD as well as elderly Down’s syndrome patients with AD harbor a third type of brain amyloid known as Lewy bodies formed by intraneuronal alpha-synuclein fibrils [1,6]. Thus, AD is a “triple brain amyloidosis” since three different proteins (tau, alpha-synuclein) or peptide fragments (A) of a larger A precursor protein (APP) fibrillize and aggregate into pathological deposits of amyloid within (NFTs, LBs) and outside (SPs) neurons in AD brains. The symposium is summarized here followed by reviews from symposium speakers who describe potential anti-A therapies some of which are in clinical trials [2,3]. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Neurodegenerative diseases; A; Brain amyloidosis; Drug discovery
Immune Therapy For Alzheimer’s Disease Ivan Lieberburg, Ph.D., M.D., Élan Corporation “Systemic Immunotherapy for Alzheimer’s Disease” Dr. Lieberburg summarized evidence from a number of groups that has been developed over the past 15 years supporting the role of fibrillar A amyloid deposits or various forms of the A peptide as being causal in the onset and/or progression of AD. Briefly, various A peptides, but in particular the presumably more toxic 42 amino acid long peptide (A1–42), are derived from APPs, and they are released from APP with and without further modifications following beta and gamma-secretase cleavage. Significantly, both of these enzymes are aspartyl proteases that also are targets for AD drug discovery, independent of anti-A amyloid drugs, as discussed in subsequent presentations in this symposium. Using the first transgenic mouse model of ∗
Tel.: +1-215-662-6399; fax: +1-215-349-5909. E-mail address: [email protected] (J.Q. Trojanowski).
AD-like A amyloidosis (e.g. the PDAPP line of mice) that was reported in 1995, Elan scientists discovered that immunization of young PDAPP mice with human A peptides prevented the immunized mice from developing AD-like A amyloid neuropathology at 6 months of age as well as beyond this time point throughout the life span of these mice which is normal. Moreover, similar immunization of 1 year old PDAPP mice, which have substantial A amyloid neuropathology by this age, stabilized, prevented further A deposition, or even reversed this pathology with advancing age for six months beyond the time of initial immunization. These results have been repeated and extended by several other groups using other lines of transgenic mice that develop AD-like A amyloidosis (e.g. the Tg2576 line), and Elan scientist have gone on to show that passive immunization with some, but not all, anti-A antibodies yield similar therapeutic effects in these PDAPP mice. Current thinking by Elan scientists is that the anti-A antibodies cross the blood brain barrier (BBB) into the central nervous system (CNS), bind to A fibrils in amyloid deposits and promote clearance of these plaques by brain microglia. Further, other groups have reported that human A peptide immunized
0197-4580/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 7 - 4 5 8 0 ( 0 2 ) 0 0 1 2 3 - 9
986
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Fig. 1. Emerging Alzheimer’s disease therapies: focusing on the future.
transgenic mouse models of AD-like A amyloidosis that show evidence of clearance of their age related accumulations of A amyloid also are able to learn more effectively than non-immunized transgenic mice. In their initial studies, Elan scientists conducted extensive toxicological testing of this immunotherapy in several mammalian species, and they did not detect evidence of toxicity, but clinical trial studies in human AD and control subjects have not demonstrated any correlation between the presence or absence of anti-A antibodies or levels thereof with the disease state. Elan scientists completed phase 1 clinical trials of this immunotherapy in the USA and the UK that did not show any deleterious effects of the A peptide vaccine, and they then initiated a multi-site phase 2 clinical trial study in >350 mild to moderate AD patients. However, recent reports on this clinical trial show that some of the patients have developed cerebri-
tis or meningoencephalitis and Elan recently terminated this clinical trial (for recent updates see http://www.elan.com/, and http://www.alzforum.org/). The implications of these perplexing and unanticipated complications are incompletely understood at this time, but it is important to emphasize that the cause of these complications have not yet been elucidated, and vaccine therapy using other types of A peptides (see below) or passive immune therapeutic strategies should still be considered a viable and potentially effective novel approach to treating AD A amyloidosis. However, since there is no evidence that these approaches will alter the existing tangle burden in the brains of AD patients, immune mediated removal of A deposits may leave these patients with impairments similar to those found in tangle only tauopathies such as Pick’s disease or Guam dementia.
