© 2004 Nature Publishing Group http://www.nature.com/naturemedicine
CO R R E S P ON D E N C E Senile plaque cores isolated from Alzheimer disease cases have been observed to be phagocytosed in vitro by rat microglia7, indicating by a separate approach the ability of microglia to digest Aβ plaques. Nicoll et al.1 concluded that the reduction in plaques observed in their case was caused by a specific immune response to the Aβ vaccine. This would be analogous to results from similar vaccinations of transgenic mice overexpressing the human amyloid precursor protein8. But if Aβ-specific antibodies were responsible for the senile plaque clearance, then a significant difference should have been observed between the intensity of IgG staining of plaques in the vaccinated case compared with other Alzheimer disease cases. No such differences were found1, although it could be argued that Aβ-specific antibody production had declined since the time of vaccination. The role of amyloid precursor protein, the parent protein of Aβ, is unknown, although it is upregulated after brain injury9,10. It is possible, then, that inflammatory stimuli may increase amyloid precursor protein production by neurons, thus increasing Aβ accumulation. This might account for the appearance of plaques in areas of the vaccinated Alzheimer brain such as the cerebellum, which contains only diffuse deposits in most cases of Alzheimer disease. Inflammation is a double-edged sword. It is normally a beneficial process, in which viable host tissue is spared while invaders and damaged tissue are dispatched. But the line can be crossed when an autotoxic process is started11. In this situation, the higher the level of inflammation, the greater the degree of damage to viable host tissue. One process favoring phagocytosis of plaques is their ability to activate complement. Aβ itself is a powerful complement activator12, and plaque material is richly decorated by C3d and C4d, which are products of complement activation13. Microglia and macrophages express very high levels of complement receptors (Fig. 1c). The danger of strong complement activation, as far as neurons are concerned, is generation of the membrane attack complex. This complex can attack dystrophic neurites and neuropil threads in Alzheimer disease13,14. Mechanisms to defend against such attack, such as the upregulation of protectin or CD59, do not seem to occur in Alzheimer disease15. Our observation of neuronal damage in the ischemic area, as well as the
118
observation of neuronal fiber loss by Nicoll et al. in the area affected by encephalitis in the vaccinated Alzheimer brain, might be due in significant part to this phenomenon. Beneficial effects were observed when transgenic mice were vaccinated with human Aβ8. Mouse C1q recognizes human Aβ poorly16, and the inflammatory response observed in transgenic mice is much lower than that in human Alzheimer disease17. Stimulating an inflammatory response through vaccination might thus be beneficial in transgenic mice, but not necessarily in human Alzheimer disease patients17. It will be important to obtain further in vivo data from vaccinated cases through magnetic resonance imaging or, if the opportunity arises, further autopsy information. Intensive investigations into the mechanisms of vaccination-induced Aβ removal and encephalitis are needed. ACKNOWLEDGMENTS We thank M. Kato of Soga Hospital for providing the brain sample. This research was supported by grants to P.L.M. from the Jack Brown and Family AD Research Fund and the Alzheimer Society of Canada/CIHR/AstraZeneca. This study was approved by the ethical committee of the Tokyo Institute of Psychiatry. COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests.
Haruhiko Akiyama1 & Patrick L McGeer2 1Tokyo Institute of Psychiatry, Tokyo, 156-8585, Japan. 2Kinsmen Laboratory of Neurological Research, University of British Columbia, Vancouver, British Columbia, V6T 1Z3, Canada. Correspondence should be addressed to H.A. (
[email protected]) or P.L.M. (
[email protected]). 1. Nicoll, J.A.R. et al. Nat. Med. 9, 448–425 (2003). 2. Wisniewski, H.M., Barcikowska, M & Kida, E. Acta Neuropathol. 81, 588–590 (1991). 3. Akiyama, H. et al. Neurosci. Lett. 219, 115–118 (1996). 4. Akiyama, H. et al. Neurosci. Lett. 206, 169–173 (1996). 5. Akiyama, H. et al. Glia 25, 324–331 (1999). 6. Sheng, J.G., Price, D.L. & Koliatsos, V.E. J. Neurosci. 22, 9794–9799 (2002). 7. DeWitt, D.A., Perry, G., Cohen, M., Dollar, C. & Silver, J. Exp. Neurol. 149, 329–340 (1998). 8. Schenk, D. et al. Nature 400, 173–177 (1999). 9. Shigematsu, K. & McGeer, P.L. Brain Res. 593, 117–123 (1992). 10. Iwata, A., Chen, X.H., McIntosh, T.K., Browne, K.D. & Smith, D. J. Neuropath. Exp. Neurol. 61, 1056–1068 (2002). 11. McGeer, P.L. & McGeer, E.G. Neurobiol. Aging 22, 799–809 (2001). 12. Rogers, J. et al. Proc. Natl. Acad. Sci. USA 89, 10016–10020 (1992). 13. McGeer, P.L., Akiyama, H., Itagaki, S. & McGeer, E.G. Neurosci. Lett. 107, 341–346 (1989). 14. Webster, S., Lue, L.F., Brachova, L., Tenner, A.J. &
McGeer, P.L. Neurobiol. Aging 18, 415–421 (1997). 15. Yasojima, K., McGeer, E.G. & McGeer, P.L. Brain Res. 833, 297–301 (1999). 16. Webster, S.D., Tenner, A.J., Paulo, T.L. & Cribbs, D.H. Neurobiol. Aging 20, 297–304 (1999). 17. McGeer, P.L. & McGeer, E.G. Neurobiol. Aging 24, 391–395 (2003). 18. Akiyama, H. & McGeer, P.L. J. Neuroimmunol. 30, 81–93 (1990).
