. Extracellular amyloid formation and associated pathology in neural grafts. Nat Neurosci. 2003 Apr;6(4):370-7. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on Primary Papers and News

  1. The paper by Meyer-Luehmann et al. takes a novel approach to investigating the mechanisms of cerebral Aß accumulation in vivo: The researchers transplanted embryonic cortical and hippocampal neurons from transgenic AßPP23 mice and wild-type B6 mice into B6 and AßPP23 hosts, respectively. They found that AßPP23 grafts in B6 hosts developed significantly fewer amyloid deposits than in comparably aged AßPP23 transgenic mice, despite the fact that the grafts produced ample AßPP. Conversely, B6 grafts introduced into AßPP23 hosts showed evidence of congophilic amyloid deposition and neuritic and glial pathology as early as three months after transplantation. These results provide compelling evidence that factors other than local Aß production are important in determining whether amyloid pathology will occur. Moreover, these results show that Aß can diffuse over considerable distances in the interstitial fluid to cause pathology at distal sites. Finally, these results highlight the relevance of extracellular amyloid to amyloid pathology, since both neuritic and glial abnormalities characteristic of AD were observed in B6 grafts transplanted into AßPP23 mice.

    At first glance, these intriguing results seem to suggest that overall intracranial Aß levels might be the relevant factor in determining whether amyloid deposition occurs. Indeed, this idea fits well with accumulating evidence that peripheral antibodies or other Aß-binding molecules such as gelsolin can—by "pulling" Aß across the blood-brain barrier through a mechanism resembling dialysis—effectively prevent amyloid deposition by lowering intracranial Aß levels (e.g., DeMattos et al., 2001; Matsuoka et al., 2003). However, Meyer-Luehmann and colleagues go on to show that the likelihood of amyloid deposition is influenced at a still more local level, since grafts transplanted into brain subregions with relatively high Aß levels (e.g., hippocampus) show amyloid deposition more often than do grafts in brain regions with lower Aß levels (e.g., striatum). These data appear to point to the presence of concentration gradients of Aß within the cerebrum, and raise several important questions. What is the nature of the "sinks" within the brain that are responsible for Aß removal? Could shunts provide an artificial means of lowering intracranial Aß levels? Whatever the answer to these questions, this paper provides further evidence that mechanisms of Aß removal are at least as important as those of Aß production in determining amyloid pathology. The neural graft paradigm may provide an elegant way to test some of these fundamental questions.—Malcolm Leissring, Brigham and Women’s Hospital, Boston, Massachusetts.


    . Extracellular amyloid formation and associated pathology in neural grafts. Nat Neurosci. 2003 Apr;6(4):370-7. PubMed.

    . 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 Jul 17;98(15):8850-5. PubMed.

    . Novel therapeutic approach for the treatment of Alzheimer's disease by peripheral administration of agents with an affinity to beta-amyloid. J Neurosci. 2003 Jan 1;23(1):29-33. PubMed.

    View all comments by Malcolm Leissring
  2. Alzheimer’s Disease-like Molecular Pathology in Neural Grafts
    Meyer-Luehmann and colleagues published an interesting paper describing the extracellular formation of amyloid plaques and pathology in a neural graft model. The authors transplanted cell suspensions of embryonic cortical and hippocampal tissue from AβPP23-transgenic mice into three-month-old wild-type and transgenic hosts. Such transgenic grafts into wild-type animals showed no amyloid deposits up to 20 months after transplantation. The grafted neurons showed no reduction in hAβPP expression, Aβ levels, or evidence of a humoral response to the grafted tissue. However, when transgenic or wild-type tissue was transplanted into transgenic hosts, amyloid deposition was observed as soon as three months after transplantation. Reactive gliosis was observed at the host-graft interface. These Aβ deposits were surrounded by activated astrocytes and microglia, dystrophic synaptic boutons, and abnormally stained acetylcholine-positive fibers. Some of these dystrophic neurons also showed evidence of abnormal tau phosphorylation. The presence of Aβ deposits in grafts derived from wild-type vs. the absence of Aβ in transgenic grafts into wild-type mice lead the authors to suggest that there was extracellular diffusion of soluble and insoluble Aβ into and out of the grafted tissue, respectively, in these two graft conditions. These results suggest that human Aβ‚ together with other unknown factors that influence its diffusion, clearance and deposition, may be central to the pathogenesis of Alzheimer’s disease (AD).