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Cynthia Lemere, Ph.D., Harvard Medical School “Intra-nasal Immunotherapy for Alzheimer’s Disease” Dr. Lemere’s presentation summarized the efforts and progress of her group to develop intra-nasal immunotherapy for AD in transgenic mouse models of AD-like amyloidosis by delivering a human A peptide vaccine through the nasal route. She reviewed the components of the nasal epithelium immune system and then reported results similar to those of the Elan group with respect to the prevention and/or reversal of brain deposits of A amyloid in transgenic mice. However, she also reviewed data suggesting an alternative mechanism for removal of A amyloid deposit clearance involving the peripheral elimination of A from brain leading to increased levels of plasma A levels, and a redistribution of brain A through a “sink”-like effect to the plasma for subsequent clearance. Differences in the mechanisms of A clearance between Elan scientists and other groups notwithstanding, it is increasingly evident that A immunotherapy is solidly grounded in a wealth of preclinical data, but, as noted above, the outcome of ongoing clinical trials of this therapy in AD patients will determine the extent to which it will prove to be an efficacious intervention for the treatment or prevention of AD. Blas Frangione, M.D., Ph.D., New York University School of Medicine “Modification of Soluble Aβ Derivatives: An Approach to Alzheimer’s Disease Therapy” Dr. Frangione’s colleague, Dr. E.M. Sigurdsson, gave this presentation on another alternative for immunotherapy of AD. Dr. Sigurdsson reviewed recent studies of transgenic mice with AD-like brain amyloidosis that were immunized with aggregated human A1–42 derivatives showing that these mice have a reduced cerebral A amyloid burden. However, Dr. Sigurdsson noted that the use of native or unmodified human A1–42 peptides in patients may have drawbacks because, if they cross the BBB into the CNS and form toxic fibrils, they could promote or seed A fibril formation in the brain, even if they are present in very small quantities. Indeed, these investigators suggest that these safety issues are of particular concern in the elderly because older individuals do not always mount an adequate immune response to vaccines. Thus, Dr. Frangione, Dr. Sigurdsson and colleagues developed an A homolog, A1–30NH2 with polylysine attached to its N-terminus (K6A1–30) that is highly soluble, non-amyloidogenic and non-toxic in human neuronal cell culture systems. Moreover, they recently reported that immunization with K6A1–30 in 11–12 month old transgenic mice that model AD-like brain amyloidosis (Tg2576 line of mice) for 7 months reduced the brain A amyloid burden by >80%, while brain levels of soluble A1–42 were reduced by 57%. Further, rami-
987
fied microglia expressing interleukin-1 that are commonly associated with the A plaque deposits were absent in the immunized mice, and this is consistent with reduced inflammation following this vaccination protocol. While additional studies of this method of immunotherapy are in progress in these transgenic mice, based on their data, these investigators suggest that immunization with soluble A derivatives represents a potentially safer therapeutic approach to reduce A amyloid burden in AD patients. Significantly, the complications that arose in the Elan vaccine clinical trial were reported several months after this symposium, and it remains to be determined if the concerns mentioned by Sigurdsson and colleagues account for any of these adverse events. Hormonal Modulation Of A Levels As Therapy For Alzheimer’s Disease Samuel Gandy, M.D., Ph.D., Thomas Jefferson Medical School “Gonadal Hormones Control Alzheimer’s β-Amyloid Levels In Vivo” Dr. Gandy began his presentation by pointing out that 14 of 15 epidemiological studies reported since 1990 have shown that hormone replacement therapy (HRT) reduced the relative risk of AD in postmenopausal women by about 50%. Thus, Dr. Gandy and colleagues have hypothesized that HRT might modulate AD risk through hormonal control of A metabolism, and they have tested this hypothesis by examining the effect of ovariectomy on A levels in the brains of experimental animals and in the circulation of elderly men with prostate cancer who were treated with anti-gonadal therapy. Notably, compared to controls, brain levels of A40 and A42 were increased by about 150% in guinea pigs subjected to ovariectomy, but brain levels of A returned nearly to control values when the ovariectomized guinea pigs were treated orally with 1 or 5 mg/kg per day or 17-estradiol as HRT. Dr. Gandy and colleagues also have extended these studies to a model of the human clinical situation by examining six men who were undergoing therapeutic androgen suppression therapy for treatment of prostate cancer, and Gandy’s team reported that plasma levels of A doubled in association with declining levels of circulating testosterone and estrogen. Further, the elevated plasma A levels persisted for at least the first 6 months of therapy. Thus, these data suggest that gonadal hormones play a key role in controlling A levels in vivo in experimental animals as well as in human patients. However, other studies report a lack of efficacy in treating advanced AD with HRT in recent clinical trials using estrogens, but it remains possible that earlier HRT intervention in individuals with mild cognitive impairment, a prodromal phase of AD, or even in early AD, could show some therapeutic benefit by slowing disease progression.