Nicoll et al. reply: We thank Akiyama and McGeer for pointing out, in this issue of Nature Medicine (see page 117), that amyloid-β (Aβ) removal by microglia in our previous studies1 may be a nonspecific consequence of brain inflammation, rather than a specific result of Aβ immunization mediated by Aβ-specific antibodies. As Akiyama and McGeer indicated, in brain infarction all tissue components undergo necrosis and are phagocytosed by microglia, including Aβ plaques if they are present. A further point in favor of their argument is that nonspecific stimulation of microglia by injection of lipopolysaccharide into brains of aged amyloid precursor protein (APP) –transgenic mice results in clearance of Aβ2. Microglial activation also occurs in response to traumatic brain injury, resulting in reduced Aβ deposition in aged APP–transgenic mice3. Reduced Aβ deposits were not found in aged human survivors of traumatic brain injury4, however, despite the presence of activated microglia. Indeed, traumatic brain injury may be a trigger of Alzheimer pathology5 rather than a protective event. Akiyama and McGeer show that there may be small areas with reduced plaque density in unimmunized Alzheimer disease (their Fig. 1a). In marked contrast, the patient immunized with Aβ in the Elan trail had extensive areas of cortex that were devoid of Aβ plaques5 (for example, see our Fig. 1d)1. Current evidence suggests two distinct mechanisms of Aβ removal when Aβ-specific antibodies enter the brain6. First, diffuse, nonfibrillar Aβ may be rapidly removed between 4 and 24 h, by a process not involving a cellular reaction. Second, compact fibrillar Aβ plaques may be removed by 3 d, possibly by microglial phagocytosis of opsonized Aβ. In immunized mice, removal of Aβ from the brain is associated with improvement of cognitive function 7. This is supported by evidence from patients in the Elan trial. Patients who responded to immunization by producing plaque-binding, Aβ-specific antibodies showed stabilization of cogni-
VOLUME 10 | NUMBER 2 | FEBRUARY 2004 NATURE MEDICINE
© 2004 Nature Publishing Group http://www.nature.com/naturemedicine
CO R R E S P ON D E N C E tive function over 12 months, whereas patients who did not produce such antibodies continued to show cognitive decline8. It is tempting to speculate that in the patients who produced plaque-binding antibodies, Aβ was removed from the brain, and this stablized cognitive function. We were careful to point out that some or all of the features we described might have been a chance finding unrelated to immunization1. However, the brains of additional patients immunized in the Elan trial are now being examined (ref. 9 and E. Masliah, personal communication), and show essentially the same findings1. These data provide strong evidence that removal of Aβ plaques is a direct consequence of immunization, although we agree that the detailed mechanisms of Aβ removal remain to be established. We described two different types of brain inflammation microglial activation and Tlymphocyte infiltration1. Either of these
inflammatory processes alone might have resulted in the acute neurological events termed ‘encephalitis’, so it is important to know which was responsible. Either cell type can have harmful effects, but whereas microglia are probably involved directly in the (presumably beneficial) Ab removal, the T lymphocytes are unlikely to be directly involved. This is important because efforts in devising the next generation of Aβ immunotherapy are being targeted at avoiding the T-lymphocyte reaction, but it may be the microglial activation that causes the harmful side effect. Furthermore, is harmful encephalitis necessary for removal of Aβ, or did the ~95% of patients in the trial who did not have encephalitis also experience potentially beneficial removal of Aβ plaques? Clearly, we still have much to learn from this trial. COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests.