    We would like to draw attention to Sykovà et al., 1999 (2) and Martins et al., 2001 (3), studies that also describe a fetal neural transplantation model. Transplants of fetal cerebral cortex onto the midbrain region of neonatal rat hosts results in extensive and chronic reactive gliosis within the grafted tissue. This gliosis is evident as early as one month after transplantation, and is associated with altered AβPP metabolism, a decrease in presenilin-1 levels, and an increase in apolipoprotein E (ApoE) protein levels. Furthermore, altered extracellular matrix deposition and a change in the diffusion characteristics of the extracellular space are observed. Conversely, when the same embryonic tissue is transplanted homotopically into the cortex of neonatal rats, minimal gliosis is observed and no consequent change in AβPP metabolism occurs.

    The graft environment mirrored what is observed in the AD brain. However, rodent Aβ is less toxic and less fibrillogenic than human Aβ (4), so we extended our chronic gliosis model to Tg 2576 transgenic mice that express human Aβ (5). This transgenic tissue was grafted into wild-type hosts and also into ApoE knockout hosts. We again observed chronic gliosis in a region-specific manner, which was associated with altered AβPP metabolism and a significant increase in levels of the C-99 fragment, the precursor of Aβ. Consistent with Meyer-Luehmann et al., after 10 months we found no frank Aβ deposition in transgenic grafts into wild-type hosts. However, our data do suggest that chronic reactive gliosis may alter the metabolism of AβPP to favor Aβ production. As such, the phenomena of astrocytic and microglial reactivity found in AD brains may be a primary pathological trigger, rather than a secondary response that facilitates plaque formation.

    Meyer-Luehmann and colleagues also investigated the possibility that activated glia in wild-type grafts may be responsible for Aβ deposits. They performed a stab injury to the hippocampal region of six-month-old transgenic mice and examined the material three months later. No Aβ deposits were observed, despite some microgliosis, suggesting that a single episode of surgical trauma per se was an inadequate insult to trigger Aβ deposition. However, this is consistent with a stab injury being associated with an acute effect on gliosis and inflammation (6,7), unlike the sustained and chronic reactivity observed in our transplantation model.

    It is apparent from these studies that Aβ deposition is a region-specific event. In our neonatal transplant model, altered Aβ metabolism and reactive gliosis is only observed in cortical grafts to the midbrain. Meyer-Luehmann et al. observed Aβ deposition in areas known to exhibit the most severe neuropathology, such as the hippocampus and thalamus in the AβPP23 mouse. Aβ deposits were not seen in grafts placed on the striatum, a region of relatively low AD neuropathology.

    Such transplantation studies provide a useful tool to study the mechanisms of reactive gliosis and the pathogenesis of AD. Further studies are clearly warranted into the transport of all forms of Aβ between different brain regions, Aβ clearance mechanisms, and the role of the blood-brain barrier in this process. In this regard, molecular chaperones such as the major AD genetic risk factor, ApoE, may play key roles in isoform-specific binding and clearance of Aβ (8,9). This process also depends on the amount of ApoE produced (10), which is most pertinent to the region-specific effects described above.

    The extracellular milieu may also be integral to the pathogenesis of AD. Chronic gliosis is associated with altered production of extracellular matrix components, and this, coupled with hypertrophied astrocytic processes, may affect the diffusion of neurotransmitters, and other substances such as Aβ. It is becoming clear that reactive gliosis is a complex event, as evidenced by the fine balance between pro- and antiinflammatory signals in normal brain. One can easily envisage a situation where an accumulation of insults over time may disturb this balance, resulting in the cascade of neurodegenerative events observed in AD. Transplantation studies should facilitate investigation into the process of reactive gliosis, the effect of molecular chaperones and the extracellular space in Aβ clearance, and the toxic effects of Aβ on both neurons and glial cells. Thus, transplantation studies should provide valuable insights into the molecular mechanisms of AD pathogenesis and serve as an excellent model for the evaluation of potential therapeutic agents.