988
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
Amyloid Plaque And A Fibril Disrupting Therapy For Alzheimer’s Disease Ashley Bush, M.D., Ph.D., Harvard Medical School “Metal Complexing Agents as Therapies for Alzheimer’s Disease” According to Dr. Bush, A is rapidly precipitated by Zn2+ at low physiological concentrations, and Cu2+ and Fe3+ also induce A aggregation. Further, Dr. Bush noted that Zn/Cu-selective chelators markedly enhance the solubilization of A deposits from post-mortem AD brains, while he and his colleagues also have purified A from human brain to demonstrate that it is metallated with Zn and Cu. Hence, Dr. Bush asserted that A should be regarded as a metalloprotein. Moreover, zinc, copper and iron are markedly elevated in the AD brain, and they are especially highly concentrated in A deposits in humans and in APP transgenic mice that model AD-like A amyloidosis. Significantly, when it binds Cu2+ or Fe3+ , A reduces the metal ions and produces H2 O2 by electron transfer to O2 . This is important because of growing evidence for oxidation injury in AD, and Dr. Bush speculated that A cytotoxicity could be mediated by Cu2+ :A interactions that generate H2 O2 , but this process might be blocked or attenuated by chelators of zinc and/or copper. However, since complications from the Elan vaccine clinical trial were reported after this symposium, it is not yet clear if the mechanisms alluded to here by Dr. Bush account for any of these adverse events. For the reasons summarized above, Dr. Bush and his colleagues have pursued metal-A interactions as a target for therapeutics, and they recently reported that A1–42 possesses an attomolar affinity Cu2+ binding site that is among the strongest known binding affinities, and this is similar to the affinity of Cu2+ for the antioxidant enzyme superoxide dismutase 1. Dr. Bush and his team also found that the Cu2+ -mediated H2 O2 production by A is suppressed by co-binding of Zn2+ , and indeed, despite precipitating A into amyloid, Zn2+ rescues A neurotoxicity in cell culture systems. These and other studies have led Dr. Bush and his colleagues to conclude that AD amyloid plaques may represent the redox-silencing and entombment of A by zinc. Hence, they speculate that SPs may not be the pathogenic culprit in AD, and there is evidence to suggest that the toxic forms of A may be the soluble species or the diffused non-fibrillar A collections in the brains of cognitively impaired patients. On the basis of growing information about Zn2+ and Cu2+ interactions with A mediating the biochemistry of A, Dr. Bush and colleagues recently embarked on a trial of copper-chelators to attempt to inhibit A accumulation in Tg2576 transgenic mice that model AD-like A amyloidosis. Treatment with clioquinol (CQ), a retired antibiotic and a bioavailable Cu/Zn chelator, induced a 49% decrease in brain A deposition in a blinded study of Tg2576 mice treated orally with CQ for 9 weeks. Further, there was no ev-
idence of neurotoxicity or increased non-amyloid pathology. This drug may work by a combined action that facilitates disaggregation of the A collections, while also inhibiting H2 O2 production. Importantly, CQ treatment did not induce a loss in metal levels systemically, probably because it is a relatively weak chelator, and these metals may be redistributed rather than excreted following CQ administration. Thus, CQ treatment appears to be a potent inhibitor of A accumulation. CQ may be the first credible drug candidate based on the amyloid hypothesis of AD, and Dr. Bush reported that a phase II double-blind clinical trial on the effects of CQ in AD patients is currently in progress. Virginia M.-Y. Lee, Ph.D., University of Pennsylvania School of Medicine “Amyloid Binding Ligands as Alzheimer’s Disease Therapies” Dr. Lee summarized studies conducted by investigators in CNDR in collaboration with the laboratory of Dr. H. Kung and Dr. M.