James A R Nicoll1,2, David Wilkinson3, Clive Holmes3, Phil Steart2, Hannah Markham1,2 & Roy O Weller1,2 1Division of Clinical Neurosciences, University of Southampton, Southampton General Hospital, Southampton, SO16 6YD, UK. 2Neuropathology, Department of Pathology, Southampton General Hospital, Southampton, SO16 6YD, UK. 3Division of Clinical Neurosciences, University of Southampton, Memory Assessment and Research Centre, Moorgreen Hospital, Southampton, SO30 3JB, UK. e-mail:
[email protected] 1. Nicoll, J.A. et al. Nat. Med. 9, 448–452 (2003). 2. DiCarlo, G., Wilcock, D., Henderson, D., Gordon, M. & Morgan, D. Neurobiol. Aging 22, 1007–1012 (2001). 3. Nakagawa, Y. et al. Exp. Neurol. 163, 244–252 (2000). 4. Macfarlane, D.P., Nicoll, J.A.R., Smith, C. & Graham, D.I. NeuroReport 10, 1–4 (1999). 5. Nicoll, J.A., Roberts, G.W. & Graham, D.I. Nat. Med. 1, 135–137 (1995). 6. Wilcock, D.M. et al. J. Neurosci. 23, 3745–3751 (2003). 7. Janus, C. et al. Nature 408, 979–982 (2000). 8. Hock, C. et al. Neuron 38, 547–554 (2003). 9. Ferrer, I. et al. Brain Pathol. 14, 11–20 (2004).
International cooperation on xenotransplantation To the editor: We are writing on behalf of the Ethics Committee of the International Xenotransplantation Association (IXA). Our committee recently expressed concern about the need for international guidelines for overseeing clinical xenotransplantation. We have published letters in scientific journals and corresponded with representatives of the World Health Organization (WHO) regarding these concerns. Although clinical xenotransplantation provides a potentially promising solution to the shortage of human organs and tissues, the potential to introduce new infections from xenogeneic source animals into the human populace is a concern that mandates extreme caution. In view of this risk, considerable effort has gone into the development of guidelines and oversight procedures for husbandry of source animals and recipient monitoring in several countries. However, clinical xenotransplantation is also carried out in countries lacking such guidelines and oversight. Some of these transplants are commercial enterprises providing cosmetic and medical treatments, whereas others are clinical trials. Given that individuals may freely travel from one country to another to undergo such procedures, international cooperation is needed to develop universal oversight
procedures and standards, including ethical and monitoring guidelines. Last October, international health officials representing more than 20 member states, members of the transplantation community and WHO representatives met in Madrid to discuss policies on transplantation. This meeting resulted in a report by the WHO secretariat, which has recently been published online (http://www.who. int/gb/EB_WHA/PDF/EB113/eeb11314.pdf). This document includes a draft resolution to be considered by the WHO’s executive board in January 2004. The draft resolution recognizes both the potential benefit and the potential infectious risks associated with xenotransplantation. Most importantly, it urges the adoption of appropriate regulation and surveillance of xenotransplantation by national health authorities of member states, as well as the development of protective measures to prevent transmission of infections that may arise after xenotransplantation. Member states are further urged to support international collaboration on the prevention and surveillance of infections resulting from xenotransplantation. Our committee believes that adopting all of these measures would be of enormous value in minimizing the potential risks of xenotransplantation. The draft resolution further requests that
NATURE MEDICINE VOLUME 10 | NUMBER 2 | FEBRUARY 2004
the WHO’s director-general support such international cooperation, create a global evidence base for evaluation of xenotransplantation practices, and provide technical assistance and expertise to support xenotransplantation efforts and oversight in member states. The IXA’s ethics committee is highly encouraged by the drafting of this resolution and strongly urges its adoption by the WHO’s executive board. If this resolution is ultimately adopted by the fifty-seventh WHO assembly, it will be a major step toward minimizing infectious risks associated with xenotransplantation. Without organized international cooperation, the best efforts at minimizing these risks in countries with appropriate regulatory oversight may be thwarted by the free travel of individuals undergoing unmonitored xenotransplantation in countries lacking such regulation. Megan Sykes1, Mauro Sandrin2 & Emanuele Cozzi3 1Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts 02129, USA. 2Austin Repatriation Medical Center, Austin Research Institute, Heidelberg, VIC 3084, Australia. 3Department of Surgery & Medical Sciences, Clinica Chirurgica IV, University of Padua, Padova 35128, Italy. e-mail:
[email protected]
119