    . Extracellular amyloid formation and associated pathology in neural grafts. Nat Neurosci. 2003 Apr;6(4):370-7. PubMed.

    . Astrocytes, oligodendroglia, extracellular space volume and geometry in rat fetal brain grafts. Neuroscience. 1999;91(2):783-98. PubMed.

    . Altered expression of apolipoprotein E, amyloid precursor protein and presenilin-1 is associated with chronic reactive gliosis in rat cortical tissue. Neuroscience. 2001;106(3):557-69. PubMed.

    . Amyloidogenicity of rodent and human beta A4 sequences. FEBS Lett. 1993 Jun 14;324(2):231-6. PubMed.

    . Chronic gliosis triggers Alzheimer's disease-like processing of amyloid precursor protein. Neuroscience. 2002;113(4):785-96. PubMed.

    . Phenotypic diversity and kinetics of proliferating microglia and astrocytes following cortical stab wounds. Glia. 1996 Apr;16(4):368-82. PubMed.

    . Diffusion barriers evoked in the rat cortex by reactive astrogliosis. Glia. 1999 Oct;28(1):40-8. PubMed.

    . Characterization of the binding of amyloid-beta peptide to cell culture-derived native apolipoprotein E2, E3, and E4 isoforms and to isoforms from human plasma. J Neurochem. 1997 Feb;68(2):721-5. PubMed.

    . Apolipoprotein E promotes the binding and uptake of beta-amyloid into Chinese hamster ovary cells in an isoform-specific manner. Neuroscience. 1999;90(4):1217-26. PubMed.

    . Expanding the association between the APOE gene and the risk of Alzheimer's disease: possible roles for APOE promoter polymorphisms and alterations in APOE transcription. J Neurochem. 2003 Mar;84(6):1215-36. PubMed.

  3. This is an interesting study by Mathias Jucker and colleagues that attempts to define the role of extracellular versus intracellular β-amyloid in plaque formation. The pivotal experiment of the paper is their injection of embryonic brain tissue from wild-type mice into the brain of plaque-forming AβPP23 transgenic mice. I recall my surprise when first hearing that wild-type grafts in AβPP transgenic mice appear to develop amyloid plaques even prior to plaque formation in the transgenic host when they presented this work at the 2001 Society for Neuroscience meeting. Their remarkable findings raise intriguing questions, such as by what mechanism the wild-type mouse tissue, which in its normal surroundings would never develop plaques, actually develops plaques prior to the host plaque-forming tissue.

    Nevertheless, I wonder how the investigators were able so precisely to demarcate graft versus host tissue. Their observation that plaques develop at the periphery of grafts could also suggest that plaques occur in injured host tissue just outside the margins of the graft. Indeed, injury and/or inflammation are associated with increased plaque formation and could explain why the injured transgenic host tissue develops more extensive plaques. The needle stab control was one of their attempts to explore this possibility.

    Overall, the paper provides a unique approach to investigate plaque formation in AD. Unfortunately, it again moves us away from what appeared to be an evolving consensus (see recent Alzforum chat) for a critical role of both extracellular and intracellular β-amyloid in plaque formation back to one favoring mainly extracellular β-amyloid, though this is unlikely to be the final chapter in this now century-old debate on the origin of plaques.