-P. Kung at Penn that seek to identify new therapies for AD. The therapeutic targets of the studies they have conducted and propose for the future are the extracellular SPs formed by fibrillar A. Dr. Lee noted that, although, A vaccines and inhibitors of amyloidogenic secretases are potential AD therapies, multifaceted strategies may be needed to effectively interrupt A amyloidosis and prevent/arrest AD. Thus, she and her colleagues have specifically targeted the inhibition of A fibrillization as a potential therapy for AD. Accordingly, Dr. Lee noted that certain amyloid-binding molecules, such as Congo red (CR) and chrysamine G (CG), can indeed arrest the formation of A fibrils; however, CR and CG are unsuitable for AD therapy because they do not cross the BBB. Therefore, working with the Kung lab at Penn, Dr. Lee and colleagues in CNDR have generated novel CG derivatives and other small molecules that specifically recognize fibrillar A in vitro, arrest the formation of A fibrils, and cross the BBB of transgenic mice that model AD amyloidosis. As proof of their ability to cross the BBB and of their high specificity for A fibrils in vivo, Dr. Lee and colleagues showed that following intravenous injection in transgenic mice these compounds specifically label AD-like brain deposits of fibrillar A. Building on these preliminary data, they propose to test the hypothesis that small amyloid-binding compounds that inhibit A fibrillization in vitro and cross the BBB are potential AD therapeutic agents. To accomplish this, Dr. Lee described a multi-disciplinary research program, including the design and synthesis of novel amyloid-binding compounds, biophysical, biochemical, molecular biological and ultrastructural studies using cell culture and animal models of AD-like A amyloidosis to develop such novel amyloid-binding compounds and evaluate their potential as therapeutic agents for the treatment of AD. While these studies are still at the preclinical stage in model systems, amyloid disrupting compounds hold promise not only for targeting A amyloidosis, but also
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
other types of brain amyloid including those due to accumulations of tau and alpha-synuclein fibrils. Secretase Inhibition To Reduce A Levels As Therapy For Alzheimer’s Disease Martin Citron, Ph.D., Amgen Incorporated “Beta-secretase Inhibitor Therapy for Alzheimer’s Disease” Dr. Citron began his presentation by emphasizing that the cerebral deposition of A is an early and critical feature of AD, and that A is released from APP by the sequential action of two proteases, beta-secretase and gamma-secretase. Thus, Dr. Citron noted that these proteases are prime targets for therapeutic intervention, and for this reason he and his colleagues recently cloned a novel aspartic protease, BACE1, with all the known properties of the beta-secretase that is responsible for the release of A from APP. Dr. Citron then briefly summarized how he and his colleagues conducted studies that led to the initial identification of BACE1, and he also described some of the distinct properties of BACE1. He then went on to review the crystal structure of BACE1 as well as data from studies conducted by his group on BACE1 knockout mice that he and his colleagues recently generated. This presentation concluded with a discussion of the pros and cons of BACE1 inhibition as a therapeutic strategy compared to other amyloid therapies. While a lively discussion of this approach to AD therapy followed, the approach is still largely theoretical since there is little published preclinical data from studies in model systems upon which to base solid predictions on the relative merits of this approach to the therapy of AD compared to the other therapies that currently target the levels of a A and the burden of A deposits in the CNS directly. Yue-Ming Li, Ph.D., Merck and Company “Gamma-secretase: Identification and Implications for Alzheimer’s Disease Therapy” Dr. Li began his presentation by reviewing the known biology of gamma-secretase, a membrane-bound protease that cleaves within the transmembrane region of APP to generate the C-termini of the two predominant forms of A known as A40 and A42, which are major constituents of SPs in AD brains. Further, Dr. Li pointed out that presenilin 1 (PS1) and presenilin 2 (PS2) are polytopic membrane spanning proteins that harbor mutations which give rise to early onset familial AD. Dr. Li then reviewed recent biochemical studies which provide compelling evidence that presenilins are novel aspartyl proteases that mediate gamma-secretase activity when engaged in the putative gamma-secretase macromolecular complex. PS1 and PS2 activities can be discriminated from one another on the basis of their susceptibility to inhibition by a potent gamma-secretase inhibitor.
989
Thus, Dr. Li concluded that presenilin/gamma-secretase or the complex that mediates this enzyme activity is a potential target for AD therapy and plays an important role in regulated intramembrane proteolysis. However, as with the efforts described above to develop beta-secretase inhibitor therapy for AD, there is little published preclinical data on gamma-secretase inhibitors upon which to base solid predictions on the merits of this approach to the therapy of AD. Summary And Conclusions On Emerging Therapies For Alzheimer’s Disease Sangram Sisodia, Ph.D., University of Chicago “Current Status Of Emerging Therapies For Alzheimer’s Disease” Dr. Sangram Sisodia provided a summary and critique of the foregoing scientific presentations wherein he outlined the strengths and weaknesses of each of the approaches discussed at the symposium. A number of these points, as well as others made by some of the >300 symposium participants are incorporated into the summaries of each of the presentations described above. Christopher Clark, M.D. and Jason Karlawish, M.D., University of Pennsylvania “What Are The Next Steps? The Challenges Of Transforming Laboratory Advances Into Treatments For Patients” Finally, the symposium was concluded by a stimulating session that was designed for attendance by scientists and clinicians as well as for the caregivers and families of AD patients. This session began with tandem presentations by Dr. Clark and Dr. Karlawish who summarized the key elements that are needed for the ethical and informative conduct of human clinical trials of new AD therapies. They noted a number of unfinished tasks for the design of such trials including issues such as informed consent for trials involving subjects who may not be able to give consent themselves, but instead require surrogates for this, and the importance of the “dyad” of patient and caregiver in judging the merits and efficacy of new drugs for AD. They also emphasized the need for better outcome measures for testing new drugs including better imaging strategies and reliable AD biomarkers that can be used in conjunction with clinical outcome measures to assess the response of the disease process to therapeutic interventions, and they discussed the types of patient populations that are most suitable for evaluating new AD therapies in a timely and informative manner. Notably, in addition to the efforts of industry to carry out clinical trials of proprietary drugs for AD, the National Institute on Aging of the National Institutes of Health has established an infrastructure of AD Centers around the USA for this and a funding mechanism through the AD Clinical Studies program to conduct clinical trials of other potential therapies for AD [4,5,7]. This
990
J.Q. Trojanowski / Neurobiology of Aging 23 (2002) 985–990
presentation stimulated robust discussion and dialogue among all the participants on the prospects of translating laboratory advances in AD research into effective therapies for AD patients.
the research leading to the identification of better therapies for AD would not be possible without the support of the families of our AD patients. References
Acknowledgments The co-directors of CNDR who organized this symposium kindly acknowledge a grant from Nastech Pharmaceutical Co., Inc. to support this meeting, as well as generous support for the meeting from Janssen Pharmaceutica. Other support came from the Memory Disorder Clinic, the Institute on Aging and the National Institute of Aging AD Center Core at Penn as well as from all of the members of CNDR, but especially Ms. Gayle Viale, who provided key design, logistical and administrative support. In addition, the National Institute on Aging of the National Institutes of Health, and the Oxford Foundation made the symposium possible. Dr. V.M.-Y. Lee is the recipient of the John H. Ware 3rd Chair for Alzheimer’s Disease Research. Additional information on AD and related neurodegenerative diseases can be obtained at http://www.med.upenn.edu/cndr, the CNDR website. Finally, all of the speakers are thanked for their insightful presentations and discussions at the meeting, and
[1] Galvin JE, Lee VM-Y, Trojanowski JQ. Synucleinopathies: Clinical and pathological implications. Arch Neurol 2001;58:186–90. [2] Irizarry MC, Hyman BT. Alzheimer disease therapeutics. J Neuropath Exper Neurol 2001;60:923–8. [3] Mayeux R, Sano M. Treatment of Alzheimner’s disease. New Engl J Med 1999;341:1670–9. [4] Morris JC, Ernesto C, Schafer K, Coats M, Leon S, Sano M, et al. Clinical dementia rating training and reliability in multicenter studies: The Alzheimer’s disease cooperative study experience. Neurol 1997;48:1508–10. [5] Sano M, Ernesto C, Klauber MR, Schafer K, Woodbury P, Thomas R, et al. Ratinale and design of a multicenter study of selegiline and alpha-tocopherol in the treatment of Alzheimer disease using novel clinical outcomes Alzheimer’s disease cooperative study. Alzheimer Dis Assoc Disord 1996;10:132–40. [6] Trojanowski JQ, Lee VM-Y. “Fatal attractions” of proteins: A comprehensive hypothetical mechanism underlying Alzheimer’s disease and other neurodegenerative disorders. Ann New York Acad Sci 2000;924:62–7. [7] Trojanowski JQ. Alzheimer’s disease centers and the dementias of aging program of the national institute on aging: A brief overview. J Alzheimer’s Dis 2001;3:249–51.