    View all comments by Gunnar Gouras
  4. Comment by Berislav Zlokovic—Posted 5 March 2003.
    The paper by Meyer-Leuchmann et al. uses transplant biology to provide convincing evidence that CNS transport of soluble Aß is a major factor implicated in amyloid formation and neurotoxicity. The significant delay in amyloid deposits in AßPP23 cellular neuronal embryonic grafts in wild-type hosts versus AßPP23 hosts, and the development of amyloid in wild-type neuronal grafts in AßPP23 host, are highly suggestive that extracellular factors and CNS transport play a primary role in regulating brain Aß levels and amyloid deposition. To give a different flair on this study from an alternative "vasculocentric" view, the altered vascular biology could also be a contributory factor in the "transplant" paradigm. Namely, wild-type grafts in AßPP23 hosts at the time of grafting (six months) are neovascularized by AßPP-primed vasaculature, while AßPP23 grafts in wild-type hosts are neovascularized by healthy host vascular cells. Since the vascular system plays an important role in brain efflux of Aß, as well as in the influx of circulating Aß into the CNS, this systemic vascular factor may well work in favor of early deposition of Aß in wild-type grafts in AßPP23 hosts, and delayed deposition into AßPP23 grafts in wild-type hosts. Once AßPP23 grafts in wild-type hosts change the host vascular system into one of the AßPP23 types, the pendulum swings back, leading to amyloid accumulation in the graft.

    View all comments by Berislav Zlokovic
  5. These elegant experiments by Meyer-Luehmann and colleagues demonstrate that the proteopathic corruption of tissue can be more strongly influenced by the environment in which the tissue develops than by the lineage of the tissue itself (I will forego the obvious sociological analogies). Specifically, brain tissue grafted from AβPP-overexpressing transgenic mice into nontransgenic host mice remains remarkably refractory to the deposition of Aβ, whereas even tissue from nontransgenic mice develops diffuse and congophilic plaques when grafted into AβPP-transgenic mice. Given the high neuronal expression of mutant human AβPP and the corresponding increase in Aβ in the transgenic mouse brains, the results seem counterintuitive. But are they? The authors argue, in essence, that Aβ moves down a concentration gradient in brain; the mechanism remains unclear (passive diffusion? transport of some kind?), but the tendency of the transgenic host's Aβ to collect in the artificial sink created by the graft might explain the relatively rapid and intense plaque formation in both transgenic and nontransgenic grafts.

    Conversely, in a nontransgenic host, the Aβ formed within a transgenic graft flows out of the graft and becomes so diluted in the surrounding tissue that, in most instances, neither graft nor host brain is sullied by deposits. The idea has been around for years that lowering the local concentration of amyloidogenic proteins can reduce their tendency to aggregate, but the results from the Jucker lab add weight to the argument that simply augmenting a concentration gradient (e.g., by immunotherapy or CSF shunting) could have therapeutic benefit. Enhancing removal of Aβ by stimulation of intracellular degradation or active transcytosis also has been suggested. Indeed, the Basel group found that grafts into the transgenic striatum, an area that produces less Aβ than does cortex and hippocampus, are much less likely to develop deposits, indicating that a relatively small reduction in local Aβ concentration may be sufficient to impede the pathogenic cascade.

    What is next? Part of the value of these experiments (in addition to the considerable effort that went into their execution) is their attempt to explain β-amyloidogenesis in the complex environment of the brain. Not unexpectedly, additional questions arise that might be addressed in the graft model. For example, the authors note the presence of acetylcholinesterase-positive neurites around some plaques; many of the cholinesterase-positive axons in neocortex and hippocampus arise from the cholinergic neurons of the basal forebrain cholinergic system, although cholinesterase is not specific to cholinergic cells. What is the origin of these neurites? Ingrowing axons, or some intragraft source? Is there evidence of stem cell-related activity in or near the grafts? And how do the cellular populations in the grafts (which appear to be largely self-contained) relate to the normal makeup of the host tissue? Specifically, has there been phenotypic drift in neuronal transmitter specificities, for example, and does this influence the pathological response of the graft? How would grafts from plaque-resistant brain areas such as cerebellum or striatum fare in the cortex or hippocampus of a transgenic host (i.e., are there region-specific features that render these areas resistant to amyloidogenesis)? The grafting model described by Meyer-Luehmann and colleagues presents promising new opportunities for applying the important lessons we have learned from molecular and cellular studies of β-amyloidogenesis to the milieu in which it matters most—the living brain.

    View all comments by Lary Walker
  6. Meyer-Luehmann et al. provide convincing evidence, through a set of elegant experiments, that diffusion of soluble Aß in the extracellular space is sufficient for the formation of plaques. They observed Aß deposits in tissue from wild-type mice when grafted into the brains of AßPP23 transgenic mice. These transgenic mice overexpress an AD-mutated form of the amyloid precursor protein (AßPP), display abnormally high levels of AßPP and Aß, and an age-dependent deposition of Aß in the brain. Importantly, no changes in the expression of AßPP or neuronal accumulation of Aß were observed in the grafts. The Aß deposited in the grafts appears to have originated entirely from the host, carried to grafts via passive extracellular diffusion. The few mature amyloid deposits present in these grafts were associated with neuronal and neuritic pathology. The latter finding is consistent with the large body of evidence indicating that fibrillar Aß, which is present extracellularly in mature plaques, exerts a toxic effect on neurons (10), particularly in the primate brain (3). In summary, extracellular diffusion of Aß, which most likely originates from the proteolytic processing of AßPP at the membrane, appears to be sufficient for the deposition of Aß in plaques and the subsequent plaque pathology.

    A superficial interpretation of the above findings would be that intracellular production and accumulation of Aß does not contribute to the pathology caused by this peptide. This interpretation, however, runs counter to accumulating evidence. As summarized in the report by Meyer-Luehmann et al., both in vitro and in vivo investigations have indicated that at least one pool of Aß is produced intracellularly, and appears to accumulate in neurons. Accumulation of Aß in cortical pyramidal and nonpyramidal neurons has been reported in normal elderly individuals (4), Alzheimer’s disease (AD) (2,5), individuals with mild cognitive impairment (4), Down’s syndrome (6), and in AßPP-presinilin-1 transgenic animals (9). Significantly, intracellular accumulation of neuronal Aß in many of the above models precedes the formation of plaques.

    It is quite likely that accumulation of high concentrations of intracellular Aß exerts detrimental effects on neurons. Consistent with this possibility, expression of high levels of intracellular Aß have been implicated in damage to neurons and other cells (1,2,7). Microinjection of Aß-42 or cDNA expressing cytosolic Aß-42 into cultured primary human neurons resulted in rapid and selective cell death (11). In normal elderly individuals and particularly those with AD, intraneuronal Aß has been shown to accumulate in multivesicular bodies localized in synapses, and to be associated with synaptic pathology (8). Of particular interest, recent evidence suggests that intraneuronal Aß may form aggregates and cause lysis of neurons, leading to an extracellular deposit (2,4). Thus, in addition to its likely deleterious effects on neurons, intraneuronal Aß may serve as one source of Aß deposited in plaques.

    Based on the evidence for the abnormal effects of both accumulated intracellular Aß and extracellularly deposited Aß, we have proposed a hypothetical two-component model of Aß pathology in AD (see Alzforum live discussion). According to this model, early accumulation of Aß in neurons represents the first component of Aß pathology and causes neuronal and synaptic dysfunction, and perhaps neuronal loss. Extracellular deposition of Aß in plaques, which, as demonstrated by Meyer-Luehmann et al., need not be related to intraneuronal accumulation of Aß, represents the second component and exerts toxic effects on neurons and processes.—Changiz Geula, Beth Israel Deaconess Medical Center, Boston, Massachusetts.


    . Fresh and globular amyloid beta protein (1-42) induces rapid cellular degeneration: evidence for AbetaP channel-mediated cellular toxicity. FASEB J. 2000 Jun;14(9):1233-43. PubMed.

    . Evidence that neurones accumulating amyloid can undergo lysis to form amyloid plaques in Alzheimer's disease. Histopathology. 2001 Feb;38(2):120-34. PubMed.

    . Aging renders the brain vulnerable to amyloid beta-protein neurotoxicity. Nat Med. 1998 Jul;4(7):827-31. PubMed.

    . Intraneuronal Abeta42 accumulation in human brain. Am J Pathol. 2000 Jan;156(1):15-20. PubMed.

    . Abeta42-positive non-pyramidal neurons around amyloid plaques in Alzheimer's disease. Lancet. 2000 Jan 1;355(9197):42-3. PubMed.

    . Intraneuronal Abeta42 accumulation in Down syndrome brain. Amyloid. 2002 Jun;9(2):88-102. PubMed.

    . Beta-amyloid peptide expression is sufficient for myotube death: implications for human inclusion body myopathy. Mol Cell Neurosci. 2001 May;17(5):793-810. PubMed.

    . Intraneuronal Alzheimer abeta42 accumulates in multivesicular bodies and is associated with synaptic pathology. Am J Pathol. 2002 Nov;161(5):1869-79. PubMed.

    . Intraneuronal Abeta accumulation precedes plaque formation in beta-amyloid precursor protein and presenilin-1 double-transgenic mice. Neurosci Lett. 2001 Jun 22;306(1-2):116-20. PubMed.

    . Mechanisms of neuronal degeneration in Alzheimer's disease. Neuron. 1996 May;16(5):921-32. PubMed.

    . Selective cytotoxicity of intracellular amyloid beta peptide1-42 through p53 and Bax in cultured primary human neurons. J Cell Biol. 2002 Feb 4;156(3):519-29. PubMed.

    View all comments by Changiz Geula
  7. This publication has to be applauded for the lengthy and painstaking analysis of a two-way graft-host transplantation model, even and although grafting suspended embryonic cells into adult brain is not exactly a minor surgical procedure. The outcome, however, elicited less surprise than a reaction of “seen this before”, since indeed the “pathology” around the plaques in the grafts is, judging from the brief description provided here (fig 7) very “classical” and as such present in old APP23 mice like in our APP[V717I] mice (Moechars et al, 1999; Van Dorpe et al, 2000; Dewachter et al, 2002) as well as in other transgenic amyloid models. The main observation of this study is that Aß accumulates as plaques in grafts (transgenic or WT alike) that are transplanted into APP transgenic (but not in WT) mouse brain. This evidently must be due to import (active or passive) of extracellular amyloid that must be present at high enough levels in the surrounding tissue, i.e. hippocampus or thalamus, not striatum. This clear case of law of mass-action builds a very strong case for a mechanism whereby plaques result simply from the fact that Aß cannot “exit”, in this case from the grafts, and by extension from specified brain-regions in ageing mice - and humans! This being the case, the paper is interesting for what is not (directly) addressed: what is the cellular organization and particularly, the vascularization inside the graft ? If faulty, that single piece of information would up-grade the mechanism of “peri-vascular drainage” to be the most important for clearance of extracellular Aß – and explain CAA as “obligate” pathological lesion in AD brain! Unfortunately, and I agree with commentator Gunnar Gouras, in AD brain we must count with intra- and extra-cellular as well as with plaque Aß as the “toxic” species. This needs not to be (too) problematic if one takes into account the time-scale: early subtle symptoms are due to intra-cellular and synaptic Aß, while the more dramatic later clinical stages are caused by plaque Aß and the various cellular reactions to it. Neuronal death and brain atrophy is then the late stage, observed by pathologists, postmortem.


    . Early phenotypic changes in transgenic mice that overexpress different mutants of amyloid precursor protein in brain. J Biol Chem. 1999 Mar 5;274(10):6483-92. PubMed.

    . Prominent cerebral amyloid angiopathy in transgenic mice overexpressing the london mutant of human APP in neurons. Am J Pathol. 2000 Oct;157(4):1283-98. PubMed.

    . Neuronal deficiency of presenilin 1 inhibits amyloid plaque formation and corrects hippocampal long-term potentiation but not a cognitive defect of amyloid precursor protein [V717I] transgenic mice. J Neurosci. 2002 May 1;22(9):3445-53. PubMed.

    View all comments by Fred Van Leuven

This paper appears in the following:


  1. Aβ, Shifty Drifter? Tissue Grafting Sheds Light on Plaque Formation